80 Papers from the Department of Marine Biology. 



to 3^ minutes. With other males fertilized at the same time, eggs from 

 the same females formed no membrane at all. Correspondingly wide 

 variations occurred in the other experiments (6/9, 7/3, 7/2, 7/19). 



Eggs that form membranes rapidly by one male may develop them 

 slowly by another of the same age and apparently similar condition; 

 but usually the rate of membrane formation is conditioned by the eggs. 

 While the rate may vary with different males, yet each female tends to 

 keep the same relative rate i. e., there are slow-membrane-forming 

 females and rapid-membrane-forming females. 



In every experiment where different females were fertilized by the 

 same male marked differences in the rate of membrane formation 

 occurred. The correlation with cleavage on the one hand and jelly 

 layers and size of the eggs on the other will be given later. 



Since the rate of membrane formation varies with different males 

 and with different females under the same experimental conditions, 

 any comparison of results of one group of eggs with another must be 

 made with great caution. 



VARIATION IN CLEAVAGE. 



Cleavage affords a definite, finely graded, and practical means 

 of measuring variation in the eggs of different females. I have there- 

 fore given considerable attention to the variations in rate and in total 

 cleavage in all three sea-urchins. 



TOXOPNEUSTES. 



The total cleavage is dependent upon the kind of male as well as the 

 kind of female, just as we have observed in the case of the fertilization 

 membrane. For example, 5 samples of the eggs of a given female were 

 fertilized by 5 different males; the percentage that cleaved in 1 hour in 

 these 5 samples was 80, 23, 35, 68, and 43, respectively, a range of 57 per 

 cent. (See table 5, experiment 7/8.) 



When different samples of eggs of a single female were fertilized by 

 the same male the maximum difference was within 5 per cent i. e., the 

 experimental error was below 5 per cent. 



When different females were fertilized by the same male, all suspen- 

 sions being standardized, the variation in cleavage was surprisingly 

 large. It varied from 1 per cent in experiment 7/2B to 98 per cent in 

 many experiments, 7/21, 7/19, 7/14, 7/12, etc. The range of varia- 

 bility as expressed by the difference between maximum and minimum 

 cleavage for any given experiment in which a given male fertilized 

 different females, was 11, 15, 16, 39, 48, 51, 52, 53, 54, 55, 57, 72, 74, 75, 

 and as large as 87 per cent. 



In a second series of experiments different males were used either 

 with samples of a given female or with different females. (See experi- 

 ments 7/1, 7/13.) 



