224 
F. E. Fritsch. 
are formed in a cell. The same author has also observed the 
liberation of gonidia from the heterocysts of Nostoc commune 1 and 
N. microscopicum,- and subsequently adds that similar structures 
also develop from ordinary vegetative cells in these species; 3 they 
are stated to differ in no way, from those derived from the 
heterocysts, but Brand was unable to observe their mode of origin, 
and bases his conclusions on the fact, that he met with short rows 
of gonidia, which were in direct contact with ordinary vegetative 
cells on each side. 1 He concludes from analogy however that, as in 
other cases, they are formed by rejuvenescence and are liberated 
by the membrane of the original cell being cast off. 
In some material of Anabaena. which had been separated from 
the rest and not examined for some months, processes, recalling 
gonidia-formation, as described by Brand, had set in. The cells of 
the filaments differed in shape from the normal one; they were 
generally more or less spherical with somewhat flattened ends, and 
differed from the usual elongate-elliptical shape (Fig. 29). Numerous 
heterocysts were however present, so that this appears to be a 
constant feature of the species. In some of the filaments the cells 
midway between two heterocysts were slightly larger than those 
immediately adjacent to the latter (cf. Fig 29), but there were no 
true indications of typical spores. The detailed structure of the 
investment, as I have described it elsewhere and briefly above, 
was remarkably clear, and it seems possible that some of the cells 
were in a sporogenous condition ; however, the cells were arranged 
in coherent rows and were never separated from one another as is 
the case in mature spores ; nor did the cell-sheath in most cases 
seem to have extended round the ends of the mother-cells of the 
gonidia. In many of these cells (both in the enlarged and the 
ordinary ones) the contents were markedly rounded off and con¬ 
tracted, and could in most cases be seen to have acquired a new 
membrane of their own (frequently exhibiting two complete invest¬ 
ments) inside that of the mother-cell (Figs. 27 and 29). Large 
numbers of cases were found in which the cell-membrane had 
1 Brand, Bemerkungen iib. Grenzzellen u. lib. spontan rothe 
Inhaltskorper d. Cyanophyceen. Ber. Deut. Bot. Ges. Vol. 19. 
1901. p. 154, et seq. 
2 Brand, Morphol.-physiol. Betrachtungen, etc. Loc. cit. p. 48. 
3 Loc. cit., p. 49. 
4 I do not quite gather what were Brand’s actual reasons for 
regarding these cells as gonidia, and it seems just possible 
that they may be the results of spore-germination according 
to the second of the above types. 
