THE CANADIAN ENTOMOLOGIST. 183 



the incisures. These stripes meet on the anterior part of the third thoracic 

 segment, and are progressively reduced from the fifth to eighth abdominal 

 segments. There is also a line running along the substigmatal fold from 

 the third thoracic to the last abdominal segment, and an obscure spot on 

 each of the first six abdominal segments above the spiracle line. Plate 2, 

 fig. T, shows the dorsal aspect of the larva immediately after the second 

 moult. 



The final moult occurs when the caterpillar has attained a length of 

 about S.T, mm., but no definite period can be set as the duration of this 

 stage, its length depending greatly upon the health and appetite of the 

 individual. The same is true of the stage immediately preceding pupa- 

 tion, though in general it may be stated that when a caterpillar has had an 

 unusually long penultimate stage the ultimate stage is abbreviated. This 

 last moult is accomplished within twenty-four hours. 



Ultimate stage. — Scarcely different from preceding stage. Head 

 gray-green, sometimes more or less yellowish. Dorsal shield inconspicu- 

 ous, light brown near the incisure, shading to creamy-white on its anterior 

 margin. Mediodorsal stripe, dorsal and substigmatal lines as before. 

 From first to eighth abdominal segment a series of oblique dashes of the 

 lighter tint, one to a segment, meeting the dorsal line just back of the 

 anterior incisure and running downward and backward almost to the 

 spiracle line. These aie well developed only on the second, third, fourth, 

 fifth and sixth abdominal segments, where they are sometimes accom- 

 panied by a small spot beneath and in front. 



Variation. — The larv?e are remarkably constant, the only variation 

 being in the relative values of the two tints. The lighter of these is 

 occasionally so lacking in chroma as to appear by contrast almost white. 

 Usually the difference is slight, and the markings in consequence are 

 obscure. I endeavoured to raise this species on Vaccinium to determine 

 how much variation might be due to a difference in the food. My failure 

 was conspicuous ; fifteen caterpillars just from the egg died rather than 

 touch corymbosuj?i, vacillaiis or petuisylvaiiicinn^ and others, further 

 developed, placed upon these plants forsook them immediately. With 

 the same purpose I furnished an equal number of newly-born larvae with 

 plum twigs. One only found the young fruit to its liking ; the others died 

 without making any attempt to feed. The single exception lived to 

 complete the second moult. The ground colour was a little deeper and 

 of a more bluish-green, the markings nearly white. This is the larva 



