268 GEO. H. HORN. 



proper. Motscluilsky's genus Creniphilus, although merely indicated 

 and not fully described, has for its types punehihdus ( = ceneiis Germ.) 

 and Hmbatus ( = globulus Payk.), which are the types of Paracymus 

 and Anaccena. For this reason Creniphilus is adopted, indicating, 

 as it does, my idea that the other two genera cannot be properly re- 

 tained as distinct. In his most recent work (Biol. Cent. Am. vol. i, 

 2) Dr. Sharp has gone a step still further back, and retains Hydro- 

 hius as adopted by Lacordaire and the authors of the Catalogus. 



In glancing over the analytical table there will be observed a 

 very important character relegated to second place, namely, the oc- 

 currence of 8-jointed antennae in two species. To use this character 

 as a means of defining another genus would associate two very unlike 

 species, unlike not only in form, but also in the structure of the two 

 sterna, and the logical result would be the formation of two other 

 o-enera. If, then, we return to those species with 9-jointed antennae, 

 one certainly will be an Anaccena, while the others must be again 

 separated by the plane (Paracymus) or protuberant mesosternuni. 

 Thus five genera will be named, all with a more or less complicated 

 nomenclature confusing to the student and without benefit to science. 



The question might here be answered as to the reason for ignoring 

 such an important character as the loss of an antennal joint. It 

 will be admitted that, in every family of Coleoptera, there is a stand- 

 ard or normal number of antennal joints, to which in some fiimilies 

 all the members conform, while in other families there is a variation 

 in the number, sometimes either below or above the standard, and 

 in the Scarabeeidie, especially, both the latter styles of variation 

 occur within the family limits. In fact, such is the plasticity in the 

 Melolonthide Scarabpeidje, that instances are by no means rare of 

 antennae having the standard number on one side and a less number 

 on the other, and more even than this, we find more rarely specimens 

 havino- both antennae reduced in number of joints from the standard 

 of the species to which the individual undoubtedly belongs. 



The antennse are organs of some sense beyond that of the sonie- 

 what mechanical sense of touch, but what that sense is need not be 

 discussed at this time Whatever it is, the sense is indicated by and 

 resides in specially organized surfaces indicated by punctuation of a 

 peculiar kind, small fovea?, pubescent patches, or otherwise as is 

 familiar to those who go over the entire coleopterous series. 



The entire antenna is not sensitive. With very few exceptions it 

 may be stated, as a general rule, that at least three basal joints are 



