Meas c=. ee Supplement I. will be published on March 1, 1905. FO ig 2 ae Re fre eee 4-8 The Fauna and Geography of the Maldive and Laccadive Archipelagoes Being the Account of the Work carried on and of the Collections made by an Expedition 3 during the years 1899 and 1900 Edited by J. Srantey Garpiner, M.A. Fellow of Gonville and Caius College and late Balfour Student of the University of Cambridge. VOLUME Il. PART IV. With Plates LXVII—LXXXVII and Text-Illustrations 127—139 CAMBRIDGE : at the University Press. Cambridge University Press Warehouse C. F. Cray, Manager. Lonpon: Ave Maria Lane, E.C. Price Fifteen Shillings net. MBL/WHOI! OAT 0 0301 00405915 e With the compliments of the Syndics of the Cambridge University Press. The Fauna and Geography of the Maldive and Laccadive Archipelagoes VOLUME Il. PART IV. CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, C. F. CLAY, Manacsr. ZLondon: AVE MARIA LANE, E.C. AND H. K. LEWIS, 136, GOWER STREET, W.C. Glasgow: 50, WELLINGTON STREET. Uripsig; F. A. BROCKHAUS. few Work; THE MACMILLAN COMPANY. Bombay and Calcutta; MACMILLAN AND CO., Lrp. [Al Rights reserved.] The Fauna and Geography ~~ of the Vie. Maldive and Laccadive Archipelagoes Being the Account of the Work carried on and of the Collections made by an Expedition during the years 1899 and 1900 Edited by J. STANLEY GARDINER, M.A. Fellow of Gonville and Caius College and late Balfour Student of the University of Cambridge. VOLUME Il. PART IV. With Plates LXVII—LXXXVII and Text-illustrations 127—139 CAMBRIDGE: at the University Press. e285) CAMBRIDGE: PRINTED BY JOHN CLAY, M.A. a AT THE UNIVERSITY PRESS. Wi; atl Ue iy ose WS, mreereree sien EGON ‘ ERIM Boyt \' oy GA Riu 7 # CONTENTS OF) VOR? tise aka aN. Reports. PAGE 1. The Alcyonaria of the Maldives. Part III. The Families Muriceadae, Gorgonellidae, Melitodidae, and the Genera Pen- natula, EKunephthya. With Plate LX VII : ; ; ; 807 By Sypyey J. Hickson, M.A., D.Sc., F.R.S. 2. Marine Crustaceans. XIV. Paguridae. With Plate LXVIII_ . 827 By Masor Atcock, F-.R.S. 3. Hydroids. With Plate LXIX é : ; ; 5 : 2 836 By L. A. Borrapaite, M.A. 4. Notes on Parasites - s P F : : 5 ; . 846 By A. E. Sarprry, M.A., F.R.S. 5. Rhynchota . : : : ' A ; , ; E ; 847 By W. L. Distan. 6. Marine Crustaceans. XV. Les Alpheidae. (Plates LXX—LXXXVII et Texte-Figures 127—139.) . 5 ; : ; , ; 852 Par Professeur H. Coutizre. THE ALCYONARIA OF THE MALDIVES. PART II]. THE FAMILIES MURICEIDAE, GORGONELLIDAE, MELI- TODIDAE, AND THE GENERA PENNATULA, EUNEPHTHYA. By Sypney J. Hickson, M.A., D.Sc, F.R.S., Beyer Professor of Zoology in the Victoria University of Manchester. (With Plate LXVIL) Our knowledge of the Alcyonarian fauna of the Indian Ocean is very imperfect. It is true that Gray, Milne-Edwards and Haime and many other writers of the earlier part of the last century have described species from the “East Indies,” but the descriptions are in most cases so superficial and the localities given so vague or doubtful that their records are of very little value. The only modern paper on the subject is that by Ridley (17) on the Alcyonaria of the Mergui archipelago. Neither the “Astrolabe,” “Alert,” the “Challenger,” nor the “Gazelle” visited the district, and the Alcyonarians collected by the naturalists of the “Investigator” have not yet been described. ; It is not surprising therefore that of the thirteen species of Gorgonacea described in this memoir from the Maldives no less than six appear to be new to Science. The classification of the Gorgonacea is still m a very unsatisfactory condition, but a thorough investigation of the anatomy of some of the leading genera is necessary before it can be placed on a more scientific basis. In the mean time I have decided to adopt as closely as possible the classifi- cation used by the authors of the Challenger volume on the Alcyonaria. According to that classification then the Gorgonacea collected by Mr Gardiner belong to the families Briareidae, Melitodidae, Muriceidae and Gorgonellidae. The important families of Gorgonacea which, according to this classification, are not represented in the collection, are the Sclerogorguidae, Coralliidae, Dasygorgidae, Isidae, Primnoidae!, Plexauridae and Gorgoniidae. The absence of any precious corals may have been disappointing but not surprising, as the family seems to be confined to regions of cooler water than that of the Maldives. The only statement we have concerning the occurrence of the precious coral in Indian seas is that made by Alcock (1. p. 291), “At one spot, not far from Trincomalee, M’Ardle found abundance of a beautiful pink coral which appears to be identical with the precious coral.” The absence of Primnoidae* and Dasygorgiidae is not surprising, as the members of these families are with a few exceptions usually found in deeper water than any investigated 1 As I was correcting the proofs of this paper Mr Gardiner _laris (W. and 8.) discovered by the Challenger in 600 fathoms sent me a fragment (25mm. long) of a Primnoid from Mahlos _ off the Kermadecs, Atoll, 24 fathoms. It is closely related to Stachyodes regu- Conus 104 808 SYDNEY J. HICKSON. by Mr Gardiner. The absence of Isidae however is noteworthy. Jsis was described by Steenstrup (fide W. and S.) from the Indian Ocean. It is common in some shallow water localities in the Pacific Ocean and coasts of the Malay archipelago. As regards the absence of Plexauride and Gorgoniidae it may be remarked that, although there are no pronounced types of these families in the collection, some of the specimens approach them very closely in their general characters. One specimen of Gorgonella verriculata, for example, has such a thick coenenchym and such a black flaky axis that it would have been placed with certainty in the family Plewawridae had it not been found in the same collection with the other specimen which is clearly a Gorgonella. The new species Elasmogorgia flewilis again might with justice be placed in the family Gorgoniidae, the slight roughness of the surface being the only character at present known which justifies the inclusion of the genus in the family Muriceidae. The boundary line between the family Muriceidae and the Gorgoniidae and that between the Plexawridae and the Gorgonellidae are quite artificial; and it is probable that a rearrangement of the genera included in these four families will be found necessary in the near future. The species of Gorgonacea described in this paper are Family Muriceidae. Acanthogorgia flabellum n. sp. Bebryce sp.? Acamptogorgia horrida un. sp. Elasmogorgia flewilis n. sp. Muricella perramosa (Ridley). Acis sp.? Family Gorgonellidae. Gorgonella verriculata (M. E. and H.). Juncella juncea (Pall). Juncella flexilis (Studer). Juncella elongata (Val). Scirpearia furcata n. sp. Scirpearella indica n. sp. Family Melitodidae. Melitodes variabilis n. sp. I have also described briefly the only Pennatulid sent to me, Pennatula murrayi Kolliker, and a specimen of Nephthyid I received too late to be included in Part I, Hunephthya maldivensis n. sp. The species of the family Briareidae were described in Part I (9). THE ALCYONARIA OF THE MALDIVES. 809 Family Melitodidae. Melitodes variabilis n. sp. Fig. 11. Locality. This species may be said, in general terms, to occur throughout the Archipelago in shallow water. In my last paper on the Alcyonaria of the Maldives I published a preliminary note on these specimens, in which I pointed out the difficulty of the problem they present to the systematist. Since then I have made a further careful anatomical investigation of the specimens, not sufficient perhaps to enable me to write very definitely and clearly on the subject, but sufficient to convince me that without further researches on the reef I should not be justified in recognising more than one species. The problem presents itself to my mind in this form. There are three characters in which the specimens show pronounced variations: 1. the colour of the spicules, 2. the length of the internodes, 3. the size of the spicules. These characters are, so far as I can judge, fairly constant within the individual colonies. All the specimens being fragmentary, however, this is a point that cannot be determined with absolute certainty. 1. The colour of the spicules may be a good character for the determination of a species of Alcyonarian, but as we know that at the Cape of Good Hope (Hickson 7) the spicules of the genus Melitodes do show striking variability im colour, and that in another Aleyonarian from the Maldives Chironephthya variabilis (9) there are very remarkable differences in the local distribution of spicule colour, it is at least very doubtful whether in this species the colour scheme of the spicules can be regarded as a good specific character. 2. The length of the imternodes is just one of those characters which varies with local circumstances. It varies in an individual colony, the imternodes of the main stem and primary branches being always shorter in proportion to their diameter and to the length of the nodes, than they are in the terminal branches. This, therefore, is not a character that can be used for determination of species without exact and precise knowledge of the local circumstances. 3. That the size of the spicules varies proportionately with the diameter of the internode, is, as a general statement of the facts, I believe, true. Melitodes variabilis is one of those species in which the size of the spicules from any one part of an internode varies considerably. It is only possible, by measuring the length of a considerable number of spicules from one internode and taking an average of their length, to arrive at any conclusion as to the figure to be taken to represent the length of the spicule of that internode. To feel strict confidence in the statement I have made it would be necessary to compare a large number of such averages from internodes of the same and different colonies. I have had neither sufficient time nor material to do this. I can only say that the comparisons of averages I have made point to this conclusion. In other characters than these three I can find, at present, no noteworthy variations. The verrucae are of approximately the same size, similarly situated, similarly retracted in all cases, and although there are still many points that might with advantage be investigated there are no other characters that seem to be very variable. This seems to me to be the case against the constitution of many species. 104—2 810 SYDNEY J. HICKSON. On the other hand Mr Gardiner writes to me as follows: “The point you have raised re Melitodes has interested me very much as it was one frequently discussed by my companion and myself. I held the opinion that they were all one species, and that it was only a question of looking to find two or more varieties in one colony, while Cooper held a contrary opinion. I then tried to find a mixture but without avail. This led me to look into the matter more closely, and then I noticed that they grew in patches, generally one form under each stone, though three or more forms were often in close proximity. I examined them to see if some were not more in the shade or subject to different conditions than others, but this was not so. I was surprised at this and therefore came to the conclusion that they were distinct species.” This valuable evidence from field notes is not conclusive, but it is of sufficient weight to shake one’s confidence in the contrary conclusion arrived at from work in the laboratory. Further investigations are now needed on the reefs, and to assist these I will add in a tabulated form my notes on the specimens, passing in succession from the North to the South of the Maldive archipelago. A study of this table shows that the variation in the specimens throughout the Archi- pelago is considerable and that with the present supply they cannot be grouped very satisfactorily into discontinuous groups, varieties or facies. Moreover we have evidence (vde Mr Gardiner’s letter) that “three or four forms were often in close proximity,” and from the table that the same or very similar forms may be found in very distant localities. As examples of this latter statement I would point out that the form with red nodes and red internodes occurs at Miladumadulu in the extreme north of the Archipelago and at Addu in the extreme South, and that the form with yellow nodes and yellow internodes occurs as far apart as Mahlos and Addu. The spicules of all the forms I have examined are of three kinds: (1) bent spindles with the tubercles longer on the convex side than on the concave side, (2) plano-convex spicules with tubercles longer on the convex side, (3) short smooth rods. The first kind occur in the coenenchym of the internodes, the second kind in the nodes, and the third kind in the anthocodiae. In addition to these a few smaller straight spindles, clubs or other forms of spicules, occur in the internodes (fig. 11). The spicules of the third kind are, in all the specimens I have examined, of very nearly the same size (0°07 mm. in length). Those of the second kind too are uniformly 0:13 x 0°07 mm. in size. The first kind however is very variable. In the Red-Yellow variety from Hulule, Male, they vary from 0-17—0:25 mm, in length; in the Red-Red variety from 0:14—0°27 mm., and in the Red-Grey variety from 0°15—025mm. The averages were found to be approximately ; Red-Yellow with a diameter of the internode 1 mm., average length of the bent spindles 0:214 mm, ; the Red-Red variety with an average diameter of the internodes 2 mm., bent spindles 0°228 mm. ; and the small Red-Grey variety with an average diameter of the internode 0°75 mm.: the bent spindles averaged 0:205 mm. in length. Taking all these points into consideration I think the best course will be to regard all the specimens from the Maldives as belonging to one variable species, which for the present may be regarded as new. Several species of the genus have been described by Studer (21), Wright and Studer (26), and one by Ridley (19). Of these the species that seem to approach it most closely are Melitodes laevis (W. and S. from Amboina) which has a similar mode of Colour Gray Red Red Pink Yellow Yellow Yellow Red Yellow Red Red Red Red Red Red Yellow Red 1Yellow Orange Red Hulule, Male Atoll. THE ALCYONARIA OF THE MALDIVES. NODES Diameter ee Colour INTERNODES AN Length Miladumadulu Atoll, 24 fathoms. 1°75 mm., swollen Gray 10 mm. Fragmentary Red Fragmentary 9 Orange as Mahlos Atoll, 22 fathoms. 2°25 mm. Orange 20 mm. Fragmentary | Yellow | Fragmentary North Male, 24 fathoms. 2mm., swollen | 1°75 mm., swollen Dirty white White 15 mm., fairly constant 20 mm. Hulule, Male Atoll, 15—20 fathoms. 4 mm., swollen 2mm., not swollen 1 mm., swollen 3 mm., slightly swollen 1 mm., slightly swollen 2mm., not swollen 1:25 mm., slightly swollen 1 mm., not swollen Red White White Pink Gray Red Pale Yellow Salmon 7—13 mm. 22—24 mm. 5 mm. 15 mm. 10 mm. 7 mm., very variable 8 mm., very variable 6 mm. Felidu Atoll, 25 fathoms. (Young colony) Koluma (Delicate fragment) | Orange Gray | (Young colony) dulu Atoll, 22 fathoms. Addu Atoll, 12 fathoms. 2°25 mm., swollen (Fragmentary) Addu Atoll, 25 fathoms. 5 mm., swollen 1:75 mm., slightly swollen 175 mm., slightly swollen 1 This specimen was a main stem. The others are apparently all branches. | | | Yellow Red Red Yellow Red 9 mm. Outer slope. 8 mm. 7 mm. 7 mm. 811 = EEeEaN Diameter 1 mm. 1 mm. 15 mm. 1:25 mm, 2 mm. 2 mm. 0°5 mm. 2 mm. Under dead coral masses close to W. Reef. 0°75 mm. 2 mm. 1 mm. J mm. 1:25 mm. 4mm. 15 mm. 15 mm. 812 SYDNEY J. HICKSON. branching, thin coenenchym, and similar spicules, and Melitodes rubeola (W. and 8.) from the Arafura Sea which shows a similar colour pattern, pink nodes with white mternodes, to that of some of the varieties of our species, but it differs from the former in the colour pattern, and from the latter in its mode of branching. The species may be described as “A delicate species branching in all planes with frequent anastomoses. Colour very variable; but the colour of the nodes and internodes frequently different.” Family Muriceidae, Verrill. This large family is well represented in the collection. The characters are ill-defined, and it might be fairly maintained that such genera as Muricella and Elasmogorgia would be more correctly placed in the family Gorgoniidae. The genera represented in the family are Acantho- gorgia, Bebryce, Acamptogorgia, Elasmogorgia, Muricella and Acis. Acanthogorgia flabellum un. sp. (figs. 1 and 2). Locality. Addu Atoll to seaward of atoll reef, 25 fathoms. The specimens of this species are in three pieces, and possibly they all belonged to one colony. The largest piece retains the main stem and a portion of the dise of attachment. It is 140 mm. in height and 140 mm. in width. The main stem is 25 mm. in diameter just above the disc of attachment, and the branches of the flabellum are on an average 1 mm. in diameter. The branching is entirely in one plane and the branches anastomose freely to form a reticulate flabelluam with meshes about 15 mm. from angle to angle. The coenenchym is very thin and the surface is rough, being beset with numerous regularly arranged tuberculate warts. The anthocodiae are cylindrical, 0‘5—1 mm. in length, with slightly projecting spines at the free edge. They are numerous and crowded on most parts of the flabellum, and situated principally in the plane of the flabellum. Young zooids are interposed between the full-grown ones. The colour of the coenenchym and anthocodiae is white. The axis is black in the older parts, brown in the younger. The axis is horny, but contains a core of bead-like calcareous bodies in definite chambers as in Acamptogorgia spinosa (Hiles 10). The spicules are colourless and very variable in shape. The spicules of the superficies of the coenenchym are so firmly interlocked and fused that they cannot be separated without injury. They are irregular tuberculated foliaceous clubs and branched spindles. In the anthocodiae a number of slightly bent spindles, from 2—3 mm. in length, are found. These are peculiar in that one half of their length is usually smooth and the other half tuberculated. At the free edge a few very long 05 mm. dagger-shaped spicules are found. In addition to these a number of very irregular foliaceous spicules occur both in the anthocodiae and coenenchym. This new species is the first species of the genus to be found in the Indian Ocean. The other species occur in the Atlantic and Pacific Oceans. Some species of the genus branch in one plane only and anastomoses of the branches are recorded in some of these (A. inermis Hedlund, 6; A. schrammi, D. and M., 2), but the formation of a reticulate flabellum such as we find in this species is unusual. The anthocodiae are distinctly smaller than usual. In several species described by Studer (20) they range from 1—2 mm. in A. truncata, to 4—8 mm. in A. verrilli. In this species they are rarely over, and THE ALCYONARIA OF THE MALDIVES. 813 more frequently under 1 mm. in height. In the literature of the subject there is very little reference to the structure of the axis. It is usually described as being horny (Hedlund, Wright and Studer), and it is horny and fibrous in the specimens of A. armata from the Cape (8), and A. spinosa from Funafuti, which I have had the opportunity to examine. The species therefore in this respect seems to be peculiar. In a genus with such variable spicules as Acanthogorgia it is difficult to determine species or to consider the relations of species by this character alone. The spicules of the new species are not unlike some of those figured by Hedlund from A. aspera, but they also resemble the spicules described by Ridley (19) from Hehinogorgia flabellum, and by Kolliker (13) from Paramuricea spinosa. Studer (22) has some hesitation in accepting the identification of Hedlund’s species as A. aspera (Pourt), but the figures Studer gives of Acanthogorgia horrida have a general family resemblance to those of Hedlund’s A. aspera and of A. flabellum. As a matter of fact it is really very difficult to draw any important distinction between the spicules of Paramuricea, Echinogorgia and Acanthogorgia, and there can be no doubt that the three genera are closely related. I had some hesitation in placing the new species in the genus Acanthogorgia, thinking it might be the same as the Indian species Paramuricea cancellata of Studer (Lchinogorgia flabellum Ridley), but an examination of Ridley’s specimen in the British Museum convinced me that it is more correctly placed in the genus Acanthogorgia. Of the species of Acanthogorgia it seems to be most closely related to Hedlund’s A. imermis from 30 fathoms off Hong Kong. The species may briefly be described as follows: Colony branching in one plane to form a reticulate flabellum. Coenenchym thin. Anthocodiae small, ‘5—1 mm. in height. Axis with a core of calcareous beads. Bebryce, sp.? Locality. S. Nilandu, 30 fathoms. S. Passage of Atoll. One small fragment of a Gorgonian having a superficial resemblance to a Muricella perramosa, but light brown in colour instead of crimson-red, belongs probably to a species of Bebryce. The specimen is 50 mm. in height and bears three short branches formed probably by dichotomy. The diameter of the branches is 175mm. Many of the Zooids are only partially retracted, the whole of the calyx being above the surface of the coenenchym, but in other places the coenenchym almost completely covers the calyx, forming verrucae 1 mm. in height. The axis is horny, fibrous, not calcareous. The remarkable character of the spicules is one of the features of the genus, and the spicules of this specimen agree fairly well with the description and size given by Kolliker (13). The genus has been found in the Mediterranean (B. mollis von Koch, 12) and a new species was described by Whitelegge (25) from Funafuti B. studeri. The Maldive specimen does not belong probably to either of these species, but as it is so fragmentary I do not feel justified in making a new species for it. If Whitelegge and I are correct in referring our specimens to this genus, its geographical distribution is most remarkable. 814 SYDNEY J. HICKSON. Acamptogorgia horrida, n. sp. (figs. 3 and 4). Localities. S. Nilandu, 25 fathoms. Suvadiva. Lagoon, coarse sand and rubble, 39 fathoms. There were two fairly complete colonies attached to stones and five fragments from Suvadiva. Judging from the specimens it may be inferred that the colonies are about 100 mm. in height, and they consist of about nine or ten branches, which in some cases branch again, given off from a short main stem. In some places the branching is apparently dichotomous, but it is not mvariably so. The branches meet the main stem at an angle which is nearly a right angle. The branches are not given off in one plane. The main stem and the basal parts of some of the branches are in the specimens bare, zooids and coenenchym alone occurring on the younger parts. The coenenchym is thin, the terminal branches are 0°7 mm. in diameter. The calices are not very prominent, but each is surrounded by a crown. of spines as figured by Miss Hiles (10) in A. spinosa. These spines project 0-1 mm. The diameter of the calyx as measured from spine to spine is 0°3 mm. The axis is black or brown, and none of the pieces I have examined contain lime; the core of the axis however is divided into a series of chambers by vertical partitions as described by Hiles for Acamptogorgia spinosa. The surface of the calices and coenenchym bristles with the smooth spines of the remark- able spicules so characteristic of the genus. Each of these consists of a tripod base, the limbs of which are about 0°15 mm. in length and covered with numerous rough tubercles, and a single sharp smooth spine 0'2 mm. long which projects freely at the surface. The base varies according to the position. On the coenenchym there are usually three limbs to each spicule, on the calices two, but many intermediate forms are found in which one of the three limbs is very small. Four lmbs are sometimes found. The limbs too are sometimes foliaceous and may interlock with one another or with the limbs of neighbouring spicules. In addition to these spicules there are some large slightly bent tuberculated spindles at the surface of the coenenchym which are about 0°6 mm. long x ‘1 mm. broad. The species seems to be most closely related to Acamptogorgia spinosa (Hiles) from Funafuti, but differs from it in being more robust in habit, less finely branched, much more spiny in appearance, and in the presence of the large spindles of the coenenchym. Elasmogorgia flewilis, n. sp. (figs. 5, 6 and 7). Locality. Suvadiva, 20 fathoms. In a passage towards the lagoon. And 37 fathoms. Hard sand near a passage, Some very delicate thread-like forms must be provisionally referred to the genus Hlasmo- gorgia. Studer (26), who established this genus, considered that it might be the same as Gray’s (5) Filigella, but owing to the loss of the type specimen and the inadequate description given by Gray, he very properly proposed a new name for it. I have compared the specimens obtained by Mr Gardiner with the description and figure given by Gray, and think that it is highly probable that they belong to his genus. For the reasons given by Studer, however, I refer it to the genus Hlasmogorgia. THE ALCYONARIA OF THE MALDIVES. 815 There are in all ten or eleven specimens varying in length from 240—300 mm. Two of them are adherent to stones with a small disc of attachment. As in so many other cases, I had some hesitation in placing these specimens in any of the known genera. They seem to differ from Muricella in being much less ramified and in having much smaller spicules. They agree better in the character and size of their spicules with the genus Stenogorgia (Verrill), but this genus is more robust and much more ramified. They differ from the description of Filigella gracilis given by Gray in two particulars. Gray describes the axis as “rather rigid”; the axes of these specimens are very flexible. Gray further describes the verrucae to be 1 inch or # inch apart; they are much closer set in these specimens. The specimens agree much more closely with the description given by Studer of Elasmo- gorgia filiformis, but our specimens are rather more delicate and the spicules much smaller. The spicules of Hlasmogorgia filiformis are from 013 to 0°62 mm. in length, whereas in our specimens they rarely exceed 0°2 mm. in length and the great majority are not over "15 mm. in length. They differ moreover in having fewer and coarser tubercles. There can be little doubt that this is a new species, and it can be easily distinguished by the size and character of the spicules. Elasmogorgia filiformis (Studer) was found in 28 fathoms S. of Papua, Filigella gracilis (Gray) off Cape Frio, Patagonia. All the specimens are unbranched except one, and this gives off only a single branch at right angles to the main stem. The average diameter of the stem is between 0:5 and 1 mm., and at the free extremity there is a knob-like swelling 1:2 mm. in diameter very similar to that shown in Filigella by Gray. The coenenchym is thin but not transparent. The verrucae project slightly from the coenenchym and completely enclose the calices. Their longest diameter, that parallel to the axis, is 1 mm. They are situated at distances of 2 or 3 mm. apart, and are not arranged in definite rows on opposite sides of the stem, but are irregularly distributed round it. The spicules of the coenenchym are stout, rough, irregular, warty spicules of various sizes, usually 0°15 mm. in length, but rarely exceeding 0:2 mm. x ‘08 mm. In the tentacles there are a few delicate needles. The axis is brown to dark-brown in colour and quite devoid of hme. The colour of all the specimens is white. Muricella perramosa Ridley (18). Localities. S. Nilandu, 30 fathoms, S. passage of Atoll, 1 specimen. Suvadiva Atoll, 33 fathoms. In passage, 2 specimens. Several species have been placed in this genus, but few of them have been at all adequately described. Muricella humosa (Esper) and Muricella tuberculata (Esper) were pro- bably obtained in the East Indies. The specimens from the Maldives agree most closely however with Muricella perramosa from Mauritius, 90 fathoms, described by Ridley (18). The specimens are all small, being 80, 60 and 40 mm. respectively in height. Ridley’s largest specimen was 500 mm. in height by 475 mm. in extreme breadth. It is noteworthy G. I. 105 816 SYDNEY J. HICKSON. that the specimens from both localities exhibit some branches which are bare or torn, sug- gesting that they might possibly have been bitten by fish; but they are also permeated by a parasitic alga(?) which has a milk-white flocculent appearance in spirit. The spicules are of the characteristic shape, those of the cortex being about 25 mm. x ‘07 mm., and those of the verrucae 0:15 mm. x ‘06 mm., but they exhibit a good deal of variation in the size. The colour of the specimens is crimson-red. Genus Acis, sp. ? Locality. South Nilandu, 30 fathoms. North Male, 20 fathoms. When this paper was nearly ready for the press and I was arranging the collection in separate bottles, I came across three fragments of Gorgonians which clearly belonged to species distinct from any I had described. These I now find undoubtedly belong to the genus Acis which is characterised by the large thick spindle-shaped spicules of the cortex. Three species of this genus are known: one (A. guadalupensis, D. and M.) occurs in the West Indies, another (A. orientalis, Ridley) in the Mauritius, and a third was discovered by the Challenger (A. pustulosa, W. and §.) off Japan. The occurrence of the genus in the Maldives is therefore a matter of some interest. In some respects our specimens come close to A. orientalis, and I was tempted to regard them all as belonging to this species, but the remarkable colour differences and the fragmentary condition of our specimens render it necessary for me to consider the species doubtful. The first specimen is a delicate branching form with yellow calices and orange-coloured coenenchym. This orange colour is due to the fact that most of the large coenenchym spicules are yellow with a dark red core. The coenenchym is thin, the branches about 1 mm. in diameter. The second specimen consists of two branches 40 mm. and 25 mm. in length. They are much more robust than the branches of the first specimen, being 2 mm. in diameter with a much thicker coenenchym. The spicules of the outer layer of the coenenchym are white and of the inner layer red. The third specimen—the only one from N. Male—resembles the second specimen from S. Nilandu in form and size, but the spicules are all red. The axis of all the specimens is horny with a core of lime. Family Gorgonellidae Verrill. Gorgonella verriculata, Milne-Edwards and Haime. Syn. Gorgonia reticularis, E. and S. (?). Localities. N. Male, 20 fathoms W. of Atoll; across a passage. Mahlos Atoll, 23 fathoms ; inner part of passage to S.E. One dried specimen of which the record of locality is lost. It is with some hesitation that I place these specimens in Milne-Edwards and Haime’s species. The description given by these authors is very inadequate, and Studer’s (20) account of a specimen which he attributes to the species does not add much to our knowledge. The THE ALCYONARIA OF THE MALDIVES. 817 specimens are placed in the genus Gorgonella on the grounds that the axes contain a considerable amount of lime, that the coenenchym is moderately thick, that the verrucae are slightly or not at all prominent, and that the spicules are of the general size and shape common to many species of the genus. The specimen described by Ridley (17) under the name Gorgonella umbella Esper, resembles our specimens to some extent, but as the author does not state what the character of the axis is, it may be a matter of some doubt whether they really belong to this family at all. There is some difficulty in drawing a hard line of distinction between this genus and Verrucella which it resembles in many respects. But Verrucella appears to have a more delicate habit, to have smaller spicules and to have more prominent verrucae. The verrucae are moreover situated principally on the sides of the branches and not scattered all over as they are in our specimens. The important characters which distinguish our specimens may be summarised as follows :— Corallum branching profusely in one plane to form an elaborate reticulate flabellum. Verrucae very slightly or not at all prominent, unevenly scattered over the surfaces of the main branches and branchlets. Coenenchym moderately thick. Axis containing a considerable quantity of caleium carbonate. Spicules of the coenenchym, spindles and a few double clubs. The colour of our specimens is very pale brown or white. In this respect they differ from Studer’s specimen which he describes as “ lackroth.” The largest specimen (from N. Male) is about 125 mm. in height and 150 mm. in width. It branches profusely in one plane and the branches anastomose freely to form meshes, many of which are 20—30 mm. long by 5 mm. broad, but in the course of the reticulum there are many much smaller meshes, 5mm, x 38mm. or less. The terminal branches are about 2 mm. in diameter. The anthocodiae are withdrawn completely, but the retracted tentacles can be clearly seen through an open gap in the coenenchym. The verrucae are very slight in most parts of the colony, rather more pronounced in others. The general appearance of this colony is very similar to Ellis’s picture of Gorgonia reti- cularis (3), but unfortunately this author gives no description of the species. The spicules of the coenenchym are of two kinds. The majority are blunt spindles with crowded exfoliating tubercles and their length is fairly uniformly 0°71 mm.; but in addition to these there are a few small double-heads (043 mm. in length. In the tentacles there are spindles, some straight and some slightly bent, smoother and flatter than the spindles of the coenenchym, and usually a trifle longer. The only statement I can find about the spicules of this species is that of Studer (20) “Die Spicula sind warzige Spindeln und Doppelkeulen.” The second specimen (from Mahlosmadulu Atoll) is in several pieces, and it is impossible to estimate at all accurately the size and shape of the colony as a whole. The coenenchym is rather thicker than in the N. Male specimen and the zooids are even more contracted, so that the verrucae have the appearance of small pores at the summit of very slight prominences. In fact, taking all the superficial characters together, it is more like a Plexaurid than a Gorgonellid. The investigation of the spicules and other microscopic characters, however, leaves little doubt that it belongs to the same species as the N. Mahlos specimen. The dried specimen from an unknown locality is only a fragment, and is noteworthy only from the fact that it is of a rufous colour. 105—2 818 SYDNEY J. HICKSON. Genus Juncella. In the family Gorgonellidae there are four genera recognised by systematists, which are undoubtedly closely related. These are Juncella, Ellisella, Scirpearia and Scirpearella. They are all either unbranched or, as compared with other Gorgonellids, slightly branched, and the coenenchym is usually thick. All the known species were at one time included in the genus Juncella and the first attempt to separate them was made by Studer (20) in 1878. This author placed the species with prominent verrucae and with club- or double club-shaped spicules in the genus Juncella, and those with not prominent verrucae and without club-shaped spicules in the genus Ellisella. The use of the distinction between prominent and not-prominent verrucae for generic purposes is very inconvenient and the source of many mistakes. There are many specimens that on the ground of the character of the spicules should be placed in the species Juncella (Ellisella) elongata which have prominent verrucae. Ridley (18) points out moreover that in his specimen of Juncella juwncea the verrucae are withdrawn above, but prominent and appressed below. The determination of the question whether any particular verruca is to be called prominent or not prominent is in many cases very difficult, and can be of little scientific importance unless all the specimens compared have been killed and preserved in the same manner. The question therefore becomes one in which we are driven to determine the generic distinction between two alcyonarians by the character of the spicules alone. In some cases perhaps this may be necessary and possible, but the spicules technically known as clubs show much variation, and when the handle is enlarged are very similar to spicules known as double- clubs. I doubt very much whether it is wise, in the present state of knowledge concerning the growth of spicules, to put too much stress upon fine distinctions of their shape. It may be pointed out in this connection, that the Cape specimens, which in a former paper (7) were referred to, the species Juncella elongata would, if judged by the character of the spicules, alone be placed in a distinct genus. On these grounds therefore I have little hesitation in referring all the species hitherto included in the two genera Juncella and Ellisella to the former genus, thereby suggesting the abolition of the latter. The genera Scirpearia and Scirpearella, however, appear to me to be still good genera. The arrangement of prominent verrucae in two rows in the former genus and in a spiral manner in the latter, combined with other characters, renders them relatively easy of identification. The principal species of the genus Juncella in its wider sense that are now known may be arranged as follows: Juncella juncea (Pallas). Juncella gemmacea (Milne-Edwards). Juncella flexilis (Studer). Juncella fragilis (Ridley). Juncella barbadensis (W. and S.). Juncella racemosa (W. and S.). With club-shaped spicules. : Juncella elongata (Val). Juncella calamus (Studer). Juncella maculata (Studer). Juncellu spiralis (Hickson). Without club-shaped spicules. THE ALCYONARIA OF THE MALDIVES, 819 Juncella hepatica Kz (11), from the Red Sea, has been obtained only once in a frag- mentary condition and may not be distinct, Juncella funiculina M. and D. (2), from the W. Indies, and Juncella laevis Verrill (23), from Hong Kong, are not sufficiently well known to classify in this system. Of these species the most widely distributed and best known are: Juncella juncea, from Bourbon (Gray), Mauritius (Lamouroux), Ceylon (Ridley), Ceram (Rumphius), Torres Straits (Challenger), N.W. Australia (Studer), Queensland (Ridley), Singapore (Studer, 21), Muscat (Man. Mus.). It appears to be confined to shallow water, 7—50 fathoms. Juncella gemmacea, from the Red Sea (Kiz., etc.), Torres Straits (Challenger), N.W. Australia (Studer), Burmah (fid. Ridley), Queensland (Ridley), Singapore (Studer and Klz.). It extends from shallow water to 80 fathoms. Juncella elongata, from Atlantic Ocean (Pallas), West Indies (E. and 8.), N.E. Australia (Ridley), Cape of Good Hope (Hickson). It may be regarded as very doubtful, however, whether the species was in all cases correctly identified by these authors. Of the less widely distributed species: Juncella flewilis was found between Flat Island and Mauritius, 25 fathoms (Studer), and at Ternate (Germanos). Juncella fragilis, Queensland and Mergui (Ridley), in shallow water. Juncella barbadensis, Sombrero Island, 450 fathoms (Challenger). Juncella racemosa, Japan, 345 fathoms (Challenger). Juncella calamus, N.W. Australia, 50 fathoms (Studer) and Queensland, 4 fathoms (Ridley). Juncella maculata, Torres Straits and Banda (Challenger), N.W. Australia, 50 fathoms (Studer). Juncella spiralis, Cape of Good Hope, 25 fathoms (Hickson, 8). The principal difficulty the systematist meets with is to distinguish between J. gemmacea and J. juncea. These two species agree in having more or less prominent verrucae according to the method employed of capture and preservation, in having a thick fleshy coenenchym, in being slightly branched or unbranched, in their great size (700—900 mm. in the case of large specimens), and in having coenenchym spicules consisting of clubs and double stars from 0:°05—0'1 mm. in length. In Juncella juncea, however, the verrucae are not usually crowded, there is only a single bare track in some specimens, the colour is usually dark-red, rarely white. Most of the specimens are unbranched and may reach a great length without branching. In Juncella gemmacea the verrucae are usually more crowded, and there is a bare track on both sides of the coenenchym, which in larger specimens is usually invaded by verrucae ; the colour is much more variable, orange-red, orange, coral-red, white or cream coloured. Specimens when of large size are usually branched. The smaller specimens of J. gemmacea are unbranched, hence the difficulty. As pointed out by Ridley (17) these species are so variable that it is very difficult to distinguish them. Perhaps they should be merged into one variable, widely distributed species. 820 SYDNEY J. HICKSON. As regards Juncella fragilis, Ridley himself felt doubtful whether it was a good species. Probably it is a young form of J. juncea. It is unbranched. Juncella flemilis may also be but a young form of another species. It is unbranched, dark-red in colour, and distinguished by its slender axis and thin coenenchym. The verrucae are arranged in two rows on the lower part of the stem, but in the upper parts are not so restricted. Juncella spiralis is of relatively small size when compared with the larger specimens of Juncella juncea and J. gemmacea, and is distinguished by the pronounced single bare track and by the spiral form. Juncella elongata, J. maculata and J. calamus were formerly placed in the genus Ellisella and agree in having no typical club-shaped spicules. J. elongata is a large, branching species, very similar to J. gemmacea in form. It may be red or pale pinkish-yellow. The surface may be smooth (Studer) or the verrucae very promi- nent (Ridley). It is very difficult to distinguish J. maculata from J. elongata. The specimen described by Studer was only 25 cm. long. The colour was orange-red, with the coenenchym immediately surrounding the verrucae red. J. calamus may be but a variety of the same species. It was found in the same dredging as J. maculata, and the only material feature in which it differs from it is that in the lower part the verrucae are arranged in five rows. It is unbranched. J. barbadensis and J. racemosa are both from deep water and appear to be quite distinct species. The former seems to be related to J. flewilis, but may be distinguished by the very long peduncles of the verrucae and the larger size of the club-shaped spicules. J. racemosa is distinguished by the peculiar and exceptional method of its branching. Juncella juncea, Pallas. Localities. S. Nilandu. W. passage of Atoll, 25 fathoms (Spec. A). N. Male Atoll, 20 fathoms, W. of Atoll. Across a passage (Spec. B). Specimen A was fragmentary, 70 mm. long, 5 mm, in diameter. Colour dark-red. The verrucae project but slightly and are about 1:3 mm, in length and breadth. They are directed towards the apex. The spicules are clubs and very spiny double clubs, the former about 0:1 x 0:025 mm., the latter 011x005 mm. There are no spindles or double stars, The spicules of this specimen are very similar to those figured by Kollker, Specimen B is 370 mm. long and about 5 mm. in diameter. It was not branched. Colour dark-red. Verrucae numerous, projecting, turned towards apex. The spicules are very similar to those of Specimen A. It is impossible to determine whether these specimens are more correctly placed in this species or in the species J. gemmacea, but as there is no sign of ramification I have placed them with J. juncea. THE ALCYONARIA OF THE MALDIVES, 821 Juncella flewilis, Studer (20). Locality. Hulule, Male, 25—30 fathoms. A very delicate unbranched specimen, may be provisionally referred to Studer’s species. It is 140 mm. long and barely 1 mm. in diameter. The verrucae are prominent but not so prominent as in the type. The coenenchym is thin. The spicules are clubs with thick handles (i.e. they are intermediate in form between typical clubs and double clubs) ‘09 mm. in length, and a few tuberculate spindles. The colour is red. The specimen seems to be more delicate than the type which, judging from the figure, must have been at least 2 mm. in diameter; but in the thinness of the coenenchym, the shape and size of the spicules and the colour, it is very similar to the description of the species. The type specimen was obtained off Mauritius in 25 fathoms. It is quite possible that these specimens, however, are only young stages of Juncella juncea. We have no information at present as to the mode of growth of the Alcyonarians of this genus, and it may be that in the early stages they grow rapidly in length in order to reach a level in the water where food can be more readily obtained, and that the growth in the thickness of the coenenchym takes place at a later stage. Juncella elongata (Val). Localities. Hulule, Male Atoll, 25—30 fathoms, close to edge of reef. Specimen A. S. Nilandu, 25 fathoms. W. passage of Atoll. Specimens B and C. S. Nilandu, 30 fathoms. W. passage of Atoll. Specimen D. As there are only fragments of the specimens, which I believe should be placed in this species, the identification may not be regarded as very satisfactory. All the specimens are unbranched. Specimen A is in three pieces and reached a total length of 315mm. The total diameter is 34 mm., and the axis is 2 mm. in the middle region. Nearer the base the coenenchym is relatively thin or very thin, and nearer the apex much thicker. The colour of the coenenchym is pale-pink and the verrucae are throughout shallow domes, white in colour. Specimen B is a fragment 60 mm. long by 3 mm. in diameter. The surface is almost smooth and pale-red in colour. Specimen C is about the same length but much more slender. Specimen D is 230 mm. long, with verrucae scattered and becoming prominent towards the distal end, but with an almost smooth coenenchym near the base. Colour orange-red. The spicules are double clubs, warted spindles, and a few more elongated spindles with fewer tubercles arranged in regular rows. The warted spindles and double clubs vary in size from 0°08—0-085 mm. in length. Some of the poimted spindles are 0-1 mm. in length. There is evidently a good deal of variation in the shape of the spicules. Im A and C the double clubs predominate, in D the warted spindles predominate. In C there are many unsym- metrical double clubs similar to those found in Juncella juncea. The colour, the prominence of the verrucae and the definiteness of pronounced tracks free from verrucae, are also characters in which the species shows much variation. 822 SYDNEY J. HICKSON. Scirpearia furcata, n. sp. Figs. 8 and 9. Localities. S. Nilandu, 25 fathoms. W. passage of Atoll. N. Male, 20 fathoms. W. across a passage. Two specimens were obtained from S. Nilandu. One was 90 mm. long x 15 mm. diameter, the other 75 mm. x 4mm. diameter. Both exhibit an orange-red coloured coenenchym with dark-red dome-shaped verrucae, closely crowded but separated into two groups by broad spirally directed bare tracks. The former more delicate specimen has a single branch which is bifur- cated at its extremity. The spicules are principally warted spindles of a dark orange colour, 0°l mm., a few paler double clubs 0°1 mm., and double stars 0°75 mm. The specimen from N. Male is only a fragment. The species differs from S. ochracea and S. flagellum in having less prominent verrucae, in the size of the spicules, and more particularly in the character of branching which is shown in one specimen. Scirpearia furcata, variety ? Locality. S. Nilandu, 24 fathoms. There is a long 200 mm. slightly branched specimen from this locality, which differs somewhat from the type. It is more delicate in build and has less prominent verrucae, and the colour is not such a pure red but tinged with orange. The spicules are of the same size and variety, but there are in proportion more double clubs. Scirpearella indica, n. sp. Fig. 10. Locality. S. Nilandu, 30 fathoms. W. passage of Atoll. There are six pieces of this species in the bottle from this locality: A, a fragment 280 mm. long, white coenenchym with prominent verrucae, red at the apex. B, a complete specimen about 330 mm. long, verrucae not quite so prominent as in A. C, D, E, three pieces which probably represent one complete specimen 210 mm., 70 mm. and 150 mm. in length respectively, 480 mm. in total length if they belong to the same colony. F, a specimen which was apparently broken off and has undergone repair, 220 mm. long. The Colour of the specimens is remarkable: In A the coenenchym is white, but the pointed tips of the verrucae are red. In B the verrucae are white throughout, but there are streaks of pink running irregularly along the coenenchym and uniting at the base to give the coenenchym a general pale-red colour. The other specimens are entirely white. The only basal dise present is 12 mm. in diameter. All the specimens are unbranched. The diameter of the specimens varies very little and is in all about 3}—4 mm. The apex is blunt, the base enlarges abruptly to form the disc of attachment. THE ALCYONARIA OF THE MALDIVES, 823 The verrucae vary considerably. In A they are pointed and about 1 mm. in height, at the base of B they are broader and less prominent. In places they have an appearance like “a shallow ledge that reminds one of the edible nests of the swallow (Callocalia),” similar to that described by Wright and Studer for S. profunda (26). The verrucae are arranged in 6 or 7 slightly spiral rows. The only specimens of this genus that have hitherto been described were found in the Pacific Ocean. S. profunda, New Hebrides, 130 fathoms; S. gracilis, New Hebrides, 130 fathoms ; S. moniliforme, Amboina, 100 fathoms; S. rubra, off Japan, 345 fathoms. The specimens from Nilandu should be placed in a new species for which I propose the name Scirpearella indica. The principal characters that distinguish the species are that the spicules rarely exceed 0°85 mm. in length and are in general smaller than they are in other species, and that the verrucae are arranged in 5—7 rows. Order PENNATULACEA. Family Pennatulida. Pennatula murrayi, Kolliker (14). Locality. Suvadiva, 43 fathoms. Lagoon. Only two Pennatulids were sent to me and both of them belong to this species. The species was originally discovered by the Challenger in 129 fathoms on the S.E. coast of Ceram. It was subsequently described by Moroff! from the coast of Japan, and has recently been re- discovered by the Siboga in 112—397 metres in the Ceram Sea. The species is characterised by the enlargement and modification in structure of one of the siphonozooids at the base of each of the leaves. The structure of this remarkable siphono- zooid will be fully described in a subsequent paper. The two specimens were respectively 305 mm. and 60 mm. in total length. The larger one exhibits 75 leaves on each side of the rachis, many of them being quite rudimentary, and the largest leaves are composed of eight autozooids. The smaller specimen has only 19 pairs of leaves, the largest with four autozooids. The colour is, as in the other specimens of the species, perhaps the most delicate and beautiful of any species of Pennatulid. The terminal portion of the stalk is pink, but at the base of the lowermost leaves there is a broad band of dark-red, the rest of the rachis is pale-red. The leaves are yellow but each of the largest anthocodiae of the leaves bears a crown of red spicules. Both of these Maldive Pennatulas are probably young ones, as was also the type specimen from Ceram. Many of the Siboga specimens which are now in my hands for description are much larger, the largest of all being 600 mm. in total length. 1 Th. Moroff, Zool. Jahrbuch xvu. 3, System. G. IL 106 824 SYDNEY J. HICKSON. Order ALCYONACEA. Family Nephthyidae. Eunephthya maldivensis, n. sp. Fig. 12. Locality. Kolumadulu Atoll, between 22 and 45 fathoms. A single specimen was obtained. It has a total height of 40 mm. with a stalk 20 mm, in length, and numerous short pointed branches. The largest anthocodiae have a width across the crown of 0°6 mm. and, varying according to the condition of their retraction, protrude 06—1'5 mm. from the coenenchym. The anthocodiae are not collected in bundles, but large and small are scattered evenly on all sides of the branches. The tentacles are either expanded or folded over the peristom. They are about 0°3 mm. in length in the spirit specimen, and each bears 7—9 pinnules on each side. The spiculation of the specimen is remarkable. The surface of the stalk bears a crowd of very irregular stellate (fig. 12 b), rod-like and short spindle-shaped spicules, with an average length of about 0°2 mm. In the mesogloea of the stalk there are numerous stout tuberculated spindles 0°5 mm. x 0°1 mm. On the surface of the branches the irregular spicules are wanting, but a few giant boomerang-shaped spindles are scattered about very irregularly. These large spicules have a remarkable resemblance to the spicules of the coenenchym of Spongodes, in shape, in the irregularity of their distribution and their great variability in size. Many spicules can be found among them which are 15 mm. in length, but the average length would probably be found to be about 1°2 mm. and some are less than 0°5 mm. (fig. 12 a). The aboral sides of the tentacles are armed with small irregular spicules, mostly rod-shaped with long tubercular processes having a length of ‘05—1 mm., and a few of the large spicules are occasionally found on the body wall (fig. 12 c). The specimen has very much the form and size of the specimen described by Kiikenthal (15) from Ternate, as Paraspongodes crassa, but it differs from it in having the branches rather more pointed, and in some of the details of the spiculation. It has also some relation to the genus Lemnalia as recently described by Bourne, but differs from it in the scarcity and great size of the spicules of the coenenchym. In a recent paper Studer (22) has given reasons for retaining the genus Hunephthya, and the new definition he gives of it accords more nearly with the species from the Maldives than that of any other genus. Kiikenthal!, moreover, in his last paper also accepts the genus Hunephthya. According to the definition of the genus given by either of these authors the specimen clearly belongs to it. As it seems to differ, however, from any species that has yet been described I propose to call it #. maldivensis. 1 W. Kiikenthal, Haeckel’s Festschrift, 1904. THE ALCYONARIA OF THE MALDIVES. 825 LITERATURE. 1. A. Aucock. “A Naturalist in the Indian Seas.” 1902. 2. P. DucHassainc and T. Micuenorri. “Mémoire sur les Coralliaires des Antilles.” 1860—1866. J. Extis and D. SoLanper. “Natural History of the Zoophytes.” 1786. 4. J. E. Gray. “Catalogue of Lithophytes.” 1870. 5. sa Ann. Mag. Nat. Hist. 1868. Part 11. p. 443. 6. T. HEDLUND. Bihang K. Svenska Vet. Akad. Handl. Bd xvt. tv. 6. 7. S. J. Hickson. “The Alcyonaria of the Cape of Good Hope.” Part 1. 1900. 8. — “The Alcyonaria of the Cape of Good Hope.” Part 1. In the Press. 9. —— “The Alcyonaria of the Maldives.” Part 1. This Series, Vol 11. Part I. 10. I. Hines. Proceedings Zool. Soc. x1. 11. C. B. Kiunzincer. “Die Korallthiere des Rothen Meeres.” 1847. 12. G. von Kocn. “Die Gorgoniden.” Fauna und Flora Neapel. xv. 13. A. K6LLIKER. “Icones Histiologicae.” 1865, 14, “The Pennatulida.” Challenger Reports, Vol. 1. 15. W. Kixentuat.