170 THE BIOLOGY OF STENTOR 



and the anterior half of the stentor rotated 180° on the posterior, 

 there was no extension of the primordium from the cut end and 

 very short membranellar bands were produced (h). This suggests 

 that increase in length occurs in all parts of the primordium and 

 not merely at its ends. 



When the stripe pattern is abnormal after certain grafting 

 operations V-shaped and even looped primordia may be formed 

 (Fig. 41, I and j). The former can produce mouthparts of normal 

 appearance, and the latter make an attempt to do so, though the 

 conditions for invaginations are certainly quite atypical. 



These and other types of oral development show that cyto- 

 differentiation is not so delicately precise a process that inter- 

 ferences cannot be surmounted. They also demonstrate the 

 important point made by Driesch that the same result in organic 

 development can be achieved by several routes, even exceeding the 

 usual experience of the organism. 



3. Determination, or the progressive specification of the 

 oral anlage 



Only early-stage anlagen are resorbable (Weisz, 1956; Tartar, 

 1958c). Stage 3 appears to be the time of transition, after which 

 oral primordia become self developing systems resistant to resorp- 

 tive influences, though some early stage-4 anlagen, not actually 

 resorbed, developed astomatously or with incomplete mouthparts. 

 Weisz cut off the tails of late dividers including the posterior end 

 of the primordium and found that the anlage then produced no 

 mouthparts. Similarly, Lund (19 17), working with Bursaria 

 truncatella, stated that minor injury to the anlage of the gullet did 

 not prevent development of the oral primordium but led to 

 abnormality in the oral structures produced. He concluded that 

 " there appears to exist in the anlage of the gullet a definite part 

 which corresponds to a definite structure in the fully differentiated 

 gullet ". 



These isolated indications led to a comprehensive demonstration 

 on the cell level of something very much like the determination 

 of parts in developing embryos (Tartar, 1957c). It is well known 

 that embryonic anlagen are at first modifiable but later not. This 

 developmental principle seems to be simply a statement that once 

 a complex is well underway it cannot be modified and there is no 



