AARAR RR AAAAA a\el A AAA ala aA AP ’ Pe AY | WV WA sale) Be yy, VVYVY J IATA A) al) ARK t AAA WAAAAAA aa AAaae ANA 3B > yD 3 D>) DF DyyF yy >> Ds > yy Q >> AAA As ARPA, Baan AARRAR AAA A AaaaaNaona J » > » » ) > y » Sy - 33 By BP. » Dp» ey > py 5 => S D> Says AA Arn , ANANANAAAAAL NS f HAAAAA. | aaaannetnnn AAA 533». SRARAI PY NAAR a a) a) a a) a EW NS A A A Was AGRRAAAn AARAA RAR RARER AAA WAN AAARARAY Ateananer? RAnaAnnnn nanos WARRRAAARAA AA RAARAARAR AAA aaaannaen AA AAANAAAA aan AAA = val al PP\alala! LAA maar ARAARA APA RANA AA panaasnaa AMA AARAAAAAAA AAA aa aaaaalas AMARA As AAA A AAR AAANAAAAACAA A AARAAAAaa m ban | ban Van Jan Ei = Ze.’ fe ~—RAAAR AAR AAA mn RRAAA , AAAA AAA RAP PVA AAAZ SN AARE AAA | yaAAAAAAN. sAAAAAAAAnAAAAAA AMnancaneene AAeenaan aeaas ARAL ANNA AAWal AAARRAAAA, SAAN NAA AAAAAAAAAANA we AsaAAanara Aalaalania 2 > 2 > > > iD) >» > »D 22 => > aaaaaaaaas lal RARARARA AA al AAAZA PAAR RA Raa seas caeace cena rw Ww AA AAA AAARARRAAAAR? Aap AAAAAARA AARaA lA ANA RARAACCAN AAABAZ nana AA ACAAAANAAARAAE A VA ary i A's P AA) ANA, LAAN ANAAAAAAIA. AAA A - AAAYAV.-V\. AAA JalAAAAAWAAAAAAAAAA WY, AAAA RRR AnAaaAAaaanaal AAA ap AVAL YW AAA ABRARAAARA 2 SAAAAnns AAA ARANSAADEEAIAMANWAMANAMMA MQ AAA A Ap AMOI PMO eee AARAAAA | A WAAL AR) AAAs NAAAAAA AAAAal AAAAAAA A AAR aes AGRA ‘ | A Af SAAAAAA AE a Race RAAAWAAAAAAARARC AAA Ar MAA ARAARAEY NAA va Ap \ ARA \AAA\AG ~nAAAAR A&A & A AARAAZ RGEC ceeG CR RATAPIIY <5 anc 555 sRoAMAMARRNRROMACON A BIA \e ) A ancaacn ee rt praca AAAAARARAAAAAA AAA nice AANA A AARARAAARARALY ' aon di CONTENTS. On the Genus Actinometra, Miill., with, a Morphological Account of a new Species | {A.) polymorpha from the Philippine Islands,—Part I. By P. Herpert CarPenter, M.A., Assistant Master at Eton College. (Communicated by W. B. CARPENTER, _ C.B., M.D. LLD., F.R.S., PLS.) (Plates 1-8.) . . . . . . Pages 1-122 on Ms) (eer eee ye hae ‘ MAN bias: READER, AND DYER, PATBRNOSUER-ROW. ’ C&¥&C246 1§79 INVZ TRANSACTIONS OF tHE LINNEAN SOCIET Y, I. On the Genus Actinometra, Wiill., with a Morphological Account of a new Species (A.) polymorpha from the Philippine Islands.—Part Il. By P. Herpert CARPENTER, M.A., Assistant Master at Eton College. (Communicated by W. B. CARPENTER, Ch. 2D. CED. 2.BS., LOLS.) (Plates I.-VIII.) Read June 21st, 1877. THE principal part of the investigations, of which the results are set forth in the fol- lowing pages, were carried on during the year 1876, in the Zool.-Zootomical Institute of the University of Wirzbure, under the superintendence of its director, Prof. C. Semper. I would here express my most sincere thanks to Prof. Semper, not only for the generous manner in which he placed at my disposal all his specimens of Act. polymorpha, which he had himself collected in the Philippine Islands, but also for the ready and valuable advice which he constantly afforded me during the progress of my work, and for the free use which he permitted me to make of his extensive library. Iam also greatly indebted to Dr. Sandberger, Professor of Geology in the University of Wiirzburg, to the authorities of the University Library, and to Professor Dr. Halm, of the Royal Library at Munich, for the means of reference to many books which would otherwise have remained inaccessible to me; and I desire to record my best thanks to all these gentlemen for the ready kindness with which they met my frequent and numerous wants. I. Historicat. (§ 1) In the remarkable work of Linckius! upon the “ Sea Stars,’”’ the modern family Comatulide (J. Miiller) is described under the name of Crinitee, or Comatee Stelle, as a eroup distinct not only from the Asteride, but also from the Ophiuride and from Astrophyton (Euryale, Lamarck), with which they have been united by many later systematists. 1 Jowannis Henntct Lrxcxrr Lipsiensis ‘ De Stellis Marinis liber singularis, Lipsiw, 1738, p. 53. SECOND SERIES.—ZOOLOGY, VOL. II. 1 2 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. Linck included three genera in this family, or, as he called it, ‘‘ Classis.” The first of these he named Acxaxynuoc, to indicate “ stellam marinam decem caudis crinitis radi- antem ;” and he referred to it three species :—(1) The “ Crocea zaffarana Neapolitanorum,” or Sexadacvaxrwoedye Of Fabius Columna !, whose description he quotes ; (2) the Decempeda Cornubiensium of Llhuyd*, which Linck figured and named “ Stella dexaxvnnoc rosacea ;” and (8) the “ Ackékynuog fimbriata Barrelieri” *, which was named by Linck barbata, as he supposed it to be different fromthe other two. All three, however, are really identical, being simply local varieties of one and the same species, viz. the British Antedon rosacea, or the Comatula mediterranea of Lamarck. ‘Thus, Fabius Columna described the Neapolitan variety, and Barrelier another obtained at the mouth of the Tiber, while Llhuyd based his description upon specimens found upon the coast of Cornwall near Penzance. Linck’s second genus, the Tpicxawexaxvnuoc, was based upon a specimen with thirteen arms, previously described by Petiver* as ‘“ Stella chinensis ;” this specimen, however, was suspected by Linck to have been mutilated. His third genus he called “ Caput- Meduse,” and described it as including those specimens which ‘ex centro corporis parvi umbonatique per quinque truncos primum bifidi, mox nullo constanti numero multifidi, in 60 et plures surculos geniculatos rectos simplices abeunt, quos gracilescentes fibrille alize pilorum instar vestiunt.”’ Linck referred two species to this genus, viz. Caput Meduse cinereum and C. brunnum ; and he gave good figures of both (tab. xxi. n. 33, and tab. xxii. n. 34), from which it may be determined with tolerable certainty that they represent species now known to belong to two different types among the Comatule—namely, to the genera dAntedon and Actinometra respectively. (§ 2) Although Llhuyd’, and after him Rosinus*, had explicitly pointed out the relationship between the recent Comatulee and the fossil Crinoidea, and although Linck, while supporting and repeating Llhuyd’s views, had clearly differentiated the former from the Asteroidea and Ophiuroidea, yet Linnzeus’, instead of adopting the more correct views of some of his predecessors as to the true relations of the Crinoidea, was so misled by the jointed structure of their stems as to rank them among zoophytes in his genus Jsis, whilst he grouped the Comatulide, together with all the other Starfish, under one common name Asterias. Linck’s three species of Decacnemus were rightly regarded by himas identical; and he placed them, together with Petiver’s Stella chinensis, in one species, Asterias pectinata, to which he also referred a specimen previously described by Retzius*. We now know, however, that this last is an Actinometra, dif- 1 Phytobasanus, sive Plantarum aliquot Historia. Neapoli, 1592. * Epvarvi Lut ‘ Lithophylacii Britannici Ichnographia’ p. 149. Londini, 1699. 3 Jaconr Barretieri ‘ Plante per Galliam, Hispaniam et Italiam observate’ Paris, 1714, p. 131. ‘Gazophylacium Nature et Artis,’ Londini, 1711; and also ‘ Aquatilium animalium Amboinensium Icones et Nomina,’ 1713. 4 5 Prelectio de Stellis marinis Occani Brit. nee non de Asteriarum, Entrochorum, et Encrinorum Origine, pp. 149- 155, Oxford, 1733. ° Tentaminis de Lithozois ac Lithophytis olim marinis, jam vero subterraneis, prodromus ; sive de stellis marinis quondam, nunc fossilibus, disquisitio. Hamburg, 1719. 7 ‘Systema Natur,’ editio decima tertia (Lipsiw, 1788), pars vi. p. 3166. § Nova Acta, Stockholm, 1783, p. 234, n. 12. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 3 fering very considerably from the type represented by Decacnemus (Antedon) rosacea, although resembling it in only having ten arms. In like manner both of Linck’s species of his genus Caput-Meduse, the one an Antedon and the other an Actinometra, were united by Linnzus, together with a many- armed specimen described by Retzius, into one species, Asterias multiradiata. (§ 3) For some years after the commencement of the present century the Linnean nomenclature held its own, and the few species of recent Comatulze with which the naturalists of that time were acquainted were described as different species of the Linnean genus Asterias. The first among the post-Linnean zoologists who recognized the claims of this form of Sea Star to a distinct generic rank was De Freminville 1, who in 1811 presented to the Société Philomatique de Paris a “ Mémoire sur un nouveau genre de Zoophytes de l’Ordre des Radiaires,”’ to which he gave the name of Antedon. His definition of the genus was as follows :—‘ Animal libre, 4 corps discoide calcaire en dessus, gélatineux en dessous, environné de deux rangées de rayons articulés, pierreux, percés dans leur largeur d’un trou central; ceux du rang supérieur plus courts, simples, et d’égale grosseur dans toute leur longueur; ceux du rang inférieur plus longs, allant en dimi- nuant de la base 4 la pointe, et garnis dans toute leur longueur d’appendices alternes égale.oent articulés; bouche inférieure et centrale.” It is not very clear which of the two apertures on the ventral (or, as he called it, inferior) side of the disk was regarded by De Freminyille as the mouth; it is very pro- bable that, as he was only able to examine a spirit-specimen, he failed to recognize more than one—that namely, which, placed at the extremity of a long tube projecting from a point near the centre of the disk, we now know to be the anus. Adams’, who had studied living specimens of Linck’s Decacnemus rosacea, had, how- ever, pointed out some years previously the existence of two orifices to the digestive cavity; but his observations seem to have escaped notice; for Lamarck *, Miller *, and many other naturalists, all regarded the aperture at the end of the anal tube either as the mouth alone or as a combined mouth and anus; and it was not till 1828 that the exist- ence of distinct oral and anal orifices was fully recognized. It was announced as a new discovery by Leuckart®*, in a letter to Von Schlotheim ; and he was followed shortly afterwards by Meckel‘, Gray’, and Heusinger *. 1 Nouv. Bull. d. Scien. par la Soc. Philomat. tom. ii. p. 349. Paris, 1811. 2 «¢ Description of some Marine Animals found on the Coast of Wales,” Trans. Linn. Soe. vol. y. p. 7 (1800). 3 Systéme d’Animaux sans Vertébres, 2™° éd. (Paris, 1816), tom. ii. p. 532. 4 A Natural History of the Crinoidea (Bristol, 1821), p. 128. 5 Von Schlotheim, Nacht. z. Petrefact. Abth. ii. p. 48 (Gotha, 1823); and Leuckart, “Einiges iiber Asteriden Geschlecht Comatula Lam. iiberhaupt, und iiber Com. mediterranea insbesondere,” Zeitsch. fiir organ. Physik, ii. p. 385 (1833). 6 « Ueber die Oeffnungen des Speisekanals bei den Comatulen,” Meckel’s Archiy fiir Physiol. Band iii. p. 470 (1823). 7 «Notice on the Digestive Organs of the Genus Comatula and on the Crinoidea of Miller,” Ann. of Philos. n. s. vol. xii. p. 392 (1826). 8 « Bemerk. iiber d. Verdauungskanal der Comatulen, Meckel’s Archiv” fiir Physiol. 1826, p. 317; and “Anat. Un- tersuch. d. Comatula mediterranea,” Zeitsch. fiir organ. Physik, iii. p.366 (1828). 1* 4 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. De Freminville’s specimen was found on the keel of a vessel which had come from a warm climate; it had ten arms and twenty cirrhi, and was named by him Antedon gor- gonia, He gave no further description of it, but simply referred to the figure of Stella decacnemus rosacea, Linck, in the ‘ Encyclopédie Méthodique’’, which represents the ordinary European Comatula rosacea, as it is now called. This species, however, is not identical with De Freminyille’s Antedon gorgonia, which was referred by Lamarck” to his Comatula carinata, under which name he described some specimens brought by Peron from the Isle de France. Nevertheless the two species resemble one another in some important points, viz. the presence of ten arms, of a central or subcentral mouth, and of an excentric anal tube. In 1815 Leach* rescued the three genera contained in Linck’s classis Crinite from the confusion of the Linnean genus Asterias, and united them into one genus, Alecto, comprising three species, viz. Alecto europea (= Decacnemus rosacea, Linck), Alecto horrida (= Capul-Meduse, Linck, or Asterias multiradiata, Linn.), and Alecto carinata (which seems to have been the same as De Freminville’s Antedon gorgonia). Leach defined Alecto as having the “ os inferius, irregulare,” a description which would suit equally well either for the true mouth or for the anal opening, though perhaps more applicable to the former. He seems, however, like his predecessor De Freminville, to haye regarded the mouth as situated at the extremity of the anal tube; for in the explanation to Schweigger’s figure* of Leach’s specimen of Alecto horrida the latter is described as the “réhrenformig hervorstehender Mund.” It is obvious, therefore, that we cannot make any use for systematic purposes of the definitions of Antedon and Alecto as given by Leach and De Freminville respectively, as far as the position of the mouth is concerned. Schweigger’s figure of the disk of Alecto horrida shows clearly enough that the five trunks of the ambulacral grooves converge towards the centre of the disk, as in Antedon rosacea (Alecto europea, Leach), Plate I. fig. 1. Leach’s Alecto horrida was therefore a true Antedon in the modern sense of the term, although belonging to that division of the genus in which the repetition of the bifurcation of the ten primary arms is carried to a great extent. (§ 4) Leach was apparently unacquainted with the memoir of Freminville; but the same was evidently not the case with Lamarck (1816), who, like Leach in the previous year, united Linck’s three genera into one, to which he gave the very appropriate name Comatula®. Tis definition of the genus differs but little from that given for Antedon five years previously by De Freminville, whose original specimen Lamarck seems to have examined; and it is difficult to see why he did not adopt the name Antedon to designate the genus, which, like Leach and De Freminville, he clearly distinguished as belonging to a different type from the Asteridx, Ophiuridxe, and Luryale. 1 Partie des Vers, pl. 42. fig. 6. 2 Tom cit. p. 534. 3 Zoological Miscellanies, vol. ii. p. 61: London, 1815. 4 Beobacht. auf naturhistorischen Reisen, p. 66, and Taf. iv. fig. 42: Berlin, 1819. 5 Tom. cit. p. 530. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 5 Lamarck included eight species in his new genus Comatula. In only three of these is the mouth at or near the centre of the disk, viz. C. mediterranea (= Stella decacnemus rosacea, Linck), C. carinata (= Antedon gorgonia, Prem. ?), and C. adeone. In all the other five species described by Lamarck, viz. C. solaris, C. brachiolata, C. rotalaria, OC. fimbriata (= Stella chinensis, Petiver), and C. multiradiata (= Caput- Meduse brunnum, Linck), the mouth (as I know from examination of the collection of Comatulz in the Paris Museum, which still contains many of Lamarck’s original spe- cimens) is nowhere near the centre of the disk, which is occupied by the anal tube, but is excentric, or even marginal. At the same time the five primary groove-trunks do not converge towards the centre of the disk, as in Antedon rosacea (Com. mediterranea, Lam.) (Pl. I. fig. 1), but they unite more or less completely into a horseshoe-shaped furrow, at one point of which is situated the excentric mouth (PI. I. figs. 2-5). This type will be described further on, under the name of Actinometra. Lamarck, who found it in more than half of the species constituted by him, seems to have regarded it as common to all Comatule. His description of it is worth quoting, as it is the first notice of a true Actinometra that I have been able to find. He says*:—‘‘Le disque inférieur ou ventral offre un plateau orbiculaire plus large que le dorsal, entouré de rayons simples, cirreux. Prés de la circonférence de ce plateau, on apercoit un sillon irrégu- liérement circulaire qui s’ouvre sur la base des rayons pinnés, et se propage le long de leur face inférieure, aussi que de celle des pinnules. Ce sillon néanmoins, ne s’approche point de la bouche [7.e. the anal tube] et ne vient point s’y réunir, comme cela a lieu pour la gouttiére des rayons dans les Astéries. Au centre du disque inférieur ou ventral des Comatules la bouche, membraneuse, tubuleuse, ou en forme de sac, fait une saillie plus ou moins considérable suivant les espéces.” Although Antedon and Alecto were both constituted previously to Comatula, yet Lamarck’s authority was sufficient to establish the latter name, and to bring it into general use, though Cuvier adopted Leach’s genus Alecto, and used it in preference to Comatula. The latter, however, was more generally employed by all subsequent observers (who pointed out Lamarck’s error respecting the position of the mouth) thenceforward till the time of Johannes Miiller. (§5) During this period the skeleton both of the recent and of the fossil Crinoids was made the subject of careful investigations by Miller? and Goldfuss*. The latter author divided his class Stellerites into two Orders :—(a) Stilasteritee, or Gestielte Seesterne; and (d) Asterites liberi, or Freie Seesterne. The former he again divided into Articulata and Inarticulata, placing in the first group a number of fossil Mesozoic Crinoids, viz. Hugeniacrinites, Miller, Solanocrinites, Goldf., Pentacrinites, Encrinites, and Apiocrinites. In his descriptions of these genera he adopted and considerably improved the some- what inapplicable system of nomenclature, introduced by Miller for the parts of the skeleton of the fossil Crinoids and of Comatula. Portions of his diagnoses of Eugeniacrinites and Solanocrinites are of considerable interest, both zoologically and morphologically. Of the former, he says * :— 1 Tom. cit. p. 532. ? Loe. cit. % Petrefacta Germaniz, i. (Dusseldorf, 1826-35). * Tom. cit. p. 162. 6 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. * Die kurze runde mit einem runden Kanale durchbohrte Siiule besteht aus wenigen walzigen verlingerten Gliedern, und nimmt am obern Ende allmiilig zu. Ihre Gliede- rung wird oft nur durch Ringe angedeutet, unten endet sie in starken Wurzeln. Das letzte verdickte Siulenglied vertritt die Stelle des: Beckens und articulirt durch eine Gelenkfliiche mit den Rippengliedern.” The term “ Becken” is here meant to signify the circlet, of basals which, in Penta- crinus and other stalked forms, intervenes between the stem and the circlet of first radials, the ‘ Rippenglieder ”’ of Goldfuss. These basals are also present in Solanocrinites, which genus, as Goldfuss well remarks, ** hat in verschiedener Hinsicht Aehnlichkeit mit den Pentacriniten, und bildet zugleich einen Uebergang zur Gattung Comatula.” Its most important characters are as follows !:— ** Die Siiule ist sehr kurz, beinahe so dick als der Kelch, fiinfseitig und an ihrer Basis nicht mit Wurzelsprossen, sondern mit ausstrahlenden Runzeln versehen. Ihre Glieder sind mit einander verwachsen und haben an den Seitenflichen Gelenkhéhlungen fiir den Ansatz zahlreicher dicker Hilfsarme. Die oberste Gelenkfliiche der Saiule zeigt fiinf strahlenférmige Erhabenheiten, auf welchen das Becken articulirt. Das Becken wird nicht durch das oberste Siiulenglied vertreten, sondern es besteht aus fiinf Gliedern, welche zwischen die Nihte der fiinf Rippenglieder eingefiigt sind, oder sie bedecken. Die fiinf Glieder des Beckens bilden entweder nur schmale Strahlen, die zwischen die Nahte der Rippenglieder einsenken (S. costatus, S. scrobiculatus), oder sie sind breiter, stossen seitlich an einander, und stellen eine tiefe mit 5 Strahlenfurchen ausgehdhlte Gelenkfliche dar (8. jegeri).” The first genus in Goldfuss’s order ‘“ Asterites liberi” is the Comatula of Lamarck, of which he says*, “ Diese Gattung bildet den Uebergang von den Stilasteriten zu den freien Seesternen und steht zuniichst mit den Solanocriniten in nichster Verwandtschaft.” Besides describing five fossil species, he gives some account of two recent ones, dwelling more especially upon the structure and composition of the skeleton. Thus “ Bei der C. mediterranea besteht die Saule aus drei Gliedern: das Becken fehlt, und die Rippenglieder sitzen unmittelbar auf dem letzten Saulengliede.” ‘ Bei der in den Ostindischen Meeren lebenden C. multiradiata Lam. hingegen, finden sich Becken- glieder, so dass man berechtigt sein kénnte sie als eine eigene Gattung zu betrachten. Ihr Saulenrudiment besteht aus einem einzigen schiisselformigen Gliede, an dessen Rande fiinf schmale dreieckige Beckenglieder ansitzen und mit ihm verwachsen sind. Diese stossen mit ihren Seitenflichen nicht an einander, sondern stehen so weit entfernt dass die ersten Rippenglieder unmittelbar zwischen ihnen auf dem Siulengliede ansitzen, und sie durch einen Abschnitt der unteren Ecke zwischen sich aufnehmen. In der Mitte des innern unteren Randes jedes Beckengliedes entspringt ein zahnformiger Fortsatz, der sich als knospeliger schmaler Streifen bis zum Mittelpunckte des Sau- lengliedes verliingert, in welchem er durch eine Rinne aufgenommen ist. Diese Becken- glieder sind also denen der Solanocriniten vollkommen analog.” ($6) Goldfuss, who, though acquainted with the name of Alecto, yet used Comatula 1 Tom. cit. p. 166. 2 Tom. cit. p. 202. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 3 in preference to it, paid very little attention to the soft parts of either of the Comatule which he dissected. In the following year, however, De Blainville’ described the visceral mass at some length. Like his predecessors, he adopted Lamarck’s genus Comatula, making it the only representative of his section, ‘‘ Les Astérencrinides libres,” while at the same time he acknowledged the prior claims of de Freminville’s name Antedon. He was, of course, acquainted with Lamarck’s error respecting the position of the mouth, which he described as “‘ assez antérieure, isolée, membraneuse, au fond d’une étoile formée par cing sillons bifurqués.” The species which he dissected was a foreign one preserved in spirit ; it had a large number of arms; and from the not very clear description which he gives of its ventral surface it would seem to have been a true Actinometra. After speaking of the tentacular furrows on the ventral surface of the arms, he says’, “En suivant ces espéces de sillons dont le nombre est proportionnel a celui des digitations du rayon, on arrive par un sillon unique pour chacun d’eux et qui en occupe la base, au centre d’une sorte d’étoile & bords épais, frangés, et par suite 4 la bouche qui est au fond. L’étoile formée par la réunion des sillons des rayons n’est pas symétrique, c’est & dire que ses branches sont trés-inégales: les unes que j’appellerai les antérieures, étant bien plus courtes que les autres, ou postérieures. I] en est résulté que la bouche n’est pas au centre de l’étoile, mais bien plus proche d’un cété que de l’autre: elle est assez difficile & voir au contraire d’un autre orifice, dont il va étre question, et que M. de Lamarck paroit avoir pris pour elle. Elle est profondément enfoncée dans V’étoile des sillons : elle est ronde, sans aucune armature et conduit immédiatement dans l’estomac.” The above description implies, if I rightly understand it, that the mouth of De Blain- ville’s specimen was nearer to one side of the disk than to the other, so that the primary trunks of the ambulacral grooves were of unequal lengths. This will be subsequently seen (section 14) to be the principal distinctive character of the genus Actinometra. De Blainville evidently attached no importance to the position of the mouth as a character of systematic value in the determination of the species of recent Comatule ; and from his definition of it as “ assez antérieure,” it would almost appear as if he supposed the other species to agree in this respect with the one dissected by him. This is, in fact, the case in five out of the eight species described by Lamarck, with which De Blainville was probably acquainted, and to which he added no new ones, except that he gave the name of Comatula barbata to Linck’s third species of Decacnemus, the fimbriata of Barrelier, or “ barbata” of Linck. Lamarck had been uncertain to which of his species he should refer it, although, as we have seen above (section 1), it is really only a local variety of his C. medilerranea. Like the other naturalists of his time (1838), Agassiz * also adopted Comatula in preference to the other generic names of this type, but defined it as having the “ bouche centrale en- foncée,”’ and with the five “ rayons du disque bifurqués,” thus limiting the number of arms in the genus Comatula to ten only. At the same time he erected Lamarck’s species C. multiradiata, with sixty or more arms, into a new genus, Comaster, which he defined as Manuel d’Actinologic, (Paris, 1834) p. 249. 2 Op. cit, p. 251. « Prodrome d’une Monographie des Radiaires ou Echinodermes,” Ann. des Scien. Nat. 2° série, Zool. vii. p- 257. 8 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. follows :—*‘* Ce genre a la méme organisation que le précédent ; mais les espéces ont les bras ramifiés au lieu de les avoir simplement fourchus.” Agassiz consequently used Comatula simply as equivalent to the Decacnemus of Linck, while his new genus Comaster was Linck’s Caput-Meduse, or the Comatula multiradiata of Lamarck. Of the seven other species constituted by the last-mentioned naturalist, only two, C. rotalaria and C. fimbriata, have more than ten arms ; in both of these the number of arms is usually twenty, though it may reach twenty-four, or possibly even more. Strictly speaking, therefore, these two species, according to the above definition, should be referred to Comaster, and not to Comatula. This character, the number of arms, upon which Agassiz founded a generic distinction, is, in fact, extremely variable, and by no means of generic importance; in fact, as Goldfuss! remarked a little later, ‘“‘ Wollte man mit Agassiz die Theilung der Arme als hinreichendes Gattungsmerkmal ansehen, so wiirde man folgerecht gezwungen sein fast jede Art der Crinoideen als Gattung aufzustellen.” Leach and Lamarck had already recognized this fact in uniting Linck’s three genera under a common name; and it is not a little strange that Agassiz should have seen fit to separate them again. His doing so, however, led to somewhat important consequences from a systematic point of view. Turning to the fossil Comatule, we find that Agassiz erected the C. pinnata of Goldfuss into a new genus, Plerocoma, and grouped together his other three species, C. tenella, C. pectinata, and C. filiformis, under the generic name Saccoma; while he expressed his belief thaf Solanocrinus was really related to the Comatule, and more especially to the problematical fossil described by Goldfuss under the name of Glenotremites, which he rightly recognized as the centrodorsal piece of a free-living Crinoid. In the year 1840 a new fossil Comatula was described by Hagenow ’ under the name of Hertha mystica. The specimen, consisting of the united first radials and hemispherical centrodorsal piece, was somewhat worn; but Hagenow was able to recognize the resem- blance between it and the remains of Solanocrinus, and the corresponding parts of Goldfuss’s Comatula multiradiata, except that he was unable to find any trace of the external basals which Goldfuss had described in both the above cases; and though he seems, and (as we now know) correctly, to have suspected “das Vorhandensein etwa verdeckt-liegender Beckenglieder,” he was, of course, unable to come to any satisfactory conclusion upon the point. (§ 7) The year 1840 is a noteworthy one in the history of our knowledge of the Crinoidea 5 for it marked the appearance of the first of a series of classical memoirs by Johannes Miiller, who laid the foundation of nearly all our knowledge of the zoology and morpho- logy of the group: the first* of these was devoted to an anatomical account of the recent and very rare genus Pentacrinus, together with many observations upon Comatula. 1 « Beitr. z. Petrefactenkunde,” N. Acta Acad. Leop.-Carol. Nat.-Cur. xix. a. p. 348. 2 « Monogr. d. Riigen’schen Kreide-Versteinerungen, II. Abtheil. Radiarien und Annulaten,” Neues Jahrb. Mine- ralogie, 1840, p. 664. 3 « Ueber den Bau des Pentacrinus caput-Meduse,” Abhandl. d. Berlin, Akad. 1843; Abstract in Monatsb. der- selben, 1840; also in Wiegmann’s Archiv f. Naturgesch. 1840,1. p. 307. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, | Miller seems at first, not unnaturally, to have supposed that Goldfuss was right in referring the many-armed specimen dissected by him to the Comatula multiradiata of Lamarck, for which species he adopted Agassiz’s name Comaster’. But he did not use it precisely in the same sense as Agassiz, who, in his definition of the genus, makes no mention of the external basals, the presence of which was regarded by Goldfuss as the principal character distinguishing Comaster from Comatula. Miiller adopted Comaster? in the sense in which Goldfuss used the name; and when he subsequently discovered * that the Comatula multiradiata or Comaster of Goldfuss was not specifically identical with the specimen described as Comatula multiradiata by Lamarck, he retained the name Comaster for Goldfuss’s specimen only, which, like Solanocrinus, is remarkable for having “kleine dasalia zwischen den Insertionen der Kelchradien, oder sogennanten Beckenstiicke welche den eigentlichen Comatulen giinzlich fehlen’”*. At the same time he gave a careful description® of Lamarck’s original specimen of Comatula multiradiata, based upon an examination of it by Troschel; but as he regarded Comaster and Solanocrinus only as one subgenus of Comatula, he gave it a new specific name “ multifida,” on the ground that “die Comatula multiradiata Goldfuss, als die zuerst genau beschriebene, den Speciesnamen multiradiata behalten muss.” Lamarck’s specimen was thus restored by Miiller to its previous position among the “ Comatulen im engern Sinne, nimlich Gattung Alecto, Leach (Comatula, Lamarck),’’ which he grouped together with Comaster into one family, Comatuline. The fossil Solanocrinus was regarded by him as identical with the latter form, while he referred the Hertha mystica of Hagenow, and Pterocoma, Ag. (C. pinnata, Goldf.), to Comatula or Alecto; for at that time (1841) he used the two names indifferently, considering them (as, indeed, they originally were) equivalent to one another. Goldfuss put forward about the same time a somewhat similar classification. Ina subsequent abstract (with additions) of his ‘ Beitraige zur Petrefactenkunde,’ [which had been published two years previously (1839)] he speaks of the two species dissected by him as “die T'ypen der zwei nichst bezeichneten Genera (Comatula, Comaster), welche daher nebst den zwei zuletzt folgenden (Solanocrinites, Gasterocoma) als Verzweigungen des Lamarck’schen weiten Geschlechts Comatula zu betrachten sind.” He did not, how- ever, agree with Miiller in regarding Solanocrinites and Comaster as identical, partly, apparently, because nothing was known of the arms of the former, and partly because of the differences in the form of the “ Knopf,” or centrodorsal piece, which he called a short stem—although, as Miller showed, this is not a character of any generic value. Although Goldfuss had at first supposed’ that the basals were really absent in Coma- tula mediterranea, and that the first radials therefore rested directly upon the top of the centrodorsal piece, or, as he expressed it, on the last stem-segment, he seems subsequently to have changed his opinion; for in his definition * of the genus Comatula, given in 1839, 1 « Beitriige zur Petrefactenkunde,” loc. cit. p. 349. ? Wiegmann’s Archiy, 1840, i. p. 309. 3 “ Ueber die Gattungen und Arten der Comatulen,” Wiegmann’s Archiv, 1841, i. pp. 140, 147. 4 “Bau des Pentacrinus,” loc. cit. p. 27. 5 Wiegmann’s Archiv, 1841, i. p. 147. 5 Neues Jahrbuch fiir Mineralogie, 1841, p. 818. * Petrefacta Germanise, tom. cit. pp. 203, 204. § Beitriige, &c. loc. cit. p. 349. SECOND SERIES.—ZOOLOGY, VOL. II. 2 10 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. he says, “‘ Auf dem letzten Saitlengliede ruhen fiinf Beckenglieder, und auf jedem derselben _ein Rippen- (=second radial) und ein Schulterglied (third or axillary radial), auf welchem zwei einfache Arme eingelenkt sind,” from which it is evident that he was wrongly led to regard the first radials as representing the basals of Comaster and Pentacrinus. This mistake is hardly a surprising one when we consider the remarkable metamorphosis undergone by the embryonic or primitive basals, and their concealed condition in the adult Comatula mediterranea. Miiller, who examined a very large number of species of Comatula, never found one in which the basals appeared externally, as described and figured by Goldfuss in Comaster, and remarked ':—* Daraus geht hervor, dass die Gegenwart wirklicher Basalia ohne Zer- legung bei einer lebenden Comatule, auch dann, wenn sie wirklich solche besitzt, schwer zu erkennen sein muss. Die Unterscheidung der Comaster und Comatula wird daher bei der Ordnung der lebenden Comatulen unpractisch.” In fact he appears to have given up the genus Comaster altogether; for he addsina note :—* Kiirzlich habe ich die einzige im Museum zu Bonn befindliche Comatula multiradiata (nicht das von Goldfuss zerlegte Exemplar, wovon ich nichts mehr vorfand) untersucht. Ich habe daran nichts von Beckenstiicken erkennen kénnen. Die Gattung Comaster ist daher wohl zu un- terdriicken.”” He seems finally’ to have thought that it might possibly be identical with the C. Bennetti of the Leyden Museum. As, however, Comaster has not been seen by any naturalist since the time of Goldfuss, its position must still remain in doubt. ($8) Up to the time of Miiller no one paid any attention, from a systematic point of view, to the arrangement of the tentacular furrows on the ventral perisome of the disk of Comatula; but Lamarck and De Blainville had, as we have already seen, examined and described, with more or less accuracy, a condition which we now know to differ very considerably from that presented by the Decacnemus of Linck, or the Antedon of De — Freminville. Both these observers seem to have regarded the former condition as the normal one, and as common to all Comatule. Miiller, who does not seem to have been acquainted with their descriptions (for he makes no mention of them), took up the subject systematically, and soon discovered that, using the distribution of the tentacular furrows as a basis of classification, he could distinguish two, as he thought, very distinct types of the genus Comatula, which he named Alecto and Actinometra respectively. In his earlier communications*® on the subject he described the ordinary Comatula and Penta- crinus as having a central mouth and symmetrically distributed tentacular furrows ; i.e. the five main trunks formed by the union of the furrows of the five groups of arms converge directly towards the centre of the disk, being separated by five “interpalmar” areas, one of which, slightly larger than the rest, is occupied by the anal tube, which is therefore excentric in its position (Pl. I. fig. 5, An.). During his visit to Vienna in 1840 Miller had an opportunity of examining an un- 1 ‘Ueber die Gattung Comutula, Lam., und ihre Arten,’ Separatabdruck aus den Abhandl. Berlin. Akad. 1849, p. 8. 2 Ibid. p. 29. 3 « Bau des Pentacrinus,” loc. eit. p. 47, and Wiegm. Archiv, 1840, i. p. 311. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. a usually large specimen of the Comatula solaris of Lamarck—unfortunately, however, only a dry one, which he found to differ so greatly from the other Comatule then known to him, that he described it under the name of Actinometra imperialis*, ‘ welche gene- risch von andern durch die Bildung ihres Scheitels verschieden zu sein schien. Auf dem Scheitel der mit blumenartigen Kalkblittchen bedeckt ist, ist keine Spur von den Furchen zu sehen, die bei den Comatulen von den Armen zum Munde fiihren. Auch ist dort nichts vom Munde zu sehen. Die Mitte der Bauchseite nimmt eine Réhre ein. Die Arme haben die ventrale Furche der Comatulen, die Furchen der zehn Arme miinden aber in gleichen Abstiinden in eine die Scheibe am Rande umziehende Cirkelfurche. Diese eigenthiimliche Bildung liesse sich durch eine unsymmetrische Vergrosserung desjenigen Intertentacularfeldes, worin die Afterréhre steht, tiber den ganzen Scheitel und auf Kosten der anderen Intertentacularfelder erkliren, so dass der Mund aus der Mitte des Scheitels ganz an die Seite zwischen je zwei Armen gerath; es ist mir aber nicht gelungen den Mund hier zu finden.” (PI. I. fig. 2.) Ina subsequent visit to Lund, Miller examined two dry specimens of Comatule, which had been described by Retzius * many years previously under the names of Asterias pec- tinata® and Asterias multiradiata. These he found to belong to the same type as the Vienna specimen, which he had already designated Actinometra imperialis, and which he supposed to be distinct from the true Comatula solaris of Lamarck. When he visited Paris, however, in 1844, he examined Lamarck’s original specimen of this species, and convinced himself of its specific identity with his Vienna Actinometra. Consequently he withdrew the specific name “imperialis,” and described the type simply as Actinometra solaris’. Miiller was unable to determine the position of the mouth in the dry specimens of the Lund and Vienna Museums on which he founded his new genus Actinometra; but subsequently he was able to examine many spirit-specimens both of his typical species, Act. solaris, and also of other “‘Comatulen von jener Anordnung der Furchen, sowohl zehnarmige als vielarmige...... Siehe die beistehende Figur von Comatuia Wahl- berghii*” (Pl. I. fig. 3). This last species he describes a few pages further on as Coma- tula (Actinometra) Wahlberghit. He did not, however, appear to regard the position of the mouth as of any systematic importance; for he goes on to say:—‘ Der Mund ist bei der in Frage stehenden Ab- ? Wiegm. Archiy, 1841, i. p. 141; and “Neue Beitr. z.|Kenntniss der Arten der Comutalen,” Wiegm. Archiy, 1843, i. p. 132. * Dissertatio sistens species cognitas Asteriarum. Lunde, 1805. * The specific name “ pectinata,” Linneus, included both Retzius’s specimens and the Decacnemus of Linck. These belong, however, to two very different types of the genus Comatula, and must be carefully distinguished from one another. The former is, as above mentioned, an Actinometra, while the latter was called Alecto by Miiller, being simply the common Comatula mediterranea, Lam. Pennant, Adams, and others naturally employed the Linnean name for this last species ; but Dujardin, following Retzius and Miiller, applied it also to Retzius’s original specimen, which is really an Actinometra, and not an Antedon, like C. mediterranea. This has given rise to much confusion in the synonymy of these two species, 4 Wiegmann’s Archiy, 1843, i. p. 133. 5 «Gattung Comatula,’ pp. 12, 13. © Ibid. p. 9. 2% 12 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. weichung allerdings vorhanden, er liegt ganz zur Seite, doch ist dies nicht die Ursac h des Unterschiedes, es gibt vielmehr auch Comatulen von der gewohnlichen Anordnung der Furchen, bei denen gleichwohl der Mund seitlich, die Afterréhre central steht. Fig. von C. multiradiata (Pl. I. fig. 4). Die fragliche Abweichung beruhrt vielmehr darauf, dass die fiinf Furchen nicht symmetrisch fiir die fiinf Gruppen der Arme vertheilt werden, sondern dass von den fiinf Furchen einzelne herrschend werden und Aeste an die meisten Arme abgeben. Indem diese Hauptfurchen, nachdem sie die Scheibe umzogen, sich wieder annahern, so entsteht der Schein eines Cirkels. An in Weingeist aufbewahrten Exemplaren sieht man indess, dass es kein geschlossener Cirkel ist ;” and further on (p. 10) he says, “ Ich werde daher bei den Arten wo fiinf centripetale Furchen beobachtet sind, den Namen A/ecto in Klammer dem Gattungsnamen Comatula beifiigen, wo aber weniger Furchenstimme den excentrischen Mund erreichen, den Namen Acti- nometra demselben Gattungsnamen Comatula folgen lassen. Also z. B. Comatula (Alecto) europea; Comatula (Actinometra) solaris.” Miiller does not appear to have been acquainted with De Freminville’s name of An- tedon; but he distinctly states that Leach’s genus Alecto was constituted a year earlier than Lamarck’s Comatula. He had in his earlier communications employed the two indifferently and as equivalent to one another; but when he became acquainted with the type represented by Comatula solaris and elevated it into a new genus, or rather sub- genus, Aclinometra, in contradistinction to Alecto, he retained Lamarck’s name Coma- tula, probably on account of its being so well known, and employed it to designate the genus in which he included the subgenera Alecto, Actinometra, and Comaster. Thus the sole character by which Miiller distinguished the first two of these subgenera from one another was the number of groove-trunks reaching the peristome, irrespective of the position of the mouth. It is therefore easy to understand that, as many of the spe- cimens which he examined were dry, and as in others, although preserved in spirits, the arms were contracted over the disk so as completely to conceal it, he was unable satis- factorily to determine more than three species of Actinometra. Two of these, Act. solaris and Act. Wahlberghii, have been already mentioned ; the third was the small Comatula rotalaria of Lamarck. ($9) The Asterias pectinata of Retzius, which presented the same “ Bildung des Scheitels’ as Act. solaris, resembled it so greatly in other respects, that Miiller regarded the two as almost identical, or, at any rate, as presenting only varietal differences '. He seems also to have come to the conclusion that the other species described by Retzius, the Asterias multiradiata, Linn., had a prior claim to this specific name over either of the similarly named types described by Goldfuss ( Comaster) or Lamarck (Co- matula multiradiata); for he described it as ‘* Comatula (Alecto) multiradiata nob.’”. It is difficult to understand why he called it Alecto; for he had already* described this «Gattung Comatula, p, 52. * Ibid. p. 25. ° Wiegmann’s Archiy, 1843, i. p. 133, and ‘ Gattung Comatula,’ pp. 9, 10. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 13 Lund specimen as belonging to the type of his new genus Actinometra, in which not five, as in Alecto, but “ weniger Furchenstiimme den excentrischen Mund erreichen.”’ Miller further examined a specimen of the “so-called” C. multiradiata in the Bonn Museum; and although he did not actually include it in his type under that name, yet he seems to have been inclined to do so; for he says’ that “es stimmt durch den Besitz der Syzygien an den Axillaria der Arme mit Comatula multiradiata Retz.,” but adds, ‘Maul excentrisch, 5 Furchen der Scheibe sammeln die Furchen der respectiven Arme und kommen am Mund zusammen.” (See Pl. I. fig. 4.) Here, again, it is evident that Miiller’s description of A/ecto will not hold good; for according to his own descrip- tions, the Lund and Bonn specimens of Comatula multiradiata, Mill., however much alike in other respects, differ so greatly in the distribution of the ambulacra on the disk that one is Actinometra and the other Alecto. Miller also referred three specimens contained in‘the Paris collection to this type; and he was perhaps thinking of the condition of the ambulacra presented by them when he added the following sentence to his previous description of the Lund specimen, and named the type Alecto*. ‘Mund excentrisch, aber an Weingeistexemplaren ergibt sich dass die fiinf zum Munde fiihrenden Furchen sich ganz symmetrisch fiir die fiinf Gruppen der Arme vertheilen.” This arrangement, which he called the “ gewohnliche Anordnung der Furchen,” had been already * figured by him as occurring in C. multiradiata, which, as he says, differs from the ordinary C. mediterranea in the excentric position of the mouth (Pl. I. figs. 1, 4). It is thus evident that, according to Miiller’s own nomenclature, two types, differing only in the “ Bildung des Scheitels,” but almost precisely similar in every other respect, viz. the Lund specimen, on the one hand, and the Paris specimens, on the other, were referred by him to the same species, Alecto multiradiata, Miller. It will, however, be shown further on that the distinction drawn by Miiller between Alecto and Actino- meira is not a real one, and that the Lund and one of the Paris specimens, both of which have an excentric mouth and a central or subcentral anal tube, really belong to one and the same species, Aclinometra multiradiata. ($10) For a short time after the publication of Miiller’s Comatula-memoirs the genera Alecto and Actinometra remained as he left them, both being regarded as subordinate types of Lamarck’s genus Comatula. A singularly minute fossil species, discovered by Philippi‘ between the valves of an Isocardia cor from the Sicilian Tertiaries, was named by him Alecto alticeps because of the height of its ‘* Kelchstiick,” a character found both in Alecto Lschrichtii and in A. phalangiwm, as Miller had already pointed out. A few new fossil species of a more or less doubtful nature have been since described, and variously referred either to Miil- ler’s family Comatulinze or to new and distinct types. The typical genus of this family, Comatula, Lam., has undergone numerous changes in its definition. Roemer, who at first revived Linck’s name Decacnemus, subsequently 1 , and comb. 3, Actinometrasolaris &c.havean excentric ,, and comb. 3, Comatula meridionalis has an excentrie ,, and no comb. Leaving Phanogenia out of consideration for the present, as it was unknown to Johannes Miiller, the following scheme will represent the relations of Antedon and Acti- nometra as used by Dr. Liitken and myself, to Alecto and Actinometra as used by Miller :— Alecto. Ambulacra symmetrically | mouth central. Antedon. Oral pinnules not specially distributed on the disk. { mouth excentric. distinguished, ? Actinometra. Oral pinnules nearly | always have a ter- ef} minal comb. Actinometra. Ambulacra unsymmetri- cally distributed on the | mouth excentric. disk. ($15) We are now in a position to investigate which species of the numerous Coma- tule described by Miiller can be referred to Actinometra under its new definition, and what further subdivisions of the genus are possible according to the principles of classi- fication introduced by Miiller. Before doing so, however, it will be advisable to devote a little time to a consideration of the descriptive terminology which he employed, and of the manner in which it has been modified by later writers. In Pentacrinus and Comatula Miller regarded the arms as starting directly from the five radial axillaries. The two primary arms borne by each of these might either remain ' Loc. cit. No. 11, p. 355. Pourtales here evidently uses Antedon as equivalent to Alecto, and not in the sense in which it was employed by Mr. Norman, viz. to designate those forms only in which the mouth is central (or nearly so) and the anus lateral. ‘To avoid confusion, I shall speak of this species simply as Comatula meridionalis. * « Phanogenia, ett hittills okiindt sliigte af fria Crinoideer,’ Ofver. af Kongl. Vetensk.-Akad. Férhandl. 1866, No. 9, p. 223. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 21 simple, as in Ant. rosacea, or divide more or less frequently into secondary, tertiary, &c. arms, as in Act. multiradiata and in Pentacrinus; and every segment, like the radial axillary, preceding a bifurcation, was called by Miiller a “ brachial axillary.” In some of the Tessellate Crinoids, however, the arms do not become free at the radial axillary, but “der Kelch setzt sich noch weiter fort; die Radien zerfallen dann in zwei Distichalradien mit radialia distichalia, die jedes mit einem distichale axillare enden, wie bei Actinocrinus moniliformis und Hucalyptocrinus”'. In this case the distichal radii represent the primary arms of Comatula and Pentacrinus, though Miiller never used the name “distichals”’ in his descriptions of the species of Comatula; for, as in the Tessellata the segments composing two adjacent distichal radii are united laterally with one another by intermediate plates, he regarded them as forming a part of the calyx, and considered the arms of this group as starting from the distichal axillary, and not from the radial axillary, as in the Articulate Crinoids. The two primary arms, or distichal radii, borne upon a single radial axillary, were called by Miiller a “distichium;” and the interval between two successive distichia dorsally between the calcareous segments, or ventrally between the corresponding grooves on the disk, was spoken of by him as “ intervpalmar,” while the interval between the two primary arms or distichal radii borne by the same radial axillary, or, as Miller called it, “die Kluft eines Distichiums,” was ‘“ interbrachial” or “ intrapalmar.” The words “interambulacral,” ‘‘ interradial,” and ‘‘intertentacular,”’ have been also used by Miiller and others to designate the interpalmar areas on the disk of Comatula. Hither of these is preferable to “‘interpalmar,” for reasons which will presently appear, though “ intertentacular” is not universally applicable, as in certain Actinometre the posterior ambulacral grooves bounding the large area in which the anal tube is situated are not provided with tentacles at their sides. The term “ interbrachial” is decidedly preferable to “intrapalmar,’ which was used by Miiller to designate the small areas on the disk, bounded by the two branches of each of the five primary groove-trunks. “Intrapalmar” does not convey any clear idea of the relation of these areas to the divisions of the skeleton, while ‘“ interbrachial’’ dis- tinctly indicates that they correspond to the intervals between the two primary arms borne by every axillary radial. Reemer? adopted Miiller’s nomenclature for the fossil Crinoids, and, like him, con- sidered that the distichal radii, when present and united laterally to one another, formed a part of the calyx; so that the arms were regarded by him as commencing from the axillary distichals, while he distinguished their different divisions simply as rami of the first, second, and third order. De Koninck and Le Hon’, however, regarded the arms as commencing from the first bifurcation, ¢.e. from the axillary radial, whether they become free at once or whether they remain united with the calyx for a longer or shorter distance. Nevertheless they distinguished the arm-segments by different names in the two cases, using the expression “ piéces brachiales” for the distichals of Miiller, 7. e. for those segments which are immovably united with the calyx, while they gave the name “articles brachiaux” to the movable segments, the brachials of Miller. 1 Bau des Pentacrinus, p. 31. 2 Lethza geoenostica, Bandi. Theil 2, pp. 210, 215. 3 Recherches sur les Crinoides du Terrain Carbonifére de la Belgique, (Bruxelles, 1854) p. 69. ’ 22 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. This view, although unquestionably correct in the case of the Articulate Crinoids (Comatula, Pentacrinus, &e.), is, as Reemer has pointed out, beset with some difficulties in its application to the fossil Tessellata; and Schultze’ accordingly reverted to the original view of Miiller, saying, “ Die Arme (brachia) beginnen unverdnderlich da, wo eine deutliche Gelenkfacette eines festen Kelchstiickes, ihren Ursprung anzeigt.” In describing the divisions of the arms, he speaks of the brachial axillaries of the first, second, and third order, without giving them any special names. ‘These are perhaps scarcely necessary when the number of segments between each division varies so much in different specimens and in different arms of the same specimen as it does in many fossil Crinoids, and in Pentacrinus. Among the Comatule, however, the number and character of the segments between the successive divisions of the arms exhibit variations which are to a certain extent constant in different species, and thus give us the means of classifying them into larger or smaller groups. Miiller has availed himself of this character to a certain extent in the scheme which he gives” of a classification of the Comatule according to the presence or absence of syzygia in the various brachial axillaries ; but though, in his descriptions of the different species, he furnishes the material for carrying this classification much further, and for separating species which, in his scheme, stand very near to one another, he never made any use of it, simply classifying the Comatule in the groups which he had constituted, according to the number of their arms—10, 20, 40, or more. Under these circumstances he would have been puzzled where to place Act. polymorpha, in which I have found the number of arms to vary from 13 to 39. (§ 16) It has been already stated that the arms proper of Comatula begin from the radial axillaries. In many cases they are united by perisome as far as their second or third division ; and in Act. multifida this perisome contains numerous small calcareous plates, which render the union of the arms with one another and with the calyx some- what firmer than usual; but they are never so united as to be immovable, as their various segments are connected with one another, except, of course, at the syzygia, by muscles and ligaments. ‘There is one point about the nature of this union which has not, I think, received suflicient attention; and as it shows clearly that the arms of Comatula and Pentacrinus begin from the radial axillaries, it is worth considering here. It is this: the first and second segments beyond every axillary, whether radial or brachial, are nearly always united together in the same manner as the second and third (axillary) radials. Thus, for example, in Act. solaris, and in the forms allied to it, the second and third radials are united by a syzygy. The same is the case with the first and second brachials. In Ant. rosacea, and in the various species which are closely allied to it, there are no muscles between the second and third radials; but their opposed articular faces present a vertical and not a transverse ridge, and are so united by ligament that the two segments are only capable of a lateral movement upon one another, and cannot take part in any movements of flexion or extension, in which they act as a single segment only. The first and second brachials are united in precisely the same manner. 1 Monographie der Echinodermen des Eiflerkalks, (Wien, 1866) p. 5. * Gattung Comatula, p. 11. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 23 In both these groups the primary arms do not subdivide; so that the total number of arms is limited to ten; and we are as yet unacquainted with any Comatula in which the second and third radials are united by a syzygy and there are more than ten arms!. This is, however, the case in Pentacrinus Miilleri, in which, in like manner, the first and second segments beyond every brachial axillary are also united by a syzygy”. On the other hand, Pentacrinus asteria, L. (=P. caput-Meduse, Miller), is remarkable for having muscles between the second and third radials as well as between the first and second’. In the same manner the first and second segments beyond every axillary are united by muscles, and the syzygium is between the second and third segments ‘. In nearly all the Comatule with which we are acquainted, with the exception of Act. solaris and the species most nearly allied to it, the second and third radials are united by ligament only, as in Ant. rosacea, their opposed faces being marked by a vertical articular ridge (Pl. VII. figs. 2b, 3a,5b,6a, 7). Inalmost all of these species which have more than ten arms, the first and second segments beyond every axillary are united by ligament only, just like the second and third radials®*. Thus in Ané. Savignii every third segment, so long as the division lasts, is an axillary, and the first and second segments beyond each axillary are united by ligament only. But in Ant. palmata only two segments follow each bifurcation, the second of which is again axillary; it is nevertheless united to the first one by ligaments only. I have found these same two conditions to occur together in Act. polymorpha (PI. II. fig. 8), in which the normal number of segments between every two points of division is three (Pl. II. figs. 7, 9, 10), of which the third is axillary with a syzygium, as in Ant. Savignii, while the second is united to the first by ligament only. In exceptional instances, however (PI. II. figs. 8, 11), the second segment may be axillary, and united to the first by ligaments only, as in Ant. palmata. In every case, after the division has ceased, the union of the first and second brachials * There are three Coimatule in the ‘ Challenger ’ collection which answer to this description. In two of them the first and second distichals and the first and second brachials are united by syzygies, like the second and third radials. But in the third species there is a curious exception to the rule. The rays may divide eight times ; and in the primary divisions there are three distichal joints, the first two of which are united by ligaments and not by syzygy. But in all the subsequent divisions the first two joints beyond each axillary form a syzygy, like the second and third radials. * This agreement between the mode of union of the second and third radials, and of the first and second brachials respectively, is seen also in Hncrinus moniliformis, in which these segments are united by syzygia asin P. Miilleri. See ‘ Petref. Germ.’ Taf. liv. figs. F, G. The same is the case in Rhizocrinus (Sars, ‘ Crinoides vivants,’ pp. 15, 22). * Bau des Pentacrinus, p. 30, and Taf. ii. fig. 8. This does not appear, however, to be always the case ; for Miiller described the syzygium as uniting the first and second arm-segments in the specimen examined by him, while I have found the same to be the case in a specimen of this species contained in the Zoological Museum of the University of Wiirzburg, in which there is certainly no syzygial union between the second and third radials. In Pentacrinus the opposed faces of two elements, which are united by a syzygium, are simple, and not radially striated as in Antedon. Sars has found this to be the case in Ehizocrinus also; but in its predecessor, Apiocrinus obconicus, Goldfuss, the radial striation of the syzygial surfaces is very distinct (Petref. Germ. Taf. lvii. fig. 5). 5 The ‘Challenger ’ collection includes two very abnormal species which present a singular exception to the rule. The rays divide three times ; and the first two segments (distichals) of each of the ten primary arms are united by ligaments only, like the second and third radials. So far the rule holds good; but with the next arm-division there is a new point of departure. The third or axillary distichal bears the secondary arms, which consist of one axillary segment only. But this segment is itself primitively double, i. ¢. it consists of two parts united by a syzygy; and the first joints of each of the ultimate arms borne by tis axillary a gree with it in being syzygial segments. 24. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, of the free and undivided secondary or tertiary arms is of precisely the same nature as the union of the first and second segments of the primary arms borne by the radial axillary. In fact, it is not at all uncommon for one of the primary arms to remain simple and the other to divide, as in Pl. IT. fig. 9, c, D, B, which shows that the arms taken in the strict sense of the word, cannot be regarded as commencing from any point but the axillary radials. (§ 17) In practice, however, it is more convenient to regard the arms of Comatula as beginning from the last bifurcation, i.e. from that axillary the two branches borne by which do not further divide, but remain composed of a series of simple brachial seg- ments (,, b.),&e. Inthe ten-armed Comatule the brachials are, of course, borne directly by the radial axillary. But in those forms, such as Act. multiradiata, in which the sub- division of the ten primary arms is carried to a very great extent, it is most convenient to regard as brachials only the segments of the ultimate branches borne by the last axillaries, and to give special names to the segments composing the primary and secondary arms; for we have already seen that the number of the segments composing these arms, 7. e. between every two successive axillaries, varies in different species, and it consequently becomes desirable to have some system of nomenclature by which these differences can be briefly indicated. Under these circumstances, therefore, the term distichals may be applied to the segments composing the ramified primary arms of e Articulate as well as of the Tessellate’ Crinoids, but only on the distinct under- standing that they are really arm-segments and do not enter into the formation of the calyx, as in the Tessellata; so that the name is purely a conventional one, employed for greater convenience in the description of species. Supposing the secondary arms borne by the distichal axillaries to divide again, we may consider them as composed of two, three, or four palmar” segments, of which the last is a palmar axillary ” (figs. 10,11, p.a.), and bears two tertiary arms. These may either remain simple and composed of brachial segments, or they may continue to divide more or less frequently. The latter case, however, is somewhat rare ; for if complete series of distichals and palmars be developed on each radius, the total number of arms rises from 10 to 40; and there are not many Comatu/e in which this number is exceeded. If we apply this nomenclature to the species of which mention has already been made, we should describe Antedon Savignii with 20 arms, as characterized by the presence of three distichals composing each primary arm and bearing the brachials directly, while in Antedon palmata, with 30 or 40 arms, there are only two distichals, which are fol- lowed by two palmars in the secondary arms. Act. polymorpha, again, normally has three distichals and three palmars (Pl. II. figs. 7, 10), while Act. multiradiata, Mill, has 1 J. e. As understood by MM. de Koninck and Le Hon. 2 T have been accustomed for some years past to use the term “palmar” to designate the secondary arms of the Crinoids. Professor Huxley, in whose lectures I first heard it used in this sense, informs me that he believes it to haye been so employed by Miiller; but I have searched in vain through Miiller’s works for any definition of the term. In his description of the Tessellate Crinoids, however, he describes the plates which continue the series of inter- radials and interaxillaries in a peripheral direction as ‘ interpalmaria;” and as these partially correspond to the intervals between the secondary arms, when such are developed, the latter may perhaps be not incorrectly regarded as is not a very suitable designation ) composed of “ palmar” segments. It will now be apparent why ‘ interpalmar ’ for the strictly “ interradial” areas on the disk of Comatula, as was remarked in section 15. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 25 only two palmars in each secondary arm, although the number of distichal segments in the primary arnis is usually three. Another proof, if proof were wanted, that the arms proper of the articulate Crinoids begin from the axillary radial is seen in the fact that whenever there are three segments in a distichal or palmar series, the second of these, which is united to the first by ligament only, always bears a pinnule, while the third, or axillary, is a double segment, ¢.e. it consists of two primitive segments united by a syzygium. ‘This is in precise accordance with what we find in all the ten-armed Comatule, even in those forms in which, as in Acé#. solaris, the first and second brachials are united by asyzygium like the second and third radials. In these the second brachial or the epizygial element of the syzygium bears a pinnule, while the next segment is also a double one, and corresponds with the compound third brachial of Ant. rosacea and of the ordinary ten-armed Comatule. In these last the second brachial is laterally movable upon the first, and bears a pinnule as in Acé. solaris, while the third has a syzygium (i. e. is a double segment). This is exactly what we find to be the case in those primary and secondary arms of the multiradiate Comatule which consist of more than two segments. ($18) The principal character of the genus Actinometra is, as we have seen in sect. 14, that the mouth is situated excentrically, while the centre of the disk is occupied by the anal tube. The position of the mouth relatively to the radii or ambulacra, however, is not the same in all Actinometre; thus in Act. solaris (Pl. I. figs. 2, 5) the mouth lies in a radial or ambulacral plane, while in Act. Wahlberghii and many other species (PI. I. figs. 3, 4, 6-16) it is interradial or interambulacral. If we place the disk of an ordinary Antedon in such a position that the interradial area containing the anal tube is nearest to us (Pl. I. fig. 1), the odd ambulacrum lies in front of the mouth. Let us designate this as ambulacrum or radius A, and the two branches of its groove-trunk corresponding to the two primary arms as A, and A, respectively, A, being that on the left of the mouth, Proceeding round the disk in the direction of the hands of a watch, we may call the other four ambulacra B, C, D, E respectively, and their primary divisions B, B, ....-H, E,. The anal area is then bounded by the two postero-lateral ambulacra C, D; and a plane passing through the mouth and anus, so as to divide the disk into two symmetrical halves, passes along the odd ambulacrum or radius A, in front of the mouth, which may therefore be regarded as radial in position. In Act, solaris the same is the case, as may be seen in Miiller’s somewhat diagrammatic figure (Pl. I. fig. 2)’, and still better in Pl. I. fig. 5, which was drawn from a spirit- specimen, and not from a dry one like Miiller’s figure. Here, as in Antedon, the odd ambulacrum is in front of the mouth, which, although excentric in position, lies in the radial half of a plane which passes through the mouth and anus, so as to divide the disk into two symmetrical halves. The same is the case in a new Actinometra from the 1 Tt is not usual to meet with specimens of Actinometra in which the branches of the ambulacral grooves are distributed with such symmetry as is represented in Miiller’s diagrams (V’l. I. figs. 2-4) and in Pl. I. fig. 1. Thus, for example, Muller’s figure of Act. solaris (Pl. I. fig. 2) is remarkably regular, much more so than that represented in fig. 5; and I have examined other specimens with more than 20 arms and a radial mouth, in which the regularity is by no means so distinct as in Pl. II. fig. 1. A great range of variation in this respect is seen in PI. J. figs. 6-16, which represent the disks of eleven different individuals of Act. polymorpha, no two of which are alike; the position of the mouth, however, is constant in all individuals of the same species. SECOND SERIES.—ZOOLOGY, VOL. I. 4 26 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. Philippines (PI. II. fig. 1) with 23arms. In both these species the dividing plane passes in front of the mouth between the two primary divisions, A, A, of the odd anterior ambulacrum A, while behind the mouth it is interradial, and separates the two postero- lateral ambulacra C, D. In Act. Wahlberghii (P\. I. fig. 3), Act. multiradiata (fig. 4), and Act. polymorpha (figs. 6-16) the case is different. If, as in Antedon, we designate the two ambulacra bounding the anal area as C and D respectively, we find that the latter is the odd ambulacrwm, and that a plane cutting the mouth and anus is radial behind the mouth, in front of which it passes between the two ambulacra A and B; so that if the centre of the disk be regarded as the centre of radiation, the mouth lies in an interradial or inter- ambulacral plane. This is clearly seen when we turn to the dorsal side of the disk, in which the radii converge to a central point, and not to an excentric one, like the ambu- lacra of the ventral side. Thus in Pl. II. figs. 9-11, the position of the mouth relatively to the radii is indicated by a x, which in each case is between the two anterior radii A and B, or interradial. So far as my experience goes, this type of Actinometra, in which the mouth is inter- radial and the odd ambulacrum lies behind it, is slightly more common than the simpler type, in which the mouth is radial and the odd ambulacrum anterior, as in Antedon'*. ($ 19) We are now in a position to investigate which of the numerous species of Comatule described by Miiiler belong to the type of Axtedon, and which to Actinometra, and into what groups the latter may be divided according to the principles of classifica- tion discussed in the last four sections. Miiller’s specific descriptions afford very little information in this respect; for though he says that the mouth, in some instances, is excentric, and that in others the oral pinnules have a terminal comb, he does not always do so; and he makes no use whatever of these two characters in his distinction between Alecto and Actinometra. In some cases he simply designates a species as Comatula, without attemping to name it more exactly. This is often, no doubt, simply due to the fact that, in the specimens which he examined, the arms were so closed over the disk that he was unable to investigate the distribution of the ambulacra. This, however, is not the ease in the C. trichoptera of the Paris Museum, the disk of which can be readily ex- amined; and I believe that Miiller did not define this species more precisely because he was unable to decide whether it should be referred to Alecto or to Actinometra ; for in one of the two Paris specimens five groove-trunks start from the excentric peristome, while in the other there are only four. This example alone suffices to show the unsatis- factory nature of the only distinctive character established by Muller between his genera Alecto and Actinometra, ‘The Paris Museum?’ contains a very large majority of the 1 Since the above was written, I have examined three large Comatula-collections :—(1) that of the ‘ Challenger ;’ (2) that made by Prof. Semper in the Philippines; and (3), thanks to the kindness of Dr. Giinther, that in the British Museum. I haye been able to determine the position of the mouth in 80 species of Actinometra. In 45 of these it is inter- radial, as in Act. polymorpha ; while in the remainig 35 it is radial, as in Antedon and in Act. solaris. * [ would here express my most hearty thanks to Mons. Edmond Perrier, Assistant-Naturalist at the Museum of Natural History, Jardin des Plantes, who has charge of the Echinoderm collection, and also to his two Assistants, for the kindness which they showed me during my stay in Paris, and for the readiness with which they afforded me every possible facility in the prosecution of my work, MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 27 - species of Comatula described by Miiller, who personally examined this collection. Last autumn (1876) I was also able to examine it for myself, and thus to determine which species should be removed from Miiller’s genus A/ecto and transferred to Actinometra, in the sense in which this name is understood by Dr. Liitken and myself. In the following scheme all those species to which no note of interrogation and the name of no authority is attached, have been determined by myself as Actinometre, i.e. as having an excentric mouth and a terminal comb on the oral pinnules. Actinometra. Act. solaris, Mill. Second and third radials ce - oie ll as ao ct. brachwwlata. unite Y, % 8Y2Y8Y-* Com purpurea? 1 Mouth radial. ies oe ta j; Com. rosea ? With 10 arms. \ Act. robusta, Liitken. united by ligaments only. Mouth ? Second and third radials { Com. echinoptera? 2. Com. meridionalis? 3. Act. rotalaria, Mull, 4, Ce ( 2 Distichals. 3 palmars. Axillary has no 4 Axillary has a 8YZYSy- syZysy- Act. polymorpha. ' Mouth interradial. 4 Act. trichoptera. | Tope Act. Wahiberghii, Mill. \. Mouth radial .... Act. fimbriata. 5. _ With more 2 palmars. ) than 10 Axillary has Act. multifida. 6. arms. no syzygy. / 3 Distichals. Axillary hasa< 2 palmars. Second and syzygy. Axillary has third radials a syzygy. Mouth interradial. Act. multiradiata. united by ligaments only. 3 palmars. | Act. parvicirra. 7. axillary has a Act. polymorpha. 8. syZzygy. ‘Act. Bennettii, Bohlsche. 9. L REMARKS. 1. I have not personally examined either C. purpurea or C. rosea. Miiller seemed to think that the former might be a young condition of Acé. solaris'; so thatit is most pro- bably a true Actinometra. C. rosea, however, presents a difficulty; for Miiller says expressly” that the first pinnules are not specially distinguished; but, except in this point, he regarded it as very closely related to C. brachiolata, which is a true Acti- nometra. 1 <«Gattung Comatula, p, 13. * Ibid. p. I4. 4* 28 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 2. C. echinoptera, on the other hand, has, according to Miiller’, a very marked comb on the oral pinnules. He says nothing, however, about the position of the mouth ; and I have unfortunately not been able to examine the species for myself. 3. According to Pourtales* the mouth is excentrie in Com. meridionalis ; but he makes no mention of a comb on the oral pinnules. If it should be absent in this species, and also in Com. rosea, while it is present in Phanogenia, in which the mouth is central, it becomes obvious that the only external character, besides the shape of the calyx, on which we can rely with any certainty in the determination of the generic position of any Comatula, is the nearly central or the excentric position of the mouth. 4, According to Miiller® there are only two radials in Act. rotalaria which are united by a syzygium, while they bear the distichal axillaries directly ; and these are also syzy- gial segments. Although, like Miiller, I examined Lamarck’s original specimen of this species, I cannot confirm the above statement. It is true that only two radials are visible externally; but this is often the case in Comatule, with a wide centrodorsal piece; and Iwas quite unable to satisfy myself that they are united by a syzygium, as Miiller describes, and as is the case with the second and third radials of Act. solaris, while I was equally unable to determine a syzygial union between the two segments of which the distichal series is composed. Lamarck’s original specimen of this species in the Paris Museum is wrongly labelled C. brevicirra, Troschel. 5. The dry specimen of Act. jimbriata in the Paris Museum, from the voyage of Peron and Lesueur, is labelled C. multiradiata, Lam.; while Reynaud’s original speci- mens from the Strait of Soude are labelled C. brevicirra, Trosch., together with Act. rotalaria and the Vavas variety of Miiller’s C. parvicirra. 6. In Act. multifida the tertiary arms borne by the palmar axillaries may divide again several times. In every case there are only two segments between each division, and all the successive axillaries, like the palmar axillaries, have no syzygia. Although Miiller mentions this in his description of the species, it is placed in his scheme in a group in which the axillaries of the arms have syzygia. In reality, however, this is the case with the distichal axillaries only. 7. The type specimen of this last species does not exist in the Paris Museum under that name, nor even under C. brevicirra, Troschel, which seems to have been used as an equivalent for it; but I believe that three small spirit-specimens from the voyage of Peron and Lesueur in 18038, which are classed, with two specimens of Act. pectinata, under the name of C. simplea, are really those which were described by Miller as C. parvicirra. 8. It will be noticed that Act. polymorpha has already appeared higher in the list as a species in which palmars are not developed. In some individuals but few of the 10 primary arms bear axillaries; so that the total number of arms is less than 20 (PI. IT. fig. 9); while in others all the primary arms divide again, and so do the resulting secon- dary arms, so that the total number is little short of 40 (Pl. II. figs. 10,11). This is very unusual; for I only know of two other species which present the same kind of 1 *Gattung Comatula, p. 14. * Loc. cit. No. 11, p. 355. 5 *Gattung Comatula,’ p. 20. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 29 irrecularity. Asa general rule all the individuals of a species agree in the presence or absence of distichals and palmars. 9. In Act. Bennettii there are more than 70 arms; but all the axillaries are like the first one (distichal), and not different from it, as in Act. multifida. According to Miiller’, every fourth segment is an axillary without a syzygium; but BohIsche’ has found this to beincorrect. There are, indeed, four segments between every two points of division; but the last two are united by asyzygium; so that the formula becomes three distichals, palmars, &e., of which the axillary has a syzygium. Béhlsche’s figure of the disk of his specimen is noteworthy; for though five groove-trunks leave the excentric peristome, as in Alecto, yet their distribution to the arms is not by any means symmetrical, so that he seems to have decided upon calling it Actinometra. Miller nameca it simply Comatula. (§ 20) In the above scheme are included all the species of Comatula which have been determined by myself or by others*, as far as I know, to belong to the type Actinometra. Fourteen of these were known to Miiller; and of the remaining 28 species described by him I have been able to refer 16 to Antedon, viz. :— Ant. adeone. Ant. macrocnema. Ant. phalangium. Ant. articulate. Ant. Milberti. Ant. Philiberti. Ant. carinata. Ant. Milleri. Ant. Reynaudii. Ant. Eschrichtii. Ant. palmata. Ant. rosacea. Ant. Jacquinoti. Ant. petasus. Ant. Sarsii. Ant. Savignit. To these must be added Ant. armata, Pourt. Ant. cubensis, Pourt. Ant. Hagenii, Pourt. Ant. bicolor, Mus. Paris. Ant. dividua, Mus. Paris. Ant. polyactinis, Mus. Paris. Ant. celtica, Barrett. Ant. Dubenii, Bolsche. Ant. rubiginosa, Pourt. The following list contains the seven remaining species of Comatula described by Miiller which I have not been able to examine, and of which I know no descriptions from which it is possible to obtain any information as to the position of the mouth or the character of the oral pinnules. C. Cumingii. C. elongata. C. flagellata. C. japonica. C. nove Guinee. C. tessellata, and C. timorensis. To which must be added C. brevipinna, Pourt. ILI. Exrernat CHARACTERS OF ACT. POLYMORPHA, AND SPECIFIC DIAGNOSIs OF THE TYPr. (§ 21) In Act. polymorpha, as in all Actinometre, the mouth (Pl. I. figs. 6-16, m) 1 «Gattung Comatula, p. 28. wo “Ueber Actinometra Bennettii und eine neue Comatula-Art (Antedon Dubenii),” Wiegm. Archiy, 1866, i. p. 90. ’ Dr. Liitken has named several new species of Actinometra besides Act. robusta—for example, Act. tena and Act. trachygaster. But his descriptions have not, as far as I know, been published; and I have had no opportunity of examining any specimens of his new species except Act. robusta; so that I am unable to place them in the classi- fication given in the previous section. Grube has described a new Actinometra from Borneo, and two new species of what he calls Comatula. His de- scriptions (Jahresber. d. Schlesisch. Gesellsch. 1875, Nat.-Hist. Sect. pp. 54, 55) are, unfortunately for me, not to be obtained in this country. 30 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, does not occupy a central or subcentral position on the ventral surface of the visceral mass as it does in Antedon, but is placed more or less excentrically, and may be some- times almost marginal (Pl. I. fig. 11). It occupies the centre of the peristome, P, and is bounded by two lips, a large anterior and a smaller posterior one ; so that its opening is very inconspicuous, and usually so much extended in a direction transverse to the antero-posterior diameter of the disk, that it presents the appearance of a simple slit, as is well seen in Pl. IT. fig. 2. The circumoral portion of the peristome, or the peristome proper, is a more or less oval depression in the ventral perisome of the disk, which completely surrounds the oral opening, and gives origin to the ambulacral grooves or, more shortly, the ambulacra. Beneath this depression lies the water-vascular rmg which gives off a trunk under each of the ambulacra radiating from it. The number and distribution of these are very variable, as is seen in Pl. I. figs. 6-16. This principally depends upon the way in which the ambulacra divide, so as to give rise to the groove-trunks corresponding to the ten primary arms. Asa general rule, the two ambulacra corresponding to the radii D and E unite into one large posterior trunk, from which the branches are dis- tributed to the various arms into which these radii divide (Pl. I. figs. 8-10, 12-16). In other cases the left lateral ambulacrum, E, leaves the peristome alone (figs. 6,11) ; while in others it is partially united with the posterior ambulacrum, D, its anterior division, E,, leaving the peristome by a separate trunk, while its posterior division, E,, unites with the posterior ambulacrum (fig. 7). : As a general rule, the right lateral ambulacrum, C, leaves the peristome alone, and supplies the arms of the corresponding radius ; but in figs. 9 & 15 itis seen to unite with the posterior division, B,, of the right anterior ambulacrum, B. The mode of division of the two anterior ambulacra is excessively variable. As a general rule there are no principal trunks corresponding to the two radii A and B, and the primary divisions, A,, A», B,, B., start directly from the peristome. In the specimens with but few arms, however, each pair may be united for a longer or shorter distance (Pl. I. figs. 6, 7), as in Antedon (fig. 1). Not unfrequently the posterior divisions A,, B., of these two anterior ambulacra unite for a longer or shorter distance with the two large aboral groove-trunks, to form an open horseshoe-shaped curve bounding the anal area (figs.12, 15, 16). The position of the anal tube in this area, and also with regard to the whole surface of the disk, varies somewhat with the position of the mouth; it is rarely, if ever, absolutely central. Its appearance differs very much according as it is full or empty: sometimes its aperture is so completely closed as to be scarcely discernible, though the tube below is widely distended; and sometimes the aperture is patent with its edges everted and crenate, and the tube leading to it quite shrunk and flaccid (Pl. IT. fig. 2). (§ 22) In Antedon the median line of the ventral perisome of all the arms is occupied by an ambulacral groove, with a floor of ciliated epithelium. This groove extends also on to all the pinnules, with the exception of those borne by the second distichals and second palmars, &c. (when present), and by the lowest brachial segments. Beneath it lie the radial water-vascular and blood-vascular trunks, between which last and the MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 31 ciliated epithelium of the floor of the groove lies a fibrillar structure, to which I have given the name of the “ subepithelial band’’', and to which a nervous character has been attributed by myself and by all the other observers who have described it. Each side of the ambulacral groove is bounded by an elevated fold of perisome, the edge of which is not straight, but cut out into a series of minute valvules, the crescentic or respiratory leaves (Wyv. Thomson), or “‘ Saumliappchen”’ of the German authors. At the base of each leaf, and to some extent protected by it, is a group of three tentacles, one of which, the more distal one, is larger than the other two. This trifid group of tentacles and the cavity of the respiratory leaf adjacent to them receive a common branch from the radial water-vessel. These groups of tentacles alternate on the opposite sides of the ambulacral groove from the base to the tip of each arm, and are distributed in the same manner at the sides of the ambulacra of the disk, though they are not so markedly developed, especially near the peristome, where every lateral branch of the water-vessel supplies only one tentacle. The crescentic leaves at the sides of the groove are also far less distinct than in the arms, the edges of the folds of perisome bounding the groove being only marked by a faint wavy line, and not distinctly cut out into “ Saumliappchen.” In many Actinometre, however, the above description only applies to the arms of the two anterior radii, A, B (Pl. II. figs. 2, 4), and to more or fewer of the antero-lateral _‘arms, C, and E,. The arms of the posterior radius, D, and of the posterior divisions of the lateral radii, C, and E,, are often entirely devoid of tentacles; and in many of them the ventral perisome not only exhibits no ambulacral groove, but is, on the contrary, convex, as in the oral pinnules of Antedon (Pl. IT. figs. 5, 6). We have just seen that in Act. polymorpha, as in all Actinometre with an interradial mouth, the anal area is bounded by two large aboral groove-trunks, which start from the posterior angles of the peristome, and form a horseshoe-shaped curve, the limbs of which are unequal in size (Pl. II. fig. 2). The smaller right limb is formed by the right lateral ambulacrum, C; while the larger left limb represents the posterior ambulacrum, D, com- bined with part or the whole of the left lateral ambulacrum, E. In neither of these limbs are the tentacular groups and crescentie leaves so well developed as they are in the two anterior ambulacra. After the branches to the two antero-lateral primary arms, C, and E,, have been given off, or sometimes even sooner (Pl. I. figs. 18, 15), the ten- tacles at the sides of the two aboral groove-trunks become more and more insignificant, and finally disappear altogether, while the position of the crescentic leaves is only indicated by a very faint wavy line at the edge of each groove. In small specimens with but few arms (Pl. I. figs. 6, 9) the grooves of the posterior (D) and postero-lateral arms (C,, E,) may remain in this condition; but in larger speci- mens with many arms all trace of the crescentic leaves disappears, and the two edges of the groove meet and unite so as to produce the condition represented in Pl. II. figs. 5 & 6, where the ventral surface of the arms and pinnules is convex, and does not show the least trace of a groove of any description. “ Remarks on the Anatomy of the Arms of the Crinoids. Part I.,’ Journ. of Anat. and Physiol. vol. x. p. 579. 32 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. The position of the point at which the two folds of perisome bounding the sides of the original ambulacral groove meet and unite, varies extremely. The fusion may, though rarely, take place on the disk; sometimes it is at the base of the arms, and sometimes not till near their middle or terminal portions. In any case, however, the fusion, when- ever it occurs, is so complete that all trace of the original ambulacral groove is entirely obliterated. (§ 28) The bearings of this fact upon the different views advanced by Greeff’ and Ludwig? respecting the homologies of the ambulacral grooves of the Crinoids will be best discussed at a later period, when the changes undergone by the various structures underlying the grooves are described and illustrated. One point, however, must be noticed here on account of its importance with respect to the two views now entertained regarding the nervous system of Comatula. As long ago as 1865 it was stated by Dr. Carpenter * that the cord which traverses the length of the arms between the subtentacular and cceliac canals, “ and which was regarded by Professor Miiller as a nerve, really belongs to the reproductive apparatus. But it will also be shown that a regular system of branching fibres proceeding from the solid cord (described by Professor Miiller as a vessel) that traverses the axial canal of each calcareous segment of the rays and arms, is traceable on the extremities of the mus- cular bundles ; and reasons will be given for regarding these fibres as probably having the function of nerves, though not exhibiting their characteristic structure.” During his residence in the Philippine Islands, Professor Semper had also discovered that the arm-nerve of Miiller is really a part of the generative system; and in a short paper * published some time after his return he announced this fact, and suggested at the same time, “dass der bisher immer als Gefiss aufgefasste Strang im innern des Kalkskelettes ein Nervenstrang sei, und dann wire wohl das im Kelch liegende sogenannte Herz als ein Ganglion anzusehen.” These observations of Dr. Carpenter’s and Professor Semper’s were unfortunately overlooked for many years, so that even as late as 1874 Miiller’s erroneous statements with regard to the nervous system of Comatula were repeated in the valuable text-book of Gegenbaur® and in many smaller works. At the commencement of 1876, however, two very different views respecting the nervous system were put forward nearly simul- taneously by Greeff and by Dr. Carpenter. The former’ described the whole floor of the ambulacral grooves on the arms and disk of Ant. rosacea as constituting a radial neryous system, starting from an oral nervous ring in the peristome, and corresponding “Ueber den Bau der Echinodermen. III. Mittheilung,” Sitaungsb. der Gesell. z. Beforder. d. gesamm. Naturwiss. zu Marburg, 1872, No. 11, p. 155. 2 « Beitr. z. Anat. der Crinoideen,” Nachrichten yon der Kénigl. Gesells. der Wissens. u. der G. A. Universitit zm Gottingen, 1876, No. 5, pp. 107, 108. * « Researches on the Structure, Physiology, and Development of Axtedon rosaceus. Part I.,” Philos. Trans. yol. elvi. p. 705. 4 «“ Kurze anatom. Bemerk. iiber Comatula,” Arbeit. aus d. zool.-zootom. Inst. zu Wiirzburg, Band i. (1874), p. 262. ® « Ueber den Bau der Crinoideen,” Marburg. Sitzungsb. 1876, No. 1, Jan. 13, p. 21. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, 33 in position and histological structure with the typical Echinoderm nerves. At the same time he denied the nervous nature of both the structures described as nerves by Miiller and Dr. Carpenter respectively, viz. the genital cord, the so-called “ rachis,’”’ on the one hand, and the axial cords in the centre of the calcareous segments on the other. A week after the publication of Greefi’s views, Dr. Carpenter’ announced his belief that a complicated apparatus, “ consisting of the outer cylinder of the Crinoidal stem, of the five-chambered central organ formed by the dilatation of that axis within the centro- dorsal basin, and of the cords proceeding from it to the arms and cirrhi,’”’ should be regarded as the central portion of a nervous system. This view was based both upon anatomical and upon physiological considerations :— (a) That while a single arm may be made to coil up by irritating one of its pinnules, the whole circlet of arms closes together when an irritation is applied to the pinnules, which arch over the mouth—an act which affords a strong indication of the “ inter- nuncial” action of a definite nervous system. (0) That stimulation of the central quinquelocular organ (“ heart” of Miiller and Greeff) contained in the calyx, with which the axial cords of the arms are in connexion, is followed by sudden and simultaneous flexion of all the arms. (c) That these axial cords give off successive pairs of branches, which ramify upon the muscles connecting the arm-segments. Shortly after the announcement of these views on the part of Dr. Carpenter, Ludwig? described a ventral nervous system as existing in Comatula in common with all the other Echinoderms. He attributed a nervous character, not to the whole epithelial floor of the ambulacral grooves, as was done by Greeff, with whose researches he was unac- quainted, but to a fibrillar layer beneath it, and more or less distinctly separated from it. This layer, which was also discovered independently by myself* and Teuscher*, and was regarded by us both as of a nervous nature, is the “ subepithelial band” men- tioned in sect. 22. Ludwig, like Greeff before him, denied the nervous character of the dorsal axial cords of the arms; Teuscher discussed it as possible, but hesitated to accept it on account of the morphological difficulties involved in such a view. Baudelot®, who seems to have been unacquainted with Dr. Carpenter’s earlier state- ments, was apparently struck with the nature of these cords, though he could not regard them as nervous. After describing their structure and their union in the calyx to form the pentagonal commissure, he adds, ‘‘Ainsi donc chez les Comatules il existe des parties qui évidemment n’appartiennent point aw systéme nerveux (!), et qui dans leur disposition aussi bien que dans leur structure offrent une analogie presque compléte avec les cordons nerveux des autres Echinodermes.” It must be remembered that Baudelot wrote before the discovery of the so-called ‘ventral nerve” of Comatula; but, in any case, I do not quite see the force of his “ évidemment.” 1 «On the Structure, Physiology, and Development of Antedon rosaceus,” Proc. Roy. Soc. no. 166, Jan. 20th, 1876, pp. 219-226. 2 Gottingen Nachrichten, no. 5, Feb. 23rd, 1876, p. 106. 3 Journ. Anat. Phys. x. p. 578. 4 « Beitr. z. Anat. der Echinodermen, I. Comatula mediterranea,” Jenais. Zeitsch. B. x. p. 253. 5’ “ Contribution 4 histoire du syst¢me neryeux des Echinodermes,” Arch. de Zool. Exp. et Gén. i. p. 211. SECOND SERIES.—ZOOLOGY, VOL. II. 5 34 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. In the centre of every segment of the skeleton of Act. polymorpha and of all the other Comatule which I have examined, from the first radials to the ends of the arms and pinnules, and also in the cirrhus-segments, these axial cords increase considerably in size, and give off four principal branches. ‘Two of these run towards the ventral side, and in the calyx disappear in the neighbourhood of the muscles connecting the segments, though I must confess that I have never been able to trace them any further (Pl. VIII. fig. 3, n'). In the arm-segments, however, they continue their course towards the ventral surface and break up into numerous branches, some of which, as I have already described’, extend to the tips of the crescentic leaflets at the sides of the tentacular furrow. The two inferior or dorsal trunks run towards the surface of the skeleton; and while some of their branches are lost in the plexus of tissue forming its organic basis, others seem to become connected with epidermie structures in a manner which will be described at length further on. Not one of the German observers makes any mention of these branches, although two of them at least have examined Antedon Eschrichtii, while they have all cut sections of the arms of species of Actinometra, in which genus I find them to be particularly distinct. It is obvious that the facts above stated strongly support the view expressed by Dr. Car- penter and by myself, that the axial cords of the arms are of a nervous nature; and the experiments made by Dr. Carpenter’ at Naples have shown conclusively :— 1. That the quinquelocular organ is the instrument of the perfect coordination of the swimming movements of the arms, which involve the conjoint contraction of several hundred pairs of muscles. 2. That nothing contained in the visceral mass is essential to the perfect coordination of the swimming-movements, and that therefore the subepithelial band or ambulacral nerve of the German authors has no immediate relation to those movements, even if it be a nerve at all. 3. That section of the subepithelial band in an arm does not prevent its playing its usual part in the regular swimming-movements. 4. That destruction of the axial cord of an arm by the application of acid causes the arm to become rigidly stretched out, while all the others work as usual. Since the publication of these experiments Greeff seems to admit the nervous nature of the axial cords, and of the yellowish fibrillar envelope (Pl. VIII. figs. 1-8, N) of the quinquelocular organ from which they proceed. Ludwig*, however, while allowing their force, cannot admit the existence in the Crinoids of an antiambulacral nervous system, of which we know as yet no homologue in the other Echinoderms, but sees no difficulty in regarding the quinquelocular organ, its fibrillar envelope, and the axial cords proceeding from it, as parts of a blood-vascular system, like that of the other Echinoderms, although he admits (p. 87) that “ihnen vergleichbare Gebilde sind bis jetzt bei anderen Echinodermen nicht bekannt geworden.” The axial cords of the 1 Journ. Anat. Phys. x. p. 584. 2 «Supplemental Note to a Paper on the Structure, Physiol. and Develop. of Antedon rosaczus,” Proc. Roy. Soc. no. 169, 1876. 2 « Beitr. zur Anat. der Crinoideen,” Sep.-Abdruck aus der Zeitsch. f. wissensch. Zool. B. xxviii. Heft 3, p. 81. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 35 calcareous segments are regarded by him (pp. 80, 86) as “ unverkalkt gebliebene Theile der bindegewebigen Grundlage der Kalkglieder, deren Aufgabe es ist, aus dem Blutge- fasssystem, genauer aus den fiinf Kammern die kernnihrende Fliissigkeit aufzunehmen und den Arm- und Pinnulagliedern zuzufiihren.” Without going into the question as to how far the organic basis of the calcareous skeleton can be regarded as of a connective-tissue nature, I would only remark that it is difficult to see why the axial cords, which Ludwig supposes to consist of uncalcified connective-tissue fibres, should give off branches the terminations of which are entirely outside the skeleton, as is the case with those which reach the crescentic leaves at the sides of the tentacular groove, and which therefore cannot take any part in the nutrition of the tissue forming the organic basis of the skeleton. (§ 24) This is not the place for a full discussion of Ludwig’s views on the nervous system of Comatula; but one point must be briefly referred to. I have already’ stated that in some arms, and in most of the pinnules, of many Actinometre, the subepithelial band or nerve of Ludwig is entirely absent, and also that “if the axial cords are not nerves, and if the subepithelial bands are to be regarded as the only nervous structures in the whole Crinoid organization, the difficulty presents itself that the oral pinnules of the European Crinoids, and more than half the arms, with the majority of the pinnules of some forms of Actinometra, are entirely devoid of a nervous supply. *‘The oral pinnules of Antedon have been shown by Dr. Carpenter’ to be extremely susceptible of irritation; when they are touched in the living animal, the whole circlet of arms is suddenly and simultaneously coiled up over the disk, while irritation of one of the ordinary pinnules is simply followed by flexion of the arm which bears it. “The structure of these oral pinnules, which are borne in Antedon rosacea by the second brachials, differs very considerably from that of the pinnules borne by the other brachial segments; for not only are they sterile, but they have neither tentacular appa- ratus nor ambulacral groove, their ventral surface being slightly convex instead of being concave as in the ordinary arms and pinnules. This has been mentioned by Teuscher’* ; but he has omitted to state that the ordinary ciliated epithelium of the ambulacral groove, with its subjacent nervous layer and nerve-vessel, are also absent.” Ludwig entirely ignores this argument, althongh he confirms the above statement concerning the oral pinnules of Antedon; in the text he is obliged to confess that “ Fraglich est mir geblieben ob die oralen Pinnule einen Zweig des radiiren Nerven besitzen oder nicht” (p. 75); while his figure of a section of an oral pinnule (pl. xvii. fig. 55) entirely con- firms the statement quoted above, to which, however, he makes no reference. This condition, which is limited in Ant. rosacea to the oral pinnules, sometimes exists in whole arms and in all the pinnules borne by them in many species of Actinometra. Even in the arms which come off from the anterior or oral side of the disk the ambulacral groove does not give off regular branches to the pinnules borne by the third and succes- sive brachial segments; but a variable number of these first pinnules (sometimes only » « Remarks on the Anatomy of the Arms of the Crinoids, part ii.,” Journ. of Anat. and Physiol. vol. xi. October, 1876, p. 89. 2 Proc. Roy. Soc. no, 166, p. 226. 3 Jenaische Zeitschrift, x. p. 249. 5* 36 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. three or four, sometimes as many as forty) resemble in this respect the oral pinnules, their ventral surface being convex, and devoid of any ciliated epithelium or subepithelial band; while their water-vessel is simple, without any lateral extensions to respiratory leaves and tentacles. In these oral arms, however, branches of the ambulacral groove enter the pinnules sooner or later, so that the terminal ones are always provided with a distinct tentacular apparatus, while the floor of their median groove is of the usual cha- racter, consisting of a ciliated epithelium and a subepithelial fibrillar band. We have seen in sect. 22 that in many cases the ambulacral grooves going to the aboral arms become less and less distinct as they get further and further from the peri- stome, and that their tentacles diminish and finally disappear. At the same time the floor of the groove becomes very much reduced in extent, its epithelial layer thinner and thinner, and the subepithelial band almost invisible, until, in those cases in which the two sides of the groove meet and unite, the ciliated epithelium and subepithelial band disappear altogether. Consequently, when this union takes place on the disk, whole arms are entirely devoid of any nervous supply, if we suppose, with Ludwig, that the anti- ambulacral axial cords are not of a nervous nature, and that the ‘ subepithelial bands ” are the only nervous structures in the arms. In such cases it would naturally be expected that these arms would be incapable of performing the regular swimming-move- ments like those in which there is an open ambulacral groove and a subjacent ‘‘ ambu- lacral nerve;”’ but Professor Semper, who has kept Actinometre in his aquaria for weeks together, informs me that he never saw the least trace of any irregularity in the alternating movement of their arms while swimming. The gradual obliteration of the ambulacral grooves by the approximation and fusion of the elevated folds of perisome at their sides, which may occur to so great an extent in Actinometra, is found also at the ends of the arms and pinnules of Antedon Eschrichtii. Ludwig states (p. 75) that their terminal segments have no ambulacral groove or tentacles; and he gives a figure of a section through the end of a pinnule (pl. xiii. fig. 12), the ventral surface of which is convex, while there is no ciliated epithelium or subepithelial band (ambulacral nerve), although in the text Ludwig makes no mention of their absence. I have found the gradual obliteration of the groove in these cases to take place in precisely the same manner as in Actinometra, the only difference being that the point at which the sides of the groove meet and fuse is much further from the disk in the one case than in the other. If we suppose, with Ludwig, that the subepithelial band is the sole structure of a nervous nature in the whole Crinoid organization, it is difficult to understand the fact, which Ludwig himself admits (p. 10), that it gives off no branches except those which go to the tentacles. It is true that in the Ophiuridea the ambulacral nerve does give off branches which go to the muscles, besides those proceeding to the tentacles, as described by Lange’, Teuscher*, and Simroth*; but the researches of the first-mentioned observer render it very doubtful whether the representative in the Ophiuridea of the * « Beitr. z, Anat. und Histiol. d. Asterien und Ophiuren,” Morphol. Jahrb. ii. Heft 2, p. 241. * « Beitr. &e., I. Ophiurids,” Jenais. Zeitsch. x. p. 274. ~i * « Anat. und Schizogonie der Ophiactis virens, Sars,” Zeitsch. f. wissensch. Zool. xxvii. p. 473 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 37 subepithelial band of Comatula takes any part in the formation of these branches. Ludwig further admits that he has been quite unable to find any sense-organs at the ends of the arms or pinnules of Comatula like those which exist in the Asteridea, and, in discussing the views of Greeff, expresses it as his opinion (p.78) that “die subepi- theliale Faserlage, welche durchsetzt wird von fadenformigen Verlingerungen des dariiber gelegenen Epithels allein den Nerven darstellt.” There can, I think, be little doubt that this subepithelial band is of the same nature in the Crinoids and Asterids; and it is therefore very interesting that the nervous nature of this structure in the Asterids has recently been disputed by Lange’, who regards as nervous only some cellular masses separated from the subepithelial band by a lamella of connective tissue, and projecting into the lumen of the two nerve-canals. He believes these cell masses to swell into a large ganglionic mass beneath the pigment-spot; while, in his opinion, the sub- epithelial band, together with the ciliated epithelium and the cuticula, constitutes a pro- tecting integumentary layer. Lange finds a corresponding condition in Ophiura textu- rata, in which the radial nervous system is better developed than in the Asterids, and consists of a series of paired ganglionic masses, connected with one another by transverse and longitudinal commissures. On the ventral side of this ganglionated cord is a longi- tudinal band, which Lange regards as the homologue of the protecting integumentary layer forming the floor of the ambulacral groove of the Asterids, and which, as is uni- versally admitted, corresponds to the subepithelial band, epithelium, and cuticula of the ambulacral grooves of the Crinoids. Lange’s views have been partially accepted by Simroth’®; but the correctness of them is altogether denied by Teuscher’, who regards Lange’s nervous cell-masses in the Asterids simply as the “ geschichtetes Epithel”’ on the wall of the nerve-canals; while the terminal ganglionic mass under the eye-spot described by Lange is represented by Teuscher (pl. xix. fig. 22) simply as a “ bindegewebiges Polster.’’ Ludwig‘, too, speaks of the nervous cell masses as local thickenings of the epithelium of the nerve-canals, which are not present in every species. This is naturally a very strong argument against Lange’s views ; but Ludwig omits to apply similar reasoning to his own opinions regarding the Crinoid nerves. The subepithelial bands (his nerves) are not constant in every arm of many species of Actimometra. Still less do Teuscher and Lange agree about the nervous system of the Ophiurids; Lange’s ganglionic masses are described as artificial by Teuscher, who, as in the Asterids, regards as the nerve only the fibrillar structure repre- senting the subepithelial band of Comatula. The question is still an open one; andit is therefore of no slight interest to learn that the supposed ambulacral nerve, or subepithelial fibrillar band, is not always present in the arms of Comatula, and that even when it exists it is certainly not motor in function, even if it be a nerve at all’, * Morph. Jahrb. ii. 274. 2 Zeitsch. f. wiss. Zool. xxvii. pp. 556-560. * « Beitr. &e., III. Asteriden,” Jen. ertsch. x. p. 513. “ “ Beitriige zur Anatomie der Asteriden,” Zeitschr. fiir wiss. Zool. xxx. p. 191. ° It is worth noticing here that the “ ambulacral nerve” of Comatula must be derived either from the mesoblast or from the hypoblast of the embryo. It is developed immediately beneath the tentacular atrium of the pentacrinoid larva, which Gotte has shown to be the most anterior portion of the left peritoneal sac. This is lined by hypoblast, 38 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. (§ 25) We have seen in sect. 22 that in certain of the arms of Actinometra the water- vessels are simple tubes, like the integumentary water-vessels of the Molpadide, and are not in connexion with any tentacular apparatus. Whether the mouth be radial or interradial, the non-tentaculiferous arms are invariably the aboral ones; so that in the latter case they belong to the trivium (PI. I. figs. 6-15), and in the former to the bivium (Pl. I. fig. 5)". This last, however, is not always the case; for I have a specimen of Act. solaris in which an anterior arm (C,) of one of the two ambulacra of the bivium is tentaculiferous, while a posterior arm (E,) in the trivium has no tentacles; it is never- theless aboral in position, as may be seen from Pl. I. figs. 2, 5. In only one individual of Act. polymorpha (P1. I. fig. 15) have I found a non-tenta- culiferous arm on one of the two anterior radii (A, B.); but this was a very remarkable case. Out of 31 arms, 19 were entirely devoid of a tentacular apparatus; and in 15 of these the fusion of the two sides of the ambulacral grooves had taken place either on the disk or in the basal arm-segments, so that an ‘‘ ambulacral nerve” was wanting in nearly half the total number of arms. In the other four non-tentaculiferous arms the groove remained open for a short distance, and then closed in the manner above described. Three of these four arms constituted the anterior division (E,) of the left lateral ambulacrum ; but the fourth was the first arm of the left anterior ambulacrum (A,), and was borne upon the same palmar axillary as a well-developed ordinary tentaculiferous arm. Pieces of the middle portions of these two arms are represented in Pl. IT. figs. 3 and 5, and their terminations in figs. 4.and6. With this exception, I have invariably found the non-tentaculiferous arms on the aboral side of the disk; their number and distribution, however, vary extremely, not only in different species but in different individuals of the same species. Thus in Act. polymorpha, in Plate II. fig. 8, the former is as low as 3% of the total number of arms, while in fig. 15 it reaches 3%. Even in two individuals with the same number of arms it may be different; thus in figs. 8 and 9 it is ~% and 3} respectively, and in figs. 12, 13 it is $3 and 43. The individual represented in fig. 12 was also remarkable for the fact that one of its aboral arms belonging to the posterior division of the left lateral ambulacrum (E,) was tentaculiferous, while those on either side of it were not so. In all the specimens of the type of Act. polymorpha which I have examined, and in three of its varieties, of which I have, unfortunately, only single specimens, more or and appears to be separated from the hypoblastic epithelium lining the water-vascular ring by a remnant of the meso- blastic tissue which occupied the blastoceel of the Echinopedium. One or other of these two layers, the hypoblast lining the atrium, or the mesoblast between it and the epithelium of the water-vascular ring, must give rise to the “ambu- lacral nerve,” which cannot be in any way derived from the epiblast. I am inclined to believe that the “nerve” is most probably of mesoblastic origin, and that the remainder of the mesoblast (in this position) is converted into the muscular layer of the ventral wall of the water-vascular ring; while the blood-vascular ring is a remnant of the primitive blastoccel. Huxley (‘Anatomy of Invertebrata,’ p. 559) has suggested a similar origin for the nerye-canals (periheemal canals, Ludwig) of the Asterids. ‘ Tn all these figures (PJ. I. figs. 5-16) the tentaculiferous ambulacra are indicated by dark lines, and the non- tentaculiferous grooves by fainter lines, MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 39 fewer of the arms have no tentacular apparatus; but in the fourth variety (Pl. I. fig. 16) all the arms are of the usual character, with open ambulacral grooves fringed with erescentic leaves and groups of tentacles. I have found the same variation to occur also in Act. solaris. In this case the number of arms is limited to ten, which may be all ten- taculiferous, or from one to four of the aboral arms may have no tentacular apparatus. [ Note. February 1879.—No less than 23 out of 48 species of Actinometra brought home by the ‘Challenger’ have more or fewer grooveless arms. I have cut sections of these arms in two species, and have obtained the same results as with Act. polymorpha and Act. solaris. The “ventral nerve” and ambulacral epithelium are conspicuous by their absence, while the axial cords in the skeleton give off branches freely in the centre of each arm-joint, as I have already described for other species, both of Antedon and of Actinometra. ‘Two points are noteworthy. In one species one of the posterior ambu- lacra stops quite abruptly on the disk, and the two arms to which it would naturally have gone, with its “nerve,” tentacles, &c., receive no branches from any of the adjacent grooves to supply the deficiency. Again, in one of the largest Comatule I have ever seen (a ‘Challenger’ specimen from the Philippines) there are more, than 100 arms, many of which are both grooveless and nerveless, as I have found from sections. But these abnormal arms are not limited to the hinder part of the disk as is usually the case ; for there are several on each radius. | The distribution of the non-tentaculiferous arms in Acé. polymorpha varies, like their number, toa very great extent. In any case they always occur on the odd posterior radius, D (Pl. I. fig. 8); when more are developed they may occur on the posterior divisons, C, and E,, of the two lateral radii, C, E; and they may then be called postero-lateral (PL. I. figs. 6, 12-14) ; and when the proportion of non-tentaculiferous to tentaculiferous arms becomes very great, more or fewer of the antero-lateral arms, C,, E,, belong to the former class (Pl. I. figs. 7, 9-11, 13), while in exceptional cases non-tentaculiferous arms may occur even on the anterior radii (fig. 15, A). (§ 26) The condition of the ambulacral groove and of the tentacular apparatus is not the only point in which the anterior or oral may differ from the posterior or aboral arms. The former taper very slowly, contain far more segments, and are much longer than the latter, while the form of their terminal portions and of the pinnules which these bear is altogether different (Pl. I. figs. 4,6). When viewed from the dorsal side (PI. II. fig. 7) the basal portions of the two kinds of arms are precisely similar ; they widen slightly between the first’ and second syzygia, i.e. from the third to the tenth brachial, remaining uniform till the third syzygium on the fourteenth brachial, after which they begin to taper. Up to about the twenty-fifth or thirtieth segment the oral and the aboral arms decrease in width at about the same rate; but from this point onwards there is a great difference between them. The arms borne by the two anterior radii, A and B, taper very slowly, the length of their segments increasing considerably, while the breadth only diminishes very gradually; at the same time the middle and terminal pinnules, in which no genital glands are developed, become very long and filiform, and ‘remain so until the last few segments, when their length suddenly diminishes very considerably (Pl. II. fig. 4). 40 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. T have never been able to ascertain what is the precise mode of termination of these anterior arms; even when the arm ends in such a manner that there is no reason to suppose that its terminal segments have been broken off, its few last pinnules appear simply as immature, and the last pair are separated by a delicate prolongation of the arm-stem, on which no pinnules have been as yet developed. Dr. Carpenter‘ has found the same ‘‘ growing-points”’ at the ends of the arms of Axt. rosacea, all of which are of the same character as the oral tentaculiferous arms of Actinometra; and he was never able satisfactorily to determine the normal mode of termination of the arms. With the posterior arms of Actinometra, however, the case is different. From the twenty-fifth segment onwards they taper very rapidly, and instead of reaching a length of 145 millims., as the anterior arms with some 150 segments may do, they have only some 80 segments, and rarely attain a greater length than 60-70 millims. At the same time their terminal pinnules are little, if at all, longer than those of the middle portion of the arm (PI. IT. figs. 5, 6); and the centre of the dorsal half of each of their segments is occupied by a dark-brown egg-shaped body, of a peculiar cellular nature, which I have reasons for believing to be a sense-organ® (Pl. II. fig. 6, 0.0). These bodies commence to appear in the pinnules at about the beginning of the second third of the length of the posterior arms, and are continued to their extremities. The pinnules of the last few segments decrease very slowly in size; and the arm ends in an axillary segment which bears ¢wo pinnules of the ordinary character, each provided with the brown ovoid bodies or ‘* sense-organs”’ (Pl. II. fig. 6, 0.0). These bodies, which may occur, though but rarely, on one or more of the anterior tentaculiferous arms, do not exist in all the specimens of Act. polymorpha which I have examined. In three out of my eight specimens of the type they are entirely wanting ; and they are also absent in all the single specimens of the four varietal forms which I have investigated. I have also failed to find them in the non-tentaculiferous arms of Act. solaris’. Between these two kinds of arms, the long anterior ones on the radii A, B, witha wide ambulacral groove and a well-developed respiratory apparatus, and the short posterior ones of the radius D with a closed groove and no external respiratory appa- ratus, all possible forms of transition may occur. As a general rule, more or fewer of the antero-lateral arms, C, and E., are tentaculiferous ; but they never reach such a great length as the anterior arms, and their terminal pinnules are by no means so long and slender. At the same time the postero-lateral arms, C, and E,, although generally non- tentaculiferous, have, except in rare cases, a more or less open groove for the greater part of their length, which, while greater than that of the posterior arms of the radius D, is less than that of the antero-lateral arms of C, and E,; and their pinnules increase slightly in length from the middle till near the end of the arm. 1 phil. Trans. 1865, p. 723, plate xxxviii. fig. 4. * Journ. Anat. & Phys. vols. x. xi. loce. citt. 3 Sense organs occur in two of the ‘ Challenger’ species—one from Banda (which is probably the young of Act. polymorpha), and one (a new species) from the Admiralty Islands. In both cases they are limited to the hinder arms, some of which are grooved and others not. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. Al The arms of Act. polymorpha may. thus be roughly classified as follows :— Anterior, on radii A and B, 120-150 segments. Pinnules increasing in length to the terminal ones, which are very long and slender. 'Tentaculiferous. Anterolateral, on C, and E,, 100-120 segments. Terminal pinnules long and slender. 'Tenta- culiferous. Posterolateral, on C, and E,, 80-100 segments. Terminal pinnules stout, and rather longer than the median ones. Usually have “ sense-organs”’ and narrow ambulacral grooves, but are non-tentaculiferous. Posterior, on radius D, 60-80 segments. Terminal pinnules stout, but shorter than median ones. Sense-organs. Usually no grooves. Non-tentaculiferous. Another difference between the anterior and posterior arms is that the genital glands in the latter are far more developed than in the former. Not only is their number greater, although the total number of pinnules on a posterior arm may not be much more than half that of an anterior arm, but they also attain a very much greater size ; the basal and median pinnules of an anterior arm being very much less swollen than the corresponding pinnules of a posterior arm. A similar inequality in the development of the genital glands has been noticed by Alex. Agassiz’ as occurring in the Echini. This difference in length in the anterior and posterior arms of Act. polymorpha, and in the character of their terminal pinnules, seems to be to a certain extent dependent upon the condition of the respiratory apparatus occupying their ventral surface. When this is well developed the arm seems to have the power of indefinite growth; for in the single specimen (PI. I. fig. 16) in which all the thirty-three arms were normal and ten- taculiferous as in Antedon, there was no very appreciable difference in the lengths of the anterior and posterior arms?. The shape of the terminal pinnules, however, was of a slightly different character in the two cases, though the development of the genital glands was about the same; and we have just seen that those arms are the shortest in which the ambulacral groove entirely closes, and the water-vessel is reduced to a simple tube without any lateral tentacular branches, while it is in these arms only that any definite mode of termination is known. This may occur before half the number of segments have been developed which are commonly met with in an anterior ten- taculiferous arm. (§ 27) The ventral surface of some of my specimens of Act. polymorpha is marked by small calcareous concretions, somewhat resembling the “ blumenartige Knétchen mit mehreren blattartigen Fortsitzen”’ described by Miiller* in the Vienna specimen of Act. solaris. When present, they are usually scattered around the peristome, and 1 * Reyision of the Echini,’ part iv. pp. 680, 681. * Not only are the arms of different lengths in the ‘Challenger’ species of Actinometra, which have ungrooved hinder arms, but there are three species in which the anterior arms are longest, although all, anterior and posterior alike, are grooved and bear tentacles. In another species the arms are all grooved and all equal in length, but tho distribution of the syzygia is quite different in the anterior and posterior arms. 3 «Gattung Comatula, p. 12. Oo SECOND SERIES.—ZOOLOGY, VOU. If. 42 Mk. P. H. CARPENTER ON THE GENUS ACTINOMETRA. disposed along the edges of the primary groove-trunks proceeding from it, and there are generally some upon the sides of the anal tube. ‘They are particularly well de- veloped in the dark variety from Ubay, in which all the arms are tentaculiferous. (§ 28) The “oral pinnules” of Act. polymorpha, those, namely, which arch over the disk so as to protect it, are borne by the second distichals and second palmars when these are present, but in any case upon the second brachials, those of the distichals and palmars being the longest. They are all very long and slender, consisting of some 30 or 40 segments ; and their terminal portions exhibit the peculiar characteristic comb made up of processes which rise from the outer margin of the ventral surface of each calcareous segment (Pl. ITI. fig. 2), just as in Acé. solaris and Act. pectinata (Pl. III. fig. 1). The number of segments on which these processes may be developed varies from 10-12 on a distichal pinnule, to 6-8 ona brachial pinnule; but in cases in which no second distichals or palmars are developed, so that the pinnule on the second brachial is the first of the series, it is much longer than usual, and more of its terminal segments bear the comb-like processes. The oral pinnules of the dark Ubay variety of Act. polymorpha differ considerably from those of the type and of other Actinometre ; not only are they much stouter, but their terminal comb is differently constituted (Pl. LIL. fig. 3). Asis usually the case, the lower processes gradually develope themselves from the outer margin of the ventral surface of each calcareous segment; but towards the end of the pinnule they gradually come to rise less and less from the outer margin, and more and more from the median portion of the ventral surface of each segment, until finally, on the last two or three segments, they are developed from the inner margin. Consequently the comb, when viewed from aboye, is seen not to lie altogether on the outer side of the pinnule, as is usually the case, but to start from the outer side, cross its ventral surface, and finally come to lie on the inner side of each pinnule, 7. e. on the one nearest the arm. Both in the type of Act. polymorpha and in all the four varieties, the pinnules diminish in length from that of the second distichal (when present) to those borne by the fourth and fifth brachials; that of the sixth brachial is longer, and usually contains well-developed genital gland, so that it is slightly swollen. From this point onwards the pinnules increase in length till about the thirtieth brachial, after which their length and character vary according as the arm is tentaculiferous or non-tentaculiferous. (§ 29) The dorsal aspect of Act. polymorpha differs from that of most Antedons, and especially from that of Ant. rosacea, in the fact that the plane of the second and third radials, like that of the first, is parallel to the vertical axis of the calyx, and not inclined to it, as in Anfedon; so that the dorsal surfaces of the whole of the pieces of the calyx lie in one horizontal plane. The centrodorsal piece is circular (PI. IT. figs. 9, 10, ed), or pentagonal (fig. 11), and conceals a large portion of the pentagon formed by the first radials, less in young specimens with but a few arms (fig. 9) than in large and full-grown specimens with many arms (PI. IT. figs. 10,11 & Pl. VI. fig. 2). It is usually a flattened plate with a slight concavity in the centre of its outer surface ; and around its margins MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 43 are disposed some 20 or 25 cirrhi in one row, but with occasional traces of a second, in which the cirrhi alternate in position with those of the first row. The number of segments in each cirrhus is normally from 11 to 14, of which the last forms a recurved claw, while a more or less distinct spine is usually visible upon the dorsal edge of each of the three or four penultimate segments (PI. IIT. figs. 8-11). In Pl. II. fig. 8, is seen an abnormal condition of the centrodorsal piece, which is of an irregular oval form, and so extended as to conceal large portions even of the second radials. ‘These last are usually more or less completely united with one another late- rally. The amount of their union is to a certain extent dependent upon the number of arms developed. ‘Thus in the small specimen with only 13 arms, represented in Pl. II. fig. 9, the second radials are not united laterally for more than half their length ; in fig. 10 (26 arms) the union is somewhat more complete, and even more so in fig. 8 (28 arms), while in the variety with 39 arms, represented in fig. 11, the second radials are completely and closely united with one another all round. This rule, however, appears to be only a specific one, and not generally applicable to all Comatule ; for in the 80-armed Phanogenia the second radials, as figured by Lovén', do not appear to be united with one another any more closely than they are in the small 13-armed speci- men of Act. polymorpha (Pl. II. fig. 9). In Act. polymorpha the two segments (first distichals, palmars, or brachials) borne by any axillary are united to one another laterally to about very much the same extent as the second radials are; 7.e. when the number of arms is small, their first segments, whether primary, secondary, or tertiary, are not laterally united in pairs with such com- pleteness as when the division of the ten primary arms is carried to any considerable extent (Pl. IT. figs. 8-11, d,, p,, b,). When the arm-division is unequal it is generally carried further in the trivium or posterior radii, C, D, E, than in the two anterior radii, A, B, which form the bivium. This is well seen in Pl. II. fig. 9, in which no distichals are developed on either of the two anterior radii; and again in fig. 10, in which, while distichals are developed all round, the division is carried no further in one of the anterior radii, while in each of the others from one to three palmar series may be developed. In only four normal cases have I found an anterior radius to bear more arms than a posterior one. In each of these the total number of arms was considerable, and one at least of the two posterior radii bore the same number of arms as the abnormal anterior one. ‘Thus, for example, in Pl. II. fig. 11, each of the radii bears eight arms, with the exception of the posterior one (D), on which only seven are developed. ‘This, however, is an abnormal case of fracture of the whole radius between its second and third segments. The new portion is con- siderably smaller than the old, the proximal articular face of the new axillary being far less wide than the corresponding distal face of the old second radial; while both the distichal series which it bears are imperfect and abnormal, so that the absence of a further division in one of the secondary arms is not particularly remarkable. (§$ 30) The number of arms that may be developed in Acé. polymorpha is a character 1 « Phanogenia,” loc. cit. p. 230. SECOND SERIES.—ZOOLOGY, VOL. II. 7 4A MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. of extreme variability. In the specimens I have examined it varies from 13 to 39; so that, with one remarkable exception (PI. II. fig. 8), the ten primary arms do not, at the most, divide more than twice, while in two specimens with 18 and 13 arms respectively two and seven of the primary arms remain undivided. I believe, how- ever, that, as a general rule, an axillary is developed on each primary arm, and that the amount of further division is variable, but that a tertiary division is probably excep- tional, so that the number of arms in this species will be found rarely to exceed 40. It will have been already apparent from the position assigned to Act. polymorpha in the classification given in sect. 20, that I consider the typical number of distichals and palmars in this species to be three, of which the second (d,) bears a long pinnule, while the third or axillary segment (d a) consists of two primitive segments united by a syzygium. 3. Comb. From 2nd distichal to 6th brachial pinnules, and then, at intervals, to about 20th brachial but no further. The basal pinnules of the arms havea faint dorsal keel, and the distal ends of their segments are rather wider than the proximal ends, Diameter. 105 millims. Colour. Greyish brown. Locality. Ubay. Variety 2. Centrodorsal piece. Small, but rather thick. Cirrhi. 10, of 11 or 12 segments, with a terminal claw; the fourth and fifth are longest ; the spines on the dorsal border begin from the middle segments, and the opposing process on the penultimate segment is well marked. Radials. Second radials only incompletely united; second and third very convex, and much higher than the first ; median dorsal line of skeleton marked by a white line with dark borders, which is lost about the middle of the arms. Arms. 29. Syzygialinterval. Usually 3, but varying from 1-10 segments ; generally to >3. Comb. Limited to distichal, palmar, and first five brachial pinnules ; those of the 6th and next succeeding brachials have a dorsal keel, and the distal ends of their segments are much wider than the proximal ones. Colour. Greyish brown. Locality. Cabulan. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 53 Variety 3. Centrodorsal piece large and thick, with only 3 cirrhus-scars. Radials. Second radials completely united all round. Centre of dorsal surface of the skeleton, from the centrodorsal till near the end of the arms, marked by a faint white line with dark borders. Arms. 39. Syzygial interval. Usually 3, but varying from 1-7 segments; generally to <3. Comb. On second distichal, palmar, and brachial pinnules, and occasionally also on those of the 3rd-5th brachials, but on no others. Colour. Reddish brown. Locality. Bohol. Variety 4. Centrodorsal piece large and thick, with only 3 cirrhus-scars. Radials. Second radials closely united all round. Median white line on dorsal surface of skeleton very marked. Arms. 33, all tentaculiferous, and tolerably uniform in length and in the character of their pinnules. Syzygial interval. Usually 3, but varying from 0-6 segments; generally to >3. Pinnules. Oral pinnules much stouter than in the type ; that of third brachial but little shorter than that of second. Comb limited to these and to the distichal and palmar pinnules, and the processes forming it gradually come to rise from the ventral surfaces of the calcareous segments instead of from their outer margins. Colour. Blackish brown. Locality. Ubay. IV. THE SKELETON. (i.) Zhe Skeleton generally with its Ligaments and Muscles. (§ 35) The general structure of the skeleton of Actinometra, and of the ligaments and muscles which connect its component pieces, is precisely the same as in Antedon; and as this has been already described by Dr. Carpenter’, there is no need to repeat it here: afew points, however, must be treated somewhat more in detail. The component pieces of the skeleton of Actinometra, as of all the other Echinoderms, consist of a calcareous reticulation formed by the calcification of an organic basis of a protoplasmic nature, in which numerous nuclei and pigment-granules are imbedded. This “ nuclear tissue,” as Simroth’ has called it, is in the form of a network, around the meshes of which the calcareous material is deposited. The character of the calcareous reticulation varies greatly in different parts of the skeleton, being much closer at the synostoses and syzygia and at the articular surfaces than in the interior of the segments ; and in correspondence with this greater compact- ness of the calcareous tissue, the organic plexus which forms its basis becomes remark- ably modified at these points, as will be seen further on. The various modes of union of the different pieces of the Crinoid skeleton have been closely investigated by Miiller and by Dr. Carpenter. The former® described the stem-segments of Pentacrinus as united to one another in two different ways—(1) by the tendons which traverse the whole 1 Phil. Trans, loc. cit. p. 702. 2 Op. cit. p. 433. ° Bau des Pentacrinus, pp. 17-20. S* 54 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. length of the stem, passing through the substance of its various segments, and (2) by the * elastic interarticular substance”’ between the individual segments. The substance of these tendons consists of a white fibrillar tissue very like the ten- dinous tissue of the higher animals; but Miiller supposed the elastic interarticular sub- stance to be of a totally different nature, consisting of “lauter senkrecht stehenden Fasersiiulchen, die durch Reihen bogenformiger Schlingen einfacher Fasern verbunden sind,” and ‘ diese Schlingen gehen mit den regelmiissigsten Arkaden in ganz gleichen Abstiinden aus einem Fasersiiulchen in das andere iiber.”’ This substance fills up the whole space between the successive stem-segments which is not occupied by the tendons, and is connected in the closest possible manner with the opposed surfaces of every pair of segments, even extending for a short distance into their superficial cal- careous tissue. Each of the arcades above mentioned consists of a single primitive fibre, the terminations of which are lost in the ‘ Fasersiulchen ;” and the passage of these fibres in loops from one fibrous column to another gives an elasticity to the whole tissue, viz. a power of contraction after lateral displacement, and of extension after vertical compression, although the individual fibres are not elastic in the ordinary sense of the word. Miiller described the basals of Pentacrinus as united with the top stem-segments in the same manner as the successive stem-segments with one another, namely, by this elastic interarticular substance, while their sides simply “stossen an einander”’ (p. 25). This mode of union between the stem-segments was generally called by him a “ Nath,” or suture; but he sometimes spoke of it as an articulation, though he usually employed this last term only in those cases in which two segments are movable on one another through the intervention of muscles and ligaments which pass between them. He further described the union between the first radials and the basals of Pentacrinus and between the first radials and the centrodorsal piece of Comatula as a suture, which name he also gave to the lateral union of the five first radials with one another (pp. 28, 29); but he does not seem to have supposed that in these cases the various elements were connected by the elastic interarticular substance which he found between the likewise suturally united stem-segments. In fact, in speaking of the syzygia, which he called an immovable sutural union of two segments, he said expressly that not only the muscles but also the elastic interarticular substance was absent. On the other hand, the latter is to be found between the segments which are capable of motion upon one another, whether ligaments and muscles be present, as between the first and second radials and between most of the brachials, or ligaments only, as between the first and second brachials and the second and third radials ; for Miiller described the ligaments connecting two mutually movable segments as having essentially the same structure as the elastic interarticular substance of the stem, except that their surface is plain and not “ krausenartig gefaltet”’ (pp. 30-38). In those more common cases in which there is a muscular union between two segments, such as the first and second radials, in contact by transverse articular ridges upon their opposed faces (Pl. VII. figs. 1b, 2a, 4b, 5a; 7), Miiller drew no distinction between the pair of ligamentous bundles on the ventral side of the articular ridge, and the single MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 55 mass which occupies the whole space between the opposed faces on the dorsal side of this ridge, describing them as alike consisting of elastic interarticular substance, the function of which is extensor and antagonistic to the flexor action of the muscles. Dr. Carpenter’, however, regards the former as interarticular, with the special function of holding the pieces together, but allowing a certain amount of movement between them, while he describes the single dorsal mass as elastic, and as antagonizing by its extensile powers the action of the flexor muscles. Histologically he finds no differ- ence between them, both consisting of minute, straight, and nearly parallel fibres, very much, in fact, like those which Miiller described as composing the tendons of the stem of Pentacrinus. At their points of attachment to the pieces of the skeleton these fibres pass into their basis substance and become incorporated with it. The union of the first radials with one another and with the centrodorsal piece, which had been spoken of as sutural by Miiller, is described by Dr. Carpenter as “an adhesion of expanded surfaces closely fitted together, and held together by the con- tinuity of their sarcodic basis substance” (p. 704); so that the different elements are cemented together by a “thin layer of sarcodic substance, continuous with that which occupies the meshwork of their own calcareous reticulation.” This mode of union may be conveniently described by the word “ synostosis,’ which has been employed by Simroth to designate the mode of union of two faces which “ verkitten sich’’ in the skeleton of the Ophiuride. It is essentially the same as the syzygial union which occurs between certain pairs of the primitive arm-segments, although differing from it in points of detail. (§ 86) We have seen that Miiller regarded the tendinous tissue of the stem of Penta- erinus, and the fibrous ligamentous bundles, or, as he called it, the elastic interarticular substance uniting the movable elements of the skeleton both of Pentacrinus and of Comatula, as distinct from one another. I believe, however, that they are fundamentally identical, not only with one another, but also with the so-called “ cement-substance”’ between two segments which are united by synostosis. ‘This last consists, in the Ophiu- ride, according to Simroth’, of connective-tissue fibres which lose themselves in the organic basis of the skeleton, and are of the same nature as the substance of the masses of connective tissue uniting two articulating surfaces, both tissues staining deeply with picro-carmine. I find the same to be the case in ComatuJa and Pentacrinus. The tendons of the stem of the latter genus, the ligamentous bundles, composed, according to Miiller, of elastic interarticular substance, which connect every pair of movable arm-segments, and, lastly, the ‘‘cement-substance,”’ uniting the first radials to one another and to the centrodorsal piece, all stain very deeply with picro-carmine, and are of essentially the same histological structure. Fig. 4, on Plate III., represents a portion of a horizontal section through the suture, or, as I prefer to callit, the synostosis of two of the first radials of Pentacrinus. In the immediate neighbourhood of their apposed lateral faces there are none of the nuclei nor pigment-granules which are imbedded so abundantly in the more internal portions of their protoplasmic ground-substance, and the threads of the plexus of which it is composed ? Phil. Trans. loc. cit. pp. 703-714. = Op. cit. p. 435. 56 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. become excessively attenuated and disposed with great regularity almost parallel to one another. At the same time the meshes of this organic plexus become greatly elongated in the intervals between the parallel threads or fibres, which are connected with one another by very delicate fibrils passing in the form of loops from one fibre to another. These loops, which forcibly recall Miiller’s description of the arcades connecting the fibrous columns of the elastic interarticular substance in the stem of Pentacrinus, are simply the expression of the ends of elongated meshes of the protoplasmic plexus forming the organic basis of the skeleton. In the neighbourhood of each of the two opposed surfaces the fibrous elements of this plexus assume the character of closely placed parallel connective-tissue fibres, with no pigment-granules nor nuclei imbed- ded in them, but staining deeply with picro-carmine, while the normal protoplasmic basis of the interior part of the calcareous segments is but little affected by this reagent. ‘These fibres pass from the organic basis of the one segment into that of the other so that the two are firmly united, and the superficial denser layer of calcareous tissue is deposited around their ends, which corresponds with Miiller’s deseription of the elastic interarticular substance of the stem of Pentacrinus as extending for a short distance into the caleareous substance of the opposed faces of the segments. The superficial layer of calcareous reticulation which occupies the small intervals between the ends of the fibres thus becomes extremely close and compact; but the central portion of the fibrous tissue (Pl. III. fig. 4, Z) does not calcify, remaining as a thin layer of fibrous cement-substance between the two opposed surfaces, precisely like the inter- articular substance in the stem of Pentacrinus, with which I believe it to be identical. It is, at any rate, of the same nature as the substance of the ligaments connecting the first and second radials, which Miiller described as identical with that connecting the stem-segments; for at the angles of the radial pentagon the fibres of the cement-sub- stance connecting the adjacent first radials with one another in pairs pass directly into the fibres of the ligamentous bundles between the first and second radials. These, which are of precisely the same character as the ligamentous bundles between the successive brachial segments, also stain deeply with picro-carmine, and only differ from the cement- substance in the greater length of their fibrous element. At the points of attachment of the ligaments to the pieces of the skeleton, the meshes of the organic plexus become greatly elongated, and its fibrous bars regularly disposed and connected with one another by loops, as above described. As, however, the distance between the two articulating faces is very much greater than in a synostosis, several of these minute primitive fibrils unite to form one of the larger fibres composing the liga- mentous bundle, at the other attachment of which these primitive fibres again separate, become connected with one another by transverse loops, and finally pass into the bars of the protoplasmic plexus forming the ground-substance of the next segment. ($ 87) I have found the fibres composing the ligamentous bundles between the arm-segments of Antedon to terminate in the manner above described for Pentacrinus ; but in Act. polymorpha they do not pass so directly into the organic basis of the seg- ment. At the ends of the ligamentous bundles, where their component fibres begin to break up into primitive fibrils, the latter cross one another in all directions, very much MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, 57 as described by Simroth’ in Ophiactis virens, so as to form a network of delicate threads without any imbedded nuclei, although it may contain pigment-granules ; and this network passes very gradually into the nucleated protoplasmic plexus forming the organic basis of the brachial segments (PI. ITI. fig. 7, Z,). The tendons of the stem of Pentacrinus are, I believe, of precisely the same character ‘as the ligamentous bundles between the arm-segments, although, of course, enormously longer. They stain deeply with picro-carmine, and are composed of parallel fibres, which may be teased out into very much finer ones, and their upper ends pass into the organic ground-substance of the five basals, precisely in the same manner as the fibres of the arm-ligaments pass into the protoplasmic network composing the organic basis of the successive segments. In Pentacrinus Wyville-Thomsoni, in which the five basals are completely in contact with one another in pairs, the two elements of every pair are united by a synostosis, and the union of the basals with the radial pentagon above them is of the same cha- racter. The first radials of Comatula are connected with one another and with the centrodorsal piece in the same manner, as is seen in Pl. IIT. figs. 5, 6, where Z, 7 repre- sent the tracts of fibrous tissue connecting the first radials with one another and with the centrodorsal piece respectively. The terminal portions of this fibrous tissue become calcified to form the compact superficial layers of calcareous substance on the apposed faces, while the middle portion remains as the fibrous cement-substance uniting the two calcareous segments, which is thus essentially of a connective-tissue nature. The mode of union of the segments of the calyx of the Tesselate Crinoids, none of which are connected with one another by a muscular articulation like the first and second radials of Pentacrinus and Comatula, was most probably a synostosis of the same nature as those just described. The immovable sutural unions between certain of the brachial segments to which Miiller gave the name of “ syzygia,” are, in Pentacrinus, of precisely the same nature as the synostoses between the segments of the calyx, the organic basis of the one segment being continuous with that of the other through the fibrous cement-substance, which forms a thin layer between the whole of the two simple opposed surfaces. This was described by Miiller’ as a very delicate membrane, of a different nature from the elastic interarticular substance between the likewise sutu- rally united stem-segments. In Comatula, however, the apposed surfaces of the two elements united by a syzygium are not plain and simple, as in Pentacrinus and Rhizocrinus*, but marked by a series of radiating ridges, as in Apiocrinus obconicus, Goldf.* The ridges of the two surfaces correspond in position, and when the surfaces are in contact are closely applied to one another, and united by fibrous cement-substance as in an ordinary synostosis. The fibrils are very numerous and placed very close to one another, so that the calcareous reti- culation forming the ridges is remarkably dense and compact, being formed around the ends of these fibrils where they pass into the organic basis of the segments; and these ridges thus correspond to the whole of the syzygial surfaces in Pentacrinus and Rhizo- 1 Op. cit. p. 435. ? Bau des Pentacrinus, p. 29. 3 Sars, loc. cit. p. 22. 4 Petref. German. Taf. lvii. fig. 5. IE Pp >P > P 58 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, crinus. In the intervals between them the organic basis of the one segment is directly continuous with that of the other, little or no fibrous tissue being interposed. The muscular fibres of Actinometra correspond very closely with those of Antedon, as described by Dr. Carpenter’ and Ludwig’; their expanded terminations are simply applied to the surfaces of the calcareous segments to which they are attached, not passing into their substance as the ligamentous fibres do (PI. ITT. fig. 7, 6 m.), and there is no trace either of a sarcolemma or of transverse striation. (ii.) Lhe Dorsal Cirrhi. (§ 88) The Dorsal Cirrhi of Actinometra do not appear, so far, at least, as my observa- tions have extended, to be developed over such a large surface of the centrodorsal piece asis the casein Antedon. In all the specimens which I have examined the cirrhi are limited to its margin, while its central portion is entirely free from them and usually slightly concave. There is generally only one row of these appendages; but small and rudimentary ones may occasionally be found interposed between the large and full-grown ones at the extreme circumference of the plate, thus forming the commencement of a second row. The number of cirrhi existing at any one time upon the plate-like centro- dorsal piece of Act. polymorpha varies, I believe, between 15 and 20. Three or, in large specimens (PI. VI. figs. 1, 2), four are attached on each side of its more or less distinctly pentagonal margin, while in var. 1 (Pl. VI. fig. 14), and in one specimen of the type (fig. 7), the total number reached 25. In var. 2 (fig. 16) there are only 10; while in vars. 3 (fig. 20) and 4 there is no evidence, in the single specimens which I have examined, of the existence of more than three perfect cirrhi in the adult state, as there are no sockets around the margins of the centrodorsal plate for the attachment of a larger number ; though there may be minute openings here and there, which appear to have corresponded with the central canals of lost cirrhi, the sockets of which have been obliterated by a later calcareous deposit. It is not a little singular that the dorsal cirrhi of Act. polymorpha, like the centro- dorsal piece which bears them, should exhibit such a very slight range of variation, not only in size but also in number (the three varieties just mentioned of course excepted) ; for in nearly every other part of the skeleton the range of variation is very great. In Antedon rosacea the reverse appears to be the case; for the composition of the skeleton is fairly constant in its simplicity, but the cirrhi vary considerably both in number and in size. In a fully developed cirrhus of Act. polymorpha (Pl. III. fig. 8a) the number of segments varies from 11-14, being usually 12 or 13, the last of which is in the form of a strong sharp claw. This is attached by simple suture to the penultimate segment, which is prelonged at the base of the claw into a short opposing process on its concave or aboral margin. The diameter of the basal segment somewhat exceeds its length; but in the second and third segments this disproportion between the length and breadth is reduced, and in the fourth it becomes reversed, the length of this segment being slightly greater than its 1 Phil. Trans. loc. cit. p. 704. ? Beitriige ke. loc. cit. p. 40. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 59 diameter. In the fifth and sixth the proportion between the length and breadth reaches 3:2; anda very slight degree of lateral compression is visible in the latter segment, while the fifth, like the four basal segments, is cylindrical, or nearly so. ‘These are the two longest segments of the cirrhus; and from this point onwards the length of the segments gradually decreases, until in the tenth and following segments it becomes again less than the dorso-ventral diameter. At the same time the transverse diameter, which in the first five cylindrical segments is equal to the dorso-ventral diameter, under- goes in the seventh and eighth segments a sudden decrease. Petref. Germ., Taf. lvi. fig. 3, ¢. 5 Petref. Germ., Taf. lvii. fig. 2, p. 7 « Bau des Pentacrinus,” loc. cit. pp. 16, 25. SECOND SERIES.—ZOOLOGY, VOL. II. 15 108 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. upper ends) to the lower surfaces of the five basals in the same way as the ligaments of the arms are attached to the brachial segments, the fibres of the one passing gradually into the protoplasmic basis of the other. The basals cannot, however, be regarded as simply ossifications in this fibrous tissue of the same nature as the basal rays of Actino- metra; for, as shown above, there is every reason to believe that they are developed, like the other elements of the skeleton, out of the embryonic basal plates ; although, so far as position is concerned, they are precisely homologous with the calcareous deposits within the tendinous cords of the stem. On the other hand, the basal rays of Actinometra, which are similar in position, though not in origin, to the peripheral portions of the basals of Pentacrinus, are of the same nature as the caleareous tissue of the leaf-like areas of the stem-segments, being simply the result of the deposition of calcareous material around and between connective-tissue fibres. In many of the fossil Articulate Crinoids the lateral union of the basals with one another is so very complete that the lines of junction between them are not always visible, and the “basis” has therefore been described either as entirely absent or as replaced by the uppermost stem-segment, which, according to Miiller’s view, it is sup- posed to represent. This is particularly the case in the Apiocrinide and in Hugeniacrinus. Miller, who was the first to describe the latter type !, mistook the first radials of LZ. caryophyllatus for the basals, and described them as firmly anchylosed to what he supposed to be the “superior columnar joint.” Goldfuss*, however, rightly determined this last to be a part of the first radials, which are very much prolonged downwards, while, at the same time, he described the basals as replaced by the enlarged uppermost stem-segment, which articulates with the inferior surface of the elongated first radials. Roemer * did not accept this view of Goldfuss’s, although he recognized that the “superior columnar joint” of Miller was simply a dorsal prolongation of the first radials; but, like Miller, he described these last as the basals. It is most probable that Goldfuss’s view is the truer one, as in Hagenow’s figure * of Eugeniacrinus Hagenowii, in which the first radials are not prolonged downwards as in E. caryophyllatus, the piece on which they rest, representing that which Goldfuss called the enlarged uppermost stem-segment of Z. caryophyllatus, is seen to be distinctly com- posite ; for its external surface is marked by five sutural lines, alternating in position with those between the first radials, and evidently indicating the lines of union of five basals. The geological collection of the British Museum, which I have been able to examine, thanks to the kindness of Mr. Henry Woodward, contains a very interesting series of specimens from the Chalk which are labelled Apiocrinus ellipticus. In some of them the basals form a complete ring, separating the radials from the upper stem-joint, which is very much enlarged. But in other specimens the basals appear 1 Op. cit. p. 111. * Petref. Germ. tom. cit. p. 162. 5 Letheea Geognostica, ii. Theil 4, p. 115. ‘ Min. Jahrb. loc. cit. ix. p. 13. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 109 externally merely as small triangular pieces, not meeting laterally; so that they exhibit the same differences as the basals of Pentacrinus and of Solanocrinus. In many specimens the sutures between the basals, radials, and top stem-joint are clear and distinct ; but in others there is no trace of them at all, just as in some examples of Hugeniacrinus and Rhizocrinus; but this is hardly a satisfactory reason for supposing the basals to be internal and concealed, as has been done in the case of the last-named genus. Even in some species of Pentacrinus the basals appear to be very closely united to one another, and to assume the form of an uppermost stem-segment. Thus in P. sca- laris, Goldfuss 1, there is no appearance whatever of small wedge-shaped basals, such as are found in P. driareus and in P. asteria; but, as remarked by the Messrs. Austin 2, they appear to be united into a single plate, which resembles an “enlarged columnar joint.” The same was probably the case in the Jurassic genus Jsocrinus, described by Von Meyer *, though it is, of course, possible that in both these cases the basals may have been internal and concealed, as in Comatula. I have endeavoured to show elsewhere* that in the recent Rhizocrinus we find a strikingly similar case to that presented by Hugeniacrinus, viz., the sutures between the basals, visible externally in one species and not in another, or, rather, not invariably in another. In &. lofotensis the first radials rest upon a large and expanded apparently simple segment, which was described by Sars® as the expanded uppermost stem-segment ; and a small circular plate situated in the central vacuity between the first radials, with which, as well as with the enlarged uppermost stem-segment, it is closely connected, was regarded by him as representing the metamorphosed embryonic basals of Comatula. Pourtales’s observations ®, as well as my own subsequent ones, have led me to believe that the piece called the enlarged uppermost stem-joint of 2. loftensis by Sars and Ludwig’ is composed (if not entirely, at any rate in great part) of five closely anchylosed basals. Schliiter 5 is evidently not acquainted with the evidence on which this view rests, or he would scarcely suggest that R. Rawsonii might not be a Rhizocrinus at all, because its basals differ from those described in BR. loftensis by Sars and Ludwig; although these two observers are not themselves in accordance as to which parts of the interior of the calyx are to be regarded as concealed basals. Sir Wyville Thomson? takes the same view as Pourtales and myself; for he describes how “in Rhizocrinus the funnel-shaped piece formed by the coalescence of the basals with the fused first radials above and the dilated upper joint of the coalesced upper joints of the stem beneath, makes up a large part of the cup;” and his descriptions of the calices of Hyocrinus and Bathycrinus, both genera allied to Rhizocrinus, together with ’ Petref. Germ. tom. cit. p. 173, Taf. lx. fig. 10, s. * A Monograph on Recent and Fossil Crinoidea,’ p. 121 (Bristol, 1845). * « Tsocrinus und Chelocrinus,’ Museum Senkenbergianum, p. 251 (Frankfurt, 1837). 4 «© Pentacrinus and Ehizocrinus,” loc. cit. pp. 47-53. 5 Crinoides vivants, loc. cit. p. 4. © Hassler Expedition, loc. cit. pp. 28, 29. 7 Rhizocrinus lofotensis, loc. cit. pp. 121, 122. ® Op. cit. p. 29. Schliiter was unfortunately unable to make himself acquainted with Pourtales’s memoir. * “ Notice of new living Crinoids belonging to the Apiocrinide,” Journ. Linn. Soc. Zool. vol. xiii. p. 48. 15* 110 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. the analogies of Apiocrinus and Eugeniacrinus, strengthen Pourtales’s view still more completely. The occasional fusion of the upper stem-joints with the lower part of the calyx, as described above in the Apiocrinide, is an excellent illustration of Miller’s idea respecting the correspondence between the basis and the stem-joints. This correspondence, however, is by no means entitled to rank as a serial homology. The earliest condition of the basals shows them to be five separate plates developed in a spiral around the aboral ecelom of the Crinoid embryo!. They have distinct homologies in the apical system of the other Echinoderms’; while the stem-segments, surrounding the aboral ccelom much in the same way as the basal circlet, are simple undivided pieces from the first, and seem to be almost or quite unrepresented in the other Echinoderms. (vi.) The Second and Third Radials. (§ 74) The second radial of Act. polymorpha (PI. VIL. fig. 2), like that of Ant. rosacea, is an oval, somewhat discoidal plate, having two nearly parallel faces—one internal or proximal, articulating with the first radial, the other external or distal, articulating with the third radial. The internal face (fig. 2a) closely resembles the external face of the first radial (Pl. VII. fig. 1b), with which it articulates, being divided transversely by a large articular ridge (7) into a dorsal and a ventral portion; the former is entirely occupied by the fossa lodging the elastic ligament (/), which is particularly deep just below the opening of the central canal (¢.c). From the ventral margin of this opening arise the two ridges which bound the intermuscular furrow (/;), and are joined near their upper extremities by the transverse secondary ridges separating the large fossee (h) that lodge the interarticular ligaments from those (f) lodging the flexor muscles of the ray; the latter are excavated ina pair of thin lamelle, which extend upwards from the proper ventral margin of the plate, as is seen in a view of the distal face (fig. 2 b, g). Besides the above-mentioned ridges and fossee, which correspond to similar ones on the distal face of the first radial, the proximal face of the second radial shows two lateral processes, in which shallow fossz (4) are excavated. These processes represent the outer portions of the distal face, which is somewhat wider than the proximal one, as the lateral faces are not set at right angles to the two terminal ones, but form an oblique angle with the proximal face, so that the outline of the radial, when seen from the dorsal or ventral side, is trapezoidal in form (fig. 2c, d). The shallow fossee which are exca- vated in these lateral faces lodge the ligamentous substance by which the second radials are united with one another in pairs: the extent of this union is, as above remarked, very variable in different specimens, being generally greatest where the number of arms is largest (Pl. II. figs. 9, 11). The external or distal face (fig. 2b) is much simpler in character than the proximal one, as no muscles are attached to the vertical lamellae which rise from its ventral margin above the articular face proper. This last is divided by a vertical ridge (i) that ' Gotte, loc. cit. pp. 595, 620. * P. H. Carpenter, “ On the Oral and Apical Systems of the Echinoderms,” Quart. Journ. Mier. Sci. xviii. (1878) pp. o71, 3&2. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. LE passes round the opening of the central wanal (c.c) into a pair of lateral fossee (h), which give attachment to the large interarticular ligament connecting the second with the third radial. The proximal face is not quite vertical, but slightly inclined towards the distal one, so that the ventral face is not much more than an edge. When the piece is viewed from the ventral side, therefore (fig. 2¢), little else is visible but the fosse for the muscles (f) and interarticular ligaments (/) of the proximal face and the intermuscular furrow (f) descending along its median line. The second radials of var. 1 are very similar to those of the type, except that, as in the first radials, the muscular fossze are relatively somewhat larger. In var. 2, however, they are very much smaller (Pl. VII. fig. 5a, f); and there are no vertical lamellee pro- jecting from the ventral margin of the distal face (fig. 5b), as is the case in the type. The lateral fossze (%) lodging the ligamentous substance which connects the second radials with one another are somewhat more marked, as the union of the second radials in pairs is more complete than in the type, though not so complete as in varieties 1, 3, and 4. The two latter also agree with var. 2 in the fact that the proximal and distal faces of the second radials are nearly parallel, and less inclined to one another than in the type and in var. 1; so that the fossze for the muscles and interarticular ligaments are barely visible when the piece is seen from the ventral side (fig. 5 ¢), as there is a proper ventral face. Its median line is occupied by a continuation of the furrow on the ventral surface of the first radial (figs. 4c, 5c, v.7.f), while its lateral portions are divided up, in the same way as those of the first radial, into secondary ridges and furrows. (§ 75) The third or axillary radial of Act. polymorpha, which gives attachment to two primary arms, presents three articular surfaces—an internal one corresponding to the distal face of the second radial, and two external ones, inclined to one another, with which the bases of the arms articulate. The proximal face (Pl. VII. fig. 3a) is precisely similar in character to the distal face of the second radial, being divided, like it, by a vertical ridge into two lateral fossze (h) which lodge the interarticular ligaments. Its articular margin, when viewed from the dorsal side (fig. 3d), is perfectly straight, and does not project in the middle as in Ant. rosacea ; so that the possible amount of lateral movement between the second and third radials must be extremely slight. Two vertical lamellee (g) project from the uppermost margin of the internal face; but they do not form part of the surface of articulation with the second radial, as they are excavated into fosse on their outer side for the attachment of the proximal ends of the outer muscular bundles passing between the axillary radial and the lowest segments of the primary arms. The two inner muscular bundles are attached to the two sides of a projecting wedge-shaped process (cl) on the external or distal face, the “ clavicular”’ of Schultze’, which occupies the angle between the two ob- liquely placed articular faces for the basal arm-segments. These are of precisely the same character as the external faces of the first radials (fig. 3b), consisting, besides the muscular fossve (f°) just mentioned, of two others for the interarticular ligaments (/), and of a large dorsal fossa (7) lodging the elastic ligament, and separated from the other two by a transverse articular ridge (7), in the centre of which is the opening of the central canal (c.c). * Loe. cit. p. 5. 112 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, The median line of the ventral face (fig. 3¢) is oceupied by a ventral radial furrow continuous with that on the ventral face of the first radial ; it divides into two branches, one of which passes on either side of the clavicular, in order to be continued on to the basal arm-segments. The proximal end of this furrow is indicated by the deep notch separating the two vertical lamellee which project upwards from the proper internal face (fig. 3.a, g), and through which the base of the clavicular is seen. In the type and in var. 1 the lateral portions of the ventral face of the third radial are plain, and not sculptured; but in varieties 2-4 they are divided up by secondary ridges and furrows (fig. 6c), just like the ventral faces of the first and second radials (figs. 4e,5¢). In these varieties also there are no vertical lamelle projecting upwards from the ventral margin of the internal face (fig. 6 a), which is also the case in the ex- ternal face of the second radial (Pl. VII. fig. 5 b), as the muscular bundles passing between the first and second radials, and between the third radials and basal arm-segments, are smaller than in the type. In var. 2 there would appear to be more power of lateral movement between the second and third radials than is the case in the type; for although, as in the type, there is no projection in the middle of the proximal articular margin of the third radial, yet the distal articular margin of the second radial shows a slight indication of such a median prominence (fig. 5 d), which is absent in the type. It would seem though, to be replaced to a certain extent by the greater thickness of the vertical articular ridge (¢) around the opening of the central canal, which is seen, in Pl. VIL. fig. 2d, to project a little beyond the level of the dorsal surface of the radial; so that when the opposed ridges of the second and third radials are in contact with each other, the third may possibly have a very slight power of lateral movement upon the second, though by no means so great as in Ant. rosacea, in which the median pro- minence on the internal articular margin of the third radial is very marked. The second and third radials of var. 2 differ from those of the type of det. polymorpha and of all the other varieties in the very marked convexity of their dorsal surfaces, which renders them considerably higher than the first radials; so that when the whole calyx is viewed from the exterior, the inner circle of first radials, which are only very little con- cealed by the small centrodorsal piece, seems somewhat sunk within the outer cirele formed by the second and third radials. This marked convexity is well seen in Pl. VII. fig. 5,a, d, and fig. 6, b, d, especially when these figures are compared with those of the corresponding parts in the type (fig. 2, a, d, fig. 3, b, d). List oF WORKS REFERRED TO. 1. Adams, J., “ Description of some Marine Animals found on the coast of Wales,” Trans. Linn. Soe. vol. y. p. 7 (London, 1800). 2. Agassiz, L., “ Prodrome d’une Monographie des Radiaires ou Echinodermes,” Annales d. Sci. Nat. 2™° sér. Zool. vii. p. 288. 3. Agassiz, Al., “ Revision of the Echini,” Illust. Cat. Mus. of Comp. Zool., No. vii. (Cambridge, U.S. 1872-74). 4. Austin, T. and T., Jun., ‘A Monograph on Recent and Fossil Crinoidea’ (Bristol, 1845). MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 113 5. Barrelieri, Jac., ‘Plante per Galliam, Hispaniam et Italiam observate ’ (Paris, 1714). 6. Beyrich, E., ‘‘ Ueber die Crinoideen des Muschelkalks,” Abhandl. d. k. Akad. zu Berlin, 1857. 7. de Blainville, H. M. D., ‘Manuel d’Actinologie’ (Paris, 1834). 8. Bohlsche, W., “Ueber Actinometra Bennettii und eine neue Comatula-Art (Antedon Dubenii),” Wiegm. Archiv fiir Naturgesch. 1866, i. p. 92. 9. Bronn, H. G., ‘Die Klassen und Ordnungen des Thierreichs, Band ii, Aktinozoen (Leipzig, 1860). 10. Carpenter, P. Herbert, “ Remarks on the Anatomy of the Arms of the Crinoids, Part I.,” Journ, of Anat. and Physiol. vol. x. April, 1876, pp. 571-585. 11. Carpenter, P. Herbert, The same, Part II., Journ. of Anat. and Physiol. vol. xi. Oct. 1876, pp. 87-95. 12. Carpenter, W. B., “‘ Researches on the Structure, Physiology, and Development of Aniedon rosa- ceus, Part I.,’ Phil. Trans. vol. 156, pp. 671-756, pls. xxxi.—xli. 13. Carpenter, W. B., “On the Structure, Physiology, and Development of Antedon rosaceus,” Proc. Roy. Soc., No. 166, 1876, pp. 211-231, pls. 8, 9. 14. Carpenter, W.B., “ Supplemental Note to the above Paper,” P. R. S., No. 169, 1876. 15. Columna, Fabius, ‘ Phytobasanus, sive Plantarum aliquot Historia’ (Neapoli, 1592). 16. Dujardin and Hupé, ‘ Histoire Naturelle des Zoophytes Echinodermes’ (Paris, 1862). 17. ‘Encyclopédie Méthodique,’ “ Histoire Naturelle des Vers, des Mollusques, des Coquillages et Zoophytes ” (Paris, 1789-1832), Partie des Vers. 18. De Freminville, “Mémoire sur un nouveau Genre de Zoophytes de l’Ordre des Radiaires,” Nouv. Bull. d. Sci. par la Soc. Philom. tom. ii. p. 349 (Paris, 1811). 19. Gegenbaur, C., ‘Grundriss der vergleich. Anat.’ (Leipzig, 1874). 20. Goldfuss, G. A., ‘ Petrefacta Germanie ’ (Dusseldorf, 1826-35). 21. Goldfuss, G. A., “Beitr. zur Petrefactenkunde,” N. Acta Acad. Leop.-Carol. Nat.-Cur. xix. . 1839, p. 348. 22. Goldfuss, G. A., “ Beitr. zur Petrefactenkunde,” Neues Jahrb. f. Mineral. (1841), p. 86. 23. Gétte, Al., “ Vergleich. Entwickelungsgesch. der Comatula mediterranea,’ Archiv f. mikros. Anat. Band xii. 1876, pp. 583-648, Taf. xxv.—xxviii. 24, Gray, J. E., “ Notice on the Digestive Organs of the Genus Comatula, and on the Crinoidea of Miller,” Annals of Philosophy, N.S. vol. xii. p. 592 (1826). 25. Greeff, R., “ Ueber den Bau der Echinodermen, 3. Mittheilung,” Sitzungsb. d. Gesellsch. z. Be- ford. der gesam. Naturwiss. zu Marburg, No. 11, 1872, pp. 155-169. 26. Greeff, R., The same, 4. Mittheilung, “Ueber den Bau der Crinoideen,” Marburg. Sitzungsb. No. 1, 1876, pp. 16-29. 27. Greeff, R., The same, 5. Mittheilung, “ Ueber das Herz der Crinoideen,” Marburg. Sitzungsb. No. 5, 1876, pp. 83-95. 28. Von Hagenow, Fr., “ Monographie der Riigen’schen Kreide-Versteinerungen, II. Abtheilung, Radiarien und Annulaten,” Neues Jahrb. fiir Mineralogie, 1840, p. 664. 29. Heusinger, C. F., “ Bemerkungen iiber den Verdauungskanal der Comatulen,” Meckel’s Archiv f, Physiol. 1826, p. 317. 30. Heusinger, C. F., “ Anatomische Untersuchungen der Comatula mediterranea,” Zeitsch. f. organ. Physik, B Band iii. 1833, pp. 8367-374, Taf. x., xi. 31. De Koninck, L., and H. Le Hon, ‘ tes ches sur les Crinoides du Terrain Carbonifére de la Bel- gique’ (Bruxelles, 1854). 32. Lamarck, J., ‘Systtme d’Animanx sans Vertébres,’ 2™° édit., Paris, 1816, tom. ii. p. 532. 33. Lange, W., “ Beitr. z. Anat. und Histiologie der Asterien und Ophiuren,” Morphol. Jahrb. Band ii., 1876, pp. 241-286, Taf. xv.—xviil. 114 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 34. Leach, W. E., ‘ Zoological Miscellanies,’ vol. ii. p. 61 (London, 1815). 35. Leuckart, PF. S., “ Emiges tiber das Asteroiden-Geschlecht Comatula, Lam., tiberhaupt, und iiber C. mediterranea iusbesondere,” Zeitsch. f. organ. Physik, Band iii. 1833, p. 385. 36. Linckii, Johannis Henrici, Lipsiensis, ‘ De Stellis Marinis liber singularis’ (Lipsie, 1733). 37. Linnzus, K., ‘Systema Nature,’ editio decima tertia, pars vi. p. 3166 (Lipsiz, 1788). 38. Lovén, S., ‘ Phanogenia, ett hittills okiéndt sligte af fria Crinoideer,” Ofvers. af Kongl. Ve- tensk.-Akad. Férhandl. 1866, No. 9, pp. 223-233. 39. Ludwig, Hubert, “Beitr. zur Anat. der Crinoideen,’ Nachrichten von der Kénigl. Gesellsch. der Wissensch. und der G. A, Universitat zu Gottingen, No. 5, 1876, pp. 105-114. 40. Ludwig, Hubert, ‘ Morphologische Studien an Echinodermen, I. Beitr. zur Anat. der Crinoideen ” (Leipzig, 1877) ; Separat-Abdruck aus der Zeitsch. f. wissensch. Zool. Band xxviii. 41. Ludwig, Hubert, “Zur Anatomie des Rhizocrinus lofotensis, M. Sars,’ Zeitschr. f. wiss. Zool. Band xxix. pp. 101-130, Taf. v., vi. 42. Luidi, Eduardi, ‘ Lithophylacii Britannici Ichnographia’ (Londini, 1699). 43. Luidi, Eduardi, ‘ Prelectio de Stellis marinis Oceani Britannici, nec non de Asteriarum, Entro- chorum, et Encrinorum origine’ (Oxford, 1708). 44, Meckel, J. F., ‘“ Ueber die Oeffnungen des Speisekanals bei den Comatulen,”’ Meckel’s Archiv f. Physiol. Band i. 1828, p. 470. 45. Metschnikoff, E., “ Beitr. zur Entwickelungsgesch. einiger niederen Thiere,” Bull. de ? Acad. Imp. des Sci. de St. Pétersb. tom. xv. 1871, pp. 502-509. 46. Meyer, C. E. H., “ Isocrinus und Chelocrinus,’ Museum Senkenbergianum (frankfurt, 1837). 47. Miller, J. 8., ‘A Natural History of the Crinoidea’ (Bristol, 1821). 48. Miiller, J., “Ueber den Bau des Pentacrinus caput-Meduse,’ Abhandl. d. k. Akad. zu Berlin, 1843; Abst. in Wiegm. Archiv f. Naturgesch. 1840, i. 49. Miiller, J., “Ueber die Gattungen und Arten der Comatulen,”’ Wiegm. Archiv f. Naturgesch. 1841,i. 50. Miiller, J., “ Neue Beitr. zur Kenntniss der Arten der Comatulen,” Wiegm. Archiv f. Naturgesch. 1843, i. 51. Miiller, J., “ Nachtrag zu der Abhandlung iiber die Comatulen,” Monatsb. d. Berlin. Akad. 1846, p. 177. 52. Miiller, J., “ Ueber die Gattung Comatula, Lam., und ihre Arten,”’ Abhandl. d. k, Akad. z. Berlin, 1849. 53. Miiller, J., “ Ueber den Bau der Echinodermen,” Abhandl. d. k. Akad. z. Berlin, 1853. 54. Norman, A. M., “ On the Genera and Species of the British Echinodermata,” Annals and Mag. of Nat. Hist. ser. iii. vol. xv. p. 98. 55. @Orbigny, A. D., ‘ Cours élémentaire de Paléontologie et de Géologie stratigraphique’ (Paris, 1850-52). 56. Petiveri, Jacobi, ‘ Gazophylacium Nature et Artis’ (Londini, 1711). 57. Petiveri, Jacobi, ‘ Aquatilium animalium Amboinensium Icones et Nomina’ (Londini, 1713). 58. Philippi, R. A., “ Alecto alticeps, n. sp., eine tertiire Comatula-Art von Palermo,” Neues Jahrb. f. Mineral. 1844, p. 540. 59. Pictet, F. G., ‘ Traité de Paléontologie ? (Paris, 1857). 60. de Pourtales, L. F., ‘ Contributions to the Fauna of the Gulf Stream at Great Depths,” Bull. of the Mus. of Comp. Zool. vol. i. no. 6. 61. de Pourtales, L. F., “ List of the Crinoids obtained on the coasts of Florida and Cuba by the United States Coast Survey Gulf-Stream Expeditions, in 1867, 1868, 1869,” Bull. of the Mus. of Comp. Zool. vol. i. no. 11, pp. 355-358. 62. de Pourtales, L. F., “ On a new Species of Rhizocrinus from Barbadoes,” Zool. Results of the Hassler Expedition, Illust. Cat. ofthe Mus. of Comp. Zool. no, viii. pp. 27-31, pl. v. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 115 63. Retzius, A. J., ‘ Nova Acta,’ Stockholm, 1783, no. 12, p. 234. 64. Retzius, A. J., ‘Dissertatio sistens species cognitas Asteriarum’ (Lunde, 1805). 65. Roemer, F., In ‘ Lethza Geognostica,’ dritte Auflage, 1851, Theile iv. v. 66. Rosinus, M. R., ‘ Tentaminis de Lithozois ac Lithophytis olim marinis, jam vero subterraneis, prodromus ; sive de stellis marinis quondam, nunc fossilibus, disquisitio’ (Hamburg, 1719). 67. Sars, M., ‘ Mémoires pour servir 4 la connaissance des Crinoides vivants’ (Christiania, 1868). 68. von Schlottheim, S., ‘ Nachtrige zur Petrefactenkunde, Abth. ii. (Gotha, 1823), p. 48. 69. Schultze, L., ‘ Monog. der Echinodermen des Eiflerkalkes ’ (Wien, 1866). Besonders Abgedriickt aus dem xxvi. Bande der Denkschr. der mathemat.-naturwissenschaft. Classe d. kaiserl. Akad. d. Wissensch. 70. Schweigger, A. F.,‘ Beobacht. auf naturhistorischen Reisen ’ (Berlin, 1819). 71. Semper, C., “ Kurze anatomische Bemerkungen iiber Comatula,” Arbeiten aus dem zoolog.- zootom. Institut zu Wiirzburg, Band i. 1874, pp. 259-263. 72. Simroth, H., “ Anatomie und Schizogonie der Ophiactis virens, Sars, Theil I.,” Zeitsch. f. wiss. Zool. Band xxvii. Heft 4. 73. Teuscher, R., ‘‘ Beitr. zur Anatomie der Echinodermen, I. Comatula mediterranea,’ Jenaische Zeitsch. f. Naturwissensch., Band x. 1876, pp. 243-262, Taf. vii. 74, Teuscher, R., “ Beitr. zur Anat. der Echinodermen, II. Ophiuride,” Jenais. Zeitsch. Band x. 1876, pp. 263-280, Taf. viii. 75. Teuscher, R., ‘ Beitr. zur Anat. der Echinodermen, III. Asteride,” Jenais. Zeitsch. Band x. 1876, pp- 493-514, Taf. xviii., xix. 76. Thomson, Wyy., “ On the Embryogeny of Antedon rosaceus, Linck (Comatula rosacea of Lamarck),” Phil. Trans. vol. 155, 1865, pp. 513-544, pls. xxili—xxvii. ADDENDA. 77. Baudelot, E., “ Btudes générales sur le systéme nerveux. Contribution & Vhistoire du systéme nerveux des Echinodermes,” Arch. de Zool. expérim. et générale, tome i. pp. 177-216. 78. Carpenter, P. Herbert, “On some points in the Anatomy of Pentacrinus and Rhizocrinus,” Journ. of Anat. and Physiol. vol. xii. Oct. 1877, pp. 85-53. 79. Carpenter, P. Herbert, “On the Oral and Apical Systems of the Echinoderms, Part i.,” Quart. Journ. Mier. Sci. vol. xviii. new series, Oct. 1878, pp. 351-383. 80. Gegenbaur, Carl, ‘Grundriss der vergleichenden Anatomie,’ zweite Auflage (Leipzig, 1878). 81. Grube, Descriptions of three new Comatule (C. levissima, C. Mertensi, Act. borneensis) in Jah- resber. d. schles. Gesellsch. 1875, Nat. Hist. Sect. pp. 54, 55%. 82. Ludwig, H., “ Beitrage zur Anatomie der Asteriden,” Zeitschr. f. wissensch. Zool. Band xxx. pp. 150-212, Taf. v.—vii. 83. Schliiter, Clemens, “ Ueber einige astylide Crinoiden,” Zeitschr. d. deutsch. geol. Gesellsch. Jahrg. 1878, pp. 28-66, Taf. i—iv. 84. Thomson, Sir C. Wyville, ‘‘ Notice of new living Crinoids belonging to the Apiocrinide,”’ Journ. Linn. Soc., Zool. vol. xiii. pp. 47-55. * T haye, unfortunately, been unable to get a sight of this paper, and only know of it from the reference to it in Leuckart’s ‘ Jahresbericht.’ SECOND SERIES.— ZOOLOGY, VOL. II. 16 116 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. DESCRIPTION OF THE PLATES. The following letters denote the same parts throughout all the Plates. A, B, C, D, E. The five Radii or Ambulacra. A,, A,, B,, B,, C,, Cy, D,, Dz, Ey, E,. The ten Primary Arms. a.c. Axial ccelom. ap. Axial prolongation. a.i.c. Axial interradial canal. a.r.c. Axial radial canal. a.i.f. Axial interradial furrow. arf. Axial radial furrow. An. Anal tube. b,, bo, b3, bs, by. First, second, third, sixth, and tenth brachials. 6.b. Basal bridge. b.f. Basal fold. b.g. Basal groove. b.m,. Muscles between the radial axillary and the first brachials. b.m,. Muscles between the second and third brachials. c.c. Central canal of the calcareous segments of the rays and arms. c.c’. Central canal of the cirrhus-segments. ch. Chambers of the quinquelocular organ. cd, Centrodorsal piece. cir. Cirrhus. cd.c. Centrodorsal ccelom. cel. Clavicular. cn. Caleareous network in the central vacuity of the pentagonal base of the calyx. co.c. Commissural canals in the first radials. dif. Dorsal interradial furrow. cv.c. Circumvisceral ccelom. d.rf. Dorsal radial furrow. d,, d,, d.a. First, second, and axillary distichals. ep. Epithelial wall of the alimentary canal. F. Central funnel-shaped space enclosed within the pentagonal base. f. Muscular fossve. g. Vertical lamellze of the calcareous segments. jf; Intermuscular furrow. gx. Their superior margins. f2. Notch representing it in Ant. celtica. go. Their inner lateral margins. gs. Ridge formed by the union of these in the first radials of Ant. celtica. h. Fossz lodging the interarticular ligaments. i. Transverse articular ridge. i.co. Interradial commissure. i.e. Interradial elevations on the centrodorsal piece. iv.c. Intervisceral ccelom. j. Fosse lodging the elastic ligaments. k. Fosse lodging the ligamentous substance between the sides of the second radials. L. Ligamentous substance between the sides of the first radials. 1. Ligamentous substance between the first radials and the centrodorsal piece. L'. Incompletely decalcified portions of the skeleton. L,. Interarticular and ) ligaments between the first and second radials, or second and third 1,. Elastic brachials. L,. Interarticular ligaments between the second and third radials. M. Mouth. N. Fibrillar nervous envelope of the quinquelocular organ. n. Axial nervous cords of the rays and arms. o. Interradial spout-like processes of the rosette. n. Their branches. ' go’, Interradial triangular processes. n.c, Axial nervous cords of the cirrhi. o.b. Ovoid bodies. P. Peristome. Pe. Uncealcified perisome between the radii. p. Radial processes of the rosette. p'. Small curved processes at the sides of a single basal. Dry P» p-2. First, second, and axillary palmars. Q. Radial openings on the dorsal surface of the pentagonal base. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA, nal Q'. Notches on the inner margins of the dorsal faces of the first radials. q. Radial depressions on the centrodorsal piece. R. Rosette. Ty, To, 7.a. First, second, and axillary radials. r.ar. Radial areas on the centrodorsal piece. r.c’. Diverticulum from the radial ccelom into the substance of the first radial. r.co. Intraradial commissure. r.o, Central opening of the rosette. r.m. Muscles between the first and second radials. 7s. Radial space. S. Rays of the basal star. S,, Sy, S3. Diverging, vertical, and lateral fibres effecting the interradial portions of the synostosis between the centrodorsal piece and the radial pentagon. s. Depressions at the central ends of the rays of the basal star. sp, Ssp.a. First and axillary suprapalmars. Sy. Syzygium. t. Short processes at the angles of the centrodorsal piece. U. Sockets for the attachment of the dorsal cirrhi. u. Inner openings of the cirrhus-canals in the centrodorsal piece. vif. Ventral interradial furrow. v.rf. Ventral radial furrow. V, W,X, Y, Z. The five primary basal cords proceeding from the angles of the quinquelocular organ. Vi, Vo, Wi, We, X1, Xo, V1, Yo, Z, Z. The ten secondary basal cords produced by the bifurcation of the primary ones. Vyy Voy Wy Woy Ly) Loy Yq» Yoo Zt) a» The apertures in the basals through which the secondary cords pass. v, v, w,w', x, 2, y,y!, z, 2. The corresponding apertures of the central canals on the internal faces of the first radials. In Pl. II. figs. 9-11 indicate the position of the mouth (ventral) relative to the radial skeleton (dorsal). * In Pl. V. fig. 4: indicates the passage of the ventral radial canal into the central calcareous network within the radial pentagon by two openings, instead of by only one as usual. Puate I. Diagrams of the distribution of the ambulacra on the disks of different species of Comatula. The red lines mark the interradial intervals. Figs. 1-4 copied from Miiller. Fig. 1. Antedon rosacea. Fig. 3. Act. Wahlberyhii. 2. Actinometra solaris. 4. Act. multiradiata. In figs. 5-16 the tentaculiferous grooves are marked by dark lines, and the non-tentaculiferous grooves by fainter lines. Fig. 5. Act. solaris. Proportion of non-tentaculiferous arms, 0 6. Act. polymorpha: Type. Ss +5 Ze oP) 2 ord +8 8. » » » 20 ah 5 Var. 1. is at 10. ” Type. 2) 33 ll. » » 2 38 12. » » ” 28 13. > 2) 9 a3 14. es Var. 2. 33 19 isp ¥3 Type. » 4? 16. ty Var. 4. All the arms tentaculiferous. iG? 118 oo a: 10. Wie MR. P, H. CARPENTER ON THE GENUS ACTINOMETRA. Prats II. Diagram of the distribution of the ambulacra in a new Actinometra from the Philippines. Superior or ventral aspect of the disk of Act. polymorpha, var. 2, the oral pinnules having been cut away near their bases. x 3. . Piece of an ordinary tentaculiferous arm of Act. polymorpha, from about the middle of its length, seen from above. x 4. Terminal portion of the same arm. x 4. . Piece of the middle portion of a non-tentaculiferous arm, borne upon the same axillary as that represented in figs. 3&4. Ventral view. x 4. . Termination of the same arm. xX 4. Distichal, palmar, and lower brachial segments of one of the radii of Act. polymorpha, var. 4, showing the white line which occupies the middle of the dorsal surface of the skeleton. x 4, Centrodorsal piece and one radius of a monstrous specimen of Act. polymorpha, showing the very irregular form of the centrodorsal piece (ed) and the imperfect condition of one of the palmar series, which consists simply of one axillary segment bearing brachials upon one of its distal faces, and two suprapalmars (sp, sp.@) upon the other. x 4. Diagram of the calyx of a small thirteen-armed specimen of Act. polymorpha: Type. x 4. A similar diagram of a larger specimen with 26 arms. x 3. A similar diagram of a specimen of var. 3, with 89 arms. x 2. The * in these three figures (9, 10, 11) indicates the position of the mouth on the ventral side of the disk relatively to the radial skeleton of the dorsal side. Fig. 1, 2. 3. 4. 5. 6. ~ 4. os 8. 9. 10. iit. Figs. 1 3 Puate IIT. Terminal comb of an oral pinnule of Act. pectinata. x 20. Oral pinnule of Act. polymorpha, Type. x 10. Oral pinnule of Act. polymorpha, Var. 4. x 10. Portion of a horizontal section through the synostosis of two first radials of Pentacrinus Wyville- Thomsoni. x 110. Lower portion of a vertical section through the peripheral end of the synostosis of two first radials of Act. polymorpha. x 110. Lower portion of a similar section, taken rather nearer the centre of the radial pentagon, showing the disposition of the fibres which effect the synostosis of the first radials with the centrodorsal piece, both in the radial (/) and the interradial planes (S,, S,, 83). x 110. Longitudinal section through one of the muscles (4.m,) and interarticular ligaments (Z,) con- necting the second and third brachials. x 110. 10. Cirrhi of Act. polymorpha. Type, and vars. 1 & 2. Type. x 6. a, adult; 6, very young ; s; c, nearly mature. Isolated cirrhus-segments of the Type. x 40. a, a basal segment; 6, a terminal segment. Cirrhus of var. 2. x 6. Cirrhus of var. 1. x 6. Puate IV. Figs. 1-8 of Ant. celtica. All x 7. & 2. Centrodorsal piece, after removal of the cirrhi, as seen from its dorsal (fig. 1) and ventral (fig. 2) sides. & 4. Pentagonal base of the calyx, as seen from its dorsal (fig. 3) and ventral (fig. 4) sides. Fig. 10. 14. 15. MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 119 . Interior of the calyx, as seen after removal of the visceral mass. . Lateral view of the base of the calyx with the centrodorsal piece in situ. . Pentagonal base of the calyx of a smaller variety, as seen from its dorsal side. . Lateral view of the base of the calyx of the same variety, with the centrodorsal piece in situ. Figs. 9-11 of Ant. Eschrichtii. . Two united first radials, together with the portion of the rosette which is in connexion with them, as seen from within. x 7. Pentagonal base of the calyx as seen from its dorsal side after removal of the rosette occu- pying its central cavity. x 3. . Centrodorsal piece seen from its ventral side. x 34. Figs. 12-17 of Ant. rosacea, all x 7, except fig. 138, which is x 15. . Isolated first radial. a. ventral, b. dorsal, c. internal aspect. . Abnormally developed rosette, with two spout-like interradial processes (0) and a basal bridge (6. 6.) connecting the ends of two of the radial processes (p). x 15. a. ventral, 6. dorsal aspect. Lateral view of the base of the calyx with the centrodorsal piece in situ. Centrodorsal piece seen from its ventral side. 16 & 17. Pentagonal base of the calyx as seen from its dorsal (fig. 16) and ventral (fig. 17) sides. Puate V. The figures all x 7, except fig. 8, whichis x 15. Figs. 1-4 of Act. solaris. Figs. 1 & 2. Centrodorsal piece as seen from its dorsal (fig. 1) and ventral (fig. 2) sides. 3 & 4. Pentagonal base of the calyx as seen from its dorsal (fig. 3) and ventral (fig. 4) sides. Figs. 5-9 of Act. pectinata. Fig. 5. Interior of the calyx as seen after removal of the visceral mass. 6 & 7. Centrodorsal piece as seen from its dorsal (fig. 6) and ventral (fig. 7) sides. 8. An isolated compound basal. x 15. a. ventral, b. dorsal aspect. 9. An isolated first radial. a. ventral, b. dorsal, c. internal aspect. Figs. 10-15 of Act. robusta. Fig. 10. Internal aspect of an isolated first radial. 11-13. Two united first radials, together with those portions of the rosette which are in con- nexion with them, as seen from above (fig. 11), below (fig. 12), and within (fig. 13). In these four figures (10-13) I. indicates a bristle passed along the axial radial canal; II. another passed along the axial interradial canal; and III. a third, entering the central canal by one of the aper- tures on the internal face (z'), and coming out through the aperture of the commissural canal (co.c.) on the lateral face. Figs. 14 & 15. Centrodorsal piece as seen from its dorsal (fig. 15) and ventral (fig. 14) sides. Puate VI. All these figures are x 7, except figs. 6, 18, 19, 22, which are all x 15. Figs. 1-11 of Act. polymorpha, Type. Figs. 1-6, from one specimen. Fig. 1. Lateral view of the base of the calyx, with the centrodorsal piece in situ. 2. The same parts seen from the dorsal side. 120 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. Fig. 3. Ventral aspect of the centrodorsal piece. 4 & 5. Pentagonal base of the calyx as seen from its dorsal (fig. 4) and ventral (fig. 5) sides. 6. Two united compound basals as seen from their ventral side. x 15. 7,8, & 9, from a second specimen. 7 & 8. Centrodorsal piece as seen from its dorsal (fig. 7) and ventral (fig. 8) sides. In fig. 8 three of the rays of the basal star are seen occupying the basal grooves (db. g), their proper connexion with the rosette having been broken. 9. Pentagonal base of the calyx as seen from its dorsal side after removal of the rosette and basal star. 10 & 11, from a third and abnormally developed specimen. 10. Centrodorsal piece seen from its ventral side. 11. Dorsal aspect of the pentagonal base of the calyx. Figs. 12-15 of Act. polymorpha, vay. 1. Figs. 12 & 13. Pentagonal base of the calyx as seen from its ventral (fig. 12) and dorsal (fig. 13) sides. 14 & 15. Centrodorsal piece as seen from its dorsal (fig. 14) and ventral (fig. 15) sides. Figs. 16-19 of Act. polymorpha, var. 2. Figs. 16 & 17. Centrodorsal piece as seen from its dorsal (fig. 16) and ventral (fig. 17) sides. 18 & 19. Two united compound basals as seen from their ventral (fig. 18) and dorsal (fig. 19) sides. Figs. 20-22 of Act. polymorpha, var. 3. Figs. 20 & 21. Centrodorsal piece as seen from its dorsal (fig. 20) and ventral (fig. 21) sides. 22. An isolated compound basal as seen from its ventral (a) and dorsal (5) sides. Figs. 23 & 24 of Act. polymorpha, var. 4. Pentagonal base of the calyx as seen from its ventral (fig. 23) and dorsal (fig. 24) sides. Pratt VII. In this Plate are shown the first, second, and third radials of the type of Act. polymorpha (figs. 1-3) and of var. 2 (figs. 4-6). The different aspects shown are designated as follows:—a. Internal or proximal face; 4, External or distal face ; c. Ventral or superior face; d. Dorsal or inferior face. Act. polymorpha, Type. Act. polymorpha, var. 2. Fig. 1. First radial. Fig. 4. First radial. 2. Second radial. 5. Second radial. 3. Third or axillary radial. 6. Third or axillary radial. Puate VIII. All the figures are x 18. Figs. 1 & 2 of Act. polymorpha, Type. Figs. 1 & 2. Two successive oblique sections through the base of a decalcified calyx, viewed from their dorsal side. Fig. 1 is the more inferior, i.e. nearer the dorsal surface. Its left-hand lower portion shows the centro- dorsal piece only, with its marginal cirrhi (cir.) which receive fibrillar cords (n.c.) from the MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. 121 central mass (N) enveloping the quinquelocular organ. Proceeding outwards from the centre are seen five dark rays (S,), which represent the closely fibrillar organic basis of the five rays of the basal star. In the upper part of the figure two of them are very short, only their central ends being visible, as the section has here passed above the level of their outer ends through the substance of three first radials (7). Fig. 2. In this section only two of the basal rays (S;) are visible, as the greater part of it has passed above the level of the synostosis (/) between the first radials and the centrodorsal piece. In the centre are seen the chambers of the quinquelocular organ (ch), with their ventral openings into the vessels contained within the axial prolongation. At the right of the figure are seen the lower ends of the axial canals, both radial (@.r.c.) and imterradial (a.i.c.); their cavities are generally crossed by transverse septa, which divide them up into two or three intercom- municating smaller ones. The interval between every two of these canals is occupied by one of the secondary basal cords (Y2, Z,, Z:, &e.), produced by the bifurcation of the short primary cords (V, Y, &c.) proceeding from the angles of the quinquelocular organ. They are connected with one another laterally by interradial and intraradial commissures (é.co. & 7.co.) and enter the central canals of the first radials (r,) in successive pairs, so that the axial nervous cord (n) of each radius is composed of fibres derived from two primary basal cords (Y,, 4, &ce.), just as in Antedon. Fig. 3. A vertical longitudinal section through a decalcified calyx, passing on the right through the synostosis of two first radials (A, B) and the fibrillar basis (S,) of one of the basal rays, ana on the left through the segments (7r,,72,7.@.) of radius D. ‘The first of these is united to the centrodorsal piece by connective-tissue fibrils (/) similar to, but less abundant than, those in the interradial portion of the section, around which, in the natural condition, the cal- careous material forming one of the basal rays is deposited. The passage of the ventral furrows (v.rf., v.if.) into the axial canals (a.7.c., a.i.c.) is also well seen in this section. Its centre is occupied by the quinquelocular organ, from the ventral portion of which the axial prolongation (a.p.) rises into the circumvisceral celom (c.v.c.), which, together with the lower end of the wide axial ccelom (a.c), occupies the space between the ventral surface of the skeleton and the lower or dorsal wall of the convoluted alimentary canal. It is traversed by numerous connective-tissue septa, which divide it up into a system of spaces, communicating freely with those both of the intervisceral and of the axial ccelom (iv.c., a.c.). Fig. 4. Transverse section through the synostosis of two first radials (A, B) near the peripheral margin of the centrodorsal piece (cd.), showing the radial (2) and the interradial (S,) fibres which effect the synostosis between it and the united first radials. The latter form the organic basis of one of the rays of the basal star. Figs. 5-8. Four vertical sections, selected from a series, through a decalcified calyx of Act. pectinata. 5. Section through the adjacent inner ends of two first radials (4, B), showmg the axial inter- radial canal (a.i.c.) between them, and the open outer ends of the radial spaces (7.s.) between their dorsal surfaces and the ventral surface of the centrodorsal piece. The central ends of their axial nervous cords (”.) are cut very obliquely. 6. A section rather nearer the centre, showing the closed central ends of the radial spaces (r.s.) of the same two radii (4, B) and their axial canals (a.r.c.) ; also the four secondary. basal cords which unite in successive pairs (Z, V; and Vz, W,) to form their axial cords, cut obliquely. 7. A section from a little the other side of the centre, through the outer end of one of the chambers (ch) of the quinquelocular organ, corresponding to radius D. The radial spaces (r.s) of C and E are cut almost longitudinally; and above them, in the interior of the radials, are seen the axial nervous cords, with one of the two secondary basal cords 122 MR. P. H. CARPENTER ON THE GENUS ACTINOMETRA. (X,, Y.) by which each is connected with the central nervous envelope of the quinquelocular organ. ‘The other branches (X,, Y;) of the two primary cords (X & Y) combine to form the axial cord of the radius D. The inner end of its first radial is seen in the centre of the upper part of the figure (r,), separated from those of C & E by the axial interradial canals (a.i.e). Fig. 8. A section rather further from the centre of the calyx, showing the first radial of D cut trans- versely, with the closed central end of its radial space (7.s.). At the sides of the latter are the expanded dorsal ends of the axial interradial canals seen in fig. 7 ; they are received in depressions (s) at the central ends of the rays of the basal star, which are ossified around the vertical fibres (S,) only, and not, like the stouter more peripheral portions of the rays, around both vertical and diverging fibres, as is seen in Plate IIT. fig. 6, S,, S,. PHC PE: Traws. LINN. Soc. SER. 2 Zoou.Vou.11. Pu.1. Lt. Carpenter welt. ’ C. Berjeau, Lith DIAGRAMS OF AMBULACRAL DISTRIBUTION, SPECIES OF COMATULA Trans. LINN. Soc. SER. 2 Zoou.Vou.II. Pu. 2 Rebus &PH, Carpenter delts. ey ptt. VCE Banks & Co, Edin* C. Berjeau, Lith, SP AMBULACRAL DISTRIBUTION ACTINOMETRA 2-11. DISK, ARMS, BRACHIAL SEGMENTS, CENTRODORSUM & CALYX OF ACT, POLYMORPHA. abe é Rabus del. TrRaws. LINN. Soc. SpeR. 2 Zoou.Vou.Il. Pu.3 C, Berjeau, Lith, Beaks & Go, Edint ORAL PINNULES, SECTS RADIALS, MUSCLES & CIRRHI OF ACT. POLYMORPHA Type & Vars.1-4 & Fig 4. SECT. RADIALS -PENTACRINUS WYV. THOMSONI Rabus del. : Trans. LINN. Soc. SER. 2 Zoou.Vou.ll.Pu. 4. CALYX OF ANTEDON G Berjeau. Lith A, GELTICA, A. ESCHRICHTIL & A. ROSACEA Banks & Co, Edin® — mm» Trans. Linn. Soc. Spr. 2 Zoo. Vou Jl Pu.6 : GALYX OF ACTINOMETRA, A. SOLARIS, A. ROBUSTA & A. PECTINATA. Bays & Oo, Ean® n ae? 2 — q Rabus del. Trans. Linn. Soc. SER. 2 Zoou.Vou.I1. PL. 8. C. Berjeau, Lith, CALYX, BASALS AND CGENTRODORSAL PIECE OF ACTINOMETRA POLYMORPHA Type & Vars. 1-4 COL fi g.6 See ee ae. a C. Berjeau, Lith, DIFFERENT ASPECTS, FIRST, SECOND AND THIRD RADIALS ACTINOMETRA POLYMORPHA Type & Var2 Rasus del. Trans. Linn. Socsser. 2 Zoou.Vou. Il. Pu. 7. d Trans. Linn. Soc. SER. 2 Zoou.Vou.II. Pu.8. Fig. 2 ate C. Berjeau. Lith Banks & Co.. Edin® SECTIONS CALYX. BASAL RAYS AND RADIALS, ACT, POLYMORPHA Type. AND. ACTINOMETRA PECTINATA Pal The Fellows are entitled to receive, gratis, all Volumes, or Parts, of the oie or Journal that may fo published after they shall have paid the Admission Fee and First Annual Subscription ; and they may be supplied id with any of the Society’s Publications at a reduction of 25 per cent. under the common selling prices. Me Complete sets of the first 20 Volumes of the Transactions will be supplied to the Fellows at the reduced 34 ia * price of £20; or of the first 25 Volumes, with the General ss to the whole, for £30. h Any Fellow purchasing, at one time, any ten or more volumes of the Transactions, may obtain them at the | reduced price of £1 each for Vols. I. to XX., and £2 each for Vols. XXI. to XXV.; and single volumes or parts, teag complete sets, may be obtained at the original prices. A few copies of the several papers contained in the under~ mentioned yolumes have also been printed off for separate sale ; and most of them may still be had. The following parts of the Transactions have been published since the 25th yolume :— 5 ; Vol. When Price to the Price to, Vol. When Price to the ee Published. Public. Fellows. | Published. Public. Rist £8. “d. rig Maen ig ee & y evi Part 1608. .... 2 20 .... 21176 | Sencha - B af Dae a T.\Part: 21; 187ix fae Ooms Part III. 1869, .... 1 DD. >. Maud ob Pete 11 1iee Be eee 0th Oke Dn ro De eam Part Ill. 1876...., 012 0. RAVI. Part 1. 1969.....1 0 0.... 015 0 | Pat 0-1 ie c Part IL. 1870.....2 0 0....110 0 Pat v.13 Par U1. TAP eed ek 2. Dee OO Part VI. 1879.0 eae “ Bt TV. A87ieo-.. 1-0. O-.. 5. O15 20 XXVIII. Part UST. ea iy de OO 015. 0 Srconp Serres.—Zoonoey. Si Part Il. 1872..... 2 8.0. 116 0 | I; Part . I. 1987p eee 0185 Ores Pert iil. 1373. ac, 0 12-0 7.20 9 0°] Part II. 1875.....0 6 0 0. £6872 meer ary. 18S. wns 0492.00 .... F00'9. 50" Part ‘III. 1876355.) 3 650 1 te XXIX.Pert I. 1872.....2 2 0: 111 6 Part IV. 1877.:..) 0 16 0 1c. Olan Pee Tl: Aya aes 1 L681 Sr 20-4 Part V. 1877. .«.» 0 18 -Olassae Omen Part Tl. 1875). 8 000... 2 bi 04 Part VI. 18777 016 6) emer. Tt, 1874... 210 0.6.51 26a Part VII. 1878)..,. 1 160 202 tT, A074...) 5 4d 0 0 18 80: | Part VIII. 1879. ..,.0 1. Ose OL Part II. 1875..... 1 4 0....018 0 | II. Part 1. 1879... ae ee Members are requested to apply to Mr. Kirrrsr, the Librarian, for such volumes as they may be entitled to, or be desirous to purchase ; but no Volume can be deliyered gratis to a Fellow whose yearly Contributions are in arrear, nor can any be delivered unless applied for within five years from the time of publication. The Society also publishes an 8vo Journal, of which the Zoological and Botanical portions are s paged, so that either section may be taken separately,—the price to the Public being 12s, for each of the first L? _ eight volumes of the entire Journal, or 8s. per volume for either the Zoological or Botanical, section—the separate Re , Nos. being charged 3s. for the whole, or 2s. for either section. From the commencement of the 9th volume, Ag es the Botanical and Zoological sections are entirely separate, and each volume consists of eight Nos. bo! 2s, each, : 4 instead of four. om Ps - x ieck Dae Y ee volume, and to "Fellows 2s 3d. per Secebide. or 18s. the volume. a: EEA te The Journal is sent, post free, to Fellows resident in Britain, or to the Agents in. ane of efes = ebroad. Jae AA AAA SAAAAAAAA YAAAAAAALY, ; AARAAA AP a4 . saa AAP a\A ae La A AAAAAAAY anna Qa A ARAAAAAry \AAARAAS AAA 1), ,AsBARAAAA- ic Ran nach eae RR Fee Vi AAA JA oa a Oe PS AAA | N eeceascons is satannasneee ecco AAA “V\AAAARAL A Y AA : AAAS AAV AAA AYY. | AA RAAAAAS AP =| AAA RANA AAA ARAAAAA aaa steep P AVNIAYA ane a Aa AA ; A AAA RAR iP a ANAAANAAL ah AAAAAAR EE AAAAAA aR | AANA AA aaian RRA AANA A) . WARN AAA sAPARAARRARRARRAR AAA A; AAAAAR AY, ipa rrnanenna DAWAAAAAAAAAAAAAA AIA? ‘a sRARARRAAAE A AAAARAAAAAAS Banansaa aA aaa? Aaaaaaaa AY AAARA a , mann AAMBanannacne tt ay AAAA NAIA A AA AAR AANS saapoasaaeevereeeneron iN f a2 E RAAB RARARAAAAAR MAAAaaa nny AAA AAA SAAA A AAAAAARA o < ‘a \ i AAA A AAAAAAAAAARNNS paar RARAAAGGAGA AAANAAAAAAARAAAAnADAARARAA AANA AANA AF AA RARE js RA ap AANA nw APAAAAAR AAA RAAAAaan RARAAA rn An an. RAR ARAAAAAAAA AAA AAA AAAAAA RA RiAaalaiaa nia AAARAARAARTAN A} AR AARAAR ) : ' | AAA AAARAAA ae Fra.