In the Anura (Spengel) all the segmental tubes are compound, and an enormous number of peritoneal funnels are present on the ventral surface, but it has not yet been definitely determined into what part of the segmental tubes they open.
Before dealing with the further changes of the Wolffian body it is necessary to return to the segmental duct, which, at the time when the pronephros is undergoing atrophy, becomes split into a dorsal Wolffian and ventral Müllerian duct. The process in Salamandra (Fürbringer) has much the same character as in Elasmobranchii, the Müllerian duct being formed by the gradual separation, from before backwards, of a solid row of cells from the ventral side of the segmental duct, the remainder of the duct constituting the Wolffian duct. During the formation of the Müllerian duct its anterior part becomes hollow, and attaching itself in front to the peritoneal epithelium acquires an opening into the body cavity. The process of hollowing is continued backwards pari passu with the splitting of the segmental duct. In the female the process is continued till the Müllerian duct opens, close to the Wolffian duct, into the cloaca. In the male the duct usually ends blindly. It is important to notice that the abdominal opening of the Müllerian duct in the Amphibia (Salamandra) is a formation independent of the pronephros, and placed slightly behind it; and that the undivided anterior part of the segmental duct (with the pronephros) is not, as in Elasmobranchii, united with the Müllerian duct, but remains connected with the Wolffian duct.
The development of the Müllerian duct has not been satisfactorily studied in other forms besides Salamandra. In Cœciliidæ its abdominal opening is on a level with the anterior end of the Wolffian body. In other forms it is usually placed very far forwards, close to the root of the lungs (except in Proteus and Batrachoseps, where it is placed somewhat further back), and some distance in front of the Wolffian body.
The Müllerian duct is always well developed in the female, and serves as oviduct. In the male it does not (except possibly in Alytes) assist in the transportation of the genital products, and is always more or less rudimentary, and in Anura may be completely absent.
After the formation of the Müllerian duct, the Wolffian duct remains as the excretory channel for the Wolffian body, and, till the atrophy of the pronephros, for this gland also. Its anterior section, in front of the Wolffian body, undergoes a more or less complete atrophy.
The further changes of the excretory system concern (1) the junction in the male of the anterior part of the Wolffian body with the testis; (2) certain changes in the collecting tubes of the posterior part of the mesonephros. The first of these processes results in the division of the Wolffian body into a sexual and a non-sexual part, and in Salamandra and other Urodeles the division corresponds with the distribution of the simple and compound segmental tubes.
Since the development of the canals connecting the testes with the sexual part of the Wolffian body has not been in all points satisfactorily elucidated, it will be convenient to commence with a description of the adult arrangement of the parts ([fig. 400] B). In most instances a non-segmental system of canals—the vasa efferentia (ve)—coming from the testis, fall into a canal known as the longitudinal canal of the Wolffian body, from which there pass off transverse canals, which fall into, and are equal in number to, the primary Malpighian bodies of the sexual part of the gland. The spermatozoa, brought to the Malpighian bodies, are thence transported along the segmental tubes to the Wolffian duct, and so to the exterior. The system of canals connecting the testis with the Malpighian bodies is known as the testicular network. The number of segmental tubes connected with the testis varies very greatly. In Siredon there are as many as from 30-32 (Spengel).
The longitudinal canal of the Wolffian body is in rare instances (Spelerpes, etc.) absent, where the sexual part of the Wolffian body is slightly developed. In the Urodela the testes are united with the anterior part of the Wolffian body. In the Cœciliidæ the junction takes place in an homologous part of the Wolffian body, but, owing to the development of the anterior segmental tubes, which are rudimentary in the Urodela, it is situated some way behind the front end. Amongst the Anura the connection of the testis with the tubules of the Wolffian body is subject to considerable variations. In Bufo cinereus the normal Urodele type is preserved, and in Bombinator the same arrangement is found in a rudimentary condition, in that there are transverse trunks from the longitudinal canal of the Wolffian body, which end blindly, while the semen is carried into the Wolffian duct by canals in front of the Wolffian body. In Alytes and Discoglossus the semen is carried away by a similar direct continuation of the longitudinal canal in front of the Wolffian body, but there are no rudimentary transverse canals passing into the Wolffian body, as in Bombinator. In Rana the transverse ducts which pass off from the longitudinal canal of the Wolffian body, after dilating to form (?) rudimentary Malpighian bodies, enter directly into the collecting tubes near their opening into the Wolffian duct.
In most Urodeles the peritoneal openings connected with the primary generative Malpighian bodies atrophy, but in Spelerpes they persist. In the Cœciliidæ they also remain in the adult state.
With reference to the development of these parts little is known except that the testicular network grows out from the primary Malpighian bodies, and becomes united with the testis. Embryological evidence, as well as the fact of the persistence of the peritoneal funnels of the generative region in the adults of some forms, proves that the testicular network is not developed from the peritoneal funnels.