1917] Phytogeny of the ElateridcB 243 



I do not consider the Buprestidas as at all related to this super- 

 family, and have not yet studied the Throscidae. Mr. F. C. 

 Craighead has very recently loaned me a larval exuvium of 

 Drapetes geminatus Say. This species, heretofore referred to 

 the Throscidae, is certainly very closely allied to the Elateridae 

 of the subfamily Oestodinae in its larval characters. Through 

 the kindness of Dr. Boving I have received specimens of Throscus 

 dermestoides from Mr. C. E. Rosenberg, of Copenhagen, which 

 are most assuredly not closely allied to the Elateridas, and 

 conclude that on the present evidence it will be necessary to 

 remove the genus Drapetes from the Throscidae. 



The hitherto supposed relationship of the Elateridae and 

 Buprestidas, using the Eucnemidas as the connecting link, is 

 undoubtedly erroneous, the larval resemblance being purely 

 superficial. The characters formerly used to differentiate the 

 Eucnemidae from the Elateridae have failed to hold even with 

 the few specimens I have examined. They are not as has 

 generally been conceded, universally apodus as is clearly to 

 be seen in an undoubted Eucnemid larva occurring in rotten 

 logs in the Eastern States, in which distinct, though rudi- 

 mentary, legs are evident. The mouthparts are adapted to 

 wood boring, so their unique structure cannot be given over- 

 much taxonomic weight, and the tenth abdominal segment is 

 situated ventrocephalad to the ninth as in the Elateridae. 

 The careful anatomical studies made by Schiodte^ (1847 and 

 1865) of the adults separate the Elateridae and Buprestidas 

 very widely. The old group Sternoxes, as we now understand it, 

 consists of three groups : the superfamily Elateroidea ; the family 

 Throscidae, part of which may eventually be placed in this 

 superfamily; and the family Buprestidae. The superfamily 

 Elateroidea embraces the families Cerophytidae, Cebrionidae, 

 Plastoceridae, Elateridae and Eucnemidae. The Cerophytidae 

 and Plastocerid^ are unknown as larvae. Their very doubtful 

 position cannot fail to be somewhat more positively determined 

 by their larvae when these are described. The plate (Fig. 1) 

 indicates in outline the various concepts of this comply. 



9 (Naturhist. Tidssk. ser. 3, Vol. Ill, 1865) and Kongo. Danske Viden. Selsk. 

 Forh. 1847). 



