J. Davidson 63 



Experience in breeding apliids shows that in investigating the re- 

 production on dii?erent hosts, the following factors must be considered : 

 {a) The physiological condition of the plant, especially with regard 



to its age, temperature, light and food. 

 (h) Temperature and humidity conditions, 

 (r) The history of the previous generations, the generation of the 



aphids used, and the kind of plants on which they have been 



reared. 

 {(l) Whether winged or apterous agamic females are used, 

 (e) The variabihty in reproductive power of the individual agamic 



females. 



(6) Influence of Food Plants on the Characters of the Species. 



The quantity and quahty of the food are very important factors in 

 affecting the appearance of Aphis runticis, especially in regard to colour 

 and size. When kept for several generations on Euonymns the aphids 

 in the later generations are very much smaller than the earlier genera- 

 tions, but by cutting back young trees, thus producing a succession of 

 young growth, the aphids were carried on the spindle tree for several 

 months. They were somewhat smaller than the earher generations, and 

 considerably smaller than the corresponding generations on broad beans, 

 also the reproductive capacity on Euonymus was greatly inferior. On 

 broad beans the aphids were always large, healthy individuals, except 

 in the case of very heavily infested plants. On poppies they were 

 somewhat smaller. On mangolds, sugar beet, and red beet they were 

 also smaller than on beans. On dwarf peas they were quite small indi- 

 viduals but healthy in appearance, while on dwarf French beans they 

 were extremely small, dwarf individuals with very low producing power. 



There is in a general way some correlation between the size of the 

 individuals on the host plant, and the number of aphids produced. This 

 correlation between size and fertiUty is also indicated in Warren's (7) 

 statistical examination of Hyalopterous trirhoda Walker. 



Ewing{3) in his interesting statistical study of Aphis avenue Fab. 

 has shown the great range of fluctuating variations over a number of 

 parthenogenetic generations, these variations being individual and not 

 inherited. Extremely high and low variations were traced to the effects 

 of food and temperature conditions. Temperature will of course also 

 indirectly affect the food factor, in so far as it influences the metabolism, 

 of the plants. 



