36 PROCEEDINGS OF THE NATIONAL MUSEUM vol.88 



identities separate. This indicates a common origin of these species 

 that have been separated by geographical barriers for a sufficient 

 length of time to be differentiated b}' small morphological characters, 

 differences in biology, and host preferences. These individual pe- 

 culiarities are maintained even when the forms are brought together 

 in the same habitat either by artificial or other means of dissemina- 

 tion. P. vulgaris and P. pecosensis are examples of this. Several 

 others will be cited in this paper. P. vulgaris apparently accom- 

 panied Pieris rapae in iis spread to the Nearctic realm, whereas P, 

 pecosensis is a Nearctic species. 



Townsend published a good description of the species. A few of 

 the minor differences from vulgaris that this latest study indicates 

 are noted : A blacker, distinctly more subshining species. Head with 

 front of male at narrowest 0.35 to 0.38 (in five specimens) and front 

 of female 0.37 to 0.38 (in five specimens) of head width; frontal 

 bristles exhibiting a tendency to be fewer in number than in vulgaris, 

 but there is no clear-cut demarcation, usually only two reclinate up- 

 per frontals (prevertical bristles) ; facial ridge bristly on lower one- 

 half to two-fifths ; antenna black, third segment in male four times or 

 slightly more and in female nearly three times as long as second 

 segment; arista thickened on basal half, penultimate joint elongate; 

 palpus usually black, varying to brownish black. 



Thorax black, subshining; when held at the same angle and in a 

 more or less upright position in which the mesonotal vittae of vul- 

 garis behind the suture are easily seen, the mesonotal vittae of peco- 

 sensis are not visible. 



Sides of second abdominal segment in male reddish and in female 

 usually black or only slightly red. 



Length 7-9 mm. On the average this species is slightly smaller 

 than Phryxe vulgaris. 



Type locality. — Pecos National Forest, N. Mex. 



Distribution. — Maine 2, New Hampshire 8, Vermont 5, Massa- 

 chusetts 11, New York 5, Montana 2, Idaho 2, California 1, Wash- 

 ington 1, British Columbia 1. Published records: Quebec, several 

 places (Tothill) ; New Brunswick (Squires); and Illinois (Forbes). 

 There exist additional doubtful records that can be confirmed only 

 from an examination of the specimens involved. Essig's record 

 probably refers to the specimens listed above under Idaho and 

 Washington. 



Hosts. — Erannis tiliaria (Harris) 56, Archips cerasivorana (Fitch) 

 15, Ennomos subsignarius (Hiibner) G, Cirphis unipuncta (Ha worth) 

 4, Cingilia catenaria (Drury) 3, lepidopterous pupa 3, Danaus plex- 

 ippus (Linnaeus) 2, unidentified noctuid 2, Archips argyrospila 



