238 PROTOZOA 



which does not lead to conjugation. These varieties, together with 

 varieties 1, 4, 5, 7, 11, 12, and 13, show no other sexual reactions 

 except between the two mating types of the same variety. Strains 

 of the same variety do interbreed, although little or no informa- 

 tion is in the record concerning later generations. Presumably, 

 however, each of these nine varieties constitutes a group of popu- 

 lations among which gene flow is possible, but between which 

 there is no gene flow. 



The remaining seven varieties, 2, 8, 9, 10, 14, 15, and 16, con- 

 stitute what may be called group 2. (Gilman has spoken of it as 

 the "variety 2 complex.") Certain combinations of mating types 

 of different varieties in group 2 give sexual reactions and all the 

 varieties of this group are interlinked in this way directly or 

 indirectly. The intensities of these intervarietal reactions vary 

 from the weak sort, mentioned above as occurring between types 

 VI and XI, up to intensities equal to the maximal reaction nor- 

 mally obtained between complementary types of the same va- 

 riety. Nevertheless, according to Gilman, each mating type of 

 each variety has a uniquely defined pattern of mating reactions. 



Whether the varieties of group 2 are sharply marked off from 

 one another genetically, as the varieties of group 1 are sexually, 

 seems to be as yet unsettled. Gilman has been studying this, but 

 has not yet published his results. Meanwhile, Pringle (1955) and 

 Johnson (1955) have been obtaining remarkable results on a 

 considerable number of European strains. All these strains belong 

 to group 2, but many of them have not yet been assigned to 

 definite varieties. Indeed Pringle and Johnson find it difficult or 

 impossible to sort their strains into varieties. Whether Gilman can 

 do this, as lie is now attempting to do, remains to be seen. The 

 two former workers have the impression that matings between 

 i titers of the same local population usually give a fair propor- 



ii '»! viable V\ clones, whereas crosses between representa- 



diflerent local populations give highly variable results 



isually a less viable Fl. Pringle suggests that each local 



popul ttion may have a gene pool which is to a considerable 



degree isolated from the gene pools of other local populations. 



Certainl) the results at present are difficult to fit into a system 



