PEECEPTION OF ULTEA-VIOLET KAYS. 217 



In this case tlie preference for ultra-violet over dark 

 was very marked. 



May 18. — I again tried them with the ultra-violet 

 rays, using three divisions — namely, further ultra-violet, 

 ultra-violet, and dark. The numbers were as follows, 

 viz. under the — 



Exp. 1 

 „ 2 



105 



To my eye there was no perceptible difference be- 

 tween the further ultra-violet and the ultra-violet 

 portion ; but slightly undiffused light reached the two 

 extreme divisions. It may he asked why the still 

 further ultra-violet division should have been entirely 

 deserted, while in each case two or three Daphnias were 

 in the darkened one. This, I doubt not, was due to the 

 fact that, the darkened division being next to the ultra- 

 violet, one or two in each case straggled into it. 



Not satisfied with this, I tried another test. There are 

 some liquids which, though transparent to the rays 

 we see, are quite opaque to the ultra-violet rays. 

 Bisulphide of carbon, for instance, is quite colourless 

 and transparent : it looks just like water, but it entirely 

 cuts off the ultra-violet rays. If, then, we place the 

 trough containing Daphnias, as I had previously done 

 my nest of ants, in the ultra-violet part of the spectrum, 

 and then place over one half of it a flat bottle contain- 

 ino- water, and Over the other half a similar bottle con- 

 taining bisulphide of carbon, botli halves will seem 

 equally dark to us, but the ultra-violet rays reach one 

 half of the vessel, while they are cut off from the other. 



