616 PROCEEDINGS OF THE ACADEMY OF [Sept., 



other half of the cell with a dark-violet pane transmitting, as 

 shown by the spectroscope, all the colors of the spectrum. I could 

 scarcely discern the ants through the violet pane. But during a 

 residence of several weeks in this cell the ants, after two removals 

 from the colorless to the violet side, remained for many consecutive 

 days under the transparent glass, and I repeatedly interchanged the 

 two panes without causing the ants to remove from under the one 

 or the other. 



The same indifference to the intensity of illumination was shown 

 by ants placed in cells covered one-half with a double layer, the 

 other half with a single layer of green glass; one-half with a 

 double layer, the other half with a single layer of yellow glass ; or 

 one-half with a double layer, the other half with a single layer of 

 blue gljass. The relative proportions of the primary colors, from 

 red to indigo, transmitted by the various colored panes varied 

 greatly, as the spectroscope showed. But the ants apparently 

 disregarded these proportions, and were hasty or deliberate in their 

 action in direct ratio to the number of rays from the violet end of 

 the spectrum. 



It is probable that ultra-violet rays were transmitted with the 

 violet ones in all of my experiments. 



It must be constantly borne in mind that the ants are instinc- 

 tively photophob. They always take shelter under an opaque 

 covering when a choice is offered between darkness and any rays of 

 light, even pure red. 



