EXPEEIliENTS WITH COLOURED GLASSES. 



189 



Adding these numbers together, there were, in the 

 twelve observations, under the red 890, under the 

 green 544, under the yellow 495, and under the violet 

 only 5. The difference between the red and the green 

 is very striking, and would doubtless have been more 

 so, but for the fact that when the colours were trans- 

 posed the ants which had collected under the red 

 sometimes remained quiet, as, for instance, in cases 

 7 and 8. Again, the difference between the green and 

 yellow would have been still more marked but for the 

 fact that the yellow always occupied the position last 

 held by the red, while, on the other hand, the green 

 bad some advantage in coming next the violet. In 

 considering the difference between the yellow and 

 green, we must remember also that the green was 

 d3cidedly more opaque than the yellow. 



The case of the violet glass is more marked and 

 more interesting. To our eyes the violet was as opaque 

 as the red, more so than the green, and much more so 

 than the yellow. Yet, as the numbers show, the ants 



