POLARITY 203 



supplied sketches from my own experiments (including Fig. 44), 

 which may therefore be regarded as generally confirmatory. 



Fig. 56. Observations regarding induction of stomatogenesis by 

 the posterior end. 



A. In an oral half of a stage-3 divider the anlage was not 

 resorbed but extended all the way to the posterior pole and 

 produced no mouthparts, presumably because inducing region 



is anterior to the pole itself. 



B. Two extra tail poles engrafted led to multiple stomato- 



genesis, with complete but ectopic gullet (x). 



C. Specimen with two tail poles due to shift of primordium 

 site produced a stentor with unusually large mouthparts, 



possibly due to the double tail. 



Primordia far from the posterior end produce no mouthparts 

 (Figs. 47D, and 26c). That the oral-inductive gradient stops short 

 of the posterior pole is indicated in Fig. 5 6a, showing incomplete 

 oral differentiation in a primordium extending too far posteriorly. 

 It is possible, also, that inductive action may be compounded by 

 the presence of multiple posterior ends. When three tails were 

 grafted, double mouthparts were produced in the host (b), and in 

 another case three posterior poles may have been responsible for 

 unusually large mouthparts formed (c). Astomatous oral differen- 

 tiation in large fusion masses may be due to the mutual canceling 

 of oral induction gradients in these random grafts. A similar initial 

 astomatous development in isolated sectors bearing division pri- 

 mordia (Tartar, 1958c) may likewise have been due to the frag- 

 ments at first containing insufficient polar regions, a situation later 

 corrected by regeneration of the posterior pole. 



