278 
NORTH AMERICAN MYRMOSIDAE 
transverse carina; the lateral angles prominent, sharp, and strongly punc- 
tured, the punctures becoming obsolete on the lower part of the sides; the 
posterior face st<i’ongly declivous and hirsute. 
Wing venation as shown in figure 6; the veins and stigma Ijrown, Sc-|-R + 
M -\-ery dark l)rown, the margin beyond the stigma slightly and indefinitely 
infuscated. 
Anterior metatarsus with a few inconspicuous slender spines among the hairs 
which cover it; middle tibia with a group of ten or twelve prominent coarse 
spines on the apical half of its outer surface, two or three near the apex 
of the posterior edge being coarse and peg-like; posterior tibia with about 
six stout spines along its posterior margin and a few small ones scattered over 
the apical part of the outer surface and largely concealed by the hairs thereon. 
Abdomen sessile; the basal segment short, broad and thick, enlarging di- 
recth' from the l)ase without any anterior neck, separated by a deep constriction 
from the second ventral; this segment with a convex horizontal surface, and 
a nearly vertical, plane, basal surface, the two separated by an undulating 
transverse chitinous ridge, entirely characteristic of the species; first segment 
coarsely and closely punctured, the others very polished and sparsely punc- 
tured; pygidium obtuse, its lateral margins indefinite, with a very slightly 
raised median area and numerous coarse lateral punctures. 
Sagitta much reduced, each forming a thin edged, longitudinally rolled plate 
without apical processes or thickening; volsella reduced to a chitinous con- 
cealed tubercle; ramus large, without a hirsute patch on its inner side; squama 
thick, Ijlunt, its apical margin notched but not deeply so; uncus of a form pe- 
culiar to this species, being deflexed and with the apical portion bent back- 
ward at an acute angle, its tip bifid and spread out, resembling the digits of a 
tree-toad (figures 21, 22, and 2.3). 
Type material . — Holotype from Coalinga, Fresno County, 
California, June 1 to 3, 1907, below 500 ft. elevation (the author), 
[Cornell University, No. 125.1]. One paratopotype: same date, 
[Cornell University, No. 125.2]. One paratype from Calexico 
in the Imperial Valley, California, August 11, 1914, (the author), 
[Cornell University, No. 125.3]. Three paratypes from Nogales, 
Arizona, June 21 and 22, and July 10, 1903, (E. J. Oslar), [Cornell 
University, Nos. 125.4 to 125.6]. One paratype from Durango, 
Colorado, July 17, 1900, (E. J. Oslar), [Cornell University No. 
125.7]. 
This species finds its closest relatives in stygia, castanea and 
thermarum. The transverse ridge on the second ventral segment, 
and the peculiar shape of the uncus with its upturned apex, abun- 
danth' distinguish it from any other known species. 
