174 On the Genetics of the Ciliate Protozoa 



descendants of a pair of exconjugants must form two fairly homogeneous 

 lots — in order that any comparison between them is possible. By taking 

 the mean for some lines descended from each exconjugant, such a homo- 

 geneity may be introduced when it does not really exist. For if the 

 progeny of an exconjugant differ from one another — the differences 

 being, urgumenti causa, caused by conjugation — how can any real 

 comparison be made between all the different organisms which could 

 be obtained from one exconjugant — the number is unlimited — and 

 those from its partner ? It is physically impossible to study the rate 

 of fission in more than a few lines descended from an exconjugant, for 

 at each fission the number of possible lines to be studied is multiplied 

 by two. Granted that conjugation causes variation, so that the lines 

 derived from different progeny of one individual exconjugant display 

 different fission rates, how is it possible to reach any definite conclusion 

 by studying certain arbitrarily selected lines representing only an in- 

 finitesimal fraction of all possible lines ? It seems to me that if 

 conjugation gives rise to variations of this sort — as certain experiments 

 seem to show — then any real demonstration of " biparentaj inheritance " 

 in fission rate is impossible. I cannot comprehend how Jennings's 

 results in this connexion can have any value beyond suggestiveness. 

 111. Calkins and Gregory (1913) have recently published an 

 account of the variations observable in different lines of Paramecium 

 derived from a single exconjugant. In one case as many as 30 lines 

 from one exconjugant were studied — that is, a single line from each of 

 30 out of the 32 individuals formed by the first five fissions after con- 

 jugation. The authors conclude: "The results of this study show that 

 physiological and morphological variations in the progeny of a single 

 exconjugant of Paramecium caudatum are fully as extensive as the 

 variations between progenies from different exconjugants. The argu- 

 ments based upon the latter variations to the effect that conjugation 

 is for the purpose of originating variations cannot therefore be sustained." 

 The authors appear to believe that they have invalidated Jennings's 

 general conclusion that conjugation causes variation. If so, their 

 argument is a palpable non sequitur. What they have shown is that 

 the progeny of an exconjugant differ from one another — that there is 

 considerable variability among them. Whether the variability is the 

 effect of conjugation or not, there is no means of judging: for no 

 comparative study of non-conjugants and split conjugants of the same 

 race appears to have been made — a study which is es.sential if the effects 

 of conjugation are at issue. The real importance of these observations 



