168 PROTOZOA 



mviable as the Fl generation. Thus, although a complete bar 

 i gene flow was neither demonstrated nor controverted, intro- 

 gression of genes from one variety into the other must be either 

 very rare or nonexistent. 



The other four possible crosses between varieties, 1 by 3, 1 

 by 5, 1 by 7, and 3 by 8, regularly yield a viable Fl. The F2 

 generations and backcrosses prove, however, to be completely or 

 nearly completely nonviable. Butzel (1953) found abnormalities 

 even in the Fl of the cross between varieties 1 and 7. Like 

 Levine, he was unable to prove that recombinants occur in the 

 F2 or backcrosses. I have obtained a little evidence, not fully 

 satisfactory, that the rare survivors in the F2 or later generations 

 may sometimes be recombinants. Some clones behave as if they 

 were the plus (or minus) mating type of both parental varieties, 

 but a descendant of such a clone after a later fertilization may 

 behave as if it had a mating type of only one of the parental 

 varieties. Whether this is due to change of dominance relations in 

 a heterozygote or to recombination is not clear. Further, descend- 

 ants of the hybrids have been maintained for a year or more. 

 They reproduced slowly, but mated avidly when given a chance. 

 It seems unlikely that life could have been maintained so long 

 without recurrent fertilizations, although it is possible, in view 

 of some of our recent work (Sonneborn, unpubl. ), that these 

 may have survived by repeated regenerations of functional nuclei 

 from the original hybrid macronucleus instead of developing 

 macronuclei from the fertilization nucleus after each fertilization. 



Thus, knowledge of the consequences of intervarietal crosses, 

 in the five combinations in which they can occur, is still not 

 sufficient to give a definite answer to the question of whether 

 gene flow can occur among those varieties. On the other hand, 

 the difficulties in analyzing the hybrids are due to the low viabil- 

 ity of the later generations and to the abnormality of the survi- 

 vors. This shows that, if genes flow between varieties, the flow 

 can be at most but a trickle and iu great danger of drying up. 

 In short, the genes of every variety are virtually isolated from the 

 genes of every other variety. Before the significance of this ge- 

 netic isolation of the varieties can be properly evaluated, it is nee- 



