AND SODIUM FLAME. 209 



It will be remembered that if an opaque substance is 

 placed over a part of a glass nest, other things being 

 equal, the ants always congregate under it ; and that if 

 substances of different opacity are placed on different 

 parts of a nest, they collect under that which seems to 

 them most opaque. Over one of my nests of Foi'mica 

 fusca, therefore, I placed two pieces of dark-violet glass 

 4 inches by 2 inches ; and over one of them I placed a 

 cell containing a layer of bisulphide of carbon, an inch 

 thick, slightly coloured with iodine. In all these experi- 

 ments, when I moved the liquids or glasses, I gave 

 the advantage, if any, to the one under which experi- 

 ence showed that the ants were least likely to congre- 

 gate. The ants all collected under the glass over 

 which was the bisulphide of carbon. 



I then thought that though no doubt the iodine 

 rendered the bisulphide more completely impervious to 

 the ultra-violet rays, I would try the effect of it when 

 pure and perfectly colourless. I therefore tried the 

 same experiment with pure bisulphide, moving the two 

 u'.tsses from time to time in such a manner that the 

 ants had to pass the first violet glass in order to reach 

 that over which was the bisulphide. 



At 8.30 the ants were all under the glass over which 

 was the bisulphide of carbon : I then changed 

 the position. 



8.45 .. 





 9.15 



