VIVIPAROUS FISHES OF THE PACIFIC COAST. 
421 
large number of females at this same time (July 29) spermatozoa were found in 
all or nearly all the ovaries, but the spermatozoa were inactive , not showing a particle 
of their great mobility of December. Sections made through the entire length of the 
ovaries of this time showed large quantities of spermatozoa, extending between the 
ovarian sheets to the anterior end of the ovary, but especially abundant in the oviduct, 
just behind the ovarian sheets. Sections of ovaries of October and November also 
show spermatozoa, so that there is no doubt about the early transfer of spermatozoa. 
Development of ovarian eggs . — This subject, which properly belongs here, will be 
dealt with in a subsequent part of the paper when the development of the sexual 
organs themselves is described. 
The mature egg . — I have not succeeded in keeping eggs alive for more than an 
hour after they were taken from the mother ; and, in most cases, they perished after 
they had been removed from the ovary but a few minutes. The successive processes of 
segmentation and gastrulation could not therefore be observed. The methods could 
only be inferred from the different stages obtained from different individuals. Although 
I opened several hundred individuals, the series is not as complete as desired, because 
the individuals had to be opened when they were found, and there was no means of 
telling externally the length of time the contained eggs had been developing. In order 
to section developing eggs it is absolutely necessary to cut the zona. Many of the first 
eggs collected were not available for study because this was not done. The eggs are 
freed from the follicle before segmentation begins. In all probability fertilization 
takes place just before or just after they are freed. In a single case observed the 
second polar globule is being extruded and the male pronucleus is formed in an egg 
still inclosed in the follicle. The cells of the follicle show evident signs of degenera- 
tion. In other members of the family ( Amphistichus ) segmentation is carried on (to 
completion ?) before the eggs become free from the follicle. In an ovary with eggs 
with two cells there were also found others with four and with eight cells. 
The egg before it is freed from the follicle, measures on an average about 280 p 
(sometimes more than 300 p). During maturation the size of the egg proper is reduced 
to a diameter of 200 to 230 p, or to half to one-third of its original volume. In this 
manner a considerable breathing chamber is formed within the zona which retains its 
original extension. If younger eggs are put under pressure a similar contraction some- 
times takes place, leaving a space between the egg and the zona, which is traversed 
by a multitude of fine threads. That these green eggs can be made to contract by 
rupturing the zona probably indicates that the contraction of the ripe egg is due to the 
fact that at maturity the ovarian fluid gains access to the yolk on account of with- 
drawal of the granulosa cells from the zona. Similar reductions in the bulk of the yolk 
take place in other teleost eggs, as has been noted by several observers. 
The mature egg, just before segmentation, is divided into two portions by a well- 
marked constriction. The larger of these and the one of an apparently homogeneous 
substance is the germ ; the smaller, composed of very small spheres, the yolk spheres, 
is the deutoplasm. Situated in the middle point of the surface of the latter, or just at 
the entodermic pole of the egg, is a more homogeneous transparent mass sometimes 
containing one or two small spheres. The appearance of this structure is similar to the 
germinal vesicle and its nucleoli of the green egg, with the exception that its outlines 
are less regular, angles projecting in among the yolk spheres. A like structure has 
