C. Do BELL 171 



content with the baldest statements and criticisms of Jennings's general 

 conclusions. It may be said at once that the programme of his work is 

 admirable. He has studied both "wild" and pure strains of Furamecium 

 caudatum and P. aurelia, by means of observation and experiment and 

 where possible by biometric methods. Certain characters (fission rate, 

 size, etc.) were studied in individuals or their progeny belonging to 

 the four classes (a) non-conjugants, (6) conjugants, (c) exconj ugants, 

 (d) "split" or "unpaired" conjugants — i.e. individuals forcibly separated, 

 and bred further, after they had united for conjugation \ Knowledge 

 of the behaviour of all these classes of individuals should obviously give 

 definite information concerning the effects of conjugation. 



106. The first general conclusion to which Jennings comes is that 

 conjugation causes variation'^ — " Conjugation produces within a pure 

 race heritable differentiations ; so that as a result races diverse in their 

 heritable characters arise from a single race with uniform heritable 

 characters." This conclusion is drawn — speaking generally — fi-om the 

 demonstration that, in a given race, the progeny of non-conjugants and 

 split conjugants are alike, but differ from the progeny of exconjugants. 

 There is greater variability among the progeny of the last. Since 

 conjugation is the only known factor which differentially affects the 

 two groups, it seems justifiable to conclude that it is in some way a 

 " cause " of variation. Yet this conclusion is remarkable. We have 

 seen that non-conjugating "pure lines" are constant in character — the 

 differences which the constituent individuals display being temporary 

 and not heritable (§ 65). We have seen further (§ 90) that within the 

 pure line " assortative mating " occurs, so that the members of a con- 

 jugating pair of organisms are more alike than those of a non-conjugating 

 pair selected at random. And yet after conjugation these like individuals 

 produce progeny which are unlike themselves and their race. 



107. Jennings's second conclusion is at first sight even more strange. 

 It is that " conjugation results in biparental inheritance." (See Jennings 

 and Lashley, 1913, 1913 a.) The meaning of this misleading expression^ 

 will be clear from the evidence upon which the conclusion rests. We 

 have seen (| 106) that the progeny of non-conjugants and split con- 

 jugants, within a pure race, are alike : and that the progeny of conjugants 



1 E. Hertwig's experiment {§ 103). 



'' This, of course, is a doctrine long ago promulgated by Weismann. 



' I say "misleading" although Jennings explains the term (Jennings and Lashley, 

 1913, p. 457): for nothing is here "inherited" from the "two parents" save the power 

 to differ from them in a given respect. 



