ACTIVITIES OF COLONIAL ANIMALS 353 
under water. If an inflated Renilla in a basin of sea-water is 
touched slightly or the whole basin jarred, contraction begins, 
the membranous edges of the axial siphonozoéid withdraw, leav- 
ing the aperture freely open, through which emerges a jet of 
water strong enough to ruffle the surface of that in the basin. 
If a number of expanded Renillas are watched in a shallow aqua- 
rium, one or other of them will be seen to contract from time to 
time—an operation that is invariably accompanied by the open- 
ing of the pore of the axial siphonozodid and the free discharge 
of water through it. 
The internal pressure necessary to bring about the opening of 
this pore may be tested in the following way: if a vertical glass 
tube be tied securely into the superior canal of the peduncle of a 
submerged Renilla—a canal that leads directly to the inner end 
of the axial siphonozodid—and the tube be filled with a solution 
of methylen blue in sea-water, it can be shown that the axial 
pore will not open till the internal pressure has reached 20 to 25 
em. of water. Since this canal communicates not only with the 
axial pore, but also with the pores of the lateral siphonozodéids, 
this experiment is always accompanied with the oozing of the 
methylen-blue solution at these pores, but even under the high 
pressure used this is never profuse, and it is very probable that 
the recorded pressures of from 20 to 25 em. of water represent the 
real pressures necessary to open the axial pore. It is thus clear 
that a much higher pressure is needed to open this pore than to 
reverse the currents of the lateral siphonozod6ids, but it is also clear, 
since the volume of the colony does not begin to show an appreci- 
able diminution till the axial pore is opened, that the lateral 
siphonozoéids are of no great significance in discharging water 
from the colony as a whole. This conclusion is supported by the 
observation that when a Renilla begins to contract, methylen- 
blue solutions or carmine discharged over the superior face of its 
rachis give no evidence of outward currents from the pores of 
the lateral siphonozoéids or from the mouths of the autozodéid. 
If, under such circumstances, water does escape through these 
apertures, it must be very small in amount. 
