On the Developinent of tlie Oviduct in Teleosteans. 343 



beneath tlie urinary vesicle, the coalescence of these pro- 

 longations together, and the formation of an opening to the 

 exterior. 



The size of the furrow and tube in Belone acus, according 

 to the figures of MacLeod, bears a very small proportion to 

 the size of the whole genital ridge. The state of things which 

 I have observed in the young sprat is somewhat different. 

 The genital ridge in all Tdeostei, as MacLeod observes, is very 

 late in developing ; in an alevin of Salmo Levenensis three 

 days after hatching, the structure can be by careful scrutiny 

 made out as a few rather conspicuous cells in the peritoneal 

 epithelium. In Clupea sprattus I was unable to find indica- 

 tions of the formation of the ovarian tube in young fish under 

 the length of 5 cm. In specimens of that size, when cut into 

 a series of transverse sections, the ovary is seen to be already 

 tubular in the anterior portion, while posteriorly the tube is 

 incomplete and open. The tubular ovary has a large cavity ; 

 the whole of the mediad and of the ventral wall is very thick, 

 consisting of a mass of young ova ; the other parts of the wall 

 are thinner, a rather thick epithelium extending from the ger- 

 minal region round the rest of the inner surface of the tube, 

 while the external surface of the ovary is formed by a 

 layer of fibrous connective tissue covered by endothelium. 

 Posterior to the ovarian tube we find the germinal mass, 

 together with the mediad wall of the ovary, projecting freely 

 like a genital ridge, while opposite to it externally is a slight 

 ridge of the wall of the body cavity continuous with the outer 

 wall of the complete ovarian tube. The way in which the 

 ovarian tube is formed is thus perfectly clear; the only 

 question is whether the ridge which, by coalescing with the 

 edge of the ovary, forms the tube is to be regarded as an in- 

 dependent projection from the wall of the body cavity, or as a 

 portion of the genital ridge itself. The latter view is perhaps 

 the most probable, and in that case the process in the sprat 

 does not differ from that in Belo'ne in kind, but only in degree; 

 the size of the ovarian canal being very large in the sprat in 

 comparison with the thickness of the walls. 



What takes place then in the sprat is this. Along the 

 dorsal wall of the body cavity external to the mesentery on 



