CARBONATE OF AMMONTA ON CHLOROPHYLL-BODIES. 271 
most of the cells containing these spheres not a grain of chloro- 
phyll could be distinguished. That the green surrounding zones 
had been derived from the chlorophyll-grains is, I think, certain ; 
for the purple colour of the central spheres showed that the cell- 
contents had not been originally green. Other cells in these 
same sections included irregularly-shaped masses of a purplish- 
green colour; and these were observed slowly to change their 
formsin the usual manner. When acetic acid was added to them, 
the green transparent spheres and the zones of similar green 
matter round the purple spheres instantly disappeared, either 
from being dissolved or, as seems more probable, from being 
killed and suddenly disintegrating. On another occasion boiling 
water and alcohol produced the same effect on the spheres. Ten- 
tacles still retaining their ehlorophyll-grains, but with many very 
pale-coloured homogeneous aggregated masses (which were seen 
in movement), were irrigated with acetic acid ; and it was eurious 
to observe how instantaneously they became filled with small 
transparent spheres. Ina short time, however, the outlines of the 
larger masses were alone left; then these disappeared, and finally 
the small enclosed spheres. On the other hand, some dark- 
coloured solid aggregated spherical masses did not disappear 
When left for 24 hours in acetic acid. 
The effect of the ammonia solution (4 and 7 to 1000) on the 
epidermal cells of the upper surface of the disk was now more 
especially observed. In some cases all these cells which, as 
already stated, invariably contain many chlorophyll-grains, in- 
cluded after immersion in the solution only a single or several 
green transparent spheres; but more commonly the spheres 
were very dark purple or brown. Sometimes a central sphere, 
which was so solid that it could be cracked, was surrounded by a 
well-defined paler zone. Numerous gradations could be traced, 
showing that several small spheres and irregularly shaped glo- 
bules often coalesce, and thus form the larger rounded masses. 
It was repeatedly observed that when the epidermal cells con- 
tained only one or two large spheres, not a single grain of chlo- 
rophyll eould be seen. It is surprising that dark purple or 
brown or almost black spheres should be formed in the epidermal 
cells of green leaves; for before immersion the cell-contents 
were colourless, with the exception of the chlorophyll-grains ; 
but the fact is less surprising when it is known that these cells 
turn more or less red as they grow old if they are exposed to a 
