H. C. FALL 307 



the case in a recent review of our species of Diplotaxis, so in 

 Pachyhrachys there seem to be ahnost no characters which afford 

 the means of divichng the genus into natural groups. As in 

 Diplotaxis, so here, a small number of species — less than one- 

 tenth of the whole number — are separable with certainty by- 

 reason of the pubescent upper surface. Two other species — 

 hepaficus and microps — are differentiated by good characters, 

 but the great mass of species seems incapable of further reduction 

 in this manner. In this great complex of species I have been 

 forced to adopt color as a basis for primary grouping, and this 

 notwithstanding the fact that color — in its smaller details — 

 is the most variable of all the characters used. In the l)roader 

 sense, however, it is quite serviceable, and it is usuall}- possible 

 for even the inexperienced student to decide whether his speci- 

 men should be looked for among the black, the yellow, the vitiate 

 or the maculate species. The great difficulty comes of course in 

 the last named group, for here the variation may be so great that 

 individuals of the same species may be referred to either the 

 black, variegated or yellow groups. This difficulty has been in 

 large measure overcome by tabulating in more than one group 

 those species known to be except ionably variable. It has 

 manifestly not been possible to provide for all contingencies of 

 this sort, and it would therefore be well if the student doss not 

 find his species in the group in which it would appear to belong 

 to try the next most probable group before giving up the search. 



In the reading of the specific descriptions which follow, it 

 must be remembered that there are no fixed characters; every- 

 thing wdthout exception is subject to individual variation. The 

 study is really a most difficult one and the student nuist not 

 expect to be able at all times to identify uniques, expecially if 

 they be females. Even with a good series and considerable ex- 

 perience the problem will often be difficult enough. 



It is hoped that the short diagnosis before the more detailed 

 descriptions will prove of service by enabling the student to 

 very quickly decide upon the possibility, or at least probability, 

 of that particular species being the one in hantl. The average 

 length given in the short diagnosis is in general that of either a 

 large male or a small female, and of course will serve only as a 

 general guide, the actual known limits of size being given at the 



TRAXS. AM. ENT. SOC, XLI. 



