LIBRARIES SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHillNS SBiaVdaiT LIBRARIES SMITHSONIAN INSTITUTION NOIlfUliSNI LIBRARIES SMITHSONIAN CO INSTITUTION NOlinillSNI NVINOSHil/JS SBIBVaail LIBRAR CO — , CO c/, ^ O o NOlinillSNI “"NViNOSHims saiavaan’^LiBRAR i es^smithsonian"* institution ^Noiiniu fi/ — U) ' — CO _ to LIBRARIES SMITHSONIAN INSTITUTION NOliniilSNI NVINOSHIIWS S3IBVaan LI > t!2 z Y CO 2 ^ ^ ■%%! * 3> \n;Ain>^ ^ V n^V j, - ^ Xi^6s«ii>Z > 2 CO 2 CO 2 to NoiiniiisNi NviNosHiiiNS S3iavaan libraries smithsonian institution Noiiniij > (I m - m ^ X.lv:^ m IS0NIAN~INSTITUTI0N‘^N0liniliSNrNVIN0SHims'^S3 I dVH a IT ^Ll B R A R I E S^SMITHSONIAN ^INS ^ z V ^ ^ i .. ^ ^ S < X if) o >' s >■ ^ ,^. 2 CO )SHll!Ns‘^S3iavyan^LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI _ NVINOSHimS^S 3 CO X ^ z 7 Q Xj\osv^ HSONIAN'^INSTITUTION"^NO!iniliSNl“'NVINOSHilWS^ S3 I 8 VH a ll_ LI B RAR 1 ES^ SMITHSONIAN^ I NS o )SHiiiNS^S3 1 avaa n u b rar i es^smithsonian institution NOiiniiiSNi NViNOSHims^sa ^S0NIAN^INSTITUTI0N^N0liniliSNI^NVIN0SHil^\s‘^S3 IdVaanfuBRARIES SMITHSONIAN _ IN' CO — 'r O 5 ™ o ''^CSiLDcJZ " O DSHillNS^S3 iyvyan~'LIBRARI ES^SMITHSONIAN"" INSTITUTION NOliniliSNI NVINOSHillAIS^S 3 Z |— Z z ^ _ „ o o i: x-^ovox o ^ 03 VO > IpI 73 ^ cn ^ ^ ^ HSONIAN~INSTITUTION‘^NOIiniliSNl“NVINOSHilWS S3 I 8 V8 a IT L I B RAR I ES SMITHSONIAN INI ^ i ^jr < ’ s i 5: XSa _ JmW/, — ^ — . /L.<"S*^QA :r 0SHimS^S3 I 8V8 a n^LI B rar I ES^SMITHSONIAN institution NOIiniliSNI_NVINOSHill^JS s. — CO X ^ ^ 2 \ 2 LU if) VOLUME 14 • PART 3 Palaeontology AUGUST 1971 PUBLISHED BY THE PALAEONTOLOGICAL ASSOCIATION LONDON THE PALAEONTOLOGICAL ASSOCIATION The Association was founded in 1957 to further the study of palaeontology. It holds meetings and demonstrations, and publishes the quarterly journal Palaeontology and Special Papers in Palaeontology. Membership is open to individuals, institutions, libraries, etc., on payment of the appropriate subscription: Institute membership £10*00 (U.S. $26.00) Ordinary membership ..... £5*00 (U.S. $13.00) Student membership .... £3*00 (U.S. $8.00) There is no admission fee. Institute membership is only available by direct appli- cation, not through agents. Student members are persons receiving full-time instruc- tion at educational institutions recognized by the Council; on first applying for membership, they should obtain an application form from the Membership Treasurer. All subscriptions are due each January, and should be sent to the Membership Treasurer, Dr. A. J. Lloyd, Department of Geology, University College, Gower Street, London, W.C. 1, England. COUNCIL 1971-2 President: Dr. W. S. McKerrow, Department of Geology, Oxford Vice-Presidents: Professor M. R. House, The University, Kingston upon Hull, Yorkshire Dr. Gwyn Thomas, Department of Geology, Imperial College, London, W.C. 7 Treasurer: Dr. J. M. Hancock, Department of Geology, King’s College, London, W.C. 2 Membership Treasurer: Dr. A. J. Lloyd, Department of Geology, University College, Gower Street, London, W.C. 1 Secretary: Dr. W. D. I. Rolfe, Hunterian Museum, The University, Glasgow, W. 2 Editors Mr. N. F. Hughes, Sedgwick Museum, Cambridge Dr. Isles Strachan, Department of Geology, The University, Birmingham 15 Dr. R. Goldring, Department of Geology, The University, Reading, Berks. Dr. J. D. Hudson, Department of Geology, The University, Leicester Dr. D. J. Gobbett, Sedgwick Museum, Cambridge Other members of Council Dr. E. N. K. Clarkson, Edinburgh Dr. L. R. M. Cocks, London Dr. R. H. Cummings, Abergele Dr. W. J. Kennedy, Oxford Mr. M. Mitchell, Leeds Dr. Marjorie D. Muir, London Dr. B. Owens, Leeds Dr. A. D. Wright, Belfast Dr. W. H. C. Ramsbottom, Leeds Dr. Pamela L. Robinson, London Dr. E. P. F. Rose, London Dr. C. T. ScRUTTON, Newcastle Dr. V. G. Walmsley, Swansea Dr. Julia Hubbard, London {co-opted) Overseas Representatives Australia : Professor Dorothy Hill, Department of Geology, University of Queensland, Brisbane Canada: Dr. D. J. McLaren, Institute of Sedimentary and Petroleum Geology, 3303- 33rd Street NW., Calgary, Alberta India: Professor M. R. Sahni, 98 The Mall, Lucknow (U.P.), India New Zealand: Dr. C. A. Fleming, New Zealand Geological Survey, P.O. Box 30368, Lower Hutt West Indies and Central America: Mr. John B. Saunders, Geological Laboratory, Texaco Trinidad, Inc., Pointe-a-Pierre, Trinidad, West Indies Western U.S.A.: Professor J. Wyatt Durham, Department of Paleontology, Univer- sity of California, Berkeley 4, California Eastern U.S. A. : Professor J. W. Wells, Department of Geology, Cornell University, Ithaca, New York © The Palaeontological Association, 1971 LOWER DEVONIAN CORALS AND BRYOZOA FROM THE LICK HOLE FORMATION OF NEW SOUTH WALES by A. E. H. PEDDER Abstract. Corals and a bryozoan are described from the middle part of the Lick Hole Formation. Reasons are given for supposing that the fauna is intermediate in age between the Zelolasma gemmiforme and Chalcido- phylluin recession faunas of the Wee Jasper and Taemas areas of New South Wales. The new assemblage, which is referred to as the Tropidophyllum hillae fauna, is of Praguian age and may correlate approximately with the Bell Point fauna at Waratah Bay, Victoria. Tropidophyllum hillae gen. et sp. nov., Chalcidophyllum discorde giandarrense subsp. nov. and Heterotrypa rapinae sp. nov. are erected. This article describes corals and bryozoa from the Lick Hole Formation of the Devo- nian inlier at Ravine, near Kiandra in the Snowy Mountains of southern New South Wales. Although limited in species the fauna evokes interest in that it appears to corre- late with some part of the poorly fossiliferous interval represented by the Majurgong Formation in the standard Lower Devonian section of southern New South Wales at Wee Jasper. It is also important in that, in conjunction with previous work, it sheds further light on the possible range of Polygnathus Unguiformis dehiscens in Australia, as well as on the problem of the precise age of the Bell Point Limestone of Victoria. Previous work. The first account of the Lick Hole Formation was given by Andrews (1901) in his work on the geology of the Kiandra Goldfield. Andrews referred to it in- formally as the Lobb’s Hole Limestone and surmised, incorrectly as is now clear, that the Lick Hole and nearby Yarrangobilly Limestones are lenticles of similar age in a questionably Upper Silurian sequence. Harper briefly described the formation in 1913 and a few years later Came and Jones (1919, p. 380) published an assay of a limestone sample from it. The formational name Lick Hole and the first regional maps covering the outcrop area stem from geological surveys, begun in 1949, for the Snowy River Hydro-Electric Scheme (Adamson 1957; Adamson et al. 1966; Moye, Sharp, and Stapledon 1969u, b). The most detailed mapping and biostratigraphic studies of the Lick Hole Formation available at the present are those of Flood and the writer (Flood 1969). W. S. Dun provided faunal lists for Andrews (1901, p. 16; Dun 1902, p. 175) and Harper (1913, p. 179); the first were said to purport an Upper Silurian age, the other an undoubted Devonian age. R. Etheridge also identified forms for Andrews (1901, p. 16). Eossils identified by H. O. Fletcher for Adamson (1957, p. 15) were said to indi- cate a lower Middle Devonian age. The widely accepted correlation (David and Browne 1950, p. 233; Browne 1959, p. 126) between the Lick Hole Formation and Murrum- bidgee Group was first put forward in print by Benson (1922, p. 94). The most signifi- cant palaeontological works to date have been the description of cricoconarids by Sherrard (1967), in which the true Lower Devonian age of the Lick Hole Formation was [Palaeontology, Vol. 14, Part 3, 1971, pp. 371-86, pis. 67-68.] C 8216 c C 372 PALAEONTOLOGY VOLUME 14 TEXT-FIG. 1. Map of New South Wales and Victoria showing localities mentioned in the text. first indicated, and figures of conodonts, principally in Flood (1969), but also in Pedder, Jackson, and Philip (1970, pi. 40, figs. 15, 16, 19, 21, 23), which refine the correlation between the Ravine and Wee Jasper/Taemas sequences. STRATIGRAPHY The thickness of the Lick Flole Formation was given as 1830 ft (558 m) by Adamson (1957, p. 8) and as 1400 ft (420 m) by Moye, Sharp, and Stapledon {\969b, p. 145). The 1600 ft (488 m) figure given below was obtained by P. G. Flood and the writer using compass and tape before the earlier measurements were available to us. The section begins on the Kiandra/Ravine road at Grid Reference point 29421760 and continues along the road for the lower 1300 ft; the upper 300 ft were measured up the PEDDER: LOWER DEVONIAN CORALS AND BRYOZOA 373 slope from the road to Round Top Trig. Station at Grid Reference point 29461753 (see also Flood 1969, fig. 1). Unit Thickness in feet and metres no. Unit Total from base 5 4 3 2 1 Mudstone, olive-grey, poorly bedded, poorly exposed; no fossils seen. Contact with overlying Round Top Formation distinct and probably conformable 300 (91-5) Mudstone, olive-grey, with dark micro-crystalline cal- careous nodules; rare brachiopods 120 (36-6) Mudstone, calcareous and partly slightly silty, bluish- grey, with numerous biogenic nodular layers; upper few feet ledge-forming; Tropidophyllum hillae, Clialcido- phyllum discorde giandarrense, Heterotrypa rapinae, brachiopods (currently under study by P. G. Flood), Tentacidites chapmani, bivalves, ostracods, etc. 360 (109-8) Mudstone, olive-grey, some thin biogenic nodular lime- stone interbeds; brachiopods and some bivalves present 522 (159-1) Mudstone, bluish-grey and siltstone, calcareous, dark grey; few brachiopods present. Contact with underlying Milk Shanty Formation distinct and apparently con- formable 298 (90-8) Total thickness of Lick Hole Formation 1600 ft (487-8 m) 1300-1600 (396-3-487-8) 1180-1300 (359-7-396-3) 820-1180 (249-9-359-7) 298-820 (90-8-249-9) 0-298 (0-90-8) CORRELATION AND SIGNIFICANCE OF THE LICK HOLE FAUNA The best Lower Devonian reference sections in south-east Australia are those around Wee Jasper and Taemas on the flanks of the Narrangullen Anticline, some 50 miles (80 km) north-north-east of the Ravine Inlier. The oldest fauna known in these sections is the Zelolasma gemmifonne assemblage of the Cavan Limestone, which is separated from higher faunas by the generally poorly fossiliferous Majurgong Formation at Wee Jasper (Tedder, Jackson, and Philip 1970), and by the Majurgong and lower beds of the Taemas Limestone at Taemas (Browne 1959). The stratigraphically important conodont element Polygnathus linguifonnis dehiscens is common to both the Zelolasma gemmiforme and Lick Hole faunas (Flood 1969). The corals of the two assemblages, however, are entirely different. This may be due to changes of facies, or to faunal provincialism. But the facies are not disparate, nor are the faunas separated by any great geographical distance. This leads one to suspect that the Lick Hole fauna may be either slightly younger or slightly older than the Z. gemmi- forme fauna. The corals suggest that it is younger, because the widely distributed genus Chalcidophyllum is unknown elsewhere in beds as old as the Z. gerntniforme fauna and a species closely related to Tropidophyllum hillae occurs in the post Z. gemmiforme Bindi Limestone of Victoria (Philip and Tedder 1968). As the fauna of unit 3 of the Lick Hole Formation appears to be intermediate in age between the gemmiforme and recessum assemblages of Philip and Tedder (1967, 1968) it is now formally designated as the Tropidophyllum hillae assemblage. In view of the Praguian age of the Z. gemmiforme and Chalcidophyllum recessum faunas (Tedder, Jackson, and Philip 1970, p. 212), the Tropidophyllum hillae assemblage should also be referred to the Praguian Stage. Two significant points emerge from the composition and assumed correlation of the 374 PALAEONTOLOGY VOLUME 14 new fauna. First, an approximate correlation is implied, by the identification of the Lick Hole Chakidophyllum as a subspecies of C. discorde, between the Lick Hole and Bell Point Limestone of Waratah Bay, Victoria. Secondly, the known range of Polygnathus lingiufonnis dehiscens in Australia is extended upwards into beds equivalent to some part of the Majurgong Formation at Wee Jasper. If correct this would reduce considerably the gap between the previously known upper limit of P. linguifonnis dehiscens and the lower limit of P. linguifonnis foveolatus in Australia (Philip and Jackson 1967, text-fig. 1). Such a gap is not known in the Carnic Alps where both subspecies have now been recognized (Skala 1969). SYSTEMATIC PALAEONTOLOGY Unless otherwise stated all specimens were collected by the writer in January 1967 from the Kiandra/Ravine road section described above. Depositories of fossils are abbreviated as follows : AM : Australian Museum, Sydney, New South Wales. GSNSW: Geological Survey of New South Wales, Sydney. UNE: Department of Geology, University of New England, Armidale, New South Wales. Phylum COELENTERATA Order rugosa Milne-Edwards and Haime 1850 Family cyathophyllidae Dana 1846 Genus tropidophyllum nov. Type species. Tropidophyllum hillae sp. nov. Diagnosis. Tetracoral, solitary except for rare buds that are not known to have developed beyond an early stage, or to have produced a genuinely fasciculate colony. Calicular platform absent, or if present narrow and flat to weakly excert. Outer wall with broad subdued interseptal ridges ; either invaginated and continuous with septal bases, or em- bedding them. Septa radially arranged, faintly to highly carinate; adaxially dilated and spinose; apart from scattered axial lobes they are markedly withdrawn from the axis. Exceptionally strongly carinate septa may be slightly retiform. Septal trabeculae mona- canthine with widely divergent fibres, inwardly projecting and subparallel to, less commonly divergent within the plane of the septum. Septa contiguous or united by sclerenchyme in the adaxially dilated region, usually forming an inner wall. Dissepiments in a few or several rows, generally inwardly sloping, although locally the outermost are flat-lying and may be rhomboid. Tabulae mostly broad to complete, and commonly, together with peripheral tabellae, form flat to concave tabularial surfaces which are characteristically periodically invested with sclerenchyme. Derivation of name. Greek, tropos = keel (carina) and phyllon = leaf. Remarks. The genus is also represented by an undescribed species in the Bindi Limestone (Praguian) of Victoria and possibly by calicular moulds figured by Talent as Rugosa indet. E (1963, pi. 12, figs. 1-7) from the Kilgower Member at Tabberabbera, Victoria. The trabeculae, variation in the relationship of the septal ends to the wall, and the adaxially expanded septal ends relate the new genus to Zelolasma (see Pedder, Jackson, and Philip 1970, p. 232). The two genera are distinguished by growth habit, which is colonial in Zelolasma and solitary in Tropidophyllum. PEDDER: LOWER DEVONIAN CORALS AND BRYOZOA 375 The solitary species Ceratophyllum shandiense Zheltonogova (in Khalfin 1961, p. 403) and Cyathophyllum pannosum Hill and Jell (1969, p. 6) from beds said to be of Eifelian age in the Salair and Queensland respectively, also appear to be related to the new genus. However, the dilation of their septa does not seem to be associated with sclerenchymal investment of the dissepiments and the relationship of wall to septa has not yet been TEXT-FIG. 2. TropidophyUum hillae gen. et sp. nov., all figures enlarged x3: a, transverse section of paratype 11, UNE FI 1675, a specimen with strongly dilated and carinate septa; b, transverse section of paratype 14, AM AM6456c, a large specimen with weakly dilated and weakly carinate septa; c, longitudinal section of paratype 2, UNE FI 1666, a trochoid specimen; d, longitudinal section of paratype 4, UNE FI 1668, a ceratoid specimen. described in these species. Ceratophyllum, interpreted on the group of species close to C. typiis Giirich (Fedorowski 1967), differs from TropidophyUum in being prominently bilaterally symmetrical in early stages and in having a more consistently fan-shaped trabecular arrangement. Cyathophyllum, based on C. dianthus Goldfuss (Birenheide 1963, p. 376), is a colonial genus with fine septa and no thickening of the dissepimen- tarial or tabularial surfaces. Gurievskiella Zheltonogova (see also Hill and Jell 1969, p. 11) is distinguished from 376 PALAEONTOLOGY VOLUME 14 the new genus by its everted dissepimentarium, weakly divergent fibres within the trabeculae and domed tabularium composed mostly of numerous vesicular tabellae. TropidophyUum hillae sp. nov. Plate 67, figs. 1-14; text-figs. 2a-d 1901 Amplexits (?); Dun, W. S., in Andrews, p. 16. 1902 Amplexits (?); Dun, p. 175. 71913 Cyatliophyllimr, Dun, W. S. (‘the Palaeontologist’) in Harper, p. 179. 1922 Amplexits, spp. ind.; Benson (in part), p. 143 (Lobb’s Hole specimen only). 71922 Cyatliopliyllum, sp. indet.; Benson (in part), p. 146 (Lobb's Hole specimen only). 1969 Solitary rugosan; Flood (in part), p. 7. Derivation of name. Patronym in honour of Professor Dorothy Hill. Type series. Holotype and paratypes 1-4, UNE FI 1664 to FI 1668 respectively, 943-970 ft above the base and 630-657 ft below the top of the Lick Hole Formation. Paratypes 5, 6, UNE FI 1669, FI 1670, 823-849 ft above the base and 751-777 ft below the top of the Lick Hole Formation. Paratype 7, UNE FI 1671, 889 ft above the base and 711 ft below the top of the Lick Hole Formation. Paratype 8, UNE FI 1672, 890-897 ft above the base and 703-710 ft below the top of the Lick Hole Formation. Para- type 9, UNE FI 1673, 970-990 ft above the base and 610-630 ft below the top of the Lick Hole For- mation. Paratypes 10, 11, UNE FI 1674, FI 1675, 990-1000 ft above the base and 600-610 ft below the top of the Lick Hole Formation. Paratypes 12-16, AM AM6456a-e (all on one large thin section) and F43438 (remaining hand specimen), unrecorded horizon within the Lick Hole Formation. Paratypes 12-16 were collected by the Geological Survey of New South Wales in 1949; although the museum label reads ‘Middle Devonian, Yarrangobilly, N.S.W. (Lobb's Hole)’ there is no doubt that they come from the Lick Hole Formation. Paratype 17, GSNSW F14710, unrecorded horizon within the Lick Hole Formation. This specimen was collected by E. C. Andrews at about the turn of the century on ‘O’Hare’s Creek, Lobb’s Hole, Kiandra’ and should have come from approximately 1 mile south-west of the type locality; it is the specimen identified in print as Amplexits (?) sp. Twenty-eight thin sections have been cut from the type material. Diagnosis. Corallum trochoid to ceratoid; maximum length approximately 3 cm, diameter 2 cm. Exterior walls confluent with, or embracing the septal bases. Septa smooth to highly carinate, radially arranged, 26x2 to 32x2 in number at maturity. EXPLANATION OF PLATE 67 Figs. 1-14. TropidophyUum hillae gen. et sp. nov.. Lick Hole Formation, Kiandra/Ravine road section, southern New South Wales. 1, holotype, UNE FI 1664, part of longitudinal section showing non- divergent monacanthine trabeculae, X 10. 2, paratype 8, UNE FI 1672, exterior view before section- ing, X 2. 3, paratype 2, UNE FI 1666, part of longitudinal section showing divergent monacanthine trabeculae, x 10. 4, paratype 10, UNE FI 1674, part of transverse section of a markedly carinate specimen, X 10. 5, holotype, UNE FI 1664, part of transverse section showing sclerenchyme and septal bases that are not confluent with the wall, x 10. 6, paratype 1, UNE FI 1665, part of trans- verse section of a specimen having thick septa with bases confluent with the wall, X 10. 7, paratype 3, UNE FI 1667, transverse section showing well-developed inner wall, x3. 8, holotype, UNE El 1664, transverse section near the base of the calice, x 3. 9. holotype, UNE FI 1664, typical longi- tudinal section, x 3. 10, paratype 8, UNE FI 1672, transverse section of a specimen with thin septa, x3. 11, paratype 9, UNE FI 1673, longitudinal section showing outwardly inclined dissepiments and abundant axial lobes, x3. 12, paratype 1, UNE El 1665, transverse section across the calice of a specimen with profuse sclerenchyme, x3. 13, paratype 9, UNE FI 1673, transverse section, note budding and the locally retiform septa, x 3. 14, paratype 11, UNE FI 1675, transverse section of a specimen having thick, strongly carinate septa, x 3. Palaeontology, Vol. 14 PLATE 67 PEDDER, Lower Devonian corals PEDDER: LOWER DEVONIAN CORALS AND BRYOZOA 377 Major septa only slightly longer than the minor septa. Sclerenchyme and/or septal dila- tion form an interior wall in the inner region of the dissepimentarium. Trabeculae mona- canthine, parallel, inclined at 45° peripherally, variably flattened adaxially. Dissepiments variable, in up to 10 rows. Tabulae broad, mostly forming flat or concave tabularial surfaces. Sclerenchyme periodically developed on dissepimentarial and tabularial sur- faces. Description. Corallites are trochoid or ceratoid. When complete the largest were prob- ably a little over 3 cm long and at least 2T cm in diameter; more typically they are 2*0- 2- 5 cm in length measured along the convex side and T2-1-5 cm across the top of the calice. Minor rejuvenations are common. Some topotypes were evidently eroded prior to fossilization ; where exteriors are well preserved there are abundant fine growth ridges and usually pronounced septal grooves and broad, rather flat interseptal ridges. Offsets are known in only one specimen (PI. 67, fig. 13) and these did not develop beyond an early stage. At first sight they appear to have originated by axial gemmation, but closer inspection reveals that the formation of two of them preceded that of the third and that all are probably better regarded as peripheral offsets. The calice, which is about as deep as it is wide, generally has a rather flat, broad base and an extremely narrow peripheral platform. Thickness of the exterior wall ranges from 0-05 to 0-4 mm but is usually from OT to 0-2 mm. In some cases the axial plate and wall are strongly invaginated at the septal bases and pass imperceptibly into them, in others the septal base is distinct and embraced by the wall. Septal arrangement is radial. Despite considerable variation in thickness the septa are invariably relatively dilated in a peritabularial zone and commonly in collaboration with sclerenchyme, but in some cases by contiguity alone, form a distinct inner wall close to the inner margin of the dissepimentarium. Carinae may be totally suppressed, although more commonly zigzag and yardarm carinae are present, and locally are so highly developed that the septa are retiform. The major septa, which are only a little longer than the minor septa, extend between one and two thirds of the distance to the axis. Short axial lobes are commonly present in the periaxial region of the tabularium. Trabeculae are monacanthine with widely divergent fibres, parallel and distally form strongly denticulate septal edges. Peripherally they are directed upwards and inwards at about 45°, adaxially they usually flatten, in places so much that they become nearly, or quite flat. In the following table measurements are in millimetres, Dt is the diameter of the tabularium and Dc of the calice or corallite. Dissepiments are confined to one or two rows in early stages; later there are up to 10, usually 6 or 7 rows, except following a rejuvenation when there are typically 4 or so rows. The dissepiments vary greatly in shape and size. Most are well inflated and moderately inclined inwards, but some in the outermost 1 or 2 rows are highly arched and even outwardly inclined, while others in the inner rows are elongated and almost vertical. Some dissepiments, especially in the more carinate specimens fail to cross the interseptal loculi. A few tabellae may be present. Tabulae are mostly gently sinuous, broad and commonly complete. Where they are closely spaced, 3, or exceptionally 4 may be counted over a vertical distance of 1 mm; the most widely spaced tabulae are 3- 4 mm apart. Sclerenchymal thickening of the dissepimentarial and tabularial surfaces is developed periodically. PALAEONTOLOGY VOLUME 14 378 Specimen Mean or measurable Dt Mean or measurable Dc DtjDc No. of septa Orientation of thin section Paratype 2 1-8 3-9 0-46 — longitudinal Paratype 15 2-5 4-0 0-62 16x2 transverse Paratype 5 3-6 60 0-60 16x2 transverse Holotype 50 6-5 0-77 — longitudinal Paratype 3 50 8-7 0-57 24x2 transverse Paratype 11 5-5 130 0-43 27x2 transverse Paratype 1 60 11-5 0-52 27x2 transverse Holotype 6-4 11-5 0-56 26x2 transverse Paratype 16 — 13-8 — 25 xl transverse (cal ice) Paratype 2 6-8 14-0 0-49 — longitudinal Paratype 8 70 12-6 0-55 27x2 transverse Paratype 4 7-5 14-2 0-53 27x2 transverse Paratype 6 — ■ 14-5 — 29x2 transverse (calice) Paratype 8 7-7 14-2 0-55 28x2 transverse Paratype 9 8-5 130 0-65 — longitudinal Paratype 10 8-8 160 0-55 27x2 transverse Paratype 14 90 15-4 0-58 28x2 transverse Paratype 9 — 190 — 32x2 transverse (calice) Paratype 9 — 21 0 — 32x2 transverse (calice) Genus chalcidophyllum Redder 1965 1965 Chalcidophyllum Pedder, p. 204. Type species. Chalcidophyllum campaueuse Pedder 1965, p. 204. Bell Point Limestone (Praguian), Waratah Bay, Victoria. Diagnosis. Corallum solitary and possibly weakly compound; corallites trochoid to cylindrical. Wall invariably invaginated at the septal bases and continuous with them; septal grooves prominent. Peripheral stereozone developed in some species. Septa nor- mally smooth, usually radially arranged, apparently without trabeculae and typically highly differentiated into two orders. Minor septa may be almost entirely suppressed. Dissepiments numerous in large species, relatively less numerous in others, inosculating and forming a herringbone dissepimentarium in some species ; at the periphery they are small, but adaxially become larger and more elongate. Tabulae broad and commonly depressed axially. Remarks. The following taxa now comprise the genus: Chalcidophyllum campanense Pedder 1965, C. campanense var. nanum Pedder 1965, C. angulare (Hill 1950), C. discorde Pedder 1965, C. discorde giandarrense subsp. nov., C. gigas Pedder 1970, C. recession (Hill 1940), and C. vesper Pedder 1970. The genus is currently known with certainty only from the Emsian and possibly the late Siegenian of eastern Australia. Chalcidophyllum discorde Pedder 1965 71940 Gen. et sp. indet.; Hill, pi. 9, fig. 9. 1954 Mictophyllum cressweUi (Chapman) var. cylindricum Hill, p. 109, pi. 7, figs. 9a, b. 1954 Mictophyllum sp. or Disphyllum sp.; Hill (in part ?), p. 110, pi. 7, fig. 11. 1965 Chalcidophyllum discorde Pedder, pp. 206, 7, pi. 30, figs. 14, 15; pi. 34, fig. 1. Remarks. The holotype and paratype are from the Bell Point Limestone at Bell Point, Waratah Bay, Victoria. Other material tentatively included in the above synonymy is PEDDER: LOWER DEVONIANICORALS AND BRYOZOA 379 d e f TEXT-FIG. 3. Chalciclopliyllum cUscorde giandarrense subsp. nov., all figures enlarged X 3 : a, longitudinal section of paratype 4, UNE FI 1680, a large specimen with wide dissepimentarium; b, transverse section of paratype 4, UNE El 1680, a large specimen with little septal dilation; c, longitudinal section of paratype 1, UNE El 1677, a specimen with peripherally dilated septa invested with sclerenchyme ; d, transverse section of paratype 1, UNE El 1677, a specimen showing pinnate septal arrangement and peripherally dilated septa invested with sclerenchyme; e, transverse section of paratype 2, UNE FI 1678, a specimen with undilated septa; /, longitudinal section of paratype 2, UNE FI 1678, a speci- men with a narrow dissepimentarium. apparently from the Bird Rock Member of the Waratah Limestone at Walkerville, Waratah Bay and the Taemas Limestone at Cavan, near Taemas, New South Wales. Chakidophyllum discorde giandarrense subsp. nov. Plate 68, figs. 1-11 ; text-figs. 3a-f 1969 Solitary rugosan; Flood (in part), p. 7. 380 PALAEONTOLOGY VOLUME 14 Derivation of name. Giandarra, name now corrupted to Kiandra. Type series. Holotype and paratype 1, UNE FI 1676, FI 1677 respectively, 1 168-1180 ft above the base and 420-432 ft below the top of the Lick Hole Formation. Paratype 2, UNE FI 1678, 970-990 ft above the base and 610-630 ft below the top of the Lick Hole Formation. Paratype 3, UNE FI 1679, 990- 1000 ft above the base and 600-610 ft below the top of the Lick Hole Formation. Paratype 4, UNE FI 1680, 1000-1002 ft above the base and 598-600 ft below the top of the Lick Hole Formation. Paratype 5, UNE FI 1681, unit 3 of the Lick Hole Limestone. This specimen was collected by the author in 1964 from the Kiandra/Ravine road section before the section had been measured. Sixteen thin sections have been prepared from the type material. Diagnosis. Corallum solitary, typically ceratoid, up to 5-5 cm in length and 2-8 cm in diameter. Wall invaginated at septal bases. Septa typically 29x2 to 31 x 2 in number, maximum 38x2. Septal dilation and sclerenchymal investment variable, where pro- nounced major septa leave an axial space of about 4 mm diameter, where subdued or absent, the axial space is normally 6-8 mm in diameter. Dissepiments steeply inclined, typically in 2-5 rows, exceptionally 7 or 8 rows. Tabulae broad where septa are short. Description. The solitary corallum is usually ceratoid, rarely subcylindrical, and may grow to a length of at least 5-5 cm and a diameter at the distal end of 2-8 cm; most specimens, however, are a little less than 2 cm across the calice. The calice, which is not freely exposed in the type material, would be deep with steep sides and have a broad flattish base. Well-defined septal grooves and growth ridges mark the exterior of the coral. The wall, which is 0 05-0-22 mm thick, is invariably invaginated at the septal bases and may also be flexed midway between the septa to form a faint interior ridge and EXPLANATION OF PLATE 68 Figs. 1-11. ChalcidophyUiim discorde giandarrense subsp. nov.. Lick Hole Formation, Kiandra/Ravine road section, southern New South Wales. 1, holotype, UNE FI 1676, part of longitudinal section showing fine structure of a septum, X 10. 2, paratype 4, UNE FI 1680, exterior view before section- ing, note the prominent septal grooves, x 2. 3, paratype 4, UNE FI 1680, part of transverse section showing invagination of the wall at the septal bases and fibronormal septal structure, X 10. 4, para- type 1, UNE FI 1677, longitudinal section showing sclerenchymal investment of the dissepimen- tarium, X 3. 5, holotype, UNE FI 1676, longitudinal section showing sclerenchymal investment of the tabularium, x 3. 6, holotype, UNE FI 1676, transverse section showing unequally dUated septa, X 3. 7, paratype 5, UNE FI 1681, longitudinal section showing narrow dissepimentarium in early stages, x3. 8, paratype 2, UNE FI 1678, longitudinal section of a specimen with uniformly thin septa, X 3. 9, paratype 4, UNE FI 1680, transverse section of a specimen with uniformly thin septa, X 3. 10, paratype 1, UNE FI 1677, transverse section through the lower part of the calice, note the weakly pinnate arrangement of the septa, carinae and septal dilation, x3. 11, paratype 2, UNE FI 1678, transverse section of an early stage of a specimen with short undilated septa, X 3. Figs. 12-16. Heterotrypa rapinae sp. nov.. Lick Hole Formation, Kiandra/Ravine road section, southern New South Wales. 12, holotype, UNE FI 1682, longitudinal section of the exozone and the outermost part of the endozone, note acanthopores and moniliform mesopores, approx. x25. 13, paratype 3, UNE FI 1685, randomly orientated section showing rejuvenation on the right, passing laterally into an overgrowth on the left, X 15. 14, paratype 2, UNE FI 1684, tangential section in the exozone showing an area of thickened zoecial walls and atypically high concentration of acanthopores, approx. x25. 15, holotype, UNE FI 1682, tangential section in the outer part of the endozone showing large acanthopores, approx. x25. 16, holotype, UNE FI 1682, tangential section within the exozone, note ‘amalgamate’ structure, approx, x 25. Palaeontology, Vol. 14 PLATE 68 12 14 15 PEDDER, Lower Devonian corals and bryozoa 16 ( i I } I PEDDER: LOWER DEVONIAN CORALS AND BRYOZOA 381 corresponding exterior groove. Septa merge with the wall at the periphery. They are in two orders and are usually radially arranged. The length and dilation of the septa and the thickness and persistence of sclerenchyme on dissepimentarial and tabularial sur- faces vary appreciably from specimen to specimen and also periodically within a single specimen. Where the septa are short there is little or no dilation, or sclerenchyme in- vestment, and the axial space is typically 6-8 mm in diameter. Where the septa are longer and dilated the dissepimentarial surfaces are coated with sclerenchyme, which may or may not be continuous with the dilated parts of the septa, and the axial space is gener- ally about 4 mm in diameter. Long septa may be irregularly sinuous at the axis and the axis may be somewhat excentric. Septa are typically smooth, although coarse carinae may occur in the peripheral parts of strongly dilated septa. The fine structure is fibro- normal, even, as far as can be discerned, in the carinate parts of the septa. In the follow- ing table measurements are in millimetres, Dt is the diameter of the tabularium and Dc is the diameter of either the calice or the corallite. Specimen Mean or measurable Dt Mean or measurable Dc DtjDc No. of septa Orientation Oj thin section Paratype 1 20? 20 — 21 undiflF. transverse Paratype 1 3-3 50 0-66 21 major transverse Paratype 2 50 5-9 0-85 21x2 transverse Paratype 5 5-0 7-5 0-67 — longitudinal Holotype 60 130 0-46 — longitudinal Paratype 3 70 130? — 31x2 transverse Paratype 2 80 100 0-80 — longitudinal Paratype 5 80 110 0-73 25x2 transverse (calice) Paratype 1 80 11-5 0-70 — longitudinal Holotype 90 16-5 0-55 29x2 transverse Paratype 2 9-5 12-5 0-76 29x2 transverse Paratype 4 9-5 160 0-59 29x2 transverse Paratype 1 10-5 13-2 0-79 29x2 transverse Paratype 4 14-5 23-5 0-62 35x2 transverse Paratype 4 — 28 0 — 38x2? transverse (calice) In stages during which the coral enlarges to a diameter of 3 mm dissepiments are absent or rare and confined to one row. Subsequently there are typically 2-5 and excep- tionally as many as 7 or 8 rows of dissepiments. The dissepiments are steeply inclined and mostly of moderate size, although some are larger and markedly more elongated. The tabularium consists of marginal tabellae, in places inosculating, and tabulae. Where the major septa are short the tabulae are commonly broad and in some cases complete. Remarks. The new subspecies dilfers from the nominate species in having a generally narrower dissepimentarium and shorter, more dilated and less regular major septa. Specimens of the new subspecies that have undilated septa may resemble Cbalcido- phyllum recession, but examples of that species normally have fewer septa at given diameters (e.g. 23x2 to 27x2 at 16-0 mm in C. recession', 29x2 to 31 X 2 at 16-0 mm in C. discorde giandarrense). In C. vesper the major septa are consistently long and attenuate in the tabularium and there is a uniformly well-developed peripheral stereo- zone. 382 PALAEONTOLOGY VOLUME 14 Phylum BRYOZOA Order trepostomata Ulrich 1882 Family heterotrypidae Ulrich 1890 Genus heterotrypa Nicholson 1 879 1879 Monticiilipora (Heterotrypa) Nicholson, pp. 291, 93, 94. 1882 Dekayella Ulrich, p. 155. 1966 Heterotrypa Nicholson; Boardman and Utgaard, pp. 1105-1107, pis. 140-142. Remarks. Boardman and Utgaard (1966, p. 1090) recognize two kinds of acanthopores in Ordovician species of Heterotrypa, based primarily on their spatial relationships to the zoaria. A histogram (text-fig. 4) of the frequencies of acanthopore diameters sug- gests that the acanthopores of the Lick Hole species fall into two groups. These, how- ever, do not seem to correspond to the endacanthopores and exacanthopores of the Ordovician species, as the larger ones occur both at and between the zoecial corners and both groups may originate in either the outer endozone or exozone. The same authors (1966, p. 1091) stress that the presence or absence of two kinds of acanthopores is not absolutely diagnostic of the genus. Nevertheless until intermediate Silurian forms are better known concern will remain regarding the generic assignment of the Devonian species. Some Devonian species presently ascribed to Leioclema (Ulrich 1882, pp. 141, 54) resemble the new species, but Owen (1969, p. 631), who has examined the type material of Leioclema, notes that it appears to be related to such genera as Rhombopora and other rhabdomesids which are normally placed in the order Cryptostomata. In view of this Owen erected Asperopora, based on Callopora aspera Hall, for several Silurian and Helderbergian bryozoa that had previously been accommodated in Leioclema. The Lick Hole species resembles Asperopora apart from its fine structure, which is essentially leioclemid (see Boardman 1960, p. 30) and therefore very different from the fine struc- ture of Asperopora aspera depicted by Tavener-Smith (1969, fig. 1). Paralioclema (Morozova 1960, p. 93) is likely related to the new species, but is probably better re- stricted to forms with especially large acanthopores. Such forms are presently only known from the Givetian and Frasnian of the Minusinsk and Kuznetsk Basins of Siberia. The relationship of the new species to Thallostigma (Hall 1883, p. 154; abstract 1881, p. 12), which has usually been regarded as a synonym of Leioclema, cannot be assessed until T. intercellata Hall (1883, p. 154; abstract 1881, p. 13; 1886, pi. 32, figs. 15-20), the type species, has been more fully described. Heterotrypa rapinae sp. nov. Plate 68, figs. 12-16; text-fig. 4 1957 Favosites sp. (A species with very fine corallites); Fletcher, H. O., in Adamson, p. 15. 1969 bryozoan; Flood, p. 7. 19697) Favosites sp.; Moye, Sharp, and Stapledon, p. 145. Derivation of name. The Old French word ravine, which is derived from the Latin rapina meaning robbery. Hence rapinae (genitive) indicating that the types are from the locality known as Ravine. Type series. Holotype, UNE FI 1682, 889 ft above the base and 71 1 ft below the top of the Lick Hole Formation. Paratype 1, UNE FI 1683, 890-897 ft above the base and 703-710 ft below the top of the Lick Hole Formation. Paratypes 2-4, UNE FI 1684-El 1686, 943-970 ft above the base and 630-657 ft 383 PEDDER: LOWER DEVONIAN CORALS AND BRYOZOA below the top of the Lick Hole Formation. Paratypes 5-7, AM F43419-F43421, unrecorded horizon within the Lick Hole Formation. Paratypes 5-7 were collected by the Geological Survey of New South Wales in 1949 and are the specimens identified by Fletcher in Adamson (1957) as Favosites with very fine corallites. The museum label cites their occurrence as ‘Middle Devonian, Yarrangobilly, N.S.W. (Lobb’s Hole)’ but the matrix, associated fauna, and Adamson’s work indicate that they are from the Lick Hole Formation. The type series includes 13 thin sections. 18 r 16 - 14 - > ’2 - o m ’0 - 3 C 8 - UJ QC ^ 6 - 4 - 2 - 0 L- — - — — •02 -03 -04 -05 -06 -07 *08 *09 -10 *11 DIAMETER IN MM TEXT-FIG. 4. Histogram of the frequency of diameters of 60 randomly chosen acanthopores from the holotypes and paratypes 1-5 of Heterotrypa rapinae sp. nov. Ten measurements were taken from each specimen. Note that the majority of acanthopores are either 0-3-0-6mm in diameter, or 0-8 mm in diameter. Description. Zoarium ramose or less commonly subramose. Rejuvenation may occur within the exozone and overgrowths, usually with a well-developed basal plate, may also be present. One slide of UNE FI 1685 shows a rejuvenation passing laterally into an overgrowth. Matrix obscures zoarial surfaces in the type series, but groups of thick- walled zoecia suggest the presence of monticles. In early stages the zoecia closely parallel the zoarial axis; later they curve outwards gently, except in the outermost region of the endozone where there is a marked increase in curvature. Normally the entire exozone is directed perpendicularly to the zoarial sur- face. Diameter of the endozone is usually between 8 and 18 mm and that of the zoarium between 14 and 22 mm. In overgrowths the endozone is abbreviated and for the most part corresponds only to the outer endozone of unrejuvenated zoaria. Inside the boundary between the endozone and exozone the thickness of the walls is approximately 0-007 mm. At the boundary the walls thicken in the normal treposto- matous manner and in certain exozonal areas, which seem to represent monticles, are as much as 1-3 mm thick. Wall laminae are highly arched over the centre of the walls and commonly continue on to the diaphragms of the mesospores. Zoecia and meso- spores, which are polygonal in cross section within the endozone, lose so much of their 384 PALAEONTOLOGY VOLUME 14 angularity in the endozone that their interior may be almost rounded. Zoecial dia- phragms are usually 1-2 mm apart in the endozone and 0-1 5-0-75 mm apart in the exozone. Although some mesopores originate deep in the endozone, in some cases at a previous growing tip, most first become evident in the outer endozone. Contractions at the dia- phragms impart a characteristic moniliform appearance to the mesopores, especially in the endozone. Diaphragms within the mesopores are 0- 1-0-2 mm apart. As the histogram represented in text-fig. 4 shows, acanthopores tend to be of two sizes — 0-03-0-06 mm diameter and 0-08 mm diameter. They originate either in the outer endozone or exozone and usually continue to the surface. Those that originate in the endozone do so invariably at zoecial corners. Those that originate in the exozone com- monly do so between the zoecial corners. There is no correlation between the size of the acanthopores and their place of origin. Most of the acanthopores remain fixed relative to other structures, but a few can be seen to migrate within the walls (PI. 68, fig. 12). Zoecial walls may or may not be inflected by the acanthopores. In the following table of quantitative data measurements are from inter-monticular exozonal areas. In each case the number of measurements per zoarium is 10. Character Range Mode Mean Standard deviation No. of measurements Zoria measured No. of whole zoecia per 1 mm^ 6-15 9 9 1-75 60 6 No. of whole mesopores per 1 mm^ 1-13 4 5 2-76 60 6 No. of whole acanthopores per 1 mm^ 7-22 14 14 3-95 60 6 Thickness of wall in mm 003-0 10 007 007 0021 60 6 Maximum diameter of zoecial void in mm 018-0-30 0-23 0-23 0034 60 6 Remarks. Heterotrypa pontensis Ross (1961, p. 40) from the Lower Devonian Garra Formation near Welhngton is a similar species. It is distinguished from H. rapinae by its finer zoecia (orifices 0-15-0-20 mm in diameter), narrower walls (average inter- monticular width 0-03 mm) and acanthopores, which apart from rare large ones of 0-06 mm diameter, have an average diameter of 0-03 mm. H. australis Etheridge (1899, p. 34) from the Lower Devonian Tabberabbera Formation of Victoria is unidentifiable (Talent 1963, p. 53), but in any case is certainly a much coarser species with zoecial diameters of 1 mm. ^Heterotrypa rusJiworthensH Chapman (1920, p. 174) from the Lower Devo- nian of Rushworth, Victoria, was founded on an inorganic object (Ross 1961, p. 106). H. yakovlevi Sheynmann (1926, p. 921) and variety palmipedalis (1926, p. 922) from the Givetian of the Minusinsk Basin are distinguished from the new species by their narrower walls and smaller zoecia (0-17-0-20 mm diameter in H. yakovlevi and even less in var. palmipedalis). Later Russian authors have referred H. yakovlevi to the genus Lioelema (Nekhoroshev 1948, p. 62; Morozova 1960, p. 87). None of the Silurian species of Heterotrypa described to date (Astrova 1959, p. 40; Perry and Hattin 1960, p. 702; Ross 1961, p. 39) is similar to H. rapinae. PEDDER: LOWER DEVONIAN CORALS AND BRYOZOA 385 Acknowledgements. Mr. P. G. Flood, formerly of the University of New England, accompanied the writer in the field in January 1967 and later made available prior to publication the results of his studies of the Lick Hole conodonts and brachiopods. Dr. J. W. Pickett of the Geological Survey of New South Wales located and loaned specimens collected and referred to in publication by Andrews. Mr. H. O. Fletcher and Dr. A. Ritchie of the Australian Museum arranged for the loan of specimens identified by Fletcher in Adamson 1957. The work was undertaken at the University of New England, Armidale, as part of a larger project on the Devonian biostratigraphy of eastern Australia and was supported financially by the Australian Research Grants Committee. REFERENCES ADAMSON, c. L. 1957. Reconnaissance geology of the Snowy Mountains area. Progress report no. 10. Tumbarumba. Tech. Rep. Min. Dep. N.S.W. 2 (for 1954), 7-15, map. et al. 1966. Australia 1:250,000 Geological Series Sheet SI 55-15, Wagga Wagga. N.S.W. Dep. Mines, Sydney. ANDREWS, E. c. 1901. Report on the Kiandra Lead. Miner. Resour. N.S.W. 10. ASTROVA, G. G. 1959. Siluriyskie mshanki tsentral’noy i zapadnoy Tuvy. Trudy paleont. Inst. 79. BENSON, w. N. 1922. Materials for the study of the Devonian palaeontology of Australia. Rec. geol. Surv. N.S.W. 10, 83-204, pis. 13-14a (separates paginated 1-122). BiRENHEiDE, R. 1963. Cyatliopliyllum- und Dolmrophyllum- Acrien (Rugosa) aus dem Mitteldevon der Eifel. Senckenbergiana Lethaea, 44, 363-458. BOARDMAN, R. s. 1960. Trepostomatous Bryozoa of the Hamilton Group of New York State. Prof. Pap. U.S. geol. Surv. 340, and UTGAARD, J. 1966. A revision of the Ordovician bryozoan genera Monticulipora, Peronopora, Heterotrypa, and Dekayia. J. Paleont. 40, 1082-1108, pis. 133-142. BROWNE, I. A. 1959. Stratigraphy and structure of the Devonian rocks of the Taemas and Cavan areas. Murrumbidgee River, south of Yass, N.S.W. J. Proc. roy. Soc. N.S.W. 92, 115-128, pis. 4-7. CARNE, J. E., and JONES, L. J. 1919. The limestone deposits of New South Wales. Miner. Resour. N.S. W. 25. CHAPMAN, F. 1920. Silurian silicified corals and a polyzoan from Rushworth. Rec. geol. Surv. Viet. 4, 171-175, pis. 13-15. DAVID, T. w. E. ‘edited and much supplemented by’ browne, w. r. 1950. The Geology of the Common- wealth of Australia, vol. 1. Edward Arnold & Co., London. DUN, w. s. 1902. Miscellaneous determination. Rep. Dep. Min. N.S.W. 1901, 175. ETHERIDGE, R.(fil.) 1899. Descriptions of new or little-known Victorian Palaeozoic and Mesozoic fossils, no. 1. Progr. Rep. geol. Surv. Viet. 11, 30-36, pis. A, B. FEDEROWSKi, J. 1967. A revision of the genus Ceratophyllum Giirich, 1896 (Tetracoralla). Acta palaeont. pol. 12, 213-222, pis. 1-3. FLOOD, p. G. 1969. Lower Devonian conodonts from the Lick Hole Limestone, southern New South Wales. J. Proc. roy. Soc. N.S.W. 102, 5-9, pis. 1, 2. HALL, J. 1883. Bryozoans of the Upper Helderberg and Hamilton Groups. Trans. Albany Inst. 10, 145- 197 (extracts separately paged and distributed in 1881). 1886. Bryozoa of the Upper Helderberg Group [plates and explanations only]. Ann. Rep. N.Y. State Geol. 5 (for 1885), pis. 25, 27, 29-32, 40, 41, 44, 45, 48, 50, 51, 53. HARPER, L. F. 1913. Notes on certain areas geologically examined during the year: — County Buccleuch (southern half). Rep. Dep. Min. N.S.W. 1912, 179. HILL, D. 1940. The lower Middle Devonian rugose corals of the Murrumbidgee and Goodradigbee Rivers, N.S.W. J. Proc. roy. soc. N.S.W. 74, 247-276, pis. 9-11. 1954. Devonian corals from Waratah Bay, Victoria. Proc. roy. Soc. Viet. 66, 105-118, pis. 6-9. and JELL, J. s. 1969. Devonian corals from Ukalunda. Publ. Qd. geol. Surv. 340 palaeont. Pap. 16 (authorship overprinted Jell and Hill on some copies). KHALFiN, L. L. (ed.) 1961 (1960 on title-page). Biostratigrafiya paleozoya Sayano-Altayskoy Gornoy oblasti, Tom II, Sredniy Paleozoy. Trudy sib. nauchno-issled. Inst. Geol. Geofiz. miner. Syr\ 20. MOROZOVA, I. p. 1960. Devonskie mshanki Minusinskikh i Kuznetskoy kotlovin. Trudy paleont. Inst. 86. 386 PALAEONTOLOGY VOLUME 14 MOVE, D. G., SHARP, K. R., and STAPLEDON, D. H. 1969n. Snowy Mountains region (Silurian System, Southern and Central Highlands Fold Belt). /. geol. Soc. Aiist. 16, 114-119. 1969b. Snowy Mountains area (Devonian System, Southern and Central Highlands Fold Belt). Ibid. 16, 143-146. NEKHOROSHEV, V. p. 1948. Devoiiskie mshanki Altaya. Paleontologiya SSSR, tom 3, chast' 2, vyp. 1. Akad. Nauk SSSR, Moscow and Leningrad. NICHOLSON, H. A. 1879. On the Structure and Affinities of the ‘'Tabulate Corals' of the Palaeozoic Period with Critieal Descriptions of Illustrative Species. William Blackwood & Sons, Edinburgh and London. OWEN, D. E. 1969. Wenlockian Bryozoa from Dudley, Niagara, and Gotland and their palaeogeographic implications. Palaeontology, 12, 621-636, pis. 114-116. PEDDER, A. E. H. 1965. A rcvision of the Australian Devonian corals previously referred to Micto- phyllum. Proc. roy. Soe. Viet. 78, 201-220, pis. 30-34. JACKSON, J. H., and Philip, g. m. 1970. Lower Devonian biostratigraphy in the Wee Jasper region of New South Wales. J. Paleont. 44, 206-251, pis. 37-50. PERRY, T. G., and HATTiN, D. E. 1960. Osgood (Niagaran) bryozoans from the type area. Ibid. 34, 695- 710, pis. 85-90. PHILIP, G. M., and JACKSON, J. H. 1967. Lower Devonian subspecies of the conodont Polygnathus lingui- formis Hinde from southeastern Australia. Ibid. 41, 1262-1266. and PEDDER, A. E. H. 1967. A correlation of some Devonian limestones of New South Wales and Victoria. Geol. Mag. 104, 232-239, pi. 12. 1968. Stratigraphical correlation of the principal Devonian limestone sequences of eastern Australia. Internat. Synip. Dev. System, Calgary, 1967, 2, 1025-1041. Alberta Soc. Petrol. Geol., Calgary. ROSS, J. p. 1961. Ordovician, Silurian, and Devonian Bryozoa of Australia. Bull. Bur. min. Resour. Aust. 50. SHERRARD, K. M. 1967. Tentaculitids from New South Wales, Australia. Proe. roy. Soc. Viet. 80, 229- 245, pis. 37, 38. SHEYNMANN, YU. M. 1926. Trepostomata iz srednego devona Minusinskogo uezda Eniseyskoy gub. Izv. geol. Kom. 44, 917-936, pis. 25, 26. SKALA, w. 1969. Fin Beitrag zur Geologie und Stratigraphie der Gipfelregion des Poludnig (Karnische Alpen, Osterreich). Jb. geol. Bundesanst. Wien, 112, 235-264, pis. 1-3. TALENT, J. A. 1963. The Devonian of the Mitchell and Wentworth Rivers. Mem. geol. Surv. Viet. 24. TAVENER-SMITH, R. 1969. Wall Structure and acanthopores on the bryozoan Leioclema asperum. Lethaia, 2, 89-97. ULRICH, E. o. 1882. American Palaeozoic Bryozoa. J. Cincinn. Soc. nat. Hist. 5, 121-175, pis. 6-8. A. E. H. PEDDER Institute of Sedimentary and Petroleum Geology 3303 33rd Street N.W. Calgary 44, Alberta Canada Manuscript received 8 September 1970 BRACHIOPODS FROM THE LOWER DEVONIAN MANDAGERY PARK FORMATION, NEW SOUTH WALES by NORMAN M. SAVAGE Abstract. The basal limestone of the Mandagery Park Formation, New South Wales, thought to be of early Siegenian age, has yielded a rich fauna of brachiopods. Almost all previous work on Lower Devonian brachio- pods in eastern Australia has been concerned with faunas preserved in shale horizons and relatively little is known about faunas from Lower Devonian limestones. The brachiopods from the Mandagery Park Formation basal limestone described herein differ considerably from the assemblages previously described from eastern Australian Lower Devonian shale localities and show a marked resemblance to brachiopod faunas from Lower Devonian limestones in central and eastern Europe, the Urals, Siberia, and western North America. The evi- dence of these fossils supports the views of earlier authors who have suggested that frequent migrations of ben- thonic invertebrates occurred between Eurasia and eastern Australia during Lower Devonian time. The fossils described herein are Dolei orthis packhami sp. nov., Isorthis festiva Philip, Platyorthis, sp., Gypidida cf. victoriae Chapman, Anastrophia magnifica australis subsp. nov., Leptaena cf. goldfussiam Barrande, Drahauostrophia burrenensis sp. nov., Machaeraria cf. formosa (Hall), ZHchorhynchitsI sp., Linguopugnoides careiis (Barrande), Cryptatrypa cf. philomela (Barrande), Meristella subovata sp. nov., Nucleospira cf. inelegans (Barrande), and Sturtella mandageriensis gen. et sp. nov. During Lower Palaeozoic times the Lachlan Geosyncline of south east Australia con- sisted of several parallel troughs separated by relatively stable blocks or geanticlines. Thick deposits of greywacke and shale accumulated in the troughs while limestones were often deposited in the marginal areas. Periodic extrusions of acid and intermediate lavas are represented in the troughs by frequent horizons of volcanic detritus which often forms a major component of the greywacke beds. The thickest deposits occurring in the troughs are grey wacke-shale sequences of Silurian and Devonian age, and these are par- ticularly well developed in the Hill End Trough (Packham 1968) and the Cowra Trough (Pakham 1960, Ryall 1965, Savage 1968a). Much of the trough sediment has yielded only a sparse graptolitic fauna and in parts of the Hill End Trough great thicknesses of sediment have yielded no fossils at all (Packham 1968). However, towards the margins of the troughs, and where the more stable areas have been subject to periodic trans- gressions, rich shelly faunas are common and mixed graptolitic-shelly faunas also occa- sionally occur. The brachiopods described in this paper are from an area towards the western margin of the Cowra Trough where several small limestone lenses were deposited during the early Devonian. At about this time the trough was narrowing and shallowing after a major expansion during the late Silurian. Along the eastern margin of the Cowra Trough a much thicker accumulation of limestone was beginning to develop at about the same time (Strusz 1965, 1967). Stratigraphy. The Mandagery Park Formation is exposed in the vicinity of Manildra, New South Wales (text-fig. 1) and is the second of three dominantly calcareous forma- tions deposited in the Cowra Trough during early Devonian times which together com- prise the Gregra Group (Savage 1968a). Near Manildra it is underlain by the richly [Palaeontology, Vol. 14, Part 3, 1971, pp. 387-422, pis. 69-74.] C 8218 D d 388 PALAEONTOLOGY VOLUME 14 fossiliferous olive-green Maradana Shale and 5 miles to the south-east, near Cudal, it is overlain by the limestones and calcareous shales of the highly calcareous Garra For- mation. Its precise relationship to the latter formation is uncertain and possibly the upper part of the Mandagery Park Formation near the western margin of the Cowra Trough is equivalent in age to the lowest part of the Garra Formation further east. TEXT-HG. 1. Location of the Mandagery Park Formation at the type area. During most of the Silurian the Cowra Trough appears to have been downwarping steadily and at least 6000 ft of greywackes, shales, and siltstones were deposited in the region immediately to the east and south of Manildra (Savage 1968a). In the early Devonian, conditions appear to have more nearly approached equilibrium and the Gregra Group sediments were deposited in a shallowing neritic environment. The grey- wacke facies of the Silurian pass upwards into the calcareous facies of the Lower Devonian. The maximum preserved thickness of the Mandagery Park Formation is about 1600 ft. The richly fossiliferous basal limestone, which is about 350 ft thick at its maxi- mum development, grades upwards into poorly fossiliferous tuffaceous sandstone and SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 389 unfossiliferous arkosic sandstone with occasional cross-bedding. In the lowest horizon of the formation sihcification makes possible the extraction of large numbers of well- preserved fossils. These are mainly brachiopods, but corals, gastropods, and conodonts are also present. From this horizon the genera Anastrophia, Quadrithyris, Proreticularia, Eoglossinotoechia, Linguopugnoides, Cryptatrypa, Ogilviella, and Atrypina have all been recorded from Australia for the first time. Elsewhere these genera are most commonly found in assemblages from limestones, particularly in central and eastern Europe, the Urals, and other areas forming part of the Bohemian-Uralian-western North American early Siegenian faunal province. It seems likely that these forms will be found eventually in other Australian limestones at about this horizon. The stratigraphy and structure of the Manildra area have been described in more detail elsewhere (Savage 1968u) together with faunal lists and a discussion of possible local correlations. This present paper is primarily concerned with the silicified brachio- pods from the Mandagery Park Eormation. All the material described here is from that formation at Locality 1, 3 miles south of Manildra (text-fig. 1). Similar assemblages have been collected from Localities 16 and 17, at the same horizon. Parts of the fauna have been described already (Savage 1968 b, c, 1969, 1970). The remaining brachiopods are described in the systematic treatment below. Faunal affinities. Although papers have been written on brachiopods from Lower Devonian shales in Victoria and Tasmania (Gill 1942, 1948, 1950; Talent 1963, 1965; Phihp 1962), little is known about brachiopods from Lower Devonian limestones in these areas. The assemblages described are generally similar to the shale faunas at Manildra but very different from the limestone faunas. Lower Devonian brachiopods have also been described from New Zealand by Allan (1935, 1947) and Shirley (1938) but here again the material is from shale horizons and the faunas have very little in common with the Manildra limestone faunas on account of differences in both facies and horizon. It is with the assemblages from some of the classical Lower Devonian areas of central and eastern Europe that the Mandagery Park Eormation brachiopods have the closest affinities. It has often been observed that there is a fundamental difference between the faunas associated with the predominantly psammitic facies of the Rhinelands and those of the calcareous facies of Bohemia (Shirley 1938; Erben 1962, 1964; Boucek, Horny, and Chlupac 1966). As Boucot, Johnson, and Talent (1969, p. 5) have recently noted, these faunal differences are not determined solely by environmental-lithological factors, for alterations in hthology in a single area do not result in alterations of Bohemian and Rhenish biofacies. There is presumably a lithological control within each faunal province, but the generic composition of assemblages from different rock types has a strong provincial ffavour and is apparently determined by palaeogeographical factors. Pre- continental drift relationships were undoubtedly of great importance in the delimitation of these Lower Devonian provinces. In Bohemia, where the Lower Devonian calcareous facies is so well developed, a tripartite stratigraphic division into the Lochkovian, Praguian, and Zlichovian stages is usually accepted. The Lochkovian, which is correlated here with the Gedinnian and part of the Siegenian, consists chieffy of limestone with some interbedded shales and cherts. It includes the Kotys Limestone and the Lower Koneprusy Limestone, both highly fossiliferous. The Praguian is correlated with the 390 PALAEONTOLOGY VOLUME 14 middle and upper Siegenian together with the lower Emsian. This includes the Radotin Limestone, Kosof Limestone, Upper Koneprusy Limestone, and the Dvorce-Prokop Beds. Many of the brachiopods collected from these two stages show a striking resem- blance to the Manildra brachiopods and most of the forms from the limestone horizon of the Mandagery Park Formation have a close equivalent in the Kotys and Koneprusy Limestones. The similarity between the atrypids and the spiriferids in the two areas is very marked and this similarity is of particular interest as close equivalents of the Manildra species are rare elsewhere. The forms Quadrithyris robusta molongensis Savage, Howellella nucula australis Savage, Ambocoelia praecox dorsiplicata Savage, Proreti- cidaria beddiei Savage, Linguopugnoides carens (Barrande), Gypidida cf. victoriae Chap- man, and Reticidatrypa fairhillensis Savage are represented in Bohemia by Quadrithyris robusta (Barrande), Howellella nucula (Barrande), Ambocoelia operculifera Havlicek, Proreticularia carens (Barrande), Linguopugnoides carens (Barrande), Gypidida caduca (Barrande), and the non ‘punctate’ forms commonly included within the 'granulifera'' group of atrypids (Barrande 1879; Havlicek 1951, 1959; Savage 1969, 1970). The faunas from the Borszczow Stage of the classic sections of Podolia are chiefly of Bohemian facies but there is less limestone and more calcareous shale than at most of the Bohemian localities (Kozlowski 1929, p. 7; Boucot and Pankiwskyj 1962, p. 8). This increase in the argillaceous nature of the substratum can be expected to have affected the composition of the fauna. Furthermore, the Pridolian age Borszczow assemblages are considerably older than the early Siegenian assemblages from the Manildra lime- stone. Despite these factors, a remarkable number of species from Manildra are con- specific, or almost conspecific, with forms from Podolia. Anastrophia magnifica australis subsp. nov., Ambocoelia praecox dorsiplicata Savage, Howellella nucula australis Savage, Cyrtina praecedens Kozlowski, Machaeraria cf.formosa (Hall), Reticulatrypa fairhillen- sis Savage, Linguopugnoides carens (Barrande), and Nucleospira cf. inelegans (Barrande) from the Manildra limestone closely resemble the Podolia species Anastrophia magnifica Kozlowski, Ambocoelia praecox Kozlowski, Howellella laeviplicata (Kozlowski), Cyr- tina praecedens Kozlowski, Machaeraria althi (Kozlowski), Reticulatrypa nieczlaviensis (Kozlowski), Linguopugnoides carens (Barrande), and Nucleospira robusta Kozlowski (Kozlowski 1929; Savage 1969, 1970). The presence of Anastrophia magnifica australis in the Manildra fauna is particularly significant. Anastrophia is a relatively rare genus and forms with a transverse outline appear to comprise a distinct group characteristic of the Bohemian-Uralian-Tasman- Nevadan early Siegenian faunal province (cf. Old World Province Bohemian Com- munity of Boucot, Johnson, and Talent 1969). As might be expected from the discussion above, the Mandagery Park Formation fauna is very different from the Appalachian Lower Devonian faunas. The Appalachian assemblages do not include many of the genera which are present in Bohemia and other Old World Province areas. These include Quadrithyris, Proreticularia, Eoglossinotoechia, Linguopugnoides, Reticidatrypa, Zlichorhynchus, Carinatina, Ogilviella, Falsatrypa, Sep- tatrypa, and Cryptatrypa. Also absent from the Appalachian Lower Devonian deposits is the transverse form of Anastrophia mentioned above. The western North American early Siegenian brachiopod faunas, reported from cen- tral Nevada and the Yukon (Johnson 1965; Lenz 1967, 1968) are quite distinct from the early Siegenian Appalachian faunas and have very clear Bohemian-Uralian alfinities. SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 391 Illustrations of most of these early Siegenian forms from Nevada and the Yukon have yet to be published but it is clear from the faunal lists of Johnson (1965) and Lenz (1967), and from material and photographs sent to me by these workers, that the Manildra limestone brachiopods have close affinities with early Siegenian species from the Great Basin, Nevada and from Royal Creek in the Yukon. This is indicated by the occurrence in western North America of the eastern European forms Anastrophia cf. magnifica Kozlowski, Gypidula cf. pelagica (Barrande), Quadrithyris cf. minuens (Barrande), HowelleUa cf. micula (Barrande), Ambocoelia cf. praecox Kozlowski, Reticidatrypa nieczlaviensis (Kozlowski), Linguopugnoides cf. praecox (Barrande), Spirigerina cf. supra- marginalis (Khalfin), and Cymostrophia cf stephani (Barrande) (see Johnson 1965, Lenz 1967). During late Gedinnian and early Siegenian times the eastern and western parts of North America were apparently covered by two distinct epicontinental seas with western North America part of a faunal province stretching from central Europe, through Asia, to the Yukon and Nevada. An offshoot of this great Eurasian-Western North American migration route appears to have passed into eastern Australia. The Mandagery Park Formation faunas have little in common with the ‘Austral’ Lower Devonian faunas of South Africa, South America, the Falkland Islands, and Antarctica. Faunas from these areas described by Reed (1903, 1904, 1906, 1908, 1925), Knod (1908), Clarke (1913), Kozlowski (1923), Boucot and Gill (1956), Boucot, Caster, Ives, and Talent (1963), and Boucot, Johnson, and Doumani (1965) contain a prepon- derance of terebratulids, leptocoeliids, and chonetids. The composition of these assem- blages probably reflects the combined effects of geographical isolation and the brackish, shallow water conditions which are thought to have prevailed in the Gondwanaland basins at that time. However, as we have seen, the Mandagery Park Formation brachio- pods are decidedly ‘Boreal’ in character and strikingly close to the Bohemian brachiopod assemblages. Terebratulids, leptocoeliids, and chonetids are conspicuously absent and there is no evidence of ‘Austral’ affinities in the Cowra Trough, or indeed in any part of the Australian mainland, during Lower Devonian time. In the systematic treatment below, specimen numbers are those of the Palaeontology Collection, Department of Geology and Geophysics, University of Sydney. SYSTEMATIC PALAEONTOLOGY Phylum BRACHIOPODA Order orthida Schuchert and Cooper 1932 Superfamily orthacaea Woodward 1852 Family hesperorthidae Schuchert and Cooper 1931 Subfamily dolerorthinae Opik 1934 Genus dolerorthis Schuchert and Cooper 1931 Type species. Orthis interplicata Foerste 1909, by original designation. Dolerorthis packhami sp. nov. Plate 69, figs. 1-32; text-fig. 2 Diagnosis. A Dolerorthis with a sulcate dorsal valve, a high ventral interarea, sub- parallel delthyrial margins, and only a small number of secondary costae. 392 PALAEONTOLOGY VOLUME 14 Material. Of about 200 specimens only 12 are complete shells with the valves conjoined, the remainder consisting of isolated dorsal and ventral valves with the latter predominating. Specimen SU 19522 is the holotype and other illustrated specimens are paratypes. Description. Exterior. The shell is subquadrate to semi-elliptical in outline with the greatest width between mid-length and the anterior margin. The lateral profile is convex- plane to biconvex. Five mature ventral valves with lengths from 9-5 to 12T mm have width/length ratios ranging from T09 to T38 with a mean of T23. The sole mature conjoined specimen, which has a length of 8-7 mm, has a width/length ratio of T20 and a thickness/length ratio of 0-49. The ventral valve is deeper than the dorsal valve be- cause of the long, flat, apsacline interarea (PI. 69, figs. 25, 31). A high, open delthyrium is slit-like with subparallel margins. The dorsal valve is gently convex and has a short, concave, anacline interarea (PI. 69, figs. 25, 31) and an open notothyrium which includes an angle of 55-65°. A distinct sulcus is present in the dorsal valve but there is little or no fold in the ventral valve and the anterior commissure is straight. The ornament is of strong angular costae which multiply by lateral branching from 10 to 12 primaries, the secondaries arising at varying distances from the beak, giving a fascicostate ornament marginally. In 7 mature ventral valves with a length exceeding 10 mm the number of costae at 5 mm from the beak varies from 14 to 18 with a mean of 16, and at 10 mm from the beak the number varies from 27 to 30 with a mean of 29. Ventral interior. Short, receding dental lamellae diverge anteriorly and merge with a thickened muscle area which extends anteriorly for one-quarter of the valve length (PI. 69, fig. 32). Strong triangular teeth have their outer edges almost perpendicular to the hinge line (PI. 69, fig. 13). The short, triangular muscle area consists of deeply impressed diductor scars situated either side of a more elevated adductor field. The diductor scars extend well up onto the dental lamellae. EXPLANATION OF PLATE 69 Figs. 1-32. Dolerorthis packhami sp. nov. 1-5, Ventral, dorsal, posterior, anterior, and lateral views of SU 19524 showing the long apsacline ventral interarea and short anacline dorsal interarea. 6-10, Ventral, dorsal, posterior, anterior, and lateral views of ventral valve SU 19529. 1 1-15, Ventral, dorsal, posterior, anterior, and lateral views of ventral valve SU 19530 showing the typically sub- parallel delthyrial margins and strong triangular teeth. 16-20, Dorsal, ventral, posterior, anterior, and lateral views of dorsal valve SU 19521. 21-25, Dorsal, ventral, posterior, anterior, and lateral views of SU 19522 (holotype) showing the linear cardinal process and straight blade-like brachio- phores. 26-30, Ventral, dorsal, posterior, lateral, and anterior views of SU 19527, a young valve showing a wide hinge line and a V-shaped delthyrium. 31, Ventral view of dorsal valve SU 19526 showing the cardinalia and the adductor muscle field. 32, Antero-dorsal view of ventral valve SU 19533 showing the thickened muscle field and the strong, receding dental lamellae. (Figs. l-30x 3; figs. 31, 32 X 6.) Figs. 33-48. Isorthis festiva Philip. 33, Interior of dorsal valve SU 19539 showing the large quadri- partite muscle field and cardinalia. 34-35, Exterior and interior of relatively large ventral valve SU 19535 showing the shape of the muscle field with distinct boundaries to the adductor scar on the low median ridge. 36-38, Dorsal, ventral, and posterior views of dorsal valve SU 19536 showing the large muscle field. 39^3, Ventral, dorsal, anterior, lateral, and posterior views of ventral valve SU 19537 showing the deeply impressed muscle field, distinct lateral umbonal cavities, and shallow crural fossettes. 44-48, Dorsal, ventral, posterior, anterior, and lateral views of complete shell SU 19534. (All figures X 2.) Palaeontology, Vol. 14 PLATE 69 SAVAGE, Devonian brachiopods SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 393 TEXT-FIG. 2. Dolerorthis packhami sp. nov. Dorsal mantle canal system drawn from gerontic specimen SU 22674. TEXT-FIG. 3. Dolerorthis packha?ni sp. nov. Scatter diagrams and limiting polygons of dimensions plotted in millimetres, a. Width and length of 35 dorsal valves, b. Width and length of 35 ventral valves. Dorsal interior. The cardinal process is a linear ridge, long and high, and shghtly thickened anteriorly (PI. 69, fig. 22, 23). Long, blade-Uke brachiophores thin distally and diverge at 55-65° with the inner edges continuous with the notothyrial margins. Small shallow sockets deepen and expand laterally. The large adductor muscle field extends half the distance to the anterior margin and is divided by a broad median ridge. Sub- triangular anterior adductors are separated from slightly smaller, oval, posterior adduc- tors by low, anteriorly directed ridges (PI. 69, fig. 3 1). Distinct diductor scars are visible on the notothyrial platform each side of the cardinal process. In gerontic specimens a digitate to pinnate mantle canal system is sometimes visible (text-fig. 2). PALAEONTOLOGY VOLUME 14 394 Measurements (in mm) SU 19522 SU 19523 SU 19524 SU 19525 SU 19530 Dorsal valve (Holotype) Dorsal valve Complete shell Complete shell Ventral valve Length Width Thickness 8-4 10-8 — 9-3 140 — 4-2 60 2-7 40 4-9 2-9 8-8 111 — Ontogeny. In young individuals the hinge line is at least as wide as the more anteiior parts^’of the shell (PL 69, fig. 26). This relationship is not maintained during growth and in mature specimens the cardinal angles are distinctly obtuse with the hinge line only about four-fifths the width at mid-length (PI. 69, fig. 11, 21). Another change is in the form of the delthyrium which has divergent lateral margins in the young specimens but is more parallel-sided and slit-like in mature specimens (PI. 69, fig. 28, 13). The internal features do not alter with growth apart from a gradual thickening of most structural details, the appearance of a distinctly impressed dorsal muscle field and mantle canal system' (PI. 69, fig. 31; text-fig. 2), and the progressive elevation of the ventral muscle field (PI. 69, fig. 32). Although the position of maximum width of the valves may change, the relationship of overall width to length varies very little during growth and a hnear width-length plot results (text-fig. 3). Discussion. This Manildra form resembles the Wenlock Limestone species Dolerorthis rigida (Davidson) but differs in possessing a distinct dorsal sulcus. Several other late Silurian orthids appear to be closely related to the Manildra material but have been re- ferred by most workers to the genus Schizonema (senior objective synonym of Schizo- ramnia- see Bassett 1970) on the basis that they possess accessory ridges flanking the cardinal process. Boucot (1960, p. 294), Williams (1965, p. H318), Amsden (1968, p. 28), and Bassett (1970, p. 18) all accept this distinction. In the absence of knowledge con- cerning the functional and genetic significance of these notothyrial platform ridges it is difficult to see why they should be afforded generic importance. In most other respects the two genera are much alike and it is an indication of the confusion associated with the Dolerorthis-Schizonema relationship that Schizoramina (= Schizonema) has been illustrated in the brachiopod treatise (Williams 1965, pp. H317-H318, in Moore) with photographs of the species Schizoramma hami Amsden, now referred to Dolerorthis by Amsden (1968, p. 28). A distinction between the dorsal mantle canal systems of the two genera is made in the treatise. The Manildra species has a system more nearly digitate than apocopate (text-fig. 2) and this is suggestive of Schizonema. However, if the form is a Schizonema it is notable in completely lacking notothyrial ridges and an apical plate in the delthyrium. The North American Upper Silurian species Dolerorthis hami is close to the Manildra Dolerorthis. Externally there is a difference only in the height of the ventral interarea, the Manildra species being considerably higher. Internally Dolerorthis packhami can be distinguished by its much longer brachiophores. Dolerorthis was recorded by Schuchert and Cooper (1932, p. 89) as ranging only from the Lower to Middle Silurian, and by Williams (1965, p. H316) as ranging from the Middle Ordovician to Upper Silurian. Although Boucot (1960) suggests a range extend- ing into the late Gedinnian, this would still appear to be too restrictive. Several forms SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 395 have been recorded from the middle or late Lower Devonian and early Middle Devonian which are Dolerorthis as presently defined. Dolerorthis persculpta Phihp 1962 is from the Boola Beds, Victoria, now thought to be of Siegenian age, Johnson (1965) has a Doleror- this from the early Siegenian of Nevada, and Wright (1967) has recorded Dolerorthis from Eifelian age deposits in New South Wales. Dolerorthis persculpta is the only species of Dolerorthis hitherto described from Australia. The external ornamentation and cardinafia are similar to those of Dolerorthis packhami but the Victorian species possesses a distinctly convex ventral umbo, a shorter ventral interarea, and a wider delthyrium (Philip 1962, p. 194, fig. 8n, b). Superfamily dalmanellacea Schuchert 1913 Family dalmanellidae Schuchert 1913 Subfamily isorthinae Schuchert and Cooper 1931 Genus isorthis Kozlowski 1929 Type species. Isorthis szajnochai Kozlowski 1929, by original designation. Isorthis f estiva Philip 1962 Plate 69, figs. 33^8 1962 Isorthis festiva Philip, p. 201, pi. 31, figs. 3-6, text-fig. 12. Material. The collection totals 33 silicified specimens of which only 4 are complete shells with con- joined valves. Of the remainder 8 are dorsal valves and 21 are ventral valves. Description. Exterior. The shell is subcircular in outline with the maximum width at about mid-length. The lateral profile is unequally biconvex with the ventral valve deeper than the dorsal valve (PI. 69, fig. 48). A mature conjoined specimen with a length of 10-5 mm has a width/length ratio of 1-02 and a thickness/length ratio of 0-64. A low broad sulcus in the dorsal valve extends from the beak to the gently sulcate anterior commissure but there is no distinct fold on the ventral valve (PI. 69, figs. 46, 47). Both valves are multicostellate with the costellae multiplying by bifurcation and intercalation to number, in the case of 3 dorsal valves, 32, 34, and 38 at 3 mm from the dorsal beak, spaced at 6 per mm medially, and 56, 58, and 58 at 6 mm from the beak, spaced at 4 per mm medially. Ventral interior. Anteriorly divergent dental lamellae advance one-quarter to one-third of the distance to the anterior margin and enclose narrow but distinct lateral umbonal cavities (PI. 69, fig. 40). Strong triangular teeth have shallow crural fossettes on their antero-medial surfaces (PL 69, fig. 40). The muscle field, which extends one-third of the distance to the anterior margin, consists of two deeply impressed elongate diductor scars bounded laterally by the dental lamellae, and by low ridges extending anteriorly from the lamellae (PI. 69, figs. 35, 40). The diductor scars are separated medially by a low adductor ridge. Dorsal interior. A narrow cardinal process has a bilobed or irregularly multilobed myophore. Deep antero-laterally expanding sockets are supported on distinct fulcral pads and high brachiophores diverge anteriorly at about 90° (PI. 69, figs. 37, 38). The large muscle field is oval to subquadrate in outline and extends at least half the distance 396 PALAEONTOLOGY VOLUME 14 to the anterior margin (PI. 69, figs. 33, 37). It is usually elevated rather than depressed and is bounded by prominent lateral ridges. The rounded posterior adductors are sepa- rated from the larger oval anterior adductors by transverse ridges and the entire muscle field is divided by a low median ridge. Measurements (in mm) Length Width Thiekness SU 19534 Complete shell 10-5 10-9 6-7 SU 19535 Ventral valve 14-5* 15-7 — SU 19537 Ventral valve 7-5 8-7 — SU 19539 Dorsal valve 7-5 * Estimated 6-5 ' — Discussion. This Manildra material differs from the type species, Isortliis szajnochai Koslowski, and from most other eastern European species, in having a very pronounced dorsal sulcus. There is, however, a marked resemblance to Barrande’s illustrations of his species Orthis pinguissima Barrande (Barrande 1879, pi. 62, fig. I, la, b, 2a-f) and it is possible that the material described herein should be referred to that species. Another species of Isorthis with a pronounced dorsal sulcus is the Silurian form from Tennessee, Isorthis arcuaria, but that species is easily distinguished from the Australian form by its strongly incurved ventral beak (Hall and Clarke 1892, pi. 5c, fig. 21 ; Amsden 1949, pi. 1, fig. 14) and less broad dorsal muscle field (Amsden 1949, pi. 1, fig. 16). The type specimens of Isorthis festiva were collected from probable Siegenian age deposits in Victoria. They are preserved as moulds and have suffered distortion but it is clear that they are markedly ventribiconvex with a distinct dorsal sulcus and possess a large dorsal muscle field. The Manildra specimens cannot be distinguished from them in any significant feature. Another Victorian isorthid is Reeftonia alpha (Gill) from the Emsian age Tabberab- bera Formation. In the past this species has been referred most often to Isorthis (see Phihp 1962, Talent 1963) and undoubtedly many of the specimens from the type locahty have distinct Isorthis characteristics. The species, however, appears to be highly variable (Talent 1963) and the specimen figured by Johnson and Talent (1967, pi. 22, fig. 9), which Dr. Johnson has assured me is undistorted, has a dorsal muscle field quite unlike Isorthis. Family RfflPiDOMELLiDAE Schuchert 1913 Subfamily platyorthinae Harper, Boucot, and Walmsley 1969 Genus PLAXYORXfflS Schuchert and Cooper 1931 Type species. Orthis ptanoconvexa Hall 1859, by original designation. Platyorthis sp. Plate 74, figs. 16-21 Material. Only two specimens have been found, both dorsal valves. Description. Exterior. In outhne the dorsal valve is transversely elliptical with the antero- lateral and postero-lateral margins evenly rounded (PI. 74, fig. 16). The width/length ratio of the larger valve is 1-30 and that of the smaller valve 1-43. In profile both dorsal SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 397 valves are weakly convex with a slight sulcus (PL 74, figs. 18, 20). The valve surface is costellate with the costellae multiplying mainly by bifurcation to number, in the case of the larger specimen, 28 at the 3 mm growth stage, spaced at 4 per mm medially, and 48 at the 6 mm growth stage, spaced at 3 per mm medially. Away from the central part of the valve the costellae curve outwards so that the most posterior costellae curve to intersect the postero-lateral margin (PI. 74, fig. 16). Dorsal interior. The posteriorly directed cardinal process is supported on a long, in- clined shaft lying between the thick, erect brachiophores (PI. 74, figs. 17, 18, 20). The sockets are not well preserved in this material. The quadripartite muscle field extends just anterior to mid-length and consists of a pair of elliptical posterior impressions sepa- rated by the anterior part of the cardinal process shaft, and a pair of larger, more widely separated anterior adductor impressions. The valve margins bear flattened crenulations, the raised parts having a distinctive double barrelled appearance (PI. 74, fig. 17). Measurements (in mm) SU 20676 Dorsal valve SU 20677 Dorsal valve Length Width 6-4 8-3 3-7 5-3 Discussion. The transversely elliptical outline with evenly rounded lateral margins, the short hinge line, the almost flat profile, and the long inclined cardinal process, are all characteristic features of the genus Platyorthis. The Manildra material appears to differ from the type species, Platyorthis plano- convexa from the Deerparkian of Maryland, in not possessing a clearly bilobed cardinal process, but this may be only a preservation feature of the two valves available. Platyor- this has not been recorded from Austraha previously. Order pentamerida Schuchert and Cooper 1931 Superfamily penteramacea M’Coy 1844 Family pentameridae M’Coy 1844 Subfamily gypidulinae Schuchert and LeVene 1929 Genus gypidula Hall 1867 Type species. Gypidula typicalis Amsden 1953, by subsequent designation of Amsden, 1953. Gypidula cf. victoriae Chapman 1913 Plate 72, figs. 1-11 1913 Gypidula victoriae Chapman, p. 106, pi. 11, fig. 12. Material. All 59 specimens are isolated valves and they are mostly broken. There are only 2 whole dorsal valves in a total of 22, and only 1 whole ventral valve in a total of 37. Description. Exterior. In outline the shell is transversely elliptical with the greatest width at about mid-length. The hinge-line is short and has a width only about one-quarter that of the shell (PI. 72, fig. 1). The ventral valve is strongly convex with an inflated umbo and a small, erect beak (PI. 72, fig. 6). An apsacline, concave interarea has an apical angle of about 100° and is bordered by low distinct beak ridges (PI. 72, fig. 10). The wide, triangular delthyrium includes an angle of 60-70° (PI. 72, fig. 10). The dorsal valve is 398 PALAEONTOLOGY VOLUME 14 convex but much less so than the ventral valve. It has a beak which is erect to sub-erect (PI. 72, fig. 9) and a low poorly defined interarea which is divided by a wide notothyrium (PI. 72, fig. 1). A prominent fold and sulcus are developed anteriorly. Four angular plications are present on the fold and three in the sulcus (PI. 72, figs. 5, 6). Ventral interior. The dental lamellae converge downwards and anteriorly to form a V- shaped spondylium supported by a median septum (PI. 72, figs. 4, 10, 11). The median septum is variable in length and is shorter than the spondylium in some specimens but advances up to one third the distance to the anterior margin in others. Long, narrow tooth ridges border the delthyrial margins and project as short, strong teeth, sub- triangular in section (PI. 72, figs. 10, 11). The musculature within the spondylium is indistinct. Dorsal interior. Short hinge plates extend about one-quarter the width of the valve and are separated by a moderately deep notothyrial cavity (PI. 72, figs. 1, 8). This is divided by a deep, narrow pit under the beak (PI. 72, fig. 1). The sockets form triangular notches between the lateral edges of the hinge plates and the valve wall. Converging downwards from the hinge plates are large, triangular inner lamellae. These unite along their lower, dorsal edges with long, subvertical brachial processes which diverge slightly anteriorly (PI. 72, figs. 1, 3). Supporting the latter are a pair of thin, outer lamellae which converge dorsally and almost meet before uniting with the valve floor. Anteriorly the edges of the outer lamellae extend almost to the valve mid-length. Measurements (in mm) Length Width Thiekness SU 20677 Ventral valve 13-3 101 6-3 SU 20680 Dorsal valve 150* 200* — SU 19557 Dorsal valve 9-3 11-6 3-3 * Estimated Discussion. The Manildra material appears to be conspecific with Gypidula victoriae Chapman 1913, from Victoria, although further collecting from Chapman’s locality may show differences not apparent from the single specimen he described. A considerable variation in the depth of the dorsal valves occurs in the Manildra specimens and it is possible there would also be some variation in the form of the specimens at the Victorian locality. G. vultura Talent 1963, also from Victoria, differs from G. victoriae in being completely smooth. A form from Europe close to the Manildra specimens is Gypidula cadiica (Barrande 1879). This is gently plicate and has a wide, angular ventral fold limited each side by a distinct plication as in the Australian material. The posterior half of the shell is also smooth (see Barrande 1879, pi. 22, fig. \a-h). When more is known of G. caduca it will be possible to compare the forms more constructively and it would not be surprising if they prove to be conspecific, with G. caduca the senior synonym. As Chapman observed (1913, p. 107) G. victoriae also resembles G. galeata (Dalman), a species from the Upper Silurian of Gotland which is common in the Wenlock and Ludlow of Britain (Davidson 1871, p. 145; Alexander 1947, p. 154) and which has also been recorded from eastern Europe (Kozlowski 1929, p. 135). The Manildra form is similar in general shape to G. galeata but has plications which are larger and more angular. SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 399 Superfamily porambonitacea Davidson 1853 Family parastrophinidae Ulrich and Cooper 1938 Genus anastrophia Hall 1867 Type species. Pentanierus verneuiU Hall 1857, by original designation. Anastrophia magnifica australis subsp. nov. Plate 70, figs. 1-34 Diagnosis. A form of Anastrophia close to Anastrophia magnifica Kozlowski 1929 but with less prominent cardinal extremities, fewer costae, and a less swollen dorsal umbo. Material. In a collection of over 300 silicified specimens 72 are complete shells with the valves con- joined. The remainder consists of isolated dorsal and ventral valves about equally represented. Specimen SU 19554 is the holotype and other illustrated specimens are paratypes. Description. Exterior. The shell is transversely sub-elliptical in outline with the greatest width usually at mid-length but sometimes at the wide, straight hinge line. The cardinal extremities are auriculate (PI. 70, fig. 10). In lateral profile the shell is strongly biconvex (PI. 70, fig. 5). Ten mature shells with lengths from 5-6 to 8-0 mm have width/length ratios ranging from 1-06 to 149, with a mean of T33, and thickness ratios ranging from 0-63 to 1-02, with a mean of 0-84. A wide, apsacline ventral interarea is bounded by low but distinct beak ridges. The apical angle is difficult to measure with accuracy but is estimated to vary from 125 to 135°. The open delthyrium, which includes an angle of 70-80°, is obscured by the dorsal beak, A dorsal fold and ventral sulcus become well defined in the anterior half of the shell and the anterior commissure is strongly uniplicate (PI. 70, fig. 28). Both valves are costate with 4 or 5 costae on the dorsal fold, 3-4 in the ventral sulcus, and 5-6 either side on the lateral slopes. The costae are high and angular and occasionally increase by bifurcation (PI. 70, figs. 25, 30). The shells are often asym- metrical with one of the subcentral costae strongly overdeveloped on one side (PI. 70, figs. 4, 10, 33). Ventral interior. The dental lamellae converge to form a spondylium supported by a short median septum. The spondylium extends approximately one-third the distance to the anterior margin and the median septum extends slightly further (PI. 70, figs. 11, 13, 14). The prominent teeth diverge as a continuation of stout tooth ridges along the delthyrial margins (PI. 70, figs. 10, 11). Shallow crural fossettes face antero-medially (PI. 70, figs. 11, 12). Dorsal interior. Sub-parallel crural lamellae rise directly from the valve floor and ex- tend anteriorly one-quarter of the valve length. Projecting laterally from them are small flange-like alate plates with their free ends directed dorso-laterally (PI. 70, fig. 8). The crural lamellae support brachial processes consisting of subvertical curved ridges with the convex sides directed inwards (PI. 70, figs. 7, 8). At the posterior ends of the brachial processes the hinge plates curve laterally to form concave socket plates united with the valve wall (PI. 70, fig. 7). The sockets are small but deep. No muscle scars are evident in the material available. 400 PALAEONTOLOGY VOLUME 14 Measurements (in mm) (see also text-fig. 4) SU 19547 Complete shell Length 41 Width 54 Thickness 31 SU 19548 Complete shell 5-6 80 4-5 SU 19549 Complete shell 7-2 8-9 5-5 SU 19550 Ventral valve 6-4 100 — ■ SU 19552 Dorsal valve 5-2 7-6 — SU 19554 Broken shell (Holotype) 7-2 8-3 — Ontogeny. A sufficient range of immature specimens is available (PI. 70, figs. 15-30) to show that the auriculate cardinal extremities, characteristic of the mature shells, are not i present in the younger shells. In addition to possessing rounded cardinal margins the > young forms also differ from the mature forms in lacking an incurved dorsal beak and j a prominent fold and sulcus (PI. 70, figs. 18, 24, 28, 29). The disproportionately greater I width of older specimens is evident from the width and length dimensions plotted in ; text-fig. 4. The allometric increase in thickness is less marked. \ Discussion. The Manildra material is characterized by a wide hinge line with sharp cardinal extremities. In outline it is similar to Anastrophia magnifica Kozlowski from the Borszczow Stage of Podolia but the specimens from Podolia have more costae and a more swollen dorsal umbo. Johnson (1965) has recorded cf. magnifica from the Wind- ' mill Limestone in Nevada and Lenz (1967) has recorded A. cf. magnifica from his Mono- graptus yukonensis zone in the Yukon. Both workers assign an early Siegenian age to their horizons (Lenz 1967, p. 595; Johnson, Boucot, and Murphy 1967). Johnson has also recorded A. magnifica from a Gedinnian age horizon in the Roberts Mountain Formation, Nevada (pers. comm.). EXPLANATION OF PLATE 70 Figs. 1-34. Anastrophia magnifica australis subsp. nov. 1-5, Dorsal ventral, posterior, anterior, and lateral views of complete shell SU 19548. 6, Antero-ventral view of dorsal valve SU 19553 showing the crural lamellae, brachial processes, and alae. 7-8, Ventro-lateral and ventro-anterior views of dorsal valve SU 19552 showing the sub-parallel crural lamellae, the brachial processes, and the small alae. 9, Conjoined specimen SU 19554 (holotype) with the dorsal valve partly broken away to show the internal features and the articulation. 10-12, Posterior, dorso-lateral, and dorsal views of ven- tral valve SU 19550 showing the auriculate cardinal extremities, small prominent teeth, shallow crural fossettes, and spondylium supported on the median septum. 13-14, Dorso-lateral and dorso- i anterior views of broken ventral valve SU 19551 showing the spondylium and the median septum. | 15-19, Dorsal, ventral, posterior, anterior, and lateral views of a very young immature specimen ^ SU 19545 showing the short, non-auriculate hinge line, the distinct notothyrium, and the absence of ; a fold and sulcus. 20-24, Dorsal, ventral, posterior, anterior, and lateral views of an immature I specimen SU 19546 showing many of the features present in SU 19545 but having grown sufficiently ' to develop a bifurcation of the central costa of the ventral valve. 25-29, Mature specimen SU 19549 showing a long auriculate hinge line, an incurved dorsal beak, a distinct ventral interarea, and a prominent fold and sulcus. 30-34, Dorsal, ventral, posterior, anterior, and lateral views of an im- mature specimen SU 19547. (All figures X 3-5.) Figs. 35-43. Machaeraria cf. formosa (Hall). 35-39, Ventral, dorsal, posterior, lateral, and anterior views of conjoined specimen SU 19569 showing well the typically rounded contours. 40-41, Antero- ventral and ventral views of dorsal valve fragment SU 19573 showing the form of the hinge plates, the crura, and the cardinal process. 42-43, Antero-dorsal and dorsal views of broken ventral valve SU 19571 showing the slender dental lamellae, the shallow crural fossettes, and the long muscle scars. (Figs. 35-39, 42, 43, x 3; figs. 40, 41, X 6.) Palaeontology, Vol. 14 PLATE 70 4. 20 21 ^ ■*» ^ 22 *'k*,* 18 ■iS 19 SAVAGE, Devonian brachiopods A SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 401 The type species from New York, Anastrophia verneuili, is less transverse than the Manildra form and does not possess the long hinge line. It also has more costae and a more swollen umbo. A. acutiplicata Amsden, from the late Silurian Brownsport Forma- tion of Tennessee has a narrow hinge line like A. verneuili but possesses fewer costae. The two other North American species are A. grossa Amsden, from the Lower Devonian of Oklahoma, and A. internaseens Hall, from the middle Silurian of New York. Both are more rounded in outline than the Australian material and have more swollen umbones. A. deflexa (Sowerby), from the Wenlock Limestone of England, and A. podolica (Wenjukow), from the Lower Silurian Kitaygorod Formation of Podolia, are also rounded species with swollen umbones. TEXT-FIG. 4. Anastrophia magnifica australis subsp. nov. Scatter diagrams and limiting poly- gons of 35 specimens plotted in millimetres, a. Plot of width to length, b. Plot of thickness to length. Species of Anastrophia appear to be divisible into two general groups on the basis of external shape. One group comprises species characterized by a short hinge line and a subcircular outline. Species included in this group are A. deflexa from the Upper Silurian of Britain, Gotland, and Bohemia, A. podolica from the Lower Silurian of Podoha, A. acutiplicata, A. internaseens, and A. delicata from the Upper Silurian of eastern North America, and A. verneuili and A. grossa from the Lower Devonian of eastern North America. The second group comprises forms with a long hinge line and a transversely elongate outline. Forms referred to this group are A. magniflca from the Lower Devonian of Podolia, A. praemagniflca and A. magniflca from the Upper Silurian and Lower Devonian of the Altai Mountains, A. magniflca australis from the Lower Devonian of eastern Australia, and A. magniflca from the Lower Devonian of western North America, The distribution of these species is shown in text-figures 5-7. It is appreciated, however, that there are considerable differences in size between average specimens of the various species included within these two groups and that further distinctions can be drawn based on variations of the internal lamellae, the prominence of the umbones, the in- chnation of the beaks, and the arrangement of the costae. 402 PALAEONTOLOGY VOLUME 14 The transversely elongate species of Anastrophia are first recorded from the Altia Mountains of Siberia in Upper Silurian times and later occur westwards in eastern and central Europe and eastwards in eastern Australia and western North America (text- fig. 6). By Lower Devonian times their occurrence gives a reliable indication of the TEXT-FIG. 5. Geographical and stratigraphical distribution of rounded and transverse forms of Anastrophia. extent of the Bohemian-Uralian-Tasman-Nevadan faunal province (text-fig. 7). The rounded species of Anastrophia are widely distributed in eastern North America and across Europe at intervals during the Silurian but they appear to have become restricted to the Appalachian and Midcontinent areas of North America by the early Devonian (text-fig. 7). This withdrawal to the North American continent east of the Rocky Moun- tains coincides with the isolation of a distinctive Appalachian fauna in these areas during late Gedinnian and early Siegenian times. Order strophomenida Opik 1934 Superfamily strophomenacea King 1846 Family leptaenidae Hall and Clarke 1895 (1894) Genus leptaena Dalman 1828 Type species. Leptaena nigosa Dahnan 1828, by subsequent designation of King, 1846. SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 403 TEXT-FIG. 7. Known Lower Devonian occurrences of Anastrophia. E e C 8216 404 PALAEONTOLOGY VOLUME 14 Leptaena cf. goldfussiana Barrande 1848 Plate 74, figs. 1-8 1848 Leptaena depressa var. goldfussiana Barrande, p. 235, pi. 22, fig. 4. 1879 Strophomena rhomboidalis Wilck. sp. = Leptaena depressa Dalm.; Barrande, pi. 41, figs. 11-13. 1967 Leptagonia goldfussiana (Barrande 1848); Havlfcek, p. 103, pi. 17, figs. 10-18. Material. Of a total of 6 specimens, 1 is almost complete with the valves conjoined, 2 are broken ventral valves, and 3 are fragments of dorsal valves. Description. Exterior. The shell is subrectangular in outline with the greatest width along the hinge line. A concavo-convex lateral profile is accentuated by the pronounced geni- culate anterior trail. The ventral valve is flat to gently convex over the non-geniculate part, but with a slightly swollen umbo and a swollen geniculate rim. The angle of geni- culation is about 100°. Anteriorly the trail descends steeply but is less steep along the sides. A short, apsacline interarea includes an apical angle of about 140° (PL 74, figs. 2, 3). The delthyrium is largely closed by a gently convex pseudodeltidium leaving a small foramen wliich appears to encroach on the umbo in old age (PI. 74, figs. 2, 8). The dorsal valve is planar or gently concave away from the lateral and anterior margins which are sharply reflexed to correspond with the ventral trail. The interarea is very short, almost linear, and the notothyrium is filled by an inflated cardinal process (PI. 74, figs. 2). The surface of both valves is covered by coarse concentric corrugations which pro- gressively increase in size away from the umbones up to the line of geniculation, after which they cease (PI. 74, fig. 1). From the few specimens available it appears that 10-14 corrugations are typical in a mature specimen. Fine radiating ridges cover the whole of the exterior, including the geniculate rim. Ventral interior. A prominent muscle field extends one-quarter to one-third the valve length (PI. 74, fig. 8). The raised rims of the flabellate diductor scars join rudimentary dental lamellae which diverge anteriorly at about 110°. The diductors are separated by a broad median adductor area which is strongly elevated and continuous from the pedicle foramen to the anterior limit of the muscle field (PI. 74, fig. 4, 8). A narrow myo- phragm separates the adductor scars medially (PI. 74, fig. 4). Prominent teeth are strongly divergent at 70-80° as continuations of the tooth ridges along the delthyrial margins (PI. 74, fig. 3). Dorsal interior. A large bipartite cardinal process extends just posterior of the hinge line. Continuous with the cardinal process lobes are massive socket ridges (PI. 74, fig. 7). These extend antero-laterally around moderately deep sockets which are open postero- laterally. Ill-defined muscle scars are excavated into a raised muscle platform which extends half the valve length. The rounded posterior adductors are larger and more widely spaced than the elongate-oval anterior adductors. At the anterior and lateral margins of the dorsal disc is a distinct diaphragm which has no equivalent furrow on the external surface of the valve. Measurements (in mm) Length Width Thickness SU 20688 Ventral valve 10-4 190* — SU 20689 Complete shell 8-8 140* 8-4 * Estimated SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 405 Discussion. Although only a few specimens of this Manildra form are available, several important features are sufficiently well preserved to demonstrate a close relationship with the Bohemian species Leptaena goldfussiana Barrande. The ventral valve of the Manildra form has a very well-developed pedicle collar and a large foramen (PI. 74, fig. 1), both features given by Havlicek (1967, p. 103) to distinguish this species. The sharp geniculation of the trail and complete absence of a ventrally directed ridge at the periphery of the ventral disc are further diagnostic features (Havhcek 1967, p. 103). The irregular nature of the corrugations and widely flaring lateral geniculation of the Manildra form (PI. 74, fig. 8) are also very apparent in the material illustrated by Barrande (Barrande 1879, pi. 41, fig. 12a). The same features can also be seen in speci- mens of this species from the Koneprusy Limestones, kindly sent to the writer by Dr. Havlicek. Superfamily davidsoniacea King 1850 Family schuchertellidae Wilhams 1953 Genus drahanostrophia Havlicek 1967 Type species. Drahanostrophia ficneri Havlicek 1967, by original designation. Drahanostrophia burrenensis sp. nov. Plate 73, figs. 1-21 Diagnosis. A planoconvex Drahanostrophia with posteriorly recurved socket plates. Material. Of 32 silicified specimens only 1 is complete with the valves in contact. The remaining 19 dorsal valves and 12 ventral valves are mostly broken or fragmentary. SU 19558 is the holotype and other figured specimens are paratypes. Description. Exterior. The shell is semicircular to transversely oval in outline with a wide, straight hinge line (PI. 73, figs. 1, 2). The lateral profile ranges from unequally biconvex in younger specimens, to planoconvex, convexoplanar, or resupinate in mature forms (PI. 73, figs. 3, 15, 19). The ventral valve is slightly convex over the costellate part of the valve but has considerable depth because of the high interarea. The height of the interarea is commonly about one-third of the width, and it is planar and apsacUne, making an angle of about 50° with the umbo. The delthyrium is almost closed by a con- vex pseudodeltidium, with a slight opening at the base (PI. 73, figs. 3). A low notothyrium is completely filled by a large cardinal process ankylosed to posterior extensions of the strongly recurved inner socket plates (PI. 73, figs. 20, 21). A small hood-hke chihdium forms a narrow arch over the cardinal process. Neither valve has a fold or sulcus and the anterior commissure is rectimarginate. The surface ornament is multico stellate with costellae which increase repeatedly by inter- calation to number, in the case of the holotype, 40 at 3 mm from the dorsal beak, 64 at 6 mm from the beak, and 116 at 12 mm from the beak. The primary costellae are the most strongly developed and they are distinct to the valve margin (PI. 73, fig. 2). Numerous irregular growth-lines cross the radial ornament (PI. 73, fig. 10). Ventral interior. Prominent tooth ridges project anteriorly from the delthyrial margins and support short teeth. Dental lamellae appear to be totally absent (PI. 73, figs. 6, 8, 18). No muscle field is visible in the material examined. 406 PALAEONTOLOGY VOLUME 14 Dorsal interior. The cardinalia occupy only one-sixth of the hinge line. The cardinal process is bilobed with each lobe deeply grooved and posteriorly directed (PI. 73, figs. 4, 21). An angular cleft divides the lobes which are divergent ventrally at about 25° (PI. 73, fig. 21). The socket plates arise from the valve floor but are strongly recurved posteriorly and medially to overhang the sockets and to fuse with the notothyrial mar- gins (PI. 73, figs. 20, 21). The sockets are widely expanding and open laterally (PI. 73, figs. 3, 5). A weakly impressed muscle field is large and subcircular, extending two-fifths of the distance to the anterior margin and laterally just beyond the socket ridges (PI. 73, fig. 1). It is bounded by barely perceptible elevations. Measurements (in mm) SU 19558 Dorsal valve (Holotype) SU 19564 Complete shell SU 19566 Ventral valve * Estimated Ontogeny. Young and mature stages of both valves have been collected. The young dor- sal valves show the posteriorly directed cardinal process and postero-laterally opening sockets present in the mature specimens but differ in having a relatively longer hinge line and no sign of a muscle field. In a gerontic dorsal fragment a deposit of callus has formed between the socket plates where it obscures the posterior part of the earlier formed muscle striations (PI. 73, fig. 20). The young ventral valves have a less acute angle between the interarea and the valve floor than in the large mature specimens but even at a very young stage there seems to be some distortion of the ventral umbo, indi- cating an early apical attachment. Also present at this stage is a convex pseudodeltidium which completely closes the delthyrium. Discussion. The only occurrence of Drahanostrophia described previously is that of the type species, D. ficneri, from the Middle Devonian of Moravia. The Manildra material dilfers from the younger species in having posteriorly recurved socket plates instead of simple, non-concave plates (see Havlicek 1967, p. 202). Order rhynchonellida Kuhn 1949 Superfamily rhynchonellacea Gray 1848 Family rhynchotrematidae Schuchert 1913 Subfamily orthorhynchulinae Cooper 1956 Genus machaeraria Cooper 1955 Type species. Rhynchonella formosa Hall 1857, by original designation. Machaeraria cf. formosa (Hall 1857) Plate 70, figs. 35^3 1859 Rhynchonella formosa Hall, p. 236, pi. 35, figs. 6n, 6p, 6r. 1893 Stenoschisma formosa (Hall); Hall and Clarke, p. 187, pi. 56, figs. 41-45. 1955 Machaeraria formosa (Hall); Cooper, p. 55, pi. 13b, figs. 13-29. 1963 Machaeraria formosa (Hall); Kulkov, p. 44, pi. 3, figs. 17-19. 1967 Machaeraria formosa (Hall); Gratsianova, p. 74, pi. 7, fig. 3. Length Width Thickness 160* 25-0* — 5-2 6-7 2*8 4-3 6-8 — SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 407 Material. Of 12 silicified specimens 3 are complete shells with the valves conjoined and the remainder are isolated valves and posterior fragments showing the articulation. Description. Exterior. The shell is transversely ovate to subtriangular in outline with rounded contours and with the greatest width anterior to midlength (PL 70, figs. 35, 36). In lateral profile the shell is subequally biconvex with the greatest thickness at about mid-length. Two conjoined specimens, with lengths of 9-0 mm and 8T mm have width/ length ratios of IT 3 and 1T8 and thickness/length ratios of 0-63 and 0-71. The ventral valve is gently convex with a strongly curved umbo terminating in a small sub-erect beak (PI. 70, fig. 38). The delthyrium is partly closed by disjunct deltidial plates which leave a narrow foramen extending to the posterior extremity of the valve. The dorsal valve is more convex than the ventral valve with a broad, prominent umbo incurved beneath the ventral interarea (PI. 70, figs. 37, 38). A pronounced dorsal fold and ventral sulcus arise at about one-third of the shell length and extend to the strongly uniphcate anterior com- missure (PI. 70, fig. 39). In specimen SU 19569, 21 angular costae are present on each valve with 3 on the fold and 2 in the sulcus. In specimen SU 22676, 15 costae are present on each valve with 4 on the fold and 3 in the sulcus. In both of these specimens the costae on the fold bifurcate very close to the beak and the central costa of the sulcus arises by intercalation. Ventral interior. Slender dental lamellae are slightly convex inwards and bound narrow umbonal cavities (PI. 70, fig. 42). The teeth diverge anteriorly and have shallow crural fossettes on their antero-medial surfaces (PI. 70, fig. 42). A narrow ventral muscle field extends half the distance to the anterior margin. It is restricted to the sulcus and has an elongate, subtriangular outline with a relatively straight anterior margin (PI. 70, figs. 42, 43). Only diductor scars have been observed. Dorsal interior. The hinge plates are deeply divided by a narrow notothyrial cavity containing a long, septate cardinal process (PI. 70, figs. 40, 41). The deep, triangular sockets are bounded by the valve margin posteriorly and by large subtriangular outer hinge plates antero-medially (PI. 70, fig. 41). From the inner edges of the hinge plates, and with their bases bounding the notothyrial cavity, arise long ventrally curved crura which are crescentic in cross-section with convex outer surfaces. (PI. 70, figs. 40, 41). No median septum is present. In the few specimens available the dorsal muscle field has not been preserved. Measurements (in mm) SU 19569 Complete shell SU 22671 Complete shell Length Width Thickness 90 lOT 5-7 81 9-6 5-8 Discussion. This Manildra form closely resembles the type material from the Helder- bergian of New York (Cooper 1955, pi. 13b), the only difference of note being the greater prominence of the parietal slopes of the fold and sulcus of the North American form. Australian material has been assigned to Machaeraria previously by Talent (1963). His specimens from the Emsian Tabberabbera Formation of Victoria are not sufficiently well preserved to permit a detailed comparison but as they possess a dorsal median sep- tum they are far removed from M. fonnosa. 408 PALAEONTOLOGY VOLUME 14 The species described by Kozlowski (1929) as Stenochisma althi is very probably a Machaeraria but differs from the type material in having more evenly sized costae. The costae in the sulcus of M. formosa are spread much more widely than those of M. althi (cf. Cooper 1955, PI. 13b; Kozlowski 1929, pi. 5) and the parietal slopes of the fold and sulcus are more strongly developed. Machaeraria has been described from several areas of Siberia. The form from the Altai Mountains, described as M. formosa by Kulkov (1963, p. 44, pi. 3, figs. 17-19) is close to the Manildra and New York material externally but the serial sections of Kulkov (p. 45) show relatively long dental lamellae and large lateral umbonal cavities. Also from the Altai Mountains is M. kurjensis Gratsianova 1967. This Lower Devonian species is more rounded than M. formosa and has a lenticular lateral profile (Gratsianova 1967, pi. 6, fig. 9). A further Asiatic form is M. pygmaea Alekseeva 1967, recently described from the Lower Devonian of the Cherskogo Mountains in north-east Siberia. This is a small form of the genus with a maximum width more posteriorly placed than in the type species (Alekseeva 1967, pi. 6, fig. 2, 3). Genus zlichorhynchus Havlicek 1963 Type species. Zlichorhynchus hiatus Havlicek 1963, by original designation. Zlichorhynchus 1 sp. Plate 71, figs. 30-34 Material. Only a single specimen has been found. This shows the external details very well but nothing is known of the internal features. Description. The shell is subovate in outline with the greatest width anteriorly. The EXPLANATION OF PLATE 71 Figs. 1-29. Meristella subovata sp. nov. 1-5, Dorsal, ventral, posterior, anterior, and lateral views of conjoined specimen SU 16678 (holotype). 6-10, Dorsal, ventral, posterior, anterior, and lateral views of smaller conjoined specimen SU 16676. 11-12, Ventral and ventro-lateral views of broken dorsal valve SU 16680 showing the small concave cardinal plate supported by the high median septum. 13-14, Dorsal and antero-dorsal views of broken ventral valve SU 16686 showing the advancing dental lamellae and the absence of a distinct muscle field. 15, Dorsal view of broken ventral valve SU 16684 showing the deltidial plates and advancing dental lamellae. 16, Lateral view of broken specimen SU 16682 showing a complete spire and one of the primary lamellae attached to a crus. 17, Anterior view of broken specimen SU 16679 showing the articulation of the valves. 18, Antero-lateral view of fragment SU 16681 showing the broad jugal saddle resting on the median septum (but slightly displaced posteriorly), the primary lamellae resting on the valve floor and no longer attached to the crura, and the broken remains of one of the jugal loops. 19-23, Dorsal, ventral, posterior, anterior, and lateral views of an immature specimen SU 22673 showing the narrower outline, more rounded anterior margin, and rectimarginate anterior commissure which occur in the earlier growth stages. 24-28, Dorsal, ventral, posterior, anterior, and lateral views of SU 16677. 29, Antero-dorsal view of broken ventral valve SU 16683 showing the prominent dental lamellae and absence of an impressed muscle field. (All figures x 4.) Figs. 30-34. Zlichorhynchusl sp. Dorsal, ventral, posterior, lateral, and anterior views of SU 20684 showing the rounded features, the low fold and sulcus, and the large hypothyridid foramen. (All figures x4.) Figs. 35-38. Nucleospira cf. inelegans (Barrande). Lateral, posterior, and two anterior views of SU 20687 showing the profiles and the strong hinge plate with its distally pointed lobes. (All figures x 4.) Palaeontology, Vol. 14 PLATE 71 SAVAGE, Devonian brachiopods SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 409 posterior margin is bluntly rounded, the lateral and anterior margins gently rounded, and the antero-lateral margins strongly rounded (PI. 71, fig. 30). In lateral profile the shell is subequally biconvex with the ventral valve the deeper and with the greatest thickness just posterior to midlength (PI. 71, fig. 33). The width/length ratio of the sole specimen is 0-94 and the thickness/length ratio 0-58. The ventral valve curvature in- creases strongly towards the umbo and the beak is prominent and erect (PI. 71, fig. 33). The delthyrium is large with an apical angle of about 75°. Along the delthyrial margins small deltidial plates are present (PI. 71, fig. 32). A broad, poorly defined dorsal umbo is incurved beneath the ventral interarea. A low dorsal fold and ventral sulcus become discernible at about two-thirds the shell length and extend to the weakly uniphcate anterior commissure (PI. 71, fig. 34). The shell surface is costate with 20 angular costae on each valve. No bifurcation or intercalation of the costae occurs and increase is by further growth of weak costae on the postero-lateral margins (PI. 71, fig. 30). There are 4 costae on the fold and 3 in the sulcus. Towards the antero-lateral margins the costae curve outwards slightly at about three-quarters the shell length. Measurements (in mm) SU 20684 Complete shell Length Width Thickness 70 6-6 4 1 Discussion. Without a knowledge of the internal features this species cannot be un- questionably referred to Zlichorhynchus but the external features strongly suggest that genus. The differences between this specimen and Z. hiatus are the presence in the Manildra form of a low fold and sulcus together with lateral costae which curve gently outwards near the anterior margin. Further material may show that this is a new species of Zlichorhynchus or even of Machaeraria. However, the type species of the latter differs from this Manildra specimen in possessing a strong fold and sulcus, more pointed um- bones, and a delthyrium largely closed by disjunct deltidial plates. Family pugnacidae Rzhonsnitskaya 1956 Genus linguopugnoides Havlicek 1961 Type species. Rhynchonella carens Barrande 1879, by original designation. Linguopugnoides carens (Barrande 1879) Plate 72, figs. 12-19 1879 Rhynchonella nympha var. carens Barrande, pi. 122, fig. 4, 5. 1906 Rhynchonella nympha Barrande; Siemiradzki, p. 170, pi. 7, fig. 2. 1929 Camarotoechia carens (Barrande); Kozlowski, p. 155, pi. 9, figs. 5-17. 1937 Camarotoechia carens (Barrande); Nikiforova, p. 37, pi. 6, fig. 11. 1954 Camarotoechia carens (Barrande); Nikiforova, p. 101, pi. 10, figs. 6-9. 1961 Linguopugnoides carens (Barrande); HavUcek, p. 94, pi. 11, figs. 1, 2. Material. Of 9 silicified specimens, 2 are sufficiently complete to show the general form of the shell. The others are posterior and anterior fragments which show the important internal features and the nature of the articulation. Description. Exterior. The shell is transversely ovate to subtriangular in outline with the greatest width anterior to mid-length. The lateral profile is unequally biconvex, the 410 PALAEONTOLOGY VOLUME 14 dorsal valve being much deeper than the ventral valve (PI. 72, fig. 15). The ventral valve is weakly convex but with a strong flexure at the anterior margin to form a prominent tongue (PI. 72, fig. 12). The umbo is narrow and gently curved, and the beak is small and erect to incurved. The interarea is very short and narrow with a small, mesothyrid fora- men (PI. 72, fig. 16). The dorsal valve is very strongly convex with a broad, inflated umbo, which is incurved beneath the ventral beak. A broad ventral sulcus extends most of the valve length to the anterior margin where it is very pronounced and forms a con- spicuous vertical tongue within the dorsal fold (PI. 72, fig. 12). The posterior parts of both valves are smooth but strong phcations develop anteriorly. The sulcus bears 2 or 3 plications and the fold 3 or 4, with a further 2 or 3 on the anterior parts of the lateral slopes (PI. 72, fig. 14). Ventral interior. Short, slender dental lamellae are widely divergent anteriorly and are placed close to the valve walls. The lateral umbonal cavities are narrow but distinct (PI. 73, fig. 18). Small, anteriorly thickened teeth project from the dorsal edges of the dental lamellae. The muscle field has not been observed. Dorsal interior. The hinge plates are divided by a narrow cleft (PI. 72, fig. 19). The outer plates are triangular and slightly concave upwards. Their inner edges support the crural bases and define the sides of the septalium, and their outer edges form low socket bounding ridges (PI. 72, figs. 17, 18). The sockets diverge at about 110° and are elevated well above the valve margins. Long, ventrally curved crura are crescentic in cross-section with the convex faces directed ventro-laterally (PI. 72, fig. 17). The septalium is very short but the median septum is long and slender, extending about one-third of the dis- tance to the anterior margin (PI. 72, fig. 17). The muscle field cannot be seen in this material. Measurements (in mm) SU 20681 Length Width Thickness Conjoined valves 10-9 12-4* 91* * Estimated Discussion. This Manildra material is closely related to the eastern European type species, Linguopugnoides carens (Barrande). The figured type specimen (Barrande 1879, EXPLANATION OF PLATE 72 Figs. 1-1 1 . Gypidula cf. victoriae Chapman. 1-2, Ventral and dorsal views of large broken dorsal valve SU 20679. The outer crural lamellae rise from the valve floor to join the anteriorly divergent brachial processes which then join the inner lamellae posteriorly. The small ridges on the inner lamellae possibly mark the posterior limit of the adductor muscle field. 3, Ventral view of dorsal valve SU 20680 showing most of the features visible in SU 20679. 4-7, Dorsal, anterior, lateral, and posterior views of ventral valve SU 20677 showing the galeate form and the anteriorly developed angular plications. 8-9, Ventral and lateral views of dorsal valve SU 19557. (All figures x 3.) Figs. 12-19. Linguopugnoides carens (Barrande). 12-16, Anterior, dorsal, ventral, lateral, and pos- terior views of broken conjoined specimen SU 20681 showing the acuminate antero-dorsal margin and the strongly inflated dorsal valve. 17, Anterior view of conjoined fragment SU 20699 showing the long median septum, the short septalium, the strongly upcurved crura, and the ventrally con- verging dental lamellae. 18, Anterior view of conjoined fragment SU 20683 showing the articula- tion. 19, Ventral view of dorsal posterior fragment SU 20682 showing the hinge plates and crural bases. (All figures x 4.) Palaeontology, Vol. 14 PLATE 72 14 15 16 SAVAGE, Devonian brachiopods '(' • SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 411 pi. 122, fig. IV-4; Havlicek 1961, pi. XI, fig. 1) is less deep than the Manildra material and has 4 plications on the fold compared with three on the Australian specimen with this part preserved. However, these differences appear to have no specific significance. The specimen sectioned by Havlicek (1961, fig. 34) is quite as deep as the Manildra material and the studies of Kozlowski on Lingitopugnoides carem from Podolia showed that over 80 % of his measured specimens had a thickness 0-7 of the shell length or greater (Kozlowski 1929, p. 157). From these observations it would appear that the majority of specimens of L. carem are much deeper than the type specimen chosen by Havlicek. The statistical studies of Kozlowski also show that about 10% of his measured specimens had 3 plications on the fold or less (Kozlowski 1929, p. 157). Internally, the Manildra material is very like the specimen from Bohemia figured by Havlicek (1961, fig. 34) and the specimen from Podolia figured by Kozlowski (1929, fig. 48). Linguopugnoides has been described from the Altai Mountains of Siberia by Gratsia- nova (1967). Internally the species L. strigata (Khodalevich) and L. remissus Gratsianova do not appear to differ significantly from the type species (Gratsianova 1967, fig. 30, 31). Externally the greater width and more plicate flanks of L. strigata are distinctive whilst L. remissus appears to be more like the type material in shape apart from pos- sessing only 2 plications on the fold. Both of these Altai forms appear to fall within the range of variation demonstrated by Kozlowski for the Podolia material. Order spiriferida Waagen 1883 Superfamily atrypacea Gill 1871 Family lissatrypidae Twenhofel 1914 Subfamily septatrypinae Kozlowski 1929 Genus cryptatrypa Siehl 1962 Type species. Terebratida philomela Barrande 1847, by original designation. Cryptatrypa cf. philomela (Barrande 1847) Plate 73, figs. 22-27 1847 Terebratida philomela Barrande, p. 387, pi. 15, fig. 7. 1879 Atrypa philomela (Barrande); Barrande, pi. 84, fig. II, pi. 134, fig. I, pi. 145, figs. I, II. 1962 Cryptatrypa philomela (Barrande); Siehl, p. 196, pi. 26, fig. 6, pi. 27, figs. 1, 2, pi. 37, figs. 7, 8. Material. Only 2 specimens are present in a collection from this locaUty totalling several thousand shells. One of these was split to expose the interior. Deseription. Exterior. The shell is circular in outline and has a biconvex, lenticular profile (PI. 73, figs. 25, 26). The width/length ratio of a specimen 3-2 mm in length is T06 and the thickness/length ratio is 0-47. The ventral valve is evenly convex with a narrow umbo and a straight beak projecting well beyond the short hinge line (PI. 73, figs. 22, 26). No interarea is evident. The delthyrium has a width about one-tenth the maximum valve width and includes an angle of about 80°. It is flanked by narrow del- tidial plates which reduce the size of the triangular foramen only slightly (PI. 73, fig. 22). The dorsal valve is also evenly convex apart from a small umbo which projects slightly into the foramen (PI. 73, fig. 25). There is no sign of a fold or sulcus and the shell surface is without ornament. The anterior commissure is reetimarginate (PI. 73, fig. 24). 412 PALAEONTOLOGY VOLUME 14 Ventral interior. A narrow delthyrial cavity is bordered by subparallel dental lamellae which extend about one-fifth of the valve length (PI. 73, fig. 27). The anterior edges of these lamellae rise almost vertically from the valve floor to terminate in simple, robust teeth. No muscle field has been observed. Dorsal interior. Small hinge plates diverge at about 90° and are well separated by a wide notothyrial cavity. Downwards the hinge plates are curved to meet the posterior wall of the valve, forming the floor of the widely divergent sockets. From the distal ends of the hinge plates prominent pointed crura project antero-ventrally (PI. 73, fig. 27). In the only internal valve available the teeth have broken off during separation of the valves and nothing of the articulating surface of teeth or sockets is visible. The muscle field, which extends about one-fifth of the distance to the anterior margin, consists of two weakly impressed lanceolate scars separated by a very low median ridge. Measurements (in mm) SU 20680 Conjoined valves Length Width Thickness 3-2 3-4 1-5 Discussion. Cryptatrypa is a poorly known genus. The interior features were described by Siehl (1962) but little is known of the form of the crura or the musclature. Charac- terized by a simple lenticular form and a smooth surface, Cryptatrypa appears to pro- vide Httle of diagnostic value below the generic level. Barrande’s figures of Cryptatrypa pliilomela (Barrande 1879, pi. 84, fig. II, pi. 134, fig. I, pi. 145, figs. I, II) show that some of the larger Bohemian specimens are longi- tudinally oval and develop a gentle dorsal fold. This is also evident in Siehl’s specimens from the Greifensteiner Kalk (Siehl 1962, pi. 37, figs. 7, 8). However, the smaller speci- mens have a circular outline and a rectimarginate commissure. The Manildra material is here tentatively referred to the European type species as no significant differences are evident in the small number of specimens available. EXPLANATION OF PLATE 73 Figs. 1-21. Drahanostrophia burrenensis sp. nov. 1-5, Ventral, dorsal, lateral, posterior, and ventro- lateral views of large dorsal valve SU 19558 (holotype) showing the external ornamentation with the distinct primary costellae and the posteriorly directed cardinal process fused to the short socket plates. 6-7, Anterior and posterior views of ventral fragment SU 19563 showing the prominent tooth ridges, the short teeth, and the deUcately ornamented deltidium. 8, Antero-dorsal view of ventral valve SU 19567 showing the tooth ridges and the absence of dental lamellae. 9-10, Posterior and ventral views of large broken ventral valve SU 19562 showing the planar, apsacline interarea, the convex deltidium, and the distorted umbo. 1 1-15, Antero-ventral, ventral, anterior, dorsal, and lateral views of conjoined specimen SU 19564 showing the unequally biconvex form of this young specimen and the attachment distortion of the ventral umbo. 16-19, Ventral, anterior, antero- dorsal, and lateral views of young ventral valve SU 19566. 20-21, Ventral and posterior views of dorsal fragment SU 19559 showing the cardinal process, sockets, and socket plates of a large gerontic specimen. (All figures x 3.) Figs. 22-27. Cryptatrypa cf. pliilomela (Barrande). 22-26, Dorsal, ventral, anterior, posterior, and lateral views of conjoined specimen SU 20680. 27, The same specimen after breakage to reveal the widely separated hinge plates and the antero-ventraUy directed, pointed crura. (Figs. 22-26 x6; fig. 27 X 14.) Palaeontology, Vol. 14 PLATE 73 SAVAGE, Devonian brachiopods SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 413 Superfamily athyridacea M’Coy 1 844 Family meristellidae Waagen 1883 Subfamily meristellinae Waagen 1883 Genus meristella Hall 1859 Type species. Atrypa laevis Vanuxem 1842, by subsequent designation of Miller 1889. Meristella subovata sp. nov. Plate 71, figs. 1-29 Diagnosis. An equally biconvex Meristella with an erect beak and a recti marginate com- missure. Internally a broad jugal saddle bifurcates to form accessory loops. The dental lamellae are short and the ventral muscle field is not distinctly impressed. Material. Of over 150 silicified specimens 41 are complete shells with the valves conjoined. The re- mainder are separate dorsal and ventral valves or fragments of valves. Specimen SU 16678 is the holotype and other illustrated specimens are paratypes. Description. Exterior. The shell is longitudinally pyriform to subpentagonal in outline with the maximum width just anterior of midlength (PI. 71, figs. 1, 6, 24). The lateral profile is equally biconvex. Ten mature conjoined specimens with lengths from 5-5 to 8-0 mm have width/length ratios ranging from 0-71 to 0-97, with a mean of 0-83, and thickness/length ratios ranging from 0-53 to 0-71, with a mean of 0-58. The ventral valve has a narrow, erect beak terminating in a small, rounded foramen (PI. 71, figs. 5, 6, 13). No interarea is evident. The delthyrium is wide with an apical angle estimated to vary from 85 to 95° and the deltidial plates meet medially to form the anterior margin of the foramen (PI. 71, figs. 1, 13). The dorsal valve is equally convex and possesses a distinct umbo. A broad shallow sulcus is developed in the anterior part of the ventral valve but the dorsal valve is without a corresponding fold and often bears a slight sulcus. The anterior commissure is rectimarginate or gently uniplicate (PI. 71, figs. 4, 27). Ventral interior. Short dental lamellae enclose narrow umbonal cavities laterally (PI. 71, fig. 29). Small, rather flattened teeth project antero-medially from the valve margin (PI. 71, fig. 13). No muscle impressions have been preserved. Dorsal interior. A small concave cardinal plate is present (PI. 71, fig. 11). The deep, narrow sockets are bounded laterally by the up-curved posterior edges of the short, pointed crural lamellae (PI. 71, figs. 11, 12). A high median septum supports the cardinal plate and projects slightly up through it (PI. 71, fig. 1 1). The septum extends almost half the valve length. Laterally directed spiraha each make 6 or 7 volutions (PI. 71, fig. 16). Primary lamellae are attached to the crura at a sharp angle. They turn downward and anteriorly along the valve floor and, at about one-third the valve length, they become very deep and are joined together by a broad, anteriorly projecting jugal saddle sup- ported on the high median septum (PI. 71, fig. 18). The jugum bifurcates posteriorly forming two accessory lamellae which then curve back anteriorly as loops to rejoin the jugum. No muscle impressions are preserved. Measurements (in mm) Length Width Thickness SU 16676 Complete shell 4-6 40 2-8 SU 16677 Complete shell 5-6 4-9 3 0 SU 16678 Complete shell (Holotype) 7-2 7-0 4-4 414 PALAEONTOLOGY VOLUME 14 Ontogeny. The youngest stages of this species have the maximum width at about mid- length and have a rectimarginate anterior commissure. Later stages have the maximum width well towards the anterior of the shell and a very gentle ventral sulcus, without a corresponding dorsal fold, resulting in a slightly emarginate anterior margin (PL 71, fig. 2). The ventral beak is almost straight in the younger forms but later becomes in- creasingly erect (PI. 71, figs. 5, 23). Throughout growth the relationship of width and thickness to length is almost linear (text-fig. 8). TEXT-FIG. 8. Men'stella subovata sp. nov. Scatter diagrams and limiting polygons of 40 specimens plotted in millimetres, a. Plot of width to length, b. Plot of thickness to length. Discussion. For many years meristellids possessing jugal loops but lacking the ‘shoe- lifter’ structure of Merista have been commonly placed in the genus Meristella. Boucot, Johnson, and Staton (1964) state that in the subfamily Meristellinae they ‘will give primary consideration to the presence or absence of dental lamellae together with the nature of the muscle field configuration’ and they have included within the genus Meristina species traditionally considered as Meristella but which possess dental lamellae . and have sub-parallel ventral muscle bounding ridges. They restrict to Meristella species with obsolescent dental lamellae and a widely flaring ventral muscle field. The Manildra j species differs from both these genera as defined above in that it has distinct dental j lamellae but no visibly impressed muscle field. Furthermore, it is equally biconvex and has an erect beak, features normally observed only in young forms of Meristella and j Meristina. jj Several other meristellid genera possess dental lamellae and a slightly impressed |j ventral muscle field but differ from the Manildra species in other very significant details. Whitfieldella has similar ventral internal features and, in addition, usually has an erect beak and an equally biconvex form. However, the structure of the jugum in Whit- j fieldella, as described for the type species (Hall and Clarke 1893, p. 59), is quite different from that of the Manildra species and although jugal structures seem to be very variable and are far from being fully understood, they remain a key factor in generic and familial i classification within the Athyridacea. I Meristella sp. has been recorded from Yass, New South Wales, (Brown, 1941) but has j SAVAGE: LOWER DEVONIAN BRACHIOPODS FROM NEW SOUTH WALES 415 not been described. The form described as Meristina (?) australis by Dun (1904) from Molong, New South Wales, was later redescribed by Mitchell and Dun (1920) and assigned to a new atrypid genus, Atrypoidea. Meristella sp. has been recorded at Tyers and at Heathcote, Victoria. The Tyers specimen is considered by Talent (1965, p. 37) to be a hssatrypid but this is not evident from the illustration of Philip (1962, pi. 35, fig. 3). The Heathcote specimens figured by Talent (1965) are known only from ventral internal moulds. From the photographs they appear to be costate (pi. 15, figs. 4, 6, 8) and may be incorrectly assigned. They are distinct from the Manildra specimens both in the form of the muscle field and in the shell outline. Of the eastern European species, Meristella upsilon Barrande, from the Upper Silurian of Bohemia, is externally very like M. subovata but the internal features of Barrande’s species have not been described and a detailed comparison is not possible. In Podolia only M. wisnioskii is known from the late Silurian and early Devonian and this species is far removed from the Manildra material, being less elongate and having a tightly incurved ventral beak (Nikiforova 1954, pi. 17, figs. 3a-4a-b, from the Lower Chalk of Ventnor, Isle of Wight ; designated by Wright and Wright (1951, p. 38). This specimen is a smaU, weU-preserved, whoUy septate, phosphatic internal mould, retaining traces of matrix which is grey, slightly glauconitic chalk. It is moderately involute, just under half the previous whorl being covered. The intercostal whorl section is compressed, with flat sides and a narrowly rounded venter. The costal section is rather similar, with the greatest breadth at the umbiUcal bulla. The umbiUcus is small and shallow, with a rounded umbilical wall. There are 28 ribs on the outer whorl. Up to a diameter of 18 mm these ribs are very weak; long, flat EXPLANATION OF PLATE 79 All figures are of natural size. Specimens are coated with ammonium chloride. Figs. \a-b. Mantelliceras saxbii (Sharpe) (the holotype of Mantelliceras hyatti Spath, from the Chalk Marl (ManteUi Zone) of Bonchurch, Isle of Wight. Side and ventral views of BMNH 50288 ; figured by Sharpe 1857, pi. 18, figs. Aa-b. Figs. 2a~b. Mantelliceras aff. saxbii (Sharpe) (the syntype of Mantelliceras batheri Spath figured by Mantell 1822, pi. 22, fig. 1, as Ammonites mantelli var. costata), from the Lower Chalk of Lewes, Sussex. Side, front, and ventral views of BMNH 5694. Fig. 3. Mantelliceras saxbii (Sharpe); phosphatized internal mould from the phosphate bed (ManteUi Zone, M. saxbii assemblage horizon) above the Glauconitic Marl of Gore Cliff, Isle of Wight. Side view of WJK 8761. Figs. Aa-b. Mantelliceras aff. saxbii (Sharpe); limonitic internal mould from the Lower Cenomanian 700 m north-east of Koudiat el Assel, 1 1 km north-north-east of Bou Khadra viUage, eastern Algeria (9865.2971). Side and front views of authors’ coUection M6; differs from M. saxbii in possessing fewer ribs although the ribbing on the venter is stronger and the tubercles on the shoulders are sharper. Figs. 5a-b. Mantelliceras aff. saxbii (Sharpe); phosphatized internal mould from the Glauconitic Marl (ManteUi Zone, H. carcitanensis assemblage horizon) at Gore Cliff, Isle of Wight. Side and front views of WJK 9958; more inflated than M. saxbii. Palaeontology, Vol. 14 PLATE 79 KENNEDY and HANCOCK, Cenomanian ammonites KENNEDY AND HANCOCK: MANTELLICERAS 439 ribs arise at the umbilical shoulder, each with a very faint umbilical bulla, and pass straight up the flank; where the whorl starts to contract the ribs turn gently forwards to meet the shoulder at small, pointed, ventro-lateral tubercles. These long ribs are separated by two or three shorter ribs which arise just below mid-flank, but are otherwise similar to the long ribs. Beyond 18 mm the ribbing becomes much stronger, each long rib bears a distinct umbilical bulla, small lower ventro-lateral tubercles appear, whilst the upper ventro-lateral tubercles become stronger and clavate as the diameter increases. In the suture line the first lateral saddle is moderately broad and evenly bipartite; the accessory lobe here is as large as the second lateral lobe (U2). The first lateral lobe is not as long as the external lobe, but has a narrow stem and the end is divided into four narrow and elongate accessory lobes. Discussion. Sharpe’s figure is good, although reversed and somewhat restored. This species is clearly based on a juvenile specimen. A good idea of the middle growth stages of the species can be gained from the following descriptions of the holotype of Mantelliceras hyatti Spath, BMNH 50288 (PI. 79, figs. \a-b). This specimen cannot be separated specifically from the lectotype of M. saxbii at comparable diameters ; it also is a phosphatized, wholly septate, internal mould. The shell is involute, about half the previous whorl being covered. The whorl section is compressed, the greatest breadth being just above the umbilical bulla. The intercostal section is rounded, and there is a broadly rounded venter in costal section; the whorl sides are slightly inflated. The umbilicus is small and rather deep, the umbilical wall is steep and undercut, the umbilical shoulder is quite sharply rounded. There are 36-37 ribs at a diameter of 55 mm. Up to a diameter of 30 mm there are 2 short ribs separating the longer ribs, and such pairs of short ribs sometimes occur up to a diameter of 45 mm. The longer ribs arise at the umbilical seam, with an umbilical bulla, developed to varying degrees on successive ribs. Up to 40 mm diameter the ribs are faintly flexed, swinging forwards across the inner part of the flank, backwards across mid-flank and forwards across the ventro-lateral shoulder, where they broaden slightly. There is an angular change in the costal whorl section a little below the shoulder without lower ventro-lateral tubercles, but strong sharp clavate upper ventro-lateral tubercles are developed on all ribs. The venter is flat with a width about half the whorl thickness, and the upper ventro-lateral tubercles are connected by broad, slightly elevated ribs. The intercalated short ribs are similar to the long ribs except that they arise at various positions on the sides, from just above the umbilical bulla to the mid-flank, and occur in ones or twos between the long ribs. Over the last part of the outer whorl, at diameters greater than 45 mm, the long and short ribs alternate regularly. On some ribs there is a faint indication of a mid-lateral swelling, but never a pronounced tubercle. A later growth stage of Mantelliceras saxbii is illustrated by one of the syntypes of M. batheri Spath (BMNH 5694, the original of Mantell 1822, pi. 22, fig. 1, from the Lower Chalk of Sussex, bearing Mantell’s original label ‘xxii, i’) although it is sfightly more inflated than the type of M. hyatti. This specimen is a well-preserved, slightly worn and distorted, wholly septate, composite internal mould (‘normal’ chalk preservation of Jefferies 1962) in hard limestone. The shell is moderately evolute, about half the previous whorl being covered. The whorl section is compressed, and the greatest breadth is at the umbilical bullae. The intercostal section is rounded; the costal section is also approximately rounded, but there is an angular change on the upper part of the sides and a flattened, broadly convex venter. The umbilicus is narrow and deep, the umbilical wall is overhanging and the umbilical shoulder rounded. There are 37 ribs at a diameter of 92 mm, alternately long and short; up to a diameter of 50 mm there are sometimes 2 short ribs together. The long ribs arise at the umbilical seam, and develop strong umbiUcal bullae immediately above the umbilical shoulder. These ribs are flat and rounded, the same width as, or slightly narrower than the interspaces. The ribs pass straight across the sides, developing clavate ventro-lateral tubercles on each side of the narrow venter. The ribs are connected across the venter by a low rounded rib. The shorter, intercalated ribs arise about a third of the way up the sides, and are similar to the long ribs. The adult body-chamber of M. saxbii appears at about 100 mm diameter in British material, and extends for over half a whorl. A crushed chalk example is figured as PI. 80, figs. \a-b\ a phosphatic 440 PALAEONTOLOGY VOLUME 14 example (slightly more inflated than the type would be, and better referred to as M. alf. saxbii) is figured as Plate 81, fig. 1. The ribs are broad and rounded, and one long rib alternates with one (rarely two) shorter ribs. The venter is flattened between distinct, rounded and faintly clavate, ventro-lateral tubercles. Close to the aperture (traces are preserved in the specimen shown in PI. 80, fig. 2), the ribs weaken and approximate; there are many coarse striae which become distinctly inclined forwards. The aperture itself is simple. TABLE 1. Measurements in mm of Mantelliceras saxbii (Sharpe). The last three specimens are from the phosphate bed above the Glauconitie Marl (Mantelli Zone, M. saxbii faunal horizon). Gore Cliff, Isle of Wight. Collection key: GSM = Geological Survey Museum, Institute of Geological Sciences, London; BMNH = British Museum (Natural History); WJK and JMH = authors’ collection. Measurements of diameter, whorl height, and whorl width have been made between ribs. No. of Whorl Whorl Width of ribs on No. of No. of Specimen Diameter height width umbilicus last whorl primaries secondaries Lectotype GSM Geol. Soc. coll. 7763 25-9 11-7 8-6 5-4 28 8 a. 20 a. M. diyatti' BMNH 50288 560 26-4 19-4 11-6 36 a. 14 22 a. M. "batheri' BMNH 5694 92-4 44-4 37-7 a. 19-8 36 17 a. 19 = M. aff. saxbii WJK 8761 68-3 28-8 23-7 19-3 3,1 18 19 JMH CC761 43-8 a. 22-1 a. 14-8 a. 8-4 WJK 8938 540 24-8 20-6 130 Characters of Mantelliceras saxbii. The distinctive characters of M. saxbii are : (i) when young, a markedly compressed whorl section with flat parallel sides; (ii) when young, the mid-flank decoration is weak, but there are markedly sharp ribs on the shoulders and sharp upper ventro-lateral tubercles; (iii) in youth there are numerous short ribs, and pairs of short ribs are maintained to middle age; (iv) numerous ribs — some 28 on nuclei, increasing to 35-38 in the middle stages, but decreasing again to approximately EXPLANATION OF PLATE 80 All figures are of natural size. Specimens are coated with ammonium chloride. Figs. \a-b. Mantelliceras saxbii (Sharpe); composite chalk mould from the upper part of the phosphate bed (Mantelli Zone, M. saxbii assemblage horizon) above the Glauconitic Marl, Gore Cliff, Isle of Wight. Side and ventral views of JMH CC760; note the retention of tubercles on the shoulders although the whorl compression and probably the clavateness of the tubercles have been exaggerated by crushing. Fig. 2. Mantelliceras saxbii (Sharpe); composite chalk mould from the phosphate bed (Mantelli Zone, M. saxbii assemblage horizon) 3 m above the Glauconitic Marl, Eastbourne, Sussex. Side view of WJK 2689 ; an adult showing striae between ribs on the body chamber. Figs. ?ia-d. Mantelliceras saxbii (Sharpe); the lectotype from the Lower Chalk (Mantelli Zone) of Ventnor, Isle of Wight. GSM Geol. Soc. coll. 7763 (Institute of Geological Sciences photograph). Figs. Aa-b. Mantelliceras saxbii (Sharpe); ILmonitic internal mould from the Lower Cenomanian 700 m north-east of Koudiat el Assel, 11 km north-north-east of Bou Khadra village, eastern Algeria (9865.2971). Side and front views of authors’ collection M5. Figs. 5a-b. Copies of Coquand’s original figures of Ammonites martimpreyi (1862, pi. 1, figs. 7-8). Figs. 6a-c. Mantelliceras ventnorense Diener; the holotype from the Chalk Marl (Mantelli Zone) of Ventnor, Isle of Wight. Ventral, side and front views of GSM Geol. Soc. coU. 7759. Palaeontology, Vol. 14 PLATE 80 KENNEDY and HANCOCK, Cenomanian ammonites KENNEDY AND HANCOCK: MANTELLICERAS 441 30 on an adult ; (v) short ribs and ventro-lateral tubercles are maintained on to the adult body-chamber. The ratio of the whorl height: whorl width is T36 in both the lectotype and the holo- type of M. "hyattV, and this compression immediately distinguishes the species from M. mantelli (J. Sowerby). Only occasionally does one get more compressed specimens with a ratio as high as T5, but less-compressed specimens ranging away from M. saxbii s.s. are common (e.g. Plate 81, figs. \a-b; ratio 1T7-1T8). Up to diameters of about 25 mm, as in the lectotype itself, the width of the venter is around 0-5 of the whorl width; with further growth the whorl section widens faster than the venter, and the venter width may be as low as 0-34 of the whorl width. In M. mantelli the width of the venter is more constant at almost exactly half the whorl width. THE STATUS OF MANTELLICERAS MARTIMPREYI The most important synonym of Mantelliceras saxbii is M. martimpreyi (= Ammo- nites martimpreyi Coquand 1862, pp. 172-173, pi. 1, figs. 7-8) because it has been adopted by many authors as the zonal ammonite for the lowest part of the Cenomanian. The original description and figures of this species are in a rather obscure publication; they are therefore reproduced here as PI. 80, figs. 5a-b. Coquand’s original description is as follows: 15. Ammonites Martimpreyi H. Coq. pi. 1, fig. 7 et 8. Diametre 21 millimetres. Coquille comprimee, ornee, en travers et par tours, de 28 a 30 petites cotes inegales, dont les unes un peu plus saillantes que les autres, et au nombre de 6 a 8, partent de Pombilic et viennent se terminer a la Peripherie: chacune de ces cotes renferme dans I’intervalle qu’elles laissent entre elles, un nouveau systeme de cotes, au nombre de 3 a 5, moins nettement accusees et qui vont en s’attenuant vers I’om- bilic, ou elles se montrent bijugees; chaque cote porte vers son extremite superieure deux tubercules, dont I’un, plus saillant, est limite franchement vers le dos, et dont I’autre est place a deux millimetres en dessous. Dos forme par une carene, s’Hevant sensiblement au-dessus des tubercules terminaux; ombilic peu ouvert. Cette espece, par ses cotes Inegales, par les deux tubercules dont elles sont ornees, ainsi que par sa carene, se distingue des autres Ammonites du terrain cretace. EUe a ete decouverte par M. Ville, a Berouaguia, province d’Alger, dans I’etage rhotomagien. Je me suis fait un devoir de la dedier au general de Martimprey, sous-gouverneur de I’Algerie, qui m’a fourni les plus grandes facilites pour penetrer dans les parties les moins connues des possessions franqaises. In the summer of 1965, one of us (W. J. K.) was able to examine Coquand’s collection in the Geological Institute, Nepstadion Korut, Budapest (part of Coquand’s collection is in the National Museum, but none of that material is relevant here). The specimens are uncatalogued, and are stored by locality. There are nine specimens in one tray bearing the following label: ‘‘Ammonites martimpreyi Coquand. Roth’, [indecipherable symbol] ‘Berrouaghia’ [indecipherable word] ‘Aumaule’. This presents an immediate problem, for Coquand does not mention more than one specimen in his description, and figures only one, which would thus be the holotype by monotypy. Pervinquiere figured seven of these nine ammonites in 1910 and considered one of them (Pervinquiere 1910, pi. 4, figs, la, lb\ PI. 81, figs, la-b herein) as ‘tres 442 PALAEONTOLOGY VOLUME 14 probablement le type figure dans “Geol. Pal. S. Constantine”, pi. 1, fig. 7-8’. Pervin- quiere’s figure shows this ammonite to have a diameter of about 21 mm (allowing for the over-pyritization visible) which would agree with Coquand’s measurement, but Pervinquiere failed to state that his figures are appreciably reduced: the original is larger; moreover, it does not agree with Coquand’s description of the decoration. The main difficulty is that Coquand stated, indeed emphasized, that his specimen had a keel, and this is clearly shown in his figure. If this was an original feature of the shell. Ammonites martimpreyi could hardly be a Mantelliceras, but it could be referred to Cottreauites, a genus with close affinities to Mantelliceras (see Sornay 1955). It is true that the smoothly keeled species of Cottreauites that have been described from north Africa only possess a mere trace of a keel (e.g. C. subboulei Sornay) but the Malagasy species C. boulei Collignon is more distinctly keeled. There are also ammonites with a similar Mantelliceras-type decoration on the sides but possessing a tuberculated keel which have been referred to Acompsoceras, e.g. A. dubourdieui Sornay. But the keel may not have been genuine: in our experience limonitic ammonites from clays in north Africa have quite often developed a ‘keel’ from being crushed, and, as discussed below, several of the ‘co-types’ of A. martimpreyi have such false keels. Suture lines cannot help because Coquand did not figure one. This matter can probably never be fully resolved without neotype designation. We therefore select the original in the Coquand collection figured by Pervinquiere 1910, pi. 4, figs, la-b (reproduced here as PI. 81, figs, la-b) as neotype of Ammonites martim- preyi Coquand. Of the seven surviving figured specimens, the originals of figs. 4, 6, 7, and 8 are all specifically in- separable from M. saxbii, although there is some variation amongst them for instance in the stage at which ribs become prominent across the venter, and the stage where the lower ventro-lateral tubercle begins to weaken. The original of fig. 8 has faint mid-lateral swellings on the long ribs, although no true tubercle is developed; crushing has produced false keels on the originals of figs. 4 and 8. The neotype (fig. 7) is a perfectly preserved, wholly septate, limonitic internal mould. It is almost the same size as the lectotype of M. saxbii, but loses the lower ventro-lateral tubercles earlier, although the costal whorl section continues to be markedly angular at this level on the side. There are up to three short ribs between any two full length ribs. The original of fig. 9 is very close to M. saxbii but differs in having stronger ribbing on the sides (and the venter?), in having a higher ratio of long to short ribs, develops lower ventro-lateral tubercles earlier (and possibly retains them longer), and possesses faint mid-lateral swellings on some of the long ribs. EXPLANATION OF PLATE 81 All figures are of natural size. Specimens in figs. 1, 11-13 are coated with ammonium chloride. Figs. \a-b. Mantelliceras aff. saxbii (Sharpe); uncrushed and partly phosphatized specimen from the Glauconitic Marl/Popple Bed junction (Mantelli Zone), Dead Maid Pit, Mere, Wiltshire. Side and ventral views of WJK 5341. Figs. 2-10. Copies of Pervinquiere 1910 pi. 4, figs. 2-10, the ‘cotypes’ of Mantelliceras martimpreyi (Coquand). Pervinquiere’ s original figure numbers are retained; figs, la-b are the neotype; figs. \0a-b are a specimen of Mantelliceras gr. tuberculatum (Mantell); fig. 4 is generically indeterminate. Originals in the Geological Institute, Nepstadion Korut, Budapest. Figs. Wa-lW. Stoliczkaia (Stoliczkaia) spp. juv.; phosphatic specimens from the Dispar Zone am- monite bed. Upper Greensand, south Dorset. C. W. Wright collection, nos. 7450 (llo-c), 7145 {I2a-b), 231 38 (13a-b). These specimens show the similarities of nuclei of S. (Stoliczkaia) and M. saxbii. Palaeontology, Vol. 14 PLATE 81 KENNEDY and HANCOCK, Cenomanian ammonites V, ? tf f t KENNEDY AND HANCOCK: MANTELLICERAS 443 The original of fig. 2 has a genuine notched keel and may be a Cottreauites. The original of fig. 10 is better judged from Pervinquiere’s figures and, as Pervinquiere noted, can be compared with M. mantelli. The other three specimens preserved in Budapest do not correspond with figs. 3 or 5, and were thus not figured by Pervinquiere. TABLE 2. Measurements in millimetres of the syntypes of Ammonites martimpreyi Coquand in the Geological Institute, Budapest, figured by Pervinquiere 1910, pi. 4; reproduced here in Plate 81, figs. 2-10. Fig. Whorl Whorl Width of No. No. Identification Diameter height width umbilicus of ribs 2 Cottreauites! 10 4-5 3 3 original not found in museum 4 M. saxbii 20 11 7 5 original not found in museum 6 M. saxbii 22 11 8 4 7 M. saxbii (neotype of A. martimpreyi) 25 12 11 5 34-35 8 M. saxbii crushed and distorted 9 M. aff. saxbii 36 19 11 7 34-35 10 now in decomposing fragments There are many ammonites figured as Mantelliceras martimpreyi but not all of them are synonymous with M. soxbii. Pervinquiere (1907) discussed M. martimpreyi at length and with discernment; he stressed the similarity with Sharpe’s var. B of Ammonites mantelli which from Sharpe’s description would appear to be represented by Sharpe 1856, pi. 18, fig. 4, i.e. the holotype of M. hyatti. Pervinquiere also explains correctly the differences from M. mantelli which is more inflated and has mid-lateral tubercles. However, Pervinquiere at this time figured a variety of species under the name Acantho- ceras martimpreyi (1907, pi. 16). Fig. 1 is closer to Mantelliceras ventnorense Diener because the ribbing is flexuous. Figs. 2 and 3 are juveniles, probably not specifically determinable, but belong to the M. saxbii group. Figs. 4 and 5 are more inflated and probably belong to another species. Fig. 16 was renamed Eucalycoceras tymense by Spath (1926). We have examined the original in the Ecole des Mines in Paris and it is a true Mantelliceras, but too inflated for M. saxbii. Fig. 17 is difficult to determine. Fig. 18 was made a syntype of Mantelliceras batheri by Spath (1926). We have examined the original in the Ecole des Mines in Paris: it is close to M. costatum (Mantell) and only differs from Mantell’s species in having abrupt beginnings to the long ribs on the umbilical edge, and a sulcate venter at the tuberculate stage. Proper assessment of all the small pyritic and limonitic specimens from Madagascar, North Africa and Texas must await detailed study of the plentiful material that we have collected stratigraphically from Algeria and Tunisia. As far as can be judged at present not a single limonitic or pyritic ammonite figured by Collignon (1929, 1964) from Madagascar belongs exactly to M. saxbii ; this is possibly only a reflection of the vari- ability of nuclei, and need not mean that the species is absent from Madagascar. 'Submantelliceras" worthense (Adkins) as figured by Adkins (1928, pi. 20, fig. 1 1) appears inseparable from M. saxbii at comparable diameters; Scott (1926) actually called some of Adkins’ specimens Acanthoceras martimpreyi. 444 PALAEONTOLOGY VOLUME 14 SPECIES COMPARABLE TO MANTELLICERAS SAXBII ManteUiceras ventnorense Diener This is the most puzzling relative of ManteUiceras saxbii. The holotype, by the ori- ginal designation of Diener (1925, p. 170) is the original of Ammonites feraudianus Sharpe (1857, p. 51; pi. 23, figs. 6a-c) non d’Orbigny, GSM 7759, from the ‘Grey Chalk at Ventnor’ (figured here as PI. 80, fig. 6). This specimen is a small composite internal mould in hard chalk with a hmonitic coat. It is distorted into an ellipse (major diameter 28-7 mm), and the venter is abraded over the last third of a whorl. This poor preservation makes comparison with M. saxbii difficult, but the decoration is similar. The chief distinctions of M. ventnorense appear to be: (i) the much earlier loss of a dis- tinct lower ventro-lateral tubercle (there is the merest trace of such a tubercle at the smallest diameter visible — 12 mm?); (ii) the much more flexuous ribbing. Both these features could be exaggerated in the type by the abrasion, crushing and distortion; the apparently greater inflation is almost certainly artificial. However, examples of Man- teUiceras which are compressed and have flexuous ribbing are known (e.g. Pervinquiere 1907, pi. 16, fig. \a-b)\ several authors have described what are probably larger speci- mens of this species (e.g. M. ventnorense in Fabre 1940, p. 236, pi. 8, fig. 8; and M. ventnorense in Collignon 1964, p. 87, pi. 347, fig. 1547). We leave M. ventnorense here as a separate species, but further work may well indicate that it would be better treated as a subspecies (or variety) of M. saxbii. ManteUiceras couloni (d’Orbigny) A neotype for this species was designated by Collignon (1937, p. 56, pi. 11) who dis- tinguished it from M. hyatti (i.e. M. saxbii) in that: (i) the ribs do not reach the centre of the venter so that between the high ventro-lateral tubercles there is a sinus; (ii) the narrowness of the venter combined with the gentle convexity of the flanks produces a perfectly oval intercostal whorl section. One of us (J. M. H.) has re-examined the neotype (from Lamnay, Sarthe, France), several topotypes and a number of other examples from the Sarthe. They show that up to moderate diameters (50 mm, possibly more) M. couloni is not easily distinguished EXPLANATION OF PLATE 82 All figures are of natural size. Specimens are coated with ammonium chloride. Figs. \a-b. ManteUiceras aff. couloni (d’Orbigny); internal mould with traces of cast of shell from the basement bed of the Wilmington Sands (ManteUi Zone), Hutchins’ Pit, Wilmington, south Devon. Ventral and side views of JMH CC337. Fig. 2. ManteUiceras saxbii (Sharpe). Front view of specimen figured in Plate 79, fig. 3. Figs. ~ia-b. Utaturiceras vicinale (Stoliczka) ; internal mould from the Utatur Group, southern India. Side and front views of Oxford University Museum KY 311. Fig. 4. ManteUiceras aff. saxbii (Sharpe); phosphatized internal mould from the phosphate bed (Mantelli Zone, M. saxbii assemblage horizon) above the Glauconitic Marl, Gore Cliff, Isle of Wight. Front view of WJK 8763; the nucleus is close to the lectotype but the whorl section in the middle stage is appreciably more inflated. Fig. 5. ManteUiceras aff. saxbii (Sharpe); phosphatized internal mould with traces of phosphatized shell from the Popple Bed (Mantelli Zone), Dead Maid Pit, Mere, Wiltshire. Ventral view of JMH CC739; angular shoulders with strong tubercles have been retained to a middle growth stage. Palaeontology, Vol. 14 PLATE 82 KENNEDY and HANCOCK, Cenomanian ammonites ■« i ^ iril PiraV KENNEDY AND HANCOCK: MANTELLICERAS 445 from M. saxbii: in M. couloni there is a narrow venter in relation to the inter-costal whorl thickness (1:4), and umbilical bullae are absent. Around diameters of 70 mm M. couloni develops exaggerated ventro-lateral tubercles whilst retaining a narrow venter; the tubercles are only slightly clavate. In some individuals these great tubercles are con- tinued to diameters of 135 mm; in others the adult body chamber begins around 100- 110 mm diameter, on which the ventro-lateral tuberculation rapidly disappears and thereafter the venter merely shows a slight flattening. Not only are the younger stages possibly indistinguishable, but there are all grada- tions between M. couloni and M. saxbii. However, the two extremes have not yet been found together at the same horizon, and one is probably the descendant of the other. The exact horizon of the neotype is not reeorded, but its lithology combined with the locality of Lamnay suggests that it was from the top third of the Lower Cenomanian. Wright and Wright (1951) recorded M. couloni from south-west England, correctly noting that the English examples are less compressed than the type. Moreover, the tuberculation is probably never so strong. We figure an example (PL 82, figs. \a-b) from Wilmington which is less compressed than the types of both M. couloni and M. saxbii, but in other features is intermediate between the two species. This specimen has a venter- width 0-36 of the whorl-width. Utaturiceras vicinale (Stoliczka) This ammonite has had a somewhat chequered history. Spath (1926) used "Acantho- ceras' vicinale as a zonal index for the top of the Cenomanian, as did Collignon (1959) and Wright (1957). In 1956 Wright made it the type species of a new gQmx5— Utaturiceras. As Casey (1960) notes, this Upper Cenomanian age is erroneous (it probably originated with Kossmat’s inference of horizon from the matrix), and it is a Lower Cenomanian genus. He concluded that Utaturiceras was a synonym of Mantelliceras, and Ammonites vicinalis could be matched with Mantelliceras of the group of M. saxbii. We had been inclined to follow the view of Casey until the revision of Utaturiceras vicinale by Matsumoto et al. (1966) combined with the features seen in a specimen in the Oxford University Museum (PI. 82, figs. ?>a-b) showed that the species are slightly diflerent. In particular, juveniles of U. vicinale have more flexuous ribbing, whilst the middle and adult stages are more compressed, high whorled and with more numerous subdued ribs (40:35) which are very weak across the venter, whilst the width of the venter is less than a third of the inter-costal whorl thickness. These subtle but important differences give Ammonites vicinalis a quite distinct appearance. We are still uncertain whether it merits generic or even subgeneric separation from Mantelliceras', it could still be an extreme variant of the M. saxbii group. Matsumoto and Sarkar were unfortunate in being dependent on specimens from the Isle of Wight for suture lines of Mantelliceras. The suture line differences they list are not so distinct when one compares U. vicinale with well preserved Mantelliceras limestone internal moulds from Devon (text-fig. 1). For these reasons we are sorry that Matsumoto (1969, p. 291) should have felt it neces- sary to place Utaturiceras in a separate subfamily (Utaturiceratinae) from Mantelliceras. Other species described from Europe Mantelliceras cressierense Renz (1963, p. 1 105, pi. 4, figs. 3n, 2>b), from the Swiss Jura, is based on a single crushed specimen; it differs from M. saxbii in its coarser, fewer ribs 446 PALAEONTOLOGY VOLUME 14 (29:37 per whorl at 70 mm diameter), but is probably a closely allied species. Renz’s M. saxbii (pi. 4, fig. 1) differs from that species by its possession of a marked mid-lateral tubercle, but his M. aff. saxbii and M. hyatti are both M. saxbii, as is possibly his M. ventnorense (pi. 4, fig. 2). Busnardo (in Busnardo et al. 1966) has figured several Mantelliceras from the French Jura whose preservation is rather poor and which are therefore difficult to assess properly. His M. albanense (pi. 12, figs. 1, 2) resembles the adult of the more inflated M. saxbii such as we figure as Plate 8 1 , fig. 1 , but the inner whorls of M. albanense are said to have ribs of equal length. TEXT-FIG. 1. External suture-lines above the umbilical shoulder of Mantelliceras and Utatiiriceras. All X 2. Figs, b, d-f drawn by J. M. H. a. Mantelliceras saxbii (Sharpe) Kyushu University no. 9307, T. Matsumoto collection from the Glauconitic Marl, St. Catherine’s Point, Isle of Wight. After Matsumoto in Matsumoto, Sastry, and Sarkar 1966. b. Utaturiceras vicinale (Stoliczka) Oxford University Museum KY 311 from the Utatur Group of southern India. c. Utaturiceras vicinale (Stoliczka) Lectotype, Geological Survey of India 190, from the Utatur Group, Odium, southern India. After Matsumoto in Matsumoto, Sastry, and Sarkar 1966. d. Mantelliceras aff. saxbii (Sharpe) JMH CC461 from the Mantelliceras-nc\\ bed, Mr. Hutchins’ pit, Wilmington, Devon. e. Mantelliceras saxbii (Sharpe) WJK 8761 from the upper phosphate bed. Gore Cliff, Isle of Wight. /. Mantelliceras ventnorense Diener JMH CC459 from the Mantelliceras-rkh bed, Mr. Hutchins’ pit, Wilmington, Devon. In the list below of the features which Matsumoto and Sarkar (1966) say distinguish the sutures of Utaturiceras from those of Mantelliceras, the exceptions are noted afterwards in brackets: (1) They are fairly deeply incised (also seen in fig. d)\ (2) There are numerous descending auxiliaries (also seen in fig. d). In addition they note in the suture lines of Utaturiceras: (3) The deeply bipartite external saddle (also seen in fig. e)\ (4) The first lateral lobe is deeper than broad and is bipartite at the bottom; its median foliole is much lower than and overhung by the adjacent lateral folioles (also seen in fig. /) ; (5) The first lateral lobe is deeper than the external lobe E (also seen in fig. e but not shown by the Utaturiceras in fig. b)\ (6) The inner (i.e. dorsal) branch of the external saddle is slightly taller than the outer, and narrowed by deeply incised lobes (also seen in fig./); (7) The saddle between the lateral lobe L and the second lateral lobe U2 is higher than that between the external lobe and the lateral lobe (also seen in fig. d and fig./, but barely shown by the Utaturiceras in fig. Z>); (8) The saddle inside the second lateral lobe U2 and the auxiliary saddles are arranged on a des- cending line (also seen in fig. d) ; (9) The second lateral lobe U2 is much smaller than the lateral lobe L (if anything this is less marked in the Utaturiceras fig. b). It has also been our experience that: (a) the first lateral saddle (= external saddle) in Mantelliceras of the saxbii group is broader than in the specimen figured by Matsumoto and Sarkar (fig. u); {b) the outer side of the lateral lobe is lower on the side than the lower ventro-lateral tubercles. KENNEDY AND HANCOCK: MANTELLICERAS 447 448 PALAEONTOLOGY VOLUME 14 Species described from Madagascar Collignon (1964) has figured several forms which belong here, or are closely related. His Mantelliceras hyatti (p. 71, pi. 340, figs. 1518-1520) is carefully compared with the type and is therefore synonymous with M. saxbii; fig. 1519 illustrates the slightly more inflated form such as we figure as Plate 79, fig. 5. This greater inflation is also charac- teristic of M. hyattiforme Collignon (p. 71, pi. 340, figs. 1521-1524) but the chief dis- tinction of this species is the greater number of ribs (42-45). Mantelliceras pseudohyatti Collignon (1964, p. 73, pi. 341, fig. 1525) also has more ribs than M. saxbii (43 at a diameter of 92 mm) ; apart from this rib density, its only other marked distinctions from the English syntype of M. batheri described on p. 439, are the slight forward twist of the ribs above the lower ventro-lateral tubercle and the considerable strength of the ribs on the venter. Mantelliceras biroi Collignon (1964, p. 84, pi. 346, figs. 1540, 1541; pi. 351, fig. 1540) is characterized by its broad ribs which lose their ventro-lateral tubercles comparatively early (around 40-60 mm diameter). However, it is doubtful if the form identified as M. saxbyi (pi. 351, fig. 1558) really belongs there: the figure is poor but one can see mid- lateral tubercles and markedly strong lower ventro-lateral tubercles. Eucalycoceras spp. Eucalycoceras is an Upper Cenomanian genus, type species E. pentagonum (Jukes- Browne & Hill), and therefore not likely to give rise to confusion if the horizons of the specimens are known. However, the middle stages and particularly the nuclei of com- pressed species are superficially similar to M. saxbii. They can be distinguished by {a) the possession of a strong umbihcal tubercle (usually bullate) on every rib at the um- bilical shoulder; {b) the sharpness of the ribbing, particularly noticeable on the venter where it is stronger than in M. saxbii. Graysonites spp. Wiedmann (in Basse 1959, p. 807) remarked that he did not believe that nuclei of M. i ^martimpreyV could be equated with true Mantelliceras', it would be necessary to prove that they were not really nuclei of Graysonites, a genus which occurs in the basal Ceno- manian of Texas. i This genus, as described by Young (1958), shows great variation in the suture lines. | Although that of the type species, G. lozoi Young (1958, text-fig. If), is distinct in pos- sessing an exceptionally wide first lateral saddle with two strong, but unequal, accessory f lobes, all the suture lines are drawn at diameters too large to compare with nuclei from either England or north Africa. The decoration of the young of G. lozoi (Young 1958, I pi. 27, figs. 1, 2, 6, 7, 9, and 10) show it to have stronger and longer ribbing on the sides j than is to be found in Mantelliceras of the saxbii group, and possibly stronger than in ; any old world Mantelliceras. Mature Graysonites have great horns on the shoulders | reminiscent of a compressed Euomphaloceras. I STRATIGRAPHY A Zone of Mantelliceras martimpreyi has long been widely quoted as the lowest zone of the Cenomanian stage, e.g. Wright 1957, Collignon 1959, Thomel 1965. There have j KENNEDY AND HANCOCK; MANTELLICERAS 449 been occasional authors who have referred to a bottom pre-Martimpreyi subzone, e.g. Spath 1926, Muller and Schenk 1943, although generally only as a subzone of the Martimpreyi Zone. Always this Martimpreyi Zone has been placed below a Zone of M. cantiaimm (Spath 1926), or M. mantelli (Collignon 1937, 1959; Wright 1957; Thomel 1965), or Schloenbachia varians (Muller and Schenk 1943). Although we do not suggest that any of these authors were basing their zonation on single species, there can be no doubt that they considered M. martimpreyi itself to occur at, or very close to, the base of the Cenomanian. Because M. martimpreyi had supposedly not been found in northern Europe, it was believed that there was a zone missing at the base of the Cenomanian in this region. We contend that M. martimpreyi itself is a synonym of M. saxbii (as de- scribed above) and that its horizon in northern Europe is well above the base of the Cenomanian. This conclusion is not entirely novel; Dubourdieu (1956), in a penetrating analysis of the ammonite succession in the Cenomanian of the Ouenza region on the borders of Algeria and Tunisia, showed that M. martimpreyi did not enter the succession until above the occurrence of Idiohamites alternatus and Hypoturrilites of the groups to which H. gravesianus and H. carcitanensis belong. The same is true of the English suc- cession as was foreshadowed by Spath (1926). THE MANTELLI ZONE IN SOUTHERN ENGLAND Subdivisions of the Lower Cenomanian. Recent fieldwork in southern England (Kennedy 1969, to which reference should be made for full faunal lists) has shown that it is possible to divide the Lower Chalk into three Zones of Mantelliceras mantelli, Acanthoceras rhotomagense, and Calycoceras naviculare. These divisions correspond to those recog- nized by one of us (Hancock 1959) in the type Cenomanian of the Sarthe. Within the lowest zone (Mantelli Zone) occur three distinctive and widely recog- nizable ammonite assemblages. It would be premature to call these subzones because the faunas are largely concentrated at discrete horizons separated by relatively unfossili- ferous beds which may correspond to levels which elsewhere yield other distinctive assemblages. (i) At the base is a Hypoturrilites carcitanensis assemblage. In this the commonest ammonites are Schloenbachia which probably outnumber all other ammonites put to- gether. But the characteristic, and still common, members of the assemblage are various inflated Mantelliceras such as M. mantelli (J. Sowerby), M. tuberculatum (Mantell), and M. cantianum Spath; and H. carcitanensis (Matheron). Particularly characteristic but infrequent are Idiohamites collignoni Spath, I. alternatus (Mantell), /. alternatus var. vectensis Spath, I. ellipticus (Mantell), I. ellipticus var. radiatus Spath, several possible new species of Idiohamites, Anisoceras auberti (Pervinquiere), A. armatum (J. Sowerby), and A. aff. picteti Spath. Other frequent finds are Hypoturrilites gravesianus (d’Orbigny), H. tuberculatus (Bose), Mariella cenomanensis (Schluter), Mar. lewesiensis (Spath), Mantelliceras cos- tatum (Mantell), M. tenue Spath, M. aif. saxbii, M. aff. ventnorense — the last two both more inflated than the types (see PI. 79, figs. 5a-b, with a whorl height: whorl width ratio of 1-09). (ii) In the middle is the Mantelliceras saxbii assemblage. In this Schloenbachia is still a common genus but not so markedly as in the carcitanensis assemblage, and the inflated 450 PALAEONTOLOGY VOLUME 14 forms have become still less common. Amongst the Mantelliceras, compressed species, e.g. M. saxbii, M. ventnorense, M. tenue, and related forms, are now dominant over inflated species, but some inflated species still survive, e.g. M. cantiammi, M. mantelli, and M. tuberculatum. The Idiolmmites and Anisoceras so typical of the carcitanensis assemblage have disappeared. Hyphoplites falcatus (Mantell) and other Hyphoplites spp. become locally common. Hypoturrilites is, in general, scarcer, although H. tuberculatus is sometimes frequent and H. carcitanensis is replaced by forms best referred to as H. alf. carcitanensis which differ from the earlier form in details of ornament. Austiniceras austeni (Sharpe) and Scaphites obliquus J. Sowerby can be common here, but are not characteristic. (iii) Near the top is the Mantelliceras dixoni assemblage. This is not a rich assemblage and the most abundant ammonites are compressed forms of Schloenbachia. This level is characterized by the appearance of Mantelliceras of the group of M. dixoni Spath and M. aff. souaillonense (Renz), which are accompanied by M. lymense and M. cf. mantelli. Occurrence of Mantelliceras saxbii. M. saxbii is common in the middle assemblage horizon, to which it lends its name, over the whole of south-east England, including the Chilterns (Childrey, Chinnor) and the Mere-Warminster region (especially from the Popple Bed). It occurs also m the remanie Lower Cenomanian faunas of the Chalk base- ment beds of the south-west, whilst this and related forms are also to be found in the phosphatic conglomerate of the top of the Eggardon Grit of the Hooke Valley; in Bed A2 of the Cenomanian Limestone on the Devon coast and at Storridge Hill (Chard- stock) ; and in the basement bed of the Wilmington Sands. It must not be assumed from these records that all these fossiliferous beds in south-west England are exactly on the M. saxbii assemblage-horizon. Related forms, which differ from the type chiefly in their greater inflation, occur in both the saxbii and carcitanensis assemblages over the whole of southern England. Horizon of the types o/M. saxbii and M. hyatti. In describing Ammonites saxbii, Sharpe (1857, p. 45) recorded that his figured specimen (now the lectotype) came from the Grey Chalk of Ventnor, Isle of Wight. He recorded that the original of pi. 18, fig. 4 (the holo- type of M. hyatti) came from the Grey Chalk of Bonchurch, also in the Isle of Wight. Both specimens are preserved as phosphatic internal moulds and retain traces of grey-buff, silty glauconitic chalk. This preservation might suggest that they were from the Glauconitic Marl, but it should be noted that Sharpe was very careful to state which specimens were actually from this horizon, e.g. explanation of pi. 18, figs. 6 and 7 {Mantelliceras mantelli and M. tuberculatum). Recent field work in the Isle of Wight has shown that there is a phosphate bed above the Glauconitic Marl (i.e. in the ‘Grey Chalk’) which is identical in lithology with the sediment that adheres to the two types, which yields M. saxbii in a preservation identical with that of the two types, and is clearly the source of both of them. The relationship of this saxbii assemblage-bed to the base of the Chalk as it is traced across the Isle of Wight is summarized in text-fig. 2. Whilst the horizon of the two types is resolved, the actual locahty is far from certain. Glauconitic Marl and the overlying Chalk Marl have been exposed at several locahties at Ventnor; thus there is a section 300-400 m west of the promenade (National Grid reference SZ 559772) and a large pit next to the old station (SZ 561779). Bonchurch is immediately east of Ventnor, and Lower Chalk fossils could well be from landslipped KENNEDY AND HANCOCK: MANTELLICERAS 451 boulders on the beach (see Middlemiss and Bromley 1962, p. 450), from a small pit below Bonchurch Down (SZ 572782), or possibly from a pit below Nansen Down, north of the A3055 (SZ 578790). S. Saxby, the collector of both types, lived at Bonchurch, and therefore might even have got them from a temporary exposure. TEXT-FIG. 2. Stratigraphical occurrences of the Mantelliceras saxbii assemblage in the Isle of Wight. Horizon of the English syntype of Mantelliceras batheri. Mantell figured his specimen (pi. 22, fig. 1) amongst a group of fossils from ‘the chalk and chalk marl, near Lewes’. The preservation suggests an horizon in the lower part of the Lower Chalk, a formation within which both the carcitanensis and saxbii assemblages are prominent near Lewes. STATUS OF THE M ARTIMPREYI ZONE We ourselves do not use ‘Martimpreyi Zone’ in southern England but its equivalent is clearly the saxbii assemblage-horizon. Inflated Mantelliceras, such as M. mantelli, occur both above and below this, and are actually more prominent below. The same is true of the more inflated Schloenbachia such as S. varians. Therefore if Martimpreyi Zone is used, it should not be placed at the base of the Cenomanian. Under various names Mantelliceras saxbii or closely related forms have been de- scribed from Lower Cenomanian rocks in Europe (Germany, Switzerland, northern and C 8216 H h 452 PALAEONTOLOGY VOLUME 14 southern France), north Africa (Algeria, Tunisia), Madagascar, and possibly North America. Flow is it that collectors have thought that it marked the base of the Ceno- manian stage ? One of the origins of error has been the mis-identification of the nuclei of other species of Mantelliceras, and even of Stoliczkoia, as M. 'martimpreyF (as was pointed out by Pervinquiere in 1907). We figure here (PI. 81, figs. 11-13) examples of Stoliczkaia nuclei from the Dispar Zone of Dorset which could easily be mistaken in this way. Pervinquiere (1907) actually recorded many of his Tunisian specimens of M. martimpreyi as Vraconian (= top Albian), but, from our experience in Algeria and Tunisia, Mantelliceras never appears until well above the disappearance of such diag- nostically Albian genera as Mortoniceras. The careful records of Thomel (1965, 1966) show that the ammonite succession in south-east France probably does not differ from that of southern England, in spite of the fact that Thomel has now fisted M. martimpreyi from undoubted top Albian (Thomel 1968). His ‘zone a Mantelliceras martimpreyi' corresponds in large part with the Hypotwrilites carcitanensis assemblage-horizon and his ‘zone a Mantelliceras man- telli' corresponds in large part with the M. saxbii assemblage-horizon. Acknowledgements. It is a pleasure to acknowledge yet again our indebtedness to Mr. C. W. Wright for allowing us access to his collections, many useful discussions, and for critically reviewing an early draft of this paper. Mr. D. Phillips has very kindly read the final version, and with his usual perspi- cacity has saved us from a number of errors. We are grateful to Dr. M. K. Howarth, Mr. D. Phillips, Dr. J. Fiilop, Dr. A. Horvath, Dr. J. Sornay and his colleagues, and Mr. C. J. Wood for allowing us to examine specimens in their care, and to Dr. N. J. Morris for technical help. We have been helped by the technicians of the Departments of Geology at King’s College (London) and Oxford; in particular we thank Miss H. Cooper for help with the photography. Thanks are due to the Central Research Fund of London University (J. M. H.) and the Natural Environment Research Council (J. M. H., W. J. K.) which have both supported part of this work with grants. REFERENCES ADKINS, w. s. 1928. Handbook of Texas Cretaceous fossils. Univ. Texas Bull. 2838, 1-385, 37 pis. BASSE DE MENORVAL, E. 1959. Lc domaiuc d’influence boreale. C.R. Congr. Soc. sav. Paris, et Depts. — ■ Dijon 1959: Colloque siir le Cretace superieur frangais, 799-814. BUSNARDO, R., ENAY, R., LATREiLLE, G., and ROQUET, p. 1966. Lc crctacc moycn detritique a Cephalo- podes pres de Poncin (Jura meridional). Trav. Lab. Geol. Fac. Sc. Lyon, n.s. 13, 205-228, pis. 12-14. CASEY, R. 1960. Cenomanian ammonite zones. Geol. Mag. 97, 173-175. COLLIGNON, M. 1928-1929. Les cephalopodes du Cenomanien pyriteux de Diego-Suarez. Ann. Paleont. 17 (1928), 139-160, pis. 15-19; 18 (1929), 1-56, pis. 1-2. 1937. Ammonites cenomaniennes du Sud-Ouest de Madagascar. Ann. geol. Serv. Mines, 8, 29-72, pis. 1-11. 1959. Correlations sommaires entre les depots du Cretaee superieur de Madagascar et ceux de I’Europe oecidentale, en particulier de la France. C.R. Congr. Soc. sav. Paris et Depts. — Dijon 1959: Colloque sur le Cretace superieur frangais, 41-52. 1964. Atlas des fossiles caracteristiques de Madagascar (Ammonites), 11 (Cenomanien), xi+ 1-152, pis. 318-375. Tananarive. COQUAND, H. 1862. Geologie et Paleontologie de la region sud de la province de Constantine. Mem. Soc. d' Emulation de la Provence, 2, 341 pp., 35 pis. DiENER, c. 1925. Ammonoidea neocretacea. Fossilium Catalogus 1.- Animalia, 29, 1-244. DUBOURDiEU, G. 1953. Ammonites nouvelles des Monts du Mellegue. Bull. Serv. Carte geol. Alger. ser. 1 (Pal.), 16, 76 pp., 4 pis. KENNEDY AND HANCOCK; MANTELLICERAS 453 DUBOURDiEU, G. 1956. Etude geologique de la region de I’Ouenza (Confins Algero-Tunisiens). Bull. Carte geol. Alger. 10, 1-659, 22 pis., map. FABRE, s. 1940. Le Cretace superieur de la Basse Provence occidentale 1. Cenomanien et Turonien. Ann. Fac. Sci. Marseille, ser. 2, 14, 1-355, 10 pis. HANCOCK, J. M. 1959. Les ammonites du Cenomanien de la Sarthe. C.R. Congr. Soc. sav. Paris et Depts. — Dijon 1959: Colloqite siir le Cretace superieur frangais, 249-252. HYATT, A. 1903. Pseudoceratites of the Cretaceous. Monogr. U.S. geol. Surv. 44, 351 pp., 47 pis. JEFFERIES, R. p. s. 1962. The palaeoecology of the Actinocamax plenus Subzone (Lowest Turonian) in the Anglo-French Basin. Palaeontology, 4, 609-47, pis. 77-79. KENNEDY, w. J. 1969. The Correlation of the Lower Chalk of South-East England. Proc. Geol. Assoc. Lond. 80, 459-560. MANTELL, G. 1822. The Fossils of the South Downs. 327 pp., 42 pis. London. MATSUMOTO, T., MJRAMOTO, T., and TAKAHASHi, T. 1969. Selected Acanthoceratids from Hokkaido. Mem. Fac. Sci. Kyushu Univ. ser. D, 19, 251-296, pis. 25-38. SASTRY, M. V. A., and SARKAR, s. s. 1966. Notes on Some Cretaceous Ammonites from Southern India. 1, Utaturiceras vicinale (Stoliczka) from Southern India, by T. Matsumoto and S. S. Sarkar. Ibid. ser. D, 17, 295-309, pis. 32-33. MiDDLEMiss, F. A., and BROMLEY, R. G. 1962. Field Meeting on the Lower Greensand and Chalk in the Isle of Wight. Proc. Geol. Assoc. Lond. 73, 449-453. MULLER, s. w., and SCHENCK, H. G. 1943. Standard of Cretaceous system. Bull. Amer. Assoc. Petrol. Geol. 27, 262-278. PERViNQUiERE, L. 1907. Etudes de paleontologie tunisienne, I. Cephalopodes des terrains secondaires. Carte geol. Tunisie, 438 pp., 27 pis. 1910. Sur quelques ammonites du Cretace algerien. Mem. Soc. geol. Fr. Pal. 42, 1-86, pis. 1-7. PICTET, F. J., and CAMPiCHE, G. 1858-1860. Descriptions des fossiles du terrain cretace des environs de Sainte-Croix. Mater. Paleont. Suisse, ser. 2, no. 2. RENZ, F. o. 1963. In RENZ, o., LUTERBACHER, H., and SCHNEIDER, A. Stratigraphisch-palaontologische Untersuch ungen im Albien und Cenomanien des Neuenberger Jura, pp. 1089-1112. Eclog. geol. Helv. 56, 1073-1116, 9 pis. SCHLUTER, c. 1871. Die Cephalopoden der oberen deutschen Kreide. Palaeontographica, 21, 1-24, pis. 1-8. SCOTT, G. 1926. Etudes stratigraphiques et paleontologiques sur les terrains cretaces du Texas. Univ. Grenoble These Faculte Sci. 218 pp., 3 pis. SHARPE, D. 1853-1857. Description of the fossil remains of Mollusca found in the Chalk of England, Part 1, Cephalopoda. Palaeontogr. Soc. {Monogr.), 68 pp., 27 pis. SORNAY, J. 1955. Ammonites nouvelles du Cretace de la region des Monts du Mellegue (Constantine). Mem. Pal. Serv. Carte geol. Alger. 18, 1-41, 2 pis. SOWERBY, J. 1814. The Mineral Conchology of Great Britcdn; or coloured figures and descriptions of those remains of Testaceous Animals or Shells, which have been preserved . . . in the earth, 1, pis. 45- 78. London. SPATH, L. F. 1925. On Upper Albian Ammonoidea from Portuguese East Africa, with an Appendix on Upper Cretaceous Ammonites from Maputoland. Ann. Transvaal Mus. 11, 179-200, pi. 28-37. 1926. On the Zones of the Cenomanian and the Uppermost Albian. Proc. Geol. Assoc. Lond. 37, 420-432. 1937. The nomenclature of some Lower Chalk ammonites. Geol. Mag. 74, 277-281. STOLICZKA, F. 1863-1866. The fossil Cephalopoda of the Cretaceous rocks of southern India; Ammo- nitidae. Palaeont. Indica, 1, 41-216 (-f-xiii), pis. 26-94. THOMEL, G. 1965. Zoneostratigraphie et Paleobiogeographie du Cenomanien du sud-est de la France. C.R. 90” Congr. Soc. Sav., Nice 1965, 2, sect. Sc. 127-154. 1966. In PORTHAULT, B., THOMEL, G., and DE viLLOUTREYS, o. 1966. Etude biostratigraphique du Cenomanien du bassin superieur de I’Esteron (Alpes-Maritimes). Le probleme de la limite Ceno- manien-Turonien dans le sud-est de la France. Bull. Soc. geol. Fr. ser. 7, 8, 423-439, pis. 8-11. 1968. A propos de I’apparition precoce de genre Mantelliceras dans le Sud-Est de la France. C.R. Somm. Seanc. Soc. geol. Fr. 1968, 102. 454 PALAEONTOLOGY VOLUME 14 TRUEMAN, A. E., and WEIR, J. 1946. A monograph of British Carboniferous non-marine Lamellibran- chia. Part 1. Palaeontogr. Soc. [Monogr.]. WRIGHT, c. w. 1956. Notes on Cretaceous ammonites. 3. Utaturiceras gen. nov. and the Metoicocera- tinae. Ann. Mag. nat. hist. ser. 12, 9, 391-393. 1957. In ARKELL, w. J. et al.. Treatise on Invertebrate Palaeontology, Pt. L, Mollusca 4, Cephalo- poda Ammonoidea, L80-L465. Univ. Kansas Press and Geol. Soc. Am. and WRIGHT, E. V. 1951. A survey of the fossil Cephalopoda of the Chalk of Great Britain. Palaeontogr. Soc. (Monogr.) 40 pp. YOUNG, K. 1958. Graysonites, a Cretaceous ammonite in Texas. J. Paleont. 32, 171-182, pis. 27-29. W. J. KENNEDY Department of Geology and Mineralogy Parks Road Oxford, 0X1 3PR J. M. HANCOCK Department of Geology King’s College Final typescript received 4 Nov. 1970 London, WC2R 2LS SILURIAN CALYMENID TRILOBITES FROM UNITED STATES, NORWAY, AND SWEDEN by H. B. WHITTINGTON Abstract. The type species of Papillicalymene and Liocalymene are redescribed from type and other material ; each represents a distinct morphological type. Papillicalymetie is known from rocks of Lower Ludlow age in Gotland and Podolia, Liocalymene from late Llandovery strata in central and eastern United States. The type specimen of Calymene breviceps, Waldron Shale, Indiana, is redescribed and specimens of Calymene from the Brownsport Formation, Tennessee, and Ludlow age rocks in Norway briefly discussed. The species celebra from the Racine Dolomite, Illinois, is redescribed and referred to Flexicalymene. The phylogeny of early Ordovician to Devonian calymenids is considered, on the assumption that Shirley’s two groups, one with, the other without, the papillate-buttress structure constitute separate evolutionary stocks. The pattern of evolution is like that in other trilobite families, showing widespread long-ranging genera from which short-ranging forms are derived at different times and in different areas. Morphological characters appear in differing combinations in each of the main stocks; the adaptive value of particular characters is difficult to assess. In 1916 Raymond discussed four species of Silurian calymenids from the United States, illustrating two of them but giving no new figures of the species on which he based his new genus Liocalymene. Among the genera erected by Shirley (1936) was Papillicaly- mene, the type species being Calymene papillata Lindstrom, 1885, a species which neither Shirley nor any subsequent author has redescribed. Thus in preparing an account of calymenids for the Treatise on Invertebrate Paleontology, I gave only brief diagnoses of these two genera and no new illustrations. The present redescriptions are intended to remedy this deficiency, though the available samples of each species are small, they do not reveal all details of morphology, and some specimens are not precisely localized. Nevertheless, in the characters revealed these two type species are morphologically quite distinct, and each is limited in stratigraphical and geographical range, so that their recognition as genera appears valid and useful. Exceptionally well-preserved material of two additional species of Raymond (1916) is also described here, one referred to Ca/v- mene, the other here placed in Flexicalymene, the latter genus better known from Ordo- vician rocks. I have attempted a partial sketch of calymenid phylogeny, and discussed taxonomic characters and their value in particular genera. It is clear that a revision of Silurian species is critical to further understanding of relationships and descent in this family. Terminology of description follows the Treatise of Invertebrate Paleontology (Moore, 1959, pp. 0117-0126), except that here glabella includes the occipital ring and I have preferred to use ‘cheek’ and ‘branch’ of the facial suture. The term ‘anterior border’ is used, and is synonymous with ‘preglabellar field’ of Shirley (1936). I have adopted Campbell’s (1967, text-fig. 4) terms ‘border sector’ and ‘doublure sector’ for the two parts of the rostral plate which are separated by a sharp flexure. I have considered the vertical plane to be parallel to the posterior margin of the occipital or axial ring; use of ‘upward’ or ‘downward’ is with reference to this plane. I have subdivided the Ordo- vician System into Lower and Upper (Whittington, 1966, table 1). [Palaeontology, Vol. 14, Part 3, 1971, pp. 455-477, pis. 83-89.] 456 PALAEONTOLOGY VOLUME 14 Acknowledgements. I am indebted to Dr. C. P. Hughes for critically reading the manuscript, analysing measurements of Papillicalymene, and helping with the illustrations. Dr. V. Jaanusson loaned speci- mens from the Palaeozoological Section, Natural History Museum, Stockhohn (catalogue numbers prefixed by Ar), Dr. A. Martinsson, Uppsala University, supplied information on the Gotland locali- ties, and Dr. D. L. Bruton, Palaeontological Museum, Oslo University, provided a translation of part of Lindstrom (1885) and loaned one specimen (catalogue number prefixed PMO). I am indebted to Dr. G. Arthur Cooper, U.S. National Museum, Washington, D.C., for the loan of material (catalogue numbers prefixed by USNM), and I have also studied material in the Museum of Comparative Zoo- logy, Harvard University (MCZ). Professor Arthur J. Boucot, Oregon State University, kindly supplied information on the probable age of the United States material. TAXONOMIC CHARACTERS AND EVOLUTION OF CALYMENIDAE An outline sketch of phylogeny and geographical distribution of Ordovician caly- menids has been given (Whittington 1966, pp. 716, 726, text-fig. 12). In that account I included Neseurelus and Pharostoma as calymenids, but Dean (1966, pp. 297-298) has advocated placing each of these genera in separate families, admitting that the use of this category rather than subfamily is somewhat arbitrary. Here I put this question aside and discuss only calymenid trilobites of post-Arenig age and their relationships (text- fig. 1). I have omitted the recently proposed Devonian taxon Paracalymene Fillet 1968, regarding it as of doubtful validity, and Protocalymene Ross 1967n, because of the diffi- culty of determining the systematic position of the small specimens on which it is based. Flexicalymene appears in Britain in the lower Llanvirn (Hughes 1969, pp. 81-83, pi. 8, fig. 7; pi. 9, figs. 1, 2, 4, 7, 8) and in North America and Scandinavia in about mid- Caradoc time. As shown below, this genus is widespread in the Upper Ordovician and appears to be present in the middle Silurian. Gravicalymene (not figured in my 1966 diagram) appears in Britain at the base of the Upper Ordovician (if not earlier. Dean 1963, pp. 226-227), in North America in the mid-Caradoc (Ross \961b), and is wide- spread geographically later in the Ordovician. Species referred to this genus are also widespread in the Lower Devonian (see discussion of F. celebra). These two genera, as well as several other Ordovician genera, belong within Shirley’s (1936, p. 394) group A which lacks papillate glabellar lobes or buttresses on the fixed cheeks. In such genera (PI. 87, fig. 5; Evitt and Whittington 1953, pi. 10, fig. 1; Ross 19676, pis. 3-5) the axial furrow is a deep open, steep-walled trench between cheek and glabellar lobes, the furrow narrowest and the walls steepest beside lateral glabellar lobe Ip. The eye ridge may be faintly developed as a broad, low ridge directed forward and inward from the eye lobe across the cheek and down the abaxial wall of the axial furrow, dying out at the base of the furrow (Evitt and Whittington 1953, pi. 10, fig. 1). Shirley’s (1936, p. 395) group B has lateral glabellar lobe 2p papillate, with a buttress projecting from the cheek (i.e. from the abaxial wall of the axial furrow) to meet it (PI. 83, fig. 7; PI. 84, fig. 2). The papillate-buttress structure juts out over the axial furrow, which passes beneath it as a tunnel. Additional papillae and buttresses may be developed in front of the 2p pair, the structures not in all cases being paired, and even if so, not necessarily being in contact at their extremities. Plate 84, figs. 4, 5, illustrate incipient buttressing without corre- sponding papillae in Calymene. Plate 84, figs. 8, 9; Plate 85, figs. 4-6, 10, 11, illustrate the papillate-buttress structure at its maximum known development, and show that these structures jut out over the axial furrow. The latter may be filled with matrix (PI. 84, fig. 9), but when partially freed of matrix by weathering (PI. 86, figs. 1, 3, 7, 8, 11, etc.) WHITTINGTON: SILURIAN CALYMENID TRILOBITES 457 Group A? Group A No buttresses, papillae rare Group B with papillate structure buttress Uncertain Relationship Z < z o > z < E? U > O Q D ■? u ) the preglabellar furrow may be narrow or broad (sag. and exs.), the anterior border is relatively broad (sag. and exs.), stands well above the preglabellar furrow and may have a flattened upper surface. This surface may be approximately horizontal, or may slope forward and upward; at the posterior edge it joins the wall of the preglabellar furrow, at the anterior edge it curves abruptly over into the outer face of the border, which slopes downward and backward. I am not aware of any Silurian species referred to Graviealymene, but a number of Lower Devonian species from Bohemia, North Africa, Turkey, Australia, and New Zealand have been referred to this genus (Shirley 1938, p. 487, pi. 44, fig. 17 ; Philip 1962, p. 231 ; Talent 1963, pp. 105-106; Strusz 1964, pp. 94- explanation OF PLATE 88 Figs. 1-10. Flexicalymene celebra (Raymond, 1916). 1-7, 9, 10. Internal mould of exoskeleton and latex cast from counterpart, USNM 59405, Racine Dolomite, Grafton, Illinois. 2-4, 6, cast, dorsal, anterior, posterior, left lateral views, x2. 1, 5, mould, dorsal, right lateral views, x2. 7, 9, 10, enlarged views of cast to show external surface, x 6, x 6, x 4-5. 8, internal mould of exoskeleton, MCZ 639, original of Raymond 1916, pi. 3, fig. 10, Racine Dolomite, Grafton, Illinois. Outer parts of thoracic segments, showing mould of doublure (d) and pillars infilling canals through exoskeleton, x6. Palaeontology, Vol. 14 PLATE 88 WHITTINGTON, Silurian calymenids WHITTINGTON: SILURIAN CALYMENID TRILOBITES 473 95; Talent 1965, p. 49; Haas 1968, pp. 101-103; Alberti 1969, pp. 413-414). In these species the preglabellar furrow is deep, the anterior border of moderate width (sag. and exs.) and semicircular in cross section, not flattened on the upper surface. Certain Australian authors have expressed doubt that these species are congeneric with Ordo- vician species, and this question remains open — -the Devonian species may be derived from a different ancestor. The present species celebra has the anterior border rising in a flat slope upward and forward from the base of the preglabellar furrow, and at the anterior margin it turns abruptly over to slope downward and backward as the outer face of the border. This shape is typical of Ordovician species of FlexicaJymene (Dean 1962, 1963; Whittington 1965; Ross \961b), as is the lack of 2p papilla and buttress, though the outline of the glabella is slightly bell-shaped rather than parabolic. On balance I prefer to place it in this genus, and the hypostome attributed to it is like that of species of Flexicalymene rather than of Calymeue (Campbell 1967, pi. 10, figs. 1-7; Haas 1968, pi. 29, figs. 6, 7, text-fig. \6b, d,f). Haas (1968, pp. 100-101) has recently described a high Llandovery to Wenlock aged species that he attributes to Flexicalymene, and Temple (1969, pp. 224- 228) an early Llandovery form. The anterior border is not well preserved in Haas’s material, and the cast figured by Temple (1969, pi. 6, fig. 1) has an anterior border ap- parently of Gravicalymene type. Thus celebra is the only Silurian species that I consider with some confidence may be placed in Flexicalymene. Genus liocalymene Raymond 1916 Type species. Hemicryptiirus clintonii Vanuxem 1842. Diagnosis. Glabella bell-shaped in outline, three pairs of lateral glabellar lobes, lobe Ip large, subtriangular in outline, maximum length (exs.) as much as half length (sag.) of glabella, lobes 2p and 3p progressively shorter (exs.) and frontal lobe shorter (exs.) than lobe 3p; lateral glabellar furrows narrow. Axial furrow beside glabella a narrow slit throughout its length, so that cheek, lateral lobes and frontal lobe approach each other closely, but papillae and buttresses (as projections across a broad, deep furrow) are apparently not developed; anterior border low, gently convex; transverse line joining midpoints of palpebral lobes crosses junction of axial and Ip lateral furrow. Hypostome unknown. Thorax of thirteen segments. Pygidium with median part of axis smooth behind third ring; pleural regions smooth except for narrow first pleural furrow; behind axis a short (sag.), slightly swollen band unites the pleural regions. Stratigraphical and geographical range. Rocks of Clinton Group and equivalents, of late Llandovery age, of central and eastern United States (New York, Pennsylvania, Maryland, Virginia, and Ken- tucky). Discussion. This diagnosis is intended to be read in conjunction with my {in Moore 1959, pp. 0450-0451) diagnosis of the family, lines 2-4 of which should read ‘widest across occipital ring or pre-occipital (i.e. Ip) lateral lobes; may or may not project in front of genae’. In erecting this genus Raymond (1916, p. 29) drew attention to the ‘smooth pleural lobes’ of the pygidium. The cephalon, with its relatively large basal glabellar lobe and narrow lateral glabellar and axial furrows, which bring lateral glabellar lobes and cheek close together without development of papillae and buttresses, appears to be 474 PALAEONTOLOGY VOLUME 14 equally distinctive. No species other than the type is known, and its relationships are obscure, for no late Ordovician or early Silurian calymenid exhibiting similar characters appears to have been described. Liocalymene clintoni (Vanuxem 1842) Plate 89, figs. 1-7, 9, 10, 12 1842 Hemicryptiinis clintonii Vanuxem, pp. 79, 80, fig. 11, 2. 1843 Hemicryptiinis, Hall, p. 77, fig. 19, 2; plates, no. 9, 2. 1852 Calymene clintoni (Vanuxem); Hall, p. 298, pi. A66, figs. 5 a, d. 1915 Calymene clintoni (Vanuxem); Bassler, p. 166, giving additional early references. 1916 Liocalymene clintoni (Vanuxem); Raymond 1916, p. 29. 1923 Liocalymene clintoni (Vanuxem); Swartz and Prouty, pp. 706-707, pi. 34, figs. 1-4. 1941 Liocalymene clintoni (Vanuxem); Butts, pi. 104, figs. 16-19. 1959 Liocalymene clintoni (Vanuxem); Whittington, in Moore, p. 0452. Material. Neither the original of Vanuxem or that of HaU (1852) is preserved in the New York State Museum (pers. comm, from Dr. D. W. Fisher, 10 June 1969), and the original of Hall is not in the Field Museum of Natural History, Chicago (pers. comm, from Dr. M. H. Nitecki, 19 June 1969). MCZ 8082 (PI. 89, fig. 1), from the Clinton Group, Clinton, Oneida County, New York State, C.D. Walcott coll., is the specimen mentioned by Raymond in erecting the genus. MCZ 8081 is from the Clinton of Pennsylvania, H. D. and W. B. Rogers coll.; MCZ 8083 from Rose Hill Shale, road cut, U.S. highway 22 and 522 at Mt. Union, Pennsylvania. USNM 154493-154496, figured herein. Crab Orchard Group, upper part, four miles south-east of Tolsboro, western part of Lewis County, Ken- tucky, coll. E. M. Kindle; similar specimens from Mastigobolbina typiis zone, Clinton Group, Wills Creek, Cumberland, Maryland (U.S. Geol. Surv. locality 286i^). USNM 116344, internal mould, entire exoskeleton, from west portal of first tunnel on Pennsylvania turnpike, Gunther Valley, east of Roxbury, Pennsylvania. Geological horizon. Fisher (1960) has recorded the range of this species as mid-part of the Clinton Group, late Llandovery age, in New York State. Other material I have examined is from rocks referred to the Clinton Group, and that described from Maryland (Swartz and Prouty 1923) is from the M. typus zone of the highest Rose Hill Formation and the overlying Keefer Sandstone member of the Rochester Formation. These beds are regarded by Professor A. J. Boucot (pers. comm.) as of late Llandovery age. Description. Glabella of maximum width (tr.) across basal part of Ip lobes, in large cranidia this width is X 1-3 the length (sag.); gently convex. Occipital ring longest (sag.) medially, becoming progressively shorter (exs.) distally, and constricted adjacent to the shallow axial furrow where lateral glabellar lobe Ip bulges posteriorly. Occipital furrow broad and shallow medially, deep, narrow and sigmoidal behind lateral lobe Ip. Latter subtriangular in outline, in large cranidia (PI. 89, figs. 10, 12) of maximum length (exs.) EXPLANATION OF PLATE 89 Figs. 1-7, 9, 10, 12. Liocalymene clintoni (Vanuxem 1842). 1, internal mould of exoskeleton, MCZ 8082, Clinton Group, New York State. Dorsal view, x2. 2-7, 9, 10, 12, latex casts from external moulds. Crab Orchard Group, Kentucky. 2, 4, incomplete pygidium, USNM 154496, dorsal, pos- terior views, x3. 3, 5, incomplete pygidium, USNM 154495, dorsal, right lateral views, X3. 6, incomplete cranidium, USNM 154496, dorsal view, x4-5. 7, 9, 10, incomplete cranidium, USNM 154493, anterior, left lateral, dorsal views, X3. 12, incomplete cranidium, USNM 154494, dorsal view, X 3. Figs. 8, 1 1. Flexicalymene celebra (Raymond 1916)? External mould of part of dorsal exoskeleton with internal mould of hypostome in place, USNM 154492, Racine Dolomite, Racine, Wisconsin. Oblique and ventral views, x 3. Palaeontology, Vol. 14 PLATE 89 WHITTINGTON, Silurian calymenids m . ■•T" ■I ;irf. ■ V' I WHITTINGTON: SILURIAN CALYMENID TRILOBITES 475 about half length (sag.) of glabella, defined by narrow, deep furrow Ip which has a slightly sigmoidal course, directed at about 45° to the sagittal line, adaxially curving to run inward and slightly forward; gently convex, standing above narrow (about one- quarter the maximum width of the glabella), depressed portion of median lobe which separates lobes Ip. Lateral lobe 2p defined anteriorly by straight, narrow, inward, and slightly backwardly directed furrow 2p which extends adaxially one-fifth to one-quarter width at this point ; lobe 2p not inflated or separated from median lobe. Lateral lobe 3p defined anteriorly by straight, narrow, inwardly directed furrow 3p which is shorter than furrow 2p; lobe 3p slightly inflated. Anterior margin of glabella curved gently convexly forward, abaxially frontal lobe of length (exs.) less than length of lobe 3p. Course of axial furrow sigmoidal, curving abaxially anteriorly; wide and moderately deep beside occipital ring and most basal part of lobe Ip, in front of here a narrow sht so that cheek and glabella are closely juxtaposed. Preglabellar furrow broad, moderately deep, an- terior border widest (exs.) abaxially, moderately convex and gently curved transversely. Cheek gently convex, posterior border longest (exs.) abaxially, antero-lateral border poorly known; eye lobe moderately convex, midpoint on a transverse line that passes approximately through junction of axial and Ip furrow. Anterior branch of suture runs straight forward and inward to cross border furrow and curve inward over border just outside the line of the axial furrow; rostral suture traverses outer surface of anterior border. Posterior branch of suture curves sigmoidally outward and backward to the genal angle. Rostral plate, doublure, and hypostome unknown. Thorax of thirteen segments. Axial rings longest (sag.) medially, abaxially curving forward and slightly inflated. Inner part of pleura horizontal, outer part curves down to become vertical distally; pleural furrow broad, slightly diagonal, narrowing and dying out beside large facet. Axis of pygidium divided by articulating furrow and eight ring furrows, the posterior one-third smooth, gently convex; first and second ring furrows complete, but median part of second shallow, third and succeeding furrows show a pro- gressively wider smooth median zone, and the abaxial remnants of the furrows are directed progressively more strongly backward as well as outward. Pleural region with large facet and narrow, shallow first pleural furrow ending against it; no further pleural or interpleural furrows; behind axis a short (sag. and exs.), slightly inflated band joins the pleural regions. Doublure of thorax and pygidium unknown. External surface may be smooth, tubercles lacking; one cranidium (PI. 89, fig. 12) suggests the presence of fine granulation on occipital ring, glabella and fixed cheek. Discussion. Examination of calymenids in the U.S. National Museum and at the Museum of Comparative Zoology did not reveal any other North American species like Liocalymene clintoni. The material used herein is limited and not adequate to reveal indi- vidual variation or changes in proportion with size, so that ratios of dimensions given refer to the largest cranidia only. REFERENCES ALBERTI, G. K. B. 1969. Tfilobiten des jiingeren Siluriums sowie des Unter- und Mitteldevons. 1. Abli. Senckenb. naturforsch. Ges. 520, 1-692, pis. 1-52. BASSLER, R. s. 1915. Bibliographic Index of American Ordovician and Silurian Fossils. Bull. U.S. Nat. Mas. 92, vol. 1, pp. viii, 1-718, vol. 2, pp. iv, 719-1521, pis. 1-4. 476 PALAEONTOLOGY VOLUME 14 BUTTS, c. 1941. Geology of the Appalachian Valley in Virginia. Bull. Virginia geoL Siirv. 52, vol. 1, pp. xxxii+568, vol. 2, pp. iv+271, pis. 1-135. CAMPBELL, K. s. w. 1967. Henryhousc trilobites. Bull. Oklahoma geol. Surv. 115, 1-68, pis. 1-19. DEAN, w. T. 1962. The trilobites of the Caradoc Series in the Cross Fell Inlier of Northern England. Bull. Br. Mus. nat. Hist. (Geol) 7, 65-134, pis. 6-18. 1963. The Ordovician trilobite faunas of South Shropshire, III. Ibid. 7, 213-254, pis. 37^6. 1 966. The Lower Ordovician stratigraphy and trilobites of the Landeyran valley and neighbouring district of the Montagne Noire, south-western France. Ibid. 12, 245-353, pis. 1-21. EViTT, w. R., and Whittington, h. b. 1953. The exoskeleton of Flexicalymene (Trilobita). J. Paleont. 27, 49-55, pis. 9, 10. FISHER, D. w. 1960. Correlation of the Silurian rocks in New York State. New York State Mus. Sci. Serv., geol. Surv., map and chart Series, 1. HAAS, w. 1968. Trilobiten aus dem Silur und Devon von Bithynien (N.W.-Tiirkei). Palaeontographica, A130, 60-207, pis. 26-37. HALL, J. 1843. Geology of New York, IV, Survey of the Fourth Geological District. Nat. Hist. New York, pp. xxii+683, pis. 1-19. 1852. Palaeontology of New York, vol. 2. Ibid., pt. 6, pp. viii+362, pis. 1-84. HEDE, J. E. 1921. Gottlands SUurstratigrafi. Sverig. geol. Unders. C, 305, 1-100. 1960. The Silurian of Gotland. 21st Int. Geol. Cong. Norden, guide book for excursions A22 and Cl 7, pp. 44-89. HUGHES, c. p. 1969. The Ordovician trilobite faunas of the Builth-Llandrindod Inher, Central Wales. I. Bull. Br. Mus. nat. Hist. (Geol.) 18, 39-103, pis. 1-14. KOBAYASHi, T. 1951. On the Ordovician trilobites in Central China. /. Fac. Sci. Univ. Tokyo, Sec. II, 8, 1-87, pis. 1-5. 1960. Some Ordovician fossils from east Tonkin, Viet Nam. Japan. J. Geol. Geogr. 31, 39-48, pi. 5. KUMMEROw, E. 1928. Beitragc zur Kenntnis der Fauna und der Herkunft der Diluvialgeschiebe. Jahrb. Preuss. geol. Landes. 48, 1-59, pis. 1-2. LINDSTROM, G. 1885. Gotlands Siluriska Crustaceer. Ofv. K. Vetensk. Akad. Fdrhh, Stockholm 6, 37- 100, pis. 12-16. MARTiNSSON, A. 1962. Ostracodcs of the Family Beyrichiidae from the Silurian of Gotland. Bull. geol. Instn. Univ. Uppsala, 41, 1-369. 1967. The succession and correlation of ostracode faunas in the Silurian of Gotland. Geol. Fdren. Stockholm Forh. 89, 350-386. MAXIMOVA, z. A. 1967. Late Silurian and early Devonian trilobites of central Kazakhstan. International Symposium on the Devonian System, Calgary, 2, 777-787. 1968. Middle Palaeozoic trilobites of Central Kazakhstan. Trudy Vses. nauch.-issled. geol. inst., 165, pp. 1-208, pis. 1-35 (in Russian). MOORE, R. c. (ed.) 1959. Treatise on Invertebrate Paleontology. Part O. Arthropoda 1. Geol. Soc. Am. and Univ. Kansas Press. NIKIFOROVA, o. L, and PREDTECHENSKiJ, N. N. 1968. Third International Symposium on Silurian- Devonian boundary and Lower and Middle Devonian Stratigraphy. A guide to the geological excursion on Silurian and Lower Devonian deposits of Podoha (Middle Dnestr river). VSEGEI, Leningrad, pp. 1-58, figs. 1-30. PHILIP, G. M. 1962. The palaeontology and stratigraphy of the Siluro-Devonian sediments of the Tyers area, Gippsland, Victoria. Proc. R. Soc. Victoria, 75, 123-246, pis. 11-36. PILLET, J. 1968. Les Calymene devoniens d’Europe et d’Afrique du Nord. Ann. Paleontol. Invert. 54, 67-89, pis. A-H. RAYMOND, p. E. 1916. New and old Silurian trilobites from southeastern Wisconsin, with notes on the genera of the Illaenidae. Bull. Mus. Comp. Zool. {Harvard), 60, 1-41, pis. 1-4. RICHTER, R., and RICHTER, E. 1954. Die Trilobiten des Ebbe-Sattels. Abh. Senckenb. naturforsch. Ges. 488, 1-76, pis. 1-6. ROSS, R. J. 1967fl. Some Middle Ordovician brachiopods and trilobites from the Basin Ranges, western United States. U.S. geol. Surv. Prof. Paper, 523D, 1-43, pis. 1-11. 19676. Calymenid and other Ordovician trilobites from Kentucky and Ohio. Ibid. 583B, 1-19, pis. 1-5. WHITTINGTON: SILURIAN CALYMENID TRILOBITES 477 SCHRANK, E. 1970. Calymcniden (Trilobita) aus silurischen Geschieben. Ber. Deutsch. Ges. geol. fViss., A, Geol.-Paldont. 15, 109-146, pis. 1-12. SHIRLEY, J. 1933. A redescription of the known British SUurian species of Calymene (s.l.). Mem. Proc. Manchester lit. phil. Soc. 77, 51-67, pi. 1. 1936. Some British trilobites of the Family Calymenidae. Q. Jl geol. Soc. Loud. 92, 384-422, pis. 29-31. 1938. The fauna of the Baton River beds (Devonian), New Zealand. Ibid. 94, 459-506, pis. 40-44. STRUsz, D. L. 1964. Devonian trilobites from the Wellington-Molong district of New South Wales. /. Proc. R. Soc. N.S.W. 97, 91-97, pi. 1. SWARTZ, c. K., and prouty, w. f. 1923. Trilobita, in Systematic palaeontology of Silurian deposits, in Maryland geol. Surv., Silurian, pp. 1-794, pis. 1-67. TALENT, J. A. 1963. The Devonian of the Mitchell and Wentworth rivers. Mem. geol. Surv. Victoria, 24, 1-118, pis. 1-78. 1965. The Silurian and early Devonian faunas of the Heathcote district, Victoria. Ibid. 26, 1-55, pis. 1-27. TEMPLE. J. T. 1969. Lower Llandovery (Silurian) trilobites from Keisley, Westmorland. Bull. Br. Mus. nat. Hist (Geol.) 18, 197-230. pis. 1-6. TILLMAN, c. G. 1960. Spatliacalymene, an unusual new Silurian trilobite genus. J. Paleont. 34, 891-895, pi. 116. TOMCZYKOWA, E. 1970. Silurian Spatliacalymene Tillman, 1960 (Trilobita) of Poland. Acta palaeont. pol. 15, 63-94, pis. 1-5. VANUXEM, L. 1842. Geology of New York, III, Survey of the Third Geological District. Nat. Hist. New York, pp. 1-306. WHITTINGTON, H. B. 1954. Ordovician trilobites from Silliman’s Fossil Mount. In miller, a. k., young- QUiST, w., and collinson, c., Ordovician cephalopod fauna of Baffin Island. Mem. geol. Soc. Amer. 62, 119-149, pis. 59-63. 1956. Silicified Middle Ordovician trilobites: the Odontopleuridae. Bull. Mus. comp. Zool. (Har- vard), 114, 155-288, pis. 1-24. 1965. The Ordovician trilobites of the Bala area, Merioneth. Part II. Palaeontogr. Soc. [Mongr.], 33-62, pis. 9-18. 1966. Phylogeny and distribution of Ordovician trilobites. J. Paleont. 40, 696-737. 1971. A new calymenid trilobite from the Maquoketa Shale, Iowa. In dutro, j. t. jr. (ed.). Paleozoic Perspectives: A Paleontological Tribute to G. Arthur Cooper. Smithsonian Contrib. Paleobiol. 3, 129-136, pis. 1, 2. H. B. WHITTINGTON Department of Geology Sedgwick Museum Typescript received 17 July 1970 Cambridge DEVELOPMENT OF GLYPTOGRAPTUS HUDSONI SP. NOV. FROM SOUTHAMPTON ISLAND, NORTH-WEST TERRITORIES, CANADA by D. E. JACKSON Abstract. The development of Glyptograptus hiidsoni sp. nov. is described based upon specimens isolated from an argillaceous limestone of Late Ordovician age on Southampton Island. The material described was obtained from a slab of limestone collected by Dr. S. J. Nelson on Southampton Island in 1966. The horizon was referred to by Nelson {in Nelson and Johnson 1966) as the ‘Oil Shale Horizon’ which he considered to be of latest Ordovician age. Associated with the described material was Amplexograptus sp. indet. as well as several pygidia of IPseudogygites. The graptoloids were freed from the argillaceous calcilutite using dilute (5 %) hydro- chloric acid and those specimens which carried clay particles benefited by treatment with hydrofluoric acid. Early growth stages were found to be naturally translucent and were mounted directly in ‘Plastimount’ on concave glass slides whereas the more advanced growth stages required clearing in Schulze’s Solution prior to being studied. All speci- mens described (UA907-924) are in the possession of the Department of Geology, Uni- versity of Alberta, Edmonton. Acknowledgements. The author expresses warm thanks to Dr. S. J. Nelson for originally submitting the material to the author and for subsequently providing pertinent stratigraphic and geographic data and to G. W. Cormie for preparing microtome thin-sections. Invaluable laboratory assistance was made possible through National Research Council of Canada grant A-2631. The author is indebted to Professor O. M. B. Bulman for valuable discussions on graptolite morphology and to Professor T. S. Westoll for his assistance and interest in this work which was completed during a sabbatical leave at the University of Newcastle upon Tyne. SYSTEMATIC PALAEONTOLOGY Order graptoloidea Lapworth 1875 Family diplograptidae Lapworth 1873 Genus glyptograptus Lapworth 1873 Glyptograptus hiidsoni sp. nov. Plate 90, figs. 1-6; text-figs. 1-3 Diagnosis. Rhabdosome small, aseptate, 1T-L2 mm wide across th H-th P and 2-0 mm across th 6Mh 6^. Sicula 1 -3 mm long, furnished with conspicuous virgella and a pair of apertural spines. Thecae have pronounced glyptograptid profile; th H carries subapertural spine; apertural margins, with lappets except on th U; there are 7 thecae in 5 mm. Development prosoblastic with th P rather straight and reclined. [Palaeontology, Vol. 14, Part 3, 1971, pp. 478-486, pi. 90.] JACKSON: DEVELOPMENT OF GLYTOGRAPTUS HUDSONI SP. NOV. 479 Holotype. UA 924, text-fig. 1a. Paratypes. UA 907-923. Material. More than 200 isolated growth stages and fragmented rhabdosomes are available. Type locality and horizon. Slab of limestone probably from base of Oil Shale interval of Nelson (in Nelson and Johnson 1966, p. 567); field designation N 64-8a; near East Bay, Southampton Island; collected by Dr. S. J. Nelson in 1966. Uppermost 50 ft of Ordovician at this locality consists of inter- bedded ‘oil shale and limestone’ and is considered to be Richmond in age. Derivation of name. The species is named for the navigator Henry Hudson who entered Hudson Bay in 1610. Description. Rhabdosome small, not exceeding 5-5 mm in length, widening from 1-1 to 1-2 mm across first pair of thecae to 1-4 to 1 -5 mm across fourth pair to 2-0 mm distally. Rhabdosome aseptate, rectangular in cross-section, and slightly depressed axially, usually exhibits prominent virgula. Sicula of diplograptid type, straight, 1-30 mm long and 0-40 mm in diameter across aperture, furnished with a conspicuous virgella 0-35 mm long, and a pair of apertural spines 0-20-0-30 mm long directed obhquely downward and outward. Apex of sicula extends to between 3rd and 4th pairs of thecae (see text- fig. 2c). The free ventral wall of th P carries a spine about 0-20 mm below aperture. No spines exist on subsequent thecae. Free ventral walls of proximal thecae have pronounced glyptograptid curvature be- coming less marked distally ; thecae overlap one-third to one-half. Apertural margins of th P and th P straight, whereas subsequent thecae have margins slightly excavated at dorsal edge; margins distinctly lip-like in mature thecae. There are 7 thecae in 5 mm. Blister-like abnormalities of the cortical tissue are not uncommon among the larger specimens and tend to be localized on the obverse and reverse sides of the proximal end of the rhabdosome (text-fig. 1a). They may have a parasitic origin, on the other hand Urbanek (1958, p. 36) considered similar abnormalities in Monograptus haupti Kiihne to be a feature of old age. Development of rhabdosome Prosicula. The prosicula is a flask-shaped body comprising the nema, the narrow hollow neck (nema prosicula) and the parallel-sided distal part (PI. 90, figs. 1-6). The prosicula is 450-570 pm long exclusive of nema and has a maximum diameter of 170-220 ^m in mid-length. Prior to budding of the first theca, the nema is merely 100 |U.m long and is probably a hollow structure. It is not uncommon to find the prosicula lacking a nema which usually breaks off immediately above the ends of the longitudinal fibres. In no instance could the author observe the longitudinal fibres extending proximally to form or to reinforce the nema. Instead, these fibres appear to terminate in the neck as shown in text-fig. iG (cf. Barrass 1954, fig. 12). The prosicula is constructed of a brown transparent sheath of structureless periderm which is thinnest just below the neck and thickest at the distal extremity. In about one- fifth of the specimens this sheath contains the spiral fine or schraubenlinie (see PI. 90, fig. 1) having approximately twelve rotations in either a right-handed or left-handed sense and individual spirals more widely spaced just below the neck. Longitudinal fibres are always visible in early growth stages and consist of four or five long primary fibres which are formed contemporaneously with the secretion of the 480 PALAEONTOLOGY VOLUME 14 sheath, and several shorter secondary fibres wliich are deposited between the primaries somewhat later. Prosiculae are seen with as few as four and as many as thirteen fibres. Those prosiculae with only four to eight fibres show no metasicular growth and are im- mature with respect to development of these strengthening fibres since measurements suggest that each sicula had the full complement of fibres present by the time metasicular growth began. The longitudinal fibres all terminate just a little above the aperture of the prosicula but, whereas the primary fibres extend proximally almost to the top of the nema prosiculae (text-fig. 1g), secondary fibres fade out at various levels prior to reach- ing the neck (cf. Cox 1933, figs. 1-5; Barrass 1954, fig. 12). A dark brown line at the base of the neck (see PI. 90, fig. 2) is seen in several early growth stages and may constitute an apical diaphragm (= membrane of Kraft 1926). No further light can be shed on whether or not this feature is a transverse partition. Length and width measurements of 49 prosiculae indicate that the prosicular sheath has no growth stages thus supporting the idea that it was secreted fully grown as sug- gested by Kozlowski (1966). Metasicula. The metasicula is subcyhndrical about 0-80 mm long, widening from 0-20 mm diameter proximally to 0-40 mm distally. There are 60-70 fuselli in the mature metasicula and these bands are about 0-01 mm wide at the distal end. The metasicular aperture is adorned by a stiff virgellar spine 0-40-045 mm long and a pair of apertural spines 0-25-0-30 mm long which are directed downward and outward. The development of the metasicula begins with the accretion of several superposed partial fuselli which form the virgella (see PI. 90, figs. 2, 3, 4). These partial fuselli are joined along a zigzag line and laterally overstep foregoing fuselli to pinch out on the rim of the aperture of the prosicula. After several such fuselli have been formed on either side of the zigzag fine subsequent fuselli encircle the rim of the prosicula and meet in a zigzag line on the dorsal side. The virgella continues to elongate and is transformed into a spine after fifteen to twenty complete fuselli have been deposited (see PI. 90, fig. 6). At a distance of approximately ten fuselli above the metasicular aperture, growth is retarded on the dorsal side of the sicula and a notch forms between a pair of denticles that later develop into the apertural spines (see text-figs. Id, e). Theca T. Growth of the metasicula ceases with the appearance of paired apertural spines opposite the virgella and the next important development is the formation of the fora- men of the initial bud. The resorption foramen is circular or oval, and about 80 ixm in diameter; it is positioned on the reverse side of the virgella two-thirds of the way down the metasicula (see text-fig. 1b). Initially the bud takes the form of a spht tube ankylosed to the sicula growing down- ward along the virgella (see text-fig. Ic) becoming tube-hke as it turns out from the sicular aperture (text-fig. Id). A little below the sicular aperture, direction of growth of TEXT-FIG. 1a-g. Glyptograptus hudsoni sp. nov. A, reverse view of holotype UA 924, with epicortical blister-like abnormalities, magnification x 22. b-f, reverse views of early growth stages ; b, sicula with foramen in metasicula UA 909, x 47 ; c, sicula with initial bud of th 1 UA 910, x 47 ; d, development of th P low down on th U, UA 911, x47; e, further growth of th U, UA 912, x47; f, early develop- ment of th P and th 2^ by diagonal build up of fuselli at base of U-shaped bend in th P, UA 913, x 45; G, prosicula of UA 908 with primary fibres terminating in nema prosicula, x 1 37. JACKSON: DEVELOPMENT OF GLYPTOGRAPTUS HUDSONI SP. NOV. 481 482 PALAEONTOLOGY VOLUME 14 th P is outward then upward to form a U-shaped bend (text-fig. 1e); when mature, the aperture of th P is about level with the initial bud. Approximately 0-20 mm below the aperture, the free ventral wall carries a spine 0-20-0-30 mm long. This spine has a hollow base and is full grown upon completion of growth of th P. Theca T. This theca develops from the obverse side of th P (text-fig. 3 section 18), grows horizontally toward the reverse side through the U-shaped bend of th P and in so doing bridges the outer side of the initial bud and the inner side of the distal portion of first theca, thus forming a conspicuous fenestrule (text-fig. 1e). Further growth is achieved by deposition of bands along the inner edge of th P commencing in the U-bend (see text-fig. If) and building diagonally upwards to become ankylosed with the wall of the metasicula. Th P crosses the reverse side of the sicula just above the aperture and be- comes tube-like just before it turns upwards at 45° to project slightly beyond the sicula. The apertural margin is slightly thickened and inclined obliquely outward. Theca 2^. Theca 2^ is an outgrowth of the proximal part of th P (see text-fig. 2c), and is rather erect and straight, the free wall exhibits glyptograptid curvature and the aper- tural margin has a slight excavation dorsally. Growth of this theca is entirely on the virgellar side of the sicula but in UA 914 can be seen to onlap on to the metasicular wall (see text-fig. 2a). Theca 2^. After the deposition of a dozen or so fuselli of th 2^ the bands separate to form two discrete thecal tubes ; th 2^ proceeds to grow diagonally upwards across the reverse side of the sicula (text-fig. 2c) and is approximately half grown when growth of th 2^ is complete. The free ventral wall exhibits glyptograptid curvature, overlap by th 3^ is about one-quarter, and the apertural margin is undulatory with a thickened rim and a slight excavation at the dorsal edge. Theca 3^ and later thecae. The initial development of th 3^ is entirely on the reverse side of the sicula, as it is for all earlier thecae, and arises through the budding of th 2^ to form a crossing canal. As the theca elongates the obverse wall progressively onlaps the sicula (see text-fig. 2b) and this finally results in the apex of the sicula being completely enveloped by th 3^ and 4^. The free ventral walls of th 3^ and subsequent thecae have a glyptograptid profile and the apertural margins are concave to shghtly undulating, with thickened rims. The development of subsequent thecae is by alternate budding as in all aseptate diplo- graptids (see text-fig. 2d). Virgula. The virgula is a hollow rod of 30 fim diameter which extends from the apex of the sicula to beyond the distal extremity of the rhabdosome. Beyond th 3^ it lies in the common canal and is without supporting structures. TEXT-FIG. 2a-d, early growth stages of Glyptograptus hudsoni sp. nov. A, b. Obverse views of UA 914 and UA 915, magnifications x47 and x44 respectively, c. Reverse view of UA 916, magnification X 42. AU three specimens have been rendered transparent by chemical treatment, d. Schematized illustration of development of two thecal series in Glyptograptus hudsoni sp. nov. as deduced from UA 916. C 8216 484 PALAEONTOLOGY VOLUME 14 Discussion. Bulman (1963) has indicated that early Ordovician diplograpti of the Glyp- tograptus austrodentatus group have a proximal-end development whieh exhibit atavistic characteristics, namely, a strong downward component of th P, th 2^ and oeeasionally th 2^. For this type of development Bulman (1963, p. 671) proposed the term strepto- blastic. From it is derived the prosoblastic type charaeterized by the tendeney to straighten out th P as exemplified by the G. dentatus group which appears to be ancestral to most of the later Ordovician glyptograptids, orthograptids, and climaco- graptids. This trend in proximal-end development has a tendency to be aecompanied by a deferment in the development of the dicalycal theea resulting ultimately in an aseptate rhabdosome. The late Ordovician diplograptids Amplexograptus inuiti (Cox) and G. hudsoni n. sp. represent end members of these two separately operating trends. By comparison with the development details that have been worked out for other diplograptids some specific differences and similarities are noted. The appearance of the virgella during the earhest developmental stages of the metasicula of Glyptograptus hudsoni n. sp. finds a parallel in Amplexograptus inuiti (Cox 1933), A. elongatus and A. prominens Barrass 1954. In contrast to this is the delayed formation of the virgella in A. cf. nmxwelli Decker (in Walker 1953) and Climacograptus alf. scalaris (Hisinger) in Barrass 1954. The manner in which th F grows horizontally through the crook in the U-shaped bend of th F and the erect nature of th 2^ (see text-fig. 2d) can be likened to the develop- ment of Amplexograptus inuiti and is quite different from that of Orthograptus gracilis (Roemer) in Kraft (1926) where the initial growth of th F has a downward component thus producing an X-shaped arrangement of fuselli. Comparison. The presence of four proximal sicular and thecal spines together with the laek of a median septum renders Glyptograptus hudsoni sp. nov. distinct from other European glyptograptids. A closely allied species probably exists in G. lorrainensis (Ruedemann) from late Middle to early Upper Ordovician of eastern North America. Using Riva’s (1969, fig. 6d-f) concept of the species as a basis for comparison in pref- erence to Ruedemann’s (1947) inadequate illustrations, we see that G. lorrainensis is slightly narrower and the thecal apertures are horizontal whereas they are slightly in- troverted in G. hudsoni n. sp. Perhaps even more closely allied to, if not conspeeific with, G. hudsoni is a new species of Glyptograptus figured by Riva (1969, fig. 61) from the English Head Formation, Antieosti Island. This glyptograptid has the same width, thecal spacing, and the slightly introverted thecal apertures seen in G. hudsoni sp. nov. Only the laek of a subapertural spine on th P makes the comparison a priori less than perfect. This comparison appears especially satisfactory from a stratigraphical viewpoint because Riva (1969, table 1) EXPLANATION OF PLATE 90 Figs. 1-6. Siculae of Glyptograptus hudsoni sp. nov. 1, prosicula with longitudinal fibres and spiral line, no metasicular growth, UA 917, magnification X 133. 2, dark line in neck of prosicula may be apical diaphragm, metasicula damaged, UA 918, magnification Xl41. 3, metasicular growth commences with local build-up of partial fusellar bands to form virgella, UA 919, magnification X 143. 4, 5, 6, further growth stages of metasiculae, respectively UA 920, magnification x 141 ; UA 921, magnification X 143; UA 922-, magnification X 146. Palaeontology, Vol. 14 PLATE 90 JACKSON, Glyptograptus ] JACKSON: DEVELOPMENT OF G LYPTOG RAPTU S HUDSONI SP. NOV. 485 30 , 27 75 thl-i 70 thf thl2 QsdCXy 'CO- 0 (£0::) ■^17 14 • 37 thl' . 9 35 » 5 r- • 30 V , 1 is TEXT-FIG. 3. Selected microtome thin sections of Glyptograptiis hiidsoui sp. nov., UA 923, magnifi- cation X 18-J-, 1 unit = 20 pm. Sections oriented with obverse side of rhabdosome facing upward, as = apertural spines, is = interthecal septum, f = foramen at origin of th P, r = reverse aper- tural spine, s = sicula, ss = sub-apertural spine, v = virgella. 486 PALAEONTOLOGY VOLUME 14 records his glyptograptid as ranging through the Upper Ordovician zones of Climaco- graptus manitoulinensis and Dicellograptus complamtus into his ‘zone of Climacograptus prominens-elongatus' which he (1969, p. 551) considers to be post-Ashgillian but pre- Llandoverian. REFERENCES BARRASS, 1954. Graptolites from Anticosti Island. Q. Jl geol. Soc. Loud. 110, 55-75. BULMAN, o. M. B. 1963. On Glyptograptus dentatus (Brongniart) and some allied species. Palaeontology, 6, 665-689, pis. 96, 97. cox, I. 1933. On Climacograptus iniiiti sp. nov. and its development. Geol. Mag. 70, 1-19, pis. 1, 2. KOZfcowsKi, R. 1966. On the structure and relationships of graptolites. /. Paleont. 40, 489-501. KRAFT, p. 1926. Ontogenetische Entwicklung und Biologie von Diplograptiis und Monograptus. Palaont. Zeitschr. 7, 207-249, pis. 3-17. NELSON, s. J., and Johnson, r. d. 1966. Geology of Hudson Bay Basin. Bull. Canadian Petrol. Geol. 14, 520-578. RiVA, J. 1969. Middle and Upper Ordovician Graptolite Faunas of St. Lawrence Lowlands of Quebec, and of Anticosti Island. North Atlantic-Geology and Continental Drift. A symposium. Amer. Assoc. Petrol. Geol. Memoir, 12, 513-556. RUEDEMANN, R. 1947. Graptolite of North America. Geol. Soc. America Memoir, 19. URBANEK, A. 1958. Monograptidae from erratic boulders of Poland. Palaeont. polon. 9, 1-105, 5 pis. WALKER, M. 1953. The development of a diplograptid from the Plattevtlle Limestone. Geol. Mag. 90, 1-16. D. E. JACKSON Department of Earth Sciences Open University Revised typescript received 28 July 1970 Walton, Bletchley THE RESSERELLINAE— A NEW SUBFAMILY OF LATE ORDOVICIAN TO EARLY DEVONIAN DALMANELLID BRACHIOPODS by V. G. WALMSLEY and A. J. BOUCOT Abstract. A new subfamily Resserellinae of the Dalmanellidae is proposed to include Resserella Bancroft, Fascicostella Schuchert and Cooper, Dedzetina Havlicek, and Visbyella Walmsley, Boucot, Harper, and Savage. It is considered that Dedzetina (late Ordovician) gave rise to Visbyella (late Llandoverian to Wenlockian) and Resserella (late Llandoverian to Emsian). R. sefinensis sp. nov. (late Llandoverian, Cx) is the oldest Resserella known. Four lineages of Resserella are recognized: (1) R. concavoconvexa (late Llandoverian, Cj-Ce, to early Wenlockian) from Anticosti, ?New Brunswick, ?Quebec, ?South Wales, and Estonia; R. elegantulina (early Wenlockian), Shropshire; R. basalis (Wenlockian) of Gotland and the Welsh Borderland; and R. waldronensis (Wenlockian) of Ohio. (2) R. springfieldensis (late Llandoverian, C4-C6) of Podolla, (early Wenlockian) of Britain, (Wenlockian) of Ohio; R. canalis (Wenlockian and Ludlovian) of Gotland and Britain; R. elegantula (Wenlockian) of Gotland; and R. logansportensis sp. nov. (Pridolian) of Indiana. (3) R. brownsportensis (late Wenlockian) of South Wales and (Ludlovian) of Tennessee; R. amsdeni sp. nov. (Ludlovian) of Oklahoma; R. elegantidoides (Gedinnian) of Podolia and Nevada; and R. triangularis (Emsian) of Rhineland and Kazakhstan. (4) R. crassicostata (Ludlovian) of Tennessee and Oklahoma ; R. impensa (Siegenian) of Victoria, Australia ; and R. pragensis (Emsian) of Bohemia. The latter lineage is considered to have given rise in the Gedinnian to Fascicostella gervillii (Gedinnian to Eifelian) of Europe and North Africa, F. imdulata sp. nov. (early Devonian) of Belgium and F. batonensis sp. nov. (Siegenian) of New Zealand. The genera and species of dalmanellid brachiopods here described have a sufficiently discernible natural affinity to permit their being gathered together into a separate sub- family. The new subfamily Resserellinae is proposed, to encompass the genera Res- serella Bancroft 1928, Fascicostella Schuchert and Cooper 1931, Dedzetina Havlicek 1950, and Visbyella Walmsley, Boucot, Harper, and Savage, 1968. This paper attempts to clarify the morphology, taxonomy, stratigraphic distribution, phylogeny, and geographic distribution of these genera and their member species so as to render the group more useful stratigraphically. During a study made by Walmsley in 1963 at the California Institute of Technology, of dalmanellid material collected by Boucot from many locahties in Gotland, Britain, Oklahoma, Tennessee, and Indiana, several species of Resserella were recognized in- cluding three new ones. Type and figured material was then borrowed from museums in an attempt to describe the remaining Resserella species. During a visit to the Sencken- berg Museum, Frankfurt in 1965, Walmsley examined Early Devonian dalmanelhd material and recognized Orthis triangularis Maurer as a resserelhd. Maurer’s original material was examined at the Hessisches Landesmuseum Darmstadt in 1966. Study of Dalman’s material at the Natural History Museum, Stockholm, enabled Walmsley to select the lectotype of R. basalis. Boucot provided locality and stratigraphic data for his collected material and general guidance during the laboratory study and in the writing of this paper. There has been much confusion over some of the species and the genus Resserella here assigned to the new subfamily Resserellinae. Since Dalman (1828, p. 117, and pi. II, fig. 6) erected the species Orthis elegantula, many dissimilar shells have been assigned IPalaeonfology, Vol. 14, Part 3, 1971, pp. 487-531, pis. 91-102.] 488 PALAEONTOLOGY VOLUME 14 to it. Investigation of material from Gotland has revealed that at least four species of resserellinid brachiopods occur there (R. elegantula, R. canalis, R. basalis, and V. visby- ensis) and as no types had been selected, some confusion was inevitable. Dahnan’s figures of O. elegantula have proved inadequate and indeed even misleading in that the species is represented as being more elongate than the type material. Material from the Swedish Museum of Natural History, which was collected by Hisinger, is the collection on which Dalman based his species O. elegantula. The lectotype SMNH Br. 1202201 has been selected from this collection and is figured here Plate 91, fig. 5. J. de C. Sowerby {in Murchison 1839, pp. 630 and 640, pi. 13, fig. 12a and pi. 20, fig. 8) erected Orthis canalis now known to be based on two distinct species. One, of Ordovician (Caradocian) age, was renamed Paucicrura sowerbyii by Cave and Dean i (1959, p. 295) and subsequently assigned by Williams (1963, p. 389) to Howellites as H. • antiquior (McCoy). The other of Silurian (Wenlockian to Ludlovian) age, is Resserella eanalis. Bancroft (1928, p. 54) erected the genus Resserella and cited O. eanalis J. de C. Sowerby as the type species but without selecting a lectotype. When Schuchert and Cooper (1932 p. 126) selected a lectotype they cited O. canalis Sowerby 1839, plate 13, fig. 12a, the Wenlock age shell which is congeneric with O. elegantula Dalman. As the latter had been selected by them as the type species of their genus Parmorthis, Parmorthis became a subjective synonym of Resserella. Cooper pointed this out in 1956 (p. 956), and in most subsequent papers, species congeneric with O. elegantula have been assigned to Resserella. However, in pre-1956 literature this name refers to the Ordovician genus and caution is therefore necessary in interpreting the name Resserella in a particular reference. The fourth resserellinid species from the Gotland material has recently been re- described as the type species V. visbyensis (Lindstrom), of a new genus Visbyella (Walms- ley, Boucot, Harper, and Savage 1968, p. 306). In erecting Fascicostella, Schuchert and Cooper (1931, p. 246) were aware of its close | similarity internally to Parmorthis [= Resserella of this paper] and Havhcek (1950, p. 33), recognized the resserellinid affinities of Dedzetina by making it a subgenus of 1 Parmorthis. In regarding the Dalmanellidae as being better placed in the superfamily Dalmanel- lacea Schuchert 1913, than in the Enteletacea Waagen 1884, we subscribe to the view of Johnson and Talent (1967, pp. 142-143) already adopted by Walmsley (in press). SYSTEMATIC PALAEONTOLOGY Suborder dalmanelloidea Superfamily dalmanellacea Schuchert 1913 [= rhipidomellacea Alichova 1960] Family dalmanellidae Schuchert 1913 Subfamily resserellinae nov. Diagnosis. Aseptate or septate plano-convex, shghtly concavo-convex or ventribiconvex Dalmanellidae, commonly shield-shaped, with a prominent beak on the deep pedicle valve. Commonly having a distinctive asymmetrical pattern of bifurcating costellae in the medial region of the brachial valve. Costellae may be evenly or unevenly ramicos- tellate or fascicostellate. Brachiophores widely divergent, and cardinal process bilobed WALMSLEY AND BOUCOT: RESSERELLINAE 489 or trilobed. Ventral muscle jfteld small, cordate. Teeth and sockets commonly crenu- lated and associated with accessory articulation. Comparison. The Resserellinae are generally distinguished from other members of the Dalmanellidae (the Dalmanellinae, Isorthinae, and Cortezorthinae — see Walmsley, Boucot, and Harper 1969, text-fig. 4), by the extreme inequality of convexity of the valves, the pedicle valve always being very much more convex than the brachial valve, which may be slightly convex, flat, or even slightly concave. Associated with the deep convexity of the pedicle valve is the commonly prominent, strongly curved beak which in some cases overhangs the hinge line. The shield-shaped outline is fairly characteristic though not confined to this subfamily (see Walmsley 1965, p. 474), and the branching pattern of the medial costellae in the brachial valve is usually distinctive. The brachial valve muscle field and cardinalia, with the commonly crenulated sockets and widely divergent brachiophores are distinct from those of other dalmanellids as are the gener- ally ponderous teeth, often with crural fossettes. The major variations within the sub- family, recognized in its member genera, concern the bundling of the costellae, the nature of the cardinal process, the attitude of the cardinal area in the brachial valve, and presence or absence of an apical plate or pedicle callist. Genera assigned to Resserellinae Resserella Bancroft 1928 [= Parmorthis Schuchert and Cooper 1931] Fascicostella Schuchert and Cooper 1931 Dedzetina Havlicek 1950 Visbyella Walmsley, Boucot, Harper, and Savage 1968 Morphological features of the Resserellinae. Schuchert and Cooper (1932, p. 129) have already commented on the distinctive morphology of Resserella [Parmorthis] and espe- cially the strength of the articulatory parts. Not only are the teeth and sockets usually large and crenulated (see our PI. 95, fig. 8u, and PI. 96, fig. Ad), but the articulation is further secured by the fitting of peg-like brachiophore processes (see PL 93, fig. 5a, and PI. 96, fig. Ad, g) into usually deep crural fossettes, excavated in the antero-medial faces of the teeth (see PI. 95, fig. 8c and PI. 99, fig. 6a). Commonly the dorsal surfaces of the hinge teeth also bear accessory dental sockets (see PI. 95, fig. 8a and PI. 97, fig. 3a) which engage small accessory teeth (outer socket ridges) formed along the posterior edges of the sockets in the brachial valve. Crural fossettes and accessory teeth and sockets are also present in Visbyella and Fascicostella. An apical plate is present in Visbyella (Walmsley, Boucot, Harper, and Savage 1968, p. 307), and a pedicle callist is present in both Dedzetina and Resserella. The vascula media are divergent in Dedzetina but sub-parallel in Resserella, Visbyella, and Fasci- costella. The dorsal interarea is hypercline in Dedzetina and Visbyella but anacline in Resserella and Fascicostella. The relative size, position, and orientation of the articulatory features help to charac- terize resserellid species. However, the major distinguishing characters of resserellid species are the outline, relative convexity of the valves (especially the brachial valve), the development of incurved beaks and especially the ribbing pattern. The distinctive nature of the resserellid pattern of asymmetrical bifurcation of the medial costellae in the brachial valve was pointed out and illustrated by Walmsley (1965, 490 PALAEONTOLOGY VOLUME 14 XI X Ca. < cq 3 s ^ '-o Urn .£l > *o (U c o W) I2 .t: := cd rt 5 a E a id rt . C S •< p c ^■ E o ^ " a rt c S. « 2 ?■£ a .2 t^-2 *^•—03 •s to 5 B o • ^ ? cd -C ft. s o c W) a .5 a: 3 BO o - c TJ to s *5 ^ .«» W ^ S I S5 ^ 5 -H S s g. 2 s 'c .S c t *2 ,2^ <3 '5 5 « i- .2 a “ ;2 t; "o o 5 B -c E o E a > .2 I § “ a <£ O.S ■= “■.! tj •" £ ^ c E G ^ < X o o “O >. id ? o to C X 60 O s ^ s-a ^ § 2 ^ .Si ^ 4 . >• X T ^ - .c > j*“ o S°i6 ill U a « S « c ! C. I 1 o ^ ‘ ^ >A • 60 2 -I « -S O C3 *f 'H BD*0 V 111 w ^ : WALMSLEY AND BOUCOT: RESSERELLINAE 491 pp. 456-457 and text-fig. 1). It may be clearly seen in a number of the shells figured in the present paper (see PL 96, fig. Id) and in the species of Visbyella (see Walmsley, Boucot, Harper, and Savage 1968, pi. 60, fig. 4u). Even when the preservation does not permit the tracing of the primary costellae, the pattern is usually evident as a medial triangular sector often of finer costellae, distinguishable from lateral areas in which the costellae curve. In detail, the pattern developed in these medial sectors and the density, number, and degree of uniformity of the costellae provide useful diagnostic features. In a general way (text-fig. 1) it is possible to pick out four main types of ribbing pattern in ResserelJa. TEXT-FIG. 1. Showing three different styles of ribbing in the medial panel of the brachial valve of Resserella species and the fascicostellate pattern as seen in Fascicostella. Fascicostella is strongly fascicostellate. Dedzetina, the probable ancestral genus has not clearly developed the asymmetrical pattern. In Resserella, what may be referred to as the eanalis type of ribbing, is even, medium coarse, with a median depression on the pedicle valve which is narrow, has finer costellae and is almost parallel-sided (see descriptions of species for details). On the brachial valve, the medial sector of slightly finer costellae is acutely triangular (see R. spring- fieldensis, R. eanalis, R. elegantula, R. logansportensis, and R. brownsportensis). The basalts type of ribbing is also even and medium coarse, but the medial costellae in the brachial valve branch relatively early and result in more nearly parallel medial costellae and a more parallel sided medial sector (see R. basalts, R. waldronensis, R. elegantulina, and possibly R. concavoconvexa). 492 PALAEONTOLOGY VOLUME 14 EIFELIAN O > Ld Q EMSIAN SIEGENIAN GEDINNIAN PRIDOLIAN LUDLOVIAN cr 3 (/) WENLOCKIAN LLANDOVERIAN B < U > O Q CL O ASMGILLIAN FASCICOSTELLA RESSERELLA VISBYELLA \Li; \ a -o ■'i I \ E. — I I \|/ 6 o/ : oi <= / TEXT-FIG. 2. Time-Stratigraphic ranges of resserellinid species and suggested phylogeny of species and genera. WALMSLEY AND BOUCOT: RESSERELLINAE 493 TEXT-FIG. 3. The geographic distribution of resserellinid species and genera. 494 PALAEONTOLOGY VOLUME 14 The elegantuJoides type is uneven, tending to be semifascicostellate and possibly ori- ginated in R. brownsportensis (see R. amsdeni, R. elegantuJoides, and R. triangularis). The pragensis type is a coarser, uneven, and somewhat fascicostellate pattern developed probably from R. crassicostata (see R. impensa and R. pragensis). Distribution of resserellinids in time and space. Until all the resserellinid species have been recognized and recorded both stratigraphically and geographically, any attempt to plot resserellinid distribution may appear to be premature and incomplete. Nevertheless, the material described in this paper was derived from many parts of the world and even tentative suggestions may prove helpful in guiding further searches. The time-stratigraphic ranges of resserellinid species and the suggested phylogeny of species and genera are illustrated in text-fig. 2. This is based on morphological compari- sons and relative age (so far as correlations are reliable). The known geographic distri- bution of resserellinids is summarized in text-fig. 3, the details of localities are given under each species. A study of these two text-figures together suggests possible migratory routes and indicates areas from which further data may be forthcoming. Genus resserella Bancroft 1928 Type species. Orthis canalis J. de C. Sowerby in Murchison 1839, p. 630, plate 13, fig. 12a. Diagnosis. Plano-convex to ventribiconvex shield-shaped resserellinids having a bilobed cardinal process, an anacline dorsal interarea and an even or only semi-fascicostellate ornament commonly with an asymmetrical pattern of bifurcating costellae in the medial region of the brachial valve. Lacking a dorsal median septum and an apical plate. Species assigned to Resserella Orthis elegantula Orthis basalis Orthis canalis Orthis elegantulina Orthis triangularis Dalmanella springfieldensis Dalmanella waldronensis Dalmanella concavoconvexa Dalmanella elegantidoides Parmorthis crassicostata Parmorthis brownsportensis Parmorthis pragensis Resserella impensa Resserella amsdeni Resserella logansportensis Resserella sefinensis Dalman 1828 Dalman 1828 J. de C. Sowerby 1839 Davidson 1881 Maurer 1889 Foerste 1917 Foerste 1917 Twenhofel 1927 Kozlowski 1929 Schuchert and Cooper 1932 Amsden 1949 Havlicek 1956 Philip 1960 sp. nov. sp. nov. sp. nov. Species rejected from Resserella. Orthis {Dalmanella) elegantula Dalman var. nov. drummockensis, Reed 1917, p. 850, pi. IX, figs. 11-15; Dalmanella elegantula mut. primitiva McLearn 1924, p. 53, pi. 3, fig. 7 [this is D. primitiva (Harper in correspondence)]; D. elegantula mut transversaria McLearn 1924, p. 54, pi. 3, figs. 8, 9 [this is Salopina submedia (Harper in correspondence)]; R. canalis (Sowerby) Bancroft 1928, p. 54; R. para- cyclica Bancroft 1928, p. 56, pi. 1, figs. 6-9; R. cyclica Bancroft 1928, p. 56, pi. 1, fig. 10; R. canalis (Sowerby) Bancroft 1929, pp. 36-38;? R. canalis (Sowerby) Schuchert and Cooper 1932, p. 126, pi. 17, figs. 14-16; R^ [Orthis] media (Shaler), Schuchert and Cooper 1932, p. 129; {O. media Shaler was referred with doubt to Par- morthis by Schuchert and Cooper; it is here rejected from Resserella because its circular outline, fairly convex brachial valve, symmetrical costellae bifurcation, and muscle fields are not resserellid in character); R. visbyensis Lindstrbm), Schuchert and Cooper 1932, p. 129; R. canalis (J. de C. Sowerby) Whittington 1938a, p. 51, pi. 6, figs. 8-12; R. cruralis, Whittington 19386, p. 247, pi. X, figs. 8-11, R. canalis (Sowerby), Bancroft (945, pp. 193, 195, and 203. R. (Howellites) intermedia, Bancroft 1945, p. 205, pi. 27, figs, l^t, pi. 28, figs. 5-6; R. {Howellites} WALMSLEY AND BOUCOT: RESSERELLINAE 495 striata Bancroft 1945, p. 204, pi. 26, figs. 2-10, pi. 27, figs. 13 and 15, pi. 28, figs. 1-3; R. (Howellites) ultima Bancroft 1945, p. 209, pi. 27, figs. 10-12, pi. 28, figs. 9-10; Parmorthis vamliemeni Gill 1948, p. 65, pi. 8, fig. 30 (= Dalejina, see Boucot, Johnson, and Walmsley 1965, p. 337); R. immatiira Williams 1949, p. 165, pi. 8, figs. 1^; R. immatura var. plana Williams 1949, p. 167, pi. 8, figs. 5-6; R. corpulenta (Sardeson), Wang 1949, p. 37, pi. 3a, figs. 1-9; R. corpulenta circularis Wang 1949, p. 38, pi. 3b, figs. \-%\ Parmorthis (Dedzetiiia) macro- stomoides Havlicek 1950, p. 34, pi. V, figs. 1-3, 5-7, P. {Dedzetina) honorata (Barrande), Havlicek 1950, p. 34, pi. XI, fig. 9, P. aff. allani (Shirley), Gill 1950, p. 253, pi. 1, fig. 24 (Shirley’s Baton River species allani is an isorthid). R. (Howellites) altera (Barrande), Havlicek 1950, p. 30, pi. VI, figs. 1, 2, 6, 11, 13, pi. XIII, fig. 1; P. visbyensis var. pygmaea Whittard and Barker 1950, pp. 575-576, pi. VIII, figs. 9-15 ; R. llandoveriana Williams 1951, p. 96, pi. IV, figs. 7-10; R. (Howellites) sp. Whittington and Williams 1955, p. 408, pi. 38, figs. 31-35; R. crassa (Lindstrom) Nikiforova 1954, p. 50, pi. II, figs. 7-1 1 (= an isorthid); R. altaica Kul’kov 1963, p. 15, pi. 1, fig. 2; R. visbyensis (Lindstrom) Rubel 1963, p. 137; R. cf. visbyensis (Lindstrom) Boucot, Johnson, Harper, and Walmsley 1966, p. 14, pi. Ill, figs. 18-20. Species requiring further study before assignment. Orthis dorsoplana Freeh 1888, p. 34, pi. Ill, figs. 5a-c\ Orthis lodanensis Freeh 1888, p. 32, pi. Ill, fig. 4; Dalmanella postelegantula Weller 1903, p. 232, pi. XX, figs. 21-24; D. eugeniensis Williams 1919, p. 118, pi. VII, figs. 1-6; D. eugeniensis var. palaeoelegantula Williams, 1919, p. 118, pi. VII, figs. 7-8; D. (Fascicostellal) clarionda Le Maitre 1944, pp. 25-27, pi. VI, figs. 13-17; Parmorthis crassa Borisyak 1955 (non Lindstrom 1861), p. 18, pi. 1, fig. 2; P. alata Borisyak 1955, p. 19, pi. 1, fig. 3; Par- morthis sp. Castellaro 1959, p. 42, pi. 1, figs. 18-21; P. balaensis Kaplun 1961, pp. 66-67, pi. VII, figs. 1-4; Resserella alf. elegantuloides (Kozlowski) Kul’kov 1963, p. 14, pi. 1, fig. 1. Origin of Resserella. The oldest undoubted Resserella species described are R. spring- fieldensis (Kataigorod Formation, Podolia, late Llandoverian, C4-C6) and R. concavo- convexa (Jupiter Formation, Anticosti Is., late Llandoverian Cg-Ce). The oldest Visbyella species is V. pygmaea (late Llandoverian Cg Britain) (Walmsley, Boucot, Harper, and Savage 1968, p. 310). One of the main differences between Resserella and Visbyella is the dorsal interarea which in typical Resserella is anacline and in Visbyella is hyperchne. It seems significant that one of the earliest species of Resserella, R. concavocotivexa, has a dorsal interarea which is intermediate in position, i.e. is catacline (see Twenhofel 1928, p. 179, and pi. XVI, fig. 3). This is also clear from our own Plate 91, fig. 6a. A comparison of R. con- cavoconvexa (PI. 91, fig. 6) and V. pygmaea (Walmsley, Boucot, Harper, and Savage 1968, pi. 61, fig. 1) shows a close general similarity and in view of their similar earUest appearances in middle late Llandoverian it seems likely that they shared a common origin. In beds older than late Llandoverian Cg, no typical Resserella or Visbyella have yet been recognized. However, in a collection made by Boucot from Q beds of the Llan- dovery area, a new species here assigned to Resserella and described as R. sefinensis, has a brachial valve interior which is in general resserellid but the exterior of the brachial valve has ribbing which is coarser and does not develop the resserellid asymmetrical pattern in the medial area. This species could be regarded as a form transitional from Dedzetina to Resserella or as a primitive Resserella. Williams (1951, p. 94) pointed out that no true representatives of Parmorthis [= Res- serella of this paper] had been collected from beds older than Late Llandoverian (Q). ^Parmorthis"-\ike shells collected from the middle, and early late Llandoverian were con- sidered by Williams to have some morphological features not typically ‘parmorthid’. It is possible that the species not described by him but listed (Williams 1951, p. 129) as Parmorthis sp. from Q, is the species here described as R. sefinensis. It seems clear that the transition from some earlier form to Resserella and possibly Visbyella was taking place at about Llandoverian time. Dedzetina was erected by Havlicek (1950 p. 103) as a subgenus of Parmorthis, of Late 496 PALAEONTOLOGY VOLUME 14 TABLE 2. Diagnostic characters of species of Resserella Profile of Beak of Brachial valve pedicle valve Outline Ribbing R. amsdeni Plano-convex Not overhanging Transverse Fine (80 per 9 mm). Raised median area in dorsal hinge line, high semi-fascicostellate sulcus, median depression along interarea ventral arch R. basalis Plano-convex Not overhanging Triangular Medium coarse (45 Teeth elongate, dental lamellae hinge line, widely per 9 mm) median advanced and sub-parallel separated from beak of brachial valve costeUae sub-paraUel R. brownsportensis Plano-convex Overhangs hinge Elongate- Medium coarse (60 Teeth and crural fossettes, small hne sub-circular per 9 mm) but fine in wide medial panel R. canalis Plano-concave Overhanging hinge Elongate Even medium coarse Teeth, wide, project across line (60 per 9 mm) commissural plane. Dental lamellae medial panel recessive narrow R. concavoconvexa Resupinate Overhangs hinge line widely separated from beak of brachial valve Transverse Even, fine Dorsal interarea catacline R. crassicostata Plane Not overhanging Transverse- Even, very coarse Most coarsely ribbed species hinge line sub-circular (35 per 9 mm) with smooth medial areas R. cleganlula Convex Incurved, may Transverse Even, medium Teeth wide, massive, heavily overhang hinge Hne coarse (60 per 9 mm) crenulated R. elegentulina Convex Overhangs hinge Sub-circular Even, coarse Small species, brachiophores, thick. line stubby R. elegantuloides Convex Not overhanging. Sub-circular Uneven, semi- high interarea. fascicostellate, very widely separated fine (100 per 9 mm) from beak of with intercalated brachial valve strong costae R. impensa Plane Overhangs hinge Sub-circular Uneven, semi- Brachiopores, long, very widely Hne fascicostellate. divergent coarse R. logansportensis Plano-convex Overhangs hinge Elongate Even, medium Teeth long and narrow, dental Hne coarse (50 per lamellae erect, sub-paraUel and 9 mm) advanced. Raised median ridge in pedicle valve R. pragensis 7 7 Sub-circular Uneven, semi- Strong dorsal median ridge fused fascicostellate. posteriorly with bases of widely coarse divergent brachiophores R. sefinensis Plane 7 Sub-circular Even, coarse Dorsal muscle field not impressed or elevated R. springfieldensis Plane-faintly 7 Elongate Even, coarse. Dental lamellae, thick and long. convex medial panel wide long dorsal median ridge R. triangularis Plane Not incurved Sub-circular to Even, medium fine Large species with long dorsal shield-shaped (65-70 per 9 nun) median ridge and thick long brachiophores. Teeth large, dental lamellae short R. waldronensis Plano-convex Small, well Sub-circular- Even, medium separated from triangular coarse (60 per 9 mm) beak of brachial medial costellae valve sub-parallel Ordovician (Ashgillian) age. In many characters Dedzetina is clearly related to Res- serella [Parmorthis] but the outline is transversely elliptical, the ribbing is relatively coarse and symmetrical and the dorsal interarea is hypercline. It is most unfortunate that the dorsal interarea of R. sefinensis is not clear in our material (it appears to be hyperchne) but in its ribbing it shows affinity with Dedzetina macrostomoides, the type species of Dedzetina, (see Havlicek’s pi. V, fig. 2, our PI. 91, figs. 3n, b, and compare our PI. 92, fig. lb). R. sefinensis (Q) appears to have given rise to R. springfieldensis (C3-C5) 497 I. Ribbing even II. Ribbing uneven (semi-fascicostellate) WALMSLEY AND BOUCOT: RESSERELLINAE TABLE 3. Key to Resserella species A With triangular panel of divergent fine costellae in median sector of brachial valve B With sub-parallel median costellae in median sector of brachial valve C Lacking fine costellae in median sector D With fine costellae E With coarse costellae c. Outline transverse Outline elongate Dental lamellae short and Median depression of pedicle R. elegantida divergent valve narrow (2 fine costellae) R. canalis Median depression of pedicle valve broad, (4-5 fine costellae) R. brownsportemis Dental lamellae long and Raised median ridge in subparallel pedicle valve No median ridge in pedicle R. iogansportensis valve R. springfieldensis Beak of pedicle valve Dorsal interarea cataline overhangs hinge line Brachial valve resupinate Dorsal interarea anacline R. concavoconvexa Brachial valve convex R. elegantidina Beak of pedicle valve not overhanging Beaks widely separated R. basalis hingeline Beaks narrowly separated R. waldwnensis R. sefinensis With raised median area in brachial valve sulcus and wide median depression along pedicle valve R. amsdeni Without ditto Ventribiconvex R. elegantidoides Large and plano-convex R. triangularis Short brachiophores Very coarse ribbing with narrow medial panels of fine costellae R. crassicostata Medium coarse semi- fascicostellate ribbing R. impensa Long, very divergent brachlophores R. pragensis and possibly to R. concavoconvexa (Cg-Ce) and itself to have been derived from Dedze- tina or some intermediate form of early or middle Llandoverian age. R. concavoconvexa and V. pygmaea may have had their origins in some as yet undescribed form transitional between them and R. sefinensis. Four lineages may be distinguished in Resserella, the coneavoconvexa-basalis, the springfieldensis-canalis, the brownsportensis-triangularis, and the crassicostata-pragensis lineages. The genus Visbyella seems to be most closely related to the first of these and Fascicoslella to the last of them. Resserella canalis (J. de C. Sowerby) Plate 97, figs. 1, 4—7; Plate 98, figs. 1-2; Plate 100, fig. 4 1839 Orthis canalis, J. de C. Sowerby in Murchison, pp. 630 and 640, pi. 13, figs. \la, non pi. 20, fig. 8. 498 PALAEONTOLOGY VOLUME 14 1847 Orthis elegantiila Dalman; Davidson, p. 62, pi. 13, figs. 9-11, non Dalman 1828. 1848 Orthis elegantiila Dalman; Davidson, p. 321, pi. 3, fig. 23, non Dalman 1828. 1869 Orthis elegantiila Dalman; Davidson, p. 2\l, pars. pi. 27, figs. 1-8, ? fig. 9, p. 205, figs. 1-2, non Dalman 1828. non 1928 Resserella canalis (J. de C. Sowerby); Bancroft, p. 54. non 1932 Resserella canalis (J. de C. Sowerby); Schuchert and Cooper, pi. 17, figs. 14-16. Diagnosis. Elongate Resserella with sub-parallel lateral margins and rounded anterior margin. Brachial valve plane to slightly concave. Costellae even, medium coarse, with narrow medial panel of finer costellae along pedicle valve. Teeth, project across the commissural plane and bear crural fossettes and accessory sockets. Comparison. R. canalis may be distinguished from R. elegantula by its more elongate outline (see PI. 97, figs, la and 2a), its much flatter or even concave brachial valve, its smaller teeth and sockets, smaller brachiophores and absence of fulcral plates. From R. springfieldensis it may be distinguished by its much narrower medial panel of finer cos- tellae along the pedicle valve. It differs from R. basalt's in its less triangular outline and more incurved pedicle valve beak, and from R. brownsportensis by its finer costellae and narrower medial panel of finer costellae along the pedicle valve. Description. Exterior. Plano-convex to slightly concavo-convex, convexity decreasing anteriorly. Brachial valve with shallow median sulcus widening anteriorly. OutUne elongately elliptical to shield-shaped. Length slightly greater than width and twice the thickness. Hinge line straight, three-quarters greatest width which is near mid-length. Cardinal angles slightly rounded, anterior and lateral commissures crenulate and shghtly flexed. A faint, narrow, flattened or depressed area marks the mid-line. Pedicle valve projects one-fifth total length posterior to hinge line with strongly incurved beak over- hanging brachial valve. Interarea anacline in brachial valve, twice as long and strongly curved in pedicle valve, lateral margins sharp. Delthyrium open, triangular, enclosing about 65°. Notothyrium open, but filled with protruding myophore which extends into delthyrium. Costellae relatively coarse, 2-5 per mm at 5 mm length, 60 on a 9 mm wide shell. Interior of pedicle valve. A broad, flat, barely elevated median area extending to mid- length, separates very faint muscle impressions hmited to delthyrial cavity, anterior limit of which on larger shells is marked by a faint rim. Thick dental lamellae meet steep walls of delthyrial cavity at about half their height, bound deep lateral cavities and sup- port small triangular teeth. Teeth, wider than long, bear distinct laterally directed acces- sory sockets and deep crural fossettes. In larger shells, teeth project beyond dental lamellae. All these features confined to posterior one-third of shell length. Anterior crenulations of shell margin, strong, rounded, and separated by equally wide interspaces. Interior of brachial valve. Adductor muscle field, confined to median third of posterior half of valve, slightly elevated with raised margins. Margins merge into brachiophore bases and anteriorly curve into midhne. Median ridge low, rounded, one-third width of muscle field, narrows between anterior pair of muscle impressions. Faint, oblique antero- lateral ridges separate smaller anterior impressions from posterior pair. Brachiophores diverging anteriorly about 70°, are curved plates, thickened distally and inchned laterally to merge with socket pads which support wide, triangular, crenulated sockets. Cardinal process with broad, grooved shaft and crenulated myophore. WALMSLEY AND BOUCOT: RESSERELLINAE 499 Type specimen. The original of pi. 13, fig. 12a of J. de C. Sowerby in Murchison’s The Silurian System, is specimen no. GSM 51550, of the Geological Survey Museum (Insti- tute of Geological Sciences), London. It is here designated the lectotype and refigured (PI. 100, figs. Aa-e). Distribution. R. canalis occurs in the Wenlockian and Ludlovian of the Welsh Border- land and Gotland. Dr. M. G. Bassett (in prep.) provides details of the locahties. He reports that R. canalis is fairly common in the Wenlock Shale and Wenlock Lime- stone of the Welsh Borderland, especially common in the upper Wenlock Shale and Wenlock Limestone of Dudley, and Walsall and occurs in the late Wenlockian mud- stones of Pen-y-lan, Cardiff but is unknown from South Central Wales or Pembroke- shire. R. canalis occurs in material collected by Boucot from the following Gotland locali- ties: canal bank on main road 2 km west and slightly to the south of Klinte Church; uppermost Slite Marl (Wenlockian) ; quarry of cement plant in Slite, Slite Marl (Wen- lockian); just north-east of the main road 2 km north of Grotlingbo church. Eke Marl (Ludlovian). Remarks. There are few determinations given as R. canalis in the literature of the Wen- lockian and Ludlovian but many references to Resserella [= Parmorthis] elegantula. It seems most likely that since both species occur in Gotland, although the presence of R. canalis there has not hitherto been recognized, there have been many misidentifications. R. elegantula is at present known only from the Mulde Marl of Gotland. All references to R. elegantula occurring elsewhere should, therefore, be treated with caution. Resserella elegantula (Dalman) Plate 91, figs. 5a-ct, Plate 95, figs. 8o-c, 9 a-e; Plate 96, figs. \a-e, la-e, 3a-e, 4a-h; Plate 97, figs. 2a, b, 3a, b 1828 Orthis elegantula Dalman, p. 117, pi. II, figs. 6a-e. non 1847 Orthis elegantula Dalman; Davidson, p. 62, pi. 13, figs. 9-11. non 1848 Orthis elegantula Dalman; Davidson, p. 321, pi. 3, fig. 23. non 1 869 Orthis elegantula Dalman ; Davidson, p. 21 1 pars pi. 27, figs. 1-8, ? fig. 9, p. 205, figs. 1-2. 1932 Parmorthis elegantula (Dalman); Schuchert and Cooper, p. 128, pi. 21, figs. 2, 3, 9, 10, 13, 14, 16, 29. Diagnosis. A transversely shield-shaped (length only slightly greater than width) Res- serella, having a convex brachial valve (half as deep as the pedicle valve). Beak of pedicle valve strongly incurved, sometimes overhanging the brachial valve. Teeth, massive with curved crenulated faces. Comparison. R. elegantula is distinctive in its combination of transverse outline and convex brachial valve. Description. Exterior. Ventribiconvex with brachial valve half as deep as pedicle valve, convexity decreasing antero-laterally. Brachial valve with narrow posterior median sulcus, widening and fading anteriorly. Outline transversely shield-shaped to sub- circular, slightly rounded cardinal angles, length slightly greater than width, and one and a half times thickness. Hinge line straight, three-quarters of maximum width, which is at mid-length. Pedicle valve outline triangular posterior to hinge line. Beaks close, separated by a distance of one-thirteenth maximum length. Anterior commissure faintly C 8216 L 1 500 PALAEONTOLOGY VOLUME 14 unisulcate and crenulate, lateral commissure crenulate and straight, crenulations may fade laterally. Wide and short interareas, half as long in brachial valve as pedicle valve, growth lines parallel to hinge line; pedicle valve interarea strongly concave dorsally, apsacline, with sharp margins; brachial valve interarea plane, anachne. Open triangular delthyrium, enclosing about 40°, triangular notothyrium, enclosing about 80°, open but filled with protruding cardinal process. Costellae fine, rounded, 4 per mm at 5 mm length, 60 on 9 mm wide brachial valve; median costellae finer and asynunetrically branched (PI. 96, fig. Id). Median costellae of pedicle valve finer in a narrow depression. Interior of pedicle valve. Muscle field faintly impressed, confined to delthyrial cavity, anterior limit of which is marked by a slight step at about one-third length. Thick, short dental lamellae support massive teeth, lateral cavities deep. Teeth triangular in plan and cross-section and bluntly rounded in profile, bear deep crural fossettes at the junction with the dental lamellae, and oblique crenulations which engage those of the sockets. Antero-lateral accessory dental socket on dorsal surface of tooth. Anterior crenulations extend over one-quarter shell length and are single ridges with pointed ends separated by deep interspaces. A median pair of crenulations reflect the distinctive median line in the costellae. Interior of brachial valve. Adductor muscle field occupies median one-third and pos- terior two-thirds of valve, bounded by raised lateral margins which merge with brachio- phore bases and decrease anteriorly as they converge towards end of median ridge. Median ridge low, rounded, one-fifth width of muscle field, separates wide impressions. Braehiophores, diverging anteriorly at about 70°, are thick, straight, erect plates, tri- angular in lateral profile with anterior edge normal to commissural plane, and bearing small peg-like projections which articulate with the crural fossettes (see PI. 96, fig. Ad). Deep, triangular sockets, supported by socket pads or fulcral plates lie partially beneath EXPLANATION OF PLATE 91 Figs. 1-4. Dedzetiua macrostomoides, Ashgillian of Bohemia, Kralodvorske bridlice Karlik. \a, b. Internal mould of pedicle valve USNM 165876, and rubber impression of mould, x 3. la, b, in- ternal mould of brachial valve USNM 165877, and rubber impression of mould, X 3. 3a, b, external mould of brachial valve USNM 165878, and rubber impression of mould, X 3. 4a, b, internal mould of brachial valve USNM 165879, and rubber impression of mould, x 3. Fig. 5. Resserella elegantida (Dalman), Gotland. 5a-e, brachial valve, pedicle valve, anterior, pos- terior and side views x 2, of whole shell, number Br. 1202201 of Swedish Museum of Natural His- tory, Stockholm, here designated the lectotype. Fig. 6. Resserella concavoconvexa (Twenhofel), Jupiter Formation (late Llandoverian) of Est River Anticosti Island, locality F-26-ER, collected by E. Roche 1939. 6a-e, side, posterior, anterior, brachial valve and pedicle valve, of whole shell USNM 165880, x4. Figs. 7-10. Resserella basalts (Dalman), Visby Marl (late Llandoverian Ce to early Wenlockian), Got- land, specimens loaned by Swedish Natural History Museum, Stockholm, la, b, interior and anterior views of pedicle valve, x 2. 8, interior view of brachial valve, X 2. 9, interior view of brachial valve, x 2. lOu, b, interior and posterior views of brachial valve, x 2. Figs. 11, 12. Resserella brownsportensis (Amsden), late Wenlockian mudstone, old quarry about one- third of mile south-east of Golden Grove Park, about 2 miles south-west of Llandeilo, Britain, grid reference SN 601696. Wa-d, internal mould of brachial valve USNM 165881, and rubber impression and external mould of same valve and rubber impression, X 3. \la-d, external mould of brachial valve USNM 165882 and rubber impression and internal mould of same valve and rubber impression, X3. Palaeontology, Vol. 14 PLATE 91 WALMSLEY and BOUCOT, Resserellinae -J'- i V ■m; AJ' WALMSLEY AND BOUCOT: RESSERELLINAE 501 the interarea. Concentric crenulations on the socket floor mark progressive growth stages which in older specimens produce a fulcral plate and slight lateral cavity beneath. Cardinal process with short, thick shaft and bifid myophore, on which a median groove persists on to posterior face which bears chevron-shaped crenulations. In some shells, median half of each myophore lobe projects slightly further posteriorly so that together these projections simulate a middle member of a trifid myophore. Crenulations of the valve margin as in pedicle valve. Type specimens. No holotype was designated for R. elegantula. Dalman used Hisinger’s collections and several specimens from Djupvik (Mulde Marl) with labels written by Hisinger, are regarded at the Swedish Museum of Natural History as syntypes. From these, specimen no. SMNH Br. 1202201, a whole shell, is here selected as lectotype (see PI. 91, figs. 5a-e). Distribution. Known only from the Mulde Marl (late Wenlockian) of Gotland. Locality, Old Brickyard, 3-3 km south-west of Klinte, Gotland. Map reference, CJ 313606. The species appears to be very localized. References to R. elegantula from other areas where checked, are found to be misidentifications. Remarks. Neither Dalman nor Lindstrom recognized the presence in Gotland of the species named in 1839 as O. canalis J. de C. Sowerby, nor have subsequent authors. Consequently resserellinids from Gotland, other than V. visbyensis and R. basalts have been taken to be R. elegantula (Dalman). Unfortunately Dalman’s figures (1828, pi. II, figs. 6a-b), which show only external views, give the impression of a somewhat elongate form and it is understandable that later authors, having only these figures would be likely to apply the name R. elegantula to the elongate Wenlockian-Ludlovian resserellid — ^which in fact is R. canalis. Thus Davidson in 1847 (p. 62), 1848 (p. 321), and 1869 (p. 211) (see synonomy) assigned to ‘O.’ elegantula, British Wenlockian forms which are certainly not conspecific with the true R. elegantula from the Mulde Marl. Without in- terior views we cannot be certain of all, but most of the specimens figured by Davidson are clearly of R. canalis (J. de C. Sowerby) which Davidson put into synonomy with ^Dalmanelld’ elegantula Dalman, no doubt under the impression (from Dalman’s figures) that elegantula was an elongate form. Many subsequent misidentifications stemmed from this decision of Davidson. A measure of the unrehability of Dalman’s figure (1828, pi. II, fig. 6) is the fact that the outline appears to be distinctly triangular — especially the anterior half. In fact, it appears as if more sharply triangular than the outline of R. basalts given in the same plate (pi. II, fig. 5). R. basalts which can be clearly recognized on other criteria (see later) is the resserelhd with the most triangular anterior outline and Dalman’s plate II, fig. 6, is a misleading representation of R. elegantula. Resserella basalts (Dalman) Plate 91, figs, la, b, 8, 9, 10a, b \ Plate 98, figs. la-f\ Plate 99, figs, la-e, 3a-e, 4a-e, 5, 6a, b; Plate 100, figs. 3a, b, 5a~e 1828 Orthis basalts Dalman, p. 116, pi. II, figs. 5a-e. 1837 Orthis basalts Hisinger, pi. 20, figs. 12a, b. non 1838 Orthis basalts von Buch, p. 60, pi. 2, fig. 9. non 1869 Orthis basalts Dalman (?); Davidson, p. 217, pi. XXVII, figs. 10-11. 502 PALAEONTOLOGY VOLUME 14 non 1912 non 1917 1932 non 1934 non 1952 non 1956 Orthis {Dahnanelld) basalis Dalman; Reed, p. 22, pi. 9, figs. 6-9. Orthis (Dalmanella) basalis, Dalman; Reed, p. 849, pi. IX, fig. 7. Fannorlhis basalis (Dalman); Schuchert and Cooper, p. 129. Pannorthis basalis (Dalman) ; Lamont, fig. 11. Dalmanella basalis (Dalman 1827); sensu Lindstrdm, Dahmer, p. 90, pi. II, figs. 19-21. Pannorthis basalis (Dalman); Curtis, p. 150. Diagnosis. Ventribiconvex Resserella with distinctly triangular anterior outline and beaks widely separated. Bifurcation of the median costellae in the brachial valve is com- pleted relatively early so that the medial costellae in the anterior region are sub-parallel. Comparison. R. basalis is distinguished from R. elegantula by its more triangular outline, flatter braehial valve, longer, less incurved interarea of the pedicle valve, much smaller teeth, thinner dental lamellae, thinner, more curved brachiophores, different cardinal process, and relatively shorter muscle field in the brachial valve. R. basalis is close to R. waldronensis, especially in the arrangement of medial costellae in the brachial valve. However, R. basalis has a more triangular outline and greater separation of the beaks. R. concavoconvexa also has widely separated beaks but the pedicle valve beak is more incurved, crossing the commissural plane and the dorsal interarea is erect (i.e. normal to the commissural plane) whereas in R. basalis it is anacline. Moreover these two species may be distinguished by the relative convexity of the valves, the brachial valve of R. concavoconvexa becoming concave anteriorly. R. eleganluloides also shows a high apsacline ventral interarea and wide separation of the beaks but is easily distinguished by its distinctive ribbing. Description. Exterior. Ventribiconvex, convexity decreasing antero-laterally. Brachial valve only slightly convex posteriorly, with shallow median sulcus widening anteriorly. Outline sub-triangular to shield-shaped. Length equal to width and twice the thickness. Hinge line straight, long, about five-sevenths of greatest width which is at one-third shell length. Cardinal angles slightly rounded. Anterior commissure crenulate and sulcate, lateral commissures gently flexed. Pedicle valve projects one-fifth total length posterior to hinge line with gently curved beak not overhanging hinge line. Brachial valve beak slightly incurved. Beaks widely separated by a distance equal to one-fifth of maximum length. Ventral interarea one-quarter long as wide, gently curved, apsacline, lateral mar- gins sharp. Dorsal interarea half length of ventral interarea, anacline slightly curved apically. Delthyrium open, triangular enclosing about 30°. Notothyrium open, partially occupied by myophore which does not extend into delthyrium. Costellae 3 per mm at 5 mm length, about 45 on 9 mm wide shell. Branching pattern in median area of brachial valve asymmetrical, with few relatively early bifurcations. Interior of pediele valve. Muscle field short and wide, almost confined to delthyrial cavity. Median ridge absent. Laterally inclined dental lamellae support small triangular teeth. Lateral cavities, deep and conical, penetrate beneath interarea. Teeth bear crural fossettes elongate normal to commissural plane. Interior of brachial valve. Adductor muscle field confined to median third of posterior half of valve. In some specimens muscle field slightly elevated with raised margins. Median ridge broad and low, extends to mid-length and separates weak impressions, an- terior pair of which are separated from larger posterior pair by transverse ridges normal to median ridge. Postero-laterally directed ridges connect median ridge to brachiophore WALMSLEY AND BOUCOT: RESSERELLINAE 503 bases (see PL 91, figs. 8-10). Brachiophores, diverging anteriorly about 50°, are tri- angular with anterior edge normal to comiTiissural plane. Small sockets are supported on socket pads or fulcral plates. Cardinal process with triangular bilobed myophore. Crenulations of anterior margin low, rounded and separated by rounded interspaces. Type specimens. No holotype was designated for ‘O.’ basalis. Dalman, however, used Hisinger’s collection from Klinteberg, Gotland, and from this material in the Swedish Museum of Natural History, Stockholm, specimen no. Br. 2298, the original of Dalman’s pi. II, figs. 5a-e, is here selected as lectotype. It is refigured on our Plate 100, figs. 5a-e. Distribution. R. basalis is known from the Visby Marl (late Llandoverian Cg to early Wenlockian) and the Hogklint Group (Wenlockian), of Gotland. Also from the Wen- lock Limestone of the Welsh Borderland. Localities. Wenlock Limestone (middle nodular member), west side of Wren’s Nest Hill, Dudley, Staffordshire, grid reference SO 935921 ; Wenlock Limestone, quarry opening on Ledbury-Malvern road, one-third of a mile east of Ledbury, grid reference SO 378716; Wenlock Limestone, old quarry at Iron Bridge in Benthall Wood, grid reference SJ 665034; Shales immediately above the Wenlock Limestone (?Wenlockian — see Walmsley, 1959, p. 487), old limestone workings south of the road from Monkswood to Glascoed, Usk inlier, Monmouth- shire, grid reference SO 333016. The range is thus Wenlockian — possibly extending into late Llandoverian. Remarks. Of the known Resserella species occurring earlier than R. basalis, R. concavo- convexa seems to be the closest and is possibly ancestral. The specimens figured by Davidson 1869 (pi. XXVII, figs. 10 and 11), from Falfield, near Tortworth and referred by him to "Orthis basalis Dalman (according to Lindstrom)’, is not R. basalis Dalman and appears to be a new species now being described by Dr. M. G. Bassettt. This is the species referred to by Curtis (1956), and may possibly be the species referred to by Reed (1912 and 1917) and Lamont (1934) — see synonymy. Resserella concavoconvexa (Twenhofel) Plate 91, fig. 6a-e, Plate 100, fig. 6a-e 1927 Dalmaneila concavoconvexa Twenhofel, p. 179, pi. XVI, figs. 1-3. 1963 Resserella concavoconvexa (Twenhofel, 1927); Rubel, p. 134, pi. V, figs. 1-7. Diagnosis. (Tentative only, because no internal material available.) Transversely shield- shaped Resserella, ventribiconvex posteriorly, brachial valve becoming concave an- teriorly (i.e. resupinate). Beaks well separated, dorsal interarea normal to commissural plane (catacline), ventral interarea incurved apsacline. Anterior commissure broadly sulcate, lateral commissures curved. Comparison. R. concavoconvexa is similar to R. elegantula in outline but is distinguished by its resupinate brachial valve, catacline dorsal interarea, and greater extension of the pedicle valve beak posterior to the hinge line. From R. canalis it is distinguished by its greater relative width and its resupinate brachial valve. In its well separated beaks it resembles R. basalis but differs in the attitudes of the interareas as well as its rounded anterior margin. Description. As no material showing internal features of this species has been available, no complete description can be given. 504 PALAEONTOLOGY VOLUME 14 Types. The holotype, figured by Twenhofel 1927 (pi. XVI, figs. 1-3), is YPM 10339 of the Peabody Museum of Natural History, Yale University. It is refigured in our Plate 100, figs. 6a-e. Distribution. R. concavoconvexa is recorded by Twenhofel (1927, p. 179) from the Jupiter Formation (late Llandoverian, Cg-Cg) of Anticosti Island. R. cf. concavoconvexa was recorded by Boucot, Johnson, Harper, and Walmsley 1966 (p. 13, pi. Ill, figs. 15- 17, 21-25), from unnamed beds (late Llandoverian C4-C5), Back Bay, New Brunswick, GSC locality 55050. R. cf. concavoconvexa has been noted in a collection made by Ollerenshaw (1961), from the Matepedia Valley, Quebec, (Loc. nos. NCO 571, Jan. 34- 37), of late Llandoverian, C3-C5 age. Rubel 1963 (p. 134, pi. V, figs. 1-7), recorded R. concavoconvexa from Estonia and on p. 137 gave the range of this species in Estonia as Jaani and Jaagarahu stages. According to Rubel (1963, p. 110), these horizons are Wenlockian. On p. 138 Rubel records R. visbyensis ^vom the same horizons. On Gotland Visbyella visbyensis is known from the Visby Marl (late Llandoverian Cg-earliest Wen- lockian). R. cf. concavoconvexa occurs, poorly preserved, in material collected by Walmsley from two localities in South Wales. From mudstone (late Llandoverian, C4) at the south end of the rock face, close to the stream junction at Mandinam, Llandovery district, grid reference SN 742282 and from Canaston Beds (late Llandoverian, ?C5-Cg), south bank of Eastern Cleddau River, about 5 miles east of Haverfordwest, Pembroke- shire, grid reference SN 048139. The range of R. concavoconvexa is thus late Llando- verian (Cg-Cg) to possibly early Wenlockian and from Anticosti, ?New Brunswick, ?Quebec, ?South Wales, and Estonia. Resserella waldronensis (Eoerste) Plate 93, figs. 2a-e, la-d, Aa, b, 5a, b 1879 Orthis elegantida Dalman; Hall, p. 150, pi. 21, figs. 11-17, non Dalman. 1917 Dalmanella waldronensis Foerste, p. 245. Diagnosis. Plano-convex to slightly biconvex Resserella with sub circular outline. Medium coarse costellae, sub-parallel in medial region of brachial valve where bifurca- tion is complete relatively early. Pedicle valve, only slightly (one-tenth) longer than brachial valve, interareas meet at an angle of less than 90°. Margins of delthyrium sub- tend almost 90°. Comparison. R. waldronensis resembles R. basalis in the arrangement of medial costeUae of the brachial valve but is distinguished by its much smaller but more incurved beak of the pedicle valve, which results in closer proximity of the beaks. In R. basalis the pedicle valve also projects considerably further posteriorly beyond the hinge line. In out- line and relative biconvexity, R. waldronensis resembles R. brownsportensis but the latter has a distinctive, finely costellate median depression along the pedicle valve and a medial triangular panel of finer costellae on the brachial valve. Moreover, in R. brownsportensis the dorsal muscle field extends into the anterior half of the valve and is bounded by a strong margin whereas in R. waldronensis it barely extends to mid-length and lacks a well-developed margin. In R. brownsportensis the beak of the pedicle valve projects further posterior to the hinge line and the angle subtended by the margins of the del- thyrium is less than in R. waldronensis. WALMSLEY AND BOUCOT; RESSERELLINAE 505 Description. Exterior. Ventribiconvex, almost plano-convex, brachial valve slightly con- vex posteriorly with shallow median sulcus widening anteriorly so that valve becomes almost plane. Outline sub-circular, commonly somewhat triangular anteriorly. Length equal to width and two and a half times the thickness. Hinge line straight equals two- thirds maximum width which is near mid-length. Cardinal angles slightly rounded, anterior commissure crenulate and broadly sulcate. Lateral commissure slightly flexed. Pedicle valve projects only one-tenth total length posterior to hinge hne, with barely incurved beak. Interareas meet at less than 90°. Ventral interarea curved, apsacline, one- fifth as long as wide, lateral margins sharp. Dorsal interarea two-thirds length of ventral interarea, plane anacline. Delthyrium and notothyrium both open and triangular en- closing about 90°. Notothyrium filled by protruding myophore. Costellae 3 per mm at 5 mm length, about 60 costellae on 9 mm wide shell. Interior of pedicle valve. Floor of delthyrial cavity concave, smooth, lacking median ridge or impressed muscle tracks. Dental lamellae erect, not extended forward of teeth. Teeth small, blunt, triangular, with distinct accessory sockets. Crural fossettes at junc- tion with dental lamellae. Lateral cavities small. Crenulations of anterior margin low, rounded, separated by narrow interspaces. Interior of brachial valve. Adductor muscle field occupies median one-third and pos- terior two-fifths of valve. Median ridge, broad, low, one-quarter width of muscle field, narrows sharply anteriorly. Posterior impressions strong and bounded by raised lateral margins which merge into brachiophore bases. Anterior impressions weak. Brachio- phores diverging anteriorly about 60° are erect, triangular plates, thickened distally but terminating in fine points. Socket pads support wide, triangular crenulated sockets. Cardinal process with broad, commonly carinate shaft and bifid myophore with crenu- lated posterior face protruding beyond notothyrium. Type specimens. The specimens figured by Hall 1879 (p. 150, pi. 21, figs. 11-17), bearing catalogue number 1765/3, in the American Museum of Natural History, New York. Distribution. R. waldronensis occurs in the Waldron Shale (Wenlockian) of Indiana. Localities. Paps Crossing and Vail Quarry, Sandusky, Indiana. Resserella elegantulina (Davidson) Plate 99, figs. la-c\ Plate 100, figs. \a-e, 2a-c 1881 Orthis elegantulina Davidson, p. 152, pi. V, fig. 12. 1883 Orthis elegantulina Davidson; Davidson, p. 219, pi. XIII, fig. 17. Diagnosis. Relatively small, thick-shelled, ventribiconvex, dorsally sulcate Resserella with sub-circular outline and costellae which are moderately coarse and even. Anterior commissure deeply sulcate. Comparison. In its outline and broadly sulcate anterior commissure, R. elegantulina re- sembles R. concavoconvexa, but is distinguished by its small size, relatively coarser rib- bing and its anacline dorsal interarea. From R. canalis, R. elegantulina is distinguished by its more transverse outline, more convex brachial valve and its coarser and more even 506 PALAEONTOLOGY VOLUME 14 costellae in the medial areas of both valves. In R. canalis both the medial depression of the pedicle valve and the narrow triangular medial panel of the brachial valve are occu- pied by finer costellae. R. elegantulim has a relatively shorter hinge line and more rounded cardinal angles than has R. canalis. Description. Small, thick-shelled, ventribiconvex. Sulcus of brachial valve widens an- teriorly. Outline transversely sub-elliptical to sub-circular. Length almost equal to width and one and a half times thickness. Hinge line straight equals two-thirds width. Cardinal angles rounded. Anterior commissure crenulate and strongly unisulcate, lateral com- missures flexed. Pedicle valve projects one-ninth total length beyond hinge line, beak commonly incurved. Ventral interarea curved apsacline with sharp lateral margins, dorsal interarea plane, anachne. Delthyrium triangular enclosing about 65°. Costellae broadly rounded, even, 3 per mm at 5 mm length. Interior of pediele valve. Floor of delthyrial cavity smooth, lacking median ridge. Dental lamellae erect, anterior edges normal to commissural plane. Teeth bluntly tri- angular in plan and profile, bear crural fossettes and accessory sockets. Interior of brachial valve. Adductor muscle field elevated on a platform reaching almost to anterior margin along median one-third of valve. Medial ridge one-third width of muscle field, extends two-thirds length. Brachiophores, stubby distally, fuse with socket pads to form deep triangular sockets. Bifid cardinal process protrudes from notothyrium. Anterior crenulations strong and medially fuse with anterior edge of muscle platform. Type specimens. The type lot ol R. elegantulina consisted of 85 whole shells all registered in the British Museum (Natural History) under one number B.5649. From these a lecto- type has been selected and re-registered as BB. 32232. Its dimensions are length 5-3 mm. EXPLANATION OF PLATE 92 Figs. 1-5. Resserella brownsportensis (Amsden), late Wenlockian mudstone, old quarry about one- third of mile south-east of Golden Grove Park, about 2 miles south-west of Llandeilo, Britain, grid reference SN 601696. In, b, external mould of brachial valve USNM 165883, and rubber im- pression, X 3. 2a-d, internal mould of brachial valve USNM 165884, and rubber impression and external mould of same valve and rubber impression, X 3. 3a-c, lateral and ventral views of internal mould of pedicle valve USNM 165885 and rubber impression, X 3. 4a, b, rubber impression and internal mould of brachial valve USNM 165886, x 3. 5a, b, internal mould of pedicle valve USNM 165887 and rubber impression, X 3. Figs. 6, 7. Resserella sefinensis sp. nov., late Llandoverian (Cj), about 50 ft east of rock step on east side of small stream entering River Sefin (‘Afon Bran’ of Ordnance Map), about 30 ft from the junction of the two streams, about 2 miles east of Llangadock, Britain, grid reference SN 742282. 6fl, b, internal mould of brachial valve USNM 165888, and rubber impression, X 4. la-d, external mould of brachial valve USNM 165889 (here designated holotype), and rubber impression and internal mould of same valve and rubber impression, X 4. Figs. 8-11. Resserella logansportensis sp. nov., Kenneth Limestone (Pridolian), USNM locality 12347, abandoned quarry at Kenneth Station, about 5 miles west of the centre of Logansport, Indiana. Sa-c, interior of pedicle valve USNM 165890, oblique view of interior and lateral view to show teeth, x5. 9a, b, interior and anterior views of pedicle valve USNM 165891, X5. 10, interior view of pedicle valve USNM 165892, X 5. lla-e, brachial valve, pedicle valve, anterior, posterior, and lateral views of USNM 165893, x4, here designated holotype. Palaeontology, Vol. 14 PLATE 92 WALMSLEY and BOUCOT, Resserellinae 'jf’" ;P. I. wy ■Aj ■f ^.r r r WALMSLEY AND BOUCOT: RESSERELLINAE 507 width 5-7 mm, and thickness 2-9 mm. Its locality was Loc. 29 of Davidson and Maw 1881, from the Buildwas Beds, early Wenlockian of Shropshire. Distribution. R. eJegantuIina occurs in the Buildwas Beds (early Wenlockian) of Shrop- shire, Britain. Locality. North bank of River Severn, about 400 yd south-east of Buildwas, Shropshire, grid reference SJ 639046. Remarks. R. elegantulina appears about the same time as R. basalis and it is possible that both were derived from R. concavoconvexa. These three species and R. waldronensis have similar ribbing patterns by which they may be distinguished from other Resserella species. Resserella logansportensis sp. nov. Plate 92, figs. Sa-c, 9a, b, 10, \\a-e\ Plate 93, figs. \a-e Diagnosis. Elongate, plano-convex to slightly biconvex weakly sulcate Resserella with moderately coarse even costellae. Median ridge of pedicle valve wide, flat, and raised. Dental lamellae erect, teeth long and curved. No interiors of brachial valves were available. Comparison. R. logansportensis is closest to R. canalis in its elongate outline and ribbing pattern, but is even more elongate and has a more convex brachial valve. The interiors of the pedicle valves are easily distinguished. In R. canalis the dental lamellae are widely divergent and the median ridge is not raised, whereas in R. logans- portensis the dental lamellae are erect and the median ridge is distinctly raised. Description. Exterior. Small. Ventribiconvex to almost plano-convex with shallow median sulcus widening anteriorly on brachial valve. Outline elongately elliptical. Length one-third greater than width and twice the thickness. Pedicle valve projects one- fifth maximum length posterior to hinge line with incurved beak overhanging hinge hne. Hinge line straight, two-thirds maximum width which is near mid-length. Cardinal angles slightly rounded, anterior margin pointed. Anterior commissure crenulate and gently sulcate, lateral commissures gently flexed. Ventral interarea, curved apsacline, lateral margins rounded. Dorsal interarea one-third length of ventral interarea, plane, anacline. Both notothyrium and delthyrium open and triangular. Latter enclosing about 60°, notothyrium occupied by protruding myophore. Costellae 4 per mm at 5 mm length, about 50 on 9 mm wide shell. Interior of pedicle valve. Narrow, deeply impressed diductor tracks are separated by a broad, low, slightly rounded median ridge which increases in height to mid-length where it ends in broad gentle anterior slope. Vertical walls of delthyrial cavity extend forward as thick, erect dental lamellae supporting long, curved, pointed teeth, tips of which project posterior to hinge line. Crural fossettes mark junction of teeth and lamellae. Lateral cavities small and deep. A pedicle callist fills apex of delthyrium. Type specimens. Specimen USNM 165893, figured Plate 92, fig. lla-c, is designated holotype. Specimens USNM 165890-165892 figured Plate 92, figs. 8-10 are paratypes. Distribution. R. logansportensis occurs in the Kenneth Limestone (Pridolian age). 508 PALAEONTOLOGY VOLUME 14 Logansport, Indiana. USNM locality 12347, abandoned quarry at Kenneth Station about 5 miles west of the centre of Logansport, Indiana. Remarks. The material from the Kenneth Limestone is silicified and two whole shells and about twenty pedicle valves were available. Although no interiors of brachial valves have yet been seen, it is clear that this is a distinct species of Resserella — possibly derived from R. canalis. Resserella amsdeni sp. nov. Plate 93, fig. 6a~e\ Plate 94, figs. \a-d, 2a-e, 3, Aa-e Diagnosis. Transversely shield-shaped, plano-convex to slightly biconvex Resserella with a raised median area in the dorsal sulcus and a deep median depression along the arch of the pedicle valve. The ribbing is finely fascicostellate. Comparison. R. amsdeni is distinguished from all other species of Resserella by its deep median depression of the pedicle valve and opposing raised medial area in the dorsal sulcus. The slightly fascicostellate ornament is seen also in R. elegantuloides, but these species are easily distinguished by the high pedicle valve interarea of R. elegantuloides which also lacks the distinctive medial areas of R. amsdeni R. brownsportensis has a slight median depression in the pedicle valve but does not have a raised medial area in the dorsal sulcus and is readily distinguished from R. amsdeni by its non-fascicostellate ornament. Description. Exterior. Plano-convex to slightly biconvex, convexity decreasing antero- laterally. Median triangular sulcus extending to anterior margin of brachial valve, encloses a triangular median raised area. Pedicle valve has a complementary median depression. Outline transversely shield-shaped, width greater than length, thickness equals two-fifths of length. Cardinal angles gently rounded. Beak of pedicle valve pro- jects one-sixth maximum length posterior to hinge line which it does not overhang. Anterior commissure weakly bisulcate and crenulate, lateral commissures straight. Hinge line straight, equals three-quarters of maximum width, which is near mid-length. Ventral interareas one-tenth as long as wide, concave apsacline, dorsal interarea half as long as ventral interarea, plane anacline. Both notothyrium and delthyrium open, triangular, enclosing 100° and 90° respectively. Notothyrium filled by protruding rnyo- phore. Ornament semi-fascicostellate, fine costellae, 5 per mm at 5 mm length, even finer along median areas of each valve, about 80 on 9 mm wide shell. EXPLANATION OF PLATE 93 Fig. 1. Resserella logansportensis sp. nov., Kenneth Limestone (Pridolian), USNM locality 12347, abandoned quarry at Kenneth Station, about 5 miles west of the centre of Logansport, Indiana. \a-e, brachial valve, pedicle valve, anterior, posterior and lateral views of USNM 165894, x4. Figs. 2-5. Resserella waldronensis (Foerste), Waldron Shale (Wenlockian) of south-east Indiana. 2a-e, lateral, anterior, posterior, brachial valve and pedicle valve of USNM 165895, x3. 2>a-d, posterior, internal, external, and lateral views of brachial valve USNM 165896, x 3. Aa, b, anterior and interior views of specimen USNM 165897, X 3. 5a, b, posterior and oblique interior views of brachial valve USNM 165898, x4. Fig. 6. Resserella amsdeni sp. nov., Henryhouse Formation (Ludlovian) locaUty Amsden 1958, P4, small quarry south-east side of road, SWJ SW|^ sec. 3 T. 2N, R.6E, Pontotoc County, Oklahoma. 6a-e, posterior, anterior, interior, lateral, and external views of brachial valve USNM 165899, x4. Palaeontology, Vol. 14 PLATE 93 WALMSLEY and BOUCOT, Resserellinae ■* WALMSLEY AND BOUCOT: RESSERELLINAE 509 Interior of pedicle valve. Muscle field very faintly impressed, confined to delthyrial cavity in which median area only faintly raised. Dental lamellae very short, attached high on delthyrial walls, support small teeth which project beyond dental lamellae. Lateral cavities small and deep. Deep, rounded crural fossettes immediately below junc- tion of teeth and lamellae. Teeth, triangular in plan and cross section, with bluntly rounded lateral profiles, bear antero-lateral accessory dental sockets. Anterior crenula- tions rounded, with median groove and separated by deep narrow interspaces, extend over one-sixth shell length. Interior of brachial valve. Muscle field occupies median third of posterior half of valve. Median ridge, one-quarter width of muscle field extends to mid-length. Raised lateral margins fuse with brachiophore bases and anteriorly converge on median ridge. Anterior adductor impressions smaller than posterior pair, not separated by distinct ridges. Short, straight, erect brachiophores diverging anteriorly at about 70°, have sub-triangular pro- file with anterior edges inclined slightly posteriorly near distal extremities which bear small peg-like projections. Small sockets on low socket pads extend partially beneath interarea and have faint transverse crenulations. Cardinal process with short, broad shaft and bifid myophore with distinct median groove and crenulated posterior face. Type specimens. Specimen USNM 165903, figured Plate 94, fig. 4a-e is designated holo- type. Specimens USNM 165899 figured Plate 93, fig. 6a-e and USNM 165900-165902, figured Plate 94, figs. 1-3, are paratypes. Distribution. R. amsdeni is known from the Henryhouse Formation (Ludlovian) of Oklahoma. Localities. Amsden 1958, P4, small quarry, south-east side of road SWl SWJ sec. 3 T. 2N, R. 6E, Pontotoc County, Oklahoma ; Amsden 1 958, P6, small glade south of road SEj- SWJ sec. 4, T. 2N, R. 6E, Pontotoc County, Oklahoma: Amsden 1958, P 7, small roadside (west) outcrop of Henryhouse, NE| SEJ sec 32 T. 3N, R. 6E, Pontotoc County Oklahoma. Remarks. This species seems to be the most likely ancestor to R. elegantuloides and was possibly derived from R. brownsportensis, or directly from R. canalis. Resserella brownsportensis (Amsden) Plate 91, figs. Wa-d, \2a-d\ Plate 92, figs, la, b, 2a-d, 3a-c, 4a, b, 5a, b\ Plate 98, figs. 3a-e, Aa-d, 5a-d, 6 1860 Orthis elegantida Roemer {non Dalman), p. 62, pi. 5, fig. 7. 1949 Pannorthis brownsportensis Amsden, p. 42, pi. 1, figs. 1-6. non 1951 Pannorthis brownsportensis Amsden; Amsden, p. 74, pi. 16, figs. 17-23. 1958 Resserella brownsportensis (Amsden); Amsden, p. 148. Diagnosis. Plano-convex Resserella with elongate shield-shaped to sub-circular outline. Costellae broad, rounded, except in median depression of pedicle valve and median triangular panel of brachial valve where costellae are much finer. Comparison. R. brownsportensis is closest to R. canalis, especially in outline, relative convexity of valves, interior of the pedicle valve and presence of a median depression on pedicle valve. However, in R. brownsportensis the lateral margins tend to be sub- parallel, the anterior margin straighter, the beak is commonly less incurved, the smaller 510 PALAEONTOLOGY VOLUME 14 teeth do not project across the commissural plane, the umbo of the pedicle valve is relatively narrower and the costellae are broader and more rounded. In its relatively coarse and broad costellae and relatively narrow umbo it is the closest species to R. crassicostata but the latter is distinctive in its extremely coarse ornament. From R. amsdeni, R. brownsportemis is most easily distinguished by its more elongate outline and its more even and broader costellae as well as by its lack of the distinctive raised median area in the dorsal sulcus and the deep median depression along the pedicle valve. Description. Exterior. Plano-convex, pedicle valve decreasing in convexity antero- laterally and having a wide shallow median depression, brachial valve with median sulcus widening anteriorly, median portion flat. Outline elongate shield-shaped to sub-circular. Length greater than width and twice the thickness. Pedicle valve projects one-fifth total length posterior to hinge line with incurved beak overhanging hinge line. Cardinal angles barely rounded, anterior commissure crenulate and weakly unisulcate, lateral com- missures slightly flexed. Hinge line straight equals two-thirds maximum width which is at mid-length. Ventral interarea curved, apsacline to orthocline, with sharp lateral mar- gins; dorsal interarea shorter, plane, anacline. Both notothyrium and delthyrium open, triangular, latter enclosing about 45°, notothyrium filled with protruding myophore. Costellae broad, rounded, 3 per mm at 5 mm length, about 60 on 9 mm wide shell. Finer costellae in median areas. Interior of pedicle valve. Narrow, elongate, well-impressed diductor tracks flank a broad, flat median area extending two-fifths shell length before ending in a gradual slope to mid-length. A fine median ridge probably separated adductor muscles whose track is presumably represented by the flat median area. Adjustor impressions present along lower part of delthyrial walls. Sub-parallel vascula media extend forward from diductor tracks. Wide pedicle callist present. Laterally divergent dental lamellae support small, triangular teeth with curved anterior edges, small deep crural fossettes and wide acces- sory sockets. Lateral cavities small and deep. Interior of brachial valve. Large, slightly raised adductor, muscle field, half width and two-thirds length of shell. Raised margins fuse with brachiophore bases and anteriorly EXPLANATION OF PLATE 94 Figs. 1-4. Resserella amsdeni sp. nov., Henryhouse Formation (Ludlovian), Pontotoc County, Okla- homa. \a-d, locality Amsden 1958, P4, small quarry south-east side of road, SWf SWf sec. 3T. 2N, R. 6E. Ifl, interior pedicle valve USNM 165900, x 5. \b, c, internal and external views of same valve, x 3. Id, oblique interior view of same valve showing teeth and crural fossette, x 5. 2a-e, locality Amsden 1958, P6, small glade south of road, SEf SWl sec. 4, T.2N, R.6E, lateral, anterior, posterior, brachial valve and pedicle valve of whole specimen USNM 165901, X5. 3. locality Amsden 1958, P7, small roadside (west) outcrop, NEf SEJ sec. 32 T. 3N, R. 6E, interior of brachial valve USNM 165902, x 5. 4a-e, locality Amsden 1958, P6, small glade south of road, SEi SWi sec. 4, T. 2N, R. 6E, brachial valve, pedicle valve, anterior, posterior and lateral views of whole shell USNM 165903, x 3, here designated holotype. Fig. 5. Resserella crassicostata (Schuchert and Cooper), Henryhouse Formation (Ludlovian), locality Amsden 1958, P6, small glade south of road, SEf SWf sec. 4, T.2N, R. 6E. 5a-c, brachial valve, pedicle valve, and anterior views of whole shell USNM 165904, x 5. Palaeontology , Vol. 14 PLATE 94 '/fe'P. 5a 5b WALMSLEY and BOUCOT, Resserellinae WALMSLEY AND BOUCOT; RESSERELLINAE 511 fade as they converge on median ridge which is low, rounded, quarter width of muscle field and narrows anteriorly. No transverse ridges present but posterior impressions deeper. Erect brachiophores diverge anteriorly at 70° and thicken distally becoming triangular in plan. Sockets small, crenulated, and supported on socket pads. Cardinal process has thick broad shaft and distinctly bilobed myophore with crenulated triangular posterior face. Anterior crenulations of shell margin strong, rounded, with pointed tips and separated by deep interspaces. Type specimen. The holotype is specimen no. YPM 17510 of the Peabody Museum, Yale University. It was figured by Amsden 1949 (pi. 1, figs. 2-3). Specimen no. YPM 17512 from the same locality is refigured here, our Plate 98, fig. 5a-d. Distribution. R. brownsportensis is known from the Brownsport Formation (Ludlovian), Tennessee. For details of the many localities see Amsden 1949, p. 42. Amsden 1951 recorded this species from the Henryhouse Formation (Ludlovian) of Oklahoma but the specimens figured by him (pi. 16, figs. 17-23) are more coarsely ribbed and less parallel-sided and are considered to be R. crassicostata (Schuchert and Cooper 1932). R. brownsportensis has been found in material collected by Boucot from late Wen- lockian mud-stones of the Llandeilo district. South Wales. Locality. Old quarry about one-third mile south-east of Golden Grove Park, about 2 miles south-west of Llandeilo, grid reference SN 601696. Remarks. R. brownsportensis is somewhat intermediate in morphology between R. canalis, R. amsdeni, and R. crassicostata. It seems likely that it was derived from R. canalis during late Wenlockian time. R. crassicostata which occurs in both the Brown- sport Formation and the Henryhouse Formation, of Ludlovian age, may have been derived from R. brownsportensis or have had an independent origin from R. canalis during the Wenlockian. Resserella crassicostata (Schuchert and Cooper) Plate 94, figs. 5a-c\ Plate 95, figs. \a-f, 2a-c, 3a, b, Aa, b, 5a-e, 6a, b, 1 1932 Parmorthis crassicostata Schuchert and Cooper, p. 129, pi. 21, figs. 4-5. 1951 Parmorthis brownsportensis Amsden; Amsden, p. 74, pi. 16, figs. 17-23, non Amsden 1949. Diagnosis. Transversely shield-shaped to sub-circular plano-convex Resserella having relatively few and coarse sub-angular costellae, except in the median areas of both valves, where the costellae are fine. Comparison. R. crassicostata is distinct amongst resserelhds because of its very coarse costellae. In general shape and outline it resembles R. amsdeni but has distinct ornament and lacks the characteristic median fold and depression in the dorsal sulcus and pedicle valve. In its pattern of broad, strong costellae enclosing finer costellae in the median areas of both valves, it somewhat resembles R. brownsportensis but is distinguished by the number and coarseness of its costellae. Description. Exterior. Plano-convex, brachial valve with shallow sulcus widening an- teriorly, pedicle valve with finely costellate slightly depressed median area. Outhne 512 PALAEONTOLOGY VOLUME 14 shield-shaped to sub-circular. Length equal to width and two and a half times the thick- ness. Pedicle valve projects one-sixth total length posterior to hinge line, beak gently incurved, not overhanging hinge line. Cardinal angles obtuse. Anterior commissure crenulate and unisulcate, lateral commissures straight. Hinge line straight, almost three- quarters maximum width which is at mid-length. Ventral interarea curved, apsacline with rounded lateral margins. Dorsal interarea shorter, plane, anacline. Both noto- thyrium and delthyrium open, triangular, latter enclosing about 70°. Notothyrium filled with protruding myophore. Costellae sub-angular, coarse, 2-5 per mm at 5 mm length, finer costellae in median areas. Only about 35 costellae on 9 mm wide shell. Interior of pedicle valve. Muscle field faintly impressed, confined to median third of posterior third of valve, barely extending beyond delthyrial cavity. Wide, short diductor tracks separated by faintly raised median area. Short, laterally divergent dental lamellae support blunt, triangular teeth which project anterior of lamellae. Deep semi-cylindrical crural fossettes at junction of teeth and lamellae. Interior of brachial valve. Adductor muscle field extends beyond mid-length in median third of valve, bounded by raised margins which merge with brachiophore bases and anteriorly fade towards median ridge. Median ridge low, rounded, one-quarter width of muscle field, narrows anteriorly. Faint transverse ridges separate smaller anterior im- pressions. Short, thick, straight brachiophores, laterally and anteriorly divergent, have pointed tips, and anterior edges normal to commissural plane. Wide triangular sockets. Cardinal process with short thick shaft and bilobed myophore. Crenulations of anterior shell margin coarse, wide, rounded, with median groove and deep narrow interspaces. Type specimen. The holotype is specimen Cat. No. 913, Schuchert Collection, Yale University, and is recorded by Schuchert and Cooper 1932, p. 129, from Martin Mills, western Tennessee. Distribution. R. crassicostata occurs in the Brownsport Formation (Ludlovian) of Ten- nessee and the Henryhouse Formation (Ludlovian) of Oklahoma. Localities. Amsden 1951, Collection 3, Henryhouse Formation; east side of road in bluff for half a mile, NW]- SWJ, sec. 4, T. 2 N, R. 6E, Pontotoc County, Oklahoma, Amsden 1951, Collection 15, Henryhouse Formation (upper); SWJ NWj, sec. 33, T. 3N, R. 6E, Pontotoc County, Oklahoma; Amsden 1951, Collection 16, Henry- EXPLANATION OF PLATE 95 Figs. 1-7. Resserella crassicostata (Schuchert and Cooper). Figs. 1, 2, and 6, Brownsport Formation (Ludlovian) locality, glade 150 yd south of Mount Lebanon Community Centre, Perry County, Tennessee. Figs. 3, 4, 5, and 7, Henryhouse Formation (Ludlovian), locality Amsden 1958, P6, small glade south of road, SEJ SWJ sec. 4, T. 2N, R. 6E, Pontotoc County, Oklahoma, la-f, external, lateral, internal, lateral, posterior, and anterior views of brachial valve USNM 165905, x4. 2a-c, interior, lateral, and exterior views of pedicle valve USNM 165906, x3. 3a, b, 4a, b, posterior and interior views of brachial valve USNM 165907, x 5. 5a-e, brachial valve, anterior, posterior, pedicle valve, and lateral views of whole shell USNM 165908, x 5. 6a, b, brachial and pedicle valves of whole sheU USNM 165909, x3. 7, interior of brachial valve USNM 165910, x5. Figs. 8, 9. Resserella elegantula (Dalman), Mulde Marl (Wenlockian), locality old brickyard, 3-3 km south-west of Klinte, Gotland, grid reference CJ 313606. 8a, oblique view of interior of pedicle valve USNM 16591 1, X 5, showing large crenulated teeth. 86, c, interior and anterior views of same specimen, X 2. 9a-e, lateral, posterior, anterior, brachial and pedicle valves of whole shell USNM 165912, x4. Palaeontology, Vol. 14 PLATE 95 WALMSLEY and BOUCOT. Resserellinae It T' 1 ' WALMSLEY AND BOUCOT: RESSERELLINAE 513 house Formation, (upper coral beds); NJ SW|-, sec. 4, T. 2N, R. 6E, Pontotoc County, Oklahoma. Amsden 1958, P. 6, Henryhouse Formation, small glade, south of road, SE)^ SW-i sec. 4, T. 2N, R. 6E, Pontotoc County, Oklahoma. Brownsport Formation, glade, 150 yd south of Mount Lebanon Community Centre, Perry County, Tennessee. Remarks. This species was first described, by Schuchert and Cooper (1932, p. 129), from the ‘Niagaran’ of Martin’s Mills, Western Tennessee. It is considered to have diverged from early forms of R. brownsportensis or possibly to have arisen directly from R. canalis. Resserella springfieldensis (Foerste) Plate 101, figs. Aa-e 1917 Dalmanella springfieldensis Foerste, p. 245, pi. XI, figs. 5a-e. 1954 Parmorthis elegantula (Dalman); Nikiforova, p. 46, pi. II, figs. 1-2 {non Dalman). 1960 Parmorthis {Partnortliis) elegantula (Dalman); Sarycheva, pi. XIII, fig. 12 {non Dalman). Diagnosis. Resserella with thick, long, anteriorly divergent dental lamellae extending almost to mid-length. Delthyrial cavity wide and smooth or with weak median ridge. Brachiophores short, thick, widely divergent. Median ridge in brachial valve almost reaches anterior margin. Comparison. R. springfieldensis is distinguished from R. sefinensis by its shorter, thicker brachiophores, more pronounced median ridge in the brachial valve and finer costellae. From R. eanalis, R. springfieldensis is distinguished by relatively longer and less diver- gent dental lamellae and by its wider medial panel of finer costellae in the pedicle valve, but otherwise these species are close and R. springfieldensis may have given rise to R. eanalis. R. eoncavoconvexa differs from R. springfieldensis in its resupinate brachial valve and proportionately greater width. Deseription. Exterior. Plano-convex with non-sulcate brachial valve and strongly arched pedicle valve. Outline elongately elliptical. Length greater than width and twice the thickness. Pedicle valve projects one-fifth total length posterior to hinge line with slightly incurved beak. Cardinal angles obtuse, anterior commissure crenulate and recti- marginate, lateral commissures straight. Hinge line straight, equal to four-fifths maxi- mum width which is slightly posterior to mid-length. Ventral interarea curved, apsacline, two-fifths as long as wide, lateral margins rounded. Delthyrium triangular, open. Cos- tellae rounded, 3 per mm at 5 mm length, finer costellae along median area of pedicle valve. Interior of pediele valve. Faintly impressed muscle field confined to wide, deep, smooth delthyrial cavity with only slightly elevated median area. Strong, thick dental lamellae inelined at 45° to commissural plane, project anterior to smooth triangular teeth, and extend almost to mid-length. Crural fossettes deep and curved. Interior of brachial valve. Faintly impressed muscle field confined to median third of valve, extends into anterior half, with low rounded median ridge one-sixth width of muscle field, extending to three-quarters length. Weak lateral margins merge into brachiophore bases but fade anteriorly. No distinct transverse ridges, but anterior pair of impressions appear to be the larger. Brachiophores, short thick divergent plates. Sockets small, raised on socket pads. Cardinal process with short shaft and small com- pressed myophore. 514 PALAEONTOLOGY VOLUME 14 Type specimens. Eleven specimens numbered 87122 are labelled as syntypes in the Smith- sonian Institution, U.S. National Museum. Distribution. R. springfieldensis was recorded by Foerste (1917 p. 245), from the Cedar- ville Dolomite (Wenlockian) of Eastern Mills Quarry, south-west of Springfield, Ohio. The specimens figured by Nikiforova (1954, pi. 11, figs. 1-2), as Parmorthis elegantula from the KJtaigorod Formation (late Llandoverian C4-C6) of Podolia are considered to be conspecific. Material collected by Dr. A. M. Ziegler from the Woolhope Limestone (early Wenlockian) of May Hill, Welsh Borderland, and housed in the Oxford Univer- sity Museum, includes a resserellid referred to ‘i?. cf. elegantula’ . Examination of this material by Walmsley reveals that this species is R. springfieldensis. Localities. 1. Northwest of Old Oaks Farm, grid reference SO 68692244. 2. Stream section, 180 yd S. 83° W., of Hill Farm Glass house, grid reference SO 70552103. Two poorly preserved specimens of Resserella collected by Dr. M. G. Bassett from the Woolhope Limestone are also considered to be conspecific with R. springfieldensis. Localities. 1. Main quarry at Scutterdine, Woolhope, Herefordshire, grid reference, SO 577372. 2. Old quarry, north side of road between Alfrick and Crews Hill (Malverns area), grid reference SO 73995291. The known distribution of the species is thus late Llandoverian (C4-C6) of Podolia, early Wenlockian of the Welsh Borderland and Wenlockian of Ohio. Remarks. R. springfieldensis is considered to have given rise to R. canalis in the Wen- lockian. Resserella elegantuloides (Kozlowski) Plate 98, figs. 8u, b; Plate 99, figs, la-e 1929 Dalmanella elegantuloides Kozlowski, pp. 63-67, text-figs. 9a, 10, 11; pi. II, figs. 1-16. 1932 Dalmanella ? crassifonnis Kozlowski; Paeckelmann and Sieverts, p. 27, pi. 1, figs. lOa-d. 1954 Parmorthis elegantuloides (Kozlowski); Nikiforova, pp. 48-50, pi. II, figs. 3-6. non 1963 Resserella aff. elegantuloides (Kozlowski); Kul’kov, p. 14, pi. 1, fig. 1. Diagnosis. Ventribiconvex, dorsally sulcate Resserella with relatively high ventral inter- area and distinctive ornament in which 12-14 more prominent costellae are developed amongst numerous fine costellae. Comparison. In its semi-fascicostellate pattern of ornament R. elegantuloides is closest to R. amsdeni from which it may be distinguished, however, by the absence of the latter’s characteristic fold in the dorsal sulcus, and sulcus in the ventral fold. Nevertheless the slightly depressed median area of the pedicle valve of R. elegantuloides supports the view EXPLANATION OF PLATE 96 Figs. 1-4. Resserella elegantula (Dalnian) Mulde Marl (Wenlockian), locality, old brickyard, 3-3 km, south-west of Klinte, Gotland, grid reference CJ 313606. \a-e, lateral, anterior, posterior, brachial and pedicle valves of whole shell USNM 165913, x 2. la-e, lateral, anterior, posterior, brachial and pedicle valves of whole shell USNM 165914, x3. la-e, anterior, posterior, lateral, brachial, and pedicle valves of whole shell USNM 165915, x 4. 4a, b, anterior and posterior views of brachial valve USNM 166535, x3. 4c, d, anterior and posterior views of same specimen X 5, showing brachiophores and crenulated sockets. 4e-h, lateral, posterior, internal, and external views of same specimen, x 3. Palaeontology, Vol. 14 PLATE 96 WALMSLEY and BOUCOT, Resserellinae 4 ■■■ < ')! i I I ^5t :“l: , \flK-; ■ !„;;J hi&'. f i 1 r> y '"f ’" ■' ■' s~- \ , WALMSLEY AND BOUCOT: RESSERELLIN AE 515 suggested by the similar style of ornament and similar high ventral interareas that these species may be fairly closely related. Description. Exterior. Ventribiconvex with shallow median sulcus widening anteriorly on gently convex brachial valve. Outline sub-circular. Length equal to width and twice the thickness. Pedicle valve projects one-fifth of total length posterior to hinge line, beak erect, barely incurved. Beaks separated by distance equal to one-quarter length of hinge line. Cardinal angle slightly rounded. Anterior commissure crenulate and broadly uni- sulcate, lateral commissures straight. Hinge line straight, equals three-quarters greatest width which is near mid-length. Ventral interarea, apsacline, almost orthocline, slightly incurved at apex, three-tenths as long as wide, lateral margins sharp. Dorsal interarea half as long, plane, anacline. Notothyrium and delthyrium both open and triangular, enclosing 60° and 30° respectively. Myophore does not project from notothyrium. Costellae rounded, 3 per mm at 5 mm length, about 100 on 9 mm wide shell. About 12 costellae on pedicle valve stronger than average, producing an uneven semi-fasci- costellate aspect. Median sectors of both valves have finer sub-parallel costellae. Interior. No material showing the interior of either valve was available. However, Kozlowski (1929, fig. 10) diagrammatically illustrated the internal features, which are clearly resserellid. Type specimens. Kozlowski (1929, pp. 63-67) did not designate a holotype but the speci- men figured by him in his pi. 2, fig. 1, which he described as ‘Echantillon typique’ is here selected, with his agreement, as lectotype. Distribution. R. elegantuloides was stated by Kozlowski (1929, p. 67) to be charac- teristic of the Borszczow stage (Gedinnian), especially the middle part. It also occurs in the Gedinnian of Nevada (see Johnson, Boucot, and Murphy 1967, p. 684). Remarks. R. elegantuloides may have been derived from R. amsdeni. No descendant species is known in the Siegenian, but R. triangularis (Emsian) has similar fine ribbing and may be derived from R. elegantuloides. Resserella sefinensis sp. nov. Plate 92, figs. 6a, b, la-d Diagnosis. Small Resserella with plane to slightly sulcate brachial valve and relatively few broad costellae, lacking finer costellae in median sector of brachial valve. Pedicle valve not known. Comparison. In the coarseness of its ornament R. sefinensis somewhat resembles R. crassicostata but the latter bears very fine costellae along the median areas of both valves. The absence of a distinct median panel of finer costellae also distinguishes R. sefinensis from R. brownsportensis which also has relatively thicker brachiophores. Description. Exterior of brachial valve. Small, plane, sulcate with shield-shaped outline. Costellae relatively few and uniformly coarse. A median triangular panel may be dis- tinguished from the lateral areas but the distinctive asymmetrical branching pattern of later resserellids is not seen. Hinge line straight, three-quarters greatest width which is posterior to mid-length. Cardinal angles obtuse. Interarea anacline. C 8216 M m 516 PALAEONTOLOGY VOLUME 14 Interior of brachial valve. Muscle field faintly impressed, not clearly delimited by raised margin or elevated on a platform. Median area slightly raised as a broad, low median ridge. Brachiophores thin, erect to postero-laterally inclined plates diverging anteriorly at 90°. Sockets, deep, conical, and crenulated. The small cardinal process is a lobed myophore arising directly from the posterior end of the median ridge without a shaft. Crenulations of the anterior shell margin broad, flat, and confined to the periphery. Type specimens. Specimen USNM 165889, figured Plate 92, figs, la-d, is designated holotype. Specimen USNM 165888, figured Plate 92, fig. 6a, is a paratype. Distribution. R. sefinensis is known from beds of late Llandoverian (Ci) age from the Llandovery area Wales. Locality. About 50 ft east of rock step, on east side of small stream entering River Sefin (‘Afon Bran’ of Ord- nance Map), about 30 ft from the junction of the two streams, about 2 miles east of Llangadock. Grid reference SN 742282, Camarthenshire. Remarks. This species is assigned to Resserella despite lack of pedicle valve material be- cause the interior of the brachial valve is resserellid in most characters. However, the poorly impressed, undefined, and unraised muscle field is not typical of the genus. Externally, the ribbing is only resserellid in its pattern of a median triangular panel flanked by lateral areas with slightly curved costellae. It is considered that R. sefinensis is a primitive Resserella and it is certainly the oldest species yet recognized. It seems that it was during late Llandoverian (Q) time that the genus Resserella made its first appearance. R. sefinensis is thus understandably not completely typical. Resserella triangularis (Maurer) Plate 100, figs. 7, 8a, Z>; Plate 101, figs. 1-3 non 1857 Orthis triangularis ZQiler, p. 49, pi. 4, figs. 14-16 [= Platyorthis, see Harper, Boucof, and Walmsley 1969, p. 86], 1889 Orthis triangularis Maurer, p. 160, pi. Ill, figs. 8-12. 1961 Parmorthis triangularis (Zeller); Kaplun, p. 67, pi. 7, figs. 8-15. EXPLANATION OF PLATE 97 Fig. 1. Resserella canalis (J. de C. Sowerby), Slite Marl (Wenlockian), Gotland. \a-e, brachial and pedicle valves, anterior, posterior and lateral views of whole shell USNM 166536, X 3. Figs. 2, 3. Resserella elegantula (Dalman), Mulde Marl (Wenlockian), locality, old brickyard, 3-3 km south-west of Klinte, Gotland, grid reference CJ 313606. 2a, exterior of brachial valve USNM 166537, X 5. 2b, interior of same specimen, X 3. 3a, interior of pedicle valve USNM 166538, X 3. 2b, exterior of same valve, X 5. Figs. 4-7. Resserella canalis (J. de C. Sowerby), figs. 4a-rf, Slite Marl (Wenlockian), locality, canal on main road, 2 km west and slightly to south of Klinte Church, Gotland. 4a, b, anterior and posterior views of brachial valve USNM 166539, X 3. 4c, exterior of same specimen, X 5. Ad, lateral view of same specimen, x 3. 5a, b, Slite Marl (Wenlockian), locality, quarry of cement plant in Slite, Gotland, interior and anterior views of pedicle valve USNM 166540, X3. 6a-c, la-e. Eke Marl (Ludlovian), locality, just north-east of main road 2 km north of Grotlingbo church, Gotland. 6a-c, lateral, posterior, interior, anterior, and exterior views of brachial valve USNM 166541. X3. la-e, anterior, posterior, lateral, brachial, and pedicle valves of whole shell USNM 166542, x 3. Palaeontology , Vol. 14 PLATE 97 WALMSLEY and BOUCOT, Resserellinae \ • V. y ■■ ■■■ I ' °S ... 'jm WALMSLEY AND BOUCOT: RESSERELLINAE 517 Diagnosis. A large Resserella with fine, even ribbing. Brachiophores, long, widely diver- gent anteriorly and posteriorly fused to prominent median ridge which extends beyond mid-length. Very narrow median ridge in pedicle valve extends from delthyrial cavity to mid-length. Comparison. In the cardinalia and especially the long, strong brachiophores and median ridge, R. triangularis resembles R. pragensis (see Havlicek 1956, pi. X, fig. 3). However, the ribbing is quite different. R. triangularis has numerous fine, even ribs whereas R. pragensis has fewer and stronger ribs some of which are more prominent than others. In the fineness of ornament R. triangularis is closest to R. elegantuloides (see Kozlowski 1929, pi. 2, fig. 2) but the latter has semi-fascicostellate uneven ribbing, and less promi- nent brachiophores. R. elegantuloides is ventribiconvex whereas R. triangularis is plano- convex and much larger. Description. Exterior. Large, plano-convex, sulcus on brachial valve widening and deepening anteriorly. Outline subcircular to shield-shaped. Length equal to width and twice the thickness. Pedicle valve projects one-fifth total length posterior to hinge line, beak barely incurved. Cardinal angles slightly rounded. Anterior commissure unisulcate and crenulate, lateral commissures straight. Hinge line straight, three-quarters of maxi- mum width which is near mid-length. Ventral interarea apsacline, dorsal interarea anacline. Both notothyrium and delthyrium open and triangular, latter enclosing about 100°. Notothyrium filled with protruding myophore. Costellae fine, even, 3 per mm at 5 mm length, about 70 on 9 mm wide shell. Interior of pedicle valve. Strongly impressed muscle field barely extends beyond del- thyrial cavity. Slightly raised median area separates elongate adductor scars and con- tinues to mid-length as a fine median ridge. Elongate diductor impressions extend to beginning of median ridge beyond which point parallel vascular trunks continue to mid- length. Adjustor scars on delthyrial walls extend almost to teeth. Teeth large, triangular, with large, antero-lateral crural fossettes and accessory sockets, are supported on short dental lamellae. Pedicle callist present. Interior of brachial valve. Slightly elevated triangular muscle field with weakly defined margins extends almost two-thirds valve length and is almost half as wide as greatest width. Median ridge narrow, prominent, extends length of muscle field and posteriorly merges with notothyrial platform and shaft of cardinal process which has median groove and expanded bifid myophore. Brachiophores thick, diverging anteriorly at about 90° and inclined posterolaterally. Sockets, strongly crenulated, antero-laterally aligned, lack fulcral plates or socket pads and bear small denticles along posterior edges. Crenula- tions of anterior shell margin narrow, rounded, separated by slightly wider interspaces are confined to periphery. Type and figured specimens. The material figured by Maurer 1889 (pi. Ill) is housed in the Hessisches Landesmuseum, Darmstadt. Specimen no. Mu. 5409 is the original of his fig. 8 and is labelled from Lahneck. Specimen no. Mu. 4568 is the original of his fig. 9 and specimen no. Mu.4566 is the original of his figs. 10 and 10a. Specimen no. Mu.5409 is here selected as lectotype. C 8216 Mm2 518 PALAEONTOLOGY VOLUME 14 Distribution. Specimens studied in the Senckenburg Museum, Frankfurt, bear the follow- ing locality labels : Specimen no. XVII 542a and b, Unt. Kobl. Sch. — Oppershafen » 542c ,, ,, ,, — Str. Usberg „ 542d Ob. Kobl. Sch. — Dreslendorf „ 542e — Strasserbersb „ 542f >» — Brandobernd „ 542g Unt. Devon. — Haiger „ 563c Ob. Kobl. Sch. — Haiger „ 563e J — Karstel b. Ob< SMF 19983 ,, — Helmesthal/K Unnumbered specimens „ — Karstel „ „ Ob. Ems. Helmroth, Western The species appears therefore to occur in the Emsian of the Rhineland. The material figured by Kaplun (1961, pi. 7, figs. 8-15) was from the Lower Devonian and lower Middle Devonian of Kazakhstan (Sardzhal’sk and Kazakhok beds). Resserella impensa Philip 1962 Resserella impensa Philip, p. 200, pi. 30, figs. 20-24, and fig. 11 (p. 199). Diagnosis. Resserella species with coarse and uneven, semi-fascicostellae ribbing and very widely divergent dental lamellae. Comparison. The pattern of ribbing relates this species to R. pragensis and possibly R. EXPLANATION OF PLATE 98 Fig. 1. Resserella canalis (J. de C. Sowerby), Slite Marl (Wenlockian), locality, canal on main road, 2 km west and slightly to south of Klinte Church, Gotland, la, b, interior and anterior views of pedicle valve USNM 166543, x3. Fig. 2. Resserella canalis (J. de C. Sowerby), Eke Marl (Ludlovian), locality, just north-east of main road 2 km north of Grotlingbo Church, Gotland. 2a, b, interior and anterior views of pedicle valve USNM 166544, x3. Figs. 3-5. Resserella brownsportensis (Amsden), figs. 2>a-e, Brownsport Formation (Ludlovian), glade 150 yd south of Mount Lebanon Community Centre, Perry County, Tennessee, anterior, posterior, brachial valve, pedicle valve and lateral views of whole shell USNM 166545, x 4. Aa-d, Brownsport Formation (Ludlovian), locality Amsden 1949, 18-(2), glade 30-45 ft above the Dixon-Brownsport contact. Blue Mound Glade, 0-2 miles north of road leading from Cedar Grove Church to Mt. Carmel Church, 0-8 mile north-east of Cedar Grove Church, Perryville quadrangle, Tennessee, ex- terior, posterior, interior and lateral views of brachial valve YPM 17511, x3. 5a-d, Brownsport Formation (Ludlovian), locality Amsden 1949, 9-(3) approximately 50-60 ft above Dixon-Browns- port contact, north side of State Highway 114, 2f miles south-east of Bath Springs Church, Bath Springs Quadrangle, Tennessee, lateral, interior, posterior and exterior views of pedicle valve YPM 17512, x3. Fig. 6. Resserella brownsportensis (Amsden), Brownsport Formation (Ludlovian), glade 150 yd south of Mount Lebanon Community Centre, Perry County, Tennessee, exterior of pedicle valve USNM 166546, X3. Fig. 7. Resserella basalis (Dalman), Hogklint (Wenlockian), Visby, where Kopparsvikgaten and Cutevagen cross, Gotland, la, brachial valve, X 5. Ib-f, anterior, lateral, posterior, pedicle and brachial valves of same specimen, USNM 166547, x2. Fig. 8. Resserella elegantidoides (Kozlowski), Borszczow Formation, Lanowce, Podolia. 8a, pedicle valve USNM 166588, x 3. Sb, oblique view of same specimen, x 5, showing uneven ribbing. Palaeontology, Vol. 14 PLATE 98 WALMSLEY and BOUCOT, Resserellinae WALMSLEY AND BOUCOT: RESSERELLINAE 519 crassicostata. It is less coarse than in R. crassicosta and slightly fascicostellate. In this it is close to R. pragensis but the cardinalia are clearly different. Description. This species was described in 1962, by Philip. As no material has been available to us, no further details can be added. Type specimens. The holotype is specimen number M 3430 of Melbourne University Geology Department. Figured by Philip 1962 (pi. XXXI, figs. 20-22). Distribution. R. inipensa is recorded Philip (1962, p. 201) from the Boola Beds and Con- glomerate Phase of the Cooper’s Creek Formation, Tyers, Victoria, Australia. In his addendum (pp. 244-246), Philip regarded the upper part of the Boola Beds as Early Gedinnian. As a result of subsequent conodont studies, Philip (1965, pp. 97-98) recon- sidered the horizon of the Tyers fauna and suggested a late Gedinnian or early Siegenian age. We regard these units as of Siegenian age. Remarks. R. inipensa may be a derivative of R. crassicostata and may have led to R. pragensis of Upper Emsian age. Resserella pragensis (Havlicek) 1956 Parmorthis pragensis HavUcek, p. 539, pi. 10 (47), figs. 3-6. Diagnosis. Resserella species with coarse, uneven, semi-fascicostellate ribbing and long, very divergent brachiophores. Comparison. In the cardinalia, especially the long and very divergent brachiophores and the prominent median ridge in the brachial valve, R. pragensis is similar to R. triangu- laris. However, the ribbing style is quite different. R. pragensis has fairly coarse uneven ribbing whereas R. triangularis has numerous fine, even, costellae. From R. impensa, R. pragensis may be distinguished by its less fascicostellate ribbing as well as by its more prominent brachiophores and median ridge in the brachial valve. Description. No material has been available to augment the description of this species given by Havlicek 1956 (p. 539). Type specimens. The holotype is the pedicle valve figured by Havlicek 1956 (pi. 10, fig. 5). Distribution. R. pragensis was described by Havlicek from the Zlickov Limestones, (upper Emsian) of Bohemia. Remarks. The similar style of ribbing and relative ages suggest that R. pragensis may have been derived from R. impensa. R. pragensis is the youngest Resserella species yet recorded. Genus dedzetina Havlicek, 1950 Type species. Dedzetina macrostomoides Havlicek 1950. Diagnosis. Multicostellate, ventribiconvex, transversely elliptical resserellinid with hypercline dorsal interarea, very widely divergent brachiophores and pedicle callist present. 520 PALAEONTOLOGY VOLUME 14 Comparison. HavHcek (1950, pp. 33-34) erected Dedzetina as a subgenus of Parmorthis Schuchert and Cooper 1931 [= Resserella of this paper]. His description and figures of D. macrostomoides the type species, shown on his pi. V, figs. 1-3 and 5-7, indicate clearly the affinity between Dedzetina and Resserella. Figures of D. maerostomoides shown here (our PI. 91, figs. 1-4) illustrate the typically resserellid pattern of muscle impressions in the convex pedicle valve (fig. In) and the much flatter brachial valve (figs, la, b, and 4a). The widely divergent brachiophores, the bilobed cardinal process and the presence of a pedicle callist all indicate a close relationship with Resserella. However, there are differences which we consider warrant generic distinction. The trans- verse form and somewhat convex brachial valve are not typical of Resserella although both are shown by R. elegantida. The interarea of the brachial valve of Dedzetina is hypercline, whereas in Resserella and Faseicostella it is anacline. It is interesting to note that Visbyella, which appears at about the same time as Resserella, both possibly sharing a common ancestry from Dedzetina, also has a hypercline interarea in the brachial valve, although its cardinal process is quite different. The brachiophores in Dedzetina are very widely divergent, more so than is typical of Resserella and the anterior pair of adductor scars are larger than the posterior pair which is the reverse of the situation in both Resserella and Faseicostella. Havlicek (1950, p. 33) drew attention to the widely divergent vascula media in the pedicle valve of Dedzetina. In Resserella (see PI. 92, fig. 5a; PI. 98, fig. 5b) the vascula media are sub-parallel anterior to the muscle field of the pedicle valve. The characteristic medial triangular panel of finer costellae seen in the typical Resserella brachial valve is absent from Dedzetina which is easily distinguished from Faseicostella by its non-fascicostellate ribbing. Discussion. The main changes required for Resserella to have evolved from Dedzetina would therefore be : a more elongate form with a flatter brachial valve, modification of the angle of the dorsal interarea, modification of the ribbing to produce the medial panel and convergence of the vascula media in the pedicle valve. R. sefinensis, the oldest Resserella known, although distinctly resserellid in its flat brachial valve and cardinalia (see PI. 92, figs. 6a, b, Ic, d) nevertheless shows a ribbing pattern (PI. 92, figs, la, b) more EXPLANATION OF PLATE 99 Fig. 1. Resserella elegantiiloides (Kozlowski), Borszczow Formation, Lanowce, Podolia. \a-e, brachial valve, posterior, anterior, pedicle valve and lateral view of whole shell USNM 166549, x3. Figs. 2-6. Resserella basalts (Dalman), Wenlock Limestone, figs. 2, 3, and 5, locahty, middle nodular member, west side of Wren’s Nest Hill, Dudley, Staffordshire, grid reference SO 935921. 4, locality, old quarry at Iron Bridge in Benthall Wood, Shropshire, grid reference SJ 665034. 6, locality, quarry opening on Ledbury-Malvern road, one-third of a mile east of Ledbury, grid reference SO 378716. 2a-e, brachial and pedicle valves, anterior, posterior and lateral views of whole shell USNM 166550, x3. 3a-e, lateral, brachial valve, posterior, anterior, and pedicle valve of whole shell USNM 166551, x 5. 4a-e, lateral, anterior, exterior, posterior, and interior views of brachial valve USNM 166552, x4. 5, interior of brachial valve USNM 166553, x3. 6a, b, two interior views of pedicle valve USNM 166554, x 3. Fig. 7. Resserella elegantidina (Davidson), Buildwas Beds (Wenlockian), north bank of River Severn, about 400 yd south-east of Buildwas, Shropshire, grid reference SJ 639046. la-c, posterior, interior, and lateral views of brachial valve USNM 166555, x 5. Palaeontology, Vol. 14 PLATE 99 WALMSLEY and BOUCOT, Resserellinae WALMSLEY AND BOUCOT: RESSERELLINAE 521 like that of Dedzetina — especially as figured by Havlicek (1956, pi. V, fig. 2). Its appear- ance in late Llandoverian (Q) is consistent with the view that it may be regarded as an intermediate between Dedzetina (Ashgillian) and the more typical Resserella spp. of later Llandoverian. Species assigned to Dedzetina. In addition to the type species D. macrostomoides, Havlicek 1950, p. 34, assigned to Dedzetina, with some doubt, the species Orthis honorata Barrande, 1879, as redefined by him. Stratigraphic range of Dedzetina. Dedzetina is known only from the Ashgillian (Krai v Dvur Shales) of Bohemia. Genus visbyella Walmsley, Boucot, Harper, and Savage 1968 Type species. Orthis visbyensis Lindstrom 1861, p. 366, pi. XII, fig. 8. Diagnosis. Plano-convex to slightly concavo-convex or ventribiconvex resserellinids having a trilobed and extroverted (dorsally facing) cardinal process and hypercline dorsal interarea. Apical plate present, dorsal median septum developed in some species. Comparison. Visbyella is distinguished from Resserella by its hypercline dorsal interarea, its trilobed extroverted cardinal process and its tendency for the dorsal median ridge to develop anteriorly into a median septum. From Fascicostella, it is distinguished by its non-fascicostellate ornament and from Dedzetina by its ‘resserellid’ type asymmetrical ornament, though both Visbyella and Dedzetina have hypercline dorsal interareas. This genus and its four species: V. visbyensis (Lindstrom), V. nana (McLearn), V. pyginaea (Whittard and Barker), and V. ciinmockensis Walmsley, Boucot, Harper, and Savage, were described by Walmsley, Boucot, Harper, and Savage (1968, pp. 306-316, pis. 60-62). The genus is known to range from late Llandoverian (C3, possibly Cx) to late Wenlockian (M. testis zone) (see text-fig. 2, and p. 495 for a discussion of its possible origin). Genus fascicostella Schuchert and Cooper 1931 Type species. Strophomenes gervilii Defrance 1827, p. 152. Diagnosis. Coarsely fascicostellate resserellinid with bilobed cardinal process and ana- cline dorsal interarea. Comparison. In its internal features, Fascicostella closely resembles Resserella except that in both valves the muscle fields tend to be relatively wider and shorter and in the pedicle valve a very thin median ridge may be developed. Externally, the ornament is distinctive in its excessive fascicostellation with 9-16 sectors (bundles of ribs) developed, sometimes so strongly as to give a semi-plicate appearance. From both Visbyella and Dedzetina, Fascicostella is distinguished by its anacline dorsal interarea as well as by its ornamentation, and from Visbyella also by its bilobed cardinal process. Discussion. Fascicostella was erected by Schuchert and Cooper (1931, p. 246) with Strophomenes gervilii Defrance (1827, p. 152) as type species. 522 PALAEONTOLOGY VOLUME 14 Schuchert and Cooper (1932, p. 130) also assigned to Fascicostella, Or this dorso- plicata Beclard 1891 and O. sedgwicki D’Archiac and Verneuil 1842. Of these, O. dorso- plicata is here assigned with doubt to Fascicostella. Beclard (1891, pp. 99-100) dis- tinguished O. dorsoplicata from O. lodanensis and O. dorsoplana of Freeh 1888 and also from O. triangularis Maurer 1889, mainly on the presence in dorsoplicata of a narrow sinus on the pedicle valve and a narrow ridge on the brachial valve. We agree that this species is different, but of Bedard’s figures (pi. 3, figs. 6-8) only fig. 8 appears to show fascicostellate ribbing, and it seems possible that more than one species was figured. O. sedgwicki D’Archiac and Verneuil is certainly not a Fascicostella and is here re- jected from this genus. It appears to be a stropheodontid as was pointed out by Shirley (1938, p. 467). However, the shells figured by Schuchert and Cooper (1932, pi. 22, figs. 6, 7, 10, 11, and 16), under this name, are of Fascicostella sp. and require a name. They are figured here (PI. 101, figs. 5-7; PI. 102, figs. 1-2) as F. imdidata sp. nov. Of the other species previously assigned to Fascicostella, F. belgica Mailfieux (1941, p. 17, figs. 1-lfl) is too poorly figured for generic determination and F. ? speciosa Poulsen (1943, p. 14, pi. 1, fig. 21) is not a Fascicostella but possibly an atrypid. The Baton River material from New Zealand includes a new species of Fascicostella here described and figured as F. batonensis. Species assigned to Fascicostella. F. gei villii (Defrance) 1827 ? O. dorsoplicata Freeh 1888, (in part) F. batonensis sp. nov. F. undidata sp. nov. Species rejected from Fascicostella. O. sedgwicki D’Archiac and Verneuil, Schuchert and Cooper 1932, p. 130, pi. 22, figs. 6, 7, 10, 11, and 16, and Sarycheva 1960, pi. 13, fig. 19. O. dorsoplana Freeh, Maillieux 1941, p. 17. F. ? speciosa Poulsen 1943, p. 14, pi. 1, fig. 21. Dalmanella {Fascicostella ?) clarionda Le Maitre 1944, pp. 25-27, pi. 6, figs. 13-17. EXPLANATION OF PLATE 100 Figs. 1, 2. Resserella elegantulina (Davidson), Buildwas Beds (Wenlockian), north bank of River Severn about 400 yd south-east of Buildwas, Shropshire, grid reference SJ 639046. \a-e, pedicle valve, posterior, anterior, brachial valve and lateral views of whole shell USNM 166556, x 5. 2a-c, lateral, posterior, and interior views of brachial valve USNM 166557, x 5. Fig. 3. Resserella basalis (Dalman), shales above Wenlock Limestone (Wenlockian), locality, old lime- stone workings south of the road from Monkswood to Glascoed, Usk inlier, Monmouthshire, grid reference SO 333016. 3u, b, internal mould of brachial valve (Wahusley collection DU. 629), and rubber impression X 4. Fig. 4. Resserella canalis (J. de C. Sowerby), the original of pi. 13, fig. 12a of J. de C. Sowerby in Murchison’s The Silurian System, Geological Survey Museum, London, no. GSM 51550, here designated lectotype. 4a-e, pedicle valve, brachial valve, posterior, anterior, and lateral views, x 3. Fig. 5. Resserella basalis (Dalman), Klinteberg, Gotland, Swedish Museum of Natural History no. Br. 2298, here designated lectotype. 5a-e, pedicle valve, brachial valve, lateral, anterior, and posterior views, x 3. Fig. 6. Resserella concavoconvexa (Twenhofel), zone 9, Jupiter Formation (late Llandoverian), South Point, Anticosti Island, holotype, no. YPM 10339, Peabody Museum of Natural History, Yale University. 6a-e, brachial valve, pedicle valve, posterior, anterior, and lateral views, x 2. Figs. 7, 8. Resserella triangularis (Maurer), Lower Devonian, Germany, specimens in the Senckenberg Museum, Frankfurt. 7, locality Haiger, no. 542g, internal mould of pedicle valve, x 2. 8a, b, locality Karstel (upper Coblenzian), rubber impression and external mould of brachial valve, x 2. Palaeontology, Vol. 14 PLATE 100 WALMSLEY and BOUCOT, Resserellinae WALMSLEY AND BOUCOT: RESSERELLINAE 523 Species requiring further study before assignment. O. dorsoplicata Bedard 1891, pp. 99-100, pi. 3, figs. 6-8. F. belgica Maillieux 1941, p. 17, figs. 1-la. F. gervillei (Defrance), Gill 1942, p. 37, pi. 6, figs. 3-5. Stratigraphic range of Fascicostella. Fascicostella appears to be confined to the Lower Devonian. The reference by Schuchert and Cooper (1932, p. 130) to F. gervillei in the Silurian, was based on Kozlowski’s report of this species in the Borszczow of Podolia, now recognized as Lower Devonian (Early Gedinnian). Fascicostella is known from Europe, North Africa, and New Zealand. Fascicostella gervillii (Defrance) Plate 102, figs. 3-5 1827 Strophomenes gervilii Defrance, p. 152. 1948 Orthis gervillei Barrande, p. 48, pi. 19, fig. 10. 1879 Orthis gervillei Barrande, pi. 58, fig. 10, pi. 126, figs. 3, 4c. 1886 Orthis gervillei Defr. sp; Oehlert, p. 44-45, pi. 4, figs. 45-55. 1912 Orthis gervillei (Defrance); Herrmann, p. 349, pi. 21, fig. 4, ?fig. 5. 1922 Orthis edgelliaiia (Salter MSS.) Davidson; Barrois, Pruvost, and Dubois, p. 80, pi. 11. figs. 21-22 non Davidson. 1929 Dalmanella gervillei (Defrance); Kozlowski, p. 70, pi. 1, fig. 32. 1932 Fascicostella gervillei (Defrance); Schuchert and Cooper, p. 130, pi. 22, figs. 12, 15. non 1938 F. gervillei (Defrance); Shirley, pp. 466^67, pi. 41, figs. 4-6. 1942 F. gervillei (Defrance); Renaud, p. 18, pi. 4, fig. 3. 1952 Dalmanella gervillei (Defrance); Le Maitre, p. 101, pi. 19, figs. 19-22. 1960 F. gervillei (Defrance); Sarycheva, pi. 13, fig. 18. Diagnosis. Fascicostella with non-plicated shell and with finely fascicostellate ribbing. Comparison. Of the three species of Fascicostella here recognized, F. gervillii seems generally to be the smallest. It differs from the other two species in its ribbing which is finely fascicostellate, with no great variation in the size of the costellae. The surface of the shell is not plicated as in the other species. No internal details were available for study. Description. Exterior. Plano-convex, brachial valve with shallow median sulcus widening anteriorly. Outline transversely elliptical to subcircular. Length equals four-fifths width and twice thickness. Ventral beak moderately incurved, not overhanging hinge line, dorsal beak inconspicuous. Cardinal angles slightly rounded. Anterior commissure crenulate and faintly unisulcate. Ventral interarea one-eighth long as wide, gently curved anacline. Costellae uneven, rounded. On pedicle valve, a median bundle of three or four low costellae, flanked by bundles of coarser, elevated costellae. On brachial valve, a triangular median panel of straight fine costellae, is flanked by bundles of more elevated costellae which curve antero-laterally. Distribution. F. gervillii has been recorded from the Lower Devonian of France, Ger- many, Bohemia, Podolia, Turkey, Spain, and North Africa. France. Oehlert (1866, pp. 44-45) from the Calcaire de Vire (Sarthe) of the Brest-Laval syncline, Renaud (1942, p. 19) from the Siegenian of Finistere, from the Siegenian of Ille-et-Vilaine (Bois-Roux), from the limestones of Mayenne (St. Pierre-sur-Erve), from the lower Couvinian of Mayenne near St. Jean and from the Calcaire de Vire (Sarthe), of the Brest-Laval syncline. Germany. Assmann (1910, p. 161) from the Erbsloch-Grauwacke of Kellerwald, Herrmann (1912, p. 394) from the Unreiner kalk of Marburg area. 524 PALAEONTOLOGY VOLUME 14 Bohemia. Barrande (1848, p. 48) from the Koneprusy Limestone. Podolia. Kozlowski (1929, pp. 70-71) from the Borszczow stage. Turkey. Huffner (1917, p. 292), Paeckelmann (1925, p. 116), Paeckelmann and Sieverts (1932, p. 31). Spain. Barrois (1882, p. 238) from the Calcaire de Moniello and Calcaire d’Arnao. North Afriea. Gigout (1951, p. 316) from West Morocco, Le Maitre (1952, p. 102) from Eifelian of El Kseib and Erg Djemel areas of Sahara. It is not certain, however, that all these forms are conspecific. The species of Fasci- costella in the Baton River beds of New Zealand, which was referred to F. gervillei by Shirley (1938, p. 466) is described here as a new species, F. batonensis. Gill (1942, p. 37) recorded F. gervillei from Yeringian Strata (Siegen-Ems) of Victoria, Australia, but con- sidered that the Australian species was conspecific with the Baton River form. It there- fore seems unlikely that it is F. gervillii. Remarks. Kozlowski (1929, p. 70) compared F. gervillii to Dalmanella edgelliana (Salter MSS.) Davidson 1869 (p. 228, pi. 32, figs. 1-4) from the Wenlockian of Britain and re- ferred to the record of D. edgelliana from the calcaire de Lievin, then believed to be Ludlovian (Aymestry), of the north of France, by Barrois, Pruvost, and Dubois (1922, p. 80, pi. 9, figs. 21-22). D. edgelliana should probably be assigned to Proschizophoria. However, the shell figured by Barrois et al. is not P. edgelliana but is most likely F. gervillii and its horizon (calcaire de Lievin), is now recognized as post Ludlovian. Fascicostella batonensis sp. nov. Plate 102, figs. 6-10 1938 Fascicostella gervillei (Defrance); Shirley, p. 466, pi. 41, figs. 4-6. Diagnosis. Fascicostella with finely fascicostellate plicated shell. The brachiophores very widely divergent but not projecting anterior to the cardinal process. Comparison. F. batonensis is distinct from both F. gervillii and F. undulata in its ribbing, which is coarse, angular, and fairly evenly distributed, although the internal moulds show similar but less-prominent plication to that seen in F. undulata. The median ridge in the pedicle valve narrows anteriorly compared with the more sharply defined and parallel-sided ridge in F. undulata and the brachiophores are so widely divergent that their tips do not project anterior to the cardinal process. The anterior limit of both is EXPLANATION OF PLATE 101 Figs. 1-3. Resserella triangularis (Maurer), Lower Devonian, Germany, specimens in the Senckenberg Museum, Frankfurt. \a, b, locality Helmesthal, internal mould of brachial valve, SMF 19983, and rubber impression, x 2. 2a, b, locality Karstel, rubber impression and internal mould of specimen no. 563e, X2. 3n, b, locality Strassebersb, internal mould of pedicle valve no. 542e and rubber impression, x2. Fig. 4. Resserella springfieldensis (Foerste), Cedarville Dolomite (Wenlockian), locality Eastern Mills Quarry, south-west of Springfield, Ohio, syntypes numbered 87122 in Smithsonian Institution, U.S. National Museum. 4n, internal mould of brachial valve, X 2. Ab, c, pedicle valve and lateral view, X 2. Ad, e, internal moulds of pedicle valve, X 1 . Figs. 5-7. Fascicostella undulata sp. nov. Early Devonian, Vire, Sarthe, France. 5a, b, brachial and pedicle valves of whole shell (Harvard Museum 1359), x2. 6a-e, lateral, anterior, posterior, brachial valve and pedicle valve of whole shell (Harvard Museum 1 359b), x 2. 7, interior of brachial valve (Harvard Museum 1359c), X 1. Palaeontology, Vol. 14 PLATE 101 r WALMSLEY and BOUCOT, Ressereliinae WALMSLEY AND BOUCOT: RESSERELLINAE 525 marked by a straight edge to the notothyrial platform, parallel to the hinge line. This is quite different from the arrangement in F. undulata (see PI. 102, figs. 2b and 6c). Description. Exterior. Plano-convex, with flat median area on pedicle valve and an- teriorly widening shallow median sulcus on brachial valve. Outline shield-shaped, width greater than length and about three times thickness. Pedicle valve projects one-seventh total length posterior to hinge line, beak barely incurved. Cardinal angles shghtly rounded. Anterior commissure crenulate and very broadly and gently unisulcate, lateral commissures straight. Hinge line straight, equal to five-eighths greatest width. Ventral interarea slightly curved, apsacline, lateral margins sharp. Dorsal interarea plane, ana- cline. Delthyrium and notothyrium both open and triangular, latter enclosing about 90° and partially occupied by protruding myophore. Costellae sharply angular and fasci- culate, arranged in bundles in which medial costella is higher and more sharply angular. Medial panel of pedicle valve, bears three costellae, central one rather flat. Four curving lateral sectors have four or five costellae each. Median panel of brachial valve has asym- metrically branched costellae. Costellae average 1 per mm at 5 mm length, about 30 costellae on 9 mm wide shell. Interior of pedicle valve. Muscle field wide and well impressed, extends one-third valve length with anterior limit marked by distinct rim. Median ridge low, narrow, separates narrow, elongate, poorly impressed diductor tracks, laterally bounded by slightly raised margins. Adjustor scars on curved floor of wide delthyrial cavity. Laterally divergent dental lamellae support small, triangular teeth with deep crural fossettes at junction, and laterally directed accessory sockets. Deep lateral cavities extend beneath interarea. Internal surface of valve anterior to muscle field, marked by broad undulations reflecting external fascicostellate pattern. Interior of brachial valve. Weakly impressed adductor muscle field occupies median one-third of posterior half of valve. Median ridge, broad, low, rounded, one-third width of muscle field, narrows slightly anteriorly. Weak transverse ridges curving antero- laterally from median ridge, separate smaller anterior impressions from posterior pair which are bounded laterally by slightly raised curved margins. Median ridge thickened between brachiophores forming a steep wall with anterior edges of brachiophores. Brachiophores widely divergent, sharply pointed triangular plates, laterally inclined. Socket pads support triangular, crenulated sockets. Thick shaft of cardinal process ex- pands into bilobed myophore with crenulated posterior face. Type specimens. Specimen USNM 166563 figured here, Plate 102, figs. M-d, is here selected as holotype. Specimens USNM 166561-166562 and 166564-166565 figured Plate 102, figs. 6, 7, 9, and 10, are paratypes. Distribution. F. batonensis is known from the Baton River Beds, New Zealand. As indi- cated by Boucot, Johnson, and Talent (1967, p. 1242), the brachiopod fauna of these beds suggests correlation with the Coopers Creek Formation of Gippsland, Victoria, Australia, the age of which they take to be Siegenian. Philip and Pedder (1967, p. 233 and text-fig. 1), also suggest a Siegenian age for the Coopers Creek Formation, based on conodont evidence. 526 PALAEONTOLOGY VOLUME 14 Fascicostella imdulata sp. nov. Plate 101, figs. 5-7; Plate 102, figs. 1, 2 1932 O. sedgwicki D’Archiac and Verneuil; Schuchert and Cooper, p. 130, pi. 22, figs. 6, 7 10, 11, and 16. Diagnosis. Fascicostella with a plicated and coarsely fascicostellate shell on which 10-12 ‘bundles’ of costellae are developed on each valve. Brachiophores which extend anterior to the cardinal process diverge from the prominent median ridge at an angle of about 45°. Comparison. F. imdulata is readily distinguished from the other two species of Fasci- costella by its distinctive ribbing in which 10-12 prominent ‘bundles’ of costellae on each valve, coincide with plications which are reflected on the internal moulds as broad undulations. The brachiophores are less divergent than in F. batonensis and extend anterior to the cardinal process (PL 102, figs. 2b and 6c). Both the median ridge and the margin of the muscle field in the brachial valve are more strongly developed than in F. batonensis and in the pedicle valve, the median ridge is more sharply defined and parallel- sided than that of F. batonensis. No interiors of F. gervillii were available for comparison. Description. Exterior. Plano-convex, convexity of pedicle valve decreasing antero- laterally, convexity of brachial valve increasing slightly laterally. Length slightly less than width and two and a half times thickness. Outline sub-circular, cardinal angles slightly rounded, pedicle valve projects one-seventh total length posterior to hinge hne with slightly incurved beak. Commissure crenulate. Hinge line straight, equal to three- quarters maximum width which is at mid- length. About 10 or 12 angular primary costae each bifurcate into bundles of 5 or 6 fine costellae enclosing a high angular median costa. These bundles reflected on inner surface as some 12 broad undulations. Ventral interarea one-fifth as long as wide, slightly curved, apsacline lateral margins sharp. Dorsal inter- area plane, anacline. Both delthyrium and notothyrium open, triangular, latter filled with protruding myophore. EXPLANATION OF PLATE 102 Figs. 1, 2. Fascicostella imdulata sp. nov. Early Devonian ? Vire, Sarthe, France, la-c, posterior, interior, and exterior views of brachial valve (Harvard Museum 1359d), here selected holotype, x 3. la-e, posterior, dorsal, ventral, anterior, and lateral views of steinkern (Harvard Museum 1359a), x2. Figs. 3-5. Fascicostella gervillii (Defrance), Koneprusy Limestone (Pragian), Bohemia. 3, brachial valve USNM 166558, xl. 4, brachial valve USNM 166559, x2. 5a-g, antero-ventral, antero- dorsal, lateral, brachial valve, pedicle valve, postero-ventral, and posterior views of whole shell, USNM 166560, X IJ. Figs. 6-10. Fascicostella batonensis sp. nov., member 3 of Willis, Baton River Formation (Siegenian), USNM locality 1 1979, bluffs 300 ft above track, north bank of Baton River between 500 and 2000 ft upstream of mouth of Heine Stream and about 400 yd upstream of a swingbridge that crosses Baton River, New Zealand, approximate grid reference S19 062273. 6a-d, internal mould of pedicle valve of steinkern USNM 166561, and rubber impression, and internal mould of brachial valve and rubber impression, x 2. la-d, internal mould of brachial valve USNM 166562, and rubber im- pression and external mould of same specimen and rubber impression, x 2. 8a-ff, external mould of brachial valve USNM 166563, and rubber impression and internal mould of same specimen and rubber impression, x 2. 9a, b, external mould of pedicle valve USNM 166564, and rubber impres- sion, x2. 10a, b, internal mould of pedicle valve USNM 166565, and rubber impression, x2. Palaeontology , Vol. 14 PLATE 102 WALMSLEY and BOUCOT, Resserellinae WALMSLEY AND BOUCOT: RESSERELLINAE 527 Interior of pedicle valve. Muscle field confined to median one-third of posterior one- third of valve, delimited by distinct rim. Median raised area, one-third width of muscle field separates wide diductor impressions and is continued as a low, narrow, median ridge to two-thirds valve length. Interior of brachial valve. Muscle field occupies median one-third of valve extending two-thirds valve length. Median ridge, broad, low, and rounded, extends length of muscle field. Faint antero-lateral transverse ridges. Thick stubby brachiophores diverge antero-laterally at 90° from each other and are laterally inclined into socket pads. Sockets triangular and crenulated. Narrow shaft of cardinal process expands into a bulbous protruding myophore. Type specimens. Specimen no. 1359d, a brachial valve, figured here Plate 102, figs. \a-c, is here selected as holotype. Specimen no. 1359a, figured Plate 102, figs. 2a-e, is a paratype. Distribution. This species has been described from the material figured by Schuchert and Cooper (1932, pi. 22, figs. 6, 7, 10, 11, and 16) and referred by them to F. sedgwicki (D’Archiac and Verneuil). Their figures were of specimens numbered 1359 in the collec- tion of the Museum of Comparative Zoology, Flarvard. These specimens are refigured here in our Plate 101, figs. 5-7, Plate 102, figs. 1-2. They are labelled as coming from the Devonian, Vise, Belgium. No further details are known. Acknowledgements. We are very much indebted to a number of people who have donated or loaned specimens. Dr. H. Mutvei and Dr. V. Jaanusson of the Natural History Museum, Stockholm, gave assistance in tracing Dahuan’s original material, helped in the selection of lectotypes, of R. elegantula and R. basalis and arranged for us to borrow spechnens. Dr. J. E. Hede kindly provided additional material of R. basalis from Gotland. Mr. J. D. D. Smith arranged the loan of Sowerby’s figured speci- men of R. canalis from the Geological Survey Museum, London. Dr. V. Havlicek very kindly sent us material of Dedzetina macrostomoides for examination and Dr. W. Struve arranged the loan of speci- mens of R. triangularis from the Senckenberg Museum. Dr. G. A. Cooper loaned us syntypes of R. springfieldensis from the National Museum, Washington. Dr. K. Waage of the Peabody Museum, Yale, loaned R. concavoconvexa type material and specimens of R. brownsportensis. Additional material of R. brownsportensis was sent by Dr. T. Amsden. Professor H. B. Whittington arranged the loan of F. sedgwicki and O. media from the Museum of Comparative Zoology, Harvard, and Mr. Powell of the Oxford University Museum kindly loaned resserellid material collected by Drs. A. Ziegler and L. R. M. Cocks. We are grateful to Mrs. G. Lewis and Mr. F. Cross for drawing the text-figures and to Mrs. R. Powell for the typing. Also to Dr. M. G. Bassett for help in photographing specimens of R. triangu- laris-, to Dr. L. R. M. Cocks of the British Museum (Natural History) for help in selecting the lecto- type of R. elegantulina and to Dr. J. G. Johnson for reading the paper. The collection of material by Boucot was made possible in 1955-1956 by a John Simon Guggenheim Fellowship. The work in Pasadena by Walmsley and collecting subsequent to 1957 was supported by grants supervised by Boucot, to the Massachusetts Institute of Technology and the California Institute of Technology by the National Science Foundation. Natural Environment Research Council Grant No. B/SR/1734 supported the work in Swansea by Walmsley and enabled him to collect from Gotland localities and to select lectotypes of Dalman’s material at the Natural History Museum, Stockholm, and study resserellid material at the Senckenberg Museum, Frankfurt, and the Hessisches Landesmuseum, Darmstadt. We gratefully acknowledge this assistance. 528 PALAEONTOLOGY VOLUME 14 REFERENCES AMSDEN, T. w. 1949. Stratigraphy and paleontology of the Brownsport formation (Silurian) of western Tennessee. Bull. Peabody Mas. fiat. Hist. 5, 1-138, pis. 1-34. 1951. Brachiopods of the Henryhouse Formation (Silurian) of Oklahoma. J. Paleont. 25, 69-96, pis. 15-20. 1958. Stratigraphy and Paleontology of the Hunton Group in the Arbuckle Mountain Region. Pt. 2 — Haragan Articulate brachiopods. Pt. 3 — Supplement to the Henryhouse brachiopods. Pt. 4 (with A. J. Boucot) — New genera of brachiopods. BuU. Okla. geol. Surv. 78, 1-199, pis. 1-14. d’archiac, e. j. a. d. de ST. s. and verneuil, p. e. p. de, 1842. On the fossils of the Older Deposits of the Rhenish Provinces, preceded by a General Survey of the Fauna of the Palaeozoic Rocks and followed by a Tabular List of the Organic Remains of the Devonian System in Europe. Trans. geoL Soc. ser. 2, 6, 303-410, pis. 25-38. ASSMANN, p. 1910. Die fauna der Erbsloch-Grauwacke bei Densberg im Kellerwald. Jb. K. preuss. geol. Landesanst. 31 (1), 136-72, pis. 6-11. BANCROFT, B. B. 1928. On the notational representation of the rib-system in Orthacea. Mem. Proc. Maiicbr. lit. phil. soc. 72, 53-90, pis. 1-3. 1929. Some new genera and species of Strophomenacea from the Upper Ordovician of Shrop- shire. Ibid. 33, 33-65, pis. 1, 2. 1945. The brachiopod zonal indices of the Stages Costonian to Onnian in Britain. J. Paleont. 19, 181-252, pis. 22-38. BARRANDE, J. 1848. Liber die Brachiopoden der silurisclien Schichten von Boehmen aus den Naturwissen- schaftliclien Abhandlungen, 2 (2), 1-153, pis. 15-23. 1879. Systeine siliirien dii centre de la Boheme, 5, Brachiopodes, 1-226, 153 pis., Prague, Paris. BARROis, c. 1882. Recherches sur les terrains anciens des Asturies et de la Galice. Mem. Soc. geol. die Nord, 2 (1), 1-630, pis. 1-20. PRUvosT, p., and dubois, g. 1922. Description de la faune siluro-devonienne de Lievin (Brachio- podes), Ibid. 6 (2), 71-225, pis. 10-17. BASSETT, M. G. Wenlockian articulate brachiopods of the Welsh Borderland and South Wales. Un- published Ph.D. thesis. University of Wales, 1-403, pis. 1-40. In preparation. The articulate brachiopods from the Wenlock Series of the Welsh Borderland and South Wales. Palaeontogr. Soc. (Monogr.). BECLARD, F. 1891. Fossiles nouveaux de devonien inferieur de la Belgique, Bull. Soc. beige. GeoL Paleont. Hydrol. 5, 96-102, pis. 3-4. BORiSYAK, M. A. 1955. Siluiiaii (Wenlockian) brachiopods from Karaganda oblast. Material on the stratigraphy and fauna of the Ordovician and Silurian sediments of Central Kazakhstan, No. 1, Mater, vses. nauclino.-issled. geol. inst. Paleont. i Stratigr. 3, 1-92. BOUCOT, A. J., JOHNSON, J. G., and WALMSLEY, v. G. 1965. Revision of the Rhipidomellidae (Brachio- poda) and the affinities of Mendacella and Dalejina. J. Paleont. 39, 331-340, pis. 45, 46. HARPER, c. w., and walmsley, v. g. 1966. Silurian Brachiopods and Gastropods of Southern New Brunswick, Bull. Can. geol. Surv. 140, 1^4, pis. 1-18. and TALENT, J. A. 1967. Lower and Middle Devonian Faunal Provinces based on Brachio- poda. Internat. Symp. Dev. Syst. Calgary, Alberta Soc. Petrol. Geol. 2, 1239-1254. BUCH, L. VON. 1938. Uber Delthyris oder Spirifer und Orthis. Abh. K. Wiss. Berlin, 1-80. CASTELLARO, H. A. 1959. Braquiopodos gotlandicos de la Precordillera de San Juan. Revta. Assoc. geoL argent. Diet. 13, 41-65, pis. 1-5. CAVE, R. and dean, w. t. 1959. Four British Ordovician species of DahnaneUoid Brachiopod. Palaeon- tology, 1, 292-297, pi. 53. COOPER, G. A. 1956. Chazyan and related brachiopods. Smithson, misc. Colins, 127, Pub. 4253, i-xvi, 1-1245, pis. 1-269. CURTIS, M. L. K. 1956. Lower Palaeozoic. In macinnes, c. m., and whittard, w. f. (eds.), Bristol and its Adjoining Counties, 3-7. Brit. Assn. Adv. Sci. Bristol. DAHMER, G. 1951. Die Fauna der nach-Ordovizischen Glieder der VerseSchichten mit Ausschluss der Trilobiten, Crinoiden und Anthozen. Palaeontographica, Abt. A, 101, 1-152, 12 pis. dalman, j. w. 1828. UppstaUning och beskrifning af de i Sverige funne Terebratulifer. Svenska Vetens- kaps-academiens, Handlingar, ser. 3, 15-16, 1827, 85-155, 5 pis. WALMSLEY AND BOUCOT: RESSERELLINAE 529 DAVIDSON, T. 1847. Observations on some of the Wenlock Limestone Brachiopoda with descriptions of several new species. London Geol. Jl, 1, 52-65, pis. 12-13. • 1848. Memoire sur les Brachiopodes du systeme silurien superieur d’Angleterre. Bull Soc. geol. Fr. ser. 2, 5, 309-338, 370-374, pi. 3. 1869. A Monograph of the British fossil Brachiopoda, vol. 3, Palaeontogr. Soc. {Monograph), 1864-71, 1-397, 50 pis. 1881. Descriptions of new Upper Silurian Brachiopoda from Shropshire. Geol. Mag. dec. 2, 8, 145-156, pi. 5. ■ 1883. A Monograph of the British fossil Brachiopoda, 5 (2), Silurian Supplement. Palaeontogr. Soc. [Monograph], 63-242, pis. 4-11. and MAW, G. 1881. Notes on the physical character and thickness of the Upper Silurian rocks of Shropshire, with the Brachiopoda they contain grouped in geological horizons. Geol. Mag. dec. 2, 8, 100-110. DEFRANCE, M. J. L. 1827. Li CUVIER, F. G. (cd.), Dictiotmalre des sciences naturelles, 51, Paris, Strasbourg. FOERSTE, A. F. 1917. Notes on Silurian Fossils from Ohio and other Central States. Ohio Sci. 17, 187- 258, pis. 8-12. FRECH, F. 1888. Geologic der Umgegend von Haiger bei Dillenburg (Nassau). Abhandl. zitr geol. Special karte von Prenssen and den Thuringischen Staaten, 8 (3) 1-36, pis. 1-3, Berlin. GiGOUT, M. 1951. Etudes geologiques sur la Meseta Marocaine Occidentale. Notes et MOnoires No. 86, Maroc Serv. Geol. Travaux de Vlnst. Sci. Cherifien, 1 (3), 259-396. GILL, E. D. 1942. On the thickness and age of the Type Yeringian strata, Lilydale, Victoria. Proc. R. Soc. Viet. 54 (1), 21-52, pis. 4-6. 1948. Eldon group fossils. Rec. Queen Viet. Mus., Launceston, Tasmania, 2, 57-74. 1950. Palaeontology and palaeoecology of the Eldon group. Pap. Proc. R. Soc. Tasm. 231-258. HALL, J. 1879. The Fauna of the Niagara Group in Central Indiana. 28th Rep. New York State Mus. Nat. Hist. 99-203, pis. 3-34, Albany. HARPER, c. w., BOUCOT, A. J., and WALMSLEY, v. G. 1969. The Rhipidomellid brachiopod subfamilies Heterorthinae and Platyorthinae (new). J. Paleont. 43, 74-92, pis. 15-17. HAVLiCEK, V. 1950. The Ordovician Brachiopoda from Bohemia. Ramenonozci ceskeho Ordoviku. Rozpr. Vstfed. list, geol., Praha, 13, 1-135, pis. 1-13. • 1956. The Brachiopods of the Branik and Hlubocepy Limestones in the immediate vicinity of Prague. Sbornik Ustfed. listavu geol., svazek XXII-1955 — oddll paleontologicky, pi. I (381-XII (49), 535-665. HERRMANN, F. 1912. Das hcrcynischc Unterdevon bei Marburg a. L. Jb. Preuss-geol. Landesanst. {Berg. Akad.) 23 (1), 305-395, pis. 18-23. HisiNGER, w. VON. 1837-1841. Lethaea svecica seu Petrefacta svecica. HUFFNER, E. 1917. Bcitragc zur Kenntnis des Devons von Bithynien. Jb. preuss. geol. Landesanst. {Berg. Akad.) 27 (2) 260-357, pis. 28-30. JOHNSON, J. G., BOUCOT, A. J., and MURPHY, M. A. 1967. Lower Devonian faunal succession in Central Nevada. Internat. Symposium on the Devonian System, Calgary, Alberta. Alberta Soc. Petrol. Geol. 2, 679-691. and TALENT, J. A. 1967. Cortezorthinae, a new subfamily of Siluro-Devonian dahnanellid brachio- pods. Palaeontology, 10, 142-170, pis. 19-22. KAPLUN, L. i. 1961. Lower Devonian brachiopods of the northern Pribalkhash in Materials for the geology and guide fossils of Kazakhstan, pt. 1 (26), Stratigraphy and palaeontology. Minist. geol. okvan nedr. Kazakh. SSR, 64-114, pis. 7-19. [In Russian.] KOZLowsKi, R. 1929. Les brachiopodes gothlandiens de la Podolie polonaise. Pu/ucuul. pu/. 1, 1-254, 12 pis. kul’kov, n. p. 1963. Brachiopodi solovikhinskikh sloev nizhnego Devona gornogo Altai. Akad. nauk SSSR, 132 pp., 9 pis. LAMONT, A. 1934. Brachiopod Morphology in relation to Environment. Cement, Lime and Gravel, 8, 216-219. le maitre, d. 1944. Contribution a I’etude du devonien du Tafilet I. La faune coblencienne de Haci Remlia (S.W. de Taouz). Notes et mem. serv. geol. Protectorat de VEtat frangais au Maroc, 61, Laval. 530 PALAEONTOLOGY VOLUME 14 LE MAiTRE, D. 1952. La faunc du devonien inferieur et moyen de la Saoura et des abords de TErg el Djemel (Sud-Oranais). Mater. Carte geol. Alger, ser. 1, Paleont. 12, 1-170. LiNDSTROM, G. 1861. Bidrag till Kannedomen om Gotlands Brachiopoder. Fork. Ofvers Vetenskaps- akad. 17 (60), 337-382. MAiLLiEUX, E. 1941. Lcs Brachiopodcs de I’Emsien de I’Ardenne: Mem. Mas. R. Hist. nat. beige, 96, 1-74. MAURER, F. 1889. Palaeontologische Studien im Gebeit des rheinischen Devon. 8 Mitheilungen iiber Fauna und Gliederung des rechsrheinischen Unterdevon. Neiies Jb. Miner. 2, 160, pi. 3. MCLEARN, F. H. 1924. Palaeontology of the Silurian rocks of Arisaig, Nova Scotia. Mem. Geol. Siirv. Can. 137, 1-179, pis. 1-30. NIKIFOROVA, o. I. 1954. Stratigraphy and brachiopods of the Silurian deposits of Podolia. Trudy Vses. nauclmo.-issled. geol. inst. 1-218 [in Russian]. OEHLERT, D. 1886. Etude sur quelques fossiles devoniens de I’ouest de la France. Ann. Sci. Geol. 19, 1-80, pis. 1-5. PAECKELMANN, w. 1925. Beitrage zur Kenntnis des Devons am Bosporus insbesondere in Bithynien Abh. Preuss. geol. Landesanst. n.f. 98, 1-152, pis. 1-6. and siEVERTS, h. 1932. Neue Beitrage zur Kenntnis der Geologic, Palaeontologie und Petro- graphic der Umgegend von Konstantinopel; 1. Obersilurische und devonische Faunen der Prin- zeninseln, Bithyniens und Thraziens. Ibid. 142, 1-79, pis. 1^. PHILIP, G. M. 1962. The Palaeontology and Stratigraphy of the Siluro-Devonian sediments of the Tyers area, Gippsland, Victoria. Proc. R. Soc. Viet. 75 (2), 123-246, pis. 11-36. 1965. Lower Devonian conodonts from the Tyers area Gippsland, Victoria, Ibid. 79, 95-118. and PEDDER, A. E. H. 1967. A correlation of some Devonian Limestones of New South Wales and Victoria. Geol. Mag. 104, 232-239. POULSEN, c. 1943. Brachiopoda, pt. 11 of the fauna of the Offley Island Formation, pt. 11 of the Silurian faunas of north Greenland: Kommissionen for videnskabelige undersogelser. i Gronland. Meddr om Gronland. 72 (3) 58 pp., 6 pis. REED, F. R. c. 1912. Silurian fossils from Kashmir. Records Geol. Surv. India, 42 (1), 16-33, pi. 9. 1917. The Ordovician and Silurian Brachiopoda of the Girvan District, Trans. R. Soc. Edinb. 51 (4), 795-998, pis. 1-24. RENAUD, A. 1942. Le devonien du Synclinorium Median Brest-Laval. Meju. Soc. Geol. Min. Bretagne 1 (2), 1-200, pi. 1-14. ROEMER, F. 1860. Die silurische Fauna des westlichen Tennessee, Breslau, 100 pp., 5 pis. RUBEL, M. 1963. Silurian orthid brachiopods of Estonia. Geoloogia Instituudi Uurimused. Eesti N.S.V. teaduste akadeemia, Tartu, 109-160, pis. 1-8 [in Russian]. SARYCHEVA, T. G. (cd.), 1960. Osnovy paleontologii, mshanki, brakhiopodi, Moscow, 1-343, pis. 1-7, 1-75. scHUCHERT, c., and COOPER, G. A. 1931. Synopsis of the brachiopod genera of the suborders Orthoidea and Pentameroidea with notes on the Telotremata. Am. J. Sci., 5th ser. 22, 241-251. 1932. Genera of the suborders Orthoidea and Pentameroidea. Mem. Peabody Mas. Nat. Hist., New Haven, 4 (1), 1-270, pis. A and 1-29. SHIRLEY, j. 1938. The fauna of the Baton River beds (Devonian), New Zealand. Q. Jl geol. Soc. Land. 94, 459-506, pis. 40-44. sowERBY, J. DE c. 1839. On the fossil shells of the Silurian Rocks. In murchsion, r. l, The Silurian System, 2, 608-722, pis. 1-27, London. TWENHOFEL, w. H. 1927. Gcology of Anticosti Island. Mem. Geol. Surv. Can. 154, 1-481. WALMSLEY, V. G. 1959. The geology of the Usk inlier (Monmouthshire). Q. Jl geol. Soc. Lond. 114, 483-521. 1965. Isorthis and Salopina (Brachiopoda) in the Ludlovian of the Welsh Borderland. Palaeon- tology, 8, 454-77, pis. 61-65. (In press). Silurian dalmanellid Brachiopods of Britain. 3rd Internat. Symp. Siluro-Devonian, Leningrad 1968, Symposium Volume. BoucoT, A. J., and harper, c. w. 1969. Silurian and Lower Devonian salopinid brachiopods. J. Paleont. 43, 492-516, pis. 71-80. ■ and SAVAGE, n. m. 1968. Visbyella — a new genus of resserellid brachiopod. Palaeon- tology, 11, 306-316, pis. 60-62. WALMSLEY AND BOUCOT; RESSERELLINAE 531 WANG, y. 1949. Maquoketa Brachiopoda of Iowa. Mem. geol. Soc. Am. 42, 55 pp., 12 pis. WELLER, s. 1903. The Palaeozoic faunas. New Jersey Geol. Surv. Kept, on Palaeontology, 3, 1-462, 53 pis. WHiTTARD, w. F. and BARKER, GLENYS H. 1950. The Upper Valentian Brachiopod Fauna of Shropshire — Pt. 1. Inarticulata ; Articulata, Protremata, Orthoidea. Ann. Mag. nat. Hist. {12), 3, 553-590, pi. v-viii. WHITTINGTON, H. B. 1938u. Fauna of the Finest Quarry, Llanfyllin {Wattsella horderleyensis Superzone) and its correlation. Proc. Geol. Ass. 49, 49-54, pi. 6. 1938^. New Caradocian brachiopods from the Berwyn Hills, North Wales. Ann. Mag. nat. Hist. {11) 2, 241-259, pis. 10, 11. and WILLIAMS, a. 1955. The fauna of the Derfel Limestone of the Arenig District, North Wales. Phil. Trans. R. Soc. ser. B, 238, 397-430, pis. 38-40. WILLIAMS, A. 1949. New Lower Ordovician brachiopods from the Llandeilo-Llangadock District. Geol. Mag. 86, 161-174, 226-238, pis. 8-9. 1951. Llandovery brachiopods from Wales with special reference to the Llandovery district. Q. Jl geol. Soc. Lond. 108, 85-136, pis. 3-8. ■ 1963. The Caradocian brachiopod faunas of the Bala district, Merionethshire. Ball. Br. Mas. nat. Hist. {Geol.) 8 (7), 327-471, pis. 1-16. WILLIAMS, M. Y. 1919. The Silurian geology and faunas of Ontario Peninsula and Manitoulin and adjacent Islands. Mem. geol. Surv. Can. Ill, 117-125, pis. 6-20. ZEiLER, F. 1857. Versteinerungen des iilteren rheinischen Grauwacke. Verb, natiirh. Ver. preiiss. Rheinl. 14, pp. 45-51, pis. 3-4. v. G. WALMSLEY Department of Geology University College of Swansea University of Wales A. J. BOUCOT Department of Geology Oregon State University Corvallis, Oregon 97331 Revised manuscript received 16 January 1971 1 J PALAEONTOLOGY The journal Palaeontology is devoted to the publication of papers (preferably illus- trated) on all aspects of palaeontology and stratigraphical palaeontology. Four parts at least are published each year and are sent free to all members of the Association. Members who join for 1971 will receive Volume 14, parts 1 to 4. All back numbers are still in print and may be ordered from B. H. Blackwell, Broad Street, Oxford, England, at £5 per part (post free). A complete set. Volumes 1-13, consists of 51 parts and costs £255. SPECIAL PAPERS IN PALAEONTOLOGY This is a series of substantial separate works published by the Association. The sub- scription rate is £6 (U.S. $16.00) for Institute Members and £3 (U.S. $8.00) for Ordinary and Student Members. Subscriptions and orders by members of the Asso- ciation should be placed through the Membership Treasurer, Dr. A. J. Lloyd, Depart- ment of Geology, University College, Gower Street, London, W.C. 1, England. The following Special Papers are available. Members may obtain them at reduced rates through the Membership Treasurer. Non-members may obtain them from B. H. Blackwell, Broad Street, Oxford, England, at the prices indicated. Special Paper No. 1 (for 1967); Miospores in the Coal Seams of the Carboni- ferous OF Great Britain, by A. H. V. Smith and M. A. Butterworth. 324 pp., 72 text-figs., 27 plates. Price £8 (U.S. $22.00), post free. Special Paper No. 2 (for 1968): Evolution of the Shell Structure of Articulate Brachiopods, by A. Williams. 55 pp., 27 text-figs., 24 plates. Price £5 (U.S. $13.00). Special Paper No. 3 (for 1968): Upper Maestrichtian Radiolaria of California, by Helen P. Foreman. 82 pp., 8 plates. Price £3 (U.S. $8.00). Special Paper No. 4 (for 1969): Lower Turonian Ammonites from Israel, by R. Freund and M. Raab. 83 pp., 15 text-figs., 10 plates. Price £3 (U.S. $8.00). Special Paper No. 5 (for 1969): Chitinozoa from the Ordovician Viola and Fernvale Limestones of the Arbuckle Mountains, Oklahoma, by W. A. M. Jenkins. 44 pp., 10 text-figs., 9 plates. Price £2 (U.S. $5.00). Special Paper No. 6 (for 1969): Ammonoidea from the Mata Series (Santonian- Maastrichtian) of New Zealand, by R. A. Henderson. 82 pp., 13 text-figs., 15 plates. Price £3 (U.S. $8.00). Special Paper No. 7 (for 1970): Shell Structure of the Craniacea and other Calcareous Inarticulate Brachiopoda, by A. Williams and A. D. Wright. 51 pp., 17 text-figs., 15 plates. Price £1-50 (U.S. $4.00). Special Paper No. 8 (for 1971): Cenomanian Ammonites from Southern England, by W. J. Kennedy. 5 tables, 64 plates. In the press. Special Paper No. 9 (for 1971): Fish from the Freshwater Lower Cretaceous OF Victoria, Australia, with Comments on the Palaeoenvironment, by M. Waldman. With 37 text-figs, and 18 plates. Price £5. Special Paper No. 10 (for 1971): Upper Cretaceous Ostracoda from the Carnarvon Basin, Western Australia, by R. H. Bate. With 43 text-figs, and 27 plates. Price £5 SUBMISSION OF PAPERS Typescripts on all aspects of palaeontology and stratigraphical palaeontology are invited. They should conform in style to those already published in this journal, and should be sent to Mr. N. F. Hughes, Department of Geology, Sedgwick Museum, Downing Street, Cambridge, England, who will supply detailed instructions for authors on request (these are pubhshed in Palaeontology, 10, pp. 707-12). PALAEONTOLOGY VOLUME 14 • PART 3 CONTENTS Lower Devonian corals and bryozoa from the Lick Hole Formation of New South Wales. By a. e. h. pedder 371 Brachiopods from the Lower Devonian Mandagery Park Formation, New South Wales. By norman m. savage 387 Shell structure of the siphonotretacean Brachiopoda. By gertruda biernat and ALWYN WILLIAMS 423 Ontogeny of Vesicaspora, a Late Pennsylvanian pollen grain. By John w. hall and BENTON M. STIDD 431 Mantelliceras saxbii, and the horizon of the Martimpreyi Zone in the Ceno- manian of England. By w. i. Kennedy and J. m. Hancock 437 Silurian calymenid trilobites from the United States, Norway, and Sweden. By H. B. WHITTINGTON 455 Development of Glyptograptus hudsoni sp. nov. from Southampton Island, North-West Territories, Canada. By D. E. jackson 478 The Resserellinae — a new subfamily of Late Ordovician to Early Devonian dalmanellid brachiopods. By v. G. walmsley and A. J. boucot 487 PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS, OXFORD BY VIVIAN RIDLER, PRINTER TO THE UNIVERSITY VOLUME 14 • PART 4 Palaeontology NOVEMBER 1971 PUBLISHED BY THE PALAEONTOLOGICAL ASSOCIATION LONDON THE PALAEONTOLOGICAL ASSOCIATION The Association was founded in 1957 to further the study of palaeontology. It holds meetings and demonstrations, and publishes the quarterly journal Palaeontology and Special Papers in Palaeontology. Membership is open to individuals, institutions, libraries, etc., on payment of the appropriate subscription; Institute membership £10-00 (U.S. $26.00) Ordinary membership £5-00 (U.S. $13.00) Student membership £3-00 (U.S. $8.00) There is no admission fee. Institute membership is only available by direct apph- cation, not through agents. Student members are persons receiving full-time instruc- tion at educational institutions recognized by the Council; on first applying for membership, they should obtain an application form from the Membership Treasurer. All subscriptions are due each January, and should be sent to the Membership Treasurer, Dr. A. J. Lloyd, Department of Geology, University College, Gower Street, London, W.C. 1, England. COUNCIL 1971-2 President'. Dr. W. S. McKerrow, Department of Geology, Oxford Vice-Presidents : Professor M. R. House, The University, Kingston upon HuU, Yorkshire Dr. Gwyn Thomas, Department of Geology, Imperial College, London, W.C.7 Treasurer'. Dr. J. M, Hancock, Department of Geology, King’s College, London, W.C. 2 Membership Treasurer: Dr. A. J. Lloyd, Department of Geology, University College, Gower Street, London, W.C. 1 Secretary: Dr. W. D. I. Rolfe, Hunterian Museum, The University, Glasgow, W. 2 Editors Mr. N. F. Hughes, Sedgwick Museum, Cambridge Dr. Isles Strachan, Department of Geology, The University, Birmingham 15 Dr. R. Goldring, Department of Geology, The University, Reading, Berks. Dr. J. D. Hudson, Department of Geology, The University, Leicester Dr. D. J. Gobbett, Sedgwick Museum, Cambridge Other members of Council Dr. E. N. K. Clarkson, Edinburgh Dr. B. Owens, Leeds Dr. L. R. M. Cocks, London Dr. W. H. C. Ramsbottom, Leeds Dr. R. H. Cummings, Abergele Dr. Pamela L. Robinson, London Dr. Julia Hubbard, London {co-opted) Dr. E. P. F. Rose, London Dr. W. J. Kennedy, Oxford Dr. C. T. Scrutton, Newcastle Mr. M. Mitchell, Leeds Dr. V. G. Walmsley, Swansea Dr. Marjorie D. Muir, London Dr. A. D. Wright, Belfast Overseas Representatives Australia : Professor Dorothy Hill, Department of Geology, University of Queensland, Brisbane Canada: Dr. D. J. McLaren, Institute of Sedimentary and Petroleum Geology, 3303- 33rd Street NW., Calgary, Alberta India: Professor M. R. Sahni, 98 The Mall, Lucknow (U.P.), India New Zealand: Dr. C. A. Fleming, New Zealand Geological Survey, P.O. Box 30368, Lower Hutt West Indies and Central America: Mr. John B. Saunders, Geological Laboratory, Texaco Trinidad, Inc., Pointe-^-Pierre, Trinidad, West Indies Western U.S.A. : Professor J. Wyatt Durham, Department of Paleontology, Univer- sity of California, Berkeley 4, California Eastern U.S.A. : Professor J. W. Wells, Department of Geology, Cornell University, Ithaca, New York © The Palaeontological Association, 1971 PALYNOLOGY OF LOWER CRETACEOUS (SWAN RIVER) STRATA OF SASKATCHEWAN AND MANITOBA by GEOFFREY PLAYFORD Abstract. A varied and generally well-preserved suite of spores and pollen grains is disclosed by palynological study of two sequences (one subsurface, the other type-outcrop) of the Swan River Group in the western Canadian provinces of Saskatchewan and Manitoba. The Swan River Group represents the initial Cretaceous continental-transgressive phase of the far eastern portion of the western Canada sedimentary basin, and rests unconformably upon Jurassic and older rocks; its uppermost horizons contain microplanktonic indications of a marine environment and pass transitionally upwards into the shaly marine Ashville Group. The palynological data indicate a wholly Albian age for the subject Swan River samples by comparison with other palynologically studied North American Lower Cretaceous strata (particularly of Alberta). More precisely the age appears to be Middle Albian, possibly to early Late Albian. The microflora includes 108 species of miospores, some of which are accorded detailed systematic description. Kraeiiselisporites hastilobatus, Crybelosporites bremerii, and Coptospora wiUiamsii are instituted as new species ; Romeisporites Pocock 1962 is shown to be a junior synonym of Triporoletes Mtchedlishvili 1960. Natural affini- ties of the bulk of the Swan River palynomorphs are with the Coniferophyta, Pteridophyta (notably Filicales), and Bryophyta. Simple tricolpate angiospermous grains appear in younger horizons of the group; their introduc- tion serves as a useful biostratigraphic marker, recognized elsewhere in North America. Closest extra-North American comparisons can be made with Albian palynological floras of Siberia and eastern Australia. This paper contains results of palynological studies of Early Cretaceous sediments penetrated by a borehole near Yorkton, south-eastern Saskatchewan and exposed to the north-east in the Swan River area of south-western Manitoba (text-fig. 1). This region is situated close to the eastern limit of the western Canada sedimentary basin, where Lower Cretaceous sandy sediments, of largely non-marine character, form a relatively thin veneer along the western margin of the Canadian Precambrian Shield. Despite extensive drilling activity the Lower Cretaceous stratigraphy of the area awaits detailed integrated study, following the pioneer, broadly based work of Wickenden (1945). Price (1963) investigated the Lower Cretaceous succession of southern Saskat- chewan (principally the south-eastern portion of the province); he recognized an in- formally designated ‘Basal Cretaceous Sandy Group’ succeeded conformably by the Ashville Group (marine shale, glauconitic beds). According to Price, the basal sandy unit is divisible into two formations: Cantuar Formation (type section, 130 feet thick) and overlying Pense Formation (type, 116 feet) which was said to be finer grained, more indurated, and less heterogeneous than the Cantuar. The Pense Formation was regarded by its designator as environmentally transitional between the Cantuar, considered to be non-marine (perhaps deltaic), and the grey marine shale of the overlying Joli Fou Formation, which constitutes the basal portion of the Ashville, and Colorado, Groups. Price’s two-fold division of the ‘Basal Cretaceous Sandy Group’ is typified by well sections in the Swift Current (Cantuar Formation), and Regina (Pense) areas and, according to him (1963, p. 5), ‘the combinations of facies typical of each division are broadly consistent over most of southern Saskatchewan and probably a considerably greater area’. The divisions were not, however, recognized in the correlative section of [Palaeontology, Vol. 14, Part 4, 1971, pp. 533-565, pis. 103-107.] C 8385 N n 534 PALAEONTOLOGY, VOLUME 14 the Bredenbury well, sampled during this study ; nor in the Swan River, western Mani- toba exposures, to which the name Swan River Group was originally applied (Wickenden 1945) for similar sandy /shaly sediments (reaching thicknesses of as much as 400 feet) developed beneath the marine Ashville Group. Price (1963, pp. 2, 35, 40) commented on likely diachroneity of the contact between the marine shale and underlying sandy strata. He considered that the demarcation between the two groups becomes younger north- eastward from southern Saskatchewan to the Swan River area in its marginal shield situation. Assignment of the Swan River Group and laterally contiguous strata to stages of the Lower Cretaceous has been hindered by their lack of marine fossils, or at least obvious ones. Their general Early Cretaceous age is, however, clearly established. The sediments rest with prominent unconformity on Jurassic or older strata, which constitute the eroded undulating surface on which the Cretaceous sedimentation regime was initiated through- out the western Canada sedimentary basin ; and are succeeded, seemingly conformably, by marine shaly strata dateable as (late) Middle Albian (Joli Fou Formation and correlatives). East-west lateral relationships of the Lower Cretaceous strata are well displayed on Rudkin’s (1964, fig. 11-7) cross-section of the southern Plains, stretching from east of Brandon, Manitoba, to the southern Alberta Foothills west of Lethbridge. G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 535 From this it may be seen that the Early Cretaceous arenaceous sequence, resting on a Jurassic erosion surface, thickens westward, passing laterally, in nomenclatural terms, from the Swan River Group into the Mannville Group (see Glaister 1959, Williams 1963) of Alberta and Saskatchewan, and thence into the thick Blairmore Group of the Foot- hills. The Blairmore covers a much greater stratigraphic interval than either the Mann- ville or the Swan River, since its upper strata intergrade with the shaly Colorado/ Ashville strata that succeed the Mannville and Swan River on the Plains (Mellon 1967). In contrast to the essentially non-marine Blairmore and Swan River sequences, the inter- vening Mannville includes, principally in Alberta, prominent marine intercalations, recording periodic advances of the boreal sea (depositing notably shales and glauconitic sandstones of the Clearwater Formation; see Williams 1963). The present geographic relations of the three groups are depicted by Price (1963, fig. 1). The marine strata carry faunal elements which permit the Clearwater to be dated as Middle Albian (see summary by Singh 1964, p. 13), which is also applicable to the other, overlying component of the Upper Mannville, the Grand Rapids Formation. The precise age of the Lower Mann- ville (McMurray Formation and correlatives) is less clear, but as its relationship with the Clearwater is conformable, it is presumably not greatly older. It is also apparent that being the initial continental-transgressive Cretaceous unit, its base would hardly be expected to be time-concordant over the western interior. Pocock (1962) ascribed a Neocomian (Berriasian-Late Barremian) age to the Lower Mannville strata (from 18 localities in Alberta/southern Saskatchewan) on the basis of his biostratigraphic appreciation of the microflora. In contradistinction, the age of the Calcareous Member at the top of the McMurray Formation has been adduced as Aptian from its ostracode assemblage (Loranger 1951) and Albian from foraminiferal evidence (Mellon and Wall 1963). The detailed work of Singh (1964) on the palynological flora of three cored borings in the Mannville Group of east-central Alberta evidenced that the base of the McMurray Formation of that area (i.e. Deville Member) is of Late Neocomian (Late Barremian) age. Singh further adduced palynologically that the middle portion of the McMurray (Ellerslie Member) is Aptian; and he supported Mellon and Wall’s contention that the upper McMurray (Calcareous Member) is referable to the Albian (Early to early Middle Albian). Vagvolgyi and Hills (1969) analysed the palyno- logical content of subsurface samples from the McMurray Formation and succeeding basal Clearwater Formation as penetrated by Socony Vacuum Hole No. 27, north- eastern Alberta (supplementary type section of the McMurray: Carrigy 1959). Vagvolgyi and Hills reported the occurrence of mainly long-ranging palynomorphs but, by refer- ence to Singh’s (1964) data, were able to infer an Early to early Middle Albian age for the McMurray Formation in the type well from the presence of a few spore species of known restricted vertical distribution. Collectively, therefore, the biostratigraphic evidence signifies an Early Cretaceous age for the Mannville strata, with the base perhaps no older than Late Neocomian and the top indubitably of Middle Albian age. Up to the present, virtually no palaeontological or palynological data were available on the Swan River or ‘Basal Cretaceous Sandy’ Groups. Consequently the dating of their bases, and, hence, of the inception of Cretaceous sedimentation in the south- eastern Saskatchewan-south-western Manitoba area has been depicted on correlation charts in a fairly vague manner. It has, however, been generally regarded as somewhat 536 PALAEONTOLOGY, VOLUME 14 later than in the better understood, westerly regions (e.g. Price 1963, table I). Rudkin’s (1964, fig. 11-1) chart suggests that the base in southern Saskatchewan is as old as Late Neocomian (as elsewhere to the west) and that it becomes progressively younger towards Manitoba, where the lowest Swan River Group is shown as young as Early Albian. Hence in places (extreme eastern part of the basin) the Swan River has been regarded as coeval in toto with Upper Mannville strata (cf. also Rudkin 1964, figs. 11-2, 11-3). The purpose of the present work is to seek evidence pertinent to the precise correlation and dating of the Swan River Group, the easterly correlatives of the Mannville. Because, firstly, the palynological sequence of the latter and the succeeding Lower Colorado Group have come to be known in a certain amount of detail, and secondly, the Swan River appears to be virtually bereft of marine megafossils, the best available facility for Swan River-Mannville comparisons is certainly palynological. It is emphasized that the current work, being restricted to one subsurface section and to samples from only the uppermost exposed Swan River section, constitutes essentially a reconnaissance study. PREVIOUS WORK ON LOWER CRETACEOUS PALYNOLOGY OF THE WESTERN CANADA PLAINS Pocock (1962), Singh (1964), and Vagvolgyi and Hills (1969), all investigated Mann- ville Group palynology of Alberta and also, in the case of Pocock, of southern Saskat- chewan. In a later paper, Pocock (1965) discussed and illustrated the morphology of EXPLANATION OF PLATE 103 All figures X 500 and from unretouched negatives. Fig. 1. Biretisporites potoniaei Delcourt and Sprumont 1955; median focus; GO24c/20, 106-7 13-4; Y.1337. Fig. 2. Stereisporites antkpiasporites (Wilson and Webster) Dettmann 1963; proximal focus; G024c/ 15, 106-2 7-3; Y.1338. Fig. 3. Kiiylisporites lunaris Cookson and Dettmann 1958; median focus; G046/3, 95-6 14-7; Y.1339. Figs. 4, 5. Undulatisporites pannuceus (Brenner) Singh 1971 ; proximal and distal foci; G045/3, 101-7 3-2; Y.1340. Figs. 6, 7. Pilosisporites spp. 6, P. veriis Delcourt and Sprumont 1955; proximal focus; G046/8, 102-0 9-7; Y.1341. 7, P. trichopapillosus (Thiergart) Delcourt and Sprumont 1955; median focus; G045/1, 109-0 19-2; Y.1342. Figs. 8, 9. Concavissimisporites spp. 8, C. minor (Pocock) Delcourt, Dettmann, and Hughes 1963; proximal focus; G045/3, 119-0 3-4; Y.1343. 9, C. variverrucatus (Couper) Brenner 1963; median focus; G064/2, 113-9 15-3; Y.1344. Fig. 10. Lophotriletes babsae (Brenner) Singh 1971; proximal focus; G065/97, 108-8 8-9; Y.1345. Fig. 11. Lycopodiacidites intraverrucatus Brenner 1963; median focus; G065/48, 114-8 16-2; Y.1346. Figs. 12, 13. Lycopodiumsporites spp. 12, L. austroclavatidites (Cookson) Potonie 1956; proximal focus; G024c/66, 113-3 8-3; Y.1347. 13, L. marginatus Singh 1964; distal focus ; G065/2, 1 1 1-0 19*1 ; Y.1348. Fig. 14. Reticidisporites elongatus Singh 1971; median focus; G059b/1, 113-5 12-8; Y.1349. Figs. 15, 16. Microreticidatisporites imiformis Singh 1964; proximal and distal foci; GO45/20, 105-4 2-8; Y.1350. Fig. 17. Kliikisporites pseiidoreticulatus Couper 1958; proximal focus; G043b/2, 81-7 16-9; Y.1351. Fig. 18. Tigrisporites scitrrandiis Norris 1967; median focus; G065/36, 103-5 17-3; Y.1352. Figs. 19-25. Cicatricosisporites spp. 19, C. hallei Delcourt and Sprumont 1955; median focus; G024c/ll, 86-6 17-0; Y.1353. 20, C. australiensis (Cookson) Potonie 1956; distal focus; G069/18, 110-2 8-2; Y.1354. 21, 22, C. spiralis Singh 1971; proximal and distal foci; G024c/74, 120-5 10-7; Y. 1355. 23, C. potomacensis Brenner 1963; proximal focus; 89-4 13-8; Y.1356. 24, 25, C. hughesi Dettmann 1963; proximal and distal foci; G045/2, 124-7 20-9; Y.1357. Palaeontology, Vol. 14 PLATE 103 PLAYFORD, Lower Cretaceous miospores G. PLAYFORD; PALYNOLOGY OF SWAN RIVER STRATA 537 certain elements of the Upper Mannville (late Middle Albian) palynological flora. The forms treated by Pocock were attributed to the gymnospermous category Chlamydo- spermidae (principally the Ephedraceae) and to the pterophyte family Schizaeaceae; the core samples concerned were from two wells in the Saskatoon area, Saskatchewan. Steeves and Wilkins (1967) also isolated and described dispersed spores from Lower Cretaceous sediments of the Saskatoon area. Although their paper contains no explicit information on the locality and horizon from which their core sample was collected, Dr. W. G. E. Caldwell (Department of Geological Sciences, University of Saskatchewan, Saskatoon) has informed the writer (pers. comm.) that it came from within the interval 1620-1880 feet (below K.B.) of Potash Corporation of American Saskatoon No. 2 well (location: Lsd. 6, Sec. 16, Tp. 36, R. 3, W. 3rd Mer.). Dr. G. D. Williams’s reading of the electric log of the latter indicated that the particular interval is probably all Upper Mannville equivalent. Steeves and Wilkins (p. 2330) intimated that a subsequent work of theirs will include description of the pollen elements of the microflora and a discussion of the age of the enclosing sediment. Norris (1967) studied in detail the spore-pollen sequence of the Lower Colorado Group (Joli Fou Formation, Viking Formation, Upper Shale Unit) as intersected in Fort Augustus No. 1 well, east-central Alberta. This represented an upward continua- tion of Singh’s (1964) study, which was concerned (in part) with Mannville Group palynology of the same well. In Fort Augustus No. 1, and elsewhere, a disconformable relationship has been postulated (Mellon and Wall 1963) between the Upper Mannville (Grand Rapids Formation) and the Lower Colorado (Joli Fou Formation). Norris’s studied interval comprised cores from the base of the Joli Fou to 50 feet below the Fish- scale marker bed separating the Upper Shale Unit of the Lower Colorado from the Upper Colorado; in age its range was cited as late Middle Albian to Late Albian, the latter in acceptance of the conventional Lower-Upper Cretaceous boundary at the Fish- scale marker bed. Singh (1971) has described the morphology and the stratigraphic distribution of spores, pollen, and non-calcareous microplankton from Middle-Upper Albian stratal sections in the lower Peace River area, north-western Alberta. Rock units studied by Singh are the Loon River Formation (upper part). Peace River Formation, and Shaftes- bury Formation (lower part) (see Table 2). MATERIAL AND METHODS Samples used in the current investigation consist of fine clastic sediments (light to dark-grey siltstones, shales, fine-grained silty sandstones) from two western Canadian localities, as cited below and representative of the Lower Cretaceous Swan River Group. The latter appears in both instances to have a conformable relationship with the over- lying Ashville Group. Localities are designated according to the Canadian Provincial Survey System (explained by Pocock 1962, fig. 1). 1. Swan River area, south-western Manitoba (type area, Swan River Group). {a) Samples 69-W-26 and 69-W-27 : transitional beds of uppermost Swan River Group-lowermost Ashville Group from north bank of Swan River (location : NE. J, Sec. 6, Tp. 37, R. 26, W. P. Mer.). {b) Samples 69-W-28 and 69-W-29 : upper Swan River Group, about 20 feet below transitional beds, north bank of Swan River (SE. J, Sec. 8-SW. J, Sec. 9, Tp. 37, R. 26, W. P. Mer.). 538 PALAEONTOLOGY, VOLUME 14 (c) Sample 69-W-32: upper Swan River Group, about 45 feet below transitional beds, north bank of Swan River (SW. i. Sec. 10, Tp. 37, R. 27, W. P. Mer.). The above were collected by G. D. Williams and the writer, 12 June 1969, and all proved to be productive palynologically. For geology see Wickenden (1945, Mafeking map 637a). 2. South-West Potash Bredenbury No. 11-36 well, south-eastern Saskatchewan. The Lower Cretaceous sequence intersected in this well (located at Lsd. 11, Sec. 36, Tp. 22, R. 1, W. 2nd Mer. — about 30 miles south-east of Yorkton, Sask.) is divisible according to G. D. Williams (pers. comm.) as follows: base of Fish-scale marker bed at 645 feet (below K.B.); top of Swan River- Group at 1275 feet; ? top of Jurassic sequence at 1495 feet. Cored material representative of the following intervals (in feet below K.B.) were processed and examined palynologically: 1273-1278*, 1278-1283*, 1283-1288*, 1288-1293*, 1293-1300*, 1300-1305, 1305-1310*, 1310-1315*, 1315-1320*, 1320-1325*, 1325-1327, 1329-1335, 1340-1345*, 1345-1349, 1349-1353, 1353-1358*, 1367-1372, 1372-1377*, 1377-1381*, 1386-1391*, 1391-1395*, 1395-1399, 1403-1408, 1408-1413*, 1413-1417, 1441-1445, 1462-1466, 1491-1495. Sample intervals asterisked in the foregoing list were palynologically productive; the others were not. Of the productive samples, all but the last one cited are clearly Lower Cretaceous. The lowest spore-containing sample (1408- 1413 feet) yielded a well-preserved Middle Jurassic palynological flora (determined as such from Pocock, 1970) and is not discussed further in this paper. From these data it is evident that the base of the Cretaceous, presumably disconformable, lies in the depth-interval 1395-1408 feet; and the thickness of Swan River Group in the welt is approximately 125 feet. The samples were subjected to conventional laboratory procedures for extraction and concentration of the contained acid-resistant microfossils. About 3-7 grams of sediment were treated in each case, depending on quantity available. Following dissolution of carbonates with dilute hydrochloric acid, several days’ immersion in cold 40 per cent, hydrofluoric acid sufficed for the removal of silicates; resultant insoluble fluoride precipitates were then removed through several treatments with warm dilute hydrochloric acid. On average, only about 5 minutes’ treatment with Schulze solution was appropriate for maceration of organic material; this was followed by rapid washing (by centrifugation) with very weak (ca. one per cent.) ammonium hydroxide and then, repeatedly, with distilled water. Where present, obtrusive amounts of structureless organic debris were removed by a single washing with Darvan No. 4. The residues were mounted in glycerine jelly (unstained, or lightly stained with Safranin ‘0’) under No. 0 coverslips as both strew-slides and single-spore mounts. All slides were thoroughly sealed with gold-size varnish. Mounted residues of the productive samples contain spores and pollen in varying concentrations and in fair to excellent states of preservation. A cross-reference between samples and preparation numbers (prefixed ‘G’) is to be found on Table 1. TAXONOMIC LIST OF FORMS IDENTIFIED All spore and pollen specimens illustrated in the current work are housed in the micropalaeontological type collection of the Department of Geology and Mineralogy, University of Queensland, Brisbane. Each specimen is designated, as in the plate explanations and in ensuing text, according to the following sequence: preparation/slide number, east-west and north-south mechanical-stage readings (of Zeiss Photomicro- scope II, no. Mx3237, of the above-cited Department) and registered specimen number (prefixed ‘Y’). The following list consolidates the spore and pollen species identified in the subject samples of the Swan River Group. For convenience of reference, the species are categorized according to the morphographic scheme of R. Potonie (1956, etc.); in the case of the Anteturma Sporites certain of the modifications and innovations proposed by Dettmann (1963) and Smith and Butterworth (1967) are adopted. Most of the specific taxa listed are illustrated herein, as specified by the square-bracketed plate/ figure numbers. Comprehensive systematic-descriptive coverage of all species has not G. PLAYFORD; PALYNOLOGY OF SWAN RIVER STRATA 539 been undertaken because the majority have been adequately described, figured, and satisfactorily treated taxonomically in existing publications. Those which are accorded systematic appraisal later in this paper are asterisked below; they are mainly new species, new combinations, or species for which emended or restated circumscriptions or brief comments are deemed appropriate. Anteturma sporites H. Potonie 1893 Turma triletes Reinsch emend. Dettmann 1963 Suprasubturma acavatitriletes Dettmann 1963 Subturma azonotriletes Luber emend. Dettmann 1963 Infraturma laevigati Bennie and Kidston emend. R. Potonie 1956 Cyathidites australis Couper 1953 Cyatliidites minor Couper 1953 Stereisporites antiquasporites ( Wilson and Webster) Dettmann 1963 [PI. 103, fig. 2] Biretisporites potoniaei Delcourt and Sprumont 1955 [PI. 103, fig. 1] Undidatisporites panmiceus (Brenner) Singh 1971 [PI. 103, figs. 4, 5]* Infraturma apiculati Bennie and Kidston emend. R. Potonie 1956 Concavissimisporites minor (Pocock) Delcourt, Dettmann, and Hughes 1963 [PI. 103, fig. 8] Concavissimisporites piinctatus (Delcourt and Sprumont) Brenner 1963 Concavissimisporites variverrucatus (Couper) Brenner 1963 [PI. 103, fig. 9] Osnnmdaeidites wellmanii Couper 1953 Bacidatisporites cornawnensis (Cookson) R. Potonie 1956 Lophotriletes babsae (Brenner) Singh 1971 [PI. 103, fig. 10] Pilosisporites trichopapiUosus (Thiergart) Delcourt and Sprumont 1955 [PI. 103, fig. 7] Pilosisporites verus Delcourt and Sprumont 1955 [PI. 103, fig. 6] Kuylisporites lunaris Cookson and Dettmann 1958 [PI. 103, fig. 3] Infraturma murornati R. Potonie and Kremp 1954 Lycopodiumsporites austroclavatidites (Cookson) R. Potonie 1956 [PI. 103, fig. 12] Lycopodiumsporites marginatus Singh 1964 [PI. 103, fig. 13] Lycopodiacidites intraverrucatus Brenner 1963 [PI. 103, fig. 11] Reticulisporites elongatus Singh 1971 [PI. 103, fig. 14] Microreticidatisporites iiniformis Singh 1964 [PI. 103, figs. 15, 16] Klukisporites pseudoreticidatus Couper 1958 [PI. 103, fig. 17] Tigrisporites scurrandus 'Norris 1967 [PI. 103, fig. 18] Cicatricosisporites australiensis (Cookson) R. Potonie 1956 [PI. 103, fig. 20] Cicatricosisporites ha/lei Delcourt and Sprumont 1955 [PI. 103, fig. 19] Cicatricosisporites hughesi Dettmann 1963 [PI. 103, figs. 24, 25] Cicatricosisporites patapscoensis Brenner 1963 Cicatricosisporites potomacensis Brenner 1963 [PI. 103, fig. 23] Cicatricosisporites spiralis Singh 1971 [PI. 103, figs. 21, 22] 540 PALAEONTOLOGY, VOLUME 14 Cicatricosisporites sp. B of Singh 1964 Costatoperforosporites foveolatus Deak 1962 [PI. 104, fig. 1] Bahneisporites sp. cf. B. holodictyus Cookson and Dettmann 1958 Arcellites disciformis Miner emend. Ellis and Tschudy 1964 Subturma zonotriletes Waltz 1935 Infraturma auriculati Schopf emend. Dettmann 1 963 Appendicisporites bifurcaliis Singh 1964 [PI. 104, fig. 10] Appendicisporites bUateralis Singh 1971 [PI. 104, fig. 4] Appendicisporites crimensis (Bolkovitina) Pocock 1965 [PI. 104, figs. 2, 3] Appendicisporites jansonii Pocock 1962 [PI. 104, fig. 9] Appendicisporites matesovae (Bolkhovitina) Norris 1967 [PI. 104, fig. 11] Appendicisporites potomacensis Brenner 1963 [PI. 104, figs. 5, 6] Appendicisporites problematicus (Burger) Singh 1971 [PI. 104, figs. 7, 8] Trilobosporites apiverrucatus Couper 1958 [PI. 104, fig. 14] Trilobosporites hannonicus (Delcourt and Sprumont) R. Potonie 1956 [PI. 104, fig. 26] Trilobosporites hwnilis Delcourt and Sprumont 1959 [PI. 104, figs. 15, 16] Trilobosporites marylandensis Brenner 1963 [PL 104, fig. 13] Trilobosporites purveridentus (Verbitskaya) Dettmann 1963 [PL 104, figs. 24, 25] Ischyosporites sp. cf. I. crateris Balme 1957 Infraturma tricrassati Dettmann 1963 Gleicheniidites senonicus Ross 1949 [PL 104, fig. 12] Clavifera rudis Bolkhovitina 1968 [PL 104, figs. 18, 19]* Asbeckiasporites wirthi von der Brelie 1964 [PL 105, figs. 11, 12]* Sestrosporites pseudoalveolatus (Couper) Dettmann 1963 [PL 104, fig. 27] Coronatispora va/densis (Couper) Dettmann 1963 [PL 104, figs. 20, 21] Caniarozonosporites ainbigens (Fradkina) comb. nov. [PL 104, figs. 22, 23]* Infraturma cingulati R. Potonie and Klaus emend. Dettmann 1963 Foraniinisporis asymmetricus (Cookson and Dettmann) Dettmann 1963 [PL 104, fig. 17] Foraminisporis dailyi (Cookson and Dettmann) Dettmann 1963 [PL 105, fig. 1] Foraniinisporis wonthaggiensis {Cookson and Dettmann) Dettmann 1963 [PL 105, fig. 2] Polvcingidatisporites rediincus (Bolkhovitina) Playford and Dettmann 1965 [PL 105, figs. 3, 4] Taurocusporites segmentatus Stover 1962 [PI. 105, fig. 5] Cingutriletes claviis (Balme) Dettmann 1963 Suprasubturma laminatitriletes Smith and Butterworth 1967 Subturma azonolaminatitriletes Smith and Butterworth 1967 Spore Type A [PL 105, figs. 9, 10]* G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 541 Subturma zonolaminatitriletes Smith and Butterworth 1967 Infraturma cingulicavati Smith and Butterworth 1967 Kraeuselisporites hastilobatus sp. nov. [PI. 105, figs. 23-25]* Limdbladispora reticingula (Brenner) comb. nov. [PI. 105, fig. 8]* Densoisporites circwnundulalus (Brenner) comb. nov. [PI. 105, figs. 6, 7]* Densoisporites microrngulalus Brenner 1963 [PI. 105, fig. 17] Suprasubturma perinotrilites Erdtman 1947 Crybelosporites breunerii sp. nov. [PI. 105, figs. 18-20]* Perotrilites pammceus Brenner 1963 Turma monoletes Ibrahim 1933 Suprasubturma acavatomonoletes Dettmann 1963 Subturma azonomonoletes Luber 1935 Infraturma laevigatomonoleti Dybova and Jachowicz 1957 Laevigatosporites ovatus Wilson and Webster 1946 [PI. 105, fig. 22] Cicatricososporites auritus Singh 1971 [PI. 105, fig. 21] Turma hilates Dettmann 1963 Coptospora williamsii sp. nov. [PI. 106, figs. 8-13]* Aeqidtriradites spinulosiis (Cookson and Dettmann) Cookson and Dettmann 1961 [PI. 106, fig. 7] Couperisporites complexiis (Couper) Pocock 1962 [PL 107, figs. 22, 23] Cooksonites variabilis Pocock 1962 Triporoletes involucratus (Chlonova) comb. nov. [PI. 105, figs. 13-15]* Triporoletes laevigatus (Pocock) comb. nov. [PI. 105, fig. 16]* Triporoletes radiatus (Dettmann) comb. nov. [PI. 106, figs. 3, 4]* Triporoletes reticidatus (Pocock) comb. nov. [PI. 106, fig. 5]* Triporoletes simplex (Cookson and Dettmann) comb. nov. [PI. 106, fig. 2]* Triporoletes singularis Mtchedlishvili in Mtchedlishvili and Samoilovich 1960 [PI. 106, fig. 1]* Anteturma pollenites R. Potonie 1931 Turma saccites Erdtman 1947 Subturma monosaccites Chitaley emend. R. Potonie and Kremp 1954 Infraturma saccizonati Bhardwaj 1957 Tsugaepollenites dampieri (Balme) Dettmann 1963 Subturma disaccites Cookson 1947 Vitreisporites pallidus (Reissinger) Nilsson 1958 [PI. 107, fig. 2] Alisporites bilaterolis Rouse 1959 [PI. 107, fig. 1] 542 PALAEONTOLOGY, VOLUME 14 Alisporites grandis (Cookson) Dettmann 1963 Podocarpidiles canadensis Pocock 1962 Podocarpidites sp. cf. P. ellipticus Cookson 1947 Podocarpidites nndtesinnis (Bolkhovitina) Pocock 1962 [PI. 107, fig. 21] Podocarpidites radiatus Brenner 1963 Parvisaccites radiatus Couper 1958 [PI. 107, fig. 19] Phyllocladidites inchoatus (Pierce) Norris 1967 Rugiibivesicu/ites rediictus Pierce 1961 [PI. 107, fig. 20] Turma aletes Ibrahim 1933 Subturma azonaletes Luber emend. R. Potonie and Kremp 1954 InaperturopoUenites Umbatus Balme 1957 [PI. 107, fig. 8] Araiicariacites australis Cookson 1947 Reticidatasporites dupliexinous Brenner 1963 Subturma zonaletes Luber 1935 Perinopollenites elatoides Couper 1958 Turma plicates Naumova emend. R. Potonie 1958 Subturma praecolpates R. Potonie and Kremp 1954 Euconuniidites minor Groot and Penny 1960 [PI. 107, fig. 3] Eucommiidites troedssonii Erdtman 1948 [PI. 107, fig. 4] Subturma polyplicates Erdtman 1952 Equisetosporites sp. cf. E. concinnus Singh 1964 Equisetosporites jansonii Pocock 1965 [PI. 107, fig. 6] Equisetosporites nndticostatus (Brenner) Norris 1967 [PI. 107, fig. 7] Subturma monocolpates Iversen and Troels-Smith 1950 Clavatipollenites hughesii Couper emend. Kemp 1968 [PL 107, fig. 13] ClavatipoUenites rotundus Kemp 1968 [? = Liliacidites dividuus (Pierce) Brenner 1963] [PI. 107, fig. 14] Liliacidites peroreticulatus (Brenner) Singh 1971 Subturma triptyches Naumova 1939 Retitricolpites georgensis Brenner 1963 [PL 107, figs. 11, 12] Retitricolpites prosimilis Norris 1967 [PL 107, fig. 18] Retitricolpites vulgaris Pierce 1961 [PL 107, fig. 17] Tricolpites sagax Norris 1967 [PL 107, fig. 16] Eraxinoipollenites venustus Singh 1971 [PL 107, figs. 9, 10] Striatopollis paraneus (Norris) Singh 1971 G. PLAYFORD; PALYNOLOGY OF SWAN RIVER STRATA 543 Turma poroses Naumova emend. R. Potonie 1960 Subturma monoporines Naumova 1939 Exesipollenites tumulus Balme 1957 [PI. 107, fig. 15] Circuliua parva Brenner 1963 [PI. 107, fig. 5] ClassopoUis classoides Pflug emend. Pocock and Jansonius 1961 SPORAE INCERTAE SEDIS Schizosporis reticulatus Cookson and Dettmann 1959 [PI. 106, fig. 6] The distribution of these specific taxa in samples utilized in this study is set out in Table 1. The stratigraphic significance of the palynological assemblages is discussed in a later section of this paper. SYSTEMATIC SECTION Genus undulatisporites Pflug in Thomson and Pflug 1953 Type (by original designation). Undulatisporites niicrocutis Pflug in Thomson and Pflug 1953 Undulatisporites pannuceus (Brenner) Singh 1971 Plate 103, figs. 4, 5 1963 Alsophilidites pannuceus Brenner, p. 56; pi. 12, figs. 5, 6. 1971 Undulatisporites pannuceus (Brenner) Singh, pi. 20, figs. 9, 10. Description. Spores radial, trilete. Amb subtriangular with somewhat acute apices and slightly concave to slightly convex sides. Laesurae distinct, sinuous, lipped, extending to equatorial margin; sometimes with terminal bifurcation; lips elevated, maximum width (usually at pole) up to 6 /xm, tapering equatorially. Proximal exine laevigate; distal exine undulant with relatively broad smooth sinuous elevations, in places with subradial arrangement, separated by very fine, sinuous, irregularly branching channels (vermiculi). Exine thickness 1-5-2 /xm. Dimensions (12 specimens). Equatorial diameter 24 (36) 43 /xm. Remarks. The Swan River specimens are on average somewhat larger than those reported previously, but in other respects accord well with the original diagnosis. Although Undulatisporites Pflug is not as yet clearly differentiated from other simple, acavate, trilete genera, it provides a more suitable repository for the species than does Alsophilidites Cookson ex Potonie 1956. U. sp. cf. U. undulapolus Brenner 1963 (Norris 1967, p. 87; pi. 10, figs. 6, 7), from the Late Albian of central Alberta, differs from U. pannuceus (Brenner) Singh in having rounded amb apices, narrower lips, and irregular distal channels which are probably corrosion effects. Previous records. U. pannuceus has been reported previously from all formational units of the Potomac Group (Barremian-Albian) of Maryland as studied by Brenner (1963); and from the uppermost Peace River and Lower Shaftesbury Formations (late Middle- Late Albian) of north-western Alberta (Singh 1971). 544 PALAEONTOLOGY, VOLUME 14 Genus clavifera Bolkhovitina 1966 For synonymy see Dettmann and Playford (1968, p. 76). Type species (by original designation). Clavifera triplex (Bolkhovitina) Bolkhovitina 1966. Discussion. Clavifera is one of four form-generic entities that have been circumscribed and discussed by Bolkhovitina (1966, 1968) in her comprehensive discourse on glei- cheniaceous-Iike trilete spores. It differs from the other three {Gleicheniidites Ross emend. Bolkhovitina 1968, Ornamentifera Bolkhovitina 1966, a.nd Plicifera Bolkhovitina 1966) in possessing bulbous or clavate projections at equatorial radii (as well as inter- radial equatorial thickenings typical of the Tricrassati) and exine that is essentially devoid of fine sculpture. Clavifera rudis Bolkhovitina 1968 Plate 104, figs. 18, 19 1968 Clavifera rudis Bolkhovitina, p. 48; pi. 13, figs. 9-18; pi. 14, figs. 1-15; pi. 15, figs. 1-12. Description. Microspores radial, trilete; distal surface arched. Amb subtriangular with almost straight sides and relatively short appendices. Laesurae long, extending near to equatorial margin, with narrow, slightly thickened lips. Exine 1-2 j^m thick; thicker at equator. Equatorial thickenings consisting of three, smooth, interradial crassitudes, EXPLANATION OF PLATE 104 All figures x 500 and from unretouched negatives. Fig. 1. Costatoperforosporites foveolatus Deak 1962; distal focus; G065/2, 91-6 16-5; Y.1358. Figs. 2-1 1 . Appendicisporites spp. 2, 3, A. crimensis (Bolkhovitina) Pocock 1965 ; proximal and distal foci; G061/1, 118-8 2-9; Y.1359. 4, A. bilateralis Singh 1971; median focus; G045/16, 109-9 3-6; Y.1360. 5,6, A. potomacensis 'Qrenmv \963. 5, Proximal focus; GO65/10, 104-8 13-2, Y. 1361. 6, Distal focus; GO24b/50, 101-0 19-3; Y.1362. 7, 8, A. problematicus (Burger) Singh 1971; proximal and distal foci; G065/112, 119-1 9-7; Y.1363. 9, A. jansonii Pocock 1962; proximal focus; G061/ 11, 108-8 13-1; Y.1364. \0, A. bifurcatus Singh 1964; proximal focus; G064/6, 93-5 11-6; Y.1365. ll, A. matesovae (Bolkhovitina) Norris 1967; median focus; G021b/23, 124-9 20-0; Y.1366. Fig. 12. Gleicheniidites senonicus Ross 1949; median focus; G031b/3, 107-1 7-9; Y.1367. Figs. 13-16, 24-26. Trilobosporites spp. 13, T. marylandensis Brenner 1963; proximal focus; G065/2, 100-0 17-8; Y.1368. 14, T. apiverrucatiis Couper 1958; proximal focus; G024c/4, 112-8 11-7; Y.1369. 15, 16, T. humilis Delcourt and Sprumont 1959; proximal and distal foci; G064/18, 109-5 15-7; Y.1370. 24, 25, T. purveriilentus (Verbitskaya) Dettmann 1963. 24, Median focus; G065/39, 109-5 20-5; Y.1371. 25, Proximal focus; G065/3, 100-5 7-7; Y.1372. 26, T. hannoniciis (Delcourt and Sprumont) Potonie 1956; median focus; G063/6, 104-8 8-6; Y.1373. Fig. 17. Foraminisporis asvmmetriciis (Cookson and Dettmann) Dettmann 1963 ; distal focus; G024b/ 43, 129-1 14-6; Y.1374.' Figs. 18, 19. Clavifera rudis Bolkhovitina 1968; proximal and distal foci; G064/13, 98-6 4-2; Y.1375. Figs. 20, 21. Coronatispora valdensis (Couper) Dettmann 1963; proximal and distal foci; GO64/20, 104-8 13-0; Y.1376. Figs. 22, 23. Camarozonosporites ambigens (Fradkina) comb, nov.; proximal and distal foci; G021b/ 31, 111-3 8-0; Y.1377. Fig. 27. Sestrosporites pseudoalveolatus (Couper) Dettmann 1963; proximal focus; G021b/37, 108-3 17-1; Y.1378. Palaeontology, Vol. 14 PLATE 104 PLAYFORD, Lower Cretaceous miospores r "!*■ . * ■ M' 4®* ■:v •V-?-' G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 545 with maximum width (6-12 yu.m) at centres of interradii, and gently undulating margins. Proximal surface laevigate. Distal surface with conspicuous rugulate and/or verrucate elevations typically confined to triangular region, the sides of which parallel amb and are 6 /xm or more from latter. Rugulae/verrucae smooth, width 3-9 /xm, usually disposed ^parallel to radii; projecting as rounded appendices (4-10 |Ltm long), one at each apex of amb. Dimensions (8 specimens). Overall equatorial diameter 55-78 /xm. Previous records. Bolkhovitina’s (1968) records are from Russian Aptian to Cenomanian strata. Genus asbeckiasporites von der Brelie 1964 Type species (by original designation). Asbeckiasporites wirthi von der Brelie 1964. Discussion. In a German Lower Cretaceous palynological study, von der Brelie (1964, p. 141) founded the genus Asbeckiasporites for the reception of certain trilete, ‘cingulate’ miospores in which the equatorial thickening is prominent interradially and reduced by invagination or truncation at the three apices of the subtriangular amb. Such difterential development of the thickening served to separate von der Brelie’s genus from Murospora Somers 1952, which has been applied to otherwise similar Upper Jurassic-Lower Cretaceous miospores, viz. M. fiorida (Balme) Pocock 1961. Asbeckiasporites wirthi von der Brelie 1964 Plate 105, figs. 11, 12 1964 Asbeckiasporites wirthi von der Brelie, pp. 141-142; pi. 8, figs. 7, 8; pi. 9, figs. 1-6. 1971 Murospora truncata Singh, pi. 18, figs. 1, 2. Description. Spores radial, trilete. Amb subtriangular with concave to almost straight sides; equatorial contour distinctly to slightly undulating. Laesurae usually distinct, simple, straight; extending to spore-cavity margin. Equatorial thickening smooth and homogeneous, or showing fine radial striations; strongly developed interradially (maxi- mum width 8-12 /xm), restricted at each of the 3 radial positions where it is distinctly embayed or truncated. Outline of spore cavity (polar view) subtriangular with concave to almost straight sides and pointed to rounded apices. Unthickened exine (i.e. delimiting spore cavity proximally and distally) laevigate, scabrate, chagrenate, or punctate. Dimensions (20 specimens). Overall equatorial diameter 40 (48) 60 /xm. Remarks. Von der Brelie (1964, p. 142) noted that the distal exines of some of his specimens possess sparse verrucate projections, circular to elongate-oval in surface view and irregular in distribution. He regarded these as ‘secondary features without taxonomic significance’. A few of the Canadian specimens encountered during the current work display similar verrucate modification. Previous records. The species was originally described from Lower Cretaceous (? Aptian or Albian) sediments of north-western Germany (von der Brelie 1964). The subsequent 546 PALAEONTOLOGY, VOLUME 14 Canadian record by Singh (1971) is from north-western Alberta strata of Late Albian age (Paddy Member, Peace River Formation-Lower Shaftesbury Formation). Genus camarozonosporites Pant ex R. Potonie emend. Klaus 1960 Type species (by subsequent designation of Potonie 1956, p. 65). Camarozonosporites cretaceus (Weyland and Krieger) R. Potonie 1956. Camarozonosporites ambigens (Fradkina) comb. nov. Plate 104, figs. 22, 23 1963 Lycopodiacidites cerniidites (non Ross) Brenner, pp. 43-44; pi. 5, fig. 2. 1964 Lycopodiacidites cerniidites (non Ross) Brenner 1963; von der Brelie, p. 138; pi. 6, 'figs. 12-14. 1964 Camptotriletes ambigens Fradkina in Fradkina and Kiseleva, p. 70; pi. 1, figs. 3-5. 1967 Camarozonosporites insignis Norris, pp. 96-97; pi. 13, figs. 12-16. Description. Spores radial, trilete, with convexly subtriangular to subcircular amb. Laesurae distinct, simple, straight, extending two-thirds to three-quarters of the distance to equator. Exine tricrassate — ^at amb apices, exine is 1-2 ftm thick; at equatorial inter- radii, 3-6 pm. Sculpture rugulate (usually much more conspicuously so on distal surface). Proximal rugulae usually sparse and with subradial orientation. Distal rugulae well defined, sinuous, smooth, typically with irregularly pleated appearance; width 1-5-4 pm, height 1 -5-2-5 pm; separated by sharply defined channels about EXPLANATION OF PLATE 105 All figures X 500 unless otherwise specified; from unretouched negatives. Figs. 1, 2. Foraminisporis spp. 1, F. dailyi (Cookson and Dettmann) Dettmann 1963; median focus; G031b/2, 110-5 16-6; Y.1379. 2, F. wonthaggiensis (Cookson and Dettmann) Dettmann 1963; proximal focus; G031a/1, 105-4 14-6; Y.1380. Figs. 3, 4. Polycingnlatisporites rediinciis (Bolkhovitina) Playford and Dettmann 1965; proximal and distal foci; G065/2, 96-6 16-7; Y. 1381. Fig. 5. Taiirociisporites segmentatiis Stover 1962; median focus; G065/88, 107-8 17-7; Y.1382. Figs. 6, 7, 17. Densoisporites spp. 6, 7, D. circiimnndiilatns (Brenner) comb. nov. 6 ( X 750), Proximal focus; GO65/70, 109-8 11-2, Y.1383. 7, Medianfocus; G024c/14, 1 10-4 4-3 ; Y.1384. 17, D.micro- rugidatiis Brenner 1963; proximal focus; G065/8, 104-2 18-0; Y.1385. Fig. 8. Lundbladispora reticingida (Brenner) comb, nov.; proximal focus; G065/1, 89-6 5-4; Y.1386. Figs. 9, 10. Spore type A; proximal and distal foci; G065/7, 110-1 10-0; Y.1387. Figs. 11, 12. Asbeckiasporites wurthi von der Brelie 1964; proximal foci. 11, G021b/35, 109-4 17-4; Y.1388. 12, G062/13, 110-1 7-0; Y.1389. Figs. 13-16. Triporoletes spp. 13-15, T. involucratus (Chlonova) comb. nov. (X750). 13, 14, Low and high foci; G031b/3, 111-0 2-9; Y.1390. 15, Median focus; G065/79, 105-8 10-9; Y.1391. 16, T. laevigatas (Pocock) comb, nov.; median focus; G024c/16, 107-0 6-6; Y.1392. Figs. 18-20. Crybelosporites brennerii sp. nov.; lateral views. 18, Holotype; G024b/15, 106-2 4-1; Y.1393. 19, GO46/10, 106-4 16-0; Y.1394. 20, G029c/1, 101-3 14-0; Y.1395. Fig. 21. Cicatricososporites auritus Singh 1971 ; lateral aspect; G065/25, 107-5 11-7; Y.1396. Fig. 22. Laevigatosporites ovatus Wilson and Webster 1946; median focus; G024c/65, 117-0 10-7; Y.1397. Figs. 23-25. Kraeuselisporites hastilobatus sp. nov. 23, 24, Holotype; median and distal foci; G026/ 15, 108-2 6-7; Y.1398. 25 (XlOOO), showing distal sculptural projections; G065/100, 95-2 8-9; Y.1399. Palaeontology, Vol. 14 PLATE 105 PLAYFORD, Lower Cretaceous miospores G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 547 0-5-1 -5 |Lim wide. Inner exine layer (nexine) uniformly thin. Equatorial contour normally crenate. Dimensions (30 specimens). Equatorial diameter 28 (43) 54/xm. This accords closely with the size ranges quoted by previous authors [Fradkina 1967, p. 99 — 27 (43) 60 /xm; Norris 1967, p. 97 — 30 to 55 /xm]. Previous records. According to Fradkina (1964, 1967), Camarozonosporites ambigens (Fradkina) comb. nov. occurs in Albian-Senonian strata of Vilyui and Lena River Basins, Yakutsk, U.S.S.R. It occurs also in imprecisely dated Fower Cretaceous sedi- ment of north-west Germany (von der Brelie 1964). From North America, Brenner (1963) noted its rare occurrence in the Albian Patapsco Formation, Maryland; Norris (1967) described it from the Late Albian, central Alberta; and Singh (1971) plots the initial entry of the species (as C. msigiiis) in the lower part of the Cadotte Member (late Middle Albian) of the Peace River Formation and shows its consistent presence higher in the sequence (Shaftesbury Formation). Another (probably late Middle) Albian record is from Oklahoma (Hedlund and Norris 1968). Spore Type A Plate 105, figs. 9, 10 Description. Spores radial, trilete, cavate, with subcircular to convexly subtriangular amb having notched periphery. Laesurae simple or narrowly lipped, perceptible to distinct, extending about one-half to two-thirds of distance to equator. Exoexine relatively thin proximally on (usually) well-defined, laevigate-scabrate, subtriangular contact area {ca. 30-40 /xm in diameter). Remainder of exoexine sculptured with pits (foveolae) and/or fine sinuous channels (vermiculi), especially in distal and equatorial regions, less commonly on equatorial region of proximal surface; unsculptured exo- exine (outside contact areas) 2-5-3-5 /xm thick. Foveolae subcircular to polygonal or irregularly elongate in surface view, diameter ranges from less than 1-10 /xm, spacing up to 8 /xm apart. Vermiculi, where present, have similarly variable spacing, are usually less than 0-5 /xm broad, and average about 1-2 /xm in length. Sculptural elements incised 2 /xm or less in exoexine. Inner layer (intexine) featureless, about 0-5 /xm thick, more or less distinctly separated from exoexine, forming central body with subcircular-roundly subtriangular outline in polar view. Dimensions (5 specimens). Overall equatorial diameter 48 (54) 65 /xm; intexinal body diameter 32 (40) 52 /xm. Remarks. So far as the author is aware no closely comparable form has appeared in the literature. Thus although a new taxon appears to be represented, formal systematic treatment must be withheld, pending the discovery of many more specimens than the five currently available. Genus kraeuselisporites Leschik emend. Jansonius 1962 For synonymy and discussion see Playford and Helby (1968, p. 112). Type species (by original designation). Kraeuselisporites dentatus Leschik 1955. 548 PALAEONTOLOGY, VOLUME 14 Kmeuselisporites hastilobatus sp. nov. Plate 105, figs. 23-25 Diagnosis. Spores radial, trilete, zonate. Amb convexly subtriangular; margin undulate to somewhat crenulate. Laesurae frequently indistinct but accompanied by well-defined, narrow (individually 1 ^<.m or less in width), elevated, sinuous lips extending to equatorial margin ; in most examples a distinct subtriangular gape in the exoexine (and often also in the intexine), defined by dehiscent lips/laesurae, occurs over the spore cavity to which it approximates in extent. Exoexine zonate, with well-developed distal sculpture usually confined to area over spore cavity (i.e. absent or rare on zona). Sculptural projections consist of large, discrete spinae, tapering from circular bases (1-5-4 /xm broad) to sharp, rounded, or truncate tips; length of spinae 4-12 /xm, spacing about 1-11 ;um. Proximal surface of exoexine featureless apart from fine minor compression-folding. Zona of approximately uniform width on a given specimen. Intexine thin, occasionally recogniz- able as a fairly distinct inner body partly separated from outer layer. Dimensions (14 specimens). Overall equatorial diameter 66 (82) 103jtxm; width of zona 5 (8) 12 /mi. Holotype. Preparation G026/15, 108-2 6-7, Y.1398. PI. 105, figs. 23, 24. Distal aspect. Amb roundly subtriangular. Well-defined, narrow laesurate lips defining proximal, convexly subtriangular gape, 34|um in diameter, and thereafter extending radially to equatorial margin. Distal exoexine, excluding zona, bearing discrete, simple spinae, mostly with pointed apices. Dimensions of spinae — length 8-10 jum, basal diameter 3-5 /xm; spinae spaced up to 8/xm apart. Exoexine otherwise laevigate apart from radial compression folds on proximal face. Intexine very thin, situated close to inner margin of exoexine. Overall diameter 85 /xm. Zona averages 11 /xm; margin crenulate. Type locality. Saskatchewan, S.W.P. Bredenbury 11-36, core, 1353-1358 feet; Swan River Group. Comparison. The Austrian Carnian form, Kraeuselisporites cooksonae (Klaus 1960, p. 141 ; pi. 31, figs. 29, 31) Dettmann 1963, differs from A'. sp. nov. inhavinga conate (rather than spinose) sculpture, shorter laesurae, and an almost entire periphery. Genus lundbladispora Balme emend. Playford 1965 Type species (by original designation). Lundbladispora willmotti Balme 1963. Lundbladispora reticingiila (Brenner) comb. nov. Plate 105, fig. 8 1963 Cingulatisporites reticinguliis Brenner, p. 42; pi. 4, figs. 2, 3. Description. Spores radial, trilete, cavate, cingulate. Amb convexly subtriangular. Laesurae accompanied by membraneous lips, 1-3 /xm in overall width, extending to inner margin of cingulum. Exoexine, with finely spongeous appearance (superficially scabrate), bearing discrete simple spinae on distal surface and at equator; spinae 1- 5-5-5 /xm in basal diameter (bases circular in surface view), 3-8-5 /xm long, spaced 2- 10 /xm apart, sides regularly tapering, apices sharp. Cingulum 3-6-5 /xm wide (some- what variable on given specimen). Intexinal body usually well defined, wall about 1 /xm thick, homogeneous; in polar view body shape is subtriangular, diameter 20-28 /xm; often only partly detached from exoexine. G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 549 Dimensions (8 specimens). Overall equatorial diameter 35 (40) 43 /rm. Remarks. The few Swan River specimens described above are in close morphological accord with those described originally (Brenner 1963, p. 42). Brenner’s generic allocation of the species cannot be upheld because Cingulatisporites Thomson {in Thomson and Pflug 1953), though narrowly cingulate, is acavate and where sculptured is only finely so (see Pocock’s (1961) clarified circumscription of the latter genus). The species does, however, find appropriate inclusion in Lundbladispora Balme as emended by Playford (1965). The spores figured by Singh (1971, pi. 17, figs. 10, 1 1) almost certainly belong to the species under consideration. Previous records. Initially recorded from the Lower Cretaceous (Barremian-Albian) Potomac Group of Maryland (Brenner 1963), this species has also been reported from the Albian of Oklahoma (Hedlund and Norris 1968). Singh’s possible record (1971) is from the Cadotte Member (late Middle Albian), Peace River Formation, north-western Alberta. Genus densoisporites Weyland and Krieger emend. Dettmann 1963 Type species (by original designation). Densoisporites velatus Weyland and Krieger 1953 emend. Krasnova 1961 (= D. perinatus Couper 1958). Densoisporites circunnmdidatus (Brenner) comb. nov. Plate 105, figs. 6, 7 1963 Psilatriletes circiimundiilatiis Brenner, pp. 67-68; pi. 20, figs. 4, 5. Description. Spores radial, trilete. Amb subtriangular with convex to almost straight sides and rounded apices. Laesurae distinct, usually sinuous, length at least three quarters of spore radius; bordered by elevated thickened lips, 1-3-5 pm in overall width, often displaying terminal bifurcation. Exine two-layered, cavate, consisting of a thin, well-defined, featureless intexine (0-5-1 pm thick) showing slight and partial contraction from enclosing exoexine (separation of two layers rarely more than 4 ;um). Exoexine laevigate or with fine irregular pitting (? result of corrosion); thickness similar to that of intexine except at equator where a continuous thickening (i.e. cingulum, 2-5-5 pm wide) is developed. Exoexine frequently but not invariably contorted with radial folds; where present these give an undulating appearance to equatorial margin. Arcuate compression folds developed marginally on intexinal body of some specimens. Dimensions (10 specimens). Overall equatorial diameter 34 (40) 46 /xm. Remarks. The original generic attribution of this species to Psilatriletes van der Hammen ex van der Hammen 1956 (or to Poroplanites Pflug 1953, regarded by Potonie (1956, p. 18) as a senior synonym of Psilatriletes) cannot be maintained on account of the two- layered, cavate nature of its exine, which in being virtually unsculptured conforms with Densoisporites. Previous records. Densoisporites circumundulatus (Brenner) comb. nov. is known sparsely from Lower Cretaceous (Barremian-Albian) strata of Maryland (Brenner 1963) and from north German sediments of Barremian- Aptian age (Doring 1966). o o C 8385 550 PALAEONTOLOGY, VOLUME 14 Genus crybelosporites Dettmann 1963 Type species (by original designation). Crybelosporites striatus (Cookson and Dettmann) Dettmann 1963. Crybelosporites brennerii sp. nov. Plate 105, figs. 18-20 1963 Perotriletes [«c] striatus non Cookson and Dettmann; Brenner, pp. 66-67, pi. 19, fig. 3, pi. 20, fig. 1. Diagnosis. Spores radial, trilete, spheroidal. Sclerine stratified, proximally cavate, consisting of: a relatively thick, homogeneous inner layer, 1-1 -5 pm thick; and an outer layer (sculptine), about 0-5 ^im thick, that is in close proximity to inner layer around distal surface and equator but projected proximally (and thus distinctly separated from inner layer by a cavum) to form a gula-type projection over proximal pole. Sculptine very finely sculptured (scabrate to microrugulate) ; sculpture usually even less con- spicuous on proximal projection which may be superficially almost laevigate. Laesurae usually very difficult to detect; where visible, are simple, straight slits extending about two-thirds of the distance to equator. Compressional folds of sculptine common. Dimensions (12 specimens). Equatorial diameter 33 (39) 46pm; polar diameter 42 (51) 60 ^am; diameter of inner layer (measured along polar axis) 30 (37) 44 pm. Holotype. Preparation G024b/15, 106-2 4-1, Y.1393. PI. 105, fig. 18. Equatorial aspect. Spore spheroidal; equatorial diameter 42 /xm, polar diameter 57 |ixm. Sclerine 2 /xm thick. Sculptine very thin, scabrate to microrugulate (lumina not delimited); microrugulae less than 0-5 ;um in width, up to 7 jixm long. Inner layer of sclerine about 1-1-5 /xm thick, featureless apart from several large-scale folds resulting from compression; diameter 40 pm (measured along equatorial axis) X 39 pm (along polar axis). Type locality. Saskatchewan, S.W.P. Bredenbury 11-36, core, 1391-1395 feet; Swan River Group. Remarks and comparison. From specimens currently available it is not entirely certain whether the sculptine is one- or two-layered. The former seems to be the case, though in all other respects the species is clearly conformable with Crybelosporites Dettmann 1963. ''Perotrilites' bursatus Hall 1963 (pp. 434, 436; figs. 16, 17) from Iowa (U.S.A.) Ceno- manian strata is close to, possibly identical with C. brennerii sp. nov. Although Hall’s pictures are none too clear, his species seems to be less distinctly sculptured than C. brennerii and the sculptine is thicker and forms a relatively minor proximal projection. C. striatus (Cookson and Dettmann) Dettmann 1963 (p. 81; pi. 18, figs. 1-6), which occurs in Late Aptian-Turonian sediments of eastern Australia (Dettmann and Playford 1969), differs from the present species in the sculptural characteristics of the outer exine layer, which is distinctly reticulate. Previous records. Brenner (1963) initially illustrated the form described above and noted its distribution throughout the Lower Cretaceous Potomac Group of Maryland. Genus coptospora Dettmann 1963 Type species (by original designation). Coptospora striata Dettmann 1963. G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 551 Coptospora williamsii sp. nov. Plate 106, figs. 8-13 Diagnosis. Spores hilate, biconvex; amb circular, subcircular, or convexly subtriangular. Exine stratified, comprising two layers that are closely adpressed or show only slight, partial detachment from one another (maximum separation up to 7 pm). Exoexine 5-9-5 pm thick (often variable, same specimen), very finely sculptured (scabrate, punc- tate, microvermiculate, granulate, to microrugulate). Distinct, circular to subcircular, potentially hilate area, centrally located on distal surface, and 52-65 /xm in diameter; area with sharply defined, incised, undulating margin and somewhat coarser, sparser sculpture (vermiculate-punctate) than remainder of exoexine to which it approximates in thickness. Hilum results from detachment of this caplike structure (see PI. 106, fig. 8). Intexine 0-5-1 pm thick, homogeneous, featureless. Tetrad mark absent or (occasionally) represented by poorly defined, low, subradial folds, up to 3-5 pm wide and half the spore radius in longitudinal extent. Dimensions (12 specimens). Equatorial diameter 82 (104) 140 jum. Holotype. Preparation G065/2, 98-7 23-2, Y.1408. PI. 106, figs. 9, 10. Distal aspect. Diameter 91 pm, amb broadly roundly subtriangular. Exoexine averaging 8 /um thick, with very fine sculptural pattern of microvermiculi and punctae, which are sparser and better defined on potentially hilate area. Latter area is subcircular in outline, 56 pm in diameter, with clearly incised, undulating margin. Intexine about O'SyLtm thick, in part contracted (by up to 2pm) from exoexine. No tetrad mark. Type locality. Manitoba, Swan River area, SW. |, sec. 10, Tp. 37, Range 27 W. P. Mer. ; Swan River Group (sample 69-W-32), about 40 feet below base of Ashville Group. Comparison. Coptospora williamsii sp. nov. is readily distinguishable from other species ascribed to the genus. The most closely comparable form appears to be C. dettmannae, which was described by Doring (1966, p. 110; pi. 6, figs. 1-4) from the German Neo- comian. The latter is distinct, however, from the Manitoba species in being differently and more coarsely sculptured (verrucate) and in having an apparently one-layered exine. Genus triporoletes Mtchedlishvili 1960 emend. 1960 Triporoletes Mtchedlishvili in Mtchedlishvili and Samoilovich, pp. 127-128. 1962 Rouseisporites Pocock, pp. 52-53. Emended diagnosis. Spores invariably devoid of proximal aperture; tetrad mark, where evident, comprises three radial ridges that are faintly or distinctly developed. Amb con- vexly subtriangular to subcircular. Sclerine consists of two layers; the outer being thin, membraneous, often loosely enveloping, zonate. Zona with a flask-shaped to conical invagination at each radial position of equator. Distal surface bearing muroid ridges, which may anastomose to form a reticulum; proximal surface smooth to reticulate. Type species (by monotypy). Triporoletes singularis Mtchedlishvili in Mtchedlishvili and Samoilovich 1960 (pp. 128-129; figs. 13, 14). Synonym: Rouseisporites triangularis Pocock 1962 (p. 54; pi. 7, figs. 110, 111). Illustrated herein as PI. 106, fig. 1. Occurrence: U.S.S.R., Siberia; Albian-Turonian (Mtchedlishvili and Samoilovich 1960; Samoilovich et al. 1961; Chlonova 1969). Western Canada, Lower Cretaceous (Pocock, 1962; Singh 1964, 1971; Vagvolgyi and Hills 1969; this paper). PALAEONTOLOGY, VOLUME 14 552 Other species. 1. Triporoletes granospeciosus (Delcourt and Sprumont) comb. nov. Synonymy: Cingulatisporites g ranospeciosiis Ddcourt and Sprumont 1955, p. 39; pi. 4, fig. 2. Occurrence: Belgium, Wealden. 2. Triporoletes involucratiis (Chlonova) comb. nov. [described, illustrated, and formally combined below]. 3. Triporoletes laevigatus (Pocock) comb. nov. (PI. 105, fig. 16). Synonymy: Rouseisporites iaevigatiis Pocock 1962, pp. 53-54; pi. 7, figs. 106-109. Occurrence: Western Canada, Lower Cretaceous (Pocock 1962; Singh 1964; this paper). U.S.A. (Oklahoma), Albian (Eledlund and Norris 1968). 4. Triporoletes radiatus (Dettmann) comb. nov. (PI. 106, figs. 3, 4). Synonymy: Rouseisporites radiatiis Dettmann 1963, p. 98; pi. 23, figs. 13-17. Occurrence: South-eastern Australia, Lower Cretaceous (Dettmann 1963). Western Canada, Lower Cretaceous (Singh 1971 ; this paper). 5. Triporoletes reticidatus (Pocock) comb. nov. (PI. 106, fig. 5). Synonymy: Rouseisporites reticulatus Pocock 1962, p. 53; pi. 7, figs. 101-105. Occurrence: Western Canada, Lower Cretaceous (Pocock 1962; Singh 1964, 1971; Vagvolgyi and Hills 1969; this paper). U.S.A. (Oklahoma), Albian (Hedlund and Norris 1968). South-eastern Australia, Lower Cretaceous (Dettmann 1963). Siberia, Lower Cretaceous (Chlonova 1969). Rumania, Albian (Baltes 1967). Argentina, Lower Cretaceous (Barremian- Aptian) (Archangelsky and Gamerro 1967). 6. Triporoletes simplex (Cookson and Dettmann) comb. nov. (PI. 106, fig. 2). Synonymy: Cingu- latisporites simplex Cookson and Dettmann 1958, p. 110; pi. 17, figs. 7, 8. Occurrence: South- eastern Australia, Lower Cretaceous (Aptian-Albian) (Dettmann 1963). Western Canada, Lower Cretaceous (this paper). In addition, several other relatively poorly known species regarded as possibly comparable with Rouseisporites (see Dettmann 1963, p. 96) must be considered similarly as akin to Triporoletes. Discussion. The type species of Mtchedlishvili’s genus Triporoletes {in Mtchedlishvili and Samoilovich 1960), T. singular is Mtchedlishvili, has been fully described in two Russian works — Mtchedlishvili and Samoilovich (1960) and Samoilovich et al. (1961). There remains no doubt as to its synonymy with Rouseisporites triangularis, a similarly well-documented species initially described (along with the genus) by Pocock (1962). Triporoletes, being instituted two years earlier, takes priority over Pocock’s genus. The diagnosis of Triporoletes given here is essentially based upon Dettmann’s (1963, p. 96) ‘restated’ version of the original Rouseisporites diagnosis. Delcourt, Dettmann, and Hughes (1963, p. 219) pointed to Zlivisporis Pacltova 1961 and Seductisporites Chlonova 1961 as being comparable with Rouseisporites. Further EXPLANATION OF PLATE 106 All figures X 500 unless otherwise specified; from unrelouched negatives. Figs. 1-5. Triporoletes spp. 1, T. singularis Mtchedlishvili 1960; median focus; G026/8, 113T 10-5; Y.1400. 2, T. simplex (Cookson and Dettmann) comb, nov.; distal focus; G046/6, 104-2 20-4; Y.1401. 3, 4, T. radiatus (Dettmann) comb. nov. 3, Median focus; G046/1, 135-2 9-6; Y.1402. 4, Distal focus; G065/117, 104-7 11-9; Y.1403. 5, T. reticulatus (Pocock) comb, nov.; distal focus; G026/23 109-4 14-8; Y.1404. Fig. 6. Schizosporis reticulatus Cookson and Dettmann 1959; G021b/26, 104-9 1-6; Y.1405. Fig. 7. Aequitriradites spinulosus (Cookson and Dettmann) Cookson and Dettmann 1961; median focus; G069/6, 94-6 9-1 ; Y.1406. Figs. 8-13. Coptospora williamsii sp. nov. 8, Lateral aspect; GO65/106, 109-0 13-5; Y.1407. 9, 10, Holotype; distal and proximal foci; G065/2, 98-7 23-2; Y.1408. 11, 12, Proximal and distal foci. 13, distal focus (X 1000); G065/2, 96-7 15-4; Y.1409. Palaeontology, Vol. 14 PLATE 106 PLAYFORD, Lower Cretaceous miospores •5 I. s. t\ ■; * \ G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 553 information on the two first named genera could result in their being attributed to Triporoletes. Ricciaesporiles Nagy 1968, from Hungarian Miocene strata, is closely similar to Triporoletes (al. Rouseisporites) as Nagy (1969, p. 317 footnote) has commented. Aside from their considerable age-disparity, it does not seem appropriate at this stage to merge the two generic taxa, for both species allocated by Nagy to Ricciaesporiles show a much closer resemblance to extant ricciaceous spores (in particular to those of Riccia — see Nagy 1968, pi. 7) than they do to the type (and other) species of Triporoletes. Affinity. Mtchedlishvili and Samoilovich (1960) implied that Triporoletes is angio- spermous; but this is considered much less likely than the hepatic relationship that has subsequently) been suggested (see Dettmann 1963, p. 96). The latter author considered it probable that the (distal) muroid elevations represent germinal exits as in extant Ricciaceae and Cleveaceae. Triporoletes invo/ucratns (Chlonova) comb. nov. Plate 105, figs. 13-15 1969 Rouseisporites involucratiis Chlonova, p. 55; pi. 9, figs. 8-12. Description. Spores circular to subcircular in equatorial outline; no tetrad mark evident. Sclerine two-layered: very thin, hyaline, outer layer enveloping thicker (0-75- 1 p.m thick) inner layer and extending beyond the latter equatorially as a discontinuous zonate structure. Except for zonate extensions, the two layers are in close contact. Outline of inner layer (polar view) undulating; zona attains maximum width (up to 10 /xm) within smoothly concave embayments, and is narrowest or non-existent (1-5 pm or less) about convex projecting areas (‘blades’ of Chlonova 1969). Embayments un- equal, number per specimen varies from 4 to 7. One surface (? distal) partly sculptured with fine, close-spaced, somewhat sinuous ridges, which are about 0-5-1 pm high and wide; sculpture typically displays radial or subradial orientation and extends about half the distance to equator. Remainder of spore surface smooth to faintly scabrate. Dimensions (15 specimens). Overall equatorial diameter 35 (44) 50 /u,m. Remarks and comparison. The species shows obvious morphological similarities with Triporoletes (al. Rouseisporites), but because only vague indications of pore-like or cone-like marginal invaginations were seen on an occasional specimen, the generic allocation must be regarded as provisional. Cookson and Eisenack (1962, p. 271 ; pi. 37, figs. 6-8) have illustrated and described Halophoridia xena, an incertae sedis palyno- morph from probable Albian-Cenomanian of Western Australia which shows some resemblance to Triporoletes involucratiis (Chlonova) comb. nov. The Australian species, however, is unwrinkled and displays a much more regular inner body or capsule with 4 strongly concave sides (hence ‘hour-glass-shape’), together with a more extensively encompassing outer layer. Previous records. The first and only previous report of T. involucratus is from Albian- Cenomanian deposits of western Siberia (Chlonova 1969). 554 PALAEONTOLOGY, VOLUME 14 NOTE ON RETITRICOLPITES AND TRICOLPITES The dispersed pollen genus Retitricolpites van der Hammen 1956 appears to be of confused status and uncertain validity (see discussion by Srivastava 1966, p. 547, and Potonie 1966, pp. 163-164). Two Tectogenerotypes’ have been nominated independently at different times — viz. R. ovalis van der Hammen and Wymstra 1964 and R. vulgaris Pierce 1961 by van der Hammen and Wymstra (1964, p. 235) and Potonie (1966, p. 163), respectively — since the type originally chosen, R. oniatus v. d. H., came from an extant species, Neea mocrophyl/a Poepp. and Endl. (van der Hammen 1956, p. 60). Even accepting the ultimately correct typification of Retitricolpites, the problem remains as to whether it can meaningfully be separated from Tricolpites Cookson ex Couper 1953 which also accommodates reticulate, tricolpate, fossil pollen. Such a problem can only be solved by a thorough reappraisal of the type and other contents of the two taxa. For the present purpose, no alteration is made or suggested in the generic allocation of species hitherto placed in Retitricolpites. DISCUSSION OF THE PALYNOLOGICAL FLORA Swan River stratal occurrences of the spore-pollen forms identified are shown on Table 1, which also notes the presence of non-calcareous microplankton (dinoflagellate cysts and acritarchs) in certain of the samples. These latter palynological components are not discussed further in this work; their presence may be taken as a reasonable indication of marine conditions or at least of nearby marine influences. Biostratigraphic significance Although a detailed palynological scheme has yet to be established for the western Canada Lower Cretaceous, a reasonably accurate assessment of the age and correlation of the Swan River sediments is possible by reference to existing published information and particularly to Singh (1971). The latter, in the most intensive study carried out to date in the region, emphasizes the biostratigraphic utility of the introduction of angio- spermous microfossils which in his studied section is charted at a horizon in the Harmon Member (Middle Albian) of the Peace River Formation. Useful also in a biostrati- graphic sense are Norris’s (1967) data which demonstrate a distinct ‘microfloral break’, marked especially by abrupt angiosperm introduction, at the Grand Rapids-Joli Fou contact. Norris (pp. 81-82) commented on this at some length, suggesting the possibility of a disconformable relationship between the two formations. The Early Cretaceous age which has generally been held for the Swan River strata is exemplified by their palynological flora, which possess numerous spore-pollen elements known previously from Lower Cretaceous sediments of Canada and U.S.A., and of many distant localities, viz. in western Europe, U.S.S.R., and Australia. Such affiliations with coeval extra-North American assemblages will be discussed in an ensuing section of this paper. The subsurface Saskatchewan sequence studied herein (Bredenbury no. 1 1-36 well) shows some microfloral variation within its 122 feet of Lower Cretaceous Swan River strata (see Table 1). Certain forms, known from previous work to have biostratigraphic utility, show restricted vertical distribution, viz. certain angiospermous elements; trilete spores attributed to Camarozonosporites ambigens and Asbeckiasporites wurthi', and SWAN R. DISTRICT S.W.P. BREDENBURY No. 11-36 ' ^ ^ III ^ to V. < S ^ •5-5-5- 5 S- f? 1^ to Iq. 5 15:^ c; d; ■§ <5 to to -5 ■5-5 O O to Q, Q..5 S .<0 .to Ts ta C> Ol ^'=TCLCU .to < -5 ^1 to -$ 5> =j "5 ^ ^ ? s: tb 5-5-^ ^ ^ ^ to to .5 ■§•§ < «o to to •$ tj :o Cl > -Q to to ^ ^■§1 c 2; ? - to:5 to -5 5 tiJ ti> 5: ^ ^ ^ to § -^ ^ §- -5 S ^ X'^'t 5- Q.Ci:-5 to 5j to I'- .to 5 a" (A ^ S,S 'll to ' tb Jo tb ^ to to Q : to to-^ Q =j.to t; to-^ C; < ^ tb 5 5 to to cc cjc cJ! ki Ly I s to ki 5 to «4_* o ^ cL-5^ to <0 tb .8^ >“ ^ to to --. - tOi'-— lo < to ^ to to t>) tb C; '5; C» C 9 ki to k) -j ^ ^tio C^Qc k ^ ;5 t; to to <; tb tb .to. to k o 5-tO -5-1 ^ -5- o V .to to .to 5 t to to 5 5 '^k:5 o ^ to k: 5 to CO. to I .to to Q o ^3 .N ^ S kgl? ko to ko O G 0 G 02 G O' G024, Pref Numb } — 1 CVJ »N ro N r-: CD N N- r- 00 N 00 r- 0 CO CD CVJ 00 m CO 'd- CO 10 00 00 0 00 00 CD CD 0 cn 1 CVJ CD ro CD CD 10 CD CD CD f^ CD CD CD CD 0 0 2j CVJ 0 ro 0 10 N 0 00 0 CD 0 > — • 1 9 9 1 9 9 9 9 9 9 9 9 9 9 1 9 9 9 9 9 • 9 9 j 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 ) — * 9 9' 9 ! 9 9 9 9 9 9 9 f 9 9 9 ■n 9 9 9 1 9 1 9 9| 1 "1 n 1 i X 9 1 1 P 9 1 F m □ 9 9 9 9 n 9 9 9 □ — !■ 1 r: □ 9 9 [■: t- r: □ 9 9 □□EH 9 9 9 9 9 □ tj □ n 9 9 9 i □ 9 9 9 9 9 1 9 9 rm 1 c □nn I9I9I M w M M 191 M TiTi nx9i 1 m 191 m \ i9n □nnnnnc 1 I9M M M9M9M M M M M M M M M9M \ □nn □nn 9 9 r r 9 □ P P 9 9 9 )! □nn □nn 1 9 9 9 9 9 9 9 n 9 9 1 !f i 9 9 9 9 9 9 9 P 9 9 9 9' 9 m • 9 9 9 9 9 j 9 9 9 9i 9 9 9 ) ), ^ W 1 M □□□ 1 ■1.9j9L □□□□ 1 I9I9I 1 1 1 191 191 1 □EIG P □nnnn G 0 6h G 06 G 06 [•I m nnn rrrp nnnnn M |9| M M 1 □□n □□□ □nn nnnn l^lol m\ i#i» □□□■nFinnn lAlAIAlAl nnn«nFinnn Slereisporites onhquasporites Birehsporites pofoniaei Undulatisporites pannuceus ConcavissimisporUes minor Concovissimisponte s punctotus ConcavissimisporUes vanverrucatus Osmundacidites wellmanli Boculotisporites comaumensis Lophofn/etes babsae Pilosisporites trichopapillosus < 3~| Pilosisporites verus Kuylisporiles hjnoris £] Lycopodiumsporites austroc/avahdifes Lycopodiumsporites morgmatus Lycopodioadites mtroverrjcatus ReticuUsporites e/ongotus Microreticu/otisporites uniformis Klukisporites pseudoreticulalus Tigrisporites scurrondus Cicotricosisporites austrahensis Cicatricosisporiles haUei Cicotricosisporites hughesi Cicotricosisporites potapscoensis Cicotricosisporites spiralis Cicotricosisporites potomacensis Cicotricosisporites sp. B Costotoperforospontes foveolatus Bolmeisporites sp.cf. Bbotodictyus Arcelliles disaformis Appendicisporites bifurcalus Appendicisporites bi lateralis Appendicisporites cnmensts Appendicisporites jansonh Appendicisporites matesovae Appendicisporites potomacensis Appendicisporites problemoticus Trilobosporites apiverrucatus TrUobasporites hannomcus Trilobosporites humihs Trilobosporites marylandensis Trilobosporites purverutentus Jschyosporites sp, cf. / crater is deicheniidites senonicus Clavifero rudis Sestrosporttes pseudoalveolatus Coronatisporo vatdensis Camarozonosporites ambigens Foraminisporis osymmetricus Foraminisporis daUyi Foraminisporis wonthoggiensis Polycingulotispontes reduncus Taurocusporites segmentatus Asbeckiasporiles wurthi CingufrUetes dovus Spore type A Kroeuselisporiles hostUobotus Lundbladispora reficingula Densoisporites circumundulatus Densoisporites microrugulatus Crybelosporifes brennerii PerotriUtes ponnuceus Laevigatosporites ovatus Cicotricososporites auritus Coptospora wiHiamsii Aequitrirodites spinulosus Couperisporites complexus Cooksonifes variabiUs Triporoletes invotucrotus Triporotetes loevigatus Triporoletes radiatus Triporoletes retlculotus Triporoletes simplex Triporoletes singutaris TsugaepoUenites dampieri Vitreisporites pallidas AUsporites biloteralis AUsporites grandis Podocarpidites canadensis Podocarpidites sp.cf P elliplicus Podocarpidites muUesimus Podocarpidites radiatus Porvisaccites radiatus Ptiyilocladidites inchootas Rugubivesiculites reductas InoperturopoHeniles Hmbotus Araucariacites australis Reticulalasporiles dupHexinous PerinopoUenites etotoides Eucommiidites minor Fucommiidites troedssonii Equisetosporites sp.cf E. concinnus Equisetosporites Jansonii Equisetosporites mutticostatus Clavotipollenites hughesii Clavotipollenites rotundus Liliacidites peroreticutotus Retitricolpites georgensis Retitricotpites prosimiiis Retitricolpites vulgaris Tricotpites sogax Fraxinoipolleniles venustus StnatopoUis paraneus ExesipoUenites tumulus CircuUna porva Schieosporis reticulatus Classopollis dassoides Organic-wolled marine microplonkton G. PLAYFORD; PALYNOLOGY OF SWAN RIVER STRATA 555 bisaccate grains of the taxa Phyllocladidites inchoatus and Rugubivesiculites reductus. These all appear in the Bredenbury well at 1340-1345 feet and higher levels and are present also in the sampled type outcrops of the upper Swan River Group of the Swan River area, Manitoba. Accordingly, the latter are correlative with the Bredenbury well interval 1273-1345 feet. As noted previously a Middle Jurassic spore-pollen suite is recognizable in the Breden- bury well at a depth of 1408-1413 feet. The overlying sampled interval between 1395 and 1273 feet is clearly of Early Cretaceous age, as testified by association of such spores as Cicatricosisporites australiensis, C. hallei, Pilosisporites trichopapiUosus, Appeudici- sporites spp., Trilobosporites apiver meatus, T. marylandensis, T. purvendentus, Foramini- sporis asymmetricus, Aequitriradites spimdosus, Triporoletes (al. Roiiseisporites) spp., ClassopoIIis classoides, Circulina parva, ExesipoUeuites tumulus, Clavatipollenites spp., Alisporites bilateralis,A. grandis, and Podocarpidites muJtesimus. In addition, many other less consistently occurring species conform with an Early Cretaceous dating for both subsurface and outcrop material; e.g. Undulatisporites pannuceus, Concavissimisporites spp., Kuylisporites hmaris, Microreticulatisporites uniformis, L. margiuatus, Cicatricosi- sporites hughesi, Crybclosporites brennerii, Couperisporites complexus, Cooksouites variabilis, Equisetosporites multicostatus, Liliacidites peroreticulatus, and Podocarpidites canadensis. The subsurface Bredenbury section below 1345 feet (i.e. sampled section 1353-1395 feet) is devoid of such angiospermous elements as Retitricolpites, Tricolpites sagax, Eraxinoipollenites, and Striatopollis. From this it is considered that the lower Bredenbury Swan River sequence is a Mannville Group equivalent, being just older than the Harmon Member of the Peace River Formation and certainly antedating the Joli Fou Formation (cf. Singh 1964, Norris 1967). In a more positive sense, the lower Bredenbury section includes Clavatipollenites rotundas (probable synonym: Liliacidites dividuus) which is unknown from pre-Albian strata (Kemp 1968, Brenner 1963, Hedlund 1966, Hedlund and Norris 1968); Triporoletes involucratus for which the same age-restriction seems likely, at least in western Siberia (Chlonova 1969); Cicatricosisporites spiralis, which in Singh’s (1971) Peace River section occurs only in the Loon River Formation (early Middle Albian); Trilobosporites purvendentus (not known indubitably from below the Aptian: Dettmann 1963); and Triporoletes simplex (Aptian-Albian: Dettmann 1963). Moreover, the distinctive ‘‘Cicatricosisporites sp. B’ of Singh (1964), a component of the oldest Cretaceous sample of the Bredenbury well, is known elsewhere only from Upper Mannville (Grand Rapids Formation) and younger Alberta strata (cf. Singh 1964, Norris 1967). Collectively the palynological evidence suggests that the immediately supra-Middle Jurassic Bredenbury section of lower Swan River strata is Albian in age, probably early Middle Albian, and correlative with the Upper Mannville/Loon River segments of the Alberta Lower Cretaceous to the west (see Table 2). The upper Swan River sequence of the Bredenbury well (i.e. 1273-1345 feet) and of sampled type exposures near Swan River contain undisputed angiospermous elements together with other palynomorph species not encountered in the lower well section. Stratigraphic significance of the entrance levels of certain of these forms has already been connoted by Singh (1971) and Norris (1967), although as mentioned earlier the sequence studied by the latter was probably interrupted by non-deposition in the later Middle Albian. The following appear, from previous knowledge of their North American 556 PALAEONTOLOGY, VOLUME 14 LOWER PEACE RIVER LOWER ATHABASCA RIVER S.E. SASKATCHEWAN and S.W. MANITOBA UPPER SHAFTESBURY FORMATION AAAAAAA A 'FISH-SCALE MARKER BED'. aaaaaaaa AAAAAAAA AAAAAAAA AAA lower a a a AAAAAAAA SHAFTESBURY FORMATION AAAAAAAA aaaaaaaa AAAAAAA A AAAAAAAA LA BICHE FORMATION ASHVILLE GROUP • UPPER , SHALE UNIT* A A A A A A^A a. a . A A Paddy Member A A lA 5iA a| a Acc A A >A PELICAN FORMATION VIKING FORMATION A A A A A A A A A A A A Cadotte Member A A A A A A A A A A A A AAAAAAA JOLI FOU FORMATION JOLI FOU FORMATION* A _A _A _A A Notikew.n Member GRAND RAPIDS FORMATION AA AAAAAA AAAAAAAA LOON RIVER FORMATION A AAAA.AAAA Bluesky Member CLEARWATER FORMATION O O Wabiskaw Member o GETHING FORMATION s 5 OOOOOOOO O OOOOOO O O McMURRAY FORMATION O OOOOOOOO OOO OOOOO ^ GRAND RAPIDS ^ + + -F + + + + + + + + + + + + + + -F + + -F-F + + ■F+ + + -F + + + f + SWAN RIVER + + + + GROUP + + - + + + +-F + + + + + + + + + + -F + + -F + + -F + + + + + + ‘ FACIES “X- J(^X X xf'jr " CLEARWATER X _X ^CIK _X Wabjskaw Mem^r X Calcareous Member x X X X X X X X X X Elierslie Member X X X X X ^ _X_ X Deville Member w XXX X X -Singh 1964 — Norris 1967 OOO O O Vagvolgyi & Hills 1969 < < < < < Singh " 1971 ■F -F + TABLE 2. Correlation chart of western Canadian Lower Cretaceous strata which have been studied palynologically. In addition to authors documented, Pocock (1962) studied Upper Jurassic-Lower Cretaceous material from various localities in Alberta and Saskatchewan; and Pocock (1965) and Steeves and Wilkins (1967) investigated Upper Mannville correlatives in the Saskatoon area, Saskat- chewan. Stratigraphic nomenclature, intraprovincial correlation, and age of the Alberta columns are extracted from Singh (1971). G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 557 distribution (cited below), to be stratigraphically important components of these upper horizons : Retitricolpites vulgaris: ? Cenomanian, Minnesota (Pierce 1961); Albian, Maryland (Brenner 1963); (? late Middle) Albian, Oklahoma (Hedlund and Norris 1968); Joli Fou-Upper Shale Unit, Alberta (Norris 1967); Harmon-upper Shaftesbury, Alberta (Singh 1971). FraxinoipoUemtes vemistus: Cadotte-upper Shaftesbury, Alberta (Singh 1971). Retitricolpites prosimilis: Joli Fou-Upper Shale Unit, Alberta (Norris 1967); Harmon- upper Shaftesbury, Alberta (Singh 1971). Tricolpites sagax: Joli Fou-upper Shale Unit, Alberta (Norris 1967); (? late Middle) Albian, Oklahoma (Hedlund and Norris 1968); Cadotte-upper Shaftesbury, Alberta (Singh 1971). Camarozonosporites ambigens: Albian, Maryland (Brenner 1963); (? late Middle) Albian, Oklahoma (Hedlund and Norris 1968); Viking-Upper Shale Unit, Alberta (Norris 1967); Cadotte-upper Shaftesbury, Alberta (Singh 1971). Asbeckiasporites wurthi: Paddy-lower Shaftesbury, Alberta (Singh 1971). Phyllocladidites inchoatus: ? Cenomanian, Minnesota (Pierce 1961); Joli Fou-Upper Shale Unit, Alberta (Norris 1967); Cadotte-upper Shaftesbury, Alberta (Singh 1971). Retitricolpites georgeusis: Albian, Maryland (Brenner 1963); (?late Middle) Albian, Oklahoma (Hedlund and Norris 1968); upper Joli Fou-Upper Shale Unit, Alberta (Norris 1967); Harmon-upper Shaftesbury, Alberta (Singh 1971). StriatopoUis paroneus: (? late Middle) Albian, Oklahoma (Hedlund and Norris 1968); Joli Fou-Upper Shale Unit, Alberta (Norris 1967); Harmon-lower Shaftesbury, Alberta (Singh 1971). Rugubivesicidites reductus: ? Cenomanian, Minnesota (Pierce 1961); Albian (younger Patapsco), Maryland (Brenner 1963); Viking-Upper Shale Unit, Alberta (Norris 1967); Cadotte only, Alberta (Singh 1971). Undidatisporites pammceus: Upper Barremian-Albian, Maryland (Brenner 1963); Cadotte-lower Shaftesbury, Alberta (Singh 1971). Lophotriletes babsae: Albian, Maryland (Brenner 1963); Cadotte, Alberta (Singh 1971). Pilosisporites tricliopapillosus: youngest documented occurrence is upper Middle Albian (Cadotte), Alberta (Singh 1971). Microreticulatisporites imiformis, Appendicisporites bifiircatus: youngest known occur- rences are lowest Upper Albian (Paddy), Alberta (Singh 1971). From the occurrence of these forms (see Table 1) it is evident that the upper Swan River strata (1273-1345 feet, Bredenbury well; type exposures. Swan River) are late Middle Albian, possibly to earliest Late Albian in age ; and that they are correlatives of the Peace River Formation (probably mainly of its Harmon and Cadotte Members) of north-western Alberta (see Table 2). No microflora is available from the lowest Peace River member, the Notikewin Member. The sequence of entrance levels of certain species cannot be matched precisely with those documented by Singh (1971) in the Peace River Formation, but this is not altogether surprising in view of the rarity of many of the taxa and the possibility of hiatuses, however minor, in neritic strata like the Swan River. There appears, in fact, to be no palynological or other evidence of a pronounced break in the subject sequence. 558 PALAEONTOLOGY, VOLUME 14 which from the above is adduced as being wholly Albian (largely, if not entirely, Middle Albian) in age. Indication of the presence or proximity of marine conditions is found only in the uppermost part of the Bredenbury sequence (1273-1288 feet) and in the upper Swan River of its type area. The overlying strata of the Ashville Group are, of course, marine, and although lithostratigraphically correlative with the Joli Fou of Alberta are on the palynological data presented here somewhat later Albian in age. These considerations accord with the comments of Price (1963, pp. 34-35) that north-eastwards from southern Saskatchewan, towards Swan River, ‘the sand-shale boundary (i.e. Swan River- Ashville boundary) rises stratigraphically above the first occurrence of distinctly marine beds’ (cf. also Price, table 1, p. 3). An alternative possibility concerning the correlation and age of the Swan River Bredenbury section should not be overlooked. Keeping in mind the somewhat erratic distribution and general rarity of the tricolpate angiosperm palynomorphs, their seeming absence below the 1345-foot level might be one of fortuity and not necessarily truly indicative of non-existence. Were this the case, deposition of the studied Swan River might wholly post-date introduction of tricolpate angiosperm pollen and could be regarded as coeval with Joli Fou accumulation. This alternative idea would conform with Stelck’s (1958, p. 6) suggestion that the Swan River Group, together with the St. Edouard Member and other ‘basal Colorado sands’ to the west, represents an early onlap phase of the Joli Fou sea, regarded as having transgressed northwards from the Gulf of Mexico late in the Early Cretaceous (post-Mannville). As documented above, monosulcate grains of possible angiospermous alliance (Clavatipollemtes hughesii, C. rotimdus, Liliacidites peroreticulatus) do occur in the lower Swan River portion of the Bredenbury drill-hole. Although none of these has yet been recorded from Alberta strata older than the Harmon Member, they are known elsewhere (e.g. U.S.A., England) to range considerably lower in the Lower Cretaceous than the here-recorded initial tri- colpate forms (cf. Brenner 1963, Doyle 1969). Taking all the available data into account, the proposition that the Swan River strata are entirely of post-Mannville age, belonging wholly to the basal Colorado gulfian depositional realm, seems rather less tenable than the view expressed earlier herein that the sediments are partly correlative with the Upper Mannville. Farther south, correlation of the Swan River strata with the Albian sector (Patapsco Formation) of the Maryland Lower Cretaceous (Brenner 1963) is also evident from the above discussion. Fredericksburgian sediments (Antlers Sand and ‘Walnut’ clay) of Oklahoma (Hedlund and Norris 1968) include many palynological features in common with the Saskatchewan and Manitoba sediments. However, the Oklahoma material contains a considerably greater diversity of angiospermous pollen (viz. tetra- and penta- colpate and tetra- and pentachotomosulcate grains) suggesting that it may be somewhat younger than the Swan River Group; or alternatively such disparity could reflect differing phytogeographic situations. Natural relationships Attribution of dispersed spores and pollen grains to natural botanical groups still remains a difficult objective so far as pre-Tertiary palynological floras are concerned. Some authors who have studied the Canadian Lower Cretaceous (e.g. Singh 1964, 1971 ; G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 559 Pocock 1962, 1965) endeavour to utilize a natural scheme, to family level, for supra- generic placements of the morphographically founded taxa; whereas others (e.g. Norris 1967; Steeves and Wilkins 1967; present study) apply the artificial (Turma) scheme in the systematic-descriptive portions of their papers. As Norris (1967, p. 83) has men- tioned, natural affiliations of many sporae dispersae can be elucidated, albeit with vary- ing precision, on the basis of morphological comparisons with in situ spores and pollen from fructifications of fossil and extant plants. Synopses compiled by Potonie (1962, 1967) show how far palaeopalynology has progressed towards the desirable objective of natural classification of plant microfossils. On the basis of current knowledge of affinities of Mesozoic sporomorphs, the follow- ing tabulation depicts the most likely natural allocation of the Swan River taxa: bryophyta: Stereisporites antiquasporites, Foraminisporis asymmetricus, F. dailyi, F. wonthaggiensis, Ciugutriletes claviis, Coptospora williantsii, Aequitriradites spintdosus, Couperisporites complexus, Cooksonites variabilis, Triporoletes involucratiis, T. laevigatus, T. radiatus, T. relicidatus, T. simplex, T. singidaris. PTERIDOPHYTA. Lycopodiales ! Lycopodiiimsporites austroclavatidites, L. marginatus, Sestrosporites pseiidoalveolatiis, Camarozonosporites ambigens, Kraeuselisporites hastilobatus, Luiidbladispora reticingida, Densoisporites circiimimdidatus, D. micronigidatiis. Filicales: Cyathidites australis, C. minor, Biretisporites potoniaei, Coucavissimisporites minor, C. punctatiis, C. variverrucatus, Osmwida- cidites wellmanii, Bacnlatisporites comaumensis, Pilosisporites trichopapillosiis, P. veriis, Microreticidati- sporites imiformis, Kliikisporites pseiidoreticidatiis, Cicatricosisporites australiensis, C. hallei, C. Imghesi, C. patapscoensis, C. potomacensis, C. spiralis, C. sp. B, Costatoperforosporites foveolatus, Balmeisporites sp. cf. B. holodictyus, Arcellites disciformis, Appendicisporites bifurcatus, A. bilateralis, A. crimensis, A. jansonii, A. matesovae, A. potomacensis, A. problematicus, Trilobosporites apiverrucatus, T. lian- nonicus, T. Inimilis, T. marylandensis, T. purverulentus, Ischyosporites sp. cf. I. crateris, Gleicheniidites senonicus, Clavifera rudis, Crybelosporites brennerii, Cicatricososporites auritiis. Pteridophyta incertae sedis: Undulatisporites panniiceiis, Lopliotriletes babsae, Kiiylisporites limaris, Lycopodiacidites intra- verrucatus, Reticulisporites elongatiis, Tigrisporites scurrandiis, Coronatispora valdensis, Polycingulatis- porites reduncus, Taurocusporites segmentatus, Asbeckiasporites wurthi, Laevigatosporites ovatus. CAYTONIALES: Vitreisporites pallidas. CONIFEROPHYTA. Conifcrales: Tsugaepollenites dampieri, Alisporites bilateralis, A. grandis, Podocar- pidites canadensis, P. sp. cf. P. ellipticiis, P. multesimus, P. radiatus, Parvisaccites radiatus, Pliyllo- cladidites inchoatus, Rugubivesiculites reductus, Araucariacites australis, Perinopollenites elatoides. Coniferophyta incertae sedis: Eucommiidites minor, E. troedssonii, Exesipollenites tumulus, Circulina parva, Classopollis classoides. GNETALES. Ephcdraceae: Equisetosporites sp. cf. E. concinnus, E. jansonii, E. multicostatus. ANGIOSPERMOPHYTA : Clavatipollenitcs hugliesii, C. rotundus, Liliacidites peroreticulatus, Retitricolpites georgensis, R. prosimilis, R. vulgaris, Tricolpites sagax, Eraxinoipollenites venustus, Striatopollis paraneus. SPORES AND POLLEN GRAINS INCERTAE SEDIS: PerotrUites pannuceus, Inaperturopollenites limbatus, Reticulatasporites dupliexinous, Schizosporis reticulatus. From the above data, it is evident that the Swan River palynological flora is composed, qualitatively and quantitatively, of predominantly pteridophytic-coniferophytic vegeta- tional elements; this is comparable to the situation adduced elsewhere in coeval North American strata (e.g. Brenner 1963, Norris 1967). Ferns were prominent and diverse contributors, especially those of the family Schizaeaceae which is represented by a variety of cicatricose-sculptured trilete spores belonging to the genera Cicatricosisporites and Appendicisporites. The family Gleicheniaceae is also represented, albeit much less diversely, by abundant and persistent occurrence of Gleicheniidites. Water ferns con- tributed spasmodically, as suggested by the rare presence of Balmeisporites, Arcellites, 560 PALAEONTOLOGY, VOLUME 14 and Crybelosporites. Lower plants of the division Bryophyta produced a variety of trilete and hilate spores, which are, however, numerically inconspicuous in the subject flora. Abundance of coniferalean plants is attested by high frequency of bisaccate pollen attributed to the genera AUsporites, Podocaipidites, and Parvisaccites. From the rare occurrence of Ephedra-type pollen, it appears that the Gnetales made strictly minor contributions. The class Caytoniales is persistently represented by small bisaccate grains of the species Vitreisporites pallidus. A poorly differentiated angiosperm flora is suggested by the palynological analysis. Grains of possible angiospermous afliliation (belonging to the genus ClavatipoIIenites), occur in strictly minor amounts and are associated in the younger Swan River sediments with simple dicotyledonous (tricolpate) pollen. Angiosperm or possible angiosperm components are greatly subordinate to the pteridophytic, bryophytic, and gymnosperm elements. It is uncertain whether this reflects a truly minor and unspecialized representation of angiosperms in the contem- porary terrestrial flora or is a function of the sedimentology of the strata concerned. However, the suggested neritic depositional environment could well explain the presence of diverse pteridophytic and bryophytic palynomorphs. Comparison with assemblages beyond North America Several authors concerned with Late Mesozoic palynology (e.g. Mtchedlishvili and Samoilovich 1962, Verbitskaya 1966) have taken cognizance of the ubiquity of many spore and pollen elements of this age through northern and southern hemisphere EXPLANATION OF PLATE 107 All figures x 500 unless otherwise specified; from unretouched negatives. Fig. 1. AUsporites bilateralis Rouse 1959; median focus; G024c/57, 115-7 18-8; Y.1410. Fig. 2. Vitreisporites pallidus (Reissinger) Nilsson 1958; median focus; G024b/29, 103-9 8-8; Y.1411. Figs. 3, 4. Eucommiidites spp. 3, E. minor Groot and Penny 1960; G046/1, 122-1 19-7; Y.1412. 4, E. troedssonii Erdtman 1948; G062/15, 104-3 14-8; Y.1413. Fig. 5. Circidina parva Brenner 1963; median focus; G024c/24, 107-3 17-8; Y.1414. Figs. 6, 7. Equisetosporites spp. 6, E. jansonii Vocock 1965; G065/64, 110-9 15-0; Y.1415. 7, Eqidse- tosporites midticostatns (Brenner) Norris 1967; G029b/1, 119-0 20-6; Y.1416. Fig. 8. liiaperturopollenites limbatus Balme 1957; G021b/16, 104-1 16-8; Y.1417. Figs. 9, 10. Fraxinoipollenites vemistus Singh 1971; lateral views (xlOOO). 9, G061/3, 118-8 3-3; Y.1418. 10, G061/1, 93-1 22-1; Y.1419. Figs. 11, 12, 17, 18. Retitricolpites spp. ( X 750). 11,12, Retitricolpites georgensis Brenner 1 963 ; lateral aspect, high and low foci; G059b/1, 106-9 4-7; Y.1420. 17, R. vulgaris Pierce 1961; lateral aspect; G059b/1, 127-2 20-7; Y.1425. 18, R. prosimilis Norris 1967; lateral aspect; G064/1, 105-6 13-6; Y.1426. Figs. 13, 14. ClavatipoIIenites spp. (x 750). 13, C. hughesii Couper emend. Kemp 1968; median focus; G065/1, 113-1 12-6; Y.1421. 14, C. rotundas Kemp 1968; distal focus; G061/2, 100-5 16-9; Y.1422. Fig. 15. Exesipollenites tumulus Balme 1957 (x750); G065/1, 89-6 8-7; Y.1423. Fig. 16. Tricolpites sagax 'Horns, 1967 (x750); polar aspect; G023b/1, 114-5 13-2; Y.1424. Fig. 19. Parvisaccites radiatus Couper 1958; oblique aspect; G062/1 1, 1 12-2 12-0; Y.1427. Fig. 20. Rugubivesiculites reductus Pierce 1961; proximal focus; G061/5, 101-0 12-0; Y.1428. Fig. 21. Podocarpidites multesimus (Bolkhovitina) Pocock 1962; median focus; GO24c/60, 96-3 19-2; Y.1429. Figs. 22, 23. Couperisporites complexus (Couper) Pocock 1962; proximal and distal foci; G046/7, 104-9 18-2; Y.1336. Palaeontology , Vol. 14 PLATE 107 22 23 PLAYFORD, Lower Cretaceous miospores G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 561 deposits that have been subjected to such study. This is particularly true of the basal part of the system (Neocomian-Aptian) which may thus be correlated widely on palynological grounds. Dettmann and Playford (1969, pp. 198-199), for instance, have noted strong microfloral uniformity among Neocomian-Aptian assemblages of Australia, South America, and India, and further that certain (bryophytic and pteridophytic) elements from these areas occur also in the Soviet Union (especially Siberia) and western Canada. Likewise the Wealden facies of north-west Europe displays conspicuous palynological uniformity over its area of development (cf. Couper 1958; Hughes 1958; Doring 1965, 1966; Burger 1966; Levet-Carette 1966; Hughes and Moody-Stuart 1967). In turn, the Wealden microfloras show certain similarities to those from the Early Cretaceous of eastern U.S.A. (see Stover 1964) and to a lesser extent of western Canada and U.S.S.R. There is less published palynological information available on the Albian than on the earlier Cretaceous. The most notable widespread palaeobotanical feature of the Albian is the occurrence, in many areas and in significant proportions, of undisputed angio- sperms following their presumed inception in mid-late Early Cretaceous times (Hughes 1961). This feature is displayed by the Swan River sequence, in the younger horizons of which tricolpate grains of presumed dicotyledonous type are introduced. These early angiosperm grains are very small and morphologically simple. Their taxonomy tends to have somewhat limited (localized) application; this is probably a result of the difficulty of illustrating their critical morphological features with conventional optical equipment. With more sophisticated techniques, the taxa may well prove to have wider geographic application than is currently documented. Table 3 shows currently known extra-North American Albian occurrences of specific miospore taxa that have been identified in the Swan River samples of this study. Apart from Clavatipolleuites and Schizosporis, the taxa are of bryophytic, pteridophytic, or coniferophytic derivation. Omitted from this tabulation are the following cosmopolitan and relatively long ranging forms which are common to all or most of the regions: Stereisporites antiquasporites, Cyathidites australis, C. minor, Osmundacidites wellmanii, Bacidatisporites comaimiensis, Lycopodiumsporites austroclavatidites, Gleicheniidites senonicus, Laevigatosporites ovatus, Araucariacites australis, Vitreisporites pallidis, and CJassopoUis classoides. The table emphasizes the paucity of Albian palynological data available on a world-wide scale. It indicates a close palynological similarity between the Saskatchewan-Manitoba assemblages and those from the Siberian and eastern Australian Albian (embracing all or part of the Coptospora paradoxa and Tricolpites pannosus Zones of Dettmann and Playford 1969). The apparently only slight resemblance between the Saskatchewan-Manitoba assemblages and those of England, Portugal, Germany, and Rumania may reflect, at least in part, the strictly minor amount of pertinent palynological data relating to those European areas. Eor instance the English Albian column is founded upon a single paper (Kemp 1968) that is concerned exclusively with Barremian-Albian pollen of probable angiosperm origin. A possibly Albian assemblage from south-west Nelson, New Zealand (Norris 1968), bears little similarity to those of the western Canadian Albian. The same is true of Albian microfloras illustrated by Jardine and Magloire (1965) from Senegal and Ivory Coast, west Africa. Acknowledgements. This work was largely undertaken at the Department of Geology, University of Alberta, Edmonton, during the writer’s 1969 sojourn there as Visiting Professor; it was completed 562 PALAEONTOLOGY, VOLUME 14 subsequently at the University of Queensland. The assistance and helpful advice given by Drs. G. D. Williams, C. R. Stelck (both of University of Alberta) and Dr. C. Singh (Research Council of Alberta) are gratefully acknowledged. Selected species in Swan River strata Albian occurrences outside North America England Germany Portugal Rumania U.S.S.R. Australia Biretisporites potoniaei cf. Concavissimisporites punctatus . . . X X X Pilosisporites trichopapillosus .... X X Pilosisporites venis X Kuylispohtes lunaris X X Khikisporites pseudoreticidatus . . . X X Cicatricosisporites australieusis. . . . X X Cicatricosisporites hughesi X Bahneisporites cf. holodictyus .... X Appendicisporites crimensis X Appendicisporites matesovae X Trilobosporites apivermcatus .... X X Trilobosporites hannonicus X Trilobosporites purvendentus .... X X Clavifera nidis X Sestrosporites pseudoalveolatus . . . X X Camarozonosporites ambigeus .... X X Foramimsporis asymmetricus .... X X Foram in isporis daily i X X Foraminisporis wonthaggiensis. . . . X X Asbeckiasporites wurthi X Cingutriletes clavus X X Aeqiiitriradites spimdosus X X X Cooksonites variabilis X Triporoletes involucratus X Triporoletcs radiatns X Triporoletes reticidatus X X X Triporoletes simplex X Triporoletes singularis X Tsugaepollenites dampieri X Alisporites grandis X X Podocarpidites cf. ellipticus X Podocarpidites multesimtis X cf. cf. Parvisaccites radiatns X Eucommiidites minor X Eiiconimiidites troedssonii X Clavatipollenites hughesii X X Ciavatipollenites rotundus X Schizosporis reticulatus X X TABLE 3. Albian occurrences of Swan River palynomorphs outside North America. Sources of data: England, Kemp 1968; Germany, von der Brelie 1964; Portugal, Groot and Groot 1962; Rumania, Baltes 1967; U.S.S.R. (principally Siberia), Bolkhovitina 1956, 1959, 1961, 1966, 1968, Chlonova 1969, Fradkina 1967, Kotova 1968, Samoilovich etal. 1961, Verbitskaya 1962; (eastern) Australia, Dettmann 1963, Dettmann and Playford 1969. REFERENCES ARCHANGELSKY, s. and GAMERRO, j. c. 1967. Sporc and pollen types of the Lower Cretaceous in Patagonia. Rev. /*«/>’«. 1, 21 1-217, pis. 1-2. BALTES, N. 1967. The microflora of the Albian ‘Green Sands’ in the Moesic platform (Rumania). Ibid. 5, 183-197, pis. 1-4. G. PLAYFORD; PALYNOLOGY OF SWAN RIVER STRATA 563 BOLKHOViTiNA, N. A. 1956. Atlas of sporcs and pollen in Jurassic and Lower Cretaceous deposits of the Vilyui Basin. Trudy geo! . hist. Mask. 2, 185 pp., 25 pis. [In Russian.] 1959. Spore-pollen complexes of Mesozoic deposits in the Vilyui Basin and their stratigraphic significance. Ibid. 24, 185 pp., 8 pis. [In Russian.] 1961. Fossil and Recent spores in the Schizaeaceae. Ibid. 40, 176 pp., 41 pis. [In Russian.] 1966. The fossil spores of the ferns of the family Gleicheniaceae (taxonomy and distribution). In The importance of palynological analysis for the stratigraphic and palaeofloristic investigations, 65-75. Moscow. [In Russian.] 1968. The spores of the family Gleicheniaceae ferns and their importance for the stratigraphy. Trudy geol. Inst. Mask. 186, 116 pp., 16 pis. [In Russian.] BRELiE, G. VON DER 1964. Fine unterkretazische Mikroflora aus dem nordlichen Saueiiand. Fortschr. Geol. Rheinld. Westf 12, 117-168, pis. 1-13. BRENNER, G. j. 1963. The spores and pollen of the Potomac Group of Maryland. Bull. Md. Dep. Geol. Mines, 27, 215 pp., 43 pis. BURGER, D. 1966. Palynology of uppermost Jurassic and lowermost Cretaceous strata in the eastern Netherlands. Leid. geol. Meded. 35, 209-276, pis. 1-39. CARRiGY, M. A. 1959. Geology of the McMurray Formation; Pt. 3, General geology of the McMurray area. Mem. Res. Conn. Alberta, 1, 130 pp. CHLONOVA, A. F. 1969. Spores and pollen characteristic of Cretaceous deposits of Zeya-Bureya depres- sion. Trudy Inst. Geol. Geofiz. sib. Otd. 91, 5-66, pis. 1-13. [In Russian,] COOKSON, I. c. and eisenack, a. 1962. Some Cretaceous and Tertiary microfossils from Western Australia. Froc. R. Soc. Viet. 75, 269-273, pi. 37. couPER, R. A. 1958. British Mesozoic microspores and pollen grains, a systematic and stratigraphic study. Palaeontographica, 103B, 75-179, pis. 15-31. DELCOURT, A. F., DETTMANN, M. E. and HUGHES, N. F. 1963. Revision of some Lower Cretaceous micro- spores from Belgium. Palaeontology, 6, 282-292, pis. 42^5. and SPRUMONT, g. 1955. Les spores et les grains de pollen du Wealdien du Hainaut. Mem. Soc. belg. Geol. (n.s.), 5, 1-73, pis. 1-4. DETTMANN, M. E. 1963. Upper Mesozoic microfloras from south-eastern Australia. Proc. R. Soc. Viet. 77, 1-148, pis. 1-27. and PLAYFORD, G. 1968. Taxonomy of some Cretaceous spores and pollen grains from eastern Australia. Ibid. 81, 69-93, pis. 6-8. 1969. Palynology of the Australian Cretaceous; a review. In Stratigraphy and palaeontology. Essays in honour of Dorothy Hill (K. S. W. Campbell, ed.), 174-210, pis. 11-13. Canberra. DORiNG, H. 1965. Die sporenpaliiontologische Gliederung des Wealden in Westmecklenburg (Struktur Werle). Geologic, Jahrg. 14, Beih. 47, 1-118, pis. 1-23. 1966. Sporenstratigraphischer Vergleich zwischen dem Wealden Norddeutschlands und Stiden- glands. Ibid., Jahrg. 15, Beih. 55, 102-129, pis. 1-7. DOYLE, J. A. 1969. Cretaceous angiosperm pollen of the Atlantic Coastal Plain and its evolutionary significance. J. Arnold Arbor. 50, 1-35. FRADKINA, A. F. 1967. Mcsozoic spore-polleii complexes of western Yakut. Leningrad. [In Russian.] and KISELEVA, a. V. 1964. New palynological data on the precise boundary between the Lower and Upper Cretaceous in central districts of Yakut. Uchen. Zap. nauchno-issled. Inst. Geol. Arkt. (Paleont. Biostratigr.), 4, 65-81. [In Russian.] GLAiSTER, R. p. 1959. Lower Cretaceous of southern Alberta and adjoining areas. Bull. Am. /Iss. Petrol. Geol. 43, 590-640. GROOT, J. J. and groot, c. r. 1962. Plant microfossils from Aptian, Albian, and Cenomanian deposits of Portugal. Commun(des Servs. geol. Port. 46, 133-175, pis. 1-10. HALL, J. w. 1963. Megaspores and other fossils in the Dakota Formation (Cenomanian) of Iowa (U.S.A.). Pollen Spores, 5, 425-443. HAMMEN, T. VAN DER 1956. A palyiiological systematic nomenclature. Boln geol., Bogota, 4, 63-101, 1-12. and WYMSTRA, t. a. 1964. A palynological study on the Tertiary and Upper Cretaceous of British Guiana. Leid. geol. Meded. 30, 183-241, pis. 1-3. 564 PALAEONTOLOGY, VOLUME 14 HEDLUND, R. w. 1966. Palynology of the Red Branch Member of the Woodbine Formation (Ceno- manian), Bryan County, Oklahoma. Bull. Okla. geol. Surv. 112, 1-69, pis. 1-10. and NORRIS, G. 1968. Spores and pollen grains from Fredericksburgian (Albian) strata, Marshall County, Oklahoma. Pollen Spores, 10, 129-159, pis. 1-9. HUGHES, N. F. 1958. Palaeontological evidence for the age of the English Wealden. Geol. Mag. 95, 41-49. 1961. Fossil evidence and angiosperm ancestry. Sci. Prog. London, 49, 84-102. • and MOODY-STUART, J. c. 1967. Palynological facies and correlation in the English Wealden. Rev. Palaeobot. Palyn. 1, 259-268. JARDINE, s. and MAGLOiRE, L. 1965. Palynologie et stratigraphie du Cretace des bassins du Senegal et de Cote d’Ivoire. Mem. Bur. Rech. Geol. Min. 32, 187-245, pis. 1-11. KEMP, E. M. 1968. Probable angiosperm pollen from British Barremian to Albian strata. Palaeontology, 11, 421-434, pis. 79-81. KLAUS, w. 1960. Sporen der karnischen Stufe der ostalpinen Trias. Jh.geol. Bundesanst. Wien, Sonderbd. 5, 107-183, pis. 28-38. KOTOVA, 1. z. 1968. Cretaceous spores of the group H Hates. Trudy geol. Inst. Mosk. 191, 17-30, pis. 1-6. [In Russian.] LEVET-CARETTE, J. 1966. Microflore wealdienne provenant d’un puits naturel a la fosse Vieux-Conde (groupe de Valenciennes). Ann. Soc. geol. N. 86, 153-176, pis. 14-16. LORANGER, D. M. 1951. Uscful Blairmore microfossil zone in central and southern Alberta, Canada. Bull. Am. Petrol. Geol. 35, 2348-2367, pis. 1-3. MELLON, G. B. 1967. Stratigraphy and petrology of the Lower Cretaceous Blairmore and Mannville Groups, Alberta Foothills and Plains. Bull. Res. Conn. Alberta, 21, 270 pp. ■ and WALL, J. H. 1963. Correlation of the Blairmore Group and equivalent strata. Bull. Can. Petrol. Geol. 11, 396-409. MTCHEDLiSHViLi, N. D. and SAMOiLOViCH, s. R. 1960. New species of angiosperms. In New species of plants and invertebrates of U.S.S.R., part 1. Moscow. [In Russian.] 1962. Common elements in the Mesozoic and Cenozoic floras of western Siberia and Australia. In Reports of Soviet palynologists for 1st International Conference on Palynology (Tucson, U.S.A.), 94-104. Moscow. [In Russian.] NAGY, E. 1968. Moss spores in Hungarian Neogene strata. Acta bot. hung. 14, 113-132, pis. 1-8. 1969. Palynological elaborations of the Miocene layers of the Mecsek Mountains. Magy. allami fbldt. Intez. Evk. 52, 235-649, pis. 1-56. NORRIS, G. 1967. Spores and pollen from the Lower Colorado Group (Albian-? Cenomanian) of central Alberta. Palaeontographica, 120B, 72-115, pis. 10-18. 1968. Plant microfossils from the Hawks Crag Breccia, south-west Nelson, New Zealand. N.Z. Jl Geol. Geophys. 11,31 2-344. PIERCE, R. L. 1961. Lower Upper Cretaceous plant microfossils from Minnesota. Bull. Minn. Geol. Surv. 42, 86 pp., 3 pis. PLAYFORD, G. 1965. Plant microfossils from Triassic sediments near Poatina, Tasmania. J. geol. Soc. Aust. 12, 173-210, pis. 6-11. and HELBY, R. 1968. Spores from a Carboniferous section in the Hunter Valley, New South Wales. J.geol. Soc. Aust. 15, 103-119, pis. 9-11. PococK, s. A. J. 1961. The microspore genus Cingulatisporites Thomson, 1953. J. Paleont. 35, 1234- 1236. 1962. Microfloral analysis and age determination of strata at the Jurassic-Cretaceous boundary in the western Canada plains. Palaeontographica, lllB, 1-95, pis. 1-15. 1965. Pollen and spores of the Chlamydospermidae and Schizaeaceae from Upper Mannville strata of the Saskatoon area of Saskatchewan. Grana palynol. 5, 129-209, pis. 1-7. 1970. Palynology of the Jurassic sediments of Western Canada, Part 1. Palaeontographica, 130B, 12-136, pis. 5-26. POTONiE, R. 1956. Synopsis der Gattungen der Sporae dispersae. I. Teil: Sporites. Beih. geol. Jb. 23, 1-103, pis. 1-11. 1962. Synopsis der Sporae in situ. Ibid. 52, 1-204, pis. 1-19. 1966. Synopsis der Gattungen der Sporae dispersae. IV. Teil: Nachtrage zu alien Gruppen (Turmae). Ibid. 72, 1-244, pis. 1-15. G. PLAYFORD: PALYNOLOGY OF SWAN RIVER STRATA 565 POTOINE, R. 1967. Versuch der Einordnung der fossilen Sporae dispersae in das phylogenetische System der Pflanzenfamilien. I. Teil: Thallophyta bis Gnetales. II. Teil: Angiospermae. ForschBer. Landes NRhein-Westf 1761, 310 pp., 20 pis. PRICE, L. L. 1963. Lower Cretaceous rocks of southeastern Saskatchewan. Pap. Geol. Siirv. Can. 62-29, 55 pp. RUDKIN, R. A. 1964. Lower Cretaceous. In Geological history of western Canada, 156-168. Calgary. SAMOILOVICH, s. R. et al. 1961. Pollen and spores of western Siberia; Jurassic to Paleocene. Trudy vses. neft. nauchno-issled. geol.-razv. Inst. 177, 659 pp., 84+65 pis. [In Russian.] SINGH, c. 1964. Microflora of the Lower Cretaceous Mannville Group, east -central Alberta. Bull. Res. Conn. Alberta, 15, 238 pp., 29 pis. 1971. Lower Cretaceous microfloras of the Peace River area, northwestern Alberta. Ibid. 28, 310 pp., 38 pis. SMITH, A. H. V. and butterworth, m. a. 1967. Miospores in the coal seams of the Carboniferous of Great Britain. Spec. Pap. Palaeont. Assoc. 1, 324 pp., 27 pis. SRivASTAVA, s. K. 1966. Upper Cretaceous microflora (Maestrichtian) from Scollard, Alberta, Canada. Pollen Spores, 8, 497-552, pis. 1-11. STEEVES, M. w. and WILKINS, L. R. 1967. Spores and pollen from the Lower Cretaceous of Saskatchewan, Canada. Part I. Sporites. Can. J. Bot. 45, 2329-2365, pis. 1-8. STELCK, c. R. 1958. Stratigraphic position of the Viking Sand. J. Alberta Soc. Petrol. Geol. 6, 2-7. STOTT, D. F. 1968. Lower Cretaceous Bullhead and Fort St. John Groups, between Smoky and Peace Rivers, Rocky Mountain Foothills, Alberta and British Columbia. Bull. Geol. Surv. Can. 152, 279 pp. STOVER, L. E. 1964. Comparison of three Cretaceous spore-pollen assemblages from Maryland and England. Spec. Pubis Soc. econ. Palaeont. Miner., Tulsa, 11, 143-152, pis. 1-2. THOMSON, p. w. and pflug, h. d. 1953. Pollen und Sporen des mitteleuropaischen Tertiars. Palaeonto- graphica, 94B, 1-138, pis. 1-15. VAGVOLGYI, A. and HILLS, L. V. 1969. Microflora of the Lower Cretaceous McMurray Formation, northeast Alberta. Bull. Can. Petrol. Geol. 17, 154-181, pis. 1-8. VERBITSKAYA, z. I. 1962. Palynological evidence and stratigraphic subdivision of Cretaceous deposits of the Suchansky coal basin. Trudy Lab. Geol. Uglya, 15, 1-165, pis. 1-24. [In Russian.] 1 966. On the question of the relations of the Early Cretaceous floras of the southern part of the maritime region with the floras of the other regions of the earth (according to palaeopalynological data). In Methods of palaeopalynological research. Materials for 2nd International Conference on Palynology (Holland, September 1966), 139-146. Leningrad. [In Russian.] wiCKENDEN, R. T. D. 1945. Mesozoic stratigraphy of the eastern plains, Manitoba and Saskatchewan. Mem. geol. Surv. Brch. Can. 239, 87 pp., 5 pis. WILLIAMS, G. D. 1963. The Mannville Group (Lower Cretaceous) of central Alberta. Bull. Can. Petrol. Geol. 11, 350-368, pis. 1-2. Typescript received 19 November 1970 GEOFFREY PLAYFORD Department of Geology and Mineralogy University of Queensland St. Lucia, Brisbane Queensland 4067 Australia C 8385 pp LIOPLEURODON ROSSICUS (NOVOZHILOV)— A PLIOSAUR FROM THE LOWER VOLGIAN OF THE MOSCOW BASIN by L. BEVERLY HALSTEAD Abstract. The species Pliosauriis rossicits Novozhilov is redescribed. The short mandibular symphysis places the species in the Liopleiirodon-Stretosaiirus group and the triradiate scapula in the genus Liopleurodon itself. The teeth are trihedral in section; hence the material, being specifically distinct from L. ferox, is assigned to L. rossiciis. In 1948 Novozhilov described two pliosaurs from the Lower Volgian of the Moscow Basin which he referred to the new species Pliosawus rossicus and Peloneiisles irgisensis. In the review of Upper Jurassic pliosaurs by Tarlo (1960), it was not possible for me to discuss these species adequately nor assign them to any particular place in the scheme proposed for the other pliosaurs. It was suggested that these two species should be provisionally retained until further description was forthcoming — in particular details of the mandibular symphysis and cervical vertebrae. The reference by Novozhilov of one form to Peloneustes was probably due to the false impression culled from the literature that Pliosaurus had a short mandibular symphysis. Subsequently Novozhilov (1964) figured the skull and pectoral girdle of Pliosaurus rossicus and placed the other species in the new monotypic genus Strongylokroptaphus. During my visit to Moscow in 1961 I had an opportunity to examine the type material of P. rossicus, thanks to the generous help of Professor C. C. Flerov and Dr. L. P. Tatarinov. Since there is no description available of those parts of the skeleton, which are of prime taxonomic importance, I propose to rectify this omission. SYSTEMATIC PALAEONTOLOGY Family pliosauridae Seeley 1874 Genus liopleurodon Sauvage Type species. Liopleurodon ferox Sauvage. Diagnosis. Mandible with short symphysis bearing 5-7 pairs of large caniniform teeth; total of 25-28 teeth in each ramus. Teeth circular in section; trihedral in Kimmeridgian and ‘Tithonian’ forms. Cervical vertebrae short, length half or less than half width or height. Scapula triradiate with dorsal process directed laterally and ventral plate expanding medially. Epipodials long; short in Kimmeridgian and ‘Tithonian’ forms. Liopleurodon rossicus (Novozhilov) Pliosaurus 1 grandis Owen; Rozhdestvenski 1947, pp. 197-199, text-figs. 1, 2. Pliosaurus rossicus Novozhilov 1948, p. 115, fig. la. Pliosaurus rossicus Novozhilov; Tarlo 1960, p. 174. Pliosaurus rossicus Novozhilov; Novozhilov 1964, p. 331, text-fig. 329. [Palaeontology, Vol. 14, Part 4, 1971, pp. 566-570.1 L. B. HALSTEAD: LIOPLEURODON ROSSICUS 567 Diagnosis. Teeth trihedral in section, ornamentation of coarse well-spaced ridges on lingual surface, flat and smooth buccally. Epipodials short. Holotype. Associated skeleton no. 304/1, 1938, housed in the Palaeontological Institute of the Academy of Sciences, Moscow. Description of Holotype: Skull. The skull was described in detail by Novozhilov (1948, 1964). He noted that the teeth were trihedral in section with the external side smooth as in PHosaurus grandis Owen (= Stretosaurus inacromerus Phillips). This tooth form appears to characterize all Kimmeridgian and ‘Tithonian’ pliosaurs (Tarlo 1960); an example is here figured (text-fig. 1). The mandibular symphysis is short, containing only 6 pairs of teeth. The length is 280 mm, width at the fifth tooth 1 50 mm, and maximum depth (at third tooth) 100 mm. The two anterior pairs of teeth are small but the following three pairs have long crowns with heights of 120 mm 3rd tooth, 125 mm 4th, and over 130 mm 5th tooth. Text-fig. 2 is a sketch of the mandibular symphysis in ventrolateral view. TEXT-FIG. 1. Tooth, a. Occlusal view to show trihedral section, b. linguo-distal view to show ornament of coarse radial ridges, c. bucco-medial view to show smooth flat external surface of crown. X TEXT-FIG. 2. Mandibular symphysis in ventro-lateral view, x Vertebral column. A number of cervical vertebrae are preserved and measurements, in mm, of a selected number are given in the following table: Cervical vertebrae Length Height Width at articular surface Number of rib facets 36 90 93 2:2 37 90 96 2: ? 38 95 108 2:2 39 99 104 2:2 40 90 100 2:2 43 99 no 1:1 45 102 112 1:1 47 98 116 1:1 Pectoral vertebrae 568 PALAEONTOLOGY, VOLUME 14 The cervical vertebrae and the pectorals — the last three in the table — are much shorter proportionately than those of the contemporary Strelosaurus, but this difference is undoubtedly due to the difference in age of the individuals and hence size. The rib facets seem to be carried higher up the lateral surface of the centra and the single facets of the pectorals have their major axis on the vertical transverse plane, contrasting with Strelosaurus where they are aligned obliquely (see Tarlo \959b). The ventral surface of the centrum is devoid of any keel, but instead has a characteristic rugosity reminiscent of that of Pliosaurus brachyspondyhis Owen (see Tarlo 19596f). This is illustrated (text- fig. 3) together with lateral and anterior views of a pectoral vertebra and a lateral view of rib facets of an anterior cervical. In some of the vertebrae there is some evidence of osteophytosis or lipping on the ventral margins of the centra, which suggests that the individual in question was somewhat arthritic. TEXT-FIG. 3. Vertebral centra, a. cervical in anterior view, b. pectoral in lateral view to show single rib facet, c. cervical in ventral view to show characteristic rugosity, d. cervical in lateral view to show double rib facet, x I. Pectoral girdle and forelimb. The coracoid and scapula are preserved. The right scapula (text-fig. 4) has a wide dorsal process which is set off from the expanded ventral plate by a small angle forming a ridge on the lateral surface of the bone. The proportions of the dorsal process relative to the ventral plate are very different from those of the scapula of Pliosaurus sp. figured by Tarlo (1958). Here again the differences in proportions are L. B. HALSTEAD: LIOPLEURODON ROSSICUS 569 simply due to the size of the individual, and hence its age. The humerus is of normal proportion and its measurements in mm are given in the table below: Head Mid-shaft Distal end Total width thickness width thickness width thickness length Humerus 115 130 90 73 200 64 455 Femur 165 160 135 75 240 75 590 The epipodials are short as in all post-Oxfordian pliosaurs. TEXT-FIG. 4. Right scapula in ventral view showing dorsal set off from ventral plate by ridge. Pelvic girdle and hind limb. The ilium and ischium are preserved as well as the femur and epipodials. The measurements of the femur are given in the above table from which it can be seen that this bone is markedly longer than the humerus but proportionately is more slender, again contrasting with the situation in the giant Stretosaurus (Tarlo \959b) The epipodials are short and the girdle bones are comparable to those of other pliosaurs. Description of ‘P. ? grandis'. This specimen consists of a single articulated hind limb and the tip of the snout. The limb is perfectly normal, the femur measuring 820 mm in length. The mandibular symphysis is 450 mm in length but the anterior part is broken, so that in all probability it would have been about 500 mm. There are six pairs of teeth in the symphysis. The 3rd and 5th teeth have a diameter of 60 mm, also they are trihedral in cross-section with a smooth flat outer surface of the crown. The maximum depth of the symphysis is 180 mm and the diameter of the entire snout at the posterior end of the symphysis is 450 mm in width and 500 mm in height. 570 PALAEONTOLOGY, VOLUME 14 DISCUSSION From the above descriptions it is evident that the only differences between the two animals are quantitative. The P. 'grcmdis' specimen belonged to an individual almost twice the size of theP. rossicus specimen. The femora cannot be distinguished. The nature of the mandibular symphyses and the teeth are identical. The teeth merely confirm that the two specimens are of post-Oxfordian age. On the other hand the short mandibular symphysis is characteristic of the Liopleurodon-Stretosaurus group of pliosaurs. Pre- viously, the only post-Oxfordian species of this group known was Stretosaurus macro- merus, which was characterized by its unusual scapula. In all other respects it was similar to Lioplewodon ferox. The Russian material under discussion has the same type of scapula as is found in all pliosaurs except Stretosaurus', the short mandibular sym- physis, however, distinguishes it from PUosaurus itself and thus the species can be referred to the genus Liopleurodon. L. rossicus differs from the two other species of the genus, L. ferox and L. pachydeirus, in that the teeth are trihedral in cross section. The ridges on the crowns are reminiscent of L. ferox but the flat smooth outer surface clearly separates L. rossicus from them, although of course it does not enable this species to be separated from the other post-Oxfordian pliosaurs. The rugose nature of the ventral surface of the cervical vertebrae is a useful feature, but even here great care must be taken not to confuse it with the similar rugosity of PUosaurus brachyspondylus. The significance of the Russian species is that it confirms that the Liopleurodon line, as well as giving rise to the unusual Stretosaurus, continued without any modification of the pectoral girdle having taken place. Thus the Lower Cretaceous Kronosaurus which has a normal pectoral girdle can be readily derived from the Oxfordian Liopleurodon without the unfortunate gap in the Kimmeridgian and Tithonian which existed until the Russian form was discovered. Acknowledgements. Thanks are due to Professor C. C. Flerov and Dr. L. P. Tatarinov for their generous assistance. The visit to Mos:ow was made possible by a grant from the Department of Scientific and Industrial Research during the tenure of a Senior Fellowship. The text-figures were prepared by Mr. John Smith. Dr. A. J. Charig kindly translated Novozhilov’s papers. REFERENCES NOVOZHILOV, N. I. 1948. Two new pliosaurs from the Lower Volga beds Povolzhe (right bank of Volga). Dokl. Acad. Naiik SSSR, Moscow, 60, 1 15-1 18. [In Russian.] 1964. Order Sauropterygia. Osnovy Paleontologii, 12, 309-332. [In Russian.] ROZHDESTVENSKi, A. K. 1947. New discovery of giant Pliosanrns in Zavolzhe (left bank of Volga). Dokl. Acad. Nauk SSSR, Moscow, 56, 197-199. [In Russian.] TARLO, L. B. 1958. The scapula of Pliosanrns tnacromenis Phillips. Palaeontology, London, 1, 193-199, pis. 36, 37. 1959a. Pliosanrns brachyspondvlns (Owen) from the Kimmeridge Clay. Palaeontology, London, 1, 283-291, pis. 51, 52. \959b. Stretosaurus gen. nov. a giant pliosaur from the Kimmeridge Clay. Palaeontology, London, 2, 39-55, pis. 7-9. 1960. A review of Upper Jurassic pliosaurs. Bull. Brit. Mas. Nat. Hist. (GeoL), 4, 145-189. L. BEVERLY HALSTEAD Department of Geology University of Reading Reading, Berkshire Final typescript received 16 March 1971 PERIODICITY STRUCTURES IN THE BIVALVE SHELL: EXPERIMENTS TO ESTABLISH GROWTH CONTROLS IN CERASTODERMA EDULE FROM THE THAMES ESTUARY by GEORGE E. FARROW Abstract. Growth patterns of living cockles from one year-group were analysed from a wide range of substrates occurring at different heights on the tidal flats. At similar population densities and tidal heights cockles from well-sorted sand grow larger and show more disturbance rings than examples from mud. Degree of tidal cover affects size only in the first year, but internally, erratically fluctuating microbanding characterizes high shore shells. In dense populations daily increment patterns suggest a much shorter growing season, growth diminution being greater on the higher parts of the flats. In any one population of cockles in their third year microstructural banding reveals ‘hardy’ and ‘sensitive’ groups whose total number of growing days varied from 795 to 650. Mean monthly sea temperature does not correlate well with mean monthly shell growth, though internal patterns of growth may be related to air temperature variations. In susceptible individuals winter rings coincide with frosts over the cockle-banks: cockles are apparently more susceptible in their first winter, and may stop growing for several periods of up to a fortnight. A warning is given against the casual use of growth banding in fossils for providing geophysical data on the history of the Earth’s rotation rate. T HE current interest in fossil growth lines shown by geologists and geophysicists stems from the demonstration by Wells (1963) of daily increments in the epithecae of corals. However, the recognition of daily increments in corals is governed by exceptional pre- servation of the epitheca, and since corals are abundant at only a limited number of horizons in the geological column, other organisms of more widespread occurrence whose internal banding is not seriously affected by abrasion were sought. Barker (1964) sliced several Recent bivalve shells and suggested that some periodic markings were probably of daily origin, although experimental data were not presented to support the hypothesis. The suggestion had in fact already been made as a result of observations under natural conditions by Davenport (1938, p. 514) for scallops kept in a tide box, and by Hopner Petersen (1958, p. 14) for Cerastodenna lamarcki (Reeve). Ten years later, almost simultaneously, three works appeared which provided proof of the daily origin of bivalve microbanding and demonstrated its widespread occurrence in the group as a whole. Clark (1968) kept twelve juvenile Peclen diegensis for 51 days in tanks at the Kerckhoff Marine Laboratory, California, and stressed the importance in a population of noting the maximum not the average line count. House and Farrow (1968) showed that plots of growth increments in C. edide from South Wales indicated a correlation between tidal phases and bunching of the microbands such that a daily origin was beyond question. Pannella and MacClintock (1968) planted notched Venus (Mercenaria) mercenaria Linnaeus in Barnstable Harbour, Massachusetts, both inter- tidally and subtidally, killing selected specimens after one and two years. Shells showed the same number of growth bands as there were days between notching and killing, and also showed tidally controlled variation in the type and thickness of growth bands. (Palaeontology, Vol. 14, Part 4, 1971, pp. 571-588, pi. 108.] 572 PALAEONTOLOGY, VOLUME 14 Similar banding, also thought to be diurnal in origin, was illustrated for Glycymeris and Tridacna {Chametrachea) squamosa Lamarck. Daily growth lines in present-day bivalves have since been recognized in many other groups. The Veneracea and Cardiacea, because of favourable shell structure, have very well preserved banding, but preservation is not favoured by the shell structure of some groups, such as the Tellinacea, and the identification of banding in such common intertidal forms as Scrobicularia plana (da Costa) and Macoma balthica (Linnaeus) is extremely difficult. In the Pectinidae, periodicity structures are extremely well shown on the surface but cannot be studied by preparing thin section or peels as the foliated shell structure prevents their recognition. Likewise the larval stages of the oysters Ostrea ediilis Linnaeus, O. lutraria Hutton, and O. chilensis Philippi show well-developed exterior diurnal growth layers (Millar 1968) though these become difficult to analyse either on the highly irregular surfaces of adult oysters or internally, owing to the foliated and vacuolar shell structure. HCtGHT ABOV£ LWST 0-2 mM. r-. m mm TEXT-FIG. 1. Distribution map showing sampling localities along the north bank of the Thames Estuary. Data on cockle population density (for flats west of Mulberry Harbour only) from Mr. G. Pickett, Fisheries Laboratory, Burnham-on-Crouch. Extinct groups of fossils also show banding homologous to the daily structures present in Recent groups. Among fossil forms, Pannella, MacClintock, and Thompson (1968) have used Eocene Crassatella and Cardita, Cretaceous Limopsis, Cucullaea and Tan- credia, Triassic Cardita and Carboniferous Conocardiwn. Among the Bivalvia, perhaps the most outstanding are the radiolitid rudists whose banding indicates very rapid growth, and at some reef horizons extremely severe tidal control. TECHNIQUES AND DATA PROVIDED Eighteen samples were taken from intertidal mud- and sand-flats exposed between Southend Pier and the Maplin Sands, a distance of seven kilometres. One group of samples was collected on 24 July 1965, other groups at intervals throughout August G. E. FARROW: PERIODICITY STRUCTURES IN CERASTODERMA 573 1967. All specimens belong to the 1963 year-group which dominated the flats until the widespread spatfall of 1967. For each sample the nature of the substrate, cockle population density, presence and abundance of other organisms, and position with respect to low water springs were noted. Approximately thirty-five specimens were collected per sample, and these were killed by boiling for twenty minutes on the evening of collection; paired valves were retained and numbered. Daily records of sea tempera- ture, air maximum and minimum temperature, sunshine and tidal range were obtained for the period 1963-1967 for comparison with the growth records preserved in the micro-structural banding of the cockles. External growth-ring analysis. A summary of the macroscopic growth increments of all shells studied in the Southend and Shoebury populations (text-fig. 1) is shown in Table 1. Measurements were made with dividers, using height along the maximum radius from the umbo as the index of growth rather than length, since this is the direction along which peels were prepared; observations were made to the nearest half millimetre below. TABLE 1. Macroscopic annual increments of Cerastoderma edide population (1963 year-group) from the Southend Flats and Shoebury Sands, Thames Estuary: measurements in millimetres expressed as x±q:, where a = cr/Vn (Hancock 1965). * Populations with densities of more than 500 cockles/m- B and D = 1800; J = 500 (1963 year-group), 1000 (1967 spatfall); Q == 1000. Localities shown on text-fig. 1. Popula- tion n Substrate A 32 Dry sand B* 55 Wet, muddy sand C 64 Fine sand D* 89 Wet mud F 30 Fetid mud G 34 Rippled sandy mud H 35 with Arenicola Mud J* 40 Muddy sand L 14 Soft mud M 30 Dry hard mud N 28 Soft mud O 35 Fine sand, edge of P 29 channel Thixotropic sandy Q* 27 mud Soft mud with algal R 16 plastering Soft mud Height above L.W.S.T. (feet) 1963 1964 + 4 9-6±0-29 16-7 + 0-48 + 3 80±0-15 9-5+0-16 + 3 8-5±0-14 14-2 + 0-24 -j- 3 8-3±011 9-6+0-14 + 5 7-3±0-21 12-3+0-22 + 3 7-4±0-22 12-9+0-26 + 5 7-7±0-16 12-2 + 0-34 0 8-2±0-19 11-8+0-27 -t-1 7-7±018 13-9+0-35 +4 7-6±0-17 13-5+0-21 + 6 7-5±0-23 13-4+0-25 + 6 7-5±0-17 13-5+0-21 + 6 7-3 + 0-24 12-9+0-22 + 6 7-3+0-16 11-3+0-20 + 7 6-9+0-25 14-4+0-30 1965 1966 Total (1963- 1966) — — — z z z 4-9+0-16 5-8±0-25 30-3 5-2+0-18 5-9 + 0-23 31-4 4-8+0-13 5-3-1 0-21 30-0 5-5 + 0-15 5-4+0-18 30-9 6-5+0-33 6-0+0-32 34-1 6-4+0-18 6-1+0-16 33-6 6-0+0-20 5-6 + 0-23 32-5 6-2 + 0-16 6-1+0-15 33-3 5-8 + 0-26 6-1+0-24 32- 1 4-2+0-12 5-1+0-20 27-9 5-4+0-13 5-6+0-25 32-3 Internal growth-ring analysis. Acetate peels of shell surfaces were prepared by the following method. After cleaning the shells by boiling in hydrogen peroxide, cockles were sliced radially from umbo to ventral margin along the longest rib using a rubber bonded 240 grade carborundum cutting wheel 0-010" in thickness and 6" in diameter. Care was necessary at this stage to prevent slicing obliquely to the direction of growth and so crossing ribs, as this exaggerated the reflexed growth banding and rendered daily increments difficult to measure (PI. 108, fig. 2). The sliced valve was then polished with 574 PALAEONTOLOGY, VOLUME 14 successively finer carborundum powders up to 1200 grade, and finished on a polishing wheel, using 5/20 fast-cutting alumina at 700 revs/min. Etching with 10°/, HCl for 10 seconds was sufficient to enhance the details of the growth banding without over-strengthening the crossed lamellar structure, which with over-etching tends to obliterate tangential growth increments. Once the etched shell was dry its surface was wetted with acetone and pressed on to 1/1500 acetate paper. The resulting peel was mounted between glass slides and examined under a Zeiss Ultraphot microscope. Photographic traverses with a final magnification in the region of X 160 were prepared for the whole length of each peel. The increments shown on the traverses were measured along the surface of maximum growth (Pannella and MacClintock 1968, fig. 1) and plotted on calendar graph paper for comparison with instrumental records. Viewed by transmitted light, individual daily increments of growth in Cerastoderma edule are represented by a doublet consisting of a thin white band of strong relief bordered by dark Becke lines, and a broad grey band relatively depressed (PI. 108, fig. 1). Examination of the ventral extremities of peels prepared from shells killed on a summer evening clearly shows that the broad grey band is that most recently formed, and that the thin band therefore corresponds to the nocturnal element in the shell secreting cycle. Staining of etched shell slices with Rose Bengal demonstrates that the thin night band is rich in organic matter, and also that there is appreciable organic matrix in the day band. At ultrastructure level transmission electron micrographs reveal the presence of projections of organic material stemming from the nocturnal band. Examination of acetate peels prepared from selected cockles serves to pinpoint those seasons of the year when high density exerts the greatest influence on growth, and also enables the shell secretion pattern of cockles from sandy and muddy substrates to be compared. However, it is first necessary to note the degree of variation in microstructural elements revealed by one supposedly homogeneous population. There are two ways in which variation in growth rate between shells either in the same population or in different populations can be assessed microscopically. The first is somewhat generalized; the spacing in millimetres of units of thirty daily bands is plotted as a monthly total. Much detail is lost with this method, although it is rapid and can be carried out using a precision slide carrier with vernier X and Y scales. The second method, adopted by House and Farrow (1968), involves plotting each diurnal increment on calendar graph paper. This ultrasensitive growth trace makes possible the correlation of small-scale fluctuations in growth with other diurnally recorded environmental phenomena. ENVIRONMENTAL SUSCEPTIBILITY OF INDIVIDUAL COCKLES Since the yearly increments of cockles from sample A show the greatest standard deviation (Table 1) peels were prepared of all shells to ascertain the variation shown in the daily banding pattern; Table 2 gives the number of daily bands counted between annual rings on each specimen. For the number of days’ growth during 1965 up to the collection date in midsummer the values quoted are probably accurate to within ±2 or EXPLANATION OF PLATE 108 Figs. 1, 2. Periodicity structures in Cerastoderma edule. 1. Rapidly secreted shell during the autumn following settlement, indicated by large daily increments; note the thin outer shell layer. Xl20. 2. Slower daily growth pattern during the second autumn, producing marked thickening of outer shell layer; note the effect of oblique slicing. X 120. Palaeontology, Vol. 14 PLATE 108 FARROW, Growth banding in Cerastoderma SELLWOOD, Tlialassinoides with Glyphaea IS r V 'ii! G. E. FARROW: PERIODICITY STRUCTURES IN CERASTODERMA 575 3 days. In shells with pronounced black winter rings crossing the peel, estimates of the total age of the animal would be in error due to a stoppage in growth. It is possible, however, to obtain an indication of absolute age from those shells which do not show such black rings. In the sample of 33 there were five such shells which indicate spat originating in 1963 at intervals between 21 May (795 days) and 4 August (720 days). Shells with indications of slight stoppage approach these dates by a matter of a few days or fall within the range, though a group of six highly sensitive individuals falls well outside this range, their total number of growth days amounting to only 650. Even if TABLE 2. Variation in number of days growth per year in Cerastoderma edule (population A) from sandbank east of Southend Pier, as measured by daily growth bands. Specimens collected 24 July 1965 (1963 year-group). Figures in brackets are external height in mm. Specimen Number 1965 1964 1963 TOTAL 1 205 (4t) — — — 2 213 (8) — — — 3 235 (8)t 360 (16) — — 4 235 (4)t 360 (22) — — 5 210 (6)t 300 (14i)X 135 (10) 645 (30j] 6 165 {4)X 315 (15i)X — — 7 107* (4)t 354 (18) — — 8 205 (7) 310 (75)t — — 9 198 (5)1 405 (17)1; 185 (14) 788 (36) 10 210 (5i-)t 382 (18-|)t 163 (14) 755 (38) 11 200 (6i) — — — 12 226J 363 — — 13 210 (5J)t 425 (19)t 160 (11) 795 (354; 14 195 (7)t 340 (19) — — 15 230 (6i)t 405 (18)t 142 (9) 777 (334; 16 207 (6^)1 363 (14)t — — 18 200 (6)t 370 (161)1: — — 19 225 (7i-)t 360 (191) — — 20 210 (8)t 363 (15)t 147 (91) 720 (32i; 21 207 (8)t 383 (131)1: 165 (91) 755 (31) 22 204 (5)t 366 (191)t 140 (91) 710 (34) 23 205 (54)t 360 (181)1; — — 24 220 (8)t 396 (15)t — — 25 209 (5J)t 372 (16)t 145 (91) 726 (31) 26 207 (4i)X 307 (14)X 134 (8D 648 (27) 27 228 (6)t 385 (151)t — — 28 220 (6)j 283 (13)1; 152 (10) 655 (29) 29 213 (44)t 407 (20l)t 140 (11) 760 (36) 30 185 (5*)| 343 (14)1; — — 31 210 (54)t 420 (184)t 146 (10) 776 (34) 23 170(7i)X 300 (14i)X — — 34 214(6i)t 375 (19)1; 139 (81) 728 (34) 35 236 (5m 370 (15) 167 (121) 773 (33) * Pathological specimen. Sensitive individuals in italics, t No indication of growth cessation Hardy individuals in bold type, during winter. t Sharply defined winter ring. 576 PALAEONTOLOGY, VOLUME 14 these shells started life on the latest date indicated by continuously growing shells they must have stopped growing for at least one month in both winters, perhaps nearer two. It should scarcely require stressing that the facile analysis of fossil material by counting the number of supposedly daily growth bands between external annual rings is liable to extreme error; in population A estimates of the number of days in the year would vary from 283 to 425 days (Table 2). Since the palaeontologist is rarely able to select as large a sample as A, the use of this method on individual specimens must be avoided. Daily growth patterns as an indication 0/' hardy' and 'susceptible' individuals. Scrutiny of Table 2 will demonstrate that population A can be split nicely into two end members which represent extremes of susceptibility to external environmental pressures. The suggested grouping is indicated on Table 2, where it can be observed that correlation between individuals from year to year in the two groups is often very close. (There is striking comparison here with the statistical methods of dendrochronologists [e.g. Ferguson 1968, p. 843] who speak of ‘complacent’ and ‘sensitive’ growth traces.) The two groups are most easily separated by their second-year growth (text-fig. 2), where 140 160 ISO 200 260 300 340 380 420 200 220 240 TEXT-FIG. 2. Histogram showing number of days growth per year in cockles from population A. 1963, following summer spatfall (n = 15) 1964, whole year (n =-- 30) 1965, up to collection on July 24 (n = 32) the sensitive individuals plot in isolation from the majority, which shows a sharp mode at 360 to 379 days growth and includes 40% of the population. For the first year of growth immediately following settlement data are scanty because of the difficulty of taking peels from the very thin shell, and also as a result of abrasion of the critical outer shell layer in the umbonal region. Daily banding in one shell, A35, could be counted throughout, but some others were so poorly preserved that no values could be obtained for 1963. In any specimen, however, where more than 100 bands were counted for the first-year growth estimates for the umbonal region were made using A35 as standard. G. E. FARROW; PERIODICITY STRUCTURES IN CERASTODERMA 517 There is a suggestion in Table 2 that the sensitive individuals settled somewhat later in the year than the more hardy group. It is possible in this instance that because of their smaller size they were more prone to environmental fluctuations than the slightly older individuals, but this cannot be proved without additional material. A more likely explanation perhaps is that the factor of susceptibility is under genetic control, and hence independent of the degree of development attained in the initial months of growth. Nevertheless one feature is clear, that at the age of 2 years susceptible cockles are on average 6 mm smaller than hardy individuals from the same environment. ENVIRONMENTAL INELUENCES Substrate. Fifteen of the eighteen samples contained a suflicient proportion of 1963 year-group to be suitable for statistical treatment. Most striking is the high growth rate of specimens from flne, well-sorted sand which dries at low water of ordinary tides. Comparing sample A from sand with samples F and H from fetid mud (text-fig. 3), all of which are situated at similar heights above low water spring tide and have relatively low population densities (30-100/m^), it will be seen that in both 1963 and 1964 growth was 25% greater in the sandbank population. Variation between specimens is also greatest in this population, where 30% of the cockles show pronounced disturbance rings in the autumn of their first year. Despite this, their growth is greater than that of Shell Height TEXT-FIG. 3. Cumulative growth curves for cockles from different habitats in the Thames compared with that from R. Yealm (Orton 1926). Localities shown on text-fig. 1. A: sand, low density L; low density, near low water spring tide R: low density, high shore E and H: mud, low density J: high density, near low water spring tide Q : high density, high shore 578 PALAEONTOLOGY, VOLUME 14 any other cockle population on the north bank of the middle estuary, and this higher growth rate cannot be ascribed to the earlier settling of spat on the sandbank than on the surrounding mudflats, for the daily growth line totals for 1963 are comparable. Population density. To assess the importance of population density, two samples of different density from near low water springs L and J and two from the higher flats R and Q will be considered; these are plotted on text-fig. 3. Dense populations of cockles from near Low Water Spring Tides (J) are on average 3-2 mm smaller at their fourth winter ring than more scattered examples (L), this difference increasing to 44 mm in more elevated populations. The fact that the growth diminution caused by over-crowding is greater on the higher parts of the flats than near low water springs may be attributed to weaker currents and consequent poorer food supply further onshore. The supply is clearly sufficient for low density populations but not when densities of third-year cockles in excess of 1,000/m^ occur. A gelatinous brown algal plastering at the site of sample Q may further have reduced the available food supply by trapping much of the plankton and detritus. In both examples differences in growth are greater between crowded and sparse populations at the same level on the flats than between high- and low-level populations. This, together with the demonstrably large effect of a sandy substratum, effectively swamps any controlling effect of tidal height on the macroscopic growth characteristics of the Thames cockles. Examination of text-fig. 4 enables the effect of population density to be studied both in terms of the maximum daily increments achieved and with respect to the length of the growing season. Growth traces for B9 and J13 are from populations which occur at a density of adult cockles of more than 1 ,000/m^ ; A 1 2, A26, R22 are from cockles occurring at about 30/m^. The most prominent characteristic of the former profiles is their very low late summer and autumn growth in the second year. After maximum daily incre- ments of around 125 pm in early or mid-May (when all the Southend shells were growing most rapidly: cf. Burry Inlet, House and Farrow 1968, text-fig. 2) the amount of shell added per individual in dense populations gradually tails off to 12-15 pm per day: this low level was reached by mid-August in population B, but was almost a month earlier in the low water springs population J. In the following (third)-year growth commenced earlier, and increments were higher than in less dense populations. In the case of the comparison with population A, it seems that substrate differences may be responsible for this, for with their rapid second-year growth shells inhabiting sand had already attained a height of 264 mm by their second winter, whereas population B from the mud nearby had only reached 17-5 mm. By early June 1965, however, the characteristic summer reduction in growth had set in, two months earlier than in the second year. Summarizing, daily growth increment patterns suggest that dense populations of cockles have a much shorter growing season than sparse populations. The absolute daily maxima in first and second years are not, however, appreciably lower; here the effect of height above L.W.S.T. is more important. Tidal cover. Comparison of the annual increments of shells from near low water springs with those from higher on the fiats does not convincingly show a diminution of growth amongst inshore populations (cf. Hancock 1967, p. 141). Differences in growth are more conspicuous along the length of the Estuary (text-fig. 1, Table 1). Thus Thorpe Bay cockles (F, G, H) situated a little lower on the fiats than those around the Boom at G. E. FARROW: PERIODICITY STRUCTURES IN CERASTODERMA 579 — : — « S ^ W) C3 a 00 -rt w ^ 00 C3 C > . -o a § « T3 W C O 03 g G -G u T3 H c G 3 G Q •£ c t> "O O C/3 > G c/3 'o o ]-i 3 a •S '-S 03 Om G G -G o 03 G -G o C c5 TJ o Ui CJ) O G o c/3 d •» X) o^ G G s c O ’5b G O a o t4-H 9 ' ° § •r o '^;§ o c tJ 0> • S ^ c/5 — c C^J C3 3 £ 73 y .5 "" "5 Q S O (yi ^ (j 6 S § E 6 -e H 8 5 u — .23 H "O 580 PALAEONTOLOGY, VOLUME 14 comparable densities (N, O, P, R) are on average 2 mm smaller after their fourth winter. The salinity gradient along the Thames could possibly be of sufficient magnitude to effect such a difference in growth, but it might also be explained by the seaward location of the Boom beds in an area of more complex current activity and hence of richer food supply. The inverse relationship which seems to exist for the Shoebury populations J to R between growth in the first year and height above low water springs (text-fig. 5: cf. Cole 1956, p. 79) does not hold when the Southend Pierhead populations are also considered. This discrepancy may well be explained by differing settlement dates of spatfall in the two regions. No more than a week’s separation of spat settlement is required to account for the difference, since Orton (1926, p. 261) has shown that small individuals may be growing shell at the rate of 1 to 1| mm per week. Shell Height + LWST Height of Station TEXT-FIG. 5. First-year growth (1963) compared with height of station above low water spring tide for Thames cockles. The linear relationship shown by Thorpe Bay and Shoebury populations (F to R: triangles) does not hold for the Southend communities (A to D: dots). In the first year of growth, although the peaks of the diurnal increment profiles for A26 and R22 correspond (text-fig. 4), the more emergent specimen has a lower August and September maximum, but a greater October and November contribution. This extension of the growth season relative to lower shore cockles is seen again in the second year where the profile is strongly bimodal. After an early May maximum (substantially less than other shells) growth was sharply reduced in June and July, but recovered in August. There is evidence in the spring of both 1964 and 1965 that the higher shore cockle was later in starting growth, notably when compared with high density popula- tions. By the time the third year of growth is reached diurnal increment totals are reversed, and the high shore example shows the greatest spring increase. Its extraordinary G. E. FARROW: PERIODICITY STRUCTURES IN CERASTODERMA 581 fluctuations make the curve quite different in character from all the others, and judging by similar amplitude variations amongst Burry cockles to be described in a second paper of this series this seems a particularly promising environmental indicator for fossil molluscs. House and Farrow (1968, text-fig. 3) demonstrated a tidally controlled cyclicity in the growth of Burry cockles, especially in their first year. Southend cockles very rarely show this, though the sensitive low shore example A26 (text-fig. 10) may conceivably do so. Why the higher shore form R22 does not show lunar periodicity seems problematical : it might reasonably have been expected to show more evidence. However, it may be that the sandy substrate of A allows the water to drain completely at low water springs, whereas on the muddy banks interstitial water is retained. TABLE 3. Monthly growth increments for early 1965 for 32 individuals of 1963 year-group Cerasto- derma edule from sandbank population A, Southend, determined by spacing of units of 30 daily growth bands. Measurements in mm. Jan. Feb. March April May June July AIR 0-2 0-4 0-2 0-6 0-7 0-8 10 A2R 0-6 0-7 1-3 1-5 0-9 1-3 1-9 A3R 0-3 0-7 10 1-3 1-2 1-8 M A4R 0-5 0-4 0-4 0-5 0-8 0-7 1-3 A5R 0-3 0-3 0-4 10 1-3 0-9 1-3 A6R 0 0-2 0-5 0-5 0-5 0-6 0-5 A8R 0-3 0-5 10 M 1-2 M 1-4 A9R 01 0-3 0-5 0-8 0-9 0-8 0-9 AlOR 0-2 0-3 0-7 0-7 0-9 0-7 0-6 AllR 0-7 0-3 0-3 0-8 M 10 0-6 A12R 01 0-2 0-3 0-3 0-7 0-7 0-9 A13R 0-2 0-3 0-5 0-7 0-7 0-8 0-7 A14R 0-2 0-8 0-7 M 1-3 1-3 M A15R 0-3 0-4 0-8 M M M 11 A16R 0-2 0-6 0-5 10 10 M 1-5 A18R 0-3 0-3 0-7 0-9 M 0-9 0-9 A19R 0-5 0-8 M 1-3 M 0-9 M A20R 0-7 0-7 10 M 1-3 M M A21R 0-3 0-5 0-9 1-7 1-4 1-3 1-3 A22R 0-3 0-7 0-3 0-3 0-5 10 M A23R 0-5 0-3 0-3 0-8 M M M A24R 0-5 0-7 1-3 1-3 1-6 1-5 1-4 A25R 0-6 0-4 0-7 M 0-8 0-7 0-6 A26R 0-5 0-6 0-5 0-7 0-7 0-5 10 A27R 0-3 0-7 0-5 M M 0-9 0-7 A28R 0-6 0-6 0-6 0-9 M 0-9 M A29R 0-3 0-9 0-3 0-5 0-9 0-7 0-8 A30R 0 0-4 0-6 0-5 0-8 0-9 1-3 A31R 0-2 0-3 0-6 0-9 M M M A33R 0 0-4 0-4 2-7 1-7 1-3 2-3 A34R 0-3 0-3 0-7 M 1-4 1-3 M A35R 0-3 0-3 0-6 0-9 0-9 0-6 0-8 Mean 0-33 0-49 0-63 0-97 103 0-98 109 Stand. Dev. 018 019 0-35 0-46 0-29 0-28 0-33 C 8385 Qq 582 PALAEONTOLOGY, VOLUME 14 Temperature: the relationship between mean monthly shell growth and sea temperature. Table 3 gives for each member of population A the distance in millimetres separating units of 30 daily growth bands measured from the ventral tip back towards the umbo. Observations were confined to 1965 growth so that ambiguity resulting from possible winter stoppages would not affect the plots. Since the shells were killed on 24 July 1965 the increments do not exactly correspond with the calendar months on the table but overlap the previous month by six days. In January only three of the shells showed a complete absence of growth, while in the remainder increments varied from 0T5 to 0-7 mm. February’s growth was within similar limits, but all shells had begun to grow. Standard deviations are greatest in March and April, and the much lower spread of Mean Monthly Sea Temperature --0 0-- Mean Monthly Shell Growth TEXT-FIG. 6. Mean monthly shell growth during early 1965 (determined by analysis of daily incre- ments) for 1963 spat cockles from population A compared with mean monthly sea temperature (recorded daily from Southend Pierhead within f mile of the living community). One standard devia- tion is shown on either side of each mean shell growth value. The high, easily discernible, April dispersion could be a useful palaeoclimatic indicator. values for May is striking (text-fig. 6). This is to be expected in view of the extreme day- to-day variations in air temperature experienced in the spring, when mortality rates among cockles are also high (Kreger 1940, p. 191). Thus in April A22 put on only 0-25 mm of shell, while A33, making a late start, put on 2-7 mm. Thereafter the increment curve flattens off, with average monthly growth of 1 mm retained for four months regardless of a steep rise in the sea temperature curve. It is quite clear from text-fig. 6 that mean monthly shell growth does not parallel mean monthly sea temperature. Obviously some other major factor such as plankton supply must be exerting an over- riding influence in early summer of the third year. Winter rings and frosts. Text-fig. 4 shows diurnal increment plots for five shells: A12, A 26, B9 were collected on 24 July 1965; R22, J13 in August 1967. Placement on the calendar scale is based on the date of killing of the 1965 specimens, but difficulties arise with 1964 and 1963 increments since there are sharp black bands cutting across the shell structure during the winters of 1964/5 and 1963/4 which clearly indicate that cessation of growth had occurred. The problem is to decide for how long growth had ceased. A reasonable attempt can be made to estimate this by back-counting A26, for it is found that the first of the four 1964/5 winter rings coincides with a period of sub-zero tempera- tures. The number of growth lines which were counted between the next three rings is in close agreement with the number of days when air temperature rose above zero. The suggested correlation is indicated on text-fig. 7. The placement of R22 and J13 is based similarly on an assumed broad relationship to air temperature. The evidence for the G. E. FARROW: PERIODICITY STRUCTURES IN CERASTODERMA 583 50JU r Daily Growth Increment TEXT-FIG. 7. Suggested correlation of winter rings in A26, a sensitive individual (Table 2) with sub-zero air temperatures: 1964/5. 100 r Daily Growth Increment 0 L M TEXT-FIG. 8. Possible correspondence of extreme growth fluctuations in R22 with air temperature variations in the spring of 1965. 584 PALAEONTOLOGY, VOLUME 14 correlation is presented in text-fig. 8, which shows the possible correspondence of extreme growth fluctuations in the high shore R22 with air temperature variations; the start of growth coincides with a period of rapid temperature rise after severe frost. The sudden increase in growth in mid-June may apparently be related to a similar trend in the sea temperature curve. Correspondence of disturbance rings and growth fluctuations in the summer of 1964 indicates that growth stoppages were not prolonged in the cockles’ second winter, being of the order of two to five days, but if the same analytical techniques are applied to the growth curves for 1963, viz. that the first winter ring corresponds to the first winter frost (text-fig. 9), it is evident that growth stoppage was more prolonged during the first winter. Here, considering A26 (a sensitive individual. Table 3), growth ceased for a period of from ten days to a fortnight on four occasions. 25 15 - Air Temperature °C 5 - -5 - 150 r Daily Growth Increment 0 - A S O N D TEXT-FIG. 9. First-year growth in A26 compared with air temperature: autumn 1963. Placement on calendar scale based on assumed correlation of first winter ring with first frost over the cockle banks (cf. text-fig. 7). Even allowing for a slightly greater incidence of frost in 1963/4 than in 1964/5 the differences seem too large to be explained completely by temperature, and this tends to support the notion of House and Farrow (1968, p. 1386) that young cockles are more susceptible to environmental variables in their first winter than in their second (cf Craig and Hallam 1963, p. 740). This is seemingly at variance with the evidence of external growth rings, which are usually more clearly marked in later life. The explanation for this discrepancy seems to be that in the first winter growth before and after the stoppage is rapid, and hence the ring is very narrow ; whereas in the second and subsequent winters, although growth may never actually cease for more than two or three days at a time, the increments before and after the absolute minimum are low, and so a broader more easily seen ring is formed. G. E. FARROW; PERIODICITY STRUCTURES IN CERASTODERMA 585 CONCLUSIONS Practical contribution of the technique. Marine biologists currently involved in studies on shell growth will find certain critical problems in the interpretation of external growth rings alleviated by studying internal periodicity patterns in the manner indicated in this paper. One major difficulty concerns the recognition of the first winter ring, which may be badly abraded in certain specimens or ill defined externally in others from areas enjoying relatively high winter temperatures (Cole 1956, p. 78). In both instances internal examination by acetate peels should settle the question. Another outstanding problem, and one which has hampered the adoption of growth ring analysis by palaeon- tologists, is to separate disturbance rings from those of annual origin. Craig and Hallam (1963) achieved such a separation statistically, but this method requires large numbers of shells. If only small samples or isolated individuals are available analysis of micro- banding can readily distinguish a disturbance ring, with its sudden reduction in daily increment compared with relatively stable background values from the annual ring, where daily increments gradually diminish over a period of weeks. In studies on population dynamics it is of considerable importance to know the degree to which the members of any one population react simultaneously to environ- mental changes. The contribution which analysis of daily growth banding can make to such studies may be assessed from the fact that it is possible by back-counting from midsummer collection dates to pinpoint exactly for each shell the date at which resump- tion of growth took place after the winter cessation. Using a vernier stage it is a rela- tively rapid process to count the banding of a statistically significant number of shells and calculate the date of the spring surge in growth rate, which may then be compared with diurnally recorded environmental variables and correlations sought. Thus in the Thames, even in the most variable cockle population, 75% started to grow within a period of one month at the beginning of their third year. The diurnal growth plot for any shell provides moreover an absolute record of the animal’s growth, which in many circumstances may be more useful than existing relative measures. This permits a true estimate of the actual length of the growing season, and one can demonstrate thereby that mature cockles living under crowded conditions have a much shorter growth season than more scattered examples. It would also be most instructive to compare the seasonal growth pattern of cockles living in conditions of reduced salinity with more normal examples: this could well be done in Poole Harbour (Cole 1956, pp. 82-83). The validity of the technique of daily growth line analysis is demonstrated nicely by the close agreement of many of its results with those reached through shell length measurements on successive collections of a single population. Two examples will suffice. First, in the second-year growth rate as determined by daily banding increases rapidly in April to maximum daily values in May; this agrees with macroshell measure- ment (Orton 1926). Several shells showed midsummer reduction in growth which could not be correlated with temperature variations. It is possible that this phenomenon is associated with spawning activity, which is known to produce an external ring in Pecten maximus (Mason 1957), though to establish this firmly for Cerastoderma edule requires more specific collecting. Second, back-counting daily bands in continuously growing members of populations sampled in the summer of 1965 gave spatfall dates ranging 586 PALAEONTOLOGY, VOLUME 14 from late May to early September 1963, in very good agreement with observations in the Thames (D. A. Hancock, personal communication 1968) and generally accepted values for C. edule (Cole 1956, p. 84). The particular value of the Thames experiments is that they were conducted on in situ specimens. Examination of periodicity structures in cockles removed to the laboratory showed that such tank-kept individuals were never normal after collection (cf. Rhoads and Pannella 1970, fig. 5), and caution is necessary when extrapolating results. Similarly, specimens which were drilled for the insertion of electrodes in connection with the work of Trueman et aJ. (1966) on burrowing activity also possessed abnormal shell growth patterns. Not only can the date of origin of a population be determined from daily growth analysis but also, for dead shell accumulations, the season of death. This has relevance to both the ecologist and palaeoecologist concerned with mortality rates and the season of greatest mortality. On the potential significance of the method in these fields see also Hancock and Simpson (1962), Craig and Hallam (1963), Craig (1967) and Hallam (1967); already the technique has been used for seasonal dating in archaeology (Coutts 1970). PaJaeoecological significance of daily growth analysis. Rhoads and Pannella (1970) have recently reviewed the general field of periodicity structures in bivalves. One or two supplementary points of interest are, however, raised by the cockle work in the Thames. The first concerns the influence of substrate on growth rate. In his fascinating general study of cockle growth rates Cole (1956) noted considerable uniformity over a wide range of substrates, such factors as exposure to wave action being more important in limiting growth. Nevertheless it does seem that within a single area where exposure, temperature, and salinity values are closely similar in adjacent cockle beds the influence of substrate can be perceived. The observed higher growth rate in sand than in mud is paralleled in other infaunal bivalves such as Mya arenaria (Swan 1952), and it would be interesting palaeontologically to study molluscan growth rates in relation to biofacies variation, especially where widespread burrowing species are recorded from a range of sedimentary types. The second point concerns the role of intertidal exposure in reducing shell growth. Elsewhere, particularly in the Burry Inlet, this is striking; but in the Thames its effect on periodicity structures is hard to detect. The large tidal range and abnormally high situation of the Burry beds (Hancock and Urquhart 1966, p. 16) must have a major bearing on this difference since certain cockles are completely emersed for several days during low amplitude neap tides. Not all cockles situated at the level of neap high water are necessarily of reduced size, however, as Cole (1956, p. 79) has pointed out in the case of the fascinating Eurzey Island cockles from Poole Harbour, which are protected from desiccation by the roots of Spartina townsendii and are only marginally smaller than examples from mean tide level. In the Thames the effects of variations in population density and substrate effectively prevent the clear recognition of tidal control on shell growth. One important palaeoecological point to emerge from this work is the demonstration of the very different response shown by various members of one supposedly homo- geneous population to environmental factors such as sub-zero temperatures. Some G. E. FARROW: PERIODICITY STRUCTURES IN CERASTODERMA 587 sensitive shells may stop growing on four separate occasions during the winter while others show no evidence of cessation. Variation in microtopography may conceivably be responsible, for Petersen (1958, p. 21) has already drawn attention to the importance of this in his ecological work on Danish cockles. Alternatively the shells may reflect genetic differences in growth pattern, and the possibility that the beds have been stocked from two distinct sources should be borne in mind; presumably a predominantly sublittoral population would provide more susceptible individuals. In a fossil population care should be taken to sample adequately in order subsequently to assess the range of susceptibility shown by each of its members; see Table 2 for the Recent model. This intrapopulation variability in growth increments, even when averaged over monthly periods, is so great that the use of periodicity structures in palaeotemperature investiga- tions is likely to prove far from straightforward; currently it is certainly less promising than other aspects of the work. Correlation between mean monthly growth and mean sea temperature is low, but the sharp rise in growth rate between March and April of the third year, easily decipherable from internal daily microbanding, could be a useful seasonal indicator, especially as this is also very clearly the season of greatest increment variation between specimens (text-fig. 6). Cole (1956, p. 86) has noted anomalies in the relationship of cockle size to sea temperature, observing that although within Poole Harbour, the region with the highest summer temperatures, cockles are large, they are even larger off Barra (Walton 1919) where the summer temperatures are much lower. Availability of food and a relatively sheltered situation must be more important factors. Acknowledgements. This work forms part of a project supported by N.E.R.C. which is being carried out at the University of Hull under the general direction of Professor M. R. House. The interest and collaboration of Dr. D. A. Hancock and Mr. A. C. Simpson of the Fisheries Laboratory, Burnham-on-Crouch, is a pleasure to record. They kindly made available population density data of cockles on the Southend Flats collected by Mr. G. Pickett. Messrs. J. Clapison and K. G. Walker were responsible for peel preparation and photography. Sea temperature and other daily records were supplied by Southend Pier and Foreshore Manager, Mr. G. S. Dawson. REFERENCES BARKER, R. M. 1964. Microtextural variation in pelecypod shells. Malacologia, 2, 69-86. CLARK, G. R. 1968. Mollusc Shell: daily growth lines. Science, 161, 800-802. COLE, H. A. 1956. A preliminary study of growth-rate in cockles {Cardium edide L.) in relation to commercial exploitation. /. Cons. perm. int. Explor. Mer. 22, 77-90. couTTS, p. J. F. 1970. Bivalve growth patterning as a method for seasonal dating in Archaeology. Nature, 226, 874. CRAIG, G. Y. 1967. Size-frequency distributions of living and dead populations of pelecypods from Bimini, Bahamas, B.W.I. J. Geol. 75, 34—45. and HALLAM, A. 1963. Size-frequency and growth-ring analyses of Mytihis edidis and Cardium edule and their palaeoecological significance. Palaeontology, 6, 731-750. DAVENPORT, c. B. 1938. Growtli lines in fossil pectens as indicators of past climates. /. Paleont. 12, 514-515. FERGUSON, c. w. 1968. Bristlecone pine: science and esthetics. Science, 159, 839-846. HALLAM, A. 1967. The interpretation of size-frequency distributions in molluscan death assemblages. Palaeontology, 10, 25-42. HANCOCK, D. A. 1965. Graphical estimation of growth parameters. /. Cons. perm. int. Explor. Mer. 22, 77-90. 1967. Growth and mesh selection in the edible cockle {Cardium edule L.). J. appl. Ecol. 4, 137-157. 588 PALAEONTOLOGY, VOLUME 14 HANCOCK, D. A. and SIMPSON, A. c. 1962. Parameters of marine invertebrate populations. In e. d. le GREN and M. w. HOLDGATE, editors. The Exploitation of Natural Animal Populations, pp. 29-50. Oxford. — and URQUHART, A. E. 1966. The fishery for cockles {Cardium edule L.) in the Burry Inlet, South Wales. Fishery Investigations, Ser. II, 25, 1-32. HOUSE, M. R. and farrow, g. e. 1968. Daily growth banding in the shell of the cockle, Cardium edule. Nature, 219, 1384-1386. KREGER, D. 1940. On the ecology of Cardium edule L. Arch. Neerl. Zool. 4, 157-200. MASON, J. 1957. The age and growth of the scallop, Pecten maximus (L.), in Manx waters. J. Mar. biol. Assoc. U.K. 36, 473-492. MILLAR, R. H. 1968. Growth lines in the larvae and adults of bivalve molluscs. Nature, 217, 683. ORTON, J. H. 1 926. On the rate of growth of Cardium edule. Part 1 . Experimental observations. J. Mar. biol. Assoc. U.K. 14, 239-279. PANNELLA, G. and MACCLiNTOCK, c. 1968. Biological and environmental rhythms reflected in molluscan shell growth. In Paleontological Society Memoir, 2, 64-79. and THOMPSON, m. n. 1968. Paleontological evidence of variations in length of synodic month since late Cambrian. Science, 162, 792-796. PETERSEN, G. H. 1958. Notes on the growth and biology of the different Cardium species in Danish brackish water areas. Meddelelser fra Danmarks Fiskeri- og Havimdersogelser, n.s. 2, no. 22. RHOADS, D. c. and PANNELLA, G. 1970. The use of molluscan shell growth patterns in ecology and palaeoecology. Lethaia, 3, 143-161. SWAN, E. F. 1952. The growth of the clam Mya arenaria as affected by the substratum. Ecology, 33, 530-534. TRUEMAN, E. R., BRAND, A. R. and DAVIS, p. 1966. The dynamics of burrowing of some common littoral bivalves. J. exp. Biol. 44, 469-492. WALTON, c. L. 1919. On the shell of Cardium edule. Lane. Seafish Fab. Rep. no. 28, 47 pp. WELLS, J. w. 1963. Coral growth and geochronometry. Nature, 197, 948-950. GEORGE E. FARROW Department of Geology The University, Hull Typescript received 28 November 1970 A THALASSINOIDES BURROW CONTAINING THE CRUSTACEAN GLYPHAEA UDRESSIERI (MEYER) FROM THE BATHONIAN OF OXFORDSHIRE by B. W. SELLWOOD Abstract. A Thalassinoides burrow from the White Limestone contains the macrurous crustacean Glyphaea udressieri (Meyer). The Glyphaea was probably the excavator of the burrow and a link is tentatively suggested between Glyphaea and Thalassinoides and the faecal pellet Favreina. Kennedy et ciL (1969) recorded the faecal pellet Favreina within Thalassinoides burrows in the White Limestone and Forest Marble at Kirtlington, Oxfordshire. They suggested that the pellets were produced by crustaceans and favoured the palinuran Meeochinis clypeatus (Carter) as the producer. Recently, a specimen of Glyphaea has been found within a Thalassinoides burrow at the same locality. This occurrence suggests an association between Thalassinoides, Favreina, and Glyphaea. Thalassinoides is a largely horizontal burrow system showing Y-shaped branching patterns and has been figured by many authors (e.g. Hantzschel 1962, Farrow 1966, Kennedy 1967, Sellwood et al. 1970). Stratigraphy and oeeurrence. The stratigraphy at Kirtlington (Ref. SP 494199) has been described by McKerrow et al. (1969). The specimen figured was obtained from beds not described in their paper, but some 2 to 2-5 m below Bed a in Profile 1 (op. cit., p. 58). These beds still belong to the White Limestone Group (McKerrow, pers. comm.). Thalassinoides is common with abundant moulds of disarticulated Modiolus, Trigonia, Goniomya Pinna, pectinids, lucinoids, and numerous veneroids. Brachiopods are absent. The majority of the bivalves are preserved in their current stable positions indica- ting considerable winnowing and sorting of the shells. Deposition is believed to have occurred in slightly deeper-water conditions than those envisaged by McKerrow et al. for the deposition of the remainder of the White Limestone (inter-tidal to sub-tidal). Limitations of the exposure make it impossible to define the relations of this facies to the rest of the White Limestone above. Glyphaea has been recorded from the same locality by Phillips (1871) and Dr. Gwyn Thomas has recently obtained three specimens from higher levels. Flowever, it is not clear whether these specimens are within burrows. The relationship between Glyphaea and the burrow. Plate 108 (lower half) shows part of a Thalassinoides burrow with the articulated remains of the macrurous (palinuran) crustacean Glyphaea udressieri (Meyer) lying on its left side at the bottom of the burrow. The left cheliped and three walking legs are visible but due to the hardness of the matrix, no further parts can be safely exposed. The matrix and the burrow-fill are of identical composition being lignitic and quartzose bio-microsparites. The burrow does not bear any crustacean scratch-marks, but the size of the crustacean is compatible with its being [Palaeontology, Vol. 14, Part 4, 1971, pp. 589-591, pi. 108.] 590 PALAEONTOLOGY, VOLUME 14 the burrow-excavator. Alternative explanations are that the animal was washed into the burrow, or that it was merely occupying a burrow constructed by another species. From the preservation and attitude of the fossil, the former is unlikely; the latter impossible to substantiate. On circumstantial grounds I prefer to regard the Glyphaea as the constructor of the burrow. The relationship of Glyphaea to Favreina. The White Limestone at Kirtlington contains abundant Thalassinoides burrows identical to that figured and these are often filled with Favreina. However, the figured specimen is not associated with Favreina but in view of the similarity in the forms of the Thalassinoides burrows it is nevertheless likely that Glyphaea formed the faecal pellets. Other crustaceans which might well have been responsible are Eryma, recorded from the Great Oolite by Woods (1925-1931), and Orhomalus, associated with Thalassinoides (recorded by Dr. A. Kendall (pers. comm.)). Remarks. Although crustaceans have been considered responsible for a variety of burrow-types, including Thalassinoides, the crustaceans themselves are but rarely preserved. Waage (1968) has figured an Ophiomorpha burrow with associated Callianassa debris, and Shinn (1968) has also figured part of a burrow system containing ‘a shrimp, the presumed architect’, both from the Cretaceous of North America. Ehrenberg (1944), who erected the ichnogenus Thalassinoides, noted its association with Callianassa in the Austrian Miocene, while Fiege (1944) suggested that either Glypheoids or Thalassinoids produced some Thalassinoides burrows in the Muschelkalk. The rarity of crustacean remains in their burrows may be explained in several ways. Firstly, burrowing crustaceans often have thin and reduced skeletons (cf. Callianassa, Alpheus, and Upogebia) which would not readily be preserved. Only the chelipeds are strongly calcified and, as Bromley (1967) noted in the Chalk, these are the parts most frequently found. Secondly, when moribund, many burrowers move to the surface, where their chances of preservation are much reduced (Schafer 1962). The preservation of this type of animal probably requires some form of catastrophe involving rapid burial (Bromley 1967), possibly by the whole burrow- system becoming almost instantaneously filled with sediment causing the death of the occupant. Glyphaea is not a member of the Thalassinoidea but Glaessner (1969) regards them as having evolved from the Glypheoidae. Similarities in burrow-style in these groups may reflect their related histories. Conclusions. Glyphaea iidressieri (Meyer) is believed to have constructed a Thalassinoides- type burrow in a sub-tidal environment. Elsewhere, at the same locality, Thalassinoides, is associated with Favreina, thus suggesting that Glyphaea produced both the burrows and the pellets. EXPLANATION OF PLATE 108 Fig. 3. Undersurface of part of Thalassinoides burrow with Glyphaea iidressieri (Meyer) arrowed, xO-5. Fig. 4. Close-up showing left cheliped and parts of the carapace and three walking legs, X 3. B. W. SELLWOOD: THALASSINOIDES BURROW WITH GLYPH AE A 591 Acknowledgements. I am grateful to Drs. W. S. McKerrow, W. J. Kennedy, R. Goldring, and A. Hallam for their advice and to Dr. G. Thomas and Dr. A. Kendall in permitting me to mention their material. REFERENCES BROMLEY, R. G. 1967. Somc observations on burrows of thalassinidean Crustacea in chalk hardgrounds. Q. Jl. geol. Soc. Land. 123, 157-182. EHRENBERG, K. 1944. Erganzcndc Bemerkungen zu den seinerzeit aus dem Miozan von Burgschleinitz beschreibenen Gangerkern und Bauten dekapoder Krebse. Paldont. Z. 23, 345-359. FARROW, G. E. 1966. Bathymetric zonation of Jurassic trace-fossils from the coast of Yorkshire, England. Palaeogeog. Palaeoclim. Palaeoecol. 2, 103-151. FiEGE, K. 1944. Lebensspuren aus dem Muschelkalk Nordwestdeutschlands. Neues Jb. Miner. Geol. Paldont. Abh. B88, 401-426. GLAESSNER, M. F. 1969. Decapoda. In moore, r. c. (ed.). Treatise on Invertebrate Paleontology, Part R, Arthropoda, R400-R566, Geol. Soc. Am. and Univ. Kansas Press. HANTZSCHEL, w. 1962. Trace fossils and Problematica. In moore, r. c. (ed.). Treatise on Invertebrate Paleontology, Part W, W177-W245, Geol. Soc. Am. and Univ. Kansas Press. KENNEDY, w. J. 1967. Burrows and surface traces from the Lower Chalk of Southern England. Bull. Br. Mils. Nat. Hist. {Geol.), 15, 127-167. JAKOBSON, M. E. and JOHNSON, R. T. 1969. A Favreina-Thalassinoides association from the Great Oolite of Oxfordshire. Palaeontology, 12, 549-554. MCKERROW, w. s., JOHNSON, R. T. and JAKOBSON, M. E. 1969. Palacoccological studies in the Great Oolite at Kirtlington, Oxfordshire. Palaeontology, 12, 56-83. PHILLIPS, J. 1871. The Geology of Oxford. SCHAFER, w. 1962. Aktiio-Paldontologie nach Stiidien in der Nordsee. Frankfurt (Kramer). SELLWOOD, B. w., DURKIN, M. K. and KENNEDY, w. J. 1970. Field Meeting on the Jurassic and Cretaceous rocks of Wessex: report by the directors. Proc. Geol. Ass. 81, 715-732. SHINN, E. A. 1968. Burrowing in recent lime sediments of Florida and the Bahamas. J. Paleont. 42, 879-894. WAAGE, K. M. 1968. The type Fox Hills formation, Cretaceous (Maestrichtian), South Dakota. I. The Stratigraphy and Paleoenvironments. Bull. Peabody Mus. Nat. Hist. 27, 175 pp. WOODS, H. 1925-1931. A Monograph of the fossil Macrurous Crustacea of England. Palaeontogr. Soc. [Monogr.]. B. W. SELLWOOD Department of Geology and Mineralogy University of Oxford Parks Road Typescript received 11 March 1971 Oxford OCCURRENCE OF THE AMMONITE PTYCHOCERAS ADPRESSUM (J. SOWERBY) IN THE UPPER ALBIAN OF KENT, ENGLAND by JULIAN D. HOLLIS Abstract. The small heteromorph ammonite Ptychoceras adpressiim (J. Sowerby) is abundant at one horizon near the middle of the Varicosum Subzone {Mortoniceras inflatum Zone) in Kent. Previously it was only known in Britain from Folkestone, Kent. It also occurs less frequently low in the Varicosum Subzone. The species was probably adapted to a benthonic habitat. The rich and diverse Upper Albian ammonite fauna of Kent includes many strange heteromorph forms, including the peculiar Ptychoceras adpressum (J. Sowerby 1814). Its shell, which is seldom over 2 cm long, consists of two straight, parallel septate shafts and a short recurved body-chamber. It was probably the last representative of the Lower Cretaceous family Ptychoceratidae Meek, which was widespread in the Tethys but seldom ranged as far north as Britain. The speeies has been described and figured from Folkestone by Spath (1941, pp. 656-659), as belonging to Mastigoceras Boehm 1925. This is generally regarded as a synonym of Ptychoceras (see C. W. Wright 1957). Other isolated occurrences of the species are from Ootmarsum, Holland (Boehm 1925) and doubtfully, from Escragnolles, Var, France (Parona and Bonarelli 1897). Although considered a rarity in the British Gault, recently opened exposures in the Varicosum Subzone show that it is in fact abundant at one horizon where it forms a useful strati- graphic marker. It may be present over the entire 90 km outcrop of the Kentish Gault. TEXT-FIG. 1. Ptychoceras adpressum (J. Sowerby), Upper Gault, Mid-Varicosum Subzone. a. Reconstruction X2; nature of aperture not known. Last septum position shown by cross, b, c. Whorl sections through septate shafts and body-chamber, respectively. d. Extreme example of constricted variety X 1 -6. Occurrence 1. Ford Place Pit, Wrotham (Nat. Grid reference TQ 636591): {a) Mid-Varicosum Subzone, middle of Bed 80 (of Milbourne 1963). Poorly exposed. (9 specimens collected). (Jb) Low Varicosum Subzone, approximately 0-3 m below bed rich in Nielsenicrinus; ?Bed 69 (28 specimens). 2. Paddlesworth Pit, Snodland (TQ 692617): (a) Mid-Varicosum Subzone, Division 24 on text-fig. 2. (Abundant — over 1500 specimens). (b) Low Varicosum Subzone, Division 1 5. (Very rare — 5 specimens.) [Palaeontology, Vol. 14, Part 4, 1971, pp. 592-594.1 J. D. HOLLIS: PTYCHOCERAS 593 3. Naccolt Pit, Wye. (TR 049445): (а) Mid-Varicosum Subzone, Division 19 on text-fig. 2. (Abundant — over 250 specimens). {b) Basal Varicosum Subzone, Division 1 1 . (Very rare — 5 specimens.) 4. Folkestone Coast (TR 243365): (n) Varicosum Subzone, horizon and abundance not specified (Jukes-Browne 1900 and Spath 1941, p. 659). Section now obscured. (б) Auritus Subzone. Price (1874) (quoted by Jukes-Browne 1900, p. 82) recorded Ptychoceras sp. from his Bed XI and regarded it as peculiar to that bed. It has not been possible to sub- stantiate this occurrence. ■ PTYCHOCERAS locality outcrop of Gault PADDLESWORTH ■Top of Varicosum Subzone . 79 FORD PLACE (Modified ’rom 28 Milbourne, 1963) 27 83 * 26 82 * 81 25 24 80 G . • CT5 , ^ " 79 22 . 21 77 75 20 74 b 19 70 18 68 ' ' _ ^ 17 64 62 15 60 14 0 Coasf SecHon FOLKESTONE PhosphaMc nodule bed. Scattered phosphatic nodules . Lenticles of c lay -ironstone PTYCHOCERAS AOPRESSUM EOSCAPHITES SUBCIRCULARIS Grey c lay (plain) . Horizon of Inoceromus concentricus appeoronce a Upper PTYCHOCERAS Horizon . b IDIOHAMITES Spp . obundont c Lower PTYCHOCERAS Horizon. FOLKESTONE (Jukes-Browne, 1900) Top of Voricosum Subzo TEXT-FIG. 2. Ptychoceras adpressiim localities in the Mortoniceras inflatum Zone of the Gault and correlation of occurrences. Figures to left of columns refer to beds or divisions. 594 PALAEONTOLOGY, VOLUME 14 At the first three localities, the mid-Varicosum Subzone Ptychoceras horizon is a grey clay about 0-5 m thick. It contains scattered phosphatic nodules and fossils. Idiohamites turgidus var. subannulata Spath is a common associate probably confined to this horizon. At all three pits, the clay up to 2 m below yields Eoscaphites subcircularis Spath, which is particularly common at Naccolt. Lenticles of clay-ironstone are present at the same level at Ford Place and Paddlesworth. The Ptychoceras horizon is most extensively exposed at Paddlesworth, although specimens are equally abundant at Naccolt. Apart from the rare occurrences in the low Varicosum Subzone, P. adpressum appears to be restricted to the bed in the mid- Varicosum Subzone. When fresh, many P. adpressum are preserved as uncrushed pyrite moulds ; however, very crushed pyrite-impregnated clay moulds are more common and are easily over- looked. Body chambers are scarce and no complete individuals have been seen. Most specimens are merely septate shafts around a centimetre long. Specimens from Paddlesworth are identical to those from Folkestone figured by Spath. Fragments of the initial shaft usually show septa and the second shaft possesses a marked dorsal furrow. About 3% of individuals show a more or less strong constriction after the commencement of the second shaft, very similar to that shown by Spath (1941, text-fig. 241 i,j). He remarked that this constricted variety is ‘almost like a miniature replica of Ptychoceras puzosiammi, d’Orbigny’. Faint costation is seen on the second shaft of some individuals, but most are smooth. Specimens tend to be clustered, and it is tentatively suggested that the adults were benthonic. On a square metre of bedding surface at Paddlesworth were two groups of closely spaced adult individuals ; one with seven specimens, the other with five. 1 n addition, four isolated adult individuals occurred on the square. Juveniles may have been pelagic, thus explaining their virtual absence from the fossil assemblage. The occurrence of the species at restricted horizons may reflect brief immigrations of stock from more southern waters into temporarily favourable environments. Acknowledgements. I would like to thank Dr. J. M. Hancock for his most useful and constructive criticism; also Mr. A. Gale for his help and interest in the field. REFERENCES BOEHM, J. 1925. Zur Fauna des oberen Alb mit Mastigoceras adpressum J. Sowerby sp., bei Ootmarsum (Holland). Z. dt. geol. Ges. 77, Abh. No. 2, 198-205. JUKES-BROWNE, A. J. 1900. The Cretaceous Rocks of Britain. 1. The Gault and Upper Greensand of England. Mem. geol. Surv. U.K. MiLBOURNE, R. A. 1963. The Gault at Ford Place, Wrotham, Kent. Proc. Geol. Ass. Lond. 74, 55-79. PARONA, c. F. and bonarelli, g. 1897. Fossili albiani d’Escragnolles, del Nizardo e della Liguria occidentale. Paleontogr. ital. 2 (1896), 105. PRICE, F. G. H. 1874. On the Gault of Folkestone. Q. Jl. geol. Soc. Lond. 30, 342-368. SOWERBY, J. 1814. The Mineral Conchology of Great Britain, 1, pis. 45-78. London. SPATH, L. F. 1941. A monograph of the Ammonoidea of the Gault, pt. 14, Paleontogr. Soc. [Monogr.]. WRIGHT, c. w. 1957. In moore, r. c. (ed.). Treatise on Invertebrate Palaeontology, Part L, pp. L2 15-216. JULIAN D. HOLLIS Department of Geology Kings College University of London Strand, W.C. 2. Typescript received 25 April 1971 OSTRACODA FROM THE UNDERBARROW, KIRKBY MOOR AND SCOUT HILL FLAGS (SILURIAN) NEAR KENDAL, WESTMORLAND by R. W. L. SHAW Abstract. The fauna collected from the rocks previously referred to the Kirkby Moor Flags, permits their subdivision and correlation with the succession at Ludlow and with the Baltic sequence. In this the ostracodes play an important role. They are here described and their value in correlation is illustrated. A new genus, Huttoniella, and a new species, Neobeyrichia confluens, are described. Ostracodes occurring in the ‘Kirkby Moor Flags’ have previously received little attention. Sedgwick and McCoy 1855 identified Beyrichia Kloedeni McCoy, the presence of which is again referred to by the Geological Survey (Aveline and Hughes 1872; Aveline, Hughes, and Tiddeman 1872; Aveline, Hughes, and Strahan 1888). No further identifications of ostracodes were made until the work of Llewellyn, published in a paper by Furness, Llewellyn, Norman, and Rickards 1967, when the occurrence of Beyrichia lauensis Kiesow was noted from the Longsleddale area, to the north of the area under consideration. The present work is the result of the revision of the faunal stratigraphy of the rocks succeeding the Bannisdale Slates in the type area of the Kirkby Moor Flags. Particular attention has been paid to the ostracodes in view of their stratigraphical importanee. In conjunction with this, especially in regard to those apparently indicating a Downtonian age, the author collected and described ostracodes from the Ludlow area (Shaw 1969). STRATIGRAPHY The classification of the strata succeeding the Bannisdale Slates by the author is comparable with that outlined by Sedgwick (1842, 1845, 1852, and 1859) and by the Geological Survey in 1872 and 1888. By the systematic collecting and identification of the fauna the stratigraphical and faunal divisions have been made with greater definition. Four faunal divisions have been recognized: the Lower and Upper Underbarrow Flags, the Kirkby Moor Flags, and the Scout Hill Flags. The Lower Underbarrow Flags correlate with the Upper Leintwardinian stage; the Upper Underbarrow and Kirkby Moor Flags are of Whitcliffian age, and the Scout Hill Flags are Downtonian. Lithologically the succession forms two main units, a finer, rather ill-sorted, closely laminated group, of silt grade comprising the Lower and Upper Underbarrow Flags and a series of tough, well-sorted silts forming the upper two faunal divisions. The Scout Hill Flags additionally include 500 ft of red, laminated silts. The important fossils in each of the four divisions may be tabulated: Scout Hill Flags Frostiella gwenvaUiaiia. Kirkby Moor Flags Acastella prima, Homalonotus knight i, Neobeyricha confluens. [Palaeontology, Vol. 14, Part 4, 1971, pp. 595-611, pis. 109-113.] 596 PALAEONTOLOGY, VOLUME 14 Upper Underbarrow Flags Acastella prima, Neobeyrichia confluens. Lower Underbarrow Flags Atrypa reticularis, Chouetes lepisma, Chonetoida grayi, Neobeyrichia lauensis, Neobeyrichia nutans, Nodibeyrichia scissa, Encrinurus stubblefieldi, Lapworthura miltoni. At the base of the Upper Underbarrow Flags there is a marked decline in the variety of species, this despite the lack of any visible lithological change. In contrast to this trend is an increase in the abundance of certain species, particularly but not exclusively, at the advent of the Kirkby Moor-Scout Hill Flag lithology. Among the brachiopods this is seen especially in Protochonetes hidloviensis and Salopina lunata. The gastropods behave similarly above the base of the Kirkby Moor Flags with Cyclonema carinata, Naticopsis glaucinoides, and Holopella cingulata becoming especially abundant. The lamellibranchs, relatively rare in both variety and abundance in the lower parts of the succession, diversify considerably in the Kirkby Moor Flags. Details of the fauna from each division and its classification into assemblages have been included in another paper (Shaw 1971). SYSTEMATIC DESCRIPTIONS There are important limitations on the systematic value of the fauna occurring in this succession. Chief among these are their mode of occurrence and preservation. In the Kirby Moor Flags and the Scout Hill Flags fossils are preserved almost exclusively in lenses set in well-sorted silts. The lenses are calcareous but induration of the quartz grains, which form the bulk of the detrital material, is such that fresh, unweathered samples yield good specimens only after patient effort with a mounted needle. Weathered samples, although yielding more abundant specimens, have their limitations. In this condition the sediment is extremely friable and fossils are preserved only as internal and external moulds. These are usually coated with limonite, which, depending on the amount, either preserves or completely obliterates the ornament. In addition, due to the relatively coarse grain size (compared with the relief of the ornament), preservation of the ornament, in moulds lacking a thin film of limonite, is always poor. As a result of these problems, the identification of the fossils often relies on occasional samples in which the preservation is unusually good with additional information being supplied by the abundant less well-preserved specimens. From the samples collected, however, details of the ostracode fauna in the succession can be gained and useful material for purposes of correlation is available. Techniques used in the study of the ostracodes varies with the nature of the rock. In the weathered samples it was found useful to make rubber casts of the external moulds. These, due to the friable nature of the rock, yielded variable results. To some extent it was possible to recement the rock before applying the rubber solution by impregnating the specimen with a weak solution of Perspex in chloroform. A fairly successful method of handling the unweathered rock is to roast it for a short time and then quench in cold water. This has the effect of weakening the rock such that fracture will occur around rather than across the specimens. Specimens from the two divisions of the Underbarrow Flags are more easily studied. They occur as diffuse lenses or isolated specimens in unsorted silt whose clay fraction is high. Thus, although specimens are usually weathered and fossils are preserved as moulds, the ornament is clearly imprinted in the sediment and excellent casts are easily made recording all observable details of the ornament without damaging the specimens. The photographic method used was that described by Martinsson 1962 and magnesium oxide was used to coat the specimens. Illumination is from the north-west unless this is prevented by the con- figuration of the rock enclosing the specimen. The terminology used in the descriptions is that of Martinsson 1962. R. W. L. SHAW: SILURIAN OSTRACODA 597 In all references to localities and strata from which the described and illustrated ostracodes were found their mode of occurrence is similar. Details of this information can adequately be furnished at this point. The successions through the Lower and Upper Underbarrow Flags are lithologically monotonous and the mode of occurrence of the fossils is similarly uniform. There is a general scatter of fossils throughout the sediment but a tendency to concentration approaching lens-like accumulations exists often associated with slightly better sorted silt. In the Kirkby Moor and Scout Hill Flags the sediments again form a monotonous sequence in this case of well-sorted silts. The fossils are almost exclusively restricted in their occurrence to calcite enriched lenses, within or occasionally at the base of these silt members. In all four divisions the clearest guide to fossil localities is by way of grid reference and stating which of the divisions the material comes from. Unfortunately there can be no clearer guide by reference to the localities relationship to obvious lithological, sedimentary, or other geological marker as these do not occur. The abundances of the fossils are stated as the percentage of fossiliferous localities, in each division, yielding each species. Order palaeocopida Henningsmoen 1953 Suborder beyrichicopina Scott 1961 Superfamily beyrichiacea Matthew 1886 Family beyrichiidae Matthew 1886 Subfamily amphitoxodontidinae Martinsson 1962 Genus hemsiella Martinsson 1962 Hemsiella maccoyiana (Jones 1855) Plate 109, figs. 1-4 1855 Beyrichia maccoyiana Jones, p. 88, pi. 5, fig. 14. 1962 Hemsiella maccoyiana (Jones); Martinsson, fig. 2b. Remarks. This is a highly distinctive species, characterized particularly by a syllobium which possesses a depressed area isolating the remainder of the syllobium from a marked cuspidal portion. This depressed area is also responsible for the formation of a lunate roll on the posterior face of the syllobium. The lobes bear a characteristic reticulate ornamentation and are weakly connected ventrally. The velum is rather flat, with strongly developed tubules and distinct cross fibres. The heteromorph has the same syllobial depression, is similarly ornamented, and has a large crumina with a striate ornamentation. Occurrence. H. maccoyiana is found in the Upper Underbarrow Flags (< 5%), throughout the Kirkby Moor Flags (48%), and in the Scout Hill Flags (59%). It is identified in the Ueba I core (Martinsson 1964) and in two of the Polish cores described by Witwicka (1967). Genus huttoniella gen. nov. Derivation of name. Feminine diminutive of Latin Huttonius, associated with, or, inhabitant of. Old Hutton near Kendal. Type species. Huttoniella contract a n. sp. Diagnosis. Hinge straight, shorter than maximum length of slightly amplete valve, ventral margin smoothly curved. Lobal connections extremely weak. Preadductoral lobe narrow in the tecnomorphic valves, and very narrow in the heteromorphs. Lobal R r C 8385 598 PALAEONTOLOGY, VOLUME 14 ornamentation finely reticulate, which in the heteromorphic valves is stretched to a striate type. Velum narrow, straight, with tubules and lacks a denticulate margin. In the heteromorph the velum is considerably reduced. Remarks. While the new genus clearly belongs to the Amphitoxotidinae it differs from all other described genera. In view of the cruminal and lobal morphology, however, it is probably related to the Cryptopholobus-Lophoctonella part of the subfamily. Huttoniella contracta sp. nov, Plate 109, figs. 5, 6 Derivation of name. From the Latin contractus, narrow, referring to the narrowness of the preadductoral lobe. Holotype. Left valve, Plate 109, fig. 5. Type Stratum and Type Locality. Kirkby Moor Flags, about ^ mile S. of Old Hutton, Kendal, Westmorland (G.R. 5642 8774). Diagnosis. As for genus. Description. Hinge straight and shorter than the maximum length of the valve. Syllo- bium weakly connected to the preadductoral lobe which is narrow. Anterior lobe broad but barely connected with the preadductoral lobe. All lobes have a well-developed reticulate ornamentation. Small tubercles occur on the posteroventral part of the an- terior lobe, and the posterodorsal part of the syllobium. Velum narrow, strongly tubulose and with a non-denticulate border. Subvelar field narrow, limited ventrally by a marginal frill. The heteromorph, apart from the addition of a large crumina, shows considerable differences from the tecnomorph. Syllobium large but rather narrow, preadductoral lobe very narrow, anterior lobe largely absorbed into the crumina. Ornamentation of lobes stretched from reticulate in the tecnomorph to a striate type in the heteromorph. Velum considerably modified in the heteromorph into a thick rolled marginal structure extend- ing only as a very weak feature half way across the crumina. Dimensions. Hinge length and sulcal height in microns: Tecnomorph 710-490, 850-610. Heteromorph 1090-730, 1170-820. Occurrence. Kirkby Moor Flags (33%) and Scout Hill Flags (29%) around Kendal. EXPLANATION OF PLATE 109 Figs. 1-4. Hemsiella maccoyiana (Jones). 1 , right valve, GSM Z19029, Scout Hill Flags, G.R. 5378 8389. 2, internal mould of left valve, GSM Z19030, Kirkby Moor Flags, G.R. 5854 9396. 3, left hetero- morph valve, GSM Z19032, Kirkby Moor Flags, G.R. 5537 9340. 4, internal mould of right hetero- morph valve, Kirkby Moor Flags, GSM Z19030, G.R. 5854 9396. Figs. 5, 6. Huttoniella contracta gen. et sp. nov. Kirkby Moor Flags. 5, holotype, left valve, GSM Z19033, G.R. 5642 8774. 6, left heteromorph valve, GSM Z19052, G.R. 5549 8449. Figs. 7, 8. Macrypsilon salterianum (Jones), Kirkby Moor Flags, G.R. 5544 8267. 7, right heteromorph valve, GSM Z19055. 8, right valve, GSM Z19054. All figures x 40. Palaeontology , Vol. 14 PLATE 109 SHAW, Silurian ostracodes <1 V'l' ., .Ji 599 R. W. L. SHAW: SILURIAN OSTRACODA Genus lophoctonella Martinsson 1962 Lophoctonella cf. scaneusis (Kolmodin 1869) Plate 1 10, figs. U6 1869 Beyrichia scaneusis Kolmodin, p. 19, fig. 11. 1962 Lophoctonella scaneusis (Kolmodin); Martinsson, figs. 6a-c. Description. Hinge straight, shorter than maximum valve length. Valve preplete, margin smoothly curved. Lobation well developed. Syllobium connected weakly to preadductoral lobe, anterior lobe separate. All three lobes ornamented by a cristal loop; that on the syllobium 2|-3 times the width of the limbs of the loop; preadductoral loop narrower, being equal in width to the limbs; that on the anterior lobe very narrow, the field between the limbs only just being discernible. Lobes reach but do not extend beyond the hinge. Velum broad, fairly flat-lying dorsally, upturned to near vertical anteriorly, strongly developed tubules each with two denticulations. The morphology of the heteromorph is strikingly different from that of the tecno- morph. In addition to the large crumina set over the deep anteroventral depression, the velum has weakly developed tubules without denticulations and is considerably thicker. Syllobium divided into a cuspidal part extending beyond the hinge line and a relatively raised ventral part. Neither part bears a cristal loop, but the ventral part is ornamented by randomly distributed tubercles. Preadductoral lobe partially absorbed into the crumina, dorsally it bears a small but well-marked cristal loop. Anterior lobe relatively small, unornamented, projects beyond the hinge line. Crumina ornamented by a striate pattern. The velum reaches much of the way across the crumina; below it is a toric ridge. Remarks. This species differs from the type species L. angustilaqueata Martinsson in the small size of the cristal loop on the preadductoral lobe in the heteromorph and the presence of a cristal loop on the anterior lobe in the tecnomorph. The velum also differs. It does not rise anteriorly into a strongly elevated comb-like structure and lacks the denticulations on the margin in the heteromorph. The specimens from the Lake District show many characters in common with L. scaneusis, especially in the disposition of the lobes, nature of the velum and ornamentation of the crumina, but differ to some degree in the lobal ornamentation. The heteromorph illustrated by Martinsson (1962, fig. 122b) shows only a very weak indication of a cristal loop on the dorsal face of the preadductoral lobe, and lacks the tuberculation on the ventral part of the syllobium. No cristal loop is seen on the anterior lobe of a tecnomorph figured by Martinsson (1962, fig. 6c). Occurrence. This species is recorded throughout the succession in the Kendal area : Lower Underbarrow Flags (6%), Upper Underbarrow Flags (< 5%), Kirkby Moor Flags (46%), Scout Hill Flags (59%). There is no apparent change in its form, though this is difficult to confirm from the generally poor preservation. Genus macrypsilon Martinsson 1962 Macrypsilon salterianwn (Jones 1855) Plate 109, figs. 7, 8 1855 Beyrichia salteriana Jones, p. 89, pi. 5, fig. 15a. 1962 Macrypsilon salterianiim (Jones); Martinsson, fig. 2d. 600 PALAEONTOLOGY, VOLUME 14 Remarks. The tecnomorphic valves are characterized by the occurrence of narrow sulci which isolate the lobes and join ventrally to give a Y-shape. The lobes are full and well rounded, with a very fine reticulate ornamentation. There is characteristically a slight swelling at the dorsal end of the adductoral sulcus. The velum is narrow and weakly tubulose. In the heteromorph, the crumina is large. The syllobium has a tendency to deflate posteriorly, with a lip or rim marking its posterodorsal extremity. Occurrence. This species occurs in the upper part of the Kirkby Moor Flags (8%) and in the Scout Hill Flags (12%). It is not a common species but is regarded as characteristic of these beds. It is a species characteristic of strata of high Ludlovian and Downtonian age, being recorded from the Baltic in erratic blocks by Martinsson (1962), from the Stonehouse Formation, Nova Scotia (Copeland 1964) and from the Chlapowo core, Poland (Witwicka 1967) in addition to the above-mentioned Lake District occurrence. Subfamily beyrichiinae Matthew 1886 Genus neobeyrichia Henningsmoen 1954 Neobeyrichia confluens sp. nov. Plate 111, figs. 1-5 Derivation of name. Latin confluens, flowing together of two rivers, referring to the merging of the ventral lobule of the syllobium and the velum. Holotype. Right tecnomorphic valve. Plate 111, figs. 2 and 4. Type Stratum and Type Locality. Kirkby Moor Flags, about -t mile S. of Old Hutton, Kendal, Westmorland (G.R. 5642 8774). Diagnosis. Ventral lobule of syllobium merges with the velum. Velum lacks tuberculate margin. Syllobial groove shallow. Surface ornamentation weak. Description. Hinge straight, valves slightly preplete. Syllobium divided by a shallow groove. Ventral lobule of syllobium merges with velum and is weakly connected with the preadductoral lobe. Preadductoral lobe narrow, straight, and reaches the hinge line. Anterior lobe slightly arcuate and isolated from the other lobes by a deep anteroventral depression. Velum well developed, but rather narrow, lacks tuberculate ornamentation on its margin. Undersurface of velum shows a marked toric ridge with evidence of two weaker ridges above. Heteromorph has a relatively small crumina. Syllobium strongly divided by a groove. Lobes ornamented by irregularly distributed small tubercles which also occur on posterior part of velum. EXPLANATION OF PLATE 110 Figs. 1-6. Lophoctonella cf. scanensis (Kolmodin). 1, left heteromorph valve, GSM Z19035, Kirkby Moor Flags, G.R. 5642 8774. 2, right valve (rubber cast), GSM Z19053, Scout Hill Flags, G.R. 5987 8104. 3, left heteromorph valve, GSM Z19036, Kirkby Moor Flags, G.R. 5642 8774. 4, left valve, GSM Z19036, Kirkby Moor Flags, G.R. 5642 8774. 5, left heteromorph valve showing details of cruminal ornamentation, GSM Z19038, Kirkby Moor Flags, G.R. 5642 8774. 6, ventral view of 1, showing trace of velum across crumina. All figures X 40. Palaeontology, Vol. 14 PLATE no SHAW, Silurian ostracodes R. W. L. SHAW: SILURIAN OSTRACODA 601 Dimensions. Hinge length and sulcal height in microns; Tecnomorph 740-540, 1050-790, 1620-1080, 1970-1430. Heteromorph 1970-1370. Remarks. The ontogeny of this species is interesting. In the smallest specimens in which the shell is preserved, the margin of the velum is ornamented by a number of fine spines. In the smaller specimens the syllobium is undivided. The division is introduced and becomes increasingly incised in successive instars. The species is closely related to Neobeyrichia lauensis, but is clearly distinguished by the confluence of the ventral lobule of the syllobium with the velum, the lack of tubercles ornamenting the margin of the velum, the lack of surface ornamentation on the lobes, the relative weakness of the syllobial groove, and the extension of the preadductoral lobe to the hinge-line. Stratigraphically, Neobeyrichia confluens appears to succeed the lauensis form of the genus. Occurrence. N. confiiiens is a common species in the Upper Underbarrow (16°o) and Kirkby Moor Flags (49%). It is recorded only as a single specimen in the Lower Underbarrow Flags and is rarely encountered in the lowest beds of the Scout Hill Flags (< 5%). Neobeyrichia lauensis (Kiesow 1888) Plate 1 1 1, fig- 6 1888 Beyrichia lauensis Kiesow, p. 8, pi. 2, figs. 1, 2. 1962 Neobeyrichia {Neobeyrichia) (Kiesow); Martinsson, p. 318, figs. 10, 177, 178. Remarks. The species is distinguished by a well-marked differentiation of the syllobium into a shorter dorsal lobule and a longer ventral one. The latter is connected by a weak col to the preadductoral lobe which does not reach the hinge-line. The surface is orna- mented by small evenly distributed granules with a number of superimposed small tubercles. The velum is highly distinctive, being thick and with a number of large tubercles along its ventral margin. In these characters it is readily distinguished from Neobeyrichia confluens sp. nov. Occurrence. N. lauensis is characteristic of the Upper Leintwardine Beds of the Welsh Borderland, where, in association with the brachiopod Chonetoidea grayi it is regarded as definitive of the Upper Leintwardinian age. N. lauensis is common in the Lower Underbarrow Flags (31°o) and there occurs together with C. grayi. The range of this ostracode in the Lake District strata extends beyond the extinction of its Upper Leintwardinian suite of fossils into the Upper Underbarrow Flags (45%) and occurs with N. confluens. N. lauensis is also described from the ‘Hemse’ Beds of Gotland (Martinsson 1967) and is recorded in the succeeding Eke Beds (Martinsson 1962). Neobeyrichia nutans (Kiesow 1888) Plate 112, fig. 3 1888 Beyrichia buchiana var. nutans Kiesow, p. 7, pi. 1, figs. 11-14. 1962 Neobeyrichia (Neobeyrichia) nutans (Kiesow); Martinsson, p. 321, figs. 40b, 179. Remarks. The syllobium is broad and divided into a dorsal cusp, a median lobule and a ventral lobule by two furrows. The ventral lobule is connected to the preadductoral 602 PALAEONTOLOGY, VOLUME 14 lobe. The anterior lobe is isolated and projects beyond the hinge line. The velum is thick with two rows of small spines. The cruminal ornamentation is distinctive, there being a striate area ventromarginally with the remainder covered by small elongate tubercles. Occurrence. This species is characteristie of the Lower Underbarrow Flags where it occurs fairly commonly (9%). In the ‘Hemse’ Beds of Gotland, Martinsson (1967) regards it, together with Ham- mariella pulchrivelata, as part of the ostracode fauna preceding the N. laueusis~N. scissa association. If this is the case then its range in the Lake District succession is somewhat extended. Neobeyrichia torosa (Jones 1855) Plate 111, figs. 7, 8 1855 Beyrichia kloedeni var. torosa Jones, p. 167, pi. 6, figs. 10-12. Description. Hinge straight, slightly shorter than the maximum length of valve. Ventral margin smooth and valve preplete. Syllobium fairly narrow, divided by a groove into a shorter cuspidal part and a longer ventral part. Connection of syllobium with pre- adductoral lobe weak. Preadductoral lobe knob-like, not reaching the dorsal margin. Anterior lobe isolated, developed into a pronounced cusp dorsally. Syllobium and anterior lobes marked by extension dorsally of cusps into long hollow spines. Lobes ornamented by fairly large tubercles. Tendency for ventral part of anterior lobe to be separated from cuspidal part. This is not a clearly defined feature, but evidently illustrates the approach of the species to the genus Nodibeyrichia. Velum well developed, with a large number of long slender spines. No heteromorph of this species has been encountered in the limited amount of Lake District material. Remarks. Specimens of Neobeyrichia torosa collected from the Whitclitfe Beds show some variations on the Lake District material. Although the bulk of the specimens are closely comparable a number of Welsh Borderland individuals develop a node on the ventral part of the syllobium. Occurrence. Neobeyrichia torosa is recorded from the Lower Underbarrow Flags (< 5%), the Kirkby Moor Flags (< 5%), and the Scout Hill Flags (10%) in the Lake District succession but occurs only rarely. It is a common fossil in the Ludlow area ranging from the Lower Leintwardinian to the top of the Whitcliffian. It is not recorded outside this country. EXPLANATION OF PLATE 111 Figs. 1-5. Neobeyrichia confluens sp. nov. Kirkby Moor Flags. 1, right valve, GSM Z19039, G.R. 5642 8774, x40. 2, ventral view of fig. 4 showing subvelar field, X 30. 3, left valve, GSM Z19040, G.R. 5642 8774, x40. 4, holotype, right valve, GSM Z19041, x 30. 5, left heteromorph valve, GSM Z19059, G.R. 6051 8688, X 30. Fig. 6. Neobeyrichia lauensis (Kiesow). Right valve (rubber cast), GSM Z19060, Upper Underbarrow Flags, G.R. 6030 8902, X 30. Figs. 7, 8. Neobeyrichia torosa (Jones). Scout Hill Flags, G.R. 5696 8283. 7, right valve (rubber cast), GSM Z19066, X 30. 8, left valve, internal mould, GSM Z19066, X 35. Palaeontology, Vol. 14 PLATE 111 SHAW, Silurian ostracodes R. W. L. SHAW: SILURIAN OSTRACODA 603 Genus nodibeyrichia Henningsmoen 1954 Nodibeyrichia scissa Martinsson 1962 Plate 112, figs. 1, 2 1962 Neobeyrichia {Nodibeyrichia) scissa Martinsson, p. 323, fig. 180. Remarks. This species is distinguished by the division of a fairly broad syllobium into cuspidal, median, and ventral lobules. The ventral lobule is connected weakly to the preadductoral lobe which fails to reach the hinge line. The anterior lobe is particularly distinctive being divided by an oblique sulcus into an anteroventral and dorsal lobule. The sulcus does not completely divide the anterior lobe, but reaches the greater part of the way across it. The velum is quite thick and bears some evidence of the occurrence of small spines. The ventral surface of the velum is ornamented by a toric ridge. No hetero- morphic specimens of this species have yet been identified from Lake District material. In these characters it closely resembles the type material described and figured by Martinsson (1962, pp. 323 and 324, fig. 180). Slight differences may be noted in the ornamentation which in the type material occurs as a fairly coarse tuberculation of the syllobium. This has not been noted in the Lake District material. Occurrence. In Gotland, Nodibeyrichia scissa occurs in the ‘Hemse’ Beds where Martinsson (1967, p. 371 ) regards it, in association with Neobeyrichia laiiensis, as part of a fauna which correlates with the Leintwardinian. N. scissa is characteristic of the Lower Underbarrow Flags, though it is not very common (13%). N. cf. scissa is reported to occur in the Upper Leintwardine Beds near Leintwardine (Martinsson 1967, p. 371). Subfamily kloedeniinae Ulrich and Bassler 1923 Genus frostiella Martinsson 1963 Frost ielJa groenvalliana Martinsson 1963 Plate 113, figs. 1-5, 7 1963 Frostiella groenvalliana Martinsson, p. 29, figs. 7c, 8, 14-17. Description. Hinge straight, shorter than the maximum length of valve, outline of the valve rounded, and preplete. Lobes quite strongly defined; anterior lobe and syllobium are connected ventrally by a ventral lobal body; preadductoral lobe connected to this by a shallow col. Ornamentation generally lacking, but shell pitted. Preadductoral lobe bears a cristal loop which extends down its dorsal face. Within this loop there is some evidence of striate ornamentation. Heteromorph distinguished by swelling of the anteroventral part of ventral lobal body. Striate area occurs between velar ridge and margin of valve. Remarks. The relationship of these specimens to F. groenvalliana is fairly clear. There are, however, some differences including an increased tumidity of the lobation and the striate area on the ventral face of the crumina is rather broad. In these characters, an approach to F. lebiensis Martinsson is seen. Occurrence. F. groenvalliana is definitive of the Scout Hill Flags of the Lake District succession (51%). It is also recorded from Bed 4 at Oved-Ramsdsa, and in the Downton Castle Sandstone group. This species therefore, is seen to be indicative of strata of Downtonian age. 604 PALAEONTOLOGY, VOLUME 14 Superfamily drepanellacea Ulrich and Bassler 1923 Family aechminidae Boucek 1936 Genus aechmina Jones and Holl 1869 Aechmina sp. Plate 112, fig. 5 Remarks. Specimens are rare and usually poorly preserved. Those which have been found do not permit specific identification. The shape of the valve is more quadrate than that of A. cuspidata Jones and Holl (1869), which is rounded and extended pos- teriorly. In shape it more closely resembles A. molengraajfi Botke (1916). Unfortunately the spine in the Kirkby Moor Flag material is not well preserved. In all cases it is broken off near the base. It is clear, however, that it is rather narrower than that of A. molen- graajfi. Occurrence. Aechmina sp. has been found only in the Kirkby Moor Flags where it is rare (< 5°o)- Superfamily hollinacea Swartz 1936 Family ?hollinidae Swartz 1936 Gen. A sp. Plate 112, figs. 6, 7 Description. Hinge straight, shorter than maximum length of valve. Slightly amplete. Shallow adductoral sulcus extends from middle of hinge to a pit in centre of valve. Right valve slightly smaller than left. Surface lacks lobes and ornamentation. Some specimens, particularly the internal moulds, show the presence of a small preadductoral lobe. Velum forms a weak shelf. Marginal frill present. Remarks. The above description permits the possible assignment of the fossil to the family Hollinidae. Its relationship to the members of that family is unknown so the present material is referred to as gen. A sp. Occurrence. This material has been recorded from the Upper Underbarrow Flags (< 5%), the Kirkby Moor Flags (9%) and the Scout Hill Flags (13%). EXPLANATION OF PLATE 112 Figs. 1, 2. Nodibeyrichia scissa Martinsson. Lower Underbarrow Flags, G.R. 4703 9043, x40. Rubber casts of left valves. 1, GSM Z19064; 2, GSM Z19065. Fig. 3. Neobeyrichia nutans (Kiesow), rubber cast of left valve, GSM Z19062, Lower Underbarrow Flags, G.R. 4703 9043, x 30. Fig. 4. Amygdalella sp., right valve, GSM Z19046, Kirkby Moor Flags, G.R. 5642 8774, X40. Fig. 5. Aechmina sp., internal mould of left valve, GSM Z19077, Kirkby Moor Flags, G.R. 5824 8500, x40. Figs. 6, 7. Genus A sp., Kirkby Moor Flags, G.R. 5642 8774, x40. 6, right valve; 7, ventral view of same, GSM Z19043. Figs. 8-10. Cavellina circulata Neckaja, Kirkby Moor Flags, G.R. 5642 8774, X40. 8, right valve, GSM Z19047; 9, left valve overlapped by right, same specimen; 10, dorsal view of same specimen. Palaeontology, Vol. 14 PLATE 112 SHAW, Silurian ostracodes R. W. L. SHAW; SILURIAN OSTRACODA 605 Superfamily primitiopsacea Swartz 1936 Family primitiopsidae Swartz 1936 Subfamily leiocyaminae Martinsson 1956 Genus amygdalella Martinsson 1956 Amygdalella sp. Plate 112, fig. 4 Remarks. Only one unweathered specimen of this genus has been found. Moulds are more common and the identification of a specimen with the shell intact pennits their tentative assignment to the genus Amygdalella. The chief character the valve displays is the straight hinge line which is overhung by the dorsal part of the valve. It is roughly symmetrical but there is an indication that a dimorphic pouch is present at the posterior end of the valve. In these characters the specimen clearly belongs to the genus Amygda- lella. Occurrence. The above specimen is described from the lower part of the Kirkby Moor Flags, Old Hutton, near Kendal. It has not been identified in strata other than Kirkby Moor Flags (< 5°o). Suborder metacopina Sylvester-Bradley 1961 Superfamily healdiacea Harlton 1933 Family CAVELLmiDAE Egorov 1950 Genus cavellina Coryell 1928 Cavellina eireulata Neckaja 1958 Plate 112, figs. 8-10 1958 Cavellina eireulata Neckaja, p. 360. Remarks. The right valve is large and overlaps the left valve. The shell is strongly arched dorsally especially towards the anterior end. The right valve is slightly pointed anteriorally but posteriorally is more truncated. The surface of the valve is smooth except for a shallow groove running along ventral margin of left valve. Internals show a slight sulcation and a muscle spot. Occurrence. Underbarrow Flags (< 5%), Kirkby Moor Flags (47%), and Scout Hill Flags (41%) in the Kendal area. It is common in both the Kirkby Moor and Scout Hill Flags. Witwicka (1967) reports the species from the Lebork, Weyherowo, and Clapowo cores in Poland. The type material comes from the ‘Ludlovian’ of the U.S.S.R. Family bairdiocyprididae Shaver 1961 Genus cytherellina Jones and Holl 1869 Cytherellifia siliqua (Jones 1855) Plate 113, fig. 6 1855 Beyrichia siliqua Jones, p. 90, pi. 5, fig. 22. Remarks. This species is oval, elongate, rounded at the hinge, and straight along the ventral margin. The surface of the shell is smooth and unornamented. Internally the shells are thickened in two areas extending from the dorsal to the ventral margin. 606 PALAEONTOLOGY, VOLUME 14 Occurrence. C. siliqua extends throughout the Underbarrow Flags (7 and 19%), the Kirkby Moor Flags (52%) and the Scout Hill Flags (36%). Cytherellina cf. siliqua Plate 113, fig. 8 Remarks. No externals of this form have been found but the internal moulds are highly characteristic. The general shape of the valves is the same as in C. siliqua but the internal sulcation is markedly different. The valves are not traversed by two sulcae but at the anterior end a step or shelf occurs pointing anteriorally. The relationships of this species are unknown. Its general form is very much that of C. siliqua but in the quite striking differences internally it is distinct from that species. However, for lack of abundant good material with which fully to understand this species, it must be referred to C. cf. siliqua. Occurrence. It is not found in the Underbarrow Flags and is rather rare in the Kirkby Moor Flags and the Scout Hill Flags. STRATIGRAPHICAL VALUE OF THE OSTRACODES In text-fig. 1 the stratigraphical distribution of the ostracodes in the Lake District succession is illustrated. While it is clear that a number of the species extend through the whole succession, there are a number of species which are useful in the division of the Lake District succession and in correlation both within this country and abroad. (o) Lower Underbarrow Flags. The ostracode fauna is characterized by the occurrence of Neobeyrichia nutans, N. lauensis, and Nodibeyrichia scissa. N. nutans and N. scissa are confined to these beds, while N. lauensis continues into the Upper Underbarrow Flags. The macro- and micro-fauna identified from the Lower Underbarrow Flags indicates an Upper Leintwardinian age when compared with the faunal lists of Holland, Lawson and Walmsley 1963. The part played in this by the ostracodes is the common occurrence of N. lauensis, in particular, and the tentative identification of N. scissa (Martinsson 1967). When the Lower Underbarrow Flag association, N. lauensis, N. scissa, and TV. nutans is compared with the successive ostracode faunas identified by Martinsson (1967) in Gotland, a useful correlation may be made. He identifies as Leintwardinian a TV. lauensis-N. scissa association. The Lower Under- barrow Flags as indicated above additionally include TV. nutans. This species, according to the faunas determined on Gotland (Martinsson 1967), is included in an assemblage somewhat older than the lauensis-scissa fauna; in Martinsson’s Fig. 2 it is drawn extending down into the upper part of the EXPLANATION OF PLATE 113 Figs. 1-5, 7. Frostiella groenvalliana Martinsson, Scout Hill Flags. 1, right valve, internal mould, GSM Z19068, G.R. 5585 8307, x40. 2, right valve, GSM Z19072, G.R. 6027 7958, x40. 3, right heteromorph valve, GSM Z19073, G.R. 6027 7958, x40. 4, dorsal view of fig. 3, showing cristal loop on preadductoral lobe, X40. 5, ventral view of fig. 3, showing ornamentation on ventral face of crumina, x 40. 7, left heteromorph valve, internal mould, GSM Z19074, G.R. 6027 7958, X 30. Fig. 6. Cytherellina siliqua (Jones), left valve, internal mould, GSM Z19076, Kirkby Moor Flags, G.R. 5434 9501, x40. Fig. 8. Cytherellina cf. siliqua (Jones), left valve, internal mould, GSM Z19031, Kirkby Moor Flags, G.R. 5854 9396, x40. Palaeontology, Vol. 14 PLATE 113 SHAW, Silurian ostracodes It i' • rr*.- “i,' % ■-C^ ■M !• at R. W. L. SHAW: SILURIAN OSTRACODA 607 Bringewoodian stage. It is evident, therefore, that in the Lake District succession its range is probably extended. The other species occurring in the Lower Underbarrow Flags are long-ranging extending through the remainder of the succession, with the exception of TV. torosa, which has not been found in the Upper Underbarrow Flags. (b) Upper Underbcirrow Flags. The Upper Underbarrow Flags are the lithological continuation of the Lower Underbarrow Flags, but are also the lateral equivalent of part of the Kirkby Moor Flags, the base of the Kirkby Moor Flags being diachronous from north-west to south-east. They are marked by the loss of TV. nutans and TV. scissa. TV. lauensis might have been expected to be lost at this point but its continuation into strata of Whitcliflfian age indicates an extended range in the Lake District. In support of this contention is the addition of some species which characterize the Kirkby Moor Flags. Underb Lower arrow Flags Upper Kirkby Moor Flags Scout Hilt Flags C cifculata C. siliqua Hemsiella sp L. cf scanensis N lauensis N nutans N. scissa N torosa indet. smooth ostracodes Genus A sp H maccoyiana N confluens Pnmitiopsis sp Aechmina sp Amygdalella sp C cf siliqua H contracta M salterianum F. groenvalliana — TEXT-FIG. 1. Stratigraphical distribution of ostracodes. The most significant ostracodes occurring in the Upper Underbarrow Flags are Hemsiella maccoyiana and Neobeyrichia confluens, whose association is characteristic of the Kirkby Moor Flags. They illustrate the faunal change occurring at the Upper Leintwardinian-Whitcliffian boundary in the Lake District. As mentioned above the Upper Underbarrow Flags, in addition to including these new species, retains TV. lauensis. It is important, however, in such circumstances to regard more highly the appear- ance of new forms into succession rather than the continued occurrence of older species. (c) Kirkby Moor Flags. The Kirkby Moor Flags are a lithologically distinct division. They mark the influx of thick units of well-sorted silts in which the fauna occurs in discrete lenses of restricted vertical and lateral extent; outside these lenses fossils do not normally occur. This distinctive lithological sub-division is marked by the appearance of species of ostracodes new to the Lake District succession. The more important include Fluttoniella contracta and Macrypsilon salterianum. The species particularly common in this division is TV. confluens. As mentioned above in the discussion of the Upper Underbarrow Flags, these contain TV. confluens which is important in demonstrating faunally the lateral equivalence of the Upper Underbarrow Flags and the lower part of the Kirkby Moor Flags. In addition this species is occasionally found in the lowest horizons of the Scout Hill Flags. However, in the overwhelming number of cases, the existence of TV. confluens is indicative of Lake District strata which correlate with the Whitcliffian stage. 608 PALAEONTOLOGY, VOLUME 14 The identification of H. contracta is significant not only from the point of view of its value in correlation but that it also emphasizes the progressive faunal changes occurring through the succession since its appearance in the succession is at the commencement of the Kirkby Moor Flags. In correlation the ostracode fauna of the Kirkby Moor Flags is rather problematical. It does not readily fall into the ostracode faunal succession identified in Gotland by Martinsson (1967) and its correlation with the type succession at Ludlow is restricted by the general lack of variation of ostracode species occurring there. Martinsson (1967, pp. 374-376) has recognized two ostracode faunas which he correlates with the Whitcliffian. These are the Cryptolopholobus-Jiiviella-Neobeyrichia regnans fauna and the Hemsiella maccoyiana-Neobeyrichia regnans fauna. None of the species characteristic of the first of these faunas has been identified in the Kirkby Moor Flags. The second fauna, however, is more promising, since H. maccoyiana has been recognized and this is regarded as a diagnostic element in the Baltic. In addition Macrypsilon salterianiim and Amygdalella sp. occur and, although rare, are important in that they constitute a part of the H. maccoyiana-N. regnans assemblage. It is clear that the correlation of these Whitcliffian Kirkby Moor Flags with the faunas of Gotland is rather difficult. At present there is a lack of species in common to both the Lake District and Gotland to suggest a firm correlation. Lake District species unknown from Gotland include Huttoniella contracta, Neobeyrichia confluens, and N. torosa. It appears that during the Whitcliffian conditions had diverged sufficiently for the faunal provinces to develop relatively independently. The ostracodes yielded by the Whitcliffian of the type area near Ludlow are rather few in variety. They include species of Hemsiella, Neobeyrichia torosa, Lophoctonella sp., Cytherellina si/iqua, and a number of smooth indeterminate forms. It is clear from this faunal list that more work is required on the Whitcliffian ostracodes until which time it may only be stated that correlation of the Upper Underbarrow Flags and Kirkby Moor Flags, on the basis of ostracodes alone, is tenuous. Using the macro- and the micro-fauna together, however, this correlation may be made. The basis for the correla- tion is in having demonstrated the existence of strata of Upper Leintwardinian age below (the Lower Underbarrow Flags) and the occurrence of rocks of Downtonian age (the Scout Hill Flags) above. The strata which lie between must therefore be Whitcliffian. This is essentially the argument upon which the type Whitcliffian was erected since no fossil found in that stage identified it as such. The Upper Leintwardine/Whitcliffe boundary is clearly documented at the loss of a readily recognizable assemblage. This is precisely the faunal change occurring at the upper limit of the Lower Underbarrow Flags with the exception of the extension of N. lauensis through the Upper Underbarrow Flags. The upper limit of the Kirkby Moor Flags is defined at the appearance of Frostiella groenvalliana, an important member of the fauna succeeding the Ludlow Bone Bed. It is on this evidence that a correlation is made between the Upper Underbarrow and Kirkby Moor Flags and the Whitcliffian Neobeyrichia confluens and the trilobite Acastella prima form a diagnostic association in the Lake District but are of no value in correlation with the Welsh Borderland where neither occurs and a definitive assemblage is badly needed. {d) Scout Hill Flags. The commencement of the Scout Hill Flags is not marked by dramatic changes in the ostracode fauna. As mentioned above Neobeyrichia confluens is lost at the base of this division, being found only on very rare instances in the lowest beds. The most marked change is in the gain of Frostiella groenvalliana. This is a particularly valuable ostracode as a means of correlation and is an easily recognizable species. At Ludlow, Frostiella groenvalliana, appears above the Bone Bed, together with Londinia arisaigensis. The latter is closely related to Londinia kiesowi which together with the former is regarded as a Downtonian assemblage (Martinsson 1963, 1964, and 1967). Frostiella groenvalliana appears in the Lake District for the first time in the Scout Hill Flags, which, by this token, may be regarded as Downtonian age. The Scout Hill Flags can be correlated with the type Siluro-Devonian succession at Ludlow. There, above the base of the Ludlow Bone Bed, F. groenvalliana and Londinia arisaigensis are introduced into the succession (Shaw 1969) at the commencement of the Downtonian. This clearly correlates with similar faunal changes at the base of the Stonehouse Formation (Nova Scotia) and in the Baltic area. In the vicinity of Llandeilo, south of Ludlow, the Long Quarry Beds which are considered to be Downtonian (Potter and Price 1965) contain Frostiella groenvalliana which, as in the Scout Hill Flags, occurs in the absence of Londinia. The value of this horizon, i.e. the introduction of Frostiella and Londinia of the arisaigensis-kiesowi R. W. L. SHAW: SILURIAN OSTRACODA 609 type into the succession is of considerable importance in that it permits a ready correlation of strata with the Siluro-Devonian boundary near Ludlow. In addition it has been shown that this horizon can be correlated with some accuracy using both ostracodes and trilobites with the graptolite sequence recorded on the Continent (Shaw 1969). The results of the correlation demonstrate the equivalence of the Ludlow Bone Bed horizon approximately with the e/Sl-e/32 boundary of the sequence in Bohemia. With present information the closest this correlation may be made is with the Monograptus iiltimus zone, though no doubt with further research into the faunas both of the type section at Ludlow and of the continental sections this correlation could be rendered more precise and will continue to provide the correct and most workable horizon for the Siluro-Devonian boundary. LUDLOW LAKE DISTRICT OSTRACODE ZONES (KENDAL AREA) (MARTINSSON) z < z o DOWNTONIAN Temeside Shales > UJ Q Oownlon Castle Sandstone Scout Hill Flags FrostieUa groenvalliana & Londinia kiesowi Ludlow Bone Beds 1 1 1 1 1 1 1 1 1 WHITCLIFFIAN Whitcliffe Beds Kirkby Moor Flags Hemsiella maccoyiana Neobeynchia regnans z < ct: 1 1 1 1 ! Upper Underbarrow Flags Neobeyrichia regnans Juviella juviensis Cryptopholobus semilaqueafus 3 _i (/) ii 1 1 1 1 LEINTWARDINIAN Upper Lemtwardine Beds Lower Under barrow Flags Neobeynchia lauensis Nodibeyrichia scissa 1 1 1 1 1 Lower Leintwardine Beds Barrrrisdale Slates TEXT-FIG. 2. Correlation table for the Lake District succession near Kendal. CONCLUSIONS 1. The ostracodes of the Lake District succession, although rather difficult to use systematically, yield important and useful results for the division of the succession and its correlation. 2. The Lower Underbarrow Flags and the Scout Hill Flags yield faunas which can be correlated with the ostracode faunas on Gotland which Martinsson (1967) correlates with the Leintwardinian and the Downtonian respectively. 3. The Upper Underbarrow Flags and the Kirkby Moor Flags are less easily cor- related directly with Martinsson’s ostracode faunas. There is some evidence, however, that the upper part of the Kirkby Moor Flags correlates with the upper of the two ostracode faunas which Martinsson correlates with the Whitcliffian. 610 PALAEONTOLOGY, VOLUME 14 4. The Upper Underbarrow Flags and the Kirkby Moor Flags contain two new species of ostracode, Huttoniella contracta and Neobeyrichia confluens. The former extends through the Scout Hill Flags but the latter is characteristic of the lower beds. 5, The presence of Frostiella groenvalliaua, defining the Scout Hill Flags, demonstrates the existence of strata of Downtonian age in the Lake District. This was suggested in the early history of the study of Lake District geology although it was not then sup- ported by fossil evidence (Marshall 1839; Sedgwick 1845, 1852, and 1859; Aveline, Hughes, and Tiddeman 1872; and Aveline, Hughes, and Strahan 1888). Acknowledgements. I wish to thank Dr. J. Shirley for his help and encouragement during the prepara- tion of this work. Dr. A. Martinsson has been most generous in his assistance with the identification of ostracodes and has kindly critically read the manuscript. The work was conducted under the tenure of a Studentship from the Shell International Petroleum Company Limited. REFERENCES AVELINE, w. T. and HUGHES, T. MCK. 1872. The geology of the country around Kendal, Sedbergh, Bowness and Tebay. Explanation of Quarter Sheet 98 N.E. Mem. geol. Surv. G.B. and TIDDEMAN, R. H. 1872. The geology of the neighbourhood of Kirkby Lonsdale and Kendal. Explanation of Quarter Sheet 98 S.E. Mem. geol. Surv. G.B. and STRAHAN, A. 1888. The geology of the country around Kendal, Sedbergh, Bowness and Tebay. Explanation of Quarter Sheet 98 N.E. 2nd Ed. Mem. geol. Surv. G.B. COPELAND, M. J. 1964. Canadian fossil Qstracoda: some Silurian Species. Bulk geol. Surv. Canada, 118, 1-25. FURNESS, R. R., LLEWELLYN, p. G., NORMAN, T. N. and RICKARDS, R. B. 1967. A review of Wenlock and Ludlow Stratigraphy and sedimentation in N.W. England. Geol. Mag. 104, 132-147. HENNINGSMOEN, G. 1953. Classification of Palaeozoic straight hinged ostracodes. Norsk, geol. Tidsskr. 31, 185-288. 1954. Silurian ostracodes from the Oslo Region Norway, 1, Beyrichiacea with a revision of the Beyrichiidae. Norsk, geol. Tidsskr. 34, 15-71. HOLLAND, c. H., LAWSON, J. D. and WALMSLEY, V. G. 1963. The Silurian rocks of the Ludlow District, Shropshire. Bull. Br. Mus. nat. Hist. {Geol.), 8 (3), 93-171, 7 pis. JONES, T. R. 1855(3. Notes on some Palaeozoic bivalved Entomostraca. No. 1. Some species of Beyrichia from the Upper Silurian Limestone of Scandinavia. Ann. Mag. nat. Hist. [2], 16, 81-92, pi. 5. lS55b. Notes on some Palaeozoic bivalved Entomostraca. No. 2. Some British and foreign species of Beyrichia. Ann. Mag. nat. Hist. [2], 16, 163-176, pi. 6. — — and HOLE, H. B. 1869. Palaeozoic bivalved Entomostraca. No. 9. Some Silurian species. Ann. Mag. nat. Hist. [4], 4, 211-229, pis. 14, 15. KiESOW, J. 1888. Ueber die gotlandische Beyrichien. Z. dt. geol. Ges. 40, 1-16, 2 pis. KOLMODIN, L. 1869. Bidrag till kdnnedomen om Sveriges siluriska ostracoder. Uppsala. 22 pp., 1 pi. MARSHALL, J. G. 1839 (publ. 1840). Section from the Shap Granite to Casterton Fell. Rep. Br. Z5.?. Advmt. Sci. 9th Meeting, 67. MARTINSSON, A. 1962. Ostracodes of the family Beyrichiidae from the Silurian of Gotland. Bull. geol. Instn. Univ. Upsala, 41, 1-369. 1963. Kloedenia and related ostracode genera in the Silurian and Devonian of the Baltic area and Britain. Bull. geol. Instn. Univ. Upsala, 42, 1-63. 1964. Palaeocope ostracodes from the well Leba 1 in Pomerania. Geol. For. Stockli. Fork. 86, 125-161. ■ 1967. The succession and correlation of the ostracode fauna in the Silurian of Gotland. Geol. For. Stockh. Fork. 89, 350-386. NECKAJA, A. I. 1958. [New species and genera of Qrdovician and Silurian ostracodes from the north- western Russian Platform. Microfauna of the USSR, IX.] Tr. VNIGRI, n.s. 115, 349-373. [In Russian.] R. W. L. SHAW: SILURIAN OSTRACODA 611 POTTER, j. F. and PRICE, j. H. 1965. Comparative sections through rocks of Ludlovian-Downtonian age in the Llandovery and Llandeilo Districts. Proc. Geol. Ass., Lend. 76, 379-401. SEDGWICK, A. 1842. Three letters on the geology of the Lake District, in A complete guide to the scenery of the Lakes of England by W. Wordsworth. 1st Ed., Keswick. ■ 1845. On the classification of the fossiliferous slates of Cumberland, Westmorland and Lanca- shire. Q. Jl. geol. Soc. Lond. 2, 106-131. 1852. On the classification and nomenclature of the Lower Palaeozoic rocks of England and Wales. Q. Jl. geol. Soc. Lond. 8, 136-168. 1859. Supplementary letter on the geology of the Lake District, in A complete guide to the scenery of the Lakes of England by W. Wordsworth. 5th Ed., Keswick. and MCCOY, F. 1855. Britsh Palaeozoic Eossils. Cambridge. SHAW, R. w. L. 1 969. Beyrichiacean ostracodes from the Downtonian of Shropshire. Geol. Eor. Stockh. Eorh. 91, 52-72. 1971. The faunal stratigraphy of the Kirk by Moor Flags of the type area near Kendal, Westmor- land. Geol. J. 1 (2), 359-380. WITWICKA, E. 1967. Malzoraczki z gornego syluru Polski. Kwart. Geol. 11, 39-61, 3 pis. R. W. L. SHAW Anglo-American Corp. (Australia) Ltd., T. and G. Building Queen Street, Brisbane 4000, Revised typescript received 20 July 1970 Queensland, Australia. THE TRILOBITE PLIOMERINA CHUGAEVA FROM THE ORDOVICIAN OF NEW SOUTH WALES by B. D. WEBBY Abstract. Two new pliomerid species, PHomerina austrina and P. prima, are described from Ordovician lime- stones of New South Wales. The well-preserved, silicified material reveals new information about the nature of the genus and its relationships. Representatives of PHomerina have been described and illustrated previously only from China, Korea, and Kazakhstan. The Australian and Asian occurrences suggest a close faunal con- nection during Gisbornian and Eastonian (about Lower-Middle Caradocian) time. With the possible exception of fragmentary asaphids, the most common trilobites in the Ordovician limestones of central-western New South Wales are pliomerids identified as belonging to the genus PHomerina Chugaeva. While other elements of the fauna exhibit North American relationships, PHomerina establishes a linkage with Asian faunas (Packham 1967). The type species, PHomerina marteUii (Reed), is described from the Upper Llanvirnian? of Yunnan, China (Reed 1917), and a form seeming to bear affinities to it is recorded from the Tsiubon Limestone (Lower Caradocian) of South Korea (Kobayashi 1934). Three species have been described by Chugaeva (1958) from the Ordovician of Kazakhstan, U.S.S.R., P. sulcifrons (Weber) from the Anderken horizon (Lower Caradocian), P. unda Chugaeva from the Otar horizon (Middle Cara- docian), and P. dulanensis Chugaeva from the Dulankarin horizon (uppermost Cara- docian to lowest Ashgillian). Whittington (1966) has reported the genus from the Gordon Limestone of Tasmania, and has taken it to be a representative of the Caradocian fauna. The Ordovician successions containing PHomerina and other trilobites in central-western New South Wales appear to be broadly of Caradocian (Gisbornian-Eastonian) age. The trilobite-bearing beds seem to be of post-Nemagraptns gracilis age, and may be as young as the Zone of Dicranograptus hians (i.e. Upper Eastonian). For discussion of age and correlation of the deposits, see Packham (1969), Webby (1969), and Moors (1970). In the lower part of the Cliefden Caves Limestone, PHomerina prima sp. nov., and unidentified asaphids are the most abundant trilobites. Less common forms include Remopleurides sp., Eobronteus sp., and a trinucleid, probably a species of ParkesoHthiis Campbell and Durham. In the upper part of the Cliefden Caves Limestone, PHomerina austrina sp. nov., ParkesoHthus sp., and fragmentary asaphids are represented. The overlying Malongulli Formation at Trilobite Hill and nearby Copper Mine Creek has produced MalonguHia oepiki Webby, Moors, and McLean, Encrin- uraspis optimus Webby, Moors, and McLean, ParkesoHthus sp., Remopleurides sp., Eobronteus sp., and asaphid fragments. Encrinuraspis optimus has also been recorded from the Malongulli Formation of the Regan’s Creek area (Webby, Moors, and McLean, 1970). The Bowan Park Limestone contains PHomerina prima sp. nov. in the lower part, P. austrina sp. nov., ParkesoHthus sp., Amphilichas sp., an encrinurid, possibly an Encrinuraspis, and fragmentary asaphids in the middle part, and a trinucleid, an illaenid, and indeterminate asaphids at the top of the succession. A species of ParkesoHthus has also been found by Mr. M. Tuckson in shales (possibly equivalent in age to the Malongulli Formation) north of Cheeseman’s Creek, 16 miles west of Orange. [Palaeontology, Vol. 14, Part 4, 1971, pp. 612-622, pis. 114-115.] B. D. WEBBY: PLIOMERINA FROM NEW SOUTH WALES 613 Ordovician limestones at Billabong Creek have yielded an abundant silicified fauna including Pliomerina austrina, and less common Remopleurides sp., Amphilichas sp., Sphaerocoryphe sp., Parke- solithus sp., an encrinurid, probably an Encriniiraspis, and fragmentary asaphids. Campbell and Durham (1970) have described Parkesolitfuis gradyi from unnamed siltstones overlying the limestone near Billabong Creek. Of these forms, MalonguUia, Encrinuraspis, and Parkesolithiis seem to be endemic genera to Australia. Pliomerina has a restricted distribution in Australia and Asia (text-fig. 1), and Remopleurides, Sphaerocoryphe, Eobronteus, and Amphilichas appear to have more cosmopolitan relationships, since they are found also in North America, TEXT-FIG. 1. Map showing known occurrences of the genus Pliomerina, and possible limits of the Pliomerina fauna in Middle-Upper Ordovician time. Europe, and Asia. The New South Wales fauna thus seems to comprise an admixture of Whittington’s (1966) Caradocian ‘northern region’ fauna (with Remopleurides, Sphaero- coryphe, and Amphilichas) and his ‘southern region’ or ‘Encrinurella' fauna (mthPliomer- ina). Since the genus Encrinurella Reed has not been confirmed in Australian ‘Caradocian’ successions, it is suggested that the faunal province bearing its name, extending from Kazakhstan, China, South Korea, Thailand, North Vietnam, and possibly the Hima- layas to New South Wales and Tasmania, be renamed the Pliomerina fauna (text-fig. 1). Pliomerids have been reported from a number of other parts of Australia, and seem to indicate that the family was well represented in the region throughout the Ordovician. s s C 8385 614 PALAEONTOLOGY, VOLUME 14 The genus Protopliomerops Kobayashi is recorded from the Digger Island Formation (Lancefieldian or Tremadocian) at Waratah Bay, Victoria (Singleton, in Lindner 1953), and a species of Pliomerops Raymond is listed as coming from the Caroline Creek Sandstone and the Florentine Valley Mudstone (Middle-Upper Arenigian) of Tasmania (Banks 1962). Undescribed pliomerids have also been mentioned as occurring in the Ordovician section of Samphire Marsh No. 1 Well on the southern side of the Canning Basin, Western Australia (Gilbert-Tomlinson 1961). These have been considered by Gilbert-Tomlinson to have a late Tremadocian-Arenigian age. Others have been reported from Western Australia, in the upper part of the Emanuel Limestone and in the Gap Creek Dolomite near Fitzroy Crossing, in beds broadly ranging from Llanvirnian to Caradocian in age (Guppy and Opik 1950), and from the Lower Ordovician of the Cambridge Gulf area (Opik 1957). SYSTEMATIC DESCRIPTIONS Family pliomeridae Raymond 1913 Genus pliomerina Chugaeva 1956 (= Pliomeraspis Harrington 1957) Type species. PUomera mortellii Reed 1917. Diagnosis. Glabella forwardly expanding, with large frontal lobe and three lateral lobes ; lobe 3p longer (exsag.) than lobes Ip or 2p; lateral furrow 3p intersecting preglabellar furrow at or just in front of antero-lateral angle of glabella; lateral parts of anterior border exhibited on dorsal side but flexed to slope downwards and backwards sagittally; relatively broad (tr.), trapezoidal, posteriorly narrowing rostral plate, with maximum width slightly more than one-half greatest width of glabella; eye lobe prominent, situated with mid-point on about same transverse line as lateral furrow 2p, only short distance out from axial furrow ; genal angle in mature stage rounded ; hypostome typically pliomerid with rounded, tongue-like posterior margin; ten thoracic segments in one species; pygidium of Pliomerella type with axis of five rings and large terminal piece; five pairs of pleurae with moderately blunt, rounded ends and deeply impressed, backwardly curving interpleural furrows; tips of pleurae fused and curved under to form doublure. Discussion. Reed (1917) based the description of PUomera martellii on two incomplete cranidia from the Llanvirnian? of Pupiao, Yunnan, China, observing that the true affinities of the species were somewhat uncertain. He noted the unusual feature of the species, namely the occurrence on the frontal lobe of ‘a pair of small marginal notches on its anterior edge’. Chugaeva (1956, 1958) designated Reed’s species the type species of her new genus Pliomerina. She emphasized the occurrence in the frontal part of the glabella of the two ‘prefrontal furrows’ (Reed’s ‘marginal notches’) which separate the two small, lateral ‘prefrontal lobes’ from the much larger, medial frontal lobe. It is now clear from study of well-preserved, silicified Australian material that Chugaeva was incorrect in interpreting the lateral ‘prefrontal lobes’ as part of the glabella. In fact the ‘prefrontal furrows’ prove to be lateral portions of the preglabellar furrow, and the ‘prefrontal lobes’, lateral portions of the anterior border. However, despite her misinter- pretation, the genus Pliomerina remains valid and distinct from all other pliomerids. B. D. WEBBY: PLIOMERINA FROM NEW SOUTH WALES 615 The cephalon of Pliomerina is perhaps nearest to that of Pliomerops, a widespread genus from the Middle-Upper Ordovician of North America, Europe, and Asia. The type species, P. canadensis (Billings), however, has a relatively narrower, straight-sided glabella, a relatively much narrower (tr.) and longer (sag.) rostral plate, a much smaller frontal lobe with lateral furrow 3p usually further forward, and it has eye lobes placed considerably further out from the axial furrow (Whittington 1961; Shaw 1968). Furthermore, Pliomerops has a quite different thorax with 14-19 segments, and pygidium with a short, triangular terminal axial piece. The genus Pliomerella Reed (type species P. serotina Reed) from the Middle-Upper Ordovician of Scotland, Virginia, and north-eastern U.S.S.R. has a similar pygidium to Pliomerina with five axial rings, a large sub-rectangular to slightly tapering terminal axial piece and five pleurae (Cooper 1953). However, the glabella of Pliomerella is markedly different, with only two pairs of lateral furrows (Reed 1941), and the thorax consists of 19 segments (Cooper 1953). A C TEXT-FIG. 2. Diagrams of the cephalon of the holotype of Pliomerina austrina sp. nov. ; a, dorsal view, B, anterior view, c, ventral view; X 6. Note inferred position of hypostome (lightly stippled and outlined by dashed lines) in anterior and ventral views. Pliomerina austrina sp. nov. Plate 114, figs. 1-28; Plate 115, figs. 1-12; text-fig. 2 Material. Holotype (SUP 22940) and thirty-seven paratypes (SUP 22941-49, 23900-06, 23921-41) from Ordovician limestone at Billabong Creek. Also paratypes from the upper part of the Cliefden Caves Limestone at Trilobite Hill (SUP 13908) and Licking Hole Creek (SUP 23943), and from the middle part of the Bowan Park Limestone at Quondong (SUP 23907). Registration numbers of specimens in the palaeontological collections of the Department of Geology and Geophysics, University of Sydney. Diagnosis. Species of Pliomerina with relatively broad glabella ; lateral furrows extending inwards to slightly more than one-third of glabella width (tr.); furrow 3p intersects preglabellar furrow well in front of antero-lateral corner of glabella; lobes lp-3p subequal in length (exsag.) at inner ends of furrows; mesial tubercle on occipital ring usually absent, though may be developed as tiny tubercle on small individuals; bluntly rounded and fused outer ends of pleurae on pygidium; dorsal surface of cephalon, thorax, and pygidium, and ventral surface of hypostome smooth to irregularly finely granulated. 616 PALAEONTOLOGY, VOLUME 14 Description. Cephalon sub-semicircular in outline, transversely and sagittally gently convex. Glabella sub-pentagonal; widest across lateral lobes 3p, narrows gradually backwards to occipital ring, de- limited by deep, relatively broad axial furrows, and narrows more abruptly forwards. Occipital furrow deep, running in gentle curve convex forward. Axial furrows and occipital furrow form broad, raised ridges on undersurface. Three relatively narrow lateral glabellar furrows, not deeply impressed except at outer ends, extending to one-third or just more than one-third of glabella width; furrow Ip trans- versely directed with very slight forward curve at inner end; furrow 2p transversely directed with slightly backwardly curved inner end; and furrow 3p, which commences at preglabellar furrow a short distance in front of antero-lateral corner of glabella, inclined inward and backward at angle of about 50° to axis. Lobes Ip, 2p, and 3p subequal (exsag.) at inner ends; however, lobes Ip and 3p expand laterally, giving club-shaped form, whereas lobe 2p has almost straight sides or even tapers slightly laterally; consequently, towards outer ends, lobe 3p is longer (exsag.) than lobe Ip, and lobe Ip is longer than lobe 2p. Lateral glabellar furrows form only shallow ridges on undersurface, dying out inwards (PI. 114, fig. 6; PI. 115, fig. 8). Preglabellar furrow shallow except at intersection with axial furrow, in anterior view (PI. 1 14, fig. 3 ; text-fig. 2b) its course curves gently down to mid-line from more sharply up-arched lateral ends; in large specimen (PI. 115, fig. 2), smooth downward curve of pre- glabellar furrow disrupted by sharp nick a short distance away from mid-line, giving rise to very short, tongue-like mesial extension of frontal lobe. Frontal lobe relatively large, slightly longer (sag.) than total projected length (sag.) of lateral lobes; slightly wider (tr.) than long (sag.). Anterior border broader than glabella, and short (exsag.); longer laterally than mesially; bounded laterally by anterior end of axial furrow and anterior branch of facial suture; only outer parts exhibited on dorsal surface, arched downward and backward sagittally (text-fig. 2c). Prominent anterior boss situated on ridge- like undersurface of axial furrow at mid-point between front and back lateral extremities of anterior border (PI. 115, fig. 8). Rostral plate trapezoidal, short (sag. and exsag.), and having maximum width just more than one-half maximum width of glabella, narrowest at posterior margin; defined laterally by faint, short connective sutures and anteriorly by faint rostral suture; limited posteriorly by deeply impressed medial portion of hypostomal suture (PI. 1 15, fig. 6; text-fig. 2b-c). Cheek sub-triangular, gently convex, sloping moderately steeply outward and forward; posterior and lateral border furrows prominent. Fixed cheek L-shaped, narrowing anteriorly. Eye lobe relatively large, raised, subrounded, situated short distance out from axial furrow, with mid-point in about same transverse line as lateral furrow 2p\ length (exsag.) just less than one-third of length (sag.) of glabella; curved visual surface slopes steeply outward, presumably thin and consequently not pre- EXPLANATION OF PLATE 114 Figs. 1-28. Pliomerina aiistrina sp. nov. 1 , from upper part of Cliefden Caves Limestone, Trilobite Hill, 2-28, from Ordovician limestone at Billabong Creek. 1, dorsal view of relatively small specimen with incomplete cephalon, thorax of 10 segments, and pygidium; paratype SUP 13908, x5. 2-7, dorsal, anterior, oblique antero-lateral, ventral, ventral (reversed aspect), and posterior views of cephalon of holotype SUP 22940, X 5. 8-1 1 , dorsal, posterior, lateral, and anterior views of pygidium of paratype SUP 22942, x4. 12, dorsal view of incomplete cephalon and 7 thoracic segments; paratype SUP 22941, x5. 13-14, dorsal and ventral views of pygidium of paratype SUP 22943, x4. 15-16, anterior and ventral views of moderately large, incomplete hypostome; paratype SUP 23903, x4. 17-19, oblique, lateral, and ventral views of hypostome; paratype SUP 23904, X5. 20, dorsal view of small pygidium ; paratype SUP 23906, X 5. 21, oblique view of enrolled exoskeleton, showing incomplete cephalon and thorax; paratype SUP 23902, x 3. 22-24, anterior, posterior, and dorsal views of thoracic segment; paratype SUP 22946, X4. 25, view of undersurface of thoracic segment of paratype SUP 22946, X 8. Note panderian notch. 26, oblique posterior view of incom- plete pygidium showing cut-a-way with inturned posterior margin forming doublure; paratype SUP 22944, X 4. 27, ventral view of articulated thoracic segments of paratype SUP 22945, X 4. 28, detailed view of undersurface of thoracic segments of paratype SUP 22945, X 8 (anterior to right side). Note slot-like opening in inner part of pleura, raised anterior area and flat, posterior articulating facet in doublure of outer part of pleura. Palaeontology, VoL 14 PLATE 114 WEBBY, Pliomerina S' i’t t. % f J- B. D. WEBBY: PLIOMERINA FROM NEW SOUTH WALES 617 served; palpebral lobe moderately broad, slightly raised, bounded on inside by deep palpebral furrow which extends forward to intersect anterior part of suture obliquely, opposite lateral lobe 3p, and to rear curves sharply outward to meet posterior part of suture (PI. 115, fig. 7). Posterior border gently inclined and convex, becoming steeply inclined and expanding (exsag.) distally; posterior furrow deep and broad, becoming less strongly impressed distally, and eventually dying out just inside margin; genal angle rounded, in smaller specimens more sharply rounded; doublure extends inwards from genal angle to fulcrum beneath posterior border. Anterior branch of suture runs forward from eye lobe, intersects anterior end of axial furrow, then turns sharply inward and extends along outer edge of anterior border to junction with rostral and connective sutures, at a point approximately half the distance (measured around margin) from lateral end of anterior border to mid-line. Posterior branch of suture swings sharply outward on posterior side of eye lobe, intersects palpebral furrow, and then continues outward and backward across lateral border, reaching margin a short distance in front of genal angle. Lateral border moderately convex, sloping steeply, broad (as seen in lateral aspect), almost keel-like posteriorly, becoming narrower (in lateral view) with complementary increase in width of doublure anteriorly; deeply impressed antennular notch on doublure of anterior part of lateral border, with only short, curved flange of lateral border extending forward of antennular notch to abut against rostral plate; this flange together with rostral plate and sagittal portion of anterior border comprise anterior part of doublure (PI. 114, figs. 5-6; text -fig. 2c). Lateral furrow deeply impressed, extending from intersection with posterior branch of suture forward along lateral border, becoming confluent with anterior branch of suture between lateral and anterior borders. Portion of eye on free cheek includes broad, steeply inclined eye socle surmounted by visual surface; bounded by gently curved, firmly impressed furrow which extends forward and inward, confluent with anterior end of palpebral furrow at its point of junction with anterior branch of suture, and backward and inward to again come into confluence with palpebral furrow and posterior branch of suture (PI. 115, fig. 4). Occipital ring long sagittally, gently convex and slightly elevated above adjacent posterior borders; usually smooth (PI. 114, fig. 12; PI. 115, fig. 1) but in few, small specimens possesses tiny mesial tubercle (PI. 115, fig. 10); occipital furrow which curves convexly forward is not strictly con- fluent with posterior border furrow; posterior margin of occipital ring straight; doublure of occipital ring moderately long, almost reaching forward to occipital furrow (PI. 114, fig. 6); apodemes only weakly developed on inside of lateral ends of occipital furrow. Gently convex median body of hypostome, widest in line with anterior wings, narrows backward; posterior border almost flat passing anteriorly into gently convex, narrower lateral border with more deeply impressed border furrow which curves sharply outwards opposite anterior wing. Posterior margin smooth, tongue-like in outline; anterior margin weakly prow-shaped; weakly developed lateral bulges merge with median body just in front of anterior wings; anterior wings long, upwardly and slightly outwardly directed (upper ends poorly preserved); deep lateral notch directly behind anterior wing, with weak, angulate shoulder below, projecting into short, inward and backwardly directed posterior wing. One specimen shows fragment of hypostome attached by hypostomal suture to rostral plate and anterior end of lateral border (PI. 115, figs. 4 and 6), and rod-like structure extending downward and inward from anterior boss, presumably representing fused broken piece of anterior wing (PI. 1 15, fig. 5). Hypostome seems to be firmly linked by hypostomal suture to rostral plate and anterior part of lateral border. Thorax of ten segments in one specimen (PI. 114, fig. 1); width diminishing gradually backwards; one and three-quarter times length of cephalon ; axis about one-third total width of thoracic segment, transversely convex; defined by very deep, relatively broad, slot-like axial furrows; articulating furrow straight in vertical plane, moderately deep, narrow and rather sharply impressed, especially distally, with development of pair of relatively large apodemal pits; apodemes only weakly developed beneath. Articulating half ring moderately long sagittally, inclined upwards and forward on to what appears to be crest-like ridge, becoming shorter, and inclined obliquely forward and outward, and with ridge appearing to die out laterally; articulating half ring only projects short distance forward of ridge; ridge in close contact with doublure of axial ring. Axial ring consists of convex (sag. ) posterior part, separated from relatively flat (sag.) anterior part by shallow furrow, curving convexly backward; posterior part of axial ring widest and highest, and shorter sagittally than laterally; anterior part of axial ring becomes shorter (exsag.) laterally, and narrower (tr.) than posterior part; posterior margin of axial ring reflected as doublure, extending from posterior side slightly more than one-third distance forward 618 PALAEONTOLOGY, VOLUME 14 beneath axis. Posterior surface of axial ring flattened with relatively expanded, flat lateral areas and finely sculptured ridge arching across mid-line (on lower edge) and down to ring processes (on inner edges); confluent with posterior flanges of pleurae. From deep, slot-like axial furrow to fulcrum, pleura gently convex (exsag.) and gently inclined out- wards; undersurface flat and horizontal with wide, straight-sided cavity (deepest proximally); sharply flexed at fulcrum, with outer part of pleura moderately steeply inclined (about 30° from vertical); outer part of pleura gently convex (exsag.), curved gently forward (concave forward) and may be slightly expanded (exsag.), paddle-like; terminations rather blunt distally but with small, anteriorly directed tips. Doublure extends from distal end to fulcrum. Articulating ledges extend to fulcrum along lower, horizontal edge of pleura ; straight, transverse, ridge-like anterior flange and groove-like posterior flange. Ring process (posterior side) and socket (anterior side) situated on outer and lower edge of axial ring with adjacent, smaller axial socket (posterior) and process (anterior) on line with axial furrow. At fulcrum, fulcral process (anterior) and socket (posterior) developed, with smaller secondary process below fulcral socket, and complementary secondary socket below fulcral process. Similar articulating structures formed on posterior surface of cephalon and anterior end of pygidium. Straight hinge line with flanges and points of articulation on inner pleura and outer margin of axial ring remain- ing in contact on enrolment or in extended position. Pleural furrows not developed. Outer part of first thoracic segment during enrolment fits beneath posterior border and genal angle of cephalon (PI. 1 14, fig. 21). In succeeding segments break-in-slope on outer part of pleura between flat, anterior, sub-triangular articulating facet and more convex, outwardly tapering posterior region, with comple- mentary break-in-slope imprinted on doublure separating gently raised triangular-shaped anterior region from posterior articulating facet; outer, anterior articulating facet slides beneath inner posterior articulating facet of preceeding pleura (PI. 1 14, figs. 13 and 17), and break-in-slope acts as stop during enrolment. Outer parts of thoracic segments do not become exposed even in extended position owing to longer (exsag.) convex and facet-like areas. Keyhole-like panderian notch extends into doublure between raised anterior and posterior flanges of fulcral articulating structures (PI. 114, fig. 25); anterior flange (= anterior edge of panderian notch) does not appear to act as stop in enrol- ment. Pygidium sub-pentagonal, gently convex (strongly convex in region of postero-lateral margin); about equal to cephalon in length. Axis tapering posteriorly, with five axial rings, and large, slightly tapering terminal piece of length (sag.) equivalent to 4-5 axial rings; anterior end of terminal piece may exhibit faint, paired lateral bulges; backward taper of terminal piece gradual until near posterior margin, where apex narrowly rounded; anterior rings slightly bowed forwards. Axial furrow deeply impressed over most of length (exsag.) of pygidium but curves slightly inwards against terminal piece and dies out close to posterior margin. Ring furrows deep, rather sharply impressed. Five gently raised, unfurrowed pleurae corresponding to (though posteriorly not strictly confluent with) axial rings, sharply downflexed distally; deep, sharp interpleural furrows, arcuate, widening slightly, be- coming progressively more posteriorly directed backwards, and shorter, with last pair shortest and orientated directly backwards alongside terminal piece, coming into contact just behind terminal piece on posterior border (PI. 114, fig. 9). First pleural ridge exhibits rather abrupt flexure between inner and outer parts, coinciding with fulcrum of preceding thoracic pleura; in larger specimens (PI. 114, fig. 10) outer part of first pleura has facet, appearing to show some wear from movement of preceding thoracic pleura over it. Ends of pleurae bluntly rounded, fused and only projecting slightly outward; extending into pygidial doublure of ventral surface (PI. 114, figs. 14 and 26) to fulcrum at anterior end, and to one-third length of terminal piece posteriorly. No spines. Surface of exoskeleton mainly smooth, though scattered granules of variable size may be seen on frontal and lateral glabellar lobes, anterior and lateral borders and fixed cheek (PI. 115, fig. 12), on median body of hypostome, on dorsal, anterior, and posterior surfaces of thoracic segments, and on dorsal surface and postero-lateral margin of pygidium of well-preserved specimens. Largest glabella (SUP 23939) has maximum width of 9 mm. It is estimated that this particular form would have had a maximum width across cephalon of about 18 mm, and an overall length (sag.) of about 36 mm. Other large specimens (SUP 23901, 23927) have glabella width of 7-4 mm and length (sag.) of cranidia of 8-5 mm. Holotype has cephalon width of 7-0 mm, glabella width of 3-6 mm, and length (sag.) of cranidium of 4T mm. Smallest cranidia in collection, presumably representing small holaspides, have length (sag.) of just over 2-3 mm and glabellar width of 2 mm (PI. 115, fig. 10). B. D. WEBBY: PLIOMERINA FROM NEW SOUTH WALES 619 Remarks. The type species, Pliomerina martellii (Reed) from the Middle Ordovician of Yunnan, China, differs from P. austrina sp. nov. in having shorter (tr.) lateral glabellar lobes and furrows, the lateral furrow 3p intersecting preglabellar furrow almost at junction with axial furrow and in line with maximum width of glabella, a pair of short, curved grooves on surface of frontal lobe towards mid-line, and a slightly more prominent mesial tubercle on occipital ring (Reed 1917). Of the species from Kazakhstan, P. sulcifrons (Weber) from the Lower Caradocian has the lateral furrow 3p also inter- secting preglabellar furrow near junction with axial furrow, shorter and straighter lateral furrows, and seemingly a shorter (exsag.) lateral lobe Ip. P. unda Chugaeva from the Middle Caradocian has a relatively narrower glabella, the lateral furrow 3p inter- sects at or very near junction of preglabellar and axial furrows, and the lateral lobe Ip is short (exsag.), especially at inner end. The uppermost Caradocian to lowest Ashgil- lian species, P. dulanensis Chugaeva, differs in having a relatively wider and blunter frontal lobe, a very short (exsag.) lateral lobe Ip and an almost rectangular terminal axial piece on the pygidium (Chugaeva 1958). Pliomerina prima sp. nov. Plate 115, figs. 13-29 Material. Holotype (SUP 23908) and five paratypes (SUP 23909-13) from the ‘brachiopod unit’ in the lower part of the Bowan Park Limestone 1^ miles north-east of Quondong. Also paratypes from the lower part of the Cliefden Caves Limestone at Fossil Hill (SUP 23914-20, 23944-49, 24900-03). Numerous other fragmentary specimens have been collected from the lower part of the limestone at Fossil Hill. Diagnosis. Species of Pliomerina with lateral glabellar furrow 3p intersecting preglabellar furrow a short distance in front of antero-lateral corner of glabella; lateral furrow Ip sharply concave forward at inner end; length (exsag.) across club-shaped lateral lobe Ip at inner ends of furrows much less than across lobes 2p or 3p\ development of moderately prominent mesial tubercle of occipital ring, even on larger specimens ; bluntly pointed free ends of pleurae on pygidium ; more conspicuous and even fine granulation, especially on free cheeks, hypostome and pygidium. Description. Cephalon sub-semicircular; gently convex in both directions, except for moderately convex frontal part of glabella sagittally. Glabella subpentagonal, relatively broad, with widest part towards front end of lateral glabellar lobes 3p, narrows gradually back to occipital ring; bounded by deep, broad axial furrows. Occipital furrow deep, broad, bowed forwards. Lateral furrows become more deeply impressed at outer ends, extending to slightly more than one-third of glabella width; furrow Ip slightly backwardly directed with strong forward bend at inner end; lobe Ip with markedly club-shaped form and very narrow (exsag.) inner end; furrow 2p transverse to slightly backwardly directed, subparallel to furrow Ip, bounding almost straight-sided lobe 2p-, furrow 3p inclined more strongly backwards at angle of about 50° to axis, and commences at preglabellar furrow just in front of antero-lateral corner of glabella; maximum length (exsag.) of lobe 3p (measured across outer expanded part) slightly greater than lobe Ip and much greater than lobe 2p. Preglabellar furrow shallow except at distal ends, curved downward across sagittal region and up-domed laterally. Frontal lobe relatively large, subequal to (projected) length (sag.) of lateral lobes lp-3p, and wider than long (sag.). Anterior border slightly wider than glabella, short (slightly shorter medially than laterally); only outer parts seen in dorsal view, inner, sagittal part reflected downwards and backwards (PI. 115, figs. 17 and 19). Undersurface of axial and occipital furrows form strong ridges; prominent anterior boss beneath anterior end of axial furrow, midway between front and back lateral extremity of anterior 620 PALAEONTOLOGY, VOLUME 14 border; lateral glabellar furrows only faintly impressed on undersurface. Occipital ring gently convex with centrally placed mesial tubercle (most prominent in small specimens) and straight posterior margin. Doublure of occipital ring reaches forward to occipital furrow. Cheek subtriangular, sloping gently outward and forward, with conspicuous eye lobe and posterior and lateral furrows. Eye lobe relatively large, raised, situated just out from axial furrow, and having mid-point in similar transverse line to lateral furrow 2p\ length (exsag.) just more than one-third length (sag.) of glabella; steep, outwardly curved eye surface not preserved; palpebral lobe moderately broad, tapering anteriorly and posteriorly, convex in exsagittal line; deep palpebral furrow runs longitudinally along inner margin of palpebral furrow, curving at anterior and posterior ends to be confluent with furrow bounding eye socle of free cheek. Anterior branch of facial suture runs for- ward from eye lobe, intersects anterior end of axial furrow and curves sharply inward and forward to pass between anterior border and lateral border ; posterior branch swings abruptly outward on posterior side of eye lobe, and runs outward and backward across lateral border, intersecting margin just in front of genal angle. In large specimens genal angle is rounded (SUP 23920), but in smaller specimens posterior border is elongated into short genal spine (PI. 115, figs. 16-19). Deep, posterior furrow not strictly confluent with occipital furrow, dying out laterally. Posterior border expanding (exsag.) EXPLANATION OF PLATE 115 Figs. 1-12. Pliomerina aiistn'na sp. nov. 1-8, 10-12, Ordovician limestone at Billabong Creek. 9, middle part of Bowan Park Limestone at Quondong. 1-2, dorsal and anterior views of large in- complete cephalon of paratype SUP 23901, x4. Note prominent nicks in preglabellar furrow to either side of mid-line and fine granular ornament. 3, dorsal view of 5 thoracic segments of paratype SUP 22949, X 4. 4—6, oblique lateral, ventral and oblique anterior views of incomplete cephalon showing prominent eye and smalt fragment of hypostome attached by hypostomal suture to rostral plate and anterior part of lateral border, and downward extension from anterior boss; paratype SUP 22948, x4. 7, oblique anterior view of incomplete cephalon of paratype SUP 22947, x4. 8, ventral view of paratype SUP 23932 showing anterior boss, X 4. 9, dorsal view of rather smooth, incomplete pygidium; paratype SUP 23907, x4. 10, dorsal view of small incomplete cephalon with tiny mesial tubercle on occipital ring; paratype SUP 23905, X 5. 11, dorsal view of small pygidium of paratype SUP 23900, X 5. 12, detail of antero-lateral part of cephalon to show ornamentation; paratype SUP 23932, x 10. Figs. 13-29. Pliomerina prima sp. nov. 13-23, lower part of Bowan Park Limestone ^ miles north- east of Quondong; 24-29, lower part of Cliefden Caves Limestone, Fossil Hill. 13, dorsal view of incomplete cephalon of holotype SUP 23908, X 5. Note short (exsag.) inner ends of lobe Ip and prominent mesial tubercle on occipital ring. 14, detail of antero-lateral part of cephalon to show ornamentation; holotype SUP 23908, X 10. 15, dorsal view of small pygidium of paratype SUP 23913, X5. Note fine granulation on surface. 16, dorsal view of small, incomplete cephalon of paratype SUP 23912, showing short genal spine, x 5. 17-19, dorsal, ventral, and oblique antero- lateral views of small, incomplete cephalon of paratype SUP 23909, X 5. Note large eye, short genal spine, and fine granulation on lateral border; and on the undersurface antennular notch and anterior boss. 20-22, oblique ventral, oblique dorsal, and dorsal views of incomplete hypostome showing anterior wing and wing process, posterior wing, doublure and outer surface granulation; paratype SUP 23910; 20, X 5; 21-22, X 8. 23, dorsal view of small, incomplete cephalon of paratype SUP 23911, x5. 24, dorsal view of large incomplete cephalon of paratype SUP 23914, x4. Note tiny mesial tubercle on occipital ring and short (exsag.) inner end of lobe Ip. 25, dorsal view of incomplete pygidium (possibly slightly distorted); paratype SUP 23916, x5. 26, dorsal view of incomplete cephalon of paratype SUP 23919, x4. Note mesial occipital tubercle and short (exsag.) inner end of lobe Ip. 27, dorsal view of slightly distorted pygidia; paratype SUP 23917, x4. Note more fragmentary pygidium exhibits pleurae with bluntly pointed ends. 28, ventral view of incom- plete hypostome showing lateral borders and granulated median body; paratype SUP 23918, X 5. 29, dorsal view of incomplete cephalon of paratype SUP 23915, X4. Note tiny mesial tubercle on occipital ring and short (exsag.) inner end of lobe Ip. Palaeontology, Vol. 14 PLATE 115 WEBBY, Pliomerina B. D. WEBBY; PLIOMERINA FROM NEW SOUTH WALES 621 distally. Deep lateral furrow extends forwards from junction with posterior branch of suture, along inner edge of lateral border to intersection with anterior branch of suture; then confluent with suture between anterior border and lateral border (PI. 115, fig. 19). Lateral border broad, gently curved; antennular notch on doublure of lateral border with only short curved flange of lateral border in front of notch. On undersurface antennular notch narrows backwards and slightly outwards, flanking lateral portion of hypostomal suture (PI. 115, fig. 18). Doublure of lateral border narrower to rear; continues on to posterior border and inwards to fulcrum. Fine, moderately uniform granulation on dorsal surface of cephalon, especially well developed on anterior and lateral borders (PI. 115, figs. 14 and 19). Hypostome with gently convex median body, widest in line with anterior wings; lateral furrows sharp, shallowing towards back; posterior margin with tongue-like outline; anterior margin bluntly V-shaped; moderately developed lateral bulges on median body just in front of anterior wings; no anterior border medially; lateral border commences opposite anterior wing, with weakly angulate shoulder in front of posterior wing, continuous with posterior border. Doublure above posterior region similar in length (sag. and exsag.) to posterior border, extending in to dorsally and inwardly projecting flange-like posterior wing, and between anterior and posterior wings as lateral notch. Anterior wing moderately large, sloping steeply dorso-posteriorly, carrying steeply dorso-anteriorly inclined wing process on inner, anterior surface (PI. 115, figs. 20-22). Attachment of hypostome at hypostomal suture is with rostral plate and anterior part of inner edge of lateral border (PI. 1 15, fig. 1 8). Presumably hypostome also supported by wing process resting close to or against anterior slope of anterior boss. Antenna may have passed outward and forward through lateral notch in hypostome and beneath anterior wing, and then forward along antennular notch to the exterior. External surface of hypostome granular, especially on median body and postero-lateral border. Thorax incomplete; maximum of seven articulated segments seen n one specimen (SUP 24902); appears to be closely similar to thorax of P. aiistrina. Pygidium sub-semicircular, gently convex, about equal to cranidium in length. Axis extending to posterior margin, tapering steadily, with five well-defined axial rings and large terminal piece having narrowly rounded apex (differences in shape of terminal piece as seen in PI. 1 15, figs. 25 and 27 seem to be due to differential compression or compaction). Axial rings occupy slightly more than half length of axis; anterior rings slightly curved forward. Axial furrow deep, dying out at posterior end of ter- minal piece. Ring furrows also deep. Five pairs of backwardly curved, slightly widening, unfurrowed pleurae, corresponding to, though not strictly confluent in posterior part with five axial rings; succes- sive pleurae shorter in length, and end in short bluntly pointed free ends. Interpleural furrows short and deep. Surface granular. Largest specimen (PI. 115, fig. 24) has maximum glabellar width of 4-6 mm and cranidial length (sag.) of 5-4 mm. Another moderately large, crushed specimen (SUP 23920) with rounded genal angle, has a cephalon 7-5 mm wide (before crushing probably nearer 8-3 mm wide), glabellar width of 3-4 mm (also minimum value owing to crushing), and cranidial length of 4-4 mm. The holotype has glabellar width of 3-4 mm and cranidial length of 4 0 mm, and the smallest specimen (SUP 24900) studied has a cephalon width of 2-8 mm and cranidial length of 1 -3 mm. Remarks. Of the described species of Pliomerina, P. prima bears closest similarities to P. aiistrina. However, it may be distinguished chiefly by having a much shorter (exsag.) inner end to the club-shaped lateral lobe Ip, more prominent mesial tubercle on occipital ring, more pointed outer ends of pygidial pleurae and, in well-preserved material, more uniform and conspicuous fine granulation on outer surfaces of the exoskeleton. Acknowledgements. The author thanks Professor H. B. Whittington (Cambridge) for critical reading of the manuscript and for identifying Amphilichas, Eobrontens, and Spbaerocoryphe in the faunas. Dr. G. H. Packham (Sydney) for discussion and co-operation in allowing me to include the silicified Billabong Creek material, and Mr. V. Semeniuk (Sydney) for providing silicified specimens from the lower part of the Bowan Park Limestone. The work has been supported by a grant from the Australian Research Grants Committee. 622 PALAEONTOLOGY, VOLUME 14 REFERENCES BANKS, M. R. 1962. Ordovician System. The geology of Tasmania. J. geol. Soc. Aiist. 9, 147-176. CAMPBELL, K. s. w. and DURHAM, G. J. 1970. A New Trinucleid from the Upper Ordovician of New South Wales. Palaeontology, 13, 573-580, pi. 111. CHUGAEVA, M. N. 1956. New trilobite genera from the Middle and Upper Ordovician of southern Kazakhstan. Dokl. Akad. Nauk SSSR, 111 (6), 1336-1339. [In Russian.] 1958. Trilobites of the Ordovician of Chu-Ili mountains. In The Ordovician of Kazakhstan, III. Trad. geol. Inst. Akad. Nauk SSSR, 9, 5-138. [In Russian.] COOPER, B. N. 1953. Trilobites from the Lower Champlainian Formations of the Appalachian Valley. Mem. geol. Soc. Amer. 55, 1-69. GiLBERT-TOMLiNSON, J. 1961. Preliminary report on lower Palaeozoic fossils of Samphire Marsh No. 1. Bur. Min. Res. Aust., Petroleum Search Sub. Act Publ. no. 5, 29-36. GUPPY, D. J. and OPiK, a. a. 1950. Discovery of Ordovician Rocks, Kimberley Division, W.A. Aust. J. Sci. 12, 205-206. HARRINGTON, H. J. 1957. Notes on New Genera of Pliomeridae (Trilobita). J. Paleont. 31, 811-812. KOBAYASHi, T. 1 934. The Cambro-Ordovician Formations and Faunas of South Chosen. Palaeontology. Pt. I. Middle Ordovician Faunas. J. Fac. Sci. Imp. Univ. Tokyo, Sect. II, 3 {8), 329-519. LINDNER, A. w. 1953. The Geology of the Coastline of Waratah Bay between Walkerville and Cape Liptrap. Proc. Roy. Soc. Viet. 64, 77-92. MOORS, H. T. 1970. Ordovician graptolites from the Cliefden Caves Area, Mandurama, N.S.W., with a re-appraisal of their stratigraphic significance. Proc. Roy. Soc. Viet. 83, 253-287. OPIK, A. A. 1957. Cambrian Geology of the Northern Territory. Bull. Bur. Min. Res. Aust. 49, 25-54. PACKHAM, G. H. 1967. The Occurrence of Shelly Ordovician Strata near Forbes, New South Wales. Aust. J. Sci. 30, 106-107. 1969. Ordovician System. The Geology of New South Wales. J. geol. Soc. Aust. 16, 76-103. REED, F. R. c. 1917. Ordovician and Silurian fossils from Yunnan. Palaeont. indica, N.S. 6, Mem. 3, 1-70. 1941. A new genus of trilobites and other fossils from Girvan. Geol. Mag. 78, 268-278. SHAW, F. c. 1968. Early Middle Ordovician Chazy Trilobites of New York. N.Y. State Mus. Mem. 17, 1-163. WEBBY, B. D. 1969. Ordovician stromatoporoids from New South Wales. Palaeontology, 12, 637-662. MOORS, H. T. and mclean, r. a. 1970. Malongullia and Encrinuraspis, New Ordovician Trilobites from New South Wales, Australia. J. Paleont. 44, 881-887. WHITTINGTON, H. B. 1961. Middle Ordovician Pliomeridae (Trilobita) from Nevada, New York, Quebec, Newfoundland. J. Paleont. 35, 911-922. 1966. Phylogeny and distribution of Ordovician trilobites. J. Paleont. 40, 696-737. B. D. WEBBY Department of Geology University of Sydney Sydney, N.S.W. 2006 Typescript received 29 December 1970 Australia TWO NEW PLIOCENE SPECIES OF NEOMERIS (CALCAREOUS ALGAE) FROM THE BOWDEN BEDS, JAMAICA by LASZLO RACZ Abstract. Neomeris bowdenensis sp. nov. and Neomeris sp. nov. are described from the Lower Pliocene. The vegetative secondary rays and the reproductive cysts of these species are arranged according to a geometrical pattern on both the Inside and outside of the calcareous wall. It is supposed that the Neomeris specimens, found together with a rich bryozoan, molluscan, foraminiferal, and coral assemblage, are derived from an extremely shallow, sheltered environment. The fragile nature of the algal structure suggests that they have not been transported very far. Neomeris dasycladacean algae are common in the Cretaceous and Tertiary in southern Europe, North Africa, the Middle East, India, and the Caribbean region. In modern tropical seas they have been collected at several places within a geographical zone delimited by the marine surface isocrymes of 20 °C (Konishi and Epis 1962). Fragmen- tary fossil representatives of the genus are often found in carbonate rocks. Well- preserved calcareous tubes have occasionally been recognized, especially in some Tertiary deposits (Morellet and Morellet 1913; Weisbord 1966). As far as I know no algal discoveries have as yet been reported from the Bowden Beds. The Neomeris fragments found in the Bowden Shell Bed cannot, to my knowledge, be considered as similar to any of the Neomeris species so far described. Acknowledgements. The algal material was donated by Dr. R. Lagaaij, who I would like to thank sincerely for his help and encouragement. Thanks are also due to Mr. W. Geluk by whom the photo- graphs were taken and processed. I am indebted to the Bataafsche Internationale Petroleum Maats- chappij N.V. (Royal Dutch/Shell Group) for permission to publish this article. The material described was collected from the Bowden Beds, Jamaica, and is deposited in the British Museum (Natural History) under the numbers V.53924-V.53929. SYSTEMATIC DESCRIPTIONS Phylum CHLOROPHYCOPHYTA Family dasycladaceae Kiitzing 1843, orth. mut. Stizenberger 1860. Tribe neomereae Pia 1920. Genus neomeris Lamouroux 1816. Neomeris bowdenensis sp. nov. Plates 116 and 117, figs. 1, 2 Diagnosis. Elongate tubular Neomeris species, showing a relatively thick but delicate wall, perforated by numerous fine pores or canals ; differs from the known Recent and Neogene species by having both a greater diameter and a thieker calcareous wall. [Palaeontology, Vol. 14, Part 4, 1971, pp. 623-628, pis. 116-117.] 624 PALAEONTOLOGY, VOLUME 14 Holotype. The specimen figured in Plate 116, figs. 1,2; Plate 117, figs. 1, 2 from the Bowden Shell Bed, Jamaica. B.M. (N.H.) V.53924. Pamtypes. Three specimens, B.M. (N.H.) V.53925-V.53927, from the same locality as the holotype. Other material. One thin section, B.M. (N.H.) V.53928, of a specimen from the same locality as the holotype. Age. Lower Pliocene. Although the Bowden Beds were previously considered to be Upper Miocene in age (Lagaaij 1959), Dr. Lagaaij has, in a written communication, drawn the writer’s attention to an article of Banner and Blow (1965, pp. 1 164-1166) in which they state that the planktonic foraminiferal content of the beds is, in fact, of Lower Pliocene age. Description. The thallus is tubular in form, straight or gently curved, consisting of a hollow central part and a relatively thick, delicate calcareous wall. The incomplete thallus varies between 3-5 mm long, and is a few mm wide. Regularly spaced nodular swellings, arranged in annular concentric rows on the exterior of the thallus, are the surface expressions of the gametangia (sporangial cases). Where the abrasion of the surface has reached an advanced state, the swellings have been removed and the game- tangia are visible as deep cylindroid casts (PI. 117, fig. 2). Fine pores, surrounding the gametangia according to a geometrical pattern, represent the distal ends of the radially arranged secondary rays. Numerous, very fine, irregularly spaced openings penetrate into the calcareous body. These usually show an inclination to the central longitudinal axis. The interior surface of the thallus is annulated longitudinally by equally spaced, integrally calcified, sporangial rings. Each sporangial ring is itself divided by diagonally running rows of protuberances and furrows. Due to stronger calcification the protuber- ances may sometimes form a continuous ‘ridge’. Abraded specimens and a transverse section have shown that the protuberances are bored by minute openings and that every opening leads into a gametangium (PI. 116, figs. 3, 4). Slight depressions, which are to be considered as the inner terminations of the secondary rays, are arranged in an offset pattern in the furrows. The biological and structural composition of the central stem is uncertain. The original non-calcified stem cell may have been composed of amorphous gelatinous material, which has disappeared and been replaced by poorly cemented post-mortem sediments. The calcareous wall is perforated by minute pores or canals (secondary rays or branches) and oval-shaped gametangia. These are arranged radially in transverse section and perpendicular to the central axis. The distribution of the gametangia and the secondary rays seems to be regular in both transverse and longitudinal sections. The interpores are filled with calcite or aragonite cement produced by the reproductive organs and the vegetative secondary rays. EXPLANATION OF PLATE 116 Figs. 1, 2, 3, 4. Neomeris bowdenensis sp. nov.. Lower Pliocene, Jamaica. 1, 2 Holotype. 1, Exterior view of tubular fragment, x20. 2, Transverse section of the same specimen, x20; B.M. (N.H.) V.53924. 3, Longitudinal section, showing the wall and interior surface, X 20. 4, Thin section of another specimen showing the shape of the gametangia and their proximal and distal terminations, x22; B.M. (N.H.) V.53928. Palaeontology, Vol. 14 PLATE 116 RACZ, Pliocene alga l1 'v' ■J a ,r S-. i f 'V J r. ; ' V ' J'V I:'/ . ;.;,i 99 L. RACZ; PLIOCENE NEOMERIS 625 TEXT-FIG. L Diagrammatic representation of the external surface of a Neomeris thallus, showing the relative positions of the main features, o distal end of gametangium, • distal end of secondary ray. Since the primary rays are not preserved it is difficult to reconstruct their exact position and to trace their biological function. The secondary vegetative rays surrounding the sporangia are regularly arranged radially as well as longitudinally. They seem to be straight, cylindroid portions of the thallus, although a slight tapering towards the distal end has been observed. The gametangia are regularly arranged ovoid, shiny walled cases with a micro-aperture at both the proximal and distal ends (PI. 116, fig. 4). Geometry. On the exterior and interior surfaces the openings of the gametangia and the secondary rays are arranged according to a geometrical system (text-fig. 1). As is illus- trated on the figure, six secondary rays surround one sporangial case, and three sporan- gial cases surround one secondary ray. In other words a pseudo-hexagonal honeycomb structure has been built up around a sporangial case and a pseudo-triangular structure has been built up around a secondary ray. 626 PALAEONTOLOGY, VOLUME 14 Because of the offset patterns present on both sides of the calcareous wall, an imaginary line con- necting the secondary rays in two adjacent rows would bisect a similar line joining the gametangia of adjacent rows. This offset pattern has caused the oblique parallel columns of pores along the length of the thallus. This phenomenon can be clearly shown with the distal ends of the gametangia on an abraded surface where the columns form an angle of about 30° to the longitudinal axis of the thallus. Relative positions of the main features (text-fig. 1) (o) Angle from a gametangium to two successive secondary rays 60° (b) „ of the connecting-line between secondary rays in two successive rows 120° (c) „ from a secondary ray to two gametangia of the same row 120° (d) „ of oblique columns of gametangia with longitudinal axis 30° Measurements (in ;um): T1 TD Td Wt Psd R1 Rd Rad Rn Pn Up to 5000 2000-2700 110-1600 350-450 40-60 220-250 110-170 20-30 30-38 30-38 Explanation of symbols; T ^ thallus (500) D = outer diameter W wall d = inner diameter P = pores, branches, rays t = thickness R - reproductive organs 1 = length a = aperture s = secondary n = number Rd: inner diameter of the ovoidal gametangia (sporangial case) at thickest part. In comparing N. bowdenensis sp. nov. with other similar species from the Neogene of the Caribbean area it beeomes apparent that the main differences are in the proportions of the thallus. The only two Neomeris species known to the author from tropical America are N. venezuelensis Weisbord, and N. (Vaginopora) sp. Morellet and Morellet (1939, p. 25, pi. 1, fig. 1). However, the measurements of both the diameter of the thallus and the thickness of the calcareous wall of N. bowdenensis are approximately double those of the species mentioned above. Neomeris sp. nov. Plate 1 17, fig. 3 One nearly complete, fairly well-preserved segment has been found. Although the calcification was very strong, the basic structural features, described in N. bowdenensis sp. nov., are also recognizable in this alga. However, the distinct differences in the size of the thallus, the calcareous wall, the secondary rays, and the reproduetive organs, suggest that this is another species. EXPLANATION OF PLATE 117 Figs. 1 , 2. Neomeris bowdenensis sp. nov.. Lower Pliocene, Jamaica. Holotype. 1 , Details of wall struc- ture showing the position of the reproductive organs and secondary rays, X 60. 2, A detailed view of the exterior surface, showing the distal ends of the reproduetive organs and secondary rays, x 40; B.M. (N.H.) V.53924. Fig. 3. Neomeris sp. nov.. Lower Pliocene, Jamaica. General view of a slightly damaged specimen, X 20; B.M. (N.H.) V.53929. Palaeontology, Vol. 14 PLATE 117 RACZ, Pliocene alga L. RACZ: PLIOCENE NEOMERIS 627 Material. The specimen figured in Plate 117, fig. 3. B.M. (N.H.) V.53929 is from the Bowden Shell Bed, Jamaica. Age. Lower Pliocene. Measurements (in /xm) : TI TD Td Wt Psd R1 Rd Rad Rn Pn 3500 1500 600-700 400-440 20 140 75 _ _ _ Depositional environment of the Bowden Beds The algal species described form a minor part of an assemblage consisting of foram- inifers, molluscs, bryozoans, and corals. Because of the variety of these fossil remains, the environment of deposition, especially the bathymetrical conditions, has attracted the attention of several workers. Vaughan (1919) identified 17 coral species and on their evidence concluded ‘that the depth probably was not so much as 20 fathoms’. Woodring (1928) stated that the molluscan fauna present is characteristic for neritic conditions (up to about 100 fathoms). Palmer (1945) studied the foraminiferal assemblage and reached the conclusion that ‘the probable depth habitat ... of 60 fathoms or slightly more’ was very acceptable. Recent work by Brouwer (oral information given to the author) in the form of a com- puterized quantitative statistical study has produced rather different results. According to him the probable depth of water during deposition was confined to between 1 1 and 42 fathoms. The discrepancies in the latter two results may be due to the fact that Palmer attached too much significance to fossils which formed only a minor part of the assem- blage. Lagaaij (1959) has described several bryozoan species which were new to Bowden. Recent representatives of some of these species are found at depths of between 0 and 71 fathoms. In order to determine the original bathymetrical conditions an extensive study of the whole bryozoan fauna was carried out by the same worker, who concluded that a water depth of between 24 and 35 fathoms during deposition was the most likely (Lagaaij, oral information). Data concerning the bathymetrical position of recent representatives of Neomen's have shown that they appear to be present in very shallow, warm seas or lagoons. Their most frequent occurrence is in sheltered areas, but provided that the water energy is low, they may also occur in the littoral zone or very near to the low-tide level in the inner neritic (infra-littoral) zone, within a 0-10 m depth-range (Konishi and Epis 1962; Klement 1966). The few species of recent Neomen's which have been found in deeper water were probably transported from their usual habitat by current action (Klement 1966). There are convincing arguments that suggest a shallow water environment for the Cretaceous N. pfenderae Konishi and Epis from the Mural Limestone in Arizona (Konishi and Epis 1962). The co-oceurrence of N. venezuelensis Weisbord with other shallow water faunal and floral elements in a bioherm of the Pliocene Playe Grande Formation in N. Venezuela suggests the same ecological conditions (Weisbord 1966). The presence of Neomeris fragments within the assemblage therefore suggests that the algal particles have been transported by current action from shallower areas of the sea into the deeper environment, prior to deposition. Because the relatively long, delicate 628 PALAEONTOLOGY, VOLUME 14 algal fragments could not offer much resistance to current action, it is believed that they can have been transported only a short distance. This theory of transportation and mixing of the fossil organisms has also been put forward, rather picturesquely, by Woodring (1965), who states that The Bowden fauna represents not a particular faunal facies, but a whole series of faunal facies, ranging from leaf litter on the forest floor (six species of land snails), through brackish-water courses of streams and mangrove swamps (Neritina, Mytilopsis, mangrove oysters), beach vegetation {Tralia, Planaxis), sand flats (Olivella, approximately 1,000 specimens of Oliva), inner and outer shelf (the bulk of the fauna), to a depth greater than 200 m, possibly as great as 500 m. It is as though these land, brackish-water (beach, shallow-water), and moderate-depth shells were swept together by a giant broom and dumped down a steep slope to be successfully buried and preserved in granule-gravel along with autochthonous deep-water and planktonic species. REFERENCES BANNER, F. T. and BLOW, w. H. 1965. Progress in the planktonic foraminiferal biostratigraphy of Neogene. Nature, 208, 1164-1166. KLEMENT, K. w. 1966. Studies on the ecological distribution of lime-secreting and sediment trapping algae in reefs and associated environments. Neiies Jb. Geol. Palaont. Abh. 125, 363-381. KONisHi, K. and epis, r. c. 1962. Some early Cretaceous calcareous algae from Cochise County, Arizona. Micropalaeontology, 8, 67-76, pi. 1. LAGAAU, R. 1959. Some species of Bryozoa new to the Bowden Beds, Jamaica, B.W.I. Micropaiaeon- toiogy, 5, 482^86. MORELLET, L. and MORELLET, J. 1913. Lcs Dasycladacees du Tertiaire parisien. Mem. Soc. geol. Fr. 21, 1-43, 3 pis. 1939. Tertiary siphoneous Algae in the W. J. Parker collection with descriptions of some Eocene Siphonae from England. London. Brit. Mus. (Nat. Hist.), 55 pp., 6 pis. PALMER, D. K. 1945. Notes on the Foraminifera from Bowden, Jamaica. Bull. Am. Paleont. 29, No. 115, 1-75. VAUGHAN, T. N. 1919. Fossil corals from Central America, Cuba, and Porto Rico, with an account of the American Tertiary, Pleistocene and Recent Coral Reefs. Bull. U.S. Nat. Mus. 103, 189-524, 85 pis. WEiSBORD, N. E. 1966. A new species of dasycladacean alga from the Playa Grande Formation (Pliocene) of northern Venezuela. Tulane Stud. Geol. 5, 49-52, pi. 1. WOODRING, w. p. 1928. Miocene molluscs from Bowden, Jamaica. II. Gastropods and discussion of results. Publ. Carnegie Inst. 385, 564 pp., 40 pis. 1965. Endemism in Middle Miocene Molluscan Faunas. Science, 148, 3672, 961-963. L. RACZ Shell U.K. Exploration and Production Ltd. Shell Centre London, S.E. 1 Typescript received 8 September 1970 THE NATURE OF ACICULELLA PIA (CALCAREOUS ALGAE) by GRAHAM F. ELLIOTT Abstract. The dasycladacean alga Aciciilella ogilvie-gordonae sp. nov. is described from the Upper Permian of northern Italy: the first Permian record of a hitherto Triassic genus. Julius Pia’s type-description of Aciciilella as an originally calcified endospore stem-cell filling, and not an acicularian spicule, is examined in the light of subsequent research and of fresh evidence from the Permian fossil, and is confirmed. Records of Aciciilella, including some possible new species, are summarized. It is concluded that the endospore stem-cell fillings referable to Aciciilella are possibly of diverse generic origins. Aciculella is a somewhat problematic microfossil, recognized by Julius Pia as dasycla- dacean in origin and named by him probably on account of resemblance to Acicularia. The morphology of Aciculella is simple and not ambiguous; the interpretation to be accorded to the structures seen, as remains of a former living alga, is difficult. Pia ( 1927) gave an excellent diagnosis of the genus, and somewhat characteristically named but did not describe or figure the type-species A. bacillwn, which came from the Middle Triassic (Ladinian Stage) of the Central European Carpathians. This deficiency he remedied later (Pia 1930) when describing other problematic dasycladacean remains, in part comparable. During a recent examination of algal limestones in the Ogilvie-Gordon collection in the British Museum (Natural History), I was surprised to find a Permian species of Aciculella in the Upper Permian Bellerophon Limestone from northern Italy. The preservation of this little fossil, and its association in the same rock with another problematic endospore alga, permits a re-examination and clarification of the problem. To facilitate subsequent discussion, the new species is described first. SYSTEMATIC PALAEONTOLOGY Order dasycladales (Pascher) Feldmann 1938 Family dasycladaceae Kiitzing orth. mut. Stizenberger 1860 Genus aciculella Pia 1927 Diagnosis (after Pia). Cylindrical solid calcareous rod or baton, without central tubular perforation, but with numerous subperipheral globular cavities, each opening externally by a pore. Considered to be the stem-cell filling of an otherwise uncalcified endospore dasycladacean. Type species. A. baciHum Pia; Middle Triassic (Ladinian) of Central Europe. Aciculella ogilvie-gordonae sp. nov. Plate 1 18, fig. 1 ; Plate 119, figs. 1, 2, 5 Diagnosis. Aciculella differing from the type-species in that the peripheral cavities are [Palaeontology, Vol. 14, Part 14, 1971, pp. 629-636, pis. 118-119.] C 8385 T t 630 PALAEONTOLOGY, VOLUME 14 smaller, more numerous, crowded, and ovoid or subdeltoid, rarely spherical. These differences are summarized below. A. bacillum A. ogilvie-gordonae Length — 2 03 mm-f Diameter 0-4-0-7 mm 0-4-0-5 mm Diameter of cavity OT mm; spherical 0 07-0 08 mm; ovoid or subdeltoid Number of cavities on cross-section 15 24 approx. Description. Solid cylindrical rod or baton, circular in cross-section, with length of 2-03 mm (incomplete) and diameters of 0-4~0-5 mm seen. The material forming the rod is coarse yellow-tinted crystalline calcite; the matrix of the rock outside is finely crystalline greyish calcite. Immediately within the outer periphery, as seen in both vertical and transverse section, the rod is lined with numerous closely-set ovoid bodies, often widen- ing from within outwards (subdeltoid), and only rarely spherical. They show diameters of 0-07-0-08 mm, and are closely set, often with interstices of only 0-01 mm across. When irregular in shape they have a crowded appearance. These bodies are usually filled with very dark, very finely crystalline calcite, much finer in grain than that of the matrix. The dark colour may be due to original plant matter. Careful examination shows that the yellow calcite often extends between the peripheral bodies, but that sometimes the grey calcite of the matrix, whether due to the surface being worn originally or to later replacement, extends inwards for a short distance between them. Very rarely an occasional ovoid body is to be seen centrally in the main calcite filling; this phenomenon was also noted by Pia in A. bacillum. Associated fossils are very numerous small foraminifera, rolled pieces of the codiacid alga Tauridium, recognizable by its distinctive thread-pattern, pieces of the dasycladacean EXPLANATION OF PLATE 118 Fig. 1. AciciileUa ogilvie-gordonae sp. nov. Holotype: longitudinal thin-section, slightly oblique, x48. Marginal sporangial bodies embedded in a common mass of organic calcium carbonate. Upper Permian, Bellerophon Limestone; Plesch dTnez, Grddener Valley, South Tirol, N. Italy. Brit. Mus. (Nat. Hist.) Dept. Palaeont., Reg. no. V.53904. Fig. 2. Aciciilaria (Bn'ardiiia) sp. Thin-section of spicule, X 60. Palaeocene-Lower Eocene, Sahil Maleh, Batinah Coast, Oman, Arabia. Reg. no. V.52033. Fig. 3. Atractyliopsis darariensis Elliott. Longitudinal thin-section of fragment, x40. In this genus the marginal sporangial bodies seen in Aciculella have their own individual coatings of organic calcium carbonate. Upper Permian, Darari Eormation; Ora, Mosul Liwa, N. Iraq. Reg. no. V.52037. Pig. 4. Acetabidaria sp. ‘The Mermaid’s Wine-glass’, x 4-5. The disc or cup is built of fused radial calcified elements, each of which is somewhat similar to the single spicule seen in fig. 2 above. Recent. Tunis, N. Africa. Fig. 5. Diplopora phanerospongia Pia. Oblique-transverse thin-section, X 24. A species in which cal- cified sporangial bodies, marginal in the stem-cell as in Aciculella and Atractyliopsis, are encased in the calcified detail of the verticillate branches of the plant, which were not calcified in the other two genera. Upper Triassic, derived in Upper Cretaceous Hawasina Formation, Oman, Arabia. Palaeontology, Vol. 14 PLATE 118 ELLIOTT, Aciculella Pia G. F. ELLIOTT: ACICULELLA PIA 631 Atractyliopsis lastensis Accordi, small rolled radiate growths of a myxophyte thread- alga, and occasional examples of the microproblematicum AeoJisacciis dimningtoni Elliott. This is for the most part a size- and density-sorted assemblage of small organic objects gently rolled and accumulated in shallow water not far from the place of growth of the algae. One small example of A. dimningtoni (Elliott 1958), now known to range from Permian to Lower Cretaceous, appears to show a tiny initial spherical structure, not seen in larger examples (PI. 119, fig. 3). This supports the view that these ubiquitous little microfossils may have been pelagic in origin. Holotype. The specimen figured in PI. 118, fig. 1, from the Bellerophon Limestone, Upper Permian; Plesch d’lnez, Grodener Valley, S. Tirol, N. Italy (Ogilvie-Gordon 1927, p. 5; 1927<7, p. 72 etc.). Brit. Mus. (Nat. Hist.), Dept. Palaeont., Reg. no. V. 53904. Paratypes. The specimens figured in PI. 1 19, figs. 1, 5, same locality and horizon as holotype, reg. nos. V.53903, V.53905. Other material. Several examples in thin-sections made from the same rock sample. This species is dedicated to the late Dame M. M. Ogilvie-Gordon (1864-1939), a pioneer Scottish woman geologist, distinguished for her work on the Dolomites, and a friend of Julius Pia. The material now described has been prepared from rock specimens collected by her. COMPARISON AND EVALUATION Acicu/ella is a somewhat problematic dasycladacean fossil: it can only be evaluated by comparison with other genera now to be considered. Aciciilaria is a common Tertiary micro-fossil. Species are represented by solid or occasionally hollow calcareous bodies or spicules, elongate-cuneiform longitudinally, showing various shapes in cross-section, and set with numerous small peripheral spherical cavities (PI. 118, fig. 2). At first regarded as animal in origin, their algal origin was indicated by Munier-Chalmas (1877), who recognized them as the dissociated radial segments of the calcified terminal discs of an alga similar to the living Acetabuloria or ‘Mermaid’s wine-glass’ (PI. 118, fig. 4). There is one surviving living species of Aciciilaria, and this is regarded taxonomically as a section of Acetabidaria by botanists (Egerod 1952), but palaeontologists, with the much greater importance of Aciciilaria over Acetabidaria in the fossil record in mind, maintain it as distinct. Shorter spicular bodies referable to Halicoryne have also been described from the Tertiary (Valet and Segonzac 1969). Associated with Aciciilaria in the Tertiary are the flattened or spherical solid calcareous sporangial bodies of Terqiieinella. These are known to have originated from dasycla- dacean genera of normal tubular form without reproductive discs (Morellet and Morellet 1913, 1922). In the Mesozoic, however, calcareous sporangial bodies of variably inter- mediate form between the two occur, and are usually referred to Aciciilaria s.l., their generie origins being quite uncertain (some may be referable to Halicoryne). This differ- ence in age and morphology between Aciciilaria s. str. and Aciciilaria s.l. has led to considerable confusion amongst those not closely familiar with this problem. In eonsidering the Triassic Aciciilella bacilliim Pia thought it highly improbable that it came from an alga with the very specialized organization of Acicularia s. str. (‘which seems to have developed from Terqueniella in late Cretaceous times only’; Pia 1930, p. 180). This was a strictly morphological recognition of the nature and geological age of 632 PALAEONTOLOGY, VOLUME 14 true Acicularia spp. However, I have since pointed out (Elliott 1968) that the reproductive disc, which in Acetabidaria and Acicularia s. str. is a choristospore development, is found also in Clypeina (Permian-Oligocene), where it is probably a cladospore develop- ment. In other words, it is evidence of the achievement of similar morphology by an earlier, more primitive alga. Moreover, in the living species Acetabularia clavata Yamada and A. tsengiana Egerod, the lightly calcified terminal reproductive rays are not united to form a disc, but are largely separate (Egerod 1952). If heavily calcified, these would give rise to structures like Aciculella. Such a development may well have existed in extinct cladospore genera, and would be represented by dissociated fossil elements. If, therefore, Pia’s rejection of Aciculella being similar in origin to Acicularia s. str. is to be supported, it must be on other grounds, and this is further dealt with below, as is his observation that Aciculella shows no pointed termination as in Acicularia s. str. Holosporella (Triassic) and Atractyliopsis (Permian, ? Carboniferous) were interpreted by Pia (1930, 1937) as the remains of endospore dasycladaceans whose sole calcification was around the reproductive elements or sporangia (cf text-fig. 1). These lay peri- pherally within the stem-cell, the exterior of the stem-cell and the outer branch-systems being uncalcified and hence quite unknown. The fossil thus consists of a hollow cylin- drical arrangement of hollow spheres with calcified walls, adjacent, touching, or fused to varying degree (and with a very variable amount of secondary calcification, depending on subsequent diagenesis) (PI. 118, fig. 3; PI. 119, fig. 4). This apparently improbable interpretation is supported by two pieces of evidence. One is the Triassic species Diplopora phanerospongia Pia (PI. 118, fig. 5), a diplopore showing normal heavy calification of both stem-cell and detailed trichophore branch-systems, but also showing calcified peripheral sporangia within the stem-cell cavity, as postulated for Atractyliopsis and other genera. Due to the waxing and waning stem-cell morphology of D. phanerospongia, accidental post-mortem insertion of originally separate and distinct straight calcareous algal structures one within the other (cf. Elliott 1968, pi. 24, fig. 2) is impossible. The endospore nature of this diplopore is thus authenticated. This vital species, known only to Pia from a Museum specimen, has now been rediscovered in the Upper Triassic, and other new similar species described (Ott 1967, Fenninger 1969). The second point is from the known ontogeny of living dasyclads. Here the large nucleus is housed within the holdfast or basal portion throughout most of the growth of EXPLANATION OF PLATE 119 Figs. 1,2. Aciculella ogilvie-gordonae sp. nov. 1, Oblique-transverse thin-section, paratype, x93. 2, Longitudinal section, x50. Upper Permian, Bellerophon Limestone; Plesch d’Inez, Grodener Valley, South Tirol, N. Italy. Reg. no. V.53903. Fig. 3. Aeolisaccus diinningtoni Elliott: longitudinal thin-section, X 330. Small example showing initial spherical chamber. Upper Permian, Bellerophon Limestone; Plesch dTnez, Grodener Valley, South Tirol, N. Italy. Reg. no. V.53903. Fig. 4. Atractyliopsis lastensis Accordi. Thin-section of fragment, X 1 10, for comparison with Aciculella ogilvie-gordonae in the same matrix and rock sample. Upper Permian, Bellerophon Limestone; Plesch dTnez, Grodener Valley, South Tirol, N. Italy. Reg. no. V.53906. Fig. 5. Aciculella ogilvie-gordonae sp. nov. Transverse thin-section to show central as well as marginal sporangial bodies, paratype, x 160. Upper Permian, Bellerophon Limestone; Plesch dTnez, Grodener Valley, South Tirol, N. Italy. Reg. no. V.53905. Palaeontology, Vol. 14 PLATE 119 ELLIOTT, Aciculella Pia a 't y li Ff' G. F. ELLIOTT: ACICULELLA PIA 633 TEXT-FIG. 1. Reproductive ontogeny of an extinct dasycladacean leading to production of a fossil Aciculella. a. Living plant growing on stone on sea-floor. b. Diagrammatic section of {a) showing tall central stem-cell with lateral branches, attached by holdfast to stone, large nucleus in holdfast, no calcareous structures in green plant. c. Break-up of nucleus and migration of small nuclei up the stem-cell. cl. Migration of nuclei in stem-cell to peripheral positions to form uncalcified ‘sporangia’. e. Rapid calcification of reproductive upper part of stem-cell, shedding of side-branches. /. Stump of plant with regenerative nucleus on stone; calcified body with encysted gametes lying in mud of sea-floor. the plant, but eventually breaks up into numerous smaller nuclei which migrate up the stem-cell to peripheral positions in the branches to initiate the reproductive bodies, around which calcification frequently originates (Hammerling 1931, 1944; Egerod 1952). Pia’s classification (1920) of dasycladacean reproductive structures as endospore, cladospore, and choristospore in that order approximately in geological time, suggests that the ontogeny is a recapitulation of the phylogeny, as Rezak (1959) has commented. 634 PALAEONTOLOGY, VOLUME 14 The existence of endospore genera, in which the sporangia are eventually distributed peripherally in the stem-cell but do not leave it for the branches (cladospore) or special reproductive outgrowths (choristospore), is thus indirectly substantiated. In abnormal living choristospore dasyclads ‘endospore’ sporangia are occasionally seen (cf Valet 1969, p. 569). In the thin-sections of Bellerophon Limestone now studied, Atractyliopsis lastensis and Aciculella ogilvie-gordome occur together. The calcareous wall-thickness of the sporangial globules of the former is often visible, plus a varying amount of secondary calcihcation which may partially or wholly fill the main stem-cell cavity, which is other- wise filled with normal matrix. In one case only was the filling of yellowish calcite as in Aciculella, and in this instance the whole fossil was embedded in a mass of such calcite extending external to the fossil. The usual preservation of Atractyliopsis is thus quite different to that of Aciculella, embedded in the same matrix. Pia’s explanation of his Triassic Aciculella may now be quoted in full. . . . we have to deal with the content of an ‘endospore’ axial cell. It must be supposed that at the time when calcification began, assimilation and circulation were stopped. The axial cell was gradually filled with lime from the growing point downward, mainly from the supply of material stored in the root-cell. When the alga died and broke to pieces, the gametes or the cysts in which they were contained, escaped from the cavities by the pores. (Pia 1930, p. 180.) This explanation has always seemed doubtful, since it gives no reason as to why he considered the calcification original-organic in origin, and not post-mortem diagenetic filling, so necessitating an explanation like that given. Study of the new Permian species supplies evidence in support, however. In A. ogilvie-gordomie the peripheral cavities show no trace at all of calcareous wall thickness. If they were originally uncalcified plant-matter at the death of the alga, as this suggests, their chanee of distinct preserva- tion from the plant-matter of the surrounding stem-cell was negligible. If preserved at all, such an Aciculella would be matrix-filled and the outer outline alone distinguishable under favourable conditions, and subsequent diagenesis could only affect what was originally preserved. It seems likely, therefore, that the stem-cell was in fact filled by organically formed aragonitie calcium carbonate spreading around and inwards from the sporangial bodies after these had formed and were mature. Calcification in living algae has not received the study it deserves, and it is still not known why some green algae form deposits of calcium carbonate on and within their tissues whilst growing and photosynthesizing side by side in the same environment with other green algae which never calcify. Between members of the Dasycladales this calcification is extraordinarily varied, though taxonomically eonstant for the ge.iera and species themselves. In the case of Aciculella the calcification must have been very rapid, once conditions were physiologically right, in order to preserve the occasional sporangia seen centrally in the stem-cell. The lime was not from the ‘supply of material in the root-cell’, as Pia supposed, though the biochemistry of its formation and timing was probably controlled from the residual nueleus in the basal holdfast (cf. Hammerling 1953 and Brachet 1965 on living Acetabularia). It is probable that this calcification extended only to the fertile middle and upper parts of the stem-cell; Pia noted the absence of a proximal ‘pointed end’ in Aciculella bacillum, and A. ogilvie-gordotiae ends irregularly proximally also. This may be due to breakage, but a ragged end to incomplete calcification is to be expected. After the calcified portion, probably containing encysted gametes in the cavities, broke G. F. ELLIOTT: ACICULELLA PIA 635 away, the nucleate holdfast and residual stump may have been capable of successive regenerations (cf. Valet (1969, p. 560) on living Bonwtella, HaJicoryne, Acetabiilaria, etc.). The calcareous rod, if not broken up with release of the sporangial contents, would be buried in calcareous mud. Subsequent diagenesis altered the aragonitic rod to coarse yellow calcite, the plant-filled cavities to dark finely-grained crystalline calcite, and the lime mud to lighter coloured, more coarsely-grained calcite, to give the appearances now seen in thin-section. Pia’s Holosporella, figured side by side with A. hacillum (Pia 1930, pi. 4), probably served him for discriminative comparison in the same way that Atractyliopsis has with Aciculella ogilvie-gordonae. This is not clear from his brief paper, though it probably accounts for his explanation, with which I agree. Other records. Aciculella ogilvae-gordouae is the first described Permian species, so far as I know: the Acicidaria sp. of Giiveng (1965, pi. A8), from the Turkish Permian, may be referable to Aciculella. Since Pia’s original paper (1930) the Triassic A. bacillum has been recorded from the Ladinian of Czechoslovakia by Andrusov (1939). Pia (1942) listed it again, together with undescribed Aciculella spp. from the Anisian and Carnic stages of central Europe. These records were repeated by Bystricky (1964), who also figured Acicidaria spp.’ from the Ladinian. Aciculella sp. nov. is recorded and figured from the Moroccan Infralias by Lemoine and Raoult (1962); the same paper quotes a verbal record of J. Sigal of A. cf. bacillum from the Upper Triassic of Spain. Some of the very miscellaneous bodies referred to Acicidaria s.l. from higher Mesozoic levels may prove to be referable to Aciculella. In conclusion, it must be noted that species of Aciculella may be the endospore stem- cell fillings from dasycladaceans of very different braneh-morphology, which would certainly be referred to difierent genera if they had calcified more completely. Aciculella is thus a form-genus for Dasycladaceae incerta sedis in much the same way as is Griplioporella. REFERENCES ANDRUSOV, D. 1939. Rolc des Thallophytes dans la constitution des roches sedimentaires des Carpathes tchecoslovaques. Mem. Soc. r. Sci. Boheme (1938), 11, 1-32. BRACHET, J. L. A. 1965. Acetcibularia. Endeavour, 24, 155-161. BYSTRICKY, J. 1964. Sloveusky kras: stratigrafia a Dasycladaceae mezozoika Sloveiiskeho krasu. Bratislava. 204 pp., 38 pis. EGEROD, L. E. 1952. An analysis of the siphonous Chlorophycophyta. With especial reference to the Siphonocladales, Siphonales and Dasycladales of Hawaii. Univ. Calif. Pubis Bot. 25, 325-454. ELLIOTT, G. F. 1958. Fossil Microproblematica from the Middle East. Micropaleoutology, 4, 419-428. 1968. Permian to Palaeocene calcareous algae (Dasycladaceae) of the Middle East. Bull. Br. Mus. nat. Hist. (Geol.), Suppl. 4. FENNiNGER, A. 1969. Ein Beitrag zur Flora und Fauna im Raume von Golpaygan (Iran). Verb. geol. Bundesanst., Wien, 1969 (1), 22-32. GUVENC, T. 1965. Etude stratigraphique ei micropaleontologique du Carbonifere et du Permien des Taurus occidentaux dans Tarriere-pays d'Alanya (Turquie). Acad. Thes., Univ. Paris. HAMMERLiNG, J. 1931. Entwicklung und Formbildungsvermogen von Acetabiilaria mediterranea. Biol. Zbl. 51, 633-647. 1944. Zur Lebensweise, Fortpflanzung und Entwicklung verschiederner Dasycladaceen. Arch. Protistenk. 97, 7-56. 1953. Nucleo-cytoplasmic relationships in the development of Acetabiilaria. hit. Rev. Cytol. 2, 475-498. LEMOINE, M. and RAOULT, J.-F. 1962. Algues calcaires dans I’infralias et le lias inferieur et moyen du chainon du Haouz (Rif septentrional, Maroc). C.r. somm. Seaiic. Soc. geol. Er. 1962 (7), 214-215. 636 PALAEONTOLOGY, VOLUME 14 MORELLET, L. and MORELLET, J. 1913. Lcs dasycladacecs du tertiaire parisien. Mem. Soc. geol. Fr. (paleont.), 47, 1-43. 1922. Nouvelle contribution a Fetude des dasycladacecs tertiaires. Mem. Soc. geol. Fr. {paleont.), 58, 1-35. MUNiER-CHALMAs, E. p. 1877. Observations sur les algues calcaires appartenant au groupe des siphonees verticillees (Dasycladees Harv.) et confondues avec les foraminiferes. C.r. hebd. Seanc. Acad. Sci. 85, 814-817. OGiLViE-GORDON, M. M. 1927. Das Grodcncr-, Fassa- und Enneberggebiet in den Sixdtiroler Dolomiten. I, II. Stratigraphie-Tektonik. Abh. geol. Bimdesanst., Wien 24 (1). 1927a. III. Palaontologie. Ibid. 24 (2). OTT, E. 1967. Dasycladaceen (Kalkalgen) aus der nordalpinen Obertrias. Mitt, bayer. Staatssamml. Palaont. 7, 205-226. PIA, J. 1927. Thallophyta: 1. Abteilung. In Hirmer M. Handbiich der Paldobotanik. Munich, Berlin. 1930. Upper Triassic fossils from the Burmo-Siamese frontier. A new Dasycladacea, Holosporella siamensis nov. gen., nov. sp., with a description of the allied genus Aciculella Pia. Rec. geol. Snrv. India, 62,, \n-m. 1937. Die wichtigsten Kalkalgen des Jungpalaozoikums und ihre geologische Bedeutung. C.r. 2me Congr. Avanc. Ft. strat. Carb. 2, 765-856. 1942. Uebersicht iiber die fossilen Kalkalgen und die geologischen Ergebnisse ihrer Untersuchung. Mitt, alpenl. geol. Ver. {Mitt. geol. Ges. Wien), 33, 11-34. REZAK, R. 1959. New Silurian Dasycladaceae from the southwestern United States. Colo. Sch. Min. Quart. 54, 115-129. VALET, G. 1969. Contribution a Fetude des Dasycladales. 2. Cytologic et reproduction. 3. Revision systematique. Nova Hedwigia, 17, 551-644. — ^ — and SEGONZAC, g. 1969. Les genres Chalmasia et Halicoryne (Algues Acetabulariacees). Bull. Soc. geol. Fr. {!), 11, 124-127. G. F. ELLIOTT Department of Palaeontology British Museum (N.IT.) London, S.W. 7 Typescript received 20 November 1970 A NEW FOSSIL ALGA FROM THE ENGLISH SILURIAN by GRAHAM F. ELLIOTT Abstract. A new fossil alga, Inopinatella lawsoni gen. et sp. nov., is described from the Upper Bringewood Beds (Aymestry Limestone), Upper Silurian. It is reminiscent of an early growth stage of the modern green alga Neomeris (Chlorophyta, order Dasycladales), and it may be a primitive adult non-calcified dasycladalean, with a simple structure now only seen in the early ontogeny of dasycladaceans. T HE alga which forms the subject of this paper was collected by Dr. J. D. Lawson during his investigation of the Aymestrey area and was entrusted to me for study. Although a new genus and represented by numerous examples, the material was at first sight unpromising, by reason of the simplicity of the structures preserved. However, considera- tion of the correct taxonomic reference of this alga has led to a surprising estimate of its possible relationships. SYSTEMATIC PALAEONTOLOGY Algae INCERTA SEDis (? Order dasycladales; chlorophyta) Genus inopinatella gen. nov. Diagnosis. Non-calcified alga with long thin main stem showing regularly spaced slightly thickened levels from each of which diverge much thinner primary branches, usually four at each level. Primary branches slightly swollen immediately beyond point of junction with main stem, then thinning and extending for some distance before dividing into two or more thinner secondary branches. Reproductive structures not seen. Type species. I. lawsoni sp. nov. Upper Silurian, Upper Bringewood Beds (Aymestry Limestone); England. Inopinatella lawsoni sp. nov. Plates 120, 121; text-fig. 1 Diagnosis. Inopinatella with primary and secondary branches about equal in length. Description. The remains of this alga lie flattened on planes of parting in the rock, much as an assemblage of Recent sea-weeds might be pressed for the herbarium, but without definite orientation. The original thallus is represented by a black carbonaceous residue, cracked and friable; where this has fallen away, the smooth-surfaced impressions of stems and branches are seen in the granular rock. In an account by Ruedemann (1909, p. 201) of a somewhat similar genus similarly preserved in the American Lower Palaeozoic, Callithamnopsis, the longitudinal cracking of the carbonaceous filling was considered to indicate flattening of an originally hollow stem. This has not been seen [Palaeontology, Vol. 14, Part 4, 1971, pp. 637-41, pis. 120-121.] 638 PALAEONTOLOGY, VOLUME 14 in JnopinateUa, but one example was noted of a thin black carbonaceous-walled circular cross-section at right angles (vertical) to the plane of parting, interpreted as a cross- section of an uncrushed hollow stem. For this reason the diameters given below are probably slightly greater than those of the living plant. TEXT-FIG. 1. A. Diagrammatic sketch of abnormal early growth-stage of living Neomeris sp., showing whorls with small number of unusually long side-branches, x 50 approx. Based on Valet (1968). b. Reconstruction of the Silurian hiopiuatella as in life, for comparison with (a). Upper portion of a plant, X 10 approx. EXPLANATION OF PLATE 120 Figs. 1-3. Inopinatella lawsoiii gen. et sp. nov., from the Silurian Upper Bringewood Beds (Aymestry Limestone), Aymestrey, Shropshire. 1, 2. Holotype, showing main stem and primary branches X 12-5, and the basal portion enlarged x 30. Brit. Mus. (Nat. Hist.), Dept. Palaeont., Reg. no. V.31268. 3. Paratype, Xl2-5. Reg. no. V.31263. Palaeontology , Vol. 14 PLATE 120 ELLIOTT, Silurian alga G. F. ELLIOTT: SILURIAN ALGA 639 The main central stem of the plant is remarkably uniform. Examples lie straight, curved, or flexuous, and were originally several cm long: up to 30 mm (incomplete at both ends) has been traced. Over this length the diameter diminishes only from 0-286 mm to 0-195 mm: in another incomplete specimen of 1 5 mm the diminution is from 0-260 mm to 0-234 mm. The stem shows a regular swelling or slight increase of diameter at the levels from which the branches spring; these are spaced apart very regularly along most of the length at 0-390 mm apart, though closer in the terminal (distal) part of the stem. In the 15 mm example quoted the diameter of 0-260 mm increases to 0-312 mm at branch level, and the diameter of 0-234 mm similarly to 0-260 mm. The black carbonaceous filling of the stem impressions is shrunken and cracked. Where these cracks are regularly and equidistantly transverse they give the appearance of an original plant structure, but this is an illusion as longitudinal, diagonal, and irregular cracking may also be found. In Callithamnopsis Ruedemann (1909) recorded transverse lines suggesting segmentation of the main stem at branch level: this has not been seen in Iiwpinatella. The point is further discussed below. The branches lie in some confusion on the planes of parting of the rock, particularly where several individual thalli are tangled together. When their junction with the main stem can be distinguished clearly at any separate branch level, they are seen to originate in fours. Damage or tangling obscures this with many junctions, and it may be that occasionally there were more than this, but four appears to be normal. Each typical primary branch swells quickly from a small insertion on the stem-cell to a diameter of 0-156 mm and thins to 0-104 mm: at a length of about 0-7 mm they divide into two secon- daries of about the same length and 0-104-0-078 or less diameter. There may sometimes have been more than two secondaries to a primary and also short tertiary branching, but I have not been able to distinguish this clearly on primaries visibly attached to a main stem. The main stems appear to end in a terminal bunch of shorter finer branches: this is probably due to the closer spacing of branch levels with their young growing branches at the distal, growing point of the plant. In Callithamnopsis Whitfield (1894) considered that one specimen probably showed a terminal growing point: I have not seen this in Inopinatella. No reproductive structures were found, and no recognizable holdfast was seen. I have pleasure in dedicating this species to its discoverer. Dr. J. D. Lawson. The generic name refers to the unexpected possible relationships of the plant : inopinatus, a, urn; that happens contrary to expectations, unexpected. Holotype. The specimen figured in PI. 120, figs. 1, 2, from the Upper Silurian, Upper Bringewood Beds (Aymestry Limestone), shale band several feet below Dnvw-beds. Small quarry above road, c. 594 m W. 40 N. of Aymestrey Church, Shropshire, England (Map. ref. 32/421655); J. D. Lawson Coll. Brit. Mus. (Nat. Hist.), Dept. Palaeontology, Reg. no. V.31268. Paratypes. The specimens figured in PI. 120, fig. 3 and PI. 121, figs. 1-3 same locality and horizon, Reg. nos. V.31256, 31262, 31263, 31278; also V.31253 not figured. Other material. Numerous examples on rock fragments from the same sample. COMPARISON AND AFFINITIES Much the closest fossil alga with which Inopinatella may be compared is Callithani- nopsis from the Ordovician (Trenton) of U.S.A. Callithamnopsis fruticosa (Hall) Whitfield was described as an alga by Whitfield (1894), and re-examined and a second 640 PALAEONTOLOGY, VOLUME 14 species (C. deJicatula) described by Ruedemann (1909), whilst Johnson (1961) has figured photographs of the type-specimens, the earlier accounts being illustrated by line drawings. Callithanmopsis is similarly preserved to Iiiopinatella, with the same consequent uncertainties of interpretation. Like the latter genus, it shows a long central main stem and numerous side-branches of lesser thickness. Branching of the central stem occurs in C. delicatula. The side-branches originate in pairs, one branch opposite the other, up the lower part of the central stem, which is said to be jointed at branch levels. In the upper part the side-branches are said to be verticillate (several originating at the same level). The ‘growing tip’, distally, is rounded and distinct. Whilst the side- branches agree with those of Inopinatella in proximal swelling, the long primaries give rise to four or more very short secondaries, and these to four or more minute tertiaries. The taxonomic position within the algae of such an extinct form is difficult to decide. The Chlorophyta, Phaeophyta, and Rhodophyta (green, brown, and red algae) are all possible. The living CaUiihanudou, which Whitfield thought his alga resembled, is a rhodophyte. Pia (1927, p. 67) followed by Johnson (1961), referred Callithanmopsis very doubtfully to the Dasycladales, a chlorophyte order. The verticillate branching of the upper stem, swollen-based primaries and the nature of the secondary and tertiary branching are dasycladacean characters. The pairs of opposite branches of the lower stem and the jointed stem (if correctly interpreted) are not dasycladacean characters. Some living brown and red algae show paired, alternate branching, or the appearance of branches in a close irregular-spiral order, but none to my knowledge are identical with Callithanmopsis. Inopinatella itself was a slender thin-stemmed alga with simple consecutive whorls of four thin simply-dividing branches. New whorls were formed at the growing tip, and enlarged to uniform spacing as the stem does not show jointing, and there is no trace of calcification, which would have been likely to have been preserved as there are calcareous fossils associated. Presumably each plant was attached by a holdfast, and although flexible each grew vertically, in spreads or thickets on the sea floor. Burial was by rapid muddy sedimentation which entombed them before decay, and prevented immediate subsequent decay so that, whatever the detailed diagenesis to which they have been subjected, they are today preserved as carbonaceous fillings in impressions. There is not, to my knowledge, any living species of alga like this. But there exists one startlingly similar developmental stage in a living dasycladacean. Neomeris is a living tropical dasycladacean alga, represented by some seven species. The adult algae are upright cylindrical, rather like little green corn-cobs, with the short branches and branchlets closely adpressed and moderately heavily calcified. But in an early uncalcified developmental stage the appearance is very different : the central stem- cell is thin, with widely-spaced thin branches. In the normal ontogeny of the plant this EXPLANATION OF PLATE 121 Figs. 1-3. 1, Paratype, portions of two individuals, x6-5. Reg. no. V.31278. 2, Paratype, area of terminal branches, X 14. Reg. no. V. 31262. 3, Paratype, general disposition of algae on slab, X 8. Reg. no. V.31256. Palaeontology, Vol. 14 PLATE 121 ELLIOTT, Silurian alga G. F. ELLIOTT: SILURIAN ALGA 641 thin stage is impersistent, with one or two such whorls near the apex and only scars of earlier deciduous branches below. However Valet (1968, pi. 18, fig. 3) figures an abnormal young Neomeris sp. at about the 3 mm growth-stage (adult Neomeris range from 15 to 30 mm high approximately). In this growth-stage a thin central stem-cell shows regularly-spaced whorls each with a small number of branches, unusually long in the primaries for such a developmental stage of Neomeris. Allowing for the difference in size, it is very much as hwpinateUa must have appeared in life. Is Inopiuatella, then, to be regarded as a primitive non-calcified ancestral dasycladacean alga ? The apparently flexuous character of the fossil plants could be explained by the conditions of burial. Its stem is not jointed, and this accords with the long central stem-cell of a modern dasycladacean. The co-existence in the older Palaeozoic of highly organized and calcified dasycladaceans, e.g. Mastopora, is not an argument against this interpretation, for similar phylogenetic development probably took place in different stocks at different times, as happened in other biological groups. Recapitulation is never an exact record of phylogeny, and it may well be that Valet’s exceptional developmental stage of Neomeris, abnormal by reference to the usual developmental pattern, mirrored a pattern of thallus long since abandoned in the adult. This interpretation of the affinities of Inopinatella must remain speculative; there seems, however, to be more evidence for it than for a similar correlation with a living brown or red alga. REFERENCES HOLLAND, c. A., LAWSON, j. D. and wALMSLEY, V. G. 1963. The Silurian rocks of the Ludlow District, Shrophire. Bull. Br. Mas. nat. Hist. (Geol.), 8, 95-171. JOHNSON, j. H. 1961. Review of Ordovician Algae. Quart. Colo. Sch. Min. 56 (2). PIA, J. 1927. Thallophyta: 1. Abteilung. In Hirmer M. Handbuch der Paldobotanik. Munich, Berlin. RUEDEMANN, R. 1909. Some marine algae from the Trenton Limestone of New York. Bull. N.Y. St. Mus. 133, 194-210. VALET, G. 1968. Contribution a I’etude des Dasycladales. 1. Morphogenese. Nova Hedwigia, 16, 21-82. WHITFIELD, R. p. 1894. On new forms of marine algae from the Trenton Limestone, with observations on Buthograptus la.xus Hall. Bull. Am. Mus. nat. Hist. 6, 351-358. G. F. ELLIOTT Department of Palaeontology British Museum (N.H.) London, S.W. 7 Typescript received 10 December 1970 REVISION OF SOME LOWER LIAS OSTRACODA FROM YORKSHIRE by ALAN LORD Abstract. Lower Liassic Ostracoda from Redcar, Hotham, and Cliffe are described and assigned to 17 species belonging to 9 genera. This work is substantially a revision of the ostracods described by T. R. Jones (1872) and J. F. Blake (1876). British Lower Jurassic ostracod faunas have received remarkably little attention from palaeontologists, our total knowledge consisting of 5 papers. Jones (1872) described 4 species from the ‘Infralias’ of East Yorkshire in an appendix to a stratigraphic account by Blake, and Blake (1876) described 16 species, mainly from Redcar, including the 4 species of Jones. A more general account of Rhaetic and Liassic ostracods was given by Jones (1894). Anderson (1964) discussed some Lower Jurassic species in a description of Rhaetic ostracods and has attempted to clarify the confusion caused by Jones (1894), who had inadvertently described some Purbeckian forms with Lower Jurassic and Rhaetic taxa. Hettangian Cytherellidae from the Dorset coast were described by Field (1966). Redescription of the species described by Jones (1872) and Blake (1876), all of which are from Yorkshire, has become necessary in view of the frequent reference to these taxa which has been made, often erroneously, by continental workers; the taxo- nomic discussion will serve to illustrate this point. The species of Jones (1872) lacked illustrations but are nevertheless valid; this omission was remedied by Blake (1876) who illustrated his species, and those of Jones, by line drawings. Although these drawings are usually quite good they have rather stylized ornament, are frequently very small, and should therefore be used with caution. It must be emphasized that this paper is in no way a full record of Yorkshire Lower Lias ostracod faunas, but an account of the ostracods found in the 2 main areas from which Jones and Blake obtained their material. The excellent Lower Jurassic section in Robin Hood’s Bay was not sampled. Stratigraphy. Although the material described from Yorkshire by Jones and Blake came from a number of localities the majority of species came from 2 places: the foreshore scars at Redcar, and the area of North and South Cliffe and Hotham, which lie along the line of the Lower Lias escarpment between Market Weighton and the River Humber. Since no type specimens were defined, and since many of the species were recorded impartially from a number of places, it was felt that redescription from these 2 areas was justified by the large numbers of species which the sediments contained. Samples were collected from sections between North Cliffe (SE 873370) and Hotham (SE 885335). The stratigraphy of the rocks in this area has been discussed by Blake (1872), Tate and Blake (1876), Fox-Strangways (1892), and Neale (in de Boer, Neale, and Penny 1958). Blake (1872, p. 140) figured a section showing the Cliffe area and indicating pits which exposed the lowest part of the Lower Lias. Of the 6 pits numbered, only 2 can [Palaeontology, Vol. 14, Part 4, 1971, pp. 642-665, pis. 122-123.] A. LORD: LIAS OSTRACODA 643 now be identified with any certainty — Nos. 5 and 6 in the north-east and south-east angles of Hotham crossroads (SE 884342 and SE 885339). Traces can be seen at the present time of some 8 pits along the scarp slope between North Cave and North Cliflfe, of which the 2 southernmost are pits 5 and 6 of Blake. The Elotham pits are somewhat overgrown but expose ^pvt-planorbis and planorbis zone beds; poor exposure, together with failure to find zonal ammonites, makes it difficult to say which zones are represented in the higher strata. In his section Blake recorded the zones of Ammonites planorbis and Ammonites angulatus, while Neale (1958, p. 161) mentioned 6 feet of beds belonging to the zone of Schlotheimia angulata as the highest strata cropping out in the Cliffe area. Samples were taken from the pits labelled on text-fig. 1 as ‘a’, ‘b’, ‘g’, and ‘h’. Pit ‘a’ is at the crest of the scarp above North Cliffe and the sample from this locality would appear to be stratigraphically the highest taken in this area. The scars at Redcar provide the best exposures of lowest Lower Lias in the north of England. Unfortunately, this section has not been redescribed since the original account of Tate and Blake (1876), and when the samples were collected it was not found possible to follow their divisions. Professor D. T. Donovan and Mr. T. Getty are presently examining the angulata zone and have kindly commented upon the distribution of the samples. The beds belonging to the angulata zone are those to the south of ‘Jenny 644 PALAEONTOLOGY, VOLUME 14 Leigh’s Scar’ from which samples R1 to R4 were taken. ‘Jenny Leigh’s Scar’ marks the base of the biicklandi zone and the remaining samples (R5 to R1 1) are from this zone or higher. When sampling was carried out exposure was limited by beach cover over the scars, and the bucklandi beds were more readily visible because they form more promi- nent scars. It is to be hoped that a full revision of this fine section will not be long delayed. The residues from the 2 sample areas were generally quite rich in microfossils; foraminifera, ostracods, and holothurian spicules were often abundant in the CliflFe and Hotham material, while that from Redcar was usually moderately rich in foraminifera, ostracods, and (frequently) micro-gastropods. The microfauna of sample R7 was almost exclusively composed of micro-gastropods, and occurrences like this may suggest a possible brackish influence in the environment from time to time due to the proximity of a western land mass. The most common ostracods were members of the metacopid genus Ogmoconcha, an ubiquitous faunal element in the Lower and Middle Lias, which appears to have possessed the ability to tolerate a wide range of environment. SYSTEMATIC PALAEONTOLOGY Figured specimens are deposited in the collection of the Department of Geology, University of Hull, and bear the catalogue numbers of that collection. The classification employed is that of the several authors in Moore (1961, Treatise on Invertebrate Paleontology, part Q). A. LORD: LIAS OSTRACODA 645 Subclass OSTRACODA Latreille 1806 Order myodocopida Sars 1866 Suborder cladocopina Sars 1 866 Family polycopidae Sars 1866 Genus polycope Sars 1866 Type species. Polycope orbicularis Sars 1 866, Polycope cerasia Blake 1876 Plate 122, figs. 1, 2 1876 Polycope cerasia Blake, in Tate and Blake, p. 434, pi. 17, fig. 16. 1938 Ostracode (151), Wicher, pi. 27, fig. 9. 1961 Polycope cerasia Tate and Blake; Fischer, p. 500, fig. 1. Material. 2 (73) carapaces, 2 valves. Distribution. Redcar, samples R4, 5, 8, 11. Dimensions (in mm). Length Height Width Valve HU.52.J.1 0-33 0-34 0-10 Carapace HU.52.J.2 0-34 0-36 0-22 Diagnosis. As Blake (1876, p. 434). Description. Shape sub-circular, slight cardinal angles on dorsal margin, anterior margin angled below mid-height at junction with ventral margin, other margins well rounded, although dorsal margin usually straight or only slightly convex. Valves uniformly inflated. Greatest height and width at about mid-height, greatest length at height of antero- ventral angle. Details of hingement, muscle-scars, and marginal zone not observed. Ornament essentially reticulate, but frequently in centre of valve walls of reticulations are thickened so that ornament is almost punctate; reticulation shows sub-concentric arrangement with elongation parallel to circumference at valve margins. In carapace, ornamentation of the 2 valves may differ slightly in strength of reticulation. Remarks. This species was described from the angulatus and bitcklandi zones at Redcar, a distribution in accord with the present record. In the Blake collection in the British Museum (Natural History) there are 5 specimens labelled ‘? Lower Lias, Redcar’ which vary slightly. These specimens are supposedly syntypes and 1 could be selected as lecto- type, but in view of the doubtful origin of this material it would be preferred to erect as neotype 1 of the specimens described above which was collected at Redcar, the only original locality. Blake’s original figure (1876, pi. 17, fig. 16) is misleading since it shows a perfectly round specimen with a coarsely punctate ornament and a marginal rim. On the speci- mens available from Redcar there was no trace of a marginal rim. P. cerasia is distinguished from other Liassic species by its reticulate ornament; however, P. pumicosa Apostolescu from the jamesoni and davoei zones of the southern Paris Basin appears superficially similar in ornament from the illustration of the holo- type (Apostolescu 1959, pi. 1, fig. 1). Further study might reveal a close relationship between the 2 species. C 8385 u u 646 PALAEONTOLOGY, VOLUME 14 Order podocopida Muller 1894 Suborder platycopina Sars 1866 Family cytherellidae Sars 1866 Genus cytherella Jones 1849 Type species. Cytherina ovata Roemer 1840. Cytherella drexlerae Field 1967 1958 Cytherelloidea iuflata Drexler, p. 504, pi. 21, fig. Aa-c. 1966 Cytherella mflata (Drexler); Field, p. 96, pi. 13, figs. 19-22. 1967 Cytherella drexlerae Field, p. 534. Material. 3 male valves. Distribution. Redcar, R6, 9,11. Diagnosis. As Field (1966, p. 96). Remarks. The material is closely comparable to that described from England by Field (1966). The species differs from Cytherella concentrica Field (1966) only in as far as the latter possesses a reticulate ornament, and it is thought likely that the 2 species are sub-species of the same species. Cytherella concentrica Field 1966 Plate 122, figs. 3, 4 1952 Ostracode 8, Usbeck, p. 406, pi. 18, fig. 65. 1966 Cytherella concentrica Field, pp. 97-99, pi. 13, figs. 23-27. Material. 3 carapaces, 1 valve. Distribution. Redcar, samples R6, 8, 10, 11; North Cliffe, pit ‘a’. Dimensions (in mm). Length Height Width Carapace HU.52.J.3 0-70 0-42 0-26 EXPLANATION OF PLATE 122 All figures X 50. Figs. 1, 2. Polycope cerasia Blake. Carapace, HU.52.J.2, Redcar — R5; 1, left view; 2, dorsal view. Figs. 3, 4. Cytherella concentrica Field. Carapace, HU.52.J.3, Redcar — RIO; 3, right view; 4, dorsal view. Figs. 5-7. Cytherelloidea pulchella Apostolescu. 5, 7, Carapace, female, HU.52.J.9, pit ‘a’; 5, left view; 7, dorsal view; 6, right valve, female, HU.52.J.8, pit ‘a’, external view. Figs. 8, 9. Cytherelloidea circumscripta (Blake). Carapace, female, HU.52.J.5, pit ‘b’; 8, left view; 9, dorsal view. Figs. 10-12. Bairdia aff. B. Apostolescu. 10, 12, Carapace, F1U.52.J.11, pit ‘b’; 10, right view; 12, dorsal view. 11, Right valve, HU. 52. J. 10, pit ‘b’, internal view. Fig. 13. Bairdia cf. B. carinata Drexler. Carapace, HU.52.J.13, Redcar — Rll, left view. Figs. 14, 15. Bairdia tatei Coryell. Carapace, HU.52.J.15, Redcar — R8; 14, right view; 15, dorsal view. Figs. 16, 17. Bairdiacyprisl sartriensis Donze. Left valve, HU.52.J.16, pit ‘h’; 16, external view; 17, dorsal view. Figs. 18-21. Klinglerella moorei (Jones). 18, 20, Left valve, HU.52.J.60, pit ‘b’; 18, external view; 20, dorsal view. 19, 21, Right valve, HU.52.J.61, pit ‘b’; 19, external view; 21, dorsal view. Palaeontology, Vol. 14 PLATE 122 II 14 17 15 LORD, Lower Lias Ostracoda m' : Ai‘ "\'...-^-i '^^idi^M .J-' '■ m- I i •tfj< lt.x' j|--Vjj;'; , :;it^’\llw - jij! ,j'4 Hi H ""^ !)•• i' (••^■1 i)l U.:‘ '%. ■.vjr. I'f* i'T-iit A. LORD: LIAS OSTRACODA 647 Diagnosis and Description. As Field (1966, p. 97). Remarks. Despite the presenee of ornament, Field assigned this species to Cytherella on the grounds that in females there is only 1 posterior internal depression, whereas females of Cytherelloidea possess 2. In certain respects the morphology of this species corresponds with the description of Cythere bJakei Jones (1872) and it is possible that the 2 are synonymous. C. concentrica was recorded from the angidata and bucklandi zones in Dorset but was found only in the bucklandi zone in Yorkshire. Blake (1876, p. 431) recorded Cythere bJakei from the planorbis zone at Cliffe and the angulatus zone at Redcar. Genus cytherelloidea Alexander 1929 Type species. Cythere {Cytherella) williamsoniana Jones 1849. Cytherelloidea circumscripta (Blake 1876) Plate 122, figs. 8, 9 1876 Cytherella circumscripta Blake, in Tate and Blake, p. 434, pi. 17, fig. 14. 1966 Cytherelloidea circumscripta (Blake); Field, pp. 88-93, pi. 11, figs. 1-8; pi. 12, figs. 9-13; text-fig. 2. 1966 Cytherelloidea circumscripta (Tate and Blake); Donze, p. 127, pi. 6, fig. 47. 1966 Cytherelloidea cf. circumscripta (Tate and Blake); Donze, pp. 127-128, pi. 6, fig. 46. Material. 2 carapaces, 63 valves. Distribution. Redcar, samples R2, 5, 10, 11. Hotham, pit ‘h’. South Clift’e, pit ‘b’. North Cliffe, pit ‘a’ Dimensions (in mm). Length Height Width Left valve, female HU.52.J.4 0-75 0-43 016 Carapace, female HU.52.J.5 0-74 0-43 0-26 Left valve, male HU.52.J.6 0-68 0-40 Oil Right valve, male HU.52.J.7 0-69 0-45 012 Diagnosis and description. Description by Blake (1876, p. 434), short diagnosis and description as Field (1966, p. 89). See text-fig. 3. Remarks. C. circumscripta was described and discussed in some detail by Field (1966), who found distinct stratigraphic variation in ornament in the species in the Lower Lias of the Dorset coast. Ornamental differences between specimens, including both males and females, from Redcar and East Yorkshire are shown in text-fig. 3. Female specimens from East Yorkshire resembled Field’s Variant II but often showed traces of an antero- median rib which occurs in his Variant I and presumably represents an intermediate stage in development. Redcar specimens resembled Variant IV and all appeared to be female; they are certainly from a younger horizon than those from East Yorkshire. As Eield noted, his Variant III most closely resembled the ostracod illustrated by Blake (1876, pi. 17, fig. 14). Sexual dimorphism is plainly discernible by ornamental differences and by the posterior inflation of the females, which is caused by 2 internal depressions, called ‘brood cavities’ by van Morkhoven (1963, p. 23). Usbeck (1952, p. 404) placed this species in Cytherelloidea without any discussion. His specimens came from the Swabian Lias alpha 1-2 and from his figured specimen 648 PALAEONTOLOGY, VOLUME 14 (pi. 18, fig. 5Sa) the assignment is correct, but his fig. 5Sb is CythereUoidea pulchella Apostolescu (1959). CythereUoidea valsoldensis Conti (1954) seems to be the same as C. circumscripta. The species, originally described from the angidatus zone at Redcar, was recorded by Field (1966) from the angidata and bucklandi zones of the Dorset coast, while the Yorkshire material was found in the planorbis, angidata, and bucklandi zones. The Yorkshire distribution is comparable with that of material from Siebeldingen/Pfalz described by Drexler (1958). TEXT-FIG. 3. Ornamental variation in CythereUoidea circumscripta (Blake). CythereUoidea pulchella Apostolescu 1959 Plate 122, figs. 5, 7 1959 CythereUoidea pulchella Apostolescu, p. 802, pi. 1, figs. 4-6. 1964 CythereUoidea pulchella Apostolescu; Contini and Pariwatvorn, p. 38, fig. 4. 1966 CythereUoidea pulchella Apostolescu; Field, pp. 93-96, pi. 12, figs. 14-18. 1966 CythereUoidea buisensis Donze, pp. 126-127, pi. 6, figs. 35^5. Material. 1 carapace, 7 valves. Distribution. Redcar, Sample R2. North Clifife, pit ‘a’. South ClifiTe, pit ‘b’. Hotham, pit ‘h’. Dimensions (in mm). Length Height Width Right valve, female HU.52.J.8 0-61 0-36 0T2 Carapace, female HU.52.J.9 0-65 0-38 0-24 Diagnosis. As Apostolescu (1959, p. 802). Description. As Field (1966, p. 93). A. LORD: LIAS OSTRACODA 649 Remarks. C. pulchella differs from C. circumscripta in its smaller size, more oval shape, and in the pattern of ornament — a feature more constant than in C. circumscripta, where ornament can vary with sex, valve, or horizon. The species has been recorded from the Rhaetic (Anderson 1964), Lias alpha (Usbeck 1952), planorbis zone (Donze 1966), angulata zone (Field 1966) and bucklandi zone (Apostolescu 1959). Donze’s (1966, pp. 126-127) species C. buisensis differs in outline from the type and more closely resembles the illustration given by Anderson (pi. 13, figs. 74-78). These 2 records seem to represent a more primitive stage within the species. Anderson (p. 151) noted that this species may be the same as Cythere terquemicma Jones (1872), but it does not fit the original description and, as is described below, there is reason to suppose that specimens of Cythere terquemiana have been found in the type area. It is tempting to identify C. pulchella with Blake’s Cytherella crepidula (see Field 1966, p. 94) and similarities in shape and ribbing exist if the original drawing of C. crepidula is regarded as being somewhat impressionistic. However, C. crepidula was only recorded from 1 locality and 1 horizon (the Ammonites capricornus zone at Huntcliff, near Staithes), and this leaves a gap of over 1 stage between the highest record of C. pulchella and the recorded occurrence of C. crepidula in Yorkshire. The validity of this comparison can only be tested when original or topotype material of Blake’s species has been found. The original material is missing and in the course of this study no examples were found. Suborder podocopina Sars 1 866 Superfamily bairdiacea Sars 1888 Family bairdiidae Sars 1888 Genus bairdia M’Coy 1844 Type species. Bairdia curtiis M’Coy 1844. Remarks. 36 specimens of this genus were found at Redcar. The species described below as Bairdia tatei Coryell (1963) was readily distinguishable, but the remaining specimens showed variation in shape and could not be assigned to existing species although, in certain respects, a few resembled Bairdia molesta Apostolescu (1959). Possibly 1 or 2 species are present and in view of their variation and poor state of preservation it is preferable to leave them under open nomenclature until more material is available. None of the specimens of Bairdia resembled Bairdia ! hettangica Donze (1966) from the Hettangian of Ardeche. Bairdia aff. B. molesta Apostolescu 1959 Plate 122, figs. 10-12 aff. 1959 Bairdia molesta Apostolescu, pp. 806-807, pi. 2, fig. 31. Material. 1 carapace, 3 valves. Distribution. South Cliffe, pit ‘b’. Dimensions (in mm). Length Height Width Right valve HU.52.J.10 0-67 0-37 0 12 Carapace HU.52.J.11 0-72 0-41 0-29 Discussion. In shape the specimens are very similar to B. molesta, but have a slightly more rounded aspect. While some variation in shape is found in B. molesta through its 650 PALAEONTOLOGY, VOLUME 14 range (Lower Sinemurian to Domerian), these specimens seem to represent a pre- B. molesta stage from which the species evolved during the late Hettangian, and they are accordingly placed as B. alf. B. molesta. Bairdia cf. B. carinata Drexler 1958 Plate 122, fig. 13 cf. 1958 Bairdia carinata Drexler, pp. 512-513, pi. 22, figs. 2a-c. Material. 2 carapaces. Distribution. Redcar, sample Rll. Dimensions (in mm). Length Height Width Carapace HU.52.J.12 0-59+ 0-31 0-22 Carapace HU.52.J.13 0-65 0-31 0-24 Discussion. Both specimens have an extended posterior end and most closely resemble Drexler’s pi. 22, fig. 2b \ they are somewhat smaller than Drexler’s specimens and may perhaps be instars, perhaps the moult stage immediately preceding the adult, or the one prior to that. Bairdia tatei Coryell 1963 Plate 122, figs. 14, 15 1876 Bairdia elongata Blake, p. 431, pi. 17, fig. 5. 1963 Bairdia tatei Coryell, p. 462. 1966 Isobythocyprisl tatei (Coryell); van den Bold, pp. 222-223. 71959 llsobythocypris elongata (Tate and Blake); Apostolescu, p. 808, pi. 2, figs. 24, 25. 71964 Pontocyprella elongata (Blake); Contini and Pariwatvorn, figs. 4, 6, 8. 71966 Isobythocypris elongata (Blake); Magne and Obert, pp. 267, 271. non 1958 Bythocypris cf. elongata (Tate and Blake); Drexler, pp. 515-516, pi. 23, figs. 1 a~f\ pi. 24, figs. 7-9; pi. 27, figs. 1, 2. non 1964 Hungarella elongata (Blake); Anderson, pp. 148, 149; pi. 15, figs. 118-121. Material. 5 carapaces, 1 valve. Distribution. Redcar, samples R4, 5, 8. Dimensions (in mm). Length Height Width Carapace HU.52.J.14 0-72+ 0-39 0-26 Carapace HU.52.J.15 0-78 0-39 0-27 Diagnosis. As for Bairdia elongata Blake (1876, p. 431). Description. Valves uniformly inflated with greatest width and height about mid-length, and greatest length just below mid-height. Left valve larger than right, which is over- lapped dorsally and along mid-part of ventral margin. Dorsal margin gently arched but ventral margin straight with slight concavity just anterior of mid-length. Anterior margin evenly rounded, but posterior short and low and dorsal margin curves steeply down to meet it. Internal details not seen, but on 1 carapace line of inner margin was visible, with narrow vestibule at anterior end and a few straight marginal pore canals. Details of muscle-scars and hingement not seen. Valve surface smooth and unorna- mented. A. LORD: LIAS OSTRACODA 651 Remarks. The specimens described were most closely comparable to Blake’s species B. elongata. This species seems to be a genuine Bairdia; the forms preceded by an interro- gation mark in the synonymy, however, apparently do not belong in that genus, because, from the illustrations available, they differ in shape, especially posteriorly. Many of the species in the synonymy seem to belong to the group discussed below as Pontocyprella, but the situation is confused and the exact position of Isobythocypris within this group of smooth elongate-oval ostracods is obscure. Anderson (1964, pp. 148, 149) described a species from the ? basal Lower Lias as Himgarella elongata (Blake) and also figured (pi. 15, fig. 119) a specimen as lectotype (BMNH Io3194, Blake Collection, biickkmdi zone, Redcar). The evidence that this specimen is Blake’s species is unconvincing, and it is unlike any species found from the Redcar samples, with the possible exception of Ogmoconeha sp. A. Comparison of Blake’s figure (1876, pi. 17, fig. 5) with the lectotype, and with Anderson’s figures (1964, pi. 15, figs. 118-121) of the lectotype and a specimen (GSM Mik (J) 285001) from the lowest Lower Lias of the Plattlane Borehole, Shropshire, is not sufficiently close to establish convincingly that they belong to the same species. Blake’s figure shows a carapace which appears to be elongate-oval in shape with a straight dorsal margin which curves down to the low posterior; the ventral margin appears straight. The ventral and posterior margins merge into the dorsal margin, the anterior margin being essentially a continuation of the venter. The posterior end of the shell is rounded and rather acuminate, its most distal point being below mid-height. The larger left valve overlaps the right valve all round and no details of hingement, muscle-scars, or pore canals are visible. Blake’s description (1876, p. 431) is vague: ‘Carapace large, long, like a narrow bean; ends nearly alike, sharply curved ; ventral side slightly concave, dorsal side greatly convex, not compressed.’ The specimens described here are regarded as morphologically closer to B. elongata than the lectotype erected by Anderson. Anderson’s figures show an ostracod which differs from B. elongata in possessing a more convex dorsal margin and differently shaped anterior and posterior ends. Admittedly, shape is not always a reliable criterion when knowledge of variation within the species is lacking, but in this case it provides the only evidence. Drexler (1958, pp. 515, 516) compared specimens from the Lias alpha 3 of Pfalz with this species and placed them in the genus Bythocypris, but Apostolescu (1959, p. 808) considered the species as possibly belonging to his new genus Isobythocypris. Several authors in the Lias Colloque (1961) called this species Pontocyprella elongata (Blake) and were followed by Oertli (1963) and Contini and Pariwatvorn (1964), although Magne and Obert (1966) followed Apostolescu and used the name Isobytho- cypris elongata (Blake) for ostracods from the oxynotum zone of Arbois and the Domerian of Andelot in eastern France. A complication is that numerous homonyms exist for Bairdia elongata and Bytho- cypris elongata. As has been shown by Coryell (1963) and van den Bold (1966), Bairdia elongata Blake is a homonym of Bairdia elongata (Munster 1830) and as such is not available for use, so that Coryell’s new name Bairdia tatei Coryell (1963, p. 462) has priority over Himgarella elongata (Blake) of Anderson (1964, p. 148). Genus bairdiacypris Bradford 1935 Type species. Bairdiacypris deloi Bradfield 1935. 652 PALAEONTOLOGY, VOLUME 14 Bairdiacyprisl sartriensis Donze 1966 Plate 122, figs. 16, 17 ?1952 Ostracode 2, Usbeck, p. 404, pi. 18, fig. 60. 1966 Bairdiacyprisl sartriensis Donze, p. 131, pi. 7, figs. 81-86, table 2. Material. 1 valve (adult), 1 carapace (instar). Distribution. Hotham, pit ‘h’, planorbis zone. Dimensions (in mm). Length Height Width Left valve HU.52.J.16 0-75 0-35 0 14 Holotype, carapace (from Donze) 0-75 0-35 — Diagnosis. Donze (1966, p. 131). Description. Oval-elongate in shape; tripartite dorsal margin, anterior section inclined antero-ventrally, median section inclined slightly towards posterior, and posterior section short, curving round to merge with rounded posterior margin. Anterior margin low, rounded. Ventral margin medianly concave, giving sinuous outline to valve. Left valve larger than right, and overlaps it at anterior and posterior ends of dorsal margin and along mid-part of ventral margin. Greatest height at anterior end of mid-part of dorsal margin in larvae, but at about mid-length in adult; greatest length at mid-height and greatest width close to posterior end. Valve surface smooth, unornamented. Hinge in adult simple, adont. No details of muscle-scar pattern seen. Remarks. The left valve was encrusted internally with small calcite crystals which obliterated any trace of muscle scars; the duplicature seemed to be very narrow, suggest- ing that the species does not belong to the family Bairdiidae. The 2 specimens are closely comparable to the original figures of Donze (1966, pi. 7, figs. 81-86) and the adult has the same dimensions as the holotype (p. 131). The assign- ment of this species to Bairdiacypris may be correct; certainly it compares quite well with illustrations of the type species B. deloi from the Carboniferous, but it differs in its possession of a more rounded posterior margin. Bairdia dispersa Blake (1876) may be the same as this species, although it comes from a higher horizon than any from which B.l sartriensis has been recorded. Ostracods from the Hoganas Series of Scania were placed in B. dispersa by Troedsson (1951, p. 245) but the validity of the assignment is questionable. Superfamily cypridacea Baird 1845 Family paracyprididae Sars 1923 Genus paracypris Sars 1866 Type species. Paracypris polita Sars 1866. Paracypris cf. P.l semidisca Drexler 1958 cf. 1958 Paracypris^. semidisca Drexler, pp. 519, 520, pi. 23, figs. Aa-d. 1966 Paracyprisl cf. semidisca Drexler; Donze, p. 132, pi. 7, figs. 69-74. Material. 1 left valve. Distribution. Redcar, sample R8. A. LORD: LIAS OSTRACODA 653 Discussion. A single, poorly preserved, left valve with a broken posterior end. Dorsal margin convex, curving down to a low rounded anterior margin and to the acuminate posterior. Ventral margin almost rectilinear. Surface smooth and unornamented. No other details visible. Apparently belonging to Paracypris, this specimen most closely resembles Drexler’s P. ? semidisca. In her synonymy Drexler tentatively included Bairdia lacrimae Blake (mis-spelling for B. lacryma) but this differs in shape especially with regard to the posterior margin, which is distally extended in the original figure (Blake 1876, pi. 17, fig. 3). The identification of B. locryma with P.l semidisca is a possibility but one which for the moment must remain unsettled. Barbieri (1964) compared ostracods from the top of the Carixian of Sicily with B. lacryma but from his illustration the identification seems doubtful. Genus pontocyprella Lyubimova 1955 Type species. Bairdia harrisiana Jones 1 849. A number of poor specimens from Redcar resemble in shape, overlap, and smooth valve surface the species Pontocyprella aureola Lyubimova 1955 or Pontoeyprella akta- gensis Mandelstam 1956; they resemble the type species less closely. More than 1 genus is represented by the species which have been placed in Pontocyprella. As in other cases cited previously, any definitive naming of the material is precluded by the small number of poor specimens. Superfamily cytheracea Baird 1850 Family progonocytheridae Sylvester- Bradley 1948 Genus klinglerella Anderson 1964 Type species. Procytheridea glabellata Klingler and Neuweiler 1959. Remarks. Klinglerella was first described by Anderson (1964, p. 141), who questionably placed 2 Rhaetic species in the genus. However, in his discussion of the genus, Anderson noted a number of Procytheridea species, including many of those described by Klingler and Neuweiler (1959), which he believed to belong to Klinglerella. The genus appears to represent a varied and important Lower Lias group of ostracods, many of the species of which have been erroneously placed in Procytheridea in the past. The problem of Procytheridea will be discussed elsewhere; suffice it to say that Procytheridea does not oecur in the Lower Jurassic or, probably, anywhere in Europe. Klinglerella moorei (Jones 1872) Plate 122, figs. 18-21 1872 Cythere moorei, G. S. Brady MS, Jones, p. 146. 71908 Bairdia moorei Jones; Issler, p. 95, fig. 343, non 344. 71952 Ostracode 4, Usbeck, pp. 404, 405, pi. 18, fig. 616, non 61a, c, d. 7par. 1959 Procytheridea litxuriosa Apostolescu, pp. 808, 809, pi. 2, figs. 33-36. 1966 Procytheridea praeluxuriosa Donze, pp. 133, 134, pi. 7, figs. 87-91. Material. 2 carapaces, 17 valves. Distribution. South Cliffe, pit ‘b’. 654 PALAEONTOLOGY, VOLUME 14 Dimensions (in mm). Left valve HU.52.J.58 Length 0-49 Height 0-30 Width 014 Left valve HU.52.J.59 0-52 0-31 0-15 Left valve HU.52.J.60 0-50 0-28 015 Right valve HU.52.J.61 0-53 0-26 0-14 Diagnosis. Jones (1872, p. 146). Description. Shape oval. Dorsal margin straight or slightly convex and gently inclined towards posterior; posterior low and rounded, anterior margin almost straight and gently inclined downwards for short distance from anterior cardinal angle, then broadly rounded; ventral margin straight or slightly convex with median selvage groove evident in some specimens. In side view ventral margin hidden by ventral inflation. Valves inflated, with low rim bordering anterior margin, which is ornamented with a weak peripheral rib. This frequently has minor ribs running into it causing rim to be split up into small sections. Left valve larger and higher than right. Greatest length at mid-valve height, greatest height at anterior cardinal angle, and greatest width medianly or postero-medianly. Valve surfaces ornamented with pattern of ribs which is essentially triangular, basal set being longitudinal with minor ribs and in centre an area of reticulate ornament produced by intersection of different rib trends. Ribs on ventral side of shell and on ventral surface of inflation are longitudinal. Posterior margin low, unornamented. Hinge merodont; in left valve an anterior loculate groove with 6 sockets, median bar, and posterior loculate groove with 6 sockets; and in right valve an anterior dentate ridge, median groove, and posterior dentate ridge. Marginal zone moderately wide, inner margin and line of concrescence coincide. Marginal pore canals simple and straight. No sexual dimorphism evident. Remarks. The species corresponds well with that described by Jones (1872). Blake (1876) quoted Jones in his re-description but added that the ornament was ‘. . . like parallel irregularly carved furrows’, and his illustration (pi. 17, fig. 9) does not really match the original description. It seems likely that Blake described a form which is close to, or identical with, the species described by Apostolescu (1959) as Procytheridea luxuriosa, from the Sinemurian (biicklandi and semicostatum zones) of the Paris Basin. 1 specimen close to this latter species was found at Redcar in the youngest fossiliferous horizon sampled (sample Rll, left valve, HU.52.J.62, length 0-62 mm, height 0-35 mm, width 0T7 mm). At the type area (South Cliffe) P. luxuriosa was not found and the species described above, which fits Jones’s original description, must be taken as Cythere moorei. Klinglerella moorei is thought to be closely related to P. luxuriosa. In the ‘Colloque sur le Lias Francais’ (1961) ostracods have been compared with this species by G. Bizon and Oertli (p. 116), G. Bizon (p. 436), Champeau (p. 438), and J. J. Bizon (p. 452). Their forms, illustrated by drawings and without descriptions, appear to show a significant similarity in outline and ornament to each other and to the species which is regarded here as KlingJerelia moorei. The ostracod illustrated by Usbeck (1952, pi. 18, fig. 6lb) from the Lias alpha of Swabia may well be C. moorei, but Usbeck’s photograph is poor and no firm decision is possible. Issler (1908, p. 95) placed the species in Bairdia, and Apostolescu (1959, p. 808) tentatively included it in Procytheridea luxuriosa. A. LORD: LIAS OSTRACODA 655 Klinglerella aff. K. triebeli (Klingler and Neuweiler 1959) Plate 123, figs. 1-3 aff. 1959 Procytheridea triebeli Klingler and Neuweiler, pp. 381, 382, pi. 13, figs. 11-16; pi. 14, figs. 17, 18. Material. 4 carapaces, 34 valves. Distribution. South Cliffe, pit ‘b’. Redcar, sample R6. Dimensions (in mm). Length Height Width Left valve, male HU.52.J.17 0-59 0-33 0-16 Right valve, male HU.52.J.18 0-61 0-32 014 Left valve, female HU.52.J.19 0-50 0-31 013 Right valve, female HU.52.J.20 0-51 0-29 014 Carapace, female HU.52.J.21 0-53 0-32 0-27 Description. Shape oval; dorsal margin straight but inclined posteriorly; from dorsal cardinal angle anterior margin is straight and moderately inclined, then curves down, becoining rounded but slightly asymm_etric. At its posterior end dorsal margin joins rounded posterior margin which in turn joins ventral margin; latter has median con- cavity and selvage groove. Shape of right, or smaller, valve differs a little from that of left, particularly in males which have rather more strongly angled dorsal margin. Left valve larger than right. Greatest length at, or just below, mid-height; maximum height just anterior of mid-length and maximum width medianly or just a little to posterior. Slight ventral inflation evident. Sexual dimorphism present; presumed females shorter, higher, and appear more rounded than males. Anterior and posterior margins possess narrow bordering rims, rest of valve being inflated except for short groove which runs from near anterior end of dorsal margin for short distance in direction of antero- ventral margin. This groove may be weakly or strongly developed. Surface of valves ornamented with ribs forming an open reticulate pattern of 4 or 5 sided cells. Some ribs are rather strong and may be traced across valve. Towards margins ornament becomes weaker and may degenerate into weak punctation. By nature of their preservation some specimens show little ornament even when stained. Hinge hemimerodont ; in left valve an anterior loculate groove, (?) smooth median bar, and posterior loculate groove, matched in right valve by an anterior dentate ridge, median furrow, and posterior dentate ridge. Median bar in left valve and median furrow in right valve may be respectively denticulate and locellate, but this is difficult to ascer- tain due to preservation and may mean that hinge is antimerodont rather than hemi- merodont. Adductor muscle scar pattern a vertical row of 4 scars, largest being the central 2, with a single V-shaped frontal scar. Normal pore canals simple, about 45 in number, fairly evenly distributed over valve surface. Marginal pore canals few, short, straight. Marginal zone quite wide anteriorly, less so on ventral and posterior margins; line of concrescence and inner margin coinci- dent. Sexual dimorphism apparent. Remarks. This record is from an horizon lower than that given for K. triebeli by Klingler and Neuweiler (1959). The specimens are also smaller than the type material. For these reasons and also because of the poor ornamentation on some of the specimens, these ostracods are regarded as K. aff. K. triebeli rather than Klinglerella triebeli [sensu strict o\ 656 PALAEONTOLOGY, VOLUME 14 The smooth nature of some of the specimens may suggest a comparison with Cythere blakei Jones despite differences from the original description. This group of ostracods differs from Klinglerel/a moorei in shape, ornament and the presence of dimorphism. The specimens from Redcar are more inflated than those from Cliffe and may not belong to the same taxon. Kliuglerellal translucens (Blake 1876) Plate 123, figs. 4, 5 1876 Cythere translucens Blake, pp. 432, 433, pi. 17, fig. 10. 71923 Bairdia translucens (Tate and Blake); Pratje, p. 253. non 1908 Bairdia translucens (Tate and Blake); Issler, p. 94, fig. 340. Material. 3 carapaces, 2 valves. Distribution. Redcar, sample R6. Dimensions (in mm). Length Height Width Carapace HU.52.J.22 0-62 0-34 0-28 Carapace HU.52.J.23 0-65 0-34 0-29 Diagnosis. Blake (1876, pp. 432, 433). Description. Shape elongate, oval-triangular. Dorsal margin rectilinear and inclined posteriorly from anterior margin. Ventral margin straight, but obscured in lateral view by weak ventral inflation. Posterior margin short, rounded, slightly asymmetric. Anterior and posterior margins have bordering rims, while rest of valve is gently inflated with exception of flatter area in antero-dorsal position. Left valve larger than right. Position of greatest height at anterior cardinal angle, of greatest length slightly below mid-height, and of greatest width posteriorly. Valve surfaces irregular with no distinct ornament discernible except on ventral sides of ventral inflations where weak longitudinal ribs may be observed. Some surface irregularities may be expression of normal pore canals. No internal features observed. EXPLANATION OF PLATE 123 All figures x 50. Figs. 1-3. Klinglerella afF. K. triebeli (Klingler and Neuweiler). 1, Left valve, female, HU.52.J.19, pit ‘b’, external view. 2, 3, Right valve, male, HU.52.J.18, pit ‘b’; 2, external view; 3, dorsal view. Figs. 4, 5. Klinglerella'I translucens (Blake). 4, Carapace, HU.52.J.22, Redcar — R6, left view. 5, Carapace, HU.52.J.23, Redcar — R6, dorsal view. Figs. 6-8. ^Cythere' terquemiana Jones. 6, Left valve, HU.52.J.25, pit ‘b’, external view. 7, 8, Left valve, HU.52.J.24, pit ‘b’; 7, external view; 8, dorsal view. Figs. 9-13. Ogmoconcha ellipsoidea (Jones). 9, Left valve, HU.52.J.32, pit ‘g’, external view. 10, Right valve, HU.52.J.33, pit ‘g’, external view. 1 1, Right valve, HU. 52. J. 37, pit ‘a’, internal view. 12, Left valve, HU.52.J.36, pit ‘a’, internal view. 13, Carapace, HU. 52. J. 30, pit ‘g’, dorsal view. Figs. 14-16. Ogmoconcha hagenowi Drexler. 14, Left valve, HU.52.J.50, Redcar — R9, external view. 15, Right valve, HU.52.J.49, Redcar — R9, external view. 16, Carapace, HU.52.J.44, Redcar — R8, dorsal view. Figs. 17, 18. Ogmoconcha sp. A. 17, Left valve, HU.52.J.55, Redcar — R6, external view. 18, Cara- pace, HU.52.J.53, Redcar — R6, right view. Palaeontology, Vol. 14 PLATE 123 LORD, Lower Lias Ostracoda V "V '• ) ( A A. LORD: LIAS OSTRACODA 657 Remarks. The weak or indeterminate ornament precludes a firm specific identification since the original description mentions a network of distinct ribs. Nevertheless, the original figure shows an ostracod with an irregular surface, and whilst the original figure and description may not be reliable, for the present these specimens are placed in C. transhicens because of their similarity to the original figure and their occurrence in the bucklaudi zone at Redcar. The material differs from any of the species described by Klingler and Neuweiler (1959). Issler (1908, p. 94) assigned this species to the genus Bairdia. Family Uncertain Remarks. The species described below is considered to belong to a new and undescribed genus. Since only 2 specimens were found the original nomenclature of Jones (1872) has been retained pending collection of more material. 'Cythere' terquemiana Jones 1872 Plate 123, figs. 6-8 1872 Cythere terquemiana Jones, p. 147. 1876 Cythere terquemiana Jones; Blake, p. 432, pi. 17, fig. 7. Material. 2 valves. Distribution. South Cliflfe. Dimensions (in mm). Length Height Width Left valve HU.52.J.24 0-45 0-24 010 Left valve HU.52.J.25 0-45+ 0-26 O il Diagnosis. Jones (1872, p. 147). Description. Shape sub-rectangular. Dorsal margin concave, margin being highest at anterior cardinal angle; anterior margin broadly rounded; ventral margin slightly concave; posterior rounded with an angulation at mid-height. Anterior and posterior margins bordered by low flat rims. Anterior marginal rim has low peripheral rib with minor ribs running from it in posterior direction, thus subdividing rim into approxi- mately 9 small cells. Posterior rim is flat, unornamented. Rest of valve inflated, except for sulcus, which starts from antero-median part of dorsal margin, behind inflated and raised area at anterior cardinal angle, and runs down to antero-median area of valve. Surface of inflated area irregular, with tubercle at postero-ventral extremity, whilst traces of another (which does not appear to be an eye-spot) exist at anterior cardinal angle, and postero-dorsal area is a complex ridged region which is in part tuberculate. Larger of 2 valves studied was more developed and showed signs of tubercles in median and ventral areas also. Surfaces of valves and postero-ventral tubercles were covered with coarse and irregular reticulate pattern. Internal features poorly preserved. Hinge tripartite, and in left valve has terminal sockets and median (?) smooth bar; impossible to say whether the hinge is lophodont or hemimerodont. Remarks. To the original description Blake (1876, p. 432) added ‘Elevated on the anterior and postero-ventral portions’, which matches the most inflated parts of the specimens described here. 658 PALAEONTOLOGY, VOLUME 14 The genus is reminiscent of Lophodentina, from which it differs in shape, especially posteriorly, and in its tuberculate character. The 2 may well be closely related, although how closely is not possible to say without more details of the internal features. The resemblance to Trachycy there is also striking in shape and ornament in that both possess tubercles and reticulation, although of different strength. Sohn (1968, p. 14) postulated an evolutionary sequence of CoruigeUa-JudaheUa-Trachycythere-Orthonotacythere based on shape, size, ornament, and a suggested hinge development. It is conceivable that the species here described belongs to, or is an offshoot from, the lineage proposed by Sohn, but it is impossible to be certain. It is also possible that the Lower Rhaetic Trachycytherel tuber osa Anderson is also closely linked; however, on balance it would seem unlikely that Anderson’s species belongs in Trachycythere [semu stricto]. The material closely resembles Cythere terquemiana in its original description and in the original figure, which is an impression rather than an accurate representation, and the identification is regarded as fairly certain. Suborder metacopina Sylvester-Bradley 1961 Superfamily healdiacea Harlton 1933 Family healdiidae Harlton 1933 Genus ogmoconcha Triebel 1941 Type species. Ogmoconcha contractula Triebel 1941. Remarks. It is inappropriate to discuss the details of the possible synonymy of the Lower Jurassic Ogmoconcha with the Triassic genus Himgarella Mehes (1911), and it is intended to examine this problem elsewhere. The name Ogmoconcha is employed because the genus was described from the Lower Jurassic and represents an important group of Lias ostracods to which the species described below certainly belong. 1872 1876 1938 71952 71954 71954 71958 71959 71961 7par. 1961 71961 71961 7par. 1961 71962 71964 71964 non 1894 Ogmoconcha ellipsoidea (Jones 1872) Plate 123, figs. 9-13 Bairdia (7) ellipsoidea G. S. Brady MS, Jones, p. 146. Bairdia liassica Brodie; Blake, p. 430, pi. 17, fig. 1. Ostracode (521), Wicher, pi. 27, fig. 1. Ogmoconcha sp. (1), Usbeck, p. 404, pi. 18, figs. 59a, b. Cytheridea ellipsoidea (Jones, Brady MS) var. rediicta Conti, p. 228, pi. 12, figs. 1-5. Cytheridea ellipsoidea (Jones); Conti, p. 229, pi. 12, figs. 6-8. Healdia aspinata Drexler, pp. 505, 506, pi. 21, figs. 5a-e\ pi. 25, figs. 1-4. Ostracoda B, Apostolescu, p. 817, pi. 2, figs. 20-23. Ostracoda B, Apostolescu; Cousin and Apostolescu, p. 428, table 1. Himgarella sp. B (Apostolescu); G. Bizon, p. 433, table 2. Ostracode sp. (200), Champeau, p. 438, table 3. Ostracoda B, Apostolescu; Apostolescu, p. 448, table 4. Himgarella sp. B. (Apostolescu); J. J. Bizon, pp. 452, 455, table 5. Healdia aspinata Drexler; Klingler, p. 79, table 7, pi. 12, figs. 1, 2. Ogmoconchella aspinata (Drexler); Griindel, pp. 470, All, figs. 5-7. Himgarella owthorpensis Anderson, pp. 147, 148, pi. 14, figs. 96-101. Cytheridea ellipsoidea (Jones); Jones, pp. 164, 165, pi. 9, figs. 6a-c. Material. 97 carapaces, 518 valves. A. LORD: LIAS OSTRACODA 659 Distribution. Redcar, samples R4, 8, 9, 11. North Cliffe, pit ‘a’ and hillside excavation. South ClifTe, pit ‘b’. Hotham, pits ‘g’, ‘h’. From planorbis, angulata and bucklandi zones. Dimensions (in mm). Length Height Width Hotham, pit ‘h’ Left valve HU.52.J.26 0-58 0-43 0-18 Right valve HU.52.J.27 0-50 0-32 0-13 Right valve HU.52.J.28 0-39 0-25 012 Right valve HU.52.J.29 0-32 019 007 Hotham pit ‘g’ Carapace HU.52.J.30 0-59 0-42 0-32 Left valve HU.52.J.31 0-54 0-39 0-18 Left valve HU.52.J.32 0-55 0-39 0-17 Right valve HU.52.J.33 0-54 0-35 015 Right valve HU.52.J.34 0-52 0-34 0-16 North Cliffe, pit ‘a’ Left valve HU.52.J.35 0-59 0-42 018 Left valve HU.52.J.36 0-59 0-43 019 Right valve HU.52.J.37 0-55 0-36 0-14 Right valve HU.52.J.38 0-56 0-36 014 Redcar, sample Rll Carapace HU.52.J.39 0-56 0-38 0-31 Left valve HU.52.J.40 0-59 0-43 019 Right valve HU.52.J.41 0-59 0-39 017 Right valve HU.52.J.42 0-53 0-35 015 Right valve HU.52.J.43 0-48 0-29 0-14 Diagnosis. Jones (1872, p. 146). Description. Shape triangular, ovate. Dorsal margin asymmetrically arched with highest point posterior of mid- length; both anterior and posterior margins evenly rounded although slight angle may be evident at posterior end of dorsal margin; ventral margin straight. Greatest height posterior of mid-length, greatest length at mid-height, and greatest width close to posterior end. Valves inflated. Shape of right valve differs slightly from that of left described above in that it is less high and anterior portion of dorsal margin is inclined more steeply so that anterior margin is lower and more sharply rounded. Some variation in shape within species can be seen in degree of asymmetry of shell. Left valve larger than right, with strong overlap. Valve surface smooth. Valves normally thick and heavily calcified. Muscle-scar pattern situated at, or just posterior of, mid-length; composed of 6 scars, which tend to be triangular in shape, within ring of 12 smaller scars. Inner lamella not properly known. Crenulated selvage of smaller (right) valve fits into prominent peri- pheral groove in larger valve. This contact groove is particularly well developed beneath anterior and posterior ends of dorsal margin, thus giving appearance of sockets in hinge structure, although the 2 strongly developed portions are joined by a weaker groove. Contact groove particularly weak anteriorly, just below junction of dorsal and anterior margins, and also in middle of ventral margin. Sexual dimorphism not observed. Pore canals not visible. A few instars possess weak postero-ventral spines. Remarks. The name Bairdia liassica is derived from Brodie (1845, p. 80), who mentioned 'Cypris liassica', but since this species was not described or figured it is therefore void 660 PALAEONTOLOGY, VOLUME 14 according to Anderson (1964, p. 133). Blake (1876, p. 430) considered Bairdia ellip- soidea Jones (1872) a synonym of his species B. liassica. Since 'Cypris Jiassicd’ is a nomen nudum and the species described by Blake is synonymous with B. ellipsoidea Jones, Blake’s name is incorrect and that of Jones has priority. Cypris liasica as described by Terquem (1854) from near Metz is the true species of this name and may be identical with the species regarded as B. ellipsoidea Jones in this work. It must be stressed that these conclusions are purely bibliographic; the problem of the actual similarity of B. liassica Blake and B. ellipsoidea Jones is discussed below. Although Blake considered B. ellipsoidea synonymous with B. liassica, to judge from the horizons at which Blake found B. liassica (1876, p. 430) it seems probable that he used the name for any vaguely rounded, triangular species of Ogmoconcha. The original material of B. ellipsoidea described by Jones was without doubt from the Hettangian of Yorkshire. Troedsson (1951, p. 245) regarded B. liassica as synonymous with Bairdia amalthei (Quenstedt) (= Cypris amalthei Quenstedt 1858 = Ogmoconcha amalthei (Quenstedt) of Triebel 1950), and seems to have been unaware of Triebel’s (1950) paper and his designation of a lectotype. Jones’s (1894, p. 164) description of Cytheridea ellipsoidea includes the following synonymy: Bairdia ? ellipsoidea (G. S. Brady, MS) Jones 1872, p. 146. Bairdia liassica (not Brodie’s sp.) Blake 1876, p. 430, pi. 17, figs. 1 and \a. It is debatable whether Bairdia liassica (of Blake 1876) = Bairdia (?) ellipsoidea Jones 1872, is the same as Cytheridea ellipsoidea (Jones). The published illustrations show similarly shaped valves, but inspection of the material in the British Museum (Natural History) revealed that the specimen of C. ellipsoidea (Lower Lias, Westbury-on-Severn, Brodie Collection, I 6090) is unlike those of B. liassica (Blake Collection, D, no locality given; 3 carapaces and 1 valve). The material described here is from the Hettangian of the area from which Jones described Bairdia ( ?) ellipsoidea, i.e. Cliffe, and is very closely comparable with both the original description (although Jones incorrectly orientated his specimens) and with the illustration given by Blake (1876, pi. 17, fig. 1). There is no doubt that this was the species described by Jones. A few specimens from Hotham (pit ‘g’) were slightly more rounded and show more similarity to one of Blake’s illustrations (pi. 17, fig. In) than to the other. From the synonomy it will be clear that 2 other species may also belong to this species. Healdia aspinata Drexler 1958 is at present the only species which Griindel (1964, pp. 469, 470) considered to belong in his genus Ogmoconchella, which is differentiated from Healdia and Ogmoconcha on muscle-scar patterns. The difference between the pattern of Ogmoconcha and Griindel’s (1964, p. All, fig. 7) drawing of the pattern in Ogmoconchella is in terms of the central group of muscle-scars which lie, in both cases, within a ring of small scars. In the latter genus the central scars are rounded, while in Ogmoconcha these scars may also be rounded but are frequently triangular. The central scars in O. ellipsoidea are of Ogmoconcha type, and tend to be triangular in shape. The author believes that the differences between the 2 patterns are insufficient to warrant splitting the genus. Ogmoconchella aspinata is from the Lias alpha of Germany and may well belong to O. ellipsoidea. Another species of similar form is Hungarella owthorpensis A. LORD: LIAS OSTRACODA 661 Anderson (1964) from the English Rhaetic, in which is included ‘Ostracoda B’ Aposto- lescu (1959) from the Lower Lias of the Paris Basin. Anderson gave dimensions for the holotype only (L. — 0-625 mm; H. — 0-450 mm; W. — 0-380 mm), which is larger than the present material of O. ellipsoidea. H. owthorpensis is differentiated from B. liassica in that the latter possesses a ‘faintly punctate surface’ (Anderson 1964, p. 148), but the material of O. ellipsoidea Jones (— B. liassica) from East Yorkshire is, in fact, quite smooth. H. owthorpensis, in the original figures, appears more inflated and possesses a more marked degree of overlap than O. ellipsoidea. Some of the records quoted in synonymy from the Lias Colloque (1961) have a range from Hettangian to top Pliensbachian and are clearly omnibus records without great significance. Apart from the species discussed above, O. ellipsoidea can be distinguished from other species of the genus by its shape, position of highest point on dorsal margin, degree of overlap, and degree of inflation. Ogmoconcha hagenowi Drexler 1958 Plate 123, figs. 14-16 1952 Bairdia sp. (9), Usbeck, p. 406, pi. 19, fig. 68. 1958 Ogmoconcha hagenowi Drexler, pp. 508-510, pi. 21, figs. 8a-/; pi. 26, figs. 1, 2. 71959 Ostracoda A, Apostolescu, p. 816, pi. 1, figs. 9-11. 1961 Hungarella hagenowi (Drexler); Pietrzenuk, p. 88. 1962 Ogmoconcha hagenowi Drexler; Klingler, p. 80, table 7; pi. 12, fig. 4. 1963 Hungarella hagenowi (Drexler); Oertli, pi. 7, fig. 2; pi. 8, figs. 1, 2. Material. 224 carapaces, 508 valves. Distribution. Redcar , samples R2, 3, 4, 5, 7, 8, 9, 10, 11. Dimensions (in mm). Length Height Width Carapace HU.52.J.44 0-73 0-56 0-42 Right valve HU.52.J.45 0-68 0-48 0-20 Carapace HU.52.J.46 0-54 0-38 0-29 Carapace HU.52.J.47 0-48 0-34 0-28 Carapace HU.52.J.48 0-43 0-28 0-21 Right valve HU.52.J.49 0-69 0-50 0-22 Left valve HU.52.J.50 0-73 0-55 0-23 Right valve HU.52.J.51 0-67 0-48 0-26 Left valve HU.52.J.52 0-70 0-50 0-25 Diagnosis. Drexler (1958, p. 509). Description. Shape rounded, triangular. Dorsal margin strongly arched with highest point at, or anterior to, mid-length; anterior margin evenly rounded, merging without modification into dorsal margin and into convex ventral margin; posterior margin rounded but may show slight asymmetry. Left valve distinctly larger than right, overlap fairly strong, but weak posteriorly and anteriorly. Position of greatest length at, or just below, mid-height, position of greatest height antero-medianly, and position of greatest width close to posterior end. Valve surface smooth, unornamented. In left (larger) valve there is a strong, crenulated contact furrow which received selvage of right valve. This contact groove strongest beneath crest of dorsal margin. Above contact groove on dorsal margin of left valve another x x C 8385 662 PALAEONTOLOGY, VOLUME 14 slight furrow is present and may represent an accommodation groove or ligament pit. Details of muscle-scars and pore canals not observed. No sexual dimorphism observed. Remarks. Despite the abundant material, internal features were largely unobserved because the valves were invariably filled with sediment, and efforts to clean the interiors of valves by ultrasonic or other physical means failed. On the margins of the specimens examined pore canals were not observed. No marginal denticles as figured by Drexler (1958, pi. 21, fig. 8Z?) were seen. This species is easy to distinguish from other species of the genus by means of its strongly triangular and convex shape, and by the position of the highest point on the dorsal margin. The species has also been recorded from France and Germany: Drexler (1958) Lias alpha 1, 2, 3. Usbeck (1952) Lias alpha 2. This work Lias alpha 2, 2>a. Oertli (1963) Lower Sinemurian (Lias alpha 3). Klingler (1962) Lias alpha 2, 3; Lias beta 1, 2. Pietrzenuk (1961) Lias beta. Ostracoda A Apostolescu (1959) bucklandi to obtusum zones = Lias alpha 3o, 6; and Lias beta \a, b. Ogmoconcha sp. A Plate 123, figs. 17, 18 Material. 35 carapaces, 34 valves. Distribution. Redcar, samples R6, 7, 1 1 . Dimensions (in mm). Length Height Width Carapace HU.52.J.53 0-61 0-41 0-27 Carapace HU.52.J.54 0-60 0-40 0-31 Left valve HU.52.J.55 0-60 0-41 0-17 Right valve HU.52.J.56 0-54 0-32 012 Description. Shape oval, almost symmetrical about mid- length; dorsal margin convex, highest point being at mid-length, posterior and anterior margins almost identically and symmetrically curved, ventral margin gently convex. Greatest length at mid-height, greatest height at mid-length, and greatest width at posterior end. Left valve larger than right and overlaps it all round, but not strongly. Selvage of smaller valve fits contact groove in left valve; contact groove most strongly developed beneath dorsal margin. Valve surface smooth, unornamented. Internal features not observed because of sedi- ment infilling valves. No sexual dimorphism apparent. Remarks. This species is distinguished from the 2 species described above as follows : O. ellipsoidea O. hagenowi O. sp. A Shape rounded, asymmetrical triangular rounded, symmetrical Position of highest point on dorsal margin posterior of mid-length anterior of mid-length at mid-length Degree of overlap strong fairly strong, weak at anterior and posterior weak, especially at anterior and posterior A. LORD: LIAS OSTRACODA 663 Since some of the material is poorly preserved the species is left under open nomen- clature. An instar, apparently belonging to this species, had a small postero-ventral, back- wardly directed spine base and also a tiny spine on the mid-part of the anterior margin of the right valve. In shape, spinosity, and size this specimen resembles Himgare/la caudata Anderson (1964) from the Rhaetic. Anderson’s right valve holotype (GSM Mik (J) 286001) was 0-450 mm long and 0-275 mm high, whilst the present carapace (instar HU.52.J.57) measures 0-470 mm by 0-290 mm. H. caudata is similar to Ostracod Nr. 5 of Klingler (1962, p. 83, table 7, pi. 12, fig. 8) but the latter ranges from Lias beta to upper delta and has the following dimensions (calculated from the original figure); length 0-583 mm, height 0-350 mm. One cannot state definitely that the spinose instar belongs to Ogmoconcha sp. A, but spinose instars of smooth adults are not an unknown phenomenon. Oertli (1957), for example, in his work on Upper Jurassic ostracods from the Paris Basin sondage Vernon 1, described and illustrated (pp. 676, 677, pi. 7, figs. 229-233) Indet. gen. (Progonocytherinarwnl) sp. A, a smooth form as an adult but with spinose instars. If the author’s interpretation is correct, this suggests that Ogmoconcha sp. A is, in fact, Ogmoconcha caudata (Anderson) and that that species was based on an immature specimen. ' — — LOCALITIES SPECIES REDCAR R5 R6 R7 SOUTH CLIFFE NORUH_^IFFE a ex b HOTHAM R1 R2 R3 R4 R8 R9 RIO R11 Polycope cerasia BLAKE 1876 Cythereila drexlerae FIELD 1967 C concentrica FIELD 1966 Cythereiloidea circumscripta (BLAKE 1876) C, pulchella APOSTOLESCU 1959 Bairdia aff. B. molesta APOSTOLESCU 1959 B. cf B. carinata DREXLER 1958 B tatei CORYELL 1963 Bairdiacypris? sartriensis DONZE 1966 Paracypris cf. P.? semidisca DREXLER 1958 Klinglerella moorei (JONES 1872) K, aff. K. triebeli (K, & N. 1959) K ? translucens (BLAKE 1876) Cythere' terquemiana JONES 1872 Ogmochoncha ellipsoidea (JONES 1872) B 0 hagenowi DREXLER 1958 0, sp. A K. & N. = Klingler & Neuweiler TEXT-FIG. 4. Distribution of Lower Lias Ostracoda in Yorkshire. Other Species of Jones (1872) and Blake (1876) Besides the species described above reference has been made by later workers to other Yorkshire species: Bairdia redcarensis Blake (1876) possibly belongs to the genus Paracypris according to Apostolescu (1959, p. 806). Cythere arcaefonnis Blake (1876) was called Bairdia arcaeformis by Pratje (1923, p. 253). As in the case of Issler’s (1908) 664 PALAEONTOLOGY, VOLUME 14 placing of C. moorei and C. translucens in Bairdia, this assignment is probably errone- ous; C. moorei and C. translucens do not belong in Bairdia and it is unlikely that this species does either. Cythere blakei Jones (1872) was placed by Coryell (1963, pp. 621, 877) in the synonomy of Darwinula liassica, which would seem to be incorrect. Of the 16 Yorkshire Lias ostracods described by Jones and by Blake (allowing for the synonomy of Bairdia ellipsoidea with B. liassica) only 3 species, Cythere triangulata Blake, C. redcarensis Blake, and Cytherella ? paupercida Blake have gone unremarked since their original description, and the present work threw no light on these 3 species. Acknowledgements. I am indebted to Professor D. T. Donovan and Mr. T. Getty for stratigraphic advice concerning the Redcar section, and to Dr. J. W. Neale, who read the manuscript and greatly aided the author during the preparatory work, which was carried out during the tenure of a University of Hull Research Studentship. Mrs. J. Irvine and Mr. P. Scott kindly prepared the photographs using the scanning electron microscope of the School of Environmental Sciences, University of East Anglia. REFERENCES ANDERSON, F. w. 1964. Rhaetic Ostracoda. Bull. geol. Siirv. G.B. 21, 133-174. APOSTOLESCU, v. 1959. Ostracodes du Lias du Bassin de Paris. Rev. Inst.fr. Petrole, 14 (6), 795-826. in COUSIN, N., ESPITALIE, J., siGAL, J. and APOSTOLESCU, v. 1961. Sud de bassin, region d’Argenton- sur-Creuse et de la Charte (Departements du Cher et de I’lndre), pp. 445^49, Colloque sur le Lias Frangais. Mem. Bur. Recherches Geol. Min. no. 4. BARBiERi, F. 1964. Micropaleontologia del Lias e Dogger del Pozzo Ragusa 1 (Sicilia). Riv. ital. Paleont. 70 (4), 709-831. BizoN, G. 1961. Lorraine, region de Nancy et Thionville, pp. 433-436, Colloque sur le Lias Frangais. Mem. Bur. Recherches Geol. Min. no. 4. and OERTLi, H. J. 1961. Conclusions, pp. 107-119, ibid. BIZON, j.-j. in RiouLT, M. and bizon, j.-j. 1961. Basse-Normandie (regions d’Isigny et sud de Bayeux), pp. 451-458, ibid. BLAKE, J. F. 1872. On the Infralias in Yorkshire. Q. Jl geol. Soc. Lond. 28, 132-146. in TATE, R. and blake, j. f. 1876. The Yorkshire Lias. 475 + xii+xii pp. London. BOLD, w. A. VAN DEN. 1966. Homonyms of Bairdia elongata and Bythocypris elongata. J. Paleont. 40 (1), 222-223. BRODiE, p. B. 1845. A History of the Fossil Insects. 130 pp. London. CHAMPEAU, H. 1961. Etude de la microfaune des niveaux marneux du Lias dans le sud-est du bassin de Paris, pp. 437-443, Colloque sur le Lias Frangais. Mem. Bur. Recherches Geol. Min. no. 4. CONTI, s. 1954. Stratigrafia e paleontologia della Val Solda (Lago di Lugano). Mem. descr. Carta geol. Ital. 30, 1-248. coNTiNi, D. and pariwatvorn, p. 1964. Etude comparee de la microfaune du Lias superieur d’Asel- fingen (Jura du Randen) et de Pouilley-les-Vignes (Jura Franc-Comtois). Annlsscient. Univ. Besancon, 2 ser., Geol. 18, 35-44. CORYELL, H. N. 1963. Bibliographic index and classification of the Mesozoic Ostracoda, 1175 pp., 2 vols., Univ. Dayton Press. COUSIN, N. and apostolescu, v. in cousin, n., espitalie, j., sigal, j. and apostolescu, v. 1961. Ardennes, region de Mezieres (Departement des Ardennes), pp. 423^31, Colloque sur le Lias Francais. Mem. Bur. Recherches Geol. Min. no. 4. DE BOER, G., NEALE, J. w. and PENNY, L. F. 1958. A guide to the geology of the area between Market Weighton and the Humber. Proc. Yorks, geol. Soc. 31 (2), 157-209. DONZE, p. 1966. Ostracodes de I’Hettangian entre Aubenas et Privas (Ardeche). Trav. Lab. Geol. Univ. Lyon. N.s. 13, 121-139. DREXLER, E. 1958. Foraminiferen und Ostracoden aus dem Lias alpha von Siebeldingen/Pfalz. Geol. Jb. 75, 475-554. A. LORD; LIAS OSTRACODA 665 FIELD, R. A. 1966. Species of the family Cytherellidae (Ostracoda) from the Lower Lias of South Dorset, England. Senckeiiberg. leth. 47 (1), 87-105. 1967. Cytherella drexlerae nom. nov., a replacement name for Cytherello inflata (Drexler, 1958). Ibid. 48 (6), 534. FISCHER, w. 1961. Neue Arten der Ostracoden-Gattung Polycope Sars 1865 aus den oberen Lias (Wurttemberg). Neues Jb. Geol. Paldont. Mb. 10, 497-501. FOX-STRANGWAYS, c. 1 892. The Jiirassic Rocks of Britain, 1. Yorkshire. Mem. geol. Surv. U.K. 551 -fix pp. GRUNDEL, j. 1964. Zur Gattung Healdia (Ostracoda) und zu einigen verwandten Formen aus dem unteren Jura. Geologie, 13 (4), 456-477. ISSUER, A. 1908. Beitrage zur Stratigraphic und Mikrofauna des Lias in Schwaben. Palaeontographica, 55, 1-105. JONES, T. R. 1872. On some Bivalve Entomostraca from the Lias (Tnfralias’) of Yorkshire. Q. Jl geol. Soc. bond. 28, 146-147. Appendix in Blake (1872). 1894. On the Rhaetic and some Liassic Ostracoda of Britain. Ibid. 50, 156-169. KLiNGLER, w. in SIMON, w. and BARTENSTEiN, H. (cds.) 1962. LeitfosslUen der Mikropaldontologie. 1, 432+viii pp. ; 2, 59 pi., 22 tab. Berlin. and NEUWEiLER, F. 1959. Leitende Ostracoden aus dem deutschen Lias beta. Geol. Jb. 76, 373-410. MAGNE, j. and OBERT, D. 1966. Niveaux a microfaunes dans le Lias de la region d’Arbois (Jura). Rev. Micropaleont. 8 (4), 265-21 h. MEHES, G. 1911. Uber Trias-Ostrakoden aus dem Bakony in Resultate der Wissenschaftlichen Erfor- schung des Balatonsees. Anhang palaeo. der Umgebnng des Balatonsees, 6, 1-38. MOORE, R. c. (ed.) 1961. Treatise on Invertebrate Paleontology. Q., Arthropoda 3, Crustacea, Ostracoda. 442+xxiii pp. Lawrence, Kansas. MORKHOVEN, F. p. c. M. VAN. 1962-63. Post-Palaeozoic Ostracoda. 1, General, 204 pp. (1962). 2, Generic descriptions, 478 pp. (1963). Amsterdam. OERTLi, H. J. 1957. Ostracodes du Jurassique superieur du Bassin de Paris (sondage Vernon 1). Rev. Inst. fr. Petrole, 12 (6), 647-695. 1963. Faunes d'Ostracodes du Mesozoicpie de France! Mesozoic ostracod faunas of France. 57 pp., 90 pi. Leiden. PIETRZENUK, E. 1961. Zur Mikrofauna einiger Liasvorkommen in der Deutschen Domokratischen Republik. Freiberger ForschHft. 113, 1-129. PRATJE, o. 1923. Der Fossilinhalt der Rhat-und Liasreste am westkucgeb Schwenzwaldrande. Geol. Arch. 1, 196-258. SOHN, I. G. 1968. Triassic ostracodes from Makhtesh Ramon, Israel. Bull. geol. Surv. Israel. 44, 1-71. TATE, R. and BLAKE, J. F. 1876. The Yorkshire Lias. 475 + xii + xii pp. London. TERQUEM, o. 1854-5. Palcontologic de I’Etage Inferieur de la Formation liasique de la Province Luxemburg. Mem. Soc. geol. Fr., Ser. 3, 1 (3), 81-125. TRiEBEL, E. 1950. Die taxonomische Stellung der Ostracoden-Gattung Ogmoconcha und der Lectotypus von. O. amalthei. Senckenbergiana, 31 (1/2), 113-120. TROEDSSON, G. 1951. On the Hoganas Series of Sweden (Rhaeto-Lias). Acta Univ. Lund. A1 (1), 1-268. USBECK, I. 1952. Zur Kenntnis von Mikrofauna und Stratigraphic im unteren Lias alpha Schwabens. Neues Jb. Geol. Paldont. Abh. 95 (3), 371-476. wiCHER, c. A. 1938. Mikrofaunen aus Jura und Kreide insbesondere Nordwestdeutschlands I. Teil: Lias alpha bis epsilon. Abh. preuss. geol. Landesanst. n.f., 193, 1-16. A. R. LORD School of Environmental Sciences University of East Anglia University Village Norwich, NOR 88C Revised typescript received 1 October 1970 MARSUPIATE TERTIARY ECHINOIDS FROM SOUTH-EASTERN AUSTRALIA AND THEIR ZOOGEOGRAPHIC SIGNIFICANCE by G. M. PHILIP and R. J. foster Abstract. A review of brooding and sexual dimorphism in living and fossil sea urchins reveals the wide variety of marsupia that can be developed. In living sea urchins, brooding, in some species with the concomitant development of marsupia, is an outcome of direct development. In fossil echinoids the presence of marsupia probably indicates a similar omission of free larval stages. Marsupia have been reasonably substantiated in five Cretaceous species, eleven Tertiary species (ten from Australia) and twenty-eight living species. Of these latter, twenty-five are from polar waters, which accords well with the interpretation that direct development is an adaptation to cold or deep waters. In Australia the abundance of marsupiate echinoids decreases through the Cainozoic and there is now but one living species. It is suggested that the occurrence of marsupiate species in the Australian Tertiary supports recent suggestions of the late separation of Australia from Antarctica (in the Late Cretaceous or Eocene) and that these marsupiate echinoids had their origin in Antarctic waters. The following marsupiate species, of Upper Eocene, Oligocene, or Miocene age, are described and figured: the temnopleurids Pamdoxechinus iiovus Laube, P. granulosus sp. nov., P. profundus sp. nov., P. stellatus sp. nov., Pentechinus mirabilis gen. et sp. nov., the clypeasteroids Fossulaster balli Lambert and Thiery, F. exiguus sp. nov., Willungaster scutellaris gen. et sp. nov., the spatangids Peraspatangus brevis gen. et sp. nov., P. depressus gen. et sp. nov. The family Fossulasteridae is erected for Australian Tertiary clypeasterinid echinoids with simple ambulacral plates in the petals. Although a revision of the prolific Tertiary echinoids of south-eastern Australia is being published elsewhere (Philip 1963dr, 1964, 19656, 1969), it is useful to discuss separately aspects of the fauna of more general interest. This paper is concerned with the marsupiate urchins. Of the ten such species described here, only two have been named previously. As only one marsupiate species of Tertiary age is known from else- where, opportunity is taken to review the occurrence and structure of marsupia in echinoids, and to speculate on their significance. MARSUPIA IN ECHINOIDS In living echinoids the sexes are separate although hermaphroditic individuals of normally dioecious species occur as rare anomalies (Hyman 1955, p. 478, q.v.). Generally the tests of individuals of different sexes are indistinguishable. Some claims have been made that different test shapes may obtain in the different sexes (e.g. Ikeda 1931, who identified as females the low tested forms of the highly variable temnopleurid species Temnopleurus toreumaticus Leske), but such interpretations have been rejected by subse- quent workers (cf. Mortensen 1943, p. 79). Some echinoids, however, are known to brood their young and with such forms sexual dimorphism may obtain. Brooding urchins, in contrast to others, are thought to develop directly with the omission of the echinopluteus (or free swimming larval stage). The few large yolky eggs of the female develop around her peristome or apical system or even in specially produced brood pouches or marsupia. [Palaeontology, Vol. 14, Part 4, 1971, pp. 666-695, pis. 124-134.] PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 667 Brooding is known best in cidaroids and spatangoids from Antarctic waters, where, apparently, a viviparous habit accords some benefit to the urchins. With the sole exception of Austrocidaris canaliculata (Agassiz), which bears its young around the apical system, brooding cidaroids carry their offspring around the peristome, protected by primary radioles. In only two cidaroid species {Rhynchocidaris triplopora (Morten- sen) in which an annular depression is present and Ctenocidaris geliberti (Koehler) in which the edge of the peristome is sunken) is the corona of females modified because of this brooding habit. In the former species Mortensen (1909) has reported the relocation of the genital pores at the edge of the peristome in females infested by the parasite Echinopbyces. In all species of the Antarctic spantangoid genera Abatus, Arnphipneiistes and Tripyhts, the petalloid ambulacra of the females are deeply sunken to accommodate the young (vide Mortensen 1951). Apart from cidaroids, undoubted brooding in living regular sea urchins appears to be confined to Hypsiechinus coronatus Mortensen. In this Arctic temnopleurid the females possess an elevated apical system in which the plates are strongly inflected to form a prominent knob. The young are carried around this knob on the adapical surface of the test. In all other living species in which a marsupium is present in the females, it consists of an apical depression containing the apical system (the neolampadids Anochanus sinensis (Grove) and Tropholampas loveni (H. L. Clark); the clypeasteroid Fibidaria nutriens H. L. Clark and the meridosternous spatangoid Plexechinus nordenskjoldi Mortensen). Excluding the fossil species described here, marsupia have been suggested previously in a number of fossil species, but many claims lack reasonable substantiation (Kier 1969, gives seven species, including one redescribed here). Kier (1967a) has described an undoubted marsupium in the middle Eocene clypeaster- oid Pentediwn curator Kier in which the females possess a deep adapical depression which includes the apical system and the petalloid ambulacra. Again, Lambert (1933, p. 27, pi. 4, figs. 6-7) has described and illustrated young in the deeply sunken petals of his Cretaceous species Tripylus pseudoviviparus (referred questionably to the genus Abatus by Mortensen 1951, p. 276). Rosenkrantz (in Mortensen 1951, p. 577) reported that in some specimens of Brissop- neustes danicus Schliiter the apical system is deeply sunken. He suggested that this depression is a marsupium, and Mortensen (loc. cit.) was in agreement. Among fossil regular echinoids, the reports are by no means as convincing. The Cretaceous species Goniopygus royoi Lambert and G. minor Sorignet have deep inter- ambulacral depressions originating at the edge of the genital plates. These depressions have been viewed as marsupia by Lambert (1928, p. 156). Similar depressions occurring in some specimens of Thylechinus said (Peron and Gauthier 1881, pi. 19, figs. 4, 6) have been given the same interpretation (cf Mortensen 1935, p. 468). Gregory (1892) regarded the adapical sutural depressions of the British Pliocene species Tenmechinus excavatus Lorbes as ‘marsupial pouches’ and gave a similar explanation to some obscure adapical interambulacral depressions in ‘‘Echinus' henslovi Lorbes. Examination of British Pliocene sea urchins in the collections of both the British Museum (Nat. Hist.) and the Sedgwick Museum, Cambridge, revealed that no 668 PALAEONTOLOGY, VOLUME 14 dimorphism exists in these last two species. Accordingly, Gregory’s interpretation is discounted. On the other hand, in Goniopygus royoi, G. minor, and Thylechinus said individuals have been reported which lack depressions, so that Kier (1969) concluded that sexual dimorphism existed in these Cretaceous stirodonts. Of the marsupiate species discussed here, two previously have been described and so interpreted. The adapical depression of Paradoxechinus novas Laube was seen as a marsupium by Bittner (1892), Mortensen (1943), and Philip (1969) who described speci- mens lacking the depression and which, accordingly, were identified as males. Hall (1908), Durham (1955), and Kier (1969) have interpreted the deep adoral bilobed depression anterior of the peristome of Fossulaster halli as a marsupium, although Mortensen (1951, p. 577) retained doubts, for specimens of the males were unavailable to him for study. Much new information concerning both of these species is provided here. The other species described not only increase two-fold the number of known marsu- piate fossil sea-urchins, but also have marsupia the architecture of which is not known elsewhere in the class. The remarkable interambulacral pouches of Pentechinus mirabilis gen. et sp. nov., are particularly noteworthy. Two new clypeasteroid species are described in which the marsupia are depressions on the oral surface anterior of the peristome. Echinoids with a marsupium in this position are known only from Australian Tertiary strata. The marsupia of the species of Paradoxechinus find a close analogue in Hypsiechinus coronatus, and the apical depression of Peraspatangus is similar to those of other irregular echinoids. From the foregoing discussion it can be seen that marsupia are widespread throughout different echinoid groups and that marsupia different in location and structure may be developed within the one family. This may be seen best in the following table that summarizes the different types of marsupia presently known. MARSUPIUM ADAPICAL A. CENTRAL 1 . Apical system raised a. without annular depression CAMARODONTA * Hypsiechinus coronatus Mortensen b. with annular depression CAMARODONTA Paradoxechinus granulosus sp. nov. P. novus Laube P. profundus sp. nov. P. stellatus sp. nov. EXPLANATION OF PLATE 124 Figs. 1-3. Stereopairs of the holotype (P27942) of Pentechinus mirabilis gen. et sp. nov. 1, Adapical view, X 5. 2, Adoral view of the interior of the adapical half of test, showing the five flattened sac- like marsupia, and the genital pores at the extremities of the introverted genital plates, X 5. 3, Oblique adapical view showing the interambulacral marsupial slits, x 8. Palaeontology, Vol. 14 PLATE 124 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 669 2. Apical system sunken CLYPEASTEROIDA *Fibularia nutriens H. L. Clark Pentedium curator Kier NEOLAMPADINA *Auochamis sinensis (Grove) *Tropholampas loveni (H. L. Clark) SPATANGOIDA *Plexechinus nordenskjoldi Mortensen Peraspatangus brevis gen. et sp. nov. P. depressiis gen. et sp. nov. Brissopneustes danicus Schliiter B. LATERAL 1. Ambulacral SPATANGOIDA *Abatiis 9 spp. *Ampliipneiistes 8 spp. *Tripylus 4 spp. (vide Mortensen 1951) ''Tripylus' pseudoviviparus Lambert 2. Interambulacral CAMARODONTA Pentechinus mirabilis gen. et sp. nov. STIRODONTA Goniopygus royoi Lambert G. minor Sorignet Thylechinns said (Peron and Gauthier) MARSUPIUM ADORAL A. CIRCUM ORAL CIDAROIDA *Ctenocidaris geliberti (Koehler) * Rhynchocidaris triplopora (Mortensen) B. ANTERIOR CLYPEASTEROIDA Fossidaster halli Lambert and Thiery F. exiguus sp. nov. Willungaster scutellaris gen. et sp. nov. * living species To date marsupia have been reasonably substantiated in five Cretaceous species, eleven Tertiary species (ten from Australia), and twenty-eight living species. It is customary to associate direct development and brooding in sea-urchins with cold water or deep water forms (Hyman 1955). Thorson (1950) has pointed out that many groups of marine benthonic organisms that normally have pelagic larval stages tend to lack these in polar seas. He suggested that this is an adaptation to accommodate to slow 670 PALAEONTOLOGY, VOLUME 14 larval development at low temperatures and to paucity of phytoplankton for much of the year. In such a situation the greater the size of the individual when it achieves independent existence, then the smaller its relative food requirements and the better its chance of survival (cf. Kier 1969, p. 216). Of the twenty-eight known living marsupiate species, twenty-three are from Antarctic waters and two inhabit the Arctic Ocean. So striking is the present-day association of marsupiate urchins with cold water conditions, that Fischer (1963, pp. 289-290) took the occurrence of marsupia in the two late Cretaceous spatangoids mentioned above as evidence indicative of cooling during that epoch. On first consideration, the abundance and diversity of Australian Tertiary marsupiate echinoids would appear to speak against such substantive uniformitarianism. They occur in Upper Eocene to Miocene sediments, which have been thought (at least for the Miocene) to have been deposited in sub-tropical seas. This conclusion was derived from consideration of the diverse and prolific fauna, with its tropical Indo-Pacific elements (David 1950, p. 587) and also oxygen isotope studies (Dorman and Gill 1959). Yet it now seems probable that the marsupiate echinoids from the Australian Tertiary had their ultimate origin in Antarctic waters. According to present interpretations of magnetic anomalies in the southern Indian Ocean, and reconstructions based on plate tectonics, Australia was not detached from Antarctica until the late Cretaceous (Dietz and Holden 1970). It was not until the late Eocene that the Australian plate began its rapid flight northwards (Le Pichon and Heirtzler 1968) away from the stable Antarctic plate, opening up what are now Antarctic waters. That a brooding habit had been established in southern seas by Cretaceous times is shown in the occurrence in Madagascar of ‘‘Tripyhis' pseudovivipanis Lambert (1933), a specimen of which even shows the young still preserved in the sunken petals. If deductions of workers on plate tectonics are accepted, then the oldest of the marsupiate urchins described here in fact inhabited epicontinental seas approximately 20° to the south of where they were col- lected. Within the Australian Cainozoic record from the Oligocene there is a progressive decrease in the abundance and variety of marsupiate echinoids, until at present only one such living species, Fihularia nutriens, inhabits eastern Australian coastal waters. This fact is given new meaning by suggesting that the adaptive advantage of the marsu- piate echinoids diminished as their habitat moved northward into more temperate climes. Another aspect of the south-eastern Australian Tertiary echinoid fauna supports current views on continental displacement. Present-day Antarctic and sub-Antarctic seas are inhabited by a specialized group of cidaroids, the ctenocidarines, to the apparent exclusion of all other cidaroids. From the Australian record it has been established that this group evolved from the stereocidarines in the Paleogene (Philip 1964, pp. 452-453). EXPLANATION OF PLATE 125 Figs. 1-2. Paradoxechiniis novus Laube. 1, Oblique adapical stereopair of apical system of P27943 ($), X 8. 2, Stereopair of adapical view of same specimen, X 5. Fig. 3. Fossulaster halli Lambert and Thiery. Stereopair of P27948 ($), a specimen prepared to show the internal supports of the test. Note portion of adapical surface of test ankylosed to one of the lobes of the marsupium, X 8. Palaeontology, Vol. 14 PLATE 125 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 67( The group is represented by a typical species of Austrocidaris in the Lower Miocene (op. cit., p. 462), but, thereafter, ctenocidarines disappear from the Australian Cainozoic record. It may be noted that sexual dimorphism of the apical system is not uncommon in living echinoids. The genital pores of females may be larger, or located differently, or both. The difference in size is particularly conspicuous in urchins thought to have direct development, as the larger pores of the females are necessary for the passage of the larger eggs. However, dimorphism is known in many species which apparently possess echinopluteus larvae. For example, in species of the temnopleurid Prionechinus, the genital pores of females are large and rounded, and mounted toward the centre of the genital plates; those of the males are small and slit-like and mounted distally (Morten- sen 1943, p. 294). In the echinid Psammechmus milians the genital pores of males are mounted on short papillae; those of females are not, and in the clypeasteroid Echino- cyamus pusiHus the papillae of males are longer than those of females. Dimorphism of the apical system attributed to sexual differences has been reported in a number of fossil groups (e.g. neolampadids, Philip 19636; oligopygids, Kier 19676; clypeasteroids, Kier 1967a, 1968) and is discussed in detail by Kier (1969). All specimens with prefix ‘P’ housed in the fossil collection of the National Museum of Victoria, Melbourne; those with prefix 'UNE’ in the collections of the Department of Geology, University of New England, Armidale, N.S.W. SYSTEMATICS Subclass REGULARIA Order camarodonta Jackson Suborder temnopleurina Mortensen Family temnopleuridae A. Agassiz Remarks. Philip (1969) divided representatives of this family into two groups; those with sculpture and those with pitted sutures. All Australian Tertiary temnopleurids belong to the former group, including the new genus Pentechinus. All known marsupiate temno- pleurids are sculptured forms. Genus paradoxechinus Laube 1869 Paradoxechinus Laube, p. 186. 1969 Paradoxechinus Laube; Philip, p. 262 (with full synonymy). Type species. Paradoxechinus novus Laube 1869, by monotypy. Diagnosis. Small to moderate sized sculptured temnopleurids with crenulate tubercles and shallow gill slits. Pore-pairs uniserial or very weakly arcuate within each triad. Sculpture varying from prominent ridges bearing secondary tubercles and granules, to multiple ridges running between primary tubercles. Perignathic girdle of relatively large, spatulate auricles just united above the ambulacra and connected by low apophy- ses. Test dimorphic, with inferred females possessing a wide, deep adapical depression. Apical system caduceous (known only in one female), dicyclic, with extremely elongate oculars, proximally with all plates strongly inflected upwards to form a prominent knob containing the periproct. 672 PALAEONTOLOGY, VOLUME 14 Remarks. The relationship of this genus to similar temnopleurids has been given else- where (Philip 1969). At that time topotype material of the type species alone was available for study and the character of the apical system was unknown. In new material from the St. Vincent Basin, S.A., the apical system is preserved in one specimen. Also, material from elsewhere now permits recognition of three other species. The above generic diagnosis accommodates this new information. Bittner (1892) and Mortensen (1943) have previously interpreted the adapical depres- sion of P. novas as a marsupium and this interpretation was enhanced by the recognition of specimens from the type horizon lacking such depressions (Philip 1969). The apical system and marsupium are discussed in detail below. Paradoxechinus novas Laube Plate 125, figs. 1,2; Plate 126, fig. 1 ; Plate 127, fig. 1 ; Plate 129, figs. 7, 12, 13, 17; text-figs. 1, 2. 1869 Paradoxechinus novas Laube, p. 188, fig. 2. 1875 Paradoxechinus novas Laube; Etheridge, p. 449. 1877 Paradoxechinus novus Laube; Duncan, p. 65. 1878 Paradoxechinus novus Laube; Duncan, p. 415. 1891 Paradoxechinus novus Laube; Tate, p. 274 (in part). 1892 Paradoxechinus novus Laube; Bittner, pp. 344-345, pi. 4, fig. 5. 1892 Paradoxechinus novus Laube; Tate, pp. 192-193 (in part). 1910 Paradoxechinus novus Laube; Lambert and Thiery, p. 230. 1943 Paradoxechinus novus Laube; Mortensen, pp. 350-351, fig. 210^. 1946 Paradoxechinus novus Laube; H. L. Clark, p. 308 (in part). 1949 Brochopleurus australiae Fell, pp. 18-19 (in part, paratype P4688 only). 1966 Paradoxechinus novus Laube; Fell and Pawson, fig. 3176 (not fig. 317« = Ortholophus bittneri Philip). 1969 Paradoxechinus novus Laube; Philip, pp. 263-266, pi. 11, figs. 1-12; pi. 12, fig. 1 ; text- fig. la-f. Diagnosis. A small, depressed species of Paradoxechinus with prominent sculptural ridges bearing secondary tubercles and granules, and zig-zagging between the primary tubercles. Pore-pairs embedded in the raised ‘epistroma’ neighbouring the primary ambulacral tubercles. Material. In addition to material described previously from the type horizon (Mannum Formation, South Australia), 54 specimens (26$; 2SS) from the Adelaide Cement Holdings quarry, 2-5 miles north of Wool Bay, Yorke Peninsula, St. Vincent Basin, S.A. (See McGowran et al. 1971 for discussion of Australian Tertiary correlations.) Port Vincent Limestone, Guemhelitria stavensis zone (Stuart 1970, p. 174) of the Oligocene and Zone p. 21 of Blow (1970). EXPLANATION OF PLATE 126 Fig. 1. Paradoxechinus novus Laube. Oblique adapical stereopair of adapical depression of a small specimen, UNE 11522 ($), x8 approx. Fig. 2. Paradoxechinus stellatus sp. nov. Stereopair of adapical view of holotype P27944 ($), showing pentalobate adapical marsupium, x 5. Fig. 3. Peraspatangus brevis gen. et sp. nov. Stereopair of adapical view of holotype, P27956 ($), showing deep adapical marsupium, approx, x 5. Palaeontology, Vol. 14 PLATE 126 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 673 Remarks. In a female specimen from St. Vincent Basin the apical system (PI. 125, figs. 1, 2; text-fig. \a) is dicyclic, with very elongate plates, particularly the oculars. The distal margins of the plates interlock with those of the corona at the deepest part of the adapical depression in a manner which has previously been inferred (cf. Philip 1969, p. 264, text-fig. la). However, proximally all of the plates are strongly inflected upwards to give a knob which contains the small periproct. This knob rises well above the level of the outer margin of the adapical depression so that in side view of the test it is clearly visible (PI. 125, fig. 1). The madreporite cannot be recognized, but this presumably is because the surface of the depression has been overgrown with secondary calcite, obscuring surface detail. c TEXT-FIG. 1. Paiadoxechiniis novus l^aube. o, Apical system of P27943 ($); dashed line marks margin of apical depression, x 8 approx, b. Analysis of ambulacral plating of P18223 (d'X X 10. c, Oblique view of plating structure within apical depression of P18243 (?), XIO. The structure and profile of the adapical depression is similar in all species of Para- doxechinus except P. steUatiis sp. nov. (q.v.). In P. novas it begins developing at a size of about 8 mm horizontal diameter and in large specimens may be over two-thirds of the horizontal diameter (text-fig. 2d). The depression is deep (up to half of the vertical diameter) and sharply bounded by the sudden inflexion of the coronal plates at its borders. The portion of the corona within the depression is slightly concave. The sculpture and primary tubercles cease abruptly at the margin of the depression and are replaced by granules which become smaller and more distant toward the apical system. In each zone two large interambulacral plates as well as a series of high ambu- lacral plates are contained in the depression (text-fig. Ic). At the rim of the depression 674 PALAEONTOLOGY, VOLUME 14 these simple ambulacral plates pass abruptly into normal echinoid compound plates. At the adradial apex of each of the enlarged interambulacral plate is a small pit which does not pierce the test. An earlier comparison of the females of P. novus and Hypsiechinus coronatus Morten- sen (Philip 1969) is now seen to be particularly apt. The females of the latter also possess a prominently elevated apical system in which the plates are strongly inflected to form a distinctive knob. The earlier comparison was based on the character of the coronal plates. In both species the upper interambulacral plates are remarkably high, lack primary tubercles, and are covered with small granules. In addition, the upper ambu- lacral plates (those within the marsupium of Paradoxechinus) are simple primaries. It has previously been suggested that the high interambulacral plates of the females of P. novus might afford a barrier to prevent the introduction of new interambulacral plates after the marsupium was initiated. This is born out in text-fig. la. The females show from 19 to 21 interambulacral plates in each zone independent of size. In contrast the number of plates increases with size to 24 in the males. Other features of growth shown in text-fig. 2 are: 1. Peristome width does not vary significantly between the sexes (text-fig. lb). 1. The males possess a significantly higher test than the females (although no doubt this difference is accentuated by the absence of apical systems in the measured specimens). Of the 49 specimens it was possible to measure for the purposes of text-fig. 2, the average horizontal diameter of the 26 males was 12-4 mm and that of the 23 females was 10-9 mm. An analysis of the echinoid ambulacral compounding in P. novus previously given by Philip (1969, text-fig. 7c), as published, is inverted. A new figure is given here (text-fig. \b). Paradoxechinus granulosus sp. nov. Plate 129, figs. 4, 5, 9, 10, 18, 19; Plate 134, fig. 3 Diagnosis. A small species of Paradoxechinus with a depressed test and somewhat incised uniserial ambulacra. Surface of test covered with crowded secondary tubercles and granules so that, in large specimens, the zig-zagging sculptural ridges may be obscured. Material. Holotype P27947 (9) and paratypes UNE 11757 (9), UNE 11761 (c?) and two other specimens, Airey’s Inlet, Anglesea District, Victoria, from the Point Addis Limestone of Junjukian (Upper Oligocene) age and probably Zone p. 22 of Blow. Seven poorly preserved specimens (59; 2S) from Wool Bay, Yorke Peninsula, S.A., from the Giiembelitria stavensis zone of the Oligocene in the Port Vincent Limestone, are also referred to this species. EXPLANATION OF PLATE 127 Fig. 1. Paradoxechinus novus Laube. Stereopair of adapical view of UNE 11523 (9), x5. Fig. 2. Willungaster scutellaris gen. et sp. nov. Stereopair of adoral view of holotype, P27951 (9), showing marsupial depression to the anterior of the peristome, X 5. Fig. 3. Fossulaster halli Lambert and Thiery. Stereopair of adoral surface of P19178 (9), showing bilobed marsupium, X 8. Palaeontology, Vol. 14 PLATE 127 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 675 b 10 a i S 5 Z3 a. 3 a). Marsupia with elongate interambulacral openings which extend from the apical system to about two-thirds of the distance to the ambitus. Internally they consist of flattened sac-like structures which extend beneath adjacent ambulacral columns. Much of each marsupium is formed by the proximal pair of interambulacral plates, which are elongated and meet along the floor of the marsupium in a straight median suture. One or two smaller interambulacral plates contribute to the distal portion of each marsupium. The proximal terminations are the distal ends of the introverted genital plates. The interambulacral plates comprising the marsupia all have surface expression, where they are covered by closely spaced granules. The inner surfaces of the marsupia are smooth. Males unknown. Measurements (in mm). Horiz. Vert. Peristome Apical No. No. diam. diam. diam. system diam. ambs. interambs. UNE 11512 9-5 6-6 3-2 3-8 11 (12) 9 Remarks. By analogy with other urchins it must be supposed that the remarkable marsupia of Pentechinus developed in the growing test at or immediately prior to sexual maturity. It is difficult to envisage how this could take place except through a complex 680 PALAEONTOLOGY, VOLUME 14 a b TEXT-FIG. 4. Pentechiniis mirabiUs gen. et sp. nov. a, Adapical view of holotype, P27942. Specimen tilted to show genital pores in three of the marsupia. b. Oblique adapical view of same specimen, x 10. PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 681 interaction of growth and resorption, or even growth within an ossicle, rather than at its margins. No further comment would be merited were it not for Durham’s (1966^?, p. 390) recent statement that in echinoids ‘Resorption . . . does not occur between adjacent plates’. Yet such resorption is well known. Hawkins (1913) used the term ‘plate crushing’ to describe the formation of ‘plates lessened in height and restricted in width’ during growth. The formation of demiplates in the anterior ambulacrum of Echinocardium cordatiim he ascribed to this process, to the satisfaction of later workers concerned with spatangoid ambulacral morphology (e.g. Chesher 1963, p. 555). Philip (1957) substituted the term ‘plate atrophy’ when describing the diminution in size with growth of the ambital interambulacral plates of the Miocene spatangoid Lovenia woodsi. Among regular echinoids, the very complex aggregation of pores seen in some living diadema- tids is a secondary development, late in the ontogeny of the urchin, and is brought about by resorption of the median zone (plus tubercles) of the adoral ambulacra (cf. Mortensen 1940, pp. 163-164). Order clypeasteroida A. Agassiz Suborder clypeasterina A. Agassiz Family eossulasteridae nov. Diagnosis. Small, flattened clypeasteroids with separate auricles, and lacking demi- plates in the petalloid ambulacra. Ambulacral food grooves simple or absent; apical system with four genital pores. Periproct supramarginal; interambulacra not extending on to the adoral surface. Included genera. Fossidaster Lambert and Thiery 1925, Scutellinoides Durham 1955, Willungaster gen. nov. Remarks. The three genera listed above, all from the Tertiary of south-eastern Australia, stand well apart from other clypeasteroids in possessing separate auricles and simple plates in the petalloid ambulacra. Mortensen (1948) divided the Clypeasteroida into two suborders, as advocated by Philip (1965u) and supported by Kier’s (1970, p. 105) study of clypeasteroid lantern supports. All previously described clypeasterinid echinoids have, beside their separate auricles, demi-plates in the petals. Scutellinoides was referred by Durham (1955, 1966) to the family Arachnoididae on the assumption that the genus possessed demi-plates. However, Durham regarded Fossulaster as incertae sedis with the suggestion (1955, p. 131) that the ambulacral plates may be simple. Scutellinoides is based on Scutellina patella Tate 1891, but, in reality, has been interpreted through Scutellina morgani Cotteau 1891 (e.g. Durham 1966o, p. U468) in the belief that these species are the same. The lectotype of Scutellina patella Tate is here chosen as Adelaide University Geology Department T27A, the largest of Tate’s eleven unfigured syntypes and the specimen that conforms most closely to the first of his measured specimens. Its locality is given as the ‘Murray River CliflFs’ S.A., where its parent stratum is interpreted as the Morgan Limestone. The lectotype and the abundant material of this species available from the Morgan Limestone indicate that ambulacral petals are lacking. 682 PALAEONTOLOGY, VOLUME 14 Scutellina morgani Cotteau is based on a single specimen housed in the Ecole des Mines, Paris, and is from the Gambier Limestone at Mount Gambier, S.A., where the species is common. Examination of the holotype reveals that it is correctly portrayed by Cotteau, i.e. it has well-developed ambulacral petals. S. morgani and S. patella are similar in possessing separate auricles, and in details of their plating structure. Elowever, because of differences in the ambulacra, here they are not considered to be congeneric. An account of these species, together with closely related forms, will be given elsewhere. Genus fossulaster Lambert and Thiery 1925 Fossulaster Lambert and Thiery, p. 577. 1955 Fossulaster Lambert and Thiery; Durham, p. 129. Type species. Fossulaster halli Lambert and Thiery. Diagnosis. Small clypeasterinids with ovate outline; periproct supramarginal, separated from the ambitus by 2 or 3 pairs of interambulacral plates. Ambulacral petals poorly defined, when apparent, open and with small non-conjugate pores. Ambulacral food groove simple and often poorly defined. Apical system with four genital pores. Inter- ambulacral columns narrow and extending only to the ambitus. Girdle of small poorly developed, separate auricles ; marginal concentric supports moderately developed ; inner supports variable. Females with a marsupium on the adoral surface to the anterior of the peristome. Fossulaster halli Lambert and Thiery Plate 125, fig. 3; Plate 127, fig. 3; Plate 130, figs. 1-10; Plate 131, fig. 6; Plate 132, fig. 10; text-figs. 5, 6 71898 Scutella marsupiata Tate, p. 196 (nomen nudum). 1908 Scutellina sp. (75. patella Tate) Hall, pp. 140-142, text-fig. 1916 Echinocyamus patella (Tate) Chapman, pp. 182-183, pi. 45, fig. 23. 1925 Fossulaster halli Lambert and Thiery, p. 577. 1946 Scutellina patella Tate; H. L. Clark, p. 351 (in part). 1955 Fossulaster halli Lambert and Thiery; Durham, pp. 129-131, figs. 3Sa-b. 1966 Fossulaster halli Lambert and Thiery; Durham, p. U468, fig. 3. 1969 Fossulaster halli Lambert and Thiery; Kier, pi. 14, figs. 1-4. Diagnosis. A moderately large species of Fossulaster in which females possess a deep, bilobed marsupium. Ambulacral petals poorly developed. Material. Hypotypes P19178 (?), P19179 ($), P19180 (cj), P19181 (<^), P19182 ($), P19183 (?), P19184 ($), and P19185-P19196 (8(^, 2$, 2 indet.) from Mt. Gambier, from the Gambier Limestone of Long- fordian (Lower Miocene) age. P27942 (?), P27948 (?), UNE 11762, UNE 11763, UNE 11764, and 88 other specimens (46,^, 32?, 10 indet.) from the Gambier Limestone at Nelson, Victoria. Fourteen specimens (9S, 5?) from Coobowie, just north of bridge over Salt Creek, Yorke Peninsula, S.A., from the Port Vincent Limestone from the Janjukian or Longfordian (Upper Oligocene or Lower Miocene). EXPLANATION OF PLATE 130 Figures X 5 unless otherwise stated. Figs. 1-10. Fossulaster halli Lambert and Thiery. 1, Adapical view of UNE 11762, X 10. 2, Lateral view of UNE 11760 (cj). 3, Lateral view. 4, Adapical view. 8, Adoral view of P19180 (cj). Circular hole in posterior part of test due to some boring organism. 5, Lateral view of P27942 (?). 6, Lateral view of P19178 (?). 7, Adapical view of UNE 11763, X 10. 9, Adoral view of P19179 (?), X 10. 10, Adapical view of P27942 (?), x 10. Palaeontology, Vol. 14 PLATE 130 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 683 f g h TEXT-FIG. 5. Fossulaster halli Lambert and Thiery. a. Oblique view from posterior, showing supple- mentary plates at the anterior end of the marsupium, P27948, X 8. b. Plating of adapical surface, interambulacral columns stippled, P19181 (cj), X 8 approx, c, Adapical view of small specimen, UNE 11762, X 10. r/. Plating of adapical surface, interambulacral columns stippled, PI 9182 (?), X 8 approx. e, Plating of adoral surface, P19183 (?), X 5 approx. /, Adapical view of UNE 11764, X 10 approx. g, Plating of adoral surface of a small female, PI 91 84, X 5. h, Adapical surface showing ambulacral petals, P27942 (?), x 5. 684 PALAEONTOLOGY, VOLUME 14 Remarks. New information on the morphology of this species is listed below. Marsupium. The structure of the bilobed marsupium has been discussed by Durham (1955) and Kier (1969). The marsupial depression appears at a length of about 7 mm and thereafter deepens rapidly. In large females it deepens to such an extent that it deforms the upper surface of the test so that a bulge is developed to the anterior of the apical system. During growth, adult females therefore become higher than males (text- fig. 6a) and their position of maximum height moves more to the anterior end of the test. Plate 125, fig. 3 is of a specimen prepared to show the internal features of the marsupium; portion of the upper surface of the test is firmly ankylosed to one of the lobes of the marsupium. The marsupium is formed by enlarged ambulacral plates, essentially those second and third from the peristome in the anterior ambulacrum. Initially it is only the second pair which form the marsupium, but with growth the third pair, as well as those of adjacent columns, contribute. In one large specimen supple- mentary plates are present in the central part of the marsupium (text-fig. 5a), but, as sutures were found difficult to develop, it could not be decided whether or not these were of more general occurrence. Growth. In small specimens (text-fig. 5c; PI. 130, fig. 1) the periproct abuts against apical system. At a length of about 5 mm it becomes detached (text-fig. 5/; PI. 130, fig. 7) and migrates to its adult position toward the posterior margin, which it achieves at a length of about 7 or 8 mm. Text-fig. 6c shows a growth series for the collection from Nelson. Noteworthy is the fact that females are consistently narrower than males. Study of the plates will show that the females also develop a more pentagonal outline. Apical system. The apical system shows considerable variation both in the size and location of the genital pores (text-fig. 5b, d). No correlation of these features, however, was found in relation to the presence or absence of a marsupium. Internal features. The separate auricles are small and insignificant. The so-called ‘radial’ partitions (actually interradial in position) are extremely variable. In males there are basically five pairs, although usually one or more pairs are absent or poorly developed ; in others a pair may be duplicated. Plate 132, fig. 10 shows a specimen with three supports in the posterior interambulacrum. In females, it is only the posterior pair, on each side of the periproct, which is consistently developed; all others may be absent. The floor of the test is furrowed by five radial grooves which are interpreted as traees of the radial water vessels. Smaller furrows extend out at right angles to these, and contain small pores, presumably to aceommodate the accessory tube feet (PI. 125, fig. 3; PI. 132, EXPLANATION OF PLATE 131 Figures X 5 unless otherwise stated. Figs. 1-5, 7. Willungaster sciiteUaris gen. et sp. nov. 1, Lateral view of paratype, UNE 12008 (1$, 71$, 8 indet.) from Edithburg, Yorke Peninsula, S.A., from a Janjukian or Longfordian (Upper Oligocene of Lower Miocene) horizon in the Port Vincent Limestone. Description. Test rounded in outline, in lateral view gently arched to reach maximum ! height at the apical system. Periproct rounded, just supramarginal, and maintaining this position during growth; in small specimens (PI. 131, fig. 5) located between two enlarged • posterior ambulacra; in mature specimens abutting against the same plates and the termination of the posterior interambulacral column (PI. 131, fig. 7; text-fig. 8a). Interambulacral columns narrow, in small specimens reaching about half-way to the ambitus; in larger ones extending up to three-quarters of this distance (PI. 131, fig. 7). Apical system somewhat posterior, small and with four genital pores. Ambulacra with well-developed open petals with slightly conjugate rounded pores mounted in the horizontal sutures. Adoral surface flattened in males, in females with a marked depression to the anterior ; of the rounded peristome. Plating of adoral surface consisting of three circlets of enlarged ambulacral plates, with a supplementary plate in column Ilia, present in both males and females (text-fig. ^b, c). Marsupium of females floored by the supplementary plate and plates of column III second from the peristome. Specimens with a marsupium usually more arched in lateral profile (cf. PI. 131, figs. 1 and 2). Basicoronal inter- ambulacral plates not visible. Surface of test covered with small, closely spaced, sunken tubercles, except within the marsupium where the test is smooth or sparsely granulated. Interior of test with small separate auricles, and five well-developed paired inter- i radial pillars (PI. 133, fig. 9). Up to two series of concentric supports within the ambital | margin of the test. | f i EXPLANATION OF PLATE 133 I Figures x 2-5 unless otherwise stated. | Figs. 1, 2, 4-6, 8. Peraspatangus brevis gen. et sp. nov. 1, Adoral view. 6, Adapical view of paratype, I UNE 11758 ($). 2, Lateral view. 5, Adapical view. 8, Adoral view of holotype, P27956 ($), X5. 4, Adapical view of paratype, P27955 (cj), X 5. ( Fig. 3. Peraspatangus depresses gen. et sp. nov. Lateral view of paratype, P27954 {$). 1 Figs. 7, 9. Willungaster sciitellaris gen. et sp. now. 7, Adapical view of paratype, UNE 12008 (t^), X 5. > 9, Adapical view of paratype, UNE 12010 (cj), X 5. Palaeontology, Vol. 14 PLATE 133 8 9 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 689 b C TEXT-FIG. 8. WiUimgaster scutellaris gen. et sp. nov. a. Plating of adapical surface of paratype, UNE 11767, interambulacral columns stippled, x5 approx, b. Plating of adoral surface of paratype, UNE 11765 (cJ), X 5 approx, c. Plating of adoral surface of paratype, UNE 11766 ($), X 5 approx. Order spatangoida Claus Suborder amphisternina Mortensen Family spatangidae Gray Genus peraspatangus gen. nov. Generic name. Erom the Greek pera, a bag. Type species. Peraspatangus brevis sp. nov. Diagnosis. Test small and ovate in outline; anterior notch not developed. Ambulacral petals flush with test and poorly defined, usually with fewer pore-pairs in the anterior columns; anterior ambulacrum narrow, with small simple pores. Apical system ethmo- lytic, with four genital pores. Adapical surface with small, closely spaced tubercles, or with several slightly enlarged tubercles. 690 PALAEONTOLOGY, VOLUME 14 On the adoral surface the ambulacra form only an inconspicuous phyllode. Labrum short, not extending past the neighbouring first ambulacral plates. Subanal fasciole well developed, with three ambulacral plates (and two pore-pairs) of each of the posterior ambulacra within the fasciole. Females with prominent adapical marsupium containing the apical system which in females is relatively larger and has larger genital pores; pore-pairs of petals not developed within marsupium. Remarks. Peraspatangus resembles most closely the living Indo-Pacific genus Nacospa- langus A. Agassiz. Both are of similar size, possess similar rudimentary ambulacra (including inconspicuous phyllodes), and lack prominent primary tubercles on the adapical surface. These features tend to separate Nacospatangus and Peraspatangus from other spatangids and invite comparison with palaeopneustids, particularly with the subfamily Palaeotropinae. Indeed, the distinguishing characters of this subfamily given by Mortensen (1950, p. 285) are those which unite Nacospatangus and Peraspatangus. Much confusion exists as to the relationship of the genera placed by Mortensen (1950) in the family Palaeoneustidae (= suborder Asterostomatina Fischer). Strongly con- trasted views on their phytogeny and classification are given by Mortensen (1950) and Fischer (1966). All that it is necessary to conclude here is that Peraspatangus has as its closest relative Nacospatangus, and Mortensen’s classification is followed to the extent that both are retained in the family Spantangidae. Peraspatangus differs from Nacospatangus in possessing four genital pores, even more rudimentary ambulacral petals, a shorter labrum and marsupiate females. Peraspatangus brevis sp. nov. Plate 126, fig. 3; Plate 133, figs. 1, 2, 4, 5, 6, 8; text-fig. 9b, c, d Material. Holotype P27956 ($), paratypes P27955 (cj), UNE 11758 ($), and 4 other specimens (2$, 2(J), from right bank of Grange Burn, just above junction with Muddy Creek, Hamilton district, Victoria, from the Bochara Limestone, Batesfordian (Lower Miocene) age. Diagnosis. A species of Peraspatangus with a narrow test and with females which possess a deep adapical marsupium. Adapical surface with small closely spaced tubercles, uniform in size. Description. Test small, oval in outline, with greatest width at midlength; adapical surface somewhat arched; centrally flattened or depressed. Adoral surface flat with a well-defined plastron. Anterior ambulacrum narrow, not contained in groove, consisting of high plates lacking pore-pairs (text-fig. 9b). Ambulacra straight, petals inconspicuous and flush with test; anterior columns of anterior petals containing fewer pore-pairs than posterior columns. Pore-pairs absent in ambulacral plates adjacent to the apical system within the EXPLANATION OF PLATE 134 Figures X 15. Fig. 1. Paradoxechinus stellatus sp. nov. Enlargement of ambital ornament of holotype, P27944 ($). Fig. 2. Paradoxechinus profundus sp. nov. Enlargement of ambital ornament of paratype, P27946 (cj). Fig. 3. Paradoxechinus granulosus sp. nov. Enlargement of ambital ornament of holotype, P27947 ($). Palaeontology, Vol. 14 PLATE 134 PHILIP and FOSTER, Marsupiate echinoids PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 691 C d TEXT-FIG. 9. a, Peraspatangiis depressus gen. et sp. nov. Adapical surface of holotype, P27953 ($), X 4 approx, b-d, Peraspatangiis brevis gen. et sp. nov. b, Adapical surface of paratype, P27955 {$), x5. c, Adapical surface of paratype, UNE 11758 (?), x5. d, Adoral surface of holotype, P27596 (?), x4 approx. 692 PALAEONTOLOGY, VOLUME 14 marsupium of females. Adapical surface covered with small closely spaced sunken tubercles. Apical system central in males, somewhat to the anterior in females ; ethmolytic with fused genital plates and with four genital pores. Apical system and genital pores smaller in males than in females. Adoral surface with anterior peristome with well-developed labrum; ambulacral phyllode poorly developed. Labrum short, only extending back beyond the first ambu- lacral plates; sternum broad and coarsely tuberculated; posterior ambulacra closely granulated. Subanal fasciole well defined, circular, containing three plates of the adjacent ambulacral columns; four pore-pairs within fasciole. Circular periproct marginal. Marsupium of females elongate and deep, at the anterior end with vertical or even overhanging sides, shallowing gradually toward its posterior end. Remarks. P. brevis differs from P. depressiis in its smaller, narrower test, in the straight posterior ambulacra, in the deeper adapical depression of females, and their smaller genital pores. In addition, several somewhat enlarged tubercles are present on the adapical surface of P. depressus. Peraspatangus depressus sp. nov. Plate 128, fig. 2; Plate 132, figs. 1, 4, 5, 7; Plate 133, fig. 3; text-fig. 9a Material. Holotype P27953 ($), paratype P27954 {$), and one other specimen, from the mouth of Ingle’s Creek, 3 miles east of Port Campbell, Victoria, from the Rutledge Marl Member of the Port Campbell Limestone of Bairnsdalian (Middle Miocene) age. Diagnosis. A species of Peraspatangus with a comparatively broad test and with females which possess a shallow adapical marsupium. Adapical surface with several somewhat enlarged interambulacral tubercles, in females bordering the marsupium. Description. Test moderate sized, oval in outline, with greatest width at about midlength. Adapical surface centrally flattened or depressed. Adoral surface strongly flattened with a well-developed plastron. Anterior ambulacrum lacking pore-pairs and not contained in a groove. Petals flush with test and inconspicuous; anterior columns of petals containing fewer pore-pairs than posterior columns (text-fig. 9a). Posterior ambulacra markedly sinuous in females. Pore-pairs lacking in ambulacral plates adjacent to the apical system (i.e. within the marsupium) of females. Adapical surface with a group of somewhat enlarged tubercles in each of the paired interambulacra. In females these border the marsupium. Apical system somewhat anterior, ethmolytic, with fused genital plates and with four genital pores. Apical system and genital pores smaller in males than in females. Adoral surface with anterior peristome with well-defined labrum; ambulacral phyl- lodes poorly developed. Labrum short, sternum broad and coarsely tuberculated as are the areas outside the ambulacra. Subanal fasciole rounded, apparently as in P. brevis. Periproct circular and marginal. Marsupium of females an elongate and comparatively gentle depression, with the apical system at its anterior end. Remarks. A fourth, poorly preserved specimen which possibly represents an additional male test was also collected at the type locality. Comparison with P. brevis is given in discussion of that species. PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 693 Acknowledgements. Philip’s research was latterly supported by a University of New England Research Grant. This paper was completed while he was National Foundation Foreign Scientist Fellow, Univer- sity of Iowa. Much of the material described, including the types of Fossulaster exigiius and Willimgaster sciitel- laris, comes from the collection of South Australian echinoids made by F. Foster and R. J. Swaby. J. M. Lindsay measured, sampled, and dated the coastal section of Port Willunga Beds just south of Port Noarlunga, South Australia, which has provided the key for allocating ages to the Eocene and Oligocene species described here. REFERENCES BITTNER, A. 1892. Uber Echiniden des Tertiars von Australien. Sitz K. Akad. Wissen. Wien {Math. Nat. CL), 101, 331-371, pis. 1-4. BLOW, w. H. 1970. Validity of biostratigraphic correlations based on the Globigerinacea. Micro- paleontology, 16, 257-268. CHAPMAN, F. 1916. Cainozoic geology of the Mallee and other Victorian bores. Rec. geol. Surv. Viet. 3, 327-430, pi. 63-78. CHESHER, R. H. 1963. The morphology and function of the frontal ambulacrum of Moira atropos (Echinoidea: Spatangoida). Bull. mar. Sci. Gulf Carribbean, 13, 549-573. CLARK, H. L. 1946. The echinoderm fauna of Australia. Its composition and origin. Publ. Carnegie Instn. 566, 1-567. COTTEAU, G. H. 1891. Echinidcs nouveaux ou peu connus. 2 (10) Bull. Soc. Zool. France 1891, 149- 162, pi. 20. (Pagination, etc. of separate publication.) PERON, p. A. and GAUTHIER, v. 1881. Echinides fossiles de L'Algerie. 7-8, Etage Senonien, 1-182, pi. 20. Paris. DAVID, T. E. w. 1950. The geology of the Commonwealth of Australia (ed. w. R. browne). 1, 1-747, Edward Arnold and Co. London. DIETZ, R. s. and holden, j. c. 1970. Reconstruction of Pangea: breakup and dispersion of continents, Permian to present. J. Geophys. Res. 75, 4939^956. DORMAN, F. H. and GILL, E. D. 1959. Oxygen isotope palaeotemperature measurements on Australian fossils. Proc. roy. Soc. Viet. 71, 73-98, pi. 8. DUNCAN, p. M. 1877. On the Echinodermata of the Australian Cainozoic (Tertiary) deposits. Quart. J. geol. Soc. Lond. 33, 42-71, pis. 3-4. 1887. A revision of the Echinoidea from the Australian Tertiaries. Ibid. 43, 411-430. DURHAM, J. w. 1955. Classification of clypeasteroid echinoids. Univ. California Publ. geol. Sci. 31, 73-198, pis. 3-4. 1966fl. Clypeasteroids in moore, r. c. (ed.). Treatise on invertebrate paleontology. Part U, U450- 491, Univ. Kansas Press and Geol. Soc. Amer. \966b. Evolution among the Echinoidea. Biol. Rev. Camb. Phil. Soc. 41, 368-391. ETHERIDGE, R. Jr. 1875. On the occurrence of the genus Hemipatagus Desor, from the Tertiary rocks of Victoria, with notes on some previously described species from South Australia. Quart. J. geol. Soc. Lond. 31, 444-450, pi. 1. FELL, H. B. 1949. An echinoid from the Tertiary (Janjukian) of South Australia, Brochopleurus australiae sp. nov. Mem. nat. Mus. Melb. 16, 17-19. and PAWSON, d. l. 1966. Echinacea in moore, r. c. (ed.). Treatise on invertebrate paleontology. Part U, U367^40, Univ. Kansas Press and Geol. Soc. Amer. FISCHER, A. G. 1963. Essay review of Descriptive Palaeoclimatology (ed. a. e. m. nairn), Amer. J. Sci. 261, 282-293. 1966. Spatangoids in moore, r. c. (ed.). Treatise on invertebrate paleontology Part U, U543-628, Univ. Kansas Press and Geol. Soc. Amer. GREGORY, J. w. 1892. A revision of the British fossil Cainozoic Echinoidea. Proc. geol. Assoc. 12, 16-60, 2 pis. HALL, T. s. 1908. On the occurrence of a marsupium in an echinoid belonging to the genus Scutellina. Proc. roy. Soc. Viet. 20, 140-142. C 8385 zz 694 PALAEONTOLOGY, VOLUME 14 HAWKINS, H. L. 1913. The anterior ambulacrum of Echinocardium cordatum, Penn., and the origin of compound plates in echinoids. Proc. zool. Soc. Loud. 12, 169-181. HYMAN, L. H. 1955. The invertebrates, 4: Echinodermata. The coelomate Bilateria. 1-763, McGraw-Hill, New York. iKEDA, H. 1931. A biometric study of the sexual dimorphism and sex ratio in Temnopleiirus toreumatictis (Klein). Annot. zool. Japan, 13, 233-242. KiER, p. M. 1967n. Sexual dimorphism in an Eocene echinoid. J. Paleont. 41, 988-993, pis. 129-130. 19676. Revision of the oligopygoid echinoids. Misc. Coll. Smithson. Inst. 152, 1-147, 36 pis. 1968. Echinoids from the middle Eocene Lake City Eormation of Georgia. Ibid. 153, 1-45. 1969. Sexual dimorphism in fossil echinoids in westermann, g. e. g. (ed.). Sexual dimorphism in fossil Metazoa and taxonomic implications. 215-222, pi. 14, E. Schweizerbart, Stuttgart. 1970. Lantern support structures in the clypeasteroid echinoids. J. Paleont. 44, 98-109, pis. 23-24. LAMBERT, J. 1928. Note sur quelques echinides du Cretace d’Espagne communiques par M. le Royo y Gomez. Bol. Real Soc. Hist. nat. Espanola, 28, 145-157. 1933. Echinides de Madagascar communiques par M. H. Besairie. Ann. geol. serv. Mines Mada- gascar, 3, 7-49, pis. 1-4. and THiERY, p. 1909-1925. Essai de nomenclature raisonnes des echinides. 1-607, 15 pis. Chaumont. LAUBE, G. c. 1869. fiber einige fossile Echiniden von den Murray Cliffs in Siid-Australien. Sitz. k. Akad. Wissen. Wien (Math. Nat. CL), 59, 183-198, 1 pi. LE piCHON, X. and heirtzler, j. r. 1968. Magnetic anomalies in the Indian Ocean and sea-floor spread- ing. J. Geophys. Res. 73, 2101-2117. LINDSAY, J. M. 1970. Port Willunga Beds in the Port Noarlunga-Seaford area. S. Aust. Mines Dept. Pal. Rep. 12/70, 1-10. MCGOWRAN, B., LINDSAY, J. M., and HARRIS, w. K. 1971. Attempted reconciliation of Tertiary biostrati- graphic systems, Otway Basin. Ch. 14 in The Otway Basin in southeast Australia. Spec. Bull. geol. Sur vs. S. Aust. Viet, (in press). MORTENSEN, T. 1909. Die Echinoiden in Rept. Deutsch. Siidpolar-Exped. 1901-3, 2 (3), 1-113, 19 pis., G. Reimer, Berlin. 1935. A monograph of the Echinoidea II, 1-647, 89 pis., C. A. Reitzel, Copenhagen. 1940. Idem III (1), 1-370, 77 pis. 1943. Idem III (3), 1-466, 66 pis. 1948. Idem IV (2), 1-471, 72 pis. 1950. Idem V (I), 1-432, 25 pis. 1951. Idem V (2), 1-593, 64 pis. PERON, p. A. and GAUTHIER, V. 1881 . Ecliitiides fossUcs de L' Algcric. 7, Etage Tiironian 1-1 10, 9 pis. Paris. PHILIP, G. M. 1957. Interambulacral plate atrophy in Lovenia woodsi (Etheridge fil.) Geol. Mag. 94, 402^08. 1963fl. The Tertiary echinoids of south-eastern Australia. I. Introduction and Cidaridae (1). Proc. roy. Soc. Viet. 76, 181-226, pis. 21-26. 19636. Two Australian Tertiary neolampadids, and the classification of cassiduloid echinoids. Palaeont. 6, 718-726, pis. 106-107. 1964. The Tertiary echinoids of south-eastern Australia. II. Cidaridae (2). Proc. roy. Soc. Viet. 77, 433-474, pis. 58-67. 1965n. A classification of echinoids. J. Paleont. 39, 45-62. 19656. The Tertiary echinoids of south-eastern Australia. III. Stirodonta, Aulodonta and Camarodonta (1). Proc. roy. Soc. Viet. 78, 181-196, pis. 26-29. 1969. Idem. IV. Camarodonta (2). Ibid. 82, 233-276, pis. 3-16. STUART, w. J. 1970. The Cainozoic stratigraphy of the south-eastern coastal area of Yorke Peninsula, South Australia. Trans. R. Soc. S. Aust. 94, 151-178. TATE, R. 1891. A bibliography and revised list of the described echinoids of the Australian Eocene, with descriptions of some new species. Trans. Proc. roy. Soc. S.A. 14, 270-282. 1892. Critical remarks on A. Bittner’s ‘Echiniden des Tertiars von Australien’. Ibid. 15, 190-194. PHILIP AND FOSTER: AUSTRALIAN TERTIARY MARSUPIATE ECHINOIDS 695 TATE, R. 1898. On deep-seated Eocene strata in the Croydon and other bores. Ibid. 22, 194-199. THORSON, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. Canib. phil. Soc. 25, 1-45. R. J. FOSTER B.H.P. Oil and Gas Division 440 Collins Street Melbourne, Vic. G. M. PHILIP Department of Geology University of New England Typescript received 19 December 1970 Armidale, N.S.W. A NEW MICROMORPHIC RHYNCHONELLIDE BRACHIOPOD FROM THE MIDDLE JURASSIC OF ENGLAND by P. G. BAKER Abstract. Recent work on micromorphic brachiopod faunas from the Inferior Oolite reveals the presence of a new rhynchonellide species attributable to Nannirhynchia Buckman but with diagnostic beak characters. The new species N. longirostra is proposed. Although the presence of a pedicle collar has not previously been recorded in the Norellinae, the form of the shell, the definitely arcuifer crura and the way in which they curve ventrally into the pedicle valve are thought to justify its inclusion in the sub-family. Juveniles are easily recogniz- i' able but no characteristic transition to adult form is noted. Investigation of the shell microstructure leads to the I conclusion that deltidial plates are not invariably developed. The possibility of synonymy with Spiriferiiial minima Moore, is discussed. The species investigated was recovered from the Oolite Marl, a weakly coherent j interbedded marl and biomicrite (Folk 1959) of Lower Bajocian age occurring in the ( Inferior Oolite of the mid-Cotswolds. Organo-detrital residue obtained from treated ‘ I marl yields numerous other brachiopod micromorphs, notably Moorellina granulosa |j (Moore), juveniles of terebratulides such as ZeiUeria leckenbyi (Walker) and 'Terebratula' ‘ whitakeri Walker, MS., also rare specimens of Zellania davidsoni Moore. A detailed ; account of the techniques adopted for preparation of the material for study may be (. found in Baker (1969, 1970). \ Although the detail of separated valves is well preserved, the interior of uncrushed j , complete shells is invariably coated with minute crystals which makes determination of I jj the form of the crura difficult. Fortunately, the strong beak resists diagenesis very well j i and its internal characters may be easily ascertained by serial sectioning. | ;| Acknowledgements. The author is indebted to Professor A. Williams, Department of Geology, Queen’s I il University, Belfast, for advice regarding certain microstructural features, also to Professor D. V. Ager, | .| Department of Geology, University College, Swansea, for confirmation of the form of the crura. | Thanks are due to Dr. J. D. Hudson, Department of Geology, University of Leicester, for general | discussion of the work and to Mr. G. McTurk for preparing the stereoscan negatives. Finally I wish i to thank Professor P. C. Sylvester-Bradley for use of the research facilities of the University of Leicester. 1 I Registration of Material. The holotype and topotypes, together with sectioned material in the form of : original and duplicate cellulose acetate peels are to be housed in the British Museum (Natural History) under the register numbers BB 45820-45829. Reference numbers quoted in the plate figures refer J to stereoscan negatives of figured specimens which are to be retained in the negative library of the ) Department of Geology, University of Leicester. \ Order rhynchonellida Kuhn 1949 I Family DiMERELLiDAE Buckman 1918 ' Genus nannirhynchia Buckman 1918 , |j Nannirhynchia longirostra sp. nov. I | Plates 135-137; text-figs. 1, 2 / | S [Palaeontology, Vol. 14, Part 4, 1971, pp. 696-703, pis. 135-137.] i j P. G. BAKER: MICROMORPHIC BRACHIOPOD 697 Diagnosis. Very small, biconvex Nannirhynchia, up to about 2-5 mm in length, 2-0 mm in width, and 1-0 mm in thickness. Pyriform in outline with well-developed intraplicate anterior commissure. Rostrate. Radial striae clearly seen on specimens in which the primary shell layer is weathered. Type specimens. Holotype and nine topotypes to be housed in the British Museum (Natural History), register number BB 45820. Distribution. Geographic distribution unknown. All the material studied was collected from Westington Hill Quarry, grid ref. SPI 42368, near Chipping Campden in Gloucestershire. Stratigraphically the species is so far known only from a single horizon in the Lower Inferior Oolite (Lower Bajocian, Murchisonae Zone), notably at the base (Baker 1970) of the Oolite Marl, where it is associated with other micromorphic brachiopods such as Moorellina granulosa (Moore) and Zellania davidsoni Moore. Dimensions of Holotype. Length 2 3 mm, width 1-8 mm, thickness 10 mm. Description. External Characters. Study of 39 specimens and numerous shell fragments shows that the recorded dimensions of the holotype are very close to the maximum for the species and that the beak characteristics are stable throughout the material studied. The arguments advanced previously (Baker 1969), in favour of regarding the associated micromorphic brachiopods as adults, are equally applicable to this study. Ontogenetic stages are available which show that adults are characterized by a well-developed, rounded, central plication with complementary sulci on either side and occasionally a single, low, lateral costa on each side. The shell is typically non-strophic sensii Rudwick and is biconvex throughout growth. In young forms, both the brachial and pedicle valves have a median sulcus and thus a ligate form. The central plication is produced by bifurcation of the brachial sulcus and subsequent deepening of the thus formed complementary sulci. Therefore as Ager (1967) has shown for N. pygniaea the plication in TV. longirostra also is a consequence of the development of the sulci and thus intraplicate sensii Buckman, rather than uni- plicate with lateral sulci. Striae when visible are a diagenetic development. The beak is strong, sharp, projecting, and slightly incurved with narrow palintropes. The delthyrium is open, hypothyridid, and bounded by strongly developed tooth ridges, with or without narrow deltidial plates. Internal Characters: Pedicle valve. The most characteristic feature of the pedicle valve is the well-developed, almost sessile pedicle collar with its clearly defined growth lines (PI. 136, fig. 1). The hinge teeth of this species are relatively large. Dental plates are present. These are inclined and show a strong lateral deflection so that the lateral umbonal cavities remain very small (PI. 137, figs. 3, 4, text-fig. 2). The dental plates are easily overlooked in adult shells as the lateral umbonal cavities quickly become filled with subsequently deposited secondary shell so that the dental plates appear to form part of the umbonal wall (PI. 136, fig. 1). In young forms the dental plates are sometimes visible and the partially infilled lateral umbonal cavities appear as small perforations (PI. 136, fig. 2). The delthyrium is bounded by narrow widely disjunct structures usually interpreted as deltidial plates. Reference to these structures will be made later. 698 PALAEONTOLOGY, VOLUME 14 TEXT-FIG. 1 . A. Block reconstruction of the posterior region of the right half of the shell of N. longi- rostra, based on superimposed cellulose acetate peels. The anterior faces of the block segments correlate with peels 37549/43, 26, 12, and 8 respectively, x60 approx, b, c. Impressions of the ventral (b) and dorsal (c) muscle fields (stippled) of N. longimstra reconstructed from visible boundaries and the distribution of terminally pitted secondary fibres, X 17. D. Trace of a stereoscan montage x225, of part of peel 37549/15 showing the orientation of the secondary shell fibres in the various structures depicted, c. ventrally deflected arcuifer crus, d.pl. dental plate, h.pl. hinge plate, i.c. infilled lateral umbonal cavity, p.c. pedicle collar, p.l. primary layer, t. hinge tooth, t.r. tooth ridge. Patches of recrystallized shell cross-hatched. EXPLANATION OF PLATE 135 Stereoscan photomicrographs of specimens of Nannirhynchia longirostra sp. nov. Material of all figures coated with evaporated aluminium before photography. Figs. 1-3. Brachial, anterior, and lateral views of the holotype 37550, X 30. Figs. 4-7. Brachial, anterior, posterior, and lateral views of a juvenile 37551. The posterior view (fig. 6) shows the tooth ridges surmounted by small deltidial plates, x40. Fig. 8. Pedicle view of a juvenile 37552, showing the laterally deflected beak and radial striae, X 30. Fig. 9. Interior of a brachial valve 37558, showing the hinge plates and the absence of a cardinal process and median septum. Crura broken, X 32. Fig. 10. Interior of a pedicle valve 37546, showing the pedicle collar, hinge teeth, and denticula, X 32. Palaeontology, Vol. 14 PLATE 135 " ,'v •^•iYA,: .'V- 1 _•!•;§' '-r. " .•r.t WOy. BAKER, Micromorphic brachiopod P. G. BAKER: MICROMORPHIC BRACHIOPOD 699 Small denticula are present, clearly visible on separated valves (PL 135, fig. 10; PL 1 36, fig. 2) but not well defined in transverse sections. Owing to the effects of diagenesis on the internal surface of shells it has not been possible to determine the true configuration of the muscle scars. The interior of separated valves, however, often shows impressions of the muscle fields and diagenesis has not completely obliterated the stellate pitting on the terminal faces of the secondary fibres lying within these areas (PL 136, fig. 7). The form of the muscle fields is shown in text- fig. iB, c. 6 8 0-12 0-16 10 12 14 — O O o 0-2 0-24 0 28 15 _ 16 - 0-3 0-32 0-38 0-66 0-7 0-72 0-74 0- 8 0-84 0-86 0-9 TEXT-FIG. 2. Nannirhynchia longirostra sp. nov., series of 21 serial transverse sections through specimen 37549, drawn from cellulose acetate peels, showing all the features of the species. Numbers refer to specific peels illustrated (top left) and the distance from the beak. Subsequent shell deposition stippled. All X20. Brachial Valve. There is no cardinal process but the posterior end of the valve protrudes into the delthyrial cavity. Hinge plates are present (PL 135, fig. 9; PL 136, fig. 4) but not well developed and in the material examined they are horizontal and not fused posteriorly so that there is nothing approaching a septalium and no median septum (text-figs. 1a, 2). Crural bases arise dorsally from the hinge plates (PL 137, fig. 7) and give rise to ventrally deflected, arcuifer crura (PL 137, figs. 5, 6, 8). Outer socket ridges are present but as in the case of the denticula they are only clear in separated valves (PL 136, fig. 4). Microstructure. The microstructure of shells is characteristically rhynchonellide (Williams 1968). The primary layer and the terminal faces of secondary fibres often 700 PALAEONTOLOGY, VOLUME 14 show some effects of diageiiesis but the fabric of the secondary layer itself is generally remarkably well preserved (PI. 136, figs. 6, 8; PI. 137, figs. 1, 2). Individual fibres are very long, often extending from the junction with the primary layer, right through the secondary layer to its internal boundary. This produces a rather weak shell and half valves, fractured longitudinally are most common. Secondary fibres often show well- defined growth lines representing stages in the repeated advance of the terminal face during fibre growth (PI. 137, fig. 2). The overall impression is that the secondary fibres are relatively flatter and much larger than in many other brachiopods. They do not show any marked decrease in size in the muscle fields but exhibit the curious stellate pitting of myotest fibres (PI. 136, fig. 7). Williams (personal communication) suggests that these pits may represent differential diagenetic solution but in the author’s opinion their regularity indicates a fundamental property of the underlying fibre. In section the fibres show the characteristic blurring of outline (PI. 136, fig. 8) and at high magnifica- tion, the minute trails associated with myotest. Ontogeny. Complete juveniles (PI. 135, figs. 4-8) are apparently much more common than complete adults but there is no difficulty in assigning these to the same species owing to the fact that their typical morphology can easily be seen in the early growth lines of adult shells (PI. 135, fig. 1). The pedicle collar as may be anticipated from its growth lines, is present in even the smallest pedicle valves studied. Brachial valves of a complementary size show broken crura so that the development of crura also is a very early feature. No characteristic ontogenetic stages are therefore present and the only EXPLANATION OF PLATE 136 Stereoscan photomicrographs (except fig. 8) of specimens of N. longirostra. Material of all figures coated with evaporated aluminium before photography. Fig. 1. Interior of a pedicle valve 37546, tilted to show the relatively large hinge teeth and the pedicle collar with its characteristic growth lines. The dental plates are apparently merged with the shell wall owing to infilling of the lateral umbonal cavities. Deltidial plates not developed. Tilt angle 68°, x75. Fig. 2. Interior of a juvenile pedicle valve 37559, tilted to show the denticulum and the hinge tooth supported by a dental plate with the relic of the left lateral umbonal cavity (arrowed). Tilt angle 65°, Xl65. Fig. 3. Enlarged view of the umbonal region of specimen 37551, showing the concave terminal face of ' each deltidial plate, x 100. Fig. 4. Enlarged view of the umbonal region of the interior of a brachial valve 37547, showing the inner and outer socket ridges, the crura, broken, the small hinge plates, and the absence of a cardinal process and median septum, x 95. Fig. 5. Enlarged view of the striae on the external surface of the shell of specimen 37552 to show that j they are an expression of the diagenesis of the primary shell layer, X 1000. Eig. 6. Junction of the primary and secondary layers in an exfoliated shell 37564, showing the well- | preserved secondary mosaic, left, and the primary shell with independently aligned striae, X 1000. j Fig. 7. Enlarged view of a portion of specimen 37562 showing the stellate pitting of the fibres lying 1 in the depressed areas interpreted as muscle fields. Location of section, floor of the brachial valve 4 approaching the mid-line, X 4500. Eig. 8. Electron micrograph of a two-stage replica of an oblique section through the shell underlying a muscle field, showing the characteristic blurring of the outline of fibres associated with myotest. ( Location of section, approximately as in Fig. 7, X 2200. Palaeontology, Vol. 14 PLATE 136 - V A- : A ; f •>? BAKER, Micromorphic brachiopod P. G. BAKER: MICROMORPHIC BRACHIOPOD 701 significant change during growth is the rapid transition from opposite to alternate folding. Buckman (1918, p. 81) used the term pliciligate for the change from opposite to alternate folding by the formation of a fold within the dorsal sulcus. However, pliciligate as defined by Williams and Rowell (1965, H150) implies only a low folding in the brachial valve. The amplitude of the folding in N. longirostra is symmetrical through the plane of the commissure. As the distinction between intraplicate and plicili- gate semu Buckman is not clear and in order to avoid confusion, the term intraplicate adopted by Ager (1967, p. 139) is adhered to. Remarks. Attention has been paid to the radial striae of Nanuirhynchia (Buckman 1918, p. 67; Ager 1967, p. 139). Stereoscan photomicrographs of N. longirostra (PI. 136, figs. 5, 6) show that the appearance of striae is an expression of diagenesis of the primary shell, which may but does not normally, reflect the orientation of the secondary fibres. Work in progress on the origin and growth of deltidial plates shows that they arise from the dorsal surface of the tooth ridges. In numerous instances in N. longirostra and probably many of the other micromorphic forms with so-called very narrow deltidial plates, the epithelium responsible for initiating the development of the deltidial plates never received the necessary stimulus and the deltidial plates failed to develop. In which case the narrow ridges bounding the delthyrium and having the appearance of deltidial plates are in fact the exposed dorsal surfaces of the tooth ridges. Where deltidial plates are present (size of the animal seems to be no criterion) their development parallels the anterior extension of the teeth (PI. 135, fig. 6) and they show a characteristic concave terminal face (PI. 136, fig. 3). The separation of the pedicle collar from the shell wall closely resembles that described by Copper (1965, p. 363, text-fig. 3a) in the atrypoid Mimatrypa insquanwsa (Schnur.). Unfortunately, although the absence of lateral umbonal cavities is noted, the text- figure does not include a key to the shell-material type so it is not possible to determine whether they are secondarily infilled. In Ager (1967) five nominal species have been previously recognized as belonging to the genus Nannirhynchia. These are N. milvina Buckman, N. ? minima (Moore), N. ? moorei (Davidson), N. pygmaea (Morris), and N. suhpygmaea Buckman. N. milvina may be distinguished from N. longirostra by its shorter beak, episulcate folding, and later occurrence (Upper Bajocian). The original of Spiriferinai moorei is apparently lost and Davidson was probably correct in assuming that it was a juvenile Spiriferina. Ager (1967) discussed the possibility that the specimen might have been a stunted adult and stated that there is no particular reason for attributing it to Nannirhynchia as suggested by Buckman (1918). Distinction between N. pygmaea and N. subpygmaea is not indis- putably established but Ager prefers to keep them taxonomically separate in view of the present lack of knowledge of N. subpygmaea and its later occurrence (N. pygmaea. Upper Lias; N. subpygmaea. Upper Bajocian). Both species may be easily distin- guished from N. longirostra by their beak characters. It is probable that the species is very closely related to the form described by Moore (1861, p. 190, pi. 2, figs. 19, 20) as Spiriferinal minima and the specimen illustrated in fig. 19 undoubtedly shows great similarity with the holotype of N. longirostra. However, significant differences exist between N. longirostra and the illustration of S.‘t minima in Davidson (1874, Suppl. pi. 11, fig. \la) which in the context of the sentence ( Davidson 702 PALAEONTOLOGY, VOLUME 14 1874, p. 103) purports to be a careful drawing of one of Moore’s specimens. David- son illustrates a correctly defined (Williams and Rowell 1965, H145) hinge line. His figure implies the presence of an interarea and therefore a strophic shell (Rudwick 1959, p. 19) differing from the non-strophic shell of N. longirostra. Moore himself states that the area is broad and flattened and refers to a broad hinge line but figures what is essentially a cardinal margin similar to that of N. longirostra. An interesting feature is the way Moore (1861, pi. 2, fig. 20) illustrates a lateral deflection of the beak. This character is commonly observed in young forms of N. longirostra in the material collected from Westington Hill quarry (PI. 135, fig. 8). Buckman (1918, p. 68) noted the similarity between 5.? minima and N. siibpygmaea (Walker) but also commented on the large beak and apparently straight hinge line depicted by Davidson. Later (p. 74) Buckman assigns these characters to the genus Rectirhynchia and suggests both Moore and Davidson may have unwittingly combined details from more than one specimen, i.e. combined the characters of two species. Both Moore and Davidson refer to the presence of a triangular deltidium yet none of the figures clearly shows the structure. Davidson’s figure shows a constriction at the anterior of the delthyrial region, very similar to the disjunct deltidial plates of certain young terebratulides, whilst Moore (1861) figures either an open delthyrium bounded by very narrow deltidial plates or has shading to indicate a concave plate. Reference to their accounts of other micromorphic forms precludes the possibility that both authors could be mistaken with regard to the presence of a plate situated in the delthyrium. The conclusion must therefore be that 5.? minima possessed a concave plate within the delthyrium, identifiable with the pedicle collar of N. longirostra. Unfortunately a careful search through the micromorphic brachiopods in the Moore collection has failed to reveal the types of S.‘t minima. They were not listed by Dr. Wallis when he catalogued the collection in 1927 so it appears that the originals are lost. The degree of affinity between N. longirostra and S.^. minima must therefore remain a matter of speculation. Demonstration of the close similarity between the two forms introduces the possibility that we are here, simply recording the discovery of 5”.? minima EXPLANATION OF PLATE 137 Stereoscan photomicrographs (except figs. 5, 6) of N. longirostra. Material of all figures coated with evaporated aluminium before photography. Fig. 1. Portion of an exfoliated shell showing the extension of individual secondary fibres from the outer to the inner (upper left) boundary of the secondary layer, X 200. Fig. 2. Surface of an exfoliated secondary fibre showing growth lines, X 2200. Fig. 3. Enlarged portion of peel 37549/15 showing the dental plate and subsequently infilled lateral umbonal cavity. T.S. X 150. Fig. 4. Enlarged portion of peel 37576/19 showing the pedicle collar developed from the inner surface of the dental plate and the very small infilled lateral umbonal cavity, close to its posterior termination. T.S. Xl50. Fig. 5. Photomicrograph, reflected light. Transverse section through specimen 37549 showing the crystallite coated interior of the shell and the crystallite encrusted, arcuifer crura, x 40. Fig. 6. Retouched copy fig. 5. Fig. 7. Enlarged portion of peel 37576/26, showing the crural base {c.b.) adjacent to the inner socket ridge and arising from the dorsal surface of the hinge plate (h.p.). T.S. X 150. Fig. 8. Horizontal section through specimen 37587 showing the form of the left crus, X 175. Palaeontology, Vol. 14 PLATE 137 BAKER, Micromorphic brachiopod P. G. BAKER: MICROMORPHIC BRACHIOPOD 703 in a new locality, and raises the question of the validity of the new species. In view of the very limited knowledge of SJ minima and in view of the absence of the original types, it is felt that there is justification in the erection of a new species to include the Westington Hill material. REFERENCES ACER, D. v. 1965. Mesozoic and Cenozoic Rhynchonellacea, in moore, r. c. (ed.), Treatise on inverte- brate paleontology, part H, Brachiopoda, H597-632. Univ. Kansas. 1967. The British Liassic Rhynchondlidae, Part IV. Palaeont. Soc. (Monogr.), London, 121, 137-172. BAKER, p. G. 1969. The ontogeny of the thecideacean brachiopod Moorellina granulosa (Moore) from the Middle Jurassic of England. Palaeontology, 12, 388-399. 1970. The growth and shell microstructure of the thecideacean brachiopod Moorellina granulosa (Moore) from the Middle Jurassic of England. Ibid. 13, 76-99. BUCKMAN, s. s. 1917 (1918). The Brachiopoda of the Namyau Beds, Northern Shan States, Burma. Palaeont. Indica, new ser. 3, 299 pp. COPPER, p. 1965. Unusual structures in Devonian Atrypidae from England. Palaeontology, 8, 358-373. DAVIDSON, T. 1874. Supplement to the British Jurassic and Triassic Brachiopoda, British Fossil Brachio- poda, pt. 4, 103, Sup. PI. 11. Palaeont. Soc. (Monogr.), London. FOLK, R. L. 1959. Practical petrographic classification of limestones. Bull. Amer. Ass. Petrol. Geol. 43, 1-38. MOORE, c. 1861. On new Brachiopoda and the development of the loop in Terebratella. Geologist, 4, 190-194. RUDWiCK, M. j. s. 1959. The growth and form of brachiopod shells. GeoL Mag. 96, 1-24. WILLIAMS, A. 1968. Evolution of the shell structure of articulate brachiopods. Spec. Pap. Palaeont. 2, 55 pp. and ROWELL, a. j. 1965. Morphology, in moore, r. c. (ed.). Treatise on invertebrate paleontology, part H, Brachiopoda, H57-155. Kansas. p. G. baker Department Biological Sciences Derby and District College of Technology Derby DE 3 1GB. Typescript received 15 January 1971 CYCLOCYSTOIDES FROM EARLY MIDDLE CAMBRIAN ROCKS OF NORTHWESTERN QUEENSLAND, AUSTRALIA by R. A. HENDERSON and J. H. SHERGOLD Abstract. A new species of Cyclocystoides, C. primotica, is described from the early Middle Cambrian Ordian Stage of the Mt. Isa district, Queensland, Australia. It comprises the first record of the Cyclocystoidea from rocks older than Middle Ordovician, and from the Southern Hemisphere. Derivation of the class from the Cystoidea is unlikely on stratigraphic grounds; its earliest representative suggests no close affiliation with con- temporaneous echinoderm taxa. It is suggested that undulation of the marginal ring provided feeding currents in life. Two well-preserved external moulds of the aboral surface of a single species of Cyclo- cystoides are represented in a fossil collection made by Mr. F. de Keyser (formerly of the Australian Bureau of Mineral Resources) during a detailed mapping programme promoted by the search for phosphorite in 1968. The specimens were picked out by one of us (JHS) following routine examination of the collection, and were identified as Cyclocystoides by the second author during a visit to the Bureau of Mineral Resources. In consequence, this report has been prepared as a joint project. The specimens were found at locality QP 243, which is three miles WSW. of Cornford Bore, and approximately 28 miles WSW. of Mt. Isa, at latitude 20° 49-5' S., longitude 139° 03-5' E. (text-fig. 1). They are from the Yelvertoft Bed, the basal unit of the Beetle Creek Formation in this district. The cyclocystoids are preserved in laminated chert, associated with silicified coquinite, algal chert, and siltstone. These rocks overlie basal Cambrian sandstone and conglomerate, which are unconformable on Precambrian sediments (Mingera Beds). The fossiliferous horizon lies about 20 ft above the unconformity. The cyclocystoids are associated with new species of the trilobite genus Redlichia (see PI. 138, fig. 3) which are of late Ordian (early Middle Cambrian) age. Similar species of Redlichia occur at a neigh- bouring locality, M262, which is shown on the Bureau of Mineral Resources Mt. Isa 1:250,000 geological series sheet, F/54-1. Opik (1967, p. 150) has listed fossils from this locality, and described the species of Redlichia (in press). The cyclocystoids are dated as older than the Redlichia chinensis fauna of western Queensland, which is of latest Ordian age, but younger than that of Redlichia forresti which occurs in the Northern Territory and Western Australia earlier in the Ordian (Opik, personal communication). The late Ordian Stage of Australia correlates with the Albertella Zone of the North American Rocky Mountains, and pre-dates the paradoxidian Middle Cambrian of Europe (Opik 1967, p. 144, table 1). Following early description of Cyclocystoides as representative of a distinctive echino- derm grouping by Salter and Billings (1858), little progress has been made in describing its complete morphology as new material has been scarce. Thorough morphological analyses of the Cyclocystoidea in the light of all accumulated information, have been given recently by Kesling (1963, 1966). Until the discovery of the fossils reported here, the class was unknown outside of North America and north-western Europe, and its established geological range was from Middle Ordovician to Devonian, post-dating the appearance of most of the better known echinoderm classes. Phyletic relations of [Palaeontology, Vol. 14, Part 4, 1971, pp. 704-10, pi. 138.] R. A. HENDERSON AND J. H. SHERGOLD: CYCLOCYSTOIDES 705 TEXT-FIG. 1. Sketch map of the bedrock geology of part of the Mt. Isa 1 : 250,000 geological sheet (black on the inset map) showing the position of the cyclocystoid locality, QP243; based on air photograph 0105, Mt. Isa, run 7, and unpublished geological interpretation by F. de Keyser. 706 PALAEONTOLOGY, VOLUME 14 the group therefore require review in terms of the affinities of the species described here with contemporaneous free-living echinoderm taxa. SYSTEMATIC DESCRIPTIONS Class CYCLOCYSTOiDEA Miller and Gurley 1895 Genus cyclocystoides Salter and Billings 1858 Cyclocystoides primotica sp. nov. (Plate 138, figs. 1-3) Material. Holotype, Commonwealth Palaeontological Collection (CPC) 11395, external mould of an aboral surface. Paratype, CPC 11396, external mould of an aboral surface. Occurrence. Yelvertoft Bed, basal Beetle Creek Eormation, 3 miles WSW. of Cornford Bore, 28 miles WSW. of Mt. Isa, western Queensland, Australia. Age. Late Ordian (early Middle Cambrian). Diagnosis. Plate surfaces of the submarginal ring exposed on the aboral surface are slightly longer than wide, about twice the width of adjacent plates of the marginal ring and aboral disc. Plates of the aboral disc are arranged in a mosaic lacking a geometric pattern; they show no gradation in size from the centre to the periphery. All plates of the aboral surface bear a fine striation radiating from the disc centre and continuous to the periphery of the marginal ring. Description. Known only from aboral surfaces which are large for the genus. Submar- ginal ring entire, consisting of subequal plates, often with indistinct sutures, numbering approximately 42 and 53 on the two specimens to hand. As exposed on the aboral surface the plates are rectangular in outline and slightly longer than wide. Marginal ring well developed, entire, of at least six, possibly eight, plates in thickness from the submarginal ring to the periphery. Plates adjacent to the sub-marginal ring are largest, form a regular circlet, and are subrectangular in outline; plate size, regularity of shape, and the geometric arrangement of plates all decline towards the periphery. Plates are separated by distinct sutures and there is no suggestion of a radial arrangement. Aboral disc composed entirely of plates, as far as can be seen, although the disc centre is imperfectly preserved on both specimens. Specimen CPC 11395 suggests some inclined, central plates, perhaps abutting a central aperture, but such structure cannot be con- firmed. The remainder of the disc surface is covered in a mosaic of small irregularly polygonal plates separated by distinct interspaces and lacking geometric arrangement. Specimen CPC 11395 has the suggestion of a spiral plate arrangement but close examina- tion shows it to be more apparent than real. Plates are mainly of uniform size but some smaller elements are intercalated in a random manner; there is no gradation in plate EXPLANATION OF PLATE 138 Pigs. 1-3. Cyclocystoides primotica sp. nov. 1 , latex cast of holotype, CPC 1 1 395. Aboral surface, N 2. 2, latex cast of paratype, CPC 1 1 396. Aboral surface, x 2. 3, slab surface with paratype, CPC 1 1 396, showing the orientation of Redlichia cephala, X 1-5. Palaeontology, Vol. 14 PLATE 138 HENDERSON and SHERGOLD, Cyclocystoides ■t <♦ J V ■1 i i?i; ';oX O m •Si^ — rr, ° I- 3 CO s]i NviNOSHiit^is S3iavaan l 2 to 2 < ~ ^ Z XO>« Ity — ^ m BRARIES SMITHS0N1An'~INSTITUTI0N NOIiniliSNI NVINOSHil^ , to Z . CO z 2 xSvwoXv < S .wife, < ;S SMITHSONIAN INSTITUTION NOIiniiiSNI_NVINOSHillNS S 3 I a V a 8 IT_L I B R A R I E S SMITHSONM Ni NViNOSHims S3iavaan libraries smithsonian institution NoiiniiisNi NviNosHim z ~ ;s SMITHSONiAN“lNSTITUTION NOIiniliSNI NVINOSHilWS S3iavaai3 LIBRARIES SMITHSONI/ NI^NVIN0SHillMs"'s3 I ava a n~ LI B RAR I ES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHiU — “ I — . CO a: < 5 5 o ^ Z 3 Z -I z ES SMlTHS0NIAN_INSTITUTI0N_N0lini!iSNI_NVIN0SHillNS^S3 I a Va a 13^ L I B R AR I ES^SMITHSONL O ZSvtitiyX. CD CO ' IT| ^ xjvAst^jx m ■■ ^ rn z _ to ■ — CO NI^NVINOSHillAIS S3iavaaH LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHiU 2; CO Z , W ^ ^ Z ^ I I ^ ES SMITHSONIAN INSTITUTION NOliniliSN I _ NVINOSHimS S 3 I a V a a I3_ L I B R A R I E S SMITHSONI Mf KiwiMHCu I iiAic c T I w NJ a n cR>iiTUcnMiAM IMCTITMTION MniinillSNI NVINOSHIU yvaan libraries Smithsonian institution NoiiniiiSNi nvinoshiiims saiavaan lie ce =1 < ce ca _ O' ' TITUTION NOIiniliSNI~'NVINOSHillAIS S3 1 H VB a IT ""lI B R AR I ES^SM1THS0NIAN~INSTITUTI0N NOI z t- , z. r- z f~ z m ' ^ in ^ m cn ' — tn ~ CO BRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIWS S31BVaan LI . CO Z . CO z CO 2 z -H d wWym:, V oo z t ^ t - p 2 ~> ‘' 5 VV\osvii>^ >■ CO ■' Z CO ■’'“' 2 CO 2 OO riTUTlON NOlinillSNI NVINOSHillNS S3IHVdan LIBRARIES SMITHSONIAN INSTITUTION NOI avaan libraries smithsonian institution NoiiniiiSNi nvinoshihais saiavaan lie T- Z ^ Z o m NO'vbcj>' 'ii m CO _ CO TITUTION NOlinillSNI NVINOSHIHAIS SHIBVaS V CO z ^ ^ BRARIES SMITHSONIAN INSTITUTION avaan^LIBRARIES^^SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHillMS SBIBVaa — >s) oo r:: .. co « ^ cn 31 < a: CO TITUTION N0liniliSNl“NVIN0SHillAJS^S3 I avaail^'LlBRARI ES^SMITHSONIAN'^INSTITUTION ^ NOI 2 f— , 2 I— 2 Z CO CO O /TCTstito^x XV ^ O y^~>a V(^ O TO > ;o — m ^ rn 'tl m _ CO _ CO _ CO iz avaan libraries smithsonian institution NoiiniiiSNi nvinoshiiws saiavaan li 2, CO 2*CO 2 CO 2 >irfsihuC)\ = y, cr o WM S E z K15^ H z -I Wl^y/y,. z / \ ^ ^ C/> O ■— \vism^~^y — y 2 ■ V^' > 2 >■ 5 > 2 CO Z 00 ■•* z C/j 2 CO riTUT'-;' NOlinillSNI NVINOSHIIWS SBiavaan libraries smithsonian institution NOI * \ CO — CO — CO