C)2 
P. E. Fritsch. 
p. 39). If we further take into consideration the characteristic 
differences in the structure of the heterocyst, according as it is 
intercalary or terminal in position, we are, I think, compelled to 
attribute more than a mere limiting function to these cells. It is 
also difficult to conceive of an explanation for the occurrence of a 
number of heterocysts side by side (as also obtains in many species 
of Tolypothrix ), J if their function is solely to split up the thread 
into a number of vegetative regions. 
I have shown, that early stages in the development of a 
heterocyst are characterised by the presence of a bright granule on 
the wall, adjacent to the vegetative cells, and that a similar granule 
is often to be seen on the adjoining wall of the neighbouring vege¬ 
tative cell. In appearance and in their behaviour towards Eosin 
these granules are quite identical to the so-called cyanophycin- 
granules, which almost always occur in considerable numbers in the 
ordinary cells and spores. 2 In later stages these granules are no 
longer present, whilst granules, which stain very faintly with Eosin, 
are frequently to be seen amongst the contents of the heterocysts 
and more rarely a deeply-stained granule occurs. It seems to me 
therefore, as though these cyanophycin-granules wandered over into 
the young heterocyst from the neighbouring vegetative cells, which 
in later stages often appear collapsed or devoid of the granules. 
Inside the heterocyst the granular substance becomes modified in 
some way, so that it loses its former properties. Probably the 
change is of the nature of the formation of a reserve-substance, 
for when the richly-granular spores of this Alga are placed under 
unfavourable conditions, necessitating a period of rest, the granules 
also gradually disappear and the contents become more or less 
homogeneous. The changed granules are scarcely affected by Eosin, 
whereas they are now capable of assuming a green colour with 
methyl green. 3 With regard to the differences between the young 
heterocysts and the ordinary cells in their behaviour towards 
Iodine, the brown colouration of the vegetative cells is probably 
due to the staining of the blue colouring-matter; this has evidently 
already undergone some change in the young heterocysts and in the 
1 During the wiuter of 1902-1903 a species of Tolypothrix with 
sometimes as many as six contiguous heterocysts was 
abundant in the Pen Ponds in Richmond Park. I was 
unable at the time to examine this species in greater detail. 
2 cf. also Pal la C93), pp. 544, 546. 
3 This need not only be due to a change in the contents, but 
also to a change in the permeability of the cell-membrane. 
Since methyl-green more or less dissolves or corrugates 
this latter, it is probably able to penetrate to the contents 
far more readily than Eosin. 
