644 
T. H. MORGAN 
turns as a whole unless roughly handled. In consequence the 
spermatozoon entering the micropyle passes by the nearest path 
to the center of the future blastodisc. 
Contact with sea-water causes the surface of the egg soon to 
separate somewhat from the inner surface of the membrane, so 
that the egg may turn within the membrane. Any sudden or 
irregular movements of the egg at this time may cause the egg to 
shift, so that the center of the blastodisc no longer corresponds 
with the micropyle. Several years ago I showed that if the egg 
of the fish is not fertilized within a few minutes after removal to 
sea-water, polyspermy occurs, and in the light of the facts to be 
recorded this result is, beyond doubt, connected with the rotation 
of the egg within its membranes. The process of '^dry fertiliza- 
tion" so often recommended by pisciculturalists owes, perhaps, its 
success to the fact that in the absence of water the separation of 
egg and membrane may not occur until after fertilization; hence 
a larger percentage of normal embryos. 
Despite the close contact between the egg and its membrane, 
the centrifuge causes the egg to rotate as a whole within its 
envelope, as stated above. In consequence another part of the 
egg surface comes to lie beneath the micropyle. In order to 
reach the artificial blastodisc the sperm, entering through the 
micropyle, must pass into a foreign part of the protoplasm, and 
then to reach the blastodisc must be transported over the surface 
of the egg in the thin layer of protoplasm. The consequences 
of this disjunction of normal conditions lead to several results. 
In the first place the results show that spermatozoa may enter 
at any part of the surface, and that even if there were a localized 
region of the protoplasm for the entrance any other region does 
as well, so far as mere entrance alone is concerned; but in regard 
to polyspermy the results are different. There is evidence that 
several spermatozoa enter in most of the centrifuged eggs. The 
first that enter move or are carried by the shortest path towards 
the blastodisc. Those that enter later remain near the micropyle, 
where a small accumulation of protoplasm appears. The most 
probable interpretation of these facts seems to be that the 
occluding reaction in the egg, by means of which accessory or 
