AMERICAN COLEOrXERA. 221 



which the family seems to be as well defined an aggregate as very 

 many of the families ia the coleopterous series. 



As constituted in the present essay the family differs in its com- 

 position from that at present given in the books. Lacordaire and 

 Duval include Leptinus, the Catalogus (p. 741), adds ISphaerius, and 

 Leconte (Class. Col. N. A. pp. 48 and 52), adds Brathinus as a 

 subfamily. 



Regarding Leptinus I can only say that I fully agree with Leconte 

 in separating it as a distinct family, and have nothing to add to the 

 views expressed by him (Proc. Acad. 18GG, p. 3G8), except that it 

 seems more closely allied to this family than the Hydrophilidae. 

 I am not aware of any reason why Sphpeiu'us should be added and it 

 seems that there are very many to the contrary, so that this does not 

 seem to need discussion here. 



The addition of Brathinus even as a subfamily introduces a very 

 disturbing element to any system of classification, and after a careful 

 examination I think the view originally expressed by Dr. Leconte 

 (Proc. Acad. 1852, p. 150), is the correct one, and that the genus 

 should take its place among the Scydmsenidae for the following reasons : 

 The head is suddenly constricted behind to a neck which is gradually 

 broader posteriorly, approaching the semiglobose form of Sci/dnisenus ; 

 the anterior coxae are as in that genus and similarly with trochantin, 

 the middle coxae slightly prominent and separated, the posterior promi- 

 nent, slightly transverse at base but contiguous, the trochanters oval, 

 rather flat and not in the axis of the thighs and the elytra have no 

 epipleurae. The greater length of the maxillary palpi and the form 

 of the hind coxae have been urged as objections to placing Brathinus 

 in Scydmsenidae. If the palpar character has any weight (and it 

 seems to me to have no greater than generic value), the preponderance 

 is greater against the Silphidae than the Scydmaenidae. The structure 

 of the hind coxae seems to me essentially that of the latter family, 

 the apparently transverse character being due to their contiguity. 

 The absence of epipleurte seems to be general in the Scydmaenidae. 

 In Brathimis it will be observed that the first four abdominal seg- 

 ments have on each side near the posterior edge a deep setigerous 

 puncture. I have not been able to assure myself that this occurs in 

 the Scydmaenidae. Finally the general appearance of Brathinus is 

 rather that of Sci/dmsenus than of any Silphide, and the comparison 

 of it with Leptodirus seems singularly inappropriate. 



Having thus disposed of the genera which seem foreign to the 



