274 PROTOZOA 



The Breeding System. Superficially the mating relations seem 

 like a very fumbling and inefficient arrangement to encourage 

 outbreeding, inefficient because not all combinations of diverse 

 types can mate, and fumbling because it can induce selfing as 

 well. However, there appear to be good reasons to suppose the 

 system is not really as bad as it seems. In fact, when one takes 

 into account all the information discovered in the laboratory, but 

 makes allowance for the differences between the laboratory con- 

 ditions used in analysis and those that probably prevail in na- 

 ture, the system appears to be well suited for outbreeding and 

 neither fumbling nor inefficient. Actually, it seems if anything to 

 be a more nearly perfectly developed outbreeding system than 

 that of any multiple type system in Paramecium or Tetrahy- 

 mena. I shall now try to make this clear. 



First, the six known types were derived from but two neighbor- 

 ing individuals. Both were heterozygotes for the hormone genes 

 — in itself a little indication that we are dealing with an out- 

 breeder that does not prevailingly self in nature. Three of the 

 four genes at the hormone locus differed in these two individuals. 

 These are the three alleles the combinations of which yield the six 

 known mating types. Is it reasonable to suppose that by chance 

 these two individuals possessed all the hormone alleles that exist 

 in nature in this species? Decidedly not, in my opinion. It seems 

 far more likely that a wide sampling of various natural popula- 

 tions would have revealed a large number of such alleles. The 

 situation might even be comparable to the one known for incom- 

 patibility alleles in Basidiomycetes and higher plants, in some of 

 which a hundred or more alleles have been estimated to occur. 

 Certainly it would be a miracle if all the alleles that exist in this 

 variety of E. patella were the three found in the only two in- 

 dividuals studied. With many multiple alleles, there would be 

 many mating types; and multiple mating types are, as has been 

 repeatedly emphasized in this paper, an adaptation for outbreed- 

 ing- 



Second, Kimball reports the regular existence of an immature 

 period of a month or more following conjugation and Katashima 

 (1952) found a comparable immature period (three weeks or 



