Description of a New Species of Moniligaster from India

At first I imagined that it would be one of Dr. Bourne’s species, but it agrees with none of them in the few—the very few—characters given by him; he practically only gives the position of the gizzard in his seven species, and even this in such a way as is scarcely sufficient for identification; for he does not state how many constrictions are present, i.e. whether there is a separate portion of gizzard to each somite passed through. We await impatiently a more detailed description of the several interesting species, especially of M. sapphirinaoides, which has been so long promised to us.

The position of the various genital pores in Moniligaster by the repeated contributions of Mr. Beddard (1, 2, 3, 4, 5), as well as by those of Rosa, Bourne, and Michaelsen, has now become definitely ascertained; and I much regret that my figure representing the genital organs of this genus, published in my “Attempt to classify Earthworms,” on p. 295, is wrong, owing to my having followed Dr. Horses account of M. Houtenii, which appears to have the organs posterior to their position in the other species.

This octochætous, meganephric genus is distinguished, as has been pointed out by recent writers, by the following external characters:

(a) The male pores (a single pair) between Somites x and xI.

(b) A single pair of spermathecal pores between Somites VII/VIII.

(c)A pair of oviducal pores between Somites xI and XII.

(d) Nephridiopores in. line with the dorsal (or outer) couple of chætæ.

Internally the following features may be taken as characteristic:

(e) The gizzard constricted by one to four narrow, fibrous, annular bands at the septa, giving the “moniliform “appearance indicated by the generic name.

(f) Absence of calciferous or other enteric diverticula.

(g) A single pair of spermathecæ in Somite VIII with very long, narrow ducts. ¹

(h) A single pair of spermsacs, supported by the septum between Somites ix and x, but belonging to the former; the sacs are independent, and each encloses the testis and the spermiducal funnel.

(i) A single pair of atria (prostates), of varied size and shape.

(j) A large ovisac, which may extend through several somites.

M. indicus, n. sp., which forms the subject of the present paper, may be diagnosed as follows:

(a) Length 5½ inches (13·75 centimetres).

(b) Body consists of about 150 somites.

(c) Ventral intersetal space is rather wider than the lateral space (or, as a formula, I—I > II—III).

(d) The septa VI/VII, VII/VIII, VIII/IX are thick, but not so thick as in other species.

(e) Gizzard occupies Somites xIII, xIv, xv, xvI, with annular bands separating the four muscular regions.

(f) Atrium (prostate) large, spherical, lying in Somite XI.

(g) Spermatheca nearly spherical; the duct opens to the exterior through a muscular sac, subdivided into an anterior and a posterior lobe by the septum VII/VIII.

The size of the present species has just been stated, and it appears to be of some importance, as there is a great range amongst the species; for example, M. japonicus and M. Barwelli is only 28 mm., M. Deshayesi 150 mm., whilst M. Houtenii attains the length of one and a half metres. Of Bourne’s seven species, only M. ruber is definitely measured, its length being 100 mm.

As in all other species, with the exception of M. sapphirinaoides, no clitellum was present; the worm was evidently not quite sexually mature, as deduced from the condition of the genital organs. As Beddard (5) and myself (6) have previously suggested, the clitellum may be a more transient structure than in other macrodrilous Oligochseta; and it would be very interesting, and even important for systematic purposes, to know the minute structure of the clitellum in Bourne’s.species. Does it agree more closely with the microdrilous clitellum, or is it several cells thick, as in Lumbricus, &c.?

The colour of the worm during life is unknown to me. In the spirit specimens the dorsal surface is very much darker—a greeny brown—than the ventral and lateral surfaces; there is a very sharply marked line of separation between the two colourings, and not a gradual paling out as in many other worms.

The worm is rather flattened, and the posterior end, which is lighter in colour, tapers off to a point, apparently in the same sort of way as in M. robustus, A. G. B., for which species I at first took it; but in that species the gizzard differs in extent and position.

The prostomium is rounded, not embedded in the peristomium at all, but is retracted in the same sort of way as Beddard (5) found in his species.

The somites are separated by deep intersegmental grooves; the first and second, as Beddard points out in his last contribution (5) to the anatomy of M. Barwelli, are very short, the two together barely reaching the length of the third somite.

The chætæ are of the normal shape (fig. 2), and do not agree with those figured by Horst for M. Houtenii, nor with those figured by Beddard (5); they are small, as in the other species, measuring 2 mm.; the free end is rather more hooked than in M. Barwelli. All four couples of chætæ in the present species are on the ventral surface, so that none are visible from above. Their arrangement, as I have inferred above, is of some specific importance. In all species of Moniligaster the individuals of each couple are very close together, and the space between the two ventral couples—i. e. the ventral space—may be equal to, greater than, or less than the lateral space or distance between the dorsal and ventral couples of either side (see fig. 1).

Thus in M. japonicus, Mich., M. Beddardii, Rosa, the ventral space (V) = the lateral space (L); in M. Barwelli, M. uniquus, A. G. B., V < L; in M. grandis, A. G. B., as in the present species, V > L; whilst in M. Houtenii V = 2L.

In the remainder of Bourne’s species no reference is made to this point.

The nephridiopores appear to commence in Somite vi, but on dissection I find nephridia anterior to this.

I failed to discover any dorsal pores, either by means of a lens or by mounting a piece of the body-wall in Canada balsam; but in other species they are apparent, viz. in M. Barwelli and M. Beddardii.

The genital pores have the position normal for the genus, and I found it quite easy to recognise all three pairs (fig. 1); with regard to the male pores, however, I was at first deceived by an apparent slit in the hinder part of Somite x; this turned out to be merely a groove (fig. 1, x), due, no doubt, to the shrinkage of the large mass of muscle attached here around the atrium. The groove is bounded anteriorly by a fairly prominent ridge, and presented the appearance figured by Beddard (5, fig. 1), and indicated by him as the male pore itself. I find, however, that the true pore does not correspond with the position assigned to it by him forM. Barwelli; and in this point I am in agreement with Michaelsen’s description of M. japonicus (9), where the pore is at the apex of a small tubercle, in front of which is a transverse groove. It is probably a matter of preservation whether the tubercle is prominent or not. My own specimen was very soft and illpreserved.

The position of the atriopore (fig. 1, ♂) between the dorsal and ventral couples of chætæ agrees with that ascribed to it by previous writers.

The oviducal pores are small but distinct slits in the groove between Somites xI and xII, in line with the ventral couple of chætæ.

The spermathecal pores are larger, and, as in other species, are in line with the dorsal couple of chætæ; in M. De shay e si, however, the pore is stated to be in line with the “inferior chæta?’

It is to the generative organs that the lumbricologist first turns his attention in opening, or otherwise examining, a worm, and naturally these organs are more engaging, so to speak, in a genus about which such diversity of statement exists, or has till quite recently existed, as to the position of the various constituents. I have already referred to the uncertainty as to the precise somites to which the different organs belong, which has resulted from the descriptions of Perrier, Beddard, and Horst. But now that a greater number of species are known, some of them of fairly large size, we find Perrier’s original statements are correct with regard to the position of most of the genital organs, and that Horst’s numbers alone differ by being greater by a single unit. Beddard’s original description has been modified by his discovery that his apparent first somite really consists of Somites i and u. The positions attributed to the organs in the species examined by Rosa and Michaelsen, and in the present one, all agree with Beddard’s amended and Perrier’s original numbers. But nevertheless there still remain certain points on which some discrepancy or uncertainty exists, both with regard to exact position and minute structure. Some of these I hope to clear up, or at any rate to advance, though, owing to the poor condition of the worm, my contribution to the histology of Moniligaster—of which we know very little indeed—is very imperfect.

The single pair of sperm-sacs (figs. 3—5) occupies partly Somite ix and partly x. On the animal’s left side the sac appeared to be strung up by the Septum ix/x attached round its equator. In reality the septum is pouched, and the sperm-sac, as figs. 3 and 4 will show, projects into Somite ix through the wide aperture of this pouch.

On the right side, however, the pouch was deeper, so that at first sight the sac appeared to lie entirely in Somite x; but I could trace the Septum Ix/x behind the sac, and sections show me that the sperm-sac belongs, as Beddard has already suggested, to Somite Ix.

Each sperm-sac is nearly spherical, and is not so simple as in M. Barwelli, for I find that the cavity is not undivided, as Beddard figures and states, but is traversed by bundles of muscular fibres in different directions (fig. 5, trab.), between which lie the groups of developing spermatozoa. The trabecular structure is certainly not so highly developed as in Perichseta, Lumbricus, and other genera, but it is as certainly not entirely absent.

Beddard has laid great stress on the simplicity of the sperm-sac in associating Moniligaster with the Microdrili or “Limicoline section” of the Oligochæta. Here, then, we have still another apparent distinctive character between “aquatic” and “terrestrial” Oligochæta breaking down, and that within the limits of one and the same genus; and it is gradually becoming more and more impossible to characterise differentially the two groups, for in fact they verge into one another at almost every point.

The two sperm-sacs are entirely independent of one another. Each contains a testis and spermiducal funnel.

The testis (fig. 5) is digitate and attached to the hinder wall of the sac, i. e. to the Septum ix/x. The funnel—we cannot call it a “rosette”—of the sperm-duct is a simple, wide, ciliated tract of the hinder wall of the sac (fig. 5). The cells are tall, and thus readily distinguishable from the flattened cells which elsewhere form the wall around it.

This flat funnel can be seen through the thin wall on dissection, and the sperm-ducts can be seen passing away from it (fig. 4).,

As to the point of attachment of the testis, I find that it is independent of the funnel. Rosa figures it, both for Moniligaster Beddardii and for Desmogaster, attached to the funnel near to its centre. Beddard also in one figure (fig. 8) represents the testis arising from the funnel, whilst in fig. 9 he shows it at its side. In my own sections the testis lies nearer the middle line of the body than the funnel, so that the two structures are not at their best in any one given section. Perhaps Rosa’s diagram is in this point merely diagrammatic.

The sperm-duct is very delicate and very greatly convoluted (fig. 4), much more so than is indicated by either Bed-dard’s or Rosa’s figures; it more nearly approaches the condition represented by Perrier (pl. iv, figs. 81, 83), though this zoologist appears to have been disinclined to regard the “feuillets blancs” as forming part of the sperm-duct, for it would, says he, give such an enormous length to it.

I may mention that in Perrier’s species the sperm-duct, which is represented as a comparatively thin coiled duct, appears to pass through a mass of isolated white leaf-like structures set round the duct as an axis. A microscopical examination of these “feuillets “showed that they consist of a very fine coiled duct surrounded by a thin membrane (his fig. 83), and I think there can be no doubt, from the condition present in M. indicus, that this duct in the “feuillet” is part of the sperm-duct itself, and not an appendix thereto. In other species, so far as descriptions go, the duct is not of so great a length. The cells lining the duct are striated as in other worms (fig. 12), but I have never been able to detect hitherto the boundaries of the cells; here, however, they are very plain, and the cells large. In many earthworms the nuclei are flattened, but here, as in aquatic forms also, they are spherical.

In M. indicus the greater part—nearly the entire length— of this fine, convoluted sperm-duct lies in Somite ix below the sperm-sac (figs. 4, 5). It passes through the septum near the body-wall, which it enters, traverses the short distance between this septum and the prostate, embedded in the longitudinal muscle, and enters the atrial wall (prostate) at its anterior end (fig. 5); passes along the dorsal wall of the prostate, perforates the glandular coat, and opens into the atrial cavity at the apex of a slight prominence in the wall (figs. 5, 8). This condition of the duct is very similar to that described by Rosa for Desmogaster doriæ, and so far as the prseseptal portion is concerned agrees with Michaelsen’s description of M. japonicus.

The prostate is a hemispherical structure occupying the greater part of the length of Somite xi, at the anterior margin of which it opens to the exterior; it also encroaches to a slight extent on to Somite x, to which it probably belongs morphologically.

It is about half (linear) the size of the sperm-sac, larger than the ovoid prostate of M. Barwelli and M. Beddardii, and unlike these is bound down to the body-wall by special muscles; if it were placed on end with its morphologically dorsal (its actual posterior) end upwards it would resemble the condition in Beddard’s species. The hemispherical glandular part is seen in sections to be provided with a duct dorso-laterally compressed and underlying the glandular part. In the other species, so far as is described, the prostate is elongated, and is very large and extensive in M. Deshayesi.

We have no information as to its shape, which is apparently a good specific character, in any of Bourne’s species.

There is a very peculiar arrangement of muscles in connection with the atrium which has not hitherto been recorded in any earthworm.

A sheet of muscles, or rather a series of flattened bundles of muscle-fibres, passes obliquely across the body-cavity from their origin in the body-wall above the lateral line to their insertion in the wall of the prostate. On dissection, when the sides of the body are pinned down, this sheet is of course horizontal; in the natural position, however, it is an oblique, transverse sheet, such as exists almost universally in the Poly-chæta, where, however, it is generally in relation with the chætophores at its outer end.

In Somite x, immediately in front of the atrium, a similar sheet of muscles exists, reaching nearly to the level of the inner chætæ ventrally, and having the same origin as that in Somite xi. In Somites vII, vIII, ix quite a similar arrangement obtains, though, as we get forwards, the muscle bundles are smaller and smaller.

These recall the “arciform muscles” recently described by Cerfontaine ¹ in Lumbricus, but which pass, not from the lateral to the ventral surfaces, but from one side to the other, inserted at each end at the level of the ventral chætæ. Another point of difference there is, in that in Lumbricus these muscles do not traverse the body-cavity—they are embedded in the longitudinal muscles of the body-wall. The “arciform muscles” are confined to those somites between the male pore on Somitexv and the hinder end of the clitellum at Somite XXXVII, and serve to produce during copulation the groove on the ventral surface, along which the spermatozoa travel from the male pore of one worm to the spermathecse (?) of the other.

Now in Moniligaster, though we have no knowledge of the mode of copulation, yet there can be little doubt that these oblique muscles are connected with that process. By their contraction they would raise the ventral surface of the body between the male pores and the spermathecal pores, not, as I believe, for the passage of spermatozoa, but to help in holding the two worms together.

The minute structure of the prostate of M. Barwelli ¹ has been dealt with by Beddard in more than one contribution; in one of these (4, p. 121, pl. xii, fig. 11) he figures a transverse section of this structure, and it will be convenient to refer to this briefly before passing to a consideration of what I find to be the condition in the present species. In M. Barwelli, then, the cavity of the prostate (or atrium) is lined by columnar granular cells; outside this is a layer of muscles; this, again, is surrounded by “an external covering composed of large granular cells, which are separated into groups by partitions. Each cell is prolonged into a fine process, which extends at least as far as the muscular wall; indeed, it is difficult to believe that the cells do not in some way or other reach the lumen of the atrium, and there discharge their glandular secretion.”

In M. indicus the hemispherical glandular portion of the atrium, which portion may be termed the prostate, has a structure differing from that of the more compressed duct leading to the external aperture. The latter is provided with an irregular lumen, lined by a low columnar, non-glandular epithelium, the nuclei of which are oval and occupy the middle of the cell (fig. 10); outside this is a thick coating of muscles in various directions, with of course numerous blood-vessels. But after a short course this simple epithelium gives way to a glandular coat (fig. 11), recalling the structure of the “prostate.”

The “prostate” or glandular wall of the atrium (fig. 9), however, is lined by a layer of tall, thin, glandular cells, with nuclei (? at any rate deeply stained structures) placed at the sides of the cells, as if compressed; somewhat similar to, though much narrower than, the epithelium figured and described by Beddard.

The muscular coat, consisting of circular, longitudinal, and oblique fibres, does not immediately surround this epithelium as in M. Barwelli, but the two are separated by a considerable area, about twice as deep as the epithelium itself, which appears to be occupied partly by certain longer cells (gl.) than the lining epithelium, and partly by the necks of the extra-muscular multicellular glands (see figs. 8, 9).

In a foot-note on p. 573 of his memoir on Ocnerodrilus, ¹ Beddard refers to the fact, as observed in M. Barwelli, that there is “hardly any distinction between a glandular and non-glandular section of the atrium.” In the present species, as is shown, there is such a distinction very well marked.

Outside the muscular coat are the groups of glandular cells figured by Beddard. Each group (fig. 9) consists in a section of some eight to ten club-shaped cells, the narrow necks of which pass through the muscular coat, and can be traced in bundles nearly up to the epithelium. Each group is, in fact, a multicellular gland. This arrangement is quite evident in my sections, and the necks or ducts of the glands are quite noticeable, though I have been unable to trace the necks right up to the epithelium; the latter, however, is traversed by narrow clear tubes, resembling ducts, and which differ from the epithelial cells in having no inner boundary (as at a, fig. 9). The nucleus of each cell of the multicellular gland is naturally at the bottom, the widest part of the cell. But I can give no detailed description of the finer histology owing to the ill-preservation of my material; I can detect granules, as if arranged around vacuoles, but very indistinctly. Outside the glands is a very delicate membrane (cæ. ep.), sparsely provided with flattened nuclei (n. per.), which I take to be the cœlomic epithelium, and which dips down, for some distance at any rate, between the groups of cells.

In Beddard’s figure the muscular coat is made to be continuous, and no indication of “ducts “or communication between the gland-cells and the lumen is represented, though, as the above quotation shows, Beddard expected to find such “ducts.” My description of M. indicus agrees pretty closely with Rosa’s description of the prostate of M. Bed-dardii, and figure relating to the prostate of Desmogaster.

Beddard has already (4, p. 125) pointed out the resemblance between the gland (cement-drüse, prostate) attached to the atrium of Tubificidæ and the glandular lining of the atrium of Perichæta and other earthworms; at the same time he has denied the homology of the gland-cells of Moniligaster with those of Acanthodrilus, Perichæta, and other worms, chiefly, if not entirely, on the ground that in Moniligaster and in Rhynchelmis there is (according to his observations) no peritoneal membrane surrounding the glands; whilst the position of the gland-cells outside the muscular coat led him to regard these cells as representing the peritoneal membrane. Bnt I have seen what I believe to be the peritoneal membrane both in M. indicus and in Rhynchelmis; and further, the prolongation of the extra-muscular gland-cells up to the epithelium of the atrium in Moniligaster and Desmogaster seems to me to point conclusively to the derivation of those gland-cells from that epithelium, in the same way that the “cement-drüse” of Tubificidse has been shown by Vejdovsky to be developed as an outgrowth from the lining of the atrium. Therefore probably the “prostate” of Tubifex is homologous with the “prostate” (using the word in a narrow sense to mean the glandular cells) of Moniligaster.

Telmatodrilus, Eisen, ¹ one of the Tubificidae, presents an intermediate condition between the arrangement met with in Tubifex and that in Moniligaster; for in this worm, instead of the single group of gland-cells communicating with the atrium, as in Tubifex and other of its allies, there are eight or ten such isolated groups (fig. 16, b) arranged around the upper part of the atrium, that is the “glandular part of the atrium,” as I have called it in my description of Heterochæta. ²

If now the atrium were to shrink, so to speak, these isolated groups of gland-cells would be approximated, and there would result a continuous coating of gland-cells, communicating here and there through the muscular coat with the cavity of the atrium, as we find in Moniligaster and Rhynchelmis.

In other earthworms, such as Acanthodrilus, Perichætæ, &c., the structure of the prostate has been dealt with by Beddard (4); the chief difference from Moniligaster being, as I believe, due to the restriction of the muscular coat to the neck of the gland, and the consequently greater compression of gland-cells against one another (fig. 16, e).

In the “intermediate condition “(fig. 16, d) I have imagined the muscular coat becoming more definitely confined to the “duct,” though some of the glands still perforate it; whilst others are in a condition similar to that in Deino-drilus, &c.

The spermatheca presents certain interesting features. The nearly spherical or slightly pyriform sac, lying above the oesophagus in Somite VIII, is provided with a very long delicate, shining duct, ³ several times coiled, which passes downwards to reach the body-wall in the anterior of the somite in a line with the lateral chætæ (fig. 3). On dissection there appears to be an accessory gland in front and one behind the point of entrance of the duct into the body-wall, similar to that in many Perichæta and others; but in reality, as sections show, these two ovoid structures are due to the constriction, by Septum vII/vIII, of a single sac, with thick muscular walls, which opens externally in the intersegmental groove (figs. 5, 13, cop. sac.). This sac is lined by a columnar glandular epithelium, the nuclei of the cells being at their inner ends (fig. 14) similar to that of the spermatheca itself. The spermathecal duct perforates the dorsal wall of this sac at about its central point, where it is nipped by the septum, and owing to its cubical epithelium can readily be traced; the wall of the duct is very thick, being provided with a circular coat of muscles (fig. 15). It appears to me that possibly this sac is a copulatory sac in the true sense of the word, and that the muscular duct of the atrium is to some slight extent eversible, and is received into the copulatory sac.

Amongst the other species, in which the genital organs have been described, an arrangement at all comparable to this occurs only in M. Deshayesi. All recent authors agree that Perrier’s interpretation of the organs lying in Somite vII as “prostates” connected with an anterior pair of testes is erroneous. Horst regarded these pyramidal glandular organs as “spermathecæ,” the ducts of which unite to form a short common duct, receiving a more delicate “vas deferens ‘‘(Perrier). But Bed-dard has conjectured, rightly as it appears to me, that these glands are comparable to such accessories of the spermatheca as are usual amongst the Perichsetidse.

Perrier’s fig. 79, pl. iv, represents a section through one of the pyramidal organs of Somite vII, which shows its glandular character, but an absence of muscle in its wall. In the present species the glandular lining is relatively thin, the muscular coat being very greatly developed. But in M. indicus no ducts such as Perrier figures exist. I would suggest that the wide bilobed sac in my worm corresponds to these pyramidal organs. In M. Deshayesi a greater differentiation has occurred, the glandular portion being distinctly marked off from the muscular duct, whereas in M. indicus no such separation exists. If the bilobed structure of my worm were to be drawn out in such a way as to produce a couple of glands communicating by short ducts with a common duct, much the same arrangement would be produced.

The ovary closely resembles that represented by Rosa for M. Beddardii, and occupies the space between the displaced septum x/xi and that between Somite xI and xII, being therefore in Somite xI.

The oviduct is provided with a wide expanded funnel (fig. 8), very much as Perrier describes it; it is quite independent of the ovisac—that is, this latter structure opens into the coelom outside the edge of the funnel, and not through it, as in some earthworms. Nor is there any reflected portion of the funnel entering the sac, as I have recently described in Sparganophilus. ¹

The ovisac is not so extensive as it is in several of the species, e. g. in M. minutus, A. G. B., where it traverses Somites xII—xv, but it is readily visible in dissection as a narrow white structure lying in Somite xII, and attached to its anterior septum.

The worm was not fully mature. I find no spermatozoa in the spermatheæ; those in the sperm-sacs are none of them fully formed, except some attached to the funnel of the duct. At first I took the muscular trabeculae for spermatozoa, but by following their course from section to section I soon ascertained that this was not the case.

The ova in the ovary are quite small, and there are none in the ovisac.

I will now pass to the consideration of a few other points in the anatomy of M. indicus.

—The usual muscular coats are present, but they show a feature which has hitherto not been noticed; the lateral regions both of the circular and the longitudinal layers are very much thicker than the dorsal and ventral regions. In Plutellus Perrieri ¹ I have noticed a somewhat similar thickening of the ventral body-wall, due, however, to the greater development of only the longitudinal coat. The longitudinal coat is, in Moniligaster indicus, interrupted by four continuous grooves, visible on simple dissection. It is in these grooves—marked i. ch. and o. ch. in fig. 7—that the chætophores are situated. On the ventral side of the lower groove is a special bundle of longitudinal muscles (v. long, mus.) which appears to run throughout the greater part of the worm, though I am not certain whether it extends forwards into the genital region, for I cut only longitudinal sections of this part. The longitudinal muscles exhibit a pinnate arrangement so well known in the group, but only in its thicker portions; it gradually merges into an irregular arrangement dorsally and ventrally (fig. 7).

In some of the anterior somites peculiar bundles of muscles traverse the body-cavity in an obliquely vertical direction, attached at one extremity just within the ventral chætophoral groove, and at the other to the outer part of the dorsal surface (fig. 6). These muscles I have referred to in speaking of the generative apparatus, and they are represented as seen in a dissected specimen in fig. 3. These oblique muscles are limited to Somites VII to x, for those which lie apparently in Somite xI, morphologically, as I believe, belong, like the atrium with which they are connected, to the preceding somite. I have referred above to the suggested purpose of these muscles.

The septa present the same variability in thickness which has been remarked in so many worms. In the present species the three septa, vI/vII, vII/vIII, vIII/Ix, are thick and muscular, the remainder thin (figs. 3, 5). In other species of Moniligaster we find the same three, or an additional one in front thickened.

I believe none of the septa are absent, as Bourne states is the case for some of his species. Certainly on dissection some of the septa appear to have been lost; this is due to the peculiar shifting of the septa first pointed out by Horst and by Rosa.

The three thickened septa are nearly in their normal position, but are inserted in the body-wall slightly behind the intersegmental groove. The same is true for Septum Ix/x, which is bagged out by the sperm-sac. Septum x/xI is, however, thrown back a whole somite; it is confluent with the Septum XI/XII along a certain part of its course laterally (fig. 5), but becomes separate towards the middle of the body, and is there inserted into the body-wall just in front of the groove XI/XII. At the level of the atrium, however, it can be traced below this organ to its insertion just behind the groove x/xi; that is, in a position corresponding to that of the preceding septa (fig. 8). The Septum XI/XII, again, is inserted in a normal position above the groove. There are, then, no septa absent in the present species.

Rosa’s diagram of the arrangement in M. Beddardii shows it to be much less abnormal than in the present species.

In M. Houtenii, allowing for the fact that the various organs are situated one (testes, spermathecæ) or two (ovary, ovisacs) somites further back, the arrangement is the same; i. e. Septum x/xr is attached to body-wall about the middle of Somite xi, Septum XI/XII to the body-wall near the hinder part of Somite xII, Septum xii/xm close to hinder boundary of Somite XIII and immediately in front of Septum xm/xiv, which has almost the normal position.

Neither Beddard nor Michaelsen makes any remarks about this shifting of the septa, which may turn out to be a normal feature for the genus; and as it occurs too in Desmogaster, perhaps even for the family.

The nephridia closely agree with those of M. Houtenii; they commence in Somite I, and are continued in each somite in a regular manner, and the fact that only one pair of nephridia occur in the wide stretch between Septum Ix/x and that which I have referred to as Septum x/xi confirms me in this interpretation.

The nephridia in Somite iv are slightly larger than those immediately following, as Beddard has noted for M. Bar-welli, the convoluted tube being thicker and whiter. I have not been able to assure myself as to the external opening, whether it is in the body-wall or in the stomodseal wall, but I believe the former.

The nephridia are provided, as in other species, with a very large muscular duct, forming a diverticulum. The coiled tube is embedded in tissue, as in Lumbricus, and a fairly extensive plexus of blood-vessels can be detected around the tube, as in all other “earthworms “except Ocnerodrilus, as opposed to “waterworms.”

With regard to the vascular system I may merely state that the dorsal vessel is single, and is not accompanied by a supra-intestinal vessel. Four pairs of lateral hearts lie in Somites vI, vII, vIII, and ix; they are dilated and twisted, but are not moniliform. The lateral hearts in M. Deshayesi are in Somites vI, vIII, and Ix; in M. Barwelli in vI—xIv.

In M. Houtenii the dorsal vessel is doubled as in Acan-thodrilus, and hearts in Somites vI—xI A subneural vessel exists, as Rosa has already stated; this has not yet been observed in any Microdrilid, but as it is frequently absent in Microchæta, Photodrilus, and others amongst the Macrodrili, no great stress can be laid upon its absence or presence.

The alimentary canal is constructed on the plan described by previous authors. The gizzards, of which there are four, lie in Somites xIII—xvI; they increase in size from before backwards, the fourth being nearly twice the width of the first (fig. 3) they are very distinct, but are not so deeply marked off from one another as Beddard figures (5, fig. 5) for M. Barwelli. Behind each gizzard is an annular fibrous band, apparently of connective tissue, for it appears white, and not glistening, as do the gizzards. Perrier and Horst mention them. Similar fibrous bands occur at the hinder end of Somites xI and xII in front of the septum, and the wall of the gut in Somite xn is distinctly more muscular than that of xi, but very much less so than that of xIII, and not sufficiently muscular to be called a “gizzard.” There is thus a transition between oesophagus and gizzard; and it may possibly be that the gut in these somites has been or is in near allies or varieties of this species more muscular than is the case with the single specimen which I have had the opportunity of examining; in fact, these regions may be potential gizzards. In other species the gizzards are either more or less extensive.

The position and extent of the gizzards appear to be one of the chief specific characters of the genus, and M. indicus differs from M. robustus, A. Gr. B., in this point, for in Bourne’s species the gizzard occupies Somites xI—xv, presumably five in number; so that, as this is the only anatomical feature mentioned by Dr. Bourne, and notwithstanding the external resemblance with regard to the tail, I am justified in creating a new species.

There are no æsophageal glands nor other diverticula of the gut. In M. japonicus Michaelsen finds that the wall of the “oesophagus in Segments xv—xvm (?)”—i. e. behind the gizzard—is much folded and very vascular. In the cavities of the folds he found a granular mass, which “must be regarded as concretions of carbonate of lime;” but he gives no evidence that this is the case. In none of the other species is any mention made of the presence of lime, though of course there is no reason why it should not occur.

The “tubular intestine” immediately behind the gizzard is narrow and cylindrical, remaining thus through Somites xvn to about xx. It then commences to enlarge till it attains its full diameter in about Somite xxx, where it occupies nearly the whole width of the cœlom. At each septum the “sacculated intestine” is very deeply constricted, so that quite a narrow neck connects the thin-walled dilated regions.

There is no typhlosole; the walls of the gut are very thin, and provided with a dense network of longitudinal and circularly disposed blood-vessels.

Illustrating Mr. W. Blaxland Benham’s paper, “Description of a New Species of Moniligaster from India.”

FIGS. 1—15.—Moniligaster indicus.

FIG. 1.—The body-wall of six segments, cut through along the dorsal line and spread out. Viewed ventrally, × 5. d. Dorsal line. v. Ventral midline. o. ch. Inner series of chsetse. o. ch. Outer series, neo. nephridiopore. spih. Spermathecal pore. ♂. Male pore. ♀. Oviducal pore. x. Groove in front of male pore.

FIG. 2.—One of the chætæ.

FIG. 3.—A dissection of the left half of the animal, to illustrate the arrangement of the genital organs, and the position of the gizzards and the oblique muscles. Enlarged about 10 times. The somites are numbered, as also are some of the septa. Note the shifting of Septum x/xi. cop. sac. Bilobed copulatory sac, through which the spermathecal duct opens to exterior (see Fig. 5). ann. band. Annular bands between the gizzards, giz. 1, 2, 3, 4. The four gizzards, obi. m. Olique muscles (see Fig. 6) in certain of the somites, o. ch. Groove, due to interruption in longitudinal muscles throughout the body, along the line of the outer cliætæ (a similar groove exists in connection with the inner chætæe), as. (Esophagus, ovd. Funnel of oviduct, sp. d. Sperm-duct. sp. sac. Sperm-sac, lying in a pouch formed by Septum Xx/x. splh. Spermatheca.

FIG. 4.—View of male organs on a smaller scale. The pouch of Septum ix/x is removed, and the sperm-sac (sp. sac) turned forwards, showing the greatly convoluted sperm-duct (sp. d.) lying in Somite ix. The spermiducal funnel is seen by transparency through the thin wall of the sperm-sac. obi. m. Oblique muscles attached to prostate, the duct of which is now seen. o. ch. The groove in line of outer chætæe.

FIG. 5.—A semi-diagrammatic longitudinal section through the genital somites. It is diagrammatic in so far as it shows both spermathecal pore and atriopore, the entire ovary, and other features which really exist in neighbouring sections. The external rings are labelled with Roman numerals, the cavities of the somites with Arabic numerals; the septa are lettered. al. o. Male pore, or atriopore, circ. Circular muscles of the body-wall. cop. sac. Copulatory sac, showing entrance of spermathecal duct. Ig. m. Longitudinal muscles of the body-wall. neph. Portions of nephridial tube. neph. bl. Caecal bladder of nephridium, obi. mus. Oblique muscles cut across. sp. d. o. Aperture of sperm-duct into atrial cavity; its course is shown from the septum, backwards in the longitudinal muscles of body-wall, and through the prostate, sp. funnel. Funnel of sperm-duct, which does not actually occur in the same section as the testis, spoa. Spermatozoa in sperm-sac; some are seen attached to funnel, splh. Spermatheca, splh. d. Spermathecal duct. splh. o. Aperture of spermatheca.

FIG. 6.—Diagrammatic transverse section through half the body in anterior region, to show position of oblique muscles, i. ch. Groove, along which are inserted the inner cheetse. o. ch. Similar groove for outer chsetæ. n. c. Nerve-cord. sub. n. v. Subneural vessel.

FIG. 7.—Half a transverse section through the intestinal region, in order to show the peculiar band of longitudinal muscle (". long, mus.) which traverses this region of the body. The lateral thickening of the body-wall is also seen in these two sections.

FIG. 8.—A longitudinal section of the atrium and neighbouring parts. The Septum x/xi is seen passing forwards underneath the atrium to be inserted near anterior boundary of the somite (cf. Fig. 5) where this septum coalesces with Septum XI/XII. circ. m. Circular muscles of the body-wall. etc. ep. Cœlomic epithelium outside the prostate, gl. The glandular region of the duct of the atrium, long, m. Longitudinal muscles of the body-wall, a portion of which is continuous with the muscular wall of the atrial duct. mult.cell, gl. Pyriform groups of gland-cells projecting outside the muscular coat (mus. coat) of the atrium, and whose necks traverse this coat. neph. Portion of nephridium in Somite xi. obi. m. Oblique muscles, some amongst the longitudinal and others amongst the muscles of the atrial wall. on. Portion of ovary, ovd. Edge of the funnel of the oviduct, p. Prominence of the lining of the atrium, on the apex of which (in another section) the spermduct opens, sp. d. Portion of sperm-duct in the wall of the atrium.

FIG. 9.—A portion of the prostate (wall of the atrium) greatly magnified. One group of gland-cells is represented in detail; the necks of the cells are readily traced through the muscular coat (mus.), but not so readily traced up to the epithelium (ep.). The latter is, however, traversed by narrow ducts (as at a), which are, no doubt, the necks of the gland-cells. Between the epithelium and the muscular coat are large pyriform glands, the details of which could not be made out owing to the preservation of the material, b. v. Blood-vessels, n. per. Nucleus of flat coelomic epithelial cells, which surround the multicellular glands and dip inwards between them.

FIG. 10.—A portion of the epithelium of the atrial duct. m. Muscles. b. v. Blood-vessel.

FIG. 11.—A portion of the glandular region of the atrial duct (see Fig. 8, gl.). ep. Epithelium, gl. Gland-cells, m. Muscles.

FIG. 12.—Transverse section of sperm-duct. ep. Large striated epithelial cells, mus. Muscles, spoa. Spermatozoa.

FIG. 13.—Longitudinal section of the copulatory sac, showing its constriction by Septum vii/vm, and the entrance into it of the spermathecal duct (sptb. duct), ep. Epithelium, mus. Muscular coat.

FIG. 14.—A portion of the epithelium of the copulatory sac.

FIG. 15.—Section of spermathecal duct, to show the very thick coat of circularly disposed muscles.

FIG. 16.—A series of diagrammatic sections illustrating my view as to the development of the prostate in earthworms from the condition in Tubifex. The lining epithelium of the atrium is represented as consisting of cubical cells, the muscular coat by thick lines, the coelomic epithelium by interrupted lines.