{ 4 WENA sis t cwere! ane Cee et au aan vat) " Hilo oa ee te ie yh pert a ta I Hai LidqadsSeaeat Ht any sarily a Tara jibe yt ht aeae Vio hes bina Peak doatys peas igh tines uh isa} i i ie! bol athe oat rat Per teaearaeh tereees: awa it oi stat pent pectal fi te Vhs rt ina: Arts ” ieee ‘ ah a Tari eh Baa ai i ce J Na Mit 0 j a ‘ omits agee eee n " i fi CaM nt) ences a Sea ante? i ie die iter ty ies + iad 11m) Hater i i iaaetdene eau aban rm ty eset Mit a x aed ot tbe 4 Caeti Joke the! erie " ished rates 4 guess ae ett pits eo ah i esas rants ‘eager id, ey yi boo? fe “sh ‘ rome tit re tesla milatittte pete Hp aeahas at t ieee ay ie ‘ Pl * ee ta dh fig i ‘ ih inate pao : ase dw ¥ < rae, eae bytes ty ee ; setehh Peace pane kale . Sipeg to Me oe had bare wy gl : cut : et bel Ad akan ae eared eo ios mad arr vw et et Ve or) apy kets erp agetier on ot) te oe Sen pe ve pe pe ih Fang tine tame eats be Sy eh ant eran parses sabei @vodcoben iad se ee lak aol - ches ph rtet tas sins rerrie etches Loneeoasenel - a Ayko pers 5 hee ICIS Ee nesta hey start whet bites Pelatattes TRANSACTIONS OF THE ROYAL SOCIETY OF VICTORIA. Voted, PART: T: (Pranms, 1, 25.35: 45.00; 6..] THE ANATOMY OF MEGASCOLIDES AUSTRALIS (THE GIANT EARTH-WORM OF GIPPSLAND.) BY W. BALDWIN SPENCER, B.A. PROFESSOR OF BIOLOGY IN THE UNIVERSITY OF MELBOURNE. MELBOURNE: PUBL'SHED FOR HE ROYAL SOCIETY By STILLWELL & CO., 78 COLLINS STREET EAST. 188&. } ' ‘ . oP f ’ P zi \ ie P 745° {* HAS been decided that the Society shall, in future, issue two separate publications, of which one shall be called the « Transactions,” the other the “Proceedings of the Royal Society of Victoria.” This forms the first part of the first volume of the “ Transactions.” August, 1888. I. On the Anatomy of Megascolides australis, the Giant Earth-worm of Gippsland. By W. Bapwin Spencmr, B.A., Professor of Biology in the University of Melbourne. [Pirates 1, 2, 3, 4, 5, 6.] HIstToRICAL. Tue first figure and description of this, which is probably the largest earth-worm yet known, were those given by Professor McCoy,* to whom it had been sent from Gippsland. ‘This description is unfortunately incorrect, the worm being placed in a wrong family, and the first correct one is that given by Mr. Fusrcuer, in his “ Notes on Australian Earth-worms.”t To this author we are indebted for a large series of careful studies, so far as their macroscopic anatomy is concerned, of Australian Earth-worms. In November 1887, this paper, on the anatomy of Megascolides australis, was read in abstract before the Royal Society of Victoria; I was unaware at that time that Mr. Fiutcumr had in September, two months previously, described its macroscopic anatomy, and our work was carried on quite independently. Before this Mr, Firrcumr had described two species of HEarth-worms, which he placed in a new genus, Notoscolex, and in the paper which contains his description of the worm with which we are now dealing, the latter is described as Notoscolex gippslandicus; at the same time Mr. Fiercuer described two other new species belonging to the same genus, viz, NV. tasmanicus and N. tuberculatus. { It is unfortunate, masmuch as we owe our knowledge of this very interesting genus almost entirely to Mr. Frzrcusr, that from reasons of priority his name of Notoscolex must give way to that of Megascolides, the name first given to the type of the genus which now includes five species. The characters of the genus, as given by Mr. Fiercumr, are as follows:— Megascolides McCoy (Notoscolex Fietcumr). Intra-clitelian worms, with clitellum, comprismg some or all of segments XITI.—XXITI.; male pores two, on segment XVIII, on papille, in a line with * Prodromus to the Zoology of Victoria. Dec. 1, pl. 7. + Proc. Linn. Soc. N.S.W., 28th Sept., 1887. See also the same author's series of ‘“‘ Notes on Australian Earth-worms” in the same journal. t WV. camdenensis and NV. grandis. See Proc. Linn. Soc., N.S.W., 30th June, 1886, pl. VIII, pages 1-6. B 2 ita enamine RS a i a aS i 4 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, the intervals between the inner couples of sete; oviducal pores on XIV.; setze in eight longitudinal rows. HApsitat. In September 1887, and at subsequent periods, I procured many specimens of Megascolides australis in the neighbourhood of Warragul and Brandy Creek, - and my thanks for valuable assistance are due to Messrs, Hueu and Henry Coprstann, and to Messrs. C. A. Topp and A. H. S. Lucas; without their kindly aid it would not have been possible for me to have secured entire specimens of the worm. I have also to thank the Rev. W. Manreuu for the gift of the first specimen of the worm which reached me, and upon the examination of which my preliminary account, presented to the Royal Society in October 1887, was based. The worm, which is known as the “ Giant Harth-worm of Gippsland,” appears to be confined to the latter districts. The other species of the genus are found, one in Tasmania, two in Camden, N.S.W., and one in Gippsland. None is as yet reported from the Western or Northern: parts of Australia, though probably species of the genus will be found in all portions of the continent when carefully searched for. Of all the species yet known, this one seems to be the largest, and is apparently confined to Gippsland ; it is, when found at all, somewhat abundant, and lives principally on the sloping sides of creeks. At times it is found beneath fallen logs, and may be turned out of the ground by the plough. When first seeking it, we were somewhat puzzled by some of those who were evidently well acquainted with the worm assuring us that the entrance to its burrow was indicated by a distinct “casting ;” whilst others, evidently equally well acquainted with the animal, were quite as positive in asserting that it never produced any “casting.” Whilst searching, we found what I believe to be the explanation of these contradictory statements, and soon discovered that the surest test of the presence of the worm underground was a very distinct gurgling sound, made by the animal retreating in its burrow when the ground was stamped upon by the foot. When once heard this gurgling sound is unmistakeable, and we at once learnt to regard it as a sure sign of the worm’s presence. The worm very frequently lives in ground riddled by the holes of the land- crab, as it is popularly called ; this animal has a small circular burrow leading down to a chamber hollowed out underneath containing a pool of water, and through these chambers the worms’ burrows frequently pass. The “crab” almost invariably has a large conical “ casting” at the entrance to its hole, and may raise this to a height of even a foot and more; but the true worm burrow never, so far as yet observed, has any “casting” at its entrance, and all trace of this is wanting where the crab-holes are absent. The very frequent association of the “crab” and worm leads to the idea that the latter forms a cast ; but one of the most noticeable features of the ground, which is at times riddled with worm burrows only, is the entire absence of “ castings.” What THE GIANT EARTH-WORM OF GIPPSLAND. 5 the worm does with the immense quantity of earth which it passes through its body I cannot at present say, and it must also be noticed that, only on very rare occasions, can any trace be detected of leaves dragged down into the burrows. It is no easy matter to extract the worm without injury, owing to its length, the coiling of the burrow, the rapidity of movement which it possesses when underground, and its power of distending either the anterior or posterior ends of the body, or both. Directly the burrow is laid bare, the worm is seen gliding rapidly away, often producing the curious gurgling sound as it passes through the slimy fluid, always present in a burrow containing the living animal. Sooner than allow itself to be drawn out, it fixes, if held in the middle, both ends of its body, by swelling them out till they are tightly jammed against the sides of the burrow ; under these circumstances pulling merely results in tearing the body. The worm has been described as brittle, but this term is most inapplicable, as its body is very soft, and capable of a great amount of extension before tearing. Its curious smell, when living, resembling somewhat that of creosote, has been already observed by Professor McCoy, and when dead it is worse than ever, and very strong and characteristic ; the body in decaying passes into an oily fluid which, we were assured by one or two old natives of the district, is very good for rheumatism. Fowls refuse to touch the worm living or dead. When held in the hand, the worm in contracting its body throws out jets of a milky fluid from its dorsal pores to a height of several inches; if the burrow be examined carefully its sides are seen to be very smooth, and coated over with a fluid exactly similar to that ejected from the pores. Whatever be the primary function of the fluid when within the body cavity, there can be no doubt that it has the important and perhaps secondary function when it has passed out of the body of making the burrow walls smooth, moist and slippery, and of thus enabling the animal to glide along with ease and speed. The worm in its burrow moves rapidly by swelling up its anterior or posterior end as the case may be, and then using this as a fixed point, in doing which the sete perhaps help, though to a minor extent, it strongly contracts the rest of its body. In the next movement, the end free in the first instance will be swollen out and used as a fixed point from which expansion forwards can take place. These changes of motion follow each other so rapidly that in the burrows the appearance of continuous gliding is given. Outside the burrow when the whole body surface is not mm contact with the earth, the worm makes no attempt whatever to move, lying passively on the ground. Anyone who merely sees the beast removed from its burrow imagines it to be of a very sluggish temperament, and can form no idea of its active and rapid movements when underground. So far as locomotion is concerned, its setze seems to be of little or no use to it. The pericheete worms, on the contrary, when taken from the burrow, move along on the ground with remarkable speed, certainly using their setze as aids to progression. 6 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, The burrows of the large worm measure 3-1 in. in diameter, and in disused ones are often found (1) casts of the worms, or rather, what are probably the earthy contents of the alimentary canal, with clear indications marked upon them of the segments of the body, and (2) more rarely cocoons. The latter measure 14-2 in. in length, vary from light yellow to dark brown in colour according to their age, and contain only one embryo each, which I have at present only been aoe to obtain in a somewhat highly developed state. The cocoon itself is somewhat thin, and made of a tough leathery material, with a very distinct stalk-like process at each end; it contains a milky fluid closely similar to that found in the body cavity of the worm. I hope soon to be able to increase the number of my stock of worm embryos, and to be able to work out in detail its development. It is interesting to note the fact that at the present time we know of three especially large kinds of earth-worms ; that of these, one comes from South Africa, another from the southern parts of India and Ceylon, and the third from the South of Australia. We know as yet little about the distribution of earth-worms, but the same laws which governed the distribution of other animals must also have governed theirs, and it is just possible that these great earth-worms may be the lingering relics of a once widely-spread race of larger earth-worms, whose representatives at the present day are only found, as occurs with other forms of life, in the southern parts of the large land masses of the earth’s surface. Possibly careful search will reveal the existence of a large earth-worm in the southern parts of South America. ANATOMY. 1.—External Anatomy. The worm certainly deserves its name of the “Giant EKarth-worm.” Large earth-worms have been described by Brnuam, from South Africa, and Brpparp from Ceylon, but this one from Gippsland seems to be the largest yet described. When dead the body is capable of great extension ; but the largest living one (measured by Professor McCoy) had a length of 6 feet. I have taken numerous worms direct from the burrow, and the longest one measured 56 in when alive; the average length being 44—48 in. and the breadth ? in. The following description of the external anatomy agrees in all important points with that already given by Mr. Fiutcumr.* The segments in the sexually mature worm vary from 300-500, or perhaps even more in number. When fully expanded, the anterior end of a large worm has the size and form represented in Figure 1, plate 1, seen from the ventral surface. The body tapers somewhat towards the first segment, broadening out till the eighth, then becoming narrower to the twelfth, then broadening again till the nineteenth, and then _ * Proc. Linn. Soc., N.S.W., 1887, p. 603. The worm is described under the name of Notoscolex g-ppslandicus. i ee) ieee Ties - ~ THE GIANT EARTH-WORM OF GIPPSLAND. 7 diminishing in size till the twenty-fourth or twenty-fifth, after which the same size is maintained till the very posterior end, where the segments again become enlarged. The prostomium overhanging the mouth is ribbed by longitudinal folds, and the first or buccal segment is divided into two by a groove; the second and third segments are of nearly the same width as the first, and each is similarly divided into two annuli by a median groove; the fourth seement is broader and has its groove nearer to the anterior than the posterior end. After the fourth segment, and as far back as the thirteenth, the grooves dividing the segments into annuli are somewhat irregular and frequently incomplete ; the figure represents them in the worm taken for description, though slight variations in the grooves are to be found in different specimens. Posterior to segment thirteen, the segments are faintly triannulate. There is no difficulty, especially in an expanded worm, in counting the segments, as the grooves marking them off are much more definite than those marking off the annuli. The setze vary in distinctness in various parts of the body and in different examples, and I have not been able to detect them anteriorly in front of the eleventh segment, though Mr. Fiurcusmr appears to have done so. ‘There is not the slightest difficulty in all specimens, at all events in certain parts of the body, in seeing them, or rather, the little papille from the tops of which they protrude. There are, as described by Professor McCoy and Mr. Fuxrcusr, eight longitudinal rows forming four series of pairs, but, as stated by the latter, the sete of the outer pairs of each segment are further apart than those of the inner pairs. The sete also agree well in shape with the description given by Mr. Fuercumr, they have a slight sigmoid flexure and a distinct swelling, about one-third of the way from the free pointed tip. Each has also a slightly swollen rounded extremity at the internal end where the protractor muscles are attached, which can be well seen when sections of the body wall containing the setze are cut. (Plate 3, fig. 13.) ‘The setee only project very slightly beyond the surface of the body and are probably of but very little service to the animal in progression. None are specially modified in connection with the male genital aperture, and those which should oceupy the position of the latter—that is the ventral pair on either side of the body in the eighteenth segment—cannot be seen either macroscopically or by means of sections. The dorsal pores which are very evident oval openings in the mid dorsal line, com- mence about the fifteenth segment, that is immediately behind the remarkably strong septa in the anterior part of the body. ach pore is very clearly situated exactly in the line of the groove separating two segments from one another. (Fig. 2, dp.) The pores can be traced back to the very last segment. The general body surface has a dirty pinkish flesh colour, and is quite intransparent. In segments 13-21 inclusive, however, the skin is dark purple in colour, and distinct from that elsewhere in the possession of a large glandular development immediately beneath the epidermis and external to the muscle-layers. (Plate 2, fig. 6.) The whole of this part may, with 8 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, Mr. Fircusr, be called the clitellar region, as its curious glandular development corresponds with that characteristic of the same part in other earth worms. In addition to this, there are present usually three and rarely four light-coloured and very prominent bands* (Plate 1, fig 1, clt.) on the ventral surface in the clitellar region. The first of these lies, as in the figure, partly on segment 17 and partly on segment 18, the second in a similar manner on segments 18 and 19, the third on segments 19 and 20, and the fourth, when present, on segments 20 and 21. The openings connected with the genital organs are not difficult to see im Specimens in anything like a good state of preservation. 1.—Spermathecal Pores (Fig. 1 rs‘ 0., 7s’ 0.) on the ventral surface very slightly in front of the grooves separating segments 7 and 8 and segments 8 and 9 respectively. Each pair is surrounded, as in the figure, by a slight line-marking on the surface of the skin, though these are too strongly indicated in the figure. 2.—Oviduct Pores (Fig. 1, fo.), just in front of but nearer to the mid ventral line than the innermost setze in segment 14. 3.—Vas Deferens Openings (Fig. 1, mo.) These are always evident and are placed on two slight papille within the second ridge in the clitellar region, one in the position of each of the two most ventral pairs of sete. As above said, the setz corresponding to this position cannot be found. No nephridiopores are visible. 2.—Septa. Immediately on opening the body cavity, one of the most striking features presented is the enormous development of the anterior septa.t (Fig. 2, sept.) For the first fourteen segments they form, as it were, deep cups, with their concavities facing forwards (Figs. 2, 3, and 4), each cup fitting into the one lying posteriorly to it. The septa are connected with each other and with the body wall by strong muscular slips. (m.) Passing backwards, the septa become gradually * This happened in the case of two out of some three or four dozen specimens examined by myself. Both these worms were found in March, but, save in this particular respect, agreed with all other specimens procured at other times of the year. + In Professor McCoy’s description, these ridges are said to represent an imperfect clitellum or cingulum. In view of the fact stated above, that the whole region from segments 13-21 has the skin modified in the manner which is so characteristic of a clitellum in other earth worms, itis perhaps preferable to call this portion of the body the clitellar regions, and to regard these ridges as special developments of the same part. In Professor McCovy’s description, these ridges are stated to occur between the 32nd, 33rd, and 34th rings (the word “ring” in his description is evidently, from its relationship to the setz, &c., used as the equivalent of segment), a mistake which must be attributed to the counting of the annuli as segments. The position of the structures on the segments named by Mr. Fuercuer and myself can easily be verified by counting the septa internally. { Cf. Bennam. WMicrocheta rappi. Q.J.M.S. Feb. 1886. Plate 15. THE GIANT EARTH-WORM OF GIPPSLAND. 9 thinner in segments 14-19 (Fig. 2), the long muscular slips running from segment to segment, until after the twentieth segment the septa assume the usual membraneous appearance, and retain this nature until the posterior end of the body is reached (Plate 4, fig. 20), where they again become strongly muscular. In this part, comprising some twenty segments, they form flat plates encircling and standing at right angles to the alimentary canal. There are very well-developed muscular slips, passing (Fig. 20) radially from the walls of the alimentary canal to the posterior face of each septum. The first complete septum bounds anteriorly, as may be seen from the external indications of the segments, the fifth segment. In front of this the body is filled with a-dense mass of muscle fibres surrounding the pharynx, all the spaces between being filled with the modified nephridia forming the salivary glands. Throughout the body the septa are incomplete only, as usual, on the ventral surface, where in the mid line is an arched space through which passes the ventral blood-vessel and the nerve cord. (Plate 5, fig. 21, sept, 0.) Special supporting muscles run to the body wall from each side of the arch. It is curious to note that all through the body, and especially in the middle and hinder regions, the insertions of the septa do not correspond with the grooves separating the segments ; each septum is very distinctly joined to the body wall posteriorly to the groove which externally forms the anterior limit of the segment of which it forms internally the anterior boundary. (Plate 3, fig. 14.) Hach septum (Plate 5, fig. 22, and plate 3, fig. 14) has, as it were, a slight pouch extending forwards in the mid dorsal line, and into which opens the dorsal pore, the latter therefore lying slightly in front of the segmental chamber with which it is thus connected by a short canal: in transverse section, this actually appears as a canal. At the pores the epidermic layer of cells dips down and lines the pore as far as the level of the inner side of the circular muscle-fibre layer. The longitudinal fibres run up towards the body surface around each pore with which they may happen to be in contact, and are inserted into the connective tissue immediately beneath the epidermis at its lowest position in the pore. The septa, as usual, are composed of radially arranged muscle fibres, of which only a thin layer is present, save in the very strong and thick septa anteriorly ; on either side is a thin epithelium of distinctly nucleated flattened cells, directly con- tinuous with the epithelium lining the body wall internally. 3.—Body Wall. This may structurally be divided into two parts—(1.) That over the whole surface except the clitellar region ; (2.) That in the clitellar region. In the first of these two regions the structure is, on the whole, that which is characteristic of most earth-worms, though in minor points it shows variations from that of Lumbricus. The epidermis is represented in fig. 7, the whole body wall in fig. 18. Most externally is the cuticle (cut.) The cells of the epidermis have the usual elongate columnar form, with very distinct oval nuclei; amongst them lie numerous goblet cells (gob.) much larger than the others, and distinguishable also by containing C tO" ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, clear spaces surrounded by a granular non-staininge material; towards the base of each can usually be detected a nucleus pushed to one side. These goblet cells push aside the others, only parts of which are hence sometimes cut in section. The internal ends of the columnar cells cannot generally be clearly made out, as they appear to run into a kind of thick basement membrane, in which nuclei (n.) are often seen. Into this membrane, and thus amongst the bases of the columnar and goblet cells penetrate numerous branches of blood-vessels (bv.) These vessels actually penetrate between the epidermis cells, and, though unfortunately not shown in the figure, form small coils amongst the cells reaching nearly to the surface. Internally to the epidermis is the usual layer of circular muscle fibres (Plate 2, fig. 6, and plate 4, fig. 18.) On its external aspect, the fibres are thinner and more loosely arranged than internally, as they are also in sections of Lumbricus ; they are also thrown into much more strongly marked folds than is the case with the internal fibres, which are thicker and somewhat closely packed together. Great numbers of blood-vessels — branch amongst the fibres. The layer of longitudinal muscles is, seen in transverse section, much deeper than that of the circular ones, and is again much more strongly developed ventrally than dorsally. The fibres differ considerably from those of Lwmbricus, not presenting that definite arrangement of the strands resembling branches on each side of a series of radially placed stems of connecting tissue, which is so characteristic of Zwmbricus. In Microcheta rappi* there is apparently no gathering of the longitudinal fibres into groups, whilst in Megascolides we meet with a form intermediate between the two extremes seen in Microcheta and Lumbricus. The strands (musc., long.) are first of all gathered into smaller groups of 3-12, surrounded by connective tissue (ct.), and these smaller groups are again gathered into much larger and irregular ones, separated from each other by radial strands of connective tissue passing from the circular muscles, inwards to the body cavity. Along these breaks, between the muscles, pass blood-vessels and the ducts of the numerous nephridia to be described later on. The strands are also, as in Zumbricus, angular and not rounded in section. Internally again to the muscle layers is a distinct layer of connective tissue, from which strands of tissue of a homogeneous appearance may be seen radiating into the body wall amongst the muscles (ct.), and within this is the epithelium, the irregularly shaped cells of which line the body cavity, becoming modified in various parts. In the second—the clitellar region—the skin is much modified as usual, but differs in structure from that in Lumbricus or in Microchetat as recently figured by BENHAM. * BennamM. Q.J.M.S. (Plate 16, bis., fig. 39.) + Op. Cit.—For the description of that in Zwmbricus I have not been able to consult the original paper of CLAPAREDE, and so have taken advantage of the description quoted by Mr. Benuam. THE GIANT EARTH-WORM OF GIPPSLAND. it Within the cuticle (Plate 2, fiz. 6) lies the epidermis (epi.), the cells of which are, asin Lwmbricus, somewhat shorter than those elsewhere ; their oval nuclei are very distinct, and their inner ends seem to merge into the thick basement membrane mentioned above. A few goblet cells can be seen, but not nearly so many as in the extra-clitellar region. There cannot be detected the narrow elongated cells con- taining granules similar to those found in the goblet cells, and remarkable for being some three times the leneth of the epidermic cells, such as are described in Lumbricus and Microcheta. Within the epidermis in Megascolides lies a great development of glandular cells (gl.) packed more or less closely together, and of very varying sizes. Some are almost globular, others almost, save for their rounded bases, columnar. All are filled with a granular material arranged in a meshwork in which les a very distinct oval nucleus. These cells have long ducts leading towards the exterior and swollen internal ends, and are many times longer than the epidermic cells. Possibly some of the more elongate external lying ones may correspond to those mentioned above as present in Lumbricus and Microcheta, but they, at all events, differ from these in not having branched bases, a condition which does not seem to obtain in any cells in the clitellum of Megascolides. Bznuam in Microcheta describes special strands of connective tissue as running down amongst the gland cells, and continuous with the network at the base of the epidermic cells; this is doubtless equivalent to a curious development seen in Megascolides, which consists of radial, somewhat wavy-looking, strands of a perfectly homogeneous material, which pass down amongst the gland cells from the basement membrane beneath the epidermis, till the circular muscles are reached, upon which the strands branch and lie flat (ct, fig. 6). This same material forms a surrounding to the openings of the nephrial tubes as they pass through the clitellum (Fig. 28, ct.) This glandular portion is very rich in blood-vessels, and the latter have a very definite arrangement, usually forming distinct coils (Fig. 6 and 28, bv.) One of these coils always accompanies each nephridial tube, and all have a definite radial arrangement. Bepparp figures in Acanthodrilus a blood-vessel, accompanying the nephridial tube, and forming a very distinct loop around the latter, just beneath the epidermis ; this loop enclosing the nephridial tube seems to be absent in Megascolides, but the two forms agree in having a special vessel running with the tube apart from, and in addition to the copious supply of blood-vessels on the nephridium itself. Numerous more irregularly arranged blood-vessels are present, and pass up close to the bases of the epidermic cells, between which they may penetrate. The clitellar region has a very marked development of pigment. The dark red-brown colouring matter is distributed principally in irregularly arranged stellate masses, close beneath the epidermis, and also along the paths of the blood-vessels. To this pigment, and the great development of blood-vessels, the clitellar region owes its dark colour. c 2 12 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, The nephridial tubes of this part, which are quite distinct from those of the rest of the body, will be described later. 4,.—Alimentary Canal.—On opening the body cavity by a median dorsal incision, the alimentary tract with its very thin brown coloured walls, and its lumen full of earth, is seen to occupy almost the whole of the body cavity in the middle and posterior part of the animal, save the last twelve to twenty segments, where the septa, as above described, are strongly developed, and there is consequently a very distinct space maintained between the body wall and the intestine. Anteriorly however, the canal is quite concealed from view by the immensely strong cup-shaped septa (Fig. 2). A median longitudinal section gives very much the appearance, as far as the alimentary tract is concerned, which is represented diagramatically in Fig. 3. The short buccal chamber is distinguishable by the possession of very thin walls, which can be protruded. This opens immediately into the pharynx, the walls of which are as usual very thick and muscular, with special strands passing from them to the body wall. The pharynx extends through four somites, that is as far back as the first septum which bounds anteriorly the fifth segment. The space between the pharynx and the body wall is occupied partly by muscle strands, but principally by a large mass of minute tubules resembling exactly, in appear- ance, the numerous small nephridia present in the other parts of the body. Examined with a lense in situ, these structures, which there is little doubt are salivary glands, are seen to be a mass of minute coiled tubules, with a very strong development of blood-vessels upon them. Somewhat similar structures are described by Bznnam as occurring in Trigaster lankestert ;* in this worm “there are three pairs of grape- like glands” around the pharynx and cesophagus in the somites, [V., V. and VI. In Megascolides, on the other hand, they are much more irregularly arranged, and are confined to the pharynx. ‘Transverse sections show them to have precisely the same structure as nephridia, and after a careful search I was enabled to trace their openings into the pharynx. (Plate 3, fig. 10.) The glands are composed of an immense number of small tubules embedded in connective tissue. Each contains an intra-cellular duct. The structure of these is exactly similar to that of the nephridia from which the salivary glands, anatomically, are indistinguishable. There is present also the great development of blood-vessels forming a network upon the tubules. The latter are massed together in great numbers, and the ducts seem at intervals to join together into a common duct, which pierces the muscular walls of the pharynx, and opens into the alimentary canal. Where the opening takes place, the columnar cells lining the pharynx are, just as in the case of the epidermis where the nephridia open, modified so as to form what, at first sight, resembles somewhat a taste bulb. (Fig. 10.) The cells are arranged, as it were, along what would correspond to the lines of longitude on the surface of a sphere, being swollen out medianly where is the nucleus and tapering * Bunyam. Q.J.M.S., August 1886, page 96, plate 9, fig. 35. THE GIANT EARTH-WORM OF GIPPSLAND. 13 towards both ends, and the duct passes right through the axis of the sphere (cf. for nephridia, plate 2, fie. 6, and plate 6, fig. 28.) From their relation to the alimentary canal, there is no doubt that they are of the nature of salivary glands, and their structure even to minute details shows them to be modified nephridia. The only difference observable between the salivary glands and the nephridia, is the absence in the former of the inter-cellular duct. The whole duct is intra-cellular until the epidermic lining of the pharynx is reached. This modification of nephridia agrees with that observed in Peripatus and almost certainly in Acanthodrilus multiporus and Trigaster lankestert.* It will doubtless be found to exist in many other worms. Segment five is occupied by the very short cesophagus, and the gizzard. The intestine commences in segment six, and runs hence to the extremity of the body. There is no trace whatever either of cesophageal glands or of typhlosole. The only modification in the intestine consists of very distinct dilatations in segments 12-18 inclusive. (Figs. 3 and 4.) In these parts the walls are highly vascular, and devoid of any strong muscular development. Sections (Plate 5, fig. 24) show that the hypoblastic linme is thrown into a series of longitudinal folds. Each fold is made up of thin columnar cells with very distinct oval nuclei, the cells of each side of a fold being very close to each other, and separated by a network of blood-vessels (bv.) which are in connection with another network running round the intestine beneath the circular muscle fibres. This great supply of blood-vessels gives the dilatations of the intestines a reddish colour. Outside the blood-vessels lie (1) the layer of circular, and (2) the layer of longitudinal muscle fibres, both of which are comparatively feebly developed. In and amongst the latter is a small amount of connective tissue, and outside is the layer of flattened peritoneal epithelium cells. Posteriorly to segment 18 the intestine forms a simple tube, the walls of which are, as usual, brown coloured. In sections (Fig. 5) they are seen to have the usual structure. Most internally lies a layer of deep columnar cells with large spherical nuclei, each with a distinct aggregation of chromatin fragments in the centre. - There are no cilia to be seen, and when the walls are uncontracted the columnar cells form a smooth coating to the tube, with an absence of folds such as are described and figured by Bznyuam in Wicrocheta rappi. External to the columnar epithelium are the circular muscle fibres, which are well developed, especially in the posterior segments of the body, from which part the section represented is cut. Here, owing to the strength of the fibres, the canal retains its tubular shape when taken from the body. Externally to the circular, lie the longitudinal fibres (Long. M.), which seem to be best developed in the median dorsal line beneath the blood-vessel. In the walls of the intestine are two networks of blood-vessels. From the dorsal blood-vessel (Figs. 5 and 8, d. bv.) pass off two branches on each side of every segment to the walls of the alimentary canal. (Plate 2, fig. 8, al, bv.) Hach pierces the longitudinal fibres, and a network of vessels is 14 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, formed immediately beneath these. From this network branches pass down, piercing the layer of circular fibres (Fig. 5), and another network is formed at the bases of the columnar cells, amongst which branches penetrate. The intestine is covered externally with an epithelium of cells, whose extremities, facing into the body cavity, are much larger than their internal ends, which run down between the longitudinal fibres. Still finer processes can at times be traced on from these cells, leading through the layer of muscular fibres till they reach the intestinal epithelium (these are represented too clearly in the figure). Hach cell contains in its swollen end a great number of minute yellow-brown granules, which give the characteristic colour to the intestine in this part. The nucleus is very distinct, somewhat spherical in shape, and placed where the cell begins to narrow. In the very median dorsal line these cells are absent, and a tubular structure is present, connected with the wall of the alimen- tary canal, and surrounding the dorsal blood-vessel, to which reference will be made later on. The brown cells are merely special modifications of the peritoneal epithelium, with which they are directly continuous around the dorsal tube just mentioned. There is nothing corresponding to the typhlosole, nor to the intestinal glands present in certain earth-worms. 5.—Vascular System. The vascular system is, in comparison with most earth worms, somewhat simply developed. It consists of the following vessels :—(1) dorsal trunk ; (2) ventral trunk; (8) transverse vessels; (4) lateral vessels. As- in Pleurocheta, Pericheta, Pontodrilus, and Microcheta, there is no sub-neural trunk. 1.—Dorsal Trunk. ‘This runs from the posterior to the anterior end of the body, in close connection with the dorsal wall of the alimentary canal. In the fifth segment it breaks up into numbers of small branches, which run forward and form a plexus in the walls of the gizzard and pharynx. (Plate 1, fig. 2; plate 2, figs. 8 and 9, d, bv.) In segments 14-6 inclusive, it is somewhat swollen and its walls more muscular, and hence firmer than in the posterior region. In Plate 4, fig. 20, is a drawing representative of a structure confined to the hinder half, or even less, of the body. Such a structure does not appear to have been described in any other earth-worm as yet, and its exact nature is very doubtful. When the hinder part of the body is opened, the dorsal blood-vessel is seen to be surrounded by a white-coloured tubular structure, from which more or less solid diverticula are given off on each side. (Plate 4, fig. 20; plate 2, fig. 5, , and dw.) In the section (Fig. 5), half of the tubular structure surrounding the dorsal blood- vessel is represented. It is lined throughout by the peritoneal epithelium (ep.), which in the median line, as mentioned above, is not modified into brown cells, though certain of the cells lying nearest to these become somewhat columnar. The walls of the tube are formed of connective tissue (ct.), in which lie a certain number of circularly disposed muscle fibres, which, where the tube rests upon the alimentary canal, pass both towards the central line, lying upon the longitudinal fibres, and outwards for a short distance amongst the bases of the brown cells. Into the tube THE GIANT EARTH-WORM OF GIPPSLAND. 15 open the diverticula, the walls of which are formed of connective tissue, lined externally by peritoneal epithelium. Each process has the form of a bag, more or less filled with a mass of polygonal-shaped cells (y.), with very distinct nuclei; between the masses of these cells, often stretching from one side of the diverticulum to the other run strands of connective tissue. Sometimes (dv.) the processes are full of these cells, at others (dv.’) they are comparatively empty. The main tube (z) opens into the ccelom in each segment anteriorly by a slit-like aperture on its ventral surface just where it becomes constricted to pass with the dorsal blood-vessel through the septum. In this part there is very little space between the wall of the blood-vessel and that of the tube in question. The cells from the diverticula, which in general appearance, save their white colour, much resemble those of the liver of higher forms, can pass into the main tube and thence into the ccelom, and may perhaps serve as stores from which are formed the numerous white corpuscles present in the ccelomic fluid. The structure has no direct connection with the vascular system, nor are any blood-vessels whatever to be found in or about the diverticula. In a worm of some 500 segments it could be traced anteriorly to about the 120th segment from the head end, where it ceased. As far as the 150th the tube could be seen, but very few diverticula, which were well developed from the 200th to the posterior extremity. °2.—Ventral Trunk. This runs, as the dorsal one, from the posterior to the anterior end, where, in the fifth seoment, it breaks up into small branches, distributed principally to the walls of the gizzard and pharynx. It lies at some distance below the alimentary canal (Plate 2, figs. 8 and 9; v. bv.), from which it is. suspended by a very definite longitudinal mesentery. It is placed just above the nerve cord, and passes, as usual, with this through a definite opening left in each septum in the median line ventrally. (Plate 5, fig. 21; sept. 0.) 3.—Transverse Vessels. Branches are given off both from the dorsal and ventral trunks, and the nature and distribution of these branches differs very much in the ; regions behind and in front of the 14th segment. (a) In the segments behind the 14th (Plate 2, fig. 8), three branches are given off from the dorsal vessel on either side. Of these, one (sept. bv.) runs to the anterior face of the septum bounding the seoment posteriorly, and gives branches to its upper half, the other two (al. bv.) run round the alimentary canal, and pierce the muscular coats, forming as already described, two vascular networks, one on either side of the circular muscle layer. These may be called the intestinal vessels. From the ventral vessel arises one main branch on either side, close to the septum, bounding each segment posteriorly. A short mesentery runs out laterally from the ventral blood-vessel to the body wall, and serves to support on either side one of the two branches into which this lateral vessel divides. The upper of the two branches (sept. bv.) supplies the lower half of the septum ; the lower runs across the mesentery to the body wall, and supplies the latter with blood-vessels, and also gives off the very numerous smaller branches, which form such a rich plexus around the nephridia. (8) In the 14th segment the dorsal blood- 16 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, vessel becomes muscular and swollen out, and gives off the same three branches already described. From the ventral vessel arises a single one, as in the posterior sections, which supplies the septum and body wall. (This has inadvertently been omitted in the diagram Fig. 9, but should have been drawn as in Fig. 8, immediately above.) In addition to these there is also a lateral vessel present on either side of the alimentary canal, to which, as well as the septum, it gives branches. It is connected with the lateral’ vessels of the anterior segments, but appears to have no connection with any vessels in the 14th segment. (c) In the segments 13-6 inclusive, the blood-vessels are very differently developed. These are first the usual dorsal and ventral trunks, both somewhat muscular, and the former swollen out in each somite. From the dorsal arises on each side posteriorly a single large muscular vessel, the “heart,” (h.) passing down to join the ventral vessel. Hach transverse vessel, or heart, gives off a branch from its antero-internal surface. Hach of these again divides into two, of which one runs to join its fellow of the other side, and these two uniting form a small median vessel lying immediately upon the dorsal surface of the alimentary canal (al.* bv.) ‘The second branch runs downward, and this again divides, one branch running to join the lateral vessel (lat. bv.), the other, and larger, passing on to the face of the septum, and giving off one important branch ventrally to the body wall and nephridia. The lateral vessel on either side gives off branches to the alimentary canal. In front of the sixth segment, the dorsal, ventral, and lateral branches all break up into smail divisions, which supply the body wall and the alimentary canal and salivary glands. There are no blood-vessels in connection with the nerve cord—neither sub. nor lateral-neural vessels. The Blood itself has the usual red colour, due to the presence of hemoglobin in the fluid, and contains (1) very numerous nucleated corpuscles of definite outline, which may be either oval or round in shape, and measure about 0016 mm. in diameter (Plate 3, fig. 17, a.), (2) fewer more irregularly shaped nucleated corpuscles, from which few or many stiff pseudopodia-like processes may be extended (Fig. 17, b.) These are somewhat larger than the former, the body of the cell, exclusive of process, measuring about ‘0025 mm. in diameter. The Celonuc Fluid, on the other hand, is of a milky-white colour, and opaque. Its numerous corpuscles are all of more or less irregular shape, and precisely similar in size and appearance to the second kind described im connection with the blood. (Fig. 17, b.) Possibly, as suggested above, the masses of cells present in the diverticula of the tube surrounding the dorsal blood-vessel posteriorly are merely special developments of the peritoneal layer lining the ecelom which serve as supplies of these corpuscles. The latter, passing into the ccelom, lose their regular shape, and put out processes—becoming, in fact, ameceboid cells. THE GIANT EARTH-WORM OF GIPPSLAND. iy. There is no doubt that the worm passes the ccelomic fluid out by its dorsal pores into the burrow, and with the fluid must pass the corpuscles, a constant fresh supply of which will hence be needed. . 6.—Nervous System. ‘This has, in the main, the form usual in earth-worms. It consists of a pair of cerebral ganglia, united by commissures to a ventral cord. The cord cannot be clearly divided either by its external shape or by its structure into well-defined ganglia and commissures, and is of about the same width throughout its whole length. Where the cerebral ganglia join the commissures on either side, a nerve arises which runs forward and sends branches to the body wall in the prostomial region. (Plate 3, fig. 16, In.) A little lower down, a curiously flattened mass of nerve fibres arises from the postero-internal face of the commissure, and immediately breaking up into branches enters the muscular walls of the pharynx. (Fig. 16, ph.,n.) In section the branches are seen ramifying amongst the connective tissue and muscle fibres of the pharyngeal wall. (Plate 3, fig. 10, nc.) This constitutes the stomato- gastric system. Still lower down the commissures arises on either side another nerve which runs forward, and breaking up into branches supplies the body wall ventrally in the region of the mouth. The ventral cord gives off in each segment three pairs of branches. ( Plate 5, fig. 21, v3. n. n3.) Of these three branches, two are placed somewhat close together mn the posterior region of the segment, whilst one is situate slightly in front of the median line. Where the nerve cord passes (Fig. 21, sept. 0.) through the septum it is supported by a definite mesentery, which arises from either side of it, and is inserted into the body wall ventrally. This structure runs for some little distance both anteriorly and posteriorly to the septum. The histological structure is much the same as that described for other earth- worms. The whole nervous system is covered by the coelomic epithelium, which over the greater portion of the surface consists of nucleate flattened cells. On the ventral nerve cord, however, as the under surface is reached, these merge into cubical cells, and such also cover the supporting mesenteries. (Fig. 19, ep.) Within the epithelium is a definite development of connective tissue, forming a sheath for the whole nervous system. As in Pontodrilus, and certain species of Microcheta and Plewrocheta, neither sub-neural nor latero-neural blood-vessel is present. With regard to the nervous tissue itself, the cerebral gangha consists of (1) a sheath of ganglion cells on the anterior and ventral aspects, and (2) a mass.of fibres dorsally and posteriorly, amongst which a strong commissural band, passing from side to side, is developed, such as is not met with elsewhere in the system. As previously noted, no nerves seem to be given off from the cerebral ganglia directly, though one arises on each side where the commissures begin. The commencement of these is indicated by the absence of ganglion cells, none of which are found in the commissures or, as far as could be ascertained, by means of a long series of consecutive sections, im D 18 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, the stomato-gastrie system. The walls of the pharynx are well supplied with nerve fibres, which run backwards from the commissures. (Plate 3, fig. 10, ne.) Where the two latter join the ventral cord ganglion cells again appear, and are developed along the whole length of the ventral cord. Along the cord the connective tissue is arranged in a very definite manner. There is first of all the external coat immediately beneath the ccelomic epithelium. On either side this passes off in the region of a septum into a flat mesentery supporting the cord (Fig. 19), and as two strands of tissue are given off, one running from the upper and the other from the Jower side of the cord, a somewhat triangular space is enclosed in which longitudinal muscle fibres are enclosed. Similar fibres are present in other worms, as figured by Benuam in Microcheta rappi.* Within the cord itself are definitely arranged strands of connective tissue, in addition to the still finer tissue penetrating in and amongst the nervous elements and serving for their support. The cord is clearly seen in sections to be composed of two lateral halves, and may be divided into (1) a central double division, consisting of two halves, lying closely side by side and separated from each other and surrounding structures by a thin, but distinct, layer of connective tissue. These two parts are each composed of a meshwork of nerve fibres, which, in very many cases, can be traced into connection with larger or smaller ganglion cells lymg externally. The fibres, many of which are seen to curve in towards the middle line, very frequently have slight varicose swellings; and amongst them, more especially towards the ventral surface, and near the middle line, are a certain number of small nucleated ganglion cells. (Fig. 19.) The central part of the nerve cord is noticeable by reason of its not staining nearly as deeply (with borax-carmine) as the outer layer, and from the very much looser way in which its fibres are arranged. (2) An outer part confined to the lateral and ventral aspects of the cord, composed of a much denser meshwork of fibres, containing ganglion cells of various sizes. These cells do not appear to be aggregated into definite ganglia, but are present in great abundance in the parts of the cord which lie between the points of origin of the three pairs of nerves which arise in each segment. The larger cells are distinctly pear-shaped (Fig. 17), and as a rule arranged so that their thin pointed end protrudes into the central part, and from it passes off a single fibre, curving round towards the centre of the cord. Some of the smaller cells appear to have a fibre passing off from each end, but the large ones never have more than the single one. Occasionally a cell may be so large as to stretch across the space between the outer and inner layer of connective tissue. This external layer takes stain more readily than the internal one, and containing numerous nerve cells, as well as a closer meshwork of fibres, appears in section much darker. Where the lateral nerves enter, the fibres run right through to the central part, with which alone they appear to be in connection. In this region, ganglion cells are * Q.J.M.S., Feb. 1886, Plate 16, bis., fig. 37. THE GIANT EARTH-WORM OF GIPPSLAND. 19 much less frequent, and as the nerves curve upwards, so as to enter nearer the dorsal than the ventral side, the few which are present are almost confined to the under surface. In Microcheta rappi,* Brennan figures the ganglion cells as present in greater numbers in the region of the lateral nerves, exactly the reverse of that which obtains in Megascolides. (3) An upper region where “giant fibres” are present. Each of these fibres is enclosed in a distinct encasement of connective tissue, continuous (see Fig. 19, gf) with the external and internal layers of this already described. The numbers of the “giant fibres” seen in section, varies in different parts of the cord. Right in the very front of the body, in the region of the sub- pharyngeal ganglion, only one, the central upper one of the figure is present (cf. Fig. 10, gf-), a little further back two lateral ones are present in addition, and in very many sections, though not in all, a fourth median one is present, as in the figure. The great length of the worm precludes the cutting of continuous sections, so as to trace the exact relationship of the fibres, but they were traced, as in the figure, completely through one segment in the hinder region of the body. In another series of sections from the same region, the lower one of the four gave branches off to the two lateral ones and disappeared, the connective tissue septum then passing right up the median line through the space occupied in the figure by the “ giant fibre.” The three upper ones appear to be the main ones, and of these, the first seen in a series of sections passing backwards from the cerebral ganglia is the central one. (ef. Fig. 10, in which in the region of the pharynx, just posterior to the sub-pharyngeal ganglion, only the central fibre is seen). It commences in the very anterior region of the ventral cord, where it simply thins out and merges into the external connective tissue sheaths of the cord. The two lateral ones appear in a similar manner a little further down the cord, but no giant fibres appear, as in other earth-worms, to either (1) pass up the pharyngeal commissures, or (2) to have any connection with the nervous tissues. As to the structure of the fibres, they are describedt as being composed, of a “‘doubly-contoured sheath, with clear contents,” and as being “separated from the nerve cord by the inner neurilemma,” and “ embedded in a connective tissue-sheath, containing reticulate cells.” Benuam{ figures them with the double contour, and surrounded by a certain amount of connective tissue. In Megascolides (Fig. 19) each one is enclosed, as said above, in a very definite compartment of tissue (ct.), and each fibre itself has the nature of a rod, formed of a perfectly homogeneous material, which stains moderately with borax-carmine, and has, owing doubtless to the action of reagents, a somewhat irregular outline in section. At times a nucleus may, as in the figure, be * Op. cit. Fig, 37. + “Forms of Animal Life,” Rotneston and Jackson, 2nd Edit., p. 212. t Op. cit., Fig. 38, 20 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, seen in the rod. Hach rod is again surrounded by a very clearly-marked sheath of tissue, which in transverse section seems to consist of fine wavy fibres, arranged, as in the figure, in a circular manner, with nuclei here and there. In longitudinal sections, the rods are very clearly seen to be formed of a homogeneous structure, perfectly continuous alone apparently the whole body. The “ giant fibres” can in no way be called “neural canals,” the name suggested by CunnincHam* for what must be regarded as homologous structures in the Polychete. ‘Neural canal” is a term which already has a definite meaning, and the structure to which this is usually applied has, of course, no connection whatever with the structure now under consideration. The “giant fibres” of Megascolides are remarkable (1) for the very definite central rod of homogeneous gelatinous material, and (2) for the equally definite enclosing sheath of connective tissue. No connection exists between the fibres and any of the nervous elements, and it seems probable, more perhaps because it is difficult to suggest any other use for them, than because of any direct evidence in its favour, that the usually accepted idea of their possessing solely a supporting function is the correct one. They doubtless serve also, as suggested by Cunniyeuan, “to prevent the nerve cords being bent at a sharp angle, causing them always to remain in curves, and so to escape injury during the wriggling and burrowing of the worm.” Under these circumstances, the name of “neurochord,” given to them by Vejdovsky, appears to be the most applicable. I hope very shortly to be able to study their development in Megascolides, in which in its adult state they form striking features in sections of the nerve cord. 7.—Nephridia. The nephridia form, perhaps, the most interesting structures in Megascolides. During the last two or three years, our knowledge of the nephridia of various earth-worms has been very greatly enlarged, owing in large part to the descriptions of Brnnam and Brpparp. The arrangement of the nephridia in Megascolides has many points of agreement with that described by Brpparpt as obtainme in Pericheta aspergillum and Acanthodrilus multiporus, and has already been shortly described.{ The main features in connection with the system are (1) the presence of numerous nephiidia im each segment, (2) the modification of the nephridia in various parts of the body, (3) the connection of the ducts of the various nephridia. Purrinr was the first apparently to draw attention to the presence of several nephridia in one segment, and in 1885, Bepparp§ described the same occurrence as taking place in a species of Acanthodrilus, from New Zealand, which, according to him, had eight nephridia in each segment. Brnxam|| mentions the fact that in certain worms (a Pericheta from * Q.J.M.S., Nov. 1887, p. 267. § Proc. R.S. May 1885, p. 459. + Q.J.M.S., Feb. 1888. || Q.J.M.S. Feb. 1886, p. 256. t Proc. R.S. Victoria. Oct. 1887. aes THE GIANT EARTH-WORM OF GIPPSLAND. 21 the Phillipines) numerous small nephridia occurred in each segment; whilst FLetcHer * noted that in various species of Australian earth-worms there were numerous small vascular tufts in each segment attached to the body wall, and probably nephridial in nature. Such occur, according to him, in Digaster armifera, Cryptodrilus saccarius, Perissogaster excavata, Pericheta tenax, P. gracilis, P. queenslandica, P. stirlingi. He noted the same fact also in various species of his genus Notoscolex. ( = Megascolides McCoy.) Bepparp, lastly, has described in two worms very numerous nephridia as being present in each segment, the whole system having a curiously close resemblance to that which obtains in Megascolides. In respect of having these numerous nephridia in one segment, the Oligochate agree with certain Polychate (Capitellide), and the connection of successive nephridia has now been shown to be common to the two groups.t As already described, when dealing with the salivary glands, these are. to be regarded without doubt as modified nephridia. They are formed out of the organs belonging to the segments as far back as, and including the fourth. They are arranged with great irregularity, forming a mass of minute vascular tubules connected with the walls of the pharynx, and filling up all the spaces amongst the muscle bands passing from the alimentary canal to the body wall. Similar structures have been already described and figured by Bryuamt in his account of Zrigaster lankesterr. He does not, however, appear to have traced these modified nephridia into actual connection with the alimentary canal, and in Trigaster also, they are much more definitely grouped than in Megascolides, and are confined to segments 4, 5, and 6, but they possess the same intra-cellular duct, and the network of minute blood-vessels, so characteristic of nephridial tubules. In all the segments behind the first four, the nephridia are seen, when the worm’s body is opened, to have the form of numerous minute vascular-looking tufts attached to the body wall. (Plate 1, fig. 2; plate 5, figs. 21 and 22.) Variations in the number, distribution, and structure of the nephridia are to be detected in various parts of the body, and the whole system is, in certain respects, more interesting than that of any earth-worm yet described. There are two distinct hinds of nephiridia present —(1) An enormous number of small nephridia (cf. Plate 1, fie. 2; plate 5, fig. 22), which lies so close to one another that the shape of each separate one cannot be distinguished. Though they vary somewhat in form, each appears to be always composed of at least two divisions—(1) A small, somewhat straight, tube, and (2) a larger coiled part. These small nephridia are present in every segment after the fourth. They are most largely developed in the clitellar region, where they form an almost complete * Proc. Linn. Soc. N.S.W., various papers on Australian Earth-worms. 1886 and 1887. + CunninecHam. Q.J.M.S. Nov. 1887. Nature, June 16,1887. Brepparp, Op. cit. Spencer, Proc. R.S. Victoria. Oct. 1887. t QJ.M.S. August 1886. 22 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, investment for the body walls. (Fig. 1 and fig. 22.) They are absent in the very median dorsal and ventral lines, though they extend very closely on to these. On the ventral surface, the nerve cord separates the nephridia of the two sides. In Fig. 22 is represented a portion of the dorsal body wall, cut away from the hinder clitellar region, where the septa are thinner than in front. Three segments are shown, and in the median line is seen a distinct groove, which separates the nephridia of the two sides, and out of the roof of which open the dorsal pores. Passing backwards, the nephridia lessen in number, and after about the 25th seoment, form simply a ring round the posterior region of the segment (Fig. 21) between the two posterior and the anterior pairs of nerve branches. ‘They are more or less irregularly scattered in the ring, which in the middle and posterior parts of the body seem to be continuous across the mid-dorsal line, to be only broken ventrally where the nerve cord runs. It is quite impossible to count the number of these small nephridia, of which certainly more than a hundred are present in each clitellar segment. None of them have any trace whatever of an imternal opening, though I have searched for such most carefully, both by means of examining numerous whole nephridia, and large series of sections. 2.—A series of much larger nephridia, which are only present in the posterior region of the body, and occur in the same segments with the smaller nephridia (Fig. 21), from which there is no difficulty in distinguishing them—(1) On account of their position ; (2) on account of their size, and (3) on account of their possessing internal openings. There is only one pair of these larger nephridia in each segment, and each lies close to the ventral nerve cord, occupying the anterior part of the segment, whereas the smaller nephridia are placed in the posterior. Each is attached to the body wall in front of the anterior nerve branch, and not very far behind the septum. There is no difficulty in recognising on the anterior wall of every septum, close to the median ventral line, a small ciliated funnel, which has the appearance during life of a minute white speck, attached by a short stalk to the septum. (Fig. 21, neph. 7. 0.) With care, it can be dissected away, and under the microscope is recognisable at once as a ciliated trumpet-shaped opening, the lower lip of the trumpet being somewhat shorter than the upper. (Fig. 26.) The opening leads back into a tubular duct, in which cilia may be seen working, and which is evidently continuous with the anterior prolongation of the large nephridium. The latter consists of two parts—(1) A smaller straight one (7.), and (2) a larger coiled portion, which seems to vary somewhat in shape and size, and to consist of a large number of smaller coiled tubules aggregated together, and enclosed in a common surrounding of connective tissue. As above said, these large nephridia are only found in the posterior region of the body, and are quite wanting anteriorly, where the smaller ones are alone present. THE GIANT EARTH-WORM OF GIPPSLAND. Ze There is, however, a very curious, and, at the same time, instructive series of gradations to be seen if the nephridia be carefully studied in situ. Thus in worms containing some 500 segments, the large nephridia, with internal openings, can be recognised through some 220 segments, counting from the posterior end. In front of this no internal openings can be detected, but for some 15-20 segments anteriorly, a nephridium larger than the rest, but gradually decreasing in size, can be recognised. It does not, however, stand alone, but forms one of a group of nephridia, the large individual one, on passing anteriorly, gradually decreasing in size and predominance as the group increases until, after twenty segments or thereabouts, the group remains, but no one nephridium is larger than the rest. This ventral group, as the anterior end is reached, tends to merge more and more into the general ring of small nephridia (though generally detectable by careful examination), which at the same time becomes more widely spread over the body wall, until, in the clitellar region, the whole body wall becomes covered with the smaller nephridia. THE SPECIALIZATION IN NEPHRIDIA APPEARS THUS TO CoMMENCE AT THE PosTERIOR Ewp, and To Pass Grapuatty Forwarps, tHe Anterior Brine 1n 4 Muce More Primitive Conpition THAN THE Posrmrior Eyp or tHE Bopy. Structure of the Nephridia. 1.—Small nephridia.—These, when cut in section, are seen to have the structure typical of the nephridia of earth-worms. Lach consists of a coiled tube with an intra-cellular duct (Fig. 18). The tube is surrounded with an investment of a large number of connective tissue cells, the boundaries of which cannot be detected, though their nuclei are clearly seen scattered about irregularly. The effect is produced of a somewhat homogeneous mass surrounding the tubules and continuous with the body wall. This agrees closely with the structure, described by Bennamas existing in the part of the nephridia of Microcheta rappt, farthest away from the vesicular end. There is also a thin definite layer of tissue surrounding closely the tubules, consisting of fibres with nuceli (ct. neph.) In the connective tissue, ramify and branch a great number of minute blood-vessels (bv.), which, as usual in earth-worms, are seen when the nephridium is viewed as a whole, to form an investing network over the nephridial tubules. The latter have the usual form of a series of large cells, placed end on, and pierced by an intra-cellular duct. Their nature makes them at once easily distinguish- able from other structures in the body. The cell consists of a homogeneous ground work, in which are scattered minute granules, so set as to produce in section the effect of lines radially arranged with regard to the duct. The bounding line of the latter is very clear and distinct. This structure again resembles exactly that figured by Breyuam in Microcheta rappi and others, and by Bepparp in Pericheta aspergillum and Acanthodrilus multiporus. Where the nephridium joins the body wall the duct passes into the latter (Fig. 18) sometimes directly, sometimes after running along for some 24 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, distance parellel to the length or circumference of the body, and giving off or joining another similar duct from some other nephridium (neph.) The nature of the duct opening to the exterior varies in the clitellar and non-clitellar regions, and will be dealt with after the structure of the larger nephridia has been described. 2.—Larger nephridia.—Their structure agrees in the main with that of the small ones, so far as the body of the nephridium is concerned. In both cases the ducts are so complicated, that as remarked by Bzepparp in connection with Acantho- drilus multiporus, it 1s simply impossible to map out the course of the duct. The differences between the large and small nephridia are, (1) the greater size of the larger, (2) the consequent greater number of the ducts cut in section. Of these some are thicker walled than others, and correspond exactly to those present in the smaller ones. The thinner walled ones, which are otherwise similar, are probably equivalent to the secondary ducts of other forms (such as Microcheta, &c.) and are absent in the part nearest to the attachment to the body wall. The supply of minute blood-vessels forming a complete network around the ducts is even more strongly marked than in the smaller ones. The blood-vessels to the large nephridia arise from a pair of vessels given off one on either side of the body from the ventral blood-vessel. The lateral vessels thus given off divide into two parts, of which one (Fig. 8, sept. bv.) runs to the septum, the other to the body wall ventrally and the nephridia. In Fig. 26 (Plate 6) is drawn a longitudinal section through the internal opening (Fig. 21, neph. v0.) of a large nephridium. The duct leads through the septum and swells out on the anterior face of this into the funnel-shaped opening. It is encased in connective tissue, amongst which are a certain number of muscle fibres (m.), circularly and obliquely arranged. External to the connective tissue and muscles is the layer of pavement ccelomic epithelium (c. ep.) The duct, almost up to the opening, is merely a continuation of the intra-cellular duct of the nephridium, and is ciliated. It is exceedingly difficult to determine exactly where the intra-cellular duct ends, and the inter-cellular begins. The cells lming the whole of the funnel have the curious homogeneous appearance characteristic of the nephridium, and the granules present in the cells in the body of the latter are here not so distinctly marked or radially arranged. It appears in many sections as if the duct right to the opening were intra-cellular, but after examining with high powers several series of sections, both transverse and longitudinal, I think that the drawing represents the real state of the case, and that the funnel has at its open extremity some three rows of cells serving to form the mouth of the funnel, and the internal opening of the nephridial duct into the body cavity. The cilia at the mouth are long, and at the entrance turned with their free ends towards the body cavity. External Openings of the Nephridia.—Brpparp has recently published a description of the relationships of the various nephridia and their external openings in Pericheta and Acanthodrilus, which has a most curious resemblance in many points THE GIANT EARTH-WORM OF GIPPSLAND. 25 to that present in Megascolides. The work upon the latter was done quite independently of Brpparp’s, and the fact of the existence of a connection between the nephridia stated in a preliminary paper read to the Royal Society of Victoria in October, 1887. Bovrnez,* in Pontobdella, described a complete nephridial network with a pair of internal and external openings in each segment of the body. Brpparp has described a continuous network in Pericheta aspergillwm, and a network not continuous from segment to segment, or on both sides of the body in Acanthodrilus multiporus. In the former, the external openings are arranged in a row round the body; in the latter, they are very numerous and irregularly arranged in the anterior part of the body, less numerous, more grouped together, and having somewhat definite relationships to the sete in the posterior part of the body. In the anterior region of Megascolides the nephridial ducts are connected together, and branches pass in the body wall from the ducts of the nephridia on one side to those of nephridia on the other side of the dividing septa, forming a nephridial network. In the posterior region of the body, all the small nephridia in the ring are united by a network of ducts, and on the ventral surface on either side of the mid line runs a clearly marked single duct (d. Fig. 27), which opens into (1) the small nephridia on the ventral side in the posterior part of the segment, which are also connected with the nephridial network, uniting the ring of small nephridia, and (2) the large single nephri- dium anteriorly. This duct passes forwards and through the body wall immediately beneath the insertion of the septum appears to send forward a small prolongation into the next segment to the region of the small nephridia. From these the duct (d.) passes on again to the next large nephridium, and so on. The longitudinal duct in transverse sections is seen to lie internally to the ventralmost setz on either side. There is no doubt whatever as to the presence of the network uniting the smaller nephridia and the longitudinal duct on either side, into which the network opens, and which runs forwards to the large nephridium. The duct passing beneath the septum is finer and more difficult to trace, but Fig. 27 represents in diagrammatic longitudinal section what I believe to be the relationship of the nephridia on either side of the body. There is no connection ventrally between the nephridia of the two sides of the body, but the ring of small nephridia is continuous across the middle dorsal line. In the anterior segments there is thus present a communication between the external ducts of the nephridia at a number of points in the circumference of the body wall, whilst in the posterior segments this communication is limited to the ventral surface, and we then find that there is here in each segment a longitudinal duct connecting together (1) a ring of smaller nephridia, (2) a single larger nephridium. *QJ.M.S. July 1884, Plate 24, fig. 3, and Plate 26, fig. 8. 26 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, The ducts leading to the exterior of the body vary slightly in structure in different parts of the body. 1.—In the clitellar region (Figs. 6 and 28) each duct passes through the longitudinal muscle layer, where, close to the ccelom, it may apparently be joined by branches from other ducts, and then runs on through the circular fibres. So far it is distinctly intra-cellular (Fig. 6, neph.) and is always somewhat coiled, so that the duct when cut in section has the characteristic appearance of a large number of sections of little ducts so arranged as to form a continuous string (cf. fig. 18.) This can be seen most clearly in the clitellar part, where the body wall is thicker. The duct on its way through the muscle layers is surrounded by a sheath of connective tissue fibres, and is always accompanied by a very distinct blood-vessel. (bv.) This vessel may sometimes, as represented by Bzpparp in Acanthodrilus,* form a loop round the nephridium, but more generally (as drawn in figs. 6 and 28) forms a loop at one side of the duct when the latter is passing through the special glandular development, which is characteristic of the clitellum. The intra-cellular duct passes just within the layer of gland cells, and there opens into the extra-cellular duct leading to the surface. The structure of this part was found to be most clearly shown in worms which had been killed by dropping them into a warm saturated solution of corrosive sublimate. In the ordinary spirit specimens, the structure, though discernible, after having once realised its true nature by means of the better preserved ones, is not nearly so clearly visible, owing doubtless to the greater contraction of the tissues.: The duct is of large size, and owing to its not running in a perfectly straight line, but being thrown into a series of small folds, it presents the appearance in section of a series of vesicular spaces lying closely one above the other, and often showing openings into each other. The walls are extremely fine and thin. In section the thin film-like wall of one side of the duct may be often seen (neph., fig. 28) its protoplasm, rendered visible by the presence of very fine granules indeed and oval nuclei of a large size, containing scattered chromatin fragments. The outlines of the cells, which thus form an extremely thin pavement epithelium, are not distinguish- able. Sections of the tube show at the edges the nuclei bulging out the thin walls (n.) When the surface of the body is reached, a very curious differentiation in the epidermis is seen. A certain number of them become so arranged as to form a sphere. The different cells composing this are swollen out medianly where the nucleus is placed, and taper off towards each end. The two ends are attached respectively to the two poles of the sphere, and along the axis of the latter passes a tubular cavity opening at the one end to the exterior, and at the other into the nephridial duct, the lining cells of which are directly continuous with those lining the axial cavity of the sphere. The cuticle does not pass down the duct. * Op. cit. Plate 30, fig. 10. THE GIANT EARTH-WORM OF GIPPSLAND. 27 There are one or two nuclei (n.*) close to the internal pole, the cells connected with which cannot be detected. They are similar in size and appearance to the nuclei of the cells forming the sphere, which again seem to be somewhat larger than those of the ordinary epidermic cells. Just where the duct enters the epidermic sphere are some 4-6 muscle cells, often with branched ends, which, from their arrangement, may be regarded as forming a sphincter for the opening of the duct, to the long axis of which they are placed at right angles (m.) Bxpparp, in describing a species of Acanthodrilus* in 1885, states that at the opening of the nephridium the columnar cells bend over towards each other on either side of the aperture. This appearance may perhaps be due to the development of a structure, such as is above described, but in his recent description of A. multiporus, he figures a different arrangement, in which the small cubical cells of the duct simply swell out into a sub-spherical shape within the epidermis. In the clitellar region, the ducts appear to pass right through the muscle layers, without branching as they approach the exterior. 2.—In the posterior region of the body (Figs. 18 and 27) the ducts are somewhat different, and resemble much those of Acanthodrilus. A single duct has the appearance represented in Fig. 18. The intra-cellular portion resembles exactly that above described in structure, but when this has passed through the longitudinal muscle fibre layer, it opens just within the circular fibre layer into an extra-cellular duct, which may (Fig. 18) run straight to the exterior without branching, or may give off one or more branches. Its cells are cubical, and this part forms a strong contrast in structure to the same division in the clitellar region. The external opening shows precisely the same modification of the epidermic cells into a sphere, through the axis of which runs the duct. The muscle cells do not appear to be developed. The blood-vessel accompanying the duct is always present, though not in the form of a coil, and runs forward to form branches, which will ultimately ramify amongst the muscle fibres and the epidermic cells. Very frequently, in fact most generally, the duct leading to the exterior may, as soon as it has passed through the layer of longitudinal fibres, branch. This branching commences when the duct becomes inter-cellular (Fig. 27), and each duct of which, there may be two or three, has its own separate opening. There is no distinction between the ducts of the larger and smaller nephridia, and ducts passing to the exterior arise from the longitudinal duct which are not definitely in connection with special nephridia, and the branches of the various ducts may pass over in the body wall from one segment to another. In no part of the body is there any relationship between the nephridiopores and the sete, even when the nephridia become more localised. Such a relationship is described by Brpparp as occurring in P. aspergillwm, in the posterior region of the body. * Proc. RS. June 1885, p. 460. 28 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, Reproductive Organs.—The macroscopic appearance of these structures has been described by Mr. Fiurcuer. When the body of the earth-worm is opened (Fig. 1), a series of racemose structures (vs.) are seen in certain of the anterior segments, overlapped and almost hidden from view by the strongly developed septa. They vary in development according, doubtless, partly to the age of the worm, and partly to the time of the year at which the animal is examined. In the eighteenth segment, a long closely coiled tube is seen on either side, pushing through the comparatively thin walls bounding the segment in which it lies. This is the prostate gland. Figs. 3 and 4 represent diagrammatically the relative positions of the different reproductive organs. Fig. 3 is supposed to be seen from the side, Fig. 4 from the dorsal surface. 1.—Male Organs. These consist of two pairs of testes (¢.’ ¢.”), attached close to the ventral surface to the posterior faces of the septa, bounding anteriorly the tenth and eleventh segments respectively. In addition to these there may occasionally be found an extra pair in the same position in the twelfth segment (¢.”) Hach testis is about 1-5 mm. in diameter, and consists of a small central portion, from which radiate numerous stiff processes, most largely in the plane of the septum, close to which the testis lies. In the tenth and eleventh segments are the two pairs of large ciliated openings of the vasa deferentia. These are white coloured, and have their margins, as usual, thrown into very deep folds. The whole internal surface of the funnel-shaped opening is ciliated, and full of ripe spermatozoa. No trace of the vasa deferentia can be seen on dissection; they lie in the connective tissue beneath the peritoneal epithelium, and run back in this position to the eighteenth segment, where they enter the prostate gland. The two vasa deferentia are completely separated from each other throughout their whole course (Fig. 30.) The great thickness and strength of the septa render it difficult to cut this part of the animal in section, but a complete series was obtained of longitudinal sections showing the whole course of the two vasa deferentia, which certainly run through a length of three, or in the fully expanded worm, even four inches in the body wall. The funnel is very close to the septum, and is composed of small ciliated cubical cells. It looks very thin when cut in sections, and forms—with its small cubical cells, only covered externally by a very small quantity of connective tissue, and its widely open mouth—a strong contrast in form and structure to the nephridial openings. The duct (Fig. 30) runs from each funnel through the septum, which is here very thick, but in doing so, does not run straight, but first of all curves upwards, coils about to a certain extent (not shown in the diagram), and then passing downwards towards the ventral surface, crosses over a special small bridge of muscle fibres and connective tissue (a), and so reaches the ventral body wall. Probably this THE GIANT EARTH-WORM OF GIPPSLAND. 29 extra length and the special bridge of muscle may be contrivances to allow of the sudden expansion and contraction of the great muscular septa in this part of the body, without hurt to the vas deferens. In the body wall the duct runs directly backwards (Figs. 3 and 4, vd.,' vd.’ and Fig. 30, vd.) The second opening and vas deferens have precisely the same relationship to the septum as the first ; and the two vasa deferentia in passing backwards lie close together, the second immediately above the first. This relationship is retained (Fig. 30) until the prostate gland (pr.) is reached. Through- out the entire course each tube contains, in its circumference, some eight cubical cells, never more apparently, and is always richly ciliate the whole length. In the eighteenth segment the two ducts run upwards within the connective tissue of the prostate gland, and then turning downwards again (Fig. 30) follow the course of the duct leading to the exterior, running through the strong circular and longitudinal muscles surrounding the opening of the prostate, until they enter side by side. The cilia stop exactly where the ducts enter the prostate. The Prostate Gland (Figs. 3 and 4, pr.) is, as said before, a large coiled mass on either side in the eighteenth segment. What its function is must be regarded as doubtful. It may be divided in Megascolides into two distinct parts, (1) an outer smaller part leading to the exterior, and serving both as the opening for the gland and for the vasa deferentia. This has simply (Fig 30) the form of a tube lined by a layer of deep and distinctly nucleated columnar cells, directly continuous with the epidermic cells at the mouth of the gland (mo.) In its upper part this tube is much coiled. It is surrounded by a great development of circular and longitudinal muscle fibres. Possibly the straight portion of the tube into which the vasa deferentia open and pour their contents (vo.), and which opens on the papilla before referred to in the swollen part of the clitellum in the eighteenth segment, may be eversible. There is no trace whatever, so far as could be seen macroscopically and by means of sections, of any penial setee. 2.—The second portion of the prostate, comprising almost all the coiled duct part, has a very different structure ; its walls are much thicker, and consist of two distinct layers—an inner and an outer. The inner is composed (Fig. 29, sp.’) of a single layer of cells, though occasionally these may be very long and extend into the second part. They are columnar in shape, and placed with their long axes radiating from the lumen of the tube (int.) Their chief peculiarity consists in the way in which they absorb staining material (borax carmine and hematoxylin), and hence appear in section to be very much darker than the outer-lying cells. The cells are, as a general rule, filled with a granular protoplasm, and much resemble in form those of an ordinary columnar epithelium, save that they are much more loosely placed together, and each has an internal rounded end, whilst (as represented in the figure) many of them may be larger than the others and extend far into the outer layer. Some of the cells have an empty appearance—as if they had poured their contents into the lumen—and in none of them can any nuclei be detected. 30 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, The cells of the outer layer on the other hand are much more numerous, and form a layer more than twice as thick as the inner ones. Lach cell has the form of a unicellular gland, with a more or less swollen external end and a somewhat narrow neck internally, which may be seen sometimes passing in between the cells of the inner layer. These stain much less readily than those of the outer layer, and in them, as a general rule, nuclei are to be seen. Possibly, the deep staining of the inner cells may hide the nuclei, but these are not even to be detected in occasional cells which are more lightly stained. Between the two layers is a shght development of connective tissue and a layer of blood-vessels, which are sometimes seen, as in the figure, branching along the plane of division between the inner and outer cells. Other branches ramify in and out amongst the gland cells of the organ. Externally is an encasing layer of connective tissue (ct), beneath the surface of this are numerous little masses of yellow-brown pigment spots, which give a general brownish tinge to the prostate gland. It is difficult to say what is the function, if any, or origin of these little definite masses of colouring matter. The connective tissue over the various coils of the ducts is continuous, and thus the latter are massed together to form on each side of the body a structure of considerable size. | The Vesicule Seminales. These are seen as racemose glands attached to the anterior surface of certain of the anterior septa, directly the body is opened from the dorsal surface. (Fig. 2, vs.) They differ very much in appearance from those of Lumbricus, not being pouch-like, and are usually present in the eleventh to the fourteenth segments inclusive, but may not be so largely developed in younger specimens. They are quite diagramatically shown in Fig. 4. Their appearance at first sight is very much like that of both testis and ovary, and they have once been described as the former. The organ is generally best developed in the fourteenth and least in the eleventh segment. Sections at once show its real nature and prove that it is the place in which the spermatozoa undergo almost their entire development. How they get in is not easy to understand. _ The testis (Fig. 12) consists in part of a great mass of germ cells, which have undergone but very little of their definitive development, and in part of a mass of protoplasm in which numbers of nuclei are scattered about forming merely a syneytium. The stalk of the testis attaching it to the septum is eomposed mainly of muscle fibres (m.) and connective tissue, which serve as a support for the structure. These run out into the ray-like processes, one of which is supposed to be cut in section in Fig. 12. It consists almost entirely of a mass of protoplasm, in which cell outlines can scarcely and only in very few parts (though this might possibly vary at different periods of the year) be distinguished, but in which are an immense number of nuclei with very distinct specks of chromatin. The nuclei are of two or three different sizes, and here and there the outline of a cell enclosing one or two of them may be seen (a.) In the central part of the testis the nuclei are small and similar, or even more minute ones may be seen in the stalk. These nuclei always appear to have a single spot of THE GIANT EARTH-WORM OF GIPPSLAND. 31 chromatin. The larger ones are in the rays where they all appear to be nearly of one size, and have very clear chromatin spots irregularly arranged around the periphery close to the nuclear membrane. No further development seems to be undergone in the testes, which can be found apparently at any time of the year. Worms secured in June, September and March contained them. They have at all times the same external appearance and the same microscopic structure, even in different worms in which the development (so far as size and extent is concerned) of the seminal reservoirs varies considerably. The evidence that these structures are really testes is both positive and negative. (qa) Positive (1) inasmuch as they contain structures indentical with the earliest stages in the development of spermatozoa, not only as found in other worms, but also as present in the seminal reservoirs of Megascolides. (2) Their position with regard to the segments of the body and relationship to the septa. (b) Negative—inasmuch as they differ uniformly in structure from the other reproductive organs of which it might be possible to regard them as early stages in development, viz., vesicule seminales and ovaries. In the vesiculz seminales, spermatozoa are, on the other hand, seen in every stage of development. The whole structure, besides occurring in scattered divisions in four distinct segments separated from each other by thick muscular septa, 1 divided up into a great series of capsular chambers, in which very different stages of development are met with at the same time. How the testicular cells get into the seminal reservoirs cannot be stated, and it is difficult to imagine how from the testes in the tenth and eleventh segments the germ can pass backwards into the fourteenth segment, as the basal opening allowing of the passage of the ventral nerve-cord is very small indeed, and the septa very thick. It is to be remembered, however, that the testes lie on the ventral side, and so close to the openings through which, in alternate expansion and contraction, currents of the ccelomic fluid doubtless flow. In Lumbricus, the testis is enclosed during growth of the seminal reservoirs, and in Microcheta the testis is enclosed in a special ccecum of the reservoir, but, as stated above, the testis of Megascolides is never enclosed, and can be easily found at all times of the year. ; Each capsule of the reservoir is encased by connective tissue, and within the walls ramify numerous blood-vessels. The same seminal capsule, as a general rule, contains sperm in very different stages of development. The different stages are represented diagramatically in Fig. 15, which is supposed to be a single capsule cut in section, and containing within itself the different stages, not necessarily seen in the one capsule, but to be found at the same time in the same animal. The youngest stage is probably the small mass of cells (3) still jomed together, but with their outlines distinguishable and their nuclei resembling exactly the largest ones 32 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, in the testis. Probably small masses of testicular cells break away from the testis, and in some manner are taken up by the vesicule seminales. The cells then separate from one another, and come to lie freely within the capsule. The great supply of blood-vessels is probably connected with the further development of the sperm cells, which increase greatly in size. Their nuclei divide, but so far as can be seen, the protoplasm of the cell does not do so. At first the nuclei, as in the figure in the left- hand corner (2), are somewhat spherical, and contain distinct chromatin spots. A later stage shows a number of oval dark-staining nuclei, arranged roughly round the periphery of a granular central mass of protoplasm. To this may now be given the name of spermatosphere. In certain of the spermatospheres, the nuclei take stain throughout, as in the figure. A somewhat later stage is to be found, in which processes of protoplasm project from the surface of the spermatosphere (sp.), each containing an oval nucleus, and in the latter the chromatin forms a crescent-shaped structure at the outer end—a curious stage corresponding to that already described and figured by Bloomfield as occurring in Lumbricus.* At this stage the central granular mass of protoplasm is clearly differentiated as the sperm-blastophore. Later on, the nuclei come to lie at the inner ends of the processes, and the protoplasm to gradually elongate into a pointed thread-like structure (sp.) At the same time, the developing spermatozoa all come to point in the one direction. Hach sperm element elongates, the nuclear end forming the head, the clear protoplasm the tail, and all the tails lie close together, so that the whole mass of spermatozoa forms a flame-shaped structure, in the broad end of which is the sperm-blastophore. When the former are mature, and ready to separate from one another, the latter cannot be detected, though possibly it is concealed from view by the elongate rod-shaped heads of the spermatozoa, which take the stain deeply. The ripe sperm comes finally to lie im the ccelom, probably either by discharge ot a capsule in which it is contained, or else by means of forcing its way through the capsule walls. There is no direct connection between the openings of the vasa deferentia and the seminal reservoirs, which are most largely developed in the three segments behind the one containing the second pair of openings of the male ducts. Within the vesicule seminales are, as usual, found capsules containing the embryos of a gregarine. 2.—Female Organs. These consist of a single pair of ovaries attached to the posterior face of the septum, between the twelfth and thirteenth segments, close on either side of the ventral line. ach has externally much the same appearance as a testis, but the projections from the small central mass are rather more racemose than radiate. The open end of the oviduct lies close to each ovary (Figs. 3 and 4, od.), and has the usual ciliate funnel shape. It is somewhat smaller than the corresponding * Q.J.M.LS. 1880, THE GIANT EARTH-WORM OF GIPPSLAND. 33 male opening, but is formed of the same small cubical cells. The oviduct leads down through the posterior septum of the thirteenth segment into the ventral body wall of the fourteenth. Its walls (Fig. 31) are composed of columnar ciliate cells, much more numerous than in the case of the vasa deferentia. The two ducts incline towards the median line, and open one on each side, very close together, in the middle of the fourteenth segment. They have no relation to the sete, opening in front of and to the ventral side of these. Though the testis and ovary closely resemble each other externally, sections at once show a great difference between the two. (Figs. 11 and 12.) At all times, so far as can be ascertained—certainly in winter, spring, and summer—the ovary contains large well-developed ova. A section through an ovary of a worm obtained in March is represented in Fig. 11, as seen under a magnifying power of 2,200 diameters. Only one of the projections from the central part is represented. The whole ovary has a diameter of rather more than 1 mm. The central part is composed, very much as the testis, of a mass of protoplasm, amongst which are muscle fibres and connective tissue serving as supporting structures (ct.) In the ovary ramify numerous blood-vessels (6v.) In parts outlines of cells may be distinguished, but as a rule the protoplasm forms an indefinite mass, in which many nuclei of various sizes are scattered. The smallest (0.’) are very minute, and resemble those in the testis stalk. Others, gradually increasing in size, are seen (0.* 0.2) Around these the protoplasm is difficult to distinguish, since it takes stain but slightly. Larger nuclei (0.’) lie further away from the centre of the ovary, and around them the protoplasm is becoming marked off into distinct areas, and also differs in its ability to take stain. In the nuclei the chromatin fragments are arranged roughly in a circle near to the external surface. At the extremities of the projections, the cells are still larger (0.’), and have the appearance of mature ova. These cells have undergone great growth, and their nuclei are very large and distinct, and generally spherical in shape. In the largest one, which forms the extremity of one projection, its inner end being buried deeply amongst smaller cells, the nuclear membrane can be distinguished, together with a nuclear network with chromatin fragments, and a somewhat eccentrically placed mass of chromatin, in which lies a still more darkly-stained portion. In the other large cells, owing doubtless to the different results produced by reagents, the chromatin appears distributed in various ways, but there is always one large main portion placed eccentrically, and besides this, there may be smaller fragments irregularly scattered. These ova grow in size, nutriment being possibly afforded to them by the absorption of the smaller cells at their inner ends, as well as by the numerous blood-vessels in the ovary, and when mature they fall into the ccelom, and enter the open ciliated end of the oviduct, which lies close beneath the ovary on either side. As far as development goes, a small number of cocoons have been secured, and I hope goon to be able to procure many more, and to work out the development F 34 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, completely. The cocoon is very large—13-2 in. in length and 3-1 in. in breadth. It is leathery in consistence, rounded at either end, with a short coiled string left where the ends were finally closed. The cocoon corresponds to the size of the burrow, and each contains only one embryo. What size the embryo attains before leaving the cocoon I cannot say, but have found them coiled up within the cocoon, and reaching the length of 5-6 in. The cocoon is filled when the embryo is young with a milk-white fluid, corresponding to, though rather thicker than, the coelomic fluid, and is probably composed in part of the latter. The spermathece are four in number, and very large. There is one pair in the eiohth and one in the ninth segment. (Figs. 1, 3, and 4, 7.s.) Each one has the form (Plate 5, fig. 23) of a bag with a pointed extremity at its free end, and a short stalk for attachment to the body wall at the swollen end. Hach lies on the ventral surface of the body, and their ducts to the exterior pass slightly forward, so as to open just (Fig. 1, r.s.* 0., r,s.” 0.) within the posterior edge of the segment next in front of those in which they really le. A slight line on the surface of the worm encloses, as shown in the figure, a distinct area in which the receptacula open. No setze can be detected so far forward in the body of the worm I have examined, but these openings lie close to the segment boundaries, and would seem to have no relations to sete. They open at about the same distance from the median line as the ventral pair of sete, but quite in the posterior region of the segment. The receptacula are marked by very distinct grooves running longitudinally (Fig. 23), and up each side passes a clearly-marked muscular slip (musc.) This may perhaps be useful in ejecting the contents of the receptacle, as on contraction the two slips would force the contained material to the exterior. The grooves all converge towards the apex, and on examining the surface carefully, circular muscle fibres can be detected running beneath the longitudinal ones on the exterior. At the attached end, and very close to the body wall, is a small diverticulum, as in other species of the genera described by Fiercuer, though in them (M. (Notoscolex) camdenensis and M. grandis) the spermathecze are much narrower and more tube-like, and in the former the diverticulum more prominent and further away from the external opening. The surface of the diverticulum in M. australis is marked by projecting coils, which are not well represented in the lithograph. When cut in section longitudinally, the structures represented in Fig. 25 are seen. Internally, the sac is lined by a layer of columnar epithelium, the oval nuclei of which lie close to their outer ends. Most externally are the longitudinal and inter-twined muscle fibres ; just within these the circular fibres, much fewer in number, are cut through. Between the latter and the epithelium is a considerable space filled with connective tissue, amongst which are many nuclei, whilst next to the epithelial cells is a layer formed of very numerous branching and interlocking blood-vessels. The great development of the blood-vessels in connection with the columnar epithelium cells indicates the fact that the inner lining of the spermatheca has a secretive function. In examining the contents of the THE GIANT EARTH-WORM OF GIPPSLAND. 35 spermathecz in several worms at different seasons, I was surprised to find no trace of spermatozoa, but simply remains of a fluid containing granules and masses of nucleate corpuscles. The latter are aggregated into the form of spheres, and it is difficult to say what is their real nature. Sections showed that the spermatozoa, at all events in those examined, were confined to the diverticulum, near the base, and in its hollow spaces they were densely crowded together. The space follows the external markings, and the whole communicates with the main cavity of the spermatheca by means of a narrow passage. The walls of the diverticulum are devoid of the columnar epithelium, and formed mainly of connective tissue. It is possible that the walls of the main cavity may secrete a fluid from the blood, which is necessary to maintain alive the spermatozoa until such time as they are needed to be placed in the cocoon. According to Jackson,* Vespovsky states that the fluid forms the spermatophores. Remarks on the Nephridial System.t—The most interesting points in the anatomy of Megascolides are concerned with the nephridial system. During the last few years our knowledge of the structure and arrangement of the nephridia of Chetopods has been very much enlarged, and though in the absence of fuller information concerning the development of the organs, it is not perhaps possible to arrive at any final conclusions, still the data available is sufficient to warrant a comparison of the Chetopod nephridia with those of certain other worms. Such a comparison has lately been attempted by Bepparp in the light of his discoveries with regard to the structures in Acanthodrilus and Pericheta. The nephridia of Megascolides may perhaps serve to render some points still more clear.{ Bzpparp has argued very forcibly in favour of the view of a direct relationship existing between the nephridia of Hirudinea and Chetopoda on the one hand, and of the Platyhelminthes on the other. Certainly the existence of provisional larval nephridia in various Hirudinee, Polychete, and Oligochete is a point of great difficulty in homologizing the permanent nephridia of the latter forms with those of Platyhelminthes. Burau, as quoted by Bepparp, holds that the above mentioned larval structures, together with those of Mollusca, are homologous with the nephridial system of the Platyhelminthes. Whilst on the one hand, Bereu asserts that the larval nephridia of the Hirudinea are * Forms of Animal Life. 2nd edit., p. 206. + The suggestions made in the following pages differ somewhat from those made in a letter on the same subject, which was published in Wature, June 28, 1888. t In the absence of access to any but a very meagre supply of original literature on this subject in Melbourne, I am indebted both for the facts stated with regard to other worms, and the views held by other investigators, to the valuable series of papers by Bennam and Bepparp, and the second edition by Jackson of “The Forms of Animal Life,” and to the memoir of Bourne for facts regarding the Hirudinece. Pp 2 36 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, quite independent of the adult ones, Wuirman states that they are not. Brpparp further points out that the larval nephridia of Oligochete discovered by Vuspovsky “occur at the anterior end of the body, where no permanent nephridia are developed. Furthermore, these organs lie in the ccelom, perforating the mesentery which separates the first from the second segment; hence Brereu’s objection to the homology between the larval and permanent nephridia, on the score that the former do not lie in the true coelom, is removed.” In the Platyheluunthes the excretory system has the form of a series of fine tubes formed of perforated cells, the terminal one of each branchlet being a flame cell. The latter may be wanting in the larval nephridia of other worms, though otherwise the structure of both sets is similar; and in others, again, where no free larval stage is present, the structures may still more lose their resemblance to one another. In Dinophilus also, which, according to Wetpon,* is “a form representing in its main features a stage in the evolution of Chetopods,” and possessing undoubted Turbellarian affinities, the excretory system resembles that of a Polychete larva. ‘The difficulties seem to lie in—(1) the presence of larval nephridia, which have, as Brrcu supposes, no connection with the adult nephridia ; (2) the structural resemblances between the adult nephridia of Platyhelmunthes, and the larval ones of Chetopods and Hirudinee. The question of connection with the cclom of the organs, supposed by Bzraxu to be homologous in the three latter groups, is not perhaps of such importance. In the first place, according to Brpparp, the larval nephridia of Oligochete, discovered by VEsDOVSKY, do open into the ecelom ; and beyond this it would scarcely be safe to conclude that structures present in Chetopods and Hirudinece, and evidently homologous in the adults of the two latter groups, were not homologous with structures in Platyhelminthes, simply because of the nature of the cavities into which they opened ; especially also taking into account the fact quoted by Bzpparp, that in Capitellidee the nephridia le within the mesoderm, and not in the ccelom, and that in Polygordius the greater part of the nephridium is similarly situated. As to the difficulties arising out of the fact that two sets of nephridia are present, it must be remembered, first of all, that, apart from Harscumr’s observations on Polygordius, WuitMaN states that there is a connection between the structures in Hirudinee. Further, it is a point of importance, as Brepparp points out, that these are always placed at the anterior end of the body, and so do not correspond in position with the nephridia of Platyhelmanthes. It is possible again that these simple larval nephridia are purely larval structures, not phylogenetically related to those of the adult forms of their ancestors; or that, again, they are portions of the adult nephridia precociously developed, just as portions of other structures are developed in the larve for use during larval stages, in either of which cases the great obstacles to comparing the adult nephridia of earth-worms and Platyhelminthes would be done away with. * Q.J.M.S. August 1886, p. 117, THE GIANT EARTH-WORM OF GIPPSLAND. S7/ These, of course, must remain mere conjectures, unless the existence of a definite connection between the larval and adult structures be established, and Hatscuex’s and Wuirman’s observations confirmed. Meanwhile, it is interesting and suggestive to note the variations in structure met with in the nephridia amongst the members of the groups Cheetopoda and Hirudinea, leading back, as it were, to the Platyhelminth nephridial system. There are now known a series of gradations between the single pair of highly developed nephridia, quite separate from each other, and placed one in each segment of the body, and a well developed irregular nephridial network with no internal funnel-openings, and only one pair of posteriorly placed external openings, or an enormous number of minute nephridial tubules, with a connecting network, and a great number of irregularly arranged openings in each segment. It must be remembered that in Platyhelminthes there are no structures which can be exactly said to be homologous with the nephridial tubules of Chetopods, the network of ducts lying in the body wall beneath the peritoneal epithelium in such forms as Pericheta and Acanthodrilus, and more espetially in the Hirudinean Pontobdella, being, I would suggest, really the structures directly homologous with the Platyhelminth network of tubes, whilst the nephridia of Chetopods themselves are to be regarded as outgrowths and special developments of these tubes, their formation being to a large extent associated with the development of special sinuses and spaces within the mesoderm into which they depend. BeEpparp is of opinion that “it is unnecessary to regard the funnels of the Annelida as new structures,” and draws attention to the fact (1) that in Stylaria the single cell, which by its proliferation forms the funnel, becomes ciliated, and acquires a lumen before it undergoes division ; (2) that in Clepsine the funnel only consists of two cells. Lane, on the other hand, regards them as new structures, not represented in Platyhelnunthes, and Ep. Merysr, as quoted by Harmer,* has shown that in Polycheta (Terebella) the ciliated funnel arises quite independently of the body of the nephridium—a fact strongly in favour of Lane’s view, as is also the varied deyelop- ment of the nephridia, as seen in a series of adult forms within the limits of the Oligochate. The development of the funnels in Clepsine, and more especially in Pontobdella, would be of great interest and importance in connection with this question. Meanwhile, seeing further that the funnels are not present in what must be regarded as the most primitive nephridial system amongst Chetopods, and only appear in forms in which the organs are somewhat highly differentiated, it is perhaps safer to conclude that at all events in Chetopoda the funnels are new structures, not represented in Platyhelawnthes. From this point we may deal separately with the Hirudinea and the Chietopoda, as I would suggest that though the nephridial system of each is derived from one of a *Q.J.M.S,, 1885. P, 280. 38 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, Platyhelminth type, yet the development of the two groups has proceeded, starting from the same point, along somewhat divergent lines. The difference is due to the greater development of the ccelomic space in the Cheetopoda than in the Hirudinea, and the more complete division of the body of the former by means of septa into a series of compartments, each almost completely separated off from the ones on either side of it. There is at the present time a strong contrast in the two groups in this respect, and it is one which has probably held true since the time at which each group separately branched off from common Platyhelmunth-like ancestors to pursue its own course of development. We may believe this to be true, even though at the same time we agree with Bournz* that the ccelom of the leeches has once been more highly developed than it is now, and that “the leeches have thus had an ancestor which, in possessing a ccelom, was already a great advance upon any Platyhelminth form.” It was the development of this definite space in the mesoderm, which resulted in, or at all events was closely connected with, the formation of the definite internal openings of the nephridia in the leeches and, as I would suggest, of the present nephridial tubes of the Chetopoda. 1.—The Hirudinea.—The simplest form of nephridial system known in leeches is that of Branchellion. As described by A. G. Bournz,t this consists only of a network of tubules, with no internal openings, and only a single pair of posteriorly placed external openings. The only indication of segmentation lies in a slight crowding of the tubules together at intervals. There can be little question of the close relationship of this to the system as found in the Platyhelminthes, and it might possibly be that further examination of the living form would result in the finding of flame cells. In Pontobdellayt a similar network is present, but there 1s in addition a segmental arrangement clearly indicated by the presence in each segment of an internal pair of ciliated funnels, and also of a pair of external openings. The former open into a definite coelomic space on each side of the body. The resemblance between the network, again, of Pontobdella and a Platyhelminth is close; the difference between the two lies in the presence of the two sets of openings. It must be noted that Lane has shown that a definite, though rudimentary, metameric arrangement of the nephridial tubes obtains in some planarians in which secondary external openings may be developed by means of branches from the longitudinal trunks, and that a paired arrangement of these may even be seen. The ciliated funnel is united to the network by a very short tube. Its development, as above suggested, may be supposed to be connected with the presence of the definite Op. cit. BP, 499. + Op. cit. P. 481. t Op. cit, BP. 478. THE GIANT EARTH-WORM OF GIPPSLAND. 39 ccelomiec space. Whether it is an entirely new structure, or developed as a modification of a pre-existing flame cell, it is quite impossible to say definitely, though the simplicity of the structure in such leeches as Clepsine may perhaps indicate its development from a flame cell, more especially if, as suggested by Hartoe,* a flame cell be really an internally ciliated cell. The segmental arrangement of the network, almost absent in Branchellion, and indicated in Pontobdella, is carried to a much greater extent in other leeches, the segmentally arranged portions losing their connection with each other, and each one acquiring its internal and external openings. The ducts, though aggregated into a definite tubular structure, still lie embedded in the mesodermic tissue, only their internal funnels lying freely in the ccelomic spaces, and having short tubes which connect them with the ageregated portion of the network. In Hurudo itself there still persists a part which Bourne regards as the remnant of a formerly existing connection between one nephridium and another. The stages of development in AHirudinea may be, briefly, somewhat as follows :— (a) A complete network of tubules (certainly at first with “flame cells,”) and with posterior openings, and no, or only the faintest, trace of segmental arrangement. Such a stage is now possibly seen in Branchellion. (6) A complete network of tubules, with paired external openings, and internal ciliated funnels opening into definite coelomic spaces. This stage is now seen in Pontobdella. The internal openings may be developed by modification of pre-existing flame cells, or be entirely new formations. (c) An aggregation of the network into portions segmentally arranged, forming definite paired nephridial tubes, each independent of the others, and with its own internal and external openings. This now exists in the adults of most leeches, as Hirudo and Clepsine. 2.—Inthe Chetopoda. An important difference in the nephridia of this group, as compared with those of Hirudinea, seems to have been brought about by the development of much larger and more extensive spaces within the mesoderm, which have moreover a segmental arrangement, and but very little lmear connection with each other. Taking the group as a whole, we find two sets of structures present in connection with the nephridial system :— (a) A network of tubules in the body wall, and hence embedded in mesoderm. (b) A series of regularly or irregularly arranged coiled tubes depending from (a) into the successive ccelomic chambers. With them may, or may not, be associated definite nephridial funnels. * Ann. Mag. Nat. Hist. 1887. P. 326. 40 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, Both of these structures are present in what are to be, without much doubt, regarded as the more primitive form of nephridia in Chwtopods; whilst in more modified ones, the second set of structures is present only. Of the two, the first is to be regarded as directly homologous with the network present in Platyhelminthes, whilst the second is a secondary development. This gives a somewhat new view as to the relationship existing between the various nephridia of the groups, and differs from that of Brpparp and others in regarding the highly developed nephridia of such forms as, for example, Lwmbricus, as secondary developments, formed primitively as outgrowths from a nephridial network homologous with that of Platyhelnunthes and still persisting in Pericheta, their existence being intimately associated with the formation of definite sinuses and- spaces within the mesoderm. A series of gradations is found amongst adult forms in the number, structure, and arrangement of the nephridia. In what may be regarded as the more primitive ones—A. multiporus, P. aspergillum, &&.— there is present a regular network of ducts with coiled nephridial tubules depending from it into the ccelom, and very many in number in each seoment. In P. aspergillwm the simplest form is seen in which the nephridia are quite irregularly scattered, and the network is a continuous one from segment to segment. In A. multiporus an ageregation of the nephridia takes place in the posterior region of the body, with a consequent lessening of the extent of the network, which is not continuous from seoment to segment. In the more highly developed forms the nephridia are restricted in number, usually, though not always, (as, for example, in Capztellide amongst the Polychetw) asingle pair being present in each segment, as in Lumbricus. Each nephridium is in Chwtopods considerably larger and more highly developed than in the case of those where many are present. In one adult Polychwte—Lanice conchilega*—a longitudinal duct connecting consecutive nephridia exists in the adult. A similar (?) longitudinal duct has been described as present in the embryo of Lumbricus and Criodrilus, placing the various nephridia in connection with one another, and Harscunk’s well known description of the development of the nephridia in Polygordius indicates a definite connection between the successive ones in this form. : Between the two extreme forms, in one of which a complete network with very numerous small nephridia with no internal openings are present, and the other, in in which a pair of large nephridia with internal openings is developed in each segment in no serial connection with each other, at all events in the adult, a most important intermediate stage is found in Megascolides australis. ‘This serves in a remarkable manner to bridge over the wide interval, and to show the path by which the nephridia of Luwmbricus may perhaps have been developed from the primitive original network of ducts in a Platyhelminth ancestor. It possesses both large and * CunnincHAamM. Q,.J.M.S. Nov. 1887, p. 250. THE GIANT EARTH-WORM OF GIPPSLAND. Al small nephridia, the latter alone present at the anterior end of the body, whilst both are developed in each segment posteriorly. The latter only have ciliated funnels opening internally. Each consists of two distinct portions, in one of which the duct is tracellular, in the other the duct is intercellular, and opens to the exterior. The latter may be supposed to represent the more highly developed nephridial opening with vesicular portion present in other worms. In the larger nephridia, a third part is present at the intercellular opening of the nephridial funnel into the ccelom. This may in various worms be of greater or less extent, but in Megascolides appears to be very small. In A. multiporus,* Bspparp describes an aggregation of the nephridial tubules into eight definite tufts, one corresponding to each seta in the hinder part of the body. In some cases two nephridiopores correspond to a single seta. The aggregation is not quite perfected, as “occasionally a single tubule was observed to perforate the body wall between the sete” in positions corresponding to septa of connective tissue, which break the continuity of the longitudinal muscle layer. Bepparp suggests that such septa represent the last trace of setee, which have now disappeared. These irregularities are interesting, as showing that the present condition of the nephridia in the posterior part of the body in A. mwultiporus is due to an aggregation ot the nephridial tubules, each tuft connected with a seta being the result of a massing together of a certain number of nephridial tubules to form one mass, as evidence of which may be instanced the occasional presence of more than one nephridiopore. In Megascolides, the differentiation has been at once carried further, and along somewhat different lines, and it must be noticed that the differentiation in - both A. multiporus and Megascolides commences in the posterior region of the body— a point of importance when we come to consider the relationship of the nephridia to the genital ducts. In Megascolides, the small nephridia are very much more numerous in the anterior part of the body than in the posterior; but side by side with this, a secondary development of large nephridia has taken place in the latter. If we examine the nephridia in situ we find, as previously described, that the large ones are only present in the anterior half of the last hundred or so segments of the body, a ring of smaller ones being present in the posterior half of each segment. In the middle region of the body, the large nephridia can for a certain number of segments be distinguished, but with no internal openings, and gradually come to form one of a group of nephridia slightly larger than the rest. Passing forwards, the single one becomes less and less distinguishable in size from the other members of the group of which it thus forms one, *I cannot tell from Brepparp’s paper whether the Acanthodrilus described there is the same form which he had previously described in the P.R.S. (No. 238. 1885) as possessing eight nephridia in each segment—one corresponding to each seta. Both worms came from New Zealand. In the first paper the nephridia are described as being single structures, in the second as being tufts of nephridial tubes, the external orifices of which are, as a general, though not universal, rule, associated with the sete. If they are not identical forms, it is very interesting to see the aggregation perfected in the first worm described, with the result that one definite nephridium corresponds to each seta. G 4g ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, until it completely merges into the rest, and nothing but small nephridia are present, which, as the anterior end of the body is approached, cover the whole body wall internally. In addition to this, where the larger nephridia are present, there exist (1) a network of ducts connecting the smaller nephridia in each segment, and (2) a well-developed longitudinal ventral duct on each side, running from the smaller nephridia forwards to the larger nephridium, and then continued on, but much finer in structure, to the ventral surface of the next segment, where it communicates with the longitudinal duct passing to the next large nephridium. These series of structures reveal (1) the method of formation of the large nephridium from an aggregation of small ones, with a subsequent formation of an internal opening, and (2) the loss of a continuous network in the anterior part of the segment, as the smaller nephridia have become aggregated towards the ventral surface, still retaining their connection with the other nephridia, and hence the final result of a ventral single duct when the single nephridium becomes established. In Megascolides, all the smaller nephridia are not, so to speak, used up in forming the one pair of large nephridia, but we can easily suppose a case in which, as in Pericheta aspergillum, there is primitively a large number of nephridial tufts in each segment, with a network continuous from segment to segment. If these become aggregated, so as to form a single pair of nephridia in each segment, then as they pass down towards the ventral surface (supposing the aggregation, as has usually happened, to take place in that direction), the network still connecting the nephridia, if it persists at all, will eventually assume the form of a longitudinal duct, passing from one group of nephridial tufts or one single nephridium, according to the stage of development arrived at, to the next in order along the ventral surface. It will be seen from this that there is no necessity to assume that a longitudinal duct, when present, is the homologue of a longitudinal duct of a Platyhelmunth, but is rather to be regarded as a modification of the network of ducts brought about by aggregation of the nephridia (1) into a series of tufts, and (2) the further develop- ment of these into large nephridia. The simplest nephridial systems of Hirudinea and Chetopoda alike, present no structures similar to the longitudinal duct of a Platyhelminth. This, at all events, does away with the necessity of supposing a double origin for the nephridial systems of Chetopoda,* one from ancestors with a network of ducts, the other from ancestors with a pair of longitudinal ducts, and at the same time, upon different grounds, brings us to the same conclusion as BEpDARD— “that the longitudinal ducts of Lumbricus, Lanice, &e. (we may now add Megascolides), have not any relationship to that of the Platyhelminths.” One great difficulty remains in the fact that, according to Wutson, the longitudinal duct in the embryo Lumbricus is epiblastic in origin. Until the development of the nephridial systems in the various forms has been consistently * BEDDARD, Op. cit. P. 408. THE GIANT EARTH-WORM OF GIPPSLAND. 43 studied, it will be impossible to establish any certain homologies. ‘Meanwhile, it is important to remember that in Chwtopods there is a very clear distinction of the nephridial duct into two parts—(1) an intracellular part, and (2) an intercellular part leading to the exterior, connected with which is the vesicular part. It may be that the first part is always mesoblastic in origin, the second epiblastic. In this case, the longitudinal duct and network of Megascolides and the network of other forms are mesoblastic, and hence different in origin from the duct present in the embryo Lumbricus. These points can only be determined by further, and especially embryological, investigations. The various stages in the development of the Chetopod nephridia may perhaps be somewhat as follows :— (1) A stage (in some Platyhelminth-like ancestor) in which in an unsegmented body a continuous network of nephridial tubules, with flame or internally-ciliated cells, the former uniting to form longitudinal canals leading to the exterior. (2) A modification (as seen in Dinophilus gyrociliatus) in which the excretory organs are still in the form of a network, with flame cells, but with secondary external openings in each segment, irregularly arranged, as in some planarians, regularly arranged, as in Dinophilus. (3) A further modification, resulting in the formation of numerous irregularly- arranged outgrowths from the nephridial network, having the nature of coiled tubules, which are directly continuous and identical in structure (duct intracellular) with the network. These form the nephridia of the more highly-developed worms. Their development is to be regarded as intimately associated with that of segmentally- arranged coelomic chambers, such as are at any rate but feebly represented in the Hirudinea, in which the tubes of the network itself, still lying within the mesoderm, become aggregated to form the nephridia. (4) In connection with these numerous nephridial tubes, many external openings leading into the still persisting network are formed. (P. aspergillwm.) (5) The aggregation of the small nephridia into groups, commencing in’ the posterior region of the body (as in A. multiporus and M. australis). The nephridia are sometimes aggregated in relationship to the sete, and as the aggregation proceeds so the external openings diminish in number and the network lessens in extent. (6) The formation of large nephridia either out of an aggregate of small nephridia, or by the special growth of one of an aggregation of small nephridia. Each large nephridium acquires secondarily an internal opening into the ccelom. These openings, which have a very definite relationship to the ccelomic chambers, are formed apparently later in the developmental history of the nephridia in Chwtopod than Hirudinea, and cannot be supposed to be related to the flame cells, but to be new formations within the group. Their formation also proceeds from the posterior towards the anterior end, and as it goes on the small nephridia gradually diminish in Ga 2 44 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, number, and the network undergoes corresponding decrease with, it may be, as in Megascolides, the formation of a single duct on each side of the segment in connection with a network in the parts where the small nephridia still persist. (7) The final disappearance of all trace of the small nephridia, and with them of the network and longitudinal duct. There then remains in each segment, as in most adult earth-worms, a limited actor ate one pair—of large nephridia, with internal and external openings. Connection of Nephridial Organs and Genital Ducts.—LanxustER was the first to suggest a probable connection between the two structures in earth-worms, and also to suggest that in the earth-worm two sets of nephridia had primitively been developed,* of which one now persists as the nephridial organs; the other only remains in the segments bearing the genital organs, and has become modified into genital ducts, the nephridia of this system having hence disappeared everywhere save in the genital segments. These two sets of nephridia were supposed to be related to the two pairs of setee—the persistent nephridia to the ventral, and the nephridia now persisting as genital ducts to the dorsal pair of sete. Subsequent investigations have shown that he was perfectly right in supposing that primitively there was more than one pair of nephridia to each segment, but it is now very doubtful indeed whether there is much evidence in favour of his theory that the genital ducts are to be regarded as modified nephridia in the Terricole. The evidence derived both from the structure of the two sets of organs, and from a consideration of the state of development of the respective structures in different earth-worms is opposed to the truth of his view. . First, with regard to the structure, which is very suggestive, though not perhaps of so great importance as the second class of evidence. The general structure of a highly developed (but not primitive) nephridium in an earth-worm is that of a tube with a funnel-shaped opening into the ccelom at one end, and an opening on to the surface of the body at the other extremity. In some part of the funnel-shaped structure the duct is inter-cellular, but the funnel, apparently without exception, leads back into an wntra-cellular duct, always of considerable length and complication, and it is this «intra-cellular duct which forms the most characteristic portion of a nephridium, and which is never absent, however slightly the wter-cellular portion may be developed. The more primitive the nephridium (as in Acanthodrilus, Megascolides, &¢.,) the less developed becomes the inter-cellular part, which forms a duct leading from the intra-cellular part to the surface, and in connection with which—but only in the most highly developed forms—a vesicle may be developed. On the other hand, the whole of the genital duct ws inter-cellular, so that of the vasa deferentia or oviducts correspond to nephridia, all the intra-cellular portion of the duct between the funnel and the inter-cellular duct opening externally must have been entirely lost. * Q.J.M.S., 1865, p. 18. On the Anatomy of LZumbricus. THE GIANT EARTH-WORM OF GIPPSLAND. 45 Benuay, in his valuable general summary* of the various organs of earth-worms, dealing with the question of the homology of the genital ducts and nephridia, says that the modification which the nephridium undergoes to form a genital duct consists either in— (a) A fusion of a series of nephridia, or (b) A disappearance of a part of the nephridium, or (c) A shifting of the position of the pore. Now, so far as our present information goes, we have no actual evidence of any one of these occurrences taking place to form a genital duct. Brnuam says, “In the case of the male duct, each of these modifications is exhibited. In the somites, in which le the ciliated rosettes, the external extremity of the nephridium has disappeared. In the somite carrying the male pore, the funnel region of the nephridium is absent, whilst in the intervening somites both these regions have aborted, and a fusion of these various parts has taken place to form the more or less elongated duct.” Now, as above stated, the whole of the male duct is inter-cellular in nature, whereas, according to this suggestion, the only parts of the nephridia which are inter-cellular have disappeared, save the ciliated rosettes at the commencement of each tube and the external openings at their terminations, leaving in the intermediate segments merely the typical intra-cellular parts of the nephridia to form a duct, the vas deferens which is characterised by its uniformly inter-cellular character ! Other difficulties again arise in connection with the male duct, as, for example, the presence of two perfectly distinct ducts running side by side in Megascolides, and of even “four separate sperm ducts, each with its external pore in Acanthodrilus and Monihigaster.”+ Sumilarly m the oviduct all intra-cellular portions of the duct must have disappeared, and the same again with the spermathece; in fact, the total disappearance of what forms the most characteristic portion of the nephridia is, even on the assumption that when converted into genital ducts the nephridia lost their primitive function, a very remarkable occurrence. If this difficulty stood alone it need not perhaps be regarded as an insuperable objection, but there is in addition the second class of evidence derived from an investigation of the respective development of the two sets of structures in various earth-worms. The latter we may, simply as far as their nephridia are concerned, divide into three sets :— (1) Those with very numerous small nephridia in each segment, with no internal but many or several external openings in each segment, which may or may not have a definite relationship to the sete. (Pericheta, Acanthodrilus, &c.) * Q.J.M.S., Feb. 1886, p. 265. + Benuam. Op. cit., p. 260. 46 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, (2) Those with numerous small nephridia in each segment, similar to those in (1), and with, in addition, large nephridia in certain segments with internal openings (only example known as yet, Megascolides). (3) Those with a limited number of large nephridia only in each segment— usually one pair (Lumbricus, &c.) Now of these three divisions, there can be little doubt that the one characterised by the presence of a great number of nephridia scattered over the body wall without definite arrangement is in a more primitive condition, as far as its nephridia are concerned, than the one characterised by the presence of a single highly-developed pair in each segment of the body. Such a worm as P. aspergillum, with its very numerous scattered nephridia, stands in very much the same relationship to other worms in this respect as that in which Peripatus stands to other Arthropods in the matter of tracheal tubes. There is only one other course open, and that is to regard the condition found in LIumbricus as a primitive one, and that in Acanthodrilus as not more highly developed, which could scarcely be held but degenerate. This view might be held, but there is, ~ so far as I can see, no evidence in its favour, and it would be contrary to all analogy. We must regard P. aspergiulwm as more primitive in its nephridial system than Iumbricus. Between them and bridging over the great differences in structure in these two extreme types come (1) Acanthodrilus, and (2) Megascolides. In both of these we get differentiation, rendering them more highly developed than Pericheta, and giving indications of the final state reached in Lumbricus. The most primitive nephridium with which we are hence acquainted in Olugochete has no internal opening resembling in this respect the ancestral Platyhelminth, from which it may be supposed to have been derived. Such nephridia are characteristic of P. aspergillum and A. multiporus, which do not appear to have yet reached the stage in which the secondarily-formed ciliated funnels place the nephridial ducts in connection with the ccelom. In the Perichete worm, the small nephridia are in their most primitive condition, and still irregularly scattered; but in Acanthodrilus and Megascolides the system is somewhat more highly differentiated. In both of them the modifications commence at the posterior end of the body, resulting in the former in the aggregation of the nephridia corresponding with the setze in position, and in the latter in the formation posteriorly of large nephridia with internal openings. At the anterior end the small nephridia are present in such enormous numbers, more especially in the segments containing the genital organs, that they line closely the whole body wall. At the same time, the genital ducts and copulatory pouches are equally highly developed in all the forms. It thus follows that genital ducts, highly developed, and consisting, amongst others, of vasa deferentia, traversing no fewer than seven or eight segments, are developed in certain earth-worms (Pericheta and Acanthodrilus), in which the very numerous primitive nephridia have no internal openings; and in another THE GIANT EARTH-WORM OF GIPPSLAND. A7 (Megascolides) in which highly-developed nephridia, with internal openings, are only _ present in the posterior region of the body. If the hypothesis with which we started be correct, viz., that the Pericheta and the Acanthodrilus are primitive forms, so far as their nephridial system is concerned, it follows that in these forms, and hence presumably in all Terricole, the genital ducts have no connection with the nephridia, and are not to be regarded as nephridia specially modified to serve the purpose of conveying genital products to the exterior. A nephridium, before it can function as a genital duct, must have an internal ciliated funnel, and in Pericheta and Acanthodrilus the nephridia have not reached this stage of development, and yet genital ducts are present, which not only have internal openings, but differ in important respects histologically from the nephridia. The relationship of the openings of the ducts to the setze, or the absence of such arrangement, can by itself be regarded as evidence of but slight value, and yet it is really the only direct evidence, if such it can be considered, in favour of a connection between the two sets of structures. It is perfectly possible that the same causes which have apparently operated, with the result that a relationship often exists between the nephridial openings and the setz, operated quite independently to produce the same result in the case of the genital ducts. Further investigations, especially with regard to the development of the genital ducts, may, if for example it could be proved that the nephridial system of P. aspergillum, Megascolides, &c., are degenerate, give some clue to the homologies of the genital ducts. At the present time our knowledge seems to indicate that in earth-worms an homology does not exist between these and the nephridia. 48 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, EXPLANATION OF THE PLATES. PLATE 1. bc ... buccal cavity. clt ... Special swellings in the clitellar region. dil ... dilatation of alimentary canal. dp ... dorsal pore. dv ... dorsal blood-vessel. fo ... female genital opening. £2 --. . gizzard. int ... Intestine. m ... mouscular slips passing from septum to septum. mo ... male genital opening. | n ... ventral nerve cord. ne ... small nephridia. 0 sn OVALY. od ... oviduct. oes ... esophagus. phar ... pharynx. pr ... prostate gland. 1S 0S ... spermathecee. rs'0 YS°0 ... openings of spermathecze SF eee setice: sept ... septum. sg ... cerebral ganglion. ae ... testis iy v ... transverse blood-vessel. vd vd' vd? vasa deferentia. vd'o vd’o ... internal openings of vasa deferentia. US ... vesicule seminales. a3 ... ¢litellum. Fig. 1.— Represents a somewhat diagrammatic view of the ventral surface of the worm in the anterior region of the body to show the segments, annuli and openings of the reproductive organs. The openings of the receptacula lie in the seventh and eighth segments, very close to the boundaries of the segments. Hach pair has a line on the surface enclosing it. The female openings are on the 14th segment, in front of, and having no relation to the sete. The male openings (mo.) are in the special ridges in the 18th segment, and correspond in position to the innermost pair of sete. In some worms another ridge is seen occupying the posterior part of the THE GIANT EARTH-WORM OF GIPPSLAND. 49 20th, and anterior part of the 21st segment. The sete are seen as far forward as the 12th segment, in front of which they cannot be detected. The outlines of the segments are clear, and much more strongly marked than the annuli. The figure shows the tapering anterior end, the swollen part succeeding this, then a somewhat narrower portion, and then the swolien clitellar region. Fig. 2.—Represents a mature worm, dissected from the dorsal surface. The body wall is cut open by a median incision, and the organs represented in situ. Anteriorly are the very strong cup-shaped septa (sept.), bound together by special muscular slips (m.), and completely concealing the alimentary canal from view. Passing backwards, they gradually become much thinner, and more membranous in appearance, until, in the 19th segment, they entirely lose their muscular nature. In especially segments 15-18 inclusive, the muscular slips are very long, and pass from one segment to another. The septum of segment 15 forms a special covering for. the dilatation of the alimentary canal (dil.) The blood-vascular system is coloured red. Anteriorly, only the strong muscular lateral vessels (t7. v.), or “hearts,” can be seen. Behind the 15th segment, the dorsal blood-vessel is seen, with a pair of transverse vessels passing off from each side in every segment to the walls of the alimentary canal (tr. v.) In segments 12, 13, and 14 the racemose vesiculz seminales (vs.) project from between the strong septa, and in segment 18 the two prostate glands push aside the septa, which are here membranous and transparent enough to allow of the coiled glands being, in part, seen through them. The walls of the body cavity are lined by an enormous number of small coiled tubules—the nephridia which are especially abundant in the clitellar region. Behind the part which is cut open, four segments are shown, with the large (very evident dorsal) pores (dp.) lying in the lines of division. Fig. 3.—A diagrammatic longitudinal vertical section through the anterior part of the worm. The alimentary canal runs through the centre, divided into a thickly walled pharynx (phar.), as far back as the end of the 4th segment ; a short cesophagus and thick walled gizzard in the 5th segment; a tubular portion of the intestine in segments 6-11 inclusive ; a dilatation of the intestine in each of the segments 12-18 ; the tubular intestine behind these. The reproductive organs are represented. The two receptacula seminis in the 8th and 9th segments, with their ducts running forwards so as to open just with the next segment in front of that which contains the organ. Hach receptaculum (vr. s. 0.) has a slight diverticulum near the fixed end. The testes, (t¢., the second one is marked by mistake vd.) attached by short stalks to the anterior septa of segments 10 and 11. The two internal openings of the vasa deferentia (vd. 0., and vd.2 0.) The two vasa deferentia, running quite distinct from each other (vd., and vd.) to open in the 18th segment into the prostate gland (pr-), which opens to the surface in the same segment (mo.) The ovary in segment 13 (o.) The internal opening of the oviduct in segment 13 (od.) The external opening in segment 14 (fo.) The vesicule seminales (vs.) in segments 11-14 inclusive. The H 50 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, septa are shown cup-shaped anteriorly. The first one bounds the 5th seoment, and no septa are present in front of this. For the sake of clearness, the anterior septa have not been represented in their real thickness. The nerve cord is shown (1.) with the cerebral ganglia (spg.) Fig. 4.—A diagrammatic longitudinal horizontal section of the anterior part of the body to show the blood-vessels and the reproductive organs. The latter are shown as in Fig. 3. In addition, in segment 12, is the extra testis (¢.”), sometimes present. The dorsal blood-vessel gives off in the posterior part two lateral vessels on each side, supplying the walls of the itestine; and in front of, and including the 13th segment, only one, which runs round the intestine to the ventral vessel. Each of these forms a “heart.” PLATE 2. al ... alimentary canal. al bv ... lateral blood-vessel to walls of alimentary canal. al’ bu ... blood-vessel on dorsal wall of alimentary canal. b ... basement membrane layer beneath cells of epidermis. br c ... brown cells. bv ... blood-vessel. circ m ... layer of circular muscle fibres. ct ... connective tissue. cut Pepecuuicle: d bv ... dorsal blood-vessel. dv ... diverticulum of the tube enclosing the dorsal blood-vessel. dv o ... opening of diverticulum into tube. ep ... ecelomic epithelium. ept ... epidermis. ep* ... epithelium of alimentary canal. gl ... gland cells of clitellum. gob ... modified gland cells of epidermis (Fig. 7), and of epithelium of alimentary canal (Fig. 5.) gob n ... nucleus of modified gland cell. glo ... opening of gland cell of clitellum. h ... lateral blood-vessel modified into heart. lat bv ... blood-vessel running by the side of the alimentary canal. long m ... layer of longitudinal muscle fibres. m neph ... muscle cells at external opening of nephridial duct. neph ... nephridial duct. n ... nuclei scattered in the basement membrane (Fig. 7.) THE GIANT EARTH-WORM OF GIPPSLAND. 51 n ... ventral nerve cord. nn n? ... branches of ventral nerve cord. pig ... pigment masses. sept sop SSO. sept bv ... blood-vessel supplying septum. v bu ... ventral blood-vessel. y ... cells in diverticulum of tube around dorsal blood-vessel. x ... tube surrounding dorsal blood-vessel. Fig. 5.—Transverse section through a small portion of the dorsal wall of the alimentary canal in the hinder region of the body. Half of the dorsal blood-vessel (d. bv.) is represented, and half of the tubular structure (+.) surrounding’ this, together with two diverticula (dv.) arising from it. Of these, the upper one is almost completely filled with large nucleated cells pressed closely together (y.), with strands of connective tissue (cf.) crossing from side to side. The lower one is almost empty, though some cells still remain, the others having doubtless passed into the tube (v.) Both diverticula open into the tube, and the latter is lined internally and externally by cells of the peritoneal epithelium. The walls of the alimentary canal are coated internally by columnar epithelium cells, with spherical nuclei (¢f.1) Amongst them goblet cells (gob.) are present. The layer of circular muscle fibres is well developed, and above and below this is a network of blood-vessels, with connecting vessels passing through the muscle layer. More externally lie the longitudinal fibres, and outside these again, the cells of the peritoneal epithelium, modified into brown cells. These are absent in the mid-dorsal line within the tube (w.), and from them long thin processes can be traced through the muscle layers to the epithelium lining the alimentary canal. (These processes are finer, and not so clearly marked as in the figure.) Hach brown cell has a very distinct spherical nucleus placed where it begins to grow narrow. Fig. 6.—Section through the modified skin of the clitellar region. The epidermic cells pass inwards, and their pointed internal ends lie in a thick homogeneous basement membrane. In this lies also a great number of red-brown pigment masses, which also pass downwards, surrounding the blood-vessels. Beneath the epidermic cells lies a great development of unicellular glands, each of which has a swollen internal end, and a narrow tubular part passing upwards towards the surface (gl.) Some of the glands are much more swollen than others, and each contains a definite nucleus. Amongst them are coiled blood-vessels, and the openings of the nephridia (neph.), the inter-cellular part of the ducts of which are very vesicular in nature, with extremely thin walls. Amongst the gland cells are pillars of connective tissue (ct.), which divide into forks resting upon the circular muscle fibres. These pillars surround the nephridial ducts, and are continuous with the basement membrane beneath the epidermic cells. H 2 52 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, Fig. 7.—Section of epidermic cells, with the cuticle externally, in the region outside the clitellum. The inner ends of the cells pass down into the basement membrane (b.), and are not clearly defined. In the membrane lie scattered nuclei, and blood-vessels pass up amongst the bases of the cells, and sometimes form loops (not shown in the figure), which reach very close to the surface. Certain of the cells are transformed into large irregularly shaped goblet cells (gob.), the interior of which shows a granular protoplasm with vacuolar spaces, and usually a nucleus pushed to one side. Beneath the basement membrane lie the circular muscle fibres. Fig. 8.—Diagrammatic representation of one complete segment of the body, and a portion of another behind the 14th segment, to show the blood vascular system. The body wall is supposed to be completely removed, and the septa in position. The alimentary canal is represented as a simple tube, passing through the septa. Above it lies the dorsal blood-vessel (d. bv.) This gives off—(1) two vessels in each segment, one on either side, to the septum (set. bv.); (2) two pairs of vessels in each segment, one pair on either side, to the walls of the alimentary canal (al. bv.) Beneath the alimentary canal is the ventral blood-vessel (v. bv.), from which arises one branch on either side in the posterior part, supplying the anterior face of each septum. From this vessel a branch passes off on each side to the body wall, supplying the nephridia, &. The ventral blood-vessel, together with the nerve cord, passes through an opening in the septum ventrally, and from the nerve cord arise three branches on each side (w.’ n.? 1.) Fig. 9.—A similar representation of segments 12, 18, and 14, and part of 11. Segments 12 and 13 are alike, and show the development of the vascular system peculiar to segments 13-6 inclusive. The dorsal vessel is dilated, and gives off, mm the posterior part of the segment, a large vessel (/.), which runs round and joins the ventral vessel. This is muscular and dilatable, and forms the “heart.” From it arises a branch, which runs towards the dorsal wall of the alimentary canal, and divides into two. parts, of which one joins its fellow of the other side, and the two together form a single median vessel on the roof of the alimentary canal (a/.* bv.); the other runs downwards, and again divides into two, one of which joins a lateral blood-vessel at the side of the canal (/at. bv.), the other passes to the septum, and also gives branches to the body wall. No other branches beyond the “hearts” are in connection with the ventral blood-vessel in these segments. In the 14th segment the vessels are similar to those in the segments behind (cf. Fig. 8), with the exception that the lateral blood-vessel is continued into it, and lies by the side of the alimentary canal, without having any connection with the other vessels in the segment. By an unfortunate mistake, the two septal blood-vessels arising from the ventral blood-vessel posteriorly, just as in Fig. 8, have been omitted. These are present in every segment behind those in which the hearts are developed. THE GIANT EARTH-WORM OF GIPPSLAND. 53 PLATE 3. al alimentary canal. bv blood-vessel. com nerve commissure. ct connective tissue. dp dorsal pore. d bv dorsal blood-vessel. ep epithelium lining pharynx. g ganglion cells. ef giant fibre. Ln} nerve branches to prostomial region. Ln? nerve branches to lower lip region. MUSC CLYC circular muscles. mus muscle fibres. n nuclei. ner ventral nerve cord. ne nerve fibres. 0 mature ovun. Oe ORO? OF OB orp various stages in the development of ova within the ovary. phn nerve supply to walls of pharynx. p stiff pseudopodial processes of ccelomic corpuscles, sal salivary gland. sal d salivary duct. sal o opening of salivary gland. Sog cerebral ganglia. sp sperm-blastophore. sperm developed spermatozoa. 123 early stages in the development of spermatozoa. sept septum. Fig. 10.—Transverse section through a portion of the pharynx wall, together with the attached nephridia, modified into salivary glands (sal.) The wall consists of epithelial cells surrounded by a mass of connective tissue and muscle fibres, blood- vessels, and nerve fibres. The salivary glands are encased in connective tissue, and resemble in structure nephridia, their ducts being intra-cellular. The ducts join together (sal. d.), pierce the walls of the pharynx, and open into the latter ; the cells of the epithelial lining (sal. 0.), becoming arranged in such a manner as to forma spherical body, through whose diameter the duct runs. Amongst the salivary glands, are bands of longitudinal muscles (Jong. musc.), beneath the pharynx lie the blood-vessels, into which the single ventral vessel (cf. Figs. 8 and 9) divides, and the ventral nerve cord. The latter is cut through shortly behind the spot at which it 54 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, divides into the two commissures. Only one giant fibre (g./.) is present so far forwards. Fig. 11.—Section through a single one of the numerous racemose projections from the surface of the ‘ovary. The upper end of the figure represents the part attached to the body of the ovary, and consists of a syneytium—a mass of protoplasm not defined into cells, and amongst which numerous nuclei are scattered. The nuclei are of various sizes—o’, 0*, 0°, o°—with well-marked chromatin fragments, usually, in the above, arranged peripherally. The more highly-developed ova lie on the outside (o’), and contain large and very distinct nuclei, with nuclear membranes, and one especially large mass of chromatin. The end of the projection is occupied by a single mature ovum, with large nucleus inside, which is a network and a large mass of chromatin, smaller chromatin fragments being scattered through the nucleus. ZuIss, apo. obj. 2°0 mm. Ocular compens. 18. Magnifying power 2250. Fig. 12.—Section through one of the projecting rays of the testis. The whole structure 1s a syncytium, amongst which are blood-vessels and numerous strands of connective tissue and muscles. The nuclei are of different sizes, the smallest, which are very minute, being placed in the stem. Sometimes, a line of demarcation can be seen around one or two of the larger nuclei, as in Fig. 12a. The little mass of protoplasm measured ‘05 mm., and included two nuclei. Zxiss, apo. obj. 2°0 mm. Ocular compens. 18. Fig. 13.—A single seta, showing the slightly swollen internal end, to which muscles are attached. The swelling one-third from the free tip and the pointed free end. The setz are very small, and the one figured measured 5 mm. in length. Fig. 14.—Represents a diagrammatic longitudinal vertical section through the body, to show the relationship of the septa to the external markings of the segments. The left is the anterior end, and the septa are inserted behind the line indicating externally the boundary of the segment. A result of this is that dorsally the septa are pouched in the middle line, into which opens the dorsal pore, which always lies in the line of division between the segments (d.p.) Fig. 15.—A capsule of the vesiculze seminalis, in which are represented the various stages of development of the spermatozoa seen in the organ at the same time, though not necessarily in the one capsule. The figure, though showing structures, each of which exists, as represented in the vesicule seminalis, is hence diagrammatic, 3 shows a mass of cells not long ago split off from the testis. 1 shows the separate cells. 2 show the increase of nuclei without corresponding division of protoplasm and separation into separate cells. Sf. show various stages in the development of the spermatospheres, the developing spermatozoa being attached to the central sperm-blastophore. Sperm. shows a mass of fully-developed spermatozoa, with their deeply staining rod-shaped heads massed together, doubtless around a sperm- blastophore at one end, and their tails, forming a flame-shaped structure, at the other THE GIANT EARTH-WORM OF GIPPSLAND. 55 end. The developing spermatozoa are contained in capsules of connective tissue of varying size, which are richly supplied with blood-vessels (0.v.) Fig. 16.—Represents the anterior end of the alimentary canal, removed with the nervous system. Dorsally are the cerebral ganglia (s.o.g.) | On each side pass off the commissures, from which arise (1) near the dorsal end a nerve forwards to the prostomial region, (2) near the ventral surface a nerve forwards to the lower portions of the mouth, (3) at the side of the pharynx a large flat sheet of nervous tissue, which spreads out upon and gives branches off into the walls of the pharynx. Fig. 17.—Group of cells from the blood (a) and from the ccelomic fluid (0), the latter with stiff pseudopodial processes. PLATE 4. ac bv ... lateral blood-vessels to wall of alimentary canal. b ... basement membrane. bv ... blood-vessels. ct ... connective tissue. ct neph ... Special connective tissue fibres around the nephridial ducts. d bv ... dorsal blood-vessels. dv ... diverticulum of tube surrounding dorsal blood-vessels. ep ... ccelomic epithelium. ept ... special epidermic cells modified to form the openings for the nephridial duct. gang ... ganglion cells. oF ... giant fibre. gob ... goblet cell. mt ... Intestine. m ... special muscle fibres at the sides of the nerve cord. musc circ ... circular muscle fibres. musc long ... longitudinal muscle fibres. nN ... nuclei of connective tissue cells surrounding nephridial duct. neph ... nephridium. neph int ... intercellular portion of nephridial duct. sept ... septum. a ..._ tube surrounding the dorsal blood-vessel. Fig. 18.—Transverse section through the body wall at the point of attachment of one of the smaller nephridia. Externally lie the cuticle and epidermic cells, amongst which are very numerous goblet cells. The epidermic cells have their bases embedded in the basement membrane (b.), within which lies the layer of circular muscle fibres, 56 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, which are smaller, and thrown into folds externally. Amongst the fibres are numerous blood-vessels. The layer of longitudinal fibres consists of groups of fibres, each of which is angular in section. The groups are surrounded by connective tissue, a layer of which lies between them and the ccelomic epithelium. The nephridium consists of a number of intracellular ducts cut in section, and surrounded by a mass of connective tissue cells, whose outlines cannot be distinguished. Amongst them ramify very many blood-vessels (b.v.) A special blood-vessel accompanies the duet through the body wall. The duct is thrown into folds in the latter, and hence presents the appearance of a series of sections. The intracellular opens into the intercellular, just within the circular layer of muscle fibres, and the intercellular part runs straight to the exterior opening through the special spherical structure formed by the epidermic cells. Fig. 19.—A_ transverse section through the ventral nerve cord in the posterior region of the body. The section is taken through the part lying between the origins of the lateral nerve branches. The whole is surrounded by connective tissue, which on either side passes off to the supporting laminz running to the body wall, Special muscle fibres (m.) running along each side of the cord are cut through. The nervous matter is distinguishable into two parts, (1) an inner double cord of nerve fibres, and (2) a darker looking outer layer of fibres and ganglion cells, confined to the lateral and ventral aspects. The ganglion cells are large, and as a rule have but one process, which runs into the central part. Dorsally are four giant fibres, each with a special sheath of wavy-looking connective tissue fibres, The ccelomic epithelial cells are more columnar on the ventral than the dorsal surface. Fig. 20.—A portion of the alimentary canal within the last twenty segments of the body, together with the body wall ventrally. The strong septa at right angles to the canal are seen with their supports radially arranged. The dorsal blood-vessel (d. bv.) is surrounded by a white tubular structure (w), from which in each segment diverticula, irregularly arranged, are given off. The smaller nephridia are seen to lie in the posterior part of each segment. PLATE 5. bv ... blood-vessel. ct ... connective tissue. dp ... dorsal pore. ep ... epithelium of alimentary canal (Fig. 24.) of receptaculum seminis (Hig. 25). 1 ... straight portion of larger nephridium. musc cive ... circular layer of muscle fibres. musc long ... longitudinal layer of muscle fibres. THE GIANT EARTH-WORM OF GIPPSLAND. a Ww n’ ... lateral nerve branches. neph ... smaller nephridia. neph 1 ... larger nephridia. neph 1o ... internal openings of larger nephridium. sept ... septum. : sept o --. ventral opening in septum to allow of passage through of nerve cord. v bu ... ventral blood-vessel. Fig. 21.—Somewhat diagrammatic representation of the ventral part of the body wall in the hinder region of the body, cut away to show the two kinds of nephridia in situ. ‘The body wall is flattened out, and all but the lower portion of the septa cut away. In the mid-ventral line is the nerve cord, running through the apertures in the septa, and giving off three branches on each side in each segment. Of these, two (n'n*) lie near the posterior end, and one (7°) nearly in the middle of the segment. The smaller nephridia form a row between the two anterior nerves, and are numerous. In each segment is a single pair of large nephridia. These lie in front of the nerve branches, and at the same level as the ventral-most amongst the smaller ones. Each has a branch passing forward to the septum through which it passes, and opens by a ciliated funnel. The smaller ones are devoid of internal openings. Fig. 22.—A portion of the dorsal body wall in the hinder part of the clitellar region cut away, so as to show the dense mass of small nephridia with which it is coated. These are separated into two lots by a median dorsal groove, out of the roof of which pass the dorsal pores (d. p.) Each septum is pouched in the median dorsal line, so that the dorsal pore lies slightly in front of the internal attachment of the septum. Fig. 23.—This, by mistake, is numbered 33. One of the receptacula seminis removed from the body. The surface is marked by longitudinal grooves, and up each side runs a strong muscular slip (musc.) Close to the attached end is a diverticulum (dvt.), the surface of which is thrown into folds. Fig. 24.—A transverse section through the wall of one of the intestinal dilatations, showing the folds into which the epithelium is thrown. The columnar cells of either side of a fold being only separated from each other by a network of blood-vessels. Externally lies a thin layer of circular fibres, and outside these the longitudinal fibres, with a small amount of connective tissue. Fig. 25.—Section through the wall of the diverticulum, taken along a line parallel to the length. The internal surface is coated by a deep columnar epithelium, beneath which is a network of branching blood-vessels embedded in connective tissue. External to this lie, first, the circular fibres, and next, the more strongly developed longitudinal fibres. I 58 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS, MUSC CLYC musc long PLATE 6. a bridge of connective tissue, across which passes the vas deferens. b intra-cellular part of the nephridial funnel. bv blood-vessel. c ep ecelomic epithelium. Gi connective tissue. d ventral longitudinal duct connecting the various nephridia. ep epidermis ep’ ep” special modifications of the epidermie cells at the external opening of the nephridia. ept columnar ciliated cells lining the oviduct. int internal surface of the prostate. m muscle fibres surrounding the nephridial funnel (Fig. 26). Fibres forming a sphincter at the external nephridial opening (Fig. 28). . m oO external opening of male organs. circular and longitudinal layer of muscle fibres. neph small nephridia. neph 1 large nephridia. neph 1 o internal opening of large nephridium. neph o external opening of large nephridium. n nuclei of cells forming nephridial duct. We special nuclei where the duct enters the epidermis. ov oviduct. pig pigment masses. br prostate gland. sept septum. sp’ sp™ cells forming the inner and outer parts of the walls of the prostate gland. vd .. vas deferens. vdo,vdo’... internal openings of the vasa deferentia. v0 opening of vasa deferentia into prostate. 123 the three rows of cells which line the entrance to the nephridial funnel. Fig. 26.—Section through the nephridial funnel. The upper lip overhangs the lower. The duct, after passing through the septum (seft.), enlarges, being still 1 intra-cellular (b.) At the open end are three rows of cells (1, 2, 3,) increasing in size towards the outermost, which line the margin of the funnel. These bear long cilia, THE GIANT EARTH-WORM OF GIPPSLAND. 59 which are continued down into the duct, and extend backwards into the body of the nephridium. The funnel is enclosed in a sheath of connective tissue, amongst which _lie many circularly disposed muscle fibres. Fig. 27.—Diagram to show the relationship of the nephridia. Three smaller nephridia are shown, and one larger one. Two of the smaller belong to the same segment as the larger one (neph. 1.) These two smaller ones open into a longitudinal intra-cellular duct (d.) running beneath the ccelomic epithelium ; into this opens the larger nephridium, from which a finer duct runs forward through the septum to the nephridial funnel. A small duct passes beneath the septum to the next segment, opening into the longitudinal duct, which is connected with the small nephridium. The ducts are indicated in the figure by the parts more darkly shaded. From the longitudinal duct arise at intervals (and having no constant relationship to the nephridia) a series of ducts leading down through the body wall. Within the muscle layers, the ducts change from being intra to inter-cellular in nature. The inter-cellular ducts usually branch, and each branch has its own external opening, at which the cells of the epidermis become modified. Fig. 28.—Section through the external opening of a small nephridium in the clitellar region. The duct, with its very thin walls, is vesicular in nature, and the thin film-like pavement epithelium is distinguishable by its slightly granular appearance, and its large oval nuclei, irregularly arranged. Where an optical section is obtained at the edges, the nuclei are seen to cause bulgings in the wall, which is darkly coloured in the drawing. ‘The epidermic cells are arranged at the external opening so as to form a spherical structure, through the centre of which the duct runs, its walls being continuous with the cells forming the axis of the sphere. These are more darkly stained (ef.') than the other cells (é.”), and no nuclei could be detected in them. The cuticle is not prolonged into the duct, and just where the latter enters the epidermis are a few muscle cells, acting, doubtless, as a sphincter (m.) Zeiss, obj. apo. 2mm. Ocular compens. 18. Magnifying power, 2,250. Fig. 29.—Section through part of the wall of the prostate gland. The cells lying internally (sf.*) are more darkly stained than the outer ones, into which they sometimes project, and contain no nuclei. The outer cells (sf.") have swollen ends, and ducts running towards the inner surface of the gland, and contain nuclei. Pigment masses of a yellow-brown colour occur on the outer edge of the organ, beneath the enclosing connective tissue. Zerrss F., ocular 4. Fig. 30.—Diagrammatic longitudinal vertical section through the segments containing the vasa deferentia. The openings into the body cavity of the two are shown, the funnels being composed of cubical ciliated cells, and being hence very unlike the nephridial funnels. The vasa deferentia run back quite independently of each other into the prostate gland, certain of the segments being omitted in the figure. The ducts lie in the connective tissue just above the longitudinal muscle 12 60 ON THE ANATOMY OF MEGASCOLIDES AUSTRALIS. fibres, and never unite to form a single one, They are ciliated up to the point at which they open into the duct of the prostate gland (v. 0.) The latter is lined by columnar epithelial cells, continuous, at the opening of the structure, with the. epidermic cells. Fig. 31.—Section through the oviduct, showing the deep columnar ciliated cells of which it is formed. Those of the vasa deferentia being cubical. ‘iy aa j ae cory Page ne oy on ¥ I , i ; ¥ “ i , Pp f We 5 « é 5 Trans. Royal Society Vietoria 1888. Pinte | a Fig. 3. i J te ut f2 : i | | | } P ees i s a= | (Gncummmn a , a 2 — \) pea nN f Pe NUUAUELUR DEG RERGRTUTUTEREeRETeOTTEE sea { SLT, iti WH e NN iT o> | i ee Vj iy me ER ey] 1 ] 4 af/ Dis vii QQ Jal ty th / / Be Be) a om, NN | T Vio To Wao T! Vs. tr) Trans. Royal Sociely Victoria 1888. Plate 2. Trans. Royal. Society Victoria 1868. Plate 3. ey \\ }) 1 AN \ Ni) DP > \ society Victoria 1666. Plate. 4. ee, Ore } j ee ee del. JJ.Wald Lith. MEGASCOLIDES _ AUSTRALIS. Troedel « 0° Print. y Victoria 1668. Plate 5. a, 0cle Trans. Royal St Se AUSTRALIS MEGASCOLIDES ca Troedel x 0? Print. 1 JJWild Lith. YT 60°? Tapsoay, Oe Mee eed y Re Ey J sity 2a NY le La ‘Q Fd QQQy mL0704 Ayonas hoy swosyz SG OG; ESSN = Ee \ AG) eee 4 00T7> ! a” mee ANANTH AMAL TT] a) ON oe) ears 4, Gor \S Q WAN, NY ory € 2p 212 wedg'g TRANSACTIONS ROYAL, SOCIETY Or VACTOREA: VOL. (erie AGES Mir CONTI ‘ENTS. RECORDS OF OBSERVATIONS. . WILLIAM MACGREGOR’ is HIGHLAND PLANTS FROM NEW GUINEA, sy Baron von Musrtier, K.C.M.G., M. anv Px.D., F.B.S._ - = : = 1 THE PREPARATION OF ALKYL-SULPHINE, SHLENINE, AND PHOSPHONIUM SALTS, sy Orme Masson, M.A., D.-Sc., Proresson oF CHEMISTRY IN THE University or MELBOURNE, AND J. B. Kirguanp, F.C.S. - - - - - - - Set) MEY ANATOMY, OF “AN” AUS'IERAELTAN