The Comatulids of Torres Strait. 109 
selected specimens did not differ, so far as could be seen, in the manner 
of the response, but only in the rapidity. Handling individual comat- 
ulids had no apparent subsequent effect on the specimen. 
Fatigue seems to be, as a rule, easily induced, but in no other respect 
is there so much individual diversity as in this. Whether the fatigue 
noted was real or only apparent, there was no means of determining. 
Nothing was noted with reference to breeding habits. Dr. Tennent 
found mature spermatozoa in many males of Comatula purpurea, but 
mature eggs were not noted in any species. Apparently the breeding 
season was just opening when we came away, the last of October. The 
shortness of our stay prevented any observations on regeneration or 
growth. In the following pages, under the heads locomotion, food and 
feeding, response to light, response to heat, and response to other stimuli, 
the facts observed are set forth. 
LOCOMOTION. 
To speak of the methods of locomotion in crinoids, or even in comat- 
ulids, is like speaking of locomotion among birds, in that there is as 
much difference between different kinds of comatulids as between 
different kinds of birds, and as birds may either fly or run preponderat- 
ingly, so comatulids may either commonly swim or creep. So far as the 
species studied at Maér are concerned, the Comasteride are creepers, 
while the other families represented there are swimmers. It was a 
surprise and disappointment to discover that none of the common 
comatulids would swim, even in the deep water by the live-car, so 
that it was not possible to make the observations and experiments 
with reference to the swimming of comatulids which I had planned. 
When any of the species of the Comasteride were placed in the water 
beside the live-car, they invariably sank to the bottom (about 20 feet). 
In the case of Comatella maculata, it was noted that, as arule, the 
individual would close the arms orally and vertically over the disk, 
so that it sank almost like a stone. As soon as it touched bottom, 
the arms were opened out and the animal began to creep. In the other 
species, however, the arms were not closed up and so the comatulid 
floated downward to the bottom much more gradually. With one 
exception, all efforts to induce swimming movements of the arms in a 
comasterid completely failed. 
In no case when the individual was on the bottom did mechanical 
stimulation result in any attempt to swim or in accelerated movements 
of any kind. Suspending specimens in the water by means of a slender 
thread gave no results; the arms moved about slowly and with no 
codrdinated effort and with no resulting locomotion. Several indi- 
viduals were provided with cork floats, so attached as not to interfere 
with arm-movements; but, with the one exception already referred to, 
none of them made any effort to swim. The one exception was a 
