het . Is Poty 3 i 5 2 3? srrstite yas 0a gh weet Yoo sae os phatitd Soop eter: ed coakee = - peatsha ee Ly te, v a ‘ BRITISH MUSEUM (NA iL HISTORY - F 21 JUN 1991 Ea Es GEE E 'TEn me a | L oe Zoology Series “VOLUME 57 NUMBER 1 30 MAY 1991 The Bulletin of the British Museum of Natural History, instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an Historical series. The Zoology Series is edited in the Museum’s Department of Zoology Keeper of Zoology: Dr C. R. Curds Editor of Bulletin: Mr N. Merrett Assistant Editor: Dr B. T. Clarke Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff and by specialists from elsewhere who make use of the Museum’s resources. Many of the papers are works of reference that will remain indispensable for years to come. A volume contains about 256 pages, made up by two numbers: published Spring and Autumn. Subscriptions may be placed for one or more of the series on an Annual basis. Individual numbers and back numbers can be purchased and a Bulletin catalogue, by series, is available. Orders and enquiries should be sent to: Sales Department, Natural History Museum Publications, British Museum (Natural History), Cromwell Road, London SW7 5BD Telephone: 071-938-9386 Fax: 071-938-8709 World List abbreviation: Bull. Br. Mus. nat. Hist. (Zool.) © British Museum (Natural History), 1991 ISBN 0 565 05049 | Zoology Series ISSN 0007 — 1498 Vol 57 No | pp 1-110 British Museum (Natural History) Cromwell Road London SW7 5BD Issued 30 May 1991 Typeset by J & L Composition, Filey, North Yorkshire Printed in Great Britain by Henry Ling Ltd, at the Dorset Press, Dorchester, Dorset Bull. Br. Mus. nat. Hist. (Zool.) 57(1) 1-16 Issued 30 May 1991 Morphology and biometry of twelve soil testate amoebae (Protozoa, Rhizopoda) from Australia, Africa, and Austria Abt ot MUSEUM (NATURAL HISTORY) , Th v 51 PRESENTED | ZOGLOGY LGRARY Corresponding author: Dr. Gabriele Liiftenegger, Universitat Salzburg, Institut fiir Zoologie, == = Hellbrunnerstrasse 34, A-5020 Salzburg, Austria Gabriele Luftenegger & Wilhelm Foissner Universitat Salzburg, Institut fiir Zoologie, Hellbrunnerstrasse 34, A-5020 Salzburg, Austri } CONTENTS 1, WRTEROG MEO ae ee aS oe a NE rk ae tn, A sk OG A OR fg 1 Pre midbemialsrand methods... SUI ta ERIE! + = Rots ak Beale Cabs, sys wx supe tae pte eds per ee eS aan a5, 1 PRD CS CUNO LES CCIOS pe cages ncaa ORNS wc «ENS ee hss le, RAD tay beeen tat a Weta en seer 1 PP CAMO COD EMCI S 8 cai Me gate Che Vo RRO RN «= > AEE Bri: ENS, SaaS) Bo cor Mn RRNA eed tax DO aaah oe US DRAG ICL Ce Sumner Shader... ig See RS » « Tepper Same Poeun eka nero Eee ten no Gekeed nee ee ha aye ibs) SYNOPSIS. The morphology of 12 species of testate amoebae from soils of Australia (Bullinularia gracilis, Centropyxis cryptostoma, Heleopera sylvatica, Assulina muscorum, Corythion asperulum, Euglypha compressa), Africa (Paraquadrula irregularis) and Austria (Cyclopyxis kahli, Difflugia stoutii, Nebela tubulata, Tracheleuglypha dentata, Trinema enchelys) were investigated by light and scanning electron microscopy. All species are char- acterized morphometrically and an ‘ideal individual’ for each species is constructed by means of the morphometric data. A first record for Austria is Difflugia stoutii, first records for Australia are Bullinularia gracilis, Centropyxis cryptostoma, Corythion asperulum and Heleopera sylvatica. In P. irregularis, numerous cells united at their apertures have been observed. Protargol silvered material convincingly shows fusion of the nuclei and nucleoli of such pairs. It is, however, uncertain whether the uniting cells are parent and daughter (autogamy) or are from 2 different specimens (conjugation). lucida using an oil immersion objective (100 x; eyepiece, 10 X) and bright field illumination. The following sample statistics were calculated according to ) Schénborn et al. (1983): X, arithmetic mean; M, median (this value is used to construct the ideal individual); SD, standard deviation; SE, standard error of the arithmetic mean; CV, | coefficient of variation in %; Min, Max, minimum and maximum values; n, sample size. Most shell variables were given a number which defines the corresponding character in the drawing of the ideal individual. The respective characters are identified by these numbers in the tables (Luftenegger et al., 1988a). For scanning electron microscopy specimens were cleaned by several transfers through distilled water before being manipulated by a single-hair brush as individuals and INTRODUCTION Many species of testate amoebae have been described and redescribed without providing reliable biometric data. The present paper is the second (Liftenegger et al., 1988a) of a series of publications intended to improve the diagnosis of such species. This hopefully will also help ecologists with the often difficult species determinations. MATERIALS AND METHODS For the sources of material see Table 1. Testate amoebae were isolated with a pipette as individuals from soil suspensions (0.2 g fresh soil + 5 ml distilled water) and stained either with protargol silver or phenolic aniline blue solution (Foissner, 1983; Liftenegger et al., 1988b). The shells of some species were made transparent overnight in a drop of albumin- glycerin (as used in histological techniques to fix sections on a slide). All drawings were produced with the help of a camera mounted on glass slides covered with a special adhesive (Mixtion a Dorer Clarifeé, Fa. Lefranc & Bourgeois). Further procedures see Liftenegger et al. (19882). Determinations of species follow the original descriptions. One slide of each population has been deposited in the British Museum (Natural History) in London. Registration number is given in the species description. Several species have the same Reg. No since they are on the same slide. 5 fable 1. Localities of the populations. Species Date Locality Reference number 23.10.1986 Bush in the Brisbane Water National Park, 50 km north of Sydney, Australia. E 152° S 33°. Sea-level 150 m. Upper soil layer (0-2 cm soil depth) with litter and moss on sandstone. pH 5.1. 23.10.1986 See Bullinularia gracilis. 10.10.1987 Aiglern near Aigen in the Ennstal, Austria, E 14° N 48°. Sea-level 650 m. Brown earth (loamy sand) of a meadow (0-5 cm). pH 6.7. (For detailed site description see Foissner et al., 1990). 10.10.1987 See Cyclopyxis kahli. 18.10.1986 Coastal wood, Royal National Park, south of Sydney, Australia, E 150°S 35°. Sea-level c. 100m. Upper soil layer (0-5 cm) with litter and brown sand. pH 4.5. 20.9.1988 Top of the Gaisberg, Salzburg, Austria, E 13° N 48°. Sea-level 1250 m. Rendzina of a spruce forest with Luzula, Calamagrostis, Sesleria (0-5 cm). pH 5.2. Mzima Springs, Tsavo National Park West, about 50 km east of Mt. Kilimanjaro, Kenya, E 38° S 4°. Humus layer with litter, many fungi, under acacia (0-5 cm). pH 7.0. 23.10.1986 See Bullinularia gracilis. 23.10.1986 See Bullinularia gracilis. 23.10.1986 See Bullinularia gracilis 10.10.1987 See Cyclopyxis kahli. 10.10.1987 See Cyclopyxis kahli. Bullinularia gracilis Centropyxis cryptostoma Cyclopyxis kahli Difflugia stoutii Heleopera sylvatica Nebela tubulata Paraquadrula irregularis 8.5.1985 Assulina muscorum Corythion asperulum Euglypha compressa Tracheleuglypha dentata Trinema enchelys DESCRIPTION OF SPECIES Bullinularia gracilis Thomas, 1959 Figs 1-8, Tables 1,2 BM (NH) Reg. No. 1990. 4. 26.1. Shell yellowish to brownish, almost hemispheric, always broader than long, ventral side smooth, structureless, covered by thin, fragile layer (Fig 7), posterior margin and dorsal side rough, covered with xenosomes (Figs 4,5). Aperture obscured by anterior lip, under which ventral side projects (Fig. 6). Approximately 30 pores, about 1—2 ym in diameter, irregularily arranged in the anterior third of shell (Fig. 8). Characters (1)-(3) show normal variability (CV 14.0 and 12.3%), whereas characters (4) and (5) have high coefficients of variation (CV 20.0 and 30.0 %; Table 2). Thomas (1959) states 120 um on average for the major axis of the shell. The specimen photographed by Bonuet (1961) shows a similar GABRIELE LUFTENEGGER & WILHELM FOISSNER Table 2. Morphometric characterization of Bullinularia gracilis. All measurements in pm. Character x M SD SE CV Min Max n (1) 93.2). A885." 13:0, 0 '5.33q, 14.02. 83 tee 46 (2) 130.3 125.0 182° 4:70 140° 112 tee 15 (3) 150.9 1500 185° 4.78 123 108 foams 15 (4) STi com $400 eA 2046.00 nena iaeee eel (5) 1,5), 10.0. 3:4) (0:89, 30:0: «Suenos size, however, detailed biometric data are not provided. Regarding this character, the values of our population are about 25 % higher, which is in agreement with Golemansky (1968). Hoogenraad & De Groot (1952b) describe two shell types for B. indica. The ‘extraordinary’ type is firm and inflexible, the ‘normal’ type is flexible and rather tenacious, consisting of a homogenous matrix, as stated also by Penard (1912). We suggest that the outer organic layer, which seems to be somewhat elastic (Fig. 7), is responsible for the above mentioned flexibility of shells in some Bullinularia populations. Centroypyxis cryptostoma Bonnet, 1959 Figs 9-13, Tables 1,3 BM (NH) Reg. No. 1990. 4. 26.2. Shell brownish, rectangular with rounded ends in ventral view, compressed, ventral side fairly smooth, with flat xeno- somes, dorsal side with rough particles. Aperture sub-apical, reniform. Posterior lip extends slightly inside shell as curved elongation of ventral side, anterior lip overhanging (Figs 10,12). Separation from C. capucina and C. halophila is mainly by smaller size and occurrence in different habitats (pH optimum for C. halophila 8.5-9; Bonnet & Thomas, 1960a). Coefficients of variation of characters (1)-(4) and (7) are less than 10 %; measurements of aperture show greater variability (CV 12.5 and 14.9; Table 3). Bonnet (1959) states 45 um for the length, 35 um for the breadth and 27-30 um for the depth of the shell. The individuals of our population are a little more flattened, which is in agreement with the measure- ments by Schénborn (1966). All other characters match well with the description by Bonnet (1959). Cyclopyxis kahli (Deflandre, 1929) Deflandre, 1929 Figs 14-16, Tables 1,4 BM (NH) Reg. No. 1990. 4.26.3 Shell brownish, hemispheric, composed of xenosomes, apertural surface smooth and distinctly invaginated, aboral surface rough. Aperture surrounded by distinct rough particles (10-20 % of specimens studied), or more or less smooth (c. 80-90 %). Shell measurements fairly variable (CV 10.6-16.0; Table 4). The data of Deflandre (1929) and Ogden (1988) agree well with ours, whereas other authors state slightly higher values (Bonnet & Thomas, 1960a; Ogden & Hedley, 1980; Ogden, 1984; Rauenbusch, 1987). The Indian population described by Guru & Dash (1983) measures only 50-65 pm in diameter (sample size not given). TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA Figs 1-8 Bullinularia gracilis, light microscopic (Fig. 1) and SEM-aspects (Figs 3-8) and ideal individual (Fig. 2) . 1 Ventral view. 2 Lateral and ventral view. 3 Ventral view, x 400. 4 Lateral view, x 500. 5 Detail of rough dorsal shell surface, x 1500. 6 Aperture, x 1200. 7 Detail of smooth ventral shell surface, x 3700. 8 Pore, x 12000. Scale bar divisions 10 pm. Table 3. Morphometric characterization of Centropyxis cryptostoma. All measurements in um. Character x M SD SE CV Min Max n 1 46.1 460 2.0 064 44 42 48 10 2 S24 S20 26" (082-0 28 35 10 3 25:2 ee 06522828" 20. 27 10 4 14 G:Ome 4 10545, "9.3: 1.13 18 10 5 6.4 6 = 09030) 1429 5) 8 10 6 5.6 CO OFF 022" 25 4 6 10 i re ISO” 14a 045 9.7 13 19 10 Table 4. Morphometric characterization of Cyclopyxis kahli. All measurements in pm. Character x 1 44.3 Dp, 70.1 3 18.0 4 11.8 SD ie Oana SE CV Min Max 1.16 1.41 0.51 0.47 1 1 1 1 NWon lente oo wo, 35 58 13 10 n ae —_ . eS ee A ~ Z ”, <4 Gas ee ea ya ye mi te Ait: iy Bay = i WP, 5 ly! Dear isin IN ge. ESS, at “. =< f e Ae Brg S| Cy 5 AY wnt. a ey Figs 9-13. Centropyxis cryptostoma, light microscopic (Figs 9,10) and SEM-aspects (Figs 12,13) and ideal individual (Fig. 11). 9-11 Ventral and lateral views. 12,13 Ventro-lateral and ventral view, < 1900, x 1600. Scale bar divisions 10 pm. mie ie Figs 14-16 Cyclopyxis kahli, light microscopic (Fig. 14) and SEM-aspects (Fig. 16) and ideal individual (Fig. 15). 14 Ventral view. 15 Ventral and lateral view. 16 Ventral view, X 1100. Scale bar divisions 10 um. TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA Figs 17-23 Difflugia stoutii, light microscopic (Figs 17-19) and SEM-aspects (Figs 21-23) and ideal individual (Fig. 20). 17-20 Broad lateral, narrow lateral and ventral views. 21 Ventral view, X 1200. 22,23 Broad and narrow lateral view, x 1600, x 1200. Arrow marks aperture. Scale bar divisions 10 pm. Table 5. Morphometric characterization of Difflugia stoutii. All measurements in um. Character x M SD SE CV (1) SO 50.0 7 21360 14-0. 38. 61 27 (2) Bicone SOR S130) 07 17.0" 22-40" 24 (3) PONE T=20L0- 4 B.0\0.98.- 15 16°. 25 15 (4) RiGee S Sienlebee 0.3919 6:40 16 (5) Sete aie See O20BS 19iBe Sa, 6) goud Bonnet & Thomas (19605) described a variety, cyclostoma, lacking the rim of rough particles around the aperture. We suggest that such elements are easily lost, since transitions in the formation or reduction of the apertural rim are common (see above and Cotitteaux, 1976, Fig. 1J; Ogden & Hedley, 1980, Pl. 24; Rauenbusch, 1987 Pl. 16, Fig. a). At the present state of knowledge it seems wise to classify our population as C. kahli. For detailed ecological data see Bonnet (1989a). Table 6. Morphometric characterization of Heleopera sylvatica. All measurements in pm. Character x M SD SE CV Min Max n (1) 6595 “G500% "S10 10:62 was 159 70 25 (2) 440 440 22 046 50 39 48 23 (3) Dees Oe 2S (O-AT 9 if: 9re 25 33 23 (4) 3.4 S10 0.98 OM9. 26.9 2 5 23 Difflugia stoutii Ogden, 1983 Figs 17—23, Tables 1,5 BM (NH) Reg. No. 1990. 4.26.4 Shell brownish, elongate ellipsoid, slightly flattened, com- posed of overlapping, flat xenosomes, very fragile, thus shells often collapse when air-dried (Fig. 21). Aperture elliptic. Shell measurements, especially characters (2) and (4), show considerable variability (Table 5). Ogden (1983), who Sansviee’: aes eae GABRIELE LUFTENEGGER & WILHELM FOISSNER Figs 24-32 Heleopera sylvatica, light microscopic (Figs 24,25) and SEM-aspects (Figs 27-32) and ideal individual (Fig. 26). 24, 25 Broad and narrow lateral view. 26 Broad lateral view. 27,28 Broad lateral views, X 1000, x 1200. 29 Detail of surface, x 2100. Note euglyphid apertural platelet (arrow). 30 Aperture, x 1200. 31,32 Details of surfaces, x 1800, x 1200. Scale bar divisions 10 um. discovered this species in a Sphagnum sample from North Wales (England), states the following values (n = 4): 47-59 um length, 33-36 um breadth, 9-12 pm diameter of aperture. Our measurements agree well with those data (Table 5). However, our population is slightly compressed and has an elliptic aperture, possibly caused by living in the soil. As far as we know, this is the first record since the original description. Heleopera sylvatica Penard, 1890 Figs 24~-32, Tables 1,6 BM (NH) Reg. No. 1990.4.26.5 Shell slightly yellowish, transparent, obovoid, flattened about 2:1, composed of siliceous shell platelets from other testaceans and sometimes rough xenosomes, found mainly in aboral region. Some individuals are completely covered with such acquired idiosomes (Fig. 28). The aboral region often has euglyphid platelets (Euglypha, Trinema) extending at right-angles like spines (Figs 28,29,31,32). Aperture terminal, slightly convex in broad view, slit-like to elliptic with small border of organic cement (Fig. 30). Nucleus with several nucleoli (protargol impregnation). Shell measurements with exception of character (4) are fairly constant (Table 6) and agree well with the data of Penard (1890). Cash & Hopkinson (1909) state a smaller shell breadth (25-30 ym). TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA 7 1 Figs 33-39 Nebela tubulata, light microscopic (Figs 33-35) and SEM-aspects (Figs 37-39) and ideal individual (Fig. 36). 33-35 Broad lateral, narrow lateral and ventral view. 36 Broad and narrow lateral view. 37,38 Broad and narrow lateral view, x 1100, x 1200. 39 Aperture, x 2500. Scale bar divisions 10 pm. The drawings in Penard (1890, Pl. VIII, Figs 79,80,84,86,87) clearly show individuals having peculiar spine-like structures in the aboral region and are even mentioned in his descrip- tion. Hoogenraad & De Groot (1940) also state ‘eigentiim- liche dornartige Fortsatze’ for H. petricola, and Penard (1890) assumes very narrow idiosomes erected on the shell surface for H. picta. This suggestion is now confirmed by means of scanning electron microscopy. No other study describing such platelets is known to us. The paper by Cotiteaux & Munsch (1978, PI. III, Fig. 3), however, includes a scanning electron micrograph of a specimen designated as Placocista lens. We suggest that it is a Heleopera with spine- like siliceous platelets. (Unlike to Heleopera, the genus Placocista is an euglyphid taxon, having filose pseudopodia and shells with regularly arranged, self-made platelets). Presumably, all species of the genus Heleopera have the ability to erect platelets. For detailed ecological data see Bonnet (19895). Nebela tubulata Brown, 1910 Figs 33-39, Tables 1,7 BM (NH) Reg. No. 1990.4.26.6 Shell colourless, flask-like with distinctly separated, parallel- sided neck, slightly compressed, fragile, composed of different sized circular and elliptic platelets from other Table 7. Morphometric characterization of Nebela tubulata. All measurements in pm. Character x M SD SE CV Min Max n (1) 60:62 2610" = 20) 0.58 35; 57 rede i6 (2) 214 2900'S = 0.60) BO. 28) —Sa8eees (3) 292, 920, SOB AAT 1 200) 68 (4) 1430 140) 9102 8024 62 13) 16m) 13 (5) Sie 50 NOSE OOS sts. 455 oes 11 (6) S310 MESA 2.400 0.65 7-029. © 37> iB (7) 2659 22610. 2108" 0555: 7.5" 22), 29. = 13 (8) as iaOe 16, 043 107° 12 17 43 (9) 0.6, MOO 9 15° St 15.8 9 12 7 testaceans (Fig. 33). Aperture terminal, elongate elliptic, concave in lateral view, surrounded by organic collar (Fig. 39) Shell measurements with exception of characters (5) and (9) are fairly constant (Table 7). Character (1) has lowest variability (CV 3.5). Our values agree well with those of Brown (1910), Wailes & Penard (1911), Deflandre (1936), Hoogenraad & De Groot (1940, 1952a), Gauthier-Liévre (1957) and Ogden & Hedley (1980). Nebela tubulata is separated from N. lageniformis by its smaller size (N. lageniformis always larger than 90 pm), and from N. militaris by the abrupt narrowing of the shell gn yy 50° Figs 40-50 Paraquadrula irregularis, light microscopic aspects of living (Figs 40-42) and protargol silver impregnated specimens (Figs 44-50), and ideal individual (Fig. 43). 40 Broad lateral view. 41,42 Light microscopic photographs of specimens united at their apertures, both x 1500. Note cyst in Fig. 42. 43 Broad lateral, narrow lateral and ventral view. 44 Specimen showing 1 pseudopodium. Note nucleus with central nucleolus (arrow). 45 Dividing cell. 46 Parent and daughter cell. Note microfilaments stretching between cells (arrows). 47 Fusion of 2 cells. 48 Nuclear fusion. 49 Early stage of encystation; nucleus contains 2 nucleoli. 50 Resting cyst; nucleus contains 1 nucleolus. Figs 44-50 x 1000. Scale bar divisions 10 ym. TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA 9 Table 8. Morphometric characterization of Paraquadrula irregularis. All measurements in pm. Character x M SD SE CV Min Max 0n (1) 2935 290047 74 26 32 Di (2) 26.6 26.0 1.9 0.41 7.0 24 30 21 (3) 19.3 19.0 1.2 0.30 6.2 16 21 16 (4) 10.3 10.0 1.5 0.36 14.8 8 13 18 (5) 3294 One ele0) OZ 24.95 3 6 13 Platclessencth .. 5.9 - 6:0 —0.9° 0:19 15.1, 4.5 8 21 towards the parallel-sided neck. Lateral pores as found by Hoogenraad & De Groot (1940) in some individuals did not occur in our population. Paraquadrula irregularis (Archer, 1877) Deflandre, 1932 Figs 40-50, Tables 1,8 BM (NH) Reg. No. 1990.4.26.7 Table 9. Morphometric characterization of Assulina muscorum. All measurements in ppm. Character x M SD SE CV Min Max n (1) 43:0. 49.0 Haest0 lI) 1160838 7 Sane 20 (2) apiaulrsihO MS 13 1516-26 4s 20 (3) [SS SAOOMMNL 204405. wom 2 220 16 (4) Se5 phat SOMO 044 03.0) soe 18 a0) (5) Acs: _.5:0SemOsG 6 Olgas. sae 16. iG (6) 3M 3.0.50 0.9) 90:22,90101 12, 0 5 1G shell colourless, transparent, circular in broad view, slightly flattened laterally, composed of more or less quadratic calcareous platelets, sometimes irregularly arranged (Figs 40,42). Aperture elliptic. Nucleus with central nucleolus. Separation from the very similar P. discoides by lesser flattening in P. irregularis. Characters (1)-(3) fairly constant, while measurements of Figs 51-56 Assulina muscorum, light microscopic (Figs 51,52) and SEM-aspects (Figs 54-56) and ideal individual (Fig. 53). 51 Broad lateral view of atypically broad specimen. 52 Broad lateral view of typic specimen. 53 Broad lateral, narrow lateral and ventral view. 54,55 Broad lateral and ventral view, x 2100, x 1700. 56 Aperture showing organic cement, x 3400. Scale bar divisions 10 ym. 10 aperture and platelets show greater variability (Table 8). Cash & Hopkinson (1909) describe a maximum shell length of 30-38 um, Grospietsch (1954) about 45 wm, Bonnet & Thomas (1960a) 30-35 um and Decloitre (1961) 30-40 um. The shell measurements of our population deviate slightly, being smaller (Table 8). Strikingly often 2 differently sized individuals can be found united at their apertures. In most cases the larger of the two specimens contains a cyst (Fig. 43) which has, especially when older, a jagged cyst wall. This phenomenon, which is frequent among all members of the genus Paraquadrula, has been reported by many authors (Penard, 1902; Deflandre, 1932; Gauthier-Liévre, 1953; Sch6nborn, 1965) and might be inter- preted as a form of sexuality (see Figs 45-50). Although sexual processes have been described for some testacean species, they are poorly documented. Some authors reported conjugation, followed by encystation, but did not observe nuclear fusion (Penard, 1902; Awerintzew, 1907; Cash & Hopkinson, 1909; Pateff, 1926; Deflandre, 1932; Chardez, 1960), e.g. the cysts contained 2 nuclei. It is evident from protargol silver impregnated specimens of P. irregularis that nuclear fusion takes place. Figures 45—S0 document the following steps: A cell divides into 2 daughter cells (Fig. 45; this stage occurred in 5.5% of 200 investigated specimes = 100%), each having a nucleus with central nucleolus (Fig. 46; 11%); the plasma of 2 cells fuse (Fig. 47; 12%); the 2 nuclei fuse (Figs 48,49; 14.5%), followed by the fusion of their nucleoli (Fig. 50; 17%); a cyst is formed (Figs 42,50). Without doubt, the processes photographed in Figs 47-50 show fusions of 2 cells and not binary fissions, since the shells of the united specimens are complete, and even food residues of the former cells are easily to be seen in the abandoned shells (cp. Fig. 45, representing binary fission). Unfortunately, we could not follow these processes in living specimens; thus it is unclear, whether the uniting cells are parent and daughter or are from 2 different specimens, which would be a true conjugation. The former suggestion is confirmed by Schénborn (1965), who documented a reunion of parent and daughter cells without nuclear fusion, speculat- ing that this phenomenon is caused by formation of an undersized daughter test. However, we also found cysts in smaller or equal sized shells. Such cases have been attributed to plasmogamy by Schénborn (1965). On the other hand, Pateff (1926) and Deflandre (1932) described copulation of strikingly different sized specimens in Difflugia mammillaris and in P. irregularis (without nuclear fusion), and Penard (1902) reported for Cryptodifflugia oviformis that cysts are always formed in the smaller of the two tests. Valkanov (19625) has observed copulation of equal sized specimens in P. madarica, followed by nuclear fusion. He also provided photographic evidence of a syncaryon in Euglyphella delicatula (Valkanov 1962a, Abb. 1). Our figures 45-50 are, with exception of Valkanov’s picture, the first photographic document of nuclear fusion in testate amoebae. Further investigations are needed to determine, whether autogamy takes place or not, and to follow the fate of the syncaryotic cysts. For detailed ecological data of this species see Bonnet (1989b). Assulina muscorum Greeff, 1888 Figs 51-56, Tables 1,9 BM (NH) Reg. No. 1990.4.26.8 GABRIELE LUFTENEGGER & WILHELM FOISSNER Young shells yellowish, older ones light to dark brown, compressed, composed of elliptic, sometimes irregularly arranged platelets. Aperture terminal, elliptic, with more or less pronounced collar of organic cement (Fig. 56; cp. Ogden & Hedley, 1980; Ogden, 1981, 1984). Collar distinctly lobed, contrary to A. collaris, Kufferath, 1932, which is, in our opinion, a doubtful species—but see Schénborn & Peschke (1988). Parameters (1) and (3) fairly constant (CV 8.3 and 8.5), variability of characters (4) and (5) relatively high (CV 17.5 and 23.3; Table 9). The data of Cash et al. (1915) correspond with ours, while those of Ogden & Hedley (1980) and Ogden (1984) are moderately higher. Hoogenraad & De Groot (1937) studied different populations of A. muscorum, of which the so-called ‘Middel’-population (Fig. 4 of Hoogenraad & De Groot, 1937) matches our population well. This is also true for the Thuringian population analyzed biometrically by Sch6nborn & Peschke (1988), despite the great geographic distance! Even the coefficient of variation of each single character coincides strikingly well. Corythion asperulum Schonborn, 1988 in Schénborn & Peschke, 1988 Figs 57-62, Tables 1,10 BM (NH) Reg. No. 1990.4.26.2. Shell colourless, ovoid, flattened, composed of irregularly arranged elliptic shell platelets (Fig. 57). Numerous about 3 um long, siliceous, ‘flame-like’ spines projecting from junctions of idiosomes over entire shell except in apertural region (Figs 60,61). Organic cement plentiful, can be seen as small border surrounding each platelet (Fig. 62). Separation from the very similar C. dubium var. spicatum by means of the spines, which are longer, in 1 single row and consist of chitin in C. dubium var. spicatum. Table 10. Morphometric characterization of Corythion asperulum. All measurements in pm. Character x M SD SE CV Min Max n (1) 44.7 45.0. 3:9 1.03 -8570 S25 eas 14 (2) 32:7 32:0, “3532 0L92 10 es 36 18) (3) 18:9 19:0' . 3.0% “"OlS7 “16:07 ans ‘ap 12 (4) 14.0 _ 14:0/~ 2:3° (0:63) 1622 8 16 3) (5) 10.6 10.0 19” Orsi 7s 6 13 13 (6) SV S20) 10:9 0:24 278 2 5 13 (7) 3.0 3.0 0-7 “O19 32777 2 4 13 Table 11. Morphometric characterization of Euglypha compressa. All measurements in pm. Character x M SD SE CV Min Max n (1) 87.4 88.5 10.8 2.42 12.4 65 115 20 (2) 61.2 61.0 7.0 1.56 11.4 50 76 #820 (3) 32.4 33.0 2.6 0.97 7.9 29 35 7 (4) NOD OYA) 3.4 0.76 14.1 18 3220 (5) 14.6 14.0 2.8 1.07 19.4 10 19 7 (6) 8.0 8.0 0.9 0.23 11.8 6.5 10 + 16 (7) 22 2250 3.3 0.78 15.0 16 29 18 (8) 8.1 8.0 15 0.36 18.3 6 11 17 (9) 4.6 4.0 I-25 e0: 29 26a 3 7 17 TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA 11 Figs 57-62 Corythion asperulum, light microscopic (Fig. 57) and SEM-aspects (Figs 59-62) and ideal individual (Fig. 58). 57 Ventral view. 58 Ventral and lateral view. 59,60 Dorsal and ventral view, x 1200, x 1700. 61 Spines, x 12300. 62 Detail of surface, x 3200. Note organic cement surrounding platelets (arrows). Scale bar divisions 10 pm. Shell measurements show rather high variability; only parameter (1) has a CV less than 10% (Table 10). The biometric data of Schénborn & Peschke (1988) correspond almost perfectly to our own, despite the extreme geographic distance. Only character (3) shows a slightly higher variability in our population. Schénborn & Peschke (1988) did not mention the organic cement surrounding the platelets (Fig. 62). As far as we know, this is the the first record since the original description, and especially remarkable, being from Australia! Euglypha compressa Carter, 1864 Figs 63-68, Tables 1,11 BM (NH) Reg. No. 1990.4.26.1 Shell colourless, obovoid, compressed 2:1, composed of about 300 platelets (often with notched narrow side; Fig. 68). Several 16-19 um long spines projecting singly from junctions of shell platelets close to lateral margins (Fig. 66). Aperture elliptic, about 15 apertural platelets. Platelets thickened at denticulate margins, carrying 1 large median tooth. Presum- ably always 3 pairs of lateral teeth, of which only 2 are easy to see; third is bent inward at right-angles to apertural opening, very small and thus difficult to recognize (Fig. 67). Shell measurements moderately varying (Table 11), cor- responding well with those of Wailes & Penard (1911), Cash et al. (1915) and Ogden & Hedley (1980). The values of Carter (1864) are slightly higher, those of Ogden (1981) lower (61-75 um length; n = 31). The peculiar notching of the idiosomes, which give the platelets a hexagonal appearance (Penard, 1902; Cash et al., 1915) is also evident in the scanning electron micrographs of Ogden & Hedley (1980), Ogden (1981) and Rauenbusch (1987, for E. compressa var. glabra). The drawings in Cash et al. (1915) show 5 different types of spines. Ogden (1981) also emphasizes that 2 types exist within this species. The spines of 12 GABRIELE LUFTENEGGER & WILHELM FOISSNER Figs 63-68 Euglypha compressa, light microscopic (Fig. 63) and SEM-aspects (Figs 65-68) and ideal individual (Fig. 64). 63 Broad lateral view. 64 Broad lateral, narrow lateral and ventral view. 65Broad lateral view, x 800. 66Detail of surface showing spines projecting from junctions of platelets, x 2500. Note points where spines are lost (arrows). 67 Apertural plates, x 5800. Third pair of teeth are hardly recognizable (arrows). 68 Shell plates, x 4000. Note notching of platelets (arrows). Scale bar divisions 10 pm. TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA 13 Figs 69-75 Tracheleuglypha dentata, light microscopic (Figs 69,70) and SEM-aspects (Figs 72-75) and ideal individual (Fig. 71). 69-71 Lateral and ventral views. 72-74 Apertures with differently shaped collars, x 3800, x 3900, x 4200. 75 Ventro-lateral view of specimen without distinct collar, X 1500. Scale bar divisions 10um. our population represent the type photographed in Pl. 78 of Ogden & Hedley (1980). Presumably such populations should be separated at species level. Tracheleuglypha dentata (Penard, 1890) Deflandre, 1928 Figs 69-75, Tables 1,12 BM (NH) Reg. No. 1990.4.26.3 Shell colourless, obovoid, circular in transverse section, composed of about 100 circular, regularly overlapping plate- lets, usually about 7 wm in diameter (often smaller in apertural region). Circular aperture terminal, usually, but not always, surrounded by chitinous collar (Figs 72-75). Coefficients of variation are between 11.3 and 17.5%. Character (3) and diameter of shell platelets show the greatest variability (Table 12). Our measurements correlate well with those of Penard (1890, 1902), Cash et al. (1915), Thomas & Gauthier-Liévre (1959) and Rauenbusch (1987), whereas the population of Ogden & Hedley (1980) is slightly larger. A comparison of our values with the average values (1 free-living and 3 cultivated populations combined) of Ogden & Cotiteaux (1987) shows a high conformity, even in standard deviations. The measurements for T. acolla, given by Bonnet & Thomas (1955), also agree with ours. As already mentioned, specimens of our population may or may not have a collar. Traditionally, individuals without collar are considered as a separate species, T. acolla (Bonnet & Thomas, 1955). The transitions between individuals with and without collar are manifold and have indeed been docu- mented by numerous authors using scanning electron micro- scopy. Thus, Fig. 6in Ogden & Cotiteaux (1987), Pl. 34 Fig. a in Rauenbusch (1987) or Pl. 90 Fig. B in Ogden & Hedley (1980), which are described as T. dentata, correspond with the scanning electron micrographs of 7. acolla in Bonnet (1975), Grospietsch (1982) and Chardez & Rassel (1985). Ogden & Coititeaux (1987, 1988), however, suggest that the collar serves in holding parent and daughter cells together during division, and may be absent, especially in empty shells taken from field samples (presumably by natural causes such as predation or influence of bacteria). The larger shell platelets of 7. acolla are considered an additional criterium in separating it from 7. dentata (Bonnet & Thomas, 1955). However, Chardez (1960) found 4 different types of platelets in various populations of T. acolla. Our population matches the type with large (about 9 pm) circular platelets (Fig. 75). Regarding the drawings of Bonnet & Thomas (1960a), which unfortunately do not give measure- ments for the diameters of platelets, it is evident that the platelets of their species designated as T. acolla have a maximum diameter of 5 pm. This is even smaller than in our population. These data strongly suggest synonymy of T. dentata and T. acolla as already indicated by Ogden & Cotiteaux (1987). Trinema enchelys (Ehrenberg, 1838) Leidy, 1878 Figs 76-83, Tables 1,13 BM (NH) Reg. No. 1990.4.26.4 Shell colourless, elliptic, almost circular in transverse section, composed of about 60 circular, scarcely overlapping large platelets. Many smaller, different sized platelets fill space between large ones (Fig. 83). Aperture circular, invaginated, surrounded by 2 rows of small idiosomes and about 30 14 LS ao! Bax GABRIELE LUFTENEGGER & WILHELM FOISSNER Figs 76-83 Trinema enchelys, light microscopic (Figs 76,77) and SEM-aspects (Figs 79-83) and ideal individual (Fig. 78). 76-78 Ventral and lateral views. 79,80 Ventral and lateral view, x 1400, x 1000. 81 Apertural platelets (arrows), " 13r 036 “15:7 I 11 14 TWELVE SOIL TESTATE AMOEBAE FROM AUSTRALIA, AFRICA, AND AUSTRIA 15 10%, others have greater variability (Table 13). The popu- lations investigated by Liiftenegger et al. (1988a) are considerably smaller (< 50 pm), but the coefficients of variation—especially those of the (PII)—agree quite well with the new data (Table 13). The high interpopulation variability of T. enchelys (Chardez, 1956) is evident from data in the literature, ranging all the way from 40 um to 140 um (Hoogenraad & De Groot, 1940; Rauenbusch, 1987). How- ever, its usual size is 50-60 um, which agrees well with our own data and with those of Ogden & Hedley (1980). Shell shape and shape and arrangement of platelets are very similar in all populations investigated with the scanning electron microscope (Ogden & Hedley, 1980; Rauenbusch, 1987; Figs 79-83). ACKNOWLEDGEMENTS. The authors wish to thank Univ-Prof Dr Hans Adam for institutional help. The photographic assistance of Mrs Karin Bernatzky and Mr Rudolf Hametner is greatly acknowledged. This study was supported by the ‘Fonds zur Foérderung der _ wissenschaftlichen Forschung, Projekt Nr P 5889’ and the Australian Biological Resources Study. REFERENCES Awerintzew, W. 1907. Uber die Si®wasserprotozoen der Insel Waigatsch. Zoologischer Anzeiger 31; 306-312. Archer, W. 1877. Occurrence in Scotland of Euglypha tincta, Quadrula irregularis, and Ditrema flavum, Archer. Quarterly journal of micro- scopical science 17: 103-104. Bonnet, L. 1959. Nouveaux thécamoebiens du sol. Bulletin de la Société d Histoire Naturelle de Toulouse 94: 177-188. 1961. L’emission pseudopodique, chez les thecamoebiens endogés (I). Bulletin de la Société Zoologique de France 86: 17-28. 1975. Types morphologiques, écologie et évolution de la théque chez les thécamoebiens. Protistologica 11: 363-378. — 1989a. Données écologiques sur quelques Centropyxidae (Thécamoebiens) des sols. Bulletin de la Société d'Histoire Naturelle de Toulouse 125: 7-16. — 1989b. Données écologiques sur quelques Hyalospheniidae et Paraquadrulidae (Thécamoebiens) des sols (Premiére partie). Bulletin de la Société d’ Histoire Naturelle de Toulouse 125: 17-22. — & Thomas, R. 1955. Etude sur les thécamoebiens du sol (I). Bulletin de la Société d’ Histoire Naturelle de Toulouse 90: 411-428. 1960a. Thécamoebiens du sol. Faune terrestre et d’eau douce des Pyrénées-Orientales 5: 1-103. — 1960b. Etude sur les thécamoebiens du sol (II). Bulletin de la Société a Histoire Naturelle de Toulouse 95: 339-349. Brown, J. M. 1910. Freshwater rhizopods from the English lake district. The Journal of the Linnean Society of London. Zoology 30: 360-368 + PI. 50. Carter, H. J. 1864. On freshwater Rhizopoda of England and India; with illustrations. The Annals and magazine of natural history, zoology, botany and geology 13: 18-39 + PI. I, Il. Cash, J. & Hopkinson, J. 1909. The British Freshwater Rhizopoda and Heliozoa. Vol. I Rhizopoda part Il. XVIII + 166p + Pl. XVI-XXXII. Ray Society, London. , Wailes, G. H. & Hopkinson, J. 1915. The British Freshwater Rhizopoda and Heliozoa. Vol. Ill Rhizopoda, part Ill. XXIV + 156p + Pl. XXXIII- LVII. Ray Society, London. Chardez, D. 1956. Variations morphologiques et teratologie chez quelques rhizopodes testaces. Biologisch Jaarboek Dodonaea 23: 265-276. & Rassel, A. 1985. Etude en microscopie electronique a balayage de la theque de quelques thecamoebiens en provenance de trois biotopes d’un etang. Acta Protozoologica 24: 211-215 + PII. Coiiteaux, M.-M. 1976. Etude quantitative des thécamoebiens d’une savane a Hyparrhenia a Lamto (Cote-d’Ivoire). Protistologica 12: 563-570. & Munsch, A. 1978. Thécamoebiens de mangroves. Revue d’écologie et de biologie du sol 15: 391-399. 1960. Thécamoebien des terres de Belgique II. Hydrobiologia 16: 197-202. Decloitre, L. 1961. Le genre Paraquadrula (Thekamoebina). Internationale Revue der gesamten Hydrobiologie 46: 321-330. Deflandre, G. 1928. Deux genres nouveaux de rhizopodes testacés. Annales de Protistologie 1: 37-43. 1929. Le genre Centropyxis Stein. Archiv fiir Protistenkunde 67: 322-375. 1932. Paraquadrula nov. gen. irregularis (Archer). Conjugaison et enkystement. Comptes rendus des séances de la Société de Biologie 109: 1346-1347. — 1936. Etude monographique sur le genre Nebela Leidy (Rhizopoda— Testacea). Annales de Protistologie 5: 201-286 + Pl. X-XXVII. Ehrenberg, C. G. 1838. Die Infusionsthierchen als Vollkommene Organismen. 547p + Tafel I-LXIV. L. Voss, Leipzig. Foissner, W. 1983. Protagol silver staining: a useful method in testacean taxonomy at the light microscopical level. JI. Int. Workshop on Taxonomy and Ecology of Testate Amoebae, Aachen (Abstract). , Buchgraber, K. & Berger, H. 1990. Bodenfauna, Vegetation und Ertrag bei Okologisch und konventionell bewirtschaftetem Griinland: Eine Feldstudie mit randomisierten Blocken. Mitteilungen der Osterreichischen Bodenkundlichen Gesellschaft 41: 5-33. Gauthier-Liévre, L. 1953. Les genres Nebela, Paraquadrula et Pseudonebela (Rhizopodes testacés) en Afrique. Bulletin de la Société d Histoire Naturelle de l Afrique du Nord 44: 324-366. 1957. Additions aux Nebela d’Afrique. Bulletin de la Société d Histoire Naturelle de l Afrique du Nord 48: 494-523. Golemansky,V. 1968. Nouvelle contribution a la connaissance des théca- moebiens (Rhizopoda, Testacea) en Yougoslavie. Bulletin de l'Institut de Zoologie et Musée 27: 57-63. : Greeff, R. 1888. Land-Protozoen. Sitzungsberichte der Gesellschaft zur Beférderung der Gesamten Naturwissenschaften 3: 90-158. Grospietsch, T. 1954. Studien tber die Rhizopodenfauna von Schwedisch- Lappland. Archiv ftir Hydrobiologie 49: 546-580. 1982. Untersuchungen tiber die Thekamobenfauna (Rhizopoda testacea) im Murnauer Moos. Entomofauna, Linz, Suppl. 1: 57-88. Guru, C. & Dash, M. C. 1983. A taxonomic study of Testacea (Protozoa) in some soils of Orissa, India. Indian Biologist 15: 11-17. Hoogenraad, H. R. & De Groot, A. A. 1937. Biometrische Untersuchungen an St8wasserrhizopoden. (Rhizopoden und Heliozoen aus dem Sifwasser er Niederlande. VI). Archiv fiir Hydrobiologie 31: 101-132. 1940. Moosbewohnende thekamoebe Rhizopoden von Java und Sumatra. Treubia 17: 209-259 + PI. 1-9. 1952a. Thekamébe Moosrhizopoden aus Nordamerika. Archiv fiir Hydrobiologie 47: 229-262. — 1952b. Uber ratselhafte Elemente in der Schale der thekamében Moosrhizopode Bullinula indica Penard. Archiv fiir Hydrobiologie 47: 1-8 + Pl. 1. Kufferath, H. 1932. Rhizopodes du Congo. Revue de zoologie et de botanique africaines 23: 52-60 + PI. Ill, IV. Leidy, J. 1878. Species of Euglypha, Trinema, Pamphagus, and Cyphoderia, with synonyma and description of new forms. Proceedings of the Academy of Natural Science of Philadelphia, year 1878: 171-173. Liftenegger, G., Petz, W., Berger, H., Foissner, W. & Adam, H. 1988a. Morphologic and biometric characterization of twenty-four soil testate amoebae (Protozoa, Rhizopoda). Archiv ftir Protistenkunde 136: 153-189. ——,, Petz, W., Foissner, W. & Adam, H. 1988b. The efficiency of a direct counting method in estimating the numbers of microscopic soil organisms. Pedobiologia 31: 95-101. Ogden, C. G. 1981. Observations of clonal cultures of Euglyphidae (Rhizopoda, Protozoa). Bulletin of the British Museum (Natural History), Zoology 41: 137-151. 1983. Observations on the systematics of the genus Difflugia in Britain (Rhizopoda, Protozoa). Bulletin of the British Museum (Natural History), Zoology 44: 1-73. 1984. Notes on testate amoebae (Protozoa: Rhizopoda) from Lake Vlasina, Yugoslavia. Bulletin of the British Museum (Natural History), Zoology 47: 241-263. — 1988. Fine structure of the shell wall in the soil testate amoeba Cyclopyxis kahli (Rhizopoda). Journal of Protozoology 35: 537-540. — & Hedley, R. H. 1980. An Atlas of Freshwater Testate Amoebae. 222p. British Museum (Natural History), Oxford University Press, Oxford. & Cotiteaux, M.-M. 1987. The biology and ultrastructure of the soil testate amoeba Tracheleuglypha dentata (Rhizopoda: Euglyphidae). European Journal of Protistology 23: 28-42. —— 1988. The effect of predation on the morphology of Tracheleuglypha dentata (Protozoa: Rhizopoda). Archiv fiir Protistenkunde 136: 107-115. Pateff, P. 1926. Fortpflanzungserscheinungen bei Difflugia mammillaris Penard und Clypeolina marginata Penard. Archiv fiir Protistenkunde 55: 516-544 + Taf 15,16. Penard, E. 1890. Etudes sur les rhizopodes d’eau douce. Mémoires de la Société de Physique et d’ Histoire Naturelle de Genéve 31: 1-230 + PI. I-XI. 16 — 1902. Faune Rhizopodique du Bassin du Léman. 714p. Kindig, Genéve. — 1912. Notes sur quelques sarcodinés. Trosiéme partie. Revue suisse de zoologie 20: 1-29 + PI. 1,2. Rauenbusch, K. 1987. Biologie und Feinstruktur (REM-Untersuchungen) terrestrischer Testaceen in Waldb6den (Rhizopoda, Protozoa). Archiv fiir Protistenkunde 134: 191-294. Schénborn, W. 1965. Beobachtungen an der Zellteilung von Paraquadrula (Testacea). Limnologica 3: 235-236. 1966. Untersuchungen tber die Testaceen Schwedisch-Lapplands. Ein Beitrag zur Systematik und Okologie der beschalten Rhizopoden. Limno- logica 4: 517-559. —— & Peschke, T. 1988. Biometric studies on species, races, ecophenotypes and individual variations of soil-inhabiting Testacea (Protozoa, Rhizopoda), including Trigonopyxis minuta n. sp. and Corythion asperulum n. sp. Archiv fiir Protistenkunde 136: 345-363. , Foissner, W. & Meisterfeld, R. 1983. Licht- und rasterelektronenmikros- kopische Untersuchungen zur Schalenmorphologie und Rassenbildung GABRIELE LUFTENEGGER & WILHELM FOISSNER bodenbewohnender Testaceen (Protozoa: Rhizopoda) sowie Vorschlage zur biometrischen Charakterisierung von Testaceen-Schalen. Protistologica 19: 553-566. Thomas, R. 1959. Les thécamoebiens muscicoles et terricoles: notions d’écologie générale et comparative. Procés verbaux de la Société Linnéenne de Bordeaux 98: 27-53. & Gauthier-Liévre, L. 1959. Note sur quelques Euglyphidae d’ Afrique. Bulletin de la Société d’ Histoire Naturelle de l’ Afrique du Nord 50: 204-221. Valkanov, A. 1962a. Euglyphella delicatula n. g. n. sp. (Rhizopoda-Testacea) und ihre Kopulation. Comptes rendus de I’ Académie bulgare des Sciences 15: 207-209. 1962b. Paraquadrula madarica n. sp. (Rhizopoda-Testacea) und ihre Kopulation. Comptes rendus de Il’ Académie bulgare des Sciences 15: 423-426. Wailes, G. H. & Penard, E. 1911. Rhizopoda. Proceedings of the Royal Irish Academy 31: 1-64 + PI. I-VI. Manuscript accepted for publication 9 May 1990. Bull. Br. Mus. nat. Hist. Zool. 57(1): 17-59 A revision of Cothurnia (Ciliophora: Peritrichida) and its morphological relatives ALAN WARREN Department of Zoology, British Museum (Natural History), Cromwell Road, London SW7 5BD & JAN PAYNTER Department of Parasitology, University of Queensland, St Lucia, Brisbane, Queensland CONTENTS AMSG CUI CEO reese Sc = Re AMOR RR RTE NRE A. isch wk ee erences esha ye ape Se RN Wa av ES oh gS yoe VG Gencialmorpnolosy and CaxOmOmic CharaChelS ok igs seen Sais eee ee we eh a a a oe RT Os chee eas 7 SERIO (CIITA LIA ee aca im Raa ARNE ae ranean ea Pa ee te eee A SL eee ae 20 See TMISO OT LILO SIS eae emer ce tan MOMs ee ae ee Se ye ead «a Lape s Be enaeBas Gee aae aaeee 51 Ser ONIC OOM ET ER A Oe eC ME tere ce ete eel ace Weta d eh ORs dich aerial Mpsn 2 Namieneetre cc akong aoaks SZ EIS On LOCOULUTIVIR mn Metre. ee Nine eae ee ooh oul SEES Ul longs Eo, estate cee CMM te dec a sa eee WINGET UIE, SSG S adie deeb enact Sates, bce raaite: sry ne decree card re ee earn eae Cae eee SS ee ea eae eee 54 PIC RMONVIC ORCI HEGNae: Crane... SANE, MPEAR, MAREE DEIR oe cs cme egmen seule bh ae R Rew nae wae Gee se poet Oo 54 SVSNETENG ER hed La EAGER os A es OU A eS | ee ein ian Cera a eae eee eee ane See ie 54 INP QISSE kD) BOOSIE soy MCR malbatcten he cy cA eae Uke hear ee er”. (RR Mee A ie eee age ae 56 SYNOPSIS. The loricate peritrich genera Cothurnia, Cothurniopsis, Cyclodonta and Dimorphocothurnia are revised and a key to the constituent species of each is given. Several species are synonymised or transferred to other genera and one, Cothurnia felinska, is erected. Four closely related genera, Baikalotheca, Semicothurnia, Tesnotheca and Issued 30 May 1991 Daurotheca are discussed. All extant species are described and figured. INTRODUCTION In recent years four genera belonging to the family Vaginicolidae have been taxonomically revised, namely Pyxicola (Trueba, 1978), Thuricola and Pseudothuricola (Trueba, 1980), and Platycola (Warren, 1982). The present publication aims to extend the revision of the Vaginicolidae to those genera in which the loricas are held erect on an external stalk, and which lack a special valve or possess only a simple means of closing the aperture (Cothurniopsis). The major genus dealt with here is Cothurnia Ehrenberg, 1831, a well known genus that contains a large number of nominal species. Many species of Cothurnia have previously been transferred to other genera but until recently no check list of those remaining was available, thus making it difficult for the taxonomist and ecologist to identify them. In this paper Cothurnia and three other closely related genera, Cothurniopsis, Cyclodonta and Dimorphocothurnia, are taxonomically revised. A key to species and a check list of nominal and extant species of each genus are also provided. GENERAL MORPHOLOGY AND TAXONOMIC CHARACTERS Morphology Accounts of the biology and morphology of genera belonging to the family Vaginicolidae were given by Kralik (1961), Trueba (1978, 1980) and Warren (1982), so it is sufficient here to give a short summary highlighting those features common to the cothurnids. Figure 1 shows the main morphological features of a typical species of Cothurnia. It consists of a lorica which has an aperture at the anterior end, a stalk at the posterior end and contains either one or two zooids. The zooids are typically cylindrical or trumpet-shaped and when relaxed may extend far beyond the aperture, but when contracted withdraw completely within the lorica. Zooids are attached to the base of the lorica either directly by means of one or two non-contractile stalks, or by a series of membranes (e.g. Cyclodonta). The lorica is attached to its substrate by a non-contractile external stalk. In many species the stalk appears to be smooth and comparatively featureless, while others possess lines or stripes which run longitudinally down the stalk. Stalks posses- sing such stripes have been described as ‘fibrillar’ (Felinska, 1965) or ‘striated’ (Lang, 1948). Examination by TEM reveals that the stalk is bounded by an outer limiting membrane and contains tubules arranged in cylindrical groups (“Tubular Units’) within a microfibrillar matrix (Vogelbein and Thune, 1988). This is broadly consistent with the findings of Randall and Hopkins (1962) who reported that the longitudinal stripes observed in the non-contractile stalks of aloricate peritrichs are also due to tubular structures in the stalk matrix. In this revision, these lines shall be referred to as longitudinal striae. Transverse folds or furrows may also be present on the stalk surface. In some species the external stalk penetrates the lorica wall via a special tube at the posterior end (Banina and 18 Fig. 1 A generalised Cothurnia showing the principal morphological features. A—aperture; BD—basal disc; CV—contractile vacuole; E—endostyle; ES—external stalk; FV—food vacuole; I— infundibulum; L—lorica; LS—longitudinal striae; Ma—macronucleus; Me—mesostyle; Mi—micronucleus; PD—peristomial disc; PL— peristomial lip; S—striations; Se—septum; TS—transverse striae/ folds; Z—zooid. Polyakova, 1977). No tube was observed by Vogelbein and Thune (1988). At its point of attachment to the substrate the stalk is flared forming the basal disc. The basal disc rests on a layer of adhesive material which extends up to 10m beyond the disc edge. The adhesive is thought to be secreted during the process of attachment to the substrate (Vogelbein and Thune, 1988). In many cothurnids the lorica contains an extra layer of material forming an internal lining or septum. The septum is most easily seen in the posterior region where it encloses a space at the base of the lorica. In these cases, the zooid is attached to the base of the lorica via a non-contractile middle stalk or ‘mesostyle’. The mesostyle is usually short and broad with conspicuous longitudinal striae. Another stalk, the endostyle, may also be present and connects the zooid either to the lorica or (when present) to the septum. The endostyle is usually short and slender. Except in Cothurniopsis, there is no mechanism of closing ALAN WARREN & JAN PAYNTER the aperture of the cothurnid lorica. In Cothurniopsis the border of the aperture is pliable and folds in to close the aperture when the zooid contracts. According to Stokes (1893) there do not appear to be any special structures involved, but rather the border closes passively as a result of the contraction of the zooid. The precise mechanism by which the aperture closes has yet to be elucidated and is therefore considered a doubtful generic character. Nevertheless, until a redescription is available, Cothurniopsis remains a separate genus. Cothurnids are found in fresh, brackish and marine waters and have a cosmopolitan distribution. They have been recorded on plant, animal (commonly as epizoites of Crustacea) and inanimate substrates. They are not generally regarded as parasitic sensu Stricta. Taxonomic characters Like most other peritrich groups, it is comparatively easy to characterise cothurnid genera but difficult to identify individual species. Many species have only been described once and, in some cases, descriptions were based on poorly fixed or unhealthy specimens. Consequently, important taxonomic data are unavailable for several taxa. The characters used as a basis for this revision are as follows. Lorica shape The dominant feature used for classifying cothurnids, and indeed vaginicolids in general, is the shape of the lorica. Identification keys published by Kahl (1935), Wailes (1943) and Stiller (1971) rely extensively on lorica shape for separa- ting species. Two advantages of the lorica are, (i) it is unaffected by fixation or other methods of preservation, and (ii) it has been described for every known species. Nevertheless, opinions differ over the usefulness of lorica shape as a taxonomic character since variability is well documented (Kralik, 1961; Trueba, 1980; Warren, 1982). The cothurnid lorica is typically cylindrical in shape with a rounded, often bulbous posterior end (Fig. 2). There is an aperture at the anterior end that is circular when viewed from above (Fig. 3). One or more clefts may be present in the aperture border. The main lorica axis is normal to the substratum and extends in straight line from the point of attachment to the external stalk. The aperture is usually at right angles to the main lorica axis although in some cases the neck just below the aperture is curved tilting the aperture obliquely to the main lorica axis (Fig. 4). Occasionally, the whole lorica may be curved (Fig. 5). The lorica may be, (i) compressed or elongated along the main lorica axis resulting in oblate (Fig. 6) or prolate (Fig. 7) forms, or (11) compressed in a plane at right angles to the main lorica axis (Figs 8 & 9) producing dorso-ventral flattening (Precht, 1935). Variations of the first type are readily ob- served, but dorso-ventral compression becomes apparent only if the lorica is viewed either from above, or from both the dorso-ventral and lateral aspects (Fig. 10). Most descriptions of cothurnids are based on observations from one direction only. Yet with all its limitations, the shape of the lorica remains a useful taxonomic character. Lorica size The dimensions most commonly used in species diagnoses are lorica length, maximum lorica width, and width of the A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES 3 2 4 10 . 19 8 e : Variations in shape of cothurnid loricas (following the terminology of Precht, 1935). Fig. 2 Normal lorica. Fig. 3 Normal aperture viewed from above. Fig. 4 Lorica with a curved neck. Fig. 5 Curved lorica. Fig. 6 Oblate lorica. Fig. 7 Prolate lorica. Figs 8-10 Dorso-ventrally compressed lorica, fig. 8 ventral view, fig. 9 lateral view, fig. 10 aperture viewed from above. aperture. Ideally several individuals should be measured and the range of variation given. Among other peritrichs it has been suggested that the ratio of length: width is a more meaningful character than size alone (see Warren, 1987), and this may also prove to be the case for cothurnid loricas. Stalks All cothurnids possess a non-contractile external stalk attach- ing the lorica to the substratum. Some authors (Entz, 1884; Jankowski, 1985) have suggested that the size of the stalk and the presence of transverse folds on its surface are generic characters. In this revision stalk size and surface folding are not employed at the generic level although they are used for separating species. A few species appear to have a special tube at the posterior end of the lorica through which the external stalk penetrates the lorica wall (Banina and Polyakova, 1977). Mesostyles and endostyles are frequently included in descriptions of cothurnids and have long been recognised as useful characters for separating species. Jankowski (1985) suggested that the presence or absence of mesostyles and endostyles should be regarded as generic characters although acceptance of this decision must await further studies. Zooids The taxonomy of aloricate peritrichs is based largely on characters relating to the zooid (Foissner, 1981; Warren, 1987), frequently omitted from species descriptions of cothurnids. Descriptions available are often from studies made on fixed, contracted cells yielding information of limited value. The following zooid characters are useful; size and shape of zooid, peristomial lip, disc and macro- nucleus; position within zooid of contractile vacuole; number and spacing of pellicular striations. However, the following points should be noted with reference to these characters. 20 i) Zooid size; the proportion of the total zooid length which extends beyond the aperture may be more significant than size alone. ii) Size and shape of peristomial lip and disc. Felinska (1965) suggested that the shape of the disc is an important taxonomic character since it ‘has no transitory forms and does not undergo modifications’, and divided thirteen species of Cothurnia into two groups—those with convex discs and those with flat, oblique discs. While the shape of the disc is probably a useful character in healthy peritrichs it should be noted that, among aloricate forms at least, the shape of the disc varies in response to changes in environmental conditions (see Warren, 1987). iii) Position within zooid of contractile vacuole (CV). The CV often lies either in or just below the peristomial region although in some descriptions its position is described in relation to the infundibulum. If, however, the infundibulum is short or orientated obliquely, differences may be subtle. iv) Size and shape of macronucleus. Entz (1884) cited the shape of the macronucleus as one of the principal differences between Cothurnopsis (compact macronucleus) and Cothurnia (band-like or vermiform macronucleus). In this revision the macronucleus is employed as a taxonomic character at the species level only. v) Pellicular striations. Biometric analysis of pellicular stria- tions are used increasingly in the taxonomy of aloricate peritrichs (Foissner, 1979, 1981; Foissner and Schiffmann, 1974, 1975), although few studies of this type have been carried out on loricate peritrichs. Nevertheless it is usually possible to determine whether the striations are widely spaced (clearly visible) or narrow spaced (fine or incon- spicuous), and whether the ribbing between the striations is convex, concave or normal (Foissner and Schiffmann, 1975). Colony and pseudocolony formation Among the vaginicolids there are typically either one or two zooids per lorica, although in some genera examples of three or even four zooids per lorica have been reported (Kralik, 1961). Jankowski (1985) suggested that the presence of more than one zooid in each lorica represents a colony, and that coloniality should be regarded as a generic character among loricate peritrichs in the same way as it is among the aloricates. Unfortunately, as Jankowski (1985) himself points out, usu- ally no attempt is made to determine the number of zooids per lorica other than at the time of observation, so it is almost impossible to know whether species described in the literature as ‘solitary’ are in fact immature ‘colonial’ forms. Therefore, due to the paucity of data available, ‘coloniality’ (sensu Jankowski, 1985) is not recognised as a generic character in this revision, although it is employed for the separation of species. At least two species of Cothurnia (C. bavarica and C. variabilis) form pseudocolonies; that is they develop as chains with each individual attached to the lorica of its neighbour via its external stalk. A third species, C. nebaliae, also forms pseudocolonies and, in addition, exhibits a dimorphic lorica structure with the individual in contact with the substratum (‘basont’ Jankowski, 1985) having a substantially longer external stalk than the others in the chain. Jankowski (1980) erected the genus Dimorphocothurnia for Cothurnia nebaliae on the basis of its lorica dimorphism. ALAN WARREN & JAN PAYNTER Key to genera 1 Aperture border rigid; no closure of aperture........... Z Aperture border pliable, used to close aperture on contraction Of ZOOIG(S)) «p42. aks ee COTHURNIOPSIS 2 Zooid(s) attached to inside of lorica either directly or via Stalk .cisnus Fee eho een eee ate se eee 3 Zooid(s) attached to inside of lorica via series of mem- BrABeS cca: sarjaiaigins dmalesgerinaewes ame CYCLODONTA 3. Forms pseudocolonies; lorica dimorphism with individual in contact with substratum having longer external stalk than those of rest of chain ........ DIMORPHOCOTHURNIA Usually does not form pseudocolonies; not exhibiting lorica dimorphism Ss... 220. 22. eee COTHURNIA Genus COTHURNIA Ehrenberg, 1831 Cothurniopsis sensu Penard, 1914 Cothurnopsis Entz, 1884 Semicothurnia Jankowski, 1976 Sincothurnia Jankowski, 1985 Tesnotheca Jankowski, 1985 The name Cothurnia was first mentioned by Ehrenberg (in Hemprich & Ehrenberg, 1828) as part of an identification table in which loricate peritrichs with stalks (Cothurnia) were separated from those without stalks (Vaginicola). The original diagnosis of Cothurnia was founded on the descrip- tion of Cothurnia imberbis Ehrenberg, 1831. A second species (C. ? mystacina Ehrenberg, 1831) was described at the same time but subsequently transferred to the genus Acineta (Ehrenberg, 1833). C. imberbis thus became the type species by monotypy. In his discussion of Cothurnia, Jankowski (1985) states that it is ‘highly undesirable’ for C. imberbis to be the type species because it is a ‘very difficult? and ‘in practice unstudied’ organism; in its place he designated C. maritima Ehrenberg, 1838 as the type species. Jankowski was apparently unaware of Bacon’s (1968) extensive study of C. imberbis, as a result of which more is known about this than any other species of Cothurnia. C. imberbis may therefore be properly designated as the type species of Cothurnia. The taxonomy of Cothurnia remained largely unchanged over the next fifty years although several new species were added to the genus. In 1884 Entz erected the genus Cothurnopsis for cothurnid species with large, transversely folded external stalks and compact macronuclei. Kahl (1935) later submerged Cothurnopsis (which he misspelt ‘Cothurniopsis’) into the genus Cothurnia, a decision supported by Matthes (1958) but opposed by Jankowski (1985). In a recent review of the cothurnids, Jankowski (1985) divided the species of Cothurnia into those with two zooids per lorica, i.e. ‘colonial’ (Sincothurnia), and those with a single zooid per lorica (Cothurnia sensu stricta). A third genus, Dimorphocothurnia Jankowski, 1985, was erected for cothurnids which form pseudocolonies and possess dimorphic loricas. For the reasons stated in the previous section the concept of coloniality as a generic character has not been applied in this revision so Sincothurnia is synonymised in the genus Cothurnia. Other genera erected by Jankowski for ‘aberrant’ Cothurnia species include Baikalotheca Jankowski, 1985, Tesnotheca Jankowski, 1985, Semicothurnia Jankowski, 1976 and A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES 21 Various junctions between stalk and lorica found in Cothurnia Fig. 11 External stalk only. Fig. 12 Endostyle and external stalk. Fig. 13 Mesostyle, endostyle and external stalk. Fig. 14 Endostyle and external stalk which penetrates the lorica wall via a special tube. Daurotheca Jankowski, 1987. Baikalotheca does not possess a true external stalk and so falls outside the scope of this paper. The main distinguishing feature of Tesnotheca is its asym- metrical zooid, while that of Semicothurnia is its prominent and distinctive stalk; neither character is considered sufficient to separate the genera, so the species of Tesnotheca and Semicothurnia are transferred to Cothurnia. Daurotheca is characterised by its unusual lorica shape and a tendency to form spines on the lorica wall. Daurotheca is of uncertain taxonomic status and is dealt with here under Incertae Sedis. Diagnosis of Cothurnia Marine, brackish or freshwater loricate peritrichs usually with one or two zooids per lorica. Lorica borne on stalk and attached to aquatic animals, plants or inanimate objects. Lorica without valves or other means of closing the aperture. Inner layer or septum sometimes present enclosing a space at posterior end of lorica; septum connected to base of lorica via mesostyle. Zooid(s) attached to base of lorica (or septum) directly or via endostyle. Key to the species of Cothurnia Although the number of zooids per lorica has been used in this key, this character should be applied only to fully developed individuals; a lorica with one zooid may be a partly developed ‘colonial’ (sensu Jankowski, 1985) species. iieeendostyle and mesostyle absent ..2../00..0. 0.20000 005 Zz Endostyle and/or mesostyle present................... 25 URL) Gwe eg ne ae a 5. FUSING IIST. ie ei hae ici hr ge tenga a ae en eee 10 3 Diameter of aperture = maximum lorica width ......... 4 Diameter of aperture < maximum lorica width ......... 6 feeeeonica with ridges or f{UrTOWS. ..- 1.2... eee ee ee 5 Lorica without ridges or furrows.................+.- C. innata 5 Lorica 35-40 wm long with annular ridge near posterior LID) 2.0 oo SeaRORe esate eee at: 9 ee ee eS C. amoyensis Lorica 50-60 wm long with two or more ridges and furrows on anterior half oflorica ............. C. antarctica 6 External stalk without transverse folds or striae......... 7 External stalk with transverse folds and striae ....C. subglobosa 7 Zooids reach as far as or just beyond aperture .......... 8 Zooids extend at least one third of their length beyond IPSC Rt tet ech yee eee aeons oR ee Se ot 2) 8 Lorica symmetrical with annular ridges; aperture border BATU ELMS Ea. Mate Sede tens A tase e an hes ists whe eh C. curvula Lorica asymmetrical and without ridges; lateral borders of aperture with distinct cleft(s) ............... C. inclinans 2) 10 11 12 13 14 15 16 it, 18 19 20 21 Wi 23 24 Macronucleus lies transversely in body; zooid extends up to half its length beyond aperture ............... C. mobiusi Macronucleus lies longitudinally in body; zooid extends up to one third of its length beyond aperture ....... C. ovalis External stalk penetrates lorica wall via special tube (Figs Ie ee cess ae Che ee ee Ree Fe eee ee Aa te BS i. External stalk does not penetrate lorica wall via special tubex (ie Sl) hoi Aires: ale: Aiea Cte ea eee 12 Lorica with annular ridge(s); two zooids per lorica 5 ald.o Bip shtatec Wists ear eae tera ecarnucts, Sent ak Lee eae C. monoannulata Lorica without annular ridge(s); one zooid per lorica OE es a eee ae nis a ee ........ C. oviformis Diameter of aperture > maximum lorica width ......... 13 Diameter of aperture < maximum lorica width ......... 15 Lorica with'two zooids 3959 2 Se 28 BPR Pen, Fe 14 iKorica withone Z001dy Mae ee ee Pa ee C. spissa Forms pseudocolonies with several loricas attached in a chain; zooids extend just beyond aperture ...... C. bavarica Not forming pseudocolonies; zooids extend up to half their lenoth beyond apertureicie) 20 see C. patula Lorica with distinct annular ridges or furrows........... 16 Lorica without annular ridges or furrows............... 17 Lorica cylindrical with evenly spaced furrows down entire To) 0 ee oe cae eae ack Ae eae ace C. lapponum Lorica rotund with three distinct centrally located ridges CSc tear PR ire ae ish nat ia eas aig ch he Serra i C. pupa LORCA. Wit ONE: ZOO 3. 6c. Gh, own okie sola 18 Lorca with tWOZO00IdS!, (sac... c n-ne ee ete Sone C. kahli Macronucleus long and vermiform.................... 19 Macronucleus short, compact and slightly curved or (OAS TIO G aeeS ees kOe EM aS OR eer Greece Manes DD, Diameter of aperture = greatest lorica width ........... 20 Diameter of aperture < greatest lorica width Macronuclens'straig hts ee ee. Pe ee 3 heise see = 21 Nacronicleus serpentine. 4555025505 -cleectsons se C. soldida Lorica 75 wm long x 40 wm wide; zooid extends up to one sixth of its length beyond aperture ........... C. macrodisca Lorica 125 um long X 75 wm wide; zooid extends up to one third of its length beyond aperture ............ C. ovata Diameter of aperture < greatest lorica width ........... 23 Diameter of aperture = greatest lorica width ........... 24 Lorica curved; typically forms pseudocolonies with several loricas attached inachain.............. C. variabilis Lorica not curved; does not form pseudocolonies ...... C. vaga Peristomial lip thick and prominent; external stalk with- out transversetiolds/or striaes. -22 5) 2 ee C. brevistyla Peristomial lip not prominently thickened; external stalk With! (ransverse'striae: i. SPs. Aes MAO C. astaci 25+ A Marine or brackish steaks aes tee ean eee 26 Freshwater) \: oc. 27s Pa... Rees: eae 73 26, Mesostylejpresent:<. 26 of maximum lorica width; lorica without ridges or furrows... 74. .54..0050-9n6 C. complanata Lorica 150 4m long; CV lies in centre of zooid...... C. elongata Lorica 50-62 um long; CV lies just below peristome . C. parvula Pellicular striations ConspiCuOUS, ........25..4 sole eee 44 Pellicular striations inconspicuous .................... 47 Macronucleus straight Se 2.25. Sees. See eee 45 Macronucleus serpentine?!) Wit alk SE ee 46 Zooid extends up to one third of its length beyond aperture; lorica without ridges and furrows..... C. ceramicola Zooid extends up to one quarter of its length beyond aperture; lorica with ridges and furrows ........ C. coarctata Brackish, lorica 9S Am lOMp. ws se ee ee C. obliqua Marine; lorica 50-60 wm long ..................-. C. simplex Zooid extends = one third of its length beyond aperture 48 Zooid extends < one third of its length beyond aperture 50 Lorica with ridges and furrows ................0.---55 49 Lorica without ridges and furrows ............. C. stylarioides Lorica 121 wm long; length of external stalk one quarter loricatlemethe ge eta: Shee es ook ool a eee C. cyclopis Lorica 50-60 wm long; external stalk < one quarter lorica length. 265.2 2s. borne ses eee ee C. cypridicola External stalk with inconspicuous longitudinal striae .... Sy! External stalk with conspicuous longitudinal striae 9 Pi, pice eer AEE ot AOR MM he ome aed 3% 0 C. harpactici Lorica 120 wm long; external stalk = one quarter lorica FST 0 aie Beat se ons; ute Shoat Bes C. cordylophora Lorica 60-70 «um long; external stalk < one quarter lorica length 103 Ming t tans Poh de lye ia C. collaris Lorica with two:zooids®: .Sa ive 3.2.0). UL eee 53 Lorica:withone Zoids: sp. eas eee ee 54 Lorica curved; macronucleus short and C-shaped SeSeed goes. PY. Re PR ee ee C. limnoriae Lorica not curved; macronucleus vermiform and coiled irregularly. . (swe leas .. so. eee ae C. trophoniae External stalk with conspicuous longitudinal striae ...... 55 External stalk with inconspicuous longitudinal striae .... a Miacronucleus straight co... maximum lorica width ......... 62 Lorica curved, 100 wm long x 50 wm wide ........ C. recurvata Lorica not curved, 60 um long X 30 wm wide......... C. inflata Lorica cyhndrical, 110;uivlone :: 222. ats ee C. fecunda Loncaiconical,, 50-56 jemilone. 2.25 oo snc oan: C. acuta orca withitworzooids® 2 s....0 5 f0.20 eee as oe ae ae 64 Eerica with ORG 2OGIG ohh. <5. ocjecy ine soe na eto 66 Aperture border withiclefts © 25.555 65 0. . Ak eters 65 Aperture border without clefts... . 2... ..6. 66.5008 C. nodosa Lorica 90-100 xm long; zooid 150 «m long and extends up to one third of its length beyond aperture ...... C. sinuosa Lorica 70-80 um long; zooid 85-90 um long and extends up to one fifth of its length beyond aperture ......... C. entzi Peihicular striations conspicuous |. <7 4. ea eee a 67 Pellicular striations MconspicuOUS . 2.6.0.6... esse oP 70 Lorica without ridges or furrows; zooid extends > one quarter of its length beyond aperture................ 68 Lorica with ridges and furrows; zooid reaches just beyond APPCMCUSE ears one erred tem enone ea eae ae C. recurvata Diameter of aperture < maximum lorica width ......... 69 Diameter of aperture > maximum lorica width ... C. amphicteis Lorica 40-45 ym long; peristomial lip prominently thickened ®. 2... pea. . Peo, asl C. membranoloricata Lorica 35 ym long; peristomial lip not prominently thickened tars yhewsieel Ses See ed C. maritima Lorica without annular ridges or furrows............... ips Lorica with annular ridges and furrows .............. C. cohni fapervore border without clefts 2.2.5. 552 4.) 0b ss aes 72 Aperture borderwith clefts... .%... 6c. 4.003... C. compressa Lorica 100 um long; length of external stalk = lorica Pa of ASIN wy ist tn. anche eR BARN HAE ls C. longipes Lorica 65-72 um long; length of external stalk < lorica Sap Sie oth: cake RR: os eed 5 Sa C. subtilis MICSOStVIC PLCSENE oo: pear MN Eee ae eos Hae sn SEs 74 Riesosivic absent sia tee 8 ik SOs oes 83 MaGOstVie PLOSCUts. 0... one third of its length beyond APEMUEE ie sates a as so oe Se eee 81 Zooid only reaches as far as or just beyond aperture ..... 82 Macronucleus long and vermiform................. C. ruthae Macronucleus short and thick .................... C. richtersi Zooid with prominent annular ridge(s); pellicular stria- TONS CONSPICUOUS: o.oo tt ot ke ae eee C. elegans Zooid without annular ridge(s); pellicular striations in- CONSPICUOUS «2c cnae cde eee toc: eee ee C. angusta Lorica with two zooids ................00- eee eee eee 84 Loriea with one Zooid ... 25.) oe on Oe eee 86 Diameter of aperture < maximum lorica width ......... 85 Diameter of aperture > maximum lorica width .... C. clausiens Lorica 30-40 um long; zooid only reaches as far as or just beyond aperture (2.2225 shot cee eee C. anomala Lorica 71-82 um long; zooid extends up to one third of its lengthibeyond aperture=-... ... «see eee C. oblonga External stalk does not penetrate lorica wall via special GUBE . oe oe ees dae hate teen ane Eee 88 External stalk penetrates lorica wall via special tube (Figs LN) es, aed ois aul eam eee ad ee 87 Zooid with annular ridge(s); pellicular striations CONSPICUOUS! eAge sv .eeeb see. Se fee Se C. annulata Zooid without annular ridge(s); pellicular striations INCONSPICUOUS 4) Ea oe Se eee C. asymmetrica External stalk with transverse folds and ridges.......... 89 External stalk without transverse folds and ridges ....... 91 Diameter of aperture = maximum lorica width ......... 90 Diameter of aperture < maximum lorica width ....... C. curva Macronucleus short and slightly curved; pellicular stria- LONSIMEONSPICUOUS. 52 in4-6 -ehe see eee C. canthocampti Macronucleus C-shaped; pellicular striations inconspicuous spite eegtmty suonatern. Recduapbisets by dete aes ec epee Ree C. lata Zooid reaches as far as or just beyond aperture ....:.... 92 Zooid extends > one quarter of its length beyond APEMUTC sss oes ee ot. ace ah meee a eee 94 Lorica 48-55 um long; external stalk with inconspicuous longitudinal striae... : ss. sees ae 93 Lorica 70-72 mm long; external stalk with conspicuous longitudinalistriae: 6 She oo 2.2). Paneer C. carinogammari External stalk 10-15 um long; zooid with posterior anntilar ride (S)in..tse Spates so eaeat ee C. irregularis External stalk 2 wm long; zooid without annular ridge(s) C. asimmetrica EoneaJ25 pom lOmg.2 one ne ee one eee 95 Figs 27 & 28 Cothurnia antarctica, after Daday, 1911, bar = 50 um. Figs 29 & 30 Cothurnia aplatita; fig. 29 after Felinska, 1965 (called Cothurnia aplatita var. flexa); fig. 30 after Stiller, 1939, bar = 50 um. Fig. 31 Cothurnia apseudophila, after Lang, 1948, bar = 100 um. Figs 32-34 Cothurnia arcuata; figs 32 & 33 after Mereschkowsky, 1879, bar = 100 um; fig. 34 after Gourret and Roeser, 1886 (called Cothurnia fusiformis). Figs 35 & 36 Cothurnia asimmetrica, after Banina and Polyakova, 1977, bar = 25 wm. Figs 37 & 38 Cothurnia astaci after Matthes and Guhl, 1973; fig. 37 bar = 50 um; fig. 38 showing variation in lorica and external talk. Figs 39 & 40 Cothurnia asymmetrica, after Sommer, 1951. Figs 41-43 Cothurnia auriculata; figs 41 & 42 after Stiller, 1939; fig. 43 after Felinska, 1965, bar = 100 wm. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES «\ wi) iit \ “tt ‘ae | mg, } GF Zs 26 Loriea 35-40 pum donige 2 4%..200 cl RO 4 A, 2 C. minutissima 95 Lorica 200 um long; pellicular striations conspicuous Hi 2 TAMU See OS RR 2 ee SE oc oe C. irregularis Lorica 127 um long; pellicular striations inconspicuous RAS Ot Oe SOS See o een oom ae ph act C. plachteri Species descriptions Cothurnia imberbis Ehrenberg, 1831 Sincothurnia imberbis (Ehrenberg, 1831) Jankowski, 1985 DESCRIPTION (Figs 15-17). Lorica 45-85 um long x 20-40 wm wide. Aperture 10—20 um in diameter. External stalk 15 wm long, mesostyle short and broad, endostyle short and incon- spicuous. External stalk and mesostyle with conspicuous longitudinal striae. Zooid 57-95 um long x 8-18 wm wide, extending about one fifth of its length beyond aperture. Annular ridge in region of telotroch band. Peristomial lip well developed, 13-18 wm in diameter. CV situated just below peristome. Macronucleus straight, curved anteriorly and lies longitudinally in zooid. Pellicle with fine striations. HABITAT. Freshwater, attached to a variety of plant and animal substrates. Note. This, the type species of the genus, was redescribed by Bacon (1968). Cothurnia acuta Levander, 1915 Semicothurnia acuta (Levander, 1915) Jankowski, 1976 DESCRIPTION (Figs 18 & 19). Lorica conical, 50-56 wm long x 30 wm wide. Aperture 31 um in diameter. External stalk broad, up to 8 wm long; endostyle also broad, up to 4 wm long; mesostyle absent. Zooid 60 wm long xX 20 wm wide and extends up to one third of its length beyond aperture. Peristomial lip 18 ~m in diameter. Macronucleus lies longi- tudinally in zooid, vermiform and curved anteriorly. Pellicular striations conspicuous. HABITAT. Brackish water, originally isolated as an epizoite of the polychaetes Harmothoe sarsi and H. imbricata (Levander, 1915). Note. Jankowski (1976) erected the genus Semicothurnia for C. acuta on the basis of its broad external stalk. The stalk is not sufficiently distinctive to separate a new genus so this species is retained in the genus Cothurnia. Cothurnia amoyensis Wang, 1935 DESCRIPTION (Fig. 20). Lorica 35-40 um long x 18-20 wm wide with annular ridge near posterior end; lorica cylindrical above the ridge and conical below. External stalk 6 wm long with transverse annulations. Mesostyle and endostyle absent. Zooid extends just beyond aperture, 45 wm long X 15 wm wide. Peristomial lip well developed, 18 wm in diameter. Disc convex. CY situated one quarter of way down zooid. Macro- nucleus vermiform, curved at both ends and extends almost entire length of zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally found in a laboratory culture containing decomposing algae (Wang, 1935). ALAN WARREN & JAN PAYNTER Cothurnia amphicteis Lang, 1948 Semicothurnia amphicteis (Lang, 1948) Jankowski, 1985 DESCRIPTION (Fig. 21). Lorica conical, 53-56 wm long X 35 pm wide. Aperture 35—37 «wm in diameter. External stalk 54— 57 um long, lower half with conspicuous longitudinal striae, upper half granular, and with ring-like thickening at point of attachment to lorica. Endostyle short, broad and with con- spicuous longitudinal striae. Mesostyle absent. Zooid conical, 85 um long x 40 wm wide, and extends between one third and one half of its length beyond aperture. Peristomial lip well developed, 45 4m in diameter. Disc convex. CV lies just below peristome. Macronucleus elongate, slightly curved anteriorly. Pellicular striations conspicuous. HABITAT. Marine, originally found attached to the setae of Amphicteis gunneri in coastal waters of Sweden (Lang, 1948). Cothurnia angusta Kahl, 1933 DESCRIPTION (Fig. 22). Lorica elongate, 60 wm long Xx 15-23 sm wide, rounded posteriorly. Aperture 10 wm in diameter. External stalk about 10 wm long, mesostyle short, endostyle absent. Zooid 70 wm long xX 12 wm wide and extends just beyond aperture. Peristome 18 wm diameter. CV situated about one sixth of way down zooid. Pellicular striations inconspicuous. HABITAT. Brackish or freshwater, originally found attached to cyprid ostracods (Kahl, 1933; Precht, 1935); also found on Enteromorpha intestinalis (Sommer, 1951). Cothurnia annulata Stokes, 1885 Cothurniopsis annulata (Stokes, 1885) Penard, 1922 DESCRIPTION (Fig. 23). Lorica 55 wm long X 25 um wide and rounded posteriorly. External stalk short; endostyle short and with conspicuous longitudinal striae; mesostyle absent. Zooid slender, 65 wm long X 12 wm wide and with centrally located annular ridge. Peristomial lip 15 wm in diameter. CV situated about one sixth of way down zooid. Macronucleus straight and lies longitudinally in centre of zooid. Pellicular striations conspicuous. HABITAT. Freshwater, originally isolated from North American ponds attached to Myriophyllum (Stokes, 1885, 1888); also found on the aquatic plants and algae Lemna, Hydroictyon reticulatum and Enteromorpha _ intestinalis (Sommer, 1951). Note. Although the name C. annulata was originally men- tioned by Stokes in 1885, a detailed description and diagram did not appear until three years later (Stokes, 1888). C. annulata was redescribed by Penard (1922) and by Sommer (1951). Cothurnia anomala Stiller, 1951 DESCRIPTION (Figs 24-26). Lorica 30-40 xm long x 15-25 wm wide. Diameter of aperture equal to maximum lorica width. External stalk broad, 5 wm long; endostyle short and incon- spicuous; mesostyle absent. Two zooids per lorica, one not reaching as far as aperture, the other extending just beyond A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES aperture. Peristomial lip well developed, 20 wm in diameter. CV situated just below peristome. Macronucleus slightly curved and lies longitudinally in zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, originally isolated from Lake Balaton as an epizoite of the amphipod Corophium curvispinum (Stiller, 1951). NogtE. If coloniality (sensu Jankowski, 1985) is accepted as a generic character among cothurnids, this species should be included in the genus Sincothurnia. Cothurnia antarctica (Daday, 1911) n. comb. Cothurniopsis antarctica Daday, 1911 DESCRIPTION (Figs 27 & 28). Lorica 50-60 wm long x 30 wm wide with 1-4 annular constrictions on anterior half. Aper- ture 30 wm in diameter. External stalk 40-50 um long and with transverse folds. Endostyle and mesostyle not observed. Zooid extends just beyond aperture. Macronucleus compact. Pellicular striations inconspicuous. HABITAT. Marine, originally found in the Antarctic region as an epizoite of the ostracod Philomedes laevipes (Daday, ICH Cothurnia aplatita Stiller, 1939 Cothurnia aplatita var. flexa Felinska, 1965 DESCRIPTION (Figs 29 & 30). Lorica 45-65 um long Xx 25-30 fm wide and typically covered with particles of sand and detritus. External stalk and endostyle short; mesostyle short, broad and with conspicuous longitudinal striae. Two zooids per lorica, each 70-80 wm long xX 20-25 wm wide and extending up to one quarter of its length beyond aperture. Peristomial lip 25 um in diameter. CV situated just below peristome. Macronucleus elongate. Pellicle with fine striations. HABITAT. Marine, originally isolated from laboratory aquaria at Helgoland attached to Campanulariaceae (Stiller, 1939); also found in marine aquaria at Plymouth attached to green algae (Felinska, 1965). Norte. According to Felinska (1965) C. aplatita and C. aplatita var. flexa are identical apart from minor differences in the shape of the lorica. These differences are considered to be insufficient for the recognition of the two organisms as separate taxa. Furthermore, if coloniality (sensu Jankowski, 1985) is accepted as a generic character among cothurnids, this species should be included in the genus Sincothurnia. Cothurnia apseudophila Lang, 1948 Sincothurnia apseudophila (Lang, 1948) Jankowski, 1985 DESCRIPTION (Fig. 31). Lorica cylindrical, 88-92 wm long x 29-31 wm wide, and rounded posteriorly. Aperture 29-31 wm in diameter. External stalk and mesostyle 7-8 «m long with conspicuous longitudinal striae; endostyle 1.5 um long. Two zooids per loria one of which is always larger than the other; larger zooid 164 um long X 20 wm wide, extending up to one half of its length beyond aperture; smaller zooid 128 um long x 17 um wide, extending up to one third of its length beyond aperture. Peristome 25-30 um in diameter. CV situated just 2 below peristome. Macronucleus straight and lies longitudin- ally in zooid. Pellicular striations conspicuous. HABITAT. Marine, originally found on the setae of Apseudes spinosus in Swedish coastal waters (Lang, 1948). Cothurnia arcuata Mereschkowsky, 1879 Cothurnia arenata (Mereschkowsky, 1879) Zelinka, 1928 Cothurnia fusiformis (Gourret & Roeser, 1886) Zelinka, 1928 DESCRIPTION (Figs 32-34). Lorica 125 wm long xX 60-65 wm wide, cylindrical and curved. External stalk up to 40 um long, broad and with large basal disc. Mesostyle short and broad. Endostyle absent. Zooid 125 wm long X 35 wm wide and just reaches aperture. Peristomial lip 25 wm in diameter. CV situated near centre of zooid. Macronucleus C-shaped and lies in centre of zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally isolated from the White Sea (Mereschkowsky, 1879); also isolated from the Port of Marseille (Gourret & Roeser, 1886). Cothurnia asimmetrica Banina and Polyakova, 1977 DESCRIPTION (Figs 35 & 36). Lorica 48-52 wm long X 25 wm wide and typically inclined at an angle to substrate. Aperture 12-18 um in diameter. External stalk 2 wm long and pene- trates lorica wall via special tube; endostyle 1 um long; mesostyle absent. Zooid 50 wm long X 17 wm wide and just reaches aperture. CV situated one third of way down zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found on Cladophora (Banina and Polyakova, 1977). Cothurnia astaci Stein, 1854 Cothurniopsis astaci (Stein, 1854) Entz, 1884 DESCRIPTION (Figs 37 & 38). Lorica 65-90 wm long x 35-45 jm wide and rounded posteriorly. Aperture 40 wm in dia- meter. External stalk 15 um long, slightly curved and with transverse folds. Mesostyle and endostyle absent. Zooid slender, 75 wm long xX 20 «wm wide and extends just beyond aperture. CV lies one third of way down zooid. Macronucleus short, C-shaped and situated in centre of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, found on a variety of substrates includ- ing the crayfish Astacus fluviatilis and A. leptodactylus (Matthes and Guhl, 1973), and on Entomostraca (Kahl, 1935). Norte. A recent redescription of this species was given by Matthes and Guhl (1973). Cothurnia asymmetrica Sommer, 1951 DESCRIPTION (Figs 39 & 40). Lorica 50 wm long Xx 25 wm wide and with prominent bulge about two thirds of way down. Aperture 12 wm in diameter. Neck region below aperture curved. External stalk slender, about 10 wm long; endostyle short; mesostyle absent. Zooid 50 wm long X 15 wm wide and extends just beyond aperture. Peristomial lip 18 mm in diameter and prominently thickened. CV large, ovoid and lies below peristome. Pellicular striations inconspicuous. 28 ALAN WARREN & JAN PAYNTER itty he [f_2 A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES HABITAT. Freshwater, originally found attached to Cladophora and Enteromorpha intestinalis (Sommer, 1951). Cothurnia auriculata Stiller, 1939 Cothurnia auriculata var. flexa Felinska, 1965 Sincothurnia auriculata (Stiller, 1939) Jankowski, 1985 Sincothurnia flexa (Felinska, 1965) Jankowski, 1985 DESCRIPTION (Fig. 41-43). Lorica irregular, 110-150 wm long x 45-60 wm wide. Aperture narrow with two clefts in aperture border. External stalk short and with broad basal disc; mesostyle short, broad and with conspicuous longitudinal striae; endostyle slender and variable in length. Zooid cylin- drical in shape, 20 wm wide and just reaches aperture. Peristomial lip 25 xm in diameter. Disc convex. CV lies just below peristome. Macronucleus vermiform with anterior end curved horizontally across peristome. Pellicle with fine striations. HABITAT. Marine, originally isolated from the North Sea attached to Cladophora (Stiller, 1939); also reported from marine aquaria at Plymouth (Felinska, 1965). Cothurnia bavarica Matthes and Guhl, 1973 DESCRIPTION (Fig. 44). Lorica 85-93 um long x 49-59 wm wide. Aperture 60 wm in diameter. External stalk 33-35 wm long X 10-12 um wide and with transverse folds. Mesostyle and endostyle absent. Forms pseudocolonies with loricas of several individuals attached to each other via external stalks to form a chain. Two zooids per lorica, each 87-118 wm long x 20-35 «wm wide and extends just beyond aperture. Peristo- mial lip well developed, 35 wm in diameter. Disc raised and umbilicate. CV situated near mid-region of infundibulum which reaches centre of body. Macronucleus _ thick, horseshoe-shaped and lies in centre of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found attached to the crayfish Astacus leptodactylus and Cambarus affinis (Matthes and Guhl, 1973). Note. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia brevistyla Nenninger, 1948 DESCRIPTION (Figs 45 & 46). Lorica cylindrical 43.5—54.9 wm long X 20-25 um wide, and rounded posteriorly. External stalk short and with broad basal disc. Endostyle and mesostyle absent. Zooid 58 wm long X 20 wm wide. Peristomial lip well developed, 25 um in diameter. Disc convex. CV lies just Fig. 44 Cothurnia bavarica, after Matthes and Guhl, 1973, bar = 100 um. Figs 45 & 46 Cothurnia brevistyla, after Nenninger, 1948, bar = 50 wm. Figs 47 & 48 Cothurnia butschlii, after Zelinka, 1928, bar = 50 wm. Fig. 49 Cothurnia canthocampti, after Stokes, 1886, bar = 100 wm. Fig. 50 Cothurnia carinogammari, after Stiller, 1953, bar = 50 wm. 29 below peristome. Macronucleus thick, slightly curved and lies longitudinally in zooid. Pellicle with fine striations. HABITAT. Freshwater, originally found attached to algae from canal water in Germany (Nenninger, 1948). Cothurnia butschlii Zelinka, 1913 DESCRIPTION (Figs 47 & 48). Lorica 53.2 wm long X 27 wm wide and tapers towards stalk. Aperture 25 um in diameter. External stalk 10 wm long X 4 wm wide; endostyle short; mesostyle absent. External stalk and endostyle with contin- uous longitudinal striae. Zooid 25-30 wm long when con- tracted. Macronucleus C-shaped and lies longitudinally in zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally found attached to echinodera (Zelinka, 1913). Note. The data above is based on observations of contracted specimens only. Uncontracted specimens of C. butschlii have yet to be described. Cothurnia canthocampti Stokes, 1886 Cothurniopsis canthocampti Monard, 1919 DESCRIPTION (Fig. 49). Lorica 84-95 um xX 40-42 um wide. Aperture 25 ym in diameter. External stalk 20-30 wm long and with regular transverse folds; endostyle short and incon- spicuous; mesostyle absent. Zooid 85 xm long xX 30 wm wide extending just beyond aperture. CV situated one third of way down zooid. Macronucleus short, curved and lies longitudin- ally in centre of body. Pellicular striations conspicuous. HABITAT. Freshwater, originally found on Canthocamptus minutus in North America (Stokes, 1886). Cothurnia carinogammari Stiller, 1953 DESCRIPTION (Fig. 50). Lorica 70-72 wm Xx 40-42 um wide. Aperture 25-30 wm diameter. External stalk 10 wm long; endostyle short and broad; mesostyle absent. Zooid 75 wm long X 25 mm wide and reaches just beyond aperture. Peristomial lip 25 wm in diameter. Macronucleus elongate, C- shaped and lies longitudinally in zooid. Pellicle with fine transverse striations. HABITAT. Freshwater, originally found attached to the crust- acean Gammarus (Stiller, 1953). Cothurnia ceramicola Kahl, 1933 DESCRIPTION (Figs 51-53). Lorica 65-77 wm long x 30 wm wide and rounded posteriorly. Aperture 25 wm in diameter. Figs 51-53 Cothurnia ceramicola; fig. 51 after Precht, 1935, bar = 50 wm; figs 52 & 53 after Kahl, 1933. Figs 54-56 Cothurnia clausiens, after Stiller, 1951, bar = 25 um. Figs 57 & 58 Cothurnia coarctata; fig. 57 after Felinska, 1965, bar = 50 wm; fig. 58 after Kahl, 1935. ALAN WARREN & JAN PAYNTER atte" ~ A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES External stalk and endostyle short; mesostyle short, broad and with conspicuous longitudinal striae. Two zooids may be present per lorica, each 80 wm long X 15 wm wide and extending up to one third of its length beyond aperture. Peristomial lip 20 4m in diameter. Disc convex. Macronucleus vermiform, extending almost entire length of zooid. Pellicular striations conspicuous. HABITAT. Marine, originally found as an epibiont of the alga Ceramium (Kahl, 1933). Also found attached to the bryozoa Crisia eburnea and Cribilina punctata, the polychaete Spirorbis spirorbis and the cnidarian Laomedea loveni (Precht, 1935), and by Felinska (1965) attached to algae collected at sea and from laboratory aquaria at Plymouth. Note. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia clausiens Stiller, 1951 DESCRIPTION (Figs 54-56). Lorica asymmetrical, 35-50 um long X 20-35 wm wide. External stalk about one third lorica length, slender and curved; endostyle short; mesostyle absent. Two zooids may be present per lorica, each 25—35 wm long when contracted. CV lies near centre of zooid. Macronucleus vermiform and situated longitudinally in body. Pellicle with fine striations. HABITAT. Freshwater, originally found attached to arthropods from Lake Balaton (Stiller, 1951). Note. Stiller (1951) stated that the CV is filled by a second CV nearby. This second CV is most probably a canal in which water collects before passing into the CV. Furthermore, if coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia coarctata Kahl, 1933 DESCRIPTION (Figs 57 & 58). Lorica 66-75 um long X 30 wm wide with transverse ridges and furrows. Aperture 20-25 um in diameter. External stalk slender, 15—20 xm long; mesostyle and endostyle both short. Zooid 85 wm long X 25 wm wide, extending about one quarter of its length beyond aperture. Peristomial lip 30 «wm in diameter. Disc flat and obliquely raised. Macronucleus straight and lies longitudinally in body. Pellicular striations conspicuous. HABITAT. Marine, originally found attached to debris from marine aquaria at Kiel (Kahl, 1933). NOTE. C. coarctata was redescribed by Felinska (1965) from specimens found on green algae at Plymouth. 31 Cothurnia cohni (Cohn, 1866) Kent, 1881 Cothurnia pupa Cohn, 1866 DESCRIPTION (Fig. 74). Lorica 60 wm long xX 20 wm wide and pale red in colour. Upper two thirds of lorica conical, lower third rounded with several conspicuous annular ridges and furrows. External stalk broad, 15 wm long; endostyle 50 wm long and with transverse striae; mesostyle absent. Zooid 50 pm long extending just beyond aperture, slender and with an equatorial swelling. Peristomial lip well developed, 15-20 wm in diameter. HABITAT. Marine. NoTE. C. pupa Cohn, 1866 was renamed C. cohni by Kent (1881) in order to solve the problem of homonymy with C. pupa Eichwald, 1849. Cothurnia collaris Kahl, 1933 C. collaris var. incisa Felinska 1965 DESCRIPTION (Figs 59-61). Lorica rotund, 60-100 um long x 30-45 wm wide. Aperture 15 wm wide and inclined at an angle oblique to the main lorica axis. Neck short and curved. Aperture border occasionally with cleft. External stalk short and slender; mesostyle short and broad with conspicuous longitudinal striae; endostyle short. Zooid cylindrical, 70-150 wm long X 15-25 um wide, extending up to one third of its length beyond aperture. Diameter of peristomial lip slightly greater than maximum body width. Macronucleus elongate and lies longitudinally in centre of body. Pellicular striations inconspicuous. HABITAT. Marine or brackish, originally found attached to algae from laboratory tanks at Kiel (Kahl, 1933). NoTE. C. collaris was redescribed by Stiller (1939) and Felinska (1965). Cothurnia complanata Precht, 1935 DESCRIPTION (Figs 62 & 63). Lorica 78-81 wm long x 32-35 pm wide. External stalk slender, 12-15 wm long; mesostyle short and broad; endostyle short and inconspicuous. Zooid 80 wm long X 24 wm wide and extends just beyond aperture. Peristomial lip 30 um in diameter. CV lies just beneath peristome. Macronucleus straight and lies longitudinally in zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally found as an epibiont of the polychaete Stylarioides plumosus (Precht, 1935). Figs 59-61 Cothurnia collaris, bar = 100 ym; figs 59 & 60 composite after Felinska, 1965; fig. 61 after Kahl, 1935. Figs 62 & 63 Cothurnia complanata, after Precht, 1935, bar = 100 um; fig. 62 ventral view; fig. 63 lateral view. Figs 64-68 Cothurnia compressa; figs 64 & 65 after Claparede and Lachmann, 1858, bar = 100 um; figs 66 & 67 after Kahl, 1935 (called Cothurnia compressula); fig. 68 after Kahl, 1935 (called Cothurnia flexa). Figs 69 & 70 Cothurnia curva; fig. 69 composite after Kent, 1881 (called Cothurnia gracilis); fig. 70 after Stein, 1867, bar = 50 wm. Figs 71-73 Cothurnia curvula; figs 71 & 72 after Entz, 1884, bar = 50 wm; fig. 73 after Stiller, 1971. ALAN WARREN & JAN PAYNTER NY m7 ay itil! \ A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES Cothurnia compressa Claparede and Lachmann, 1858 Cothurnia compressa var. flexa Wailes, 1928 Cothurnia compressula (Wailes, 1928) Kahl, 1933 Cothurnia flexa Kahl, 1935 Sincothurnia compressa (Claparede and Lachmann, 1858) Jankowski, 1985 Sincothurnia compressula (Wailes, 1928) Jankowski, 1985 DESCRIPTION (Figs 64-68). Lorica 100-140 wm long X 50 wm wide. Aperture elliptical when viewed from above, 45 wm X 15 pm. Aperture border with two deep clefts. External stalk short, endostyle short and broad, mesostyle absent. Con- tracted zooid 110 um long x 30 um wide. CV situated one quarter of way down zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally found attached to bryozoa and algae (Claparede and Lachmann, 1858). NOTE. Stiller (1939) and Felinska (1965) described cothurnids which they identified as C. compressa, although both of these organisms differ from Claparede and Lachmann’s (1858) C. compressa in that they possess a mesostyle. C. compressa Stiller, 1939 appears to be synonymous with C. auriculata, while C. compressa Felinska, 1965 closely resembles C. ceramicola. Cothurnia cordylophorae Kahl, 1933 DESCRIPTION (Fig. 75). Lorica cylindrical, 120 wm long x 35 pm wide. Aperture 35 wm in diameter. External stalk 40 wm long; mesostyle short and broad; endostyle short and incon- spicuous. Zooid elongate, 160 wm long x 15 wm wide and extends one sixth of its length beyond aperture. Pellicular striations inconspicuous. HABITAT. Brackish water, originally found as an epizoite of the cnidarian Cordylophora sp. (Kahl, 1933); also found by Precht (1935) on Cordylophora caspia. Cothurnia curva Stein, 1867 Cothurnia gracilis Kent, 1881. DESCRIPTION (Figs 69 & 70). Lorica curved, 70-100 um long x 35-40 um wide, colourless when young becoming red when mature. Aperture 15—20 um in diameter. Neck region just below aperture narrow and curved. External stalk broad; endostyle slender; mesostyle absent. Endostyle and external stalk with transverse striae. Zooid 90 wm long xX 15 wm wide and extends up to one third of its length beyond aperture. Macronucleus short and lies longitudinally in centre of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found attached to Entomostraca (Stein, 1867); also occurs as an epizoite of the Fig. 74 Cothurnia cohni, after Kahl, 1935, bar = 50 wm. Fig. 75 Cothurnia cordylophorae, after Kahl, 1935, bar = 100 um. 38 affinis crayfish Astacus leptodactylus and Cambarus (Krucinska and Simon, 1968). Cothurnia curvula Entz (1876), 1884 Cothurnia imberbis var. curvula Entz, 1876 DESCRIPTION (Figs 71-73). Lorica 60 wm long x 30 wm wide, occasionally with three centrally located annular furrows. Aperture 15 jm in diameter. External stalk short with bulbous thickening at its point of attachment to lorica. Mesostyle and endostyle absent. Zooid 60 um long x 8-10 jum wide and extends just beyond aperture. Macronucleus straight and lies longitudinally in zooid. Pellicular striations inconspicuous. HABITAT. Marine, found attached to harpacticoid copepods from the Gulf of Neapel (Entz, 1884). Cothurnia cyathiforme Stiller, 1939 Cothurnia compressa var. cyathiformis Felinska, 1965 Sincothurnia cyathiformis (Stiller, 1939) Jankowski, 1985 DESCRIPTION (Figs 76-78). Lorica asymmetrical, 95-100 wm long X 45 um wide. Aperture 30 wm wide and with shallow cleft in border. External stalk and endostyle short; mesostyle broad with conspicuous longitudinal striae. Two zooids per lorica, each 70 wm long when contracted. Pellicular striations conspicuous. HABITAT. Marine. NoTE. The data above are based on observations of contracted specimens only. Uncontracted specimens of C. cyathiforme have yet to be described. Cothurnia cyclopis Kahl, 1933 DESCRIPTION (Fig. 79). Lorica 60 wm long X 25 wm wide with irregular ridges and furrows. External stalk slender 15 um long; mesostyle short and broad; endostyle short and slender. Zooid 80 xm long X 20 wm wide, extending up to one third of its length beyond aperture. Pellicular striations inconspicuous. HABITAT. Marine, originally found attached to harpacticoid copepods (Kahl, 1933). Cothurnia cylindrica Sommer, 1951 DESCRIPTION (Fig. 80). Lorica cylindrical, 61 wm long x 29 pm wide. Aperture 27 um in diameter. External stalk 5 wm long with a broad cylindrical swelling at its point of attachment to lorica. Two zooids per lorica, each 81-95 wm long x 13-16 jm wide and extending up to one fifth of its length beyond Figs 76-78 Cothurnia cyathiforme; figs 76 & 77 after Stiller, 1939, bar = 50 wm; fig. 78 after Felinska, 1965 (called Cothurnia compressa var. cyathiformis). Fig. 79 Cothurnia cyclopis, after Kahl, 1933, bar = 50 um. Fig. 80 Cothurnia cylindrica, after Sommer, 1951, bar = 50 um. Figs 81 & 82 Cothurnia cypridicola, after Kahl, 1933, bar = 50 um. Fig. 83 Cothurnia cytherideae, after Kahl, 1933, bar = 50 um. ALAN WARREN & JAN PAYNTER CTI AANA tl! (TN \ A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES aperture. Peristomial lip 20-22 wm in diameter. Disc obliquely elevated. CV situated just below peristome. Pellicular stria- tions inconspicuous. HABITAT. Freshwater, originally found attached to Enteromorpha intestinalis (Sommer, 1951). Note. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia cypridicola Kahl, 1933 DESCRIPTION (Figs 81 & 82). Lorica 50-60 wm long 20-25 fm wide and with irregular ridges and furrows. Aperture ovoid when viewed from above, 25 wm X 15 um. External stalk short; mesostyle and endostyle short and broad. Zooid elongate, 70-80 um long x 10-12 wm wide and extending up to one third of its length beyond aperture. Pellicular striations inconspicuous. HABITAT. Marine or brackish water, originally found on cyprids (Kahl, 1933); also found on the gastropods Hydrobia ulvae and H. jenkinsi, and on the ostracod Cyprideis litoralis (Precht, 1935). Cothurnia cytherideae Kahl, 1933 DESCRIPTION (Fig. 83). Lorica cylindrical, 70 wm long x 35 pm wide and rounded posteriorly. Aperture 30 um in dia- meter. External stalk slender 35 um long; mesostyle broad; endostyle slender. Zooid ovoid when contracted, 45 wm long x 20 wm wide. Macronucleus curved and lies longitudinally in zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally found as an epibiont of the hydroid Cytheridea (Kahl, 1933). Cothurnia elegans Stiller, 1940 DESCRIPTION (Fig. 84). Lorica 65—75 um long X 30 wm wide. Aperture 20 wm in diameter. External stalk 6-10 um long; mesostyle short and broad; endostyle absent. Zooid 65-75 pm long X 10-30 um wide with centrally located annular ridge. Zooid just reaches aperture when extended. Peristome 30 zm in diameter. CV large and situated just below peristome. Macronucleus curved anteriorly and lies longitudinally in the body. Pellicular striations conspicuous. HABITAT. Marine or freshwater, originally found attached to algae from Lake Holstein (Stiller, 1940); also isolated from laboratory aquaria at Plymouth attached to green algae (Felinska, 1965). Fig. 84 Cothurnia elegans, after Stiller, 1940, bar = 50 um. Fig. 85 Cothurnia elongata, after Felinska, 1965, bar = 100 um. 35 Cothurnia elongata Felinska, 1965 Sincothurnia elongata (Felinska, 1965) Jankowski, 1985 DESCRIPTION (Fig. 85). Lorica 150 wm long X 65 wm wide. Aperture 40-45 wm wide. External stalk 30-40 um long; mesostyle short; endostyle short with conspicuous longitudinal striae. Zooid 150 wm long x 45 wm wide and just reaches aperture when extended. Peristomial lip 45 wm in diameter. CV lies near the centre of body. Macronucleus straight and lies longitudinally in anterior part of zooid. Pellicular striations conspicuous. HABITAT. Marine, originally isolated from laboratory aquaria at Plymouth (Felinska, 1965). Cothurnia entzi Kahl, 1933 Cothurnia nodosa Entz, 1884 DESCRIPTION (Fig. 86). Lorica 70-80 um long xX 30-40 um wide and with two or three centrally located annular furrows. Aperture border with shallow cleft. External stalk slender, up to 50 wm long and with bulbous swelling at point of attach- ment to lorica; endostyle short and slender; mesostyle absent. Two zooids per lorica, each 85—90 wm long X 20 wm wide and extending up to one quarter of its length beyond aperture. Peristomial lip well developed, 25 sm in diameter. CV situated near centre of infundibulum. Macronucleus straight or slightly curved and lies longitudinally in zooid. Pellicular striations conspicuous with convex ribbing between striations. HABITAT. Marine. Cothurnia fecunda Stokes, 1893 DESCRIPTION (Fig. 87). Lorica cylindrical, 110 wm long x 35 jm wide. Aperture 30 wm in diameter. External stalk 15 um long; endostyle inconspicuous; mesostyle absent. Two zooids per lorica, each 80 wm long X 20 wm wide when contracted, although only one zooid may contract completely at any time. Pellicular striations conspicuous. HABITAT. Brackish water, originally found attached to fila- mentous algae in North American canal water (Stokes, 1893). NoTE. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Uncontracted specimens of C. fecunda have yet to be described. Cothurnia felinska (Felinska, 1965) n. sp. Cothurnia compressa var. compressula Felinska, 1965 DESCRIPTION (Figs 88 & 89). Lorica 110 wm long x 40 um Fig. 86 Cothurnia entzi, after Entz, 1884, bar = 50 xm (called Cothurnia nodosa). Fig. 87 Cothurnia fecunda, after Stokes, 1893, bar = 100 um. Figs 88 & 89 Cothurnia felinska, after Felinska, 1965 (called Cothurnia compressa var. compressula); fig. 88 bar = 100 um; fig. 89 nuclei. Fig. 90 Cothurnia fibripes, after Kahl, 1933, bar = 50 um. Fig. 91 Cothurnia floscularia, after Perty, 1852, bar = 100 wm. Figs 92 & 93 Cothurnia gammari, after Precht, 1935, bar = 50 um; fig. 92 ventral view; fig. 93 lateral view. Figs 94 & 95 Cothurnia halacaricola, after Precht, 1935, bar = 50 um; fig. 94 ventral view; fig. 95 lateral view. 36 wide with irregular ridges and furrows. Aperture 25 um wide with two deep clefts in border. External stalk short and broad; mesostyle short; endostyle short and broad with conspicuous longitudinal striae. Two zooids per lorica, each 60-100 wm long X 20-25 wm wide when contracted. Macro- nucleus lies longitudinally in zooid with posterior end curved upwards. Pellicular striations conspicuous. HABITAT. Marine, originally found attached to algae from marine aquaria at Plymouth (Felinska, 1965). Note. Felinska (1965) called this organism Cothurnia compressa var. compressula because it possesses features common to both C. compressa Claparede and Lachmann, 1858 and C. compressula Kahl, 1935. However it also differs from both of these taxa in terms of its nuclear shape, mesostyle and the longitudinal striae on the endostyle. This last feature, along with the shape of the aperture, serves to separate C. felinska from another similar species, C. cyathi- forme Stiller, 1939. Cothurnia fibripes Kahl, 1933 DESCRIPTION (Fig. 90). Lorica 60 wm long xX 30 um wide, rounded posteriorly and with irregular ridges and furrows. Aperture 15 um in diameter. External stalk short and broad; mesostyle broad and with conspicuous longitudinal striae; endostyle absent. Zooid elongate, 90 wm long x 10 wm wide, extending up to one third of its length beyond aperture, and with a centrally located annular ridge. Peristome 15-20 um in diameter. Pellicle with fine striations. HABITAT. Marine, originally isolated from laboratory aquaria at Kiel (Kahl, 1933). Cothurnia floscularia Perty, 1852 DESCRIPTION (Fig. 91). Lorica 85 wm long X 40 um wide. Aperture 30 wm in diameter. External stalk slender, 8 wm long; mesostyle and endostyle absent. Zooid 90 um long x 20 zm wide, extending just beyond aperture. Pellicular striations inconspicuous. HABITAT. Freshwater. Cothurnia gammari Precht, 1935 Sincothurnia gammari (Precht, 1935) Jankowski, 1985 DESCRIPTION (Figs 92 & 93). Lorica compressed dorso- ventrally, 53-55 um long X 25-34 wm wide, occasionally with irregular ridges and furrows. Aperture ovoid when viewed from above, 25 wm X 10 wm. External stalk 10-20 um long; mesostyle short and broad; endostyle absent. Mesostyle and ALAN WARREN & JAN PAYNTER external stalk with conspicuous longitudinal striae. Zooid 65 wm long X 13-22 wm wide, reaching just beyond aperture when extended. Peristomial lip well developed, 15-20 um in diameter. Disc obliquely elevated. Macronucleus situated longitudinally in zooid, usually curved at both ends. Pellicular striations conspicuous. HABITAT. Marine, originally found attached to Gammarus locusta (Precht, 1935); also found attached to G. oceanicus and G. duebeni (Fenchel, 1965). Cothurnia halacaricola Precht, 1935 DESCRIPTION (Figs 94 & 95). Lorica 70 wm long X 23-35 wm wide. Aperture ovoid 28 zm X 17 wm. External stalk 15 um long; mesostyle short and broad; endostyle inconspicuous. All three stalks with conspicuous longitudinal striae. Zooid cylindrical, 70 wm long X 15 um wide and reaches just beyond aperture. Peristomial lip well developed, 20 wm in diameter. CV situated just below peristome. Macronucleus straight and lies longitudinally in body. Pellicular striations conspicuous. HABITAT. Marine, originally found as an epibiont of the halacarid Copidognathus fabriciusi (Precht, 1935). Cothurnia harpactici Kahl, 1933 DESCRIPTION (Figs 96 & 97). Lorica cylindrical, 75-88 m long X 28-35 wm wide, occasionally with three or four annular ridges and furrows. Aperture 30 «m in diameter. External stalk 15 ~m long and sometimes curved; mesostyle short and broad; endostyle short. All three stalks with conspicuous longitudinal striae. Zooid 105 wm long x 17-20 gm wide, extending up to one fifth of its length beyond aperture. Peristomial lip 20 wm in diameter. Macronucleus lies longitudinally in body, proximal end slightly curved. Pellicular striations conspicuous. HABITAT. Marine, originally found as an epizoite of harpacti- coid copepods (Kahl, 1933); also found by Precht (1935) on the harpacticoids Cletocamptus confluens and Mesochra lillgeborgi. Cothurnia inclinans Felinska, 1965 DESCRIPTION (Fig. 98). Lorica compressed dorso-ventrally, 57 wm long X 40 wm and with irregular ridges and furrows. Aperture 25 zm wide with a deep cleft in each lateral border. External stalk long and slender with a distinct swelling at point of attachment to lorica; mesostyle and endostyle absent. Zooid 65 wm long X 25 wm wide, extending up to one third of its length beyond aperture. Peristome 40 wm in diameter. Figs 96 & 97 Cothurnia harpactici, after Precht, 1935, bar = 100 um; fig. 96 ventral view; fig. 97 lateral view. Fig. 98 Cothurnia inclinans, after Felinska, 1965, bar = 50 wm. Fig. 99 Cothurnia inflata, after Stokes, 1893, bar = 50 um. Fig. 100 Cothurnia inflecta, after Stiller, 1939, bar = 50 um. Figs 101-103 Cothurnia innata; fig. 101 after Miller, 1786, bar = 50 wm; figs 102 & 103, after Hofker, 1930. Fig. 104 Cothurnia irregularis, after Fromentel, 1874, bar = 50 um. Figs 105 & 106 Cothurnia lapponum; fig. 105 after Penard, 1922, bar = 100 um; fig. 106 after Wang, 1977 (called Cothurnia lapponum naidongensis). Fig. 107 Cothurnia longipes, after Mereschkowsky, 1879, bar = 100 um. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES 37 38 Disc convex. CV situated just below peristome. Macronucleus straight and lies in upper two thirds of zooid. Pellicular striations inconspicuous. HaBITaT. Marine, originally found attached to algae from marine aquaria at Plymouth (Felinska, 1965). Cothurnia inflata Stokes, 1893 DESCRIPTION (Fig. 99). Lorica 60 wm long X 30 wm wide, rounded posteriorly and with irregular ridges and furrows. External stalk short and slender; endostyle short and with conspicuous longitudinal striae; mesostyle absent. Zooid 75 um long X 12 wm wide, extending up to one fifth of its length beyond aperture. Peristomial lip 15 wm in diameter. Disc obliquely elevated. CV situated just below peristome. Macro- nucleus straight, extending almost entire zooid length. Pellicle with fine striations. HABITAT. Brackish water, originally found attached to fila- mentous algae from Coney Island, New York (Stokes, 1893). Cothurnia inflecta Stiller, 1939 DESCRIPTION (Fig. 100). Lorica 70 wm long xX 35 wm wide with irregular ridges and furrows. Aperture 25 wm in diameter. External stalk slender, 10 wm long; mesostyle short and broad; endostyle short and slender. Zooid 70 wm long Xx 20 jm wide and reaches just beyond aperture. Peristomial lip 25 um in diameter. Macronucleus vermiform. Pellicular striations conspicuous. HABITAT. Marine, originally found attached to green algae from the North Sea near Helgoland (Stiller, 1939). NoTE. The original diagram of this species was labelled ‘Cothurnia inflexa’ (Stiller, 1939); this is assumed to be a misspelling of C. inflecta. Cothurnia innata Miller, 1786 Cothurnia hofkeri (Hofker, 1930) Kahl, 1933 DESCRIPTION (Figs 101-103). Lorica cylindrical, 40-100 um long X 20-27 wm wide. Aperture 10-15 wm in diameter. External stalk 20 mum long; endostyle short and broad; mesostyle absent. Two zooids per lorica, each 40-50 um long x 10-15 wm wide and not reaching as far as aperture. Peristomial lip well developed, 15 4m in diameter. Pellicular striations inconspicuous. HABITAT. Marine, found by Hofker (1930) attached to Hydrobia spp. ALAN WARREN & JAN PAYNTER Cothurnia irregularis Kent, 1881 Cothurnia nodosa Fromentel, 1874 DESCRIPTION (Fig. 104). Lorica 55 wm long X 35 wm wide, anterior end conical, posterior end broad and rounded. Aperture 20 wm in diameter. External stalk 10-15 um long;’ endostyle short and inconspicuous; mesostyle absent. Zooid 55 wm long xX 15 um wide and extending just beyond aperture. Peristome 20 um in diameter. Pellicular striations inconspicuous. HABITAT. Freshwater. Cothurnia kahli Banina and Polyakova, 1977 DESCRIPTION (Fig. 108). Lorica 43—-SO wm long xX 21-29 wm wide. Aperture 18 wm in diameter. External stalk 3-4 wm long and penetrates lorica wall via special tube. Two zooids per lorica, each 50-56 wm long X 15 wm wide and extends just beyond aperture. Peristome 18 4m in diameter. CV lies just below peristome. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found attached to Cladophora (Banina and Polyakova, 1977). NotE. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia lapponum Penard, 1922 Cothurnia lapponum naidongensis Wang Jiaji, 1977 DESCRIPTION (Figs 105 & 106). Lorica cylindrical 92-116 wm long X 30-49 um wide and with regular annular furrows along the entire length with exception of posterior end. External stalk 7-8 wm long; mesostyle and endostyle absent. Two zooids per lorica, each 140 wm long xX 15 wm wide and extends one quarter of its length beyond aperture. Peristome 25 wm in diameter. CV situated near centre of infundibulum. Macronucleus straight and lies longitudinally in zooid. Micro- nucleus large, ovoid and lies below posterior end of macro- nucleus. Pellicle with fine striations. HABITAT. Freshwater, originally found on Sphagnum from Haparanda, Sweden (Penard, 1922); also isolated from marsh water at Naidong, Tibet, 400 m above sea level (Wang Jiaji, 1977). NoTE. C. lapponum naidongensis was described by Wang Jiaji (1977) as a subspecies of C. lapponum. Wang Jiaji cited the shape of the lorica as the main difference between the two taxa. However, no description of an uncontracted C. lapponum naidongensis was given and no mention was made of the endostyle, which is absent in Penard’s animal but appears to Fig. 108 Cothurnia kahli, after Banina and Polyakova, 1977, bar = 50 um. Figs 109-111 Cothurnia lata, after Kellicott, 1883, bar = 50 um; fig. 109 ventral view; fig. 110 aperture, anterior view; fig. 111 lateral view. Figs 112-114 Cothurnia limnoriae, after Dons, 1928; fig. 112 bar = 50 wm; figs 113 & 114 showing variation in lorica and macronucleus. Fig. 115 Cothurnia macrodisca, after Stiller, 1971, bar = 50 wm. Fig. 116 Cothurnia magna, after Yunfen, 1980, bar = 200 um. Fig. 117 Cothurnia membranoloricata, after Stiller, 1968, bar = 50 um. Figs 118-121 Cothurnia maritima; fig. 118 after Felinska, 1965, bar = 50 wm; fig. 119 after Jankowski, 1965 (called Cothurnia cyathus); fig. 120 after Andrussowa, 1886 (called Cothurnia marina); fig. 121 after Ehrenberg, 1838. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES Nill if “OH Oy y| ) M1) es 39 40 be present in Wang Jiaji’s. Until more data are available there are not considered to be sufficient differences between C. lapponum and C. lapponum naidongensis for their recognition as-separate taxa. Cothurnia lata Kellicott, 1883 Cothurnia lata (Wailes, 1928) Kahl, 1935 DESCRIPTION (Figs 109-111). Lorica compressed dorso- ventrally, 70 wm long x 40-45 wm wide x 20 um deep. Aperture elliptical when viewed from above. External stalk 20-25 wm long and curved; endostyle short; mesostyle absent. External stalk and endostyle with transverse furrows. Zooid 70 wm long X 20 wm wide and just reaches aperture. CV situated at base of infundibulum. Macronucleus short, C- shaped and lies in centre of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found as an epibiont of the copepod Diaptomus (Kellicott, 1883b). Cothurnia limnoriae Dons, 1928 Tesnotheca limnoriae (Dons, 1928) Jankowski, 1987 DESCRIPTION (Figs 112-114). Lorica curved, 70-90 um long x 33-40 wm wide. Aperture 30-40 um in diameter. External stalk 20-40 um long; mesostyle short and broad; endostyle absent. Two zooids per lorica, each 60 wm long x 40 um wide. Macronucleus C-shaped, 30 um long. Pellicular striations inconspicuous. HABITAT. Marine, originally found as an epizoite of Limnoria lignorum (Dons, 1928). Note. Jankowski (1987) erected the genus Tesnotheca for C. limnoriae on the basis that following binary fission the zooid grows asymmetrically with the peristome inclined to one side, whereas typical Cothurnia zooids are symmetrical. The symmetry of the zooid, however, has never previously been used for separating peritrich genera, and furthermore no description or diagram of an uncontracted zooid of C. limnoriae has been found in the literature. Therefore, C. limnoriae is retained in the genus Cothurnia. Cothurnia longipes Kellicott, 1894 Cothurnia longipes (Mereshkowsky, 1879) Kahl, 1935 Cothurnia nodosa var. longipes Mereschkowsky, 1879 DESCRIPTION (Fig. 107). Lorica 100 wm long x 35 wm wide. Aperture 30 wm in diameter. External stalk slender, 100-120 um long; endostyle slender, 20-40 um long; mesostyle absent. Zooid 55 wm long X 25 wm wide when contracted. Pellicular striations inconspicuous. Fig. 122 Cothurnia minutissima, after Penard, 1914, bar = 50 wm. Fig. 123 Cothurnia mobiusi, after Stiller, 1939, bar = 50 um. Fig. 124 Cothurnia monoannulata, after Banina and Polyakova, 1977. Fig. 125 Cothurnia nereicola, after Precht, 1935, bar = 50 wm. ALAN WARREN & JAN PAYNTER HABITAT. Marine, originally isolated from the White Sea (Mereschkowsky, 1879); also found on sea-weeds and polyzoa from tide pools in North America (Kellicott, 1883a). Note. This taxon was first described by Mereschkowsky (1879) under the name C. nodosa var. longipes, although it was renamed C. longipes by Kahl (1935). In the meantime, however, Kellicott (1894) isolated a similar organism from North America which he called Cothurnia longipes. Kellicott (1894) made no reference to Mereschkowsky’s (1879) work. The two organisms appear to be synonymous thus solving the problem of homonymy. Cothurnia macrodisca Stiller DESCRIPTION (Fig. 115). Lorica 75 wm long xX 40 wm wide. Aperture 35 wm in diameter. External stalk 5-6 um long, attached to substratum via conspicuous ring-like basal disc; mesostyle and endostyle absent. Zooid 90 wm long x 15 wm wide, extending up to one sixth of its length beyond aperture. Peristomial lip well developed, 25 um in diameter and inclined at angle to main body axis. Disc convex. CV situated just below peristome. Macronucleus vermiform and lies longitudinally in zooid. Pellicle with fine striations. HABITAT. Freshwater, originally found attached to Spirogyra (see Stiller, 1971). Note. The original description of C. macrodisca could not be located. Stiller (1971) cited Stiller as the authority for this species but did not give a date or reference. Cothurnia magna Yunfen, 1980 DESCRIPTION (Fig. 116). Lorica 200 wm long X 56 «wm wide, cylindrical and tapering posteriorly towards stalk. Aperture 56 wm in diameter. External stalk 95 um long X 3 wm wide; endostyle short and slender; mesostyle absent. Zooid elong- ate, 280-300 um long X 25 wm wide and extending up to one third of its length beyond aperture. Peristomial lip 52 wm in diameter. Disc convex. CV situated just below peristome. Macronucleus vermiform and extends almost entire body length. Pellicle with fine striations. HABITAT. Freshwater, originally isolated from Lake Dong, China (Yunfen, 1980). Cothurnia maritima Ehrenberg, 1838 Cothurnia cyathus Jaworowski, 1893 Cothurnia marina Andrussowa, 1886 DESCRIPTION (Figs 118-121). Lorica 35-56 wm long x 28-33 wm wide, occasionally with irregular ridges and furrows. Aperture 25 wm X 17 wm. External stalk 10-15 um long; Figs 126 & 127 Cothurnia nitocrae, after Precht, 1935, bar = 50 wm; fig. 126 ventral view; fig. 127 lateral view. Figs 128 & 129 Cothurnia nodosa; fig. 128 after Claparede and Lachmann, 1858, bar = 50 wm; fig. 129 after Kahl, 1935, bar = 50 wm. Fig. 130 Cothurnia obliqua, after Bock, 1952, bar = 100 wm. Fig. 131 Cothurnia oblonga, after Kahl, 1935, bar = 50 wm. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES UT} [ACHING tan HI na {( Ii) qq wii : 4] 42 endostyle short and broad; mesostyle absent. Zooid 60 wm long X 12-15 um wide, extending up to one third of its length beyond aperture. Peristomial lip 20-25 um in diameter. CV situated near centre of infundibulum which is about one third zooid length. Macronucleus straight and lies longitudinally in body. Pellicle with fine striations. HABITAT. Marine, attached to a variety of substrates. Note. Several authors (Quennerstedt, 1865; Andrussowa, 1886; Mobius, 1888; Kahl, 1935; Stiller, 1939; Felinska, 1965) have described cothurnias under this name. The descriptions given by Quennerstedt (1865) and Felinska (1965) conform with Ehrenberg’s (1838) original description, whereas those given by Mobius (1888), Kahl (1935) and Stiller (1939) show several differences; C. maritima sensu Mobius, 1888 was redescribed by Stiller (1939) under the name C. mobiusi mihi (= C. mobiusi); C. sp. (‘bei maritima’) Kahl, 1935 = C. obliqua Bock, 1952; and C. maritima Stiller, 1939 possesses a mesostyle whereas C. maritima Ehrenberg, 1838 does not. Cothurnia membranoloricata Stiller, 1968 DESCRIPTION (Fig. 117). Lorica 40-45 um long x 18-23 wm wide. Aperture 20 wm in diameter. External stalk 8-12 um long and attached to substrate via basal disc, 4-6 wm in diameter; endostyle short and broad; mesostyle absent. Zooid 55-57 um long X 12-15 wm wide, extending up to one quarter of its length beyond aperture. Peristomial lip well developed, 5 wm thick xX 15—18 um in diameter. Disc convex. CV lies just below peristome. Pellicle clearly striated with 14-15 striations per 10 um. HABITAT. Marine, originally found attached to Cladophora repens and C. heteronema in Yugoslavian coastal waters (Stiller, 1968). Note. C. membranoloricata was found by Viljoen and van As (1983) attached to the alga Spirogyra; it was misspelt C. membraniloricata. Cothurnia minutissima (Penard, 1914) Kahl, 1935 Cothurnia sinuata Kahl, 1933 Cothurniopsis minutissima Penard, 1914 DESCRIPTION (Fig. 122). Lorica 35-50 wm long X 18-27 wm wide and rounded posteriorly. Aperture 10-12 um wide. External stalk slender, 6-17 wm long; endostyle short and slender; mesostyle absent. Zooid 50-55 um long x 10 um wide, extending up to one third of its length beyond aperture. Peristomial lip well developed, 10 wm in diameter. Disc convex. CV large and situated near base of infundibulum. Fig. 132 Cothurnia ovalis, after Kahl, 1933, bar = 50 um. ALAN WARREN & JAN PAYNTER Macronucleus straight and lies longitudinally in posterior half of zooid. Pellicle with fine striations. HABITAT. Freshwater, originally found attached to moss (Penard, 1914); also found on detritus (Kahl, 1935) and on Enteromorpha intestinalis (Sommer, 1951). Cothurnia mobiusi (Mobius, 1888) Stiller, 1939 Cothurnia maritima Mobius, 1888 Cothurnia mobiusi mihi (MObius, 1888) Stiller, 1939 DESCRIPTION (Fig. 123). Lorica 50 wm long X 30 wm wide. Aperture 20 wm in diameter and sometimes inclined at an angle to main lorica axis. External stalk short; mesostyle and endostyle absent. Zooid 70 wm long Xx 15 wm wide, extending up to one half of its length beyond aperture. Peristomial lip 20 jm in diameter. CV situated one quarter of way down zooid. Macronucleus C-shaped and lies transversely in centre of zooid. Pellicular striations inconspicuous. HABITAT. Marine, isolated by Stiller (1939) from the North Sea near Helgoland. Note. Stiller (1939) described a Cothurnia which conforms closely with C. maritima Mobius (1888) but clearly differs from Ehrenberg’s (1838) C. maritima. Stiller (1939) called this organism C. mobiusi mihi, although the subspecies epithet mihi may be disregarded. Cothurnia monoannulata Banina and Polyakova, 1977 Sincothurnia monoannulata (Banina and Polyakova, 1977) Jankowski, 1985 DESCRIPTION (Fig. 124). Lorica cylindrical, 68 wm long x 32 jm wide and with a centrally located annular ridge. Aperture 29 wm in diameter. External stalk 4 wm long and penetrates lorica wall via special tube; mesostyle and endostyle absent. Two zooids per lorica, one larger than the other; larger zooid typically curved anteriorly, 90 wm long X 15 wm wide and extends up to one third of its length beyond aperture; smaller zooid not curved, 80 wm long X 12 wm wide and extends only one sixth of its length beyond aperture. Peristomial lip 18—20 jm in diameter. CV situated just beneath peristome. Pelli- cular striations conspicuous with convex ribbing between the striations. HABITAT. Freshwater, originally found attached to Cladophora (Banina and Polyakova, 1977). Cothurnia nereicola Precht, 1935 Semicothurnia nereicola (Precht, 1935) Jankowski, 1976 Figs 133 & 134 Cothurnia ovata, after Fromentel, 1874; fig. 133 bar = 100 um; fig. 134 macronucleus. Fig. 135 Cothurnia oviformis, after Banina and Polyakova, 1977, bar = 50 um. Fig. 136 Cothurnia parva, after Bock, 1952, bar = 50 wm. Fig. 137 Cothurnia parvula, after Felinska, 1965, bar = 50 um. Fig. 138 Cothurnia patula, after Fromentel, 1874, bar = 50 um. Figs 139 & 140 Cothurnia peloscolicis, after Precht, 1935, bar = 50 um; fig. 139 ventral view; fig. 140 lateral view. Fig. 141 Cothurnia plachteri, after Matthes and Guhl, 1973, bar = 100 um. Fig. 142 Cothurnia plectostyla, after Stokes, 1885, bar = 100 wm. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES as UT TEE Nan \\\\ AT \ 43 44 DESCRIPTION (Fig. 125). Lorica almost hemispherical in shape, 33 um long X 30 wm wide. Aperture 30 um in diameter. External stalk short and attached to substratum via basal disc, 10 wm in diameter; mesostyle and endostyle short and broad. All three stalks with continuous longitudinal striae. Zooid 65 wm long X 18 wm wide, extending up to two thirds of its length beyond aperture. Peristomial lip well developed, 20 wm in diameter. Disc convex. CV situated just beneath peristome. Macronucleus straight and lies longitudinally in zooid. Pellicular striations conspicuous. HABITAT. Marine, originally found as an epizoite of the polychaete Nereis diversicolor (Precht, 1935). Note. Jankowski (1976) erected the genus Semicothurnia for this species (and for C. acuta Levander, 1915) on account of their unusual stalks. The stalk of C. nereicola is not, however, considered to be sufficiently distinct for the separation of a new genus. Cothurnia nitocrae Precht, 1935 DESCRIPTION (Figs 126 & 127). Lorica 93-97 um long x 30-35 um wide. Aperture oval when viewed from above, 30 wm X 15 wm. External stalk, mesostyle and endostyle all short with continuous longitudinal striae. Zooid 95 wm long xX 20 wm wide and extends just beyond aperture. Peristomial lip 25 wm in diameter. CV situated near anterior end of macronucleus which is straight and 70 wm long. Pellicular striations conspicuous. HABITAT. Marine, originally found as an epizoite of the harpacticoid copepod Nitoura spinipes (Precht, 1935). Cothurnia nodosa Claparede and Lachmann, 1858 Cothurnia sahrhagei (Sahrhage) Kahl, 1933 DESCRIPTION (Figs 128 & 129). Lorica 58 wm long xX 26 wm wide and with one or more annular ridges. Aperture 20 um in diameter. External stalk slender, 15-55 um long; endostyle slender; mesostyle absent. Two zooids per lorica, each 70 wm long X 15-18 wm wide and extending up to one third of its length beyond aperture. Peristomial lip 20 wm in diameter. CV situated beneath the peristome. Pellicle with fine striations. HABITAT. Marine, originally found attached to algae (Claparede and Lachmann, 1858). Cothurnia obliqua Bock, 1952 Cothurnia sp. Kahl, 1933 DESCRIPTION (Fig. 130). Lorica 95-96 wm long xX 44-45 um wide. Aperture 37—38 um in diameter. External stalk 5-7 wm long; mesostyle short and broad; endostyle short and incon- spicuous. Zooid 110-115 ym long x 28-30 wm wide, extend- ing up to one fifth of its length beyond aperture. Peristomial lip 35 ym in diameter and inclined at an angle to main zooid axis. CV situated just below peristome. Macronucleus vermi- form and curved at both ends. Pellicle with fine striations. HABITAT. Brackish water, found by Kahl (1933) attached to green algae, and by Bock (1952) attached to Ceramium diapharum, both at Kiel. Note. Kahl (1935) described a cothurnid (Cothurnia spec.) which he considered to be closely related to C. maritima ALAN WARREN & JAN PAYNTER Ehrenberg, 1838. Bock (1952) subsequently isolated C. obliqua from the same location, noting the similarity between his and Kahl’s (1933) organisms. The two are considered synonymous. Cothurnia oblonga Kahl, 1935 DESCRIPTION (Fig. 131). Lorica 71-82 wm long x 30-33 wm wide. Aperture 28-30 wm in diameter. External stalk 5 um long; endostyle short and broad; mesostyle absent. Two zooids per lorica, 90-110 wm long X 18 wm wide. One zooid typically larger than the other, the larger zooid extending up to one third of its length beyond aperture while the smaller one extends only one fifth. CV situated just below peristome. Macronucleus straight, about 70 ym long. Pellicular striations conspicuous. HABITAT. Freshwater, attached to a variety of plant and animal substrates including Lemna, Cladophora and Enteromorpha intestinalis (Kahl, 1935; Sommer, 1951). Note. C. oblonga was redescribed by Sommer (1951). Cothurnia ovalis (Wailes, 1928) Kahl, 1933 Cothurnia innata Wailes, 1928 DESCRIPTION (Fig. 132). Lorica 40 wm long X 22 um wide. Aperture oval when viewed from above, 27 wm X 14 wm. External stalk up to 27 wm long; mesostyle and endostyle absent. Zooid 50-58 um long x 12-18 wm wide, extending up to one third of its length beyond aperture. Peristomial lip 22 jm in diameter. Macronucleus short, curved and lies longi- tudinally in centre of zooid. Pellicular striations inconspicuous. HABITAT. Marine, originally isolated from Departure Bay, Vancouver (Wailes, 1928). Cothurnia ovata Dujardin, 1841 Cothurniopsis ovata (Dujardin, 1841) Zelinka, 1928 DESCRIPTION (Figs 133 & 134). Lorica cylindrical, 125 «wm long X 75 wm wide. Aperture 75 wm in diameter. External stalk short and broad; mesostyle and endostyle absent. Zooid 175-200 wm long X 25 «wm wide, extending up to one third of its length beyond aperture. Peristomial lip 50 wm in diameter. Macronucleus elongate. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found attached to Conferva (Fromentel, 1874). NoTE. Zelinka (1928) transferred C. ovata to Cothurniopsis (= Cothurnopsis) Entz, 1884, but with the submergence of Cothurnopsis by Kahl (1935), this species returned to the genus Cothurnia. Cothurnia oviformis Banina and Polyakova, 1977 DESCRIPTION (Fig. 135). Lorica asymmetric with one side flattened, 48 xm long X 22 «wm wide and tapers at both ends. Aperture 16 «4m wide. External stalk 5 wm long X 4 wm wide and penetrates lorica wall via special tube; mesostyle and | endostyle absent. Zooid 73 wm long X 12 wm wide, extending up to one half of its length beyond aperture. Peristomial lip 15 jm in diameter. CV situated just below peristome. Pellicular striations inconspicuous. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES HABITAT. Freshwater, originally found attached to Cladophora (Banina and Polyakova, 1977). Cothurnia parva Bock, 1952 DESCRIPTION (Fig. 136). Lorica 46-50 um long x 30-34 wm wide. Aperture 23-25 um in diameter. External stalk 11-14 pum long; mesostyle short and broad; endostyle inconspicuous. All three stalks with conspicuous longitudinal striae. Zooid 49-54 xm long X 17 wm wide, extending just beyond aperture. Peristomial lip 19-21 um in diameter. CV situated just below peristome. Macronucleus straight and situated in upper two thirds of zooid. Pellicular striations conspicuous. HABITAT. Brackish water, originally found attached to Ceramium diaphanum (Bock, 1952). Cothurnia parvula Felinska, 1965 DESCRIPTION (Fig. 137). Lorica 50-62 um long X 24 wm wide with irregular ridges and furrows. Aperture 15 wm wide and with deep cleft in border. External stalk, mesostyle and endostyle all short and broad; mesostyle with conspicuous longitudinal striae. Zooid 55 «wm long X 13 wm wide, extend- ing just beyond aperture. Peristomial lip 17 «m in diameter. CV small and situated just below peristome. Macronucleus 20 pm long and lies longitudinally in centre of zooid. Pellicular striations conspicuous. HABITAT. Marine, originally found attached to algae from laboratory aquaria at Plymouth (Felinska, 1965). Cothurnia patula Fromentel, 1874 Sincothurnia patula (Fromentel, 1874) Jankowski, 1985 DESCRIPTION (Fig. 138). Lorica 55 wm long X 45 um wide. Aperture 55 xm in diameter. External stalk short and broad; mesostyle and endostyle absent. Two zooids per lorica, each 110 xm long X 20 wm wide and extending up to one half of its length beyond aperture. Peristomial lip 40 wm in diameter. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found attached to algae (Fromentel, 1874). _ Cothurnia peloscolicis Precht, 1935 ee DESCRIPTION (Figs 139 & 140). Lorica compressed dorso- ventrally, 81 wm long < 27-41 «wm wide. Aperture oval when viewed from above, 15 wm xX 33 wm. External stalk 20 wm long; mesostyle short and broad; endostyle absent. External stalk and mesostyle with conspicuous longitudinal striae. Zooid 80 wm long X 18-22 um wide, extending just beyond aperture. Peristomial lip 27 wm in diameter. CV small and situated in peristomial region. Macronucleus straight, 50 wm long. Pellicular striations inconspicuous. HABITAT. Marine, originally found as an epizoite of the oligochaete Peloscolex benedeni (Precht, 1935). Cothurnia plachteri Matthes and Guhl, 1973 DESCRIPTION (Fig. 141). Lorica elongate, 127 um long x 30 fm wide and tapering at posterior end. External stalk short and slender; endostyle broad; mesostyle absent. Zooid 118- 45 170 wm long X 26 wm wide, extending up to one third of its length beyond aperture. Peristomial lip 34 um in diameter. CV situated just below peristome. Macronucleus elongate and curved at both ends. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found attached to the cray- fish Astacus fluviatilis and A. torrentium (Matthes and Guhl, 1973) Cothurnia plectostyla Stokes, 1885 DESCRIPTION (Fig. 142). Lorica 80-100 um long x 30-40 um wide. Aperture 15-20 wm wide. External stalk 30 um long, curved with lateral furrows; mesostyle broad with conspicuous longitudinal striae; endostyle short with irregular lateral furrows. Zooid 90 wm long X 25 wm wide, extending just beyond aperture. Peristomial lip 25 wm in diameter. CV situated one quarter of way down zooid. Macronucleus ovoid, 20 wm long, and lies in centre of zooid. Pellicle with fine striations. HABITAT. Freshwater, originally found attached _ to Canthocamptus (Stokes, 1885). Cothurnia propinqua Kahl, 1933 DESCRIPTION (Fig. 143). Lorica 60 4m long x 23 wm wide and tapering posteriorly. Aperture 18 4m in diameter. External stalk 10-15 wm long and with irregular transverse furrows; mesostyle slender; endostyle inconspicuous. Zooid 55 wm long X 10 wm wide, extending just beyond aperture. CV situated near base of infundibulum. Macronucleus short, C- shaped and lies obliquely across central region of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater. Cothurnia pupa Eichwald, 1849 DESCRIPTION (Fig. 144). Lorica ovoid, length about x2 width, and with three distinct centrally located annular ridges. External stalk slender, about one quarter of lorica length. Zooid conical in shape, extending just beyond aperture. Peristomial lip broad. Pellicular striations inconspicuous. HABITAT. Freshwater. Norte. No dimensions were given in the original description and C. pupa has never been redescribed. Cothurnia recurva Claparede and Lachmann, 1858 DESCRIPTION (Figs 145 & 146). Lorica curved and irregular, 45-75 wm long X 17-35 wm wide. Aperture 13-17 um in diameter. External stalk up to 18 wm long, curved with transverse furrows; endostyle short and broad; mesostyle absent. Zooid 40-70 um long xX 10-17 um wide, extending just beyond aperture. Diameter of aperture slightly greater than that of lorica. CV small and situated one quarter of way down zooid. Macronucleus straight and lies longitudinally in centre of zooid. Pellicular striations conspicuous. HABITAT. Marine, found as epizoites of cyclopoid and harpac- ticoid copepods (Kahl, 1933; Felinska, 1965). 46 ALAN WARREN & JAN PAYNTER ay flMMCICELMI ii WIN ny Me (ferret K: SJ A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES Cothurnia recurvata Kahl, 1928 DESCRIPTION (Fig. 147). Lorica 100 wm long xX 50 um wide. Aperture 30 um in diameter with curved neck. External stalk slender, 25 um long attached to substratum via basal disc, 15 pm in diameter; endostyle short and broad; mesostyle absent. Zooid 75 wm long X 40 wm wide when contracted. HABITAT. Brackish water, originally found attached to algae (Kahl, 1928). NOTE. C. recurvata is very similar to C. recurva, the principal differences being those of size, host and habitat. It is possible that re-examination will show these taxa to be synonymous. Cothurnia rhabdota Bock, 1952 Sincothurnia rhabdota (Bock, 1952) Jankowski, 1985 DESCRIPTION (Fig. 148). Lorica cylindrical, 125-130 wm xX 40-42 um wide with several (about eight) annular furrows. External stalk 88-93 um long; mesostyle short; endostyle 20-22 wm long. Mesostyle and endostyle with continuous longitudinal striae. Two zooids per lorica, each 55 xm long X 28 wm wide when contracted. Macronucleus short, C-shaped and lies longitudinally in centre of zooid. HABITAT. Marine, originally found attached to the red alga Polysiphonia nigrescens (Bock, 1952). Note. Uncontracted specimens of this species have yet to be described. Cothurnia richtersi (Penard, 1914) Kahl, 1935 Cothurniopsis richtersi Penard, 1914 DESCRIPTION (Figs 150-152). Lorica 45-60 wm long x 30-35 fm wide, rounded posteriorly. Aperture oval when viewed from above, 25 wm X 13 wm. External stalk slender, 20 wm long; mesostyle short and slender; endostyle absent. Two zooids per lorica, each 80 wm long x 20 um wide, extending up to one third of its length beyond aperture. Peristomial lip 20 um in diameter. CV situated near mid-region of infundibu- lum which reaches about one third zooid length. Macronucleus 20 ym long and lies either longitudinally or horizontally in posterior part of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, reported by Penard (1914) from both France and Antarctic regions. Note. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Fig. 143 Cothurnia propinqua, after Kahl, 1933, bar = 50 um. Fig. 144 Cothurnia pupa, after Kent, 1881 (dimensions not available). 47 Cothurnia ruthae Lupkes, 1974 DESCRIPTION (Fig. 149). Lorica conical in shape, 66 ~m long x 30 wm wide, posterior end flattened and broad. Aperture 12 wm in diameter. External stalk slender, 27 wm long; mesostyle short; endostyle absent. Two zooids per lorica, each 93 wm long X 12 wm wide with annular constriction near posterior end. Peristomial lip 10 wm in diameter. Disc obliquely raised above peristome. Macronucleus irregular and lies longitudinally in posterior half of zooid. Pellicle with fine striations. HABITAT. Freshwater, originally isolated from interstitial groundwater from the Fulda Valley (Ltupkes, 1974). Cothurnia sieboldii Stein, 1854 DESCRIPTION (Fig. 153). Lorica 113-132 wm long x 30-49 um wide with two horn-like processes projecting up and back- wards on either side of aperture. External stalk and mesostyle both short and broad; endostyle absent. Two zooids per lorica, each 130 wm long X 30 wm wide. Peristomial lip well developed, 30 «m in diameter. Infundibulum broad, almost reaching centre of zooid. CV situated about one third of way down zooid. Macronucleus short, curved or U-shaped, and lies in centre of zooid. HABITAT. Freshwater, found on Entomostraca (Kahl, 1935) and also on the crayfish Astacus fluviatilis and A. torrentium (Matthes and Guhl, 1973). Note. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia simplex Kahl, 1933 DESCRIPTION (Fig. 154). Lorica 50-60 um long X 23 wm wide. Aperture 18 «wm in diameter. External stalk short and broad with longitudinal striae; mesostyle and endostyle short and inconspicuous. Zooid 90 wm long X 15 wm wide, extending up to one quarter of its length beyond aperture. Peristomial lip 20 wm in diameter. CV situated just below peristome. Macronucleus vermiform and irregularly coiled. Pellicle with fine striations. HABITAT. Marine, originally found attached to algae at Helgoland (Kahl, 1933), and at Plymouth (Felinska, 1965). NOTE. C. simplex was redescribed by Felinska (1965). Figs 145 & 146 Cothurnia recurva; fig. 145 after Felinska, 1965, bar = 50 um; fig. 146 after Claparede and Lachmann, 1858. Fig. 147 Cothurnia recurvata, after Kahl, 1935, bar = 100 um. Fig. 148 Cothurnia rhabdota, after Bock, 1952, bar = 100 um. Fig. 149 Cothurnia ruthae, after Liipkes, 1974, bar = 50 um. Figs 150-152 Cothurnia richtersi, after Penard, 1914; fig. 150 bar = 50 wm; fig. 151 nuclei; fig. 152 lorica with two contracted zooids. Fig. 153 Cothurnia sieboldii, after Matthes and Guhl, 1973, bar = 100 um. Fig. 154 Cothurnia simplex, after Kahl, 1933, bar = 50 um. Fig. 155 Cothurnia subglobosa, after Daday, 1911, bar = 50 um. 48 ALAN WARREN & JAN PAYNTER A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES Cothurnia sinuosa (Wailes, 1943) n. comb. Cothurnia compressa? Kahl, 1935 - Cothurnia compressa var. sinuosa Wailes, 1943 Sincothurnia sinuosa (Wailes, 1943) Jankowski, 1985 DESCRIPTION (Figs 156 & 157). Lorica compressed dorso- ventrally, 90-100 um long x 35-40 um long, tapering at both ends and with annular ridges. Aperture 40 um X 15 wm with two deep clefts. External stalk 10 wm long; endostyle short and inconspicuous; mesostyle absent. Two zooids per lorica, each 150 wm long X 15 wm wide and extending up to one third of its length beyond aperture. Peristomial lip 25 wm in diameter. Pellicular striations inconspicuous. HABITAT. Marine, originally found attached to algae from North America (Wailes, 1943). Cothurnia soldida Vuxanovici, 1962 _ DESCRIPTION (Figs 158 & 159). Lorica 68 wm X 30 «wm wide, cylindrical and rounded posteriorly. Aperture 30 wm in diameter. External stalk short; mesostyle and endostyle as aperture. CV situated near anterior end of macronucleus which is vermiform and curved at both ends. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found on decomposed plant material from Lake Fundeni, Bucharest (Vuxanovici, 1962). Cothurnia spissa Fromentel, 1874 Cothurnia calix (Wailes, 1928) Kahl, 1933 _ DESCRIPTION (Figs 160 & 161). Lorica inverted bell-shaped, 60 um long x 30-35 wm wide and rounded posteriorly. Aperture 36 wm in diameter. External stalk 10-12 um long and slender; mesostyle and endostyle absent. Zooid 40 um long X 15 wm wide when contracted. CV centrally located. Pellicle clearly striated. _ HasIrAarT. Freshwater, originally found attached to Confervae _ (Fromentel. 1874). Note. C. calix (Wailes, 1928) Kahl, 1933 shows only minor differences from C. spissa with which it is here synonymised. _ Cothurnia stylarioides Precht, 1935 _ DESCRIPTION (Fig. 162). Lorica cylindrical, 121 4m long X 32 | wm wide. Aperture 30 wm in diameter. External stalk 12 wm ' long and slender; mesostyle short and broad; endostyle inconspicuous. Zooid elongate, 145 wm long x 20 um in | diameter, extending up to one third of its length beyond Fig. 162 Cothurnia stylarioides, after Precht, 1935, bar = 100 um. Figs 163 & 164 Cothurnia subtilis, after Stiller, 1939, bar = 50 um. absent. Zooid 65 wm long X 20 «wm wide, not reaching as far Figs 156 & 157 Cothurnia sinuosa, after Wailes, 1943, bar = 100 um. Figs 158 & 159 Cothurnia soldida, after Vuxanovici, 1962, bar = 50 um. | Figs 160 & 161 Cothurnia spissa; fig. 160 after Kahl, 1933, bar = 50 um (called Cothurnia calix), fig. 161 after Fromentel, 1874. 49 aperture. Peristomial lip 30 wm in diameter. CV situated just beneath peristome. Pellicular striations inconspicuous. HABITAT. Marine, originally found as an epizoite of the polychaete Stylarioides plumosus (Precht, 1935). Cothurnia subglobosa (Daday, 1911) n. comb. Cothurniopsis subglobosa Daday, 1911 DESCRIPTION (Fig. 155). Lorica ovoid, 45-50 wm long x 46-52 pum wide. Aperture 28-34 um in diameter. External stalk 35— 37 xm long; mesostyle and endostyle absent. Two zooids per lorica. Macronucleus C-shaped and lies longitudinally in centre of zooid. HABITAT. Marine, originally isolated from the Antarctic region as an epizoite of the ostracod Cythereis (Daday, 1911). NoTE. With the submergence of the genus Cothurniopsis by Kahl (1935), this species was transferred to Cothurnia. If, however, coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia subtilis Stiller, 1939 DESCRIPTION (Figs 163 & 164). Lorica 65-72 wm long Xx 28 ym wide. Aperture 20 wm in diameter. External stalk 25—30 wm long; endostyle 10-15 wm long and slightly swollen at distal end; mesostyle absent. Contracted zooid 50 wm long x 15 wm wide. Macronucleus J-shaped. Pellicular striations inconspicuous. HABITAT. Marine, originally found attached to Mytilus edulis (Stiller, 1939). Note. Uncontracted specimens of this species have yet to be described. Cothurnia tekirghiolica Tucolesco, 1962 DESCRIPTION (Figs 165-167). Lorica 80 um long xX 30 um wide. Aperture oval when viewed from above, 30 wm xX 20 wm. External stalk 15 um long with irregular ridges and furrows; mesostyle short; endostyle short and inconspicuous. All three stalks with continuous longitudinal striae. Zooid 80 pm long X 20 um wide, extending up to one third of its length beyond aperture. Peristomial lip well developed, 25 um in diameter. CV situated near base of infundibulum. Macro- nucleus C-shaped and lies longitudinally in the centre of zooid. Pellicle clearly striated. HABITAT. Freshwater, originally found attached to Cladophora from Lake Tekirghiol (Tucolesco, 1962). _ Figs 165-167 Cothurnia tekirghiolica, after Tucolesco, 1962; fig. 165 with developing telotroch, bar = 50 um; fig. 166 external stalk and _ endostyle; fig. 167 contracted zooid. | Figs 168 & 169 Cothurnia triangula, after Precht, 1935; fig. 168 aperture, anterior view; fig. 169 bar = 50 um. | Fig. 170 Cothurnia trophoniae, composite after Dons, 1946, bar = 100 um. | | 50 ALAN WARREN & JAN PAYNTER 175 | Hy ‘ 179 A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES Cothurnia triangula Precht, 1935 DESCRIPTION (Figs 168 & 169). Lorica curved, 45 um long X 18 um wide. Aperture triangular when viewed from above, 14 pm across. External stalk curved, 10-15 um long; mesostyle short and broad; endostyle short and inconspicuous. All three stalks with conspicuous longitudinal striae. Zooid 40-45 xm long X 12 um wide, extending just beyond aperture. Peristo- mial lip well developed, 14 xm in diameter. CV situated near anterior end of macronucleus which is 27 wm long and lies longitudinally in zooid. Pellicle clearly striated. HABITAT. Marine, originally found as an epibiont of the halacarid Copidognathus fabriciusi and the ostracod Cythereis tuberculata (Precht, 1935). Cothurnia trophoniae Dons, 1946 Cothurnia pedunculata Dons, 1918 Cothurnia nodosa Mereschkowsky, 1879 (in part) Dons, 1918 DESCRIPTION (Fig. 170). Lorica roughly cylindrical, 125-140 pm long X 40-50 um wide. External stalk 65-135 wm long; mesostyle short and broad with conspicuous longitudinal striae; endostyle short. Two zooids per lorica, each 70 wm long X 35 wm wide when contracted. Macronucleus vermiform and coiled irregularly. Pellicular striations inconspicuous. HABITAT. Marine, found as an epibiont of the alga Polysi- phonia violacea (Dons, 1918), the polychaetes Trophonia plumosa and Stylarioides plumosus (Dons, 1946; Precht, 1935), the cnidarian Laomedea loveni and the pantopod Pallene brevirostris (Precht, 1935). Norte. Dons (1946) applied the name C. trophoniae to C. nodosa Mereschkowsky, 1879, an organism which Dons (1918) had previously redescribed under the name C. _ pedunculata. It is unclear why Dons (1946) chose not to use _ the name C. pedunculata for this species as it would appear to have priority over the name C. trophoniae. Furthermore, if _coloniality (sensu Jankowski, 1985) is accepted as a generic _ character, this species should be included in the genus Sincothurnia. C. trophoniae was redescribed by Precht (1935). _ Cothurnia trophonicola Dons, 1946 _ DESCRIPTION (Fig. 171). Lorica irregular and variable in _ Shape, 60-85 um long x 31-35 wm wide. Aperture oval when _ viewed from above, 12-16 wm X 30-33 wm. External stalk 5— _ 15 wm long x 3-4 wm wide; mesostyle 4-6 wm long X 6-8 wm _ wide; endostyle absent. Two zooids per lorica, each 50 wm _ long X 25 wm wide when contracted. Macronucleus vermiform _ and coiled irregularly. Pellicular striations inconspicuous. | HABITAT. Marine, originally found as an epizoite of the _ polychaete Trophonia plumosa (Dons, 1946). | Fig. 171 Cothurnia trophonicola, after Dons, 1946, bar = 50 um. _ Fig. 172 Cothurnia vaga, after Roux, 1901, bar = 100 um. _175—called Cothurnia variabilis var. emarginata). | composite after Stokes, 1893. St NotE. If coloniality (sensu Jankowski, 1985) is accepted as a generic character, this species should be included in the genus Sincothurnia. Cothurnia vaga Roux, 1901 DESCRIPTION (Fig. 172). Lorica cylindrical, 100 wm long x 45 jm wide and rounded posteriorly. External stalk short and slender; mesostyle and endostyle absent. Zooid 140 um long x 30 wm wide, extending up to one third of its length beyond aperture. Peristomial lip 45 wm in diameter. CV situated beneath peristome. Macronucleus short, curved and lies in centre of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, originally found as an epizoite of the crustaceans Cyclops and Gammarus (Roux, 1901). Cothurnia variabilis Kellicott, 1883 Cothurnia variabilis var. emarginata (Kellicott, 1883a) Stokes, 1888 Daurotheca variabilis (Kellicott, 1883) Jankowski, 1987 Daurotheca marginata (Kellicott, 1883) Jankowski, 1987 DESCRIPTION (Figs 173-175). Lorica 75-130 wm long, variable in shape but usually rounded posteriorly and with curved neck. One or more spines may be present either near aperture or at posterior end of lorica. Aperture 30-40 um wide. External stalk short and broad; mesostyle and endostyle absent. May form pseudocolonies with several loricas attached to each other via their external stalks to form chains. Two zooids per lorica, each 75—130 um long X 25—45 wm wide and extending just beyond aperture. Diameter of peristomial lip slightly less than maximum body width. Infundibulum broad and reaches one third of zooid length. CV situated near centre of zooid. Macronucleus C-shaped and lies either in centre or in lower half of zooid. Pellicular striations inconspicuous. HABITAT. Freshwater, found as an epizoite of the crayfish Cambarus diffinis, C. propinquus, C. bartonii, C. affinis, Astacus leptodactylus and A. fluviatilis (Kellicott, 1883a; Nenninger, 1948; Krucinska and Simon, 1968). NoTE. Jankowski (1987) transferred this species and C. variabilis var. emarginata Kellicott, 1883 (as Daurotheca marginata) to Daurotheca. However, since Daurotheca is of uncertain taxonomic status (see Incertae Sedis), C. variabilis is retained in the genus Cothurnia. Genus COTHURNIOPSIS Stokes, 1893 The genus Cothurniopsis was erected by Stokes (1893) for peritrichs which resemble Cothurnia in every respect save | Figs 173-175 Cothurnia variabilis; fig. 173 chain of individuals forming a pseudocolony, bar = 100 um; figs 174 & 175 after Kellicott, 1883 (fig. _ Figs 176-179 Cothurniopsis valvata; figs 176 & 177 composite after Penard, 1914, bar = 50 um (called Cothurniopsis elastica); figs 178 & 179 | Figs 180-182 Dimorphocothurnia nebaliae; figs 180 & 181 after Jankowski, 1985, bar = 100 um; fig. 182, after Dons, 1928. 32 one, that is the lateral borders of the lorica are pliable and are used to close the aperture when the ciliate contracts. By contrast Cothurnia has no mechanism of closing its aperture. The name Cothurniopsis has, on several occasions, been erroneously confused with Entz’s (1884) genus Cothurnopsis. According to Entz (1884) the principal characters by which Cothurnopsis is distinguished from Cothurnia are the posses- sion of, (i) a large, transversely folded external stalk, and (ii) a compact macronucleus. Like Cothurnia, Cothurnopsis has no mechanism for closing its aperture. Unfortunately the similarity of the two names Cothurniopsis and Cothurnopsis has led to some confusion. For example Penard (1914), Zelinka (1928) and Kahl (1935) have either transferred existing species or described new species using the generic name Cothurniopsis. Yet in almost every case, the organisms concerned conformed to Cothurnopsis (sensu Entz, 1884) rather than Cothurniopsis (sensu Stokes, 1893). Jankowski (1985) cites other examples of such errors. Stokes (1893) described a single species of Cothurniopsis, C. valvata, which became the type species by monotypy. Cothurniopsis elastica Penard, 1914 is the only cothurnid described since which possesses a closeable aperture. Diagnosis of Cothurniopsis Lorica erect and borne upon a stalk. Aperture border pliable and closes aperture when zooid contracts. Key to the species of Cothurniopsis 1. Anterior lateral borders pliable and close aperture upon contractiomofzooid(s)mers tas. S27 see Cae ea C. valvata Species description Cothurniopsis valvata Stokes, 1893 Cothurnia elastica (Penard, 1914) Kahl, 1935 Cothurnia valvata (Stokes, 1893) Kahl, 1935 Cothurniopsis elastica Penard, 1914 DESCRIPTION (Figs 176-179). Lorica 50-70 wm long x 25-35 zm wide and rounded posteriorly. Aperture oval when viewed from above, 20 wm long X 10 wm wide. External stalk 10-18 wm long. Zooid 75-80 um long xX 10-15 um wide, extending up to one third of its length beyond aperture. CV situated in anterior half of body. Macronucleus straight and lies longitudinally in posterior half of body. Pellicle with fine striations. HABITAT. Brackish or freshwater, originally found attached to filamentous algae from Coney Island, New York, USA (Stokes, 1893); also isolated from moss in Europe (Penard, 1914). NoTE. Penard (1914) described several species under the generic name Cothurniopsis although only one, C. elastica, has pliable lateral borders enabling the aperture to be closed. C. elastica shows only minor differences from C. valvata so the two are synonymised. Genus CYCLODONTA Matthes, 1958 The genus Cyclodonta was erected by Matthes (1958) for Cothurnia bipartita Stokes, 1885, which differs from other ALAN WARREN & JAN PAYNTER species of Cothurnia in that the zooid is attached to the inside of the lorica via a series of membranes rather than via an endostyle or mesostyle. In all other respects, Cyclodonta resembles Cothurnia. C. bipartita is the only species of Cyclodonta and so becomes the type species by monotypy. Cyclodonta is commonly included in_ the family Vaginicolidae (Corliss, 1979; Foissner, 1979), although according to Stiller (1971) it should belong to the family Lagenophrydae because of its mode of development. Diagnosis of Cyclodonta Lorica borne upon a short stalk and without valves or other means of closing aperture. Zooid attached to lorica via a series of membranes. Mesostyle and endostyle absent. Single species genus. Key to the species of Cyclodonta 1 Zooid attached to the lorica via a series of membranes es, AARP I ees en ew, RL EM I BM So oo a oc C. bipartita Species description Cyclodonta bipartita (Stokes, 1885) Matthes, 1958 Cothurnia affinis (Kahl, 1935) Matthes, 1958 Cothurnia bipartita Stokes, 1885 Cothurniopsis rheotypica (Stiller, 1931) Felinska, 1965 Cothurnia trilobata (Sramek-Husek, 1957) Matthes, 1958 Cothurnia voigti (Voigt, 1902) Kahl, 1935 Cothurniopsis longipes Voigt, 1902 DESCRIPTION (Figs 183-189). Lorica curved, 78-132 um long x 40-70 um wide with fine, longitudinal striae. Aperture 40— 70 wm in diameter. External stalk curved, 20 um long, with transverse furrows. 2-6 membranes present at posterior end of lorica; mesostyle and endostyle absent. Zooid 40-105 wm long X 29-62 um wide, not extending as far as aperture. CV empties into infundibulum via short channel. Macronucleus variable in shape, from short and curved to vermiform. Pellicle clearly striated with convex ribbing between striations. HABITAT. Freshwater, commonly found attached to harpacti- coid copepods (Matthes, 1958; Jankowski, 1985); also found attached to the crayfish Cambarus affinis and Astacus lep- todactylus (Krucinska and Simon, 1968). Note. Cyclodonta bipartita has been isolated by several workers including Voigt (1902), Penard (1914), Kahl (1935), Matthes (1958) and Foissner (1979), and in every case the host was a freshwater harpacticoid copepod. Indeed Jankowski (1985) has suggested that the absence of Cothurnia from these hosts is due to its substitution by Cyclodonta. However, Cyclodonta does not appear to be confined exclu- sively to harpacticoid copepod hosts as suggested by Jankowski (1985), since it has also been isolated (as Cothurnia bipartita) from the crayfish Cambarus affinis and Astacus leptodactylus (Krucinska and Simon, 1968). Genus DIMORPHOCOTHURNIA Jankowski, 1980 The genus Dimorphocothurnia was first mentioned by | Jankowski (1980), although a full generic description did not appear until five years later (Jankowski, 1985). Dimorpho- | | A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES 53 —— SS. aK NS Hitiny ~~ \ yi NN My, NX ( flip N Z y, 187 186 183 191 | ' : | L | | | | | _ Figs 183-189 Cyclodonta bipartita; fig. 183 after Stamek-Husek, 1957 (called Cothurnia trilobata); fig. 184 after Stokes, 1885; figs 185 & 186 after Matthes, 1958 (fig. 185 bar = 50 wm; fig. 186 macronucleus); fig. 187 after Voigt, 1902 (called Cothurnia longipes); fig. 188 after Stiller, 1931 | (called Cothurniopsis rheotypica); fig. 189 after Foissner, 1979, bar = 50 um. _ Fig. 190 Daurotheca tespa, after Jankowski, 1987, bar = 50 wm. Fig. 191 Daurotheca transoceanica, after Jankowski, 1987, bar = 50 um. Fig. 192 Daurotheca ussurina, after Jankowski, 1987, bar = 50 um. _cothurnia was erected for Cothurnia nebaliae Dons, 1928 Diagnosis of Dimorphocothurnia which typically forms pseudocolonies with several individuals attached to each other via their external stalks to form chains. Although other species of Cothurnia, e.g. C. bavarica and C. _variabilis, also form pseudocolonies they differ from Dimor- _ Phocothurnia in one important respect; that is, in C. bavarica _ and C. variabilis all the individual loricas of any chain are identical, whereas Dimorphocothurnia exhibits lorica / dimorphism with the individual in contact with the substratum _ (basont’ Jankowski, 1985) having a longer stalk than those of | the rest of the chain. D. nebaliae is the only species of | Lorica without valves or any other means of closing aperture. Forms pseudocolonies with several loricas joined together via their external stalks to form chains. Exhibits lorica dimorphism with lorica in contact with substratum (‘basont’) having longer external stalk than those in rest of chain. Single species genus. Key to the species of Dimorphocothurnia Dimorphocothurnia and is the type species by monotypy. 1. Forms pseudocolonies and exhibits lorica dimorphism Nn = with individual in contact with substratum having longer external stalk than others inchain ................ D. nebaliae Species description Dimorphocothurnia nebaliae (Dons, 1928) Jankowski, 1985 Cothurnia nebaliae Dons, 1928 DESCRIPTION (Figs 180-182). Lorica 100-120 wm long x 40-55 wm wide. Aperture 20-35 wm in diameter. Forms pseudocolonies with several loricas joined together via their external stalks to form chains. External stalk of individual in contact with substratum (‘basont’), 30-40 wm long xX 6-7 wm wide; those of rest of chain, 5 wm long X 6—-7 wm wide. Mesostyle short and broad; endostyle absent. Zooid 140 um long X 60 um wide and extends just beyond aperture. Peristomial lip well developed, 50 wm in diameter. Disc convex. Macronucleus vermiform, 40-50 um long and irre- gularly coiled. Pellicular striations inconspicuous. HABITAT. Marine, originally isolated from Norwegian coastal waters attached to the phyllocarid Nebalia bipes (Dons, 1928); also isolated from phyllocarids by Jankowski (1980, 1985). INCERTAE SEDIS Genus DAUROTHECA Jankowski, 1987 The genus Daurotheca was erected by Jankowski (1987) for cothurnids which have asymmetrical loricas and zooids, and exhibit a tendency to form spines on their loricas. Neither feature has previously been used for separating peritrich genera, and data on their reliability as taxonomic characters is scarce. In all other respects Daurotheca resembles Cothurnia. Two cothurnids, Cothurnia variabilis Kellicott, 1883 and C. variabilis var. emarginata Kellicott, 1883 were transferred by Jankowski (1987) to Daurotheca. However, until the taxo- nomic status of Daurotheca has been established these taxa should remain in the genus Cothurnia. Jankowski (1987) also described three new species, Daurotheca transoceanica, D. tespa and D. ussurina. Diagnosis of Daurotheca Lorica asymmetrical, often with spines. Attached to substratum via external stalk. Aperture narrow, inclined at oblique angle to main lorica axis. Neck short. Zooid asymmetrical and bends to one side upon contraction. Key to the species of Daurotheca k” Tegried WHNSOMIER G57 oi ss cat's, oe ete ere aa ee 2 LOLCA WIHNOME SPINES i492 cian cca nibs arson. sec D. ussurina 2 __Lorica with two spines, one subapical, one basal RE ccrarertan iin tictrrse. patience teres Apiencnci nie D. transoceanica Dasalhai se ccak.. ARM in eee cc ee RO Eee D. tespa ALAN WARREN & JAN PAYNTER Species descriptions Daurotheca tespa Jankowski, 1987 DESCRIPTION (Fig. 190). Lorica trapezoid, 86-92 um long x 58-64 «wm wide and with four spines, one broad apical (20 wm long), one slender subapical (14 4m long), and two basal (8 wm long). Apical and subapical spines point upwards, basal spines horizontally. Aperture held at angle to main lorica axis on short, broad neck. External stalk short; meso- style and endostyle absent. HABITAT. Freshwater, originally found attached to the gills of the crustacean Pacifastacus lenisculus (Jankowski, 1987). Daurotheca transoceanica Jankowski, 1987 DESCRIPTION (Fig. 191). Lorica roughly triangular, 69-74 wm long X 60-75 um wide. Anterior region constricted to form a neck. Dorsal side of neck straight, 29-33 um long. Aperture 22-25 xm wide. Lorica with two spines, one subapical 21 wm long, the other basal 10 um long. External stalk up to 8 wm long; mesostyle short and inconspicuous; endostyle absent. Zooid 45 wm long X 45 wm wide when contracted. HABITAT. Freshwater, originally found attached to the crusta- cean Pacifastacus lenisculus (Jankowski, 1987). Daurotheca ussurina Jankowski, 1987 DESCRIPTION (Fig. 192). Lorica somewhat rotund, 69-74 wm long X 42-46 wm wide. Aperture 20-23 wm wide. Neck narrow. External stalk 7-8 wm long X 3 wm wide; mesostyle short and broad; endostyle absent. Zooid 65 wm long x 45 jum wide when contracted. Macronucleus C-shaped and lies horizontally in anterior part of zooid. HABITAT. Freshwater, originally found attached to the crusta- ceans Cambaroides, Cambarus, Orconectes and Pacifastacus (Jankowski, 1987). ACKNOWLEDGEMENTS. The authors would like to thank Dr D. Roberts, British Museum (Natural History), for reading the manuscript and making many helpful comments. REFERENCES André, E. 1910. Sur quelques infusoires marins parasites et commensaux. Revue Suisse de Zoologie 18: 173-188. Andrussowa, I. 1886. Infusorii Kerchenskoe buxte. Trudy St. Petersbourg Obshchestva Estestvoispytatalei 17: 236-259. Bacon, A. L. 1968. Biology of Cothurnia imberbis. Dissertation Abstracts International 29B: 3968. Banina, N. N. & Polyakova, L. A. 1977. (New species of Peritricha Sessilia on the aquatic surface of Ropsha ponds—in Russian). /zvestiya Gosudarstven- nogo Nauchno-Issledovatel ’skogo Instituta Ozernogo: Rechnogo Rybnogo Khozyaistva, Leningrad 119: 12-23. Blochmann, F. 1886. Die mikroscopische Thierwelt des Susswassers. pp. 122, Haering, Braunschweig. Bock, K. J. 1952. Einige gehausebauende Peritrichen aus der Kieler Bucht. Zoologischer Anzeiger 149: 273-276. Claparede, E. & Lachmann, J. 1858. Etudes sur les infusoires et les rhizopodes. Mémoires de I’Institut National Genevois, Geneve 5 (year 1857): 1-260. Cohn, F. 1866. Neue infusorien im seeaquarium. Zeitschrift fur Wissenschaft- liche Zoologie 16: 253-302. Corliss, J. O. 1979. The Ciliated Protozoa: Characterization, Classification and Guide to the Literature, 2nd edition, pp. 455, Oxford, Pergamon. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES Cuénot, L. 1891. Infusoires commensaux des Ligies, Patelles et aronicoles. Revue Biologique du Nord de la France 4: 81-89. Curds, C. R. 1985. A revision of the Suctoria (Ciliophora, Kinetofragminophora) 1. Acineta and its morphological relatives. Bulletin of the British Museum (Natural History). Zoology. London. 48: 75-129. Daday, J. 1907. Plancton-Tiere aus dem Victoria Nyanza. Sammelausbeute von A. Borgert, 1904-1905. Zoologische Jahrbiicher. Systematik 25: 245-262. — 1911. Deux infusoires nouveaux de la région Antarctique. Allattani K6ézlemenyek, Budapest 10: 97-99, 114-115. Davis, J. 1879. On a new species of Cothurnia. Journal of the Royal Microscopi- cal Society, London 2: 653-655. Dons, C. 1918. Neue marine Ciliaten und Suctorien. Tromsg Museums Aarschefter 38-39 (year 1915-16): 75-100. 1922. Notes sur quelques Protozoaires marins. Videnskabelige Meddelelser Dansk naturhistorisk Forening i Kjobenhavn 73: 49-84. 1928. Neue und wenig bekannte Protozoen. Kongelige Norske Videnska- bernes Selskabs Skrifter 7 (year 1927): 1-17. — 1946. Zwei neue Cothurnien. Det Kongelige Norske Videnskabers selskabs 12 (year 1939-40): 119-121. Dujardin, F. 1841. Histoire Naturelle des Zoophytes. Infusoires. pp. 678, Paris, De Fain et Thunot. Ehrenberg, C. G. 1831. Uber die Entwicklung und Lebensdauer der Infusions- thiere: nebst ferneren Beitragen zu einer Vergleichung ihrer organischen System. Abhandlungen der Akademie der Wissenschaften der DDR. Berlin. year 1832,: 1-154. — 1833. Dritter Beitrag zur Erkenntniss grosser Organisation in der Richtung des kleinsten Raumes. Abhandlungen der Akademie der Wissenschaften der DDR. Berlin. year 1835: 145-336. — 1838. Die Infusionsthierchen als Volkommene Organismen. pp. 612, Leipzig. Eichwald, E. 1849. Zweiter Nachtrag zur Infusorienkunde Russlands. Bulletin de la Societe des Naturalistes de Moscou 22: 400-548. Entz, G. 1876. A tordac és szamosfalvi sostavak azalagfaunaja. Magyar orvosok és termeszétvizsgalok nagygylilésének évkényvei. Budapest 18. 1884. Uber Infusorien des Golfes von Neapel. Mitteilung der zoologischen station in Neapel 5: 289-444. — 1904. Einiges uber das variieren der Infusorien. Mathematische und Naturwissenschaftliche Mittheilungen aus den Sitzungsberichten der Kongilich Preussischen Akademie der Wissenschaften zu Berlin 19: 125-144. Felinska, M. 1965. Marine ciliata from Plymouth Peritricha, Vaginicolidae. _ Journal of the Marine Biological Association of the United Kingdom 45: im 229-239. Fenchel, T. 1965. On the ciliate fauna associated with the marine species of the | amphipod genus Gammarus J. G. Fabricius. Ophelia 2: 281-303. _ Foissner, W. 1979. Peritrichen Ciliaten (Protozoa: Ciliophora) aus Alpinen | Kleingewassern. Zoologische Jahrbiicher Systematik 106: 529-558. | | | —— 1981. Morphologie und Taxonomie einiger Heterotricher und Peritricher Ciliaten (Protozoa: Ciliophora) aus Alpinen Béden. Protistologica 17: 29-43. — & Schiffmann, H. 1974. Vergleichende Studien an argyrophilen Strukturen von vierzehn peritrichen Ciliaten. Protistologica 10: 489-508. & 1975. Biometrische und morphologische Untersuchungen uber die Variabilitat von argyrophilen | Protistologica 11: 415-428. | Fromentel, E. de 1874-1876. Etudes sur les Microzoaires ou Infusoires Propre- ment Dits, Comprenant de Nouvelles Reserches sur Leur Organisation, Leur Classification, et la Description des Espéces Nouvelles ou Peu Connues. pp. 364, Paris, G. Masson. Gajewskaja, N. 1933. Zur oekologie, morphologie und systematik der infusorien _ des Baikalsees. Zoologica, Stuttgart 32: 113-216. Gourret, P. & Roeser, P. 1886. Les Protozoaires du Vieux-Port de Marseille. Archives de Zoologie Expérimentale Génerale 4: 443-534. | Greeff, R. 1888. Land-Protozoen. Sitzungsberichte der Gesellschaft zur | Beforderung der gesammten Naturwissenschaften 1888: 90-158. Gruber, A. 1879. Observations on new infusoria. Journal of the Royal Microscopical Society 2: 899-900. 1880. Neue Infusorien. Zeitschrift fiir Wissenschaftliche Zoologie 33: _ 439-466. Hemprich, F. G. & Ehrenberg, C. G. 1828. Symbolae physicae. Pars Zool. 1V Animalia Envertebrata, Berlin. / Hofker, I. 1930. Faunistische beobachtungen in der Zuidersee wahrend der trochenlegung. Zeitschrift fur Morphologie und Okologie der Tiere Berlin 18: 189-216. _ Hutton, F. W. 1878. On a new operculated infusorian from New Zealand. | Journal of the Royal Microscopical Society, London 1: 49-50. 1981. On a new infusorian parasitic on Patella argentea. Transactions and _ Proceedings of the New Zealand Institute 11 ( year 1878): 330. | Jankowski, A. W. 1976. Reviziya ostryada Sessilida (Peritricha). Materialy I] | | ) Strukturen bei peritrichen Ciliaten. vsesoynznogo s’ezda protozoologov. Chast (part 1) I, obshchaya protozoo- logiya. Kiev, Nankora Dunka 1976: 168-170. 55 1980. (A new conspectus of the systematics of Ciliophora—in Russian). Trudy Zoologischeskogo Instituta. Akademiya Nauk SSSR. Leningrad 94: 103-121. 1985. (Life cycles and taxonomy of generic groups Scyphidia, Heteropolaria, Zoothamnium and Cothurnia (Class Peritricha)—in Russian). Trudy Zoo- logicheskogo Instituta. Akademiya Nauk SSSR. Leningrad 129: 74-100. — 1987. (New and little known genera of ciliated protozoa (phylum Ciliophora)—in Russian). Trudy Zoologicheskogo Instituta. Akademiya Nauk SSSR. Leningrad 144 (year 1986): 72-88. Jaworowski, A. 1893. (Fauna of the wells in Krakau and Lemburg—in Polish). Sprawozdania Komisji Fizyograficznej oraz Materyaly do Fizyografii Kraju 28: 29-48. Kahl, A. 1928. Die Infusorien (Ciliaten) der Oldesloer Salzwasserstellen. Archiv fiir Hydrobiologie 19: 189-246. 1933. Ciliata libera et ectocommensalia. Jn Grimp, G. & Wagler, E. (Eds) Die Tierwelt der Nord und Ostsee Bd 1. (Tiel II. c3): 29-183. 1935. Wimpertiere oder Ciliata (Infusoria), eine Bearbeitung der freilebenden und ectocommensalen Infusorien der Erde, unter Ausschluss der marinen Tintinnidae. 4. Peritricha und Chonotricha. Jn. Grimp, G. & Wagler, E. (Eds) Die Tierwelt der Nord und Ostsee 30: 651-886. Kellicott, D. S. 1883a. On some infusoria found on the cray-fish. Proceedings of the American Society of Microscopists 1883: 105-111. — 1883b. Cothurnia lata n. s. American Microscopical Society Proceedings 6: 113-114. — 1894. On certain marine infusoria observed at Woods Hole, Mass. Report of the Ohio State Academy of Science 2: 10-11. Kent, W. S. 1869. On some new infusoria from the Victoria Docks. Monthly Microscopical Journal 1: 289-293. 1881-1882. A Manual of the Infusoria vol II: part 5 pp. 577-720, part 6 pp. 721-913, London, David Brogue. Kralik, U. 1961. Ein Beitrag zur Biologie von loricaten peritrichen Ziliaten insbesondere von Platycola truncata Fromentel, 1874. Archiv fur Protisten- kunde 105: 201-258. Krucinska, J. & Simon, E. 1968. (On the parasites and epibionts of the branchial cavity in the crayfish at Wroclaw vicinity—in Polish). Przeglad Zoologiczny 12: 288-290. Labbé, A. 1895. Note sur les protozoaires marins de Roscoff. Archives de Zoologie Expérimentale et Génerale 3: XI (abstr.). Lang, K. 1948. On Rhabdostyla commensalis Mobius and two new Cothurnia species (Ciliata, Peritricha) from the west coast of Sweden. Arkiv for Zoologi 41-B: 1-6. Levander, K. M. 1915. Uber eine neue marine Cothurnia-Art. Meddelanden et Societas Profauna et Flora Fennica 10: 86-88. Liipkes G. 1974. Drei neue stygobionte Peritrichen aus dem interstitiellen Grundwasser. Archiv ftir Hydrobiologie 73: 394-402. Maggi, L. 1879. Intorno alle Cothurnie parassit delle branchie dei Gamberi nostrali. Instituto Lombardo di Science e letice Milan Rendiconti 12: 439-448. Maskell, W. M. 1887. On the freshwater infusoria of the Wellington District. Transactions and Proceedings of the New Zealand Institute 20: 3-19. Matthes, D. 1958. Das peritriche Ciliat Cyclodonta bipartita (Stokes) nov. gen. Archiv fur Protistenkunde 102: 481-500. & Guhl, W. 1973. Sessile Ciliaten der Flusskrebse. Protistologica 9: 459-470. Mereschkowsky, C. 1879. Studien uber Protozoen des nordlichen Russland. Archiv ftir Mikroscopische Anatomie 16: 153-244. 1881. On some new or little known infusoria. Annals and Magazine of Natural History. Series 5. 7: 209-219. Mobius, K. 1888. Bruchstiicke einer Infusorienfauna der Kieler Bucht. Archiv fiir Naturgeschichte 1888: 81-116. Monard, A. 1919. Sur la faune profunde du Lac de Neuchatel. Revue Suisse de Zoologie 26 (year 1918): 341-359. Miller O. F. 1786. Animalcula Infusorian Fluviatilia et Marina. pp. 367, Havniae et Lipsiae. Nenninger, U. 1948. Die Peritrichen der Umgebung von Erlangen mit beson- derer Berticksichtigung ihrer Wirtsspezifitat. Zoologische Jahrbucher. Systematik 77: 169-266. Penard, E. 1914. Les Cothurnidés Muscicoles. Mémoires de la Societé de Physique et d’ Histoire Naturelle de Genéve 38: 19-65. 1922. Etudes sur les Infusoires d’Eau Douce. pp. 331, Genéve, Georg et Cie. Perty, M. 1852. Zur Kenntnis Kleinster Lebensformen nach Bau, Funktionem, Systematik mit Specialverzeichneiss der in der Schweiz beobacteten. pp. 228, Bern, Jent u. Reinert. Precht, H. 1935. Epizoen der Kieler bucht. Deutsche Akademie der Natur- forscher Nova Acta Leopoldina (Halle) N.F. 3: 405-474. Quennerstedt, A. 1865. Sidrag till Sveriges Infusorie-fauna I. Acta Universitatis Lundensis 2: 1-64. Randall, J. T. & Hopkins, J. M. 1962. On the stalks of certain peritrichs. Philosophical Transactions of the Royal Society of London 245: 59-79. 56 Roux, J. 1901. Faune infusorienne des eaux stagnantes des environs de Geneve. Mémoires de I’ Institute National Genvois 19: 1-153. Shubernetskii, I. V. 1978. Novbievidbi krugoresnichn, ikhiafuzorii iz vodoe- mob moldabii. /zvesitsya Akademi Nauk Moldavskoi SSR, Seriya Biologi- cheskikh Nauk 1978: 62-67. Sommer, G. 1951. Die Peritrichen Ciliaten des Grossen Ploner Sees. Archiv fiir Hydrobiology 44: 349-440. Sramek-Husek, R. 1957. K poznani nalevniku ostravského kraje. Acta Societatis Zoologicae Bohemoslovenicae 21: 1-24. Stein, F. 1854. Die Infusionsthiere auf ihre Entwicklungsgeschichte untersucht. pp. 265, Leipzig, Verlag von Wilhelm Engelmann. — 1867. Der Organismus der Infusionsthiere nach eigenen Forschungen in Systematischer Reihenfolge Bearbeitet I]. pp. 355, Leipzig. Stiller, J. 1931. Die Peritrichen infusorien von Tihany und Umgebung. Arbeiten des Ungarischen Biologischen Forschungs-Institutes 4: 171-205. 1939. Die peritrichenfauna der Nordsee bei Helgoland. Archiv fur Protistenkunde 92: 415-452. 1940. Beitrage zur peritrichenfauna des Grossen Ploner Sees in Holstein. Archiv fiir Hydrobiologie 36 (year 1939): 263-285. 1951. Epizoische peritrichen aus dem Balaton II. Archiva Biologica Hungarica 19: 15-37. 1953. Die Protozoen des Précsely-Baches in Ungarn. Annals Historico- Naturales Musei Nationalis Hungarici 4: 47-71. 1968. Peritriche Ciliaten 6kologisch verschiedener Biotope von Rovinj und Umgebung. Acta Zoologica Academiae Scientiarum Hungaricae 14: 185— 211. — 1971. Szajkoszorus Csillosok—Peritricha. Fauna Hungaricae 105: 203- 213. Stokes, A. C. 1884. A new infusorian belonging to the genus Pyvxicola. American Monthly Microscopical Journal 5: 24-25. — 1885. Notices of new fresh-water infusoria—III. The American Monthly Microscopical Journal 6: 121-127. —— 1886. Notices of new fresh-water infusoria—V. The American Monthly Microscopical Journal 7: 81-86. 1887. Notices of some new fresh-water infusoria. Proceedings of the American Philosophical Society 24: 244-255. — 1888. A preliminary contribution toward a history of the fresh-water infusoria on the United States. Journal of the Trenton Natural History Society 1: 71-344. — 1893. VII—Notices of some undescribed infusoria from brackish waters of the eastern United States. Journal of the Royal Microscopical Society Yr 1893: 298-302. Swarczewsky, B. 1930. Zur Kenntnis der Baikalprotistenfauna. Die an den Baikalgammariden lebenden Infusorien. VII. Lagenophrys, Vaginicola & Cothurnia. Archiv fur Protistenkunde 69: 455-532. Trueba, F. J. 1978. A taxonomic revision of the peritrich ciliate genus Pyxicola. Beaufortia 27: 219-243. —— 1980. A taxonomic revision of the peritrich ciliate genera Thuricola and Pseudothuricola. Beaufortia 30: 125-138. Tucolesco, J. 1962. Etudes protozoologiques sur les eaux Roumaines. I. Espéces nouvelles d’infusoires de la Mer Noire et les bassins sales paramarins. Archiv fur Protistenkunde 106: 1-36. d’Udekem, J. 1864. Description des infusoires de la Belgique. Mémoires de l’ Academie Royale des Sciences, des Lettres et des Beaux-Arts de Belgique 34: 1-34. Vejdovsky, F. 1882. Tierische Organismen der Brunnenwdsser von Prag., pp. 68, Prag., Selbstverlag. Viljoen, S. & van As, J. G. 1983. A taxonomic study of sessile peritrichians of a small impoundment with notes on their substrate preferences. Journal of the Limnological Society of South Africa 9: 33-42. Voigt, M. 1902. IV. Neue organismen aus ploner Gewassern. Forschungs- berichte aus der Biologischen Station zu Plén 9: 33-46. Vogelbein, W. K. & Thune, R. L. 1988. Ultrastructural features of three ectocommensal protozoa attached to the gills of the red swamp crawfish Procambarus clarkii (Crustacea: Decapoda). Journal of Protozoology 35: 341-348. Vuxanovici, A. 1962. Contributii la systematica Ciliatelor (Nota III). Studii si Cercetari de Biologie. Seria Biologie Animala 14: 549-573. Wailes, G. H. 1928. Fresh-water and marine protozoa from British Columbia. Vancouver Museum and Art Notes 3: 25-36. 1943. Canadian Pacific Fauna. 1. Protozoa. If Ciliata, 1g Suctoria. pp. 46. Toronto, University of Toronto Press for the Fisheries Research Board of Canada. Wang, C. C. 1935. Notes on the marine infusoria of Amoy. Marine Bio- logical Association of China. 3rd Annual Report (year 1934), pp. 50-70. & Nie, D. 1933. A survey of the marine protozoa of Amoy. Contributions from the Biological Laboratory of the Science Society of China, Nanking. Zoology Series 8: 285-385. ALAN WARREN & JAN PAYNTER Wang Jiaji 1977. Protozoa from some districts of Tibetan Plateau. Acta Zoologica Sinica 23: 131-160. Warren, A. 1982. A taxonomic revision of the genus Platycola (Ciliophora: Peritrichida). Bulletin of the British Museum (Natural History). Zoology. London. 43: 95-108. —— 1987. A revision of the genus Vorticella (Ciliophora: Peritrichida). Bulletin of the British Museum (Natural History). Zoology. London. 50: 1-57. Wright, T. S. 1858. Description of new protozoa. Edinburgh New Philosophical Journal 7: 276-281 Wrzesniowski, A. 1866. Spis Wymoczkow Sporstrzeganych w Warszawie ijej okolicach w latach 1861 do 1865. Wyzak Szkoly Glownej Warszawskiej 5: 15-28. — 1870. Beobachtungen Uber Infusorien aus der Umgebung von Warsaw. Zeitschrift fiir Wissenschaftliche Zoologie 20: 467-511. Yunfen, S. 1980. Descriptions of six new species of periphytic protozoa in Lake Dong Hu of Wuhan. Acta Hydrobiologica Sinica 7: 245-251. Zelinka, C. 1913. Zwei Ektoparasiten der Echinoderen aus der KI. der Ciliaten. Bericht tiber die Verhandlungen der Naturforschenden Gesellschaft in Leipzig 85. 1928. Monographie der Echinodera, pp. 396, Leipzig, Engelmann. Manuscript accepted for Publication 16 August 1989 INDEX TO SPECIES The following is an annotated list of nominal species as well as an index of extant species. Cothurnia C. acuta Wang & Nie, 1933 = Vaginicola wangi (Wang & Nie, 1933) Kahl, 1935. C. acuta Levander, 1915 (p. 26). C. affinis Blochmann, 1886 = Pyxicola pusilla Wrzesniowsky, 1866) Trueba, 1978. C. affinis Kahl, 1935 = Cyclodonta bipartita (Kahl, 1935) Matthes, 1958 C. amoyensis Wang, 1935 (p. 26). C. amphicteis Lang, 1948 (p. 26). C. amphora Kahl, 1928 = Vaginicola amphora Kahl (1928), 1935 C. amphorella Maskell, 1887 = Vaginicola amphorella (Maskell 1887) Kahl, 1935. . angusta Kahl, 1933 (p. 26). . annulata Stokes, 1885 (p. 26). . anomala Stiller, 1951 (p. 26). . antarctica (Daday, 1911) n. comb. (p. 27). . aplatita Stiller, 1939 (p. 27). . aplatita var. flexa Felinska, 1965 = C. aplatita n. comb. . apseudophila Lang, 1948 (p. 27). arcuata Mereschkowsky, 1879 (p. 27). arenata Kent, 1882 = C. arcuata (Kent, 1882) Zelinka, 1928. asimmetrica Banina and Polyakova, 1977 (p. 27). . astaci Stein, 1854 (p. 27). asymmetrica Sommer, 1951 (p. 27). . auriculata Stiller, 1939 (p. 29). . auriculata var. flexa Felinska, 1965 = C. auriculata n. comb. . bavarica Matthes & Guhl, 1973 (p. 29). . bipartita Stokes, 1885 = Cyclodonta bipartita (Stokes, 1885) Matthes 1958. C. brevistyla Nenninger, 1948 (p. 29). C. butschlii Zelinka, 1913 (p. 29). C. calix (Wailes, 1928) Kahl, 1935 = C. spissa Fromentel, 1874 n. comb. C. canthocampti Stokes, 1886 (p. 29). C. carinogammari Stiller, 1953 (p. 29). C. carteri Kent, 1881 = Pyxicola constricta Stokes, 1884 (Kahl, 1935). C. castellensis Penard, 1914 = Thuricola kellicottiana (Stokes, 1887) Kahl, 1935. AAAAAIAAADAGAAAADAO DFYV™YV A AGXYVLMQ YOOX XUOYQUS, A Aiea | | | / | A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES C. ceramicola Kahl, 1933 (p. 29). C. ceratophylli Penard, 1922 = Vaginiola ceratophylli (Penard, 1922) Kahl, 1935. C. chaperoni Penard, 1922 = Vaginicola chaperoni (Penard, 1922) Kahl, 1935. . clausiens Stiller, 1951 (p. 31). . coarctata Kahl 1933 (p. 31). . cohni (Cohn, 1866) Kent, 1881 (p. 31). . collaris Kahl, 1933 (p. 31). . collaris var. incisa Felinska, 1965 = C. collaris. . complanata Precht, 1935 (p. 31). compressa Claparede & Lachmann, 1858 (p. 33). . compressa var. compressula = C. felinska (Felinska, 1965) n. sp. compressa var. cyathiformis Felinska, 1965 = C. cyathiforme Stiller, 1939. . compressa f. flexa Wailes, 1928 = C. compressa. . compressa var. ovata Dons, 1922 = Vaginicola ovata (Dons, 1922) Kahl, 1935. compressa var. sinuosa Wailes, 1943 = C. compressa. . compressula (Wailes, 1928) Kahl, 1935 = C. compressa. . cordylophorae Kahl, 1933 (p. 33). . corrugata Davis, 1879 = Pyxicola socialis (Trueba, 1978). . cothurnoides Blochmann, 1886 = Pyxicola pusilla (Trueba, 1978). . cratera Swarczewsky, 1930 = Vaginicola cratera Kahl, 1935. . crystallina sensu Entz, 1884 = Vaginicola crystallina (Entz, 1884) Kahl, 1935. C. crystallina sensu Entz, 1904 = Pseudothuricola dyonsii pro parte (Entz, 1904) Trueba, 1980, and Thuricola kellicottiana pro parte (Entz, 1904) Trueba, 1980. C. crystallina sensu Penard, 1922 = Thuricola gracilis (Penard, 1922) Trueba, 1980. C. curva Stein, 1867 (p. 33). C. curvula Entz, 1884 (p. 33). C. curvula sensu Dons, 1922 = Vaginicola curvula (Dons, 1922) Kahl, 1935. . cyathiforme Stiller, 1939 (p. 33). . cyathus Jaworowski, 1893 = C. maritima Ehrenberg, 1838. . cyclopis Kahl , 1933 (p. 33). . cylidrica Sommer, 1951 (p. 33). . cypridicola Kahl, 1933 (p. 35). . cytherideae Kahl, 1933 (p. 35). . doliola Stiller, 1939, was originally described as a provisional species (‘sp. prov.’) and has yet to be designated species status. C. doliolum Penard, 1914 = Vaginicola doliolum (Penard, 1914) Kahl, 1935. C. dubia Stiller, 1939 was originally described as a provisional species (‘sp. prov.’) and has yet to be designated species status. C. elastica (Penard, 1914) Kahl, 1935 = Cothurniopsis valvata Stokes, 1893. C. elegans Stiller, 1940 (p. 35). C. elegans Swarczewsky, 1930 = Vaginicola elegans (Swarczewsky, 1930) Kahl, 1935. C. elongata Felinska, 1965 (p. 35). C. elongata Fromentel, 1876 = Vaginicola elongata (Fromentel, 1876) Kahl, 1935. C. endostyla Jankowski. Although this species was mentioned by Jankowski (1985), no description of it could be located. C. entzi Kahl, 1933 (p. 35). C. fecunda Stokes, 1893 (p. 35). C. felinska (Felinska, 1965) n. sp. (p. 35). C. fibripes Kahl, 1933 (p. 36). C. flexa Kahl, 1935 = C. compressa Claparede and Lachmann, 1858. C. floscularia Perty, 1852 (p. 36). DWwOXVOQVAO) SG _ C. furcifer Hutton, 1878 = Pyxicola pusilla (Hutton, 1878) Trueba, 1978. C. fusiformis Gourret & Roeser, 1886 = C. arcuata (Zelinka, 1928). C. gammari Precht, 1935 (p. 36). >) C. gigantea d’Udekem, 1864 = Vaginicola gigantea (d’Udekem, 1864) Kahl, 1935. C. globosa d’Udekem, 1864 = Vaginicola globosa (d’Udekem, 1864) Kahl, 1935. C. gracilis Kent, 1881 = C. curva. C. halacaricola Precht, 1935 (p. 36). C. harpactici Kahl, 1933 (p. 36). C. havniensis Ehrenberg, 1838 = Acineta compressa Claparede and Lachmann 1859 (Curds, 1985). C. hofkeri Kahl, 1933 = C. innata Miller, 1786. C. imberbis Ehrenberg, 1831 (p. 26). C. imberbis var. limbata Stiller = Pyxicola limbata (Stiller) Kahl, 1935. C. incisa Daday, 1907 = Thuricola incisa (Daday, 1907) Trueba, 1980. . inclinans Felinska, 1965 (p. 36). . inflata Stokes, 1893 (p. 37). . inflecta Stiller, 1939 (p. 37). . innata Miller, 1786 (p. 37). irregularis Kent, 1881 (p. 37). kahli Banina and Polyakova, 1977 (p. 37). . kellicottiana Stokes, 1887 = Thuricola kellicottiana (Stokes, 1887) Kahl, 1935. lapponum Penard, 1922 (p. 38) lapponum naidongensis = C. lapponum Wang Jiaji, 1977. lata Kellicott, 1883 (p. 40). . lata (sensu Wailes; 1928) Kahl, 1935 = C. lata Kellicott, 1883. . ligiae Cuénot, 1891 = Pyxicola ligiae (Cuénot, 1891) Kahl, 1935. . limnoriae Dons, 1928 (p. 40). . lobata Daday, 1907 = Vaginicola lobata (Daday, 1907) Kahl, 1935. C. longipes Kellicott, 1894 (p. 40). C. longipes (sensu Mereschkowsky, 1879) Kahl, 1935 = C. longipes Kellicott, 1894. C. longipes sensu Voigt, 1902 = Cyclodonta bipartita (Stokes, 1885) Matthes, 1958. . macrodisca Stiller (p. 40). . magna Yunfen, 1980 (p. 40). . marina Andrussowa, 1886 = C. maritima Ehrenberg, 1838. . maritima Ehrenberg, 1838 (p. 40). . maritima sensu Mobius, 1888 = C. mobiusi (Mobius, 1888) Stiller, 1939. . membranoloricata Stiller, 1968 (p. 42). . minutissima (Penard, 1914) Kahl, 1935 (p. 42). . mobiusi (Mobius, 1888) Stiller, 1939 (p. 42). . mobiusi mihi (Mobius, 1888) Stiller, 1939 = C. mobiusi. . monoannulata Banina & Polyakova, 1977 (p. 42). . nebaliae Dons, 1928 = Dimorphocothurnia nebaliae (Dons, 1928) Jankowski, 1980. C. nereicola Precht, 1935 (p. 42). C. nitocrae Precht, 1935 (p. 42). C. nodosa Claparede & Lachmann, 1858 (p. 44). C. nodosa sensu Fromentel, 1874 = C. irregularis (Fromentel, 1874) Kent, 1881. C. nodosa var. longipes Mereschkowsky, 1879 = C. longipes (Mereschkowsky, 1879) Kahl, 1935. C. obliqua Bock, 1952 (p. 44). C. oblonga Kahl, 1935 (p. 44). C. operculigera Kent, 1869 = Pyxicola operculigera Kent (1869), 1881. C. operculata Gruber, (1879), 1880 = Thuricola valvata (Gruber, 1879, 1880) Kahl, 1935. C. ovalis (Wailes, 1928) Kahl, 1935 (p. 44). C. ovata Fromentel, 1874 (p. 44). C. oviformis Banina and Polyakova, 1977 (p. 44). C. paguri André, 1910 = Vaginicola paguri (André 1910) Kahl, 1935. Aga aane DADDY OO DALAM A© AAA aA OO 58 C. parallela Maskell, 1887 = Vaginicola parallela (Maskell, 1887) Kahl, 1935. C. parva Bock,.1952 (p. 45). C. parvula Felinska, 1965 (p. 45). C. patellae Hutton (1878), 1881 = Mantoscyphidia patellae (Hutton (1878), 1881) Jankowski, 1985. C. patula Fromentel, 1874 (p. 45). C. pedunculata Dons, 1918 = C. trophoniae Dons (1918), 1946. C. peloscolicis Precht, 1935 (p. 45). C. plachteri Matthes & Guhl, 1973 (p. 45). C. plectostyla Stokes, 1885 (p. 45). C. poculum (Wailes, 1928) Kahl, 1933 = C. patula. C. pontica Mereschkowsky, 1881 = Vaginicola pontica (Mereschkowsky, 1881) Kahl, 1935. C. propinqua Kahl, 1933 (p. 45). C. pupa Eichwald, 1849 (p. 45). C. pupa Cohn, 1866 = C. cohni (Cohn, 1866) Kent, 1881. C. pusilla Wrzesniowski, 1870 = Pyxicola pusilla (Wrzesniowski, 1870) Kahl, 1935. C. putanea Jaworowski, 1893 = C. imberbis Ehrenberg, 1831 (Kahl, 1935). C. pyxidiformis d’Udekem, 1864 = Pyxicola pyxidiformis (d’Udekem, 1864) Trueba, 1978. C. pyxidiformis var. lacustris Maggi, 1879 = Pyxicola pyxidiformis var. lacustris (Maggi, 1879) Trueba, 1978. C. recurva Claparede & Lachmann, 1858 (p. 45). C. recurvata Kahl, (p. 47). C. regalis Penard, 1914 = Thuricola valvata Wright, 1858 (Trueba, 1980). C. rhabdota Bock, 1952 (p. 47). C. rheotypica Stiller, 1931 = Cyclodonta bipartita Matthes, 1958 (Felinska, 1965). C. richtersi (Penard, 1914) Kahl, 1935 (p. 47). C. ruthae Lipkes, 1974 (p. 47). C. sahrhagei (Sahrhage) Kahl, 1933 = C. nodosa. C. sediculum Penard, 1914 = Vaginicola sediculum (Penard, 1914) Kahl, 1935. C. sieboldii Stein, 1854 (p. 47). C. simplex Kahl, 1933 (p. 47). C. sinuata Kahl, 1933 = C. minutissima. C. sinuosa (Wailes, 1943) n. comb. (p. 49). C. socialis Gruber (1879), 1880 = Pyxicola socialis (Gruber 1879, 1880) Kent, 1881. C. soldida Vuxanovici, 1962 (p. 49). C. spinosa Labbé, 1895. Only a brief description of this species exists and it has never been figured. C. spinosa is therefore declared a nomen nudun. C. spissa Fromentel, 1874 (p. 49). C. striata Gourret & Roeser, 1886 = Vaginicola striata (Gourret & Roeser, 1886) Kahl, 1935. . Stylarioides Precht, 1935 (p. 49). . subglobosa (Daday, 1911) n. comb. (p. 49). . subtilis Stiller, 1939 (p. 49). . sulcata Kahl, 1928 = Vaginicola sulcata Kahl (1928), 1935. . tekirghiolica Tucolesco, 1962 (p. 49). . terricola (Greeff, 1888) Penard, 1914 = Vaginicola terricola Greeff, 1888. C. thuricolae Shubernetskii, 1978 = Pyxicola thuricolae (Shubernetskii, 1978) n. comb. C. triangula Precht, 1935 (p. 51). C. trilobata Sramek-Husek, 1957 = Cyclodonta bipartita Stokes, 1885 (Matthes, 1958). C. trophoniae Dons (1918), 1946 (p. 51). C. trophonicola Dons, 1946 (p. 51). C. vaga Roux, 1901 (p. 51). C. valvata (Stokes, 1893) Kahl, 1935 = Cothurniopsis valvata Stokes, 1893. CVG Gre: Ga ALAN WARREN & JAN PAYNTER C. valvata Dons, 1922 = Thuricola valvata Dons, 1922 (Trueba, 1980). C. variabilis Kellicott, 1883 (p. 51). C. variabilis var. emarginata Kellicott, 1883 = C. variabilis. C. vas Swarczewsky, 1930 = Vaginicola vas (Swarczewsky, 1930) Kahl, 1935. C. virgula Penard, 1914 = Vaginicola virgula (Penard, 1914) Kahl, 1935. C. voigti (Voigt, 1902) Mereschkowsky, 1933 = Cyclodonta bipartita Stokes, 1885 (Matthes, 1958). Cothurniopsis C. annulata (Stokes, 1885) Penard, 1922 = Cothurnia annulata. C. antarctica Daday, 1911 = Cothurnia antarctica (Daday, 1911) n. comb. C. astaci (Stein, 1854) Entz, 1884 = Cothurnia astaci. C. aurea Gajewskaja, 1933 = Vaginicola vas (Swarczewsky, 1930) Kahl, 1935. C. canthocampti Monard, 1919 = Cothurnia canthocampti Stokes, 1886. C. dionysii Penard, 1914 = Pseudothuricola dionysii (Penard, 1914) Kahl, 1935. C. elastica Penard, 1914 = C. valvata Stokes, 1893. C. entzii Stiller, 1931 = Pyxicola entzii (Stiller, 1931) Kahl, 1935. C. longipes Voigt, 1902 = Cyclodonta bipartita Stokes, 1885 (Penard, 1922). C. minutissima Penard, 1914 = Cothurnia minutissima (Penard, 1914) Kahl, 1935. C. ovata (Dujardin, 1841) Zelinka, 1928 = Cothurnia ovata. C. richtersi Penard, 1914 = Cothurnia richtersi (Penard, 1914) Kahl, 1935. C. subglobosa Daday, 1911 = Cothurnia subglobosa (Daday, 1911) n. comb. C. urceus Gajewskaja, 1933 = Vaginicola vas (Swarczewsky, 1930) Kahl, 1935. C. valvata Stokes, 1893 (p. 52). C. vejdovskii (Vejdovski, 1882) Zelinka, 1928 = Cothurnia oblonga. Cyclodonta Cyclodonta bipartita (Stokes, 1885) Matthes, 1958 (p. 52). Daurotheca D. marginata (Kellicott, 1883) Jankowski, 1987 = Cothurnia variabilis. D. tespa Jankowski, 1987 (p. 54). D. transoceanica Jankowski, 1987 (p. 54). D. ussurina Jankowski, 1987 (p. 54). D. variabilis (Kellicott, 1883) Jankowski, 1987 = Cothurnia variabilis. Dimorphocothurnia D. nebaliae (Dons, 1928) Jankowski, 1985 (p. 54). Semicothurnia S. acuta (Levander, 1915) Jankowski, 1976 = Cothurnia acuta. S. amphicteis (Lang, 1948) Jankowski, 1985 = Cothurnia amphicteis. S. nereicola (Precht, 1935) Jankowski, 1976 = Cothurnia nereicola. Sincothurnia S. apseudophila (Lang, 1948) Jankowski, 1985 = Cothurnia apseudophila. S. auriculata (Stiller, 1939) Jankowski, 1985 = Cothurnia auriculata. S. compressa (Claparede and Lachmann, 1858) Jankowski, 1985 = Cothurnia compressa. A REVISION OF COTHURNIA AND ITS MORPHOLOGICAL RELATIVES S. compressula (Wailes, 1928) Jankowski, 1985 = Cothurnia compressa. S. cyathiformis (Stiller, 1939) Jankowski, 1985 = Cothurnia cyathiforme. elongata (Felinska, 1965) Jankowski, 1985 = Cothurnia elongata. flexa (Felinska, 1965) Jankowski, 1985 = Cothurnia auriculata. imberbis (Ehrenberg, 1831) Jankowski, 1985 = Cothurnia imberbis. isonica Jankowski. Although figured by Jankowski (1985), no description of this species could be located. S. isonica appears to resemble Cothurnia tekirghiolica. S. S. S. gammari (Precht, 1935) Jankowski, 1985 = Cothurnia gammari. S. S. 59 S. monoannulata (Banina & Polyakova, 1977) Jankowski, 1985 = Cothurnia monoannulata. S. paguri (André, 1910) Jankowski, 1985 = Vaginicola paguri (Kahl, 1935). S. patula (Fromentel, 1874) Jankowski, 1985 = Cothurnia patula. S. rhabdota (Bock, 1952) Jankowski, 1985 = Cothurnia rhabdota. S. sinuosa (Wailes, 1943) Jankowski, 1985 = Cothurnia sinuosa. Tesnotheca T. limnoriae (Dons, 1928) Jankowski, 1987 = Cothurnia limnoriae. i, jetmahis chittiad Wlpaind WS dealasiasStealiomesineees satan ven vrretnnarhy deat ads mA, own, eA Us < pbs Penns, by! pesky yeu einer 20h sire ject (Pet, Frat s witli, 2 ; AV wii? ip jhiverpag chap Ligand >. =e 2 ee ey a ee ufnal {Lal wba wouas 2 iPr @ shirentet. arty Sari Wh » « 7 Xi t ws ye “wee e MyT “arr Ls \ t = Pi * - hi i f uae 7. =e =/* ats -_ ey" a mw -) © a i y a . aad - Uy — (ee ° Le WvIgw-s a i; * s v AOy.aie ae a“) cp i. | SUAS *;. o! | oe, ovedignes) doet (Hei ines -sncaes 1 i 1, (GIS 2 ©. unas iA P aviideria wally { - bd 1 z + ey 1 .c i ” 7 <( a‘) L= 7 1 : ot awe Ch oe ee ! = % . j » so iy {= a. — -. s z q a ; - —— — — n 6 = > = a os ° -_ : = é ‘ = m. @ s .* = 1e f 7 a 4 A 7 Z —= eh Or “ i a hs , ie Dive - , 2 $ » -. - 7, ; >» P ond +d = : = ¢! re e —-_ + . . . . ) { 1 ~ a cores = nil : f 4 = >% ip tas ‘ u ’ wv = > a an 1 3 4 — - ! | = 1 ate F a) ‘i j J codhat OED ria LO Co wd > thes 7% rating anpae ae i aupiitins < ( geiyar® (led Teas ‘ ee re ; ? gcee pte, a ee en ba ie hed ment iiFtPA 4 \ 5 oe L we? | 7 om | oo Teaeitt oi 7%. 53 Mi é "a —g j fF oh ; ; tt) Toten Ty = . i hatesuret (v1) «Cae ies a \- af 14 = ® ’y sh f diy iH bo rey i mo gin \ a : i J bie ; cial " b ‘ que is f 4 7 j | 2 ee — J sy nai > ~~. / , ; rj ”" oe. i a ve i © s oi ? ik. i} 0 a* ’ ’ ; ; 2 1 s y eT, ves } {, t tai Gy s t Bull. Br. Mus. nat. Hist (Zool.) 57(1): 61-70 Issued 30 May 1991 Indian Ocean echinoderms collected during the Sindbad Voyage (1980-81): 2. Asteroidea LOISETTE M. MARSH Western Australian Museum, Francis St., Perth, Western Australia, 6000. ANDREW R. G. PRICE* Tropical Marine Research Unit, Department of Biology, University of York, York YOI 5DD, U.K. CONTENTS Se M MUG CCOMINE bo... Ba a oe wisi be ane haw a gp well doe PIGRMONWICCGeCIMeMiS £hNAe cau cena ceed ee adn dee dewens EVES SD EES eke a UC ECE RR ee Ea Ee RO” ES SYNOPSIS. Fifteen species of shallow-water Asteroidea, representing nine genera in four families, are recorded from collections made during the Sindbad Voyage (Oman to China) from the Lakshadweep (Laccadive) Islands, Sri Lanka and Pula Wé (Sumatra). Paraferdina laccadivensis James 1973 is re-described and a new species (P. sohariae) is described. Range extensions are recorded for P. laccadivensis, Fromia nodosa and Thromidia seychellesensis. INTRODUCTION The coral reef asteroids of the islands of the western Indian Ocean are fairly well known e.g. Aldabra (Sloan, Clark and Taylor 1979), the Seychelles (Mébius 1880, A. M. Clark 1984, Jangoux & Aziz 1984), Maldives (Clark & Davies 1966, Jangoux & Aziz 1984), Mauritius (Mobius 1880, de Loriol 1885), and Réunion (Guille & Ribes 1981). The echinoderms of Sri Lanka (Ceylon) were described by Déderlein (1889) and H. L. Clark (1915) and of India and adjacent areas by Koehler (1910). The asteroids of the eastern half of Indonesia are also fairly well known (de Loriol 1893, Déderlein 1896, 1917, 1920, 1924, 1935, 1936, Pfeffer 1900, Engel 1938a and b, Guille & Jangoux 1978, Jangoux 1978, Aziz 1976, 1979). However the eastern Indian Ocean is less well known; small collections have been described from the Cocos (Keeling) Islands (A. H. Clark 1950) and Christmas Island (Bell 1887, Fisher 1934 and Gibson-Hill 1947) but the Sumatran fauna is poorly known. The present collection of 53 specimens of 15 species from the Lakshadweep (Laccadive) Islands, Sri Lanka and Pula We (Sumatra) will therefore help to fill some of the gaps. The collection includes one new species (Paraferdina sohariae) and provides range extensions for Paraferdina laccadivensis (from the Laccadives to Sumatra), Fromia nodosa (from the Maldives to Sumatra) and Thromidia seychellesensis (from the Seychelles to Sumatra). * Present address: Centre for Tropical Coastal Management Studies, Department of Biology, The University, Newcastle upon Tyne, NE1 7RU, UK The collections were made during a voyage across the Indian Ocean from Oman to China. The expedition, Sindbad Voyage, was undertaken in 1980-81 using a replica of an ancient Arab sailing vessel, Sohar. Details of the holothurian fauna collected during the Sindbad Voyage have already been published (Price & Reid 1985). MATERIALS AND METHODS Specimens were collected by one of us (A.R.G.P.) and other expedition members at localities from Chetlat island, Lakshadweep (Laccadives), Sri Lanka and Pula Wé (Sumatra). Details of sampling localities are shown in Fig. 1. Sampling was undertaken principally on coral reefs using scuba. At each locality, details of habitat type and depth range were recorded, along with the number of individuals of each asteroid species. Material was fixed and preserved using standard methods (Lincoln & Sheals, 1979). Specimens were ientified by L. M. M. in Perth. The specimens have been divided between the British Museum (Natural History) and the Western Australian Museum (WAM) with a paratype of the new species lodged at the National Museum of Natural History (USNM). An addi- tional specimen of P. sohariae from the Phuket Marine Biological Centre (PMBC), Thailand was examined. SYSTEMATIC ACCOUNT Throughout this account the synonymy has been confined, where possible, to a single reference from which the original LOISETTE M. MARSH & ANDREW R. PRICE re) te) ie) Lo) te) (o) 50 E 60 70 80 90 100 Muscat OMAN Chetlatp ANDAMANS$ ie LACCADIVES «° ‘a : SRI LANKA NICOBARS* * Pula Tangalla ge? ° Galle e 3 MALDIVES & Ug Bau Fig. 1 (A) map of northern Indian Ocean showing sampling areas (@) during Sindbad Voyage, with inset (B) for Pula Wé, Sumatra. description can be traced. Where this is not appropriate the SEE. Clark & Rowe, 1971: 34, 54, pl. 7, figs 3-5. original description is quoted. MATERIAL. 810425A/1 (BMNH_ 1989.11.1.1), 810425D/3, (BMNH_ 1989.11.1.3), 810424B/2, (BMNH 1989.11.1.2), 810422E/5, (BMNH 1989.11.1.5), 810504C/6, 810428E/4, ASTEROIDEA (BMNH 1989.11.1.4), 810420A/9 (WAM 577-89), 810428E/3 OREASTERIDAE (WAM 578-89). COLLECTION SITES. Ug Tapa Gadja, Seulakée, N. Klah I., E. Klah 1. Culcita novaeguineae Miiller and Troschel, 1842. I., S. Klah I., N. Rubiah I., all Pula Wé, Sumatra (8 specimens). INDIAN OCEAN ECHINODERMS COLLECTED DURING THE SINDBAD VOYAGE (1980-81) 63 HABITAT AND DEPTH. Rock, coral, sand and gravel, 2—30 m. REMARKS. The specimens vary from those having numerous pointed tubercles on the reticulum and in the papular areas to one with no tubercles abactinally. All have coarse granules on the actinal surface and a marginal pore-free band. Specimen WAM 577-89 has a few small tubercles in the papular areas while WAM 578-89 has no abactinal tubercles at all. Both approach C. schmideliana in having groups of coarse granules, corresponding to the underlying actinal plates, but have small granules amongst the large ones. Their general facies is closer to that of C. novaeguineae than to C. schmideliana. 2. Choriaster granulatus Litken, 1869 SEE. Clark & Rowe, 1971: 34, 53, frontispiece. MATERIAL. 810424B/3 (BMNH 1989.11.1.6). COLLECTION SITE. Seulak6e, Pula Wé, Sumatra (1 specimen). HABITAT AND DEPTH. Rock, 20-35 m. OPHIDIASTERIDAE 3. Linckia guildingi Gray, 1840 SEE. Clark & Rowe, 1971: 36, 61, fig. 14b, pl. 8, fig. 7. MATERIAL. 810420A/7 (WAM 574-89) (BMNH 1989.11.1.10). COLLECTION SITE. S. W. Klah I., Pula Wé, Sumatra (2 specimens). - HABITAT AND DEPTH. Coral reef, 4—5 m. 4, Linckia laevigata (Linnaeus, 1758) | SEE. Clark & Rowe, 1971: 36, 62. MATERIAL. 810420A/6 (BMNH 1989.11.1.11), 810420A/8 (WAM 575-89). COLLECTION sITE. S. W. Klah I., Pula Wé, Sumatra (2 specimens). HABITAT AND DEPTH. Coral reef, 4—5 m. 5. Linckia multifora (Lamarck, 1816) _ SEE. Clark & Rowe, 1971: 36, 62. MATERIAL. 810124A/2a (WAM _ 589-89) and (BMNH —-1989.11.1.12-15), 810125/B2 (BMNH _ 1990.4.27.13), —810423A/6 (BMNH_ 1990.4.27.8-9, 810423B/1 (WAM 580-89) and (BMNH 1990.4.27.5), 810423C/10 (WAM _ 579-89) and (BMNH 1990.4.27.6-7), 810428A/11 (BMNH _ 1990.4.27.10-12). _ COLLECTION SITE. Ala Gala (Galle), Sri Lanka (7 specimens). | | | HABITAT AND DEPTH. Rock, 8-15 m. _ COLLECTION SITES. Ug Bau, S. Ug Bau, Pula Wé, Sumatra (12 | Specimens). | HABITAT AND DEPTH. Rock, 20-40 m; coral and sand, 10-20 m. _ REMARKS. Several specimens are parasitized by gastropods, _ Thyca sp. (external) and another species (? eulimid) internally. 6. Fromia indica (Perrier, 1869) SEE. Clark & Rowe, 1971: 34, 62. MATERIAL. 810124A/2G (BMNH 1989.11.1.7). COLLECTION SITE. Ala Gala (Galle), Sri Lanka (1 specimen). HABITAT AND DEPTH. On rock, 10-15 m. 7. Fromia monilis Perrier, 1875 SEE. Clark & Rowe, 1971: 36, 62. MATERIAL. 801210B/4 (WAM _ 590-89) 1990.4.27.18). COLLECTION SITE. S. end Chetlat, Laccadive Archipelago (2 specimens). and (BMNH HABITAT AND DEPTH. Coral reef, 18 m. REMARKS. These specimens differ from F. nodosa in having five plates across the base of the ray, between the supero- marginals, and in lacking prominent convex carinal plates. The superomarginals are convex but do not encroach on the abactinal surface as much as in F. nodosa. R/r of the two specimens is 42/10 mm = 4.2/1. 8. Fromia nodosa A. M. Clark, 1967 SEE. A. M. Clark, 1967: 189-191, pl. 6, figs 1-3; Clark & Rowe, 1971: 36, 62, pl. 8, fig. 8. MATERIAL. 810421A/6 (BMNH 1990.4.27.17), 810424B/1G (BMNH_ 1989.11.1.9), 810427D/4 (WAM _ 585-89) and (BMNH 1990.4.27.14-15), 810422E/2G (BMNH 1989.11.1.8), 810425F/6 (BMNH_ 1990.4.27.16), 810428A/11 (WAM 584-89), 810424C/1 (WAM 587-89), 810422C/2 (WAM 586-89), 810423C/10 (WAM 583-89). COLLECTION SITES. S. Ug Bau, Ug Bau, N. Klah I., Seulakée, N. Udjung Lo Me, Ug Seukundo, Ug Murung, all Pula Wé, Sumatra, Indonesia (11 specimens). HABITAT AND DEPTH. Sand, coral gravel, coral and sand, rock 2-35 m. REMARKS. The type locality of this species is Amirante Islands, W. Indian Ocean; Clark also records it from the Maldives. The range was extended to the Marshall Is. (W. Pacific) by Oguro & Sasayama (1984), but this is the first record from Indonesia. The specimens have less prominent carinal plates than the holotype but otherwise agree closely. The superomarginals are large and rounded, nearly as broad as long. There are generally only three plates across the base of the ray between the superomarginals. The specimens range in size from R/r of 42/9 mm to 31/7 mm (R’/r varies from 4.1 to p/p 9. Nardoa frianti Koehler, 1910 SEE. Koehler, 1910: 158-161, pl. 17, figs 3, 4. MATERIAL. 810424B/la (WAM _ 573.89) 1989.11.1.16). COLLECTION SITE. Seulak6e, Pula Wé, Sumatra (2 specimens, R/r = 123/12 mm (larger specimen). and (BMNH HABITAT AND DEPTH. Coral gravel, 20-30 m. REMARKS. The present specimens agree closely with Koehler’s LOISETTE M. MARSH & ANDREW R. PRICE Fig. 2 Paraferdina laccadivensis, a, abactinal view (BMNH 1990. 4.27.1, R/r = 36/12 mm); b, abactinal surface, denuded (WAM 77-90, R/r = 33/11 mm). description and figures. N. frianti differs from N. mamillifera Livingstone, 1930 principally in the disposition of the actinal plates. In N. frianti they are poorly developed and never extend to more than half the arm length. Frequently the series consists of a few plates at the base of the ray with a few more small plates between the large inferomarginals which abut the adambulacrals. If a second series is present at all it consists of a very few plates in the arm angle. In N. mamillifera the actinal plates are well developed with two or more rows at the base of the ray. The main series extends to about two thirds of the arm length with plates nearly as large as those of the inferomarginals in some specimens. Both species have alternating large and small superomarginal plates. N. frianti and N. mamillifera cannot therefore be considered synony- mous, as suggested as a possibility by Clark and Rowe (1971). 10. Nardoa galatheae (Litken, 1864) SEE. Clark & Rowe, 1971: 36, 64. MATERIAL. 810213A/1 (WAM 572-89), 810428B/4 (BMNH 1989.11.1.17). COLLECTION SITES. Tangalla, Sri Lanka (1 specimen, R/r = 69/ 12 mm); S. Ug Bau, Pula Wé, Sumatra (1 specimen, R/r = 140/17 mm). HABITAT AND DEPTH. On rock, 3—20 m. REMARKS. The smaller specimen (apart from having six rays) is indistinguishable from individuals from north-western Australia, which have convex abactinal granules, while the large one has much less convex granules, similar to others of comparable size from the Moluccas. Since one of the latter has a regenerating arm with more convex granules, as found in the smaller specimen, it appears that the more convex shape of the granules is a juvenile character. 11. Neoferdina cumingi (Gray, 1840) SEE. Clark & Rowe, 1971: 36, 65; Jangoux, 1973; 786-789. MATERIAL. 810430A/39 (BMNH_ 1990.4.1.2), 810504B/3 (WAM 581-89) and (BMNH_ 1990.4.27.3), 810423C/10 (WAM 582-89), 810422E/2 (BMNH 1990.4.27.4). COLLECTION SITES. N. Klah, N. Rubiah I., Ug Seukundo, Ug Bau, all Pula Wé, Sumatra, Indonesia (5 specimens). HABITAT AND DEPTH. Rock, coral and sand, 2-10 m. INDIAN OCEAN ECHINODERMS COLLECTED DURING THE SINDBAD VOYAGE (1980-81) 65 PARAFERDINA James 1973 TYPE SPECIES. Paraferdina laccadivensis James 1973. TYPE LOCALITY. Minicoy Island, Lakshadweep (Laccadive) Islands, Indian Ocean. For the sake of completeness James’ (1973) diagnosis of this poorly known genus is repeated here. DIAGNOsIs. A genus of Ophidiasteridae with polygonal aboral plates, irregular in size with spaces for papular pores; supero- marginal plates regular and uniformly covered by granules; actinal plates embedded in tissue; no actinal papulae; a single row of short furrow spines with tips projecting to the outside. REMARKS. Paraferdina differs from Ferdina principally in having a regular series of superomarginals and from Neoferdina in lacking bare plates. 12. Paraferdina laccadivensis James, 1973 SEE. James, 1973: 556-557, pl, la. MATERIAL. 810428A/12 (BMNH_ 1990.4.27.1), 810423C/10 (WAM 77-90), 810422E/2C (WAM 76-90). COLLECTION SITES. Ug Bau, S. Ug Bau, N. Klah I., all Pula Wé, Sumatra, Indonesia (3 specimens). HABITAT AND DEPTH. Mixed coral and sand, 2-8 m; rock 20-30 m. DIAGNOsIS. A species of Paraferdina with short, stout arms, largest known R/r is 42/13 mm, range of R/r ratio 3.0 to 3.2/ 1; abactinal plates irregular in shape but all of similar size, flat to slightly convex aligned along arms or irregularly arranged, 7-9 across base of ray; carinal series distinct but not promi- nent; superomarginal plates large, conspicuous from above, squarish to semiciruclar; actinal plates in 4-5 rows at base of ray; granules on abactinal plates uniform, 9-10/linear mm, on marginals 12/mm and on actinal plates 10—12/mm; abactinal surface uniform light orange, superomarginals magenta, arm tips cream (from one preserved specimen). DESCRIPTION. (Figs 2, 3) The three specmens have R/r of 42/ 13, 36/12 and 33/11 mm giving R/r of 3.2 and 3.0/1. The abactinal plates are circular to irregular in shape, 1-2 mm in diameter and are notched for papulae (Figs 2a, b, 3a). They are nearly flat and vary from being aligned along the arm to being irregularly disposed particularly on the outer third of the arms; the carinal row of 18-20 plates is distinct nearly to _the arm end; a few plates are slightly more convex than the _ remainder and in some cases are larger than the lateral plates. _ At the base of the ray there are 3-4 more or less distinct _ longitudinal rows of plates on either side of the carinal series _decreasing to two distally and becoming more irregular in _ arrangement. On one specimen slightly raised plates form up | to five indistinct rows across the arms. The terminal plate is | rounded, 1.5 to 1.8 mm in diameter. | The superomarginal plates are large, conspicuous when _ viewed from above (Figs 2b, 3a) and number 10-11 (the third | from the arm angle measures 3.0 to 3.5 mm long by 2 to 3 mm | broad) and are regularly arranged except sometimes near the | arm end, they vary from more or less square to rectangular | and D shaped; the last superomarginals meet across the upper Surface of the arm. The inferomarginals number 12-16, the _ Smaller number when they match the superomarginals, the / / / Fig. 3. P. laccadivensis, a, base of arm, denuded, carinal series at top, first superomarginal on left (WAM 77-90); b, ambulacral furrow, showing outer, granule covered face and inner surface of furrow spines (WAM 77-90). (Scale bar = 1 mm.) larger when the plates become smaller and less regular near the arm end. The actinal plates are in four rows at the base of the rays except in the largest specimen where there are 2-3 plates of a fifth row. The abactinal plates are covered by close packed, convex, polygonal granules (9-10/linear mm) except near the arm ends where they are coarser and irregular in size (5—7/mm). In the largest specimen the granules surrounding the papular pores are slightly elongate. The marginal plates have a finer granulation (12 granules/mm) while the actinal plates have 10-12 granules/mm. The adambulacral armature is limited to a single row of paired furrow spines, which are stout, blunt ended, equal to subequal in size (in one specimen measuring 0.8 by 0.5 mm) but somewhat irregular in size and shape. The actinal granulation conceals all but the tips of the furrow spines (Fig. 3b). The papulae are generally single with 6 equally spaced around a hexagonal carinal plate; however, where the plates are less regular, papulae occur in groups of two or three and there may be up to 14 around one plate. The colour of one specimen (preserved in Steedman’s solution) is striking, the abactinal surface is light orange bordered by bright magenta superomarginals except for the arm tips which are cream. The inferomarginals and actinal areas are paler, shading to cream at the furrow; the arm ends and madreporite are cream. One dry specimen has the abactinal surface burnt orange, including the arm tips, with the marginals and actinal surface faded magenta, the largest specimen (dry) has the abactinal surface faded to deep cream; magenta is still evident on the marginals and actinal surface but the colour does not extend to the arm ends. REMARKS. The present specimens, in excellent condition, allow further descriptive details and variation of the species to be noted. The R/r ratio differs from that of the holotype which had R.r of 35/14 mm (2.5/1) (James 1973). The illustration of the holotype, however, indicates longer arms relative to r despite damage to the arm ends. Since it would have been difficult to estimate R accurately the difference in R/r ratio from the present specimens is not regarded as significant. The major difference is in the number of superomarginal plates (15 in the holotype but no more than 12 in the present specimens) and in their shape (the holotype has the supero- marginals broader than long while the Sindbad specimens have them as long as broad or longer). In other respects these specimens closely resemble the holotype. Unfortunately direct comparison with the holotype could not be made and the Sindbad specimens are referred to this species with a little hestitation. 66 The colour is described for the first time. We agree with James (1973) that this species belongs in a genus between Ferdina and Neoferdina and consider it to be more closely related to the latter because of the plate arrangement and prominent (but not bare) superomarginal plates. 13. Paraferdina sohariae sp. nov. Ferdina offreti: (pt.) Koehler, 1910: 143, 147, pl. 16, Figs 4, 5. HOLOTYPE. 810125A/1 (WAM 78-90). TYPE LOCALITY. Deumba Gala, Galle, Sri Lanka, on rock 12- 15 m. MATERIAL. 810125A/1 four paratypes (WAM 79-90, BMNH 1990.4.4.2-3, USNM E40225) same data as holotype; PMBC 3048, Similan I., Andaman Sea, coral reef, 14 m, 12 Feb. 1979. HABITAT AND DEPTH. Rock, 12-15 m; 62 m (Koehler, 1910). DIAGNOsIS. A_ species of Paraferdina with somewhat flattened, tapering arms, largest known R/r is 50/15 mm, range of R/r ratio 3.3 to 4.3/1; abactinal plates markedly variable in shape, size and convexity, 5—7 across base of ray; usually a prominent carinal series of elevated, slightly convex plates sometimes alternating with smaller convex or flat plates; usually smaller flat plates among the larger ones on disc or arms; superomarginal plates large, conspicuous from above, D-shaped; actinal plates in 4-5 rows at base of rays; granules on abactinal plates very variable, from 5-6 to 9-10/ linear mm on convex plates, 12/mm on interstitial and flat plates, 9-12/mm on actinal plates; granulation is coarser near arm ends than proximally; 1-5 enlarged granules sometimes present on convex abactinal plates; colour in life unknown, preserved specimens have buff or yellow-brown granules on the convex plates contrasting with orange to deep rose-red granules on the interstitial and flat plates. DESCRIPTION OF HOLOTYPE. (Figs 4, 6) Rays 5, tapering, R/r = 40/11 mm, = (3.6/1), br = 11.5 mm. The abactinal surface of the disc and rays is convex, with convex, rounded to poly- gonal, granule covered abactinal plates and large supero- marginals (Fig. 4a, b). There are 22-23 carinal plates; on the outer two thirds of the ray prominent convex carinals alternate irregularly with smaller flat plates. At the base of the ray there are five to seven plates across the arm, two or three each side of the carinal series, by about half the arm length this is reduced to one each side of the carinal series. The abactinal plates are irregular in shape and size, 0.8-2.7 mm diameter, close fitting and slightly notched for the papulae. On the disc there are five prominent primary interradial plates (including the madreporite) and a prominent radial plate at the beginning of each carinal series. The remainder of the disc is covered by smaller, irregularly arranged plates. The madreporite is small, rounded triangular, convex, situated interradially, half way between the anus and the margin. The superomarginals number 12-13, the first pair are wider than long (3 X 2 mm) but the remainder are about as wide as long except distally where they are again wider than long. They are prominent, more or less semicircular and gradually decrease in size distally, the last 2-4 meet mid radially. The inferomarginals number 15 and are aligned with the supero- marginals except at the arm tips; they are longer than wide (3 x 1.5 mm) except for the first one which is square. LOISETTE M. MARSH & ANDREW R. PRICE The actinal plates are in 4 rows proximally, (Fig. 4c), with a few plates of a 5th row in some arm angles. The innermost row extends nearly to the arm end (to the 11" or 12™ infero- marginal), the second row extends to the 9" or 10" infero- marginal, the third row to the 6" or 7™ infero- marginal and the fourth to the 3"? or 4" inferomarginal. There is a single row of adambulacral spines along the furrow margin, only the tips of which project above the adjacent granules. The furrow spines are subequal, truncate, 1.05 mm long, two or occasionally three per plate (Fig. 6a). The four oral spines are undifferentiated from the furrow series. The entire surface is granule covered; the abactinal convex plates are more coarsely granulated (9-10/linear mm) than the smaller plates and the marginals (12/mm). In addition some of the convex plates bear 1-5 enlarged tubercle-like granules (Fig. 6b). Near the arm ends the granulation of abactinal, terminal and marginal plates is very coarse (6—7/ mm). The actinal plates have 10-12 granules/mm. The papulae occur singly, occasionally in groups of 2, not in regular rows but around the plates e.g. there are 6-8 around a carinal plate. There are no actinal or intermarginal papulae. Pedicellariae are absent. Colour (in alcohol).The granules covering the raised abac- tinal plates, marginals and actinal plates are buff, those covering the flat plates and grooves between the marginal and actinal plates are orange. OTHER MATERIAL. The four paratypes show considerable variablility in the R/r ratio and in the number and disposition of convex abactinal plates giving a very different appearance. The colour pattern is constant but the shade of orange is darker in some. Paratype WAM 79-90 (Fig. 5a) has R/r of 36/11 mm (3.3/1). The carinal plates do not form a regular series on all the arms, fewer of them alternate with smaller plates than in the holotype and there are more large convex plates (Fig. 6c); the arm plates are from 1—3 mm in diameter with coarse granula- tion on the larger plates (7—8/linear mm), finer on the smaller abactinal plates and on the actinals (10-11/mm); there are 5 plates across the base of the ray; 10-11 superomarginals and 13-15 inferomarginals. The actinal plates are in 4 rows with a few extra plates in the arm angle; the third row extends to the 5 or 6 inferomarginal, the fourth to the third inferomarginal. Paratype BMNH 1990.4.4.2 (Fig. 4d) has R/r of 37/10 (3.7/1). The carinal series is irregular except near the end of the arms and the abactinal surface is covered by large convex plates. On the proximal half of the arm there are three irregular rows with a small plate either side on some rays. There are 13 superomarginals and 15—16 inferomarginals. The actinal plates are in 3 rows proximally, with a few plates of a 4" row. Paratype BMNH 1990.4.4.3 is an irregular specimen with three arms regenerating. R/r = 38/10 mm (3.8/1). The abac- tinal plates are very irregularly arranged and the carinal series in only evident on one ray. Paratype USNM E40225 (Fig. 5b) has R/r of 43/10 (4.3/1). The carinal series is only regular on one ray where large and small plates alternate distally. There are 5 plates across the base of the ray; 12-13 superomarginals and 16 inferomarginals. The actinal plates are in 4 rows proximally. The granulation is much coarser than on the holotype with 5—6 granules/mm on INDIAN OCEAN ECHINODERMS COLLECTED DURING THE SINDBAD VOYAGE (19 = ) Fig. 4 Paraferdina sohariae n. sp. a, b, abactinal views of holotype (WAM 78-90, R/r = 40/11 mm); c, actinal view of holotype; d, abactinal view of paratype (BMNH 1990.4.4.2, R/r = 37/10 mm). 68 LOISETTE M. MARSH & ANDREW R. PRICE Fig. 5 P. sohariae n. sp. a, abactinal view of paratype, arm partly denuded (WAM 79-90, R/r = 36/11 mm); b, abactinal view of paratype (USNM E40225, R/r = 43/10 mm). Fig.6 P. sohariae n. sp. a, ambulacral furrow, showing outer granule covered face and inner surface of furrow spines (holotype, WAM 78-90); b, granulation of an abactinal plate with two tubercular granules (holotype) c, base of arm denuded (paratype WAM 79-90). (Scale bar=1 mm). the abactinal and marginal plates with a few tuberculate granules and 9-10/mm on the actinal plates. Only two of the paratypes have any tubercle-like granules on the abactinal plates and these are fewer than in the holotype. PMBC 3048 (Fig. 7) has R/r of c. 50/15 mm (3.3/1). The R value is approximate because of damage to the arm ends. The carinal plates are enlarged, convex, alternating irregularly with smaller, flat plates. There are five plates across the base of the ray, distally three; 14 superomarginals and about the same number of infero-marginals, distally the last 2-3 supero- marginals meet across the ray. The actinal plates are in three rows with a fourth in the arm — angle, the third row extends to the fifth or sixth inferomarginal, the fourth row to the second or third inferomarginal. The whole of the abactinal and actinal surfaces are covered by a uniform coat of fine granules, 11/linear mm. A colour slide of the dried specimen, photographed in 1982, shows the convex abactinal plates and marginals as light yellow brown; the disc, some small plates and grooves between the convex plates and marginals were deep rose-red while the actinal plates were cream to pink outlined in old rose, yellowish at the arm ends; the furrow spines were INDIAN OCEAN ECHINODERMS COLLECTED DURING THE SINDBAD VOYAGE (1980-81) 69 | Fig. 7 P. sohariae, dried (PMBC 3048, R/r = 50/15 mm). cream. This example differs from the Sindbad specimens in its larger size, lack of convex plates on the disc and uniform _ granulation of the plates. However, in view of the variability shown by the five specimens from Sri Lanka the differences are outweighed by the similarities and we consider that this _ specimen also represents Paraferdina sohariae. A small specimen from Ceylon (Sri Lanka), with R of 13.5 mm, referred to as a juvenile Ferdina offreti by Koehler (1910) and believed to represent Fromia nodosa by A. M. Clark (1967) cannot be placed in Fromia because of the _ adambulacral armature. Koehler included it in Ferdina as it had a single row of spines on the adambulacral plates but noted that the marginal plates were granule covered, unlike those of the holotype. This specimen is now referred to | Paraferdina and clearly represents P. sohariae. RANGE OF VARIATION. The size (R) ranges from 36 to 50 mm _ with an R/r ratio of 3.3 to 4.3/1 (mean 3.7/1). They have five to seven plates across the base of the rays between the _ Superomarginals, however there is considerable variation in | the relative numbers of convex and flat plates. There are _14 superomarginals in the Thai specimen, 12-13 in three _ specimens, 10-11 in one and those of the remaining specimen are too irregular to count. Inferomarginals number 15-16 in _ three specimens. 13-15 in another. The abactinal plates are _ coarsely granulated with 9-10 granules/linear mm, while the |Marginals have 12/mm except near the arm ends where _abactinal, terminal and marginal plates have very coarse granulation (6—-7/mm). Actinal plates have 10-12 granules/ mm in the holotype, 9—-10/mm in one of the paratypes. Three of the paratypes have slightly coarser granulation than the holotype while one has a much coarser granule cover. The Thai specimen has a uniformly fine granule cover with 11/linear mm. ETYMOLOGY. Named after the expedition ship, Sohar. REMARKS. This species clearly belongs in the genus Paraferdina since it has conspicuous granule covered marginal plates, polygonal abactinal plates, irregular in size, no actinal papulae and a single row of furrow spines. P. sohariae differs from P. laccadivensis in having more tapering rays, more convex abactinal plates which are very variable in size and up to 3 mm in diameter. The differences are emphasised by the distinctive colour pattern of each species. DISTRIBUTION. This species is known from only two localities—the type locality (Sri Lanka) and Similan I., Andaman Sea. MITHRODIIDAE 14. Thromidia seychellesensis Pope & Rowe, 1977. SEE. Pope & Rowe, 1977: 207-210, figs 7, 8, 11. MATERIAL. 810504D/1 (WAM 576-89). 70 COLLECTION sITE. N. W. Klah I., Pula Wé, Sumatra (1 specimen). HABITAT AND DEPTH. Coral rubble, 10 m. REMARKS. The specimen is badly distorted but R/r = c. 170/50 mm (3.4/1) compared to the holotype and paratype which have R/r of 124/22.5 mm (5.5/1) and 135/27.5 mm (4.9/1) respectively. Papular areas are 24 mm in diameter with 20— 100 pores on the disc and arm bases, decreasing to single pores near the arm ends. On the distal fifth of the arms, convex, granule covered plates form a cobbled pavement as described by Pope and Rowe. The granulation is as described for the holotype. The present specimen with R 45 to 35 mm longer than the holotype and paratype confirms Pope and Rowe’s (1977) suggestion that their specimens were not fully grown. This is the first record of this species outside the Seychelles Islands and extends the range of the species from the western Indian Ocean to the eastern Indian Ocean. ACANTHASTERIDAE 15. Acanthaster planci (Linnaeus, 1758) SEE, Clark & Rowe, 1971: 38, 71,'pl. 11, fig. 3. MATERIAL. 810424E/1 (WAM 588-89), 810425G/1 (BMNH 1989.11.1.18). COLLECTION SITES. Nr Rubiah I., Pula Wé, Sumatra (1 specimen R/r = c. 150/70 mm); W. Rubiah I., Pula Wé, Sumatra (1 specimen, R/r, = c. 130/70 mm). HABITAT AND DEPTH. Coral, 1.5—6 m. ACKNOWLEDGEMENTS. We are grateful to Dr R. Dalley, P. Hunnam, P. Dobbs and D. Tattle for their considerable assistance during field work and to Mr Somchai Bussarawit of the Phuket Marine Biological Centre for the loan of a speci- men of Paraferdina sohariae from Thailand. One of us (A. R. G. P.) would also like to thank Tim Severin, leader of the Sindbad Voyage, for the kind invitation to participate in the expedition which was made possible by generous support from the Ministry of National Heritage and Culture, Sultanate of Oman. We also thank A. M. Clark, formerly BM(NH), for helpful comments on the manuscript. REFERENCES Aziz, A. 1976. Echinoderms collected during the Rumphius Expedition II. Oseanologi di Indonesia 6: 35-38. 1979. Checklist of the echinoderm collections of the Rumphius Expedi- tion II. Proceedings of the International Symposium on Marine Bio- geography and evolution in the Southern Hemisphere, Auckland, 17-20 July 1978. New Zealand DSIR Information Series 137: 717-726. Bell. F. J. 1887. VIII Echinodermata. Jn: Report on a zoological collection made by the officers of HMS Flying Fish at Christmas Island, Indian Ocean. Proceedings of the Zoological Society of London: 523. Clark, A. H. 1950. Echinoderms from the Cocos-Keeling Islands. Bulletin of the Raffles Museum 22: 53-67. Clark, A. M. 1967. Notes on asteroids in the British Museum (Natural History) V. Nardoa and some other ophidiasterids. Bulletin of the British Museum (Natural History), Zoology 15: 169-198, 6 pls. 1984. Echinodermata of the Seychelles. In Stoddart, D. R. (ed.), Biogeography and ecology of the Seychelles Islands, Junk, The Hague: 83-102. & Spencer Davies, P. 1966. Echinoderms of the Maldive Islands. Annals and Magazine of Natural History Series 13 8: 597-612, pl. 18. LOISETTE M. MARSH & ANDREW R. PRICE — & Rowe, F. W. E. 1971. Monograph of shallow-water Indo-west Pacific echinoderms. Trustees of the British Museum (Natural History): 238 pp, 31 pls. Clark, H. L. 1915. The echinoderms of Ceylon (other than holothurians). Spolia Zeylanica, 10(37): 83-102. Déderlein, L. 1889. Echinodermen von Ceylon. Bericht tiber die von den Herren Dres. Sarasin gesammelten Asteroidea, Ophiuroidea und Echinoidea. Zoologische Jahrbuecher 3: 822-846, pls 31-33. — 1896. Bericht tber die von Herrn Prof. Semon bei Amboina und Thursday Island gesammelten Asteroidea. Jenaische Denkschriften, (Semon, R. Zoologische Forschungreisen, 5)8: 301-322, pls 18-22. 1917. Die Asteriden der Siboga Expedition. I. Die Gattung Astropecten und ihre Stammesgeschichte, Siboga Expedition 46a: 1-191, pls 1-17. 1920. Die Asteriden der Siboga Expedition. I]. Die Gattung Luidia und ihre Stammesgeschichte, Siboga Expedition 46b: 193-291, pls 18-20. — 1924. Die Asteriden der Siboga Expedition. II. Pentagonasteridae Siboga Expedition, 46°. 49-69, pls 14-19. — 1935. Die Asteriden der Siboga Expedition. III. Oreasteridae. Siboga Expedition, 46°. 71-110, pls 20-27. — 1936. Die Asteriden der Siboga Expedition. III. Die Unterfamilie Oreasterinae. Siboga Expedition, 46c: 295-369, pls 21-32. Engel, H. 1938a. Astéries du voyage aux Indes Orientales Néerlandaises (1932) de LL. AA. RR. le Prince et la Princesse Leopold de Belgique. Musée Royal Histoire Naturelle de Belgique. Bulletin, 14(18): 1-4, 2 figs. 19385. Résultats scientifiques du voyage aux Indes Orientales Néerlandaises de LL. AA. RR. le Prince et la Princesse Leopold de Belgique. Astéries et Ophiures. Musée Royal Histoire Naturelle de Belgique. Memoires (Hors Serie), 3(18): 1-31, 4 pls. Fisher, A. K. 1934. Note on a starfish from Christmas Island, Indian Ocean. Bulletin of the Raffles Museum, 9: 74. Gibson-Hill, C. A. 1947. The Echinodermata. Museum, 18: 22-26. Guille, A. & Jangoux, M. 1978. Astérides et ophiurides littorales de la region d’Amboine (Indonésie). Annales Institut Océanographique, Paris, 54(1): 47-74. & Ribes, S. 1981. Echinodermes associés aux Scléractiniaires d’un récif frangeant de Vile de La Réunion (Océan Indien). Bulletin du Muséum national d’ Histoire naturelle, Paris, (4) 3 section A, no 1: 73-92. James, D. B. 1973. Studies on Indian echinoderms -5. New and little known starfishes from the Indian Seas. Journal of the Marine Biological Association of India, 15(2): 556-559, 1 pl. Bulletin of the Raffles Jangoux, M. 1973. Le genre Neoferdina Livingstone (Echinodermata, Asteroidea: Ophidiasteridae). Revue de Zoologie Africaine, 87(4): 775-794, 1 pl. — 1978. Biological results of the Snellius Expedition. Echinodermata, Asteroidea. Zoologische Mededelingen Leiden, 52: 287-300. —— & Aziz, A. 1984. Les astérides (Echinodermes) du centre-ouest de l’océan Indian (Seychelles, Maldives et iles Mineures). Bulletin du Muséum national d Historie naturelle, Paris, (4)6, Sect. A., no. 4: 857-884, pls 1-4. Koehler, R. 1910. Shallow-water Asteroidea. Echinoderma of the Indian Museum. Calcutta, 258 pp. 20 pls. Lincoln, R. J. & Sheals, J. G. 1979. Invertebrate Animals: Collection and Preservation. British Museum (Natural History) & Cambridge University Press, pp. 1-150. Loriol, P. de 1885. Catalogue raisonné des échinodermes recueillis par M. V. de Robillard a I’ Ile Maurice. II. Stellérides. Memoires de la Société de Physique et d’ Histoire naturelle de Genéve, 29(4): 1-84, pls 7-22. —— 1893. Echinodermes de la Baie d’Amboine. Revue Suisse de Zoologie, 1: 359-426, pls 13-15. Mobius, K. 1880. Beitrage zur Meeresfauna der Inseln Mauritius und der Seychellen. Berlin, : 1-352, 22 pls. [Echinoderms on pp. 46-50]. Oguro, C. & Sasayama, Y. 1984. Occurrence of Fromia nodosa A. M. Clark (Asteroidea, Ophidiasteridae) from the Marshall Islands, the Western Pacific. Proceedings of the Japanese Society for systematic Zoology, no. 27: 101-106. Pfeffer, G. 1900. Echinodermen von Ternate, Echiniden, Asteriden, Ophiuriden und Comatuliden. Abhandlungen senckenbergischen Naturforschenden Gesellshaft, 25: 81-86. Pope, E. C. & Rowe, F. W. E. 1977. A new genus and two new species in the family Mithrodiidae (Echinodermata: Asteroidea) with comments on the status of the species of Mithrodia Gray 1840. Australian Zoologist,19(2): 201-216. Price, A. R. G. & Reid, C. E. 1985. Indian Ocean echinoderms collected during the Sindbad Voyage (1980-81): 1. Holothuriodea. Bulletin of the British Museum (Natural History) Zoology, 48: 1-9. Sloan, N. A., Clark, A. M. & Taylor, J. D. 1979. The echinoderms of Aldabra and their habitats. Bulletin of the British Museum (Natural History) Zoology, 37(2): 81-128. Manuscript accepted for publication 31 August 1990 Bull. Br. Mus. nat. Hist (Zool.) 57(1): 71-75 The identity and taxonomic status of Tilapia arnoldi Gilchrist and Thompson, 1917 (Teleostei, Cichlidae) PETER HUMPHRY GREENWOOD Visiting Research Fellow, The Natural History Museum, Cromwell Road, London SW7 5BD CONTENTS NiPM GGULGEL © Ia NP aA ne er en he Be See Pd cb vm vg woe Yraaciae o eprady elo hs de epee oie eee oe ee ee 71 NEN NURI GATEHOUSE AS ills cya 2 ao hata ah eS Cee eg ea RC Perea Pe re ee 71 SceSCrintionmammrernOla nIGIOEy eG. None ete ee ss wah doe waddle wen een ee Seas eamoencck. se. dz BRIS fax ONO Miles FAUNIS ORM (APIA OIILOLAU het AEN soit euncthe nm ieiaveeoanas > Srtiadken in ne Pyne Se a en ae wea ee ee 73 PVCRMO WLC Cp cae DRS Reei Eien ee eee 5 ver are Wt es ac ccalshelelauen A ae Seow sey Sig ono, Redd alee tae deklatea ee ee 74 EOC LC ICES nNOS en ea rT eee EEN SAN, MPR Pe Sh RE ME Oe ae Mee ys SEES TC POLY oe 74 SYNOPSIS The type and only specimen of Tilapia arnoldi Gilchrist and Thompson (1917) is redescribed (as far as its poor condition allows) and its taxonomic status reassessed in the light of that study. It is concluded that T. arnoldi should be considered a junior subjective synonym of Tilapia sparrmanii A. Smith, 1840, and not, as previously Issued 30 May 1991 thought, of Oreochromis mossambicus (Peters), 1852. INTRODUCTION Tilapia arnoldi was described by Gilchrist and Thompson (1917) on the basis of a single specimen collected by G. Arnold from the Mazui River, Zimbabwe. Their description, although detailed, lacks a figure, and no specifically diag- _ nostic features are given. However, in a synopsis of the Southern African cichlid species described in that paper the authors paired Tilapia arnoldi with a taxon then considered to be T. melanopleura Dum., 1859, but which is now identified as T. rendalli (Blgr.) 1896 (see Thys van den Audenaerde, 1964). In Gilchrist and Thompson’s synoptic key, 7. arnoldi is distinguished from T. rendalli by its having a lower lateral- _ line scale count (27 cf 28-32) and a higher number of gill-rakers on the lower part of the anterior gill-arch (i.e. 15 cf 8-12). Regan (1922), without examining the type specimen of T. _arnoldi, and without giving any reasons for his opinion, suggested that ‘T. arnoldi may be a synonym of Tilapia natalensis’ (T. natalensis [Weber] 1897 is now consider a synonym of Oreochromis mossambicus [Peters] 1852; see Trewavas [1983] for a detailed synonymy). There matters stood until Barnard (1948a: 49 and 54, b: 448) formally synonymised T. arnoldi with O. mossambicus | (then Tilapia mossambica). Again no reasons were given, but _ judging from Barnard’s (1948 a,b) key to the southern African Tilapia species it was the supposed number of gill- takers (i.e. 15) and the general morphology of T. arnoldi that prompted his action. Barnard’s synonymy has been accepted by subsequent workers (Jackson, 1961; Jubb, 1961, 1967), although Trewavas (1983) omits, without comment, any specific reference to 7. arnoldi in her detailed synonymy of O. mossambicus. However, since she gives references, again without comment, to the Jubb and Jackson papers in that synonymy, one can assume, at least by implication, she too accepted Barnard’s conclusions. Only Bell-Cross and Minshull (1988) have departed from this consensus; they list T. arnoldi under T. sparrmanii in their check-list, but give no reasons for so doing. I have recently examined the holotype of T. arnoldi and, despite its now very poor condition (see Fig. 1), have concluded that it is conspecific with Tilapia sparrmanii A. Smith, 1840 and thus should be considered a junior subjective synonym of that species. Throughout this paper most further references to Gilchrist and Thompson (1917) will, in the interests of brevity, be abbreviated to ‘“G & T’, and references to the papers of Thys van den Audenaerde will be cited as “Thys’. MATERIAL AND METHODS The system of counts and measurements employed follows that detailed by Trewavas (1983). The length of the lower pharyngeal bone is measurd, perpendicularly, from the anterior tip of the blade to a horizontal line through the posterior margin of the bone’s dentigerous surface; the overall breadth of the bone is taken as the horizontal distance between the outer tips of its articular horns. Various diagnostic tilapiine features used by Thys (1964, 1968) and Trewavas (1983) at both the generic and specific levels have been used, as was information contained in the more general accounts of 7. sparrmanii, T. rendalli (A. PETER HUMPHRY GREENWOOD Fig. 1 Dum.), T. ruweti (Poll and Thys) and O. mossambicus given by Jubb (1967) and by Bell-Cross and Minshull (1988). Specimens used for comparative purposes are: Tilapia arnoldi: holotype, South African Museum (SAM) 10862; now housed in the Albany Museum, Grahamstown, under that number. Tilapia rendalli: J. L. B. Smith Institute of Ichthyology, Grahamstown (RUSI) 26579; a single specimen from the Incomati river; RUSI 27988, 3 specimens from the Sabi River, Kruger National Park. Tilapia ruweti: RUSI 30126, a single specimen from the Thamalakane River, Okovango, Botswana; 10 specimens from lot RUSI 30343, Gomoti system, Okovango, Botswana. Tilapia sparrmanii: RUSI 22600, 3 specimens from the Palala River, Limpopo River system at Muisvogelkraal 20° 00’S, 24° 30’E, Transvaal. Oreochromis mossambicus: RUSI 26135, 7 specimens from weir R2M 10, above Laing Dam (Fort Murray), Buffalo River system, Eastern Cape Province. All specimens of the four latter species are of approximately the same size as the type of 7. arnoldi, and were chosen for that reason. REDESCRIPTION OF T. ARNOLDI HOLOTYPE The specimen is now in very poor condition (Fig. 1) having at some time been partially dried-out and suffered fairly extensive damage to the body and unpaired fins. It is also partly cleared since the alimentary tract is readily visible through the body wall. No trace of chromatophores remains on the body and fins, all of which are now a light caramel-brown in colour. Unfortunately the preserved coloration and colour- patterns as described by G & T are of no taxonomic value. Tilapia arnoldi holotype in left lateral view. Actual standard length: 46.0 mm. The head is extensively damaged and almost detached from the body; the right operculum and suboperculum are missing, as is the entire lower jaw. Some branchiostegal rays have been lost on both sides, and the branchiostegal membrane is badly torn; the complete hyoid arch, including the urohyal is, however, present. All four gill-arches are preserved on the left side, but only the third and fourth arches remain on the right; what seems to be the second arch of that side is detached and lies loose in the jar. The remnants of the pharyngobranchial skeleton have lost their attachment to the skull and pectoral girdle. The lower pharyngeal bone was still attached to the upper pharyngeal elements, but has now been dissected-out for detailed examination. Damage to most of the fins is also extensive. The upper half of the caudal fin is missing and the distal portions of all but three branched dorsal fin rays broken off, as are all the branched anal rays, although these are still attached to the fin proximally. The pelvic fins are virtually undamaged, as is the right pectoral fin; the left pectoral, however, is broken distally. With the aid of radiographs, and using the undamaged fins, the following ray counts and measurements can be made: Dorsal with 15 spinous and 11 branched rays; the penultimate and ultimate spines are the longest of the series, are of equal length, and are as long as the third anal spine. Pelvic fins 12.0 mm in length, their tips reaching the anus but not extending to the first anal spine insertion. Pectoral fin (right) ca 14 mm long, its tip reaching a vertical through the anus but not to that through the insertion of the first anal spine. On the left side most scales are missing from the anterior part of the body below the upper lateral-line, but the squamation is fairly complete over the entire right side and on the posterior part of the left side. All scales are cycloid, those on the flank below the lateral-line over approximately the anterior half of the body have moderately rugose exposed surfaces. There is a very gradual transition in size between the antero-ventral scales on the flanks and those situated laterally and ventrally on the chest. Consequently the chest scales, especially in the midline, are not markedly smaller than the scales lying above them. THE IDENTITY AND TAXONOMIC STATUS OF TILAPIA ARNOLDI The upper section of the lateral-line has 15 pored scales, the lower part 11 or possibly 12; the lateral scale count is 26 or 27. There are two rows of imbricating scales on the cheeks, and these completely cover the underlying muscles. Although it is difficult to make a precise count because of damage to the caudal peduncle, there are apparently not more than 14 circumpeduncular scales. No circumpeduncular count was given by Gilchrist and Thompson, but our other counts are in close agreement except for a difference of 3 in the pored scales of the upper lateral-line. Body measurements which could be made with any degree of accuracy are: Standard length (S. L.) 46.0 mm; head length (H. L.) ca 15 mm (i.e. about 33.3% of S. L.); body depth ca - 23 mm (i.e. about 50% of S. L.). Preorbital depth 3.0 mm (i.e. about 20% of H. L.), least interorbital width 6.0 mm (i.e. about 40% of H. L.), eye, measured as the horizontal _ diameter of the bony orbital margin, 6.0 mm (i.e. about 40% of H. L.), depth of cheek 3.0 mm (i.e. about 20% of H. L.). Depth of caudal peduncle 1.3 times its length. These proportional measurements, most of which were taken between bony fixed points, agree with those given by G & T, as do those for the pelvic and pectoral fins. The length (46.0 mm), however, is two millimetres less than recorded by G & T, but that difference could well be attributed to shrinkage with time, and the fact that the head is now almost free from the body. Shrinkage and damage would not account for the dif- ference in our counts of branched anal fin rays (11 cf 9 according to G & T: see p. 72 above). This discrepancy probably is due to the difficulty often encountered in deciding whether the last two or three rays of this fin are separate entities or branches of a single ray. The radiograph now available clearly shows 11 separate rays. The gill-rakers were described by G & T as short and thick, with 15 on the lower part of the first arch. The holotype now has only the first gill-arch of the left-side remaining. The upper eight outer-row rakers on the lower (ceratobranchial) part of this arch are now fine and slender structures, the three lower rakers being short and flaccid. Desiccation and poor preservation could well account for this alteration in gill-raker shape, but not for there being 11 instead of 15 rakers on the lower part of the arch. Since the spacing of the eleven rakers is regular and without gaps, and there is no damage to the surface of the arch, it seems very unlikely that four rakers have been lost by shrinkage or any other cause. Thus one must assume that G & T’s count of 15 is an error, perhaps resulting from the inadvertent inclusion of upper (i.e. epibranchial) rakers in their count. Unfortun- ately that supposition cannot be checked because much of the skin covering the epibranchial has been lost and only the lowermost epibranchial raker now remains. There also remains the problem of why Barnard (1948 a,b) too gave a count of 15 rakers since he had the holotype available for study. I can only proffer the suggestion that he accepted Regan’s (1922) opinion, based solely on G & T’s paper, and did not check the actual number of rakers. The lower jaw of T. arnoldi holotype is now missing, so dental characteristics can only be determined from the premaxillary teeth. Outer row teeth on this bone are un- equally bicuspid, with the very small minor cusp rather bluntly conical, and the crown of the major cusp obliquely truncate and not drawn-out. In other words, these teeth have a form common amongst many Tilapia and other tilapiine species (see figures in Thys, 1964, and Trewavas, 1983). 73 Apparently no teeth are missing from this row, and their total number is 36, a figure much lower than the ‘about 50 in the upper jaw’ givenbyG & T. Inner row premaxillary teeth are arranged irregularly in 2 or 3 series anteriorly and antero-laterally, and in a single row posteriorly. All are small and equally tricuspid. Since the lower pharyngeal bone was still in situ until I removed it, one can assume that previous workers had not studied the bone in any detail, especially as none has described it or the pharyngeal dentition The bone’s dentigerous surface is broadly and almost equilaterally triangular in outline (Fig. 2). Its posterior margin is gently biconvex, with the convexities joined me- dially by a short and shallow concavity. The overall width of the bone is slightly greater than its length and the anterior blade slightly shorter than the median tooth row. Fig 2. Tilapia arnoldi holotype. Lower pharyngeal bone in occlusal view. Drawn by Elaine Grant. Except for the large, distinctly more robust and clearly bicuspid teeth of the two posterior transverse tooth-rows, all other lower pharyngeal teeth are slender and moderately spaced, are ‘kukri’-shaped (see fig. 30 in Greenwood, 1987) and very weakly bicuspid. Only one osteological character of note is revealed by the radiographs, namely that there are 27 vertebrae, comprising 14 abdominal and 13 caudal elements (including the urostylar centrum). THE TAXONOMIC STATUS OF TILAPIA ARNOLDI Because neither Regan (1922) nor Barnard (1948 a,b) gave reasons for, respectively, their suggested or actual synonymis- ing of 7. arnoldi with Oreochromis mossambicus it 1s impossible to tell what particular character or character combination shared by the two taxa led to those decisions. With hindsight it seems likely that the reasons lay mainly in the supposed number of gill-rakers in T. arnoldi (15 according to G & T), and possibly in its low number of cheek scale rows. However, despite the present poor condition of T. arnoldi holotype, I am certain that there were only 11 rakers on the lower limb of the first gill-arch, a number below that of the 74 fewest (14) recorded in O. mossambicus, and then very rarely. Furthermore, two rows of cheek scales are of rare occurrence in O. mossambicus, in which species three rows are modal and two rows are rarely encountered (Trewavas, 1983: 295; personal observations). Thus neither of these features can be used to identify the specimen as O. mossambicus, and there are no other meristic or morphometric features of T. arnoldi which are diagnostic for O. mossambicus. The proportions and shape of the lower pharyngeal bone in T. arnoldi can also be used to argue against the holotype being identified as a member of the species O. mossambicus. Indeed, these features indicate that the specimen is not even a member of the genus Oreochromis as currently defined, and that it should be referred to the genus Tilapia (sensu Trewavas 1983). In Oreochromis species the blade of the bone is as long as, and generally longer than, the median tooth row, and the bone’s overall length is noticeably greater than its width (see Thys, 1964, 1968; Trewavas, 1983). In T. arnoldi, as in other members of the genus Tilapia, these ratios are reversed and the length of blade visibly contributes less to the bone’s overall length than it does in species of Oreochromis (see Thys, 1964: fig. 5). When the lower pharyngeal dentition of Tilapia arnoldi holotype is compared with that of comparable-sized O. mossambicus specimens, the teeth in T. arnoldi are seen to be more widely spaced and relatively coarser. In these respects the teeth closely resemble those of Tilapia sparrmanii and T. rendalli (whose pharyngeal bone’s proportions are also like those in T. arnoldi). Tilapia arnoldi also differs from Oreochromis mossambicus in two other features, namely its vertebral count and the size of its chest scales relative to those on the antero-ventral aspects of the flanks and anterior belly region. In O. mossam- bicus the range of vertebral numbers is from 28 to 31, with only three of the 23 specimens examined having 28 vertebrae (Trewavas, 1983). The count in T. arnoldi holotype is 27. The chest scales in O. mossambicus are noticeably smaller than those on the antero-ventral flanks and belly, whereas in 7. arnoldi the chest scales are but slightly smaller. Finally, mention should be made of three further features which distinguish the two species, namely the shorter pectoral fin (reaching the anus in T. arnoldi but to the first or second anal spine in comparable-sized O. mossambicus), the lower lateral scale count in T. arnoldi (26 or 27 cf 30-32), and the larger eye and wider interorbital in that species (data from personal observations and Trewavas, 1983). Taken in concert, the characteristics discussed above strongly indicate that the type specimen of T. arnoldi is not conspecific with Oreochromis mossambicus, while, as noted earlier, the nature of its pharyngeal bone and dentition show that it is a member of the genus Tilapia. If those conclusions are accepted, there remains the question of its specific identity within the genus Tilapia. On zoogeographical grounds and on the overall levels of morpho- logical similarity in preserved specimens, the resolution of that problem involves comparisons with Tilapia rendalli, T. sparrmanii and T. ruweti, the two former species being widely distributed in the Zambezi system, the latter, in Zimbabwe, restricted to the Upper Zambezi river (Bell-Cross and Minshull, 1988). Tilapia ruweti is readily distinguishable from T. arnoldi by its shallower body and rounded dorsal head profile, shorter and distinctly rounded pectoral fin (not reaching the level of PETER HUMPHRY GREENWOOD the anus), 3 or 4 (cf 2) rows of cheek scales, fewer (6-8 cf 11) gill-rakers which also appear to be shorter and stouter in T. ruweti (although this difference could be a consequence of the poor conditon of 7. arnoldi holotype) and, at least in specimens of a similar size, by its having the eye diameter about three-quarters that of the interorbital width and not equal to it as in T. arnoldi. In many respects T. arnoldi closely resembles T. rendalli, but it differs in its larger scales as evidenced by the lateral scale count of 26 or 27 cf 29-32 in T. rendalli, the circum- peduncular count of 13 or 14 cf 16 and by the larger size of the scales on the ventral aspect of the chest. The two taxa also differ in the posterior extent of the pectoral fin which, in T. rendalli, reaches to the level of the first, or even the second anal fin spine, but only to the level of the anus in T. arnoldi. At least in specimens of a comparable size, the teeth situated anteriorly and antero-laterally in the outer premaxillary row are larger (i.e. have wider tips) than those in T. arnoldi, with the result that 7. rendalli has 26-30 teeth in the outer row, whereas there are 36 in T. arnoldi. None of the features distinguishing 7. arnoldi from T. ruweti and T. rendalli serves to distinguish that species from T. sparrmanii, and I can find no others which do so. Further- more the vertebral count in 7. arnoldi (27) is that modal for T. sparrmanii and not for T. rendalli (28), although the ranges in both these species do overlap (26—28 and 27-29 for the taxa respectively). On the basis of the similarity between T. arnoldi and T. sparrmanii, and in the absence of any detectable contra- dictory evidence (here the poor state of 7. arnoldi holotype must be taken into account, as must the absence of informa- tion about its live coloration) I would conclude that Tilapia arnoldi Gilchrist and Thompson, 1917, should be treated as a junior subjective synonym of Tilapia sparrmanii A. Smith, 1840, and not, as Regan (1922) first suggested and Barnard (1948a,b) subsequently formalized, a synonym of Oreochromis mossambicus (Peters), 1852. ACKNOWLEDGEMENTS. It is with pleasure that I thank Jim Cambray of the Albany Museum, Grahamstown for allowing me to borrow the type specimen of Tilapia arnoldi and other specimens described by Gilchrist and Thompson. I am especially grateful to Paul Skelton of the JLB Smith Institute of Ichthyology for the many discussions we have had about 7. arnoldi, and for his critical reading of the manuscript. From the same Institute it is a great pleasure to thank Elaine Grant who provided the artwork, Robin Stobbs who produced the photograph and radiograph of the type specimens, and Huibre Tomlinson for her typing skills and patience when employing them. I have benefited greatly from the generosity and hospitality of the JLB Smith Institute of Ichthyology and the Department of Ichthyology and Fisheries Science of Rhodes University, to whom I offer my warmest thanks and gratitude. REFERENCES Barnard, K. H. 1948a. Revision of South African Cichlidae. Rep. inl. Fish. Dept. Admin. C. Good Hope, No. 5: 48-61. 19486. Report on a collection of fishes from the Okovango river, with notes on Zambesi fishes. Ann. S. Afr. Mus., 36(5): 407-458. Bell-Cross, G. & Minshull, J. L. 1988. The fishes of Zimbabwe. National Museums and Monuments of Zimbabwe, Harare. Boulenger, G. A. 1896. Descriptions of new fishes from the Upper Shire THE IDENTITY AND TAXONOMIC STATUS OF TILAPIA ARNOLDI River, British Central Africa, collected by Dr Percy Rendall. Proc. Zool. Soc. Lond., 1896, 915-920. Dumeril, A. H. A. 1859. Reptiles et poissons de l’Afrique Occidentale. Archs, Mus. Hist. nat. Paris, 10: 137-268. Greenwood, P. H. 1987. The genera of pelmatochromine fishes (Teleostei, Cichlidae). A phylogenetic review. Bull. Br. Mus. nat. Hist. (Zool.), 53(3): 139-203. Gilchrist, J. D. F. & Thompson, W. W. 1917. The freshwater fishes of South Africa. Ann. S. Afr. Mus., 11: 465-575. Jackson, P. B. N. 1961. The fishes of Northern Rhodesia. The Government Printer, Lusaka. Jubb, R. A. 1961. An illustrated guide to the freshwater fishes of the Zambezi River, Lake Kariba, Pungwe, Sabi, Lundi and Limpopo Rivers. Stuart Manning, Bulawayo. 1967. The freshwater fishes of southern Africa. A. A. Balkema, Capetown and Amsterdam. 75 Peters, W. C. H. 1852. Diagnosen von neuen Flussfischen aus Mossambique. Ber. Akad. Wiss. Berlin, 1852: 681-685. Poll, M. & Thys van den Audenaerde, D. F. E. 1965. Deux Cichlidae noveaux du sud du bassin du Congo. Revue Zool. Bot. afr., 72: 322-333. Regan, C. T. 1922. The classification of the fishes of the family Cichlidae. II. On African and Syrian genera not restricted to the Great Lakes. Ann. Mag. nat. Hist., ser. 9, 10: 249-264. Smith, A. 1840. /llustrations of the Zoology of South Africa, vol. 4, Pisces. London. Thys van den Audenaerde, D. F. E. 1964. Revision systématique des espéces Congolaise du genre Tilapia (Pisces, Cichlidae). Annls Mus. r. Afr. cent. Ser. &°. Sci. Zool. , 124: 1-155. 1968. An annotated bibliography of Tilapia (Pisces, Cichlidae). Documn. zool. Mus. r. Afr. cent. Ser. 8°. Sci. Zool., 14: x1 + 1—406. Weber, M. 1877. Zur Kenntniss der Susserwasser-Fauna von Siid-Afrika. Zool. Jb. Syst. 10: 135-200. Manuscript accepted for publication 30 August 1990 — ree } : ~ ¢ ’ ty rile Caz . Pe Boe Le Pei on tg " at oh maak are {he at > ee ber Stade af the _ Ab heals sperma of aang Mtge f taper a ow eng a eo ¢Gre che feu éiae”* gute) wdbantors¢ateraty Gy mode oi f nat eee ep oe ware 4 (qari epg = nate erates}. gre dares (1.» “hase waller pe? fe ti ane iy sited (sar Type, WR Tere, PED) bo) Yer Pal eee 7 reer) ¥ i= ih yh Mae i’: Goakt 2) harthas wee fa pees gia Oe ee | wheres ie o8 ar Ont paar vesdgel eaat tw Magi ot beds veri pesvieni Wale + Mowe of Gap. Sakasi raat ge ters ai. we i hee he weite oF iulee atl a. ‘pedal mes on - 4. TS & : “Fe T. ai Leremearn, : not Ora p< Vows — % ; ; > 2 lilopra arhceiii . nore, the wv pret ‘cowie pany (heme a sf fk yeabhes "8 be ere tt SO mwa gua, ifn. te ) sued) ave eet toe... 1H Oe @ weet és 2k a —s é Atm Tet te Gow. oc 0 frre ; het 34, s ‘gk opi a? . iy dhs Oy im a? tts pamdlaniy tte aad Pe 7S » > atve’s cntipeletines jay erecta; ae in ee wturnce Al: uti ON S44 iN ‘ . ‘ + ‘Cie =, dehonice fers tie peur ere They j ‘ 5 —— ora a muit bo Aken o (Tat, 4 : a Seis agl! _ vere i. tre > entae GPr- oe eteast ic ler codavemuurey 4 ae - — —_ Te quran Cath yi aivl be ; a vis nl te pes 7 rth " ‘yetiecs «= qurnywltise ore eyes I > : 1a t : al : 5; t 2 { = —— seal 7" ; : rath), 1240. aad got 2 Mngan 19 i aw, Wieemed (240g, 4) Geant ee ae o~ Tears i , j Tx6-¢4 3 dew! 7 | opeotines oe negra Ns bitin 4 = ; is ve - ~~ ha ; — | , a : P ; vil 1G : aie ‘ —T - . ; “ i s n — Sow a8 7 ue eee 8 «nani AO & =_ et ‘ ao aaris Bal mig sy se ; : 1 Ge Afters Uheown, 17 — ss a « - { . 4 x ; s wee? 6h ) ee = wo.’ Ske | , ¥ aint el ‘ah tao ee Vey ae - ah , 2 = . . y= Te ue? cael are eat } Pe : ; —_ ~~ ite Reh 2 —e. i Pla wt’ ¥ Cus X pes jm oe 4 ees fn" aga: ‘ —_ ; : Bayes Se? be age ft or ye bein 3 / ad? | =r Tuent*cues Ge ve '¢ 9 Aan | eatiuame . Ge. 1 Fe - tes _ te 4. +8 ihe 4 Sas mies a MPs Togs and a 8 : ‘ a) fi Lares ecu quit Fie a a fihwese t i } Tug sey Weg Pret Wace eve mites, & i z i ’ ‘ = Cha’ » j Ae J 4 : fe > i eae = 2 — ae RENCES ‘ : i . * a Si 4 : ; a et 5 sets hee ger en ee hath Aho Cage . "4 pet 7] what hee & at Oni "he Oesbwed! a oad : ~ _o4 a0 aoe .oe 6 ete aes, Go Cee ! : ; 4 i“ r hd 5 e=s an) oe Oo @s. ee 4‘ Ay ig % . etre’ oe Atv ¢. » eee tk a - ~ oe — <; od ostenale = A ‘Wai fsb aye ee eo. * Bull. Br. Mus. nat. Hist (Zool.) 51(1): 77-110 Anatomy, phylogeny and taxonomy of the gadoid fish genus Macruronus Gunther, 1873, with a revised hypothesis of gadoid phylogeny GORDON J. HOWES Department of Zoology, British Museum (Natural History), Cromwell Rd, London, SW7 5BD CONTENTS MCT RIGL IEC BLOM BA cad Nena ive rced dora ied iinet s Ss ayqlaoeline oblonga ttercs > + gery) + < ouehdaye sn! aepayt > aaa. Wi ALE ASEAN HNOGS) cots. choo, Ses ptecw Mp pe teats. Ws Sti a's «Fagg tcl ape WAS Beer ote cage oc BA ppusncliap Dagehe seunee tat 78 {5 SIGNET ECC TH CUS Se RT nN cha 20 LS Oe CE PO SP SUN” AERO ene Ow SUPE 78 mamatomy Gf Macruronus and comparison/with Merluccius 2 eee nye ee ee ee nin ed nf Oe dle we ey ne 80 ZT SATE NOTE DOC) oP ROI ae en Ly ae ee Pe, Ne Eg PRAY RR aan ee meme peer ee 80 “TEV IIT SS od of ee ees ie ER OR Sete aR ae eee TEN ent a ge AE NE Ee 80 PO SHETINS DOT ce Rene peo OT CRETE RSE, Amn AIP ai SR Soa. cul gel ane nee .. 84 2 SURVTS OLDE ETAS gk sons Pace Aa Oe S ago ie ae Acne mcm DN ot et a a 84 MDL ABIES 20 ae Meiaia.c Fun snr a erg! akems ani! in Hiei E seajiemisl aa see eagle alte oe Ree Mme ates «er ees ae ea ae we oME ae Snes 84 IM peiea MOG VCLUAW Sr: Met ces or ak ae ate Ne are O Ne Cem w es OpWinlsc amo Em Gand cNule anise w oleae 84 LE DETECT on 6 cpg beatmaminestetne-setloeenly veer ta ame ai ot tae nee oa ne et ee ee en nea fe ae eo 85 “CNCISIE(DLIG 1 | DSERVE a tr od are a ee I Sr ea ea rar 88 EMVOMRARCNCS ate heey eta: Mee arene, 2 SIMA Scobie pewala es ds gate aa ee ee Cone EA. A. 88 Eecameitaltanen)< ee sae, Seles, Sle SER. |. Sh. 2, meena re rere erat et reece emcee agen ae inns 89 | PEC) COI TSG |i OOO ch nc cc ol Ae 5 ARMOR 90 PSN O Ce ese accaiilers 9 Ue NS. 5 Sa S| es ee a ae Reena ee Ae a 90 Mette Ond Ceol eae nn mene ACETONE... RENE oc ca he PEN eRe aa be tar Rr REPRE. nae oi UGE AIMEE NG COM. 215. se2.5 5 255s, = ME eee as: Sh; ROMS RAUITTA SPE Per hey EE Ds od Se 94 RACES CIEINSCHMLON res att. a JeaEIO! . . Sek nd Se ea ae . Oca. Ieee hae eee 96 ICEIES/GOMMIVUIES TINUSCIE rte: wabereeth syne chs EE cs. | DAR ARs elas Le Oe he eee aed ae 96 PETITES GLCT ALES ACCCSS OLIN MCIVChy ncmos unk. . >. eee | > s Denes fomtheniwonwvews. 24. i areesmeee Lat 96 Pr SHOCONG Chat Ate hin ptensand Pep etie wuss... ot. oo < Geabaeearanrdes : bh deo. Amgen y ones. PT etl oF; LDL ECUSSIC TG Sante ene oo MOOR Fa eM OT one eee ey ee ee ee ae ee ee we 101 Relationships of Macruronus with Lyconus and Ly GOnOdEeS) 2:5 «i, ccvec suspen» mp eo4y orn ene sens eae «ops » yp era eet 101 itelationships ef the Macruronidae with other gadoidfamillies |... . =... 2c je nee ee ete eee ee oie 103 Bee TE CDN ATs SE RA se PRR Sa Moan Ma aac A ceili A non ak dl atts Cie eR AY 2 niente hepa Seger 106 SUPT se RGN RE Fae Le ee ea lense ec I eR cs Ra Ge a a BA 108 Sh SMOMUGNIINYA Ok CHE NIACTIROUIG AC 2g sere an a 45 OR eas Oy ee eth dak ic SSN syste a soa cm ae asa 108 RRO OC TCIM Gian nee ee Wee ese yearn ein aa eee ar os hata Sl oi: Bese Re Sas ue aud sao 2, etree 109 | EL EIEIO S Sd AUS Raed ab ane eo Bin apes A ane eer 5 4 | ? 4 | | ? | | 3 e dae 5 melano™ Fig. 35 Cladogram of gadoid relationships. Synapomorphies relating families: 1, Levator arcus palatini muscle lies lateral to adductor mandibulae; 2, Absence of pars jugularis (a single opening serves for the transmission of all cranial nerves and vessels. 3, Rectus communis muscles attached to sternohyoideus; 4. X and Y bones in caudal skeleton (lost in Melanonidae, Macruronidae, Lotidae and Gadidae; presumably also in Trachyrincidae, Bathygadidae and Steindachneriidae; 5, Caudal fin muscles modified, interradiales connect caudal fin rays with dorsal and anal rays; 6, Palatine contacts the mesethmoidal part of lateral ethmoid; 7, Medial shift of muscle Alb; 8, Complete fusion of upper hypurals into single plate. (According to Markle (in litt.) upper hypural plate of Raniceps is bifurcate and ontogenetically displays more than two upper hypurals). This demonstration of ontogenetic fusion could thus be considered a character reversal; distribution of other synapomorphies do not support the view that Raniceps occurs at a lower phylogenetic level; 9, Loss of lateral ethmoid-palatine ligaments; 10, 1st infraorbital contacts posterior face of lateral ethmoid wing; 11, Posterohyal articulates with an interopercular fossa whose rim forms a ‘stop’ to its posterior movement; 12, Rectus dorsalis muscle inserts on pharyngobranchial 4 in addition to 3 (exceptions: Merluccius, inserts only on 3; Muraenolepis, inserts only on 4); 13, Hyomandibular with lateral shelf; 14, Cone-shaped lateral ethmoid wing; 15, Attrition of the prootic anterior border with often, in conjunction with pterosphenoid; formation of separate foramen for trigeminal and hyomandibularis nerves; 16, Medial prootic shelf. | Synapomorphies for lineages: a, Macrouroidei; loss of interopercular- subopercular ligament; many other synapomorphies, discussed in Howes & Crimmen, 1990; b, Trachyrincidae; Adductor mandibulae A1 muscle a single element, nasal bones extended, palatine immobile (Howes, 1987, 1988; 1989), caudal fin skeleton ‘rudimentary’ (Howes, 1988b and this text); c, Bathygadidae; pectoral branch of RLA nerve hypertrophied (Howes & Crimmen, 1990); d, Steindachneriidae; luminescent organs, wide separation between anus and urogenital opening, enlarged anal radial and compound first anal fin ray (Fahay, 1989); e, Melanonidae; specialized neuromast pattern, ectopterygoid teeth, loss of intermandibularis muscle; f, Moridae; primitive otophysic connection (Paulin, 1983), swimbladder with horizontal septum (Paulin, 1988), parasphenoid with transversely aligned ascending process (this text); g, Euclichthyidae; jugular placed and reduced pelvic girdle, pelvic rays and anterior rays of anal fin extended, lateral ethmoid wing transversely convex; h, Macruronidae; retroarticular with anteroventral prolongation, prootics form a wall across the midline, modified caudal fin skeleton, modified infraorbital bones, adductor arcus palatini muscles originates partly from a ventral palatine fossa (this text); i, Bregmacerotidae; single first dorsal fin ray articulating with an angled radial lying along the supraoccipital crest, advanced symmetry of the caudal fin (Markle, 1982) reduced metapterygoid and operculum (pers. obs adductor mandibulae muscle reduced to a single element (Howes, 1988); j, Muraenolepididae; derived palatine morphology (Howes, 1987), rectus dorsalis muscle inserts only on pharyngobranchial 4 (Howes, 1988); k, Ranicipitidae; first dorsal fin with 3 rays, lateral line reduced; 1, Phycidae (Phycis, Urophycis) epaxialis muscle inserts on operculum (also shared with the Muraenolepis and Lota), frontals fused (shared with Gadidae), anterior diverticulum of swimbladder adnate to exoccipital (pers. obs.); m, Lotidae, no well corroborated synapomorphies for Molva, Lota and Brosme, those given by Markle, 1982 are, absence of pterotic spine, initial pelvic fin ray formation prior to flexion, and delayed acquisition of adult complement; n, Phycidae (Gaidropsarus, Motella, Ciliata, Rhinonemus), first dorsal fin comprised of several filamentous rays contained in a dorsal groove, and supported by modified radials (Markle, 1982); 0, Gadidae; three dorsal fins, reduced and enclosed ‘mucous’ cavity of the frontals, frontals fused in most taxa, swimbladder with elaborate anterior diverticulae; p, Merlucciidae; enlarged vertebral parapophyses, hyomandibular with two long, lateral ventrally directed processes, levator arcus palatini muscle originates from a lateral palatine fossa, medial prootic shelves form pseudo-posterior myodome (this text) From this cladogram, it follows that the Gadoidei are embraced by character 3. Since, however, this feature is also encountered in some macrouroids (Howes, 1988a) its status as a synapomorphy is weakened. Alternatively, recognition of the Gadoidei by characters 5 and 6 places the Trachyrincidae, Bathygadidae, Steindachneriidae and Melanonidae as incertae sedis. This cladogram has since been superseded by another (Howes, 1990) in which the Muraenolepididae is aligned with the Lofidae and the second part of the Phycidae. 108 Merluccius. In other words, they accept the near-traditional view of the Merlucciidae (recognised as a family by Okamura and a subfamily by Inada. Both authors use characters which are regarded therein either as plesiomorphic or homoplastic. Inada’s sole ‘synapomorphy’ for allying the three genera is the V-shaped frontal ridge, a feature discussed here and deemed plesiomorphic for gadoids. Okamura presents no evidence for grouping the three genera other than ‘All characters are opposite, that is, primitive states of those of Steindachneria’. I do not believe that either author has addressed the problem of merlucciid monophyly and none of the characters they present are new or are polarized through adequate out- group analysis. APPENDIX Taxonomy of the Macruronidae Family MACRURONIDAE Macruroninae Norman 1966:196 Distinguished from other gadoid families in arrangement of dentition, having large, compressed caniniform teeth in outer row of both jaws and an inner row of small, horizontally aligned and medially directed teeth; anteriorly extended anteroventral margin of retroarticular; densely ossified Sth infraorbital and reduced 6th contained in a sphenotic-frontal notch. TWO GENERA: Macruronus Ginther, 1873; Lyconodes Gilchrist, 1922. Macruronus Gunther, 1873: 103; type species Coryphaenoides novaezelandiae Hector 1889: 196 Lyconus Ginther, 1887: 158; type species Lyconus pinnatus Gunther, 1887: 158. Subgenera: Macruronus; Cynogadus (p. 103). Species: Macruronus (Macruronus) novaezelandiae (Hector, 1889) Coryphaenoides novae-zelandiae Hector, 1889: 157 Macruronus novae-zelandiae Ginther, 1873: 103; 1887: 157 For description, see Waite, 1911: 180, and for bioeconomic data, Kuo & Tanaka, 1984a,b,c. and Patchell, 1982; for larval development, Patchell et al., 1987; Bruce, 1988. DISTRIBUTION: New Zealand, Tasmania. Macruronous (Macruronus) magellanicus Lonnberg, 1907 Macruronus magellanicus Lonnberg, 1907: 15 Macruronus novae-zelandiae (non Hector) Giinther, 1880: 22 Macruronus argentinae Lahille, 1915: 22 For description, see Norman (1937:49) and Inada (1986), and GORDON J. HOWES for bioeconomic data, Hart (1946) and Torno & Tomo (1980). DISTRIBUTION: coastal areas and banks off southern South America. Macruronus (Macruronus) capensis Davies, 1950 Macruronus capensis Davies, 1950: 512 Described from a single specimen, Davies distinguished this from other species on the presence of a small ‘spine’ and 13 soft rays in the first dorsal fin. However, there is a reduced first ray present in all Macruronus species and there are 13 longer rays in M.magellanicus. Davies also noted a supposed difference in dentition, mistakenly believing that teeth were absent in the upper jaw of M.magellanicus and that two rows of upper jaw teeth were a feature unique to M.novaezelandiae. The inner row is, however, present in M.magellanicus and is visible only when tissue covering the inner surface of the jaw is removed (see Fig. 11C), and so it seems likely that Davies overlooked these teeth. Davies also maintained that the locality (off the Cape) and depth of capture (280 fms) further signified ‘specific distinc- tiveness’. Since Davies’s description further specimens have been reported (Cohen, 1986: 325). DISTRIBUTION: Off South African Cape. Macruronus (Macruronus) caninus Maul, 1951 Macruronus caninus Maul, 1951: 45 Known from three specimens; provenance uncertain, but most likely the southern coast of Madeira. Macruronus (Macruronus) maderensis Maul, 1951 Macruronus maderensis Maul, 1951: 49 Known from eight specimens, all juveniles (the largest 85mm TL), taken from the stomach of an Alepisaurus ferox; off Madeira. Macruronus (Macruronus) pinnatus Lyconus pinnatus Gunther, 1887: 158. The type (Fig. 33) and other known specimens are all juveniles. Recorded from the south Atlantic and Indian Oceans. Macruronus (Cynogadus) brachycolus Lyconus brachycolus Holt & Byrne, 1906: 424 Known only from a single specimen from the Irish continental shelf (Fig. 34). The species of Macruronus are poorly known and the distinc- tions between them are slight. Macruronus (M.) magellanicus is distinguished from M.novaezelandiae on the basis of its smaller eye and longer premaxillary; the two species from Madeira supposedly have fewer rays in the first dorsal fin, both are described from juvenile specimens which may be conspecific with the southern Atlantic species. Only the collection and study of more material will determine whether the described taxa represent distinct species or populational variants of a single cosmopolitan one. ANATOMY, PHYLOGENY AND TAXONOMY OF THE GADOID FISH GENUS MACRURONUS 109 ACKNOWLEDGEMENTS. Iam most grateful to Humphry Greenwood, Douglas Markle, Colin Patterson and Alwyne Wheeler for their critical readings of the manuscript and their many helpful comments. My thanks are also due to Patrick Campbell, Oliver Crimmen and Mandy Holloway for their many and varied technical services. REFERENCES Boulenger, G. A. 1902. On the classification of teleostean fishes IV. On the systematic position of the Pleuronectidae. Annals and Magazine of Natural History (7) 10: 295-304. Brauer, A. 1912. Die Tiefsee-Fische. 1, Systematischer Teil. Wissenschaftliche Ergebnisse der deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’, 1898-1899, 15: 1-432. Bruce, B. D. 1988. Larval development of blue grenadier Macruronus novaezelandiae (Hector), in Tasmanian waters. Fishery Bulletin, Seattle 86 (1): 119-128. Cohen, D. M. 1984. Gadiformes: overview. In H. G. Moser (Ed.-in-chief). Ontogeny and systematics of fishes. American Society of Ichthyology and Herpetology Special Publication 1: 259-265. — 1986. Merlucctidae. In M. M. Smith & P. C. Heemstra (Eds) Smith’s Sea Fishes: 324-326. Davies, H. 1950. A new species of Macruronus from South Africa. Annals and Magazine of Natural History (12) 3: 512-515. Fahay, M. P. 1989. The ontogeny of Steindachneria argentea Goode & Bean, with comments on its relationships. Papers on the systematics of gadiform fishes. Science series No. 32. Natural History Museum of Los Angeles County: 143-158. Fahay, M. P. & Markle, D. F. 1984. Gadiformes: development and relationships. In H. G. Moser (Ed.-in-chief). Ontogeny and systematics of fishes. American Society of Ichthyology and Herpetology Special Publication 1: 265-283. Farugi, A. J. 1935. The development of the vertebral column in the haddock (Gadus aeglifinus). Proceedings of the Zoological Society of London 1935 (2): 313-332. Freihofer, W. C. 1970. Some nerve patterns and their systematic significance in paracanthopterygians, salmoniform, gobioid and apogonid fishes. Proceed- ings of the California Academy of Sciences (4th ser.) 38 (12): 215-264. Gilchrist, J. D. F. 1922. Deep-sea fishes procured by the S. S. ‘Pickle’ (part 1). Union of South Africa Fisheries and Marine Biological Survey, Report No. 2 for 1921: 41-79. Gosline, W. A. 1963. Considerations regarding the relationships of the percop- sifom, cyprinodontiform and gadiform fishes. Occasional Papers of the Museum of Zoology, University of Michigan No. 629: 1-38. Gregory, W. K. 1933. Fish skulls. A study of the evolution of natural mechanisms. Transactions of the American Philosophical Society 23: 75-481. Ginther, A. 1873. Pisces. Zoological Record for 1871: 103. —— 1887. Report on the deep-sea fishes collected by H.M.S. ‘Challenger’ during the years 1873-76. Report of the scientific results of the voyage of H.M.S. Challenger 1873-1876 (Zoology) 22 (57): 1-335. —— 1880. Report on the shore fishes procured during the voyage of H.M.S. ‘Challenger’ in the years 1873-1876. Report of the scientific results of the voyage of H.M.S. ‘Challenger’ 1873-1876 (Zoology) 1 (6): 1-82. Hart, J. J. 1946. Report on trawling surveys on the Patagonian continental shelf. Discovery Reports23: 223-408. Hector, J. 1889. On a new species of fish, Coryphaenoides Novae Zelandiae; native name, Okarari. Transactions of the New Zealand Institute 3: 136. Holt, E. W. L. & Byrne, L. W. 1906. On a new species of Lyconus from the North-east Atlantic. Annals and Magazine of Natural History (7) 18: 432-426. Howes, G. J. 1987. The palatine bone and its associations in gadoid fishes. Journal of Fish Biology 31: 625-637. —— 1988. The cranial muscles and ligaments of macrouroid fishes (Teleostei: Gadiformes); functional, ecological and phylogenetic inferences. Bulletin of the British Museum (Natural History), Zoology series 54 (1): 1-62. —— 1989. Phylogenetic relationships of macrouroid and gadoid fishes based on | cranial myology and arthrology. In D. M. Cohen (Ed.) Papers on the | Systematics of gadiform fishes. Science series No. 32, Natural History Museum of Los Angeles County: 114-128. —— 1990. The syncranial osteology of the southern eel-cod family Muraen- depididae, with comments on its phylogenetic relationships and on the biogeography of sub-Antarctic gadoid fishes. Zoological Journal of the Linnean Society 100: 73-100. | —— & Crimmen, 1990. A review of the Bathygadidae (Teleostei:Gadiformes). ] | Bulletin of the British Museum (Natural History), Zoology series 56 (2): 155-203. Inada. T. 1981. Studies on the merlucciid fishes. Bulletin of the Far Seas Fisheries Research Laboratory no. 16: 1-172. 1986. Merlucciidae. In (I. Nakamura Ed.) Important fishes trawled off Patagonia Japan Marine Fishery Resource Research Center: 120-127. —— 1989. Current status of the systematics of Merlucciidae. In D. M. Cohen (Ed.). Papers on the systematics of gadiform fishes. Science series No. 32, Natural History Museum of Los Angeles County: 197-207. Kusaka, T. 1974. The urohyal of fishes University of Tokyo Press. 320pp. Kuo, C. -L. & Tanaka, S. 1984a. Feeding habits of the hoki Macruronus novaezelandidae (Hector) in waters around New Zealand. Bulletin of the Japanese Society of Scientific Fisheries 50 (5): 783-786. 1984b. Distribution and migration of hoki Macruronus novaezelandidae (Hector) in waters around New Zealand. Bulletin of the Japanese Society of Scientific Fisheries 50 (3): 391-396. 1984c. Maturation and spawning of hoki Macruronus novaezelandidae (Hector) in waters around New Zealand. Bulletin of the Japanese Society of Scientific Fisheries 50 (3): 397-402. 1984d. Otolith features and reliability for age determination of hoki Macruronus novaezelandidae (Hector) in waters around New Zealand. Bulletin of the Japanese Society of Scientific Fisheries 50 (8): 1349-1355. Lahille, F. 1915. Sobre dos Peces macrurides de las costas de la Provincia de Buenos Aires. Anales del Museo de Historia Natural de Buenos Aires 26: 21-29. Lauder, G. V. & Liem, K. F. 1983. The evolution and interrelationships of the Actinopterygian fishes. Bulletin of the Museum of Comparative Zoology, Harvard 150 (3): 95-197. Lonnberg, E. 1907. Fische. Sammelreise. 8 (6): 1-16. Markle, D. F. 1982. Identification of larval and juvenile Canadian Atlantic gadoids with comments on the systematics of gadid subfamilies. Canadian Journal of Zoology 60 (12): 3420-3438. —— 1989. Aspects of character homology and phylogeny of the gadiforms. In D. M. Cohen (Ed.) Papers on the systematics of gadiform fishes. Science series No. 32, Natural History Museum of Los Angeles County: 59-88. Marshall, N. B. 1966. The relationships of the anacanthine fishes Macruronus, Lyconus and Steindachneria. Copeia 1966: 275-280. 1973. Macrouridae. In. Check-list of the fishes of the north-eastern Atlantic and of the Mediterranean (Clofnam), UNESCO, Paris 1: 287-302. & Cohen, D. M. 1973. Order Anacanthini (Gadiformes). Characters and synopses of families. Memoir, Sears Foundation for Marine Research 1. Fishes of the western North Atlantic (6): 479-495. Maul, G. E. 1951. Monografia dos peixes do Museu Municipal do Funchal. Familia Macrouridae e Merlucciidae. Boletim do Museu Municipal do Funchal 5, art 12: 5—56. Norman, J. R. 1937. Coast fishes. Part II. The Patagonian region. Discovery Reports 16: 1-150. —— 1966. A draft synopsis of the Orders, Families and Genera of Recent fishes and fish-like vertebrates. British Museum (Natural History), 649pp. Okamura, O. 1970. Studies on the macrouroid fishes of Japan-morphology, ecology and phylogeny. Reports of the Usa Marine Biological Station, Kochi University 17 (1-2): 1-179. —— 1989. Relationships of the suborder Macruroidei and related groups, with comments on Merlucciidae and Steindachneria. In D. M. Cohen (Ed.) Papers on the systematics of gadiform fishes. Science series No. 32, Natural History Museum of Los Angeles County: 129-142. Patchell, G. J. 1982. The New Zealand hoki fisheries 1972-82. New Zealand Fisheries Research Division Occasional Publications 38: 23pp. Patchell, G. J., Allen, M. S. & Dreadon, D. J. 1987. Egg and larval development of the New Zealand hoki Macruronus novaezelandiae. New Zealand Journal of Marine and Freshwater Research 21: 301-313. Patterson, C. 1975. The braincase of pholidophorid and leptolepid fishes, with a review of the actinopterygian braincase. Philosophical Transactions of the Royal Society of London B269 (899): 279-579. —— & Rosen, D. E. 1989. The Paracanthopterygii revisited: order and disorder. In D. M. Cohen (Ed.) Papers on the systematics of gadiform fishes. Science series No. 32, Natural History Museum of Los Angeles County: 5— 36. Paulin, C. D. 1983. A revision of the family Moridae (Pisces: Anacanthini) within the New Zealand region. Records of the National Museum of New Zealand 2 (9): 81-126. — 1988. Swimbladder structure in morid cods (Pisces: Gadiformes). Copeia, 1988 (2): 450-454. Regan, C. T. 1903. On the systematic position and classification of the gadoid or anacanthine fishes. Annals and Magazine of Natural History (7) 11: 459- 466. Rosen, D. E. 1985. An essay on euteleostean classification. American Museum Novitates No. 2827: 1-57. Ergebnisse Hamburger Magalhaensische 110 & Patterson, C. 1969. The structure and relationships of the paracanthop- terygian fishes. Bulletin of the American Museum of Natural History 141 (3): 357474. Svetovidov, A. N. 1948. Gadiformes. In E. N. Pavlovskii & A. A. Shtakel’berg (Eds) Fauna of the U.S.S.R., Fishes 9 (4): 1-304. English Translation, 1962. Israel Program for Scientific Translations. Symmons, S. 1979. Notochordal and elastic components of the axial skeleton of fishes and their functions in locomotion. Journal of Zoology, London 189: 157-206. Torno, A. E. & Tomo, A. P. 1980. Nuevos aportes al conocimento de la merluza de cola (Macruronus magellanicus Lénnberg) del mar Argentino. Revista del Museo Argentino de Ciencias Naturales ‘Bernardino Rivadavia’ Zoologia 12 (14): 177-187. GORDON J. HOWES Videler, J. J. 1975. On the interrelationships between morphology and movement in the tail of the cichlid fish Tilapia nilotica (L.). Netherlands Journal of Zoology 25 (2): 143-194. Waite, E. R. 1911. Scientific results of the New Zealand Government Trawling Expedition, 1907. Pisces, Part II. Records of the Canterbury Museum 1: 157-258. Winterbottom, R. 1974. A descriptive synonymy of the striated muscles of the Teleostei. Proceedings of the Academy of Natural Sciences, Philadelphia 125: 225-317. Manuscript accepted for publication 8 March 1989 “4 Wenieal, &.% feet tetas. le? a doh arenes 9: (RS B, Pdate ot bole teal! Tom tatte awe be ree ee ne Wine, » ap po V/caeriah aki posiecsamenaine : re Leman th Soaregenn lems A ) is Se — 6a Oe. 2 Neem gpuines od Caer iH dae ids as haem tanner ag ha Nap Paes dx tae ee eT Savdieees, 79 (rap 7.40 = : = ei, - : ; ; — — : , . 5 i x 4 i ts “ > me i” ; = 7 7 < ; = 7 ~~ 9 . 2 § ~ i ‘Ss ‘i. r aie = j 2 di t > & 4 : ~ - ae 2 . r. ~ = > -- 1 a : 7 s 7% i - . se vs a) fo Ar \es ) he hm cy Pe Se BG, ake, he! ee E rsa)" * Pe a 5 hhe ‘a? Lago ‘ we es A 2 Cae vt oe : ~ : ¥ t ‘ % 7 J Aa) ps y t ne th 7 t ~ A r 7 ~ ae : a oe 1 ‘ 7 : 1 q Fy - ~ a : 8 ¥. ) - , i q 1 - : 4: on . . ‘ , 1 , ~. A= } f ‘ ad Py 7 —_ > Bas = a” t 4 . \ ~ = a “= < ind ¢ = “. i y > = 5 p> ate 7 . =P} . * ; ay ; . t a . - i 2 a t va = x Ne a i = \ a os n ; 2 ‘ So = . f / 4 rat a : y S i _— , . i - = boas J ‘ e —— y i te ‘ 3 ~ ; = 1 | e i , . x ts * 7 ? ® ra i J, 1 ' i 7 = = ae ‘ ise + q q ° 7 * 4 u = _ . i t . Ee : Me hi + : ; ‘i = i pe a : de 4) “d : i > 7 Mi 1 Se a re P , tae ‘ x a owt - 1 u 1 > ‘ Me 3 J &. 1 * % h) \ ‘ sic - ~~ 4 1 A é Me 7 4 4 n J i ~4 i] = i : f ’ i ri * NC Z i i f 7 x i ieee ~ » n . ~ vr i ' ned - : rf 7 r - i c \ NOTES FOR THE GUIDANCE OF AUTHORS Policy. The Bulletin of the British Museum (Natural History) Zoology, was established specifically to accommodate manu- scripts relevant to the Collections in the Department of Zoology. Preference is given to original contributions in English whose contents are based on the Collections, or the description of speci- mens which are being donated to enhance them. Acceptance of manuscripts is at the discretion of the Editor on the understand- ing that they have not been submitted or published elsewhere and become the copyright of the Trustees of the British Museum (Natural History). All submissions will be reviewed by at least two referees. Manuscripts. The Bulletin provides an outlet for the publi- cation of taxonomic papers which because of their length, prove difficult to publish elsewhere. However, prior agreement with the Editor is required before submission of these longer manu- scripts. Shorter manuscripts, which do not exceed 8000 words (about 8 printed pages), including references and text figures can be submitted at any time. Three complete copies should be submitted, only clean, clear photocopies are acceptable. They must be accurate and in their final format, because corrections at proof stage may be charged to the author. Additions at proof stage will not normally be allowed. Manuscripts must be submitted in the style and format used in the Bulletin. The text must be typed double spaced throughout, including references, tables and legends to figures, on one side of A4 paper with a 2.5cm margin at each side. All pages should be numbered consecutively in the top right hand corner, with the title page counting as p.1. Manuscripts on floppy disc are acceptable but must be ac- companied by three clean print outs. Full details of hard- and software used should be supplied, do not key in design instruc- tions, but caps, bold and italic should be used as appropriate. Layout. 1. Title page should be arranged with the full title of the article in capitals; first name(s) in full and surname(s) of authors without academic titles; full institutional addresses, in upper and lower case; running title; key words; and address for correspondence. 2. The synopsis begins p.2 and should not exceed 200 words. It should present the purpose of the investigation, important methods, the main findings and the principal conclusions. 3. A maximum of six key words referring to the subject of the publication should be supplied. The editors may change the key words. 4. The text should be divided into five main headings: Intro- duction, Materials and Methods, Results, Discussion, Ac- knowledgements and References. There are usually three levels of text headings and subheadings. The main text heading (A) should be typed ranged left, the (B) subheading should be ranged left and any further subheadings should also be ranged left, they should all be in upper and lower case with only generic and specific names, to be set in italic, underlined. Systematic head- ings denoting class, order, family etc. are termed DA headings and should be typed left with the class/order/family/generic name in capitals. The species heading (C) in upper and lower case. Footnotes are not acceptable. 5. References should be listed alphabetically, authorities for taxonomic names are not to be cited unless the appropriate pub- lication is referred to in the text. The date should immediately follow the authors’ names after a single space. Where an author is listed more than once, the second and subsequent entries should be denoted by a long dash, these entries should be in date order. Joint authorship papers follow the entries for the first author, an ‘&’ should be used instead of ‘and’ to connect joint authors. Journal titles should be entered in full. Examples: (i) Journals: England, K.W. 1987. Certain Actiniaria (Cnidaria, Anthozoa) from the Red Sea and tropical Indo- Pacific Ocean. Bulletin of the British Museum ( Natural History ), Zoology 53: 206-292. (ii) Books: Jeon, K.W. 1973. The Biology of Amoeba. 628p. Academic Press, New York & London. (iii) Articles from books: Hartman, W.D. 1981. Form and distri- bution of silica in sponges. pp.453-493. Jn: Simpson, T.L. & Volcani, B.E. (Eds) Silicon and Siliceous structures in Biological Systems. Springer-Verlag, New York. 6. Each table should be typed on a separate sheet designed to extend across single or double column width of a Journal page. They should have a brief specific title, be self-explanatory and supplementary to the text. Limited space in the Journal means that detailed results like locality lists, tables of measurements, details from mathematical derivations and computer print-outs will be stored in the Department of Zoology under the sectional records of the animal group concerned. These records will be considered as part of the Collections and reference should be made to them in the text. Illustrations. Figures should be designed to go across single (83 mm wide) or double (170mm wide) column width of the Journal page, type area 235 x 173 mm. Drawings should be in black on white, stiff, card with a line weight and lettering suitable for same reduction either 50%, 30% or 25%. After reduction the smallest lettering should be not less than 10 pt (3mm). All photographs should be prepared to the final size of repro- duction, mounted on stiff card and labelled with press-on letter- ing, components of figure-plates should be abutted. All figures should be numbered consecutively as a single series. Legends, brief and precise, must indicate magnifications and explain symbols and letters. Reprints. Twenty-five reprints of each paper will be furnished free of charge. Orders for additional reprints can be submitted to the publisher on a form provided with the proofs. Later orders cannot be accepted. CONTENTS 1 Morphology and biometry of twelve soil testate amoebae (Protozoa, Rhizopoda) from Australia, Africa and Austria. Gabriele Luftenegger & Wilhelm Foissner 17 Arevision of Cothurnia (Ciliophora: Peritrichida) and its morphological relatives. Alan Warren & Jan Paynter 61 Indian Ocean echinoderms collected during the Sindbad Voyage (1980-81): 2. Asteroidea. Loisette M. Marsh & Andrew R. G. Price 71 Theidentity and taxonomic status of 7i/apia arno/di Gilchrist and Thompson, 1917 (Teleostei, Cichlidae). Peter Humphry Greenwood 77 Anatomy, phylogeny and taxonomy of the gadoid fish genus Macruronus Ginther, 1873, with a revised hypothesis of gadoid phylogeny. Gordon J. Howes Bulletin British Museum (Natural History) ZOOLOGY SERIES Vol. 57, No. 1, May 1991 Bulletin © a: ~ British Museum (Natural History) Zoology Series The Bulletin of the British Museum of Natural History, instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an Historical series. The Zoology Series is edited in the Museum’s Department of Zoology Keeper of Zoology: Dr C. R. Curds Editor of Bulletin: Mr N. Merrett Assistant Editor: Dr B. T. Clarke Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff and by specialists from elsewhere who make use of the Museum’s resources. Many of the papers are works of reference that will remain indispensable for years to come. A volume contains about 256 pages, made up by two numbers: published Spring and Autumn. Subscriptions may be placed for one or more of the series on an Annual basis. Individual numbers and back numbers can be purchased and a Bulletin catalogue, by series, is available. Orders and enquiries should be sent to: Sales Department, Natural History Museum Publications, British Museum (Natural History), Cromwell Road, London SW7 5BD Telephone: 071-938-9386 Fax: 071-938-9212 World List abbreviation: Bull. Br. Mus. nat. Hist. (Zool.) © British Museum (Natural History), 1991 ISBN 0 565 05050 8 Zoology Series ISSN 0007 — 1498 Vol 57 No 2 pp 111 —219 British Museum (Natural History) Cromwell Road London SW7 5BD Issued 28 November 1991 Typeset by J & L Composition, Filey, North Yorkshire Printed in Great Britain by Henry Ling Ltd, at the Dorset Press, Dorchester, Dorset Bull. Br. Mus. nat. Hist. (Zool.) 57(2) 111-132 Issued 28 November 1991 r me 7 f= 8 5 yy mo The pharyngobranchial organ of mugilid (NAT biter iol UT 13 DEC 199) PD P2 %, Ew r Pr RIP Py fishes; its structure, variability, ontogeny. possible function and taxonomic utility. IAN J. HARRISON * Dipartimento di Biologia e Fisiologia Generali, Universita’ degli Studi di Parma, 43100 Parma, Italy GORDON J. HOWES Department of Zoology, The Natural History Museum, Cromwell Road, London, SW7 5BD. CONTENTS TESAIONCH TG CLO Ute cian Peete ba eee es cee re Mes At a Oe Rat, cand Reap, + Sa none ee Ste ee Soe 111 bE OCS SING) IF CTTSTAE SS ie Neng. SNe aula th UT er rat ae rar ee oe eee et a ae ee ee eins 112 SiMmUCHMC One PuArynoO branchial OLSAM CE BO) 5... 56 80.0 mm (n=8) 27.3-36.0 30.9 22.2-29.3 25.6 Cheek* 16.0-31.4 23.9 Lower jaw* 34.8-48.9 41.2 Cp. 15.9-21.0 17.9 Cp.I/d 1.2-1.7 Mode 1.5-1.6 D. XIV or XV (Mode XV) 10-12 (Modes 11 & 13) A. III 8 or 9 (Mode 8) Ll.sc. Cheek scales Gr. 30-34 (Modes 32 & 33) 4-6 (Modes 4 & 5) 7-10 (Mode 9) Okavango system n= 58 Size range 27.5-102.0mm SL Kazungula (Zambezi system) n=5 Size range 53.0-75.0mm SL M M 30.8-40.6 Bk) 30.2-35.7 32.4 31.0-37.7 33.8 34.7-36.7 35.9 14.3-24.2 20.5 15.8-20.8 18.5 14.3-22.9 19.9 15.4-20.8 iyi) 25.0-36.7 32.2 31.6-35.6 33.5 <70.0 mm S.L.(n=31) 28.0-36.4 31.8 29.2-31.6 29.6 >70.0 mm S.L. (n=27) 25.4-21.3 28.3 17.6-27.0 24.7 20.5-22.9 DMR) 35.7—43.6 39.7 37.5-40.0 38.4 15.4-21.8 17.6 16.9-19.5 18.1 1.2-1.8 Mode 1.5-1.7 1.4-1.8 XIV-XVI (Mode XV) XIV or XV 11-14 (Mode 12) 11 or 12 III 7-10 (Mode 8) Ill 8 31-35 (Mode 33) 32-34 4-5 (Mode 4) 3 or 4 7-12 (Mode 10) 8 or 9 | REVISION OF PHARYNGOCHROMIS 51 Table 2 Chromis jallae Pelmatochromis darlingi Pelmatochromis multiocellatus Chromis acuticeps Holotype, 59.0mm SL Holotype, 84.5mm SL Holotype, 97.0mm SL Lectotype, Paralectotype, 59.5mm SL 68.0mm SL Depth* 28.8 34.3 34.0 32.0 32.4 Head* 30.5 34.3 Sey) 37.0 37.9 Preorb.* 222 20.7 18.6 1529 1557 Interorb.* 16.6 20.7 20.1 20.5 20.8 Snout* 33:3 33.7 33.1 295 3373 Eye* 27.8 293 259 ifs) ies) Cheek* 25.0 Zi 24.4 220 PENS) Lower Jaw* 36.1 41.4 43.1 40.9 43.1 Cp.* 21.8 175 15-3 18.5 17.6 Cp. V/d 1.8 1.4 ie 1.6 SS) D: XV, 10 eve lel XVI, 10 XV, 11 XV Lt A. Ill, 8 Ill, 8 Ill, 9 0t, © V5) ENse, BY) ca.30 Sy Sil ca.32 Ck:se. 6 5 5 4 5 Gr. 8 10 9 10 2) Rhodes University) gave so much of her time to help with the dull task of measuring specimens; to both of them, my warmest thanks. For considerable technical assistance in providing photo- and radiographs it is a pleasure to thank Robin Stobbs of the J.L.B. Smith Institute, as it is to thank Elaine Grant for exercising her artistic talents when producing the figures, and Huibre Tomlinson for her patience, skill and tenacity when preparing the typescript. Finally, there is my indebtedness to Professor Mike Bruton, Director of the Institute, who arranged and financed my two visits to the Okavango swamps, without which my contact with P. acuticeps would have been far less intimate and informative, and to Professor Tom Hecht of Rhodes University’s Department of Ichthyology and Fisheries Science for his, and his Department’s, hospitality, the loan of optical equipment, and companionship in the field. REFERENCES Balon, E.K. & Muyanga, E.D. 1974. The Darling’s dwarf bream, Haplochromis darlingi (Boulenger 1911). pp. 417-419 Jn: Balon, E.K. & Coche (Eds) Lake Kariba: a man-made ecosystem in central Africa. Dr W. Junk. The Hague. — & Stewart, D.J. 1983. Fish assemblages in a river with unusual gradient (Luongo, Africa-Zaire system), reflections on river zonation, and descrip- tion of another new species. Environmental Biology of Fishes 9 (3/4): 225-252. Barnard, K.H. 1948(a). Report on a collection of fishes from the Okovango river, with notes on Zambezi fishes. Annals of the South African Museum 36 (5): 407-458. — 1948 (b). Revision of South African Cichlidae: synonymy and notes on the species. Provincial Administration of the Cape of Good Hope, Report No. 5 (1948): 48-61. Bell-Cross, G. 1975. A revision of certain Haplochromis species (Pisces: Cichlidae) of Central Africa. Occasional Papers of the National Museums and Monuments of Rhodesia, Series B, Natural Sciences 5 (7): 405-464. — Minshull, J.L. 1988. The Fishes of Zimbabwe. 294 p. Trustees of the National Museums and Monuments of Zimbabwe. Harare, Zimbabwe. Boulenger, G.A. 1896. Liste des poissons recueillis par le R.P. Louis Jalla a Kazungula, haut Zambese. Bolletino dei Musei di Zoologia et di antatomia comparata della R. Universitata die Torino 11: 260. — 1911. Description of a new cichlid fish (Pelmatochromis darlingi) from Mashonaland. Annals and Magazine of Natural History (8) 7: 377. — 1915. Catalogue of the Fresh-water Fishes of Africa in the British Museum (Natural History) 3. xii + 526 p. London. Bowmaker, A.P., Jackson, P.B.N. & Jubb, R.A. 1978. Freshwater fishes. pp. 1183-1230. In: Werger, M.J.A. & van Bruggen, A.C. (Eds) Biogeogra- phy and Ecology of Southern Africa: Monographiae Biologicae, 36. Junk. The Hague. Cracraft, J. 1989. Speciation and its ontology. pp. 28-59. In: Otte, D. & Endler, J.A. (Eds) Speciation and its Consequences. 679 p. Sinauer Associ- ates Inc. Massachusetts. Gilchrist, J.D.F. & Thompson, W.W. 1917. The freshwater fishes of South Africa. Annals of the South African Museum 11: 456-575. Greenwood, P.H. 1974. The cichlid fishes of Lake Victoria, east Africa: the biology and evolution of a species-flock. British Museum (Natural History) Zoology Supplement No. 6: 1-134. — 1979. Towards a phyletic classification of the ‘genus’ Haplochromis (Pisces, Cichlidae) and related taxa. Part 1. Bulletin of the British Museum (Natural History) Zoology 35 (4): 265-322. 1981. The Haplochromine Fishes of the East African Lakes. 839 p. Kraus-Thomson Organization GmbH. Munich. 1984. The haplochromine species (Teleostei, Cichlidae) of the Cunene and certain other Angolan rivers. Bulletin of the British Museum (Natural History) Zoology 47 (4): 187-28. — 1991. Speciation. pp. 86-102. In: Keenleyside, M. (Ed.) Cichlid Fishes. Chapman & Hall. London. Hert, E. 1989. The function of egg-spots in an African mouth-brooding cichlid fish. Animal Behaviour 37: 726-732. Hustler, K. & Marshall, B.E. 1990. Population dynamics of two small cichlid fish species in a tropical man-made lake (Lake Kariba). Hydrobiologia 190: 253-262. Jackson, P.B.N. 1961. The Fishes of Northern Rhodesia. xv + 140 p. Govern- ment Printer. Lusaka. Jubb, R.A. 1952. Some notes on freshwater fishes in Southern Rhodesia. Family Cichlidae, continued. Rhodesian Agricultural Journal 49 (4): 235-237. — 1961. An Illustrated Guide to the Freshwater Fishes of the Zambezi River ix + 171 p. Stuart Manning. Bulawayo Rhodesia. — 1967. The Freshwater fishes of Southern Africa. vii + 248 p. A.A. Balkema. Cape Town & Amsterdam. —— 1968. A new Chetia (Pisces, Cichlidae) from the Incomati river system, eastern Transvaal, South Africa. Annals of the Cape Provincial Museums (Natural History) 6 (7): 71-76. — Gaigher, I.G. 1971. Check list of the fishes of Botswana. Arnoldia 5 (7): 1-22. Kenmuir, D. 1983. Fishes of Kariba. p. iv + 132. Wilderness Publication. Harare, Zimbabwe. Leviton, A.E., Gibbs, R.H., Heal, E. & Dawson, C.E. 1985. Standards in herpetology and ichthyology: Part 1. Standard symbolic codes for institu- tional resource collections in herpetology and ichthyology. Copeia 1985 (3): 802-832. Otte, D. & Endler, J.A. (Eds). Speciation and its consequences. 679 p. Sinauer Associates Inc. Massachusetts. Paterson, H.E.H. 1985. The recognition concept of species. pp 21-29. In: Verba, E.S. (Ed.). Species and Speciation. Transvaal Museum Monograph (4). Pretoria. Pellegrin, J. 1920. Poissons noveaux du haut Zambeze, recueillis par M.V. Ellenberger (2° Note). Bulletin de la Societe Zoologique de France 45: 148-152. 2 P.H. GREENWOOD Nn —— 1936. Contribution a lichthyologie de Angola Archivos de Museu Steindachner, F. 1866. Ichthyolgische Mittheilungen (IX). I. Ueber einige neue Bocage, Lisboa 7: 45-62. Susswasserfische von Angola. Verhandlungen der Zoologisch-botanischen Poll, M. 1967. Contribution a la faune ichthyologique de Angola. Museu do Gesellschaft in Wien 16: 764-5. Dundo, Publicacoes Culturais No. 75: 1-381. Trewavas, E. 1961. A new cichlid fish in the Limpopo basin. Annals of the Regan, C.T. 1922. The classification of fishes of the family Cichlidae: II. On South African Museum 46 (5): 53-56. African and Syrian genera not restricted to the Great Lakes. Annals and —— 1973. II. A new species of cichlid fishes of rivers Quanza and Bengo, Magazine of Natural History (9) 10: 249-264. Angola, with a list of the known Cichlidae of these rivers and a note on Skelton, P.H., Bruton, M.N., Merron, G.S. & van der Waal, B.C.W. 1985. The Pseudocrenilabrus natalensis Fowler. Bulletin of the British Museum (Natural fishes of the Okavango drainage system in Angola, South West Africa and History) Zoology 25 (1): 27-37. Botswana: Taxonomy and distribution. /chthyological Bulletin of the J.L.B. — 1983. Tilapiine fishes of the genera Sarotherodon, Oreochromis and Smith Institute of Ichthyology No. 50: 1-21. Danakilia. viii + 583 p. British Museum (Natural History). London. Bull. Br. Mus. nat. Hist. (Zool.) 58(1): 53-59 Issued 25 June 1992 Description of a new species of Microgale (Insectivora: Tenrecidae) from eastern Madagascar PAULINA D. JENKINS Department of Zoology, British Museum (Natural History), Cromwell Road, London SW7 5BD Synopsis. A new species of Microgale (shrew-tenrec) from primary forest in eastern Madagascar is described. Morphological comparisons are made with other members of the genus, in particular with those of the Microgale cowani and M. gracilis clusters, with which it shows greatest affinity. INTRODUCTION Eleven species of shrew-tenrecs belonging to the genus Microgale are currently accepted (MacPhee, 1987; Jenkins, 1988). The genus was believed to be much more diverse until the revision by MacPhee (1987) showed that over half of the named forms of Microgale were merely juveniles or morpho- logical variants. Fortunately in the present case, adults of both sexes, a subadult and a juvenile were included in the small sample collected, so that adults could be distinguished with a good measure of confidence from those of other species and some observations were possible on the decidu- ous dentition. Distinctive cranial and dental features charac- terise the new species, which is further distinguished from most other species by its large size. Brief comments are included on the other species of Microgale collected from the same locality. MATERIALS AND METHODS Small mammal trapping was carried out for two months by the Madagascar Environmental Research Group in primary forest of the Ambatovaky Forest Reserve (Barden, in prep.). A mixed collection of rodents and small tenrecs was col- lected, including the undescribed form. Preliminary identifi- cations based on external features were made by members of the Research Group, who then brought the specimens to The Natural History Museum [BM(NH)] for formal identifica- tion, which in these mammals, requires the examination of a suite of craniodental characters in addition to external fea- tures. The preliminary identifications were recorded (Nicoll & Rathbun, 1990), some of which were found to be incorrect following detailed examination. Measurements were taken with dial calipers and are given in millimetres. The dental nomenclature follows that of Mills (1966), Swindler (1976), Butler & Greenwood (1979) and MacPhee (1987). The following abbreviations are used in the text c.—circa, GCL—greatest cranial length, HB—head and body length. Microgale dryas sp. nov. HOLOTYPE. BM(NH)91.230, collector's number RAN 33610, adult female, in alcohol, skull extracted. Collected 20 February 1990 by Tanya Barden and Christopher Raxworthy, Madagascar Environmental Research Group from Site 1, Ambatovaky Special Reserve, [northeast] Madagascar, 16°51’S 49°08°E, in primary rainforest, between 600-750 metres altitude. Paratypes: BM(NH) 91.227, collector’s number RAN 33592, adult male; BM(NH)91.228, collector’s number RAN 33593, juvenile female; BM(NH)91.229, collector’s number RAN 33596 subadult male. All with the same collection data as the holotype but collected 14 February 1990. RESULTS Diagnosis Intermediate in size between the smaller M. thomasi Major, 1896 and M. gracilis (Major, 1896), and the larger M. dobsoni Thomas, 1884 and M. talazaci Major, 1896. Braincase narrow relative to skull length. Upper second and third premolars (P* and P*) with well defined anterior ectostyle, separated by a notch from the distinct posterior ectostyle and distostyle. Mid region of guard hairs of the dorsal pelage flattened and broadened in cross-section. Description Size large; external measurements follow with those of the holotype in brackets: head and body length 105.5—-113.5 (105.5), mean 110.9, SD 3.16; tail length 68.0-70.5 (70.4), mean 69.5, SD 1.02; hindfoot length 18.1-18.7 (18.7), mean 18.5, SD 0.23; weight 38-40 grams (38), mean 39.25, SD 0.83. Tail greater than half as long as head and body length: 60-67% (67%), mean 62.8%, SD 2.68. Dorsal pelage dark reddish or greyish brown, with a grizzled appearance; bases of hairs grey, distal portion light brown or red brown, some with black tips; interspersed with long guard hairs which are grey at the base but black for most of their length; unlike any other member of the genus, these hairs are flattened and 54 broadened in cross-section in their mid portion, rounded in cross-section distally. Hairs of ventral pelage grey at the base with light grey tips; colour of dorsal pelage merges gradually with that of the venter. Forefeet grey brown dorsally, light ventrally; hindfeet grey brown above and below; tail uni- formly grey. The claws of the forefoot are elongated; claws of the third digit of the hindfoot are 65.1—78.9% (65.1%), mean 71.28, SD 5.63 of the length of those of the forefoot. The skull is long and gracile (see Figs 1-3); the rostrum is elongated but moderately robust; the interorbital region is narrow and slightly concave in dorsal view; the braincase is long and deep, yet narrow; the squamosal region is not inflated and the superior articular facet is angular and clearly visible in dorsal view; the sinus canal forms a markedly peaked curve (see Fig. 3). The mandible is long, moderately robust, the corpus is sinuous in profile, with a moderately deep and broad coronoid process; the angle between the dorsal articular facet and the coronoid process is shallow; the mental foramen lies below the anterior portion of P3. The dentition is moderately robust and illustrated in Fig- ures 4 to 7. Interproximal diastemata are present between all the upper incisors and canine, those on either side of the first upper premolar (P7) are large. Posterior basal cusps (disto- styles) are well defined on all three upper incisors, that on the first incisor (I’) is robust and more than half the height of the principal cusp; anterior accessory cusps are scarcely evident on the upper incisors; the upper canine lacks an anterior accessory cusp and mesiolingual cusp, while the distostyle is small, slender and approximately one quarter of the height of the principal cusp; the first upper premolar (P7) is robust, P.D. JENKINS with well defined anterior and posterior basal cusps; the second upper premolar (P*) has a slender paracone, the anterior ectostyle is well developed and separated by a notch from the posterior ectostyle and distostyle, the talon is moderately large and the lingual shelf well developed; the third upper premolar (P*) is similar in structure to P® in buccal aspect but the paracone is more robust and the anterior ectostyle separated from the well defined posterior ectostyle and distostyle by an even more distinct notch, the talon is large with a small yet well defined cusp; the upper molars are similar to those of other members of the genus, but the talons of M! and M? are broad and deep, while that of Mp? is lingually extended. Small diastemata are present on either side of the third lower incisor (I,), the lower canine (C) and the first lower premolar (P,); a small anterior accessory cusp is present on the lower canine; the first lower premolar (P,) is large, being only slightly smaller than the second lower premolar (P3); the anterior and posterior accessory cusps of P, are well marked, the main cusp is ‘anteroflexed’ due to the short convex anterior slope and the longer, concave posterior slope; the paraconid is well developed on P, and the third lower premolar (P,), and on the molars (M, to M;); the anterior face of the protoconid of P, and all the molars is markedly convex. : Etymology The name of this species is derived from the greek opvac, dryad or wood nymph. Soest inekoas ns sorrerepeanntnncnstnins scat nspenareetennetaaninttecenenanin Fig. 1 Dorsal view of skulls, from left to right Microgale dryas, M. gracilis, M. thomasi and M. cowani. Scale 500 mm. MICROGALE DRYAS SP.N. 35 AUREETT AAU EREGEARROANDGRDGOOOAOOUGOOORRDOROOONE Fig. 2 Ventral view of skulls, from left to right Microgale dryas, M. gracilis, M. thomasi and M. cowani. Scale 500 mm. Comparison with other species Microgale dryas (HB 105-114, GCL 30-32) is intermediate in size between Microgale thomasi (HB < 98, GCL c.27) and M. gracilis (HB c.93, GCL c.29), and M. dobsoni (HB < 103, GCL > 29) and M. talazaci (HB > 115, GCL > 34). It is considerably larger than the other known species of Micro- gale (HB < 83, GCL < 25), see MacPhee (1987, table 2). It is readily distinguished from M. talazaci and M. dobsoni in which I, is larger than the lower canine, while, as in the other species of Microgale, I, is smaller or subequal to the lower canine. M. dryas is distinguished from all other species by the dorsal pelage, in which the guard hairs are flattened and broadened in their mid region. On cranial and dental charac- ters it is clearly associated with the cowani cluster [see MacPhee (1987), p.9], which includes M. cowani Thomas, 1882, M. parvula Grandidier, 1934, M. pulla Jenkins, 1988 and M. thomasi, and the gracilis cluster (M. gracilis). All members of the cowani and gracilis clusters have gracile skulls with a long, narrow rostrum and diastemata between the anterior teeth. The skull of M. dryas is larger than any of the other members of the cowani or gracilis clusters and is intermediate in elongation of the rostrum between M. tho- masi and M. gracilis. M. gracilis shows the greatest degree of attenuation of the rostrum, which is slender, with very long diastemata between the anterior teeth, in M. dryas the diastemata are moderately long (especially between the upper canine and P’) and the rostrum is narrow (in these dimensions the new species resembles M. cowani) but in M. thomasi the diastemata are small and the rostrum is relatively broader and shorter (see Table 1). The interorbital region of M. dryas is narrow and slightly concave in dorsal view, in contrast to other members of the cowani and gracilis clusters in which the interorbital region increases in size from the anterior to the posterior region. The braincase of M. dryas is narrower relative to skull length than any of the other members of the genus; it is slightly narrower but deeper than that of M. gracilis, yet markedly narrower but deeper than that of M. thomasi (see Table 1). The squamosal region dorsal to the bulla is scarcely inflated in M. dryas, slightly inflated in M. gracilis, inflated in M. cowani and markedly inflated in M. thomasi; the sinus canal follows a shallow curve in M. cowani, M. thomasi and M. gracilis but forms a peaked curve in M. dryas. The corpus of the mandible of M. dryas is sinuous as in M. cowani and M. gracilis, unlike the straighter profile of M. thomasi; the mandible is shallower at the coronoid process in M. dryas, M. cowani and M. gracilis, than in M. thomasi. Although only slightly larger than M. gracilis and with similarly elongated claws on the manus, M. dryas differs markedly from it in the cranial features given above and the following dental features. The toothrow length in M. gracilis is not markedly shorter than that of M. dryas, due to the much longer diastemata between the anterior teeth of M. gracilis, than those of M. dryas. The teeth of M. gracilis are smaller in all dimensions than those of M. dryas (buccal length x crown height of P, 1.08 in M. gracilis but 1.64—1.85 in M. dryas). The most marked dental difference between the P.D. JENKINS Fig. 3 Lateral view of skulls, top row left Microgale cowani, right M. gracilis, bottom row left M. thomasi, right M. dryas. Scale 500 mm. M. dryas and M. gracilis is in the size of the talon of the molariform maxillary teeth; this is large in M. dryas but in M. gracilis is effectively absent and more reduced than in any other species. Microgale dryas and M. thomasi differ in the following dental features. The distostyle of I’ is more robust and greater than 50% of the height of the principal cusp in M. dryas, while it is more slender and less than 50% of the height of the principal cusp in M. thomasi. A mesiolingual accessory cusp is present on I? in M. thomasi but absent in M. dryas. A mesiolingual cusp is present and the distostyle is larger, more robust and approximately one third of the height of the principal cusp in M. thomasi, while in M. dryas the mesiolin- gual cusp is absent or reduced to a ridge and the distostyle is small, slender and approximately one quarter the height of the principal cusp. In M. dryas the anterior ectostyle of P? is well defined and separated from the distostyle, and the talon is large, unlike the condition in M. thomasi in which the anterior ectostyle is not separated and the talon is small. The posterior ectostyle and distostyle of P* are moderately well defined and separated from the anterior ectostyle by a notch, the talon is large with a well defined cusp in M. dryas but in M. thomasi there is no posterior ectostyle, the distostyle is barely evident and merges with the anterior ectostyle, and while the talon is moderately large it lacks a well defined cusp. A posterior ectostyle is present on M’ and the talon is large and unicuspid or bicuspid in M. dryas but in M. thomasi there is no posterior ectostyle and the talon is medium sized and unicuspid. In all the molariform teeth the talon of M. dryas is larger than that of M. thomasi. There are fewer differences in the mandibular teeth of the two species. The incisors are similar but there are no diastemata between the incisors of M. thomasi, while in M. dryas a diastema is present between I, and the canine of all specimens and between I, and I, of three of the four specimens. An anterior accessory cusp is present on the canine in M. dryas but not in M. thomasi. Although P, is similar in both species, there is a slight difference in shape, in M. dryas the tooth is anterof- lexed and tends to be caniniform, while in M. thomasi it is not anteroflexed and more molariform in appearance. P, and P* in both species are larger relative to the rest of the toothrow than in any other species (see Table 1). The molariform teeth (P, to M,) are similar in the two species except that the anterior face of the paraconid of M, and M, is markedly convex in M. dryas but only slightly convex in M. thomasi. DISCUSSION Microgale is a taxonomically complex genus containing many named forms, over half of which were shown to be juveniles or morphological variants (MacPhee, 1987). In his revision, MacPhee demonstrated the high morphological within- species variation found in the genus, and described and MICROGALE DRYAS SP.N. 57 Table 1 Comparing M. dryas, M. thomasi and M. gracilis Tail length 73-81! 62-70! 68-71 om : mean 78.0 67.2 69.53 M. gracilis M.thomasi M. dryas SD 3.46 4.91 1.02 n 4 3 4 Condyloincisive length 29.0 25.9,26.8 30.4-31.0 Ratio of tail to head 0.73-0.87' 0.66-0.75! —0.60-0.67 mean 30.63 and body length SD 0.23 mean 0.83 0.69 0.63 n 4 SD 0.03 Upper toothrow length 14.4 125513 MH 15.2-1526 =, : 2 : mean 15.43 SD 0.18 Note: | data from MacPhee (1987) n 4 Length of anterior teeth 8.4 6.6 7.5-8.0 (I'—anterior of P?) mean 7.78 SD 0.18 n 4 Breadth of rostrum Med) 3.5, 3.6 3.5-3.6 (P2-P?) mean Shap) SD 0.05 n 4 Ratio of length of 0.58 0.50,0.53 0.49-0.52 anterior teeth (I'—P’) to upper toothrow length mean 0.50 SD 113 n 4 Ratio of breadth of 0.17 0.27,0.28 0.22-0.24 rostrum (P?—P”) to upper toothrow length mean 0.23 SD 0.47 n 4 Braincase breadth 11.3 ke NaS) 11.2-11.6 mean 11.4 SD 0.14 n 4 Braincase height Fig, D550, 7.9-8.2 mean 8.05 SD Ort n 4 Ratio of braincase breadth 0.39 0.43,0.44 0.37-0.38 to condyloincisive length mean 0.37 SD 0.36 Fig. 4 Buccal view of left maxillary dentition of Microgale dryas n 4 (top), M. thomasi (middle), M. gracilis (bottom). Scale 1 mm. UES ale ria i sigs Agia ss illustrated the deciduous and adult dentitions of most species. mean 0.71 Since the small sample of the new species contained a SD 0.36 juvenile, subadult and two adults, it was possible to be “ 4 confident that the specimens did indeed represent an unde- Ratio of mandible height 0.27 0.33, 0.36 0.26-0.30 scribed species. MacPhee suggested the existence of growth at coronoid process to curves, unusual in mammals, in which some subadults may mandible length exceed the average size of adults. This feature may be a ae indicated by M. dryas, in which both the juvenile and subadult specimens are slightly larger than the adults in head : . and body length, although the small sample size precludes Buccal length x crown 1.08 1.49 1.64-1.85 any meaningful comparison. al Ores 174 MacPhee divided the genus into six ‘clusters’ on the basis of SD 0.08 dental traits and body proportions; he emphasised that this n 3 was a phenetic, not a phylogenetic arrangement. On the basis Head and body length —c.93 91,97 qoseenigss Obmtheacharactemermployed:by.Mackhee, pM, wihyasyerouns macad j 110.88 with the cowani and the gracilis clusters and is intermediate in SD 3.16 many features between M. thomasi and M. gracilis. Since M. n 4 thomasi, M. gracilis and M. dryas are known from such small Fig. 5 Buccal view of left mandibular dentition of Microgale dryas (top), M. thomasi (middle), M. gracilis (bottom). Scale 1 mm. Fig. 6 Lingual view of left P? to M? of Microgale dryas (top), M. thomasi (middle) and M. gracilis (bottom). Scale 1 mm. samples, it is impractical to speculate about possible relation- ships. Eisenberg & Gould (1970) divided Microgale into four behavioural classes on the basis of external morphology. However this classification was challenged by MacPhee (1987) because of redefinition of within-species variation and lack of field study data to support the theory. Specimens of three other species of Microgale: M. cowani, M. principula Thomas, 1926 and M. talazaci were collected from the same locality as M. dryas. This sympatric association of several different species is apparently common in Microgale (see MacPhee, 1987; Nicoll & Rathbun, 1990). Regrettably, some P.D. JENKINS Fig. 7 Occlusal view of left P? to M? of Microgale dryas (top), M. thomasi (middle) and M. gracilis (bottom). Scale 1 mm. of the species recorded for this locality by Nicoll & Rathbun were based on incorrect preliminary field identifications. Although found in the same habitat, it seems likely that these four species are occupying different ecological niches. Eisen- berg & Gould (1970) hypothesised that M. principula was a climbing form on the basis of its long tail, which is naked on its distal dorsal surface, and long hindfeet. Although MacPhee (1987), pointed out that there was no evidence of the long tail being prehensile, and studies that might confirm such locomotor behaviour are lacking, these morphological differences do suggest adaptations to a specialised life-style. Studies were made on M. talazaci (Eisenberg & Gould, 1970), which show that it is scansorial and shows some burrowing behaviour; its much greater size suggests that it may take larger prey than the smaller species. Since there is no field data for M. dryas, no speculation about its ecology or behaviour is attempted here. A species of rice-tenrec, Ory- zorictes talpoides and two species of rodent, Eliurus minor and E.myoxinus were also collected from the same locality. The specimens of M. principula from this locality represent a northern extension of the recorded range for this species (see MacPhee, 1987). ACKNOWLEDGEMENTS. I am very grateful to Tanya Barden and Chris- topher Raxworthy of the Madagascar Environmental Research Group who collected shrew-tenrecs and rodents and donated them to the National Collection. I thank Ross MacPhee, Department of MICROGALE DRYAS SP.N. Mammalogy, American Museum of Natural History for reviewing the manuscript and providing constructive advice and criticism. I am also grateful to Iain Bishop, Department of Zoology, The Natural History Museum for reading and commenting on the manuscript. REFERENCES Barden, T. in prep. Report on the Madagascar Environmental Research Group expedition to Ambatovaky Special Reserve. Butler, P.M. & Greenwood, M. 1979. Soricidae (Mammalia) from the Olduvai Gorge, Tanzania. Zoological Journal of the Linnean Society 67: 329-379, 18 figs. Elcenberg, J.F. & Gould, E. 1970. The tenrecs: a study in mammalian behavior and evolution. Smithsonian Contributions to Zoology (27) :1-138. Grandidier, G. 1934. Deux nouveaux mammiféres insectivores de Madagascar Microgale drouhardi and Microgale parvula. Bulletin du Muséum National a9 d Histoire Naturelle. Paris (2) 6: 474-477. Jenkins, P.D. 1988. A new species of Microgale (Insectivora: Tenrecidae) from northeastern Madagascar. American Museum Novitates (2910): 1-7. Major, C.I.Forsyth 1896. Diagnoses of new mammals from Madagascar. Annals and Magazine of Natural History (6) 18: 318-325. MacPhee, R.D.E. 1987. The shrew tenrecs of Madagascar: systematic revision and holocene distribution of Microgale (Tenrecidae, Insectivora). American Museum Novitates (2889): 1-45. Mills, J.R.E. 1966. The functional occlusion of the teeth of Insectivora. Journal of the Linnean Society (Zoology) 47: 1-25, pl.1. Nicoll, M.E. & Rathbun, G.B. 1990. African Insectivora and elephant-shrews: an action plan for their conservation. [UCN, Gland, Switzerland. Swindler, D.R. 1976. Dentition of living primates. London, Academic Press. Thomas, [M.R.] Oldfield 1882. Description of a new genus and two new species of Insectivora from Madagascar. Journal of the Linnean Society 16: 319-322. — 1884. Description of a new species of Microgale. Annals and Magazine of Natural History (5) 14: 337-338. — 1926. On some small mammals from Madagascar. Annals and Magazine of Natural History (9) 17: 250-252. r BD at * ee ~~ pes “ wat to Vire . Nevery tat ie Br fa fod atl ate 1 mylinee se niet oul at & Praia =" JInveechresnd) sel halted j \ veges jo nosis suisse Sa a if moritee@nl! othe ei ws AwsT | AA lke " = *, ae \ da) cali ean 4 MA . ari oR. iit i. Matt v a = : ~etebs od i ie wrt cares WAS wah rode tevin’ ai ymiiy | Arateen Vine oemmeeny of be thee come, pare vole) 40 lige? > > - “oyagead (nib Se a ee mt Sheth 014) nee ieTEnt® sit) at (athwrarte ht) snbiahind | 8 Pian tthe sommes h u's tanh. wrnmibald real eapAbaagit Ws «> 21 MOREE Teta ean fh Yo Tom vag Sasha ee en ee xg) Sra ae ae it~ - p a8) P ang , WU Bra 5 Cone - otvnsid Beara i ching 9 Ths: > f o¥2 Mo Pow geal 4 Htemeryerts Bip a — : Pe Cae Cw LE SIV) y aay Thana! ¥ —- ; ~ LF “= " va, Le i Mi vf / ‘ ; ‘ Ley rv ‘Sel : : are vi mifat . ete — . ae SN Me AR os , difteremt Gealdg : a ae og oe Ceowdd (1970; Sy pet coh : oe lag for Oey the b A. foi tall, Wilk : 1 “theta uriace. sad hong athe Vg : - 1a In? ), oie? cut them hee bed : > a ¥ a 7 ¢. t i . rs ra - 7 .< ' ' x rt. 5 tou ait i. —- P ee 7 ' bavweit ch great ” , ; - © ti Lak reed necy tian tiie ane F eo * Adtatet i Ad. eivvas, 0 1 HPACHLATOR : , : org P ; ary. 4 ; tude ThA a rir mp woo tC. f s} Fie et ‘a Ms 5 i | Site et | ang (des ain) Oo apoge ta we vibe fit, Ek es - oa € vag nytt . ~~ 4 uu vile to frog ) the ne - Ties arecte em Ol ak! lat rig thie, loeatily rh : ren) NSO the roe shed 0 range ‘fer th ? < rg 4 G . < ° I Ary sy | Z ' . hey , 1 . iW rue M a sigs Pe en 9 wes fae lop tea sel aoa -_ * 4 j } S om 7 dh sik Kom ‘Ma Bull. Br. Mus. nat. Hist. (Zool.) 58(1): 61-93 Studies on the deep-sea Protobranchia (Bivalvia): the family Nuculidae P.M. RHIND* AND J.A. ALLEN University Marine Biological Station, Millport, Isle of Cumbrae, KA28 0EG CONTENTS SV ADLO) OSS * cates ite, Behe. enemas contre io maar? een Gober strat ian eNOERn See © Tene mo See a PTA INO SNES ES 61 MEXR OUINCM OME tear Me Pere te rete. i. cee aoe un tagnc cr cmeaet oe siini ack MeeMiaane Santana RETR nr second sea Muara uinachaats agains’ 61 Mic eREaeANGh WLC UNOOS mer erate tc Sertee nme etinae toe etme ce cision Sux enabwind tnotnenaine Stas msacsie sates Gudseanens tara elons ott trem eatin 62 SR ANO MOMMY Pe. fe era: RT eH Aaa iste Nc RPE. Noiseless oe opematoatyes om nqahaset analune + sebiscneeian ster acotidenns ce sn stuns 63 ROPASSINIC ALONG): Petes ck whee cat eee Rene nbc NRO Eee ts cvs cdhreamae ashe tulet oie ie neta cs tein sitiednete tebauk teks see thateeentond 64 PMI ATICIC DICE ISCAS PECIESIN SRE. Maat TIA ME tad dane Reon sinned ean ata S Arcee stadia ene MeSan de Sats oatecn cept eny die Wet weeee 65 Deminucula aiacelland (JEMGeyS)oe. «see... vetetced tevionasnceer oh pote tee oie eh oe aes Meee emer es as eea Teen a abeete Ht. GU, 65 DRESCUAREG I CLCACHTIAG (UDNAIN) | Ruaveseten ee nee ace aAdac kis cs wacisiscou Moeee es cc Rabek 4 RR Leadon dee AMORUED «geet es eine edbblemenmulan tetiohads TD MEMOIICED USHBel(DONIMUISS)) Seer iee tc, «eee ee ea wsdl okt enltaer ans dsce hepa dee deh eta cthatatetenaek adden. dacediemetanieek + 73 IN ELEU GIG CADET OUCCTISIS'(SUMIUM) ii. zi ees ads Cay seeea ss nas Wee aelnate ns ath lddtadac es sdldunadede sine dae dade canuhh. seldbedddenae 75 INCH OINGNDET{OTAIGN HOW SPECIES ry apcpes deatefeaarcdhioauaa sae cckk nGhbee os Le ote ee ce oe Ce SEER « DueT es (Gage Se cbues 76 NGULOMG, CrAanulosa (WetwA) Esc! dalek flee ach See towed 045 00 opieldeioe ade tee Ses «Ghee Wes ded age use dake oa oe gedaan shachph stsaree de 78 INE GHOVH ESTER LSA GLE W, SPECIE SHAMIL 510 at EER Ute Sah saaana tasers sesh oceans cert tonne aaa hee paqatteen uentes pana eeaee geteeeent 80 INC UIONIE CLONGGIG, (HEN, SPECIES INO sic doe Eee hm Sedaka onic oe cae roared ede sean ee tere bnc cece ns yelbcelss sntcactadednrbeast dteathes 82 IB LEUITUCIN GLVCrIIIA Pall). 0. fe a hates ms Qette RE Tide sntied Aes dalle vc SM AN cod S Naat MORON ARN a 83 IBrevInUCHI GMD Inia eHIATIS (MEWISPECIES) i iztancse Gael... ..dre eoea tenes cackecetion du senate omohatelesclenelldon ce able soeeahbealecee 85 GthemMeep-seapnelantic NUCuMGsS: csgscch been pentasia... na. Sentra eee Mee eee a eee eee eee eaten 87 Oriping Antiquity 62 Diversity of the, DeepsSeaiINucultds ...{.3s25 ...t fae. sanemseemaaraaeas Mes «ciate. diane ee eemeeeane ese saaaenas 88 SGUSSIO MeN wetted awacdy tncctee tine redant esto t geek qantas dha ve. SRAMAEL CB ne ObasteaMan halle GcedbGhal es enens ene « «mousey 89 IRERCROHCCS MENU, (ou Perdinn steiutd iS actus aie: aise Senatermnuattssaee aap + «eon ROT gon aeRat ahha Ja betemadynancHehstlaecadhhigaetheGielun aeaqat 91 Synopsis. The morphology, geographical distribution and adaptations to life in deep water of ten species of the family Nuculidae (Bivalvia: Protobranchia) from the Atlantic are described. These include four new species. The. evolution of the family and the origins of the deep-water species are discussed and the taxonomy of the family Issued 25 June 1992 reassessed. * Field Studies Council Research Centre, Fort Popton, Angle, Pembroke, SA7 1AD INTRODUCTION This paper forms part of a continuing investigation of the biology of deep-sea protobranch bivalves of the Atlantic. (Allen & Hannah, 1989; Allen & Sanders, 1973, 1982; Sanders & Allen, 1973, 1977, 1985). The present study includes an appraisal of the biology, systematics and antiquity of ten species of the deep-sea Nuculidae. From the time of Linneaus taxonomists have had difficulty in classifying the Nuculidae (eg. Hancock, 1846; Adams, 1856; Sowerby, 1871; Seguenza, 1877; Jeffreys, 1881; Verrill, 1884; Reeve, 1885; Dall, 1886; Hedley, 1902; Iredale, 1931, Schenck, 1934, 1939; Allen, 1954; James, 1972; Lubinsky, 1972). The nuculids were originally regarded as members of the family Arcidae. The type species of the genus Nucula, and of the family Nuculidae, is Arca nucleus Linné. This is now generally accepted as equivalent to Nucula nucleus auct. As with other protobranch groups (Allen, 1978) taxonomic difficulties arise because of the very great conservativeness of shell and body form. The Subfamily Nuculacea has a long fossil record possibly dating from the Ordovician. The apparent stasigenesis involv- ing few cladogenetic events possibly results from an early adaptation to a common homogeneous environment that has persisted throughout much of the zoic period. The magnitude of this early adaptation appears to have left little room for selective pressure change. For example, Levinton & Bam- bach, (1975) in a comparative study of Silurian and Recent deposit-feeding bivalve communities found that non- siphonate deposit-feeding nuculaceans were dominant forms in the non-compacted, watery sediments of both environ- ments, even though separated by 400 million years. The Silurian counterpart to Nucula was a species of Praenucula which is similar in size and morphology to Nucula and probably had similar life style as an active deposit feeder, just below the sediment-water interface. Rhoads & Young, (1970) found that intensive near-surface reworking of the 62 sediment by non-siphonate protobranchs (Nucula proxima) creates an unstable, fluidized sediment. Similar conditions are thought to have been created by Praenucula in corre- sponding Silurian environments (Levinton & Bambach, 1975). According to Rhoads & Young, (1970) this has the effect of excluding siphonate forms and suspension feeders from the deposit-feeder biotope. The reduced competition resulting from the exclusion by non-siphonate protobranchs of more advanced forms, especially those of the massive post-Palaeozoic radiation of infaunal, siphonate heterodonts (Stanley, 1968), may further help to explain why the nuculoid forms have persisted virtually unchanged for such an enor- mous period of time. This ability to thrive in fluid sediments also pre-adapted the nuculids for a deep-sea existence. The nuculids have been regarded as being close to the stem stock of many proposed phylogenies of the bivalves (eg. Pelseneer, 1911; Yonge, 1939) and, even though they are not now regarded as being similar to the ancestral bivalve (Atkins, 1938; McAlester, 1964; Allen & Sanders, 1969; Morris & Fortey, 1976; Babin, 1977; Allen, 1985), they are probably close to the ancestral stock of the subclass Proto- branchia. An examination of the fossil record has been carried out to endeavour to establish which of the extant species are close to the ancestral stock furthermore, because the nuculids are such an archaic group it was thought that it might provide a contribution to the debate on the origin and antiquity of deep-sea fauna. MATERIAL & METHODS Specimens used in this study were sampled as follows: 1) Marine Biological Association of the United Kingdom (1967)—Bay of Biscay, R.V. ‘Sarsia’; 2) Institute of oceano- graphic Science, U.K. (1968)—between Fuerteventura and the West African coast, RRs ‘Discovery’; 3) Centre National de tri d’Oceanographic Biologique, France—Walda (1971), between Nigeria and the Angola, R.V. ‘Jean Charcot’; Biogas (1972-74)—Bay of Biscay, R.V. ‘Jean Charcot’, R.V. ‘La Perle’ and R.V. ‘Cryos’; Incal (1976), Rockall Trough and the North East Atlantic, R.V. ‘Jean Charcot’; 4) Woods Hole Oceanographic Institute, U.S.A.—North America Basin (1961-73), Gayhead-Bermuda transect, R.V. “Atlan- tis’, R.V. ‘Atlantis II‘, R.V. ‘Chain’, and R.V. ‘Knorr’; Cape Verde Basin (1967), between Dakar and Recife, R.V. ‘Atlan- tis I’; Angola Basin (1968), between Walvis Bay and Luanda, R.V. ‘Atlantis II’; Argentine Basin (1971), transect shelf slope break to abyss, R.V. ‘Atlantis II’; Guiana Basin (1972), transect shelf slope break to abyss, R.V. ‘Knorr’; West European Basin (1972), between Ireland and Woods Hole, R.V. ‘Chain’; 5) Scottish Marine Biological Associa- tion (1973-86), Rockall Trough, R.R.S. ‘Challenger’; 6) Naval Ocean Research and Development Activities N.S.T.L. Station, U.S.A. (1981), Venezuela Basin, USNS ‘Barlett’. The vast majority of samples were obtained using various forms of epibenthic sledge (ES/ET/ED). The remaining sam- ples were obtained using a Sanders Dredge (SD), Agassiz Trawl (AT), Anchor Dredge (AD), Otter Trawl (OT), Anchor Box Dredge (ABD), Spade Box Corer (SBC) and Beam Trawl (CP = chalit a perche). The samples were elutriated on board using sieves (mesh 0.42 mm USA and U.K., 0.25 & 0.50 mm France), fixed in 4% or 10% formal P.M. RHIND AND J.A. ALLEN saline (4% formaldehyde in the case of S.M.B.A. samples) and then after 24 hours, washed and transferred to 70% or 95% ethanol. Histological Procedures For taxonomic comparison, distortion of the sections had to be reduced to a minimum. Much distortion occurs with excessive heat during wax infiltration and section flattening. This was minimized by using low melting point polyester wax (37°C) (Steedman, 1957). This wax is miscible in alcohol and obviates the need to use xylene prior to hydration thus further reducing distortion. Most sections were cut at 25 microns, below this they fail to expand to their original proportions during flattening process (Aumonier, 1938). Standard, Mayer’s haematoxylin was used with eosin as a counter stain and supplemented by sections stained with Azan to differentiate between acidophilic and basophilic cytoplasm (Humason, 1967). for muscle and connective tis- sue, the trichrome staining methods of Pantin, (1946) and Pollak, (1944) were used. Three-dimensional reconstruction requires a sequence of perfectly alignable serial sections. At least two static refer- ence points are necessary and which run the length of the embedded specimen. The technique of Pusey, (1939) was found to be the most satisfactory. The block was painted with an indelible colloidal stain made up of Indian ink and bile salts. The procedure has been used mainly to determine configuration of the coils of the hindgut. Stained whole mounts were also prepared using Ehrlich’s haematoxylin. Shape Analysis: Specimens from nine abyssal Atlantic Basins were drawn in lateral view using a Wild MS microscope with a drawing attachment. In the case of large samples specimens were selected randomly by spreading them over graph paper and then selecting them by X, Y co-ordinates specified by computer-generated random numbers. Using a Hewlett Pack- ard Shape Analysis System the drawings were analysed with reference to the area and perimeter of each shape. Then, a series of linear measurements were made (Fig. 1), quantify- ing nine variables in each specimen. A further ‘dummy’ variable was added recording the presence or not of external radial shell markings. Analysis was by computer-based statis- tics. Standardization of the data to z-scores was achieved by substracting the mean (X) from each variate (Y) and dividing by the standard deviation (S) to give the rth standardized value Zr = (Yr-X/)/S The data set so derived has a mean of zero and a standard deviation of one. In morphometric terms this transformation has three advantages. First, all values less than the mean have a negative sign and all greater are positive, with the result that distance (or degree of similarity) between individuals can be inspected visually. Second, it is possible to relate the distance from one individual to another or from one individ- ual to a point of reference, (eg. a centroid) independent of the unit of measurement of the original data. Third, for multivariate analysis it is an advantage to have a standard variance of one in that each variable is on the ‘equal footing’ in a geometrical sense such that each variable is treated symmetrically and independent of the variance. It also was found that because the range in size of the specimens was STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 63 A Fig. 1 Measurements (mm) A, B, C, D, E, F & G together with the area perimeter length for each specimen. large it was necessary to eliminate a size effect by dividing each linear variable by length, and area by length squared. This is important because if shape differences are subtle (as many are) then size will tend to predominate in any clustering procedure and clusters will form on the basis of size rather than shape. The morphometric data was subjected to multivariate analysis using the ordination technique of principal co-ordinate analysis (Gower, 1966). This was carried out using the software package GENSTAT on the Edinburgh University mainframe computer. Like principal component analysis, principal co-ordinates analysis is used to reduce dimensionality but, instead of being based on~a P X P dispersion or correlation matrix where there are measure- ments on P variates for each of N individuals, it is based on an N X N association matrix and is therefore a Q rather than an R technique (Sokal & Sneath, 1963). The procedure gener- ates a set of vectors that are a linear combination of the individuals scores. These are the principal co-ordinates. If the first K principal co-ordinates give an adequate representation of the data in K dimensions, the remaining N-K can reason- ably be disregarded as uninformative. Abbreviations to text figures aa anterior adductor muscle apr anterior pedal retractor muscle by ‘byssal’ gland cg cerebral ganglion dd digestive duct dh dorsal hood di digestive diverticula ft foot ga gill axis gi gill gs gastric shield ht heart hy hypobranchial gland ky kidney ma mantle oe oesophagus Ov Ovary pa posterior adductor muscle pb palp proboscis pg pedal ganglion pp palp ppr posterior pedal retractor muscle ps posterior sorting area Si sorting ridges SS style sac st stomach ts testis ty typhlosole vg visceral ganglion TAXONOMY Early systematic studies on the Nuculidae (eg. Montagu, 1808; Reeve, 19885; Adams, 1856) were based on differences in shell morphology. A more-detailed attempt was made by Schenck, (1939) who also classified Recent and fossil species by shell characters alone, though he pointed out that it was impossible to say what characters should be accorded the greater weight, and whether emphasis should be placed on ‘soft or hard part’ morphology. Moore, (1931a, 1931b) distin- guished between shallow-water nuculid species on the basis of the pattern of grooves on the faecal pellets, however Allen, (1954) was to show that this was not an infallible method. Indeed, sections of the hindgut show that the pellets of deep-sea species have no systematic value usually being round or oval in cross-section and without distinctive grooves. Heath, (1937) described variations in the muscula- ture and the digestive system of various species and found that the configuration of the hindgut could be used as a consistent diagnostic feature. This remains true (Allen, 1978), however the hindgut of deep-sea nuculid species is usually smaller in diameter and far more extensively coiled as compared with shallow-water species, and to determine dif- ferences, delicate dissection is required, supported by exami- nation of small whole-mount specimens and _ sections. Furthermore, there appears to be a degree of intraspecific variation. The musculature of deep-sea species per se has little systematic value but differences in the orientation of the body and the size and shape of the foot are reflected in the deposition of the pedal musculature. Interspecific differences in the morphology of the stomach were also noted. Morphological orientation in bivalves is frequently related to a series of axes, (Allen, 1985). Definitions are by no means consistent, thus, Fischer, (1886) defined the antero-posterior axis as a straight line touching the lower margins of the adductor muscles, whereas, Jackson, (1890) considered it to pass through the mouth and the middle of the posterior adductor muscle. The cardinal or hinge axis as distinct from the body axis is most used as the basic criterion for orienta- tion, the latter being assessed by the change from parallel of the hinge and oro-anal axes (Kauffmann, 1969). Unfortu- nately, the hinge axis is not easily determined in the Nucul- idae. Differing orientations of the foot appear to be correlated with the degree of alignment of the adductor muscle axis as compared with the axis through cerebral and pedal ganglia (Fischer, 1886). Such axes cannot be deter- mined with mathematical precision and cannot be used in a strict taxonomic sense, nevertheless they show how interspe- cific transformations are related to various anatomical refer- ence points. The family is very conservative with regard to its body 64 anatomy. Thus, Schenck, (1944) working with Nucula tamat- avica and Heath, (1937) with N. nucleus found that these species could not be separated from others on the basis of their anatomy. Similarly Moore, (193la) and Allen, (1954) could not find any significant anatomical differences between the British species of Nucula. Shell morphology, therefore, remains of paramount importance to the taxonomy of the Nuculidae. On the basis of shell morphology, Schenck, (1934) split the nuculids into three main “Taxonomic Units’. (A) Forms with crenulate inner ventral margins. ( 1) Nucula Lamarck, (1799) ( 2) Pronucula Hedley, (1902) ( 3) Pectinucula Quenstedt, (1930) ( 4) Linucula Marwick, (1931) ( 5) Deminucula Iredale, (1931) ( 6) Lamellinucula Schenck, (1944) ( 7) Gibbonucula Eames, (1951) (B) Forms with smooth inner ventral margins. ( 1) Ptychostolis Tullberg, (1991) ( 2) Nuculoma Cossmann, (1907) ( 3) Nuculopsis Girty, (1911) ( 4) Nuculoidea Williams & Breger, (1916) ( 5) Nuculopsis Woodring, (1925) ( 6) Leionucula Quenstedt, (1930) ( 7) Palaeonucula Quenstedt, (1930) ( 8) Ennucula Iredale, (1931) ( 9) Brevinucula Thiele, (1934) (10) Austronucula Powell, (1939) (11) Trigonucula Ichikawa, (1949) (12) Habonucula Singh & Kanjilal, (1977) (13) Condylonucula Moore, (1977) (c) Forms with divaricate sculpture. ( 1) AcilaH. & A. Adams, (1958) ( 2) Truncacila Schenck, (1931) Thereafter, Cox, (1940) described species such as Nucula obliqua Lamarck, Nucula expansa Reeve and Nucula superba Hedley, as intermediate between ‘crenulate’ and ‘non- crenulate’ forms in that they possessed very fine crenulations only visible with microscopic aid. Vokes, (1949) who found a similar condition in the Palaeozoic species Nuculoidea opima Hall suggested that such species should be placed in a separate group. Many of the taxa listed above, are known only as fossils. Of these Thiele, (1934) was of the opinion that Ennucula and Leionucula are congeneric and Allen & Hannah, (1986) synonymized Linucula with Nucula, ajudged that Pronucula is a subgenus of Nucula and that Nuculopsis Woodring (non Girty), Leionucula, Ennucula and Austronucula are synony- mous with Nuculoma. On the basis of shell structure, Recent non-divaricate species have been divided into two genera, Nucula and Nuculoma (Poel, 1955). species in which the outer shell layer is constructed of radial elements and, consequently, have a crenulate inner ventral shell margin are included in the genus Nucula. Those with a uniform, non-radial, structure with a smooth inner ventral shell margin are included in Nuculoma. Taylor et al., (1969) describe the shell of Nucula as being composed of three layers of aragonite, the inner being nacrous. Unfortunately, Taylor et al., (1969) did not report on Nuculoma, however, we can confirm that Nuculoma P.M. RHIND AND J.A. ALLEN species described here have an inner nacrous layer. It has been assumed that species of the Nuculacea and Nuculanacea can be distinguished on the basis that nuculaceans have an inner nacreous layer, which is absent in nuculanceans, how- ever, Moore (1977) in his description of a tropical, shallow- water nuculacean Condylonucula states it is without an inner nacreous layer. Similarly Cox, (1959) states that early fossil members of the Nuculacea and Nuculanacea are not easily distinguished in that some Jurassic species of Nuculana have an inner nacreous layer. All deep-sea nuculaceans so far examined have an inner nacreous layer. CLASSIFICATION As a result of this study on the shell morphology and anatomy of the Nuculidae of the deep Atlantic and on the Recent and fossil material from the Natural History Museum, London (BM(NH)) and elsewhere, and examination of descriptions in the literature, a new listing of the extant genera of the family Nuculidae has been arrived at. Family NUCULIDAE Gray, 1824 Genus NUCULOIDEA Williams & Breger, 1916 TYPE SPECIES: Cucullea opima Hall, (1843) Shell ovate, inequilateral with concentric sculpture; faintly prosogyral; inner ventral margin not crenulate but with internal marginal micropectinations; resilifer perpendicular to hinge plate; inner nacreous layer present. Originally Williams & Breger, (1916) proposed the subge- neric name Nuculoidea for Palaeozoic species of Nucula having a non-crenulate inner ventral margin and with other characters intermediate between those of the extinct family Ctenodontidae and the Recent members of the family Nucul- idae. These latter include a cartilage pit as found in Cteno- donta albertina Ulrich (upper Ordovician). Williams & Breger, (1916) divided the species of Nuculoidea into three groups each based on a described species: (1) Nuculoidea opima (Hall) with umbones twisted to a vertical position or faintly prosogyrate, shell usually anteriorly elongate and semi-lunate; (2) Nuculoidea aquisgranensis (Beuschausen) with opisthogyrate umbones, anterior and convexly rounded and usually the larger, posterior outline semi-lunuliferous; (3) Nuculoidea niotica (Hall) with opisthgyrate umbones, posterior margin truncate and nearly vertical, anterior margin also nearly straight, producing a characteristic vertically triangular outline. Later, McAlester, (1962) upgraded Nucu- loidea to generic rank. Similarly, Soot-Ryen, (1964) and Liljedahl, (1983, 1984) in a systematic account of nuculoid pelecypods from the Silurian (Palaeozoic) formations of gotland described a number of nuculoids with characters common with the Nuculoidea opima group and raised these species to the generic rank. We are reluctant to create a new genus for Recent forms that match the description given above, particularly in regard to the faintly prosogyral nature of the beaks, the micropecti- nate margin and the vertical ligament. Furthermore, this relates to the fact that many deep-sea bivalves belong to groups with a long geological record. STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 65 Genus NUCULOMA Cossmann, 1907 TYPE SPECIES: Nucula castor d’Orbigny, (1850) Shell ovate or triangular, smooth or with concentric sculp- ture; opisthogyral; ventral margin not crenulate as seen externally but may be microscopically crenulate internally corresponding with marginal limits of radial shell elements; resilifer oblique to hinge plate; inner nacrous layer present. Cossmann, in Cossmann & Thiery, (1907) proposed the subgeneric name Nuculoma for Jurassic nuculids character- ized for the most part by a narrow oblique chondrophore. Later Schenck, (1934) raised Nuculoma to generic rank. Genus BREVINUCULA Thiele, 1934 TYPE SPECIES: Nucula guineensis Thiele, (1931) Shell deeply triangular with smooth surface; opisthogyral; inner ventral margin not crenulate; resilifer absent or very small, inner nacrous layer present. Thiele, (1934) originally proposed the subgeneric name Brevinucula for a deep-sea species Nucula guineensis (p. 83) from the Gulf of Guinea which was later synonymized with Nucula verrilli (Verrill & Bush, 1898). Genus CONDYLONUCULA Moore, 1977 TYPE SPECIES: ?Condylonucula cynthiae Moore, (1977) Shell very robust, ovate with concentric sculpture, heavily sculptured cap-like prodissoconch; opisthogyral; inner ven- tral margin not crenulate; resilifer perpendicular to hinge plate; inner nacreous layer absent. Species of Condylonucula are confined to shallow-water sediments of the tropical western Atlantic. Genus NUCULA Lamarck, 1799 TYPE SPECIES: Arca nucleus Linné, (1758) Shell ovate-triangular, inequilateral with reticulate sculpture; opisthogyral; inner ventral margin crenulate; hinge line angu- late; resilifer oblique to hinge plate; inner nacrous layer present. Genus DEMINUCULA Iredale, 1931 TYPE SPECIES: Nucula praetenta Iredale, (1931) Shell triangular with reticulate sculpture; opisthogyral; inner ventral margin crenulate; hinge line angulate; resilifer not well-defined, inner nacrous layer present. Schenck, (1934) questioned whether Deminucula should be included in the Nuculidae in that the original specimens (Smith, 1891) showed that they were without a ‘chondro- phore’. A possible explanation of the difference between Deminucula and Nucula is that there has been retention of juvenile shell characteristics into adulthood (neoteny) in Deminucula (p. 67). This is evident when D. atacellana is compared with the juveniles of a shallow water species such as Nucula sulcata. Genus PRONUCULA Hedley, 1902 TYPE SPECIES: Pronucula decorosa Hedley (1902) Shell triangular with reticulate sculpture; opisthogyral; inner ventral margin crenulate; hinge line arched; resilifer perpen- dicular to hinge plate, inner nacrous layer present. This genus differs from Nucula in having an arched instead of an angulate hinge line, the hinge teeth series do not overlap beneath the umbos, and the resilifer is perpendicular rather than oblique to the hinge plate. Since the description by Hedley, (1902), Cotton, (1930) and Marwick, (1931) and Clarke, (1961) have described several other Recent and Tertiary species from the southern hemisphere. Genus ACILA (Adams & Adams, 1858) TYPE SPECIES: Nucula divaricata Hinds, (1843) Shell ovate- triangular with divaricate sculpture; opisthogyral; hinge line arched; inner ventral margin crenulate; resilifer oblique to hinge plate; inner nacrous layer present. Adams & Adams, (1858) proposed the generic name Acila for nuculid species mainly characterized by a divaricate sculpture. Extant species are restricted to the Indo-Pacific and to relatively shallow depths (Schenck, 1934). ATLANTIC DEEP-SEA SPECIES OF THE FAMILY NUCULIDAE Deminucula atacellana (Schenck, 1939) Figs 2-4, 5a, 6, 7a TYPE LOCALITY: Porcupine St 16, off NW Coast of Ireland, 1476 and 1215 fms. Designated Schenk (1939). TYPE SPECIMEN: Holotype: not designated, Lectotype: USNM No. 197154 selected by Schenck, (1939) PI. 5, figs. 4, 5, 9, 10, 13, 16. ( 7 specimens from Atlantis II, Sta. 131, 39°38.5'N. 70°36.5'W, 2178m lodged in BM(NH) No. 1990010). Nucula reticulata Jeffreys, 1876, p. 429; 1879, p. 583; (non Hinds, 1843); Smith, 1885, p. 229. Nucula cancellata Jeffreys, 1881, p. 951; (non Meek & Hayden, 1856, p. 85); Verrill, 1884, p. 280, 285; Dall, 1890, p. 258; Dautzenberg & Fischer, 1897, p. 203, 204; Verrill & Bush, 1897, p. 854, pl. 81, Fig. 3, pl. 86, Fig. 5; Johnson, 1934, p. 15; Dautzenberg, 1927, p. 288. Nucula (Nucula) atacellana Schenck, 1939, p. 27, pl. 5, Figs 4,5, 8, 10, 13, 16. DEPTH RANGE: 1102-4938 metres. MATERIAL Cruise Sta Depth No Lat Long Gear Date (m) NORTH AMERICA BASIN Atlantis II 62 2496 82 39°16.0'N 70°33.0'W ET 21.08.64 64 2886 1 38°46.0’N 70°06.0'W ET 21.08.64 73 1470 361 39°46.5'N 70°43.3’W ET 25.08.64 Chain 76 2862 1 39°38.3’N 67°57.8’W ET 29.06.65 77 3806 54 38°00.7'N 60°16.0'W ET 30.06.65 78 3828 12 38°00.8’N 69°18.7'W ET 30.06.65 85 3834 69 37°59.2'N 60°26.2'W ET 5.07.65 87 1102 48 39°48.7'N 70°40.8’W ET 06.07.65 87 1102 48 39°48.7'N 70°40.8'W ET 06.07.65 66 Cruise Sta DepthNo Lat Long Gear Date (m) Atlantis II 115 2051 2131 39°39.2'N 70°24.5'W ET 16.08.66 Atlantis II 126 3806 3 39°37.0'N 66°47.0'W ET 24.08.66 128 1254 12 39°46.5'N 70°45.2’W ES _ 16.12.66 131 2178 1022 39°38.5'N 70°36:5'W! ES 18.12.66 Chain 210 2064 409 39°43.2'N 70°49.5'W ES — 22.02.69 335 3882 3 40°23.3’'N 46°30.0'W ES = 31.08.72 Knorr 340 3264 25 38°14.4’'N 70°20.3'W ES — 24.11.73 WEST EUROPEAN BASIN Chain 313 1500 155) S1232°20N\ _12235.9!W) BS) 1 17/'08:72 316 2209 531 50°57.7'N 13°01.3'W ES 18.08.72 318 2506 7 50°27.3'N, 13°20:9'W ES 19!08°72 321 2890 4 50°12.3’N 13°35.8'W ES — 20.08.72 Incal CP08 2644 1 50°14.7'N 13°13.5'W CP 27.07.76 ROCKALL TROUGH Challenger 2 2857 2 55°04.0'N 12°33.0'W ES 04.06.73 3 1997 1 56°46.0'N 10°02.0’'W ABD 05.06.73 4 1993 24 56°52.0'N 10°01.0'W ES _ 05.06.73 6 2900 547 55°02.0'N 12°29.0'W ES _ 02.07.73 8 2900 5 54°45.0’N 12°10.0'W ES — 03.07.73 10 2540 147 56°37.0’N 11°04.0'W ES — 04.07.73 12 2076 6 56°49.0'N 10°15.0'W ES — 20.09.73 13 1842 8 56°45.0'N 09°50.0'W ABD 22.09.73 15 1632 13 56°44.0’N 09°28.0'W ES = 22.09.73 18 1392 16 56°44.0’N 09°20.0'W ES = 22.09.73 27 2880 456 54°40.0’N 12°16.0'N ES 03.11.73 34 2515 536 56°36.0'N 11°30.0'N ES 10.05.75 12°04.0'W SBC 07.09.75 12°05.0'W SBC _ 07.09.75 12°03.0'W SBC _ 07.09.75 48 2875 1 55°04.0'N 49 2875 1 55°03.0'N S15 2875 1 55°03.0'N 55 2878 52 54°40.0’N 12°16.0'W ES 17.11.75 56 2886 105 54°40.0’N 12°16.0'W ES _ 01.03.76 57 2950 67 54°41.0’N 12°23.0'W ES 21.06.76 12°17.0'W SBC 21.06.76 12°20.0'W ES _ 21.06.76 09°49.0'W SBC 25.06.76 58 2900 2 54°41.0'N 59 2900 96 54°40.0’N 63 1800 1 56°37.0’N 69 1050 24 59°39.0'N 07°12.0'W ES 02.07.76 110 2886 92 54°41.0’N 12°14.0'W ES — 22.10.76 111 2886 104 54°40.0’N 12°16.0'W ES 22.10.76 118 2910 68 54°39.0'N 12°14.0'W ES — 28.01.77 121 2910 1 54°37.0'N 12°09.0'W AT 29.01.77 124 2900 184 53°30.0’N 13°15.0’'W AT — 30.01.77 135 2900 158 54°39.0’N 12°16.0'W ES 07.08.77 137 2900 69 54°34.0'N 12°19.0'W ES — 22.02.78 174 2885 1 54°44.0'N 12°18.0'W SBC 22.05.80 176 2245 168 57°15.0’N 10°26.0'W ES — 28.05.80 185 2907 32 54°44.0'N 12°15.0'W ES 10.04.81 231 2898 69 54°42.0'N 12°12.0'W ES 17.05.83 Incal DSO1 2091 3 57°5S9.7'N = 10°39.8’W SD 15.07.76 DS02 2081 1 57°58.8’N 10°48.5'W SD _ 16.07.76 CP03 2466 4 55°38.0'N 11°64.4’.W CP 17.07.76 CP04 2483 3 5693-2 Nia S/W CR 107-76 DS06 2491 24 56°26.6'N 11°10.5'W DS 18.07.76 DS07 2884 37 55°00.0'N 12°31.0'W SD _ 19.07.76 CPO0S 2884 4 55°00.0'N 12°29.0'W CP 19.07.76 CP06 2888 4 55°02.3'N 12°40.3'W CP 19.07.76 CP07 2895 4 55°03.4’N = 12°49.2"W CP 20.07.76 DS09 2897 625 Sd:07-7 IN WIZS2'6'W SD 20107576 CP08 2644 ff S0ALTNG 1SA86 WwW CR. 27 07676 BAY OF BISCAY Sarsia S44 1739 14 43°40.8'N 03°35.2".W ES 16.07.67 $50 2379 1 43°46.7'N 03°47.8'W ES 18.07.67 $65. 1922 3 46°15.0'N 04°50.0’N ES — 25.07.67 Biogas II DS32 2138 2 WATSZ.25N, “O8,05:3'W SD 1804573 Biogas III DS37 2110 1 47°31.8'N 08°34.6'W SD _ — 24.08.73 DS41 3548 1 47°28.3'N 09°07.2'W SD — 26.08.73 DS49 1845 29 44°05.9'N 04°15.6';W SD __ 01.09.73 P.M. RHIND AND J.A. ALLEN DSS50 2124 1 44°08.9'N 04°15.9'W SD _ 01.09.73 Biogas IV DSS1 2430 5 44°11.3'N 04°15.4°;W SD 18.02.74 DS52 2006 30 44°06.3'N 04°22.4’.W SD _ 18.02.74 DS62 2175 1 47°32.8'N 08°40.0'W SD _ — 26.02.74 DS63 2126 1 47°32.8'N 08°35.0'W SD — 26.02.74 DS64 2156 3 47°29.2'N 08°30.77;W SD — 26.02.74 Biogas V CP07 2170 7 44°09.8'N 04°16.4°;W CP 21.06.74 Biogas VI DS71 2194 3 47°34.3'N 08°33.8'W SD 20.02.74 DS86 1950 105 44°04.8'N 04°18.7'W SD — 31.10.74 DS87 1913 79 44°05.2'N 04°19.4"°W SD _ 01.11.74 DS88 1894 7 44°05.2'N 04°15.7,;W SD _ 01.11.74 Polygas DS18 2138 1 47°32.2'N 08°44.9'W SD 22.10.72 DS25 2096 5 44°08.2'N 04°15.7,;W SD _ 01.11.72 DS26 2076 12 44°08.2’N 04°15.0'W SD _ 01.11.72 GUYANA BASIN Knorr 293 1518 9 08°58.0’N 54°04.3'W ES — 27.02.72 299 2076 2 O7°5S5.1’N 55°42.0'W ES 29.02.72 301. 2500 12 08°12.4’N 55°50.2'W ES — 29.02.72 303 2953 7 08°28.8'N 56°04.5'W ES _ 01.03.72 CANARIES BASIN Discovery 6697 1564 3 27°57.0'N 13°46.0'W ED 15.03.68 6701 1934 6 27°45.2'N 14°13.0'W ED 16.03.68 6704 2129 14 27°44.9'N__ 14°25.0'W_ ED 17.03.68 CAPE VERDE BASIN Atlantis II 141 2131 3 10°30.0'N 17°51.5'W ES 05.12.67 142 1796 48 10°30.0'N 17°51.5'W ES 05.12.67 148 3828 1 10°37.0’N 18°14.0'W ES 07.12.67 149 3861 2 10°30.0'N 18°18.0'W ES 07.02.67 GUINEA BASIN Walda DS20 2514 1 02°32.0'S O08°18.1'E SD - DS29 3547 2 02°57.0'S 04°28.1'E SD = ANGOLA BASIN Atlantis II 195 3797 20 14°50.0'S 09°54.0'E ES — 19.05.68 196 4630 6 10°29.0'S 09°04.0'E ES 21.05.68 197 4595 34 10°04.0'S 09°04.0'E ES 21.05.68 198 4566 16 09°47.0'S 10°29.0'E ES 21.05.68 200 2754 1 09°43.5'S 10°57.0'E ES — 22.06.68 201 2031 8 09°05.0'S 12°17.0'E ES 23.05.68 ARGENTINE BASIN Atlantis II 243 3822, 213 37°36.81S) 522336 WE Smee 4203571 245 2707 77 =36°55.7'S_ — 53°01.4°W_ ES 14.03.71 256 3917 356 37°40.8’S 52°19.3’W ES 24.03.71 259 3305 862 37°13.3'S 52°45.0'W ES 26.03.71 SHELL DESCRIPTION The original description by Jeffreys, (1876) is detailed and accurate and only a summary description is given here. Shell ovate, ventricose, lustrous, transluscent, equivalve, inequilaterial; umbo posterior to midline, slightly opisthogy- rate or orthogyrate; when postero-dorsal margin is orientated vertically the antero-dorsal margin is above horizontal; lunule and escutcheon not obvious, hinge teeth often visible through thin dorsal area; surface with fine concentric ridges which disappear towards the umbo, radial striations give impression of reticulation; chondrophore absent, resilifer inconspicuous, not oblique; hinge plate moderately broad; teeth chevron- shaped, the number varies with the size of the specimen (Fig. 2). Thus a shell 3.86 mm in length has 8 teeth in the anterior series and 5 in the posterior and for shells of 2.64 mm and 1.46 mm in length the numbers of teeth are 6/4 and 4/2 respectively. There appears to be a continuous gradation in the external characters of Deminucula atacellana and of Nucula callicre- STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 67 Fig. 2} Deminucula atacellana. Internal view of the left valve and hinge region of the right valve of a shell from Station 131. (Scale = 1.0 mm). demna (Fig. 7b) (p. 69). While Dall, (1890) regarded N. callicredemna as belonging to the group which includes D. atacellana, both Schenck, (1939) and Knudsen, (1970) believed the species to be conspecific with D. atacellana. In contrast James, (1972) states that there is a change in the shape of the outline of N. callicredemna with age with the shell becoming more ovate. This allomorphic variation taken with the fact that adult populations of D. atacellana also show a large amount of variation in shape is probably the reason why authors have debated the species. In fact, some mature specimens of D. atacellana appear identical to small speci- mens of D. callicredemna however, small immature speci- mens of N. callicredemna do not look like large mature specimens of N. callicredemna nor of D. atacellana of similar length. The small immature N. callicredema are far more triangular in shape (James, 1972) (Fig. 7). Mature specimens of N. callicredemna and D. atacellana can be clearly sepa- pa Fig. 3 Deminucula atacellana. The gross anatomy as seen from the right side of a whole mount with the shell removed. (Scale - 1.0 mm; for key to abbreviations see p.63). Fig. 4 Deminucula atacellana. Enlarged lateral view as seen from the right side of the gill and surrounding organs. (Scale = 0.2 mm; for key to abbreviations see p.63). rated on the basis of the internal shell morphology, and apart from the features outlined in the shell descriptions of the two species, they also differ in that mature specimens of D. atacellana are typically deminuculoid and without a well- defined resilifer (Fig. 4), whereas mature specimens of N. callicredemna have a well-developed chondrophore (Fig. 8). N. callicredemna grows to a much larger size than does D. atacellana. ff Uy, Fig. 5 A. Deminucula atacellana. Semidiagrammatic view of a transverse section immediately anterior to the posterior adductor; B. Nucula delphinodonta inner and outer gill plates as drawn by Drew 1901. (Scale = 0.5 mm; for key abbreviations see p.63). 68 INTERNAL MORPHOLOGY D. atacellana being the most common abyssal nuculid, is used to provide a basic account of morphology against which the other species can be described and compared. Particular emphasis is placed on features that show adaptation to its deep-water existence. The mantle edge of D. atacellana is relatively unmodified and typically consists of an inner muscular lobe, a middle sensory lobe and an outer lobe which secretes the outer layer of the shell. The periostracum arises from the junction of the outer and middle layer and there is a broad zone of attach- ment of the pallial muscle. Trichrome staining shows num- bers of gland cells scattered throughout the outer and inner epithelium of the mantle. Each labial palp is made up of a proboscis, pouch and paired lamellae (Figs 3a & 4). They are not greatly larger in overall size to those of more shallow-water nuculid species (Allen, 1978) and their structure compares well with the description given by Hiraska, (1927) for Nucula nucleus. The inner surfaces are covered with a layer of columnar epithelial cells with long cilia and with a cytoplasm which is strongly acidophilic. The epithelial cells of the outer surfaces are squamous. The outer surfaces are smooth whilst the inner surfaces are ridged dorso-ventrally. The fine structure of the ridges compares well with that of N. nucleus. The ridges get smaller towards the anterior part of the palp and their structure in cross-section changes slightly. The number of ridges varies and ranges from 12 in specimens ca. 1 mm total length to 27 in specimens ca. 3 mm total length. The palp proboscis extending from the posterior dorsal limit of the palps is muscular and contractile. The form in life is difficult to determine for in preserved specimens they are always in an extreme state of contraction (Fig. 4). Neverthe- less, they have a similar morphology to that of Nucula B aa P.M. RHIND AND J.A. ALLEN delphinodonta (Drew, 1901), and Nucula nucleus (Hirasaka, 1927). Morphologically they are an extension of the distal oral groove and a shallow V-shape in section. The concave surface and its margins consist of a single layer of columnar epithelial cells with long cilia. The convex surface has a single layer of more cuboidal epithelial cells whch appear to be devoid of cilia. Trichrome staining shows large numbers of mucous secreting cells, especially in the concave surface. Each proboscis has well-developed longitudinal muscle fibres amongst connective tissue. The fibres are continuous with the musculature of the body wall and are asymmetrically distrib- uted so that on contraction the proboscis curls towards the concave surface. Each proboscis is supplied with a large nerve that originates in the cerebral ganglion and runs within the dorsal margin of the palps. There is also a large blood space. Palp pouches are features only found in the Nuculoidea. They are spoon-shaped and attached to the body at the point of attachment of the proboscis to the palp (Fig. 4). They estab- lish continuity between the point of food collection and the distal oral groove of the palp. They have a simple structure consisting of two layers of cuboidal epithelium, separated by a thin layer of connective tissue. Within the latter is a material with similar staining properties to chitin. This mate- rial is also found in the gill plates where it forms the skeletal rods. Short cilia are present on the concave surface and margins of the pouch but not on the convex surface. The gills are suspended obliquely from the postero-dorsal edge of the viscero-pedal mass across the posterior mantle cavity adjacent to the ventral margin of the posterior adduc- tor muscle (Figs 3 & 4). The number of gill plates on the axis varies with the size of the specimen and ranges from 5 pairs in specimens ca. 1 mm in length to 24 pairs in mature specimens ca. 3 mm in length. Drew, (1901) found that 20 pairs was the common number in full grown specimens of the shallow- Fig. 6 Deminucula atacellana. A. Dorsal view of hind gut in situ; B. Course of hind gut as seen from right side—elucidated by dissection; C. Posterior view of stomach and hind gut—note a section of gut immediately posterior to the stomach is missing; D & E. Intact stomach as seen from the left and right side respectively. (Scale = 0.1 mm; for key abbreviations see p.63). STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 69 Fig. 7 A. Deminucula atacellana and B. Nucula callicredemna. A comparison of the external features of the two species as seen from right lateral and dorsal views. (Scales = 1.0 mm). water species Nucula delphinodonta Mighels & Adams, 1842. The gill plates of D. atacellana have a different shape to those of Nucula delphinodonta (Figs 5a & b). In the latter case the plates are more triangular in shape. The finger-like plates of D. atacellana appear to be similar to those of Nucula prox- ima, another shallow-water species (Hampson, 1971), but when surface areas are compared, that of D. atacellana is less than half that of N. proxima. This may relate to the fact that deep-sea species have different respiratory requirements, but may also be a reflection of the small quantities of suspended material in deep water (Allen, 1978). Shallow-water species (eg. N. nucleus) may be facultative filter feeders (Caspers, 1940). The distal ends of the inner plates of each gill meet in the midline posterior to the foot. The gills of D. atacellana are supported by a skeletal framework, trichrome stain indicates that chitin occurs in the gill axis and in the ventral part of the gill plates. A few muscle fibres are present in the axis. The nervous system of D. atacellana appears to be similar to that of N. delphinodonta (Drew, 1901). Cerebral ganglia are located on each side of the oesophagus and are connected by a short commissure that passes between the outer wall of the oesophagus and the anterior adductor muscle. From each cerebral ganglion there are two posteriorly directed commis- sures. The more dorsal passes through the viscera and joins with an elongate visceral ganglion anterior to the posterior adductor muscle. The ventral commissures extend into the foot to connect with the pedal ganglia. Adjacent and postero- dorsal to the pedal ganglia are a pair of statocysts which contain crystalline fragments, the statoliths. Cephalic sense organs (Vlés, 1905) are present on either side of the mouth close to the proximal palp ridge and immediately ventral to the cerebral ganglia and are innervated from the latter. In N. Nucleus (Hirasaka, 1927) the cephalic sense organs comprise pigment cells with a cornea, in contrast to D. atacellana where there is a thickened epithelium which is innervated from the cephalic ganglia, but with no evidence of differentiation into cornea and pigment cells. The hypobranchial gland (Pelseneer, 1891), consists of a layer of swollen columnar cells covering the surface of the suprabranchial cavity and the outer surface of the membrane suspending the gill axis (Fig. 5). The gland in D. atacellana is similar to that described for Nucula delphinodonta (Drew, 1901) and Nucula turgida (Yonge, 1939). According to Yonge, (1939) the function of the gland is to produce a secretion that binds any fine material which passes between the gill filaments into the suprabranchial cavity. In Nucula delphinodonta it has the additional function of forming a brood sac (Drew, 1901). Heath, (1937) believed that the gland produced different types of secretion in different spe- cies. In some species he found the gland stained deeply with Delafield’s haematoxylin, whilst in others it remained unstained. In the case of D. atacellana different staining characteristics were found in different specimens. Sometimes it stained only with Orange G, sometimes with Aniline Blue, sometimes in part with Orange G and in part with Aniline Blue, and sometimes it would not take up any stain and remained colourless. The reasons for the apparent changing chemistry of this gland are unclear. Similarly the function of the gland is yet to be determined. The foot has a similar shape and proportions to that of shallow-water species (Fig. 3). The sole is divided sagitally and there is a prominent heel. The margins of the sole are fringed with papillae. The number of papillae (c. 10) does not appear to vary with the size of the animal. A large, dorso- ventrally elongate, ‘byssal’ gland is present in the heel of the foot and from it a short duct opens at the posterior limit of the sole. The pedal muscles of N. delphinodonta (Drew, 1901) are arranged in three pairs, two of which are attached antero- dorsally and the third postero-dorsally to the shell. The arrangement of D. atacellana differs from this and corre- sponds more closely with the arrangement in Acila divaricata (Heath, 1937). Thus, there is only one pair of anterior muscles lying immediately posterior to the oesophagus. Twin medial pedal muscles attach to each valve and then merge and pass to the outside of the stomach to serve the central 70 region of the foot. There is a pair of posterior pedal muscles which pass to either side of the hindgut. In Acila divaricata the anterior pair are the largest, (Heath, 1937). This is not the case in D. atacellana, where all are approximately equal in size. The mechanics of foot movement in the protobranchs was described by Driscoll, (1964). The median and posterior pedal muscles are retractors, the anterior pair which control the anterior movement of the foot are protractors. Heath, (1937) believed that the posterior part of the anterior muscles also acted as a retractor. The ‘byssal’ gland in the heel of the foot of protobranchs opens to the median groove of the foot, however, nothing comparable to byssus threads have been observed. Its structure differs greatly from the byssus gland of lamellibranchs and there is no evidence that it is a vestige of a post-larval gland (Heath, 1937). In the protobranchs it varies greatly in size in different species although it is fairly consis- tent in its composite structure. In D. atacellana it is compara- tively large. The lumen of the gland may contain traces of a secretory material. The surrounding cells are usually swollen and vacuolated. In general the form of the gut of D. atacellana compares well with descriptions given by Yonge, (1939) and Owen, (1956) for shallow-water species. The mouth is situated mid-ventrally a short distance behind the anterior adductor muscle (Fig. 3). At the point of its junction with the distal oral grooves it is laterally distended and bilobed in cross- section. The oesophagus enters the stomach antero-dorsally and a little to the left of the mid-line. This is in contrast to Nucula sulcata where the oesophagus enters the stomach to the right of the mid-line (Owen, 1956), however, in D. atacellana most of the right side of the body is occupied by the extremely elongate hindgut (see below). The oesophageal epithelium is much folded and is made up of a mixture of cuboid and columnar cells. The epithelial cells are richly ciliated and there are mucous cells present between them. There is a layer of subepithelial circular muscle that is progressively thicker towards the stomach, so much so that it probably forms a sphincter at the junction with the stomach. Functionally this is essential to prevent regurgitation of the stomach contents when under pressure. The combined stomach and style sac consist of a dorsal globular region—the stomach proper—and a ventral conical tapering style sac (Fig. 6). The dorsal hood (‘dorsal pouch’ of Graham, 1949) extends along the left wall of the stomach to end blindly on the left dorsal side. As in most protobranchs three ducts enter the stomach anteriorly from the digestive diverticula. In D. atacellana there is one on the left and two adjacent to each other close to the mid-line and ventral to the oesophagus. There is no small caecum similar to that found on the right of the mid-dorsal line in the stomach of N. sulcata (Owen, 1956). A grooved ciliated sorting area which extends over most of the right side of the stomach is visible through the stomach wall. In a specimen approximately 3 mm long seven grooves are present. This compares with more than 25 in a specimen of N. sulcata 15 mm in length. A reduction in the number of the ciliated grooves is a consistent feature in deep-sea species. The left wall of the stomach is lined with a gastric shield and in the region of the dorsal hood this is thickened to form a pronounced tooth. Previously regarded simply as an inert structure serving to protect the underlying epithelium and assist in the trituration of the gastric contents, Halton & Owen, (1968) showed it has a far more elaborate structure and function. There are two components, an inner part consisting of numerous microvilli of underlying epithelial P.M. RHIND AND J.A. ALLEN cells which extend into the secreted matrix and, an outer matrix of non-cellular chitin. At intervals on certain microvilli vescicular swellings are present. Histochemical studies show the shield to be enzymatically active. The epithelium underly- ing the shield is of columnar cells which are tallest in the region of the gastric tooth. Distally these cells are packed with pigment granules. Yonge, (1939) and Owen, (1956) suggest that they are probably excretory. Histochemical studies suggest they are related to lipofuschin pigments which are generally regarded as by-products of lipid metabolism. Underlying the epithelium of the stomach is a layer of collagen and a system of muscle fibres. The hindgut is extremely long. This is characteristic of deep-sea nuculids and it is particularly elongate in D. atacel- lana (fig. 6). This may relate to the fact that deep-sea sediments make a poor diet with many of the organics resistant to digestion and therefore demanding of time for digestion, however, the discovery of complete forameniferans in the stomach perhaps indicate more living matter is ingested than was previously thought. Heath, (1937) used the configuration of the hindgut as a diagnostic feature and found the simplest and possibly the most-primitive nuculoid configuration is that found in species of Acila. The more-complex extensive overlapping of the coils in D. atacellana makes the course of the gut difficult to determine except by three dimensional reconstruction (Fig. 6). The digestive diverticula of D. atacellana consist of numer- ous blind-ending tubules which communicate with the stom- ach by a system of unbranched secondary ducts and much- branched main ducts. The main ducts are lined with cuboidal cells with a spherical or slightly oval nucleus and numerous pigmented cytoplasmic granules. Owen, (1956) in studies on living N. sulcata found these latter to be lipo-pigments staining strongly with Sudan Black B, suggesting that they were a bi-product of lipid metabolism and, from frozen sections, he confirmed the presence of lipids in the basal region. The cells are without cilia but have a well-developed brush border. Secondary ducts in D. atacellana are similar to main ducts. This is in contrast to N. sulcata where secondary ducts have a dense coat of long cilia which extend almost to the centre of the lumen. While individual tubules of N. sulcata are round or oval in cross-section (Owen, 1956), those in D. atacellana are much more irregular. This appears to be a consistent feature in deep-sea protobranchs. Two types of cell occur. The most numerous have an irregular outline and are highly vacuolated. They contain two types of cytoplasmic granule, one large and blue staining and the other small and usually red in trichrome staining. The blue colour indicates an external membrane of basophilic mucus, the red colour indicates an acid mucopolysaccharide. A second, much less common, type of cell is present in small clusters scattered within the tubules. They are distinguishable by a lack of granules, a more-prominent nucleus and a more-darkly stain- ing cytoplasm. They tend to be triangular in outline and distally narrow. According to Owen, (1956), they possess a single flagellum but this was not seen in D. atacellana. The dark staining cells have been regarded as excretory and secretory, but it remains unclear what function they perform (Owen, 1956). No evidence of intra-cellular digestion was seen and it is generally agreed that digestion in protobranchs is exclusively extra-cellular. Studies of D. atacellana by Sanders and Hessler, (1969), STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 7 Scheltema, (1972) and Grassle & Sanders, (1973) from the North America Basin suggests this species has non-periodic reproduction. Preliminary evidence from samples taken in February, June, October and November from Biscay sup- ports this view. Sexes are separate and of equal proportions. Gonads are simple, paired, and overlie the viscera. The gonoducts open into the supra-branchial cavity. Maximum diameter of the ova is 135 um. Approximately 200 ova are present in a specimen 3.3 mm total length. No brooding of the eggs occurs, development is probably lecithotrophic (Scheltema, 1972). COMPARATIVE SHELL MEASUREMENTS OF SPECIMENS OF D. atacellana Dimension (mm) No. Mean SD Max. Min. NORTH AMERICA BASIN Length 45 3.07 0.88 5.04 1.51 Width 1.88 0.57 3:23 0.94 Height ail 0.79 4.55 1.34 WEST EUROPEAN BASIN Length vi 2.96 1.14 4.87 0.78 Width 1.85 0.75 2.89 0.48 Height 2.58 0.96 4.02 0.76 BAY OF BISCAY Length 134 2.68 0.84 4.53 0.69 Width 1.67 0.56 2.96 0.45 Height 2.34 0.72 3/5 0.68 S.E. ATLANTIC Length 58 2.46 C59 5.54 0.84 Width 1.42 0.93 3.33 0.45 Height DAG 1.37 4.77 0.76 GUYANA BASIN Length 17 3.67 1.74 6.50 1.34 Width 2.04 0.99 3.75 0.75 Height 3.13 1.49 5.50 1.16 CANARIES BASIN Length 14 7a 0.42 2.78 1.37 Height 1.05 0.24 1.64 0.83 Height 1.56 0.34 2.41 1.23 CAPE VERDE BASIN Length 7 2.26 0.73 3.49 1.46 Width 1.30 0.44 2.10 0.88 Height 1.95 0.60 2.92 1.29 ARGENTINE BASIN Length 23 3.50 1.10 5.20 1.14 Width 1.96 0.64 3d 0.66 Height 3.18 1.00 4.74 1.05 DISTRIBUTION D. atacellana is common and widespread. The species was originally described from specimens taken from the West European Basin by Jeffreys, (1881). It also was recorded from the North America Basin (Verrill, 1884), the Surinam Basin (Dall, 1890) and the Canaries Basin (Dautzenberg and Fischer, 1897). The species occurs throughout the North Atlantic, as far north as the Rockall Trough in the east and the Davis Strait in the west. In the South Atlantic it occurs in the Guinea, Angola and Argentine Basins but to date has not been taken from the Brazil Basin or the Sierra Leone Basin nor is it recorded by Clarke, (1961) in the Cape Basin. It appears to be absent from high latitudes—the Antarctic (Thiele, 1912; Dell, 1972), Arctic (Clarke, 1960; Bernard, 1979) and Norwegian Basins (Bouchet & Warén, 1979). D. atacellana is the most eurybathic of the deep-sea nucu- lids with a vertical range of approximately 4000 m. It is not found in depths less than 1000 m. Shallow-water nuculids appear not to occur much beyond the continental shelf edge eg. Thus for European seas with the possible exception of Nuculoma tenuis and Nucula tumidula Malm shelf species which penetrate down slope to some degree, there appears to be little overlap of shelf and slope species at the extremities of their ranges. More sampling will be necessary to confirm this but there appears to be a gap between 200 m and 500 m in which few species of Nucula dominate. D. atacellana consis- tently occurs deeper in the Basins south of the equator than in the north. This may possibly relate to the absence of Brevinu- cula verrilli (p. 85) from the southern Basins. Nucula callicredemna Dall, 1890 Figs 7b & 8 TYPE LOCALITY: Albatross Sta. 2754, Northeast of Tobago, Lat. 11°40'N., Long. 58°33'W, 1609 m. TYPE SPECIMEN: Holotype. No holotype was designated. The type lot (USNM-95431) contained 11 valves, 1 conjoined specimen and several fragments (James, 1972). Re-examined (1991, JAA) 6 valves (4 in poor condition) and one conjoined specimen and fragments. Conjoined specimen selected as lectotype. Nucula callicredemna Dall 1890, p. 258, pl. 31, Fig. 9; James 1972, p. 29, 34-36, Figs 5-14, map 2. Nucula crenulata var. obliterata Dall 1886 p. 247, (in part); (non N. crenulata var. obliterata Dall, 1881). Nucula obliterata Knudsen, 1970, p. 177; Pequegnat, 1972, p. 74. Nucula aureliae Métivier, 1982, p. 39-42, Fig. 1, pl. 1, Figs 1-2. DEPTH RANGE: 3411-4077 metres. The taxonomy of this species is complicated by the fact that Dall, (1886) incorrectly identified a specimen of N. callicre- demna (USNM. 63133) from ‘Blake’ Sta. 236 as Nucula crenulata var. obliterata (Knudsen, 1970). It is also possible that certain specimens in past records reported as N. cancel- lata (= D. atacellana) could be small specimens of N. callicredemna (p. 66). We believe that Nucula aureliae which Métivier, (1982) has described from 3360 metres off the Azores Archipeligo is synonymous with N. callicredemna and, if true, its range is now extended to the Canaries Basin. Métivier, (1982) who regarded N. aureliae as being similar to N. cancellata (= D. atacellana) failed to mention N. callicre- demna. N. callicredemna also appears to be identical, cer- tainly in shell morphology, to the Arctic deep-sea species Nucula zophos (Clarke, 1960), although this cannot be cer- tain until the internal morphologies have been compared. Dall, (1890) also regarded N. callicredemna as being similar to the Pacific deep-sea species Nucula niponica (Smith, 1885). 72 N. callicredemna originally taken from off Tobago in 1609 metres (Dall, 1890) is now known to be common in the Gulf of Mexico in depths ranging from 2122 metres to 3563 metres (James, 1972). The one specimen taken off Recife in the Brazil Basin now extends the range of the species approxi- mately 2180 km south. MATERIAL Cruise Sta. DepthNo Lat Long Gear Date (m) BRAZIL BASIN Atlantis IT 156 3459 1 00°46.0’'S —29°28.0’'W_ ES 14.02.67 GUYANA BASIN Knorr 306 3392 4 09°31.1'N 56°20.6’'W ES 02.03.72 VENEZUELA BASIN Norda 33 =3800- +1 15°08.0’N 69°12.0'W OT 24.10.81 4035 35 3954- 16 15°08.0’N 69°12.0'W OT 25.10.81 4017 36. 3951— 24 15°08.0'N 69°12.0'W OT 26.10.82 4044 37 =3995- 11 15°08.0'N 69°12.0'W OT 27.10.81 4097 38 3995- 15 15°08.0’N 69°12.0'W OT 28.10.81 4013 39 = 3993-1 15°08.0'N 69°12.0'W OT 28.10.81 4065 40 3967- 11 15°08.0’N 69°12.0'W OT — 28.10.81 4009 87 3482- 22 13°30.0’N 64°45.0’'W OT 25.11.81 3518 88 3516— 25 13°30.0'N 64°45.0'W OT 25- 3550 26.11.81 90 3422- 19 13°30.0'N 64°45.0'W OT 26 3464 27.11.81 91 3459- 12 13°30.0'N 64°45.0'W OT 27.11.81 3503 92 3476- 34 13°30.0'N 64°45.0'W OT 27- 3518 28.11.81 93 3411- 21 13°30.0'N 64°45.0'W OT 28.11.81 3459 94 3428- 11 13°30.0’N 64°45.0’'W OT 28 3476 29.11.81 SHELL DESCRIPTION The original shell description by Dall, (1890) is detailed, although as pointed out by James, (1972) it is in reverse ie. in the description anterior is posterior and vice versa. A sum- mary description is given here. Shell large, ovoid, lustrous, equivalve, inequilateral, sur- face covered in fine concentric ridges, radial striations give surface a fine reticulated appearance; umbos (frequently eroded) posterior to the midline, opisthogyrate; when poste- rior margin orientated vertically the anterior dorsal margin is above the horizontal; lunule long, escutcheon short and broad, dentition often visible dorsally through the shell, especially beneath the lunule; resilifer oblique to narrow hinge plate, extends ventral to anterior proximal teeth, teeth typically chevron-shaped but obtusely angled, number varies with the size of the specimen (6 posterior and 8 anterior teeth in a specimen 3.1 mm in length and 15 posterior and 29 anterior teeth in a specimen 18.9 mm in length (James, 1972), ligament tear-drop shaped. P.M. RHIND AND J.A. ALLEN Nucula callicredemna: MEASUREMENTS (FROM JAMES, 1972) Sta. Vv L H B A P 6813-9 l 18.9 14.0 3.6 29 15 j [ 18.9 14.0 3.5 30 14 6787-4A ] 12.2 9.4 3.2 21 11 ‘ r WA? 9.4 3.2 21 11 QV. Well Vp 19 10 : it 2 Tell DD, 19 10 65A14-6 ] 6.1 Sy? 1.4 14 7 ; r 6.1 S22 1.4 13 7 ] 3.1 3.0 0.9 10 6 it 3h 3.0 0.9 8 6 V = valve, r = right valve, 1 = left valve, L = length (mm), H = height (mm), B = breadth (mm), A = number of anterior teeth, P = number of posterior teeth. Characteristically, N. callicredemna has a well-developed chondrophore that extends ventral to the anterior proximal teeth and is similar to that in shallow-water species such as Nucula sulcata (Fig. 8). It has similar proportions to that of many shelf species but the large size of the shell (19 mm total length), is unusual in a deep-sea species (James, 1972). It is by far the largest deep-sea nuculid in the Atlantic. INTERNAL MORPHOLOGY Discounting differences related to size the internal morphol- ogy is not greatly different from that of D. atacellana. The palp is large and deep covering much of the foot and in a specimen 16 mm total length there are 65-70 palp ridges. The gill is narrow lying diagonally across the posterior part of the mantle cavity. The axis is parallel to the postero-dorsal shell margin. The same specimen bears 32 gill plates. The course of the hind gut is similar to that of D. atacellana. There are fourteen ridges on the posterior sorting area of the stomach. Fig. 8 Nucula callicredemna. A. Internal view of left valve; from Station 156; B. Detail of hinge of same valve. (Scales = 1.0 mm). STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 73 Nuculoidea bushae (Dollfuss, 1898) Polygas DS18 2138 1 47°32.9'N 08°44.9'W SD 22.10.72 Figs 9-11 DS25 2094 11 44°08.2'N 04°15.7’;W SD _ 01.11.72 DS26 2076 11 44°08.2'N 04°15.0°'W SD _ 01.11.72 TYPE LOCALITY: U.S. Bureau Fisheries Sta. 2171, Lat. 37°59'30"N, Long. 73°48’40"W. GUYANA BASIN Knorr 295 1022 28 08°04.2'N 54°21.3’W ES 28.02.72 TYPE SPECIMEN: Holotype: USNM No. 40474. GUINEA BASIN Nucula subovata Verrill & Bush, 1898, p. 858, pl. 81, Fig. 8; Walda DS28 1261 37 04°21.2/N 04°35."E SD 00.00.71 pl. 83, fig. 5; Johnson, 1934, p. 15; (non D’Orbigny, 1850). Nucula bushi Dollfuss, 1898, p. 180. OS EeSIN Nucula bushae emend Schenck, 1939, p. 37. Atlantis II 203 542 22 08°48.0’'S 12°52.0’'E ES 23.05.68 DEPTH RANGE: 509-2876 metres. SEIN SESS Discovery 6696 1780 3 28°06.0’N 13°28.0'W ES 15.03.68 6697 1564 2 17°57.0'N 13°46.2'W ES 15.03.68 MATERIAL 6701 1934 12 27°45.2‘N 14°13.0'W ES 16.03.68 6704 2129 17 27°44.9'N 14°25.0°'W ES 17.03.68 Cruise Sta’ DepthNo Lat Long Gear Date CAPE VERDE BASIN st) Atlantic II 138 1976 30 10°36.0'N 17°52.0°W ES 04.02.67 139 2187 21 10°33.0'N 17°53.0'W ES 04.02.67 141 1796 38 10°30.0’N 17°51.5'W ES 05.02.67 142 1796 101 10°30.0'N 17°51.5°'°W ES 05.02.67 Atlantis D1 509 13 39°54.5’N 70°35.0'W AD _ 23.05.62 143, 2095 18 10°35.0’N 17°44.0'W ES 05.02.67 Chain 105 530 235 39°56.6’N 71°03.6'W ES 05.05.66 144 2357 39 8 10°36.0’N 17°40.0'W ES 05.02.67 145 2185 25 10°36.0’'N 17°49.0'W ES 06.02.67 NORTH AMERICA BASIN WEST EUROPEAN BASIN Sarsia S44 1739 67 43°40.8'N 30°35.2’W ES 16.07.67 S65 1922 1 46°16.5'N 04°44.0'W ES — 25.07.67 : ; Chain 313. 1500 46 51°32.2’N 12°35.9'W ES 17.08.72 A closely related species appears to be Nucula aegeensis Chain 314 1015 42 51°54.6’N_12°27.3'W ES 18.08.72 Forbes, (1844). Like many Forbes types, that of N. aegeensis 316 2209 10 = 50°58.7'N_ 13°01.6'W ES 18.08.72 has been lost, and although specimens named N. aegeensis Challenger 4 1993 85 56°52.0’N 10°01.0'W ES 05.06.73 : . . : 12 2076 19 56°49.0'N 10°15.0’'W ES 20.09.73 exist and have been examined, eg. Jeffreys collections in the 13 1842 93 56°45.0’'N 09°50.0’W ABD 22.09.73 BM(NH) No. 85.11.5.560-4 and 85.11.5.567-570 and USNM, 15 1632 32 56°44.0'N 09°28.0'W ES 22.09.73 there is no certainty that these are in fact the species that 18 1392 64 56°44.0'N 09°20.0'W ES = 22.09.73 Forbes described from off Crete. Indeed Jeffreys, (1879) is Beene! TONS ) eee Nb 10-8 Wiles) 12s ue uncertain and states ““Assuming this to be Forbes’s species 22 1028 392 56°41.0’N 09°11.0'W ES — 23.09.73 : Swe ; : a 57°08.0'N 12°09.0’W. SBC. 23.06.76 (although his description is too scanty to be satisfactory). 61 2000 63 1800 56°37.0'N 09°49.0'W SBC 25.06.76 64 1400 56°39.0'N 09°29.0'W SBC 26.06.76 56°39.0'N 09°40.0'W SBC 26.06.76 56°39.0’N 09°23.0'W SBC 26.06.76 56°39.0’N 09°13.0'W SBC 26.06.76 58°42.0’N 09°43.0'W SBC _ 01.07.76 60°00.0'N 10°35.0'W ES 09.07.76 nN nN — i>) S £ = wom 56°16.0'N 09°44.0'W SBC 06.08.83 Incal DS02 2081 57°58.8'N 10°48.5'W SD _ 16.07.76 105 1600 15 58°27.0’N 12°35.0'W ES 10.07.76 155 1330 6 48°27.0’'N 10°20.0'W SBC 04.08.79 156 1310 6 48°27.0'N 10°21.0'W SBC 05.08.79 159 2036 3 50°55.0'N = 12°21.0'W SBC_ 08.08.79 188 2876 1 54°40.6’N 12°16.1';W SBC _ 15.08.81 215 2001 2 57°02.0'N 09°47.0'W SBC 03.08.82 220 1608 8 59°05.0'N 08°51.0'W SBC 04.08.82 222 1101 15 59°43.0'N 07°43.0'W SBC _ 05.08.82 232 2105 8 57°17.0'N 10°16.0'.W ES 19.05.83 ips 7ies\o) 1 56°40.0'N 10°30.0'W SBC _ 05.08.83 273 2185 10 56°05.0’N 10°28.0'W AT 05.08.83 1 1 3 CP02 2091 57°58.4'N 10°42.8'W SD _ 16.07.76 BAY OF BISCAY Biogas I DS11 2205 1 47°35.5'N 08°33.7'W SD _ 08.08.72 Biogas III DS49 1845 20 44°05.9'N 04°15.6’W SD _ 01.09.73 DS50 2124 3 44°08.9'N 04°15.9'W SD _ 01.09.73 Biogas IV DS52 2006 24 44°06.3’N 04°22.4'W SD 18.02.74 Biogas V CP07 2170 7 44°09.8'N 04°16.4".W CP 21.06.74 Biogas VI DS71 2194 1 47°34.3'N _08°33.8'W SD _— 20.10.74 CP24 1995 5 44°08.1’N 04°16.2,;W CP 31.10.74 DS86 1950 9 44°04.8'N 04°18.7,W SD 31.10.74 DS87 1913 14 44°05.1’N 04°19.4'W SD _ 01.11.74 : : DS88 1894 14 44°05.2'N 04°15.7';W SD _ 01.11.74 Fig. 9 Nuculoidea bushae. Lateral and dorsal views of a shell from Station 87. (Scale = 1.0 mm). 74 Specimens of Jeffreys labelled N. aegeensis differ markedly in shape and fringe detail from N. bushae. Likewise Dall, (1886) also identified N. aegeensis from the Atlantic off Havana in 320 m and 823 m, off Morro Light 479 m and off St Vincent in 847 m regarding this as a geographical race of Nuculoma tenuis a view not supported by Jeffreys, (1879). There is certainly confusion, an easily accessible example is a compar- ison of the photograph in Abbott, (1974) of a West Atlantic specimen against a well-figured specimen in Locard, (1898). They are dissimilar in details of hinge and outline. SHELL DESCRIPTION The description by Verrill & Bush, (1898) is detailed and accurate. Shell small, elongate-ovate, equivalve, inequilateral; semi- transparent, dentition often visible through shell, surface smooth, lustrous with fine concentric lines; umbos posterior to midline, orthogyrate, to slightly opisthogyrate; when pos- terior margin orientated vertically anterior dorsal margin is well above the horizontal; lunule often bordered by tiny perforations, escutcheon indistinct; teeth typical chevron- shape, number varying with size of specimen, 3 posterior and 4 anterior teeth in a specimen 1.64 mm in length and 6 posterior and 9 anterior teeth in a specimen 4.47 mm in length, ligament slightly oblique to moderately narrow hinge plate, does not extend ventral to anterior proximal teeth. COMPARATIVE SHELL MEASUREMENTS OF Nuculoidea bushae Dimension (mm) No Mean SD Max Min NORTH AMERICA BASIN Length ZS 2.70 1.05 4.78 1.05 Width 1°23 0.479 2.28 0.53 Height 2.01 0.764 3.51 0.82 WEST EUROPEAN BASIN Length 23 2.05 0.75 4.11 0.86 Width 1.06 0.35 2.13 0.49 Height 1.64 0.55 3.26 0.60 BAY OF BISCAY Length 54 2.68 0.85 3.99 1.25 Width 1.44 0.51 2.36 0.68 Height Helis, 0.69 3.14 1.01 ANGOLA BASIN Length 55 2.23 0.95 Sell 0.73 Width 1.00 0.45 2.67 0.35 Height 1.72 0.72 4.38 0.59 GUYANA BASIN Length 28 1.18 0.46 2.54 0.76 Width 0.57 0.19 1.03 0.400 Height 0.92 0.33 1.82 0.61 CANARIES BASIN Length 5 3.24 0.99 3.85 1.47 Width 1.83 0.57 Di pepd 0.83 Height 2.65 0.818 3.17 1.21 CAPE VERDE BASIN Length 29 2.82 0.76 S292 1.34 Width 1.46 0.44 2.15 0.73 Height Zee 0.50 2.93 1.09 P.M. RHIND AND J.A. ALLEN INTERNAL MORPHOLOGY N. bushae differs from Deminucula atacellana in having a much more extensive foot with less pronounced papillae but with a marked heel. The ‘byssal’ gland is large. In N. bushae the adductor muscles are oval in cross-section and the ante- rior is slightly larger than the posterior. The ‘quick’ and ‘catch’ portions are approximately equal in size. The gill axis lies parallel to the postero-dorsal margin of the shell. The gill plates number 30 in a specimen 3.1 mm in length. The palps are relatively large compared with D. atacellana but do not extend beyond the limits of the foot. The palp ridges number 30 in a specimen ca. 2 mm in length. The palp proboscides, although highly contracted, appear to have the same relative proportions as in D. atacellana. The stomach differs from that of D. atacellana in being more slender and the hind gut is less extensive (Fig. 11). The maximum recorded diameter of ova from these samples is 140 um. This is in contrast to Scheltema (1972) who recorded an egg diameter of 270 um. We believe that this is because Scheltema (1972) had examined speci- mens of N. similis a species of very similar appearance. There is no evidence of seasonal reproduction and fully mature specimens were recorded in February, June, August and October. DISTRIBUTION Nuculoidea bushae occurs in the North America, Angola, Guinea, Guyana, Canaries, Cape Verde and West European Basin, the latter includes the Rockall Trough and the Bay of Biscay. Verrill & Bush, (1898) thought that Nuculoidea bushae to be closely related to the shallow water species Nuculoma tenuis. The shell is, however, much less oblique and more elongate than in the latter species, while the hinge plate of N. bushae is broader and the resilifer is more rounded, less oblique and does not extend ventral to the anterior teeth. Verrill & Bush, (1898) collected N. bushae from four U.S. Fisheries Commission stations in the North American Basin between Lat. 40°00.0’N. Long. 71°14'30’W, and Lat. 37°8’N, Long. 74°33’W., in 287-812 m but, until now, the species has not been recorded from other regions of the Atlantic. Fig. 10 Nuculoidea bushae. Internal view of left valve from Station 314. (Scale = 1.0 mm). STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 75 Fig. 11 Nuculoidea bushae. A. Semidiagrammatic view from the right side of the body organs; B. Lateral view of part of body from the left side to show the position of the ovary in a mature specimen; C. Course of hind gut as seen from the right side. (Scale = 1.0 mm; for key to abbreviations see p.63). Nuculoidea pernambucensis (Smith 1885) Figs 12-14 TYPE LOCALITY: Challenger Sta. 120, Lat. 08°37'00"S, Long. 34°28'00"W, 675 fms. TYPE SPECIMENS: Syntypes, 5 valves, BM(NH) reg.no. 1887.2.9.2910-11. (Examined by PR). (Note: Two specimens from Knorr, Sta. 293, 08°58.0’N, 54°04.3’W, 1518 m lodged in BM(NH) No. 1990012). Nucula pernambucensis Smith, 1885, p. 227, pl. 18, Figs 10—-10a. Nucula cymella Dall, 1886, p. 246; 1889, p. 42; 1890, p. 258, pl. 13, Fig. 1; 1903, p. 42; Johnson, 1934, p. 16; Clarke, 1962, p. 48; Knudsen 1970, p. 220. Nucula turnerae Clarke, 1961, p. 367, 368; pl. 2, Figs 2, 3; 1962, p. 49. DEPTH RANGE: 587-2976 metres. Nucula cymella Dall, (1890) was synonymized with Nucu- loidea pernambucensis by James, (1972). Clarke, (1960) noted that Nucula turnerae from the Argentine Basin was similar to Nuculoidea pernambucensis and that it was possibly a juvenile specimen of the latter species. Verrill & Bush, (1898) also noted Nuculoidea bushae resembled both Nucu- loidea pernambucensis and Nuculoma tenuis. N. pernam- bucensis appears to be closely related and probably congeneric with Nucula sp. A described by James, (1972) from the Gulf of Mexico. MATERIAL Cruise Sta Depth No Lat Long Gear Date (m) GUYANA BASIN Knorr 293 151823" -.08°58,0’N,_54°04°3’W ES . 27.02.72 299 2076 3 O7°55.1°N 55°42.0'W ES 20.02272 BRAZIL BASIN Atlantis II 167 975 4 07°58.0'S 34°17.0'W ES 20.02.67 169 587 12 08°03.0’'S 34°23.0’'W ES 21.01.67 SHELL DESCRIPTION The original description by Smith, (1885) is not very detailed. Shell broadly ovate, somewhat triangular, equivalve, inequilateral, transluscent, with dentition often visible through shell, strong concentric ridges; umbos posterior to the midline, orthogyrate; when postero-dorsal margin orien- tated vertically antero-dorsal margin well above horizontal; lunule often bordered by tiny perforations, escutcheon indis- tinct; hinge-teeth, small, pustular, 3 posterior and 4 anterior teeth in specimen of 1.97 mm in length, and 9 posterior and 11 anterior teeth in specimen 5.70 mm in length, ligament perpendicular to very broad hinge plate, does not extend ventral to anterior proximal teeth. COMPARATIVE SHELL MEASUREMENTS OF Nuculoidea pernambucensis Dimension (mm) No Mean SD Max Min GUYANA BASIN Length 19 3.69 0.85 5.06 LOY Width 2.00 0.54 3.02 0.97 Height 2.99 0.69 4.06 1.56 BRAZIL BASIN Length 9 3p 0.73 5.70 3.25 Width AAT] 0.57 3.54 1.64 Height 3.26 0.63 4.73 2.60 N. pernambucensis appears to be unique among extant nuculids in possessing an extremely broad hinge plate with narrow obtuse and somewhat pustulate teeth (Fig. 12). This feature is fairly common among Palaeozoic species, for example species of Tancrediopsis Beushausen (McAlester, 1963; Bradshaw, 1970) and Nuculoidea Williams & Breger (Vokes, 1949). This type of hinge is also present in an unusual nuculoid Tironucula jugata from the Lower Ordovician of Spitzbergen (Morris & Fortey, 1976). This latter species originally placed in a family Praenuculidae (McAlester, 1969) was later regarded by Morris & Fortey, (1976), on the basis of 76 Fig. 12 Nuculoidea pernambucensis. Lateral and dorsal views of a shell from Station 293. (Scale = 1.0 mm). its hinge morphology, as intermediate between taxodont and actinodont forms. Actinodont dentition is probably the more primitive and gave rise to the taxodont condition. Allen & Sanders, (1973) point out that the lack of taxodont teeth per se can no longer be used as the prime criterion for excluding fossil bivalves from the Nuculoidea. N. pernambucensis also resembles the Silurian fossil Nuculoidea pinguis pinguis (Soot-Ryen, 1964); N.p. pinguis together with members of the primitive nuculoid family Ctenodontidae, are thought by Soot-Ryen, (1964) to have lived in a deep-sea environment. This is based on the fact that they lived in muddy sediments so tranquil that valves did not separate after the animal’s death. INTERNAL MORPHOLOGY The adductor muscles are oval, the anterior being slightly larger than the posterior. The ‘quick’ and ‘catch’ portions are approximately equal in size. The gill axis lies parallel to the posterior dorsal margin of the shell and gill plates number 20 in a specimen 3.9 mm in length. The palps are relatively longer than those of D. atacellana and palp ridges number 40 in a specimen 4 mm in length. The ‘byssal’ gland is large. DISTRIBUTION Because only dead valves had been collected from abyssal depths Knudsen, (1970) did not regard Nuculoidea pernam- bucensis as truly abyssal, and James, (1972) on the basis of collections from the Gulf of Mexico, also supported this view. The results here confirm that it occurs at abyssal depths at least in the Guyana Basin. P.M. RHIND AND J.A. ALLEN HMS ‘Challenger’ originally collected N. pernambucensis off Pernambuco (Recife) on the slope of the Brazil Basin. Nucula cymella was recorded from the Yucatan Straits, Florida Strait, south of Cuba, east of Tobago and north of Ceara, Brazil (Dall, 1890) and N. turnerae was collected in the Argentine Basin at ‘Vema’ Sta. 12 (Clarke, 1961). James, (1972) collected N. pernambucensis in the north east and the north west of the Gulf of Mexico at depths of 732-1683 m but living specimens were only taken in depths of 1000-1494 m. Since specimens are found at upper slope depths it is possible that N. pernambucensis may have evolved following a downward migration of an adjacent shelf species of the tropical shelf of South America and/or the Gulf of Mexico. Such a theory may explain why this, unlike other deep-water species, is large and has unusually strong concentric ridges and why the hinge plate is heavily calcified. Tropical molluscs tend to be more heavily calcified (Vermeij, 1978; Nicol, 1964; 1965, 1966 and 1967) and Moore, (1977) lists a number of heavily calcified endemic tropical nuculid species with strong concentric ridges. Conversely, reduced calcification is charac- teristic of deep-water species (Nicol, 1967) and is related to low temperatures and to high pressure (Grause, 1974). Thus, N. pernambucensis is unusual in this respect. Unfortunately no living shallow-water species similar to N. pernambucensis occurs in the tropical western Atlantic today (Moore, 1977), nor is there a similar species in the Holocene formations of Surinam even though a number of nuculids are present (Altena, 1968). N. pernambucensis has the second-largest size of ovum of the species recorded here. The largest diameter recorded was 204 um, sufficiently large to predict direct development rather than the more-typical lecithotrophic development with a free-swimming stage (Ockelmann, 1965). Nuculoma perforata (new species) Figs 17-19 TYPE LOCALITY: Atlantis II. Sta. 236, 36°27.0’S, 53°31.0'W, 518 m. TYPE SPECIMEN: Holotype BM(NH) No. 1990008 from type locality; paratypes at present lodged at Woods Hole Oceano- graphic Institution (10 specimens in lot). Fig. 13. Nuculoidea pernambucensis. Internal view of the left valve of a shell from Station 293. (Scale = 1.0 mm). STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE We Fig. 14. Nuculoidea pernambucensis. Semidiagrammatic view from the right side of the body organs. (Scale = 1.0 mm). See figures 5 & 15 for identification of the parts. DEPTH RANGE: 518—2323 metres. MATERIAL Cruise Sta DepthNo Lat Long Gear Date (m) ARGENTINE BASIN Atlantis II 236 «©6518 15 36°27.0'S_ 53°31.0'W ES 11.03.71 237 101d A 9 N86232.69S" . 53°23.0'W ES’ 41.03.71 239 1679 37 36°49.0'S 53°15.4;W ES _ 11.03.71 ZA0M (2323.1, |) S6°5314/S) S3°10!2'W WES © 1203.71 ee 480 — SGaS.’S 52°79" WES ~27.03771 DEA 2048 BCD 5242 YW AD 28003971 SHELL DESCRIPTION Shell small, elongate-ovate, equivalve, inequilateral; denti- tion usually visible through the dorsal part of the shell, smooth with fine concentric lines; umbo posterior to midline, opisthogyrate; when posterior margin orientated vertically, anterior dorsal margin is above horizontal; lunule usually bordered by tiny perforations, escutcheon, indistinct; resilifer oblique to narrow hinge plate, except in small specimens (<2 mm) extends ventral to anterior proximal teeth; hinge teeth chevron-shaped, the number varies with size, 3 poste- rior and 4 anterior teeth, in a specimen 1.6 mm in length and 4 posterior and 8 anterior teeth, in a specimen 4.7 mm in length. SHELL MEASUREMENTS OF Nuculoma perforata Dimension (mm) No Mean SD Max Min ARGENTINE BASIN Length 20 713/53) ied 9.04 1.42 Width 1.46 0.68 3.67 0.79 Height 2.24 MDL 6.54 1.07 eal abe ‘ hese Leer Me WN he ue ee Meni , Fig. 15 Nuculoma perforata. Lateral and dorsal views of a shell from Station 239. (Scale = 1.0 mm). INTERNAL MORPHOLOGY The adductor muscles are oval in cross section the anterior being larger than the posterior (not obvious in Fig. 17 because of angle of view). The ‘quick’ and ‘catch’ portions are approximately equal. The gill axes lie parallel to the postero- dorsal margin of the shell. Gill plates number 18 in a specimen 3.7 mm in length. The palps are slightly smaller than those of N. subovata and do not extend to the posterior limit of the foot. There are also relatively fewer palp ridges (15 in a specimen 3.7 mm in length). The palp proboscides are typical but the ‘byssal’ gland is smaller than that of D. atacellana. This species has an exceptionally large stomach (Fig. 17). DISTRIBUTION Nuculoma perforata is restricted to the Argentine Basin. It appears to be closely related to the much-larger species Nucula puelcha d’Orbigny, 1842 from the adjacent continen- tal shelf (Schenck, 1939). Nucula puelcha is similar to Nucu- loma tenuis and should now be classified as a member of the genus Nuculoma. N. puelcha is present in WHOI collections from shallow-water stations adjacent to the Argentine Basin and we have considered the possibility that N. perforata may be young slow-growing specimens of Nuculoma puelcha, close to the limit of their bathymetric range. Although no mature specimens of Nuculoma perforata are present in the samples they differ in the configuration of the hindgut (Fig. 18) and furthermore they were more abundant at 78 Fig. 16 Nuculoma perforata. Internal view of left valve of shell from Station 236. (Scale = 1.0 mm). St Fig. 17 Nuculoma perforata. Semidiagrammatic view of right side of the body to show arrangement of organs. (Scale = 1.0 mm; for key to abbreviations see p.63). Fig. 18 Nuculoma perforata. The course of the hind gut as seen from the right side. 1670 m than at 507m. We believe N. perforata to be a separate but closely related species to N. puelcha. The species has some similarity with specimens dredged from relatively shallow water off the coast of Guyana (exact locality unknown) and named Nucula surinamensis by Alt- ena, (1968). P.M. RHIND AND J.A. ALLEN Nuculoma granulosa (Verrill, 1884) Figs 19-22 TYPE LOCALITY: Fish Hawk Sta. 892, Lat. 39°46'00"N, Long. 71°10'00"W, 891 m. TYPE SPECIMEN: Holotype USNM No. 52561. (Examined JAA). (Note: 5 specimens from Sarsia, Sta. S63, 46°17.5'N, 04°45.2’W, 1336 m lodged in BM(NH) No. 1990015). Nucula granulosa Verrill, 1884, p. 280; Dall, 1889, p. 42; Verrill & Bush, 1898, p. 853, pl. 81, Fig. 2, pl. 88, Fig. 8; Johnson, 1934, p. 15. Nucula cortica Grassle, 1977, p. 618; 1978, p. 42. DEPTH RANGE: 811-2178 metres. MATERIAL Cruise Sta DepthNo Lat Long Gear Date (m) NORTH AMERICA BASIN 2 Atlantis F1 1500 33 39°47.0'N 70°45.0'W AD 24.05.61 Atlantis II 73 1470 1143 39°46.5'N 70°43.3’W ES 25.08.64 Chain 87 1102 844 39°48.7’N 70°40.8’W ES 06.07.65 103 2022 6 39°43.6'N 70°37.4’.W ES 04.05.66 105 530 232 39°56.6’N 70°03.6'W ES 05.05.66 Atlantis II 128 1254 5 39°46.5’N 70°45.2'W ES 16.12.66 131 2178 1 39°38.5’N 70°36.5'W ES 18.12.66 Chain 207. +811 161 39°51.0’N 70°56.4".W ES — 21.02.69 209 1693 887 39°46.0'N 70°51.5’'W ES 22.02.69 210 2064 6 39°43.2’N 70°49.5'W ES — 23.02.69 WEST EUROPEAN BASIN Sarsia $63 1336 31 46°17.5'N 04°45.2";W ES — 24.07.67 Challenger ES20 1271 48 56°46.0’N 09°15.0'W ES = 23.09.73 Verrill, (1884) originally described the species as occurring in the North America Basin at U.S. Fisheries Commission stations 892, 1880, 1883 and 2072. This was later updated by Verrill & Bush, (1898) to include six stations between Lat. 39°43'45"N, Long. 70’7 W. and Lat. 36°47’N, Long. 73°9'30"W, in 2086-3340 m (station numbers were not given). Dall, (1889) describes the species as having a latitudinal range off the east coast of N. America from Georges Bank in the North to Cape Lookout in the South. This study extends its range to the Bay of Biscay and the Rockall Trough. SHELL DESCRIPTION The original description by Verrill, (1884) is not very detailed. Shell very small, broad-ovate, equivalve, inequilateral, opaque surface smooth with fine concentric lines; umbo posterior to midline, opisthogyrate; when posterior margin orientated vertically, anterior dorsal margin is below horizon- tal; lunule bordered by angular ridge, escutcheon, indistinct; hinge teeth chevron-shaped, the number varying with size, 2 posterior and 3 anterior teeth in a specimen 0.94 mm in length, and 4 posterior and 6 anterior teeth in a specimen 1.98 mm in length; ligament large, perpendicular to relatively broad hinge plate, does not extend ventral to the anterior proximal teeth. Although the external shell morphology is somewhat variable, it must be regarded as a single species. STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 79 Fig. 19 Nuculoma granulosa. Lateral and dorsal views of a shell from Station 73. (Scale = 1.0 mm). COMPARATIVE SHELL MEASUREMENTS OF Nuculoma granulosa NORTH AMERICA BASIN Length 22 2.05 0.28 2.45 1.39 Width 1.18 0.45 1.75 0.85 Height 1.81 0.27 2.26 1.22 BAY OF BISCAY Length 10 1.76 0.16 1.93 1.42 Width 1.06 Om2 hey 0.83 Height 1.54 0.14 1.69 2 INTERNAL MORPHOLOGY The adductor muscles are oval and the anterior is larger than the posterior. The ‘quick’ and ‘catch’ portions are approxi- mately equal. The gill axis lies parallel to posterior margin of the shell and the gill plates number 14 in a specimen ca. 2 mm in length. The palps do not extend to the posterior limit of the foot. The palp ridges number 16 in a specimen ca. 2 mm in length. The foot has a similar orientation and relative size to that of Deminucula atacellana, but the ‘byssal’ gland is narrower. The hind gut configuration is also similar to that of D. atacellana. The maximum recorded diameter of the egg is Fig. 20 Nuculoma granulosa. Internal view of left valve of shell from Station 563. (Scale = 1.0 mm). Fig. 21. Nuculoma granulosa. Semidiagrammatic views of A, right and B, left side of body of different specimens to show arrangement of body organs. (Scale = 1.0 mm; see Fig. 19 for identification of parts and for key to abbreviations see p.63). 80 120 um (Scheltema, 1972) with approximately 220 eggs in a specimen 2.2 mm total length. DISTRIBUTION It is an upper slope stenobathic species. It is closely related to Nucula corticata Moller, 1842 (Sowerby, 1871) a Greenland shallow-water species suggesting, as in the case of Nuculoma similis (see p.81 et seq.), a high latitude shallow-water deriva- tion. Nuculoma similis (new species) Figs 23-2 TYPE LOCALITY: Chain Sta. 96, Lat. 39°55.2’N, Long. 70°39.5'W, 498 metres. TYPE SPECIMEN: Holotype: BM(NH) No. 199005; Paratypes No. 1990006 (5 specimens in lot). a DEPTH RANGE: 498—2064 metres. MATERIAL Cruise Sta DepthNo Lat Long Gear Date (m) NORTH AMERICA BASIN Atlantis 3) "823 37 39°59:5'N 70°35.0.W AD 25.05.61 Atlantis II 73, 1470 =126 =39°46.5’'N 70°43.3'W ET 25.08.64 Chain 87 1102 270 39°48.7'N 70°40.8'W ET 06.07.65 88 478 34 39°54.1'N 70°37.0'W ET 06.07.65 96 = 498 86 39°55.2'N 70°39.5'W ET 27.04.66 103 2022 2 39°43.6’'N 70°37.4';W ET 04.05.66 Atlantis II 128 1254 27 + 39°46.5’N 70°45.2’,W ES 16.12.66 Chain 207 811 1348 39°46.5'N 70°56.4’'W ES 21.02.69 209 1693 200 39°46.0’N 70°51.5’W ES 22.02.69 210 2064 1 39°43.2'N 70°49.5'W ES 23.02.69 SHELL DESCRIPTION Shell subtriangular, oblique, ventricose, equivalve, inequilat- eral; surface lustrous with fine concentric lines; umbos poste- rior to midline, opisthogyrate; when posterior margin orientated vertically, the antero-dorsal margin is below hori- zontal; indistinct lunule, chordate shaped escutcheon; hinge teeth chevron-shaped, the number varying with size, 1 poste- rior and 3 anterior teeth in a specimen 1.6 mm in length, and 3 posterior and 6 anterior teeth in a specimen 2.6 mm in length; ligament oblique to narrow hinge plate, extends slightly ventral to anterior proximal teeth. SHELL MEASUREMENTS OF Nuculoma similis Dimension (mm) No Mean SD Max Min NORTH AMERICA BASIN Length 71 1.97 0.420 3.03 1.19 Width 1.18 0.255 2.87 0.73 Height 1.62 0.351 1.61 1.0] INTERNAL MORPHOLOGY The adductor muscles are approximately equal in size, more or less oval in shape. The anterior part of the gill axis lies P.M. RHIND AND J.A. ALLEN parallel to the dorsal margin of the shell but posteriorly it curves to lie parallel to the posterior margin. The gill plates number 15 in a specimen ca. 3 mm in length. The palps are small with 16 palp ridges in a specimen of ca. 3 mm in length. In preserved specimens the foot is laterally compressed and the ‘byssal’ gland is extremely small. The stomach has an unusually small style sac. The hind gut configuration is not as complex as that of Deminucula atacellana. N. similis has the largest ovum recorded for any species described in this paper (maximum diameter 244 um). Schel- tema, (1972) recorded a maximum egg size of 270 um for N. subovata which was almost certainly a misidentification for N. similis (see p. 73). There can be little doubt that this species has direct development. DISTRIBUTION This species is restricted to the North America Basin. In view of the intensity of past sampling in this Basin, it is curious that it has never been described previously, particularly as it is more abundant and has a greater vertical depth range than that of Nuculoidea subovata (p. 73). Nuculoma_ similis appears to be closely related to the Arctic shallow-water species Nuculoma bellotii (Fig. 26) (cf. 2 specimens from 181 m off Baffin Island BM(NH) No. 1990014). The taxonomic status of N. bellotii has been disputed. Three Arctic species, Nucula inflata Hancock, 1846, Nucula expansa Reeve, 1855, and Nucula bellotii Adams, 1856, have been regarded as being conspecific with, or varieties of, Nuculoma tenuis (Gould, 1870; Whiteaves, 1901; Soot-Ryen, 1932; Filatova, 1948; Thorsen, 1951; Ockelmann, 1958; MacGinitie, 1959). Schenck, (1939) regarded the three Arctic species as synonymous and because the names Nucula inflata and Nucula expansa are homonyms the first available name is Nucula bellotii. Schenck accepted the subgenus Ennucula (Iredale) and named the species Nucula (Ennucula) bellotii. Schenck, (1939) also synonymized N. bellotii with the North oe Pg Fig. 22 Nuculoma granulosa. A. Detail of stomach and anterior part of nervous system and B. The course of the hind gut as seen from the right side. (For key to abbreviations see p.63). STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 81 Fig. 23 Nuculoma similis. Lateral and dorsal views of shell from Station 207. (Scale = 1.0 mm). Pacific species N. quirca Dall. Filatova, (1948) reports that an inflata-expansa complex (N. bellotii) as occurring in deep waters of the Barents and Bering Seas, while Lubinsky, (1972) also reported that specimens collected by Krause, (1885) from the Bering Straits and by MacGinity, (1959) from Point Barrow, and identified as N. tenuis Mont., N. inflata Hancock and N. expansa Reeve, were N. bellotii. Lubinsky’s study indicates that N. bellotii occurs throughout the Cana- dian Arctic and she regards it as a circumpolar, panarctic species and separate from N. tenuis, however, she inferred that N. bellotii might be a subspecies of N. tenuis, although from comparisons of N. tenuis from Scottish waters, we doubt that. Both are species of the genus Nuculoma. Bernard, (1979) collected specimens of Nuculoma bellotii from 132 stations throughout the Beaufort Sea. Abundant in depths <200m it was also taken from depths down to 2560 m, showing that although N. bellotii is large (up to 20 mm in length) it is capable of deep-water existence. Recently Farrow et al., (1984) recorded N. bellotii from the Baffin Island fjords McBeth and Cambridge. These latter specimens were made available for comparative study (see Table below). Wagner, (1968) regarded fossil Nuculoma bellotii, collected by Bell, (1879) in the Hudson Bay area as part of the faunal assemblage of the Tyrrel Sea. Due to uplift much of this sea which reached its maximum extent between 5000 and 6000BC, disappeared with only remnants such as Hudson Bay and James Bay now remaining. Nuculoma similis and N. bellotii share similar external shell features (eg. The characteristic ‘pouting’ of the anterior dorsal margin). As is usually the case with deep-sea species, specimens of Nuculoma similis are by far the smaller. The Table below lists the ratios of the height/length and width/ length of N. bellotii taken from various localities (Lubinsky, 1972) compared with those of Nuculoma similis. Those of Nuculoma similis fall well within the range of N. bellotii. As in Nuculoma similis, the h/| and w/I ratios of N. bellotii are normally distributed (Lubinsky, 1972) and it is highly proba- ble that N. bellotii and that inflated specimens (= Nucula inflata Hancock) and elongate, narrow ones (= Nucula expansa Reeve) are extreme variants of a unimodal popula- tion of a single species. If Nuculoma similis is a sibling species of N. bellotii it would help to explain why the former species only occurs in the North America Basin. Shallow-water specimens of N. bellotii from Baffin Bay and the Labrador Sea would presumably have been the origin of a sibling population adapted to deep water and early isolated in the North America Basin. That N. bellotii is not present on the New England shelf is probably due to it being a cold-water stenothermic species. The absence of Nuculoma similis from other Basins is probably related to the fact that it has direct development. Note that N. bellotii also has direct develop- ment, but from a much smaller egg (maximum diameter 163 um). Stenothermy would also tend to isolate it to the deep sea and prevent it crossing boundary ridge systems. Both species are absent from the Norwegian Basin (Bouchet & Warén, 1979) which would suggest their centre of origin to be in the North Pacific. SHELL MEASUREMENTS OF Nuculoma bellotii (from Lubinsky, 1972) Species Location No Length h/l wil (mm) Nuculoma bellotii Foxe Basin 40 817 0.73-0.85 0.49-0.65 : : : Hudson Bay 80 612 0.70-0.90 0.49-0.67 ; ; ; Ponds Inlet 10 714 0.77-0.82 0.47-0.67 : : : WellingtonCh& 14 10-16 0.78-0.86 0.49-0.68 Frobisher Bay . : : Labrador 16 10-16 0.79-0.90 0.54-0.64 Nuculoma similis N. America 71 #1.19- 0.79-0.89 0.54-0.65 Basin 3.03 Other Atlantic Arctic species to which Nucula granulosa Fig. 24 Nuculoma similis. Internal view of left valve of shell from Station 96. (Scale = 1.0 mm). Fig. 25. Nuculoma similis. Semidiagrammatic view of right side of body to show arrangements of organs. (Scale = 1.0 mm; see Fig. 19 for identification of parts). Fig. 26 Nuculoma bellotii. Lateral view of right valve of shell from Hudson Bay. (Scale = 1.0 mm). and WN. similis have possible similarity are N. corticata and N. delphinodonta. N. delphinodonta has a predominantly arctic shelf distribution, but off the Eastern Seaboard of the USA it extends south from Labrador to Maryland (Ockelmann, 1958; Abbott, 1974). Thus, its distribution is adjacent to that of N. similis. Furthermore, it has a large egg which has direct development, eggs being brooded in a secretion of the pallial gland attached to the posterior part of the shell. Although N. similis has a large egg, we have no evidence to indicate that brooding occurs. Although similar in shape and size, N. delphinodonta a short, stout species with a truncate posterior margin, can be distinguished by its coarse concentric growth lines, and a slight dorsal carination. Other differences include an average anterior to posterior hinge teeth ratio of 3:7 as compared with 3:6 for N. similis; the hinge teeth of N. similis are stouter and the hinge plates are broader particularly the anterior above which the antero-dorsal margin is more raised with greater curvature. Nevertheless, the two species appear to be closely related. In comparison with N. granulosa, there is less similarity. N. granulosa is a much shorter species with fewer anterior hinge P.M. RHIND AND J.A. ALLEN teeth and a shell outline that is significantly different from N. delphinodonta. N. granulosa has eggs of much smaller size, and almost certainly has a short, non-feeding, planktonic larva. Comparison with N. corticata is much more difficult, simply because so little is known of this species. It was described as a Greenland species in Moller, (1842) and further described and illustrated by Sowerby, (1871) in Volume 18 of Reeve’s Conchologia Iconica. There is little mention of it in the literature on the molluscan fauna of Greenland and Eastern Canada since then except that Soot-Ryan, (1966) records three specimens from the Michael Sars Expedition at 1100 m previously recorded as N. tenuis by Greig, (1920). Posteriorly the shell is rounded and not truncate as in N. similis and N. granulosa. Indeed, in outline N. corticata would appear to be similar to N. tenuis expansa. Nuculoma elongata (new species) Figs 27-28 TYPE LOCALITY: Knorr Sta. 297, Lat. 7°45.3’, Long. 54°24’W. TYPE SPECIMENS: Holotype: BM(NH) No. 1990007 from type locality; paratypes at present lodged at Woods Hole Oceanographic Institution (2 specimens in lot). DEPTH RANGE: 508-523 metres. MATERIAL Cruise Sta Depth No _ Lat Long Gear Date (m) GUYANA BASIN Knorr 297 508-523 8 07°45.3'N 54°24.0'W ES — 28.02.72 SHELL DESCRIPTION Shell very small, elongate-ovate, equivalve, inequilateral; surface smooth with fine concentric and radial striations, umbo posterior to midline, orthogyrate; when postero-dorsal margin orientated vertically antero-dorsal line is above the horizontal, posterior margin, dorsal to midline with faint concavity, lunule, bordered by tiny indistinct perforations, escutcheon, indistinct; ligament oblique to narrow hinge plate, not extending ventral to the anterior proximal teeth; 3 posterior and 4 anterior teeth in a specimen 2.1 mm in length. SHELL MEASUREMENTS OF Nuculoma elongata Dimension (mm) No Mean SD Max Min GUYANA BASIN Length 5 JF) 0.24 2.01 1.47 Width 0.84 0.08 0.93 0.73 Height 1.30 0.15 1.47 1.11 INTERNAL MORPHOLOGY The adductor muscles are oval in shape and approximately equal in size. The ‘quick’ and ‘catch’ portions are also STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 83 approximately equal in size. The gill axis lies parallel to the posterior margin. There are eight gill plates and twelve palp ridges in a specimen ca. 2 mm in length. The foot is large and has a similar orientation to that of Nuculoidea and when retracted extends anterior to the anterior adductor muscle. The ‘byssal’ gland is large. REMARKS All specimens were collected from one station in the Guyana Basin from the upper slope in a depth of 508-523 m. A number of the larger specimens had mature gonads, indicat- ing that they were not the young of a larger species. There is no species of similar shell shape and characteristics in adja- cent waters. Brevinucula verrilli (Dall, 1886) Figs 29-32 TYPE LOCALITY: Albatross Sta. 2229, Lat. 37°38.40", Long. 73°16.30"W, 2604 m. TYPE SPECIMEN: Holotype: USNM No. 45752. Examined (JAA). (Note: 4 specimens from Knorr Sta. 307, 12°34.4’N, 58°59.3'W, 3862 m, BM(NH) No. 1990016). Nucula trigona Verrill 1885, p. 438; (non Bronn, 1849; non Seguenza 1877). Nucula verrilli Dall, 1886, p. 248; 1889, p. 42; 1890, p. 257, pl. 14, Fig. 4; Bush 1893, p. 240, pl. 1, Figs 6, 7; Verrill & Bush, 1898, p. 853, pl. 95, Fig. 10; Clarke, 1962, p. 49; Pequegnat, 1972, p. 74. Brevinucula verrillii Schenck, 1934, p. 41; James, 1972, p. 24-27, Figs 3, 4; map 1. Brevinucula verrilli Knudsen, 1970, p. 19, Fig. 3, pl. 1, Figs 6, 7. Nucula (Brevinucula) verrillii Haas, 1949, p. 7. Nucula guineensis Thiele, 1931, p. 35; pl. 2, Figs 35, 35a; Clarke, 1962, p. 48, Knudsen, 1970. Nucula (Brevinucula) guineensis Thiele, Fig. 788. Brevinucula guineensis Schenck, 1934, p. 4; pl. 5, Figs 22, 2c. Nucula aequalis Barnard, 1964, p. 3365; Fig. 1c. 1934, p. 786; DEPTH RANGE: 1976-4749 metres. The synonymy of this species has been much debated. Thiele, (1931) noted the similarity between Nucula guineensis and B. verrilli, however, while Schenck, (1934) regarded the two species as congeneric and probably conspecific, Clarke, (1962) recorded them as separate species. Later Knudsen, (1970) recorded them as conspecific and this view was upheld by James, (1972) who compared specimens from the Gulf of Mexico with the type (N. trigona, USNM No. 45752, and Brevinucula guineensis, Valdivia Sta. 56, ZMHU). Knudsen, (1970) also believes that Nucula aequalis (Barnard) is also synonymous with Brevinucula verrilli. MATERIAL Cruise Sta. DepthNo Lat Long Gear Date (m) NORTH AMERICA BASIN Chain 77 =3806 11 38°07.0'N 69°16.0'W ES 30.06.65 78 3828 6 38°08.0'N 69°18.77;W ES 30.06.65 84 4749 7 36°24.0'N 67°56.0'W ES 04.07.65 85 3834 21 37°59.2'N 69°26.2";W ES _ 05.07.65 Atlantis II 119 2223 11 32°16.1'N 64°32.6'W ES 19.08.66 340 3356 12 38°14.4’N 70°20.3’'W ES 24.11.73 WEST EUROPEAN BASIN Chain 321 2879 5 50°12.3'N 13°35.8'W ES 29.08.72 323 3356 63 50°08.3’N 13°53.77.W ES 21.08.72 329 4632 29 50°43.3’N 17°44.77W AD 23.08.72 GUYANA BASIN Knorr 291 3868 10 10°06.6’N 55°15.4".W ES 26.02.72 299 2076 2 07°55.1'N 55°42.0'W ES — 29.02.72 301 2500 18 08°12.4"N 55°50.2";W ES = 29.02.72 303 2853 9 08°28.8'N 56°04.5';W ES 01.02.72 306 3429 27 09°31.1'N 56°20.6';W ES 02.03.82 307 3862 15 12°34.4’N 58°59.3'W ES 03.03.72 CANARIES BASIN Discovery 6704 2129 23 27°44.9'N 14°25.0'W ES 17.03.72 6707 2593 2 27°29.2'N 15°26.5'W ES 17.03.68 6709 2351 10 27°29.8'N 15°20.1'W ES 18.03.68 6710 2670 6 27°23.6'N 15°39.6'W ES 19.03.68 6711 2988 1 27°14.9'N _15°36.3'W ES 19.03.68 CAPE VERDE BASIN Atlantis II 138 1976 2 10°36.0'N 17°52.0'W ES 04.12.67 139 2187 3 10°33.0'N 17°53.0'W ES 04.12.67 141 2131 1 10°30.0'N 17°51.5'W ES 05.02.67 143 2095 4 10°35.0'N 17°44.0°.W ES 05.02.67 144 2357 9 10°36.0'N 17°49.0'W ES 05.02.67 145 2185 10 10°36.0'N 17°49.0'W ES 06.02.67 146 2891 4 10°39.5'N 17°44.5'W ES 06.02.67 147 2934 51 10°38.0’N 17°52.0'W ES 06.02.67 148 3828 2 10°37.0’N 18°14.0'W ES 07.02.67 149 3861 7 10°30.0'N 18°18.0'W ES — 07.02.67 MID ATLANTIC Atlantic II 55) SI 2 00°03.0'S_ _27°48.0’.W ES 13.02.67 GUINEA BASIN Walda DS25 2470 56 02°19.8’N 07°49.2'W SD _ 00.00.71 SHELL DESCRIPTION Except that he omits the fact that B. verrilli is without a well-defined resilifer, the original description by Verrill, (1885) is detailed and accurate. Shell deeply triangular, robust, lustrous, equivalve, inequi- lateral; surface smooth, lustrous sometimes with fine irregu- lar radiating striations; umbo approximately medial, orthogyrate, with faint ridge running from umbo to the postero-lateral angle; when postero-dorsal margin orientated vertically, antero-dorsal margin is below horizontal; lunule lanceolate, faint, escutcheon more distinct; resilifer not well- defined; ligament approximately triangular; hinge plate stout, hinge teeth chevron-shaped, the number varies with size, 4 posterior and 5 anterior teeth in a specimen 2.6 mm in length and 7 posterior and 9 anterior teeth in specimen 4.1 mm in length. S84 SHELL MEASUREMENTS OF Brevinucula verrilli Dimension (mm) No Mean SD Max Min NORTH AMERICA BASIN Length 7 2.31 1.05 3.94 1.06 Width 1.21 0.53 2.08 0.60 Height 2.25 1.11 4.06 0.99 WEST EUROPEAN BASIN Length 29 2.35 0.75 3.95 1.00 Width 1.18 0.30 1.83 0.59 Height PIR) 0.77 4.04 0.93 CANARIES BASIN Length 7 2.64 1.43 4.55 1.25 Width 1.37 0.68 2.41 0.73 Height 713) 1.43 4.47 1.20 CAPE VERDE BASIN Length 19 4.01 1.02 5.19 1.63 Width 2.02 0.53 2.87 0.97 Height 3.99 1.03 5.09 1.59 INTERNAL MORPHOLOGY The posterior adductor muscles are more-or-less oval with the anterior muscle the more elongate and convex posteri- orly. The ‘quick’ and ‘catch’ portions are approximately equal. The gill axis lies parallel to the postero-dorsal margin Fig. 27 Nuculoma elongata. Lateral views of three specimens from the right side illustrative of the size range of specimens from Station 297. Body organs as seen through the transparent shell. (Scale = 1.0 mm). P.M. RHIND AND J.A. ALLEN of the shell. The gill plates number 16 in a specimen 3.9 mm in length. The palps are moderately large and, in the same specimen, the palp ridges number 27. The ‘byssal’ gland is large. The stomach is relatively small and the hind gut is extensive (Fig. 31). Note the course of the gut is complex even to the extent that a few loops are to the left of the stomach. The species has large characteristic loosely arranged gonads. The maximum recorded egg diameter is 125 um and development is likely to be lecithrotrophic. There are 260 eggs in a specimen 4.3 mm total length (Scheltema, 1972). The species was originally described as having a triangular ‘chondrophore’ and, as Vokes, (1949) points out, although this is unique among Recent nuculids, it occurs in the Palaeozoic species Nuculoidea opima Hall and Nuculopsis girtyi Schenck. More-recent studies show that although the hinge plate is broad B. verrilli is without a chondrophore but has a well-developed resilifer (Allen & Hannah, 1986). This may also be the case in the Palaeozoic species. According to Moore, (1969) B. verrilli is not present in the fossil record prior to the Miocene. DISTRIBUTION B. verrilli is restricted to the deep sea and no comparable shallow-water counterpart exists. - Verrill, (1885) collected the species from the North Amer- ica Basin at U.S. Fisheries Commission Stations 2194, 2228 and 2229 at depths ranging from 2086-4077 m. Dall, (1890) reported specimens from Station 2754, east of Tobago and Station 2760, north from Ceara, Brazil in 1610 and 1865 metres respectively and summarized the then known geo- STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 85 graphical distribution as from latitude 39°44’, south to Yucatan at depths from 538 to 3340 metres (Dall, 1927). Thiele, (1931) described N. guineensis from the Guinea Basin (‘Valdivia’ Stations 56, 63 and 71) and Knudsen, (1970) set the latitudinal limits from 40°N to 12°S in the western Atlantic and from 4°N to 6°S in the eastern Atlantic. Pequegnat, (1972) reported the species from the gulf of Mexico and this was confirmed by James, (1972) who collected from 12 Stations in depths from 914 to 3475 metres. The report (Knudsen, 1970) that the species occurs at 12°S in the western Atlantic is not confirmed. Specimens taken from this latitude differ from Brevinucula verrilli. Little material was taken from the Brazil Basin and the absence of B. verrilli may be related to insufficient sampling. B. verrilli is predominantly a North Atlantic species. Its distribution in the temperate eastern Atlantic is now extended to 50°N. Despite intensive sampling it is not recorded from the Rockall Trough and Bay of Biscay and the reason for this is not immediately apparent. It has a vertical range that extends into depths shallower than both these regions. No comparable species occurs in the angola and the Argentine Basins, however, Barnard, (1962), described Nucula aequalis from off Cape Point, South Africa and this may be synonymous with B. verrilli (Knudsen, 1970). If this is correct it could be that this is a case of a species with bipolar distribution Ekman, (1953). This phenomenon is unusual among deep sea bivalves and would be the first reported case of an abyssal bivalve to have such a distribution. B. verrilli occurs at greater depths than any other member of the Nuculidae and its distribution extends on to the abyssal plain. Brevinucula subtriangularis (new species) Figs 33-34 TYPE LOCALITy: Atlantis II Sta. 167, Lat. 7°58’S Long. 34°17'W. TYPE SPECIMEN: Holotype: BM(NH) No. 1990009; paratypes at present lodged at Woods Hole Oceanographic Institution (5 specimens in lot). DEPTH RANGE: 943-1007 metres. a Fig. 28 Nuculoma elongata. Internal view of left valve of shell form Station 297. (Scale = 1.0 mm). Fig. 29 Brevinucula verrilli. Lateral and dorsal view of a shell from Station 85. (Scale = 1.0 mm). MATERIAL Cruise Sta. DepthNo Lat Long Gear Date (m) BRAZIL BASIN Atlantis II 167 943- 31 07°58.0'S 34°17.0'W ES 20.02.67 1007 SHELL DESCRIPTION Shell small, robust, lustrous, very smooth, equivalve, inequi- lateral; umbo posterior to midline, opisothygrate; when pos- terior margin orientated vertically, antero-dorsal margin is horizontal; lunule indistinct, escutcheon marked by faint ridge from umbo to the postero-ventral angle; resilifer large; ligament approximately triangular; hinge plate stout, hinge teeth chevron-shaped, 5 posterior and 9 anterior teeth in a specimen 4.9 mm in length. SHELL MEASUREMENTS OF Brevinucula subtriangularis Dimension (mm) No Mean SD Max Min BRAZIL BASIN Length 2 4.95 0.06 5.00 4.91 Width 2.98 0.44 3.29 2.67 Height 4.58 0.15 4.69 4.48 86 P.M. RHIND AND J.A. ALLEN Fig. 30 Brevinucula verrilli. Internal view of left valve of shell from Fig. 32 Brevinucula verrilli. A. Lateral view of left side of Station 85. stomach, and coils of hind gut on the left side of stomach; B. Lateral view of the right side of stomach to show posterior sorting area; C. The coils of the hind gut to the right side of the stomach, INTERNAL MORPHOLOGY the stomach outlined by dashed lines. (For key to abbreviations The adductor muscles are more or less oval. The anterior see p.63). muscle is larger than the posterior and both have a shape similar to those of B. verrilli. The ‘quick’ and ‘catch’ portions are approximately equal in size. The gill axis is parallel to the posterior margin of the shell with 15 gill plates in a specimen Fig. 31 Brevinucula verrilli. Semidiagrammatic view of right side of body to show arrangement of organs. (Scale = 1.0 mm; see Fig. 33. Brevinucula subtriangularis. Lateral and dorsal view of a Fig. 19 for identification of parts). shell from Station 167. (Scale = 1.0 mm). STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 87 5 mm in length. The palps are moderately large and there are 28 palp ridges in the same specimen. The foot is similar in proportion and orientation to that of B. verrilli. The ‘byssal’ gland is large. The stomach is relatively small and the hind gut is extensive. The gonads are large and loosely arranged. Although this species differs in shape from B. verrilli eg. The shells are slightly longer than high, they have features in common. They have a very large resilifer, and have a non-crenulate inner ventral shell margin. Both species have rounded anterior and posterior shell margins. B. subtriangu- laris is slightly heteromyarian, and the shape of the anterior adductor muscle is similar to that of B. verrilli. Both species have a comparatively small stomach, and large, loosely arranged gonads. DISTRIBUTIONS Brevinucula subtriangularis was collected from one Station at mid-slope depth from the Brazil Basin. No comparable species are present in the Holocene formations of Guyana (Altena, 1968), even though species of Brevinucula appear to have been common in Jurassic and Triassic shallow seas. Fig. 34 Brevinucula subtriangularis. Internal view of left valve of shell from Station 167. (Scale = 1.0 mm). OTHER ATLANTIC DEEP-SEA NUCULIDS Here we list species not discussed so far in the text that have been described as having a deep-sea distribution. This latter we construe as having a population ‘peak’ below 500 metres, clearly below the shelf-slope break. We have excluded spe- cies from the Norwegian Basin in that its abyssal bivalve fauna is almost without exception self-contained (Bouchet & Warén, 1979). Nucula crenulata Adams, 1856 TYPE LOCALITY: Guadaloupe. TYPE SPECIMEN: Probable syntype, BM(NH) 1991012 (exam- ined PR). Nucula crenulata Adams, 1856; Dall, 1886, p. 247, pl. 7, Fig. 2 (in part); 1889b, p. 42, pl. 8, Fig. 2. Nucula crenulata var obliterata Dall, 1881, p. 123; Clarke, 1962, p. 48. Nucula culebrensis obliterata Johnson, 1934, p. 16. Nucula obliterata James 1972, p. 39-40, Figs 19, 20, Map 2; (non Nucula obliterata Knudsen, 1970). Dall, (1881, 1886, 1890) reported this species from the Florida Straits, the south east of the Gulf of Mexico, the southeastern Caribbean, and off Cape Hatteras from depths of 168 to 1472 metres James, (1972) also recorded the species from the Gulf of Mexico from depths of 969 to 1280 metres. N. crenulata resembles the Tertiary species Nucula striatis- sima Seguenza. Nucula culebrensis Smith, 1885 TYPE LOCALITY: ‘Challenger’ Sta. 14, Lat. 18°38’30’N, Long. 65°05’30"W, 714 m. TYPE SPECIMEN: Holotype not designated; syntypes (5 valves, BM(NH) 1887.2.9.2912-3) (examined PR). Nucula culebrensis Smith, 1885, p. 228, pl.18, fig. 11, 11a; James, 1972, p. 36-38, Figs 15, 16, Map 2. REMARKS The original description by Smith, (1885) is detailed. The species is similar in shape to Deminucula cancellata but differs in having tubercles that surround the lunule. The species originally taken off the West Indies is also recorded in the Gulf of Mexico 417-1518 metres (James, 1972), however, these latter shells were badly eroded and were thought to have been vertically displaced. Deminucula fernandinae (Dall, 1927) TYPE LOCALITY: ‘Albatross’ Sta. 2668, Lat. 30°58’N, Long. 79°38'W, 678 m. TYPE SPECIMEN: Holotype not designated, but syntypes (two pairs conjoined valves and one valve) exist: USNM No. 108198 (examined JAA). Nucula fernandinae Dall, 1927, p. 8; Johnson, 1934, p. 16. Dall, (1927) gives only a brief description but compares D. fernandinae with N. culebrensis and N. crenulata var obliter- ata. This species was described from two whole specimens and one larger valve (Dall, 1927). A single left valve of this species is also recorded by James, (1972) from the Gulf of Mexico in 2528 metres but is probably vertically displaced. This species is similar to D. atacellana and like this species has a prominent concentric sculpture and is without a well- defined resilifer. Nucula zophos Clarke, 1960 TYPE LOCALITY: ‘Alpha’ Sta. 6, Lat. 84°28’N, Long. 148°28’W, 1691 m. TYPE SPECIMEN: Holotype: Mus. Comp. Zool. Harv. No. 222067 (examined JAA). Nucula zophos Clarke, 1960, p. 5; pl. 1, Figs 15-18, 1963, 88 p. 99; Paul & Manzies 1973, p. 127; Bernard, 1979, p. 11, Fig. 2; Knudsen, 1985, p. 98. The original description by Clarke, (1960) is detailed. The species was found by Clarke, (1960) about 800 miles north of Port Barrow, Alaska, in depth ranges from 1464-1660 metres. Bernard, (1979) has also collected the species from 2377 m in the Beaufort Sea. Bernard, (1979) states that the species is widely distributed in the archibenthal of the Lau- rentian Basin. Knudsen, (1985) records the species off north- east Greenland and from 530-2237 metres—a much greater range than found by Bernard, (1979). The species appears to be restricted to the Polar and Laurentian Basins and has not been reported south of the Iceland-Faeroe ridge nor, surpris- ingly in the Norwegian abyssal basin (Bouchet & Warén, 1979). It bears a close resemblance to N. callicredemna. Pronucula benguelana Clarke, 1961 TYPE LOCALITY: R/V Vema Sta. 14, Lat. 30°14.9’S, Long. 13°03’E, 3116 m. TYPE SPECIMEN: Holotype: Mus. Comp. Zool. Harv. No. 224964 (examined JAA). Pronuncula benguelana Clarke, 1961, p. 368-369, pl. 3, Figs 9, 11. Nucula (Pronucula) benguelana Barnard, 1962, p. 446; Fig. 11a. The original description by Clarke, (1961) is detailed, but although Clarke, (1961) described this species as having a non-crenulate ventral margin, Pronucula benguelana has radial sculpturing which is typical of species with marginal crenulations and we confirm Barnard, (1962) who describes the species as having a crenulate margin. Note: Barnard regards Pronucula as a subgenus of Nucula. This species, which is not present in any of the Atlantic abyssal material reported on in this study, appears to be restricted to the Cape Basin (Clarke, 1961). It has been described by Barnard, (1962) as occurring west of Cape Point, South Africa and, according to Clarke, (1961) it is a member of a deep-sea species complex present in the south- ern Indian and Pacific Oceans. Finally brief mention should be made of Nuculoma corbu- loides (Seguenza, 1877) and Nucula striatissima (Seguenza, 1877). Both species were described from Tertiary fossils. Thereafter Recent specimens from the Mediterranean and the Eastern Atlantic are listed in Jeffreys, 1879, Locard, 1898, and Massey, 1930. These specimens are relatively few in number and taken from slope depths. Despite extensive sampling by French workers in the Bay of Biscay these species have not been recorded in recent years. ORIGIN, ANTIQUITY & DIVERSITY OF THE DEEP-SEA NUCULIDS The abyss has been regarded as a sanctuary for an archaic relict fauna (Ekman, 1953; Dahl, 1954; Zenkevitch & Birstein, 1956, 1960; Birstein, 1959, 1969; Zenkevitch, 1969) and there are many claims that ‘living fossil’ representatives of groups occur eg. Radiolaria (Haecker, 1908), Hexactinell- P.M. RHIND AND J.A. ALLEN ida (Ijima, 1927), Crinoidea (Clark, 1915), Asteroidae (Zen- kevitch & Birstein, 1956), Holothurioidea (Theel, 1882), Harpacticoida (Zenkevitch & Birstein, 1956), Mysidacea (Tattersall, 1921), Isopoda (Wolff, 1956a), Tanaidacea (Wolff, 1956b), Decapoda (Doflein, 1904; Balss, 1925, 1955), Pisces (Andraishev, 1953) and Mollusca (Locard, 1898; Bon- nevie, 1912; Zenkevitch & Birstein, 1960; Parker, 1962; Filatova, 1959; Filatova et al., 1968). The majority of deep sea bivalves belong to groups whose geological record dates back to the Ordovician (Allen, 1978). The dominant bivalve group of the deep sea, the Proto- branchia, is present in the earliest assemblages of the fossil record and protobranch genera such as Tindaria, Malletia, Neilonella, Neilo, Nuculana, Yoldia, Yoldiella and Nucula are regarded by Zenkevitch & Birstein, (1960) as ancient elements of the deep-sea fauna. In contrast relatively few protobranch species occur in shelf seas and these are almost entirely restricted to the families Nuculidae and Nuculanidae (Allen, 1978). Of the divisions of the Nuculidae as defined by Schenck, (1939) (p. 64) two occur in the deep Atlantic namely those with crenulate shell margins and those with smooth shell margins. Species with divaricate shell sculpture do not occur (nor do they occur in shallow waters of the Atlantic). ~ Species with smooth margins are found throughout geo- logic time. One of the earliest from the Pennsylvanian of Iowa, Nuculopsis ventricosa (Hall) has a shell morphology that differs little from Recent species (Schenck, 1939). Spe- cies of Nuculoidea date from the Silurian and Devonian (McAlester, 1962; Soot-Ryen, 1964). Nuculoma dates from the Jurassic (Cossmann & Thiery, 1907). In contrast crenu- late species have a far less protracted geologic history and extend no further back than the Cretaceous (a possible exception is the Mississippean Carboniferous species Nucula schumardiana (Cox, 1940; Vokes, 1949). Vokes, (1949) also reported the Palaeozoic species ‘Nucu- loidea opima Hall’ as having microscopic crenulations. Because this did not fit into the Schenck, (1939) classification Vokes, (1949) suggested that such species should be placed in a separate group from the crenulate and non-crenulate forms. This includes the species of Brevinucula. Moore, (1969) states that Brevinucula does not appear in the geological record prior to the Miocene. Thus, the present study supports the concept of antiquity. Species with non-crenulate inner ventral shell margins are descended from a more-archaic group than species with a crenulate inner ventral margin and the former, which have representatives in the Ordovician, are far the more abundant in the deep sea. In contrast, the crenulate species which did not appear in the fossil record until the Cretaceous are more abundant in shallow water. Of 12 shallow-water nuculids of the North Atlantic, 4 have non-crenulate shell margins, whereas of 8 deep-sea Atlantic species (not including Brevi- nucula) 6 have non-crenulate margins. As Knudsen, (1970) suggests, there probably has been a descent of shallow-water species into the abyssal zone throughout geological history. Therefore it would be expected that along with species of deep-water origin there exist deep-water species closely related to shallow-water species. Zenkevitch & Birstein, (1960) suggest that it is possible to distinguish between ancient and secondary deep-water species by their patterns of distribution. In the case of the secondary species, their diversity decreases with increasing depth whereas the ancient groups tend to show increasing diversity with increasing STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 89 depth which diminishes only when encroaching lower abyssal depths or the perturbed trenches and vent systems. The distribution pattern of deep-sea ‘non-crenulate’ nuculids would suggest that they are an ancient rather than a second- ary deep-water group. It has been assumed that one of the more important factors restricting downward migration is temperature and that polar coastal regions yield species of the necessary eurybathic capacity to invade the deep sea (Bruun, 1956; Kussakin, 1973). The close affinity between the Arctic shallow-water species, N. bellotii and Nuculoma similis and the Greenland species N. corticata with N. granulosa would appear to support this conclusion. Kussakin, (1973) noted that primitive isopods are princi- pally confined to tropical waters, whereas isopods of cold and temperate regions and especially the deep-sea are considered phylogenetically recent. He speculated that tropical faunas gave rise to temperate faunas which then gave rise to cold- water faunas and that from the cold-water faunas of the Antarctic shelf came the initial and main source of deep-sea faunas. The adaptation of the shelf fauna to deep-sea condi- tions was thought to have been promoted by climatic cooling and subsequent glaciation which occurred in the Miocene and approximately a million years earlier than in the northern hemisphere, and therefore Antarctica became the sole zone supplying cold oxygen bearing water into the abyssal zone. The thickening ice was thought to cause the gradual isostatic dipping of the continent and shelf by as much as 900 m and so the Antarctic shelf-fauna was slowly forced deeper. The movement of cold Antarctic water into the abyss and then northwards would have then facilitated subsequent distribu- tion. In contrast to Kussakin, (1973), Hessler et al., (1979) also on the basis of biogeographical studies of deep-sea isopods, argue that the presence of species closely related to deep-sea species in shallow high latitudes is the result of subsequent emergence and that many deep-sea species evolved in situ. Curiously, although evidence suggests that a few deep-sea nuculids had their origin in the Arctic region, there is no indication that a similar process occurred in the Antarctic. There is a dearth of nuculids in the deep Antarctic (Dell, 1972) and to date only N. notobenthalis Thiele, has been described and this does not appear to be similar to any other deep-sea Atlantic species. It is perfectly feasible that ‘non- crenulate’ nuculids, which are more diverse in the deep sea, may have given rise to shallow water species such as Nucu- loma tenuis and N. bellotii. Some abyssal genera are cosmo- politan and thus they are potentially available for emergence into any suitable shallow-water environment (Hessler et al., 1979). This would explain the ubiquity of N. tenuis. There are two strict shallow-water nuculids in the Arctic (Clarke, 1960; Bernard, 1979), 7 or 8 species occur in the temperate zone (5 on the east coast of the North Atlantic (Allen, 1954), and 3 or 4 on the west coast (Mighels & Adams, 1842; Hampson, 1971; Abbott, 1974). Six species occur in the western tropical region between the Caribbean and Surinam (Dautzenberg, 1900; Weisbord, 1964; Altena, 1968; Moore, 1977). There does not appear to be any certain information relating to the number of species present in the eastern tropical Atlantic, but there is likely to be a further assemblage of species. On this basis nuculid diversity would appear to increase slightly towards the tropics. Prior to the radiation of the other major shallow water bivalve deposit- feeding group the Tellinacea, in the middle Cretaceous nuculid diversity in the tropics was probably far greater and, since the ‘crenulate’ nuculid species did not evolve until the Cretaceous, the assemblages would presumably have been mainly composed of ‘non-crenulate’ forms. Tellinaceans are not present in the abyss and only two species are present at lower slope depths (Allen & Sanders, 1966), but massive radiation of tellinaceans may have resulted in the extinction of the shallow water tropical nuculid predecessors and may eventually lead to the loss of others in higher latitudes and the deep-sea. This totally ignores physiological aspects which presently constrain the colonization of the deep-sea by telli- naceans and favour the nuculacean presence there (Allen, 1978). Sanders & Hessler, (1969) found that in general, diversity in abyssal basins was of a similar order of magnitude to that found in some shallow tropical seas and well exceeds that in cold temperate shelf seas. Unlike the nuculanids the vertical diversity of the nuculids does not conform to this generaliza- tion. Although, there is an increase in the diversity of ‘non-crenulate’ species with depth, overall nuculid diversity decreases. For example, on the temperate. North East Atlan- tic shelf (<200 m) there are 5 species (Allen, 1954), whereas in the West European Basin (>1500 m) there are only 3 species. In the western tropics around Guyana the disparity is 7 shelf-species (Altena, 1968; Moore, 1977) and 4 abyssal species. In the temperate western Atlantic there are 4—5 shelf species (Mighels & Adams, 1842; Hampson, 1971; Abbott, 1974) as compared with 5 abyssal species in the North America Basin. The reduction in nuculid diversity with increasing depth in the north east and tropical west Atlantic may be related to the fact that ‘crenulate’ species were not present in shallow water before the Cretaceous and therefore only comparatively recently have they invaded the deep sea. DISCUSSION The species described here are based on morphological characters and therefore would be regarded by numerical taxonomists as morpho-species rather than biological species as defined by Mayr, (1949). There is no biological reason why inability to interbreed should always be traceable to morpho- logical differences (Savory, 1962), and classifications based on similarity alone may not always reflect genetic affinity. Nevertheless, the morphological differences between intra- basin species are sufficiently large to more-or-less exclude the possibility that they are variants of a single species. Widely distributed species such as D. atacellana with numerous populations as far apart as the Rockall Trough and the Argentine Basin (approximately 7000 miles) may turn out to be more than one species on the basis of, say, electrophoretic evidence. Overall the deep-sea bed is not subject to large amounts of environmental heterogeneity, consequently it is hypothesized that populations in this environment would maintain little genetic variability (Manwell & Baker, 1970; Grassle, 1972; Grassle & Sanders, 1973). In fact deep-sea species are far from genetically depauperate and may main- tain levels of genetic variability which are average or above average for marine invertebrates (Gooch & Schopf, 1972; Ayala & Valentine, 1974; Valentine & Ayala, 1975; Ayala et al., 1975; Murphy et al., 1976), and have as much if not more potential for change as organisms from other environments. 90 Although the nuculids are preadapted for deep-sea exist- ence (eg. feeding and shell composition) their life in deep water has been accompanied with adaptation. This includes reduction in the area of the gill and number of plates, the elongation of the hindgut, a reduction in the number of ciliated grooves in the stomach, an increase in egg size and a reduction of body size and low productivity. Probably all are adaptations to nutritional impoverishment but are also likely to be accompanied by physiological adaptation to low tem- perature and high pressure. Nevertheless, none of the Atlan- tic deep-sea nuculids can be regarded as truly abyssal since no species reaches an optimal population density at abyssal depths (ie. >4000 m). Only two species, B. verrilli and D. atacellana, have distributions which exceed 4000 m and both of these are also found at slope depths. Indeed, most species are restricted to the continental slope with diversity reaching a maximum between 1000-2000 m. Reduction in diversity as abyssal depths are approached is recorded in a number of macrofaunal groups (Sanders et al., 1965; Rex, 1973, 1976) and may be a result of extremely low productivity at these depths. One adaptation to low productivity is the reduction of body size, but there must be a limit to this particular adaptation. Rex, (1973) suggests that productivity affects macrofaunal diversity only when the lower limit of adaptation in size is approached. Co-existing protobranchs having closely similar feeding habits and apparently identical food in their guts has to be explained against the apparent lack of potential niches within the soft monotonous abyssal sediment. Originally ‘niche’ was defined as the subdivision of the environment within which the species lives (Grinnell, 1917). Hutchinson, (1957) refor- mulated the concept in terms of set theory such that if each environmental variable is given a co-ordinate in an N-dimensional space, then a niche can be defined as a multi-dimensional hypervolume in which the fitness of the individual is positive. Real separation of species may be based mainly on trophic rather than physical or chemical factors (Green, 1971), and this may well apply in the deep-sea where if protobranch niche demarcation does exist in the deep-sea environment, it appears to be very subtle, and may be a product of biochemical specialization and differing spatial requirements (Grassle & Sanders, 1973; Allen, 1983). The distribution of cosmopolitan species such as Deminu- cula atacellana and Nuculoidea subovata seems to be attribut- able to the production of pelagic larvae. Species which have direct development, such as Nuculoma similis and Nucula delphinodonta have restricted distributions. Pelagic larvae of deep-sea protobranchs have never been collected from sur- face waters, therefore, transportation must take place in deep water. The velocity of deep-sea currents range from 1.5 to 44 cm/sec (Knauss, 1965; Webster, 1969; Schmitz et al., 1970; Hogg, 1983; Saunders, 1983), and if the duration of larval transport is only a few days, larvae could in theory be transported many kilometres. Pelagic larvae are not only important as a means of dispersal but also for maintaining genetic continuity. The lack of morphological differences between widely separated popu- lations of D. atacellana and N. subovata suggest frequent genetic exchange between these populations and that long- lived larvae exist in the deep sea. Discontinuities in the distribution of deep-sea nuculids are difficult to explain but could be related to a dispersal controlled by prevailing currents and ridge systems. The results of this study and that of Bouchet and Warén, P.M. RHIND AND J.A. ALLEN (1979) show that none of the North Atlantic abyssal nuculid species occur north of the North Atlantic Transversal Ridge. Although both D. atacellana and N. subovata are common in the Rockall Trough, adjacent to the ridge, neither are found in depths less than 1000 m and it must be that their larvae are unable to travel north into the Norwegian Basin against the prevailing deep cold water currents originating in high lati- tudes (Lynn & Reid, 1968). There also appear to be no nuculids common to the abyss of both the Atlantic and the Antarctic (Soot-Ryen, 1951; Powell, 1960, 1965; Dell, 1964, 1972) even though there appears to be no barrier. The Cape Basin also appears to have few affinities with the rest of the Atlantic as far as nuculids are concerned. There are no species common to the Cape Basin and the adjacent Angola Basin, yet two species found in the Angola Basin are found in other Basins on both sides of the equator. Clarke, (1960) only found one nuculid species, Pronucula benguelana, in the Cape Basin, and neither this species nor the genus is found in any other Atlantic Basin. REFERENCES Abbott, R.T. 1974. American sea shells: the marine Mollusca of the Atlantic & Pacific Coasts of North America. 663 p. Van Nostrand Reinhold Co, New York. Adams, A. 1856. Description of thirty-four new species of bivalve molluscs (Leda, Nucula, Pythia) from the Cunningham collection. Proceedings of the Zoological Society of London. 24: 47-53. Adams, J. & Adams, A. 1858. The genera of Recent Mollusca. Vol. Il, 659 p. Van Voorst, London. Allen, J.A. 1954. A comparative study of the British species of Nucula and Nuculana. Journal of the Marine Biological Association of the U.K. 33: 457-472. 1978. Evolution of the deep-sea protobranch bivalves. Philosophical Transactions of the royal Society of London B., 284: 387-401. 1983. The ecology of deep-sea molluscs. pp. 29-75. In: Russell-Hunter, W. (Ed.) The Mollusca, 6. Academic Press, Orlando. —— 1985. The Recent Bivalvia: Their Form and Function. pp. 337-403. In: Trueman, E. & Clarke, M. (Eds) The Mollusca, 10. Academic Press, Orlando. — & Hannah, F.J. 1986. A reclassification of the Recent genera of the Subclass Protobranchia (Mollusca: Bivalvia). Journal of Conchology, 32: 225-249. —— & —— 1989. Studies on the deep sea Protobranchia. The Subfamily Ledellinae (Nuculanidae). Bulletin of the British Museum (Natural History), Zoology 55: 123-171. —— & Sanders, H.L. 1966. Adaptations to abyssal life as shown by the bivalve, Abra profundorum. Deep Sea Research, 13: 1175-1184. — & Sanders, H.L. 1969. Nucinella serrei (Protobranchia) a monomyarian solemyid and possible living actinodont. Malacologia, 7: 381-396. — & Sanders, H.L. 1973. Studies on the deep sea Protobranchia. The families Siliculidae and Lametilidae. Bulletin of the Museum of Comparative Zoology Harvard, 145: 263-310. — & Sanders, H.L. 1982. Studies on the deep sea Protobranchia. The subfamily Spinulinae (Family Nuculanidae). Bulletin of the Museum of Comparative Zoology Harvard, 150: 1-30. Altena, C.O. 1968. The Holocene and Recent marine bivalve Mollusca of Surinam. Studies on the Fauna of Curacao, 10: 153-179. Andriashev, A.P. 1953. Archaic deep sea and secondary deep sea fishes and their role in zoogeographical analysis. pp. 58-64. In: Tarasov, N.I. (Ed.) Outlines on the general problems of ichthyology, Moscow. (In Russian). Atkins, D. 1938. On the ciliary mechanisms and inter-relationships of lamelli- branchs. Part 7. Latero-frontal cilia of the gill filaments and their phyloge- netic value. Quarterly Journal of Microscopical Science, 80: 346-436. Aumonier, F.J. 1938. Notes on the distortion of paraffin sections. Journal of the Royal Microscopical Society, 58: 253-257. Ayala, F.J. & Valentine, J.W. 1974. Genetic variability in a cosmopolitan deep water ophiuran, Ophiomussium lymani. Marine Biology, 27: 51-57. ——,, Valentine, J.W., Hedgecock, D. & Barr, L.G. 1975. Deep sea asteroids: high genetic variability in a stable environment. Evolution, 29: 203-212. Babin, C. 1977. Etude comparée des genres Babinkia Barrande et Coxiconcha STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 91 Babin (Mollusques bivalves de l’Ordovicien intérét phylogenique. Geobios, No. 10, fasc 1: 51-79. Balss, H. 1925. Macrura der Deutschen Tiefsee-Expedition 1. Palinura, Astacura, und Thalassinidea. Wissenschaftliche Ergebnisse der Deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’, 20: 189-216. — 1955. Decapoda, VI. Okologie. Bronn’s Klassen und Ordnungen des Tierreichs, V. Abt I, 7.10: 1285-1367. Barnard, K.H. 1962. New species and records of South African marine Mollusca from Natal, Zululand and Mocambique. Annals of the Natal Museum, 15: 247-254. Bell, R. 1879. Report on the country between Lake Winnipeg and Hudson Bay, 1878. Progress Reports of the Geological Survey Branch of Canada, 1877-78: 1-31. Bernard, F.R. 1979. Bivalve mollusks of the western Beaufort Sea. Contribu- tions in Science, 313: 1080. Birstein, J.A. 1959. Paleotemperatures and the problem of deep-sea fauna origin (In Russian). Priroda, 5: 21-28. — 1969. Some aspects of the evolution of the deep-sea fauna (In Russian). Okeanologia, 9: 907-908. Bonnevie, K. 1912. Pteropoda from the ‘Michael Sars’ North Atlantic deep sea expedition, 1910. Reports of the Scientific Results of the Michael Sars North Atlantic Deep Sea Expedition 1910, 3(1) Zoology: 1-86. Bouchet, P.H. & Warén, A. 1979. the abyssal molluscan fauna of the Norwegian Sea and its relation to other faunas. Sarsia, 64: 211-243. Bradshaw, M.A. 1970. The dentition and musculature of some middle Ordovi- cian (Llandeilo) bivalves from Finistere, France. Palaeontology, 13: 623-645. Brunn, A. 1956. The abyssal fauna: its ecology, distribution and origin. Nature, 177: 1105-1108. Caspers, H. 1940. Uber Nahrungserwerb und Darmverlauf bei Nucula. Zoolo- gischer Anzeiger, 129: 48-55. Clark, A.H. 1915. A study of Recent crinoids which are congeneric with fossil species. American Journal of Science, 4: 60-66. Clarke, A.H. Jr. 1960. Arctic archibenthal and abyssal mollusks from drifting station Alpha. Breviora, 119: 1-17. — 1961. Abyssal mollusks of the South Atlantic Ocean. Bulletin of the Museum of Comparative Zoology Harvard, 125: 345-387. — 1962. Annotated list and bibliography of the abyssal marine molluscs of the world. Bulletin. National Museum of Canada, 181: 1-114. Cossmann, M. & Thiery, P. 1907. Notes sur le Callovien de Haute-Marne et spécialement sur un gisement situé dans la commune de Bricon. VI Congres de l’Association Franc-Comtoise (Vesoul 1906) Société d’Agriculture Lettres Sciences et Arts de la Haute-Saodne: 55-56. Cotton, B.C. 1930. Pelecypoda of the ‘Flindersian’ region, southern Australia. Records of the South Australian Museum 4: 223-240. Cox, L.R. 1940. The Jurassic lamellibranch fauna of Kuchh (Cutch). Memoirs of the Geological Survey of India, Palaeontologia indica, 9 (3): 9-25. —— 1959. The geological history of the Protobranchia and the dual origin of taxodont Lamellibranchia. Proceedings of the Malacological Society of London, 34: 200-209. Dahl, E. 1954. The distribution of deep-sea Crustacea. Jn: On the distribution and origin of deep-sea bottom fauna. Union Internationale des Sciences Biologiques, Ser. B: 43-48. Dall, W.H. 1881. Report on the results of dredging by the United States Coast Survey Steamer ‘Blake’. 15. Preliminary report on the Mollusca. Bulletin of the Museum of Comparative Zoology Harvard, 9: 33-144. — 1886. Reports on the results of dredging under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80) by its coast survey steamer ‘Blake’, Lieut-Commander C.D. Sigsbee, U.S.N. and Commander J.R. Bartlett, U.S.N. commanding. XXIX Report on the Mollusca, Part I, Brachiopoda and Pelecypoda. Bulletin of the Museum of Comparative Zoology Harvard, 12: 171-318. — 1889. A preliminary catalogue of shell bearing marine mollusks and brachiopods of the Southeastern Coast of the United States with illustrations of many of the species. Bulletin of the U.S. National Museum, 37: 1-221. — 1890. Scientific results of explorations by the United States Fish Commis- sion Steamer ‘Albatross’. 7. Preliminary report on the collection of Mollusca and Branchiopoda obtained in 1887-1888. Proceedings of the U.S. National Museum 12: 219-362. — 1927. Small shells from dredgings off the southeast coast of the United States by the United States Fisheries Steamer ‘Albatross’ in 1885 and 1886. Proceedings of the U.S. National Museum 70 (18): 1-134. Dautzenberg, P. 1900. Croisiéres du yacht Chazalie dans l’Atlantique. Mol- lusques. Mémoires de la Société Zoologique de France, 13: 145-365. —— & Fischer, H. 1897. Dragages effectues par L’Hirondelle et par la Princesse-Alice. Mémoires de la Société Zoologique de France, 10: 139-234. Dell, R.K. 1964. Antarctic and Sub-Antarctic Mollusca: Amphineura, Sca- phopoda, Bivalvia. ‘Discovery’ Reports, 33: 99-250. —— 1972. Antarctic benthos. Advances in Marine Biology, 10: 1-216. Dolfein, F. 1904. Brachyura. Wissenschaftliche Ergebnisse der Deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’, 6: 1-314. Drew, G.A. 1901. The life history of Nucula delphinodonta. Quarterly Journal of Microscopical Science, 44: 313-391. Driscoll, E.G. 1964. Accessory muscle scars, an aid to protobranch orientation. Journal of Paleontology, 38: 61-66. Ekman, S. 1953. Zoogeography of the sea. 471 p. Sidgwick and Jackson, London. Farrow, G.E., Atkinson, J. & Moore, P.G. 1984. SAFE: 1983 Macrobenthos and bioturbation. In: Sedimentology of Arctic fjords. Experiment: H.U.83—028. Data report, 2. (compiler: J.P.M. Syvitski). Canadian data report of hydrology and ocean sciences, 28: 6.1-6.35. Filatova, Z.A. 1948. Class of bivalve molluscs. pp. 405-446. In: Gajevskaia, N.S. (Ed.) Keys to the fauna and flora of the Northern Seas of the U.S.S.R., Moscow. — 1959. Bivalve molluscs of the abyssal zone of the north-western Pacific. Proceedings of the Fifteenth International Congress of Zoology, London (1958): 221-222. ——,, Sokolova, M.N. & Levenstein, R.Y.A. 1968. Mollusca of the Cambro- Devonian class Monoplacophora found in the Northern Pacific. Nature, 220: 1114-1115. Fischer, P. 1886. Manuel de conchyliologie et de paléontologie conchyliologique Histoire naturelle des mollusques vivants et fossiles: pt. 10: 897-1008, Paris. Forbes, E. 1844. Report on Mollusca and Radiata of the Aegean Sea, and on their distribution, considered as bearings on Geology. Report of the British Association of the Advancement of Science for 1843: 130-193. Gooch, J.L. & Schopf, T.J.M. 1972. Genetic variability in the deep sea: relation to environmental variability. Evolution, 26: 545-552. Gould, A.A. 1870. Report on the Invertebrates of Massachusetts. 2nd ed., 524 p. W.G. Binney, Boston. Gower, J.C. 1966. Some distance properties of latent root and vector methods used in multivariate analysis. Biometrika, 53: 325-358. Graham, A. 1949. The molluscan stomach. Transactions of the Royal Society of Edinburgh, 61: 737-778. Grassle, J.F. 1972. Species diversity, genetic variability and environmental uncertainty. pp. 19-26. Jn: Battaglia, B. (Ed.) Fifth European Marine Biological Symposium, Piccin Editone, Padua. & Sanders, H.L. 1973. Life histories and the role of disturbance. Deep Sea Research, 20: 643-659. Grause, R.R. 1974. Latitudinal trends in the shell characteristics of marine gastropods. Lethaia, 7: 303-314. Green, R.H. 1971. A multivariate statistical approach to the Hutchinsonian Niche: bivalve molluscs of central Canada. Ecology, 52: 543-556. Greig, J.A. 1920. Brachiopoda, Scaphopoda, Gastropoda and Lamellibranchi- ata. Reports of the Scientific Results of the Michael Sars North Atlantic Deep Sea Expedition, 3: 1-16. Grinnell, J. 1917. Field tests of theories concerning distribution control. American Naturalist 51: 115-128. Haecker, V. 1908. Tiefsee Radiolarien. Wissenschaftliche Ergebnisse der Deut- schen Tiefsee Expedition auf dem Dampfer ‘Valdivia’, 14: 1-336. Halton, D.W. & Owen, G. 1968. the fine structure and histochemistry of the gastric cuticle of the protobranchiate bivalve Nucula sulcata Bronn. Proceed- ings of the Malacological Society of London, 38: 71-81. Hampson, G.R. 1971. A species pair of the genus Nucula (Bivalvia) from the Eastern coast of the United States. Proceedings of the Malacological Society of London, 38: 333-342. Hancock, A. 1846. A list of shells dredged on the west coast of Davis Strait; with notes and descriptions of 8 new species. Annals and Magazine of Natural History (1), 18: 323-338. Heath, H. 1937. The anatomy of some protobranch mollusks. Mémoires du Musée r. d’Histoire naturelle de Belgique, ser. 2, 10: 1-26. Hedley, C.W. 1902. Mollusca, Part 1. Brachiopoda and Pelecypoda (Scientific Results of the trawling expedition of H.M.C. ‘Thetis’). Memoirs of the Australian Museum, Sydney IV, 1: 290-291. Hessler, R.R., George, D.W. & Thistle, D. 1979. The deep sea isopods: a biogeographic and phylogenetic overview. Sarsia, 64: 67-75. Hirasaka, K. 1927. Notes on Nucula. Journal of the Marine Biological Association of the U.K. 14: 629-645. Hogg, N.G. 1983. A note on the deep sea circulation of the western North Atlantic: its nature and causes. Deep Sea Research, 30: 945-961. Humason, G.L. 1967. Animal tissue techniques. 468 p. W.H. Freeman and Company. San Francisco and London. Hutchinson, G.E. 1957. Concluding remarks. Cold Spring Harbor Symposium on quantitative Biology 22: 415-427. Ijima, I. 1927. The Hexactinellida of the Siboga Expedition. Siboga-Expeditie, 6(101): 1-384. Iredale, T. 1931. Australian molluscan notes 1. Records of the Australian Museum, 18: 201-235. Jackson, R.T. 1890. Phylogeny of the Pelecypoda. The Aviculidae and their allies. Memoirs of the boston Society of Natural History 4: 277-400. 92 James, B.M. 1972. Systematics and biology of the deep sea Palaeotaxodonta (Mollusca: Bivalvia) from the Gulf of Mexico. Ph.D. Thesis submitted to the Graduate College of Texas A & M University. Jeffreys, J.G. 1876. Preliminary reports of the biological results of a cruise in H.M.S. ‘Valorous’ to Davis Straits in 1875. Proceedings of the Royal Society of London, 25: 177-237. Jeffreys, J.G. 1879 & 1881. On the Mollusca procured during the ‘Lightning’ and ‘Porcupine’ Expeditions 1868-1870. Part II: Proceedings of the Zoologi- cal Society of London 1879, 553-558; Part III: Proceedings of the Zoological Society of London, 1881, 693-724. Kauffmann, E.G. 1969. Form, function and evolution. pp. N130-N205. In: Moore, R.G. (Ed.) Treatise on invertebrate paleontology (N) Mollusca 6, Bivalvia, 1 & 2. Geological Society of America and University of Kansas Press. Knauss, J.A. 1965. A technique for measuring deep ocean currents close to the bottom with an unattached current meter and some preliminary results. Journal of Marine Science, 23: 237-245. Knudsen, J. 1970. The systematics and biology of the abyssal and hadal Bivalvia. Galathea Reports, 11: 1-241. — 1985. Abyssal Mollusca of the Arctic Ocean. Journal of Conchology, 32: 97-107. Krause, A. 1885. Ein Beitrag zur Kenntnis der Mollusken Fauna des Bering- meeres. I. Brachipoda et Lamellibranchiata. Archiv fiir Naturgeschichte 51: 14-40. Kussakin, O.G. 1973. Peculiarities of the geographical and vertical distribution of marine isopods and the problem of deep sea fauna origin. Marine Biology, 23: 19-24. Levinton, J.S. & Bambach, R.K. 1975. A comparative study of Silurian and Recent deposit feeding bivalve communities. Paleobiology, 1: 97-124. Liljdahl, L. 1983. Two silicified Silurian bivalves from Gotland. Sveriges Geologiska Undersékning. Afhandlinger. Ser. C., Nr. 799: 1-51. — 1984. Silurian silicified bivalves from Gotland. Sveriges Geologiska. Undersékning Afhandlinger Ser. C., Nr. 804: 1-82. Locard, A. 1898. Mollusques testacés. Expeditions scientifiques du Travailleur et du Talisman, 2: 1-515. Lubinsky, I. 1972. Canadian Arctic bivalves. Ph.D. Thesis submitted to the faculty of Graduate Studies and Research, McGill University. Lynn, R.J. & Reid, J.L. 1968. Characteristics and circulation of deep and abyssal waters. Deep Sea Research, 15: 577-598. McAlester, A.L. 1962. Upper Devonian pelecypods of the New York Chemung stage. Bulletin of the Peabody Museum of Natural History, 16: 1-23. — 1963. Revision of the type species of the Ordovician nuculoid pelecypod genus Tancrediopsis. Postilla, 74: 1-19. —— 1964. Preliminary suggestions for a classification of nuculoid bivalves. Journal of Paleontology, 38: 397-400. — 1969. In: Moore, R.C. (Ed.) Treatise on invertebrate paleontology, Part N. Mollusca 6, Bivalvia 1 & 2. Geological Society of America & University of Kansas Press. Macginite, N. 1959. Marine mollusca of Point Barrow, Alaska. Proceedings of the U.S. National Museum, 109: 59-208. Marwick, J. 1931. The Tertiary Mollusca of the gisbourne District. Palaeonto- logical Bulletin. Geological Survey Wellington, No. 13: 1-177. Massey, A.L. 1930. Mollusca (Pelecypoda, Scaphopoda, Gastropoda, Opistho- branchia of the Irish Atlantic slope, 50-1500 fathoms. Proceedings of the Royal Irish Academy 39B: 232-342. Manwell, C. & Baker, C.M.A. 1970. Molecular biology and the origin of species. 394 p. University of Washington Press, Seattle. Mayr, E. 1949. Systematics and the origin of species. 334 p. Columbia Biologi- cal series, Columbia University. Métivier, B. 1982. Description d’espéces abyssales de Nuculacea des Acores. Journal of Conchology, 31: 39-44. Mighels, J.W. & Adams, C.B. 1842. Description of 24 species of the shells of New England. Boston Journal of Natural History 4: 37-54. Moller, H.P.C. 1842. Index molluscorum Groenlandiae. Naturhistorisk Tidsskrift 4: 76-97. Montagu, G. 1803-1808. Testacea Britannica. 2 vols. Supplement London. Moore, D.R. 1977. Small species of Nuculidae (Bivalvia) from the tropical Western Atlantic. Nautilus, 91: 19-128. Moore, H.B. 1931a. The systematic value of the study of molluscan faeces. Proceedings of the Malacological Society of London, 19: 281-290. 1931b. The specific identification of faecal pellets. Journal of the Marine Biological Association of the U.K., 17: 359-365. Moore, R.C. 1969. Treatise on invertebrate paleontology (N) Mollusca 6. Geological Society of America & University of Kansas Press. Morris, N.J. & Fortey, R.A. 1976. the significance of Tironucula gen. nov. to the study of bivalve evolution. Journal of Palaeontology 50: 701-709. Murphy, L.S., Rowe, G.T. & Haedrich, R.L. 1976. Genetic variability in deep sea echinoderms. Deep Sea Research, 23: 339-348. P.M. RHIND AND J.A. ALLEN Nichol, D. 1964. Lack of shell attached peoecypods in Arctic and Antarctic waters. Nautilus, 77: 92-93. —— 1965. Ecological implications of living pelecypods with calcareous spines. Nautilus 78: 109-116. —— 1966. Size of pelecypods in Recent marine faunas. Nautilus, 79: 109-113. — 1967. Some characteristics of cold-water marine pelecypods. Journal of Paleontology, 41: 1330-1340. Ockelmann, K.W. 1958. The zoology of East Greenland. Marine Lamellibran- chiata. Meddelelser om Grgnland, 122: 1-258. —— 1965. Development types in marine bivalves and their distribution along the Atlantic coast of Europe. pp. 25-35. In: Cox, L.R. & Peake, J. (Eds) Proceedings of the first European Malacological Congress, London, 1962. Owen, G. 1956. Observations on the stomach and digestive diverticula of the Lamellibranchia. II Nuculidae. Quarterly Journal of Microscopical Science, 97: 541-567. Pantin, C.F.A. 1946. Notes on microscopical techniques for zoologists. 75 p. Cambridge University Press. Parker, R.H. 1962. Speculation on the origin of the invertebrate faunas of the lower continental slope. Deep Sea Research, 8: 286-293. Pelseneer, P. 1891. Contribution a l’étude des lamellibranchs. Archives de Biologie, 11: 147-312. — 1911. Les Lamellibranches de l’expédition du Siboga, Partie Anatomique. Siboga-Expeditie 71: 1-125. Pequegnat, W.E. 1972. A deep bottom current on the Mississippi Cone. pp. 65-87. In: Capurro, R.H. & Reid, J.L. (Eds) Contributions on the physical oceanography of the Gulf of Mexico. Texas A & M University Oceanographic Studies 2. Gulf Publishing Co, Houston. Poel, L. Van Der 1955. Structure du test et classification des nucules. Bullertin de I’ Institut r. Science Naturelle Belgique, 11: 1-68. a Pollak, O.J. 1944. A rapid trichrome stain. Archives of Pathology, 37: 294. Powell, A.W.B. 1960. Antarctic and Sub Antarctic Mollusca. Records of the Auckland Institute Museum, 5: 115-193. — 1965. Mollusca of the Antarctic and Subantarctic seas: Biography and Ecology in Antarctica. Monographiae Biologicae, 15: 333-380. Pusey, H.K. 1939. Methods of reconstruction from microscope sections. Journal of the Royal Microscopical Society, 59: 232-244. Reeve, L. 1885. Account of the shells collected by Capt. Sir Edward Belcher, north of Beechey Island. pp. 392-399. In: Belcher, E. (Ed.) The last Arctic voyages, 2. Rex, M.A. 1973. Deep-sea species diversity: decreased gastropod diversity at abyssal depths. Science, 181: 1051-1053. — 1976. Biological accommodation in the deep sea benthos: comparative evidence on the importance of predation and productivity. Deep Sea Research, 23: 957-987. Rhoads, D.C. & Young, D.K. 1970. The influence of deposit-feeding organisms on sediment stability and community trophic structure. Journal of the Marine Research, 28: 150-178. Sanders, H.L., Hessler, R.R. & Hampson, G.R. 1965. An introduction to the study of the deep-sea benthic faunal assemblages along the Gay Head, Bermuda transect. Deep Sea Research, 12: 845-867. —— & —— 1969. Ecology of the deep-sea benthos. Science, 163: 243-282. — & Allen, J.A. 1973. Studies on deep-sea Protobranchia (Bivalvia); Prologue and the Pristiglomidae. Bulletin of the Museum of comparative Zoology Harvard, 145: 237-261. — & 1977. Studies on the deep-sea Protobranchia (Bivalvia); the family Tindariidae and the genus Pseudotindaria. Bulletin of the Museum of Comparative Zoology Harvard, 148: 23-59. —— & —— 1985. Studies on deep-sea Protobranchia (Bivalvia); the family Malletiidae. Bulletin of the British Museum (Natural History), Zoology, 49: 195-238. Saunders, P.M. 1983. Benthic observations on the Madeira Abyssal Plain: currents and dispersion. Journal of Physical Oceanography, 13: 1416-1429. Savory, T. 1962. Naming the living world: an introduction to the principles of biological nomenclature. 128 p. English University Press. Scheltema, R.S. 1972. Reproduction and dispersal of bottom dwelling deep sea invertebrates: a speculative summary. pp. 58-68. Jn: Brauer, R.W. (Ed.) Barobiology and the experimental biology of the deep sea. N. Carolina Sea Grant Programme. University North Carolina, Chapel Hill. Schenck, G.H. 1934. The classification of nuculid pelecypods. Bulletin de la Musée r. Science Naturelle du Belgique, 10: 29-31. — 1939. Revised nomenclature for some nuculid pelecypods. Journal of Paleontology, 13: 21-41. —— 1944. Lamellinucula a new subgenus of the nuculid pelecypods. Journal of Paleontology, 18: 97-99. Schmitz, W.J., Robinson, A.R. & Fuglister, F.C. 1970. Bottom velocity observations directly under the Gulf Stream. Science, 170: 1192-1194. Seguenza, G. 1877. Nuculidi terziarie rinvenute nelle provincie meridionali d'Italia. Atti dell’ Accademia del Lincei Memorie, ser 3, 1: 1163-1190. STUDIES ON THE DEEP-SEA PROTOBRANCHIA (BIVALVIA): THE FAMILY NUCULIDAE 93 Smith, E.A. 1885. Report on the Lamellibranchiata collected by H.M.S. ‘Challenger’. Zoologist, 13: 1-341. — 1891. Descriptions of new species of shells from the ‘Challenger’ Expedition. Proceedings of the Zoological Society of London, 1891: 436-445. Sokal, R.R. & Sneath, P.H.A. 1963. Principles of Numerical Taxonomy. 359 p. W.H. Freeman & Co, San Francisco & London. Soot-Ryen, H. 1964. Nuculoid pelecypods from the Silurian of Gotland. Arkiv f6ér Mineralogi och Geologi, 3: 489-519. Soot-Ryen, T. 1932. Pelecypoda, with a discussion of possible migrations of Arctic pelecypods in Tertiary times. Scientific Results of the Norwegian North Polar Expedition with the Maud, 5: 1-35. — 1951. Antarctic pelecypods. Scientific Results of the Norwegian Antarctic Expeditions, 3: 1-46. — 1966. Revision of the pelecypods from the Michael Sars North Atlantic deep-sea expedition 1910. Sarsia, 24: 1-31. Sowerby, G.B. 1871. Monograph on the genus Nucula. In: Conchologia Iconica, 18: Reeve & Co., London. Stanley, S.M. 1968. Post palaeozoic adaptive radiation of infaunal bivalve molluscs—a consequence of mantle fusion and siphon formation. Journal of Palaeontology, 42: 214-229. Steedman, H.F. 1957. Polyesterwax. A new ribboning embedding medium. Nature, 179: 1345. Tattersall, W.M. 1921. Report on the Stomatopoda and the macrourous Decapoda collected by Mr Cyril Crossland in the Sudenese Red Sea. Journal of the Linnean Society Zoology, 34: 345-398. Taylor, J.D., Kennedy, W.J. & Hall, A. 1969. The shell structure and mineralogy of the Bivalvia. Introduction Nuculacea-Trigonacea. Bulletin of the British Museum (Natural History) (Zoology) Supplement 3: 1-125. Theel, H.J. 1882. Report on the Holothurioidea dredged by HMS ‘Challenger’ during the years 1873-76. Report on the Scientific Results of the Voyage of H.M.S. Challenger, Zoology, 4: 1-76. Thiele, J. 1912. Die Antarctischen Schnecken und Muscheln. Deutsche Stidpo- lar Expedition 1901-1903. 13. Zoology, 5: 183-285. 1931. In: Thiele & Jaeckel. Muscheln der Deutschen Tiefsee Expedition. Wissenschaftliche Ergebnisse der Deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’, 21: 159-268. — 1934. Handbuch der Systematishcen Weichtierkunde. Dritter Teil. pp. 779-1154. Jena. Thorsen, G. 1951. Scaphopoda, Placophora, Solenogastres, Gastropoda, Prosobranchiata, Lamellibranchiata. The Godthaab Expedition 1928. Med- delelser om Grgnland, 81: 1-117. Valentine, J.W. & Ayala, F.J. 1975. Genetic variation in Frieleia halli, a deep sea branchipod. Deep Sea Research, 2: 37-44. Vermeij, G.J. 1978. Biogeography and adaptation. 332 p. Harvard University Press. Cambridge, Massachusetts and London. Verrill, A.E. 1884. Second catalogue of Mollusca recently added to the fauna of the New England coast and the adjacent parts of the Atlantic consisting mostly of deep-sea species, with notes on others previously recorded. Transactions of the Connecticut Academy of Arts and Science, 6: 139-294. —— 1885. Third catalogue of Mollusca recently added to the fauna of New England and the adjacent parts of the Atlantic. Transactions of the Connect- icut Academy of Arts and Science, 6: 395-452. —— & Bush, K.J. 1898. Revision of the deep water Mollusca of the Atlantic coast of North America with descriptions of new genera and species. Part I. Bivalvia. Proceedings of the U.S. National Museum, 20: 775-901. Vieés, F. 1905. Sur un nouvel organe sensitif de Nucula nucleus. Bulletin Société Zoologique de France, 30: 88-90. Vokes, H.E. 1949. The hinge and marginal pectinations of Nuculoidea opima (Hall), type of Nuculoidea Williams and Bregar, Journal of the Washington Academy of Science, 39: 361-363. Wagner, F.J.E. 1968. Faunal study, Hudson Bay and Tyrrell Sea. In Hood, P.J. (Ed.) Earth Science Symposium on Hudson Bay Survey of Canada, Geological Survey Paper, Mines and Geology Branch, Canada, 68-53: 7-48. Webster, F. 1969. Vertical profiles of horizontal ocean currents. Deep Sea Research, 16: 85-98. Weisbord, N.E. 1964. Late Cenozoic pelecypods from Northern Venezuela. Bulletin of American Paleontology, 45: 1-564. Whiteaves, J.F. 1901. Catalogue of marine Invertebrata of Eastern Canada. Publications of the Canadian Geological Survey, 727: 1-271. Williams, H.S. & Breger, C.L. 1916. The fauna of the Chapman sandstone of Maine. Professional Paper of the U.S. Geological Survey, Paper 89: 173-174. Washington D.C. Wolff, T. 1956a. Isopoda from depths exceeding 6000 metres. Galathea, Report 2: 85-150. — 1956b. Crustacea Tanaidacea from depths exceeding 6000 metres. Galathea Report 2: 187-241. Yonge, C.M. 1939. The protobranchiate Mollusca; a functional interpretation of their structure and evolution. Philosophical Transactions of the Royal Society of London (B), 230: 79-147. Zenkevitch, L.A. 1969. The origin and the antiquity of the deep sea fauna. pp. 235-243. In: Zenkevitch, L.A. (Ed.) The Pacific Ocean. (In Russian) Moskva: Indatelstvo Nauka. —— & Birstein, J.A. 1956. Studies on the deep water fauna and related problems. Deep Sea Research, 4: 54-64. —— & Birstein, J.A. 1960. On the problems of the antiquity of the deep sea fauna. Deep Sea Research, 7: 10-23. , roy Oe Seat wanes wo ocak od) baap.banlans. a andy wcohoehes coetvodt WOOT A lel hm : \ ee aol aad Soy ms. = 4! ' mye 1 Sa. meade \ cardia | _ streak eral onan : nba hie wer yh istics sais bay User du gue 200) ri i (bea watlons fas Kong! — digs Oth BE simul Sh airoipolend “Tale facta H re enced , Nica ep grurtecech ey bya, em menial at fa a =, 97, S-Snit: Rivet be niewZ On ; sap jp ‘eaten Ray P hugs quaitt trully sag mayA ogo, \Baliy crt shat 4: rl _ fot earay, al, oa dgeny a ; ; IM, 1Ot 0, sorwent Wy yee ( : ae ‘pabsed ., tan — gull pnababl.. doen lancw'h, Are wf tee _ eyes iam es ho co wigilens cine rer . thew ont, Hy ye piece, Steet Aru ap 4.9 YOY Oe ee Ta aneteais MPR er = a see) a, 1 9, ces) dame gary! v ir A ape Pie r —— eh Jue alae A & on echoes = ae te. i ft ber enliiiey lnouton >: mie cemmtaers anibbodras grin won A ne newunteoriils : ‘ined tied Y 1 K vase one 4 2 — a5 a a ae Negee =. eaini ane FL. hell, Jeremietayy muy . fa , Very agp eel an thing 6.05 nade ne eT den ork; 4a we a i Af Ph veowknete seine se eiiohls & t GT hows eens om. 64 t a a? Put. a x Awe: > — oe tL ao Aor aie Chg An’ wiseuedioiadt Ons thy Ma sie ret , in rar) i r) A eet "| rr * i werent i) MG) Lt eee A A pcunellllW ma repel jai. AFM: own damier) eitinerttd 2 ~~ drag ubibin 2 AY) Wyo wan \aerticenices minh ty wi 5 : th gpamrlenad (nw kanye “Wii. Verena . fiat 2 es i i ~ "stalled PA ree aries sm A HAR ait ei ' UO yorenEte | i EY wre aly : m4 is Fe map F ' iA : ype rare PNR 7 St Derererte C1) -@8\;4 they te 4 ’ i=deei Dae uordind p RRL od Kh ‘ pas Steph) soley aio ved... Lopate Veils ® shear ae nag) nay al '¢. vi NOR SETS ver rary prt sare Ba in he rin —— | vind svndige att ere ae vidjiay of) ual L biyaate isks ag ; ies M8; Des 4 {Sie nana Medgiee a vee iewecggad yy wi Sere Schin oi 49 o: 42) aye tue rate nalts peorwe ao! Lire Memiepin, sor 7, th mA ; rs on & at ‘Brit Menéeik (Natural History y) igs Publications’ ‘ | , LARE TAMA “vim Ay ND iTS LID. Tiereelin: | Monsenwy -1 Meike A &- Sere gtas ti vacances od Nirican kates’ ars) 0) Clot Gece) v4 whe Os ORD aRI Me gue SereEPamicttal "Sl came 460 Fete ot | (mie sthon ac the tebe Woe) foe ailatoles Tie vente’ « a at © ofaunt | 466 re ated Tei MVE ar eGPe ED, a aot et 39? oe, batftons wa bee ibasy. porns The cn «ee dite: Reed! University P ae re ce te o , 4 “A WORLD LIST OF MAMMALIAN SPEC Th —_— ace ai! sai b calivines, Contghett: Saciec tre 0 tivimg prad recamtly certiet ack! -., - ae Upon The mast up-fo-dite teponal works and recep awe « ¢ oe iirc. Engtinn Shine apd ditzibymom. Al esicibiee egy it) oe 1? Ke, quickly faiusnad . | sy * 1 Eo PP. illustrated 1 witli drarw: Ty i ian: i‘ qey@as ’ Oxford Umiverality Press, All track order ot OL} og ¢ , —— { ; *, sf MAMMALS OF THE INDOMAI a’ An bee \ CIE i - re ; oi bans "etn Mega re $& i eis Pe ae with abut 4 thoteand Aer & « work gresenls an aunt Of res Gere fe | wm tuli, od plaping ve: wits 4 Lasonv Uwe se oS - ae Ciscpeabinabaied ith Chali I ee wae r eee ane Lk a ee bate he 0 ant ene 5 : oe - a°4 a r 4 be hd a aa 6 e” ">> <= ra ae << «ies 4 math 4 7 » are British Museum (Natural History) Publications LAKE TANGANYIKA AND ITS LIFE G.W. Coulter (Editor) With contributions from J.J. Tiercelin, A. Mondegeur, R. Hecky & R. Spigel Lake Tanganyika is the largest in volume of the African lakes and is the second deepest lake in the world. To protect the lake against exploitative and environmental threats and because of its large natural resources, it is important that all information on this lake is easily available. The primary object of this work has been to collate the scientific knowledge about Lake Tanganyika, and to give an analytical and interpretive treatment. 1991, 276 X 215 mm, 352 pp, halftone and line illustrations. Three colour plates. Co-published with Oxford University Press. All trade orders to OUP. 0 19 858 525 X Hardback £60.00 A WORLD LIST OF MAMMALIAN SPECIES G.B.Corbet & J.E. Hill Extensively revised edition. Complete, concise list of living and recently extinct species of mammals of the entire world based critically upon the most up-to-date regional works and research reports. Over 4300 species are listed, each with its scientific name, English name and distribution. An extensive bibliography allows further details of any species in the list to be quickly found. 1990, 3rd Edition, 240 x 160 mm, 300 pp, illustrated with drawings throughout. Co-published with Oxford University Press. All trade orders to OUP. 0 19 854017 5 Hardback £30.00 MAMMALS OF THE INDOMALAYAN REGION G.B. Corbet & J.E. Hill The Indomalayan (or Oriental) Region has been recognized for over a century as one of the major zoogeographical regions of the world, with about a thousand species of mammals, over 20 per cent of the world’s mammalian fauna. This work presents an account of every species of mammal in the region, documenting the names and synonyms in full, and placing them within a taxonomic framework that takes account of all relevant published research. Autumn 1992 publication. Co-published with Oxford University Press. Approximate cost hardback £60.00. All trade orders to OUP. 21 53 61 CONTENTS The morphology and phylogeny of the Cerastinae (Pulmonata: Pupilloidea) P.B. Mordan A redescription of the uniquely polychromatic African cichlid fish Tilapia guinasana Trewavas, 1936 P.H. Greenwood A revision and redescription of the monotypic cichlid genus Pharyngochromis (Teleostei, Labroidei) P.H. Greenwood Description of a new species of Microgale (Insectivora: Tenrecidae) from eastern Madagascar P.D. Jenkins Studies on the deep-sea Protobranchia (Bivalvia): the family Nuculidae P.M. Rhind and JA. Allen Bulletin British Museum (Natural History) ZOOLOGY SERIES Vol. 58, No. 1, June 1992 Zoology Series ~ VOLUME 58 - NUMBER 2 — — — — 26 NOVEMBER 1992 - 4 DEC 1992 PRESENTED The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology. The Zoology Series is edited in the Museum’s Department of Zoology Keeper of Zoology: Dr C.R. Curds Editor of Bulletin: Mr N. Merrett Assistant Editor: Dr B.T. Clarke Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff and by specialists from elsewhere who make use of the Museum’s resources. Many of the papers are works of reference that will remain indispensable for years to come. A volume contains about 160 pages, made up by two numbers: published Spring and Autumn. Subscriptions may be placed for one or more of the series on an annual basis. Individual numbers and back numbers can be purchased and a Bulletin catalogue, by series, is available. Orders and enquiries should be sent to: Intercept Ltd. P.O. Box 716 Andover Hampshire SP10 1YG Telephone: (0264) 334748 Fax: (0264) 334058 World List abbreviation: Bull. Br. Mus. nat. Hist. (Zool.) © British Museum (Natural History), 1992 Zoology Series ISSN 0007 — 1498 Vol. 58, No. 2, pp. 95-170 British Museum (Natural History) Cromwell Road London SW7 5BD Issued 26 November 1992 Typeset by Ann Buchan (Typesetters), Middlesex Printed in Great Britain by The Alden Press, Oxford Bull. Br. Mus. nat. Hist. (Zool.) 58(2): 95-131 Issued 26 November 1992 Notes on the anatomy and classification of ophidiiform fishes with particular reference to the abyssal genus Acanthonus Gunther, 1878 F LICTADV RALICE GORDON J. HOWES f MHOTORY WUSEL Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD PAEOER ITE D oa ZOOLOGY LIBRA WG CLAN CRORE a ar a ce Aare ces saa oa xk “Epon snd Maa eta ino tie Se RRM s NeIREE aM aie asks oh clgaisintige sana oop 95 MicEbods: and MAtCKIAl Ses... tos recntan ok gow Ae iain ac asl te ats come Mea lia c's Lore mat BEE Tees Megan oe aes sna ctaleines sas sola ct eee 96 POEM IALIOMSTUSCH INTC MIS UIECS Sse ecco ae ae cn on an cprgdea acne Atte cogkae sig cat PRenee eet ames Seeiace 42 eee cieaie scisaicsanGiciscined +e 97 STREET CA1! TERETE NO IA RR SOS REE 8 INE SI SE ea MAS 2 5 ne sn ic aE 98 ERMOVOMCKING TePlOM ANG PAAUNES coo ses «ns saeaiaees ne soe teeeeaeamerecactvesdneaeecssceeescnee sce sresseetes 102 inigenminal verve LOLaMitia ANG NEIVE COURSES tr s.iee esse sigengeme sos ¢+« de dike sade tase Ra eae esas wane en na Meee ne ee 104 PROLSOCLAMIAIIMOLPHOLOSY: Vesa: sets. ke secs s cee tas bolas sis on Sen Se « seele sane aa: Aen eee col ae wn zag eencemeteeet 105 Ollacronysmnanvc EHACtS. sere A hele, LI Moatiides ocnn agen eteees hoe sea age auRee Rom atael leas Pee nE EERE «maples se sermthas ae eee « Nodtogenseiee 126 RESUS SHS 5 recs PELE Btn iad DSB Sno RGhe < oppeitlds «noe ae acide bnsisistonuahe splcaaes Astakeieaanaspein ini temebi Mtanae aan’ 127, BS Se ALARCON RACAI OTIS) 6. 0B SEH. 6 ocBracraceb o.s nes >» seaeee pak eek aaaneee Ry LAs eG exces pee Sloaies Yass SesiteBisiss ‘econ 127 Relationships of Acanthonus and the classification of ophidiiforms .................. cc ceceeeeeeeeee eens eeeeneeeeeeeed 128 AGRI THIEG OTE a, ooo MERE nen nee oe car eee 2 ey rc Re nee eens Smee Rare 131 RS gee eer cee ae Eh cates SER ee acta aia on ic ented ae dee aE A TE eee IS wee ne ian ddim nna ream arytiee 131 Synopsis. Although the dominant group of benthic deep-sea fishes, Ophidiiformes are virtually unknown anatomically. This study focuses on anatomical features of the abyssal genus Acanthonus a taxon which previous authors have considered specialised. Of ca 80 ophidioid and bythitoid genera, 36 have been utilised for comparative study. The characters investigated are those commented upon by previous authors and ones not previously described. Characters identified as synapomorphic are the hypertrophy of the RLA-PP (pelvic) nerves, divergence of the supraorbital trunk of the trigeminal nerve complex external to the trigeminal foramen, reduction and loss of epipleural ribs and posterior shift of the swimbladder. The distribution of these characters upholds the classification of Cohen & Nielsen (1978) with respect to the recognition of a monophyletic Brotulinae and Ophidiinae, but indicates that the Neobythitinae is paraphyletic with part (Hypopleuron) being the sister-group to other ophidioids and bythitoids and other parts being closely related to respectively, brotulines and aphyonids. The Bythitidae also appears non-monophyletic with respect to one genus, Brosmophyciops whose phylogenetic position is ambiguous. The majority of Neobythitinae and Bythitidae are considered as a monophyletic assemblage, the detailed relationships of which are yet to be ascertained. It is confirmed that Acanthonus possesses several autapomorphies, and synapomorphies which support Cohen & Nielsen’s (1978) contention that its close relationship lies with Tauredophidium and Xyelacyba. INTRODUCTION and Cohen & Nielsen (1978). Nearly fifteen years on the Gosline (1953) noted the dearth of anatomical accounts of situation remains almost the same; there is no broad com— ophidiiform (brotulid) fishes, reiterated by Cohen (1974) parative anatomical treatment oof the — group. 96 Gosline (1960) also drew attention to the size and heteroge- neous nature of the group noting that ‘known variation in osteological features in greater than between many families of . . . Percoidae’. In current systematic parlance this varia- tion and heterogeneity can be attributed to an array of apomorphic features most of which appear to be autapomor- phic at the generic ‘level’, witness the many (twenty-four among eighty-two [29%]) monotypic genera. No attempt is made here to provide anything approaching a comprehensive anatomical survey of ophidiiform taxa, instead certain anatomical features are detailed from the standpoint of Acanthonus, a monotypic abyssal genus (Fig. 1). Acanthonus is presently placed in the majority ophidiiform group, the Neobythitinae; it appears to possess a number of specializations correlated with its abyssal life (Nielsen, 1966; Horn et al., 1978) most of which appear to be unique but offering the possibility that some might be shared with anatomically uninvestigated taxa. It is these ‘unique’ features together with certain ophidiiform features com- mented upon by previous authors which are used as starting points in making a broader comparative anatomical investiga- tion. Regan (1903; 1912) and Gosline (1953) had made asser- tions concerning certain ophidiiform features and it is these which have been taken as a starting point for a broader comparative investigation. The taxonomic breadth of the comparisons is, however, limited by the availability of mate- rial. Among the ca eighty ophidioid and bythitoid genera, thirty-six (45%) have been used and then not for every character complex. By examining a combination of osteolog- ical, myological and neurological features it was hoped that synapomorphies would be revealed which would identify at least basal monophyletic assemblages. Thus far, only one potential synapomorphy, involving the morphology of the cranial articular facets and their connection with the vertebral column, has been recognised as splitting ophidiiforms into the Carapidae plus Ophidiidae and the Bythitoidei, the latter being the derived assemblage (Patterson & Rosen, 1989). Monophyly of the Carapidae has recently been satisfactorily demonstrated by Markle & Olney (1990) while that of both the Ophidiidae and Bythitoidei remains suspect. METHODS AND MATERIALS Although Patterson & Rosen (1989) recognised the Bythitoi- dei as a monophyletic group they were not explicit in the content of that group naming only certain genera lacking the synapomorphic ‘cod-like’ exoccipital facets and which were classed as ‘ophidiids’. For the purposes of the anatomical descriptions the usage of ‘ophidiiform(s)’ embraces the Ophi- doidei and Bythitoidei sensu Cohen & Nielsen (1978), the Carapidae being treated as an outgroup along with other paracanthopterygians. Specimens examined (the majority have been dissected and radiographed and from some the cranium has_ been extracted): Abyssobrotula galatheae Nielsen Uncat. ‘Discov- ery’ Stn 10652, 5112 m (149 mm SL); Acanthonus armatus Gnthr 1990.8.21:78a-f (6 specs 290-350 mm SL), off Cape Verde, 3120 m; 1887.12.7.55 (holotype, 285 mm SL), Philip- pines; 1887.12.7.56 (Paratype, 310 mm SL) North of New Guinea; Aphyonus gelatinosus Gnthr 1887.12.7:59 (Holo- G.J. HOWES type, 130 mm SL), N.E. Australia; Barathrodemus cf. manat- inus Goode & Bean 1991.11.14.1 (107 mm SL), 24°25’N, 77°22'W; Barathronus bicolor Goode & Bean 1961.9.7:1 (104 mm SL), Puerto Rico; Bathyonus sp. Uncat. ‘Discovery’ Stn 12179 (190mm _ SL); Bassozetus taenia (Gnthr) 1990.8.21:97-8 (205 mm SL, skull prep.), Cape Verde; Bas- sozetus sp. 1991.7.9:10-33 ‘Discovery’ Stn 10884 (ca 156, 170 mm SL); Brosmophyciops sp. 1991.11.14:2 (52 mm SL) N.W. Gulf of Aqaba; Brotula multibarbata Temminck & Schlegel 1983.3.25:1128-30 (103 mm SL), Fiji; Carapus acus (Brunn.) 1952.11.25:1-4 (180 mm), Naples; Carapus bermu- densis (Jones) 1985.6.6:138-183 (alizarin/alcian), Bimini; Cataetyx sp. 1990.8.21: 184-8 (226, 250 mm SL; skull prep.), Porcupine Seabight; Cherublemma emmelas_ (Gilbert) 1985.6.6:136-7 (205 mm SL), Baja California; Dicrolene intronigra Goode & Bean 1939.5.24:1441—-4 (135 mm SL), S. Arabian coast; Dinematichthys sp. 1983.3.25:1123-5 (92 mm SL), Fiji; Diplacanthopoma brachysoma Gnthr 1972.10.24:4 (185 mm SL) 9°03'N, 81°18’W; Echiodon drummondii Thompson 1969.5.4:3-5 (300 mm SL) 58°N, 9°W; Genypterus blacodes (Schneider) 1936.8.26:1052-7 (290mm SL) S. Atlantic; 1896.6.17:73, (skeleton) Melbourne Market; Glyp- tophidium macropus Alcock 1939.5.24:1458-65 (170 mm SL, skull prep.) Gulf of Aden; Hoplobrotula armata (Temminck & Schlegel) 1938.6.23:27-8 (193 mm SL), Chousi, Japan; Hypopleuron caninum Smith & Radcliffe 1986.10.6:63—65 (330 mm SL) Indonesia; Lamprogrammus fragilis Alcock 1939.5.24:1493-6 (130 mm, body truncated: 220 mm SL) S. Arabian coast; L. niger Alcock 1939.5.24:1483-7 (300 mm SL; skull ex ca 380 mm SL) maldive area; Lepophidium profundorum (Gill) 1984.8.8:263-6 (105 mm SL) Baja Cali- fornia; Lucifuga dentatus Poey 1981.10.27:1-4 (104 mm SL) Cuba; Monomitopus metriostoma (Vaillant) 1964.8.6:47-54 (135 mm SL) 7°55'N, 12°38'E; Neobythites gilli Goode & Bean 1967.11.9:1-6 (100 mm SL) Caribbean; N. steatiticus Alcock 1910.1.31:11 (skeleton) Sea of Oman; Nybelinella erikssoni (Nybelin) 1991.5.7:2 (58 mm SL) ‘Discovery’ Stn 11261, 5440 m; Ogilbia cayorum Evermann & Kendall 1985.6.6:128-135 (69, 72 mm SL) Bahamas; Ophidion rochei Muller 1971.12.17:6-8 (175 mm SL) and Uncat. (170 mm) Black Sea; Penopus sp (285 mm SL) ‘Discovery’ Stn 12177 (No. 2); Petrotyx sanguineus (Meek & Hildebrand) 1976.7.15:205 (410 mm SL) B.V.I.; Porogadus trichiurus (Alcock) 1939.5.24:1453-5 (175 mm SL) Zanzibar; Pyc- nocraspedum squamipinne Alcock 1939.5.24:1497-8 (114 mm SL) Zanzibar; Sirembo imberis (Temminck & Schlegel) 1905.2.4:450—4 (137 mm SL; skull prep.) Wakanoura, Japan; Spectrunculus grandis (Gnthr) Uncat. (125 mm SL, alcian stained spec.) ‘Discovery Stn 51015, 2540m; 1990.8.21:106-111 (420; 250; 220 mm SL, skull prep) Porcu- pine Seabight; Tauredophidium hextii Alcock 1890.11.28:38 (type, 100 mm SL) Ganjam Coast; 1992.2.4:3-5 (89 mm SL) 11°31'S, 86°55'E; Thalassobathia pelagica Cohen 1967.11.8:1 (240 mm SL) Donegal; Typhlonus nasus Gnthr 1887.17.7:58 (Holotype, 205mm, tail missing) N. of Celebes; 1887.12.7:587 (Paratype, 215 mm SL) N.E. of Australia; 1992.2.4:6 (190 mm SL) Indian ocean; Xyelacyba myersi Cohen USNM 320096 (120 mm SL) 24°51'N, 90°00'W; USNM 212087 (73 mm SL) 38°35'N, 72°23'W; 1992.2.4:1-2 (170 mm SL) 25°S, 88’E. Pediculati: Antenharius nummifer (Cuvier) 1888.12.29:143 (skeleton); 1898.12.24:103-112 (49 mm SL) Persian Gulf; Porichthys porosissimus (Val.) 1937.9.30:163-7 Bahia Blanca; 1890.11.15:150 (skeleton); Lophius piscatorius Linn. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES 1970.2.17:630-2 (95 mm SL) S.W. Ireland; Uncat. (skull). Abbreviations used in the figures Al,2,a,b Divisions of adductor mandibulae muscle aa anguloarticular ahf anterior hyomandibular fossa asp autosphenotic bl Baudelot’s ligament blf fossa for attachment of Baudelot’s ligament bo basioccipital boc basioccipital cavity for articulation with first cen- trum bof basioccipital facet c centrum (numbered) clp anterior process of first centrum cc cerebellar corpus cl cleithrum cle cleithral extension cs spinal cord de dentary do dilatator operculi muscle dRLA dorsal branch of ramus accesorius lateralis nerve dsp dermosphenotic ect ectopterygoid ent entopterygoid epd dorsal epaxial muscle epl anterior lateral segment of epaxialis muscle epm anterior medial segment of epaxialis muscle epo epioccipital epr epipleural rib eps lateral band of epaxial muscle connecting Ist and 4th ribs epx epaxial muscle es ethmoid spine exc exoccipital condyle exf exoccipital facet exo exoccipital fc frontal crest fg foramen for glossopharyngeal nerve fh foramen for hyomandibularis nerve fli lateral ethmoid-first infraorbital facet flp lateral ethmoid-palatine facet (lateral) fm foramen magnum fmp lateral ethmoid-palatine facet (medial) foo foramen for optic-olfactory and trigeminal nerves fon foramen for occipital nerves fr frontal fRLA foramen for ramus lateralis accessorius nerve fsn spinal nerve foramen ft facial trunk of trigeminal nerve fv vagus nerve foramen gaS Sth gill-arch hb hyoideus branch of hyomandibularis nerve hy hyomandibular hyb2 2nd hypobranchial hyx hypaxial muscle ic intercalar ica infracarinalis anterior muscle ih interhyal i0 interoperculum ki kidney lal labial ligament lap levator arcus palatini muscle O77 compound ligament part of compound ligament connecting swimblad- der with first rib lateral ethmoid ethmoid-infraorbital ligament liver interneural ligament mandibular-interopercular ligament neural-rib ligament levator operculi muscle preopercular-opercular ligament palato-vomerine ligament swimbladder-vertebral ligament mandibular branch of hyomandibularis nerve outer subbranch of above mesethmoid ethmoid cartilage medial opening of mandibular canal (dentary) metapterygoid maxilla cranial nerves first neural arch second neural arch auditory nerve mandibular branch of trigeminal nerve maxillary branch of trigeminal nerve neural spine vagus nerve olfactory foramen olfactory bulb olfactory lobe notch in lateral ethmoid for olfactory nerve perforation in orbital septum for olfactory nerve operculum optic lobe orbital septum palatine palatine prong parietal pelvic bone pharyngoclavicularis externus muscle postcleithrum peritoneum posterohyal pharyngohyoideus muscle pleural rib pelvic muscles medial ridge of parasphenoid premaxilla posterior (pterotic) hyomandibular fossa pelvic fin ray prootic parasphenoid parasphenoid process pterotic pterosphenoid posttemporal quadrate first enlarged rib (pleural or epipleural) ribs (pleural or epipleural) retroarticular rostral cartilage retractor dorsalis muscle rostrodermosupraethmoid 98 RLA-PEL pelvic branch of RLA nerve RLA-PP _pectoral-pelvic branch of RLA nerve rm recti muscles sb swimbladder sbc sclerified cap of swimbladder shl lateral segment of sternohyoideus muscle shm medial segment of sternohyoideus muscle smx supramaxilla sn spinal nerve so supraoccipital sol semi-ossified ligament sop suboperculum sot supraorbital trunk of trigeminal nerve sp sphenotic ste sclerified tunica externa of swimbladder sy symplectic tepm tendon of medial epaxial (swimbladder) muscle ti tunica interna tnp perforation in orbital septum for trigeminal trunk uh urohyal ut tendon-ligament connecting sternohyoideus with urohyal vh ventrohyal vo vomer vps vertebral parapophysis vst ramus of third spinal nerve vvp vertebral ventral process ANATOMICAL FEATURES Ethmo-vomerine region and palatine In Acanthonus the ossified dorsal surface of the ethmoid (rostrodermosupraethmoid, RDS) is produced anteriorly into a broad, dorsally channelled strut with a bifurcate tip which G.J. HOWES pierces the skin (Figs 1;2): this is the bifid spine referred to by Cohen & Nielsen, (1978:18). The posterior border of the RDS is only partially overlain by the frontals. The anterior surface of the underlying ethmoid is sharp-edged and overlies a thick ethmoid cartilage which is visible only laterally (emc, Fig. 2). The lateral ethmoid (le, Fig. 2) is broad with a sloped anterior face abutting the mesethmoid cartilage and a thin lateral wing. The ventrolateral surface of the lateral ethmoid bears a well-developed almost vertically directed medial (fmp) and a weak laterally directed articulatory process (flp, Fig. 2). These processes articulate respectively with the medial and anterior surfaces of the palatine prong. The medial part of the lateral ethmoid is deep and contacts the parasphenoid ventroposteriorly and the frontal dorsally. Each lateral ethmoid is separated from its partner in the midline. The olfactory foramen is variously developed. In the specimen illustrated the ‘foramen’ appears as a notch on the posterior rim of the lateral ethmoid wing. The olfactory nerve merely passes over the edge of the bone and crosses its anterior face to where the olfactory rosette is situated. In other specimens a thin posterior stem of bone provides either a complete or incomplete closure of the notch. The vomer (vo, Fig. 2) has a deep and broadly triangular head and a short narrow shaft the posterior tip of which lies in line with the posterior borders of the lateral ethmoids. The tooth-patch is deep and the conical teeth are directed not only ventrally but also anteriorly and anterodorsally (Fig. 2C). The ethmovomerine region of Acanthonus differs most noticeably from the other ophidiiform taxa examined in three respects, 1) elongation and bifurcation of the RDS; 2) position of the palatine articulatory facets on the underside of the lateral ethmoid; 3) disposition of the vomerine teeth. As in Acanthonus in nearly all other taxa the anterior of the mesethmoid capping the ethmoid cartilage slopes forward. However, the slope is variable from being almost vertical (Ophidion and Genypterus; Fig. 3A) to shallowly sloped (Cataetyx and Lamprogrammus; Fig. 3F). In Glyptophydium Fig. 1 Acanthonus armatus, life appearance based on type specimens, photograph in Horn et al. (1978) and Winther’s drawing in Nielsen (1966). Scale in cms. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES 99 leh BAN YS ra BOWS wh os =>) — on os Fig. 2 Acanthonus armatus ethmovomerine region in A, lateral, B, ventral and C, dorsal views. Scale in this and subsequent figures in mm divisions. (Fig. 3C), Sirembo and Spectrunculus the ethmoid region is indented and, notably in the latter genus, the frontals extend far anteriorly rising medially to form a crest which in Glypto- phidium (Fig. 3C) is particularly pronounced. Hoplobrotula and Tauredophidium are the only other taxa examined which possess an anteriorly exiended RDS spine (also reported by Machida, 1990). Only Cataetyx, Brotula, Glyptophidium and Lamprogrammus have a prominent lateral wing on the lateral ethmoid. In other taxa, including Acanthonus, the wing is feebly developed but always has a lateral facet which articu- lates with the first infraorbital (fli, Fig. 3). The disposition of the palatine articulatory facets on the lateral ethmoid is a variable and perhaps important classifica- tory feature. Unlike the condition in Acanthonus where the 100 G.J. HOWES Fig. 3 Ethmovomerine regions of A and B Genypterus blacodes in lateral and ventral views; C, Glyptophidium macropus (lateral); Brotula multibarbata in D, lateral and E, ventral views; F, Lamprogrammus niger (lateral). lateral facet is reduced, in other taxa examined both facets are equally well-developed. In Brotula (Fig. 3D,E), Genypterus (Fig. 3A), Ophidion, Bassozetus, Glyptophidium (Fig. 3C) and Spectrunculus the medial facet is large and ventroposteriorly directed and the outer facet is usually narrowly but clearly separated by a well-defined depression (exceptionally in Brotula the outer facet is juxtaposed to the inner and is continued on to the lateral flange of the lateral ethmoid; Fig. 3E). The lateral facet lies on the outer rim of the lateral ethmoid and is angled downward. The facets contact apposing surfaces on the palatine head, the medial articulating with the dorsomedial surface of the head, the lateral with the base of the palatine prong. In Lamprogram- mus there is only a medial facet, the base of the palatine prong being attached to the lateral ethmoid rim by a thick ligament (Fig. 3F). In Cataetyx the facets lie in an almost straight line but are well-separated. The palatine-lateral ethmoid articulation in Acanthonus is rigid and the leading edge of the palatine is tightly juxtaposed to the vomer so that the vomerine and palatine teeth are ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES contiguous (Fig. 4B). A particular feature of the palatine teeth in Acanthonus is their lateral placement (Fig. 4B); in all other taxa examined the tooth patch is orientated along the medial surface of the bone so that nearly all the teeth point inward and only a single row is usually visible along the ventral margin of the bone. The palatine is secured to the lateral ethmoid by two ligaments (present in all taxa exam- ined), one running from the dorsal indentation of the head of the bone or its medial rim to the lateral ethmoid wall, the other, which may sometimes be divided, from the inner face of the palatine prong to the lateral ethmoid cavity above the vomer (lpv, Fig. 3B). The latter ligament corresponds to Stiassny’s (1986) anterior palato-vomerine ligament (VI). In all the taxa examined which embrace Cohen & Nielsen’s (1978) brotulines, ophidines and neobythitines, this ligament invariably stems from the lateral ethmoid itself rather than the vomer. A medial palato-lateral ethmoid ligament is a widespread teleostean feature. There is, however, no poste- rior palato-vomerine ligament (Stiassny’s ligament V) in any of the examined taxa. This absence is most likely a secondary loss associated with the development of lateral ethmoid- palatine articulatory facets. The shape of the palatine, the angle at which the prong is directed and the proximity of the anterior palatine margin to the vomer are largely dictated by the orientation of the palatine facets with the corresponding articular surfaces of the lateral ethmoid. Usually the articulatory facets of the palatine are discrete, the medial one forming a right-angle to that on the neck of the prong so that the base of the prong lies in the saddle of the articulatory processes. As noted above, in Cataetyx the articulatory facets on the lateral ethmoid lie nearly in tandem, consequently the palatine prong extends directly forward rather than pointing ventrolaterally and the anterior lateral ethmoid facet appears to act as a stay. In its rigidity with the lateral ethmoid and juxtaposition with the posterior rim of the vomer, the palatine articulation of Cataetyx most closely approaches that of Acanthonus (Fig. 4A). There are, however, noticeable differences between the two taxa in the angle of the palatine prong and shape of articulatory surfaces which suggest that the resem- blances have been independently derived. Palatine-lower jaw bite The reason for the rigidity of the palatine-ethmovomerine connection in Acanthonus becomes clear when it is realised that the palatine occludes with the lawer jaw (Fig. 4B). The palatine tooth patch is long, 75% of the dentary and when occluded the posterior tips of the palatine and dentary tooth-patches coincide (Fig. 4B). In other ophidiiforms (and indeed, other teleosts) there is no or only partial (see below) occlusion between the palatine and lower jaw; the length of the palatine tooth patch is usually only half that of the dentary and the teeth are angled medially and fall inside the lower jaw when the mouth is closed so that teeth are rarely visible in lateral view (Fig. SC). To what extent, in life, the palatine can be rotated laterally so as to occlude with the lower jaw is not clear merely from the manipulation of the jaws of preserved specimens but certainly the connection between the palatine and ethmovomerine region is loose enough to allow some lateral rotation. Of the four exceptions among examined taxa two display partial and two (apparently) complete occlusion between the palatine and dentary. In Lamprogrammus and Spectrunculus 101 Fig. 4 Above, palatine and vomer of Cataetyx sp. (right lateral); below, palatine-lower jaw occlusion in Acanthonus armatus; the lower jaw is shown in the closed position and the elements are orientated with respect to the vertebral axis. Dashed outline indicates hidden border of the vomer. D y ie tty, Uiffe_7 GLAS: KY Li Fig.5 Above, palatines of A, Lamprogrammus niger, B, Spectrunculus grandis, C, Neobythites steatiticus (right lateral views). Below, articulatory surfaces of lower jaws of D, Acanthonus armatus, E, Lamprogrammus niger, F, Genypterus blacodes, G, Spectrunculus grandis (dorsal view, medial surface at top). only the posterior half of the palatine is in direct contact with the lower jaw when the mouth is tightly shut, the anterior half curving medially. The dentigerous surface of the palatine in Spectrunculus is longer than that of other taxa (Fig. 5B) but 102 nonetheless its posterior tip extends backward beyond the posterior tip of the dentary tooth patch, a feature common to all other genera examined. In both Lamprogrammus and Spectrunculus the palatine dentigerous area is exposed later- ally (Figs SA & B). From radiographs and partial dissections of Tauredophidium and Xyelacyba it would seem that these taxa also have complete or nearly complete occlusion of the palatine and lower jaw. In Xyelacyba the palatine is deep with a strong and steeply angled prong (as in Acanthonus) and in both Tauredophidium and Xyelacyba the palatine tooth patch is narrow with the teeth laterally exposed. Acanthonus is not unusual among ophidiiforms in lacking upper and lower jaw occlusion, indeed it seems to be the common condition among this group of fishes that the lower jaw is surrounded by the upper when the mouth is shut. The feature is most clearly seen in Spectrunculus where the rather sharp-edged lip of the lower jaw forms a tight seal with the overlapping upper lip. Typhlonus is unusual among ophidii- forms in possessing long premaxillary ascending processes which also allow the upper jaw to envelope the lower. The dentigerous surface of the upper jaw rarely contacts that of the lower aithough in Cataetyx the tooth bands contact one another anteriorly. In Lamprogrammus there is a prominent dentary symphyseal process which lodges in a symphyseal notch in the upper jaw and although Acanthonus also bears a similar dentary process it is completely covered when the upper jaw is closed. Quadrate-preopercular modification The occlusion between the palatine and lower jaw teeth in Acanthonus is allowed principally by modifications in the relationship of the quadrate to the lower jaw and the contact between the quadrate and preoperculum. Acanthonus differs principally from other taxa examined in having the quadrate outwardly curved and the arc of that curvature continued through the ectopterygoid thus considerably displacing the palatine laterally (Fig. 6). In other ophidiiform taxa the anterior border of the quadrate is either perpendicular or inwardly curved so that the palatine lies either in the same vertical plane as the quadrate condyle or only slightly later- ally (Fig. 6). Only in Spectrunculus is there a noticeable outward curvature of the quadrate and lateral displacement of the palatine. The shape of the Acanthonus quadrate condyle shows no special modification although it is some- what shallower and its medial face more attenuated ventro- medially than that of other taxa (Fig. 6). When the lower jaws of various taxa are positioned at the same angle with respect to the quadrate it is seen that in Acanthonus the upper anterior face of the bone and the lower border of the ectopterygoid are obscured laterally by the anguloarticular; in the other taxa, there is a wide separation between the quadrate-ectopterygoid border. Only Lampro- grammus approaches Acanthonus in the separation of the elements (radiographs of Tauredophidium indicate a similar condition). Acanthonus and Lamprogrammus also possess a similarly modified anguloarticular condyle and its shape possibly dictates the alignment of the lower jaw (the condi- tion in Tauredophidium is unknown). In the majority of ophidiiform taxa examined the anguloarticular is elongated, terminating in a posteriorly directed pointed or rounded process (Figs SF & G). The transverse saddle of the condyle is well defined and the articulatory surface is elongate and extends on to the medial side of the condyle. In contrast both G.J. HOWES Acanthonus and Lamprogrammus have a short articulatory condyle with an angled, slightly upturned blunt posterior process and there is a well-defined saddle with articulatory facets equally or nearly equally disposed on either side (Figs 5D & E). The articulation between the ventral surface of the quad- rate and the lower limb of the preoperculum in Acanthonus is absent in the majority of other taxa examined. The posterov- entral surface of the quadrate is flared and sits across a similarly widened preopercular flange (Fig. 7). A preopercu- lar flange, which extends ventrally to cover the neuromasts, is common to all the taxa examined but in Acanthonus it extends shelf-like from the bone and terminates in a ventrally extended triangular process (Fig. 7). Lamprogrammus and Xyelacyba also have a broad based quadrate and an extended preopercular lateral flange (radiographs of Tauredophidium indicate a similar condition). In Lamprogrammus, however, the flange does not turn ventrally but extends laterally well beyond the quadrate border. Acanthonus differs from all these taxa in that it is only the posterior half of the quadrate which contacts the preopercular limb. In the others nearly the entire quadrate lies on the preopercular limb so that its anterior tip comes close to the quadrate condyle. In Acanthonus the preoperculum has a slender upright limb which has a slight anterior ridge and a short horizontal limb (Fig. 7). In addition to the lateral ventral spine (see above) the posteroventral border is also produced into a long stout spine. In other examined taxa there are two principal condi- tions of the preopercular upright limb, it is either slender as in Acanthonus or short and broad with a rounded posteroven- tral margin. In those six taxa with a slender limb, five (Neobythites, Hoplobrotula, Monomitopus, Dicrolene and Tauredophidium) have the posteroventral border developed into two or three spines. Tauredophidium has especially long spines similar to those of Acanthonus. Brotula lacks preoper- cular spines and most of the lateral face of the upright limb is covered by a pronounced flange; the horizontal limb is also elongated and modified (see p. 122). Of those taxa with short, rounded upright preopercular limbs none bear spines but an anterior groove and flange is often present. In the taxa with a slender preopercular limb neuromasts are poorly developed but in those with a short limb are either moder- ately or well-developed (Lamprogrammus, Dicrolene, Monomitopus) or absent (Genypterus, Cataetyx). A particular feature of Acanthonus is a ligament which stretches from the posterodorsal rim of the preoperculum across the hyomandibular articulation to the anteroventral margin of the operculum (Ipo, Fig. 17); this feature is other- wise present only in Tauredophidium and Xyelacyba. In Typhlonus a strong ligament extends from a groove in the centre of the lateral face of the preoperculum to insert on the lower articulatory margin of the operculum and does not extend across it. Hyomandibular articulation and foramina In Acanthonus the hyomandibular is short, lacks a pro- nounced stem and has a slightly indented anterior margin and a short opercular articular process (hy, Fig. 7). The postero- dorsal part of the bone has an acute mesad slope so that the perpendicular plane of the body and stem of the bone comes to lie at some distance lateral to its articulation with the cranium. The posterior half of the dorsal border of the hyomandibular curves sharply mesad so that the posterior ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES Fig. 6 Palato-pterygoid orientation right side, viewed anteriorly, in Acanthonus armatus (right) and Neobythites steatiticus (left). The anterior part of the palatine has been cut through (hatched). The position of the lower jaw in Acanthonus is shown by a dashed line. Fig. 7 Acanthonus armatus; hyomandibular and palato-pterygoid series, right side, in medial view (main figure); above, dorsal view of hyomandibular, medial surface to the bottom, the arrow indicates the surface articulating with the pterotic; below, ventral view of quadrate and preoperculum (dashed line shows shape of quadrate base), lateral surface at top; below left, lateral view of quadrate showing its contact with the preoperculum. articulation is with a shallow fossa which curves around the posterior border of the pterotic and terminates in a notch (ppf, Fig. 8). The anterior articulation is with a deep fossa on the posterior surface of the sphenotic. The anterior fossa extends posteriorly as a shallow groove a short way along the 103 pterotic and is widely separated from the posterior fossa (ahf, Fig. 8). The foramen for the hyomandibular trunk of the facial (VII) nerve lies in the centre of the anterior field of the bone and the nerve passes directly through it to immediately turn ventrad on the lateral face. In no other ophidiiform taxon examined does the upper part of the hyomandibular curve mesad as in Acanthonus and in only three genera (Spectrun- culus, Bathyonus, Bassozetus) is the nerve foramen situated in the anterior field. In other taxa the foramen perforates the dorsoanterior strut of the bone. The nerve enters a medial opening and courses posteroventrally through the strut to exit from a lateral opening situated at a lower level than the medial (Figs 9B & C). Markle & Olney (1990) report that in Brotula multibarbata the hyomandibular foramen is small and located close to the edge of the bone. Furthermore, they note two conditions of the foramen in carapids whereby it either pierces the anterior strut or the area below the strut. These conditions correspond to the two states observed in ophidio- ids. Markle & Olney’s observation of a small, anteriorly located foramen in Brotula is either erroneous or based on an aberrant specimen (in their figure (24) the foramen seems too small to transmit the large diameter trunk of the hyomandib- ular nerve). As such their polarity assignment of a derived state to the larger carapid foramen is questionable. A foramen piercing the hyomandibular anterior strut appears to be the widespread euteleostean condition since it occurs in virtually all of a wide range of taxa examined (clupeomorphs, ostariophysans, berycoids, atherinomorphs, percomorphs). Among other paracenthopterygians the hyo- mandibular has been modified to a greater or lesser extent with subsequent relocation of the foramen and nerve course. In gadiforms attrition of the anterior strut has resulted in the foramen appearing at the anterior margin of the hyomandib- ular rather than the medial face of the bone (Howes, 1989), a character which Gosline (1968) used to define Gadiformes. In the bathygadid Gadomus attrition of the anterior strut has led to complete absence of an anterior hyomandibular foraman, a feature also common to (?all) macrouroids (Howes, 1989, figs 8 & 9A). Interestingly, the absence of a dorsoanterior strut from the hyomandibular of Acanthonus suggests a direct correlation with the loss of the pars jugularis. In the pedicu- late paracanthopterygians examined (Lophius, Antennarius, Porichthys) the hyomandibular trunk travels through the central stem of the bone (Fig. 9A) and upon exiting via a lateral foramen at the base of the stem the hyoideus branch departs from the mandibularis branch. In gadiforms the hyoideus departs from the mandibularis prior to entering the hyomandibular and travels separately within or across the bone (Howes, 1989, Fig. 9). In ophidiiforms and other teleo- sts examined the hyoideus branch departs from the mandibu- laris immediately on leaving the foramen and crosses the lower face of the hyomandibular (Fig. 9C). Brotula is unique amongst the ophidiiforms examined in that the hyoideus branch runs interiorly through the stem of the hyomandibular before exiting posteroventrally (Fig. 9B). The ‘precursor’ of the Brotula condition occurs in Hoplo- brotula where a strong flange lies diagonally along the lateral face of the hyomandibular following the line of the central stem and partially covering the nerve. The conditions in Brotula and the pediculate paracanthopterygians are clearly not homologous since in the former only the hyoideus branch of the hyomandibular trunk runs internally and exits from a separate foramen. The Pediculati are derived with respect to 104 fr eaeaae os onp es a ahf G.J. HOWES pte pat SO epo exo fv exo ic ‘9 ue Fig. 8 Acanthonus armatus Neurocranium; postorbital and occipital regions in left lateral view. other Gadoidei in the enclosed passage of the nerve through the hyomandibular. The majority of ophidiiforms possess a condition corresponding to that in other euteleosts where the nerve travels a short distance within the anterior strut of the hyomandibular. The transmission of the nerve directly through a foramen in the anterior lamina of the bone is considered derived in Acanthonus and Spectrunculus. The hyomandibular opercular process in Acanthonus is short (Fig. 7), in which respect it is like that of a wide spectrum of taxa including Carapidae, Hypopleuron, Glypto- phidium, Tauredophidium and Brotula (Fig. 9B). Among other taxa, however, the process is long (Neobythites, Sirembo, Monomitopus, Genypterus, Ophidion) and is espe- cially so in Spectrunculus, Cataetyx and Bathyonus. Trigeminal nerve foramina and nerve courses In Acanthonus the trigeminal, facial, optic and olfactory nerves exit the cranium through a common aperture. How- ever, they pass through separate openings in a membranous septum which extends between the pterosphenoid-prootic and parasphenoid (Fig. 8). In having a common cranial nerve foramen Acanthonus differs from all other ophidiiforms examined (with the exception of Bassozetus, Abyssobrotula and Lucifuga); among those other taxa are major differences in the arrangement of the trigeminal-facialis foramina. A narrow prootic lateral commissure is common to all the examined taxa with the exception of Genypterus and Cata- etyx, both of which have a long lateral commissure (Fig. 10A). In all taxa the hyomandibular trunk foramen is narrowly separated from the trigeminal but the distance between the trigeminal foramen and opticolfactory aperture is variable. In Genypterus the distance is greatly increased due to elongation of the parasphenoid ascending process and pterosphenoid. In Brotula (Fig. 10C), Spectrunculus and Cat- aetyx the distance between the foramina is half that in Genypterus whereas in Glyptophidium, Neobythites (Fig. 10B), Sirembo and Monomitopus the separation is narrow. In Lamprogrammus it is intermediate between the first and second group of taxa. Bassozetus, Abyssobrotula and Lucifuga resemble Acanthonus in lacking a separate trigeminal foramen and in all these genera the common cranial nerve foramen is bordered by a lateral commissure. The size of the common cranial nerve foramen is variable but in Brotula, Sirembo and Genypterus it is narrow, exception- ally so in Brotula (foo, Fig. 10C) where an anteriorly directed laminar process rises from the midline of the parasphenoid and serves to separate the nerve tracts of either side (psp, Fig. 10C). According to Markle & Olney (1990) this process is a basisphenoid. However, in my specimen there is no suture to indicate that it is a separate element and the process stems from a medial ridge along the parasphenoid (pmr, Fig. 10C). In Neobythites the parasphenoid bears prominent paired ridges (pmr, Fig. 10B) but lacks a medial process. A basisphenoid does not occur in any ophidiiform or gadiform examined. In Acanthonus the supraorbital trunk of the trigeminal nerve complex diverges from the infraorbital trunk at the point of emergence of the nerve bundle from the common cranial nerve foramen (Fig. 11). The RLA branch of the supraorbital (not shown in figure) extends dorsad entering a foramen between the sphenotic and pterosphenoid to pass caudad beneath the sphenotic and pterotic bones. Among other ophidiids, bythitids and aphyonids examined ANATOMY AND CLASSIFICATION OF OPHIDITFORM FISHES Fig. 9 Hyomandibular and its nerve courses in A, Porichthys porosissimus, B, Brotula multibarbata, C, Neobythites gilli (right lateral views). two groups of genera can be distinguished on the basis of whether the supraorbital trunk branches from the infraorbital trunk external or internal to the facialis chamber in the prootic. In Brotula, Genypterus, Parophidium, Cherub- lemma, Ophidion (Group A, Table 1) (Figs 12E & F), branching occurs within the chamber, the supraorbital nerves departing from the cranial cavity through separate foramina, as is the usual condition in teleosts (Fig. 12E). In Dicrolene, Hoplobrotula, Sirembo and Neobythites branching occurs at the point of emergence and the supraorbital nerves run dorsad along a channel in the outer surface of the sphenotic (Fig. 12A). In all other examined taxa (Group B, Table 1) branching occurs external to the chamber whether or not the exit foramen is common to the optic and olfactory nerves as in Acanthonus, Bassozetus (Fig. 12D), Abyssobrotula and Lucifuga or a separate trigeminal foramen as in the other above-cited genera (Fig. 12B, C & F). Among gadoids and macrouroids the supraorbital nerves branch from the trigeminal trunk just prior to the latter’s exit from the common optic-trigeminal foramen. The nerves then run medial to the pterosphenoid, departing from the cranial cavity via a foramen between the pterosphenoid and frontal to lie against the roof of the orbital cavity. Branching of the supraorbital nerves external to the cranial cavity and lack of separate formina are considered derived conditions. The situation in Hoplobrotula, Sirembo and Dic- rolene where marginal branching exists but a channel serves to conduct the supraorbital nerves dorsally is hypothesised to be a condition intermediate with the external branching and 105 dsp Fig. 10 Postorbital connections in A, Genypterus blacodes, B, Neobythites steatiticus, C, Brotula multibarbata (left lateral views). The extent of the trigeminal foramen covered by the lateral commissure of the prootic is indicated by dashed lines. loss of trochlear and oculomotor foramina. The derivation of the derived state may be viewed as an anterior displacement of the nerve complex, possibly the result of temporal read- justments in the ontogeny of the brain and nerve trunks relative to the osteological development of the cranium (Fig. 12G). Loss of the pars jugularis giving rise to a common optic- olfactory-trigeminal foramen is considered a synapomorphy for Gadiformes, linking Gadoidei with Macrouroidei (Howes, 1989; 1990; 1991). That a similar large-scale modifi- cation should also occur among ophidiiforms in what are apparently separate lineages questions the homologous ver- sus homoplastic nature of the macrouroid and gadoid charac- ter. Dorsocranial morphology Gosline (1953) remarked on the nature of the parietals in ophidiiforms. Of the three taxa he examined, only in one, Dinematichthys, do the parietals meet in the midline. Cohen 106 G.J. HOWES cc op! of! oF nt pro Fig. 11 Acanthonus armatus: brain and cranial nerves in dorsal view; the semicircular canals are shown in light shading. (1974) shows that in Brotulotaenia the parietals almost meet anterior to the supraoccipital but are otherwise widely sepa- rated. In all the genera I have examined, the parietals are separated by the supraoccipital. In Brotulotaenia and in Enchelybrotula the parietals are extensive, being half the length of the frontals (Cohen, 1974; 1982) and although there is some variation in size in the genera examined, most do not exceed 25% of the frontal length. In Acanthonus the parietals are widely separated by the supraoccipital (the separation almost equals the width of the parietal; Fig. 13A); such wide separation occurs only in Cataetyx and Bathyonus (Figs 13C & 14A), taxa with broad and depressed crania. There is considerable variability in frontal morphology among the taxa examined. Acanthonus differs from other genera in having the anterior border of the frontal laterally flared. In other taxa the frontal tapers anteriorly. The excep- tion is Glyptophidium (Fig. 14E) where the frontal has a broadly rounded anterior margin and a straight lateral bor- der. Genypterus, Brotula (Fig. 15B), Sirembo and Spectrun- culus (Fig. 14B) have an extensive laterally open frontal canal. Anterior frontal crests, arising at the point where the bones meet the ethmoid bloc occur in Lamprogrammus, Monomitopus, Neobythites, Sirembo and Glyptophidium, the latter also having posterior crests at the junction of the frontal and supraoccipital (Fig. 14E). Frontal crests similar to those of Glyptophidium also occur in carapid ophidiforms (Fig. 21B of Onuxodon in Markle & Olney, 1990). A deep anterior medial cavity in the frontal midline, similar to the ‘mucosal cavity’ of gadoids occurs in Monomitopus and Lamprogrammus (Fig. 13B). The autosphenotic of Acanthonus has a long, thick posteri- orly curved lateral process (Fig. 13A). Of the taxa examined, only in Spectrunculus (Fig. 14B), Monomitopus (Fig. 14C) and Brotula (Fig. 15A) is the sphenotic process directed posteriorly but in these taxa it is small and spine-like. In other genera the process is either laterally or anteriorly directed. In Genypterus the sphenotic extends anteriorly to halfway beneath the length of the frontal (Fig. 15B). The pterotic of Acanthonus has a prominently rounded posterior margin, a condition approached only by Bathyonus (Fig. 14A). In other genera examined the pterotic has a blunt or pointed posterior border. Regan (1929) surmised that direct contact between frontal and parasphenoid diagnosed ophidioids (and blennioids); he had earlier (1912) illustrated this condition in Brotula where the frontal meets the anterior part of the parasphenoid ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES 107 ofl nl Op! of; pill. Ul GOTT Zh op! ofl oe In eau gqynunn gunn titi a dRLA G Fig. 12 Cranial nerve courses (semi-diagrammatic) in A, Neobythites gilli; B, Lamprogrammus niger, C, Cataetyx sp., D, Bassozetus sp., E, Genypterus blacodes, F, Penopus sp. Trigeminal branching external to the facialis foramen (ventrolateral view), G, Hypothesised shift of branching point of infraorbital and supraorbital nerves. Left, plesiomorphic ophidiiform condition (circular outlines represent foramina); right, derived condition with exit via single foramen and branching external to it. Shaded strips denote cranial wall. 108 ascending process. Gosline (1953) also recorded the condi- tion in Brotula and Dinematichthys. Of the taxa I have examined, apart from Brotula (Fig. 10C), the feature occurs only in Sirembo and is thus certainly not diagnostic for ophidiiforms. However, among the Carapidae the frontal appears to contact the parasphenoid ascending process in nearly all the taxa illustrated by Markle & Olney (1990), the exception being Encheliophis. In view of the established monophyly of the Carapidae (Markle & Olney, 1990), the similar condition in the ophidiids (Brotula, Sirembo) and bythitid (Dinematichthys) appears to have arisen indepen- dently. asp G.J. HOWES Olfactory nerve tracts Gosline (1953) noted that the olfactory nerve pathway dif- fered between Brotula and Dinematichthys. In the former the tracts are enclosed in a canal formed by the frontal and diverge from one another anteriorly, while in the latter, the tracts are free and diverge immediately from the olfactory lobes of the brain. In the taxa examined the variability is similar. In Acanthonus the tracts lie parallel for some distance before diverging (Fig. 11); the width of the cranial cavity allows a broad, unimpeded divergence of the tracts which lie medial to the sagittal interorbital membranous septa. This situation obtains in Bassozetus, Lamprogrammus, Glypto- phidium and Genypterus (Figs 12B, D & E) but in other Fig. 13. Crania in dorsal views of A, Acanthonus armatus; B, Lamprogrammus niger; C, Cataetyx sp. In A, the posttemporal is indicated by dashed lines; below is a ventral view of the occipital condyles to show the extension of the exoccipital facets beyond the basioccipital facet. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES genera the tracts are broadly divergent. In part these patterns are due to the distance between the forebrain and the olfactory organs which in Brotula is short, principally on account of the brain being particularly large and extending well forward. In Monomitopus a somewhat different situation occurs in that the forebrain enters the orbital cavity, the tracts running parallel for a short distance then diverging. Position of glossopharyngeal foramen Gosline (1953) referred to the varying position of the glossopharyngeal nerve foramen, apparently perforating the exoccipital in Dinematichthys and the intercalar in Brotula. Patterson & Rosen (1989) report that in all ophidiiforms they examined the foramen occurs in the intercalar. In Acantho- nus an external examination of the cranium fails to locate the glossopharyngeal foramen. The fretted and honeycombed nature of the bone disguises any small foramina and its papery thinness is easily accidentally pierced giving rise to artificial ‘foramina’. By following the route of the glossopha- ryngeal nerve within the cranial cavity it can be ascertained that the nerve exits from the posterodorsal border of the intercalar above the articulation of the lower limb of the posttemporal (Fig. 8). In other taxa examined for this feature the foramen is usually located in the centre of the intercalar either beneath or, more rarely, posterior to the site of attachment of the posttemporal limb. IC exo 109 Exoccipital, occipital condyles and first neural arch Rosen & Patterson (1968: 425) noted that one of the features distinguishing gadiforms and ophidioids was the presence in the latter of a ‘. . . complex basioccipital joint with the first vertebra, involving the formation of a bony arch between the foramen magnum and the basioccipital facet’. Rosen (1985: 29) recognised that the batrachoid-lophiiform and ophidiiform-gadiform groups have the first neural arch and spine ankylosed with the first centrum and joined firmly to the cranium. Rosen (op. cit.: 50) in referring to gadiforms and ophidiiforms also noted that in some cases (specific examples were not cited) neural arches are incorporated into the exoccipitals and supraocciptal. Howes & Crimmen (1990) recognised such an incorporation in the neobythitine ophidii- form Lamprogrammus in which they identified both an incorporated neural spine and arch which formed the ‘exoc- cipital’ condyle and the first centrum which formed the ‘basioccipital’ condyle. Although it is evident in all the ophidiiform taxa I have examined that there is a complete arch forming the rear of the cranium and bordering the posterior margin of the supraoc- cipital it appears that it is the occipital arch and not an incorporated neural arch. Evidence that it is the exoccipital is the presence of a suture with the entire length of the basioccipital and an unbroken suture with the supraoccipital. In many specimens the suture with the basioccipital is almost impossible to discern but the maceration of crania in both KOH and by boiling in water readily reveals the exoccipitals exo Fig. 14 Crania in dorsal views of A, Bathyonus sp.; B, Spectrunculus grandis; C, Monomitopus metriostoma; D, Neobythites steatiticus, 1B, Glyptophidium macropus. 110 asp asp par exo Fig. 15 Crania in dorsal views of A, Brotula multibarbata; B, Genypterus blacodes. as single elements. Furthermore, in cleared and stained specimens cartilaginous laminae border the bones along the midline where they meet to form the floor of the foramen magnum. A foramen lying posterior to that of the vagus nerve transmits the occipital nerve trunk. The posterodorsal extension of the exoccipital with conse- quent exclusion of the supraoccipital from the posterior cranial margin is a feature unique to ophidiiforms. In the Carapidae although the exoccipital is dorsally extended it does not rise high enough to occlude the posterior tip of the supraoccipital (Fig. 18C). Rosen (1985: 50) noted that a possible synapomorphy between gadiforms and some ophidiiforms is the position of the exoccipital facets with respect to the basioccipital and the corresponding anterior extension of the prezygapophyses of the first centrum onto the occiput to meet the exoccipitals. Of the taxa examined here only in Acanthonus do the exoccipital condyles extend posteriorly beyond the basioccipital facet (Figs 13A, 18A), a feature otherwise peculiar to pediculate paracanthopterygians (cf. Rosen’s 1985 fig. 35B and D). In all other taxa the posterior edge of the basioccipital facet lies in the same vertical plane or slightly beyond that of the exoccipital facets. Rosen (1985) and Patterson & Rosen (1989) recognised two groups of ophidiiforms by their different occipital condyle morphology: Ophidiidae possess a plesiomorphic ‘planar’ arrangement of the exoccipital facets and Bythitoidei ~ have derived ‘cod-like’, widely separated, tubular and cartilage-filled condyles. Rosen (1985, Fig. 36) illustrated Ophidion and Ogilbia as representing respectively the ophi- diid and bythitoid conditions (Rosen’s drawing of Ogilbia is reproduced here as Fig. 17B). Patterson & Rosen (1989) G.J. HOWES Fig. 16 Cranial-vertebral articulation in A, Neobythites steatiticus and B, Ophidion rochei (lateral views). observed the planar condyle feature in Brotula, Genypterus and Neobythites as well as Ophidion but they did not list those taxa in which the supposed ‘cod-like’ articulation occurs. In the taxa examined here I do not find these marked differ- ences (cf. Figs 17 with 16 and 18). In Ophidion (Fig. 19D) the exoccipital facets are rectally deep and are united to form a continuous articulatory sur- face; the lateral part of each facet is extended posteriorly into a ‘condyle’ and turned ventrad. The basioccipital facet is separated from the exoccipital by a deep forwardly directed cavity (boc). With two exceptions (see below) all other taxa (including the bythitoids Ogilbia and Lucifuga) have a ‘pla- nar surface formed by medially united exoccipitals (Fig. 19A,B) although not as deep as in Ophidion; the lateral, condylar part of the exoccipital (exc) is variously angled and in Glyptophidium (Fig. 19F) faces directly ven- trad. The basioccipital facet is separated from the exoccipital by a dorsally directed bursa-like cavity (boc) which accepts the ventral facet of the first centrum (clp, Figs 16B, 19D); see also Rosen & Patterson, 1968, fig. 10D of Dinematichthys and Howes & Crimmen, 1990, fig. 27 of Lamprogrammus. Apart from Ophidion the two exceptions to this condition are Genypterus (Fig. 19E) and Acanthonus (Fig. 19C), the only two taxa which have medially separated exocciptal facets. In Genypterus the separation is marginal but in Acanthonus it is extensive, and the facets appear as true ‘cod-like’ condyles. In Acanthonus the exoccipital facets extend beyond the basiocciopital facet (see above) and there is no davity between the basioccipital and overlying exoccipitals. Genypterus however, closely resembles Ophidion in having a horizontal cavity between the basioccipital and exoccipital facets into which inserts the anteriorly directed medial facet of the first centrum (Fig. 16B). In ophidiiforms Baudelot’s ligament is always confined to the basioccipital and is a consistent marker in separating the dorsal and ventral (which posteriorly become lateral and medial) sections of the epaxial muscle which inserts respec- tively on the anterior ribs and medial region of the swimblad- der (see below). In Acanthonus Baudelot’s ligament stems not from a fossa on the underside of the basioccipital but from the rim of the facet (Fig. 18A). The first neural arch and spine display several modifica- tions amongst ophidiiforms. In Ophidion, Brotula and Genypterus the bases of the neural arch are flattened anteri- orly and expanded laterally which in Ophidion form wing-like ANATOMY AND CLASSIFICATION OF OPHIDITFORM FISHES so exo r1(epr) hyx 111 nal fsn sb Fig. 17 Cranial-vertebral articulation (left, lateral views) in A, Ogilbia cayorum, also showing swimbladder connections (drawn from dissected and stained specimen); B, copy of Rosen’s (1985) figure 36B showing the supposed designation of the first neural arch (guide-line absent in the original figure) and ‘cod-like’ condyles. structures to which attaches a segment of epaxial muscle (Fig. 16B; Rose, 1961 and below). In some taxa (e.g. Ogilbia, and Diplacanthopoma, only the anterior base of the arch is autogenous with the centrum (Fig. 17A). In others including Acanthonus, Bathyonus and Penopus, (Figs 18A; 24B & C) the arch is reduced to a slender bone, the base of which is supported by only a narrow pedestal on the lateral ridge of the centrum. In other taxa (Lucifuga and Spectruncu- lus, Figs 23A; 25A) the neural spine has been lost, leaving the base of the neural arch autogenous and attached by post- and prezygapophyseal ligaments respectively to the succeed- ing neural arch base and the epioccipital. It seems that among ophidiiforms and pediculates the skull has been posteriorly extended, in the former group by the dorsoposterior expansion of the exoccipital and in the latter by expansion and ankylosis of the first neural arch. Although there are similar modifications to the exoccipital-vertebral contact in pediculates and ophidiiforms the resemblances are not as close as recognised by Rosen (1985) and Patterson & Rosen (1989) so as to unite pediculates with bythitoids. As noted above, in almost all ophidiiforms the exoccipital articu- latory facets form a continuous medial surface and the lateral margins retain a condylar form even though the articular surfaces may be variusly shaped and angled. In pediculates, however, there is no continuous exocciptal facet and the lateral ‘condyles’ are so modified as to have lost their condylar nature and in those taxa where they appear strut- like, their articular surfaces are sutured with those of the prezygapophyses of the first vertebra. There is no truly ‘cod-like’ cranial-vertebral articulation in any ophidiiform examined here and cartilage cores are present in all their exoccipitals. The most notable feature of the ophidiiform articulation is the presence in the majority of taxa of the angled bursa-like cavity lying between the exoccipitals and basioccipital and which occurs in all ophidiiforms and cara- pids (Fig. 18C). Markle & Olney (1990: 277) have drawn attention to this feature in referring to the medial ventral facet of the first centrum and it would seem that the modified basioccipital-vertebral articulation is a synapomorphy for ophidiiforms (including Carapidae). Swimbladder and its connections Acanthonus lacks a swimbladder, in which respect it appears to be derived (see below). The 28 genera examined for this feature display a plethora of conditions of swimbladder- vertebral column association. The variable nature of the association is revealed in muscle attachments, number and degree of hypertrophy of anterior ribs, their position on the vertebral column and features of the swimbladder itself. In addition the nature of the first free neural arch (discussed above) appears to be intimately connected with specific types of swimbladder-vertebral associations. There are two basic types of swimbladder-vertebral con- nections: (1) having the first (enlarged) rib attached to the first centrum and it and ribs 2 and 3 thickened with ribs 2 and 3 often expanded, extending horizontally and partially covering the anterodorsal surface of the swimbladder; the distal tips of all 112 ) G.J. HOWES exo Fig. 18 Cranial vertebral articulation in A, Acanthonus armatus, also showing associated musculature (left, lateral view); B, Barathronus bicolor, membrane (hatched area covers space between exoccipital and first neural arch and spine; right side, lateral view; drawing reversed); C, Carapus bermudensis, (left, lateral view of cleared and stained specimen). three ribs are attached to the swimbladder wall and well- developed muscles (lateral and medial) run between the cranium and first rib-swimbladder wall. The swimbladder is situated well forward, its anterior wall thickened by sclerifica- tion or ossification and sometimes the outer wall forms a flap or ‘door’ or separate elements (rocker bones) which are ligamentously attached to the ribs. A subgroup of taxa can be distinguished in which the first rib, although thickened is never expanded. (2) having the first (large) rib attached to the second or third centrum, it and ribs 2 and 3 normal, vertical or somewhat posteriorly directed, all are free from the swimbladder. A narrow, outer band of epaxial muscle connects the first rib to the basicranium and a medial band runs to the dorsoanterior wall of the swimbladder. The swimbladder is situated far back, between the fifth and eighth centra and its anterior wall lacks any sclerification or thickening. Some of the variation within these groupings is detailed below. Group 1. Examined taxa with these features are Brotula, Cherublemma, Genypterus, Hypopleuron, Lepophidium, Ophidion, Parophidion Rose (1961), Svetovidov (1961) and Courtenay (1971) have made detailed descriptions of the swimbladder and its cranial- vertebral connections in Ophidion (given as Rissola by Courtenay). These authors showed that sexual dimorphism is exhibited in the morphology of the swimbladder connections, the principal differences being the presence of an anteroven- tral element in males termed the ‘rocker bone’ by Rose (1961). The anterior pair of ribs are extended into wing-like structures which support the edges of the rocker bone (Fig. 16B). The expanded third rib forms a bony sheet ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES exo 113 | Fig. 19 Posterior views of occipital condyles of A, Neobythites steatiticus; B, Ogilbia cayorum; C, Acanthonus armatus; D, Ophidion rochei; E, Genypterus blacodes; F, Glyptophidium macropus. covering the anterior region of the swimbladder. In these taxa the first neural arch is also modified (see above). Apart from Ophidion, Brotula is the only other examined genus to possess a rocker bone. In Cherublemma (Fig. 20) the third and fourth-sixth ribs (possibly a combination of vertebral parapophysis and rib) are enlarged to form an ossified shield over the anterior part of the swimbladder; the antericr wall of the swimbladder (tunica externa) is almost completely excised, being attached by a ventral hinge and forming a ‘trap-door’ (sbc, Fig. 20) which is attached on either side of its upper rim to the tip of the first rib by a ligament (Icc, Fig. 20). When opened the door exposes a fenestra covered by the tunica interna (ti, Fig. 19B). The ribs of the third vertebra join together in the ventral midline to form a broad bony surface with paired depressions on which the dorsal lip of the ‘door’ closes. Hyplopleuron (Fig. 21) also has the first-third ribs expanded to the same degree; the tunica externa of the swimblad- der is dorsally sclerified and laterally and ventrally the swimblad- der is sequentially constricted, resembling the body of a caterpillar. 114 G.J. HOWES Fig. 20 Cherublemma emmelas (female). Anterior vertebral column and swimbladder in A, right lateral and B, ventral views. In B the cap of the swimbladder is shown reflected and with the ligaments cut away. Owing to its expanded first rib and neural spine, Lepophid- ium (Fig. 22A) is also included in this group but there are several differences between it and the other included taxa. Ribs 2 and 3 are slender and rib 1 is attached to the head of the swimbladder by a thick, semi-ossified ligament which runs from the distal tip of the rib to bifurcate behind the second rib, the two branches (of normal ligamentous consistency) attaching to respective sites on the anterolateral and dorsal areas of the swimbladder wall. Taxa recognised as forming a subgroup (see above) are: Brosmophyciops, Dicrolene, Glyptophidium, Lamprogram- mus, Monomitopus, Neobythites, Ogilbia, Pycnocraspedium, Petrotyx, Sirembo. In these genera the first-third ribs are stout but the first, unlike that of other Group 1 taxa, is not produced into a wing-like structure but resembles them in having the distal tips of the ribs ligamentously united and attached to the wall of the swimbladder (Fig. 22B). A separate muscle attaching to the medial part of the swimbladder occurs in some taxa (in Brosmophyciops it is particularly well-developed); the lateral band runs from the epi- and exoccipital regions of the cranium to join the compound swimbladder-rib ligament. Lamprogrammus (Fig. 21) has a distinctly modified swim- bladder and attachment. The expanded first-third ribs are entirely enclosed in silvery connective tissue which appears to be an extension of the compound ligament that invests the distal tips of the ribs in other Group 1 taxa; the nature of the tissue suggests that it might even incorporate the tunica externa of the swimbladder. The swimbladder itself is finger- like with a bulbous anterior cap, partially detached and hinged on its underside to the body of the bladder (sbc). When closed the dorsal rim of the cap is covered by a strong lip extending from the dorsal midline of the body of the swimbladder. A segment of the compound ligament attaches to either side of the cap. A paired muscle (epm, Fig. 21B,D) runs from the basicranium to insert directly on the swimblad- der cap either side of the midline. The outer section of muscle (epl, Fig. 21B) runs to the proximal part of the first rib. A thick median ligament (Isv, Fig. 21B,C) stretches from between the ventrally directed parapophyses of the fifth centrum to the anterior lip of the swimbladder. The interior of the swimbladder is packed with dense, almost fibrous material. In Glyptophidium the anterior ribs have a similar associa- tion with the swimbladder as in Lamprogrammus except that the swimbladder is more extensive and the tunica externa is firmly united to the underside of the centra by thick connec- tive tissue. The swimbladder has a well-developed sclerified lip and is large and heart-shaped with a ventral opening exposing the tunica interna. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES nal epl ste sp nal Fig. 21 Swimbladder and its vertebral attachments in A, Hypopleuron caninum (male), right lateral view; B-D, Lamprogrammus niger (male), in B, right lateral view; C and D, dorsal and anterior views of swimbladder. In A, the ribs are shown as they appear upon superficial dissection, enswathed in ligamentous tissue; in D, the cap of the swimbladder is shown reflected. 115 116 G.J. HOWES ~b/ rd sbec Ic Fig. 22 Swimbladder and its vertebral attachments in A, Lepophidium cf. profundorum (female); B, Monomitopus metriostoma (female), right lateral views. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES Petrotyx and Ogilbia stand somewhat apart from the other taxa assigned to this group in having the swimbladder less directly connected to the anterior three ribs (Figs 17A, 22C). In Ogilbia (Fig. 17A) the first rib is scarcely thicker than the second although the third and fourth are proximally expanded. In Petrotyx (Fig. 23) all three anterior ribs have some expansion, the second only along its proximal part, and both the lateral and medial bands of epaxial muscle insert on the first rib although upper fibres of the medial band also insert on the second rib. The upper part of the outer (dorsal) muscle band passes over the first rib to become continuous with the lateral body epaxial muscle. Luciobrotula possibly also belongs with this group. The second and third ribs are expanded and joined with the first to the all of the swimbladder by a compound ligament to which a lateral epaxial muscle attaches; there is no medial swimbladder muscle in the specimen examined (a female). Group 2. Examined taxa with these features are: Abyssobrotula, Barathrodemus, Bathyonus, Bassozetus, Cataetyx, Diplacanthopoma, Penopus, Spectrunculus Taxa of this group exhibit a broad spectrum of conditions but are united on the basis of lacking any great enlargement of the anterior three ribs and in having the first (longest) rib attached to the second or third centrum. The swimbladder is usually located in a posterior position and its anterior rim lies below the fifth-eighth vertebrae; it is elongate and extends to the 20—24th vertebrae. The muscles linking the swimbladder with the cranium have variable morphology; in Penopus (Fig. 24A) the muscle has its cranial attachment to the intercalar, pterotic and exoccipital and is divided by Baude- lot’s ligament, the outer segment attaching to the distal portion of the first rib, the medial segment extending caudad to insert on the dorsal surface of the swimbladder. The tips of the ribs are interlinked with hypaxial muscle but remain free of any direct connection with the swimbladder; in Bathyonus the tip of the 1st rib is attached to the postcleithrum (Fig. 24B). Bathyonus and Bassozetus have a similar arrange- ment except that in the former the medial muscle is attached to the swimbladder by a long thin tendon (Fig. 24B). In female Spectrunculus (Fig. 25A) the medial segment of the muscle (it passes ventral to Baudelot’s ligament) attaches not to the swimbladder but to the peritoneum to which the anterior wall of the swimbladder is itself connected. A pair cf large, medial ventral projections, joined in the midline, extend from the fifth centrum in front of which the retractores dorsales take their origin. In the male (Fig. 25B) these medially united vertebral processes extend posteriorly in the form of a long deep strut to beneath the eighth or ninth centrum and the lateral epaxial muscle band inserts along the face of the strut. Separate, paired muscles run horizontally from the posterior border of the united vertebral processes above the strut inserting on either side of the dorsal midline of the swimbladder. This muscle is assumed to be a ‘detached’ part of that medial segment of the epaxialis which inserts on the vertebral ventral strut in the male and the peritoneum in the female. In Cataetyx and Diplacanthopoma (Fig. 26) there is no separate medial muscle linking the swimbladder (both males and females examined), instead the lateral segment of the epaxial muscle inserts on a compound ligament with the distal 117 tip of the first rib and the swimbladder wall. This situation is similar to that in taxa of the subgroup in Group 1 but differs in that the first nb of Diplacanthopoma and Cataetyx is curved anteriorly and a band of epaxial muscle distinct from the lateral myomeres runs diagonally backwards from the point of attachment of the first rib to the fourth or fifth rib (eps). My observations on a single specimen of Barathrodemus sp. indicate that it belongs within this group of taxa but they also conflict somewhat with those of Carter & Musick (1985) on many specimens of B. manatinus. I do not find a separate pair of muscles connecting the fourth pair of ribs with the prootic; the only muscles in this position are the retractores dorsales which extend from the fourth centrum to the pharyn- gobranchials. Furthermore, in my specimen (a male) the medial muscles connecting the cranium with the swimbladder are separated from the lateral bundle by Baudelot’s ligament (the usual condition; see above) and appear to stem from the exoccipital rather than the prootic. The muscles are tendi- nously joined to the anterodorsal surface of the swimbladder which lies below the eighth or ninth centrum and not the third as indicated in Carter & Musick’s specimens. In most respects Barathrodemus most closely resembles Penopus. Although Acanthonus has no swimbladder, features of the vertebral column such as the slender anterior ribs, the first rib articulat- ing with the third centrum and the first neural arch reduced and basally supported on a narrow pedestal (Fig. 18A) indi- cate its inclusion in this group. That Acanthonus has second- arily lost a swimbladder is suggested by the insertion of the lower (medial) segment of epaxial muscle to the first rib, rather than the upper (lateral) segment. Since it is the medial segment which inserts on the swimbladder, indications are that this segment has ‘replaced’ the original lateral connec- tion. Lucifuga (Fig. 23A,B) resembles taxa of Group 2 in having unexpanded ribs, however, the first rib associated with the swimbladder articulates with the base of the first centrum and the swimbladder is situated anteriorly as in Group 1 (Fig. 23A). The lateral muscle band attaches to the distal tip of the first rib and the swimbladder wall and there is no medial swimbladder muscle (in the male specimen exam- ined), all features of the subgroup of Group 1 taxa. The lateral muscle band is, however, undivided by Baudelot’s ligament which suggests that the lower and medial part of the muscle normally inserting on the swimbladder has been lost. Furthermore, Baudelot’s ligament passes between two branches of the vagus nerve (Fig. 23A). In the only aphyonid examined for these features, Barath- ronus (Fig. 18B), there is no swimbladder, the first pair of ribs articulating with the third centrum but having no muscu- lar connections. According to Nielsen (1969) aphyonids lack ribs but the articulation with the ventral cavity of the 3rd centrum indicates that these processes are pleural ribs (see below). Markle (1989) states that ophidiiforms typically have epi- pleural ribs on at least the first two centra and pleural ribs on all succeeding centra, citing Porogadus as exceptional in lacking epipleurals from the first two centra (Carter & Sulak, 1984). However, one group of ophidiiform, classified above as Group 2, consistently lack ribs from the first and often the second centrum and when ribs do occur they are short (e.g. Spectrunculus, Fig. 25A). Markle (1989), Markle & Olney (1990) and Patterson & Rosen (1989) consider the anterior ribs to be epipleurals, a 118 G.J. HOWES /nr Fig. 23. Swimbladder and its vertebral attachments in A, Lucifuga dentata (male), right lateral view; B, dorsolateral view of the first neural arch; the near-side has been removed to expose medial aspect of opposite half of the arch and the central spinal nerve and its associated branch; C, Petrotyx sanguineus (sex not determined), right lateral view. 119 ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES epm nalts Ny 21) pe 3 . ATI) YS . r1 (ptr) en Se ee) P24 A Ki gut nal (female), right lateral views. Fig. 24 Swimbladder and its vertebral attachments in A, Penopus sp. (male); B, Bathyonus sp. 120 : a ene S ie S / UME 05) ae Rie yy — a ye es (ae ‘ | =e Yue as ?epm i i rd on 5, G.J. HOWES epr Fig. 25 Swimbladder and vertebral attachments in Spectrunculus grandis. A, female; B, male, dashed line indicates margins of vertebral process, right side lateral views. ANATOMY AND CLASSIFICATION OF OPHIDITFORM FISHES eps Fig. 26 Swimbladder and vertebral attachments in Diplacanthopoma brachysoma (sex not determined), right side lateral view. view based principally on their topographical arrangement (see caption to fig. 8 in Patterson & Rosen, 1989). I would suggest that a further indication of their identity is given by the nature of their articulation, namely to the base of the neural spine (in Ogilbia cayorum the first epipleural articu- lates directly with the stem of the neural spine, Fig. 17A). In taxa of Group 2 all the ribs, apart from those on the anterior centra of Spectrunculus and Bathyonus, are pleural, that is they articulate with the ventral cavity of their respective centrum (Rosen, 1985: 50), none are expanded or directly associated with the swimbladder. Carter & Musick (1985) identify both a stout pleural and an epipleural rib associated with the first centrum in female Barathrodemus whereas in the male only an epipleural is present. Markle & Olney (1990) state that ligaments connecting the epipleural ribs are absent in Brotula but present in the other ophidiids they examined; they also identify a compound swimbladder ligament uniting the tips of the anterior epipleu- rals as a synapomorphy for Carapini. I find interconnecting epipleural ligaments in all ophidiiforms examined, including Brotula, and a compound swimbladder ligament in those taxa assigned to the subgroup of Group 1. To summarise the essential features of the swimbladder- vertebral column association: two assemblages of taxa can be recognised on the relationship of the swimbladder with the anterior (epipleural) ribs: those which have direct contact between the swimbladder and the thickened and expanded ribs (Group 1 above) and those where the swimbladder is isolated from the ribs which are nearly all pleural and never expanded (Group 2 above). Both assemblages contain taxa in which muscles connect the swimbladder with the cranium and in which sexual dimorphism is apparent in the presence/ absence of medial muscles. In the assemblage with expanded epipleurals there are modifications to the anterior wall of the swimbladder taking the form of sclerification and ossification (rocker bone and hinged openings). The same two assemblages can be recogn- ised on the basis of the first neural arch morphology. In that assemblage with expanded ribs and modified swimbladder it is autogenous, usually thick and posteriorly inclined, whereas 21 in the other assemblage the arch is usually separated into its two halves which are reduced in size joined ligamentously across the midline and connected to the centrum by a narrow bony or cartilaginous pedestal. It is assumed, based on the condition in other paracanthopterygians, that the expansion of epipleural ribs and their intimate contact with the swim- bladder is a synapomorphy as are the various modifications to the anterior wall of the swimbladder. Similar features of the anterior vertebrae and swimbladder, including the develop- ment of a rocker bone, also occur in Carapidae (Courtenay & McKittrick, 1970; Markle & Olney, 1990). Markle et al. (1983) suggested that the paired sclerified structures of the swimbladder to which the anterior pleural ribs attach (Rose, 1961) may represent the precursor of the rocker bone. The presence of a semi-ossified ligament in Lepophidium (see above) which connects the swimbladder to the first epipleural rib suggests that such a structure could be implicated in the evolution of the bone. Markle & Olney (1990) dismiss the rocker bone and sclerification of the swimbladder in carapids as being synapomorphic with other ophidiiforms on the grounds that carapid monophyly is attested by a suite of six other synapomorphies, none of which occur in members of the wider group. The modification of the first neural arch in both assem- blages (expansion and thickening in one and reduction in the other) are seen as alternative derived states from the plesio- morphic slender, caudally sloped condition (Markle & Olney, 1990: 278). Paired muscles having their connections with the swimblad- der are common to both assemblages although a complex, medial division is absent from one subgroup (in Group 1 above) and some taxa of Group 2. The muscles identified here as lateral (dorsal) epaxial (epl) correspond to Rose’s (1961) muscle M2 + M3 and Courtenay’s (1971) ‘dorsal sound producing muscle’ and the medial (ventral) epaxial (epm) to Rose’s (1961) muscle M1 (which in Ophidion and Brotula connects the rocker bone to the prootic) and Courte- nay’s (1971) ‘ventral sound producing muscle’. Similar mus- cles in the Carapidae are named by Courtenay & McKittrick (1970) as respectively the secondary and primary sound producing muscles. The lateral (dorsal) band appears to be a segment of that part of the anterior epaxial muscle which has its attachment primitively to the posterior cranial roof and inserts on the first neural arch (see fig. 5 in Courtenay, 1971); the medial (ventral) segment likewise appears to be that part of the epaxialis which is primitively divided from the lateral part by Baudelot’s ligament (see above), interconnects the ribs and its innervated by one of the occipital nerves. The presence of homologous muscles in Carapidae suggest they are primitive for Ophidiiformes sensu Cohen & Nielsen (1978). The retractores dorsales muscles commonly originate from the third centrum in most ophidiiforms and Carapidae (Mar- kle & Olney, 1990) but in taxa of one of the two assemblages recognised above as Group 2, the muscle originates from the fourth or fifth centrum (in Spectrunculus it runs from the anterior border of the united ventral vertebral processes, Fig. 25). It is beyond the scope of this paper to attempt ascribing functional attributes to the various swimbladder arrange- ments described here. Suffice it to say that the swimbladder, anterior ribs and first free neural arch all appear to be elements of a single functional unit. Modifications to the first neural arch may have been signaled by the ‘release’ of the 122 constraints imposed by the epaxialis due to its reorientation to the anterior ribs. The absence of a swimbladder is, judging by its widespread presence among ophidiiforms, a secondary loss which appears to have occurred independently in Acan- thonus, Typhlonus and aphyonids. It is noted that the two morphological groups recognised above correspond with ecological groupings. Group 1 are taxa which all inhabit shallow to mid-depths while Group 2 are taxa which are all deep to abyssal. Cranial muscles There are three features in which Acanthonus differs from other ophidiiforms: (1) almost the entire lateral face of the levator arcus palatini (LAP) is exposed, (2) muscle A1b is reduced to a narrow band originating lateral to LAP, (3) the dilatator oper- culi (do, Fig. 27) is extensive and exposed dorsally. epx do mbo G.J. HOWES In all other taxa examined much of LAP is laterally covered by the adductor muscle (Ala and A2) but in Acantho- nus only the lower border of the Jevator is covered by the adductor (Fig. 27). LAP is posteriorly divided and the hyo- mandibularis nerve runs through this division before bifurcat- ing into mandibular and hyoid branches medial to the adductor; a separate, thin external branch of the mandibu- laris breaks through the muscle to innervate the tissue of the posterior part of the dentary. In other taxa the mandibularis externus is well-developed. Among the taxa examined, the size of Acanthonus LAP is equalled by Xyelacyba and Taure- dophidium and exceeded only by that of Spectrunculus (Fig. 28). In Acanthonus muscle A1b is shallow and thin, originating from the anterior border of the hyomandibular and overlap- ping the anteroventral tip of LAP. In other examined taxa muscle Alb is usually deep and thick, having a bifurcate Fig. 27 Acanthonus armatus, right lateral view of head showing musculature; the operculum has been elevated to expose the suboperculum and so obscures the adductor operculi muscle. Nerves passing medial to muscles are indicated by dashed lines. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES origin from either side of the LAP (Fig. 28). The medial segment stems from the metapterygoid or even the ectoptery- goid and in Bassozetus, Dicrolene and Spectrunculus (Fig. 28B) joins the ventral edge of the adductor arcus palatini; in Genypterus some fibres of Alb even run from the palatine. The lateral portion always originates from the hyomandibular, which in some taxa (e.g. Bassozetus, Lam- programmus; Howes, 1988, Fig. 25) may be the posterior border of the bone. nmd 123 Muscle Ala was defined by Howes (1988) as that lying lateral to the ramus mandibularis of the trigeminal nerve even though that element might insert on the lower rather than the upper jaw (a more usual definition of Ala is that it inserts on the upper jaw). In ophidiiforms, Howes (1988) reported that the ramus mandibularis passes through the outer muscle bloc and so that section of the muscle lying lateral to the nerve is, by definition, Ala whereas that lying medial to it is A2; Acanthonus is no exception to this situation and the major lap pap A> Fig. 28 Facial muscles and nerves of Spectrunculus grandis. Above, right lateral view; below, dorsal view of right muscle Alb showing its relationships with surrounding musculature. Nerves and medial extent of muscles indicated respectively by short and long dashed lines. 124 part of the adductor muscle which lies medial to the mandib- ularis nerve and inserts on the lower jaw is regarded as A2 (Fig. 27). As in gadiforms, muscle Alb lies lateral to the ramus mandibularis, or at least its outer division does; the divided ‘Alb’ of ophidiiforms appears to be synapomorphic for the group. A similar modificiation of ‘A1b’ is a diagnostic character of supragadoids (sensu Howes, 1991b) where ‘A1b’ lies entirely medial to A2 (Howes, 1988). Acanthonus dis- plays a secondarily derived condition in having only an outer segment of Alb, the inner having been lost. The backward looping of the ramus mandibularis around muscle Alb in Ophidion (Howes, 1988: 35, Fig. 23A) sets this genus apart from other ophidiiforms examined. In Acanthonus the dilatator operculi (do, Fig. 27) is exten- sive and occupies a cranial fossa in the pterotic and posterior part of the sphenotic. It is not covered by the adductor mandibulae or LAP muscles. In all other ophidiiform taxa examined, with the exception of two (see below) the DO runs behind the adductor and LAP muscles and originates from a narrow pterotic fossa situated on either the lateral or ventro- lateral surface of the bone (Fig. 28, top). Only in Tauredo- phidium and Xyelacyba is the DO expanded and lies in a dorsally situated fossa and in the former the muscle is even more extensive than in Acanthonus. Pelvic girdle and innervation In ophidiiforms the pelvic girdle, when present, is reduced to a small triangular bone (in the Carapidae examined it is a slender rod with a cartilaginous anterior tip sandwiched between the tips of the horizontal cleithral limbs, making it in effect, a ventral appendage of the pectoral girdle (Fig. 29B). The anterior shift of the pelvic girdle has also involved modification of the muscles and nerves which motivate the pelvic fin rays. The sternohyoideus muscle, as in other teleosts, runs from the lower part of the cleithrum and inserts along the urohyal, but unlike the common situation, a ventromedial segment detaches completely or partially from the main body of the muscle, and its ventral attachment is to the cleithral symphysis above the pelvic bones (Fig. 29A). In most ophidiiforms the posterior border of the urohyal is near to the cleithral anterior margin and the medial segment of the sternohyoideus lies, for the most part, between the two outer segments. In Acanthonus, however, the urohyal is reduced in size and widely separated from the cleithrum and the medial segment of the sternohyoideus is well-separated from the lateral elements which extend further posteriorly along the cleithral limb (Fig. 29A). The pelvic muscles are restricted to the pelvic bone and lack any connection with the sternohyoi- deus. Hypaxial muscle runs forward between the cleithra and attaches to their medial walls; the infracarinalis anterior also extends forward to attach to each pelvic bone (Fig. 29C). Other than Acanthonus, the distinct separation of the medial section of the sternohyoideus is present only in Hoplo- brotula, Sirembo and Dicrolene (all members of the ophidiid subfamily Neobythitinae sensu Cohen & Nielsen, 1978) whereas in Brotula (Brotulinae) it is integrated to the same degree as in other neobythitines (i.e. Bassozetus). The situation in Genypterus, Lepophidium and Ophidion (all members of the Ophidiinae sensu Cohen & Nielsen, 1978) is more complex than in other ophidioids. Cohen & Nielsen (1978) recognised the subfamily Ophidiinae princi- pally on the possession of a modified cleithrum where the horizontal limb extends forward as a bony filament, thus G.J. HOWES advancing the pelvic girdle which is supported between the cleithral extensions. It should be noted that it is the medial bony laminae of the cleithra which extend forward and the cleithral tips remain in the approximate position of those of other ophidiiforms, viz beneath the basioccipital, but instead of meeting symphyseally they diverge and are connected in the midline only by hypaxial musculature. Although Cohen & Nielsen (1978) show the pelvic girdle sandwiched between the extended laminae, in the taxa examined here the cleithral bar attaches ligamentously to the pelvic bone (cle, Fig. 30). The sternohyoideus muscle is distally divided into lateral and medial portions. The majority of the posterior fibres of the lateral section attach to the urohyal while the ventral fibres converge into a tendon which attaches to the base of the urohyal; the medial portion runs directly to the inner face of the pelvic bones. The urohyal is extensive and canopy-like, forming a cover above the medial section of the sternohyoi- deus (ut, Fig. 30B); posteriorly the urohyal is firmly attached to the cleithral limbs by the sternohyoideus; anteriorly it attaches by a ligament to the 2nd hypohyal as well as to the ventrohyal (Fig. 30A). Of the taxa examined, Lamprogrammus lacks a pelvic girdle and consequently any medial division of the sternohyoi- deus (Cohen & Rohr, 1992 detect the presence of a rudimen- tary pelvic girdle in L. shcherbachevi); the supracarinalis anterior attaches to the inner cleithral symphysis. The innervation of the ophidiiform pelvic girdle has been described by Freihofer (1970) in Brotula and Ogilbia and by Machida (1988) for Neobythites. Both authors describe a hypertrophied branch of the ramus lateralis accessorius (RLA-PP) exiting from the posterodorsal region of the cranium and following a path along the posterior margin of the cleithrum which turns mesad and meets the third ventral spinal nerve (VSR3) to enter the two pelvic fin rays. In Brotula, Hoplobrotula, Sirembo, Dicrolene, Bathyonus and Spectrunculus the RLA-PP nerve crosses from the cleithrum to the pelvic bone beneath the hypaxial muscle, making it readily visible when the skin is removed (Fig. 30D). Each VSR3 descends to the ventral midline to run together between the infracarinales anteriores before meeting the RLA-PP. In the ophidiines Ophidion, Genypterus, Parophid- ion and Lepophidium, the nerve course is similar except that the RLA-FP and VSR3 pass together between the cleithral limbs and beneath the canopy-like urohyal (Fig. 30). In all other taxa examined (including Ogilbia and Neobythites) the RLA-PP is relatively thin, passes mesad to the cleithrum and hypaxial muscles and meets the vsr3 directly above the pelvic bone, both nerves exiting together between the infracarinales anteriores and entering the pelvic rays. In summary, the respective conditions in the taxa examined are: RLA-PP hypertrophied, running close to cleithral border and exposed laterally and ventrally, joining VSR3 posteri- orly: Brotula, Hoplobrotula, Sirembo, Dicrolene, Bathyonus, Spectrunculus. As above but RLA-PP and VSR3 running together through modified cleithrum and urohyal: Ophidion, Genypterus, Lepophidium. RLA-PP thin, not exposed laterally or ventrally, joining VSR3 anteriorly: Acanthonus, Abyssobrotula, Aphyonus, Barathronus, Bassozetus, Brosmophyciops, Cataetyx, Dipla- canthopoma, Glyptophidium, Hypopleuron, Lamprogram- mus, Lucifuga, Monomitopus, Neobythites, Ogilbia, Penopus, Porogadus. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES ga5 125 ica hyx shl RLA-PEL (od | ica RLA-PP Fig. 29 Pectoral-pelvic girdle associations. A, Acanthonus armatus, right lateral view showing principal muscles. B, isolated pelvic bone with fin ray attached. C, Bassozetus sp., right lateral view of sternohyoideus muscle (dashed lines show limits of the medial part of the muscle and lower border of the urohyal). D, Spectrunculus grandis, ventral view; infracarinalis anterior removed from left side of fish to display nerves. Pharyngohyoideus muscle In all ophidiiforms and carapids examined the pharyngohyoi- deus (rectus communis) muscle extends from the urohyal to the fifth gill-arch (Fig. 29A). In the majority of taxa it runs from the anterodorsal tip of the urohyal and connects both elements with no intermediate attachment but in Monomito- pus and Neobythites it also attaches to the third hypohyal and in Spectrunculus to the fourth. In the ophidiines Ophidion, Genypterus and Lepophidium, the pharyngohyoideus is a broad band-like muscle extending posteriorly from the face of the urohyal and attaching ligamentously to the fourth gill- arch as well as the fifth. Howes (1988) noted that in gadoids and a few macrouroids pharyngohyoideus (rectus communis) attaches directly to the 126 shm shm SSS > Se aes G.J. HOWES ut uh vh RLA-PP pb uh pm ~~ ee vh Fig. 30 Genypterus blacodes, pelvic girdle anatomy. Above, right lateral; below, ventral view. In the lateral view, the course of the RLA-PP nerve beneath the urohyal is indicated by a broken line. In the ventral view the muscles and nerves of the specimen’s right side have been removed to expose the urohyal. anterodorsal tip of the urohyal or via the sternohyoideus (as in ophidiiformes and carapids) whereas in most macrouroids it attaches to the bone’s lateral face. The sternohyoideus mediated condition of the pharyngohyoidus in gadoids and some macrouroids was considered by Howes (1988) as having been derived from direct attachment to the urohyal keel, the (derived) acanthomorph state. In the group of ophidiform genera which possess the latter condition (ophidiines) there is no median keel, but merely a dorsal ridge so that the muscle inserts on the canopy-like base-plate of the bone (see above). Thus, the condition in ophidiines is considered to have been independently derived from that in other acanthomorphs. Anterior mandibular pores Acanthonus, in common with 50% of ophidiiform genera examined has a single opening in the dentary close to the symphysis which is the anterior opening of the mandibular sensory canal. In the other 50% of taxa there is an additional opening on the medial rim of the dentary, posterior to the ventral symphyseal opening. The medial aperture opens into the cleft where the isthmus (protractor hyoideus muscle) joins the dentary symphysis and is often covered by a fold of skin extending along the medial rim of the dentary (in Dinemat- ichthys there are separate flaps of skin covering the medial openings). In Spectrunculus, Monomitopus, Diplacan- thopoma, Ogilbia, Dinematichthys and Lucifuga the medial opening is large and in the latter three genera the dentary rim is notched at the medial opening (med, Fig. 31A). In Sirembo, Hoplobrotula, Ophidion, Lepophidium and Genypterus the medial opening is small and narrowly sepa- rated in the midline from its antimere (Fig. 31B) but in Porogadus and Tauredophidium the medial openings are extensive and widely separated. In Brotula the so-called barbels are tubular extensions of the medial openings which communicate directly with the mandibular canal. ANATOMY AND CLASSIFICATION OF OPHIDITFORM FISHES Fig. 31 Mandibular pores of A, Ogilbia cayorum; B, Sirembo imberis. DISCUSSION Specializations of Acanthonus Acanthonus possesses a series of uniquely derived inter-linked osteological, myological and neurological cranial features. It has a lower jaw-palatine bite, a consequence of the lateral shift of the palatopterygoid articulation by outward curvature of the quadrate, in turn laterally shifted through outward curvature of the dorsal region of the hyomandibular. The relocation of the anterior hyomandibular strut has necessitated the hyomandibu- laris nerve piercing the anterior lamina of the bone rather than the strut itself. The preoperculum gives additional rigidity to the quadrate by its widened horizontal limb. The loss of upper jaw biting strength is reflected in the reduction of adductor muscle Alb whereas the increased importance of lower jaw-opercular series coupling is reinforced by hypertrophy of the dilatator operculi muscle. Howes (1989) hypothesised that the loss of a trigeminal chamber in gadiforms was a response to increased buccal expansion so giving more freedom of passive movement to the nerve tracts. That hypothesis is here extended to the similar condition in Acanthonus where expansion of the buccal cavity is indicated and also accompanied by loss of a trigeminal chamber. Nielsen (1966) commented on the biology of Acanthonus and examined stomach contents of six specimens, the major- ity of which contained crustaceans and dorsal filt of polycha- etes. Another specimen, however, contained bones of a conspecific. In the three specimens here examined (310, 290, 240 mm TL) the stomach contents contained a ‘mush’ in which only crustacean elements are discernible. Horn et al. (1978) reported on the large size of the cranial cavity of Acanthonus and the fact that it was filled with a low-density fluid. The authors hypothesised that since the head is the heaviest part of the body, the concentration of positively buoyant fluid provides lift and allows the fish to maintain a horizontal position with minimum swimming energy. Horn et al. suggest that the head might function as an acoustic dome. Certainly the cranial wall is exceedingly thin and the labyrinth is, compared with other ophidiforms (and gadiforms), extensive (Fig. 11). The fluid in which the brain is bathed is retained in the cranium by the membrane covering the large common optic- olfactory-trigeminal opening. Compared with other ophidiiforms Acanthonus appears to have a relatively poorly developed olfactory system, the olfactory tract being thin and the nasal rosette having only five or six lamina (cf. 8-18 in other taxa examined (Fig. 32A). 127 Only Tauredophidium and Typhlonus have reduced nasal rosettes with six laminae (Fig. 32B). In nearly all the speci- mens of Acanthonus available the nasal rosette is missing; since it is small and firmly attached to the connective tissue surrounding the nasal opening it appears to have become detached with the skin during collecting (the specimens are all badly damaged). In the two type specimens of A. armatus the rosettes are clearly visible in only one specimen (BMNH 1967.18.7:58). In the type specimen of Typhlonus nasus the nasal rosette is attached to and surrounded by fatty tissue. In Xyalecyba the nasal ‘rosettes’ appear to be in the form of three pear-shaped plates. Of the other taxa examined most have large nasal rosettes bearing ten-twelve laminae (Fig. 32C), Monomitopus and Genypterus both have eighteen and Spectrunculus has in excess of twenty (Fig. 32D), Dic- rolene and Glyptophidium have the lowest number, viz eight. Compared with most other ophidiiforms the eyes of Acan- thonus are small (9.1-9.7% of head length) and the recti muscles are thin (Fig. 27); the lens is large and surrounded by a ring of pigment. Of 23 genera in which the eye/head length proportion has been calculated, only three approach the ratio of Acanthonus: Genypterus, Dinematichthys and Tauredo- phidium. In the two former genera the ratio is equal to or slightly less than that of Acanthonus, respectively, 6.8-8.2 and 9.0-9.8% of head length. The relatively small ratio in Genypterus is a reflection of its disproportionately elongate skull due to the lengthened parasphenoid ascending process and pterosphenoid (Fig. 10A). Tauredophidium has an eye diameter/head length ratio of 5.0% (the eye is covered with skin). In the majority of genera the proportional range is 10.3% (Sirembo) — 24.6% (Lepophidium); mode, 20.9% of head length. This range includes Xyelacyba (12.5 & 13.9% in the two specimens measured). Scales are lacking in all six specimens of Acanthonus examined but according to Nielsen (1966), small, thin, cycloid scales occur on the head and body. Mats of sponge spicules were found associated with all the specimens, in the orbital cavities, gill chambers and mouths. The spicules apparently belong to a hexactinellid of the family Pherone- matidae which tend to carpet extensive areas of the substrate in parts of the north Atlantic; these sponges play host to many invertebrates (S. Stone, pers. comm.). It is perhaps not an artefact of capture that the fishes are covered with sponge spicules but that they actively feed within the sponge beds and even bite through the sponge basket, which may be one explanation for the derived palatine-lower jaw bite. Horn et al. (1978) surmise that Acanthonus conforms to Childress & Nygaard’s (1973) hypothesis that deep-sea fishes conserve energy and discourage predation. They point out that the enlarged head of the fish and a correlated large mouth allow a wide size-range of prey. It does not necessarily follow that an expanded mouth is a correlate of an enlarged cranium (cf. the large-headed, small-mouthed macrouroids, Echinomacrurus and Squalogadus) and in Acanthonus it is not simply the capacity of the mouth but its derived function (lower jaw-palatine bite) which presumably further widens its prey spectrum (see above). Relationships of Acanthonus and the classification of ophidiiforms Cohen & Nielsen (1978) placed Acanthonus in the neo- bythitine tribe Sirembini along with Sirembo, Hoplobrotula, 128 Tauredophidium, Dannevigia and Xyelacyba. Of these gen- era I have examined all except Dannevigia and find that Tauredophidium and Xyelacyba are closest to Acanthonus in overall morphology with a long, stout ventroposterior preo- percular spine, a preopercular-opercular ligament covering the articulation of the operculum, a narrow operculum, occlusion between the palatine and lower jaw teeth and a ‘platform’ articulation between the preoperculum and quad- rate and narrow operculum. Tauredophidium also possesses a short, single ethmoid spine (contrary to Cohen & Nielsen, 1978) but a spine is lacking in Xyelacyba. A superficial and radiographic examination of Xyelacyba suggests that although lacking an ethmoid spine it more closely resembles Acanthonus than does Tauredophidium in the morphology of its suspensorium, palatoquadrate and the first neural arch. Hoplobrotula also possesses an ethmoid spine (Machida, 1990), lacking in Sirembo, but both share an almost identical pelvic-hyoid morphology where the medial section of the sternohyoideus is well-separated from the lateral parts and the urohyal is reduced to a small triangular element. These are also specialized characters possessed by Acanthonus. Despite these apparent synapomorphies I would agree with Cohen & Nielsen’s (1978) doubt that their Sirembini is a natural assemblage, since Acanthonus lacks the synapomor- phic condition of the RLA-PP of Sirembo and Hoplobrotula and also possesses the derived external branching of the supraorbital trunk of the trigeminal nerve. The other neobythitine division, Neobythitini, was charac- terized by Cohen & Nielsen (1978) solely on the basis of pelvic fin position (at the level of the preoperculum) and it includes ca 30 genera of which 12 have been examined. None of the characters reviewed here appear to be synapomorphic for this sample of genera and it seems unlikely that the Neobythitini as presently construed is monophyletic. The characters described above and the limited number of taxa examined do not permit an in-depth evaluation of the existing classification of ophidiiforms (Cohen & Nielsen, 1978). They certainly uphold the division of Ophidiinae and Brotulinae. Modification of pelvic and pectoral girdle struc- ture, associated musculature and innervation, urohyal mor- phology and swimbladder-vertebral association characterize the former group (six of the eight included genera have been examined). The path of the hyoideus branch of the facial nerve through the hyomandibular characterizes the Brotuli- nae (Brotula only). The Brotulinae shares derived features with several other taxa, e.g. narrow optic-olfactory foramen and separate medial section of sternohyoideus muscle with Hoplobrotula, Sirembo and Dicrolene, and hyomandibular flange for hyoideus branch of VII with the former; frontals contacting the pterosphenoid with Sirembo and Dinematich- thys; and forward position of the forebrain with Monomito- -Y = Fig. 32 Nasal rosettes of A, Acanthonus armatus; B, Tauredophidium hextii; C, Ophidion rochei; D, Spectrunculus grandis. Viewed as in situ, posterior to the left. G.J. HOWES pus (the latter is most likely a convergent character since both taxa have short neurocrania). Cohen & Nielsen (1978) established a classificiation of ophidiiforms in which two orders, Ophidioidei and Bythitoi- dei were distinguished primarily on their respective oviparity and viviparity and secondarily on the position of the anterior nostril, either well or just above the upper lip. As pointed out by Patterson & Rosen (1989) the bythitoid characters are probably apomorphic, whereas the ophidioid ones are plesio- morphic. Within the Ophidioidei were included the Carap- idae and Ophidiidae. The former have been dealt with most recently by Markle & Olney (1990) who have discovered synapomorphies which reaffirm the family’s monophyly. These authors did not specifically address the problem of carapid relationships and merely point out that the Carapidae does not appear as ‘. . . an obvious sister group’ to any other ophidiiform lineage. The Carapidae appear to be the sister- group to other ophidiiforms on the basis of their possessing a smaller, shallower exoccipital but which, like other ophidii- forms excludes the supraoccipital from the foramen magnum. Like brotulines, some neobythitines and ophidiines, carapids possess enlarged anterior epipleural ribs, similarly modified swimbladders and a well-developed basioccipital cavity with a corresponding ventromedial process on the first centrum; paired ventro-lateral processes on the lateral ethmoid articu- lating with the palatine, all features which characterise Ophi- diiformes. The Ophidiidae are diagnosed by Cohen & Nielsen (1978) on two plesiomorphic features: a supramaxilla and presence of body scales and on the dorsal fin rays being equal to or longer than the anal rays. This latter feature was used principally to distinguish the Ophidiidae from the Carapidae but as equal length dorsal and anal fin rays occur amongst bythitoids it is not a diagnostic character. Of their four subfamilies the Ophidiinae and Brotulinae are clearly united on shared anatomical features of the pectoral girdle and hyoid musculature (Cohen & Nielsen, 1978; this paper). The other two subfamilies are Brotulotae- niinae, characterized by derived squamation and spinous gill-rakers (in other osteological characters it appears plesio- morphic) and Neobythitinae, diagnosed on plesiomorphic features (i.e., absence of ophidiine characters). The Bythitoidei comprises two families, Bythitidae and Aphyonidae, the latter characterized by absence of scales and swimbladder, reduced eyes and other generally neotenic features. The Bythitidae comprises two subfamilies, Bythiti- nae and Brosmophycinae recognised respectively by the continuity or separation of the caudal with the dorsal and anal fins. Part of the Brosmophycinae (recognised as a tribe, Dinematichthyini) seems well characterized by a synapomor- phic intromittent organ (in males) having ‘ossified’ elements. Of the two members of the Brosmophycini examined, Bros- mophyciops is most like the Dinematichtyinae in its anterior vertebral column and swimbladder morphology. Lucifuga does not, on the basis of its vertebral column-swimbladder anatomy, appear to be closely related to Brosmophyciops, nor for that matter, to the group which includes the deep- water taxa (Group 2 above). Of the characters discussed in this text, the Aphyonidae retain the plesiomorphic states. A swimbladder is absent, which must certainly represent sec- ondary loss since it is present in virtually all other ophidii- forms. However, the derived morphology of the opercular and hyoid bones supports Cohen & Nielsen’s (1978) recogni- tion of aphyonid monophyly. ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES In the three aphyonid genera examined, Aphyonus, Bar- athronus and Nybelinella, the interoperculum is reduced to a narrow strip of bone having nearly the same width as the ligament connecting it to the mandible (Fig. 33). The interopercular-mandibular ligament is elongate, twice or more the length of that in other ophidiiforms (cf. Fig. 33B). Posteriorly, the interoperculum is connected to the interhyal by connective tissue and a short, strong ligament. Distally, the interhyal is also syndesmotically connected to the tip of the posterohyal and the entire junction of the posterohyal- interhyal-interoperculum is wrapped in thin connective tis- sue; proximally the interhyal is attached to the posterior medial border of the preoperculum (Fig. 33A). The suboper- culum is also apparently reduced but its shape cannot be readily discerned in the non-cleared and stained specimens dissected. In all these features aphyonids differ from other ophidiiforms where the interoperculum is deep, triangular or boomerang-shaped; the interopercular-mandibular ligament short and the interhyal distally connected with the posterior half of the interoperculum and with the posterohyal, and proximally attached to the central region of the preoperculum (Fig. 33B). The majority of ophidiiform taxa are contained in the non-monophyletic Neobythitinae and Bythitidae, the former containing ca 50 genera, the latter ca 28 (Cohen & Nielsen, 1978). According to Patterson & Rosen (1989) these two assemblages could be distinguished on the basis of differences in their cranial-vertebral articulation, that of bythitids being derived. As discussed above (p. 111) this is a doubtful synapomorphy and is not supported by the synapomorphies identified here (hypertrophy of the RLA-PPD nerve (p. 122), branching of the supraorbital trigeminal trunk external to the trigeminal foramen (p. 104), and vertebral- swimbladder interconnections). These features also support a major dichotomy of ophidiiforms but different to that pro- posed by Patterson & Rosen (1989); Table 1. Those taxa (except for two, see below) with the derived form of RLA-PPD (Group A) have the plesiomorphic, internal branching of the superficial-trigeminal nerve trunks while those with the derived external branching pattern (Group B) possess the plesiomorphic pelvic nerve condition. Taxa of Group A comprise the Ophidiinae (sensu Cohen & Nielsen, 1978) and Brotulinae (Brotula). The ‘neobythitines’ Hoplobrotula, Sirembo, Neobythites and Dicrolene display an intermediate connection of nerve branching (p. 105) and Hoplobrotula, Sirembo and Dicrolene also share derived features which indicate that they form a monophyletic group Fig. 33. Interoperculum and its associations in A, Barathronus bicolor (dissected specimen, lateral view); B, Monomitopus metriostoma (cleared and stained specimens, right lateral view). 129 Table 1 Group membership of the genera examined. Group A. RLA-PP nerve hypertrophied, exposed laterally and ventrally; some taxa with modified cleithrum and urohyal; supraorbital trunk of trigeminal primitively dividing within the cranium; anteromedial mandibular pore openings small. Group B, RLA-PP normal; cleithrum and urohyal unmodified; supraorbital trunk of trigeminal dividing external to facialis foramen; anteromedial mandibular pore openings usually large. Group 1, swimbladder anteriorly situated, intimately associated with expanded first-third (epipleural) ribs the first of which occurs on the first centrum, swimbladder often with modified anterior cap and with muscle attachment; first neural arch short. Group 2, swimbladder posteriorly situated, not associated with anterior ribs which are unexpanded, the first of which occurs on the third centrum (pleural); swimbladder not modified but with muscle attachment; first neural arch reduced and autogenous. Abbreviations: X = ‘intermediate’ condition; O = features of these groups absent; ? = condition unknown. Under ‘Present classification’, A = Aphyonidae, B = Bythitoidei, N = Neobythitinae, O = Ophidiidae. Groups A B 1 z Present classification Abyssobrotula Acanthonus Barathrodemus Bassozetus Bathyonus + Brosmophyciops Brotula + Cataetyx Cherublemma Dicrolene Diplacanthopoma Genypterus + Glyptophidium Hoplobrotula Hypopleuron' Lamprogrammus Lepophidium oe Lucifuga Monomitopus Neobythites Xx Ogilbia Ophidion + Parophidium tf Petrotyx 2 Penopus Porogadus Pycnocraspedum Sirembo Spectrunculus Aphyonidae: Aphyonus O Barathronus O t+++4+4++ +++4++4+ + > + +++ 4” 4+ + ++ + +++ t++t++++ett+4+4++ ~ ++ Pe BZ 22 22 ZC Ctr ZZ Wie 2 2 2A Zio WZ 10 ZO Z 2.2 ZIZ +N CO +MY 4+4~ O O O O ' Hypopleuron has a normal RLA-PP and the supraorbital trunk divides primitively within the trigeminal chamber so that although the taxon belongs to Groups A and B it lacks the derived features of those groups. (Fig. 34). In turn they share a similar, apparently derived, mandibular sensory canal pore pattern with ophidiines (p. 126). The taxa belonging to Group B contains those other taxa presently assigned to the Neobythitinae (of the Ophidi- idae) and the Bythitidae. The neobythitine Hypopleuron and the bythitoid Aphy- onidae possess neither of these derived characters. Loss of epipleural ribs places the Aphyonidae with Group B taxa. 130 Carapidae 7 Piacd 4 5 6 giuh 8 G.J. HOWES Groups Neobythitinae (part, A) Brotulinae Neobyth. (part,B) |A Ophidiinae Neobyth.(pt, C) Bythitidae(pt) Aphyonidae Neobyth. Bythitidae (remainder) Je |e Fig. 34 Distribution of synapomorphies among investigated taxa and their correspondence with the groupings given in Table 1. The Carapidae are indicated as being the sister-group to other ophidiiforms on the basis of their sharing an enlarged exoccipital which excludes the supraoccipital from the foramen magnum but which still allows the dorsal part of the supraoccipital to form the posterior margin of the skull. They also share the ophidiiform features of expanded anterior ribs, modified swimbladder and associated muscles, basioccipital-vertebral articulation and well-developed paired lateral ethmoid facets (see text). Synapomorphies: 1) exoccipital dorsally expanded to exclude supraoccipital from forming the dorsoposterior cranial border; 2) RLA-PP nerve hypertrophied and exposed ventrally to hypaxial muscles; 3) olfactory-optic nerve foramen reduced to narrow aperture; separation of medial part of sternohyoideus muscle (occurs convergently in Acanthonus); enlarged mandibular sensory pores; 4) anteriorly extended cleithra; urohyal canopy-like with modified course of RLA-PP and VSR3 nerves; pharyngohyoideus muscle expanded; 5) supraorbital trunk of trigeminal nerve complex dividing externally to facialis chamber; 6) reduction or loss of anterior ribs; 7) opercular bones modified; 8) first neural arch reduced and autogenous; swimbladder situated posteriorly. Neobythitinae (part A) = Hypopleuron; (part B) = Hoplobrotula, Sirembo, Dicrolene: (part C) = Monomitopus, Lamprogrammus, Glyptophidium; Bythitidae (part) = Brosmophyciops. Groups 1 and 2 characterised on the basis of (divergently) derived swimbladder-vertebral connections correspond for the most part with the A and B groupings. Most taxa of Group 2 also belong to Group B and an overlap between Groups 1 and B occurs only amongst five genera, biz. Brosmophyciops, Glyptophidium, Lamprogrammus, Monomitopus and Ogilbia (Table 1). Membership of both Groups A and B is restricted to two genera, Bathyonus and Spectrunculus. Both possess the respective derived conditions of these groups, namely, hypertrophy and ventral exposure of RLA-PP nerve and external division of the supraorbital trunk of the trigeminal complex. Both are also members of Group 2 having the derived conditions of rib loss and posterior place- ment of the swimbladder. Since these genera lack any syn- apomorphies which characterise ophidiines, brotulines and the few ‘neobythitine’ genera included in Group A, it must be assumed that the RLA-PP nerve characters have been evolved independently. The division between ophidiods and bythitoids made by Cohen & Nielsen (1978) on the basis of ovo- and viviparity is transgressed by two genera, Brosmo- phyciops and Ogilbia both of which, by virtue of their anteriorly situated swimbladder and expanded ribs belong to Group 1. These and the aphyonids apart, the viviparous taxa are included within that assemblage possessing external branching of the supraorbital-trigeminal nerves, namely, Group B. The ‘neobythitine’ Hypopleuron stands phylogenetically apart from other taxa in lacking any of the synapomorphies recognised here. The majority of taxa referred to the Neo- bythitinae and Bythitidae belong to a derived assemblage whose sister-group is the Aphyonidae, characterised by mostly reductive features (see above). The distribution of synapomorphies and their correspondence with the groups recognised above as A, B, 1 and 2 are shown in Fig. 34. As an aside, it may be pointed out that the arrangement of caudal fin muscles in ophidiiforms correspond more closely to those of pediculate paracanthopterygians and acanthoptery- gians than to gadiforms. Howes (1991a) noted that gadoids have ‘reduced’ caudal muscles insofar as a superficial interra- diales, hypochordal longidorsalis, flexores dorsales and ven- trales are lacking. Ophidiiforms possess all these muscles, albeit that the superficial interradiales are reduced to a thin band of widely spaced fibres. Regrettably, this study has not conclusively resolved the phylogenetic relationships of Acanthonus but available char- acter data from Tauredophidium and Xyelacyba (p. 128) suggest that the three taxa form a monophyletic group. As yet the anatomy of Tauredophidium and Xyelacyba is too imper- fectly known to resolve the trichotomy. The shared features of hyoid muscle anatomy and an ethmoid ‘spine’ which indicate relationship of Acanthonus and Tauredophidium with Hoplobrotula and Sirembo are doubtful in the light of ANATOMY AND CLASSIFICATION OF OPHIDIIFORM FISHES the distribution of the trigeminal and pelvic nerve characters. No synapomorphy has emerged from this study which can further distinguish separate groups from within the larger body of taxa (in terms of the number investigated here) placed in Groups B and 2 (Fig. 34). ACKNOWLEDGEMENTS. I gratefully acknowledge the assistance and many helpful criticisms for improving the manuscript given by Dan Cohen, Nigel Merrett, Jorgen Nielsen and Colin Patterson. I also thank Drs R. Vari and Y. Scherbachev for the gifts and loans of material from their respective institutions (USNM and Shirov Insti- tute). Special thanks are due to my colleagues Patrick Campbell for preparing radiographs and Shirley Stone for identifying sponge spicules and to vacation student Neil Browne for recording meristic and other data. REFERENCES Carter, H.J. & Musick, J.A. 1985. Sexual dimorphism in the deep-sea fish Barathrodemus manatinus (Ophidiidae). Copeia 1985 (1): 69-73. —— & Sulak, K.J. 1984. A new species and a review of the deep-sea fish genus Porogadus (Ophidiidae) from the western North Atlantic. Bulletin of Marine Science 34 (3) 358-379. Childress, J.J. & Nygaard, M.N. 1973. The chemical composition of midwater fishes as a function of depth occurrence off southern California. Deep Sea Research 20: 1093-1109. Cohen, D.M. 1974. A review of the pelagic ophidioid fish genus Brotulataenia with descriptions of two new species. Zoological Journal, Linnean Society 55 (2): 119-149. —— 1982. The deep sea fish genus Enchelyobrotula (Ophidiidae): descriptions of new species, notes on distribution and osteology. Bulletin of Marine Science 32 (1): 99-111. —— & Nielsen, J.G. 1978. Guide to the identification of genera of the fish order Ophidiiformes with a tentative classification of the order. NOAA Technical Report. NMFS Circular 417: 1-72. — & Rohr. 1992. Description of a giant circumglobal Lamprogrammus species (Pisces: Ophidiidae). In press. Courtenay, W.R. 1971. Sexual dimorphism of the sound producing mechanism of the striped cusk-eel, Rissola marginata (Pisces: Ophidiidae). Copeia 1971 (2): 259-267. — & McKittrick, F.A. 1970. Sound producing mechanism in carapid fishes with notes on their phylogenetic significance. Marine Biology 7: 131-137. Freihofer, W.C. 1970. Some nerve patterns and their systematic significance in paracanthopterygian, salmoniform, gobioid and apogonid fishes. Proceed- ings California Academy of Sciences 38 (12): 215-264. Gosline, W.A. 1953. Hawaiian shallow-water fishes of the family Brotulidae, with the description of a new genus and notes on brotulid anatomy. Copeia 1953: 215-225. — 1960. Hawaiian larva-flow fishes. Part IV. Snyderidia canina Gilbert, with notes on the osteology of ophidioid families. Pacific Science 14 (4): 373-381. 131 — 1968. The suborders of perciform fishes. Proceedings of the United States National Museum 124 (3647): 1-78. Horn, M.H., Grimes, P.W., Phleger, C.F. & McClanahan, L. 1978. Buoyancy function of the enlarged fluid-filled cranium in the deep-sea ophidiid fish Acanthonus armatus. Marine Biology 46: 335-339. Howes, G.J. 1988. The cranial muscles and ligaments of macrouroid and gadoid fishes; functional, ecological and phylogenetic inferences. Bulletin British Museum (Natural History) Zoology, 54 (1): 1-62. —— 1989. Phylogenetic relationships of macrouroid and gadoid fishes based on cranial myology and arthrology. In D.M. Cohen (Ed.) Papers on the systematics of gadiform Fishes. Science series No. 32. Natural History Museum of Los Angeles County: 113-128. — 1990. The syncranial osteology of the southern eel-cod family Muraenole- pididae, with comments on its phylogenetic relationships and on the biogeog- raphy of sub-Antarctic gadoid fishes. Zoological Journal of the Linnean Society 110: 73-100. — 1991a. Anatomy, phylogeny and taxonomy of the gadoid fish genus Macruronus Gunther, 1873, with a revised hypothesis of gadoid phylogeny. Bulletin British Museum (Natural History) Zoology 57 (1): 77-110. — 1991b. Biogeography of gadoid fishes. Journal of Biogeography 18: 595-622. Machida, Y. 1988. Preliminary study of nerve pattern in Neobythites sivicola (Ophidiidae, Ophidiiformes). Memoires Faculty of Science, Kochi University 9: 49-56. — 1990. A new ophidiid species Hoplobrotula badia from Sagami Bay, Central Japan Japanese Journal of Ichthyology 37 (3): 209-214. Markle, D.F. 1989. Aspects of character homology and phylogeny of the Gadiformes. In D.M. Cohen (Ed.) Papers on the systematics of gadiform fishes Science Series No. 32, Natural History Museum of Los Angeles County: 59-88. — Williams, J.T. & Olney, J.E. 1983. Description of a new species of Echiodon (Teleostei: Carapidae) from Antarctic and adjacent seas. Proceed- ings Biological Society of Washington 96 (4): 654-657. — & Olney, J.E. 1990. Systematics of the pearlfishes (Pisces: Carapidae). Bulletin of Marine Science 47 (2): 269-410. Nielsen, J.G. 1966. On the genera Acanthonus and Typhlonus (Pisces, Brotul- idae). Galathea Report 8: 33-46. — 1969. Systematics and biology of the Aphyonidae (Pisces, Ophidioides). Galathea Report 10: 7-88. Patterson, C. & Rosen, D.E. 1989. The Paracanthopterygii revisited: order and disorder. In D.M. Cohen (Ed.) Papers on the systematics of gadiform fishes Science series No. 32. Natural History Museum of Los Angeles County: 5-36. Regan, C.T. 1903. On the systematic position and classification of the gadoid or anacanthine fishes. Annals and Magazine of Natural History (7) 9: 459-466. —— 1912. The classification of the blennioid fishes. Annals and Magazine of Natural History (8) 10: 265-280. —— 1929. Fishes. Encyclopaedia Britannica 14th ed. 9: 305-328. Rose, J.A. 1961. Anatomy and sexual dimorphism of the swimbladder and vertebral column in Ophidion holbrooki (Pisces: Ophidiidae). Bulletin of Marine Science of the Gulf and Caribbean 11 (2): 280-308. Rosen, D.E. 1985. An essay on euteleostean classification. American Museum Novitates 2827: 1-57. Stiassny, M.L.J. 1986. The limits and relationships of acanthomorph teleosts. Journal of Zoology, London (B) 1: 411-460. Svetovidoy, A.N. 1961. On European species of the family Ophidiidae and on the functional importance of the structure specificity of their fish-sound. Voprotsky Ikhtiology 17: 3-13. ud sates th a2 oe a ‘td int 4a at W afm we hal ’ - pK leaibinldh and | | . . ibm a ; waren Herve Me aBiner: ail tine owls leiarmony eo TS Ute | Fabdoen 3 : , ti it ; i haiy bare (aly Deo la wot anus (oa fazi . -Brotutiogs: bo =e lows beeen, i ted iy ent —— roe Puc af? te oe SC ) rs al Deal eis rind es ban b nen amen ot) Sabo ; a r wr 5 Vie a IB wmv ty peal ts a enna erent ory mt Y eomgiolth it lee ; — b oceeretidt nite dba evel ot Aaa Ad, Shan 2 nel Wie stig sey wl edabet rose OF bee hd ‘eaten eeepeaa re | aang en on SoD ote al it? dri ad fi riatue ‘i he oe wire 4 : ‘ iv v @x , 3) 9 ; ; ve RAVE wi ( ‘ital blac yn wy ih eal lo : , iva tate eed, Stn ‘pe H phe o = > 4 LA dre FT wi8 Hawai) avon) (sat sf dag Risin ny ta ; pf brat ~ ny vii kc bye PORNO 6 city «& i “ww eo TREK). Pe fag in al 4 cabald vhethe g feet = . t “a pe Bull. Br. Mus. nat. Hist. (Zool.) 58(2): 133-148 Issued 26 November 1992 Morphology and morphogenesis of the soil ciliate Bakuella edaphoni nov. spec. and revision of the genus Bakuella Agamaliev & Alekperov, 1976 (Ciliophora, Hypotrichida) WEIBO SONG College of Fisheries, Ocean University of Qingdao, Qingdao 266003, China NORBERT WILBERT Zoologisches Institut der Universitat Bonn, Poppelsdorfer Schloss, D-5300 Bonn, Germany HELMUT BERGER Institut fiir Zoologie der Universitat Salzburg, Hellbrunnerstrasse 34, A-5020 Salzburg, Austria CONTENTS Ri EOI TACDIN: — ete c= ee ote eI ise cc es wrostolte aes Pe tsnicie ais vcs aaa RMR cat wtmcniaiqs bd SVE oboe. cone bho cee Saorerae erate 133 Maaterialssmethods, and termmolocy <2u.fia.ceccdseteerues sos vcbeemrattensiaatts Mieke Mbeha rer wares wectnge ls Lemay awwmupmmdetess 133 Description of Bakuella edaphoni nov. spec. ..............sececeeeeeeceeeeneeeeeenes ban aa sean ae a atatetapraate otenve ce era atte ae 134 Revision(of the genus Bakuella Agamaliev & Alekperov, 1976. 2... ccsced. caicseeestnseneveoeatdellnesins seisasiasenaseeadiion 141 SY EONS CICS Pte SORE ts PO soe te Mieke cen len onaianiid ao eNee An ya iaee oeeatieas Cemee ent att meaNeS AED EN SEED 141 PD ESCHIGEINE OLISMECICS. BSR! wa. Aa Pose cae nite ts ohh atlas nnigg eSB nascent area deta om eee sees aah g atROL eR ee. naa MaRC REE 142 Unidentifiable taxa and species excluded from the genus Bakuella ............ 0.0. cccccccecececececececeeeeeensnenenenenens 144 PREKHOWISADEMICHIS HR a5: SST sie one Seciecon Mine oft sn andccnae ciseeeh eee nebeteic eat aes ane aa tae he sh os peta das, ska toes Reece 147 BTEC MCCS ts Mote a ee Acne eos tiie diaghsinnmenaracae ganiinlaminmt deen cemdde dada te + Sahas a £EE ese eit « Reece seta 147 Synopsis. The morphology and morphogenesis of the hypotrichous ciliate Bakuella edaphoni nov. spec. from a soil in Qingdao, China, are described by observation of living and protargol impregnated specimens. The morphogenesis of cell division does not significantly differ from that of B. salinarum. Bakuella Agamaliev & Alekperov, 1976 comprises 6 species: B. marina Agamaliev & Alekperov, 1976 (type species), B. crenata Agamaliev & Alekperov, 1976, B. edaphoni nov. spec., B. kreuzkampii nov. spec., B. salinarum Mihailowitsch & Wilbert, 1990, and B. walibonensis nov. spec. Bakuella imbricata Alekperov, 1982 and B. polycirrata Alekperov, 1988 are junior synonyms of B. marina and B. crenata, respectively. Bakuella agamalievi Borror & Wicklow, 1983, B. variabilis Borror & Wicklow, 1983, and B. pulchra (Buitkamp, 1977) Jankowski, 1979, do not belong to the genus Bakuella. The obliquely arranged ventral rows behind the midventral cirri and the 3 enlarged frontal cirri could place the genera Bakuella and Parabakuella Song & Wilbert, 1988 in the subfamily Bakuellinae Jankowski, 1979, within the family Holostichidae Fauré-Fremiet, 1961. INTRODUCTION Agamaliev & Alekperov (1976) established the genus Bakuella for two new hypotrichous ciliates with short, obliquely arranged cirral rows at the posterior end of the midventral row. Since then new representatives of this genus have been described by Alekperov (1982, 1988), Borror & Wicklow (1983), and Mihailowitsch & Wilbert (1990). A further new species is here described, the morphology and morphogenesis of which have been examined in detail. The genus Bakuella, which now contains 6 species from saline water, freshwater, and soil, is revised. MATERIALS, METHODS, AND TERMINOLOGY Bakuella edaphoni inhabits the upper soil layer (0-2 cm) on a hill in the city of Qingdao, China (36° 08’ N; 120° 43’ E). Excystation occurred about 3 days after the incubation of the air-dried soil was begun (for method see Buitkamp, 1977). Some individuals were taken from this initial culture and transferred to petri dishes containing Eau de Volvic as culture medium, with a few crushed wheat grains to promote bacte- rial growth. The infraciliature was revealed by the protargol impregna- tion method according to Wilbert (1975). All counts and 134 SONG, WILBERT & BERGER measurements were made at a magnification of x 1500. The Table 1 Morphometrical characterization of Bakuella agamalievi drawings of the protargol impregnated specimens were made (aga, uncertain species, from Agamaliev, 1972; wet silver with a camera lucida. For clarity, in the morphogenic stages impregnation), B. crenata (cre, from Agamaliev & Alekperov, : Be : 1976; wet silver impregnation), B. edaphoni (eda, original data; of B. edaphoni, the parental cirri are shown only by outline, protargol impregnation), B. kreuzkampii (kre, from whereas new ones are shaded. The terminology is according Mihailowitsch & Wilbert, 1990; protargol impregnation), B. to Kahl (1932), Borror (1972), Borror & Wicklow (1983), marina (mal, from Agamaliev & Alekperov, 1976; wet silver Corliss (1979), Foissner (1982), and Hemberger (1982). impregnation. ma2, original data of the population of Wilbert, Wiackowski (1985, 1988) designates both the zigzag arranged 1986; protargol impregnation. ma3, B. imbricata from Alekperov, 1982; wet silver impregnation. ma4, original data; protargol impregnation), B. salinarum (sal, from Mihailowitsch & Wilbert, 1990; protargol impregnation), and B. walibonensis (wal, from ventral cirri and the short rows posteriad of them as midven- tral cirri, which is correct in terms of morphogenesis. How- ever, the term ‘midventral row’ is here used only for the sum Mihailowitsch & Wilbert, 1990; protargol impregnation). All of the zigzag arranged ventral cirri (= midventral cirri). The measurements in um. ? = sample size unknown; if only 1 value is more or less obliquely arranged short rows are designated as known it is listed in column x, if 2 values are available they are ventral rows. For details of the morphometrical analysis see listed as Min and Max. Max = maximum value, Min = minimum Berger et al. (198 4). value; n = number of individuals examined; SD = standard deviation; Vr = coefficient of variation in %; X = arithmetic mean. DESCRIPTION OF BAKUELLA EDAPHONI Character Species x SD Vr Min Max n NOV. SPEC. yD Sen hee. = eae ee Body, length aga — - - 140 220 ? DIAGNOSIS. In vivo 190-300 x 50-85 um. More than 100 Gig! Ue hi - 120 150 ? macronuclear segments. 7 buccal cirri, 3 frontoterminal cirri, eda 219.6 41:5) . 18:9'- 1SSeezs3elG 8 transverse cirri, 9 pairs of midventral cirri, 7 ventral rows, kre - 155.8 -16:4 105°” 1305eio® 3 and 39 adoral membranelles on average. mal - = = 120 140 ? ma2 266.7 27.6 — 250) “310. 25 TYPE LOCALITY. Upper soil layer on a hill in the city of ma3— - - 110 130 ? Qingdao, China. ma4 101.4 57/ = 91 108 18 sal 307.4 26.6 8:6... 272 348 - 10 TYPE MATERIAL. The slide of holotype specimens and 1 slide wal 202.0 16.3 80 180 229 10 of paratype specimens are deposited in the collection of Budyewidth as 69.400-Rig bo Gatphensg R216 microscopic preparations of the College of Fisheries, Ocean ve AGGe br QOeRtGablademrs 6 University of Qingdao, China. 1 paratype slide (reference ma2 85.0 148 — Woe 40. 11 number 1990:11:19:1) is deposited in the Natural History ma4 616 63 - 45°\"70° 18 Museum, London. sal 99.1 18.6 18.7 87 145 10 wal 72:3. 85a, Ley 62-5 163. 7 DESCRIPTION (Figs 1-5, Table 1). Long elliptical, posteriorly distinctly narrowed. About 2:1 dorsoventrally flattened, flex- Pe gee ee Ces | ae rs ; ‘ < : membranelles, ible. Macronuclear segments ellipsoidal, rather regularly dis- length tributed in cytoplasm. 2-7 (x = 4.6, SD = 2.0, Vr = 43%, n= 16) micronuclei. Contractile vacuole at level of doral ; er 7 “= 2530 g cytostome (Fig. 1). Cortical granules absent. Cytoplasm ee oem en 3g 9 % P 7 3 a - 16 colourless, often densely packed with up to 5 um large, ine 33.7 23 #68 30 37.7 colourless globules (mitochondria?), readily visible after pro- mini eee re z. 39-36? targol impregnation (Fig. 3). Feeds on bacteria, diatoms, ma2 40.4 42 - BAT e vot 25 zooflagellates, testaceans, and small ciliates. Moderately ma3— = = a8 40) 7 ma4 32.4 2.4 - 28 36, 12 rapid movement. Buccal area large. 1-7, usually 2 or 3, cirri immediately ; : ; i 30.2 41 105 330 is posterior to right frontal cirrus (Fig. 4, arrow). Midventral es row terminates at about level of cytostome. Anterior ventral Frontoterminal = cre’_— 10.0 - 5 > tle 1 rows with 3-5 cirri, posterior rows with 7-14 cirri each. Both cirri, number oe oy; 0.8 24.3 ; 2 22 marginal rows terminate in median of cell without overlap- che Oi. é aie fF, ping (Figs 1, 4). ma 7.6 20 = re ery, st = ne COMPARISON WITH RELATED SPECIES. Bakuella edaphoni dif- a, ae wie 14 fers from the type species (Figs 21-35) in habitat and in that it wal ried wes 10 has more buccal cirri and fewer frontoterminal cirri wal > One= = 17 (Table 1). Bakuella salinarum has distinctly more pairs of midventral cirri and only 2 frontoterminal cirri (Fig. 41). In A ot sn na 74 “13 77 o B. kreuzkampii and B. walibonensis the ventral rows are few ice fsQ) hiss a 7 in number and rather short (Figs 51, 52). Furthermore, both mal —- - - ? species have only 2 frontoterminal cirri. Recently, P. Eigner ma2 30d ny Di aia 34 & W. Foissner (pers. comm. to H. Berger) found a rather ma3 - 7 a ? ma4 Ni MVS) = sal dele 10'S" 1083 wal Sy OA similar terrestrial species which has, however, conspicuous cortical granules (visible in vivo and after protargol impreg- nation!) arranged in longitudinal rows. At superficial in vivo ANNAN WRK NW NWN HNorunrnhfre oh NHONWO ~ BAKUELLA EDAPHONI NOV. SPEC. 135 observation B. edaphoni can be easily confused with large Holosticha species (Foissner, 1982). The soil species Parauro- Character Species xX SD Vr Min Max n styla pulchra Buitkamp, 1977, which was probably errone- = > aa. ? 4 “7 , : , The first event in morphogenesis is the formation of small an icy fo. a igs groups of basal bodies close to the posterior end of the middle a) LAAs 4 25" is ventral rows (Fig. 6) and near the left transverse cirrus (Fig. sal 26.0 7.9 304 22-38 7 7, arrow). The basal bodies increase in number forming a wal 13:82 O04 .3.0. 13614, 15 longish field from which some migrate forward to the poste- Meeeetstveitral cre! 3.0? — > — 1 rior portion of the midventral row (Fig. 8). The parental cneemmih Seri. eda 26 10 37.0 {ie ase endoral membrane and the anterior third of the paroral kre 140 2.6 “IRS 10° "See 7. membrane dedifferentiate (Fig. 8, small arrows). Simulta- re eral cre! aie 7" i. , neously the posterior buccal cirri are modified to the anlage II et > oda as 07 166 ae of the proter (Fig. 8, large arrow). The pharyngeal fibres fen Sicitri disappear. The macronuclear segments fuse to about 15 ellipsoidal segments with reorganization bands. Total number of cre’ 10.0 - A =. 2 I The adoral membranelles of the opisthe develop in a pentitows a Es a oe : rs = posteriad direction (Fig. 9, large arrow). At this stage some ee ee en ee or, 9 of the posterior cirri of the middle ventral rows are dediffer- ma2 i — A. £80 1G entiated and the posterior portion of the paroral membrane ma3- - - 5 a), becomes disorganized. Between the dedifferentiated buccal ma4 43> 054 = te) S15 row and the unaltered midventral row some small rows of sal 5.8 2.9 184 13° 21 10 basal bodies occur (Fig. 9, small arrows). The macronuclear wal em 12 2 Ries segments differ in size. Transverse cirri, aga 5.0 - - ae =? In the middle stages of morphogenesis most of the cirri of number cre! Te = =" 1 the ventral rows are modified to primordia of the opisthe. a Be i ae : © ” The parental buccal cirri are now a narrow band (Fig. 10, KG Lo ee eS arrow). Some midventral cirri are obviously incorporated in m2 FO Ine = 5 arr the formation of the obliquely arranged primordial streaks. ma3 7) — oa - ? In both the marginal rows and in all dorsal kineties 2 ma4 | ES i le 5 9 20 primordia occur (Figs 10, 11). The macronucleus is a large sal 9.5 Dp, Pek ses 1 1 10) single mass. wal 5.20 07 13.1 4 6 15 Somewhat later the proximal parental adoral mem- Right marginal row,aga — = = AS 30? branelles begin with reorganization. Possibly the posterior number of cirri cre — - ~ a0). 455 7 end of the anlage I of the proter is involved in this process eda 506 41 81 43 55 10 (Fig. 12, small arrow). Both in the proter and in the opisthe, kre 41.0 5.2 12.7 34 46 ue about 20 short anlagen and the new left frontal cirrus Lae Ka Aabe 7 6 S sy os 33 (Fig. 12, large arrows) are recognizable. et ns i oe In the next stage the oral primordium of the opisthe is ma4 AD waiSkG ex 34 54 °«O@9 already distinctly bent to the right and completely modified to sal 622, SS Ss 8666.10 adoral membranelles (Fig. 13). The reorganization of the wa §6960:5 63 5104 St 66 11 parental adoral zone of membranelles is restricted to about 10 Left marginal row, aga 38.0 — “ a os 9 proximal membranelles. In both the proter and the opisthe number of cirri cre 225 - as 38 40 ? the numerous obliquely arranged streaks begin with the eda ASA 37 -" 7:6, 44 “56 10 differentiation of cirri. The large arrows in Fig. 13 mark the kre 37:6 45 | 12:0", 30: 40 7 penultimate streaks. These streaks appear to form the fronto- mal - = a 52 56? terminal cirri (Figs 13, 15) although this is unusual because, oe gee Bea. 3 - = = in general, the frontoterminal cirri originate from the anterior a a ee 5 33 «9 part of the posteriormost anlage. The morphogenesis of the sal Sq soe 1000-45 60» 10 dorsal infraciliature shows no peculiarities. The macronu- wal 51S 5 88 a2. ASO cleus and the micronuclei begin to divide (Fig. 14). ee meties Aa 20? — " nee Figure 15 and 16 show a late morphogenetic stage. All new panier AL diet 3.0 0 0 3 3 24 fronto-ventral-transverse cirri are differentiated and they kre 3.0 0 0 3 Bre] begin to migrate to their final positions. Immediately before ma2 3.0 0 - SS iat 2 the separation of the proter and the opisthe, most of the ma3 S20 = he eal parental cirri and dorsal kineties are resorbed (Figs 17, 18). ma4 SOO. ra i fa" 2 The morphogenesis of the dorsal infraciliature shows _ 2 ; A : 3 _ unequivocally that no caudal cirri are present (Fig. 18). The postdividers have a wide elliptical outline 1 From Fig. 38. (Figs 19, 20). Now the pharyngeal fibres impregnate again. 2 See taxonomy of this species. The macronucleus commences with segmentation. In SONG, WILBERT & BERGER 136 “€-[ SONOUTY [eSIOP = €-] ‘MOI [eIJUSAPIUL 9Y} PUTYaq SMOI [eIUA = YA ‘HAI9 osIoasue = DI :MOI Jeulsiew 1y3U = YWY ‘oeueiquiow jerored = yd ‘oyjsido oy) Jo wnipsiownd yeio = g ‘MOl [eUSApIU = YIN ‘SnoponuoIOIU = TA] ‘yUOWIZas IvaTONUOIORU = RIA] {MOI [RUIZIeW IO] = YA] :o[nuessd jeurwse;dopus = O {HII jeuluIaj0}UOTy = LJ ‘Wild [eyUo’ = OJ -ourIquiowl [e1opue = We {111d yeoong = Og ‘Se][eueIqUuIOUI JO 9UOZ [e1Ope = WZV ‘Wr 082 ‘MIA [eIJUDA Ul 98e}s SHOUSSOYdIoW ATIVS AJOA Q “SMLINO [eIWOIJ YS Oy) pUTYaq IID oY) SyeW p “SLY UI MOLE OY “Wn OpZ “MOA [eSIOP PUR [eIUSA UT oANjeTO’IJU] § “p ‘({eUpuoYyoo}IU) so_nqoys ‘19}0UFeIp UI Wri ¢ 0} dn ‘poyoed I ‘(uoneusoidun osiejo1d ‘9-¢ ‘oala ur Ajasuap oy) Moys 0} sporfjed oy} Jo Weg ¢ “Wr CTZ ‘WIN CZZ ‘SMOTA [e1O}e] PUR [eNUSA Z ae Rc o~G B N- aon Thee ~ 6 SN SS 2 ee ¢ Z ‘[ ‘sjeulsiio ‘g-]) ‘sods ‘aou 2uoydypa vjjanyog = 9-| SB1q BAKUELLA EDAPHONI NOV. SPEC. DA AQ BWBBswewMBes vue i a a Reb! ee, ge, 275 um. The arrow marks the primordium protargol impregnation). 7 Very early sta , originals. Morphogenetic stages in ventral view after , 250 um. Small arrows, , formation of the adoral membranelles of the opisthe Figs 7-9 Bakuella edaphoni nov. spec. (7-9 137, ; large arrow, dedifferentiated buccal cirri. 9 Early -transverse streaks. aroral (only the anterior part) membrane ; small arrows, primordia of the fronto-ventral dedifferentiated endoral and p close to the transverse cirri. 8 Early stage stage, 250 um. Large arrow 138 _— - —~ ee SONG, WILBERT & BERGER entral view, 200 um. Large arrow, anlage II of the dle stage in ventral view, 270 um. Large arrows, new left frontal cirrus of , originals. Morphogenetic stages after protargol impregnation). 10 Middle stage in v proter. 11 Middle stage in dorsal view, 200 um. The arrows mark the primordia of the new dorsal kineties. 12 Mid Figs 10-12 Bakuella edaphoni nov. spec. (10-12 anlage I of the proter; primordium of the left marginal row; OP = the proter and the opisthe; small arrow, reorganizing proximal part of the parental adoral zone of membranelles. LMP = RMP primordium of the right marginal row. BAKUELLA EDAPHONI NOV. SPEC. arses oe -: tee, s < CS) cave seaqareenanesett® sw ih en, aad O aw D> vereannnness RC ae gh Bar AB LAN = -s Fm eee tts eee - my “Mv wass we «, QstNS sete —— AN oat 3 Bo aon we 139 new dorsal kineties 1-3. , 14 Middle stage in ventral and dorsal view, 220 um. Large arrows. , hew frontoterminal cirri. 1-3 protargol impregnation). 13 , 205 um. Large arrows , originals. Morphogenetic stages after 15, 16 Late stage in ventral and dorsal view Figs 13-16 Bakuella edaphoni nov. spec. (13-16 anlagen of the frontoterminal cirri. 140 SONG, WILBERT & BERGER l row. 1-3 = anlagen I, II, and III 17, 18 Very late stage in ventral and dorsal view, 220 ym. The arrows mark same anlage are connected by dotted lines; the origin of the , 120 um. The cirri which originate from the, for details see text). The arrow in Fig. 20 marks the posterior end of the left margina , originals. Morphogenetic stages after protargol impregnation). 20 Postdivider in ventral and dorsal view frontoterminal cirri could not be precisely determined (arrow with ?; and dorsal kineties 1-3, respectively. Figs 17-20 Bakuella edaphoni nov. spec. (17-20 the new frontoterminal cirri. 19, BAKUELLA EDAPHONI NOV. SPEC. Figure 19 those cirri which originate from the same streak are connected by dotted lines. DISCUSSION. The morphogenesis does not significantly differ from that of B. salinarum (Figs 41-50). Mihailowitsch & Wilbert (1990) describe a complete reorganization of the parental adoral zone of membranelles. This is, however, obviously a misinterpretation. Figs 46, 47 show that, just as in B. edaphoni and in Keronella gracilis Wiackowski, 1985, only the proximal membranelles are reorganized. Figure 45 does not show a morphogenetic stage as assumed by Mihailowitsch & Wilbert (1990), but a reorganizer. The streaks (= anlagen) of the fronto-ventral-transverse primordia in the proter have the following origin: streak I, from the parental undulating membranes; it forms the new undulating membranes and the left frontal cirrus. Streak II, from the parental buccal cirri; it forms the middle frontal cirrus and the buccal cirri. Streak III, probably de novo; it forms the right frontal cirrus and 1 or more cirri behind this cirrus. Streaks IV to n, probably de novo and by modification of parental midventral cirri, respectively (Figs 9, 10); they form the numerous cirral pairs of the midventral row, the ventral rows, and the transverse cirri. The origin of the frontoterminal cirri could not be unequivocally recognized (Fig. 19, arrow with ‘?’). It gives the impression that they originate from the last but one streak which is, however, not in accordance with any other morphogenetic pattern of related species (Fig. 13). In the opisthe streaks I-n are derived from the oral primordium and modified ventral rows (Figs 8-10). The morphogenesis of the dorsal infraciliature proceeds as in B. salinarum and e.g. in Kahliella franzi (Foissner, 1982) Berger & Foissner, 1988 or Gonostomum strenua (Engel- mann, 1862) Sterki, 1878 (see Song, 1990), viz. according to type 1 of Foissner & Adam (1983). REVISION OF THE GENUS BAKUELLA Agamaliev & Alekperov, 1976 Bakuella Agamaliev & Alekperov, 1976 1976 Bakuella Agamaliev & Alekperov, Zool. Zh., 55: 128 — Type species (original designation): Bakuella marina Agamaliev & Alekperov, 1976. 1977 Bakuella Agam. & Alek. — Corliss, Trans. Am. microsc. Soc., 96: 137 (see nomenclature and taxonomy). 1979 Bakuella Agamaliev & Alekperov, 1976 — Corliss, Ciliated protozoa, p. 309. 1979 Bakuella Agamaliev & Alekperov, 1976 — Tuffrau, Trans. Am. microsc. Soc., 98: 526. 1979 Bakuella Agamaliev & Alekperov, 1976 — Jankowski, Trudy zool. Inst., Leningr., 86: 50. 1979 Bakuella (Bakuella) Jankowski, Trudy zool. Inst., Leningr., 86: 50 — Type species (original designation): Bakuella marina Alekperov & Agamaliev, 1976. 1979 Bakuella (Loxocineta) Jankowski, Trudy zool. Inst., Leningr., 86: 51, 57 — Type species (original designation): Bakuella crenata Alekperov & Agamaliev, 1976. 1983 Bakuella Agamaliev & Alekperov, 1976 — Borror & Wicklow, Acta Protozool., 22: 113. 141 1985 Bakuella — Small & Lynn, Phylum Ciliophora, p. 450. 1987 Bakuella Agamaliev & Alekperov, 1976 — Tuffrau, Annls Sci. nat., 8: 115. 1989 Bakuella Agamaliev et Alekperov 1976 — Alekperov, Ecology of marine and freshwater protozoans, p. 7. 1990 Bakuella Agamaliey & Alekperov, 1976 — Mihailowitsch & Wilbert, Arch. Protistenk., 138: 208. DIAGNOSIS. Medium sized to large Holostichidae; obliquely arranged ventral rows situated behind midventral row; 3 slightly to distinctly enlarged frontal cirri; 2 or more fronto- terminal cirri; transverse cirri present; 1 left and 1 right marginal row; caudal cirri absent. ADDITIONAL CHARACTERS. Most of the species are described only from silver impregnated material. Thus the important character of the presence or absence of cortical granulation is unknown in these species. Outline usually long elliptical, both ends rounded. Body flexible. Adoral zone of mem- branelles 30-40% of body length. Undulating membranes bent and crossing. Bases of transverse cirri not conspicuously enlarged. 3 dorsal kineties (possibly a generic character; a redescription of B. crenata following protargol impregnation will probably reveal that it also has 3 dorsal kineties; Table 1). Dorsal cilia in vivo 3-5 um long. Edaphic, limnic, and marine species. REMARKS. Corliss (1977, p. 111, 137) erroneously desig- nated Bakuella as a nomen nudum. The suggestion of Jan- kowski (1979) that the number of macronuclear segments should be used for the splitting into the 2 subgenera Bakuella and Loxocineta does not appear to be justified. Three other genera are known which have obliquely arranged ventral rows behind the midventral row, viz. (i) Parabakuella Song & Wilbert, 1988 (3 frontal cirri; transverse cirri absent; 1 left and 1 right marginal row; caudal cirri present), (ii) Keronella Wiackowski, 1985 (many frontal cirri, i.e. the anterior part of the midventral row, form a ‘bico- rona’; transverse cirri present; 1 left and 1 right marginal row; caudal cirri present; see also Wirnsberger, 1987), and (iii) Metabakuella Alekperov, 1989 (many frontal cirri form a ‘bicorna’; transverse cirri present; 1 [or more?] right and 2 or more left marginal rows; caudal cirri absent?). Two of these genera, viz. Bakuella and Parabakuella, can be united in the subfamily Bakuellinae Jankowski, 1979 for which we propose the following improved DIAGNOSIS: Holostichidae with more or less obliquely arranged ventral rows behind the midventral row; 3 more or less distinctly enlarged frontal cirri. TYPE GENUS: Bakuella Agamaliev & Alekperov, 1976. Nomenclatural remarks: Jankowski (1979, p. 74) estab- lished the family Bakuellidae. According to Article 50c (i) of the ICZN (1985) a change in rank of a taxon within the family group does not affect the authorship of the name of the taxon. Thus Jankowski is the author of the subfamily Bakuel- linae and not Alekperov (1988), who erroneously wrote ‘subfamily Bakuellinae (Jank.) comb. n.’). KEY TO SPECIES 1 Species from terrestrial habitats — Species from limnic or marine habitats 142 2 2macronuclear segments ................. B. crenata (Figs 36-40) — Many (usually >100) macronuclear segments (e.g. Fig. 22) . 3 3. Posteriormost ventral row terminates roughly in the middle or at the end of the 2nd third of the cell; ventral rows with only about 3-4 cirri (RigsiS1o52)uhous. stewalean: teves. nosatet dea... 4 — Posteriormost ventral row terminates at about the level of the transverse cirri; ventral rows with up to 13 cirri (e.g. BS 21 AL) okt comes neat toaew as na tickwsmmestecinen cmcaoasessateeee 5 4 1 buccal cirrus; an additional row between the anterior end of the midventral row and the right marginal row RS Se ES. SRT B. kreuzkampii (Fig. 52) — 5-6 buccal cirri; no such row .......... B. walibonensis (Fig. 51) 3 2 frontoterminal cirri; 22-38 pairs of cirri in the midventral row. att. !..lemstcn:. hates: B. salinarum (Figs 41-50) — 5-11 frontoterminal cirri; 4-12 pairs of cirri in the midventral TOE sense ncs gh dea cae caeastopeane eee ter ath a B. marina (Figs 21-35) DESCRIPTION OF SPECIES REMARKS. Characters which are mentioned in (i) the diagno- sis, (ii) the section ‘additional characters’ of the genus, (iii) the key, or (iv) the tables, are not repeated in the descriptions below. Thus, some of them are rather short. Bakuella crenata Agamaliev & Alekperov, 1976 1976 Bakuella crenata Agamaliev & Alekperov, Zool. Zh., 55: 130 (Fig. 38). 1979 Bakuella (Loxocineta) crenata — Jankowski, Trudy zool. Inst., Leningr., 86: 51. 1982 Bakuella crenata — Alekperov, Zool. Zh., 61: 1253. 1983 Bakuella crenata Agamaliev & Alekperov, 1976 — Borror & Wicklow, Acta Protozool., 22: 113. 1988 Bakuella polycirrata Alekperov, Zool. Zh., 67: 778 (nov. syn.; Figs 36, 37, 39, 40). TAXONOMY. Bakuella polycirrata is almost certainly only a postdivider of B. crenata (Figs 36, 37). This is indicated by the enormous relative size of the adoral zone of mem- branelles (more than 50%), the unusual position of the porus of the contractile vacuole, and the immature ventral infracil- iature (cp. Figs 19 and 36). The only significant difference is the number of dorsal kineties; only 2 in B. crenata and 5 in B. polycirrata (Fig. 37). This difference may be due to the wet silver impregnation method employed, which yields inaccu- rate data for the dorsal infraciliature (e.g., Stylonychia mytilus sensu Agamaliev, 1978). Another possibility is that in Fig. 37 parental dorsal kineties are still preserved. A detailed redescription of this taxon is necessary. MORPHOLOGY (Figs 36-40, Table 1). In vivo (?) up to 210 um long. Each macronuclear segment with a micronu- cleus. Marginal rows distinctly separated posteriorly. OCCURRENCE. Locus classicus is the Djeiranbatansky fresh- water reservoir in Azerbaijan. Bakuella polycirrata was found in the same (?) freshwater habitat. Not found since. SONG, WILBERT & BERGER Bakuella marina Agamaliev & Alekperov, 1976 1976 Bakuella marina Agamaliev & Alekperov, Zool. Zh., 55: 129 (Figs 21, 22). 1979 Bakuella (Bakuella) marina — Jankowski, Trudy zool. Inst., Leningr., 86: 50. 1982 Bakuella marina — Alekperov, Zool. Zh., 61: 1253 (Figs 26, 27). 1982 Bakuella imbricata Alekperov, Zool. Zh., 61: 1253 (nov. syn.; Figs 31, 32). 1983 Bakuella marina A. et A., 1976 — Borror & Wicklow, Acta Protozool., 22: 113. 1983 Bakuella marina Agamaliev et Alekperov, 1976 — Agamaliev, Ciliates of Caspian Sea, p. 105. 1986 Bakuella marina Agamaliev — Wilbert, Symposia Biologica Hungarica, 33: 251 (Figs 28-30). NOMENCLATURE AND TAXONOMY. ‘Bakyella marina’ in Aga- maliyev (1976, p. 91) is an incorrect subsequent spelling. Bakuella marina is probably the senior synonym of B. imbri- cata because the important biometric data are rather similar (cp. Figs 21, 26, 31, Table 1). Bakuella marina sensu Wilbert (1986) is distinctly larger. Other characters, however, agree well, so that the identification can be accepted for the time being. MORPHOLOGY (Figs 21-35, Table 1). In vivo (?) about 200 um long. Contractile vacuole porus at level of cytostome (Figs 21, 23, 26). Posterior ventral rows with up to 13 cirri (Figs 21, 24, 26, 31). Marginal rows usually distinctly sepa- rated posteriorly. Sometimes with 4 dorsal kineties (Fig. 30). OCCURRENCE AND ECOLOGY. Locus classicus is the Caspian Sea (see also Agamaliyev, 1976). Wilbert (1986) found it in saline lakes in Saskatchewan, Canada. According to Hammer (1986) B. marina occurs at salinities of 2-37%. Chaouite et al. (1990) recorded it in mineral and hot springs in France. Locus classicus of the synonymous B. imbricata is the Djei- ranbatansky freshwater reservoir, Azerbaijan. Bakuella salinarum Mihailowitsch & Wilbert, 1990 1990 Bakuella salinarum Mihailowitsch & Wilbert, Arch. Protis- tenk., 138: 208 (holotype slide deposited in Institut fir Land- wirtschaftliche Zoologie und Bienenkunde, University of Bonn, Germany). MORPHOLOGY (Figs 41-50, Table 1). In vivo 280-350 um long. Contractile vacuole roughly in middle of cell. 2 micron- uclei. Cells brownish (no information is given why). 2 cirri situated behind right frontal cirrus. Anteriormost buccal cirrus slightly enlarged. Midventral row terminates slightly behind middle of cell. Ventral rows rather densely arranged. Marginal rows slightly overlapping. Very probably during morphogenesis the parental adoral zone of membranelles is reorganized only proximally and not completely as supposed by Mihailowitsch & Wilbert (1990; see discussion of B. edaphoni). OCCURRENCE AND ECOLOGY. Locus classicus is a salt-loaded ditch in Bad Waldliesborn, Lippstadt, Germany (for details see Mihailowitsch, 1989). Feeds on bacteria. Mihailowitsch & Wilbert (1990) give the following autecological data (n = 28): 5.3-16.3 °C, pH 7.0-7.7, 19-187 mg I"! CO, (free), 2.2-9.6 mg I! O,, 0.13-7.3 mg I! NH,*-N, 0.02-0.4 mg I* BAKUELLA EDAPHONI NOV. SPEC. PB. "e Alo ” Ve vo 143 Figs 21-32 Bakuella marina (21, 22, from Agamaliev & Alekperov, 1976; 23, from Wilbert, 1986 slightly modified; 24, 25, originals of a population from a saline lake in Saskatchevan, Canada; 26, 27, 31, 32, from Alekperov, 1982; 28-30, from Wilbert, 1986. 21, 22, 26, 31, wet silver impregnation; 23-25, 28-30, protargol impregnation; 27, 32, Feulgen staining). 21-32 Infraciliature in ventral and dorsal view and nucleus apparatus, 21, 22 = 120 wm, 23-25 = 230-310 um, 26, 27 = ? um, 28 = 210 um, 29 = 240 um, 30 = 212 um, 31, 32 = 110-130 um. NO, -N, 0.63-10.6 mg I’ NO;-N, 300-12763 mg I? CI, 216-3590 mS m" spec. conductivity. Bakuella walibonensis nov. spec. 1990 Bakuella spec. 1 — Mihailowitsch & Wilbert, Arch. Protistenk.., 1384213. TAXONOMY. Redescription necessary, especially in vivo observations. Differs from the rather similar B. kreuzkampii (Fig. 52) in that it has more buccal cirri and no additional ‘frontal’ row between the anterior end of the midventral row and the right marginal row. DIAGNOsIs. After protargol impregnation about 180-230 x 60-80 um. More than 100 macronuclear seg- ments. 6 buccal cirri, 5 transverse cirri, 14 pairs of midventral cirri, and 39 adoral membranelles on average. 2—5 ventral rows and consistently 2 frontoterminal cirri. TYPE LOCALITY. Salt-loaded ditch in Bad Waldliesborn, Lippstadt, Germany. MORPHOLOGY (Fig. 51, Table 1). Contractile vacuole situated slightly behind the middle of the cell. 3 cirri left of anterior end of midventral row. Ventral rows terminate at about level we SONG, WILBERT & BERGER Figs 33-35 Bakuella marina (originals of a population from a saline lake in Saskatchevan, Canada). 33 Ventral view in vivo. 34, 35 Infraciliature in ventral and dorsal view, protargol impregnation, 91-108 um. Figs 36-40 Bakuella crenata (36, 37, 39, 40, from Alekperov, 1988; 38, from Agamaliev & Alekperov, 1976. 36—40, wet silver impregnation; nuclear apparatus after Feulgen staining). 36, 37 Infraciliature of a postdivider in ventral and dorsal view, 66 wm. 38 Infraciliature in ventral view, 195 um. 39 Late morphogenetic stage in ventral view. 40 Reorganizer in ventral view. of contractile vacuole and marginal rows almost exactly in median of cell. Bakuella kreuzkampii nov. spec. 1990 Bakuella spec. 2 — Mihailowitsch & Wilbert, Arch. Protistenk., 138: 213. TAXONOMY. Redescription necessary, especially in vivo observations. The rather similar B. walibonensis has more buccal cirri and no cirral row between the anterior end of the midventral row and the right marginal row (Fig. 51). DIAGNOsIs. After protargol impregnation about 135-175 x 40-50 um. More than 100 macronuclear seg- ments. 5 transverse cirri, 14 pairs of midventral cirri, and 34 adoral membranelles on average. 3-5 ventral rows and con- sistently 1 buccal cirrus and 2 frontoterminal cirri. TYPE LOCALITY. Salt-loaded ditch in Bad Waldliesborn, Lippstadt, Germany. MORPHOLOGY (Fig. 52, Table 1). A short row with about 6 cirri between anterior end of right marginal row and midven- tral row. Ventral rows terminate at about level of contractile vacuole and marginal rows almost exactly in median of cell. UNIDENTIFIABLE TAXA AND SPECIES EXCLUDED FROM THE GENUS BAKUELLA Bakuella agamalievi Borror & Wicklow, 1983 1972 Holosticha manca Kahl, 1932 — Agamaliev, Acta Protozool., 10: 21 (misidentification; Figs 54, 55). 1983 Bakuella agamalievi Borror & Wicklow, Acta Protozool., 22: 114, 117, 120. REMARKS (Figs 54, 55, Table 1). This species was estab- lished for the rather superficially described Holosticha manca sensu Agamaliev (1972). Almost certainly Keronopsis rubra sensu Agamaliev (1974, p. 72, Fig. 10) represents the same taxon. Whilst in both descriptions buccal and frontoterminal Figs 41-47 Bakuella salinarum (from Mihailowitsch & Wilbert, 1990. 41-47, protargol impregnation). 41, 42 Infraciliature of a non-dividing specimen in ventral and dorsal view, 225 um. 43, 44 Very early and early morphogenetic stage in ventral view. 45 Reorganizer. 46, 47 Late stages in ventral view. 145 BAKUELLA EDAPHONI NOV. SPEC. 22828 se950 0000000099599 ws. Risteee ve & LAA tt ee eT = 7 Paes FR: LY Say 5° « : no PERSE dite Mt a ET, 69 we Whe ee eee eee eee” 0999 = ae eae =a ey & ae were ss eee sents S8 ty Sn SASS y ae \ ee we? Bele, Bed oe eee, oe eg - o,f, oe, sete" eee cers ite versie | doa. ° weve ese lerrees OND papa ued evra, rT] Li ys Cog teen gsr pee ested Se aL 2] ee Sher a o ee ad "Se0050 ooooF So goo09 9" 146 “a it wo ate, e SONG, WILBERT & BERGER oe COPIS “Sa 3 PSPASD PP PRSSSSR STREET ERENE? P Pe Re re errery sncrerePossss ° ° e ? e eur err pert mrmee rece ee RE , f ‘ é < ¢ 4 é é ‘ C4 6 & i z . 2 ¢ t 4 PA bd ri % ra be & 4 7 t f VA a“ Figs 48-50 Bakuella salinarum (from Mihailowitsch & Wilbert, 1990. 48-50, morphogenetic stages after protargol impregnation). 48, 49 Late and very late stage in ventral view, 303 um, 340 um. 50 Late stage in dorsal view, 303 um. cirri are absent, it is assumed that they were overlooked, possibly due to the inappropriate wet silver impregnation method used. Because no ventral rows are recognizable, it must be excluded from Bakuella. Possibly it belongs to the genus Holosticha. Length about 2.5 times width. Both mar- ginal rows terminate at level of transverse cirri. Number of dorsal kineties unknown. Locus classicus is the Caspian Sea. Not found since. Paraurostyla pulchra Buitkamp, 1977 1977 Paraurostyla pulchra Buitkamp, Decheniana, 130: 119 (Fig. 53). 1979 Bakuella pulchra — Jankowski, Trudy zool. Inst., Leningr., 86: 83. REMARKS (Fig. 53). ‘Parayrostyla pulchra’ in Jankowski (1979, p. 83) is a misspelling. Its inclusion in the genus Bakuella is not justified, because it has no midventral row. However, there is no doubt that some other characters, such as the absence of caudal cirri, the 3 dorsal kineties, the large number of macronuclear segments, the short cirral row near the anterior end of the right marginal row which simulates the frontoterminal row, and more than 1 buccal cirrus, remind one of members of the genus Bakuella. Thus only a detailed redescription, including an investigation of the morphogene- sis, will definitively elucidate the correct systematic position of Paraurostyla pulchra. Roughly spindle-shaped. Contractile vacuole slightly above middle of cell. 2 ventral cirri in front of transverse cirri, which are circa 18 um long and only slightly enlarged. Right marginal row begins at about level of poste- rior buccal cirri and terminates at level of transverse cirri; left marginal row terminates in median of cell. Feeds on diatoms and ciliates. Locus classicus of P. pulchra is the upper soil layer (0O-S cm) of a pasture near Bonn, Germany. Not found since. Buitkamp (1979) counted 79 individuals g™' dry soil 6 days after incubation at 30°C. Biomass of 10° individuals about 300 mg (Foissner, 1987). Bakuella variabilis Borror & Wicklow, 1983 1979 ?Bakuella sp. — Borror, J. Protozool., 26: 547 (Fig. 56). 1983 Bakuella variabilis Borror & Wicklow, Acta Protozool., 22: 111 (Fig. 57). 1989 Metabakuella variabilis (B. & W.) — Alekperov, Ecology of marine and freshwater protozoans, p. 7. REMARKS. Bakuella variabilis has 2-5 left marginal rows. Furthermore, the ventral rows are rather long and are not obliquely arranged behind the midventral row as in all other species of the genus (Fig. 57). The arrangment of the ventral rows reminds strongly on Urostyla grandis Ehrenberg, 1830. Urostyla, however, has many frontal cirri which form a bicorona (see Foissner et al., 1991, Ganner, 1991). A classifi- cation in the genus Metabakuella is also uncertain, because the type species, M. perbella (Alekperov & Musayev, 1988) Alekperov, 1989, has a bicorona too. The correct generic classification of B. variabilis can only be determined after a detailed description of the morphogenesis. Possible it needs a genus of its own. In vivo 225-240 um long. Flexible and opaque. More than 100 small macronuclear segments. Con- tractile vacuole at about the level of the cytostome. Cortical BAKUELLA EDAPHONI NOV. SPEC. 147 eo 7 ¢ 2 ‘ : = e é é a e! ¢ £ z ¢ rs ¢ & & t € ¢ ¢ é & % é o¢ e $ ¢ % ¢ t ae f é f # ¢ ¢ é ‘ é t Soni ig A é - t ¢ t f 4 f 2 *s é t e gf é $ f 3 ‘ > # ? éé © ¢ *, é ite *, # 44 ba SS a ee ee ', een YS eS % % 4%> mo &e Sim eng w Stans a & Fig. 51 Bakuella walibonensis nov. spec. (from Mihailowitsch & Wilbert, 1990). Infraciliature in ventral view, protargol impregnation, 165 um. The arrows mark the frontoterminal cirri. Fig. 52 Bakuella kreuzkampii nov. spec. (from Mihailowitsch & Wilbert, 1990). Infraciliature in ventral view, protargol impregnation, 165 um. The arrows mark the frontoterminal cirri. Fig. 53 Paraurostyla pulchra (from Buitkamp, 1977). Infraciliature in ventral view, protargol impregnation, 158 um. Figs 54,55 Bakuella agamalievi (from Agamaliev, 1972). 54 Infraciliature in ventral view, wet silver impregnation, 95 um. 55 Nucleus apparatus, Feulgen staining. granules on entire surface, 2-4 granules near each cirrus, on dorsal surface oblique rows, 3-10 granules per row. Feeds on flagellates. Adoral zone of membranelles about 1/3 of body length. 3 frontal cirri, 8-10 buccal cirri, about 12 transverse cirri. 1 right marginal row. Frontoterminal cirri present. Locus classicus of B. variabilis is a temporary pool in a flooded agricultural field in Lee, New Hampshire, USA (70°58’ W. Long., 43°8’ N. Lat.). Bakuella sp. in Wiackowski (1988) is not figured. Thus an identification is impossible. 4 f “a | ACKNOWLEDGEMENTS. We thank Prof Dr W. Foissner, University of Salzburg, for some critical comments on the manuscript. The first Liar ig vm VU irae eeans Svat aneveqannuedt VANESSA BUHFEV HV OSGer ag Be > g oe Z SS, Sone RRS oa 2 = author wishes to acknowledge the financial support of the ‘Shandong Sfaiue : = Foundation for Natural Sciences’. (oi ae oe = is A Se 5 e Re = A a Gaus ps eee sse 2 Poa eo yl tee oe 28 = eva’ Ses 5 tet => DGaucsss foto C6 oe ea Ss Meee Bans 4 ge ese DOR SS 2 fen 2 : ren of, ase Figs 56,57 Bakuella variabilis (56, from Borror, 1979; 57, from ar 7 we Borror & Wicklow, 1983). 56, 57 Infraciliature in ventral view, protargol impregnation and nigrosin-butanol staining, 270 um, 175 um. Right from Fig. 56 a middle morphogenetic stage. 148 REFERENCES Agamaliev, F.G. 1972. [Ciliates from microbenthos of the islands of Apseron- skij and Bakinskij archipelagos of the Caspian Sea]. Acta Protozoologica 10: 1-27 (in Russian with English summary). 1974. [Ciliates of the solid surface overgrowth of the Caspian Sea]. Acta Protozoologica 13: 53-83, Plates I-IV (in Russian with English summary). 1978. Morphology of some free-living ciliates of the Caspian Sea. Acta Protozoologica 17: 419-444, Plates I-IV. 1983. [Ciliates of Caspian Sea. Taxonomy, ecology, zoogeography]. Nauka, Leningrad Branch. 232 pp & Plates I-XX (in Russian). —— & Alekperoy, I. K. 1976. [A new genus Bakuella (Hypotrichida) from the Caspian Sea and the Djeiranbatansky water reservoir]. Zoologicheskii zhurnal 55: 128-131 (in Russian with English summary). Agamaliey, F.G. 1976. [Planctonic and periphytonic ciliophora in the eastern part of the Middle Caspian]. Izvestiya Akademii nauk Azerbaidzhanskoi SSR. (Biologicueskie) Year 1976: 90-94 (in Russian). Alekperov, I. H. 1982. [Bakuella imbricata sp. n. (Ciliophora, Hypotrichida) from the Djeiranbatansky water reservoir]. Zoologicheskii zhurnal 61: 1253-1255 (in Russian with English summary). —— 1988. [Two new species of infusoria (Ciliophora, Hypotrichida) from fresh waters of Azerbaijan]. Zoologicheskii zhurnal 67: 777-780 (in Russian with English summary). —— 1989. [Revision of the genera Bakuella Agamaliev et Alekperov 1976 and Keronella Wiackowski 1985 (Hypotrichida, Ciliophora)]. Jn: Poljansky, G.I., Zhukov, B.F. & Raikov, I.B. (Eds). [Ecology of marine and fresh- water protozoans. Proceedings of the II Symposium]. Academy of Sciences of the USSR. The All-Union Society of Protozoologists, Yaroslavl, p. 7, Abstract (in Russian). —— & Musayev, M. A. 1988. [New and rare free-living infusoria from fresh waters and soil of the Apsheron Peninsula]. Zoologicheskii zhurnal 67: 1904-1909 (in Russian with English summary). Berger, H. & Foissner, W. 1988. The morphogenesis of Kahliella franzi (Foissner, 1982) nov. comb. and Oxytricha gigantea Horvath, 1933 (Cilio- phora, Hypotrichida). Archiv ftir Protistenkunde 136: 65-77. ——,, Foissner, W. & Adam, H. 1984. Taxonomie, Biometrie und Morphogen- ese einiger terricoler Ciliaten (Protozoa: Ciliophora). Zoologische Jahr- biicher (Abteilung Systematik, Okologie und Geographie der Tiere) 111: 339-367. Borror, A. C. 1972. Revision of the order Hypotrichida (Ciliophora, Proto- zoa). The Journal of Protozoology 19: 1-23. — 1979. Redefinition of the Urostylidae (Ciliophora, Hypotrichida) on the basis of morphogenetic characters. The Journal of Protozoology 26: 544-550. —— & Wicklow, B.J. 1983. The suborder Urostylina Jankowski (Ciliophora, Hypotrichida): morphology, systematics and identification of species. Acta Protozoologica 22: 97-126, Plate I. Buitkamp, U. 1977. Uber die Ciliatenfauna zweier mitteleuropdischer Boden- standorte (Protozoa; Ciliata). Decheniana 130: 114-126. — 1979. Vergleichende Untersuchungen zur Temperaturadaptation von Bodenciliaten aus klimatisch verschiedenen Regionen. Pedobiologia 19: 221-236. Chaouite, J., Groliere, C.A. & Pepin, D. 1990. Etude du peuplement infusorien des eaux minérales et thermominérales en Auvergne. The Journal of Protozoology (Supplement) 37: 31A, Abstract 149. Corliss, J.O. 1977. Annotated assignment of families and genera to the orders and classes currently comprising the Corlissian scheme of higher classifica- tion for the phylum Ciliophora. Transactions of the American Microscopical Society 96: 104-140. —— 1979. The Ciliated protozoa. Characterization, Classification and Guide to the Literature. Pergamon Press, Oxford, New York, Toronto, Sydney, Paris, Frankfurt. 455 pp. Ehrenberg, C.G. 1830. Beitrage zur Kenntnif der Organisation der Infusorien und ihrer geographischen Verbreitung, besonders in Sibirien. Physikalische Abhandlungen der Koniglichen Akademie der Wissenschaften zu Berlin Jahr 1830: 1-88, Tafeln I-VIII. Engelmann, T.W. 1862. Zur Naturgeschichte der Infusionsthiere. Zeitschrift fiir wissenschaftliche Zoologie 11: 347-393, Tafeln XX VIII-XXXI. SONG, WILBERT & BERGER Fauré-Fremiet, E. 1961. Remarques sur la morphologie comparée et la systématique des ciliata Hypotrichida. Compte rendu hebdomadaire séances de l’Académie des sciences, Paris 252: 3515-3519. Foissner, W. 1982. Okologie und Taxonomie der Hypotrichida (Protozoa: Ciliophora) einiger 6sterreichischer Béden. Archiv fiir Protistenkunde 126: 19-143. — 1987. Soil protozoa: fundamental problems, ecological significance, adap- tations in ciliates and testaceans, bioindicators, and guide to the literature. Progress in Protistology 2: 69-212. & Adam, H. 1983. Morphologie und Morphogenese des Bodenciliaten Oxytricha granulifera sp. n. (Ciliophora, Oxytrichidae). Zoologica Scripta 12: 1-11. —— Blatterer, H., Berger, H. & Kohmann, F. 1991. Taxonomische und Okologische Revision der Ciliaten des Saprobiensystems. I: Cyrtophorida, Oligotrichida, Hypotrichia, Colpodea. Informationsberichte des Bayerischen Landesamtes fiir Wasserwirtschaft, Miinchen 1/91: 478 pp. Ganner, B. 1991. Zur Taxonomie einiger Ciliaten (Protozoa) aus Fliessge- wassern. Dissertation Universitat Salzburg, Salzburg. 159 pp. Hammer, U.T. 1986. Saline lake ecosystems of the world. Monographiae biologicae 59: I-XI, 1-616. Hemberger, H. 1982. Revision der Ordnung Hypotrichida Stein (Ciliophora, Protozoa) an Hand von Protargolpraparaten und Morphogenesedarstellun- gen. Dissertation Universitat Bonn, Bonn. 296 pp. ICZN (The international Commission on Zoological Nomenclature) 1985. Inter- national code of zoological nomenclature. Third edition adopted by the XX General Assembly of the International Union of Biological Science. Univer- sity of California Press, Berkeley, Los Angeles. 338 pp. Jankowski, A. W. 1979. [Revision of the order Hypotrichida Stein, 1859. Generic catalogue, phylogeny, taxonomy]. Trudy Zoologicheskogo instituta. Akademiya nauk SSSR. Leningrad 86: 48-85 (in Russian with English summary). Kahl, A. 1932. Urtiere oder Protozoa I: Wimpertiere oder Ciliata (Infusoria) 3. Spirotricha. Die Tierwelt Deutschlands und der angrenzenden Meeresteile 25: 399-650. Mihailowitsch, B. 1989. Taxonomische und 6kologische Untersuchungen an Ciliaten (Protozoa, Ciliophora) in solebelasteten Fliessgewdssern. Disserta- tion Universitat Bonn, Bonn. 291 pp. —— & Wilbert, N. 1990. Bakuella salinarum nov. spec. und Pseudokeronopsis ignea nov. spec. (Ciliata, Hypotrichida) aus einem solebelasteten FlieBgewasser des 6stlichen Miinsterlandes, BRD. Archiv fiir Protistenkunde 138: 207-219. Small, E.B. & Lynn, D.H. 1985. Phylum Ciliophora Doflein, 1901. In: Lee, J.J., Hutner, S.H. & Bovee, E.C. (Eds). An illustrated guide to the protozoa. Society of Protozoologists, Lawrence. pp 393-575. Song, W. 1990. A comparative analysis of the morphology and morphogenesis of Gonostomum strenua (Engelmann, 1862) (Ciliophora, Hypotrichida) and related species. The Journal of Protozoology 37: 249-257. —— & Wilbert, N. 1988. Parabakuella typica nov. gen., nov. spec. (Ciliata, Hypotrichida) aus dem Edaphon eines Standortes in Qingdao, China. Archiv fiir Protistenkunde 135: 319-325. Sterki, V. 1878. Beitrage zur Morphologie der Oxytrichinen. Zeitschrift fiir wissenschaftliche Zoologie 31: 29-58, Tafel IV. Tuffrau, M. 1979. Une nouvelle famille d’hypotriches, Kahliellidae n. fam., et ses consequences dans la repartition des Stichotrichina. Transactions of the American Microscopical Society 98: 521-528. — 1987. Proposition d’une classification nouvelle de ’ordre Hypotrichida (Protozoa, Ciliophora), fondée sur quelques données récentes. Annales des Sciences Naturelles, Zoologie, Paris, 13 Série 8: 111-117. Wiackowski, K. 1985. The morphology and morphogenesis of Keronella gracilis n. gen., n. spec. (Hypotrichida, Ciliophora). Protistologica 21: 81-91. 1988. Phenetic and cladistic numerical estimates of phylogenetic relation- ships in Urostylina (Ciliophora: Hypotrichida). Acta Protozoologica 27: 1-20. Wilbert, N. 1975. Eine verbesserte Technik der Protargolimpragnation ftir Ciliaten. Mikrokosmos 64: 171-179. — 1986. Ciliates from saline lakes in Saskatchewan. Symposia Biologica Hungarica 33: 249-256. Wirnsberger, E. 1987. Division and reorganization in the genus Pseudoker- onopsis and relationships between urostylids and oxytrichids (Ciliophora, Hypotrichida). Archiv fiir Protistenkunde 134: 149-160. Bull. Br. Mus. nat. Hist. (Zool.) 58(2): 149-156 Issued 26 November 1992 A new genus and species of freshwater crab from Cameroon, West Africa (Crustacea, Brachyura, Potamoidea, Potamonautidae) NEIL CUMBERLIDGE Department of Biology, Northern Michigan University, Marquette, Michigan 49855, USA. PAUL F. CLARK Department of Zoology, The Natural History Museum, Cromwell Road, London, SW7 5BD, UK. CONTENTS RANT OGM CANOMINS cee emits ecole occa nc citar rire Mere a tics ete cs eendte wsle Seren vice eM eata ac MME cis va OR Dee ec a5 Sea eaiele APSE Mev ac auetuceins 149 SS Siena MC oS CC OUI rere eee nena Neer mtn tem ee eee ne aiae i ste en asetan Gnahiemetariaavtabamnnteses qs babi suas Sagecrank acces 149 POMOC USIOCM TE Menno naas seeeeet tes (aerate se ete tan Seid dds vn Sen tkly ccn sc SMR RG ia ne ab ob bE veces vadtyeafoR awe Meee aeRS 149 PP SILER aE OMUOU ESSE aera cet eee onthe eee rene eae nee ok scale close Sk aiscis ol etchlcle na Shiva nln cine Ounuaaurcisnatevaedebeta Suacendlishalen 150 DIS GERD UNAOM ee etre Biase cama teal ena ana scales acinine wales elves Selo walk oasis dae AORTA ania sav aii ewan ncahe eee renee mee ee 150 Maton ameter cedicrassnaraassdateesocataaeectace tamer es tat tes cag ce ssisiecljenngerasae eo ree crecis Wick che uactcnonsomeaey velvet stamens 153 PCSCIAO COM ep note cree eer te eaten de ene ORE nis one canis seo rsiclein ality ainn vie leliecd cele Sampreplan voteueesenesanngenarounnes 154 COLO PIC AIINOMES tern cee etn is sence eaten ae ae te te oe Ae asi ceisbss hie-casaieis «aieuR Models sidan dun staamevanes toc cemsigees ese guaeemaee 155 PAP OMOMIC MCMIAGKS fo cec. poten rack n das anctreapactcvoan acleaen cetee da steuble dao ounces tad hc sad Motel Basne soeattes «edd Ae beceses ead. 155 IES ei ha cr etc ate cc ate pin ie reine Se since sche setaags a encd lt mea aieetb ss nCibale hath da'e Sosa clabiaatnaigaeites 155 ENC KGa OG OPE MAC ENES Greeters Gr een. mmCE nes air de to einraaiomeietntia a an ast age dee tieprapape han cAqueniteiches ap celpwedhveg «atharesinoe eta 155 MERCH MNCS Merrett crete ta lic sire cen cotapne n technic sins wie sieln naire eae AES cine rot ™ Ost 0.337 "033 aap oem? A 16.5 Kies) 7.3~ “O°70== 0531 0.31 MNHU—14141 Yaounde, Paratype Fe Or 94.6 17.8 8.6 833 Yer O72", 0:35) (n 034 MNHU—20162 Bamenda, Paratype Sher GelS5:Sy B49 11.6 LONE COROmret0:33; 0:30 MNHU—20183 Bamenda, Paratype Die S29. Dae 210 19 Ort O ieee O39 0:38 10;%.Ou8 208 inl 7.0 7.0 0.74 0.34 0.34 MNHU—20201, Paratype LO ig 3D-4ed, 24-5 12.0 11 Zan, + 0269.".. 0534... 0232 MNHU—20203 Douala, Paratypes 1 Fl © ins |< | 724 fe | 12.4 2. 0.71 033" 70132 sm oO” 220" 102 10.1 0.71 03357035 1 Ses Sa] m2 7, US ee O72 a iL 0.30 i 30.5, 21.9 10.5 10/0) 2072 034 5. 033 16:00 Gun 28.4007 W204 95 O71 baxOW2h BOS3i/, 0:32 Table 2 Potamonemus mambilorum sp. n. length of the legs of the type from Cameroon (CW = 34.5 mm). P1—P5 = pereiopods 1-5; 1 = length of segment; Total Leg Length = sum of length of all five segments; Proportion Total/CW = ratio of total leg length to carapace width. Leg No. Pi P2 P3 P4 Segment Ishium 4.0 4.0 4.0 4.0 Merus 12.0 12.0 16.0 16.0 Carpus 17.0 7.0 8.0 8.0 Propodus 37.0 8.0 7.0 8.0 Dactylus 26.0 10.0 10.0 11.0 Total Leg Length (mm) 96.0 41.0 45.0 47.0 Proportion Tota/CW 2.78 LA9. 1.30 1.36 end segment. End segment of gonopod 2 very short (one tenth the length of the penultimate segment); end segment of gonopod 1 curving outwards away from the medial line when viewed from the lateral aspect; end segment of gonopod 1 with a longitudinal groove on posterior aspect. Second sternal CUMBERLIDGE AND CLARK groove incomplete, reduced to two short notches at the lateral edges of the sternum at the bases of the chelipeds. HOLOTYPE. NHM reg. 1991:183, from Somié Village, Tikow Plain, Cameroon (6°30’N, 11°30’E), 760 metres. Collected by David Zeitlin. Potemonemus mambilorum sp. n. (Figs 1-3, Tabs 1-2, pl. 1a, b) Potamonautes anchetiae ; Balss 1929:117 (part). Potamonautes (Isopotamonautes) anchetiae; Bott 1955:247 (part). DIAGNOsIs. Major (right) cheliped of male enlarged, longer and higher than left (minor) cheliped, dactylus of cheliped arched forming permanent gape between the fingers when the chelipeds are closed; end segment of gonopod 1 with curving longitudinal groove on posterior surface; exorbital and epibranchial teeth small and low; vertical flank groove meeting anterolateral margin at the base of the epibranchial tooth; carapace smooth, no deep grooves. DISTRIBUTION. The forested highlands of southwest Cameroon (Bamenda, Bambulae, Fsou Grasslands), and the forested lowlands of south Cameroon (Douala, Yaounde). Fig. 1 Potamonemus mambilorum sp. n., from Tikow Plain, Cameroon, male, type, carapace width = 34.5 mm. a, cephalothorax, dorsal aspect; b, cephalothorax, frontal aspect; c, right cheliped, lateral aspect; d, left cheliped, lateral aspect. Scale bars equal 10.0 mm. 151 NEW FRESHWATER CRAB FROM CAMEROON Fig. 2 Potamonemus mambilorum sp. n., from Tikow Plain, Cameroon, male, type, carapace width = 34.5 mm. a, abdomen; b, right third maxilliped showing details of the exopod; c, right mandible showing details of palp. Scale bars equal 10.0 mm (a), 5.0 mm (b, c). 152 CUMBERLIDGE AND CLARK “ed = } Fig. 3 Potamonemus mambilorum sp. n., from Tikow Plain, Cameroon, male, type, carapace width = 34.5 mm. a, right gonopod 1, ventral aspect; b, right gonopod 2, ventral aspect; c, right gonopod 2, dorsal aspect; d, right gonopod 1, dorsal aspect showing groove. Scale bars equal 5.0 mm (a,d), 1.0 mm (b,c). NEW FRESHWATER CRAB FROM CAMEROON 153 TT Plate 1.a, Potamonemus mambilorum sp. n. Dorsal view, scale bar in mm; b, Potamonemus mambilorum sp. n. Frontal view, scale bar in mm. Photos by Phil Crabb, NHM Photo Unit. MATERIAL HOLOTYPE. NHM reg. 1991:183. Paratypes: Fifteen addi- tional specimens of P. mambilorum are held in the MNHU. All re-identified by one of the authors (NC). This material was formerly identified as Potamonautes anchetiae Brito- Capello 1871 by Dr H. Balss. MNHU-13498, from Bambulae near Bamenda, Cameroon, Collected by Oberl. Ader, 2/IV/ 1909, 1m; MNHU-13593, from Fsou Grassland 1100m, Cameroon, Collected by Oberl. Bartsch, 2m, 2f£; MNHU- 154 Carapace Dimensions (mm) Carapace Lenath (mm) Carapace Thickness (mm) Front Width (mm) Carapace Width (mm) Fig. 4 Comparisons of 16 specimens of Potamonemus mambilorum sp. n. ranging in size from CW = 20.30 mm to 37.8 mm, all from Cameroon. Dimensions of the carapace (CL, CT, FW) compared to body size (CW). Relationships are described by the following regression equations: CL = 0.77 + 0.619 CW, r = 0.99; CT = 0.97 + 0.31 CW, r = 0.933; FW = 1.06 + 0.29 CW, r = 0.975. All r values indicate a highly significant correlation (P < 0.001), at 14 degrees of freedom. CW = carapace width at the widest point; CL = carapace length, measured along the median line; CT = cephalothorax thickness, the maximum depth of the cephalothorax; FW = front width, the width of the front measured along the anterior margin; r - correlation coefficient. 14141, from Yaounde station, south Camercon, Collected by Oberl. v. Somerfeld, 1f; MNHU-20162, from Bamenda, Cameroon. Collected by Adametz and Naumann, 12/XI/ 1909, 1m; MNHU-20183, from Bamenda, Cameroon. Col- lected by Lt Naumann, 15/1/1912, 2f; MNHU-20201, from Cameroon. Collected by Weibel, 1m; MNHU-20203, from Douala, Cameroon. Collected by Thorbecke, 29/X/1912, 2m, 3f. DESCRIPTION The following description is based on the male holotype (NHM reg. 1991:183, CW = 34.5 mm). CARAPACE (Figs 1a,b; Tab. 1; pl. 1a,b). The cephalothorax is ovoid, widest at the anterior third (ratio of CL to CW = 0.36) and distinctly arched, with the maximum depth in the anterior region (ratio of CT to CW = 0.74). The anterior margin of the front is straight and wide, one-third the width of the carapace (ratio of FW to CW = 0.33). The carapace texture is smooth, no granules are visible, even under magnification, and there are no deep grooves; the cervical and semi-circular grooves are faint. The exorbital tooth is blunt and low and the epibranchial tooth is almost undetectable. There is no detectable intermediate tooth on the anterolateral magin between the exorbital and epibran- chial teeth. The anterolateral margin is smooth (lacking teeth, spines, or granulations) behind the epibranchial tooth, CUMBERLIDGE AND CLARK ” ca ° 2 -_ ° Qa ° _ oO O cucw A cT/cw KS O Fwicw oO Qa oO _ oO (S) Carapace Width (mm) Fig. 5 Comparisons of 16 specimens of Potamonemus mambilorum sp. n. ranging in size from CW = 20.30 mm to 37.8 mm, all from Cameroon. Relative proportions of the carapace (CL/CW, CT/CW, and FW/CW) compared to body size (CW). The relationships are described by the following regression equations: CL/CW = 0.745 — 0.001 CW, r = 0.313; CT/CW = 0.356 - 0.001 CW, r = 0.148; FW/CW = 0.356 — 0.001 CW, r = 0.456. The r values for CL/CW, CT/CW, and FW/CW indicate no significant correlation (P > 0.05), at 14 degrees of freedom. CW = carapace width at the widest point; CL = carapace length, measured along the median line; CT = cephalothorax thickness, the maximum depth of the cephalothorax; FW = front width, the width of the front measured along the anterior margin; r = correlation coefficient. and this margin is smoothly continuous with the posterior margin. The anterolateral margins do not curve inward over the surface of the carapace in the branchial regions. The postfrontal crest extends laterally across the entire carapace, meeting both anterolateral margins at the epibranchial teeth; there is a short groove at the mid point of the postfrontal crest. Each flank has a long longitudinal groove dividing the subhepatic region from the pterygostomial region, and a shorter vertical groove in the subhepatic region beginning at the longitudinal groove and ending at the anterolateral mar- gin at the base of the epibranchial tooth. These two grooves divide the flanks into three parts. The first sternal groove is complete, and the second sternal groove is reduced to two small notches at the sides of the sternum. The third maxilli- peds fill the entire oral field, except for the efferent openings, which are oval. There is no flagellum on the exopod of the third maxilliped (Fig. 2b). The mandibular palp is two- segmented with a single, undivided, end segment (Fig. 2c). The first five segments of the male abdomen are broad but short and taper inward; the last two segments are long and narrow (A6, A7), the last segment (A7) is rounded at the distal margin (Fig. 2a). A small to medium-sized species, mature at CW = 29.0 mm. Gonopops (Figs 3a,d). The end segment of gonopod 1 is long, (half as long as penultimate segment), lacking a longitu- dinal groove on the anterior surface, but with a twisting NEW FRESHWATER CRAB FROM CAMEROON longitudinal groove visible on the posterior surface, running from the junction with the penultimate segment to the tip of the end segment. The end segment of gonopod 1 curves outwards away from the medial line when viewed from the lateral aspect, is widest at the half way point, and ends in a pointed tip. The distal region of the penultimate segment of gonopod 1 is slim, tapering to the junction with the end segment. Gonopod 2 is shorter than gonopod 1 (reaching only the junction between the segments of gonopod 1). The end segment of gonopod 2 is extremely short, only one-tenth as long as the penultimate segment. The end segment of gonopod 2 is not solid, its sides are thin and folded inward enclosing an inward-facing hollow; the tip is rounded. The penultimate segment of gonopod 2 is widest at its base, tapering sharply inward about one-third along its length, the last two-thirds form a long, thin, tapering, upright process which supports the end segment. CHELIPEDS (Figs 1c,d; pl. 1b). The chelipeds of the male are markedly unequal, the right is much longer (37.0 mm) and higher (14.5 mm) than the left (26.5 mm and 8.5 mm respec- tively). The length of the ventral margin of the propodus of the right cheliped alone (37.0 mm) is greater than the width of the carapace at its widest point (34.5 mm). The palm of the right dactylus is swollen, the fingers gape widely when closed; the ends of the dactylus and pollex end in sharp, overlapping points (pollex over dactylus). The left cheliped is small, and lacks the arched dactylus and swollen palmar region; its dactylus and pollex touch along the entire length when closed except for a small gap in the proximal region. The pereiopods are slender and the proportions of the various segments are given in Table 2; P4 is the longest leg, P5 the shortest leg; the dactylus of PS is very short. GROWTH. The dimensions of the carapace (CL, CT and FW) of P. mambilorum increase with increasing body size (CW) in the manner described in Fig. 4. The carapace thickness was found to be almost equal to the front width over the range of sizes from CW = 20.3 mm to CW = 38.1 mm. These dimen- sions are expressed as a ratio of the carapace width in Fig. 5, which indicates that the relative proportions of the carapace (CL/CW, CT/CW and FW/CW) do not alter with increasing body size. The r values of 0.313, 0.148 and 0.456 respectively showed no significant correlation (P > 0.05, d.f. = 14) over a range of sizes, indicating that larger crabs have proportionally the same sized carapaces as smaller specimens (Fig. 5). ECOLOGICAL NOTES The collector of the type specimen provided the following comments on the crab. Potamonemus mambilorum is a riverine crab and is eaten by the Mambila tribe especially during the dry season, when crabs are caught by bailing out drying sections of river beds. The Mambila people call this species the ‘Kap’ crab. A second species of freshwater crab is sympatric with P. mambilorum and was identified as Sudanonautes faradjensis (Rathbun, 1921). The local name for this species is the ‘Nyar’ crab. 155 TAXONOMIC REMARKS Not all of the material assigned by Balss (1929) to Potamo- nautes anchetiae Brito-Capello 1871 was examined because the MNHU only has 15 specimens of the original Cameroon collection. It is these that are redescribed here as Potamone- mus mambilorum. Bott (1955) has referred the MNHU material identified by Balss to Potamonautes (Isopotamonau- tes) anchetiae together with additional specimens collected from Zaire. However the specimens from Zaire and Cameroon differ. For example, the specimens from Zaire possess two deep sternal grooves, the anterolateral margin of the carapace curves inward over the surface of the carapace in the branchial region, and they have a thread-shaped end segment on gonopod two. In contrast, in the specimens from Cameroon only the first sternal groove is deep and clear, the second is absent except for two side notches, the anterolateral margin is continuous with the posterolateral margin, and the end segment of gonopod 2 is very short indeed. Potamonemus mambilorum has a two-segmented mandibu- lar palp, witn a single (simple) end segment, which places it in the superfamily Potamoidea and the family Potamonautidae, rather than in the Gecarcinucoidea. This new species could not be assigned to any genera of the Potamonautidae (Pota- monautes, Sudanonautes and Liberonautes) for the following reasons. Potamonautes and Liberonautes are characterised by a long thread-like end segment of gonopod 2. The short end segment of gonopod 2 of P. mambilorum prevents the inclusion of this species in Potamonautes and Liberonautes (Bott, 1955; Cumberlidge, 1985; Cumberlidge and Sachs, 1989a,b). The very short end segment of gonopod 2 is a characteristic of Sudanonautes. However, the Cameroon material cannot be assigned to Sudanonautes since all speci- mens lack a flagellum on the exopod of the third maxilliped, a character considered here and elsewhere to be of taxonomic significance at the genus level and above (Bott, 1959, 1969, 1970; Cumberlidge, 1987). In addition, P. mambilorum lacks an intermediate tooth, the presence of which is another characteristic of Sudanonautes. These features justify the recognition of a new genus. ETYMOLOGY. The new genus has been named Potamonemus to recognise the distribution of the species in a tropical rain forest habitat. The species is restricted to a forested highland area of southwest Cameroon and a forested lowland area in south Cameroon. These areas are included within the bound- aries of one of Africa’s Pleistocene forest refuges (Kingdon, 1989). Potamonemus (Neuter) from the Greek; Potamo a contraction of the family name and nemos meaning forest. The species has been named for the Mambila people of Cameroon who know Potamonemus mambilorum well, and use this species as a seasonal food source. ACKNOWLEDGEMENTS. We should like to thank Mr David Zeitlin of Cambridge University, Cambridge, U.K. for bringing the material to our attention. We should also like to acknowledge the contributions of Prof H.-E. Gruner of the Zoologisches Museum, Museum fur Naturkunde der Humboldt-Universitat zu Berlin, Germany, Dr M. Tirkay, of the Senckenberg Museum, Frankfurt, Germany and Phil Crabb (NHM Photo Unit) for the photographs reproduced in this paper. Part of this work (NC) was supported by a Faculty Grant from Northern Michigan University, Marquette, Michigan, USA. 156 REFERENCES Balss, H. 1929. Potamonidae au Cameroon. In: Contribution a l'étude de la faune du Cameroun. Faune Colonies Frangaises 3: 115-129. Bott, R. 1955. Die Siisswasserkrabben von Afrika (Crust., Decap.) und ihre Stammesgeschichte. Annales du Musée Royal du Congo Belge Tervuren, C-Zoologie Série III, III 1(3): 213-349. — 1959. Potamoniden aus West-Afrika. Bulletin de l'Institut Francaise Afrique Noire Series A 21 (3): 994-1008. — 1969. Die Flusskrabben aus Asien und ihre Klassificaten (Crustacea, Decapoda). Senckenbergiana Biologie 50: 359-366. — 1970. Die Siisswasserkrabben von Europa, Asien, Australien und ihre Stammesgeschichte. Abhandlungen der Senckenbergischen Naturfor- CUMBERLIDGE AND CLARK schenden. Gesellschaft Deutsch 526: 1-338. Cumberlidge, N. 1985. Redescription of Liberonautes chaperi (A. Milne- Edwards, 1886) n. comb., a freshwater crab from Ivory Coast (Brachyura, Potamonautidae). Canadian Journal of Zoology 63: 2704-2707. —— 1987. Notes on the taxonomy of West African gecarcinucids of the genus Globonautes (Brachyura, Decapoda). Canadian Journal of Zoology 65: 2210-2215. Cumberlidge, N. & Sachs, R. 1989a. A key to the crabs of Liberian freshwaters. Zeitschrift fiir Angewandte Zoologie 76: 221-229. — 1989b. Three new subspecies of the West African freshwater crab Liberonautes latidactylus (DeMan, 1903) from Liberia, with notes on their ecology. Zeitschrift fiir Angewandte Zoologie 76: 425-439. Kingdon, J. 1989. Island Africa. 287 p. Princeton University Press, Princeton, New Jersey. Bull. Br. Mus. nat. Hist. (Zool.) 58(2): 157-170 Issued 26 November 1992 On the discovery of the male of Mormonilla Giesbrecht, 1891 (Copepoda: Mormonilloida) R.HUYS' Laboratoria voor Morfologie en Systematiek, Rijksuniversiteit Gent, K.L.Ledeganckstraat 35, B-9000 Gent, Belgium, & Centre for Estuarine and Coastal Research, Vierstraat 28, EA 4401 Yerseke, The Netherlands G.A.BOXSHALL Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD, England R.BOTTGER-SCHNACK? Institut fiir Hydrobiologie und Fischereiwissenschaft, Zeiseweg 9, D-2000 Hamburg 50, Germany” CONTENTS UTERO UCUOME emer eee secs Aen cae nie oe sioesia sete Oa ic nO PRE SE RE eres NSO Te cleans else Poors Nias viens cial nate Saw Anau cee aeons 157 a MITTELS Yaa LJ ELI OG Ss se es adaatiaeocanenee ton Hende cases etddoo Sop setanee steeeecanedo Gor oot meat Erce OL tt att aer BEESeenntiae co monneoneonon Sy WAAR WVU OGIPDORPRLS EDs FRIAS TOE linea ease «is! Vesna ofa Mae ere cioia snare abiaten tas Mics desharce Petseta wes ascites sea atiabiadis tie falatletle eiemretaat eile 158 WA BUS JY OA ON EEL La GOTO? Redhat Biictosbot oie co DBE GE odo acti SBOE RUE deco RERE BED DS ctice DO OSE pho tad ab CHEER GEE COD Mee eH en anacoee 165 FIREPNOGUCHVEHIOIOPY, faa ovsny: teeewss Aces coseateeiinstirwa daaelnn Gaels omsicsis «oR ni einahlebeeciadmanpsida/ubbriciebeuied opening Seana coseuw ce «aR 165 IDSISCUGSTOTN Gate gett tone cae cite Coe RRD BB GREE o GCC nOaERR Pere o RTT Cccp E TE eEROn. Ain oiinten AGUS cL scan ata tite du Gate deeee mes ena teamer cer 165 PRC IH O SS [CR PC INE ES ers ea. Means datas coin ois as eepiea Rook cule mERG onions nia na ok s SER pS as MMST Oe fn RRR re rineccnie see ymemwmmmsins 170 FRE LETC MICE Siete fee mete eyh e S Se e es te e I REN o ese eae Rts are arn «oi oa See AE RAE tee ce ass ick Eso stam nosroeaanaisie a eeiantiens 170 Synopsis. Males of the two known species of Mormonilla Giesbrecht are described in detail for the first time. They have reduced or vestigial mouthparts and are probably non-feeding. Males have a single testis and produce a single spermatophore. Male antennules are geniculate. A system of probable segmental homologies for the antennulary segments is proposed and the supporting evidence for this system discussed. Females carrying egg sacs are reported and figured for the first time. Mormonilla is unusual amongst copepods in producing paired egg sacs from a common medial genital aperture. * Present addresses: ' Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD. 2 Institut fiir Meereskunde an der Universitat Kiel, Diisternbrooker Weg 20, D 2300 Kiel 1, Germany. INTRODUCTION The order Mormonilloida was established by Boxshall in 1979 to accommodate the family Mormonillidae which consists of a single genus, Mormonilla, containing two species, M.phasma Giesbrecht, 1891 and M.minor Giesbrecht, 1891 (Boxshall, 1979). Both species have a very widespread distribution in the midwater oceanic plankton, especially at depths between 400 and 1000m. The skeletomusculature and external morphol- ogy of the feeding apparatus of adult females of Mormonilla indicate that they are small particle feeders (Boxshall, 1985). In a brief review of their biology Boxshall (1986) noted that no males of Mormonilla, or females carrying egg sacs have ever been reported, although he had observed a single female with a spermatophore attached to its median genital aperture. Mormonilla is relatively common in the Arabian Sea, in the deep eastern Indian Ocean, and in the Gulf of Aden; in the Red Sea its occurrence is restricted to the southernmost parts only (Beckmann, 1984; Bottger-Schnack, unpubl.). In zoo- plankton samples collected using a very fine mesh net (0.055 mm) several males were found, in addition to numer- ous females bearing egg sacs or spermatophores and some copepodid stages. The males of both species are described below in detail and short notes are presented on the repro- ductive biology of the genus. MATERIALS AND METHODS The specimens of Mormonilla were collected during cruises of the German Research Vessel Meteor using a multiple opening-closing net (Weikert & John, 1981) of mesh size 0.055 mm hauled vertically. The majority of the material was taken in the Arabian Sea and in the Strait of Bab el Mandab, the southern entrance of the Red Sea, during Meteor cruise 5/3a+b and 5/5 respectively (see Table 1). A single ovigerous 158 HUYS, BOXSHALL & BOTTGER-SCHNACK Table 1. Mormonilla specimens from the Arabian Sea, the Red Sea, the Mediterranean Sea and the eastern Indian Ocean. Station Position Date Arabian Sea 347 20°44'N 59°40’E 5.4.87 347 a 5.4.87 347 3 5.4.87 496 18°00'N 66°25'E T2587 Bab el Mandab, Red Sea ily) 12°2'N 43°24.5’E 6.8.87 Eastern Mediterranean Sea 34/35 34°25'N 26°14’E 20.1.87 34/35 34°25'N 26°4’E 20.1.87 Eastern Indian Ocean 11 04°47’S 87°14’E 24.1.77 11 ‘ 24.1.77 i ‘ 24.1.77 11 24.1.77 female of M.phasma was also found in zooplankton samples collected during Meteor cruise 64 at a depth of 550-650 m in the upwelling area of the N.W. African coast in spring of 1983 using a multiple opening-closing net of 0.3 mm mesh size. The specimens were dissected in lactic acid and examined as temporary mounts in lactophenol. All drawings were prepared using a camera lucida on a Leitz Dialux 20 interfer- ence microscope. All appendage segments and setation ele- ments are named and numbered using the system established by Huys & Boxshall (1991). The material is stored in the collections of The Natural History Museum, London, BM(NH) Reg.Nos. 1992.38-81. DESCRIPTION OF MALE Mormonilla phasma Body slender, cyclopiform (Fig. 1A—B). Body length of fig- ured specimen 1.42 mm; range 0.90 to 1.42 mm (based on 7 specimens). Prosome 5-segmented, comprising cephalosome and 4 free pedigerous somites; urosome 5-segmented, com- prising fifth pedigerous, genital and 3 free abdominal somites. Genital somite with functional genital aperture on left side only. Anal somite about 1.67 times longer than wide; ornamented with patches of fine spinules on ventral surface (Fig. 6A). Rostral region poorly developed, with pair of sensilla (Fig. 2A). Caudal rami elongate, about 18 times longer than wide (Fig. 5C); armed with 6 setae; seta I lacking, seta II located 34% of distance along ramus, setae III to VI postioned around distal margin, seta VII located Depth Number of specimens 1250-1450m 1 2 M.phasma (ovigerous) 2 O M.phasma 1450-1650m 42 2 M.phasma (2 ovigerous) 1 2 M.phasma (+ spermatophore) 1 CO M.phasma 57 copepodid stages 34 2 M.minor (2 ovigerous) 1650-1850m 1 2 M.phasma (ovigerous) 1 2 M.minor (+ spermatophore) 1650-1850m 2 2 M.phasma (ovigerous) 1 2 M.phasma (+ spermatophore) 2 2 M.minor (ovigerous) 125-150m 7 O M.minor 100-150m 2 2 M.minor 1 CO M.minor 2 copepodid stages 600-750m 1 2 M.minor (ovigerous) 1 CO M.minor 200m 1 CO M.minor 600m 2 O M.phasma 800m 2 CO M.phasma 1000m 2 0 M.phasma dorsally, just anterior to rear margin. Ramus ornamented with fine denticles. Antennule indistinctly 8-segmented (Fig. 2A); with well developed geniculation separating segments 7 and 8. Part proximal to geniculation comprising long proximal segment armed with 1 long and 6 short naked setae and 1 large aesthetasc, a middle section consisting of 5 poorly separated segments bearing 2, 2, 2 + 1 aesthetasc, 2 and 2 elements, and a long distal segment armed with 1 long and 2 short setae and 1 aesthetasc proximally, 2 hirsute setae at midlength (Fig. 2B) and a curved subapical seta. Apical segment slen- der, bearing 3 subapical setae, one of which with biarticulate base. Antenna biramous (Fig. 2C) with partly fused coxa and basis, both unarmed. Endopod 2-segmented; proximal seg- ment indistinctly separated from basis, unarmed; distal seg- ment representing fused second and third endopodal segments, armed with 11 setae around apex. Exopod 9-segmented, setal formula 1,0,0,0,0,1,1,1,3. Seta on first exopodal segment plumose, setae on distal part of ramus sparsely spinulate. Labrum (Fig. 3A) a simple rounded lobe. Paragnaths small, ridge-like lobes. Mandible (Fig. 4A) biramous; coxal gnathobase extremely reduced with poorly developed blades on margin (Fig. 4B); basis unarmed; endopod 2-segmented, first segment unarmed, second segment with 5 plumose setae; exopod 4-segmented with 2,1,1,2 setal formula (Fig. 4C). Maxillule biramous (Fig. 3B); praecoxa with weakly devel- oped arthrite (Fig. 3C); coxa and basis fused, bearing single ON DISCOVERY OF MALE OF MORMONILLA Fig. 1 Mormonilla phasma male. A, Entire, dorsal; B, Entire, lateral. 159 160 HUYS, BOXSHALL & BOTTGER-SCHNACK Fig. 2 Mormonilla phasma male. A, Antennule, dorsal; B, Detail, showing compound segment proximal to geniculation, anterior; C, Antenna, anterior. ON DISCOVERY OF MALE OF MORMONILLA 161 Fig. 3 Mormonilla phasma male. A, Cephalosome, ventral view showing cephalosomatic appendages on right side only; B, Maxillule, posterior; C, Detail of praecoxal arthrite of maxillule; D, Maxilla, anterior; E, Maxilla, lateral. 162 HUYS, BOXSHALL & BOTTGER-SCHNACK E72, Me PH: ren nee H | | i Sie 2 a ee Vat 3S \ C | an i \ FA av ve Zz eee \ at Z a NM ee va ee new, / f/ ys) i // / Fig. 7 Mormonilla minor male. A, Entire, dorsal; B, Entire, lateral; C, Caudal ramus, dorsal. 167 ON DISCOVERY OF MALE OF MORMONILLA oe = ATI TEI tI oe BEES Le Fig. 8 Mormonilla minor male. A, Mandible, anterior; B, Urosome, ventral; C, First leg, anterior. 168 HUYS, BOXSHALL & BOTTGER-SCHNACK \\ x \ N \ \ NS RSA AKG AER IT ASS Ss Tread a SS = N SS ORS A Taare Xf aX : sss — —- SA SERS La / (| aN 50 yw RONSSSSO. % bs . A-C ERE Le OR AS) LUKENS Pa a <> 4 J ' £ ? aha 4 aa So Soca Sn pons -iecareteserrerse ketet rt n . >~. phere Saige "a trascesew ovaeae _ : : : 4 3 ; : : = SEPSIS ices Sata So ieceng tang ines “ - Sr z - ae soak peeees SEES i Seictere sane ; aes! 15-33 : : Sh wee Meaperssesier tes : > 2 ; ; Spiotges gin sree ne] Ls enw kaos : : : ic - SETAE : gro $IOF posse Setpataae lonely ceeseee a om sary ce 2 f xs - omy ens reste 1uasintateae deerysaaes