It corresponds to the more or less pyriform vesicle into which the glandular part of the nephridium opens in some terrestrial and most of the aquatic Oligochaeta and the Leeches, and is probably therefore derived from an invagination of the hypodermis, as is the vesicle of the aquatic forms, the rest of the nephridium being mesoblastic. The pore by which the duct opens is generally placed slightly in front of the setae in the ventral row, but it may be placed in a similar position relative to the dorsal row. There seems to be much variability in the position of the aperture in the same individual, and the apertures of the right and left organs in the same somite may not correspond (cf. Claparede, Z. W. Z. xix. 1869, p. 620; Hering, Z. W. Z. viii. 1856-57, note, p. 401). The variation is perhaps due to the loss of nephridia originally present in each somite, and opening in different positions relatively to the setae. Perrier observed in Plutellus an alternation in the position of the nephridial apertures between the upper and lower rows of setae.

Beddard has noted a similar but irregular alternation in a species of Acanthodrilus in which the dorsally placed nephridia have the duct represented by a large thin-walled muscular sac, with a small diverticulum; whereas those ventrally placed open laterally into a long muscular sac, much as do the nephridia of aquatic Oligochaeta. There is therefore an anatomical difference between the two sets of nephridia opening dorsally and ventrally. The last-named anatomist has also described the nephridia of a New Zealand Acanthodrilus (A. multiporus), in which there are four nephridia on each side in each somite of the body, corresponding to the setae. The two ventral nephridia on each side are intertwined and anastomose (?). In the posterior region of the body the organs open externally, each by a single aperture placed dorsally to each of the eight separate setae; in the anterior region, however, there are innumerable minute orifices in a single circular row, and in the same line with the setae, due to the branching of the ducts which appear to form a circular canal round the somites in this region. It is quite possible therefore that in other Earthworms certain of the nephridia in each somite are aborted, but that it is sometimes one, sometimes another pair which is missing.

Benham states (Q. J. M. xxvi. p. 256) that numerous small nephridia occur in each of the somites of a small Perichaeta from the Philippine Islands. Details are wanting at present.

1 The Oligochaete Plutellus is an exception to this rule. Its nephridia (and oviducts) open both externally and internally in the same somite, as in many Polychaeta.

The aquatic Oligochaeta have a single pair of nephridia in each somite opening near the ventral set of setae, and composed of the same parts as in Lumbricus. The duct, which is sometimes glandular, is usually vesicular, and the gland-tube opens into it laterally. The funnel is absent in Chaetogastridae. Vejdovsky finds that both funnel and gland-tube are produced by the growth of peritoneal cells covering the septa: the duct by an invagination of the hypodermis.

Nephridia of the ordinary shape appear to be absent in Megascolex and Typhaeus, and their place is taken by tufted bunches of tubes, the structure of which has not been investigated. But they appear to be closely similar to the first.pair of nephridia of Urochaeta, and the nephridia in general of Microchaeta. Each organ in the last named consists of a muscular vesicle opening externally to which are attached (1)a series of U-shaped loops, each loop containing intra-cellular tubes, and (2) a single simple tube which perforates a septum in the usual manner and opens into the coelome by a small contracted funnel. The mode of connection of the tubes to the vesicle has not been ascertained as yet.

The Earthworm is hermaphrodite like most of its congeners. The organs are figured in P1. xii. The male apparatus consists of two pairs of.testes, three pairs of vesiculae seminales, and a paired vas deferens. The testes are situated near the nerve cord on the posterior faces of the anterior septa dividing the tenth and eleventh somites. Each testis is described by Blomfield as 'a white translucent body of irregularly quadrangular form, rarely more than one-tenth of an inch in diameter, much flattened, and attached by one side to the coelomic epithelium of which it appears to be a local modification.' In the Chaetogastridae the spermatozoa begin to develope in the coelome and complete their growth in the sperma-thecae, but in most Oligochaeta, as in the Earthworm, they develope in vesiculae seminales which are caecal outgrowths of the septa. This is their probable origin in Lumbricus. The immature vesiculae in this worm are, according to Blomfield, 'six small light-coloured vascular growths on the three septa 9-10, 10-11, 11-12, arranged in three pairs.' The first and second pairs are anterior outgrowths of the septa to which they belong: the third is a posterior outgrowth of the septum 11-12 (Benham, note, p. 259, Q. J. M. xxvi. 1886). The cavities of the vesiculae are traversed by connective tissue trabeculae and capillary bloodvessels.

The sperm-polyplasts and fully formed spermatozoa are lodged in the interstices of the trabeculae, in which may also be found the Gregarine Monocystis Lumbrici in all stages of its life-history. When the Earthworm is sexually mature the first four vesiculae 'form a central body covering in the rosettes and testes of the tenth segment.' A similar coalescence occurs in the eleventh somite between the two last vesiculae. Each vas deferens consists of two ciliated funnels or rosettes, one situated in the tenth, the other in the eleventh somite, which join a common duct. The latter opens near the ventral set of setae in the fifteenth somite. It therefore traverses three somites.

The two ovaries occupy a similar position to the testes but are lodged in the thirteenth somite. Each is invested by a layer of flat peritoneal cells. When mature it has the form of a pear with a long stalk - the latter being formed of a single string of ripe ova. It consists, when immature, like a testis, of a finely granular protoplasm with scattered nuclei, in other words, of a syncytium. The ova of the aquatic Oligochaeta, e. g. Tubificidae, break away from the ovary and ripen in the coelome or in an 'egg-sac' similar to the vesiculae seminales. Each oviduct commences with a wide ciliated aperture vis a vis to the ovary. The duct perforates the septum between somites thirteen and fourteen, and opens in the fourteenth somite near the ventral row of setae. A receptaculum ovorum, or 'egg-sac,' is attached to the oviducal funnel and receives the ripe ova. It contains 1-5 ova, and like a vesicula seminalis, it is a diverticulum of a septum, - that which separates somite thirteen from somite fourteen.