T. M. SONNEBORN 279 



probability of doing so is less in two-type varieties, autogamy 

 may serve as a safety device to cover the risk of failure to meet 

 a mate. 



Sonneborn (1955) agreed with some of these views and to 

 some extent also does so in the present paper. Yet, taking into 

 account other facts and more comprehensive correlations, I have 

 here developed in detail the thesis that all of them have their 

 primary significance as mechanisms adapted to different systems 

 of breeding, either inbreeding or outbreeding. The association 

 of a considerable period of immaturity with multiple mating 

 types is not interpretable on Siegel's hypothesis. These two fea- 

 tures seem to act at cross purposes: one favors mating, the other 

 prevents it. Their association is, however, adapted to outbreeding. 

 In varieties that have these features, it is important to find a 

 stranger, not just to undergo fertilization indiscriminately. The 

 absence of autogamy and the long life cycle in such varieties 

 have the same meaning, as do the other features discussed 

 throughout this paper. As implied here and developed in detail 

 above, the length of the life cycle is highly variable from variety 

 to variety; there is no need for quick fertilization in an outbreeder 

 and a short life cycle is adequate for inbreeders with their various 

 devices: little or no immaturity, much selfing, early autogamy, 

 and various genetic means of assuring presence of both mating 

 types. Actually, the situation seems to be just the reverse of the 

 one postulated by Siegel: in two-type systems, the whole biology 

 of the organism is designed to assure early meeting of related 

 mates; whereas, in multiple type systems, usually the meeting of 

 mates is prevented during early life so as to increase the chance 

 of mating with a stranger. 



So far as O. bifaria is concerned, the variety examined by 

 Siegel seems clearly to be an outbreeder. It rarely selfs; it has 

 no autogamy; it has multiple mating types; it has a considerable 

 period of immaturity, long maturity, and a long life cycle. Only 

 the questions of inbreeding degeneration and outbreeding via- 

 bility remain unsettled. The high death rate after selfing fits the 

 picture, but the variable results of outcrosses need clarification. 



