94 
F. E. Frit sell. 
Great attention should be paid to those cases, in which spore- 
formation takes place centripetally (Sphaerozyga); in these cases 
one would imagine, that the above function would be taken over by 
the intercalary heterocysts. That raises the question, why does 
spore-formation in one case commence immediately next to the 
heterocyst and in another case remote from it? This indicates 
some influence on the part of the heterocyst, which is as yet quite 
unknown to us. It is not a point either, which is inherent in the 
plant; although spore-formation was almost invariably centrifugal 
in the species under discussion, I met with a few isolated examples 
of the reverse (cf. Palla ’93, p. 545 for Sphaerozyga). 
Until w r e have further evidence of the ultimate fate of the 
