CYTOLOGICAL STUDIES OF CENTRIFUGED EGGS 641 
ally produced may lie in any relation to the micropyle. It was 
important therefore to determine whether the artificially produced 
blastodisc is formed in a different region of the egg from that 
occupied by the normal blastodisc, or whether the result is due 
to the turning of the egg within its membranes. 
It seemed to me that if serial sections were studied there would 
be no difficulty in determining whether the egg rotated within 
its membrane, or whether the blastoderm formed at an}' part of 
the surface. The nucleus of the normal egg lies in that part of 
the protoplasmic envelope immediately beneath the micropyle. 
Agassiz and Whitman have shown for this same egg that the proto- 
plasmic shell is thickest at the micropyle. They record for the 
egg of Ctenolabrus that at this region the protoplasm is thickest 
thinning out gradually towards the opposite pole where it is very 
thin. If the artificially produced blastodisc corresponds in posi- 
tion with the natural blastodisc, we should find the egg nu- 
cleus, or the polar spindle, at or near the center of the artificial 
blastodisc. But if the blastodisc forms in a new place it might be 
without a nucleus at first, or the nucleus might often lie in an 
eccentric position in the new blastoderm. If the polar spindle 
had developed before centrifuging, it might fail to be carried into 
the blastodisc, or if transported, it might lie at first obliquely at 
the periphery of the blastodisc. 
Serial sections were prepared of a number of eggs in different 
stages. The preparations were excellent and the series in most 
cases complete; yet I have searched in vain through some of the 
series to find either nucleus or spindle. I have found the egg 
nuclei or conjugating nuclei in onl}^ seven eggs; in two cases the 
polar spindle was formed. In all of these cases the nucleus or 
spindle lay exactly in the middle of the blastodisc. The failure 
to find the nucleus or spindle in other cases I attribute to the 
following conditions. The polar spindle of the fish egg is extremely 
small, and easily overlooked; any disturbance in the protoplasm 
such as that produced by the sudden heaping up of cytoplasm 
might make the discovery of the spindle still more difficult. 
Whether, in fact, the spindle would be buried in the mass of cyto- 
plasm or lifted to its surface, I do not know, since in one case the 
