March, 1910.] BrUES : PARASITIC HyMENOPTERA. 13 



of new and aberrant genera have also been discovered which are 

 referred to the family, so that during the past five or ten years our 

 knowledge of the group has been greatly increased. 



During this time it has become apparent that the group is of 

 very heterogeneous composition and that it must include several 

 series of forms which have been derived from different sources. 

 Unfortunately a number of the genera are at present known only by 

 their wingless or subapterous females which increases the difficulty 

 of determining their affinities, since the wing venation is perhaps 

 the most important single character so far discovered in defining 

 the limits of many of the groups of Hymenoptera. Even in forms 

 with wings, the neuration of these organs is much reduced or atro- 

 phied so that the primitive types from which they must have been 

 derived are difficult of determination. With these important char- 

 acters lacking, many forms exhibiting reduced neuration or atrophied 

 wings have been included in the family on account of their similar 

 appearance notwithstanding the possibility or even likelihood that such 

 a condition is the result of convergence rather than an indication of 

 real genetic relationship. On this account I have been led to attempt 

 a study of the characters of some of the members of the family 

 previously known, and, at the same time to present some conclusions 

 derived from several genera, both recent and fossil which are here 

 discussed for the first time. 



The Bethylinse are characterized by the elongate, flattened form 

 of the head, 12-13-jointed antennae, more or less elongate prothorax 



Fig. I. Palccobethylus longicollis Brues MS.; wing. 



and usually thickened legs. The wingless females in most cases 

 have the thorax much constricted between the meso- and metatho- 

 racic segments. This group appears to be more closely related to the 

 Ampulicidse than to any other family of the fossores and this rela- 

 tionship is very strongly shown by a peculiar genus which I have 



