‘ a ee: eri \ An STUDIES FROM THE MORPHOLOGICAL LABORATORY With Mr Sedgqwick's Compliments. From the Batrour Lrprary, New Museums, CAMBRIDGE. The Balfour Library will be glad to receive publications in exchange for the “Studies.” Vol. VI. London: Cc. J. CLAY AND SONS, CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, AVE MARIA LANE. 1896 STUDIES FROM THE MORPHOLOGICAL LABORATORY IN THE UNIVERSITY OF CAMBRIDGE. EDITED BY ADAM SEDGWICK, M.A., F.R.S. FELLOW AND LECTURER OF TRINITY COLLEGE, CAMBRIDGE, * AND READER IN ANIMAL MORPHOLOGY. Vol. VI. London: Cc. J. CLAY AND SONS, CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, AVE MARIA LANE. 1896 ' a < ce a ye om ® i 7 i a ragweed \ i aes vi CONTENTS. 1S. F, Harmer. On the occurrence of Embryonic Fission in Cyclo- stomatous Polyzoa. Plates I.—III. 28. J. Hickson. The Early Stages in the Development of Disticho- pora violacea, with a Short Essay on the Fragmentation of the Nucleus. Plate IV. 3A, Sepewick. On the Law of Development commonly known as von Baer’s Law; and on the Significance of Ancestral Rudiments in Embryonic Development 4A, Sepewick. On the Inadequacy of the Cell Theory, and on the Early Development of Nerves, particularly of the Third Nerve and of the Sympathetic in Elasmobranchii . ; : 3 5J. J. Lister. Contributions to the Life-History of the Foraminifera. Plates V.— VIII. 6J. Granwam Kerr. On some points in the anatomy of Nautilus Pompilius. Plates IX. and X. 7A, Sepewick. Further Remarks on the Cell Theory, with a Reply to Mr Bourne. : : ; s 3 : : : 7E. W. MacBripz. The Development of Asterina Gibbosa. Plates XI.—XXII. : F : : ; : : PAGE 45 75 93 108 181 213 221 1 Reprinted from the Quarterly Journal of Microscopical Science, Vol. 34. 2 Tbid. Vol. 35. 3 Ibid. Vol. 36. 4 Tbid. Vol. 37. 5 From the Philosophical Transactions of the Royal Society, Vol. 186, 1895. 6 From the Proceedings of the Zoological Society of London, 1895. 7 From the Quarterly Jownal of Microscopical Science, Vol. 38. On the Occurrence of Embryonic Fission in Cyclostomatous Polyzoa. By Sidney F. Harmer, M.A., B.Sc., Fellow of King’s College, Cambridge, and Superintendent of the University Museum of Zoology. With Plates I, IL & III. Tue results of the present paper have formed the subject of a preliminary communication made to the Cambridge Philo- sophical Society (16). The case of embryonic fission which I have now to describe in greater detail appears to me, on the assumption that my explanation of the observed facts is the correct one, to be without parallel in the animal kingdom. My observations refer entirely to the genus Crisia, and in particular to a form common at Plymouth, which I have described as a new species under the name C. ramosa (17). The general results may be stated as follows :— (i) The ovicell, which is morphologically equivalent to a zocecium, develops at the growing-point in the same way as an ordinary zoccium. (ii) A polypide-bud is found in the young ovicell, consisting of tentacle-sheath and a part which represents the alimentary canal of a polypide. (iii) Small egg-cells are present in various parts of some of the growing-points. One of these acquires a close relation to the potential alimentary canal of the ovicell-polypide. (iv) This potential alimentary canal grows round the ovum, losing its previous form, and becoming a compact multi- VOL. VI. 1 2 SIDNEY F. HARMER. nucleated follicle surrounding the egg, which at first lies in an excentric cavity in the follicle. (v) The ovum segments! and the blastomeres may, in early stages, be completely separated from one another, The rela- tions of the segmenting egg to its follicle are similar to those described by Salensky (28) in Salpa (cf. Salensky’s figs. 12, 13, on pl. x). (vi) The ovicell is meanwhile maturing, and by the end of the segmentation of the ovum has been shifted to some dis- tance from the growing-point by the superposition of new zocecia above it. Its non-calcified aperture, which, at an earlier stage, formed the wide end of a large funnel, has become constricted, and has grown out into a long tubular orifice. (vii) At the end of segmentation, the embryo consists of a small mass of undifferentiated cells, lying near the distal end of the follicle, which has increased largely in size, and now forms a spherical knob projecting freely into the interior of a spacious tentacle-sheath. A complicated arrangement of cells connected with the aperture has meanwhile been formed. (viii) The follicle becomes vacuolated, and is soon trans- formed into a nucleated protoplasmic reticulum. The tentacle- sheath loses its distinctness. (ix) The number of blastomeres increases, cell-limits being indistinguishable at this, as at all other stages, excepting the very earliest. (x) The embryo, having thus considerably increased in size, although remaining a solid mass, without differentiation of organs, grows out into several finger-shaped processes, which are generally directed towards the distal end of the ovicell. (xi) The finger-shaped processes are divided up by a series of transverse constrictions into rounded masses of cells, each of which becomes a complete larva. (xii) This process of embryo-formation continues during the whole functional period of the life of the ovicell, and is still actively proceeding at a stage when many of the embryos are mature, or nearly mature. The number of (secondary) embryos 1 The occurrence of a process of fertilisation was not made out. EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 3 present in an ovicell at any one time may exceed one hundred, and these have all been produced by budding from the above- described “ primary embryo.” (xiii) Each of the “secondary embryos” acquires its well- known two-layered condition at the time of its separation from the budding mass of embryonic cells. It develops in a vacuole of the protoplasmic reticulum, which presumably supplies it with nutriment since the embryo rapidly increases in size, be- coming ciliated externally, and ultimately escaping through the tubular aperture of the ovicell as a characteristic Cyclo- stome larva. Taking the above history into consideration, it is not sur- prising that, as is actually the case, the Cyclostome larva differs considerably in structure from that of other marine Polyzoa. This history also explains the fact that no observer has ever succeeded in giving an account of any process corre- sponding to egg-cleavage in Cyclostomata. The protoplasmic mass surrounding the embryos has been figured by Smitt (84), who has alluded to the yellow colour so characteristic of the contents of the ovicell. This colour is contained principally in the protoplasmic reticulum, although the embryos themselves have a yellowish colour. The first satisfactory account of the Cyclostome larva was, however, given by Barrois (1), who calls special attention to the fact that no previous observer had been able to discover ‘genital products” in any Cyclostome, and adds, “ Je n/’ai pour ma part encore réussi qu’é suivre les morulas jusqu’a des stades composés d’un nombre d’éléments de moins en moins nombreux et plus volumineux, sans réussir encore 4 constater d’une maniére bien certaine la présence de lceuf ; ”” although supposing that the Cyclostomes do not really differ from other Polyzoa in this respect.” Barrois’ failure to under- stand the early development of the embryos is readily explained if my own account be correct; and it is not surprising, con- 1 See his pl. iv, fig. 2. 2 L.c., pp. 58, 59, note. 4. SIDNEY F. HARMER. sidering the great difficulty of making out anything of the nature of the early stages except by means of sections. Barrois expressly states that the earliest stage to which he succeeded in tracing his morulas with certainty is that repre- sented in his pl. ili, fig. 83. This stage exactly corre- sponds with the condition at which I have found the embryos to be constricted off from the budding primary embryo (ef. Pl. II, fig. 11). Barrois was, how- ever, once successful in finding a cell, the egg nature of whieh he considers uncertain (his pl. 111, fig. 1); and in another case in finding what may have been an egg divided into two blasto- meres. It is not easy to say whether the former cell was really an egg, or whether it was merely one of the giant-cells described below. The rest of Barrois’ account contains an erroneous history of the later stages, which he himself was the first to correct (2). I am compelled to doubt altogether Barrois’ account here given (not accompanied by any figures) of the supposed occurrence of a process of segmentation of the egg, accom- panied by the formation of an epibolic gastrula. In a later paper (3) Barrois figures quite accurately the “ morula” at the stage at which it becomes independent (his pl. i, fig. 26), although he wrongly supposes that the inner layer of cells disappears in the later stages (his pl. iii, figs. 29, 30). Although Metschnikoff (23, pl. xx, figs. 61—64) gives admirable figures of the early embryos of Discoporella radiata, the earliest stage observed by that author is the stage at which the “secondary” embryo becomes free from the budding mass of embryonic cells. Ostromnoff (25) is no more fortunate in elucidating the early history of the embryo of Cyclostomata. I, Development of the Ovicell. This process, which takes place fundamentally in the same manner in all the species of Crisia which I have examined, has been to some extent described by Smitt (34), although most writers have paid little attention to the difference EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 5 between the form of the adult ovicell and that of the younger stages of the same organ. The ovicell is developed at the growing point, and it is here that the early stages in the development of the egg take place. A young internode! may be described as an acute-angled isosceles triangle with two sub-equal sides(AB, AC). Within the triangle a calcareous septum occurs parallel to AB, cutting off the oldest zocecium of the internode from the others. The next septum is parallel to AC, and is nearer to the base of the triangle. The formation of septa, alternately parallel to AB and AC, gives rise to a series of alternate zoccia, an arrangement characteristic of Crisia. The oldest zocecia are, of course, those nearest to the apex of the triangle, and the central part of the base is the region from which, with con- tinued growth, fresh zoccia are cut off. It remains to be stated that the growing-point, like the adult internode, is flattened, and that the openings of the zoccia are lateral, and are directed towards one of the flat surfaces of the branch. As the internode elongates, its proximal zocecia acquire their full length, and cease to take part in the formation of the growing-point. Or, explaining this by the former illustration, let the internode grow to twice its former length, the growing point remaining of the same width throughout its growth. By producing the lines of the septa already present it will be seen that if the growing point does not grow wider the older zocecia will be excluded from it, their growth being completed. It follows that the zoccia, several of which occur in a young state at the end of the branch, become successively shifted to the edges of the growing-point, preparatory to leaving it altogether. The growth of the zoccia and of the ovicells takes place by the apposition of fresh material at the distal end. The proxi- mal end of each unit of the colony is first laid down, and the last-formed portion is the aperture. Thus, by drawing a line transversely at any level across an internode, whether the in- ternode bears an ovicell or not, we obtain an accurate idea of 1 Compare Pl. IIT, fig. 15. 6 SIDNEY F. HARMER. the condition of the branch when the growing-point was at the level of that line. It follows from the shape of the ovicell, that an ovicell which is half grown will have the form of a wide-mouthed funnel, as shown in the figures of Smitt and others. But although it is easy to recognise a young ovicell at this stage, it is anything but an easy matter to distinguish the ovicell while it is still a sub-median member of the growing-point. The ovicell is indeed merely a modified zoccium, as is shown by the method of its development, as well as by its internal structure. Further evidence for this statement is afforded by the occasional occurrence of abnormal units of the colony, intermediate in form between the zoccia and ovicells (17, pl. xii, fig. 12). In Pl. III, fig. 19, the proximal portion of the ovicell is already developed. The first, second, and third units of the internode are zoccia, the fourth being an immature ovicell. The growing-point is formed, on the right side, by the base of a lateral branch, which would have been borne by the fifth member of the internode. There follow, in order from right to left, the fifth unit, the actual growing-point capable of pro- ducing fresh zocecia, the sixth unit, and the ovicell. The last occurs at the left side of the growing-point; but while its proximal end is in the same plane with the zoecia of the inter- node, the open end of the funnel is already projecting forwards (i. e. in the direction of that surface of the internode on which the zocecia open). This condition becomes more prominent at a later stage, so that the ovicell, in its most swollen portion, projects considerably beyond the level of the general surface of the internode. By referring to pl. xii, fig. 11, of my former paper (17) it will be seen that the zoccium “5” (in fig. 19) would have formed its aperture at the level of the middle of the ovicell, while “6” would have completed its growth at a very short distance above it. The young ovicell has, at first sight, the appearance of an open funnel. This is not really its condition, since its end is EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 7 closed by a chitinous uncalcified membrane (ectocyst), This is the condition of the ovicell, and of the growing-points generally, at all stages before their growth is completed and the definitive apertures are formed. The funnel, which is, in fig. 19, the most conspicuous part of the ovicell, is consequently merely that part of the ovicell in which calcification has occurred. In fig. 20 (in which the arrangement of the lateral buds does not correspond with that in fig. 19) the zocecia have been numbered in such a way as to facilitate comparison with fig. 19. The zoccium “5” is already complete, while “6” is beginning to free itself from the growing-point. The growth of the ovicell has progressed, the most swollen part is already completed, and the aperture (still closed by a membrane of uncalcified ectocyst) is beginning to constrict. In fig. 21 the zoccia “6” and “7” are complete. The aperture of the ovicell is still further constricted, and now consists of a slit-like portion which will soon close completely, and of a wider portion which will become the base of the tubular aperture so characteristic of this species (C. ramosa), A comparison of figs. 19—21 with one another will show that the method of the growth of the ovicell has been such as to bring its distal portion on to the front of the branch, while its proximal portion is lateral, and in series with the zoccia. The base of the tubular aperture thus comes to be situated at about the middle line of the internode. The valve of the ovicell (17, pl. xii, fig. 10) is formed as a ridge from the back of the ovicell at a stage between figs. 20 -and 21. The growth of the ovicell will be completed by the outgrowth of the tubular aperture. So far as I have been able to make out, the aperture is closed by the uncalcified membrane of ectocyst at all stages of its development, and does not become actually perforated until the escape of the first larva. I am quite unable to say when and how the process of fertilisation is effected. 8 SIDNEY F. HARMER. II. The Male Sexual Elements. There can, however, be no doubt of the existence of sperma- tozoa in Cyclostomes, although I am not aware that they have previously been described. In Crisia I have usually found them in colonies without ovicells' (17, p. 145), although they occur in ovicell-bearing colonies in Idmonea serpens, The spermatoblasts occur in masses filling up a large portion of the body-cavity of sexual individuals. The sperm mother- cells in both Idmonea and Crisia seem to occur in groups of four (Pl. I, fig. 4); and the four flagella when first developed appear, under insufficient magnification, as if they belonged to one cell. The mature spermatozoon (fig. 4) possesses an elongated head (measuring about °0064 mm.), and a long, active flagellum. In C. cornuta it was noticed that a delicate, hyaline layer of endocyst protruded from the aperture of the zoccium, during the escape of the spermatozoa, in the form of a cone at the apex of which the spermatozoa escaped. III. The Origin of the Secondary Embryos. My observations on this part of the subject have been made almost entirely by means of sections. The ovicells were pre- served and decalcified, at one operation, by placing in a mixture of corrosive sublimate, nitric and acetic acids. The most suc- cessful staining was obtained with Grenacher’s hematoxylin or with borax-carmine, in the latter case washing with alcohol containing picric acid. The free larva of C. eburnea is well figured by Barrois (1, pl. iii, fig. 22). It is, roughly speaking, cylindrical in shape, being covered externally by a complete coating of cilia. At one end of the cylinder is an aperture leading into the “sucker,” by means of which fixation is effected; and, at the opposite end, is another aperture leading into the so-called 1 In one case, spermatozoa were found in a colony of C. cornuta, which bore a single very young ovicell, EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 9 “ mantle-cavity.” I have observed no trace of a “ pyriform organ.” Barrois (2, p. 142; 3, p. 43, pl. iii, fig. 31) states, however, that he has discovered a rudiment of this structure in the larve of Discopora. The internal structure of a young larva may be illustrated by means of fig. 23, a median longitudinal section. The sucker is already well developed, having been formed, as in most other Ectoproct larve, by a process of ectodermic in- vagination. Cilia have appeared on the greater part of the external surface, the non-ciliated part of the ectoderm repre- senting the portion which will be later invaginated to form the mantle cavity. The inner layer of cells is still perfectly dis- tinguishable, forming a thin layer, closely applied to the ecto- derm, and enclosing a cavity which occupies the whole of the interior of the embryo. The earlier stages, which alone concern us at present, may be realised by assuming that the volume of the embryo shown in fig. 23 has become largely reduced; and that the sucker has become flattened out. Slightly anterior to the stage of fig. 22, the sucker is much shallower, and opens by a wide aperture in the middle of the “ oral” surface. Still earlier, the sucker is a very slight depression of the thickened “ oral ” ectoderm. The inner layer is at this stage a layer of great tenuity, in which a nucleus is thick enough to form a swelling wherever it occurs. Before this, the embryo is plano- convex, the position of the future sucker being represented by its flat side; and, still earlier, it is rounded in section, the inner layer consisting of a few cells, completely surrounding a central cavity. Between this stage and that shown in fig. 22, the inner layer may be separated, in parts or completely, from the ectoderm; so that it would be impossible to overlook its presence in any well-preserved section. At the earliest stage at which the embryo is free in the ovi- cell, it consists of a small rounded mass (PI. II, fig. 11) The outer layer is in the form of a continuous mass of proto- plasm, enclosing one layer of nuclei. The inner layer also consists of continuous protoplasm, with a very small number 10 SIDNEY F. HARMER. of nuclei arranged in one row; and it encloses a minute central cavity. Pl. III, fig. 17, represents a median section, slightly magnified, of an ovicell of Crisiaramosa. The ovicell con- tained in all about 115 embryos, which were embedded in a loose protoplasmic reticulum, filling up most of the cavity of the ovicell. In the older embryos, the conspicuous sucker or ‘internal sac” is clearly seen; and in some of them, a slit- like space which is the mantle-cavity. The aperture of the latter to the exterior is not shown in any of these embryos. To the left of the ovicell is the structure from which all the embryos have been produced. This structure is labelled “‘ primary embryo;” the evidence that this name implies its real nature being given in the sequel. The primary embryo is produced into several processes ; and indications are seen, in at least one case, that the end of the process is being constricted off, as a rounded mass of cells, which is equal in size to the smallest of the embryos found free in the protoplasmic reti- culum. Fig. 11 (Pl. ID) is a longitudinal section of a young ovicell, at the period when the formation of “ secondary ” embryos (i.e. embryos which are developed by budding from the “ primary” embryo) has just commenced. The proto- plasmic reticulum includes one or two free embryos, the structure of which has already been described. The most conspicuous structure in the section is, however, the large primary embryo, which consists of a dense mass of granular protoplasm containing numerous nuclei, and having an ex- tremely embryonic appearance. This structure is in a state of active growth, as is shown by the occurrence of nuclei with karyokinetic figures. The proximal end of the primary em- bryo is compact and rounded, and contains centrally a group of nuclei which are distinguished by the activity with which they are undergoing division. The opposite end of the primary embryo is produced into several irregular processes, which show constrictions at intervals. From the ends of two of these processes, embryos have just been constricted off, and EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 11 are seen disconnected from the primary embryo. The con- strictions indicate the limits of as many future embryos. The “primary embryo” contains, distally, an irregular cavity. It is difficult to be sure of the exact arrangement of the embryogenic processes; but in some cases at least it is evident that the distal end of the primary embryo has the form of an irregular cup, the processes forming the wall of the cup, from which they become free at their ends. The irre- gular cavity seen in fig. 11 is part of the cavity of the cup. Towards the ends of the processes an ectodermic layer becomes clearly differentiated ; while, in the centre of each of the swellings indicating a future embryo, a small group of inner-layer cells can, in some cases, be clearly distinguished. The ectoderm of the processes is continuous with the outer nucleated layer of the primary embryo, and with the similar layer immediately lining its distally-placed cavity. The inner- layer cells of the secondary embryos are continuous with the inner nuclei of the more solid, proximal half of the primary embryo. But these nuclei and the protoplasm surrounding them are not throughout clearly differentiated from the outer layer of nuclei. I am, however, inclined to suppose that the somewhat triangular, clear mass of protoplasm at the proximal end of the primary embryo, containing actively dividing nuclei, is the region which gives rise to the inner-layer cells. This region can generally he distinguished with ease in ovicells at this stage. The primary embryo consists of a mass of embryonic cells (or, rather, nuclei embedded in continuous protoplasm) which are obscurely differentiated into outer and inner cells (or nuclei). The whole function of this embryo is to act as an embryogenic organ, or producer of secondary embryos, and it possesses no structures which can be described as its own organs, At its proximal end, the primary embryo is budding off nuclei which migrate into the protoplasmic reticulum, where they become indistinguishable from the rest of the nuclei of that reticulum. I have been unable to make out the 12 SIDNEY F. HARMER. significance of this phenomenon, which I have frequently observed. I am in a position to multiply indefinitely figures showing the important fact that the young larve are really produced as buds from a “ primary embryo.” I consider that I have the clearest possible evidence of the following statements : i. The larve are produced as buds from an em- bryonic mass of cells found in the young ovicell. ii. They are produced in no other way than that mentioned under i. The embryogenic organ is invariably present in all ovicells in which young embryos are found, and in most of the older embryo-containing ovicells as well. It is still active, even at the stage shown in fig. 17. The youngest embryos, free in the reticulum, are invariably identical in structure with the ends of the processes of the primary embryo, and there is not the slightest trace in any of the ovicells, young or old, of the de- velopment of larve by the ordinary process of the segmentation of an egg. It might, indeed, be supposed that the bi-nucleated cell shown in the upper part of the reticulum in fig. 11 had the nature of a dividing egg. This supposition is not confirmed by an examination of the actual facts. While the evidence in favour of the origin of the larve by a process of budding is unmistakably clear, there are no transitions between such cells as the large one shown in fig. 11 and the young two- layered larve. These large cells, which are normally present in the ovicells, are probably of the nature of “ giant-cells,” similar to those which are found in developing bone. This subject will be considered later; but it may be pointed out that it is possible that the supposed egg-cell figured by Barrois (1, pl. iii, fig. 1) may have been one of these giant-cells. IV. The Development of the Primary Embryo. Fig. 15 (Pl. III) is a decalcified internode of C. eburnea, possessing a very young ovicell. The internode consists of one complete zocecium, which bears the beginning of a lateral EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 13 branch; of a second zocecium, which is very nearly mature ; of the ovicell as the third member of the internode; and of the real growing-point, which contains a young polypide-bud ; although the corresponding zoccium is not yet separated off from the growing-point by a septum. The ovicell contains a structure which is the exact equivalent of an ordinary polypide-bud. This consists of (1) a thick (proximal) mass of cells, which in a zocecium would give rise to alimentary canal and tentacles; (2) a thin-walled portion, next to the above, and corresponding to the tentacle-sheath; and (3) a distal portion, indicated by two parallel lines in the sketch, and which is really an invagination of the distal body-wall of the ovicell, This is formed in a precisely similar manner in any young zocecium, where it develops into the aperture. Fig. 1 (Pl. I) is a nearly median longitudinal section of an ovicell at nearly the same stage as fig. 15. The body- cavity is, as in ordinary zoecia, largely filled up by funicular tissue, but contains an obvious polypide-bud, the distal portion of which can be clearly distinguished as a tentacle-sheath, similar in all respects to the same structure in an ordinary polypide-bud, The one fact, indeed, which enables this member of the colony to be distinguished as an ovicell is the presence of a relatively large cell, which is closely applied to one wall of the polypide-bud. The latter shows some tendency to give off cells which are growing round the large cell. This has a diameter of about ‘0176 mm., and it has a large clear nucleus with one or two nucleoli. Its structure, in fact, reminds one irresistibly of that of an egg; and I believe this cell to be the source from which all the larve produced in the ovicell are developed. ** Eggs” of this kind are found in various positions in some of the growing-points, Thus in the particular individual in question there is a second, smaller egg! in the same ovicell ; and in the next zocecium there are two eggs!, one of which is at the apex of the polypide-bud. The fact that these eggs are commonly found in the growing-points leads me to suppose ? Not visible in the particular section figured. 14 SIDNEY F. HARMER. that several are produced in each fertile internode, apparently by a modification of cells of the funicular tissue, and that their further development depends on their entering into definite relation with a polypide-bud. If this association is brought about, it may be assumed that what might at first have de- veloped into a zocecium becomes an ovicell. In abnormal cases, where several polypide-buds enter into relation with ova, two or more ovicells may be produced in the same inter- node (17, p. 166; pl. xii, fig. 13). It may further be sup- posed that the failure to bring about the association between the egg and the polypide-bud results in such abnormalities as that shown in fig. 12 of my former paper, and that this or some other cause, such as the failure to get fertilised, results in the development of the empty ovicells which are so frequently observed. On decalcifying a number of ovicells, it is soon noticed that many ovicells are either completely empty or are abnormally developed. An ovicell, with complete tubular aperture, may be absolutely devoid of any trace of primary or secondary embryos. In some cases, these empty ovicells are probably the result of degeneration which has set in after the comple- tion of the process of development of free larve. After the escape of the last larve, the remaining tissues of the ovicells degenerate, and are gradually absorbed. Many of my sections bear out this assertion. In other cases, however, the degeneration takes place in ovicells which have produced no larve. Empty ovicells which are near the growing-points are, probably, generally of this character. It is easy to obtain evidence of the fact that, in such cases, degeneration may set in at various periods—some- times after the egg has developed to a considerable extent. In some cases, this may be the result of the absence of fertili- sation—a process of which I have vainly endeavoured to prove the existence. That fertilisation does actually occur at some period can hardly be doubted, considering the fact that normal spermatozoa are developed in some colonies. In other cases, the degeneration is probably due to the atrophy of the poly- EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 15 pides in the zocecia contiguous to the ovicell. It is well known that the thick calcareous ectocyst of the Cyclostomata is perforated by pores. On decalcifying a colony, and stain- ing what is left, it can be easily shown that all the zocecia are in organic connection by means of the funicular tissue, which passes through the pores from one zocecium to another, and from the zocecia to the ovicell. It can hardly be doubted that the nutriment at the expense of which the larve develop is provided by means of the protoplasmic network which thus connects all the individuals of a colony. The ovum is ex- tremely minute, although it gives rise to a massive primary embryo; and this to numerous free larv, each of which is very many times larger than the original ovum. This rapid growth—to say nothing of the development of an extensive reticulum of funicular tissue in the ovicell itself—can only depend on the existence of pores by which the ovicell is con- nected with zoccia which possess functional polypides. In fig. 2 the ovum is completely surrounded by the polypide- bud, whose tentacle-sheath has considerably increased in size ; while in fig. 3 further alterations of importance have taken place. The tentacle-sheath has grown very much larger ; but, so far as this structure and the invagination which forms the aperture are concerned, the ovicell still resembles an ordinary zocecium. The proximal part of the polypide-bud, which in the younger ovicell was practically indistinguishable from the corresponding structure in an ordinary zocecium, has now become much modified. The egg is now completely sur- rounded by it; and the polypide-bud has in fact transformed itself into a round mass of cells which may be termed the “ follicle.’ The ovum lies partly surrounded by a cavity in this follicle. The fact that the distal endocyst is not in contact with the ectocyst is probably due to shrinkage brought about during decalcification. The side-walls of the zocecium are of course calcified (ef. fig. 19), while the distal ectocyst forms an uncal- cified membrane stretching across the mouth of the funnel formed by the ovicell. 16 SIDNEY F. HARMER. It is presumably at this stage that fertilisation takes place ; but I have in vain looked for any evidence of perforation in the terminal membrane of the ovicell, or for traces of sperma- tozoa inside the tentacle-sheath. This fact is not really sur- prising when it is remembered that the finer details of the highly calcified ovicell of Crisia can hardly be examined except by means of sections; and that the spermatozoa are very minute. In fig. 5 the whole ovicell has considerably increased in length. Its irregular form is of course due to shrinkage caused by the action of reagents. The ectocyst is not repre- sented in the figure. The ovicell was probably at about the stage represented in fig. 21. The valve (ef. pl. xii, fig. 10, of my former paper) is now developed as a fold of the ectoderm on the back wall of the ovicel], The aperture has no longer any obvious opening to ‘he exterior ; and the tentacle-sheath has increased in size, its walls.,having become very thin, except at its distal end, which is considerably thickened. The follicle is slightly larger than before, and its nuclei have obviously increased in number. In place of the egg found in the preceding stage, there are now three egg-like cells, which are not in contact with one another; and which I regard as blastomeres. Remains of the follicle-cavity are still present. Although I have no direct evidence that the “ blastomeres ” are really derived from the egg, their subsequent history leaves room for little doubt on this point. The details of the forma- tion of the primary embryo in Crisia remind one strangely of the early development of Salpa, as described by Salensky (28). This is true not merely of the segmentation of the ovum, but also of the later relations of the embryo to its follicle. Salensky states, for instance, that the blastomeres of Salpa may at first be entirely disconnected from one another (i. c. pl. x, fig. 10; pl. xxii, figs. 3, 4). In the next stages, of which I have numerous preparations, but which I have not figured, the number of blastomeres gra- dually increases. I have been unable to make out any regu- EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 17 larity in the succession of the blastomeres, which are, in fact, inextricably entangled among the follicle-cells. They are not necessarily in contact with one another, but may be separated from one another by ingrowths of the follicle-cells, so that in most preparations it is almost impossible to count the number of the blastomeres, or to distinguish all of them from the fol- licle-cells. An excellent idea of the general relation of the blastomeres to the follicle-cells (or nuclei) may be obtained by referring to some of Salensky’s figures of Salpa, as his pl. x, figs. 12,13. The only difference that I can point out between Crisia and Salpa, as regards the relation of the blastomeres to the follicle-cells, is that in Crisia the follicle is somewhat larger relatively than in Salpa, and that the blastomeres occupy only the central region of the follicle instead of filling up most of that structure, as in Salpa. Remains of the follicle-cavity may still be detected in some of these stages. In fig. 6 (which is connected with fig. 5 by numerous pr-- parations, forming a perfectly continuous series, in my pos- session) the tentacle-sheath has increased in size so much as to fill up nearly the whole of the ovicell. The follicle has largely increased in size, and now forms a sub-spherical knob, projecting freely into the cavity of the tentacle-sheath. This stage is a perfectly constant and easily recognised one. The scattered blastomeres have at last come together to form a small but compact embryo, in some of the nuclei of which karyokinetic figures are discernible. There is no trace of the differentiation of germ-layers in the embryo, which consists simply of a small rounded mass of undifferentiated embryonic cells, or rather of a continuous mass of protoplasm, containing nuclei scattered through it without any attempt to arrange themselves in definite layers. At the distal end of the embryo is a clear part of the follicle which contains small nuclei. This is apparently a constant feature of the stages near this one; but I have not been able to make out its significance. Fig. 7 is not cut quite medianly, so that it does not show that the attachment of the follicle to the tentacle-sheath is VOL, VI. 2 18 SIDNEY F. HARMER. much the.same as in the former figure. The tentacle-sheath is, however, now becoming less definite. To the left of the figure it is hardly distinguishable from the follicle, with which it probably fuses. The embryo is practically unaltered, except that it has come to the surface of the follicle; but the principal difference between this and the earlier stage concerns the follicle itself. This structure no longer forms a compact mass of granular, nucleated protoplasm, as in fig. 6; but it has become distinctly vacuolated. This vacuolation is the beginning of the process by which the follicle of earlier stages is transformed into the protoplasmic reticulum of later stages. Fig. 9 (Pl. II) well illustrates the manner in which this transformation is effected. The tentacle-sheath is not so clear as in the former stage, and can, indeed, hardly be distinguished except at its upper end. Whilst in earlier stages it filled up most of the ovicell, it has now collapsed to a large extent. The proximal portion of the follicle is in this ovicell stall solid, and is perfectly similar in structure to the solid follicle of fig. 6. Distally the follicle is almost unrecognisable, having become separated by enormous vacuoles into strands of anas- tomosing, nucleated protoplasm. These strands are, however, most unmistakably continuous with the proximal, solid portion of the follicle. The embryo is practically unaltered, still forming a small rounded mass of undifferentiated embryonic tissue lying in a part of the reticulum. The great increase in the size of the follicle and in the number of its nuclei up to the stage shown in fig. 11 is pro- bably connected with the development of a nutritive arrange- ment for the young larve. The minute egg-cell of fig. 1 gives rise, as I believe, to the embryogenic organ of fig. 11, and this to the numerous young larve with which the mature ovicell is crowded. These larve lie in the meshes of the protoplasmic reticulum, from which they are probably supplied with nutri- tive material. In figs.6 and 9 the base of the ovicell has a very characteristic structure, always noticed in young ovicells at certain stages. EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 19 Next to the ectocyst comes a very definite nucleated layer, which encloses a network of cells separated by smallish vacuoles. In later stages the outer definite layer of nuclei disappears, and the network becomes continuous with the reticulum formed by the modification of the follicle (fig. 11). The basal network of cells is obviously part of the ordinary funicular tissue, which, as has already been pointed out, forms a con- tinuous connection from zocecium to zocecium, or from zocecium to ovicell, through the pores in the calcareous septa between neighbouring individuals. It can hardly be doubted that the rich protoplasmic reticulum in which the young larve lie is the means by which nutriment is conveyed to the developing larve. Fig. 10 is a stage of which I obtained only one example, and I cannot be sure that what is there represented is really a normal process. This preparation indicates that a kind of invagination takes places in the embryo at this stage (which is certainly very near that shown in fig. 9). If this is really correct, the inner layer of the primary embryo may possibly be formed by a process of invagination, and the inner layer of the secondary embryos is then probably derived from this invaginated layer; but I make these suggestions with all reserve. Fig. 8 is interesting partly because it supports the view advanced in my former paper (17) that the species there described as C. ramosa is not identical with C. eburnea. The figure is a longitudinal section of an ovicell of C. eburnea; and so far as the general development of the ovicell goes, the age corresponds with the stage shown, for C. ramosa, in fig. 6. The tentacle-sheath is at its period of maximum develop- ment; but the relative sizes of follicle and embryo are widely different from their relative sizes in C. ramosa. This appears to be a constant difference between the two species. In C. eburnea, the follicle is reduced to a minimum, and the quantity of the protoplasmic reticulum of mature ovicells is, 20 SIDNEY F. HARMER. consequently, appreciably smaller than that in C. ramosa, although in old ovicells of the latter species even, the quantity of the protoplasmic reticulum may be considerably reduced, by the development of the larve at its expense. The primary embryo, on the contrary, is relatively very large. It has differentiated a distinct external layer of nuclei, which will give rise to the external layer of the secondary embryos. Throughout the development, the budding secondary embryo of C. eburnea differs considerably from that of C. ramosa; although the fundamental facts are the same in both species. I have, unfortunately, no satisfactory sections of the ovicells of other species, which I found more difficult to obtain than the two former species. It is necessary to have a large stock of material in order to study the development ; as it usually happens that a very small proportion of the colonies found are provided with ovicells. The stages intermediate between figs. 9 and 11 have not been figured ; but it is easy to describe their general develop- ment. After the stage shown in fig. 9 (but not until then) the embryo increases in size, and rapidly transforms itself into the characteristic mass of embryonic cells from which the young larvee are budded off. The history of the aperture of the ovicell has, so far, not been considered in sufficient detail. Its commencement as an invagination of the endocyst has been seen in fig. 3. When the egg has begun to segment (fig. 5), the opening of the invagination has closed. The distal end of the tentacle-sheath is, however, now thickened ; and the valve is commencing to develop. In fig. 8 (C. eburnea), the valve is practically complete ; the distal thickening of the tentacle-sheath has increased, but the invagination constituting the primary aperture has not materially altered. The ovicell is completely calcified except in the region of its aperture, which is beginning to grow out into its tubular form. This part is covered merely by uncal- cified ectocyst. Fig. 12 represents a slightly earlier stage in C, ramosa. EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 21 The primary aperture is still present ; the distal thickening of the tentacle-sheath being already distinct, and showing a dif- ferentiation of an external epithelial layer, and a more inter- nally placed mass of nucleated protoplasm. It is very difficult to make out with certainty the later history of the aperture. In stages previous to that at which the solid follicle has its maximum size, the connection of the primary aperture with the distal wall of the ovicell seems to be completely lost in many cases. It is perhaps the case that the original invagination remains connected with the distal wall of the ovicell by a thin cord of cells which is not easily seen in sections—accounting for the apparent discontinuity between endocyst and apertural invagination which is fre- quently remarked ; and that, later, this cord shortens, bring- ing the invagination once more nearer to the distal wall of the ovicell, where its aperture once more opens out widely. In fig. 13, representing the aperture of an ovicell in which traces of vacuolation are beginning to appear in the follicle, the primary aperture still opens to the exterior ; although, in accordance with what has just been said, its opening would probably not have been discernible at a somewhat earlier stage. The formation of the tubular definitive aperture has progressed, and the valve is complete. The differentiation of the thickened part of the tentacle-sheath into two kinds of cells, alluded to in the description of the last figure, has advanced a stage. In fig. 9 the tubular aperture is practically complete. At its end is seen an invagination which I regard as the remains of the primary aperture, but which has now become disconnected from the thickened part of the tentacle-sheath. The history of the aperture is thus, according to what I believe I have made out, as follows :—During the calcification of the distal end of the ovicell, the primary aperture, which at first opened in the middle of the mouth of the funnel, becomes shifted nearer the “ back ” wall of the ovicell, into the position where the tube of the ovicell is to be formed. The aperture is thus not closed by the calcification of the ovicell, but finally 22 SIDNEY F. HARMER. disappears in the region of the permanently uncalcified part of the ovicell; i.e. of the definitive aperture. Fig. 14 illustrates a condition of the tube of the ovicell which I have observed in one or two cases. The ectocyst is drawn out into a long narrow tube, which was probably un- calcified and which opens to the exterior. This recalls the condition described in the zocecia of certain Cyclostomata, in which the zocecium is closed by a (calcareous) lid, perforated by a small central aperture.’ I am not prepared to state whether or not this is a normal character of ovicells at any particular stage, nor can I suggest any satisfactory explanation of the meaning of the phenomenon. The central mass of cells differentiated from the thickened part of the tentacle-sheath in fig. 12 is destined to undergo certain very important modifications. In fig. 16 (more highly magnified than the previous figures, and belonging to the same ovicell from which fig. 6 was drawn) some of the nuclei of the central mass are growing larger. This is especially the case in the neighbourhood of the lumen of the tentacle-sheath, where there is a tendency for the nuclei to group themselves in small numbers. In fig. 9 some of these multinucleated masses of protoplasm are breaking off into the lumen of the tentacle-sheath. From this stage onwards the characteristic multinucleated cells, which have been formed from the thickened distal part of the tentacle-sheath, are a normal feature of the ovicell, being found in the vacuoles of the protoplasmic reticulum in which the young larve lie. One of them is seen in fig. 11, and others are shown, more highly magnified, in fig. 18 (Pl. III). In the latter figure the multinucleated cells contain nuclear and other structures which are obviously degenerating ; and they are clearly not unlike the “ giant-cells ” which are known to occur in certain tissues in Vertebrates. The giant-cells make their appearance at just that stage 1 Cf. Mesentipora meandrina (Busk, No. 8, pl. xvii, fig. 2); Reticu- lipora dorsalis (Waters, No. 36, pl. xvii, fig. 4); and other cases referred to by Waters, EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 23 when the vacuolation of the follicle is commencing ; and they appear to be closely connected with the carrying out of this process of vacuolation, although it is clear that the first small vacuoles (cf. fig. 7) make their appearance independently of the giant-cells. At later stages each giant-cell is usually seen to lie in a large, sharply-marked vacuole of the protoplasmic reticulum. The cell may be apposed to one wall of the vacuole; or may, apparently, lie quite freely within it. There is good reason to believe that the vacuoles which are at first occupied by giant- cells are later occupied by the young larva. Hach embryo, soon after its formation, comes to lie in a sharply-marked vacuole in the protoplasmic reticulum. Taking into consider- ation the facts (1) that the giant-cells are formed simul- taneously with the appearance of large vacuoles in the follicle, and (2) that they contain fragments of degenerating cells or nuclei, it may be concluded that one of the functions of the giant-cells is to excavate spaces in the follicle in which the larvee can develop. These spaces are probably filled with some albuminoid fluid, at the expense of which the embryos develop —probably by diffusion through their tissues, as they have no recognisable means of absorbing nutriment. The function of the giant-cells would thus be closely similar to that of the osteoclasts or myeloplaxes of bone “ which ex- cavate small shallow pits... . in the part which is under- going absorption” (27, p. 104). Their structure, too, is in accordance with the descriptions of various observers of the multinucleated giant-cells in Vertebrates. In the mature ovicell the remains of the distal thickening of the tentacle-sheath are always found as a dense mass of nucleated protoplasm which is attached to the ectocyst, not in the tubular aperture of the ovicell, but invariably at its base, on the side which is further from the back of the ovicell (figs. 9 and16). The valve constantly projects from the back of the ovicell into the proximal part of this mass of cells in the manner shown in figs. 9 and 16. It appears to me probable that the function of the valve is to offer an obstacle to the 24. SIDNEY F. HARMER. escape of the immature larve. When mature, the larve force their way one by one through the solid mass of protoplasm into the tubular aperture, and so escape to the exterior. The tentacle-sheath is no longer easily distinguishable in the mature ovicell. With the commencement of the vacuola- tion of the follicle its distinctness vanishes, and it becomes confounded with the vacuolated follicle. The relations shown in figs. 9 and 17 probably indicate that the follicle ultimately fills up the whole of the original tentacle-sheath, and that that part of the ovicell which is not occupied by the protoplasmic reticulum and its contents is the original body-cavity of the ovicell. The ovicells which are at their period of greatest activity can readily be recognised in the living condition by the pronounced yellow colour of their contents. This is sufficiently distinct to show clearly through the calcified wall of the ovicell. Although the embryos and larve are pale yellow, the colour of the ovicell depends mainly on the pigment in the reticulum which supports the embryos. In C. cornuta this is bright red- orange in colour, while in C. ramosa the orange colour is not quite so bright. The oldest larve, which are almost ready to escape, lie each enclosed in a distinct vacuole of the reticulum, in close contact with the thick mass of protoplasm which fills up the aperture of the valve. The production of embryos continues up to a very late stage, but embryos are always developed only from the budding mass of embryonic cells (primary embryo). ‘The budding organ has, however, a somewhat different appearance in old ovicells from that which it first had, both the secondary embryos and their nuclei being markedly smaller than in the younger ovicells. So far as my observations go, the whole of the budding organ is ultimately used up in the production of embryos. In ovicells which are nearly exhausted the embryos are few in number, and the budding organ has been reduced to small dimensions. Finally, the ovicell is found to consist merely of EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 25 a protoplasmic reticulum, which may be richly developed, and which no longer contains any embryos or larve. It must not, however, be assumed that all ovicells in this condition have passed through an embryo-producing stage. I have repeatedly obtained evidence to show that degeneration of the ovicell may start at almost any stage in its development. The external form of the ovicell may develop completely, even if the embryo and its follicle are degenerating. It is common to find an ovicell which, from its proximity to the growing- point, should be a young one, but which appears completely empty in sections. I can only assume in these cases that the protoplasmic structures which the ovicell at first possessed have been absorbed through the pores into the neighbouring zocecia. In other cases the ovicell may contain remains of a degene- rating follicle, the degeneration having clearly commenced before the follicle became vacuolated. One may, therefore, distinguish between a “ primary” degeneration of the ovicell occurring before any larve have been produced, and a “ secon- dary’ degeneration, which has taken place after the escape of the last larva. It is sometimes possible to distinguish between these two conditions by reason of the fact that the base of the young ovicell is limited by a marked epithelial arrangement of its endocyst (cf. figs. 6 and 9). I have so far avoided the use of the term “‘ endoderm” as an equivalent for the inner layer of cells of the secondary em- bryos. This layer is excessively distinct in the embryos soon after their liberation from the embryogenic mass of cells. It then forms an epithelium, lying more or less close to the inner surface of the ectoderm-cells, and completely surrounding the whole internal cavity of the embryo (fig. 22). In later stages the distinctness of the cavity becomes lost, and its lining cells send off processes which grow across the cavity and convert it into an irregular set of spaces. At the sides of the sucker these spaces disappear altogether, while between the sucker and the middle of the aboral pole the cavity remains distinct for a time longer. Its cells become, however, almost indistinguishable from the epithelium of the 26 SIDNEY F, HARMER. sucker on the one hand, and from the epithelium lining the mantle-cavity on the other. I formerly assumed (15, p. 455), on the authority of Ostrou- moff’s statements (25, pl. vi, fig. 1), that the cavity lined by the inner cells represented the alimentary canal of the larva. But, taking into account the manner in which the larve are developed, it appears to me doubtful whether any representa- tive of the endoderm occurs in them. It appears to me to be satisfactorily established that a young polypide-bud in any Polyzoon is developed at the expense of two layers, viz. the ectoderm and a layer of funicular tissue which may be re- garded as mesoderm.! The metamorphosis of the larva of Cyclostomata has been described by Barrois (3) and by Ostrou- moff (25). The observations of Barrois show that the pro- cesses of fixation and of metamorphosis take place essentially as in other Gymnolzemata. The larva fixes by the eversion of its sucker, its mantle being rolled downwards so as to come into contact with the flattened plate formed by the eversion of the sucker, and the greater part of the larval tissues undergo a process of histolysis. The larva thus enters into the condition of a zocecium containing a “ brown body,” and the young poly- pide is produced by an invagination of the body-wall from the centre of the surface opposite to the basal surface. While the inner layer of the bud is formed by an invagination of the ectoderm, Barrois was unfortunately unable to trace the history of its outer layer. Ostroumoff is but little more definite on this point. The inner layer of the bud is formed, according to this observer, not as an invagination, but as a plate of cells split off from the aboral ectoderm. The edges of this plate curve round, so as to transform the plate into a sac, to the outer side of which ‘*mesenchym-cells ” apply themselves, and form the outer layer of the bud. The origin of these “ mesenchym-cells ” is not traced. It is recognised that the “ alimentary canal” of the earlier stage disappears, but there is nothing to show how its cells are related to the “‘ mesenchym-cells” shown in 1 Cf. especially Seeliger, Nos. 82 and 38. EMBRYONIC FISSION IN CYULOSTOMATOUS POLYZOA. 27 Ostroumoft’s pl. vi, fig. 2, which, by the way, are unlike any cells which I have ever seen in a Cyclostome larva. Whatever be the origin of the outer layer of the bud which which forms the primary polypide, it is quite clear that that polypide is formed in fundamentally the same way as any other polypide in the future colony. There can be no question of the ‘alimentary canal” of the embryo passing over directly into that of the primary zocecium. In default of sufficient evidence on this point I am inclined to regard the inner layer of the Cyclostome embryo as meso- dermic rather than endodermic, and this principally on the following grounds: 1. The alimentary canal is an excessively rudimentary struc- ture in the great majority of known Ectoproct larve. 2. The peculiar character of the early development of Crisia suggests that a representative of this rudimentary structure is likely to be found in the primary embryo only, and that the secondary embryos, formed by budding from the primary one, are no more likely to possess an alimentary canal than is a young zoccium formed at the growing-point of an old colony. 3. The analogy of other Ectoprocta is in favour of this hypo- thesis.' Prouho (26), for instance, has given an account (which I can confirm in the main from my own observations) of the metamorphosis of Flustrella. Even before the end of larval life, a distinct aboral mesodermic layer is present, from which the outer layer of the bud is directly developed. In the course of the budding of an ordinary Ectoproct colony the polypide buds are formed from two distinct layers. The inner layer of the bud is developed at the expense of the ectoderm; the outer layer, either from an already definite layer of mesoderm (Phylactolzemata), or from mesoderm-cells of the funicular tissue which arrange themselves as an epithe- lium round the outside of the ectodermic portion of the bud 1 Cf. particularly the larva of the Phylactolemata, as described by Braem (5) and-by Davenport (8a), 28 SIDNEY F. HARMER. (most Gymnolemata; cf. especially Seeliger, No. 33). There is no sufficient reason for supposing that a young zocecium consists of anything but ectoderm and mesoderm. The Ecto- proct larva may be considered morphologically as a young zocecium containing a potential “ brown body” (the remains of the purely larval organs), and it is not unreasonable to suppose that the structures found in the larva of the Cyclo- stomata, developed as it is by a process of budding, are com- parable with those which are found in a zocecium. We arrive, therefore, at the provisional conclusion that the inner layer of the Cyclostome embryo is more likely to repre- sent the mesoderm than the endoderm of the larva. There can be no doubt that, on the assumption that my account of this process is in the main correct, the development of Crisia takes place in a manner to which there are few known parallels. The most frequently quoted case of embryonic fission is that of Lumbricus trapezoides, in which, according to the statements of Kleinenberg (20), the embryo normally divides into two complete embryos at the gastrula-stage. In some abnormal cases, however (1. c., p. 217), a single embryo is first formed; and this gives rise to one or more embryos produced as buds on the margin of its mouth. The segmentation of the egg is described as being much less regular than in other species of Lumbricus, in which no embryonic fission takes place. An equally striking case of the same kind had previously been described by Busch (7), in Chrysaora. In only a few cases does an egg develop into a single embryo. In the other cases, the embryo gives rise to one or two buds, apparently at the gastrula-stage ; the buds becoming free larve, and deve- loping fresh buds. Not only does Busch claim to have followed the whole process in an isolated individual, but he states that each time that the water in which the young larve were kept was changed, two thirds or so of the embryos were thrown away, and that this loss in number was compensated for, by the next day, by the gemmiparous habit of the larve EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 29 (l. c., p. 28). This account is confirmed by Haeckel (14), who observed the production, in three weeks, of 60—80 buds, from ten isolated gastrule of Chrysaora. The process of larval fission or gemmation is known to be even more remarkable in Aurelia (Haeckel, l. c.). Not only do the gastrule multiply, in some cases, by budding or by fission, but the same processes are known to occur in the Scyphostoma stage; while numerous variations are recorded in the character of the strobilation, in the multiplication of the tentacles by incomplete fission or budding, and even in the number of the highly characteristic teniolz and in that of the Ephyra-lobes. The property of giving rise to fresh individuals, whether by budding or by fission, has, in this case, become a normal feature of the species; and the process takes place even at very early periods of the development, just as is the case in Crisia. The striking variability in the number of the radii of the immature stages in Aurelia may possibly be con- nected with something in the constitution of the egg which predisposes it to develop in an unusual way. ~ For since the cells which are destined to give rise to a single individual are not normally separated off until a late stage, which varies in different individuals, the existence of a ten- dency to vary in the number of individuals produced from an egg might also, in all probability, make itself felt in variations in a different direction. If the gastrula contains in itself the power to develop into several individuals, it is hardly surpris- ing that it should in some cases develop an abnormal number of radii. Similar cases of larval budding have been recorded in other Scyphomeduse. Thus Goette (12), confirming an older ob- servation of Sars (1841), shows that the formation of a stolon may take place (presumably in Cotylorhiza tuberculata) in the larva which has just fixed, but which is still without tentacles. Ciliated buds are also given off from the Scypho- stoma of Cotylorhiza, the buds fixing’ and developing a mouth after fixation. 30 SIDNEY F. HARMER. A method of reproduction similar to the last is recorded by Bigelow (4) in Cassiopea xamachana. In Oceania armata, Metschnikoff (24) characterises the process of segmentation as a regular “ Blastomerenanarchie ” (p. 38). The first two blastomeres almost separate from one another ; while, in some cases, when the very slight connection which normally exists between them becomes ruptured, the separated blastomeres atrophy. Oceania further distin- guishes itself, at the eight-cell stage also, from other Meduse investigated by Metschnikoff; the blastomeres, instead of being arranged in an orderly manner, lying together “ ganz unregelmassig.” This extraordinary irregularity (see Metsch- nikoff’s pl. i, figs. 83—385) is equally remarkable at later stages, and ultimately gives rise to irregularly shaped masses of cells; the embryos often assuming a quite “ abenteuerliche Gestalt,” due to the fact that they multiply by divi- sion. Those embryos which do not divide form much larger larve than the others. As a converse to this may be mentioned some most interest- ing results arrived at by Driesch (9) and by Fiedler (10). Driesch showed that by violent shaking of the water contain- ing Echinus-eggs which had divided into two blastomeres, or in other ways, the two cells could be isolated from one another. Each segmented in the same way that it would have followed if it had remained connected with its fellow, i.e it developed into a half-embryo, right or left as the case might be. The segmentation cavity, at first widely open, closed up in course of time so as to form a blastophere, consisting (as appeared from measuring the cells) of half the normal number of cells, and being halfthe normal size. Three of these embryos developed into complete Plutei, which differed from normal ones only in size. In cases where the two original blastomeres had been only partially separated, seventeen cases were recorded in which the embryo distinctly consisted, at the end of the first day, of two halves. In several cases each of these embryos divided into two complete embryos, some of which were shown to develop into small normal Plutei. In EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 31 another instance an injured two-cell embryo developed apparently normally up to the end of the blastophere stage, but finally divided into two; and in another case the injury to the two-cell stage resulted in the formation of a double monster. In Crystallodes (Siphonophora) the remarkable ameeboid character of the superficial blastomeres suggested to Haeckel (13) that the embryo, at the end of the second day, could be compared to a colony of Amcebe, in consequence of the great individuality of the separate blastomeres ; and that, if this comparison were correct, an isolated portion of the embryo might be expected to have the power of further development. The experiments made to test this hypothesis were completely satisfactory. Embryos artificially divided at this stage de- veloped into normal individuals of a smaller size than usual. The cut surface became concave, the edges ultimately joining, so that the embryo again became spherical, and then proceeded to develop in its normal manner. It is hardly possible to overlook the fact that, in some at least of the above quoted cases, embryonic fission is specially connected with deviation from the normal type of segmenta- tion of the egg. This is most clearly seen in the case of Oceania, where asuperficial glance at Metschnikoff’s figures is sufficient to convince one of the extraordinarily abnormal cha- racter of the segmentation. The same fact is, however, to some extent true of Lumbricus trapezoides and of Crys- tallodes, where Kleinenberg and Haeckel respectively call attention to remarkable features in the segmentation. The segmentation of the egg of Crisia obviously belongs to an unusual type, and, as has already been pointed out, it finds its closest parallel in Salpa, an animal which is remarkable for the great extent to which asexual reproduction is carried out. Doliolum, whose life-history agrees with that of Salpa in including two remarkably different generations, offers a further analogy to Crisia in the character of its asexual reproduction. The stolon of the asexual generation segments off, according to 382 SIDNEY F. HARMER. the description of Uljanin (35), a series of buds in which there is a very small amount of differentiation. These ‘“ Ur- knospen ”’ consist of a layer of ectoderm surrounding a mass of embryonic cells which are but slightly differentiated (1. c., pl. x, fig. 3). These buds divide up into numerous similarly- constituted buds, so that the sexual individual of Doliolum takes its origin from a group of cells which is very similar to the young “ secondary embryos”’ of Crisia. The same method of reproduction characterises the remark- able Dolchinia, recently described by Korotneff (22). This animal is closely allied to Doliolum, if, indeed, it should not be placed in that genus. The only phase in its life-history which is so far known is a gelatinous axis, bearing very nume- rous Doliolum-like zooids, and which probably corresponds to the dorsal process of Doliolum. The axis bears numerous buds, wandering about on its surface by means of pseudo- podia. The buds have probably been derived from the seg- mentation of the ventral stolon of an asexual form. They increase in number by division. Should one of the daughter- buds fix itself on the base of a young zooid, it becomes a bean-shaped body, which gives rise to a large number (as many as forty) of new buds. The young buds, at the stage at which they become free, consist of a solid mass of cells in which a very small amount of differentiation has taken place. The formation of the secondary buds, as shown -in Korot- neff’s pl. xii, figs. 14, 15, has thus a striking resemblance to the mode of development of the secondary embryos in Crisia; neglecting the not unimportant difference that in the former case the budding organ is itself a bud, and in the latter case an embryo. A similar process probably takes place in Anchinia (21) ; and Uljanin (Il. ¢c., pp. 106—117) brings forward evidence to show that the same is true of some of the compound Asci- dians. The larva of Distaplia magnilarva, for instance, gives rise to structures comparable with the “‘ Urknospen” of Doliolum. Uljanin comes to the conclusion that the bud- ding of adult Tunicates is derivable from a division of “ very EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 33 young developmental stages.” A similar suggestion with regard to the origin of alternation of generations among the Hydromedusz has also been elaborated by Brooks (6), who supposes that the hydroid stage has been evolved by the ac- quirement of the power of budding by the fixed larval stage. A slight modification of the “ primary embryo” of Crisia would suffice to make it necessary to consider the life-history of that animal as a case of alternation of generations. But since, as I believe, the budding structure consists of a mass of embryonic cells, which ultimately becomes completely con- verted into “secondary embryos,” leaving nothing behind, I have preferred not to describe it as a separate generation. Enough has been said to show that in the Tunicata at least, and to a less extent in the Ceelenterata,! there are remarkable cases of the formation of buds from slightly differentiated masses of cells. These two groups, with the Polyzoa, are certainly the groups of animals in which budding in the adult condition is a more normal event than in other groups of animals. It may thus be asserted that in the Polyzoa, the Tunicata, and the Ceelenterata the asexual reproduction of certain forms takes place at a stage before the individual which is reproduc- ing asexually has had time to undergo more than the earliest steps in its development. A similar precocious formation of fresh individuals is well known in the reproduction of Trematoda.” The investigations of Driesch (9) and of Haeckel (13) have shown that blastomeres which have been artificially separated from the embryo are able, in some cases, to give rise to com- plete larve. The question suggests itself: Has the gemmi- 1 The case of Cunina, as described by Uljanin, Schulze, Metschnikoff, and Brooks, and more recently by O. Maas (‘Zoolog. Jahrbiicher,’ “ Abth. f. Anat. u. Ontog.,” Bd. v, Heft 2, 1892), is another remarkable instance of the same kind. ; 2 Compare particularly the remarkable account given by Heckert (18) of the life-history of Distoma macrostomum, and particularly the state- ments referring to its remarkable branched sporocyst, known as Leuco- chloridium paradoxum. VOL. VI. 3 34 SIDNEY F. HARMER. parous method of reproduction in the adults of the above- mentioned groups been preceded by larval fission, possibly induced by the separation from the embryo of individualised blastomeres or groups of blastomeres, or is the embryonic fission the result of the precocious acquirement of the budding habit which characterises the adult? Now in the Polyzoa, embryonic fission is by no means a common phenomenon, although the adults of all known Polyzoa possess the power of budding ; and although it is probable that the method of re- production above described in Crisia will be found to be characteristic of all Cyclostomes. I have no sufficient evidence on this point at present, but it may be pointed out that the ovicells of Cyclostomatous Polyzoa invariably (so far as I know) contain a large number of embryos. My own observations enable me to state further that the general structure of the ovicell in Idmonea serpens and in Diastopora patina agrees with that in Crisia; and I have little doubt that I shall be able to show that embryonic fission is characteristic of Cyclostomes in general. The development of the Phylactolemata possibly offers some analogies to this process. The structure of the larva is somewhat similar to that of Cyclostomes, and the early development, according to the account given by Jullien (19), is not unlike that of Crisia. Braem (5) has also given an incomplete account of the development of Pluma- tella, which suggests further resemblances to the Cyclosto- mata. The two layers which form the wall of the embryo, and which are considered by Braem to represent ectoderm and body-cavity epithelium respectively,! are obviously comparable with the two layers shown in Pl. III, figs. 22 and 23, of Crisia. The manner in which (in Plumatella) a rudi- mentary bud encloses the egg, forming the “‘ oecium,” is again strikingly suggestive of Crisia.2 The first stage in which the 1 The same conclusion is arrived at by Davenport (8a), whose valuable paper should be consulted for a comparison of the larva of Phylactolemata with that of Gymnolemata. ? Compare in particular the woodcut given by Braem in his explanation to fig. 171. - EMBRYONIC FISSION IN OYCLOSTOMATOUS POLYZOA. 35 egg is completely enclosed is shown in Braem’s pl. xv, fig. 171, while in fig. 172 the “ owcium ” has differentiated off a distal portion which may be the exact equivalent of the “ tentacle- sheath” shown in my own fig. 3 for Crisia. The tendency to precocious fission shows itself in Phylactolemata, how- ever, in the precocious formation of a considerable number of polypides, particularly in Cristatella;—a process which is of course very- different from the embryonic fission of Crisia. The ovicells of the Cheilostomata are probably not homo- logous with those of Cyclostomata. They are probably not to be regarded as modified zocecia, since the ovicell is an appen- dage of a fertile zocecium, and ordinarily contains a single embryo. Similarly in the other groups which have been mentioned precocious fission is not characteristic of the whole group, but occurs sporadically ;—in Coelenterata, in Oceania, Cunina, &c.; in Trematoda, in the Distomez ; and in Tunicata in the Thaliacea, and in some Synascidians. Although I must regard the question as a very open one, the conclusion which appears to me to be suggested by the above facts is that one is not justified in assuming that the budding of the Polyzoa, for instance, commenced with the acquirement of a habit of embryonic fission like that found in Crisia, but that the embryonic fission may be the consequence of the pre- viously acquired power of adult budding. It may be pointed out that the embryonic fission of Crisia gives rise to numerous larve, each of which may form the starting-point of a new colony. In the case of adult Polyzoa, the result of budding is merely to increase the number of individuals in a colony, with the exception of Jioxosoma (in which the bud normally becomes free) and of certain dendritic forms of colony, in which the decay of the proximal part of the colony leads to the separation as distinct colonies of what were at first merely branches, or of cases like that of Crisia itself, where new colonies are formed by the upgrowth of new stems from a creeping rootlet, which acts as a stolon for the production of 36 SIDNEY F. HARMBER. new colonies.! It should be further noted that the production of new polypides in old zoccia is one of the most characteristic ways in which the property of budding manifests itself in Ectoprocta, and that this process is most easily interpreted as a process of regeneration of lost parts. The provisional conclusion may therefore be stated as follows :—That the process of embryonic fission, which may appear abnormally in certain individuals in so many groups of animals which do not multiply by fission, has in Crisia become a normal phenomenon of the development ; and that this pro- cess is correlated with the tendency which is so strongly marked in the Polyzoa to produce buds in the adult condition. Giard (11) has recently published a note on what he terms ** necilogonie,”’ i. e. the phenomenon exhibited by certain ani- mals of developing in a more or less ‘‘ condensed” manner, in correlation with the amount of nutritive reserves in the egg, or with the conditions under which the parent is living. As examples of this process are mentioned, inter alia, the follow- ing cases :—In Leptoclinum lacazii, Gd., the same colony may produce two sorts of eggs; of these, one is poor in yolk, and gives rise to small larve, whose tail is absorbed early, and which do not begin to bud even on the third day. The other kind is rich in yolk, and produces larve which are still free- swimming on the fourth day, and which then already contain a colony of three individuals. Ophiothrix fragilis, Mill., lays eggs which develop, according to the conditions, either into perfect or into imperfect Plutei, or into embryos incapable of swimming, and which develop directly. The remarkable variations in the development of Aurelia aurita and of Pale monetes varians are also included in this category ; in the latter form the size and number of the eggs, as well as the rapidity of the metamorphoses, varying according as the animal lives in the brackish waters of the North or in the fresh-water lakes of the South. Giard’s observations suggest that the acquirement of em- 1 The statoblasts of the Phylactolemata are indeed a further exception, since each of these bodies is able to give rise to a new colony. EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 37 bryonic fission in Cyclostomes may have been connected with the presence of the nutritive conditions which are suited to induce the precocious formation of buds. Nothing can be more striking than the obvious continuity of protoplasm between the several units of the colony in a decalcified branch of Crisia. In the individuals which are modified as ovicells the protoplasmic network is particularly well developed. The embryo is thus surrounded by a rich nutritive material ; and just as the presence of a nutritive placenta in a Placental Mammal has resulted in the diminution of the size of the ovum, and in various abnormalities in its early segmentation, so in Crisia the size of the egg is reduced to a minimum, the whole of the nutritive substance being retained in the parental tissues and handed on to the egg or embryos as required, while the segmentation is entirely abnormal. Further, while the Mammalian embryo becomes easily comparable with that of any other Vertebrate embryo after a certain number of the early stages have been passed through, so the Crisia larva becomes, to some extent at least, comparable with the free larva of any other Polyzoon, although with this difference from other Polyzoa, viz. that the primary embryo has given rise to numerous lary, a process comparable with the artificial pro- duction of a complete embryo from a single blastomere of the two cell stage in the experiments of Driesch (9) and of Fied- ler (10). Attention has already been called to the similarity between the early stages of the development in Crisia and those in Salpa. The latter is another example of the modification of the first processes in the development, associated with the presence of special maternal nutritive arrangements. The embryo of Salpa develops, as is well known, in close connec- tion with akind of placenta ; and its early stages are, compared with those of most other animals, highly abnormal. The formation of buds from the individual developed from the egg does not take place at once, as in Crisia, but is deferred until the animal is mature, when buds are produced in very large numbers from the stolon. 388 SIDNEY F. HARMER. ‘Similarly the egg of Pyrosoma, like that of Salpa, makes its appearance in the same precocious manner as that of Crisia, being formed very early from the so-called “ genital string” (Sa- lensky, 29). The early development, which is modified by the presence of yolk, takes place in the interior of the old colony, and is very abnormal, the blastomeres being for a time com- pletely separated from one another (Salensky, p. 443). The result of the development is the formation of the well-known “ Cyathozooid,” with its colony of four ‘ Ascidiozooids,” the formation of which is compared by Salensky (30, p. 92) with the embryonic fission of Lumbricus trapezoides. The for- mation of a stolon (represented by the chain of four Ascidio- zooids) in the Pyrosoma-embryo is further regarded as the precocious acquirement by the embryo of the power of bud- ding already possessed by the Synascidians. Peripatus! is well known to be viviparous, and the extra- ordinary character of the segmentation of its ovum may have some relation to the presence of external sources of nutriment. The cases already quoted may be taken as showing that some of the abnormalities in the development of Crisia may be due to the nutritive conditions in which the development takes place. Just as the presence of food-yolk within the egg modifies the character of the segmentation and of the forma- tion of the layers, so the presence of copious stores of nutrient material in the maternal tissues outside the egg may also affect the early developmental processes. Thus the large number of relatively large larvee which develop from the minute egg of a Crisia could not be produced if the egg were not supplied with nutriment from outside itself. While some ofthe irregu- larity in the segmentation of the egg may be due to this cause, the extreme independence of the blastomeres at an early stage may be connected with the acquirement by the embryo of a habit of forming buds in the embryonic condition. 1 See Sedgwick, No. 81. EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA, 39 REFERENCES. 1. Barrots, J.—‘ Recherches sur l’Embryol. des Bryozoaires,’ Lille, 1877. 2. Barxots, J.—‘‘ Embryogenie des Bryozoaires, Essaie d’une théorie générale du dével.,” &c., ‘Journ. de l’Anat. et de la Physiol.,’ Xviil, 1882, 140. . Barrois, J.—* Mém. sur la Métamorphose de quelques Bryozoaires,” ‘Ann. Sci. Nat.,” ‘Zool.,’ 7¢ sér., i, 1886, Art. No. 1. 4. Bicrtow, R. P.—‘ On Reproduction by Budding in the Discomeduse,” ‘Johns Hopkins Univ. Circulars,’ xi, No. 97, 1892, 71. 5. Brarm, F.—“‘ Untersuchungen iib. d. Bryozoen d. siissen Wassers,” Leuckart and Chun’s ‘ Bibliotheca Zoologica,’ Heft 6, 1890. 6. Brooks, W. K.—“ The Life-History of the Hydromeduse,”’ ‘Mem. Boston Soc. Nat. Hist.,’ iii, 1886. 7. Buscu, W.—‘ Beob. iiber Anat. u. Entwick. einiger wirbellosen See- thiere,’ Berlin, 1851. 8. Busk, G.—‘ Monogr. of the Fossil Polyzoa of the Crag,’ Palzont. Soc., 1859. 8a. Davenvort, C. B.—‘‘ Obs. on Budding in Paludicella and some other Bryozoa,” ‘ Bull. Mus. Comp. Zool. Harvard Coll.,’ xxii, No. 1, 1891. Driescu, H.— Entwicklungsmechanische Studien,” ‘ Zeits. f. wiss. Zool.,’ liii, 1892, 160. 10. Fispier, K.—* Entwicklusgsmechanische Stud. an Hchinodermeneiern,”’ ‘Festschrift zur Feier d. fiinfzigjahrigen Doctor-Jubilaums des . . Dr. K. W. von Nageli, &c.,’ Zurich, 1891. 11. Garp, A.—“ Sur le bourgeonnement des larves d’Astellinm spongi- forme Gd. et sur la Peecilogonie chez les Ascidies composées,” ‘Comptes Rendus,’ Fév., 1891. 12. Goutrz, A.—‘Hntwickelungsgesch. d. Aurelia aurita u. Cotylorhiza tuberculata,’ Hamburg and Leipzig, 1887. 13. Harcxer, E.—“Zur Entw. d. Siphonophoren,’ ‘Naturk. Verb. Utrechtsch Genootschap,’ 1869. 14, Harcxet, E.—‘ Metagenesis und Hypogenesis von Aurelia aurita,’ Jena, 1881. 15. Harmer, S. F.—“Sur l’embryogénie des Bryozoaires Ectoproctes,” ‘Arch. Zool. Exp. et Gén.,’ 2 sér., v, 1887, 443. 16. Harmer, S. F.—‘‘On the Origin of the Embryos in the Ovicells of Cyclostomatous Polyzoa,” ‘ Proc. Cambridge Phil. Soc.,’ vii, part 2, 1890, 48. These ‘Studies,’ vol. v. iv) © 29. 30. 31. 32. 33. 34, 35. 36. SIDNEY F. HARMER. . Harmer, 8. F.—“On the British Species of Crisia,” ‘Quart. Journ. Mier. Sci.,’ xxxii, 1891, 134, &c. These ‘Studies,’ vol. v. . Hecxert, G. A.—“ Unt. ib. d. Entwicklungs- und Lebensgeschichte d. Distomum macrostomum,” Leuckart and Chun’s ‘ Bibliotheca Zoolo- gica,’ Heft 4, Cassel, 1889. . JULLIEN, J.—‘ Observations sur la Cristatella mucedo,” ‘Mém. Soe. Zool. France,’ iii, 1820, 361. . Kiernenperc, N.—“ The Development of the Harthworm, Lumbricus trapezoides,” ‘Quart. Journ. Mier. Sci..’ xix, 1879, 206. . Korornerr, A.—“ Die Knospung d. Anchinien,” ‘Zeits. f. wiss. Zool.,’ xl, 1884, 50. . Korornerr, A.—‘‘ La Dolchinia mirabilis (nouveau Tunicier),” ‘ Mitt. Zool. Stat. Neapel.,’ x, Heft 2, 1891, 187. . Metscunixorr, E.—“ Vergleichend-embryologische Studien,” ‘ Zeits. f. wiss. Zool.,’ xxxvii, 1882, 286. . Merscunixorr, H.—“ Embryolog. Stud. au Medusen,” Wien, 1886. . Ostroumorr, O.—*< Zur Entwickelungsgeschichte der cyclostomen See- bryozoen,” ‘ Mitt. a. d. Zool. Stat. zu Neapel, vii, 1887, 177. . Provo, H.— Recherches sur la Larve de la Flustrella hispida,”’ * Arch. Zool. Exp. and Gén.,’ 2 sér., viii, 1890, 409. . Rurrer, M. A—‘ Immunity against Microbes,” ‘Quart. Journ. Mier. Sci.,’ xxx, 1891, 99 and 417. . Satensky, W.—“ Neue Untersuch. ib. d. embryonale Entwicklung der Salpen,” ‘Mitt. Zool. Station zu Neapel,’ iv, 1883, 90 and 327. SaLensky, W.—“ Beit. zur Embryonalentwicklung der Pyrosomen,” ‘Zoolog. Jahrbiicher,’ ‘ Abth. f. Anat. u. Ontog.,’ iv, 1891, 424. SaLensky, W.—lIbid., continued, ‘ Zoolog. Jahrb.,’ ‘ Abth. f. Anat. u. Ontog.,’ v, Heft 1, 1891, 1. Sepewick, A.—“‘The Development of Peripatus capensis,” Part I, ‘Quart. Journ. Mier. Sci.,’ xxv, 1885, 449. SEELIGER, O.—“ Die ungeschlechtl. Vermehrung d. Endoprokten Bryo- zoen,” ‘ Zeits. f. wiss. Zool.,’ xlix, 1889, 168. SEELIGER, O.—“ Bemerkungen zur Knospenentw. d. Bryozoen,” ibid., 1, 1890, 562. Smirt, F. A.—‘‘ Om Hafs-Bryozoernas utveckl. och fettkroppar,” ‘Ofvers. af K. Vet.-Akad. Foérhandl.,’ 1865, No. 1. Ussanin, B.—“Die Arten d. Gattung Doliolum, &c.,” ‘Fauna and Flora G. v. Neapel,’ x. Monog., 1884. Waters, A. W.—Closure of the Cyclostomatous Bryozoa,” ‘Linn. Soc. Journ.,’ “ Zool.,” xvii, 1884, 400. EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 41 It hardly falls within the province of this paper to discuss the details of the normal budding in Polyzoa. Both Braem (5) and Davenport (8a) have shown that polypide-buds in general are derived from a mass of “‘embryonic”’ tissue, handed down from the beginning of the formation of the colony, some part of this tissue being left over for the production of fresh buds on each occasion when a polypide-bud is formed. Braem’s account of the formation of the statoblasts in Phylac- tolemata more nearly resembles the development of the “secondary embryos” in Crisia than any other process as yet described in Polyzoa. The funiculus is, indeed, not an em- bryo; but the young statoblasts are formed from it in much the same way as that in which the larve are developed from the “ primary embryo” in Crisia. The funiculus consists of a core of ectoderm surrounded by a sheath of mesoderm (both kinds of cells having an ‘‘embryonic” character). The stato- blasts are formed by a process which is to all intents and purposes a transverse segmentation of the funiculus. EXPLANATION OF PLATES I, II, & III, Illustrating Mr. Sidney F. Harmer’s paper “On the Occur- rence of Embryonic Fission in Cyclostomatous Polyzoa.” PLATE I. Fic. 1.—Crisia eburnea, Median longitudinal section through a young ovicell, showing the egg, which is already partially surrounded by the polypide- bud (Zeiss, DD). Fic. 2.—C. ramosa. Part of a similar section at a more advanced age, showing the complete inclusion of the ovum (Zeiss, DD). Fic. 3.—C. ramosa. Similar section at an older stage. The polypide-bud has become the “follicle.” The tentacle-sheath and the aperture are well developed... Ovicell at the “funnel-stage ” (Zeiss, DD). VOL. VI. 4, 42 SIDNEY F. HARMER. Fie, 4.—C. eburnea. Spermatozoa. To the right, three mature sperma- tozoa, drawn in the living condition (Zeiss, F) ; to the left, three stages in the development of the spermatozoa, treated with osmic acid and picro-carmine (Zeiss, F); in the middle, four immature groups consisting of four spermatozoa each, killed with osmic vapour (Zeiss, 4; immersion, 4 oc.). Fies. 5—7.—C. ramosa (Zeiss, DD). Fig. 5. (Combined from several sections of the same ovicell.) The egg has divided into three blastomeres; the valve is developing, and the distal end of the tentacle-sheath has become thickened. Fig. 6. Proximal end of a longitudinal section (more advanced). The embryo consists of a compact rounded mass lying in a large follicle, which projects freely into the tentacle-sheath. Fig. 7. A similar preparation at a stage when the vacuolation of the ‘follicle is commencing. PLATE II. Fic. 8.—Crisia eburnea. Ovicell at a stage corresponding to Fig. 6 inC. ramosa. The embryo is larger and the follicle is much smaller than in that species (Zeiss, DD). Figs. 9—14.—C. ramosa (Zeiss, D D). Fig. 9. The vacuolation of the follicle is nearly complete. The tubular aperture is formed, and the structures connected with its base are well developed. Fig. 10. Invagination (?) in a “primary embryo,’ stage as Fig. 9. Fig. 11. Considerably later stage. The follicle has become a dense protoplasmic reticulum, containing the massive ‘‘ primary embryo,” now transformed into a budding organ, which is giving rise to numerous secondary embryos, three of which are seen lying freely in the reticu- lum. At the upper end a giant-cell, derived from the thickened distal end of the tentacle-sheath (cf. Figs. 5, 13 and 9). Figs. 12—14. Illustrating the development of the aperture of the ovi- cell. In Fig. 12, the primary aperture still remains open, the distal end of the tentacle-sheath being thickened. In Fig. 13, the tubular aperture of the adult ovicell is developing; it contains the remains of the primary aperture. The thickening of the distal end of the tentacle- sheath has increased in size, and the valve is well developed. In Fig. 14, the tubular aperture is almost complete. The thickening of the tentacle-sheath still extends into its base. At its distal end a de- pression occurs, which is possibly the remains of the primary aperture. The tube ends in a cap prolonged into a narrow tube, of unknown significance. > at about the same EMBRYONIC FISSION IN CYCLOSTOMATOUS POLYZOA. 43 PLATE III. Fie. 15.—Crisia eburnea. A young internode decalcified, with a develop- ing ovicell. For explanation of the letters (a, B, and c) see p. 5 (Zeiss, A). Fies. 16—23.—C. ramosa. Fig. 16. The aperture of the same ovicell from which Fig. 6 was taken. The origin of the giant-cells from the thickened distal end of the tentacle-sheath is shown (Zeiss, F), Fig. 17. Longitudinal section of an ovicell which is filled with secondary embryos. To the left, the primary embryo (Zeiss, A). Fig. 18. Protoplasmic reticulum with giant-cells, from an ovicell at about the stage of Fig. 11 (Zeiss, F). Figs. 19—21. Development of ovicell (Zeiss, A). The ovicell has in each case been numbered 4, in order to admit of ready comparison between the three stages. Fig. 22. Young secondary embryo, in longitudinal section (Zeiss, F). Fig. 23. Older embryo, in longitudinal section (Zeiss, F). — 1 a jl. cepa) syoeaiton Lota eR T “4 Ce ee ee er a0) ht i elle et co AN) a en See “eS Se ne aoe ane 6 , ous ; io & 0 - 5 lah imped 4 ar Sul ue hates Goth Ob belie fs Dai ie T 1 ie ere he ~ -eo aig eee av mw bothed Se > bale gap Pe We : By Play . ri.” vy iat ; ; be . = : hte ace petann gi ae RRR RO Pie he . ; é i eS a ye Ox : 2 ; y = 4 H 7 c : —_ = - = yo. = i ° z > r ip = =) —— oat oe ‘, : 1 , yr i | * = A, ot Mas are se "=a oe ~ —~ < ~! mn . a a Dt, Geeta Sy Mme, 4 : ‘ se ¥ Mi ys > "4 a ical Fe eg eso i ~ _ * * 7 ‘ ; : é A+. > | 2 2 : - . ; , Fig. 3. calerfied ectocyst, .... S.F. Harmer del. Studies M.L. Vol. VI, P11. F. Huth, Lith? Edin? tubular aperture.“ gan tentacle _.-- cablorfied ectocyst S.F.Harmer del Studies M.L.Vol.VI, PL IL. F.Huth, Lith? Edin® — Leg. £7: JOLY embryos ca hg j prunary embryo pera \ ; ba \\ar € S.F. Harmer del. Studies M.L. Vol. VI, Pl. Ill. fig. Al. Fig. 20 ps Y 6 {A x ) ae ) { 7s ( Y 6 | ys « Lem ie ae \ 2 \. Uf 3 Bae: Za Fig fa. “4 go eeee 229% @o %@ eo 2°98 000 208 & F.Huth, Lith’ Edin? ae ‘by we ler re 7 ae - { DEVELOPMENT OF DISTICHOPORA VIOLAOCEA. 45 The Early Stages in the Development of Disti- chopora violacea, with a Short Essay on the Fragmentation of the Nucleus. By Sydney J. Hickson, M.A. Cantab. et Oxon., D.Sc.Lond., University Lecturer on the Morphology of Invertebrates ; Fellow of Downing College, Cambridge. With Plate IV. Tue material upon which I have made my investigations was in part collected by me in N. Celebes, and in part by Professor A. C. Haddon in Torres Straits. Some of the specimens were treated with strong alcohol alone, others with corrosive sub- limate followed by alcohol. For decalcification I have entirely used nitric acid. I have tried a great many different stains and combinations of stains. Borax carmine, Biondi’s fluid, methyl green, and hematoxylin all give fairly good results; but I find that the best treatment is to place the sections, when fastened to the slide, in a strong solution of eosin in 90 per cent. spirit for an hour, then to wash in 90 per cent. spirit and stain in weak hematoxylin for twenty minutes. This treatment gives a beautiful double stain which shows the nuclei and the chromatin granules ‘better than I have seen them in any preparations treated with carmine. My researches were entirely carried on in the morphological laboratory at Cambridge. VOL. VI. 5 46 SYDNEY J. HICKSON. I. The Early Stages in the Development of Distichopora. The ovum of Distichopora, like that of Allopora and other Stylasterids, is provided with a large amount of yolk, and lies in a cup-shaped trophodise. In young immature ova the germinal vesicle is situated in the middle of the egg, is spherical in shape, is provided with a well-defined membrana limitans, a germinal spot, and a fine net- work of protoplasmic fibrils with thickened nodes (PI. IV, fig. 1). When examined with a high power the germinal spot may be seen to contain a few clear vacuoles (fig. 2). In some ova with a full complement of yolk-spheres the germinal vesicle is irregular in shape, and provided with processes resembling the pseudopodia of ameba. The outlines of these processes are usually difficult to observe, the mem- brana limitans being apparently wanting, and the intra-nuclear and extra-nuclear protoplasm perfectly continuous (fig. 3). In these cases there may be seen a few large rod-shaped granules (the chromosomes), which stain deeply with carmine and other stains. These amceboid germinal vesicles are without doubt passing from the centre of the ovum towards the periphery. In those that are near the periphery the chromosomes are more numerous and very much smaller than they are in those nearer the centre of the ovum. In one case I have observed these bodies arranged in a row parallel to the surface of the ovum, and dividing the nucleus into two unequal halves (fig. 5). When and in what manner the polar bodies are formed I cannot say, but it is probable that in some cases the nuclei of the polar bodies are formed before the germinal vesicle reaches the periphery, and are absorbed in the substance of the ovum. The germinal vesicle finally reaches the periphery of the ovum, and when it is in that position the fertilisation most probably occurs. It is clear that the germinal vesicle must remain at the DEVELOPMENT OF DISTICHOPORA VIOLACHA. 47 periphery for a very considerable time, for of the numerous unfertilised ova that I have examined a large majority have their germinal vesicles in that position. In the next stage the membrana limitans of the inner half of the vesicle disappears, the network and the germinal spot break down into numerous very minute scattered granules (fig. 7). Then the membrana limitans entirely disappears, and lastly, the substance of the vesicle, or, as it should now be called, the oosperm nucleus, becomes scattered through the substance of the ovum. Fig. 8 is a careful drawing of a stage in which the mem- brana limitans has just disappeared, and I have three or four complete series of sections through ova in which no trace of nuclear structure can be found nor any area, such as that shown in this figure, which represents the vanished nucleus. As these two stages are of the greatest importance in the consideration of what follows, it is necessary to say that not- withstanding very careful search with high powers, no trace of karyokinetic figures could be observed. The ova of these stages are not sufficiently numerous, nor are the methods of preservation sufficiently perfect to enable one to assert that such figures do not occur. Corrosive sub- limate followed by alcohol, although giving excellent general histological results, does not always bring out the full details of nuclear division; and it will be necessary to confirm these purely negative results as regards karyokinesis by observa- tions made upon specimens preserved in Flemming’s solution and other reagents before any general statements regarding fragmentation of the oosperm nucleus of Distichopora can be accepted. Nevertheless it is my belief that we have here an instance of nuclear fragmentation, for reasons which I propose to dis- cuss in the third section of this paper. In the next stage that I have observed, a few small islands of protoplasm may be seen in the yolk (fig. 9), and the exa- mination of broken sections, in which part of the yolk has 48 SYDNEY J. HICKSON. been washed away, shows that these islands are connected together by a very coarse mesh-work of fine protoplasmic strands. In a later stage the islands are seen to be more numerous, and the protoplasmic mesh-work somewhat finer. A complete nucleus may be seen in some of these islands, but in others all that can be made out are a few deeply staining granules (figs. 10 and 12). In a later stage the nuclei have increased in number in the midst of the yolk, and a few make their appearance in the protoplasmic sheath that surrounds the ovum. In these last three stages I have described a process which can only be compared with the so-called free nuclear forma- tion in early insect embryos. Nuclei make their appearance in places which were previously apparently devoid of any nucleus or nuclear structure. Moreover nuclei of various sizes and shapes may be seen in the embryo at the same time. It is not reasonable, however, to assume on the insufficient evidence before us that “free nuclear formation” does actually occur. It seems to me to be much more probable that minute fragments of nuclear substance scattered through the proto- plasmic mesh-work collect together in places, and form by their fusion true recognisable nuclei. In other words, the pro- cess we have under observation is rather one of ‘‘nuclear re- generation’’ than one of “ free nuclear formation.” I have often noticed in ova of these stages an aggregation of the yolk into spherical, polygonal, or irregular lumps, sug- gesting that the egg has undergone some form of complete segmentation (fig. 13). This is not a true process of seg- mentation, however, since the distribution of the nuclei in the spaces between the aggregations and not in their centres shows that it affects the yolk only. It is remarkably similar in appearance to the so-called yolk segmentation of Arthropods, the appearance of the embryo at this stage being very much like that of such a form as Peripatus nove-zealandiz, as described by Miss Sheldon (53). This segmentation of the yolk seems to be only temporary, for in embryos in which the DEVELOPMENT OF DISTICHOPORA VIOLACHA. 49 ectoderm has commenced to be differentiated it cannot be observed. In later stages of the development the ectoderm is gradually formed. Nuclei appear in the peripheral sheath of protoplasm, and the protoplasm accumulates in the form of cellular blocks around each nucleus, as in Allopora. I have carefully examined the endoderm in these stages in the hope of finding out the manner in which the nuclei divide, and although I have found a few dumb-bell-shaped forms, and no satisfactory evidence of karyokinesis, I do not feel justified in asserting that the nuclei always divide amitotically. As far as the ectoderm is concerned, I can assert most positively that indirect nuclear division does occur. Numerous dumb-bell-shaped nuclei and nuclei connected together in pairs may be seen in the developing ectoderm, and in these faint achromatic lines may be seen connecting the chromatin rodlets. The nuclei are too small to enable me to make out all the details of the process, but there can be no doubt that there is a true process of karyokinesis in the divisions of these nuclei (fig. 18, a, 6, c, d, and e). Ihave not been able to decipher anything like the “spheres of at- traction.” One very remarkable and important point in the develop- ment of all the Hydrocoralline, so far as they have at present been investigated, is the fact that thereis no segmentation of the ovum, either complete or partial, nor is there any for- mation of cells with a definite outline until a very late stage. At the time when (in Allopora and Distichopora) there are ten or fifteen nuclei, the young embryo is a simple multinu- cleated plasmodium, loaded with yolk. In the later stages the nuclei have increased in numbers, and a certain number of them are arranged in a row at the periphery of the embryo. The yolk in the immediate neighbourhood of these peripheral nuclei disappears, probably by absorption, and thus they are situated in a clear peripheral sheath or envelope of protoplasm. In a later stage this peripheral sheath of nuclei breaks up into blocks, each block containing one nucleus, and thus the ectoderm is formed. 50 SYDNEY J. HICKSON. The ectoderm is, then, a differentiation of the periphery of a multinucleated plasmodium. What becomes of the inner part of the plasmodium ? We have no answer to this question so far as the Hydro- corallines are concerned ; but, judging from the other Ceelente- rates, there can be little doubt, I think, that it becomes the endoderm. In the development of Aglaophenia (Tichomiroff, 55) we find a stage that is almost precisely similar to the solid planula of Distichopora and Allopora, and in a later stage this central yolk-laden plasmodium breaks up into blocks, which become the endoderm-cells of the adult. The difficulty that we have now to face is, how can these facts concerning the origin of the germ layers be brought into line with those of other Ceelenterates ? We find in the Stylasteridz no segmentation, no process of invagination to form the endoderm, and no process that can be compared with ordinary primary delamination ; but still it is probable that this method of the formation of the germ layers, if it is not itself the primitive one, has been derived from those of other Ceelenterates, and I shall endeavour to show in the next section how the transition has taken place. II. On the Formation of the Germinal Layers in the Celenterata. During the last ten years our knowledge of the early stages of the development of the Celenterata has very considerably widened, but still we seem to be no nearer to the solution of many interesting phylogenetic questions than we were before. The various theories that have been put forward, based upon the study of a few forms, have in no instance received the un- qualified approval of the principal authorities on the group, and we find ourselves in a maze of conflicting theories, none of which seem to conform entirely to our knowledge of facts. This unfortunate state of affairs is due to the fact that in the group of the Coelenterata we find many very different types of —— -—_— DEVELOPMENT OF DISTICHOPORA VIOLACHA. 51 development, and no one of them seems to be particularly predominant. The development of a gastrula by invagination probably occurs only in the group of Scyphomeduse. The formation of a planula by delamination (1. e. the primary delamination of Metschnikoff) occurs only in the group of the Geryonide. The formation of a sterrula by secondary delamination occurs in most of the Anthozoa (MecMurrich) and in many of the Hydroids. The formation of a sterrula by hypotropic invagination occurs in many Sertularide and Campanularide. The formation of a planula or sterrula by polypolar immi- gration of cells into a hollow blastula occurs in a few forms. Lastly, the formation of a multinucleated plasmodium without segmentation, which is followed by the differentiation of epiblast-cells at the periphery of a solid plasmodium (the endoderm), occurs in the Hydrocoralline and in some Al- cyonarians. Between these various types of development many inter- mediate forms have been found, so that we have as it were a complete series of developmental histories, with the typical in- vaginate gastrula at one end and the multinucleated plasmo- dium at the other. We may represent this series by the following plan : A, Gastrula formed by invagination.. Large segmentation cavity. Examples: Cotylorhiza (Claus, 8), Pelagia nocti- luca, and Nausithoé (Metschnikoff, 42). a. Intermediate forms between type A and B are found in Aurelia flavidula (Smith, 52), in which the clump of cells that are invaginated is at first solid, and in Cyanea capillata (McMurrich, 41), in which this clump of cells remains solid longer than in A. flavidula. B. A solid planula (sterrula) formed by hypotropous im- migration of cells into a large segmentation cavity. 52 SYDNEY J. HICKSON. Examples: Clytia, Tiara, Rathkea, Obelia, Tima, /Equorea (Metschnikoff, 42), and Cyanza arctica (McMurrich, 41). 6. Intermediate forms, in which the migration takes place mainly at the hind end, occur in Mitrocoma (Metschnikoff, 42). C. A sterrula is formed by polypolar immigration of cells into a large segmentation cavity, these cells being formed by the radial fission of the cells of the ccelo- blastula. Example: Aiginopsis (Metschnikoff, 42). c. Intermediate form, in which the cells that immigrate are formed partly by radial and partly by tangential division. Example: Hydra (Brauer, 5). D. A planula is formed by primary delamination, the endoderm-cells formed by tangential division only. The segmentation cavity is large. Example: Geryonia (Metschnikoff, 42). d. Numerous intermediate forms in which the segmen- tation cavity is small. Examples: Tubularia (Brauer, 5a), Bougainvillea (Gerd, 14). E. A sterrula is formed by precocious delamination (secondary delamination of Metschnikoff). No seg- mentation cavity formed. Examples: Aglaura (Metschnikoff, 42), Rhopalo- nema (Metschnikoff), Hudendrium and Sertularella (Tichomiroff, 55). e. Intermediate forms in which the segmentation is at first incomplete. Examples: Renilla (Wilson, 63), Gorgonia (von Koch, 36), and probably other Alcyonarians. F. A multinucleated plasmodium is formed. There is no segmentation and no segmentation cavity. Examples: Algaophenia (Tichomiroff, 55), Mille- DEVELOPMENT OF DISTICHOPORA VIOLACEA. 53 pora (Hickson, 17), and the Stylasteridz (Hickson, 18 and 19). It is not my purpose to discuss fully the various views that have been put forward concerning the origin of the Metazoa from the Protozoa. The gastrula theory, the planula theory, the plakula theory, and the phagocytella theory have each received in their turn the consideration of naturalists, and nothing would be gained were an attempt made in these pages to reopen the discussions that they gave rise to. But I cannot pass on without expressing my opinion that the developmental history of the Hydrocoralline lends some support to the so-called “ plasmodium ” theory. Many years ago, Jehring (27) and Saville Kent (32) put forward the view that the Metazoa are derived from a multinucleated Protozoan like Opalina. Sedgwick (57) has supported this view, as a result of his important work on the development of Peripatus, and considers that the ancestral Metazoan was probably of “the nature of a multinucleated Infusorian, with a mouth leading into a central vacuolated mass of protoplasm.” In discussing Saville Kent’s views Metschnikoff (42) says that there is no evidence of the formation of such a multi- nucleated cell in the lowest Metazoa. Now I have already pointed out that in the earliest stages of the Stylasteridz and of Millepora the embryo is nothing more nor less than a multinucleated cell; that is to say, it is a single undivided mass of protoplasm, containing numerous nuclei. It might be urged that it is a syncytium, a number of cells fused together; but there is no more evidence for such a view than for the view that it isa single multinucleated cell. Similarly it may be urged that Tubularia (Brauer, 5a), Aglaophenia (Tichomiroff, 55), Aleyonium (Kowalewsky, 37), Gorgonia (von Koch, 86), and Renilla (Wilson, 63) all pass through ¢, stage in their development in which the embryo is simply a multinucleated cell. The fact that such a condition as this occurs in many different groups of the animal kingdom widely separated from 54 SYDNEY J. HICKSON. one other also lends support to the view that it may have some important phylogenetic significance. Instances of the occurrence of an unsegmented multi- nucleated plasmodium are found not only in the Coelenterata above mentioned, but in Peripatus, Myriapods, Spiders (Kishi- nouye, 34, and Morin, 44), Insects, Crustacea, Elasmobranchs, and probably many other forms with large eggs. It might be urged as an argument against the plasmodium theory that the multinucleated plasmodium occurs principally in the development of those forms whose ova contain a large amount of food-yolk, that the segmentation is modified by the presence of this yolk, and that consequently the phylogeny is obscured. But it does seem to me that in the ovum that is perfectly clear and homogeneous we have a cell that is any nearer to the ancestral Protozoan than the ovum that contains a moderate amount of yolk. It is almost certain that the ifs Protozoan normally contained some food-vacuoles, and it is quite as probable as not that it had some contractile or simple water-vacuoles for floatation purposes as well. It is quite as reasonable to suppose that the Metazoa are derived from an Actinospherium-like ancestor with vacuoles in the outer regions as well as in the inner mass, as it is to derive the Metazoa from a “ multinucleated Infusorian with a mouth leading into a central vacuolated mass of protoplasm.” If this is the case, then we can no longer consider the yolk- bearing eggs to be secondarily modified, and the small transparent eggs to be the primitive types from which all the others are derived; but we may expect to find in the develop- ment of eggs with a moderate amount of yolk just as much or even more evidence of ancestral history as in eggs that are practically yolkless. It must not be forgotten, moreover, that the occurrence of a multinucleated plasmodium is not confined to those cases in which the ovum contains a large amount of yolk. In the ovum of Millepora there is no yolk, and yet the DEVELOPMENT OF DISTICHOPORA VIOLAOBA. 5) oosperm nucleus fragments without any segmentation occur- ring, giving rise to a simple multinucleated plasmodium. The eggs of Aphis (Will, 62) and some other insects contain very little yolk, and do not segment until a large number of nuclei are formed. The segmentation of the ovum, then, and the subsequent formation of a morula mass of cells, are phenomena not entirely dependent upon the absence of yolk. Many, compara- tively speaking, large eggs, such as that of Rana, segment, whilst others, such as that of Aleyonium, do not. We cannot, consequently, assert that when an ovum segments it is simply repeating an ancestral phase, and that when it does not segment it is prevented from doing so by the physical obstruction of the yolk. The reverse of this is more probably true. The recent brilliant researches of Driesch (9) prove that the segmentation of the ovum is due to physical or mechanical laws, and we cannot or should not derive any phylogenetic conclusion from the phenomena of segmentation. We may even go further than this, and say that the develop- ing ovum would not segment, but would naturally pass through the stage of a multinucleated plasmodium, were it not for the action of certain purely mechanical forces, with which we are not at present fully acquainted. When these forces cannot act upon the egg, or are in some way counteracted, the ovum does not segment, whether it is laden with yolk (Stylasteride, many Insects, Elasmobranchs, &c.) or not (Millepora). III.—On the Fragmentation of the Oosperm Nucleus. It is the belief of many eminent histologists that any process of division of the nucleus other than that by karyokinesis or mitotis ig a sign of the degeneration of the nucleus, and the approaching end of the life of the cells. Flemming says, “ Fragmentation of the nucleus, with and without subsequent division of the cell, is universally a 56 SYDNEY J. HIOKSON. process in the tissues of Vertebrates which does not lead to the physiological multiplication and reproduction of cells, but, on the contrary, represents where it occurs a degeneration or aberration, or perhaps, in many cases, is subservient to the metabolism of the cell by increasing the periphery of the nucleus.” Ziegler (65), who quotes the above passage from Flemming’s work, discusses in detail some of the many instances of amitotic nuclear division, and comes to similar conclusions. He says that amitotic division of the nucleus always indicates the end of the series of divisions, and considers it hardly probable that nuclei which have arisen by amitotic division will ever again divide by mitosis. If Flemming, Ziegler, and those who agree with them are right, then it is clear that the oosperm nucleus does not and cannot fragment. It must divide regularly by karyokinesis. But Ziegler’s views are, it seems to me, altogether unten- able. By simply denying, or passing over in silence, many instances of fragmentation of the nucleus, which do not support his views, he has given undue weight to mitosis, and leaves an unsatisfactory gap in the list of cases which support his theory. Verson (56), Frenzel (12), and Lowit (40) have, since the publication of Ziegler’s paper, called attention to cases of amitotic division of the nucleus which are most certainly not followed either by nuclear degeneration or by a cessation of cell multiplication. A review of the recent literature of cell division shows that the cases given by these authors may be supplemented by many others, and, indeed, leads one to a conclusion quite different from that of Ziegler and Flemming, namely, that indirect nuclear division rarely occurs unless it is preceded by or accompanied by some partial or complete segmentation or division of the surrounding cell substance. It is undoubtedly true that in many cases amitotic frag- mentation of the nucleus is followed by its degeneration and the death of the cell. The numerous examples quoted by DEVELOPMENT OF DISTICHOPORA VIOLACEA. 57 Ziegler prove that this is the case. But I shall endeavour to show that we are by no means justified in assuming that amitotic fragmentation is a sign of degeneration. In the first place, it can be shown that there is considerable evidence for believing that the oosperm nucleus of some ova does not divide by normal karyokinesis, but does split up amitotically into a large number of minute fragments. I have already described (17 and 18) such a process of fragmentation in the case of Millepora, Allopora, and Disticho- pora, but the following considerations prove that the same is probably true of many other eggs. Celenterata.—In the development of Alcyonium the germinal vesicle entirely disappears, and no traces of the karyokinetic division of the oosperm nucleus can be found. Kowalewsky! (37) gives a figure of the ovum without any nucleus, but my own observations show that at a stage cor- responding to the one he figures the nucleus is in the form of a number of minute fragments scattered through the substance of the ovum. The failure to find karyokinetic division of the oosperm nucleus cannot be attributed to imperfect methods of pre- servation or staining, because young embryos, preserved and stained in precisely the same way as the fertilised ova, exhibit beautiful and typical karyokinetic figures. The early stages in the development of Gorgoniacevolini, described by G. von Koch (36), seem to be precisely similar to those of Aleyonium. In the unfertilised ovum there is a large germinal vesicle containing an excentrically placed germinal spot, but in the eggs that he believed to be fertilised there was no nucleus. “Thre Structur weicht von der des unbefruchteten Hies wesentlich ab. Es fehlt namlich vor allem der Kern, von dem ich keine Spur mehr auffinden konnte.”” The fact that von Koch, after carefully examining over a hundred series of sections turough fertilised ova, could find neither traces of segmentation nor the division of the oosperm nucleus, suggests 1 As Kowalewsky’s paper is written in the Russian language I am unable to read it. 58 SYDNEY J. HICKSON. very forcibly that the ovum of Gorgonia does not segment at first, and that the oosperm nucleus fragments as it does in Alcyonium. Arthropoda.—In the development of Peripatus capen- sis, Sedgwick (51) has described the division of the ovum into two blastomeres, and the large and easily seen karyo- kinetic figures which mark the first division of the oosperm nucleus, The fertilised ovum of Peripatus nove-zealandiz, however, does not segment, and Miss Sheldon (53) was unable to find any karyokinetic figures in the divisions of its nucleus. It is a very striking fact in support of my views that in two species of the same genus we should find such a well- marked difference in this respect, the ovum that does seg- ment showing clear and unmistakable nuclear mitosis, and the ovum that does not segment showing no signs of karyo- kinesis. But this is not the only example of the relation between the segmentation and the division of the nucleus. In a recent paper on the “ Embryology of the Macroura ” Herrick (6) states that it is a rule with the decapod Crustacea that the nuclei of the segmenting eggs divide with karyokinesis. There is an exception to this rule, however, in the case of Alpheus minus. “The fertile egg of A. minus is pervaded with a remarkably fine reticulum which encloses spherules of minute and uniform size. The nucleus is central or nearly so, and consists of an ill-defined mass of protoplasm, in which a fine chromatin network is suspended. In the next phase the nucleus is elongated and about to divide. Division appears to be direct andirregular. Ata somewhat later stage the phenomena of the most interest occur. Each product of the first nucleus has developed a swarm of nuclear bodies which seem to arise by fragmentation. These bodies take the form of spherical nuclei in clear masses of protoplasm. ... In the last stage obtained the whole egg is filled with several hundred very large elements, which are descended more or less directly from some of the nuclear DEVELOPMENT OF DISTIOHOPORA VIOLAOEA. 59 bodies just considered, but the intermediate stages have not been considered.” In the species Alpheus Saulcyi and Alpheus hetero- chelis (two varieties) the segmentation is normal and regular, of the centrolecithal type, and the division of the nuclei indirect. In Alpheus minus alone is the segmentation extremely irregular and the nuclear division direct. Among Mpyriapoda we find that the ovum of Julus terrestris is very similar in many respects to that of the Stylasteride. There are no signs of segmentation, and there is no formation of cells until the time when the epiblast is formed. Heathcote (20), who carefully studied the early stages in the development of this species, could not find any signs of karyokinesis in the first divisions of the oosperm nucleus. There is, according to Kingsley (33), a disappearance of the germinal vesicle of the American Limulus, and it is a suggestive fact that Kishinouye (35), in his careful paper on the development of Limulus longispina, does not refer to the first nuclear divisions. It is possible that there may be a fragmentation of the oosperm nucleus in the ova of some other Arachnida. In the development of many Insecta there are many facts that point to the conclusion that the oosperm nucleus fragments. It is noteworthy in the first place that, notwithstanding the fact that several excellent embryologists have carefully studied the development of the common blow-fly, not one of them has been able to give a satisfactory account of the first division of the oosperm nucleus. Blochman (3), who figures the spindles of the nuclear divisions in the formation of the polar bodies, and also the spindles of the nuclear divisions of the later stages of embryonic development, did not apparently observe the first division of the oosperm nucleus. He says, “Als erste Theilung des Eikernes kann man die Bilder wohl nicht auffassen, weil, wie ein Blick auf die spateren Figuren zeigt, bei Theilungen die 60 SYDNEY J. HICKSON. Tochter kernplatten stets so fort weit aus einander riicken.” Henking (21), too, was unable to find the first division of the oosperm nucleus of Musca. Now, in Musca, and in many other insects in which the early divisions of the oosperm nucleus have not been made out, the occurrence “‘ of free nuclear formation” has been described in the young embryo. Whence come these free nuclei? It can hardly be believed that they are actually formed in the cell substance from something that is not directly derived from a pre-existing nucleus. All the evidence of modern histology tends to prove that nuclei are derived from nuclei, and nuclei only, and it is only reasonable to suppose that the so-called “free nuclei” of insect embryos are formed by the growth or fusion of fragments of the oosperm nucleus. The evidence in support of this hypothesis is not the purely negative evidence of the absence of any direct proof of mitotic division of the first nuclei, but the fusion of minute chromatin bodies to form larger ones has actually been observed by Henking (23) in the embryos of Pieris, Pyrrochoris, and Lasius. But it is extremely probable that fragmentation of the oosperm nucleus is of very frequent occurrence in the eggs of imsects. In many cases, both in large yolk-laden eggs and in small yolk-free eggs, the fertilisation is followed by the appearance of numerous nuclei in the substance of the egg. In Neophalax concinnus, one of the Phryganids, the division of the oosperm nucleus was not observed by Patten (46), and the following is his account of the early stages :— “Within ten or twelve hours after oviposition—the time varying with the temperature—a clear space makes its ap- pearance at the surface of the egg, and gradually increases until it has attained the breadth of the future blastoderm. In this layer, which has been called the ‘blastema,’ the pro- toplasm has, under ordinary conditions, a very homogeneous appearance, with occasionally lighter, less refractive spots, which appear like vacuoles, but in which, when observed more closely and under slight pressure of a cover-glass, or especially when treated with a very little acetic acid, faintly marked DEVELOPMENT OF DISTICHOPORA VIOLACEA. 61 nuclei make their appearance in greater or less numbers ac- cording to the more or less advanced stage of the blastema.” It is extremely improbable, if the minute nuclei in the blastema could be observed by the simple method of treat- ment with acetic acid, that the karyokinetic divisions of the large oosperm nucleus, if they really occur, would have been overlooked. Many other instances could be given from the writings of naturalists during the last twenty years of the failure to trace the divisions of the oosperm nucleus in insect eggs, and of the occurrence of “free nuclear formation” in the eggs after fertilisation ; but in many of these instances it might be urged that sufficient patience was not exercised, or that the methods of preservation and staining were imperfect. An important paper has, however, been recently published by Henking (23) containing an extremely elaborate account of his investigations upon many different species of .insects carried on with the aid of the best modern methods of re- search. It would take me far beyond the limits of this paper to give even an outline sketch of Henking’s important results, but a brief reference to some of the points bearing upon the subject of this essay must be made. In Pyrrochoris, one of the Hemiptera, Henking finds that in the formation of the polar bodies the nucleus divides by a process of karyokinesis, the chromatin bodies being of con- siderable size and definite in number. After fertilisation a new spindle is formed with the chromo- somes arranged in an equatorial plate, but before the division is completed the chromosomes disappear. Later on the chromosomes reappear in the form of extremely minute and numerous granules, which fuse together into threads, and arrange themselves in the equatorial plate of a new spindle, Similarly, in Agelastica alni, a Coleopteran, the chroma- tin entirely disappears after the division of the segmentation nucleus. In the Hymenopteran Lasius the chromatin of the first two segmentation nuclei completely disappears, and when the VOL, VI. 6 62 SYDNEY J. HICKSON. nuclei are about to divide again reappears in the form of extremely minute granules, which fuse together to form the chromosomes of the next division. A similar disappearance has been described in the unfer- tilised egg of Rhodites, and in this form there is no mem- brane surrounding the nuclei. These researches prove, then, that in some insects there is a “ disappearance”’ of the chromatin substance of the nucleus after its first division. To what is this disappearance due? MHenking thinks that it is due to some chemical change in the chromatin substance, as in some cases the outline of the chromosomes may be ob- served after the disappearance of the colouring matter. Never- theless it is a fact that commonly the chromosomes lose their compact form during the colourless stage, and become very finely divided. We can attribute the disappearance, then, partly to the change in the chemical character of the chro- matin, and partly to the very minute and scattered condition of its elements. Further, in some cases (Rhodites) not only does the chro- matin disappear, but also the membrane surrounding the nuclear area, so that we have (as in Distichopora, &c.) a condition in which the nucleus is practically indistinguishable from the surrounding protoplasm. It is during this condition that some of the nuclear fragments may be distributed through the substance of the ovum, and give use to the nuclei of the so-called “free nuclear formation” by subsequent fusion. It must be obvious to anyone who carefully studies Henking’s figures that in many insects the spindle of the first division of the oosperm nucleus is very irregular, that the chromosomes are not always arranged with the same mathe- matical precision that they are in typical karyokinetic figures, and further, that in consequence of the disappearance and extremely fine division of the chromatin substances there are still some steps in the nuclear divisions at the commencement of development which have not been satisfactorily traced. DEVELOPMENT OF DISTICHOPORA VIOLACHA. 65 We may go further than this, though, and say that some of Henking’s figures, such as figs. 335, 336, and 337 of Lasius, can only be interpreted on the supposition that the nucleus has fragmented. The little clusters of chromatin granules, of very irregular size and indefinite arrangement, that are here figured scattered through the substance of the ovum, cannot be considered to be the product of regular mitosis. It seems to be extremely probable that in the group of insects we have a series of stages intermediate in condition between regular mitotic division of the oosperm nucleus or its immediate successors and irregular fragmentation. In Aphis (Will, 62) we may have regular karyokinesis at all stages of the segmentation, the chromosomes being divided into two equal halves at each division of the nucleus; but in Musca, in Lasius, and perhaps in several others in which the earliest stages are passed through with great rapidity, the nuclei fragment with greater or less irregularity. That the occurrence of karyokinesis is in some way dependent upon forces manifesting themselves in the cell substance of the ovum and acting upon the nuclei is rendered probable (1) by the fact that in Aphis, where the nuclei divide by karyokinesis in all stages, there is, as Will points out, a distinct aggregation of protoplasm round the nuclei, and (2) by the fact that in nearly all insects the karyokinetic figures of the nuclear divisions that take place in the formation of the polar bodies are much more regular and constant than they are in the early stages of development. But I shall discuss this point and general significance of mitosis in greater detail later on. That a similar process of fragmentation of the oosperm nucleus may also occur in some Vertebrata seems to be probable from the recent researches of Kastschenko (31) upon Elasmo- branchs. It must be remembered that the early stages of the development of Elasmobranchs and birds have been carefully studied by numerous observers for the last twenty years, and although the karyokinetic spindles in the developing blasto- derm and its surrounding yolk have been described by nearly 64 SYDNEY J. HICKSON. all of them, we have not received any account of the first division of the oosperm nucleus. It is quite unreasonable to suppose that all these observers would have overlooked a nuclear division—which we might expect, if it exists at all, to be the largest and most conspicuous of the whole series. Nor can we suppose that the methods of preservation or staining was so consistently bad at the first stage as to prevent the observation of the figure, and so fre- quently good in the later stages as to show the whole process of karyokinesis clearly and distinctly. Now Kastschenko (31) shows that in Elasmobranchs a number of nuclei appear in the blastoderm and the surround- ing yolk before the formation of the segmentation furrows, which appear not in regular sequence, but simultaneously and irregularly. ‘Die bekannte regelmassige Reihenfolge des Erscheinens des Segmentationsfurchen existiert bei Selachiern fast gar nicht. Nur in seltenen Fallen bemerkt man das urspriingliche Erscheinen einer Segmentationsfurche, welcher dann gleichzeitig mehrere andere unregelmissig sich kreuzende folgen. In den meisten Fallen aber erscheinen schon vom Anfang an mehrere Segmentationsfurchen gleichzeitig und somit zerfallt die Keimscheibe direct in mehrere verschieden grosse Segmentationskugeln, welche sich dann weiter aber nicht gleichzeitig theilen.”’ We have, then, at the commencement of the development of the Hlasmobranch a multinucleated plasmodium, and Kast- schenko is of opinion that all the nuclei of this plasmodium are formed by repeated divisions of the first segmentation nucleus- But, like all his predecessors, Kastschenko was apparently unable to observe these repeated divisions of the first nucleus, and it seems extremely probable that in Elasmobranchs, as in insects, Hydrocorallines, and others, we have at this stage a true process of nuclear fragmentation. I have already called attention to the fact that in all of these cases in which the fragmentation of the oosperm nucleus probably occurs the ovum does not segment immediately after fertilisation ; that there is, in fact, for a time in the early em- DEVELOPMENT OF DISTICHOPORA VIOLACEA. 65 bryonic development a multinucleated plasmodium without any definite cell walls or cell areas. There can be little doubt, I think, that in all holoblastic eggs, such as those of Echinoderms, worms, Amphioxus, &c., the first segmentation is accompanied by typical karyokinetic division of the nucleus. We may go further than this, and say that in many mero- blastic eggs the first divison of the oosperm nucleus is also an indirect one. Vialleton (57) and Watase (59) have observed this division in the egg of Cephalopods, and Oppel (45) has observed it in the egg of the lizard, Anguis fragilis. But in both these cases the segmentation furrows occur regularly and in sequence from the commencement of development, and we have, consequently, evidence that the same forces are at work in the protoplasm as those which produce the more or less complete blastomeres of holoblastic eggs. Even in those eges of insects in which the nuclei are known to divide by karyokinesis there is evidence of the drawing together of the protoplasm along certain lines of force in the “‘ plasmatische _ Strahlungen” of Henking, which surround the nuclei. But if there is any truth in the view that I have here put forward, that karyokinesis is primarily due to the forces which bring about cell division, and that in those cases in which cells or cell areas are not formed the nucleus may fragment or divide directly in some other way, then we should expect to find some further evidence of fragmentation of the nucleus in other tissues. There is ample evidence of this iu other tissues. In the formation of the spores in Protozoa the nucleus of the parent cell often divides long before there is any division of the cell protoplasm, and in nearly all such cases division of the nucleus is direct. In some cases the nucleus disappears, and it is probable, as in the case of the oosperm nucleus quoted above, that this may be due to the.extremely fine division oi the chromosomes and fragmentation. I will give just a few examples to illustrate these points. Wolters (64), in describing the conjugation of Monocystis magna and agilis, says, “ Kurz nach erfolgter Encystirung 66 SYDNEY J. HICKSON. soll der Kern, respective die Kerne der beiden Copulanten sehr undeutlich werden. Sie entziehen sich zuletzt dem beo- bachtenden Auge ganz und sind im Inhalte der ausgequetschten Cyste nicht mehr zu finden.’ The author figures, it is true, an achromatic spindle in the encysted forms after the extrusion of the polar bodies, but the chromatin bodies are very minuteand irregularly scattered through the substance of the protoplasm. “Es gelangzwar nicht,” he says, “eine zusammenhingende Reihe von Bildern fiir die Constatirung der mitotischen Theil- ung an den Sporogonien zusammen zustellen, doch liess sich mit Sicherheit constatiren, dass die Kernmembran an manchen Kernen der ungetheilten Sporogonie geschwunden war und die farbbare Substanz in zwei, durch einen groésseren Zwischen- raum getrennte Reihen angeordnet war.” But the evidence in favour of a process of fragmentation of the nucleus seems to be much more conclusive in the case of Clepsidrina blattarum (Wolters, l. c.). In this form nuclei are found “in denen unzahlige kleine chromatische Korner lagen, wie es schien regellos, ohne besondere Anord- nung vertheilt. Allen bisheran geschilderten Kernformen war dagegen eine scharf contourirte Kermembran gemeinsam. Im Gegensatz da zu stehen Formen, die ebenfalls haufig beo- bachtet wurden, welche einer solchen Membran entbehrten. Der Kern breitet sich sternformig mit seinen Fortsatzen in die Leibessubstanz des Thieres aus und steht mit dem proto- plasmatischen Gefiige derselben in directen unterbrochenen Zusammenhange.” This account of the fragmentation of the nucleus of Clep- sidrina blattarum is confirmed in all its essential details by the more recent work of Marshall (40a), who was unable to find at any time any traces of karyokinesis. A very similar account is given by Schneider (49) of the division of the nucleus of Klossia. It may be that in some forms, such as the Gregarina irregularis of Holothuria nigra (Minchin, 48), a regular form of division with mitosis does occur, but this does not detract from the importance of the fact that in many DEVELOPMENT OF DISTICHOPORA VIOLACEA. 67 Gregarines which form during encystment a vast number of spores, no karyokinetic figures can be observed. Many years ago Hertwig (24) described a curious method of the fragmentation of the nucleus without karyokinesis in the spore formation of Thalassicola, and more recently Brandt (4) was unable to find karyokinesis in the divisions of the nucleus to form the nuclei of the spores of the Sphzrozooids. Gruber (16) has described several instances among the ciliate Infusoria in which the nucleus apparently fragments into extremely minute granules, which become scattered through the protoplasm of the body and collect again into lumps. Jickeli (29) has described fragmentation of the nucleus of Stylonychia, Paramecium, and other Ciliata. Quite recently, too, Lister (39), in his researches upon Orbitolites, has not been able to discover any signs of karyo- kinesis in the division of the nuclei. There is probably, too, a method of fragmentation in the spermatogenesis of many animals. I have myself carefully examined the earliest divisions of the nucleus of the sperm mother-cell of Millepora, Allopora, Distichopora, and Alcy- onium, and I can find no trace of karyokinesis. It is, in fact, only in a few exceptional cases, such as Ascaris (Hertwig, 25), where the cell outlines of the spermatocytes are very early delineated, that karyokinesis has been observed in the division of the nuclei of the sperm mother-cells. Verson (56) shows that in Bombyx mori the primordial cells have at first a giant nucleus, which divides amitotically to form numerous secondary nuclei, and these divide mitotically to form the nuclei of the Spermatids. Bolles Lee (38) found amitotic division of the nuclei of the spermatogonia of Chetognatha and Nemertines and regular karyokinesis in the division of the nuclei of the spermatocytes. Dostojewski found the same thing in the spermatogenesis of Amphibia (see Waldeyer, 58, p. 39), and other examples could be quoted from the writings of La Valette St. George, Gilson, Sabatier, and others (see Waldeyer, 58, p. 39). 68 SYDNEY J. HICKSON. In some Annelid worms the nucleus of the spermatogonium disappears, and there is no evidence at present that the nuclei of the spermatocytes are derived by repeated mitotic divisions of this nucleus (Jensen, 28, and others). In the recent work on spermatogenesis, by Pictet (47) no mention is made of the manner in which the nuclei of the spermatogonia divide in Polychetes.1 A study of the literature of spermatogenesis shows that when there is a distinct division of the protoplasm to form the spermatocytes or spermatogonia, distinct karyokinesis of the nuclei may generally be seen; but when, on the contrary, multinucleated cells are formed, which eventually give rise to the spermatocytes, the nucleus of the spermatogonia either disappears or divides amitotically. It is not necessary for me to discuss in detail the numerous cases of indirect fragmentation of the nucleus that have been described by Arnold in his numerous papers in ‘ Virchow’s Archiv’ and the ‘Archiv fiir mikroscopische Anatomie,’ by Werner (61), Schottlaender (50), Hess (26), Geelmuyden (18), Beltzow (2), Strobe (54), Goppert (15), and others. Many of these cases are those of the nuclear division of giant-cells, and I believe I am quite correct in saying that in all of them the fragmentation of the nucleus is not immediately followed by cell division. The general conclusions to be drawn from the evidence before us are—1. That fragmentation of the nucleus is a normal method of nuclear division, and is not always a sign of pathological change. 2. That in many of the instances in which the nucleus is supposed to disappear there is, as a matter of fact, minute fragmentation. 3. That fragmentation only occurs where there is no cell division; and 4. That karyokinetic division of the nucleiis caused by the forces in the cell protoplasm which bring about the division of the cytoplasm. That there may be many forms of nuclear division inter- 1 It is a noteworthy point that O. von Rath (48), who believes that the nuclei of the spermatogonia and spermatocytes always divide mitotically, does not refer at all in his paper to the spermatogenesis of Polychetes. DEVELOPMENT OF DISTICHOPORA VIOLACHA. 69 mediate in character between fragmentation and bipolar karyokinesis seems to be probable from the discovery of pluripolar mitosis in the inflamed cornea by Schottlaender (50), and other atypical nuclear divisions in the spleen of the white mouse by Arnold (1), &c. We have, then, a series of phenomena in the division of nuclei, with typical karyokinesis at one end and direct frag- mentation at the other. The occurrence of any one kind or the other is, in my opinion, determined by the forces which act simultaneously upon nucleus and cell plasm. If these forces are of such a kind as to drag the cell plasm into two equal halves, the nucleus is also dragged into two equal halves with mitosis; if, on the other hand, the forces are irregular and act from many centres at the same time, the nucleus fragments irregularly. These views seem to me to be supported by the statement of Flemming (11) quoted by Sedgwick, that “the first change observable in a cell whose nucleus is about to divide is in the extra-nuclear protoplasm,” and by Biirger’s (7) recent con- clusions concerning the meaning of the spheres of attraction. List oF REFERENCES. - . Arnotp, J.— Weitere Mittheilungen iiber Kern- und Zelltheilung in der Milz,” ‘ Archiv f. mikr. Anat.,’ xxxi, p. 541. . Bextzow, A.—‘ Zur Regeneration des Epithels der Harnblase,” ‘ Vir- chow’s Archiv,’ 97, 1884. 8. Brocumann, F.—‘ Ueber die Richtungskérper bei Insecteneiern,” ‘Morph. Jb.,’ xii, 1887. 4, Branpt, K.—‘‘ Die Koloniebildenden Radiolarien,” ‘Fauna und Flora des Golfes von Neapel,’ xiii, 1885. 5. Braver, A.—‘ Ueber die Entwicklung von Hydra,” ‘Z. wiss. Zool.,’ lii, Hft. 2, 1891. 5a. Braver, A.—‘ Ueber die Entstehung der Geschlectsprodukte und die Entwicklung von Tubularia mesembryanthemum,” ‘Z. wiss. Zool.,’ lii, Hft. 4, 1891. 6. Brooxs, W. K., and Herrick, Ff. H.—“a hte 3 cA a ry i Zz SA ist - - ~~ neaiaaeagenn “ » : 7 “> Farasitic OF GANS INS. .. Lister de/ , ve Studies M.L.Vol.VI.P1.VII. y small algae. - Frotoplasm belween F is ns the young. £ (ig Primordial _- Pee chamber. chambers. Cambridge Engraving Co J. G. KERR. ON THE ANATOMY OF NAUTILUS POMPILIUS. 181 On some points in the Anatomy of Nautilus pompilius. By J. Graham Kerr, Christ’s College, Cambridge. (PLATES IX. anp X.) I. Introduction, p. 181, II. The Body-cavity of Nautilus, p. 182. Ili. The Male Genital Ducts and Penis, p. 191. IV. The Buccal Nervous System, p. 194. V. The Innervation of the ‘Inner Inferior Lobe,” p. 197. VI. The Post-anal Papille and Nerves, p. 198. VII. The Spermatophore-receiving Apparatus, p. 199. VIII. The Morphology of the “‘ Arms” of Cephalopods, p. 201. IX. The Phylogenetic Relationships of the Cephalopoda, p. 208. X. Summary of Conclusions, p. 210. Explanation of the Plates, p. 211. I. INTRODUCTION. DurinG the year 1893 Mr Adam Sedgwick very generously placed at my disposal a number of specimens of Nautilus pompilius with the suggestion that I should make an examina- tion of their structure. The specimens were twenty-five in number, of which, however, the great majority were very young and immature. Owing to the method of preservation and to several months’ sojourn in sawdust moistened with spirit, the condition of the specimens was usually such as to render them unfit for histological study. Fortunately one of them was sufficiently good to allow the use of the section-method to confirm the results of minute dissection. In the following somewhat fragmentary paper it is my purpose to touch upon what seem to me the more important points at which I have arrived, hoping at some future date, if able to obtain properly 182 J. GRAHAM KERR. preserved specimens, to extend my investigations and to fill up the obvious lacune. I can hardly adequately express the obligation under which I am to Mr Sedgwick for the generous gift by which he has made these investigations possible and opened the way to what, however poor its results are so far, has proved a study of absorbing interest, and also for much kind advice and en- couragement. To Mr Wilson also a word of thanks is due for the care with which he has attended to the illustrations. II. Tse Bopy-cavity oF NAUTILUS. It is now generally recognized that the body-cavity in the higher Metazoa may be referred to either of two very distinct types. The first of these, typically developed in Annelida and Vertebrata, is lined by a definite characteristic epithelium, from some of whose cells arise the genital products, while others become the renal excretory cells. It appears at an early stage in development as a more or less continuous space, and it communicates with the exterior by apertures in the body-wall. To a body-cavity of this type it is advisable to restrict the term Colom. The second type of body-cavity is to be found in the Mollusca and Arthropoda generally. It is part of the vascular system, through it is pumped a continuous stream of blood by the heart, and it does not communicate with the exterior. It may be looked on as being formed by the expansion of the terminal parts of the blood-vessels into large sinuses whose walls have, to a greater or less extent, disappeared, giving rise to a sponge-work more or less sparse according to the extent to which this process has gone on. This type of body-cavity was named by Sedgwick, Pseudocel; by Lankester, Hemocel. The word ccelom has been used with such looseness that Lankester’s term is perhaps to be preferred ; all the more so as it specifies in itself one of the main characteristics of this form of body-cavity. Occurring well-developed in Annelids, at least allied in all ON THE ANATOMY OF NAUTILUS POMPILIUS. 183 probability to the ancestral forms of Molluscs and Arthropods, the ccelom is to be looked on as the more primitive of the two types of body-cavity above-mentioned ; and it looks as though within each of the two latter groups it had gradually dwindled and become supplanted and replaced as the functional perivisceral cavity by the ever increasing hemoceel. In most Cephalopods the ccelom still takes a large part in the formation of the perivisceral cavity, and in Nautilus, corre- sponding with its more archaic character, this is so to a greater extent than in any of the other Cephalopods. The hemoccel of Nautilus is specially developed in the headward section of the body. A sagittal incision through the body-wall just behind the hood exposes to view a large chamber in which lies the pharynx as well as the vena cava, several large nerve-trunks, and a single loop of the intestine. This cavity is the main division of the hzmocel; ventrally it is bounded by the body-wall and the muscular substance of the hood, etc., into which it extends in numerous sinuses, while dorsally and towards the apex of the visceral hump it is bounded by a thin and delicate but complete membranous septum which forms the boundary between it and the ccelom. The inner (“ventral”) face of this septum has a rough and spongy appearance, and little connective-tissue strands pass from it to the surface of the pharynx. These delicate threads of connective tissue traversing the cavity and slinging up its contained organs at once suggest the haemoccelic nature of this part of the body-cavity: and the conjecture is confirmed on raising up the pharynx, for one then sees that the upper wall of the vena cava is perforated by numerous foramina, some of considerable size, which put its cavity into free com- munication with that part of the body-cavity now under discussion. These foramina were described and figured long ago by Owen, in his Monograph, but they appear to have been unnoticed by subsequent observers’. 1 Since writing the above I see that Pelseneer in his recent Etude des Mollusques, p. 191, says that ‘la cavité viscérale est un vaste sinus communiquant avec la veine cave par des orifices percés dans la paroi de celle-ci,”’ 184 J. GRAHAM KERR. The true ccelom (wiscero-pericardial sac, Owen) has received comparatively little attention from previous investigators, Fia. 1. g n c \ r ee SIE = Biz Sh Li) ; Ln = 4 a BAG AVS ; NY Nore: Saag ——— haem Sayittal section through Nautilus. Diagrammatic sagittal section through the animal of a young female of Nautilus pompilins, to show the general relations of the celom and hemocel. (The section really passes very slightly to the right of the mesial plane, so as to traverse one of the renal chambers.) haem, hemocel; g.c, genital division of the ccelom; p.c, pericardial division of the celom; a, aperture in septum dividing g.c from p.c; k, right inner kidney-chamber; 7, pericardial gland corresponding to this kidney-sac; h, ventricle; ov, ovary; ph, crop; giz, position of its opening into the gizzard, which lies to the left of the plane of section; int, intestine; v.c, vena cava. Grobben and Lankester being the only authors who devote to it more than a few passing words}. It is convenient to treat together the ccelom itself, the 1 Pelseneer, op. cit. p. 192, says that the celom ‘‘s’étend dorsalement, autour de l’estomac, jusque vers la moitié de ’cesophage. Il contient, outre le cur, la glande génitale, la veine cave et une partie des appendices glandulaires des vaisseaux branchiaux afférents,”—a statement which is obviously at variance with the account here given in two important respects. ON THE ANATOMY OF NAUTILUS POMPILIUS. 185 excretory and the genital organs as forming all parts of the same organic complex. On reference to the diagrammatic longitudinal section through the animal, it is seen that the cecelom is limited to the aboral end of the body, where it forms a flattened space immediately underlying the body-wall— between this latter and the thin membranous bag which limits the cavity of the hemocel. The ccelomic cavity is divided by an oblique septum into a large upper division (g.c.), the genital division of the ccelom, and a smaller lower pericardium (p.c.). The septum separating these is not quite complete, being perforated by three apertures of considerable size. One of these is indicated in the diagram at a. The Genital division of the Colom (g.c.) is, as already men- tioned, the larger of the two ccelomic chambers. It occupies the extreme aboral (dorsal) end of the body, and is lined throughout by a delicate epithelium composed of flattened plate-like cells usually hexagonal in outline, the cell-boundaries, however, being very indistinct. Each cell contains a rounded nucleus with chromatin network and one or two small nucleoli, Into this division of the ceelom project the gizzard, the greater part of the length of the intestine, and the genital gland. It must of course be remembered that all these organs are invested by the celomic epithelium, so that it is only in a certain sense that they can be said to be situated within the cavity. The genital gland being merely a specialised part of the wall of the ccelom, it may appropriately be shortly described at this point. The Ovary is a flattened ellipsoidal body attached by a mesovarium to the lower (posterior) side of the intestine, and at its oral (ventral) end having a considerable aperture which throws its cavity into continuity with that of the surrounding celom. The general characters of the ovary are shown in fig. 2, representing a sagittal section through the organ of an immature female. The ovary exists here in almost its simplest possible condition, in the form of a specialized ovigerous area of the ccelomic epithelium roofed over and protected by a simple upgrowth from the coelomic wall, The outer surface of the 186 J. GRAHAM KERR. organ is covered over by the general epithelium of the celom, having the characters already mentioned, and at the opening at the oral (ventral) end of the ovary this is inflected into its interior so as to line this likewise. The cavity of the ovary is Longitudinal section through ovary of a young Nautilus. c.ep, ccelomic epithelium covering outer surface of the ovary; cil.ep, ditto reflected into aperture of ovary; ov.foll, ovarian follicles; ov. 1, 2, 3, ova in various stages of development; b.s, blood-sinuses in the wall of the organ, thus merely an incompletely shut-off portion of the ccelom. Traced into the interior of the ovary, the epithelium about its opening assumes a columnar form and bears long cilia (czl. ep.). Along its roof the cells become shorter and eventually cubical. About two-thirds of the way from the mouth of the cavity the ovigerous region is reached, and this occupies the remainder of the roof and nearly the whole of the floor of the ovary. The ovigerous region of the cavity is thickly beset with egg-follicles of various ages (ov. foll.). In the recesses between the bases of these the lining epithelium—a thin protoplasmic layer with scattered nuclei and indistinct division into cells—thickens up into syncytial masses of protoplasm containing large round nuclei, each with a large deeply staining nucleolus, around which the protoplasm tends to segregate off more or less distinctly. The primitive ovum develops within such a heap, the nucleus increasing in size and assuming more and more the ON THE ANATOMY OF NAUTILUS POMPILIUS. 187 character of a “germinal vesicle,” and the protoplasm first becoming more distinctly aggregated round the nucleus and marked off from the surrounding protoplasm and then increas- ing rapidly in size. As the ovum increases in size the substance of the ovarian wall grows up round it to form the follicle, while the syncytium accompanying the ovum apparently gives rise to the lining-cells of the follicle. The latter are pear-shaped structures borne on stalks, which are usually simple, but occasionally branch, thus showing a tendency towards the con- dition in Argonauta, where they are much branched and tree- like. Externally the follicle is covered by a very thin epithe- lium, which distally becomes continuous with the lining layer of “follicle cells’—a layer of thick columnar cells immediately surrounding the egg. In the young follicle this layer runs concentrically with the outer surface of the epithelium, but as the egg increases in bulk an increase in the surface of this apparently nutritive organ becomes necessary and the follicle- epithelium grows inwards as a series of anastomosing folds. On this account the older eggs when removed from the follicle show on their surface a network of deep fissures formed by the follicular epithelial ingrowths. In the oldest female specimen accessible, unfortunately too macerated to make out many details, the eggs had reached a large size, over 10mm. in length, and their substance was already enormously yolk-laden, the protoplasm being practically restricted to a small cap on the end of the egg next the follicular opening. Imbedded in this was the large nucleus with densely staining nucleolus. The presence of a definite opening in at least the older follicles and the position of the egg-nucleus close to this, suggest the possibility of fertilization in Nautilus being internal ; and the great size of the eggs and their yolk-laden character point to the segmentation being meroblastic as in other Cephalopoda. The wall of the ovary, as of other important organs, is loose and spongy, traversed by extensive blood-sinuses (b.s.). Pro- longations of these pass up the stalks of the follicles, and form a specially developed layer immediately underlying the follicle- epithelium—a condition to be correlated with the provision of 188 J. GRAHAM KERR. an abundant blood-supply to satisfy the needs of the glandular epithelium. The Testis is, m its main morphological features, quite similar to the ovary; in other words, it is an invaginated area of the lining of the ceelom. Only in the testis great increase in the area of the germinal epithelium has been brought about by the involuted portion of coelomic epithelium, instead of remain- ing a simple sac, becoming divided up into a system of delicate branched tubes. In an apparently adult specimen, the testis was a large brownish organ of roughly triangular shape, its rounded apex directed upwards and towards the right side. Its apical portion was in close contact with the body-wall, while its basal part was separated from the body-wall by the pericardium. The testis is slung up by a strong ligamentous band about lcm. broad to the tunic of the gizzard, by a similar but broader band which is attached along a sagittal line to the body-wall (the root of the siphuncle being about the middle of its lime of attachment), and finally along its anterior face by a thin peritoneal fold to the loop of the intestine. Further, at its lower end the epithelium covering the outer surface of the testis is continued into that covering the pericardial septum and heart. Near the inferior angle of the organ is its aperture— a slit about 2mm. in length, bounded by two flat, much projecting lips, which, lying closely opposed to one another, project into a deep recess covered by a crescentic flap, the internal opening of the vas deferens. Thus, though the cavities of the testis and of the vas deferens open quite independently into the celom, they are at least during sexual maturity functionally continuous with one another. In a section through the testis of a young individual, the aperture of the organ is seen to lead into a vestibule into which open several straight ducts. Each of these, traced inwards, divides up into numerous tubules which end blindly and are aggregated into distinct lobes and lobules. Vestibule and tubes are lined by epithelium continuous with that of the general celom. The wall of the organ is traversed by a sponge-work of ON THE ANATOMY OF NAUTILUS POMPILIUS. 189 blood-sinuses. Between the lobules these are greatly developed, their separation walls being reduced to fine connective tissue- threads serving to bind the lobes together. Into the substance of the lobules also pass continuations of the sinuses. Regarding the character of the epithelium in different parts of the tubes, the state of the specimens does not allow me to say anything. The Pericardium or inferior chamber of the ccelom is con- siderably smaller than that already described. It immediately underlies the “ postero-dorsal” body-wall throughout its half next the mantle-flap, and its cavity is conveniently exposed by reflection of its external wall. It is then seen to be quad- rangular in outline, rather broader than long (68 mm. x 54 mm. in one specimen). From its inner (anterior) wall, in a curved row parallel to the ventral (oral) border of the chamber, project the four clusters of pericardial gland-follicles. The external pair are in such a view (2.e¢., from posterior) partially hidden by a broad freenum, which on each side connects the anterior wall of the chamber with the posterior wall. Dorsal (aboral) to the two central pericardial glands is seen the ventricle firmly bound down to the anterior wall of the chamber—the epithelium lining which is reflected over its surface. Just dorsal to the ventricle a large rounded aperture leads into the genital division of the cclom, and ventral to it is a still larger such opening. The four auricles attached to the corners of the ventricle, unlike it, hang quite free in the pericardium. In some specimens these were markedly asymmetrical, those of the left side being much more dilated than those of the nght. Each of the divisions of the ccelom above described is in open communication with the exterior. In the case of the pericardium, one finds at its ventral end that the cavity is prolonged on either side on the anterior face of the frenum mentioned. Each such prolongation forms a small somewhat triangular chamber with its greatest diameter transverse, and this at its mesiad end opens into the mantle-cavity by the tumid lipped, so-called viscero-pericardial aperture. The genital division of the ccelom primitively possesses at its ventral end also a communication upon each side with the exterior. In the VOL. VI. 17 190 J. GRAHAM KERR. actual animal, however, one of these has become closed intern- ally, as Lankester has shown, while the other persists in the female as the oviduct, in the male probably as the part of the functional genital duct extending from its coelomic opening to the inner end of Needham’s sac, On pulling the mantle dorsalwards, so as to afford a view of the interior of the mantle-cavity, such as that shown in Lankester and Bourne’s figure, one notices a little distance to the headward side of the root of each gill one of the four kidney-openings. These are arranged in two pairs. Just to the mesiad side of each of the posterior openings, one sees the slit-like viscero-pericardial apertures, leading, as above men- tioned, into the pericardium. This condition in Nautilus, where the viscero-pericardial sac opens independently of the kidney, is homologized, and no doubt rightly so, with the condition met with in Spirula and Agopsids, where the viscero-pericardial canal opens into the kidney-sac near its mouth, by supposing the opening of the latter to have migrated on to the outer surface (Grobben, Lankester), an identical process to that which has taken place in, eg., the genito-urinary passage and the rectum in Mammals. Accompanying the anterior kidney openings no such pericardio-visceral pores are seen, and in consequence of this it has been concluded that the anterior and posterior kidney-sacs are not serially homologous. All agree in regarding the posterior one as primitive, but the anterior sac is looked on as a secondary formation—either as a secondarily arising repetition of the posterior one, or as having been split off from it in correlation with the development of a new gill and new afferent vessel (Grobben). As a matter of fact, however, such a viscero-pericardial aperture is present, corresponding to the anterior kidney-open- ing. Itis the primitive genital aperture. Such is seen either in the case of the oviduct or of the rudimentary left genital duct of either sex’. This opening leads into the genital division 1 In the case of the functional genital duct of the male, a shifting of the external aperture has taken place through the, in all probability, secondary development from the adjoining body-wall of the penis. ON THE ANATOMY OF NAUTILUS POMPILIUS. 191 of the ccelom just as does the viscero-pericardial pore into the pericardium, and, like it, is situated mesiad to the kidney- opening. The only striking difference is, that this pore is normally rather farther apart from its corresponding kidney- opening than is the viscero-pericardial pore. The latter is normally quite close to its kidney-opening, but its distance from it is very variable and may reach 3 mm. It appears to me that there can be no question as to the homology of the two sets of apertures. In the genital segment, however, the migration of the coelomic aperture has gone a little further beyond the bounds of the kidney-sac. Each ccelomic duct, plus its kidney-sac, would on this view correspond to an ordinary “nephridium,” 2.¢., a tube leading from the ccelom to the exterior, part of the wall of which has taken on an excretory function. In the Dibranchs, in correlation with the disappear- ance of the anterior gill, the corresponding kidney-sac has disappeared, while its ccelomic duct persists as the genital duct. The genital ducts of the Cephalopoda in general then are nephridia?, minus their excretory sacs. Ill. Tse Mare GenitaL Ducts AND PENIS. The general disposition of the genital apparatus in the male is shown in fig. 3 (p. 192). As is well known, only the duct of the right side is functional in Nautilus. On the left side there is the “pyriform sac” of Owen, shown by Lankester and Bourne to represent the left genital duct, although the question was left open by them—whether it represented only the genital duct, or the genital duct together with the genital gland of the same side’, From the large ccelomic aperture the genital duct passes through the quadrangular “accessory gland” composed of 1 Pelseneer asserts that the genital ducts of Cephalopods are nephridia— without, however, qualifying his statement or supporting it by evidence. 2 That it represents only the duct appears to me to be shown by the condi- tion in the very young animal, in which the inner part of the genital duct has exactly the appearance of the pyriform sac in the adult—the rudiment of the gonad being quite distinct and apparently median and unpaired. 17—2 192 J. GRAHAM KERR. numerous cecal tubular outgrowths from the duct itself. Beyond this point the duct opens into the spermatophore sac— a large structure somewhat elliptical in outline when seen from the anterior (dorsal) or posterior (ventral) aspect (Pl. X. fig. 1). The vas deferens opens into this at its outer end. Internal to this opening there begins a longitudinal septum which divides the cavity of the sac through about half its length—terminating by aS . Sig ¢ ’ | 4 aa an) ry i. ‘ ry ‘ - . ® ‘ . r a 4 ‘ ‘ ’ . ‘ ~S oh Diagram of the testis, genital duct, and penis of the male Nautilus. The rudi- mentary genital duct of the left side is dotted in. T, testis: at its upper end is seen its aperture into the general cavity of the celom. Ac,Gl, accessory gland with internal opening of the vas deferens beneath the crescentic flap at its left-hand corner. Sp.Sac, sper- matophore sac. The curved line traversing the diagram from side to side represents the line along which the mantle-flap is reflected headwards. ON THE ANATOMY OF NAUTILUS POMPILIUS. 193 in a free concave edge. In the sexually mature animal the much coiled-up spermatophore mass occupies the cavity of the sac bending round the edge of the septum. The internal surface of the sac-wall is smooth to the naked eye, while a low- power lens discloses the existence of minute glandular-looking ruge running on the whole parallel to the axis of the sperma- tophore mass. At its anterior inner corner the cavity of Needham’s sac passes into the penis. This is a somewhat cylindroidal, flat- tened structure about 10 mm. in length and 8 mm. greatest breadth, attached to the body-wall within the mantle-cavity. Its walls are thick and muscular, and its cavity is divided by a sagittal longitudinal septum, which does not extend quite to the tip of the organ, into two moieties. Of these it is only the right with which the sac of Needham communicates, the left being (as will appear later) connected merely with a peculiar blind sac. The right penial cavity is somewhat semi-pyriform, becoming narrower distally. Its lining is thrown into large, smooth, glandular-looking ruge, which anastomosing with one another form a kind of raised network with elongated meshes. Outside this lining is the muscular coat about 1:5 mm. thick and largely composed of radial fibres. The muscular layer is traversed by an extensive system of blood-spaces. This is most developed towards the “posterior” end of the penis. It forms a distinct layer near the outer surface of the organ, but its spaces also, though less conspicuously, ramify hither and thither in the general substance of the muscle. The left penial cavity is cylindroidal in form, and its diameter only about half that of the right cavity at its widest part. The inner surface of its wall is also thrown into folds; but these are mainly longitudinal, parallel, and do not anasto- mose to the same extent as do those of the right cavity. The lining-tissue is of a less deep colour and less glandular-looking ; the muscular wall is thinner, and the cavernous layer is also less developed. At its “posterior” end, about the level of the point at which the right cavity becomes continuous with the sac of Needham, 194 J. GRAHAM KERR. the left cavity diverges towards the left side, much as the long axis of the Needham’s sac does towards the right, and gradually expands into a flask-shaped sac, in this specimen 6 mm. long by 3 mm. broad. This is rounded off and ends blindly. The inner surface of its wall exhibits faint longitudinal corrugations. It is difficult to believe that this left moiety of the penial apparatus does not represent the reduced fellow of the right moiety, z¢., of the right penial cavity plus the sac of Needham. On the left side, however, the rudimentary vas deferens does not communi- cate with the penial sac, but opens, as is well known, directly into the mantle-cavity. The position of this external aperture corresponds very closely to that of the opening of the vas deferens into the spermatophore sac on the opposite side. The whole arrangement strongly suggests that of the functional male genital duct, only that portion from the ccelomic aperture to its opening into the sac of Needham represents the primitive duct, and that the Needham’s sac and the penis are secondarily added structures developed from the adjacent wall of the mantle-cavity. In the young animal, the Needham’s sac being not yet expanded, the form and size of the right portion of the apparatus are in almost exactly the same condition as is the left in the adult. IV. Tue BuccaLt Nervous SYSTEM. Lankester* says, in speaking of Nautilus:—“No buccal nervous system has been observed in Nawutilus;”’ and again, “nor has an enteric nervous system been described in this animal.” In regard to both these statements, Professor Lan- kester seems to be in error, as a complicated buccal system was described and figured by H. v. Jhering’, while at least part of an enteric system was described by Keferstein’. In regard to the buccal nervous system it seems advisable to give a short Zoological Articles, p. 142. Vergl. Anat. des Nervensystems der Mollusken, p. 263. 1 3 Bronn’s Thierreich, Malacozoa, p. 1373. ON THE ANATOMY OF NAUTILUS POMPILIUS. 195 account, however, as von Jhering’s remarks are very brief, while in the construction of his diagram a curious blunder seems to have been made, which has been perpetuated by its being copied by leading text-books. In effect, what he figures as the cerebro-pharyngeal connective is really the forward prolongation of the pharyngeal ganglion, which, uniting with its fellow in the middle line, forms the anterior infra-buccal commissure. On the other hand, the two nerves figured as emerging from the pharyngeal ganglia laterally are the cerebro-pharyngeal con- nectives, of which there are not one, but two, on each side. In fact, by rotating the portion of his figure representing the buccal nervous apparatus through 180° about an axis passing through the pharyngeal ganglia and joining up at the cut ends as indicated above, one gets quite an accurate diagram. No doubt v. Jhering’s slip arose through dissecting and figuring this part of the nervous system after the buccal mass had been removed from its connection with the animal. The cerebro-pharyngeal connectives are two thick nerve- trunks on each side, taking their origin from the adoral border of the supra-cesophageal nervous mass. Enclosed in a dense sheath of connective tissue, they pass to the sides of the buccal mass. For the first part of their extent their course is highly sinuous, a character probably to be connected with the pro- trusibility of the buccal mass and the consequent very variable distance between it and the fixed circum-csophageal parts of the nervous system. Pursuing a slightly convergent course the two connectives reach the lateral aspect of the buccal mass, and there unite in the pharyngeal ganglion (fig. 4, ph. g.)—a trian- gular structure with its apex directed towards the mouth—and another of its angles external. It les on the muscles of the buccal mass immediately beneath the skin. The two pharyngeal ganglia are connected to one another by a longer anterior and a shorter posterior commissure, on the course of the latter being the slightly swollen “ buccal ganglia.” The anterior commissure (fig. 4, ant. com.), figured by v. Jhering as cerebro-pharyngeal connective, is a flattened band nearly 1 millimetre in breadth, and pursuing a f-shaped course 196 --.-- J, GRAHAM KERR. immediately beneath the skin, and just within and parallel to the margin of the lower mandible. As the ganglion tapers off into the commissure, it gives off numerous small and several larger filaments to the skin-fold surrounding the mandibles. Fia. 4, ant.com c.ca. buc. phar.con Buccal nervous system of Nautilus pompilius. ph.g, pharyngeal ganglion; buc.g., buccal ganglion; c.c, cerebro-pharyngeal connective; buc.phar.con, bucco-pharyngeal connective; ph.n, pharyngeal nerves; b.com, buccal commissure; ant.com, anterior pharyngeal com- missure. From the adoral part of the commissure also some very fine strands pass off to the same structures. The internal angle of the pharyngeal ganglion is prolonged towards the middle line into the posterior commissure, which soon swells out slightly, forming the buccal ganglion (bwc.g.). ON THE ANATOMY OF NAUTILUS POMPILIUS. 197 From this pass backwards two nerves into the sponge-work of the pharyngeal wall (ph.n.). Whether they are continued back in this along the sides of the crop to become connected with the gastric ganglion, I was not able satisfactorily to determine. From the aboral end of the buccal ganglion anteriorly a twig passes to a large elevation of the buccal lining, within which it divides: up into numerous branches. At its oral end the ganglion passes into the short convex-forwards commissure which connects it with its fellow. From this two nerves pass adorally on each side, the smaller more mesially situated immediately underlying the radula, VY. THE INNERVATION OF THE “INNER INFERIOR LOBE,” Posterior (ventral) to the buccal mass, well within the hood- tentacle complex, is a flattened lobe, bearing on each side a series of tentacles, separated by a peculiar lamellated organ which has been supposed to be sensory. This lobe is called the “inner inferior lobe” by Lankester. For its innervation there is figured by Owen, and copied by Gegenbaur and others, a small distinct ganglion on each side. In the specimens which I have dissected, however, the conditions are as follows :—Upon each side, somewhat external to the root of the funnel-nerve, there arises from the anterior sub-cesophageal nerve-cord a rather thinner nerve-trunk, which passes into the basal part of the lobe mentioned. This bends towards the middle line, pursues a curved course in the substance of the lobe, and meets with its fellow of the opposite side. The two together form in fact not two separate ganglia, but a continuous cord. The median most strongly curved part of this cord gives off about 24 slender nerve-filaments, which radiate forwards to the lamelle of the lamellated organ. The more lateral parts of the cord, on the other hand, give off a stout unbranched nerve to each of the tentacles of the lobe. These nerves, coursing as they do through the fibro-muscular substance of the lobe, are very hard to trace out in their entirety. 198 J. GRAHAM KERR, VI. THE Post-ANAL PAPILLA AND NERVES. A short distance behind the anus is a peculiar flap-like structure, arising from the body-wall and bearing four papilla. It varies much in form—sometimes being divided into two distinct halves—sometimes continuous mesially—sometimes thin and membranous—sometimes tumid and swollen. It is covered by columnar epithelium, and filled with ordinary con- nective tissue, sometimes with abundant jelly-like matrix. Fie. 5. Post-anal papilla with glands in the female. a, anus; p.a.p, post-anal papilla; g, openings of glands; n.g, nidamental gland; p.c.v, pericardio-visceral aperture; neph, opening of kidney chamber. In the female, examination of this region with a hand-lens shows the existence of a large number of apertures in the outer skin. These, to the number of about 150, form a band about 0°5 mm. in width, curving gently forwards on either side of the post-anal papilla, tapering off and terminating close to the advehent vessel of the posterior gill. In section these openings are seen to be the apertures of tubular ducts which pass inwards perpendicular to the surface for some little distance ON THE ANATOMY OF NAUTILUS POMPILIUS. 199 and then break up into several blindly ending branches. These are lined by involutions of the surface epithelium, which in the neighbourhood of each aperture increases to about twice its thickness elsewhere, its celia at the same time becoming extremely long and powerful (03 mm. in length). Once within the narrow aperture the lumen of the tube expands to about 05 mm. in diameter, and the lining epithelium becomes shorter, the remainder of the lumen being lined by comparatively short columnar cells, each with a round ellipsoidal nucleus. Arising from the posterior side of the posterior sub-cesopha- geal nerve-cord, close to the middle line, are a pair of stout nerve-trunks, which pass backwards on either side of the vena cava. The greater part of these pass off to supply the gills, but a direct prolongation of each is present, which passes backwards on either side of the post-anal papilla. This nerve is largest in the female, where it gives off nerves towards the middle line, supplying the nidamental gland. In the region of the post-anal papilla branches also pass off towards the middle line. There appears to be—although the condition of the material did not allow me to quite satisfy myself on this poimt—an anastomosis of these centrally passing branches with those of the opposite side. If this be confirmed we have here a true post-anal com- missure, such as exists in Chiton: in which case we should be compelled to regard not merely the “posterior sub-cesophageal nerve-mass,” but rather the two lateral portions of this, together with the nerve-trunks which have been mentioned as passing backwards on either side of the vena cava, as forming the homologue of the pleuro-visceral nerve-cord of Chiton. The mesial part of the posterior sub-cesophageal nerve-mass would then represent a secondary fusion between the nerve-masses of the two opposite sides. VII. THE SPERMATOPHORE-RECEIVING APPARATUS. Behind (ventral to) the buccal mass and immediately under- lying the inner inferior lobe, the kind of shelf which connects the tentacular mass of one side with that of the other has its inner surface raised into a series of curious lamelle. 200 J. GRAHAM KERR. The remarkable organ formed has been referred to by Valenciennes’, and by Lankester?, as a paired structure. Not always, however, does it seem to be so, as in one of the two specimens in which I observed it the laminz were quite con- tinuous across the middle line, the laminz appearing in fact to be mere exaggerations of the fine transverse wrinkles into which the surface of the skin is thrown behind the lamellar organ. Of the two authors referred to, the first, after some hesitation, suggests that the lamellated organ may be tactile in function, “analogous to the palpi round the Crustacean mouth.” Lankester, on the other hand, for what reasons is not stated, very definitely describes the organ as “probably olfactory®.” During the examination of a mature female somewhat startling evidence was obtained as to the true function of this organ. The lamellz were here covered with a thick coagulated material apparently secreted by them, spreading over the edges of the lamelle and passing in thin plates down between them. Partially imbedded in the coagulum on the left side and only partially visible, there appeared a peculiar brown structure which at once suggested the appearance of a spermatophore. And upon carefully clearing away the surrounding material the surmise so suggested was corroborated. The long slender spermatophore lay coiled backwards and forwards over the surface of the lamellz in the manner indicated in Pl. X. fig. 2, held firmly in position by the coagulated material. We would seem to have here a peculiar cement-secreting glandular apparatus, on whose sticky surface the spermatophore is deposited by the male. In other Cephalopods the position in which the spermato- phore is attached to the female varies: in Gigopsids, Octopods, and Sepiola, e.g., it is passed into the mantle-cavity; in other Decapods (e.g., Sepia, Loligo) it is attached to the skin on the outer surface of the buccal mass. In Nautilus the position is thus a somewhat intermediate one. 1 Arch. Mus, d’Hist, Nat. tom. ii. p. 277. 2 Zoological Articles, p. 130. 3 Op. cit. fig. 88. ON THE ANATOMY OF NAUTILUS POMPILIUS. 201 VIII. Tat Morrsoitocy or THE “ ARMS” OF CEPHALOPODS. As Grobben has justly remarked, and as Pelseneer has adopted as text to his paper on the subject “eine der schwierig- sten Fragen in der morphologischen Deutung des Cephalo- podenkérpers bildet die Morphologie der Kopfarme”; and in accordance with this, as well as with its far-reaching interest, the question has attracted from time to time a great amount of attention from morphologists. Regarding the fundamental nature of these organs, two very different views have been brought forward :— (1) That the arms of the Cephalopods are processes of the head or circumoral region. (2) That they are processes of the foot, part of which has grown up on either side so as to finally surround and almost completely hide from view the head itself. To enter in detail into the differences as to minor points in the tenets of the various upholders of these two views seems unnecessary, aS this has already been done by others’; and further, because it is proposed to consider the problem here in its most general aspect—as to whether the Cephalopod arms are cephalic or pedal. It may be advisable, in the first place, to inquire whether there is anything in the general relations of the parts to support or even suggest the second of these views. In ordinary Cuttle- fishes it is pretty obvious that there is nothing of the kind—the arms form a continuous circle round the buccal mass—one would naturally suppose they belong to the head. It is therefore important to glance at Nautilus, where, as Lankester has well accentuated, “any divergence from the condition obtaining in other forms has possibly, and even probably, a special signifi- cance,” and “is not readily to be dismissed as an ‘adaptation’ peculiar to that form.” 1 Cf. especially Pelseneer’s admirable summary, ‘Challenger’ Report, Pteropoda. 2 Quart, Journ. Micr. Science, vol. xxiii. p. 348. 202 J. GRAHAM KERR. In Nautilus the arrangement of the circumoral lobes and tentacles has been described by Bourne and by Lankester, so that it is unnecessary to go into details. Anteriorly (dorsally) is the large fibrous mass of tissue which forms the hood. Laterally, on each side, is an aggregation of tentacles. Anteriorly (dorsally) the mass of tentacle-sheaths is directly continuous with the hood. On slicing away the substance of the hood carefully, it is seen that the bases of all the outer tentacles are embedded in it. The appearance of tentacle-sheaths is due merely to the more or less distinct marking off by superficial grooves of the parts of the mass surrounding each tentacle. Hood and tentacle-sheaths together form a perfectly continuous mass lying anterior (dorsal) to the buccal mass and curving backwards (downwards) on either side of it in saddle-like fashion. In the male this is very obvious, the two limbs of the mass being connected together posteriorly merely by a thin shelf. In the female, however, this bears on its inner side the “inferior inner lobe,’ which bears on each side a group of tentacles and whose appearance suggests a bilateral origin. The main impression given by the tentacle-hood complex is that of a saddle-shaped structure, situated anterior (dorsal) to the buccal mass—its limbs passing backwards on either side of the latter. The anterior (dorsal) part of the complex here predominates : it is developed less equally all round the buccal mass than in Decapods ; its preponderating part is dorsal. The next point of interest in the gross anatomical relations of the parts lies in the funnel which, according to the upholders of the “pedal” view, is primitively continuous with the tentacle- hood mass. The Funnel—This is a large tongue-shaped structure attached to the posterior face of the body—to the roof of the mantle-cavity, into which it imperceptibly passes aborally. At its oral end it projects parallel to the axis of the buccal mass— quite free and separated by a deep groove from the hood and tentacle-mass. Tongue-like in form, its margins are inrolled about a longitudinal axis, so that one comes to overlap the other. Which does so appears to be quite inconstant in dif- ON THE ANATOMY OF NAUTILUS POMPILIUS. 203 ferent individuals, and in any one individual the right and left margins present exactly the same appearance; there being nothing to point to one in particular being kept habitually folded over the other. From this, and from the general muscular character of the funnel, I have little doubt that the living animal possesses the power of unrolling and flattening it out, possibly even of using its broad lower face to creep on or adhere to rocks. In spirit-specimens one can readily so unroll the funnel, and when this is done the appearance of the animal is very striking, as is shown in PI. IX, fig. 1, where, by the way, the mantle-flap has been partially removed so as to afford a better view of the creature. One is here impressed, first of all, by the sharp way in which the funnel is marked off from the hood-tentacle-head mass. Everywhere a deep groove separates them’. There is nothing here to suggest or even support the view that part of the foot has grown up round and become fused with the head. Again, the great size of the organ is very impressive—more especially its width from side to side,—and its entire condition is such as at once, to my mind irresistibly, to suggest that in this organ one has the representative of the whole of the foot of the ordinary Gasteropod. The general relations of the parts in Nautilus impress upon one that :— (1) The hood-tentacle complex is preponderatingly an- terior (dorsal) to the buccal mass, its posterior (ventral) parts being relatively insignificant. (2) The hood-tentacle complex is most sharply marked off from the funnel by a deep groove. (3) The funnel is enough, in itself, to represent the whole of the Gasteropod foot. Considering merely them alone, there is no suggestion of doubt that the hood-tentacle complex is cephalic; that the funnel is the Gasteropod foot. 1 In this connection the figure given by Lankester (Zoological Articles, fig. 91) seems scarcely in accord with the actual conditions as shewn in my specimens. 204 J. GRAHAM KERR. It is because, at the present time, after many years of con- troversy, the contrary view, which for shortness may be referred to as the ‘pedal’ view, has gained the ascendency and has come to be the one enunciated by the most authoritative text-books’ that the present discussion seems necessary. When Lankester published his Encyclopedia Britannica article on Mollusca, he pointed out that the view taught by Leuckart, Huxley, and himself, that the Cephalopod arms are pedal in their nature, was based upon three different sets of evidence—to wit, those derived from (1) Their ontogenetic development ; (2) Their innervation ; (3) Their homology with the sucker-bearing processes of the larval Pneumoderma. Of these (3) derived its force from the supposed pedal nature of the sucker-bearing appendages. However, it has now been satisfactorily shown* that they are purely cephalic, and there- fore this argument, if it be argument at all, tells precisely in the opposite direction. At present, therefore, the view that the Cephalopod arms are parts of the foot rests upon (1) and (2). In regard to (1), however, although it must be admitted that the facts of embryology do tend to bear up the view that the crown of arms is formed by an upgrowth from each side of the foot, it must be borne in mind how extremely unreliable any evidence, as to topographical relations, must be which is based on the phenomena exhibited in the development of enormously yolk-laden eggs. Therefore it appears that the only one of the three classes of evidence adduced above which can be considered of real weight, is that resting upon the innervation of the parts under consideration, and that this opinion is shared by other workers, is shown by its tendency in more recent writings to supplant the evidence derived from embryology. It appears, therefore, not inadvisable to submit this portion of the evidence to a short critical examination, to 1 Lang’s Lehrbuch, pp. 587, and Korschelt and Heider, p. 1176. 2 ‘Challenger’ Reports: Pteropoda, Anatomy, p. 39. ON THE ANATOMY OF NAUTILUS POMPILIUS. 205 endeavour to ascertain whether it is equal to bearing the strain of acting as main support to a view which we have seen to be inherently improbable, on the evidence afforded by gross anatomical relations. And as a preliminary it may be well to look into the general ideas now held and taught by zoologists as to the general character of the Cephalopod central nervous system. In the latest text-book of Zoology (Lang, p. 722) one reads, “Das symmetrische Nervensystem aller Cephalopoden zeichnet sich durch die sehr starke Concentration der typischen Mollus- kenganglien, auch derjenigen der Visceralconnective, aus';” and this I think I may venture to say fairly represents the views held and taught by zoologists generally: that the Cepha- lopod central nervous system consists typically of three pairs of ganglia aggregated round the cesophagus, which ganglia are homologous with the three similar pairs of, say, a Gasteropod. That a certain rough resemblance does exist between the arrangement of the ganglia round the cesophagus of a Di- branchiate Cephalopod and that met with in many Gasteropods ‘may be at once admitted ; but when it comes to be a question of precisely homologizing the individual ganglia in the one case with those in the other, one has to do with a very different matter. Supposing, for a moment, the homology to hold, then one ought to find the resemblance most marked in those Cephalopods which phylogenetically most nearly approach the common ancestral forms of Gasteropods. But what are the actual anatomical facts?—that in the Nautilus, the most primitive and oldest Cephalopod now existent, such division into three pairs of ganglia is completely absent. And then one might turn to that Gasteropod (I here use the term in its wide sense) which other evidence points to as having similarly to the greatest extent retained such common ancestral conditions—to wit, Chiton. And here again one finds a complete absence of segregation of the central nervous system into its three pairs of ganglia, and in its stead a central nervous system showing in 1 The italics are mine. VOL. VI. 18 206 J. GRAHAM KERR. many respects a strong and fundamental resemblance to that of Nautilus. The facts of Anatomy, then, are strongly opposed to any rough-and-ready homologizing of the various ganglia of the higher Cephalopod with those of the higher Gasteropod. One might go so far as to say that they demonstrate their non-homo- logy. The common ancestor of Gasteropods and Cephalopods, so far as we can see, possessed, as did and do so many other primitive forms, a nervous system consisting of thick strands ensheathed in a continuous layer of nerve-cells; and any depar- ture from this condition, in the direction of collecting and centralizing these nerve-cells into ganglia to fulfil local require- ments, 1s a process which has taken place independently within each of the two stems of descent. It follows, from this independence in phylogenetic develop- ment of these secondarily formed ganglia, that we are not justified in taking any one of the ganglia of the higher Cepha- lopods and saying this is the “pedal” ganglion (implying in the term “pedal” accurate homology with the so-named ganglia of Gasteropoda)—d, fortiori, in asserting here is an organ inner- vated by the pedal ganglion, therefore it is morphologically part of the foot. Yet it is precisely this latter line of argument which modern exponents of the “pedal” hypothesis use as theirmainstay. The central nervous system of Cephalopoda may be said, according to what we know of Nautilus, to consist primarily of— (1) A supra-cesophageal mass, connected with (2) An anterior sub-cesophageal, and (3) A posterior sub-cesophageal mass. To these is added in the Dibranchiata a separate nervous mass lying in front of (2)—the branchial ganglion ; and it is this which innervates the arms. To quote Pelseneer (Chall. Rept. p. 65):— “Regarding (1) there is no disagreement as to its nature, all recognising in it the fused cerebral ganglia. (2) “Has been universally regarded as constituted by the pedal ganglia. (3) “Corresponds to the combined visceral ganglia of other Mollusca. ON THE ANATOMY OF NAUTILUS POMPILIUS. 207 “All observers are agreed as to the interpretation of the supra-cesophageal and the two posterior sub-cesopha- geal masses (ve. (2) and (3)). The disagreement re- lates only to the branchial ganglia, which are regarded by one party as pedal and by the other as cerebral.” Pelseneer then goes on to combat the view that the bran- chial ganglion has been derived from the fusion of a downgrowth on each side of the cerebral ganglion. While protesting, in passing, against the statement that the supra-cesophageal nerve-mass is formed of “the fused cerebral ganglia,” when in reality it represents the primitive nerve-mass out of which “cerebral ganglia” have not yet become segregated, it is (2) the statements as to the “ pedal” and branchial ganglia which concern most closely the point under discussion. The one fact of independence of evolution is enough to show that the so-called pedal ganglion of Cephalopods—z.e. the anterior sub-cesophageal nerve-mass of Nautilus, which in the higher Cephalopods has, in accordance with a very general law, become condensed into a definite ganglion, supplying the various organs originally in its neighbourhood—is not in the strict morpho- logical sense the “pedal” ganglion at all. One may then accept with Pelseneer the development of the branchial ganglion by splitting off from this anterior sub-cesophageal nerve-mass, and yet be as completely without evidence as we were before that the structures supplied by it have anything whatever to do with the foot. In brief it appears to me that :—the general relations of the parts point undoubtingly to the arms of Cephalopods being processes of the head-region—that all the special evidence brought forward to support the pedal view is either erroneous, of little weight, or is permeated with fallacy—and that it there- fore behoves us in the meantime to unhesitatingly accept the first mentioned! 1 The forerunner of the hood-tentacle complex of Nautilus (and consequently of the arms of the Dibranchiata) we may probably see still persisting in the similarly innervated and highly sensitive mass which surrounds the mouth in Chiton, 18—2 208 J. GRAHAM KERR. IX. THE PHYLOGENETIC RELATIONSHIPS OF THE CEPHALOPODA. From its archaic character Nautilus might be expected to give valuable hints as to the phylogenetic relationships of the group to which it belongs. Upon the whole it appears to me that its structure affords strong evidence that the nearest living allies of the Cephalopoda are to be found in the Amphineura. And it is interesting to note that amongst these it is the Chitons in which the points of resemblance are most striking, as they are apparently the oldest and most primitive members of the group. The number of really important morphological features in which the Chitons resemble Nautilus is quite remarkable, e.9.— (1) Its bilateral symmetry. (2) The general characters of its nervous system. (3) Its possession of paired metamerically arranged ctenidia, of which in some species, believed to be phylogenetically younger, there is a tendency for those at the anterior end of the body to disappear—only those towards the posterior end persisting (mero- branchiate forms). (4) The traces of metamerism exhibited by the heart in some forms, there existing several pairs (four in Chiton magnificus) of auriculo-ventricular openings. (5) General relations of ccelom, nephridia, &c. (6) Eggs developed within follicles. In regard to (5), fig. 6 indicates diagrammatically the relationships of the parts concerned. In the case of Chiton (B) two coelomic chambers are shown, one lying in front of the other—the genital coelom and the pericardium. The pericar- dium communicates with the exterior by a pair of functional nephridia; the genital ccelom by the pair of genital ducts which from their relations can hardly be otherwise than morphologically a pair of nephridia too. In Chetoderma (A), a less primitive animal, a less primitive arrangement has ON THE ANATOMY OF NAUTILUS POMPILIUS. 209 been developed: the genital division of the ecelom has developed a communication with the pericardium through which the genital products pass—and it has lost its original genital ducts. Fig. C shows the condition in Nautilus, where again the same two ccelomic chambers are visible. Here also a communication Fig. 6. C Diagram showing the relationships of the ceelom and nephridia in Amphineura and Cephalopoda, “ A, Chetoderma; B, Chiton; C, Nautilus; D, Sepia. gc, genital division of the celom; pc, pericardiac division of the celom; gn, nephridia of genital segment; pn, nephridia of pericardiac segment ; ks, kidney-sacs in Nautilus; in Sepia the two posterior kidney-sacs are seen still in communication with the rest of the nephridia. has become formed between the two, but the two pairs of ducts to the exterior still persist—the anterior nephridium here still preserving its excretory portion—a more primitive condition than in Chiton, and probably to be correlated with the fact of its having become shut off from the main lumen of the duct. A few irregular apertures in the wall separating the two ecelomic 210 J. GRAHAM KERR. chambers point towards the still later condition to be met with in Sepia (D), where the septum has disappeared—a faint rudi- ment remaining in the form of a transverse fold rising up from the floor of the common chamber’. X. SumMMARY OF CONCLUSIONS. 1. The perivisceral cavity in Nautilus is remarkable for the almost equal participation in its formation of both ccelom and heemoccel. 2. The ccelom consists of two distinct chambers—genital and pericardial—separated by a perforated septum. 3. Each of these coelomic chambers opens to the exterior by a pair of nephridia. 4. The genital ducts of the Cephalopoda represent portions of nephridia. 5. The ovary is remarkable for its extremely archaic character—an ovigerous region of the ccelomic epithelium, roofed in by a simple upgrowth of the ccelomic wall. 6. The ova arise from syncytial masses of protoplasm. 7. The testis is also archaic in character, and similar to the ovary in its main features. Its cavity, however, has become subdivided into numerous delicate tubes for the provision of increased area of the spermatogenetic epithelium. 8. The penis is a paired structure, its left moiety, however, remaining rudimentary. 9. An elaborate buccal nervous system is present. 10. The “inner inferior lobe” is innervated not by a pair of distinct ganglia, but by a continuous nerve-cord. 11. Round the base of the postanal papilla is a curious system of skin-glands. 1 The view advocated by Grobben (Morph. Stud. p. 39) that the condition in Sepia is the more primitive, and that it represents a stage in the evolution of the condition met with in the other Mollusca, seems to me untenable, ON THE ANATOMY OF NAUTILUS POMPILIUS. 211 12. A prolongation backwards of the nerve-trunk which supplies the gills probably represents the postanal commissure of Amphineura. 13. A laminated organ lying below the mouth has a func- tion in connection with copulation—the spermatophore of the male becoming attached to it. 14. The evidence as to the “pedal” nature of the Cephalo- pod arms appears to rest on insecure foundations, and it seems desirable to abandon it for the inherently much more probable view that these structures are processes of the head region. 15. Nautilus shows many strong resemblances to the Amphineura, and it is probably amongst these latter that we have to look for the nearest allies of the Cephalopoda. EXPLANATION OF THE PLATES. PuatEe IX. Fig. 1. Side view of an animal of Vautilus pompilius, extracted from the shell, The funnel has been opened out and the mantle-flap partly cut away so as to give a better view of the various parts. h, hood ; ¢, tentacles; e, eye; 7, funnel separated by a deep groove from the hood-tentacle mass; m, cut edge of mantle-flap; g, gill; 8, siphuncle. Fig. 2. Longitudinal section through the animal of Vautilus very slightly to the right of the middle line. b, buccal cavity; 7, radula; cg, supra-cesophageal nerve-cord ; pl.g, posterior subcsophageal nerve-cord; .g, anterior ditto; f, funnel with its valve; ph, crop; it, intestine; an, anus; &, kidney-chamber with follicular appendages of advehent vein pro- jecting into it; p.foll, pericardial gland-follicles projecting into pericardium ; v.c, vena cava. 212 Fig. 1. Fig. 2. Fig. 3. J. GRAHAM KERR. PuLate X. View of penis and sac of Needham from posterior (ventral) aspect. The outer wall has been removed so as to show a and 6—the right and left halves of the penis. c, sac of Needham; d, corresponding structure of left side; e, bristle passing through opening of vas deferens into sac of Needham. Spermatophore-receiving apparatus of an adult female with sper- matophore (s) zm situ. In this specimen the laminz of the organ were continuous across the middle line. The same organ in its more usual (paired) form. tudies M L.Vol VI, Pl. IX (a! 1B) Jambridae lson.C dwin Wil h B S) J Tt Le POMPILI ayaa, ae (AUTILUS FN ATOMY A Studies ML.VolVI,Pl. X 2 Edwin Wilson.Cambridgé ANATOMY OF NAUTILUS POMPILIUS FURTHER REMARKS ON THE CELL-THEORY. 2s Further Remarks on the Cell-theory, with a Reply to Mr. Bourne. By Adam Sedgwick, F.R.S. In a paper published last autumn (this Journal, vol. 37), I called attention to the apparent inadequacy of the cell- theory. Recently a criticism upon my article has appeared from the pen of Mr. G.C. Bourne, to which I may be allowed to devote a few words. But before replying to Mr. Bourne, I should like to state my position with regard to the theory a little more fully than I have hitherto done. In my previous communication I used the word “inadequacy”’ because it seemed to me to express, as nearly as possible, my own views with regard to the theory. A theory to be of any value must ex- plain the whole body of facts with which it deals. If it falls short of this, it must be held to be insufficient or inadequate ; and when at the same time it is so masterful as to compel men to look at nature through its eyes, and to twist stubborn and -uncomformable facts into accord with its dogmas, then it becomes an instrument of mischief, and deserves condemna- tion, if only of the mild kind implied by the term inadequate. The assertion that organisms present a constitution which may be described as cellular is nota theory at at all; it is— having first agreed as to the meaning and use of the word cell —a statement of fact, and no more a theory than is the asser- tion that sunlight is composed of all the colours of the spec- trum. The theory comes in when we try to account for the cellular constitution of organisms; and it is this theoretical part of the cell-theory, and the point of view it makes many vou. 6. A 214 ADAM SEDGWICK. of us assume, that I condemn. It is not the word “ cell” which I am at issue with, for structures most conveniently called cells undoubtedly exist, as the ovum, spermatozoon, lymph- cells, &c.; and I fully agree that the phenomenon called cell- formation is very general in organic life. But at the same time I hold with Sachs and many others that it is not of primary significance, but ‘ merely one of the numerous ex- pressions of the formative forces which reside in all matter.” No one who has studied animal tissues could for one moment deny that nuclei have in many cases a relation to the surrounding protoplasm, a relation which is expressed in the arrangement and structure of that protoplasm. They have not always this relation, but it is usually present, and the question is, how are we to interpret it? That we cannot interpret it finally until we know the relative values of nucleus and extra-nuclear protoplasm, and the functional re- lation between the two, is clear; but we may form and hold provisional theories. The hypothesis or idea which holds the field at the present day is the cell-theory in its modern form. This theory, recognising the cellular structure (while not ad- miring the phrase, I must use it for want of a better one) asserts that organisms of Metazoa are aggregations or colonies of individuals called cells, and derived from a single primitive individual—the ovum— by successive cell-divisions ; that the meaning of this mode of origin is given by the evolution theory, which allows us to suppose that the ancestor of all Metazoa was a unicellular Protozoon, and that the develop- ment of the higher animals is a recapitulation of the develop- ment of the race. Thus the holoblastic cleavage of the ovum represents the process by which the ancestral Protozoon be- came multicellular, and the differentiation of the cells into groups the beginning of cellular differentiation. According to this view the order is: unicellular stage—multicellular stage—differentiation of cells into tissue elements; cellular structure preceded cell-differentiation, and to get tissues you must first have cells. And ten years ago it was commonly held that these cells were primitively separate from one another, FURTHER REMARKS ON THE CELL-THEORY. 215 and that the connections found between them in the fully formed tissues were secondary. You had your neuro-epithelial cell, and your musculo-epithelial cell, each derived from a distinct cell produced by division of the ovum ; and the question was, how do they find each other and become connected ?! Further, in studying the development of a tissue you had to find a group of cells, each of which became modified into one tissue element. Thus the primitive streak was a proliferating mass of cells which eventually gave rise to a number of meso- dermal tissues ; the nerve-crest similarly was a mass of cells which gave rise to nervous tissues; a nerve-fibre was one of these cells elongated, and before you would get your nerve-cell and fibre you must have your nerve-crest cell produced by division from the cells of the nerve-cord, and subsequently sending out a process which elongated and travelled to the periphery as a nerve-fibre. My work on Peripatus first led me to doubt the validity of this view of the origin of the Metazoon body. In the first place I found that in some forms there is no complete division of the ovum, and on examining the facts I discovered that such forms were more numerous than had been supposed. It therefore appeared that in some Metazoa the ovum divided into completely separate cells, while in others it did not so divide. The question then arose, which of these methods is primitive ? and the answer naturally was, the complete division, because this fitted in with our ideas as to the supposed evolu- tion of the Metazoa from a colonial Protozoon. But on reflection this difficulty arose: the individuals of colonial Protozoa are in protoplasmic connection, while the cells of the completely segmenting ova are separate; so that the parallel between the ontogeny and the phylogeny breaks down in an important particular. To get over this difficulty it was necessary to suppose that the isolation of the segments of incompletely segmenting ova was apparent and not real, that 1 For exposition of this view vide Flemming, ‘ Zell-Substanz, Kern u. Zell-Theilung,’ Leipzig, 1882,p. 74, and Balfour’s Address to the Depart- ment of Anatomy and Physiology at the British Association in 1880. 216 ADAM SEDGWIOK. they were really connected by protoplasmic strands which had escaped observation. But, on the other hand, there was the possibility that the completely segmenting ova were secondary acquisitions of ontogeny, and that the development in such forms as Peripatus, Alcyonaria, &c., was more primitive, and that the passage from a Protozoon to a Metazoon had taken place by way of a form more resembling a multinucleated cilated Infusorian than Volvox. In other words, that the differentiation of the Metazoa had been effected in a continuous multinucleated plasmatic mass, and that the cellular structure had arisen by the special arrangement of the nuclei in reference to the structural changes. This was the stage to which my researches on Peripatus led me. Since then I have paid attention to Vertebrata, and I have found that a number of embryonic processes have been wrongly described, amongst them such important matters as the development of nerves and the origin of the mesoderm ; and I thought that I traced the errors referred to to the dominating influence of the cell-theory in its modern form, for the facts seemed so obvious in them- selves that it would have been impossible to make any mistake about them had they been examined without the prejudice imparted by a preconceived theory. A theory which led to such obvious errors must, I thought, be wrong, and I denounced it. But my denunciation in no way implies that I fail to recognise the so-called cellular structure of organisms or their origin from the one-celled ovum. On the contrary, I was led toa re- consideration of the question, what is the meaning of the pre- dominance of the structure called cellular, which is characterised by a definite relation of the nuclei to the functional tissues, and of the fact that the organism so often passes through a unicellular stage. With regard to the former I must say that I have arrived at no conclusions which enable me to formulate to myself any satisfactory hypothesis, and, as I stated at the outset, I do not think it is possible to do this until we acquire some more understanding of the relative function of nuclei and protoplasm. But with regard to the latter there are some facts which might FURTHER REMARKS ON THE CELL-THEORY. 217 well be considered. In the first place, the unicellular origin is only found in sexual reproduction, not in asexual. The characteristic of the unicellular form is its simplicity of struc- ture, and the essential feature of sexual reproduction is the conjugation of the reproductive cells. Now in the Protozoa, in which the amount of formed tissue is generally slight and the structure of the body simple, conjugation can and does often take place between the ordinary form of the species. But in the Metazoa, in which conjugation is as necessary a phenomenon in the specific cycle as in Protozoa, conjugation is impossible between adult or ordinary individuals of a species from mechanical causes. How is this difficulty got over in nature? My answer is, by the formation of special individuals of extremely simple structure—a structure so simple that conju- gation between them is possible. To put the matter in another way, I should regard the ordinary dicecious Metazoon as a tetra- morphic species, consisting of male, female, ovum, and sper- matozoon, the two latter being individuals which are specially produced to enable conjugation to take place. Mr. Bourne, in his criticism, begins by complaining that he cannot ascertain from my article my own views on the subject of the cell-theory. Why should he expect or wish to discover them? My remarks were simply directed to show the short- comings of the theory with regard to certain anatomical facts. As explained above, my own view is that the ceil-theory is inadequate to explain the facts, and that it is not possible at present to explain them by any theory. He proceeds to state that I am abusive because I say that certain observers “ are constrained by this theory with which their minds are saturated, not only to see things which do not exist, but also to figure them” (I am referring to embryonic mesoderm of verte- brates). He calls this abuse, not argument. I venture to differ with him—it is neither abuse nor argument ; it is merely a statement of fact (unless, indeed, it be considered abusive to say that a man accepts and believes in the cell-theory). If you disbelieve it, consult the memoirs of the last twenty years in which this tissue is referred to, and in most of them ycu 218 ADAM SEDGWICK. will find the mesenchyme described or figured as consisting of branched, isolated cells. Mr. Bourne then refers to certain remarkable researches which emphasize the distinction and complete isolation of the cells formed in the segmentation of the egg; with what object is not apparent, for he proceeds on the next page to condemn those who hold that the arganism is constituted of independent and isolated units. He even maintains that no reputable bio- logist holds such a view. However that may be, I do not think that his quotation from Haeckel in support of his con- tention is a happy one, for it is perfectly clear from the quota- tion that Haeckel, who indeed goes so far as to call the units individuals, holds the view which Mr. Bourne condemns. Haeckel even calls them individuals of the first order, and says that in the adults they frequently unite to form colonies ; and he particularly implies that the loss of independence caused by their colonial union is secondary. Mr. Bourne has completely failed to grasp Haeckel’s meaning, else how can he write as he does on the same page with the quotation from Haeckel—* So that, as a matter of history, while plants used to be considered to be colonies of independent life units, animals were not.” The most remarkable part of Mr. Bourne’s criticism is that in which, after strongly animadverting on my statement that it is difficult if not impossible to enunciate the cell-theory in a manner satisfactory to every one,—indeed he quotes from Schwann and Hertwig to show how precisely it can be stated,—he proceeds to devote a dozen or more pages of his paper to a consideration of the various views which are held and which may be held as to what a cell really is! If this amount of discussion is required to arrive at the meaning of the word cell, is it likely that there will be simple agreement as to the theory which is supposed to explain and account for the so-called cellular constitution of organisms ? Again he says, referring to my description of the embryonic mesoderm as a protoplasmic reticulum with nuclei at their FURTHER REMARKS ON THE CELL-THEORY. 219 nodes: ‘ Does he accept the logical consequences of this, and say of the epithelial cells of the salamander or of unstriped muscle fibres, that they are protoplasmic reticula with nuclei at their nodes ?”’ Now, with all due respect to Mr. Bourne’s logical faculties, may I ask him where logic comes in here? If I describe London as a network of streets, with public-houses at many of the street corners, am I obliged by logic to give the same description of the Gog-Magog Hills? However, on the next page Mr. Bourne makes up for all the hard strictures he has passed upon me; for he says that, after all, reflection may induce us to abandon the cell-republic or colonial theory ; thus he admits a very important part of my contention, for the assertion that organisms present a consti- tution which may be described as cellular is not a theory at all, it is a statement of fact (having agreed to the use of the word cellular). The theory comes in when we try and account for the cellular constitution of organisms; and it is this theoretical part of the cell-theory which I condemn, and which Mr. Bourne after a great effort agrees with me in condemning. At the same time it is possible that we might still disagree as to the meaning of the word cellular. May I call attention to Mr. Bourne’s remarkable faith in the rapidity of evolutionary changes? He says (page 169) that Schwann’s assertion that “the elementary parts of all tissues are formed of cells, &c.,” is even more true to-day than when it was written. Also I should like to know how he reconciles the implication at the top of page 170, that “ specialisation is not possible in continuous tracts of proto- plasm,” with the statement a few lines further on, that “in the Protozoa there is differentiation within the limits of a single corpuscle.” The criticism on page 172 as to my use of the word empty is not quite fair. On reference to the context it will be seen that the word empty clearly means “empty of structural elements.” te | ~ i vei 4b i — od . , , >, - es bs bat Vie F 7 pee v i . hie ‘ Fs < Ss * . 1 ty ey a ae / ¥ - ey fe t pF e . il vy au f - + WY q i li y nan a anil = ell a ed am i a. ‘ 4 iF oa elec — i - fi mF Fi % a4 Tei peek i’ va yr Lae Hind oe BF ua et ( mee E ena! F hag a "| i = Sal . i i ; ? % i ; - i . ; ’ ue , ye, f ” ! yea) n f i 7 i = = veins i 7 : ! = if fi ty =) wv wy! I . ANN io" 2 i ae) Senet ae ; ft ie @ Piette a ae Pear were aes gm a a te fe i a! ; ae < bois ite Fe. ones Phe niyo ( =H “ t Ey f Es ye i ‘ o 6 4 : cA 7 a Se Te — : ~~ = A sf : = po = ~ “ ~ » } ‘ . a A o ‘ i ' uj i *| ‘ Sad a 2! , Ms Vise ga 2 a ee = 4 i 1's Nm ir 2 ; % fs “s f = a ih ir = =. “a le Pw 4h . P pica: f j Os onth i alia ~ 7 i 7 ; bj rege ay ne es y 7a 2 x : ' (Pe, ] we Gy! aly , = + 7 = i * 4 \ ‘7 : q ion, = ¥ Fy aa a Ware as 4 mus = " li . -" : ae o 4 iy ¢ ahi 4 hing! we >: ay pena i ¥ ri uw! ‘ i wr < THE DEVELOPMENT O§ ASTERINA GIBBOSA. 221 The Development of Asterina gibbosa.! By E. W. MacBride, Fellow of St. John’s College ; Demonstrator in Animal Morphology in the University of Cambridge. With Plates XI—XXII. THE investigations which form the subject of the present memoir were commenced with the object of seeking in Asterids the results which the author (14) had already obtained from the study of Ophiurids, viz, the development of the so-called heart and its accompanying sinuses. A study of the literature soon led to the conclusion that our knowledge of the development of most organs in the Asterid body was very defective, and that a thorough revision of the whole embryonic and larval history would be most desirable. This work has occupied my attention for the last two years, and I am now in a position to give a fairly complete account of the whole organogeny ; an account which will, I hope, place our knowledge of Asterid development on the same level as that to which our acquaintance with Crinoid ontogeny has been raised by the researches of Bury (1) and Seeliger (18) ; I have to express my warm thanks to Mr. Sedgwick not only for the suggestion of Asterina gibbosa as a proper type to investi- gate, but also for much assistance and advice in revising the proofs of this paper. That there was an immense lacuna in our knowledge to be 1 A preliminary account of the observations recorded in this paper was the subject of the successful essay in the competition for the Walsingham Medal of the University of Cambridge in 1893. VOM B 222 E. W. MACBRIDE. filled up will become evident when I state in the first place, that my researches have made it clear that the Crinoids are only very distantly related to the other classes of Echinoderms, and secondly, that our previous knowledge of the metamorphosis of Asterids and their allies was confined principally to the changes which take place in their external form. ) It will be most convenient, I think, to give first a general account of the development, and then to point out how far the results of other workers have been confirmed, as by this means needless repetition will be avoided. Methods adopted. My material consisted of a large number of larve of all stages including those which had just completed the metamor- phosis, and of a considerable number of young adults varying from an age of about three weeks to several months from the metamorphosis. Of these the former, with the exception of two small collections made by myself in Plymouth, 1893, and Jersey, 1894, were collected for me and preserved according to my directions by the authorities of the Naples Zoological Station ; the latter were obtained for me and preserved by myself during my stay in the Naples Station in 1892. I have to express my deep sense of my indebtedness to Prof. Dohrn for his kindness in meeting my wishes, and to Cay. Salv. Lo Bianco for the extreme care and attention with which he carried out my directions. All the stages were preserved in osmic acid, followed by 14—24 hours in Miiller’s fluid, as this method had yielded me the best results in the case of Ophiurids. It makes the speci- mens exceedingly brittle, but at the same time gives the most excellent preservation of the minute histology; preserved in this manner the various tissues are differentiated as to their staining capacities, so that the sections look almost like coloured diagrams. On account of their brittleness, and in order to avoid shrinkage in the tissues, the larvae were embedded in celloidin, and the celloidin block subsequently embedded in paraffin. THE DEVELOPMENT OF ASTERINA GIBBOSA. 223 They were then cut into series of sections in most cases 44 uw thick—in the case of the adults 7 mw; these sections were mounted on hot water on the slide to flatten them, and stained in either Grenacher’s hematoxylin or Mayer’s carmalum. Two points of interest in connection with this process may be mentioned : first, I found that when the slide was transferred from turpentine to absolute alcohol some of the sections were sure to be lost, but that this could be avoided by placing the slide for a minute or so, after taking it out of turpentine, into oil of cloves, and thence into 90 per cent. alcohol; second, that the readiness with which sections, especially when overcharged with osmic acid, will take up either hematoxylin or carmalum is greatly increased by immersing them for twenty-four hours in borax-carmine, though they do not acquire a particle of stain from it. In the youngest stages the osmic acid produces too great impenetrability for either celloidin or paraffin, and accordingly my best results were obtained from some specimens preserved for me by Sig. Lo Bianco in a mixture of three parts concen- trated aqueous solution of corrosive sublimate, and one part glacial acetic acid. This method also gives most excellent preservation, though without that fine differentiation of the tissues yielded by osmic acid and Miiller’s fluid; as during the stages in question however the larve consist almost exclu- sively of epithelial cells, this is not a matter of any importance. This second method was recommended to me by Dr. Hisig. The orientation of the specimens was one of the chief difficulties to be overcome. I found that the best results were given by horizontal sections perpendicular to the median sagittal plane of the larva, and sections parallel to the dise and perpendicular to the median axis of symmetry in the just metamorphosed star-fish. The planes, to which in these two cases the sections are cut parallel, viz. a median horizontal plane in the larva and the plane of the disc in the adult, make an angle of about 70° with each other ; and hence it is difficult to correlate sections cut parallel to the one with those cut parallel to the other. I shall call these planes the “larval” 224, BH. W. MACBRIDE. and “adult” planes respectively. A rudiment of the preoral lobe of the larva is retained, as we shall see, until the close of the metamorphosis, and by means of it I found it possible to determine the direction of the “larval” plane up till the adult form has been almost attained. Hence, by cutting sections parallel to the larval plane, one can follow the internal changes of the metamorphosis step by step; then when the metamorphosis is complete it is possible to correlate with less difficulty sections cut parallel to the two planes, and the further history may be followed vid, so to speak, the adult plane. This was the course which I adopted; and I also penetrated back a considerable distance from the adult con- dition into the stages of the metamorphosis by sections parallel to the adult plane, and so confirmed results obtained by the other method. For the youngest stages of all, which are spherical, orientation is, of course, impossible, and one has to trust to chance to getting sections in the proper direction ; but it is fairly easy to recognise from their appearance when this is so. General Account of the Development. The ontogenetic history of the Asterina gibbosa may be conveniently divided into three parts: first, the development of the bilaterally symmetrical larva from the egg; second, the metamorphosis of this larva into the young star-fish ; and lastly, the gradual development of what we may term the young adult into the sexually mature form. I have made no observations on the segmentation of the egg, nor on the gastrulation; my work, properly speaking, commences with the completed gastrula, and my material was not suitable for observing the development of the calcareous plates. On all these points I intend, however, for the sake of completeness, to say a few words, and my authority will be Ludwig, who, in his classic research (12), has on these subjects left nothing to be desired in point of view of completeness. I may add also that the figures illustrating the changes in external form are copied from Ludwig’s memoir. The three figures illustrating the relations THE DEVELOPMENT OF ASTERINA GIBBOSA. DONS) of the Asterid and Crinoid to their common ancestor were designed for me by my friend and colleague Mr. J. J. Lister, of St. John’s: in their present form they were drawn for me by a lady friend. The Development of the Larva. The eggs are laid by the parent on the under surface of stones, to which they adhere by means of their vitelline membrane. I have never discovered a male, though Ludwig says that the male twists his arms round the female whilst she is depositing her ova, and then pours out his spermatozoa upon them; it is quite certain that in the English Channel, at any rate, isolated females will lay eggs which develop with perfect regularity up to the conclusion of the metamorphosis. Cuénot (4) says that young females of a certain size develop spermatozoa in their ovaries—a statement I have not been able to verify. It may, indeed, be said that Ludwig’s statement that a kind of sexual congress takes place, Cuénot’s observations, and the experience of the authorities of the Jersey Biological Station are irreconcilable, and that the whole subject demands renewed investigation. The eggs are larger than those of most other Echinoderms ; they are about ‘5 mm. in diameter. This is a result of the yolk which they contain, and which gives them their bright orange colour. This yolk is so uniformly distributed, however, that it does not alter the type of segmentation, which is total and regular. The blastomeres, in consequence of their larger size, are more closely packed than is usual amongst Echino- derms; they are wedged into the interspaces between their neighbours, and so the strict ‘ radial ’’! type of segmentation characteristic of the group is no longer maintained. The result of segmentation is a hollow blastosphere or blastula, which on the second day of development becomes converted into a gastrula by embolic invagination. The embryo 1 For a discussion of the different types of regular segmentation see “The Cell-lineage of Nereis,’ by Prof. E. B. Wilson, ‘Journal of Morphology,’ vol. vi, 226 E. W. MACBRIDE. is not quite spherical, its long axis exceeding very slightly its transverse axis, so that we can see that the blastopore is situated in the centre of what afterwards becomes the ventral surface. The gastrula has acquired a uniform covering of cilia, and the blastopore is a round opening with well-defined lips. This well-marked stage of development, which is easy to recognise, I have called Stage A (P]. XI, fig. 1). The blastopore narrows in a peculiar manner, one of its lips becoming reflected over it (PI. XI, fig. 2), and it is finally reduced toa minute pore (Pl. XI, fig. 3). This opening, which is identical with the larval anus, gradually travels back to near the posterior end of the embryo; this is effected by differences in the rate of growth of surrounding parts. During this time the embryo has been lengthening its long axis, and on the fourth day it ruptures the vitelline membrane and escapes. It then has the form shown in Pl. XI, figs. 4—6, and as this stage is also a well-marked one, I have called it Stage B. The foregoing is Ludwig’s account; my material was not suitable for such observations, which ought to be made on the living embryos, and I had not the opportunity of observing these early stages alive. As far, however, as I could make out, Ludwig is perfectly correct in his statements. I was able to recognise Stage A, for instance, with ease. Let us turn now to the internal changes which have gone on during this time. Pl. XII, figs. 20 and 21, are two sections of an embryo of Stage A, and they form the starting-point of the changes we shall have to consider ; I may here say at once that all sections which illustrate the development of the larva and its metamorphosis are to be understood to have been cut parallel tothe larval plane except the contrary is dis- tinctly affirmed. Fig. 22 is a sagittal section of a slightly older embryo ; here mesenchyme cells have appeared. The large size of the archenteron is a remarkable feature, the blastoccele or segmentation cavity, usually spacious in Echino- derms, being reduced to a mere slit. Fig. 23 shows us that the archenteron becomes differentiated into an anterior thinner- walled vesicle, the ccelom, and a posterior thicker-walled gut ; THE DEVELOPMENT OF ASTERINA GIBBOSA. 227 and in fig. 24 we see that the celom has grown back in the form of two tongues, Jpe., rpc., lying one at each side of the gut. Fig. 25 shows us a more ventral section passing through the blastopore of the same individual, and we see that in it these coelomic lobes are absent; they are therefore still con- fined to the dorsal side of the embryo. It has been mentioned above that the larva, immediately on escaping from the egg-membrane, has the form of Stage B, and it will be observed that its anterior end has the appearance of being obliquely truncated, and that the anterior surface so constituted is surrounded by a thickened rim, which is covered with specially long cilia, and to which I give the name of larval organ. The changes of form involved in acquiring this shape are considerable, and are undergone whilst the larva is still enclosed in the egg-membrane, though superficially the ovoid shape is maintained, the larval organ and the neighbouring ectoderm being to a large extent developed as invagina- tions into the interior of the larva, exactly as the Tenia head is developed on the wall of the cyst. The histology of the embryo is illustrated in Plate XIX, figs. 124 and 125. The first is a portion of section of a larva of Stage A, the same specimen as that from which figs. 20 and 21 are taken. Both ectoderm and endoderm are seen to consist of long narrow cylindrical ceils, and there is no mesenchyme. Recent researches have gone to show that this is exceptional. Field (5) has proved for Asterias, and it has been long known in the case of Echinids, that mesenchyme is formed by the wall of the blastula before any invagination has taken place. Fig. 125 is taken froma slightly older gastrula. It shows the forma~ tion of the mesenchymatous cells by the division of the endoderm cells. I found no indication that mesenchyme continued to be formed when Stage B is reached. The anterior wall of the celom is the spot where its formation lasts longest, as in Antedon (18). The ccelomic epithelium consists of small cubical cells (see Pl. XVI, fig. 95). We must now return to Stage B, up to which we have traced the development. A stomodzum is now developed just behind 228 BE. W. MACBRIDE. the posterior wall and ventral edge of the larvalorgan. This is well shown in the sagittal section, fig. 31. The larva increases in size, and the preoral portion and larval organ alter their shape, the latter changing from a circular to an elongated elliptical form, whilst the preoral lobe extends in a vertical direction (P]. XI, figs. 7 to 9). The whole larva has now the form which Ludwig calls slipper-shaped, but which would be more correctly termed boot-shaped, the dorsal lobe of the przeoral lobe representing the toe and the ventral one the heel of the boot. In the centre of the larval organ appears an elevation (fiz.). This structure, which Ludwig did not interpret, we shall find to have a most important function during the metamor- phosis; it is, in fact, the disc by means of which the animal fixes itself. Possibly this disc also functions during free life for temporary attachment, though in a different manner ; thus when the larval organ is applied to the substratum, the retrac- tion of this disc would cause a cupping action which would be relieved by its again being protruded. It has been pointed out by Ludwig, and I have myself confirmed it again and again, that the larva is able to attach itself most strongly to the sub- stratum. The mode of life of the larva Ludwig calls ‘‘creeping.” This is not strictly correct; as far as I have seen, the larva swims by means of the cilia of the larval organ. The latter is directed downwards, and for this reason Ludwig calls what I have termed the anterior surface of the animal the ventral, and the posterior end becomes for him the dorsal end. I cannot agree with this orientation ; the proper longitudinal axis of any bilaterally symmetrical animal is the oro-anal one, and it is by this that I discriminate between the dorsal and ventral, the anterior and posterior surfaces. That the posterior end is held upwards is no more reason for calling it dorsal than the fact that the Cephalopod directs the apex of its visceral hump back- wards is reason for calling that posterior. I should mention that Ludwig calls the whole przoral portion of the body, the preoral lobe in fact, the larval organ. I wish to avoid this, since the proral lobe has functions which Ludwig did not suspect, and hence I confine the term “ larval organ” to the THE DEVELOPMENT OF ASTERINA GIBBOSA. 229 thickened ridge with long cilia, which is the locomotor organ of the larva, and is the first thing to disappear in the meta- morphosis. Stage C is the point which we have now reached, and it is characterised by the appearance of this disc for fixation. Ludwig compares the larval organ to the non-ciliated processes of the Asterid larva, the Brachiolaria. This larva appears to be merely a further stage in the development of the well- known Bipinnaria, from which it differs in the development of three stalked papille from the apex of the preoral lobe, which are presumably used forattachment. These papillearise between the anterior dorsal and the anterior ventral arms of the Bipin- naria: one of them is median and more dorsally situated than the other two, and to this arrangement Ludwig compares the occasional bifurcation of the ventral lobe of the larval organ of Asterina. Now, however, that we know the function of the adhesive disc, it is, in all probability, this which is to be com- pared to the papille of the Brachiolaria; and the larval organ with its long cilia (compare Pl. XX, figs. 1833—185) in all pro- bability represents some portion of the ciliated bands of the Bipinnaria. Garstang (6) has, in fact, recently described a Bipinnaria in which the dorsal arm of the preoral lobe exe- cutes muscular movements in the same way as Ludwig asserts for the Asterina larva. I repeat, however, that the latter can swim by ciliary action alone, without any muscular move- ment. The internal changes which have occurred between Stages B and C are numerous and important. We have already referred to the appearance of the stomodzum or larval cesophagus. About the same time the primary madreporic pore is formed; it arises by a pocket of the ccelom slightly to the left! of the mid-dorsal line, meeting a thickening of the ectoderm (fig. 26, mp.) and a perforation taking place. The pocket of the ceelom is called the ‘ pore-canal” (pe., fig. 26), and is lined by cylindrical ciliated cells. By this time the two posterior 1 This position is not shown in fig. 26; the figure represents a section which was rather oblique. 230 E. W. MACBRIDE. lobes of the coelom have extended so as nearly to meet one another in the mid-ventral line ; the mesentery formed by their apposition is seen in fig. 30, posterior tothe gut. The opening of the gut into the coelom has become closed ventrally (figs. 29 and 30); dorsally, however, it remains open for some consider- able time yet. On the left side the coelom becomes segmented into an anterior portion, a., into which the pore-canal opens, and a left posterior portion, Jpc., which we may call the left posterior celom (fig. 27); this second cavity includes a large part but not all of the left coelomic lobe mentioned above ; part of this latter is, as is seen in the figure, included in the anterior celom. The septum between the two cavities is first formed dorsally, and then extends in a ventral direction ; fig. 28 shows it in process of formation. At the same time one can notice the first indication of that predominance of the organs of the left side which is the key to the whole ontogeny of the star-fish. We see in fig. 30 that the septum between the right and left coelomic sacs is pushed over to the right, owing to the tendency of the left posterior coelom to extend over to the right on the ventral side. At no time, so far as I have seen, however, does this septum break down. Some curious trabecule are in this stage stretched across the left coelom. They are easily distinguished from the septum between the two sacs, as they consist of solid strings of cells, whereas the septum has two layers of epithelium with a slit of blastocele between in this stage. These trabeculz are very transitory; in figs. 28 and 29 (Stage B) we see them being formed, and in fig. 33 is the last trace of them (Stage C). As development proceeds the gut becomes more completely separated from the ccelom, the larval anus closes, and the short rectum (fig. 31) disappears. Shortly before this, however, the stomodzum opens into the gut, the main portion of which constitutes the larval stomach (/. stom.), the rectum being very short; but it is only for an extremely short time that the larva possesses both mouth and anus. Stage C is reached about the end of the fifth day, or the commencement of the sixth day. The division of the left THE DEVELOPMENT OF ASTERINA GIBBOSA. 231 posterior ccelom from the ccelom of the preoral lobe, which we may now call the anterior ccelom (a., figs. 832—385), is com- plete. On the right side the separation of the posterior part of the right celomic lobe, the right posterior celom, from the anterior ccelom has just commenced dorsally (fig. 82). On the left side the rudiment of the water-vascular system, or, as it is convenient to term it, the left hydrocele, has appeared (as will be related immediately a similar rudiment appears on the right side, but “ hydrocele” alone means left hydroccele). It originates as an outgrowth from the hinder end of the anterior celom ; and whilst it is as yet but faintly marked off from this cavity, indications of its five primary lobes are seen. These are arranged ina curve open anteriorly, and throughout all the figures they are denoted by the Arabic numerals; the most dorsal being No. 1, the most posterior No. 3, and the the most ventral No. 5 (see figs. 32—34). Their mutual re- lations are well shown in the sagittal section (Pl. XIII, fig. 47), though this represents a somewhat later stage. We have seen that the division of the right posterior coelom from the anterior celom has begun in exactly the same manner as happened in the case of the left posterior celom at an earlier stage. This division has not proceeded very far towards the ventral surface, when the anterior coelom buds off a vesicle from its right posterior extremity. This vesicle is homologous to the water-vascular rudiment on the left side, for which reason it will be termed the right hydroceele ; so we see that the coelom on the right side of the larva undergoes exactly the same changes as that on the left, only that they are retarded in their appearance. The first trace of the right hydrocele is shown in Pl. XVI, fig. 95; we see that it consists of a small vesicle of cubical cells arising as a thickening of the ccelomic wall. Its lumen is, in this stage, a minute slit; other pre- parations show this slit in open communication with the an- terior celom. It is important to observe that it originates from the dorsal portion of the hinder end of the anterior celom, which extends further back ventrally to it, as would be seen if a more ventral section than fig. 95 were shown. 232 E. W. MACBRIDE. Later stages of this organ are seen in figs. 35 and 36. In fig. 35 it is a conspicuous solid bud; in fig. 36 it has acquired a lumen, and is connected with the anterior ccelom by a string of cells, which soon atrophies, and it is then left as an isolated vesicle in the midst of the mesenchyme. Bury (2), indeed, has seen it in this stage, and called it ‘‘ a mesenchymatous vesicle ;” and Field (5) has described what I believe to be an homologous structure in the larva of Asterias. The right hydroceele persists in the adult as a closed sac just under the madreporite, and has been seen here by Cuénot (3), and Leipoldt (9) has described a similar sac in Echinids. It may seem rather a rash assump- tion to regard this organ as the fellow of the water-vascular system, but a complete proof that this is really its nature will be given when abnormal larvee are described. Stage D, the summit of the development of the larva, is reached on the seventh day, according to Ludwig (Pl. XI, figs. 10 and 11). The preoral lobe and the larval organ have greatly increased in size, the former having acquired a large ventral as well as a dorsal lobe. The internal changes are more striking than the external. The separation of gut from coelom was practically complete in Stage C, the last trace of connection being shown in fig. 36. The right posterior cceelom is entirely separated from the anterior celom, but, strange to say, the septum between the left posterior celom and the ante- rior ccelom has become broken down in two places. This occurs by the two layers of epithelium of which it is composed fusing, and then thinning out toa film. Of these two secondary com- munications between the two sacs, one is situated dorsal to the left hydroceele (Pl. XIII, fig. 42), and the other ventral to it (Pl. XII, fig. 41). Figs. 42 and 43 belong to the same series ; we see that the dorsal opening is formed before the separation of the right posterior coelom is complete ; the ventral opening is formed at the same time. Not having had the opportunity when I wrote my preliminary account (15) of observing younger larvee than these, I imagined that the segmentation of the celom of the left side was incomplete ab initio, a mistake which was the more excusable as both the breaches in the THE DEVELOPMENT OF ASTBRINA GIBBOSA. 233 septum dividing the two portions of the ccelom from each other become again closed during the metamorphosis. The left hydroccele has become much more sharply separated from the anterior ccelom than in the last stage, though in the region of the third lobe the hydroccele still opens widely into the anterior ccelom (Pl. XII, figs. 88—41; Pl. XIII, figs. 44— 46). We saw that the pore-canal in Stage B originated a little to the left of the middle line; now, however, owing to the in- creasing predominance of the left side, it is shifted to the right of the median plane (pe., fig. 44). The stone canal (séc., figs. 45 and 46) arises as a groove along the anterior face of the transverse septum forming the hinder wall of the anterior celom. The central portion of this groove soon becomes closed to form a canal, opening at one end into the hydroceele between lobes 1 and 2 (fig. 46), and at the other into the anterior ccelom (fig. 45); and this opening is in this stage entirely inde- pendent of the opening of the pore-canal. I have referred more than once to the predominance of the organs of the left side. This is strikingly shown in the stage we are considering by the narrowness of the right posterior coelom as compared with the left. Already in Stage B we have seen that the left posterior coelom has begun to sweep round to the right on the ventral side of the right posterior ccelom ; this occurs more and more, and in the stage we are considering in the most ventral sections (fig. 41) the right posterior coelom is entirely absent. The left not only passes under it, but toa certain extent interposes between its anterior portion and the gut (figs. 89 and 40), and here opens freely into the anterior celom! (fig. 40) by the secondary ventral communication de- scribed above. This portion of the left ceelom we may call its right ventral horn ; it plays a most importaut part in the meta- morphosis, and it is marked /‘p’c’. in all the figures. Ludwig failed entirely to recognise the left posterior celom 1 T may anticipate a little by informing the reader that the anterior celom gives rise to the axial sinus of the adult ; a space which opens to the exterior by the pore-canal and into the left hydroccele (water-vascular ring) by the stone-canal. 234 E. W. MACBRIDVE. as a sac separate from the anterior ccoelom; he states that the mesentery between the right and left coelomic lobes is absorbed ventrally. We have seen that only the posterior parts of the right and left coelomic lobes are employed in the formation of the right and left posterior cceloms respectively; the anterior parts of these lobes are continuous with the anterior ccelom, and the longitudinal mesentery between them breaks down, as Ludwig observed. Hence we see that the hinder part of the anterior coclom in Asterina is at first a double structure; in the Bipinnaria larva the anterior ccelom is at first double throughout its whole extent. At the dorsal anterior angle of the left coelom (fig. 37) an invagination of its wall takes place, giving rise to a thick- walled vesicle (or. ¢.), which communicates by a narrow slit with the coelom. This structure has been strangely misunder- stood. Ludwig saw it, but not its origin, and supposed it to arise as a “* schizoceele,” and regarded it as the rudiment of the oral blood-ring. In my preliminary account I recognised its true nature, but supposed that its upper end was the rudi- ment of the so-called heart,! with which, as a matter of fact, it has nothing to do. It is the rudiment of the oral celom, a space closely surrounding the adult cesophagus, the relations of which we shall study later. Histology of the Larva. The structure of the body-wall of the larva is shown in Pl. XX, fig. 138, and Pl. XXI, fig. 144. In the first we see that the peritoneum of the left posterior ccelom consists of 1 Tt will be observed that Bury, in his last paper (‘Q. J. M. S.,’ September, 1895), makes the same mistake. This work appeared after the present paper had been sent in for publication, and is therefore not referred to further here. The best answer to Bury’s criticisms on my observations as recorded in the preliminary account (15) is the publication of full details in the present paper. Bury’s observations contain much interesting matter, but also in my opinion many mistakes, which are due to the fact that the stages which he obtained in the development of most of the larvee he studied, did not form a series without gaps; the orientation which he adopted seems to me also not that which yields the best results. THE DEVELOPMENT OF ASTERINA GIBBOSA. 235 small cubical cells; the ectoderm is made up of exceedingly long and narrow cells bearing flagella, and the wall of the hydroceele of similar cells, but I could not make out any flagella there, Fig. 144 is taken from the posterior end of the animal on the right side; the form of the ectoderm cells is well seen, and one observes occasional goblet cells (god.) amongst them. The section goes through a peculiar patch of peritoneum, where the cells are actively engaged in budding off the amcebocytes which float in the ceelom. So far as I can make out, however, no cells are budded off at this stage into the blastoceele (i.e. the space between the ectoderm and the celomic wall), and the mesenchyme cells are as yet entirely undifferentiated. The characters of gut cells are shown in Pl. XIX, fig. 126. Although this is taken from a larva in which the metamorphosis has commenced, yet the characters of these cells do not vary till the very close of the metamorphosis. They have the same general form as the ectoderm cells, but the bases of the latter are often contracted, and leave chinks between them, whereas the endoderm cells are closely apposed to one another. Fig. 126 also shows another point of interest: here and there a small round amcebocyte may be seen applied to the basal end of the gut cells, and one discovers amongst the latter also one or two rounded cells, thus suggesting that these ameebocytes may be able to pass between the gut cells like the lymph cells in the Vertebrate intestine. Plate XX, figs. 183—135, are three sections through the larval organ which have already been alluded to. It is to be noted that in this stage the adhesive disc has short cilia, just as See- liger (18) has described for the adhesive disc of Antedon. Where I have put “‘ xerv. darv.” a thin strand of pale fibrous matter is observable with the highest powers. This is the only trace I can discover of a larval nervous system, and I am not perfectly satisfied about it, since it does not take the yellowish-brown tone with osmic acid so characteristic of the adult nervous system. Should my interpretation of it be correct, the larval nervous system would consist of a layer of “ Punktsubstanz ” underlying the larval organ. 236 rE. W. MACBRIDE. Pl. XX, fig. 187, shows the character of the wall of the pre- oral lobe. The peritoneal cells have developed fine muscular tails (muse. larv.), and it is perfectly apparent to anyone looking at sections of a number of larvee that it is the peritoneum which is the active agent in contraction. The ectoderm is often wrinkled (fig. 38), but the peritoneum never, though its cells vary in shape from cylindrical to flattened according to the state of contraction ; thus in some cases the peritoneal cells on the left side will be cylindrical, those on the right side flattened. The cceelomic wall is in this case short and straight on the one side, and on the other bulged in to the lumen of the anterior celom by a great accumulation of the fluid of the blastoceele, or rather (as we must conclude from observations which have been made on other Hchinoderms) the blastoccelic semi-fluid jelly. In fig. 187 we see some fine fibrils traversing the blasto- coele; these, so far as I can make out, are not protoplasmic, but of skeletal nature—of the same nature, that is, as the adult fibrous tissue. The Metamorphosis. On the eighth day the larva fixes itself by the adhesive dise by means of a thin secretion of mucilage (see Pl. XX, fig. 136, which represents a much later stage), and remains fixed during the whole of the metamorphosis. I had the opportunity of observing this in Plymouth in 1893 and in Jersey in 1894, and it was most instructive to observe the difference between the larvee which had thus definitely become sessile and those which, being still able to move, had attached themselves by the cupping action of the muscles of the pre- oral lobe, the larval organ being applied to the substratum. In the first case, that of truly sessile larve, if one attempted to remove them with a pipette, one failed to move them unless very strong suction was applied or they were displaced by a needle; but once displaced they were perfectly helpless, those even which had to all appearance almost completed the meta- morphosis being unable to use their tube-feet (which as yet were rudimentary) ; they could do nothing but feebly rotate by THE DEVELOPMENT OF ASTERINA GIBBOSA. 237 the action of their general covering of cilia, and they had no power of re-attachment. In the case, however, of larvee which were attached by what we may call voluntary muscular action, if one brought the pipette cautiously near so as not to alarm them, it was very easy to remove them from a stone, just as it is easy to kick a limpet off a stone if it is taken unawares; but if they were irritated they were excessively difficult to re- move, and when one finally succeeded in getting them up into the pipette, unless one promptly re-expelled them, they at- tached themselves to the glass, and it was almost impossible to detach them from it. The metamorphosis of Echinoderms is probably the most remarkable ontogenetic change known in the animal king- dom; but our knowledge of its details has been up to the present most insufficient. We possess a completely satis- factory account of only one form, viz. Antedon, for which the credit is due to the researches of Bury (1), which have been amply confirmed by Seeliger (18). As I mentioned in the introduction, I hope the account I am about to give of the metamorphosis of Asterina will compare in completeness with those I have just mentioned; and as it is of the utmost im- portance for the comprehension of the meaning of the anato- mical structure of the Asterid that its relation to the larva should be thoroughly grasped, I shall anticipate a little what I have to say in order to make the essence of the process per- fectly clear. The metamorphosis of the Asterid, then, consists in the following processes, which go on simultaneously : (1) The constriction of the body into disc or body sensu stricto, and stalk, the latter being formed from the preoral lobe. (2) The sharp flexure of the disc on the stalk [the former is bent obliquely downwards and to the left. This is not well shown in any of the figures copied from Ludwig; it is better seen in the diagram, Pl. XXII, fig. 158 (Dec., 1895) J. (3) The preponderating growth of the organs of the left side, the left posterior celom and the left hydrocele both sending out dorsal and ventral horns, which meet so as to form complete VOL. 6. Cc 238 E. W. MACBRIDE. circles, whilst the right hydrocele and the right posterior coelom remain small. (4) The gradual atrophy of the stalk. (5) The outgrowth of the adult cesophagus and the formation of the new mouth on the left side. In the Crinoid the list would stand thus: (1) The constriction of the animal into calyx and stalk. (2) The displacement of the mouth and neighbouring organs, i.e. the hydroceele, to the posterior end of the body by unequal growth. (3) The mutual displacement of the right and left posterior celoms, the left becoming posterior and the right anterior, both having a ring-shaped growth. (4) The spiral growth of the intestine and formation of anus close to primary madreporic pore. It will be seen that the Asterid metamorphosis is very different from that of the Crinoid, being much simpler: one great difference which strikes one at once being that in the former case the ends of the hydroccele grow so as to embrace the stalk, which thus appears to spring from the oral surface ; whereas in the latter case the hydroceele is carried far away from the stalk to the posterior end of the body. Much diligent search has been made in the centre of the aboral surface of Asterids for traces of a stalk, but to anyone who has grasped the foregoing explanation it will be at once obvious how futile such search must prove. Pl. XXII, figs. 158 and 159, though intended to indicate ancestral forms, illustrate the two meta- morphoses outlined above very well. The sections about to be described illustrating the meta- morphosis are nearly all cut parallel to the larval plane, and as was the case with the sections of the larva, where two or three sections from the same series are figured the most dorsal is in every case placed first, and so one can clearly see their relation to corresponding sections of the larva. As one always thinks, however, of the organs of an Asterid as related to the plane of the disc or adult plane, it will be well to repeat the relation which these two planes bear to one another. The THE DEVELOPMENT OF ASTERINA GIBBOSA. 239 adult plane makes an angle of about 70° or more with the larval plane; but without any very serious error, it may be regarded, for purposes of description, as at right angles to it: thus the direction right to left, according to the larval plane, becomes aboral to oral according to the adult plane, and dorsal to ventral according to the larval plane is nearly parallel to the adult plane. Here I may remark that the words “ dorsal ” and “ventral” will only be used with reference to the larval plane; in speaking of the adult plane the words “ oral” and *aboral”’ will be used. Pl. XI, figs. 12 and 13, show the appearance of a larva which has only been fixed for a short time. On the left side we see that the hydroceele lobes have become visible externally, since they have raised the ectoderm into protrusions which, as we shall find, are the rudiments of sensory terminal tentacles of the radial water-vascular canals. Outside the curve of these rudiments is another set of protrusions, also arranged in an open curve. These are the rudiments of the arms: they are all, as we shall see, outgrowths of the left posterior coelom, and their primary function is to form supports for the lobes of the hydroceele, to which they later become apposed. The con- striction of the przoral lobe or stalk from the body proper is hardly as yet marked, but the rounded appearance of the dorsal and ventral outgrowths of the przoral lobe is to be noticed. This is due to the disappearance of the larval organ, the opposite sides of which become approximated to each other and wrinkled, and then broken up, portions of the organ becoming invaginated into the interior and destroyed by histolysis. ‘The appearance of the remnants of it at this stage gave Ludwig the impression that one had to do with the outgrowth of a series of protrusions homologous to the adhesive disc. Thisis, of course, a mistake; the adhesive disc remains single and unaltered to the end of the metamorphosis. This well-marked phase of development we may call Stage E. Pl. XIII, figs. 48 to 50, are taken from a larva of this age; fig. 48 is of course the most dorsal section (see explanation of plates). In fig. 50 we notice the great growth of the left hydrocele, lobe 3 reaching nearly to the 240 E. W. MACBRIDE. posterior end of the body, and we can also make out an arm rudiment, which at this stage is a mere protrusion of ectoderm filled with mesenchyme cells; it forms the extreme posterior end of the section. The rudiment of the adult cesophagus a. c is also seen, and we notice the relation of the oral ccelom to it, and we may remark that the larval esophagus is by this time disrupted from the gut. Fig. 49 shows that dorsally the hydro- cele is completely shut off from the anterior celom, and shows that the oral ccelom dorsally opens into the left poste- rior celom. Fig. 48 shows that the opening of the oral ceelom is in close relation to a process of the left posterior coelom extend- ing over to the right, dorsal to the gut. This is the right dorsal horn (see p. 233 for the ventral horn) of the left posterior ceelom, and it is marked /’p’c” in all the figures. In later stages it extends ventrally for a short way, insinuating itself between the gut and the septum dividing the anterior ccelom from the left posterior one (PI. XIV, fig. 61). The opening of the oral ccelom is later shifted so as to be connected only with the right dorsal horn, and hence it came to pass that Ludwig regarded oral coelom and right dorsal horn of the left coelom as one structure, and described the oral ccclom as the oral blood-ring and the dorsal horn as the “ heart.” In common with all other growing spaces in the larve, this right dorsal horn has at its growing tip an epithelial thickening, and it was this which in my preliminary account I mistook for the rudiment of the “heart.” Figs. 51—53, taken from a slightly older larva, show the appearance of the rudiments of the perihemal spaces. It may be useful to refresh our memory of the arrangement of these spaces in the adult; this the annexed woodcut is intended to do. They are usually described as consisting of a canal situated just aboral to each radial nerve, and divided by a longitudinal septum (Pl. XXII, fig. 155). These radial canals open into a circular canal surrounding the mouth, inside which is another inner ring-canal, The longitudinal septa of the radial canals are inserted into the septum separating these two ring-canals. Into the inner of the circular canals a vertical canal opens THE DEVELOPMENT OF ASTERINA GIBBOSA. 241 which is the axial sinus, embedded in the wall of which is the stone-canal (Pl. XVIII, figs. 11O—118). This axial sinus Fie. 1. ph. 1.2, &e. Rudiments of the outer perihemal ring. a’. Axial sinus and its outgrowth the inner perihemal ring. ad. Aboral sinus. gez.r. Genital rachis. was supposed to open at its upper end into an aboral perihemal ring or pentagon, from which in each interradius two canals branched off to go to the genital organs. As is well known, these spaces were called “perihzmal” by Ludwig (10), because he imagined that he had discovered the true blood-system in the form of curious tracts of tissue embedded in the longitudinal septa of the radial canals, and in the septum separating the two circular canals. He further supposed that that curious 242 E. W. MACBRIDE. so-called heart, which projects along with the stone-canal into the axial sinus, was connected with this system, and that a string of tissue lying in the aboral ring and connected with the “heart” was also part of the vascular system. We shall, however, see later that these two latter structures (“heart ” and aboral string) are of totally different nature from the oral ring, being composed of primitive germ-cells, and have, as a matter of fact, no connection with it. The radial tracts are absent in Asterina, but the oral circular tract is well repre- sented, and we shall study its development later. The woodcut (p. 241) shows us that the foregoing description is not quite correct. In the first place, we see that one can hardly speak of an outer perihemal ring, because this space is broken up into five compartments by the prolongations of the longi- tudinal septa of the radial canals; secondly, apart from the mistake we just pointed out in reference to the nature of the “heart”? and aboral ring, we see that the axial sinus (a’) does not open into the perihemal aboral ring; and, further, that to the upper end of the axial sinus is closely apposed a small closed sac, the right hydroceele. Returning to figs. 51—53, we see that each of the five compartments of the outer oral perihemal ring arises separately as a wedge-shaped outgrowth of the celom. Ihave numbered these rudiments according to the numbers of the lobes of the hydroceele between which they occur—ph. 1.2, ph. 2.8, ph. 3.4, ph. 4.5, and ph. 5.1; the last, however, arises later, and is not seen in these figures, and the first is an outgrowth of the anterior celom (Pl. XIII, fig. 51, Pl. XIV, fig. 54): all the rest arise from the left posterior coelom. The shape and relations of these rudiments are well shown in the enlarged drawing given of one of them (Pl. XX, fig. 139) ; we see that the base of the wedge is directed outwards, and that its basal angles tend to insinuate themselves between the ecto- derm and the hydrocele. As amatter of fact, each angle grows out till it meets the adjacent one of the next rudiment. The two then become apposed to each other, and their walls, which meet, form the longitudinal septum of the radial canal, and THE DEVELOPMENT OF ASTERINA GIBBOSA. 243 both spaces grow out together underneath the growing lobe of the hydrocele, and thus the radial perihemal canal itself is formed ; we shall find later that the inner perihzmal ring arises as an outgrowth from the oral end of the axial sinus or anterior coelom, and hence it is marked a’ in the woodcut. Fig. 53 shows us that the fourth and fifth lobes of the hydroceele have extended over to the right; this being the result of the tendency of the two ends of the hydrocele, which have become entirely shut off from the anterior ccelom, to approach one another. We also see from the obliquity of the right posterior coelom (compare figs. 44—46 with figs.52 and 53) that the lateral flexure of the body on the stalk has commenced. The flexure in a downward direction cannot be well shown by sections. Pl. XXII, figs. 54—57, are sections of a larva rather older than Stage E. We see that the differentiation of the stalk from the body has been initiated by the dorsal constriction of the neck of the prxoral lobe. In consequence of this the anterior ccelom becomes divided into a stalk portion a, and a body portion a’, the latter forming the axial sinus. We see, further, that the ventral horn of the left posterior ccelom I’p’c’ has pursued its growth, extending obliquely to the right under the gut, and then upwards in a dorsal and anterior direction, and on its course the last of the five arm rudiments appears, viz. V. Fig. 57 shows the outgrowth of septa destined shortly to close the ventral communication between this right horn of the left posterior coelom and the anterior celom. The primary lobes of the hydrocele have each by this time given rise to two lateral lobes, the rudiments of the first tube-feet, the primary ones themselves being destined to form the terminal tentacles of the water-vascular system. Figs. 58 and 59 represent a larva about midway between Stages E and F. We see the final division of the hydroccle from the anterior ccelom, the last connection being in the neighbourhood of lobe 3, and also the separation of the axial sinus from the stalk celom. We see also the remains of the larval cesophagus (/ce.), which already in Stage E has broken off 24.4, E, W. MACBRIDE. from connection with the gut; the relative position of the adult cesophagus (a.@.) is also well shown. Fig. 60 is from a larva of about the same age; it shows the formation of the fifth perihemal rudiment (ph. 5.1) as an outgrowth of the ventral horn of the posterior coelom: this lies beyond the fifth hydroceele lobe, and will therefore come to lie between this and No. 1 lobe when the two ends of the hydroceele meet. We also see the process of destruction of the. stalk going on, the ectoderm of its anterior surface being invaginated in patches, and, as we shall see, each patch as it is invaginated becomes destroyed by histolysis. Fig. 61 is from a larva which has nearly attained Stage F; it shows how the dorsal horn (/’p’c’”) of the left posterior celom wedges itself in between the gut and the hinder wall of the anterior coelom (a@’). In this wall we see running from left to right (i.e. from oral to aboral sides of the disc) from the second lobe of the hydroceele, the stone-canal. The ciliated cylindrical epithelium of this has now become continuous with that of the pore-canal, but only on one side; the conjoined tubes still open to the anterior coelom, and this opening persists in the adult, a fact which Ludwig did not observe (to see this, a more dorsal section than fig. 61 would have to be shown). The reader will remember that the pore- canal is formed by a dorsally directed outgrowth of the anterior celom fusing with the ectoderm, and a perforation occurring at the point of contact, and that the stone-canal is at first a ciliated groove running along the posterior wall of the anterior celom. This groove we found became converted into a canal opening into the hydroccele on one side, and the anterior ccelom on the other just below the inner opening of the pore-canal (woodcut 2). We have now arrived at Stage F’, the external appearance of which is shown in PI. XI, figs. 14—16. We notice that the preoral lobe or stalk has become very much reduced, and that the two ends of both curves, that of the hydrocele lobes (numbered in Arabic figures) and that of the arm rudiments (numbered in Roman numerals), have become very much ap- proximated to each other. At the same time we see that oral and aboral parts of the THE DEVELOPMENT OF ASTERINA GIBBOSA. 245 future star-fish are decidedly oblique to one another, being closely apposed posteriorly, but anteriorly separated by the thick base of the stalk. We see also that a lateral shift of the \ arm rudiments has commenced, No. V having passed beyond the hydroceele lobe No. 5, and so also in the case of the others. A second pair of rudiments of tube-feet has grown out from each lobe of the hydroceele, so that they are now 9-partite. Figs. 62—69, P]. XIV,are taken from a most instructive series of sections of a larva of this age, and are intended to give a clear conception of its internal anatomy. We are struck at once by the great reduction of the stalk, although ventrally (fig. 66) the stalk ccelom still communicates with the axial sinus. In fig. 65 we see the last trace of the secondary ventral communication between the left posterior coelom (/p’c’) and the axial sinus a’ (anterior ccelom) just closing. The secondary dorsal opening persists much longer, but fig. 63 shows us that it also is beginning to be closed. Comparing figs. 64 and 65, we see that the adult cesophagus has acquired two lateral out- growths, one directed anteriorly, the other posteriorly ; there is also a third horn directed dorsally, which of course cannot be seen in the sections. Fig. 67 shows how the oral ceelom (or.c.) now half encircles the adult esophagus. As to the arm rudi- ments, the most interesting thing is to notice the wide separa- tion of No. V from the hydrocele lobe No. 1. When the intervening tissue shrinks, a change which involves a reduction in size of the axial sinus (compare a’., Pl. XV, figs. 75 and 76), the metamorphosis will be complete. The incipient shift of the other rudiments is seen, especially in the case of Nos. II and III, the latter falling between lobes 3 and 4. By a continuation of the processes referred to above, viz. the constriction of the base of the stalk, the increasing flexure of the body on it, and the continued growth of the hydroceele and left posterior ceelom, we soon reach Stage G, which is represented in Pl. XI, figs. 17 and 18. We notice the great reduction of the stalk (which is now usually directed downwards almost at right angles to the disc, though the extent of the angle between the two varies) and the completion 246 E. W. MACBRIDE. of the circle of arm rudiments, though No. I is not quite adjusted to hydroccele lobe No. 2, and the hydroceele ring is as yet incomplete. Here is a fitting place to give in a word or two the gist of Ludwig’s observations on the calcareous plates. On the oral side (fig. 17) we notice ten small calcareous stars, two at the base of each primary hydroceele lobe, situated on the inner side of the first pair of tube-feet rudiments. These are the beginnings of the first ambulacral ossicles (amb.), On the aboral side we notice eleven plates, one central (C.), five situated in the arm rudiments and destined to form the terminals (7.) (the plates which protect the terminal tentacles of the water vascular system), and five interradially situated, the basals (B.), one of which becomes the madreporite. The name “ basal” is given on account of an imagined homology with the basals of Crinoids ; the groundlessness of this assump- tion I shall point out later. All these plates make their first appearance simultaneously, rather earlier than Stage F. Fig. 19 shows the aboral surface of a young star-fish about sixteen days old. We see that the anus has been formed close to the central; that a plate has been interposed between each terminal and the central, the former maintaining its position in the tip of the growing arm, and that finally a pair of plates has appeared in each interradius, peripherally situated with regard to the basals, the latter retaining their position in the centre of the disc. These paired interradial plates are homologised by Ludwig with the interambulacrals of Echinids. Plate XV, figs. 70 and 71, are two sections of a larva of Stage G. As in all the figures the stalk is placed as nearly as possible in the same position, one can see at a glance the very great lateral flexure which the disc has undergone with reference to the stalk. We see the relation of the rudimentary larval cesophagus to the permanent one ; we further see that the oral coelom is commencing ventrally to open into the left posterior one (this is of course a secondary communication, and I may say at once that the oral coelom does not give rise to a separate space in the adult, but merely forms the part of the ccelom abutting on the inner side of the buccal membrane), and finally THE DEVELOPMENT OF ASTERINA GIBBOSA. 247 we observe the incipient bifurcation of the posterior end of the pyloric sac (which is formed from the larval stomach) to form \the pyloric czeca. Fig. 79 is a section parallel to the adult plane of a slightly younger larva; it shows beautifully the mutual relations of the water-vascular ring (wv), the axial sinus, and the oral celom. If one compares this figure with Pl. IV, fig. 53, in Ludwig’s paper, one sees at once that his supposed rudiment of the oral blood-ring is only the oral celom. Figs. 75 and 76 show the completion of the metamorphosis by the apposition of arm rudiment No. V covering the tip of the ventral horn of the left celom (I'p’c’) to hydrocele lobe No. 1. As compared with the larva represented in Pl. XIV, figs. 62—69, we notice the much smaller size of the axial sinus (a’). Fig. 75 shows also the bifurcation of the anterior end of the pyloric sac into two czxca, Comparing it with fig. 76, which is a more ventral section from a larva of the same age, we see also that the spaces between the pyloric ceca (py) and the aboral body- wall are continuations of the right posterior celom. Fig. 76 shows also the first trace of ovoid gland (“heart”) (ov.g.) arising as a ridge of epithelium including blastoccelic jelly and fibres and ameebocytes, projecting into the axial sinus, By comparing this figure with Pl. XIV, fig. 61, the shift of arm rudiment No. V can be clearly made out. Figs. 80 and 81 are sections parallel to the disc of a larva rather older than Stage G. Fig. 80 shows how the oral ccelom almost surrounds the ceso- phagus, and also that the axial sinus is commencing to form the inner perihemal ring by growth from its lower end (compare woodcut). In fig. 81 we see at the point marked * the closing of the water-vascular ring by outgrowths from the hydroccele lobes Nos. 1 and 5 respectively. We also notice what we have already seen in fig. 76, that the septum between the oral coclom and the left posterior coelom is breaking down; and in fig. 82, which is from a young star-fish in which the metamorphosis is just complete, we see that from the remnants of this septum the retractor muscles of the oesophagus or “ stomach ” are formed, The remaining figures on the plate show the finishing touches 248 E, W. MACBRIDE. of the metamorphosis. In fig. 72 the adult mouth is formed, and the sessile mode of life has been given up, the stalk being reduced to a small solid rudiment. We see also the first trace of the eye as a small knob at the base of hydroceele lobe No. 8. Fig. 78 shows the permanent anus; if we com- pare its position with that which the larval anus occupied, we find that they are by no means the same: the larval anus, if it had persisted, would be situated at the point x, though both occupy a position on the mesentery dividing the left from the right posterior celoms. Fig. 77 from the same larva shows that the left posterior celom now forms a complete ring by the breaking down of the partition between its right ventral and right dorsal horns (/‘p’c’. and Ip’e”.). In fig. 73 a dorsal section, and in fig. 74 a ventral section, we see the incipient bifurcation of the right posterior coelom in order to form the outgrowths connected with the two dorsal and the ventral pyloric ceca respectively. We see, therefore, that of the five pyloric ceeca, two are formed from the dorsal end of the pyloric sac or larval stomach, and two from its ventral end, and that their suspensory mesenteries are outgrowths from the mesentery separating right and left posterior celoms. The fifth ceecum is directed dorsally and posteriorly. In Pl. XV, fig. 82, and Pl. XVI, figs. 83, 84, we have three sections parallel to the adult plane of a specimen which had just completed the metamorphosis. Once the mouth is open, the trifid form of the adult cesophagus changes, we get the five slightly bifid lobes of the adult “stomach.” In fig. 83 we see the first trace also of the bifurcation of the pyloric ceca; I remind the reader that in each arm of the adult there are two ceca; the characteristic appearance of the axial sinus, stone-canal, and right hydrocele in a section parallel to the disc are also shown, the right hydro- cele having a crescentic form. Fig. 84 shows us the relation of the rectum and the rudiment of the rectal cecum to the pyloric ceca; we see that the mesentery which binds the bases of the pyloric czeca together is only the original mesentery between the right and left posterior (oral and aboral cceloms) ; and, further, that the mesenteric band connecting the inter- THE DEVELOPMENT OF ASTERINA GIBBOSA. 249 radius of the stone-canal with the stomach is a part of this same criginal mesentery, with which, however, is continuous a piece o) the wall between dorsal and ventral horns of the left coelom, these two horns being still separated by this wall near their right sides (aboral surfaces). Histological Changes during the Metamorphosis. Up to Stage G the histology has little changed from that of the larva before metamorphosis. The most striking alterations are those connected with the destruction of the preoral lobe. Pl. XX, fig. 136, givesa specimen of them. This figure, which is taken from the larva represented in figs. 62—69, shows that the ectoderm becomes invaginated into pockets, and then these pockets completely closed, so that no breach in the continuity of the skin is made. The invaginated portion is then destroyed by ameebocytes as shown in the figure. The peritoneum lining | the stalk ccelom contracts violently, the cells becoming cylin- drical instead of flattened, and the larval muscles very appa- rent. So far as I can make out, these cells are destroyed by amecebocytes of the ceelom. In the larva the whole hydroccele rudiment is lined by cylin- drical cells (P1. XX, fig. 188); but as metamorphosis proceeds, and the hydroceele increases in size, the cells are stretched so as to become flattened (P]. XX, fig. 139); they retain their original character only in the rudiments of the tube-feet (PI. XXI, fig. 149) and terminal tentacles. The first trace of the adult nervous system appears in Stage F in the ectoderm covering the water-vascular ring,—that is, the portion of the hydroceele between the primary lobes. The ectodermal cells become long and filamentous, with their nuclei set at different levels, and amongst their bases (P]. X XI, fig. 140) appears a tangle of fine fibrils of excessive tenuity, so that the highest magnification is required to make them out; this is the first trace of the adult nervous system. Ludwig talks of cells stretched parallel to the surface under the ectoderm, which he supposed to become the bipolar gan- glion cells of the nerve-cord; but the cells in question, if I 250 E. W. MACBRIDE. rightly identify what he means, are only the epithelial lining of the perihzemal spaces which at a later period become closely apposed to the ectoderm. The first trace of muscles in the body-wall appears much earlier. Pl. X XI, fig. 145, shows the formation of a well-marked muscular band from the wall of the right posterior coelom of a larva of Stage E. We see that it consists of indubitable myo-epithelial cells. I have traced this band into the oldest specimen I have examined for histo- lugy ; and so far as I can see it appears to become a dilator of the anus. It is very strange that it should appear. long before any other muscles of the body-wall ; it forms quite a conspicuous feature in sections of all well-preserved metamor- phosing larvae. The same figure shows the first trace of histo- logical differentiation in the mesenchyme; we see the first formation of that fibrous intra-cellular substance which gives firmness and tenacity to the adult body-wall. The cells of the gut remain unchanged till the very end of the metamorphosis, but in Stage G we can trace some differen- tiation. Pl. XIX, figs. 127, 128, show part of the lining of the adult cesophagus and of the pyloric sac of such a larva. The cells of the former are very long and narrow, and their outer portions take a clear yellow tone with osmic acid; those of the latter are ordinary cylindrical epithelium cells. Abnormal Larve. I mentioned above that the demonstrative proof that the sac I have termed the right hydroceele is of that nature is obtained from the study of abnormal larve. I suppose that about one in thirty of the larve I examined were abnormal, though in very different degrees. The commonest abnormality results from the unusually great development of the organs of the right side, and the consequent checking of the metamorphosis.! The larva of which the two sections are given in figs. 85 and 86 had about attained Stage D. The left hydroceele is perfectly normal, but the right, though not much larger than usual, is 1 The reader will remember that in the analysis of the metamorphosis which I have given on p. 355, one of the main factors recognised is ‘the preponde- rating growth of the organs of the left side.” THE DEVELOPMENT OF ASTERINA GIBBOSA. 251 divided into distinct rounded lobes lined by cylindrical epi- thelium (rfy.), in all respects similar to those of the left, and the rudiment opens by a narrow but distinct slit into the anterior ceelom. This larva also exhibits another very common abnormality, which I do not in the least understand ; this con- sists of the breaking up of the gut epithelium into a mass of cells having the appearance of mesenchyme, which choke up the lumen, but leave the walls almost denuded of epithelium, consisting chiefly of the basement membrane. This curious change can take place at any stage from the commencement of the differentiation of the celom, up to young adults a month old: in one such specimen it affected the pyloric ceca. As to what its meaning is, I confess I am entirely in the dark. Figs. 87 and 88 represent a most remarkable larva. The development of the left posterior coelom would indicate that it had reached Stage E, but the left hydroccele consists only of four lobes, and is poorly developed. There are two rudiments of a hydroceele on the right side ; the more ventral has three distinct lobes lined by cylindrical epithelium (7”hy’., fig. 88), and opens by a distinct opening into the anterior ccoelom; the more dorsal is perfectly normal (rhy., fig. 87); but, as if to emphasise the fact that, in spite of the presence of the other rudiment, it does in fact represent a hydroceele, we find in connection with it a second small stone-canal and pore- canal (p’c’. st’. c.). The relation of these to the right hydro- cele may seem unusual ; instead of the canal (conjoined stone and pore-canal) leading from the hydrocele to the anterior celom and thence to the exterior, it appears to lead from the anterior ccelom to the hydrocele and thence to the exterior. This apparent difference may be reconciled with the arrange- ment on the left side by observing the angle which stone- canal and pore-canal make with one another. Woodcut 3, p. 252, shows that this is an acute instead of an obtuse angle, and hence that stone-canal and pore-canal have coalesced laterally ; Woodcut 2 shows for the sake of comparison the normal stone- canal and pore-canal and their relationship to the left hydro- cele and the axial sinus or anterior ceelom. 252 E. W. MACBRIDE. Fig. 89 is a section of a larva of Stage D; both hydrocceles are well developed—the right, in fact, better than the left; the Fie. 2. Fic. 3. right hydroccele appears on the left side of the figure, since by an oversight the section was drawn from the wrong aspect. It took me some time in this larva to determine which side was which, but the right hydroceele is rather more dorsally situated, and opens by only a narrow slit into the anterior celom. It is also curved somewhat differently, the most posterior lobe being No. 4, not No. 8, as on the left side. Fig. 90 shows a most remarkable variation. We see a pore opening directly from the hydroceele to the exterior. If,as I shall attempt to show later, the anterior coelom may be compared to the proboscis cavity of Balanoglossus, and the two hydroceeles to the collar cavities of that animal, we see that what we may terma collar-pore may arise as a variation. Figs. 91—94 are sections taken from a larva of Stage G. Its only abnormality is that in connection with the right hydrocceele, which is of normal character, a second pore-canal and stone-canal are developed. Fig. 92 should show the opening of the second stone-canal into the hydrocele ; fig. 93 the opening of conjoined pore-canal and stone-canal (compare woodcut 3) into the axial sinus. Fig. 91 shows that the two pore-canals unite, to open by a common median pore. The above are not by any means all the variations observed, but they are sufficiently typical to in- dicate their nature, THE DEVELOPMENT OF ASTERINA GIBBOSA. 253 The History of the Young Star-fish. The just metamorphosed Asterina gibbosa has a disc of about ‘6 millimetre in diameter; if we take R to denote the length from the tip of the arm to the centre of the disc, then R equals °36 millimetre. A larva such as that figured in figs. 51—53 may be °8 millimetre from the tip of the adhesive disc to the posterior end, and measured obliquely from the dorsal end of the preoral lobe may exceed a millimetre in length. There is, therefore, a considerable diminution in size during the metamorphosis, the reason of which is evident when we consider that no nutriment is taken during this time. A full-grown specimen may have a diameter one hundred times greater than that of the just metamorphosed star-fish,—that is, it may exceed the latter one million times in bulk. The young star-fish, however, rapidly increases in size, and by the time R equals 3'7 millimetres the ovaries are visible. This is the oldest stage I have examined; my account of the histology is, however, taken from smaller specimens, in which Requals ‘8mm. The changes we shall have to consider are (1) the formation of the primitive germ cells, the ovoid gland, genital rachis, and ovaries ; (2) the dermal branchie; and (8) general histological differentiation. We have already in Fig. 76 seen the first trace of the ovoid gland. It there appears as a ridge projecting into the axial sinus ; inside this ridge there is as yet to be found only ameebo- cytes, jelly and fibres, as is the case with the other blastoceelic spaces in the larva. Later, a thickening of peritoneum takes place on the wall of the left posterior ccelom opposite the aboral end of this ridge—and from this thickened patch a cord of cells grows into the ridge, gradually forcing its way in an oral direction ; this is the characteristic core of the ovoid gland. From this same thickening of peritoneum a cord of cells grows out in a direction parallel to the disc ; this is the origin of the genital rachis. By the outgrowth of a flap of peritoneum it is enclosed in a space which is called the aboral sinus. The genital rachis and the space enclosing it both give off branches VOL. 6. D 254 E. W. MACBRIDE. one at each side of each arm. Local thickenings of these branches of the rachis constitute the genital organs. The surrounding spaces, the genital sinus (ab gon, figs. 122 and 123), is shut off from the aboral sinus by the outgrowth of a septum. Fig. 99 is the marginal portion of a section vertical to the dise of a larva of StageG. We see the rudiment of the ovoid gland (ovg.) as a fold projecting into the axial sinus. Further up we notice a thickened patch of peritoneum, which is invaginated into the septum separating the axial sinus from the left posterior celom (pr. germ. inv.). This is the rudiment from which, on the one hand, the genital rachis and, on the other, the core of the ovoid gland are derived. Figs. 1O0O—103, similar sections to fig. 99, from a just metamorphosed star-fish, illustrate this. We see that from this rudiment a cord of primitive germ cells has grown out and filled the fold which is the rudiment of the ovoid gland. The last two sections cut a more oral portion of the fold, since they are slightly oblique; we see (figs. 102 and 103) that this core has not as yet penetrated to the oral end of the fold, and, further, that the fold is attached to the oral side of the inner perihemal ring, or, in other words, that it traverses the lower end of the axial sinus, and is attached to its lower side. The original invagina- tion to form the germ cells is situated at the very tip of the right dorsal horn of the left coelom, where it meets the right ventral horn, but at this level the two horns do not open into each other (see p. 249). Figs. 104—106, again representing sections vertical to the margin of the disc, are taken from a young star-fish, in which R equals ‘4 millimetre. Fig. 104 shows the cord of cells which arises from the peritoneal invagination and penetrates the dorsal organ, and the relation of this cord to the right hydroccele and the axial sinus. We see that now this core of cells reaches to the oral end of the ovoid gland, and penetrates also a prolongation of the same, which is prolonged as a fold, hanging from the aboral wall of the inner perihzemal canal (figs. 105 and 106). Pl]. XVIII, fig. 110, which represents a similar section to figs. 99—106, shows practically the adult condition of the ovoid THE DEVELOPMENT OF ASTERINA GIBBOSA. 955 gland and neighbouring organs. We see that the madreporic pore has commenced to be divided into two by the ingrowth of a fold. It is not the case in Asterina, as far as I can make out, that the numerous pore-canals found in the fully grown adult are derived from fresh perforations, as Cuénot has stated (3). Rather the statement which he quotes from Perrier seems to give the actual method of their formation.! We see that the openings of the stone-canal proper and the pore-canal into the axial sinus are still maintained. The ovoid gland with its core is seen to reach right down to the oral end of the axial sinus, and to be attached to its oral wall. Embedded in the septum dividing the inner perihemal ring-canal (lower end of the axial sinus—see woodcut 1) from the perihemal spaces proper is the so-called oral blood-ring (sang. circ.). This is a ring-shaped tract of peculiarly modified connective tissue; the section shows that it is of a different nature from the ovoid gland, and has no connection with it. In Asterias this ring gives off radial pro- longations traversing the longitudinal septa of the radial perihemal canals, but these do not exist in Asterina, The development of this structure as far as its histology is con- cerned is shown in P]. XVII, figs. 107—109, which represent small portions of sections parallel to the disc. The first two sections are taken from the same specimen as figs. 82—84; in this specimen as we have already learned (see above, p. 248) the metamorphosis has just concluded. We see that the mesenchymatous tissue between the outer and the inner peri- hemal rings has undergone differentiation. Most of it has be- come converted into fibrous tissue, but at one level no fibres have been formed, the whole of the mesenchyme cells becoming ameebocytes (sang. circ.); this part is the rudiment of the blood-ring. In fig. 109, taken from a specimen in which R equals ‘45 millimetre, we see that the ring is completely formed ; 1 Durham, in a paper on “ Wandering Cells in Hchinoderms” (‘ Quart. Journ. Micr. Sci.,’ vol. xxxiii), has described the communication of the axial sinus and stone-canal in a young Cribrella. He also insists that we have no blood-vessels, but rather “ hemal strands ” in Echinoderms, but makes thie common error of supposing the ovoid gland to belong to this category. 256 HE. W. MACBRIDE, the intercellular jelly or plasma has acquired staining properties. To Leipoldt (9) is due the credit, in a careful paper on the anatomy of “ the so-called excretory organ of the sea-urchin,” of emphasising the fact that the ovoid gland and the oral blood- ring are of totally different nature; he describes branches from the blood-ring ramifying on the external surface of the ovoid gland. The question arises, what is the true nature of this blood- ring? Cuénot (3) answers that it is a lymphatic gland, or centre for the formation of amcebocytes; and there is a great deal to be said for this view. We must, however, remember that structures of similar nature are found accompanying the alimentary canal in Echinids and Holothurids. Ludwig (18) has given a splendid description of their arrangement in the last group. He brings out with great clearness that they are tracts of connective tissue in which the fibres are sparse. The close relation of these “ vessels ” to the alimentary canal suggests forcibly that we may have here the first attempt at forming blood-vessels. There is certainly no propulsive organ or proper circulation, but the staining properties of the plasma show that it has been chemically altered, and the idea is suggested of some secretion from the gut-cells propelling itself along these tracts by the vis a tergo force of secretion. In the Asterid no close connection with the gut is observable,—the oral ceelom, in fact, intervenes between the cesophagus and the ring, as we have seen (p. 247); but the altered character of the plasma suggests that perhaps here some substance is formed necessary for the well-being of the organism, which then diffuses out into the neighbouring celomic spaces. The blood-spaces of the higher animals are known in many cases to be remnants of the blastoccele or segmentation cavity of the embryo; this has been shown in the case of Balanoglossus with great clearness by Spengel (21). Strictly speaking, therefore, the blood and lymph spaces of other forms are represented in Echinodermata by all the spaces in the body-wall unoccupied by fibrous tissue and dermal ossicles, and traversed by amcebocytes; but the blood-ring, gut vessels, &c., may be a first attempt at specialisation. THE DEVELOPMENT OF ASTERINA GIBBOSA. 257 Figs. 113—117 are intended to illustrate the formation of the genital rachis; and they all represent portions of sections cut parallel to the disc; those portions, in fact, which are transverse sections of one of the five interradial folds of the body-wall which in the star-fish project into the body-cavity. As we see in Pl. XVI, fig. 83, the axial sinus, right hydroceele, and the stone-canal, are embedded in one of these folds. It follows that the ccelomic wall of this particular fold represents the larval septum between the anterior coelom and the pos- terior ceeloms; and its interradial position in the star-fish becomes explained when we remember that the stalk with its contained anterior ccelom lies opposite an interradius of the water-vascular ring; which interradius is constituted by the outgrowth of processes of the two lobes situated at the ends of the hydroccele, which is as yet an imperfect ring. These out- growths meet, so to speak, above the neck of the stalk. Figs. 113 and 114 are from the same specimen as fig. 109. We see the appearance of the rudiment of the germ cells in a section parallel to the adult plane, and notice the remains of the cavity of invagination (fig. 114, pr. germ. inv.). Fig. 113 shows us that one horn of the right hydrocele has become embedded in the ovoid gland, and this is one reason why it is extremely difficult to trace the continuity of the primitive germ cells by sections taken parallel to the adult plane, since the cord of cells is in some spots so narrow, that it is therefore difficult to distinguish it from the epithelium lining the right hydrocele. Longitudinal sections, such as fig. 104, show it much better. In figs. 115 and 116 (taken from a specimen in which R equals ‘7 millimetre) we see the formation of the genital rachis; this takes place by a lateral outgrowth from the primitive patch of invaginated peri- toneum, from which we have seen the core of the ovoid gland originating as an orally directed outgrowth; the aborai sinus which surrounds it (ad.) is formed at the same time, it is a portion of the cclom shut off by the outgrowth of a fold of peritoneum. Fig. 117, taken from a much older specimen, shows the genital rachis in its complete form 258 EH. W. MACBRIDE. in continuity with the original rudiment of the primitive germ cells. It is, then, not quite correct to speak of the genital rachis as being an outgrowth from the ovoid gland, as Cuénot has done (3). This statement, nevertheless, marked a step in advance in our knowledge, for it gave a hint as to the meaning of the ovoid gland. Cuénot found specimens of Astropecten with the ovoid gland, but without the genital rachis, and noting the identity of the character of the cells in the two structures, stated that the rachis was an outgrowth from the gland, though he found no intermediate stages. These were first found by me (14) in the Ophiurid Amphiura squamata, and at the same time I demonstrated the epithelial origin of both gland and rachis. It is the genital rachis which of course was formerly known as the aboral blood-vessel; in most Asterids and Ophiurids it later undergoes partial degeneration, giving rise to cells con- taining violet pigment. Ludwig, however (11), and Hamann (7) have pointed out that the central core remains unaltered ; the latter was the first to point out that in all Echinoderms, except Holothurids, a genital rachis exists, of which the genital organs are local outgrowths. In Amphiura squa- mata, however, and in Asterina gibbosa, according to Cuénot (3), the whole genital rachis remains unaltered through life; this is only one of the many points in which Asterina shows itself to be one of the most primitive of Asterids. In the plans given in text-books of the blood system, two vessels are shown proceeding from the aboral ring in the interradius of the madreporite to the pyloric sac. These are two mesen- teric bridles, remnants of the piece of septum left at this level between the two horns (right dorsal and right ventral) of the left colom. At this spot the right (aboral) coelom breaks through into the left (oral) ccelom, perforating the piece of tissue referred to, and leaving only the mesenteries. The peritoneum covering them seems to be peculiarly modified, and is possibly a place where the ameebocytes of the ccelomic fluid are formed. The genital rachis gives off, as it passes each interradius, two branches enclosed in corresponding branches of the aboral THE DEVELOPMENT OF ASTERINA GIBBOSA. 259 sinus (gen. r., woodcut 1); one of these branches runs in an oral direction down each side of the interradial septum. This septum is an ingrowth of the body-wall, which has by this time become marked, though its first beginnings date back to the end of the metamorphosis (Pl. XVI, fig. 84). A section of one of these branches in an older specimen is given in Pl. XIX, fig.119). These genital branches are formed as the rachis reaches each interradial septum before it has formed a circle; in one specimen.I have observed a rachis reaching only to the next interradius, and there giving off one genital branch. Figs. 120 and 121 (taken from the same specimen as fig. 119) show the first rudiments of the genital organs. ‘The branch of the rachis ends in a swelling accom- panied by a dilatation of the aboral sinus, and we see the begin- ning of aseptum tending to shut off the main aboral sinus from this dilatation. This septum was first described by Cuénot (3), and in it the genital duct is formed. This is shown in fig. 123, taken from the oldest specimen I examined, in which R equals 3'7 millimetres. We see that the genital duct is formed by a core of primitive germ cells burrowing its way through the body-wall. Fig. 122, from the same specimen, shows the con- tinuity of the rachis and the ovary. We notice also the forma- tion of follicles from the indifferent germ cells. We are now ina position to compare the arrangement of the ovoid gland and genital rachis and their accompanying spaces in Amphiura squamata with that found in Asterina gibbosa. In the former I described the genital rachis issuing from the oral end of the gland and accompanied by three spaces, which I named sinus @, sinus 0, and sinus c (Pl. XVIII, fig. 112). This figure is a diagram of a section parallel to the ~ long axis of the stone-canal. Fig. 111 is a diagram of a similar section of Asterina, but it is not quite accurate, since it shows both the ovoid gland and the stone-canal, and these two structures do not lie in the same radial plane in Asterina. In order to avoid obscuring the opening of the stone-canal into the axial sinus, it is necessary to indicate part of the ovoid gland by dotted lines. 260 E. W. MACBRIDE. Comparing figs. 11] and 112 we see that the axial sinus of Asterina is represented in Amphiura by sinus c, the so-called “ampulla.” The aboral sinus (ad, fig. 111, sinus a, fig. 112) is also obviously homologous in both. [Since my paper (14) was published, and since the present work was sent in for publication, I have made a careful re- examination of my sections of Amphiura squamata, and have arrived at a more complete comprehension of the structure and development of the ovoid gland and the neighbouring spaces in that animal. The space marked sinus 0’ (fig. 112) is not, as I formerly supposed, a part of sinus 4, but is quite distinct. Sinus 0’ probably represents the right hydroceele ; it is already present in the youngest specimens I examined. Sinus 5* is a portion of the ccelom shut off by the outgrowth of a flap of peritoneum; from the inner wall of this sinus the cells which at the same time give rise to the ovoid gland and to the genital rachis take their origin; it is obviously homolo- gous to the cavity of the invagination of the primitive germ cells (pr. germ inv., figs. 110 and 111), only in Asterina this space disappears.— December, 1895. ] We observe that the arrangement in Amphiura might be obtained from that in Asterina by rotating the stone-canal and accompanying structures outwards and downwards through an angle of 180°. That this is what has occurred in phylogeny is indicated, not only by the fact that in the young Amphiura the madreporite is near the edge of the disc and the stone- canal almost horizontal, whereas in the adult the madreporite is situated far in towards the mouth on the oral surface, but also by the curious undulating course of the genital rachis, which is aboral in the interradii and oral in the radii. This points to the conclusion that the aboral parts of the interradii * Jn my paper on this subject (14) sinus 0 is referred to as the axial sinus— it was formerly supposed to be continuous with sinus ¢c, though Ludwig knew this was not so. At that time the meaning of the axial sinus in Asterids which Bury first suggested (2) was not generally known, and his interpreta- tions were not accepted, and hence two different spaces were called axial sinus, one in Asterids and the other in Ophiurids. THE DEVELOPMENT OF ASTERINA GIBBOSA. 261 have greatly developed, and have grown in between the radii on to the oral surface, forcing the original oral plates to the extreme centre of the disc; and so the stone-canal has been swung round and the genital rachis pulled out of shape. Nowin Asterina gibbosa there is a trace of this process; the rachis does not, as Hamann (7) has described in Asterias, lie in one plane, but pursues an undulating course, being much more aboral in the radii than the interradii. Iam inclined to look upon this as the primitive condition from which the Asterid and Ophiurid arrangements have been derived. I may as well mention here some other facts which indicate the primitive nature of Asterina. Chief among them is, that in the family of which it is a member we meet with the most rudimentary form of those characteristic Asterid organs the pedicellariz. We have in Asterina the aboral surface covered with small spines, arranged in twos and threes, and acting on irritation like pedicellarie. It is true that some Asterids have no pedicellariz, but here the evidence from allied genera (cf. Luidia and Astropecten) suggests that they have been lost; Asterina, however, shows us pedicellariz in statu nascendi. The simple biserial tube-feet also con- stitute a primitive character. Fig. 118 represents ovoid gland and stone-canal in the latest stage examined by me. The gland is attached by an exceedingly narrow pedicle to the wall of the axial sinus. Its surface is thrown into deep folds, and the peritoneal lining of the axial sinus, which forms its outer covering, is modified, consisting of cylindrical cells with projecting rounded ends. The interior of the gland is filled with a mass of primitive germ cells supported by fibres, doubtless of mesenchymatous origin. I was unable to find any trace of a tube lined by primitive germ cells, such as was discovered by Hamann in the young Asterias. What, we may finally ask, is the function of this strange organ? Cuénot, as usual, maintains that it is a lymphatic organ. This I am disposed to doubt very strongly; the cells which it contains are of quite a different nature from the amcebocytes of the oral blood-ring, and the evidence that 262 E. W. MAOBRIDE. Cuénot brings to show that they escape by diapedesis into the axial sinus is quite insufficient. The cells of outer epithelial lining are not flattened but cylindrical, and I strongly suspect that he has mistaken their freely projecting ends for escaping amcebocytes; and I may remark that this curious outer epithelium shows its distinctive character from the time the first rudiment of the ovoid gland appears. Whatever its function may be now, there is no doubt that the ovoid gland was pri- mitively a part of the genital organ, and probably is a remnant of the arrangement of the reproductive cells before the radial symmetry was acquired. It is interesting to notice that it originates from the left posterior coelomic wall, whereas an analogous organ in Crinoids arises in the right or aboral celom, so that they are not strictly homologous. If Hamann is, as there is strong reason to suppose, right in stating that the primitive germ cells wander along the rachis into the genital organ, it seems very probable that, at any rate in the young adult, the ovoid gland is a centre of formation of the primitive germ cells; and its relation to the axial sinus may have to do with its aération, for it must be remembered that the pore-canal opens into the axial sinus, and the current in this is, as we shall see, inwards. In the fully grown adult it no doubt undergoes, to some extent, the degene- rative change noted above in the genital rachis of other genera. What the meaning of this change is, is very obscure. Obser- vations on the histology of the gland at different seasons might elucidate its meaning. Turning now to the stone-canal, we see, in fig. 118 (asection transverse to the axial sinus and stone-canal), the beginning of that curious T-shaped ingrowth which is so marked a feature of the stone-canals of Asterids, but which is much less developed in Asterina than in other genera. It is covered by short cilia, the rest of the epithelium bearing long flagella. Cuénot asserted that the stone-canal was a functionless rudi- ment, the current being neither outwards nor inwards. Ludwig! ' Ludwig, “ Ueber die Function der Madreporenplatte und des Sioa der Hehinodermen,” ‘ Zool. Anz.,’ 1890, p.377. THE DEVELOPMENT OF ASTERINA GIBBOSA. 263 subsequently showed that in the stone-canal of Holothu- rids and Echinids the direction of the current is inwards. He examined the stone-canal cut out of the living animal; I have confirmed his result by a somewhat more satisfactory method. I kept Amphiura squamata living for several days in sea water, carrying in one case carmine, and in another ~ lamp-black in suspension; and on cutting sections I found these particles in the pore-canal, and in some cases apparently ingested by the cells lining it. In view of Ludwig’s researches Cuénot comes in a later paper (4) to what I believe to be the correct solution of the question of function. He there suggests that the flagella lining the stone-canal are always tending to produce an inward current, and that thus the turgidity of the whole water-vascular system is kept up. [This is practically the old view; except that he does not assert a continuous inward current.— December, 1895. | It is obvious from the structure of the valves of the tube- feet that, in consequence of the ambulatory movements, there must be a slow loss of fluid. The ampulla and the tube-foot are shut off from the canal leading into the radial water-vascu- lar canal by a pair of valves opening only inwards. Conse- quently during the contraction of either ampulla or tube-foot the two act together as a closed system, since no fluid can escape into the radial canal. The existence of the valves however shows clearly that fluid occasionally enters the tube-foot, and this can only be rendered possible by a slow loss of turgidity owing to the osmosis of the contained fluid when under pressure. This is confirmed by considering the case of Ophiurids, where (except in the Astrophytidz), the tube-feet having lost their ambulatory function, the madreporite has only one or at most two pores, and the calibre of the stone-canal is exceedingly narrow. The dermal branchie arise when the star-fish has reached a diameter of about 1°5 millimetres (R equal ‘85 millimetre). We see that the branchia is only a very thin piece of the body- wall produced into a finger-like process (Pl. XVI, fig. 98). Around the base of the branchia is a peribranchial space lined by flattened epithelium: this space, as Cuénot has rightly 264, E. W. MACBRIDE. observed, is the only one of the great “‘schizoccelic” spaces which Hamann (8) has described in the body-wall which has any real existence, the others being merely artefacts produced by the process of decalcification. I have found one specimen showing the first trace of a dermal branchia (figs. 96 and 97). We see a slight thickening of the peritoneum, and above it the peribranchial space. Fig. 96 shows that the latter is a diverti- culum of the coelom. As I have only one section illustrating this I do not speak with absolute certainty on the point; but, with this possible though very improbable exception, there is no schizocele whatsoever in Asterina gibbosa: all spaces lined by epithelium are of celomic origin. Histological Differentiation. The cells of the gut-wall have undergone some change since the close of the metamorphosis. Specimens of the epithelium from different regions are given in Pl. XIX, figs. 129—182. These are all taken from a young adult in which R equals ‘85 millimetre. The cells of the lateral walls ofthe stomach (i.e. the adult cesophagus) have become exceedingly long and narrow; their outer ends are refracting and take a light yellow tone with osmic acid (fig. 129). The cells of the aboral wall, on the contrary, have developed numerous gland cells filled with globules; interspersed amongst them are some very narrow filamentous cells. Fig. 130 shows the spot marked X where the stomach opens into the pyloric sac and the abrupt change in the character of the epithelium. The pyloric sac is lined by uniform columnar cells; the nucleus is generally near the base of the cell, and is never placed further up than the middle, and the protoplasm is uniformly granular (fig. 181). The cells lining the rectal cecum (fig. 132) are similar in form but smaller, and the protoplasm is clearer, with the outer part more refringent. It is at least a plausible suggestion that the gland cells of the stomach secrete the poison which paralyses the prey, and that the cells of the pyloric sac give rise to a digestive ferment. The differentiation of tissues which has gone on in the THE DEVELOPMENT OF ASTERINA GIBBOSA. 265 body-wall is illustrated in Pl. XXT, figs. 146 and 147. These sections are taken from young adults in which R equals -4 mm. and ‘86 mm. respectively, and they pass through the same region as fig. 145, which is from a larva in Stage HE, and which we have already described. In fig. 146 we see that the mus- cular fibres of the muscle we may call the dilator ani are still connected with the peritoneal cells; but in fig. 147 they have become quite distinct, and the cells of the peritoneum have become quite flattened. The ectoderm has entirely changed its character, the numerous larval goblet cells have almost disappeared, and thecells in general have become shorter; many of them are inversely wedge-shaped, and are apparently about to become converted into gland cells, probably of the same histological character as those of the abora] wall of the stomach. Here and there is a narrow cell ending in a fine hair, one of the scattered sense-cells of the aboral surface ; these are shown in fig. 148, a piece of ectoderm from another individual of the same age. All observers agree in maintain- ing that the ectoderm of the adult retains its ciliated covering; but though I have been able to make out easily the cilia, or rather flagella of the metamorphosing larva, I have not been able to do so with any certainty in the aboral wall of these young adults. Probably the cilia are very delicate and fragile. The tissues of the mesenchyme have undergone marked differentia- tion. So far as my researches have extended it seems that three fates are open to mesenchyme cells, all of which are illustrated in fig. 147. They may remain practically unchanged as amoebocytes or wandering cells (ameé.), or they may become embedded in bundles of fibres so as to form connective-tissue cells (the fibres being intercellular, not outgrowths of cells) ; or, finally, they may fuse to form a syncytium having the form of a meshwork (calc.). This is the skeletogenous tissue; lime is deposited in the interstices of the meshwork. There is a fourth fate, which is not reached by any as far as I have gone, but which obviously must be the lot of some, and that is to form the muscles moving the spines or rudimentary pedicellariz. The superficial position of these muscles renders it exceedingly 266 E. W. MACBRIDE. unlikely that they are of peritoneal origin, and their position in other Asterids where, as in Asterias, for example, they occur on the skin covering the spines, growing even from their tips, makes such a supposition almost impossible. There- fore we must postulate some muscles of mesenchymatous origin for Asterina, although all those which I have examined are of epithelial origin. The development of the nervous system has advanced greatly, and has reached, as soon as the metamorphosis is complete, its final form ; this is shown in fig. 141, taken from the same specimen as fig. 146. The ectoderm cells have increased immensely in number, and become excessively filamentous, so that the nuclei are many layers deep; the fibrillar layer has increased very much in thickness. It is traversed by vertical fibres which sometimes branch and sometimes have small nuclei on them; these are in continuity with the ectoderm cells, but are probably of non-nervous character. Sections parallel to the disc show that numerous little bipolar cells are embedded in the mass of fibrils (Pl. XVII, fig. 109, dip. gang.). Since these cells are not present in the just metamorphosed form, they must be ectoderm cells which have passed in, and occasionally one sees a cell just at the boundary of the fibres apparently in the act of passing in. The perihemal spaces become closely apposed to the nerve-cord, no mesenchyme being left between (ph. fig. 141) ; the vertical fibres do not, how- ever, arise in connection with the epithelium of these cavities, since they are present before this close apposition takes place. Cuénot states that all the ectoderm cells of the nerve-cord end in the vertical supporting fibres described above. This is a bold statement which it is quite impossible to prove by sections, and which is most improbable. As a matter of fact these vertical fibres are not present in nearly large enough number to account for all the ectoderm cells ; and Hamann’s statement (8) is probably correct, that many of these end in fine processes which lose themselves in the mass of fibrils. The sense-organs of Asterina are all developed in connection with the appendages of the water-vascular system. The eye THE DEVELOPMENT OF ASTERINA GIBBOSA. 267 arises at the base of the terminal tentacle of the radial canal ; two stages in its development are given in Pl. XXI, figs. 142 and 143. In the first we see a simple ectodermic involution ; in the second we see a pit surrounded by columnar cells, pro- bably retinal, and filled up by closely fitting polygonal cells, which correspond to the layer of vitelligenous cells in an Arthropod eye. The existence of these cells has been vigor- ously denied by Cuénot (3), who maintains that we have only polygonal cuticular plates. My sections, however, remove all doubt on the subject; they show,with perfect clearness that we have to do with cells, and their nuclei can be madé out. This pit is the first only of the numerous pits which cover the “ eye ”’ of the adult, which is really essentially a small rounded swelling at the very tip of the radial nerve. The method of preservation employed seems to have dissolved the pigment. The remaining sense-organs are the tips of the tube-feet and the terminal tentacle. A longitudinal section of a tube-foot is given in Pl. XXI, fig. 150. This is taken from a specimen in which R equals ‘4 millimetre, but it holds true for specimens of a radius of a millimetre or more,— that is, for probably the first two months after the metamor- phosis. Comparing it with fig. 149, a similar section taken from a larva in Stage F, we see that the ectoderm at the tip has become thickened, and underneath it we can make out on each side a mass of nerve-fibrils. A powerful nerve leaves the radial nerve-cord to supply each sense disc; it would be more correct to speak of these branches as actual prolongations of the nerve-cord with its cells and fibrils; they are, indeed, the only conspicuous branches which it gives off. Some of the ectoderm cells of the sense disc have a peculiar regular cylin- drical form, which recalls that of the retinal cells. The facts above related justify the view that the whole radial canal with its tube-feet is to be looked on as one large branched tentacle, the main function of which was probably originally prehensile and therefore also sensory; and since a plexus of nerve-fibrils is in the adult found under the ectoderm all over the body, the central nervous system may be said to be a local 268 BH. W. MACBRIDE. concentration of this in the neighbourhood of a greatly deve- loped sensory tentacle. The support of this tentacle by the arm is a secondary matter, as we have already learned—a fact which comes out still more clearly in Crinoid development. There the primary hydrocele lobes develop into long free tentacles covered with sensory hairs. At a very late period (later than any which Seeligerobserved) these primary tentacles, according to Perrier (17) become applied to five outgrowths of the body-wall; these latter immediately bifurcate to form the ten arms, and so the free tips of the tentacles are situated each in the angle between a pair ofarms. Seeliger (18) adduces this last fact to show that the primary tentacles are not the same as the primary hydrocele lobes of Asterids, forgetting that the point where a pair of arms diverge corresponds to the tip of the Asterid arm, since in Antedon there are ten arms which have arisen by dichotomy from five. The epithelium of the water-vascular system in fig. 150 shows an interesting feature; the cells have developed muscular tails which are arranged longitudinally, and the important point is that these myo-epithelial cells persist as such for a considerable period of free life. Pl. XXII, figs. 151—154, show us that the aboral wall of the perihemal space also gives rise to muscles. These connect one ambulacral ossicle with its fellow of the opposite side, and serve, by approximating these to one another, to close the ambu- lacral groove. Figs. 151 and 152 show us that here again we have, in the first instance, to do with myo-epithelial cells. Muscles connecting one ossicle with its successor and prede- cessor are also present, but very much more feebly developed. In Ophiurids, however, as is well known, they are most power- ful, and this point gives the key to nearly all the peculiarities of this group as compared with Asterids. Presuming, as we fairly may, that these muscles are developed from the peri- heemal wall as in Asterids, we are brought face to face with a most interesting effect which this produces on the nervous system. Fig. 156 gives a section of the radial nerve-cord of an Ophiurid. We notice two great masses of cells and fibres on THE DEVELOPMENT OF ASTERINA GIBBOSA. 269 the aboral side of the nerve-cord, and Hamann (8) has shown that the nerves for the ambulacral muscles arise entirely from these. Now it has been for a long time suspected, and Cuénot has finally proved it (4), that there is a similar but feebler develop- ment of what we may call ‘‘ccelomic nervous tissue” takes place in the Asterid. None of my specimens were old enough to show this, though one can see (fig. 141) that the perihemal epithelium has come into intimate connection with the nervous matter. Pl. XXII, fig. 155, represents a transverse section of the nerve-cord of a young Asterias; we see in it that this epithelium has become thickened on each side of the median septum; one requires, however, a section of the nerve of a fully grown adult to see the celomic nervous fibrils. So we may say that from their aboral wall the periheemal spaces give rise to muscles, and from their oral wall to the corresponding nervous tissue. I ought to mention in this place that Cuénot describes a canal leading from the perihemal space into the ccelom at the level of each ambulacral ossicle; also five pores leading from the outer perihemal ring to the celom. If these communications exist, they are certainly secondary, as there is no trace of them in my specimens; but as Cuénot’s results were founded on injection I am exceedingly sceptical as to the existence of such openings. I have said above that the increasing importance of the ambulacral muscles is the explanation of the evolution of Ophiurids from Asterids. The Ophiurids have substituted the quick powerful movements of these muscles for the slow motions of the tube-feet. In correlation with this change the nervous system has become better developed, the radial cords becoming gangiiated, and the whole is removed from the ex- terior by invagination, and thus the subneural space is really a neural canal. The ambulacral ossicles have become firmly united, each to its fellow, to form a series of vertebre, and thus the cavity of the arm is reduced, and this, with the simpler food, accounts for the disappearance of the pyloric ceca. We have already pointed out that the lessened activity of vou. 6. E 270 EB. W. MACBRIDE. the tube-feet, consequent upon the loss of the locomotor function, explains the reduced stone-canal and madreporite, though probably their increased sensitiveness has helped in ~ developing the nervous system. Literature consulted. An account of the earliest publications on Echinoderm de- velopment is not given here, since a résumé of them will be found in the papers I quote; and I hold it to be a waste of time to reiterate with each new paper the whole history of the growth of our knowledge ab initio. I mention here only those authors on whose results I have, so to speak, built, or from whom I have found it necessary to differ. Ludwig’s work on the anatomy of Asterids (10) laid the foundation of our know- ledge of the hemal and perihemal systems ; though, as we have seen, many of his ideas about these structures were incorrect. Subsequently in treating of Ophiurids (11) he discovered the genital rachis. Hamann (7) extended this result, and pointed out the ameboid nature of the primitive germ cells. Then we had Ludwig’s great work on the development of Asterina gibbosa (12), the first account of the metamorphosis of any Echinoderm which had any pretence of completeness, and to which I have constant occasion to refer. His account of the changes in external form and of the developmeut of the calcareous plates can hardly be improved upon. Owing, however, to the imperfect methods in vogue at that time he failed to penetrate with equal success into the course of the internal changes. He saw nothing of the segmentation of the ceelom or of the ring-like growth of the left coelomic vesicle ; he saw nothing also of the origin of genital organs, ovoid gland, or oral celom. He did not observe the right hydroceéle or find the origin of the perihzmal spaces. He missed the fixed stage, and he does not seem to have had any clear con- ception of the relation to each other of the larval and adult planes of symmetry. We owe to him, however, the clear distinction of pore-canal and stone-caunal, and the recognition of the fact that the pore-canal is completely independent of the THE DEVELOPMENT OF ASTERINA GIBBOSA. 271 hydrocele. Bury (1) may be said to have introduced modern conceptions of Echinoderm development by his work on the development of Antedon; there he distinguished between an- terior ceelom and hydrocele, and showed that the stalk was the preoral lobe. Then he made a series of observations on Echino- derm larvze (2), and showed that generally speaking the cclom on each side becomes segmented into two vesicles, an anterior and a posterior. He, however, regarded the hydrocele as an essentially unpaired structure, an outgrowth from the anterior celom, and was greatly distressed to find that it originated from the posterior vesicle in Ophiurids, and that in Asterina the stone-canal, which in other forms represented the original neck of communication between anterior celom and hydrocele, was apparently an independent perforation. The Jast difficulty has been answered by Ludwig;! as to the former, the proof I have brought that the hydroceele is paired shows that there are really three primary divisions of the cwlom on each side, viz. the anterior celom, single in Asterina, but primitively paired in Asterias; the right or left hydrocele, and the posterior celom (right or left as the case may be); the apparent forma- tion therefore of the hydroccele from the anterior or posterior vesicle is a mere question as to whether the septum between the posterior ceelom and the hydrocele or the septum between the hydroccele and the anterior ccelom is formed first. In speaking of the Bipinnaria, Bury says that in a future paper he intends to prove that the anterior ceelom becomes the axial sinus, but up till now he has published nothing further on the subject.2 He made a few observations on Asterina 1 Bury had not seen the stage of development when the stone-canal is an open groove. 2 Since the preliminary account (15) of the present paper was published, a paper on the “‘ Organogeny of Stellerids,” by M. Achille Russo, has appeared in the ‘ Atti della Accadema reale di Napoli’ for 1894. In this work (to which I only obtained access some considerable time after the present paper was finished) M. Russo gives a description of the ontogeny and anatomy of the ovoid gland and axial sinus in Asterina gibbosa and an Ophiurid. He combats my statements about the origin of these structures in Amphiura squamata. The origin of the axial sinus in Asterina has been correctly de- scribed; it is about the only thing that is correctly described in the paper. 272 E. W. MACBRIDE. larvee of Stage D, and saw the completely closed coelomic vesicle on the right, and the imperfect transverse septum on the left side, and was at a loss how to interpret these appearances ; the right hydroccele he calls a mesenchymatous vesicle. It is curious to see how unable many zoologists have been to grasp Bury’s idea of the anterior celom; thus Seeliger, who has confirmed his work on Antedon and amplified it till it may be said that we have an exhaustive knowledge of the subject, objects to consider the structure Bury named anterior celom as such, on the supposition that Bury meant by that a fellow of the hydroceele, which it obviously is not. Seeliger calls it the “‘ parietal canal,’ but the structural facts he so accurately relates are convincingly in favour of Bury’s inter- pretation. The weak point in Bury’s observations on Plutei and other larve was that in no case were any more than a few stages taken at random examined; but I hope the account I have given in this paper will provide a more solid basis for the idea of segmentation of the celom in Echinoderms. Field (5) has published a short paper on the development of the Bipin- naria; he carriesit up only to a stage corresponding to midway between Stages B and C of Asterina. The chief points of interest in the paper are that many of the larve had two madreporic pores, and he suggests that this is a normal stage in the ontogeny ; also that the two ciliated rings characteristic of the Bipinnaria are derived from one, and that there is a preeoral sense-organ comparable to that in Antedon. This paper does not contain the discovery that the water- vascular rudiment is paired; for, as a matter of fact, in the oldest. larva examined no trace of the left hydrocele was present. The “ schizoccelic space,”’ near the madreporic pore, may represent the rudiment of the right hydroccele; needless to say, it was not recognised as such. Theel (22) has recently succeeded in following the meta- morphosis in Echinocyamus pusillus so far as the external features are concerned. He finds that already in the blastula M. Russo’s technique was obviously not equal to dealing successfully with such difficult subjects as Echinoderm larve. THE DEVELOPMENT OF ASTERINA GIBBOSA. 273 a przoral sense-organ is present; this subsequently becomes incorporated with the ciliated ring, and if this organ is homo- logous with that of the Bipinnaria, we may conclude that the ciliated band of the Pluteus corresponds only to the posterior of the two bands of the Bipinnaria, since in the Bipinnaria the sense-organ is situated between preoral and post-oral ciliated bands, and this spot corresponds to a constriction in the original longitudinal ciliated ring, not to a position on its anterior edge. Our knowledge of Echinoderm histology is largely due to Hamann (8) and Cuénot (3 and 4). The latter, as we have seen above, was the first to suggest that the ovoid gland gave rise to the genital rachis. The first account of the development of ovoid gland and rachis is given in my paper on Amphiura squamata (14), and I have there collected the fragmentary notices on this subject, which had till then appeared. [I regret that when I sent in this paper for publication I did not mention the well-known paper of Metschnikoff (‘ Studien uber die Entwickelung der Echinodermen und Nemertinen,” ‘Mémoires de l’Académie Impériale de St. Pétersbourg,’ tome xiv, No. 8), in which he describes a right hydrocele in Amphiuriasquamata. He there says that the right coelomic vesicle becomes divided into anterior and posterior portions just like the left; the anterior portion sometimes atrophies but sometimes develops into a regular five-lobed hydrocele. It has been the fashion to ignore this work, since it was not accomplished by modern methods; but after my experience with Asterina I feel morally certain that Metschnikoff was right, though of course he did not distinguish between hydroceeles and anterior celom. Bury (2) seems to have missed the importance of this observation.—Dec., 1895.] General Considerations. On reviewing the developmental history recorded in this paper, two main questions present themselves: first, what light does it throw on the affinities of the Asterids with other Echinoderms? and second, does it suggest any direction in 274 E. W. MACBRIDE. which we may look to find the origin of the group Echino- dermata as a whole? In answer to the first question, we must observe that the stalks of Asterina and Antedon are morphologically equivalent,} both being formed from the preoral lobe, and, so far as one might judge from the different shape of the latter in the two cases, the adhesive discs by which they fix themselves are situated in precisely the same position. Now no one doubts that Antedon had a fixed ancestor; it is, in fact, one of the very few Crinoids which do not remain fixed throughout their whole life. If Asterids ever had an ancestor in common with Crinoids which could be called an Echinoderm at all, it must have been one represented by the fixed larva of Antedon before it has fully acquired radial symmetry, since, as we have already pointed out, the metamorphoses of Antedon and Asterina pursue different courses. In the first case the mouth is shifted backwards and upwards, and a precisely similar thing happens to the larve of Entoproct Polyzoa, Ascidians, and Cirri- pedes when they fix themselves. In the second case, how- ever, the disc is flexed obliquely downwards on the stalk, so that the left coelomic sac and the hydrocele both come to encircle the base of the stalk ; and as consequence the aboral poles in the two cases are not homologous, for in the first case this pole is the cicatrice left by the rupture of the stalk, whereas in the second case the point where the stalk passes into the disc is quite remote from the aboral pole. The apparent correspondence of the calcareous plates of the calyx in Antedon and the so- called calyx in Asterina is simply due, in my opinion, to the 1 Since the present paper was sent in for publication, my attention has been called to some observations of Perrier’s which I regret having overlooked. In his account of the Echinoderms collected by the ‘ Mission Scientifique du Cap Horn,” he describes the larve of Asterias spirabilis, which adhere to the buccal membrane of the mother. They are attached by a pedicle which Perrier compares to the stalk of the Antedon larva and to the preoral lobe of the Asterina larva, He points out that both in the case of Asterias spira- bilisandof Asterina gibbosa the pedicle arises from the oral surface, whereas in Antedon it is aboral in its origin, but he offers no explanation of this dif- ference in position. THE DEVELOPMENT OF ASTERINA GIBBOSA. 275 fact that their arrangement is in both cases dominated by the prevailing pentamerous symmetry of the adult. The reason why the change in the position of the mouth takes place in Antedon is that this animal, like the others in which a similar change occurs, feeds on swimming or floating prey, and, so to speak, turns the mouth upwards to receive it. Asterids and their allies, on the other hand, find their food on the substratum, and therefore we must suppose that in the fixed ancestor of Asterids the body was flexed downwards so as to bring the substratum within reach of the tentacles. The difficulty suggests itself that a fixed form finding its food on the substratum might very soon devour all within its reach ; and the suggestion may be made that perhaps the ancestor of Asterids never was fixed, but that the divergence from Crinoids took place when the common ancestor was a creeping form, since we may reasonably conclude that creeping habits formed the transition stage between a free-swimming and a fixed mode of life. In this case, however, the difficulty meets us of accounting for that radial symmetry which is so deeply impressed on the organisation of Asterids and other forms. It would be rash to say that fixed life is the direct cause of radial symmetry when we consider the case of Brachiopods, Cirripedes, &c., but this symmetry is only, so far as our knowledge goes, developed in connection with a fixed life. The proximate cause of the radial symmetry of Asterids is the immense preponderance of the organs of the left side, and it is difficult to see how this could have gone on to the extent it has done in an animal which moved about with a definite part directed forwards. The motion of the Asterid when metamorphosed is vague,—that is, any part is directed forwards; and it seems to me that a fixed stage must intervene between this and the mode of motion in which the head went first. ' Some might object that Ctenophores and Meduse are radially sym- metrical, but the first are not truly so; and as to the second, I hold very strongly the view that the Medusa is only a specialised bud, which has secon- darily acquired locomotive powers in order to disperse the ova. Its radial symmetry has been inherited from fixed ancestors. 276 E. W. MACBRIDE. Therefore I feel that we are shut up to the supposition that Asterids had a fixed ancestor, and we must suppose that this form lived under conditions where enough food drifted along the bottom to meetitsdemands. Pl. XXII, fig. 157, represents the characters which I consider the common ancestor of all Echinoderms possessed when it became fixed. Figs. 158 and 159 show how these characters became modified in the cases of the Asterid and Crinoid respectively. It is probable that a fixed stage occurs in the life history of all Asterids. The larve of Echinaster and Asterias Miilleri, which are carried in brood-pouches, certainly possess one, and the three papille on the Brachiolaria larve are generally interpreted as an apparatus for fixation. The fixed stage has, however, been lost so far as we know in all other Echinoderms; and it is instructive to note in this connection that Asterids alone retain the great przoral lobe. This has completely atrophied in the Plutei both of Ophiurids and Echinids ; and in the latter case, as I have indicated above, (page 273) there is some evidence to show that a preoral ciliated band has likewise disappeared. The Auricularia still retains a trace of the preoral lobe, and it has been regarded as an ex- ceedingly primitive form because it retains the undivided lon- gitudinal ciliated band, and because the larval mouth becomes the adult one. The internal anatomy of this larva shows that, except in these two points, it is the most modified of all; the anterior ceelom so conspicuous in the Bipinnaria is represented, as Bury has shown (2), by a bud of cells which forms the secondary madreporite on the stone-canal, and the whole mode of segmentation of the celom is most erratic. I have dwelt on this subject at some length because some have regarded the Holothurids as the primitive group of the Echinoderms, and Sémon (19) has even attempted to show that the primary hydroceele lobes in them became the oral tentacles, whilst the so-called radial canals were really interradial out- growths. Ludwig (13) has, however, shown the incorrectness of this; in the Synaptide alone do the oral tentacles spring direct from the ring-canal, and it was the development of THE DEVELOPMENT OF ASTERINA GIBBOSA. 277 Synapta on which Sémon based his theory. In all other Holo- thurids the buccal tentacles spring like the buccal tube-feet of Echinids from the proximal portion of the radial canals. It is, however, difficult for me to see how anyone can doubt that the Asterids are the least modified group of the Echinoderms. I have already dealt with their relations to Ophiurids, and have also pointed out that the Asterid central nervous system is really a concentration of the diffuse nervous plexus in connection with what must be regarded as a great sensory tentacle,—that, in fact, the whole radial water-vascular canal is to be regarded as a pinnately branched tentacle for which the arm is a secondary support. Sémon himself has suggested this (20), and it comes out even more clearly in Crinoid development than in the case of Asterids. Now the long radial canals in Echinids, ending in degenerate sense tentacles, clearly at one time had arms to support them; but these supports have been drawn back into the body. The Holothurids have been probably derived from the primitive Echinids; their calcareous nodules are most likely plates and spines atrophied in order to allow of free muscular movement. The terminal sense tentacles of the radial canals have entirely disappeared, and the forward shift of the madreporite and genital opening is no more difficult to understand than the varying position of the anus in Hchi- nids. In the Asterids alone is locomotion entirely dependent on the tube-feet, and in them only we have the nervous system exposed. On the second question, viz. that of the affinities of the Echinodermata as a whole, much light is thrown by the development of Asterina gibbosa. It is of course well known that the Tornaria larva of Balanoglossus shows a strong resemblance to the Bipinnaria in the course of its ciliated bands, and in possessing a preeoral celom opening by a pore on the left. The adult Balanoglossus has five coelomic cavities, and Bateson has shown that these arise as separate pouches of the gut. The question arises whether it is legitimate to homologise with these the five coelomic cavities of the Asterina larva which arise by division of pouches already formed, but 278 E. W. MACBRIDE. which still persist in the adult as sharply separated cavities, only the most posterior pair, viz. the right and left posterior ccoeloms (oral and aboral) of the adult having partially fused with each other. The development of Antedon seems to answer this question in the affirmative. In its case the hydroceele is budded off quite independently of the posterior ccelomic sacs. Adopting, then, the view that the ceelomic sacs of the Ente- ropneusta and Asterids correspond, we find that the hydroceele represents the collar cavity. Now in Cephalodiscus the collar cavities are produced into long pinnately branched tentacles, comparable to the radial water-vascular canals, and further a branch from the central nervous system accompanies each tentacle, just as the radial nerves accompany the radial canals in Echinoderms. Now, if we suppose that the two hydroccles of Asterina were equally developed and approximated in the mid-dorsal line, the fusion of the anterior portion of the two nerve ‘rings,’ which of course would in this case be only open curves (since a ring-form is attained through the preponderating growth of one side) would give rise to a mid-dorsal nervous system like that of Cephalodiscus. Nor is that all; Professor Spengel (21) has shown in his monograph of the Enteropneusta that the currents in the proboscis-pore and collar-pore are inwards, and that by this means the animal inflates the proboscis and collar so as to render them efficient locomotor organs. We have seen that the function of the stone-canal is a similar one. We conclude, then, that the free-swimming ancestor of Echinoderms, for which we may adopt the name Dipleurula, and the Tornaria ancestor of Balanoglossus, were closely allied, This involves the assumption that they were allied to the Pro- tochordata, for, as I have elsewhere stated (16), Professor Spengel’s attempt to refute the Chordate affinities of Balano- glossus has been, in my opinion, futile. Although it may seem somewhat fanciful, I cannot help seeing hints of Vertebrate peculiarities in the anatomy of Echinoderms. Where else among all animals of higher grade than the Celenterates do we find calcareous ossicles in the dermis? Where else THE DEVELOPMENT OF ASTERINA GIBBOSA. 279 is the removal of the nervous system from the surface effected by invagination leading to the formation of a neural canal ? When we come to try and picture the characters which the Dipleurula possessed, we see at once that it must have been far more primitive than any existing form. In point of fact an Asterid is about the most undifferentiated animal above the level of Coelenterates which exists. No proper blood-vessels, no specialised excretory organ, a central nervous system which is really a local concentration of a diffuse skin plexus, perfectly simple generative ducts, a most feebly developed muscular sys- tem, the fibres being for a considerable time simply myo-epi- thelial cells,— where is such a state of things to be found outside the Coelenterata? When we further add that in the Crinoid the ambulacral nervous system nearly atrophies in the adult, and is replaced by a new system developed in a totally different position, we see that we are at about as low a level as one could well imagine, since the central nervous system in all higher forms is a most persistent structure. Assuredly Platyhelminths, which have been usually regarded as the basal group in the Celomata, or better, Triploblastica, are far more highly specialised. To say nothing of their cephalic ganglia, we have their highly developed muscular wall and their complicated excretory and genital organs to prove this. We shall not, then, go far astray in assigning the Dipleurula and the Tornaria to a group, the Protocelomata, which were not far removed from the Ccelenterates ; the colom was divided into three parts on each side, but of these the most an- terior were usually fused to form an unpaired vesicle. The Dipleurula differed from the Tornaria chiefly in possession of an aperture, the stone-canal, in the wall separating the proboscis ceelom from the collar celom. This may have been the primi- tive arrangement, or it may have been a secondary arrangement acquired in consequence of the Dipleurula having lost the collar- pores, one of which may, however, as we have seen, be developed as a variation in the Asterid larva. At the apex of the przoral lobe was a more or less developed sense-organ with associated 280 E. W. MAOBRIDE. nervous tissue. The collar cavities were probably prolonged into tentacles with which nervous tissue was associated. If this supposition is correct, the group Protocceelomata was a pelagic cosmopolitan one, and it isin accordance with what we know of wide ranging groups that some of its members should adopt changed habits and modified structure. The Echino- dermata, then, represent the earliest offshoot which took to a sessile life and acquired radial symmetry. A little later the Hemichordata branched off, a burrowing life being adopted and consequent degeneracy resulting. The main stem, how- ever, remained pelagic and gave rise to the Chordata. The Ascidians were the next offshoot, and then came Amphioxus. We see, therefore, that the track of the great Chordata phylum through past ages is traced by examining those of its members who at very different periods of its history, and at different stages in its evolution, have forsaken their high vocation, and taken to a sessile or burrowing life, with the inevitable consequence—degeneracy. The following diagram may represent these relationships a little more clearly : Protocelomata Dipleurula Hemichordata (‘Tornaria). Fixed ancestor Balanoglossus, Cephalodiseus. of Echinoderms. Crinoids. Protochordata. Asterids. Ascidians. Protoechinids. | Ophiurids. | Amphioxus. Echinids. Holothurids. Vertebrata. THE DEVELOPMENT OF ASTERINA GIBBOSA. 281 I hope in a future paper to be able to show that the Trochophore larva is also related, though much more distantly, to the Dipleurula. Zoological Laboratory, March 8th, 1895. Cambridge. [Postscript.—With reference to the point discussed on p. 275, viz. the existence of a fixed Echinoderm in which the tentacles were directed towards the substratum, it is of interest to note that in the fossil Crinoid Cheirocrinus the calyx was sharply bent on the stalk, so that the arms were directed downwards towards the substratum. | List or WoRKS REFERRED TO IN THIS Memotrk. 1. Bury, H.—‘The Harly Stages in the Development of Antedon rosacea,” ‘ Phil. Trans. Roy. Soc.,’ 1888. 2. Bury, H.—Studies in the Embryology of Echinoderms,” ‘ Quart. Journ. Mier. Sci.,’ 1889. 8. Cunnot, L.— Contributions 4 lEtude anatomique des Asterides,”’ ‘Arch, pour Zool. Exp.,’? 2me series, tome v bis. 4. Coinor, L.— Etudes morphologiques sur les Echinoderms,” ‘ Arch. de Biol.,’ tome xi, fascicles 1 and 2. 5, Frerp.— The Larva of Asterias,” ‘ Quart. Journ. Micr. Sci.,’ 1892. 6. Garstanc, W.— From the Philosophical Transactions of the Royal Society, Vol. 186, 1895. 6 From the Proceedings of the Zoological Society of London, 1895. 7 From the Quarterly Journal of Microscopical Science, Vol. 38.