ACADEMY OF NATURAL SCIENCES. 315 



contained in a uterus which, in the undeveloped state, resembles very much the 

 ovaries of the common oviparous fishes, except that its walls are thicker, and 

 that the number of ova it contains is very much smaller. In the interior of 

 the uterus, projecting from its sides, are a number of processes analogous to 

 those to which the ova are usually attached. These processes vary in number 

 in different examples, but they are so arranged that each foetal fish is in contact 

 on every side with a surface of one of these processes. • They consist apparently 

 of a membrane composed of a cellular tissue, and scattered over their surface 

 are a number of small mammillary elevations with an orifice in the center, and 

 which are probably the organs by which the peculiar secretion of the uterus, to 

 be hereafter noticed, is poured out. In an example I examined, in which im- 

 pregnation had apparently just taken place, numerous ova were found adhering 

 to these processes, although not at all in such numbers as in the ordinary fishes. 

 I counted thirty-eight in about the space of an iuch ; of these, however, but 

 few can be developed, as the number of foetuses seldom exceeds forty, and 

 sometimes is only eight. In the whole of the uterus there probably were from 

 one hundred to one hundred and fifty ova. Of the earlier stages of develop- 

 ment, however, it is not my object to treat in the present memoir, as I did not 

 commence my investigations sufficiently early to be able fully to make it out. 

 As soon, however, as the embryo has advanced sufficiently for the fins to be 

 formed, these appendages are found to be terminated by a number of digitations, 

 which project from the free edges of the fin, and are usually found situated, one 

 between each ray or spine. They are composed almost entirely of fine capillary 

 blood-vessels, united apparently by a very delicate and structureless membrane. 

 They are so delicate that unless great care is taken in removing the specimeu 

 from the uterus, they are destroyed ; nor have I ever been able to discover them 

 in specimens that have been preserved in alcohol. These processes seem con- 

 tinuous with the membrane extended between the rays of the fins, but are much 

 more delicate ; they project from the free edge of the fin, sometimes as much 

 as the eighth of an inch, and are, in the fully developed embryo, the fifteenth 

 of an inch broad. On the free margin of each digitation, a larger capil- 

 lary can be observed, which appears to be continuous all around ; it is about 

 the .003 in. in diameter, the intermediate space being filled with a net-work of 

 smaller capillaries. This system of digitations projects from the entire edge 

 of the dorsal, ventral and caudal fins, but not from the pectorals. They in fact 

 form a fringe round the entire body, with the exception of the head and that 

 part of the abdomen in front of the anus. 



Such is the -structure of the organ that evidently has some connection with 

 the nourishment of the foetus, resembling as it does so closely the early forma- 

 tion of the vascular villi and the placental tufts that proceed from the chorion 

 of the mammiferous embryo, and through which it derives its nourishment 

 before the placenta is fully formed. 



The question now presents itself as to how nourishment is conveyed from the 

 parent to the foetus through these tufts? As before stated, the lining mem- 

 brane of the uterus sends off processes which surround each foetus, without 

 however forming sheet sacks ; but although these processes are very freely sup- 



