HIPPOLYTB VAEIANS. 619 



unchanged^ and E had gained transparency and a slight 

 green tint in the middle of the body, but was otherwise un- 

 altered. Examination showed that in the two (G and H) 

 which were green the effect was due to retraction of the red 

 and extrusion of the blue, which with the yellow already out 

 gave the green tint. These two (Gr and H) were now put 

 into the jar standing on the black plate, and at 4.50 and 7.5 p.m. 

 were full dark brown ; while E and F were transferred to 

 the porcelain dish, and at 4.50 were greenish brown. 



Thus we see that alterations of light intensity and of the 

 distribution of light act as powerful stimuli to colour-changes; 

 that light of low intensity scattered evenly from the surface 

 of the vessel produces a retraction of the red pigment and an 

 evolution of the blue (possibly also of the yellow) — in other 

 words, a green colour — during the hours of daylight; that 

 light of low intensity, either absorbed by the walls of the 

 vessel or unequally scattered over its interior, effects the full 

 expansion of the red pigment, and to a less extent extrusion 

 of the yellow. 



Hippolyte avoid high light intensity by creeping under 

 the shadow of the dish. It is noteworthy that when, how- 

 ever, they are compelled to submit to it — placed, for example, 

 in a porcelain dish in the open — the high light intensity pro- 

 duces a green effect; whilst if the prawns be similarly ex- 

 posed, but in a dish with a background which absorbs the 

 light, the red is maintained fully expanded, or retracts but 

 gradually. 



Effect of Monochromatic Light. — Having ascer- 

 tained that the distribution of a pigment in the ^^chromato- 

 phores" of Hippolyte varians is profoundly influenced by 

 the quantity of light, we next endeavoured to determine 

 whether the quality of the light to which the prawns were 

 exposed exerted any similar effect. For this purpose we 

 employed the Landolt " Strahlenfilter" already described. 

 As sources of illumination we used, in some cases, diffuse 

 light, in others incandescent gas-light, and again in others 

 bright sunlight. The colour-screens do not transmit abso- 



