ASSORTIVE MATING IN CHROMODORIS ZEBRA 285 



of the animals in proper physiological condition for reproduction 

 are simultaneously brought near together; 3) the immunity of 

 the adults, and of their egg ribbons; and 4) the practice of 

 assortive mating with respect to size. 



Concerning the last of these points, it has been shown that 

 assortive coupling probably operates by conserving the genital 

 products and by insuring that as the result of any particular 

 copulation the number of eggs fertilized and liberated is, on the 

 average, greater than that which would be the outcome of ran- 

 dom pairing. On these grounds I believe it not entirely correct 

 to say, as Pearl does ('07, p. 274), that assortive mating in 

 somas is probably vastly less important than it would be if it 

 occurred in the fusion of gametes. It is well known that some 

 littoral molluscs which pass through a pelagic stage (e.g., Ischno- 

 chiton; Heath, 1899) lay vast numbers of eggs, very few of which 

 are able to attain the adult state. For the continuance of the 

 race — and for the multipUcation of its members, thus possibly 

 affording greater opportunities for its evolution — any influence 

 tending to further the production of numerous young is clearly 

 of the utmost consequence. 



From the standpoint of the theory of adaptation it is, I take 

 it, highly significant that this ''adaptive" result may be under- 

 stood to follow mechanically from the circumstance that Chromo- 

 doris of different sizes are sexually mature. This is not exactly a 

 case which can be grouped with the striking instances of adapta- 

 tion to which Eigenmann, Cuenot, Loeb ('16, p. 343), and others 

 have applied the term (or idea) of 'preadaptation.' It is rather 

 one of those eminently 'purposeful' practices — yielding no spe- 

 cific ulterior advantage to the individual, yet probably signifi- 

 cant in the history of the race — which nevertheless depend auto- 

 matically upon the structure of the organism (Parker, '13),^ in 

 itself determined by quite independent and unrelated causes. 



' In this connection it may be noted that in the interesting case of some vivip- 

 arous teleosts there is exhibited a condition which might at first be taken to 

 favor assortive mating, in the sense that homogamy of a certain type might 

 thereby be rendered "valuable to the species." There is, however, no evidence 

 of assortive pairing, and indeed, with respect to the character in question, it 

 could not occur. According to Eigenmann (1894, p. 417), in Cymatogaster the 



