if. K. K\VIN<;. 



After studying the different characters which were available 

 I finally decided to use those of the antennae. The variations 

 in the relative lengths of the third and fourth segments were 

 observed and employed. It is interesting to note that these 

 characters are the same as- those I selected two years ago when 

 working with Aphis brassicce Linn.; they are also the same as 

 those used by Kelly in his work with Aphis ntmicis. 



METHODS. 



In order to obtain as nearly as possible uniform conditions for 

 these experiments, I reared all of the aphids on young wheat 

 plants. These wheat shoots were never allowed to get to be 

 more than ten or fifteen days old, and I kept them nipped back 

 to a height of not more than three or four inches. New plants 

 were reared from seed for each generation of plant lice. All 

 of the plants were kept upon a single large table in the middle 

 of the laboratory, so that changes in heat or moisture would 

 affect all individuals of each fraternity in much the same way. 



The wheat plants used were grown in small flower pots, a single 

 plant being raised in each pot. Soon after the aphids were born 

 they were separated, and each nymph was placed on a single 

 wheat plant growing in its own pot. This aphid was confined 

 by placing a small lamp chimney, covered at the top with 

 surgeon's gauze, over the wheat plant. This would permit 

 ready observation of the developing aphid, and w y ould also permit 

 the transpiration of the wheat plant. Each individual was 

 ui\en a label which showed the generation to which it belonged, 

 and also its individual number in this generation. Thus, F 7 I4 

 would mean individual number fourteen of the seventh filial 

 generation . 



When it came to taking measurements of the third and fourth 

 segments of the antennae in order to record the same and to 

 select individuals for carrying on the strain I encountered several 

 difficulties. At first I tried to measure these in the live adults 

 by weighing dow r n the insect with a cover-glass, but this would 

 not enable me to make accurate measurements. After trying 

 -< 'veral other methods I finally decided to let each adult remain 

 on its own wheat plant, and produce her progeny; after which 



