On the Occurrence of ‘Hepato-pancreatic’ Glands in the Indian Earthworms of the Genus Eutyphoeus Mich

In 1895, in his monograph on the Oligochaeta (3), Beddard says that in all the three species of Typhoeus (T. orientalis, T. gammiei, and T. masoni), of whose anatomy we possess at all sufficient data, the intestine is furnished with a series of about six pairs of reniform glands lying on its dorsal surface. He adds that the glands are made up of a much-folded membrane, their interior being thus divided up into numerous compartments by the folds which seem to anastomose, and that the structure of the glands is extremely like that of calciferous glands. Further, in his monograph Beddard (3, p. 64) mentions these glands as occurring only in Megascolex and Typhoeus and says that they are unknown otherwise in the Oligochaeta. He drops out both Acanthodrilus and Eudrilus from the category of worms possessing ‘intestinal glands’, presumably because the descriptions of the glands in these two cases are very meagre and have not since been confirmed.

Since 1895 several other species of Eutyphoeus (the new generic name for Typhoeus) have been described ; Stephenson (10, 1930), in his recent admirable memoir on the Oligochaeta, records as many as twenty-six species of this genus, all from India, especially the Gangetic Plain. In systematic work it is not necessary to open the middle of the worm, since all characters of systematic importance can be made out by a dissection of the anterior portion of the worm alone; but in spite of this fact, these ‘intestinal glands’ on the dorsal surface of the intestine have been described in several species, and Stephenson (9, 1923) considers ‘that these glands may not improbably be a general character of the genus, though they have not, as a rule, been noted by recent observers, who have not usually opened the intestinal regions of these worms’.

Like Beddard, Stephenson (10, 1930) also records these glands only in Megascolex coeruleus and Euty- p h o e u s. The only description available so far of these glands is that given by Beddard (2) which has been quoted above, his diagram being reproduced in fig. 6, Pl. 10. It will be seen, therefore, that these glands have not been properly investigated, and that their significance has not been realized. Beddard worked only on preserved specimens, and was really concerned mainly with characters of systematic importance rather than with the minute structure and function of the glands.

In 1923 G. S. Thapar (Proc. Tenth Ind. Sci. Congress) communicated a paper, the published summary of which reads as follows: ‘In certain species of Eutyphoeus are found a few paired glands in connection with the alimentary canal. They occur behind the middle of the body, covering the intestine from above in a few consecutive segments. The author, after describing the histology of these glands and the condition of the intestine in this region, proceeds to compare these glands with other glands in connection with the alimentary canal of earthworms. A brief discussion on the probable origin of these_glands in Eutyphoeus is also given.’

While engaged in working out the morphology and physiology of these glands, I was struck by their extraordinary resemblance to the so-called ‘liver’ or ‘hepato-pancreas’ of other Invertebrates. Not only do they resemble the ‘liver’ in their position, structure and development, but their blood-supply, the digestive secretion they discharge into the gut, and the presence of glycogen granules in the cells corroborate their hepatopancreatic character.

The chloragogen cells of Oligochaeta have been credited with a nutritive function by Schneider (7, 1896), and Liebmann (5, 1926) has recently revived the view that the chloragosomes are storehouses of nutriment; but the existence of a definite localized ‘liver or hepato-pancreas ‘is so far unknown amongst the Oligochaeta, or even amongst the Annelida, although one would expect that this important organ or its representative must be present in animals at the Annelidan grade of organization.

Owing to other pressing engagements and also to the fact that these worms are available only during the monsoon months (July to September), the work has taken a long time, but a number of observations and results have already been communicated to the Indian Science Congress (1930) and to the United Provinces Academy of Sciences (December 21, 1931). In carrying out the physiological part of the work I have received considerable help from my friend and colleague Dr. S. M. Sane of the Chemistry Department at Lucknow, and my best thanks are due to him. Dr. J. G. Mukerjee of the King George’s Medical College has also been very helpful, and I have pleasure in recording my thanks to him. I am also thankful to Mr. M. L. Bhatia for his help in the illustrations.

As the paper is being published jointly, it will be appropriate to state that the senior author is responsible for the observations and conclusions, as also for the preparation of the manuscript and the diagrams. The junior author rendered valuable help in the experimental part of the work by preparing the material and generally assisting in the work. The first person singular in the text refers to the senior author.

All previous observers mention that these glands occur in about five or six successive segments in the middle of the body, but they have not recorded their exact segmental position. I have carefully counted the segments of Eutyphoeus waltoni Mich, and find that the total number of segments varies from 190 to 210 and that the glands always occupy five segments—seventy-ninth to eighty-third, both inclusive— (fig. 1, Pl. 9). If we take into account, therefore, only the segmental position of the glands, they would seem to he quite in front of the middle of the body, but this is actually not the case. The glands are really situated behind the middle of the body, the discrepancy between their segmental position and their actual site being due to the fact that the anterior segments of the worm are much larger in size as compared with the posterior segments. In fact, the anterior half of the worm is made up only of 65 to 70 large segments, while the posterior half contains as many as 135 to 140 small segments. Thus it is that the glands, although they occur in segments 79 to 83, actually lie behind the middle of the body.

One other fact with regard to the position of these glands, which has not been observed by previous workers, may be recorded here: that the glands always occupy the last five typhlosolar segments of the intestine. On opening the gut of Eutyphoeus we find that the typhlosole is very feebly developed, as for example in Pheretima, and forms a very low and narrow ridge which projects for a very short distance into the lumen of the gut (fig. 2, Pl. 9). It is merely a fold of the intestinal epithelium and corresponds to Stephenson’s (10, 1930) first type of typhlosole characteristic of the families Megascolecidae and Glossoscolecidae, in which only the ‘blood of the gut-sinus fills out the space within the epithelial fold, containing also a quantity of loose connective tissue’. This type of typhlosole is to be distinguished from the second type, i.e. that of the Lumbricidae and Microchaetinae in which all the coats of the intestine are folded in to form the typhlosole, forming a prominent ridge which projects a good distance into the lumen of the intestine. Again, the typhlosole in Eutyphoeus, as in other earthworms, does not extend along the whole length of the intestine ; it stops short of the eighty-fourth segment—so that we can distinguish an interior typhlosolar region of the intestine extending up to and including the eighty-third segment and a post-typhlosolar region¹ occupying the last 107 to 127 segments. The intestinal glands always occupy the last five typhlosolar segments of the intestine. As seen in fig. 2, Pl. 9, the typhlosole at its posterior end (segments 79 to 83) no longer remains a low and narrow ridge, but becomes very prominent and forms a thick and flat swollen mid-dorsal ingrowth, this appearance being due to the presence of ‘intestinal glands’ overlying and depressing the typhlosole in this region.

The ‘hepato-pancreatic glands’, therefore, lie within and above the epithelial fold of the typhlosole (figs. 2, 3, and 4, Pl. 9) at the posterior end of the typhlosolar region of the intestine, there being no trace of a typhlosole behind the region of the glands. In other words, the typhlosole in its last five segments carries the ‘hepato-pancreatic glands’ on its dorsal surface.

In a live worm the glands are easily recognized through the translucent skin as a very vascular red patch behind the middle of the body. When a large number of different kinds of live worms are brought to the laboratory, I have invariably sorted out Eutyphoeus from other worms by recognizing its vascular intestinal glands. On opening a freshly narcotized specimen in normal saline, the glands are seen as prominent, paired, rounded swellings on the dorsal surface of the gut lying on either side of and beneath the dorsal vessel. Each pair of glands is again bilobed through a shallow indentation about the middle of its length. The glands exhibit rhythmic contractions, the wave of contraction passing from their outer edges to the mid-dorsal line. Since the glands are richly permeated with blood-capillaries, the blood also follows the rhythmic wave of contraction, and one can easily observe under a binocular dissecting-microscope the flow of blood in the capillary network from the outer edges of the glands towards the dorsal vessel.

In the seventy-ninth segment the glands are usually small, but they go on increasing in size in the succeeding segments, the best-developed glands being the last pair in the eighty- third segment. In serial sections of a worm the glands of the seventy-ninth segment measured 616μ. ×366μ, while those of the eighty-third segment measured 1350μ × 833 μ. In fact, the last pair of glands are not only the largest in size, but they also bulge backwards and overlap the intestine, so that in transverse sections of this region the posterior bulgings of the glands appear as paired rounded structures lying separately above the intestine. We may note here that although five distinct pairs of glands are visible (fig. 1, Pl. 9) on a superficial examination, actually all of them form one continuous structure; since, as Beddard observed (2, 1889), not only are the pair of glands in each segment fused together in the mid-dorsal line, but even the. glands of successive segments are continuous with one another. In fact, we should speak of all the five pairs of glands as one large hepato-pancreatic gland extending over five consecutive segments.

The glands (figs. 3 and 4, Pl. 9) are formed of a large number of branching lamellae and lobules of varying sizes and lengths massed together. Just as the epithelial folds of the typhlosole enclose between them part of the blood of the gut-sinus, similarly each fold or lamella of the intestinal gland encloses a blood-sinus in its double fold of glandular epithelium. The spaces between the adjoining lamellae are well-defined and form the channels into which the secretions of the gland-cells are discharged. Generally, the ventral part of the gland nearest the gut shows simple or branched lamellae hanging freely in the cavity of the gland, but the dorsal and lateral parts of the gland are much more solid and consist of interlacing and anastomosing lamellae forming distinct lobules which in sections present a honeycombed appearance, each cell of the honeycomb being a lobule surrounded by a blood-sinus and enclosing a gland-ductule within its layer of glandular epithelium (fig. 5, Pl. 10). In sections passing through the greater part of the glands of the eighty-third segment, one can see the whole gland filled up with tubular lobules, each lobule being formed of a single layer of glandular epithelium surrounding a gland-ductule, and separated from the adjoining lobules by sinusoid blood-capillaries. The free lamellae hanging in the cavity of the gland are altogether absent in the posterior region of the glands. The three essential elements of a gland-structure, i.e. the secretory epithelial cells, a blood-supply, and a basement membrane holding the secretory cells in position, are all present. Each of the five glands is covered on the outside by the muscular layers of the intestine and the outer visceral peritoneal layer, while ventrally it is lined by the ciliated intestinal epithelium forming the roof of the gut. The muscular layers are very thin and lie only on the surface, but do not extend into the substance of the gland. Further, the cells of the visceral peritoneum are also very small and flat, unlike the tall swollen cells of the other parts of the gut.

The glandular cells are more or less rhomboidal—even polygonal—in outline, and are arranged in single layers along a well- marked basement membrane which encloses a blood-sinus. Some of the cells have a straight free margin, but in most cells the free margin presents a rounded swollen appearance. The nucleus lies either in the centre of the cell or near its free margin ; it is more or less rounded and shows within it distinct darkly staining spots. On an average, each cell is about 11 pμ in height and about 14μin width, while the nucleus is about 6μ in diameter. The cells bear neither cilia nor rodlets on their free surfaces, the cilia or rodlets being confined to the cells of the intestinal epithelium only. We may note in passing that while the whole of the intestinal epithelium of the gut in front of the glands and in the region of the glands themselves bears cilia or rodlets, the epithelium of the gut behind the glands, i.e. the folded epithelium of the rectum, does not bear any cilia or rodlets (fig. 8, Pl. 10).

The glandular cells are no doubt secretory, their free surface being the secretory surface. One cannot fail to notice a collection of secretory granules within the outer bulging surface of the cells as well as in the cavities of the glands. The shape and structure of the cells remind one strongly of hepatic cells. A trained observer, like my friend Capt. J. G. Mukerjee of the Medical College to whom I showed my sections, remarked that the cells were exactly like the human liver-cells. The secretions from the gland-cells are discharged into the ductules and cavities of the glands, whence they are poured into the intestine usually through two paired openings (fig. 3, Pl. 9) in each of the five glandular segments.

These openings of the glands into the intestine are distinct and well-defined ; in a series of sections through the five glands the number of openings into the intestine was found to be as follows :

(1) Seventy-ninth segment—one median unpaired opening.

(2) Eightieth segment—one median and two paired openings.

(3) Eighty-first segment—two paired openings.

(4) Eighty-second segment—two paired openings.

(5) Eighty-third segment—two paired openings.

Thus there are altogether eighteen openings of the glands into the intestinal lumen, some median (fig. 5, Pl. 10) but generally paired (fig. 3, Pl. 9). Each opening is elongated and slit-like, 30μ to 44μ in length and 11μ to 17μ in width as measured from serial sections. The ciliated intestinal epithelium extends upwards into these paired openings, so that the actual openings are lined with the intestinal epithelium (fig. 5, Pl. 10) and not with the glandular epithelium, the latter being confined to the actual substance of the glands.

As we have already mentioned, the distinguishing character of the intestinal glands in a living specimen is their blood-red appearance which makes them easily visible through the translucent body-wall of the worm. The minute sinusoid capillaries are thickly interwoven to form a close network so that the glands are richly permeated throughout with blood (fig. 2, Pl. 9 ; figs. 5 and 7, Pl. 10). That a copious supply of blood courses through the glands is easily observed by opening a freshly narcotized worm and examining the exposed glands under a binocular dissecting microscope. The glands are seen contracting rhythmically and their blood-capillaries alternately filled with and emptied of blood. The blood is seen to enter the glands along their outer edges, run through the capillaries of the glandular tissue, and to come out again to pass into the longitudinal dorsal vessel through two pairs of intestino- dorsal vessels in each of the five glandular segments. As the glands contract and relax, a wave of blood is seen passing from the outer edges to the middle line of the glands during the passage of the blood to the median dorsal vessel. As many as twenty-one to twenty-five such waves per minute were counted in freshly narcotized worms.

All along the length of the post-typhlosolar intestine (from the eighty-third segment to the posterior end of the worm), i.e. in the region of the gut behind the intestinal glands, a prominent blood-vessel—the ventral intestinal sinus— runs along the mid-ventral line of the intestine (fig. 2, Pl. 9). In transverse sections (fig. 8, Pl. 10) this sinus is a very prominent structure, and is seen bulging out from the floor of the intestine into the intestinal lumen. In size it is as big or even bigger than the dorsal vessel and is always seen full of blood in preserved specimens.

The direction of blood-flow in this sinus is from behind forwards, as has been verified by me by cutting the sinus at various places and observing from which of the cut ends the blood flows ; it always flows from the posterior cut end. In the eighty-third segment this ventral-intestinal sinus forks into two (fig. 2, Pl. 9), and each of the two branches at once runs dorsalwards and continues along the outer edges of the glands, emptying all its contained blood into them. These lateral glandular sinuses, therefore, take the place of the ventral-intestinal sinus in the region of the intestinal glands and form prominent blood-channels lying at the outer edges of the glands (figs. 2 and 4, Pls. 9 and 10). In the region of the glands these lateral glandular sinuses are in communication not only with an enveloping sinus, which goes all round the periphery of the glands between the thin outer muscular layers and the glands proper, but also with the general capillary network of the glands. The enveloping sinus is also continuous with the extensive network of sinuses permeating the lamellae and lobules of the glands. If we bear in mind the large size and extent of the ventral intestinal vessel (fig. 8, Pl. 10) and the large amount of blood contained in it, we cannot help concluding that the greater part of the blood from the intestine in the last 107 to 127 segments passes into the ventral-intestinal sinus. It flows forwards in this sinus, and is carried through the two lateral glandular sinuses into the sinusoid capillaries of the glands. The blood courses through these capillaries of the glands and passes into the dorsal vessel through the two pairs of very short intestino-dorsal vessels in each segment.

The size and disposition of the intestino-dorsal vessels in the different regions of the body of Eutyphoeus is interesting. In the region of the body in front of the glands two pairs of ring-like intestino-dorsals are very prominent in each segment : they run all round the gut and are seen bulging on its outer surface (figs. 1 and 2, Pl. 9). As they are all along in communication with the general intestinal sinus they, no doubt, carry the blood from the intestine into the dorsal vessel. Behind the region of the glands, however, they are very short blood-vessels and do not go round the gut (fig. 2), as they do in the anterior region. They emerge from the dorsal surface of the intestine and immediately enter into the dorsal vessel ; in short, the intestino-dorsals are very much reduced in size and extent in the post-typhlosolar region of the intestine. They no doubt receive a small quantity of blood from the gut and convey it to the dorsal vessel. But the greater part of the blood of the intestine of this region really goes into the ventral-intestinal sinus, which is very large and prominent—even larger than the dorsal vessel—and discharges its contained blood into the intestinal glands. In the region of the glands itself the intestino-dorsals are again very short, as the dorsal vessel comes to be very closely connected with the glands—it is almost imbedded between the glands of the two sides in each of the five segments, as shown in fig. 5, pl. 10. In fact, there can be no doubt that the blood from the glands directly enters the dorsal vessel. Not only is it actually seen doing so in a freshly narcotized worm, but that this is the direction of flow is confirmed by the disposition of the valves (fig. 5).

We can thus describe the blood-supply of the glands by saying that the two lateral glandular sinuses break up into a network of sinusoid capillaries within the substance of the gland; these capillaries join together again and form two pairs of intestino- dorsal vessels in each segment to convey the blood from the glands into the dorsal vessel.

We thus get a semblance of a portal system in connexion with these hepato-pancreatic glands: the blood from the intestine is collected into the ventral-intestinal sinus; from there it goes to the glands, in which it breaks into a close network of capillaries; these capillaries join together in each gland to form two pairs of intestino-dorsal vessels which discharge their contained blood into the contractile dorsal vessel in each of the five glandular segments.

Interesting light is thrown on the nature of the glands by a study of their development. They arise. as median dorsal outgrowths of the endodermal epithelium of the intestine in the seventy-ninth to eighty-third segments. In an embryo, about 15 mm. in length, the glands are seen to extend over five segments, but the development is more rapid in the posterior segments than in the anterior ones. The endodermal epithelium of the gut in these early embryos consists of very tall cells full of very small albumen granules which do not take up the stains, the nuclei alone being stained red with borax-carmine. In the anterior segments the mid-dorsal wall of the gut is seen to give out a hollow outgrowth; the tall albumenladen cells of the intestinal epithelium grow upwards, become gradually smaller and smaller in size until they pass into the gland-cells which are absolutely free from albumen-granules and are stained deeply with borax-carmine (fig. 9, Pl. 10). The glands themselves at this stage consist of rounded or elliptic follicles formed of a single-layered epithelium, each follicle enclosing a small or large cavity which will later on form, no doubt, one of the many ductules of the adult gland. At the regions of these glandular outgrowths one can easily make out a long narrow canal through which the lumen of the gut communicates with the lumen of the gland. The cilia or rodlets which form a characteristic feature of the gut-epithelium of this region in the adult worm are completely absent from the gut-epithelium of the embryos.

In the middle and posterior regions of the glands the mid dorsal region of the gut changes its character. Its cells are no longer large and full of albumen granules, but have undergone a proliferation into smaller cells, more or less free of albumen granules. These differentiated cells take up a deep borax-carmine stain and are seen to bud off gland-cells from their dorsal surface (fig. 10, PL 10). The intestinal cells and the gland-cells are very closely associated in this—the posterior—region of the glands. We may note here that these differentiated cells are confined only to the mid-dorsal region of the gut ; the cells of the remaining parts of the gut are still large and full of albumen granules. Further, on the dorsal surface of the glands the peritoneal cells, which are also filled with albumen granules, much larger in size than those present in the endodermal epithelial cells, also bud off cells which migrate into the substance of the glands and make their way in between the follicles of the gland-cells. These cells, originating from the mesodermal peritoneal cells, go to form the sinusoid capillaries of the glands.

We may conclude, therefore, that the glands arise as dorsal hollow outgrowths from the roof of the intestine and are reinforced in the posterior region by the budding off of cells from the mid-dorsal region of the gut. Later, cells are budded off from the peritoneal epithelium of the gut, make their way into the endodermal glandular tissue, and lead to the development of sinusoid blood-capillaries. The substance of the gland is therefore entirely endodermal, while the blood-capillaries are mesodermal in origin.

The most important question about these glands is their function. Bearing in mind Beddard’s suggestive observation that ‘the minute structure of these alimentary glands bears a very close resemblance to that of calciferous glands ‘, the first step was to compare these glands with the ‘calciferous glands’. Fortunately, Eutyphoeus itself possesses definitive calciferous glands, a single pair in segment 12, and these form a constant character of all the species of this genus, just as the ‘intestinal or hepato-pancreatic glands’ form another constant character. We get, therefore, both the intestinal glands and the calciferous glands in one and the same earthworm. The calciferous glands have been described in detail by Stephenson and Prashad (8, 1919), and a comparison between the two kinds of glands as regards their nature is interesting.

The very first consideration in the comparison of the glands is their respective position. One salient feature which marks off at once the ‘intestinal glands ‘from the ‘calciferous glands ‘of all earthworms is that while the latter are always situated in the oesophageal region, the former are situated just about the middle region of the intestine, where the greater part of the food of the earthworm would be digested and absorbed. Besides this essential difference in their respective positions, there are differences in structure between the two types of glands. There is no doubt that both kinds of glands are formed essentially of folds of the gut-epithelium, oesophageal epithelium in the one case and the intestinal epithelium in the other; but this does not indicate identity of nature or function, since we must bearin mind the fact that in other animals the gut-epithelium is known to be capable of giving rise to structures entirely different in nature and function. The origin of lungs, liver, and pancreas from the gut-epithelium of different regions of the body of a vertebrate is a case in point. In comparing, however, the structure of the calciferous and intestinal glands in an adult Eutyphoeus we must note : firstly, that the intestinal glands are much more solid structures than the calciferous glands. While the latter consist of a large number of simple vertical lamellae only sometimes united by synapticula, the former consist of lamellae which interlace and anastomose so freely with one another as to produce an almost solid honeycombed structure. Both in transverse and longitudinal sections (figs. 3 and 11) the intestinal glands present greater complexity and a higher degree of differentiation than is seen in the calciferous glands. Again, there are other differences. Stephenson and Prashad (8, 1919) record the presence of cilia or rodiets on the free borders of the cells of the calciferous glands ; these cilia or rodlets are entirely absent from the cells of the intestinal glands, being confined to the definitive gutepithelium only. Further, the sinusoid blood-capillaries of the intestinal glands are enclosed by a distinct basement-membrane, while in the case of the calciferous glands a basement- membrane, according to Stephenson and Prashad, is often not to be made out in well-fixed preparations, though well seen in badly fixed specimens.

That the intestinal glands are not functionally ‘calciferous’ is easily proved by the fact that they never secrete calcium carbonate. While the calciferous glands in segment 12 always show copious white calcareous particles of milky lime under a binocular microscope, the intestinal glands never show any traces of lime, nor could they be shown to contain any trace of calcium even on chemical examination.

—The intestinal glands from twenty specimens of Eutyphoeus were ignited on a platinum foil, and the foil, together with the ashes, was placed in dilute hydrochloric acid in a test-tube. The acid was boiled and the solution of the ash was transferred through a filter-paper into another test-tube. Ammonia was added until the solution was alkaline. On the addition of ammonium oxalate no precipitate of calcium oxalate was formed, showing the absence of calcium.

The same test was applied to the definitive calciferous glands of the twelfth segment, when whitish calcium carbonate ash could be detected immediately on ignition. This was confirmed by boiling the ash in dilute hydrochloric acid and testing with ammonia and ammonium oxalate, when a white precipitate of calcium oxalate was formed, showing the presence of calcium.

This test definitely negatived the possibility of the intestinal glands being functionally calciferous in nature.

Whatever may be the real function of the calciferous glands in segment 12, it seems highly improbable that glands identical in nature to them would occur again in the same earthworm sixty-seven segments behind.

Having satisfied myself by observation and experiment that the intestinal glands are not calciferous in function, the next step was to determine the nature of their secretion. The fact that the glands open into the gut generally by two pairs of minute apertures in each of the five segments—eighteen openings in all, to be exact—strongly indicates that the secretion of the glands is discharged into the gut through these apertures. The possibility of the glands directly absorbing nutriment from the gut is almost completely excluded, since the openings are very minute and the glands are fairly well segregated from the gut—it is practically impossible for the contents of the gut to come in contact with the surface of the cells of the gland.

Attempts were made, therefore, to determine the nature of their secretions. A preliminary test was made with the extract of the gland (strength 2·5 per cent.) to determine its action on calcified milk.

—Five cubic centimetres of the glandextract was added to 5 c.cm. of calcified milk and kept at a temperature of 40° C. Clotting took place in fourteen minutes, showing the presence of a proteolytic enzyme.

As the glands are small, it is necessary to dissect a large number of earthworms to obtain gland-extracts of appreciable strength. An idea of the extremely small weight of the gland can be gained from the fact that the glands of fifty earthworms dissected weighed only 0·884 gm., so that the gland in each earthworm weighs, on an average, only 0·0176 gm.

The next experiment was made to test the character of the enzyme in the glands.

—Two hundred worms were dissected and their glands taken out. The glands weighed 3·4 gm. They were ground up in sand and 80 c.cm. of distilled water added, giving the strength of the gland-extract at 4·25 per cent. These extracts were kept at different temperatures and were allowed to digest their own substance for twenty-four hours. The amount of amino-acids produced was estimated as shown in the table on the next page.

The control was titrated first with formol solution and NaOH (phenolphthalein was used as indicator), and all other extracts were strictly compared on titration with the control. Toluol was employed as an antiseptic. The hydrogen-ion concentration of the gland-extract was tested with bromo-thymol blue and phenol red ; the pH being 7·8.

This experiment clearly demonstrated the presence of a ‘tryptic’ enzyme. The best results were obtained between 34° and 40° 0., which is very nearly the average temperature at Lucknow during August and September.

This experiment was repeated several times with very nearly the same results. That the glands secrete a tryptic enzyme is, therefore, well established. The glands do not seem to act on starch and fats, as I have not been able to get a satisfactory test of their action on these substances.

A further piece of work affording strong evidence of the hepatic nature of these glands was carried out in the form of staining sections with Best’s carmine to test the presence of glycogen granules. The glands were fixed in acetic absolute alcohol, and imbedded in celloidin-paraffin. The sections were stained with Best’s carmine according to the technique given by Bolles Lee and Gatenby (4, 1921). Thickly set granules stained deep red were seen all along the secretory edges of the glandular-cells, clearly demonstrating the presence of glycogen granules in them (fig. 11, Pl. 10).

Students of the Oligochaeta are well aware of the fact that there is a great difference of opinion amongst various observers even with regard to the function of the calciferous glands. Michaelsen (6, 1895) holds strongly that the function of the calciferous glands is the absorption of nutriment, and he always designates them as ‘chyle-sacs’. The excretion of lime, according to him, is a secondary function. The main objection urged against his view is the situation of the calciferous glands, since they lie far forward, and the main digestive region, i.e. the intestine, lies behind them. But this objection does not hold in the case of the intestinal glands, since they lie just at the right place, a little behind the middle of the body where the typhlosole ends, and the gut still extends behind for another 107 to 127 segments. They lie, therefore, just in the region where the main work of digestion and absorption takes place.

That the glands open into the gut by as many as eighteen openings in five segments strongly indicates that their secretions are poured into the gut through these openings. The nature of the secretions seems to be tryptic, since amino-acids were formed as products of digestion in the experiments made. Further, the greater part of the blood of the last portion of the intestine (107 to 127 segments) is collected and taken to these glands, where it permeates the substance of the gland through a sinusoid capillary network. The blood from the glands is collected again by the intestino-dorsals and taken to the dorsal vessel, the entire blood-supply of the glands resembling a hepatic portal system. These facts taken together with the demonstration of glycogen granules within the glandular cells strongly suggest that the intestinal glands are of the nature of a hepato-pancreas. The structure of the gland-cells themselves and the development of the glands further corroborate these conclusions.

The observations and conclusions arrived at may now be summarized :

1. The intestinal glands of Eutyphoeus waltoni lie on the dorsal surface of the gut and extend as paired glands from segments 79 to 83. As the two glands of a pair are fused in the middle dorsal fine and the glands of successive segments are connected, we should speak of it as one large gland extending over five segments.

2. The gland forms the posterior boundary of the typhlosole of the gut, there being no typhlosole behind the region of the gland.

3. The gland is more or less solid and consists of lobules of glandular epithelium and interlacing lamellae. The lobules are separated from one another by sinusoid capillaries, while the two epithelial folds of a lamella enclose a blood-sinus between them.

4. The gland-cells are rhomboidal to polyhedral in outline, and their shape and structure strongly resemble those of liver-cells. They do not bear cilia or rodlets.

5. The whole gland opens into the gut through as many as eighteen apertures over five segments. These apertures are lined with the ciliated gut-epithelium.

6. The blood-supply of the gland resembles a hepatic portal system. The blood is collected from the gut of the last 107 to 127 segments into a ventral intestinal sinus which empties all its blood into the sinusoid capillaries of the gland. The capillaries of the gland join together to form five pairs of intestino-dorsal vessels which carry all the blood of the glands into the dorsal vessel.

7. The glands develop as dorsal outgrowths of the endodermal lining of the embryonic gut.

8. The glands secrete no calcium whatever. Calcified milk was curdled by the gland-extract in fourteen minutes. A tryptic enzyme has been demonstrated by the formation of amino-acids in digestion experiments.

9. Thickly set glycogen granules have been demonstrated by staining the gland-cells with Best’s carmine.

10. The glands are, therefore, of the nature of a hepatopancreas.

All figures, except fig. 6, are of Eutyphoeus waltoni.

PLATE 9.

Fig. 1.—A photograph of a dissection showing the hepato-pancreatic glands in situ. The segments are marked with their serial numbers. The intestino-dorsals form a pair of ring-like vessels going all round the gut in front of the glands, but are very short vessels behind the glands, just connecting the gut-sinus with the dorsal vessel, b.w., body-wall; d.v., dorsal vessel; h.p.gl., hepato-pancreatic gland; int.d.v., intestino- dorsal vessels; i.s., intersegmental septum. (× about 3.)

Fig. 2.—Same as fig. 1, but the intestine has been cut open along one side and the flap reflected to show the structures on the inside. The typhlosole extends as a low ridge up to the seventy-eighth segment, but becomes expanded into a flat oblong structure to support the glands in segments 79 to 83. The ventral intestinal sinus is exposed and is seen dividing into the two lateral glandular sinuses which supply blood to the glands. The network of sinusoid capillaries is shown in the region of the glands, d.v., dorsal vessel; int., cut edge of the intestine; int.d.v., intestino- dorsal vessels ; lal.gl.s., lateral glandular sinus ; n., diagrammatic representation of the network of sinusoid blood-capillaries in the glands ; ty, typhlosole; v.i.s., ventral-intestinal sinus.

Fig. 3.—A microphotograph of a transverse section passing through the hepato-pancreatic glands, which are seen to open into the lumen of the intestine through two contiguous openings (0). The capillaries of the glands are seen opening into the dorsal vessel. The lobules and lamellae of the gland and the sinusoid capillaries are clearly seen, b.w., body-wall; d.v., dorsal vessel; h.p.gl., hepato-pancreatic gland; nt., intestine; n.c., nervecord; O., openings of the glands into the intestine; v.v., ventral vessel. (× about 22.)

Fig. 4.—A camera-lucida drawing of a transverse section passing through the glands, b.w., body-wall ; d.v., dorsal vessel ; h.p.gl., hepato-pancreatic glands showing lamellae and lobules and the sinusoid blood-capillaries shaded black; lat.gl.s., lateral glandular sinus; lat.n.v., lateral neural vessel; n.c., nerve-cord; o., opening of the gland into the lumen of the gut; v.v., ventral vessel. (× about 22.)

PLATE 10.

Fig. 5.—A camera-lucida drawing of a transverse section of the gland showing interlacing lamellae and lobules, and the rich blood-supply of the glands. The valves guarding the entrance of the blood from the glands into the dorsal vessel are also shown, bl.sin., blood sinus; cil.int.epith., ciliated intestinal epithelium; d.v., dorsal vessel; gl.duct., one of the ductules of the gland ; gl.duct.², one of the large ducts of the gland ; inl.d.v., short intestino-dorsal vessel in the region of the glands; lal.gl.sin., lateral glandular sinus ; o., opening of the gland into the lumen of the intestine ; peri.epith., peritoneal epithelium ; v., valve guarding the entrance of the blood into the dorsal vessel. (× about 70.)

Fig. 6.—Beddard’s diagram of the structure of the glands in Eutyphoeus gammii.

Fig. 7.—A camera-lucida drawing of a small portion of the gland showing lobules of the gland interspersed with sinusoid blood-capillaries, bl.sin., sinusoid blood-capillary; gl.c., gland-cell; gl.duct., a ductule of the gland; gl.lob., a gland-lobule. (× about 200.)

Fig. 8.—A transverse section passing through the post-typhlosolar intestine behind the region of the glands. The glands are absent, but the large ventral-intestinal sinus collecting blood in the posterior segments and supplying the glands therewith is prominently visible, d.v., dorsal vessel; lat.n.v., lateral neural vessel; n.c., nerve-cord; v.i.s., ventral intestinal sinus; v.v., ventral vessel. (× about 25.)

Fig. 9.—A transverse section through the gut of an embryo showing the origin and development of the glands as mid-dorsal outgrowths from the wall of the gut. d.v., dorsal vessel; dor.div., dorsal diverticulum of the gut forming the glands; end., endodermal lining of the gut; int.gl., intestinal glands; peri.ep., peritoneal epithelium. (× about 225.)

Fig. 10.—A transverse section passing through the glands at a more advanced stage of development than that shown in fig. 9. The mid-dorsal region of the gut shows proliferation of cells, and cells are also being budded from the endoderm on both sides. The peritoneal epithelium is also budding off cells which make their way into the gland-substance, d.v., dorsal vessel ; gl.l., a gland-lobule ; int.end., intestinal endoderm full of albumen spheres; m.d.r., mid-dorsal proliferating region of the gut with cells almost free of albumen; peri.ep., peritoneal epithelium.

Fig. 11.—A microphotograph of a portion of a longitudinal section of the gland stained with Best’s carmine to show the glycogen granules. The glycogen granules stained red (black in the photograph) are thickly aggregated along the secretory edges of the cells that line the ductules of the gland, bl.c., blood-capillaries; gl.d., ductules of the gland; gl.g., glycogen granules.