CHARLES W. METZ. 93 



with P. modeshis in genitalia, and yet is quite different in 

 sculpture. A more striking example is that of P. fossata 

 n. sp. which is exceedingly different in sculpture, and even 

 in the proportions of the thorax, from P. nwdeshis, and yet 

 has similar genitalia. It is noteworthy, also, that in each of 

 these cases the face marks are very similar. 



With such characters in the males, then, there is little 

 chance for confusion in that sex, and very little excuse for 

 describing a new species that is not easily recognizable by 

 either external or genital characters. It should be a rule in 

 this genus that to warrant specific separation an insect should 

 have distinct and characteristic genitalia, or very strongly 

 marked and thoroly diagnostic external characters, or else 

 characters that are shown by a large series of specimens to 

 be reasonably constant, otherwise confusion will almost cer- 

 tainly result. It should also be made a rule to never de- 

 scribe a new species from a female specimen. The types 

 should invariably be males, and the females afterwards asso- 

 ciated with them, if anything like satisfactory work is to be 

 done ; for the descriptions of females are almost always such 

 that no student could recognize them with certainty. In the 

 material accumulated by the writer are a number of undeter- 

 mined females which could be described as new, but to do 

 so would simply add to the difficulties in the genus, so they 

 will remain among the indeterminates until the males are 

 found. Likewise in the list of uncertain described species 

 at the end, will be found a number of females, most of which 

 might be referred to almost any of the known species from 

 the descriptions. As long as there are so many species hav- 

 ing females in which there are no really plain, easily recog- 

 nizable characters, it is pure folly to describe a specimen as 

 new which could in any way be confused with others, and 

 those females which cannot be so confused are very rare. 

 These difficulties will become more apparent when the de- 

 scriptions of the females included in the present paper are 

 examined. That they are unsatisfactory is certain, but so 

 long as there is such great variation in the few diagnostic 

 characters that are available the trouble will remain. At 

 present it has seemed best to describe a typical female in 



TRANS. AM. ENT. SOC, XXXVII. (12*) 



