ACTIVITIES OF COLONIAL ANIMALS 301 
yellowish crystalline material. If finely divided carmine in sea- 
water is flooded over the surface of an expanded Renilla, sooner 
or later much of it will be found to have collected at the pores of 
siphonozodids. And if such a preparation is closely watched 
under a hand lens, very small pieces of carmine will often be seen 
darting into the siphonozoéid pores. If a piece of the superior 
surface of the rachis is separated from the rest of the colony by 
a cut approximately parallel to that surface, and if the spaces 
connected with the siphonozodids and exposed on its under face 
are closely watched, small carmine particles are seen to shoot 
out from these spaces and into the surrounding sea-water. As 
such preparations show, the siphonozodids ordinarily conduct 
water from the exterior to the interior and such currents are gen- 
erated within the siphonozo6id itself. The currents are without 
doubt the results of ciliary action. 
Not infrequently siphonozodéids will be found in which no cur- 
rents can be demonstrated. If these are watched for a time, cur- 
rents will sooner or later be seen in them, the whole condition 
recalling strikingly that of the lateral pores of sponges whose 
currents, though produced by the flagellated cells, are controlled 
by the sphincters or other like devices which close and open these 
pores. In my opinion, the siphonozoéid currents in Renilla are 
controlled by some such device. 
If methylen blue in sea-water is injected under pressure into 
the internal spaces of a living Renilla, small amounts of this 
fluid will be seen oozing from a few of the siphonozodids. If the 
peduncle of a Renilla is cut open and a vertical glass tube tied 
securely into the cut, the end of the tube and the animal being 
under sea-water, it will be found that a pressure of some 5 or 6 
cm. of water is necessary to drive a methylen-blue solution 
through the siphonozoéid pores to the exterior. This pressure 
may be taken, therefore, as the pressure necessary to counteract 
that by which the siphonozoéid currents are produced. The 
escape of methylen-blue solution thus brought about at the siphon- 
ozodids is never very general and never large in amount. It in- 
dicates that the siphonozoéid pores are not provided with any 
effective valve-like parts whereby the currents are limited in 
THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 31, No. 3 
