VARTATION. 



101 



( Costal Vjslt 



\ she'«inq posiUon &. 



^ usual sVisipe . 



(DiscoidaV lunulas 

 :\or veivilets 



.?.. 1?> SuVjsjpical spot 

 " ( nonnal mrarlonq ) 

 Z^SuVjapical spot 



I will commence Avith what appears to be the most important 

 variation, viz., the development of red scales in the posterior wings, 

 most often to be seen in the sul)marginal yellow lunules ; there are six 

 of these lunnles, and it is no uncommon thing to see nearly a whole 

 series (British) showing more or less red scales, at any rate in the first 

 one from the costa, indeed it is rather rare than otherwise to find a $ 

 entirely without it, it is less frequent in J s and less conspicuous. It 

 Avould be well here to mention that throughout this paper, the statistics 

 I give are, unless notified as otherwise, from a series of forty specimens 

 in my own cabinet, twenty males and twenty females, they were 

 selected somewhat promiscuously from a number of others, several years 

 prior to my taking an interest in the variation of the species, quite 

 independently of the development of any i)articular marking with the 

 exception of the " celldivided " varieties, and it is only a peculiar 

 coincidence that they are erpially divided lietween tlie two sexes ; all 

 are from Wicken Fen and all are bred with the exception of two " cell- 

 divided " ones. Of tliese forty specimens, the red appears in one or 

 more lunules of 19 Js and 18 ^ h ; it is most often in the first or 

 costal lunule in which it is always conspicuously larger than in any of 

 tlie others; in specimens in which it is present in more than one lunule 

 it diminishes in size in each one from the costa, but there is more 

 difference in size between the first and second than the second and tliird. 

 The lunules which have it most rarely are the 4th and 5tli, only two 

 specimens have it in the 6th and three in the 4:th, and these only to a 

 very small extent : I have prepared the following table to show exactly 

 how it is placed in the forty specimens. The top numbers represent 

 the submarginal lunules, commencing from the costa of the wing-=. 



