BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY VOL. I 1950-1953 PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM (NATURAL HISTORY) LONDON : DATES OF PUBLICATION OF THE PARTS No. i. 17 January 195 No. 2. 20 March 1950 No. 3. 31 March 1950 No. 4. 29 August 1950 No. 5. 3 November IQ5 I No. 6. 26 October IQ 5 I No. 7. 6 June 1952 No. 8. 5 August 1952 No. 9. 31 July 1952 No. 10. ii December 1952 No. ii. 13 January 1953 No. 12. 27 November 1953 PRINTED IN GREAT BRITAIN AT THE BARTHOLOMEW PRESS DORKING BY ADLARD AND SON, LTD. CONTENTS ZOOLOGY VOLUME I No. i. On some species of Lernaea (Crustacea, Copepoda : parasites of fresh- water fish). By j. P. HARDING i No. 2. On a giant squid, Ommastrephes caroli Furtado, stranded at Looe, Cornwall. By w. j. REES (Pis. 1-2) 29 No. 3. The identity of Captain Cook's kangaroo. By T. c. s. MORRISON- SCOTT and F. c. SAWYER (Pis. 3-5) 43 No. 4. Notes on Asteroids in the British Museum (Natural History). By D. DILWYN JOHN (PL 6) 51 Lernaeodiscus pusillus nov. spec., a Rhizocephalan parasite of a Por- cellana from Egypt. By HILBRAND BOSCHMA 61 No. 5. On a rare deep-sea fish Notacanthus phasganorus Goode (Heteromi- Notacanthidae) from the Arctic Bear Isle fishing grounds. By D. w. TUCKER and j. w. JONES (Pis. 7-9) 67 No. 6. The ocean sunfishes (family Molidae). By A. FRASER-BRUNNER 87 No. 7. The cestodes of seals from the Antarctic. By STANISLAW MARKOWSKI (Pis. 10-21) 123 No. 8. The ' Manihine' Expedition to the Gulf of Aqaba 1948-1949 Foreword . Station list and collectors' notes (Pis. 22-27) I 53 Preliminary hydrological report. By G. E. R. DEACON 159 Sponges. By MAURICE BURTON 163 Turbellaria . Polycladida. By STEPHEN PRUDHOE 175 Gephyrea. By A. c. STEPHEN 181 Mollusca. By w. j. REES and A. STUCKEY (Pis. 28-30) 183 Echinodermata. By AILSA M. CLARK (Pis. 31-32) 203 Tunicata. By WILLARD G. VAN NAME 215 Fishes. By N. B. MARSHALL 221 No. 9. On the species and races of the yellow wagtails from Western Europe to Western North America. By c. H. B. GRANT and c. w. MACKWORTH- PRAED (Pis. 33-35) 253 No. 10. Mammals collected by Mr. Shaw Mayer in New Guinea 1932-1949. By ELEANOR M. O. LAURIE 269 No. ii. Taxonomy of the karroo and red-back larks of Western South Africa. By J. D. MACDONALD (Pis. 36-38) 319 No. 12. Suberites domuncula (Olivi) : its synonymy, distribution, and ecology. By M. BURTON 353 Notes on Asteroids in the British Museum (Natural History) III & IV. By A. M. CLARK (Pis. 39-46) 379 Some inter-tidal mites from south-west England. By G. o. EVANS and E. BROWNING 413 Index to Volume I 423 ERRATUM Plates 7-9. For NOTOCANTHUS read NOTACANTHUS. !- J .17JANB5U- ' * ' V^k ON SOME SPECIES OF LERNAEA (CRUSTACEA, COPEPODA: PARASITES OF FRESH-WATER FISH) J. P. HARDING BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. i LONDON: 1950 ON SOME SPECIES OF LERNAEA (CRUSTACEA, COPEPODA: PARASITES OF FRESHWATER FISH) BY J. P. HARDING Pp. 1-27; 95 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. i LONDON : 1950 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is to be issued in five series, corresponding to the Departments of the Museum. Parts will appear at irregular intervals as they be- come ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. i, No. i, of the Zoological series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued January 1950 Price Seven shillings and sixpence ON SOME SPECIES OF LERNAEA (CRUSTACEA, COPEPODA: PARASITES OF FRESHWATER FISH) By J. P. HARDING (With ninety-five text-figures) SYNOPSIS Twenty-eight species of Lernaea are recognized, of which fourteen are represented in the collections of the British Museum. Nine of these are new species. In addition there are seven names in the literature which are relegated to the synonymy. A key to the twenty-eight recognized species is given and the fourteen species in the Museum are described and figured. ONE result of the renewed interest in the freshwater fisheries of Africa and other countries in recent years has been an accumulation of the parasites of these fish in the British Museum with requests for their identification. Attempts to name the species of Lernaea on the basis of existing descriptions and keys soon showed that a high proportion of them were new, and that Cunnington's (1914) and Wilson's (1917 & 1918) revisions and keys are out of date. There is no need here to repeat recriminations against Wilson (1917) for transposing the names for the genera Lernaea and Lernaeocera. The inevitable confusion caused by this strict interpretation of the Rules of Nomenclature has fortunately been lessened by the fact that subsequent workers have, however unwillingly, nearly all agreed to follow. As is often the case with degenerate parasitic forms, the characters used to distin- guish between species are often ill-defined and not easily seen. Wilson often expressed the opinion that species of Lernaea and other parasitic copepods could readily be distinguished by reference to their appendages; but I have found extraordinarily little difference between the appendages of one species and those of another. The first four pairs of pereiopods, which will be referred to as legs i to 4 in this paper, are the only appendages that are easily examined as they are flat, and as each has a number of setae and spines I expected variations in their arrangements to provide useful characters for distinguishing species. Unfortunately I could find hardly any variations. The precise arrangement of the setae and spines was investigated in fourteen species, and in thirteen of them it was identical, only one of them could be separated on this basis. Table i gives the arrangement of the setae and spines in the thirteen species. L. bistricornis is included although I was able to see only legs 3 and 4 ; all four pairs of legs were seen in the other species. L. haplocephala differs from all these species in having four setae instead of five on the terminal segment of each exopod (Table 2, p. 19). L. oryzophila, according to Monod's (1932) descrip- tion and figures, differs from the other species, having four instead of three spines on the last segment of the second exopod and two spines instead of three on the last segment of the fourth exopod. L. dolabroides (Wilson, 1918, figs. 77 and 78) seems to have a quite different setation of its legs. ON SOME SPECIES OF LERNAEA TABLE i Arrangement of the setae and spines on the legs of L. bagri, L. barbicola, L. barilii, L. barnimiana, L. bistricornis, L. cyprinacea, L. dicer acephala, L. longa, L. lophiara, L. palatae, L. piscinae, L. tilapiae, and L. tuberosa Leg i Leg 2 Leg 3 Leg 4 \ spines Ex P d (setae 1. 1. a 1.1.5 1.1.3 I.I-5 1.1.3 1.1.5 1.1.3 1.1.5 ( spines Endopod( setae O.O.2 I.I.4 O.O.2 1.2.4 O.O.2 1.2-4 O.O.2 1.2.3 The other appendages are even less useful than the legs for separating one species of Lernaea from another. Wilson (1920, p. 7) claims that L. haplocephala may be distinguished by the 'small spherical terminal joint of the maxillipedes, with its four curved claws '. This may have been true of the specimen he examined, but I find that the maxillipedes of the type specimens of L. haplocephala have five claws like any other species and that the terminal segment does not appear to have any specific shape. Fig. 34 gives the arrangement of the mouth parts as far as I have been able to see these very minute appendages. We are left with the shape of the body and with the internal anatomy for distin- guishing species. Unfortunately it is difficult to study one without destroying, or at least distorting, the other, and I have neglected a study of the internal anatomy in favour of the shape of the body and its processes, and in particular that of the anchor. I am restricting the use of the word 'head' to that small, rounded part which bears the antennae and the mouth parts. The swollen part with processes which are usually described as 'cephalic processes', 'cephalic arms', or 'cephalic horns ' I propose to call the anchor and its arms, as this describes both the appearance and the function of this part of the body. The anchor is often difficult to remove from the flesh of the fish without damage, and I have adopted the method of cutting out the part of the fish with the parasite embedded in it and placing it in a tube with a solution of potassium hydroxide to which a little chlorazol black has been added. If this is left for about twenty-four hours the tissues of the fish are usually softened sufficiently for the Lernaea easily to be removed ; the chlorazol black stains the chitin of the parasite a dark blue. The external shape of the animal is well preserved by this method and the appendages can be examined without difficulty. The egg-sacs should be removed before placing the parasite in the hydroxide or they will be destroyed. All drawings and measurements of specimens recorded in this paper have been made with the aid of a camera lucida. The total length of a specimen is understood to mean the length from the front of the head to the end of the abdomen, allowance being made for bends and curves in the body. Parts of the anchor which may project in front of the head and the furcal setae are not included in this measurement. The measurement and drawing of curved specimens was helped by the use of gimbals which enabled the specimen to be held in any position on the stage of the microscope. The gimbals, similar in principle to those of a ship's compass, but with sufficient ON SOME SPECIES OF LERNAEA 5 friction in the bearings to prevent free swinging, were made of concentric cylinders of perspex, as shown in Fig. i. The whole is submerged in formalin in a glass vessel and the specimen is placed in the central cylinder, which is closed at the bottom to form a dish. I have found the shape of the anchor and its arms to provide the most useful characters for taxonomic purposes. In spite of the fact that the shape of the anchor is liable to be distorted by meeting bones and other hard obstructions during its growth in the body of the fish, each species has a characteristic form which varies within limits which are usually definable provided sufficient material is available. FIG. i. Gimbals for tilting the specimen into positions required for drawing or measuring. The abdomen and the pregenital prominences of each species examined have been drawn with some care from more than one aspect as I have found them useful in separating species. The shape of these parts is not so easily influenced by the site of attachment of the parasite, but on the other hand there is often a difference in the shape of these parts of a young individual and those of an old one. The characters of the immature specimens are often less well defined than those of adult specimens, and the adult female seems to continue to grow and develop specific form for some time after eggs are first produced. The positions of the legs on the body have been recorded wherever possible with an estimate of the range of variability. The amount of torsion and its direction is very variable; but worth recording because some species show little torsion and in others considerable torsion is the rule. In some species, such as L. barnimiana, the torsion is not only variable in direction and extent in different specimens but often changes its direction along the length of the one individual. Cunnington (1914) remarks on the rarity of copepod parasites on the fish he collected from Lake Tanganyika. Lake Nyasa, however, seems to be very rich in species. More information is required about the seasonal distribution of the different species. Miss R. H. Low tells me that she found Lernaea on many of the specimens of Bagrus that she collected in August and that by November the Bagrus were free of parasites but the Tilapia were infected. She had the impression that the parasites 6 ON SOME SPECIES OF LERNAEA had transferred their attentions from Bagrus to Tilapia ; but an examination of the specimens shows that there are two distinct species of Lernaea involved, L. bagri and L. tilapiae, described below. Lernaea cyprinacea Linnaeus FIGS. 2-12. 1746 Lernea tentaculis quatuor Linnaeus, Fauna Svecica: 367, pi. 2. 1758 Lernaea cyprinacea Linnaeus, Syst. Nat.: 655. 1783 ,, : Barbut, Gen. Vermium: 67, pi. 7, fig. 3. 1822 Lerneocera cyprinacea (Linnaeus) Blainville, Journ, Phys. 95: 377. 1835 Lernaeocera ,, : Burmeister, Nova Acta Leap. Carol. 17: 309, pi. 24 A, figs. 1-3. 1850 ,, ,, : Baird, Brit. Entomost: 343, pi. 35, fig. 13. 1904 ,, : Hofer, Handb. Fischkrankheit, Munchen: 144, fig. 95 and p. 119 [fide Pesta 1934]. 1909 ,, ,, : Neresheimer, Brauer: Susswasserfauna Deutschl. 11: 77, fig. 326. 1913 ,, (part): T. and A. Scott, Brit. Parasit. Cop., Ray Soc. London: 154, pi. 50, figs. 4-5 [not figs. 1-3]. 1917 Lernaea cyprinacea: Wilson, Proc. U.S. Nat. Mus. 53: 4, 39. 1918 ,, : Wilson, Bull. U.S. Bur. Fish. 35: 193, 196, pi. 15, fig. 86. 1925 ,, (Lernaeocera) elegans: Leigh-Sharpe, Parasitology, 17: 245, text-figs. 1-5. 1927 ,, elegans: Nakai, /. Fish. Inst. Tokyo, 23: 39, pis. 2-4, text-figs. 1-7. 1927 ,, cyprinacea: Okada, Annot. Zool. Jap. Tokyo, 11: 185, text-figs. 1-2. 1928 ,, elegans: Matsui and Kumada, /. Fish. Inst. Tokyo, 23: 101, pis. 5-7. J 932 ,, cyprinacea: Monod. Ann. Parasit. hum. comp. 10: 362, text-figs. 8 H, n, 12. 1933 ,, ,, : Gurney, Brit. Fresh-water Cop, Ray Soc. Lond. 3: 338 , text-figs. 1969, 1971-1983. 1 933 carassii Tidd, Ohio J. Sci. 33: 465, pi., figs. 1-8. !Q34 ,, cyprinacea: Markewitsch, Ann. Mus. zool. polon. 10: 234, pi. 45, fig. 8. 1934 : Pesta, Tierwelt Deutschl. 29: 42, text-fig. 25. 1937 ' Markewitsch, Cop. Parasit. Binnengewdss. U.S.S.R., Kiev: 98, pi. 8. 1937 ,, ,, : Wagler, Tierwelt Mitteleuropas, Crust. 2, 2a: 179, text-fig. 542. 1939 ,, ,, : Yamaguti, Vol. Jubil. Prof. S. Yoshida 2: 475, pi. 30, figs. 156-165. The material in the British Museum consists of two specimens found by Dr. Gurney on a specimen of Carassius carassius (L.) from Sweden in the Museum fish collection ; a specimen from Canon Norman's collection which was labelled ' L. esoscina from Prof. Heller ' (this specimen is unfortunately without record of host or locality) ; thirty or more specimens from Japan presented by Dr. Gurney, and finally a few microscope slides of the type specimens of L. elegans presented by Mr. Leigh-Sharpe. I have little to add to the excellent descriptions and figures given by many of the authors listed above, Gurney (1933) in particular ; the shape of the anchor is, however, rather more variable than these descriptions indicate. The most typical arrangement is that of the Swedish specimen (Fig. 2), if the right dorsal arm which is distorted is ignored ; the arms are all rather long and slender and the dorsal arms are T-shaped. This is the arrangement shown in nearly all figures of European specimens from Linnaeus, 1746 to Monod, 1932, and Gurney, 1933. Very few of the Japanese speci- mens are quite like this, there is a tendency for the dorsal arms to be Y-shaped (Figs. 5-7), and the posterior fork of the Y is often reduced. Prof. Heller's specimen (Figs. 10-12) of unknown origin is very like the Japanese specimens . The pregenital ON SOME SPECIES OF LERNAEA 7 prominence of L. cyprinacea is generally described as 'simple or only slightly in- dented' (Wilson, 1918, p. 193, key). It is simple in the Swedish specimens (Figs. 3 and 4). The Japanese specimens have a distinctly double pregenital prominence (Figs. 8 and 9). Prof. Heller's specimen again agrees with the Japanese specimens FIGS. 2-12. Lernaea cyprinacea Linnaeus Figs. 2-4, specimen from Sweden; Fig. 2, dorsal view of anchor; Fig. 3, ventral view of pregenital prominence and abdomen; Fig. 4, lateral view of the same; Figs. 5-9, specimens from Japan; Figs. 1012, specimen from Prof. Heller. The line near each figure is i mm. drawn to the same scale. rather than with the Swedish ones. Markewitsch (1937) also gives a figure of a speci- men with a bilobed pregenital prominence, but he does not say from what part of the U.S.S.R. it came. None of the characters of L. elegans given by Leigh-Sharpe are valid for distinguishing between the Japanese form and the European. No ' auricular expansions' are now visible on the specimen from which Leigh-Sharpe's Fig. 3 was based ; possibly the artist saw some folds of chitin. In this figure are shown what are called 3-segmented uncinate thoracic appendages, but an examination of the speci- men shows that these are not appendages at all but are the badly fixed internal tissues which can be seen only by focusing well below the surface of the body, as Monod (1932) has already pointed out. Leigh-Sharpe's types of L. elegans are the same form of L. cyprinacea as the Japanese specimens I have seen. 8 ON SOME SPECIES OF LERNAEA I have looked at the appendages of the Swedish and Japanese specimens with some care, but have been unable to find any difference, the setation of the legs is precisely the same in both (Table i). The positions of the legs on the body of five Japanese specimens were measured; there was considerable variation, particularly of the anterior legs, but the positions of the five pairs of legs do not enable the Japanese and European specimens to be separated from one another. The positions of the five legs were 6-9, 16-20, 42-45, 73-74, and 90-92 per cent, of the total body length distant from the most anterior part of the head, respectively. There is little doubt that L. carassii Tidd is the Japanese or elegans form of L. cyprinacea. Lernaea barnimiana (Hartmann) FIGS. 13-28 1865 Lernaeocera barnimiana Hartmann, Naturges.-med. Skizze Nillander: 206. 1870 barnimii Hartmann, Arch. Anat. Phys. Wiss. Med. 1870: 726, pis. 17-18. 1871 ,, ,, : Hartmann, 5. B. naturf. Fr. Berl.: 60. 1914 temnocephala Cunnington, Proc. Zool. Soc. Lond. 1914: 827, pi. i, figs. 8-9, text-fig, i c. 1917 Lernaea barnimii: Wilson, Proc. U.S. Nat. Mus. 53: 38. 1918 ,, temnocephala: Wilson, Bull. U.S. Bur. Fish. 35: 193, 196, pi. 15, fig. 87. 1918 barnimii: ibid.: 193, 196, pi. 15, fig. 94. 1940 ,, temnocephala: Brian, Boll. Idrobiol. Caccia Pesca, 1: 50, pi., figs. A-F. 1944 barnimiana: Capart, Bull. Mus. Hist. nat. Belg. 20 (24): 2, text-fig. I. Several specimens of this species were taken from a fish, Labeo forskalii Ruppell, caught in Lake Edward by Dr. E. B. Worthington in 1931. The heads of the parasites were buried in the flesh of the head and inside the mouth of the fish. The Museum also possesses the single specimen on which Cunnington founded L. temnocephala. Thanks to the kindness of Dr. Capart I have seen three of the specimens he described from the Belgian Congo. The length of the adult female, judging from the literature, ranges from 7 mm. to 12 mm. (Capart's 1944 figs.) ; Hartmann's 1870 record of a range of from 10 mm. to 14 mm. may include the anterior arms of the anchor in the length. The British Museum specimens range from 8-2 mm. to 10-8 mm. in length. . The positions of the five pairs of legs of seven of the Lake Edward specimens give the following ranges measured in percentages of the total body length: 7-2-9-8, 19-24, 41-51, 73-79, and 89-92 per cent, respectively. The positions of the legs on the specimens kindly lent me by Dr. Capart agree, but two other specimens which he figures (Capart, 1944, fig. i, A and E) appear to have the first and second legs a little farther forward ; this may, however, be owing to the foreshortening which is inevitable when curved specimens are drawn. Hartmann's (1870) drawings are not very reliable and I attach no importance to the fact that the position of the fourth pair of legs in his Fig. i is only 65 per cent, of the body length from the anterior. The positions for Cunnington's type of L. temnocephala are 8, 18, 44, 79, and 92 per cent, respectively. The variations in the positions of the legs seems to be quite independent of the size of the specimen, i.e. there is no heterogony with respect to this character. ON SOME SPECIES OF LERNAEA The torsion of the specimens I have seen was variable ; that between successive pairs of legs never exceeded 90 and was usually much less. It could be either sinistral or dextral and frequently changed its direction. The total torsion of the whole body was not more than 120 in any of the twelve specimens examined. FIGS. 13-28. Lernaea barnimiana (Hartmann). Figs. 13-16, Cunnington's specimen from L. Tanganyika, named by him temnocephala ; Fig. 13, ventral view of anchor; Fig. 14, dorsal view; Fig 15, ventral view of abdomen and pregenital prominence; Fig. 16, lateral view of the same. Figs. 17-28, specimens from L. Edward. Fig. 17, ventral view of anchor; Fig. 18, ventral view of another specimen; Fig. 19, anterior view of the latter; Figs. 20 and 21, posterior end of this specimen; Fig. 22, posterior end of a specimen with egg-sacs; Figs. 23-28, anchors of other specimens from L. Edward, all from the same fish. The line near each figure is i mm. drawn to the same scale. The arms of the anchor are rather variable in shape and arrangement, as Capart (1944) has shown. The most usual arrangement is for the part between the head and the first legs to be swollen and more or less globular. The ventral arms are simple in shape and very short. The usual arrangement is for the ventral arms to be directed outwards ; this was so in all the Lake Edward material, in Cunnington's specimen from the Nile, and in most of Capart's material from the Belgian Congo. In some of Capart's specimens and also in the figure accompanying Hartmann's description (Hartmann, 1870, fig. i), on the other hand, the ventral processes are directed ante- riorly. With regard to the bifurcating dorsal arms, Hartmann's figure shows both branches equal to one another and both diverging slightly away from the body; but in his description he says that the anterior branch is the longer of the two and zoo. i, i. B io ON SOME SPECIES OF LERNAEA is the more outwardly directed. The normal condition seems to be for the anterior branch to diverge widely from the body while the posterior branch is directed slightly inwards (Figs. 17, 18, 23, &c.). The angle between the two branches is normally an open and continuous curve ; but sometimes as in Figs. 25, 27, 28, &c. there is a more or less distinct angle. The Y-shaped condition of the arms of the temnocephala holo- type (Fig. 14) is unusual but within the range of variation of L. barnimiana. The pregenital prominence is distinct and bilobed (Figs. 15, 16, 20, and 21) ; but from some aspects it may appear to be a single broad process. The abdomen is distinctly 3-segmented and may continue the line of the body or be set at an angle. Each segment is a little smaller than the preceding one. Hartmann does not describe the abdomen of his specimens, and his figure and those of Brian (1940) are of little value in this respect. Cunnington's temnocephala specimen (Figs. 15 and 16) is normal. The setation of the legs is the same as that of L. cyprinacea (Table i, p. 4). I have cleared Cunnington's type of temnocephala with potassium hydroxide, and this has got rid of the twists and distortions he mentions and has enabled me to examine the setation of its legs; and as with the other characters investigated I can find no difference between this specimen and the Lake Edward specimens and I have no hesitation in placing L. temnocephala (Cunnington) in the synonymy of L. barnimiana, as Capart (1944) has already suggested. Lernaea piscinae sp. nov. FIGS. 29-34 Holotype, Reg. No. 1949.8.14.1, and many paratypes, all females, in the British Museum. The parasites were found heavily infesting a Cyprinid fish, Hypophthal- micthys nobilis (Richardson) cultivated by the Chinese on a fish farm at Singapore. Four fish heavily infested with the parasites, over 50 per fish, were presented to the Museum by Mr. W. Birthwhistle in 1929. Length of holotype 10-4 mm. ; the length of io paratypes ranged from 9-7 mm. to 12-4 mm. The positions of the five pairs of legs of these eleven specimens were 5-7, 13-14, 31-38, 69-74, and 91-93 per cent, of the total length from the most anterior part of the head. All the specimens of this species were very much alike ; seven out of the eleven specimens had a curve between legs 2 and 3 as in Fig. 29. Three of the remainder were straight and the other had an additional bend between legs i and 2. The main part of the anchor forms a bar set at right angles to the body like the cross-bar of a T (Figs. 29 and 30) . The middle of the bar is considerably thicker than the part of the body joined to it, and tapers gradually towards the ends, which are also curved slightly in an antero-ventral direction. From about the middle of each half of the cross-bar there is a short dorsal process. There is also a pair of ventral processes between the head and the legs i ; these are separated by a distance about equal to the width of the head and are directed slightly inwards towards one another. Except for slight swellings at the positions of the legs the body increases in thick- ness very gradually from before backwards. The abdomen (Figs. 32-33) makes a slight angle with the body ; it is nearly i mm. ON SOME SPECIES OF LERNAEA long and less than ^ mm. wide. Ventrally it is distinctly 3-segmented ; but the dorsal profile forms an even, continuous curve. The pregenital process is double, the two lobes being small but well denned and quite separate from one another. md mxl 31 FIGS. 29-34. Lernaea piscinae sp. nov. Fig. 29, ventral view of holotype ; Fig. 30, anterior view of anchor; Fig. 31, lateral view of posterior end of a para type with egg-sacs; Fig. 32, ventral view of abdomen and pregenital prominences of holotype; Fig. 33, lateral view of the same; Fig. 34, mouth and associated appendages. /./., lower lip; md., mandible; mxa., maxilla; mxl., maxillule. The line near each figure is i mm. drawn to the same scale except that near fig. 34 which is 0-02 mm. The egg-sacs (Fig. 31) are very long, about 4 mm., three-quarters of their length projecting beyond the tip of the abdomen. The setation of the legs is the same as in L. cyprinacea (Table i). The mouth parts (Fig. 34), as far as I was able to make out, are the same as for other species. Lernaea diceracephala (Cunnington) FIGS. 35-39 1914 Lernaeocera diceracephala Cunnington, Proc. Zool. Soc. Lond. 1914: 824, pi. i, figs. 1-3, text-fig, i A. 1917 Lernaea diceracephala: Wilson, Proc. U.S. Nat. Mus. 53: 38. 1918 ,, ,, : Wilson, Bull. U.S. Bur. Fish. 35: 192, 194, pi. 15, fig. 90. 1944 " " : Capart, Bull. Mus. Hist. nat. Belg. 20 (24): 7. Holotype, Reg. No. 1914.12.2.1, and one paratype in the British Museum; I have 12 ON SOME SPECIES OF LERNAEA selected the more perfect of the two specimens as the holotype. These are the only specimens known and were taken from the gill arches of a large Clarias mossambicus Peters, caught at Sumbu, Lake Tanganyika, by Dr. Cunnington in 1904. Capart (1944) includes the species in his paper because part of Lake Tanganyika lies in the Belgian Congo. Cunnington's description of the two specimens is very good ; but he describes the left arm as being complete in the better specimen when in fact the tip has been broken off. The length of the holotype measured as if straightened out is 9-1 mm. The five pairs of legs come in positions 10, 23, 50, 71, and 92 per cent, of the total length from the most anterior part of the head. I have made drawings of the two specimens which I hope are an improvement on Cunnington's photographs, and which show how similar to one another are the bends and constrictions in the body. Lernaea bagri sp. nov. FIGS. 40-43 Holotype, Reg. No. 1949.8.14.9, and over two dozen paratypes, all females, in the British Museum. The copepods were taken from Bagrus meridionalis in Lake Nyasa by Miss R. H. Low, 14 Aug. and 22 Sept. 1946. The length of the holotype is 12-1 mm. ; that of twenty-four adult females carrying egg-sacs ranged from 9-9 mm. to 14-2 mm. The body of a few of the slender, young-looking specimens is straight, but usually it is curved as shown (Fig. 40), and in these there is a torsion which in this species nearly always changes its direction. In all the specimens I have examined for this purpose the torsion is first sinistral and then dextral. In the holotype, for example, the torsion between legs i and 2 is 40 sinistral, between legs 2 and 3 it is 80 dextral, and between legs 3 and 4 it is a further 50 dextral, after which there is no further torsion. The resultant torsion between the head and the abdomen is about 90 dextral. The arms of the anchor are heavily chitinized, in contrast to those of the next species to be described, L. tilapiae, and lie in a plane approximately at right angles to the body. The head is placed centrally over the cross formed by the four arms. The ventral arms are straight and the dorsal ones are curved towards them. There is a tendency for each arm to end with a rounded knob. The positions of the legs are rather variable in this species. Six specimens were examined, and the positions of legs i to 5 gave the following ranges respectively: 7-10, 18-22, 42-52, 73-78, and 90-93 per cent. The setation of the legs is the same as that of L. cyprinacea (Table i). The abdomen (Figs. 41-3) is set at a slight angle to the line of the body; it is straight and slightly tapering ; the three segments are very indistinctly separated. The pregenital prominence is bilobed. Sometimes the lobes are prominent and bulge laterally beyond the greatest width of the body, but usually, as in the holotype, they are not very prominent from the ventral aspect (Fig. 43). ON SOME SPECIES OF LERNAEA The egg-sacs (Fig. 41) are about 2| mm. long and \ mm. wide at their greatest width, which lies at about the proximal third of the length. FIGS. 35-39. Lernaea diceracephala (Cunnington) . Fig. 35, ventral view of holotype; Fig. 36, dorsal view of abdomen; Fig. 37, constriction between legs 3 and 4 from a slightly different aspect from that of Fig. 35 ; Fig. 38, lateral view of paratype; Fig. 39, similar view of part of holotype. FIGS. 40-43. Lernaea bagri, sp. nov. Fig. 40, dorsal view of holotype; Fig. 41, lateral view of abdomen and egg-sacs of a paratype; Fig. 42, lateral view of abdomen and pregenital prominences; Fig. 43, ventral view of the same. The line near each figure is i mm. drawn to the same scale. Lernaea tilapiae sp. nov. FIGS. 44-46 Holotype, Reg. No. 1949.8.14.17, and a few paratypes, all females, in the British Museum. The parasites were collected by Miss R. H. Low from Lake Nyasa and were taken from the mouth and gills of Tilapia squamipinnis Giinther and T. lidole Trewavas caught in Lake Nyasa 22 Nov. 1946. The length of the holotype measured from the front of the head to the tip of the 14 ON SOME SPECIES OF LERNAEA abdomen is 9-2 mm. Five other females bearing egg-sacs ranged from 7-5 mm. to ii mm. in length. The body is comparatively slender from the head to as far as legs 3 and is usually curved here, so that the anterior part of the body is at right angles to the broad part behind legs 3 (Fig. 44). In the holotype the torsion is dextral 45 between legs 2 and 3, 45 FIGS. 44-46. Lernaea tilapiae sp. nov. Fig. 44, dorsal view of holotype; Fig. 45, lateral view of abdomen, pregenital prominence, and egg-sacs; Fig. 46, ventral view of the same without egg-sacs. The line near each figure is i mm. drawn to the same scale. dextral 90 between legs 3 and 4, and dextral a few degrees beyond legs 4, the total torsion being dextral through about 140. The only torsion in one of the paratypes is a sinistral one of 45 between legs 3 and 4. The anchor bears four long straight slender arms as figured (Fig. 44) ; these lie in a plane nearly parallel to that of the body ; the posterior pair are directed backwards and are only slightly divergent; the anterior pair diverge widely, with the head placed in the angle between them. The four arms of the anchor are about equal in length to one another and more than half the length of the body. They are only lightly chitinized and are much softer than those of the last species described, L. bagri. ON SOME SPECIES OF LERNAEA 15 The legs come in positions 8, 25, 50, 77, and 90 per cent, of the body length from the anterior end. The setation of legs I to 4 is the same as for L. cyprinacea (Table i, P-4)- The abdomen is divided into three segments by transverse ventral constrictions which give it a characteristic profile (Fig. 45). The dorsal profile of the abdomen is slightly arched. The pregenital prominence is bilobed; the two lobes overhang the abdomen slightly, but their ventral surface is in line with that of the body in front (Fig. 45). The egg-sacs are about 2-5 mm. long and 0-5 mm. wide, slightly tapering towards each end. Miss Low records that in life the parasite is brown in colour and the eggs are jade-green. Lernaea barilii sp. nov. FIGS. 47-60 Holotype, Reg. No. 1949.8.14.21, and about 10 paratypes, all females, in the British Museum. The parasites were taken on a large specimen (500 mm. long) of Barilius microlepis Giinther from Lake Nyasa by Dr. Christy in 1925, a piece of the flank of the fish with the copepods embedded being preserved together with a note to the effect that there were more parasites on the tongue, &c. I have only seen the specimens from the flank. The length of the holotype is 8-3 mm. with the positions of legs i to 5 at 8-4, 20, 47, 77, and 92 per cent, of the total body length from the anterior end respectively. The positions on paratype Reg. No. 1949.8.14.24 are 8, 19, 47, 77, and 93 per cent. The setation of the legs is the same as in L. cyprinacea (Table i, p. 4). The body is straight and short, widest at the posterior end. In the two specimens which were examined in detail the torsion was about 80. In the holotype most of the torsion was between legs 2 and 3, and in the paratype examined it was between legs 3 and 4 ; it was sinistral in the holotype and dextral in the paratype. The arrangement of the anchor is best understood with reference to Figs. 47-49 and 53-55. There are a pair of lateral T-shaped arms with the cross-bar of the T running more or less parallel to the body ; the basal part of these arms is short and thick. Anterior and ventral to the dorsal arms are a pair of simple arms directed outwards, with in nearly all cases a small knob facing anteriorly. The pregenital prominences are very large and distinctly separated from one another ; they reach almost to the end of the abdomen in some specimens (Figs. 40, 50-52, 57- 60 )- The abdomen, particularly the part composed of the last two segments, is very small and set at an angle to the body. All three segments are clearly marked off from one another ventrally; the first is very much broader than the other two (Figs. 51 and 60). The egg-sacs of the holotype were about 2.75 mm. long, broadest in the middle and tapering towards each end (Fig. 58). ON SOME SPECIES OF LERNAEA FIGS. 47-60. Lernaea barilii sp. nov. Fig. 47, ventral view of holotype; Fig. 48, dorso-lateral view of anchor; Fig. 49, lateral view of the same; Fig. 50, dorsal view of posterior end of holotype; Fig. 51, lateral view of the same; Fig. 52, ventral view of the same; Fig. 53, ventral view of anchor of a paratype; Fig. 54, anterior view of another paratype; Fig. 55, ventral view of the same; Fig. 56, anchor of a specimen drawn in situ by clearing in benzyl alcohol; Fig. 57, lateral view of specimen with egg-sacs before removing from the fish (the specimen has shrunk and collapsed dorsally); Fig. 58, ventral view of holotype with egg-sacs before treatment with hydroxide; Fig. 59, ventral view of the posterior end of a paratype; Fig. 60, lateral view of the same. The line near each figure is i mm. drawn to the same scale. Lernaea palati sp. nov. FIGS. 61-64 Holotype, Reg. No. 1949.8.14.26 in the British Museum. The single specimen on which this species is based was from the roof of the mouth of a fish, Haplochromis chrysonotus (Boulenger) from Vua on Lake Nyasa, collected by Dr. Christy in 1925. The hind end of the parasite projected through a gill slit and was visible externally. The length of the specimen, allowance being made for bends, is 12-7 mm. The body from the first pair of legs to half-way between legs 2 and 3 is cylindrical, about 07 mm. thick; the section containing legs 3 is broader, about 1-2 mm. across; there is a waist between legs 3 and 4 ; the body bends backwards and to the left here and bulges again to a thickness of 1-2 mm. in front of legs 4. The abdomen is tilted dorsally at an angle of about 45 ; it is a simple cylinder rounded at the end about i mm. long and 0-7 mm. broad without any sign of seg- mentation. ON SOME SPECIES OF LERNAEA The five pairs of legs are placed in positions 8-7, 26, 55, 80, and 93 per cent, of the body length from the anterior end. The setation is the same as that of L. cypri- nacea (Table i, p. 4). There is little torsion. FIGS. 61-64. Lernaea palati sp. nov. Fig. 61, ventral view of holotype; Fig. 62, lateral view; Fig. 63, holotype embedded in roof of mouth of fish; Fig. 64, another view of the same showing egg-sacs. FIGS. 65-68. Lernaea longa, sp. nov. Fig. 65, lateral view of holotype showing anchor and swelling by legs 2. hd, head; Fig. 66. A paratype with the left ventral arm of the anchor distorted; Figs. 67 and 68, the externally visible parts of two other specimens embedded in the fish. The line near each figure is i mm. drawn to the same scale. The anchoring arms (Figs. 61 and 62) are four in number, of medium length and uniform thickness, each with a bend or a kink near the end. The ventral pair is directed slightly forwards and the dorsal pair backwards to the same extent. The head is not in line with the body, but inclined towards the angle between the ventral arms of the anchor. The egg-sacs are comparatively short and broad, being about 1-5 mm. long and 0-5 mm. broad (Figs. 63 and 64). zoo. i, i. i8 ON SOME SPECIES OF LERNAE Lernaea longa sp. nov. FIGS. 65-68 Holotype, Reg. No. 1949.8.14.27, and half a dozen paratypes, all females, in the British Museum. All the specimens were from a single specimen of Lates niloticus subsp. longispinus Worthington from Lake Rudolf, collected by Dr. E. B. Worthington in 1931. The parasites were embedded in the head and flanks of the fish. The length of the holotype is 19 mm., with the five pairs of limbs in positions 6-3, 14, 36, 64, and 77 per cent, of the body length from the anterior end. Owing to the fact that the head is held ventrally between the ventral arms of the anchor, these measurements have been made from the most anterior part of the body, i.e. the central boss of the anchor. In paratype, Reg. No. 1949.8.14.29 (Fig. 66) the total length is 22 mm. with the legs in positions 5, n, 32, 59, and 71 per cent, of the body length. The body is long and slender with a conspicuous swelling in the region of legs 2, and from this swelling a pair of rounded processes project ventrally with the second pair of legs between them. There are slight swellings in the regions of legs 3 and 4. Tw r o examples will suffice to show how the torsion varies and may change its direction in this species. In the holotype the total torsion is a sinistral one of 110, made up of a dextral torsion of 10 between legs 2 and 3 and a sinistral torsion between legs 3 and 4. In paratype Reg. No. 1949.8.14.29 the total torsion is a dextral one of 20 ; this is the resultant of a sinistral torsion of 45 between legs I and 2, and of 135 between legs 2 and 3, followed by a dextral torsion of 90 between legs 3 and 4, and of 110 between legs 4 and 5. The abdomen is very long, about a quarter of the total body length ; it is in line with the rest of the body and tapers gradually to a rounded tip without any indica- tions of segmentation. The pregenital prominence is ill defined. The anchor has normally four simple more or less cylindrical arms as shown in Fig. 65. One of the ventral arms of specimen Reg. No. 1949.8.14.29 is branched, but this is evidently an abnormality probably caused by its meeting an obstruction during its growth into the flesh of the fish (Fig. 66). The ventral arms are about 5 times as long as they are broad and are directed backwards at an angle of about 45 to the body. The dorsal arms are a little shorter and are directed more nearly at right angles to the body. The head is placed on the ventral side of the anchor in the fork between the ventral pair of arms. None of the specimens had complete egg-sacs ; the most complete was 3 mm. long, with a maximum width of 0-6 mm. Lernaea haplocephala (Cunnington) 1914 Lernaeocera haplocephala Cunnington, Proc. Zool. Soc. Lond. 1914: 826, pi. i, figs. 4-7, text-fig, i B. 1917 Lernaea haplocephala: Wilson, Proc. U.S. Nat. Mus. 53: 38. 1918 : Wilson, Bull. U.S. Bur. Fish. 35: 193, 195, pi. 15, fig. 92. 1920 : Wilson, Bull. Amer. Mus. Nat. Hist. 43 (i): 5, pi. 3, figs. 20-22. ON SOME SPECIES OF LERNAEA 19 1923 Lernaeocera bichiri Kurtz, S. B. Akad. Wiss. Wien. 131, Abt. i: 332, pi. 2, figs. i-n. 1927 Lernaea haplocephala: Brian, Faune Colon. Fr. 1: 581, figs. 26-34. 1944 ,, ,, : Capart, Bull. Mus. Hist. nat. Belg. 20 (24) : 7. The British Museum possesses the twenty-seven specimens listed by Cunnington, from three species of Polypterus from Lake Tanganyika and the White Nile. I select as holotype the single specimen, Reg. No. 1914.12.2.3, taken from Polypterus congicus Boulenger collected from Lake Tanganyika by Cunnington himself and on which his description is largely based. L. haplocephala is probably the best known of the African species of Lernaea and has been found on several species of Polypterus in the White Nile, Belgian Congo, and Cameroons. The species is easily recognized by the shape of the anchor and by the peculiar swelling in the region of legs 2 ; and it is unfortunate that Wilson (1920), in his eager- ness to find characters in the appendages, should have added as a distinguishing character 'the small spherical terminal joint of the maxillipedes, with its four curved claws'. I have examined the maxillipedes of the holotype and of some of the para- types and can find no distinguishing feature in them ; they have five claws like every other species I have looked at. Wilson may have had a specimen with only four claws Brian describes and figures only three, but they are not easy to see and are difficult to count. There is, however, a character in which the appendages of L. haplocephala differ from those of all other species of Lernaea that I have been able to examine: there are only four setae on the terminal joints of the exopods of the first four pairs of legs (Table 2) ; other species have five setae here. The setation of these legs has been correctly figured by Kurtz and by Brian. TABLE 2 Arrangement of the setae and spines on the legs of Lernaea haplocephala Leg i Leg 2 Leg 3 Leg 4 . { spines Exopod ( s tae I.Z.3 1. 1.4 1.1.3 1. 1. 4 1.1.3 1. 1. 4 1.1.3 1. 1. 4 , \ spines End P d ( setae 0.0.2 I.I.4 0.0.2 1.2-4 O.O.2 1.2.4 0.0.2 1.2.3 Lernaea lophiara sp. nov. FIGS. 69-80 Holotype, Reg. No. 1949.8.14.34, and several paratypes, all females, in the British Museum. The holotype was from the dorsal fin of Lethrinops lethrinus (Giinther) from Lake Nyasa. The paratypes included very similar specimens from the dorsal fins of the following species of fish, all from Lake Nyasa: Haplochromis prostoma Trewavas, H. sp. cf. micrentodon Regan, Rhamphochromis lucius Ahl, Pseudotropheus tropheops Regan, Diplotaxodon argenteus Trewavas, and also buried in the edge of the operculum of Lethrinops praeorbitalis Regan. Other paratypes which differ from the holotype only in having very short arms to the anchor were found in the dorsal fins of Haplo- chromis breviceps Regan and Tilapia melanopleura Dumeril. Specimens which I have left in situ and not examined but are presumably the same species were found in 20 ON SOME SPECIES OF LERNAEA the dorsal fins of Haplochromis argyrosoma Regan, H. incola Trewavas, H. johnstoni (Giinther), and H. nigritaeniatus Trewavas. The length of the holotype is 9-6 mm. with the legs in positions 6, 16, 42, 76, and 93 per cent, of the body length from the anterior end. The body is curved between legs 2 and 3, so that the axis of the anterior end of the body is approximately at FIGS. 69-80. Lernaea lophiara sp. nov. Fig. 69, lateral view of head and anchor of holotype from dorsal fin of Lethrinops; Fig. 70, anterior view of the same; Fig. 71, paratype Reg No. 1949.8.14.35 from dorsal fin of Haplochromis; Fig. 72, lateral view of anchor; Fig. 73, ventra iview of the same; Fig. 74, lateral view of abdomen and pregenital prominences of the same para- type; Fig. 75, ventral view of the same; Fig. 76, anterior view of anchor of paratype Reg. No. 1949.8.14.45 from operculum of Lethrinops; Fig. 77, general view of this paratype; Fig. 78, posterior end with egg-sacs of paratype Reg. No. 1949.8.14.46 from operculum of Lethrinops; Fig. 79, ventral view of abdomen and pregenital prominences of paratype Reg. No. 1949.8.14.45; Fig. 80, lateral view of the same. FIGS. 81-83. Lernaea cf. lophiara Fig. 81, specimen from operculum of Rhamphochromis Reg. No. 1949.8.14.47; F'ig. 82, ventral view of abdomen and pregenital prominences of this specimen; Fig. 83, lateral view of the same. The line near each figure is i mm. drawn to the same scale. right angles to the posterior end. The total torsion of the holotype is a dextral one of about 145 ; this is made up of a dextral torsion between legs 2 and 3 of 85 and a further dextral torsion of 60 between legs 3 and 4. The lengths of seven paratypes range from 6-7 mm. to 9-8 mm., the positions of the five pairs of legs ranging from 5-6-5, 17-18, 42-47, 73-79, and 93-94 per cent, of the body length from the anterior end respectively. Only in two specimens, one of ON SOME SPECIES OF LERNAEA 21 them the holotype, is the torsion of the body more than 90. There is nearly always some torsion, however, and also a curvature in the region of legs 2 and 3. The anchor has four simple arms. In most of the specimens (Figs. 69, 70, 76, 77), including the holotype from the dorsal fin of Lethrinops praeorbitalis and paratype Reg. No. 1949.8.14.45 from the operculum of the same fish, the dorsal arms are a little longer than the ventral ones and tend to splay outwards towards the ends. In two other specimens (Figs. 71-73) from dorsal fins, one from Haplochromis breviceps, the other from Tilapia melanopleura, the arms of the anchor are very short. The pregenital prominence is double in all specimens except one which is the smallest examined and measures 6-7 mm. in length. The setation of the legs is the same as that of L. cyprinacea (Table i, p. 4). The egg-sacs are spindle-shaped and two or three times as long as the abdomen. Lernaea sp. cf. lophiara FIGS. 81-83 A fish of the species Rhamphochromis lucius Ahl bore three parasites : one on the fin is a typical example of Lernaea lophiara, the other two, one on the flank and the other on the operculum, are considerably larger than normal for that species and have very much larger arms to the anchor. The shape of the abdomen is also rather different. It may be found that these two specimens belong to a new species, but the comparatively well-developed condition of the arms of the anchor might be due to there being more space for them to grow in the body of the fish than there is in the fin. Against this, however, is the fact that the specimens of L. lophiara from the operculum of Lethrinops praeorbitalis do not differ in the size of the anchor or in other respects from specimens from the fins. The lengths of the two specimens from the flank and operculum are 14-4 mm. and 13-7 mm. respectively. The anchor of the shorter specimen is damaged, that of the larger is shown in Fig. 81. The last segment of the abdomen is very small in these two specimens, the abdomen as a result being much more conical in shape than is typical for L. lophiara (Figs. 82-83). The positions of the five pairs of legs of the larger specimen are 5, 17, 46, 73, and 94 per cent, of the body length from the anterior end respectively. The setation of the legs is the same as that of the other specimens. Lernaea bistricornis sp. nov. FIGS. 84-88 Holotype Reg. No. 1949.8.14.49 in the British Museum. This species has to be described from a single specimen found at the base of a pelvic fin of Cardio-pharynx schoutedeni Poll from Lake Tanganyika. The length of the holotype is 87 mm. The body is curved evenly into a semicircle, and there is a sinistral torsion of about 90. The positions of the five pairs of legs are 8, 21, 45, 76, and 94 per cent, of the body length from the anterior end respectively. The setation of legs i and 2 cannot be seen, but that of legs 3 and 4 is the same as that of L. cyprinacea (Table i, p. 4). ON SOME SPECIES OF LERNAEA The anchor (Figs. 84-86) has six short, blunt processes, three on each side. There is a dorsal pair and a ventral pair, both of which are directed outwards and back- wards. These are similar to those of some specimens of L. lophiara ; but in addition FIGS. 84-88. Lernaea bistricornis sp. nov. Fig. 84, latera Iview of anchor; Fig. 85, ventral view; Fig. 86, anterior view of the same; Fig. 87, ventral view of abdomen and pregenital prominence; Fig. 88, lateral view of the same. FIG. 89. Lernaea barbicola Leigh-Sharpe. Abdomen, pregenital prominence, and egg-sac. FIGS. 9095. Lernaea tuber osa sp. nov. Fig. 90, ventral view of holotype; Fig. 91, abdomen and pregenital prominence; Fig. 92, lateral view of anchor; Fig. 93, the same seen as a transparent object; Fig. 94, anterior view of anchor; Fig. 95, the visible part of the parasite protruding from the hole in the side of the fish. The line near each figure is i mm. drawn to the same scale. there is a pair of small knobs on each side of the head which reach over as if to protect it. The pregenital prominence is well denned (Figs. 87-88), but appears to be simple. The abdomen is of normal length, rather tapering, and without signs of segmentation. The egg-sacs are spindle-shaped and about 1-3 mm. long. ON SOME SPECIES OF LERNAEA 23 Lernaea barbicola Leigh-Sharpe FIG. 89 1930 Lernaea (Lernaeocera) barbicola Leigh-Sharpe. Parasitology 22: 334, text-figs. 1-6. Mr. Leigh-Sharpe has kindly presented the holotype Reg. No. 1949.8.14.50 of this species mounted on a microscope slide to the Museum. It was from the tail of a species of Barbus from the Transvaal. Unfortunately the arms of the anchor have been broken and it is no longer possible to see their arrangement. Owing to the fact that the specimen is flattened on a slide the precise shape of the abdomen and pregenital prominence must remain uncertain; but I have made a camera lucida drawing (Fig. 89) which I hope is a little more accurate than Leigh- Sharpe's Fig. 2. Leigh-Sharpe's figures of the first and second pairs of legs are evidently not intended to show the precise setation of these limbs. All four pairs are visible in the preparation, and with the help of an immersion lens I have been able to make out the setation, which is precisely the same as that of L. cyprinacea (Table I, p. 4). Lernaea tuber osa sp. nov. FIGS. 90-95 Holotype, Reg. No. 194.9.8.14.51, and one paratype in the British Museum. Both specimens were from the body of the fish Engmulicypris sardella (Giinther) from Lake Nyasa. The holotype was from the flank of a specimen collected by Dr. Christy in 1925, and the paratype was from the mid-ventral line of a fish in the Museum collection, Reg. No. 1908.10.27.24-33 collected by Captain Rhoades. The length of both specimens is 11-5 mm. The positions of the five pairs of legs are 7-8, 18, 42, 72, and 91 per cent, in the holotype and 7-8, 18, 44, 76, and 93 per cent, of the body length from the anterior end in the paratype. The total torsion in both specimens is 100 in a dextral sense. In the holotype this is the result of dextral torsions between legs i and 2 of 45, between legs 2 and 3 of 40, and between legs 3 and 4 of 15. In the paratype the torsion is at first sinistral through 45 between legs i and 2, followed by dextral torsions of 105 between legs 2 and 3, 35 between legs 3 and 4, and about 5 behind legs 4. The neck of both the specimens, that is the part of the body from the anchor to nearly the third legs, is covered with little peg-like processes which immediately distinguish this species from any other known at present, and which have suggested to me the trivial name tuber osa. The anchor has four arms arranged as shown in Figs. 90 and 92-94 ; each bears a number of small finger-like processes. In spite of the apparent irregularity there is a distinct bilateral symmetry in the arrangement of the processes, and the two speci- mens are very similar to one another. The pregenital prominence is distinct but single or only indistinctly bilobed. The abdomen is of normal length and tapering, with the segments not marked off from one another. The egg-sacs of the holotype (Fig. 95) were small and rather shrunk in appearance. 24 ON SOME SPECIES OF LERNAEA KEY TO ADULT FEMALES In the following key to the species of Lernaea those which I have seen are printed in bold type and those I know only from descriptions or figures are in italics. The following species are omitted as I consider them to be synonyms : L. bichiri (Kurtz, 1922) = L. haplocephala (Cunnington, 1914) L. carassii Tidd, 1933 = L. cyprinacea Linnaeus, 1758 L. elegans Leigh-Sharpe, 1925 = L. cyprinacea Linnaeus, 1758 L. pectoralis (Kellicott, 1882) = L. catostomi (Kr0yer, 1864) L. temnocephala (Cunnington, 1914) = L. hamiTnifl.na (Hartmann, 1865) L. tortua (Kellicott, 1881) = L. catostomi (Kr0yer, 1864) L. werneri (Kurtz, 1922) = L. composita Wilson, 1924 1. Neck with many peg-like protuberances. . . L. tuberosa sp. nov. Neck smooth ........... 2 2. Anchor with four unbranched arms, confluent at their bases ... 3 Anchor with some other arrangement of its arms . . . ... 12 3. A localized swelling at least twice the width of the body present in the region of legs 2 . . . . . . . . . -4 Body not conspicuously swollen in region of legs 2 . . . . .5 4. Abdomen of normal length, less than three times its breadth . L. haplocephala (Cunnington, 1914) Abdomen very long, about a quarter of the total body length L. longa sp. nov. 5. Arms of anchor long and straight and in a plane roughly parallel to body axis 6 Arms not answering to this description ...... 7 6. Pregenital prominence bilobed ..... L. tilapiae sp. nov. Pregenital prominence with three lobes . L. pomatoides (Kr0yer, 1864) 7. Anchor with dorsal arms curved and ventral arms straight L. bagri sp. nov. Anchor not answering to this description . . . . . .8 8. Abdomen short, little, if any, longer than pregenital prominence . . 9 Abdomen distinctly longer than pregenital prominence . . . .10 9. Dorsal and ventral arms of anchor of about equal size .... L. cruciata (Lesueur, 1824) Ventral arms much smaller than dorsal arms . L. tennis (Wilson, 1916) 10. Pregenital prominence bilobed .... L. lophiara sp. nov. Pregenital prominence simple . . . . . . . .11 11. Arms not tapering, each with a kink near the end . . L. palati sp. nov. Arms thick at base and tapering rapidly . . L. composita Wilson, 1924 12. Anchor of four simple flattened arms, an anterior pair in front of legs i and a posterior pair behind these legs . . L. variabilis (Wilson, 1916) Anchor not answering to this description . . . . . 13 13. Main bulk of anchor at right angles to the body like the cross-bar of a T, with the length of cross-bar at least a third of body length . . 14 Anchor not answering to this description . . . . . 17 14. Anchor with a median dorsal process bifid at the tip .... L. dolabroides Wilson, 1918 ON SOME SPECIES OF LERNAEA 25 Anchor with no median dorsal process . . . . . . 15 15. Lateral arms of anchor unbranched . . L. parasiluri Yamaguti, 1939 Lateral arms of anchor each with a dorsal branch near the end . .16 16. Anchor with a small pair of ventral arms near middle line, body without a conspicuous constriction . . . . . . L. piscinae sp. nov. Anchor without ventral arms. Body with conspicuous constriction between legs 3 and legs 4 . . L. diceracephala (Cunnington, 1914) 17. Anchor with six short, rounded protuberances, three on each side, a dorsal pair, a ventral pair, and also an anterior pair at the sides of the head L. bistricornis sp. nov. Anchor not answering to this description . . . . . .18 18. Anchor set at right angles to body by a ventral flexure by legs 2, arms lateral with bulbous branches. Posterior part of body much swollen . L. insolens Wilson, 1919 Not answering to this description . . . . . . .19 19. Anchor with a median dorsal arm which may be branched, and lateral arms . 20 Anchor with arms in pairs, no median arm . . . . . .22 20. Dorsal arm twice bifid, posterior half of body behind legs 4 swollen and spindle-shaped . . . . . . L. lagenula (Heller, 1865) Dorsal arm unbranched, or branched only once . . . . .21 21. Lateral arms simple .... L. barbicola Leigh-Sharpe, 1930 Lateral arms branched at least once . . . L. catostomi (Kr0yer, 1864) 22. Arms of both dorsal and ventral pairs bifid . . L. oryzophila Monod, 1932 Either dorsal or ventral arms unbranched . . . . . -23 23. Ventral arms simple, dorsal arms branched . . . . . .24 Ventral arms branched, dorsal arms unbranched . . . . .28 24. Ventral arms very short, hardly longer than breadth of head . L. barnimiana (Hartmann, 1865) Ventral arms distinctly longer than breadth of head . . . -25 25. Ventral arms curved outwards, with a small swelling facing anteriorly about the middle of the curve ...... L. barilii sp. nov. Ventral arms more or less straight, without a swelling . . k .26 26. Legs 2 as well as legs i between the bases of ventral arms. Abdomen in line with the body . . . . L. ranae Stunkard & Cable, 1931 Legs 2 situated some distance behind the bases of ventral arms. Abdomen generally at an angle with body ....... 27 27. Arms not more than three times as long as they are broad. Dorsal arms nearly as short as ventral ones. Egg-sacs oval L. esoscina (Burmeister, 1835) Arms slender and cylindrical in form. Dorsal arms distinctly longer than ventral. Egg-sacs spindle-shaped . . L. cyprinacea Linnaeus, 1758 28. Branches of ventral arms unequal, main branch directed outwards and smaller one directed ventrally from it. Pregenital prominence bilobed . L. phoxinacea (Kr0yer, 1864) Ventral arms bifid at tip with resultant prongs equal and parallel. Pregenital prominence hemispherical. . . L. senegali Zimmermann, 1923 26 ON SOME SPECIES OF LERNAEA REFERENCES BAIRD, W. 1850. The Natural History of the British Entomostraca. Ray Society, London. viii + 364 pp., 36 pis. BARBUT, J. 1783. The Genera Vermium. . . . London, xx+ioi pp., n pis. BLAINVILLE, H. M. D. de. 1822. Memoire sur les Lernees (Lernaea, Linn.). Journ. Phys. 95: 372-380. BRIAN, A. 1927. Crustacea II. Copepoda parasitica. Faune Colon. Franc. 1: 571-587, figs. 1-34. 1940. Sopra una specie di Copepodo parassita raccolto dal Prof. Parenzan nel lago Ararobi nell' A.O.I. Lernaea temnocephala (Cunnington) . Boll. Idrobiol. Caccia Pesca, 1: 50-56, text-figs. A-F. BURMEISTER, H. 1835. Beschreibung einiger neuen oder weniger bekannten Schmarotzer- krebse. . . . Nova Acta Leap. Carol. 17: 269-336, pis. 23, 24, 24 A, 25. CAPART, A. 1944. Copepodes parasites des Poissons d'eau douce du Congo Beige. Bull. Mus. Hist. nat. Belg. 20 (24) : 1-24, text-figs. 1-4. CUNNINGTON, W. A. 1914. Zoological results of the third Tanganyika Expedition, conducted by Dr. W. A. Cunnington, 1904-1905. Report on the Parasitic Eucopepoda. Proc. Zool. Soc. Lond. 1914: 819-829, pi. i, text-fig, i. GURNEY, R. 1933. British Fresh-Water Copepoda, 3. Ray Society, London. xxix+3&4 pp., text-figs. 1196-2061. HARTMANN, R. 1865-1866. Naturgeschichtlich-medicinische Skizze der Nillander. Berlin. pp. vii, 419. 2 abt. (abt. 2, pp. 209-419, 1866.) 1870. Beitrage zur anatomischen Kenntniss der Schmarotzer-Krebse. Arch. Anat. Phys. Wiss. Med. 1870: 726-752, pis. 17-18. 1871. Uber das von Poren durchsetzte aussere Chitinskelet des Caliopus, Cecrops, und gewisser Lernaeoceren. S. B. Ges. naturf. Fr. Berl. 1870: 60-61. HELLER, C. (1865). Crustacea. Reise der osterreichischen Fregatte Novara um die Erde in den Jahren 1857, 1858, 1859. Zool. Theil. 2 (8) : 1-280, pis. 1-25. HOFER, B. 1904. Handbuch der Fischkrankheiten. Stuttgart, xv + 359 pp., 18 pis., 222 text- figs. KELLICOTT, D. S. 1881. Lerneocera tortua n.s. Proc. Amer. Soc. Micr. 3rd. Ann. Meeting, Detroit 1880: 41-43, pi., figs. 1-3. 1882. On certain crustaceous parasites of fresh- water fishes. Proc. Amer. Soc. Micr. 5th Ann. Meeting, Elmira 1882: 75-78. KROYER, H. 1863-1864. Bidrag til Kundskab om Snyltekrebsene. Naturhist. Tidsskr. Kjoben- havn (3), 2: 75-426, pis. 1-18. KURTZ, H. 1923. Zwei neue Arten von Lernaeocera aus dem Nil. S. B. Akad. Wiss. Wien. (Abt. i) 131: 327-337. p!s- 1-2. LEIGH-SHARPE, W. H. 1925. Lernaea (Lernaeocera) elegans n. sp. A parasitic Copepod of Anguilla japonica. Parasitology , 17: 245-251, text-figs. 1-5. 1930. Lernaea (Lernaeocera) barbicola n.sp. A parasitic Copepod of Barbus sp. from the Transvaal. Parasitology, 22: 334-337, text-figs. 1-6. LESEUER, C. A. 1824. On three new species of parasitic vermes belonging to the Linnean genus Lernaea. J. Acad. Nat. Sci. Philad. 3: 286-293, pi. n. LINNAEUS, C. 1746. Fauna Svecica. Stockholmiae. xxvi+4ii pp., 2 pis. 1758. Sy sterna Naturae. Ed. X. Tom. 1, Holmiae. 824 pp. MARKEWITSCH, A. P. 1934. Die Schmarotzerkrebse der Fische der Ukraine. Ann. Mus. zool. polon. 10: 223-249, pis. 44-45. I937- Copepoda parasitica der Binnengewdsser der U.S.S.R. (Akad. Wiss. Ukr. S.S.R.) Kiew, 222 pp., 27 pis, 10 text-figs. [Ukranian, Germ. Summary.] MATSUI, Y., and KUMADA, A. 1928. ' Ikari-Mushi ' (Lernaea elegans Leigh-Sharpe), a new para- sitic Copepod of Japanese Eel. /. Fish. Inst. Tokyo, 23: 101-107, pis. 5-7. MONOD, T. 1932. Contribution a 1'etude de quelques Copepodes parasites de Poissons. Ann. Parasit. hum. comp. Paris, 10: 345-380, text-figs. 1-23. ON SOME SPECIES OF LERNAEA 27 NAKAI, N. 1927. On the development of a parasitic copepod, Lernaea elegans Leigh-Sharpe, infesting on Cyprinus carpio L. /. Fish. Inst. Tokyo, 22 : 39-59, pis. 2-4, text-figs. 1-7. NERESHEIMER, E. 1909. Die parasitischen Copepoden. In Brauer, Die Siisswasserfauna Deutschlands, 11: 70-84, text-figs. 311-345. OKADA, Y. K. 1927. Copepode parasite des Amphibiens. Nouveau parasitisme de Lernaea cyprinacea L. Annot. zool. Jap. 11: 185-187, text-figs. 1-2. PESTA, O. 1934. Krebstiere oder Crustacea. I. Ruderfiisser oder Copepoda. Dahl, Die Tierwelt Deutschl. 29: 1-68, text-figs. 1-42. SCOTT, T. and A. 1913. The British Parasitic Copepoda. Ray Society, London, ix+256 pp., 2 pis. (Atlas: xii pp., 72 pis.) STUNKARD, H. W. and CABLE, R. M. 1931. Notes on a species of Lernaea parasitic in the larvae of Rana clamitans. J. Parasit. 18: 92-97. pi- 8. TIDD, W. M. 1933. A new species of Lernaea (Parasitic Copepoda) from the Goldfish. Ohio J. Sci. 33: 465-468, pi. i. WAGLER, E. 1937. Crustacea (Krebstiere). Die Tierw. Mitteleuropas, 2, 2a: 3-224, text-figs. 1-624. WILSON, C. B. 1916. Copepod parasites of fresh-water fishes and their economic relations to mussel glochidia. Bull. U.S. Bur. Fish. 34 [for 1914]: 331-374, pis. 60-74. 1917. North American Parasitic Copepods belonging to the Lernaeidae with a revision of the entire Family. Proc. U.S. Nat. Mus. 53: 1-150, pis. 1-21. 1918. The economic relations, anatomy, and life history of the genus Lernaea. Bull. U.S. Bur. Fish. Washington, 35 [for 1915-1916]: 163-198, pis. 6-15. 1919. A new species of parasitic copepod, with notes on species already described. Proc. U.S. Nat. Mus. 55: 313-316, pi. 21. 1920. Parasitic Copepods from the Congo Basin. Bull. Amer. Mus. Nat. Hist. 43, i : 1-8, pis. 1-3. 1924. Parasitic copepods from the White Nile and the Red Sea. Res. Swed. Zool. Exped. Egypt 6- White Nile, igoo-igoi. Pt. 5(3) : 1-17, pis. 1-3. YAMAGUTI, S. 1939. Parasitic Copepods from Fishes of Japan. Pt. 5, Caligoida, III. Vol. Jubil. Prof. S. Yoshida, Osaka, 2: 443-487, pis. 14-33. ZIMMERMANN, F. 1923. Bearbeitung der parasitischen Copepoden von Fischen. Denkschr. Akad. Wiss. Wien. Math. Nat. Klasse, 98: loi-ni, pis. 1-2, text-figs. 1-2. PRESENTED 1 V JAN WbU AXISH3AINn 3HX OX H3XNIJIJ A3XV9 S31HVHD AS dHOixo SS3Hd AXISHSAIN 3HX XV NIVXIH9 XV3HO MI I I ON A GIANT SQUID OMMASTREPHES CAROLI Furtado STRANDED AT LOOE CORNWALL W. J. REES BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 2 LONDON : 1950 ON A GIANT SQUID OMMASTREPHES CAROLI Furtado STRANDED AT LOOE ... CORNWALL BY W. J. REES, D.Sc. Pp. 29-42; Pis. 1-2; 12 text-figures; 3 wdps m BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 2 LONDON : 1950 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is to be issued in five series, corresponding to the Departments of the Museum. Parts will appear at irregular intervals as they be- come ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. i, No. 2, of the Zoological series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued March 1950 Price Four shillings ON A GIANT SQUID, OMMASTREPHES CAROLI Furtado STRANDED AT LOOE, CORNWALL By w. j. REES, D.SC. THE object of this note is to place on record some details of a female specimen of Ommastrephes caroli Furtado 1 stranded in live condition at Looe, Cornwall, in November 1940. It was acquired by the Plymouth Laboratory and was photographed before preservation by Mr. D. P. Wilson, to whom I am indebted for the excellent photographs. Subsequently it was preserved in formalin at the Laboratory, where I was able to examine it by kind permission of Mr. F. S. Russell, F.R.S. The earliest certain record of a stranding of this species, near Scheveningen in Holland in 1661, is mentioned by Steenstrup (1887), and in the same year the species was described for the first time by Furtado from Portuguese specimens in the Lisbon Museum. It was subsequently reported from the Faroes by Lonnberg (1897), and since that date there has been a number of records all strandings from British waters and one from Heligoland ; these are summarized by Clarke & Robson (1929) and, more recently, by Stephen (1944). Apart from these positive records, there are occasional reports of strandings unsupported by details, and probably also strandings on lonely coasts which are never reported, so that the number of actual strandings is possibly much more frequent than indicated in the literature. It is curious that this species is known only from strandings and that all the known specimens are females. 0. caroli most nearly resembles 0. bartrami (Lesueur, 1821), from which it can be readily distinguished by the remarkable membranes of the third arms this feature being absent in 0. bartrami and 0. pteropus. Robson (1925) described the largest example of 0. caroli yet found from a stranding at Withernsea, Yorkshire, and although the Looe specimen is a little smaller it is larger than all the others that have been measured. The standard measurements of 0. caroli from Looe are given below: Measurements in mm. Overall length (apex to tip of right tentacle) .... 1,860 Total length (i.e. including 3rd arm) ..... 1,225 Dorsal mantle length ........ 670 Ventral mantle length ....... 650 Maximum mantle width (excluding fins) .... 245 Maximum mantle width (including fins) .... 570 Width at mantle openings ....... 205 Length of head ......... 170 Interocular width . . . . . . .170 Thickness of head 100 Arm length : Right Left ist ... 360 355 2nd ...... 3rd 4th .... Tentacle length .... Tentacle, length of sucker-bearing surface 415 415 415 400 445 445 1,100 1,300 463 460 1 I have followed Winckworth (1932) in referring this species to Ommastrephes although most authors have recorded the species under the name Sthenoteuthis caroli. 32 ON A GIANT SQUID, OMMASTREPHES CAROLI Furtado The Looe specimen agrees well with Robson's Withernsea example as regards colour and most external features and I have omitted further reference to them. I have, however, thought it desirable to redescribe the tentacles, arms, and suckers in some detail. The first pair of arms are quadrate in section and carry 25-26 pairs of suckers in oblique pairs on ridges. The proximal six pairs are well spaced, then distally, the remaining pairs are set closer together and give the appearance of being alternate. Suckers in the first or proximal row have a diameter of less than 10 mm. Those of the second to the eleventh rows are 10 mm. or over in diameter, while those of rows 12-26 gradually decrease in size down to i mm. in diameter. On the right arm the largest suckers (on the fifth row) have a diameter of 13 mm. The left arm is very similar, with suckers of 14 mm. diameter in the fourth row. Both second arms are strongly keeled along their whole length and there are twenty- seven rows of suckers beginning with medium-sized proximal ones of 9 mm. in diameter. Distally there is a gradual increase in sucker width to 20 mm. in the eighth row, followed by an abrupt reduction to 12-15 mm. in the ninth row. The third arms have about twenty-eight pairs of suckers with similar appearance to those of the second arms. The proximal suckers are only 8 mm. in width, with a gradual increase distally to 13 mm. in the ninth row, followed by a gradual decrease. There is a well-developed keel which is much enlarged not far from the tip of the tentacle to form a strong crest. This is 70 mm. deep opposite the twenty-third and twenty-fourth rows of suckers. The lateral membrane, too, is very well developed and has a distinctive and characteristic shape at least in the female, for the male is unknown. It extends from the base of the arm to within 60 mm. of its tip. The membrane is greatly enlarged distally to form a large, thin flap of a curious shape (Pis. i & 2). In the left arm this has a width of 220 mm., while in the right arm it is rather torn and is estimated to have a width in excess of 160 mm. Robson (1925) has discussed the shape of this organ in relation to the differentiation of species, but it is apparent, even in this fine specimen, that little reliance can be placed on it for taxonomic purposes because of its fragile nature. The right and left ventral arms have thirty-six and thirty-four pairs of suckers respectively ; these are widely spaced on the flat, sucker-bearing face of the tentacle. On the right arm the proximal suckers have a diameter of 7 mm., and there is a gradual increase in size to 14 mm. in the seventh row. Large suckers of 12-14 mm. diameter are maintained to the tenth pair, after which there is a gradual decrease down to i mm. or less at the tip of the arm. The left arm is similar, with larger suckers of 15-16 mm. diameter in the sixth to ninth rows. The right and left tentacles respectively are 1-65 and 1-94 times the length of the dorsal mantle. The following description applies to the right tentacle. It can be conveniently differentiated into four regions to facilitate description: viz. the tip portion, the large sucker region, the locking-apparatus region, and the proximal portion devoid of suckers. The tip portion, 87 mm. long, carries oblique rows of four suckers each at the extreme tip ; these are small with a diameter of i mm. Proximally these become enlarged to 5-6 mm. diameter with only three in a row. STRANDED AT LOOK, CORNWALL 33 In the large sucker region of the manus there are eleven rows of suckers with four to each oblique row. The two median ones in each row are much enlarged, reaching a maximum size of 17-21 mm. ; those of the first and second row adjoining the tip portion are slightly smaller with diameters of 10 and 13 mm. respectively. On each side, flanking the median suckers, are smaller, long-stalked suckers of about 8 mm. diameter. These are borne on the transverse ridges. The locking-apparatus region (carpus) has three tubercules alternating with three smooth-ringed suckers and is similar in arrangement to that figured by Goodrich (1892) for 0. pteropus. These smooth carpal suckers are small with a diameter of only 3 mm. The ordinary suckers of this region, counting from the most distal tubercule, are twelve in number and diminish in size down to 5 mm. proximally. On the sucker-less part of the tentacle there are fourteen transverse ridges which become fainter and disappear towards the base. The tentacle is keeled along its dorsal surface and becomes slightly finned in the part corresponding to the distal half of the large sucker region and the proximal half of the tip portion. There are narrow, undulating fins along both sides of the sucker- bearing face. Proximally the fin on the dorsal edge is less prominent but persists as a thin ridge as far as the end of the transverse ridges. The ventral fin reaches only to the tenth transverse ridge (from the base) . Sucker rings of this species have been figured by Furtado and by Lonnberg, but unfortunately those of Furtado are not very clear and Lonnberg has failed to indicate the precise position of the suckers on the arms and tentacles. As Robson (1925) has pointed out, the dentition of the rings varies according to their position, the proximal teeth of the arm suckers being lost towards the free end of the arm. The earlier figures are therefore of little use for comparison, so new ones have been drawn from known positions on the arms (Figs. 1-3). On the basal portion of the arms the suckers are toothed all round, but the proximal teeth are small and often rudimentary (Fig. 3). Distal sucker rings have lost their proximal teeth and are of the form illustrated in Figs, i and 2. Typically these suckers have seven, long, backwardly directed teeth. The points where the proximal teeth disappear on each arm are fully discussed by Robson (1925). The tentacular sucker rings are dentate all round and also show some variation according to their position (Figs. 4-6). The distal teeth are curved inwards, while the proximal teeth, although often reduced in size, are bent outwards in the same direction as the distal ones ; .thus the teeth of the whole ring are admirably arranged for clawing. Fig. 4 illustrates a ring with twenty-one teeth, whereas that portrayed in Fig. 5 has twenty-three teeth. The larger rings of the manus are typically ommas- trephid in character with four enlarged teeth (one in each quadrant). This, the largest sucker ring of the club, has twenty-seven teeth. The stranding of giant squids of the genera Architeuthis and Ommastrephes on British coasts has aroused much interest during the past twenty years; the signi- ficance of the strandings, especially the preponderance of records along the east coast of Britain, being the subject of speculation by Clarke & Robson (1929), Robson (1933), and Stephen (1944). The known strandings of 0. caroli, 0. pteropus, and Architeuthis spp. are plotted on Id Ib 2a 2b 3a 5b FIGS. 1-3. Sucker rings from the arms in face and oblique views. 1 a & b., ring, 3-6 mm. in diameter, from 2nd left arm, zist row. 2 a & b., ring, 8-5 mm. in diameter, from $rd left arm, i2th row. 3 a & b., ring, 13 mm. in diameter, from 4th left arm, 8th row. 4a 4b 5a 5b 6a 6b FIGS. 4-6. Tentacular sucker-rings in face and oblique views. 4 a & b., medium-sized sucker-ring, 3-75 mm. in diameter, from the tip portion of the tentacle. 5 a & b., ring, 6-3 mm. diameter, from the long-stalked suckers of the manus. 6 a & b., typical ring, 18 mm. in diameter, from the middle of the manus. 36 ON A GIANT SQUID, OMMASTREPHES CAROLI Furtado Maps I-III, and it is at once evident that most of the specimens have come ashore at three places, viz. the Scarborough area, the Dunbar-North Berwick area, and at Buckie. Another feature of the strandings is that all, with the exception of a single record of 0. pteropus at North Berwick in June 1921, have come ashore during the winter months from November to March. Clarke & Robson correlate the strandings on the Yorkshire coast with hydro- graphic conditions which favour stranding, especially if the animal is enfeebled by some cause. They quote Bowman's testimony that a high percentage of drift bottles released in the north are finally stranded on the mid- Yorkshire coast and between Berwick and St. Abb's Head. Architeuthis and Ommastrephes are clearly oceanic species .which occasionally migrate into the North Sea, possibly during the summer months, and are later en- feebled by unfavourable conditions during the winter months. There is as yet no clue as to what these factors are, but it is probable that lack of suitable food, lower salinity (especially near the coast), and temperature fluctuations have an adverse effect. Various Ommastrephids are, as young animals, common in the surface waters of temperate and tropical seas, but so far the habits of the large adults are a matter for speculation. Perhaps the single record of Ommastrephes pteropus (trawled off St. Kilda, at a depth of 180-200 fathoms in September 1925) is an indication of its normal habitat on the edge of the continental slope. Robson (1933) in discussing the distribution of Architeuthis was also inclined to favour this view. If we may judge by the records plotted on Maps I-III, 0. caroli is the most frequent immigrant into the North Sea, while 0. pteropus is just as rare as Architeuthis in British waters. The British records of these giant squids are scattered in the literature, and are, for the sake of completeness, given below. BRITISH RECORDS OF OMMASTREPHES CAROLI 1. 8 Jan. 1911. Briar Dene, Northumberland; Meek & Goddard (1926). Length (including 3rd arm) 3 ft. n in. (1,175 mm.). 2. Feb. 1921. Isle of Skye; Stephen (1944). 3. 3 Jan. 1925. Withernsea, S. Yorkshire; Robson (1925). 4. 7 Jan. 1925. Cullercoats, Northumberland; Meek & Goddard (1926). Length (including 3rd arm) 3 ft. 8 in. (1,118 mm.). 5. 14 Jan. 1927. Buckie, Moray Firth; Stephen (1944). 6. March 1927. N. Berwick; Stephen (1944). 7. 18 March 1927. N. Bay, Scarborough, Yorkshire ; Clarke & Robson (1929). Length 5 ft. 7 in. 8. i Feb. 1928. Scarborough; Clarke & Robson (1929). Length (including 3rd arm) 3 ft. 6 in. 9. Jan. 1929. Buckie, Moray Firth; Stephen (1944). 10 ii. Dec. 1929. N. Berwick; Stephen (1944), 2 specimens. 12. 9 Jan. 1930. Filey, Yorkshire; Clarke (1930) & Stevenson (1935). Length (including 3rd arm) 3 ft. 9 in. 13. 10 Feb. 1930. Isle of Skye; Stephen (1944). 14. Feb. 1930. Isle of Mull; Stephen (1944). 15. March 1930. Dunbar; Stephen (1944). 16-17. 6 Jan. 1931. Dunbar; Stephen (1944), 2 specimens. 18. 22 Dec. 1931. South Sands, Scarborough; Stevenson (1935). Overall length 5 ft. 10 in. MAP I OMMASTREPHES CAROLI GEORGE PHILIP & SON. LTD The London Geographical Institute STRANDINGS OF OMMASTREPHES CAROLI FURTADO ON BRITISH COASTS ZOOL. I. 2 MAP II OMMASTREPHES PTEROPUS * Slrandinds / 6 O OtKer Records Q GtOBGF PHILIP & SON LTD The London Geographical Institute STRANDINGS AND OTHER RECORDS OF OMMASTREPHES PTEROPUS STEENSTRUP IN BRITISH WATERS MAP III ARCHITEUTHIS spp. Stranding O Other Records GEORGE PHILIP & SON. LTD The London Geographical Institute STRANDINGS AND OTHER RECORDS OF ARCHITEUTHIS SPP. IN BRITISH WATERS 4 o ON A GIANT SQUID, OMMASTREPHES CAROLI Furtado 19. 12 Dec. 1932. Buckle, Moray Firth; Stephen (1933). Overall length 6 ft. 2 in. 20. 31 Jan. 1935. South Bay, Scarborough; Clarke & Stevenson (1935). Overall length 5 ft. 21. 13 Feb. 1935. ij miles north of Scarborough; Clarke & Stevenson (1935). Overall length 5 ft. 2 in. 22. 3 Nov. 1935. Castlerock, Co. Londonderry; Stendall (1936). Determined by A. C. Stephen. 23. 24 Nov. 1937. Birsay Parish, Orkney; Stephen (1938). Overall length 5 ft. 24. 18 Dec. 1937. Stronsay, N. Orkney; Stephen (1938). Overall length 5 ft. 8 in. 25. Nov. 1940. Looe, Cornwall (present record). 26. Jan. 1941. Fair Isle, Shetland; Stephen (1944). BRITISH RECORDS OF OMMASTREPHES PTEROPUS STEENSTRUP 1. 19 Nov. 1883. Scarborough; Goodrich (1892). 2. 27 Feb. 1884. 'North Sea'; Goodrich (1892). 3. Jan. 1892. Salcombe, Devon; Goodrich (1892). 4. ? Killala, Co. Mayo; Nichols (1905, 'many years ago'). 5. ? Miltown Malbay, Co. Clare; Nichols (1905, 'a few years ago'). 6. 19 Dec. 1907. Redcar; Hoyle (1908). 7. i Mar. 1912. Redcliff, near Scarborough. Length (including 3rd arm) 3 ft. 8. June 1921. N. Berwick, Firth of Forth; Ritchie (1922). 9. ? Isle of Man ; Robson & Chadwick MS. 10. Sept. 1925. Trawled off St. Kilda in 180-200 fathoms. Overall length 6 ft. (det. Robson). BRITISH RECORDS OF ARCHITEUTHIS SPP. 1. 1673. Dingle Bay, Co. Kerry, S. Ireland; (More, 1875: 4526, as Dinoteuthis proboscideus) . 2. 1860-1861. Between Hills wick and Scallo way, W. Shetland ; (Jeffreys, 1869: 124, as Archi- teuthis monachus). 3. 25 Apr. 1875. Caught at sea off Boffin Island, Connemara, Ireland ; (More, 1875 : 123). 4. Oct. 1880. Stranded at Kilkee, Co. Clare, S. Ireland; (Ritchie, 1918: 137, as Architeuthis) . 5. 1914. In stomach of a sperm whale at Belmullet Whaling Station; (Hamilton, 1915: 137). 6. 2 Nov. 1917. Stranded at Dunbar, Firth of Forth; (Ritchie, 1918: 133, as Architeuthis harveyi). 7. Feb. 1920. Stranded at N. Uist, Outer Hebrides; (Ritchie, 1920: 57, as Architeuthis harveyi) . 8. 1921. Stranded at Caithness, Scotland; (Ritchie, 1922: 423, as Architeuthis harveyi). 9. 14 Jan. 1933. Stranded at Scarborough, Yorkshire; (Robson, 1933, as Architeuthis clarkei n. sp.). 10. 7 Nov. 1937. OS Bell Rock, Angus, E. Scotland; (Stephen, 1937, as Architeuthis harveyi). REFERENCES CLARKE, W. J., & ROBSON, G. C. 1929. Notes on the stranding of giant squids on the north-east coast of England. Proc. Malacol. Soc. Land. 18 : 154-158 ; i text-fig. & STEVENSON, J. A. 1935. Yorkshire Cephalopods. /. Conch. 20: 102. DANIEL, R. J. 1925. A large oigopsid cephalopod. 3gth Ann. Rep. Mar. Biol. Stat. Port Erin, 1925 : 34 ; i text-fig. FURTADO, A. 1 887. Sur une nouvelle espece de C6phalopode appartenant au genre Ommatostrephes. Mem. R. Acad. Lisboa, 6 (2): 1-16; 2 pis., 5 text-figs. GOODRICH, E. S. 1892. Note on a large squid (Ommastrephes pteropus Steenstrup). /. Mar. Biol. Assoc. U.K., N.S., 2: 314-321 ; text-figs. GRIEG, J. A. 1933. Cephalopods from the west coast of Norway. Bergens Mus. Aarb. 1933 (4) : 1-25 ; pis. 1-4, i text-fig. GRIMPE, G. 1925. Zur Kenntnis der Cephalopoden-Fauna der Nordsee. Wiss. Meeresuntersuch. 16 (3): 1-124; J Pi-. 34 text-figs. STRANDED AT LOOE, CORNWALL 41 HAMILTON, J. E. 1915. Belmullet Whaling Station : Report to the Committee. Rep. Brit. Ass. 1914: 125-161; 4 text-figs. HERTLING, H. 1938. Ueber eine auf Juist gestrandete Sthenoteuthis caroli (Furtado). Wiss. Meeresuntersuch. 1: 93111. HOYLE, W. E. 1908. A large squid at Redcar. Naturalist, 615: 132-133; i text-fig. JEFFREYS, J. G. 1869. British Conchology, 5. LONNBERG, E. 1897. Ofversigt ofver Sveriges Cephalopoder. Bih. Svensk. Vetenskakad. Handl. 17 (4, No. 6) : 1-41 ; i pi. MASSY, A. L. 1928. The Cephalopoda of the Irish coast. Proc. R. Irish Acad. 38 (B, No. 2) : 25-37. MEEK, A., & GODDARD, T. R. 1926. On two specimens of giant squid stranded on the Northum- brian coast. Trans. Nat. Hist. Soc., Northumb., N.S., 6 : 229-237. MORE, A. G. 1875 a. Gigantic squid on the west coast of Ireland. Ann. Mag. Nat. Hist. (4), 16: 123-124. 1875 b. Notice of a gigantic cephalopod (Dinoteuthis proboscideus) , which was stranded at Dingle, in Kerry, two hundred years ago. Zoologist, 83 : 4526-4532. NICHOLS, A. R. 1905. On some Irish specimens of a large squid, Sthenoteuthis pteropus (Steen- strup). Irish Nat. 14: 54-57; i text-fig. RITCHIE, J. 1918. Occurrence of a giant squid (Architeuthis) on the Scottish Coast. Scot. Nat. 1918: 133-139- 1920. Giant squid cast ashore N. Uist, Outer Hebrides. Scot. Nat. 1920: 57. 1922. Giant squid on the Scottish coast. Rep. Brit. Ass. 1921 : 423. ROBSON, G. C. 1925. On a specimen of the rare squid, Sthenoteuthis caroli, stranded on the Yorkshire coast. Proc. Zool. Soc. Lond. 1925 : 291-301 ; pi. i ; 5 text-figs. I933- On Architeuthis clarkei, a new species of giant squid, with observations on the genus. Ibid. 1933 : 681-697 ; pi. i, 8 text-figs. STENDALL, J. A. S. 1936. Giant cuttlefish, Sthenoteuthis caroli Furtado, ashore in Co. London- derry. Irish Nat. J. 6 : 23-24. STEPHEN, A. C. 1933. Rare cuttlefish (Sthenoteuthis caroli) washed ashore at Buckie. Scot. Nat. 1933 : 96. 1938. Rare squid in Orkney. Ibid. 1938: 119. 1944. The Cephalopoda of Scottish and adjacent waters. Trans. Roy. Soc. Edinb. 61 : 247-270; 14 text-figs. STEVENSON, J. A. 1935. The Cephalopods of the Yorkshire coast. /. Conch. 20: 104-116; pis. 3-7, i text-fig. VERRILL, A. E. 1879-1881. The Cephalopods of the north-eastern coast of America. Trans. Conn. Acad. Arts. Sci. 5 : 177-257 and 259-446; pis. 13-56. WINCKWORTH, R. 1932. The British Marine Mollusca. /. Conch. 19: 211-252. PRESENTED 1 APR JJ Bull. EM. (N.H.) Zoology, I, 2 PLATE 1 Photo. D. P. Wilson OMMASTREPHES CAROLI ; DORSAL VIEW Bull. B.M. (N.H.) Zoology, I, 2 PLATE 2 Photo. D. P. Wilson OMMASTREPHES CAROLI ; VENTRAL VIEW PRESENT D PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATEY PRINTER TO THE UNIVERSITY 1 APR 1950 OF CAPTAIN COOK'S KANGAROO T. C. S. MORRISON-SCOTT AND F. C. SAWYER BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 3 LONDON : 1950 THE IDENTITY OF CAPTAIN COOK'S KANGAROO BY T. C. S. MORRISON-SCOTT AND F. C. SAWYER Pp. 43-50; Ph. 3-5 BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 3 LONDON: 1950 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is to be issued in five series, corresponding to the Departments of the Museum. Parts will appear at irregular intervals as they be- come ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. i, No. 3, of the Zoological series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued March 7950 Price Three shillings THE IDENTITY OF CAPTAIN COOK'S KANGAROO By T. c. s. MORRISON-SCOTT and F. c. SAWYER INTRODUCTION THE identity of the kangaroo discovered by Captain Cook's expedition in 1770 has lately been the subject of some dispute. For years this kangaroo has been referred to as Macropus giganteus (Zimmermann, 1777) and was thought to have been the Great Grey Kangaroo, until Iredale & Troughton (1925) not only pointed out that giganteus is antedated by Mus canguru Miiller, 1776, which was based on the descrip- tion and plate given in Hawkesworth's (1773) account of Cook's voyage, but also threw doubt on whether Captain Cook's kangaroo was in fact the Great Grey Kangaroo. The ship's company of H.M.S. Endeavour included Sir Joseph Banks who brought with him Dr. Solander as naturalist and Sydney Parkinson as draughtsman. Iredale & Troughton (1925) published a transcript of Solander's manuscript Latin description of the kangaroos obtained by Captain Cook's party at Endeavour River (the future site of Cooktown) in June and July 1770 a description which, as Iredale & Troughton pointed out, does not accord too well with the Great Grey Kangaroo. They sup- ported their contention that the animals in question were not Great Grey Kangaroos with the statement that the latter do not occur at or near Cooktown. Even if this were true the argument would not be valid, since the non-occurrence of the species at Cooktown nowadays does not preclude its possible occurrence there in 1770, when the country was quite undeveloped. But in fact Raven (1939) records that the Great Grey Kangaroo occurred within thirty miles of Cooktown in 1897, and Tate informs us (in litt.} that he obtained three specimens about fifteen miles from Cooktown in 1947. Iredale & Troughton, though satisfied in their own minds that Captain Cook's kangaroo is not conspecific with the Great Grey Kangaroo, were unable to decide its identity but suggested that the weight of evidence pointed to a form of the robustus series. The same authors (1937) next published a paper in which they sought to show that Captain Cook's kangaroo was a northern representative of the Whiptail, or Pretty- face Wallaby usually known as Macropus (Protemnodori) parryi Bennett, but which they hold should be called Wallabia elegans Lambert. This contention rests on rather insecure foundations. Briefly, the argument is that in 1929, or thereabouts, two skins were purchased in the neighbourhood of Cooktown and that Solander's description, they say, agrees with one of these which was a Whiptail the other skin being that of a Wallaroo of the antilopinus type. But it is not at all clear why Cook's kangaroo must necessarily be restricted to one of the two species represented by these two purchased skins, nor is it clear why Iredale & Troughton abandoned their previous conclusions that the weight of evidence pointed to Cook's kangaroo having been a form of robustus. 46 THE IDENTITY OF CAPTAIN COOK'S KANGAROO The next stage in the controversy was a paper by Raven (1939), who holds that the evidence is decidedly against Cook's kangaroo having been a Whiptail, or Pretty-face Wallaby. With this the present writers concur. Raven further holds that the evi- dence supports the view that the early revisers were right in identifying Cook's kangaroo with the Great Grey Kangaroo and pleads the confusion caused by up- setting this position. Finally Tate (1948), in the course of his review of the Macropodidae, dismissed the Whiptail theory of Iredale & Troughton and agreed with Raven that, inter alia, the hip stripe and face stripe of the Whiptail are too diagnostic to have been omitted from the contemporary plate and descriptions of Cook's kangaroo had the animal in fact been a Whiptail. Tate added that the only really large species of Macropodidae that conceivably could have been found near Cooktown are the Great Grey, the Red, and Macropus robustus reginae Schwarz, one of the antilopine group. He dismissed the second and third on grounds of colour and decided that the description and plate in Hawkesworth (1773) and hence Captain Cook's kangaroo and Mus canguru Miiller, 1776 agreed most closely with the Great Grey Kangaroo. Tate avoided discussion of Solander's manuscript description, but since Solander was on board Cook's ship in his capacity as a naturalist, what he has said on the subject of the kangaroos must be examined. Here, however, we are straightway confronted with a difficulty. Iredale & Troughton (1937) say that Solander's description was based upon the small male first captured, and Troughton (1946: 202) repeats the contention, saying that it is indisputable that it applies only to an apparently adult male weighing 38 pounds. But far from being indisputable it is not at all clear why these authors take this view at all, unless it is because the only measurements given are those of the male which Mr. Gore shot on 14 July 1770 (Solander gives the weight of this animal as 24 pounds ; the difference between this and the 38 pounds of the other accounts may be the difference between the ' clean ' and ' dead weight ') . But Solander gives the weights of all three animals taken, and the description itself is clearly a composite one since both male and female genitalia are described and also the mammae, and Solander says that the size of the animal varies with age. Nor is it clear why Trough- ton refers to the 38-pound animal as apparently adult when Solander says that it was possibly two or three years old. Solander's estimate of its years may not be reliable, but he was basing his view that it was not adult on the condition of the molar teeth, as will be seen from his discussion of the latter. But on top of this, Solander may well have had three separate species as well as three separate specimens in front of him as he wrote, and it is not possible to say which animal he had most in mind while describing the various characters. He might well have been making a qualitative average of the characters of all three. So Solan- der is not much help in arriving at the identity of Cook's kangaroo and any deductions drawn from his description should be treated with reserve. With this in mind it can be said that in two particulars Solander's description does not encourage any leanings towards the Great Grey theory. The rhinarium is described as 'Rostrum breviusculum, parum compressum ; apice inter nares nudum ibique cute aterrima rugulosa vestitum'. But the Great Grey Kangaroo has hairy skin between the nostrils. Then again the THE IDENTITY OF CAPTAIN COOK'S KANGAROO 47 upper incisors are described as ' Incisores sex, approximati, lati : primum par leviter bilobum; secundum integrum; tertium lathis crassiusque, bilobum: lobis anticis minoribus '. Iredale & Troughton on the one hand, and Raven on the other, perform some agile juggling with the Latin text in support of their respective theses, but what Solander says is that the third upper pair of incisors are bilobed and that the anterior lobes are the smaller, thus suiting neither the Whiptail theory nor the Great Grey theory. However, as has already been indicated, Solander's description cannot be treated as a reliable guide in the quest for Captain Cook's kangaroo. The controversy has so far been argued in terms of Solander, Hawkesworth, and a skin obtained near Cooktown 160 years after Cook was there. It seems strange that no attempt appears to have been made to find the original specimens, especially as Iredale & Troughton (1925), quoting Hunter (1790), drew attention to the probability of a skull which Banks gave to Hunter being in the Museum of the Royal College of Surgeons. Iredale & Troughton also drew attention to the probability of a pencil drawing by Parkinson being preserved in the British Museum. But they did not pursue these two lines of research on which we now report. DRAWINGS BY SYDNEY PARKINSON AND NATHANIEL DANCE It seems certain that the plate of Captain Cook's kangaroo published by Hawkes- worth (1773) was based on a drawing by Sydney Parkinson, the draughtsman in Banks' s employ on board H.M.S. Endeavour. Search has been made in the British Museum (Natural History), and though the original of Hawkesworth's plate has not been found there are two rough sketches of kangaroos signed 'S. Parkinson' and marked in his hand 'Kangura Endeavour's River'. On the back of one of these Parkinson has added, 'The whole body pale ash colour the ears excepting the base fine specled gray iris of the eye Chestnut '. It was the practice of Parkinson, and other artists who accompanied Cook on his voyages, to make pencil sketches of animals seen, together with notes on the details of coloration, &c., the intention being to paint these in at a later date. In Parkinson's case, due to his death before the end of the voyage, many of the sketches were never completed. These drawings are inade- quate for purposes of identification but we consider them of sufficient interest to warrant publication (PI. 3). Of much greater interest, however, is a wash drawing of a complete kangaroo skull and another of its lower jaw shown separately (PL 4). These are signed 'N. Dance' and are among the collection of Parkinson drawings which came to the British Museum from Sir Joseph Banks's library. Dryander (1748-1810), in his manuscript catalogue of the drawings of animals in Banks's library, has the following entry on page 21 : Mammalia Glires, KANGURU K N.C. S. Parkinson x Cranium Nath. Dance The ' 'is Dryander's symbol for a pencil drawing and the 'x' for a coloured one; ' N.C. ' stands for Nova Cambria, as that part of Australia was called in those days. Sir Nathaniel Dance (1735-1811) was a celebrated portrait painter with a reputa- 48 THE IDENTITY OF CAPTAIN COOK'S KANGAROO tion for accuracy as a draughtsman. Captain Cook sat to him for his portrait in 1776 (fide Kitson, 1907), after which year Dance appears to have given up painting. There is no indication of the scale of the drawing, but by analogy with the series of Parkinson drawings it seems likely that the skull is drawn life-size. The skull and lower jaw are both represented on a single folio sheet. Parkinson's drawings are also on folio sheets and his practice was to draw objects life-size except where they were too big for the paper. In this case he reduced them, but he did not make drawings larger than life-size. This seems to have been the general practice of the time. The point is not pressed, but if Dance's drawing is life-size, then it is likely to be that of the skull of the 84-pound kangaroo shot 1 on 27 July 1770 ; the other two beasts, one shot by Lieutenant Gore on 14 July and another caught by Banks's greyhound on 29 July, were smaller. The skull drawn by Dance appears to be that of a young Macropus robustus. We have been unable to trace the skull itself. ANOTHER OF CAPTAIN COOK'S SPECIMENS John Hunter, in his observations on animals in White's Journal (1790), says: 'Of the Kangaroo . . . the only parts at first brought home were some skins and sculls ; and I was favoured with one of the sculls from Sir Joseph Banks.' The posthumous papers of Hunter (1728-93) edited by Owen (1861) contain the same words, but Owen has added a footnote to the last sentence quoted above, which reads: 'No. 1732 Hunt. Osteol.' Professor Wood- Jones has searched for this skull in the Museum of the Royal College of Surgeons but it cannot be found, and appears to have been destroyed by bombs along with many other Hunterian specimens during the 1939-45 War. How- ever, he drew our attention to a figure of a skull in a paper on the history of surgery by Webb-Johnson (1939). The text to this figure says: 'Kangaroo's skull, from the Hunterian collection brought from Australia ("New Holland") by Sir Joseph Banks when with Captain Cook's Expedition, 1768-71.' The figure itself is a reproduction of a photograph and it shows quite clearly the number ' 3703 ' painted on the skull. Flower's catalogue (1884) makes it plain that No. 3703 is the same specimen as No. 1732 in Owen's catalogue (1853). Webb- Johnson's figure is small and not very clear, but Professor Wood-Jones went to much trouble and eventually found a lantern slide of the same photograph. This slide (PI. 5) is probably one Webb- Johnson had made when he read his paper in 1939. The skull it represents is clearly not the same as the one drawn by Dance and it appears to be slightly younger, an impression which is borne out by the description of its dentition in Owen's catalogue (1853). The skull is from one of the three animals obtained at Endeavour River in 1770 and is probably that of the 38-pound animal shot by Lieutenant Gore on 14 July 1770. As will be seen from Plate 5, the skull is a young one and the incisors are missing. In view of its important bearing on the nomenclature of the genus Macropus, we sent the photograph to Dr. G. H. H. Tate, who has recently (1948) monographed the 1 By Lieutenant Gore, according to the journal of Midshipman John Bootie, who records the weight as 80 pounds. THE IDENTITY OF CAPTAIN COOK'S KANGAROO 49 kangaroos, and himself collected specimens in the neighbourhood of Cooktown. We are indebted to him for his detailed report on this skull which he unhesitatingly refers to the Great Grey Kangaroo amongst other characters the short ante-orbital canal and the ' zog ' in the maxillo-premaxillary suture being particularly characteristic. CONCLUSION Captain Cook's first expedition to Australia obtained three specimens of kangaroo, all from Endeavour River, Queensland, July 1770. The skull of one of these was still preserved in the Museum of the Royal College of Surgeons in 1939 but was destroyed by bombs during the late war. No trace of the other material has been found. The only figure of the original material hitherto generally known to zoologists is the plate in Hawkesworth (1773) of a not easily determinable kangaroo, or reproductions of it. Four more figures are now published. The first two are indeterminable outline drawings of the whole animal by Parkinson, who was on board Cook's ship. The third is a painting of a skull by Nathaniel Dance. This is almost certainly the skull of one of Cook's specimens ; in fact it is difficult to see where else it could have come from. It is the skull of a Wallaroo of the Macropus robmtus series. The fourth is a photograph (PI. 5) of the specimen which was destroyed in the Museum of the Royal College of Surgeons. This skull was from one of the kangaroos obtained by Cook's party at Endeavour River in July 1770. It was given by Banks to Hunter and is No. 1732 in Owen's catalogue (1853) and No. 3703 in Flower's catalogue (1884). It is the skull of a young Great Grey Kangaroo and we hereby designate it as the photo-lectotype of Macropus canguru (Miiller, 1776) 'Captain Cook's Kangaroo '. REFERENCES BANKS, J. 1768-1771. Journal. [MS. transcript, preserved in the Botanical Department, British Museum, by the Misses Mary and Hannah Turner, aunts of Sir J. D. Hooker who helped to collate it with the original which is now in the Mitchell Library, Sydney.] This MS. contains material which was omitted by Hooker in his published version of the Journal (1896). BARTON, G. B. 1893. Historical Records of New South Wales, 1 (i): 1-526. [This contains reprints of Cook's log, of the journals of his officers, including that of Midshipman J. Bootie, and of correspondence between Cook and the Admiralty and others concerning his voyages.] BOOTIE, J. 1770. See BARTON, G. B. CARRINGTON, H. 1939. Life of Captain Cook. London (Sidgwick & Jackson). [This contains a list of all the known MS. logs and journals of Cook, Banks, and other members of the ship's company of H.M.S. Endeavour.] COOK, J. 1768-1771. Captain Cook's Journal during his First Voyage. . . . [Edited by Capt. W. J. L. Wharton, London, 1893, from a contemporary MS. transcript of Cook's holograph. This transcript is now in the Australian Museum, Sydney. Cook's holograph is in the National Library, Canberra.] DRYANDER, J. [Autograph Catalogue of the Drawings of Animals in the Library of Sir J. Banks, arranged in systematic order.] (Preserved in the Zoological Department, British Museum.) FLOWER, W. H. 1884. Catalogue of Specimens . . . in the Museum of the Royal College of Surgeons of England, 2 : 708. HAWKESWORTH, J. 1773. An Account of the Voyages undertaken . . .for making Discoveries in the Southern Hemisphere ... 6y ... Captain Cook, 3 : 577 (ist edition) ; 173 (and edition) . London. HUNTER, J. 1790. See WHITE, J. 50 THE IDENTITY OF CAPTAIN COOK'S KANGAROO HUNTER, J. 1861. Essays and Observations on Natural History, Anatomy, Physiology, Psychology and Geology, 2 : 250. [Edited posthumously by R. Owen.] IREDALE, T., & TROUGHTON, E. LE G. 1925. Captain Cook's Kangaroo. Aust. Zool. 8:311. 1937. The identity of Cook's Kangaroo. Rec. A ust. Mus. 20 : 67. KITSON, A. 1907. Captain James Cook, &c. London (John Murray). Pp. xvi, 525, frontis., 16 pis., i map. MULLER, P. L. S. 1776. Des Hitters C. von Linne . . . vollstdndiges Natursystem nach der zwolften Lateinischen Ausgabe . . . Supplementsband : 62. Nurnberg. OWEN, R. (1853). Descriptive Catalogue of the Osteological Series contained in the Museum of the Royal College of Surgeons of England, 1 : 322. London. 1861. See HUNTER, J. PARKINSON, STANSFIELD. 1773. A Journal of a Voyage to the South Seas in His Majesty's Ship the Endeavour. Faithfully transcribed from the Papers of the late Sydney Parkinson, &c. : 145. London. [Sydney Parkinson died on 26 January 1771.] PARKINSON, SYDNEY. 1768-1771. [294 Original Water Colour Drawings and Pencil Sketches of Animals made during Cook's First Voyage by S. Parkinson, and (17 of Fish and 4 of Mollusca) by A . Buchan.~\ 3 vols. RAVEN, H. C. 1939. The identity of Captain Cook's Kangaroo. /. Mammal. 20 : 50. SOLANDER, D. C. 1768-1771. [Manuscript descriptions of Animals written on slips and systematically arranged in accordance with Linne' s ' Sy sterna Naturae . . . Editio duodecima reformata'.] 1 : 90. (Preserved in the Zoological Department, British Museum.) TATE, G. H. H. 1948. Studies on the anatomy and phylogeny of the Macropodidae (Marsupialia). Bull. Amer. Mus. Nat. Hist. 91: 233. TROUGHTON, E. LE G. 1942. The kangaroo family origin and earliest discoveries. Aust. Mus. Mag. 8: 17. 1946. Furred Animals of Australia. (Third revised edition.) Pp. xxx, 376, 25 pis. Sydney (Angus & Robertson). WEBB-JOHNSON, A. 1939. The George Adlington Syme Oration: Surgery in England in the making. Aust. N.Z. J. Surg. 9: 10. WHITE, J. 1790. Journal of a Voyage to New South Wales &>c. [Hunter discusses the kangaroo on p. 272.] London. ZIMMERMANN, E. A. W. 1777- Specimen zoologiae geographicae, Quadrupedum domicilia et migrationes sistens, * * ' f ** FIG. 3 ..,,,-,,,,,,,.,, .,,,.,,,,,. 1 _L L FIG. 4 PLATE 5 FIG. 5. Photograph of the skull of a young Great Grey Kangaroo obtained at Endeavour River, Queensland, by Captain Cook's party in July 1770. This plate is the photo-lectotype of Macropus canguru (Miiller). The specimen, which no longer exists, was number 1732 in Owen's Catalogue (1853) and number 3703 in Flower's Catalogue (1884). Scale unknown Bull. B.M. (N.H.), Zoology I, 3 PLATE 5 FIG. 5 PRESENTED -1 APR 1950 PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATEY PRINTER TO THE UNIVERSITY I NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NATURAL HISTORY) D. DILWYN JOHN LERNAEODISCUS PUSILLUS NOV. SPEC. A RHIZOCEPHALAN PARASITE OF A PORCELLANA FROM EGYPT H. BOSCHMA BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 4 LONDON : 1950 NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NATURAL HISTORY) fa) BY D. DILWYN JOHN LERNAEODISCUS PUSILLUS NOV. SPEC. A RHIZOCEPHALAN PARASITE OF A PORCELLANA FROM EGYPT BY DR. HILBRAND BOSCHMA Pp. 51-65; Pi 6; 4 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 4 LONDON : 1950 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is to be issued in five series, corresponding to the Departments of the Museum. Parts will appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. These papers form Vol. I, No. 4, of the Zoological series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued August 1950 Price Four Shillings NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NATURAL HISTORY) 2. SOME ASTROPECTINID SPECIES By D. DILWYN JOHN (DIRECTOR OF THE NATIONAL MUSEUM OF WALES, CARDIFF) (With Plate 6) THE first Note in this series (John, 1948) began with the statement that the Asteroids in the British Museum (Natural History) were being revised. This, the second Note, will be the last in the series by the present author, who has since left the Museum staff. It is shorter than it was intended to be and deals only with the following six Astropectinid species: Lonchotaster tartareus Sladen. Leptychaster antarcticus Sladen. Dytaster exilis Sladen. Leptychaster kerguelensis Smith. Plutonaster agassizii (Verrill). Craspidaster hesperus (Miiller & Troschel). Lonchotaster tartareus Sladen Lonchotaster tartareus Sladen, 1889, Rep. Voyage Challenger (ZooL), 30 : 104, pi. 16, figs. 1-5. The only species and the only specimens of the genus Lonchotaster remain those described by Sladen in 1889, L. tartareus from 2,400 fathoms between the Canaries and the Cape Verde Islands, and L. forcipifer from nearly 2,000 fathoms in the Southern and Antarctic Oceans south-west of Australia. The large Astropectinid described by H. L. Clark (1916: 30) as Lonchotaster magnificus was referred to Dipsa- caster by Fisher (1919: 150). Fisher, both in 1917 (p. 170) and 1919 (p. 150), makes what are, in effect, minor corrections to Sladen's account of L. tartareus, saying there is a small spine on each marginal plate and one on most of the actinal intermediate plates; he refers to Sladen's figures as bearing out his statement. As for the superomarginal plates, Fisher is wrong and Sladen's account, with which his plate agrees, is correct : ' within the interbrachial arc and at the base of the rays in the large example, a small conical tubercle is present close to the upper end of the plate, but it is not found in the smaller specimens'. For the inferomarginals neither Sladen's account nor Fisher's is quite correct. In the larger specimens there are small spines, of diminishing size, as far out as about the thirtieth plate, but not beyond; they are present on the plates of the interbrachial arc of one of the smaller, entirely absent from the other. Sladen's account of the spination of the actinal intermediate plates is correct, including the implication that there are no spines on those of the smaller specimens. 54 NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) Dytaster exilis Sladen Dy 'taster exilis Sladen, 1889, Rep. Voyage Challenger (Zoo/.), 30 : 65, pi. 2, figs. 3 & 4 ; pi. 4, figs. 9 & 10 (figs, of var. gracilis) ; Wood-Mason & Alcock, 1891: 429; Alcock, 1893: 80. The Challenger took the type of D. exilis off Valparaiso in the Pacific, those of its varieties gracilis and carinata in the Atlantic near Tristan da Cunha and off the Maryland coast of N. America respectively. The only subsequent records are those of exilis itself by Wood-Mason and Alcock from the Bay of Bengal, where it 'has several times been met with . . . between 1748 and 1924 fathoms on globigerina ooze '. They did not describe their specimens beyond giving the colour when fresh as salmon- pink. One of their specimens, from St. 117, 1,748 fms., is in the British Museum. It is dry and small: R = 47 mm., r = 9 mm., R : r is 5-2. The abactinal paxillae have four to ten finely thorny spinelets ; there are no pedicellariae among them. The supero- marginals number thirty-three. They are not confined to the lateral wall but encroach a little on the abactinal surface ; those in the inter-brachial angle do so to the extent of i mm. This is a marked difference to the type of exilis ; in the variety gracilis, on the other hand, they do encroach abactinally though not so strongly as in this specimen. When seen from the side the length of the plates is less than the height in the inter-brachial angle, greater than it in mid-arm, equal to it at the end of the arm. The large spines are missing from the plates at the ends of the arms which are abraded, but I am unable to say if they have merely been rubbed off. The inferomarginals correspond to and are of the same size as the superomarginals as seen from the side. On the actinal surface their breadth is greater than their length on the inner part of the ray. In the interbrachial angle some of the marginal plates of both series carry two spines. The enlarged spine on the adambulacral plate first appears about half-way down the arm and arises more often from the second than the first comb of spines. The latter has ten, the former eight, spines, and they are followed by a third row as Sladen describes for exilis. The actinal intermediate plates extend to about the third inferomarginal. Each bears a group of widely spaced spines, up to fourteen on the largest. They and the spines of the marginal and adambulacral plates are finely thorny. The madreporite is neither large nor conspicuous. In the shape of the superomarginal plates, the absence of pedicellariae, and the occurrence of the enlarged spine on the adambulacral plates I see this specimen as nearer to the var. gracilis than to exilis itself. Experience with other species leads me to believe it possible that more specimens may serve to bridge the gap which now appears to exist. Verrill (1895 : 131) was not able to satisfy himself that D. exilis var. carinata was distinct from the young of his D. grandis (of which D. madreporifer Sladen is a synonym) . A direct comparison leaves no doubt of its distinctness. In the first place the larger specimen described by Sladen cannot be regarded as young, having R 98 mm. The paxillae of its disk are comparatively large, those of grandis conspicuously small ; the pedicellariae on the actinal intermediate plates of carinata are larger and NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) 55 of valves more highly modified than those of grandis (Plate 6, fig. i) ; the adambulacral armature differs, for whereas grandis has only one row of strong furrow spines, carinata has two, the second being of the peculiar dagger-like form described by Sladen. Finally, the appearance of the two forms is quite different to the naked eye for, whereas D. grandis is distinguished by the strong high sides which the marginals give to its rays, in the var. carinata the marginals are comparatively poorly developed, their combined height being only a little more than half that of grandis, and the spines are correspondingly smaller (Plate 6, figs. 2 & 3). Plutonaster agassizii (Verrill) Archaster agassizii Verrill, 1880, Amer. J. Sci. 20 : 403. Plutonaster rigidus Sladen, 1889, Rep. Voyage Challenger (Zoo/.), 30 : 91, pi. 14, figs. 3 & 4 ; pi. 15, figs. 3 & 4; Koehler, 1909: 19, pi. 4, fig. 6; pi. 10, figs. 5 & 6. Plutonaster rigidus var. semiarmata Sladen, 1889, Rep. Voyage Challenger (Zoo/.), 30 : 94, pi. 14, fig- 5- Plutonaster agassizii Verrill, 1894, Proc. U.S. Nat. Mus. 17 : 248; 1895: 131; 1899: 211, pi. 27, fig. 6. Verrill (1880: 403) in his 'Notice of the remarkable Marine Fauna occupying the outer banks off the Southern Coast of New England' described the new species Archaster agassizii. Sladen (1889) made no reference to Verrill 's paper in the Chal- lenger Report. In 1894 (p. 248) Verrill placed his species in Sladen's genus Plutonaster \ listed Sladen's rigidus and rigidus var. semiarmata and a part of his bifrons, all from off the coast of North America, as synonyms ; and added to the description. In 1899 he described the species as occasionally having pedicellariae and gave a figure show- ing one. Koehler (1909: 19) used Sladen's name, rigidus, for describing a series taken in mid-Atlantic in the latitude of the Azores, explaining that he did so because he could not be sure that Verrill's agassizii and Sladen's rigidus were the same. He found Verrill's description inadequate and his attempt to have photographs of his specimens compared with Verrill's had failed. Dr. Austin Hobart Clark has generously made it possible for me to make the sort of comparison that Koehler wished to make by sending me six specimens of Verrill's species. They came from off New Jersey, 39 58' 30" N., 70 30' oo" W., 384 fms. They show that agassizii and rigidus are one. Koehler had found that the var. semiarmata of Sladen could not be maintained, so variable is the occurrence of spines on the inferomarginal plates. Verrill (1894: 248) says that there may be all grada- tions from those having no marginal spines whatever to those that have a large spine on nearly every marginal plate of both series. Koehler does not record spines on the superomarginal plates and it may be assumed that they were not present in his specimens. There is none in the six specimens from Verrill before me, but in the type of Sladen's rigidus there is on one or two plates a single slightly enlarged granule such as I have seen to occupy a similar position from which a spine often arises in other asteroids. Koehler makes no mention of pedicellariae. I find a row of four to be present actinally in the midline of one interradius of one of Verrill's specimens, arid a single 56 NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) one in another interradius. They have four or five blades. The type of rigidus has some small groups of spines in the actinal intermediate areas which are pedicellaria- like in their disposition, but the 'blades' are short and coarse. Sladen (p. 92) described the conical spinelet immediately behind the furrow spines on the outer adambulacral plates. Though Koehler did not mention it, it is to be assumed it was present since he identified his specimens with Sladen's species. It is present in VerrilTs specimens, more strongly developed in some than in others. R : r is more than 3 in one of Verrill's specimens (R = 49 mm., r = 15 mm.) ; it is less than 3 in the remaining five in which R varies from 42 to 63 mm. and r from 17 to 22 mm. Verrill included the small specimen which Sladen (p. 88) described with a query as P. bifrons in his synonymy of agassizii. It possesses a spine on each marginal plate, inferior and superior ; there is a large spine behind the furrow series on each adambula- cral plate. In view of its origin it is probably the young of agassizii, but it cannot be said with certainty that it is. 1 Leptychaster antarcticus Sladen and L. kerguelensis Smith Leptychaster antarcticus Sladen, 1889, Rep. Voyage Challenger (Zoo/.), 30 : 190, pi. 31, figs. 3 & 4; pi. 32, figs. 7 & 8. Leptychaster kerguelensis Smith, 1876, Ann. Mag. Nat. Hist. 17 : no. The type of L. antarcticus, and a second and smaller specimen taken with it (R 10-5 mm., r 4-5 mm.), are in the Museum collection. They are the only specimens recorded. Bell (1908 : 9) thought them the young of kerguelensis, but he gave no good reasons for doing so. Koehler (1917: 53) discussed the question and Fisher (1940: 83) referred to it, but, while not affirming that Bell was wrong, neither accepted his conclusion. It seemed well that I, with access to the types of both species, should re-examine them and other available specimens and report what I find. The paxillae of the greater part of the swollen abactinal surface of the type of antarcticus have lost their spines. It may have happened during transport to and from a safe place in the Second World War. They appear to have been present when the Challenger figure (pi. 31, fig. 3) was made. While Sladen's written description is of his usual excellence, fig. 4, pi. 31, is a poor representation: it is, indeed, a misrepre- sentation of the mouth plates, which are as Sladen describes them in words. It is hoped that the photograph given here conveys a better idea (Plate 6, fig. 4) . Sladen's description of kerguelensis is of a large specimen of R = 66 mm. ; though he listed smaller specimens and gave their sizes he did not otherwise describe them. He states (p. 192) that kerguelensis is distinguished from antarcticus by the longer and more cylindrically rounded rays, by the larger and more compact paxillae, by the smaller actinal intermediate areas, and, above all, by the characteristic adambulacral armature. The smallest specimen of kerguelensis in the collection was taken with three larger 1 A doubt is possible about its origin. On p. 87 Sladen gives it as St. 47, off the coast of N. America. On p. 88 he gives St. tfa. There was no Challenger station of that number but there was one by the Porcupine and it was in the Faroe Channel. NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) 57 specimens (R up to 60 mm.) in 50 fms., off Marion Is. In it R = 13-8 mm. and r = 5 mm., so that it is slightly smaller than the type of antarcticus (R = 15 mm., r = 6 mm.). A direct comparison has been made between them. The rays of the kerguelensis specimen are, in proportion, longer and more rounded, and the actinal intermediate areas are smaller; and the differences in proportion give a different facies to each specimen. But the paxillae are similar in the two specimens and as Sladen described them for antarcticus, though his figure is not very good. It is, however, far better than is that of the paxillae of kerguelensis (pi. 32, fig. i). In only three of the fifteen Museum specimens are they as shown in that figure, with the spines represented by low rounded granules, tending to be polygonal where crowded. In the others they are much more spine-like and radiate apart. Though it is not necessarily the biggest specimens in which the paxillae spines are lowest and most crowded, it is in the smallest that they are most spine-like. In short, the distinction between kerguelensis and antarcticus based upon the nature of their paxillae appears not to be real. The question of the adambulacral armature remains. It can only be said that Sladen's descriptions are correct and that his figs. 2 & 8, pi. 32, are good representa- tions. It may be added that Koehler's eight specimens of kerguelensis conformed with Sladen's description for that species, and that it is implicit in Fisher's account that his three specimens also did so. And so, since no intermediate stages have been found, it seems best to go on regard- ing kerguelensis and antarcticus as distinct species distinguished by their different adambulacral armature. The three starfishes from the Cape which Bell (1905 : 242) recorded as L. kerguelensis are Dipsacaster sladeni Alcock, as Mortensen (1933: 237) pointed out. Bell (1908: 9) also recorded the species from the Ross Sea, including one specimen in which R = 212 mm. I cannot find that specimen ; nor are there any Ross Sea specimens labelled L. kerguelensis. There are several jars labelled by Bell ' Lepty chaster young' or 'very young', and I suppose them to be the young examples to which he referred. They are, however, not Leptychaster but Odontaster and some other genera are included. Craspidaster Hesperus (Miiller & Troschel) Archaster hesperus Miiller & Troschel, 1840, Ber. preuss. akad. Wiss. : 104. Craspidaster hesperus Sladen, 1889, Rep. Voyage Challenger (Zool.), 30: 177, pi. 17, figs. 5-7; pi. 18, figs. 1-4; Doderlein, 1921: 5 (for synonymy), 8, pi. i, figs. 2-3. Craspidaster glauconotus Bedford, 1900, Proc. Zool. Soc. Lond.: 290, pi. 24, figs. 8a, b; Doderlein, 1921: 8, pi. i, figs. 4-6. Craspidaster hesperus crassus Doderlein, 1921, Siboga Exped. Monog. 46 i : 9, pi. i, figs, i & la. There are in the British Museum thirty-nine specimens. One is from an unknown locality, five are said to be from Japan but there can be no certainty of it, twenty-one from the Chusan Archipelago, one from Amoy, and another from Hong Kong (Chal- lenger) , two each from the Philippines (Challenger) and Batavia, and six specimens of Bedford's glauconotus from Malacca. Doderlein had twelve specimens and took into account, for measurements, &c., three more. He recognized three sub-species differing from one another in the length 58 NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) and width of the arm, the number, size, and spination of the marginal plates, and the number and nature of the actinal intermediate plates. Four of his specimens were from China and Japan, the remainder from East Indian or Malayan seas. The former had shorter and wider arms, and larger and on the whole, and especially in the second row fewer actinal intermediate plates. One of the Chinese specimens of un- usually plump form, with massive marginals and having only one row of actinal inter- mediate plates, he made the type of a new sub-species, crassus ; the remainder he regarded as typical hesperus. The Malayan examples, with longer more slender arms, more numerous marginals, smaller and more actinal intermediate plates especially in the second row and with, in the larger, spines on the ventral faces of the infero- marginals, he grouped with Bedford's specimens in the sub-species glauconotus. The present collection bears out Doderlein's conclusions concerning the relation of R : r, and the number of marginal plates. In the twenty-one Chusan specimens R ranges from 8-5 to 42 mm. and the relation R : r varies from 2-1 in the smaller to 3-5 in the larger. In the six specimens of glauconotus from Malacca the range of R is 18 to 67 mm. and of R : r 3-2 to 4-6. There is no doubt that the latter are conspicuously longer-armed. They have, too, a larger number of superomarginal plates. Perhaps the most telling way of making a difficult comparison is to bring together (i) a number of specimens of roughly equal sizes, as follows : Locality R in mm. R :r No. of marginals ? Japan .... Chusan .... Timor (Doderlein) Malacca (glauconotus) 34 2Q-5 29 31 3'2 3' 1 3-6 4'4 2 4 23 26 33 and (2) a number of specimens with roughly equal numbers of marginal plates : Locality No. of marginals R R :r ? Japan .... 27 4 1 3'4 Chusan .... 30 42 3'5 Hong Kong 3i 53 3-6 Philippines .... 3i 37'5 3-8 Malacca (glauconotus) 33 3i 4'4 The first list shows that Bedford '$ glauconotus is sharply marked off from the other specimens by the high value of R : r and by the large number of marginal plates ; the second, that a specimen of glauconotus with a given number of marginals is of much smaller major radius and has a markedly higher value of R : r than specimens of hesperus with the same number of marginals. 1 Each list tells the same story, but by means of different specimens. One of the Batavia specimens is roughly equal in size (R 57 mm.) to one of those from Malacca (R = 59 mm.). R : r is 4 in the former, 4-3 in the latter, and the relative numbers of marginal plates are 40 and 47. 1 The large major radius of the Hong Kong (Challenger) specimen is because of its peculiarly massive marginals; compare the type of crassus which, with only 20-22 marginals, has R = 46 mm. NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) 59 The spines on the lower surfaces of the inferomarginal plates and on the actinal intermediate plates afford a strong difference between Bedford's glauconotus and typical hesperns. They are well developed on each of the six specimens. They occur, strongly on the inferomarginal plates, poorly developed on the actinal intermediate plates, of the larger specimen (R = 57 mm.) from Batavia ; there are traces of them on the actinal-intermediate plates only of the second Batavian specimen (R = 57 mm.). There are spines, varying in number but never numerous, on the lower surfaces of the inferomarginals of (i) the Challenger specimen from Hong Kong (an odd one or two), (2) the larger Challenger specimen from the Philippines (one on each of two rays), and (3) one of the Japan specimens (one on each of the first eight plates). I find nothing to support Doderlein's implication that there is a real difference in the number of actinal intermediate plates of 'Chinese' and 'Malayan' specimens. He gives as a characteristic of some of the former that they have few and massive plates, sometimes only one row (var. crassus}. It is true that in the British Museum collection six of the smaller specimens from Chusan (R = 10-17 mm.) have only one row, but since the remaining and larger specimens have two rows, and the largest specimens have the highest number of plates, this is clearly a matter of growth. The only other specimens with no second row of actinal intermediate plates are (i) one of glauconotus of no less than R = 60 mm. (no second row in two interradii ; a single plate comprises the ' second row ' in each of the other three) ; (2) the smallest specimen of glauconotus (R = 18 mm.) ; (3) Sladen's 'young phase' (R = 22 mm.) from the Philippine Islands. The largest glauconotus (R = 67 mm.) has six to eight plates in the first, three plates in the second, row. The specimen from an unknown locality is exceptional: it has R = only 31 mm. and yet has seven to eight plates in the first row, three to four in the second, and it possesses a third row of one plate on either side. Sladen described the occurrence of a thumb-like spine on the aboral margin of the adambulacral plates of his Hong Kong specimen and its absence from those from the Philippines. It was not present in the specimens from the Philippines seen by Fisher (1919: 60). Doderlein does not mention it. 1 It is (as Bedford says) present in glauco- notus ; I find it in each specimen from the smallest (R = 18 mm.) to the largest (R = 67 mm.). It is present in the specimen from an unknown locality and in that from Amoy, in three of those from Japan (R = 35-41 mm.), but it is absent from all but a few plates of the fourth (R = 34 mm.). It is not present in the two specimens from Batavia. It is absent from twenty of the twenty-one specimens from Chusan of R = 8-5 to 29-5 mm., but is present in the twenty-first which is conspicuously larger having R = 42 mm. The conclusion appears to be that in the present state of our knowledge glauco- notus should continue to rank as a sub-species distinguished by the length of its rays, the number of its marginals, and the presence of spines on the inferomarginal and actinal intermediate plates ; but that crassus cannot be maintained. The species is seen to be variable: e.g. the Hong Kong specimen approaches Doderlein's crassus in its massive marginals and yet bears traces of spines, a glauconotus character, on some of them ; the thumb-like spine of the adambulacral plate is absent from most small 1 His fig. 6a on pi. i shows it to have been absent from his specimen from Lombok. Text-fig, i and the accompanying text do not make clear the possibility of its existence. 60 NOTES ON ASTEROIDS IN THE BRITISH MUSEUM (NAT. HIST.) specimens but it is present in one glauconotus, R = 18 mm., and it may be entirely wanting on large specimens up to R = 57 mm. REFERENCES ALCOCK, A. 1893. An account of the collection of deep-sea Asteroidea. Ann. Mag. Nat. Hist. 11 : 73-121, 3 pis. BEDFORD, F. P. 1900. On Echinoderms from Singapore and Malacca. Proc. Zool. Soc. Lond.: 271-299, 4 pis. BELL, F. J. 1905. The Echinoderma found off the coast of South Africa. 2. Asteroidea. Mar. Invest. S. Afr. 3 : 241-253. 1908. Echinoderma. National Antarctic Expedition, 1901-1904, Natural History, Zoology (Echinoderma), 4 : 1-16, 5 pis. CLARK, H. L. 1916. Report on the Sea-Lilies, Starfishes, Brittlestars and Sea-Urchins obtained by the F.I.S. Endeavour on the coasts of Queensland, New South Wales, Tasmania, Victoria, South Australia, and Western Australia. Biol. Res. 'Endeavour', 1909-1914, 4: 1-123, 44 pis. DODERLEIN, L. 1921. Die Asteriden der Siboga Expedition. I. Porcellanasteridae, Astro- pectinidae, Benthopectinidae. Siboga Exped. Monog. 46 i : 1-47, 13 pis. FISHER, W. K. 1917. Notes on Asteroidea. Ann. Mag. Nat. Hist. 20 : 166-172. 1919. Starfishes of the Philippine Seas and adjacent waters. Bull. U.S. Nat. Mus. 100 (3): 1-712, 1 56 pis. 1940. Asteroidea. 'Discovery' Rep. 20 : 69-306, 23 pis. JOHN, D. D. 1948. Notes on Asteroids in the British Museum (Natural History) i. The Species of Astropecten. Novit. Zool. 42 : 485-508, 4 pis. KOEHLER, R. 1909. fichinodermes provenant des campagnes du yacht Princesse Alice. Result. Camp. sci. Monaco, 34 : 1-317, 32 pis. 1917. fichinodermes recueilles par M. Rollier du Baty aux lies de Kerguelen, en 1913-1914. Ann. Inst. Oceanogr. Monaco, 7 (8): 87 pp., 10 pis. MORTENSEN, T. 1933. Echinoderms of South Africa. Vidensk. Medd. naturh. Foren. Kbh. 93 : 215-400, pis. 8-19. MULLER, J., & TROSCHEL, F. H. 1840. [In Report of the Session of 30 April 1840.] Ber. preuss. akad. Wiss. SLADEN, W. P. 1889. Asteroidea. Rep. Voyage Challenger (Zool.), 30. VERRILL, A. E. 1880. Notice of the remarkable Marine Fauna occupying the outer banks off the southern coast of New England. Amer. J. Sci. 20 : 390-403. 1894. Description of new species of Starfishes and Ophiurans, with a revision of certain species formerly described. Proc. U.S. Nat. Mus. 17 : 245-297. 1895. Distribution of the Echinoderms of North-eastern America. Amer. J. Sci. 49 : 127-141. 1899. Revision of certain genera and species of Starfishes with descriptions of new forms. Trans. Conn. A cad. Arts Sci. 10 : 145-234, 8 pis. WOOD-MASON, J., & ALCOCK, A. 1891. Natural History notes from H.M. Indian Marine Survey Steamer Investigator. Ser. 2, no. i. On the results of deep-sea dredging during the season 1890-1891. Ann. Mag. Nat. Hist. 8 : 427-443. PLATE 6 FIG. i. Dytaster exilis var. carinata, type, mouth-angle and actinal- intermediate area, X 5. FIG. 2. Dytaster exilis var. carinata, type, side view of the proximal portion of arm, x 4. FIG. 3. Dytaster grandis, cotype, side view of proximal portion of arm, X4- FIG. 4. Lepty 'chaster antarcticus, type, under surface of disk, x 10. ull. B.M. (N.H.) Zoology I, 4 PLATE LERNAEODISCUS PUSILLUS NOV. SPEC., A RHIZOCEPHALAN PARASITE OF A PORCELLANA FROM EGYPT By HILBRAND BOSCHMA (DIRECTOR, RIJKSMUSEUM VAN NATUURLIJKE HISTORIE, LEIDEN) IN 1936 Dr. Isabella Gordon kindly sent me twelve specimens of Rhizocephalan parasites on Porcelain Crabs collected by Dr. R. Gurney in coral rock on the Harbour Reef near Ghardaqa, Red Sea, Egypt. The hosts of these parasites were provisionally identified as Porcellana serratifrons of Nobile, nee Stimpson. The parasites appear to represent a hitherto undescribed species. FIG. i . Lernaeodiscus pusillus : a-c, dorsal view of three specimens, mantle opening in the upper part, stalk in the lower part of the figures; d-f, ventral view of the same specimens, x 18. The animals are of very small size, their greatest diameter being about 2 mm., their antero-posterior diameter (in the median plane) about i^ mm., and their smallest (dorso-ventral) diameter less than i mm. The total diameter in the antero-posterior direction is, as a rule, slightly less than the greatest diameter. The outlines of three specimens in dorsal view are given in Fig. la-c, in ventral view in Fig. id-f. The shape of the parasites is more or less roundish or somewhat trapezoid or triangular ; their contour is slightly irregular as the mantle shows a number of rather incon- spicuous lappets. The comparatively wide mantle opening, which is surrounded by a well-developed muscular wall, is found on the anterior region of the dorsal surface. As a rule the dorsal surface shows a system of three shallow grooves running from ZOOL. I. 4 H FIG. 2. Lernaeodiscus pusillus, specimen of Fig. la, d. Transverse section showing one of the colleteric glands (eg), dms, dorsal mesentery ; me, mantle cavity ; vm, visceral mass ; vms, ventral mesentery, x 60. FIG. 3. Lernaeodiscus pusillus, specimen of Fig. ia, d. Central parts of transverse sections, a from a region not far from the stalk, each following section from a more anterior region, dms, dorsal mesentery ; It, left testis ; Ivd, left vas deferens ; me, mantle cavity ; rt, right testis ; rvd, right vas deferens; st, stalk; vm, visceral mass; vms, ventral mesentery. LERNAEODISCUS PUSILLUS NOV. SPEC. 63 the centre to the mantle opening and to the lateral parts of the posterior region of the body. On the ventral surface there is a distinct groove running from the stalk in an anterior direction ; this groove varies in length and in breadth. The three specimens shown in Fig. i were sectioned transversely for the study of their internal structure. In sections from the region about half-way between the stalk and the mantle opening the colleteric glands are found ; as a rule one of these is situated more anteriorly than the other. These glands (Fig. 2, eg) are more or less cup-shaped small cavities surrounded by an epithelium with a stronger affinity for stains than the surrounding parts. The figure further shows that the dorsal surface of the visceral mass is broadly attached to the mantle, in this way forming the so- called dorsal mesentery. On the other side the visceral mass is connected with the mantle by means of a real mesentery, the ventral mesentery. Where the latter is attached to the mantle there is, externally, the longitudinal groove referred to above. In the three sectioned specimens the colleteric glands entirely agree with one another in shape, their position in the visceral mass, and their size. The male organs in two of the sectioned specimens are also similar in every respect (Fig. 3), but in the third specimen (Fig. 4) they are slightly more complicated. The male organs closely correspond with those of Lernaeodiscus okadai Boschma (cf. van Baal, 1937, figs. 18-21). The male openings, in a region about half-way between the stalk and the mantle opening, are found on each side of the ventral mesentery (Fig. ^d, e). The vasa deferentia run along the ventral mesentery until they reach the posterior part of the visceral mass. Here they turn towards the dorsal surface (Figs. 30, 40), and continue their course along the dorsal mesentery in an anterior direction. After the vasa deferentia have passed into the testes the latter extend in a lateral direction, so that the terminal part of the testes is the most lateral part of the male organs (Fig. 36, c). As remarked above, the male organs in two of the sectioned specimens have a similar shape (as represented in Fig. 3) ; in the third specimen the male organs show some differences. Here the left testis (Fig. ^d, e) does not extend in a lateral direction, whilst the terminal part of the right testis after continuing its course in a lateral direction towards the right margin of the visceral mass (a in Fig. 4) obtains a curved shape by extending towards the median plane again (p in Fig. 4). The closed end of this testis consequently lies next to the right vas deferens (Fig. 46). Besides having a course in a lateral direction the testes in all the three specimens are strongly contorted, so that in sections they appear to be divided into numerous smaller parts. It is rather difficult to define the characters by which Lernaeodiscus pusillus can be distinguished from the other species of the genus that are, like the new species, parasites of Porcelain Crabs, viz. L. porcellanae Miiller (cf . Miiller, 1862 ; Boschma, 1931) and L. okadai Boschma (cf. Boschma, 1935 ; van Baal, 1937). The external shape of Lernaeodiscus porcellanae seems to be rather constant, the animal having well-developed lappet-like expansions of the mantle. But too few specimens are known to establish this peculiarity as a constant character for full- grown as well as immature specimens. In L. okadai, van Baal (1937) has shown that the external shape is subject to a very large amount of variation. Here, as a rule, iffll FIG. 4. Lernaeodiscus pusillus, specimen of Fig. ic, f. Central parts of transverse sections, a from a region not far from the stalk, each following section from a more anterior region, a, anterior part of right testis ; dms, dorsal mesentery ; //, left testis ; Ivd, left vas deferens ; me, mantle cavity ; p, posterior part of right testis ; ro, right male genital opening ; rvd, right vas deferens ; vm, visceral mass; vms, ventral mesentery. LERNAEODISCUS PUSILLUS NOV. SPEC. 65 the lappets do not occur in young specimens but are generally distinct in mature animals. The specimens of L. pusillus have, as far as their external shape is concerned, a rather constant appearance. The colleteric glands in the genus Lernaeodiscus are of such a simple structure that they cannot furnish characters for specific distinction. The male genital organs are, to a large degree, subject to individual variation, as is evident from van Baal's (1937) elaborate researches on numerous specimens of L. okadai. The only remaining distinctive character is that of the size of the animals. On this character L. porcellanae, by its comparatively large size, is at once distinguished from L. okadai and L. pusillus. In L. pusillus the greatest diameter is about 2 mm., and the total length is but slightly smaller. The sectioned specimens are fully mature, as their mantle cavities contain large quantities of eggs. For L. okadai there are the following data (the numbers giving the length and the greatest transverse diameter in mm.) recorded by van Baal (1937) : 2| x 3 (small number of eggs) ; 4 x 5| (no eggs) ; 4| X 5 (small number of eggs) ; 4x5 (large number of eggs) ; 2f X 3^ (very small number of eggs) ; i| x 2 (no eggs) ; 6 x 7| (large number of eggs) ; 3^ x 5| (no eggs) ; 2 x 4$ (many eggs) ; 2|X4 (crowded with eggs); 5^x6 (many eggs); 4^x6 (many eggs); 4x4^ (many eggs) ; 2^x4 (without eggs). These data show that the specimens with numerous eggs are the larger ones in which at least one dimension reaches 4 mm. Moreover, when in large specimens no eggs are present in the mantle cavity they may have been recently discharged from this cavity. The data, therefore, give sufficient evidence for the opinion that L. okadai reaches its mature state at a stage in which at least in one dimension the body has a size of 4 mm. On the other hand, L. pusillus is fully mature at a size of 2 mm. Summarizing it may be remarked that though the specific characters of Lernaeo- discus pusillus may appear unconvincing there is sufficient evidence for regarding the parasite as specifically distinct from the other forms belonging to the genus. REFERENCES BAAL, I. VAN. 1937. Biological Results of the Snellius Expedition. II. Rhizocephala of the Families Peltogastridae and Lernaeodiscidae. Temminckia, 2 : 1-94, 3 pis. BO'SCHMA, H. 1931. Papers from Dr. Th. Mortensen's Pacific Expedition, 1914-1916. LV. Rhizo- cephala. Vidensk. Medd. naturh. Foren., Kbh. 89 : 273-380. 1935- Notes on Japanese Rhizocephala, with Description of two new Species. Zool. Meded. 18 : 151-160. MULLER, F. 1862. Die Rhizocephalen, eine neue Gruppe schmarotzender Kruster. Arch. Naturgesch. 28 : Bd. i : 1-9. PRESENTED 5SEP195U PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATEY PRINTER TO THE UNIVERSITY 9 - NOV 1951 ON A^SS^E DEEP-SEA FISH NOTACANTHUS PHASGANORUS GOODE (HETEROMI-NOTACANTHIDAE) FROM THE ARCTIC BEAR ISLE FISHING-GROUNDS DENYS W. TUCKER and]. W.JONES BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 5 LONDON: 1951 ON A RARE DEEP-SEA FISH NOTACANTHUS PHASGANORUS GOODE (HETEROMI-NOTACANTHIDAE) FROM THE ARCTIC BEAR ISLE FISHING-GROUNDS BY DENYS W. TUCKER, B.Sc. w AND ]) J. W. JONES, Ph.D. Pp. 67-79; Pis. 7-9 BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 5 LONDON : 1951 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is issued in Jive series, corresponding to the Departments of the Museum. Parts appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. I, No. 5, of the Zoology series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued November 1951 Price Five shillings ON A RARE DEEP-SEA FISH NOTACANTHUS PHASGANORUS GOODE (HETEROMI-NOTACANTHIDAE) FROM THE ARCTIC BEAR ISLE FISHING-GROUNDS By DENYS W. TUCKER, B.Sc. (BRITISH MUSEUM (NATURAL HISTORY)) and J. W. JONES, Ph.D. (UNIVERSITY OF LIVERPOOL) (With Plates 7-9) INTRODUCTION ON the 2yth of August 1949 the Fleet wood trawler Wyre General landed an unusual fish from the Bear Isle grounds. No information is available concerning the depth at which it was taken, but about 100 fathoms may be assumed from our knowledge of the fishery. Messrs. James Mitchell (Port Health Officer) and P. J. Fisher (Chief Sanitary Inspector), who have frequently been instrumental in obtaining rare fishes, kindly forwarded it to the Department of Zoology, University of Liverpool, where it was recognized as a rare Notacanthus and presented to the British Museum. The species is N. phasganorus Goode, new to the national collections. Only five other authenticated specimens are known, all in American museums, and of these but two have been des- cribed and figured. 1 The holotype (U.S. National Museum, Washington, No. 25972 ; Goode (1881) ; Goode & Bean (1894 = 1896)) was taken from the stomach of a Ground-shark, Som- niosus brevipinna Lesueur = S. microcephalus (Bloch & Schneider), on the Grand Bank of Newfoundland, and was partly digested and mutilated about the head. Bigelow & Schroeder (1935) describe a specimen trawled in about 100 fathoms, 20 miles south of Sable Island, which was in good condition except that the viscera had been removed, and the same authors mention a further example from the same locality (Museum of Comparative Zoology, Cambridge, Mass., Nos. 33946 and 35306 respectively) . 1 A large and originally well-preserved Notacanthus obtained in Iceland during the voyage of La Recherche and figured as N. nasus Bloch by Gaimard (1851, pi. XI) and by Cuvier (1836, pi. 55) has been tentatively referred to N. phasganorus Goode by Vaillant (18886), who was able to examine the specimen (Musee National d'Histoire Naturelle, Paris, No. A. 6864). One of us (D.W.T.) visiting Paris in October 1950 was told by Prof. L. Bertin that it could not then be found. "Ties probablement a-t-il et6 detruit a une date ancienne (vers 1889) '. We have little doubt concerning the accuracy of Vaillant's identification, but do not regard the published figures and data available as sufficiently reliable for a critical determination. See Saemundsson (1949) for further discussion and a bibliography of Icelandic material. 70 ON NOTACANTHUS PHASGANORUS GOODE In reply to a request for further information on his material Dr. William C. Schroeder disclosed that two more examples have since been taken : M.C.Z. No. 37027 in 420 fathoms at 42 18' N., 65 01' W., and No. 37037 in 100 fathoms at 44 N., 57 W. Dr. Schroeder is preparing a paper on the species in which these will be de- scribed and has kindly allowed us to use such unpublished data as are needed to establish the identity of the Bear Island specimen. We wish also to acknowledge the assistance of Mr. Ernest A. Lachner of the U.S. National Museum who re-examined the holotype for us. The illustrations to the present paper are (with the exception of Fig. i) the work of Mr. Hubert Williams and the X-ray photographs were taken by Mr. P. E. Purves. Modern papers by Matsubara (1938) on his Notacanthus fascidens and by Trotti (1939) on N. bonapartei Risso (based on the examination of 9 and 69 specimens respectively) have largely invalidated the taxonomic distinctions made by earlier workers, especially by Goode & Bean. Matsubara concludes: ' Among the characteristics used in the taxonomy of the fishes of the family Notacanthidae, the number of anal spines and the positions of the insertions and also end points of the fins, which are in reality most variable, are considered to be of most importance. . . .It would be super- fluous to say that one must re-examine whether or not each known species belonging to the Notacanthidae is an independent species by taking the above mentioned variabilities into con- sideration.' Trotti remarks similarly: ' Concludendo, la grande variabilita del profile del muso e soprattutto la mancanza di persi- stenza del rapporto tra dorsali ed anali dure . . . ci porta ad una revisione dei caratteri differenziali dei rappresentanti del genere Notacanthus e Gigliolia.' In publishing this full account of the new specimen (British Museum (Natural History), No. 1950.3.30.2) we hope to put on record material of value to such a sub- sequent revision, and to justify an identification which not only extends the known range of N. phasganorus from the western Atlantic to the Arctic but also provides the first published data on the bionomics of the species if not of the genus. But although we now identify our specimen with Goode 's species, we are conscious that in the present state of the taxonomy of the genus this name may not be final. There is need of a critical re-examination especially of the material designated N. chemnitzii Bloch 1787, N. nasus Bloch 1795, N. phasganorus Goode 1881, and N. analis Gill 1883, the inter-specific differences between which, as at present described, do not seem greater than the intra-specific variation demonstrated elsewhere by Matsubara and by Trotti. It is probable that such a re-examination of the types of these four 'species' supple- mented by observations from other material will confirm our suspicion that some or all may be identical. This is no new speculation (see, for example, Liitken, 1898), and it may reasonably be inquired why no precise solution has yet been given. The answer is that apart from the comparative paucity of material, aggravated by its wide dis- persal in study-collections, even the type-locality of Bloch's material is not certainly established though stated by him to have come from the East Indies it has since been believed to have come from Iceland and the originally bad condition of the holotype has since further deteriorated. (Cf. accounts of Bloch himself, of Cuvier & ON NOTACANTHUS PHASGANORUS GOODE 71 Valenciennes (1831), and of Hilgendorf in Goode & Bean (1896).) Even if the specimen in the Berlin Museum is still in existence, it is therefore exceedingly doubtful whether it retains characters adequate for a modern redescription of Bloch's species. We have no more material relevant to that problem in the British Museum (Natural History), but hops in a subsequent paper to redescribe the types of N. sexspinis Richardson 1844 and N. annectens Boulenger 1904, and to give accounts of the series of these and related species in our collections as a contribution towards a future full revision. A forthcoming report on the Notacanthidae collected by the Danish Thor Expeditions in the north-eastern Atlantic will provide further material. DESCRIPTION Although the body is very well preserved, three factors seriously complicate the usual table of measurements. Firstly the fish is a spawning female, greatly distended by a mass of ripe eggs : as a consequence the vent is widely dilated, blocked by a large plug of ova, and opens posteriorly rather than ventrally, while the postero-lateral walls of the abdomen project as a pair of pouches which partly embrace the vent and conceal the origin of the spinous anal fin. This general distortion of the abdomen renders measurements of body-height of doubtful value. Secondly, the head of the specimen is markedly downturned in a very ' Mormyrid ' fashion and more so than in any figure or specimen of a Notacanthid that we have seen. Though there is little support for our opinion forthcoming from other specimens of N. phasganorus we are satisfied that the X-ray photograph published as Plate 8 and other considerations (dentition-]- diet, position of operculum in relation to gill-opening) indicate that this may at least be adopted as a natural attitude, even though it may not be the attitude of rest. Accordingly we give two measurements for body-length and other distances from the tip of the snout to various points ; the first represents the measurements with the head forced into line with the body, the second with it in situ. Statements of body proportions are based on the former to facilitate comparison with other accounts ; the corresponding duplicate set may be computed from the data given if desired. Thirdly, there is some doubt regarding the tail, which may have had the tip broken off and subsequently regenerated a caudal fin. In this case it would be necessary to allow about another 5 cm. on the standard length, plus 2-3 cm. for the caudal fin. Measurements Total length Standard length Body: Depth at pectoral pel vies vent Greatest depth Greatest breadth Length, snout to vent 970 mm. (950) 945 (925) 140 170 140 1 80 5 422 (402) ON NOTACANTHUS PHASGANORUS GOODE Head: Length ...... Greatest depth ..... Greatest breadth ..... Interocular. width ..... Length of snout ..... ,, postorbital region . ,, upper jaw .... ,, mandible, to hind end of articular Breadth of gape ..... Length of maxillary spine Diameter of eye ..... Longest gill-raker ..... Dorsal : Distance from snout .... Length of base ..... Horizontal distance from pelvics Measurements (contd.) 122 mm. 92 50 25 35 80 36 39 4i .. 6 21 ,, 6 ' 352 235 12 (350) I II III IV V VI VII VIII IX X XI Lengths of spines . Intervals between spines Length of soft ray . Anal: Distance from snout ,, vent. Length of base spinous base . ,, first spine ,, longest spine (XVIII) ,, ,, soft ray Pectoral : Distance from snout Length, left .... right Pelvic : Distance from snout ,, base to vent tip . Length .... Caudal : Distance from tip to dorsal Length .... 1678: 6 15 20 21 9 ii 12 10 13 14 mm. 24 23 22 21 14 ii mm. 7 mm. 432 10 540 230 2 19 34 148 65 56 350 70 24 46 390 25 (412) (139) (330) Radial formula D. XI-i ; A. XX, 101 + ; C. I2( ?) ; P. 13 ; V. Ill, 7. Gill-rakers on first arch 3 + 1 + 13. Branchiostegal rays 9. Vertebrae 185. (Nos. 75 and 80 have double centre.) (All counts from X-ray photographs.) Scales along lateral line, about 500. Scales in transverse series, 31 above lateral line, 58 below. Pyloric caeca destroyed through decomposition. Length of the head 7-95 times in the total length ; depth at pectoral 6-92 ; depth at pelvic 5-70; distance from tip of snout to dorsal 2-75 ; from tip of snout to pectoral 6-55 ; ON NOTACANTHUS PHASGANORUS GOODE 73 from tip of snout to pelvic 277 ; from tip of snout to vent 2-29 ; tip of snout to anal 2-24 ; from tip of caudal to dorsal 2-48 ; base of dorsal 4-12 ; spinous base of anal 4-21. Snout 3-48 in head ; eye 5-80 ; postorbital part of head 1-52 ; upper jaw 3-38 ; inter- ocular space 4-88 ; mandible 3-12 ; pectoral 1-87 ; pelvic 2-65. Body elongate, compressed, considerably higher at the pelvics than at the pectorals, even allowing for the distension of the abdomen ; the greatest breadth 0-35 the height at the vent ; tapering posteriorly into a long slender tail. Head compressed, shorter than depth of body, 2-46 in the trunk and 3-54 in the length from tip of snout to anal. Snout long, fleshy, 1-4 times the interocular width and 1-66 times the diameter of the eye. Interocular space narrow, strongly convex, 1-19 times the diameter of the eye. Eye covered by semi-transparent skin, lacking an orbital fold. Nostrils close together, much nearer eye than tip of snout, the posterior slit-like, one-third the eye's diameter from the orbit, the anterior opening into a thin-walled tube protected by a small flap. The centres of the eye, of the two nostrils, and the tip of the snout lie on a straight line. Mouth inferior, broad, gently curved ; upper jaw nearly as long as length of snout ; maxilla with a posteriorly directed pungent spine on its upper margin, extending to below the middle of the eye. The integument of the mandible forms a labial fold on each side. Teeth (PL 7, fig. 4) in the upper jaw in a single row, 37 on each side, slender, inclining inward, the bases cylindrical, the tips antero-posteriorly flattened and in- trorse, mesially 3 mm. long, gradating into smaller and simpler lateral ones. Pala- tines movable vertically with two rows of about 25 rather finer teeth on each side, with sharper markedly introrse tips. Mandible with a complete innermost row of about 30 teeth on each side, resembling those of the upper jaw but more delicate, preceded by two irregular rows of fine aciculate teeth which do not extend as far laterally as those of the main series. All teeth more or less movable. Anteriorly the teeth of the upper jaw bite between the two series of the lower, but owing to the greater radius of curvature the posterior teeth bite outside those of the mandible. The palatine teeth engage with those of the lower jaw. No vomerine teeth. Gill-openings wide, membranes separate and free from isthmus. Gills four; no pseudobranch visible on superficial examination. Gill-rakers slender, pointed, in- curved, well separated, having minute bristles on their inner faces ; a little more than half the length of the gill-filaments, the longest 3-50 in the diameter of the eye. The prominent pores of the lateralis system of the head are distributed thus : 3 in the supra-temporal series, and on each side 5 in the supra-orbital (comprising 2 above the eye, I above the posterior nostril, 2 before the anterior nostril), 16 in the infra-orbital and 14 in the preoperculo-mandibular series. Lateral line gently arched over pectoral, following profile of the back, thence dropping obliquely to one-third the depth of the body over the vent, and further descending nearly to a median position at the point where it disappears two-thirds of the way along the tail. Lateral line pores conspicuous with darkly pigmented lips. Entire body scaled, even to the lips, except for the hinder margin of the opercu- lum. Scales cycloid, rectangulo-ovate, closely inset in tough sheaths; very small on the head (1-2 X i-o to 2-2 X 2-0 mm.), increasing in size posteriorly to a maximum 74 ON NOTACANTHUS PHASGANORUS GOODE of 4-5 X 3-7 mm. on the middle of the body, and thereafter becoming progressively reduced until half-way along the tail they equal those of the head. Pectorals vertically inserted at middle of body-depth, at a distance behind the gill- opening equal to length of own base ; bases broad, fleshy, scaled, pedunculate ; pos- terior edge of fin rounded, length slightly more than half length of head. Pelvics (PL 7, fig. 3) closely adjacent, separated by a narrow groove, reaching far short of the vent. Bases fleshy, pedunculate, thickly covered with scales, origin very slightly behind vertical through origin of dorsal, posterior edge rounded. The third pelvic spine has two much smaller ones set against its base, the first of these concealed by skin. First dorsal spine (PL 9, fig. 6) hidden under the skin ; last dorsal spine the longest, followed by a recurved soft ray (PL 9, fig. 7) set in a fleshy protuberance. There is a slight groove between the bases of the spines and each supports a slight membrane posteriorly which is best developed between the last spine and the soft ray. The anal commences immediately behind the vent and below the Vth dorsal spine ; the XHIth anal spine lies under the last dorsal. The anal spines are embedded in fleshy tissue (the first completely concealed, PL 9, fig. 8), from which successive spines emerge farther and farther. Caudal (PL 9, fig. 9) clearly separated from anal, but lacking a distinct peduncle and probably aberrant owing to regeneration of tip (see p. 75). Colour. Head and body dark brown, tending to be lighter on the forehead and flanks ; lips and hinder edge of operculum bluish-black, fin-rays and anal fin dusky. The fish had a glossy, varnished appearance when dry. Peritoneum and stomach and inside of buccal cavity and operculum black, intestine cream. COMPARISON WITH SPECIMENS PREVIOUSLY DESCRIBED The original description of the holotype (Goode, 1881) gives the radial formula D. X ; A. XIX (130) ; C.o ; P. (17) ; V. II, 8-9. Mr. Lachner was asked to re-examine the dorsal, pectoral, and spinous anal fins only, ascertaining whether any concealed spines and rays had been overlooked and whether a count of the pectoral rays obtained by means of an incision across the fleshy base required any modification of the above formula. He finds the right pectoral fin wanting and gives the count for the left : the revised formula now reads : Holotype: D. X-i ; A. XIX, 130; C.o; P. 18; V. II, 8-9. compared with: M.C.Z. No. 33946 D. XI-( ?) ; A. XXIV, 127 ; C. 7 ; P. 17 ; V. Ill, 7. New specimen, D. XI-i ; A. XX, 101+ ; C. i2( ?) ; P. 13 ; V. Ill, 7. Bigelow & Schroeder give A. XX for M.C.Z. No. 35306. Schroeder, in lit., provides the following additional data : M.C.Z. No. 35306 P. 1 6. One soft ray in dorsal. ,, 37027 P. 13. One ,, ,, 3737 P- l6 - Two soft rays in dorsal. 33946 Not available for re-examination. Bearing in mind the known variation in other species we may regard the counts for ON NOTACANTHUS PHASGANORUS GOODE 75 dorsal, ventral, and spinous anal fins as giving an adequate agreement. 1 The range of variation in the pectoral (13-18) is remarkable, however, even compared with Trotti's counts for N. bonapartei (12-14) and Matsubara's for N. fascidens (12-15). The discrepancies in the counts given for the caudal in part reflect the curious mis- understanding which has surrounded the problem of the tail'structure in this group. The diagnoses of Goode & Bean (1894) contain mutual contradictions : Fam. Notacanthidae. 'Anal fin ... extending ... to the caudal with which it unites.' Notacanthus. 'No caudal', although under the same generic diagnosis N. sexspinis is given a count of C. 5. In the accounts of the various species several numbers are given, including N. phasganorus with C.o. Regan (1929) gives: Order Heteromi. ' A long tail, with a long anal fin below it, tapering to a point, without caudal fin.' While the relations of anal and caudal are certainly difficult to ascertain in these fishes and really call for radiographs and alizarin preparations for their proper eluci- dation, there can be no doubt that many previous descriptions made before the use of the binocular microscope became de rigueur will prove to be erroneous when the material is re-examined. The present specimen shows a distinct separation between the caudal and anal rays, more easily studied in an X-ray photograph (PL 9, fig. 9), which shows at least 12 caudal rays. But the structure is markedly different from that of the tails of other species which we have examined, which are symmetrical, having a distinct though small caudal peduncle, already described and figured in AT", phasganorus by Bigelow & Schroeder (1935). The appearance presented in our figure suggests that the tail has lost its tip at some time and subsequently regenerated a caudal fin. Since Goode almost certainly included the caudal rays in his count for the anal fin (130) we should do likewise to obtain a comparison, and so have 134 for the fish des- cribed by Bigelow & Schroeder and 113+ for the new specimen. A truncation of the tail would also account for this lower number. Gaimard's (1851) figure of the La Recherche specimen evidently represents a tail even more markedly truncated (Vaillant, 18886) and again with a regenerated caudal fin. It seems that this condition is not uncommon in Notacanthus. 1 Vaillant's (18886) data, supplemented by counts from Gaimard's (1851) plate, give the radial formula : D. XI-i ; A. XXII, 92 + ; C. 8 ( ?) ; P. 16; V. Ill, 8 for the La Recherche specimen, which therefore comes within tie known range of N. phasganorus. For further comparison the following counts all purport to have been taken on the holotype of N. nasus by Bloch (1795), Cuvier & Valenciennes (1831), and Hilgendorf for Goode & Bean (1896) respectively: D. X; A. + C. XIII, 136; P. 16; V. II, 8. D. X-O; A. XIII, 116; C. 8; P. 17; V. I, 8. D. XI; A. XV, 118; C. ?; P. 19; V. Ill, 7 (1), 8 (r). There seems to be little useful purpose in attempting to decide the relation between N. nasus and N. phasganorus on such data, except to remark that the only serious discrepancy, the consistently low count for the spinous anal, must be considered against the range of A. IX-XVIII demonstrated by Trotti (1939) in N. bonapartei, and the anterior fin-structure shown in our PL 9, fig. 8. ZOOL. I. 5 K 7 6 ON NOTACANTHUS PHASGANORUS GOODE ANATOMY Those skeletal features discernible from X-ray photographs agree with the very full accounts given by Giinther (1887) for N. sexspinis and Vaillant (18880, b) for N. mediterraneus. Vaillant gives the more detailed account of the general anatomy. The viscera in the present specimen are in general poorly preserved, but it is possible to supplement these descriptions in certain details. The spacious body-cavity is very high, and extends posteriorly considerably behind the anus, to the level of the seventh anal spine. The kidneys are large, the deep anterior lobes flanking the rectum and not extending farther forward in any , cm P. D. i R.M. FIG. i. Gas-bladder from left side. P.D., pneumatic duct ; A. and V., artery and vein supplying bladder; R.M., retia mirabilia. The dotted portions indicate the extensions of the pneumatic duct and of one rete within the bladder. bulk, while the remainder of the kidneys extend back along the roof of the post-anal body-cavity. There is no urinary bladder preserved. The undivided liver, the gonads, and the alimentary canal appear to agree with previous accounts, but the gas-bladder shows some marked differences and merits fuller treatment. Whether the dis- crepancies are due to interspecific variation or to inaccuracies of description cannot be stated. The gas-bladder (Fig. i) is oval in shape with a small blind posterior prolongation, and lies above and extends slightly before the ventral fins. It is suspended in a fold of mesentery with a rather stronger attachment posteriorly ; the bulk of it being free anteriorly sags down into the body-cavity. The tunica externa comprises the usual two easily separable layers : an outer thin, tough, white, and muscular and an inner very dense and silvery, containing elastic fibres. The tunica interna comprises a sub- stantial basis of dense connective tissue supporting a poorly preserved series of muscular, vascular, and columnar epithelial layers. The lumen of the bladder con- tains a quantity of granular yellow matter. There is a fairly long pneumatic duct which does not approach anywhere near the oesophagus. Along it run the artery and vein supplying the bladder, and a number of streaks of yellowish tissue interpreted as pancreas. The vessels break up into two retia mirabilia before approaching the bladder with the pneumatic duct on the lower ON NOTACANTHUS PHASGANORUS GOODE 77 left side, the combination of these structures forming a laterally compressed body which Giinther regarded as a left 'cornu' of the bladder, the retia evidently being identical with his 'pair of thick muscle-like pads'. The pneumatic duct opens in the centre of the floor of the bladder towards the anterior end. The retia are of the 'rete mirabile unipolare duplex' type (Woodland, 1911, 19110), since dissection does not reveal any recombination of capillaries to form major vessels before they enter the gas- gland. The gas-gland is a small patch of spongy vascular tissue surrounding the en- trance of the pneumatic duct from which similar tracts radiate over most of the lining of the bladder. The postero-dorsal portion of the bladder has a thinner, smoother lining epithelium which probably represents a fully dilated oval (Woodland, 1913). BREEDING Though the precise date of capture is not available it may be assumed that the fish was taken about mid-August, and that the breeding season in Bear Island waters is therefore about that time. The ova, entangled in fibrous tissue, were opaque white when received and slightly elliptical, ranging from 1-20 x 1-30 down to 1-16 x 1-25 mm. diameter. They thus pro- vide a further instance of aspherical teleost eggs to be added to those discussed by Breder (1943). They contain many small colourless oil droplets, lo-yo/A in diameter. FOOD AND FEEDING The stomach was well filled with the remains of some two dozen pink and magenta- coloured Actiniarians, comprising the tops of several small anemones of 1-2 cm. dia- meter and pieces apparently bitten from the rims of much larger ones. In some cases it was possible to distinguish scapus and scapulus ; all the fragments were more or less heavily tuberculated and bore traces of a dehiscent cuticle. A consideration of structure in relation to diet leads to some interesting conclusions. i. The dentition and shearing bite of the jaws are admirably suited to feeding on Actiniarians. What would, on theoretical considerations, seem the ideal shape of the head and position of the mouth ? A terminal mouth would require the fish to stand on end in the water when feeding, a rather unlikely proceeding, or to perform move- ments like those of the Lemon Dab Pleuronectes microcephalus Donovan which re- moves tubicolous polychaetes from their burrows by 'bringing its mouth down almost vertically upon its victim by a strong arching of the anterior part of the body' (Steven, 1930). (The same species in the southern North Sea feeds largely on Cerianthus sp. ; Todd, 1907.) This last movement is hardly possible to a stout- bodied fish such as our Notacanthus. There remains only the combination of an in- ferior mouth with what degree of flexure can be attained, the condition in fact which is illustrated in PI. 8, where there is a marked downturning of the vertebral column bringing the jaws into the best position for horizontal and near-horizontal biting. From these considerations, accompanied by the fact that there is no indication of any fracture or dislocation of the skull and pectoral region, we believe that the head of our specimen is in fact being carried in a normal position, though whether this is faculta- tive or permanent cannot be decided. 78 ON NOTACANTHUS PHASGANORUS GOODE 2. The pieces of anemones present fall, as we have noted, into two size-groups, those from very small and very large individuals. The absence of remains of medium- sized ones suggests that such animals are possibly too large to be taken entire and yet too small to allow the fish to take a bite because the curvature of their body surface is so sharp that the jaws at maximum gape cannot obtain sufficient hold. With larger anemones it becomes possible to take a bite from the rim. 3. Giinther (1887) remarks of N. sexspinis: 'The osseous framework of this fish is so much wanting in the characteristic peculiarities of bathybial fishes as to throw serious doubts that this species at least of Notacanthus lives at a great depth.' The evidence from radiographs indicates that the skeleton of N. phasganorus is sub- stantially similar, and its gas-bladder is better developed than in oceanic fishes. But from its diet and the related structural adaptations it is clearly a bottom-feeding form, and it is therefore probable that specimens taken have been obtained on or near the bottom, so that a bathymetric distribution of 100 to at least 420 fathoms may be deduced from the records so far available. N. mediterraneus Fil. & Ver. is evidently another bottom-feeding form ; Vaillant (18886) records hexactinellid sponge spicules from a specimen taken by the Talisman from more than 1,200 metres. Actiniarians have been reported as of frequent occurrence in Cod stomachs obtained from Bear Island and the Murman coast (Brown & Cheng, 1946) ; off Greenland, where Cod from deep water off Nuk feed almost entirely upon them (Jensen & Hansen, 1931), and in Danish waters (Blegvad, 1916). Stephenson, in Brown & Cheng, loc. cit., provisionally identified their material as Hormathia digitata (O. F. Mull.), H. nodosa (Fabr.), and Tealia felina (L.) var. lofotensis (Dan.). Some of our material may be referable to Hormathia spp., but precise identification would be extremely difficult if indeed possible. PARASITES The gills, alimentary canal, and peritoneum lining the body-cavity have been examined for parasites, but none have been found. REFERENCES BIGELOW, H. B., & SCHROEDER, W. C. 1935- Two Rare Fishes, Notacanthus phasganorus Goode and Lycichthys latifrons (Steenstrup and Hallgrimsson) , from the Nova Scotian Banks. Proc. Boston Soc. Nat. Hist. 41: 13-18. BLEGVAD, G. 1916. On the Food of Fish in the Danish Waters within the Skaw. Rep. Danish Biol. Sta. 24: 19-71. BLOCK, M. E. 1787. "Cber zwei merkwiirdige Fischarten (Notacanthus chemnitzii und Silurus militaris). Abh. Bohm. Ges. Wiss. 3: 278-282. 1795- N aturgeschichte der ausldndischen Fische, 9, Atlas: pi. 431. Berlin. BREDER, C. M. 1943. The Eggs of Bathygobius soporator (Cuvier and Valenciennes) with a dis- cussion of other non-spherical Teleost Eggs. Bull. Bingham Oceanogr. Coll. 8 (3) : 1-49. BROWN, W. W., & CHENG, C. 1946. Investigations into the Food of the Cod (Gadus callarias L.) off Bear Island, and of the Cod and Haddock (G. aeglefinus L.) off Iceland and the Murman Coast. Hull Bull. Mar. Ecol, 3, No. 18: 35-71. ON NOTACANTHUS PHASGANORUS GOODE 79 CUVIER, G., & VALENCIENNES, A. 1831. Histoire Naturelle des Poissons, 8. Paris. CUVIER, G. 1836. Le Regne Animal ['Disciples' Edition']. Poissons. Paris. GAIMARD, P. 1851. Voyage en Islande et au Greenland. Zoologie, &c., atlas. Paris. GILL, T. N. 1883. Diagnosis of new genera and species of deep-sea fish-like vertebrates. Proc. U.S. Nat. Mus. 6: 253-260. GOODE, G. B. 1881. Notacanthus phasganorus. A new species of Notacanthide from the Grand Banks of Newfoundland. Proc. U.S. Nat. Mus. 3: 535-537- & BEAN, T. H. 1894. A Revision of the Order Heteromi, deep-sea fishes, with a description of the new generic types Macdonaldia and Lipogenys. (Sci. Res. Albatross Exped. XXIX.) Proc. U.S. Nat. Mus. 17: 455-470. 1896. Oceanic Ichthyology. Mem. Harv. Mus. Comp. Zool. 22. GUNTHER, A. 1887. Report on the Deep-Sea Fishes. Rep. Sci. Res. 'Challenger', Zoology, 22. JENSEN, A. S., & HANSEN, P. M. 1931. Investigation on the Greenland Cod (Gadus callarias L.). Rapp. Cons. Explor. Mer. 72. LtiTKEN, C. 1899. Danish Ingolf Exped. 2, Pt. I. 1-39. Copenhagen. MATSUBARA, K. 1938. Studies on the Deep-Sea Fishes of Japan. X. On a New Fish, Notacanthus fascidens, belonging to Heteromi with Special Reference to its Variations. Bull. Jap. Soc. Sci. Fish. 7: 131-136. REGAN, C. T. 1929. Fishes. [Article in] Encyclopaedia Britannica, i^th edn: 317. SAEMUNDSSON, B. 1949. Marine Pisces. Zoology of Iceland. Ed. A. Fridriksson & S. L. Tuxen. 4 (72). STEVEN, G. A. 1930. Bottom Fauna and the Food of Fishes. /. Mar. Biol. Ass. U.K. n.s. 16: 677-700. TODD, R. A. 1907. Second Report on the Food of Fishes (North Sea, 1904-1905). Rep. (Southern Area) Fish Invest. N. Sea, 1904-1905 (Mar. Biol. Ass.), 2 (i): 48-163. [Ed. 3837.] TROTTI, L. 1939. Contribute alia Conoscenza del Genere Notacanthus ed in particolare della Specie Bonapartei Risso. Ann. Mus. Stor. Nat. Genova, 60: 363-378. VAILLANT, L. i888a. Sur les rapports zoologiques du genre Notacanthus Bloch. C. R. Acad. Sci. 107: 75 T -753- Paris. 18886. Exped. Sci. 'Travailleur' et 'Talisman'. Poissons. Paris. WOODLAND, W. N. F. 1911. On the Structure and Function of the Gas Glands and Retia Mirabilia associated with the Gas Bladder of some Teleostean Fishes, with Notes on the Teleost Pancreas. Proc. Zool. Soc. Land. 1911: 183-238. 1 91 1 a. On some Experimental Tests of Recent Views concerning the Physiology of Gas Production in Teleostean Fishes. Anat. Anz. 40: 225-242. 1913. Notes on the Structure and Mode of Action of the 'Oval' in the Pollack (Gadus pollachius) and Mullet (Mugil chelo). J. Mar. Biol. Ass. U.K. 9: 561-565. PRESENTED i - NO\ 1S51 PLATE 7 FIG. 2. Notacanthus phasganorus Goode; Bear Island specimen. FIG. 3. Detail of right pelvic fin, from below. FIG. 4. A, underside of head; B, side, and C, D, front views of teeth of maxillary series; E. palatine tooth. Bull. EM. (N.H.) Zoology, I. 5 PLATE 7 CO P tf O O CO <; ffi P ffi H o O H O PLATE 8 FIG. 5. Unretouched X-ray photograph of head, showing flexure of vertebral column. Bull. B.M. (N.H.) Zoology, I. 5 PLATE 8 FIG. 5 NOTOCANTHUS PHASGANORUS PLATE 9 FIG. 6. X-ray photograph of origin of dorsal fin. FIG. 7. X-ray photograph of end of dorsal fin. I, II, &c., spines ; R, soft ray. FIG. 8. X-ray photograph of pelvic region, showing pelvic fins and girdle. AI, first spine of anal fin. FIG. 9. X-ray photograph of end of tail. (Figs. 2-4, scale indicated on drawing ; Figs. 5-8, x i ; Fig. 9, x 2.) Bull. B.M. (N.H.) Zoology, I. 5 PLATE 9 II I FIG. 6 FIG. 8 FIG. 7 FIG. 9 NOTOCANTHUS PHASGANORUS PRESENTED 9 - NOV 1951 PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATEY PRINTER TO THE UNIVERSITY 3 1 OCT 1951 OCEAN SUNFISHES (FAMILY MOLIDAE) A. FRASER-BRUNNER BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 6 LONDON : 1951 THE OCEAN SUNFISHES (FAMILY MOLIDAE) BY A. FRASER-BRUNNER Pp. 87-121; 18 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 6 LONDON : 1951 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is to be issued in five series, corresponding to the Departments of the Museum. Parts will appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. J, No. 6 of the Zoology series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued November ig5i Price Seven Shillings and Sixpence THE OCEAN SUNFISHES (FAMILY MOLIDAE) By A. FRASER-BRUNNER SYNOPSIS The relationships of the Molidae with other Plectognathi are briefly discussed. The movable lobe at the hind margin of the body, supported usually by migrant dorsal and anal rays but sometimes also by caudal rays centrally, is designated the ' clavus '. Three genera are recognized, assigned to two subfamilies. Masturus is shown to include two forms (treated as species but possibly the sexes of one). Evidence is presented to show that in this genus alone of the family some caudal rays are developed. Mola is shown to include two species, which are diagnosed and figured. Sexual dimorphism in Mola mola is described. Full synonymies are included. ON account of their curious form and the great size which they often attain, the fishes of the family Molidae, usually called Ocean Sunfishes, have attracted considerable attention from early times. A large and scattered literature exists concerning them, but although comparative studies have been made from time to time and their anatomy has received attention quite frequently, we are still far from a complete understanding of their relationships. This is mainly because all the species are rather rare, and their occurrence unpredictable, so that it is not possible to make an excur- sion for the express purpose of collecting specimens, as could be done with many other fishes, while the great size of most examples makes transportation and preservation a difficult problem. Consequently good comparative material is not easily available for study, and much reliance has to be placed upon published descriptions and figures. It is the purpose of the present work to draw attention to certain facts which have become apparent from a study of the literature, aided by the material in the national collection. My thanks are due to Mr. G. Palmer for his assistance in seeking out some of the references and checking a number of points in them. I am concerned here only with taxonomy within the family, since a full considera- tion of their relationship to other Plectognathous fishes will be included in a larger work upon the anatomy and phylogeny of the whole Order now in preparation. It can be pointed out here, however, that I have already indicated in an earlier paper (Fraser-Brunner, 1943), that the Molidae are not really as highly specialized as previously supposed. Their main peculiarity lies in the atrophy of the rear end of the vertebral column, resulting in a mechanical rearrangement of the median fin- structures closely resembling that seen in other fishes when the tail is amputated al an early age ; some interesting examples of this among Flatfishes have been given by Chabanaud (1935). The resemblance is not quite perfect, since with amputation the supporting bones of dorsal and anal fins are lost with the tail, whereas in the Molidae only the vertebral structures are lost. The later alis muscles of the trunk, deprived of their normal attachment, become inserted upon the deep muscles of the dorsal and anal fins, and progressively lose their identity in the genera Ranzania, Masturus, Mola. The result of this is that body- flexion is lost but the dorsal and anal fins gain in power, and the latter are therefore go THE OCEAN SUNFISHES (FAMILY MOLIDAE) the principal means of locomotion. The posterior parts of these two fins extend round the rear end of the truncated body to support a broad, stiff lobe which acts as a rudder. This has been called the 'pseudo-caudal' by Raven (1939 a), but this is not a very suitable term in my opinion ; any structure in this part of the body may be described as 'caudal', and even if 'pseudo-caudal fin' is used, this is not true for all the species, for I hope to demonstrate on a later page that remains of the true caudal fin are included in the structure in Masturus. For this rudder-like lobe at the end of the body in the Molidae I therefore propose using a new term, and throughout this paper it will be called ' the clavus ' (Lat. clavus, a rudder). Apart from these changes of form, all of which are demanded as mechanical consequences of the phyletic atrophy of the posterior part of the vertebral column, the Sunfishes resemble in their anatomy the more primitive of the Tetraodont fishes, and in one feature at least, the retention of the fourth gill, they are less modified even than those. They stand, therefore, near the main stem of the Tetraodonts, and attempts to derive them from the highly modified Diodontidae seem to me to be very far-fetched ; whatever resemblances the latter may show are more plausibly explained by the assumption that they are evolved from a Mo/-like type (before caudal atrophy) rather than the reverse. The Molidae show also some features in common with the Ostraciontoidea alone among Fleet ognathi, and indicate therefore the diver- gence of the Trunkfishes and Puffer-fishes from a common stock during their evolu- tion. In my classification of the Tetraodontoidea I expressed the view that only two genera of Molidae should be recognized. This was based on the belief, current at that time, that Ranzania, Masturus, and Mola were each represented by a single species, and since the latter two forms seemed to be more closely related to each other than to Ranzania, it appeared that this relationship would be better expressed by placing them together in the genus Mola. A more detailed examination of these fishes, however, has caused me to modify these views. Firstly, I find that there are two species of Mola in the limited sense one of world- wide distribution and the other apparently restricted to the Australasian region. Whitley (1931) recently revived the name Mola ramsayi Giglioli 1883 for the latter, but was apparently unaware of its distinguishing characters and assumed that all specimens from that region should be so named, whereas his main description appears to be of M. mola and the records show that both M. mola and M . ramsayi are to be found around Australasia. The type of M . ramsayi, a huge stuffed specimen, is in the British Museum (Natural History), and by a piece of good fortune one of our spirit- specimens belongs to that species, so that I have been able to make direct comparison with examples of M. mola of similar size. Secondly, a close study of the literature concerning Masturus lanceolatus, aided considerably by the excellent work of Gudger on this subject, reveals that two forms are included here also, though it is not certain that they are different species. More interesting is the apparent fact that in Masturus alone of the family a remnant of the caudal fin is included in the support of the clavus. In this and in its musculature it is a little less specialized than Mola, and it therefore now seems desirable to recognize THE OCEAN SUNFISHES (FAMILY MOLIDAE) 91 it as a distinct genus in order to express its relationship to the other genera more clearly. There has been much speculation in the past as to whether the rays supporting the clavus belong to the caudal fin or to the dorsal and anal, and even Gregory & Raven (1934), when describing the anatomy of M. mola, thought them to be caudal although their description and figure indicate that they are not (an error corrected by Raven in 1939). Apart from internal anatomy, the number of these rays is in most cases against the likelihood that they all belong to the caudal fin ; in most Plectognathi FIG. i. Diagram illustrating reduction of the caudal region in the Molidae. Persistent parts of the axial skeleton shown in black ; atrophied parts shown with broken line; the last interneural and interhaemal bones close in along the lines marked with arrows. (Based on Ryder, 1886, modified by reference to adult and larval forms.) there are only 12 caudal rays, exceptionally 13, and sometimes as few as 10. But in Ranzania and Masturus the clavus is supported by 20 or more rays, in Mola mmsayi by 16, and only in Mola mola by 12. These rays are, in the main, supported by elements which have all the appearance of belonging to the series of interspinous supports of the dorsal and anal fins, but have been rotated to lie roughly at right angles to the last normal vertebro-interspinous complex by the process which has already been suggested by Ryder (Fig. i). The skeletal supports of the clavus are accompanied by muscles which have split off from the inclinators of the dorsal and anal fins, and caudal muscles appear to have been lost with the posterior vertebral structures. Reduction of the caudal region can be shown to extend to the number of rays supporting the clavus. Thus in Ranzania, which has 18 remaining vertebrae, there are 22 rays of dorsal and anal origin in the clavus ; in Masturus and Mola, which have 17 vertebrae, we find the series : Masturus 14-18 (exclusive of caudal rays), Mola ramsayi 16, M. mola 12. Alongside this the form of the rays is of interest (see Barnard, 1935). In Ranzania 92 THE OCEAN SUNFISHES (FAMILY MOLIDAE) each ray in the clavus (except the outermost) is abruptly branched at its distal end (like those of the dorsal and anal lobes) and forms a fairly stiff fan-shaped support, closely apposed to those each side of it. In Mola this branched portion becomes hyperossified into a single plate or ossicle characteristic of the genus, the number and arrangement of these ossicles being of importance in specific diagnosis. In Masturus the rays seem never to be branched in adults, and are never ossified distally, but in young examples they may be branched at the tip like those of Ran- zania. This development can be seen by comparing the figures accompanying this paper. Between the rays in Ranzania lie elongate lobes of apparently collagenous material (shown in Fig. 3), and it is probably these which in Mola mola extend back between the widely spaced rays to form the lobes characteristic of the clavus of large speci- mens of that species. As a matter of interest, it may be remarked that Ranzania, Mola, and the Percomorphous family Carapidae (Fierasferidae) are the only fishes to which the term ' gephyrocercal' can properly be applied, as pointed out by Ryder when originally proposing the term. Raven has taken the view that Ranzania is the most specialized of these genera. I cannot agree with this. Its skeleton is much less de- generate than that of Mola, more strongly ossi- fied, and there are 18 or 19 vertebrae. The later alis muscles are still moderately developed, though inserted posteriorly on the m. dorsalis profundus ; the usual division of the dorsal por- tion into superior and inferior parts is still quite distinct anteriorly. I feel sure that Raven was mistaken in identifying the lateralis muscles as dorsal and anal depressors ; they insert on to the latter but are distinct. The gill-rakers are free and apparently functional as in more generalized fishes. Further, this species is not gigantic. It is not suggested, however, that Ranzania is completely representative of the ancestor of the other two genera ; it has retained more primitive features, but it has completely lost the caudal fin, whereas Masturus, which is otherwise a stage farther towards Mola, retains a vestige of this fin, as will be shown later. The relationships of these genera are therefore probably as shown in Fig. 2. An ancestral form in which the skeleton and musculature is still fairly normal and the caudal fin not completely lost gives rise to Ranzania on the one hand, which loses its caudal fin, and to Masturus on the other, in which the caudal fin retains a precarious hold but the skeleton and the musculature deteriorate. Further degeneration and complete loss of the caudal fin in this second line gives us Mola. The need to recognize Mastiirus and Mola as more closely related to each other FIG. 2. Relationships of the genera of the family Molidae. THE OCEAN SUNFISHES (FAMILY MOLIDAE) 93 than to Ranzania therefore still remains, and these two lines can now be expressed as subfamilies. The three genera illustrate quite well the manner in which the lateralis muscles lose their primary function of flexing the body and become successively more closely associated with the dorsal and anal fins, their added power enabling these fins to FIG. 3. Rumania laevis, adult. A specimen 515 mm. long, from Baltimore, County Cork, Ireland. become proportionately larger. The body is therefore held rigid, assisted in Ranzania by a carapace similar to that of Strophiurichthys among the Ostracionts, but with much smaller hexagonal plates; in Masturus and Mola this carapace is reduced to small denticles, and rigidity is assisted by a thick collagenous layer beneath the skin (Green, 1901). All the species pass through a remarkable metamorphosis. The newly hatched larvae are Tetraodon-like, but soon (at 1-8 mm.) develop a cuirass of broad plates with jutting triangular projections, looking more reminiscent of an Ostraciont (Richardson named this stage Ostracion boops). With the atrophy of the larval tail, Ranzania seems to pass, by reduction of the cuirass and elongation of the body, into 94 THE OCEAN SUNFISHES (FAMILY MOLIDAE) a form essentially like that of the adult though proportionately longer, but Masturus and Mola show an intermediate stage, wherein the cuirass breaks up into small denticles and the triangular projections grow into long sharp spines on broad poly- gonal grooved bases (very like those of Acanthostmcion or Lactoria). This stage is much shorter in the body than the adult. As growth proceeds the body lengthens and the spines shorten and disappear, though in Mola the bases of one on the snout and one at the chin are nearly always retained even in the largest specimens. KEY TO THE GENERA OF MOLIDAE I. Form comparatively elongate. Vertebrae 8+ 10 or n. Carapace of smooth hexa- gonal plates, 1 terminating at bases of dorsal and anal fins and clavus. Lips pro- duced forward beyond teeth as a funnel, closing as a vertical slit. Gill-rakers free. 2 uppermost branchiostegal rays coalesced. Clavus with 22 rays, all borne on interspinous bones. No secondary post-larval metamorphosis (subfamily RANZANIINAE) -. . . . i. Ranzania II. Form shorter. Vertebrae 9+8. Carapace collagenous; skin of body and clavus with small rough denticles. Lips not funnel-like. Gill-rakers concealed in thick skin. 6 distinct branchiostegal rays. A secondary post-larval metamorphosis (subfamily MOLINAE). A. Median rays of clavus not borne on interspinous bones, supporting a pronounced lobe; none of the rays bearing ossifications distally . . 2. Masturus B. All rays in clavus borne on interspinous bones, most of them terminating in an ossification distally ...... 3. Mola Genus RANZANIA Nardo ? Triurus Lacepede, 1800, Hist. Nat. Poiss. 2: 200. Type: Triurus bougainvillianus Lacepede. Ranzania Nardo, 1840, Ann. Sci. Regno Lombardo-Veneto, 5: 10, 105. Type: Ranzania typus Nardo (= Ostracion laevis Pennant). Centaurus Kaup, 1855, Arch. Naturgesch. 21 (i) : 221. Type: Ostracion boops Richardson (= Os- tracion laevis Pennant, young). The characters of this genus have been indicated concisely in the foregoing key. Lacepede 's description of Triurus bougainvillianus was based upon manuscript notes by Commerson. It could be interpreted as referring to the fish later known as Ranzania, but to describe the funnel-like lips as 'rictu fistulari' or ' le museau avance en forme de tube' and again 'un museau tres prolonge fait en forme de tube assez etroit ' requires a good stretch of imagination. Moreover, the depth of the body is given as the proportion of 18 against the body-length of 71, and no other Sunfish has been recorded as slender as that. Finally, it has to be noted that in vol. i of the same work Ranzania is figured (pi. 22) under the name 'le Tetrodon lune'. The status of the name Triurus is therefore doubtful, and I hesitate to follow Whitley in using it, particularly since the name Ranzania is so well known. A single species. 1 In young specimens (90 mm.) each plate has a prominent bony tubercle centrally. THE OCEAN SUNFISHES (FAMILY MOLIDAE) 95 Ranzania laevis (Pennant) Ostracion laevis Pennant, 1776, Brit. Zool, ed. 4, 3: 129, pi. 19. Tetrodon truncatus Retzius, 1785, K. Svenska Vetensk Akad. Handl. 6 (2) : 121 (based on Pennant) ; Lacepede, 1798, Hist. Nat. Poiss. 1: 514, pi. 22 f . 2 ; Donovan, 1808, Nat. Hist. Brit. Fish. 2: pi. 41. Orthragoriscus oblongus Bloch & Schneider, 1801, Syst. Ichth.: 511, pi. xcviii. Orthagoriscus oblongus Jenyns, 1835, Man. Brit. Vertebr. Anim.: 491; Yarrell, 1836, Hist. Brit. Fish. 2: 357, fig.; Couch, 1841, Ann. Mag. Nat. Hist. 6: 144; Bonaparte, 1846, Cat. Met. Pesci euv.\ 88; Bleeker, 1860, Natuurk. Tijdschr. Ned.-Ind. 21: 57; Couch, 1865, Hist. Fish. Brit. Is. 4: 381, pi. 246; Harting, 1868, Verh. Akad. Wet. Amst. : 12, pi. 2, fig. 2 ; Andrews, 1871, Proc. Nat. Hist. Soc. Dublin, 5: 123; Sauvage, 1891, Hist. Madagascar, 16 (Poiss.): 529; Nobre, 1935, Faun. Mar. Portugal, Vertebr.: 242. Cephalus oblongus Shaw, 1806, Gen. Zool. 5: 439, pi. 176; Swainson, 1839, Nat. Hist. Class. Fish. 2: 330. Cephalus varius Shaw, 1806, ibid. Orthragus commersoni Rafinesque, 1810, Caratt. Sicilia: 18. Orthragus oblongus Rafinesque, 1810, Indice Itt. Sicil.: 40. Tetraodon truncatus Couch, 1825, Trans. Linn. Soc. Lond. 14: 88. Cephalus elongatus Risso, 1826, Eur. Merid. 3: 173. Mola planci Nardo, 1828, Bull. Sci. Nat. Ferussac, 13: 437. Orthagoriscus truncatus Fleming, 1828, Hist. Brit. Anim.: 175; Giinther, 1870, Cat. Fish. Brit. Mus. 8: 319; Bleeker, 1873, Ned. Tijdschr. Dierk. 4: 121 ; 1879, Verh. Akad. Wed. Amst. 18: 26; Rochebrune, 1883-1885, Faune Senegambie (Poiss.): 157; Day, 1884, Fish. Gt. Brit.: 276, pi. 149 ; Beauregard, 1893, Bull. Soc. Sci. Nat. Quest, 3: 229 ; Scharff, 1906, Irish Nat. 15: 275 ; Mauro, 1906, Boll. Accad. Gioenia, Catania, N.S. 85: 16, fig. Cephalus cocherani Traill, 1832, Mem. Werner: 6. Orthragoriscus elegans Ranzani, 1839, Novi Comment. Acad. Sci. Inst. Bonon. 3: table. Orthragoriscus battarae Ranzani, 1839, ibid. Ranzania typus Nardo, 1840, Ann. Sci. Regno Lombardo-Veneto, 5: 105; Smith, 1949, Sea Fish. S. Afr.: 422, pi. 95, fig. 1212. Ostracion boops Richardson, 1844, Voy. Erebus and Terror, Fish.: 52, pi. 30, figs. 18-21 ; Giinther, 1880, Intro. Study Fish.: 175, fig. 93. Orthagoriscus planci Bonaparte, 1846, Cat. Met. pesci eur.: 88; Canestrini, 1872, Fauna d' Italia, Pesci: 149; Stossich, 1879, Boll. Soc. Adriat. Sci. Nat. 5: 36. Orthragoriscus lunaris (Gronow) Gray, 1854, Cat. Fish.: 165. Centaurus boops Kaup, 1855, Arch. Naturgesch. 21 (i): 221. Ranzania truncata Jordan & Gilbert, 1883, Bull. U.S. Nat. Mus. 16: 966; Trois, 1884, Atti 1st. Veneto, 2 (6) pt. i: 1269, pis. 12-14; P*- 2: J 543> pi- l6 ; Perugia, 1897, Ann. Mus. Stor. nat. Genova (2), 18: 140 ; Jordan & Evermann, 1898, Bull. U.S. Nat. Mus. : 47 (2) : 1755 ; Steenstrup & Liitken, 1898, K. danske vidensk. Selsk. Skr. (6) 9: 54, fig. ; Gunther, 1910, /. Mus. Godeffroy, 9 (17) : 477; Pellegrin, 1912, Bull. Soc. Zool. France, 37: 228, fig. i ; Ribeiro, 1915, Arch. Mus. nac. Rio de J. 17 (Molidae) : 4, pi. ; Thompson, 1918, Mar. Biol. Rep. Cape Town 4: 176; Buen, 1919, Bol. Pesc. Madr. 4: 295; 1935, Notas Inst. esp. Oceanogr. 2 (81): 146; Schmidt, 1921, Nature, Lond. 107: 76, figs. 2, 4, 5 ; Medd. Komm. Havunders0g. Fisk. 6 (6), fig. 2. 13, pi. i, fig. 7 ; Fowler, 1922, Copeia 112: 84 ; Vinciguerra, 1923, Comune di GenovaS: 5, fig. 3 ; Barnard, 1927, Ann. S. Afr. Mus. 21: 989, fig. 32; Fowler, 1928, Mem. Bishop Mus. 10: 475; Schmidt, 1932, Dana's Togt omkr. Jord.: 251, fig. 197 (6-n); Gudger, 1935, Nature, Lond. 135: 548; Barnard, 1935, Ann. S. Afr. Mus. 30: 655, fig. 6c; Ehrenbaum, 1936, Handb. Seefisch. Nord- europ. 2: 88, fig. 69; Gudger, 1936, Nature, Lond. 137: 947; Fowler, 1936, Bull. Amer. Mus. Nat. Hist. 70 (2): 1123, fig. 470; Ninni, 1939, Atti. Soc. Ital. Sci. nat. 78: 236; Raven, 1939, Amer. Mus. Novit. 1038, figs. 1-3; Clark, 1949, ibid. 1397: 7, fig. 9; Maul, 1949, Vertebr. Madeira 2 (Peixes) : 158. Ranzania makua Jenkins, 1895, Proc. Calif. Acad. Sci. (2) 5: 780, pi. ; Fowler, 1900, Proc. Acad. ZOOL. I. 6 M 9 6 THE OCEAN SUNFISHES (FAMILY MOLIDAE) Nat. Sci. Philad.: 514; Jordan & Snyder, 1901, Proc. U.S. Nat. Mus. 24: 262; Jenkins, 1902, Bull. U.S. Fish. Comm. 22: 486 (1903) ; Jordan & Evermann, 1905, Bull. U.S. Fish. Comm. 23 (i): 439, fig. 194; Jordan, Tanaka, & Snyder, 1913, /. Coll. Sci. Tokyo 33: 231, fig. 166; Snyder, 1913, Proc. U.S. Nat. Mus. 44: 460, pi. 63; Tanaka, 1914, Fig. Descr. Fish. Japan 16: 274, pi. 76; Jordan & Jordan, 1922, Mem. Carneg. Mus. 10 (i) : 89; Jordan, Evermann, & Tanaka, 1927, Proc. Calif. Acad. Sci. 16 (4): 680. Orthagoriscus (larva) Sanzo, 1919, Mem. R. Com. Talassogr. ital. 69: 1-7, figs. 1-4. Ranzania laevis Whitley, 1933, Viet. Nat. 49: 211, figs. 6, 7 ; Phillips, 1941, Trans. Proc. Roy. Soc. N.Z. 71 (3) : 245, pi. 41, fig. 6; Deraniyagala, 1944, /. Bombay Nat. Hist. Soc. 44 (3) : 429. Triurus laevis Whitley, 1937, Mem. Queensland Mus. 11 (2): 147; Hale, 1944, 5. Aust. Nat. 22: pt. 4, pi. i, figs. FIG. 4. Ranzania laevis. Front view of head, showing mouth open and closed. Examination of the records leaves little doubt that a single species of Ranzania ranges the seas of the whole world except the polar regions, but it seems that two subspecies can be recognized as follows: Ranzania laevis laevis (Pennant). Depth of carapace contained twice or more in its length, in adults (up to 580 mm.). Axil of pectoral fin well below level of centre of eye. Height of anal fin less than f length of head. Atlantic Ocean. R. i. makua Jenkins. Depth of carapace contained less than twice in its length, in adults (400-500 mm.). Axil of pectoral fin above level of centre of eye. Height of anal fin -f length of head or more. North Pacific Ocean. That these two forms are simply subspecific extremes in the range is shown by the records from the Indian Ocean, wherein the depth is usually given as for makua while THE OCEAN SUNFISHES (FAMILY MOLIDAE) 97 the pectoral fin is low as in laevis. A specimen from Mauritius in our collection shows these features well, and a closely similar specimen has been figured by Whitley from Australia. Whenever the coloration has been described it has been shown to be closely similar in all these forms, a pattern of pale transverse bands on a darker ground, the bands edged with spots and broken lines of black ; three bands associated with the eye are FIG. 5. Development of Ranzania laevis. A, larva (1-7 mm.) ; B, C, D, early, full, and late 'Ostracion hoops' stage; E, transition to adult form (8 mm.). (After Schmidt.) the most constant, the posterior ones being variously broken or anastomosed, some- times enclosing large oval areas of the ground colour. The colours are said to be very brilliant in life. The mouth is very curious, the lips extending well beyond the teeth and forming a funnel, the mouth being oval with the long axis vertical. The orifice closes along this axis, so that the rictus is really vertical apparently unique among fishes (Fig. 4). This feature was shown clearly in the earliest published picture of the fish (Aldro- vandi, 1613), a remarkably good representation for its period. Too little is known of the feeding habits to show whether they can be associated with the peculiar mouth, but the species has been reported to take littoral algae 9 8 THE OCEAN SUNFISHES (FAMILY MOLIDAE) (Barnard, 1927), and it is possible that the lips can suck in and close upon a frond while the teeth nip it off. The fine developmental series given by Schmidt (1932) shows that Raven was correct in supposing that elongation of the body is secondary, but it also shows that Ranzania is never so greatly shortened as the other two genera (Fig. 5). Lengthening occurs after the 8-mm. stage, until at 53 mm. the length of the carapace is about 3 times its depth. This proportion is maintained up to go-mm. size, and after that the depth of the body increases with growth, so that at 250 mm. the length of the carapace is 2-5 times its depth, at 430 mm. 2-25 times, at 515 mm. 2-1 times, and at 580 mm. only twice. These figures are for the Atlantic form as shown by our specimens ; in the North Pacific subspecies R. makua either the early lengthening is not so great or the later deepening is more rapid. The general use of the name truncatus for this species seems to date from Giinther, 1870 ; it is not clear why he chose this name rather than that of Pennant, on whose work that of Retzius was based, but possibly it was due to the fact that Pennant's description was numbered 54, while on his plate the number 54 appears beside a figure of the ' Short Diodon ' (Mola mold] , leaving the other Sunfish without a number. As both description and figure are titled 'Oblong Diodon', however, this is clearly an error in numbering, and there can be no doubt as to the identity of Ostracion laevis, which antedates Tetrodon truncatus by nine years. Ranzania laevis does not reach so enormous a size as the other members of the family, apparently not exceeding 800 mm. in length. It has been recorded from all warm seas, as far north as Scandinavia and far south as New Zealand, usually from single specimens though it was once observed in great numbers at the surface of the water off Martinique (Pellegrin, 1912). As Schmidt has pointed out, most records of larval Sunfishes to date belong to this species, and he has given us a fairly complete picture of its development from egg to adult. Genus MASTURUS Gill Masturus Gill, 1884, Proc. U.S. Nat. Mus. 7: 425. Type: Orthagoriscus oxywopterus Bleeker. The study of this genus has been greatly facilitated by the careful bibliographical work of Gudger, who studied the records over a number of years, added several new ones, and finally in 1937 published a work dealing with the structure of the caudal region and another summarizing the knowledge of the genus up to that date. The latter two works are of great value, and the remarks which I make in the pages which follow are based largely upon them and should be considered with constant reference to them. The distinctness of Masturus from Mola had already been acknowledged by Steen- strup & Liitken (1898) , and discrimination between the post-larval forms was achieved by Schmidt (1921). The secondary post-larval stage of Masturus is characterized by enormous elongation of the 'cornicles' (Fig. 6). But it remained for Gudger to dis- entangle the confusion in the literature, and it is no doubt because these necessitated a chronological arrangement of his data that he was unable to recognize the two forms involved. But an analvsis of the records leaves little room for doubt that there THE OCEAN SUNFISHES (FAMILY MOLIDAE) 99 are indeed two forms, one the generally accepted M. lanceolatus (Lienard), the other apparently taking M. oxyuropterus (Bleeker) as the earliest name. These will be diagnosed on a later page, but it is necessary to note their existence at this point in order to clarify the discussion which follows. I must stress now, however, that they are regarded here as species only because we have no knowledge to the contrary, but I suspect that they may prove to be the sexes of one species. Not one of the FIG. 6. Post-larvae of Masturus. A, 'Ostracion hoops' stage (2-8 mm.); B, ' M olacanthus ' stage (5 mm.). (After Schmidt.) specimens so far recorded has been sexed. Raven, the only person to make a dissec- tion, does not even mention the gonads. Masturus is peculiar among the Molidae in the possession of a pronounced lobe a little above the centre of the clavus. Gudger continually stressed the dorsal situation of this lobe, apparently as evidence that it could not be the remains of the larval tail ; this is not a very good point, for his own anatomical figures show that the lobe is associated with the end of the vertebral column. In other Plectognathi the vertebral column lies dorsally until it enters the caudal peduncle, where it lies approximately in the central long axis of the body. The fact that in the Molidae the vertebral column is dorsally placed at its hind end is therefore interesting as a further demonstration that the posterior part of the column is lost. It is my belief that the lobe on the clavus of Masturus can truly be called the ioo THE OCEAN SUNFISHES (FAMILY MOLIDAE) 'caudal lobe', for all the illustrations of its anatomy so far given seem to demonstrate that the slender rays supporting it are caudal rays. The first to be published was that by Ryder, after a drawing by Putnam ; it was reproduced by Gudger, and is now copied as my Fig. 7 A. It is interesting in that the dorsal and anal rays of the clavus are shown branched, a feature shown only once elsewhere in the literature (Gudger, 1939), perhaps because the tips are so often broken off in young specimens. They are INN B FIG. 7. Caudal skeletons of A, Masturus oxyuropterus, copied from Ryder, 1886; B, Masturus lanceolatus, copied from Gudger, 1937. c, caudal rays; INN, interneural bones; INK, interhaemal bones. thus distinguishable at a glance from the simple caudal rays in the middle, but it is probable that the outermost two of the latter are also of dorsal and anal origin, for they have each a small skeletal support. The interspinous bones supporting the clavus are shown completely fused with the hindmost remaining haemal spine. Comparison with other dissections shows that these must in fact have been distinct elements. The shape of the supporting bones of the dorsal and anal fin lobes is obscured by the inclinator muscles in this figure, but the drawing of these muscles is interesting in helping to show their character after the later alis mass has been removed. Each of the rays in the clavus is supported by an interspinous bone, with the excep- tion of the middle four ; these are associated with the last of the remaining vertebrae, THE OCEAN SUNFISHES (FAMILY MOLIDAE) 101 which has no neural or haemal elements. There is no apparent reason why, if they also are dorsal and anal rays, they should not have their supports ; but if they are caudal rays they cannot be expected to be borne on hypurals, since these and other posterior vertebral elements have been lost. The presence of only four of these un- supported rays and the equal length of the dorsal and anal fin bases shows that Putnam's fish was a Masturus oxyuropterus ; two other dissections of this form have been illustrated that byGudger (1937 a, p. 41, fig. 27), and that by Raven (19396). The first does not show the internal skeleton, and one of the caudal rays is doubled, as can be seen by the nature of its basal cartilage ; but Raven's illustration, drawn by Helen Ziska, is admirable, and agrees in all essentials with that by Putnam. The only illustration showing the anatomy of M. lanceolatus is that given by Gudger in the work just quoted, based upon a young specimen (the same size as Putnam's) which was stained with alizarin and cleared. During the staining process some of the elements, notably the interneural supports of the clavus, were displaced, but I am satisfied that nothing was lost. This illustration is copied here as my Fig. 7 B. Here it will be seen that the central lobe of the clavus is supported by eight rays whose only skeletal support is the last vertebra (which has no neural or haemal elements) . Above these are five rays which can be associated with the five interneural bones which have been displaced from the horizontal during preparation. Below them are nine rays which belong to nine interhaemals, the lower three of which have been dis- placed forward. The presence of eight rays in the caudal lobe of the clavus and the greater length of the base of the dorsal fin lobe as compared with that of the anal fin lobe shows this to have been a specimen of Masturus lanceolatus as identified by Klunzinger (a figure of whose specimen is given by Gudger). It is admittedly hazardous to speak of caudal rays when the hypural bones are lost, since in normal fishes caudal rays are distinguishable only by their association with the hypurals. But I feel convinced that these central rays of the clavus in Masturus are homologous with the hypocaudal rays of the more generalized forms, and it remains for me to suggest how it is possible for them to persist although their skeletal supports are lost. It has to be borne in mind that two opposing forces are involved during the de- velopment of the caudal region, interacting in different proportion at successive stages. First there is the reduction of the larval tail and the atrophy of the posterior vertebral elements, and secondly the normal growth of body and fins. The first process evidently begins at an early age, for Schmidt has figured a larval specimen in which, as Gudger has pointed out, dorsal and anal rays are present but not caudal rays ; development of the latter is retarded. To see how this fact may affect later stages it is necessary to consider what occurs in the Triacanthodidae, the most primitive family of Plectognathi. In the larval Triacanthodid (Fig. 8), the caudal rays are twelve in number, as in most Plectognathi, but the last four lie in relation to the end of the notochord, which will later become ossified as the urostyle; the anterior eight belong to the last few myotomes. Degeneration of the tail from the rear will mean that the end of the notochord is lost first, and if this occurs before the hypocaudal rays appear not more than eight of them will develop. The eight slender rays of Masturus lanceolatus thus become intelligible and significant, while the 102 THE OCEAN SUNFISHES (FAMILY MOLIDAE) presence of only four in M. oxyuropterus suggests that reduction has proceeded still farther before caudal rays begin to develop in this form. Comparison with the larval Mola figured by Sanzo (1939) is interesting in this connexion, for it will be seen that if in his specimen hypocaudal rays were developed, they would not be associated with myotomes, and this probably accounts for their absence in that genus. As the caudal rays become stronger the axial structures decrease rapidly, so that by the time the rays are brought to the homocercal position there are no hypurals for their support, nor neural or haemal elements for the last few vertebrae ; but normal FIG. 8. Caudal region of post-larval Triacanthodid fish, showing relationship of hypocaudal rays to notochord and myotomes. body growth has extended the posterior parts of the dorsal and anal fins with their supporting structures backward and downward to fill the void. This process is probably correctly demonstrated by Ryder's diagram, upon which mine is based (Fig. i), in which the region of atrophy is delineated by the broken line. The vertebrae with their neural and haemal arches and spines are lost, but the interneural and interhaemal spines develop in relation to the fins in the normal manner except that they ultimately become tilted nearly at right angles to the last developed vertebral elements (Fig. 7). The number of these interspinous bones does not give a reliable estimate of the number of vertebrae that have been lost, because reference to the dissections shows that more than one may be associated with each neural or haemal spine, while of course the last few vertebrae are probably not associated with inter- spinous bones at all. Ryder, of course, thought Putnam's young fish was a Mola and that the caudal rays were completely lost in the adult. A curious feature of the posterior migration of the dorsal and anal fins is that, while in the lobes the rays are more numerous than the interspinous bones, each of those in the clavus has its own supporting element ; this might be taken to indicate that the central rays, which I THE OCEAN SUNFISHES (FAMILY MOLIDAE) 103 regard as caudal, are simply the last dorsal and anal rays, which are therefore more numerous than their supports in this region also, but there seems no good reason why the odd rays should all be crowded at the end. A more difficult argument to combat is that the supporting elements of these last rays cannot develop because of the presence of the vertebral column ; against this I can only point out that in Putnam's fish two elements lie behind the last vertebra, and there seems no reason why, if the odd four rays were in the same series, their supporting bones should not be there also. In fact, the presence of the two elements mentioned is reminiscent of the condition INT FIG. 9. Masturus oxyuropterus, late post-larva (21 mm.), in British Museum collection. B, dis- section of same specimen, showing presence of air-bladder. AB, air-bladder; H, heart; K, kidney; L, liver; INT, intestine; SPL, spleen. shown in Cyema atrum by Trewavas (1933), who identified the two small ossifications as hypurals. But since the last caudal vertebrae are so obviously lost in the Molidae it would be incautious to speak of hypurals here. McCulloch has left us drawings of very young examples of both M. lanceolatus and M. oxyuropterus, at the stage when the larval tail is not quite lost, the small peduncle bearing its allotted quota of caudal rays and the dorsal and anal fins extending round to meet them. Knowing what a careful observer and excellent draughtsman McCul- loch was, I am prepared to accept these as good evidence. Eventually, at the better known stage of 50 mm. or thereabouts, there is no sign of the original tail, but the ' caudal rays project beyond the rest of the clavus as the basis for the ultimate central lobe. Gudger believed that even these central rays were lost, at what he called the ' square-tailed ' stage, but as this was based on the two obviously damaged specimens of Steenstrup & Liitken, this seems to be improbable a point which Dr. Gudger himself has conceded in a letter to me. As a matter of interest I may mention here that in these small specimens it appears that the air-bladder is still present ; one which I dissected (Fig. 9 B) had a very ZOOL. i. 6 N io 4 THE OCEAN SUNFISHES (FAMILY MOLIDAE) delicate, bubble-like structure at the centre of mass, which unfortunately collapsed while I was examining it. At this planktonic stage in its development such an organ is not surprising, and of course the Molidae are evolved from fishes in which the air- bladder is well developed, but it is worth noting that the statement that an air-vessel is absent in this family is probably true only of adults. FIG. 10. Masturus lanceolatus, adult. Singapore. (After Smedley, 1932.) With the development of the skeletal structures (poorly ossified though they are) atrophy proceeds no farther, and the processes of growth produce what later changes we can observe in the fish. An extension of the dermis and its collagenous sub- stratum, probably that which would develop over the caudal peduncle in a more normal fish, eventually covers the caudal lobe and the whole clavus. These are what seem to be the main features in the development of the clavus of Masturus, but there is a certain amount of individual variation. In M. lanceolatus the presence of eight caudal rays seems to be fairly consistent, but the middle ones are sometimes represented only distally whether their proximal ends atrophy in the early stages or degenerate later is not evident. In M. oxyuropterus four caudal rays THE OCEAN SUNFISHES (FAMILY MOLIDAE) 105 are usual, but may occasionally be five (as in a dissection figured by Gudger, wherein one of the rays had split or doubled as shown by its supporting cartilage) or rarely three. On published evidence the number of rays supporting dorsal fin, clavus, and anal fin respectively appear to differ very greatly, but most of these are of doubtful value, for an accurate count can only be made by dissection (except possibly in stuffed specimens) . As an example of this may be quoted the description by Gudger, a careful worker, of the specimen he obtained for the American Museum of Natural History. In this he counted, on external examination, 'D.-f-C.+A. complex = 60' ; the fish is in other respects M. oxyuropterus , so that this high count would cast doubt on its distinctness from M. lanceolatus. But later Raven dissected this same fish, and his illustration shows distinctly the total of fifty-five rays which is usual in M. oxyuropterus. Consequently it has not seemed expedient to give any definite statement of the number of rays to be found in dorsal and anal fins and clavus respectively, but only to indicate the total number, which seems to be characteristic for each species. Whether or not I am correct in calling them caudal rays, the presence of median rays unsupported by interspinous bones is characteristic of Maslurus. In the adults all the rays of the clavus are simple, without distal ossifications. There is always a median projection to the clavus, and the body is rather more elongate than that of Mola, especially in the early stages. Osseous tubercles, the remains of post-larval spines, seem never to be retained anywhere on the body of the adult. Two forms can be recognized, treated here as species, but I suspect that further study will show them to be the sexes of one. They have been taken in the same localities, and sometimes together. The sexual dimorphism found to be present in Mola mola (p. 117) lends support to this idea. But with no knowledge of the sex of any recorded individual I can but state their characteristics and apply available names to them pending further information. Since all the literature before 1939 has been fully quoted and discussed by Gudger, I have not thought it necessary to repeat it all below, particularly as a number of records cannot be assigned with certainty, but full and discriminating reference has been made to Gudger's papers. KEY TO THE SPECIES OF MASTURUS I. Profile of lower jaw more convex, usually projecting beyond the upper. Upper profile of head evenly convex. Base of dorsal lobe conspicuously longer than that of anal fin. Dorsal and anal fins and clavus with a total of 60 to 62 rays. Caudal lobe of clavus 1 longer than head in perfect specimens (often mutilated), sup- ported by eight (rarely 7 or 9) rays I. lanceolatus II. Profile of lower jaw less convex, straight or concave, not projecting beyond the upper. Upper profile of head with distinct concavity above the eyes. Bases of dorsal and anal fin lobes about equal. Dorsal and anal fins and clavus with a total f 55 to 57 rays. Caudal lobe of clavus shorter than head, supported by 4 (rarely 3 or 5) rays 2. oxyuropterus The uncertainty of authors as to where dorsal and anal fins end and clavus begins, 1 Measured from ' hinge ' of clavus to tip. io6 THE OCEAN SUNFISHES (FAMILY MOLIDAE) and the obvious inaccuracy (already mentioned) of fin-ray counts made upon external examination, militates against giving counts for individual fins, but it may be, as suggested by Gudger's cleared specimen, that in M. lanceolatus there are more FIG. ii. Masturus oxyuropterus, adult. Tahiti. (After Gudger, 1935.) rays in the dorsal lobe than in M. oxyuropterus. In the latter, on the other hand, the number of claval rays supported on interhaemal bones seems to be greater (10 to 12) than in M. lanceolatus (9). Although usually these types seem to be recognizable at an early age, there are some doubtful cases among young specimens, as might be expected if they were the THE OCEAN SUNFISHES (FAMILY MOLIDAE) 107 sexes of one species. For example, if M. lanceolatus should be the male, it might be more like the female (M. oxyuropterus) when young, as in many other fishes, and in fact small examples of the oxyuropterus type seem to be the more common. Masturus lanceolatus (Lienard) Orthagoriscus lanceolatus Li6nard, 1840, Revue ZooL: 291; 1841, Magasin Zool. (2) 3 (Poiss.) : pi. 4. Orthagoriscus mola Klunzinger, 1871, Verh. Zool.-Bot. Ges. Wien, 21: 648; Giinther, 1880, Introd. Stud. Fish.: 175, fig. 94; Perugia, 1881, Ann. Mus. Star. Nat. Genova 27: 365, fig. Mola mola Collett, 1896, Result. Camp. Sci. Monaco 10: 163, pi. 6, fig. i. Ranzania truncata Steenstrup & Ltitken, 1898, K. danske vidensk. Selsk. Skr. (6) 9: pi. 6, fig. C. (Not of Jordan & Gilbert, 1883.) Mola (Molacanthus) sp. McCulloch, 1912, Proc. Linn. Soc. N.S.W. 37 (3) : 553, pi. 58. Mola lanceolata Schmidt, 1921, Medd. Komm. Havunderssg. Kbh., Fisk. 6 (6): pi. i, figs. 4, 5; 1932, Dana's Togtomkr. Jord.: 255, fig. 167 (part.); Barnard, 1927, Ann. S. Afr. Mus. 21: 987, fig. 31 ; Ehrenbaum, 1936, Handb. Seefisch. Nordeurop. 2: 88; J. L. B. Smith, Sea Fish. S. Afr.: 422, fig. 1214. Masturus lanceolatus Hubbs & Giovannoli, 1931, Copeia, 1931: 135; Gudger, 1935, Amer. Mus. Novit. 778: i, fig. i ; Gudger & McDonald, 1935, Sci. Mon. 41: i, figs. 4-9, n, 14, 15 ; Rivero, 1936, Amer. Nat. 70: 92, fig.; Palmer, 1936, Science, 83: 597; Gudger, 1937, Ann. Mag. Nat. Hist. (10) 19: 9, fig. 6; 15, fig. 10; 31, fig. 19; 33, fig. 20; 34, fig. 21 ; 38, fig. 23; Proc. Zool. Soc. Lond. 107 (A) (3) : 353 (part.), text-figs, i, 2, 4, 6, 7, 8, 9, 14 ( ?), 16, 20, 21 ( ?) ; pi. i, figs. 3, 4 ; pi. 2, figs. 5, 6 ; pi. 4, fig. 10 ; 1939, /. Elisha Mitchell Sci. Soc. 55 (2) : 305 ; Brimley, 1939, ibid. 295, pi. 28; Fitch, 1950, Calif. Fish Game, 36 (2) : 65. Lienard's figure cannot be said to be notable for its faithful representation, but as it shows the caudal lobe indubitably much longer than the head and the base of the dorsal fin longer than that of the anal, it shows to which of our two forms the name is applicable. The fin-rays are always more easily seen in dried specimens of these fishes, and so Klunzinger's stuffed example shows the structure of the clavus very well ; it is closely similar (in this 65-in. example) to that of Gudger's 53-mm. cleared specimen. Other figures in Gudger's papers which appear to represent this form are stated in the above synonymy. Where the caudal lobe is mutilated or otherwise doubtful the broad dorsal base and rather pugnacious-looking 'chin' are the most useful distinguishing characters. It grows to a great size, the largest recorded specimen being 10 ft. long and n ft. 3 in. from the tip of dorsal to tip of anal fins. In our collection it is represented only by the post-larval specimen figured by Giinther. Recognizable records of adults are from the Atlantic, off Florida, Havana, North Carolina, and Table Bay, South Africa, from the Red Sea, and from the Pacific at Tahiti. Young specimens have been taken off Alabama, Teneriffe, and in the South Seas. As this paper goes to press Fitch (1950) states that 100 post-larvae J to 2 in. in length have been taken from the stomachs of tuna in Hawaiian waters. Masturus oxyuropterus (Bleeker) Orthagoriscus spinosus Gatchet, 1832, Act. Soc. Linn. Bordeaux 5: 253. (Not of Cuvier, 1817.) Orthagoriscus oxyuropterus Bleeker, 1873, Versl. Akad. Amst. (2) 7: 151, fig. io8 THE OCEAN SUNFISHES (FAMILY MOLIDAE) Mola rotunda Ryder, 1886, Rep. U.S. Fish. Comm. (1884): 1027, pi. 8, fig. 5. (Not of Cuvier, 1798.) Ranzania truncata Steenstrup & Liitken, 1898, K. danske vidensk. Selsk. Skr. (6) 9 (i) : 98, pi. 6, figs. D, E. (Not of Jordan & Gilbert, 1883.) Mola (Molacanthus) sp. McCulloch, 1912, Proc. Linn. Soc. N.S.W. 38 (3): 553 (part.), pi. 59. 9 FLORIDA . FLORIDA. ATLANTIC FLORIDA --J^- N. CAROLINA. ^"~ TABLE BAY FLORIPA ' RED SEA _ FLORIDA MIAMI, FLA FLORIDA FLORIDA AMBOINA N CAROLINA FEET and inches FIG. 12. Diagram showing comparative ranges of size for the two species of Masturus, based on recognizable records of adult specimens. Mola mola Townsend 1918, Bull. N.Y. Zool. Soc. 21: fig. (not of Linnaeus) ; Collett, 1896, Result. Camp. Sci. Monaco, 10: 163 (part.) pi. 6, fig. i. Pfc Mola lanceolata Schmidt, 1921, Medd. Komm. Havunders0g. Kbh. Fisk. 6 (6) (part.}: pi. i, fig. 6; Smedley, 1932, Bull. Raffles Mus. 7: 17, pi. Masturus lanceolatus Jordan & Jordan, 1925, Mem. Carneg. Mus. 10: 89, fig. 7; Gudger & McDonald, 1935, Sci. Mon. 41: i, figs. 3, 10, 12, 13 ; Gudger, 1935, Copeia, 1935: 35, figs, i, 2 ; 1937, Ann. Mag. Nat. Hist. (10) 19: i (part.), text-figs. 18, 22, 26, 27; 1937, Proc. Zool. Soc. Lond. 107 (A) (3): 353 (part.}, text-figs. 5, 10, 12, 13, 15, 18 (?), 19, 22, pi. i, figs, i, 2, pi. 3, fig- 9 ( ?), pl- 4, fig. ii, pi. 5, fig. 17; 1939, / Elisha Mitchell Sci. Soc. 15 (2) : 305 (part.), figs. THE OCEAN SUNFISHES (FAMILY MOLIDAE) 109 1-5 ; Brimley, 1939, ibid. 300, pi. 29 ; Raven, 1939, Bull. Amer. Mus. Nat. Hist. 76 (4) : 143, pi. 2 ; Hardenberg, 1939, Treubia 17 (2) : 121 ; Clark, 1949, Amer. Mus. Novit. 1397: 7, fig. 9. A high proportion of the young specimens recorded seem to belong to this form, but a possible explanation of this is given on page 107. The small number of support- ing rays in the caudal lobe, the equal bases of dorsal and anal fins, and the compara- tively weak-looking 'chin' are recognizable even in McCulloch's lo-mm. specimen. The concavity of the dorsal profile of the head, however, is not noticeable in very small specimens, but it is already apparent in the 152-mm. example figured by Gudger (1939). The latter paper is also interesting in that it shows branching at the tips of the rays of the clavus, like that illustrated by Ryder (Fig. 7 A in this paper), but very much smaller, evidently in process of reduction. Only these two records of such branching exist, probably because the tips of the rays have been damaged in most small specimens, and the branching is lost with age. This form is so often taken in the same locality as the preceding that it is almost certainly a sex of that species ; in some instances young specimens of both forms have been taken from a single predatory fish (e.g. McCulloch (1912) , whose i3-mm. specimen is M.lanceolatus and his lo-mm. specimen M. oxyuropterus ; or Gudger (1939), whose i25-mm. fish is M. oxyuropterus, whereas at least the i27-mm. fish, and possibly the i30-mm. specimen also appears to be M. lanceolatus) . It will be noticed that in each case the M. oxyuropterus is slightly the smaller, and the records of this form do tend to lie about a lower range of size (Fig. 12). The largest record seems to be the 'Miami Masturus no. Ill' of Gudger, figured by Gudger & McDonald, though the identification of this badly slung specimen is a little doubtful. It was 7 ft. in length. Recognizable records of adults of this form are from the Atlantic at North Carolina and Florida, from Singapore and Amboina, and from the Pacific at Hawaii. Young specimens have been taken at Florida, the Sargasso Sea, the Azores, and in the South Pacific. A young specimen (Fig. 9) of unknown provenance is in our collection, and a plaster cast of the specimen dissected by Raven is exhibited in the fish gallery of the British Museum (Natural History). Genus MO LA Koelreuter Mola Koelreuter, 1770, Novi Comment. Acad. Petropol. 8: 337. Type: Mola aculeata Koelreuter ( Tetraodon mola Linnaeus, young). Orthragoriscus Bloch & Schneider, 1801, Syst. Ichth.: 510. Type: Tetraodon mola Linnaeus. Cephalus Shaw, 1804, Gen. Zool. 5: 437- Type: Tetraodon mola Linnaeus. Orthragus Rafinesque, 1810, Caratt. Sicilia: 17. Type: Orthragus luna Rafinesque ( Tetraodon mola Linnaeus). Diplanchias Rafinesque, 1810, ibid. Type: Diplanchias nasus Rafinesque. Orthagoriscus Cuvier, 1817, Regne Anim., ed. i, 2'. 149. Type: Tetraodon mola Linnaeus. Pedalion Swainson, 1839, Nat. Hist. Fish. 1: 199. Type: Pedalion gigas (Guilding) Swainson. Molacanthus Swainson, 1839, ibid. 2: 329. Type: Molacanthus pallasi Swainson (= Tetraodon mola Linnaeus). Ozodura Ranzani, 1839, Novi Comment, acad. Sci. Inst. Bonon. 3: 80. Type: Ozodura orsini Ranzani. Tympanomium Ranzani, 1839, ibid., table. Type: Tympanomium planci Ranzani. no THE OCEAN SUNFISHES (FAMILY MOLIDAE) Trematopsis Ranzani, 1839, ibid., table. Type: Trematopsis willoughbii Ranzani. Pallasina Nardo, 1840, Ann. Sci. Regno Lombardo- Veneto 10: 10, 112. Type: Pallasina pallasi Nardo (larval form). Acanthosoma De Kay, 1842, Nat. Hist. New York (Zoo/.) 3: 330. Type: Acanthosoma carinatum De Kay (= Tetraodon mola Linnaeus, young). Aledon Castelnau, 1861, Mem. Poiss. Afr. austr.: 75. Type: Aledon storeri Castelnau. Closely related to Masturus, but differing in that the clavris is supported entirely by elements from the dorsal and anal fins. The form of the body is relatively shorter, conspicuously so in the young, and the post-larval spines are not entirely lost, the base of one at the chin or one on the snout, or both, remaining as a low bony boss in the largest examples. Very few post-larval specimens of Mola have been found, but the smallest, 5 mm. long, shows that there is an 'Ostracion boops' stage, and several examples of the secondary post-larval or 'Molacanthus' stage have been described ; it is not known whether the 'cornicles' are ever as long as those of Masturus. Although a number of naturalists have believed in the existence of several species of Mola, and Ranzani went so far as to recognize five genera and eleven species, it has generally been believed, especially during this century, that only one widely dis- tributed species is admissible. My studies, however, show that while Mola mola is indeed wide-ranging, it is largely or entirely replaced in the South Pacific by a second species, distinguishable as follows : KEY TO THE SPECIES OF MOLA I. Clavus supported by about 16 rays, 12 of which bear ossicles ; the ossicles much broader than the spaces between them, and forming the margin of the clavus; those borne on paraxial rays separate, much smaller than the others. No band of reduced denticles between dorsal and anal fins . . . I. ramsayi II. Clavus supported by about 12 rays, 8 or 9 of which bear ossicles; the ossicles widely separated, invested with cuticle, which grows beyond them to form lobes in large examples ; those borne on paraxial rays united to form a single ossicle larger than all the others. A band of reduced denticles, smoother to the touch, at base of clavus from dorsal to anal fin . . . . .2. mola. The term 'paraxial rays' refers to the pair of supporting rays of the clavus the proximal ends of which lie nearest to the end of the vertebral column. The smooth band between dorsal and anal fins in M. mola is usually visible, marked by a fold posteriorly, and often differently coloured from the rest of the fish ; in doubtful cases the tips of the fingers will discern that this area is less rough than the body in front of it and the clavus behind it. Mola ramsayi (Giglioli) Orthagoriscus truncatus Hutton, 1872, Fish. New Zealand: 73. (Not of Fleming, 1828.) Orthagoriscus 'mola Castelnau, 1872, Proc. Zool. Acclim. Soc. Viet. 1: 211; 1875, Res. Fish. Austral.: 3; Hutton, 1873, Trans. Proc. N.Z. Inst. 5: 271; Macleay, 1875, Proc. Linn. Soc. N.S.W. 1: 12; Johnston, 1883, Pap. Roy. Soc. Tasm.: 137; 1891, ibid.: 38; Hamilton, 1886, THE OCEAN SUNFISHES (FAMILY MOLIDAE) in Trans. Proc. N.Z. Inst. 18: 135; Williams, 1893, ibid. 25: no, pi. 8 a; Drew, 1897, ibid. 29: 286 ; Parker, 1897, ibid. : 627 ; ? Fletcher, 1929, Proc. Linn. Soc. N.S.IV. 54: 225, 227. (Not of Cuvier, 1817.) FIG. 13. Mola ramsayi, adult, 2130 mm. long, New South Wales. (Drawn from the type of the species in the British Museum collection.) Orthragoriscus ramsayi Giglioli, 1883, Nature, Lond. 28: 315; Ramsay, 1883, Cat. N.S.W. Court. Intern. Fish. Exhib.: 43. ? Orthagoriscus eurypterus Philippi, 1893, Chilen. Fische: 15, pi. 6, fig. i (not seen). Mola mola Waite, 1907, Rec. Canterbury [N.ZJ] Mus. 1: 34; 1913, Trans. N.Z. Inst. 45: 223, pi. 9; 1921, Rec. S. Aust. Mus. 2: 198, fig. 332; 1923, Fish. S. Austral.: 230, fig. ; Phillips, 1919, ZOOL. I. 6 O 112 THE OCEAN SUNFISHES (FAMILY MOLIDAE) Rep. Dom. Mus. N.Z.: 6; 1926, N.Z. J. Sci. Tech. 8 (3): 169, figs. 1-3; McCulloch, 1922, Aust. Zool. 2 (3): 130, fig. 374 a; 1930, Mem. Aust. Mus. 5 (3): 436 (part.} ; Schneider, 1930, FIG. 14. Mola ramsayi, young adult, 410 mm. long, South Australia ( ?). (From specimen in spirits in the British Museum collection.) PO, paraxial ossicles. Rev. Chil. Hist. Nat. 34: 200, figs. 36, 37 ; Fowler, 1945, ibid. 45-47: 170, fig. ; Morrow & Mauro I 95 Copeia, 1950: 108, fig. 4 c. Mola ramsayi Whitley, 1931, Rec. Aust. Mus. 18 (3) : 126 (part.), pi. 16, figs. 3, 4. THE OCEAN SUNFISHES (FAMILY MOLIDAE) 113 All the New Zealand records, most of the Australian, and the few Chilean specimens appear to belong to this species, though in many cases it is not possible to be certain. It may be assumed, therefore, that in the South Pacific it replaces the wide-ranging M . mola. The two meet, however, in the Australian area, for Stead, McCulloch, and Whitley have all figured specimens which were undoubtedly M. mola, Whitley including his specimen with one of the true M. ramsayi in the same paper under the latter name. The type of Orthragoriscus ramsayi Giglioli is in the British Museum (Nat. Hist.). Its locality was given as 'Southern Hemisphere', but a label accompanying the specimen states 'New South Wales', and it is known to have been taken on that coast (fide Whitley, 1931). It was exhibited at the International Fisheries Exhibition in London in 1883 and later presented to the Museum by the Commissioners of the Exhibition. It is a very large stuffed skin, now in a rather dilapidated condition. The total length is 213 cm. (6 ft. 8 in.). We have, fortunately, a second specimen, in spirits much smaller, of course ; it is without a definite locality, but almost certainly from South Australia, since it was in a collection of specimens presented by the Zoological Society, several of which were typical South Australian species and all of which would be likely to occur there. It agrees very well with the excellent figure given by Waite (1923), and removes any doubt as to the distinctness of the species from M. mola. The type is not by any means the largest recorded specimen of M. ramsayi. That distinction apparently goes to one taken on 12 December 1889 in Poverty Bay, and recorded by Williams as measuring 9 ft. 8 in. and weighing 3^ tons. Mola mola (Linnaeus) Tetraodon mola Linnaeus, 1758, Syst. Nat. ed. 10, 1: 334; Pennant, 1776, Brit. Zool. 3: 131, pi.; Migliorini Spinola, 1843, Poiss. Genes: 14. Tetrodon mola Briinnich, 1768, Ichth. massil.: 8; Gmelin, 1778, Syst. Nat. Linn.: 1447; Retzius, 1785, K. Svensk. Vetensk. Akad. Handl. 6: 115; Bonnaterre, 1788, Tabl. Encycl. Meth.: 25, pi. 17, fig. 54; Lacepede, 1798, Hist. Nat. Poiss. 1: 509; Retzius, 1800, Fauna Suec.: 310; Donovan, 1803, Nat. Hist. Brit. Fish. 2: pi xxv. Mola aculeata Koelreuter, 1770, Novi Comment. Acad. Petropol. 8: 337. Diodon mola Pallas, 1777, Naturgesch. Thiere 8: 41, pi. 4, fig. 7; Bloch, 1785, Naturgesch. ausldnd. Fische 1: 75, pi. 128; Jacob. 1826, Dublin Phil. J. 2: 443, pi. Mola rotunda Cuvier, 1798, Tabl. Elem. Nat. Hist. : 323 ; Jordan, 1881, Proc. U.S. Nat. Mus.: 70; Jordan & Gilbert, 1883, Bull. U.S. Nat. Mus. 16: 865 ; Petersen, 1884, Vidensk. Medd. naturh. Foren. Kbh.: 159; Smith, 1885, W. Amer. Sci. 1 (7): 45; Linton, 1897, Proc. U.S. Nat. Mus. 19: 788, 812, 824; Steenstrup & Liitken, 1898, K. Danske vidensk. Selsk. Skr. (6) 9 (i): 28, pi. i ; Murray & Hjort, 1912, Depths of the Ocean: 119, 607, 615, 697, figs. 102, 507; Schmidt, 1921, Medd. Komm. Havunders0g. Kbh. Fisk. 6: i, figs, i, 5, 6, 10 b, 12, pi. i, figs, i, 2 ; 1926, Nature, Lond. 117: 80, figs, i, 2 ; Ehrenbaum, 1936, Handb. Seefisch. Nordeurop. 2: 86, fig. 68 ; Jensen, 1940, Vidensk. Medd. nat. Foren. Kbh. 104: 319. Orthragoriscus mola Bloch & Schneider, 1801, Syst. Ichth.: 510; Turner, 1862, Nat. Hist. Rev.: 185, pi. 6, figs. 4-6; Beneden, 1871, Mdm. Acad. R. Belg. 38; Jeude, 1890, Notes Leyden Mus. 12: 189, pi. ; Roon & Pelkwijk, 1939, Zool. Meded. Leiden 22: 65, figs, i, 2. Orthragoriscus fasciatus Bloch & Schneider, 1801, Syst. Ichth.: 511. ZOOL. I. 6 O2 H4 THE OCEAN SUNFISHES (FAMILY MOLIDAE) Orthragoriscus hispidus Bloch & Schneider, 1801, ibid.: 511. Cephalus brevis Shaw, 1804, Gen. Zool. 5: 437, pi. 175; Neill, 1811, Mem. Werner. Soc. 1: 546; Mitchill, 1815, Trans, lit. phil. Soc. N.Y. 1: 471 ; Swainson, 1839, Nat. Hist. Fish. 1: 199. Cephalus pallasianus Shaw, 1804, Gen. Zool. 5: 440. Orthragus luna Rafinesque, 1810, Caratt. Sicilia: 17-18; Indice Siciliana: 40. Diplanchias mola Rafinesque, 1810, ibid. Cephalus mola Risso, 1810, Ichth. Nice: 60; Poey, 1868, Repert. Cuba 2: 433. Orthagoriscus mola Cuvier, i8ij,Regne Anim., ed. i, 2: 149; Fleming, 1828, Hist. Brit. Anim.: 175; Nilsson, 1832, Prodr. Ichth. Scandinav.: 111; Jenyns, 1835, Man. Brit. Vertebr. Anim.: 490; Storer, 1839, Fish. Massachusetts: 170, pi. 3, fig. i ; Swainson, 1839, Nat. Hist. Fish. 2: 329, fig. 107; Bellingham, 1840, Mag. Nat. Hist. (N.S.), 4: 235; Bennett, 1840, Narr. Whaling Voy. 2l 262 ; Wellenbergh, 1840, Dissert. Inaug., Lugd. Batav., pi. ; Goodsir, 1841, New Philos. J. 30: 188, pi. 4; De Kay, 1842, Nat. Hist. N.Y. (Zool.), 3: 331, pi. 59, fig. 193; Storer, 1846, Mem. Amer. Acad. Arts Sci. N.S. 2: 495; Dilwyn, 1848, Mater. Fauna Swansea: 15; Parlby, 1848, Proc. Zool. Soc. Land. 17: 6; 1850, Ann. Mag. Nat. Hist. (2) 5: 53 ; Schlegel, 1850, Fauna Japonica (Poiss.): 288, pi. 127; Costa, 1850, Fauna Regn. Napoli (Pesci, Plettognathi) : 28, pis. 63-64; Smith, 1851, Ann. Mag. Nat. Hist. (2) 8: 347; Kroyer, 1852, Danmarks Fisk. 3: 732 ; Embleton, 1854, Trans. Tyneside Nat. 2: no, pi. 3 ; Nilsson, 1855, Skandinav. Fauna: 697 ; Thompson, 1856, Nat. Hist. Ireland 4: 243 ; Kolliker, 1860, Verh. phys-med. Ges. Wurzburg 10: xxxviii; Cleland, 1862, Nat. Hist. Rev.: 170, pi. 5-6; Storer, 1863, Mem. Amer. Acad. Arts Sci. N.S. 8 (2) : 420, pi. 34, fig. 2 ; Beltremeux, 1864, Ann. Acad. la Rochelle (Faune) : 53 ; Couch, 1865, Hist. Fish. Brit. Is. 4: 377, pi. 245; Blanchere, 1868, Nouv. Diet, peches: 505, fig. 673; Schlegel, 1869, Nat. Hist. Ned. Vischen: 182, pi. 17, fig. 4; Gunther, 1870, Cat. Fish. Brit. Mus. 8: 317; Capello, 1870, /. Sci. Math. Phys. Nat. Lisboa2: 136; 1881, Mem. R. Acad. Lisboa: 41; Andrews, 1871, Proc. Nat. Hist. Soc. Dublin (1865-1869), 5 (i): 123; Putnam, 1871, Proc. Amer. Ass. Adv. Sci. 19: 255; Amer. Nat. 4: 629, figs. 134, 137; Jourdain, 1871, C. R. Acad. Sci. Paris 63: 1225; Canestrini, 1872, Fauna d' Italia (Pesci}: 148; Barker, 1876, Zoologist: 5087; Malm, 1877, Goteborgs Fauna: 599, 654; Winther, 1879, Nat. Tidsskr. (3) 12: 54; Stossich, 1879, Boll. Soc. Adriat. Sci. Nat. 5: 36; Moreau, 1881, Poiss. France 2: 74; Vignal, 1 88 1, Arch. Zool. exp. gen. 9: 369, pi. 21 ; Campbell, 1883, Proc. Nat. Hist. Soc. Glasgow (1882) 5: 176; Day, 1884, Fish. Gt. Brit. 2: 272, pi. 148; Thompson, 1888, Anat. Anz. 3: 93, figs.; 1889, Stud. Mus. Zool. Univ. Coll. Dundee 1, No. 4; Vinciguerra, 1890, Boll. Mus. Zool. Rome 1: 33; Haller, 1891, Morph. Jb. 17: 198, figs., pis. 13-15; Steindachner, 1891, Ann. Naturh. Hofmus. Wien 6: 90; Almeida & Roquette, 1892, Inquir. Industr., Lisboa 2: 377', Girard, 1894, Ann. Sci. Nat., Porto 1:31; Tagliani, 1894, Monit. Zool. ital. 5: 248 ; Grieg, 1895, Bergens Mus. Aarb. 6: n; Smitt, 1895, Skandinav. Fisk. 2: 622, figs. 153, 154 a, 156, 157, pi. 27, fig. 4 ; Osorio, 1896, /. Sci. Math. Phys. Nat. Lisboa 4: 157 ; Vieira, 1898, Ann. Sci. Nat., Porto: 24; Clarke, 1898, Zoologist 16: 439; Andersson, 1900, Ofvers. Vetensk Akad. Forh., Stockh.: 603; Parker, 1900, Anat. Anz. 17: 313, fig.; Herdman & Dawson, 1902, Mem. Lanes. Sea Fish. Comm. 2: 57 ; Grifnni, 1903, Ittiol. Ital. : 155, figs. 81, 82 ; Michailovskij, 1903, Annu. Mus. Zool., Acad. St. Petersb. 8: xlvi; Meek, 1904, Anat. Anz. 25: 217, fig.; Dall, 1908, Bull. Mus. Comp. Zool. Harv. 43 (6) : 232 ; Novikov, 1909, Dnevn. russkh. Estestroisp. 1909- 1910: 286; 1910, Anat. Anz. 37: 97; Sauvage, 1910, Mem. Soc. Hist. Nat. Autun. 23: i; Gunther, 1910, /. Mus. Godeffroy 9(17) : 477; Seabra, 1911, Bull. Soc. Portug. Sci. Nat.: 193; Le Danois, 1913, Poiss. Manche occ. : 106, fig. 182 ; Kaschkarov, 1916, Rev. Zool. Russe 1: no, figs. 1-12 ; Thompson, 1918, Scot. Nat. : 41, 59 ; Kincaid, 1919, Annot. List. Puget Sound Fish. : 23, fig. 43 ; Toni, 1921, Atti 1st. Veneto 80: 125 ; Grenholm, 1923, Stud. Floss. Teleost. Upsala: 240 ; Patroni, 1923, Ann. Mus. zool. Napoli, N.S. 5 (4), pi. i ; Jenkins, 1925, Fish. Brit. Is. 1212, pi. 85 ; Duncker & Mohr, 1926, in Grimpe & Wagler, Tierwelt Nord u. Ostsee 4 (12) : Xllg 29, figs. 4, 5; Gudger, 1928, Sci. Mon. N.Y.: 257; Burr, 1928, /. Comp. Neurol. 45: 33, figs.; Caraffa, 1929, Poiss. Corse: 50, fig.; Marine Biol. Ass. 1931, Plymouth Mar. Fauna: 318; Saemundsson, 1931, Nat. Reykjavik 1: 164; 1939, Vidensk. Medd. naturh. Foren. Kbk. 102: 207; Noronha & Sarmento, 1934, P^ixes Madeira: 121; Nobre, 1935, Fauna Mar. Portugal, Vertebr. : 240, fig. 109; Toschi, 1936, Boll. Pesca Piscicolt. Idrobiol. 12: 325 ; Sanzo, 1939, Arch. THE OCEAN SUNFISHES (FAMILY MOLIDAE) 115 zool. Torino 26: 121, pi. 7, figs. 16, 17; Andersson, 1942, Fisk. Nord. 1: 62, pi.; Roon, 1942, Zool. Meded. 23: 313, fig. Orthagoriscus spinosus Cuvier, 1817, Regne Anim. ed. 2, 2: 370; Richardson, 1844, Voy. Sulphur, Fish.: 125, pi. 62, figs. 10-12. Cephalus ortagoriscus Risso, 1826, Hist. eur. Merid. 3: 173. Diodon carinatus Mitchill, 1828, Ann. Lyceum New York 2: 264, pi. 5, fig. i. ? Mola aspera Nardo, 1828, Bull. Sci. Nat. (Ferussac) 8: 437; Bonaparte, 1846, Cat. met. pesci eur. : 87. Mola hispida Nardo, 1828, ibid. : 438. Pedalion gigas (Guilding) Swainson, 1839, Nat. Hist. Class. Fish. 1: 199, fig. 33. Molacanthus pallasi Swainson, 1839, ibid. 2: 329. Ozodura orsini Ranzani, 1839, Novi Comment. Acad. Sci. Inst. Bonon 3: 80, pi. 6. Tympanomium planci Ranzani, 1839, ibid., table. Diplanchias nasus Ranzani, 1839, ibid. Trematopsis willughbii Ranzani, 1839, ibid. Orthragoriscus retzii Ranzani, 1839, ibid. ; Bonaparte, 1846, Cat. met. pesci eur.: 87. Orthragoriscus ghini Ranzani, 1839, ibid. Orthragoriscus rondeletii Ranzani, 1839, ibid. Orthragoriscus blochii Ranzani, 1839, ibid. Orthragoriscus alexandrini Ranzani, 1839, ibid., pi. 6; Alessandrini, 1839, ibid.: 359, pis. 31-34. Orthragoriscus redi Ranzani, 1839, ibid., table. Orthragoriscus aculeatus Ranzani, 1839, ibid. Pallasina pallasi Nardo, 1840, Ann. Sci. Regno Lombardo-Veneto 10: 112. Acanthosoma carinatum De Kay, 1842, Nat. Hist. New York, Zool. 3: 330, pi. 15, fig. 179; Storer, 1846, Mem. Amer. Acad. Arts Sci. 2: 494. Molacanthus hispidus Bonaparte, 1846, Cat. met. pesci eur.: 87. Mola luna Sassi, 1846, Saggio sopr. Pesci, &c. : 35; Aradas, 1871, Ann. Min. Agric. Ind. Comm. 1, pt. i : 587- Orthagoriscus analis Ayres, 1859, Proc. Calif. Acad. Sci. 2: 31, fig. 14; 1860, ibid.: 54, fig. 5; Stearns 1867, ibid. 3: 341. Molacanthus carinatus Gill, 1861, Proc. Acad. Nat. Sci. Philad. (1860): 21. Aledon storeri Castelnau, 1861, Mem. poiss. Afr. australe: 75. Aledon capensis Castelnau, 1861, ibid.: 76. Mola nasus Steenstrup & Liitken, 1863, Overs, danske Vidensk. Selsk. Fork.: 36; Wahlgren, 1868, Acta Univ. Lund. 4: i, pi. Mola retzii Steenstrup & Liitken, 1863, ibid. ; Wahlgren, 1868, ibid. Orthagoriscus sp. Swinhoe, 1863, Ann. Mag. Nat. Hist. (3) 12: 225. Orthagoriscus ozodura Harting, 1868, Verh. Akad. Wet. Amst. 11: i, pis. 1-8. Orthagoriscus planci Stossich, 1879, Boll. Soc. Adriat. Sci. Nat. 5: 36. Orthagoriscus nasus Jeude, 1892, Notes Leyden Mus. 14: 127, pi. 5; Tijdschr. Ned. Dierk. Ver. 18: 185, pi. ii. Orthagoriscus sp. Reuvens, 1894, Notes Leyden Mus. 16: 128, pi. 5. Mola mola Jordan, 1885, Proc. U.S. Nat. Mus. 8: 393; Eigenmann, 1893, ibid. 15 (1892): 131, 175 ; Jordan, 1895, Proc. Calif. Acad. Sci. (2) 5: 491 ; Collett, 1896, Result. Camp. Sci. Monaco, 10: 163 (part.); Jordan & Evermann, 1898, Bull. U.S. Nat. Mus., No. 47, 2: 1753; H. M. Smith, 1898, Bull. U.S. Fish. Comm. 17: 85; Linton, 1898, Proc. U.S. Nat. Mus. 20: 507 et seq. ; Evermann & Kendall, 1899, Rep. U.S. Fish. Comm.: 88; Jordan & Snyder, 1901, Proc. U.S. Nat. Mus. 24: 260; Green, 1901, Bull. U.S. Fish. Comm. 19: 321 ; Jordan & Evermann, 1902, Amer. Food and Game Fish.: 492, fig. ; Gilbert & Starks, 1904, Mem. Calif. Acad. Sci. 4: 206; Hargitt, 1905, Bull. U.S. Bur. Fish. 24 (1904): 25; Stead, 1906, Fish. Austral.: 227, fig. 82; Starks & Morris, 1907, Univ. Calif. Publ. Zool. 3 (n): 205; Murray & Hjort, 1912, Depths of the Ocean: 644; Halkett, 1913, Checklist Fish. Canada: 116; Dean, 1913, Amer. Mus. J. 13 (8) : 370, fig. ; Hilton, 1914, J.Ent.Zool. 6(4): 233; Evermann, 1915, Copeia, 20: 17; Buen, 1919, Bol. Pesc. Madr. 4: 295 ; 1935, Notas. Inst. esp. Oceanogr. 2 (89) : 146; Dons, 1920, Ii6 THE OCEAN SUNFISHES (FAMILY MOLIDAE) Troms. Mus. Adrsh. 43 (6): 38, pi. 2; Jordan, 1921, Copeia, 93: 28; McCulloch, 1922, Aust. Zool. 2: 130, pi. 43, fig. 374 a; Fowler, 1923, Proc. Acad. Nat. Sci. Philad. 75: 294; Wolleboek, 1924, Norges Fiske: 224, fig. 254; Damant, 1925, Nature, Lond. 116: 543, fig.; Bigelow & Welsh, 1925, Bull. U.S. Bur. Fish. 40 (i) : 301 ; Buen, 1926, Result. Camp. int. Inst. esp. Oceanogr. 2: 56; Barnard, 1927, Ann. S. Afr. Mus. 21: 986; Fowler, 1928, Mem. Bishop Mus. 10: 473; Ulrey & Greeley, 1928, Bull. Calif. Acad. Sci. 27 (i): 24; Breder, 1929, Field Book Mar. Fish. Atlant. Coast: 236, fig.; Hubbs & Schultz, 1929, Calif. Fish Game, 15 (3): Ulrey, 1929, /. Pan-Pacif. Res. Inst. 4 (4) : n, 235 ; McCulloch, 1930, Mem. Aust. Mus. 5: 436; Myers & Wales, 1930, Copeia 1934: n ; Ancona, 1931, Faune Flore Mediter., figs, i, 2 ; Breder, 1932, Copeia (4) : 180 ; Gregory, 1933, Trans. Amer. Phil. Soc. 23 (2) : 294 ; Gregory & Raven, 1934, Copeia 4: 145 ; Barnard, 1935, Ann. S. Afr. Mus. 30: 645 ; Barnhart, 1936, Mar. Fish. South. Calif. : 95, fig. 288 ; Tibby, 1936, Calif. Fish Game 22 (i) : 49, fig. 22 ; Fowler, 1936, Bull. Amer. Mus. Nat. Hist. 170 (2) : 1123, fig. 469 ; Schultz & De Lacy, 1936, Mid-Pac. Mag. 49 (3) : 211 ; Scofield, 1937, Calif. Fish Game 23 (4): 336; Schultz, 1938, Nat. Geogr. Mag. 74 (4): 497; Brimley, 1939, /. Elisha Mitchell Sci. Soc. 15 (2): 301, pi. 30; Deranyigala, 1944, /. Bombay Nat. Hist. Soc. 44 (3) : 429 ; Mendes, 1944, Bol. Fac. Filos. Cien. Let. Univ. S. Paula, Zool. No. 8: 173, pi. ; Engel, 1945, Zool. Meded. Leiden 25: n, pi. i ; Clemens & Wilby, 1946, Bull. Fish. Res. B. Canada 68: 330, fig.. 247; Medcof & Schiffman, 1947, Acadian Nat. New Brunswick 2: 8, 63, fig.; Poll, 1947, Poiss. Mar.: 405, figs. 260, 261; Barnard, 1948, Ann. S. Afr. Mus. 36 (5): 401, pis. 12, 13; Maul, 1949, Vertebr. Madeira, ed. 2, 2 (Peixes) : 158; Clark, 1949, Amer. Mus. Novit. 1397: 7, fig. 9; J. L. B. Smith, 1949, Sea Fish. S. Africa: 422, pi. 95, fig. 1213; Tortonese, 1950, Att. Ace. Ligure Sci. 6 (i) : 112. Orthragoriscus nasus Reuvens, 1897, Notes Ley den Mus. 18: 209, pi. 3. Mola ramsayi Whitley, 1931, Rec. Aust. Mus. 18 (3) : 126 (part.}, fig. 2, pi. 16, fig. i ; 1933, Viet. Nat. 49: 210, figs, i, 2 (not of Giglioli). Mola alexandrini Barnard, 1948, Ann. S. Afr. Mus. 36 (5): 402. The above extensive synonymy illustrates the considerable literature which has accumulated concerning this species. From a perusal of this data it is possible to give a rather more complete account than for other members of the family, but there is still much of its biology that remains conjectural. The anatomy has been studied broadly and in detail by a number of workers, and from this, together with descrip- tions or figures giving reliable information about the clavus, it seems quite clear that not more than one species is involved. Published records, considered statistically, would give the impression that the species is mainly a North Atlantic one, becoming rarer southwards, in the Indian Ocean and in the Western Pacific, but this is possibly an illusion due to the much higher rate of publication in the Atlantic and Mediter- ranean countries. Certainly the Japanese form is not separable from the Atlantic form, since we have specimens from Japan in our collection for comparison; according to Jordan and Fowler it occurs at Hawaii, and it seems to be common at California, so that it is replaced by M. ramsayi only in the South Pacific. I am much indebted to Mr. W. I. Follett, of the California Academy of Sciences, for information and radiographs which enable me to identify the Calif orni an specimens. A bad practice among some authors is the borrowing of an illustration from some earlier work, especially when the specimen depicted was obtained in a locality remote from that being discussed. Mola mola has suffered much from this treatment, and in consequence it is not possible to be definite as to the identity of specimens in regions where M. ramsayi might occur also, because the distinguishing characters THE OCEAN SUNFISHES (FAMILY MOLIDAE) 117 of the clavus have been hitherto unknown and are, therefore, not described; a re- liable picture might have given the answer. Comparison of adequate descriptions and figures shows that some order underlies the variability which has been remarked upon by so many authors. After meta- morphosis the young fishes are short and deep, the snout not protuberant, the fins rather narrow, and the margin of the clavus is not conspicuously lobed. The length of the clavus from the posterior edge of the 'carapace' i.e. the anterior edge of the smooth band between dorsal and anal fins is much less than that of the head. When the fish exceeds a length of about 2 ft., however, sexual differences become apparent. The bony tubercle on the snout is either pushed forward (in the male), or upward (in the female) ; in consequence the male develops a pronounced snout, projecting forward (the 'nasus' form), while the female appears more deep-headed, with the front of the snout nearly vertical (the ' alexandrini' form). As growth proceeds the clavus develops backwards between the ossicles, forming a series of lobes which at first number between 9 and 12 in both sexes ; females do not seem to pass beyond this stage, but in large males the five median lobes become very large and the others reduced. After the formation of the lobes the clavus is probably always longer in a male than in a female of the same size, and in the biggest males it may be as long as the head. In large specimens of both sexes two prominent, swollen ridges are formed on each side of the head ; these are discernible in small examples, and are evidently analogous if not homologous with the late- ral ridges of Ostracionts, but with age they become very conspicuous. In the larger examples also the dorsal and anal fins are relatively much broader. All this is indicated by a study of the records. Comparatively few of the specimens described have been examined for sex, but in each case where the sex is stated the characters mentioned above are found to be associated with it; of particular interest is the paper by Roon & Pelkwijk (1939), who had both sexes and figured them together. Harting's (1868) plate I gives a fair representation of a female, and Whitley (1931) has given a drawing of another, together with a photograph of it (pi. xvi, fig. i), which shows the lateral ridges excellently, and Murray & Hjort's (1912) photograph, copied by Schmidt, illustrates a fine male. The various phases of development out- lined do not always coincide with a particular size or age, but are evidently dependent to some extent on environmental circumstances. M ola mola grows to a great size, the largest record being apparently that by Dean (1913), measuring 10 ft. i in. in length and n ft. from tip of dorsal fin to tip of anal fin, a male. Mikailovskij (1903) described one measuring 8 ft. 6 in. in length and weighing 1,410 kg. Jeude (1890) described a specimen 2-23 m. (7 ft.) in length, FIG. 15. Post-larvae of Mola. A. ' Ostracion hoops' stage (5 mm.). (After Schmidt) ; B. ' Molacanthus' stage (16 mm.). (From specimen in the British Museum collection.) il8 THE OCEAN SUNFISHES (FAMILY MOLIDAE) apparently a female. The specimen recorded by Giinther as ' 7 feet long, Portsmouth ' was the fish taken by Parlby (1849), wno described its capture at Chesil Beach and NA FIG. 1 6. Mola mola, adult, 600 mm. long, Plymouth. (From stuffed specimen in the British Museum collection.) NA, area of reduced denticles ; PO, paraxial ossicle. stated that it measured 6 ft. 3 in. long. It was probably a male. As a stuffed skin it remained in the British Museum collection until recently, when it was found to be in a bad state and destroyed ; my (calliper) measurement at this time reading 5 ft. 8 in., the loss being presumably due to shrinkage (unless Parlby made a contour measure- THE OCEAN SUNFISHES (FAMILY MOLIDAE) 119 ment). A number of smaller stuffed skins and several specimens in spirits remain in the collection. It is never common, the large literature being due to the great interest FIG. 17. Mola mola, young adult, 366mm. long, Chouse, Japan. (From specimen in spirits in the British Museum collection.) it arouses, almost every specimen being reported upon ; but it is more frequently met with than any of the previous species. Nevertheless its early developmental stages are less well known than those of Ranzania, and fertile eggs or early larvae have not been found ; it is not improbable that it spends a great part of its life in deep water. 120 THE OCEAN SUNFISHES (FAMILY MOLIDAE) The scarcity of young specimens is remarkable when we consider that a female 4 ft. 6 in. long contained 300 million eggs. The mode and place of breeding have yet to be found. Its migrations inshore are unpredictable, and are usually supposed to coincide with invasions of medusae, salps, and ctenophores, upon which it largely feeds. Specimens FIG. 1 8. Different tabulation of the clavus, with similar skeletal supports, in Mola mola. Drawn to the same size for comparison. That on the right is the characteristic form in large males. taken inshore, however, are usually found to be feeding on littoral forms, and the list of organisms taken from stomachs includes Crustacea, ophiuroids, molluscs, hydroids, ctenophores, corallines, and algae ; Schmidt has reported them as feeding heavily on kptocephali ; on one occasion a flounder (Platichthys flesus) was found in the throat (Reuvens, 1897), and in our collection there is a ling (Molva macrophthalma] two feet long which was taken from the stomach of Mola mola. The stomach is not infrequently found to be empty, and it is quite probable that the specimens so frequently taken without difficulty while ' basking ' at the surface are in fact sick or dying fish. Myers & Wales (1930) found young fish to be active and alert, but later found two larger fish ' disabled' at the surface. It would be interesting to know the cause of such disable- ment. Possibly the great variety of parasites with which they are often found to be infested may have some bearing on the matter. THE OCEAN SUNFISHES (FAMILY MOLIDAE) 121 REFERENCES BARNARD, K. H. 1927. A Monograph of the Marine Fishes of South Africa. Part II. Ann. S. Afr. Mus. 21 (2) : 419-1065. (Molidae, p. 984.) - I 935- Notes on South African Marine Fishes. Ann. S. Afr. Mus. 30 (5): 645-658^15. 23-25. (Molidae, p. 635, figs. 5-7.) CHABANAUD, P. 1935. Quelques Monstruosites chez des Poissons Heterosomes. Arch. Mus. Hist. Nat. Lyon, 15: 1-23, 4 pis. FRASER-BRUNNER, A. 1943. Notes on the Plectognath Fishes. VIII. The Classification of the Suborder Tetraodontoidea, with a Synopsis of the Genera. Ann. Mag. Nat. Hist, (n) 10: 1-18. GREGORY, W. K., & RAVEN, H. C. 1934. Notes on the Anatomy and Relationships of the Ocean Sunfish (Mola mola). Copeia, No. 4: 145-151, pi. GREEN, E. H. 1901. The Chemical Composition of the Sub-dermal Tissue of the Ocean Sunfish. Bull. U.S. Fish Comm. 19: 321. GUDGER, E. W. 1937 a - The Structure and Development of the Pointed Tail of the Ocean Sunfish Masturus lanceolatus. Ann. Mag. Nat. Hist. (10) 19: 1-46, pis. 1-2. !937 b. The Natural History and Geographical Distribution of the Pointed-tailed Sunfish (Masturus lanceolatus) with Notes on the Shape of the Tail. Proc. Zool. Soc. Lond. (A) pt. 3 : 353-396, pis. 1-4. McCuLLOCH, A. R. 1912. A Description and Figures of Three Specimens of Molacanthus. Proc. Linn. Soc. N.S.W. 37: 553-555, 2 pis. PELLEGRIN, J. 1912. Sur la presence d'un bane de Ranzania truncata Retzius a la Martinique Bull. Soc. Zool. France, 37: 228-230, fig. i. RAVEN, H. C. 1939 a. Notes on the Anatomy of Ranzania truncata. Amer. Mus. Novit. 1038: !-7- 1939 b. On the Anatomy and Evolution of the Locomotor Apparatus of the Nipple-tailed Sunfish (Masturus lanceolatus}. Bull. Amer. Mus. Nat. Hist. 76 (4): 143-150, pi. 2. RYDER, J. A. 1886. On the Origin of Heterocercy and the Evolution of the Fins and Fin-rays of Fishes. Rep. U.S. Fish Comm. 1884: 981-1085, n pis. SANZO, L. 1939. Rarissimi stadi larvali di Teleostei. V. Orthagoriscus mola Linn. Arch. zool. Torino 26: 143-146, pi. 7, figs. 16-17. SCHMIDT, J. 1932. Dana's Togt omkring Jorden. 1928-30. 368 pp. (Molidae, p. 249 et seq.) K0benhavn. TREWAVAS, E. 1933. On the Structure of Two Oceanic Fishes, Cyema atrum Giinther and Opisthoproctus soleatus Vaillant. Proc. Zool. Soc. Lond. (3) : 601-614, 2 pis. WHITLEY, G. P. 1931. Studies in Ichthyology No. 4. Rec. Aust. Mus. 18 (3) : 96-133, pis. 11-16. (Molidae, p. 126, fig. 2, pi. 16.) PRESENTED 3 1 OCT 1951 PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATEY PRINTER TO THE UNIVERSITY 2 JUL 1952 CESTODES OF SEALS FROM THE ANTARCTIC STANISL-AW MARKOWSKI BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 7 LONDON : 1952 THE CESTODES OF SEALS FROM THE ANTARCTIC BY STANISLAW MARKOWSKI V 6 Associate, Department of Zoology British Museum (Natural History) Pp. 123-ljO; Pis. 10-21 BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 7 LONDON : 1952 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is issued in five series, corresponding to the Departments of the Museum. Parts appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. I, No. 7, of the Zoological series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued January 1952 Price Twelve shillings and sixpence By STANISLAW MARKOWSKI ASSOCIATE, DEPARTMENT OF ZOOLOGY, BRITISH MUSEUM (NATURAL HISTORY) SYNOPSIS The present paper reviews and re-describes all known Pseudophyllidean Cestodes occurring in the five species of Antarctic Seals, namely, Weddell seal, Leopard seal, Crabeater seal, Elephant seal, and Ross seal. According to the literature, twelve species belonging to two genera have been previously recorded from these hosts. The investigation of material collected by the British Graham Land Expedition, together with a comparative study of type-specimens collected by other Antarctic expeditions, leads to the conclusion that there are nine species of Pseudophyllidean Cestodes occurring in these hosts. These belong to four genera, two of which are new and of these, one represents a new species. From the twelve species quoted in the literature, three are here placed in synonymy and one transferred to another genus. MATERIAL AND METHODS THE bulk of the material for this study was collected by the British Graham Land Expedition during 1934-1937. In addition, some samples gathered by the Discovery in 1925, 1928, and 1931, and by the ' Falkland Islands Dependencies Survey ' in 1945, have been examined for comparative purposes. The material obtained by the British Graham Land Expedition comprises 33 lots from Weddell seals (Leptonychotes weddelli), 12 from Leopard seals (Hydrurga leptonyx), 6 from Crabeater seals (Lobodon carcinophagus) , and 2 from Elephant seals (Macrorhinus leoninus). The Discovery material consists of 5 lots from Leopard seals and i from a Crabeater seal, and that of the ' Falkland Islands Dependencies Survey ' of 2 lots only from Weddell seals. Thus the 6 1 batches of specimens investigated can be summarized as follows: 35 from Weddell seals, 17 from Leopard seals, 7 from Crabeater seals, and 2 from Elephant seals. The British Graham Land Expedition's material was obtained from Graham Land (Debenham Is., Horseshoe I., Argentine Is., and Beascochea Bay), South Shetland (Deception I.), Palmer Archipelago (Melchoir I.), and South Georgia (Cooper Bay and Bay of Isles). The Discovery material was collected from South Orkneys (Coronation I.), South Georgia (Maivicken), and South Sandwich Is. The 'Falkland Islands Dependencies Survey' from Graham Land (Hut Cove, Hope Bay) and Palmer Archipelago (Port Lockroy, Wiencke I.). Preservation was in 4 per cent, formalin or, occasionally, in Bouin's Solution. The material comprises mainly portions of the gut with Cestodes attached to the walls, and individual specimens are, with a few exceptions, perfectly extended. In addition to the new material, numerous collections from the same host-species were examined. These included the type-specimens described by Baird (1853), Shipley (1907), Rennie & Reid (1912), and Leiper & Atkinson (1914). The collection 126 THE CESTODES OF SEALS FROM THE ANTARCTIC of Fuhrmann consisted of microscopical preparations of specimens, probably types, of Linstow (1892), of Railliet & Henry (1912), and material described by Fuhrmann (1920) himself. Some of the material in these collections is not in very good condi- tion, the specimens mounted whole or as serial sections having partially lost their stains. Over 1,200 slides of the present material have been made. Whole preparations have been stained with Mayer's paracarmine or alum carmine, and serial sections, 10 ju to i8/Lt in thickness, have been double-stained with Ehrlich's haematoxylin and erythrosin. I have great pleasure in expressing my heartiest thanks to Dr. H. A. Baylis, who kindly suggested this investigation and provided the necessary materials ; to Dr. H. W. Parker, Keeper of the Department of Zoology, British Museum (Natural History) ; to Dr. M. Burton ; and to Mr. S. Prudhoe for his valuable assistance. In addition, I wish to express my thanks to Professor J. G. Baer, Rector of the University of Neuchatel, for his kindness in lending me the collection of slides left by the late Professor O. Fuhrmann, and to Dr. G. C. L. Bertram, of St. John's College, Cam- bridge, for information on the Graham Land Seals. HISTORICAL Our knowledge of Cestodes occurring in the Antarctic seals has been obtained almost exclusively from various Antarctic expeditions, including five British, one Australian, one French, and two German expeditions. The first specimens were collected in 1839-1843 by Ross's Antarctic Expedition from a Phoca sp. probably the Ross seal and described by Baird (1853) as Bothrio- cephalus antarcticus. Further samples were collected from the Ross seal by the National Antarctic Expedition (Discovery) in 1901-1904 and described by Shipley (1907) under the name Dibothriocephalus antarcticus. Railliet & Henry (1912) studied this species from material collected from the Ross seal by the Second French Antarctic Expedition of Dr. J. Charcot (Pourquoi Pas ?) 1908-1910 and described it as Diphyllo- bothrium antarcticum. Finally, Fuhrmann (1920) J re-examined the type-specimens and made the species the type of a new genus, Glandicephalus , characterized by the peculiar development of the musculature of the body, by the arrangement of the testes, and by the presence of gland-cells in the scolex. Johnston (1937) mentions the species as having been found in the Ross seal by the Australasian Expedition, 1911-1914. Linstow (1892) described as Bothriocephalus tectus headless specimens from Elephant seals, collected at South Georgia by the German Expedition in 1882-1883. This was also recorded by Linstow in Shipley (1902) from the Ross seal, collected by the Southern Cross Expedition, 1898-1900, and by Johnston (1937) from the Elephant seal, in the collection of the Australasian Antarctic Expedition. Diphyllobothrium quadratum (Linstow 1892) was collected for the first time from the Leopard seal by the German Expedition, 1882-1883, in South Georgia and in 1902-1904 by the Scottish National Antarctic Expedition (Scotia). It was re-described 1 Fuhrmann's paper was published in 1920 (December) and not in 1921, as is often quoted. THE CESTODES OF SEALS FROM THE ANTARCTIC 127 by Rennie & Reid (1912) as Bothriocephalus coatsi, from the same host. Railliet & Henry (1912) recorded this species as Diphyllobothrium resimum, collected by the Second French Antarctic Expedition. Later, Fuhrmann (1920) gave a description of D. quadratum, collected by the German South Pole Expedition, 1901-1903, and regarded Bothriocephalus coatsi (Rennie & Reid, 1912) and Diphyllobothrium resimum (Railliet & Henry, 1912) as synonyms of that name. Finally, Johnston (1937) gave a few further details of the species, based on material from the Leopard seal, collected by the Australasian Antarctic Expedition. Shipley (1907) described Diphyllobothrium wilsoni and D. scotti, the first being obtained from the Weddell seal and the Ross seal, and the second from the Ross seal, both having been collected by the National Antarctic Expedition, 1901-1904. They were re-described by Fuhrmann (1920) from the material obtained by the German South Pole Expedition (1901-1905) from the Ross seal and Leopard seal. They are also reported by Johnston (1937) from the Ross seal and the Weddell seal collected by the Australasian Antarctic Expedition. Diphyllobothrium mobile (Rennie & Reid, 1912) from the Weddell seal and D. scoticum (Rennie & Reid, 1912) from the Leopard seal were collected for the first time by the Scottish National Antarctic Expedition in 1902-1904 and described by Rennie & Reid (1912) under the generic name Dibothriocephalus. Fuhrmann (1920) gives a full re-description of both these species from the Ross seal and the Weddell seal, and of D. scoticum from the Leopard seal, collected by the German South Pole Expedition in 1901-1903. Both species were studied by Johnston (1937) from material from the Weddell seal and the Leopard seal, collected by the Australasian Antarctic Expedition. Diphyllobothrium perfoliatum Railliet & Henry, 1912, described also as D. clavatum in the same work, was collected for the first time by the Second French Antarctic Expedition from the Weddell seal. It was re-described by Leiper & Atkinson (1915) from material collected by the British Antarctic (Terra Nova) Expedition 1910 and by Fuhrmann (1920) from material collected by the German South Pole Expedition under the name Dibothriocephalus perfoliatus. In both cases the material was taken from the same host. Fuhrmann (1920) recognized D. clavatum as a synonym of D. perfoliatum. This Cestode is also mentioned by Johnston (1937) in a collection made by the Australasian Antarctic Expedition. Diphyllobothrium rufum Leiper & Atkinson, 1914, collected by the Terra Nova Expedition in 1910, is considered by Johnston (1937) as 'a short-necked, precocious form of D. perfoliatum', found in the Weddel seal by the Australasian Antarctic Expedition. The last two species, Diphyllobothrium lashleyi and D. archeri, were both described by Leiper & Atkinson (1914) from the Weddell seal gathered by the Terra Nova Expedition. The first of these species was re-described by Johnston (1937) from material from the same host-species collected by the Australasian Antarctic Expedition 1911-1914. To sum up, 9 species have been described from the material collected by the British Antarctic Expeditions, 4 by the French, and 2 by the two German Expeditions. The details of the result of each expedition are shown in Table No. i. 128 THE CESTODES OF SEALS FROM THE ANTARCTIC TABLE No. i A list of the Cestodes from Seals, collected by Antarctic Expeditions during the period 1839-1914 Expeditions Parasite Host Ross's Antarctic Exp. 1839-1843 Southern Cross Antarctic Exp. 1898-1900 National Antarctic Exp. (Discovery) 1901-1909 Scottish National Antarctic Exp. (Scotia) 1902-1904 British Antarctic Exp. (Terra Nova) 1910-1913 Australasian Antarctic Exp. 1911-1914 Glandicephalus antarcticus (Baird, 1853) Diphyllobothrium tectum (Linstow, 1892) ,, scotti (Shipley, 1907) ,, wilsoni (Shipley, 1907) Glandicephalus antarcticus (Baird, 1853) ,, antarcticus (Baird, 1853) Diphyllobothrium coatsi (Rennie & Reid, 1912) ., mobile (Rennie & Reid, 1912) ,, scoticum (Rennie & Reid, 1912) Phyllobothrium sp. (larva) Rennie & Reid, 1912 Diphyllobothrium mobile (Rennie & Reid, 1912) ,, coatsi (Rennie & Reid, 1912) perfoliatum (Railliet & Henry, 1912) ,, archer i (Leiper & Atkinson, 1914) ,, lashleyi (Leiper & Atkinson, 1914) ,, rufum (Leiper & Atkinson, 1914) ,, perfoliatum Railliet & Henry, 1912 ,, lashleyi (Leiper & Atkinson, 1914) ,, wilsoni (Shipley, 1907) ,, rufum Leiper & Atkinson, 1914 ,, mobile (Rennie & Reid, 1912) ,, quadratum (Linstow, 1892) ,, scoticum (Rennie & Reid, 1912) scotti (Shipley, 1907) ,, wilsoni (Shipley, 1907) mobile (Rennie & Reid, 1912) Glandicephalus antarcticus (Baird, 1853) Diphyllobothrium tectum (Linstow, 1892) Phyllobothrium sp. (larva) Phoca sp., probably Ross seal Ommatophoca rossi Hydrurga leptonyx Leptonychotes weddelli Hydrurga leptonyx Leptonychotes weddelli Hydrurga leptonyx Ommatophoca rossi Mirounga leonina German Antarctic Exp., South Georgia, 1882-1883 German Antarctic Exp. (Gauss) 1901-1903 Diphyllobothrium tectum (Linstow, 1892) ,, quadratum (Linstow, 1892) ,, perfoliatum Railliet & Henry, 1912 ,, quadratum (Linstow, 1892) wilsoni (Shipley, 1907) mobile (Rennie & Reid, 1912) Mirounga leonina Hydrurga leptonyx Leptonychotes weddelli Hydrurga leptonyx Ommatophoca rossi, Hydrurga leptonyx Ommatophoca rossi, Leptonychotes weddelli THE CESTODES OF SEALS FROM THE ANTARCTIC TABLE No. i (continued) 129 Expeditions Parasite Host 2nd French Antarctic Exp. (Pourquoi Pas ?) 1908- 1910, Dr. J. Charcot Diphyllobothrium resimum Railliet & Henry, 1912 ,, wilsoni (Shipley, 1907) ,, perfoliatum Railliet & Henry, 1912 ,, clavatum Railliet & Henry, 1912 ,, sp. ? Railliet & Henry, 1912 Glandicephalus antarcticus (Baird, 1853) Cestoda (unidentified) Hydrurga leptonyx Leptonychotes weddelli Ommatophoca rossi Lobodon carcinophagus The species enumerated above have been listed by Meggitt (1924) and Stunkard & Schoenborn (1936), though all these authors appear to have overlooked the work of Fuhrmann (1920), who reduced the number of species to twelve, apportioned between two genera, Diphyllobothrium and Glandicephalus. 1 Wardle, McLeod, & Stewart (1947) have proposed a new classification of the genus Diphyllobothrium, but this appears to have been based mainly upon information obtained from the literature. Stunkard's (1948) criticism of the classification is fully subscribed to by the present writer. DISCUSSION After investigation of the new material it seems that the Pseudophyllidean Cestodes occurring in the Antarctic seals represent no more than 4 genera and 9 species namely : 1. Diphyllobothrium lashleyi (Leiper & Atkinson, 1914). 2. D. mobile (Rennie & Reid, 1912). 3. D. quadratum (Linstow, 1892). 4. D. scoticum (Rennie & Reid, 1912). 5. D. wilsoni (Shipley, 1907). 6. Glandicephalus antarcticus (Baird, 1853). 7. G. [Diphyllobothrium] perfoliatus (Railliet & Henry, 1912). 8. Baylisia baylisi gen. nov., spec. nov. 9. Baylisiella tecta (Linstow, 1892) gen. nov. The details of their anatomical differences are given in Table No. 2. Of the five species of Diphyllobothrium mentioned above, D. lashleyi alone possesses a well-developed distinct neck and the rudiments of genital organs at some distance behind it. In this it resembles both species of Glandicephalus. The rest of the species of Diphyllobothrium possess a very short indistinct neck, the presence of which has 1 In addition, unidentified Cestodes have been recorded from the Crabeater seal by Railliet & Henry (1912) ; and the larval stages of Phylloboihrium in the blubber of the Weddell seal by Rennie & Reid (1912) and Fuhrmann (1931). Phyllobothrium delphini (Bosc, 1802) Gervais, 1885, found by J. E. Hamilton in 1931 in the blubber of the Leopard seal at Falkland Islands is reported by Southwell & Walker (1936), and the larval stage of Phyllobothrium from Elephant seal by Johnston (1937). 130 THE CESTODES OF SEALS FROM THE ANTARCTIC Baylisiclla tecta gen. nov. Mi CO W 00 5 mm. ? absent 1 rtogi xn 61 rtoSi H o a VO 00 5 a a a VO vo * CM 33 XX a a M VO VO vo VO 555 / Elephant seal ?, j i! rtoo6 unn g uio gzi 1-3 mm. ? absent a o <^ a Os a <* O X oo a. M a p a. 'S a o * CO M ^_ vo a a VO O v v VO XX f^ a a 450 H beater seal e 3 " -s H VO <* V a X *^ PJ in >^ a a S o S _e c SEE sI5 i! a iy- *. a. vC l?a -I *O cC " s S 3 VO T Tx* I** CO 5? N CO VO u-1 j O. o U M So "' VO 0, 3 a 1 a a a | 1 III C m E E S E S i ^c a M K> * H J 8| 2 * li xx * a a 1^ VO O c vj * fl M vo tx i jj u E *,* a c re Q E o? x * 00 % ^J. ^ ~z ** U . a ^ a a a - ~ i a a S E m ,,,00 a a o >o in fx a a VC 99 M 9211x56 >. C tH a ^5 S.2 S t> fe f e .S o O 00 00 M I? <& X a o o xx 2" a a O vo to VO 17^ Leopard se Weddell se a a o fc S o CS 1 ill op * s s * a a." M a a|' |- X " a 5 || & M M N 10 e # *^ o O o CJ oo N ' s " wia ^v8 5 a b a a a _ 1 c E E SEE E E E E a a c a2 o x >^ C t- a ^ -S VO VO ^ O M w *^ 5 j i x CO XX m *? 1 1000 " V M J" * a ""age c u ra M "a vo a a vo o & g U IH in a a _ "rt 1 Q Baa E 2 S a a 2 1 a a vO y " a oo M _aj 3 ^ .E -> o\ 5 g 5. -0 tx X s a g, s * a a m 3 4V Weddell se S a a a a 1 is| ! i a o *x Sc?| O VO t-i ^ 1- XX "g a S 3 Q O^> fO M M n- 01 a o-, vo fl! K a a VO CO 1 Wedde o 5 ."2 ' C ' S 2 is 15 "O If) & . 8 . 1 Parasite &* c 2 S 'f i >. 2 1 |S i s 3 .Cirrus-sac: length widtn Vesicula seminalis: Vesicula seminalis: thickness . Distribution of test K ^ 3 So ' ' S H g "> a o o :.; co'o ^ D U /, ? , .' f: .('.. o o_ 21 20 24 FIGS. 19-24 23 1.7. THE CESTODES OF SEALS FROM THE ANTARCTIC 135 segment. In the squarish segment it is more compactly built and oval in shape ; in the elongate segments its structure is reticular and more diffuse. The eggs are 53-56 ^ x 40-44 V~ The vitelline glands, composed of large cells, are irregular with rounded lobes. They are sometimes slightly confluent in the anterior parts of the segment and much more so in the posterior region. There is an area free from vitelline glands around the genital complex. The products of the glands are collected in special ducts running through the cortical parenchyma. It seems that there occur additional collectors, transferring the yolk to the main yolk reservoir which is situated in the central part of the segment below the uterus. The reservoir is filled with large yolk cells, olive- greenish in colour. The size of the vitelline glands is 33 p to 50 /JL in diameter. The longitudinal muscles form a coat, 20 p, thick in transverse section, composed of irregularly distributed bundles. The dorso-ventral musculature is feebly developed. The excretory system consists of two main stems in the medullary parenchyma and some 28 vessels occurring in the cortical part of the segment, as seen in transverse section. Leiper & Atkinson (1914) described D. archeri from the Weddell seal as a separate species. Comparison of the type specimens of D. archeri and D. lashleyi with the new material leads to the belief that D. archeri is a synonym of D. lashleyi. Both authors dealt with specimens not fully developed and having squarish segments, but the anatomical features of the two species, as seen in serial sections of the type specimens, are closely similar. The structure of the longitudinal muscles, cirrus-sac, and the number and arrangements of testes are identical. Diphyllobothrium mobile (Rennie & Reid, 1912) [PL. 10, FIG. 2 ; PL. u, FIG. n ; PL. 13, FIGS. 25-31] Dibothriocephalus mobilis Rennie & Reid, 1912. Diphyllobothrium mobile Meggitt, 1924. Diphyllobothrium wilsoni Railliet & Henry, 1912. Dibothriocephalus coatsi Leiper & Atkinson, 1914 (nee Rennie & Reid, 1912). Host : Weddell seal (Leptonychotes weddelli) . Locality: Debenham Islands. This species is recorded from seven Weddell seals, twice in company with Glandi- cephalus perfoliatus. The infection was in most cases a mass infection. The length of the body of the specimens examined varied from 2-3 mm. to 14 mm., and the width from 345 ju, to 510 /n. The strobila is composed of about 14 segments, which in the anterior part of the strobila are wider than they are long. The terminal segment is usually oval. 1 The scolex is 675 ^-825 /u, in length and 345 ^-510 /A in width. Longitudinal serial sections show that it contains glandular tissue. The neck is little more than an unsegmented part of the scolex, 300 p, in length and 495 //, in width. 1 Beside the normally-developed segments, one abnormal strobila has been found in the writer's material. In it, the segments are split into two parts. The left-hand portion forms a kind of cul-de-sac and possesses testes and vitelline glands. The right part, which is a continuation of the strobila, contains eggs. (PI. 13, fig. 25.) 136 THE CESTODES OF SEALS FROM THE ANTARCTIC The genital rudiments occur in the immediate vicinity of the scolex. The specimens of 2-3 mm. in length have distinctly visible genital anlagen and well-separated testes. The length of the cirrus-sac, measured in sagittal section, is about ii2/u, and its height 66 p. Some very thin muscular fibres attach the proximal part of the cirrus-sac to the dorsal wall of the segment. The vesicula seminalis measures, in sagittal section, about 83 /it by 15 p, ; its walls are about 10 p thick and it is in a straight line with the cirrus-sac. The testes seem to occur only in the anterior part of the segment and are from 22 to 44 in number. There are 5 testes on each side in the transverse, and 6 to 7 in the sagittal, sections of the segment. They are 66 /n to 99 /z in diameter. The uterine openings alternate irregularly. The uterus forms a compact mass of coils. In the anterior part of the body it is situated below the cirrus-sac, and in segments in the hinder part of the strobila its coils surround the male copulatory organ. The vagina opens into the common genital opening in the vicinity of the male opening. The ovary forms two more or less elongate-oval wings. The eggs measure 56-60 //, by 40-43 ju,. The vitelline glands, about 33 p in diameter, are arranged in two separate lateral fields in each segment. A narrow transverse space free of vitelline glands is distinctly visible in the anterior part of the segment in whole preparations as well as in serial sections. The longitudinal musculature is very feebly developed. It forms a coat about 4 p, thick, composed of single, barely visible, fibres. The excretory system has not been detected, probably owing to contraction due to the fixative. Diphyllobothrium quadratum (Linstow, 1892) [PL. 10, FIG. 3; PL. n, FIG. 12; PL. 14, FIGS. 32-36] Bothriocephalus quadratus Linstow, 1892. Dibothriocephalus quadratus Zschokke, 1903. Diphyllobothrium quadratum Railliet & Henry, 1912. Cordicephalus quadratus Ward, McLeod & Stewart, 1947. Dibothriocephalus coatsi Rennie & Reid, 1912. Bothriocephalus coatsi Fuhrmann, 1920. Dibothriocephalus resimum Railliet & Henry, 1912. Host : Leopard seal (Hydrurga leptonyx) . Locality: Galindez Island, Argentine Islands; Debenham Islands; Horseshoe Island and Sandefjord Harbour, Coronation Island. This species is recorded from five Leopard seals, some of which were very heavily infested. The length of the body is from 4 to 12 cm. and the width about 4-5 mm. The specimens obtained from the mass infested hosts averaged about 4 cm. in length. In horizontal serial sections the lateral margins of the body seem to have a villous character. The segments are square, about 1-5 mm. long and i to 4-5 mm. in width. Their lateral edges in the hinder part of the strobila are slightly convex. The terminal segment is oval. PLATE 13, FIGS. 25-31 (For list of abbreviations see Plate 10) Diphyllobothriitm mobile from Weddell seal FIG. 25. Malformation of strobila. FIG. 26. Young specimen with genital anlagen. FIG. 27 (a). Sagittal section of scolex, showing glandular tissue, (b) Glandular cells enlarged. FIG. 28. Terminal gravid segment. FIG. 29. Transverse section of segment. FIG. 30. Sagittal section of segment, showing terminal part of uterus. FIG. 31. Sagittal section of male and female openings. PLATE 13 FIGS. 25-31 PLATE 14, FIGS. 32-36 (For list of abbreviations see Plate 10) Diphyllobothrium quadratum from Leopard seal FIG. 32. Scolex. FIG. 33. Gravid segment. FIG. 34. Horizontal section of segment, showing structure of ovary FIG. 35. Transverse section of segment. FIG. 36. Sagittal section of segment. PLATE 14 35 33 THE CESTODES OF SEALS FROM THE ANTARCTIC 137 The scolex of the mounted specimens is i to 1-8 mm. in length and 960 p to 1-4 mm. in width, more or less ovoid in shape and possesses internal glandular tissue. The neck although short is recognizable and measures from about 450 /n, to 1-5 mm. in length and from 587 to 870 [M in width. The genital rudiments occur in the segment next behind the neck. In whole preparations the cirrus-sac appears to be spherical; in sagittal section its length is about 300 n and height 60 /u. The vesicula seminalis is connected with the cirrus-sac almost along the same axis and is 120 /A in length and 105 /u, in width. The walls are about 17 p, thick. The almost spherical shape of the organ is sometimes rendered more or less irregular, probably by the fixative used. There are about 340 testes in each segment, 170 on each side and confluent in its anterior part. They are arranged in a single layer, though this is not very regular in some cases, probably owing to contraction. This was observed in the small 4 cm. long specimens, where some of the testes were arranged in two planes. There are 12 to 16 testes on each side in the transverse, and 12 to i6inthe sagittal, section. They are not very regular in shape and measure about 65 /u. by 78 /z. The vas deferens is situated dorsally in the segment and forms numerous coils. The uterine openings alternate irregularly. In the hinder region of the strobila the uterus is converted into an irregular sac filled with eggs. The anterior part of the uterus surrounds the cirrus-sac which, in the gravid posterior segments, is hardly visible . The ovary forms two wings, which surround the lower coils of the uterus. The thin-shelled and operculate eggs are 56 /u, by 43 /u,. The vitelline glands are very numerous, thickly arranged, irregularly lobed, and measure 50 /u, by 33 p. They obscure all other organs of the genital complex except the uterus and the cirrus-sac. The longitudinal muscles, not very well developed, form a coat about 33 p thick. They are composed of single bundles of fibres. The transverse and dorso-ventral musculature is composed of single fibres running in these two planes. The excretory system comprises two main vessels in the medullary parenchyma and about 34 trunks in the cortical part of the segment in transverse section. Diphyllobothrium scoticum (Rennie & Reid, 1912) [PL. 10, FIG. 4; PL. n, FIG. 13; PL. 15, FIGS. 37-43] Dibothriocephalus scoticus Rennie & Reid, 1912. Diphyllobothrium scoticum Meggitt, 1924. Dibothriocephalus pygoscelis Rennie & Reid, 1912. Host: Leopard seal (Hydrurga leptonyx). 1 Locality: Debenham Islands. This species has been found in four Leopard seals. The number of worms per host varied from 2 to 14. 1 Baylis (in Hamilton, 1934) assigned to Diphyllobothrium scoticum some Cestodes from the intestine of Otaria byronia, but the identification was only provisional and has not yet been confirmed, zoo. i. 7. s 138 THE CESTODES OF SEALS FROM THE ANTARCTIC The body is much longer than recorded by Fuhrmann (1920) and Johnston (1937) and ranges in the specimens examined from 52 cm. to 130 cm., with a corresponding width of 0-5 cm. to 1-8 cm. The elliptical scolex is about 3-5 mm. in length and 2 mm. in width. No glandular tissue has been found in this organ. The neck is about 495 p in length and 825 n in width. The figures for scolex and neck have been taken from a mounted specimen 52 cm. long. The segments are shorter than wide and have convex lateral edges. They are tapering in the posterior part of the body, and the terminal segment in small speci- mens is ovoid. The posterior lateral edge of the segment seems to have a semicircular thickening (PI. 15, fig. 43). The gravid segments are 5 to 8 mm. in length and 1-5 cm. to 1-8 cm. in width. The genital rudiments occur in the first segment behind the neck. The male genital openings are surrounded by numerous papillae, radially arranged, and are sometimes bordered with a semi-lunar furrow. In sagittal section the cirrus-sac measures about 231 /u, in length and 142 p in height. In the vicinity of the cirrus-sac, plainly visible in the sagittal section, occur spherical cells, probably of a glandular character. The vesicula seminalis is situated in the same main axis, as a continuation of the cirrus-sac. It is about 285 ^ in length and 180 p in width, with walls about 17 /u, thick. The vas deferens runs dorsally in numerous coils. There are about 600 testes, disposed in two separate fields about 300 on each side, arranged in a single layer and measure about 150-210 p, by 150 /x ; they are not con- fluent in the anterior part of the segment. The number of testes in sagittal section amounts to 25. It seems that there is no difference in the number of testes in sagittal section as between the large and the small gravid specimens. The number of testes counted in transverse section amounts to 14-15 on each side. The uterine openings alternate irregularly, and are situated in a transverse groove. The terminal part of the uterus is modified to form a thick- walled pocket, lined with a villous tissue. This modification of the terminal uterine duct is typical for the species. The uterus is not of a 'rosette' type but forms spiral coils, more or less distinct, 5-12 in number on each side. The ovary is reticular and irregularly palm-shaped. The eggs, some provided with a boss, are 76-79 /n by 56 p. The vitelline glands, composed of small cells, are fairly large, about 60-105/11 in diameter. They are confluent in the anterior part of the segment, leaving a free area around the genital opening. The longitudinal, transverse, and dorso-ventral musculature is very well developed. The longitudinal muscles form a coat 150 /A thick measured in transverse section, composed of numerous fibres, collected in not very distinct bundles. They are thickly arranged in the upper parts of the muscular coat, gradually becoming less dense towards the middle of the segment. The excretory system consists of 2, not always distinctly visible, trunks in the medullary parenchyma and about 20 in the cortical parenchyma. It is, however, not always possible to be sure of the number of cortical excretory vessels as the vessels may contract as a result of fixation and may not be distinct in serial sections. PLATE 15, FIGS. 37-43 (For list of abbreviations see Plate 10) Diphyllobothrium scoticum from Leopard seal FIG. 37. Gravid segment of small specimen. FIG. 38. Modification of terminal part of uterus, seen in transverse section. FIG. 39. Scolex. FIG. 40. Gravid segment of large specimen. Part of cortical parenchyma has been removed, showing distribution of testes and terminal modifica- tion of uterus. FIG. 41. Transverse section of segment. FIG. 42. Sagittal section of segment with male and female openings. FIG. 43. Outline of posterior edges of segment, in sagittal section. . - , s.q. u o 40 42 FIGS. 37-43 THE CESTODES OF SEALS FROM THE ANTARCTIC 139 Diphyllobothrium wilsoni (Shipley, 1907) [PL. 10, FIG. 5; PL. n, FIG. 14; PL. 16, FIGS. 44-50] Dibothriocephalus wilsoni Shipley, 1907. Dibothriocephalus scotti Shipley, 1907. Diphyllobothrium scotti Meggitt, 1924. Dibothriocephalus mobilis Leiper & Atkinson, 1915 (nee Rennie & Reid, 1912). Host : Weddell seal (Leptonychotes weddelli) and Leopard seal (Hydrurga leptonyx) . Locality : Deception Island ; Beascochea Bay ; Debenham Islands ; Melchior Archipelago ; Argentine Is. (Galindez I.) and South Sandwich. The species was found in eight Weddell seals and six Leopard seals. The length of the body varies from about i cm. to 5 cm. with a maximum width of 3 mm. The scolex is variable in size and shape, probably the result of contraction due to fixation ; its length is about 825/11 and the width from 450/01 to i mm. The presence of glandular tissue in the scolex has been confirmed, using Ehrlich's haematoxylin and erythrosin. The neck is very short and often impossible to distinguish, 375 /n in length and about 240 IJL in width. The first segments are shorter than wide, becoming gradually squarish, and then longer than wide. The terminal segment is usually oval or elongate-oval. The genital rudiments occur in the first segment behind the scolex. The cirrus-sac is 106-165 P l n g an d 83-99 P high m sagittal section. It is attached in its hinder part, with numerous well-developed muscular fibres, to the dorsal wall of the segment. These structures are plainly visible in a sagittal section 10 /u, thick. The contracted cirrus seems to be coiled spirally. When everted the cirrus is about 44 p, to 59 \n in length. The vesicula seminalis is about 92 /u, in length and 56 /u, in width, with the walls about 20 [L thick. The vas deferens, filled with sperm, runs dorsally and is irregularly coiled. There are about 150 testes, 75 on each side of the segment, arranged in a single layer and very closely distributed. They are confluent in the anterior part of the segment, irregular in shape, with slightly lobed outlines. There are 8 testes on each side in the transverse and 6 to 8 in the sagittal section ; their dimensions are about 132 n by 99 p. Histologically, as compared with other species, the testes seem to be more compact and stain more intensively with Ehrlich's haematoxylin. This may, however, be due to the number of spermatozoa in the testicular tissue absorbing more of the stain. The uterine openings are irregularly alternating. The uterus forms an irregular spherical sac filled with eggs and surrounds the cirrus-sac on both sides. The vagina opens close to the male genital opening in the common genital atrium. The eggs are 50 /x, by 40 /u,. The ovary is irregular in outline or kidney-shaped with more or less lobated edges. It is composed of large egg cells. The vitelline glands, measuring 66 /u, by 50 p, are irregularly spherical and somewhat 140 THE CESTODES OF SEALS FROM THE ANTARCTIC amoeboid, very numerous, and strongly developed. They form thick uninterrupted layers from segment to segment, covering all the internal organs, except the uterus. The longitudinal musculature is not very strongly developed. It forms a coat composed of minute bundles about 17 p thick in transverse section. The central, medullary excretory system consists of two main stems not very easily distinguishable. The cortical excretory system possesses approximately 14 stems, fairly large in diameter, running among the vitelline glands. It seems from a comparison of the small specimens of D. wilsoni and the type specimens of D. scotti, described by Shipley (1907), with the newly-collected material from the Leopard seal, that D. scotti is a synonym of D. wilsoni. The comparison in Table 3 of Fuhrmann's data (1920) with the figures recorded from the present TABLE No. 3 Comparison of D. wilsoni and D. scotti Fuhrmann (1920) Writer's material D. wilsoni D. scotti D. wilsoni Body: length ..... 10 mm. 9 cm. 5 cm. width ..... 1-7 mm. 2 mm. 1-5-3 mm. Scolex: length ..... 850 p. 500-900 /i 825 /i-i mm. width ..... 450 fj. 700 fJL 450 /i-i mm. Neck: length ..... ? short 375 M width ..... ? ? 240/1 Cirrus-sac: length . .... 140/4 150/1 106-165 /i width . . . . ? ? 83-99 /* Vesicula seminalis, diameter . 80/1 80/1 92x56/1 No. of testes : transverse section .... 6-9 6 8 sagittal section ..... 6 6-io 6-8 Eggs, diameter ..... 60 X 36 fj. 64 X 40 fj. 50 X 40 p No. of cortical excretory vessels 14 12 14 Thickness of longitudinal muscle-layer 12/1 14-18 /I 17/1 material shows that there is practically no difference between the two so-called species, except in the size of the body, and it is this which seems to have misled previous authors. The small, mature specimen of D. wilsoni might be considered a dwarf form, caused by mass infection and consequent unfavourable living conditions. In the Leopard seal, where the infection is not so heavy, D. wilsoni reaches a relatively large size. 2. GLANDICEPHALUS Fuhrmann 1920 Glandicephalus antarcticus (Baird, 1853) [PL. 10, FIG. 6; PL. u, FIG. 15; PL. 17, FIGS. 51-53] Bothriocephalus antarcticus Baird, 1853. Dibothrium antarcticum Diesing, 1863. Diplogonoporus antarcticus Liihe, 1899. Dibothriocephalus antarcticus Shipley, 1907. Diphyllobothrium antarcticum Railliet & Henry, 1912. Glandicephalus antarcticus Fuhrmann, 1920. Host: Ross seal (Ommatophoca rossi). PLATE 16, FIGS. 44-50 (For list of abbreviations see Plate 10) Diphyllobothrium wilsoni from Weddell seal FIGS. 44-46. Scolex in different stages of contraction. FIG. 47. Gravid segment. FIG. 48. Terminal segment with double set of gravid genital organs. FIG. 49. Transverse section of segment. FIG. 50. Sagittal section of segment. PLATE 16 O-5 mm. 44 O-5 mm. cs.- FIGS. 44-50 PLATE 17, FIGS. 51-53 (For list of abbreviations see Plate 10) Glandicephalus antarcticus from Ross seal FIG. 51. Scolex. FIG. 52. Sagittal section of several segments. FIG. 53. Transverse section ol segment. PLATE 17 I mm. d.v.m. 53 )o. i. 7. THE CESTODES OF SEALS FROM THE ANTARCTIC 141 This species, described by Baird (1853) and re-described by Fuhrmann (1920), was collected by the Ross's Antarctic Expedition. To give a complete picture of the Pseudophyllidean Cestodes occurring in Antarctic seals, however, the type-specimens have been re-described to show their generic relationship with Glandicephalus perfoliatus. The specimens re-examined were about 10 cm. in length and 7 mm. in width. The strobila is markedly imbricate. The scolex, 3 mm. in length and 2 mm. in width, is provided with protuberances as depicted by Baird, though these are not distinct in all specimens. No glandular structure was discovered in sagittal serial sections stained with Ehrlich's haema- toxylin. The neck is about 2-35 mm. long and 750 p in width. The cirrus-sac, in sagittal section, measured about 462 p in length and 231 /z in height. The vesicula seminalis is about 132 ju, by 155 /JL, with walls about 10 ^, thick. The testes are scattered in the medullary parenchyma and among the longitudinal musculature. They are arranged in irregular layers. The eggs are 43-50 /z by 33 p and the vitelline glands are about 40 /z by 50 /z. The muscular system, as described and depicted by Fuhrmann (1920), is almost identical with that of G. perfoliatus. The longitudinal muscular coat is about 450 ju, thick. There are about 30 excretory vessels, counted in the transverse section of the cortical parenchyma. Glandicephalus perfoliatus (Railliet & Henry, 1912) n. comb. [PL. 10, FIG. 7; PL. n, FIG. 16; PL. 18, FIGS. 54-59] Diphyllobothrium perfoliatum Railliet & Henry, 1912. Dibothriocephalus perfoliatus Fuhrmann, 1920. Diphyllobothrium clavatum Railliet & Henry, 1912. Diphyllobothrium rufum Leiper & Atkinson, 1914. Host: Weddell seal (Leptonychotes weddelli). Locality: Debenham Islands; Stella Creek, Deception Island; Argentine Is. (Galindez I. and Winter I.) ; Beascochea Bay, Graham Land (Hut Cove, Hope Bay) ; and Palmer Archipelago (Port Lockroy, Wiencke Island) . This species was collected from thirteen Weddell seals. It occurs usually as a mass infection, mainly in the bile-duct, overhanging into the gut. The specimens examined were at different stages of maturity from 4 mm. to 20 cm. in length. The average length of the body ranges from 12 to 14 cm., and the minimum width is 7 mm. The strobila is differentiated into two distinct parts: the anterior, ivory white, amounting to about one-third of the total length, and the posterior, yellowish, increasing in width and tapering slightly at a small distance from the posterior end of the body. This differentiation of the strobila has not been observed in specimens of 4 cm. or 5 cm. in length, in which the outline of the body is oval with a well-defined scolex. The surface of the strobila bears, a small distance behind the neck, charac- teristic strongly developed imbrications formed by the excessive development of the cortical part of the segment. i 4 2 THE CESTODES OF SEALS FROM THE ANTARCTIC The scolex is about 3-5 mm. in length, measured in formalin specimens, and 2 mm. in width. No glandular structure has been discovered in longitudinal sections of the organ. The neck is distinct and ranges from 2-5 mm. to 3 mm. in length and i mm. in width. The segments are very short. The length of a fully gravid proglottid, measured in a sagittal serial section, is about 400 /A. The terminal segment is small and bell- shaped, with a cone at the terminal part of its longitudinal axis. It contains normal genital organs and produces eggs. This peculiar shape of the segment may be the result of fixation. The genital openings are irregularly alternating and situated on the anterior surface of the imbrications. The length of the cirrus-sac, measured in a sagittal section, is about 264 /n, and its width i88/Li. The pear-shaped vesicula seminalis is 148^ long and 132 \i wide in sagittal section ; its walls are about 20 ^ thick. There are about 100 testes in the fully gravid proglottid. They are more or less distributed in a single layer, which is more readily distinguishable in the lateral part of the segment. Near the centre, close to the uterine coils, the testes are arranged in irregular clumps. In transverse section there are about 14 testes on each side, and in the sagittal plane from 3 to 6 ; they measure 116-172 /a by 73-93 p. The vagina opens immediately behind the male pore, slightly obliquely. The uterine openings are irregularly alternate on the left or right side of cirrus-sac, and the uterus forms an irregular sac with indistinct coils, filled with eggs. These, some of them provided with a boss, are about 60-66 p by 50 /A. The ovary comprises two small wings tapering towards the lateral edge of the segment. The vitelline glands, distributed mainly in the anterior region of the imbrications of the segment, are very irregular in shape and size and measure about 70 \i by 40 /*. The longitudinal musculature of this species is very unusual and almost identical in structure with that of G. antarcticus (Baird). Together with the transverse and dorso-ventral muscular system, it is distributed throughout almost the whole medul- lary part of the segment. The single bundles of the longitudinal muscular system are separated by the fibres of the dorso-ventral and transverse musculature. Examined in transverse section the two lateral systems form a kind of square which encloses, at its centre, the fibres of the longitudinal muscles. The boundary between the cortical and medullary parenchyma, so characteristic of, and fairly easily dis- tinguishable in, most other species of Pseudophyllidean Cestodes, is not very distinct in G. perfoliatus, because of the network formed by the transverse and dorso-ventral musculature, through the meshes of which run the fibres of the longitudinal muscles. In the near vicinity of the testes, or of the excretory vessels, the muscular system is more diffuse. The central excretory system in the medullary parenchyma is composed of 2 vessels, which run an undulating course through the length of the body and are about 17 /*, in diameter. The cortical excretory system seems to be composed of 16 vessels. PLATE 18, FIGS. 54-59 (For list of abbreviations see Plate 10) Glandicephalus perfoliatus from Weddell seal FIG. 54. Young immature specimen. FIG. 55. Young specimen showing differentiation of strobila. FIG. 56. Portion of strobila showing internal structure, after removal of part of cortical parenchyma. FIG. 57. Sagittal section of segments with male and female openings. FIG. 58. Sagittal section of two posterior segments. FIG. 59. Transverse section of segment, with vitellaria in portion of imbrication. PLATE 18 O-5 mm. 57 THE CESTODES OF SEALS FROM THE ANTARCTIC 143 The two main nerves, 66 p by 44 p in transverse section, are situated outside the medullary excretory system. Comparison of this material with the type specimens of Glandicephalus antarcticus (Baird) reveals that, although there is no doubt that there are two distinct species, these have features in common that appear to warrant their being grouped together in a single genus distinct from Diphyllobothrium ; the name Glandicephalus Fuhrmann (1920) is available. It is true that this name was coined by Fuhrmann in consequence of his discovery of glandular tissue in the scolex of antarcticus and that no such tissue has been found in the scolex of perfoliatus ; but this character is not at all diagnostic of Glandicephalus and occurs in several species of Diphyllobothrium, e.g. D. lashleyi, D. mobile, and D. quadratum. Features common to both antarcticus and perfoliatus, but not found in any true Diphyllobothrium, are as follows : imbrication of the segments (less well developed in the first species) ; presence of a well-separated neck, and the arrangement of the musculature. In both species the transverse and dorso-ventral muscles form a kind of network spread in the medullary parenchyma, enclosing in the 'meshes' the fibres of the longitudinal muscles. Moreover, the vitelline glands are situated in the anterior region of the imbrications of both species. Finally, they seem to show a host specificity: G. antarcticus for Ommatophoca and G. perfoliatus for Leptonychotes weddelli. Leiper & Atkinson (1914) describe another species from the Weddell seal, namely, Diphyllobothrium rufum. Johnston (1937), who examined immature specimens which he thought to be D. rufum, was apparently inclined to believe that the species was probably a 'precocious form' of G. perfoliatus. Comparison of the type-specimens of D. rufum with G. perfoliatus reveals that the only differences are in the shorter neck and the presence of a 'notch' in the posterior margin of some of the segments. The neck of D. rufum, measured in sagittal section, is over I mm. in length. To judge from the published descriptions there would seem to be a difference in egg size also, for Leiper & Atkinson report the eggs of D. rufum to be 25 p in diameter. This, however, appears to be a slip ; in the type specimens their dimensions are 59-66 /z, by 43-46^, which is identical with the size of the eggs in G. perfoliatus. Further, there are no differences in the structure of the genital apparatus or of the musculature. Diphyl- lobothrium rufum seems, therefore, to be a synonym of Glandicephalus perfoliatus. 3. BAY LI SI A gen. nov. Diagnosis : Large Cestodes in which the anterior part of the body is cylindrically modified. Scolex with cup-shaped bothria. Normal segmentation not distinct ; the body bears pseudosegmentation not corresponding to the individual sets of genital organs. Genital organs and their openings situated ventrally on both sides of the segment in double sets, regularly alternate in relation to the main axis of the body. Testes arranged in a single layer. Ovary ramified. Longitudinal muscles forming a thick coat. Excretory system situated in the cortical parenchyma. Type species : Baylisia baylisi sp. nov. Type host: Crabeater seal (Lobodon car cinophagus) . I 4 4 THE CESTODES OF SEALS FROM THE ANTARCTIC Baylisia baylisi sp. nov. [PL. 10, FIG. 8; PL. n, FIG. 17; PL. 19, FIGS. 60-68] Host : Crabeater seal (Lobodon carcinophagus] . Locality : Deception Island ; Debenham Islands. This parasite has been found in two Crabeater seals. There were two complete worms, one 35 cm. in length and i cm. in width and a second 126 cm. by about 8 mm., together with some fragments of strobila, varying from 15 cm. to 63 cm. in length and from n mm. to 15 mm. in width. The colour in formalin is ivory-white in the anterior part of the body, becoming brownish-grey in the posterior segments. The ivory-white part of the strobila, about 2 cm. in length, is more or less cylindrical. There are two double furrows running laterally along the body. The central part of the strobila is convex along the main axis, marking the position of the uterus. The scolex has two cup-shaped bothria, and measures, in the specimen mounted in Canada, balsam about 900 /A long and 1-3 mm. broad. A neck seems not to be developed since segmentation starts immediately behind the scolex. The segmentation is very distinct, but apparently does not divide the body into single genital complexes, as happens in other tapeworms. The ' segments ' are 2 cm. long and about 1-5 cm. in width, the terminal one being oval. The genital openings are situated ventrally on a segment in the longitudinal furrows, alternately left and right. There are about 30-40 double genital sets in one 'segment', which alternate in relation to the main axis of the body, being arranged in 'zigzag', in contrast to Diplogonoporus, where they occur as two genital sets in the same transverse plane. In Baylisia baylisi one set of the genital organs only may be seen in transverse section. The cirrus-sac measures 750 /A long and 180 [j. high in sagittal section. An irregularly coiled ductus ejaculatorius is situated inside the cirrus-sac. A cirrus has not been observed. The vesicula seminalis is situated slightly laterally but internally to the cirrus-sac, and is 198 //, long and 99 /j, wide in sagittal section ; its walls are about 33 p, to 50 /x thick. The internal surface of the walls seems to have a villous structure. Beside the normally developed cirrus-sac mentioned above, one abnormality has been noted. Two cirrus-sacs were joined together, with a common opening, but each with a separate vesicula seminalis (PI. 20, fig. 65). In each segment there are about 36 testes arranged in a single layer and flattened antero-posteriorly. The number of testes in transverse section amounts to about 18 on each side, and there is I testis in sagittal section. The testes measure 165 //, in diameter in transverse section and 40 /x in the sagittal plane. They are elongate- oval, 116 IJL by 40 /A in horizontal section, with the longer axis transverse to the main line of the segment. The vagina opens into the genital atrium with the male opening and runs ventrally, but is transverse to the terminal part of the uterus. PLATE 19, FIGS. 60-68 (For list of abbreviations see Plate 10) Baylisia baylisi from Crabeater seal FIG. 60. Scolex. FIG. 61. Sagittal section of fused cirrus-sac, with common genital duct and opening. FIG. 62. Transverse section of part of male and female genital apparatus. FIG. 63. Transverse section of segment, showing structure of ovary and genital ducts. FIG. 64. Sagittal section of part of segment, showing arrangement of ovaries and uterine coils. FIG. 65. Transverse section of male and female ducts (enlarged). FIG. 66. Transverse section of segment. FIG. 67. Sagittal section of segment, showing male copulatory organs. FIG. 68. Horizontal section of part of strobila, showing arrangement of genital organs. PLATE 19 4l'^Bp?3K O-5 r XI. 7 . 2 mm. FIGS. 60-68 u THE CESTODES OF SEALS FROM THE ANTARCTIC 145 The uterine openings on both sides are situated nearer the median line than the genital openings and lie posteriorly and obliquely to the cirrus-sac. The uterus itself comprises a few horizontal coils in the central part of the segment. The ovary is very characteristic and is composed of one solid compact central part, giving off a system of branches and ramifications among the uterine coils; in sagittal section it is V-shaped. The eggs are 66 ju, by 46 //,. The musculature is very well developed. The longitudinal muscles form, in trans- verse section, a continuous coat about 450 \i thick, composed of numerous bundles. The transverse muscles are also well developed but the dorso-ventral musculature is much weaker. The excretory system seems to occur only in the cortical parenchyma, and is composed of about 76 vessels, visible in transverse section. Their transverse diameters differ considerably. This cortical excretory system is situated near the surface of the segment and runs among the vitelline glands. The vitelline glands, 43 \L by 20 //,, are composed of tiny cells and form a continuous layer, interrupted by the excretory system. They are more or less irregularly elongate in sagittal section, and elongate-oval in the horizontal plane. 'Calcareous' bodies are very numerous, 17^ by 26^ in diameter, and occur in the cortical and medullary parenchyma. They have also been noticed among the longitudinal muscles. 4. BAYLISIELLA gen. nov. Diagnosis: Pseudophyllidean Cestodes with the scolex bearing two strongly de- veloped bothria, modified in the anterior part in a f oliaceous lamella. The thick strobila, composed of very short and broad segments, is tapering posteriorly. The testes, arranged in two or three layers in the medullary parenchyma, have a tendency to ascend dorsally in relation to the uterus. The longitudinal muscles form, in transverse section, club-shaped or elongately oval bundles, some of them collected in pyramids, separated by the excretory vessels. Excretory system occurs in the cortical parenchyma. Type species : Baylisiella tecta (Linstow) . Type host: Elephant seal (Macrorhinus leoninus). Baylisiella tecta (Linstow, 1892) [PL. 10, FIG. 9; PL. n, FIG. 18; PL. 20, FIGS. 69-72] Bothriocephalus tectus Linstow, 1892. Dibothriocephalus tectus Zschokke, 1903. Diphyllobothrium tectum Meggitt, 1924. Cordicephalus tectus Wardle, McLeod, & Stewart, 1947. Host: Elephant seal (Macrorhinus leoninus). Locality : Bay of Isles, South Georgia. This species was found in two Elephant seals ; there were three entire worms in one host and two in the other, one specimen being headless. ZOO. I. 7. U2 146 THE CESTODES OF SEALS FROM THE ANTARCTIC The thick belt-shaped strobila, composed of very short indistinct segments, is 32 cm. in length and 2 cm. in width. The previous descriptions of Linstow (1892) and Fuhrmann (1920) were based on headless specimens. The scolex, not previously described, is deeply embedded in the intestinal tissue and very characteristic in shape. It possesses two powerful bothria and a complicated lamellar structure of their upper part, which recalls the scolex of Pyramicocephalus Monticelli 1890. The 'cauliflower' lamellar structure seems to give rise to two lateral lamellar flaps and there is a similar differentiation on the top of the scolex. The length of the scolex is 8 mm. and the width 5 mm., and the transverse diameter across the ' flap ' is 12 mm. The neck seems to be very short or not-existent. The segments are very short, about 165 /a-2oo p in length. The genital pores are situated in a common recess, provided with numerous papillae. The cirrus-sac, measured in sagittal section, is about 450 p in length and 150 //, in height. The vesicula seminalis, situated in the same axis as the cirrus-sac, is 195 p in length and 180 /x in width ; its walls are about 30 /u thick. The vas deferens runs dorsally in numerous coils. The testes are distributed in 2 or 3 layers, some of them ascending almost dorsally to the uterus and close to it. There are about 45 testes on each side in transverse, and 2 or 3 in sagittal, sections. They measure about 136 //, by 86 p. The uterine openings are situated a little below the cirrus-sac on the right side. The uterus comprises a few irregular transverse coils. The vagina opens in the vicinity of the cirrus-sac on its right side. The ovary is bilobed and elongate. The eggs with 3 p, thick shells, measure 59-66 /z by 46 /* and are thickened opposite to the operculum. The vitelline glands, 66 ^ by 26 /x, and arranged in a very thick layer, are very numerous, spherical or oval in transverse section. They form a compact mass of glands in the cortical parenchyma. The excretory system seems to occur in the cortical part of the segment. It is very strongly developed and runs through the longitudinal muscular system with numerous transverse anastomoses. In transverse section about 108 excretory canals have been counted. The musculature is exceedingly well developed. The longitudinal muscles are collected in large irregular bundles, which, in transverse section, appear club-shaped or elongate-oval. The bundles are often separated into distinct groups between which run excretory canals and dorso- ventral muscle-fibres. The thickness of the longitudinal muscular coat is about 555 p. There is also a layer of longitudinal muscles situated in the subcuticular region of the segment, externally to the vitelline glands. This additional layer is composed of scattered, rather thick, individual fibres or very small bundles of fibres. The transverse muscles are also very strongly developed. This species represents a new genus, Baylisiella, distinguished from Bothriocephalus PLATE 20, FIGS. 69-72 (For list of abbreviations see Plate 1 0) Baylisiella tecta from Elephant seal FIG. 69. Transverse section of segment. FIG. 70. Transverse section of male apparatus. FIG. 71. Sagittal section of segment. FIG. 72. Scolex. PLATE 20 FIGS. 69-72 THE CESTODES OF SEALS FROM THE ANTARCTIC 147 and Diphyllobothrium by characteristic differences in the structure of the scolex, in the development of the muscles, and in the distribution of the testes. ? Diphyllobothrium larvae [PL. 21, FIGS. 74-75] Host : Weddell seal (Leptonychotes weddelli) : Leopard seal (Hydrurga leptonyx) . Locality : Melchior Archipelago ; Debenham Islands ; Cooper Bay, South Georgia. Beside the adult forms enumerated in this paper, juvenile stages have also been discovered in five Weddell seals and in one Leopard seal. The infection bears in some cases a mass character. The identification of these juvenile forms was not possible, because of a complete lack of morphological features. They belong probably to one of the species dealt with here, most likely to Diphyllobothrium wilsoni or D. mobile. Based on differences in the scolex, two types of larvae might be distinguished : one with a kind of papillary modification occurring at the edge of the bothria, and another with the typical scolex of Diphyllobothrium but with the bothria smooth. The maximum length of the body of the first type, from the Weddell seal, is about 3 mm. long and about I mm. in width, and the second, from the Leopard seal, is about 1-2 mm. in length and 195 \i in width. ? Diphyllobothrium sp. (larva) [PL. 21, FIG. 73] Host: Crabeater seal (Lobodon carcinophagus) . Locality: South Sandwich. This juvenile stage was collected by the Discovery on 21 March 1928 from the intestine of a Crabeater seal. The total length of the body is 7 mm. and the width 480 ju. The scolex is well separated from the rest of the body and is heart-shaped, tapering anteriorly. Its length is 675 /A and its width 480 /x. The bothria are well developed. The unsegmented body has neither genital organs nor their rudiments, making identification impossible. OCCURRENCE OF THE PARASITE AND ITS RELATIONSHIP TO THE HOST As already stated, there are nine species of Pseudophyllidean Cestodes known to occur in the Antarctic seals. An analysis of the parasites and their hosts is shown in Table No. 4. Our knowledge of these parasites is very limited and the present material is merely the tenth collection of this kind. Nevertheless, certain speculations may be justifiably made on the occurrence and the host specificity of these parasites, as well as on their 148 THE CESTODES OF SEALS FROM THE ANTARCTIC TABLE No. 4 Composition of Pseudophyllidean Cestodes in the present material Diphvllobothrium lashleyi (Leiper & Atkinson, 1914) Leptonychotes weddelli D. mobile (Rennie & Reid, 1912) . D. quadratum (Linstow, 1892) D. scoticum (Rennie & Reid, 1912) ? D. sp. (larva) .... D. wilsoni (Shipley, 1907) Glandicephalus antarcticus (Baird, 1853) G. perfoliatus (Railliet & Henry, 1912) Baylisia baylisi gen. nov., spec. nov. Baylisiella tecta (Linstow, 1892) gen. nov. Hydrurga leptonyx Lobodon carcinophagus Leptonychotes weddelli, Hydrurga leptonyx Ommatophoca rossi Leptonychotes weddelli Lobodon carcinophagus Macrorhinus leoninus host-relationship. It may be presumed that Glandicephalus perfoliatus and Diphyl- lobothrium lashleyi are specific to the Weddell seal, G. antarcticus to the Ross seal, and D. quadratum and D. scoticum 1 to the Leopard seal. The newly-described Baylisia baylisi is the first identified Cestode from the Crabeater seal, although unidentified tapeworms have been reported from this host by Railliet & Henry (1912). Baylisiella tecta seems to be closely associated with the Elephant seal. Linstow (in Shipley, 1902) identified some Cestodes collected from the Ross seal as belonging to this species, but this identification requires confirmation. The two next species, Diphyllobothrium mobile and D. wilsoni, are less selective in their hosts, the first occurring in the Weddell and Ross seals, the second in the Leopard and Weddell seals. According to information supplied by Dr. G. C. L. Bertram, who collected the Graham Land material, the most frequently and heavily infested species are the Weddell seal, the Leopard seal, and the Elephant seal. In the Crabeater seal infestation with tape- worms very seldom occurs and the present record is virtually the first. It is obvious that the nature of the food has an enormous influence on the kind and number of parasites. The Weddell seal eats fish and cephalopods. The food of the Leopard seal is composed of penguins and fish. The food of the Crabeater seal consists of Crustacea, mainly Euphausiids. These Cestodes occur in specific parts of the gut of the host. Glandicephalus perfoliatus infests mainly the bile-duct, overhanging into the lower part of the intestine. It occurs very often as a mass infection, choking the lumen of the bile-duct. Diphyllobothrium lashleyi, D. mobile, D. quadratum, D. scoticum, and D. wilsoni infest the duodenal part of the gut and Baylisia baylisi and Baylisiella tecta occur in the rectum. Except for these two and Diphyllobothrium scoticum, the rest of the species occur very often as a mass infestation occupying almost all the free surface of the gut. D. scoticum does not occur in such numbers as the others but makes up for it in the size of the individuals. It is the largest Cestode recorded from the Antarctic seals. Baylisiella tecta also is fairly large and is not numerous in the gut. Diphyllobothrium wilsoni and D. mobile have been considered by previous authors as dwarfs, and the smallest species of Diphyllobothrium, D. wilsoni, is very small, reaching in mass infestation no more than 10 mm. in length. On the other hand, specimens collected from the less heavily infested Leopard seals reach from 5 to 9 cm. 1 See footnote on p 137. PLATE 21, FIGS. 73-75 FIG. 73. ? Diphyllobothrium sp. from Crabeater seal. FIGS. 74-75. Two types of larvae from Weddell seal. PLATE 21 73 74 FIGS. 73-75 THE CESTODES OF SEALS FROM THE ANTARCTIC 149 in length. The larger specimens were considered by Shipley (1907) to be a separate species, D. scotti. It is quite probable that further investigation of the Cestodes of Antarctic seals will prove that the small size of D. mobile is also caused by the living conditions in mass-infected hosts. Where a mass infestation occurs one species of Cestode only is invariably present. Usually in the moderately infested hosts the Cestodes may represent more than one species. Glandicephalus perfoliatus was recorded in the same host together with Diphyllobothrium lashleyi, D. mobile, and D. wilsoni. The same has been shown for the occurrence of D. scoticum with D. quadratum and D. wilsoni in the same gut. No pathological changes of the gut have been observed, except in the rectum of the Elephant seal, where Baylisiella tecta provokes large nodules, about 3 cm. in diameter. This is caused by the scolex being very deeply embedded in the intestinal wall. Judging from the mass occurrence of parasites in the gut of seals, there must be a high degree of immunity on the part of the host against toxic factors provoked by the parasite. It seems from the literature that the Pseudophyllidean Cestodes found in the Antarctic seals do not occur in any other species of Pinnipeds. REFERENCES ARIOLA, V. 1900. Revisione della famiglia Bothriocephalidae s. str. Arch. Parasit. Paris, 3: 369-484. BAIRD, W. i853. Descriptions of some new species of Entozoa from the collection of the British Museum. Proc. zool. Soc. Lond. 21: 18-25. 1853. Catalogue of the species of Entozoa or intestinal worms contained in the collection of the British Museum. 132 pp. London (British Museum (Nat. Hist.)). 1855. Catalogue of the species of Entozoa or intestinal worms contained in the collection of the British Museum. Ann. Mag. nat. Hist. (2) 15: 69-76. BAYLIS, H. A., in HAMILTON, J. E. 1934. The southern sea lion Otaria byronia (De Blainville). 'Discovery' Rep. 8: 306. BERTRAM, G. C. L. 1940. The biology of the Weddell and Crabeater seals. British Graham Land Expedition, 7934-^937. Sci. Rep. 1. (i) : 1-139. BLANCHARD, R. 1894. Notices sur les parasites de 1'homme, 3 me s6rie; Sur Krabbea grandis et remarques sur la classification des Bothrioce'phalines. C.R. Soc. Biol. Paris, 46: 699-702. BRAUN, M. 1897. Vermes. Bronn's Klassen, 4 Abt. Ib: 1407-1454. 1898. Vermes. Bronn's Klassen, 4 Abt. Ib: 1535-1614. COOPER, A. R. 1921. Trematoda and Cestoda. Rep. Canad. Arct. Exped. 1913-1918. 9 (G-H) : 1-28. DIESING, K. M. 1863. Revision der Cephalocotyleen. Abtheilung: Paramacocotyleen. S.B. Akad. Wiss. Wien 48, Abt. i: 200-345. 1856. Zwanzig Arten von Cephalocotyleen. Denkschr. Akad. Wiss. Wien 12: 23-38. DRUMMOND, F. H. 1937. Reports of the expedition of the McCoy society for field investigations and research Lady Julia Percy Island. Proc. roy. Soc. Viet. 49: 401-406. FUHRMANN, O. 1920. Die Cestoden der Deutschen Siidpolar-Expedition 1901-1903. Deutsche Sudpol.-Exped. 1901-1903 (Drygalski), 16: 469-524. 1931. Cestoidea. Handb. Zool. Berl. 2: 416 pp. GERMANOS, N. K. 1895. Bothriocephalus schistochilos n. sp., ein neuer Cestode aus dem Darm von Phoca barbata. Jena. Z. Naturw. 30: 1-38. HAMILTON, J. E. 1934. The Southern sea lion Otaria byronia (De Blainville). 'Discovery' Rep. 8: 269-318. 150 THE CESTODES OF SEALS FROM THE ANTARCTIC JOHNSTON, T. H. 1937. Entozoa from the Australian hair seal. Proc. Linn. Soc. N.S.W. 62: 9-16. 1937. The Cestoda of the Australasian Antarctic Expedition. Sci. Rep. Aust. Antarctic Exped. 10 (4) : 1-74. JOYEUX, CH., & BAER, J. G. 1936. Cestodes. Faune Fr. 30: 613 pp. LEIPER, R. T., & ATKINSON, E. L., 1914. Helminthes of the British Antarctic Expedition 1910- 1913. Proc. zoo/. Soc. Land. 1914: 222-226. 1915. Parasitic worms with a note on free-living Nematode. Brit. Antarctic (' Terra Nova') Exped., 1910 Nat. Hist. Rep. Zool. 2 (3) : 19-60. LEON, N. 1910. Un nouveau cas de Diplogonoporus brauni. Zbl. Bakt. 55: 23-27. LINSTOW, O. 1878. Compendium der Helminthologie. xxii+382 pp. Hannover. 1892. Helminthen von Siid-Georgien. Nach der Ausbeute der deutschen Stationen von 1882-1883. Jb. hamburg. Wiss. Anst. 9: 59~77- in SHIPLEY, A. E. 1902. Nematoda, Cestoda. Rep. Coll. Nat. Hist. 'Southern Cross'. ix + 3 44 pp . London . LUEHE, M. 1899. Zur Anatomic und Systematik der Bothriocephaliden. Verh. dtsch. zool. Ges. 1899: 30-55- MARKOWSKI, S. 1949. On the species of Diphyllobothrium occurring in birds, and their relation to man and other hosts. /. Helminth. 23: 107-126. MATZ, F. 1892. Beitrage zur Kentniss der Bothriocephalen. Arch. Naturgesch. 1: 97-122. MEGGITT, F. J. 1924. The Cestodes of Mammals. 282 pp. London. 1924. On the life history of a reptilian tapeworm (Sparganum reptans). Ann. trop. Med. Parasit. 18: 195-204. MONIEZ, R. 1896. Traite de parasitologie animale et vegetale applique a la medecine. 680 pp. Paris. MUELLER, J. F. 1937. A repartition of the genus Diphyllobothrium. J. Parasit. 23: 308-310. NYBELIN, O., in SKOTTSBERG: 1931. Saugetier- und Vogelcestoden von Juan Fernandez. Nat. Hist. Juan Fernandez and Easter Island. 3: 493-524. RAILLIET, A., & HENRY, A. 1912. Helminthes recueillis par 1'Expedition Antarctique fran9aise du ' Pourquoi-Pas ? ' II. Cestodes des phoques. Bull. Mus. Hist. nat. Paris. 18: 153-159- RENNIE, J., & REID, A. 1912. The Cestodes of the Scottish Antarctic Expedition (Scotia). Trans, roy. Soc. Edinb. 48: 441-453, and in Rep. Scient. Res. Voyage 'Scotia' (1902-1904), 6: 243-256. SHIPLEY, A. E. 1905. Notes on collection of Parasites belonging to the Museum of University College, Dundee. Nematoda, Cestoda, with hosts. Proc .Camb. phil. Soc. (biol.) 13: 95-102. 1907. 'Cestoda.' Nat. A ntarct. Exped., 1901-1904. Nat. Hist. 3: 6 pp. (British Museum (Nat. Hist.)). SOUTHWELL, T., & WALKER, A. J. 1936. Notes on a larval Cestode from a Fur-seal. Ann. trop. Med. Parasit. 30: 91-100. STILES, C. W., & HASSALL, A. 1912. Index-catalogue of medical and veterinary Zoology: Cestoda and Cestodaria. Bull. U.S. hyg. Lab. 85: 1-467. STUNKARD, H. W. 1947. On certain Pseudophyllidean Cestodes from Alaskan Pinnipeds. /. Parasit. 33: (suppl.): 19. 1948. Pseudophyllidean Cestodes from Alaskan Pinnipeds. /. Parasit. 34: 211-228. 1949. Diphyllobothrium stemmacephalum Cobbold, 1858 and D. latum (Linn., 1758). /. Parasit. 35: 613-624. & SCHOENBORN, H. W. 1936. Notes on the structure, distribution and synonymy of Diphyllobothrium lanceolatum. Amer. Mus. Novit. 880: 1-9. WARDLE, E. R. A., McLEOD, J. A., & STEWART, I. E. 1947. Luhe's ' Diphyllobothrium ' (Cestoda) . /. Parasit. 33: 319-330. ZSCHOKKE, F. 1903. Die arktischen Cestoden. Fauna arct., Jena 3: 1-32. o ini IQI/> PRINTED IN GREAT BRITAIN AT THE UNIVERSITY PRESS OXFORD BY CHARLES BATEY PRINTER TO THE UNIVERSITY ' THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 1948-1949 BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) ZOOLOGY Vol. i No. 8 LONDON : 1952 THE 'MANIHINE' EXPEDITION " TO THE GULF OF AQABA 1948-1949 CONTENTS I. Foreword: Station List and Collectors' Notes 'fo II. Preliminary Hydrological Report: G. E. R. DEACON III. Sponges: M. BURTON IV. Turbellaria: Polycladida: s. PRUDHOE V. Gephyrea: A. c. STEPHEN VI. Mollusca: w. j. REES and A. STUCKEY VII. Echinodermata: A. M. CLARK VIII. Tunicata: w. G. VAN NAME IX. Fishes: N. B. MARSHALL Pp. 151-252, Pis. 22-32; 10 Text-figs BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) y ZOOLOGY VOL. I, No. 8 LONDON: 1952 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted in 1949, is issued in five series, corresponding to the Departments of the Museum. Parts will appear at irregular intervals as they become ready. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. This paper is Vol. i, No 8, of the Zoology series. PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM Issued August 1952 Price Twenty-five shillings THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA I. FOREWORD: STATION LIST AND COLLECTORS' NOTES DURING the winter of 1948-1949 the motor-yacht Manihine was engaged in biological investigations in the Gulf of Aqaba on behalf of the British Museum (Natural History), the work being under the supervision of Mr. N. B. Marshall. This gulf is of special interest because in it the peculiarities of the Red Sea appear at their most intense. The Red Sea is geologically young with a fauna derived from that of the Arabian Sea and, possibly, the Mediterranean. This immigrant fauna is now completely isolated from the last-mentioned and also partially isolated from the former by reason of the narrowness and shallowness of the connecting passage, the Strait of Bab-el Mandeb. It also finds itself in a region where some, at least, of the ecological conditions are very different. The most noticeable of these ecological differences is to be found in the isohaline and isothermal nature of the water below 200 metres and the complete absence of any cold, deep-water layer. The John Murray Expedition (Seymour Sewell, 1935, John Murray Exp., Reports, 1, i) recorded tem- peratures from 21-64 to 21-84 C. at depths of 1,000 to 1,900 metres in the Red Sea, but at similar depths in the Gulf of Aden the temperature was at least 10 C. lower (3'59- i:ic '53 C.)- The degree of isolation of the Aqaba fauna is greater than that of any other part of the Red Sea since the passage between the two, the Strait of Tiran, provides only a restricted channel for faunal interchange. The strait is both narrow and shallow, forming a distinct sill, with a greatest depth of less than about 300 metres ; on either side of the sill the water deepens rapidly to 1,000 metres and upwards. The hydrological conditions inside the gulf appear to be essentially similar to those in the Red Sea proper, though, as might be expected, salinities are some- what higher. In this Bulletin are reports on some of the collections that were brought back. Other reports, including a study of the interchange of heat and water vapour between the surface water and the air, will be prepared as opportunities offer, but in some instances the collections will be studied in conjunction with other material and will not form the subject of a special report. Acknowledgements and thanks are due to many individuals and institutions whose material aid or advice contributed greatly to the expedition. Foremost among them is Major H. W. Hall, O.B.E., M.C., who not only provided the ship and was respon- sible for most of the preliminary organization, but who, with Mrs. Hall, accompanied the expedition taking a large share of the actual collecting and doing most of the photography. A small selection of the photographs is published here to give a general impression of the gulf and its surroundings. Many localities could not have been visited but for the skilful pilotage of Captain Hargreaves through poorly charted 154 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA waters, and to him, and to his hard-working crew, all possible thanks are due. The Hydrographer of the Navy and the Director of 'Discovery Investigations' lent apparatus vital to the expedition and His Excellency the Egyptian Ambassador in London made arrangements that ensured pleasant and harmonious relations wherever the ship was in Egyptian waters. Lastly, thanks are due to the Government Chemist, whose department carried out the analyses of salinities. Except for the plankton and some of the fishes all material was obtained from littoral areas and coral reefs (or coral patches). Localities where collections were made are indicated on the chart. Within the Gulf of Aqaba (reading from north to south) these were : Aqaba (PI. 22, fig. i) Hobeik (PI. 23, fig. 4) Faraun Island (PI. 22, fig. 2) Dahab (PL 24, fig. 5) Graa Um Nageila (PI. 24, fig. 6) Mualla (PL 23, fig. 3) Abu Zabad Along the Sinai shores there are well-formed coral reefs at Dahab and from Um Nageila southwards. The bulk of the invertebrate material was obtained from these regions, particularly from Abu Zabad on the loth and nth February 1949 when there were low spring tides. North of Dahab there were coral patches at all localities visited, but these never become massed to form a definite reef. Outside the gulf collections were made at the following localities : Sanafir Island (PL 25, fig. 7) Sherm-el-Moiya (PL 26, fig. 9) Tiran Island (PL 25, fig. 8) Ras Muhammad Bay (PL 26, fig. 10) Sherm Sheikh Most time was spent on Sanafir Island, where there were well-formed coral reefc. Here, as elsewhere, much material was collected by diving for pieces of coral and extracting the small invertebrates and fishes. It will be observed that in the Station List no temperatures are given for depths below 40 metres. It was, however, established that at all stations where deep-water samples were taken (i.e. where salinity figures are given in the list) the temperature exceeded 18-5 C. The following are the meanings of the abbreviations used in the list. D.M. = Dredge, medium. D.S. = Dredge, small. K.T. = Kelvin tube. N. 70 V. = Vertical haul by silk plankton net with mouth 70 cm. diameter. N. 100 V. = Vertical haul with stramin plankton net with mouth 100 cm. diameter. O.T.L. = Otter trawl, large. The Gulf of Aqaba. The positions of the numbered stations are given in the station list. 156 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA No catch : net torn off. No catch. K.T. i *- H - MM 55 t O M m O CO M M MM M MM in CO in O m O -*- vo \ M M O m CO M M M m CO M M m M 8 M in M M in O M j s VMAtl in fx m m MM M MM O * co M O O * m * m M M in in m o M M M M M Tj- M m M m M O T(- ON O co M M m M O M j (syjjffiu) W(f0(j O O vo oo M CO Ol IN, IN. . * * M M u O M o OO M > > O O O O Q Q M 55 55 55 55 55 X K * 55 55 55 55 55 v| TfOO COOO CO Ov CO O M OO M OO M 00 M OO M t*s M OO M OO CO ^S 14 'Oo ' M OO "* Oi ON in ON vO IN IN vo M 00 oo IN vo vO JN. g 4 /oS O 000 66 666 6 I c O O O O O O O O O O 000 O O o * * 1 0.+-* M M M Ov P 1 1 M 1 1 M f 1 1 1 ' ! 1 1 \ ^ \ \ M 1 M 1 1 M y 1 I M 1 1 M oo ?l 1 M M m M M co m M 9 M $ (sy^ytu) wfog O O IN Tt- co IN M M O IN O IN ON O ^J" C CO CO M IN TJ M MM M if "\ O IN ^ O IN rt* " co IN co IN O ^ OO M m O * M IN co O o 'yqout) wjpuiotvg op ON M in ON M in in m in in ON ON ON MM M m in co co ON ON m CO ON m co M CO ON M co 0\ M CO ON M ON M * CO " MM M M M M M M M M M * UOtJOMld 55 * C/3 CO C/3 C/3 CO CO is CO CO CO i ( sftftfu) Sutpunos o Tj- vO in CO m ON co VO OO M * ON J CO oo M ON co ON M m O oo ON M O M'JITT *M>H O CO * M M M i? O * CO M M 1 O 1 m m in o m in mm co in M m M vo Tj- M in M VO M MM M i2 6^? ir; -^- MM MM O W m * CO M M 1 in ^~ in in M 1 m O * O OO M O 1 m m M M M M 1 m in TJ- O M M O co T(- CO M M M <_ =g ON M CO l'3 mm m MM M O O M M V* vo' M IN M M IN IN M 55 W 55W 55 W 55 W ' W 55 W 55 W 55W 55W 55 W 55W 55 W 55 W UOtJISOJ CO CO * o m co * o o vo * oo m co O co vo M VO M OO * CO M MM M M * * M M CO O VO CO O o o O CO M M M ^ VO M M in M CO O IN co m o\ * M CO O\ in IN co vo IN M in M in M in ON ^J" O\ ^ ON Tj* M CO M CO M CO MM o m Mm Mm M CO M CO in IN M ^ ON%- M CO ON in O TI- ONr|- M CO oo O O m ON Tj- M CO CO CO O * o\ * M CO M OO o * ON * M CO vO VO m <* OO Tj- M CO -OT M M M CO Tj- in vo 00 O\ M M M M CO M THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 157 Catch nil. m O 8 O CM CM m M * o CO m O O O O O CO O CO o\ O co CM in Tj- o o o M O CM m o CO o in M in -a- CO o 0* in CO O O O CM M CO CO M m m M m M o M O * N M a\ o O *! M CO O O\ o CM O Oi O * O M CO IH M oo 8 M o\ N M M m CM M O CO o CO O 00 O oo O CO O CO O co o co O OO oo O OO O CO O OO O O co m o oo oo oo u > O O o o o o O O O O > RN o O O 55 N * 55 * 55 fe * ^ * 5 ^ 55 fe * 55 O tx CM tx M CO * M tX CM tx M OO M OO m o\ 0\ tx tx U X CO > m m M VO o C3\VO VO m vo vo VO vo vo vO vo VO VO O vo oo m vo vo CO VO CO " "** S3 O O O O o O O O O O O O o o c 5 O O ( 5 O O o O O O O O O o O O O o o o a CO 1 I I CM in co M CM Jj N co * 1 1 M 1 1 CM tx y> M CM co SI 1 CM ;? 1 CM M i CM op M N O CM CM O CM M CM m CM rf- r i N o in M CM OO SI 1 N 1 1 CM 3 o tx * CO tX M CM O O tX Tt- CO tx M CM O O tx O CO CM M OO O tx C CO * M 1 > O tx < r> co i r> MI. 3 O n r> O O O tx co in M m Jx covo M ^* tx co m O tx O CO CO w tx C8 "a. CO 0\ CM m m CM in CM m in CM m in m VO ON CM in vp CM m vp M in VO Ov CM vp CM m vo 0\ CM m vp o\ CM vp Oi CM m N m ON CM vo 6-1 E o N M M - N M CM N CM CM N CO CO - CM CM M 9 Ul CO CO CO CO CO CO i i Be 55 55 55 W 55 g S5 !5 - .0 C8 CO CO O oo O\ in O o in A 1 CM * ON VO A 1 1 1 o m oo 1 1 1 1 1 1 1 in M L o 6 o CO O O\ M 1 in M O M m CM 1 m 6 M M ^- CO M 1 O co co in ^ M 1 o o N CM M M o o" O M i CO CO CM M i co o o o\ o L O CO O Tt- is O CTi O\ O 45 m o 1 m M CM CM O co 8S I o m O M 1 "> I? M M fl oo M oo' M co' co' M iH CO 0\ M ON M M M t d CM 6 CM d N N M CM CO 'jr PL] 55W W 55 W 55W 55W 55 W 55W 55 W W XU 55 W 55 W S5 W 55 w 55 W 55 W O CO in Tf oo * CM CO * CM m -t- oo -* CM CO Tj- VO in o <* Tj- CO Tj- CM CO vO vo CO CM co m * * co -* CM CO -* CO CM Tt- CO * CO OO Tt- CM CO o o CTiCO CO CO OO T)- CM CO O co CO M CO Tj- OO Tj- CM CO 5- N in CO CO OO Tj- CM CO VO -t co m CO -*j- CM co CM O M CM CM CO oo V CM CO CM CM CM tO OO Tj- CM CO in M CM CO OO Tj- CM CO 5, 8 m oo M CM co <* CM CO vo N CO M * m M CO oo * CM CO O s CO vO O co CO >* CM CO O N O Tj- Vioo O N oo % CM CO Tj-VO CM CO m e CM CO s in M vo M M 00 M ON M CM N CM CO N CM in CM vo CM CM CO N CM co M CO Legends to Plates 22-27. PLATE 22 FIG. i. Aqaba looking north-east. FIG. 2. Gezeret-el-Faraun from the south-east. PLATE 23 FIG. 3. Looking north from the anchorage at Mualla. FIG. 4. Hobeik. PLATE 24 FIG. 5. Typical gulf scenery. Coast 5 miles south of Dahab. FIG. 6. Mangrove swamps at Um Nageila. PLATE 25 FIG. 7. Sanafir Island ; Fish-eagle's nest. FIG. 8. Tiran Island, seen from Sanafir. PLATE 26 FIG. 9. Sherm-el-Moiya ; looking north-east from the entrance. FIG. 10. Manihine at anchor in Ghazulani Bay with Ras Muhammad in the distance. PLATE 27 FIG. ii. Abandoned police post at Naweibi-el-Terabin, about 45 miles south of Aqaba. FIG. 12. Arab fisherman using cast net. Bull. B.M. (N.H.) Zoology, I, 8 PLATE 22 FIG. i. AQABA FIG. 2. GEZERET-EL-FARAUN Bull. B.M. (N.H.} Zoology, I. 8 PLATE 23 FIG. 3. MUALLA FIG. 4. HOBEIK Bull. B.M. (N.H.) Zoology, I, 8 PLATE 24 FIG. 5. GULF SCENERY NEAR DAHAB FIG. 6. UM N AGE I LA Bull. B.M. (N.H.) Zoology, 1, 8 PLATE 25 FIG. 7. SANAFIR ISLAND FIG. 8. TIRAN ISLAND Bull. EM. (N.H.) Zoology, I, 8 PLATE 26 FIG. 9. SHERM-EL-MOIYA FIG. 10. GHAZULANI BAY Bull. B.M. (N.H.) Zoology, I, 8 PLATE 27 FIG. ii. NAWEIBI-EL-TERABIN FIG. 12. CAST NET II. PRELIMINARY HYDROLOGICAL REPORT By G. E. R. DEACON, F.R.S. NATIONAL INSTITUTE OF OCEANOGRAPHY The observations confirm the general picture of the water circulation described by A. F. Mohammed in Proc. Roy. Soc. B. 128, 1939, and give some new information about the surface layer. As plotted in Fig. i, the surface water at Stations 30 and 31 in the Straits of Tiran, and at Station 23 twenty miles farther north on the east side of the gulf, had a salinity less than 40-6% which can be attributed to the inflow of water from the Red Sea. There is some indication that the inward movement has a greater influence on the east side of the gulf since the surface salinity at Station 17 nearly half-way up the gulf is only 40-63% . For the remainder of the gulf, including all stations north and west of a line from Station 26 to Station 17, the water between the surface and 20 fathoms can be regarded as almost isothermal and isohaline, with a temperature of 21 to 22 C. (in January) and a salinity of 40-7 to 4O-8% . Excepting Stations 31, 30, and 25, the observed surface temperature appears to depend more on the time of day at which the measurement was made than the position of the station in the gulf. When plotted against time of day (Fig. 2) the temperatures lie fairly closely about a curve of diurnal temperature change which has a maximum at approximately 13.00 hours. The bathythermograph observations made at all the stations always show a temperature less than that measured by taking a surface sample and using a thermometer. Some of the differences can be attributed to the shallower depth of the sample scooped up in a surface sampler, and to the existence of an appreciable thermal gradient in the first foot or two of water. The differences between the thermometer and bathythermograph readings when plotted against the time of day (Fig. 3) lie fairly closely about a curve with a maximum of '55 C., which is very similar to that showing the diurnal temperature variation (Fig. 2) at 13.00 hours. The differences between the readings at the surface and a depth of 40 metres on the bathythermograph slides (Fig. 4) shows that this differ- ence, which varies between 0-2 and 0-6 C., varies according to a similar curve. It is expected that some further information about the interchange of heat and water vapour between the surface water and the air can be obtained from the data, and, when some attempt is made to smooth out the diurnal temperature variations, one or two useful indications of the surface movements ; but the best that can be done at present is to regard the upper 40 metres of water as more or less uniform, excepting Stations 31, 30, and 23. These appear to be influenced by the inflow of surface water from the Red Sea. Reference to Fig. i will also show that the stations near the eastern shore in the southern part of the gulf appear to be influenced by a more recent inflow of water than those farther north and west. FIG. i. Surface salinities. The underlined figures are the station numbers. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 161 3! 5 - 4 3O 3 - 2 25 1 - u o 22-O 13 3 "o 7 ^ *-2 ^ X X x*> S \ A * fc ^ 6 x x -5 X "3x ' 4 .6 / -21 '-18 -9 X x 5 I2/ ,, 29' '" * 27 28 4 ,A 26 / IO / 3 j' ' on 21-2 ' 24 z . 2 i i i i i i i i i O8OO O9OO IOOO MOO I2OO I3OO I4OO I5OO I6OO I7OO Hours (local Mean Time) FIG. 2. Surface temperature in relation to time of day. (Numbers refer to stations.) 162 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 31 c 0-8 ^ O-7 O> Q. -- g 0-6 01 |6 -14 - 8 9 I JTO-S r ^ ^ >. 0) 1 O-4 *F'./t *i3 ,9 XN X X 5 _w 24 x x .23 * 2 -3 X '> 3 2 10 / 5 s S. jl O-2 *$>' .1 E 3 s - a> jE :O-4 - ' HO ^ H7 H8 25 -3 X ^ X 5 -c ^' ** w "o & fo *4 2 6 v O ^27^^ 28 .5.|...2H4 . 8 |To.2 1- E - 24^ 29 ' O-l O8OOO9OO IOOO IIOO I2OO I3OO WOO I5OO I6OO I7OO Hours(L-M.T) FIG. 4. Difference between temperature at surface and at 40 metres, plotted against time of day. (Numbers refer to stations.) III. SPONGES By MAURICE BURTON, D.SC. The sponges represent thirty-three species, and although their study has resulted in little of unusual interest, a useful addition to the faunal list of the Red Sea area has been made. In addition, it has been possible to establish the correct identity of some of the forms described by Keller (1889 and 1891), which has long been in doubt. Most of the thirty-three species are common to the Indian Ocean fauna, some having been recorded also from Australia or the Indo-Pacific. It is of interest to note, however, that twelve species appear to be endemic, but this may be due largely to gaps in our knowledge of the Indian Ocean fauna. Furthermore, there are three species (Leuconia nausicae, Tethya aurantium, and Pseudosuberites mollis) belonging more properly to the Mediterranean fauna. The commonest form in the Gulf of Aqaba seems to be Callyspongia viridis, which, according to the members of the expedition, is 'abundant everywhere'. LIST OF SPECIES AND SYSTEMATIC NOTES Order CALCAREA Leucosolenia canariensis (Michlucho-Maclay) Nardoa canariensis Michlucho-Maclay, 1868: 221. Leucosolenia canariensis, Dendy & Row, 1913: 724. Occurrence. Mualla, 30.1.49, under rocks at low tide ; Sherm-el-Moiya, 3.ii.49. Remarks. A greyish white, typical specimen, 10 mm. across. Distribution. Arctic; Mediterranean; Cape Verde Islands; Canaries; Red Sea; Mauritius; NW. Pacific (Commandorski Islands). Leucosolenia tenuipilosa Dendy Leucosolenia (Clathrina) tenuipilosa Dendy, 1905: 227, pi. xiii, fig. 9. L. canariensis (pars), Thacker, 1908: 762. Clathrina tenuipilosa, Row, 1909: 185. Leucosolenia tenuipilosa, Dendy & Row, 1913: 723. Occurrence. Dahab, 14.^.49 ; Abu Zabad, n.ii.49. Remarks. There are a number of typically cushion-shaped specimens, up to 30 mm. across, which were brown or fawn in formalin, and now, in spirit, are coloured a greyish brown. Distribution. Ceylon; Red Sea; Cape Verde Islands. Grantessa glabra Row Grantessa glabra Row, 1909: 203, pi. xix, figs. 5-6; Dendy & Row, 1913: 752. Occurrence. Sherm Sheik, 11.1.49; Abu Zabad, io.ii.49, on ree f a ^ low tide. Distribution. Red Sea. 164 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Leuconia bathybia (Haeckel) Dyssycum bathybia Haeckel, 1869: 241. Leucaltis bathybia, idem, 1872: 156, pi. xxviii, fig. 2. Leucandra bathybia, Dendy & Row, 1913: 773. Occurrence. Sherm Sheik, 2.11.49, 2 fms. ; Sanafir, 6.11.49. Remarks. The four specimens may possibly represent two well-marked varieties, and, since the species was originally subdivided in this manner, it may be worth while to consider them in this light. The first specimen is the smaller, a few millimetres high, and of typical form and colour. The skeleton is arranged as Haeckel described it, and the rays of the large quadriradiates have a maximum of 0-4 by 0-032 mm. The other three range from a few millimetres high to 16 mm. high by 12 mm. diameter. Again, the external form is typical, as well as the spiculation. But in these three the rays of the quadriradiates have a maximum of 0-96 mm. by 0*09 mm. Either the first of the present specimens represents Haeckel's var. perimina and the other three var. arabica, or, what is much more likely, we have to deal with a species showing a tendency to vary widely in the measurements of the spicules. The first specimen and two out of the group of three were found at the same station, Sherm Sheik. Distribution. Red Sea ; ? Australia. Leuconia nausicae (Schufmer) Leucaltis nausicae Schuffner, 1877: 407, pi. xxiv, fig. I. Leucandra nausicae Dendy & Row, 1913: 774. Occurrence. Sanafir, 9.1.49 ; Tiran, 10.1.49 ; Abu Zabad, 11.11.49, on ree ^ a ^ l w tide. Remarks. The two specimens seem to agree closely with the description of the holotype, which is the only other recorded specimen. Presumably Row (I.e.) examined this and, as a consequence, the species was transferred to Leucandra. It is difficult, therefore, to accept Topsent's (1937: 14) remark that 'Leucaltis Nausicae Schuffner se confond vraisemblablement avec Leucetta solida (O. Schmidt) '. Distribution. Mediterranean. Kebira uteoides Row Kebira uteoides Row, 1909: 210, pi. xx, figs. 8-9, text-figs. 7-8; Dendy & Row, 1913: 785. Occurrence. Sherm Sheik, 2 fms., 2.11.49. Remarks. The single specimen, 20 mm. high, is typical, in both external appear- ance and the details of the skeleton. Distribution. Red Sea. Order TETRAXONIDA Stelletta purpurea Ridley (For synonymy see Burton, 1926.) Occurrence. Tiran, 10.1.49 ; Sanafir, 8 and 9.1.49 and 4.11.49 ; Sherm-el-Moiya, 3.11.49. Remarks. The spiculation of the several specimens shows the usual variation in THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 165 size. The main interest lies, however, in the external form. The smallest specimens, 10 to 15 mm. diameter, have the spherical or subspherical shape typical of the species, but in one or two cases these small spherical sponges have coalesced to give an ir- regular lobulated mass. In the larger specimens, 50 to 60 mm. across, on the other hand, the form is often extremely irregular, suggesting not only the coalescence of several smaller sponges but irregularities of growth due to environmental factors. Distribution. Red Sea ; Indian Ocean ; Malay ; Australasia ; Antarctic. Chondrilla sacciformis Carter (For synonymy see Burton, 1924.) Occurrence. Sherm-el-Moiya, 3.11.49. Distribution. Indian Ocean ; Malay. Chondrosia reniformis Nardo Chondrosia reniformis Nardo, 1847: 272. Occurrence. Abu Zabad, 11.11.49. Remarks. The two specimens appear to be typical except that there is a sparse accumulation of fine sand grains in the outer layers of the cortex. Distribution. Atlantic coast of Europe ; Mediterranean ; South Africa (Stil Bay) ; Indian Ocean ; Malay ; Australia. Chrotella cavernosa (Lamarck) Tethya cavernosa Lamarck, 1813: 70; 1815: 385. T. cranium var. australiensis Carter, 1886: 127. Cinachyra australiensis, Burton, 1934: 523. (For further synonymy see Burton, I.e.) Occurrence. Mualla, 30.1.49, at low tide under rocks. Distribution. Red Sea ; Indian Ocean ; Malay ; Australia ; Philippines. Tethya aurantium (Pallas) (See Burton 1924 and 1949: 122.) Occurrence. Sherm Sheik, 2.11.49, 11.1.49, an d 2 - n -49 Tiran, 10.1.49; Mualla, 30.1.49, at low tide under rocks. Remarks. The five specimens, all somewhat flattened, are fawn, orange, or red (in formalin) and measure 7, 8, 12, 18, and 21 mm. across respectively. Distribution. Arctic ; North Atlantic ; West Indies ; Mediterranean. Tethya robusta Bowerbank (For synonymy see Burton, 1924.) Occurrence. Mualla, 30.1.49, under rocks at low tide ; Abu Zabad, 10 and 11.11.49, on reef at low tide. Remarks. The six specimens measure 13, 15, 21, 25, 26, and 28 mm. across respec- tively. The colour (in formalin) is pink to red. There is, however, another specimen 166 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA consisting of five lobes set in a horizontal plane, each lobe being about 20 mm. across. Its colour was a cerise-red in formalin. Clearly this specimen has been formed by the complete coalescence of five adjacent individuals. It is not unknown for two speci- mens to fuse in this way, but five is unusual. The spiculation is typical in all but two specimens, which lack the larger micrasters. In other words, these two should be assigned to Tethya japonica Sollas. In 1924 I suggested that this so-called species was probably a reduced form of T. diploderma Schmidt (= T. ingalli Bowerbank), but it now seems that it is a mixture of the reduced forms of both T. robusta and T. ingalli. Distribution. Australia; Malay; Indian Ocean. Pseudosuberites mollis Topsent Pseudosuberites mollis Topsent, 1925 : 9, fig. 2m. Occurrence. Mualla, 30.1.49, under rocks at low tide. Remarks. The sample consists of three fragments of a soft and delicate sponge, having approximately the characters described by Topsent (I.e.). The spicules are slightly larger, 0-15 to 0-45 by 0-005 to 0-008 mm., as compared with 0-175 to 0-315 by 0-0065 mm - m the holotype, but the variations in the shape of the spicules are similar to those figured by Topsent. Distribution. Mediterranean (tang de Thau). Haliclona toxophorus (Hentschel) Gellius toxophorus Hentschel, 1912: 392, pi. xxi, fig. 46. G. toxotes, idem, I.e. : 392, pi. xxi, fig. 47. Occurrence. Sherm Sheik, 11.1.49. Remarks. The two small fragments are evidently from one sponge which formed a flattened, massive incrustation, with oscules slightly raised. Almost transparent, soft and compressible, delicate in texture, the specimen appears to be denuded of flesh, the skeleton, an isodictyal and unispicular network, being held together by spongin at the nodes. The megascleres are oxea, with a tendency to become strongy- lote at one or both ends, 0-24 by 0-012 mm. The microscleres are toxa, 0-02 to o-i mm. across. The two species described by Hentschel were sufficiently closely related, judging by the original descriptions, to suggest their identity one with the other. The inter- mediate character of the present material adds point to this. Distribution. Malay. Adocia dendyi (Burton) Toxochalina robusta Dendy, 1905: 139; idem, 1921: 29. T. dendyi Burton, 1931 : 340, fig. 26. Nee Toxochalina robusta Ridley. Occurrence. Sherm Sheik, 11.1.49. Remarks. The several specimens are all small and cushion-shaped, with con- spicuous oscules 2 to 3 mm. diameter. The colour, in spirit, is brownish grey, and THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 167 the texture soft, compressible, elastic. The main skeleton is a close-meshed reticula- tion of fibres, the ascending fibres multispicular (3 to 4 spicules), the connectives unispicular. The tangential dermal skeleton is very much as figured by me (I.e., fig. 26) and is unispicular. The spicules are oxea o-i by 0-004 mm., and toxa of about the same length. Distribution. Indian Ocean. Callyspongia viridis (Keller) Dactylochalina viridis Keller, 1889: 391, pi. xxiii, figs. 37-43. Occurrence. Sherm Sheik, 2 and 3.ii.49 ; Tiran, io.i.49 ; Abu Zabad, 10 and n.ii.49, on reef at low tide; Dahab, I3.i.49 and 14.^.49; Sanafir, 4, 5, and 6.11.49. Remarks. Of the eleven specimens, only one is almost identical with that figured by Keller (I.e., fig. 37), nine of the remainder being irregularly massive, on the whole smaller, and the eleventh being no more than a thin incrustation on a coral. All have the typical vents and the typical pore-sieves (Keller, I.e., fig. 40), although in some cases the pore-sieves are less strongly marked. In a few cases, at least, the characters of the surface have been blurred by preservation in formalin. The characters of the skeleton are comparatively uniform for the nine irregularly, massive specimens, but the typical specimen and the thin incrustation show features which merit special notice. In the nine specimens the network of the main skeleton consists of well-marked primary or ascending fibres which branch, as they run to the surface, in a somewhat irregular manner. At the centres of the fibres is a more or less continuous core of spicules arranged in an untidy manner (almost irregularly sub-plumose), often with individual spicules projecting from the fibres. The primary fibres are connected by secondary fibres, thinner than the primaries, and forming often an irregular network. In these the spicules are arranged, usually, uniserially ; but, again, individual spicules may project, at right angles to the main series, beyond the surface of the fibres. The tangential skeleton at the surface is a close-meshed network of fibres, cored by uniserially arranged oxea, and showing no obvious differentiation into primary and secondary meshes. The average diameter of the meshes is 0-04 mm. The oxea vary from 0-08 to 0-16 by 0-004 to 0-005 mm. The main skeleton of the one typical specimen (i.e. externally typical) is unlike that of the nine specimens in that it approaches the ceraochalinoid condition. It is a very close-meshed reticulation of thick fibres which appear at first sight to be aspiculous. In general it resembles that shown in Keller's fig. 39. On closer examina- tion, however, it can be seen that the spicules are present, are reduced in numbers, and seldom more than 0-002 mm. thick ; and often a spicule may be discontinuous throughout its length (as though breaking up) . As a result of comparing the external forms of these sponges, as well as the struc- ture of their skeletons, there seems little doubt that they are all conspecific and that the variation in their skeletons is unimportant. Generally speaking, it seems that in the younger sponges and the newer tissues the reticulation of the fibres is more loose and the fibres themselves more heavily cored with spicules ; that with maturity the skeleton is more closely knit and the proportion of spicule to spongin decreases (cf . 168 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Burton, 1926 : 265). One further point may be mentioned. In the specimen, described above as typical, the spicules have the appearance, as a result of their slender build and the discontinuous structure already referred to, of being dissolved or absorbed. Whether, in fact, this is the case is, however, problematical. The colour of the present specimens, in formalin, was grey to fawn. Distribution. Red Sea. Gelliodes fibulatus Ridley Gelliodes fibulatus Ridley, 1884: 427, pi. xxxix, fig. i, pi. xli, fig. b; Ridley & Dendy, 1887: 47, pi. xii, fig. 2; Lendenfeld, 1887: 793. Pachychalina fragilis, Lindgren, 1897: 481; idem, 1898: 290. Gelloides ramosa Kieschnick, 1898: 47. ? Pachychalina conulosa, idem, I.e.: 51. Gelliodes ramosa, idem, 1900: 565, pi. xliv, fig. 3. ? Pachychalina conulosa, idem, I.e. : 568, pi. xliv, fig. 8. Gelliodes fibulatus, Hentschel, 1912: 393. Sigmaxynissa fibulata, Burton, 1928: 115. Occurrence. Graa, 30.1.49; Sherm-el-Moiya, 3.11.49; Sanafir, 6.ii.49. Remarks. It is somewhat surprising to find what appear to be typical examples of this species so far west as the Gulf of Aqaba. All records previously have been for the Malay region and the Indian Ocean (Andaman Islands). Distribution. Malay; Indian Ocean; (? Australia). Mycale euplectellioides Row Esperella euplectellioides Row, 1911: 333, pi. xxxvii, fig. 12, text-fig. 16. Mycale euplectellioides, Burton, 1926: 80. Occurrence. Sherm Sheik, 2.11.49; Graa, 30.1.49; Dahab, 13.1.49; Sanafir, 4 and 6.ii.49. Remarks. The sponge occurs in irregular masses on coral, the largest being some 30 mm. across. Externally there is a close resemblance to the type, and from the condition of the several specimens, when removed from the formalin in which they were originally preserved, it is clear that a copious amount of mucus is present in life. The skeleton is typical except that microscleres are extremely rare, none being found except in a section from one specimen, which contained a few sigmata, 0-05 to 0-08 mm. chord, and one anisochela 0-024 mm - chord. Distribution. Red Sea ; Suez Canal. Mycale (Carmia) suezza (Row) Esperella suezza Row, 1911 : 338, fig. 18. Occurrence. Mualla, 31.1.49; Dahab, 14.11.49. Remarks. Two samples are assigned doubtfully to this species. The first is a thin incrustation, orange-coloured in formalin, and a larger, irregularly massive sponge, having the same colour and general appearance. The skeleton has the same structure as the holotype of Mycale suezza, but in neither specimen has it been possible to find a single microsclere. Distribution. Red Sea. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 169 Mycale (Aegagropila) erythraena (Row) Esperella erythraena Row, 1911: 340, fig. 19. Mycale erythraena, Burton, 1926: 80. Occurrence. Dahab, 4.11.49. Remarks. The single specimen forms a thin, irregular incrustation on coral. Its colour, in formalin, was grey. The arrangement of the skeleton approximate closely to the type, and the megascleres are typical in form and size ; but in spite of repeated searching not a single microsclere has been found. Distribution. Red Sea ; Suez Canal. Genus PARISOCIELLA gen. n. Type Species. Esperiopsis anomala, Ridley & Dendy, 1886 : 341. Diagnosis. Mycaleae with skeleton an irregular reticulation of spongin fibres cored by slender tylostyli; microscleres, when present, degenerate anisochelae palmatae and toxa. Parisociella anomala (Ridley & Dendy) Esperiopsis anomala Ridley & Dendy, 1886: 341 ; idem, 1887: 84. Ceraochalina gibbosa Keller, 1889: 386, pi. xxiv, fig. 44. Ophlitaspongia arbuscula Row, 1911: 347, pi. xxxix, fig. 22, pi. xl, fig. 25, text-fig. 22. O. horrida, idem, I.e.: 349, pi. xl, fig. 26, text-fig. 23. Occurrence. Sanafir, 4 and 9.11.49, along the shore among rocks; Abu Zabad, 10.11.49, on reef at low tide. Diagnosis. Sponge typically branching, surface uneven, minutely hispid ; oscules not apparent ; texture soft, elastic ; colour alive red, in spirit greyish yellow to dark grey; main skeleton an irregularly isodictyal reticulation of fibres cored by mega- scleres ; dermal skeleton of radiating brushes of megascleres ; megascleres tylostyli, slender and often appearing as styli, 0-25 to 0-3 by 0-002 to 0-005 mm - i microscleres usually absent and never plentiful, anisochelae palmatae, o-oi mm. chord, and toxa, 0-02 to 0-06 mm. long. Remarks. The diagnostic features of this species are unsatisfactory, since the microscleres, even when present, exist in such small quantities and are difficult to find. Further, the main skeleton is so like that of Mycale euplectellioides, growing in the same habitat, that only the external form remains as a guide to identification. If, therefore, the particular specimen is macerated or fragmentary the possibility of wrong identification is great. The present three specimens include a fragment of a branch, which is macerated, and two extensive, but low, incrustations on pieces of coral. The colour, in formalin, was orange and yellowish brown, in spirit, yellow or brown. No microscleres were found. Distribution. Red Sea ; Honolulu. Lissodendoryx cratera (Row) Myxilla cratera Row, 1911: 343, pi. xxxvii, fig. 13, text-fig. 20. Occurrence. Abu Zabad, n.ii.49. Distribution. Red Sea. 170 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Agelas mauritianus (Carter) Ectyon mauritianus Carter, 1883: 310, pi. xii, fig. 3. Agelas mauritianus, Ridley & Dendy, 1887: 164, pi. xxix, fig. 10. A. cavernosa Thiele, 1903: 963, fig. 28. A. mauritiana, Dendy, 1905: 174. Occurrence. Sanafir, 6.11.49. Remarks. A fairly large fragment which, in formalin, was pink outside and orange in the interior. Distribution. Indian Ocean ; Malay. Halichondria glabrata Keller Halichondria glabrata Keller, 1891: 311, pi. xvi, fig. 9; Burton, 1926: 75. Occurrence. Abu Zabad, 11.11.49. Remarks. A single, thinly encrusting specimen, in colour pale brown, both in formalin and in spirit. Distribution. Red Sea. Rhaphoxya typica Hallmann Rhaphoxya typica Hallmann, 1917: 643, pi. xxix, fig. 3, pi. xxxviii, figs. 8-9, pi. xxxix, fig. 5, pi. xlii, figs. 1-2, text-fig. 17. Occurrence. Sanafir, 6.11.49 ; Abu Zabad, 10.11.49, on reef at low tide. Remarks. The several species which may be assigned to Rhaphoxya are mainly Australian and none has been previously recorded from the Red Sea, although Anacanthaea nivea Row might conceivably belong to this genus. Yet the present two specimens clearly belong to Rhaphoxya and are almost certainly conspecific with the genotype. They are both encrusting, but their general appearance and the characters of the surface agree closely with those described and figured by Hallmann, except that the pore-areas (?), in his pi. xxxviii, fig. 8, are not so numerous in the 'Manihine' sponges. There is, also, a close agreement in the shape of the spicules and their arrangement in the skeleton, except that the trichites are not numerous and, as far as can be seen, do not form dragmata. A striking feature of the anatomy concerns the presence of numerous oval groups of cells, looking very like embryos, which they may well be, except that they vary somewhat in size, from 0-08 to 0-2 mm., with 0-12 mm. as the average, across the long axis. The tissues of the sponge contain numerous brown pigment cells in the surface layers, and the ' embryos ' lying in the surface tissues are also filled with them. Distribution. Australia. Order KERATOSA Aplysilla lacunosa Keller Aplysilla lacunosa Keller, 1889: 356, pi. xxii, figs. 19-22. Occurrence. Sanafir, 6.11.49. Remarks. A single, very small, incrusting specimen, purple in colour, showing the typical fibres (see Keller, I.e., pi. xxii, fig. 22). Distribution. Red Sea. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 171 Megalopastas erectus Row Megalopastas erectus Row, 1911: 360. Occurrence. Sherm Sheik, 11.1.49; Dahab, 14.11.49. Remarks. The two specimens form irregular encrustations, with the surfaces irregularly conulose. The colour of one, in formalin, was purple, in spirit it turned to a deep violet ; in the other it was fawn in formalin and the same in spirit. Distribution: Red Sea. Spongia officinalis Linnaeus, var. arabica (Keller) Euspongia officinalis, var. arabica Keller, 1889: 342; Topsent, 1906: 558; Row, 1911: 379. Occurrence. Abu Zabad, 10 and 11.11.49, on ree f a * l w tide ; Sherm-el-Moiya, 3.11.49 ; Sanafir, 9.1.49. Remarks. There are two typical specimens, two very small specimens in which the skeleton only remains and which are doubtfully assigned to this species, a fifth, typical but very small, and a sixth specimen which agrees in general appearance, but has the internal tissues so crowded with sand that a better identification is not possible. The colour in formalin varies from fawn (the specimens without flesh) to dark brown. Distribution. Red Sea. Heteronema erect a Keller Heteronema erecta Keller, 1889: 340, pi. xx, figs. 4, 7, 8; Topsent, 1906: 558; Row, 1911: 369. Duriella nigra Row, 1911 : 370, pi. xli, fig. 29. Occurrence. Dahab, 3.1.49 and 2.11.49 and 14.11.49, shore ; Sanafir, 5.11.49. Remarks. The type of Duriella nigra and Row's specimen of Heteronema erecta are almost identical in external form though they differ in the structure of the skeleton. Both specimens are, however, massive and lack the digitiform processes of the type of H. erecta. There is also available in the British Museum collection a preparation from Keller's type, and comparing this with Row's specimens suggested, in the first place, that the only difference between Duriella nigra and Heteronema erecta lay in the much greater amount of sand in the fibres of the latter. The 'Manihine' speci- mens, four in all, have a sufficiently general resemblance to each other, and to the specimens described by Keller and Row, to be considered alongside them. In these, two have a skeleton approximately similar to that of Duriella nigra, one is much more like Heteronema erecta, and the fourth is intermediate between the two. With seven specimens thus available for comparison it seems certain that the variation in the skeleton of this species (for Duriella nigra and Heteronema erecta are here accepted as conspecific) is similar to that shown by me (1934, figs. 18-33) f r Dysidia fragilis. In other words, that according to the amount of sand present the skeleton will vary from clearly defined ascending fibres cored with sand, connected by a secondary network free of it, to a dense network in which the spongin of all fibres is almost entirely obscured by a heavy intake of sand, with no perceptible differentiation into primary (or ascending) and secondary fibres. 172 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Supporting such a view is the fact that the amount by which the fibres are im- pregnated with sand varies from one part to another of the skeleton of any individual sponge. The colour of the 'Manihine' specimens ranged, in formalin, from brown to a deep purple-brown. Distribution. Red Sea. Carterispongia clathrata (Carter) (For synonymy and discussion see Burton, 1934: 574.) Occurrence. Sherm Sheik, 11.1.49; Mualla, 31.1.49; Dahab, 13 and 14.^.49; Sanafir, 9.1.49 and 4.ii-49 ; Sherm-el-Moiya, 3.11.49. Remarks. The several fragmentary specimens have the typical cavernous appear- ance. The skeleton differs considerably, however, from one individual to another, and these differences seem to offer a gradation from the typical skeleton of this species to that of Euryspongia lactea. It is possible, therefore, that Euryspongia may ultimately prove to be synonymous with Carterispongia. The colour of the different specimens, in formalin, ranged from fawn or brown, to purple, with occasional pink patches. Distribution. Indian Ocean; Australia; (? West Indies). Hircinia ratnosa Keller Hircinia ramosa Keller, 1889: 345, pi. xx, fig. 5. H. schulzei Dendy, 1905: 221, pi. xvi, fig. 3. H. ramosa, Row, 1911: 372; Burton, 1934: 579, pi. i, fig. n, text-fig. 16. Occurrence. Sanafir, 8.1.49 and 9.11.49, littoral, growing among rocks. Remarks.The two specimens are typical in the structure of the skeleton but show less of the ramose external form. One of them is low-lying and massive, with occa- sional ramose portions. The colour of the two specimens, in formalin, was fawn and brown respectively, in spirit it is now olive-green and brown. Distribution. Red Sea; Ceylon; Australia (Barrier Reef). Cacospongia ridleyi, sp. n Cacospongia cavernosa Ridley, 1884: 590; nee C. cavernosa, Autt. Occurrence. Abu Zabad, 11.11.49. Remarks. The name Cacospongia cavernosa has been used by many authors for sponges from the Indian Ocean, Mediterranean, and the West Indies. Pallas (1766 : 395) appears to have been the first to use the trivial name, but his Spongia cavernosa is not recognizable except as one of the Keratosa. Esper's (1794: 189) 5. cavernosa, based on Pallas 's specimen, has been inadequately re-described by Ehlers (1870: 30) ; and Lamarck's specimen (1813: 371) has been shown by Topsent (1930: 13) to be conspecific with Ciocalypta penicillus Bowerbank. Ridley (1884: 590) recorded specimens under Cacospongia cavernosa from the Seychelles, and it is with these THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 173 that the present specimens are to be identified. C. ridleyi agrees closely with C. cavernosa Schmidt (as re-described by Schulze, 1879) m external form, but the skeleton has larger meshes and the fibres are more heavily cored with sand-grains and other foreign bodies. It is, however, impossible to say, in the present state of our knowledge, whether the sponges from Seychelles and the Gulf of Aqaba represent a simple variety of the Mediterranean form. As a temporary measure at least they are here given full specific rank. Distribution. Indian Ocean. REFERENCES BURTON, M. 1924. The Genus Chondrilla. Ann. Mag. nat. Hist. (9) 14: 206-209. 1924. A revision of the Sponge Family Donatiidae. Proc. zool. Soc. Lond. : 1033-1045, i pi. 1926. Sponges [in] Zoological Results of the Suez Canal Expedition. Trans, zool. Soc. Lond. 22: 71-83, 7 figs. 1926. Stelletta purpurea, Ridley, and its variations. Ann. Mag. nat. Hist. (9) 18: 44-49. 1928. Report on some Deep-Sea Sponges from the Indian Museum collected by R.I. M.S. Investigator. Part II. Rec. Indian Mus. 30: 109-138, 2 pis., 9 text-figs. 1931. On a collection of marine sponges mostly from the Natal coast. Ann. Natal Mus., 4: 337-358, i pi., 9 text-figs. I934- Sponges. Sci. Rep. Gt. Barrier Reef Exped. 1928-29, 4: 513-621, 2 pis., 33 text-figs. 1948. The Ecology and Natural History of Tethya aurantium Pallas. Ann. Mag. nat. Hist. (12) 1: 122-130. CARTER, H. J. 1883. Contributions to our knowledge of the Spongida. Ann. Mag. nat. Hist. (5) 12: 308-329, pis. xi-xiv. 1886. Descriptions of Sponges from the Neighbourhood of Port Phillip Heads, South Australia (contd.). Ann. Mag. nat. Hist. (5) 17: 112-127. DENDY, A. 1905. Report on the Sponges collected by Prof. Herdman at Ceylon. Rep. Pearl Fish. Manaar, Suppl. 18: 57-246, 16 pis. DENDY, A., & Row, R. W. H. 1913. The Classification and Phylogeny of the Calcareous Sponges. Proc. zool. Soc. Lond.: 704-813. HAECKEL, E. 1869. Prodromus eines Systems der Kalkschwamme. Jena. Z. Naturw. 5: 236-254. 1872. Die Kalkschwamme: eine Monographic, 2 Bd. & Atlas. Berlin. HALLMANN, E. F. 1917. A Revision of the Genera with microscleres included, or provisionally included, in the Family Axinellidae. Proc. Linn. Soc. N.S.W. 40: 634-675, 9 pis., 4 text-figs. HENTSCHEL, E. 1912. Kiesel- und Hornschwamme der Aru und Kei Inseln. Abh. senckenb. naturf. Ges. 34: 291-448, 9 pis. KELLER, C. 1889. Die Spongienfauna des Rothen Meeres. Z. wiss. Zool. Leipzig, 48: 311-405, 5 pis. 1891. Die Spongienfauna des Rothen Meeres. Z. wiss. Zool. Leipzig, 52: 294-368, 5 pis. KIESCHNICK, O. 1898. Die Kieselschwamme von Amboina. 66 pp. Jena. 1900. Kieselschwamme von Amboina. Denkschr. med.-naturw. Ges. Jena, 8: 545-582, 2 pis. LAMARCK, J. B. P. A. DE M. 1813. Sur les Polypiers empates. Ann. Mus. Hist. nat. Paris, 20: 294-312, 370-386, 432-458. 1815. Suite des Polypiers empates. Mem. Mus. Hist. nat. Paris, 1: 69-80, 162-168, 331-340. LENDENFELD, R. VON. 1887. Die Chalineen des australischen Gebietes. Zool. Jb. 2: 723-828, 10 pis. LINDGREN, N. G. 1897. Beitrag zur Kenntniss der Spongienfauna des Malaiischen Archipels und der Chinesischen Meere. Zool. Anz. Leipzig, 20: 480-487. 1898. Beitrag zur Kenntniss der Spongienfauna des Malaiischen Archipels und der Chinesischen Meere. Zool. Jb. Jena (Abt. Syst.), 11: 283-378, 4 pis. MICHLUCHO-MACLAY, N. 1868. Beitrage zur Kenntniss der Spongien. Jena Z. Naturw. 4: 221- 240, 2 pis. 174 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA NARDO, G. D. 1847. Osservazioni anatomiche sopra 1'animale marine detto volgarmente Rognone di mare. Atti. 1st. veneto, 6: 267-276. RIDLEY, S. O. 1884. Spongiida. Rep. Zool. Colls. Voy. H.M.S. 'Alert', London: 366-482, 582-630, 6 pis. & Dendy, A. 1886. Preliminary Report on the Monaxonida collected by H.M.S. Challenger. Ann. Mag. nat. Hist. (5) 18: 325-351. 1887. Monaxonida. Rep. Sci. Res. Voy. H.M.S. 'Challenger', Zool. 20: 1-275, 5 1 pl g - Row, R. H. W. 1909. Report on the Sponges collected by Mr. Cyril Crossland in 1904-5. Part I. Calcarea. /. linn. Soc. Lond. Zool. 31: 182-214, 2 pis. 1911. Report on the Sponges collected by Mr. Cyril Crossland in 1904-5. Part II. /. linn. Soc. Lond. Zool. 31: 287-400, 7 pis., 26 text-figs. SCHUFFNER, O. Beschreibung einiger neuer Kalkschwamme. Jena. Z. Naturw., xi, (2) 4: 403- 433. 3 Pis. THACKER, A. G. On collections of the Cape Verde Islands Fauna made by Cyril Crossland. Proc. zool. Soc. Lond.: 757-782, i pi., 12 text-figs. THIELE, J. 1903. Kieselschwamme von Ternate. Abh. senckenb. naturf. Ges., 25: 933-968, i pi. TOPSENT, E. 1906. fiponges recueillies par M. Ch. Gravier dans la Mer Rouge. Bull. Mus. Hist. nat. Paris, 12: 557-570. 1925. fitude des Spongiaires du Golfe de Naples. Arch. Zool. exp. gen. Paris, 63: 623-725, i pi., 27 text-figs. IV. TURBELLARIA: POLYCLADIDA By STEPHEN PRUDHOE THOUGH comprising only three specimens, the collection is an interesting one, since it includes three species which apparently have not been recorded hitherto from the Red Sea. The condition of the material is satisfactory, and it has been possible to supple- ment existing descriptions of the three species with some new details of their structure, more especially of the copulatory organs. Lastly, a brief historical account of the polyclad fauna of the Red Sea is given, together with a list of the species recorded. PLANOCERIDAE Planocera crosslandi Laidlaw, 1903 (FlG. I) A young adult specimen of this species was found in the fauna associated with coral at Sherm Sheik, 2 February. It measures about 28 mm. in length and about 20 mm. in maximum width, which occurs in the middle region of the body. FIG. i. Planocera crosslandi. Arrangement of eyes (dorsal view). zoo. i. A a 176 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA In the structure of the copulatory organs the present specimen agrees very well with the original description of P. crosslandi. The posterior region of the cirrus- cavity bears three very large hook-like structures, one of which is attached to the dorsal wall and the others to the subventral walls of the cavity. These structures are directed posteriorly and lie almost entirely in the spacious male antrum. Planocera crosslandi has been recorded hitherto only from British East Africa. LEPTOPLANIDAE Notoplana gardineri (Laidlaw, 1904) (FIG. 2) A single individual, provisionally assigned to this species, was found under a rock near the low-tide mark at Sherm Sheik, 15 February. Unfortunately the specimen is damaged, and, as a portion of its hinder region is lost, it is not possible to determine the structure of the female copulatory apparatus. Transverse serial sections of the copulatory organs of the type-specimen of this species have recently been presented to the British Museum (Natural History) by FIG. 2. Notoplana gardineri. Arrangement of eyes (dorsal view). Dr. F. F. Laidlaw. The series is incomplete, but, so far as it has been possible to make out, the male copulatory apparatus of the specimen from the Red Sea is indistinguishable, structurally and histologically, from that of the type-material of N. gardineri (Laidlaw), a species known hitherto only from Ceylon. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 177 The damaged specimen is somewhat pellucid and measures about 16 mm. in length and about 9 mm. in maximum width. The body is more or less oval in outline. No tentacles have been made out. The eyes are arranged in two elongate groups (Fig. 2). Those in the hinder region of each group are distinctly larger than the remainder and probably represent the tentacular eyes present in other species of Notoplana. The mouth occurs about 10 mm. from the anterior extremity of the body and opens into the hinder region of the pharyngeal chamber. The latter measures about 4-5 mm. in length and contains about 10 pairs of shallow lateral pockets. The male pore is situated at 3-5 mm. behind the mouth. As is usual in this genus, the ovaries and testes lie in the dorsal and ventral parenchyme respectively. The vasa deferentia unite to open into the proximal end of the arcuate seminal vesicle, which possesses a very thick coat of longitudinal and circular muscle-fibres. This vesicle opens, through the ejaculatory duct, into a well-developed, somewhat pear- shaped prostatic organ lying above the proximal end of the seminal vesicle. The ejaculatory duct projects well into the prostatic organ, the highly glandular epi- thelium of which completely invests the duct. In this epithelium there are seven elongate pockets, which, together with the ejaculatory duct, open into a small chamber situated in the posterior region of the prostatic organ. From the prostatic chamber a long ductus communis or prostatic canal passes through an extremely thick sheath of muscle-fibres and enters a very small penis-papilla lying in the shallow male antrum. The thick sheath appears to be a continuation of the muscula- ture of the prostatic organ and merges with that of the penis-papilla. There are numerous nuclei present among the muscle-fibres of the sheath, and they seem to congregate more particularly around the prostatic canal. N. gardineri appears to bear a very close resemblance to Notoplana otophora (Schmarda, 1859) which was also originally recorded from Ceylon. According to Stummer-Traunfels (1933), the 'ductus communis' or prostatic canal of the type- specimen of N. otophora is invested with a deep layer of parenchymatous tissue enclosed in a thick muscular sheath. On the other hand, in N. gardineri the prostatic canal is, as stated above, invested solely with an extremely thick musculature of longitudinal and circular fibres. Nuclei are abundant in this musculature, being particularly dense immediately around the prostatic canal. This difference between the two species might be accounted for by the fact that the type-specimen of N. otophora had, when examined by Stummer-Traunfels, apparently been stored in preserving fluid for about seventy years. During this time the tissues of the speci- men had, no doubt, undergone some maceration and possibly the histology of the structure through which the prostatic canal passes might originally have been similar to that occurring in N. gardineri. In other respects, except possibly in the number of eyes, the two species appear to be identical. Notoplana cotylifera Meixner, 1907 A single specimen was found in sponges associated with coral at Graa, 30 January. It agrees very well with the description of N. cotylifera Meixner, and, as in the original material, a well-developed sucker occurs between the genital pores. The most striking feature of the female copulatory apparatus in this species is 178 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA the pocket-like structure, which Meixner regards provisionally as a rudimentary accessory vesicle, opening into the vagina interna, near the ' shell '-chamber. A somewhat similar structure occurs in the present specimen, but in this instance it appears also to open on the dorsal surface of the body, anteriorly to the female genital pore. Unfortunately the condition of the tissues in this region of the body is not very satisfactory, and the presence of a dorsal opening requires confirmation. If a study of new material were to show that the dorsal opening normally occurred in this species, the accessory structure of the female apparatus would appear com- parable with the ductus vaginalis present in some other species of Polyclads. Notoplana cotylifera has been recorded previously from the Gulf of Tadjoura, French Somaliland, which is, of course, situated near the southern entrance to the Red Sea. Thus the occurrence of this species in the Gulf of Aqaba is not unexpected. The history of the Polyclad fauna of the Red Sea apparently begins in the year 1826, when the name Planaria mulleri was given by Audouin to a planarian figured, but not described, by Savigny in the same year. Two years later (1828) Leuckart described five new forms from Tor in the Gulf of Suez. This work was shortly followed by that of Ehrenberg (1831), in which a further four new species were described from Tor and the Isle of Ras el Gusr. The descriptions of all these ten species are very incomplete, and it does not appear possible to recognize any of the species with certainty. After 1831 no further species of Polyclads seem to have been recorded from this region until Boutan (1892) mentioned the occurrence of Pseudoceros violaceus (Schmarda) at Port Tewfik. Another thirty years elapsed before Meyer (1922) described three new species from Kosseir. Since the appearance of Meyer's work, Palombi (1928) has recorded, among other species, Idioplana australiensis Wood- worth 1 from the Port of Suez, and Melouk (1940, 1941) has described two new forms from the Biological Station at Ghardaqa. The results of the sporadic work done since 1826 indicate that our knowledge of the occurrence and distribution of Polyclads in the Red Sea is, in all probability, very incomplete. It may therefore be deemed useful to tabulate the species, including those in the present collection, that have so far been recorded from the Red Sea. The taxonomy of some of the species is very uncertain, and these are marked with an asterisk in the following table : Species Locality Cestoplana polypora Meyer, 1922. . Kosseir ' Craspedomata sp. ? ' Palombi, 1928 Cryptophallus aegyptiacus Melouk, 1940 *Euryleptaflavomarginata Ehrenberg, 1831 *Eury lept a praetexta Ehrenberg, 1831 Idioplana australiensis Woodworth, 1 898 * Leptoplana hyalina Ehrenberg, 1831 Gulf of Suez El Ataka & Ghardaqa Ras el Gusr Tor Port of Suez Tor [This species, the type of the genus Leptoplana, has been regarded by most early writers as a synonym of Leptoplana tremellaris (Miiller, 1774).] 1 Judging from Palombi's description, the material determined by him as Idioplana australiensis is probably not identical with that described by Woodworth. In fact, Palombi's material appears to be more closely related to the genus Idioplanoides Barbour, 1912, than to Idioplana Woodworth, 1898. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 179 Species Locality Leptoplana nadiae Melouk, 1941 . . . . Ghardaqa Notoplana cotylifera Meixner, 1907 . . . Graa Notoplana gardineri Laidlaw, 1903 . . . Sherm Sheik Paraplanocera marginata Meyer, 1922 . . . Kosseir *PlanariabilobataL&uc]iaTt, 1828. . . . Tor *Planaria bituberculata Leuckart, 1828 . . . Tor *Planaria gigas Leuckart, 1828 .... Tor *Planaria limbata Leuckart, 1828 .... Tor *Planaria mulleri Audouin, 1826 .... [P. bituberculata and P. mulleri have been generally regarded as synonyms of Stylochus suesensis Ehrbg. If this be accepted, P. mulleri has priority over 5. suesensis and therefore becomes the type-species of Stylochus Ehrbg.] *Planaria zebra Leuckart, 1828 .... Tor Planocera crosslandi Laidlaw, 1903 . . . Sherm Sheik Pseudoceros violaceus (Schmarda, 1859) . . Port Tewfik Stylochus coseirensis Bock, 1927 [nom. nov. pro Stylochus reticulatus of Meyer, 1922] . . . Kosseir *Stylochus suesensis Ehrenberg, 1831 . . . Tor & Port of Suez REFERENCES AUDOUIN, V. 1826. Explication sommaire des planches. Annelides. Descr. Egypte, etc. Hist. nat. 1(4): 76. BOUTAN, L. 1892. Voyage dans la Mer Rouge. Rev. Biol. Nord France, 4: 173-183. EHRENBERG, C. G. 1831. Symbolae physicae, etc. [Invertebrata.] Berolini. LAIDLAW, F. F. 1903. On the marine Fauna of Zanzibar and British East Africa, from collections made by Cyril Crossland in the years 1901 and 1902. Turbellaria Polycladida. Part I. The Acotylea. Proc. zool. Soc. London, 2: 99-113, pi. ix. 1904. Report on the Poly clad Turbellaria collected by Professor Herdman, at Ceylon, in 1902. Rep. Pearl Fish. Manaar, Pt. II: 127-136, pi. LEUCKART, F. S. 1828. In LEUCKART & RUPPELL: Neue wirbellose Thiere des Rothen Meeres. Atlas zu der Reise im nordlichen Afrika von E. Ruppel. Abth. I, Zoologie: n & 15, pi. iii. Frankfurt a. M. MEIXNER, A. 1907. Polycladen von der Somalikiiste, nebst einer Revision der Stylochinen. Zeitschr. miss. Zool. 88: 385-498, pis. xxv-xxix. MELOUK, M. A. 1940. A new Polyclad from the Red Sea, Cryptophallus aegypticus nov. spec. Bull. Fac. Sci. Egypt. Univ., 22: 125-140, pis. i-ii. 1941. Leptoplana nadiae, a new Acotylean Polyclad from Ghardaqa (Red Sea). Bull. Fac. Sci. Egypt. Univ., 23: 41-49, pi. i. MEYER, F. 1922. Polycladen von Koseir (Rotes Meer). Arch. Naturgesch., Abt. A, 87: 138-158, pis. i-iii. PALOMBI, A. 1928. Report on the Turbellaria. [Zool. Results of the Cambridge Expedition to the Suez Canal, 1924. xxxiv.] Trans, zool. Soc. Lond., 22: 579-631, pi. i. STUMMER-TRAUNFELS, R. VON. 1933. Polycladida (contd.). Bronns Klassen, 4 Abt. ic, (179): 3486-3596, pi. i. V. GEPHYREA By A. C. STEPHEN ROYAL SCOTTISH MUSEUM THROUGH the courtesy of the British Museum (Natural History), I have had the privilege of examining this collection. It is a small one containing seven individuals, referable to two genera of Sipunculids and one Echiurid. With one exception they have been recorded previously from the Red Sea, the exception being Siphonosoma koreae Sato, whose status is discussed. ECHIURIDAE Ochetostoma erythrogrammon (Leuckart & Riippell) Sherm Sheik, 15.^.49. Under rock at low tide. One specimen, body 30 mm., proboscis 22 mm. This species has already been recorded from a number of localities in the Red Sea. SIPUNCULIDAE Siphonosoma koreae Sato Sherm-el-Moiya, 3.ii.49. Associated with coral. One specimen, not fully extended, 115 mm. in length. A single specimen, which agrees closely with Sato's description (Sato, 1939: 379), was secured. The body is long and thin, pink in colour, and capped at both ends by areas of yellow colour, the posterior area being much less extensive than the anterior area. The body is translucent, the muscle-bands showing through clearly. The posterior end of the body is somewhat cone-like, and the yellow cap extends for a distance of 5 mm. The introvert is not fully extended, but the yellow area occupies some 20 mm. of the body. In the specimen described by Sato the colour of the body is given as greyish white. The skin has numerous papillae, prominent and closely packed on the posterior end and at the base of the introvert, small and scattered on the rest of the body. Sato described the papillae on the posterior end in his specimen as being less prominent than those on the introvert basis. In this specimen, however, they are of similar size. On the introvert basis the area of prominent papillae extends for about 4 mm. On the introvert the papillae are small and arranged on circular ridges. The longitudinal muscle is divided into 19 bands, as in Sato's specimen. This species was described by Sato from a single specimen taken at Gunzan in Korea on 2 September 1937. In his key and text it is described as being very similar to 5. cumanense (Keferstein), separable mainly by colour differences, especially the yellow caps, and by the character of the papillae on the basis of the introvert. The 182 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA specimen from Aqaba differs from the Korean one in the colour of the body and the greater prominence of the posterior papillae. In view of the somewhat protean nature of 5. cumanense, with its three well-marked and widely distributed varieties, of which two are common to both the Red Sea and Korean waters, as well as the differences between the two known specimens, it is possible that more material may show that it is not a distinct species but only another variety of 5. cumanense. Physcosoma pacificutn (Keferstein) Abu Zabad. ii.ii.4g. O n ree f a * l w tide. Two specimens. One partially ex- tended, 12 cm. in length. The other similar in size but much contracted. Greyish brown in colour, with scattered darker patches. Tiran. 10.1.49. Associated with coral. One large specimen; not fully extended, about 13 cm. in length. Uniformly greyish brown in colour with a number of darker bands anteriorly. Dahab. 3.ii.49. Shore. Two specimens of similar size to the above, but too contracted for measurement. Greyish brown in colour, with scattered darker patches. This species is widely distributed in the Indo-Pacific area and has already been recorded from the Red Sea. REFERENCE SATO, H. 1939. Studies on the Echiuroidea, Sipunculoidea, and Priapuloidea of Japan. Sci. Rep. Tdhoku Univ. (4) 14: 339-460, 5 pis., 60 figs. VI. MOLLUSCA By w. j. REES and A. STUCKEY THE mollusca are represented by 2 Loricates, 27 Gastropods, 13 Lamellibranchs, 4 Cephalopods, and a few Nudibranchs not reported on here. As this particular area has been thoroughly worked for mollusca by numerous workers, notably McAndrew and Issel, it is not surprising that no new forms were found. The Gastro- poda and Lamellibranchia call for no special description and have been listed with notes on distribution. Although the Cephalopoda are all well-known species, they are so well preserved that we have noted features of interest, standard measure- ments, and included photographs. Callistochiton heterodon var. savignyi appears to be rare and is only known from the northern part of the Red Sea ; it was not hitherto represented in the collections of the British Museum. Other species which appear to be confined to the Red Sea are Clanculus pharaonis, Trochus erythraeus, and Lithophaga hanleyana. All the remaining species are found either in the western part of the Indian Ocean or have a wide distribution in the Indo-Pacific. In the Cypraeidae Schilder (1938) has drawn attention to races of well-known species which are becoming differentiated in various areas, including the Red Sea. The classification of Indian Ocean Lamellibranchs, and indeed all Lamellibranchs, is in a very unsatisfactory state, and in the specific names we have adopted we have followed Thiele (1929-1934), Tomlin (1927), Smith (1897), and various papers by E. Lamy. Recent work on molluscs has shown that species seemingly identical, or appearing to have only minor points of difference, have distinct larvae and life histories, revealing that they are distinct. Elaborate lists of synonyms may therefore prove erroneous, and usually we have confined ourselves to referring the specimens to species with which they appear to be identical. Among the Cephalopods Octopus macropus is common in the Red Sea and in the Mediterranean. The remainder, 0. horridus Orbigny, 0. cyanea Gray, and Sepio- teuthis lessoniana Lesson, are at the western limit of their range, which extends to the Andaman Islands in 0. horridus and throughout the tropical Indo-Pacific in the other species. The Cephalopods of the Red Sea have been reviewed by Adam (1942) and a study of his list reveals that they are all either littoral and shallow water forms or have planktonic larvae which live close to the surface during their early life. 1 As examples of the former we have species of Octopus, Sepia, Sepioteuthis, and Doryteuthis, and of the latter (oceanic species) we have Symplectoteuthis, Tremoctopus, and Argonauta. Cephalopods characteristic of deep water, and even the Cranchiidae (pelagic species which spend much of their early life in the upper 500 metres), are absent. It has been pointed out by Thompson (1939) that there is a shallow sill near 1 We have excluded Spirula spirula (L.), the shells of which are recorded from the Red Sea. It is probable that these have their origin outside the area. zoo. I. 8. Bb 184 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Hanish Islands separating the Red Sea 'proper' from the Gulf of Aden. At about latitude 13 41' N. the depth of the sill is only 100 metres, and this may act as a geographical barrier to deep-water species. This is probably one reason why bathy- pelagic forms are absent in the Red Sea, but it does not explain the absence of Cranchiidae, which as larvae often occur right up to the surface. The normal inter- change of water over the sill (see Thompson) should carry the larvae into the Red Sea and another explanation is required. There are some grounds for believing that these forms are the young of little known bathypelagic species and they may not be tolerant to high salinities of 38%,, to 4o% such as are typical of the Red Sea. The following species are known only from the Red Sea; those marked by an asterisk are insufficiently known and may prove to belong to other species: Sepia savignyi Blainville, 1827 Doryteuthis arabica Ehrenberg, 1831 * Sepia gibba Ehrenberg, 1831 *Abralia steindachneri Weindl, 1912 *Sepia elongata Ferussac & d'Orbigny, Octopus robsoni Adam, 1941 1835-1848 Sepia dollfusi Adam, 1941 *Sepia trygonina Rochebrune, 1884 Four of the above are imperfectly known, and of the seventeen species recorded from the Red Sea, only four sound species can be regarded as endemic. It is possible that even this number may be further reduced when the cephalopod fauna of the western Indian Ocean becomes better known. Class LORICATA Family CRYTOPLACIDAE Callistochiton heterodon var. savignyi Pilsbry Locality: 30.1.49, Mualla, i specimen. This small Callistochiton was taken with two specimens of Acanthopleura haddoni. It has a total length of 13 mm. and a breadth of 7-5 mm. This variety was named by Pilsbry (1892) from a figure given by Savigny (Egypte, pi. 3, fig. 8). Our specimen has the following characters. Shell oval, distinctly ridged along the median line; the sides of the valves are only slightly curved. Valves greyish white with occasional irregular darker markings. Girdle buff-coloured with faint slate-grey vertical bands. Head valve, with 10 slightly denticulate ribs, 2 of these bifurcate anteriorly. Tail valve, distinctly narrower than head valve, with n radiating rays. Other valves with distinct but not backward projecting beaks. Lateral areas raised with 2 denticulate ribs. Central areas with 7-8 narrow deeply etched riblets on each side with a central smooth tract between them. This variety has affinities with C. adenensis Smith, but differs from it mainly in having only 10-11 radiating ribs on the anterior valve instead of 22 as in Smith's species. C. heterodon var. savignyi is only known from this northern part of the Red Sea. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 185 Family CHITONIDAE Acanthopleura haddoni Winckworth Acanthopleura sp. Haddon, 1886: 24. Chiton (Acanthochites) spiniger Issel, 1819: 235 [non Sowerby]. Acanthopleura spinigera, Sykes, 1907: 34; Tomlin, 1927: 292 [non Sowerby]. Acanthopleura haddoni Winckworth, 1927: 206. Localities: 2g.xii.48, Aqaba, just below low tide mark, 4 adults. 8.1.49, Sanafir, 2 adults. 30.1.49, Mualla, attached or under stones at low tide, 2 adults. 9.11.49, Sanafir, among rocks along the shore, i adult. This large and decorative chiton is known from the Suez area of the Red Sea under the name spinigera. The earliest record is that of Savigny (Egypte, pi. 3, fig. 4). Winckworth (1927) distinguishes the species from A. spinigera Sowerby, an Austra- lian and Indonesian species, and described specimens from Aden under the name A . haddoni. According to Winckworth the living animal reaches a length of 3 in. and our largest specimen is of this size, although it cannot be measured accurately because of contraction of the foot causing the animal to be bent in the form of a crescent. In all our examples the girdle is irregularly marked with black and olive bands. In the living animal the foot is of a salmon-pink colour. It is impossible at present to give an accurate picture of the distribution of this mollusc. We know it occurs in the Red Sea (at Aden, Suez, and the localities given above), but its occurrence outside this area becomes confused with that of A. spini- gera (Sowerby). Cyril Grassland (quoted by Sykes) notes that it is 'the common high tide chiton, everywhere in E. Africa, on the cliffs of coral-rag at Djibouti, Mombasa, Zanzibar, Wasin etc. ; also on stone on the edge of reefs of the East Coast of Zanzibar'. Class CEPHALOPODA Family LOLIGINIDAE Sepioteuthis lessoniana Lesson (Plate 28, figs, i and 2 ; Plate 29, figs. 5 and 6) 3i.xii.48, Station Ai, dip net, surface, Faraun Island, i?(3). 28.1.49, Aqaba, i$(i). 4.11.49, Sanafir, cast net, surface, iC = E>B>A Diameter of largest sucker . . . . .2-25 Length of ligula ....... 2-55 Indices Mantle width index . . . . . .80 Head width index . . . . . . 13'5 Sucker index (normal) . . . . .15 Octopus macropus Risso n.ii.49, Abu Zabad, on reef at low tide, i $. 3i.xii.48, Station Ai, Faraun Island, surface, imm. cJ. This well-known octopus needs no further description, but standard measure- ments are provided for comparison with those which already exist for the Caribbean population of this species (Table IV). The measurements indicate that the Red Sea specimens fall within the limits already known for the species. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 189 Distribution. The species occurs in the Caribbean, the NE. Atlantic, the Mediter- ranean, the Red Sea, and the Indo-Pacific to Japan and Australia. Its eastern limit appears to be the Marshall Islands. It has been recorded from the Red Sea by Wiilker (1920) and Weindl (1912), to mention only two records. TABLE IV Octopus macropus Risso (Measurements in mm.) Sex ....... $ juvenile^ Total length (including 3rd arm) . . 246 40 Dorsal mantle length .... 58 16 Eye to dorsal web .... 47 6 Width of body .... 36 36 10 Width of head .... 29 7 Arms Right Left Right Left ist 246 245 34 34 2nd. ..... 228 227 28 26 3rd ...... 208 177 23 22 4th ...... 186 190 20 20 Diameter of largest sucker . . 6 0-75 No. of gill filaments ... 1 1 Web formula . . . . B>A>C>D>E B>C = A=D>E Indices Mantle width index 62 62-5 Head width index ..... 50 44 Sucker index (normal) .... 10-5 4-7 Arm length index ..... 78-5 68 TABLE V Octopus cyanea Gray I n Sex ' $ ? Total length (including longest arm) .... 420 343 Dorsal mantle length ....... 52 55 Eye to dorsal web ....... 44 Width of body 46 38 Width of head ........ 40 35 Arms Right Left ist ......... 265 205* 2nd ......... f 280 190 3rd ......... 200* 260$ 4th igoj 2ioJ Diameter of ocellus ....... 8 5 Diameter of largest sucker ..... 6 5 No. of gill filaments ....... 7-8 9 Web formula D = C > B > A = E Arm formula ........ 2.1.3.4 or 1.2.3.4 Web depth ........ 47 * Arm incomplete, tip portion missing. f Arms too tightly coiled for accurate measurements. Reeneratin. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Indices Mantle with index . Head width index . Sucker index (normal) Arm length index . Web depth index 885 77 115 82 635 91 815 168 Octopus cyanea Gray (Plate 30) Octopus cyanea Gray, 1849: 15. Octopus marmoratus Hoyle, 1886: 227. Octopus horsti Joubin, 1898: 23. Polypus fontanianus Robson, 1920: 437. Polypus horsti, Wiilker, 1920: 51. 6.1.49, Sanafir, along shore, I $. 12.1.49, Sherm Sheik, in shallow water along shore, i ?. We have referred these two specimens to Octopus cyanea Gray, but as they present a different appearance to what is usually associated with 0. cyanea, the various features worthy of note are discussed below. Typical specimens, of which we have seen a number in the collections of the British Museum, are, as Robson says, 'mainly of a warm ochreous red suffused and maculated with purple, which may be very deep so as to render the animal homogeneously blackish or deep livid (in preservative) '. Our specimens, however, are of a buff or pale brownish colour, with an olive-green sheen, which is especially marked on the dorsal surface of the web and the base of the tentacles. The top of the head, between the eyes, is a deeper brown colour. The specimens are paler ventrally and the ventral side of the arms have the character- istic zebra-like marking which Robson regards as one of the most striking and constantly associated features of cyanea as a species. Colour notes made from the living animal state that the colour of the specimen taken on 12.1.49 was brown and that the zebra-like markings on the arms were of a light blue colour. The dark purple ocellus is well marked and surrounded by an ill-defined pale ring, as mentioned by Robson for his British Museum specimens (Nos. 4 and 8) . Specimen No. I is rather contracted ; the skin of the mantle is reticulated and has a number of scattered irregularly arranged cirri, which are more numerous between the eyes and on the fore part of the head. Specimen No. II is less contracted, and has four cirri arranged in a diamond pattern on the dorsal mantle and four to five prominent cirri on the fore part of the head. The ventro-lateral and anterior portion of the mantle carries a number of scattered cirri. There is also a curious fold of skin, on either side of the neck region postero-ventral to the eye, which effectively separates the ventral funnel region from the lateral face of the head. The dorso-lateral surface of the arms in both specimens have a double row of slightly raised, buff-coloured, simple papillae which have not been mentioned by any other writer. A re-examination of Gray's type of 0. cyanea and other specimens reveals the presence of these papillae, but they are more difficult to see than in our specimens, because they are obscured by the dark, ground colour normal in this species. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 191 The number of gill filaments in the demi-branchs, normally a good diagnostic feature in octopods, appears to be rather variable in the species (7-9 in our speci- mens). Robson gives 9-10 for Gray's type of 0. cyanea, and we have found that even in the same specimen one gill may have 7 filaments and the other 9 filaments per demi-branch (i $ from the Cocos-Keeling Islands). Standard measurements are given in Table V, but it has not been possible to give measurements of the arms in specimen I because they are too tightly coiled. The only other ocellate species recorded from this area is Octopus robsoni Adam, 1941, of which a complete description has not yet been published. Adam states that this octopod 'se caracterise a premiere vue par la presence d'une pake d'ocelles pourvue d'un anneau irise blanchatre, bleuatre ou mauve '. We have mentioned this species because our specimens approach nearer to it in colour and the arrangement of the cirri than to the typical form usually found in 0. cyanea. However, the character of the ocellus, without an iridescent ring, the zebra-like markings on the ventral surface of the arms, and the various indices which fall within the limits of 0. cyanea, leaves us in no doubt as to the identity of our species. Distribution. Octopus cyanea is a littoral species well known as a reef-inhabiting octopod, with a distribution ranging through the Indo-Pacific in tropical and sub- tropical waters from Hawaii to the Red Sea. Previous records from the Red Sea are given by Robson (1929) and Wiilker (1920). Class GASTROPODA Family HALIOTIDAE Haliotis varia L. 3i.xii.48, station Ai, shore of Faraun Island, 3 specimens. 20.i.49, Dahab, on mud flats at low tide, I specimen. 5.^.49, Sanafir, found on coral, I specimen, n.ii.49, Abu Zabad, on reef at low tide, 4 specimens. 15.^.49, Sherm Sheik, under rocks at low tide, i specimen and I juvenile. Dahab, found on coral, i specimen. Issel (1869) collected two specimens of H. varia from the Gulf of Suez. From the numbers obtained in our collection it appears to be fairly common in the Gulf of Aqaba. According to Pilsbry (1890) it has a wide distribution in the Indo-Pacific, being found in the following places: Australia and Philippines to China; Mozam- bique, Red Sea, Island of Bourbon, Mauritius, Ceylon, Nicobar Islands, Malay Archipelago. Family FISSURELLIDAE D io dor a ruppellii (Sowerby) Fissurella ruppellii Sowerby, 1838: 128. Fissurella costaria Vaillant, 1865: 109. Fissurella vaillanti Fischer, 1865: 244. Glyphis ruppellii, Pilsbry, 1890: 217, pi. 39, fig. 8. Diodora ruppellii, Tomlin, 1927: 289. I5.ii.49, Sherm Sheik, under rock at low tide, i specimen. Distribution. This molluscs seems to be common almost throughout the Suez zoo. i. 8. cc 192 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Canal according to Tillier & Bavay. It has frequently been reported at Suez (see Tomlin, 1927, for previous records). D. ruppelliiis found in the Western Indian Ocean, in the Red Sea, at Aden, Mauritius, and on the East African coast. Family PATELLIDAE Cellana rota (Gmelin) Patella rota, Issel, 1869: 233. Patella rota, McAndrew, 1870: 444. Patella variegata Reeve, 1842, pi. 136, fig. i. Cellana rota, Tomlin, 1927: 299. 12.1.49, Sherm Sheik, 6 specimens. 2O.i.49, Dahab, on mud flats at low tide, 2 specimens. Both McAndrew and Issel record this species as common; the former from the Gulf of Suez and the latter from the Gulf of Aqaba. Tomlin (1927) found it in the Suez Canal zone. Distribution. Red Sea, east coast of Africa, Reunion, and Madagascar. Family TROCHIDAE Clanculus pharaonis (L.) 30^.49, Mualla, among rocks and coral at low tide, I specimen. This is one of the most characteristic molluscs of the Red Sea area ; it occurs from Suez to Aden, and was reported by Issel (1869) to be especially common in the Gulf of Aqaba. Tomlin (1927) gives previous records for the Suez area and records it from the Canal. Trochus (Infundibulops) erythraeus Brocchi 20.1.49, Dahab, on mud flats at low tide, i specimen. 2.ii.49, Sherm Sheik, asso- ciated with coral, 2 fms., i specimen. T. erythraeus has been collected from the Gulf of Aqaba by Issel (1869). Tomlin (1927) recorded it from the Gulf of Suez, and various other collectors, e.g. McAndrew (1870) and Vaillant (1865), have recorded it from the Red Sea area. Trochus dentatus Forskal 30.1.49, Mualla, among rocks and coral at low tide, 2 specimens. 2.ii.49, Sherm Sheik, associated with coral, i young specimen. T. dentatus is one of the common molluscs of the Red Sea and Persian Gulf. It has been recorded from the Gulf of Suez by McAndrew, Issel, and Vaillant. Tomlin (1927) reports it from the Suez Canal zone, and Issel (1869) states that it is abundant in the Gulf of Aqaba. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 193 Family TURBINIDAE Turbo radiatus Gmelin 6.11.49, Sanafir, found in coral, i specimen. 11.11.49, Abu Zabad, on reef at low tide, 2 specimens. T. radiatus is a common Indo-Pacific form, which is found in the Red Sea, the East African coast, and eastwards as far as the Philippines and New Caledonia. Tillier & Bavay (1905) and Tomlin (1927) record it from the Gulf of Suez and the Suez Canal zone. Family NERITIDAE Nerita forskalii Recluz 6.1.49, Sanafir, along shore of anchorage, 3 specimens. 12.1.49, Sherm Sheik, 7 specimens. 30.1.49, Mualla, found at low tide among rocks and coral, 2 specimens. This extremely variable mollusc has been recorded from the Gulf of Aqaba by Tomlin (1927) and Issel (1869). It is a common Indo-Pacific form, Tryon (1888) giving its distribution as the Red Sea, Indian Ocean, Natal, Singapore, China, the Philippines, and Viti Islands. Nerita undata var. quadricolor Gmelin 12.1.49, Sherm Sheik, I specimen. N. undata is a widely distributed species in the Indo-Pacific. In the variety quadricolor the aperture of the shell is white and the ribs are maculated with purplish black. This variety is confined to the western part of the Indian Ocean. Family PLANAXIDAE Planaxis breviculus Deshayes 6.1.49, Sanafir, along shore of anchorage, 3 specimens. This species has been reported from the Gulf of Suez by McAndrew (1870), who records it as a common species at low water. Smith (1891) reports it from Aden and refers to specimens in the British Museum from the Gulf of Aqaba and Persian Gulf. According to Tryon (1887) P. breviculus is a variety of P. sulcatus. Both forms have a wide distribution in the Indo-Pacific. Until more is known about the life-history of these periwinkles, we prefer to retain the name P. breviculus. Family CERITHIIDAE Cerithium tuber culaturn (L.) 6.1.49, Sanafir, shore of anchorage, 2 specimens. McAndrew found this species moderately common in the Gulf of Suez. It is an extremely variable species, and has been reported on numerous occasions from the Red Sea. Distribution. Widespread in the Indo-Pacific (Smith, 1903). 194 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Family MELANELLIDAE Melanella sp. 10.1.49, Tiran, I specimen. We do not feel justified in giving this specimen a name in view of the confusion which exists in the classification of the genus. Family STROMBIDAE Pterocera lambis (L.) 5.11.49, Sanafir, in coral, I specimen. This large shell was previously recorded from the Gulf of Aqaba by Issel (1869). Distribution. Widespread in the Indo-Pacific. Family NATICIDAE Natica mamilla L. A T . mamilla, Lamarck, 1838: 630. 6.1.49, Sanafir, along shore of anchorage under rocks, i specimen. N. mamilla has been previously recorded from the Gulf of Aqaba by Issel (1869). Tryon (1886) gives the distribution as the East Indies, the Philippines, New Cale- donia, and central Polynesia. Family CYPRAEIDAE Cypraea caurica (L.) 20.1.49, Dahab, on mud flats at low tide, i young specimen. Schilder (1938) recognizes seven races of this species, which has a widespread distribution in the Indo-Pacific. Cypraea arabica L. 30.1.49, Mualla, among rocks and coral at low tide, i specimen. 5.11-49, Sanafir, found in coral, i juvenile specimen, n. 11.49, Abu Zabad, on reef at low tide, i specimen. 11.11.49, Abu Zabad, on reef at low tide, 4 juvenile specimens. C. arabica is a well-known Indo-Pacific species, often recorded by workers on Red Sea fauna. Savigny (Egypte) gives a figure, and the species is recorded from the Gulf of Aqaba by Issel (1869). Schilder (1938) recognizes six races in the Indo-Pacific; our specimens conform to the E. African and Red Sea form which Schilder calls immanis. Cypraea Isabella L. Turia (Basilitrona) Isabella, Schilder, 1938: 176. 3.11.49, Sherm-el-Moiya, associated with coral, i specimen. 6.11.49, Sanafir, associ- ated with coral, i specimen. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 195 C. Isabella, of which four races are recognized by Schilder, has a widespread distri- bution in the Indo-Pacific. Our specimens belong to the typical form which is con- fined to the Western Indian Ocean and the Red Sea. Cypraea carneola L. Cypraea (Lyncina) carneola, Schilder, 1938: 188. n.ii.49, Abu Zabad, on reef at low tide, 3 specimens, n.ii.49, Abu Zabad, on reef at low tide, 2 juvenile specimens. This species is widely distributed in the Indian Ocean and also in the Pacific as far as Hawaii. Schilder recognizes four races of this species. The Red Sea form crassa is also found in the Gulf of Aden, Persian Gulf, and Karachi. Cypraea erosa L. Erosaria (Erosaria) erosa, Schilder, 1938: 137. 30.1.49, Mualla, among coral at low tide, i specimen. This species has been recorded from the Gulf of Aqaba by Issel (1869). C. erosa has a wide distribution in the Indian Ocean and in the Western Pacific. Our speci- men belongs to the typical form. Schilder (1938: 137) recognizes six races in the Indo-Pacific. Cypraea tigris L. Cypraea (Cypraea) tigris, Schilder, 1938: 186. 11.11.49, Abu Zabad, on reef at low tide, I immature specimen. Issel (1869) reports this species to be abundant in the Gulf of Aqaba. It has previously been reported from the Red Sea by many writers, including Ehrenberg (1831). Our specimen is not fully grown and we are unable to determine whether it belongs to the typical form. Cypraea tigris (sensu laid) is widely distributed in the Indian Ocean and in the Pacific. Family CYMATIIDAE Cymatium rubeculum (L.) Tritonium (Simpulum) rubeculum, Me Andrew, 1870: 434. Triton (Simpulum) rubecula, Tryon, 1881: 12. i.ii.49, Sherm Sheik, associated with coral, 2 specimens. McAndrew took 2 specimens at Jubal Island in the Gulf of Suez. Distribution. Red Sea to the Philippines. Distortrix anus (L.) Triton anus, Reeve II, Triton, pi. xii, fig. 63. Abu Zabad, on reef at low tide, i specimen. This species has been previously recorded from the Gulf of Aqaba by Issel (1869). 196 THE 'MANIHINE* EXPEDITION TO THE GULF OF AQABA Family MURICIDAE Drupa (Drupd) ricinus (L.) 30.1.49, Mualla, among rocks and coral at low tide, 4 specimens, n.ii.49, Abu Zabad, on reef at low tide, i specimen. Distribution. Red Sea, east coast of Africa, to Natal, Philippines, and Polynesia (Tryon, 1880: 184). Drupa (Drupd) elata (Blainville) 2.11.49, Sherm Sheik, 2 fms., associated with coral, 3 specimens. This well-known inhabitant of coral reefs has a wide distribution in the Indo- Pacific. It is recorded from Aden by Smith (1891). Family BUCCINIDAE Pisania ignea Gmelin 2.ii.49, Sherm Sheik, 2 fms., i specimen. 5.ii.49, Sanafir, found in coral, i specimen. Distribution. Red Sea, Singapore, and Philippines. Family CONIDAE Conus rattus Lamarck Conus rattus, Smith, 1891 : 399. Conus rattus, Dautzenberg, 1937. Conus rattus is a very variable species and has been recorded by many authorities including Smith (1891) and Dautzenberg (1937). Its distribution is very widespread in the Indo-Pacific. Conus textile L. ii.ii.49, Abu Zabad, on reef at low tide, i specimen. This poisonous cone shell is widely distributed in the Indo-Pacific and has been recorded from the Gulf of Aqaba by Sturany. Dautzenberg (1937) gives a very long list of localities for the species. Family NASSIDAE Nassa pulla L. 20.1.49, Dahab, collected on mud flats at low tide, 5 specimens. Issel (1869) records this shell from the Red Sea area. Tryon (1882) gives its distribution as the Red Sea, Java, and the Philippines. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 197 Class LAMELLIBRANCHIA Family ARCIDAE Area divaricata Sowerby Area divaricata, Tomlin, 1927: 304. 2.11.49, Sherm Sheik, associated with coral, 2 fms., 2 specimens. 15.11.49, Sherm Sheik, under rocks at low tide, 3 specimens. It has previously been recorded by Tomlin from the Suez Canal and by McAndrew from the Gulf of Suez, under the name A. plicata. A. divaricata has a wide distribu- tion in the Indian and Pacific Oceans. Area (Barbatia) decussata Sowerby 3i.xii.48, station Ai, shore of Faraun Island, i specimen. 3i.xii.48, station Ai, shore of Faraun Island, 2 specimens. 20.1.49, Dahab, mud flats at low tide, 4 speci- mens. 30.1.49, Mualla, among rocks and coral at low tide, i specimen. 9.11.49, Sanafir, among rocks on shore, i specimen. This species is known from the following places, according to Lamy (1917), Djibouti, Obock, Perim, and Aden. It is expected to have a much wider distribution, and we note a specimen in the British Museum collections from the Java Sea (off Batavia) . Family MYTILIDAE Brachidontes variabilis (Krauss) Mytilus variabilis Krauss, 1848: 25. Mytilus pharaonis Tillier and Bavay, 1905: 177. Mytilus exustus, Vaillant, 1865: 114. 20.1.49, Dahab, on mud flats at low tide, i specimen. This very common species was first described from Table Bay by Krauss, who drew attention to its similarity to specimens from the Red Sea. The earliest record from the latter locality is that of Savigny (Egypte, pi. xi, fig. 5). Lithophaga hanleyana Reeve 31.1.49, Mualla, associated with coral, 2 specimens. L. hanleyana has been previously recorded from the Gulf of Aqaba by Sturany (1899), who also recorded it from the Gulf of Suez and the Red Sea generally. It has also been recorded from the Gulf of Suez by Reeve and McAndrew. The Cambridge expedition to the Suez Canal (1924) also took the species in association with coral. Lithophaga moluccana Hanley 14.11.49, Dahab, associated with coral, i specimen. We have identified this species with Hanley 's species from Malacca. It appears IQ8 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA to differ from L. hanleyana (which is already known from the Red Sea) by the more tapering posterior part of the shell. Distribution. Indian Ocean. Family VULSELLIDAE Vulsella vulsella (L.) V. lingatula, Issel, 1869: 99. V. mylitina, Issel, 1869: 100. V. trita Reeve, 1858, pi. 2, fig. 17. I4.ii.49, Dahab, associated with coral, I specimen. Smith (1911), who has reviewed the genus, gives the distribution of this species as widespread in the Indian Ocean and eastwards to Japan, N. Australia, and New Caledonia. From the Red Sea it has been figured by Savigny (Egypte, pi. xiv, figs, i and 2) Ruppell records it as mytilina and Reeve as trita, both from the Red Sea. Family PECTINIDAE Chlamys luculentus (Reeve) Pecten luculenta Reeve, 1853, pi. 16, fig. 59. 2.ii.49, Sherm Sheik, 2 fms., associated with coral, i specimen. We have compared this specimen with the holotype of Reeve from NW. Australia and also with some specimens in the British Museum collection from Aden. There are no differences to be noted in our shell. The known distribution is the Red Sea and Indian Ocean. Family OSTREIDAE Ostrea cucullata Born 9.H.49, Sanafir, among shore rocks, i specimen. 0. cucullata is a very variable species and had been recorded from the Gulf of Suez by Vaillant (1865) and by Issel (1869). This oyster is edible and according to Jous- seaume, as quoted by Lamy (1925), is an excellent purgative. Distribution. Very common at many points in the Red Sea, attached to rocks, which are uncovered by the tide. This species is common throughout the Indian Ocean, and in the Pacific as far as Japanese waters (Lamy, 1925). Family CARDITIDAE Cardita variegata (Sowerby) Cardium variegatum Sowerby, 1841: 107. Cardita subaspersa Lamarck, 1819: 25. Cardita radula Reeve, 1843: 191. n.ii.49, Abu Zabad, on reef at low tide, 4 specimens. C. variegata is widespread in the Indo-Pacific, Red Sea, and Australian waters. Lamy (1916) records it from Suez, Massaouah, Djibouti, and Perim. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 199 Family TRIDACNIDAE Tridacna noae (Roding) Tridacnes noae Roding, 1798: 171. Tridacna elongata Lamarck, 1819: 106. 3i.xii.48, shore of Faraun Island, 2 specimens. This Tridacna has been recorded from the Red Sea, from Suez, and the Gulf of Aqaba by Issel (1869) under the name T. elongata Lamarck. Savigny gives the earliest figure from this area (Egypte, pi. x, fig. i). It has a wide range in the Indo-Pacific, including Zanzibar, Mauritius, Australia, Solomon Islands, Carolines, Marshall, and Loo Choo Isles (McLean, 1947). Tridacna squamosa Lamarck One specimen of this common Indo-Pacific form was collected ; the label appears to have been lost. Distribution. Indian Ocean, Indonesia, Australia, the Philippines, and Japan. Family VENERIDAE Circe scripta (L.) Venus scripta L. 20.1.49, Dahab, mud flats at low tide, I specimen. Sowerby gives the distribution of this as the Red Sea and Australia. According to Issel (1869) it is a rare species at Suez. REFERENCES ADAM, W. 1938. Un caractere sexuel secondaire chez Sepioteuthis lessoniana Lesson. Arch. neerl. Zool. 3 (Suppl.): 12-16. 1939. Cephalopoda. Part i. Le genre Sepioteuthis Blainville, 1824. SibogaExped. 550: 1-33, 1 pi. & 3 text-figs. 1941. Notes sur les cephalopodes. XVIII. Sur les especes de Cephalopodes de la Mer Rouge d6crites par C. G. Ehrenberg en 1831 et sur une nouvelle espece de Sepia (Sepia dollfusi sp. nov.). Bull. Mus. Hist. not. Belg. 17 (62): 1-14, 2 pis. 1942. Les cephalopodes de la Mer Rouge. Bull. Inst. oceanogr., Monaco, 822: 1-20. DAUTZENBERG, Ph. 1937. Resultats Scientifiques du Voyage aux Indes Orientales Nderlan- daises II. Gastropodes Marins 3. Famille Conidae. Mem. Mus. Hist. Nat. Belg. Hors Ser. 2 (18): 1-284, 3 Pis. FERUSSAC, A. DE, & ORBIGNY, A. D'. 1834-1848. Histoire naturelle generate et particuliere des Cephalopodes acetabuliferes vivants et fossiles. 2 vol., Paris. FISCHER, P. 1870. Sur la Faune conchyliologique marine des baies de Suez et de 1'Akabah. /. Conchyliol. 18: 161-179. HADDON, A. C. 1886. Report on the Polyplacophora collected by H.M.S. Challenger during the years 1873-1876. Rep. Sci. Res. Voy. H.M.S. Challenger, Zool. 15: 1-50, i pi. HOYLE, W. E. 1885. Diagnoses of new species of Cephalopoda. Ann. Mag. nat. Hist. (5) 15: 222. zoo. i. 8. D d 200 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA HOYLE, W. E. 1904. On the Cephalopoda. Rep. Pearl Fish. Manaar, Suppl. Rep. 14: 185- 200, 3 pis. 1905. The Cephalopoda. The Fauna and Geography of the Maldive and Laccadive Archi- pelagoes, 2: 975-988, i pi., 9 text-figs. 1907. Report on the Marine Biology of the Sudanese Red Sea VI. On the Cephalopoda. /. linn. Soc. Zool. 31: 35-43. ISSEL, A. 1869. Malacologia del Mar Rosso. 388 pp., 5 pis. Pisa. JOUBIN, L. 1898. Sur quelques c6phalopodes du Musee royal de Leyde et description de trois especes nouvelles. Notes Ley den Mus. 20: 21-28. LAMY, E. 1916. Sur quelques especes de Cardita figur6es par Valenciennes. Bull. Mus. Hist. nat. Paris, 21: 195-200. 1917. Les arches de la Mer Rouge (d'apres les materiaux recueillis par M. le Dr Jous- seaume). Bull. Mus. Hist. nat. Paris, 23: 26-34, 106-112. 1919. Les moules et les modioles de la Mer Rouge. Bull. Mus. Hist. nat. Paris, 25: 40-45, 109-114, 173-178. 1919. Les lithodomes de la Mer Rouge. Bull. Mus. Hist. nat. Paris, 25: 252-257, 344-350. 1925. Les huitres de la Mer Rouge (d'apres les materiaux recueillis par M. le Dr Jousseaume). Bull. Mus. Hist. nat. Paris, 31: 190-196, 252-257, 317-322. McANDREW, R. 1870. Report on the Testaceous mollusca obtained during the dredging excur- sion in the Gulf of Suez in the months February and March, 1896. Ann. Mag. nat. Hist. ( 4 ) 6: 429-450- MCLEAN, R. A. 1947. A revision of the Pelecypod family Tridacnidae. Notulae Naturae, Phila- delphia, 195: 1-7, 2 pis. PICKFORD, G. E. 1945. Le Poulpe am6ricain. A study of the littoral Octopoda of the Western Atlantic. Trans. Conn. A cad. Arts Sci. 36: 701-811; 14 pis. PILSBRY, H. A. 1890. Man. of Conchology (2) 6. 1892. Ibid. (2) 8. REEVE, L. A. 1841. Conchologia Systematica. 2 vol. London. 1843-1878. Conchologia Iconica, 20 vol. London. ROBSON, G. C. 1929. A Monograph of the Recent Cephalopoda. Pt. I. Octopodinae: 1-236, 89 text-figs. SCHILDER, F. A., & SCHILDER, M. 1938. Prodrome of a monograph on living Cypraeidae. Proc. Malac. Soc. Lond. 23: 119-231, 3 text-figs. SMITH, E. A. 1891. On a collection of marine shells from Aden. Proc. zool. Soc. Lond. 1891: 370-436, i pi. 1903. A list of species of Mollusca from South Africa forming an appendix to G. B. Sowerby's 'Marine Shells of South Africa'. Proc. Malac. Soc. Lond. 5: 354-402, i pi. 1911. On recent species of Vulsella. Proc. Malac. Soc. Lond. 9: 306-312, i pi. STURANY, R. 1899. Catalog der bisher bekannt gewordenen Sudafrikanischen Land- und Siisswasser-Mollusken mit besonderer Beriicksichtigung des von Dr. Penther gesammelten Materiales. Denkschr. Akad. Wiss. Wien 67: 537-642, 3 pis. SYKES, E. R. 1907. Reports on the Marine Biology of the Sudanese Red Sea V. On the Poly- placophora or Chitons. /. linn. Soc. Lond. 31: 31-34. THOMPSON, E. F. 1939. Chemical and physical investigations. The exchange of water between the Red Sea and the Gulf of Aden over the 'sill'. Sci. Rep. John Murray Exped. i933~34- 2: 105-119, i o text-figs. TILLIER, L., & BAVAY, A. 1905. Les Mollusques Testaces du Canal de Suez. Bull. Soc. zool. Fr. 30: 170-181. TOMLIN, J. R. LE B. 1937. Catalogue of Recent Fossil Cones. Proc. Malac. Soc. Lond. 22: 205-33- 1927. Report on the Mollusca (Amphineura, Gastropoda, Scaphopoda, Pelecypoda). Trans, zool. Soc. Lond. 22'. 291-320. TRYON, G. W. 1880. Man. of Conchology (i) 2. 1881. Ibid, (i) 3. THE 'MANIHINE 1 EXPEDITION TO THE GULF OF AQABA 201 TRYON, G. W. 1882. Man. of Conchology (i) 4. 1886. Ibid, (i) 8. 1887. Ibid, (i) 9. 1888. Ibid, (i) 12. VAILLANT, L. 1865. Recherches sur la faune malacologique de la baie de Suez. /. Conchyliol. 13: 97-121. WEINDLE, T. 1912. Vorlaufige Mitteilungen iiber die von S. M. Schiff 'Pola' im Roten Meer gefundenen Cephalopoden. Anz. Akad. Wiss. Wien, 49: 270-275. WINCKWORTH, R. 1927. New species of Chitons from Aden and South India. Proc. Malac. Soc. Land. 17: 206-208, 2 pis. WULKER, G. 1913. Cephalopoden der Aru- und Kei-Inseln. Abhandl. Senckenb. Naturf. Ges. 34: 451-487; i pi., 8 text-figs., i sketch-map. 1913. t)ber das Auftreten rudimentarer akzessorischer Nidamentaldriisen bei mannlichen Cephalopoden. Zoologica, Stuttgart 26: 201-210, I pi. 1920. tJber Cephalopoden des Roten Meeres. Senckenbergiana, Frankfurt, 2: 48-58. Legends to Plates 28-30 PLATE 28. SEPIOTEUTHIS LESSONIANA LESSON FIG. i. Ventral view of ? caught at the surface off Faraun Island, 3i.xii.48. FIG. 2. Dorsal view of c? caught off Sanafir Island, 4.11.49- The transverse streaks characteristic of the male and the pale areas overlying the iridescent patches are clearly shown in the photograph. PLATE 29 FIGS. 3 and 4. Sepioteuthis sp. ; dorsal and ventral views of a young immature specimen taken off Sherm Sheik, i.ii.49. FIG. 5. Sepioteuthis lessoniana Lesson; left tentacle club of $ shown on Plate 28, fig. i. FIG. 6. Sepioteuthis lessoniana Lesson ; right tentacle club of $ taken at Aqaba, 28.1.49. FIG. 7. Octopus horridus Orbigny taken at Tiran Island, 10.1.49. PLATE 30. OCTOPUS CYANEA GRAY FIG. 8. Lateral view of a $ taken at Sherm Sheik, 12.1.49. FIG. 9. Oral face of the same specimen as in Fig. 8. The so-called ' zebra ' markings on the lateral side of the arms are a constant feature in this species. Bull. B.M. (N.H.) Zoology, I. 8 PLATE 28 SEPIOTEUTHIS LESSONIANA LESSON Bull. B.M. (N.H.) Zoology, I, S PLATE 29 AQABA CEPHALOPODA Bull. B.M. (N.H.) Zoology, I, 8 PLATE 30 BB .4 A OCTOPUS CYANEA GRAY VII. ECHINODERMATA By AILSA M. CLARK THE collection of Echinoderms includes many well-known littoral species which are widespread throughout the Indo-West Pacific area, as well as some which are peculiar to the Red Sea. A few species, notably the single Crinoid Capillaster multiradiata (Linnaeus) and an Echinoid, Clypeaster fervens Koehler, have not been previously recorded from the Red Sea. The species are the following, all of them from low tide or low spring tide level except where otherwise stated. Those mentioned in more detail in the text are marked with an asterisk. References in the text giving further details are marked with a dagger. ASTEROIDEA Astropeclen polyacanthus Miiller & Troschel *Fromia ghardaqana Mortensen *Gomophia egyptiaca Gray Linckia multifora (Lamarck) . Aster ope carinifera (Lamarck) *Asterina burionii Gray . OPHIUROIDEA *Ophiocoma pica Miiller & Troschel *0phiocoma scolopendrina (Lamarck) *Ophiocoma erinaceus Miiller & Troschel Ophiocoma valenciae Miiller & Troschel *0phiocoma sp. ..... * M ' acrophiothrix hirsuta (Miiller & Troschel) Locality Number Dahab I Ras Muhammad Bay i Dahab i Abu Zabad 3 Abu Zabad I Sherm Sheik 2 Dahab I Sanafir I. i Abu Zabad 2 Sanafir I. i Dahab i Sherm Sheik I Abu Zabad 4 Dahab 5 Sherm Sheik 4 Sanafir I. 5 Mualla i Tiran 3 Abu Zabad 5 Sanafir I. 3 Dahab 8 Faraun Id. 10 Sherm Sheik 4 Abu Zabad 2 Dahab 2 Sherm Sheik I Sanafir I. 2 Abu Zabad 2 Tiran 4 Sanafir I. Sherm Sheik Abu Zabad Dahab Sherm Sheik Sherm Sheik Sanafir I. Dahab 204 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Locality Number Ophiolrichoides propinqua (Lyman) * Placophiothrix purpurea (von Martens) Ophiolepis cincta Miiller & Troschel Dahab Dahab Dahab Abu Zabad Sherm Sheik CRINOIDEA *Capillaster multiradiata (Linnaeus) . Dahab ECHINOIDEA Eucidaris metularia (Lamarck) Diadema setosum (Leske) Echinometra mathaei (Blainville) Heterocentrotus mammillatus (Linnaeus) *Tripneustes gratilla (Linnaeus) Ctypeaster humilis (Leske) *Clypeaster fervens Koehler Lovenia elongata (Gray) Sherm-el-Moiya Mualla Sherm Sheik Sana fir I. Tiran Aqaba Sherm-el-Moiya Tiran Faraun I. Abu Zabad Mualla Abu Zabad Sherm Sheik Tiran Sanafir I. Dahab Dahab Abu Zabad Sanafir I. Dahab Aqaba Dahab Dahab Dahab 4 4 5 4 10 3 3 i 6 2 I I I HOLOTHUROIDEA Synapta maculata (Chamisso & Eysenhardt) Synaptula recta (Semper) Halodeima edulis (Lesson) Halodeima atra (Jager) .... Halodeima cinerascens (Brandt) Holothuria impatiens (Forskal) * Holothuria sucosa Erwe. Holothuria pardalis (Selenka) . Holothuria curiosa var. pervicax (Selenka) Microthele difficilis (Semper) . *Microthele nobilis (Selenka) . Actinopyga miliaris (Quoy & Gaimard) . Um Nageila (in shallow i water off mangrove swamp) Sherm Sheik i (pt.) Dahab i Abu Zabad i Abu Zabad 2 Abu Zabad I Dahab 5 Dahab i Dahab 5 Graa 2 Dahab I Abu Zabad 8 Dahab 6 Ras Muhammad Bay i Faraun I. i THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 205 ASTEROIDEA Family LINCKIIDAE Fromia ghardaqana Mortensen PL. 31, FIGS, a-c Scy taster milleporellus, Miiller and Troschel, 1842: 35; [non Asterias milleporella Lamarck, 1816: 564]- Fromia milleporella (part), Gray, 1866: 14. Fromia monilis, Tortonese, 1935: 70; 1936: 213; [non Fromia monilis Perrier, 1875: 443 (p. 179 in repaged edition)]. Fromia ghardaqana Mortensen, 1938: 37. Dahab, shore ; I specimen. Abu Zabad, reef at low water springs ; 3 specimens. Description. R = 40 mm., r = 10 mm., R/r = 4-0. The arms taper evenly throughout their length to a rather pointed tip. One has been broken and is in pro- cess of regeneration. Of the five primary inter-radial plates, three are enlarged with a flat surface raised slightly above the level of the surrounding plates, the one adjacent to the madreporite is smaller but also a little elevated, while the fifth is not at all conspicuous. The madreporite is triangular in shape, with deep radiating grooves, and measures 1-4 mm. across. The carinal row of plates is not very clear proximally, where all the plates are in fact rather irregularly arranged. At the base of the arm there are about seven plates across the width. All the dorsal and ventral plates, as well as the marginals, are closely covered with uniform, smooth, rounded granules, about 7 in the length of i millimetre. These lie very close together and are polygonal on the convex plates, of which there are about 10 on the dorsal side of each arm, besides the primary inter-radial plates and the marginals. Most of the convex plates are near the tip of the arm, but an irregular series of spaced plates occupies the mid-radial distal area. The number of supero-marginal plates varies between 19 and 21 on the four com- plete arms, with the same number in each infero-marginal series. The latter plates are relatively narrower and are noticeably longer than broad. In the distal half of the arm they may bear a small tubercle in the centre as also do the last two supero- marginals. These plates, unlike the infero-marginals, are not evenly sized but, especially distally, large and small plates tend to alternate with one another, the larger ones being rather convex. The two series of marginal plates tend to alternate in position. On the ventral side the papulae are clearly visible in the angles between the plates. The granules surrounding each one are not markedly larger than the other granules. Proximally there are 3 rows of papulae, correlated with the presence of 4 rows of ventro-lateral plates. The outermost row of these consists of only 4 plates on each side of the interbrachial angle, extending to the aboral end of the second infero- marginal plate. The third series reaches the seventh infero-marginal, the second to the eleventh plate, and the innermost series to the fourteenth. The adambulacral plates bear 2, or, in the middle part of the arm, often 3 flattened furrow spines. Outside these is a single stumpy spine, shorter than the furrow spines though much thicker and slightly elongated in transverse section. On both sides of this spine and outside it are numerous granules like those of the ventro-lateral plates. 206 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Remarks. Miiller and Troschel's description of Scytaster milleporellus together with the locality of the Red Sea suggests that their specimens like many of Gray's were almost certainly Fromia ghardaqana. However, some of the latter, from Mauritius and other localities in the Indian Ocean, are the form (pi. 31, fig. d) with even-sized supero-marginal plates which is generally assumed to represent Fromia milleporella. Since Lamarck gave as the type locality 'les mers d'Europe?', and only a brief description, it is not certain which form really is milleporella. This question can only be answered by study of the type specimens if they are still in existence. Mortensen has examined the type of Fromia monilis Perrier and finds it quite distinct from the Red Sea species, although he does not give any details. On comparison of the specimen described with one of F. monilis from Macclesfield Bank, South China Sea, with R = 35 mm., it is at once seen that the granulation of the dorsal side of the latter is very much finer with at least 10 granules to a millimetre rather than only 7. Also the arms of F. monilis are relatively much narrower with R/r about 4-5 on average and the supero-marginal plates usually occupy more of the dorsal surface of the arm, so that only 3 or 5 rows of plates, rarely more, lie across the base of the arm. On the ventral side the granules around the pores are clearly enlarged unlike those of F. ghardaqana. Unfortunately there are no specimens of Fromia pacifica H. L. Clark (the species that Mortensen says is most nearly related to F. ghardaqana) in the British Museum to compare with the material from the Red Sea. That Torres Strait species apparently has even-sized supero-marginal plates and pointed granules rather than flat ones. There are also three juvenile specimens in the present collection, the two larger ones having R = 18 mm., but whereas one is much more slender with an R/r ratio of 3-5 : i, the other has the ratio only 2-8. Of the many old dry specimens in the British Museum, the R/r value varies between 3-0 and 3-7, although a co-type of F. gharda- qana from Ghardaqa sent by Dr. Mortensen has the ratio 4-0. This it seems is just about the maximum value. From all the other species of Fromia, F. ghardaqana is easily distinguished by the alternate large and small distal supero-marginals. Gomophia egyptiaca Gray PL. 32 Gomophia egyptiaca Gray, 1840: 286. H. L. Clark, 1921: 55. ^Scytaster aegyptiacus, Perrier, 1875: 428 (p. 164 in re-paged edition). Nardoa aegyptiaca, de Loriol, 1891: 30. Fisher, 1906: 1087. Koehler, 1910: 157, pi. xvii. 5, 6. Abu Zabad, reef at low water springs ; one specimen. R = 84 mm., while the type has R = 62 mm. The intermarginal plates in the arm angle are not more conspicuous than in the type and indeed are quite hidden by the granulation in one of the angles. Range. Red Sea, Mauritius, Samoa, Philippines, Fiji, Macclesfield Bank. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 207 Family ASTERINIDAE Asterina burtonii Gray Asterina burtonii Gray, 1840: 289. fG. A - Smith, 1927: 641. Asteriscus wega Perrier, 1869: 102. Asterina wega Perrier, 1876: 238 (p. 318 of re-paged edition). Sanafir ; one specimen. Dahab ; one 6-armed specimen. Abu Zabad ; 4 specimens. Sherm Sheik ; one y-armed specimen. Remarks. Since in 1876 Perrier corrected the error in his original description of A. wega, regarding the number of spines on each ventro-lateral plate, emending it to 2 or 3 rather than i, there seems to be no reason why specimens with up to 8 arms should not be regarded as Asterina burtonii. Smith accepts 6-armed specimens as such. These forms with more than 5 arms are usually juvenile and more or less obviously in process of regeneration. The y-armed specimen in the present collection has 4 arms diminutive. Perrier states that all his thirteen specimens of A. wega were undergoing regeneration. 2. OPHIUROIDEA Family OPHIOCOMIDAE Ophiocoma pica Miiller & Troschel Ophiocoma pica Miiller & Troschel, 1842: 101. H. L. Clark, 1921: 127, pi. xiii, 8 (coloured). |Ely 1942: 54, pi. xii, B.i., text-fig. 15. Ophiocoma lineolata Miiller & Troschel, 1842: 102. de Loriol, 1893: 28. Dahab ; 5 specimens. Sherm Sheik ; 4 specimens. Sanafir ; 5 specimens. Mualla ; i specimen. Tiran ; 3 specimens. Abu Zabad ; 5 specimens. All from coral at low tide. Remarks. These specimens are easily distinguished from the other Ophiocomas collected by the conspicuous stripes on the otherwise black arms and the yellowish stripes on the disk. The ratio of arm length to the disk diameter varies between 3-6 and 4-8: i. Note. It has been accepted for a very long time that 0. pica and 0. lineolata are synonymous, but both names have been retained by different authors. For instance Koehler (19220: 324) still uses lineolata although most other recent authors prefer pica. However, the latter name had page priority in Miiller & Troschel's System der Asteriden. So in spite of its previous use in manuscript by Valenciennes, which has no validity, the name Ophiocoma lineolata should be dropped. Ophiocoma scolopendrina (Lamarck) Ophiura scolopendrina Lamarck, 1816, 2: 544. \Ophiocoma scolopendrina, de Loriol, 1893: 23. H. L. Clark, 1921: 125, pi. xiii. 9. fKoehler, 1922^: 325, pis. Ixxiii. 5; Ixxiv. 1-7. Sanafir ; 3 specimens. Dahab ; 8 specimens. Faraun Island ; 10 specimens. Sherm Sheik ; 4 specimens. Abu Zabad ; 2 specimens. All from the shore under stones, zoo. i. 8. EC 208 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Remarks. The colour ranges from variegated bluish grey to dense black on the dorsal side of the disk and arms, the ventral side of the disk being always pale. Most have the arms broken but they are usually relatively long, six or more times the disk diameter. Ophiocoma erinaceus Miiller & Troschel Ophiocoma erinaceus Miiller & Troschel, 1842: 98. fde Loriol, 1893: 21. H. L. Clark, 1921 : 127. |Ely, 1942: 52, text-fig. 45, pi. xiia. Dahab: 2 specimens. Sherm Sheik; i specimen. Sanafir; 2 specimens. Abu Zabad: 2 specimens. Remarks. Except for the two specimens from Abu Zabad, these are densely black all over ; even the tentacles of those from Dahab are black ; also the arms are relatively short, the ratio of arm length to disk diameter being 4-4-8 : i. The Abu Zabad speci- mens are also densely black dorsally but are pale on the underside of the disk, although the tentacles are black. The arms of one are all broken but in the other their length is nearly seven times the disk diameter. They are thus intermediate between 0. scolopendrina (with relatively long arms and lighter colour) and 0. erinaceus, with shorter arms and a uniformly dark colour, so there was some doubt as to which species they should be. Finally they were referred to the latter species for the following reasons : besides the very dense black colour on the dorsal side, the disk granulation hardly extends below the periphery and there are two tentacle scales for quite a large part of the arm, as in erinaceus. Also, apart from these morphologi- cal characters, the fact that they were taken well out on the reef at low spring tide level in the same zone as Ophiocoma pica suggests that they belong to erinaceus, for H. L. Clark makes the distinction of habitat of the two forms scolopendrina and erinaceus an important reason for maintaining them as separate species, the former characteristically occupying a higher level on the shore which is uncovered at ordinary low tides. I fully agree with Ely that very rarely can several characters be used to distinguish intermediate specimens as belonging to one or the other species. Quite often con- flicting results are obtained by using two different characters. For instance there is a specimen in the British Museum collection from Muscat, with the proportions 170 mm./2i mm. = 6-4:1, which would on this count be called scolopendrina, but the unrelievedly black colour on the contrary suggests that it is erinaceus. In such cases only a detailed observation of the habit and habitat can produce a conclusive identification. Ophiocoma sp. Sherm Sheik ; i specimen. This is a very small specimen (disk diameter = 5 mm.) with all the arms broken and a hole through the centre of the disk. It is nearest to O.pica as there are 2 tentacle scales, 5 slender arm spines proximally, and dark bands on the arms, also the oral shields are longer than wide. However, the dorsal side of the disk is unusual in having black spots each surrounded by a lighter ring on a dark brown background. These spots vary in size and shape but are relatively much larger than those of Ophiocoma doderleini. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 209 Family OPHIOTRICHIDAE Placophiothrix purpurea (von Martens) Ophiothrix purpurea von Martens, 1867: 346. Doderlein, 1896: 296, pis. xiv. 12; xvii. 23. ^Ophiothrix lepidus de Loriol, 1893: 45, pi. xxv. i. ^Ophiothrix fallax de Loriol, 1893: 47, pi. xxv. 2. Placophiothrix purpurea, H. L. Clark, 1939: 86. Dahab ; i specimen. This specimen agrees very closely with de Loriol's description of Ophiothrix fallax from Mauritius, as it has a pale green disk and relatively long arms (disk diameter = 4-5 mm., arm length = 45 mm.). H. L. Clark has declared 0. kpida de Loriol to be a synonym of 0. purpurea, from a study of the long series of specimens obtained by the John Murray Expedition. He makes no mention of 0. fallax, but as the characters of that species are intermediate between those of the other two, it cer- tainly comes within the range of variation of Placophiothrix purpurea. Possibly Doderlein's Ophiothrix lorioli (1896: 297) from Amboina, with radial shields similar to those of 0. lepida, is also a synonym of purpurea. Both Doderlein and Koehler (1898: 102) say that 0. lepida and 0. lorioli cannot be confounded, but neither of them give any reason for this. Macr ophiothrix hirsuta (Miiller & Troschel) Ophiothrix hirsuta Miiller & Troschel, 1 842 : 1 1 1 . Marktanner-Turneretscher, 1887:311. f Koehler, 19220: 234, pis. xxxi. i, 2; xxxiii. 13; xcix. 2. Tortonese, 1949: 37. Ophiothrix cheneyi Lyman, 1861: 84. Macr ophiothrix hirsuta, H. L. Clark, 1938: 285. Ophiothrix demessa, H. L. Clark, 1939: 83. [non Ophiothrix demessa Lyman, 1861: 82.] Sherm Sheik ; i specimen. Sanafir; i specimen. Dahab ; 2 specimens. Remarks. There seems to be considerable difference of opinion as to the shape of the dorsal arm-plates in this species. H. L. Clark describes them as more or less oval in his key to the species of Macr ophiothrix, but as Tortonese points out, Miiller & Troschel's original description mentions lateral angles, a statement open to several interpretations but suggesting at least something a little more angular than an ellipse. Koehler's plate 83, fig. 13, of the arm of a Philippine specimen shows dorsal arm-plates of which the widest part is midway between proximal and distal edges, whereas all the Red Sea specimens that I have seen have the widest part distinctly distal to the half-way line with a slightly rounded angle as opposed to the very acute angle of M. longipeda. This rather fan-shaped form is shown in Koehler's plate 31, fig. i, of a specimen from the Red Sea, which also resembles the present material in the characters of the disk. That the shape of the dorsal arm-plates varies in different parts of the range is shown by the fact that Lyman 's species from Zanzibar, 0. cheneyi, which is commonly accepted as a synonym of M. hirsuta, is described as having oval, microscopically granulated dorsal arm-plates. 2io THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA The latter feature, that is the presence of more or less thorny granules on the dorsal arm-plates, is not mentioned by Miiller & Troschel, but Marktanner-Turner- etscher states that they are always somewhat granulated although this is not so marked as in 0. demessa. In fact he considers the difference in the size and thorni- ness of these granules to be the only difference separating the two species. Through the courtesy of Dr. Elisabeth Deichmann I have had the opportunity of studying some specimens of 0. demessa and as a result fully agree with Marktanner-Turner- etscher, the only other difference that I can see being that the arms seem to taper more rapidly, in younger specimens at least, of 0. demessa. The granules on the arms are distinctly more thorny than in the specimens from the Red Sea, where they may be quite unobtrusive in spirit specimens. H. L. Clark in his John Murray Report names two specimens from the Red Sea and the Gulf of Aden Ophiothrix demessa, of which the one in the British Museum is indistinguishable from M. hirsuta, and I suspect that Koehler's record of 0. demessa from the Red Sea is also based on a similar specimen. In 1946 H. L. Clark erected a new genus Amphiophiothrix to accommodate the species 0. demessa, but I cannot agree that there is a generic dis- tinction between it and Macrophiothrix hirsuta. The validity of some of the other Indo-Pacific species of Macrophiothrix has been questioned by several authorities. Some of them are possibly variants of other species such as hirsuta in which the granulation of the radial shields is reduced, for there is a tendency for such a reduction throughout the genus as there is also for the development of granules on the dorsal arm-plates, a character featuring in the descriptions of several species, such as M. rugosa H. L. Clark, and noticeable also in some larger specimens of other species. However, without seeing the types and being able to compare them with large series of specimens from different parts of the Indo- Pacific, it is impossible to add anything concrete to the suspicions already voiced. 3. CRINOIDEA Family COMASTERIDAE Capillaster multiradiata (Linnaeus) Asterias multiradiata, Linnaeus, 1758: 663. Capillaster multiradiata, A. H. Clark, 1909: 364. ^Capillaster multiradiata, A. H. Clark, 1931: 173, pis. iii. 5; xi. 30; xiii. 34; xiv. 35, 36; Ixxxi. 222, 223, also many text-figs. Dahab ; i specimen ; arms 90 mm. in length. This is the first record of this species from the Red Sea, the former known range being from Formosa south to northern Australia and west as far as the Maldive Islands, so its discovery here is most interesting. There are 36 arms, which is rather more than usual ; A. H. Clark gives 12 to 35 as the usual range, but quotes specimens with up to 43 arms. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 211 4. ECHINOIDEA Family TOXOPNEUSTIDAE Tripneustes gratilla (Linnaeus) Echinus gratilla Linnaeus, 1758: 664. Tripneustes gratilla, Loven, 1887: 77. -fMortensen, 1943, 3 (2) : 500, pis. xxxiii. 1-3 ; xxxiv. 2-6; xxxv. 3-4; xxxvii. 1-2, 4-10; xxxviii. 1-4; Ivi. n. Abu Zabad, reef at low spring tide ; 3 specimens. Sanafir ; I specimen. Dahab ; 6 specimens. Aqaba ; 2 specimens. The two from Aqaba are superficially very different from the others, having relatively few and long primary spines above the ambitus, which are white in colour and contrast sharply with the dark purple of the test, produced mainly by the numer- ous pedicellariae. The tube feet of these two specimens are black or at least have a black band around them. The other specimens are more drab in colour, several being slightly reddish and their tube feet are grey. The denuded tests are distinctly green aborally. Family CLYPEASTRIDAE Clypeaster (Rhaphidoclypus) fervens Koehler Clypeaster fervens Koehler, 1922: 45, pis. vi. i, 2, 6; xv. I. ^Clypeaster (Rhaphidoclypus) fervens, Mortensen, 1948, 4 (2): 84, pis. xiii. 2, 3; xxii. i-n; xxvi. 2; Ixv. 7-9, 12, 20. Dahab, shore ; i dead test. This specimen is easily distinguished from Clypeaster humilis by the relatively large petals and the concave oral side. It is 46 mm. in length but already has well- developed genital pores. According to Dr. Mortensen (who has very kindly confirmed my identification) in his monograph, the genital pores only begin to appear when the length is about 56 mm., that is in the John Murray Expedition material from the Indian Ocean. It seems then that in the Red Sea this species undergoes precocious genital development. 5. HOLOTHUROIDEA Family HOLOTHURIIDAE Holothuria sucosa Erwe Cucumaria hartmeyeri Heifer, 1912: 332. [non Holothuria hartmeyeri Erwe, 1913 : 383, pi. vii. 19.] ^Holothuria sucosa Erwe, 1919: 186, text-fig. 5. Panning, 1934, 3: 80, text-fig. 64. ? Holothuria ocellata, Tortonese, 1936: 235, text-figs. 5, 6. Dahab ; i specimen. The knobbed buttons have 4 or 5 pairs of holes, sometimes as many as 10 pairs. Unlike H. arenicola var. boutani Herouard, which also has multilocular, though flat buttons, the tables, which are also larger, have a complete ring of holes around the margin not interrupted by the extended four central holes. Unlike H. ocellata Jager, the great majority of buttons have more than 3 pairs of holes. 212 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA Microthele nobilis (Selenka) Mulleria nobilis Selenka, 1867: 31, pi. xvii. 13-15. \Holothuria (Microthele) nobilis, Panning, 1929, 1: 131, text-fig. 15. Microthele nobilis, Heding, 1940: 320. Ras Muhammad ; i specimen. Although shrunken in preservation this specimen still measures 24 cm. in length. The tables have mostly rather irregular disks. The other dorsal deposits are ' three- dimensional buttons', fenestrated irregularly with about 4 pairs of holes on each face. Ventrally, however, these spicules are much outnumbered by more conventional flat buttons with holes in one plane, there being usually 4 or 5 pairs of holes if not more. REFERENCES CLARK, A. H. 1909. The Crinoids of the 'Gazelle' Expedition. Zoo/. Anz. 34 (11-12): 363-376. 1931. A monograph of the Existing Crinoids. i (3). Bull. U.S. Nat. Mus. 82: vii, 1-816, 82 pis. CLARK, H. L. 1921. The Echinoderm fauna of Torres Strait. Pap. Tortugas Lab. 10: vi, 1-223, 38 pis. 1938. Echinoderms from Australia. Mem. Mus. comp. Zoo/. Harv. 55: viii, 1-596, text-figs. 1-64, pis. 1-24. 1939. Ophiuroidea. Sci. Rep. John Murray Exped. 6: 29-136, text-figs. 1-62. 1946. The Echinoderm fauna of Australia. Pub. Carneg. Instn. 566: iv, 1-567. DODERLEIN, L. 1896. Bericht iiber die von Herrn Prof. Semon bei Amboina und Thursday Island gesammelten Ophiuroidea. In SEMON, Zoologische Forschungsreisen in Australien und dent Malayischen Archipel. pp. 279-300, pis. 14-17. Jena. ELY, C. A. 1942. Shallow water Asteroidea and Ophiuroidea of Hawaii. Bull. Bishop Mus. Honolulu, 176: 1-63, text-figs. 1-18, pis. 1-13. ERWE, W. 1913. Holothuroidea. In MICHAELSEN, W., & HARTMEYER, R., Die Fauna Siidwest- Australiens. Ergebn. Hamburg, sudwest-austr. Forsch. 4: 349-402, text-fig, i, pis. 5-8. 1919. Holothurien aus dem Roten Meer. Mitt. Zoo/. Mus. Berlin, 9 (2) : 177-189. FISHER, W. K. 1906. The Starfishes of the Hawaiian Islands. Bull. U.S. Fish. Comm. 1903. 3: 987-1130, pis. 1-49. GRAY, J. E. 1840. A Synopsis of the genera and species of the class Hypostoma. (Asterias Linn.) Ann. Mag. Nat. Hist. 6: 175-184, 275-290. 1866. Synopsis of the Species of Starfish in the British Museum, iv+17 pp., pis. 1-16. London. HEDING, S. G. 1940. Die Holothurien der Deutschen Tiefsee Exped. II. Aspidochirote und Elasipode Formen. Wiss. Ergebn. Valdivia, 24: 317-375, text-figs. 1-21. HELPER, H. 1912. Uber einige von Dr. Hartmeyer im Golf von Suez gesammelte Holothurien. Mitt. Zoo/. Mus. Berlin, 6 (2) : 327-334. KOEHLER, R. 1898. fichinodermes recueillis par 1" Investigator ' dans 1'Ocean indien. II. Les ophiures littorales. Bull. Sci. France Belgique, 31: 55-124, pis. 2-4. 1910. Echinoderma of the Indian Museum. 6. Asteroidea II. pp. 1191, pis. 1-20. Calcutta. 1922. Echinoderma of the Indian Museum. 8. Echinoidea. II. Clypeastrides et Cassidulides. pp. 1-161, pis. 1-15. Calcutta. 19220. Ophiurans of the Philippine Seas. Bull. U.S. Nat. Mus. 100 (3). x, 1-486, pis. 1-103. LAMARCK, J. B. P. A. DE MONET DE. 1816. Histoire naturelle des animaux sans vertebres. II. pp. 1-568. Paris. LINNAEUS, C. 1758. Sy sterna Naturae. Ed. X. 1. 824 pp. Holmiae. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 213 LORIOL, P. DE. 1891. Notes pour servir a l'6tude des fichinodermes. III. Mem. Soc. Phys. Geneve. Suppl. 1890 (8): 1-31, pis. 1-3. - 1893. Catalogue raisonne des Echinodermes recueillis. par M. V. de Robillard a 1'Ile Maurice. III. Ophiurides et Astrophytides. Mem. Soc Phys. Geneve, 32 (i): 1-59, pis. 23-25- LOVEN, S. 1887. On the species of Echinoidea described by Linnaeus in his work 'Museum Ludovicae Ulricae'. Bih. Svensk Vetensk. Akad. Handl. (13) 4 (5): 1-185, pis- I -9- LYMAN, T. 1861. Descriptions of new Ophiuridae. Proc. Boston Soc. Nat. Hist. 8: 75-86. MARKTANNER-TURNERETSCHER, G. 1887. Beschreibung neuer Ophiuriden und Bemerkungen zu bekannten. Ann. naturh. (Mus.) Hofmus. Wien, 2: 291-316, pis. 12-13. MARTENS, E. VON. 1867. Uber vier neue Schlangensterne. Mber. preuss. A had. wiss. : 345-348. MORTENSEN, TH. 1926. Cambridge Expedition to the Suez Canal in 1924. VI. Echinoderms. Trans. Zool. Soc. Lond. 22: 117-131, text-figs. 11-13. - 1928-1948. Monograph of the Echinoidea. Copenhagen. 1. Cidaroidea, 1928; 3 (i) Aulo- donta, 1940; 3 (2) and (3) Camarodonta, 1943; 4 (2) Clypeastroida, 1948. - - 1938. Contributions to the study of the Development and Larval forms of Echinoderms. IV. K. Danske Vidensk. Selsk. Skr. (9) 7 (3) : 1-59, text-figs. 1-30, pis. 1-12. MULLER, J., & TROSCHEL, F. H. 1842. System der Asteriden. xx + i34 pp., pis. 1-12. Braun- schweig. PANNING, A. 1929-1935. Die Gattung Holothuria. Parts I-V. Mitt. Zool. St.Inst. Hamburg, 44: 91-138, text-figs. 1-21 ; 45: 24-50, 65-84, 85-107, text-figs. 22-102 ; 46: 1-18, text-figs. 103-121. PERRIER, E. 1869. Recherches sur les p^dicellaires et les ambulacres des Astries et des Oursins. I. Ann. Sci. nat. (5) 12: 197-304, pis. 17, 18. - 1875. Revision de la collection de Stellerides du Museum d'Histoire Naturelle de Paris. 384 pp. Paris. (Also published in Arch. Zool. exp. gen. 4 (1875) : 263-449; 5 (1876) : 1-104, 209-304). SELENKA, E. 1867. Beitrage zur Anatomic und Systematik der Holothurien. Z. wiss. Zool. 17: 291-374, pis. 17-20. SMITH, G. A. 1927. On Asterina burtonii Gray. Ann. Mag. Nat. Hist. (9) 19: 641-645. TORTONESE, E. 1935. Gli Echinodermi del Museo di Torino. III. Asteroidi. Boll. Mus. Zool. Anat. comp. Torino, 45 (3): 27-132, pis. i-n. - 1936. Echinodermi del Mar Rosso. Ann. Mus. Stor. nat. Genova, 59: 202-245, text-figs. 1-8. - 1949. Echinodermi della Somalia Italiana. Ann. Mus. Stor. nat. Genova, 64: 30-42, i pi. Legends to Plates 31 and 32 PLATE 31 Fromia ghardaqana Mortensen, specimen from Dahab. (a) Dorsal side ; (b) ventral side; (c) specimen 40.3.23.35; and (d) Fromia milleporella Lamarck, specimen 39.3.29.20, for comparison. PLATE 32 Gomophia egyptiaca Gray, (a) Dorsal side of the type and (b) an inter- brachial angle to show the intermarginal plates. Bull. B.M. (N.H.) Zoology, I, 8 PLATE 31 FROMIA GHARDAQANA MORTENSEN (Figs, a -c) FROMIA MILLEPORELLA LAMARCK (Fig. d) Bull. B.M. (N.H.) Zoology, I, 8 PLATE 32 GOMOPHIA EGYPTIACA GRAY VIII. TUNICATA By WILLARD G. VAN NAME AMERICAN MUSEUM OF NATURAL HISTORY THROUGH the kindness of the authorities of the British Museum (Natural History) the Tunicata collected by the M. Y. Manihine in the Gulf of Aqaba in the early months of 1949 were forwarded to me for examination. As far as I am aware no collection of Tunicata has previously been made there, but the tunicates of the Red Sea, and especially those of the Gulf of Suez, have been the subject of much study and are dealt with in several published articles. The remarkable work by Savigny (1816), which was many decades in advance of his time, and which laid the foundations of much of our knowledge of the Tunicata, as well as important articles by Hartmeyer, Michaelsen, and others during the present century, were based in large part on specimens from those waters. It was therefore hardly to be expected that new species would be found in a com- paratively small collection, especially since no specimens were obtained except in very shallow water, in no case over about 2 fathoms. All the specimens appear to be referable to species already described, but never- theless the collection contains some that are of interest, especially those of the solitary form of Salpa maxima var. tuber culata described by Metcalf, 1918, from the southern Philippines, who, however, had specimens of the aggregated form only. Since the Gulfs of Suez and Aqaba are extensions of the Red Sea and consequently of the tropical part of the Indian Ocean, their faunas are Indo-Malayan, in spite of their near approach geographically to the eastern Mediterranean. This fact is, however, not so evident in the present collection as might be expected, since it happens to contain some species that are practically circumtropical, and found both in the Mediterranean and Indian Ocean. These species are shallow water forms, and it is possible that some of them may owe their very extensive distribution to human agency, by transportation on the bottom of ships. The Tunicata in this collection appear to belong to the following 13 species, one of them (Salpa maxima] being perhaps represented by two varieties: Class ASCIDIACEA COMPOUND ASCIDIANS 1. Polyclinum saturnium Savigny, 1816 Polyclinum saturnium Savigny, 1816: 190, pi. 19, fig. I ; Michaelsen, 1920: 9. One rather thick colony measuring over 50 mm. in extent. 2. Didemnum candidum Savigny, 1816 Didemnum candidum Savigny, 1816: 194, pi. 4, fig. 3, pi. 20, fig. i. Several small colonies with abundant spicules, whose points are so short and zoo. i. 8. Ff 216 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA slightly developed that the spicules are almost spherical. Also one small colony having spicules with larger and better developed rays or points. There is also one colony, growing on coral, which has very few spicules and a great many faecal pellets in the intestinal tracts of the zooids, perhaps indicating an incipient case of the so-called 'Hypurgon' condition to which this and allied forms are subject, in which the water currents in the cloacal canals become too weak to carry off the waste material, which remains in the cloacal system and in the common test, greatly altering the character and appearance of the colony, but there does not seem to be any reason for assuming that it is of a different species. See Michaelsen, 19190: 11-13. Didemnum candidum appears to be a species of very wide distribution, being found also in American waters, very abundantly in some places. It cannot be doubted that far too many species of the genus Didemnum have been described. Apparently this is in part due to overlooking the great effects on the general appearance of the colony of its age and past history, particularly in the case of old colonies. Many or most of the species are subject to periods (in many cases seasonal) of regression and extensive degeneration of the zooids, followed by sub- sequent recovery and regrowth of the colony to its normal functional condition. During such regressive periods, though the zooids degenerate more or less com- pletely, the spicules may endure unchanged through several or perhaps many genera- tions of the zooids. The result is that in old colonies we may find a far greater abundance of spicules than the spicule-forming ability of the zooids present could possibly account for, and likewise often peculiarities in the distribution of the spicules, which one must not mistake for specific characters. Old colonies are apt to acquire a hard calcareous character in which the spicules form a far larger component than the test substance and zooids do. SIMPLE ASCIDIANS 3. Phallusia nigra Savigny, 1816 Phallusia nigra Savigny, 1816: 163, pi. 2, fig. 2; pi. 9, fig. i. Ascidia atra Lesueur, 1823: 2, pi. i, fig. 2. Ascidia nigra, Herdman, 1882: 210. Phallusia nigra, Hartmeyer, 1916: 408, figs. 5-9. Eleven specimens, all of small size. This species, widely distributed and common in shallow water in many warm regions of both hemispheres, is easily recognizable from its bluish or blue-black coloration. If Phallusia is accepted as a genus distinct from Ascidia, the present species should be placed in it, as in old and large individuals the neural duct has accessory apertures, at least in many specimens. In other respects it is a very typical Ascidia. 4. Phallusia sp., apparently Phallusia arabica Savigny, 1816 Phallusia arabica, Hartmeyer, 1916: 414, figs. 10-12. One specimen of 52 mm. body length (or 63 mm. if the obliquely forwardly ex- tending atrial siphon is included). In external features other than unusual forward THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 217 position of the atrial siphon (probably only an individual peculiarity), as well as in a majority of the internal characters, it agrees well with the descriptions of Savigny and Hartmeyer cited above. But this specimen is abnormal and defective in the slight development of the dorsal tubercle, which is practically wanting, although its aperture, which is U- shaped, with the open interval obliquely forward and to the left and with one of the ends bent down, is clearly visible, but very small. Yet I was not able to find any neural duct extending from its aperture, nor any neural gland. Even the ganglion was only doubtfully demonstrated. The neural duct should be long in this species, with accessory lateral openings as well as the terminal one in the dorsal tubercle. The tissues of this specimen were dark coloured and somewhat opaque, but that would not account for the difficulty of finding the above structures if they were present in a normal state of development. 5. Ascidia cannelata (Oken), 1820 Phallusia sulcata Savigny, 1816: 162, pi. g, fig. 2. (Name preoccupied.) Phallusia cannelata Oken, I sis, 1820 : 796. Ascidia cannelata, Hartmeyer, 1916: 400, fig. i. One specimen, 32 mm. in length, growing on coral. 6. Rhodosoma turcicum (Savigny), 1816 Phallusia turcica Savigny, 1816: 165, pi. 10, fig. i. Seven specimens, all rather small except one 45 mm. long. This, apparently the only species of its genus, is found in many tropical seas, and is readily recognizable by the two apertures being near together in a cleft of the test which can be tightly closed to give them protection. Said to be in most places a rather uncommon species ; apparently the Gulf of Aqaba is an exception, as is also the island of Curasao, West Indies. 7. Cnemidocarpa hemprichi Hartmeyer, 1916 Cnemidocarpa hemprichi Hartmeyer, 19160;: 218, figs. 6, 7. One specimen of very irregular external form, about 29 mm. long. Found asso- ciated with coral in a depth of 2 fathoms. 8. Polycarpa mytiligera (Savigny), 1816 Cynthia mytiligera Savigny, 1816: 158, pi. 8, fig. 2. Polycarpa mytiligera, Hartmeyer, igi6a: 208, figs, i, 2. Two specimens, each of which contained a relatively large symbiotic macruran crustacean in the branchial cavity. 9. Herdmania tnomus (Savigny), 1816 Cynthia momus Savigny, 1816: 143, pi. i, fig. 2; pi. 6, fig. i. Cynthia pallida Heller, 1878: 96, pi. 3, figs. 17, 18. Five specimens, all of rather small size and apparently all representing the typical variety of this widely distributed species of warm regions. 2i8 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 10. Microcosmus exasperatus Heller, 1878 Microcosmus exasperatus Heller, 1878: 99, pi. 3, fig. 19. Three very small specimens. This is also a species of extensive distribution in tropical and warm-temperate waters. ii. Halocynthia spinosa Sluiter, 1905 Halocynthia spinosa Sluiter, 1905: 15, pi. 2, figs. 8-8d. Five specimens, the largest about 20 mm. in greatest diameter. This species, more or less red or pink in colour in life, is easily recognizable from its spiny exterior, the spines about the aperture on the siphons being especially long and conspicuously provided with sharp lateral branches. 12. Molgula dione (Savigny), 1816 Cynthia dione Savigny, 1816: 153, pi. 7, fig. i. One specimen, about 22 mm. long, found on coral. Class THALIACEA PELAGIC TUNICATA All the Thaliacea in the collection are of one species, Salpa maxima Forskal, 1775, which is found in both the Atlantic and Pacific Oceans, and though reported also from the southern part of the Indian Ocean, has apparently not previously been recorded from the Red Sea. The specimens, with the possible exception of some immature ones as noted below, belong to the following variety of this species: 13. Salpa maxima Forskal, 1775, var. tuberculata Metcalf, 1918 Metcalf, 1918, Bull. U.S. Nat. Mus., No. 100, 2 (2): 87, fig. 72. Described by Metcalf (who had examples of the aggregated form only, from the southern Philippines). The 'Manihine' collection has large adult examples of both aggregated and solitary forms, collected with dip nets near the surface, in some cases with the aid of a light. Five adult specimens of the aggregated form agree well with Metcalf 's description and figures, in having the anterior and posterior processes of the body longer than in the typical 5. maxima, and in having on each side of the external body surface an oval area of the thickened test at the base of the atrial siphon, bearing small acute conical spinous tubercles as described by Metcalf, the area on left side being the larger. Four adult examples of the hitherto undescribed solitary form of the variety tuberculata, the largest about 135 mm. in length, also differ from the solitary form of the typical 5. maxima in having external spinous areas, though these are small. There are three of these in the case of the solitary form, the most conspicuous one being a narrow transverse strip of thickened test extending across the rear end of the THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 219 body just below (ventral to) the base of the atrial siphon, bearing two not very regular rows of conical spinous tubercles similar to those in the aggregated form. The rows are one above the other, and extend slightly farther on the left than on the right side. On the dorsal surface of the body, above the intestinal ' nucleus ', there is on each side a thickened area of test bearing a few conical tubercles, but both areas are of small extent, especially the one on the right side. The variety tuber culata appears to be a well-marked one, but the differences from the typical form are superficial and hardly seem to justify considering it a distinct species, especially since we do not yet know the extent to which intermediate forms may occur. The collection also contains a number (over 50) of young specimens of 5. maxima, aggregated form, measuring up to about 20 mm. in length exclusive of the anterior and posterior processes. Many of these, when collected, were still adhering together as parts of chains, but due to transportation and handling are now all separated. It is likely that they are all the young of the variety tuberculata, but as they fail, pro- bably because too young, to show the varietal characters, they have been labelled simply Salpa maxima. REFERENCES FORSK!L, P. 1775. Descriptiones animalium . . . quae in itinere orientali observavit. (Tunicata, pp. 112-117, 129, 130.) HARTMEYER, R. 1916. t)ber einige Ascidien aus dem Golf von Suez. S.B. Ges. naturf. Fr. Berl. 1915: 397-43. J 4 fi g s - igi6a. Neue und alte Styeliden aus der Sammlung des Berliner Museums. Mitt. zool. Mus., Berl., 8: 203-230, 13 figs. HELLER, C. 1878. Beitrage zur naheren Kenntniss der Tunicaten. S.B. Akad. Wiss. Wien, 77: 83-110, 6 pis. HERDMAN, W. A. 1882-1888. Rep. Sci. Res. Voy. H.M.S. 'Challenger', 6, 1882, Ascidiae Sim- plices: 1-296, 37 pis., 23 text-figs.; op. cit. 14, 1886, Ascidiae Compositae: 1-429, 49 pis., 15 text-figs. ; op. cit. 27, 1888, Pelagic Tunicata and Appendix: 1-163, JI pls-> 2 $ text-figs. LESUEUR, C. A. 1823. Descriptions of several new species of Ascidia. /. Acad. nat. Sci. Philad. 3: 2-8, 3 pis. METCALF, M. M. 1918. The Salpidae: a taxonomic study. Bull. U.S. nat. Mus., No. 100, 2 (2) : 1-193, J 5 fig 8 -. J 4 pis. MICHAELSEN, W. 1919. Ascidiae Ptychobranchiae und Dictyobranchiae des Roten Meeres. Denkschr. Akad. Wiss. Wien, 95 (10): 1-120, 20 figs., i pi. igiga. Zur Kenntniss der Didemniden. Abh. Naturw. Hamburg, 21 (i) : 1-44, 3 figs. 1920. Ascidiae Krikobranchiae des Rothen Meeres. Denkschr. Akad. Wiss. Wien, 97 (9): 1-38, i pi. OKEN, L. 1820. [Translation into German of Savigny's work of 1816 (with the plates).] I sis (von Oken), 1820. SAVIGNY, J. C. 1816. Memoires sur les animaux sans vertebres, 2: 1-239, 24 pis. SLUITER, C. PH. 1905. Tuniciers recueillis . . . dans la Golfe de Tadjourah. Mem. Soc. zool. Fr., 18: 1-20, 2 pis. 1919. t)ber einige alte und neue Ascidien aus dem Zool. Museum von Amsterdam. Bijdr. Dierk. 21: 1-21, i pi. 220 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA APPENDIX Ascidian from Mukalla Bay Apparently Ascidia savignyi Hartmeyer, 1916 A large specimen of the genus Ascidia from Mukalla Bay, South Arabia (A. Fraser-Brunner, coll. 17-12-1948) is not included in the above list of specimens as it was not from the Gulf of Aqaba. It is remarkable for its large size (about 160 mm. long by 35 mm. transversely) and greatly elongated form, due chiefly to much lengthening of the anterior half of the body, though the siphons (both of which arise at the anterior end) are short, and the branchial one is much distorted. The internal structure does not show much abnormality, though the branchial sac extends close to the anterior end of the body, and the dorsal tubercle (whose aperture is irregular S-shaped, with the upper end bent down), also the neural gland and ganglion, are close to it and very near to the circle of tentacles. The branchial sac has no intermediate papillae ; the internal longitudinal vessels are numerous (over 70 on the left and over 80 on the right side) ; the intestinal loop (about 37 mm. long) is far back in the body. It is evidently an unusually old individual ; one that has grown in a favourable position in respect to food-supply and protection from predatory fishes and crabs, but where surrounding obstructions compelled it to become unusually elongated. A similar specimen might be hard to find again, but I do not think it should be assumed to be a new and undescribed species, though such mistakes have too often been made, resulting in burdening literature with supposed species having no real existence. Such a specimen is hard to identify with certainty, but I think it is an unusually large and abnormally shaped example of Ascidia savignyi Hartmeyer, 1916. (Sitzungsber. Gesell. naturf. Freunde Berlin: 1916: 404), des- cribed from the Sinai coast and Gulf of Suez. In that article Hartmeyer mentioned (p. 407) the close relationship of A. savignyi to A. depressiuscula Heller, 1878, described from Ceylon, and common in the Philippines, which is a species that also attains rather large size. I am quite ready to agree with this opinion, and think he was also probably correct in believing it related to the European species A . virginea Mueller, but I do not consider it also related to A . paratropa (Huntsman) of the American Pacific coast, as Hartmeyer believed. That species has intermediate papillae on the branchial sac, and belongs to a different section of the genus. IX. FISHES By N. B. MARSHALL, M.A. THE collection comprises 113 species, of which i is new, while 4 sub-species have been proposed. There are n new records for the Red Sea (these being indicated by an asterisk preceding the name of the species). Collections were made from 28 December 1949 to 16 February 1950, coming from various localities along the Sinai shores of the gulf and from an area around the en- trance. These include Aqaba, Faraun Island, Graa, Mualla, Wasit, Hobeik, Dahab, Um Nageila, and Abu Zabad within the gulf and Tiran Island, Sanafir Island, Sherm-el-Moiya, Sherm Sheikh, and Ras Muhammad Bay around the entrance. For the positions of these localities reference should be made to the chart in the intro- duction to this series of reports. The fishes were captured by a variety of methods: cast-net, fish-trap, hand-lines, trolling gear, and dip-net. In addition many were taken by bringing up pieces of coral and breaking them open to obtain the enclosed fishes, while a number were obtained from pools along the reef at low water. The method of capture is indicated under each species, giving certain information on the habits of the fish. For example, those taken by cast-net occurred singly or in shoals in shallow water close to the shore, while those taken by trolling spoon or live bait were nearly always caught along the seaward edge of reefs, where they appear to station themselves to prey on smaller fishes living in association with coral. Clearly those found within pieces of coral must live in close association with it, darting back to shelter on being disturbed by the diver. Perhaps no more striking way of appreciating the direct or indirect dependence of so many tropical fishes on coral can be obtained than through the many ways necessary to obtain them as specimens. SELACHII CARCHARINIDAE Negaprion acutidens (Riippell) i specimen of length 660 mm. 1 taken close inshore in Ras Muhammad Bay. Carcharinus melanopterus (Quoy & Gaimard) i specimen of 535 mm. caught by hand-line at Sanafir Island. Carcharinus albimarginatus (Riippell) i specimen of 870 mm. caught by hand-line at Sherm Sheikh. 1 Except for the Selachii and the eels, lengths throughout this paper refer to the standard length. 222 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA RHINOBATIDAE Rhinobatus halavi (Forskal) Six specimens were taken in very shallow water in Ras Muhammad Bay. One of these is a female of length 507 mm., while the rest are males ranging from 355 to 520 mm. DASYATIDAE Dasyatis uarnak (Forskal) One specimen taken by hand-line at a depth of 10 fathoms at the anchorage in Sanafir Island. The disk is about 1,000 mm. in length and 1,250 mm. wide. The tail, from which the whip-like end is missing, has a length of about 1,250 mm. Taeniura lymma (Forskal) Three specimens were obtained by cast-net close inshore at Sanafir Island (length 570 mm.), and at Mualla (length 445 mm.) and Um Nageila (length 564 mm.) within the Gulf of Aqaba. ISOSPONDYLI CLUPEIDAE Sub-family DUSSUMIERIINAE Spratelloides delicatulus (Bennett) Individuals of this species were taken with a dip-net and Aldis lamp at night. Faraun Island: 10 specimens from 21 to 50 mm. Sanafir Island: 15 specimens from 40 to 45 mm. Spratelloides gracilis (Schlegel) Like the preceding species, this was caught by dip-net at night in the light of an Aldis lamp. Hundreds of specimens were taken at the anchorage at Sanafir Island, ranging in length from 9 to 39 mm. I have compared some of these specimens with material in the museum collections from Japan and Formosa (the type locality being along the south-east coast of Nagasaki). There are differences in the number of pectoral and anal fins as shown in the table below: Pectoral (left) Anal No. of rays Red Sea specimens . Japanese specimens . 13 H 15 3 7 3 6 II 12 13 14 5 7 i 262 On the basis of the above counts it seems not unlikely that the Red Sea populations should be separated as a distinct sub-species ; but lacking data from areas between the end points of the range of this species it is considered premature to subdivide it. THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA 223 INIOMI SYNODONTIDAE Synodus variegatus (Lacepede) One specimen of 130 mm. taken in a pool at Dahab. APODES MURAENIDAE Echidna nebulosa (Ahl) Two specimens of 444 and 460 mm. taken on the reef at Abu Zabad at low tide. Echidna polyzona (Richardson) One specimen from Abu Zabad of 195 mm. and two from Sanafir Island of 115 arid 165 mm. The latter were found in a piece of madreporarian coral. Gymnothorax meleagris (Shaw) Seven specimens were obtained from the following localities: Dahab (108 mm.), Abu Zabad (145 and 160 mm.), Sanafir Island (in, 165, and 180 mm.), Sherm Sheikh (100 mm.). Except those from Abu Zabad, which were obtained on the reef at low tide, all were found in pieces of madreporarian coral brought up for examination. Gymnothorax flavimarginata (Riippell) One specimen of 295 mm. taken on the reef at Abu Zabad at low tide and one of 880 mm. from Ras Muhammad Bay. Gymnothorax geometrica (Riippell) Two examples of 130 and 143 mm. taken from pieces of coral at Sherm-el-Moiya and Sanafir Island respectively. The body colour of these specimens was fawn with the pattern of dark pigment FIG. i. An immature specimen of Gymnothorax geometrica (Riippell) from Sherm-el-Moiya in the northern Red Sea, showing the pattern of dark spots on the head and body. spots on the head looking rather like a series of lateral line pores (see Fig. i). These spots extend down the mid flanks as a single line, extending just beyond the anus, zoo. i. 8. eg 224 THE 'MANIHINE' EXPEDITION TO THE GULF OF AQABA This spot pattern was also found in specimens from the collections labelled Gymno- thorax thyrsoidea (Richardson): 2 from Rodriguez, i from the Seychelles, i from Muscat. These specimens differed from those listed above in having a dotted and speckled body coloration, but as all these Indian Ocean examples were much larger it is likely to be a difference due to age. Riippell's (1828) figure shows a mottled body coloration. There can be no doubt that G. geometrica (Riippell) and G. thyrsoidea (Richardson) are very closely related, the only apparent difference between them being the absence of the spot pattern in the latter. However, as this pattern seems quite constant in G. geometrica and as the pattern only appears to be found in individuals from the western Indian Ocean, the two species have been kept separate. Further work may perhaps show that what is now called geometrica is a western Indian race of a widely spread species. (This species will, of course, need to be called G. geometrica this name having priority over G. thyrsoidea.} If Gymnothorax geometrica is a distinct species its distribution must include the western Indian Ocean as well as the Red Sea. Uropterygius polyspilus (Regan) One specimen from Sanafir Island of length 201 mm. taken in a piece of coral. Schultz (1943) has suggested that this species is perhaps the young of Uropterygius tigrinus (Lesson). Examination of the above specimen, together with the type specimen from Tahiti (length 183 mm.) and two specimens from Samoa (331 mm.) and Zanzibar (715 mm.), shows that polyspilus is distinct from tigrinus (a specimen of 860 mm. was examined) in the following characters: 1. The number and size of the outer maxillary teeth: 20-28 in polyspilus, which are much smaller than the inner series of maxillary teeth; 12-13 m tigrinus, which are nearly the size of the inner series. (Bleeker, Atlas Ichthyologique, 4, 1864: 113, counts 16 outer maxillary teeth for tigrinus. The figure on plate 165 shows them almost equal in size to the inner series.) 2. The proportions between trunk and tail: about equal in length in polyspilus, but in tigrinus the trunk is about 1-7-1 -8 times longer than the tail. 3. The proportions between the eye and the snout: in polyspilus the length of the snout is from 1-7 to 1-8 times the diameter of the eye, whereas in tigrinus the snout is about 2-3 times the eye diameter. SYNENTOGNATHI BELONIDAE Strongylura crocodilus (Lesueur) One specimen from Sanafir Island of 465 mm. HEMIRHAMPHIDAE Hemirhamphus far (Forskal) One specimen from Sanafir Island o