216 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



Harris states that tlio ef;f;s arc laid iluiiiijj: .Itily '-in cliisttTs on the nuderside of a leaf,, 

 generally near the end of a branch." He then observes: " When first hatched they eat only the 

 substance of the under side of the leaf, leaving the skin of the upjicrsich' and all the veins 

 untouched, but as they grow larger and stronger they devour wlioh^ leaves from the i)oiiit to the 

 stalk, and go from leaf to leaf down the twigs and branches" (Treatise, p. 425). He adds : The fully 

 grown caterpillars "rest close together on the twigs, when not eating, and sometimes entirely cover 

 the small twigs and ends of the branches. The early bi-oods come to their growth and leave the 

 trees by the middle of August, and the others between this time and the latter part of September. 

 All the caterpillars of the same brood descend at one time and disap]>ear in the night. They 

 conceal themselves under leaves, or just beneath the surface of the soil, and make their cocoons, 

 ■which resemble those of the ITnicoru Notodonta. They renmin a long time in their cocoons before 

 changing to chrysalids, and are transformed to moths toward the end of June or the beginning of 

 July " (Treatise, pp. 425-426). This habit of feeding exposed and living gregariously up to the time 

 of pupation jjn.ves tlic almost entire immunity enjoyed by tiiis caterjiillar from the attacks of birds. 

 We have also noticed iu Providence the simultaneous and suddeu disappearance of a whole bi'ood 

 from an apple tree at the end of September. 



Eegarding the habits of this species in California, Mr. Edwards states that he detected the 

 caterpillars iu tlie fall of 1875 "feeding upon willows in tlie neighbcuhood of Mount Shasta. Six 

 caterpillars taken, all feeding close together, upon a dwarf willow, their brilliant colors giving to 

 the phint at a little distance the appearance of a raceme of showy flowers. In a few days they 

 began to undergo their change, and by the 27th of August had all transformed. The i)erfect 

 insects began to appear on the 22d of December, a second followed on the !)th of January, and the 

 third on the IGth of March. The remaining specimens all died in the chrysalis state." Mr. Dyar 

 found the larvie he describes on the maple in the Yosemite Valley in August. 



The moth has been bred by Mr. Elliot from the willow, and i have fouud it in diftereut stages 

 of growth on the willow at Brunswick, Me., in August and September. It also feeds on the 

 aspen and blackberry iu Maine. 1 have also found the cater])illar feeding on the huckleberry 

 ( Vaccinitim). 



I found the eggs with tlie larv;e just hatching on the leaves of the willow at Brunswick, ^le., 

 June 24. The eggs were iu this case somewhat scattered and few in number, and the larvae did 

 not feed gregariously. The hrrv;e continue to hatch till the early part of August iu Maine, as 

 August 14 I found the larvae in Stage II and also fully grown on the aspeu. 



"This curious and. well-known caterpillar was received in August from Oregon. Mr. F. S.. 

 Matteson, of Aumsville, states that he found it in large numbers on a young api)le tree, entirely 

 denuding the branches of leaves. This mention is made as bearing upoTi the geographical 

 distribution of the species. The gregarious habits of these larviv when fir.st hatched admit of an 

 easy remedy in hand picking." (Kiley, Bep. U. S. Dcpt. Agr., 1SS4.) 



After the second molt some of the larvie are iclineunioneil. September 2 an ichneumon larva 

 had issued from the ventral side of the caterpillar and s\nui a white thin cocoon; the nearly dead 

 caterpillar was fastened by its back to the cocoon. After a day or two the caterpillar died and 

 turned wliitish, the rows of black warts becoming conspicuous. 



Kiley has observed the eggs in June; the larva' from .June to October; the moths in May and 

 August. 



Food i)lants. — Aj)ple, cherry, i)lum, rose, thorn, ])ear, lictiiUi uUm, willow, aspen, blackberry, 

 bramble, hucklebcu'ry ( VdcciniKm). I have found the larva' in ]\Iaine most commonly on the 

 willow, and it is probably from this tree that the insect has migrated to our fruit trees. In 

 California it feeds on the willow (Edwards) and maple (Dyar). 



In Beutenmiiller's list, besides the fiuit Trees already mentioned and different species of willow, 

 he has fouud it on the tlowering dogwood, sweet gum, persimmon, snowdroj) tree, bayberry, and 

 three different species of hickory. Apricot, wistaria, oak, locu.st, hickory, persimmon, poplar.. 

 (Kiley.) 



Geographical distribntioii. — This species has a wide range, extending throughout the Apjia- 

 lachian, Austrorii)arian, aiul Campestriau subprovinces from Maine and Canada to ]\Iissouri and 

 southward to Texas, (leorgia, aiul Florida. 



