VIVIPAROUS FISHES OF THE PACIFIC COAST. 
443 
Those cells which under certain conditions have the power to reconstruct the whole 
organism are the sex cells. But it was seen that in the infusorian the macronucleus 
has its function suspended soon after preparations are made for conjugation and 
that it entirely disappears after conjugation. The macronuclear functions of the sex 
cells ought therefore also to become suspended if the comparison between the two 
organisms is to be complete. This suspension must take place the moment of the ulti- 
mate division of the germinal epithelial cell or soon after, and the egg has become 
irrevocably an egg or the spermatozoon irrevocably a spermatozoon. A partial sus- 
pension of these functions is evidenced by the absence of further divisions except 
during the formation of the polar globules, which in the infusorian are seen to be 
formed from the micronucleus. The formation of the polar cells may therefore be 
looked upon as products of the micronuclear elements of the germinal vesicle and do 
not vitiate the supposed suspension of macronuclear divisions. The macronuclear 
element of the germinal vesicle is eliminated as the yolk nucleus in eggs, and as the 
Nebenkern in spermatozoa. In eggs without yolk (ForsJcalia), and whose ovarian his- 
tory is consequently short, it is eliminated just before maturation, as the metanucleus. 
In eggs developing a large amount of yolk and whose ovarian history is prolonged it 
arises at a time corresponding to the maturation of non-yolked species, i. e., at the 
beginning of yolk formation as the yolk nucleus. In those in which it appears com- 
paratively late ( ForsJcalia , Cymatogaster) it may remain during some of the early 
stages of development. In those species in which it appears early it is lost in the 
yolk long before maturation. I have explained why it may be present in one species 
and absent in another closely related one. Its suspension of activity need not be sudden, 
and it is not unlikely that it retains some of its functions in some eggs even after it is 
expelled from the germinal vesicle, and it may then be active and entirely used up in the 
building up of the yolk, as has been suggested by a number of observers. That its 
functions are in some way connected with the yolk is certain from the close associa- 
tion between yolk and yolk nucleus, which has given it its name in eggs.* 
The direct nuclear division frequently seen in degenerate tissues extends the 
possibility of comparison between protozoa and metazoa. Such cells may be compared 
with infusoria which have passed the period of maturity and in which the micronuclear 
element, which is always accompanied by indirect division, has been lost. 
The reason for the complicated process of cell division seen in karyokinesis is 
evident in all cases where an exact distribution of the halved chromosomes is essen- 
tial. In cases like the macronucleus in protozoa and the degenerate cells of metazoa 
where an exact division can be of no advantage, direct division takes place. I do 
not mean by this to insist that there is a further comparison than the above between 
the direct division of the macronucleus and that of degenerate cells of metazoa. The 
former is primitive. 
* The above account of the yolk nucleus was. written before the appearance of Weismann’s “ The 
Germplasm.” In the phraseology used in this theory I hold that each cell (except the degenerate 
ones in which direct nuclear division takes place) contains germplasm aside from ids from which all 
the determinates but those controlling the cell have been removed in carrying the cell to its final des- 
tination. All cells, the reproductive cells included, are controlled by determinants which are not 
directly derived from the ids of germplasm contained in the nucleus but from ids which have been 
simplified during ontogeny. These simplified ids are removed as the yolk nucleus. 
