322 MAINE AGRICULTURAL EXPERIMENT STATION. 1 908. 



in some individuals to a heavy marking in reddish brown or purple in 

 other individuals of the same age. Usually the third thoracic segment 

 bears a dark oblique lateral mark with the ventral end slanting cephalad 

 and the sixth abdominal segment bears a similar mark with a ventral 

 slant caudad. Both these marks are sometimes missing and very often 

 other similar marks are present on other segments. 



This species would be a fascinating subject for a study in variation 

 but nothing could really be done with it without the constant service of 

 both a photographer and an artist to reproduce the. intricate designs in 

 color as no description could serve to fix them adequately. 



Cocoon. Figure 26. The cocoon is formed by a very thin layer of 

 silk usually lining a place hollowed in the leaf mold at the base of the 

 infested tree. Sometimes the cocoon is firm enough so that it can be 

 removed with the particles of earth attached as with the specimen photo- 

 graphed. 



Pupa. Figure 25. Within ihe cocoon is a dark glistening brown pupa 

 about 19 mm. long. Just caudad of the mesoscutum is a curve of 8 dull 

 unpolished tubercles the two at the ends being triangular, and the other 

 six being nearly square. The terminal spines of the cremaster are some- 

 what boot shaped, the toe being turned laterad and the heel mesad. 



SEASONAL HISTORY. 



Packard records the following dates for this species.* 



"The eggs were found at Brunswick, Maine, as early as July 

 3, and it hatched July 11 or 12. Other larvae, as observed in 

 Maine, hatched about the 8th to loth of July, feeding on the 

 under side of the leaf, at first eating away a little irregular 

 patch. Stage I lasts 9 days, Stage II probably 4 or 5 days. 

 The last stage is reached a month later, August 9-10; one 

 belated individual occurred on the oak at Providence as late 

 as September 20 to 24. 



Riley notes larvae as occurring in July, and captures of the 

 moths in May, June, July and August." 



Doctor Dyar gives the species as "occasionally double 

 brooded" ** and Mr. Beutenmiiller says the species "is not 

 common and possibly double brooded." *** 



In view of these records the calendar for H. guttivitta during 

 the past season's outbreak in Maine is of particular interest. 



May 27, 1908, the writer visited a beech growth at North 

 Fryeburg, Maine, which had been stripped by H. guttivitta the 



* Bombycine Moths, page 234. 



** Psyche VI, p. 179, 1891. 



*** Bombycine Moths of Vicinity of New York, p. 426. 



