PRIMORDIUM DEVELOPMENT 171 



turning back, which is a rather universal generalization; yet in 

 the development of the relatively new science of biology it was 

 very important to find that the elaboration of the organism is not 

 magical and immutable but is a process which occurs in time and 

 is subject to some analysis by operative manipulations. In many 

 respects it can be shown that this principle also appHes to Stentor. 



First we shall discuss deletion experiments on the oral primor- 

 dium. Minimal excisions of parts of the oral primordium, always 

 necessarily including some of the surrounding ectoplasm and 

 endoplasm, were performed on regenerating stentors in stages 4 

 and 5 when the membranellar band is already well formed but there 

 is still no visible indication whatever of developing mouthparts. 

 When the anterior halves of such anlagen were removed, develop- 

 ment continued and complete mouthparts were formed but the 

 membranellar band was only half its normal length (Fig. 42A). 

 Re-regeneration then occurred about a day later to produce a set 

 of feeding organelles in normal proportion to the size of the cell. 

 If the posterior half or third of the primordium was excised, only a 

 considerable length of membranellar band was produced and no 

 mouthparts at all, with re-regeneration now following sooner (b). 

 Extensive removal of all but the posterior end of the anlage could 

 lead to formation of a perfect set of mouthparts without any 

 membranellar band at all (c) ; and removal of only the posterior 

 tip, at stage 5 or early stage 6, resulted in complete absence of the 

 gullet although oral pouch and membranellar band were normal (d). 

 Finally, by removing a penultimate section of the anlage at stage 5 

 it w^as possible to produce heads in which a normal gullet termin- 

 ated the membranellar band but the oral pouch was missing 

 entirely (e). Therefore, it is clear that by stage 4 the oral anlage is 

 determined and any ablation of its parts results in corresponding 

 deletions in the organelles formed. 



The same operations were then performed on early primordia. 

 Even when extensive sections of the anlage were removed, the lack 

 was then compensated by additions to the primordium, increasing 

 it to its normal length, and conlplete, proportionate feeding 

 organelles w^ere produced (f). At stage 3 some specimens showed 

 corresponding defects while others did not, so that it may be 

 concluded that fixity or determination of the primordium occurs 

 in late stage 3. 



