240 PROTOZOA 



find it with certainty. These conflicting reports suggest that 

 autogamy may occur in some varieties and not in others; but fur- 

 ther investigation is called for. The relation of selfing to autogamy 

 seems also to differ in different materials. Giese and Arkoosh 

 (1939) found that autogamy was followed by the occurrence of 

 conjugation in a culture previously pure for one mating type; 

 autogamy had resulted in the differentiation of stable comple- 

 mentary mating types. Gilman (1941) studied cultures descended 

 from single individuals and could find no autogamy, although 

 selfing occurred within single cultures. It would seem that, as 

 with different varieties of P. aurelia, some varieties of P. cauda- 

 tum. differentiate mating types without prior autogamy, while 

 others do not. However, a special peculiarity in P. caudatum 

 seems indicated. Gilman ( 1939, 1941 ) observed that the cultures 

 which selfed, presumably without prior autogamy, in some cases 

 contained animals which, when isolated, gave cultures pure for 

 different mating types. In other cases, the selfing was apparently 

 due to transient change of mating type, for animals isolated from 

 the culture, even those that were coming together to mate, all 

 gave rise to the same mating type or to predominantly the same 

 type from each isolate. These varied observations indicate that 

 there may be several mechanisms in P. caudatum which bring 

 about mating between progeny of the same ancestral individual. 

 A different type of selfing was observed by Hiwatashi ( 1951 ) 

 in variety 3, the variety that may be an outbreeder. He marked 

 cultures of the two mating types with different vital dyes and 

 observed that about 5% of the conjugant pairs obtained by mixing 

 the two cultures consisted of two animals of the same color. This 

 is selfing induced by mixture of pure types. It is not mediated by 

 detectable hormones in the medium. I doubt whether this sort ol 

 induced selfing plays much of a role in nature, for it depends 

 upon the formation of an agglutinated group consisting of at least 

 two animals of one mating type and one of the other. It is un- 

 likely that the formation of agglutinated groups occurs in the 

 diluted populations of natural waters to anything like the extent 

 observed with concentrated cultures in the laboratory. Yet this 

 sort o! selfing loo remains as a natural possibility. 



