PROTOZOA 



elusion and shows in detail how P. bursaria fits the group B sys- 

 tem. Exposure of the four macronuclear anlagen to a common 

 cytoplasm differentiates all of them to control the same mating 

 type. If my formulation of the operation of the cytoplasmic action 

 in group B is correct, then there should be four, instead of two 

 cytoplasmic differentiators in flux equilibrium in variety 1 of P. 

 bursaria. Of the four, one is in high concentration in one mate, 

 another in the other mate; the mixture of these two cytoplasms 

 thus gives high concentrations of two of the four. This accounts 

 readily for the fact that usually over 80% of the synclones pro- 

 duced from a cross are like one or the other parent in mating 

 type, and that, of these, roughly half are usually like the one 

 parent and half like the other parent. Whether temperature can 

 affect these proportions is unknown. The hypothesis also explains 

 some other remarkable observations, as will now appear. 



The phenomenon of self-differentiation has this peculiar fea- 

 ture: it always results in the same two types in any one clone, 

 but results in different types in different clones of the same 

 original mating type. That is, a clone of type A may differentiate 

 into A and B; another clone of type A may differentiate into A 

 and C. Yet if either the new A or the new B in the first clone 

 self-differentiates again, the A will always yield B and the B will 

 always yield A. Likewise, in the other clone, the new A and C 

 will always differentiate only into C and A, respectively. All 

 combinations of two types arc possible in different clones, but 

 any one clone has only one such combination of possibilities. 



The physical basis of these remarkable relations has never been 

 elucidated. Kimball (1943) suggested that each possible com- 

 bination of alternative mating typos producible by self-differen- 

 tiation was determined by a different genotype. He agreed with 

 Jennings' conclusion that the usual uniformity of mating type in 

 a synclone bespoke genie determination and suggested two rea- 

 sons for Jennings' failure' to find workable genotypic formulas. 

 First, each formula should be not for a particular mating type, 

 but for a particular pair of alternative types. Second, the formu- 

 las should take into account Chen's (1940) finding of polyploidy 

 in P. bursaria. Kimball's proposals have not thus far led to the 



