354 Journal of Agricultural Research voi. vi. no. io 



weakening, unless the trees can put out plenty of new rootlets to replace 

 those injured and destroyed. This condition has been noted especially 

 on light-clay soils where poor cultivation was employed. 



SPREAD OF THE INSECT 



In nurseries under favorable conditions the spread of the insect may be 

 rapid. A half-acre pear nursery examined on June 9, 191 5, failed to show 

 infestation, though the aphis was probably present. When visited four 

 months later, on October 16, it was found that more than half the trees 

 examined were infested, some quite heavily. In large orchards where the 

 soil is permeated throughout with rootlets the aphis doubtless is very easily 

 diffused through the soil. In young orchards conditions indicate that 

 not much spread takes place from tree to tree. Infested young orchards 

 generally point to the nursery as the source of infestation, but the possi- 

 bility of infestation through the winged forms, or migrants, must be con- 

 sidered. A knowledge of the full life cycle of the insect alone can clear 

 up this point. 



BIOLOGY AND DESCRIPTION OF THE INSECT 



The wingless individuals live chiefly on the small rootlets and less fre- 

 quently on roots and the underground portions of the sucker growth. 

 They are always somewhat elongate and are for the most part pale 

 yellowish red, but they may vary from a pale pink or yellow to deep red. 

 They are rather sparsely clothed with long, curling, woolly, or cottony 

 filaments, of which there are four or six on each segment. Toward the 

 end of each instar these filaments are longer than the body — often three 

 times as long. There is a sparse whitish powder on the body, more abun- 

 dant at the caudal end. The cornicles appear as dusky-rimmed pores. 

 The young are pale yellowish red and elongate. 



The pupae develop on the same portions of the tree as the wingless 

 forms. They are very elongate in form and are clothed as are the wing- 

 less. The wing pads are inconspicuous and are white or light gray. As 

 a rule pupae on a rootlet develop almost simultaneously. The winged 

 forms issue together, after which the narrow, elongate, cast pupal skins 

 are conspicuous in little heaps, and are easily distinguishable from those 

 of the wingless forms. 



In the Walnut Creek district pupae and winged migrants were collected 

 in appreciable numbers from August 25 to November 17, and as late as 

 December 22 a nymph was found. These forms were most abundant in 

 September, and this obser\^ation apparently holds true for other localities 

 in California. Wingless colonies collected at San Jose, Cal., on June 10 

 and thereafter, kept in Petri dishes with moist sand in a cellar, produced 

 pupae on July 20 and migrants from July 24 to August 7. This appeared 

 to be abnormally early in the year for the production of winged forms, and 



