Cuap. III. THE PROCESS OF AGGREGATION. 63 
any aggregation. Nevertheless, before being thus 
affected, they are able, at least in some cases, to excite 
aggregation in their own tentacles. 
That the central glands, if irritated, send centri- 
fugally some influence to the exterior glands, causing 
them to send back a centripetal influence inducing 
aggregation, is perhaps the most interesting fact given 
in this chapter. But the whole process of aggrega- 
tion is in itself a striking phenomenon. Whenever 
the peripheral extremity of a nerve is touched or 
pressed, and a sensation is felt, it is believed that an 
invisible molecular change is sent from one end of the 
nerve to the other; but when a gland of Drosera is 
repeatedly touched or gently pressed, we can actually 
see a molecular change proceeding from the gland 
down the tentacle ; though this change is probably of 
a very different nature from that in a nerve. Finally, 
as so many and such widely different causes excite 
aggregation, it would appear that the living matter 
within the gland-cells is in so unstable a condition 
that almost any disturbance suffices to change its 
molecular nature, as in the case of certain chemical 
compounds. And this change in the glands, whether 
excited directly, or indirectly by a stimulus received 
from other glands, is transmitted from cell to cell, 
causing granules of protoplasm either to be actually 
generated in the previously limpid fluid or to coalesce 
and thus to become visible. 
Supplementary Observations on the Process of Aggre- 
gation in the Roots of Plants. 
It will hereafter be seen that a weak solution o1 the car- 
bonate of ammonia induces aggregation in the cells of the roots 
of Drosera; and this led me to make a few trials on the roots 
of other plants. I dug up in the latter part of October the 
first weed which 1 met with, viz. Zuphordia peplus, being care- 
