to limitations of the Milllerian Hypothesis of Mimicry. 1 23 



that dardanus possesses nauseous qualities ; and if this 

 be so, the argument for Diaposematism falls to the 

 ground. 



It is not always easy to form a definite conclusion as to 

 the relative numbers of a model and its mimic, but there 

 can be little chance of error in the present instance. We 

 may first note that if the $ dardanus exhibits this angular 

 marking in such of its forms as hippocoonoides or trophonius, 

 this can have no mimetic influence upon A. echeria, because 

 those forms do not at all resemble this species, but mimic 

 two other Danaines. For this reason we must only con- 

 sider the occurrence of this angulation in the ccnea form 

 which mimics echeria. We shall find however that that 

 character occurs more rarely in cenea than in the above 

 forms, and then nearly always in the examples which are 

 least like echeria. Therefore in estimating numbers, we 

 have on one side a comparatively rare variation of a single 

 form of only one sex of the Papilio, and on the other side 

 both sexes of two very common species of Amauris. From 

 what I know of the prevalence of these two insects in 

 nature it would be a conservative estimate to reckon that 

 the latter would exceed the former in the ratio of 100 

 to 1. It has already been shown how impossible it is to 

 believe that in such circumstances the mere operation of 

 the Mtillerian factor could have compelled the Amauris to 

 mimic this variation of the Papilio. Here again the facts 

 appear to entirely forbid a diaposematic interpretation. 



W T hen we investigate the occurrence of this angular 

 . ... 



marking in other species of Amanris and Papilio, still 



further difficulties present themselves. With one or two 



exceptions all the species of Amauris have a large basal 



or sub-basal pale patch on the hind-wings. Out of 20 



of such species I find that no less than 13 possess the 



angulation in question. In psyttalea and dominicanus it 



is variable in its occurrence ; in such forms as lobengula 



and craioshayi it is very similar to what we find in 



Pap. dardanus ; in ochleides, hyalites, etc., it is more 



marked ; whilst in echeria, and albimacidata it is far more 



developed than in any Papilio. On the other hand, this 



character is a very unusual one in the genus Papilio; 



indeed I have failed to find anything which is really 



comparable with the Amauris pattern except among those 



Papilios which tend to mimic Danaines in which it already 



exists. These facts, again, are quite at variance with what 



