LIFE HISTORY OF CARPENTER ANT. 201 



placed the red plates back on the nest and the following day 

 about twenty ants were collected under one of them, but not all 

 of the ants removed to the red glass, however, until February 

 22. The glasses were now arranged as at the beginning of the 

 experiment and the carbon disulphide cells were placed over the 

 two clear plates. The experiment now proceeded as before, with 

 no essential difference in results, until the red, orange and green 

 glasses had been removed from the nest. The green plates were 

 removed on March 8, and two days later all ants were collected 

 under the two disulphide cells. I then returned the green plates 

 to the nest, and the following day ten ants were in the room 

 under one of them. The number that left the disulphide cells for 

 the green gradually grew until, on March 17, all had done so. 



Experiment II. 



This colony was now removed from the nest and, after the 

 latter was thoroughly cleaned, another colony of C. pennsylvanicus 

 containing a large number of winged queens and males was intro- 

 duced. The glass plates were again arranged as at the beginning 

 of the experiment and the disulphide cells were placed over the 

 clear plates. The colony would have filled about three of the 

 rooms, but they scattered out and occupied eight of them, omit r 

 ting entirely, however, the two indigoes and the two blues. The 

 nest was left as first arranged from March 20 to May 12, and 

 some ants were seen at all times in each of the eight rooms in 

 which they first settled, except a few days while the disulphide 

 cells were removed. During all this time only occasional strag- 

 glers were ever seen under the blue or the indigo plates. 



Experiment III. 



A number of the queens of the colony used in experiment II. 

 were removed for other experiments and the colony was reduced 

 in size until it could easily occupy two of the rooms, and then, 

 on May 12, the nest was taken into a room which admitted no 

 light from the outside and which was supplied with an arc light 

 of four hundred and eighty candle power. The nest was placed 

 about three feet below the arc light, and a little to one side so as 

 to avoid the shadows of the lamp. The glass plates were 



