370 



sary to burden this paper with a repetition of the reasons 

 already stated which led me to those conclusions. I ex- 

 pressed the opinion that the generic names which I can ascer- 

 tain to have been regarded by Lacordaire and other authors 

 as synonyms of Heteronyx ought to be treated as follows : — 



Caulobius and Haplopsis as distinct valid Australian 

 genera. 



Homaloplia and Fhilochlcenia as not occurring in Aus- 

 tralia at all (the species which Lacordaire rightly regarded 

 as incorrectly referred to them are not Heteronyces^ as he 

 took them to be). 



Sericesthis as a valid distinct genus,^though possibly one 

 or two of the species Boisduval referred to it may have been 

 Tlcteronyces. 



SilojM and Hostilina as synonyms of Heteronyx. 



Melolontha as not occurring in Australia, although the 

 name was applied by early authors to some species of 

 Heteronyx. 



Cotidia (referred to above) also seems to be probably a 

 synonym of Heteronyx. 



I pass now to some general remarks on the 

 grouping of the Heteronyces in aggregates. I have 

 failed to find any character that will avail for this 

 purpose so as to produce "natural" groups, i.e., 

 groups the members of which can be rightly regarded as 

 on the whole more nearly allied to each other than they 

 are to the members of other groups. Whether the structure 

 of the antennae, or of the labrum and clypeus, or of the claws, 

 or of the coxae (in all of which organs the diversity of form 

 is very great and extremely interesting) be taken as the basis 

 of classification, the result is always that species closely agree- 

 ing in facies are widely separated, and if the species resemb- 

 ling each other in facies be grouped together, each of the 

 groups so formed is found to contain the widest possible 

 diversity of structure in almost every organ of the body. T 

 therefore arrive at the conclusion that, to me at any rate, 

 the grouping of the Heteronyces must be a mere matter of 

 convenience for the purpose of identification, and consequently 

 I content myself with the effort to place the species in aggre- 

 gates which will serve best to that end. This view of the 

 matter is similar to that which I expressed in my former 

 series of papers on the genus, and in the main my recent 

 study of the subject, founded on a vastly increased quantity 

 of material, has confirmed in my judgment the conclusions 

 I then set forth as to the method that it is best to follow 

 in forming these artificial groups. I have found, however, 

 that one important modification of my former scheme is 



