RESEARCHES ON THE DEVELOPMENT OF STARCH-GRAINS. 293 
nutrition. This conclusion is further supported by the fact 
that when the starch-grains come into contact with other 
chlorophyll-corpuscles, prominences are developed upon them 
at the points of contact (Peperomia stenocarpa, fig. 8 ; Oxalis 
Origtesii, Dieffenbachia Seguina, fig. 13). 
The different stages in the development of the starch- 
grains, in so far as they are related to the chlorophyll-cor- 
puscles, are essentially the following :—The starch-grains 
which are produced by flattened chlorophyll-corpuscles are 
at first wedge-shaped and flattened in the same planes as 
the chlorophyll-corpuscle ; the side which is connected with 
the corpuscle is truncated, often somewhat concave or 
uneven, whereas the free end is rounded. When the for- 
mation of starch is very considerable, the chlorophyll-cor- 
puscle gradually assumes a nearly isodiametric form ; it 
diminishes in density, and subsequently also in size, until 
only a small residue of it is left, or it may disappear 
entirely. Simultaneously the starch-grain becomes thicker, 
and usually acquires an ovoid form. The growth of the 
starch-grain ceases with the disintegration of the chloro- 
phyll-corpuscle. ‘The process of development can be easily 
observed in Peperomia stenocarpa (figs. 6—8), and in Ozalis 
Origiestz (figs. 9 and 10); for observing the first forma- 
tion the cortex of Philodendron grandifolium (figs. 4 and 
5) may be recommended. ‘The starch-grains which are 
formed in chlorophyll-corpuscles which are not flattened, 
are, so far as can be ascertained from the scanty observa- 
tions, at first hemispherical, the flat surface being in contact 
with the chlorophyll-corpuscle. 
In those chlorophyll-corpuscles which are capable of 
producing starch in all parts of their substance, the starch- 
grains may of course be formed near the surface and may 
sooner or later project freely. In this case, which is by no 
means rare in the mesophyll (e.g. Tradescantia, Begonia, 
&c.), the starch-grains must be excentric. I have, however, 
not succeeded as yet in finding grains of this kind exhibit- 
ing evident differentiation. 
2. The observation of fresh sections, not too thin, of parts 
of plants which do not contain chlorophyll, shows that the 
starch-grains which are in process of development are not 
surrounded by ordinary protoplasm, but that they are con- 
ained in, or attached to, peculiar refrangible corpuscles 
which are usually spherical or spindle-shaped. These 
bodies are very unstable. So soon as the surrounding fluid 
penetrates into the cell they swell up considerably and 
then dissolve. Close observation has shown that prolonged 
