288 W. J. CROZIER 



pairs in nature also suggest that size as such, and not (?) position 

 in the reproductive cycle, is the basis of selection. . 



4. C. zebra is functionally hermaphroditic, and effective re- 

 ciprocal insemination is practiced. 



5. The number of eggs deposited in a single mass varies from 

 2,000 to 20,000, and is almost directly proportional to the 

 length of the animal. The larger animals possibly lay several 

 more egg masses in a given time than do the small ones. 



6. It is consequently of advantage to the species that large 

 individuals should mate together. In this way the numbers of 

 eggs fertilized, and presumably of larvae set free, as the result 

 of any one mating, is on the average greater than that which 

 would be produced if randon pairing were the rule. Moreover, by 

 means of assortive conjugation eggs (and sperm?) are conserved. 



7. Selective pairing has in this case a result which is there- 

 fore of a distinctly advantageous or ''purposeful" character, since 

 it makes for the multiplication of the species. This "adaptive" 

 behavior is nevertheless clearly an automatic consequence 1) of 

 the fact that the eggs are fertilized internally, necessitating the 

 copulation of adults; and 2), more immediately, of the very 

 condition that the size of the body is not identical in all sexually 

 mature individuals. 



IX POSTSCRIPT 



The preceding account is based on observations made in late 

 spring (April 5-May 5, 1917). It was thought to be of sufficient 

 importance, as part of a further study of this matter, to see to 

 what extent similar findings would result at another season of 

 the year. Early in December, 1917, at the beginning of a 

 period of shoreward flocking (Crozier, '17 c), naturally occurring 

 pairs were very infrequent in Fairyland Creek, the area dealt 

 with in the collections of the previous spring; not enough ma- 

 terial of a homogeneous order could be secured for the purpose 

 in view before the work had to be interrupted, although in the 

 isolated pairs secured it was evident that assortive conjugation 

 was the rule. Recourse was therefore had to a mating experi- 

 ment, carried out with animals secured at this time. 



