dark reflexed scales, broadest at the base, and extending one-third of 

 the length of the anterior margin. I have before me eighteen males 

 of this species, in every one of which this structure is distinctly visible. 

 My continental type of purgatana also has it, but this, I think, is only a 

 small form of maculosana. I think that if Wilkinson had noticed 

 this structure, he would have felt some perplexity about his family 

 PlicatcB. 



It thus appears to me that the costal fold is clearly proved to be 

 of no value as a family characteristic, and when we find the typical 

 fold of Lozotcdnia Podana and costana associated with the folded and 

 rolled margins of rohorana^ xylosieana, and rosana, and with the false 

 folds, with or without rolled margins, in unifasciana, Lofaiiryana, and 

 musculana, and then closely allied to them the non-folded species, such 

 as dumetana and diversana, which have also the rolled margin, the 

 raised margins in others, and in many other groups, the barely per- 

 ceptible folds in Pilleriana, candidulana, and E. maculosaoia, and, 

 finally, the intimately close alliance between folded and non -folded 

 species of Dichrorampha, the conclusion seems forced upon us that 

 this costal fold is equally valueless as a generic character. 



After studying this last Dich^orampha-Endopisa group of species, 

 I was almost inclined to doubt its value even as a specific character, 

 but in the course of careful consideration of nearly all our native 

 species, I find its presence or absence, as the case may be, so thoroughly 

 reliable in each species in other groups, and the form which (when 

 present) it takes so true and accurate in the individuals of each species, 

 as to render it highly improbable that a character so reliable in all 

 other species should be variable and uncertain in the single group of 

 the BicJirorampha. So that although it is almost impossible to find 

 reliable characters by which to distinguish the different forms which 

 I enumerated at the commencement of this article— Nos. 1, 2, 3, 4, 7, 

 and 8 — it will be necessary to search for such characters, and to deter- 

 mine if possible at some future time, the name and description which 

 fit most accurately to each. This task will be all the more difiicult 

 from the fact that scarcely any of the published descriptions are 

 minute enough to meet the case, but will generally serve for several 

 of these forms almost equally well. 



68, Camberwell Grove, S.E. : 

 April, 1885. 



