334 FISHES 



externally are nourished by the yolk contained in the egg, and 

 when development takes place within the ovary this mode of 

 nutrition is not necessarily altered. It becomes, however, pos- 

 sible for the embryo to absorb nourishment from the secretions 

 of the ovarian walls or from the maternal blood-vessels, and 

 thus the supply afforded by the yolk may be supplemented or 

 the amount of yolk may be diminished and the new supply of 

 nourishment may be to a greater or less degree substituted for 

 the old. Among the Cyprinodonts the gestation has been 

 chiefly studied in Gambusia and Fundulus, two North American 

 genera, and in Anableps, the curious South American form 

 whose eyes are divided into two halves, one for seeing things 

 above and the other things below the surface of the water. It 

 must be remembered that the ovaries of Teleosteans are usually 

 closed sacs, the inner surface of which produces the eggs, each 

 originally enclosed in a separate cavity, called the follicle, in 

 the ovarian tissue. Gambusia patruelis is a species living in 

 the fresh waters of Virginia. The adult males are only I \ in. 

 long, the females if in. The ovaries are united into a single 

 sac and at each gestation twenty to twenty-five yellowish eggs 

 are produced. The eggs do not leave the follicular sacs in 

 which they are developed ; these follicles, instead of bursting 

 as in ordinary cases and liberating the eggs into the cavity of 

 the ovary, merely form a small opening in their walls and 

 through this opening sperms enter and fertilise the egg which 

 goes on developing within the follicle. In this species the 

 wall of the ovarian sac has disappeared and the follicles are 

 exposed to the body cavity, the young escaping at birth by an 

 abdominal pore. Another peculiarity is that the egg is not 

 provided with an egg-membrane, although in the allied genus, 

 Fundulus, this structure is present. An egg-membrane is not 

 essential to an egg that develops within its follicle, and thus we 

 can understand why it has disappeared in Gambusia. The 

 walls of the follicle are richly supplied with blood-vessels, and 

 the developing fish obtains its oxygen from the maternal blood 

 within these, but although the follicle contains liquid around the 

 embryo this does not appear to contribute to the nourishment 

 of the latter, for the yolk is not exhausted till the young fish 

 is completely developed and ready for birth, when its fins 

 and fin-rays are completely formed, and it resembles its parent 



