996 JULIAN §. HUXLEY 
necessary to squirt them free. On examining them I| at first 
thought that they might be derived from the Sycon restitution- 
bodies, but dismissed the idea as improbable. Later in the 
same day I took some of them to Miss Lebour, Naturalist at 
the Laboratory, to see if she could identify them. On examining 
them under pressure with a high power, it was found that they 
contained fragments of spicules. Thus the suspicion that they 
were of sponge origin was strengthened. 
Two days later a restitution-body which had for four days 
been isolated for other purposes on a slide in a moist chamber 
was examined and was found to have subdivided into six 
spherules (fig. 2, a). Thus their sponge origin was conclusively 
proved. Meanwhile the original dish was picked over, some 
of its contents preserved, and the remainder separated into 
divided and normal undivided masses. The normal masses 
were examined two days later (the tenth day of the whole 
experiment) and found to be still undivided, many with 
active flagella and protruded collars still visible externally. 
On the thirteenth day, eight out of fourteen masses were still 
single, but the remaiming six had subdivided. They were 
similar in every way to those observed on the eighth day, 
except that they were not so closely packed, and that I could 
see no traces of a gelatinous membrane round the spherules. 
It would, however, of course be expected that those which 
subdivided earlier would be of slightly different composition 
from these later-divided ones. 
A detailed observation of one of the earlier divided masses 
on the ninth day (fig. 1, a) showed that the spherules were 
tightly packed and mutually compressed. The whole body 
was surrounded by a faint gelatinous membrane, which 
apparently caused the whole to adhere to the glass. Under 
a higher power (fig. 1, b) the single spherules were seen to 
consist of a one-layered epithelium surrounding a central mass. 
The epithelium was composed of extremely clear cells, with a few 
minute granules ; the central mass did not touch the epithelium 
at all points, and was dense and of a yellowish colour; cell 
outlines were not visible in it. The single spherules did not 
