GROWTH AND DIVISION 73 



Stage, indicate that clumping is not for the purpose of dividing the 

 nucleus equally. Hence Johnson suggested that both the clumping 

 and the preliminary constriction are a recapitulation of phylogeny, 

 harking back to the form in fission of the nucleus in less specialized 

 protozoa. Nor is there any evidence of macronuclear reorganization 

 occurring during clumping; for Johnson found that the character 

 of the macronuclear matrix remained unchanged and he stated 

 explicitly that there was no indication of linear arrangement of 

 threads or the formation of something like chromosomes. 



At the earliest, a new contractile vacuole for the opisthe makes 

 its appearance in the proper location at stage 4. Its formation is 

 therefore probably not initiated by the division furrow, which is 

 not yet visible, though Weisz (1951b) found that cutting the stripes 

 of non-dividing stentors transversely would induce the temporary 

 formation of a posterior vacuole and I have confirmed this. 



As the division line cuts around its upper end, the primordium 

 can bend more sharply and move backwards into the future 

 opisthe. Schwartz described how the anlage shifts with reference 

 to the striping on the left side so that these lines, at first parallel, 

 come to lie at right angles to the new membranellar band. The cut 

 ectoplasm of the future proter, or anterior daughter, closes together 

 immediately as the primordium migrates posteriorly, with the 

 result that a herringbone-pattern of stripes is formed which is 

 somewhat asymmetrical because the furrow ran more sharply 

 posteriorly on the left side. Anterior can be distinguished from 

 posterior daughters long after separation because this pattern may 

 persist for three days afterward in starved animals. Gradually the 

 abbreviated stripes grow posteriorly to reproduce the typical 

 ramifying zone. 



In the meantime the anterior portion of the opisthe bearing the 

 primordium has been bulging outward while gullet and oral pouch 

 have been forming and shifting forward. The clumped macro- 

 nucleus then elongates parallel to the main axis to form a long 

 rod, which begins to nodulate simultaneously at both ends. As 

 constriction continues the stripes of the proter are drawn together 

 to form its tail and possibly extend in length as they narrow to a 

 point. The half-nodulated macronucleus now divides in two; 

 Johnson thought that its division is autonomous and the same is 

 implied in Causin's (1931) report that even in regeneration the 



