114 
Psyche 
[Vol. 87 
instead of five as in Ambononeura; however, in the other members 
of the order the branches range from two to six. As in nearly all 
genera, M has a single, prominent fork. CUA arises from the base of 
M, just before the level of origin of RS, and is slightly curved as it 
runs toward the hind margin, ending somewhat vaguely in associa¬ 
tion with cross veins at the margin—a feature that occurs in many 
Caloneurodea, especially the Permobiellidae. The separation of 
CUA and CUP is greater than in any other members of the order, 
although not much greater than in the Permobiellidae. The condi¬ 
tion in Caloneura itself is of interest in this connection; CUA and 
CUP are only slightly nearer together than the other veins are to 
their adjoining veins. The structure in Amboneura is probably the 
more generalized, since the ancestral stock of the order Caloneuro¬ 
dea presumably had a more normal venational pattern than is 
present in most of the members of the order as we now know it. It 
may well be also that the reticulation of some of the cross veins is 
another generalized trait, since it is present in many Protorthoptera. 
The diagnosis of the Caloneurodea, given above, has been slightly 
changed to permit inclusion of Amboneuridae. 
I take this opportunity to discuss the relationships of Calo- 
neurella carbonaria Carpenter, 1934, the only other member of the 
order Caloneurodea recorded from the anthracite coal fields of 
Pennsylvania. This was originally placed in the family Caloneuri- 
dae, otherwise known only from the Commentry beds of France. At 
the time of the description of the genus, very few specimens of the 
order Caloneurodea had been found and their classification was 
very arbitrary. Since then the Commentry specimens have been 
revised and the number of known families, genera, and species has 
tripled. Under the revised definition of the Caloneuridae, Calo- 
neurella can no longer be placed in that family, but it can be 
appropriately assigned to the Permobiellidae. This family was based 
on the genus Permobiella from the Permian of Kansas (Tillyard, 
1937) and was originally considered to be a subfamily of the 
Neuroptera. Martynov (1938), recognizing its affinities with the 
Caloneuridae, transferred it to the Caloneurodea, where it almost 
certainly belongs (Carpenter, 1943). The family has been character¬ 
ized by the relatively short SC, which terminates just beyond mid¬ 
wing; by the presence of 3 branches on RS, a deeply forked M, and a 
slight divergence of CUA and CUP distally (figure 3B). Pseudo- 
biella, the second genus in the family, shares these features. 
