242 PROTOZOA 



pointed out that this indicates similarity between R of variety 4 

 and F, G, H, and J of variety 2, because all these mate with E, 

 K, L, and M; and between S of variety 4 and E, K, L, and M of 

 variety 2, because all mate with R. Obviously one cannot express 

 such similarities in terms of a basic binary system of plus and 

 minus types. 



Metz (1954) suggests that mating reactions nevertheless de- 

 pend upon reactions between pairs of complementary mating 

 substances, even in varieties with multiple mating types. Accord- 

 ing to this hypothesis, there are three pairs of complementary 

 substances in a system of eight interbreeding mating types: alpha 

 substances A and a, beta substances B and b, gamma substances 

 G and g. Each mating type would have one alpha, one beta, and 

 one gamma substance. The formulas for the eight mating types 

 would therefore be: ABG, ABg, AbG, aBG, Abg, aBg, abG, and 

 abg. A system of two mating types would have only one pair of 

 substances, e.g., A in one type and a in the other. The mating 

 type with the A substance would thus be complementary to the 

 four mating types with the a substance in an eight-type set. This 

 hypothesis is consistent with the observation that one mating type 

 in the two-type system of variety 4 mates with four mating types 

 of the eight-type system of variety 2. 



Isolation of the Varieties. The isolation of the varieties of P. 

 bursaria is partly sexual. All combinations of varieties, except 2 

 and 4, are completely unable to mate with each other. (Jennings, 

 1939, p. 223, remarks that very rarely a single pair of conjugants 

 is found when an immature clone is mixed with a mature clone 

 of a different variety. He suggests that such pairs are not crosses, 

 but sclfing in the mature clone. ) 



In the exceptional combination of variety 4 with variety 2 

 Jennings and Opitz (1944) found that all four of the possible 

 reactive combinations of mating types of these two varieties (R 

 with E, K, L, and M) led to the same result: most of the con- 

 jugants died without separating; those that eventually separated 

 died within four days without undergoing a single fission. Chen 

 (1946a) examined in detail the cytology of the invariably fatal 

 (losses. Although the nuclear processes proceed normally for a 



