426 
MEXICAN DERMAPTERA 
In the present material the third and fourth dorsal abdominal 
segments have well developed, brief, rounded, lateral carinae 
(the stink-glands), those of the fourth segment being the larger. 
The male pygidium is declivent, slightly longer than broad, with 
narrow proximal portion concave and large distal portion convex 
in contour, nearly circular in outline, without trace of disto- 
lateral projections. The caudal limbs have the first tarsal joint 
decidedly longer than the combined length of the second and 
third joints. 
These specimens have the head, antennae, limbs, greater 
portion of pronotum and the tegmina along the humeral angle, 
cinnamon rufous. The large meso-caudal portion of the pro- 
notum, remaining area of tegmina, the abdomen and forceps are 
chestnut brown. 
ANCISTROGASTRINAE 
The forms of this group are varied and striking, the specializa- 
tion of the male dorsal abdominal segments being in many species 
remarkable. 
We are forced to change the generic association of a number of 
species, the evidently wrong assignments being mainly due to the 
separation of genera upon the but rarely valid character of pre- 
sence or absence of fully developed wings. Had we material for 
comparison of all the species involved, the task would be less 
difficult. The presence in the collections before us, of the species 
perditus, genotype of Burr’s Praos, and championi, genotype of 
Burr’s Vlax, enables us to reach definite and important conclu- 
sions. The descriptions are exceptionally adequate and numerous 
figures are given of the other species with which we are con- 
cerned. The results may be summed up as follows. 
The structure of championi, genotype of Vlax, shows much 
closer similarity to that of perditus, genotype of Praos, than to 
any of the other species assigned to Vlax;^^ this is also true for 
perditus, when compared with impennis, assigned by Burr with 
a query to Praos.^’’ Indeed no sufficiently important features 
“ It is evident that Burr drew up his original description of Vlax from tolteca 
of Scudder, though he chose championi of Bormans as the genotype. 
The features compelling this assignment have been discussed in a paper 
on Panamanian Orthoptera, the species involved being in no way associated 
with the Mexican fauna. We would note that recently Borelli has described 
a species as Praos uncinatus, this insect being clearly a member of the genus 
Andstrogaster. That author has very kindly sent us a figure of the forceps, 
drawn from the type of this species. 
