322 JOURNAL OF ECONOMIC ENTOMOLOGY [Vol. 10 



means an isolated instance of the confusion of these species. To 

 emphasize further the degree of confusion that exists in this group it 

 may be added that the material labeled "type" of one species de- 

 scribed from this state contains specimens of three species belonging 

 to as many different genera. Another species is based upon the male 

 of a Ceroputo and the female of a Pseudococcus and there are at least 

 two others which appear to be based upon two different species. 



The existence of the present unsatisfactory conditions affords 

 grounds for a justly severe criticism of the methods of systematic 

 entomology as they have been applied in this particular group. How- 

 ever, this is a question which it would be out of place to pursue further. 

 The present paper is concerned only with a consideration of the means 

 by which the existing conditions may perhaps be changed for the better. 



The present confusion may be ascribed to two factors, operating 

 both separately and together. 



Of these factors one is to be found in the continued use by coccidol- 

 ogists, as specific criteria, of the relative lengths of the segments of 

 the antennae, expressed in the so-called antennal formulae. The weak- 

 ness of these antennal formulae as diagnostic characters has been im- 

 plied or even openly admitted even by those who- have most consist- 

 ently clung to them and has been most conclusively demonstrated by 

 Smith^ who found that the formulae of ten antennae of each of five 

 species of Pseudococcus "varied as much as the specifically diag- 

 nostic formulae published for all the species of Pseudococcus." Yet it 

 is upon such characters as these that most authors have based their 

 new species and their discussions in regard to specific relationships. 

 Other characters usually given as supplementary to the antennal 

 formulae are equally non-significant, as was also pointed out by Smith 

 in the paper already referred to. 



Attempts have been made to refine the usual method of presenting 

 these antennal formulae by embodying them in graphs, but the result 

 is hardly satisfying for no clue is afforded by these graphs to the 

 specific identity of the individuals upon which the graph is based. In- 

 asmuch as two or even more species may occur together upon the 

 same host this objection is serious. Furthermore, the construction 

 of such graphs requires several individuals and does not aid in the 

 identification of isolated specimens. 



The effect of the use of these formulae has been not only to render 

 useless practically all the published descriptions but also, because of 

 the painfully evident variability of the antennae, to induce the belief 

 that the mealy-bugs are inherently variable in all their structures, 



' Smith, P. E. Specific Characters Used in the Genus Pseudococcus. Annals 

 Entomological Society of America, vol. 4, pp. 309-327. (1911.) 



