160 Proceedings of the Royal Physical Society. 
blood of some of them, viewed in bulk, is actually milk-white (see next 
section). Orchestww in which a blue tint is visible have bright blue plasma; 
those in which no blue is to be seen, ae. the yellow-brown specimens, have ~ 
yellow plasma. Talitrus, so blanched in colour, has, as already noted by 
Bruntz (10), colourless blood. The external covering of these shore 
amphipods being devoid of chromatophores and being more transparent than 
that of most isopods, the hue of the blood largely determines the colour of the 
animal. It is for this reason that their colour is not subject to reflex 
alteration, but remains fixed in spite of change of eye illumination. 
Even isopods that possess chromatophores may derive part of their colour 
from that of the blood. Thus, some specimens of Jdotea emarginata are 
fundamentally darker than others, and no length of exposure on a white 
background can turn them into light-coloured individuals. While the blood 
of all individuals of this species 1s tinged with brown, that of the permanently 
dark specimens is deep brown. In such a case the underlying colour of the 
blood markedly limits the extent to which the animal can change its shade 
by retraction of chromatophores. Generally speaking, the residual colour 
of any isopod whose chromatophores have retracted to the utmost, is that 
of its blood plasma. It would be of interest to know whether these changes — 
in blood colour are in any way dependent on the state of the isopod as 
regards moult (cf. Tait, 3). 
The colour of the blood 4 is best determined by allowing it to escape from 
the antenne, and examining it against a white or a black background according 
to circumstances; though in many cases it is sufficiently pronounced to be 
seen with the low power of an ordinary microscope, as in making direct 
observations on arrest of hemorrhage from the antenne. To see the colour 
in its full depth a simple plan is to drop the animal, after its antennary 
flagella have been cut across, into a white porcelain dish containing sea- 
or preferably tap-water. Heemorrhage is brisk, the blood pouring out and 
settling to the bottom of the dish in quantity. By heating the dish the 
colour changes produced with rise of temperature can be readily studied. 
Thus the bright blue blood of Orchestre turns at first violet and then reddish- 
brown with heat; yellow blood from the same species remains unchanged 
or becomes brownish. 
It is generally agreed that the actual colour of the blood icpente on ie 
simultaneous presence, in different proportions, of two or three fundamental 
pigments—a blue, a yellow, and a red (see Newbigin, 11, also Winterstein, 12). 
Bauer (8) states that the prevailing greenish tint of Jdotea tricuspidata is due 
to its vegetable food. Whether the blood is greenish he does not say, but: _ 
in any case one would accept only after experimental evidence the conclusion 
