go Annals Entomological Society of America [Vol.11, 



females. Many individuals were found isolated on bean leaves, 

 but none of these succeeded in establishing colonies. The mi- 

 grants were certainly not depositing larvae on the normal leaves 

 of the Chenopodiums. About the middle of October, 1908, at 

 Forest Hills, Mass. I found many small plants of this genus grow- 

 ing in soil that had been cultivated or newly made after July. 

 These plants must have come up late in summer. They were cov- 

 ered with the galls of this species. It is evident, then, that they were 

 colonized in August. Undoubtedly the winged females observed 

 migrating at Fredonia in the same latitude were merely colonizers 

 on tender young Chenopodiums starting up in damp situations. 

 It is highly improbable that these were true migrants going to 

 some distinct host. As a rule the sexes of truly migrating plant 

 lice develop on some woody plant called the primary host, while 

 the summer forms of the species live on a succulent herb, the 

 secondary host. Since the sexes of Aphis atriplicis developed in 

 great numbers on the Chenopodittms last fall, they are undoubt- 

 edly confined in their life-cycle to these plants and their allies. 



On September 21st the perfect sexes were first noticed on the 

 white goosefoot at Forest Hills. Small, lank, wingless males of 

 a yellowish color, with blackish appendages, were actively climb- 

 ing about over the large clusters of plump oviparous females, 

 and copulation was observed as late as the 30th of October. 

 Large numbers of these females were found abundantly on the 

 goosefoot plants in this neighborhood. The males were few in 

 number, between 5 and 10 per cent, of the total number of indi- 

 viduals (by guess) . But they were making up for their fewness 

 by great activity, and copulation was often seen. Most of the 

 females had left the summer galls and were on the seed-heads, 

 where they were frequently mixed with black aphids, the so- 

 called Aphis rumicis. Only a small proportion of the individuals 

 were found in the galls during the periods of oviposition. No 

 viviparas were seen at Forest Hills, and of many specimens taken 

 throughout the fall, none proved to be viviparous on examination 

 in the laboratory. Viviparity doubtless ceases or becomes 

 exceptional in this species with the development of the perfect 

 sexes. There is evidently a general migration of these sexes or 

 of the sexuparse producing them from the galls to the seed-heads 

 in the fall, where the eggs are usually laid. Oviposition was 

 observed during the entire month of October. The eggs were at 

 first light yellow in color, then deep green, finally becoming shin- 



