RESEARCHES ON THE DEVELOPMENT OF STARCH-GRAINS. 301 
Other starch-grains which I have studied cannot he 
arranged under one or other of these types, inasmuch as I 
have not been able to make out all the details of their develop- 
ment. In many cases the development of the starch-forming- 
corpuscles was not observed, though it could be inferred 
from their position in the cell. In other cases it was 
impossible to see the first appearance of the starch-grains. 
The following conform to the type of Amomum Carda- 
momum, namely, other Scitaminee such as Thalia setosa, 
Elettaria Cardamomum, Costus Malortieanus ; the Potato, so 
far as observations made on the cortical part of young 
potatoes go (the more central portions were too opaque in 
consequence of the presence of the starch grains); the 
rhizome of Iris florentina, in which the starch-forming-cor- 
puscles have a peculiar granular appearance ; the parenchyma 
of the pith of Philodendron grandifolium. In the following 
cases the same relation between starch-grain and starch- 
forming-corpuscles was observed, although I was unable to 
ascertain the mode of development of the latter; in the 
bulbils of Ficaria ranunculoides ; in the cortical parenchyma 
of the rhizome of various species of Peperomia; in the 
cortical parenchyma of the scales of a Tydea; in the tubers 
of Dioscorea alata ; in the root of Gunnera scabra. 
Silene inflata and Lychnis dioica belong to the Melandryum 
type. 
The observations made on Phajus probably hold good 
with reference to the other allied Orchidaceous plants— 
Acanthephippium,! for instance. 
The other species of Canna resemble Canna gigantea, and 
perhaps Curcuma zedoaria does also; the minuteness and 
indistinctness of the corpuscles in this plant made it im- 
possible to observe them accurately ; in their first formation 
and in their mode of development they resemble those of 
Amomum, and at a later period they appear to become 
elongated like those of Canna. 
3. When we compare the starch-forming-corpuscles with 
other bodies contained in cells their resemblance to chloro- 
phyll-corpuscles at once suggests itself. In their composition 
they appear to be essentially the same as the leucophyll- 
corpuscles? which are found in the more internal cells of 
etiolated stems, and which are quite colourless and very un- 
stable. Further, there is a singular similarity in the mode 
of development; the mode of formation of the starch- 
1 Gris, loc. cit., p. 196. 3 
? I prefer this term (which was suggested by Sachs) to the term “ etiolin- 
corpuscles,” for these corpuscles appear frequently to contain no etiolin. 
