IN DEFENSIVE COLORATION 339 



It will be admitted that the above account may be taken 

 as typical of the way in which Mullerian associations 

 are scattered over a country, and of the changes which 

 are witnessed as we follow them from one area into 

 another. There is nothing to lead us to believe that 

 progressive unification will occur on any given area. 

 Each tract has developed its own set of important com- 

 binations. The recognition of these, and a memory of 

 the implied qualities, is the climax of one important side 

 of the education of an insect-eating animal. 



5. Seasonal Transition from Cryptic to Synaposematic 

 Defence. — The astonishing discovery by the irrefutable 

 evidence of breeding, that Precis natalensis is the wet 

 season form of P. sesamus, is briefly described on p. 208. 

 The latter forms, with a highly procryptic under surface, 

 were bred by Mr. Marshall from wet season individuals, 

 in which the conspicuous colours and pattern of the upper 

 surface are reproduced, considerably heightened in effect, 

 upon the under surface. It is particularly interesting 

 that upon both surfaces this wet season form should bear 

 a general likeness to some of the larger red and black 

 Acraeas found in various parts of its wide range. 1 

 Although the shape of these broad-winged butterflies is 

 very different from that of an Acraea, I do not doubt 

 that Mr. Marshall is correct in considering the resemblance 

 roughly mimetic. If we take different habits and modes 

 of flight into account, it is difficult to believe that the 

 resemblance would deceive a bird at any distance. It 

 might well, however, put the bird on its guard and lead to 

 a cautious attack during which a special defence would be 

 recognized by the enemy. Hence it appears to be more 

 reasonable to regard natalensis as a Mullerian than as 



1 The present writer was in error in drawing a distinction between the 

 basal ivhite-marked black patch of the under side in natalensis, and the 

 black-marked light-coloured patch which occupies the same position in 

 Acraeinae. Although the foregoing is true of the great majority of 

 African Acraeas possessing the marking, the particular large species (e.g., 

 A. acara and A. anemosa), to which we must suppose that natalensis 

 bears a general resemblance, exhibit a white-marked black basal patch. 

 Compare Tra?is. Ent. Soc, Lond., 1902, pp. 424-8, where the upper and 

 , under surfaces of natalensis are contrasted in detail. 



Z 2 



