54 H. E. EWING. 



obtained with this species have been published, 1 but none of 

 those obtained beyond the first fifteen generations. Here there 

 is given for the first time a complete report of the pure line work 

 with Aphis avence Fab. up to the last generation obtained, the 

 eighty-seventh. I regret that, at the end of the eighty-seventh 

 generation, excessive heat killed all the individuals of the line; 

 stopping further experimentation which might have given 

 results of even more importance. 



Plant lice are especially well adapted for experimental pur- 

 poses in pure lines. They belong to the order Hemiptera, a 

 group of insects showing a relatively high degree of organization. 

 They can be reared easily on many small succulent plants; are 

 so small that a score or more of their breeding cages can be kept 

 in a small laboratory, yet they are large enough to be easily 

 handled and mounted on microscope slides. Plant lice reproduce 

 very rapidly, and give a sufficiently large number of progeny. 

 The species used gave a new generation every six days with an 

 average number of progeny of from twelve to twenty individuals. 

 Aphis avence Fab. is in fact so well adapted for pure line w T ork in 

 another respect that it was easily made the choice of the species 

 available in the locality where the experiments were started. 

 This characteristic is its ability to reproduce continually asexually, 

 or parthenogenetically, under the natural conditions found on 

 the Pacific slope the place where the work was begun. Occa- 

 sionally sexual forms have been reported from these milder por- 

 tions of the United States, but never has the present writer 

 found them there. In the north central and in the northeastern 

 states the sexual forms are common a's in the case of other species 

 of aphids. The choice of this species has been, I believe, very 

 fortunate, as the results here given will attest. 



But to begin with what are our conceptions as to the effect of 

 selection in a pure line? They are simply these: Selection in a 

 pure line of self-fertilized plants or animals, or plants or animals 

 reproducing asexually, no matter for how long carried on does 

 not change the somatic characters that are dependent upon the 

 germplasm. These somatic characters, according to the theory, 



Line Inheritance and Parthenogenesis," BIOL. BULL., Vol. XXVI., 

 pp. 25-35; and, "Notes on Regression in a Pure Line of Plant Lice," /. c., Vol. 

 XXVII., pp. 164-168. (Both published in 1914.) 



