A “@. ; } ah re f aed? i @:: 4 4 By Cae x OO Male hesaad ea Sr af” § my oe LIBRARY OF THE UNIVERSITY OF CALIFORNIA. RECEIVED BY EXCHANGE Class LIBRARY OF THE UNIVERSITY OF CALIFORNIA. RECEIVED BY EXCHANGE Class eee fai: = ae ae THE DANISH TNwee@t Fr EXPEDITION. WRAL. TVA. PUBLISHED AT THE COST OF THE GOVERNMENT BY ms THE DIRECTION OF THE ZOOLOGICAL MUSEUM OF THE UNIVERSITY. COPENHAGEN. or H. HAGERUP. ~ PRINTED BY BIANCO LUNO AJs. 1903-1907. Echinoidea, I, p-1-193 (21 plates), 1903. . ‘ s OY. cae : as) . #4 ‘ EDEL LON Poe CONTIN Re Gee ae et | TH. MORTENSEN: BCHINOIDEA, = —s—S—S—S PUBLISHED AT THE COST OF THE GOVERNMENT | PG re @ - G hes. . =a ‘ a Li : | ae DECOR OF pet ZOOLOGICAL MUSEUM OF THE UNIVERSITY : - 5 Y, : : - : ; fers 3 ee : aad : , : UNIVERSITY | wide * | a : is : ; ra ee : nike Be + ‘ Z a < % ” - = H o> F & < ; rt THE DANISH INGOLF-EXPEDITION. VOLUME IV. is EL HINOTDEA. (PART L,) BY TH. MORTENSEN. WITH 21 PLATES AND 12 FIGURES IN THE TEXT. TRANSLATED BY TORBEN LUNDBECK. COPENHAGEN. PRINTED BY BIANCO LUNO. 1903. CONTENTS. Echinoidea. Page Page EERE as IR east aah wh cia pik fine 2 oN oaks es 3 I. | On the Fam. Echinometrade Gray and the Subfam. Tripl- On generic and specific Characters in the Echinoids..., 3. TENE, TSG CS egg ee Rel iene ey cp oe go. EMIS ed cig c's kee noe edabay feb Mes. s 11. | Diagnoses of the Fam. Stomopneustidee, Echinidz, Toxo- Diagnoses of the genera of the Fam. Cidaride.......... 28. pneustidee and Echinometride, with their subfamilies Dorocidaris papillata (Leske).....................0.. 31. and Gomera «+... ieee ee eee eee ee nee e ese eees 133. Ge ae are eae ORES Be 35. | Fame Behimidee: 1... eee eee eee eee ees 141 Stereocidaris ingolfiana n. sp.................0.. 000: 38. Subfam. Parechininze Ag ea ae ne eae #4t Porocidaris purpurata Wyv. Thomson................ 4l. Parechinus miliaris (Miill.)........................ 141 Table of the Cidarids occurring in the northern Atlantic UMN RU RL Sites Se aC Eel SLs ek ovine 142 SUIRUENE LOCIUCITGNEAN 5, conc eis ca cess ccsuisatvaen 42. Echinus elegans Dib. Kor....................... — Pier Meiinothatidee le. 43. SR rare ee ORE Soir, Be is od ait Diagnoses of the genera of the Fam. Echinothuride.... 62. a Se He a Be pee is SEE * Renae oe ae sae Phormosoma placenta Wyv. Thomson ........2.2.... 66. a ae eo i Bee ratte Wyv. Thomson i... essen. 7° | Fam. ROXOPMORMU Ete oe cs aes oe se ve. HUE 162. Brmosoma feneatratum (Wyv. Thomson)........ .... = Subfam. Strongylocentrotine........................ 162. Sperosoma Grimaldi Koehler....................... 75: Strongylocentrotus drobachiensis................. 162. Tromikosoma Koehleri n.g., n.sp...............-00-- 78. | Table of the Echinids of the Families Echinidze and Toxo- Table of the Echinothurids occurring in the northern pneustidz occurring in the northern Atlantic and the RAR en ONE gah pein en Wnt a ve ase 80. Meditemiionie ogee Mess clr a ye 165. Rai, “PERIOD CULES oh es eee hen ng ake. BRT ATION K eee ee risen miata ea oa” kine etn 169 Hypsiechinus coronatus n.g., m.sp.............. ..-. Port BibNODApinyes oe rar ie ain oes eee ee, 183. 166041 /4IBRARD ‘ f OF THE , UNIVERSITY OF SSALIFORNIE Echinoidea. I. By Th. Mortensen. he present work forms the first part of a planned revision comprising all the arctic Echinoderms, 2% excepting the Holothurioidea. The basis of the work is formed by the rich material of the Ingolf-Expedition together with the large collections of arctic Echinoderms found at our Zoological Museum from earlier expeditions. To the arctic fauna all the species are referred which are found in the Norwegian Sea, the Greenland Sea, the Denmark Strait, and at the coast of West-Greenland, as also in the White Sea and the Polar Sea with the Bering Strait. Of forms that are only found south of the large ridge between Greenland and Iceland, and between Iceland and the Farée Islands, only such as have been taken by the Ingolf-Expedition, have been included in the work. During the examination of the material the absolute necessity of taking into consideration also other more or less nearly related forms soon made itself felt. By and by I became aware of the fact that the classification hitherto used with regard to the families treated of here, was quite erroneous, and so I have sought to include into the examination as many forms as possible in order to be able to give the new classification that had to be made, so broad a base as possible. Inspector G. M. R. Levinsen placed the whole rich collection of Echinoids of the museum at my disposal with the greatest readiness; but as far from all species and genera are represented in this collection, I have applied to several foreign naturalists, and have everywhere been met with the most obliging kindness and friendliness, so that I have been enabled to examine almost all known genera and species com- prised in the groups treated of here. The following gentlemen have sent me Echinoids on loan or in exchange: Dr: Appelléf _(the Museum of Bergen), Prof. F. Jeffr. Bell (British Museum), Prof. E. v. Beneden (Liége), Prof. Collett (Christiania), Prof. Déderlein (Strassburg), Conservator J. Grieg (the Museum of Bergen), Prof. Koehler (Lyons), Prof. P. de Loriol (Genéve), Prof. E. v. Marenzeller (Vienna), Gel.rath, Prof. E. v. Martens (Berlin), Geh.rath, Prof. K. Mobius (Berlin), Prof. Monticelli (Naples), Prof. P. Pallary (Oran), Prof. G. Pfeffer (Hamburg), Prof. R. Rathbun (U.S. National Museum), Prof. The Ingolf-Expedition. IV. 1. I 2 ECHINOIDEA. I. d’Arcy Thompson (Dundee). By this present I beg to offer my sincerest thanks to all these gentlemen. Finally I had occasion for a short stay at the British Museum in August 1901. By the genial friendliness of Prof. Bell I was enabled to examine a great many forms, especially original specimens from the Challenger-Expedition. It will appear throughout my work, that this stay has been of material importance to me, and my best thanks are due to Prof. Bell for his liberality. Still I have to thank Dr. F. A. Bather (British Museum) for his excellent assistance in several literary questions. Copenhagen, January Igo2. The Author. «Loin d’étre nuisible aux vrais progrés de la science, cette multiplication des genres, lorsqu’ils sont établis sur des caractéres précis, ne saurait avoir d’autre effet que de rapprocher de plus en plus les espéces, que leurs caractéres naturels lient le plus étroitement. Cest la le grand avantage des petits genres, et cet avantage est surtout sensible dans les familles, dont toutes les espéces se ressemblent par leur aspect extérieur et par l'ensemble de leurs caractéres.» L. Agassiz. On generic and specific Characters in the Echinoids. Everybody who has studied Echinoids, will have felt a considerable difficulty in recognising many of the genera, at all events of the regular Echinoids. Such was, at any rate, my case at the commen- cement of my researches. I studied the excellent collection of these animals found in our museum, and found it to be more and more hopeless. A great many genera were exhibited, as: Lchznus, Psammechinus, Toxopneustes, Hipponoé, Boletia, Psilechinus, Lytechinus, Loxechinus, etc.; but it seemed to be impossible to discover the characters on which they were established, whether the naked tests, or specimens that had kept the spines, were examined. And the literature did not contribute very much to clear up the question. To be sure, some of these names (— as it will be seen, partly unjustly —) appeared to be synonyms; but nevertheless the other genera were not much _ better characterized. We learned through long descriptions that the spines were thick or thin, few and scat- tered, or many and closely packed; that the tubercles might be small or large, and that they might be placed in more or less regular series, etc. — altogether things easily enough seen, but so relative, that it was impossible to get any any firm hold. It was almost enough to drive one to despair. Still a faint hope was left. Might not the difficulty be in the literature, and the animals them- selves in reality be less intractable? A profound and careful attempt at penetrating into the mysteries of the relationship of the Echinoids was planned, and the plan was the simple, but clear one: to let literature alone for the present, while the animals were studied thoroughly. Everything had to be examined that might in any way be supposed to show systematic characters: the test, the spines, the tube-feet, the pedicellariz, the spicules, the spheeridie, etc. The beginning was to be made with the Zchinus-species. This choice seemed to be the best one, as these species have hitherto been especially notorious for their difficulty, and a very rich material of them is found in the museum of Copenhagen. The result was excellent. The animals proved to be very tractable, the species to be very well characterized (with a few exceptions). The difficulties arise from the literature containing numberless bad descriptions. And what a confusion is reigning in the literature with regard to the names. Almost every species must drag along with it a lot of synonyms, not only specific syno- nyms, but also generic ones. Several species have by and by been referred to a whole series of different genera, to end at last as a separate genus, as badly characterized as most of the other genera. To name only one instance: The genuine Psammechinus-species: variegatus (Lamk.) and semituberculatus (Val.) have by and by been referred to the following genera: Echinus, Lytechinus, Schizechinus, Toxo- pneustes, but only rarely, in recent times not at all, to the genus to which they decidedly belong. On the other hand the following extraneous species have been referred to Psammechinus: Echinus norvegicus, magellanicus, miliaris, microtuberculatus, angulosus, Strongylocentrotus Gaimardt, intermedius, r* 4 ECHINOIDEA. I. Spherechinus pulcherrimus, Evechinus chloroticus, Echinostrephus molare. — This instance may be taken as a significant illustration of the generic descriptions. Or should it be necessary also to recall the genera of Cidarids? That under such circumstances erroneous determinations have been frequent, is not to be wondered at. I have had occasion to substantiate several (far too many!) cases, and such cases too where the greatest authorities have been responsible for the determination. We ought therefore to be very cautious in using the existing statements. with regard to the geographical distribution of these forms. The characters that have hitherto chiefly been used for the distinguishing between the genera and species, are the following: the pores, the spines, the tubercles, the mouth-slits, the lining of the buccal membrane with larger or smaller plates, and the calycinal area. All these structures may give excellent characters, and, of course, they are always to be taken into consideration. But most- frequently they are so relative, that it is exceedingly difficult or impossible by means of these structures to decide whether a specimen in hand belongs to one species or another. Such is especially the case when the question is of the position of the tubercles; it may be simply irritating to read the descriptions of these in different species that are to be compared, and often the result falls very short of the exertion to get a clear view of the descriptions. To this may be added that the number, size, and position of the tubercles vary very much with age. With regard to the pores, their number and mutual position is no absolutely reliable character either. That in species with many pairs of pores their number increases with age is a well-known fact. The young Strongy- locentrotus drobachiensis has only three pairs of pores (Lovén 250); «Strongylocentrotus» lividus has only 3 pairs of pores in the lower ambulacral plates; Echinostrephus has 2—4 pairs of pores, oftenest 3 pairs ete. By these researches the pedicellarize and spicules proved to be of very great systematic importance; they give the most excellent characters we may want. To be sure, this fact is no new discovery. It has long been known that these organs and structures were more or less differently constructed in the different species and genera; much has been written about this fact, and a great many figures have been published. But nevertheless the fact has never been fully utilised. The history of the pedicellariz is highly interesting; scarcely many zoological objects will be able to vie with these organs with regard to the number of interpretations. From parasites to embryos, and even to vertebrates, and back again to parasites their history passes, until they are generally acknowledged to be what they really are: organs forming integral parts of the animal. v. Uexkiill has given an excellent account of their history (406), and so there is no reason to give it here again. I shall only here note a few less important treatises, not mentioned by v. Uexkiill, viz. by Duncan (130), Groom (175), and Stewart (381). A little note by Troschel (Verhandl. d. natur- hist. Vereins d. preuss. Rheinl. u. Westphalen. 1870 p. 137) is also to be mentioned for the sake of completeness; it contains nothing new. The histological structure of the pedicellarize has of late years been very carefully studied, especially by Foettinger (155), Hamann (184), Sladen (366), Prouho (327), and v. Uexkiill (406). The most interesting ones in this respect are the globiferous pedicellariz, which have proved to be ECHINOIDEA. I. 5 poison-apparatus of a very peculiar and complicated structure with sensitive cilia, poison-glands ete. Only a single point seems hitherto not to have been fully understood, viz. how the poison gland opens through the large tooth at the end of each of the three valves forming the skeleton of the head of the pedicellaria. Perrier’) thinks that in some there is a large -Echinoids (395) and Déderleins as excellent work on the Cidarids (116)) been treated as being of importance in the systematic works; generally they have only been mentioned as a matter of small importance beside the description proper, and often no attention at all has been paid to them. Rarely all the different forms of pedicellarize in a species are described, and still less in all species of the same genus; of one species an ophicephalous and a tridentate pedicellaria is figured, of another a valve of a globiferous one, of a third perhaps none at all, etc. In this way, of course, we shall never get a clear understanding of the systematic characters which may be found in these small organs. The pedicellarie in effect give absolutely excellent systematic characters, sometimes only specific characters, sometimes also generic ones. The use of the pedicellariz in classification is attended with great advantages; they do not change their form with age, but are in the newly metamorphosed Echinoid of the same form as in the grown one, only somewhat smaller in the small specimens. It is therefore (oftenest) possible, by means of the pedicellariz, easily to determine quite small Echinoids with absolute certainty — at all events as to genus. Another advantage is that it is not necessary to remove the spines in order to get a view of the tubercles, the specimens have not to be destroyed for the sake of determination. It may, perhaps, seem unreasonable to lay so much stress, as is done here, on so minute fea- tures as the pedicellarie — to use them for the characterizing of as well species as genera and families. But when it proves to be a real fact that these minute features give excellent, constant characters, it may be taken to be reasonable to use them without regard to their being small or large. Surely any student of Echinoids will also feel it as a great advantage not to be obliged to be contented with all these relativities, as the length and number of the spines, the size of the tubercles, the form of the test etc. To all these things, of course, regard must always be paid, and so has also been done here, as far as the material has permitted. But the pedicellariz are, at least, as important. I can completely subscribe the expressions of Stewart (381 p. 912): «It seems to me ‘most desirable that minute, and even apparently trivial, features should be given in the descriptions of species, and that when this is more done, we may find affinities between forms, we should otherwise not suspect, and be enabled by the examination of even an ambulacral tube or pedicellaria etc. to determine a species without the denudation of portions of the corona, which is sometimes not desirable», The supposition by Stewart that by an examination of the pedicellarie etc. we might find a closer relation between forms not otherwise regarded as related, has been amply justified by these researches, even to so high a degree that the classification hitherto used proves to be quite a failure (with regard to the groups treated of here). A good proof of the correctness of the new classi- fication given here, which has been found especially by the examination of the pedicellariz, is found in the fact that forms with the same kind of pedicellarie also agree in other important respects. To 8 ECHINOIDEA. I. be sure, the material has not been sufficient for a thorough examination of all characters with regard to some groups (especially the Cidarids), but I think that from the results found elsewhere we shall be justified in-supposing that it will appear everywhere to be a fact that forms with the same kind of pedicellariz in reality belong to the same natural group. It is a serious drawback that the pedicellarize cannot be used in the classification of the fossil Echinoids. Groom (175), to be sure, has described the pedicellarize of Pelanechinus corallinus in a very well preserved state, and it will, no doubt, also be possible to find them in well-preserved speci- mens of other fossil Echinoids; of course, however, it will always be a rare thing — generally we have here to be content with the tests (and the spines). These structures also often give excellent characters, but they are far from being always reliable. The former great incertainty in the determi- nation of the recent forms of regular Echinoids (and I think it is not much better with regard to the irregular ones) may be taken to imply that there cannot be any great certainty in the classification of the fossil forms either. As is well known, no less than four different kinds of pedicellarie are found in an Zchinzws, viz. globiferous pedicellarize, tridentate, ophicephalous, and triphyllous ones. Of these forms the tri- phyllous and ophicephalous ones have only very little systematic importance; they are very much alike in almost all Echini. The tridentate ones give often excellent specific characters; the globiferous ones are generally very much alike in related species, but show very characteristic differences in the different genera. Especially the latter form shows many peculiarities. The structure of the blade is highly different; it may be open or shut, the margins having coalesced on the inside; there may be many or few teeth along the edge, placed symmetrically or unsymmetrically, or teeth may be quite wanting. On the other hand no forms are known with more than one end-tooth'). When Perrier (op. cit.) says that the globiferous pedicellariz in the Echinometrids end in two hooks, one placed a little above the other, this statement is not quite correct. There is also here only one end-tooth, with the men- tioned open canal on the upper side; the other one that is placed below the former, is a lateral tooth with no poison-canal, homologous with the lateral teeth of the pedicellarize in Achinus. Here thus is only one unpaired lateral tooth. In Spherechinus, Strongylocentrotus etc. no lateral teeth are found at all, only a little obliquity is seen towards the end of the blade, a little process on one side, perhaps a reminiscence of the unpaired lateral tooth in the Echinometrids. — Some (Strongylocentrotus) have a long, muscular neck between the stalk and the head; in most forms the head is placed directly on the end of the stalk. Even the structure of the stalk is very different, in some forms it is a per- forated tube, in others some thin calcareous threads, irregularly connected by short cross-beams, or it may even be a single thin calcareous thread. Some forms have large mucous glands on the stalk. In the Cidarids the stalk is very peculiar, with an upper thin part and a lower thick one; at the transition between the two parts a limb of projecting calcareous ridges is often seen. The mentioned four different kinds of pedicellaricze are found in the old families Achinide and Echinometrade. In the Echinothurids globiferous pedicellarie are only found in a single genus (Hapalosoma); they are highly peculiar (Pl. XIII, Figs. 20, 24, 25), obviously very primitive. The calcareous skeleton consists of three simple rods lying between the three (mucous?) glands, each 1) Comp. however, the description of the globiferous pedicellarize in Stomopneustes. ECHINOIDEA. I. 9 of which ends in a fine pore at the end. The rods reach only half-way, the whole thing is coalesced to the very point; there are no muscles between the basal parts of the valves. In another genus (Areosoma) a singular kind of pedicellariee are found, the tetradactyle, with four peculiar, very ele- gantly formed valves. Also in other Echinoids a four-valved pedicellaria may now and then be found, but only as an abnormity. Ophicephalous pedicellarie') are among the Echinothuride found in only a single genus (7vomikosoma); on the other hand, triphyllous and tridentate pedicellariz are found in all of them, and especially the tridentate ones show a great variety of forms, and are of great systematic importance. In the Cidarids are found tridentate pedicellariz, and another kind occuring in a large and a small form, of substantially the same structure. They seem to be poison-apparatus as the globiferous pedicellariz of the Echinidee; but they are of a quite different structure, the gland being here placed inside the blade, quite surrounded by the calcareous skeleton, while in the Echinide it is situated on the outside of the blade. On the inside of the blade, somewhat below the point, there is a larger or smaller opening («the mouth») in the calcareous skeleton, filled with large cells, richly provided with cilia (sensitive hairs?). The efferent duct of the secretion of the gland passes up through the end-tooth, and opens on its surface. How these structures are arranged in forms with no end-tooth is unknown. The inner opening is of great systematic importance, while the glandular opening itself scarcely is of any importance in this respect. Perrier (op. cit.) gives these pedicellarize a special name «Pedicellaires armées». After the discovery of the above described form of globiferous pedicellarize in the Echinothurids?) there seems to be sufficient reason to take these pedicellariz in the Cidarids to be homologous with the globiferous pedicellariz of the Echinoids, as has also been done by Stewart (379) and Prouho (327), so that there is no cause to keep the name given to them by Perrier. There is still less reason to keep the name «Ped. inermes» for the tridentate pedicellarize of the Cidarids; there can be no doubt but that they correspond to the tridentate pedicellariz of the other Echinoids (Prouho (327), Koehler (217). Hamann (184) regards the small pedicellarize as «a sub- species of the tridactylous ones». Now it has to be admitted that sometimes it may be rather difficult to distinguish between these latter and small tridentate pedicellarie; but generally they are very easily recognised, and there is no doubt that, with regard to structure, they resemble very much the large globiferous pedicellariz. Where no pronounced difference is found between large and small pedicellariz, it may in fact be impossible to decide, whether a certain specimen is to be regarded as a large or as a small form. There seems to be no reason to give a special name to the small pedicellariz; in the present work they will the mentioned as «small globiferous pedicellarize».— Ophicephalous and triphyl- lous pedicellariz are not found in the Cidarids. O. F. Miiller3) has originally given names to the pedicellariz, viz. Pedicellaria globifera, triphylla, and tridens. These names have not been generally accepted, the reason being especially that Valentin in his classical monograph on the anatomy of Zchimus has used other appellations: Pedicellaire gemmiforme, tridactyle, and ophicephale; these names have become the common ones. Sladen (366) justly maintains that it is incorrect to use these latter names. The figures of Miiller 1) What has hitherto been regarded as ophicephalous pedicellarie in the Echinothuride, are in reality triphyllous ones. 2) Also the globiferous pedicellarize in Stomopneustes seem to form a peculiar type. They have no end-tooth, and there seems to be no poison gland on the outside of the blade. 3) Zoologia danica, 1788. pag. 16. Tab. XVI. The Ingolf-Expedition. IV. IL 2 10 ECHINOIDEA. I. are perfectly recognisable, and therefore his names ought to be restored to their rights. The name of P. triphylla of Miiller, however, no doubt includes as well ophicephalous pedicellariz as triphyllous ones. This name must then be kept for the small form the valves of which resemble clover-leaves, while Valentin’s name P. ophicephale is kept for the form described by him under this name, — Hamann (184) uses the name «Globiferen» especially of the pedicellariz where the mucous glands on the stalk have been so highly developed, that the head has become rudimentary or is even quite wanting. Thus they, as is also admitted by Hamann himself, are not a peculiar kind of organs, but only transformed pedicellariz; it may, perhaps, be as well to have a special name for these pedi- cellarie, but the name of «Globiferze» cannot be restricted to them, as has also been observed by Duncan (130). It is, in reality, contrary to all common practice not to use the names of Miiller. The reason for keeping Valentin’s names given by Geddes and Beddard (163): «both on account of their general acceptance and because they were the first names applied to pedicellariz after the determination of their real nature; Miiller’s nomenclature refers to pedicellarie as a genus of para- sitic animals», is not sufficient for a disregarding of the common rules of priority. Accordingly the namnes that ought to be used, are the following: | Pedic. gemmiforme Valentin, Perrier. Globiferous pedicellariza — Pedicellaria globifera Miller = ,; P.armé Perrier (in the Cidarids). | «Globiferen» Hamann. : : { P.tridactyle Valentin ete. Tridentate — — aes tridens — = ; Te whee \ P.inerme Perrier (in the Cidarids). Obhi a heocebhale Val is P. triphylla Miiller pro parte. me pa vel hiocepha entin = I icy su dh ca : P. buccale Valentin, Hamann. Triphyllous — a — triphylla Miiller = P.trifolié Perrier. To facilitate the understanding of the descriptions in the following, figures are annexed showing a single valve of each of the four kinds of pedicellariz together with the names used for the separate parts. To be able to study the pedicellaric, especially the calcareous skeleton, which is of particular importance for the classification, they must necessarily be treated carefully. On being boiled in a not too strong solution of potash the separate pieces of the skeleton may easily be isolated, and no very great technical skill is necessary to be able to make preparations in Canada balsam of these pieces. (They cannot be kept in glycerine, as it resolves the lime). Accordingly I can in no way subscribe to the opinion of Pomel that the pedicellarize only with difficulty can be used for the classification, because «leur ténuité en rend l'étude peu pratique» (324 p. 13). Also the spicules yield good systematic characters, even if they are not, in this respect, equal to | the pedicellariz. They only rarely yield specific characters, and are oftenest very similar in the sepa- tate genera of the same family, but they may yield excellent family characters. They may be of a simple C-shape («, none of which are mentioned) as the only recent representative; des Goniocidariens with the recent genera Gontocidaris and Dorocidaris; and les Rhabdocidariens with the genera Phyllacanthus (with the subgenus Stephanocidaris), Letocidaris and Porocidaris. The genus Schleinitzia Studer is supposed to be a Rhabdocidaris, consequently also to belong to this subfamily. Duncan (132) only admits the genus Czdaris with the subgenus Gonzocidaris; the other earlier genera are only classed as «divisions». De Loriol (245) comprises a great number of species under the name of Rhabdocidaris Desor; but he owns (p.7) that «au fond, toutes les tentatives, qui ont été faites pour demembrer le grand genre Cidaris, n’ont pas été heureuses; on trouvera toujours tant de passages entre les espéces, en apparence les plus distinctes, qu'il est douteux pour moi, s'il est vraiment nécessaire de diviser ce genre admirable, qui apparait dés la fin de Pére paléozoique et traverse dés lors tous les étages, sans manquer dans aucune, pour se retrouver enfin dans les mers actuelles sans avoir modifié aucun de ses caractéres». The most important contribution to the classification of the Cidarids has been given by Déderlein in his above quoted, large and excellent work «Die japanischen Seeigel» where he attempts to group as well the recent forms as the fossil ones according to their real relation. With regard to the recent forms the following genera are retained: Dorocidaris, Stereocidaris (known until then only as fossil from the cretaceous period), Hucitdaris, Leiocidaris, Porocidaris, and Goniocidaris. But neither is the limitation by Déderlein of these genera satisfactory; above all it holds good with regard to his genera as well as with regard to those of the other authors that nobody is able to recognise them with certainty by the diagnoses given, — when upon the whole diagnoses are given. After all it is a ECHINOIDEA. I. 13 matter of judgment, to which genus one species or another is to be referred, and most of the species more frequently mentioned have also by and by been referred to almost all the different genera. So far it is very consistently done by Duncan and Bell (73) quite to strike out all these undistinguish- able genera, and only retain the old genus Czdaris; but then on the other hand this way of proceed- ing means quite to abandon the pursuit. The reason why the result of the earlier attempts at classification has been so meagre, has to be sought in the characters used. The most important ones have been, whether the two pores of each ambulacral plate are connected by a groove or not, and whether the tubercles are crenulated or not. Further the spines, the number of plates, the breadth of the ambulacral area, and upon the whole the structure of the test have been considered of great importance. All these characters, however, are insufficient or even unreliable. As has been pointed out by both Déderlein and Dun- can, it is often impossible to decide, whether the pores are or are not connected by a groove. The crenulation is a very variable character; crenulated tubercles may be found in some individuals belong- ing to species normally without crenulation. The structure of the test, the tubercles, the number of plates etc. are very much dependent on the age of the animal. All these characters, says Duncan, are «of no physiological importance whatever»; «any classification in which these characters are used is artificial». On the other hand he thinks that «the number of interradial plates (is) of physiological importance; and there is a great temptation to consider typical Cidarids as having but a few, say not more than seven, in a vertical row» (132 p.30). This character seems to be at least as gratuitous, as the others criticised by Duncan are relative ones; neither seems the result of his systematic researches in any way to show that he has found here a systematic character of any great importance. Among the characters hitherto used in the classification, the spines seem to be one of the most reliable. They show a great richness of forms, but are at the same time of a rather constant form in the separate species. Also their microscopic structure differs to a high degree, and here, perhaps, we might find good generic characters. There are in the literature not a few examinations of the struc- ture of the spines in the Cidarids. Stewart"), Bell (57), and Agassiz (Revision of Echini and Chall. Ech.) have figured transverse sections of the spines of different species; but especially H. W. Mackintosh has rendered great services to the question by his excellent researches on the struc- ture of the Echinoid-spines (264—6s). The spines of the Cidarids differ from those of-the other Echi- noids by having a compact outer layer («». According to Agassiz (Revision of Echini) C. annulata A. Ag. is = C. tribulotdes Lamk., and C.annulata Gray = Phyllacanthus annulifera A. Ag. ‘The figured pedicellaria, however, cannot belong to any of those species, although Agassiz (Revision p.99) mentions the quoted work of Stewart under C. ¢ribuloides; it seems to be a Goniocidaris, but which species cannot be deter- mined. In (379) Stewart further gives a couple of excellent figures of globiferous pedicellarize in Doro- cidaris papillata. Also Wyville Thomson (395) gives excellent figures of the pedicellarie in Doro- cidaris papillata and Porocidaris purpurata. In «Revision of Echini» and in the «Challenger»-Echinoids (8) Agassiz figures pedicellariz of several Cidarids, but generally the figures are not good. Déder- lein (116), however, is the first author, who has tried to use the pedicellarie in a correct way in the classification of the Cidarids. He has studied the pedicellarie in a larger number of species, and ' ECHINOIDEA. I. 15 thinks that they often give excellent specific characters, but he was disappointed «in ihrer erhofften Verwendbarkeit zur Unterscheidung natiirlicher Gruppen innerhalb der Familie» (p.1). «Nur mit grosser Vorsicht diirfen Pedicellarien als systematische Merkmale bei den Cidariden beniitzt werden». The small pedicellarize are highly similar in almost all species, but they may vary very much in the separate individuals. (Only the form with a long terminal hook, occurring in Goniocidaris mikado and clypeata, is especially mentioned). The tridentate ones («léffelartige Form») are better, but they are also highly varying in the separate individuals. Most applicable for the classification is the thick- headed form, (the large, globiferous pedicellarize); it is highly constant in form and size, and shows many peculiarities, «die sehr wohl einzelne Arten, manchmal auch Gruppen charakterisiren kénnen». He also tries to group the species according to these peculiarities, without, however, attributing to them any great systematic importance, and therefore he does not mention the pedicellariz in his diagnoses of genera. The fact is that also this form of pedicellarize shows some variability, is some- times even quite wanting in some individuals, so that it is no quite reliable character. An extra- ordinary fact is «dass sehr ahnliche Formen dieser Pedicellarien bei Arten vorkommen k6nnen, die nach den iibrigen Charakteren sehr wenig Verwandtschaft mit einander bekunden» (C. metularta and verticillata). Wis final result is: «In vielen Fallen hat nun ohne Frage die Vergleichung der Pedicel- larien nicht geringen Werth fiir die Systematik; sie geben jedenfalls sehr brauchbare Charaktere zur Unterscheidung der Arten. — Zur Charakterisierung von grésseren Gruppen innerhalb der Familie finde ich aber Pedicellarien sehr wenig verwendbar» (p. 34). And so the last hope of finding good generic characters in the Cidarids seems to have vanished. Fortunately, however, my researches have given another result than that of Déderlein, viz. that the pedicellarie yield excellent generic characters, while they may only more rarely be used for distinguishing between the species. This seems to be irreconcilable with the above quoted statement of Déderlein that species not more nearly related, may have quite similar pedicellariz. As instances are only named Cydaris metularia and verticillata. Now it is quite correct that they have the same kind of pedicellarie; but then the question is whether the other characters, in which they differ, are sufficient to show that they cannot belong to the same genus. The most essential difference seems to be found in the spines, which are in C. verticillata provided with large thorns placed in circles far from each other, while in C. metularia the spines have the whole surface evenly set with homogeneous, small tubercles arranged in longitudinal series. Also with regard to the provision of the interambulacral plates with miliary tubercles a difference is found — they are almost naked in C. verticillata, closely covered in C. metularia.. As it has otherwise proved to be a fact that the characters taken from the structure of the test have been anything but good as generic characters, and as there seems to be nothing unnatural in the fact that spines as those in C. metu- laria and verticillata are found in species of the same genus, I cannot but regard the fact of the two species having the same kind of (very characteristic) pedicellarie as proving them to be nearly related, so that they will have to be regarded as not too closely allied species of the same genus. Besides there is another species of the same genus presenting considerably more resemblance to C. verticillata than the C. metularia mentioned by Déderlein. This is C. daculosa which is by Déderlein referred 16 ECHINOIDEA. I. to the same genus (Lezoczdaris) as C. verticillata, In this species the thorns are often placed in circles in a somewhat similar way as in C. verticillata. Especially the large globiferous pedicellariz are of importance in the classification, the blade and partly also the stalk offering a great variety of forms. Also the length of the stalk is very different; this fact, however, has to be used with great caution, at it is very varying. Déderlein seems to put no small weight upon it. Also the small globiferous pedicellarie are of rather great importance; more important, however, are the tridentate ones, which in a single genus, Porocidaris, are two-valved. In this genus (and perhaps in the genus /7/istocidaris) globiferous pedicellarize seem to be quite wanting; on the other hand tridentate pedicellariz are wanting in several other species — but perhaps not constantly. That the globiferous or tridentate pedicellarize may sometimes be want- ing, is mentioned by Déderlein as an objection to their being used in the classification. I cannot see, however, that this objection is sound; a corresponding fact would be, if we were to give up using the teeth of the mammals as systematic characters, because now one, now another kind, or even sometimes all of them are wanting. When we now look over the Cidarids, and place together the species with similarly constructed pedicellarize, we shall get a grouping rather differing from all hitherto given classifications. Dorocidaris papillata: the globiferous pedicellarize have a powerful hook at the point, above the large, somewhat lenghtened, not terminal opening; small pedicellariz of the same form; the triden- tate ones simple (Pl. IX, Figs. 7, 25). Quite similar pedicellarice are found in Dorocidaris Blakei A. Ag. (Pl. IX, Fig. 16), which is accordingly a genuine Dorocidaris. On the other hand the following species that have been referred to Dorocidaris: D. Bartletti Ag., bracteata Ag. and Retni Déderl. differ widely from this genus, and are moreover so different from each other that they must be referred to three different genera. D. Bartletti: the globiferous pedicellarie have a long powerful hook at the point. The opening is exceedingly small, as a fine pore, surrounded by small teeth; it is placed rather far from the point. (Pl. X, Figs. 23, 30). The stalk is most frequently provided with a limb of freely projecting calcareous ridges. The small pedicellariz are of the same structure, only the opening is larger; tridentate pedi- cellariz simple. There can be no doubt but that this species must form a separate genus; I propose the name of Tretocidaris'), To this genus must further be referred the two following new species, which I found in British Museum, both under the name of Dorocidaris papillata. Tretocidaris annulata n.sp. The globiferous pedicellarize differ somewhat from those of 7 Sartletti the inside of the blade being provided with some dentate transverse ridges and crests forming a coarse, irregular reticulation; at the upper end of the apophysis the margin of the blade is somewhat widened, highly fenestrated in a reticulate way, and bent a little outward (Pl. X, Figs. 22, 31). The stalk (Pl. IX, Fig. 4) and the other pedicellariz as in 7: Bartlett, The spines are finely annulated with brown rings, the upper spines have powerful thorns especially on the side turned up; they are tapering, about one time and a half as long as the diameter of the test; the actinal spines were wanting in the specimen, There is a rather deep, naked furrow along the median line of the interambulacral areas, and it continues between the plates outward to where the scrobicular areas join each other. 1) tpntds = bored. ECHINOIDEA. I. 17 There are 7 plates in the interambulacral areas. In the ambulacral area there is a little tubercle Oo alternately between each two primary tubercles eet as in Porocidaris purpurata. The colour of the test is redbrown, and therefore the white, naked furrow of the interambulacral areas is especially con- spicuous. —’ Locality: The West Indies (no nearer information). Should this species perhaps be Gray’s Cidaris annulata? Tretocidaris spinosa n.sp. The globiferous pedicellarie have no such reticulation as those of 7. annulata, and differ from those of 7: Bartletti by the sides forming an almost straight line from the basal surface to the opening. (Pl. X, Figs. 10, 11). The small globiferous pedicellarie as in the two other species (Pl. X, Fig. 16). On the stalk no distinct, freely projecting calcareous ridges are seen, only a marked swelling. (It is, however, possible that the limb of the stalk is found on other specimens; in the two other species it was not found either in all the large globiferous pedicellariz); tridentate pedicellariz were not found. The spines closely grooved, rather finely thorned, widened at the point, of the same length as the diameter of the test. The actinal spines smooth, not serrated, their points not widened. The small spines are strongly redbrown. There is a naked median line in the interambulacral area, but it is only little conspicuous. g plates in the interambulacral area; thus the large spines are somewhat more numerous than commonly, which gives to the animal a very charac- teristic appearance. The tubercles in the ambulacral areas as in Zi: annulata. Locality: St. Helena (no nearer information). «Dorocidaris» bracteata Ag. The globiferous pedicellarize much lengthened and narrow, with a powerful hook at the end, and a rather small, triangular opening a little below the point (Pl. X, Fig. 18); the small pedicellariz of the same structure, tridentate ones simple. This form of pedicel- lari is further found in «Phyllacanthus» annulifera (Lamk.), Pl. X, Fig.17, and Stephanocidaris bispi- nosa (Lamk.), and these species will have to be united into one genus, which must keep the name of Stephanocidaris. «Dorocidaris» Reini Déderl. The globiferous pedicellariz are of a very peculiar structure; the mouth is placed in the end of the blade, surrounded by well marked teeth on the margin which is bent a little outward. «Schnauzenahnlich vorragend» Déderlein says of the blade in this peculiar form of pedicellarize, and it really resembles a snout to some degree. On the stalk a limb of short thorns is found. The small pedicellarize are of a quite different structure, a well developed end-tooth being found here, and the large mouth situated below the point. This form of pedicellariz is found in a series of species, viz. Cidaris affinis (P|. IX, Figs.9, 22, 24) (which is in no way synonymous with Doro- cidaris papillata, as has been commonly supposed), ¢rzbeloides, galapagensis — and, I suppose, also in Dorocidaris panamensis Ag.; at all events this species, to judge by the figure, would seem to be most nearly related to Cidaris affinis and Reint; it is scarcely a Dorocidaris. The following species have pedicellariz of the same structure, but are distinguished by having a limb of long, freely projecting calcareous ridges on the stalk of the globiferous pedicellarie: Cidaris metularia, Thouarsit (according to Déderlein (116 p. 19) Cidaris Thouarsii has only a short limb on the stalk; the specimens examined by me have long limbs), verticillata and baculosa. Further has (according to the statement of Ddder- lein) Phyllacanthus imperialis the same kind of pedicellarie (whether a limb is found on the stalk The Ingolf-Expedition. IV. 1. 3 18 ECHINOIDEA. I. or not, is not mentioned; I have not been able to find any large globiferous pedicellariz in the few specimens I have examined), and the same, I suppose, holds also good with regard to Phyllacanthus dubia and parvispina Woods. Finally a similar form of giobiferous pedicellariz is found in Gomzo- cidaris florigera Ag. («Challenger»-Echinoids, Pl. I. Fig. 12) (Pl. X, Figs. 27, 29); in the latter there is no trace of a limb on the stalk. Do now all these species belong to one genus? — Surely not. We shall first have to separate Gontocidaris florigera. It has no trace of a limb on the stalk, the spines differ considerably from those of all the other mentioned species, and I suppose that a closer examination will show several other peculiarities. Déderlein (116) thinks it to be most nearly related to the species Gondocidaris clypeata and G. mikado described by him, which species are distinguished by the spines being provided with a peculiar flat widening at the base. Traces of such a widening are also found in G. florigera; but the pedicellarize of this species are so different from those of the two mentioned species that their being united into one genus is out of the question. It differs also from the genuine Gomiocedaris-species (G. tubaria etc.) by its pedicellarize; it must form a separate genus, for which I propose the name of Petalocidaris. There can scarcely be any doubt, however, that it is closely related to Gonzoczdaris. Next Phyllacanthus imperialis must form a separate genus. It has peculiar large tridentate pedicellarie, the blades of which are quite filled by a close net of meshes forming irregular longi- tudinal ridges closely set with small teeth (Pl. X Fig. 8); (the valve figured here, is from a smaller pedi- cellaria where only two longitudinal ridges are seen). The small pedicellarie have no end-tooth (Pl. IX. Fig. 6). The spines are peculiar, thick, with fine longitudinal strie. Together with this species Ph, dubia has no doubt to be placed — if upon the whole it can be kept as a separate species, of which I can have no decided opinion, as I have had no occasion to examine it. Also Phyllac. parvispina Woods must, to judge by the figure given by Woods (443), belong here; its spines resemble very much those of Ph. imperialis though Woods states them to be «entirely different from any described species». Also Ramsay (331 p.45) says of this species that on the Australian south-coast it is the «representative of P. dubia of the North Coast». — This genus, no doubt, must keep Brandt’s old name of Phydla- canthus. Brandt?) gives Czdarites dubia as the type of the section «Phyllacanthus», and observes that to this will have to be added C. zmperialis, hystrix, geranioides, and pistiliaris. The three latter can in no way be classed together with the two former; these two must keep the name of Phyllacanthus. Desor in his «Synopsis des Echinides fossiles» (1855) establishes the genus Ledocédaris (p. 48), and as the type of the genus he gives Czdaris imperialis. — ‘Thus there will be no use for the name of Leio- cidarts, it will only be a synonym of Phyllacanthus. — It will also be necessary to say some words of the much used name of Rhabdocidaris by the present occasion. The genus has been established by Desor (op. cit. p. 39) for fossil species; in a note is added: «Parmi les espéces vivantes on pourrait reporter 4 ce genre les C7zdaris tribuloides et C. impertalis, si leurs tubercules n’étaient pas complétement lisses». De Loriol (245) has later enlarged this genus to comprise: 1) The fossil species of the genus Rhabdocidaris sensu stricto, 2) the Rhabdocidaris-species with smooth tubercles, 3) the species of Letocidaris Desor and Dames (emend.), 4) the recent species of the genus Phyllacanthus Brandt, 5) the genus Stephanocidaris Ag., and 6) the genus Schleinitzia Studer. «Ainsi constitué, le genre Rhabdocidaris 1) Prodromus descriptionis animalium ab. H. Mertensio in orbis terrarum circumnavigatione observatorum. 1825 p. 68. ECHINOIDEA. I. 19 x groupera naturellement un assez grand nombre d’espéces vivantes et fossiles et me parait utile a conserver>. The advantage of such a «genus», however, seems to me to be rather illusory; with the limitation given by de Loriol Rhabdocidaris becomes still more heterogeneous than Phyllacanthus, as it is limited by Agassiz in «Revision». As the genus has originally only been used of fossil species, it is quite impossible to decide whether some of the recent forms really belong to it; by the tests and the spines alone the genera cannot at present be recognised with certainty, and no pedicel- larice of fossil species are known. Accordingly the name of Rhabdocidaris is not to be used for any recent Cidarid. On the other hand the other species with terminal opening on the globiferous pedicellariz and limb on the stalk seem to form a natural group; the shortness or length of the limb can scarcely be used as a character for the subdivision of the group. Possibly C.a/fnis and Rezni (and perhaps fanamensis) will prove to form a special group — their spines seem to differ somewhat from the other mentioned species; but this can only be decided by more thorough examinations. For the present all these species: Cidaris affinis, Reint, (panamensts ?), tribuloides, galapagensis, metularia, Thou- arsit, verticillata, and baculosa*) must form one genus, which must keep the old name of Czdaris, Linné’s «Echinus Cidaris», as has been proved by Lovén (252), being Czdarts baculosa Lamk. The name of Eucidaris Pomel, which has of late often been used for species of this group, cannot correctly be used. Pomel (324) enumerates as types of this genus some fossil forms (moriert etc.) from the trias, and «trois espéces vivantes», but he does not mention which species he means, and the fact is here, as in Rhabdocidaris, that it is quite impossible to decide whether any of the recent species belong to the same genus as the mentioned fossil ones. Besides the species mentioned here, Déderlein still enumerates «Lezocidaris» annulifera Lam. as belonging to those species, the globiferous pedicellarie of which have terminal opening and limb on the stalk; here C.annulifera is referred to the genus Stephanocidaris which has a quite different form of pedicellariz (see above) — a contradiction which can only have its origin from a difference in the interpretation of the species C.annulifera Lamk. This species together with C. daculosa Lam. have caused and still cause many difficulties to the systematists. Lamarck?) in his diagnosis of C.annulifera says: «spinis majoribus longis, tereti-subulatis, asperulatis, albo purpureoque annulatis», and in his diagnosis of C. daculosa: «spinis majoribus subteretibus, tuberculato-asperis, apice truncatis, collo guttatis»; according to this Agassiz («Revision of Echini» p. 389) states as the only certain character of the highly varying C. daculosa «the spotted base of the shaft of the spine below the milled ring, which is of a light reddish or reddish-yellow ground-color, with deep violet spots marked extremely distinctly upon the fine longitudinal striation». Loriol (243) later describes and figures a Cidarid by the name of C. annulifera Lamk.; he has had a radiole of the type-specimen of this species for comparison, and has found it completely corresponding to those of the specimen described by him. These spines have «leur base couverte sur une longueur plus ou moins grande de petites taches pourpres, formant des lignes et entremélées de petits points» — the character especially particular of C. baculosa! Thus, somehow or other, an error must have slipped in, and I think it most likely that 1) If C. pistillaris Lamk. be a good species, it must also be referred here. 2) Histoire naturelle des animaux sans vertébres. II. Ed. 1840. T. III. p. 380. 20 ECHINOIDEA. |. the spine, which Loriol has got from Paris, has really been of C. daculosa — such a changing of loose spines in a museum is not absolutely inconceivable. The C. Laitkenz described by Loriol in the same work, seems rather to be the real C. annulifera, which must then be very nearly related to C. bispinosa, perhaps identical with it. Bedford (35 p.274) also regards C. Liitkent as synonymous with C.annuilifera Lamk., but at the same time he seems to think it to be identical with Loriol’s C.annulifera, which cannot be correct. D6dderlein, who has examined a specimen of Loriol’s C. annulifera, finds this species to be highly consistent with C. daculosa. «Einen Unterschied zwischen den beiden Arten kann ich nur in der Farbung der Primarstacheln finden; denn selbst die Form der Primarstacheln kann bei bestimmten Individuen beider Arten identisch sein. — Nur die Farbung des Schaftes ist verschieden, indem ZL. annulifera Querbinden zeigt, die Z. daculosa fehlen; die eigenthiim- liche und auffallende Tiipfelung des Stachelhalses dagegen, die sonst nirgends zu beobachten ist, findet sich bei beiden Arten in gleicher Weise. Nachdem aber eine Autoritét wie Al. Agassiz auf Grund eines reichlichen Materials die Frage nach der méglichen Identitat der beiden Arten tiberhaupt nicht aufwirft, kann ich es nicht wagen bei meinem ganz unzulanglichen Materiale eine solche zu behaupten. Ich kann hier nur constatieren, dass die oben beschriebene jugendliche Z. annulifera nach ihren sdmmtlichen Charakteren, abgesehen nur von der Farbung der Stacheln, unbedingt als ein junges Exemplar von Z. daculosa gelten kénnte» (116 p.24). Prominence is also given to the fact that the pedicellariz are quite identical. In another work (245) Loriol gives a thorough description and figures of C. baculosa, but its resemblance with the C. annulifera before described by him, is not at all mentioned. Thus the fact seems to be: either Loriol’s C. annulifera is really this species — and then C. daculosa Lamk. and C. annulifera are synonyms — or it has, on account of some error or other, been wrongly determined —- and then C. annulifera is most nearly related to C. dispinosa Lamk. (perhaps synonymous with it). The latter is the more probable. An examination of the type- specimens, especially their pedicellarize, will easily decide this question. To be sure, Perrier has figured pedicellariz of these two species, but unfortunately only so little exactly and minutely that he has not at all contributed to the clearing up of the question, especially as of one species he has only figured a globiferous pedicellaria, of the other only a tridentate one. According to Déderlein (116 p. 25) Schleinitzia crenularis Studer is very nearly related to C. baculosa; Studer’s figures (386) agree also partly with it, the separately figured spines having all the characteristic spots on the neck. On the figure of the whole animal these spots, however, are not found, and as, according to informations I have received from both Geh.rath, Prof. E. v. Martens and Prof. Déderlein, spines of at least two different species are found in the glass together with the type-specimen (v. Martens has sent me some of the spines), the safest plan will be to say nothing definite of this species, till the pedicellariz of the type-specimen have been examined. Studer only figures the small form of the globiferous pedicellariz. Among the species referred to Phyllacanthus by Agassiz, still one has not been mentioned, viz. Ph. gigantea Ag. It differs from all other known Cidarids by its peculiar spines, as well primary as secondary ones; also its pedicellariz are peculiar. The large globiferous ones (Pl. X, Figs. 15, 19) have a large cordate opening the lower limit of which is formed like a highly protruding lower lip; the opening reaches to the very point, and no end-tooth is found. No limb on the stalk. The ECHINOIDEA. I. at small pedicellariz are of a somewhat different form (Pl. X, Fig. 26), and have a more or less powerful end-tooth. Tridentate pedicellariee about as in Dorocidaris papillata, only with the edge somewhat more dentate. Spicules of the common form. It is obvious that this species cannot] remain in the genus Phyllacanthus as here limited, or be referred to any of the mentioned genera; it must form a separate genus and retain the name of Chondrocidaris, originally given to it by Agassiz’). The splendid Cidaris curvatispinis described by Bell (74), is in its whole appearance so unlike all other Cidarids that it is beforehand to be supposed that it represents a separate genus. The examination of its pedicellariz also confirms this supposition. The globiferous pedicellariz (Pl. VIII, Fig. 37) have no end-tooth; the opening is large, reaching to the point, but its lower limit is remark- ably irregular — the figured one is one of the most regular; sometimes there seems to be no definite limit at all, the calcareous covering running out into irregular dents, as if it was broken off (which is, however, quite out of the question, as the pedicellaria was otherwise quite undamaged). The small pedicellarize are of the same structure, the only difference being that the lower limit of the opening is here often a rather regular transverse line. (The possibility that the described and figured pedicel- laria is really, in spite of its size, only the small form of the globiferous pedicellariz, is not excluded; but on the only known specimen, which by the kindness of Prof. Bell I had the opportunity to examine in British Museum, there seemed to be found no other kind of globiferous pedicellariz). The tridentate pedicellarize (Pl. X, Fig.9) are very peculiar, with some large, dentate crests of thin calcareous lamelle longitudinally in the blade. No limb on the stalk. The spicules of the common form. For this species I propose the generic name of Acanthocidaris. The genus Porocidaris is established by Desor (op. cit. p. 46) for some fossil Cidarids, especially distinguished by a circle of pores in the scrobicular area; to this genus Wyville Thomson (394—95) referred a Cidarid from «Porcupine» under the name of Porocidaris purpurata. Whether it really belongs to this genus cannot be decided, till the pedicellariz of the fossil species referred to it by Desor, become known. But to judge by what is hitherto known the species may well seem to be a Porocidaris, and for the present there seems to be no reason to reject this commonly used name, and P. purpurata W.Th. may then be put down as the type of the genus. Peculiarities of this genus are then the depressions in the scrobicular area (not pores as in the fossil species), the highly developed neck of the spines, the highly serrate edge of the actinal radioles?). But the most particular feature are the pedicellariz. Only one form is found which must be referred to the tridentate ones; they are two-valved, highly compressed, and exceedingly large and conspicuous. The spicules of the common form. To Porocidaris have later been referred the following species: P. elegans Ag., Sharreri Ag., Millert Ag., Cobost Ag., gracilis Sladen, gracilis Doderl., misakiensis Yoshiwara, and zucerta Koehler. Of these species P. gracilis Sladen is, no doubt, only a young P. purpurata, and this name is then to be omitted as a synonym. P. elegans (one of the type-specimens («Challenger» St.164a) examined in British Museum): the tridentate pedicellariz are widely different from those of P. purpurata. There t) List of Echinoderms sent to different Institutions in exchange for other specimens, with annotations. Bull. Mus. Comp. Zool. I. 1863. 2) Especially the latter fact is often mentioned as characteristic of the genus; this, however, is not at all reliable, as sufficiently shown by these researches. 22 ECHINOIDEA. I. are two forms, a larger and a smaller, both three-valved. In the larger form the blade is filled by an exceedingly rich net of meshes, in which the holes are rather distinctly arranged serially, and radiate in a fanshaped way from the upper end of the apophysis; this net is covered with numerous small thorns, especially towards the point. Also the upper edge of the apophysis is very broad and full of holes. (The figures in the «Challenger»-Echinoids, Pl. XLIV, 6—14, are not very good, especially not figs. 6 and 11, where it is not seen at all that the whole mass filling the blade, is really a net of meshes with innumerable larger and smaller holes). In the other, smaller form the apophysis has the common structure; the blade is highly compressed, deep, and filled with an irregular net of meshes where the holes are not at all serially arranged. Transitions are however found between the two forms, so that they cannot be said to be two distinct kinds. When Agassiz (Chall. Ech. p. 43) says of «the large-headed, shortstemmed pedicellarize» that they are «very similar to those of Dorocidaris», this is only so far correct, as tridentate pedicellariz, of course, always in some degree resemble each other; in the finer structure the large tridentate pedicellarize of this species are especially widely’ different from those of D. papillata. The small ones are much more similar. — Agassiz (l.c.) mentions one more form of pedicellarize, «shortstemmed globular abactinal pedicellariz» (Pl. XLIV, 10); they are, as I have been able to substantiate, only developmental forms of the large tridentate pedicellarie. I am a little in doubt whether globiferous pedicellariz are found. In my preparation of isolated skeleton- pieces of pedicellarize of this species is seen one valve of a small globiferous pedicellaria, which is very peculiar, with two large teeth at the point, and a rather small opening surrounded by well developed teeth (Pl. IX, Fig.2). As, however, only one such valve is found, it may be thought to have come in by chance; in this case it must be abnormal, as no other Cidarid examined by me, is possessed of such pedicellarie. For the present this must be left undecided. — It is obvious that this species has no relation with P. purpurata, and as it shows no nearer relation to any other known species, it must form a separate genus, for which I propose the name of Histocidaris. P. Sharrert: Agassiz (9) unfortunately gives no details as to the pedicellariee, and from the figure (op. cit. Pl. III) it cannot be decided whether it is a genuine Porocidaris. There seems to be no highly developed neck on the spines (in the text nothing is said of this feature); the pedicellariz might well look like those of P. purpurata, but a close examination will be necessary for the decision. By the kindness of Prof. Rathbun I have from U.S. National Museum received a specimen deter- mined as P. Sharreri («Albatross» 1885. St. 2415); it proved to be the new species Stercocidaris ingolfiana described hereafter; it has no relation to P Sharrert. Further I have in British Museum seen a specimen determined as P. Sharreri, from U.S. Fish Commission («Albatross» 1885. St. 2345). Neither seems this specimen to be identical with the real, figured P. Sharrerd, at all events it does not to any striking degree resemble the figure given by Agassiz. It is no Porocidaris. The pedicellariz (Pl. IX, Fig. 26) are much like those of Dorocidaris, only the opening of the large globiferous pedicellariz is more round and of a more definite form than is otherwise the case in this genus; but this fact might | very well be interpreted as a specific difference. Tridentate pedicellariz simple. A much more con- siderable difference is found in the spines; they are long, slender — unfortunately they were broken, so that their length and the form of their point are unknown. The base is finely pink, the outer part white. They are quite smooth and shining, as if polished, and the structure of the outer layer ECHINOIDEA. I. 23 is peculiar (Pl. XI. Fig. 24) with no trace of roughness on the surface. Perhaps the specimen of Porocidaris Sharrert mentioned by Agassiz (9 p.13) «which was of a light greenish pink color when alive, the spines white with a delicate brownish-pink base» is identical with the specimen described here — in this case this specimen mentioned by Agassiz has certainly not been of the same species as the one he figures; but this latter must, of course, keep the name of Sharrerz. There can be no doubt that the specimen described here is a new species; whether it also is to be regarded as a new genus, or belongs to Dorocidaris, can only be decided, when the systematic significance of the spines has been established. For the present it ought to be classed with Dorocidaris, under the name of D. micans n. sp. Neither is P. zzcerta Koehler (233 a), of which species Prof. v. Beneden has lent me a speci- men for examination, a Porocidaris. I have only found one form of globiferous pedicellariz on it; it has no end-tooth, the opening small, round (PI. VIII, Fig. 31). Most likely another, larger form of globiferous pedicellarize will be found in this species; but the figured form is a sufficient proof that this species has no relation to Poroctdaris. Koehler also refers it only in a doubtful way to Poro- cidaris on account of the highly dentate actinal radioles. The spicules are simple. Of the other species that have been referred to Porocidaris, P. Cobosi most likely is a genuine Porocidaris, but it cannot be decided with certainty, till the pedicellarise have been examined. For the present nothing can be said with certainty of P. Milleri and misakiensis; according to Agassiz (13) P. Millert is «closely allied to P. elegans». On the other hand it may be said with certainty that P. gracilis Déderl. is no Porocidaris. Its globiferous pedicellarize of which only one form is known, recall to some degree those of «Gontocidaris» canaliculata; tridentate pedicellarie unknown. Perhaps it ought to form a separate genus. The genera Stereocidaris and Goniocidaris to. which a whole series of species have been referred, are still left. The species referred to Stereocidaris: japonica, grandis, sceptriferoides, and the here described new species S¢. zwgolfiana agree in the structure of the pedicellarie: there is no end-tooth, and the large opening reaching to the very point is broad and well limited below, quite narrow above. The small globiferous pedicellarice chiefly of the same structure, without end-tooth; the tridentate pedicellarize seem to show no special peculiarities (they are not known in all the species). The spicules _are rather large fenestrated plates, not thorny bows, as is else the case in the Cidarids — this, however, does not apply to all the species; in SZ grandis they are of the common form, and so the spicules give no reliable generic character. There is no reason to doubt that also S¥. imdica Déderl. really belongs to this genus, although we have no informations of its pedicellarie. Déderlein further thinks (118) that Dorocidaris tiara and alcocki are perhaps only local forms of this species. Of the species S¢. tenuispinus and microtuberculatus Yoshiw. nothing can be said with certainty. — Whether this group of species really belongs to the same genus as the ‘fossil Stereocidaris-species, cannot be definitely decided, until the pedicellariz of the latter are known; but the probability is that they really belong here, and there is no reason, at all events not for the present, to reject the name of Stereocidaris for them. To the genus Goniocidaris, the only one of the hitherto admitted genera that has been com- monly acknowledged, the following species have been referred: geranioides Lamk., ‘¢udaria Lamk., 24 ECHINOIDEA. I. canaliculata Ag. (to which Cidaris nutrix W.Th., Gontoc. vivipara Studer, and G. membranipora Studer are referred as synonyms), forigera Ag., Déderleini Ag., biserialis Déderl., clypeata Déderl., umbraculum Hutton, and MJortenseni Koehler. Types of this genus are the species geranioides and tubaria, espe- cially peculiar by having rather deep pits between the plates, in each of which pits is placed an almost globular pedicellaria. These pedicellarie are very peculiar, short and broad; the opening, which is small and surrounded with distinct teeth, reaches to the point, so that no end-tooth is found (Pl. X, Fig. 20). The small globiferous pedicellarie have a powerful end-tooth; no tridentate pedicel- lariz seem to be found. Spicules of the common form. There can be no doubt that G. geranioides has the same structure of the pedicellarie as G. twbaria; the large globiferous ones are figured by Agassiz (Revision Pl. XXIV, 12—13), and they are obviously very similar to those of ¢wbarta. Perrier (op.cit. Pl. III, 12) figures a small globiferous pedicellaria, but the figure gives no clear information of the structure of the point; the text, however, leaves no doubt that it is built as in G. tubaria. Most closely allied to these two species is no doubt G. wmbraculuwm Hutton. The pedi- cellariz (Pl. X. Figs. 13, 21) show only little difference from those of the two mentioned species. Also G. biserialis Déderl. belongs here; to be sure, it is not clear from the figures and description of Déderlein, in what way the small globiferous pedicellariz are constructed, but Prof. Déderlein has kindly sent me a preparation, so that I have been able to substantiate that they are built as in the other species, with a powerful end-tooth (Pl. IX, Fig. 10). The two species G. clypeata and mikado are especially distinguished from the other Gomzocidaris-species by the spines being highly widened, and having, moreover, a peculiar basal widening; the impressions in the angles of the plates are indistinct; the pedicellariz seem also to be somewhat different from those of the typical Gonzocidaris- species, although agreeing with them in main features (no end-tooth on the large pedicellarie, an even uncommonly powerful one on the small ones). Thus there seems to be every reason to comprise these species in a separate subgenus, Dzscocidaris, as proposed by Déderlein (114). Doéderlein thinks that G. florigera must be referred to the same group, especially because it also shows the basal widening on the spines, although only as a trace. It has long been doubtful to me, whether the two forms figured by Agassiz as G. florigera (Chall. Ech. Pl. I. Figs.7 and 12), were really the same species, and my doubt was confirmed, when I had examined the type-specimens in British Museum. They are not only two different species, they will even undoubtedly have to be referred to two different genera — and moreover it appeared that among the specimens determined as G. /lori- gera still a third form was hidden, which must also form a new genus. The form meant by Déderlein when he places G. florigera together with clypeata and mikado, is the one figured in Fig. 12; it is this form of which the spines show traces of the basal widening. It has already been mentioned above, and a new genus has been established for it: Petalocidaris, its pedicellarize not admitting it to be referred to. any of the other known genera. Otherwise it is presumably most closely allied to the two mentioned species. The other form, which is figured in Fig.7, shows no basal widening on the spines, which are, upon the whole, very much different from those of Pefalo- cidaris; they are highly and rather regularly thorny, evenly tapering. In none of the three specimens (Chall. St. 204) I have examined, large globiferous pedicellarize were found, but only the small form, which is quite similar to the small pedicellarie of Dzscocidaris (Pl. X. Figs.6—7); for the present ECHINOIDEA. I. 25 therefore, I think it better to refer it to this subgenus; the spines, to be sure, show no trace of the widenings peculiar to the two other species, but the not widened spines of the latter are rather similar to those of this species, for which I propose the name of Discocidaris serrata n. sp. From st. 192 (Chall.) a specimen is found referred by Agassiz to G. florigera, which it also resembles rather well (i.e. it resembles the one figured in Fig. 12, Petalocidaris florigera). ‘The spines are much richer thorny than in this species; the ambulacral areas almost naked. The pedicellarie are very peculiar (Pl. X. Figs. 25, 28). The opening is a long, narrow slit reaching not quite to the point; a powerfully developed end-tooth is found. The small pedicellariz are essentially of the same structure, the opening only being somewhat shorter and a little broader. Such pediceilarie have not been found in any of the other known species, and accordingly this species must form a separate genus, for which I propose the name of Schizocidaris with the species Sch. assimilis n. sp.*). According to Agassiz (Chall. Ech. p. 43 seq.), Gondocidarts canaliculata is exceedingly varying; he thinks that Czdaris nutrix W.Th. must be regarded as one of the many forms of this species, and also that G. vivipara and membranipora are synonymous with it. After having examined the speci- mens of G. canaliculata in British Museum I must admit that it really appears as if they all formed only one highly varying species, in which a great number of transitional forms connect the easily recognised extreme forms. If we examine the pedicellariz, we shall get another conviction; we shall then see that at all events three different species are found among these specimens referred to G. canaliculata. There is a fact that ought to have made Agassiz hesitate in referring them all to one species. He quotes the description by Wyv. Thomson (397) how the eggs of C. matrix «are passed along on the surface of the test towards the mouth, and the smaller slightly spathulate prim- ary spines, which are articulated to about the first three rows of tubercles round the peristome, are bent inwards over the mouth, so as to form a kind of open tent, in which the young are developed». Immediately after this quotation Agassiz (op.cit. p.45) says: «The specimen (Pl. II. fig. 2) shows the manner in which they are held in a sort of marsupium by the folding of the abactinal spines over ~ the young crowded upon the abactinal system». Thus in this species not only a nursing of the brood should take place, but the young should even be placed, now round the mouth, now on the apical area. Even if this were not inconceivable, it would have been worthy of remark; but Agassiz has no word of it, though it might seem to imply that Cidaris nutrix is really specifically different from - Gonioc. canaliculata. Wyv. Thomson (397 p.66) also remarks expressly that in G. canaliculata we have the reverse of the fact in C. mutrix: «These spines ... lean over towards the anal opening, and form an open tent for the protection of the young as in Cidaris nutrix, but at the opposite pole of the body». There is also another fact that ought to raise the suspicion against the interpretation of all these forms as one species: most of the specimens are coast-forms, taken on depths of 3—150 fathoms; from this there is a far cry to a depth of 1600 fathoms and more. Beforehand it is very improbable that the same species should be found in so varying depths. This fact is not mentioned by Agassiz either. According to my examinations Cidaris nutrix is specifically different from G. *) Unfortunately I made no more thorough notes on this specimen, as during my stay at Br. Mus. I had no clear understanding of the fact that it was a genus quite different from the other specimens called G. forigera. I did not get a clear view of this fact till after my return, when I had examined the pedicellari more exactly. The peculiar pedicellarie may, however, be sufficient for the identification of the species, and therefore I do not hesitate to give it a name here. The Ingolf-Expedition. IV. 1. 4 26 ECHINOIDEA. I. canaliculata; among the deep-sea forms at all events one new species is found, and upon the whole searcely any genuine G. canaliculata is found among them. In the typical G. canaliculata the large globiferous pedicellarize do not differ much from those of Gontocidarts tubaria, or still less from those of G. wmbraculum; they are somewhat narrower, and the blade is a little curved inward below the rather large opening that reaches to the point; there is no end-tooth (Pl. VII. Figs. 8, 32). The small pedicellariz, on the other hand, are very different from those of the genuine Gonzocidaris-species, as there is no end-tooth (Pl. VIII. Fig.6). Spicules simple. — The young are carried on the apical area. «Cidaris» nutrix (Wyv. Thomson’s type specimen examined): the large pedicellariz (Pl. X. Figs. 3—4, 12, 14) very much resembling those of Stercocidaris grandis (Déderlein 116, Pl. VIII. 2); the small globiferous ones (Pl. X. Fig. 24) chiefly as in G. canal- culata, — The young are carried round the mouth. The two species are most frequently easily distinguished as to their habitus. In C. sutrix the apical area is densely set with rather long, club-shaped spines, between which large pedicellariz are found abundantly. In G. canaliculata the apical area is set with rather few and scattered, not club- shaped spines some of which are quite small, so that the area looks rather naked; generally no pedi- cellarize are found on the apical area. This difference, however, is not absolutely reliable, and without the pedicellarize the two species are not always to be distinguished with certainty. It is evident that these two species cannot be referred to the genus Gonzocidaris; especially the small pedicellarize are different from those of Goniocidaris, as they have no end-tooth. Déderlein (116. p. 18) thinks G. canaliculata to be nearly allied to Dorocidaris; to be sure it occupies an extreme position in the «Dorocidaris»-group, and perhaps it might also be regarded as the only representative of a special group. In many respects it recalls the «Hzczdarts»-group. «Wirklich nahe Beziehungen zu einer der bisher bekannten Arten von Cidariden bietet diese Form jedenfalls nicht dar». — As has already been mentioned, the pedicellariz of C. mwtrix are very similar to those of Stereocidaris grandis, and these two species would seem to have to be referred to the genus Stercocidaris; at all events there seems to be no objection of consequence to their being referred to this genus, and it might be difficult to point out a character, which would necessitate the establishing of a special genus for these species. The simple spicules are in accordance with those of S4 grandis (in the other Stereocidaris- species they are, as mentioned, large fenestrated plates). Of the species « Gontocidaris» vivipara and membranipora the former (according to Studer, 386) is synonymous with G. canaliculata, which statement I am able to corroborate from the examination of a specimen that our museum has received from the museum at Berlin. The other (also according to examination of specimens from the museum at Berlin) is identical with «Cidaris» nutrix W.'Th., as has already been supposed by Studer (385). As the paper by Wyv. Thomson (397) bears the date of June 1 1876, and that of Studer (384) the date of July 27 1876, the name of zuérvix has the priority. Now we meet here with a new difficulty. Studer says of G. membranipora (384 p. 455): «Die jungen Czdaris bleiben auf dem Analfelde der Mutter bis zu ihrer vélligen Entwicklung, von den obern Stachelreihen geschiitzt, die sich kreuzweise dariiber legen». According to this statement this species would seem nevertheless to carry the young now arround the mouth, now on the apical area. As this seems to me to be very improbable, I must suppose a mistake to have taken place, so that ECHINOIDEA. I. 27 the specimen (or specimens?), which Studer has had, with young ones on the apical area, is not G. membranipora (= nutrix), but canaliculata, and then it is scarcely from Kerguelen (comp. the fol- lowing about the occurrence of these two species). When the pedicellarie are not examined — which has evidently not been done by Studer — it is, as has been stated above, not always to be decided with certainty, to which of the two species a specimen in hand belongs; this will especially hold good, when, as the case has been here, the apical area is not to be seen. Among the rather numerous specimens of these two species exam ned by me (from «Chal- lenger» at British Museum), S¢ canaliculata was only taken at the Falkland Islands and a station near those islands, «Chall». st. 315, Sz. trix only at Kerguelen. Some specimens from st. 150 («Chall.») near Kerguelen, 150 fathoms, have pedicellariz as those of the typical S¢. atrix but the spines are much longer, three times the diameter of the test; perhaps it is a separate species) Wyv. Thomson (397) mentions C. zwtrix from Kerguelen, G. canaliculata from the Falkland Islands. In the same way Studer’s G. vivipara (= canaliculata) is from Patagonia, his G. membranipora from Kerguelen. Thus it would seem that these two species do not occur together; S¥. canaliculata is found at the southern coasts of South America, S¢ muiérix at Kerguelen. Agassiz, to be sure, mentions S¢. canaliculata from several other localities at Kerguelen, but according to what is shown here his statement is not to be relied upon. Until a definite proof of the opposite fact comes forth, I must believe that either of these species has a territory of its own, as represented here. Among the deep-sea specimens referred by Agassiz to G. canaliculata, I have only examined two from Chall. st. 156 (the South Polar Sea, 1975 fathoms). No doubt they represent another species. The large globiferous pedicellarie (Pl. VIII, Fig. 35) recall very much those of the Gonzocidaris-species, but the small ones are like those in canaliculata and nutrix; and thus it would seem that this species must also be referred to Stereocidaris. The. ground-colour is very dark, almost black; the primary spines are white, the actinal ones highly indented in the edge. Perhaps it may prove to be identical with «Porocidaris» incerta Koehler. I have not examined the specimens from st. 147 (1600 fathoms) and 153 (1675 fathoms), but that they are not identical with S¥. canaliculata or nutrix, which live on shallow water, may be said a priori with a great deal of probability. Goniocidaris Mortenseni Koehler. Koehler (2334) in his excellent description of this species mentions only one form of pedicellariz with . The description gives otherwise only very incomplete informations of this species, and no figures are given. 1) As this specimen is said by Bell (69) to have disappeared, I must observe that it has later been found again. ECHINOIDEA. I. 49 Another very distinctly marked group is formed by the species Asthenosoma varium Grube, Grubei Agass., wrens Sarasin, and heteractis Bedford, all which species I have had occasion to examine. The primary spines on the actinal side are curved, and end with a thin, but rather long, little conspicuous hoof; they are green with dark rings. All the spines on the abactinal side and the secondary ones on the actinal side are covered with skin; on the larger spines the bag of skin is repeatedly constricted (Chall. Ech. Pl. XVI), on the small spines there is only a simple bag of skin at the point (poison apparatus — Sarasin 350, 352); these skin-covered spines end in the usual point. The tube feet are placed in three dense series; in the actinal tube feet a well developed sucking disk is found. The spicules are small, irregularly branched, rarely with a single hole (Pl. XI. Fig. 20); only just below the sucking disk a few larger fenestrated plates are found. They are placed in 2-4 series, but only in the outer part of the foot, nearest to the sucking disk; in the other, larger part of the tube foot only quite few scattered spicules are found, and also in the abactinal tube feet only very few spicules are found. This feature of the spicules also separates this group of species very distinctly from all the other Echinothurids. The pedicellariz of these species are especially characteristic (while on the other hand there is only very little difference in this respect between the species themselves). Only tridentate and tri- phyllous pedicellarize are found here, but in return the tridentate ones are found in no less than three well marked forms. In the largest form the blade is narrow, only widened in the point and provided with 2—3 very coarse indentations which work into each other when the pedicellaria is shut; below the blades are then widely separated; there are no fine teeth in the edge of the blade (PI. XIV. Figs. 3, 7). Now, to be sure, I have only seen this form in A. varium and Grudet, but I think there is no doubt that it is also found in the two others. There appears, besides, some difference between A, varium and Grubei just with regard to this form of pedicellariz, they being much slenderer in A. varium than in.the other; in both they have a length of 2—2:2mm (the head). The neck is quite short. — This difference in the pedicellariz of the two species indicates that A. Grudez is really a good species, and not synonymous with A. variuwm, as Agassiz is inclined to think (Chall. Ech. p. 84). The second, smaller form of tridentate pedicellarize (Pl. XIII. Figs. 4, 27) reminds very much of ophicephalous pedicellariz; but as no indication of arcs is found here, there can be no question of referring them to this kind; they are a highly modified form of tridentate pedicellarie. The blade is short and broad, filled by a rich net of meshes, and with 2—3 large indentations in the edge, which is otherwise smooth as in the large form. When the pedicellaria is shut the blades join with the excep- tion of a quite small space at the base. Also this form has a very short neck. The length of the head 1-2—1°5"™, This form as well as the following one and the triphyllous pedicellarie are quite identical in all four species. — On Pl. XVI. Figs. 10 and 11 in the «Challenger»-Echinids Agassiz gives tolerably recognizable figures of this and the following form of pedicellariz; — «large, short- stemmed» and «small-headed, long-stemmed pedicellariz» they are called. Pl. XLIV. Fig. 34 likewise gives a rather good figure of a valve of the second tridentate form, and Fig. 36 of the third form, which is here called «large-headed»>. But it would be difficult to say what is meant by Pl. XLII. Fig.9, and Pl. XLIII. Fig. 2, although the former is given as a «long-headed, long-stemmed», the latter as a «long-stemmed, small-headed» pedicellaria of A. Grudez On the other hand the pedicellaria The Ingolf-Expedition. IV. 1. 7 50 ECHINOIDEA. I. figured on Pl. XLII. Fig. 8, which in the explanation of the figures is called a «globular-headed, short-stemmed pedicellaria» of A. Grube, is easily recognizable; but does it really belong to A. Grudei? I have not been able to find such pedicellarie, neither in A. Grudez nor in the other allied species. But it is strikingly similar to the peculiar short-headed pedicellaria of Ph. luculentuwm figured by Agassiz (Pl. X.a. Fig. 7, and Pl. XLIV. Figs. 25—26), and I must suppose a confounding to have taken place. The third, smallest form of tridentate pedicellarie (Pl. XIV. Fig.10) is more simple, but also highly characteristic. The blade is simple, but the apophysis continues into it as a high, sharp, coarsely serrate keel; in the larger specimens of this form the keel reaches to the very point of the blade, in the smaller generally only to the middle of the blade. On the sides of the keel there is a rather coarse net of meshes which is, however, far from filling the blade; in the small specimens this net of meshes is only slightly developed. The edge of the blade is finely serrate. When the pedicel- laria is shut, the edges join through the whole length, only a quite small opening is found below. This form has a rather long neck. The head o5—1™". — In the triphyllous pedicellarie the cover- plate is well developed, with a few holes; the edge of the blade is beautifully rounded and finely serrate (Pl. XII. Fig. 18). The stalks of the pedicellariz are of the common structure, only somewhat stronger than is else the case in the Echinothurids. This group of species is very sharply distinguished from all the other Echinothurids, and must form a separate genus, which will, of course, get the old name of Asthenosoma. The other species referred to Asthenosoma do not justly belong to this genus, no more than the other species referred to Phormosoma do in reality belong there. As mentioned above, Agassiz is inclined to think that A. Grudez is identical with A. variwm. Also de Loriol (246) advocates the same opinion. «La réunion de ces deux espéces me parait fort probable; cependant les exemplaires d’Amboine paraissent différer de ceux que M. Agassiz a fait figurer, par leur forme circulaire, un arrangement des plaques un peu différent dans les zones poriféres et, aussi, par la structure de l’appareil apical qui, d’aprés le dessin ne serait pas la méme> (p. 367). To this may be added the difference of the large pedicellarie pointed out above. — As I have not had both species for examination at the same time, and have moreover only seen a large specimen of 4. Grubet and a couple of small ones of 4. variwm, I shall give no decided opinion of this question. In the work quoted above de Loriol further describes a young Echinid which he calls Asthenosoma varium?? — «Il me parait extrémement probable que le petit exemplaire..., qui est un jeune d’une espéce de la famille des Echinothurides, peut étre envisagé comme celui d’ 7’ Asthenosoma vartum Grube». It is scarcely an Echinothurid at all, far less a young one of 4. variwm. As appears from the description and the figures, the arrangement of the pores (a single, regular series), the spines, the buccal membrane, the apical area are all so different from what is else characteristic of the Echino- thurids, that there can certainly be no question of its being referred there. For the present I shall express no conjecture as to where it may really have to be referred. Ludwig (257) is inclined to think that one of the specimens examined by him is a different — species from A. varium, especially because its large pedicellarie are different from those of A. variwm. The figure given shows, however, that it is only the second, broad form of tridentate pedicellariz that ECHINOIDEA. I. 51 Ludwig has found in this specimen, while he has not seen this form in the other specimens. I shall express no opinion whether it be otherwise the same species or not. Asthenosoma hystrix. The tube feet are placed in three dense series; a well developed sucking disk is found in the actinal tube feet. In the upper part of the tube foot the spicules are large, irre- gular fenestrated plates quite inclosing the foot; in the lower part of the foot they are placed in two distinctly separated series, and are more or less rod-shaped, with few holes (Pl. XI. Fig. 29). The pri- mary spines on the actinal side end in a little hoof. Only tridentate and triphyllous pedicellariz are found. Of tridentate pedicellarize two forms are found, not very sharply distinguished. In the larger form (Pl. XIV. Fig. 26) the edges of the blade are involuted, only the point is a little widened, with a remarkably irregular, finely serrate edge. In the smaller form (Pl. XIII. Figs. 17—18) the involuted part of the blade is shorter, the widened part comparatively larger and less irregular in the edge; when the pedicellaria is shut, the valves are far less separated below than in the larger form (see Wyv. Thomson: «Porcupine»-Ech. Pl. LXIV. Fig. 5). This form occurs in very varying sizes. In the triphyllous pedicellarize the cover-plate is highly developed, with a few, large holes along the median line; the edge finely serrate (Pl. XII. Fig. 34). The stalk of the pedicellarie of the common structure. It is evident that this species is not nearly allied to Asthenosoma, as here limited. Accordingly it must form a separate genus keeping the name of Calverza, which was originally given to it by Wyv. Thomson, and which it has unjustly been deprived of. To the same genus «Asthenosoma» gracile A. Agass. will further have to be referred. Its pedicellarie (Pl: XIII. Fig. 3) agree so exactly with those of C. Aystrix, that no distinct specific difference seems to be found in this feature; only the smaller form of tridentate pedicellarice is a little slenderer than in C. hystrix. The primary spines end in a small hoof as in C. hystrix; the tube feet are arranged in the same way as in this latter. The spicules are rather large, irregular fenestrated plates; in the lower part of the tube foot they are smaller and arranged in two well separated series, in the upper part they join completely, and form a close mail round the foot, as figured by Wyv. Thomson from C. hystrix («Porcupine»-Ech. Pl. LXIV. Fig. 3). The sucking disk well developed. — Agassiz, who has seen, to be sure, that this species is very similar to C. hystrix, mentions in his description of it (Chall. Ech. p. 98) some peculiarities with regard to the arrangement of the tubercles as «special characters»; in pedicellarize and tube feet no - distinct specific difference seems to be found, so that for the present we must rest satisfied with the statements of Agassiz. I discovered a very interesting feature by the examination of the type specimen of this species. Some of the secondary spines were swollen at the point (Pl. XIV. Fig. 27), and in the swollen part proved to be sitting a little parasitic Copepod. This seems to be a case of parasitism hitherto quite unknown, and in interest scarcely below that found by Koehler: the gall- forming, parasitic Copepoda in «Phormosoma uranus> (229)°). . The characters here mentioned for Calveria gracilis as well as the mentioned feature of the parasitic Copepod, apply only to the specimen from Chall. st. 200. — Of some specimens from sts. 184 and 219 Agassiz says that he refers them to this species «with considerable doubt», in which he is ') The parasite will be described by Dr. H. I. Hansen in Vidensk. Medd. fra Nat. Foren. Kobenhavn. “ 52 ECHINOIDEA. I. quite right. They belong to two different species, most likely also to different genera, and none of them has any relation to C. gracilis. The specimen from st. 219 has a remarkable form of tridentate pedicellarize; the blade is long, narrow, with uneven, finely serrate edge, deep and in the lower part filled by a net of meshes. The valve figured on Pl. XIV. Fig. 20 is from one of the smaller pedicellariz. I have only found this form of tridentate pedicellarie. The triphyllous pedicellariz (Pl. XII. Fig. 13) have a well developed cover- plate with few holes; the edge finely serrate. The stalk of the pedicellarize of the common structure. The spicules are large fenestrated plates arranged in two well separated series; the sucking disk well developed. The tube feet are arranged in three series. None of the primary spines on the actinal side are whole, so that nothing can be said of the way in which the point is formed; there is, however, certainly no skin-bag round the point. This species must probably form a separate genus. As, how- ever, no quite sufficient characterization can be given of it here, I shall propose no name for it, but be contented with having pointed out that it has no relation to C. gracilis. The specimen from st. 184 has tridentate pedicellarize somewhat recalling those of Phormosoma; but they are distinguished from the latter by the fact that the widenings from the upper end of the apophysis reach quite to the edge of the blade (Pl. XIII. Fig. 26); (in Phormosoma they, as stated above, end on the middle of the side of the blade.) The triphyllous pedicellarize are similar to those of the specimen from st. 219. The stalk of the pedicellarize of the common structure. The spicules are lengthened, narrow plates, arranged in 2—3 longitudinal series; no sucking disk is found. On the actinal side the tube feet are arranged in a single regular line (on the abactinal side the arrangement was indistinct in the specimen). All the primary spines on the actinal side are broken, so that the form of the point cannot be decided. — That this species has no relation to C. gracilis or to the specimen from st. 219 is evident. It seems to be nearly related to «Ph.» tenue, and would then have to be referred, together with this latter, to the genus Echinosoma. (See farther down p. 57.) Although iu the text Agassiz expresses a strong doubt whether the two species here men- tioned, be really «4.» gracilis, he nevertheless afterwards cites the stations from which they have been obtained, among the localities of this species without adding any interrogation; this way of proceeding is very objectionable — and this is, unfortunately, not the only case. I shall express no opinion whether the specimen(s) from st. 169 is really C. gracilis, as I have not seen it. It is not to be relied upon with certainty, until the pedicellariz etc. have been examined. «Asthenosoma» fenestratum Wyv. Thomson is by Bell (72, 73), and Koehler (229) thought to be synonymous with «4.» hystrix, It has also to be admitted that there is a striking similarity as to habitus between the two species; but a closer examination of the pedicellarie shows that the question is so far from being of one species, that they will even have to be referred to different genera. — There are three kinds of pedicellarize, tetradactylous, tridentate, and triphyllous ones. The tetradac- tylous ones, which have been so excellently described and figured by Wyv. Thomson («Porcupine» Kchinoidea. Pl. LXVII. Figs. 5—6), are something quite unique among the Echinids, and consequently an excellent character of this genus. Bell (72), to be sure, thinks it to be an abnormal form of pedi- — cellarie, as he has not been able to find it in the numerous specimens he has examined. As, how- ever, I have succeeded in finding this form also in 4. coriaceum Ag., there can, of course, be no doubt ECHINOIDEA. I. 53 that it is a normal form of pedicellarize characteristic of this group of species. Of tridentate pedicel- lariz two kinds are found. The larger form has not been seen by Wyv. Thomson, but I have found it on a fragment kept in British Museum under the name of «Calveria Phormosoma», but being undoubtedly an original specimen of Wyv. Thomson’s Calveria fenestrata. The edges of the blade are much involuted, only the point is widened and deeply indented in the edge (Pl. XIV. Fig. 32). The valves are highly curved outward, so that they are wide apart when the pedicellaria is shut. The length of the head up to 2™™. The other form is very varying according to its size (Pl. XIV. Figs. 8, 17, 18, 24). Larger specimens recall to some degree the large form, but the widened part of the blade is comparatively larger, the involuted part smaller; the edge of the widened part is coarsely sinuate. When the pedicellaria is shut the valves are only a little apart (the figure by Wyv. Thomson, Pl. LXVII, 7). In the very smallest ones only a quite small space below is involuted, and the edge of the upper part is quite straight. All transitions between these forms are found, so that they can only be inter- preted as modifications of one kind. Their neck is short, the stalk of the common structure. The triphyllous pedicellariz have the cover-plate much developed, and are lengthened and narrow; the edge finely serrate (Pl. XII. Fig. 33). — The primary spines on the actinal side are curved and end in a little hoof. The tube feet as in C. Aystrix arranged in three separated series; the spicules large, irregular fenestrated plates, in the lower part of the tube foot arranged in four separated series; the sucking disk well developed. As characteristic of this species Wyv. Thomson lays stress on the large membranous interpaces between the plates; as Bell (72) has shown that this feature is very varying this character is not reliable. For the present there is no other sure character than the pedi- cellariz, and even if the tetradactylous ones be wanting, which seems most frequently to be the case, be it now that they have fallen off, or perhaps may be quite wanting in some individuals, the tri- dentate pedicellariz are sufficiently characteristic, so that no confounding can take place between this species and Calveria hystrix. A separate genus must be formed for this species; I propose the name of Arzosoma'). —. No doubt it is this species that Agassiz (6) described as Asthenosoma Reynoldsit, but later (9) retired as a synonym of A. hystrix. To this genus will further have to be referred 4. coriacewm Ag. Of this species I have examined a specimen from Chall. st. 169. This station is not enumerated by Agassiz as a locality of the species, but according to the statement of Prof. Bell the determination of the animal has been made by Agassiz, so that it may be taken to be due to an omission that this station has not come in. — The tetradyctylous pedicellarie agree exactly with those of A. fenestratum, so that no specific difference seems to be found in this structure. They were only found on the upper side, and only a few ones, as it was almost rubbed off. Of the tridentate pedicellariz I have not found the largest form. The smaller form (Pl. XIV. Fig. 5) is especially highly developed, the head up to 2™™long. The blade is filled by a very complicated net of meshes, more developed than in 4. fenestratum. As in this latter, forms are also here found with almost straight edge, as well as such as are rather similar to the large involuted form, and all transitions between them. ‘Triphyllous pedicellarice chiefly of the same form as in A. /enestratum (PI. XII. Fig. 27). (The form figured of 4. fenestratum with the cover- plate open in the median line, is not constant; they are as commonly found with the projections 1) dpawég — thin. 54 ECHINOIDEA. I. coalesced, so that a series of large holes is found along the median line — and they may also be found of the form, figured of 4. coriacewm). ‘The pedicellarice (the tridentate ones) with short neck; the stalk of the common structure. The tube feet in three series. The spicules (Pl. XI. Fig. 15) are not so compact fenestrated plates as in 4. fenestratum, the holes are much larger and fewer. In the lower part of the tube foot the spicules are more narrow, at last only fine, thorny, irregular needles, often a little widened as small fenestrated plates in one end or in both ends, or they have a larger hole in the middle. Below they seem to be arranged in four longitudinal series, above they inclose the whole foot as a close mail. The sucking disk well developed in the actinal tube feet. The primary spines on the actinal side form a very conspicuous, regular series along the outer edge of the interambu- lacral areas; in the ambulacral areas only 5—6 large spines are found scattered on the outer plates. They are curved, and end in a little hoof. «Resembling more nearly the primary spines of Phormosoma than the characteristic flaring trumpet-shaped spines of Asthenosoma», Agassiz says of these spines (Chall. Ech. p. 88). As his «Phormosoma» contains so widely different forms as Ph. placenta and hopla- cantha this statement gives no clear information; the meaning of it is that they are similar to those of A. fenestratum; the hoof is little, short, and broad. Agassiz says of this species that it is , which seems to be an ophicephalous one. This form I have not found in the specimen I examined in British Museum (Chall. st. 200); but as, at the time, I had not noticed the mentioned figure, I have not, of course, made any special search for it, and so I dare say nothing of it. If this species should thus prove to be possessed of two kinds of pedicellariz, to which nothing corresponding is found in any other known Echinothurid, there might be some reason to establish a separate genus for it. For the present, however, I think it most correct to refer it to the genus Hygrosoma, as in so many important structures it agrees exactly with the other species referred to this genus. The last of the Echinothurids described from «Challenger», Phormosoma asterias, differs to a high degree from all the others; to be sure, its peculiarities do not appear from the description of the species by Agassiz (Chall. Ech. p. 104), but his figures give more information, and the examination of the type specimen in British Museum revealed still more interesting features. — The ambulacral areas show the quite unique feature that the small secondary ambulacral plates are wanting; there is only one tube foot for each ambulacral plate. Thus only a single series of tube feet is found, and the — distance between the feet is rather large. This highly interesting feature is seen very well on the figures of Agassiz (Pl. XII. a. Figs. 8, 9); in the description he only says that «the course of the poriferous zone is quite sporadic». (It is a matter of course that this very interesting feature ought to be examined exactly, as it is possible that traces may be found of the secondary ambulacral plates and their tube feet.) The spicules are lengthened, narrow, with few or no holes (comp. Pl. XI. Fig. 18); they are arranged parallel to the longitudinal axis of the foot, in 2—3 well separated series; in the outer part of the foot they may join completely. No sucking disk is found. — The spines are of a quite peculiar structure, that is to say they are flat and broad towards the point (Pl. XIV. Fig. 29). I can give no information whether a hoof is found on the point of these spines or on other spines of common form, as I have not made sufficient notes on this fact. The pedicellariz are not less peculiar. The blade of the tridentate pedicellariz (Pl. XIII. Fig.g) is rather flat, with a more or less well developed, perforated cover-plate below reminding of that in the triphyllous. pedicellariz. The point is hastately cut off, a little widened, with finely dentate outer edge; the apophysis and the lateral edges more or less thorny, In the triphyllous pedicellariz the cover-plate is very slightly developed, highly perforated (Pl. XII. Fig. 12). The edge shows only very slight indications of teeth, so that they are only to be seen under especially high magnifying powers. The stalk of the pedicellariz is quite different from that of all other Echinothurids, as it consists of long, thin calcareous threads, almost without any connection except in the upper and lower end of the stalk — as in an EAchinus. It is evident that this species cannot be referred to any of the other genera; it must form a separate genus, for which I propose the name of Kamptosoma?). 1) xdpntw — bend. ECHINOIDEA. I. 61 To this genus belongs further one specimen (or more?) from Chall. st.272 determined by Agassiz as Phormosoma tenue? — The spicules (Pl. XI. Fig. 18) are as in A. asterias and arranged in the same way; no sucking disk. I can give no information of the fact whether the spines are as in K. asterias, as 1 have no notice of this feature. The pedicellarize are very similar to those of K. asterias, but here moreover a larger form of tridentate pedicellarie is found (Pl. XIII. Figs.15, 21), which I have not seen in the type specimen of X. asterias. As, however, the pedicellariz agree otherwise so exactly, it may be supposed that this form will also be found in K. asterias. This larger form of pedicellarize is chiefly constructed as the smaller one; the cover-plate has only a few holes in the median line, or is quite open the edges not joining completely. The point is a little widened, broadly hastate, with exceedingly finely serrate edge; (as in the triphyllous pedicellariz the serrations are only to be seen under very high magnifying powers); the holes in the blade are beautifully arranged in curved series. They are very long-necked; the head up to o8™™; the stalk is of the structure char- acteristic of the genus Kamftosoma. The smaller form of tridentate pedicellariz resemble to a high degree those of AK. asterias the only difference being that the apophysis and edges have no thorns. The triphyllous pedicellariz are somewhat shorter and more arched than those of A. asterias, but they have the same peculiar cover-plate, and the serrations of the edge are likewise exceedingly slight. — There can be no doubt that this species also belongs to the genus Kamffosoma; but it may be doubtful whether it is a separate species, or identical with K. asterias. The small differences in the pedicellariz are suggestive of its being a distinct species; but this question cannot be decided with certainty, till a direct comparison of the two specimens has been made. Now we have only left two of the species referred to Phormosoma, viz. Ph. panamense A. Ag., and Ph. hispidum A. Ag. As to the former it has been supposed above that it may be a genuine Phormosoma, of the latter nothing at all can be said. Both species have only been preliminarily and very incompletely described. The genus Sferosoma established by Koehler (228, 229) is especially characteristic by the peculiar construction of the ambulacral areas on the actinal side. The secondary ambulacral plates are of about the same size as the primary ones; the primary ambulacral plate is divided into an outer part, in which the pore is found, and an inner part. Thus on the actinal side the ambulacral area consists of 8 series of plates. The tube feet are placed in three widely separated series. The spicules are large fenestrated plates, not arranged in series; there is a well developed sucking disk (Pl. XIV. Fig. 4). Only tridentate and triphyllous pedicellarie are found. The tridentate ones (Pl. XIV. Figs. 2, 6, 33) remind somewhat of those in Ph. placenta, especially the small forms are only with difficulty to be distinguished from those; the widenings from the upper end of the apophysis do not reach to the edge of the blade. There is a rather coarse net of meshes in the bottom of the blade, slightly devel- oped in the small forms, more developed in the larger ones, and in these latter it is set with thorns (Pl. XIII. Fig. 12.) The length of the head up to 2™™, the neck rather short in the large ones; the stalk of the common structure. In the triphyllous pedicellarie the cover-plate is rather slightly developed, with numerous small holes. The edge finely serrate. The primary spines on the actinal side curved, with a large, white hoof. Besides the species of Koehler, SP. Grimaldii, a species established by Déderlein (118), Sf. 62 ECHINOIDEA. I. biseriatum, has been referred to this especially well characterized genus; but it has not hithertho been more thoroughly described, so that for the present nothing can be said of this species. One more genus will have to be established for a large Echinothurid obtained by the «Ingolf»- Expedition. The tube feet form one irregular series on the actinal side; the spicules irregular fene- strated plates not arranged in series; no sucking disk. The primary spines on the actinal side curved, with large hoof. Three kinds of pedicellariz are found: tridentate, ophicephalous, and triphyllous pedicellarie. The tridentate ones occur in two forms; in the larger form (length of the head up to 3°5™™) the blade is filled by a coarse, very thorny net of meshes (PI. XII. Fig. 41). The edges are not involuted; the outer part of the blade somewhat widened. The neck very short, the stalk of the common structure. The smaller form resembles those in Ph. placenta, but the widenings from the upper end of the apophysis reach to the edge of the blade. The ophicephalous pedicellariz (Pl. XIV. Figs. 19, 23, 25) are very peculiar, the upper end of the valve being widened in a wing-shaped way, while the middle part is very narrow. The length of the head ca. o5™, The neck is quite short, contrary to the ophicephalous pedicellariz of the Kchinids, and the stalk is a thick, perforated tube. — As ophicephalous pedicellariz, as far as hitherto known, are not found in other Echinothurids (perhaps they are found, however, in Aygrosoma luculentum (see above p. 59—60), but then they have quite another form) they yield an excellent character for this genus. In the triphyllous pedicellariz the cover-plate is rather slightly developed, richly perforated (Pl. XII. Fig. 31). — For this genus I pro- pose the name of Tromikosoma’). According to these researches the system of the Echinothurids gets the following appearance: Phormosoma Wyv. Thomson (emend.). The primary spines on the actinal side straight, club-shaped, inclosed by a thick bag of skin; marked difference between the actinal and the abactinal sides. The areoles of the actinal side very large. The tube feet are arranged in a single series on the actinal side. The spicules large fenestrated plates; no sucking disk. Only tridentate and triphyllous pedicellarie. The tridentate ones are simply leaf-shaped, with little developed net of meshes. The widenings from the upper end of the apophysis do not reach to the edge of the blade. The stalk of the pedicellariz irregularly perforated. Species: Ph. placenta Wyv. Thomson, bursariwm A. Ag., rigidum A. Ag. Distribution: Northern part of the Atlantic, Japan, the Philippines, New-Zealand. — Archiben- thal forms. Echinosoma Pomel (emend.). The primary spines on the actinal side curved, with a little hoof at the point; the actinal and the abactinal sides look almost quite alike, only a few, large spines being found near the ambitus. The areoles large. The tube feet are placed in one almost regular series on the actinal side; the spicules large fenestrated plates, no sucking disk. Only tridentate and triphyllous pedicellarie. Of tridentate pedicellarize two forms are (always?) found, a large one, flat, with a rich net of meshes, and with the upper end of the apophysis continuing some way into the blade as a serrate crest, and a smaller one, 1) tpopixds — quivering. ECHINOIDEA. 1. 63 simply leaf-shaped, with a little developed net of meshes. The stalk of the pedicellarie irregularly perforated. Species: “ch. tenue (A. Ag.), wranus (Wyv. Thomson). Distribution: The Pacific, the northern Atlantic. — Abyssal forms. Asthenosoma Grube (emend.). Synonym: Cyanosoma Sarasin. The primary spines on the actinal side curved, with a rather long, narrow hoof; rather great difference between the abactinal and the actinal sides, on account of the numerous primary spines covering the whole actinal side; the areoles are almost of equal size on both sides. The spines on the abactinal side inclosed by a thick, annularly constricted bag of skin. The tube feet form three dense series; the spicules small branched bodies, arranged in longitudinal series. Sucking disk well devel- oped. Only tridentate and triphyllous pedicellarie. The tridentate ones occur in three distinct forms. The largest form has a long, narrow blade, widened in the point where it is coarsely serrate (not observed in all the species); the second form has a short, broad, and flat blade filled by a rich net of meshes and with coarsely sinuate edge. The third form is simply leaf-shaped, with the apophysis con- tinued to the middle of the blade, or quite to the point as a sharp, serrate crest. The stalk irregularly perforated. Species: > = og 5 EM i inv ing ovo 4 ppyschini 4. 2. The globiferous pedicellariz with the edges of the blade sharp, not connected by cross-beams; several lateral teeth on either side. The spines strongly thorny, those around the mouth curved; the spicules a little irregular, three- radiate, inot,; bihamate ..,: asi4 «04:3 5 dey pedh ee pees ea Hypstechinus. The globiferous pedicellariee with the edges of the blade thickened, with only one unpaired lateral tooth; the spicules bihamate..................... 3. 3 The. test. much grooved .5 « & «-.i¢/s «sb sFosid jie aa hydeg eialelaclag Again Toate casey Armia Trigonocidaris. — — not eT ee Ce ene ee Prionechinus. 4. The globiferous pedicellarize with the edges of the blade almost quite coa- lesced' on the inside, so that only a series of small holes is left. One very large -amal.- plates. «0%. <9ie:5 ajarepresiesinys Ws: oiniedg ls aan mala tie ce non Si hig Wes dy Genocidaris. The globiferous pedicellariz with the edges of the blade thickened, but not connected by cross-beams. No very large anal plate.................... Arbacina. 1o. Hypsiechinus coronatus n. sp. PL V. Fig. 1. Pl. VII. Figs.1—20. Pl. VIII. Figs. 5, 9, 15, 17, 18, 24, 25, 38. PL XI. Fig, 6. The test is flattened, more than twice as broad as high (the remarkably raised apical area not included); the outline most frequently beautifully round, sometimes a little pentagonal. It is not curved inward at the edge of the mouth. The mouth-slits indistinct, the peristome large. The apical ECHINOIDEA. I. 87 area is large, in ¢ and young specimens slightly raised, in the adult 2? so much raised as to form a very conspicuous knob (Pl. VII. Figs. 1-4). When both the peristome and the apical area are wanting, the test resembles a little ring. " Height Diameter. Greatest Breadth. | Number of plates. ‘ Dia- (apical Longest rahi Sat iecidaagp Perminnins| ARS. | apis, | ohmie | ates arenes: | eines 9 3°5 4 5 Niue tae 8-9 | 7 3 9 4 4 42 2 bi gee 8 8 3 8 3°5 4°5 4 | 7 3 8 32 4 3'5 15 | 3 8—9 3 g 8 35 35 5—6 2 68 2°5 3°5 2°5 1'2 2°5 8—9 7—8 ce) 4 1'8 2°5 2'2 I 1'8 5—6 5—6 3 18 2 2 4 All the measures are in millimetres. The interambulacral areas are about twice as broad as the ambulacral ones; the boundaries between the plates are very indistinct, especially in the ambulacral areas; they are given too distinctly in the figures (Pl. VIII. Figs. 24—25). Near the apical area the ambulacral plates are single, farther down they are coalesced in the common way, three and three. Here one larger tubercle is found for each compound plate, and besides some quite small ones above each primary tubercle. The ambula- cral plates are comparatively high, so that upon the whole the same number of ambulacral and inter- ambulacral plates is found. The pores form almost a straight line, but are in reality trigeminate, which fact, however, is not distinct in the upper part of the area; the upper hole of each pair of pores is larger than the lower one. The interambulacral plates, especially above, are rather broad, the horizontal boundary line between the plates bends downward in the middle; the median line of the area is only slightly sinuate, likewise in the ambulacral areas. Each interambulacral plate has a not very conspicuous primary tubercle near the sinuate lower edge and besides some miliary tubercles: In ¢ the upper plates are almost smooth, in ? these plates are very richly provided with miliary tubercles: In the adult ? the test most frequently has an irregular, grooved-netshaped surface, espe- cially between the close-set tubercles on the upper interambulacral plates. The primary spines are in the adult specimens hardly as long as the diameter of the test, in small specimens somewhat longer than the diameter; the spines around the mouth are somewhat curved in the point. All the spines are strongly indented, and end in a little, conical point, surrounded by ca. 6 smaller points (Pl. VIII. Fig. 9); the actinal spines end irregularly truncate, presumably owing to wear (Pl. VIII. Fig. 17). In transverse sections (Pl. XI. Fig.6) the spines are seen to consist of 6 longi- tudinal ridges the outer edge of which is somewhat widened; they are united with each other so as to form a little cavity in the middle, and 6 smaller cavities in a circle round this. The buccal membrane is covered by large plates, which under the microscope are seen to be common, almost smooth fenestrated plates. Those inside of the buccal plates are smaller and quite smooth, and the plates decrease likewise in size towards the edge of the peristome (Pl. VII. Figs. 11, 15). The buccal plates are more complicate, and form a little arch, as it were, over the base of the tube 88 ECHINOIDEA. I. foot, with the opening directed towards the mouth. The two buccal tube feet are not placed in quite the same line, but one a little outside of the other; this is most distinctly seen in younger specimens, and in quite small young ones of a diameter of up to 2—3™" only one tube foot of each pair is devel- oped at all. Also in a single specimen of a diameter of 6™™ only one tube foot of each pair of mouth- feet is developed; sometimes it may also be seen that one tube foot is quite wanting in one pair, rudimentary in another, while both the tube feet are well developed in the other pairs. — A similar feature is found, as stated by Agassiz, in Prionechinus, or, at all events, in a form by Agassiz wrongly referred to Prionechinus (see above p. 82—83). Spicules are not found in the buccal membrane, the small gills contain the common irregular calcareous plates (Pl. VII. Fig. 12), only, however, in the basal part; spines or pedicellarize are not found on the buccal membrane. The apical area is very peculiar, especially in 9 — a well marked sexual difference being found. In $ the apical area is only slightly raised in the middle (Pl. VII. Fig.9); the ocular plates are small, all widely separated from the periproct, the genital plates are much larger, truncate, rather regularly septangular, only the boundary line towards the ocular plates somewhat curved. Each genital plate has one rather strong tubercle or a pair of such tubercles at the inner edge, the ocular plates are quite smooth, or more rarely with a few, very small miliary tubercles. The genital pore is very small, situated about in the middle of the plate. The madreporite is very little conspicuous, has only few (2—3) pores. The periproct is covered by one larger plate and some smaller ones; in quite small speci- mens the large plate covers the whole periproct. In 2 the mutual relation of the plates is chiefly the same as in ¢, but the ocular plates and especially the genital ones have been very much elongated and bent upward, so that the whole apical. area is raised like a knob. The lower part of the genital plates and the ocular plates in their whole extent are quite smooth, but the inner (upper) part of the genital plates is very richly set with tubercles forming, as it were, a crown round the upper edge of the knob (Pl. VII. Fig. 1). The peri- proct as in $, without tubercles. The genital pores are large, and situated nearer to the outer (lower) edge. Of pedicellarize only three kinds are found: globiferous, ophicephalous, and triphyllous pedicel- lari. Tridentate pedicellarize are wanting — at all events in the specimens in hand. The globiferous pedicellariz (Pl. VII. Figs. 19, 20) remind very much of those in «Achinus» miliaris. The upper ends of the apophysis continue directly in the edges of the blade, which are sharp and run out into 2—4 teeth on either side; there are no cross-beams connecting the edges across the hollow inside of the blade; the end-tooth especially large, of the structure typical in the Echinids. The glands are quite small reaching only to the basal part; no neck. The ophicephalous pedicellariz (Pl. VII. Fig. 18, Pl. VIII. Fig. 38) have a quite short neck, but otherwise they do not, any more than the triphyllous pedicellarize (Pl. VII. Fig. 16), show conspicuous peculiarities. It is, however, to be noted that in the triphyllous pedicellarize the edge is quite smooth. — The stalks of the pedicellariz consist of longi- tudinal fibres connected by cross-beams to a rather compact reticulation; they are not hollow; they increase evenly in strength downward, but are not widened at the base. — The spheeridize (Pl. VIL Fig. 17) show no marked peculiarities; they are slightly spinulous in the point, short-stalked, often _ somewhat irregular, and more globiform than the figured one. ECHINOIDEA. I. 89 The tube feet have a typical sucking disk, as in an Echinus, but generally there are only three leaves in the rosette (Pl. VII. Fig. 10). In the mouth feet the sucking disk, as in an Echinus, is an oval, continuous ring, of a far more complicate structure than the parts of the sucking disk in the other tube feet. The spicules (Pl. VII. Fig. 13) are small three-radiate, somewhat irregular bodies. In the lower part of the tube feet almost none are found, nearest to the sucking disk they are more numerous, and are here often a little branched and larger. No spicules are found in the skin at the base of the spines, nor in the genital organs. The dental apparatus is of the structure common in the Echinoids; on the other hand the auriculz are peculiar, only consisting of a pair of small processes, not joining above. None of the specimens in hand show indication of any coloration. This little Echinid is especially interesting by nursing its brood — a fact hitherto unknown among the regular Echinids, with the exception of two Cidarids: Stereocidaris nutrix and canaliculata. As mentioned in the description there are in 2 a great many tubercles on the upper coronal plates, and on the upper edge of the genital plates. The spines of these latter are bent downwards thus joining those of the upper coronal plates. By this means a protected space is formed round the knob- like process; the genital apertures open into this space, and here then the eggs and young are placed protected by the spines (Pl. VII. Fig.5). The number of the eggs varies from 3—7; they are about o5™" in diameter. Sometimes they are all in the same stage of development, sometimes may be found in the same individual almost quite developed young and eggs or embryos where the first skeletal structures have not yet been formed. It was not possible, by means of the material in hand, to study the whole development of the young, only a few stages have been given (Pl. VII. Figs.6—8). In the youngest stage (Fig. 6) the first beginning of the teeth is seen; the buccal plates are begun, and the primary tentacles may be dis- cerned through a plate, which I take to be the terminal plate (the ocular plate). Between each pair of buccal plates, a little outside, a larger unpaired plate is found, the basal plate (the genital plate?). In the following stage (Fig. 7) the different parts of the dental apparatus are begun, and in some of the buccal plates a larger hole has appeared. In the oldest stage (Fig. 8), in each pair of buccal plates one large opening has been formed for the buccal tube foot, and this feature of only one tube foot being developed, is still found, as mentioned above, in young specimens of a diameter of 2—3™™, and sometimes in still larger specimens. The smallest individuals, in which I have found both buccal tube feet developed, had a diameter of 4™™. In the oldest stage figured, the five primary tube feet are seen distinctly, and the five first spines, interambulacral ones, are begun. In corresponding stages only one large anal plate is found (Pl. VII. Fig. 14), which may be perforated by a larger opening; accordingly it seems quite to encompass the anal aperture. Of this especially interesting little Echinid several specimens have been taken by the «Ingolf»- Expedition on the following stations: St. 73 (62° 58’ N. Lat. 23° 28’ W. L. 486 fathoms. 5°1 bottom temp. Bottom [?]). 1 specimen. i ERO te capo mme er FOO mo. AF = Mud. ). 40 — — 81 (61°44’ — 8 2a7°rr — 485 — 5°7 -- ? ). 18 _ S BgOr 56 ag a Ogg 44 oe ? page The Ingolf-Expedition. IV. 1. 12 go ECHINOIDEA. I. St. go (64° 45' N. Lat. 29° 06’ W. L. 568 fathoms. 4° bottom temp. Mud. ). 2 specimens. RT AOR RE | srt 899898 «cre ASO core ee oe Sty clos Sears Further three specimens have been taken by Ryder (1888) on 553 fathoms in the Denmark Strait. Thus this species also belongs to the rich archibenthal fauna of the northern Atlantic; it is scarcely to be doubted that it is also found in other places than in the Denmark Strait and on the ridge south of Iceland. On the Fam. Echinometrade Gray and the Subfam. Triplechinide A. Agass. It has been shown in the preceding, how little successful the previous attempts at a classifica- tion of the Cidarids and Echinothurids have been. It is still worse with regard to the forms that are to be treated here. In the former only the species and genera were confused; here not only the species and genera, but also the families have been mingled to such a degree, that species which have proved by a closer examination to belong to at least three different families, have been referred to the same genus (S¢rongylocentrotus). The «family» Echinometride and the «subfamily» Triplechinide prove to be interwoven to such a degree, that it is impossible to treat each group separately. I have examined almost all the genera and species referred to these groups, and have found the relation between these numerous forms that all look rather uniform, to be widely different from what has formerly been supposed — although these suppositions have otherwise been sufficiently different. The earlier attempts at a classification of the forms belonging here, have been put together by Liitken, to whose paper I shall only here refer‘). Gray is the first author, who has tried to arrange the genera into families; he establishes the following system?): Fam. Hipponoide. The ambulacral areas as broad as the interambulacral areas; the pores form three separate series. — Amblypneustes, Boletia, Hipponoé, Holopneustes. Fam. Echinide. The ambulacral areas half as broad as the interambulacral areas; the pores form arcs of 3. A. With pores at the sutures. JMesfpilia, Microcyphus, Salmacis, Temnopleurus. B. With- out pores at the sutures. Zchinus, Psammechinus, Heliocidaris. Fam. Echinometradz. The ambulacral areas half as broad as the interambulacral areas; the pores in arcs of 4 or more. A. Test round: Strongylocentrotus. B. Test oblong: Echinometra, «Holo- centronotus», Colobocentrotus. In the following time repeated attempts have been made to improve the system, but none of these attempts have been very successful. A short survey of these systems is given here. Troschel (403. p. 297). (No genera are named.) Fam. Echinidz. Pores trigeminate; mouth-slits insignificant; no ocular plate reaches the periproct. Fam. Tripneustidz. Pores trigeminate, mouth-slits deeper than broad; two ocular plates reach the periproct. 1) Bidrag til Kundskab om Echiniderne. Kobenhayn 1864. p. 84 f. (Vid. Medd. Naturh. Foren. Kbhvn. 1863.) *) An arrangement of the families of Echinida, with descriptions of some new Genera and species. Proc. Zool. Soc. 1855. Pp. 35—39- ECHINOIDEA. I. 9r Fam. Toxopneustide. Pores multigeminate; the test round or pentagonal. Fam. Echinometradez. Pores multigeminate; the test elliptical. Agassiz (Revision of Echini). Fam. Echinometrada. Pores multigeminate — Colobocentrotus, Heterocentrotus, Echinometra, Parasalenia, Stomopneustes, Strongylocentrotus (Subgen. Spherechinus, Pseudoboletia), Echino- strephus. Fam. Echinide. Pores trigeminate. (Subfam. Temnopleuride.) Subfam. Triplechinida. Phymosoma, Hemipedina, Echinus, Toxopneustes, Hipponoé, Lvechinus. Bell (40). Fam. Echinide. Group I. Test round. Echinine. a) The ambulacral plates formed of three primary plates. Echinus etc. Bs) — — _ — - four or more primary plates. Strongylocen- trotus ete. Group II. The morphological axis obliquely to the longitudinal axis, Echinometrine. — Il — — — at right angles to the longitudinal axis. Heterocen- trotine. Pomel (324). (In this account of the system of Pomel the fossil | genera are omitted). Les Echinométriens. Colobocentrotus, Podophora, Heterocentrotus, Acrocladia, Echinometra, Ellipsechinus, Parasalenia. Les Héliocidariens. Strongylocentrotus, Toxocidaris (= Anthocidaris Ltk.), Loxechinus, Echino- strephus, Stomopneustes, Heliocidaris (= Evechinus), Holopneustes. Les Schizechiniens. TZoxopneustes (= Boletia), Pseudoboletia, Hipponoé, Spherechinus, Ana- pesus (= Lytechinus Ag., Psilechinus Ltk., Schizechinus Pomel). Les Psammechiniens, Zchinus, Psammechinus (miliaris etc.), Arbacina (forbesiana). Duncan (132). Fam. Echinometrida. Subfam. Echinometrine. Heterocentrotus, Colobocentrotus, Echinometra, Stomopneustes, Parasalenia, Subfam. Polyporine. Strongylocentrotus, Spherechinus, Echinostrephus, Pseudoboletia. Fam. Echinide. Lchinus (Subgen. Psammechinus), Toxopneustes, Boletia; Tripneustes (Subgen, Evechinus). I. W. Gregory’). Fam. Triplechinide. Zchinus, Psammechinus, Tripneustes (= Hipponoé), Toxopneustes, Boletia, Evechinus. Fam. Strongylocentrotida. Strongylocentrotus, Spherechinus, Pseudoboletia. Fam. Echinometridz. Zchinometra, Stomopneustes, Heterocentrotus, Colobocentrotus, Parasalenia. 1) Echinoidea, in ) with the exception of Pomel and Bell. In the essay on the Echinometrids quoted above, Bell has given a thorough criticism of this feature, and has shown that it is by no means a natural principle of division, in spite of the assertion of Agassiz (Rev. of Ech. p. 423) that «this division, although it appears a numerical one, is yet one of great physiological importance, as the mode of growth of the poriferous zone in these two families is totally unlike». I must assert, still more strongly than has been done by Bell, that this division is a quite numerical one, not at all corresponding to the natural relation of the forms. Moreover it cannot be carried through at all, some species having on the lower ambulacral plates (i.e. as young individuals) trigeminate pores, on the others multigeminate ones. Besides the instances mentioned by Bell: Lchinostrephus, Strongylocentr. drobachtensis, Echinometra macrostoma and other Lchinometra- species, I can name «Strongylocentrotus» albus and lividus that have also only three pairs of pores in the lower ambulacral plates. Also in young Spherechinus granularis trigeminate pores may be found in the lower plates, and this feature, I think, may be taken to be found in all polypore forms. When Bell, in his group of Zchinine, uses the number of the pores as a base of further subdivision, I can- not agree with him; so much importance is not due to this feature, it can by no means be regarded as more than a generic character, and I should not wonder, if in some cases it should prove to be no more than a specific character. At all events the number of the pores has only slight importance or none at all with regard to the natural grouping of the genera; Pomel seems to be the only author, who has hitherto seen this fact. The breadth of the ambulacral areas is used by Gray as a distinguishing character. That it is especially unfortunate is shown by the result, as Gray thereby is brought to the uniting of dmdly- pneustes, Holopneustes, Boletia, and Hipponoé into one family, what is absolutely wrong; neither has any author followed him in this respect. The slits of the test are used by Pomel and Troschel, by the latter, however, only as a sub- ordinate character, the number of the pores being used as the first principle of division, so that only the forms with trigeminate pores are referred to his family Z7ipmeustide, while Spherechinus and Pseudoboletia are referred to the family Zoxopneustide. — Agassiz says of the deep slits of the test in Spherechinus (Rev. of Ech. p. 451): «the presence of deep, sharp cuts in the actinal system ... are simply quantitative characters, the value of which a better acquaintance with the subject will deter- 1) Synopsis des Echinides fossiles. 1855. ECHINOIDEA. I. 93 mine». The better acquaintance, however, does not grant that Agassiz is right, on the contrary we find that we have here an especially important systematic character. All the genera with deep slits of the test agree also in other respects, as will be shown hereafter, and form a separate, distinctly limited group (that is to say in such a way that not all the forms belonging to this group have deep slits of the test, but that all forms with deep slits of the test belong to this group; for in some small forms no doubt belonging here, the slits of the test are not very large). The group of «Les Schizechiniens» of Pomel is completely correct — the only correct thing in all the systems hitherto given. The form of the test plays a very great part in the previous systems; that all oblong forms belong to the Echinometride is considered as a matter of course. Even by Agassiz, who character- izes the family Achinometride as «having always more than three pairs of pores to each are», Para- salenia is referred here, although it has only three pairs of pores in each arc; but it is oblong, and accordingly it must be an Echinometrid! That the obliquity, however, is a character insufficient for being the base of a family Zchinometride, has been justly emphasized by A gassiz (Rev. of Ech. p. 436). In Stomopneustes there is in large individuals an indication of obliquity, and «there are in Echino- metra, in one and the same species, specimens in which the elongation of the axis cannot be traced». — Already Stewart (381) has called attention to the fact that Parasalenia is distinguished from the Echinometride, «to which family most would, I should think, refer Pavasalenia», in the structure of the spines and the pedicellaria. According to my examinations that quite corroborate the observa- tions of Stewart, there can be no question of referring Parasalenia to the Echinometrids. And so the obliquity of the test must be dropped as a reliable character; not every oblique Echinid can before- hand be taken to be an Echinometrid. That the obliquity is not the same, the morphological axis not being in the same proportion to the longitudinal axis in all the oblique forms, has been shown by Joh. Miller*), and again emphasized by Bell (op.cit.), who according to this fact distinguishes between Lchinometrine and Heterocentrotine. As. consequently none of the characters hitherto used, with the only exception of the slits of the test, have any greater systematic importance, we must seek other characters, by means of which we can set this chaos right. The characters, of which there can be any question, are the following: the structure of the test, the apical area, the spines, the gills, the buccal membrane, the inner ana- tomical structures, especially the dental apparatus and the auricule, the spheridiz, the spicules, and the pedicellariz. The structure of the test cannot be expected to yield more important characters; if such were to be found they would no doubt have been found long ago, as the attention has hitherto almost exclusively been directed to the form of the test, the arrangement of the tubercles etc. in the descrip- tions. The systematic attempts mentioned above, show to a sufficient degree of how little value the characters found here are. One feature of not quite small importance is found, however, which seems to have been quite overlooked by almost all later authors, viz. that in several forms only every other ambulacral plate has a primary tubercle, while in others every ambulacral plate is provided with such a one. Only in Liitken (op.cit. p.87) I have found a remark «that it is not always the case that 1) Uber den Bau der Echinodermen. Abh. d. Berl. Akad. d. Wiss. 1853. p. 128. 94 ECHINOIDEA. I. every (ambulacral) plate has its primary tubercle well developed». He has not, however, used this feature as a systematic character. On the other hand Diiben & Koren) and G. O. Sars?) have carefully noted this fact in their descriptions, and Koehler (233.a) has recently given prominence to this feature in his description of Sterechinus antarcticus. The apical area, no doubt, shows some difference: sometimes all the ocular plates are shut off from the periproct, sometimes one or more reach to it. That no greater importance can be attached to this feature is a sure fact, which may be seen with especial clearness from a case as that of Ster- echinus antarcticus (= Ech. margaritaceus), where in young individuals all the ocular plates are shut off from the periproct, while in the adult they reach, all of them, to it (Koehler, 233. a). The structure of the spines does not seem to yield very good systematic characters. Mackin- tosh (265) has given numerous excellent figures of transverse sections of spines from a great number of species. But I do not think that he has found so great and reliable differences in this feature, that it can be used as a criterion of a nearer or farther relation between the separate forms. Especially I think that a greater variation in the structure of the spines of the same species may be found, than is to be seen from the work quoted. Also the secondary spines of the different species may deserve a nearer examination. Hesse (195.a) has recently made thorough studies of the structure of Echinid- spines, especially the fossil ones. He arrives at the result, «dass fast jede der einzelnen Familien der Echinoideen ihren eigenen mikrostrukturellen Stacheltypus besitzt, und dass die histologischen Ver- haltnisse der Stacheln ein wichtiges systematisches Kennzeichen fiir die Familien und in gewissen Ziigen von secundérer Werthigkeit oft sogar fiir die Gattungen, ja fiir einzelne Arten der Seeigel liefern» (p. 204). He establishes 6 types: Czdaris, Echinus, Diadema, Clypeaster, Scutellide, and Spa- tangus, and if we take the families to be of a corresponding extent, the spines may be seen to yield «. Also common bihamate spicules are found together with these, but in small numbers. A quite unique form of spicules is found in Stomopmeustes; they are of two kinds: smaller, irregular fenestrated plates, and large, thorny, perforated tubes that may be a little branched (Stewart. Op. cit. Pl. L. Fig. 1). The spicules are especially found in the tube feet, but also in the skin round the pedicellariz (especially the globiferous ones), both on the stalk, the neck, and the head, and round the base of the spines they occur frequently. In the gills and the buccal membrane bihamate spicules are often found together with the more or less irregular fenestrated plates that are commonly found here. Also the inner organs are often richly provided with spicules that may be of a very irregular form, as has been shown by Stewart with regard to Echinometra. This, however, is of no practical importance in the classification where regard must chiefly be paid to the regular spicules of constant form in tube feet and pedicellarize. With regard to the pedicellarie we have some good informations, especially in the works by Perrier and Agassiz. From these informations it is evident that an abundance of peculiar struc- tures may be found here which are, no doubt, of great systematic importance. Thus Perrier has 1) Recherches sur les Pedicellaires et les Ambulacres des Astéries et des Oursins. Ann. Sc. nat. 5. Série. Zool. T. XII—XIII. 1869—7o. 2) On the Spicula of the Regular Echinoidea. Trans. Linn. Soc. XXV. 1865. 3) Notes on the formation and absorption of the skeleton in Echinoderms. @fvers. Kgl. Vet Akad. Férh. 1894. ECHINOIDEA. I. 97 rightly mentioned as a character of the Echinometrids that the globiferous pedicellarie «se termine(nt) par deux crochets, mais ces deux crochets naissent 4 des hauteurs différentes, quoique assez rapprochés du sommet du Pédicellaires. Even if Perrier has not understood this feature quite correctly, his figures are sufficiently clear and good. Accordingly no excuse can be found for the later authors, when they have overlooked this excellent character and in stead of it have stuck to the useless ones: the number of the pores and the form of the test. If they had made use of this character, they might have avoided the many systematical errors they have now fallen into. Beyond the peculiarity of the globiferous pedicellarize of the Echinometrids emphasized by Perrier, no attempts, as far as I know, have been made to find other characters in the structure of the pedicellariz that might be used for a limitation of larger or smaller groups inside this difficult division of the Echinids. The reason why no such characters have hitherto been found, is partly that far too few genera and species have been examined, partly that the examinations have not been made with sufficient exactness. My examina- tions have shown that in the structure of the pedicellariz such peculiarities are found as yield excel- lent characters, by which the genera may be grouped. In «, Agassiz says («Blake» Echini. p. 39), and also Wyv. Thomson classes Ech. elegans among the «critical species» (395. p. 744). In this statement I cannot at all agree with the two celebrated authors. Ech. elegans is very different from Ech. acutus; the question might rather be of referring it to another genus than of confounding it with Ech. acutus. The most essential difference is that it has a primary tubercle on all the ambulacral plates. The globiferous pedicellarize (Pl. XVIII. Figs. 2—3) have generally two lateral teeth on either side, the tri- dentate ones are somewhat shorter and broader than in the preceding species, but the edge is also here set with transverse series of small teeth. In some specimens only quite small tridentate pedicel- larize occur of a somewhat other form than the large ones (Pl. XX. Figs.9, 19), but in other specimens both the small and the large form as well as all transitional sizes are found. Apical area, buccal membrane, and spicules as in Ech. esculentus. — The difference here stated between Ech. elegans and acutus is already seen from the description of Diiben & Koren), where it is said that «de primara kn6larne bilda paa skalet, fran anus till munnen, 20 ytterst tydliga, aldrig afbrutna rader», while it is said of Ech. Flemingit (p.267): «de to rader primara kndlar, som upptaga ambulacralplatarne, aro esom- oftast afbrutna»; this feature is also emphasized by the authors under Ach. norvegicus. To be sure it is not clearly seen in the Latin diagnoses, so that it is perhaps on account of the language that this feature has been overlooked by the later authors?) to great injury for the correctness of the determina- tions; especially Zch. elegans may often have been confounded with quite red specimens of Ech. norvegicus. Ech. Wallisi Ag. In the description of this species («Blake»-Echini. p. 39) it is said that it is «readily distinguished .... by the arrangement of the pairs of pores in sets of two». If this be correct it can scarcely be an Achimus, in which genus the pores are always trigeminate; Agassiz himself, however, thinks that it is «closely allied to, if not identical with, Achimus Alexandri>, in which the pores are arranged in the conimon way. Agassiz further thinks it to be «allied to A. Flemingit and 1) Skandinaviens Echinodermer. p. 273. 2) Thus in Bell’s «Catalogue of British Echinoderms» it is said of Ech. acu/us: «each of these (the compound Ambu- lacra plates) has a large primary tubercle set about the middle of each plate». p. 146, 13* 100 ECHINOIDEA. I. E. elegans»; according to what has been stated above it cannot be closely allied to both these species, and no inference can be drawn from the quite insufficient description that is not even accompanied by figures. From U.S. National Museum I have received a specimen on loan, determined as Zed. Wallist. It is a large, fine specimen of Ech. elegans (only with somewhat shorter spines and higher than the typical form); but it is unfortunately not certain that it is really identical with Ech. Wallisi, as it does not agree very well with the description, except in the colour. Thus Ach. Wallis? must for the present remain somewhat problematic. Most nearly related to Echimus elegans are the species: gracilis, Alexandri, and lucidus, and the new species described here: Ech. affinis n.sp. and atlanticus n. sp.; they have ali of them a primary tubercle on every ambulacral plate; numerous fenestrated plates imbedded in the buccal mem- brane (this feature, however, not observed in £. ductdus); no ocular plates reach to the periproct; the spicules bihamate; all with rather strong, long, and pointed spines. ch. Alexandri is rather sharply distinguished from the other species by its tridentate pedicellarie, which are especially broad and comparatively short (Pl. XX. Fig. 1), while in the other species they are long and narrow (Pl. XVIII. Fig. 4). In the smaller forms of tridentate pedicellaricze the blade is more flat and broad, and the upper end of the apophysis is a little widened as a more or less perforated plate; in the larger forms there is some mesh-work in the bottom of the blade. As in &. elegans there are in these species all transi- tions between the largest and smallest tridentate pedicellariz; to be sure, I have only seen a few of smaller size in Ech. luctdus, but as these resemble to a high degree, those of a corresponding size in the other species it may be supposed that also in this species large tridentate pedicellariz will be found of the same form as in the other mentioned species. In all these species the tridentate pedicel- larize are upon the whole so similar, that reliable specific characters can scarcely be found in them (Pl. XVIII. Figs. 15, 21—22, 26—28). — The globiferous pedicellarie in Ech. Alexandri have generally 3—4 teeth on either side, in the other species there are most frequently 1—1 or 1—2 lateral teeth. Also the globiferous pedicellariz are very similar in all these species (Pl. XVIII. Figs. 9—11, 16—18, Pl. XIX. Fig. 18). Ech. affints is distinguished from the other species by the peculiar feature that the two series of tubercles in each ambulacral area are of unequal size or quite irregular; there is, however, always a primary tubercle on every ambulacral plate (see the particular description below). Zch. gracilis is easily distinguished from the other related species by its beautiful green coloration; the tridentate pedicellarize (Pl. XVIII. Figs.15, 21) are a little more serrate below than in the other species, it is however, scarcely a reliable character. Agassiz, in his description of it (Rev. of Ech. p. 293), says: «this species holds an intermediate position between &. Flemingii Ball and £. melo Lamk., to both of which it is allied», This, according to what is stated here, is incorrect; its nearest relations are Z. elegans and the other species named here. — Ech. lucidus, of which species Prof. Déderlein has kindly lent me a specimen for examination, is most similar to Ech. Alexandri, but may easily be distinguished from this species by its tridentate and globiferous pedicellarize (Pl. XIX. Fig. 18). In Challenger-Echinoidea (p. 114) Agassiz mentions Echinus acutus from st. 343, off Ascension, 425 fathoms. I have had occasion to examine these specimens in British Museum, and I must positively assert that it is not Ech. acutus. The test is high; the peristome very small (r5™™ in a ECHINOIDEA. I. IoI specimen of a diameter of 65™™), the edge of the mouth not bent inward. There are very few spines on the abactinal side, almost only the primary ones, and as the plates are very high, the primary spines are also widely separated; on the actinal side there are more secondary spines, they are not, however, very close-set. The primary spines are of a middle length, and do not decrease much in length towards the apical area. A primary spine is found on each ambulacral plate, and they are of equal size in both series. The buccal membrane with numerous, lengthy, simple fenestrated plates outside the buccal plates; inside of these they are small and a little less perforated, as in 2. Alexandri. The colour is beautifully red, the point of the spines white. The globiferous pedicellariz (Pl. XVIII. Fig. 17), which are very few in number, have 1—1 lateral tooth, but are otherwise similar to those of Ech. affinis; also the tridentate pedicellarie are scarcely to be distinguished from those of Z. affinis. On the other hand the ophicephalous pedicellariz are very characteristic, lengthy, and the teeth in the edge are uncommonly fine, only to be seen under especially high magnifying powers (Pl. XIX. Fig. 37). Triphyllous pedicellariee of the common form; spicules bihamate. — There can be no doubt that this is a new species of Echinus, closely allied to Z. elegans, gracilis etc.; I propose to call it Echinus atlanticus. Presumably there are among the Echinids obtained by the «Challenger»-Expedition still one or two species allied to those mentioned here. Agassiz has determined these specimens partly as Ech. elegans (from Tristan d’Acunha), partly as Ech. norvegicus (from Patagonia, st. 308, and Japan, st. 232). That these determinations are incorrect is a sure fact. «Ach. elegans» from Tristan d’Acunha is a large form, very similar to Ech. Alexandri, that is to say, to the most long-spined specimens of this species (see the description below), but its tridentate pedicellarize are narrow as in Ech. affinis. «Ech. nor- vegicus» from Japan is absolutely not this species; as far as I am able to see from my notes, it must be ch. luctdus; the pedicellarie are quite agreeing with those of that species. The speci- méns from Patagonia, at all events, are not Zech. norvegicus; they belong to two different species, of which one (3 large specimens) belongs to this group of species with a primary tubercle on all the ambulacral plates; perhaps it is Zch. affinis, but I am not able to determine it with certainty after my notes. The other species (4 small specimens) is Ech. magellanicus Phil. — The incorrect referring of these specimens to Ech. norvegicus has unfortunately given rise to the fact that this species is now constantly named among the «bipolar» animals. Ech. margaritaceus lamk. Of this species it is justly said in «Rev. of Ech.» (p. 493) that it has «very marked features», but in the description only one of its peculiarities is mentioned, viz. the nature of its covering with spines; «the plate is densely covered with minute secondary tubercles carrying short, slender, yellowish spines closely crowded together, which are a lower groundwork from which the primary spines, long, slender, and white, project prominently». This description of the spines is excellent, it is only to be added that these spinules are richly set with fine thorns, which gives them a peculiar silky gloss; further that the primary spines round the mouth are curved in the point, and that generally, but not always, some small, club-shaped spines are found on the buccal plates. Only every other ambulacral plate carries a primary tubercle. The apical area is very peculiar, all the ocular plates reach to the periproct, which is large and covered by numerous small plates among which the central plate is especially distinct. In small specimens all the ocular plates are shut off 102 ECHINOIDEA. I. from the periproct. The buccal membrane has inside of the buccal plates numerous small fenes- trated plates imbedded in the skin; just outside of the buccal plates there are a few small plates, as thick and complicate as the buccal plates, and like these set with pedicellariz. Nearest to these plates some small, fine fenestrated plates are found, but all the rest of the buccal membrane is quite naked. The globiferous pedicellariz (Pl. XIX. Fig.20) are of the same form as in Ech. elegans etc., but only one tooth is found on either side. The tridentate pedicellarie are more peculiar and of a rather varying form (Pl. XIX. Figs. 3, 33). The blade is broad and deep, without or with a quite feeble net of meshes at the bottom; the edge is more or less sinuate in the part where the valves join: sometimes almost through the whole length (Fig. 3), sometimes only in the outer half (Fig. 33); it is finely serrate, but not thickened, and has no transverse series of teeth as in the Zchinus-species men- tioned above. The «huge pedicellariz....covering the whole test», mentioned by Agassiz, are the globiferous pedicellariz, which are rather long-stalked and conspicuous, not the tridentate ones. The ophicephalous and triphyllous pedicellarize of the common form; it may, however be noted that in the latter the upper ends of the apophysis do not reach to the edge of the blade, and that there seems to be a tending to a formation of a little mesh-work in the blade. The stalks of the pedicellariz of the common structure; the spicules bihamate, very numerous. — That this species is not «most closely allied» to Ech. norvegicus, as Agassiz thinks (14. p.11) is clearly shown by the characters here mentioned. The description of Ech. margaritaceus given here agrees remarkably well with the description of Sterechinus antarticus by Koehler (233. a.), and after having examined some specimens from «Belgica» which Prof. E.van Beneden has most kindly lent me, I must positively assert that it is Ech. marga- rvitaceus; no single character can be pointed out that might be a mark of distinction between them. — Echinus diadema Studer is by Agassiz (Chall. Ech), Bernard (79), and Meissner (285) thought to be synonymous with Ech. margaritaceus. Studer (386) admits, to be sure, that they are very similar, but thinks that some difference is found in the pedicellarize — i.e. the ophicephalous ones. Now it is true that his figures show a slight difference; but the ophicephalous pedicellariz are generally of very little importance with regard to the distinguishing between the species, and yield only quite exceptionally good specific characters (as in Ech. atlanticus). In this case there can be no question of distinguishing between the two «species», either by the ophicephalous or the other pedicellariz. After having examined some specimens, determined by Studer himself as Ach. diadema, which I have received for examination from the museum at Berlin, I must decidedly follow the mentioned authors; Ech. diadema cannot be distinguished from Zech. margaritaceus. Echinus horridus A. Ag. is not closely allied to £ch. norvegicus, as stated by Agassiz (Chall. Ech. p. 116); its nearest relation is no doubt Ech. margaritaceus. The spines are quite as in this species, and also the pedicellarie are very similar to those of the latter species. The tridentate pedicellarize (Pl. XIX. Fig. 2) are rather much open and rather sinuate in the outer part, where the valves meet; they may become pretty large (a little more than 1™™), and then they have a rather strong, coarse net of meshes in the blade (it may be described as cross-beams rather far from the bottom). In the globiferous pedicellarize (Pl. XIX. Fig. 22) cross-beams are wanting between the edges of the blade (also in young Ech. margaritaceus they may be found without cross-beams), and there are ECHINOIDEA. L 103 2—4 teeth on either side. The basal part has somewhat projecting outer corners. The ophicephalous pedicellarize are of the common form, the triphyllous ones resemble those of Ach. margaritaceus. — Agassiz says, but wrongly, that only two kinds of pedicellarize are found in this species,