SOME PROBLEMS OF COELENTERATE ONTOGENY 503 
named were in early stages (two- to eight-cell) of cleavage. This 
would seem strongly to indicate their deposition at perhaps five 
or six o’clock in the morning or thereabout, as recorded in my 
notes of July 11th and 12th. In other cases development had 
reached the morula stage at nine a. m., which would lead to the 
conclusion that liberation of sexual products had occurred about 
midnight. While it is true that in many hydroids the liberation 
of eggs and sperm occurs at a fairly constant time, yet there 
are others in which this is not the case, and in which such ripen- 
ing and discharge is a more or less continuous process during the 
breeding period. _ 
The character of the egg is much like that of Pennaria, though 
it is much smaller. Both are alike in general texture of proto- 
plasm, contain yolk, and similar inclusions. There is present a 
pigment similar to that in the eggs of Pennaria, though less marked 
in color. Like those of the latter, the eggs are devoid of a defi- 
nite membrane. They are rather heavy and sink promptly when 
set free. By reason of this it was practicable to suspend colonies 
in shallow vessels within wire baskets under docks in freely cir- 
culating water and with little liability of their being lost. This 
was a matter of some importance; for, despite the best precau- 
tions, these hydroids soon deteriorate in vitality under the arti- 
ficial conditions of the laboratory, while by suspending them in 
open waters about the docks they thrive almost as if in the natural 
habitat. 
1. Cleavage. So far as I am aware the only definite work on 
cleavage of Hydractinia is that of Bunting (94). In this paper 
we have a characteristically symmetrical portrayal of the process. 
In general surface aspects it 1s represented as almost mathemat- 
ical in its regularity and symmetry. 
That the earlier cleavage phases in perhaps a majority of the 
eges conform to this in greater or less degree is probably true. 
But that it represents with any degree of accuracy the average 
behavior of this phenomenon as a whole none who had carefully 
followed it could for a moment admit. It has been difficult to 
conceive how, except by a selective process, any such account could 
have been formulated. It is quite easy to see that by directing 
