THE XEMATOCYS'TS OF EOLIDS 
123 
vided with two pairs of barbs, of which the stouter is about three 
times the length of the other. The third type of nematocyst 
(fig. 3, E) is the most common. In this the capsule is ovoid, and 
the discharged filament presents distinct regions. At the place 
where evagination occurs there is a bare projection crowned by 
four stout barbs with points directed toward the capsule. These 
four barbs originate in a crease between the barren mound and 
a considerabl}^ smaller distal swelling with many minute barbules 
pointing, like the large ones, toward the capsule. From the tip 
of this barbulated swelling the undifferentiated portion of the 
thread arises. The thread itself, however, is not without dis- 
tinguishing characteristics; it is about twice the thickness of the 
threads of the first and second type, and about one-fifth as long; 
it is devoid of barbules and ends in an excessively fine point. 
The cnidophores of eolids which prey on Tubularia contain 
nematocysts which in their discharged state agree point for point 
with one or the other of the three types found in the hydroid. 
That these three types represent developmental stages of only one 
kind of nettle seems unlikely, but even if they do, the argument 
from identity remains unchanged. 
Indigestibility of nematocysts 
If the source of eolidian nettles is coelenterate food, it follows 
that the latter is not completely digestible. In a paper on the 
physiolog}' of nematocysts ('09), I described certain methods of 
isolation whose success depends on the immunity of the capsules 
to at least two forms of digestion, peptic and putrefactive. In or- 
der to discover whether the nematocysts of eolids could be treated 
without destruction in the same manner as those of Hydra, 
Metridium, and Physaha, I subjected a large number of the cerata 
of Montagua to peptic digestion. The experiment gave a posi- 
tive answer. Without immunit}^ to digestive enzymes eolidian 
nettles could not be derived from ccelenterates, but the fact that 
they are indigestible is no argument in favor of derivation, since 
in the opposite case they would probably be equally resistant. 
