43 



heather would, under the fostering shelter of the trees, become more 

 robust, growing and spreading in all directions, and gradually cover 

 nearly the whole of the rocks, in consequence of which repandata 

 would generally rest on the tree-trunks. Arboreal insect-eating birds 

 would take up their abode in the young woods, and then commenced 

 a struggle for existence between the varieties of repandata and 

 that form best adapted by reason of its protective coloration, to 

 exist under the altered nature of the locality would be preserved. 

 Of course the best protected, and therefore predominant form of 

 to day {conversaria), did not acquire all at once the distinctive 

 pattern of wing ornamentation we now find in this insect. As the 

 nature of its habitat gradually changed from a treeless and rocky 

 wilderness into a well-timbered wood with a luxuriant undergrowth, 

 so first one, and then another of the varieties of repandata would 

 be in the ascendant, but the tendency of the variation throughout 

 must have been towards the character of marking now so fully 

 developed in conversaria. In evidence of this we have the fact that 

 the majority of the more or less typical repcndata exhibit either a 

 strong outline or a faint trace of the band of conversaria. 



There are many other places in England where the conversaria 

 form of JBoarmia repandata occurs from time to time, but not in 

 such numbers as to threaten to supplant the type. From one of 

 these localities (Bristol) I received part of a brood of larvae hatched 

 from eggs deposited by a banded female. The nineteen larvae 

 produced ten typical repandata, and nine of the form conversaria. 

 Unfortunately nothing of the male parent was known, but probably 

 as conversaria only occurred sparsely in the locality, the male was of 

 the typical form. 



Types of some species of Butterflies and their named varieties 

 are sometimes only slightly differentiated in colour and markings of 

 the upper surfaces of the wings, but the under sides of the wings 

 show considerable difference, for instance in the case of Cceno?iy/npha 

 typlwn and its varieties la'idion and philoxenus. A pair of each of 

 these insects are shown. On reference to the var. la'idion it will be 

 observed that the sub-marginal spots of the under side are either 

 very small or entirely absent. This form occurs in Ireland and 

 Scotland, but does not appear to have occurred elsewhere. The 

 variety philoxenus is only found in the North of England, and is 

 peculiar to the counties of Cumberland, Durham, and Yorkshire. If 

 you compare this form with the type it will be found to possess 

 an extra spot on the upper surfaces of the superior wings, situated 



