J. CHESTER BRADJ.EY. 119 



sculptured on the inner surface with a groove, at the base of whicli 

 is a well-marked tooth, which in PammegiscJiia and Aulacus is 

 g'reatly prolonged. I have not seen Aiilacus, but according to 

 Kieffer's figure, it is much less prolonged in Atilaeus than in Pam- 

 iiiegiscliia. These two genera are also united with each other as 

 well as with Interaulacus and Aidacunis by the absence of any veins 

 excej)t Sc + R -j- M in the hind wings (Fig. 69). Likewise in 

 these four genera the claws are simple or at most with two teeth 

 beneath (Fig. 89). Pammegischia and Aalacus are parasitic on 

 Xiphydria. The hosts of Aulaclnus and Interaulacus are not 

 known. All other genera so far as known are parasitic on the 

 larv:e of Coleoptera. So that it seems as though these foui- genera 

 may togethei' form a distinct tribe in contrast to the remaining 

 genera. There is but one known species o^ Aulacus, although most 

 members of the subfamily have been described as belonging to that 

 genus. 



Deraiodonfus Bradley is a pure synonym of PristuiclacKS Kieff'er. 

 In my former paper therefore substitute for Deruiodontus Bra<lley, 

 Pristaulaeus Kieffer, and for Pristaiducus Kiefier, the new name 

 Xeaulacus. 



The most generalized wing of the family is Aidacinus (Fig. 67). 

 Note that K5 is entire; the base of M3+4 distant from r-m. Cell 

 M4 is very small — rather a specialization, as is. also the loss of 

 venation in the hind wings. But the simple tarsal claw is a primi- 

 tive character. The wing of Pammegischia ouelletii (Fig. 68) was 

 figured to show the trouble that may arise from malformations 

 if we do not exercise care. There is an extra cell cut off from first 

 M2, and R3 enters the stigma instead of being separated from it 

 by r. Both of these characters are monstrosities. The left hand 

 wing of the type specimen is normal. The wing of Pammeqis- 

 cJiia asJaneadi (Fig. 69) is typical of the genus. A slight advance 

 is shown in the partial-loss of R5. Specialization is also shown in 

 the prolongation of the coxse of the female. Pammegischia and 

 Aulacus may together be considered to form a sort of side branch. 

 Going back to Aulacvnus we find Interaulacus (Fig. 70) the first 

 step along another line, from which probably descended the remain- 

 ing genera. The base of Ms 1 is still distant from r-m. R5 is 

 entirely lost. The hind wings are as in Pammegischia. But the 

 claws have gained a single tooth. The venation in the remaining 



TEANS. AM. ENT. SOC. XXXIV. APBIL, 1908. 



