510 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 53. 



characters, and there is apparently ground for doubt as to the validity 

 of Szepligeti's two genera above mentioned. At least the characters 

 used are of very doubtful generic value. C. rostratus, new species, 

 and C. longigenalis, new species, display the characters peculiar to 

 Paracremastus SzcpUgeti but to a less marked degree, while several 

 species placed toward the end of the key to species run in Szepligeti's 

 table and agree fairly well with the description of Pseudocremastus 

 Szopligeti. But the characters which ally them with those genera 

 prove, upon examination of a large series of species, to be subject to 

 great variation. 



Temelucha plutellae Ashmead, the genotype of Temelucha Foerster 

 is conspecific with a specimen in the United States National Museum 

 determined by Schmiedeknecht as Cremastus decoratus Gravenhorst, 

 which all European authorities concede to be a true Cremastus. 

 The genotype, Cremastus spectator Gravenhorst, I have not seen. 



Zaleptopygus, erected by Viereck for his species ohereae and Porizon 

 orhitalis Cresson, is really not at all closely related to Leptopygus Foer- 

 ster. The propodeum, in which Viereck says it is more closely 

 allied to the latter genus than to Cremastus, differs essentially from 

 that of a typical Cremastus only in being more robust, shorter, and 

 in having the areola and petiolar area separated by a carina. In 

 Leptopygus, as represented by the genotype, L. harpurus (Graven- 

 horst) the basal transverse carina and the basal portion of the 

 lateral carina are obsolete, while that portion of the lateral carina 

 that separates the apical lateral and apical pleural areas is strong. 

 In Cremastus the opposite is true. 



The prmcipal characters which distinguish this genus from Pseud- 

 eripternus Viereck are indicated in the table to genera, and the 

 more striking ones designated in the discussion of the latter genus. 

 The only character separating it from DolichoselepTius Ashmead is 

 found in the normal palpi as distinguished from the abnormally 

 long palpi of the latter genus. It is easily distmguished from 

 Neocremastus Cushman by the characters used in the generic key. 



In secondary sexual characters the males of some species differ 

 very little from the females, while othei-s exhibit very striking modifica- 

 tions. The eyes and ocelli in some species are but little larger m 

 the male than in the female, while hi others the eyes occupy the 

 whole sides of the head and the ocelli occupy nearly the entire 

 vertex. The postpetiole in most males is more depressed than in 

 the females, and the petiolar carinae and foveae arc stronger, beuig 

 sometimes present in the male when absent in the female. 



In using the key for the determination of species no character in 

 a group should be given precedence over any other, but a majority 

 of characters should be considered as indicating the way in which 



