34 THE CANADIAN ENTOMOLOGIST. 



at the sides, and ending in two long, three-sided palpi cases, which are 

 sharp and divergent, with an angular depression between them ; a cross 

 section of one of these cases would give a right angle, the long side a 

 little convex ; wing cases flaring on the dorsal side ; color of dorsum and 

 the entire abdomen, in the summer and fall examples, sordid yellow-buff, 

 the wing, antennas and palpi cases all more yellow ; the surface finely 

 streaked with brown, mostly longitudinally and always irregularly ; from 

 base of mesonotum to last segment a brown band ; the- wing cases show 

 an irregular, wavy, brown stripe on the disk and another on the costal 

 margin, and each nervule ends in a blackish dot (fig. c). 



In the spring brood the chrysalids were from .46 to .50 inch, breadth 

 .14; color blue-green, the dorsum streaked irregularly with paler or 

 whitish-green ; the wing cases finely granulated with pale green, and with- 

 out any stripe, but the nervules end as in the other form, in blackish dots ; 

 there are also two such dots on the inner side of the antennas cases, at 

 about half their length ; the outer edge of the wing cases cream color, and 

 this is continued along the head and palpi cases ; the keel of mesonotum 

 lightly marked by same color. Duration of this stage in May and June 

 8 days. 



The attitude of the larva when suspended is peculiar. From last seg- 

 ment to 4th the body hangs almost perpendicular, the dorsal side incurved 

 and ventral correspondingly curved outward ; the anterior segments are 

 bent at a right angle, the head being turned down upon second segment. 

 In Sosybius, the only other species of Coenonympha whose early stages 

 are fully known to me, the attitude of the suspended larva is that of figure 

 6. Gemma in its larval stages resembles an Apatura ( Celtis) more than 

 it resembles Sosybius. It holds the head bent under so that the horns are 

 nearly or quite in the plane with the back. The chrysalis differs 

 materially from any Satyrid chrysalis known to me, that is, from Sosybius, 

 Eurytris or Nephele. 



The figures of both larva and chrysalis of Gemma in Boisduval and 

 LeConte, Plate 62, are quite wrong, and must have been drawn from some 

 other species. The larva figured has no horns, and moreover is round 

 headed. Gemma is not uncommon in this region, but I have found it 

 restricted to certain localities, in open woods, or near woods, and at the 

 proper season I can always find examples by visiting these places. It is 

 three brooded, and, as shown above, I have raised each brood from 

 the egg. 



