POLARITY 201 



reversal is in folded aboral halves in which the striping bends and 

 breaks to form a new holdfast (see Fig. 27), for in these cases both 

 posterior and anterior ends of the severed striping meet at the new 

 posterior pole. Yet such specimens will have to be followed very 

 closely to determine whether the stripes in reversed orientation 

 are not subsequently resorbed. 



2. Rate of regeneration in relation to the polar axis 



A further manifestation of polarity is of course to be found in 

 the fact that heads are always regenerated at anterior ends of 

 fragments and tails at the posterior. As in regeneration of metazoa, 

 the organism can produce either anterior or posterior structures 

 from almost any level of the body, the choice depending on the 

 original polarity. Although coelenterates and turbellarians may 

 produce heteropolar heads on very short pieces, this does not occur 

 in stentors. Isolated heads do not become heteromorphic (see 

 Fig. 32) nor do disc-shaped fragments whose longitudinal dimen- 

 sion is brief (Fig. 25A). Nevertheless, it is conceivable that the rate 

 of oral regeneration might vary with level of cut, as in flatworms 

 (Bronsted, 1955). That this is not the case within a single cell such 

 as Stentor was first shown by Gruber (1885b) who found that 

 animals, minus the head only, regenerated a new set of feeding 

 organelles as rapidly as posterior fragments. This was confirmed 

 by Weisz (1948a), who also found (1948c) that the relative growth 

 rates of fragments from any region were the same. Contrary to the 

 experience with stentors, Sokoloff (19 13) reported that middle 

 pieces of Spirostomum regenerate faster than the ends but I think 

 this work requires checking. 



Although denying an axial gradient in speed of regeneration, 

 Weisz (1948a) stated that oral regeneration is hastened by the 

 presence of a holdfast, mid-pieces regenerating more slowly than 

 posterior fragments. Child (1949) regarded this difference as 

 probably incidental to the fact that middle fragments have to 

 accomplish two regenerations, of both head and tail. This could be 

 tested in other ways, as I have done (unpublished). If the presence 

 of a holdfast hastens oral regeneration, then stentors from which 

 the head only is excised should regenerate sooner than animals 

 from which head and tail are removed, but they do not. On Weisz's 

 assumption, the former should also regenerate in the same time as 



