122 



THE BIOLOGY OF STENTOR 



is impossible with such units. Failure of smallest pieces to re- 

 generate would then be due neither to pathological changes nor 

 to insufficiency of material but rather to structural incompatibility 

 between the size of the parts and what is to be made from them. 

 Tiny fragments can produce some oral cilia and membranelles but 

 it may well be that there is a jamming when these parts attempt to 

 coil tightly inward to produce a gullet. 



Fig. 29. Regenerated *S. coeruleiis of near minimum size, a: 

 Tiny and large stentors drawn to same scale. Note that mem- 

 branelles are of same width and length in both. Pigment stripes 

 are also of similar widths, hence minute form had only about 20 

 as compared with 100 for the large animal, b: Enlarged view 

 of regenerant, which has but one macronuclear node and very 

 few stripes in the frontal field. 



If units of ectoplasmic structure in ciliates, such as oral cilia, 

 body cilia, and trichocysts, are of a standard, nearly invariant size 

 for any species of ciliate (cf. Bonner, 1954; Ehret and Powers, 

 1959) this should simplify the problems of growth; for one would 

 then need only to explain their increase in number, and further 

 hypothesis regarding their adaptive size would not be necessary. 

 This seems to be one of the crucial theoretical points involved in 

 these small-fragment studies. The other resides in the amazing 

 fact that organic form is largely independent of size and, outside 

 the limitation just mentioned, it is possible for stentor shapes and 



