88 



ALFRED E. CAMEROX. 



but it was recognised that the narrow confines of the breeding cages tlien employed 

 were detrimental to the display of the insects' normal activities. In order to eliminate, 

 so far as possible, all objections that could be raised on tliis score, a large frame- 

 structure cage covered with coarse grade, closely woven muslin was erected in the 

 field at the farm of the Agiicaltural College, Holmes Chapel. The cage measured 

 50 feet long, 14 feet broad and 5 feet high, giving a total capacity of 3,500 cubic feet. 

 By erecting partition walls of muslin, the cage was divided into five equal-sized 

 compartments of 10 X 11 X 5 feet, which gave a, capacity for each of 700 cubic feet. 

 This allowed for sufficient space in which one might conveniently attend to the 

 experiments. It may be mentioned that in order to lessen the risk of tearing under 

 the stress of wind and wet, the nmslin was securely fastened by means of narrow 

 strips of wood nailed to the supports: Similarly, the muslin on the roof Avas attached 

 to the cross-supports in like fashion. Midway between the cross-supports of the roof 

 of each compartment and those of the sides and ends two similar strips of wood were 

 nailed to each other with the muslin held firmly between. An idea of the general 

 strurture will be obtained from fig. 1. 



Fig. 1. 



There was no communication between one compartment and another, but each 

 was supplied with a small door just large enough to allow the experimenter to enter. 

 All the doors, which were similarly placed on the same side of the cage, were covered 

 with mushn, and great care was taken that they fitted closely to the jambs. The 

 cage was absolutely fly-proof, so that those flies introduced into the compartments 

 in the course of the experiments had no chance of escape, and, similarly, no random 

 specimens could enter the cage from the outside. Several species of flies and various 

 insects, other than those used in the experiments, were collected inside the cage, and 

 of these Leptis scolojMcea, L., Leptis tringaria, L., and Empis tesselata, F., predominated 

 from June to September. Their larvae and pupae must have been present in the 

 soil in large numbers, and when the adults emerged, they were imprisoned in the cage. 



