Keactions and Pkoducts in Inteespecific Crosses 91 



Cut food wilts in a short time and with especial rapidity under high tempera- 

 tures or under arid conditions, and attempts to keep the food fresh by placing 

 it in water end in failure or in indifferent results at the best. When a healthy 

 growing plant is used as the food, the animals always breed and lay their eggs 

 exactly as in nature, and in my experiments over 90 per cent of the fertilized 

 eggs laid (by actual count) hatch, and the freshly emerged, delicate larvae can 

 at once begin feeding in a perfectly normal manner without in any way being 

 molested or injured. 



The most satisfactory type of cages thus far devised for this work consists of 

 a light, strong drum of fine wire-netting (pearl wire, No. 23 mesh), of which 

 the sides and top are of wire and the bottom of galvanized iron, with a sleeve 

 that projects down into the earthen pot that serves as the base as 8ho"\vn in 

 figure 1. The pots that I have used are 12 inches in diameter and hold a con- 

 siderable volume of earth, in the center of which is placed the food-plant in a 

 smaller pot. The breeding-cages are uniformly 18 inches in diameter and are 

 either ] 8 or 24 inches in height. A cage of this character is used for each mated 

 pair and for their eggs and young progeny. With this arrangement the breeding 

 pair has normal, healthy, and fresh food, proper conditions for egg-laying, and 

 imcrowded space about them. They have for all practical purposes natural 

 conditions for their breeding activities. 



I allow one pair to breed in such a cage, and when the progeny are half grown 

 they are put into sorting-cages in which the same construction is employed, 

 excepting that they are only 12 inches in diameter (fig. Ic) ; though if the 

 progeny are especially abundant or valuable the parents are removed to a fresh 

 cage to breed further and the first 100 or 200 are allowed to complete their 

 ontogeny in the original cage. As soon as the food-plant is defoliated a new one 

 is easily put in its place. 



Entrance to the cages is had by means of a sliding door in the side, large 

 enough to permit the passage of an 8-incli pot and food-plant. This type of 

 breeding-cage has the advantage that it is self-sustaining, and needs but little 

 daily attention aside from the routine spraying of the plant and the adding of a 

 regular amount of water to the soil in the pot. The mechanical labor of the 

 project is thus reduced to a minimum, with constant efficiency of operation. 

 With these arrangements I have been remarkably free from epidemics of disease. 

 The only epidemic that I have had to combat was not a very serious one. It was 

 introduced during my absence and was not recognized by the assistant in charge, 

 because of his never having encountered anything of the kind before. 



The larval stages are thus passed under optimum and normal conditions and 

 they pupate in the soil of the same cages in which they pass the latter portion of 

 the larval period. As a rule it has been found advantageous to keep the later 

 stages in the sorting-cages, which occupy less room and leave a larger number 

 of breeding-cages free for the mated pairs. After pupation the cage is watched 

 with care to keep the soil at the optimum condition as regards moisture, and 

 also for the adults to emerge. These are isolated at once, the males from the 

 females, in separate sterile cages, and given living food as in the breeding-cages. 

 This separation is made two or three times daily during the period of emer- 

 gence, within 1 to 4 hours after emergence, and before the post-pupal develop- 

 ment has been completed. The emerged adults are without exception allowed 

 to remain in these cages, with the sexes separated from one another, from 5 day& 



