642 
T. H. MORGAN 
spindle was imbedded, and in the other was on the surface; but 
in the latter the centrifuging was not hard and the blastodisc 
still fiat. The resting nucleus is not sharply marked off from the 
cytoplasm, and is difficult to detect, owing to the presence of vac- 
uoles in some of these centrifuged eggs. In many cases the egg is 
filled with them ; in other cases they are scarcer — owing to differ- 
ences perhaps in the state of maturity of the egg. As the centri- 
fuging was done soon after the egg was stripped from the fish it is 
possible that the germinal vesicle may have just broken do wn , which 
would render the detection of the chromatin much more difficult. 
It would have been of no avail to wait before centrifuging until the 
easil}' detected segmentation spindle formed, because at this time 
as is well known the egg may move readily within its membranes. 
Finally, although I have looked for the nucleus or spindle in other 
parts of the surface of the egg I have never seen them there. It 
would be most laborious to thoroughly search the periphery of 
every section, and since the eggs were punctured in order to 
imbed, the absence of discovery might be attributed to the nu- 
cleus lying near the puncture. 
After centrifuging, the superficial cytoplasm, except at the 
blastodisc, is reduced to a minimum. If a spindle formed elsewhere 
than in the blastodisc, an accumulation of protoplasm would be 
expected to occur, but I have seen none. I am inclined therefore 
to think that the failure to find a nucleus in the blastodisc in many 
eggs is due to technical difficulties, not to its absence. Neverthe- 
less, despite the time spent in this side of the problem, the outcome 
is on the whole negative, since there are too many failures to find 
a nucleus for the positive cases to count for much, as such cases 
might be attributed to a coincidence. I therefore returned to 
the experimental side once more, where, fortunately, I have ob- 
tained convincing evidence that the egg may move as a whole 
within its membranes. The essential part of this evidence will 
be given in the three following paragraphs. 
1. If the blastodiscs be artificially produced, and the eggs 
turned over and over so that they are no longer orientated and 
then again centrifuged for a few turns, the blastodiscs orient at 
ance. The experiments show clearly that the eggs orient within 
