RELATION OF RHYTHMS AND ENDOMIXIS. 45 



p. 455) shows that of seventeen endomixes in the four pairs of 

 cultures to be compared, 1 the synchronism was exact in only 

 six cases, approximately exact (i. e. t within one five-day period) 

 in six and inexact (differing by at least two five-day periods, 

 which in a total interval of never more than seven and seldom 

 more than five such periods is considerable) in five. These 

 facts, taken from Woodruff's own data, hardly seem to prove 

 his claim regarding the mainlenance of rhythmical constancy 

 and synchronism in Paramecium in either the same or different 

 environments. 



Were this negative evidence the only refutation of his con- 

 tention however I should hesitate to urge it too strongly. But I 

 have positive evidence from my own experiments leading to 

 directly opposite conclusions. 



The graphs in my previous paper (I. c.} show numerous cases 

 of irregularity in the time period of endomixis. In experiment 

 i (Figs, i and ia) for example line I shows a fairly regular en- 

 domictic periodicity with an average interval of eighteen days, 2 

 while line la, branched off from I on 5/8 was carried for a period of 

 forty-two days without any endomixis whatever. Here the dis- 

 turbance of the period is obviously due to changed environment; 

 I being carried for the most part in stale, and la in fresh culture 

 media. Experiment 2 shows that, whereas the first three en- 

 domicitic periods in line II, branched from I on 2/8, agree fairly 

 well with the corresponding periods in I, following the endomixis 

 on 3/5 no others occurred until 5/13, a period of sixty-nine days. 

 This change in period is probably traceable in part to changed en- 

 vironment, as suggested in my previous paper (1. c., p. 39). It 

 is probable howevei that some factor inherent in the cell itself is 

 also in part responsible since the line did not undergo the urea 

 treatment until 4/21, forty-seven days after the last previous en- 

 domixis occurred. Whatever the cause however the experiment 

 shows that marked variations in endomictic periods do occur. 



Both lines Ila and III which are derivatives of line II, while 

 showing essentially the same periodicity with lespect to each 



1 Not counting those at period 59 which may have been induced by the sudden 

 change to the stale culture medium. 



2 Omitting the doubtful record on 3/16 cf. footnote 5, p. 39 (I. c.). 



