82 THE BIOLOGY OF STENTOR 



found only among the largest animals. All we can assert, however, 

 is that whatever precipitates division is usually correlated with 

 size increase, for many circumstances demonstrate that size alone 

 is not the determining factor. Stentors of smaller than maximum 

 volume can divide, and yet when many stentors are grafted together 

 the combined ** cell ", of extraordinary mass, generally shows no 

 tendency whatever toward fission. 



The idea that division is caused by a progressive deviation from 

 some normal ratio of macronuclear volume to cytoplasmic volume 

 (Popoff, 1909; Causin, 1931) does not seem to be confirmed in 

 Stentor. It is sufficient to say that in a variety of experiments in 

 which nucleus or cytoplasm is at once added or subtracted there is 

 no clear evidence of division occurring promptly as a consequence. 

 Nor does micronuclear mitosis trigger division (Weisz, 1951b) 

 because (a) mitosis occurs near the end of fission (Johnson), 

 (b) reorganizers and regenerators also show mitoses though not 

 dividing, (c) emicronucleate stentors divide (Schwartz), (d) and 

 many other amicronucleate ciliates reproduce normally. 



In certain unpublished experiments I have found that when the 

 membrane liar band or the mouthparts alone are removed from 

 rather large specimens the stentors almost always promptly 

 divided and the resulting anterior daughters with the abbreviated 

 feeding organelles then regenerated a new set (see Fig. 39B). The 

 same operation performed on the smallest stentors in the culture, 

 however, yielded no divisions at all. Hence the combination of size 

 with this specific operation seems to have done the trick, but further 

 study is needed. 



Weisz (1956) conceived that the problem of division in Stentor 

 could be approached by determining the effect on smallest, post- 

 fissional coeruleus of large pre-divisional and dividing cells grafted 

 to them. First he fused largest animals which were soon to divide 

 but which had not yet produced a division primordium with 

 smallest cells or products of recent division which therefore would 

 not be expected to divide until they doubled their size. Division 

 of the larger partner was then greatly delayed but this could be 

 attributed to injuries of operation since ungrafted controls also 

 postponed fission if they were sliced into. After this delay a 

 division primordium finally appeared in the larger partner, 

 followed by an induced primordium in the smaller, promptly if 



