2IO THE BIOLOGY OF STENTOR 



chains, and two complete sets of feeding organelles. Above all, the 

 pattern of lateral striping is double, with two primordium sites 

 or loci of stripe contrast ; and this is seen to be crucial, for whenever 

 doublets revert to singles there is always the obliteration of one 

 primordium site, after which all other aspects of the complex 

 become single. And doublets could be converted at once into 

 singles by excising one of the primordium sites, even if the 

 bistomial head was left intact (Fig. 58A). 



It was difficult in cultures to catch doublets in the act of trans- 

 forming into singles, but something of how this occurs may be 

 indicated in the following. Figure 5 8b shows an asymmetrical 

 doublet which was in fact not a 2-mass but produced by grafting 

 a primordium site into a single animal. Such specimens remained 

 as doublets for several days, but then one of the primordium sites 

 disappeared as such, either the host site or that of the graft 

 transforming into uniform lateral striping, for there was no evidence 

 of stripe resorption. The transformation illustrated in (c) was 

 instigated in the anterior half fragment of a broad symmetrical 

 doublet. Reduction to half the original size resulted in the length 

 of the membranellar bands being greatly reduced in situ until they 

 became proportionate to the new cell size, but the mouthparts 

 remained large. One primordium site then disappeared as its 

 contrasting pigment stripes became of uniform width. While this 

 was occurring the mouth subtending these stripes separated from 

 the membranellar band and moved into the frontal field where it 

 was gradually resorbed. The two bands then joined together and 

 the final result was a single stentor produced even without the 

 formation of a reorganization primordium. 



Although there is evidently a strong tendency towards unifica- 

 tion of shape, one may speak of a reversed propensity of sets of 

 lateral striping to establish separate shapes, as if a complex which 

 cannot achieve complete singleness then settles on a frank expres- 

 sion of its multiplicity. Doublets, especially when so oriented as 

 to have two frontal fields, become double cones or Siamese twins, 

 and enduring triplets also develop *' cleavages " making them 

 triple shaped (Fig. 59A and b). 



A single animal even converted itself into a double shape when 

 the tail-pole was bent and directed forward (c). These examples 

 show again that there is no mysterious unity in the endoplasm and 



