12 W. BLAXLAND BENHAM. 
element has been described in the fluid contained in the blood- 
vessels of any Annelid. But a few worms are known in which 
the blood-vessels are absent, whilst the fluid in the celom is 
coloured ; and in these cases (Glycera, Capitellide, Poly- 
cirrus) the hemoglobin exists within the corpuscles, the plasma 
of the coelomic fluid being colourless. The coloured corpuscles 
of this fluid in Glycera and Capitellide are, however, 
entirely unlike the coloured globules in the blood of Magelona. 
In the case of Capitella I took the opportunity of examin- 
ing the corpuscles during my stay at St. Andrews, as I was 
able to get abundant material. My observations are not new; 
Claparéde and Eisig have already sufficiently demonstrated 
the character and constitution of these corpuscles. Lankester! 
was the first to show by spectroscopic examination that they 
contain hemoglobin. The corpuscles when drawn fresh from 
the worm are circular, somewhat flattened discs, yellowish red in 
colour. A distinct membrane is visible, and granules of various 
small sizes are present. Stained with picro-carmine, a small 
round nucleus is seen in each corpuscle. When fresh cor- 
puscles are treated with salt solution no effect is noticeable. 
On the addition of distilled water the hemoglobin is dissolved 
out and the nucleus is rendered distinct, and the corpuscle 
itself becomes slightly smaller. This is no doubt due to os- 
mosis, the water passing into the corpuscle and causing it to 
become spherical: it thus loses in diameter in one direction as 
it gains in the other,—a process exactly like that occurring 
when Vertebrate corpuscles are similarly treated. 
Prolonged action of water causes the outlime of the cor- 
puscles to become less and less distinct, till it is almost 
impossible to recognise it, the nucleus and refringent bodies 
alone remaining. This reaction is quite different from what 
we have seen to occur in the caseof Mageiona, but resembles 
the corresponding reaction with human blood. 
Chloroform dissolves the hemoglobin, the corpuscles become 
granular, and the outline becomes gradually less distinct, but 
the membrane remains for a long time. But there is no 
1 * Proc. Roy, Soc.,’ xxi, 1873, p. 70. 
