HEREDITY AND VARIATIONS IN PROTOZOA 579 



'regained about the usual rate.' If the result of this experiment 

 may be regarded as typical, then it indicates that the lowering in 

 fission following conjugation is transient, recovery occurring sooner 

 or later. It is demonstrated that after conjugation the organism 

 and its progen,\' are weaker, or less resistant to external conditions 

 (shown by higher mortality, lagging fission-rate, unstable size, 

 abnormalities, etc.), for a certain time; and that complete recovery 

 to the normal state preceding conjugation occurs subsequently; but 

 I find no proof that from a race with a given fission-rate, another 

 race with a permanently different fission-rate has arisen as a result 

 of conjugation" (Dobell, 1914, p. 173). 



The source of the difficulty with Jennings' conclusions lies in 

 the cause of the transient lowering of the division-rate, of the 

 increased mortality and of the increase in the number of abnormali- 

 ties. Jennings finds the cause in bi-parental inheritance and in 

 "incompatible combinations" at amphimixis, but a more probable 

 cause, as Dobell intimates, lies in the conditions of the experiments 

 and, as there is reason to believe, in the use of a culture medium 

 which is unsuitable for Paramecium during the critical reorganiza- 

 tion period after conjugation (see Calkins, 1923). It is unnecessary 

 to repeat here all of the evidence for this conclusion but the nature 

 of the evidence is indicated by the fact that mortality after conjuga- 

 tion between indi\'iduals recently brought into the laboratory is 

 about 20 per cent to 30 per cent while after conjugation between 

 individuals that have been cultivated many weeks on the standard 

 medium used by Jennings the mortality rises to 94 per cent (Calkins, 

 1904), or in some of his own cultures from 3S per cent to 59 per cent. 

 If there is such high mortality among ex-conjugants we should ex]3ect 

 to find amongst those which were not killed a large percentage of 

 individuals with lowered division-rates, of abnormalities, and an 

 increased variability in these respects, without calling upon incom- 

 patible combinations to ex])lain them. 



Except for one case of real nuitations described below, there is no 

 experimental evidence to indicate that permanent variations in 

 Protozoa arise in any such al)rupt manner as -Jennings describes. 

 In long-continued cultures changes of organization do occur; such 

 changes are sporadic at first but gradually they predominate until 

 the original character becomes sporadic. This, for example, is the 

 case with one of the morj)hological features of Uroleptus mohilis. 

 In the early series the number of macronuclei was almost uniformly 

 8 (Fig. 1), but an occasional individual was found with more than 8 

 (up to 10 or 12). In the later series the number has l)ecome almost 

 uniformly 14 to 15. It is impossible to tell when the change occurred 

 as it has been so gradual; inbreeding has been strictly adhered to 

 in every series and each series coming from a single ex-conjugant 

 has shown more or less variation in this respect. 



