THE LARCH WORM 881 



clustered together on a branch, and they ato continually, I should think, by the quick 

 work they did in stripping the trees. No juniper escaped destruction. The lower 

 limbs of some trees were left nutouched. 



C. Hill. 



We are especially obliged to Charles G. Atkins, esq., Fisb Commis- 

 sioner, and who traveled extensively during the last summer, for infor- 

 mation and specimens. He writes as follows : 



Manchester, Me., August 25, lb82. 



The editor of the Kennebec Journal wrote me that he had sent you one batch of 

 hackmatack worms, and was about to send you another. Doubtless you have all you 

 need. I did not come upon specimens until too late, though now that I have once 

 found them, I marvel that the affected trees did not sooner attract my attention. 

 They are all about here. 



I have just returned from a trip to Grand Lake Stream, Washington County, and 

 will give you the results of my observations on hackmatack insects. 



From Grand Lake Stream to Princeton, and thence to Forest Station, by stage, a 

 distance of 40 miles, the hackmatacks (there called juniper) had been attacked by 

 some insect that had shorn off the foliage of the upper part of each large tree. In 

 all that distance I did not see a dozen trees less than 25 feet high that had been 

 touched, but of those of 80 feet and upward in height 90 per cent, or more had been 

 attacked at the top and denuded (almost completely) down on an average, say 8 feet 

 or 10 feet from the top. The terminal shoots of the main stem and branches did not 

 appear to have been eaten off, but the side whorls of leaves were mostly gone. In 

 some cases the outer extremities of large limbs below the region generally denuded 

 had been attacked near their extremities. There were no worms to be seen on the 

 trees. I climbed one tree and searched it carefully, bat found nothing. On descend- 

 ing, however, I found a larva crawling on my coat-sleeve, a greenish slate color, some 

 three-fourths of an inch long, with black head, which I send you in vial. In Hinkley 

 Township I noticed some sphinx larv?e on hackmatack tips, and inclose one. I sup- 

 pose it was feeding, but did not verify supposition. 



From Forest to Bangor, wherever I saw large hackmatacks they had been gener- 

 ally denuded to a greater extent than on the first part of the route, and the work was 

 worse as I approached Bangor, and a smaller class of trees had been attacked than 

 in Washington County. 



I ascertained by inquiry that the devastations extended eastward as far as Orland 

 in that direction, beyond which I know nothing. 



From Bangor westward the depredations everywhere appeared (I came by rail to 

 Readfield), and on going to a remote part of my farm where hackmatacks grow, I 

 find they have generally suffered, but I notice here that trees under 10 feet in height 

 have generally escaped. Here I find that the dormant buds on the sides of the twigs 

 have begun to push out a new growth, which is now one-fourth of an inch long. 



I find lots of empty pupa cases in the turf under one of the trees, and send some 

 in a vial ; possibly some of them may contain pupa?. No worms to be seen now. 



Mr. A. P. Buck, of Foxcroft (postal messenger on E. & N. A. Railroad) told me 

 that they were at work in his vicinity, and had committed more havoc than anywhere 

 on the E. & N. A. Railroad, and even small trees had been completely stripped. 



Hon. Z. A. Gilbert, of East Turner (post-office), (his farm is in the northwest cor- 

 ner of Greene and southwest corner of Leeds, or near the Androscoggin River), says 

 the hackmatack worms have been operating in his vicinity for three years. After 

 the first attack the trees all leafed out. After the second some died, and now, after 

 the third, many appear likely to die. 



I showed the larva I got in Washington County to both Buck and Gilbert, and they 



thought it might be the same that they had seen in their sections, except that Mr. G. 



thought his worms were more positively green in color. He said it was characteristic 



of them to work first at the top of the tree, as I had observed in Washington County. 



5 ENT 56 



