RESEARCHES ON THE DEVELOPMENT OF STARCH-GRAINS, 295 
from it, the function of the corpuscle is indicated by the 
constant connection of the starch-grain with it, and by the 
fact that the portion of the grain which is in contact with it 
is the part which has grown the most rapidly. If the cor- 
puscle were not a starch-former, it would only hinder the 
access to the starch-grain of plastic substances, and in that 
case the growth of the side of the grain in contact with the 
corpuscle would be less than that of the other. 
I shall call these bodies in the following pages “ starch- 
forming-corpuscles” (Starkebildner). 
I will now give a short account of the development of 
these corpuscles and of the starch-grains in a few plants. 
One of the most appropriate objects for observations of this 
kind is the epidermis of the stem and petiole of Phzlo- 
dendron grandifolium (figs. 14, 15). In young cells the 
nucleus, which is either parietal or suspended in the vacuole 
by threads of protoplasm, is seen to be surrounded by a con- 
siderable number of glistening spherical bodies, which much 
resemble the nucleoli. The development of these bodies is 
essentially the following:—The nucleus of a young cell is 
surrounded by a layer of dense protoplasm, which is at first 
uniformly thick, but which subsequently becomes uneven. 
The prominences, which are at first hemispherical, round 
themselves off to form the spherical bodies mentioned above, 
whilst the intermediate substance assumes the properties of 
ordinary protoplasm. This process is probably to be inter- 
preted thus: that a substance which is at first uniformly 
distributed in the protoplasm investing the nucleus separates 
out, and collects around certain centres of attraction. These 
spherical bodies behave with reagents in the manner described 
above; they are the starch-forming-corpuscles. They de- 
velope numerous starch-grains close beneath their surface, 
which remain small, especially in the petiole, and often form 
a hollow sphere surrounding the central portion of the cor- 
puscle. In this central position it appears that no starch- 
grains are formed. The duration of these starch-grains is 
limited ; in the mature thick-walled cells of the stem 
1 The vesicles containing starch (Brutblaschen), which were discovered 
by Nageli (‘ Zeitsch. f. Wiss. Bot.,’ i, S. 149, iii, p. 109), are doubtless the 
same as our starch-forming-corpuscles. Trécul (‘ Ann. d. Se. Nat.,’ sér. 4, 
t. x, ‘ Des formations vesciculaires dans les cellules végetales ”) observed 
and correctly drew these bodies in the endosperm of certain Caryophyllee, 
Chenopodiacere, Graminer, &c. The numerous new observations which 
his paper contains have remained comparatively unknown in consequence 
of the extraordinary theories which he builds upon them, and of remarkable 
errors in matters of fact. 
