AE eee: Se De slew «nee oP Meaty er adie een th Abe DV ay: LIne the Ne hod , Wit NEP ceen ne D4 ee mrt aiy ST Z5p BY ore ole Me tey es Be de Cindi Fe edeatie lela NIV tego HA FIN INN asin ey Ny stihpe Breen Hare dh aU Me Ne frat, rar een n ce “ otha ate eye die Ne Me had ote Ned ana Wu eh a Be Ade gaat 4 We HE Da th the pe } ete OM ef Nour, . Ma Corer fe Ue eta an ey Pica st we MPLS Ne a He henley eEnees Vier tN ete Se Ale oes Sposa We Nha sy Ti SPW Ashe adets a4. uy Ptah oun ong MeN Ge? Sen tnt E Ns ed AE digs f Patan tn we Wile she sey Wen & UW tie "ets Oh die seine ne PEON egy mdi BNI EN Ufo eNO Teo We OME tog a oy TNE poor! D ein ery watt Serna grub ce Auehenebetan's {ON wet gery vin Wy OL TAY Se nee) ean ty sofas tna Nett Wa Ged eof an yh 2B Liat PAM shores Vey r Melt stint Siete Awad Nye s Lee nc, ae ONIN 2808 gen ip Nt athe y INAS eM Noe VV by Pi Weilnce NT hieunaseine Haye Cat LR a TL Nel gu, ATAU Pah tpg EE EN AOR ING Uy, AHA Mi SDE REELS AME Gn Mra ivetiy IN tTEate PEN Bot May ts yoga. VPNH ds tuts NORTE whotine yore PNP a SL Si SAVED as orn) Pat More dN AY sd 4 SEPM Loe itey + ene oe: Mey ek Mole’ IS sin, Nem datadys WN YEAN AN ae PN Aig fa Naf D8 4, ty at iba ore oe CEng MAAN | ye) Nie Seton tof, taneon Ve nee MAAS hy dhe MW Ast Scary Wo MLL f vy WO Ta Line ie citar NM bdiow Pay mn TEER er rn Mh es EM eee aC ace SHRP IN deh as vy at WAN te, (= NW Mans Body, UU Ba Phd eb y Pay w hts MME A apnerny Nidan Leper AN Pb cee Ny eI aye Cyan 4 Nosnaid « FO ps aves pean a 1.29 dh aly MAN DYE) tna ot Py Net VOL Ey tay Pubow ih sya CP Leon enn MM ened rs VEEN Pr ome Movie gy Wd 0d Nd Le ati eet rete Ps eT MSN Eat ON Gen aay PLR) Secu We te my Sb Wye st hay Wet ew ag Lene Rom Mea ed Mere PLFA over ce thd Nf gd ie aes SUR TW) weg us rake “ ‘Nu ba vues “NGL FUN MAN toa Mae Nay a VAN NN th ip N bevy Soret UVa cek. vivre wv yweseh, uw Mer thedlhy wey NN ta ay LNs dN vee A enoAe Cry Hom ¥. Mh Sian hy UPd SHON ley pias PV by hed Beatty BATES vying VSG Vad hin ae oN avin g Vest EMARAY ING tibiagt MERAH y AY Sih vvryey 0 Wann, dy DY voeiter bs ar ee er Baie en OVA ING SM ae aren pe tert NAA Ys ahs oe ay hh can % ANd AV aatiad ose ybote ys wi DRM vane of Hee Goby iets, A MetVAtin ce wa Ny utly 5 bn / Le ner irae WAN AA Denar HY Ts ears Ay AOL phy A) IN aI yey AIA wal y HM al watt ude ah Tes TTS Bo oh Dot AN ND wt atthe a Seiit sy SDPO My a yl hen Lhasa Saal \in Meet) es Paes fread ute NAN a aes SRN a ay NIM yey Qs Mit ot rb AAR aN, DS Vea hyn a Vadis 1/4, hate bee ea my Mest : Nat oh atic 4 abled AM an vq Msi, a? si BF oriM A! ee 6b od on aL H, eM ON ae seh Spat Harsh yi Rae's SN oo way He owiits Neate oe WH si alia oh Chennai dg gw CSN apelin ip ere: A Siw He ete yun PP eC al sarah aie A he WT Ne Sethe 250 5% aut os 7 CN maha oA ny Are tte wy 4 Puede tae PArad 6 ods hang Peter cin ith beth Bi fo trite has le it aay ot. ote SB lrndina se ved al fh Catt tet ake Wakeman re Vela hg, eNep thm 6 ho ra Bie: Meee ap winners . F Retern ey Drei Sea, ney se Sale asp atea a Sat e ‘AA eA ae Saba tonrey Fal aan Moms inept ee a ter ee a ANNALS OF THE ANNALE VAN DIE SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM VOLUME 99 BAND 99 ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM VOLUME 99 BAND Ale TERNS S B:8.B:B ev ti tw) \ 7 E ct 2, S LIQ, at VIO Novi: THE TRUSTEES OF THE DIE TRUSTEES VAN DIE SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM CAPE TOWN KAAPSTAD 1990-1991 SET, PRINTED AND BOUND IN THE REPUBLIC OF SOUTH AFRICA BY THE RUSTICA PRESS (PTY) LTD, NDABENI, CAPE D2109 LIST OF CONTENTS Boomer, I. D. see DINGLE, R. V. et al. Cooper, M. R. A new genus of Rutitrigoniinae (Bivalvia, Trigoniacea) from the Lower Cretaceous (Aptian) of Zululand? (PublishediFebnuary 19902) syns) oe eae DINGLE, R. V., Lorp, A. R. & Boomer, I. D. Deep-water Quaternary Ostracoda from the continental margin off south-western Africa (SE Atlantic Ocean). (Published September 1990.).................... DuRAND, J. F. A revised description of the skull of Moschorhinus (Therapsida, Therocephalia). (Rublishe he DintanyelOONs) ie cere ye tate aeie acne iaciae Ve coe setae ate ap Rse eee a GERNEKE, D. A. see WILSON, M. L. et al. Gow, C. E. A tooth-bearing maxilla referable to Lycorhinus angustidens Haughton, 1924 (Dinosauria, Ornithischia). (Published December 1990.)..................... HILxer, N. Benthic communities and sedimentary facies in the Lower Witteberg Group (evonianySouthAtnica)s(zubliswed uly: 19903) eee sodas oee ace aoe KENNEDY, W. J. & KLINGER, H. C. Cretaceous faunas from Zululand and Natal, South Africa. Hatchericeras Stanton, 1901 (Cephalopoda, Ammonoidea), from the Barremian of Zululand. (Pub- hishve GRU RIOOU Retry cee ian gerne stem neue bieteh ger bat hacte 2 eas Sag ae tua KENNEDY, W. J. see KLINGER, H. C. KLINGER, H. C. & KENNEDY, W. J. Cretaceous faunas from Zululand and Natal, South Africa. A Koloceras (Cephalo- poda, Ammonoidea) from the Mzinene Formation (Albian). (Published | SSL DYRUET EN SED Nes ceehcre cach 6 ecnumioy Se ane a ae atte ean ace ea fre ee nee a KLINGER, H. C. see KENNEDY, W. J. Lorp, A. R. see DINGLE, R. V. et al. McMILLan, I. K. Foraminifera from the late Pleistocene (latest Eemian to earliest Weichselian) shelly sands of Cape Town City Centre, South Africa. (Published May 1990.)......... PETHER, J. A new Austromegabalanus (Cirrepedia, Balanidae) from the Pliocene of Namaqua- land, Cape Province, South Africa. (Published February 1990.) ............... VAN RIJSSEN, W. J. J. see WILSON, M. L. et al. VISSER, J. N. J. Geography and climatology of the late Carboniferous to Jurassic Karoo Basin in south-western Gondwana. (Published March 1991.) .....................005- WILLIAMS, G. C. The Pennatulacea of southern Africa (Coelenterata, Anthozoa). (Published May ISIS) 5) Stes eae oe acter ty cada ba ake o's oe COR Ota ee CC Witson, M. L., VAN RUSSEN, W. J. J. & GERNEKE, D. A. An investigation of the ‘Coldstream Stone’. (Published May 1990.) ................ Volume 99 is complete in 12 parts. Page 23 245 381 367 Mal) 231 15 121 415 187 rae f . on od ¢ i 1S | aiey Pe A ‘ ey, NEW GENERIC NAMES PROPOSED IN THIS VOLUME Aulniriconial Cooper, 1990 i eee ene Rugocythereis Dingle, Lord & Boomer, 1990 Ce OO | F THE SOUTH AFRICAN — Se EUM | CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (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) 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 and neat, double spaced with 3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for illustrations should be typed separately, their positions indicated in the text. All pages should be num- bered 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; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including caption); the reduction or enlargement required should be indicated (and preferably uniform); orig- inals 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 caption; 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. If Letraset is used authors are requested to use Helvetica-style letter- ing, if possible. 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, 19696; 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, 1969b) 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 (according to the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number 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 la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140. FiscHER, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie expérimentale et générale 74 (33): 627-634. Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and Magazine of Natural History (13) 2 (17): 309-320. Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of the Bingham Oceanographic Collection, Yale University 17 (4): 1-51. THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHuLTzE, L. Zoologische und anthro- pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika ausgeftihrt in den Jahren 1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 99 Band February 1990 Februarie Part 1 Deel NX és > E> A NEW AUSTROMEGABALANUS (CIRREPEDIA, BALANIDAE) FROM THE PLIOCENE OF NAMAQUALAND, CAPE PROVINCE, SOUTH AFRICA By JOHN PETHER 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 i ACES, 8) ED, 2-5, 8 Bntl,), SOS, 5, 7D), AG aor), MODY, &, OD, My, MOS), 111-2, 5, 7, t.-p.i.), 1402), 154-5), 240), 27, 31G23)s 326) 33s 36) aa Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 106 7 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press. Pty.) Wid: Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap A NEW AUSTROMEGABALANUS (CIRREPEDIA, BALANIDAE) FROM THE PLIOCENE OF NAMAQUALAND, CARE PROVINCE, SOUITE APRICA By JOHN PETHER Department of Cenozoic Palaeontology, South African Museum, Cape Town (With 3 figures and 1 table) [MS accepted 14 November 1988] ABSTRACT A preliminary investigation of fossil barnacles from shallow marine coastal plain deposits exposed at Hondeklip, Namaqualand coast, South Africa, revealed the presence of a new species of austral megabalanine, Austromegabalanus (Notomegabalanus) kensleyi sp. nov., in an Upper Pliocene formation informally termed the 50-metre Package. This is the first extinct barnacle described from South Africa and, with the possibility of additional new species being recognized in future work, the biostratigraphic utility of barnacle assemblages for nearshore deposits of the South African margin is promising. CONTENTS PAGE HERI ROCIO TON oak ws (eurrsai aes ON NEL aleoN et ace Aree OneCare tt, a aes Sins oe 1 i DESSTIOUMIOMN, & aso. © ain een ala note cigt RC ok feat aie nace ae eee Ta mare Screaes Z | DISCUSSIOMs. J'5 h 0.5\6 6:5 ons iso.6 aro enon Ee oe area RRS otro c/a centr ae ere 7 GEOlOPICAESE UM Eee Err ett aI ioc Mee a Aare gens ercrmee eset Siete snc i cee 10 | PRS OSTROW TNE I Bere ckoio HIE eee Cet, ceo oats Sua rons Cena cera il Conichucimesne nails wien Oe ee ao eee ne a we eed LS Nau arene eee 11 ANCHO ECUSMNETUS: ag eee Coats emery guts oA Sia aS Ie org ae ae & 12 IN@UEREMCGOS ‘ose, Sosa 9. been MOTE cee Ree asec a rch CTE tho cid herb Gopeacein is 1 INTRODUCTION Shallow marine deposits, ranging in age from Early Pliocene to Late Pleistocene, underlie the terrestrial cover sands of the coastal plain along the Atlantic margin of South Africa. Mining of diamonds in these marine deposits adjacent to the Namaqualand coast creates temporary exposures that provide Opportunities to examine the sedimentology and sample the fossil content. Hitherto the fossil molluscs have received attention, the most recent work being that of Kensley & Pether (1986), which presents a list of taxa and discusses the composition and biogeographic affinities of the fauna. This paper describes the first new species of barnacle recognized from the coastal plain marine deposits of South Africa. The specimens were obtained from the diamond mine at Hondeklip Bay (Fig. 1). The stratigraphy, sedimentary geometry, depositional environ- ments, and evidence for the age of the deposits in this area are summarized in Pether (1986). Ann. S. Afr. Mus. 99 (1), 1990: 1-13, 3 figs, 1 table. D. ANNALS OF THE SOUTH AFRICAN MUSEUM SOUTH ATLANTIC Fig. 1. Provenance of Austromegabalanus (Notomegabalanus) kensleyi sp. nov. Broken lines in left insert show inland (subsurface) limits (transgressive maxima) of main marine formations underlying the coastal plain. Informal formation labels record elevation of transgressive maximum, e.g. 50-metre Package. Asterisks indicate localities where the new species was obtained. DESCRIPTION Family Balanidae Leach, 1817 Subfamily Megabalaninae Newman, 1979 Genus Austromegabalanus Newman, 1979 Subgenus Notomegabalanus Newman, 1979 Austromegabalanus (Notomegabalanus) kensleyi sp. nov. Figs 2-3 Diagnosis Relatively small, smooth shell with single row of parietal pores and porous radii; no transverse septa in either. Thick scutum of high relief, no exterior radial sculpture; small, triangular, deeply excavated, lateral depressor pit, deep adductor muscle pit, no definite adductor ridge. Tergum not beaked and with short, broad spur at less than two-thirds its own width from basi-scutal angle; broad, very shallow spur furrow; no definite depressor crests. Distribution Upper Pliocene of the Namaqualand coastal plain, west coast of South Africa. AUSTROMEGABALANUS FROM THE PLIOCENE OF SOUTH AFRICA 3 Material Holotype. SAM—PQ—-HB337a, one complete shell with opercula removed from cluster on boulder, Hondeklip Zone 12A (30°19'29"S 17°18'21"E). Paratypes. SAM—PQ-HB337b-—df, 110 complete shells from same cluster as holotype, many with opercula and bases. SAM—PQ-HB 1261, cluster of approxi- mately 30 shells on Striostrea margaritacea (Lamarck, 1819) valve, one of which contained an articulating opercular pair, Hondeklip Zone 4A. Additional material. SAM—PQ-HB321, unattached cluster of 9 shells with- out opercula, Hondeklip Zone 12A. SAM—POQ-HB180, disarticulated shells and scuta, Hondeklip Zone 12A. SAM—POQ-AV984, 4 shells without opercula, Avon- tuur A, T2/3. SAM—PQ—AV1067, 3 shells without opercula, Avontuur A, T2/3. Measurements Holotype. Height 4,2 mm, carino-rostral diameter 11,1 mm, scutum occlu- dent margin 3,7 mm, tergum scutal margin 2,6 mm. Largest shell. SAM—PQ-AV984a, height 6,3 mm, diameter 14,7 mm. Largest scutum. SAM—PQ-HB180, occludent margin 6,2 mm. Description Shell relatively small (~1 cm basal diameter). Specimens from smooth, un- crowded substrate (Fig. 2C) are low-conical with regular and smooth exteriors, have pentagonal to pointed-ovate, moderately toothed orifices about one-third carino-rostral diameter, and thin, flat bases. Those from uneven and crowded substrata (Fig. 3C) are more globulo-conical with relatively larger orifices, rugose exteriors and have cup-shaped bases (up to | cm deep), often with irregularly cellular underlayers. Parietes with single row of open longitudinal pores, ovoid in section near apex, subrectangular near basis; internally ribbed, ribs usually on lower portion but may extend almost to sheath, short secondary ribs sometimes on larger speci- mens; basal denticles conspicuous with correspondingly small tapered ridges on lower ribs; sheath about half the length of plate, lower margin usually separated from wall by moderate to very shallow concavity. Many specimens still have a rose-pink hue colouring the sheath. Darker radial lines on exteriors of parietes correspond to the underlying septa. Radii inset from parietes, transversely porous although upper pores may be infilled; sutural edges regularly and densely septate with denticles on lower edges of septa (Fig. 3D), articulating surfaces on adjacent parietes with corresponding low, upward denticulations. Alae thin and broad, crenate sutural edges sometimes visible. In unabraded specimens, such as the SAM-—PQ-HB337, the radii summits are obliquely truncated and the alae have steeply oblique summits, usually projecting above the opposite sloping radu. In more abraded specimens the summits are approximately parallel to the base. Basis radially porous, with shallow furrows corresponding to the underlying pores. 4 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 2. Holotype of Austromegabalanus (Notomegabalanus) kensleyi sp. nov., SAM—PQ- HB337a. A. Interior and exterior views of scutum. B. Interior and exterior views of tergum. C. Apertural and lateral views of shell. Scales = 1 mm. AUSTROMEGABALANUS FROM THE PLIOCENE OF SOUTH AFRICA 5 Fig. 3. Austromegabalanus (Notomegabalanus) kensleyi sp. nov. A-B. Large paired scutum and tergum from paratypes, SAM—PQ-HB1261. Scale = 1mm. C. Shell cluster from same paratypes, set on Striostrea margaritacea valve. Scale = 5mm. D. Sutural edge of radius showing austral megabalanine denticulation. Scale = 1 mm. 6 ANNALS OF THE SOUTH AFRICAN MUSEUM Scutum thick, of high relief; concave exterior with prominent growth ridges, every alternate one denticulating occludent margin; no radial sculpture; basal margin convex and about two-thirds length of occludent margin; tergal margin about three-quarters length of occludent margin and on exterior raised as a low ridge from apex to basi-tergal angle before being strongly but quite narrowly reflexed, the ridge enhancing the concavity of the exterior; a shallow groove is often present along the reflexed tergal segment. Articular ridge high, about four-fifths of tergal margin in length and reflexed over relatively narrow articular furrow; articular ridge thins basally, ending in a closely rounded-off corner above concave terminal edge; outer face shallowly concave and faint grooves are generally visible. Adductor muscle scar large, ovate and conspicuously deep, usually being visible without magnification, even in small scuta. No obvious adductor ridge but a low, rounded swelling is usually present beside adductor tergal margin and laps on to the basal flank of articular ridge. Lateral depressor pit small, narrowly triangular and deeply excavated, usually with slightly overhanging sides. Occludent margin interiorly reflexed, producing marginal ridge that basally overhangs rostral depressor insertion. Tergum carinal and basal margins convex, scutal margin straight; exterior with low growth ridges, more pronounced on scutal portion; no radial sculpture. Spur furrow broad, open and very shallow to flush; margins slightly incised. Exterior carinal portion gently convex, exterior scutal portion usually slightly outwardly deflected. Articular margin quite narrowly reflexed at approximately right angle, rugose, forming ridge along scutal margin of articular furrow; articular furrow broad, moderately deep; articular ridge relatively thin, moderately high and partially overhanging furrow near apex. Apex not beaked, area of transverse ridges small; interior of valve below apex with several irregular grooves; carinal depressor crests effectively absent but faint grooving sometimes visible. Spur short and wide, approximately as long as broad; occupying from quarter to third of basal margin, situated about half to two-thirds own width from basi-scutal angle; basal margin segments slope towards spur, basi-scutal margin concave. Scutal end of spur closely rounded off, carinal corner very widely rounded off and almost a continuous curve from basal margin to scutal end. On interior of valve spur is only little raised and has a flat to slightly concave inner face. On the largest tergum available (Fig. 3B) the spur is seen to be relatively - more developed than is the case in smaller specimens and the upper scutal margin is eroded. Etymology The species is named for Dr Brian Kensley of the Smithsonian Institution, Washington, D.C., in recognition of his contributions to the knowledge of the southern African fossil molluscan fauna. AUSTROMEGABALANUS FROM THE PLIOCENE OF SOUTH AFRICA 7 DISCUSSION The porous radii establish the Namaqualand specimens as belonging to a megabalanine genus. The classification system followed here is that of Newman (1979), with the modifications proposed by Buckeridge (1983). In the former work, Balanidae with porous radii were allocated to a new subfamily Megabalani- nae. Taxa possessing radii with secondary denticles on the lower sides of primary denticles only were recognized as pan-austral Southern Hemisphere endemics and assigned to two new genera, Austromegabalanus and Notomegabalanus. The former genus 1s distinguished by beaked terga and closed spur furrows, the latter by non-beaked terga and open furrows. However, Buckeridge (1983), in his monograph on the fossil barnacles of Australia and New Zealand, expressed the opinion that the differences in tergal morphology, with respect to spur furrow and beak, between Austromegabalanus and Notomegabalanus are not sufficiently strong to justify full generic separation. He thus proposed that Notomegabalanus be regarded as a subgenus of Austromegabalanus. Furthermore, the taxa with porous outer parietal laminae were removed from Notomegabalanus and placed in a new genus Fosterella. With secondary denticles on the lower sides of the primary radii denticles only, this species from Namaqualand is confirmed as an austral megabalanine. The open spur furrow and lack of a strongly beaked tergum place it in the subgenus Notomegabalanus. Listed in Table 1 are the fossil and extant species of Notomegabalanus, together with their approximate geological ages. Austromegabalanus s.s. and Fosterella have been included to provide a complete list of these related austral genera. The austral megabalanines have not always been confined to the Southern Hemisphere. Fossil occurrences of Notomegabalanus in Tertiary deposits on both the Pacific and Atlantic coasts of North America have been recorded by Zullo & Guruswami-Naidu (1982) and Zullo (1986). As discussed by Newman & Foster (1987), this distribution suggests that Notomegabalanus has had a Tethyan and an amphitropical history, the present austral endemism being due to Northern Hemi- sphere extinction and Neogene dispersal via the West Wind Drift. Notomega- balanus is considered more generalized than Austromegabalanus s.s., whilst Megabalanus s.s. is the most highly evolved of the megabalanines (Newman 1979). The presence of an undescribed species of Notomegabalanus in the Oligocene of Oregon (Zullo 1986) is consistent with such a relationship. The Neogene radiation of these related taxa and the extinctions due to Pleistocene glaciations (which particularly affected Fosterella and Notomegabalanus), are evident in Table 1. Austromegabalanus (Notomegabalanus) kensleyi is rendered distinct from all other species of the subgenus by its short, broad, rounded tergal spur at less than its own width from the basi-scutal angle, especially if this character is considered simultaneously with the very shallow spur furrow and subdued expression of the spur ridge on the valve interior. The comparison of A. (N.) kensleyi with the austral species of Notomegabala- nus indicates that the deep, narrowly triangular lateral depressor pit distinguishes (6L61 ‘UBWIMON) WIUUaY v]Ja1a1S0.4 (6L61 ‘URUIMAN) Saplojvjnqn] vjja421S0,7 (PZ6T ‘SIOUUM) SNIDINGNY vIJa42480.] EQ6I ‘OSPlIoyoNg SisuawomyyoYs D]Ja121S0,7 (9161 ‘Alqstiq) 0j0013]v (snuvjoqvsawojon) snuvjoqvsawuoysny “Aou ‘ds 1dajsuay (SnuvjJDGvsaOJON!) Snub DGvsawo4sny (S88T ‘ouTlLy) yWlaqduo (snuvjoqvsawmojon) snuvjoqvsawojsny ANNALS OF THE SOUTH AFRICAN MUSEUM (pSQI ‘UIMIed) SmoJap snoJap (SNUB]DgvsaUOJON) SnuDvjOGusawousny CQ6I ‘Osplioyong sisuapjAduv snsodap (Snuvjoqvsawulojon’) snuvjoqvsawousny ER61 ‘OSplioyong smodaponu (snuvjvqvsawojon’) snuvjoqvsawmousny (€961 ‘YSoYy 29 JoIUed) avyinbs (snuvjvqvsawojoNZ ) snuvjoqvsawmoasny OQ6I ‘OLINZ, snpidaj (snuvjvqvsauojoN) snuvjoqvsawousny (961 ‘SssSOy) SnnbigGo (snuvjvqvsawojoN) snuvjoqvsawousny (6961 ‘OTINZ) 2u0sjIM (snuvjvgvsawojoN!) snuvjoqvsawojsny 7861 Nplen -IWPMSNIND 2 O]NZ, snvsadsui (snuvjvqvsawuojon; ) snuvjoqvsawoysny 9R6[ ‘OIINZ ‘ds (snuvjoqvsawojoN) snuvjogvsawo“sny 6L6[ ‘URWIMON 101/NZ (SnuBjJDGUsaWUOJSnY) Snub] vGvsaWOosNny (£96 ‘UdSIIOAIS 2 SINYIJOH]) apjos1 (snuvjvqvsawousny) snuvjvqvsawousnpy (O8LT ‘UITOWID) snoupuydd (snuvjoqvsawoysny) snunjvgvsawousny (Z8LT ‘euljoy) snovpisd (snuvjoqvsawmousny) snuvjogvsawoysny L961 ‘70 Ja jourIeD sisuaoosid (snuvjvqvsawosny) snuvjoqvsawuojsny (QIQT ‘yorewey) suaosausiu (snuvjoqvsawoysny) snuvjvgvsawousny INV WVS ZN omega IN IVS ELAS ZN ZN ZN ZN NI VNad VNa VNM VNM VNM IOS IOS EAS WVS WVS Sav SNV FR6[ ‘osplroyong SISUDUOJIIA (. snuvjvqvsawmo.jsny) Snuvjvqvsawojsny “‘puelsy URDDO UIOYINOS = [OS ‘eonoIeWYy = LNV :BOMOUTY YINOS = WIVS ‘Boy YINOS = FWS ‘elfelisny = SQV ‘puejeoz MON = ZN ‘eIpuy = NI ‘eovoury YON ulojseq = VWNA ‘eoLoury YON UI9IS9A\ = VWNAM ‘90U91IND90 sIOYydsIWIO}] uIOYINOS = S {s0UDTIND00 sIOYdSIWIDFY UIOUWION = N suoNeoo] jeo1ydeiso9yH + -vjJaajso.y pue snuvjoqvsawmogsny JO Saidods jue}xo pur |Isso4 NNNNNNNnLZZ LAZLZAZNNNnNUNNHnHVHDHNHnN 1 ATaVvib AUSTROMEGABALANUS FROM THE PLIOCENE OF SOUTH AFRICA 9 the scutum of the new species from other representatives of the subgenus. The lack of a definite adductor ridge separates the new species from all other austral Notomegebalanus species except A. (N.) algicola (Pilsbry, 1916) which, however, has a very broad scutum. The absence of scutal radial sculpture is a feature in common with A. (N.) algicola and A. (N.) campbelli (Filhol, 1885). The basal end of the articular ridge of the new species is not pointed or dependant, as is the tendency in other austral species of Notomegabalanus. The lack of a tergal beak distinguishes the new species from the weakly beaked Notomegabalanus species of New Zealand. The shells of A. (N.) kensleyi are slightly larger than those of A. (N.) algicola but smaller than the New Zealand representatives of the subgenus, the ribbed A. (N.) campbelli being closest in dimensions. Whereas the latter and A. (N.) decorus (Darwin, 1854) have transverse septa in the parietal pores, the new species does not, a feature in common with A. (N.) algicola and evidently A. (N.) miodecorus Buckeridge, 1983. In the Northern Hemisphere, the extant Balanus (Megabalanus) squillae Daniel & Ghosh, 1963, possesses austral megabalanine radii. This rather cryptic and very small species from the coast of Madras is parasitic on a stomatopod, Squilla sp. It is problematic in possessing a tergum that is non-beaked but has a closed spur furrow. On the basis of the overall resemblance of the opercula to those of A. (N.) algicola, this species has been tentatively placed in Notomega- balanus by Henry & McLaughlin, 1986. The small broad scutum with an adductor ridge further distinguishes it from the new species. Austromegabalanus (N.) wilsoni (Zullo, 1969) from the Upper Pliocene of California also possesses an unbeaked tergum with a closed spur furrow and similarly has been placed in Notomegabalanus on the basis of its overall resemblance to A. (N.) algicola. Additional discoveries of such species that do not strictly conform to the definition of Notomegabalanus might render a redefinition of the subgenus or erection of another subgenus advisable. The scutum of A. (N.) wilsoni resembles that of A. (N.) kensleyi in respect of the small, deep, lateral depressor pit and lack of an adductor ridge, but differs in respect of the straight basal margin, radially aligned pits on the exterior of some Specimens and overall larger size. Austromegabalanus (N.) lepidus Zullo, 1986, from the Pliocene of the Western North Atlantic is a species of comparable size to that of A. (N.) kensleyi but the scutum, although also lacking an adductor ridge, is thin with a shallow lateral depressor pit and occasionally has exterior radial Striae. In addition to the deep spur furrow, the tergum of A. (N.) lepidus differs from the new species in the strongly concave scutal margin, presence of distinct depressor crests, and radially striate exterior scutal portion. The opercula of A. (?N.) obliquus (Ross, 1964) (Pliocene, Western North Atlantic) have not been recovered but that species is larger than the new species and has externally ribbed parietes with deeply inset radii. The opercular valves of A. (?N.) insperatus Zullo & Guruswami-Naidu, 1982 (Late Miocene, California), are also unknown but the shell is also larger than A. (N.) kensleyi and has prominent external ribs. i0 ANNALS OF THE SOUTH AFRICAN MUSEUM It is noteworthy that of all the representatives of Notomegabalanus, the opercula of the new species are overall most similar to those of the extinct genus Fosterella Buckeridge, 1983. This applies particularly to the deep, small, scutal lateral depressor pit (this feature having been considered a defining character of Fosterella) and the short tergal spur and shallow spur furrow. In general appearance the opercula of Fosterella tubulatoides (Newman, 1979) from the Late Pleistocene of the continental shelf off Tierra del Fuego are closest. However, several other features remain to distinguish the opercula of species of Fosterella, such as the presence of weak adductor ridges, pointed basal ends of articular ridges, and strongly developed tergal depressor crests (the latter not present for F. hennigi (Newman, 1979) but that species has very distinctive elongate opercula). Furthermore, the shells of Fosterella are all strongly ribbed and, in addition to the defining character of multitubate parietes, have transverse septa in the parietal pores. GEOLOGICAL SETTING Austromegabalanus (Notomegabalanus) kensleyi is abundantly represented in a formation informally called the 50-metre Package (Pether 1986). This formation is comprised of gravels and fine sands laid down during regressive shoreline progradation from a transgressive maximum near 50 m above present sea-level. Depositional environments include the nearshore shelf (locally preserved only), lower shoreface, upper shoreface, foreshore, and back-barrier settings. The age of the 50-m Package is not yet unequivocally determined but is constrained by the presence of a remanie vertebrate assemblage of Mio-Pliocene aspect at the base (Q. B. Hendey pers. comm.) and the occurrence of Equus near the top, the latter probably representing stratigraphic leakage from a closely overlying erosion surface. Furthermore, the highest elevation marine deposits (90-m Package, Fig. 1) inland of the 50-m Package have been correlated with the Early Pliocene Varswater Formation at Langebaanweg (Hendey 1981). The 50-m Package is thus considered to be Middle to Late Pliocene in age. Austromegabalanus (Notomegabalanus) kensleyi has not been found in barnacle coquina from the 30-m Package, the regressive wedge overlying the transgressive truncation of the 50-m Package. Thus it appears to have become extinct on the Namaqualand coast around the Plio-Pleistocene. The lower limits of the new species are not yet determined, due to lack of material, but it is considered quite likely to extend into the Early Pliocene 90-m Package. At least one other species of barnacle is present in the 50-m Package. This is a non-megabalanine but opercula permitting determination have not yet been recovered. In the barnacle coquina of the subsequent Early Pleistocene 30-m Package, disassociated opercula indicate the presence of three barnacle species, whereas parietal fragments indicate that Austromegabalanus, Megabalanus s.s. and Balanus species occur. The opercular material is too abraded for the purposes of unequivocal specific determination but one scutum type and one tergum type AUSTROMEGABALANUS FROM THE PLIOCENE OF SOUTH AFRICA 11 (probably from the same species) appear sufficiently unique to suggest a new species. Other fragmentary opercula suggest that the Austromegabalanus sp. indicated by parietes may by Austromegabalanus s.s. It is hoped that continued fieldwork will produce better-preserved and associated examples of these taxa. PALAEOENVIRONMENT The types of A. (N.) kensleyi were found in situ on a boulder resting on the gneissic bedrock. The palaeoenvironment of the type area, reconstructed from the overlying sediments, was the inner, shallow portion of a protected embayment. The boulder was at shallow subtidal palaeodepth (<—5 m) when the live barnacles were covered by fine sands—probably during or immediately subsequent to a storm. Disarticulated barnacle fragments are a major component of coarse, basal lower shoreface deposits in the deeper portion of the bay and, within the fine sandy portion of the lower shoreface bay infill, the new species 1s commonly found encrusting Striostrea margaritacea (Lamarck) valves. These oysters were swept from the lower intertidal to shallow subtidal margin of the bay during storms and deposited in deeper water. Lagoonal deposits within which barnacle coquina lenses are present represent the final infilling of the embayment. These lenses are interpreted as coarser lags accumulated along the very shallow subtidal bases of tidal creeks within the lagoonal system, with barnacle fragments likely to have been transported by flood tides from the proximal lagoon and adjacent coast. In exposed coastal settings, detached clusters of barnacles are found in shelly lenses within the storm-deposited lower shoreface facies. Disarticulated frag- ments are also an important component of the cross-stratified, coarse sandy upper shoreface facies. In all the above instances the dominant barnacle is referable to the new species and its preferred environment was evidently the intertidal to shallow subtidal of normal marine salinity. The presence of warm-water West African and southern African east-coast molluscs in the 50-m Package indicates that the sea temperature was warmer than the present interglacial, at least seasonally. However, upwelling influences were present, as is attested by the occurrence of authigenic phosphorite rinds within nearshore shelf and basal lower shoreface facies. It is likely that changes in faunal structure associated with cooling, sea-level fall, and altered shelf currents account for the extinction of A. (N.) kensleyi. CONCLUDING REMARKS It is appropriate and not unexpected that the first extinct barnacle to be described from the coastal plain deposits of South Africa is a species of Notomegabalanus. Newman (1979) remarked that the austral region must have been richer in austral megabalanines during the Pliocene than it is today. The subsequent discovery of Fosterella chathamensis Buckeridge, 1983, from the i, ANNALS OF THE SOUTH AFRICAN MUSEUM Pliocene of the Chatham Islands, Austromegabalanus (A.) piscoensis Carriol et al., 1987, from the Pliocene of Peru, and the present species bears out that supposition that, with inclusion of the North American fossil species, may now be amplified to include both hemispheres during the Neogene. Another species of Austromegabalanus is represented in the Early Pleisto- cene 30-m Package and may be Austromegabalanus s.s., but the abraded state of the available opercula preclude unequivocal specific identification at present. Significantly, material conclusively attributable to the extant South African Austromegabalanus species has not been recovered from the Upper Pliocene and Lower Pleistocene deposits. This suggests that A. (A.) cylindricus (Gmelin, 1780) does not have a range extending from the Pliocene, in contrast to A. (A.) psitt- acus (Molina, 1782) from South America. However, the former species is well represented in Last Interglacial beach deposits at Milnerton near Cape Town (Kensley 1985) and at other Late Pleistocene outcrops in the south-western Cape. Austromegabalanus (N.) algicola is also present in Last Interglacial deposits at Milnerton. This paper represents the results of an initial reconnaissance of fossil barnacles from the Hondeklip study area. The recognition of a new species of barnacle from Pliocene coastal plain deposits, with possibilities of others in the future and as yet undetermined lower limits for extant South African Austromegabalanus species, holds the promise that barnacles will be of value in the correlation of the Late Tertiary and Quaternary shallow marine sequences of the southern African margin. Their calcitic composition favours preservation and, together with their abundance in shallow marine deposits, barnacle assemblages should emerge as useful biostratigraphic markers in the local coastal stratigraphy. ACKNOWLEDGEMENTS I am indebted to Mr F. Hoffman of Transhex Group Ltd for access to the Hondeklip mine exposures and the mine manager and geologists for their hospitality; to Dr Q. B. Hendey of the South African Museum and Dr J. Rogers of the University of Cape Town, for their advice and support; to Marilyn Pether for photographic assistance; to the referees for their constructive comments, and to the CSIR and the South African Museum for financial assistance. REFERENCES BUCKERIDGE, J. S. 1983. Fossil barnacles (Cirrepedia: Thoracica) of New Zealand and Australia. Palaeontological Bulletin. Geological Survey, New Zealand 50: 1-151. CaRRIOL, R. P., DE Muizon, C. & SECRETAN, S. 1987. Les Crustacés (Cirrepedia et Decapoda) du Neogéne de la cdte Péruvienne. Annales de paléontologie 73 (3): 137-164. DanigEL, A. & Guosn, A. 1963. A new cirrepede of the genus Megabalanus from the stomatopod (Squilla sp.) from Madras. Annals and Magazine of Natural History (13) 6 (68): 477-479. HeENpEY, Q. B.1981. Geological succession at Langebaanweg, Cape Province, and global events of the Late Tertiary. South African Journal of Science 77 (1): 33-38. AUSTROMEGABALANUS FROM THE PLIOCENE OF SOUTH AFRICA 13 Henry, D. P. & McLauGuiin, P. A. 1986. The Recent species of Megabalanus (Cirrepedia: Balanomorpha) with special emphasis on Balanus tintinnabulum (Linnaeus) sensu lato. Zoologische verhandelingen 235: 1-19. KENSLEY, B. 1985. The faunal deposits of a Late Pleistocene raised beach at Milnerton, Cape Province, South Africa. Annals of the South African Museum 95 (2): 111-112. KENSLEY, B. & PETHER, J. 1986. Late Tertiary and Early Quaternary fossil mollusca of the Hondeklip area, Cape Province, South Africa. Annals of the South African Museum 97 (6): 141-225. Newman, W. A. 1979. On the biogeography of balanomorph barnacles of the Southern Ocean including new balanid taxa; a subfamily, two genera and three species. In: Proceedings of the International Symposium on Marine Biogeography and Evolution in the Southern Hemisphere, Auckland, New Zealand, July 1978. Information series. Department of Scientific and Industrial Research, New Zealand 137 (1): 279-306. Newman, W. A. & Foster, B. A. 1987. Southern Hemisphere endemism among the barnacles: explained in part by extinction of northern members of amphitropical taxa? Bulletin of Marine Science 41 (2): 361-377. PETHER, J. 1986. Late Tertiary and Early Quaternary marine deposits of the Namaqualand coast, Cape Province: new perspectives. South African Journal of Science 82 (9): 464-470. Ross, A. R. 1964. Cirrepedia from the Yorktown Formation (Miocene) of Virginia. Journal of Palaeontology 38 (3): 484-491. ZuLLo, V. A. 1969. Thoracic Cirrepedia of the San Diego Formation, San Diego County, California. Contribution in Science. Los Angeles County Museum 159: 1-25. ZULLO, V. A. 1986. Pliocene barnacles (Cirrepedia, Thoracica) from South Carolina, with a key to Pliocene balanoids of the Eastern United States. South Carolina Geology 29 (2): 1-18. ZuLLoO, V. A. & Guruswami-Naibu, R. B. 1982. Late Miocene balanid Cirrepedia from the basal Wilson Ranch beds (‘Merced’ Formation), Sonoma County, Northern California. Proceedings of the California Academy of Sciences 42 (2): 525-535. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu- larly 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. nov., sp. nov., comb. nov., syn. nov., 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 trans- ferred 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 (fig. 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 specimens 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, descrip- tion 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 Eliza- beth (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. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text e.8. . 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 ‘sebedth 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 Punctuation should be loose, omitting all not strictly necessary Reference to the author should preferably 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. The generic name should not be abbreviated at the beginning of a sentence or paragraph. Name 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. JOHN PETHER A NEW AUSTROMEGABALANUS (CIRREPEDIA, BALANIDAE) _ FROM THE PLIOCENE OF NAMAQUALAND, CAPE PROVINCE, SOUTH AFRICA 4 | VOLUME 99 PART 2 FEBRUARY 1990 . ISSN 0303-2515 if the \ ®& & t ANNALS OF THE SOUTH AFRICAN MUSEUM \CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (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) 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 fee are numerous. 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for illustrations should be typed separately, their positions indicated in the text. All pages should be num- bered 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; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including caption); the reduction or enlargement required should be indicated (and preferably uniform); orig- inals 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 caption; 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. If Letraset is used authors are requested to use Helvetica-style letter- ing, if possible. 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 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, 1969b) 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 (according to the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140. FiscHer, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie expérimentale et générale 74 (33): 627-634. KouNn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and Magazine of Natural History (13) 2 (17): 309-320. Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of the Bingham Oceanographic Collection, Yale University 17 (4): 1-51. THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro- pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren 1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 99 Band February 1990 Februarie Part 2 Deel CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA. A KOLOCERAS (CEPHALOPODA, AMMONOIDEA) FROM THE MZINENE FORMATION (ALBIAN) 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 i, AGES, 5-9), SUD, 25, 8, taps.) SCIR8, 5, 9-9), 6, t-sa..)), 1), 8, 902, 7), OG), 1112) 5. 7, t =p.ib), 14122), 15425) 24) 27, ides) 2G) san eeOmeeoan) Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 107 5 Printed in South Africa by In Suid-Afrika gedruk deur ithe Rustica Bressh sPiyaaslerde Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap C465 CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA. A KOLOCERAS (CEPHALOPODA, AMMONOIDEA) FROM THE MZINENE FORMATION (ALBIAN) By HERBERT CHRISTIAN KLINGER Department of Invertebrate Palaeontology, South African Museum, Cape Town & WILLIAM JAMES KENNEDY Geological Collections, University Museum, Oxford (With 2 figures) [MS accepted 22 September 1989] ABSTRACT The monospecific, micromorphic ammonite genus Koloceras (of the subfamily Aconecera- tinae), thus far only known from the Upper Albian of Estancia La Vega in Santa Cruz Province, Argentina, is recorded for the first time from the Upper Albian of Zululand. CONTENTS PAGE FICO CIC HO Maree rename ter re se eeten trans Mien ce ohm Sakae My aca oie a aae Bry lic) 4 15 SVSCCHIAUI CG CSCMIP LOM syrre ans ale oe san oe ie sk ges We Mee era ce tlw BOs) Bee tv INGO W COO CHICIOS hua dimcics: cesta rad C eekoueine ad Wat: aime ese Weber aa ae, BS le 20 IR ENIEIRSTD COG ig Bee aes Ele oe cs CR Le Oe eee Rr yt SIE eo ais eae ZAM INTRODUCTION On three separate occasions, micromorph, platyconic ammonites have been recorded from the Upper Albian of the Southern Hemisphere as Boris- siakoceras—a genus generally regarded as being of Middle Cenomanian to Turo- nian age, but which may occur as early as the Lower Cenomanian and persist up to the Coniacian (see e.g. Kennedy & Klinger 1979: 115-116). These records are: 1. Borissiakoceras sp. Nullo et al., 1981 (pl. 2 (fig. 9a—c)) from Patagonia, Argentina. 2. Falciferella breadeni Brunnschweiler, 1959 (p. 15, pl. 1 (figs 5-6)) and Falciferella reymenti Brunnschweiler, 1959 (p. 16, pl. 1 (figs 7-8)), both referred to Borissiakoceras by Cooper & Kennedy (1979: 297) from South Australia. 3. Borissiakoceras sp. nov. ? aff. reymenti Cooper & Kennedy (1979: 297, fig. 31H-I) from Porto Amboim, Angola. 5) Ann. S. Afr. Mus. 99 (2), 1990: 15-21, 2 figs. 16 ANNALS OF THE SOUTH AFRICAN MUSEUM Cooper & Kennedy (1979: 296-7) regarded the Angolan and Australian material as bridging the phylogenetic gap in the family Binneyitidae between the last appearance of the genus Falciferella Casey, 1954 (type species F. milbournei Casey, 1954: 271, pl. 7 (figs 1-5), text-fig. 3), in the Middle Albian and the first occurrence of Borissiakoceras in the Middle Cenomanian. Recent data suggest that this view seems to be too simplistic. The Australian material was recently (McNamara 1985) referred to a new genus Naramoceras (type species Falciferella breadeni Brunnschweiler, 1959: 15, pl. 1 (figs 5—6)) in the subfamily Aconeceratinae. The Patagonian material was also referred to a new genus (Riccardi et al. 1987: 167), Koloceras (type species Koloceras talenkanum Riccardi et al., 1987: 169, pl. 14 (figs 1-17b), text-figs 42— 46), also of the subfamily Aconeceratinae. Both Koloceras and Naramoceras have distinctive, little-incised suture lines with a trifid lateral lobe (L), which clearly separate them from Borissiakoceras. Koloceras in turn differs from Naramoceras in having a trifid second lateral (U,) lobe, rather than bifid as in the latter. The erection of these two (apparently endemic) genera casts doubt on the identity of the Angolan specimen. Is it indeed the only true Upper Albian Borissiakoceras, or might it belong to Koloceras or Naramoceras? Unfortunately, we have not seen the suture lines of the Angolan specimen to be able to comment further on this matter. In Patagonia and South Australia, Koloceras and Naramoceras respectively occur in association with a diverse and distinct heteromorph ammonite fauna belonging to the subfamily Labeceratinae (see e.g. Whitehouse 1926; McNamara 1978; Aguirre Urreta & Riccardi 1988). The subfamily Labeceratinae has a typi- cally South Gondwanid distribution, and is locally very common in Zululand and southern Mozambique (see Spath 1925; Klinger 1976; Klinger 1989). When Naramoceras and Koloceras were first described, our immediate reaction was to see if we had any representatives of either genus in our collections from Labeceras-bearing localities. We did not find any. Recently, in the course of routine curation of the Van Hoepen collection, housed in the South African Museum, a small ammonite was brought to the attention of one of us (H. C. Klinger), who identified it as belonging to the genus Koloceras; later a second specimen was found, which is tentatively referred to that genus. We here describe the first record of the genus Koloceras from South Africa. We cannot be certain whether this is only a fortuitous discovery of a post-mortem drifted occurrence of this apparently endemic, monospecific Patagonian genus, or whether it indeed represents part of a population as yet not localized. Its discovery should encourage the search for further material of both Koloceras and Naramoceras in the Southern Hemisphere, to clarify the interrelationship between these two genera and Borissiakoceras of the subfamily Binneyitinae. CRETACEOUS FAUNAS FROM SOUTH AFRICA NY SYSTEMATIC DESCRIPTION Superfamily HAPLOCERATACEAE Zittel, 1884 Family Oppelidae Bonarelli, 1894 Subfamily Aconeceratinae Spath, 1923 Genus Koloceras Riccardi, Aguirre Urreta & Medina, 1987 Koloceras talenkanum Riccardi, Aguirre Urreta & Medina, 1987 Figs 1-2 Borissiakoceras sp. Nullo et al., 1981: 200, pl. 2 (fig. 9a—c). ?Falciferella sp. Medina & Rinaldi, 1986: 16, pl. 2 (figs 3-5), pl. 4 (figs 6-9). Koloceras talenkanum Riccardi et al., 1987: 169, pl. 14 (figs 1-17b), text-figs 42—46. Type Holotype is the specimen (MLP 19981) figured by Riccardi et al. (1987, pl. 14 (fig. 15a—b)), an adult macroconch from the upper part of the Rio Mayer For- mation, Estancia La Vega, Santa Cruz Province, Argentina, HOMIES in the col- lections of the Museo de Ciencias Naturales, La Plata. Material NMB D380 (Van Hoepen Collection) from locality 51 and NMB D872 (Van Hoepen Collection) from locality 55, both from the Mzinene Formation, Albian IV—V, Zululand. (For precise locality data see Kennedy & Klinger 1975.) Dimensions Dimensions of specimens are in millimetres; abbreviations are as follows: D = diameter, Wb = whorl breadth, Wh = whorl height, U = umbilical diameter. Figures in parentheses are dimensions as a percentage of total diameter. Specimen i) Wb Wh Wb/Wh U D380 8,6 lis (209) 455) (6,9) 0,4 Los (192) D872 139 Les (3,1) 3,0 (4S,1) 0,51 135 (0,9) Description D380 (Figs 1E—-F, 2A) is a microconch; the phragmocone is preserved in clear calcite and the body chamber is preserved in grey glauconitic silt and occu- pies half of the outer whorl. The umbilicus is partially occluded with matrix but it seems to have steep umbilical walls. The flanks are flat and parallel, and the venter rounded. The small size, platycone shell shape, lack of ornament and simplified sutures (Fig. 2A) with a trifid lateral (L) lobe clearly identify it with Patagonian Koloceras talenkanum. D872 (Fig. 1G) is part of a phragmocone preserved in sparry calcite. The sutures are not preserved well enough to permit a definite identification. The small size and shell shape suggest that it might be a macroconch of Koloceras talenkanum. 18 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 1. Koloceras talenkanum Riccardi et al., 1987. A-C. SAM-—PC8394. D. SAM-—PC8393. E—-F. NMB D380. G. NMB D872. H-I. SAM-—PC8392. A-D, H-I. From Estancia La Vega, Santa Cruz Province, Argentina, upper part of the Rid Mayer Formation, Late Albian. E-F. From locality 51, Zululand, Mzinene Formation, Albian IV-V. G. From locality 55, Zululand, Mzinene Formation, Albian V. All x 4. CRETACEOUS FAUNAS FROM SOUTH AFRICA 19 mm Fig. 2. External suture lines of Koloceras talenkanum Riccardi et al., 1987. A. NMB D380. B. SAM—PC8392. Scale bar for size. Discussion These two specimens do not add anything to the extensive descriptions of Koloceras talenkanum provided by Riccardi et al. (1987), but they do present a palaeobiogeographic enigma. Thus far, Koloceras talenkanum, or the genus Koloceras for that matter, were only known from a single locality, 0,7 km to the south-east of the Estancia La Vega, approximately 34 km north-west of the village of Tres Lagos, where about 30-35 metres of sediment referable to the upper part of the Rid Mayer Formation are exposed. According to Riccardi et al. (1987: 118), Koloceras talenkanum seems to occur throughout the extent of the exposure. The associated ammonite fauna consists of a strange mixture of apparent endemic, South Gondwanid, and 20 ANNALS OF THE SOUTH AFRICAN MUSEUM cosmopolitan species respectively, which have been referred to the Puzosia vegaensis Assemblage Zone (Riccardi et al. 1987: 121), and includes Puzosia vegaensis Leanza, Hypophylloceras lestai Leanza, Partschiceras sp., Anagaudry- ceras pulchrum (Crick), Eomarshallites espinosum Medina & Rinaldi, E. hibridum Medina & Rinaldi (= ‘Parasilesites’ desmoceratoides (Stolley)), ? Umsinenoceras, Mortoniceras (Neokentroceras) tarense H. Leanza, Labeceras singulare (A. Leanza), L. crassetuberculatum magnum Aguirre Urreta & Ric- cardi, and Myloceras (Calliscaphites) andinus A. Leanza, (cf. A. Leanza 1970, Nullo et al. 1981; H. Leanza 1986; Medina & Rinaldi 1986; Riccardi et al. 1987; Aguirre Urreta & Riccardi 1988). The bivalve genus Maccoyella Etheridge (1892)—thus far only known from Patagonia, Australia and New Zealand—is also common (see Waterhouse & Riccardi 1970). Detailed synonymies still have to be worked out for some of the above spe- cies, but cosmopolitan forms include Phylloceras (H.), Partschiceras and Ana- gaudryceras pulchrum; the labeceratinae are typical South Gondwanid forms (see Klinger 1989 for a review); and endemic forms are Eomarshallites and Kolo- ceras (the latter with the exception of the Zululand specimen(s?)). Representa- tives of the subfamily Mortoniceratinae are conspicuously rare. Up to now only a fragment of Neokentroceras has been recorded from Estancia La Vega (H. Leanza 1985). Judging by the similarity of some of the labeceratine faunas (see Aguirre Urreta & Riccardi 1988; Klinger 1989), an open marine connection must have existed between part of the Patagonian Austral basin and the Zululand basin. We can think of no single factor to explain the selective barrier between the two basins, permitting free interchange of cosmopolitan and part South Gondwanid faunas but yet restricting interchange of Eomarshallites and Koloceras on the one side and Mortoniceratinae on the other. On a less negative note, we have to bear in mind that Estancia La Vega is thus far the only Upper Albian locality known in Patagonia. Future exploration may well diminish the apparent endemic char- acter of the Upper Albian ammonoid fauna of Patagonia. ACKNOWLEDGEMENTS We thank C. M. Joubert for bringing these specimens to our notice. Both were collected in 1922-23 by the late Dr E. C. N. van Hoepen in a section of the Mzinene River that is now flooded due to extensive damming downstream. We pay tribute to Van Hoepen’s meticulous collecting activities—the fruits of which lay undiscovered for more than sixty years. Drs M. B. Aguirre Urreta, F. Med- ina, G. Blasco de Nullo (Buenos Aires) and A. C. Riccardi (La Plata) allowed access to their collections and provided stratigraphic and other data to H.C.K. during a visit to Argentina in 1986. Dr G. Blasco de Nullo provided the figured Argentinian specimens. Their generous assistance is gratefully acknowledged. Sally Dove prepared the photographic plates and Jacqueline Blaeske did the CRETACEOUS FAUNAS FROM SOUTH AFRICA TA drafting. Financial assistance from the Foundation for Research Development to Klinger and from the NERC to Kennedy is gratefully acknowledged. REFERENCES AGUIRRE URreETA, M. B. & Riccarpi, A. C. 1988. Albian heteromorph ammonoids from southern Patagonia, Argentina. Journal of Paleontology 62 (4): 598-614. BRUNNSCHWEILER, R. O. 1959. New Aconeceratinae (Ammonoidea) from the Albian and Aptian of Australia. Bulletin. Bureau of Mineral Resources, Geology and Geophysics, Australia 54: 1-19. Casey, R. 1954. Falciferella, a new genus of Gault ammonites, with a review of the family Aconeceratidae in the British Cretaceous. Proceedings of the Geologists’ Association 65 (3): 262-277. Cooper, M. R. & KENNEDY, W. J. 1979. Uppermost Albian (Stoliczkaia dispar Zone) ammo- nites from the Angolan littoral. Annals of the South African Museum 77 (10): 175-308. KENNEDY, W. J. & KLINGER, H. C. 1975. Cretaceous faunas from Zululand and Natal, South Africa. Introduction, stratigraphy. Bulletin of the British Museum (Natural History) (Geology) 25 (4): 263-315, 1 pl. KENNEDY, W. J. & KLINGER, H. C. 1979. Cretaceous faunas from Zululand and Natal, South Africa. The ammonite superfamily Haplocerataceae Zittel, 1884. Annals of the South African Museum 77 (6): 85-121. Kuincer, H. C. 1976. Cretaceous heteromorph ammonites from Zululand. Memoirs. Geologi- cal Survey South Africa 69: 1-142. Kuincer, H. C. 1989. The ammonite subfamily Labeceratinae Spath, 1925: systematics, phy- logeny, dimorphism and distribution (with a description of a new species). Annals of the South African Museum 98 (7): 189-219. LEANZA, A. F. 1970. Ammonites nuevos 0 pocos conocidos del Aptiano, Albiano y Ceno- maniano de los Andes australes con notas acerca de su posicion estratigrafica. Revista de la Asociacion geologica argentina 25 (2): 197-261. LEANZA, H. 1986. Un nuevo mortoniceratido (Cephalopoda-Ammonoidea) del Albiano superior de Santa Cruz, Argentina. Ameghiniana 22 (3-4): 249-254. McNamara, K. J. 1978. Myloceras (Ammonoidea) from the Albian of central Queensland. Alcheringa 2 (3): 231-242. McNamara, K. J. 1985. A new micromorph ammonite genus from the Albian of South Aus- tralia. Special Publication. South Australia Department of Mines and Energy 5: 263-268, Ol, Ie Mepina, F. & Rinatpi, C. A. 1986. Ammonoideos del Albiano superior en la Formacion Rid Mayer de la Estancia La Vega, Provincia de Santa Cruz. Contribucion. Direccion Nacional del Antartico, Instituto Antarctica Argentina 324: 1-34, pls 1-S. NuLLo, F., PRosErpIO, C. A. & BLASCO DE NULLO, G. 1981. El Cretacico de la Cuenca Austral. In: VOLKHEIMER, W. & Mussacuio, E. eds. Cuenas sedimentarias del Jurdsico y Cretacico 1: 181-220. Buenos Aires: Comité Sudamericano del Jurasico y Cretacico. RIccARDI, A. C., AGUIRRE URRETA, M. B. & MepINA, F. A. 1987. Aconeceratidae (Ammonitina) from the MHauterivian—Albian of southern Patagonia. Palaeonto- graphica (A) 196 (4-6): 105-185. SPATH, L. F. 1925. On Upper Albian Ammonoidea from Portuguese East Africa, with an appendix on Upper Cretaceous ammonites from Maputoland. Annals of the Transvaal Museum 11 (3): 179-200. WATERHOUSE, J. B. & Riccarpi, A. C. 1970. The Lower Cretaceous bivalve Maccoyella in Patagonia and its paleogeographic significance for continental drift. Ameghiniana 7 (3): 281-296. WHITEHOUSE, F. W. 1926. The Cretaceous Ammonoidea of eastern Australia. Memoirs of the Queensland Museum 8: 195-242, pls 34-41. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu- larly 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. nov., sp. nov., comb. nov., syn. nov., 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 trans- ferred 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 (fig. 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 specimens 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, descrip- tion 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 Eliza- beth (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. 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 .. .’: ‘. . . 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 Punctuation should be loose, omitting all not strictly necessary Reference to the author should preferably 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. The generic name should not be abbreviated at the beginning of a sentence or paragraph. Name 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. A KOLOCERAS (CEPHALOPODA, AMMONOIDEA) FROM THE MZINENE FORMATION (ALBIAN) = 7 VOLUME 99 PART 3 FEBRUARY 1990 | ISSN 0303-2515 -ANNALS OF THE SOUTH AFRICAN ~ MUSEUM | CAPE TOWN | INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (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) 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 and neat, double spaced with 3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for illustrations should be typed separately, their positions indicated in the text. All pages should be num- bered 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; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including caption); the reduction or enlargement required should be indicated (and preferably uniform); orig- inals 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 caption; 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. If Letraset is used authors are requested to use Helvetica-style letter- ing, if possible. 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 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, 1969b) 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 (according to the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BuLLoucu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FIscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140. FiscHer, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie expérimentale et générale 74 (33): 627-634. Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and Magazine of Natural History (13) 2 (17): 309-320. Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of the Bingham Oceanographic Collection, Yale University 17 (4): 1-51. THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro- pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgeftihrt in den Jahren 1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 99 Band February 1990 Februarie ant soe Dee! A NEW GENUS OF RUTITRIGONIINAE (BIVALVIA, TRIGONIACEA) FROM THE LOWER CRETACEOUS (APTIAN) OF ZULULAND By MICHAEL R. COOPER 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 iL, AUS, 5-8), SUD, 25, 8 eo.t), SAS, 5, 7-9), Ge, toi) TOD), 8, COD, 7), MOT, AGED, 5, 7 es oat)), IAC), ICS), DAO), 27, S13), GC), 23, BIO), 45) Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 108 3 Printed in South Africa by In Suid-Afrika gedruk deur The WRustica sPressan etyenaele tae Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap A NEW GENUS OF RUTITRIGONIINAE (BIVALVIA, TRIGONIACEA) FROM THE LOWER CRETACEOUS (APTIAN) OF ZULULAND By MICHAEL R. COOPER Department of Geology, University of Durban-Westville, Natal (With 1 figure) [MS accepted 12 September 1989] ABSTRACT Trigonia (Rutitrigonia) pongolensis Rennie lacks the characters of Rutitrigonia, and is made the type species of the new genus Zulutrigonia. Other species to be referred here include T. krenkeli Lange and T. kigombana Aitken from the Neocomian—Aptian of Tanzania. CONTENTS PAGE MN GHOCU Ci Omer we ateneg eater aie para a ent cna canynd ani cic. a ore mania eat Niel tease Op SyStemavicdeSchipUOmMerme ie Atk Tae sean Crates at aiie jocles has eu: 24 INCKnOwledeementSraser tne tae: Globe Ie aati me oee se oss 28 INCH INCC Sarma ine ey Cen ines mie ener se niente aiean ees Si ue Siete a Miele 8 28 INTRODUCTION In a benchmark paper representing a marked departure from the conser- vative taxonomy of that period, Van Hoepen (1929) introduced a number of subfamilies and genera for trigoniid bivalves from the Cretaceous of Zululand. Among these was the genus Rutitrigonia, introduced for R. peregrina van Hoepen from the upper Middle or low Upper Albian. This taxon formed the basis for the monogeneric subfamily Rutitrigoniinae van Hoepen, corresponding broadly with the Excentrica group of Lycett (1879). Van Hoepen’s (1929) bold taxonomy failed to make immediate impact, and Rutitrigonia was reduced to a subgenus of Trigonia by Rennie (1936), whereas Cox (1952) and Aitken (1961) treated it as a subgenus of Megatrigonia. However, Crickmay (1932), Kobayashi (1957), Saveliev (1958), Nakano (1961, 1963, 1965, 1977), Cox (1969), Poulton (1977, 1979) and Reyes & Pérez (1978) have all used Rutitrigonia as a valid genus, and it is now widely accepted. In addition to the type species, the following species have also been assigned to Rutitrigonia: R. affinis (Sowerby), R. agrioensis (Weaver), R. amagensis (Koba- yashi), R. beyrichi (Krumbeck), R. beyschlagi (Muller), R. bornhardti (Miller), 23 Ann. S. Afr. Mus. 99 (3), 1990: 23-29, 1 fig. 24 ANNALS OF THE SOUTH AFRICAN MUSEUM R. coquandiana (d’Orbigny), R. dietrichi (Lange), R. dunscombensis (Lycett), R. excentrica (Parkinson), R. jacksonensis (Packard), R. janenschi (Lange), R. kigombana (Aitken), R. laeviuscula (Lycett), R. lerchi (Hill), R. longa (Agassiz), R. mesembris (Tenison-Woods), R. niongalensis (Lange), R. nossae Aitken, R.nyangensis Aitken, R. pongolensis (Rennie), R. sanchuensis (Nakano), R. schwarzi (Miller), R. semiculta (Forbes), R. skorkovensis Pugaczew- ska, R. syrdariensis (Archangelsk’ii), R. weaveri (Stoyanow), and R. yeharai Kobayashi. In introducing the new species Trigonia (Rutitrigonia) pongolensis, Rennie (1936: 358) noted that *. . . It is not without some hesitation that I associate with Rutitrigonia the species T. pongolensis sp. nov. and T. krenkeli. The concentric ribbing of the neanic stage is of such a character in both these species as to exclude them from /ndotrigonia in spite of adult similarities, and Rutitrigonia is the only term available for species of Trigonia having features at all like those of the last two species.’ Nakano (1963) excluded Trigonia (Rutitrigonia) pongolensis Rennie from his concept of Rutitrigonia because of its coarse oblique flank costae, referring it instead to Megatrigonia. Subsequently, he identified it (Nakano 1965) as Megatrigonia? pongolensis (Rennie). Given the large number of Rutitrigonia species now known, it is clear that Trigonia pongolensis and its allies can no longer be accommodated satisfactorily in this genus. It is proposed, therefore, to introduce a new genus for these forms. SYSTEMATIC DESCRIPTION Superfamily TRIGONIACEAE Lamarck, 1819 Family Trigoniidae Lamarck, 1819 Subfamily Rutitrigoniinae van Hoepen, 1929 Diagnosis Shell small to moderately large, subovate to pyriform, subtrigonal and subtrapezoidal. Typically with inner and marginal carina lacking, resulting in poor discrimination between the smooth escutcheon and area. The latter is ornamented only in the nepionic stages and may be weakly bipartite. In some taxa there is a more or less prominent umbonal ridge. Nepionic ornament comprises concentric costae that pass uninterrupted on to the area where they weaken across the longitudinal furrow. Later flank ornament comprises costae that are subconcen- tric anteriorly, but posteriorly may be subconcentric, strongly oblique, or effaced. Discussion The origins of Rutitrigonia are cryptic. It first appears in the Bathonian of East Africa (Venzo 1942), and its ancestry has been linked to the myophorelline Frenguelliella (Kobayashi 1957; Nakano 1963). However, it seems to be Gon- A NEW GENUS OF RUTITRIGONIINAE Ds, dwanic in origin and there is a noteworthy similarity with primitive Anditrigonia species such as the Lower Callovian A. keideli (Weaver) (cf. Leanza & Garate, 1987, pl. 3 (fig. 1)). Like Van Hoepen (1929) and Cox (1952), the writer considers the Megatrigoniinae and Rutitrigoniinae to be closely allied; they are here regarded as sister taxa. Genus Zulutrigonia gen. nov. Type species. Trigonia (Rutitrigonia) pongolensis Rennie, 1936; by original designation herein. Diagnosis Shell medium sized, subtrapezoidal, with inconspicuous subterminal umbones and opisthogyrous beaks. Anterior margin weakly convex, posterodorsal margin almost straight and respiratory margin obliquely truncate. Escutcheon narrow, indistinct, not sunken. Broad subtrigonal area essentially flat, meeting the posterodorsal commissure in an acute angle, so that the shell is fastigiate (roof-shaped) posteriorly. Ornament comprises coarse, robust, relatively sparse costae that are subconcentric anteriorly and curve strongly upwards posteriorly, cutting obliquely across the growth striae. Age: Lower Cretaceous (Neocomian— Aptian). Discussion Zulutrigonia gen. nov. is distinguished from Rutitrigonia by its subtrapezoidal outline, with subterminal umbones, almost straight posterodorsal margin and obliquely truncate siphonal margin. In addition, the shell is fastigiate postero- dorsally, due to the broadly flattened areas meeting the commissure at an acute angle, and the coarse, robust flank costae curve strongly upwards posteriorly. There are noteworthy similarities between Zulutrigonia gen. nov. and Bucho- trigonia (Syrotrigonia) (Cox, 1952). Both are characterized by moderate inflation, subterminal umbones, a straight posterodorsal margin and obliquely truncate res- piratory margin, an inconspicuous to near-obsolete escutcheon, and broad flat- tened areas that produce a strongly fastigiate profile to the shell posterodorsally. The concentric ornament of the nepionic and middle growth stages is identical. Syrotrigonia may be distinguished, however, by its more ovate outline, the posses- sion of a marginal carina that becomes a tuberculate umbonal ridge in maturity, the presence of a broad shallow antecarinal sulcus in maturity across which ribbing weakens or is effaced, and the generally V-shaped costae of the adult. Significantly, however, the V-shaped costae are far from uniformly developed and in some indi- viduals examined they simply bend strongly upwards as in Zulutrigonia gen. nov. Similarly, the strong marginal carina that was emphasized in the original diagnosis is not a true carina (the latter is restricted to the nepionic stages) but an umbonal ridge whose prominence is enhanced by the bounding antecarinal sulcus. Initially the tubercles on this ridge are no more than rib endings that are emphasized by the weakening of ornament across the antecarinal sulcus. Whether the similarities 26 ANNALS OF THE SOUTH AFRICAN MUSEUM are due to phylogeny or are the result of convergence is, at present, unclear. Syro- trigonia is here elevated to full generic status because Buchotrigonia has radial flank costae in the early to middle growth stages, an internally crenulated postero- ventral margin and transverse costellae to the escutcheon (cf. Pérez & Reyes 1980), characters that ally it with the Apiotrigoniinae and Pterotrigoniinae. Zulutrigonia pongolensis (Rennie, 1936) Fig. LA-E Trigonia (Rutitrigonia) pongolensis Rennie, 1936: 359, pl. 41 (figs 5-6), pl. 42 (figs 5-7). Megatrigonia (Rutitrigonia) pongolensis (Rennie) Cox, 1952: 59. Megatrigonia pongolensis (Rennie) Nakano, 1963: 527. Megatrigonia? pongolensis (Rennie) Nakano, 1965: 17. Type material The holotype, SAM—PCZ8528, and paratype, SAM—PCZ8529, both in the South African Museum, Cape Town. Type locality The material was collected by Drs S. H. Haughton and A. W. Rogers along the Mfongosi stream, a tributary of the Pongola, at their locality Z2. It is from the Mlambongwenya Formation (Cooper & McCarthy 1988), which here is of late Aptian age and corresponds broadly with Locality 164 of Kennedy & Klinger OTS). Description Shell medium sized (maximum length about 60 mm), longer than high (H/L = 0,73-0,80), subtrapezoidal, with low subterminal umbones and incurved opisthogyrous beaks. The valves are moderately inflated (W/L = 0,20—0,22) and maximum inflation is just behind the umbones. The broad, weakly curved anterior margin is subvertical and curves evenly into the long, almost straight, ventral margin. The broad, obliquely truncate respiratory margin meets the straight posterodorsal margin in a very obtuse angle. The dorsal and ventral margins are subparallel. The smooth, narrow, flat, lanceolate escutcheon is poorly discriminated and can be distinguished from the area only in the nepionic stages when it is ornamented. Marginal and inner carinae are lacking, but an umbonal ridge marks the position of the marginal carina. The broad subtrigonal area is weakly concave, almost flat, and with an indistinct longitudinal groove. It forms an acute angle with the opposing area at the posterodorsal commissure so that the posterodorsal part of the shell is fastigiate (roof-shaped). The ornament of the nepionic stages comprises coarse concentric costae, broader than the interspaces, which flex slightly at the umbonal ridge before crossing onto the area. The ribs thin noticeably where they cross the longitudinal groove. Later flank costae are coarse, robust, relatively distant and slightly A NEW GENUS OF RUTITRIGONIINAE Di) Fig. 1. A-E. Zulutrigonia pongolensis (Rennie). A-—C. Anterior, lateral and dorsal views of the paratype, SAM-—PCZ8529. D-E. Lateral and dorsal views of the holotype, SAM—PCZ8528. F. Zulutrigonia krenkeli (Lange). A plastercast of the holotype, in the South African Museum. All x 1. 28 ANNALS OF THE SOUTH AFRICAN MUSEUM narrower than the interspaces. They are subconcentric anteriorly but curve strongly upwards to the posterior, cutting the growth striae obliquely and approaching the area almost at right angles. In the middle growth stages the ribs terminate at the umbonal ridge but in maturity they weaken and become effaced before reaching the area. The flank costae become obsolete on the anterolateral shoulders, leaving the narrow anterior face crossed only by growth striae. Discussion Other species to be assigned to Zulutrigonia include the Tanzanian Trigonia krenkeli Lange (1914: 230, pl. 20 (fig. 2)) and T. kigombana Aitken (1961: 95, pl. 13 (fig. 3)), both from the Rutitrigonia schwarzi Beds of the Tendaguru succession and thus of Hauterivian age. Zulutrigonia krenkeli (Lange) (1914: 230, pl. 20 (fig. 2)) (Fig. 1F) differs from Z. pongolensis in having dorsal and ventral margins that converge posteriorly and with a lower, less-fastigiate posterodorsal profile. Zulutrigonia kigombana (Aitken) (1961: 95, pl. 13 (fig. 3)) differs from Z. pongolensis in being strongly produced posteriorly, with a gently concave pos- terodorsal margin, and with the posterior one-third of the flanks unornamented. Trigonia inca Fritzsche (1921: 49, pl. 3 (figs 1-3)) from the Barremian of northern Peru shows some similarity to Zulutrigonia. However, its flank costae form distinct chevrons in maturity and hence it may be a Syrotrigonia (Etayo-Serna 1985; Pérez & Reyes 1986). Occurrence Zulutrigonia pongolensis (Rennie) is currently known only from the late Aptian of northern Zululand. ACKNOWLEDGEMENTS I thank Drs M. Tashiro (Kochi) and E. Pérez d’Angelo (Santiago) for their constructive criticism of the manuscript. Betsie Greyling assisted with the photography and the University of Durban-Westville is thanked for a grant in aid of publication. REFERENCES AITKEN, W. G. 1961. Geology and palaeontology of the Jurassic and Cretaceous of southern Tanganyika. Bulletin of the Geological Survey of Tanganyika 31: 1-144. Cooper, M. R. & McCartny, M. J. 1988. The stratigraphy of the Uloa Formation. Extended Abstract, Geocongress ’88, Durban: 121-124. Durban: Geological Society of South Africa. Cox, L. R. 1952. Notes on the Trigoniidae, with outlines of a classification of the family. Proceedings of the Malacological Society of London 29 (2-3): 45-70. Cox, L. R. 1969. Trigoniacea Lamarck, 1859. In: Moore, R. C. ed. Treatise on invertebrate paleontology. Part N, Mollusca 6 (1): N471—-N489. Boulder: Geological Society of America; Lawrence: University of Kansas. Crickmay, C. H. 1932. Contributions toward a monograph of the Trigoniidae. American Journal of Science 24 (5): 443-464. A NEW GENUS OF RUTITRIGONIINAE 29 ETAYO-SERNA, F. 1985. Paleontologia estratigrafica del sistema Cretacico en la Sierra Nevada del Cocuy. Publicaciones Especiales Ingeominas, Colombia, No. 16, Proyecto Cretdacico, Capitulo 24: 1-47. FritzscHe, C. H. 1921. Neue Kreidefaunen aus Siidamerika (Chile, Bolivia, Peru, Colombia). In: STEINMANN, G. ed. Beitrdge zur Geologie und Paldontologie von Stidamerika. Neues Jahrbuch fiir Mineralogie, Geologie und Paldontologie 50: 1-56, 313-334. KENNEDY, W. J. & KLINGER, H. C. 1975. Cretaceous faunas from Zululand and Natal, South Africa. Introduction, stratigraphy. Bulletin of the British Museum of Natural History (Geology) 25 (4): 265-315. KopayasHl, T. 1957. Nipponitrigonia and Rutitrigonia in Japan. Transactions and Proceedings of the Palaeontological Society of Japan (n.s.) 26: 51-61. Lance, E. 1914. Die Brachiopoden, Lamellibranchiaten und Anneliden der Trigonia schwarzi-Schicht, nebst verleichender Ubersucht der Trigonien der gesamten Tendagur- uschichten. Archiv ftir Biontologie 3 (4): 193-289. LEANZA, H. A. & Garate, J. I. 1987. Faunas de Trigonias (Bivalvia) del Jurasico y Cretacico Inferior de la Provincia del Nequén, Argentina, conservadas en el Museo Juan Olsacher de Zapala. In: VOLKHEIMER, W. ed. Bioestratigrafia de los Sistemas Regionales del Jurasico y Cretacico de América del Sur 1: 201—255. Mendoza. Lycetr, J. 1872-79. A monograph of the British fossil trigoniae. Monograph of the Palaeontological Society: 1-52 (1872); 53-92 (1874); 93-148 (1875); 149-204 (1877); 205-245 (1879). Nakano, M. 1961. On the Trigoniinae. Journal of Science of the Hiroshima University (C, Geology and Mineralogy) 4 (1): 71-94. NAKANO, M. 1963. On the Rutitrigoniinae. Geological Report of the Hiroshima University 12: 513-529. NAKANO, M. 1965. On the Megatrigoniinae. Journal of Science of the Hiroshima University (C, Geology and Mineralogy) 5 (1): 13-20. Nakano, M. 1977. A new genus Levantotrigonia nov. gen. Journal of Science of the Hiroshima University (C, Geology and Mineralogy) 7 (4): 199-202. PEREZ, E. & Reyes, R. B. 1980. Buchotrigonia (Buchotrigonia) topocalmensis sp. nov. (Trigoniidae; Bivalvia) del Cretacico superior de Chile. Revista Geologica de Chile 9: 37-55. PEREZ, E. & Reyes, R. B. 1986. Presencia de Buchotrigonia (Syrotrigonia) Cox, 1952 (Bivalvia; Trigoniidae) en Sudamérica y descripcion de dos especies nuevas. Revista Geologica de Chile 28-29: 77-93. PouLtton, T. P. 1977. Early Cretaceous trigoniid bivalves of Manning Provincial Park, southwestern British Columbia. Geological Survey Paper. Mines and Geology Branch, Canada 76-9: 1-25. Poutton, T. P. 1979. Jurassic trigoniid bivalves from Canada and the western United States of America. Bulletin. Geological Survey of Canada 282: 1-82. RENNIE, J. V. L. 1936. Lower Cretaceous Lamellibranchia from northern Zululand. Annals of the South African Museum 31 (3): 277-391. Reyes, R. B. & PEREZ, E. 1978. Las trigonias del titoniano y cretacico inferior de la cuenca andina de Chile y su valor cronoestratigrafico. Boletin de Instituto de Investigaciones Geoldgicas de Chile 32: 1-105. SAVELIEV, A. A. 1958. The Lower Cretaceous trigontids of Mangyschlak and western Turkmenia. Trudy Vsesoyuznogo neftyanogo nauchno-issledovatel’skogo geologorazved- ochnogo instituta (VNIGRI) 125: 1-516. (In Russian.) VAN Hoepen, E. C. N. 1929. Die krytfauna van Soeloeland. 1. Trigontidae. Paleontologiese Navorsing van die Nasionale Museum van Bloemfontein 1 (1): 1-38. VeENZO, S. 1942. Trigonia (Laevitrigonia) stefaninii n. sp. del Batoniano dell’Oltregiuba. Revisione del sottogenere e sua distribuzione. Atti della Societe italiana di scienze naturali, e del Musei civico di storia naturale 81 (3-4): 210-229. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu- larly 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. nov., sp. nov., comb. nov., syn. nov., 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 trans- ferred 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 (fig. 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 specimens 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, descrip- tion 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 Eliza- beth (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. 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 ...’: ‘. . . 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 Punctuation should be loose, omitting all not strictly necessary Reference to the author should preferably 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. The generic name should not be abbreviated at the beginning of a sentence or paragraph. Name 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. MICHAEL R. COOPER A NEW GENUS OF RUTITRIGONIINAE (BIVALVIA, TRIGONIACEA) FROM THE LOWER CRETACEOUS | (APTIAN) OF ZULULAND 'OF THE SOUTH AFRICAN OR EE Mose CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (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) 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 and neat, double spaced with 3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for illustrations should be typed separately, their positions indicated in the text. All pages should be num- bered 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; 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 caption); the reduction or enlargement required should be indicated (and preferably uniform); orig- inals 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 caption; 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. If Letraset is used authors are requested to use Helvetica-style letter- ing, if possible. 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 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, 1969b) 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 (according to the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BuLtoueu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FiscuEr, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140. FiscHer, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie expérimentale et générale 74 (33): 627-634. Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and Magazine of Natural History (13) 2 (17): 309-320. Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of the Bingham Oceanographic Collection, Yale University 17 (4): 1-51. THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro- pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren 1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270. (continued inside back cover) ‘ = itn) e y ‘ : . fe ? ~ nl ¢ y A Fig. 1. Pennatulaceans of southern Africa. A-C. Actinoptilum molle Kikenthal, 1910). D-F. Virgularia schultzei Kikenthal, 1910. G. Echinoptilum echinatum (Kikenthal, 1910). H-I. Echinoptilum macintoshii Hubrecht, 1885. ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 99 Band May 1990 Mei Part 4 Deel THE PENNATULACEA OF SOUTHERN AFRICA (COELENTERATA, ANTHOZOA) By GARY C. WILLIAMS 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 il, HOES), 5-8), CW, 425, 8, Loh), SRS, 5, 7-9), GGL tat), TED), 8, HOD, 1), 1OMB), NAD, 5, 7, t.70.8.), GED), 1SCES), DAM, 5), 27, S28), SAG), 33, 36(2), 43(1), 45(1), 67(5), 84(2) Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 109 1 Printed in South Africa by In Suid-Afrika gedruk deur ihe wus ticamEnessamtyer aletcey Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap C460 THE PENNATULACEA OF SOUTHERN AFRICA (COELENTERATA, ANTHOZOA) By Gary C. WILLIAMS* Department of Marine Biology, South African Museum, Cape Town, South Africa and Department of Zoology, University of Cape Town, Rondebosch, South Africa (With 41 figures) [MS accepted 29 April 1988] ABSTRACT Nineteen species of pennatulacean octocorals from sublittoral and bathyal regions of southern African coastal waters are described and illustrated. Six additional species that are recorded in the literature are mentioned, although material was not available for examination. Two of these, Cavernulina cylindrica and Virgularia juncea, are considered dubious records and are not included in the faunal list provided for the region. Presumably young or partial colonies or problematic taxa representing seven genera (Cavernulina, Kophobelemnon, Scleroptilum, Calibelemnon, Virgularia, Halipteris, and Pteroeides) are not identified to species but are mentioned in the text and included in the list of species. A total of 23 determined and at least seven undetermined species representing 20 genera in 12 families are recorded. Three species that have been recorded from Angola, which is north of the geographic limit of this paper, are also mentioned: Virgularia tuberculata, Pteroeides griseum, and Crassophyllum cristatum; these may possibly be encountered further south. Fourteen new geographic records are established, including the first southern African records for the following ten genera: Cavernulina, Echinoptilum, Kophobelemnon, Distichop- tilum, Scleroptilum, Calibelemnon, Chunella, Amphiacme, Scytaliopsis, and Halipteris. Scanning electron micrographs of sclerite form or axial morphology are presented for many species. A key is provided to all species treated in the present work. A brief discussion of faunistic considerations is also included. CONTENTS PAGE MTGhOGUCtIOM peer teeta rm ye ein ug is retin arn OaMtve aia Seg 32 Watentalsrandumethods qs sn cae eines ottawa ca aes o.oo clea es 55 SV SUCIMAUCrACCOUMUr ier eerie raesst sie er alam sie ene me Laicia Sue lane urun leit 55 Key to the families of Pennatulacea from southern Africa.......... 315) Key to the species of Pennatulacea from southern Africa .......... 36 List of Pennatulacea recorded from southern Africa............... 39 DD CSCHIP CONS ye See etn te ioe ohio Fie e soa iennnte alee 40 Any ene Clllidae rem metre meena rye Macias ok mys. shee a oi 40 Banulyse chino ppilidacwwe. 4 omecce ne ciate ns ane ere: sheen es 50 BanulyaCophobelemmidaeyes x20. osinee se ce te ees se dante G 64 Banuilye Amino ptilidacwmiere sare sya si-) Sellers cece oi olsun rosie weary 68 RannilyaumiculiniGdaen ie ee cas ete so hae ols sd ain os so alate ale lores 69 Bamilyperocopulidace nr tei. ese | ee ear oe rea ei mal * Present address: Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, U.S.A. Sil Ann. S. Afr. Mus. 99 (4), 1990: 31-119, 41 figs. 32 ANNALS OF THE SOUTH AFRICAN MUSEUM PAGE Family Scleroptilidaci...5 3225 (ae ace re ieee ne i Family Chunellidae. 0%... 40055 eer ek nk ee ene 76 Family Umbellulidae <...0.25 29) 58s eos ee eels eee eee 81 Family Virgulamidae. .0.5..a.36 5 eae ote oot soe eee tee oe 86 Family Pennatulidaes.%..5 22) 44s ce tee er een 99 Family Pteroeididac’s 0.2). a aes a serees oehe eeee 106 DISCUSSIONS ie 8 ot occ eee tas a eae ce cs oy ee ee 112 Summary and conclusions sxe. 8. ae aaerecroe encie roe ae eee Rear ee 113 Acknowledements...4 2. s5.ncgss sake so sea Ooenak os eee ee 11S FRELETENCES > ac: oh Gif whee Ps Gece ea een cs NOIRE TRS OS te EE ne eet CC 114 INTRODUCTION At present, octocoral classification is frequently an equivocal and disputa- tious field (Weinberg 1976: 63). It is clear that our knowledge concerning this group is far from adequate; a great deal of research needs to be accomplished in order to attain the goal of a world-wide synthesis of the biology and classification of octocorals (Bayer 1981a: 11). A substantial amount of confusion has arisen in the literature because of a general lack of knowledge concerning variability; some octocoral species can be exceedingly variable and many authors have described species or even genera based on intraspecific variants. Extensive collecting in many different geographical localities and detailed comparison of material is needed to assess the degree of variation in many taxa due to genetic, geo- graphical, or ecological differences. The goal of the present study of the systematics and faunistics is to contribute toward a world-wide synthesis of the classification and distribution of octocorals by concentrating on a localized fauna that has been previously only minimally to moderately known. This work is the first of four proposed parts describing the octocorals of southern Africa. Subsequent papers will deal with additional species of pennatulaceans from the Cape, Transkei and Zululand; stoloniferous octocorals and soft corals (families Clavulariidae, Tubiporidae, Coelogorgiidae, Alcy- oniidae, and Nephtheidae); and the shallow-water gorgonians (families Antho- thelidae, Melithaeidae, Keroeididae, Acanthogorgiidae, Plexauridae, Gorgonii- dae, Ellisellidae, Chrysogorgiidae, Primnoidae, and Isididae). The four papers together will attempt to synthesize our knowledge of southern African octocorals. The studies exclude the deep-water gorgonians, primarily of the families Ellisellidae, Primnoidae, and Chrysogorgiidae, groups that are less well known regionally or that require large-scale revisions before accurate identifications can be made. Many octocoral species that have been collected still need to be identified and new species are continually being discovered. The present work is intended not only for specialists in the field of octocoral systematics but also as a guide for other biologists, e.g. ecologists, fisheries biologists and biogeographers, who share the common need to identify material from benthic surveys and other studies. Southern African octocorals are of interest to non-specialists, since they comprise such a major part of the SOUTHERN AFRICA PENNATULACEA 33 shallow-water benthic fauna. Therefore, sections have been added to the texts of these papers that would not normally be included in a work intended solely for specialists. These include brief diagnoses of the families and genera, sections on general characteristics of the octocoral groups, and illustrations of the entire colonies of all species, as well as figures depicting diagnostic features, sclerites, and other features. The monograph by Kikenthal (1915) remains the most comprehensive work on pennatulaceans of the world and is certainly the most thorough account up to that time. Modern revisions of many of the genera are obviously needed. Other major reviews on sea pens are mainly monographs produced from the large Oceanographic expeditions of the past and include K6lliker (1872a, 1872b), Kolliker (1880) (Challenger), Danielssen & Koren (1884) (Norwegian North Atlantic), Kukenthal & Broch (1911) (Valdivia), Hickson (1916) (Siboga), and Broch (1958b) (Discovery). Thomson (1915, 1917) described the pennatulacean fauna of South Africa known at that time and Bayer (1956) reviewed the known world genera. Tixier-Durivault (1954, 1960, 196la, 19666, 1972) and Tixier- Durivault & D’Hondt (1974) described material from South Africa, Mozambique, Angola, and Madagascar (including neighbouring islands). General accounts of the biology of sea pens are by Hickson (1909) and Hyman (1940). More recently, the distribution and taxonomy of pennatulaceans has been reported on by Pasternak (1960, 1961a, 1961b, 1962, 1964, 1970, 1975a, 1975b, 1980), Grasshoff (1972, 1981, 1982a, 1982b), and D’Hondt (1984a, 19845). Bayer (1981a: 7) classified various geographic regions of the world into four general categories of taxonomic knowledge pertaining to octocorals: essentially complete; moderately well known; poorly known; and minimally known. He considered South Africa, in the region of the Cape of Good Hope, to be moderately well known and described this category as follows: ‘where there is extensive literature, but many more species remain to be described and taxonomic problems to be solved, and the major patterns of distribution must yet be worked out. Much descriptive work remains to be done before ecological and experimental studies can proceed at an effective level’. The rest of southern Africa is less well known and certain regions fall into Bayer’s latter categories—being poorly or even minimally known. Two such regions include the Atlantic coast between 20°S and 33°S, and the northern Natal—southern Mozambique region of the Indian Ocean between 20°S and 29°S. Southern Africa is here defined as the African coastal region that lies south of 20°S—an arbitrary limit used by Millard (1978) in a zoogeographical analysis of southern African hydroids, which are another diverse and widespread benthic group. This region extends from the Skeleton Coast of Namibia in the Atlantic, round the coast to the vicinity of Beira, Mozambique, in the Indian Ocean. Material examined in the present work was obtained throughout this range from the continental shelf, slope and rise, and from the littoral to depths of 3 000 m. The island of Madagascar is not included here, since its affinities are strictly Indo-Pacific and not directly associated with the southern African fauna. 34 ANNALS OF THE SOUTH AFRICAN MUSEUM Pennatulaceans from geographical regions marginal or in proximity to the area of the present study have been described by Broch (19585) and Tixier-Durivault (196la, 1963) for Angola, and Tixier-Durivualt (19666, 1972) and Tixier- Durivault & D’Hondt (1974) for Madagascar. Of the 14 presently recognized pennatulacean families of the world, 12 (or 86 %) are represented in southern Africa, and of the 34 world genera, at least 20 (or 59 %) are known from this region. The only families not represented in the regional fauna are two monogeneric ones: the Renillidae—genus Renilla Lamarck, 1816, with four or five species restricted to South America and the southern part of North America; and the Stachyptilidae—genus Stachyptilum Kolliker, 1880, with three species presumably restricted to warmer parts of the Pacific Ocean. The classification system used in this study at the subordinal and familial levels conforms to Kikenthal (1915) and Bayer (1956). Terminology conforms to the trilingual glossary of Bayer et al. (1983). All illustrations, photographs, and scanning electron micrographs are by the author. The abbreviation ‘Lit.!’ in a synonymy indicates where a complete or more detailed synonymy can be found. The notation non in the synonymies indicates a misidentification by the author cited; partim indicates ‘in part’. Characteristics of the Pennatulacea The pennatulaceans, or sea pens, are considered a very distinct and greatly specialized group of octocorallian cnidarians (Hickson 1909: 358; 19306: 229; Bayer 1956: 181). Unlike other octocorals, pennatulacean colonies are formed by a very large polyp, the oozooid (also known as the axial, initial, principal, or primary polyp), that by lateral budding of its body walls gives rise to all the secondary zooids of the colony. The oozooid forms a fleshy, unbranched peduncle to anchor the colony by peristaltic contractions in soft substrata such as mud, sand, or abyssal-ooze deposits. The polyps of an individual colony (excluding the oozooid) are always dimorphic, more rarely trimorphic; the distal region of the oozooid forms a rachis bearing several to many daughter polyps, which take the form of autozooids, siphonozooids, and sometimes mesozooids. Disregarding the oozooid. the autozooids (Fig. 11E) are the largest polyps in polymorphic colonies and possess eight well-developed tentacles and mesenteries; these are the feeding polyps of the colony. Siphonozooids (Fig. 11C) are much smaller than autozooids and have strongly developed siphonoglyphs that produce currents for internal water circulation; the tentacles are reduced or absent. Mesozooids (Fig. 32B, D) are polyps that are intermediate in structure between autozooids and siphonozooids. In colonies with bilateral symmetry and well-developed polyp leaves (such as Virgularia, Pennatula, and Pteroeides), the dorsal side refers to the side of the rachis adjacent to the oldest (usually largest) autozooids, where the rachis appears mostly bare and uncovered by the polyp leaves; hence, the ventral side refers to the side of the rachis opposite to the dorsal side, where the polyp leaves cover most or all of the rachis surface. SOUTHERN AFRICA PENNATULACEA 35 Characteristics that are important in the classification of pennatulaceans include the form, number, and arrangement of autozooids; distribution of siphonozooids; the colonial symmetry, growth form, and axial development; and form and distribution of sclerites. Pennatulacean sclerites differ from those of other octocorals by commonly being smooth and unsculptured, sometimes grooved or three-flanged, with tubercles usually absent but, if present, then simply knobby or rounded, not thorny or elaborate. MATERIALS AND METHODS Recently procured material was collected by means of SCUBA, dredge, or beam trawl. Colonies were relaxed in a solution of magnesium chloride mixed with sea-water and preserved in 70 per cent ethanol. Small and fragile specimens were first fixed in 10 per cent formalin before transfer to ethanol. Sclerites and axial fragments were isolated by dissolving tissue in concentrated sodium hypochlorite, with repeated rinsing in water and finally in ethanol. Minute sclerites were pipetted from alcohol suspension and dropped on to stubs that had circular pieces of exposed and developed black-and-white photographic paper glued to them. After evaporation of the alcohol, the sclerites adhered well to the paper surface. Large sclerites and fragments of axial material were applied directly to the stub surface with SEM glue. Specimens were coated with gold/palladium (Au/Pd). Cambridge S200 and S180 scanning electron micro- scopes were used to examine sclerites and axial fragments. All material examined is housed in the marine invertebrate collection of the South African Museum (SAM), Cape Town. SYSTEMATIC ACCOUNT KEY TO THE FAMILIES OF PENNATULACEA FROM SOUTHERN AFRICA eee NayaceitpOlyps tree, Not fused to) amy degree... 2.2... .... 2520025. I 1B. Proximal portions of adjacent polyps fused to some degree, forming polyp leavestonmmndiscd mideess on ,omed only atthe bases)... /.0.........-.-- 9 2A. Polyps crowded and distributed on all sides of rachis, or rachis with a single longitudinal furrow or V-shaped region without polyps .............. 3 2B. Polyps confined to a terminal cluster, arranged biserially along rachis, or in groups of two to four arranged sparsely along rachis ................ 4 eee OLY I SAWALMOUC CAV CES « ayn nn. So yee sues dae be Veretillidae 3B. Polyps with non-retractile bifurcated calyces ............ Echinoptilidae 4A. Polyps arranged in a terminal cluster at the end of a long slender stalk PE ee a eee. eG SAME in os eb aes Umbellulidae TPE Lolypssamansed along the sides of the tachis. .7..4..:...5...0..0..- 5 SA. Coloniés clavate, distal portion of rachis distinctly wider than rest of colony 56-G.8 cig, S.9n6 Gra TRANG ne Once Oa ne ne Kophobelemnidae 5B. Colonies not clavate, but are long, thin and whip-like or staff-like..... 6 36 6A. 6B. TA. is}. 8A. SB. 9A. OB. 10A. 10B. 11A. 11B. IVAN. 1B. 2A. 2B. 3A. 318) 4A. 4B. SA. ANNALS OF THE SOUTH AFRICAN MUSEUM e Polyps numerous (> 40) and contiguous, arranged in two longitudinal series along rachis. Sclerites dense. Polyps retractile into calyces............ 8 Polyps few (<40), arranged sparsely in clusters of 2—4 at intervals along rachis. Calyces and sclerites not apparents..554 554-5 450 iy Polyp pairs 10-20. Intervals between polyp pairs <10 mm .. Scleroptilidae Polyp clusters <5. Intervals between clusters of 1-4 polyps is >20 mm pbug oie Reet ow 4 le tee ee ee See eee Chunellidae Calyxawithvercht terminaliteeth( eee ne eee Funiculinidae Calyx with 0—6.tenminalitcethy anaes ene ee eee Protoptilidae Polyps non-retractile, calyces absent. Adjacent polyps united at their bases only, not forming polyp leaves or raised ridges........... Anthoptilidae Polyps retractile into calyces or into margins of fleshy polyp leaves or raised TUDES istic is whe peal see nade oe ee 10 Polyps have tubular calyces with eight terminal teeth ...... Pennatulidae Polyps have calyces with two terminal teeth, or calyces absent....... i Polyp leaves stiffened with large white needle-like sclerites up to 10 mm that formiprojectineysuppentingenays are eee Pteroeididae Polyp leaves soft without sclerites, or polyps arranged in raised ridges with calyces having) tworterminalatee thie acer Virgulariidae KEY TO THE SPECIES OF PENNATULACEA FROM SOUTHERN AFRICA Adjacent polyps free, not united to any degree. Individual polyps arise directly from: the surface of the rachis) (e3e4 Fig tA) 2 ane Z Proximal portions of adjacent polyps fused to some degree, often forming flattened expansions (polyp leaves, e.g. Fig. 32A, D), raised ridges or swellings that emanate laterally from the rachis (e.g. Fig. 30B, D), or joined only at their bases with major portions of polyps free (e.g. Fig. 17B).............. 20 Polyps crowded and distributed evenly on all sides of rachis, or rachis with a single longitudinal furrow or elongate V-shaped region devoid of polyps. Colony radially symmetrical or somewhat bilaterally symmetrical, capitate, clavate, or, Gigitiform. «44. se. cent ores ee ee ee 3 Polyps restricted to a limited portion of rachis, arranged biserially along rachis, or concentrated in a terminal cluster, or in groups of two to four distributed! sparsely, along machisi= 442 4542 oe oe ne ee eee 11 Polyps. without:calyces: c.f ye an See ease Sa oe + Polyps with non=retractile biturcated (calyices 4254.) .5- 5000 ooo 9 Sclerites are smooth ovals, rods that are bilobed or branched at the ends, or spindles up to 0,4 mm in length. Colonies clavate or capitate ......... 5 Sclerites are small flat plates (<0,1 mm in length) medially constricted with rounded ends or with terminal denticles. Colonies cylindrical to elongate ClAV ALE: 2854) Scie Re RN Ee BAS IE 7 Rachis sclerites are primarily rods with bilobed or branched ends Te er Pn Mes ies sa ae OB Ordos So ad 5:2 Cavernulina sp. SB. 6A. 6B. WES 73% 8A. SB. 9A. OB. 10A. 10B. 11A. 11B. 12A. 12B. 13A. 13B. 14A. 14B. ISA. 1SB. 16A. SOUTHERN AFRICA PENNATULACEA 37 Rachis sclerites are ovals or spindles that are usually not bilobed or ip TeeATNC NS Gl ie es ee at Pe ce eo es tahoe ad al A Naa danse Pu Or ered UAC ke ds 6 Internal axis absent. Sclerites are mostly spindles... Cavernularia elegans Internal axis present. Sclerites are mostly ovals....... Cavernularia dayi Colonies elongate clavate. Sclerites are small biscuit-shaped plates, many Wwithecerminal GEmticleS 20g jccn 5c cians Stas yo Lituaria valenciennesi Colonies cylindrical or elongate clavate. Sclerites are flat plates with medial Consinictions andi smooth, roundediends...2..4.+.2..-5.555055-.-..- 8 Colonies plump and cylindrical with stalk shorter than rachis. Siphono- zooids aligned in longitudinal rows. Colour yellowish to orange-yellow 56 0.0.8.0 9 B16 80.0 Ne een tet Oe Se nan ar a ge ace Veretillum cynomorium Colonies long and narrow, clavate, with stalk equal to or longer than rachis. Siphonozooids not aligned in longitudinal rows. Colour cream to greyish- WHIGIRS c's oA eee gin tg ae they Sen rer ee ena ane ae eee Veretillum leloupi Colonies radially symmetrical; polyps evenly distributed and covering all sides of rachis. Calyces often not conspicuous with unaided vision. Colonies usually straight or sometimes curved............... Actinoptilum molle Colonies bilaterally symmetrical; rachis curved, concave side with longitu- dinal furrow or V-shaped region near base of rachis free of polyps. Calyces Conspicuous) givins colony a spiny appearance ..................... 10 Calyx length approximately three times the width. Calyx teeth long and thin, lanceolate. Peduncle sclerites are small ovals 0,04-0,15 mm in length. Colour of rachis often brick-red or deep reddish-purple ee ee cree on vats ech ale nnd Oa igaeees Echinoptilum echinatum Calyx length approximately twice the width. Calyx teeth conical or deltoid. Peduncle sclerites are rods or spindles 0,07—0,47 mm in length. Colour of rachis) Orange to reddish-brown. |... 24:5.-..- Echinoptilum macintoshii Polyps arranged in a terminal cluster at the distal end of a long naked DECINACKS 9:9 6 a:& ovBsh G4 CRO AOae een ee ee ear ESP ar oe 12 Roly ps anrangedsalone the sides of the rachis....0. 2. 4-0. 24-52.5.55 4. 13 Polyps and distal portion of stalk with sclerites ..... Umbellula thomsoni Colony lacking sclenites altogether... 2. 2.5.-..5... Umbellula lindahli Colonies clavate, distal portion of rachis wider than rest of colony; terminal end obtuse. Polyps large (>10 mm), non-retractile, without calyces .. 14 Colonies not clavate, but are long, thin and uniform. Distal portion of rachis NOtwiderthaminest Of COLONY i604 ae Hs el eek PR we 5 Rolyps Zon smear tenminalend Of rachis... 5. ...-- Kophobelemnon sp. Polyps numerous and arranged biserially along rachis ................ a 6'o oo Bhacs bad Cha Sich hee a Ree Kophobelemnon stelliferum Polyps numerous (usually >40) and contiguous, arranged along rachis in two longitudinal series. Polyps retractile into distinct calyces......... 16 Polyps few (<40), arranged sparsely in clusters or whorls of 2—4 at intervals AM OMemiaciisMiG@alyCesADSEME Me ic don wad Aliete yds ce alle elena dee ee LSS 18 Calyxaqwithie2=—omenminal tect a... . oc. y20 mm. Mature colonies >120 mm in length...... 19 Polyps of rachis arranged in pairs. Distal end of colony with a single, well-developed, somewhat asymmetrical terminal polyp................ ee a le Nees ee hee eG ic od Amphiacme abyssorum Polyps of rachis in clusters of one to four, but usually three. Distal end of colony with terminal polyp highly reduced or absent .. Chunella gracillima Polyps non-retractile, without calyces. Sclerites not evident. Adjacent polyps united at their bases only, major portion of polyps are free ...... ry oe ee eR LN re ORM, Sh Anthoptilum grandiflorum Polyps retractile into conspicuous calyces or into margins of fleshy polyp leaves. Fused proximal portions of adjacent polyps form prominent polyp leaves, ridges, or basal swellings which emanate laterally from rachis in two longitudinal series... os. ae eos ee ee 21 Polyps arranged in short oblique rows of four to six polyps in two longitudinal series. Adjacent polyps united at their bases only, often forming a common raised ridge or basal swelling. Polyps possess calyces that each havestworterninalatce thine seer eee Halipterus africana Adjacent polyps united most of their length to form distinct and flattened lateral expansions (polyp leaves) that emanate from the rachis....... DD Colonies lack visible sclerites (sclerites too small to be seen with unaided eye). Peduncles usually thin and flexible or limp and worm-like...... 73 Colonies with numerous and conspicuous sclerites (sclerites easily seen with unaided eye), Reduncles thick andsnieid 3. 26 Polyps usually 5-12 per leaf. Polyp leaves sparsely arranged, with conspicuous intervals of bare rachis between leaves ................ 24 Polyps >12 per leaf. Polyp leaves densely situated, often overlapping one another with little or no interval of bare rachis between leaves....... 25 Polyps never more than five per leaf. Peduncle somewhat stiff. Adjacent polyps of a particular leaf gradually become larger from the innermost to the outermost polyp. Gradual transition between peduncle and rachis EMT ee ee Bat shee 5 ett Sth sch is een 0) tals Scytaliopsis djiboutiensis Polyps 5-12 per leaf. Peduncle long and thin, vermiform. All polyps of a given leaf approximately same size. Peduncle and rachis abruptly delimited ee ST oS icin oa ny nS OG 8.4 8 Virgularia mirabilis Polyps 50—200 pemleaii eee noe Virgularia gustaviana Polyps 15=35:perdéatee sean ee ee eee ee Virgularia schultzei SOUTHERN AFRICA PENNATULACEA 39 26A. Polyps possess tubular calyces with eight terminal teeth and dense sclerites of three-flanged spindles. Sclerites usually orange in colour............. ra a8 abet inciy Ue eee Maan as eA ake, nome ine ae Pennatula inflata 26B. Polyps retract directly into outer margin of polyp leaf. Sclerites of polyp leaves are white or transparent spine-like needles up to 10 mm in length that align to form projecting supporting rays. Sclerites not three-flanged. Colour of colonies pale yellowish-grey or brownish-grey..... Pteroeides isosceles LIST OF THE PENNATULACEA RECORDED FROM SOUTHERN AFRICA Order PENNATULACEA Verrill, 1865 Suborder SESSILIFLORAE Kiikenthal, 1915 Family Veretillidae * Lituaria valenciennesi D’Hondt, 1984 Cavernulina sp. * Veretillum cynomorum (Pallas, 1766) * Veretillum leloupi Tixier-Durivault, 1960 Cavernularia dayi Tixier-Durivault, 1954 Cavernularia elegans (Herklots, 1858) Family Echinoptilidae Echinoptilum macintoshii Hubrecht, 1885 Echinoptilum echinatum (Kikenthal, 1910) Actinoptilum molle (Kukenthal, 1910) Family Kophobelemnidae Kophobelemnon stelliferum (Miller, 1776) Kophobelemnon sp. Family Anthoptilidae Anthoptilum grandiflorum (Verrill, 1879) Family Funiculinidae * Funiculina quadrangularis (Pallas, 1766) Family Protoptilidae Distichoptilum gracile Verrill, 1882 Family Scleroptilidae Scleroptilum sp. Calibelemnon sp. * Literature records only; material not available for examination. 40 ANNALS OF THE SOUTH AFRICAN MUSEUM Family Chunellidae Chunella gracillima Kukenthal, 1902 Amphiacme abyssorum (Kikenthal, 1902) Family Umbellulidae Umbellula thomsoni Kolliker, 1874 Umbellula lindahli Kolliker, 1874 Suborder SUBSELLIFLORAE Kiikenthal, 1915 Family Virgulariidae Subfamily Virgulariinae Virgularia schultzei Kikenthal, 1910 Virgularia mirabilis (Muller, 1776) Virgularia gustaviana (Herklots, 1863) Virgularia sp. Scytaliopsis djiboutiensis Gravier, 1906 Subfamily Balticininae Halipteris africana (Studer, 1879) Halipteris spp. Family Pennatulidae Pennatula inflata Kikenthal, 1910 Family Pteroeididae Pteroeides isosceles Thomson, 1915 Pteroeides spp. DESCRIPTIONS Family Veretillidae Herklots, 1858 Colonies radially symmetrical without a trace of external bilateral sym- metry. Colony cylindrical, clavate (club-shaped), or capitate (forming a head). Autozooids without calyces, completely retractile. Siphonozooids numerous between the autozooids. Axis developed to varying degrees: conspicuous, minute, or absent. Sclerites are plates, platelets, ovals, rods or spindles; none are three-flanged. Distribution widespread in shallow tropical and subtropical seas, mostly from the Atlantic and Indo-Pacific, shallow sublittoral to approximately 190 m in depth. Five genera, four of which are included below, in addition to Pollicella Gray, 1870. SOUTHERN AFRICA PENNATULACEA 41 Genus Cavernularia Valenciennes in Milne Edwards & Haime, 1850 Veretillum: Philippi, 1835: 277. Gray, 1870: 28. Cavernularia Valenciennes in Milne Edwards & Haime, 1850: 84. Kukenthal & Broch, 1911: 180 (Lit.!). Kukenthal, 1915: 12. Sarcobelemnon Herklots, 1858: 25. Stylobelemnon Kolliker, 1872a: 336. Balss, 1910: 79. ?Fusticularia Simpson, 1905: 561. Balss, 1910: 80. ?Parabelemnon J. A. Thomson & Simpson, 1909: 307. Diagnosis Colonies cylindrical, clavate or capitate. Sclerites are smooth ovals, rods or spindles; many are irregular in shape. Siphonozooids numerous and minute (<0,3 mm in diameter). Autozooids with or without sclerites. Axis variable; large to very small or absent. Colour usually white to cream, yellowish or greyish-white. A genus of approximately 14 species. Indo-Pacific, west coast of Africa, Mediterranean Sea, and Bay of Biscay. Type species. Cavernularia obesa Valenciennnes in Milne Edwards & Haime, 1850; Indian Ocean. Cavernularia dayi Tixier-Durivault, 1954 FIGS Z, Bo dks Cavernularia Dayi Tixier-Durivault, 1954: 626, fig. 1. Material SAM-H3594 and SAM-—H3595, off East London (33°02'S 27°56’E), 25-30 m, 16 July 1984, 6 specimens, dredge; G. C. Williams, R.V. Meiring Naude. SAM-—H3817 and H3818, off Umlaas River, Natal (29°58'S 31°01’'E), 50 m, 10 July 1985, 2 colonies, dredge; G. C. Williams, R.V. Meiring Naude. Description The colonies examined include three complete colonies, 23 mm, 30 mm and 70 mm in length, and five larger colonies, which have the proximal portions of the peduncles missing. These partial colonies are between 30 mm and 62 mm in length. Colonies club-shaped with tapering peduncle, length of which is 0,9-1,8 times the rachis length. Rachis conspicuously wider than peduncle. Autozooids concentrated on distal half of rachis; capable of complete retraction into rachis. Retracted autozooids less than 1,5 mm in diameter. Siphonozooids numerous, minute, c. 0,16 mm in diameter. Siphonozooids may extend down proximal portion of rachis, but are not conspicuous. Axis is a short rod (c. 8 mm long and 0,8 mm wide) in the centre of the proximal portion of the rachis. It is roughly circular in cross-section. The surface of the axis may be partially impregnated with oval sclerites. Axis length is less than 15 per cent of total colony length. 42 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 2. Cavernularia dayi. A. Entire colony. Scale = 5mm. B. Single autozooid showing placement of sclerites; polyp length 2,5 mm. C. Entire colony. Scale = 6mm. D. Entire colony. Scale = 20 mm. E. Longitudinal section of a colony showing position of axis. Scale = 20 mm. F. Axis, length 8,0 mm. G. Cross section of axis, diameter 0,8 mm. H. Sclerites from surface of rachis. I. Sclerites from interior of rachis. J. Sclerites from surface of peduncle. K. Sclerites from interior of peduncle. Scale: H-K = 0,15 mm. Sclerites are mostly small ovals, or elongate rods, usually 0,12 mm or less in length. They are present both in the rachis exterior and interior, as well as the peduncle exterior and interior. Some ovals may have a cross-pattern originating in the centre. Some sclerites of the rachis may be constricted in the middle with rounded or truncate ends. Sclerites are densely distributed, and are more or less consistent in shape throughout the colony. They vary from 0,02—0,13 mm in length. Autozooids may have a few scattered sclerites at the bases, extending up to less than one-third total length of polyp. The tentacles have many rod-like sclerites, 0,06 mm in length. SOUTHERN AFRICA PENNATULACEA 43 Fig. 3. Cavernularia dayi. A. Entire colony, length 30 mm. B. Partial colony, length 60 mm; peduncle absent. C. Entire colony, length 23 mm. D-I. Scanning electron micrographs of sclerites. D-—F. From surface of peduncle. D. 0,04 mm. E. 0,03 mm. F. 0,03 mm. G-I. From surface of rachis. G. 0,04 mm. H. 0,03 mm. I. 0,07 mm. 44 ANNALS OF THE SOUTH AFRICAN MUSEUM Colour greyish-white in life. Neck zones of autozooids may have eight longitudinal subcutaneous regions of deep purple pigmentation, which is conserved in alcohol. Distribution (Fig. 38) East London to the northern Natal coast, 25-50 m depth. Tixier-Durivault (1954: 626) described the type locality from off Lake St. Lucia, Natal at 27 m depth. The present study extends the range from the type locality south-west to the eastern Cape Province, and north to Hully Point, Zululand. This species is possibly a southern African endemic but many more records are necessary to accurately establish its geographical distribution. Remarks Cavernularia dayi is distinguished from other southern African species of Cavernularia by its clavate shape, small axis, and short, robust, oval or rod-like sclerites of the rachis and peduncle. Cavernularia elegans (Herklots, 1858) Figs 4, 40 Sarcobelemnon elegans Herklots, 1858: 25, pl. 7 (fig. 3). Veretillum elegans Richiardi, 1869: 127, pl. 13 (fig. 112). Cavernularia elegans Kolliker 1872a: 165, pl. 23 (figs 207-208). Kikenthal & Broch, 1911: 185, text-figs 13-15, pl. 18 (fig. 29). Kiikenthal, 1915: 16, fig. 20. Broch, 1958: 249. Cavernularia elegans (non Herklots, 1858) Hickson, 1900: 89. J. A. Thomson & Simpson, 1909: 303. Balss, 1910: 86, figs 26-28. Material SAM-—H3871, 67 km west of Cape Fria, Namibia (18°25'S 11°15’E), 274 m, May-June 1982, 9 colonies, trawl; collector unidentified. Description Colonies examined are 29-58 mm in length. Axis absent. Colonies clavate, with peduncle as long as the rachis or somewhat longer. Autozooids usually c. 1,3 mm in diameter. Siphonozooids conspicuous and numerous, in more or less longitudinal rows between autozooids, 0,1 mm in diameter. Peduncle slightly swollen near junction of rachis, with longitudinal striations on the external surface. Sclerites from base of autozooids as well as the surface of the rachis and peduncle are smooth elongate rods and spindles, 0,30—0,42 mm long. These are not three-flanged and may have bluntly rounded or somewhat pointed ends. In the rachis interior, similar sclerites are common; 0,2—0,4 mm long. The interior of the peduncle contains minute ovals, 0,01 mm in length. Colour yellowish-white with some brownish pigmentation in the polyps below the tentacles. SOUTHERN AFRICA PENNATULACEA B. Sclerites from surface Sclerites from surface of peduncle. terior of peduncle. Scale = 0,01 mm = & S = & = en (=) ae aA °°... baa oS 5 oe a So 3g oO eo 5.2 0 fs! @) ape 5 5 (ae) 6 ae & gy & 80 2 wn ie ot VVo® 3 2 sae s02 SS 5.4P = is So 5 8 e . oe AQ + 0 a) ee 46 ANNALS OF THE SOUTH AFRICAN MUSEUM Distribution (Fig. 40) Along the west African coast from Morocco to northern Namibia. The present study extends the range south from near Luanda, Angola, to the Skeleton Coast of Namibia, and represents a new record for southern Africa. The type locality is an unspecified station off the west African coast. Remarks One colony of Cavernularia elegans (Herklots, 1858) was recorded by Broch (1958: 249) from a depth of 65m off the coast of Luanda, Angola. Tixier- Durivault (1963: 72) recorded one specimen as C. pusilla (Philippi, 1835) from Lobito Bay, Angola, at 35m depth. These two species, together with C. madeirensis Studer, 1878, were considered synonymous by Broch (1958: 250), yet he retained C. elegans (instead of C. pusilla) as the correct binomial. I here consider C. elegans and C. pusilla as separate. Kikenthal (1915: 13) maintained that C. pusilla has irregular rod-shaped sclerites in the peduncle surface, whereas C. elegans has spindles. Balss (1910: 86) repeated Hickson’s (1900: 89) misidenti- fication of Actinoptilum molle as C. elegans, from 45 m depth off the Cape of Good Hope. Records of C. elegans from Japan (Balss 1910) and the Indian Ocean (J. A. Thomson & Simpson 1909) were considered dubious by Kikenthal (1915: 16). Cavernularia elegans is characterized by Ktikenthal (1915: 13, 16) as having dense spindles at the surface of the peduncle, whereas these sclerites are absent in the peduncle interior; axis absent; sclerites from the surfaces of the peduncle and rachis are predominantly elongate rods and spindles. The rachis interior has similar sclerites but the peduncle interior has minute ovals. Cavernularia obesa, from the Indo-West Pacific, also lacks an axis, but possesses sclerites in the interior of the peduncle that are similar to those from the surface (according to Ktikenthal, 1915: 13; also see D’Hondt 19845: 634). Cavernularia elegans is distinguished from other southern African species of the genus by its clavate growth form, absence of an axis, and sclerites of elongate rods and spindles. Remarks on the genus Cavernularia Day et al. (1970: 17) listed Cavernularia luetkeni Kolliker, 1872, from 80-102 m depth in False Bay, near Cape Town. Since the specimen upon which this identification was made cannot be traced and since no description or figure is provided, it must be considered a dubious record. A 19-mm long specimen collected at Algoa Bay, near Port Elizabeth (SAM-—H3236), was originally identified by Tixier-Durivault as C. luetkeni. I have examined this specimen and found the polyps to contain small bifurcated calyces. I therefore identify it as a young colony of Actinoptilum molle (Kiikenthal, 1910). Tixier-Durivault (1960: 366) listed C. luetkeni from Inhaca Island, southern Mozambique. Material of C. luetkeni is not available at present for examination. Kiikenthal (1915: 14) distinguished the species by the possession of a small axis in the base of the rachis and top part of the peduncle; stalk sclerites <0,19 mm long; and rachis sclerites <0,31 mm long. SOUTHERN AFRICA PENNATULACEA 47 Approximately 18 species of the genus have been named. Of these, four were considered dubious species or synonyms by Kikenthal (1915). Most of the recognized species are from the eastern Atlantic or Indo-West Pacific, and one is from the Pacific coast of Central America and the Galapagos Archipelago. The genus is in need of revision, since clear distinctions between some species are not evident from the literature. A comparison of type material is necessary to establish the validity of certain members of the genus. Other species of the family Veretillidae Four additional veretillid species have been recorded from southern Africa, but for three of these material is not presently available to the author for examination, and in one species only partial material is available. Genus Veretillum Cuvier, 1798 Pennatula: Pallas, 1766: 177 (partim). Alcyonium Linnaeus, 1767: 342 (partim). Veretillum Cuvier, 1798: 675. Herklots, 1858: 26. Gray, 1870: 33. Kolliker, 1872a: 324. Balss, 1910: 79. Kiikenthal & Broch, 1911: 175. Kiikenthal, 1915: 11. Diagnosis Colonies cylindrical and plump or elongate and clavate. Size of axis variable, or altogether absent. Autozooids are distributed over surface of rachis; siphono- zooids are contained in longitudinal rows between the autozooids. Sclerites are small plates (<0,1 mm), unbranched, often bone-shaped, biscuit-shaped, or at least constricted in the middle to some degree. Three species from southern Europe, west coast of Africa, southern Mozambique and the Malay Archipelago. Type species. Pennatula cynomorium Pallas, 1766; Europe. Veretillum cynomorium (Pallas, 1766) Figs 38, 40 Pennatula cynomorium Pallas, 1766: 177. Alcyonium epipetrum Linnaeus, 1767: 1294. Veretillum cynomorium Kikenthal & Broch, 1911: 176, pl. 13 (fig. 1), pl. 18 (figs 32-36) (Lit.!). Kukenthal, 1915: 12, figs 12, 13. Broch, 1958: 249. Tixier-Durivault, 1960: 362. Material Material was not available for examination. Remarks According to Kiikenthal (1915: 11-12) colonies are cylindrical and plump, sausage-shaped, with rachis longer than peduncle. Siphonozooids in longitudinal rows. Sclerites are thin flat plates (<0,06 mm), usually medially constricted with ends rounded. 48 ANNALS OF THE SOUTH AFRICAN MUSEUM Broch (1958: 249-250) recorded this species from the Atlantic and Medi- terranean coasts of southern Europe, and along the west African coast as far south as Walvis Bay, Namibia; the latter record follows Ktikenthal & Broch (1911: 177). Molander (1929: 15) extended the range south to the mouth of the Orange River (border of Namibia and South Africa), 13 m depth. Tixier-Durivault (1960: 362) recorded a colony from Inhaca Island, southern Mozambique. The known depth range is 13-91 m. Veretillum leloupi Tixier-Durivault, 1960 Fig. 38 Veretillum leloupi Tixier-Durivault, 1960: 363, figs 1-2. Material Material was not available for examination. Remarks According to Tixier-Durivault (1960: 363-365), colonies are long and nar- row, clavate with stalk length equal to or longer than rachis. Siphonozooids not aligned in longitudinal rows. Sclerites are bone-shaped, not flattened, with medial constrictions and rounded ends, up to 0,1 mm in length. The species is known only from the type locality at Inhaca Island, southern Mozambique, in the western Indian Ocean; depth not recorded. Genus Cavernulina Kikenthal & Broch, 1911 Cavernularia J. A. Thompson & Simpson, 1909: 302 (partim). Cavernulina Kiikenthal & Broch, 1911: 172. Kiikenthal, 1915: 9. Diagnosis Colonies clavate. Axis present. Sclerites of the rachis are mainly rods with bilobed or branched ends, usually <0,4 mm in length. Peduncle sclerites are smooth rods and spindles, most of which are not bilobed at the ends, and are mostly <0,2 mm in length. Polyps without sclerites. Three species from the Indo-West Pacific. Type species. Cavernularia cylindrica Kikenthal & Broch, 1911; Malay Archipelago. Remarks Cavernulina cylindrica Kiikenthal & Broch, 1911, was listed without description or figures by Day et al. (1970: 17) from 33-36 m in False Bay, near Cape Town, as Cavernularia cylindrica. Since the specimen upon which this record was based is not available for examination and since this is the only False Bay record of the species known, it must be considered a dubious record. It is probable that small specimens from the Cape region, previously identified as SOUTHERN AFRICA PENNATULACEA 49 species of Cavernularia, may in fact be young colonies of Actinoptilum molle in which the bifurcate calyces are neither prominent nor conspicuous, but present nonetheless. Two species were considered valid by Ktikenthal (1915: 9): Cavern- ulina cylindrica and C. orientalis (Thomson & Simpson, 1909), both from the Indian Ocean. A third species, C. grandiflora D’Hondt, 1984, has recently been described from New Caledonia. See D’Hondt (19846: 629) for a comparison of these three species. Cavernulina sp. Fig. 38 Remarks A partial colony, in which only a 30 mm long rachis is present, recently collected from 62 m depth off the Durban region of Natal, possesses bilobed sclerites 0,1—0,25 mm in length. These sclerites are similar in shape and size to those figured by Kiikenthal (1915: 10) for C. cylindrica and by D’Hondt (19845: 628) for C. grandiflora. | therefore believe that this specimen belongs to the genus Cavernulina, but a proper identification to species is not possible due to the fragmentary nature of the material, in which the peduncle is entirely missing. The present study establishes the genus Cavernulina as a new record for southern Africa. Genus Lituaria Valenciennes in Milne Edwards & Haime, 1850 Pennatula: Pallas, 1766: 179 (partim). Veretillum: Lamarck, 1816: 420 (partim). Lituaria Valenciennes in Milne Edwards & Haime, 1850: 84. Gray, 1870: 33. Kélliker, 1872a: 135. J. A. Thomson & Simpson, 1909: 311. Balss, 1910: 78. Kiikenthal, 1915: 7. Clavella Gray, 1870: 33. K6élliker, 1872a: 144. J. A. Thomson & Simpson, 1909: 311. Balss, 1910: 78. Diagnosis Colonies clavate. Axis contained in rachis, length variable, four-angled. Siphonozooids very numerous between the autozooids. Sclerites are small biscuit-shaped plates, many of which possess short branches or denticles at the ends. Other sclerites may be crosses or rods with warty ends. Nine species of the Indo-Pacific. See Light (1921: 248) for a key to the species, and D’Hondt (1984b: 636) for a discussion of the type species. Type species. Pennatula phalloides Pallas, 1766; Indian Ocean. Lituaria valenciennesi D’Hondt, 19846 Lituaria phalloides (non Pallas, 1766) Valenciennes in Milne Edwards & Haime, 1850: 84. Kolliker, 1872a: 313. Ktkenthal, 1915: 8, fig. 6. Lituaria valenciennesi D’Hondt, 1984b: 636, fig. 9. Material Material was not available for examination. 50 ANNALS OF THE SOUTH AFRICAN MUSEUM Remarks According to Kukenthal (1915: 8), colonies are clavate with rounded distal ends. Polyps mostly without sclerites. Sclerites of the rachis and stalk are small flat plates up to 0,1 mm long, many with terminal denticles. The species is recorded from Mozambique (undesignated station), the Malay Archipelago, the Andaman Islands, and Sri Lanka; shallow sublittoral. Family Echinoptilidae Hubrecht, 1885 Colony digitiform or cylindrical. Axis absent. Rachis generally longer than peduncle. Colonies radially (Fig. 11B) to bilaterally symmetrical (Fig. 9B). Autozooids with distinctive, non-retractile, bifurcated calyces that are heavily armed with sclerites (Fig. 9A, E). Sclerites mainly smooth three-flanged rods or spindles, ovals and platelets (Fig. 13). Two genera from the Indo-West Pacific to south-western Africa; shallow sublittoral to 835 m. Genus Echinoptilum Hubrecht, 1885 Echinoptilum Hubrecht, 1885: 512. Balss, 1910: 37. Kikenthal & Broch, 1911: 196. Kikenthal, 1915: 18. Actinoptilon Kikenthal, 1910: 54 (partim). Diagnosis Colony digitiform, commonly curved. Concave side with a longitudinal groove or furrow, and inverted V-shaped region at the base of the rachis devoid of polyps. This furrow may extend full length of rachis in small specimens (less than 40 mm). Proximal portion of rachis therefore bilaterally symmetrical. Distal portion of rachis may be radially symmetrical with polyp-bearing region con- tinuous in larger specimens (over 40 mm). Autozooids with prominent bifur- cated calyces, giving rachis a spiny or prickly appearance. Two terminal teeth of calyx deltoid to needle-like. Larger autozooids may be concentrated in two lateral series of longitudinal rows in colonies less than 40 mm long. Autozooids capable of total retraction into calyces. Siphonozooids numerous at base of autozooids; also with non-retractile, bifurcated calyces. Sclerites numerous; those of calyx are long three-flanged spindles, peduncle with ovals or three-flanged rods and spindles. A genus of six species, Indo-West Pacific— Hawaii to eastern Africa. Type species. Echinoptilum macintoshii Hubrecht, 1885; Japan. Echinoptilum macintoshii Hubrecht, 1885 Figs 1H-I, 5, 6, 7A—C, 8D-F, 39 Echinoptilum macintoshii Hubrecht, 1885: 512, text-figs 1-3, pl. 30 (figs 1-8), pl. 31 (figs 9-14). Echinoptilum macintoshi: Nutting, 1908: 561; 1912: 48. Balss, 1910: 38, pl. 2 (figs 1-3), pl. 5 (figs 1, 3-5). Kikenthal, 1915: 19. Utinomi, 1961: 221. Echinoptilum Macintoshii: Kikenthal & Broch, 1911: 197. Echinoptilum M’Intoshii: Hickson, 1916: 57-62. Echinoptilum mackintoshi: J. A. Thomson & Rennet, 1927: 121, pl. 9 (fig. F). Bayer, 1956: 226. SOUTHERN AFRICA PENNATULACEA ai Material SAM-H3683, off Mbotyi, Transkei (31°32’S 29°50’E), 150-170 m, August 1981, 2 colonies, dredge; Natal Museum, R.V. Meiring Naude. SAM—H3685, off Port St. Johns, Transkei (31°39’S 29°35’E), 50-70 m, 14 August 1981, 1 colony, dredge; Natal Museum, R.V. Meiring Naude. SAM-—H3819, off Mgazi River, er a av W a aoe We ee | BP AS S| Za / 4 > t. ; A Sy AM Ss IF eS it Ge f | S i SS = ) A i WA 3 ag ; ANN I \ | ey C ‘ f 1 | ( WI J a F| r II ii | 1) \ Oe y AVA i ny mld | i 2 ee A mn Wi Hl 2 $2 48 Waal File ee Sch Eee al dtd dhe ge ue as ve ieee PAY ‘va > PEE 6 8 oo Doers ate DEE eS a Fig. 5. Echinoptilum macintoshii. A. Entire colony, lateral view. Scale = 5mm. B. Five autozooids with eight siphonozooids; sclerites shown in calyx of top autozooid only. Scale = 1mm. C. Single tentacle (length 1,3 mm) showing alignment of sclerites. 52 ANNALS OF THE SOUTH AFRICAN MUSEUM Transkei (31°43,7'S 29°31,9'E), 140-145 m, 4 July 1985, 8 colonies, dredge; G. C. Williams, R.V. Meiring Naude. Description Specimens examined are 10-84 mm in length. Colonies curved with ventral groove on concave side. Calyx teeth conical to deltoid, 0,4—0,8 mm in length. Calyx length approximately twice the width. Entire calyces 1,2—2,3 mm in length. Siphonozooids have two broad teeth, often with obtuse tips. Sclerites of the calyces are smooth three-flanged spindles 0,17—1,07 mm in length (Fig. 6A); aboral surface of each tentacle with a conspicuous longitudinal medial band of red-orange three-flanged rods, which extend into the base of the Fig. 6. Echinoptilum macintoshii. A. Calyx sclerites. B. Sclerites from surface and interior of rachis. C. Tentacular sclerites. D. Sclerites from surface of peduncle. E. Sclerites from interior of peduncle. Scales: A-B, D (left) = 0,2 mm; C, E (right) = 0,1 mm. SOUTHERN AFRICA PENNATULACEA Sa ; | : : :. | 4 i A Gt OR CR i esc ibe Fig. 7. Scanning electron micrographs of sclerites. A-—C. Echinoptilum macintoshii. From surface of peduncle: 0,33 mm; 0,31 mm; 0,16 mm.D-H. Echinoptilum echinatum. D-E. From calyx: 0,17 mm; 0,55 mm. F-—H. From surface of peduncle: 0,11 mm; 0,11 mm; 0,09 mm. 54 ANNALS OF THE SOUTH AFRICAN MUSEUM anthocodia. These are all arranged parallel to the axis of each tentacle and form a band 2-3 sclerites in width. Similar but colourless rods are found at various oblique angles to the axis, near the bases of the pinnules (Fig. 5C). The rods of the tentacles vary in length from 0,13—0,17 mm (Fig. 6C). Bare areas of the rachis surface, aS well as the rachis interior, possess three-flanged spindles and rods 0,13-0,50 mm in length (Fig. 6B). Sclerites from the surface of the peduncle are longitudinally grooved (but not three-flanged) spindles or ovoid rods, 0,07- 0,47 mm in length, often with undulate margins and tapering ends (Figs 6A—C, 6D). The tissues of the peduncular interior possess scattered sclerites of minute ovals 0,008—0,020 mm in length (Fig. 6E). Rachis rust-orange to light reddish-brown. Peduncle and naked region of rachis whitish or cream. Calyces reddish-brown, often with margins edged in white. Colour of colony permanent, originating within the sclerites. Distribution (Fig. 39) Transkei coast, Japan and Hawaii. This study establishes a new record for the African coast and extends the range of the species to the western Indian Ocean. Colonies from south-eastern Africa are known from 50-170 m depth. Hubrecht (1885: 512) reported the type locality as the Sea of Japan, 130 m in depth. Nutting (1908: 561) recorded the species from the Hawaiian Archipelago, 225-258 m, and again off Japan, 150 m (Nutting 1912: 48). Remarks Echinoptilum macintoshii is distinguished from other members of the genus by having sclerites in the tentacles, and triangular calyx teeth <0,8 mm long. Echinoptilum echinatum (Kikenthal, 1910) Figs 1G, 7D—H, 8G-H, 9, 10, 39 Actinoptilon echinatum Kiukenthal, 1910: 57. Echinoptilum echinatum Kikenthal & Broch, 1911: 197, pl. 13 (fig. 3), pl. 18 (figs 38, 39), pl. 19 (figs 40-43). Kikkenthal, 1915: 19 (fig. 21). Hickson, 1916: 57-62. Material SAM-—H1595, off Durban (30°20’S 30°51’E), 406 m, 23 July 1964, 1 colony, dredge; Division of Sea Fisheries. SAM—H3680, off Mendu Point, Transkei (32°23'S 29°00’E), 450 m, 13 July 1984, 1 colony; G. C. Williams, R.V. Meiring Naude. SAM-H3681, off Qolora River, Transkei (32°48’S 28°37'E), 510 m, 14 July 1904, dredge; G. C. Williams, R.V. Meiring Naude. SAM—H3682, off Qora River, Transkei (32°35’S 28°49’E), 450-460 m, 14 July 1984, dredge; G. C. Williams, R.V. Meiring Naude. SAM—H3684, off Qora River, Transkei (32°34’S 28°50'E), 400-420 m, 12 July 1984, dredge; G.C. Williams, R.V. Meiring Naude. SAM-—H3820, off Mbashe River, Transkei (32°23,6’S 28°59,2’E), 350 m maximum, 6 July 1985, 1 colony, dredge; G. C. Williams, R.V. Meiring Naude. SOUTHERN AFRICA PENNATULACEA 55 Description Specimens examined are 30—80 mm in length. Colonies curved with ventral groove on concave side. Calyx teeth lanceolate to needle-like, 1,0-1,6 mm in length. Total calyx length (approximately 3 mm) is more than twice and often three times the width. Siphonozooids possess two narrow teeth with acute tips. Sclerites of the calyces are narrow, three-flanged spindles 0,14—0,86 mm long (Figs 7D-E, 10A). The aboral surface of each tentacle has a medially placed longitudinal row of 1-20 sclerites, arranged more-or-less parallel to the axis of the tentacles (Fig. 9C-D). These are three-flanged rods 0,10—-0,16 mm in length (Fig. 10C). Bare regions of the surface of the rachis, as well as the interior of the rachis, possess elongate three-flanged rods 0,06—0,35 mm in length (Fig. 10B). The peduncular surface sclerites are oval finger-biscuits with longitudinal striation (but not three-flanged) 0,04—0,15 mm in length (Figs 7F—H, 10D). The tissues of the interior of the peduncle possess minute ovals 0,009-0,014 mm in length (Fig. 10E). Calyces and rachis are brick red or deep reddish-purple to cream. Peduncle, as well as the naked longitudinal track on the concave side of the rachis are whitish to cream, often with reddish longitudinal lines. Colour permanent due to pigmentation of the sclerites. Distribution (Fig. 39) Transkei and Natal coasts to Somalia. South-east African specimens from 350-510 m depth. Kiikenthal & Broch (1911: 197) reported the type locality as off the Somalia coast at 628 m depth. The present study extends the range of the species southward along the coast of east Africa, and represents a new record from southern Africa. The two south-east African species of Echinoptilum apparently show divergent bathymetric ranges: E. macintoshii has a shallow-water distribution, being encountered at depths of 130-258 m, whereas E. echinatum is presumably a deeper-water form, known only between 350 m and 628 m in depth. Remarks Echinoptilum echinatum is distinguished from other species of the genus by having sclerites in the tentacles, and lanceolate calyx teeth up to 1,6 mm long. Remarks on the genus Echinoptilum Four other species of Echinoptilum were described by Hickson (1916) from the Malay Archipelago: E. elongatum, E. minimum, E. roseum, and E. asperum. These species have not been recorded outside their respective type localities. Hickson (1916: 62) presented a key to the six described species. Echinoptilum macintoshii is characterized by having pronounced curvature, spicules in the tentacles, and small calyxes, whereas E. echinatum is also described as being curved with spicules in the tentacles, but with large calyces. The southern African material agrees with Hickson’s descriptions of these two species, as well as the 56 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 8. A-C. Actinoptilum molle colonies. A. 70 mm. B. 65 mm. C. 85 . D-F. Echinoptilum macintoshii. D.25 mm. E.45 mm. F. 35 mm. G-H. Echinoptilum echinatum. G. 40 mm. H. 55 mm. SOUTHERN AFRICA PENNATULACEA 57 eS SSS = = ~—: cS A Fig. 9. Echinoptilum echinatum. A. Entire colony, lateral view. Scale = 5mm. B. Ventral view of colony showing naked tract. Scale = 5 mm. C-—D. Two tentacles, showing variation in sclerite density and placement. Each tentacle is 2,0 mm in length. E. Single autozooid, partially retracted; calyx sclerites shown, five siphonozooids at base. Scale = 1 mm. F. Single bifurcated calyx of a retracted autozooid; sclerites omitted. Scale = 1 mm. 58 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 10. Echinoptilum echinatum. A. Calyx sclerites. B. Sclerites from surface and interior of rachis. C. Tentacular sclerites. D. Sclerites from surface of peduncle. E. Sclerites from interior of peduncle. Scales: A, B (left) = 0,2 mm; C, D, E (right) = 0,1 mm. detailed descriptions of Hubrecht (1885: 512) and Kiikenthal & Broch (1911: 197). The distinctions between the various groups may not be consistent or marked enough to warrant specific separation. Hickson (1916: 58) raised the possibility that the genus Echinoptilum is monospecific and that the six described species may in fact represent one highly variable and widespread species. A comparative examination of type specimens and other known material is needed. However, the differences between E. macintoshii and E. echinatum, based upon compari- son of the available southern African material, are marked and consistent. Genus Actinoptilum Kikenthal, 1911 Cavernularia (non Valenciennes in Milne Edwards & Haime, 1850) Hickson, 1900: 89, 92. Actinoptilon Kiikenthal, 1910: 56. Actinoptilum Kikenthal & Broch, 1911: 201. Kiikenthal, 1915: 20. Actinoptinum Day et al., 1970: 17. SOUTHERN AFRICA PENNATULACEA 59 Diagnosis Colony cylindrical with rachis radially symmetrical. Polyps evenly distributed and completely covering surface of rachis. Calyces of autozooids usually less than 3 mm in length, bifurcate. Terminal teeth short and conical to deltoid or broadly lanceolate and acute. Inside of each tooth often with a fleshy rounded pro- tuberance near the apex. A monospecific genus restricted to southern Africa and possibly Madagascar. Type species. Actinoptilon molle Kikenthal, 1910; South Africa. Actinoptilum molle (Kikenthal, 1910) Figs 1A-C, 8A-C, 11-13, 39 Cavernularia obesa (non Valenciennes in Milne Edward & Haime, 1850) Hickson, 1900: 92, pl. 3 (figs A—B). | Cavernularia elegans (non Herklots, 1858) Hickson, 1900: 89. Actinoptilon molle Kiikenthal, 1910: 57 (partim). Tixier-Durivault, 1960: 366. Tixier-Durivault & D’Hondt, 1974: 259. Actinoptilum molle Kikenthal & Broch, 1911: 201, pl. 13 (figs 3-4), pl. 18 (fig. 37). Kiikenthal, 1915: 20, figs 22-24. J. S. Thomson, 1915: 3. 1924: 81. Hickson, 1916: 41. Molander, 1929: 15. Tixier-Durivault, 1954: 628. Broch, 1958b: 251. Actinoptilum molle var. maculatum J. S. Thomson, 1915: 5. Actinoptilum molle var. intermedium J. S. Thomson, 1915: 7. Actinoptilum molle var. zonatum J. S. Thomson, 1915: 8. Actinoptinum molle Day et al., 1970: 17. Material SAM-H3254, Port Elizabeth Harbour (33°58’S 25°38’E), 20—25 m, February 1984, 2 colonies, small dredge; W. R. Liltved. SAM—H3263, Algoa Bay (33°59'S 25°42'E), 20m, February 1984, 2 colonies, SCUBA; W. R. Liltved. SAM- H3164, off Jeffrey’s Bay (34°00’S 25°10’E), 30m, August 1980, 6 colonies, dredge; W. R. Liltved, R.V. T.B. Davie. SAM-—H3272, False Bay (34°09’S 18°27'E), 26 February 1984, 12m, 1 colony, small dredge; W. R. Liltved. SAM-—H3339, Algoa Bay (33°59’S 25°42'E), 16m, 17 May 1984, 3 colonies, SCUBA; G. C. Williams. Description Colonies examined range in length between 12 mm and 240 mm. Colony cylindrical, plump, sausage-shaped. Peduncle thick, tapering gradually, usually one-fifth to one-third total colony length. Rachis tapering gradually to a rounded apex. Rachis entirely covered with dimorphic polyps; thus the polyp-bearing surface of the rachis is radially arranged. Siphonozooids crowded between bases of autozooids with minute bifurcated calyces. Rachis surface often with longi- tudinal lines between columns of autozooids. Calyces of autozooids usually not conspicuous, bifurcate, non-retractile. Autozooids capable of total retraction into calyces. Calyx teeth variable in shape and length. 60 ANNALS OF THE SOUTH AFRICAN MUSEUM a 2 Met SEES SSF BSR ACS S 6 X89 Gh aS 23 FRPRERIEFIERS oy = BS ese APY PAA WVATWNHY_Y 7U® WINNT IIE LSS ep ceed Ra Sree iabe tak saben apts Fig. 11. Actinoptilum molle. A. Expanded colony in situ. Scale = 30 mm. B. Single autozooid and five siphonozooids. C. Three autozooids retracted into calyces, and numerous siphono- zooids. D. Two autozooid calyces with four smaller siphonozooids. Scales: B-D = 0,8 mm. E. Expanded autozooid. Scale = 1,8 mm. F. Preserved colony with retracted polyps. Scale = 10 mm. SOUTHERN AFRICA PENNATULACEA 61 Sclerites of the calyx are smooth three-flanged spindles 0,18—0,85 mm long (Figs 12A, 13A—D). Anthocodia (distal part of polyp that contains the tentacles, mouth, and neck zone) with small ovals or platelets in neck zone and base of tentacles, 0,02—0,04 mm in length (Figs 12B, 13E). The surface of the rachis has three-flanged spindles and rods 0,22—0,42 mm in length (Figs 12C, 13F, H-I). The tissues of the interior of the rachis possess similar sclerites 0,22—0,41 mm in length (Fig. 12D). The surface of the peduncle has ovoid finger biscuit-shaped sclerites with longitudinal surface striations 0,03-—0,20 mm long (Figs 12E, 13G). The tissues of the peduncular interior possess minute ovals 0,005—0,020 mm in length (Fig. 12F). Po()O0 20860 09 AQO Fig. 12. Actinoptilum molle. A. Calyx sclerites. B. Sclerites from anthocodial wall. C. Scle- rites from rachis surface and interior. D. Sclerites from surface of peduncle. E. Sclerites from interior of peduncle. Scales: A, C (left) = 0,2 mm; B, D (right) = 0,1 mm; E = 0,02 mm. 62 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 13. Scanning electron micrographs of sclerites of Actinoptilum molle. A-—D. From calyx: 0.41 mm; 0,21 mm; 0,45 mm; 0,44 mm. E. From tentacle: 0,04 mm. F,H,I. From surface of rachis: 0,2 mm; 0,15 mm; 0,2 mm. G. From surface of peduncle: 0,07 mm. SOUTHERN AFRICA PENNATULACEA | 63 Colour highly variable; individual colonies may be entirely yellow, white to cream, reddish-brown, pinkish or mauve; or colonies with rachis deep-purple to reddish-purple, and peduncle yellow, white, pinkish or brownish. Colour permanent, originating within the sclerites. Distribution (Fig. 39) This species is presumably a southern African endemic, known from off Cape Columbine on the west coast of South Africa to Inhaca Island, Mozambique, at depths ranging from 12 m to 333 m. However, Tixier-Durivault & D’Hondt (1974: 259) recorded, without description or figures, seven colonies from north- eastern Madagascar between 200 m and 605 m in depth. The species is perhaps the most frequently encountered pennatulacean in South African coastal waters. Observations A 60-mm long live colony from False Bay (SAM-—H3272), which was kept in a bucket of sea-water at night, was observed to luminesce vigorously when agitated; the rachis sparkled throughout with minute pinpoints of green light. SCUBA divers have observed this species inhabiting rocky reefs off Cape Agulhas at 48 m depth. A colony was observed in a small depression in rocky substratum that was filled with fine sediment (W. R. Liltved pers. comm.). Divers have also observed extensive sandy areas at 20 m depth in False Bay with a density of 4—5 colonies per square metre (P. Williams pers. comm.). Remarks Thomson (1915: 1-13) named three varieties of Actinoptilum molle based primarily on differences in external coloration: A. molle var. maculatum, A. molle var. intermedium, and A. molle var. zonatum. Because of the wide range of colour variability observed throughout the geographic extent of the species and many gradations between various colonies or populations, I believe that formal subspecies or varieties should not be designated. This species can be regarded as yet another example of the extreme range of variability exhibited by sedentary benthic organisms such as octocorallians. Thomson (1915: 8) observed a slight trace of a groove on the rachis of four white colonies that he referred to A. molle var. intermedium. A conspicuous groove or furrow on the concave side of the rachis is considered diagnostic of the genus Echinoptilum. However, according to Thomson, the four specimens show typical characteristics of Actinoptilum, such as smaller inconspicuous calyces and numerous polyps arranged almost completely radially, as well as the form, dimensions, and distribution of sclerites being similar to those described by Kukenthal for A. molle. Calyx length is also quite variable, ranging from 1 mm to 3 mm. In some colonies the calyx teeth are only visible with the aid of a hand lens or dissecting microscope, whereas in others they are very conspicuous and easily seen with the unaided eye. 64 ANNALS OF THE SOUTH AFRICAN MUSEUM Kiikenthal (1915: 20) stated that the polyps are without sclerites. The density of sclerites in the anthocodiae is variable; in a single colony some anthocodiae may contain sclerites, whereas others may contain none. The number of ovals present in a particular anthocodia is also quite variable. Actinoptilum molle is distinguished from other southern African sea pens by having radial symmetry of the rachis, in which the polyps are distributed evenly on all sides, and the possession of generally inconspicuous calyx teeth. Family Kophobelemnidae Gray, 1860 Colonies clavate (club-shaped). Axis present. Polyps arranged bilaterally along rachis; rachis may show a tendency toward radial symmetry. Autozooids large, without calyces. Siphonozooids with calyces, at base of autozooids. Rachis with sclerites of three-flanged or tuberculate spindles, or plate-like, biscuit-shaped or bone-shaped rods. Four genera from the Atlantic, Mediterranean and Indo-Pacific: Mesobelem- non Gravier; 1907; Sclerobelemnon KO6lliker, 1872a; Kophobelemnon Asbjgrn- sen, 1856; and Malacobelemnon Tixier-Durivault, 1966a. The present study represents the first record of the family for southern Africa. Genus Kophobelemnon Asbj@rnsen, 1856 Kophobelemnon Asbj@rnsen, 1856: 81. Kukenthal & Broch, 1911: 219. Kiikenthal, 1915: 29 (Lit.!). Bathyptilum Kolliker, 1872: 200. Gunneria Danielssen & Koren, 1884: 58. Diagnosis Colony clavate, elongate, stiff. Distal end slightly expanded and rounded, knob-like, or slightly pointed at the tip. Rachis and stalk about equal in length. Axis thin, often slightly quadrangular in transverse section. Sclerites are mostly three-flanged spindles that may possess tubercles. A genus of eight to ten species. Atlantic, Mediterranean, and Indo-Pacific; mostly deep sea, 40—4 400 m. Type species. Pennatula stellifera Muller, 1776; northern Atlantic. Kophobelemnon stelliferum (Miller, 1776) Figs 14-16, 40 Pennatula stellifera Miller, 1776: 255. Kophobelemnon stelliferum Kikenthal & Broch, 1911: 224, figs 54-60. Kiikenthal, 1915: 29, figs 40-41 (Lit.!). Material SAM-—H3707, off Skeleton Coast, Namibia (20°59’S 12°23'E), 486-605 m, 20 August 1982, 1 colony, bottom trawl; Sea Fisheries Research Institute, Crab Survey. SAM-—H4026 off Skeleton Coast, Namibia (c. 21°00'S 12°00’E), SOUTHERN AFRICA PENNATULACEA 65 nO a is Sis ea aa PDP Aye a =A _ 3 ¥ > Ss crate a. ss weeteas a gtk = a th aad - PATO IG a ae BPR A s t= See $5 <> SVs. Rea ot. SS > WAIT ST 2y WN. ‘> ay Th. Pe mr es AA? tm OLE aie Fig. 14. Kophobelemnon stelliferum. A. Entire colony. Scale = 50mm. B. Sclerites from surface of rachis. C. Sclerites from wall of an autozooid. D. Sclerites from surface of peduncle. Scale: B—D = 0,2 mm. E. Detail of single autozooid and seven siphonozooids. Scale = 8 mm. ANNALS OF THE SOUTH AFRICAN MUSEUM wien dba pipe: Libre iitpepiilliin tists wsthjpuilelgey dle ti Fig. 15. Scanning electron micrographs of sclerites of Kophobelemnon stelliferum. A-—C. From rachis: 0,32 mm; 0,33 mm; 0,59 mm. D. Tip of rachis sclerite: 0,08 mm. E—F. From peduncle: 0,26 mm; 0,24 mm. G. Tip of peduncle sclerite: 0,1 mm. SOUTHERN AFRICA PENNATULACEA 67 Fig. 16. Kophobelemnon stelliferum. Scanning electron micrographs of sclerites from base of peduncle: A. 0,14 mm; B. 0,14 mm; C. 0,16 mm; D. 0,13 mm; E. 0,19 mm. 450-650 m, February 1987, 10 colonies, bottom trawl, R. Melville Smith, Sea Fisheries Research Institute, Crab Survey. Description The colony, SAM—H3707, is 340 mm in length, tip of peduncle missing. Colony elongate, slender, curved. Rachis expands gradually toward distal end, which is obtuse and knob-like. Autozooids 10-15 mm long, c. 50 in number, in two longitudinal rows along the concave face of the rachis. Convex side of rachis is free of autozooids but is quite rough and has a granular appearance due to the presence of the spiculated calyces of numerous siphonozooids. These zooids are also found at the base of the autozooids. Colony densely spiculated throughout. Sclerites are tuberculated three-flanged spindles varying in length from 0,1 mm to 0,6 mm. The longer sclerites from the polyps and rachis are tuberculate only at the ends. The shorter spindles of the peduncle may have tubercles throughout the entire length but mostly concentrated at the ends. Tubercles rounded and prominent. The appearance, size, and distribution of these sclerites agrees with those figured by Ktkenthal & Broch (1911: 228) for the species. Colour in alcohol is brownish-grey. Distribution (Fig. 40) Atlantic Ocean from the Norwegian Sea to Namibia, and at scattered locations in the Pacific. In southern Africa it is known only from off the Skeleton 68 ANNALS OF THE SOUTH AFRICAN MUSEUM Coast of Namibia (450-650 m). This study establishes a new record for southern Africa. The type locality is the Grand Banks, south-east of Newfoundland. Remarks The form of the sclerites of this specimen agrees with those figured by Kikenthal & Broch (1911: 228). They are slender and many have knobby protuberances at each end. Kophobelemnon heterospinosum Kikenthal, 1910, has similarly shaped sclerites but without prominent tuberculation. Kophobelem- non affine Studer, 1894, has tubercles on the peduncular sclerites, but the colonial sclerites as a whole are more stout, being proportionally wider than those of K. stelliferum. Kophobelemnon stelliferum is distinguished from other southern African sea pens by its elongate clavate appearance with rachis that gradually widens distally, 20-40 polyps arranged biserially, and peduncular sclerites that are rods with knobby tubercles. Kophobelemnon sp. Fig. 38 Remarks A 50-mm long colony, possessing only two antozooids, has recently been collected from off Durban, Natal, at 820m in depth. Four species are characterized by having a reduced number of autozooids: Kophobelemnon hispidum Nutting, 1912; K. pauciflorum Hickson, 1916; K. molanderi Pasternak, 1975b nom. nov. (= K. pauciflorum Molander, 1929); and K. macrospinosum J. A. Thomson, 1927 (= K. biflorum Pasternak, 1960). The unavailability of pertinent literature sources precluded the making of necessary comparisons. Family Anthoptilidae Kolliker, 1880 Colonies long and whip-like, bilaterally symmetrical with polyps in two longitudinal series. Polyps in oblique rows, non-retractile, calyces absent. Siphonozooids are numerous on the rachis situated between the oblique rows of autozooids. Sclerites are absent except for minute ovals of the stalk interior. Axis present; round to somewhat quadrangular in cross section. Peduncle stout to robust, not more than one-fifth total colony length. A monogeneric family of the Atlantic Ocean and Arctic region 238-2 744 m. Genus Anthoptilum Kolliker, 1880 Virgularia (non Lamarck, 1816) Verrill, 1879: 239. Anthoptilum Kolliker, 1880: 13. Jungersen, 1904: 65. Kikenthal & Broch, 1911: 232. Kiikenthal, 1915: 32 (Lit!). Hickson, 1916: 138. Grasshoff, 1982b: 954. Benthoptilum Verrill, 1885: 510. Stephanoptilum Roule, 1905: 455. Thesioides J. A. Thomson & Henderson, 1906: 91. SOUTHERN AFRICA PENNATULACEA 69 Diagnosis The characters of the family. Two species of the Atlantic and Arctic. Type species. Virgularia grandiflora Verrill, 1879; northern Atlantic. Anthoptilum grandiflorum (Verrill, 1879) Figs 17, 40 Virgularia grandiflora Verrill, 1879: 239. Anthoptilum thomsoni KOlliker, 1880: 13, pl. 4 (figs 16-18). J. A. Thomson & Rennet, UGS NEB). Anthoptilum grandiflorum: Jungersen, 1904: 66. Hickson, 1904: 233, pl. 8 (fig. 14). Kiikenthal & Broch, 1911: 233, pl. 14 (figs 5-7), pl. 19 (figs 44-46) (partim). J. S. Thomson, 1915: 17. Kikenthal, 1915: 32, fig. 44 (partim) (Lit!). Tixier-Durivault, 1954: 629. Grasshoff, 1982a: 741, 750; 1982b: 954, fig. 35 (Lit!). Anthoptilum sertum: Kikenthal, 1915: 33. Material SAM-H1213, off Lion’s Head, Cape Peninsula (33°55’S 17°55’E), 313 m, 16 March 1900, 2 colonies, large dredge, S.S. Pieter Faure survey, station JE AASKO Description Colonies examined are 400-1 050 mm in length. Polyps 5-10 per row, 8- 25 mm in length. Adjacent polyps may be joined proximally, forming a common base for an entire oblique row, or only two or three polyps in a row may be joined, this group being free from the rest of the polyps in the row. Colour often bright red in life, yellowish brown in alcohol. Distribution (Fig. 40) In southern Africa, known from off the west coast of the Cape Peninsula, at 238-348 m. The species is also recorded from the western, eastern, and southern Atlantic, as well as the Arctic region. Recorded depth range is 238-2 500 m. Remarks This species is distinguished from other southern African pennatulaceans by its whip-like appearance, lack of calyces and sclerites, and with adjacent polyps united at their bases into groups of 5-10 polyps forming short oblique rows. Family Funiculinidae Gray, 1870 Elongate, whip-like colonies with polyps arranged biserially, often covering the lateral and ventral surfaces of the rachis. Polyps are retractile into tubular, eight-toothed calyces, which may be arranged irregularly or in somewhat oblique rows. Siphonozooids few. Sclerites numerous; three-flanged spindles, rods, and ovoid plates. A monogeneric family of cosmopolitan distribution. i ANNALS OF THE SOUTH AFRICAN MUSEUM 70 res Wh ——— ee a fi ; sls A oS ae } 7 —~ / ' Y Ths 7 ; ‘ LLG ee, C= eee Lm NS SS ee reget = 50 mm. B. Detail of group of autozooids with numerous siphonozooids at base. Scale = 10 mm. Fig. 17. Anthoptilum grandiflorum. A. Entire colony. Scale SOUTHERN AFRICA PENNATULACEA 71 Genus Funiculina Lamarck, 1816 Funiculina Lamarck, 1816: 422. Herklots, 1858: 8. Kolliker, 1872: 250. Kiikenthal & Broch, 1911: 241. Ktikenthal, 1915: 34 (Lit!). Manuel, 1981: 53. Leptoptilum Kolliker, 1880: 27. Trichoptilum Kolliker, 1880: 29. Diagnosis The characters of the family. Three species are recognized worldwide. Type species. Pennatula quadrangularis Pallas, 1766. Funiculina quadrangularis (Pallas, 1766) Fig. 38 Pennatula quadrangularis Pallas, 1766: 372. Funiculina quadrangularis Jungersen, 1904: 49. Kikenthal & Broch, 1911: 243, pl. 15 (fig. 12) (Lit.!). Kiikenthal, 1915: 34, figs 45-46. J. S. Thomson, 1915: 15, figs 45-46. Pasternak, 1975b: 104. Manuel, 1981: 53, fig. 14. Grasshoff, 1981: 227. Material Material was not available for examination. Distribution (Fig. 38) Cosmopolitan, recorded up to 2 300 m in depth. Remarks Thomson (1915: 15) recorded one 490-mm specimen from 146-183 m off Cape Vidal, Natal (28°10’S 32°40'E). According to Manuel (1981: 53), Kikenthal (1915: 34), and the above reference, the species is characterized by having a dis- tinctly quadrangular axis covered with a thin coenenchyme. Colonies may reach 2 199 mm in length. Distal portions of colonies are flexible and often tend to curve or coil spirally. Pedunclular interior with ovoid plate-like sclerites. Calyces with elongate, three-flanged spindles (up to 9,63 mm in length) forming eight, pointed terminal teeth and a transverse band of similar sclerites sur- rounding the base of the teeth. The surface of the rachis possesses three- flanged rod-like sclerites. Family Protoptilidae Kolliker, 1872 Colonies elongate, slender, bilateral. Autozooids arranged in one to three longitudinal series, retractile into spiculate calyces. Sclerites are numerous, three-flanged, mostly spindles. Axis prominent and rounded in cross-section. Three genera of the Atlantic and Indo-Pacific: Protoptilum Kolliker, 1872a; Distichoptilum Verrill, 1882, and Helicoptilum Nutting, 1912. The present study represents the first records of the family for southern Africa. WD ANNALS OF THE SOUTH AFRICAN MUSEUM Genus Distichoptilum Verrill, 1882 Distichoptilum Verrill, 1882: 362. Kiikenthal, 1915: 39. Hickson, 1916: 101. Juncoptilum J. A. Thomson & Henderson, 1905: 555. Diagnosis Polyps arranged in two longitudinal series. No polyps on ventral surface. Axial side of calyces appressed to lateral margin of rachis. The siphonozooids are two per autozooid, placed on the rachis directly above the mouth of each autozooid, one on the dorsal side and one on the ventral side of the calyx. A monospecific genus. Type species. Distichoptilum gracile Verrill, 1882; eastern North America. Distichoptilum gracile Verrill, 1882 Figs 18, 19H-I, 40 Distichoptilum gracile Verrill, 1882: 362. Jungersen, 1904: 62. J. A. Thomson & Henderson, 1906: 87. Kikenthal, 1915: 39 (Lit.!). Pasternak, 1975b: 106. Grasshoff, 1982a: 738, 741-742, map 8; 1982b: 955, figs 38-39. Juncoptilum Alcocki J. A. Thomson & Henderson, 1905: 555. Material SAM-—H3163, west of Table Bay, western Cape Province (33°49’S 16°30’E), 2744 m, 27 August 1959; many fragments of several colonies, 15-foot beam trawl, Division of Sea Fisheries. Description Colonies examined are perhaps up to 1 000 mm in length and 4 mm in width. Polyps in two longitudinal series, alternate or sub-opposite, retractile into promi- nent calyces. Distance between polyps along rachis 4—7 mm. Calyces 2—4 mm long. Calyces adhere to rachis along their inner proximal portions and face upward. Calyces usually with two terminal teeth, but these are sometimes obscure in autozooids of the lower rachis. Sclerites of the calyces are large spindles (up to 0,5 mm in length), three-flanged and smooth. Peduncular sclerites are shorter and inconspicuously three-flanged. Calyces orange, rachis yellowish to cream. Distribution (Fig. 40) Previously recorded from the eastern Pacific, Indian, and northern Atlantic Oceans, 793-4 300 m. This work represents the first record of the genus and species in southern Africa, and extends the range in the Atlantic southward from Europe and eastern North America. Remarks A considerable amount of variation is evident in this species. Kukenthal (1915: 40) stated that the calyces possess six terminal teeth and the length of calyx sclerites is up to 1,0 mm. Jungersen (1904: 62) regarded number of calyx teeth as SOUTHERN AFRICA PENNATULACEA 73 Fig. 18. Distichoptilum gracile. A. Entire colony. Scale = 50 mm. B. Detail of colony from three different parts of rachis showing calyces. Scale = 2mm. C. Sclerites from calyx. Scale = 0,2 mm. D. Sclerites from peduncle. Scale = 0,1 mm. 74 ANNALS OF THE SOUTH AFRICAN MUSEUM i | Wty Sec 2 “gail Fig. 19. Scanning electron micrographs of sclerites. A-—G. Halipteris africana. A-D. From peduncle: 0,07 mm; 0,05 mm; 0,04 mm. E. From tentacle: 0,11 mm. F. From calyx: 0,58 mm. G. Tip of calyx sclerite: 0,2 mm. H-I. Distichoptilum gracile. From calyx: 0,36 mm; 0,44 mm. J-N. Umbellula thomsoni. J-L. From anthocodial neck zone: 0,05 mm; 0,05 mm; 0,06 mm. M-N. From tentacle: 0,2 mm; 0,24 mm. SOUTHERN AFRICA PENNATULACEA WD a matter of the degree of contraction of the polyps, as some highly contracted polyps show two lobes, whereas partially contracted polyps often exhibit six distinct teeth. He also stated that the length of calyx sclerites varies between 0,528 and 0,800 mm. The calyces of the southern African material show two large teeth only and the sclerites vary from 0,28—0,50 mm in length. Distichoptilum gracile is distinguished from other southern African sea pens by its thin whip-like appearance, small polyps retractile into permanent calyces that are appressed close to the surface of the rachis and are arranged biserially in two longitudinal rows. Family Scleroptilidae Jungersen, 1904 Polyps usually situated in pairs or groups of three along rachis, separated by intervals of bare rachis. Axis thin and delicate, extending entire length of colony. Siphonozooids few and scattered on rachis. Calyces absent. Sclerites present or absent. A family of two genera from the North Atlantic and Indo-Pacific. Genus Scleroptilum Kolliker, 1880 Scleroptilum KO6lliker, 1880: 30. Balss, 1910: 24. Kikenthal & Broch, 1911: 265. Kiikenthal, 1915: 43. Diagnosis Sclerites (rods or three-flanged spindles) are present in the walls of the autozooids. A genus of at least two or three species from the northern Atlantic and Indo-Pacific. Type species. Scleroptilum grandiflorum Kolliker, 1880; Japan. Scleroptilum sp. Fig. 38 Remarks Five colonies, 35-50 mm in length, have recently been collected from off the Transkei coast at 510 m in depth. These colonies can be assigned to the genus Scleroptilum by the arrangement of the polyps in whorls or clusters of 2—3 (each cluster separated by bare areas of rachis), lack of calyces, and the possession of sclerites in the walls of the polyps. The sclerites differ from those of Scleroptilum grandiflorum KOlliker, 1880, and the material has not been identified to species. This represents the first record for the family and genus in southern Africa. Genus Calibelemnon Nutting, 1908 Calibelemnon Nutting, 1908: 562. Balss, 1910: 70. Kiikenthal, 1915: 44. Prochunella Balss, 1909: 426. 76 ANNALS OF THE SOUTH AFRICAN MUSEUM Diagnosis Sclerites completely absent or with small bodies restricted to the peduncle. A genus of perhaps two species of the Indo-Pacific. Type species. Protocaulon indicum J. A. Thomson & Henderson, 1906; Bay of Bengal. Calibelemnon sp. Fig. 38 Remarks Several colonies were collected in 1975 from near Cape Vidal, Natal, at 740 m in depth. Unlike Calibelemnon indicum (J. A. Thomson & Henderson, 1906), in which sclerites are completely absent, these colonies possess small sclerites in the peduncle, and therefore probably represent an undescribed species. This represents a new record of the genus in southern Africa and extends the range to the south-western fringe of the Indo-Pacific region. Family Chunellidae Kikenthal, 1902 Rachis slender. Axis quadrangular. Polyps in well-separated whorls. One to four polyps per whorl. Polyps without calyces. Siphonozooids few and incon- spicuous, on rachis between autozooids. Sclerites absent except for minute oval bodies in the interior of the peduncle. Two genera of the Indian Ocean. The present study represents the first record of the family in southern Africa. Genus Chunella Kiukenthal, 1902 Chunella Kiikenthal, 1902: 302; 1915: 45. Kikenthal & Broch, 1911: 271. Diagnosis With terminal autozooid highly reduced or absent. Whorls with 1-4 autozooids. A genus of three nominal species, from the eastern coast of Africa to the Malay Archipelago. Type species. Chunella gracillima Kiikenthal, 1902; East Africa. Chunella gracillima Kikenthal, 1902 Figs 20, 38 Chunella gracillima Kikenthal, 1902: 302; 1915: 45, fig. 53. Kiikenthal & Broch, 1911: 272, pl. 15 (fig. 13). Bayer, 1956: 227. Material SAM-—H3735, off Port Durnford, Natal (28°37,8’S 32°38,4’E), 1 000-1 200 m, 25 May 1976, 8 whole colonies, beam trawl, South African Museum R.V. Meiring Naude cruises. SAM—H4029, one whole colony, same data as SAM-—H3735. SOUTHERN AFRICA PENNATULACEA I a Va Fig. 20. Chunella gracillima. A. Entire colony. Scale = 30 mm. B. Transverse section of axis. Scale = 0,7 mm. C. Two groups of three autozooids with minute siphonozooids on rachis. Scale = 5mm. D. Minute sclerites from lower peduncle. Scale = 0,02 mm. E. Rudimentary polyp from distal tip of colony. Scale = 1,2 mm. 78 ANNALS OF THE SOUTH AFRICAN MUSEUM Description Colonies examined vary in length from 320-550 mm. Number of autozooids per whorl varies from one to four (most commonly three). Whorls 3-7 per colony separated by 30-60 mm intervals of bare rachis. Autozooids vary from 3 mm to 15 mm in length in a preserved state. Siphonozooids few and inconspicuous on the rachis at each autozooid whorl, consistent in form and size throughout the colony. Axis conspicuously quadrangular, 0,5—1,5 mm in width. Peduncle up to 50 mm in length and 3-4 mm in width. Sclerites absent except for minute oval bodies up to 0,02 mm long in the interior of the lower peduncle of all colonies examined. Autozooids greyish with brown tentacles, rachis and peduncle cream. Distribution (Fig. 38) Equatorial eastern Africa and Natal, 818-1 200 m. This is the first record of the genus and species in southern Africa. The type locality is Pemba Island, border of Kenya and Tanzania, 818 m in depth. Remarks This species is distinguished from other southern African sea pens by its 3-7 demarcated clusters of 2—4 polyps and possession of a single rudimentary terminal polyp. Remarks on the genus Chunella Kikenthal & Broch (1911: 275) described Chunella quadriflora for colonies with lower whorls of four autozooids, from the same locality as C. gracillima. Hickson (1916) described C. biflora (from the Malay Archipelago) for colonies with two autozooids per whorl. Kiikenthal (1915: 45) characterized C. gracillima as having three autozooids per whorl. Of the eight colonies from the same trawl sample that I have examined, one has four autozooids in a lower whorl and three in an upper whorl. Another colony has two autozooids in upper whorls and three in lower whorls. Some whorls in other colonies may have one, two, or three autozooids per whorl. It therefore seems that, since the number of autozooids per whorl is not consistent in a given colony, the character can hardly be used to separate species. As a result, I suspect that the three species may in fact be conspecific but the pertinent type material is presently unavailable for comparative examination. Kikenthal & Broch (1911, pl. 20 (fig. 49)) showed a photograph labelled ‘Chunella quadriflora’. The specimen appears to possess three whorls of two autozooids each and a single well-developed terminal autozooid. The photograph therefore appears to be that of a colony of Amphiacme abyssorum and not Chunella quadriflora. Genus Amphiacme Kikenthal, 1903 Amphianthus Kikenthal, 1902: 303. Amphiacme Kikenthal, 1903: 275; 1915: 46. Kiikenthal & Broch, 1911: 276. SOUTHERN AFRICA PENNATULACEA 719 Diagnosis With a single terminal autozooid that is well-developed and somewhat asymmetrical. Autozooids in pairs. Siphonozooids present on the keel of the terminal autozooid as well as on the rachis between the individual autozooids of each pair. A monospecific genus of the eastern African coast. Type species. Amphianthus abyssorum Kikenthal, 1902; East Africa. Amphiacme abyssorum (Kikenthal, 1902) Figs 21, 38 Amphianthus abyssorum Kikenthal, 1902: 302. Amphiacme abyssorum: Kiikenthal, 1903: 275; 1915: 47, fig. 54. Kikenthal & Broch, 1911: 276, pl. 15 (fig. 14). Tixier-Durivault, 1972: 52. Tixier-Durivault & D’Hondt, 1974: 260. Material SAM-H3736, off Jesser Point, Natal (27°38,6'S 32°52,6’E), 860 m, 21 May 1976, 1 whole colony, beam trawl, South African Museum R.V. Meiring Naude cruises. SAM-—H3737, off Port Durnford, Natal (28°37’S 32°38’E), 1 000- 1 200 m, 25 May 1976, 2 whole colonies, beam trawl, South African Museum R.V. Meiring Naude cruises. Description The colonies examined are 230-250 mm in length. Axis conspicuously quadrangular, usually <1 mm in width. Autozooids in two or three pairs along rachis, at intervals of 30-50 mm. Autozooids about 10 mm in length preserved. Terminal autozooid bilaterally symmetrical or somewhat asymmetrical, 8-10 mm long with a ventral keel. Peduncle 10-12 mm in length and 1-2 mm in width. Siphonozooids few, located on the rachis at each whorl and on the keel of terminal autozooid. Sclerites absent except for oval bodies (<0,02 mm) in peduncle. Polyps brownish-grey, rachis cream or tan, peduncle brownish. Distribution (Fig. 38) The Indian Ocean coast of Africa, from the equator to Natal, and Madagascar; 760-1 200m. The present study represents a new record for southern Africa and extends the range of the genus and species south to South Africa. The type locality is the coastal region near the border of Kenya and Tanzania, 818-1 019 m in depth. Remarks Amphiacme abyssorum is distinguished from other southern African pennatulaceans by the arrangement of the polyps in two or three demarcated pairs along the rachis, and the possession of a single modified terminal polyp. 80 ANNALS OF THE SOUTH AFRICAN MUSEUM La Z Fig. 21. Amphiacme abyssorum. A. Two entire colonies. Scale = 40mm. B. Terminal autozooid with ten minute siphonozooids on keel. Scale = 10 mm. C. Single tentacle from autozooid. Scale = 1 mm. D. Sclerites from peduncle. Scale = 0,02 mm. E. Detail of rachis with a pair of autozooids and several siphonozooids. Scale = 10 mm. F. Transverse section of axis. Scale = 0,5 mm. SOUTHERN AFRICA PENNATULACEA 81 Family Umbellulidae Lindahl, 1874 Rachis long and slender. Axis quadrangular to round in cross section. Autozooids restricted to the distal terminus, usually forming an umbellate cluster. Siphonozooids are present on the rachis at the base of the autozooids or on bare parts of the rachis. Sclerites present in peduncle and terminal cluster, or totally absent. When present, sclerites are rods, spindles or needles, three-flanged round in cross-section. A monogeneric family. Genus Umbellula Cuvier, 1798 Isis Linnaeus, 1758: 800 (partim). Pennatula Pallas, 1766: 335 (partim). Vorticella Linnaeus, 1767: 1317. Ombellula Cuvier, 1798: 675. Umbellularia Lamarck, 1816: 436. Kolliker, 1872a: 203. Umbellula Gray, 1870: 39. Kolliker; 1875: 11. Kikenthal & Broch, 1911: 282 oe Broch, 1957: 350; 1958a: 267; 1958b: 251. Grasshoff, 1982b: 956. Diagnosis Characters of the family. A genus of perhaps 12 valid species. Cosmopolitan distribution; deep sea, 250 m to over 6 100 m. Type species. Isis encrinus Linnaeus, 1758; Arctic Ocean. Umbellula thomsoni (Kolliker, 1874) Figs 19J-N, 22, 23, 40 Umbellularia Thomsonii Kolliker, 1874: 13. Umbellula Thomsonii: Kolliker, 1875: 11. Umbellula thomsonii: Broch, 1958: 253, figs 2, 4 (Lit!). Grasshoff, 1972: 2, fig. 1. Umbellula Guntheri Broch, 1913: 3, pl. 1 (fig. 1). Umbellula aciculifera J. S. Thomson, 1915: 20 (and text-fig. on p. 21), pl. 2 (figs 1-2). Umbellula thomsoni: Kolliker, 1880: 19. Kiikenthal, 1915: 54. Pasternak, 1970: 240, fig. 2; 1975b: 106. Grasshoff, 1982a: 753, map 11; 1982b: 958. Material SAM-H1166, south-west of Cape Point (34°40’S 17°45’E), 1 650 m, 21 July 1903, 1 colony, shrimp trawl, S.S. Pieter Faure survey, station PF 17026. (Holo- type of Umbellula aciculifera Thomson, 1915.) Description The colony examined is about 475 mm in length. The axis is 3 mm in diameter, rounded quadrangular near the terminal cluster. Umbellate cluster composed of 10 autozooids, each 15-20 mm long. Sclerites conspicuous and numerous. Terminal end of rachis, polyp walls, and proximal portions of tentacles with knobby, three-flanged rods and spindles (<0,3 mm). Aboral surface of tentacles and pinnules with spindles (0,5—1,3 mm); round in cross-section, not 82 ANNALS OF THE SOUTH AFRICAN MUSEUM were a ee ene Neem ee et rn eens Scope SAO. ; pee ee re EE ae an nrg pet pag gr en pags mene Sete Rene a5 Se LL. awe oy A Fig. 22. Umbellula thomsoni. A. Entire colony. Scale = 20mm. B. Transverse sections through three different areas of axis. Scale = 2 mm. C. Single tentacle. Scale = 3 mm. D. Small sclerites from proximal portion of tentacle. Scale = 0,1 mm. E. Sclerites from body wall of autozooid. Scale = 0.02 mm. F. Large sclerites from proximal portion of tentacle. G. Sclerites from pinnule. F—G. Scale = 0,5 mm. H. Sclerites from peduncle. Scale = 0,04 mm. SOUTHERN AFRICA PENNATULACEA. 83 three-flanged. Peduncle with rough ovals or rods (0,08—0,11 mm in length), having numerous low, rounded knobs or tubercles. Distribution (Fig. 40) Apparently cosmopolitan, 1 300-6 200m. This specimen is the only known southern African record, first described by J. S. Thomson (1915) as Umbellula aciculifera (type specimen). The type locality of U. thomsoni is the north Atlantic. Fig. 23. Scanning electron micrographs of sclerites of Umbellula thomsoni. A-D. From peduncle: 0,08 mm; 0,09 mm; 0,1 mm; 0,11 mm. E. Detail of surface of D; length of micrograph 0,04 mm. F-H. Large sclerites from tentacle: 0,7 mm; 0,8 mm; 0,7mm. 84 ANNALS OF THE SOUTH AFRICAN MUSEUM Remarks The colony has five solitary epizoic zooanthids (species indeterminate) attached to bare areas of the upper rachis. They vary from 5—15 mm in diameter and 10-12 mm in length (zooanthids not shown in Fig. 22A). This species is distinguished from other southern African sea pens by having the autozooids restricted to a cluster at the distal end of an elongate colony, and the possession of sclerites in the autozooids. Umbellula lindahli Kolliker, 1874 Figs 24, 40 Umbellula lindahlii K6lliker, 1874: 12. Jungersen, 1904: 75, pl. 3 (figs 37-46). Kiikenthal, 1915: 48. Umbellula miniacea Lindahl, 1874: 12, pls 1-2. Umbellula pallida Lindahl, 1874: 13, pl. 3. Umbellula lindahli: Broch, 1958: 262 (Lit.!). Grasshoff, 1982a: 741-742, 753; 1982b: 958, fig. 43. Pasternak, 1975b: 109. Material SAM-—H3738, off the Cape of Good Hope Peninsula (34°37'S 17°03’E), 2 890-2 963 m, 8 December 1959, 2 colonies, beam trawl, Division of Sea Fisheries Survey, station A315. SAM—H3739, west of Dassen Island, western Cape Province (33°36’S 16°15’E), 2 780-2 872 m, 26 August 1959, 2 colonies, beam trawl, Division of Sea Fisheries Survey. SAM—H4028, south of False Bay (35°22,3'S 18°45,4’E), 490 m, 6 July 1986, 22 colonies, beam trawl, Rob Lesley (Sea Fisheries Research Institute, West Coast Biomass Survey, station A4305046). Description Colonies examined are 770-900 mm in length. Axis conspicuously quad- rangular, 1-2 mm in width. Peduncle up to 150 mm long by 6 mm in width. Umbellate terminal cluster with 8—10 autozooids, each 20-30 mm long. Siphono- zooids numerous on keel-like terminus of rachis. Sclerites totally absent. Distribution (Fig. 40) Cosmopolitan up to 6 100 m in depth. In southern Africa, the species is presently known only from the south-western Cape (490-2 963 m). This study represents a new record of the species to southern Africa. The type locality is the Arctic region of Greenland. Remarks This species is differentiated from other southern African sea pens by the restriction of the autozooids to a cluster at the terminal end of an elongate colony, and the complete absence of sclerites. SOUTHERN AFRICA PENNATULACEA 85 Fig. 24. Umbellula lindahli. A. Entire colony. Scale = 50 mm. B. Transverse section through peduncle. Scale = 1,5 mm. 86 ANNALS OF THE SOUTH AFRICAN MUSEUM Remarks on the genus Umbellula Grasshoff (1972) recognized three species of the genus that have sclerites in the terminal autozooid region. Umbellula monocephalus Pasternak, 1964 (jun. syn. U. thieli Grasshoff, 1972) has only a single autozooid; the other two species both have multiple autozooids—U. durissima KOlliker, 1880, has both axis and sclerites that are round in cross section, whereas U. thomsoni has three-flanged sclerites and a quadrangular axis. I have found that shape of the axis in cross-section is an invalid differentiating character with regards to U. thomsoni, since different parts of the axis of the colony examined (SAM-H1166) are round or rounded quadrangular, but never sharply quadrangular (Fig. 22B). Also, the large sclerites of the tentacles and pinnules (>0,5 mm) are round in cross-section, whereas the small sclerites (<0,3 mm) of the proximal regions of the tentacles and of the entire polyp body wall are conspicuously three-flanged (Figs 19J-N, 22D—-E, 23F—H). Thus, this specimen (which is the type upon which the des-cription of U. aciculifera Thomson, 1915, was based) possesses characteristics intermediate between U. durissima and U. thom- soni. In addition, Grasshoff (1972, fig. 2B) showed three-flanged spindles from the polyp walls of U. durissima. These facts tend to negate distinctions between the two species. Examination of type material and a variety of specimens is necessary to determine if we are dealing with two distinct species or a single variable one. This material is presently unavailable to me for comparison. Broch (1958b) revised the genus Umbellula and recognized five species without sclerites in the terminal cluster. Umbellula hemigymna Pasternak, 1975a, was subsequently added from the Caribbean Sea. Of these six, U. lindahli is characterized by having quadrangular axis, large autozooids on a tassel not especially crowded, and full-grown colonies >400 mm in length. Broch (1961) compared specimens of U. encrinus (Linnaeus, 1758) with those of U. lindahli and considered them conspecific, thereby making the latter a junior synonym of U. encrinus. Subsequent workers, including Pasternak (1961b, 1964, 1975b, 1980) and Grasshoff (1982a, 1982b), did not recognize this synonymy and considered the two species as distinct. I have retained this distinction here. Grasshoff (1982b: 957-958) distinguished the two species as follows: U. encrinus—stalk thick, polyps stocky; U. lindahli—stalk thin, polyps slender. Family Virgulariidae Verrill, 1868 Colonies bilateral, long and slender to whip-like. Axis well-developed, present throughout length of colony. Adjacent autozooids fused proximally, forming polyp leaves or low ridges that are arranged in transverse or oblique rows along rachis. Sclerites three-flanged spindles, minute plates, or entirely absent. The family is of cosmopolitan distribution, with two subfamilies and six genera. SOUTHERN AFRICA PENNATULACEA 87 Subfamily Virgulariinae Verrill, 1868 Autozooids form low to well-developed polyp leaves (flattened expansions) by fusion of their proximal portions. Five genera, two of which are included below, in addition to Acanthoptilum K6lliker, 1870, Scytalium Herklots, 1858, and Stylatula Verrill, 1864. Genus Virgularia Lamarck, 1816 Virgularia Lamarck, 1816: 429. Kiikenthal & Broch, 1911: 323. Kitikenthal, 1915: 71 (Lit!). Lygus Herklots, 1858: 11. Halisceptrum Herklots, 1863: 33. Sceptonidium Richiardi, 1869: 63. Cladiscus Koren & Danielssen, 1877: 101. Protocaulon Kolliker, 1880: 26. Svava Danielssen & Koren, 1884: 57. Deutocaulon Marshall & Fowler, 1888: 453. Svavopsis Roule, 1908: 181. Diagnosis Colonies long and slender, often vermiform. Peduncle and rachis usually abruptly delimited. Autozooids form polyp leaves that are arranged in two longitudinal series along rachis. Autozooids retractile into fleshy calyces. Siphonozooids are present on polyp leaves below the free parts of the autozooids or on the rachis between the polyp leaves. Sclerites absent except for minute ovals and plates of the peduncle interior. Axis present throughout entire length of rachis and upper portion of peduncle, often projecting beyond apex of rachis. A genus of about 20 species, in need of revision. Cosmopolitan. Type species. Pennatula mirabilis Linnaeus, 1758; North Atlantic. Virgularia schultzei Kikenthal, 1910 Figs 1D-F, 25, 28C-D, 41 Virgularia schultzei Kiikenthal, 1910: 53; 1915: 72, figs 72, 73 (Lit.!). Kikenthal & Broch, 1911: 326, figs 124-125, pl. 17 (figs 21-22). J. S. Thomson, 1915: 16. Broch, 1939: 30. Tixier- Durivault, 1954: 630. Virgularia Reinwardti (non Herklots, 1858) Hickson, 1900: 87. Balss, 1910: 47 (partim). Virgularia spec. (juv.) Broch, 1910: 231. Material SAM-H3387, off Danger Point, Cape Province (34°37’S 19°24'E), 19 m, 11 April 1984, 11 colonies, SCUBA, W. R. Liltved. SAM—H3159, off Koeberg Power Station, western Cape Province (33°40’S 18°26’E), 6-20m, 24-26 November 1981, 8 colonies, SCUBA, W. R. Liltved. SAM—H3338, Algoa Bay, eastern Cape Province (33°50’S 25°40’E), 16 m, 17 May 1984, 1 colony, SCUBA, G. C. Williams. 88 ANNALS OF THE SOUTH AFRICAN MUSEUM R2 a3 4 ee 3 mY Y a : Fig. 25. Virgularia schultzei. A. Entire preserved colony. Scale = 20 mm. B. Expanded colony in situ. Scale = 20 mm. C. Polyp leaf from a 120 mm long colony. Scale = 1 mm. D. Peduncle sclerites. Scale = 0,005 mm. E. Lower surface of a polyp leaf from middle of rachis of a 400-mm long colony. Scale = 2mm. F. Two retracted polyps from a polyp leaf showing siphonozooids at base. Scale = 0,5 mm. SOUTHERN AFRICA PENNATULACEA 89 Description Colonies examined ranged in length from 85-500 mm, and 5-12 mm in width. Peduncle long and flexible. Axis up to 1,5 mm in diameter, round in cross-section. Surface of axis often with numerous perforations (canals of Kolliker) 0,003—0,004 mm in diameter. Polyp leaves well developed, numerous, transparent. Autozooids 15-35 per polyp leaf. Siphonozooids are present on the lower surface of the polyp leaves, usually one per autozooid in a band below the calyx-like free portions of the autozooids. Sclerites absent except for minute ovals of the peduncle interior (c. 0,005 mm in length). Colour variable, cream or white to yellowish or orange; uniformly coloured or distinctly mottled. Distribution (Fig. 41) Apparently endemic to southern Africa, from Lideritz Bay, Namibia, to at least as far as Kosi Bay, littoral to 222 m in depth. Day (1974b: 35) listed the species from Inhambane, Mozambique. This species is one of the most frequently encountered sea pens in South Africa. Observations SCUBA divers have observed this species with patchy or localized dis- tribution and high population density. An extensive patch about 50 m in depth off Hout Bay, Cape Peninsula, has numerous colonies that are spaced relatively uniformly (about 0,5 m apart) in what appears to be a relatively regular pattern of spacing (P. A. Hulley pers. comm.). Remarks Virgularia schultzei is distinguished from other southern African species of Virgularia by having siphonozooids on the lower surfaces of the polyp leaves, below the free distal ends of the autozooids, and 15—35 autozooids per polyp leaf. Virgularia mirabilis (Linnaeus, 1758) Figs 26, 41 Pennatula mirabilis: Linnaeus, 1758: 819. Miiller, 1776: 255. Virgularia mirabilis Lamarck, 1816: 431. Kikenthal & Broch, 1911: 329, fig. 126. Kikenthal, 1915: 73, fig. 76 (Lit.!). Manuel, 1981: 54, fig. 15. Tixier-Durivault, 19616: 254, figs 17-18; 1963: 72. Tixier-Durivault & D’Hondt, 1974: 262. Material SAM-H3240, mouth of False Bay, western Cape Province (34°22,7’'S 18°43,1’E), 78m, 10 September 1953, 1 colony, dredge, University of Cape Town Ecological Survey. SAM-—H3746, off Stony Point, Transkei (32°38,9'S 28°45,0’E), 360 m , 17 July 1984, 1 partial colony, dredge; G. C. Williams, R.V. Meiring Naude. SAM—H3869, 67 km west of Cape Fria, Namibia (18°25’S 11°15’E), 274 m, May—June 1982, many colonies, trawl. 90 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 26. Virgularia mirabilis. A. Entire colony. Scale = 10mm. B. Detail of rachis with retracted polyps. Scale = 1 mm. C. Part of rachis from colony with expanded polyps. Scale = 10 mm. D. Detail of rachis showing expanded polyps. Scale = 1 mm. SOUTHERN AFRICA PENNATULACEA 91 Description The colonies examined are up to 95 mm long, vermiform, very slender and delicate. Polyp leaves small and not crowded. Autozooids 5-12 per polyp leaf. Autozooids globular or long and narrow, about 1,0 mm long and 0,2 mm wide. Siphonozooids inconspicuous on the sides of the rachis below the polyp leaves and do extend on to the dorsal surface of the rachis. Peduncle about as long as rachis. Sclerites absent except for minute ovals (approximately 0,005 mm in length) from interior of peduncle. Colour yellowish to cream. Distribution (Fig. 41) Atlantic Ocean, including the Arctic region, Mediterranean Sea, northern and western Africa, and the south-western Indian Ocean, 9—400 m in depth. Known in southern Africa from Cape Fria and Walvis Bay, Namibia; the Cape of Good Hope region, Cape Province, and the Transkei coast, 78—360 m in depth. Tixier-Durivault (1963: 73) recorded the species from just south of Walvis Bay, Namibia. Tixier-Durivault & D’Hondt (1974: 262) reported the species from Tulear, Madagascar. The type locality is the northern Atlantic. Remarks Virgularia mirabilis is distinguished from other southern African species of the genus by the often limp and vermiform appearance of the colonies, with 5—12 autozooids per polyp leaf. Kikenthal (1915: 34) stated that up to 16 autozooids per polyp may be present. Virgularia gustaviana (Herklots, 1863) Figs 27, 28A-B, 41 Halisceptrum gustavianum Herklots, 1863: 31. Virgularia gustaviana Kikenthal & Broch, 1911: 334. Kiikenthal, 1915: 74, fig. 77 (Lit.!). Tixier-Durivault, 1954: 630; 1960: 366. Material SAM-H1193, Durban Bay (29°50’S 31°00’E), 23 May 1929, 3 colonies. SAM-H1199, off Salisbury Island, Durban Bay, Natal (29°50’S 31°00’E), 4 July 1927, 1 colony. SAM—H1509, Durban Bay (29°50’S 31°00’E), 1 colony. Description Colonies examined range between 140-419 mm in length and 10-20 mm in width. Colonies are robust and firm, tapering gradually or abruptly at the apical end. Polyp leaves large, numerous and crowded, placed opposite each other. Autozooids highly congested, more than 50 per leaf (often up to 100 or more). Siphonozooids are inconspicuous on the lateral and dorsal surfaces of the rachis between the polyp leaves. Peduncle length varies from one third to one half of total colony length. Peduncle may have a conspicuous end bulb. Sclerites are minute ovals (c. 0,01 mm) restricted to the peduncle interior. Colour yellowish. 92 ANNALS OF THE SOUTH AFRICAN MUSEUM SiG “fnew yeurtes ii aeie Fig. 27. Virgularia gustaviana. A. Entire colony. Scale = 20 mm. B. Single entire polyp leaf. Scale = 3 mm. C. Four adjacent polyps from a polyp leaf. Scale = 1 mm. SOUTHERN AFRICA PENNATULACEA 93 Fig. 28. A-B. Virgularia gustaviana. A. Dorsal view of rachis. B. Ventral view of rachis; length of photographs = 80 mm. C-D. Virgularia schultzei. C. Broken end of axis; diameter = 0,36 mm. D. Detail of axial surface; width of axis = 0,36 mm. 94 ANNALS OF THE SOUTH AFRICAN MUSEUM Distribution (Fig. 41) Indo-West Pacific. Known in southern Africa only from Inhaca Island, Mozambique, and Durban Bay, Natal; littoral and shallow sublittoral. Remarks Virgularia gustaviana is distinguished from other southern African species of Virgularia in having 50-200 autozooids per polyp leaf. Virgularia sp. Fig. 41 Remarks Four colonies, 40-80 mm in length, have recently been collected from off the eastern Cape and Transkei (60—410 m in depth). These colonies are fragmentary, therefore assignment to species is untenable at present. The thick and firm appearance of the polyp leaves clearly separates this species from the other southern African forms of Virgularia. Other species of Virgularia Tixier-Durivault & D’Hondt (1974: 262) listed Port Natal (Durban) as a locality of the Indo-Pacific species Virgularia juncea (Pallas, 1766). Since no description or figures of colonies from this locality are given, and since there are no other records or material of V. juncea known from southern Africa, I here consider this as a dubious record. Tixier-Durivault (1963: 73) recorded Virgularia tuberculata Marshall, 1887, from northern Angola. This is outside the geographical limit of the present work, but is mentioned, here nonetheless as a species that may possibly be encountered further south. It is distinguished by having tubercles or small teeth on the terminal rim of the calyx and only 3-6 autozooids per polyp leaf. Genus Scytaliopsis Gravier, 1906 Scytaliopsis Gravier, 1906: 293; 1908: 237. Kiikenthal, 1915: 67. Diagnosis Colony firm. Axis circular in cross-section. Gradual transition between rachis and peduncle. Small polyp leaves in pairs, separated by areas of bare rachis. Autozooids 4—5 per polyp leaf. Innermost autozooids of each polyp leaf smaller than adjacent ones. Anthocodiae are retractile into the globular, fleshy bases of autozooids. Siphonozooids few and inconspicuous, situated on rachis between leaves. Sclerites absent. A monospecific genus of the western Indian Ocean and Red Sea. Type species. Scytaliopsis djiboutiensis Gravier, 1906; Djibouti (Red Sea). SOUTHERN AFRICA PENNATULACEA 95 Scytaliopsis djiboutiensis Gravier, 1906. Figs 29, 41 Scytaliopsis djiboutiensis Gravier, 1906: 293; 1908: 237. Kiikenthal, 1915: 67 (figs 66, 67). Material SAM-H3711, off Qora River, Transkei (32°35,4’S 28°49,2'E), 450-460 m, 14 July 1984, 1 colony, dredge, G. C. Williams, R.V. Meiring Naude. Description The colony is 92 mm in length; rigid, straight, and shaft-like. Peduncle shorter than rachis. Polyp leaves arranged in 15—20 pairs, each leaf with not more than five autozooids. Adjacent autozooids gradually increase in size from innermost to outermost. Intervals of bare rachis up to 3 mm. Siphonozooids are very sparse and inconspicuous, usually not more than two are present on the rachis below each pair of polyp leaves. Sclerites entirely absent. Colour brownish. Distribution (Fig. 41) East African coast, littoral to 460 m in depth. This is the first record of the genus and species for southern Africa and extends the range southward from the Red Sea to the Transkei coast. The type locality is Djibouti, Red Sea. Remarks This species is differentiated from other southern African pennatulaceans by having 5 polyps per polyp leaf, which grade in size from the smallest on the dorsal side of each leaf to the largest on the ventral side. Subfamily Balticininae Gray, 1870 Fusion of proximal portions of adjacent autozooids forms raised ridges or swellings that emanate laterally from the rachis. Siphonozooids few and incon- spicuous between autozooids. One genus of the Atlantic and Indo-Pacific oceans. Genus Halipteris Kolliker, 1869 non Virgularia Lamarck, 1816: Koren & Danielssen, 1847: 269. Halipteris Kolliker, 1869: 124. Pavonaria K6lliker, 1869: 123 (non Schweigger, 1820; see Bayer, 1956: F228). Kukenthal & Broch, 1911: 305. Kiikenthal, 1915: 60 (Lit!). Balticina Gray, 1870: 13. Norticina Gray, 1870: 13. Osteocella Gray, 1870: 40. Lygomorpha Koren & Danielssen, 1877: 99. Stichoptilum Grieg, 1887: 15. Diagnosis Colonies long and slender. Axis conspicuous, extending entire length of colony, round in cross-section. Autozooids arranged in numerous oblique rows in 96 ANNALS OF THE SOUTH AFRICAN MUSEUM A Fig. 29. Scytaliopsis djiboutiensis. A. Entire colony. Scale = 20mm. B. Detail of rachis showing two polyp leaves. Scale = 2mm. C. Single polyp leaf showing gradation in size of adjacent autozooids. Scale = 2 mm. SOUTHERN AFRICA PENNATULACEA 97 two longitudinal series. Adjacent autozooids sometimes fused at their bases, often forming ridges or swellings that emanate laterally from the rachis. Autozooids with two-toothed calyces. Anthocodia retractile into calyces. Sclerites numerous; three-flanged spindles in calyx and tentacles, smooth rods or spindles in peduncle. A genus of five species; Atlantic, Pacific and Western Indian oceans. Type species. Virgularia finmarchica Sars, 1851; northern Atlantic. Halipteris africana (Studer, 1879) comb. nov. Figs 19A-G, 30, 40 Halipteris sp. Studer, 1878: 139. Pavonaria africana Studer, 1879: 672, pl. 5 (fig. 41). Kiikenthal, 1915: 61. Material SAM-—H3237, north-west of Lamberts Bay, western Cape Province (31°29,7’'S 16°03,4’E), 459 m, 15 August 1947, 1 colony, dredge, University of Cape Town Ecological Survey, AFR730G. SAM-—H4027, off Namaqualand (30°00,5’S 14°53,6’E) 461 m, 20 July 1986, 8 colonies, bottom trawl, Rob Leslie (Sea Fisheries Research Institute, West Coast Biomass Survey, station A4373046). Description Colonies whip-like, up to 1 550 mm in length and averaging 10 mm in width. Axis round to rounded quadrangular, approximately 5 mm in diameter. Peduncle stout, approximately 200 mm long and up to 15 mm wide. Peduncle distinctly quadrangular. Autozooids arranged in numerous oblique rows, 3-7 (usually 4—6) per row. Some rows form basal ridges by fusion of proximal regions of autozooids. In some rows, circular or elliptical walls of the basal ridge surround individual autozooids or groups of two or three. Aboral surface of tentacles with densely set three-flanged spindles, 0,10—0,15 mm in length, arranged longitudinally. Calyx with numerous three-flanged spindles 0,12—0,60 mm long. Terminal teeth of calyx sometimes blunt or indistinct. Peduncle with smooth spindles or rods 0,04— 0,07 mm in length. Rachis and peduncle pale red to white, calyx light blue, polyps brown. Distribution (Fig. 40) Atlantic coast of Africa, 459-659 m in depth. This study represents a new record from southern Africa and extends the range of the species south from the Senegal—Guinea region to the Namaqualand coast, South Africa. The type locality is tropical west Africa. Remarks Gray (1870: 13) described two genera: Balticina for B. finmarchica (Sars, 1851) and Norticina for N. christii (Koren & Danielssen, 1848), presumably after ANNALS OF THE SOUTH AFRICAN MUSEUM 98 ts SCRE ca. eec eee g < XI ee a FA wazre yy iH t iG a Lh LO ae eee oe ene Ay \\ hee j ( SU IS — a ac eA = 78) —— Sacee Nig as 7,5mm. D. Group of six autozooids with Fig. 30. Halipteris africana. A. Entire colony. Scale = 30 mm. B. Detail of rachis. Scale = Js “> 7mm. C. Transverse section of axis. Scale common basal ridge. Scale = 2 mm. E. Two autozooids with ten siphonozooids at their bases. Scale = 2 mm. F. Single tentacle with sclerites. Scale = 0,5 mm. G. Sclerites from tentacle. Scale = 0,1 mm. H. Sclerites from peduncle. Scale = 0,04 mm. I. Sclerites from calyx. Scale = 0,25 mm. SOUTHERN AFRICA PENNATULACEA — 99 the Baltic and Nordic (North) seas, respectively. Both genera are considered synonyms of Pavonaria Kolliker, 1869, by Kikenthal (1915: 60). However, Pavonaria was found by Kikenthal (1915: 60) and Bayer (1956: 228) to be preoccupied by Pavonaria Schweigger, 1820, a synonym of Funiculina. Therefore the name Halipteris Kolliker, 1869, which is synonymous with Pavonaria Kolliker, 1869, is the valid generic name (M. Grasshoff pers. comm.). Tixier-Durivault made the original determination for this specimen and identified it as Funiculina armata Verrill, 1879. However, the possession of ridges in oblique rows, which unify the fused bases of adjacent autozooids, and the autozooids with calyces of two terminal teeth prove this to be a misidentification. Tixier-Durivault (1954: 629) also listed F. armata from South Africa based on two specimens collected on the same day and in the same vicinity as the present specimen. I consider this a dubious record until these specimens can be examined. Kiukenthal (1915: 60-61) distinguished B. africana by the possession of five or six autozooids per row, and the presence of contiguous distal portions of the basal swellings of the polyp leaves that may surround the proximal regions of several adjacent autozooids. This latter feature is common in autozooids of the colony examined but is not universal (Fig. 31D). Jungersen (1904: 43) maintained that B. africana is probably not a distinct species but rather a southern form of the common Atlantic species, B. fin- marchica (Sars, 1851). However, he did not make a formal synonymy. The distinguishing characters recognized by Studer (4—6 autozooids per row and low ridges with wall-like extensions around individual autozooids) may indeed not be sufficient for specific differentiation. A comparison of many specimens from various localities is necessary. Remarks on the genus Halipteris Three additional colonies assignable to the genus Halipteris are present in the South African Museum’s collection. They are from the Natal south coast (900-625 m), the Transkei coast (300 m), and the Namaqualand coast (457 m). The three colonies are relatively small (105-220 mm in length) and apparently represent at least two species, which have not been identified at present. Family Pennatulidae Ehrenberg, 1834 Colonies bilateral with large polyp leaves. Autozooids in one or more rows on margin of polyp leaves. Anthocodia retractile into calyces with marginal teeth. Siphonozooids confined to rachis. Mesozooids may be present. Sclerites numerous; mostly three-flanged needles, rods, plates, or ovals. Three genera: Pennatula Linnaeus, 1758, of cosmopolitan distribution; and Leioptilus Gray, 1860, and Ptilosarcus Gray, 1860, both restricted to the Pacific coast of North America. 100 ANNALS OF THE SOUTH AFRICAN MUSEUM Genus Pennatula Linnaeus, 1758 Pennatula Linnaeus, 1758: 818. Kikenthal & Broch, 1911: 348. Kikenthal, 1915: 81 (Lit.!). Grasshoff, 19826: 959. Penna Bohadsch, 1761: 98 (partim). Diagnosis Calyces usually with eight terminal teeth. Calyces tubular with numerous sclerites, which are three-flanged needles. Inconspicuously three-flanged spindles and plates occur in peduncle. A cosmopolitan genus of probably 12 to 15 valid species and many uncertain or dubious ones. Type species. Pennatula phosphorea Linnaeus, 1758. Pennatula inflata Kikenthal, 1910 Figs 31A-B, 32, 33A-C, 34, 35, 40 Pennatula inflata Kikenthal, 1910: 52; 1915: 83, figs 81-83. Kikenthal & Broch, 1911: 350, pl. 17 (fig. 25), pl. 22 (figs 61, 62). Fennatula phosphorea (non Linnaeus, 1758) Tixier-Durivault, 1954: 630. Material SAM-—H3238, north-west of Lambert’s Bay (31°43,6’S 16°13,2’E), 457 m, 9 June 1947, 2 colonies, trawl, University of Cape Town Ecological Survey. Fig. 31. Scanning electron micrographs of sclerites. A-B. Pennatula inflata. From surface of peduncle: 0,1 mm; 0,08 mm. C-—D. Pteroeides isosceles. From surface of peduncle: 0,08 mm; 0,1 mm. SOUTHERN AFRICA PENNATULACEA 101 A Fig. 32. Pennatula inflata. A. Entire colony, lateral view. Scale 40 mm. B. Two mesozooids with sclerites. Scale = 0,3 mm. C. Group of siphonozooids. Scale = 0,2 mm. D. Single polyp leaf showing positions of autozooids (a), mesozooids (m), and siphonozooids (s). Scale = 8,8 mm. E. Group of three retracted autozooids. Scale = 1,6 mm. F. Two autozooids showing sclerite arrangement. Scale = 1 mm. G. Three sclerites from calyx. Scale = 0,3 mm. H. Seven sclerites from surface of peduncle. Scale = 0,1 mm. I. Terminal tip of a calyx sclerite. Scale = 0,035 mm. J. Sclerites from interior of peduncle. Scale = 0,03 mm. ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 33. Scanning electron micrographs of sclerites. A-C. Pennatula inflata. From calyx: 0,8 mm; 0,6 mm; 0,8 mm. D-E. Pteroeides isosceles. D. Needle from supporting ray of polyp leaf: 7 mm. E. Terminal tip of needle from supporting ray: 1,3 mm. SOUTHERN AFRICA PENNATULACEA 103 Fig. 34. Pennatula inflata. Entire colony; length 220 mm. A. Dorsal view. B. Ventral view. 104 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 35. Pennatula inflata. Two views of distal portion of rachis. Length of rachis portion shown is 150 mm. SOUTHERN AFRICA PENNATULACEA 105 Description The colonies examined are 220 mm and 400 mm in length. Peduncle stout, rigid, one-third total length of colony, oval in cross-section. Prominent bulb about midway on peduncle. Rachis spongy and flexible, with naked dorsal tract up to 10 mm in width. This tract is widest in the distal half of the rachis. Polyp leaves well developed, up to 40 mm in length and 16 mm in width, 26-37 pairs of polyp leaves per colony. Autozooids 29—42 per polyp leaf, in usually 2 or 3 ill-defined rows. Calyces tubular, up to 10 mm long, eight-toothed, with teeth equal to subequal; many calyces with one tooth more strongly developed than the others. Zooids dimorphic, dense and numerous on two narrow marginal tracts bordering dorsal face of rachis. Mesozooids with two tooth-like points composed of sclerites, common in a row along the basal dorsal margin of each polyp leaf. Siphonozooids without sclerites, crowded in narrow band on rachis above each polyp leaf. Colonies with 26—37 rows of siphonozooids on each side, representing one row per polyp leaf. Sclerites numerous; three-flanged needles in calyx, up to 1,2 mm in length; inconspicuously three-flanged spindles of peduncular surface up to 0,2 mm in length; minute oval plates in peduncular interior 0,015—0,025 mm long. Colour of peduncle, rachis and leaves yellowish-white to reddish-orange, sclerites of autozooids and mesozooids orange, siphonozooids white. Distribution (Fig. 40) East African coast and south-west coast of South Africa, 457-741 m in depth. The present study establishes a new record for southern Africa and extends the range south-west from Somalia to the Namaqualand coast, South Africa. It is the first record of the species for the Atlantic Ocean. The type locality is Somalia (Indian Ocean), 628-741 m in depth. Remarks Tixier-Durivault (1954: 630) identified these specimens as P. phosphorea. However, the possession of mesozooids on the dorsal edge of the polyp leaves proves this to be a misidentification. According to Kikenthal (1915: 82), P. phosphorea does not possess mesozooids and does not possess more than 20 polyps per mature polyp leaf. Day (1974b: 35) recorded several unidentified species of the genus Pennatula from Delagoa Bay, Mozambique. Of the twelve species of Pennatula considered to be valid by Kikenthal (1915: 81), six possess between 3 and 20 polyps per polyp leaf—P. aculeata Danielssen, 1860; P. indica J. A. Thomson & Henderson, 1906; P. murrayi Kolliker, 1880; P. pearceyi KGlliker, 1880; P. phosphorea Linnaeus, 1758; and P. prolifera Jungersen, 1904. Six additional species have 30-100 polyps per polyp leaf—P. fimbriata Herklots, 1858; P. grandis Ehrenberg, 1834; P. inflata Kiikenthal, 1910; P. moselyi KGlliker, 1880; P. naresi Kolliker, 1880; and P. rubra (Ellis in Bohadsch, 1761). Of these, only P. inflata possesses a wide 106 ANNALS OF THE SOUTH AFRICAN MUSEUM naked dorsal tract of the upper rachis with a spongy texture, and many calyces have one tooth more strongly pronounced than the others. A certain amount of variation occurs with regard to the calyx teeth in P. inflata. In the present specimens, the development of the calyx teeth is variable—many autozooids examined possess eight equal or subequal teeth. However, in a few polyps, one, two, three, or four of the teeth may be somewhat larger in size than the remaining teeth; and in many polyps one tooth 1s slightly or distinctly larger than the other seven teeth. Family Pteroeididae Kolliker, 1880 Colonies bilateral, possessing well-developed and rigid polyp leaves with long stout sclerites that are not three-flanged. Siphonozooids crowded in proximal region of leaves. Autozooids retractile into fleshy distal region of polyp leaves, or into rounded calyces that are not composed of dense sclerites. Three genera of the Atlantic and Indo-Pacific: Pteroeides Herklots, 1858; Sarcoptilus Gray, 1848; and Crassophyllum Tixier-Durivault, 1961a. Genus Pteroeides Herklots, 1858 Pteroeides Herklots, 1858: 19. K6lliker, 1872a: 124. Gray, 1870: 24. Kiikenthal & Broch, 1911: 395. Ktikenthal, 1915: 97. Godeffroyia Kolliker, 1869: 222. Argentella Gray, 1870: 22. Pteromorpha Gray, 1870: 23. Crispella Gray, 1870: 25. Diagnosis Polyp leaves with supporting rays composed of long needle-like sclerites. Needles of supporting rays smooth, except for terminal ends that are somewhat tuberculated. Sclerites of peduncle are rods or spindles. A genus of over 80 named species, including many uncertain or dubious ones; Atlantic, Mediterranean, and Indo-Pacific. Type species. Penna grisea Bohadsch, 1761 (= Pennatula spinosa Ellis, 1764); Europe. Pteroeides isosceles Thomson, 1915 Figs 31C—D, 33D-E, 36, 37, 38 Pteroeides isosceles J. S. Thomson, 1915: 17, pl. 1. Material SAM-—H1207, off East London (33°00’S 28°00’E), 58-69 m, 28 November 1906, 1 colony, large trawl, S.S. Pieter Faure Survey, P.F. 19061 (holotype). SOUTHERN AFRICA PENNATULACEA 107 Redescription of the holotype The colony is 170 mm long and up to 40 mm in width. Peduncle 18—20 mm in diameter and slightly more than half the length of the colony. Rachis with 22 pairs of polyp leaves. Larger leaves 20 mm in width by 35 mm in length, 15-18 rays per leaf in leaves from the middle region of the rachis. Rays extend through the entire length of a particular leaf and may extend 1-2 mm beyond the margin. Rays not visible in the basal plate of siphonozooids. Autozooids restricted to the outer margin of each leaf to 5 mm below the margin on both the upper and lower surfaces of each leaf; retractile directly into pockets of the fleshy leaf margin. Autozooid zone relatively free of smaller spindles. The autozooids are not arranged in distinct rows, but 5—8 autozooids are usually scattered between two adjacent rays. Mesozooids not apparent. Siphonozooids are densely arranged and restricted to a plate about 12 mm broad in the proximal region of each leaf. Needles of rays up to 10 mm long. Spindles of dorsal surface of polyp leaves 0,2—0,4 mm in length. Rods and spindles of peduncular surface are longitudinally grooved but not distinctly three-flanged, 0,10—0,17 mm long. Sclerites of peduncular interior absent. Peduncle gray, leaves cream with yellowish siphonozooids and white rays. Distribution (Fig. 38) Known only from the type locality off East London, South Africa, 58-69 m depth. Remarks This species can be differentiated from other southern African sea pens by having autozooids without calyces united into thick, fleshy polyp leaves, 11-18 conspicuous rays per leaf, and siphonozooids densely set but restricted to a plate near the base of each leaf. Remarks on the genus Pteroeides Recent collections from northern Zululand have yielded several colonies of a species distinct from P. isosceles, which has not been identified. Several damaged or presumably young colonies of Pteroeides from Natal are also present in the South African Museum collection that can not presently be identified to species but seem to be distinct from P. isosceles. It is likely that several species of the genus can be expected to occur in southern African waters. The genus is in much need of revision; well over 80 specific names can be found in the literature at present. Broch (1958: 278) recorded Pteroeides griseum (Bohadsch, 1761) from off the Angola/Namibia border region at 80-91 m. This is just north of the geographic range of the present work, but is mentioned since the species may be encountered further south. It is distinguished from P. isosceles by having numerous spindles in the autozooid zone, independent of the ray needles. In P. isosceles the ray 108 ANNALS OF THE SOUTH AFRICAN MUSEUM TN . ; : Fig. 36. Pteroides isosceles. A. Entire colony, ventral face. Scale = 60 mm. B. Detail of margin of polyp leaf, upper surface. Scale = 3 mm. C. Entire polyp leaf, lower surface. Scale = 10 mm. D. Peduncular surface sclerites. Scale = 0,1 mm. E. Ray needles from polyp leaf. Scale = 2 mm. SOUTHERN AFRICA PENNATULACEA — 109 Fig. 37. Pteroeides isosceles. Entire colony, length 170 mm. A. Ventral view. B. Dorsal face. needles are pointed and similar at both ends; the autozooid margins may have only scattered sclerites that differ from the ray needles; the siphonozooid and autozooid zones have a distinct boundary. J. S. Thomson (1915: 17) described a damaged colony from the type locality of Pteroeides isosceles as Pteroeides sp. An examination of this specimen has shown that it closely resembles the holotype of P. isosceles and thus the two specimens are very likely conspecific. 110 ANNALS OF THE SOUTH AFRICAN MUSEUM MAPUTOA RE INDIAN OCEAN EAST LONDON, PORT ELIZABETH Fig. 38. Map of south-eastern Africa showing known collecting stations for twelve pennatu- lacean species: Amphiacme abyssorum (@); Calibelemnon sp. (@); Cavernularia dayi (A); Cavernularia lutkenii (¥); Cavernulina sp. (*); Chunella gracillima (+); Funiculina quadran- gularis (©); Kophobelemnon sp. ((@); Pteroeides isosceles ({_}); Scleroptilum sp. (®); Veretillum leloupi and V. cynomorium (§). ATLANTIC SOUTHERN AFRICA OCEAN ® Fig. 39. Map showing known distribution of the family Echinoptilidae: Actinoptilum molle (@); Echinoptilum echinatum (&); Echinoptilum macintoshii (¥); Echinoptilum asperum, E. elong- atum, E. minimum, E. roseum (x). SOUTHERN AFRICA PENNATULACEA 111 ANGOLA NAMIBIA Walvis Bay ®)\Luderitz SOUTH AFRICA ST. HELENA INDIAN BAY OCEAN LANGEBAAN LAGOON 35 ATLANTIC OCEAN 16 Fig. 40. Map of south-western Africa showing known collecting stations of eight pennatulacean species: Anthoptilum grandiflorum (A); Cavernularia elegans ((@); Distichoptilum gracile ({_]}); Halipteris africana (A); Kophobelemnon stelliferum (+); Pennatula inflata (J); Umbellula lindahli (@); Umbellula thomsoni (CE); Veretillum cynomorium (Q). MOZAMBIQUE NAMIBIA MAPUTOS @ e@) LUDERITZ DURBAN/@ ATLANTIC INDIAN OCEAN SOUTH AFRICA OCEAN EAST 6 LONDON¢ D) eo PORT ELIZABETH t ) ) PCAPE TOWN 6, A @ 19 23 Fig. 41. Map showing known collecting stations for the family Virgulariidae (subfamily Virgulariinae) in southern Africa: Scytaliopsis djiboutiensis (QO); Virgularia gustaviana () Virgularia mirabilis (A); Virgularia schultzei (@); Virgularia sp. ((_]). fA ANNALS OF THE SOUTH AFRICAN MUSEUM Remarks on the genus Crassophyllum Tixier-Durivault (1961a: 428) described a new genus and species, Crasso- phyllum cristatum, from northern Angola. It is mentioned here since it may possibly be encountered further south. This form is distinguished by the lack of supporting rays in the polyp leaves, but possesses scattered spindles and needles in the region between the siphonozooid plate and the polyp leaf margin. DISCUSSION ZOOGEOGRAPHY The littoral and sublittoral benthic regions of southern Africa constitute a remarkably diverse and biogeographically unique area. The region is made up of cold-water and warm-water temperate elements, as well as a tropical component. In particular, the southernmost coast (the region between the Cape Peninsula and Port St. Johns) has a high degree of endemism in the invertebrate fauna (Ekman 1953: 189; Day et al. 1970: 2). This region is referred to as the ‘Agulhas Province’ by Briggs (1974: 149), the ‘Cape Fauna’ by Stephenson & Stephenson (1972: 101), and the ‘South Coast’ fauna by Ekman (1953: 187) and Millard (1978). The peculiarities of this region and the composition of the fauna are due to a number of factors. Most important of these are: (1) the isolated, peninsular nature of southern Africa, which is bounded by two oceans, and the extensive area of continental shelf known as the Agulhas Bank, which is a probable centre of adaptive radiation (Millard 1978, fig. 1); and (2) the convergence of oceanic waters of divergent temperatures and origins in the vicinity of the Cape Peninsula, via the colder Benguela Current of the South Atlantic and the warmer Agulhas Current of tropical Indian Ocean influence (Branch & Branch 1981: 14). A zoogeographical analysis of the southern African octocoral fauna will be dealt with in a subsequent paper. AFFINITIES OF THE FAUNA Shallow-water benthic invertebrates of southern Africa are generally recognized to have four major zoogeographical components: temperate Atlantic, Subantarctic, Indo-Pacific, and endemic (Millard 1978; Gosliner 1987). Many deep-water species have more scattered or cosmopolitan distributions. Several species of the west and south coasts, such as Lituaria valenciennes, extend further north into the Mediterranean Sea. Similarly, a number of species of the east coast, such as Calibelemnon indicum, have wider distributions in the tropical. Indo- Pacific. At least 20 species of hydroids from southern Africa have Antarctic or Subantarctic affinities (Millard 1978: 171-172). Gosliner (1987), in considering the sister relationships in opisthobranch molluscs, regarded the Subantarctic influence as perhaps the oldest influence, which may be associated with the break-up of Gondwanaland. In regard to sea pens, the cosmopolitan Umbellula lindahli is known from throughout Antarctic waters (Grasshoff 1982a: 753, SOUTHERN AFRICA PENNATULACEA 113 map 11). However, no truly Subantarctic component is presently recognized within the southern African fauna. Endemic species, such as Virgularia schultzei and Actinoptilum molle, are found throughout the region, with the centre of distribution along the south coast (the region between the Cape Peninsula and the Transkei coast). SUMMARY AND CONCLUSION The present study is the first in a two part series of a comprehensive survey of pennatulaceans from the subcontinent of southern Africa. It has increased in the number of identified species in the region from 12 to 23, the number of genera from 10 to 20, and the number of families from 8 to 12. The families recorded from southern Africa for the first time are the Kophobelemnidae, Protoptilidae, Scleroptilidae, and Chunellidae. The eleven species newly recorded for the region are Cavernularia elegans, Echinoptilum macintoshii, Echinoptilum echinatum, Kophobelemnon stelliferum, Disticho- ptilum gracile, Chunella gracillima, Amphiacme abyssorum, Umbellula lindahli, Scytaliopsis djiboutiensis, Halipteris africana, and Pennatula inflata. The genera Cavernulina, Sclerobelemnon, and Calibelemnon are recorded for the first time from the region. ACKNOWLEDGEMENTS I extend my appreciation to Dane Gerneke and Klaus Schultes of the Uni- versity of Cape Town Electron Microscope Unit for assistance in the use of the Cambridge S180 and S200 scanning electron microscopes. I am grateful to William Liltved (South African Museum), Philip Coetzee (University of Port Elizabeth), Richard Kilburn, Ruth Fregona, and David Herbert (Natal Museum), and Roy Melville-Smith, Robert Leslie, and Paul Williams (Sea Fisheries Research Institute) for field assistance and collection of specimens. Elizabeth Hoenson and Michelle van der Merwe provided curatorial assist- ance. The late Dr A. H. B. De Decker assisted in the translation of some literature. Marcelle Scheiner and Sandra Saven typed the manuscript. Elizabeth Louw and Sheryl Ozinsky aided in correcting the manuscript. I thank Professor G. M. Branch and Dr J. A. Day of the Department of Zoology, University of Cape Town; Dr M.-J. D’Hondt of the Museum National d’Histoire Naturelle, Paris; and Dr M. Grasshoff of the Forschungsinstitut Sen- ckenberg, West Germany, for their helpful comments. Partial financial support for this work was provided by the Council for Scientific and Industrial Research, Research Grants Division. Part of the material examined was procured by the South African Museum and Natal Museum during dredging and trawling cruises on board the R.V. Meiring Naude between 1975 and 1985. 114 ANNALS OF THE SOUTH AFRICAN MUSEUM REFERENCES ASBJ@RNSEN, P. C. 1856. Beskrivelse over Kophobelemnon miilleri, en ny Sofjaerslaegt. In: Sars, M., Koren, J. & DANIELSSEN, D. C. eds. Fauna Littoralis Norvegiae 2: 81-85. Bass, H. 1909. Uber Pennatuliden des Miinchener Museums. Zoologischer Anzeiger 34 (13/14): 423-431. Bass, H. 1910. Japanische Pennatuliden. Jn: DoFLEIN, F. ed. Beitrdge zur Naturgeschichte Ostasiens. Abhandlungen der Mathematisch-Physischen Classe der K6niglich Sdchsischen Gesellschaft der Wissenschaften 1 (10, suppl.): 1-106. Bayer, F. M. 1955. Contributions to the nomenclature, systematics, and morphology of the Octocorallia. Proceedings of the United States National Museum 105 (3357): 207-220. Bayer, F. M. 1956. Octocorallia. In: Moore, R. C. ed. Treatise on invertebrate paleontology. Part F. Coelenterata: 166-231. Lawrence: Geological Society of America and University of Kansas Press. Bayer, F. M. 1981la. Status of knowledge of octocorals of world seas. Seminarios de Biologia Marinha, Academia Brasileira de Ciencias Rio de Janeiro 1981: 3-102. Bayer, F. M. 19816. Key to the genera of Octocorallia exclusive of Pennatulacea (Coelenterata: Anthozoa), with diagnoses of new taxa. Proceedings of the Biological Society of Washington 94 (3): 902-947. Bayer, F. M., GrAssHorr, M. & VERSEVELDT, J. eds. 1983. Illustrated trilingual glossary of morphological and anatomical terms applied to Octocorallia. Leiden: E. J. Brill. BouwapscH,—. 1761. De quibusdom animalibus marinis, eorumque proprietabibus, orbi litterario vel nondum vel minus notis, liber cum nonnullis tabulis aeri incisis, ab auctore super vivis animalibus delineatis. Dresdae: apud G. C. Walther. Bourne, G. C. 1900. The Anthozoa. Jn: LANKESTER, E. R. ed. A treatise on zoology. Part I. The Porifera and Coelenterata. London: Adam & Charles Black. BRANCH, G. & BRANCH, M. 1981. The living shores of southern Africa. Cape Town: Struik. Briccs, J. C. 1974. Marine zoogeography. New York: McGraw-Hill Book Company. Brocu, H. 1910. Anthozoa (Nachtrag): Pennatuliden. Jn: ScHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stidafrika, 1903-1905. 4. Systematik und Tiergeographie. XIII. Denkscriften der Medicinisch- naturwissenschaftlichen Gesellschaft zu Jena 16: 229-235. Brocu, H. 1913. Pennatulacea. Report on the Scientific Results of the ‘Michael Sars’ North Atlantic Deep Sea Expedition, 1910 3 (1): 3-8. Brocu, H. 1939. Some South African shallow water octactinians. Kungliga Fysiografiska sdllskapets i Lund forhandlingar 9 (6): 1-32. BrocH, H. 1957. Pennatularians (Umbellula). Reports of the Swedish Deep Sea Expedition, 1947-48 2 (21): 348-364. Brocu, H. 1958a. The pennatularian genus Umbellula. Nature, London 182 (4630): 267-268. Brocu, H. 19585. Octocorals. Part I. Pennatularians. Discovery Reports 29: 245-280. eee 1961. Umbellula encrinus pavist i hgyantarktiske farvann. Fauna, Oslo 14 (2): Cuvier, G. 1798. Tableau élémentaire de V’histoire naturelle des animaux. Paris: Bandouin. DANIELSSEN, D. C. & Koren, J. 1884. Pennatulida. Den Norske Nordhavs-Expedition, 1876-1878. Zoologi 12: 1-84. Day, J. H. 1974a. The ecology of Morrumbene Estuary, Mozambique. Transactions of the Royal Society of South Africa 41 (1): 43-97. Day, J. H. 1974b. A guide to marine life on South African shores. 2nd ed. Cape Town: A. A. Balkema. Day, J. H., Fietp, J. G. & Penritu, M. J. 1970. The benthic fauna and fishes of False Bay, South Africa. Transactions of the Royal Society of South Africa 39 (1): 1-108. D’Honpt, M. J. 1984a. Pteroeides (Octocorallia, Pennatulacea) de Nouvelle-Calédonie. Bulletin du Muséum national d'histoire naturelle (4) 6 (A, 1): 3-29. D’Honpt; M. J. 1984b. Contribution a la connaissance de certains genres de la famille Veretillidae (Pennatulacea). Description de Cavernulina grandiflora n. sp. et de Lituaria ee ee nom. nov. Bulletin du Muséum national dhistoire naturelle (4) 6 (A, 3): SOUTHERN AFRICA PENNATULACEA eS Exuis, J. 1764. An account of the sea pen, or Pennatula phosphorea of Linnaeus; likewise a description of a new species of sea pen, found on the coast of South-Carolina, with observations on sea-pens in general. In a letter to the honourable Coote Molesworth, Esq; M.D. and F.R.S. from John Ellis, esq; F.R.S. and member of the Royal Academy of Upsal. Philosophical Transactions of the Royal Society of London 53: 419-435. EKMAN, S. 1953. Zoogeography of the sea. London: Sidgwick & Jackson. GosLinER, T. M. 1987. Biogeography of the opisthobranch gastropod fauna of southern Africa. American Malacological Bulletin 5 (2): 243-258. GrassHorFF, M. 1972. Eine Seefeder mit einem einzigen Polypen: Umbellula thieli n. sp. Die von F.S. ‘Meteor’ 1967-1970 im 6stlichen Nordatlantik gedredschten Pennatularia (Cnidaria: Anthozoa). Meteor Forschungsergebnisse (D) 12: 1-11. GrassHorfF, M. 1981. Gorgonaria und Pennatularia (Cnidaria: Anthozoa) vom Mittelatlantis- chen Rucken SW der Azoren. Steenstrupia 7 (9): 213-230. GrassHorFr, M. 1982a. Die Gorgonaria, Pennatularia und Antipatharia des Tiefwassers der Biskaya (Cnidaria, Anthozoa). Ergebnisse der franzdésischen Expeditionen Biogas, Polygas, Geomanche, Incal, Noratlante und Fahrten der ‘Thalassa’. I. Allgemeiner Teil. Bulletin du Muséum national @histoire naturelle (4) 3 (A, 3): 731-766. GrassHoFrF, M. 1982b. Die Gorgonaria, Pennatularia und Antipatheria des Tiefwassers der Biskaya (Cnidaria, Anthozoa). Ergebnisse der franzoGasischen Expeditionen Biogas, Polygas, Geomanche, Incal, Noratlante und Fahrten der “Thalassa’. II. Taxonomischer Teil. Bulletin du Muséum national @histoire naturelle (4) 3 (A, 4): 941-978. GravieR, M. C. 1906. Sur un type nouveau de Virgulaire. Bulletin du Muséum d histoire naturelle 12 (5): 291—293. GRAVIER, C. 1907. Sur un genre nouveau de Pennatulidé (Mesobelemnon nov. g. gracile nov. sp.). Bulletin du Muséum national @ histoire naturelle 13 (2): 159-161. GRAVIER, C. 1908. Recherches sur quelques Alcyonaires du Golfe de Tadjourah. Archives de zoologie expérimentale et générale (4) 8 (2): 179-266. Gray, J. E. 1848. Description of Sarcoptilus, a new genus of Pennatulidae. Proceedings of the Zoological Society of London 1848 (16): 45. Gray, J. E. 1860. Revision of the family Pennatulidae. Annals and Magazine of Natural History (3) 5: 20-25. Gray, J. E. 1870. Catalogue of sea-pens or Pennatulariidae in the collection of the British Museum. London: British Museum. GriEG, J. A. 1893. Bidrag til kjendskaben om de nordiske alcyonarier. Bergens museums arbog 1893 (2): 1-21. Herktots, J. A. 1858. Notices pour servir a l’étude des polypiers nageurs ou pennatulides. Bijdragen tot de Dierkunde 7: 1-31. HerkLots, J. A. 1863. Descriptions de deux espéces nouvelles de pennatulides des Mers de la Chine. Nederlandsch Tijdschrift voor de Dierkunde 1: 31-34. Hickson, S. J. 1900. The Alcyonaria and Hydrocorallinae of the Cape of Good Hope. Marine Investigations in South Africa 1: 67-96. Hickson, S.J. 1904. The Alcyonaria of the Cape of Good Hope. Part II. Marine Investigations in South Africa 3: 211-239. Hickson, S. J. 1909. Coelenterata: Anthozoa. Jn: HARMER, S. F. & SHipLey, A. E. eds. The Cambridge Natural History 1: 326-411. London: Macmillan & Co. Hickson, S. J. 1916. The Pennatulacea of the Siboga Expedition, with a general survey of the order. Siboga Expeditie Monogr. 14, Livr. 77: 1-265. Hickson, S. J. 1921. On some Alcyonaria in the Cambridge Museum. Proceedings of the Cambridge Philosophical Society 20 (3): 366-373. Hickson, S. J. 1930a. Some alcyonarians from the Eastern Pacific Ocean. Proceedings of the Zoological Society of London 1930 (1): 209-227. Hickson, S. J. 1930b. On the classification of the Alcyonaria. Proceedings of the Zoological Society of London 1930 (1): 229-252. Hickson, S. J. 1937. The Pennatulacea. Scientific Reports. John Murray Expedition, 1933-34 4 (5): 109-130. Husrecut, A. A. W. 1885. Ona new pennatulid from the Japanese Sea. Proceedings of the Zoological Society of London 1885: 512-518. 116 ANNALS OF THE SOUTH AFRICAN MUSEUM Hyman, L. H. 1940. The Invertebrates. Vol. 1. Protozoa through Ctenophora. New York: McGraw-Hill. JUNGERSEN, H. F. E. 1904. Pennatulida. Danish Ingolf-Expedition 5 (1): 1-95. KOLLIKER, R. A. von. 1872a. Anatomisch-Systematische Beschreibung der Alcyonariarien. I. Die Pennatuliden. Abhandlungen von der Senckenbergischen naturforschenden Gesell- schaft 7-8: 1-458. KOLLIKER, R. A. von. 1872b. Morphologie und Entwickelungsgeschichte des Pennatuliden- stammes nebst allgemeinen Betrachtungen zur Descendenzlehre. Frankfurt a. M.: Christian Winter. K6LLIkeR, R. A. von. 1874. Uber den Bau und die systematische Stellung der Gattung Umbellularia: eine vorlaufige Mittheilung. Verhandlungen der Physikalisch-medizinischen Gesellschaft zu Wurzburg 8: 13-18. KOLLIKER, R. A. von. 1875. Die Pennatulide Umbellula und zwei neue Typen der Alcyonarien. In: Festschrift zu Feier des ftinfundzwanzigjdhrigen Bestehens der Physikalisch- medizinischen Gesellschaft in Wiirzburg. KOLLIKER, R. A. voN. 1880. Report on the Pennatulida dredged by H.M.S. Challenger during the years 1873-1876. Report on the Scientific Results of the Voyage of H.M.S. Challenger during the years 1873-76. Zoology 1 (2): 1-41. Koren, J. & DanieELssen, D. C. 1847. Virgularia christii, n. sp. Nyt magazin for naturvidenskabernes 5 (3): 269-271. Koren, J. & DANIELSSEN, D. C. 1877. Contribution to the natural history of the Pennatulidae living on the Norwegian coast. Jn: SARS, M., Koren, J. & DANIELSSEN, D. C. eds. Fauna Littoralis Norvegiae 3: 82-102. KUKENTHAL, W. 1902. Diagnosen neuer Alcyonarien aus der Ausbeute der Deutschen Tiefsee-Expedition. Zoologischer Anzeiger 25 (668): 299-303. KUKENTHAL, W. 1903. Uber eine neue Nephthyidengattung aus dem siidatlantischen Ocean. Zoologischer Anzeiger 26 (694): 272-275. KUKENTHAL, W. 1910. Pennatuliden der Deutschen Tiefsee-Expedition. Zoologischer Anzeiger 36 (2/3): 51-58. KUKENTHAL, W. 1915. Pennatularia. Das Tierreich 43: i-xv + 1-132. Berlin: Verlag von R. Friedlander. KUKENTHAL, W. & Brocu, H. 1911. Pennatulacea. Wissenschaftliche Ergebnisse der Deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’ 1898-1899 13 (1) Lieferung 2: 113-576. LaAMARCK, J. B. 1816. Histoire naturelle des animaux sans vertébres. Paris: Verdiére. Licut, S. F. 1921. Notes on Philippine Alcyonaria. Part VI. New Philippine Pennatularia (sea pens) of the genus Lituaria. Philippine Journal of Science 19 (2): 247-255. LINDAHL, J. 1874. Om Pennatulid-Slagtet Umbellula Cuv. Kungliga Svenska vetenskapsakad- emiens handlingar 13 (3): 1-22. LINNAEUS, C. 1758. Systema naturae. Editio decima, reformata 1: 1-824. Holmiae. LINNAEUS, C. 1767. Systema naturae. Editio duodecima, reformata 1 (2): 533-1327. Holmiae. MANUEL, R. L. 1981. British Anthozoa. Synopses of the British fauna (n.s.) 18: 1-241. MarsHalt, A. M. 1887. Report on the Pennatulida dredged by H.M.S. Triton. Transactions of the Royal Society of Edinburgh 32: 119-152. MarsHaLL, A. M. & Fowter, G. H. 1887. Report on the Pennatulida dredged by H.M.S. ‘Porcupine’. Transactions of the Royal Society of Edinburgh 33 (2): 453-464. MarsHALL, A. M. & Fow er, G. H. 1888. Report on the Pennatulida of the Mergui Archipelago, collected for the Trustees of the Indian Museum, Calcutta, by Dr John Anderson, F.R.S., Superintendent of the Museum. Journal of the Linnean Society of London (Zoology) 21: 267-286. Mittarp, N. A. H. 1978. The geographical distribution of southern African hydroids. Annals of the South African Museum 74 (6): 159-200. MILNE Epwarps, H. & Haime, J. 1850. A monograph of the British fossil corals. Part 1. Introduction; corals from the Tertiary and Cretaceous formations. London: Palaeonto- graphical Society. Mo.anper, A. R. 1929. South and West African Octactiniae in the Gothenburg Natural History Museum. Géteborgs Kungliga vetenskaps- och vitterhetssémhalles handlingar (B) 1 (7): 1-16. SOUTHERN AFRICA PENNATULACEA 7 MULLER, O. F. 1776. Zoologiae Danicae Prodromus, seu animalium Daniae et Norvegiae indigenarum characteres, nomina, et synonyma imprimis popularium. Havniae—Vermes, Cellulana. Nuttina, C. C. 1908. Descriptions of the Alcyonaria collected by the U.S. Bureau of Fisheries steamer Albatross in the vicinity of the Hawaiian Islands in 1902. Proceedings of the United States National Museum 34 (1624): 543-601. NuttinG, C. C. 1912. Descriptions of the Alcyonaria collected by the U.S. Fisheries steamer ‘Albatross’, mainly in Japanese waters, during 1906. Proceedings of the United States National Museum 43 (1923): 1-104. Patras, P. S. 1766. Elenchus zoophytorum sistens generum adumbrationes generaliores et specierum cognitarum succinctas descriptiones cum selectis auctorum synonymis. Hagae Comitum. PASTERNAK, F. A. 1960. The deep-sea Pennatularia from the Bering Sea and Kuril-Kamtschatka Trench. Trudy Instituta okeanologii. Akademiya nauk SSSR 34: 329-335. (In Russian.) PASTERNAK, F. A. 1961a. Some new data on the specific composition and the distribution of deep-sea Pennatularia, genus Kophobelemnen, in Northern-Pacific. Trudy Instituta okeanologii. Akademiya nauk SSSR 45: 240-258. (In Russian.) PASTERNAK, F. A. 1961b. Pennatularia (Octocorallia) and Anthipatharia (Hexacorallia) obtained by the Soviet Antarctic Expedition in 1955-1958. Trudy Instituta okeanologii. Akademiya nauk SSSR 46: 217-230. (In Russian.) PASTERNAK, F. A. 1962. Pennatulids of the genus Umbellula Cuvier from the Antarctic and Subantarctic. I[ssledovaniya Fauny More 1: 105-128. (In Russian.) PASTERNAK, F. A. 1964. The deep-sea pennatularians and antipatharians obtained by R/S ‘Vitjaz’ in the Indian Ocean and the resemblance between the faunas of the pennatubrians of the Indian Ocean and the Pacific. Trudy Instituta okeanologii. Akademiya nauk SSSR 69: 183-215. (In Russian.) PASTERNAK, F. A. 1970. Sea pens (Octocorallia, Pennatularia) of the hadal zone of the Kurile—Kamtschatka Trench. Trudy Instituta okeanologii. Akademiya nauk SSSR 86: 236-248. (In Russian.) PASTERNAK, F. A. 1975a. Deep-sea pennatularians genus Umbellula from the Caribbean Sea and Puerto-Rican Trench. Trudy Instituta okeanologii. Akademiya nauk SSSR 100: 160-173. (In Russian.) PASTERNAK, F. A. 1975b. New data on the specific composition and distribution of the deep-sea pennatularians (Octocorallia, Pennatularia) of the Peru-Chile region and South Atlantic. Trudy Instituta okeanologii. Akademiya nauk SSSR 103: 101-118. (In Russian.) PASTERNAK, F. A. 1980. [Pennatularia Umbellula encrinus (L.) from the Canadian Basin of the northern Arctic Ocean.] Jn: VinoGRaDov, M. E. & MEL’Nikov, I. A. eds. The biology of the Central Arctic Basin: 236-239. Moscow: Nauka. PuHiLiprPl, R. A. 1835. Ueber Veretillum pusillum. Archiv ftir Naturgeschichte 1: 277-280. RicHtarpI, S. 1869. Monografia della famiglia dei Pennatularii. Archivio per la zoologia, l'anatomia e la fisiologia (2) 1: 1-150. Rou te, L. 1905. Notice préliminaire sur les pennatulides recueillis par le Travailleur et le Talisman, dans l’Océan Atlantique, au large du Maroc. Bulletin du Muséum d’histoire naturelle 11 (6): 454-458. Route, L. 1908. Alcyonaires d’Amboine. Revue suisse de zoologie 16 (2): 161-194. Sars, M. 1851. Beretning om en i sommeren 1849 foretagen zoologisk reise; Lofoten og Finmarken. Nyt magasin for naturvidenskaberne 6: 121-211. Simpson, J. J. 1905. A new cavernularid from Ceylon. Annals and Magazine of Natural History (7) 15 (90): 561-565. STEPHENSON, T. A. & STEPHENSON, A. 1972. Life between tidemarks on rocky shores. San Francisco: W. H. Freeman. STUDER, T. 1878. Uber die mit dem Schleppnetz angestellten Untersuchungen an der Westkus von Afrika wahrend der Reise S.M.S. Gazelle. Sitzungsberichte der Gesellschaft naturforschender Freunde zu Berlin 1878: 135-139. STUDER, T. 1879. Ubersicht der Anthozoa Alcyonaria, welche wahrend der Reise S.M.S. Gazelle um die Erde gesammelt wurden. Monatsbericht der Koéniglichen Preussischen Akademie der Wissenschaften zu Berlin 1878: 632-688. 118 ANNALS OF THE SOUTH AFRICAN MUSEUM STUDER, T. 1894. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California, in charge of Alexander Agassiz, carried on by the U.S. Fish Commission steamer ‘Albatross’, during 1891, Lieut. Commander Z.L. Tanner, U.S.N., commanding. X. Note préliminaire sur les alcyonaires. Bulletin of the Museum of Comparative Zoology at Harvard College 25 (5): 53-69. Tuomson, J. A. 1927. Alcyonaires provenant des campagnes scientifiques du Prince Albert Ier de Monaco. Résultats des campagnes scientifiques accomplies par le Prince Albert I 73: 1-77. Tuomson, J. A. & HENDERSON, W. D. 1905. Preliminary notice of the deep-sea Alcyonaria collected in the Indian Ocean. Annals and Magazine of Natural History (7) 15 (90): 547-557. THomson, J. A. & HENDERSON, W. D. 1906a. The marine fauna of Zanzibar British East Africa, from collections-made by Cyril Crossland, M.A., B.Sc., F.Z.S., in the years 1901 and 1902. Alcyonaria. Proceedings of the Zoological Society of London 1906 (1): 393-443. THomson, J. A. & HENDERSON, W. D. 19066. An account of the alcyonarians collected by the Royal Indian Marine Survey Ship Investigator in the Indian Ocean. Part I. The alcyonarians of the deep sea. Calcutta: Trustees of the Indian Museum. THomMsoN, J. A. & RENNET, I. 1927. Report on Japanese pennatulids. Journal of the Faculty of Science, Imperial University of Tokyo (4, Zoology) 1 (2): 115-143. THomson, J. A. & Simpson, J. J. 1909. An account of the alcyonarians collected by the Royal Indian Marine Survey Ship Investigator in the Indian Ocean. II. The alcyonarians of the littoral area. Calcutta: Trustees of the Indian Museum. Tuomson, J. S. 1915. The Pennatulaceae of the Cape of Good Hope and Natal. Memoirs and Proceedings of the Manchester Literary and Philosophical Society 59 (1): 1-26. THomson, J. S. 1917. The occurrence of Cavernularia Liitkenii KOll. in the Seas of Natal. Memoirs and Proceedings of the Manchester Literary and Philosophical Society 62 (3): 1-5. Tuomson, J. S. 1924. Charts and comparisons of the distribution of South African Alcyonaria. Transactions of the Royal Society of South Africa 11 (1): 45-84. TIxXIER-DURIVAULT, A. 1954. Les octocoralliaires d’Afrique du sud (II—Gorgonacea; I1I—Pennatulacea). Bulletin du Muséum national @ histoire naturelle (2) 26 (5): 624-631. TIXIER-DURIVAULT, A. 1960. Les octocoralliaires de I’Ile Inhaca. Bulletin du Muséum national @ histoire naturelle (2) 32 (4): 359-367. TIXIER-DURIVAULT, A. 1961a. Crassophyllum cristatum n. gen. et n. sp., type d’un genre de Pteroeididae (Pennatulacea). Bulletin du Muséum national d’histoire naturelle (2) 33 (4): 428-433. TIXIER-DuRIVAULT, A. 19616. Les octocoralliaires du Golfe de Guinée et des Iles du Cap-Vert (Alcyonacea, Pennatulacea). Campagne da la ‘Calypso’: Golfe de Guinée. Annales de l'Institut océanographique 39 (5): 237-262. TIxtER-DuRIVAULT, A. 1963. Alcyonacea et Pennatulacea de l’Afrique occidentale. Atlantide Report. Scientific Results of the Danish Expedition to the Coasts of Tropical West Africa, 1945-1946 7: 63-76. TIxIER-DuRIVAULT, A. 1966a. Quelques octocoralliaires australiens. Bulletin du Muséum national dhistoire naturelle (2) 37 (4): 705-716. TrxIER-DURIVAULT, A. 1966b. Octocoralliaires de Madagascar et des Iles avoisinantes. Faune de Madagascar 21: 1-456. TrxtER-DurRIVAULT, A. 1972. Nouvel rapport d’octocoralliaires de Madagascar et des Iles avoisinantes. Téthys, supplément 3: 11-68. TrxtER-DuRIVAULT, A. & D’Honpt, M. 1974. Nouvelles récoltes d’octocoralliaires a Madagascar. Téthys 5 (2-3): 251-266. Utinom1, H. 1961. Noteworthy octocorals collected off the southwest coast of Kii Peninsula, Middle Japan. Part 2. Telestacea, Gorgonacea and Pennatulacea. Publications of the Seto Marine Laboratory 9 (1): 197-228. VALENCIENNES, A. 1850. In: MtLNE Epwarp, H. & Haime, J. A monograph of the British fossil corals. Part 1. Introduction; corals from the Tertiary and Cretaceous formations. London: Palaeontographical Society. SOUTHERN AFRICA PENNATULACEA 119 VERRILL, A. E. 1864. List of the polyps and corals sent by the Museum of Comparative Zoology to other institutions in exchange, with annotations. Bulletin of the Museum of Comparative Zoology at Harvard College 1 (3): 29-60. VERRILL, A. E. 1865. Synopsis of the polyps and corals of the North Pacific Exploring Expedition, under Commodore C. Ringgold and Captain John Rogers, U.S.N., from 1853 to 1856. Collected by Dr. Wm. Stimpson, naturalist of the Expedition. With descriptions of some additional species from the west coast of North America. Proceedings of the Essex Institute, Salem, Mass. 4: 181-196. VERRILL, A. E. 1868. Notes on Radiata in the Museum of Yale College. No. 6. Review of the corals and polyps of the west coast of America. Transactions of the Connecticut Academy of Arts and Sciences 1 (2): 377-558. VERRILL, A. E. 1879. Notice of recent additions to the marine fauna of the eastern coast of North America. No. 3. American Journal of Science and Arts (3) 17: 239-243. VERRILL, A. E. 1882. Notice of the remarkable marine fauna occupying the outer banks off the southern coast of New England, No. 7, and of some additions of the fauna of Vineyard Sound. American Journal of Science and Arts (3) 24: 360-364. VERRILL, A. E. 1885. Results of the explorations made by the steamer Albatross off the northern coast of the United States in 1883. Report United States Commissioner of Fisheries 1883: 503-699. WEINBERG, S. 1976. Revision of the common Octocorallia of the Mediterranean circalittoral. I. Gorgonacea. Beaufortia 24 (313): 63-104. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu- larly 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. nov., sp. nov., comb. nov., syn. nov., 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 trans- ferred 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 (fig. 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 specimens 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, descrip- tion 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 Eliza- beth (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. 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 ...’: ‘. . . in C. namacolus (Fig. 10) . (b) The prefixes of prefixed surnames in all languages, when used in the text, if not ae 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 Punctuation should be loose, omitting all not strictly necessary Reference to the author should preferably 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. The generic name should not be abbreviated at the beginning of a sentence or paragraph. Name 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. GARY C. WILLIAMS THE PENNATULACEA OF | SOUTHERN AFRICA (COELENTERATA, ANTHOZOA) ‘OF THE SOUTH AFRICAN | KUM CAPE ‘TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (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) 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 and neat, double spaced with 3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for illustrations should be typed separately, their positions indicated in the text. All pages should be num- bered 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; 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 caption); the reduction or enlargement required should be indicated (and preferably uniform); orig- inals 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 caption; 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. If Letraset is used authors are requested to use Helvetica-style letter- ing, if possible. 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 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, 1969b) 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 (according to the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number 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 la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140. FiscHER, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie expérimentale et générale 74 (33): 627-634. Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and Magazine of Natural History (13) 2 (17): 309-320. Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of the Bingham Oceanographic Collection, Yale University 17 (4): 1-51. THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHULTzE, L. Zoologische und anthro- pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren 1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 99 Band May 1990 Mei Part 5 Deel FORAMINIFERA FROM THE LATE PLEISTOCENE (LATEST EEMIAN TO EARLIEST WEICHSELIAN) SHELLY SANDS OF CAPE TOWN CITY CENTRE, SOUTH AFRICA By I. K. McMILLAN 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 A358), HUD, 25, 8, rept), SAB, 3, TO, Gay tai), HEA), 8 TAD, 7), 1OM=9),, ING, 5, 7, tia), MiG), 1528), HO, 8), 27, BIGD), 326), 33, 36(2), 43(1), 45(1), 67(5), 84(2) Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 111 3 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, C632 FORAMINIFERA FROM THE LATE PLEISTOCENE (LATEST EEMIAN TO EARLIEST WEICHSELIAN) SHELLY SANDS OF CAPE TOWN CITY CENTRE, SOUTH AFRICA By I. K. McMILLAN Department of Geology, University College of Wales, Aberystwyth, Dyfed, United Kingdom* (With 20 figures and 4 tables) [MS accepted 29 March 1989] ABSTRACT Samples from shelly sands exposed during deep trenching in Cape Town city centre have yielded diverse and relatively well-preserved foraminifera faunas. Fifty-three species are recognized, of which four are reworked from older deposits; none are new. The total assemblage is very similar to ones from off the mouth of the Orange River, substantially further north. From foraminiferal and geological considerations, a latest Pleistocene (latest Eemian to earliest Weichselian) age is proposed for the sequence, which is regarded as having acccumulated during the fall in sea-level at the end of the Eemian. The foraminifera lived in a wave-dominated, littoral environment, and the clean sands in which they occur are considered to be part of an extensive, regressive sheet sand that originally covered almost the entire continental shelf off all three coasts of South Africa, and which is still widely preserved. Similar shelly sands, also attributable to the latest Eemian—earliest Weichselian sea-level fall, occur in the coastal parts of both the ‘Bredasdorp Formation’ and the ‘Alexandria Formation’ (both termed Unit III), as well as along the Gamtoos coast (Unit III), at Durban (the Bluff Sandstone Formation), and extensively over the Zululand coastal plain (the Uloa Formation). Variable diagenesis has resulted in considerable cementing of these deposits at some localities. The foraminifera of the same sequence at several localities on the west coast are given in three appendices: for the Llandudno—Sandy Bay coast, Cape Peninsula, the Berg River mouth, and the Hoedjiespunt Peninsula, Saldanha. CONTENTS PAGE TNCKOGIC HO Meni rarer a eet ree ncaa « leales Ws WPetdnin tus 122 JPRREAIOIOS WHOS oreo G.c dhe ei 0.5 CRO ONG Ee UCI Re a a 122 hcworaminiieraassemblAGeSine amass ese) sles Sele wot sede anor 125) EVIL One ntzOMmdepOstiOnesi we. is oe dae oe eee ee oes See 125 Stratigraphy andeconmelatiomene: see sei acide Se Cac e ean as 126 Synchronous deposits of the South African continental margin..... 126 Companisoniwitholder Pleistocene deposits .44-.4.55-6- 54045000" 129 Companisoniwith Holocene depositsinn sn.) ons ssse5 ase ea se 3h INC WOK Speman yeny ter. er en tnta uve PG ees Ahan Gan Wate clea uaa aes 1132 EALOCESS TMS MMIC LAOS Agta rae cei pet keels ecaiestin era tuniresos Soi Jiang, Sake ue oes 133 Rhotographysandsdepositioniotmatenialtes s+ ee eoe ano see oe 1133) FOKAMMMILCEAM AKON OMe ne emcee eee oh ikk jase tah a staels ee cole on 135 NCKMOWLe AS CINE MUS ameristar ids eth, claasv etal alee SRS 173 INC IC KCNC SMB rye atte race (ke Menace ac Gade weno gai ath Nea ee a 174 * Present address: SOEKOR (Pty) Ltd, P. O. Box 307, Parow 7500, South Africa. WA). Ann. S. Afr. Mus. 99 (5), 1990: 121-186, 20 figs, 4 tables. Oy ANNALS OF THE SOUTH AFRICAN MUSEUM PAGE Appendix A: The Llandudno-Sandy Bay coast, Cape Peninsula....... 179 Appendix B: Latest Pleistocene Foraminifera from Berg River mouth . . 182 Appendix C: Latest Pleistocene Foraminifera of Hoedjiespunt Peninsula, Salama. 5252 se oe Ree he EN ct ns oe eo eee 184 INTRODUCTION During deep trenching by the Municipality along several of Cape Town’s city centre streets in late 1986, clean white, rather bleached, shelly quartz sands were exposed. These deposits appear to be confined to the more seawards parts of the city (Fig. 1). The sands are distinctly bedded, and undisturbed; human interference is confined to a distinct zone overlying the sands. Despite the clean and somewhat coarse nature of the sands, studied samples yielded a fairly well-preserved and diverse foraminifera fauna that contains a number of age-diagnostic species. The whole assemblage is almost identical to that described by McMillan (1987b) from the Late Pleistocene (latest Eemian to earliest Weichselian) deposits occurring on the inner Orange Shelf, just to the north of the Orange River mouth. A summarized stratigraphic column, based on observaticns by Mr V. H. Valicenti of SOEKOR (Pty) Ltd, Cape Town, at the time the trench was open, is given in Figure 2. The total thickness of the Cape Town sequence is not known. Two samples each were collected at the corner of Hertzog Boulevard and Oswald Pirow Street, and at the corner of Oswald Pirow and Martin Hammerschlagweg. Results of the foraminifera study are given in Table 1. PREVIOUS WORK Studies of Quaternary foraminifera in South Africa were initiated by Chapman (1907), who examined samples from a drill hole in the bed of the Buffalo River, East London. The precise age of the sequence, studied to a depth of 33,2 m below the bed of the river, is probably partly of Holocene, but mostly of latest Eemian—earliest Weichselian age. The latter age is indicated by the presence of Elphidium crispum (Linné) (given as Polystomella crispa by Chapman) through most of the studied sequence, suggesting that the Holocene is less than 4,57 m thick. Further comments on the use of Elphidium crispum as a zone fossil for the latest Eemian to earliest Weichselian deposits of South Africa are given below in the section on stratigraphy (see p. 129). Most later work has involved lists of foraminifera species only, but Biesiot (1957) illustrated a number of the in situ foraminifera of the Uloa Formation at its type locality (see McMillan 1987a); these include the species of Amphistegina, Elphidium, and perhaps some of the miliolids, together with Eponides zuluensis Biesiot, ?Eponidella sp. sensu Biesiot, Rotalia beccarii sensu Biesiot, Cibicides lobatulus (Walker & Jacob), and possibly Nonion elongatum sensu Biesiot. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 123 18°25'E (e} Se ae cae a e «za (0) J 5 Km CAPE TOWN City Centre s 33°56'S Fig. 1. Location of studied samples from the latest Eemian—earliest Weichselian regressive shelly sands (Unit III) beneath Cape Town city centre. Samples 1 and 2 are from the corner of Oswald Pirow Street and Hertzog Boulevard; 3 and 4 from the corner of Oswald Pirow Street and Martin Hammerschlagweg. Based on 1 : 50 000 topographic sheet 3318CD (Cape Town), 4th edition. Street level Rubble and soil later than 1693 A.D. Latest Eemian- earliest Weichselian Bedded and Shelly Sands Fig. 2. The stratigraphy of exposures in deep trenches in Oswald Pirow Street, Cape Town city centre, as measured in November 1986, by Mr V. H. Valicenti, SOEKOR (Pty) Ltd. Sampled horizons are indicated (CTC 1 to 4). 124 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 1 Distribution of foraminifera in studied samples from the latest Eemian—earliest Weichselian sands of Oswald Pirow Street, Cape Town city centre. Sample number Sp. Species CTC-1 CTC-2 CTC-3 CTC-4 no. 1 1 1 = Spiroplectammina sp. 1 1 2 Quinqueloculina contorta 8 6 18 28 3. Quinqueloculina dunkerquiana Z 4 Quinqueloculina isabellet 2, 2, 5 Quinqueloculina cf. Q. lata 1 1 1 >) 6 Quinqueloculina seminulum 11 7 Quinqueloculina triangulans 1 Z 8 Quinqueloculina cf. Q. undulata 4 6 24 34 9 Quinqueloculina spp. (broken) y 4 10 Siphonaperta sp. 2 1 7 3) 11 Milolinella subrotunda 1 12 ?Dentalina sp. (reworked) 1 13. Lagena semilineata var. 1 14 Lenticulina sp. 1 15. = Oolina squamososulcata 1 3. 16 Oolina sp. A 1 17 _—‘-Fissurina lucida 1 18 = Fissurina cf. F. marginata 2 19 Fissurina spp. (reworked) 1 1 20 Guttulina spp. (broken) 1 21 Bulimina marginata 1 22 Uvigerina sp. (reworked) 1 23 Trifarina angulosa 1 24 ~Bolivina sp. 1 25. Brizalina spathulata 16 8 26 Cassidulina laevigata 1 27 ~~ Cassidulina crassa s.1. 1 28 ?Gavelinopsis sp. (reworked) 1 29 Planulinoides biconcavus 1 1 30 = @Rosalina bradyi 1 31 =Rosalina sp. 2 dip) 10 32 Glabratella australensis 3) 33 Cibicides fletchen s.1. 5 a, 24 13. 34 ~=©Cibicides lobatulus 4 4 35 Cibicides spp. (reworked) y 5 y) 3. 36 = Cibicidoides sp. 1 37. Ayalinea balthica 1 1 1 38 Sphaerogypsina globulus 1 39 = ?Acervulina sp. 14 3. 40 Elphidium advenum 6 1 41 Elphidium sp. A 1 4 42 — Elphidum articulatum 54 26 131 137 = 43, —Ss Elphidium crispum s.1. 2 6 130 44 -~— Elphidium macellum s.1. 1 45 Elphidium sp: B 1 46 Astrononion echolsi 9 2 47 Ammonia japonica 1 i 3 6 48 Ammonia parkinsonia s.1. 11 16 105 146 49 Pararotalia nipponica 1 50 indeterminate reworked rotalid 51. Globigerina (G.) bulloides Globorotalia (G.) inflata 53 Neogloboquadrina cf. N. pachyderma Ne N N 91 V2 429 452 Total number of tests PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 125 All the remaining fossils detailed by Biesiot (1957) are reworked from the Early and Middle Miocene (McMillan 1987a). Lists of foraminifera from the coastal ‘Bredasdorp Formation’ and the coastal ‘Alexandria Formation’ were given by Rtmke, in Spies etal. (1963) and Engelbrecht et al. (1962). Some of the samples described by Parr (1958), including those from the Bluff at Durban, are also of this age. It is now certain that the Pararotalia cf. P. inermis (Terquem) emend. Le Calvez, and the Ammonia beccarii (Linné) var. 2 sensu McMillan (1974), from the Agulhas Bank, are also of latest Eemian—earliest Weichselian age. The former can now be referred to Pararotalia nipponica (Asano), whereas the latter are poorly preserved tests of the Ammonia sp. C (Alexandria) detailed by McMillan (1987a). McMillan (1987a) illustrated Ammonia species from a variety of South African Pleistocene, Holocene and Pliocene localities as a preliminary attempt to correlate the many, geographically disparate deposits of these ages. McMillan (1987b) described in detail the stratigraphy and foraminiferal assemblages of the latest Eemian-—earliest Weichselian and the Holocene silts and sands of the Orange Shelf, southern Namibia continental margin. Diagnostic zone foramini- fera from the different aged rock units that comprise the ‘Alexandria Formation’ have been illustrated by McMillan (in press). THE FORAMINIFERA ASSEMBLAGES All four Cape Town samples yielded faunas dominated by the benthonic species Elphidium crispum (Linné) s.l. and Pararotalia nipponica (Asano). As is the case with all of the Pleistocene marine sand deposits preserved on the South African continental margin, planktonic foraminifera are rare. Species of the benthonic genera Elphidium, Ammonia, Pararotalia and Cibicides predominate, again in keeping with other Pleistocene marine sequences. ENVIRONMENT OF DEPOSITION A littoral, wave-influenced environment of deposition is indicated from both the foraminifera and the sediments. The clean nature of the sands, the abrasion and breakage of some foraminifera tests, and the presence of a number of foraminifera species that in life attach to a substrate, all clearly reflect a turbulent sea-floor, with highly oxygenated conditions both on the sea-bed and in the water column. A water depth of 10 m or less is suggested. The attached species include Cibicides lobatulus (Walker & Jacob), Rosalina sp., Rosalina bradyi (Cushman), Glabratella australensis (Heron-Allen & Earland) and Sphaerogypsina globulus (Reuss), and possibly Cibicides fletcheri Galloway & Wissler s.1. The macrofaunal remains in the sands, mainly bivalve and echinoid debris, are substantially comminuted. 126 ANNALS OF THE SOUTH AFRICAN MUSEUM It appears likely that the foraminifera tests, while evidently in situ in a stratigraphical sense, were much transported before coming to their final rest. It seems likely that these foraminifera lived in a sublittoral environment, possibly a kelp-bed, and were swept shorewards after death by wave action, to be incorporated into truly littoral deposits accumulating as the shoreline retreated. The retreat of the sea at this time would seem to have occurred over a fairly short interval (perhaps well within 5 000 years, if an analogy with the Holocene transgression holds good), so that the littoral sands laid down at any one locality may have soon no longer been subject to wave- and swell-induced abrasion, lying instead in coastal sand flats or dunes. The presence of Ammonia parkinsoniana (d’Orbigny) s.1. in small numbers in the Cape Town samples suggests that the assemblage, if regarded as from one environment, derives from an area of slightly reduced salinity conditions. Thus, river output or other run-off affected the habitats of the studied foraminifera, but the lack of land-derived plant debris, seeds, charophyte oogonia and so on, together with the relatively small numbers of Ammonia parkinsoniana, clearly indicate the freshwater influence to have been an insubstantial one, perhaps as is the case in the Cape Town part of Table Bay at the present day. It seems that a comparison of numbers of Ammonia japonica (Hada), typical of shelf environments, with numbers of Ammonia parkinsoniana, characteristic of estuarine and littoral environments, may provide a simplified environmental indicator for these latest Pleistocene accumulations around the west and south coasts of South Africa. Since most of the foraminifera species encountered in the Cape Town sands are known to live off the western Cape coast at the present day, it would appear that sea temperatures at the end of the Eemian were much the same as they are at present. However, until the full geographic ranges of the various species around the South African coast are better known, an improved understanding of water temperatures cannot be attempted. STRATIGRAPHY AND CORRELATION The foraminifera of the Cape Town sands show varying similarities to those of older Pleistocene deposits, those of deposits of the same age, and those of the Holocene. Generalized locations are given in Figure 3. SYNCHRONOUS DEPOSITS OF THE SOUTH AFRICAN CONTINENTAL MARGIN The Cape Town shelly sands can be correlated with ease to the sands and silts occurring to a considerable thickness (about 70 m according to Hoyt et al. 1969) off the mouth of the Orange River, based on the very close similarity of their foraminiferal faunas (see McMillan 1987b). Correlation is also possible with a number of deposits preserved about 8 m above high sea-level at the Hoedjiespunt Peninsula, Saldanha, along the Sandy Bay to Llandudno coastline, Cape PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 27 MAPUTO 33 5S % WA -4~—=Jc holes Bluff Sst. Formation Aliwal Shoal HOEDJIESPUNT CAPE TOWN LLANDUDNO- ‘Alexandria Formation’ =o a a ~ PLETTENBERG BAY Bredasdorp Formation ~___ Fig. 3. Distribution of latest Eemian—earliest Weichselian deposits around the South African coast, of equivalent age to the shelly sands of Cape Town centre. The areas shown for the Varswater Formation, the ‘Bredasdorp Formation’, and the ‘Alexandria Formation’ are for the entire sequence, Unit I to Unit III. In all these areas Unit III occurs as a narrow strip along the present coast. The major foraminiferal facies, based mainly on Ammonia and Elphidium species are also shown. Species listed distinguish the Unit III sands from both older Pleistocene and Holocene deposits. Peninsula, and below the level of the present Berg River mouth (see Appendices A, B, C). Deposits of the same age also occur along the south coast of South Africa: the latest, most coastal portion of the ‘Bredasdorp Formation’ between Cape Agulhas and Mossel Bay, below the mouth of the Keurbooms and Bietou rivers at Plettenberg Bay and onshore adjacent to the estuary, and the latest, most coastal portion of the ‘Alexandria Formation’ between Port Elizabeth and Port Alfred, as well as the whole of the smaller area of ‘Alexandria Formation’ along the Gamtoos coast. Deposits of this age have been termed Unit III informally by the present author (see McMillan in press). It is now known that Unit III and Unit IV of the ‘Alexandria Formation’ are slightly different facies of the same rock sequence, and are of the same age (see McMillan in press). Unit III in the eastern Cape is characterized by polished foraminifera tests, with many normal marine shelf species, and the distinctive Ammonia sp. C (Alexandria) is common and widespread. Unit IV, seen only at Woody Cape, contains foraminifera assem- blages dominated by miliolids, together with Poroeponides lateralis (Terquem), that derive from a more exclusively littoral environment. Off the south coast, Unit II] sediments are extensively preserved as a thin veneer over the majority of the Agulhas Bank. The sequence has been frequently 128 ANNALS OF THE SOUTH AFRICAN MUSEUM encountered in borehole tops as far offshore as the outermost shelf, down to a water depth of about 200 m. Foraminifera faunas remain uniform and distinctive throughout, and are easily correlated with those deposits occurring onshore at the present day. At borehole F—A 13, the latest Eemian—earliest Weichselian shelly deposits, together with the overlying Holocene interval (the two are separated by an unconformity), are no more than 0,5 m thick. Ammonia species from F—A 13 and E-B 1 have been previously discussed by McMillan (1987a). Where the latest Eemian-—earliest Weichselian Unit III deposits unconformably overlie the shelly white limey clays of the Early Miocene (Burdigalian) over most of the middle and outer shelf on the Agulhas Bank, much reworking of the Miocene has occurred, so that the latest Pleistocene sediments are often white clays, with abundant Miocene foraminifera in some cases. The thin sliver of coastal deposits preserved along the East London to Durban coast have only been examined for foraminifera by Parr (1958); the species listed by him from near Bats Cave, East London, are insufficient to determine what part of the Pleistocene (but presumably Unit III) is represented here. It seems certain now that most, possibly all, of the Bluff Sandstone Formation at Durban and the Uloa Formation, extensively developed over the Zululand coastal plain, are of the same age—latest Eemian to earliest Weichselian. Although very much thinner offshore, the sequence also occurs in the borehole tops of Jc-A 1, Jc-B 1 and Jc—C 1, where it lies unconformably between the Holocene and Middle or Early Miocene, and consists of a loose shelly sand. The eastern offshore sequence thus exactly duplicates that seen on the Agulhas Bank. Again, there is little change in the littoral, benthonic foraminifera of this unit, and Ammonia sp. C (Uloa) (possibly referable to Challengerella persica Billman, Hottinger & Oesterle), Elphidium crispum (Linné) s.1., Planorbulinella larvata (Parker & Jones) and Cellanthus craticulatus (Fichtel & Moll) have proved useful local indicators for Unit HI from Durban northward. Earlier comments on the age of the Uloa Formation have been made by McMillan (1987a), who suggested reasons for a probable middle Pleistocene age. However, the presence of Elphidium crispum s.1., together with feathered-suture Ammonias in the east-coast and the south-coast sequences, as well as their unconsolidated nature on the outer shelf off both the east and south coasts, clearly argues for the two to be of the same age. Since the offshore deposits remain unconsolidated for the most part, it is clear that they have never been subjected to a major regression of the sea and the surf-zone, and must thus date from the last fall in sea-level at the end of the Eemian. It is clear from study of the Holocene sequences around South Africa that the early Holocene transgression, in strong contrast, was a very mild event in terms of its erosive power. On the basis of the presence of Elphidium crispum (Linné), the Aliwal Shoal, south of Durban (Carter 1966; McCarthy 1967), would appear to be also of Unit III age. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 129 In all of the above rock sequences, correlation of this youngest regressive event can be achieved in part by its contained littoral foraminifera, particularly the species of Elphidium and Ammonia, and partly by its stratigraphical relationships. Although, on following Unit III around the continental margin of South Africa, the foraminifera faunas change substantially, nevertheless with sufficient density of samples, key species can be traced until they overlap with other key species, and correlation is thus maintained. Elphidium crispum (Linné) s.l. has proved the most reliable species for identifying Unit III, though there are subtle differences in test morphology between east- and west-coast examples that remain to be defined. The major foraminiferal facies recognized in Unit III deposits so far, together with Jocal age-diagnostic species, are shown in Figure 3. McMillan (1987a) has proposed the use of different Ammonia species to identify the various latest Cainozoic (Pliocene, Pleistocene and Holocene) marine deposits of southernmost Africa. Subsequent unpublished work on the Elphidium species has indicated that they are morphologically more diverse and often more distinctive than the Ammonia species, and thus would seem to be a more useful group for correlation purposes. Perhaps in response to the changing environments available for occupation during high (glacial minima) and low (glacial maxima) sea-level stands by species of the two genera, it would seem that appearances of new species are more frequent than in the planktonic foraminifera (Globigerinacea) over the timespan of the Quater- nary. Particularly in Ammonia, but also in Elphidium, the historical view has been that species of these genera show much morphological variation in their tests. On the contrary, intraspecific variation is as limited as in most other rotalid species, but the number of species in both genera is unusually large, and most appear to have survived for relatively short time periods, often markedly less than one million years in duration. COMPARISON WITH OLDER PLEISTOCENE DEPOSITS The regressive littoral sands of the end of the last interglacial can be identified as such in a number of ways. From a foraminiferal point of view, they contain the greatest number of species in common with extant littoral assemblages around South Africa. The foraminifera of the earlier Pleistocene and Pliocene deposits show a progressive decline in similarity with increasing age. A typical example of this is given by McMillan (in press) for the ‘Alexandria Formation’. In both the oldest, Pliocene part of the ‘Alexandria Formation’, and the Pliocene of the Orange Shelf (McMillan 1987b), the benthonic foraminifera show no similarity with living South African species, though in deeper-water assemblages some conservative forms persist (e.g. Pullenia, Gyroidina). Secondly, the latest Eemian—earliest Weichselian deposits can be recognized because they consist of unconsolidated or poorly cemented shelly sand, up to 86 m thick in the coastal ‘Alexandria Formation’, but often less than 1 m, that 130 ANNALS OF THE SOUTH AFRICAN MUSEUM covers most of the South African continental margin, though there are large areas where it is missing off the west coast. This sheet sand has suffered relatively little from later erosion. In contrast, relatively litthe now remains of the earlier Pleistocene regressive deposits, the remnants usually being unusually thick because of local, mild subsidence prior to deposition, and uplift after deposition has removed them from the erosive activities of subsequent regressions. Earlier Pleistocene deposits of this type are known from the west coast, especially in the vicinity of Saldanha and Langebaan, and in the older ‘Bredasdorp Formation’ and ‘Alexandria Formation’, inland from the present coast. It seems probable that the older, higher terraces (A to C) of the Oranjemund to Chameis Bay coast in southern Namibia should be included here, but foraminiferal study of them has not yet been attempted. Thirdly, there is a tendency for the Unit III shoreline at maximum advance in the Eemian to be similar to the present shoreline, with the only major differ- ence occurring on the Zululand coastal plain, which was entirely inundated during the Eemian. With the older Pleistocene and Pliocene deposits, a progressive divergence away from the present shoreline can be seen, which by Pliocene times was quite substantial, notably in the eastern Cape. Off the south and west coasts of South Africa, the marine deposits of the latest Eemian to earliest Weichselian can be distinguished from older units by the particular presence of Elphidium crispum (Linné) s.l., as well as Elphidium macellum (Fichtel & Moll) s.l., Ammonia japonica (Hada), and Ammonia parkinsoniana (d’Orbigny) s.l. (see McMillan 1987a, 1987b). In addition, Astrononion echolsi Kennett, Quinqueloculina isabellei d’Orbigny, Oolina sp. A, Elphidium cf. E. advenum (Cushman) and Cassidulina crassa d’Orbigny s.1., as well as the very distinctive Quinqueloculina cf. Q. undulata d’Orbigny, are all unique to the latest Eemian to earliest Weichselian of the west coast (McMillan 1987b). Also characteristic of these youngest marine Pleistocene deposits is the widespread presence of mollusc shell retaining its life colours: purples, blues and pinks seem the most commonly preserved. In earlier Pleistocene deposits, all the mollusc shell is white. Preliminary examination of the Varswater Formation (here regarded as later Early Pleistocene) at a number of localities, as well as the phosphatic lithofacies termed the ‘Saldanha Formation’ by Tankard (1975), has revealed the following foraminifera species: Ammonia sp. (similar to Ammonia japonica), Elphidium sp. (within the Elphidium crispum—macellum group), Glabratella australensis (Heron- Allen & Earland) locally in abundance, Planulina cf. P. ariminensis d’Orbigny and ?Gavelinopsis sp. Most of these are unknown in the latest Eemian to earliest Weichselian of both the Orange Shelf (McMillan 1987b) and Cape Town city centre. The Pelletal Phosphorite Member of the Varswater Formation in phosphate exploration borehole Q12 from near the New Varswater Quarry, Langebaanweg, has yielded a somewhat different assemblage, with Cibicides lobatulus (Walker & Jacob), Pararotalia nipponica (Asano), Elphidium advenum (Cushman), Pseudo- PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 131 nonion cf. P. chiliensis (Cushman & Kellett), Nonion boueanum (d’Orbigny), Trifarina angulosa (Williamson), Oolinasp., Elphidium spp., Rosalina cf. R. bradyi (Cushman), Planorbulina mediterranensis d’Orbigny, ?Gavelinella spp.., Cibicides spp., Ammonia spp., ?Pseudononion sp., and Globigerina cf. G. bul- loides d’Orbigny. More work is necessary on these older Pleistocene deposits but, from the abundance of Glabratella australensis at many localities, a correlation of the Varswater Formation with Unit IIA of the ‘Bredasdorp Formation’ seems likely. It must be emphasized that the majority of the foraminifera species employed in correlation of the southern African latest Cainozoic are of local value only. Ammonia japonica, Ammonia parkinsoniana, Elphidium crispum and Elphidium macellum, for example, range back to the Early Pleistocene or Pliocene in the Mediterranean or Japan. In the case of Ammonia and Elphidium, both were late arrivals to southern Africa. Ammonia is first seen in the Early Miocene of Mediterranean Europe, but in South Africa it first appears in the Early Pliocene. Elphidium is known from the Early Eocene (Loeblich & Tappan 1964), but first appears in South Africa in the Middle Miocene. COMPARISON WITH HOLOCENE DEPOSITS The bedded shelly sands exposed in the Cape Town municipal trenching lie seaward of the Holocene shoreline as it was in 1693, before the construction of the modern harbour works. The location, relative to the site of the Castle, is evident on old maps. There thus exists the possibility that the sands are Holocene in age, having accumulated prior to commencement of the land reclamation schemes associated with harbour development. However, a Holocene age can be discounted on a number of points. The latest Pleistocene Unit III sands are rich in tests of Elphidium crispum (Linné) s.1., and this species is absent everywhere in the Holocene, except rarely where it has been reworked. Similarly, Quinqueloculina cf. Q. undulata d’Orbigny and the other species listed on page 139 are not known from the Holocene (see Fig. 4 for interpreted stratigraphic ranges). The absence of Elphidium magellanicum Heron-Allen & Earland, widespread in shallow marine environments off the west coast during the later Holocene (McMillan 1987b), also supports a pre-Holocene age. Elphidium magellanicum occurs in the present-day sands washed into the canalized mouth of the Soutrivier, just a little distance north-east of the study area. Finally, nearly all the tests from the trench sands are bleached and lack their life colours, whereas tests from Holocene samples include a high proportion that retain their natural golden-brown coloration. This is probably due to the destruction of the inner tectin lining of the test by bacteria and percolating groundwater. Since the latest Eemian—earliest Weichselian sands are overlain by a mix of rubble and soil, the fate of any Holocene deposits at the site of the trenching, laid down prior to land reclamation, is unclear. 132 ANNALS OF THE SOUTH AFRICAN MUSEUM HO _TORANGE SHELF Hil STRATIGRAPHY ir S P Ey Gages Elphidium magellanicum Discammina compressa Ammonia japonica Ammonia parkinsoniana s.l. Hyalinea balthica Elphidium macellum s.l. Elphidium articulatum Elphidium advenum Pararotalia nipponica Cibicides lobatulus Glabratella australensis Elphidium Quinqueloculina cf. undulata crispum s.l. Quinqueloculina isabellei Oolina sp. A Elphidium cf. advenum Astrononion echolsi Cassidulina crassa s.l. Fig. 4. Range chart of selected time-restricted benthonic foraminifera species from the latest Quaternary littoral of the west Cape and Namibia coasts. Those species marked with an asterisk range back to near the Pliocene—Pleistocene boundary (base of Unit II) in the “Alexandria Formation’; that marked by a cross ranges back in abundance to the later Early Pleistocene (base of Unit ITA) of the ‘Bredasdorp Formation’. REWORKING A small number of the Cape Town tests are rather more poorly preserved than the majority, and are regarded as having been reworked. One, ?Gavelinopsis Sp., appears to be derived from the local equivalent of the Varswater Formation (later Early Pleistocene) sandy limestones, as it is creamy in colour and cemented; the same species occurs in the limestones at Saldanha. Occasional angular lumps of creamy sandy limestone present in the Cape Town sands are probably also from an equivalent of the Varswater Formation. The remaining reworked tests may prove to be from the Early Miocene, though this cannot be confirmed from the species present, which are ?Dentalina sp., Cibicides spp. and Uvigerina sp. (Fig. 9B). All show darker infillings to the chambers (and dark test calcite) than is typical for later Neogene or Quaternary foraminifera tests. An Early Miocene age seems most likely, because of the extensive development of very fossiliferous limy clays of this age that outcrop or subcrop below the Quaternary on the continental shelf off the west coast. Intensive reworking of Early and Middle Miocene foraminifera tests into latest Pleistocene regressive sands is known from the Orange Shelf (McMillan 1987b) and the Uloa Formation of Zululand (McMillan 1987a). The species listed above are not typical of the Varswater Formation. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 139 PROCESSING METHODS All of the samples studied, including those described in the three appendices, consist of clean or muddy, unconsolidated sands. These were easily washed with a cold-water spray over a 240 mesh/63 micron sieve. Those samples with substantial present-day plant debris were further cleaned by standing the washed residue in a beaker of water and picking off the floating debris using a pair of tweezers. PHOTOGRAPHY AND DEPOSITION OF MATERIAL Scanning electron microscope work was performed using the JEOL JSM240 microscope of the University College of Wales, Aberystwyth. The line drawings were executed with a camera lucida attachment to a light microscope. All illustrated specimens were returned to their relevant assemblage slides; they can be distinguished by their gold coating. The eleven assemblage slides have been deposited with the Ph.D. thesis material of McMillan (198765) in the foraminifera collections of the Department of Geology, University College of Wales, Aberystwyth. The slides possess the following McMillan collection numbers: CTC-1: Slide 118 C€TC—2: - Slide 119 CTC-3: Slide 120 CTC-4: Slide 121 BR-1: Slide 115 BR-2: Slide 116 SB-1: Slide 111 SB-LL1: Slide 112 SB-LL2: Slide 113 SB-LL3: Slide 114 H=1: Slide 117 A representative set of species slides has also been deposited in the Micro- palaeontology Unit of the South African Museum. FORAMINIFERA TAXONOMY The taxonomic scheme followed is essentially a combination of that of Loeblich & Tappan (1964, 1974), as revised by Haynes (1981). More detailed discussions and full descriptions of many of the species listed below can be found in McMillan (1987b). 134 ANNALS OF THE SOUTH AFRICAN MUSEUM Family Textulariidae Ehrenberg, 1838 Subfamily Spiroplectammininae Cushman, 1927 Genus Spiroplectammina Cushman, 1927 ?Spiroplectammina sp. Fig. SA-B Remarks Two badly broken tests, which seem more likely to be referable to Spiroplectammina than Textularia from their test shape. The specimens differ from Spiroplectammina atrata (Cushman), prevalent in the Holocene deposits of the middle and outer Orange Shelf (McMillan 1987b), in being less compressed and with a more rounded test periphery. The latter feature, however, is poorly preserved. The grain size of the test wall is also coarser than either Spiroplectammina atrata from off the west coast or Spiroplectammina wrightii (Silvestri), known from the Holocene of the Agulhas Bank (McMillan 1974). Family Milliolidae Ehrenberg, 1839 Subfamily Quinqueloculininae Cushman, 1917 Genus Quinqueloculina d’Orbigny, 1826 Quinqueloculina contorta d’Orbigny, 1846 Fig. 6A—C Quinqueloculina contorta d’Orbigny, 1846: 298, pl. 20 (figs 4-6). McMillan, 1974: 33, pl. 2 (fig. la—c); 1987b: 164, pl. 2 (figs 8-11). Remarks Some variation in the degree of indentation of the rectangular test margins is evident in southern African individuals. Those from the Agulhas Bank (McMillan 1974) display strong indentations, but examples from the west coast are flatter (McMillan 1987b, present study) and more in keeping with the type illustrations of the species (D’Orbigny 1846). Some of the Orange Shelf tests display distinctly inflated margins, and the rectangular feature is almost lost (McMillan 19875, pl. 2 (figs 8-9)). The toothplate of the Cape Town tests is perhaps a little more bladed and more elongate than that originally figured by D’Orbigny. The species occurs in small numbers in the more nearly littoral intervals of the latest Eemian—earliest Weichselian of the Orange Shelf (McMillan 19875), and over the middle and outer shelf on the Agulhas Bank in the Holocene (McMillan 1974). This discrepancy in environments between the latest Pleis- tocene and the Holocene is mirrored by several other species found in the Cape Town shelly sands. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 135 Fig. 5. A-B. ?Spiroplectammina sp. A. Side view, CTC 4. x 130. B. Apertural view, same specimen. X 141. C-—E. Quinqueloculina isabellei dOrbigny. C. Side view, CTC 4. xX 67. D. Apertural view, same specimen. X 520. E. Edge view, same specimen. xX 60. F. Quin- queloculina cf. Q. lata Terquem. Side view, CTC 4. x 117. G—H. Quinqueloculina triangu- laris d’Orbigny. G. Side view, CTC 4. xX 97. H. Apertural view, same specimen. xX 113. I. Quinqueloculina cf. Q. undulata d’Orbigny. Apertural view, CTC 2. xX 108. 136 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 6. A-C. Quinqueloculina contorta d’Orbigny. All same specimen, CTC 3. All x 67. A. Side. B. Opposing side. C. Apertural view. D-F. Quinqueloculina dunkerquiana (Heron- Allen & Earland). All same specimen, CTC 4. All x 67. D. Side. E. Opposing side. F. Apertural view. G—I. Quinqueloculina seminulum (Linné). All same specimen, CTC 3. All x 67. G. Side. H. Opposing side. I. Apertural view. Quinqueloculina dunkerquiana (Heron-Allen & Earland, 1930) Fig. 6D-F Miliolina dunkerquiana Heron-Allen & Earland, 1930: 56, pl. 12 (figs 9-11). Quinqueloculina dunkerquiana (Heron-Allen & Earland) Haynes, 19736: 2, pl. 1 (fig. 8). McMillan, 1987b: 166, pl. 2 (figs 12-13). Quinqueloculina cf. Q. dunkerquiana (Heron-Allen & Earland) McMillan, 1974: 33, pl. 2 (fig. la—c). Remarks A number of small Quinqueloculina specimens with test width, height and outline, and chamber shape much in keeping with the north-west European shallow marine species Quinqueloculina dunkerquiana. As with the shell figured by Heron-Allen & Earland (1930) from present-day sediments of the Plymouth district, south-western England, the maximum height of tests is little greater than their maximum width. The distinctive indentations on the two last-formed chambers give a slightly sigmoiline appearance to the test if seen in apertural view, but the internal arrangement is always clearly quinqueloculine. The species has been regarded as close to Quinqueloculina seminulum (Linné) by British authors (Heron-Allen & Earland 1930; Haynes 1973a, 19735), but the gradation of forms recognized by Haynes (19735, pl. 1) between the two species is far less evident in southern African late Quaternary assemblages. Specimens referable to Quinqueloculina dunkerquiana occur widely at the present PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 137 day in shallow marine, innermost shelf deposits around the entire South African coastline. The full stratigraphic range of the species remains unclear but on the Orange Shelf it extends throughout the latest Eemian to earliest Weichselian interval, while it is rare in the Holocene (McMillan 1987b). Quinqueloculina isabellei d Orbigny, 1839 Fig. 5C-E Quinqueloculina isabellei d Orbigny, 1839b: 74, pl. 4 (figs 17-19). Quinqueloculina seminulum (non Linné) Boltovskoy et al., 1980: 47, pl. 29 (figs 7-10, ?figs 11-13). ‘Quinqueloculina plataensis’ McMillan, 1987b: 169, pl. 3 (figs 1-4). Remarks The nearly circular aperture and strongly bifid apertural tooth, with the pronounced inflation of the final three chambers leading to a broad, evenly rounded test periphery, are distinctive features. These clearly distinguish the species from Quinqueloculina seminulum (Linné), with which it has been confused. Quinqueloculina araucana, also described by D’Orbigny (18396) from present-day sediments off the west coast of South America, is clearly very similar to Qinqueloculina isabellei, and the two may be conspecific. The test outline of the type specimen of Quinqueloculina isabellei is strongly lobate, though this feature is not evident in the well-illustrated specimens of Boltovskoy et al. (1980), nor in the southern African tests here referred to D’Orbigny’s species. Boltovskoy et al. (1980: 47) considered their specimens to be part of a range of forms that they grouped under the name Quinqueloculina seminulum. If the neotype of Quinqueloculina seminulum from Rimini, Italy, illustrated by Loeblich & Tappan (1964, fig. 349 (no. la—c)), is regarded as typical for Linné’s species, then the South Atlantic shells here referred to Quinqueloculina isabellei are clearly distinct. There is no gradation between Quinqueloculina isabellei and Quinqueloculina seminulum in either the Orange Shelf latest Eemian to earliest Weichselian shells (McMillan 1987b), or in those from Cape Town. See also remarks on Quinqueloculina seminulum (p. 138). Quinqueloculina isabellei has yet to be found in the Holocene of southern Africa, whereas most, if not all, southern South American records are from the Holocene. Quinqueloculina cf. Q. lata Terquem, 1876 Fig. 5F see Quinqueloculina lata Terquem, 1876: 82, pl. 2 (fig. 8a—c). Mathieu er al., 1971: 159, pl. 1 (figs 5-7). Remarks The few, rather damaged specimens from the Cape Town samples lack the embracing, quadrate nature of the final chambers of European littoral specimens 138 ANNALS OF THE SOUTH AFRICAN MUSEUM referred to Quinqueloculina lata. However, they are markedly more elongate than is typical for Quinqueloculina seminulum (Linné). In this respect they appear similar to the tests illustrated by Haynes (1973a, pl. 7 (figs 10, 12), 1973b, pl. 1 (fig. 1)) from Cardigan Bay, Wales, although the Cape Town tests lack such an elongate aperture. Similar forms occur in the latest Pleistocene of the Orange Shelf (McMillan 1987b) to those in the Cape Town samples but, as with the majority of Pleistocene littoral miliolids, almost all are badly damaged and usually only the kernel of each test has preserved. An exact comparison of the Orange Shelf and Cape Town shells is thus not easy to achieve. Quinqueloculina seminulum (Linné, 1758) Fig. 6G-I Serpula seminulum Linné, 1758: 786. Quinqueloculina seminulum (Linné) Loeblich & Tappan, 1964: C458, fig. 349 (no. la—c). Haynes, 1973a: 74, pl. 7 (figs 14, 19), pl. 8 (fig. 3), pl. 32 (figs 1-3), text-fig. 18 (nos 1-4). McMillan, 1974: 35, pl. 2 (fig. Sa—c); 1987b: 172, pl. 3 (figs 5-6). Martin, 1981: 26, pl. 2 (i, 1). Remarks The neotype illustrated by Loeblich & Tappan (1964), from the shore sands of Rimini, Italian Adriatic, has been relied on in establishing the identity of the various smooth-walled Quinqueloculina tests in the Cape Town samples. The neotype displays a rounded-triangular aperture with a rather short tooth that exhibits a distinctly bifid tip (apparently slightly broken on one side). In contrast, southern African tests reveal a more elongate-ovate aperture, within which lies a long, narrow tooth that extends for about three-quarters of the aperture length. The tooth is incipiently bifid at its free end, but this feature is never as well developed as the pronounced Y-shaped tooth of Quinqueloculina isabellei d’Orbigny. In keeping with the neotype, southern African tests of Quinquelocu- lina seminulum display a general test outline and chamber morphology sharply different to those seen in Quinqueloculina isabellei, notably in the sub-rounded, rather than broadly rounded test margin. However, the Cape Town tests are not as strongly triangular in cross-section as the Rimini neotype. Quinqueloculina triangularis d’Orbigny, 1846 Fig. 5G—H Quinqueloculina triangularis d’Orbigny, 1846: 288, pl. 18 (figs 7-9). Mathieu ef al., 1971: 159, pl. 1 (figs 8-10). Remarks In terms of its test morphology, this species compares closely with Quinqueloculina dunkerquiana (Heron-Allen & Earland), differing mainly in the possession of a distinctly bifid tooth, and lacking the indentations of the final two chambers. The Cape Town tests are a little more squat than those of Mathieu PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 139 et al. (1971), and more nearly as high as wide. The bifid tooth is particularly well developed in the figured specimen. D’Orbigny (1946) described Quinqueloculina triangularis from the Miocene of the Vienna Basin; Mathieu ef al. (1971) figured tests from the present-day beach sands of the Dunkerque coast, north France. Quinqueloculina cf. Q. undulata d’Orbigny, 1852 Figs 51, 7A see Quinqueloculina undulata d’Orbigny, 1852: 195. Rosset-Moulinier, 1972: 140, pl. 6 (figs 5-7). Quinqueloculina cf. Q. undulata d’Orbigny: McMillan, 1987b: 175, pl. 3 (figs 7-9). Remarks This distinctive species, with its finely grooved surface and very elongate aperture revealing a slender, tapering tooth, is at present known only from the latest Pleistocene (Unit HI) off the west coast of southern Africa. D’Orbigny’s (1852) species was partly distinguished by its undulating test margins, a feature that was also depicted, though not to so marked a degree, by Cushman (1945). Cushman’s figured specimen is much less intensely ornamented than the original test of D’Orbigny (1852). The shells illustrated by Rosset-Moulinier (1972) are not undulate; she noted that the species was close to Quinqueloculina bicornis (Walker & Jacob), but more elongate in form. None of the southern African tests, from Cape Town or the Orange Shelf, possess an undulate periphery, and they thus appear closest to those described by Rosset-Moulinier (1972) from the Brittany coast. D’Orbigny (1852) described the species from the Pliocene of Castel-Arquato and from present-day deposits near Rimini, both in Italy. The illustrated test from Cape Town possesses a damaged apertural tooth; Orange Shelf tests display a bladed tooth, which extends for about three-quarters the length of the aperture (McMillan 1987b). The depressed area at either end of the aperture is also typical of the Orange Shelf examples. Quinqueloculina spp. Remarks Many badly broken tests of smooth-walled, unornamented Quinqueloculina that probably mostly fall within the Quinqueloculina seminulum—dunkerquiana group. Genus Siphonaperta Vella, 1957 Siphonaperta sp. Fig. 7B—C Quinqueloculina agglutinans (non d’Orbigny) Martin, 1974: 85; 1981: 25, pl. 2 (fig. 8). Sigmoilopsis sp. McMillan, 1974: 37, pl. 2 (fig. 7a—c). Sigmoilopsis schlumbergeri (non Silvestri) Martin, 1981: 27, pl. 2 (fig. 13). Quinqueloculina horrida (non Cushman) Boltovskoy et al., 1980: 46, pl. 27 (figs 17-20). 140 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 7. A. Quinqueloculina cf. Q. undulata d’Orbigny. Oblique edge view, same specimen as Fig. 51. x 65. B-C. Siphonaperta sp. B. Side view, CTC 4. X 96. C. Apertural view, same specimen. x 132. D. Miliolinella subrotunda (Montagu). Side view, CTC 4. xX 110. E-F. Lagena semilineata Wright var. E. Apertural view, CTC 3. x 122. F. Side view, same specimen. X 112. G. Lenticulina sp. Side view, CTC 4. x 129. H. Oolinasp. A. Side view, CTC 4. x 168. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 141 Remarks The above, rather disparate references appear to be to the same species. The presence of an apertural neck and the rough, agglutinated test surface, as well as a quinqueloculine chamber arrangement, are indicative of the genus Siphonap- erta, rather than Quinqueloculina or Sigmoilopsis. Sectioning of tests from the Orange Shelf Holocene (see McMillan 19875, pl. 3 (fig. 11), text-fig. 2) shows the chamber arrangement to be quinqueloculine rather than sigmoiline. These tests can be distinguished from Quinqueloculina horrida, described by Cushman (1947) from off Charleston, South Carolina, in being substantially more triangular in cross-section and somewhat more squat, and lacking the elongate neck and slender bifid apertural tooth of the United States species. Quinqueloc- ulina agglutinans, in contrast, possesses subangular chamber margins, and a large, circular aperture with a phialine lip, but not developed on a neck (see Le Calvez & Le Calvez 1958: 166, pl. 9 (figs 103—104)). Analysis of wall composition of one test dissolved in dilute hydrochloric acid (McMillan 1987b) revealed the presence of small lengths of broken sponge spicules in the wall, a feature that is not evident in external view. The species ranges throughout the latest Eemian to earliest Weichselian and Holocene sequences studied by McMillan (19875), though it is substantially more abundant in the Holocene. The species does not occur in the older Pleistocene rocks of South Africa. Most of the Cape Town specimens are damaged, but the distinctive arrangement of coarse- and fine-grained quartz particles over the test wall easily distinguishes even badly fragmented tests. Subfamily Miliolinellinae Vella, 1957 Genus Miliolinella Wiesner, 1931 Miliolinella subrotunda (Montagu, 1803) Fig. 7D Serpula subrotunda dorso elevato Walker & Boys, 1784: 2, pl. 1 (fig. 4). Vermiculum subrotundum Montagu, 1803: 521. Miliolinella subrotunda (Montagu) Haynes, 1973a: 56, pl. 5 (figs 5-6, 12-13), pl. 31 (figs 8-9), text-fig. 11 (nos 1-4), text-fig. 12 (nos 1-11). Ponder, 1974: 201, pl. 1 (figs 1-2), pl. 2 (figs 6-11). Remarks British researchers have considered the wide variety of forms encountered to be referable to one species: ‘It would seem altogether inappropriate to separate some as Miliolinella, some as Scutuloris and to dispatch the irregular, wild-growing, hauerinid forms into a separate genus and even subfamily’ (Haynes 1973a: 57). This concept is followed here. South African specimens also show much variation, from tight triloculine tests to ones that are loosely and irregularly triloculine, but the wild forms are by and large missing from the cool waters of the west coast. 142 ANNALS OF THE SOUTH AFRICAN MUSEUM Miliolinella subrotunda occurs widely in the South African littoral at the present day, from cold-water west coast to warm-water east coast. The species ranges back to the latest Eemian to earliest Weichselian on the Orange Shelf (McMillan 1987b), but its earlier record in southern Africa is as yet unclear. Family Nodosariidae Ehrenberg, 1838 Genus Lagena Walker & Jacob, 1798 Lagena semilineata Wright, 1886 var. Fig. 7E—F see Lagena semilineata Wright, 1886: 320, pl. 26 (fig. 7). Lagena semilineata (non Wright) Earland, 1934: 161, pl. 7 (figs 19-20). Lagena semilineata Wright var. McMillan, 1987b: 204, pl. 5 (figs 9-11). Remarks The single specimen is not particularly well formed, and the ornamentation is not quite typical of the tests described by McMillan (1987b) from the Orange Shelf. The group of short apical spines of this specimen are poorly developed, lost perhaps through abrasion of the test. The flat-topped, tapering ribs that occupy much of the lower half of the globular part of the test are rather more irregularly formed, particularly in their varying width and length, than is usual. There also are occasional tubercular or rugose developments in the depressed areas between the ribs. However, the style of the ornamentation is so distinctive that there can be no doubt that the Cape Town specimen and those from the Orange Shelf are the same variety. Genus Lenticulina Lamarck, 1804 Lenticulina sp. Fig. 7G Remarks One broken specimen that appears to be in situ, may be referable to one of the rather conservative, unornamented species typified by Lenticulina gibba (d’Orbigny). Little can be made of the single specimen. Family Glandulinidae Reuss, 1860 Subfamily Oolininae Loeblich & Tappan, 1961 Genus Oolina d’Orbigny, 1839 Oolina sp. A McMillan, 1987 Figs 7H, 8A-B Oolina sp. A McMillan, 1987b: 220, pl. 6 (figs 13-14). Remarks This graceful, pyriform Oolina occurs rarely in the sandier, littoral deposits of the latest Eemian to earliest Weichselian of the Orange Shelf (McMillan PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 143 Fig. 8. A-B. Oolina sp. A. A. Apertural view, CTC 4. xX 156. B. Apertural view, same specimen as Figure 7H. X 173. C-—D. Oolina squamososulcata (Heron-Allen & Earland). C. Side view, CTC 4. xX 206. D. Apertural view, same specimen. X 246. E. Fissurina lucida (Williamson). Side view, CTC 3. xX 327. F-G. Fissurina cf. F. marginata (Walker & Boys). F. Apertural view, CTC 3. xX 204 G. Side view, same specimen. xX 161. 144 ANNALS OF THE SOUTH AFRICAN MUSEUM 1987b). The species is distinguished by a vertically elongated hexagonal ribbing pattern ornamenting the test surface. Some variation is evident in the height of the elongated hexagons at the maximum width of the test; those from Cape Town exhibit more nearly regular hexagons than those from further north. The narrower part of the test adjacent to the aperture is marked with small depressions and protrusions that give the surface a rather scaly, reptilian appearance. The subcircular aperture is developed on a short neck, which in well-preserved examples is almost phialine and ornamented with many short, vertically aligned grooves. Oolina squamososulcata (Heron-Allen & Earland, 1922) Fig. 8C—D Lagena squamoso-sulcata Heron-Allen & Earland, 1922: 151, pl. 5 (figs 15, 19). Oolina squamoso-sulcata (Heron-Allen & Earland) Loeblich & Tappan, 1953: 74, pl. 12 (figs 6-7). Remarks The single test compares closely with Oolina squamososulcata except that the horizontally aligned ribs ornamenting the test are more arched, and they fade away only slowly toward the apical end of the test, rather than disappearing abruptly about midway down. The holotype figured by Heron-Allen & Earland (1922) displays 13 vertical ribs; Loeblich & Tappan (1953) reported that their tests exhibit about 12 to 20; the present example possesses 15. Genus Fissurina Reuss, 1850 Fissurina lucida (Williamson, 1848) Fig. 8E Entosolenia marginata (Montagu) var. lucida Williamson, 1848: 17, pl. 2 (fig. 17). Fissurina lucida (Williamson) Murray, 1971: 97, pl. 39 (figs 1-3). Boltovskoy ef al., 1980: 32, pl. 15 (figs 17-20). Remarks One specimen, rather damaged, but closely comparable to Orange Shelf examples of this species. As with Fissurina marginata (Walker & Boys), it would seem to be confined to cool, shallow marine environments, off the west coast of southern Africa only, in both the latest Pleistocene and the Holocene (McMillan 1987b). Fissurina cf. F. marginata (Walker & Boys, 1784) Fig. 8F-G see Serpula (Lagena) marginata Walker & Boys, 1784: 2, table 1, fig. 7. see Fissurina marginata (Walker & Boys) Feyling-Hanssen, 1964: 315, pl. 15 (fig. 22). Haynes, 1973a: 97, fig. 20 (nos 7-8). McMillan, 1987b: 225, pl. 7 (figs 2-3). Remarks This single example displays several slight differences from the tests seen in the latest Pleistocene and Holocene of the Orange Shelf (McMillan 1987b). The PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 145 peripheral ornamentation is more nearly carinate rather than a rounded thickening; the lips on each side of the slit aperture are thicker but less elongate; and there are two raised areas on each side of the test, a little below the aperture. The raised features are low elongate protrusions, aligned roughly parallel to the test periphery. The degree of inflation and the general outline of the Cape Town test is about the same as seen in the Orange Shelf specimens. A rounded peripheral ornamentation would seem to be closer to Fissurina marginata in British waters (Haynes 1973a: 98), so that this Cape Town example rather falls outside the confines of the north-west European interpretation of the species. Family Buliminidae Jones, 1875 Genus Bulimina d’Orbigny, 1826 Bulimina marginata d’Orbigny, 1826 Fig. 9A Bulimina marginata dOrbigny, 1826: 269, pl. 12 (figs 10-12). Murray, 1971: 119, pl. 49 (figs 1-7). Knudsen, 1973: 181, pl. 2 (figs 10-11). Martin, 1981: 41, pl. 4 (figs 8-9). Remarks One specimen, showing rather poorly incised steps into the bases of the later chambers, but with a distinct fringe of blunt spines near the lower margins of some chambers. The test is abraded and has been subject to boring, but is clearly comparable to the more distinctive tests detailed by Martin (1981) and McMillan (19876) from further north. Bulimina marginata ranges throughout the latest Eemian-earliest Weichselian and the Holocene sequences of the Orange Shelf (McMillan 19875). Family Uvigerinidae Haeckel, 1894 Genus Trifarina Cushman, 1923 Trifarina angulosa (Williamson, 1858) Fig. 1OA-B Uvigerina angulosa Williamson, 1858: 67, pl. 5 (fig. 140). Trifarina angulosa (Williamson) Knudsen, 1971: 241, pl. 18 (figs 8-9). Haynes, 1973a: 126, pl. 10 (figs 12-13, 16-17), pl. 11 (fig. 11). Remarks One example, with the terminal part rather broken. However, it clearly shows the triserial chamber arrangement, the tricarinate test periphery, the nearly triangular cross-section to the test, and irregular, vertically aligned ribs on the test surface. The species occurs throughout the latest Eemian to earliest Weichselian and the Holocene of the Orange Shelf (McMillan 1987b), where it shows a considerable range in its surface ornamentation of vertically aligned ribs. Tests range from being almost smooth, with only the three peripheral keels, to densely 146 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 9. A. Bulimina marginata d’Orbigny. Side view, CTC 4. x 123. B. Uvigerina sp. (reworked). Side view, CTC 3. x 133. C. Brizalina spathulata (Williamson). Side view, CTC 3. xX 153. D-E. Cassidulina laevigata d’Orbigny. D. Side view, CTC 4. xX 136. E. Apertural view, same specimen. X 129. F-H. Cassidulina crassa d’Orbigny s.l. F. Side view, CTC 3. X 187. G. Apertural view, same specimen. X 177. H. Close up of foramen, showing serrate margins, same specimen. X 646. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 147 Fig. 10. A-B. Trifarina angulosa (Williamson), same specimen, CTC 4. Both x 67. A. Side view. B. Apertural view. C—D. Bolivina sp., same specimen, CTC 3. Both x 135. C. Side view. D. Apertural view. ribbed over almost the entire exterior. Variation also occurs in the intensity of the ribs, some shells displaying few strongly developed ones, whereas others reveal many closely spaced fine ribs. Family Bolivinitidae Cushman, 1927 Genus Bolivina d’Orbigny, 1839 Bolivina sp. Fig. 10C—D Remarks One small test, ornamented with fine, dense pitting over the surface. The specimen shows some resemblance to Bolivina variabilis (Williamson), but is too juvenile a shell to make a reliable identification. Genus Brizalina Costa, 1856 Brizalina spathulata (Williamson, 1858) Fig. 9C Textularia variabilis Williamson var. spathulata Williamson, 1858: 76, pl. 6 (figs 164-165). Brizalina spathulata (Williamson) Hedley et al., 1965: 21, pl. 6 (fig. 23a—b), text-fig. 6A—-G. Haynes, 1973a: 135, fig. 25 (nos 10-13). Remarks One example, rather damaged. A narrow test when compared to examples from off the Orange River, but the arrangement and coarseness of the test perforations, the sub-rounded acute margin to the test, and the angle at which the sutures decline to the margin are closely comparable. The central zig-zag suture of the Cape Town test is particularly limbate and prominent. Brizalina spathulata ranges throughout the latest Eemian and earliest Weichselian, and the Holocene of the Orange Shelf (McMillan 1987b). However, 148 ANNALS OF THE SOUTH AFRICAN MUSEUM in the latest Pleistocene, the species is widespread through the sandier, more nearly littoral parts of the sequence; these tests lack any substantial carination of the periphery (see McMillan 19875, pl. 11 (figs 10—-11)). The shell from the Cape Town samples is of this type. In contrast, in the Holocene, the species appears commonest in more offshore, shelf environments; these tests usually possess a rather bladed carina (see Martin 1981, pl. 3 (figs 12-13); McMillan 19875, pl. 11 (fig. 12)). There is thus the possibility that two varieties are represented off the west coast in the Late Quaternary. Family Cassidulinidae d’Orbigny, 1839 Genus Cassidulina d’Orbigny, 1826 Cassidulina laevigata d’Orbigny, 1826 Fig. 9D-E Cassidulina laevigata d’Orbigny, 1826: 282, pl. 15 (figs 4-5). Knudsen, 1971: 246, pl. 7 (figs 20-21), pl. 18 (fig. 12). Boltovskoy et al., 1980: 22, pl. 7 (figs 4-6). Remarks Few specimens, all distinguished from the similar species Cassidulina carinata Silvestri by a rounded, non-carinate margin, coarser test perforations and thicker walled test. The precise relationship between the two species is not clear. On the Orange Shelf, Cassidulina laevigata occurs in abundance in the littoral and sublittoral deposits of the latest Eemian—earliest Weichselian, but is present only rarely in the nearshore Holocene. Cassidulina carinata, in contrast, is rare in the latest Pleistocene, but common in the offshore (middle to outer shelf) Holocene. It remains possible that the two species are two facies-controlled formae of one species. Cassidulina crassa d’Orbigny, 1839 s.1. Fig. 9F—H see Cassidulina crassa d’Orbigny, 1839b: 56, pl. 7 (figs 18-20). Cassidulina crassa d’Orbigny ‘intermediate form’ Heron-Allen & Earland, 1932: 358. Cassidulina crassa d’Orbigny forma media Lena, 1966: 316, pl. 2 (fig. 7). Remarks One test, which like those present in the latest Pleistocene of the Orange Shelf, is not Cassidulina crassa sensu stricto but a smaller form, first recognized by Heron-Allen & Earland (1932) from the seas around the Falkland Islands. This form, termed ‘intermediate’, was believed by Heron-Allen & Earland to be a juvenile or pauperate stage of Cassidulina crassa, as first described by D’Orbigny (1839b). Tests possess a rather short aperture, extending from the interio-marginal suture, up toward, but not reaching, the test periphery. There is no extension of the aperture along the interio-marginal suture, as is seen in Cassidulina crassa PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 149 ‘small type’ of Heron-Allen & Earland (1932, pl.9 (figs 29-31)). Some subsequent listings of the ‘intermediate form’ have been made by Boltovskoy (1959, 1961) and Lena (1966) from the coast of Argentina. Cassidulina crassa s.s. would seem to be absent in southern African waters, both in the Pleistocene and the Holocene. The ‘intermediate form’ appears to be confined to deposits of the latest Eemian—earliest Weichselian, and to the cold waters off the west coast. Family Discorbidae Ehrenberg, 1838 Genus Planulinoides Parr, 1941 Planulinoides biconcavus (Jones & Parker, 1862) Fig. 11A—C Discorbina bi-concava Jones & Parker, in Carpenter, 1862: 201, fig. 32G. Discorbina biconcava Parker & Jones (sic) Parker & Jones, 1865: 385, 422, pl. 19 (fig. 10a—c). Brady, 1884: 653, pl. 91 (fig. 2a—c). Planulinoides biconcavus (Jones & Parker) Parr, 1941: 305, fig. a-c. Loeblich & Tappan, 1964: C584, fig. 458 (nos 4a—c, 5-6). Lowry, 1987: 269, pl. 17 (fig. 1a, c, ?1b). Planulinoides biconcava (Jones & Parker) (sic) McMillan, 1974: 58, pl. 5 (fig. 5a—c). Discorbinella biconcava (Jones & Parker) Carter, 1978: 19, pl. 1 (figs 2-3). Remarks This very distinctive species, widespread in shallow marine environments around south-east Australia at the present day (Carpenter 1862; Parker & Jones 1865; Brady 1884; Chapman 1909; Parr 1941, 1945; Collins 1974), also occurs sporadically around southernmost Africa. Specimens from the middle shelf of the Agulhas Bank (McMillan 1974) are always abraded and scoured, and clean tests would seem to be lacking. Whether this indicates that Planulinoides biconcavus is restricted to latest Pleistocene (latest Eemian—earliest Weichselian), Unit III deposits on the Agulhas Bank, rather than the Holocene as well, and has been reworked into the Holocene, is not fully clear at present. Fig. 11. Planulinoides biconcavus (Jones & Parker). All same specimen, CTC 3. All x 68. A. Dorsal view. B. Apertural view. C. Ventral view. 150 ANNALS OF THE SOUTH AFRICAN MUSEUM Planulinoides biconcavus occurs in small numbers on the Agulhas Bank (McMillan 1974; Lowry 1987), and Lowry recorded it between 107 m and 890 m water depth. The species appears to be absent off the west coast in the Holocene and Pleistocene, except for the single specimen from the present study. However, in the Early Miocene of the Orange Shelf, a similar, perhaps identical species occurs (McMillan 1975). A full comparison of the Miocene and Late Quaternary specimens has not yet been attempted. In Australia, the species is known from the Early Miocene (Heron-Allen & Earland 1924), the Middle Miocene (Mitchellian) (Carter 1978), and the Early and Middle Pliocene (Cheltenhamian and Kalimnan) (Parr 1939, 1941) of Victoria. The Cape Town test is typical in its shell morphology, but as is often the case with the Agulhas Bank specimens examined by McMillan (1974), the thinner- walled, less-ornamented ventral side, with the small rounded flaps over the spiral suture, is very badly broken. The elongate-ovate, oblique primary aperture is distinct, though rather damaged. Despite showing some morphological similari- ties to Planulina ariminensis d’Orbigny, even very badly broken tests of Planulinoides biconcavus are easily distinguished, since they entirely lack the coarse test perforations of D’Orbigny’s species. Some differences in size of the flaps on the ventral side, over the spiral suture, can be seen. Tests such as that illustrated by Carter (1978) display large triangular flaps, whereas southern African Quaternary examples reveal only small, rounded flaps. Genus Rosalina d’Orbigny, 1826 Rosalina bradyi (Cushman, 1915) Fig. 12A-—C Discorbina globularis (non d’Orbigny) Brady, 1884: 643, pl. 86 (fig. 8a—c). Discorbis globularis (d’Orbigny) var. bradyi Cushman, 1915: 12, pl. 8 (fig. 1a—c). Rosalina bradyi (Cushman) Hedley et al., 1967: 42, pl. 1 (fig. 3), pl. 11 (fig. 2a—c), text-figs 50-55. Remarks This essentially Indo-Pacific species appears to reside preferentially in patches of Corallina alga in the littoral (Hedley etal. 1967). It occurs in association with Corallina debris in vibracore 620/40 on the nearshore Orange Shelf, in sandy sediments of earliest Weichselian age (McMillan 1987b). The illustrated test from Cape Town is closely comparable to those from the Orange Shelf, although, being a rather juvenile shell, the limbation of the dorsal sutures is not well developed. The ‘milled edge’ of Rosalina bradyi (Cushman, 1915) is caused by the peripheral coarse perforations when viewed from the ventral side, which are visible at right angles through the glassy shell. This feature, obviously, is obscured in the scanning electron microscope photographs. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 151 Fig. 12. A-C. Rosalina bradyi (Cushman). A. Ventral view, CTC 4. xX 139. B. Side view, same specimen. X 144. C. Dorsal view, same specimen. X 135. D-F. Rosalina sp. D. Dorsal view, CTC 3. x 125. E. Side view, same specimen. xX 115. F. Ventral view, same specimen. x 120. G-I. Glabratella australensis (Heron-Allen & Earland). G. Dorsal view, CTC 4. x 81. H. Side view, CTC 4. xX 130. I. Ventral view, CTC 4. x 107. 152 ANNALS OF THE SOUTH AFRICAN MUSEUM Rosalina sp. Fig. 12D-—F Remarks One specimen, closely comparable in its test morphology to Rosalina williamsoni (Chapman & Parr), but differing in a pronounced umbilical ornament. A similar form, less ornate, occurs on the Orange Shelf (McMillan 1987b—described as Rosalina williamsoni). The Orange Shelf shells display a variable, tuberculate ornamentation in and around the umbilicus, but the Cape Town test exhibits a stronger ornament of radiating, elongate-ovate flat-topped pillars and bosses. North-west European examples of Rosalina williamsoni (see Rosset- Moulinier 1972, pl. 9 (fig. 32); Haynes 1973a, pl. 17 (fig. 13), text-fig. 31 (no. 3)) feature a relatively unornamented umbilicus, with just occasional bosses or tubercles on the umbilical terminations of the chambers in the final whorl, and within the umbilicus itself. In contrast, the tests figured by Boltovskoy et al. (1980, pl. 12 (figs 5—12)) from the coast of Argentina as Discorbis williamsoni appear in fact more closely allied to Gavelinopsis praegeri (Heron-Allen & Earland), since the umbilicus is infilled with a distinct plug. It seems probable that the Discorbis cf. D. valvulatus (d’Orbigny) of Boltovskoy et al. (1980, pl. 12 (figs 1—4)) is conspecific with Rosalina williamsoni sensu McMillan (1987b). Some similarity, in terms of style of ornamentation, exists between the Orange Shelf tests and Rosalina malovensis (Heron-Allen & Earland) from the seas around the Falkland Islands. However, none of the southern African shells feature such elongate, crescentic chambers, evident on both dorsal and ventral sides of the test, as exist in Rosalina malovensis. The radiating ornament around the umbilicus of both the Cape Town and Orange Shelf forms perhaps shows some similarity to Rosalina parri Collins (1974, pl. 3 (fig. 36a—c)). However, this Australian species is only slightly bedecked with pillars around the umbilicus; more evident are the radiating grooves and ridges that cover nearly all the ventral side of the illustrated test. Radiating ornament around the umbilicus has been one of the characteristics of the family Gla- bratellidae of Loeblich & Tappan (1964: C587), but the last chamber of many of the southern African tests of this group clearly extends a small triangular flap into the umbilical area, a feature that is rather more typical of Rosalina than Glabratella. Family Bagginidae Cushman, 1927 Genus Glabratella Dorreen, 1948 Glabratella australensis (Heron-Allen & Earland, 1932) Fig. 12G-I Discorbina pileolus (non d’Orbigny) Brady, 1884: 469, pl. 89 (figs 2—4). Discorbis australensis Heron-Allen & Earland, 1932: 416. Parr, 1939: 68. Glabratella australensis (Heron-Allen & Earland) McMillan, 1987b: 363, pl. 14 (figs 14-18), pl. 15 (figs 1-2). PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 153 Remarks A common littoral species along most of the South African coastline at the present day (Brady 1884), extending at least from Cape Town to Mzamba, Transkei (unpubl. studies). It has also been widely recorded along the south, south-east and south-west coasts of Australia (Brady 1884; Chapman 1909; Parr 1945; McKenzie 1962; Albani 1968a, 1968b; Johnson & Albani 1973; Collins 1974; Albani & Johnson 1975) and off New Zealand (Mestayer 1916; Cushman 1919). Its fossil range in Australia appears to be Middle Pliocene (Kalimnan) (Parr 1939), Pleistocene (Collins 1953), and present day. In contrast, in South Africa, it is confined to the Pleistocene, being extremely rare or absent in the earliest Pleistocene (Unit II) of the ‘Alexandria Formation’, and only coming into prominence in the later Early Pleistocene (Unit IIA) of the ‘Bredasdorp Formation’. Glabratella australensis occurs in considerable numbers at some localities of the Varswater Formation (probably Unit IIA), particularly at Hoedjiespunt, Saldanha. The species is distinguished from the South American Glabratella pileolus (d’Orbigny) in its larger number of chambers per whorl (‘eight or more’ rather than ‘four or five—Heron-Allen & Earland 1932: 416), its shallower umbilicus, and its somewhat compressed, low-spired test. Living specimens of Glabratella australensis from intertidal pools at Camps Bay and elsewhere on the north-west Cape Peninsula are a strong greenish brown in colour; perfect, unabraded tests are spectacular when viewed under the scanning electron microscope (see McMillan 1987b, pl. 14 (figs 14-18)). Tests from the Cape Town city centre samples are rather abraded, and much of the detail of the ventral ornamentation of radiating ridges and tubercles has been smoothed. Family Anomalinidae Cushman, 1927 Subfamily Cibicidinae Cushman, 1927 Genus Cibicides De Montfort, 1808 Cibicides fletcheri Galloway & Wissler, 1927 s.1. Fig. 13A—C Cibicides fletcheri Galloway & Wissler, 1927: 64, pl. 10 (figs 8a—c, 9a—c). Lankford & Phleger, 1973: 117, pl. 6 (figs 1la—c). Cibicides cf. C. fletcheri Galloway & Wissler. Boltovskoy et al., 1980: 24, pl. 8 (figs 17-21). Remarks Three rather damaged tests, compressed, and characterized by a small, rounded, domed umbilical infilling on the ventral side, and a flat and wide boss, flush with the test surface, in the dorsal umbilicus. These tests are similar to those detailed by McMillan (19875) from the latest Eemian-—earliest Weichselian and Holocene of the Orange Shelf. The correct taxonomic name for this form is difficult to determine; so many similar, though distinctly different morphotypes have been assigned to Galloway 154 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 13. A-C. Cibicides fletcheri Galloway & Wissler s.l. A. Ventral view, CTC 3. x 174. B. Side view, same specimen. X 168. C. Dorsal view, same specimen. X 174. D-F. Cibicides lobatulus (Walker & Jacob). D. Dorsal view, CTC 4. X 88. E. Ventral view, CTC 4. x 82. F. Side view, CTC 4. x 146. G-H. Cibicidoides sp. G. Side view, CTC 3. X 84. H. Ventral OY, CIE BZ, X< WY, PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 155 & Wissler’s (1927) species that it is unclear what the limits of the species should be. A comparable species is Anomalina schmitti, described by Cushman & Wickenden (1929) from present-day sediments off Juan Fernandez Islands, off the west coast of South America. It is not clear how distinct are these two species. Cibicides fletcheri was first described by Galloway & Wissler (1927) from the Pleistocene of the Palos Verdes Hills of California. Cibicides lobatulus (Walker & Jacob, 1798) Fig. 13D-F Nautilus lobatulus Walker & Jacob, in Kanmacher, 1798: 642, pl. 14 (fig. 36). Cibicides lobatulus (Walker & Jacob) Murray, 1971: 175, pl. 73 (figs 1-7). Haynes, 1973a: 173, pl. 20 (figs 1-2), pl. 21 (figs 3, 5-6), pl. 33 (figs 1-7), text-fig. 35 (nos 4-10). Boltovskoy et al., 1980: 24, pl. 9 (figs 1, 3-4, ?2). Martin, 1981: 52, pl. 5 (figs 4-6). Remarks All tests possess narrow, elongate chamber flaps over the spiral suture, from under which the secondary apertures are developed on the dorsal sides of the final whorl of chambers. The shape of the flaps along the spiral suture is in keeping with those seen on tests from the latest Pleistocene and Holocene of the Orange Shelf (McMillan 1987b), off the west coast of southern Africa (Martin 1981), on the Agulhas Bank (McMillan 1974—though not well drawn on pl. 8 (fig. 1c)), and off the coast of Argentina (Boltovskoy et al. 1980). However, north-west European tests (the type locality is Whitstable in Kent) tend to possess rather more lobate chamber flaps (Feyling-Hanssen 1964; Knudsen 1971; Murray 1971; Rosset-Moulinier 1972; Haynes 1973a). The difference is slight but it may prove to be consistent between the two regions. As with the latest Pleistocene and Holocene tests of Cibicides lobatulus from the Orange Shelf (McMillan 19876), there are substantial differences in the density of test perforations and in the overall test morphology of the Cape Town shells. Spreading, more rounded-conical forms, often with a distinct peripheral thickening, are usually densely perforate over the entire ventral side. The higher, button-shaped, more biconvex tests, with a more rounded periphery and little thickening, are generally only sporadically perforate on the ventral side, more so over the surface of the earlier chambers. Genus Cibicidoides Thalmann, emend. Loeblich & Tappan, 1955 Cibicidoides sp. Figs 13G—H, 14A-B Remarks All tests of this species are rather broken. All are coarsely perforate on both sides of the test, and exhibit ventral sutures that are more radiate and 156 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 14. A-B. Cibicidoides sp. A. Dorsal view, CTC 2. xX 112. B. Side view, same specimen as Figure 13H. x 120. C-—D. Hyalinea balthica (Schréter). Side view, CTC 3. X 205. D. Apertural view, same specimen. x 205. E-F. Sphaerogypsina globulus (Reuss). Different views of the same specimen, CTC 3. E. Xx 137. F. x 144. G-H. Elphidium advenum (Cushman). G. Side view, CTC 3. x 144. H. Apertural view, same specimen. X 146. I. Elphidium sp. A. Side view, CTC 4. x 120. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE SS7/ straight or weakly curved than is typical of the type species, Cibicidoides mundulus (Brady, Parker & Jones), which is coarsely perforate only on the dorsal side. This type of Cibicidoides is not known from the Orange Shelf (McMillan 1987b), and would seem closer to the Cibicides cf. C. praecinctus (Karrer) of McMillan (1974) from the Agulhas Bank and Martin (1981) from scattered western offshore localities. These tests are distinctly more biconvex than is typical for Cibicidoides pseudo-ungerianus (Cushman), widely found off both the south and west coasts of southern Africa (McMillan 1974; Martin 1974, 1981). Subfamily Planulininae Bermudez, 1952 Genus Hyalinea Hofker, 1951 Hyalinea balthica (Schréter, 1783) Fig. 14C—D Nautilus balthicus Schroter, 1783: 20, pl. 1 (fig. 2). Hyalinea balthica (Schréter) Murray, 1971: 173, pl. 72 (figs 5-8). Hyalinea balthica (Gmelin) Bremer et al., 1980: 23, pl. 3 (fig. 8). Remarks One specimen, rather broken and with tubular borings. The specimen is typical for the species, which occurs widely over the western continental margin of southern Africa at the present day (Pinto 1955; Martin 1974, 1981), as far south as the westernmost Agulhas Bank (McMillan 1974). Hyalinea balthica ranges throughout the latest Eemian—earliest Weichselian and Holocene deposits of the Orange Shelf (McMillan 19876). It is known from the latest Eemian—earliest Weichselian sediments of the Agulhas Bank (borehole F—A 13) (McMillan 1986), and at that time ranged as far east as Algoa Bay, where it is seen in the coastal Unit III of the ‘Alexandria Formation’ (Rimke in Engelbrecht etal. 1962; Salmon 1981; McMillan in press). In the latest Eemian—earliest Weichselian Unit III, Hyalinea balthica occurs in littoral sands but, in the Holocene and at the present day, the species is more typical of middle to outer shelf environments and deeper; Martin (1981) gave its depth range as 92 m to 995 m off the west coast. As is the case with the Cape Town test, latest Pleistocene shells would appear to be slightly thicker walled than those from the Holocene, but there seems to be no distinct morphological difference. Hyalinea balthica has not been found in earlier Pleistocene deposits in South Africa, though this may be due to the depositional facies of the rocks. Certainly, though, the species cannot be considered as a marker for the Pliocene- Pleistocene boundary on the continental shelf of southern Africa, as has been done elsewhere (see Bayliss 1969). 158 ANNALS OF THE SOUTH AFRICAN MUSEUM Family Acervulinidae Schultze, 1854 Genus Sphaerogypsina Galloway, 1933 Sphaerogypsina globulus (Reuss, 1848) Fig. 14E-F Ceriopora globulus Reuss, 1848: 33, pl. 5 (fig. 7a—c). Sphaerogypsina globulus (Reuss) Loeblich & Tappan, 1964: C698, fig. 569 (nos 1-2). Remarks Several subspherical tests, although damaged, some with a flattened surface that may have attached to a substrate, are referable to this species. One or two examples were seen in the latest Pleistocene deposits of the Orange Shelf (McMillan 1987b), but the species is more numerous in the Cape Town samples. In most of the studied tests, the thin perforated walls over each chamberlet are usually damaged or entirely worn away, and only the thicker intercameral walls remain visible, as in the illustrated test. Similar forms, though larger and more nearly spherical, occur in the Early Miocene (Burdigalian) shelly, white lime- muds at a number of localities on the South African continental margin. Family Elphidiidae Galloway, 1933 Subfamily Elphidiinae Galloway, 1933 Genus Elphidium De Montfort, 1808 Elphidium advenum (Cushman, 1922) Fig. 14G-H Polystomella subnodosa (non Minster) Brady, 1884: 734, pl. 110 (fig. 1a—b). Polystomella advena Cushman, 1922: 56, pl. 9 (figs 11-12). Elphidium advenum (Cushman) Albani, 1968a: 111, pl. 10 (fig. 6). see Elphidium advena (Cushman) Apthorpe, 1980: 211, pl. 26 (figs 10-11). Remarks Most later authors studying Indo-Pacific material appear to have followed the illustrations of Brady (1884) rather than those given by Cushman (1922) for their identification of Elphidium advenum. The septal pits of Indo-Pacific tests of the species are short and never as elongate as those of the holotype, which is from the Tortugas region of the West Indies. It may be, as suggested by Cushman (1922), that the Indo-Pacific shells assigned to this species are of a slightly different form than the tropical Atlantic Elphidium advenum s:.s. From studies of the ‘Alexandria Formation’ of the eastern Cape (McMillan in press), it is clear that Elphidium advenum, as here understood, ranges from the present day to near the base of the Pleistocene (and includes Unit II). It can thus be considered as a useful marker species for the South African Quaternary, in company with Pararotalia nipponica (Asano) and Cibicides lobatulus (Walker & Jacob). The species is extensively distributed off the west, south and east coasts of southern Africa at the present day, and tends to predominate in shallow marine PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 159 environments (Albani 1965; Martin 1981; unpubl. studies), although-off the west coast it ranges into waters as deep as 418 m (Martin 1981), and small numbers of specimens occur as deep as 127 m on the Agulhas Bank (McMillan 1974). It was also widely distributed in latest Eemian-—earliest Weichselian times in shallow marine environments around southern Africa, from the Orange Shelf (McMillan 1987b) to the Uloa Formation. Elphidium sp. A Fig. 141, 1I5A Elphidium advenum (non Cushman) McMillan, 19875, pl. 17 (figs 1-2). Remarks A small proportion of the tests, initially assigned to Elphidium advenum by McMillan (19875), lack the distinctive peripheral carina and wide, flat-topped and flush umbilical boss of that species. Instead, the umbilical boss is small, almost like a pinhead, and sited in a narrow umbilicus that is not entirely occupied by the boss. The later chambers are slightly inflated, the test size is somewhat greater, and the septal pits are more sharply defined, rather deeper, and more elongate than true Elphidium advenum. Elphidium sp. A is apparently confined to the latest Pleistocene of the west coast, and is absent in the Holocene. The differences from true Elphidium advenum are sufficient for this form to be considered a separate, possibly new species but insufficient specimens occur in the Cape Town samples for a formal taxon to be erected here. Elphidium articulatum (d’Orbigny, 1839) Fig. 1SB—C Polystomella articulata d’Orbigny, 1839b: 30, pl. 3 (figs 9-10). Elphidium articulatum (d’Orbigny) Boltovskoy, 1963: 61, pl. 6 (fig. 15). Boltovskoy et al., 1980: 29, pl. 13 (figs 1-4). Remarks Distinct, though slight differences exist in the style of the umbilical ornamentation of Cape Town tests compared with those illustrated by McMillan (1987b) from the Orange Shelf. The Orange Shelf examples reveal a rather depressed umbilicus ornamented with 20 or more thin calcite pillars. In the illustrated Cape Town shell, there are no more than ten pillars, and these are often merged together, or with the umbilical terminations of the chambers; they also tend to be larger in size than the pillars of the Orange Shelf tests. The septal pits, both in their numbers and outline, the style of the aperture, the number of chambers in the final whorl, the chamber morphology and the overall test morphology of the Orange Shelf and Cape Town tests are all closely comparable. Those from Cape Town show an umbilical arrangement somewhat closer to the 160 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 15. A. Elphidium sp. A. Oblique apertural view, same specimen as Figure 14]. x 113. B-C. Elphidium articulatum (d’Orbigny). B. Apertural view, CTC 4. x 84. C. Side view, same specimen. X 86. D-E. Elphidium crispum (Linné) s.l. D. Apertural view, CTC 4. ZS Wake 1 Be Side view, CTC 4. x 83. F-G. Elphidium macellum (Fichtel & Moll) s.1. F. Oblique apertural view, CTC 4. x 95. G. Side view, same specimen. x 95. H-I. Elphidium sp. B. H. Side view, CTC 3. xX 178. I. Apertural view, same specimen. X 178. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 161 individuals illustrated by Boltovskoy et al. (1980) from the coast of Argentina. However, the Argentinian tests are more strongly biconvex than those from southern Africa, if the apertural view (Boltovskoy et al. 1980, pl. 13 (fig. 3)) can be regarded as typical. D’Orbigny (1839b) figured a specimen that is substantially more lobate around the test periphery than is the case in more recently illustrated South Atlantic tests; his specimens were from the Rio Negro coast of Patagonia and from off the Falkland Islands. Elphidium articulatum appears to range from the latest Eemian—earliest Weichselian to the present day in southern Africa. It is typical of innermost shelf, cold-water locations on the west coast, and occurs in the present-day sands washed into the canalized mouth of the Soutrivier, just north-east of Cape Town city centre. Elphidium crispum (Linné, 1758) s.1. Fig. 1SD-E Nautilus crispus Linné, 1758: 709 (figured by Plancus, 1739, pl. 1 (fig. 2d—-f)). Elphidium crispum (Linné) Cushman & Leavitt, 1929: 20, pl. 4 (figs 3a—b, 4a—b). Hageman, 1979: 94, pl. 5 (fig. 6a—b). Remarks Preliminary details of the taxonomic complexity surrounding this species and Elphidium macellum (Fichtel & Moll), and the relationship of these two Mediterranean species to comparable southern African forms, were given by McMillan (19876). The following are supplementary notes. Examination of authors’ illustrations of Elphidium crispum from the Mediterranean Sea shows a wide range of forms. All the original figures are somewhat stylized drawings (Plancus 1739; Gualtieri 1742), and it is upon these that Linné based his concept of this species. The tests detailed by Gualtieri appear to display an immense number of chambers in the final whorl, whereas those figured by Plancus seem to contain about 20 chambers. Cushman & Leavitt (1929), in an attempt to distinguish clearly between Elphidium macellum and Elphidium crispum, re-examined the illustrations of both species as figured by Fichtel & Moll (1798), and compared these with specimens ‘from as near the type locality as possible’. In the case of Elphidium crispum, it is worth including here Cushman & Leavitt’s full description: ‘Test of large size for the genus, about 21/2 times longer than wide in peripheral view, lenticular, completely involute, the umbilical regions with a medium sized, slightly projecting, rounded boss of clear shell material, the surface of which is set with 10—12 small, shallow, rounded pits, periphery sharply angular but not sharply keeled, sometimes becoming slightly lobulate and blunter in the last portion of the adult coil; chambers very numerous, 20—40 in number, long, narrow, the last few often inflated, forward part slightly raised, giving a radially ribbed appearance; sutures somewhat sigmoid, the middle portion of each nearly 162 ANNALS OF THE SOUTH AFRICAN MUSEUM radial, partially obscured by the retral processes, set with pores, those of the earlier ones indistinct but later ones with large and deep pores; wall finely perforate with about 12 relatively long retral processes extending well up onto the side of the next added chamber, appearing as fairly widely spaced, low elevations of the wall in the earlier chambers, but as thicker and more rod-like structures in the later chambers; aperture, a row of small openings between the retral processes at the base of the apertural face, which is low, sagittate, the sides often slightly concave, the ends pointed. Diameter up to 3 mm.’ Such large-sized tests of E/phidium occur nowhere in South Africa, living or fossil. South African tests assigned to this species possess only from 17 to 22 chambers in the final whorl of adult specimens, whereas juvenile examples reveal rarely more than 12. The South African individuals tend to be more biconvex rather than lenticular, normally with a small, sharply defined, well-raised umbilical boss that exhibits a maximum of seven pits on its surface, rather than the 10-12 described by Cushman & Leavitt (1929). Closer similarity would seem to exist between the style and numbers of the septal pits and bridges, though perhaps the tests figured by Cushman & Leavitt (1929) are a little stylized in this respect. South African shells feature a rounded to weakly sub-rounded, never acute, test margin that reveals a hint of thickening along the earlier part of the final whorl, whereas the periphery of the Italian form is sharply angular. However, some latest Pleistocene tests of Elphidium crispum clearly show abrasion along the periphery, which in severe cases may wear through to the chamber cavities. It thus seems likely that the test periphery has been substantially amended on occasion by post-mortem transport, probably through rotating and cartwheeling of these shells in the littoral sands before final deposition. Nevertheless, despite the probability of some post-mortem changes to these Elphidium crispum shells, there are clearly distinct differences between southern African and Mediterra- nean tests. Although the differences are not great, and the southern African examples clearly fall within Elphidium crispum s.\., nevertheless they can be confidently regarded as a local, distinctive form, which is easily distinguished from Elphidium crispum s.s. Unfortunately, this fairly simple picture is complicated by authors’ changing attitudes to species through the twentieth century. Cushman & Leavitt (1929) evidently based their descriptions on adult specimens only, for they noted that a considerable difference exists between the test morphologies of microspheric and megalospheric, and juvenile and adult tests of both Elphidium crispum and Elphidium macellum; these changes mostly remain to be documented. Thus, for Cushman & Leavitt (1929), the two species are easily distinguishable, but later authors have tended to see the two as end-members with substantial variation between them. In addition, later authors’ illustrations are of smaller-sized (?juvenile) tests, e.g. Kruit (1955, pl. 2 (fig. 10)) from the Rhéne delta, Hageman (1979, pl. 5 (fig. 6a—b)) from PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 163 the Plio—Pleistocene of the western Peloponnesus of Greece, and Jonkers (1984, pl. 12 (fig. 4)) from the Pliocene of Crete. Hageman (1979) reported on a continuous gradation from lenticular tests with a large umbilical boss (typical Elphidium crispum) to compressed tests lacking or possessing a small umbilical boss (closely resembling typical Elphidium macellum), and Jonkers (1984) followed similar methods of interpretation. Despite this rather confusing picture of Mediterranean Elphidium crispum, the South African tests appear to remain distinct from the species in a strict sense. However, to avoid increasing the range of names available for this group of Elphidium, the southern African tests are here referred to Elphidium crispum s.1. Elphidium crispum s.\. occurs in large numbers in the Cape Town samples and elsewhere around the south-western Cape coast in deposits of the same age (see Appendices A to C). The species is not found in deposits older than latest Eemian-—earliest Weichselian in the South African Pleistocene, though similar species, of the Elphidium crispum—macellum group, occur in both the Varswater Formation and in the older ‘Bredasdorp Formation’ (both Unit IA). Elphidium crispum s.l. became extinct during the Weichselian or at the end of the Pleistocene, and it is not found in situ in any Holocene deposits in South Africa. This species occurs in smaller numbers in the latest Eemian—earliest Weichselian of the Orange Shelf (McMillan 1987b) than at Cape Town. It is also present close to the coast in the youngest parts (Unit III) of the ‘Bredasdorp Formation’ and the “Alexandria Formation’, at Plettenberg Bay, and on the Agulhas Bank (McMillan 1974); on the inner shelf it may occur in considerable numbers, reworked into the Holocene. The species is also known from the Aliwal Shoal (Carter 1966) and the Uloa Formation of Zululand (pers. obs.), but there are slight morphological differences evident in these warmer-water tests that have yet to be defined. Elphidium crispum s.1. thus seems to be an ideal zone fossil for all the littoral or sublittoral deposits of the latest Eemian—earliest Weichselian regression. It is restricted to the same time interval over which Ammonia sp. B (Alexandria) and Ammonia sp. C (Alexandria) range (McMillan 1987a), but is better developed in the colder waters of the west coast, where the two Ammonia species are absent (see Fig. 3). Elphidium crispum s.\. tests from studied localities are invariably broken or abraded, and often the entire outer whorl of chambers is missing, leading to strongly raised umbilical bosses, which tend to better survive the abrasion processes. McMillan (19876: 565) considered the possibility of Elphidium crispum s.l. and Elphidium macellum s.\. being abraded and clean tests, respectively, of the same species. On the Orange Shelf, the former is most common in the sandier, littoral parts of studied sections; the latter, always well preserved, is most frequent in the clayier intervals that presumably accumulated in quieter water conditions. There thus appears to be rather different facies controls on the two species. However, examination of the test outlines of the two, particularly their cross-sections, reveal distinct differences. Sectioned tests of Elphidium macellum s.l. are far too slender and lenticuline throughout their course of growth to 164 ANNALS OF THE SOUTH AFRICAN MUSEUM accommodate a kernel in their interior of the more strongly biconvex Elphidium crispum s.1. (or vice versa). Elphidium macellum (Fichtel & Moll, 1798) s.1. Fig. 15F-G Nautilus macellus Fichtel & Moll, 1798: 66, var. B, pl. 10 (fig. h—-k). Elphidium macellum (Fichtel & Moll) Cushman & Leavitt, 1929: 18, pl. 4 (figs la—b, 2a—b). Régl & Hansen, 1984: 50, pl. 14 (figs 2, S—6), pl. 15 (figs 1-2), text-fig. 18B. Remarks As with Elphidium crispum (Linné), local, southern African populations of Elphidium macellum, though closely similar to those from the type area of the Mediterranean Sea, nevertheless show some distinct and consistent differences. Cushman & Leavitt (1929), in their examination of Elphidium macellum from Rimini on the Adriatic coast of Italy, gave the following description: ‘Test of medium size for the genus, about 31/2 times longer than wide in peripheral view, lenticular, planispiral, completely involute, umbilical regions flat, periphery angular, slightly more rounded in the last-formed chambers, slightly keeled, somewhat lobulate; chambers numerous, averaging 17 in adults in the last-formed coil, slightly arched giving a ribbed appearance to the test; sutures slightly depressed, those of the last-formed chambers slightly more so, partially obscured by the retral processes, curved backward strongly toward the periphery, the proximal half nearly radial, set with indistinct pores; wall thin, finely perforate, usually with less than 12 relatively long retral processes, more widely spread than in E. crispum and extending from well up on the side of the chamber to the front of the preceding and appearing as narrow, fairly widely spaced, low elevations of the wall; aperture, a row of small openings between the retral processes at the base of the apertural face which is convex, sagittate, with the sides slightly convex, the lobes sharply angled, saddling the preceding coil. Diameter typically less than 1 mm; thickness 0,25 mm.’ Cushman & Leavitt (1929) concluded by distinguishing Elphidium macellum from Elphidium crispum and Elphidium striatopunctatum (Fichtel & Moll) by its keeled periphery that is often lobate, the fewer chambers in the final whorl, and the flattened or depressed umbilicus. More recently, the Fichtel & Moll collections have been rediscovered in Vienna (Hansen & Roégl 1980), and a full re-appraisal of their species of 1798, including Elphidium macellum, has been achieved (R6gl & Hansen 1984). In the case of Elphidium macellum itself, Régl & Hansen (1984: 51) described and illustrated a lectotype from the Fichtel & Moll collection, which thus must be regarded as the nomenclatural type. This specimen possesses a circular, non-lobate outline, a sharply angled and keeled periphery, 16 and a half chambers in the final whorl, curved sutures with 8 to 15 septal bridges per suture, slightly depressed umbilicus, lacking ornamentation or plug, apertural face low, sagittate, with fine tubercles, and interio-marginally an equatorially placed aperture PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 165 composed of multiple openings with small protruding lips; maximum diameter 0,79 mm; maximum thickness 0,32 mm. On comparing the South African tests with the illustrations and descriptions of Rogl & Hansen (1984), it can be seen that the local tests differ in a number of skeletal features. These shells are more compressed, more slender than either the lectotype of R6gl & Hansen (1984) or the comparative Rimini tests illustrated by Cushman & Leavitt (1929). They contain from 13 to 19 chambers in the final whorl, with an apparent increase in numbers with increasing test size (McMillan 19875). Most significantly, the umbilical area of the southern African tests is infilled with a low, irregular plug of varying size that is ornamented over its surface with one to seven pits, which in well-preserved examples can be seen to be spinose or tuberculate interiorly. The plug is never as large or as prominent as that seen in Elphidium crispum s.\. tests from the latest Pleistocene of south-western Africa, but it clearly shows variation in size and height (McMillan 19875, pl. 18 (figs 1-6)). The southern African shells are also not quite as acutely carinate on the periphery of the test as the lectotype of Rogl & Hansen (1984). Because these differences in test morphology appear to be consistent distinctions between the Mediterranean and southern African tests, the latter are here termed Elphidium macellum s.1., for the same reasons as expressed for Elphidium crispum s.1. It should be added that the southern African shells of Elphidium macellum s.l. compare closely, especially in their umbilical ornament, with some figured from the coast of Patagonia (Thomson 1978, pl. 2 (fig. 1)). The test illustrated by Brady (1884, pl. 110 (fig. 11a—b)) is of the same group, though the arrangement of pits over the surface of the umbilical infilling has not been found in so symmetrical a pattern in any of the southern African examples examined by the present author, and is perhaps a little stylized. Tests of this group have also been illustrated by Martin (1981, pl. 6 (fig. 12, and probably also fig. 10)) and McMillan (1974, pl. 9 (fig. 7a—b)). Elphidium macellum s.\. ranges from the latest Eemian—earliest Weichselian (Unit III) deposits through to the present day around South Africa, and shows a slight preference for cooler water conditions. The species is also one of the group that occurs on the shelf during the Holocene (Martin (1981) gave its single occurrence in her samples as at 118 m water depth, and it was recognized by McMillan (19875) between 137 m and 183 m on the Orange Shelf), but it is more typical of sublittoral, fairly quiet water conditions in the latest Pleistocene. Spinose juveniles, such as are typical of north-west European waters (see Haynes 1973a: 202, pl. 24 (figs 2—-3)) do not occur in South Africa, either fossil or extant. Elphidium sp. B Fig. 1SH-I Remarks One specimen, slightly reminiscent of Elphidium gunteri Cole, as described by McMillan (19876) from the Orange Shelf but differing on several points. There 166 ANNALS OF THE SOUTH AFRICAN MUSEUM are 12 chambers in the final whorl, intersected by deeply incised, curved to straight, radiate sutures that are crossed by up to five septal bridges per side, the bridges appearing only on the later sutures. The umbilical chamber terminations are rounded and slightly nodose, and the umbilicus is fairly wide and shallow, being ornamented with a granular, almost sugary surface. The test wall is finely and densely perforate; the test periphery is broadly and uniformly rounded, the margin being sub-circular and continuous. The foramen, possibly composed of two arched interio-marginal openings at the base of the terminal face, is overlain by a zone of tuberculate ornamentation. Elphidium gunteri tests display a more lobate test periphery, coarser test wall perforations, less incised sutures, and an umbilical ornamentation of small numbers of irregularly shaped calcite pillars. Genus Astrononion Cushman & Edwards, 1937 Astrononion echolsi Kennett, 1967 Fig. 16A—-B Astrononion echolsi Kennett, 1967: 134, pl. 11 (figs 7a—b, 8). Fillon, 1974: 139, pl. 6 (figs 1-3). Astrononion stelligerum (non d@’Orbigny) McMillan, 1987b: 422, pl. 19 (figs 1-3). Remarks A reconsideration of the Astrononion specimens described and illustrated by McMillan (1987b) from the latest Eemian—earliest Weichselian of the Orange Shelf suggests that they should rather be referred to the Antarctic species Astrononion echolsi. Le Calvez (1974) has examined the type specimen of Nonionina stelligera d’Orbigny, found it to be unusable, and has proposed a neotype (Le Calvez 1974: 37, pl. 9 (figs 1-4)) from D’Orbigny’s Canary Island (Teneriffe) specimens. The neotype is strongly compressed, with ten chambers in the last-formed whorl, deeply incised sutural slits outside of the umbilical cover plate(s), and a broad, shallow umbilicus almost entirely covered, except centrally, by the cover plate(s). In contrast, the present specimen and those from the Orange Shelf possess only seven or eight chambers in the final whorl, rather more inflated chambers and a lobate test periphery, wider, more open and shorter incisions of the sutures, and a less extensive cover plate revealing more of the central umbilicus. The Cape Town specimen is a little more strongly biconvex than is typical for the Orange Shelf shells. This species would seem to be confined to the cold-water influenced latest Eemian and earliest Weichselian deposits of the west coast of South Africa. Family Rotaliidae Ehrenberg, 1839 Genus Ammonia Brtinnich, 1772 Ammonia japonica (Hada, 1931) Fig. 16C-E Rotalia japonica Hada, 1931: 137, fig. 93a—c. Ammonia japonica (Hada) Matoba, 1970: 48, pl. 5 (fig. 14a—-c), pl. 6 (fig. 1la—c). McMillan, 1987a: 37, text-fig. 7a—d. Ammonia beccarii (non Linné) Martin, 1974: 84, fig. 14-1 (part.). Salmon, 1979: 77, fig. 3p. Martin, 1981: 48, pl. 3 (figs 2-3). Ammonia beccarii (Linné) var. inflata (non Seguenza) McMillan, 1974: 62, pl. 6 (fig. 2a—c). PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 167 oe ONS Fig. 16. A-—B. Astrononion echolsi Kennett. A. Side view, CTC 4. x 139. B. Apertural view, same specimen. X 157. C-—E. Ammonia japonica (Hada). C. Ventral view, CTC 4. x 111. D. Side view, same specimen. X 124. E. Dorsal view, same specimen. Xx 110. F—H. Ammonia parkinsoniana (d’Orbigny) s.l. F. Side view, CTC 4. Xx 135. G. Dorsal view, CTC 4. x 111. H. Ventral view, same specimen as F. X 142. 168 ANNALS OF THE SOUTH AFRICAN MUSEUM Remarks Subsequent to the comments on Ammonia japonica by McMillan (1987a, 1987b), additional studies have been made of the Pleistocene littoral deposits of the Saldanha—Langebaan region, the ‘Bredasdorp Formation’ and the ‘Alexan- dria Formation’ of the southern Cape coast, and the Uloa Formation and Bluff Sandstone Formation of Zululand/Natal. As was mentioned by McMillan (1987a), Ammonia japonica is best developed in the Holocene off the south and west coasts of South Africa, and is similarly distributed in the latest Eemian—earliest Weichselian (Unit III) deposits. The species is absent in older Pleistocene and Pliocene marine rocks of this region, although a similar form does occur in the Varswater Formation (Unit IIA) of the Langebaan—Saldanha area, that is of the same Ammonia-group as Ammonia Japonica. Holocene tests tend to possess numerous, rather granular calcite pillars in the umbilicus (Martin 1981, pl. 3 (fig. 2); McMillan 19875, pl. 19 (figs 12, 15-16)), but latest Pleistocene shells are usually characterized by more robust umbilical pillars, which may often merge into sinuous, irregular structures (McMillan 1987b, pl. 20 (figs 1-2)). It is probable that the difference is due to abrasion of the Pleistocene tests but not the Holocene ones. However, variation in test morphology through time is also a possible cause. Rouvillois (1974) described Pseudoeponides falsobeccarii from the continen- tal shelf (60-120 m water depth) off the west of France, and it is clear that southern African Ammonia japonica is closely related to this North Atlantic species. Neither show much similarity to the genotype, Pseudoeponides japonicus Uchio (in Kawai et al. 1950). Rouvillois (1974) identified both ventral and dorsal openings in her species, but these are so unlike those of Pseudoeponides japonicus that, together with major differences in test and chamber morphology, it seems best for the moment to regard the French species as referable to Ammonia. The ‘secondary openings’ of the ventral side of Ammonia falsobeccarii occur in various forms in all species of Ammonia, but the dorsal openings, present also in southern African Ammonia japonica, clearly separate these two species from Ammonia beccarii (Linné) s.s., Ammonia parkinsoniana (d’Orbigny) s.1., and other species typical of more littoral environments. It may be that the deeper-water, shelf species of Ammonia warrant a distinct genus. A comparison of southern African Ammonia japonica with Ammonia falsobeccarii evinces the following similarities and differences. Rouvillois’ species possesses usually eight chambers, southern African tests display eight or nine, more rarely seven or ten, chambers in the final whorl. Both species possess very strongly convex ventral sides but the dorsal side of the southern African shells is always markedly convex, whereas that of the French species is weakly convex, almost flat. Variation in the arrangement and the style of the umbilical filling is similar, with well-preserved tests exhibiting a granular or tubercular ornament over the dense array of thin calcite pillars that occupy the umbilicus. The arrangement of incisions, bordered with granules and small nodes along the spiral and septal sutures, is much the same in both groups of specimens. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 169 The detail of the secondary openings at the inner junctions of the septal and spiral sutures cannot be seen in the illustrations of Ammonia japonica given by Hada (1931) and Matoba (1967, 1970). Their figures appear to be derived from light microscope photography and drawings, and the depressions on the dorsal side may thus either be obscured by reflection effects within the shell, or they are absent in the Japanese tests. For the present, Hada’s name has been retained for these southern African tests. McMillan (1974) employed the name Ammonia beccarii (Linné) var. inflata (Seguenza) for this species on the Agulhas Bank. The dorsal sides of Agulhas Bank tests are generally less convex than those of tests from off the west coast (McMillan 1987a). Rosalina inflata was originally described by Seguenza (1862) from the ‘Pleistocene abbastanza recente’ clays of Catania, Sicily. Again, the ventral side of the test is strongly convex, the dorsal side less so. Some variation can be seen in the tests illustrated by Seguenza (1862, pl. 1 (figs 6, 6a—c)) but, although the umbilical ornamentation and the test outline are similar to southern African Ammonia japonica, the beaded bordered incisions of the sutures (in fig. 6b) may extend to the test periphery, and the ventral surfaces of the chambers may be strongly ornamented with irregular rugosities (figs 6a, 6c). Billman et al. (1980: 86, pl. 5 (figs 1-14)) have illustrated tests referred to Ammonia inflata (Seguenza) from northern Morocco, but noted that the species should be revised using topotypic material. The Moroccan tests possess incised sutures with beading over the entire ventral side; over the dorsal side the same ornamentation occurs, but it is less well developed along both septal and spiral sutures of the early part of the test. The umbilical infilling again consists of numerous calcite pillars. It is felt by the present author that Ammonia inflata, as understood by Seguenza (1862) or as differently interpreted by Billman et al. (1980), does not occur in southern Africa. The Agulhas Bank tests are hence regarded as Ammonia Japonica, with a local variation being the less high trochospire of the tests compared to those seen off the west coast. Although the species is known only as far back as the latest Eemian-—earliest Weichselian in southern Africa, Ammonia japonica ranges back to the Early Pleistocene in Japan (Matoba 1967). Ammonia parkinsoniana (d’Orbigny, 1839) s.1. Fig. 16F—H Rosalina parkinsoniana d’Orbigny, 1839a: 99, pl. 4 (figs 25-27). Ammonia parkinsoniana (d’Orbigny) forma tepida Poag, 1978: 397, pl. 1 (figs 1-4, 10-12, 17-18). Ammonia parkinsoniana (d’Orbigny) forma typica Poag, 1978: 397, pl. 1 (figs 5-9, 13-16, 19-21). Ammonia beccarii (non Linné) Martin, 1981: 48 (part.). Ammonia parkinsoniana (d’Orbigny) s.1. McMillan, 1987a: 35, figs 3a—r, 4—S. 170 ANNALS OF THE SOUTH AFRICAN MUSEUM Remarks Extensively distributed around southern Africa in shallow marine environ- ments in the Holocene and the latest Eemian—earliest Weichselian. The major morphological variations of the species in the southern African Holocene have been detailed by McMillan (1987a). Ammonia parkinsoniana s.1. occurs in the quieter water, clayier and siltier intervals, as well as the more active, sandier parts of the latest Pleistocene sections studied by McMillan (1987b) on the Orange Shelf. Genus Pararotalia Le Calvez, 1949 Pararotalia nipponica (Asano, 1936) Fig. 17A-—C Calcarina rotula (non Egger) Chapman, 1923: 3, pl. 1 (fig. 1). Rotalia nipponica Asano, 1936: 614, pl. 31 (fig. 2a—c). Rotalia ozawai Asano, 1951: 15, figs 115-117. Pararotalia taiwanica (non Nakamura) Huang, 1964: 56, pl. 2 (fig. 2a—c). Pararotalia sp. A Martin, 1974: 84, fig. 14-1 (part.). Pararotalia cf. P. inermis (non Terquem emend. Le Calvez) McMillan, 1974: 63, pl. 6 (fig. 3a—c). Pararotalia cf. P. nipponica (Asano) Martin, 1981: 48, pl. 9 (fig. 6). Pararotalia nipponica (Asano) McMillan, 1987b: 443, pl. 20 (figs 13-17), pl. 21 (figs 1-3), text-fig. 9. Remarks Studies of the ‘Alexandria Formation’ (McMillan in press) indicate that Para- rotalia nipponica ranges from the earliest Pleistocene (Unit II) to the present day; it is absent from the oldest, most inland part of the sequence (Early Pliocene, Unit I). In this respect it has a stratigraphic range comparable to Cibicides lobatulus (Walker & Jacob) and Elphidium advenum (Cushman). Chapman’s (1923) record of Pararotalia nipponica in the Late Cretaceous Mzamba Formation is clearly a contaminant from the overlying latest Pleistocene sands; similar tests, markedly different in their preservation to the in situ microfauna, were recognized by Makrides (1979). Neither Calcarina nor Para- rotalia, nor any morphologically similar genera with peripheral blunt spines, are known anywhere in the South African Late Cretaceous. The reference to Rotalia dentata Parker & Jones by Chapman (1907) from the ?latest Pleistocene of the Buffalo River, East London, almost certainly is referable here too. Siesser & Salmon (1979) reported ‘Pararotalia inermis’, supposedly from the Late Eocene Langental Beds of southern Namibia, but samples from pits dug in these outcrops yielded no trace of Eocene Pararotalia species (unpubl. studies). The illustrations given (Siesser & Salmon 1979, fig. 1Sc-d) are certainly of Pararotalia nipponica. It is evident that these Pararotalia tests, in keeping with those of Elphidium and Ammonia, were introduced into the Langental region with windblown sand derived from the littoral. The presence of Elphidium cf. E. crispum (Linné) at Langental (Siesser PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 171 Fig. 17. A-—C. Pararotalia nipponica (Asano). A. Dorsal view, CTC 4. x 98. B. Ventral view, CTC 4. x 122. C. Side view, same specimen as B. X 122. D. Globigerina (Globigerina) bulloides d’Orbigny. Ventral view, CTC 3. x 172. E-F. Globorotalia (Globoconella) inflata (d’Orbigny). E. Ventral view, CTC 3. x 181. F. Side view, CTC 3. x 235. G-H. Neoglobo- quadrina cf. N. pachyderma (Ehrenberg). G. Ventral view, CTC 3. x 200. H. Side view, Crea ex Zip: iV ANNALS OF THE SOUTH AFRICAN MUSEUM & Salmon 1979) perhaps suggests a latest Eemian—earliest Weichselian age for the windblown contaminants. Pararotalia nipponica occurs in the Varswater Formation (McMillan 1987b, Appendix A) of the Langebaan area. It has also been reported as ‘Rofalia-like foraminifers’ and ‘Rotalia’ in the Saldanha area (Visser & Schoch 1973). The species is widespread throughout the ‘Bredasdorp Formation’ (unpubl. studies), and was noted by Riimke (in Spies et al. 1963) under the name Rotalia audouini d’Orbigny, though it is distinctly different from D’Orbigny’s European Eocene and Oligocene species. In the “Alexandria Formation’ the species is widespread and abundant in the Early Pleistocene Unit II and latest Pleistocene Unit IIJ— again previously recognized by Riimke (in Engelbrecht et al. 1962) as Rotalia audouini. Specimens of ‘Rotalia sp.’ from the Bluff Sandstone Formation at Durban, described by Parr (1958), are certainly referable to Pararotalia nipponica. Some of this material was stored on grid slides at the Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, Johannesburg, and, through the courtesy of Professor M. Raath, were examined by the present author. Pararotalia nipponica also occurs, presumably in situ, in the latest Pleistocene (Unit III) Pecten bed in the lower half of the Uloa exposure, Zululand. As Streblus taiwanica (Nakamura), this species was encountered in Quaternary sands along the Inharrime to Vilanculos coast of Mocgambique (Rocha 1965). The Pararotalia nipponica-group exhibits wide morphological variation, parti- cularly in the height of the trochospire, size of the umbilical boss, and in the spinose, lobate or circular test outline. Attempts to divide the latest Pleistocene tests into discrete morphological formae proved to be impossible for the Orange Shelf material (McMillan 1987b). However, there does seem to be one tendency toward tests with a high trochospire, circular or weakly lobate periphery, and large umbilical boss (Pararotalia nipponica form), and one tendency to low trochospire, spinose periphery, and small or divided umbilical boss (Pararotalia ozawai form). These two trends resemble the distinctions made by Ujiié (1966) within the group. Bhalla (1972) considered sexual dimorphism to be partly responsible for the wide range of morphologies included within Pararotalia nipponica. Pararotalia nipponica is closely comparable to Pararotalia serrata, described by Ten Dam & Reinhold (1941) from the Pliocene of the Netherlands, and also known from similarly aged Crag deposits of eastern England (Funnell 1981). The precise differences between the two species remain to be determined. Family Globigerinidae Carpenter, Parker & Jones, 1862 Genus Globigerina d’Orbigny, 1826 Globigerina (Globigerina) bulloides d’Orbigny, 1826 Fig. 17D Globigerina bulloides d’Orbigny, 1826: 277, modéles no. 17 (1 ére livraison), no. 76 (4 éme livraison). PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 173 Remarks Only brief notes are given for this and the following planktonic species. One test, with the last-formed chamber broken away. Compared with examples encountered in the Holocene of the Orange Shelf (McMillan 1987b), this test displays a rather low-arched aperture but is otherwise typical. The species is characteristic of subpolar and transitional waters in the world oceans (Bé & Tolderlund 1971). Kennett & Srinivasan (1983) regarded Globigerina bulloides to range from the Middle Miocene to the present day. Family Globorotaliidae Cushman, 1927 Genus Globorotalia Cushman, 1927 Globorotalia (Globoconella) inflata (d’Orbigny, 1839) Fig. 17E-F Globigerina inflata d’Orbigny, 1839c: 134, pl. 2 (figs 7-9). Remarks Two specimens, one rather more lobate than is typical. Both are sinistrally coiled, as are almost all examples encountered by McMillan (19876) through the Holocene of the Orange Shelf. The species is most typical of temperate water masses (Bé & Tolderlund 1971; Stainforth et al. 1975). Its stratigraphic range is Late Pliocene to present day (Kennett & Srinivasan 1983). Genus Neogloboquadrina Bandy, Frerichs & Vincent, 1967 Neogloboquadrina cf. N. pachyderma (Ehrenberg, 1861) Fig. 17G—H see Aristerosira pachyderma Ehrenberg, 1861: 303; 1873: 386, pl. 1 (fig. 4). Remarks Two specimens of the shelf forma (non pachyderm), both sinistrally coiled and both exhibiting a kummerform final chamber. These tests show similarities to the shelf forma tests illustrated by McMillan (19875, pl. 22, figs 14-16) from a water depth of 46,5 m, close to the Namaqualand coast. However, the presence of the kummerform final chambers result in distinctly unusual apertures, and the unique apertural arrangement of Neogloboquadrina pachyderma is not seen. For this reason, the two tests are placed in the status of confer N. pachyderma. Sinistrally coiled Neogloboquadrina pachyderma appears to be associated with upwelling waters of the Benguela Current close to the west coast of southern Africa (Bé & Tolderlund 1971). The species ranges from the Late Miocene to the present day (Kennett & Srinivasan 1983). ACKNOWLEDGEMENTS I am most grateful to Mr V. H. Valicenti of SOEKOR (Pty) Ltd., Cape Town, for sending portions of samples collected by him from the municipal 174 ANNALS OF THE SOUTH AFRICAN MUSEUM trenching, for stratigraphy data on the samples, for much help in collecting the samples detailed in the three Appendices, as well as for valued discussions over many years on southern African Cainozoic and Mesozoic geology. Professor J. R. Haynes of University College of Wales, Aberystwyth, kindly made available facilities for sample processing, literature searching (including the Ellis & Messina ‘Catalogue of Foraminifera’), photographic darkroom facilities, and allowed access to the U.C.W. scanning electron microscope, during completion of a Ph.D. thesis late in 1987. Comparative material from the Varswater Formation was supplied through the courtesy of Mr V. H. Valicenti and Dr A. J. Tankard (formerly South African Museum), from the ‘Bredasdorp Formation’ by Mr J. Malan (Geological Survey, Bellville), and from the ‘Alexandria Formation’ by Messrs D. Stear and W. Smuts (both formerly University of Port Elizabeth). Material from the Bluff Sandstone Formation was kindly provided by Mr J. A. G. Cooper (University of Natal), and some Uloa Formation material generously supplied by Mr G. Kruger (Industrial Development Corporation). REFERENCES ALBANI, A. D. 1965. The Foraminifera in a sample dredged from the vicinity of Salisbury Island, Durban Bay, South Africa. Contributions from the Cushman Foundation for Foraminiferal Research 16 (2): 60-66. ALBANI, A. D. 1968a. Recent Foraminiferida from Port Hacking, New South Wales. Contributions from the Cushman Foundation for Foraminiferal Research 19 (3): 85-119. ALBANI, A. D. 1968b. Recent Foraminiferida of the central coast of New South Wales. Handbook. Australian Marine Science Association 1: 1-37. ALBANI, A. D. & JOHNSON, K. R. 1975. Resolution of foraminiferal biotopes in Broken Bay, N.S.W. Journal of the Geological Society of Australia 22 (4): 435-446. ApPTHorPE, M. 1980. Foraminiferal distribution in the estuarine Gippsland Lakes system, Victoria. Proceedings of the Royal Society of Victoria 91 (2): 207-232. ASANO, K. 1936. Fossil Foraminifera from Muraoka-mura, Kamakura géri, Kanagawa Prefecture. Journal of the Geological Society of Japan 43 (515): 603-615. ASANO, K. 1951. Illustrated catalogue of Japanese Tertiary smaller Foraminifera. Part 14 (Rotaliidae): 1-21. Tokyo: Hosokawa Printing Co. Bayuiss, D. D. 1969. The distribution of Hyalinea balthica and Globorotalia truncatulinoides in the type Calabrian. Lethaia 2 (2): 133-143. BE, A. W. H. & ToLDERLUND, D. S. 1971. Distribution and ecology of living planktonic Foraminifera in surface waters of the Atlantic and Indian oceans. Jn: FUNNELL, B. M. & RIEDEL, W. R. eds. The micropalaeontology of oceans: 105-149. Proceedings of the Symposium on Micropalaeontology of Marine Bottom Sediments, Cambridge, 1967. Cambridge: Cambridge University Press. BHALLA, S. N. 1972. Some observations on the taxonomic status of Pararotalia nipponica (Asano) and allied species of Foraminiferida. Journal of the Geological Society of India 13 (2): 175-177. Bresiot, P. G. gr. 1957. Miocene Foraminifera from the Uloa Sandstone, Zululand. Transactions and Proceedings of the Geological Society of South Africa 60: 61-80. BILLMAN, H., Horrincer, L. & OESTERLE, H. 1980. Neogene to Recent rotaliid Foraminifera from the Indopacific Ocean: their canal system, their classification and their stratigraphic use. In: Hortrincer, L. ed. Rotaliid Foraminifera. Schweizerische paléontologische Abhandlungen 101: 71-113. Bo.tovskoy, E. 1959. La corriente de Malvinas (un estudio en base a la investigacion de foraminiferos. Servicio de Hidrografia Naval, Buenos Aires H1015: 1-96. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 175 Bo.ttovskoy, E. 1961. Foraminiferos de la plataforma continental entre el Cabo Santo Tomé y la desembocadura del Rio de la Plata. Revista del Museo argentino de ciencias naturales ‘Bernardino Rivadavia’ (Zool6gicas) 6 (6): 249-346. Bo.tovskoy, E. 1963. The littoral foraminiferal biocoenoses of Puerto Deseado (Patagonia, Argentina). Contributions from the Cushman Foundation for Foraminiferal Research 14 (2): 58-70. Bo tovskoy, E., GIuSSANI, G., WATANABE, S. & WriGHT, R. 1980. Adlas of benthic shelf Foraminifera of the Southwest Atlantic. Den Haag: Dr W. Junk. Brapy, H. B. 1884. Report on the Foraminifera dredged by H.M.S. Challenger during the years 1873-1876. Reports of the Scientific Results of the Voyage of H.M.S. Challenger, during the years 1873-76 (Zoology) 9: 1-814 (2 volumes). BREMER, M. L., BRISKIN, M. & BERGGREN, W. A. 1980. Quantitative paleobathymetry and paleoecology of the Late Pliocene—Early Pleistocene Foraminifera of Le Castella (Calabria, Italy). Journal of Foraminiferal Research 10 (1): 1-30. CARPENTER, W. B., with PARKER, W. K. & Jones, T. R. 1862. Introduction to the study of the Foraminifera. London: Ray Society. Carter, A. N. 1966. Age of the Aliwal Shoal, South Africa. Nature 211 (5048): 507-508. CarTerR, A. N. 1978. Foraminifera from the Mitchellian Stage. Jn: BeELForD, D. J. & SCHEIBNEROVA, V. comp. Crespin Volume: In Honour of Irene Crespin. Bulletin. Bureau of Mineral Resources, Geology and Geophysics, Australia 192: 19-29. CHAPMAN, F. 1907. Report on Pleistocene Microzoa from a boring in the bed of the Buffalo River, East London. Records of the Albany Museum 2 (1): 6-17. CHAPMAN, F. 1909. Recent Foraminifera of Victoria: some littoral gatherings. Journal of the Queckett Microscopical Club (2) 10: 117-146 (for 1907). CHAPMAN, F. 1923. On some Foraminifera and Ostracoda from the Cretaceous of Umzamba River, Pondoland. Transactions of the Geological Society of South Africa 26: 1-6. Couns, A. C. 1953. Pleistocene Foraminifera from Port Fairy, western Victoria. Memoirs of the National Museum of Victoria 18: 93-105. CoLuins, A. C. 1974. Port Phillip Survey 1957-63: Foraminiferida. Memoirs of the National Museum of Victoria 35: 1-61. CusHMAN, J. A. 1915. A monograph of the Foraminifera of the North Pacific Ocean. Part 5. Rotaliidae. Bulletin. United States National Museum 71: 1-83. CusHMAN, J. A. 1919. Recent Foraminifera from off New Zealand. Proceedings of the United States National Museum 56 (2302): 593-640. CusHMAN, J. A. 1922. Shallow-water Foraminifera of the Tortugas region. Publications. Carnegie Institution of Washington 311: 1-85. CusHMAN, J. A. 1945. The species of Foraminifera recorded by D’Orbigny in 1826 from the Pliocene of Castel Arquato, Italy. Special Publications. Cushman Laboratory for Foraminiferal Research 13: 1-27. CuSHMAN, J. A. 1947. New species and varieties of Foraminifera from off the southeastern coast of the United States. Contributions from the Cushman Laboratory for Foraminiferal Research 23 (4): 86-92. CusHMAN, J. A. & Leavitt, D. H. 1929. On Elphidium macellum (Fichtel & Moll), E. striatopunctatum (Fichtel & Moll) and E. crispum (Linné). Contributions from the Cushman Laboratory for Foraminiferal Research 5 (1): 18-22. CuSHMAN, J. A. & WICKENDEN, R. T. D. 1929. Recent Foraminifera from off Juan Fernandez Islands. Proceedings of the United States National Museum 75 (2780): 1-16. EARLAND, A. 1934. Foraminifera. Part III: The Falklands sector of the Antarctic (excluding South Georgia). ‘Discovery’ Reports 10: 1-208. EHRENBERG, C. G. 1861. Uber die Tiefgrund-Verhiltnisse des Oceans am Eingange des Davisstrasse und bei Island. Monatsberichte der Koniglichen Preussischen Akademie der Wissenschaften zu Berlin: 275-315. EHRENBERG, C. G. 1873. Mikrogeologische Studien tiber das kleinste Leben der Meeres- Tiefgriinde aller Zonen und dessen geologischen Einfluss. Abhandlungen. Koniglichen Preussischen Akademie der Wissenschaften, Berlin 1872: 131-397. ENGELBRECHT, J. N. J., CoERTZE, F. J. & SNymMan, A. A. 1962. Die geologie van die gebied tussen Port Elizabeth en Alexandria, Kaapprovinsie: toeligting van blaaie 3325D (Port Elizabeth), 3326C (Alexandria) en 3425B. Pretoria: Geological Survey of South Africa. 176 ANNALS OF THE SOUTH AFRICAN MUSEUM FEYLING-HANSSEN, R. W. 1964. Foraminifera in Late Quaternary deposits from the Oslofjord area. Norges geologiske unders’ gelse 225: 1-383. FicHTEL, L. von & MOLL, J. P. C. von. 1798. Testacea Microscopica, aliaque minuta ex generibus Argonauta et Nautilus, ad naturam delineata et descripta (Mikroscopische und andere kleine Schalthiere aus den Geschlechtern Argonaute und Schiffer, nach der Natur gezeichnet und beschrieben). Wien: Camesina. Fitton, R. H. 1974. Late Cenozoic foraminiferal paleoecology of the Ross Sea, Antarctica. Micropaleontology 20 (2): 129-151. FUNNELL, B. M. 1981. Quaternary. Jn: JENKINS, D. G. & Murray, J. W. eds. Stratigraphical atlas of fossil Foraminifera: 286-293. Chichester: Ellis Horwood. Ga.Loway, J. J. & WissLeR, S. G. 1927. Pleistocene Foraminifera from the Lomita Quarry, Palos Verdes Hills, California. Journal of Paleontology 1: 35-87. GUALTIERI, N. 1742. Index testarum conchyliorum: 7-23. Florentia: C. Albizzini. Hapa, Y. 1931. Notes on the Recent Foraminifera from Mutsu Bay. Report of the biological survey of Mutsu Bay, Part 19. Science Reports of the Téhoku University (4, Biology) 6 (1): 45-148. HacemAN, J. 1979. Benthic foraminiferal assemblages from Plio—Pleistocene open bay to lagoonal sediments of the western Peloponnesus (Greece). Utrecht Micropaleontology Bulletins 20: 1-171. HANSEN, H. J. & ROcL, F. 1980. What is Nonion? Problems involving foraminiferal genera described by Montfort, 1808 and the type species of Fichtel & Moll, 1798. Journal of Foraminiferal Research 10 (3): 173-179. Haynes, J. R. 1973a. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur, 1962-1964). Bulletin of the British Museum (Natural History) (Zoology) Supplement 4: 1-245. Haynes, J. R. 1973b. Further remarks on Cardigan Bay Foraminifera. Publications. Department of Geology, University College of Wales, Aberystwyth 4: 1-17. Haynes, J. R. 1981. Foraminifera. London & Basingstoke: Macmillan. HEDLEY, R. H., HurDLE, C. M. & Burpbetr, I. D. J. 1965. A foraminiferal fauna from the western continental shelf, North Island, New Zealand. Bulletin of the New Zealand Department of Scientific-and Industrial Research 163: 1-46. HEDLEY, R. H., HuRDLE, C. M. & Burpett, I. D. J. 1967. The marine fauna of New Zealand: intertidal foraminifera of the Corallina officinalis zone. Bulletin of the New Zealand Department of Scientific and Industrial Research 180: 1-86. HERON-ALLEN, E. & EARLAND, A. 1922. Protozoa, Part II: Foraminifera. British Antarctic Terra Nova Expedition, 1910 (Zoology) 6 (2): 25-268. HERON-ALLEN, E. & EARLAND, A. 1924. The Miocene Foraminifera of the ‘Filter Quarry’, Moorabool River, Victoria, Australia. Journal of the Royal Microscopical Society, London 1924: 121-186. HERON-ALLEN, E. & EaRLAND, A. 1930. The Foraminifera of the Plymouth district. Journal of the Royal Microscopical Society, London (3) 50: 46-84, 161-199. HERON-ALLEN, E. & EARLAND, A. 1932. Foraminifera, Part I: The ice-free area of the Falkland Islands and adjacent seas. ‘Discovery’ Reports 4: 291-460. Hoyt, J. H., Oostpam, B. L. & SmitH, D. D. 1969. Offshore sediments and valleys of the Orange River (South and South West Africa). Marine Geology 7 (1): 69-84. Huanc, T. Y. 1964. ‘Rotalia group from the Upper Cenozoic of Taiwan. Micropaleontology 10 (1): 49-62. JOHNSON, K. R. & ALBANI, A. D. 1973. Biotopes of Recent benthonic Foraminifera in Pitt Water, Broken Bay, N.S.W. (Australia). Palaeogeography, Palaeoclimatology and Palaeoecology 14 (4): 265-276. JonkERS, H. A. 1984. Pliocene benthonic Foraminifera from homogeneous and laminated marls on Crete. Utrecht Micropaleontology Bulletins 31: 1-179. KANMACHER, F. 1798. Essays on the Microscope, etc., by C. Adams. (2nd ed.) London: Dillon & Keating. Kawal, K., Ucuio, T., UENo, M. & Hozuxi, M. 1950. Natural gas in the vicinity of Otaki, Chiba-ken. Journal. Association of Petroleum Technology, Tokyo 15 (4): 151-219. KENNETT, J. P. 1967. New Foraminifera from the Ross Sea, Antarctica. Contributions from the Cushman Foundation for Foraminiferal Research 18 (3): 133-135. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE NOY KENNETT, J. P. & SRINIVASAN, M. S. 1983. Neogene planktonic Foraminifera. A phylogenetic atlas. Stroudsburg, Pa.: Hutchinson Ross. KNUDSEN, K. L. (Ed.) 1971. Late Quaternary Foraminifera from Vendsyssel, Denmark and Sandnes, Norway—systematic part. Bulletin of the Geological Society of Denmark 21 (2-3): 185-291. KNUDSEN, K. L. 1973. The Lundergyrd Clay of Vendsyssel, Denmark, and its Foraminifera. Bulletin of the Geological Society of Denmark 22 (3): 155-192. Kruit, C. 1955. Sediments of the Rhone Delta. I: Grain size and microfauna. Verhandlingen van het K. nederlandsch geologisch-mijnbouwkindig genootschap 15: 357-499. LANKFORD, R. R. & PHLEGER, F. B. 1973. Foraminifera from the nearshore turbulent zone, western North America. Journal of Foraminiferal Research 3 (3): 101-132. Le CALvEz, J. & LE CALvEz, Y. 1958. Répartition des foraminiféres dans la Baie de Villefranche. I—Miliolidae. Annales de I’ Institut océanographique (n.s.) 35 (3): 159-234. LE Catvez, Y. 1974. Révision des foraminiféres de la collection D’Orbigny. I. Foraminifeéres des Iles Canaries. Cahiers de Micropaléontologie 2: 1-108. LENA, H. 1966. Foraminiferos Recientes de Ushuaia (Tierra del Fuego, Argentina). Ameghiniana 4 (9): 311-336. LINNE, C. von. 1758. Systema Naturae per regna tria naturae, secundum classes, ordines genera, species, cum characteribus, differentiis, synonymis, locis 1. (10th ed.) Stockholm: L. Salvius. LogeBLIcH, A. R. yr. & TAPPAN, H. 1953. Studies of Arctic Foraminifera. Smithsonian Miscellaneous Collections 121 (7): 1-150. LoeBLicH, A. R. yr. & Tappan, H. 1964. Sarcodina, chiefly ‘Thecamoebians’ and Foraminiferida. In: Moore, R. C. ed. Treatise on Invertebrate Paleontology. Part C. Protista 2: 1-900 (2 volumes). Boulder, Lawrence: Geological Society of America and University of Kansas Press. LoEBLICH, A. R. yr. & TAPPAN, H. 1974. Recent advances in the classification of the Foraminiferida. Jn: HEDLEY, R. H. & Abas, C. G. ed. Foraminifera 1: 1-53. London and New York: Academic Press. Lowry, F. M. D. 1987. Foraminiferal thanatocoenoses from the continental shelf of southern Africa. Unpublished Ph.D. thesis, University College of London. Makripes, M. 1979. Foraminifera of the Upper Cretaceous Mzamba Formation, Transkei, southern Africa. Unpublished M.Sc. thesis, University of the Witwatersrand, Johannes- burg. MarTIn, R. A. 1974. Benthonic Foraminifera from the western coast of southern Africa. Technical Report. Joint Geological Survey South Africa/University of Cape Town 6: 83-87. Martin, R. A. 1981. Benthic Foraminifera from the Orange—Lideritz shelf, southern African continental margin. Bulletin. Joint Geological Survey South Africa/University of Cape Town Marine Geoscience Unit 11: 1-75. MATHIEU, R., MOMENI, I., POIGNANT, A., ROSSET-MOULINIER, M., RouviLtois, A. & UBALDO, M. 1971. Les représentants des Miliolacea (foraminiféres) dans les sables des plages des environs de Dunkerque. Remarques sur les espéces signalées par O. Terquem. Revue de Micropaléontologie 14 (3): 157-166. Martosa, Y. 1967. Younger Cenozoic foraminiferal assemblages from the Choshi district, Chiba Prefecture. Science Reports of the Téhoku University (2, Geology) 38 (2): 221-263. Matosa, Y. 1970. Distribution of Recent shallow water Foraminifera of Matsushima Bay, Miyagi Prefecture, northeast Japan. Science Reports of the Tohoku University (2, Geology) 42 (1): 1-85. McCartny, M. J. 1967. Stratigraphical and sedimentological evidence from the Durban region of major sea-level movements since the Late Tertiary. Transactions of the Geological Society of South Africa 70: 135-165. McKenzigE, K. G. 1962. A record of Foraminifera from Oyster Harbour, near Albany, Western Australia. Journal of the Royal Society of Western Australia 45 (4): 117-132. McMiian, I. K. 1974. Recent and relict Foraminifera from the Agulhas Bank, South African continental margin. Unpubl. M.Sc. thesis, University College of Wales, Aberystwyth. McMiian, I. K. 1975. Pilot study: results of five Marine Diamond Corporation vibracore samples. Unpublished report to Marine Diamond Corporation, Cape Town. 178 ANNALS OF THE SOUTH AFRICAN MUSEUM McMittan, I. K. 1986. Cainozoic planktonic and larger Foraminifera distributions around southern Africa and their implications for past changes of oceanic water temperatures. South African Journal of Science 82 (2): 66-69. McMitian, I. K. 1987a. The genus Ammonia Briinnich, 1772 (Foraminiferida) and its potential for elucidating the latest Cainozoic stratigraphy of South Africa. South African Journal of Science 83 (1): 32-42. McMILian, I. K. 1987b. Late Quaternary Foraminifera from the southern part of offshore South West Africa/Namibia. Unpublished Ph.D. thesis, Univeristy College of Wales, Aberystwyth. McMi.tan, I. K. (in press) A foraminiferal biostratigraphy and chronostratigraphy for the Pliocene to Pleistocene upper Algoa Group, eastern Cape, South Africa. South African Journal of Geology. MestayYeErR, R. L. 1916. List of Foraminifera dredged from 15’ south of the Big King at 98 fathoms depth. Transactions and Proceedings of the New Zealand Institute 48: 128-130. Montacu, G. 1803. Testacea Britannica, or Natural History of British Shells, marine, land and freshwater. Romsey: J. S. Hollis. 3 vols. Murray, J. W. 1971. An atlas of British Recent foraminiferids. London: Heinemann. OrziGny, A. D. Dd’ 1826. Tableau méthodique de la classe des Cephalopodes. Annales des Sciences naturelles (1) 7: 245-314. OrzBicny, A. D. D’ 1839a. Foraminiféres. In: SAaGRA, R. DELA. Histoire physique, politique et naturelle d ’Ile de Cuba: 1-224. Paris: A. Bertrand. OrsiGny, A. D. dD’ 1839b. Voyage dans l’ Amérique méridionale— Foraminiféres 5 (5): 1-86. Paris: Pitois-Levrault et Cie; Strasbourg: V. Levrault. OrbBiGNy, A. D. Dd’ 1839c. Foraminiféres. Jn: BARKER-WEBB, P. & BERTHELOT. S. Histoire Naturelle des Iles Canaries 2 (2, Zool.): 119-146. Paris: Bethune. OrBiGNy, A.D. D’ 1846. Foraminiféres Fossiles du Bassin Tertiaire de Vienne (Autriche). Paris: Gide et Cie. Orsicny, A. D. dD’ 1852. Prodréme de paléontologie stratigraphique universelle des animaux mollusques et rayonnés 3: 1-196. Paris: V. Masson. PARKER, W. K. & JONES, T. R. 1865. On some Foraminifera from the North Atlantic and Arctic Oceans, including Davis Straits and Baffin’s Bay. Philosophical Transactions of the Royal Society, London 155 (1): 325-441. ParR, W. J. 1939. Foraminifera of the Pliocene of south-eastern Australia. Mining and Geological Journal 1 (4): 65-71. Parr, W. J.. 1941. A new genus, Planulinoides, and some species of Foraminifera from southern Australia. Mining and Geological Journal 2 (5): 305. ParR, W. J. 1945. Recent Foraminifera from Barwon Heads, Victoria. Proceedings of the Royal Society of Victoria (n.s.) 56 (2): 189-218. Parr, W. J. 1958. The Foraminifera of the Bluff Beds, Durban, and some other South African calcareous coastal rocks. Transactions and Proceedings of the Geological Society of South Africa 61: 103-109. Pinto, J. pos S. 1955. Contribuigao para o estudo dos foraminiferos dos sedimentos marinhos de Angola. Anais. Junta de investigacées do ultramar 10: 195-230. PLANCUS (BIANCHI), J. 1739. De conchis minus notis liber cuis accessit specimen aestus reciproci maris superi ad littus portumque arimini. Venetia: J. B. Pasqualus. PoaG, C. W. 1978. Paired foraminiferal ecophenotypes in Gulf Coast estuaries: ecological and paleoecological implications. Transactions of the Gulf-Coast Association of Geological Societies 28 (2): 395-421. PONDER, R. W. 1974. The foraminiferal genus Miliolinella Wiesner, 1931, and its synonyms. Micropaleontology 20 (2): 197-208. Reuss, A. E. 1848. Die fossilen Polyparien des Wiener Tertiarbeckens. Naturwissenschaftliche Abhandlungen, Wien 2: 1-109. Rocua, A. T. 1965. Contribuigéo para o estudo dos foraminiferos do Quaterndrio do sul da provincia portuguesa de Mocambique. Revista ‘Garcia de Orta’, Junta de investigacoes do ultramar 13 (3): 407-424. ROGL, F. & Hansen, H. J. 1984. Foraminifera described by Fichtel & Moll in 1798—a revision of Testacea Microscopica. Neue Denkschriften des naturhistorischen Museum, Wien 3: 1-143. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 179 RossET-MOouLINIER, M. 1972. Etude des foraminiféres des Cétes nord et ouest de Bretagne. Travaux. Laboratoire de Géologie, Ecole Normale Supérieure 6: 1-225. RovuviLLois, A. 1974. Un foraminifére méconnu du plateau continental du Golfe de Gascogne: Pseudoeponides falsobeccarii n. sp. Cahiers de Micropaléontologie 3: 1-9. SaLmon, D. A. 1979. Quaternary foraminifers in piston cores from the S.W. Indian Ocean. Technical Report. Joint Geological Survey South Africa/University of Cape Town Marine Geoscience Unit 11: 72-79. SALMON, D. A. 1981. Revised data on benthic Foraminiferida used as age-dating evidence in some South African Tertiary sediments. Technical Report. Joint Geological Survey South Africa/University of Cape Town Marine Geoscience Unit 12: 11-17. ScHROTER, J. S. 1783. Einleitung in die Conchylienkenntniss nach Linné 1. Halle: J. J. Gebauer. SEGUENZA, G. 1862. Prime ricerche intorno ai rizopodi fossili delle argille Pleistoceniche dei dintorni di Catania. Atti dell’Accademia gioenia di scienze naturali (2) 18: 84-126. SIESSER, W. G. & SALMon, D. A. 1979. Eocene marine sediments in the Sperrgebiet, South West Africa. Annals of the South African Museum 79 (2): 9-34. Spies, J. J., ENGELBRECHT, L. N. J.. MALHERBE, S. J. & VILJOEN, J. J. 1963. Die geologie van die gebied tussen Bredasdorp en Gansbaai: toeligting van blaaie 3419C en 3419D (Gansbaai) en 3420C (Bredasdorp). Pretoria: Geological Survey of South Africa. STAINFORTH, R. M., Lams, J. L., LUTERBACHER, H., BEARD, J. H. & JEFFoRDS, R. M. 1975. Cenozoic planktonic foraminiferal zonation and characteristics of index forms. Paleonto- logical Contributions. University of Kansas 62: 1-425 (2 volumes). TANKARD, A. J. 1975. The marine Neogene Saldanha Formation. Transactions of the Geological Society of South Africa 78 (2): 257-264. TEN Dam, A. & REINHOLD, T. 1941. Die stratigraphische Gliederung des niederlandischen Plio-Pleistozans nach Foraminiferen. Mededelingen van de Geologische Stichting (C-S) 1: 1-66. TERQUEM, O. 1876. Essai sur le Classement des Animaux qui Vivent sur la Plage et dans les Environs de Dunkerque: Deuxiéme Fascicule: 55-100. Paris: Terquem. (Also 1877— Mémoires de la Société dunkerquoise pour encouragement des sciences, des lettres et des arts 20: 146-191.) THompson, L. B. 1978. Distribution of living benthic Foraminifera, Isla de los Estados, Tierra del Fuego, Argentina. Journal of Foraminiferal Research 8 (3): 241-257. Ume, H. 1966. Shell structure of Japanese smaller Foraminifera. Part 2: Pararotalia nipponica (Asano, 1936). Transactions and Proceedings of the Palaeontological Society of Japan (n.s.) 61: 191-200. VissER, H. N. & ScHocu, A. E. 1973. The geology and mineral resources of the Saldanha Bay area. Memoirs. Geological Survey of South Africa 63: 1-150. WALKER, G. & Boys, G. 1784. Testacea minuta rariora nuperrime detecta in arena littoris sandvicensis. (A collection of the minute and rare shells lately discovered in the sand of the sea-shore near Sandwich.) London: Walker, G. WILLIAMSON, W. C. 1848. On the Recent British species of the genus Lagena. Annals and Magazine of Natural History (2) 1: 1-20. WILLIAMSON, W. C. 1858. On the Recent Foraminifera of Great Britain. London: Ray Society WRIGHT, J. 1886. Foraminifera of the Belfast Naturalists’ Field Club’s cruise off Belfast Lough, in the steam-tug Protector, June 1885; also Foraminifera found by Dr Malcomson, at Rockport, Belfast Lough. Proceedings. Belfast Naturalists’ Field Club 1885-6 (Appendix): 317-326. APPENDIX A THE LLANDUDNO-SANDY BAY COAST, CAPE PENINSULA Deposits of the same age as those described in the main body of this article occur intermittently around the Cape Peninsula. Along the Llandudno to Sandy Bay part of the coastline, ‘raised beach’ deposits occur about 10 m above present 180 ANNALS OF THE SOUTH AFRICAN MUSEUM IG°20'E \ Karbonkelberg Fig. 18. Locality map of sampling sides along the Llandudno to Sandy Bay coast, Cape Peninsula. mean sea-level as unconsolidated, white shelly sands at 34°01'05”S 18°20'00’E, just north of the Hamersteen (Fig. 18). Because of their loose nature, they are poorly exposed and tend to wash down slope, but occasional clear faces may be seen or excavated on either sides of paths leading inland from, and parallel to the beach. Large shells, particularly gastropods, are common and often well preserved. Three samples were collected from the best available exposures late in 1986, and revealed the presence of abundant, though rather poorly preserved, often leached, and abraded foraminifera tests (sample SB—LL 3). In addition, cemented shelly sandstone was noticed occasionally adhering to the large granite slabs and boulders in the present-day surf zone. One sample of sandstone revealed a microfauna very similar in composition to that of the ‘raised beach’ deposits, but severely affected by calcite leaching and precipitation, and somewhat less diverse. PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 181 The foraminiferal faunas of the four samples are closely comparable, both in species and composition, to those from Cape Town city centre. It is clear that these deposits near Llandudno are derived from the same regressional episode during the latest Eemian—earliest Weichselian. The faunas are distinguished by the predominance of Elphidium crispum (Linné) s.1., Pararotalia nipponica (Asano) and Cibicides lobatulus (Walker & Jacob), the first-named confirming the Unit III, latest Pleistocene age. Table 2 details the foraminifera recovered from the four samples. The absence of Ammonia parkinsoniana (d’Orbigny) s.1. in the Llandudno samples is of note, and probably reflects the minute rainfall catchment area of this part of the coast, with a corresponding lack of dilution of littoral sea-water. In contrast, numbers of Cibicides lobatulus are substantially greater than at Cape Town, probably indicating a more turbulent environment: a rocky shore rather than a beach. Again, a littoral to sublittoral environment is indicated by the foraminifera species present, though in such a wave-influenced shallow marine environment, considerable post-mortem transport of tests must have occurred. Macrofaunal debris is composed mainly of mollusc shell, some of which retains its life colours. In addition, barnacle plates, echinoid spines and sponge spicules occur. TABLE 2 Distribution of foraminifera in studied samples from the latest Eemian—earliest Weichselian deposits of the Llandudno to Sandy Bay coast, Cape Peninsula. Sample number Sp. Species SB-LL3 SB-LL2 SB-LL1 SB-1 O. 1 1 Spiroplectammina cf. S. wrightit 1 2 Quinqueloculina contorta 8) 2 2 5) 3 Quinqueloculina dunkerquiana 1 1 4 Quinqueloculina seminulum 3} 4 5 7 5 Quinqueloculina spp. (broken) 72 6 Siphonaperta sp. 1 7 Lagena perlucida 1 3 8 Lenticulina spp. 1 9 Bulimina gibba 1 2 10 Cassidulina laevigata 1 11 Rosalina bradyi 1 1 12. Rosalina cf. R. williamsoni iL 13 Rosalina sp. 3 1 8) 14 Glabratella australensis 17 3D) 27 71 15 Cibicides lobatulus 1 2 4 4 16 Cibicides spp. (broken) 1 1 3 2 17 Cibicidoides sp. 1 18 HAyalinea balthica 1 19 Elphidium advenum 1 2 20 Elphidium sp. A 38) sy) 36 96 21 Elphidium crispum s.1. 1 1 22 Elphidium macellum s.l. 1 23 Elphidium sp. (broken) ON 56 24 149 24 Pararotalia nipponica 80 173 105 349 Total number of tests 182 ANNALS OF THE SOUTH AFRICAN MUSEUM Many of the species present are confined to the west-coast littoral at the present day, and indicate that, as with the Cape Town fauna, water temperatures were closely comparable to those experienced today along this coast. It seems likely that the very close similarity of the foraminifera species in the latest Eemian-—earliest Weichselian of the south-western Cape and the Orange Shelf (McMillan 1987b) can be taken to indicate little difference in coastal sea-water temperatures between the two areas in the latest Pleistocene. APPENDIX B LATEST PLEISTOCENE FORAMINIFERA FROM BERG RIVER MOUTH Further north along the west Cape coast, near the mouth of the Berg River, deposits of the same age (latest Eemian—earliest Weichselian, Unit III) as those of Cape Town city centre are preserved beneath the Holocene estuarine sediments. Recent dredging and channelling operations for a new marina at Port Owen, just inside the estuary mouth, have excavated muddy and shelly sands that have been used as fill over the adjacent area. Two samples were collected in late 1986: one from a brown, clayey and lignitic sand, apparently in situ, near the base of a telephone pole next to the dredged marina channel (BR-2), and the other of muddy shelly sand dredged from the bed of the channel (BR-1). The localities of the samples are shown in Figure 19, and a list of the recovered foraminifera in Table 3. The two samples were from closely adjacent sites at 32°47'25"S 18°09’05"E. TABLE 3 Distribution of foraminifera in two studied samples from the latest Eemian—earliest Weichselian deposits of the Berg River mouth, west Cape coast. Sample number Sp. Species BR-2 BR-1 no. 2 1 Bulimina elongata 3 2 Bulimina gibba 1 3. Trifarina angulosa 1 4 Cassidulina laevigata 1 5 Rosalina cf. R. globularis 1 6 Glabratella australensis 8 6 7 Cibicides lobatulus 1 8 Cibicides sp. (broken) 11 49 9 Elphidium advenum 1 21 10 Elphidium sp. A 18 3 11. Elphidium cf. E. alvarezianum 62 87 12. Elphidium crispum s.1. 10 13. Elphidium macellum s.1. 58 68 14. Ammonia japonica 1 1 15 Ammonia parkinsoniana s.l. 113) 48 16 Pararotalia nipponica 182 293 Total number of tests PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 183 LAAIPLEK VELDDRIF Fig. 19. Locality map of sampling sites at Berg River mouth, west Cape coast. Both samples are rather gypsiferous, though this would seem to be a secondary feature derived from the overlying Holocene estuary. Present-day environments in the estuary are markedly hypersaline, and some areas are utilized for salt production. However, the foraminifera of the latest Pleistocene Unit II reveal a much more normal marine environment. Cibicides lobatulus (Walker & Jacob) occurs in fairly small numbers—as at Cape Town but unlike Llandudno-— and is suggestive of a sandy beach environment rather than a rocky shore. Ammonia japonica (Hada) in such numbers clearly illustrates a good connection to the open sea, whereas the lack of Ammonia parkinsoniana (d’Orbigny) s.1. perhaps implies little fresh water entering the Berg estuary during the latest Eemian-—earliest Weichselian. Very rare fragments of mollusc shell retain their 184 ANNALS OF THE SOUTH AFRICAN MUSEUM life colours, whereas all of the foraminifera tests are bleached. Occasional tests of Ammonia parkinsoniana s.\. with their natural brown colour are substantially better preserved than the rest of the assemblage, and are considered to be contamination from the Holocene estuarine deposits. The brown sand contains much lignite and clay, and it yielded rather more corroded foraminifera tests than the channel sample. It seems likely that the clayey sand accumulated in a non-marine or estuarine mud-flat environment, as plant fragments, seeds and non-marine gastropods also occur. The marine foraminifera shells were probably carried off the adjacent beach and into the estuary by onshore winds. However, the absence of Holocene foraminifera in the clay clearly identifies it as having accumulated before the Holocene transgression. The abundance of Elphidium crispum (Fichtel & Moll) s.1. in both samples clearly shows these deposits to have been laid down at the time of the latest Eemian-—earliest Weichselian regression. APPENDIX C LATEST PLEISTOCENE FORAMINIFERA OF HOEDJIESPUNT PENINSULA, SALDANHA Although the main body of the white, shelly and sandy limestones that cap the Palaeozoic and older rocks of the Hoedjiespunt Peninsula are the most obvious latest Cainozoic deposits of this region, there nevertheless are small residues of an even later unit. Foraminifera from shelly parts of the main limestones (the Varswater Formation, and including the ‘Saldanha Formation’) TABLE 4 Foraminifera species of one studied sample from the latest Eemian—earliest Weichselian sands of Hoedjiespunt Peninsula, Saldanha, west Cape coast. Sample Sp. Species no. no H-1 B) 1 Quinqueloculina dunkerquiana 2 2 Quinqueloculina seminulum 11 3. Quinqueloculina spp. (broken) 1 4 Siphonaperta sp. 1 5 Cassidulina laevigata 24 6 Glabratella australensis 9 7 Glabratella australensis (reworked) 5 8 Cibicides lobatulus 12 9 Cibicidoides sp. A 10 Elphidium advenum 3 11. Elphidium crispum s.1. 6 12. Elphidium sp. (reworked) 1 13. ?Haynesina sp. 3 14. Ammonia spp. (reworked) 112 15 Pararotalia nipponica 21 16 Pararotalia nipponica (reworked) 238 Total number of tests PLEISTOCENE FORAMINIFERA FROM CAPE TOWN CITY CENTRE 185 indicate a later Early Pleistocene age (see p. 132), regarded as equivalent in age to Unit ITA of the ‘Bredasdorp Formation’. However, bleached white, unconsolidated, shelly sands were recognized by Mr V. H. Valicenti about 8 m above high tide mark, below the Control Tower, on the north side of the Hoedjiespunt Peninsula, unconformably overlying the Varswater Formation limestones. Though severely affected by rainwash down- slope, and rather obscured by small plants and humus, as well as being contaminated with weathered debris from the adjacent limestones, one relatively clean sample was obtained. The locality is at 33°01'36"S 17°57'44’E, on Figure 20, and the foraminifera assemblage is listed in Table 4. Despite the apparent cleanness of the sample, reworked foraminifera from the Varswater Formation do occur, but they can be easily identified by being cemented, creamy in colour and well preserved. Jn situ tests, in contrast, are I7°57 E Outlier of Varswater Formation limestones +b ® =| HOEDJIESPUNT BOMGAT control tower Fig. 20. Locality map of the sampling site H—-1 at Hoedjiespunt, Saldanha, west Cape coast. 186 ANNALS OF THE SOUTH AFRICAN MUSEUM bleached white, rather damaged, often displaying borings, poorly preserved, and often appear rather ‘leached’. Again, the age of the shelly sands is regarded as being latest Eemian—earliest Weichselian, based on the presence of Elphidium crispum (Linné) s.l. The absence of Ammonia parkinsoniana (d’Orbigny) s.l. would seem to duplicate the situation along the Llandudno—Sandy Bay coast, with little coastal dilution of seawater. As with the previous deposits studied, the mollusc shell often retains its life colours, but unlike elsewhere, tests of foraminifera are relatively rare. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu- larly 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. nov., sp. nov., comb. nov., syn. nov., 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 trans- ferred 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 (fig. 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 specimens 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, descrip- tion 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 Eliza- beth (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. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text Sip: . 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 ea 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 Punctuation should be loose, omitting all not strictly necessary Reference to the author should preferably 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. The generic name should not be abbreviated at the beginning of a sentence or paragraph. Name 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. I. K. McMILLAN FORAMINIFERA FROM THE LATE PLEISTOCENE (LATEST EEMIAN TO EARLIEST WEICHSELIAN) SHELLY SANDS OF CAPE TOWN CITY CENTRE, SOUTH AFRICA LANAI fF Pus OUTH AFRICAN . : MUSEUM | KCAPE "TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (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) 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 and neat, double spaced with 3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for illustrations should be typed separately, their positions indicated in the text. All pages should be num- bered 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; 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 caption); the reduction or enlargement required should be indicated (and preferably uniform); orig- inals 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 caption; 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. If Letraset is used authors are requested to use Helvetica-style letter- ing, if possible. 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 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, 1969b) 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 (according to the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BuLtLoucu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FIsCHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140. FiscHER, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie expérimentale et générale 74 (33): 627-634. Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and Magazine of Natural History (13) 2 (17): 309-320. Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of the Bingham Oceanographic Collection, Yale University 17 (4): 1-51. THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHuLTZE, L. Zoologische und anthro- pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgeftihrt in den Jahren 1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270. (continued inside back cover) i ae ae Sy, Ae seit Bhp di Bi a. ‘(uassiry uea “ff “AQ Aq Ydessojoyg) “jorNsIp diopzyyeo ‘eesysjoueg wo ‘PSIOVV-NVS (D ‘201g Weenspfod, 24) “8009VV-INVS ‘A ‘S 2ARD YINO| JOATY solsepy Woy ‘g//8VV-INVS “A ToSYs Weorjspjo) oy} Woly YI0q “L8E8VW-NVS G-D “¢ eAeD YIMOW JOATY solsepy Wor ‘g//8VV-INVS “A ‘euskuy “peoH{ Wloyseq oy] WO ‘oOgegVW-INVS “V ‘“suOT}O9][00 s,winesny, UROLITW YINOS ay} UT soUO}s poauieg ‘| ‘sly ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 99 Band May 1990 Mei Part 6 Deel AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ By M. L. WILSON, W. J. J. van RIJSSEN S& D. A. GERNEKE 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 il, HAR3, S28), 02, 45, 8, tod), SAS, 5, 79), GG, baat), TOD), 8, NOB, T, 1OGE3) GED, 5, 7, East.) GD, ISCES), BAO, S), Ve, SG, BOG), 28, 36(2), 43(1), 45(1), 67(5), 84(2) Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 112 1 Printed) sini) South ay Atincas by, In Suid-Afrika gedruk deur Wine IRuBiies LESS, Ji. lil, Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap C647 AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ By M. L. WILSON, W. J.J. VAN RIJSSEN Archaeology Department, South African Museum, Cape Town & D. A. GERNEKE Electron Microscope Unit, University of Cape Town (With 10 figures) [MS accepted 23 July 1989} ABSTRACT The South African Museum has in its collections an unusually well-preserved, polychrome painted stone recovered from archaeological deposits in the southern Cape Province. Doubts have been expressed as to the authenticity of this artefact. Documentary research and micro-analysis of the paints indicate that, while there are a number of points that cannot be resolved, there is no evidence to refute the authenticity of the painting on the stone. CONTENTS PAGE MMR OGMCHO Mean mire tern ce tent taec ral, we itn ra At in os eh avtyoisyeiel dime deal 187 the documentaryevidence: 2 it osc compa ecs soos Le nea ey hor ee nae 189 | DPONAY'S CEICENE TUONO eho Slgctlaned ara che aha nny coctio te mat Cite en tar te nee ere 192 STMOSSCIOGINE INVISION gr aeeita Ste le utr ecto y eecuera ol ae ect coats NG) Cues sees area 194 Phy sicalianalySes- Othe plementS awn em onde ea. sce aa meas sae 195 IRONS TON Ts Sasa taug kpseseesmer cy et etree a ca as We idler en Oe err 195 INCI palmtsianGipicmiem(Sy nea ee ee es key ee leony 197 NVihivespaintsrandiplenmentSan= criss sacle, acess cel ented dee eee e § 201 Blackspaintsand pilemenmtS aes owes es aes a oe oe ee eta oy8 201 DISCUSSION Maw re eee ao en eet te acai eien ais dyh Wage aiayer ea 4 206 ithe stylerolsthe;pamlines s.r ark eee eas er epee he cee ee Seka 206 DISCUSSION anGd’CONnClUSIONS =. ace re ye een es aus a ek os hers havent aneia le 3 207 PNCKMOW IEC CSTE TICS Tpit ney tee eh na acer Mae 2, sia es 8 2, J RETTOT RETIN SOS a eo aastates oa been eelive, Sri ana ko. Repth aa Tater ata UAC Ds ee nea 212 INTRODUCTION In 1911 a remarkable painted stone (Fig. 1F) was reported to have been found in association with a human burial during excavations in a rock-shelter near the mouth of the Lottering River on the southern coast of the Cape Province of South Africa, in what is now the Tsitsikama Coastal National Park (Fig. 2). To this day this artefact remains unique because of the unusual state of 187 Ann. S. Afr. Mus. 99 (6), 1990: 187-213, 10 figs. 188 ANNALS OF THE SOUTH AFRICAN MUSEUM BOTSWANA MOZAMBIQUE SOUTH WEST AFRICA / NAMIBIA SWAZILAND LESOTHO SOUTH AFRICA ATLANTIC INDIAN OCEAN OCEAN TSITSIKAMA MOUNTAINS COLDSTREAM VILLAGE a OAKHURST a LOTTERINGS R. WITELS R. ELANDSBOS R. COLDSTREAM SHELTER GEELHOUTBOS R. INDIAN OCEAN Fig. 2. Maps showing the location of the Coldstream shelter and Whitcher’s Cave. AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 189 preservation of the painting and because it is the only known polychrome painted stone to have come from archaeological deposits in South Africa. This stone, together with the associated skeleton, was on display in the South African Museum for a number of years. When the display was changed the stone was put for safe-keeping in the Museum’s strong-room, stored in the wooden box made for its transport from Coldstream, and was seen by very few people, although a copy was later put on display in the Museum. The skeleton was integrated into the Museum’s physical anthropology collections with the others excavated from the shelter but, unfortunately, can now not be identified. In 1980, pending the removal of most of the Museum’s staff and collections to temporary premises during building operations, the ‘Coldstream Stone’, as it has become known, was returned to the Archaeology Department’s artefact col- lections, where it has since been seen more frequently by researchers. Sight of the stone itself and comparison of it with other painted stones in the Museum’s collections, such as those shown in Figure 1, have raised doubts that anything so fragile as a painting could have survived in so remarkable a state of preser- vation, given that it was said to have been found in the damp deposits of a coastal rock-shelter. Since it is clearly in the Museum’s interest, as well as that of other researchers, to know whether or not this unique artefact is genuine, various lines of enquiry were followed. These included documentary research, micro-analysis of the paints and comparison with paints and pigments from other sources, and the opinion was sought of those who knew the excavator. These points are dis- cussed below. THE DOCUMENTARY EVIDENCE The first record of the finding in South Africa of painted stones associated with human burials appears to have been a report by J. S. Henkel, Conservator of Forests at Knysna, to the then Director of the South African Museum, L. Péringuey. Henkel’s correspondence cannot now be traced, but Péringuey wrote to him on 12 April 1909 ‘I never heard of painted stones or slabs placed above the head. I’ll go further and say that nothing of this kind is known and your announcement took my breath away’ (S.A. Museum Letter Book 1908-1910: 184). Péringuey was obviously intrigued by Henkel’s report and possibly stimu- lated by another letter from him, for he wrote to Henkel again on 7 July ‘I have a group of, perhaps dancers, perhaps dead bodies judging from their attitude, painted on a small stone, found in one of the Knysna grottoes or caverns’ (SAM Letter Book 1908-1910: 376). Again, on 18 December, he wrote to Henkel ‘Hitherto painting or carving on stone could not be traced to these primitive people—I mean the Strand Looper—except however a small painting which is alleged, and I am sure rightly so, to have been found in a cave’ (SAM Letter 190 ANNALS OF THE SOUTH AFRICAN MUSEUM Book 1908-1910: 835). The stone referred to is one found in 1872 by the son of a Mr Chevalier in a cave in the Eastern Head at Knysna, about two feet (0,6 m) below the surface (J. Rudner 1971: 54; see also Péringuey 1911: 162-163) and which is in the Museum’s collections (SAM—AA8386— Fig. 1A). Péringuey twice again referred to this stone in letters to Henkel the follow- ing year. On 12 May 1910 he wrote ‘I have had a good deal of information regarding the contents of some of these caves in the Knysna—Plettenberg Bay coastal region and I thought we had the only painting, a monochrome, found there, until you told me of Witcher’s [sic] discovery’ (SAM Letter Book 1910-1911: 136). On 11 October he wrote ‘The painting has been carefully pre- served—it is a small fragment and had the painted stones of Coldstream not been found, I would have been left in doubt as to it having been detached from the sides or roof of the cave’ (SAM Letter Book 1910-1911: 527). C. J. Whitcher was the owner of a sawmill near the little village of Cold- stream (Fig. 2) and of extensive tracts of indigenous forest in the surrounding area. He excavated in the shelter at the mouth of the Lottering River, one of several shelters on his property, possibly in 1908 but certainly early in 1909, making the discoveries referred to in Henkel’s correspondence with Péringuey. Péringuey (1911: 154-158) published excerpts from Henkel’s correspon- dence in his seminal monograph on the Stone Ages of South Africa, in which Whitcher’s excavations are described. Initially, Henkel emphasized ‘You will observe from my account that there was no direct evidence to connect the stones decorated with drawings with skeletons, although in the first instance I was led to believe that this was so’ (Péringuey 1911: 158). However, later in the same extract, Henkel confirmed from personal observation on 14 July 1909 that from one to three stones, none of them painted, were found overlying skeletons excavated during his visit. Péringuey was extremely anxious to obtain for the Museum the artefacts and skeletons excavated by Whitcher, as well as to ensure that any future exca- vations were carried out systematically. Henkel negotiated with Whitcher on Péringuey’s behalf for the purchase of the material and on 7 July 1910 Péringuey wrote to Henkel, acknowledging receipt of it and commenting “The ostrich egg- shell beads are interesting because found on the child, and the cockle necklace is next to the painted stone—the larger one—the interesting part of the collec- tion’ (SAM Letter Book 1910-1911: 271). Péringuey’s reference to ‘the larger one’ of the painted stones, implying as it does that two were received, calls for comment. In a paper on painted ‘burial stones’ from the Cape, J. Rudner (1971: 56) described four painted stones from the Coldstream shelter. Rudner’s C1-—3 are said to have been found by Whitcher and to have the accession number SAM-[AA]8387, whereas his C4, accession number SAM-[AA]6008, is the painted stone that is the subject of this paper. According to Rudner, ‘C1 is a lower grinding stone with four human figures painted on a red background . . . C2 consists of a rough rock fragment with a black painting on the upper part carrying a long, probably cylindrical quiver and a NN AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 191 a bow . . . C3 originally contained four black figures of animals, but these have now disappeared’. The Archaeology Department’s accession register lists under SAM- AA8387 only two painted stones that, from Rudner’s descriptions, are his Cl and C2. According to Péringuey (1911: 163), ‘Mr Henkel reports that the third one became obliterated as soon as exposed to light’. It is thus possible that Péringuey actually received three stones, but omitted to mention the third, Rudner’s C3, from which the painting had vanished and which was accordingly not accessioned as a painted stone. The matter of the acquisition of Rudner’s Cl (Fig. 1D) also calls for comment. On 12 April 1912 Péringuey wrote to Whitcher ‘I should very much like you to loan me the tracing taken by Fourcade of the dancing women, on the stone found in the cave. Our copy has greatly faded and I want to re-touch it’ (SAM file M—18—d). This seems a strange request if the Museum had acquired the stone itself in 1910. Moreover, Péringuey (1911, fig. 199) had published an artist’s impression of the stone, coloured lithographs of which are still in the Museum’s possession. However, Péringuey (1911: 163) observed of the painted stones from Coldstream “They are fading so rapidly that we had a great difficulty in reproducing them’. He may therefore have considered that Fourcade’s copy would provide a more accurate representation than could be obtained from the stone itself. Whatever the case, it is evident from the foregoing that, up to the beginning of 1911, the only painted stones known to exist in South Africa were considered to be either monochrome: the Knysna Heads stone (Fig. 1A), or bichrome: the ‘dancing women’ stone (Fig. 1D) and the smallest of the three stones from the Coldstream shelter (Fig. 1C). It should be pointed out that, although—as mentioned above—Peéringuey considered the Knysna Heads stone to be a monochrome, the black figures are painted on a red ochre background, as are the ‘dancing women’. Where Rudner’s C2, the smallest of the Coldstream stones, is concerned, microscopic examination of what Rudner (1971: 56) described as ‘a white mark ... that ...may be part of the picture’ has shown that it is in fact white pigment, evidently depicting a bag, or possibly two, slung over the figure’s shoulder. Samples of the pigments on this stone were included in the analyses discussed below. Péringuey (1911: 163) did not at first consider that the ‘hook-heads’ of the figures in the ‘dancing women’ painting resulted from the loss of the white pigment used to depict the faces, a phenomenon now well known to occur in paintings on rock-shelter walls (Yates et al. 1985: 70; W.J.J.v.R. pers. obs.). He suggested that it might be a totem, but thought there was no doubt that it was ‘a distinctive badge’. However, in the final chapter of his monograph, which deals with the Coldstream shelter and appears to have been written after the main body of his text, Péringuey (1911: 209), in describing the ‘Coldstream Stone’, commented ‘The face is painted white and the crescent-shaped process... 192 ANNALS OF THE SOUTH AFRICAN MUSEUM which I took at first to be possibly the mark of a clan—stands for the hair and is black’. The conclusion is, therefore, that on the basis of the information then available, there was no reason to suppose that anything other than monochrome or bichrome painted stones would be found. DRURY’S EXCAVATION On 16 February 1911, on Péringuey’s instructions, James Drury, the Museum’s taxidermist (Fig. 3), started excavations in the shelter in which Whitcher had made his finds. The following are excerpts from Drury’s field notes (SAM file M-—18—d). In the interests of authenticity, the only alterations that have been made to Drury’s text are the omission of certain passages that are not directly relevant and the separation of the daily information into para- graphs. The days of the week, written in the margin of Drury’s notes, have been incorporated into the text. ‘Mr Whitcher had dug out a part of the cave marked in sketch, he had got to a depth of 10 feet, so I dug another 2 feet found remains of Baby skeleton very badly preserved, and as I don’t think skeletons or paintings can be of any use at that depth I gave up working and made a fresh start, at the back centre of cave as in sketch. I found the top layers sandy for a few inches then shells stones and refuse to a depth of just on 3 feet, then a layer of limy ground varying from an inch to 3 inches thick and so on in alternating layers of ashes earth lime and shells to a depth of about 4 feet 6 in. then the ground is brownish earthy and rather damp... . Thursday. I started a section next to where Mr Whitcher had worked and am now working down to the same level as first section about 11 feet deep... . At a depth of 4 feet I found my first skeleton, no. 2. . . their was no flat stones on top or side of him . . . this skeleton is very rotten (as I find all are) and I don’t think it will be any good to go too deep. Next skeleton was a baby but as before too far gone, this was close to the above skeleton, then I found another just a little deeper . . . lying on right side with a stone on top this time but their was no trace of Paintings. All through we found lots of colouring sub- stances but no trace of painting. I got charcoal, Red Ochre also Yellow Ochre but so far no white. Friday. I started another section down and when nearly the end of the layer I found the second full skeleton and the most interesting find so far. I was having layers carefully removed when I saw back part of skull protruding out of ground so started to carefully work earth away all round with trowels and hands and in so doing found a large flat stone lying on shoulder, so with great care removed the earth from on top, and thus exposed was the finest Painting I have so far seen anywhere, and with great care got it cleaned of all dirt adhering and so removed to Mr Whitcher’s house where I finished cleaning, their is three figures in colours and realy marvellous it is. The position of stone was flat on shoulder with feet end of figures next to the jaw of skeleton on painting side AN INVESTIGATION OF THE ‘COLDSTREAM STONE? 193 Fig. 3. James Drury excavating in the Coldstream shelter, February 1911. (Photograph: S.A. Museum archives.) 194 ANNALS OF THE SOUTH AFRICAN MUSEUM up... there was a flat stone lying on hips of skeleton also but there were no tracings of Paintings . . . close by him I| found for the first time two pieces of white clay which now gives all the colours found on painting, this clay was in a hole in rock just at back of where the painting was found... .’ Drury continued excavating until 28 February, removing the bulk of the deposits and finding a total of 28 skeletons. Some of the burials had stones associated with them, but none of them was painted. According to J. Rudner (1971: 56) the skeleton with the painted stone was found at a depth of 6 feet (approx. 1,8 m). This is not stated in Drury’s notes but the three sketches of burials that accompany his notes are, according to the captions, of the first three skeletons, that noted as having been found at a depth of 6 feet being identified as ‘2nd skeleton’ and ‘no. 10’. It will be seen above that on the Friday, Drury mentioned finding ‘the second full skeleton’ and Rudner’s identification may thus be correct. None of the three sketches contains infor- mation about the placement of stones apart from a note that the first skeleton had ‘no big stones on top’ (SAM file M—18-d). SUBSEQUENT HISTORY As already mentioned, Péringuey (1911: 209, pl. 27) described the ‘Cold- stream Stone’ in his monograph but illustrated only the other two painted stones from the site and the Knysna Heads stone. This tends to confirm that his chapter on the Coldstream shelter was a late addition to his manuscript and that he did not have time to have a plate prepared from an artist’s impression, as was the case with the ‘dancing women’ stone. One was prepared at some stage and sent to Germany for the preparation of coloured lithographs, but due to the exigen- cies of the First World War these did not reach South Africa until after 1920 (R. Summers pers. comm. 1985). It was left to Haughton (1926) to publish the first illustration of the ‘Coldstream Stone’, after Péringuey’s death in 1924. Burkitt (1928, fig. 20) illustrated only the ‘dancing women’. Goodwin (1929: 209, pl. 42), in discussing painted stones, merely remarked that ‘Coldstream shows three female figures and a child’ and illustrated a monochrome from the collections of the Port Elizabeth Museum (PEM 603). In a later monograph Goodwin (1946: 115-16) mentioned the coastal caves of the southern Cape and the ‘amazing efflorescence of culture’ of the area, which included painted ‘grave- stones’; and elsewhere in the same work (Goodwin 1946: 127) he discussed Péringuey’s (1911) monograph. As mentioned above, the latter publication con- tains a description of the ‘Coldstream Stone’ though not an illustration of it. Goodwin must therefore have known of the stone’s existence; and, as he was at that time honorary curator of the Museum’s archaeological collections, he could accordingly be expected to have known of the stone’s presence in the Museum’s collections. Moreover, he sent a copy of the coloured lithograph to J. D. Clark (Clark pers. comm. to J. Deacon 1986). It seems, however, that Goodwin never published a reference to this unique artefact in the years remaining before his AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 195 death in 1959. Goodwin’s apparent failure to refer to the ‘Coldstream Stone’ in any of his prolific writings is a puzzling aspect of this case for which there is now no ready explanation. On the other hand, E. M. Shaw, until recently Head of the Ethnography Department, who started at the Museum in 1933, and in whose charge the archaeological collections were until the appointment of an archae- ologist to the Museum’s staff in 1967, has said (pers. comm. 1986) that it was many years before she became aware of the existence of the ‘Coldstream Stone’; and it may be that it lay forgotten in the Museum’s strong-room. (It was only accessioned in 1956.) The ‘Coldstream Stone’, remarkable though it is, appears to have been ignored for more than thirty years after Haughton’s publication, until Clark (1959: 214, pl. 15b) mentioned ‘the magnificent painted stones of Coldstream’ and included an illustration of the stone under discussion. Since then it has been described and/or illustrated by Woodhouse (1968, 1969, 1987, 1988 in Coertze ed. 1988 (unpaginated)), Lee & Woodhouse (1970), Rudner & Rudner (1970), J. Rudner (1971), Summers (1975), I. Rudner (1982), Lewis-Williams (1984) and Willcox (1984). None of these authors commented on the fact that this is the only known polychrome in the category of painted stones, and only J. Rudner (1971: 56) and Lewis-Williams (1984: 241) commented on the remarkable state of preservation of the painting. PHYSICAL ANALYSES OF THE PIGMENTS INTRODUCTION The rarity and consequent importance of the painted stones impose severe restrictions on the size of the sample available for analysis. The sampling tech- nique must be as little destructive as possible, which means that only small quantities of the paint may be removed. The paint is nowhere thick, with an estimated maximum of 100 u; and this further affects the available size of the sample. Energy-dispersive X-ray (EDX) micro-analysis is thus the only tech- nique available for this purpose. Samples were obtained by placing a 5x 5 mm piece of silver mylar tape (Scotch 3M 850 PAU333310) with the adhesive face down on to the painted area and rubbing the back of the tape with a small, blunt object. The effect of this sampling technique was minimal and the places from which the samples were removed cannot be detected with the naked eye. The pieces of tape were then glued to Cambridge 100 stubs and carbon coated, to an interference colour blue, in a Balzer’s high-vacuum evaporator. The paint and raw pigment samples were analysed using a KEVEX 7000 EDX system, interfaced with a Cambridge $180 scanning electron microscope. Standardized operating conditions were used for all the analyses. Conditions included a calibrated 15 keV probe, 38° take-off angle and a 20 per cent amplifier dead time. The electron optic parameters gave a probe current of 1,2 x 10-° amps. All the analyses were carried out using the same electron optic 196 ANNALS OF THE SOUTH AFRICAN MUSEUM parameters so that spectra would be comparable, even though absolute values were not obtained. EDX micro-analysis is capable of detecting any element having an atomic number greater than 11 (Na) and in practice the detection limit is about 0,1 weight per cent (wt %) in the analysed volume. The volume analysed depends on sample density, but is in the order of 0,08 mm? for a sample of density 8 g/mm? and using a 15 keV probe and a scanning electron microscope magnifi- cation of 1 000 X on the visual screen. By measuring the intensity of X-ray emission for each element and comparing the experimental with the theoretical X-ray yield, the concentration of constituent elements can be determined. Unfortunately, the relative errors in this method of analysis can be high, particularly if the sample contains unde- tectable elements such as oxygen and carbon, and is not homogeneous and flat. The more satisfactory strategy involves direct comparisons of X-ray yield with similar accurate standards. The difficulties in this method are obtaining suitable standards, polishing the samples, and particle selection based on morphology. For each sample the secondary electron image was observed and analyses carried out, initially using area mode (the probe scanned in a raster over an area of about 30 x 30 u) on visually selected differences. Nine additional spot or area acquisitions (the probe stationary in one position) were undertaken on selected, morphologically similar particles or areas in order to establish the nature of the sample and to determine the average concentration of detectable elements. In selecting the area or spot for acquisition, the probe was carefully positioned to assist In minimizing topographical, refluorescence and interaction volume errors. In this connection, it must be pointed out that, although the samples analysed were small, because they were almost without exception not homogeneous, probes that, for example, included a grain of sand would yield results very different from that of pure ochre. The raw pigments may have been ground on a quartzite grindstone, using a muller of the same material, which would introduce foreign material. The pigment was mixed with some sort of binder and probably liquefied with additional fluid, both of which would introduce more foreign elements. Incontrovertible factual information regarding the nature of the binders used in rock paintings is not available, although much has been written about these, and a number of experiments carried out by various researchers. I. Rudner carried out extensive documentary research and fieldwork in regard to Khoisan paints and pigments and their relationship to rock paintings, and con- cluded (I. Rudner 1982: 264) that there is virtually no primary information regarding the nature of the binders used, and that very little of what is available can be considered as reliable. She was particularly sceptical of the claim by Den- ninger (1971) that a number of binders had been identified by means of paper chromatographic analysis of the constituent amino-acids (I. Rudner 1982: 33-34). AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 197 The results of the five analyses of each sample that were most similar were averaged and are presented in graphic form in Figures 4-10 as weight per cent (wt %) of the elements detected in the analysed volume, normalized to 100 per cent. Since the samples are not homogeneous and averages were used, the results characterize the samples rather than provide precise, absolute determinations of their constituents. When studying the results of the EDX analyses given below, it must be borne in mind that none of the samples can be considered as chemically pure. They have all been subjected to various taphonomic processes: adsorption, absorption, and leaching of material, in addition to the other forms of contami- nation mentioned above. It should also be remembered that silicon (Si) and aluminium (Al) are the two most common elements in the earth’s upper crust (hence its name, sial). For the purposes of these analyses, the presence of these elements should be considered as ‘noise’ and, for the most part, irrelevant. Included in the small collection of non-skeletal material Drury retained from his excavation (SAM-—AA1378, AA1379) are a few pieces of ochre as well as one of the pieces of ‘white clay’ that he recorded (see above) as having found in the deposits. It was therefore decided to analyse samples of these as well as of the paint on the ‘Coldstream Stone’ and, where necessary, to compare the results with those of other paints and pigments. Because of a lack of concord- ance between the results of the analyses of the various white and black pigments in the earlier series of analyses, samples of marine shell and burnt animal bone were also analysed. There is too little black paint left on the ‘dancing women’ stone (Fig. 1D) to allow for analysis, and the red ochre background is also a complicating factor. Samples of the black and white paints from the smallest of the three stones from the site (Fig. 1C) were, however, included in the analyses. RED PAINTS AND PIGMENTS Four samples of red paint from the ‘Coldstream Stone’ were analysed, and two of unprocessed ochre from the excavation that visually matched the light red and dark red paints most closely. The results are shown in histogram form in Figures 4—6. The light red paint from the front (left) figure (Fig. 4A) has less iron (Fe) and lower frequencies of the minor elements (S—Ti) than the dark red paint from the central figure (Fig. 4B); otherwise these two samples are broadly similar. Both the light red ochre from the excavation (Fig. 5A) and the dark red (Fig. 5B) have very little iron by comparison with the frequencies in the paints (Figs 4A—-B) and are generally dissimilar to the paints in most respects. The light red ochre lacks five of the minor elements present in the dark red. These minor elements in the latter may be the cause of its darker colour. Figure 6A is the analysis of the light red paint from the front leg of the last (right) figure and Figure 6B that from the back leg. It is immediately evident 198 ANNALS OF THE SOUTH AFRICAN MUSEUM | za tu O oc uu Oo. Ke cL S iu = a - Na Mg Al Si P S Cl K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT -_ = iu O oc uu Oo. | cd < S wi = Na Mg Al Si P S CI K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT Fig. 4. Histograms of amounts (wt%, normalized to 100%) of elements detected in paint samples. A. Light red paint from the ‘Coldstream Stone’. B. Dark red paint from the ‘Coldstream Stone’. AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 199 kK za WW O oO WwW oO. kK ae Oo WW Ss Na Mg Al Si P S Cl K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT 70 60 50 - za Ww O 40 o WwW Oo. - 30 GS Ww GST, = 20 10 mt WE, ot aaa an le a —-——— 7A ZI Na Mg Al Si PS Cl K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT Fig. 5. Histograms of amounts (wt %, normalized to 100%) of elements detected in pigment samples. A. Dark red ochre from the excavation. B. Light red ochre from the excavation. 200 ANNALS OF THE SOUTH AFRICAN MUSEUM = = Lu O oc Ww oO. = < 2 Ww = Na Mg Al Si P S CI K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT = = Lu O oc Lu Oo. pis <= 2 Lu = Na Mgr All Si PS) Cl 1K Ca Ti Mn Sees Co Ni MCusEa ELEMENT Fig. 6. Histograms of amounts (wt%, normalized to 100%) of elements detected in paint samples. A. Light red paint from the front leg of the last (right) figure on the ‘Coldstream Stone’. B. Light red paint from the back leg of the same figure. AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 201 that the two samples differ considerably from each other and from those of the other two figures. The paint from the front leg is similar to that of the light red ochre from the excavation (Fig. 5A), although this contains only about half as much iron. The presence of a relatively high frequency of calcium (Ca) in the sample from the back leg is not easy to explain. Unlike the front leg, there is no white paint on the back leg, which is, however, outlined in black on its front (Fig. 1). A possible source of the black paint used on the ‘Coldstream Stone’, and thus of the calcium in this sample, is given below. WHITE PAINTS AND PIGMENTS Samples of white paint from the face of the central figure on the ‘Cold- stream Stone’ and the white bag(?) carried by the figure on the smallest painted stone from the site were analysed, as was a sample of the white ‘clay’ Drury reported as having found in the excavation. The presence of a high frequency of calcium in the samples could not be explained, particularly as one of the samples analysed for comparative purposes (but not included here), from a mural paint- ing in the Transkei, only contained a minimal amount of calcium (approx. 6%). A sample of calcined marine shell from an excavation in the southern Cape was therefore also analysed. The results of these analyses are shown in Figures 7-8. The paint from the ‘Coldstream Stone’ (Fig. 7A) and that from the smallest stone (Fig. 7B) differ only in the amount of calcium present and in the presence or absence of various minor elements. The white ‘clay’ (Fig. 8A) has consider- ably less calcium than the paint samples, partly because of the increased presence of other elements. Examination of this material under a standard binocular microscope indicated that it is not clay but an agglomerate of calcined material that includes marine shell, sand and other unidentified material. Because of uncertainty as to the precise nature of the white paint, a sample of calcined marine shell, brown mussel Perna perna, without its periostracum, was also analysed (Fig. 8B). Despite differences caused by the presence or absence of small amounts of minor elements in the two paint samples, the result of the shell analysis strongly suggests that shell was the pigment used for the white paint on both stones. It also tends to confirm that the white ‘clay’ is prob- ably a lump of calcined shell with other admixtures. Another possibility, egg- shell, which is also rich in calcium, was considered later but not tested. BLACK PAINTS AND PIGMENT Samples of black paint from the quiver or bag held by the central figure on the ‘Coldstream Stone’ and from the figure on the smallest of the stones from the site were analysed. The results of these were compared with that of black paint from a mural rock painting in the Cedarberg Mountains of the western Cape. As will be seen below, this bore no resemblance to the paint on the two stones. On the basis of the discovery that the white paint on these stones had probably been made from calcined shell, it was decided to analyse a fragment of 202 ANNALS OF THE SOUTH AFRICAN MUSEUM = Zz Lu O o Lu oO. = <= S Lu = Na. Mg Al Si P S* Cl K Ca Ti Mn Fe Co NiCusBa ELEMENT = = Lu O eg LJ oO. = It e Lu = Na a tes Al K Ca Ti Mn Ti Ma Fe. ce) Noe Co Ni Cu Ba ELEMENT Fig. 7. A. Histograms of amounts (wt %, normalized to 100%) of elements detected in paint samples. White paint from the ‘Coldstream Stone’. B. White paint from the smallest of the painted stones from the Coldstream shelter. AN INVESTIGATION OF THE ‘COLDSTREAM STONP’ 203 70 60 50 40 30 WEIGHT PER CENT 20 10 LZ, 0 82 Vea ETE, A SEC EAN TERY Nae Mg Alsi PS (Cl K Ga Ti Mn Fe Co Ni Cu Ba ELEMENT WEIGHT PER CENT NaiViqurAle Site Penton Clenheu Can til Ninpte Co mNi- Cus Ba ELEMENT Fig. 8. Histograms of amounts (wt %, normalized to 100%) of elements detected in pigment samples. A. White ‘clay’ from the excavation. B. Calcined marine shell. 204 ANNALS OF THE SOUTH AFRICAN MUSEUM 70 60 50 - za Lu O 40 a Lu Oo. FE 30 Oo > 20 10 Na’ Mg) VAll "Si PS) Cl 3K Cas Ti Mn Fem Com NimCumea ELEMENT = Fz Lu O oc Wu o. = ag So Lu = Na Mg “Al” Si PS) Cl 4K Car it Mn kesiCom NiniceusBa ELEMENT Fig. 9. Histograms of amounts (wt%, normalized to 100%) of elements detected in paint samples. A. Black paint from the ‘quiver’ of the central figure on the “Coldstream Stone’. B. Black paint from the smallest of the painted stones from the Coldstream shelter. AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 205 = za Lu O oc Lu Oo. = < © Ww = Na Mg Al Si P S Cl K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT 70 60 50 | ome Zz Lu O 40 oc wu o. = 30 © Ww = 20 10 Na Mg Al Si “P= S Cl K Ca Ti Mn Fe Co Ni Cu Ba ELEMENT Fig. 10. Histograms of amounts (wt %, normalized to 100%) of elements detected in paint and pigment samples. A. Black paint from a rock painting in the Cedarberg. B. Burnt mammal bone. 206 ANNALS OF THE SOUTH AFRICAN MUSEUM burnt mammal bone, also from an excavated site in the southern Cape. The results of these analyses are given in Figures 9-10. The large amount of calcium in the paint from the ‘Coldstream Stone’ (Fig. 9A) and from the smallest of the painted stones (Fig. 9B) was surprising, especially when these were compared with the paint from the Cedarberg rock painting (Fig. 10A), which consists mostly of manganese (Mn), a substance often found in nodular form in archaeological deposits, and common in the sandstones of the Table Mountain Group. Since it was known that carbon, an organic element that is used to prepare samples for electron microscopy, would not register in the EDX analyses another source of the black pigment was sought: burnt bone. Although the result of this analysis (Fig. 10D) shows differ- ences from the paint on the two stones from the Coldstream shelter in the presence or absence of a number of minor elements, and differences in the fre- quency of others, the burnt bone sample is more like the two Coldstream paint samples than any of the three are like the Cedarberg paint sample. Although these analyses have not demonstrated conclusively that the black paint was made from burnt bone, they do show that the pigment used was not derived from a manganese-based compound. DISCUSSION As mentioned in the introduction to this section, EDX analysis is at present the only feasible method of analysing very small samples of paint from rock paintings. The method has its limitations in that it cannot identify all the con- stituents of the paints and, because of their lack of homogeneity, cannot identify them precisely. However, given these limitations, the analyses have shown that it is possible to characterize paints and raw pigments to a degree that enables it to be deter- mined whether any two samples are alike or not. Moreover, as was shown in the cases of the white and black paints, it is possible to identify possible, even prob- able, sources of the pigments from which the paints were made. The analyses have therefore proved useful in the investigation of the ‘Coldstream Stone’. THE STYLE OF THE PAINTINGS The figures on the ‘dancing women’ stone (Fig. 1D) are now only barely visible, but examination of the coloured lithograph in the Museum’s possession (see Péringuey 1911, fig. 199; Burkitt 1928, fig. 20, etc.) shows that, in general, they differ from the figure on the smallest stone only in having ‘hook-heads’ and in being more robust. This latter aspect may be a consequence of the different sizes of the paintings, though this is not true of the larger ‘Coldstream Stone’. In general, however, figures of women in rock paintings are more amply propor- tioned than those of men (W.J.J.v.R. pers. obs.). The ‘hook-heads’ of the ‘dancing women’ provide a stylistic link with the figures on the ‘Coldstream Stone’: were the men’s white faces to disappear, they AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 207 would be left with ‘hook-heads’. That apart, the stylistic differences between the paintings on the three stones from the Coldstream shelter are marked. The ‘Coldstream Stone’ appears to have the most elaborate style of the three, but it should be borne in mind that the painting of the ‘dancing women’ is now only a faint vestige of what it was originally. The style of the painting on the two smaller stones is closer to that of the figures in the mural paintings in Whitcher’s Cave and the small shelter above it (Woodhouse 1966) than to that of the ‘Coldstream Stone’. However, the Whitcher’s Cave paintings are in red only and do not have ‘hook-heads’ indica- tive of the loss of the white paint used for the faces on the ‘Coldstream Stone’ and, probably, the ‘dancing women’ stone. Whitcher’s Cave is in the southern foothills of the Tsitsikamma Mountains, some 10 km north-east of the Cold- stream shelter (Fig. 2), and is the closest known site with mural rock paintings. Relatively little has been published about the rock art of the southern Cape, but L. Abel, who has studied the art of the area and has seen an illustration of the ‘Coldstream Stone’, is of the opinion (pers. comm. 1985) that the style of the painting on the stone is similar to that of mural rock paintings in the Langkloof and in the Willowmore district, which are to the north of the Tsitsikamma range. Rudner & Rudner (1970: 126) found the figures on the stone reminiscent of some of those in the mural paintings on Kriedouwkrantz Hill, which is in the Clanwilliam district of the western Cape Province. Since there is no information as to the stratigraphic provenance of the painted stones from the Coldstream shelter, even their relative chronology cannot be established. The most that can be said, then, is that their styles fall within the range of painting styles of the rock art of the area. The last figure on the ‘Coldstream Stone’ differs from the other two figures in some respects. In these, the body and limbs were sketched in white before the red paint was applied, and the lower part of the bodies as well as the front of the legs were outlined in black. In the last figure, only the front leg and the lower part of the body were first sketched in white and the black outlining was extended to the back of the trunk and to the back of the rear leg. Parts of this figure, and particularly the rear leg, have the appearance of having been retouched, a suggestion that is reinforced by the differences indicated above in the element composition of the paint on the two legs. DISCUSSION AND CONCLUSIONS Discussing the ‘Coldstream Stone’, Lewis-Williams (1984: 241) commented ‘The extraordinarily good preservation of the painting on this stone, coming as it does from a rather damp cave deposit, has led some workers to question its authenticity. Two possibilities exist: Either the stone is a complete forgery, or it has been considerably touched up. The second of these possibilities may be correct because it seems unlikely a forger in 1911, when the stone was found, 208 ANNALS OF THE SOUTH AFRICAN MUSEUM would have known about some of the details in the painting and because, if it had been touched up, the workers of that time may not have considered it necessary to record their restoration’. The details Lewis-Williams referred to are those such as the red lines across the faces, which he interpreted as stylized depictions of nasal haemorrhage induced during trance performance, rather than the purely technical details such as the style and manner in which the painting was executed. The possibility that the painting was touched up subsequent to its discovery seems unlikely. The present condition of the paint is not uniform, most of the black, in particular, having all but disappeared. If any touching up was done, it seems most likely to have been of the rear leg and lower part of the body of the last figure; but if this was so, it was a clumsy attempt at restoration to have out- lined the back of the trunk and leg in black when the back of the front leg is outlined in white, as are the legs of the other figures. Moreover, had it been the practice to restore such paintings, this would surely have been done to the ‘dancing women’ stone which, as mentioned above, was in a very poor condition as early as 1912. At some time, attempts were made to preserve some of the Museum’s other painted stones, notably those from the caves on the Robberg Peninsula at Plettenberg Bay (SAM-—AA2822 to AA2828), by covering them with clear varnish. This has now discoloured to the extent that some of the paintings are almost obliterated, and it is fortunate that the painted stones from Coldstream, whatever their present condition, were not subjected to this treatment. Discussing the painted ‘gravestones’ of the southern Cape, Battiss (1948: 93) commented ‘It is difficult to examine the paintings on these gravestones for SO many are in a very poor State of preservation. In a letter to me Mr A. J. H. Goodwin writes: ‘““The many gravestones from Robberg have almost been destroyed. This was owing to the way in which the paint was forced out of the stone by the efflorescence of salt crystals and so formed a fine impalpable powder— just like a pastel”’’. The Robberg Peninsula is some 35 km west of the Coldstream shelter and in the same general environment, although the Cold- stream shelter is higher above sea-level than some of the Robberg caves from which the painted stones were recovered, and better protected by vegetation from wind-blown salt spray. The shielding vegetation does, however, enhance the dampness of the shelter’s deposits (M.L.W. pers. obs.). There is no indication on any of the three painted stones from Coldstream of any crystalline efflorescence, nor any indication of the paint having been forced out of the stone by such action. On the contrary, most of the paint is retained in the pores and other surface irregularities on the stones. The same is true of the painted stones from Klasies River Mouth Cave 5 (Fig. 1B, E; Singer & Wymer 1982, figs 48—49) as well as the other painted stones in the Museum’s collections. The painted stone from a rock-shelter in the Caledon River Kloof, on the farm Danielskraal in the Calitzdorp district (Fig. 1G), has a calcareous accretion on its sides and base, but it is the only one of the painted stones in the AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ 209 Museum’s collections that shows any such damage, perhaps excluding those from Robberg already mentioned, which were subjected to preservative treat- ment. The Danielskraal painted stone was one of two found on the floor of the shelter, the deposits having been washed away (J. Rudner 1971: 57). The differential preservation of the painting on the three painted stones from the Coldstream shelter may be explained partly by differences in the nature of the pigments and/or binders used, partly by the nature of the rock surface on to which they were painted, and partly by differences in the way in which the artefacts were handled during and after excavation. The nature of the binders clearly has some bearing on the way in which the paint adheres to its stone base. In the absence of any clear information in this regard, however, and particularly with regard to the painted stones from the Coldstream shelter, it is not possible to enter into any discussion on this subject. It will be seen that, in the examples illustrated in Figure 1, red is the colour that is generally best preserved, followed by white, and with black least well pre- served, although in the case of the ‘dancing women’ stone the white has completely disappeared, leaving only ‘hook-heads’. In the case of mural rock paintings, the order in which paints are best preserved is generally red, yellow, black and white (W.J.J.v.R. pers. obs.; see also Yates et al. 1985: 70). This makes it the more surprising that white is better preserved than black on the ‘Coldstream Stone’. This may have to do with the use of different types of pig- ments in different areas. The ‘dancing women’ are painted on a lower grindstone, the surface of which is accordingly smoother than that of the ‘Coldstream Stone’, although this is a small water-worn boulder. The smallest of the three stones is an irregular piece of cave rock, quartzite like the others, but with a more uneven surface. The different surface textures almost certainly play a part in the retention of the paint on its base. From the information Henkel gave Péringuey, the ‘dancing women’ stone may have been the first found by Whitcher who, being unaware of the existence of such artefacts, may not have handled it as carefully as he did the other two painted stones, although the third was left exposed to the weather. Drury began his excavation in the knowledge that painted stones might be found, and he would accordingly have treated any stones he found very carefully, as his field notes in fact reveal he did when the ‘Coldstream Stone’ was found. Figure 3 shows that the stones and other material from his excavation were laid on sacking. One of the aspects of Drury’s field notes that seemed suspicious was his mention, the day before making his great discovery, of finding red, yellow and black pigments in the deposits ‘but so far no white’. In this, he seems almost to have anticipated the discovery of a polychrome painted stone. It must be stated again, however, that the smallest of the three painted stones obtained from Whitcher is a black and white bichrome, and was already in the Museum’s pos- session. As discussed earlier, Drury had no reason to suspect the existence of 210 ANNALS OF THE SOUTH AFRICAN MUSEUM polychrome painted stones, the only three painted stones known before his dis- covery being monochromes or bichromes. Moreover, yellow ochre, one of the pigments he mentioned, appears not to have been used in the painting. He would, on the other hand, have known that polychrome mural paintings existed, as indicated by the comment in his field notes given above, that this was ‘the finest Painting I have so far seen anywhere’. The Museum’s earliest other acqui- sition of polychrome paintings, two pieces of a large slab from a mural painting in a cave on the farm Zamenkomst in the Maclear district of the eastern Cape (SAM-AA1617), was only obtained in 1912 (SAM Antiquities Register: 93). If Drury were the perpetrator of a forgery, the logistics of this act, though not insuperable, would have been considerable. The stone on which the painting was done measures 300 x 240 x 90 mm and weighs 8,5 kg. Had Drury prepared the stone in Cape Town, he would have had to take it with him on the train journey to Assegaibosch, and thence by cart on the close on 70 km journey to Coldstream. He stayed with Whitcher at the latter’s home, ‘Oakhurst’, which is about 5 km from the shelter (Fig. 2), and even though Whitcher gave him the use of a cart, its closest access to the shelter was about a mile (1,6 km) from the site (SAM file M—18—D). Drury would thus have had to transport the stone to Coldstream, get it into and out of Whitcher’s house and then to the site, all unnoticed by Whitcher, who probably accompanied him to the shelter on at least the first day. An alternative possibility is that Drury painted the stone at the site. He could have obtained the stone from the site: as mentioned above, he found a number of unpainted stones in the deposit (see Fig. 3, behind Drury); or he could have brought it up from the beach, some 20 m below. He could have used some of the pigments he recorded as having found in the deposit, although this is somewhat negated by his record that it was only after he had found the painted stone that he found two pieces of ‘white clay’ in a niche in the shelter wall behind the burial. In this case, Drury would almost certainly have had to bring a drawing, or drawings, with him, that he could use to copy on to the stone. He employed four labourers, who would have been witnesses to this action, as well as to his ‘salting’ of the deposit, or to his pretence that he had found the stone on the shoulder of the skeleton, as he reported. While neither of these scenarios is beyond the bounds of possibility, both tend to seem far-fetched, and to credit Drury with remarkable deviousness. The strongest argument that can be advanced against Drury’s having done the paint- ings is the evidence provided by the analyses of the paints and pigments. These indicate, with a high degree of probability, that the white paint was made from ground shell and the black from burnt bone. Drury thought that the white material he found at the back of the cave after he had found the painting was clay; and he said in his field notes that charcoal was the black colouring material. It seems highly unlikely that he would have considered that shell or bone could or would have been used as pigments. That there is a close match between the light red ochre from the site and the paint on the front leg of the AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ PAS last figure but not on the back leg, which is the one that might appear to have been retouched, seems to provide further support for the argument that Drury was not involved in either the painting or the retouching. Since the red colour of the first and last figures is visually similar, and different from that of the central figure, it seems more logical to have used the same pigment for both. The retouching could well have been done before the stone was buried. Apart from the fact that Drury made numerous plaster casts of rock engravings, some from as far afield as Namibia, nothing is known of his interest in, or knowledge of, rock art. He joined the Museum in 1902 but, although Péringuey is known to have had a high regard for Drury’s craftsmanship in the fields of taxidermy and life-casting (Summers 1975: 103), the touching-up of the photographic negatives Péringuey used in his monograph was done by A. R. Walker, one of the Museum assistants (Péringuey 1911: 163); and he commissioned a local artist, G. M. Winkles, to prepare the copies of the ‘Cold- stream Stone’ and the ‘dancing women’ stone from which the coloured lithographs were later prepared (Summers 1975: 107). It may be that Péringuey did not consider Drury’s artistic abilities, even as a copyist, to be adequate for such a task. There is, too, no evidence that Drury spent any time in the southern Cape, although it is possible that he may have done so during the Anglo-Boer War, which brought him to South Africa from Scotland (Summers 1975: 103). Although Drury had a reputation for eccentricity and secretiveness (Summers 1975: 128), E. M. Shaw (pers. comm. 1985) and R. H. N. Smithers (1985 in litt.), both of whom knew him for many years, stressed that he was a man of the highest integrity. Smithers also mentioned that Drury was a meticu- lous worker who would probably have taken greater care in the execution of the painting had he done it; and here the anomalous treatment of part of the last figure may be mentioned again. Presumably, the reason for anyone wanting to perpetrate such a forgery would be to acquire, or share in, the prestige attaching to such a spectacular ‘discovery’. In the event, the ‘Coldstream Stone’ did not arouse international interest, and appears to have been ignored, even locally, for nearly 50 years, apart from Haughton’s (1926) publication. Although Péringuey was an entomol- ogist by profession, rather than an archaeologist, there can be no doubt that he would not have allowed the stone to be exhibited if he had had even the slightest suspicion about its authenticity. There is, moreover, no indication that, until recently, the genuineness of this artefact has been doubted by any of those who have seen it, and then only on the grounds of the remarkable state of preser- vation of the painting. The conclusion to be drawn is that, although this investigation has shown some lacunae, if not enigmas, in the documentary evidence, the pigment analyses strongly suggest that Drury was not the author of the painting, nor responsible for the retouching. The remarkable state of preservation of the paint on the stone must be considered a taphonomic problem for which there is at present no ready answer. Ultimately, however, the evidence yielded by this 22 ANNALS OF THE SOUTH AFRICAN MUSEUM investigation suggests that there are no grounds for doubting the authenticity of the painting on the ‘Coldstream Stone’. ACKNOWLEDGEMENTS Thanks for helpful comment on earlier drafts of this paper are due to Dr M. J. Wicomb, Director of the Electron Microscope Unit, University of the Witwatersrand, Johannesburg; Dr D. Crawford, Core Consulting, London; Dr J. Deacon, National Monuments Council, Cape Town; and Dr A. J. B. Humphreys, Department of Anthropology, University of the Western Cape, Bellville. Louis Lawrence, Archaeology Department, South African Museum, produced the maps. The two painted stones from Klasies River Mouth Cave 5 (Figs 1B, E) were first published by the excavators, Ronald Singer and John Wymer (1982, pls 48—49), and thanks are due to Professor Singer for agreeing to allow them to be photographed for use in this paper. Thanks are also due to Dr G. A. Robinson, formerly Warden of the Tsitsikama Coastal National Park, now Deputy Director, National Parks Board, who first took M. L. W. to the Coldstream shelter and other sites, and made available his considerable knowledge of the archaeological sites of the area. REPERENGES Battiss, W. 1948. The artists of the rocks. Pretoria: The Red Fawn Press. Burkitt, M. 1928. South Africa’s past in stone and paint. Cambridge: The University Press. CLARK, J. D. 1959. The prehistory of southern Africa. London: Pelican Books. CoERTZE, R. D. ed. 1988. San rock art. University of Pretoria. Woodhouse exhibition. Pretoria: University of Pretoria. DENNINGER, E. 1971. The use of paper chromatography to determine the age of albuminous binders and its application to rock paintings. Special Publications. South African Associa- tion for the Advancement of Science 2: 80-84. Goopwin, A. J. H. 1929. The Wilton Industry. Jn: Goopwin, A. J. H. & VAN Ret Lowe, C. The Stone Age cultures of South Africa: 251-270. Annals of the South African Museum 27: vii—298. Goopwin, A. J. H. 1946. The loom of prehistory. South African Archaeological Society Hand- book Series 2: 1-151. HaucGuton, S. H. 1926. Note on a burial-stone. Transactions of the Royal Society of South Africa 13 (2): 105-106. Lee, D. N. & WoopuouseE, H. C. 1970. Art on the rocks of southern Africa. Cape Town: Purnell. Lewis-WILuiAMs, J. D. 1984. Ideological continuities in prehistoric southern Africa: the evi- dence of rock art. Jn: SCHRIRE, C. ed. Past and present in hunter gatherer studies: 225-252. Orlando, Fla.: Academic Press. PERINGUEY, L. 1911. The Stone Ages of South Africa as represented in the collection of the South African Museum. Annals of the South African Museum 8 (1): 1-128. Rupne_r, I. 1982. Khoisan pigments and paints and their relation to rock paintings. Annals of the South African Museum 87: 1-281. RupDNeER, J. 1971. Painted burial stones from the Cape. Special Publications. South African Association for the Advancement of Science 2: 54-61. RUDNER, J. & RuDNER, I. 1970. The hunter and his art. A survey of rock art in southern Africa. Cape Town: Struik. SINGER, R. & Wyner, J. 1982. The Middle Stone Age at Klasies River Mouth in South Africa. Chicago & London: Chicago University Press. AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ MANS) SumMERS, R. F. H. 1975. A history of the South African Museum 1825-1975. Cape Town: Balkema. WiLicox, A. R. 1984. The rock art of Africa. London: Croom Helm. WoopHousE, H. C. 1966. Whitcher’s Cave: the paintings. South African Archaeological Bulletin 21 (2): 92-93. WoopHousE, H. C. 1968. The Coldstream burial stone. South African Journal of Science 64 (9): 341-344. WoopHousE, H. C. 1969. The Coldstream burial stone. South African Journal of Science 65 (4): Ws WoopHousE, H. C. 1979. The Bushman art of southern Africa. Cape Town: Purnell. WoopHousE, H. C. 1987. The Coldstream burial stone: a review. South African Journal of Science 83 (9): 538-541. YaTES, R., Gouson, J. & HALL, M. 1985. Trance performance: the rock art of Boontjieskloof and Sevilla. South African Archaeological Bulletin 40 (142): 70-80. 6. SYSTEMATIC papers must conform to the Jnternational code of zoological nomenclature (particu- larly 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. nov., sp. nov., comb. nov., syn. nov., ete: 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 trans- ferred 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 (fig. 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 specimens 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, descrip- tion 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 Eliza- beth (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. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text eo... . the Pisure depictins 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 Punctuation should be loose, omitting all not strictly necessary Reference to the author should preferably 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. The generic name should not be abbreviated at the beginning of a sentence or paragraph. Name 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. M. L. WILSON, W. J. J. van RIJSSEN | & ‘D. A. GERNEKE AN INVESTIGATION OF THE ‘COLDSTREAM STONE’ eo vy ari Ae > Rai i wn cinta . - . _— yar Te a a PM ma A eat SPM ee ang a Z OW Arie Vist ay rs ill [ 3 9088 01206 6973 SMITHSONIAN INSTITUTION LIBRARIES | A Nes as eae Sty ay EOE Rg eei WK Ae ae ssa AVA Nein ee tit TUF orth nae AW eg, ENA n oh UN ima N ri wee yan gyi BAIS RDP ey Th a eas uray poi, he ras RA ahE Ao wy Net bina Sy en ; ; : SPUN RAR 8 WPS Gy che ns Kifer pita : “ BA hep etn + ats TANEOS Rarely Hearne ‘ eV Meee, SPP weg VOW eae NT ab My aA ton PE we Pea, eae he eal SW Uy “ ISB Ne ee ny OO Soe, tee ete, Yess Ve yp Pb wis es 4 HERA ian Peart coe 6 ith Fal pe, Ain OF Maa ef Be Tee Sere A a PEN PN IAD ain Py N Pada ied weet " “ASA oS . Nee eh rin RtRisencuye rhs ROM paw ree a a ern of SO Mays oy SORT Garey w BOE AND AoW Rn Sp op ihe py ty AOS wast mg LPN AeA ew et cel Le LOOT Nitro epee gen, ANAT AVE A: oainhass: wat Rim) ak bay gy dips oa M1 WARN NIRV cy ' se : Aru : . wae ees oo yore: NY Saas te fe a to at Cr Gr rl te vibti SONIA Ae alan nahn 2 a Ui girnindyty pe a rite \ \ 90K Vd ainorinat PANES 8h eKs elt Halen i banat SVE bP yan Vy 2 TENOR AP oy Idella Ht Nam Pee Hype meas SW Jum POY Mpneny ee By Pf swnote shah wen es! PS wicy'bond BA rere (suse, By, . Ono eeignt Se Ol To Aa ON Ry Fw Mit) wale Fee NEWT dn, inert i dint om oy tye th may, : t ; SP Gy 5 5 - Cian ak lo alee Ae Rn ems i Ney SPN iP a eBoy : INS A vaeoye dia ip eS se stow San anys, SL Sry TAC SY A Ep ay, PUNY Fs, . SPA a Apa we EA vey Kinder) CERES ele re! Mig casi Vb Seance de aoe ie ait Be APD Nee HE, ot Ng We Payee Nt ane MRR yet Dire swine! a eri eannnn vinienigp vieheahaee Fog ss Berean NN ey ee EAN pre) «FIM atent : AON Paw aay ch . Area ah Madr my dar ieee pa dachy olen Fy ARORA oy Se shy F , 1 ie he wwe, f miywrnws ¥ WU ae any ong a plvay a an : ule anata eee RAE Ny tpn ey Nye en | Ad Oe ye ORT P spor Wey He KE, be Said MMR HPAL Siviw PMG ag TOP Teg ei we oS) GrAPAbape, wWegee We Age ieee hk t Noe ee Tae ee ee be D aon 4 NMRA Ae SOS hee BN ean Bele pew A hisg Kong “DAMON wet Vy peta Se ee doped ye ah tea rong Pe ene | FVM AVE yb WP vintsingey CUPS APNE, OA FR Are CALA ee WW POE bie See ee p “ths EA NWA BUNA Pa ahi Mg y “Lie AF AWE ei Ve ¥ ib ST Atop Ap eneyp “wr Saari at Tee HO piety. Bupha ety ol OP carey PRES SPIN, Spe kv rah det an sat Sue dy Lalit Weg Se ty CENT Ae TSO tee he aye Pm Cay ny 9h m4 HE em eye y iS SO Tis Cay) COUP ay i oeh aah ok cal a DiC aa A "9 tg Chhe Cae er Pen OWE 0 ue TR SWAN Waits OUR gal cimy We Ge ls PR m8 ow Ratha AGN me NE Dace aiid a nr Prey Lo OTA ANS OFS AFD mre Sey BAR wre P Oyen) WEA Anya weeny, 26 9 Ree Ay SPEC oara) Sate ot FLEA ATR hon ander ll i POA ean TPT CARE re veren MSPS fin qa Cad WOR ay WARHOL Crp Guys Hw SPS COP GN Syke py OO ds Calaera se =P Wry te ANF OS ee em 9 vs rts PANE iE OW A Wak RN ORL CW mL Ad ates AYA, aN lhe Ut OR NC RA ne there Ha ive! giecph DOs Ont Nie lake CRT UN Gey 7 ay CER, Mads hts VVC AE Emmy Uhcaqen Peery 4 Te hry ‘ BRTN oR OLA aren Wii areas UF 8a Cetuee etek Teer uPys ae ti Vr Wey WARE A SEE Moye heh FEU py PRT yn aia wre eve pete RUE ED viene wv PURO manrce wm WAP aba, , may wis exe eurtien rid Wir WU VL eye, Vives eer PE PPLE Wy ie ee Re MAA Lt tt WINDS bein Uru stny FEY Bay NAN RIOT ge very ey UOT AB Siyey ole TMA AEE LET FEU Ndr ay pyre COAL wey ae 20 EDS WAAR up UA} Hf bat een PAPAS ener ay Dever Pele et tans ie baba Veh Wali’ Vane Mala aan re r Puen yi, ue Nit oadi wer Davia! Wea ey y Yh wiy GNP ey Urea val ae Skt, AN? Fi wy TO a i Me tinh AD ngethatiag te y a at eat heh oN Sry ‘ VEN a pe Wales an) md } ar Sty we Me Wy bey Ws Ae \ Vane bts “' PNT Nate. ee WARD : ae 5 senate ty EO OP Ab tay PD CHVEN Ry KEne TA TAIN G ene aa Ve pe », WA OL, SPUN ee AiG My aL! *e) yi alee Asap GON Ay y wh PE Vrly AAS Bron] we TA OAR HW ALOPANEN ian gee a aie WEAR woe Set dul Wiad Wh iA palates pay IESE og oars pay Niner May dew Chale see NENW 98 Uh 89H WAY Ne GM Tees al FAVE NAS ety te p DADRA ES any Bits TURES is eWay an vipa AD RS ts 8 t bid Sa Na dell iw ROTTS UPNENTS Wael tare a WAM he By Up AY erie Ry Dee i A phe bye RE ed oe