470 VvoW K. H. Poiilton on the Jtimdic X. American 



the luiul-wing iqipor surt'aoo in tho tuning foinale is in 

 nearly tho same position as the one conspicuous spot of 

 ixMerivii and troihis, while at tho apical angle of the same 

 surface of the same wing', a large red spot appears in (iirinia 

 and f roil us. Rod spots arc not i'ound in either position on the 

 upper surl'aco of phi/ oi or. Those spots arc certainly ances- 

 tral in tlie tnrmu'i fonialo, inasmucli as they ai"e present in 

 the non-mimetic female form and the non-mimetic male of 

 the same species, as well as in the allied non-mimetic species. 

 For the same reason the black-pupilled red spot at the anal 

 angle of a.^criits is ancestral. In t roil us, on the other 

 hand, both red spots are probably of recent origin on 

 the upper surface, and have been developed in relation 

 with tlio mimetic appearance. They are yellow in the 

 ancestral p?/(r;»C(i?(>'. It is probable that their red colour 

 in troiltis has boon developed in secondary mimicry of 

 aster ins and the t urn us female form of (/laucus. It is in 

 accordance with this interpretation that tho rod spot at 

 the anal angle of troihts, although without the black pupilj 

 bears considerable superlicial resemblance to that oi'asferius, 

 and that tho red spot at the apical angle is especially well 

 developed in tho female. 



The comparisi>n of these three mimetic forms also yields 

 evidence o\' an approach towards tho prin\ary model, in 

 which the ditl'erent species have made ditforeut rates of 

 progress, presumably determined in large part by the age 

 of the mimetic resemblance. 



Fajnlio troilus is evidently the oldest mimic. The 

 mimetic likeness, extending to both sexes, has boon derived 

 from a very dilierent appearance still preserved in pala- 

 medcs. The suppression in troilus of tho tifth orange-rod 

 spot of the inner row of j; >«/<?/« <(?('5 is apparently an advance 

 in the direction of the open loop of spots which is the 

 prominent feature in the hind-wing under surface of 

 pJiilenoi'. Equally clear advance is seen in tho bluish tint 

 which tho sulnnarginal yellow spots of the upper surface of 

 2niiafiiedes have gained in troilus. 



I^apilio jiolj/.irnes aste^'ivs is less perfect and presumably 

 less old as a mimic than troilus. The mimetic resemblance 

 is found on the under surface of both sexes, but on the 

 upper tho male, if a mimic at all, has only reached an 

 early stage in the resemblance. Tho evolution of the 

 two rows of dotinito orange-rod spots out of the ill-defined 

 elements found in the non-mimetic ancestors, was probably 



