AET. 7 REVISION OF COCCOPHAGUS COMPERE 5 



The following key may aid in defining the position of G occoyhagus 

 in its relation to allied genera.^ 



1. Antennae eight-jointed 2 



Antennae nine-jointed, funiele four-jointed Euxanthellus Silvestri. 



2. Scape usually fusiformly subcylindrical, not strongly expanded nor strongly 



compressed 3 



Scape strongly compressed and expanded Prococcophagus Silvestri. 



3. Flagellum not compressed, funiele joints usually plainly longer than wide 4 



Flagellum compressed, funiele joints about as wide as long or wider than 



long Aneristus Howard 



4. Mesoscutum with a reduced number of small setae, less than 18, arranged 



in bilateral symmetry 6 



Mesoscutum usually plainly densely setose, in a few species with a reduced 

 number of setae or microscopic pale setae; Coccophagoides with about 24 

 setae in bilateral symmetry 5 



5. Flagellum of antenna plainly differentiated into club and funiele. Marginal 



vein plainly more than one-half as long as the submarginal 



vein Coccophagus Westwood 



Flagellum not differentiated into club and funiele, flagellum tapering toward 

 the apex. Marginal vein unusually short, about one-half as long as the 

 submarginal vein. Stigmal vein intermediate between Coccophagus and 

 Prospaltella Coccophagoides Girault. 



6. Antennal club composed of three joints differentiated from the funiele 7 



Antennal club composed of two joints, or the club and funiele not dif- 

 ferentiated Encarsia Forster 



7. Fore wings very slender with the marginal fringe usually longer than the 



greatest width of the wing Aspidiotipiiagus Howard 



Fore wings not so narrow as in the foregoing and the marginal fringe usu- 

 ally plainly less than the width of the wing Prospaltella Hovt^ard 



MORPHOLOGICAL CHARACTERS 



Head. — Due to shrinkage if the specimens are preserved dry, and 

 because of distortion or concealment if mounted in balsam, such 

 characters as the proportions and shape of the head are not easily 

 utilized. In fresh material, these characters can be noted but the 

 taxonomist rarely obtains specimens that have not partly shriveled. 



Mandibles. — These are extremely variable, ranging from sharply 

 tridentate in some species to edentate in others, with the greater 

 number having a more or less typical mandible characterized by a 

 well-developed apical tooth and a broad dorsal truncation that may 

 be more or less incised so as to form a median and inner tooth. The 

 mandibles usually possess two stout peg-shaped ventral spines and 

 a similar but weaker dorsal spine in addition to the setae. Unfor- 



2 1 am unable to detect good generic characters to distinguish these genera, but believe 

 that since the genus is an arbitrary category established to aid in the recognition of 

 species, better results are obtained if group distinctions are perpetuated where hundreds 

 of species are involved even though certain species intergrade and exact generic concepts 

 can not be established. With a little practice it is not difficult to assign most species to 

 one genus or another and confusion results only when the intergrading species are 

 encountered. 



