1896.] METALLIC COLOUES Or BIHDS. 291 



this apical band, until such a metallic feather as that of flg. 1 is 

 produced. The course of development of the metallic edging of 

 quill-feathers may be described in a little more detail. In order 

 to illustrate this, it may be convenient to describe iu a concrete 

 example the differences between the sexes in the coloration of 

 wings and tail. The following notes were made on a female of 

 Anihreptes malaccensis and a male in nearly completed moult, but 

 the characters of the adult male were checked by reference to 

 Shelley's ' Monograph of the Sun-birds.' In the male the tail was 

 composed of black feathers with an edging of metallic violet, which 

 was widest in the caSe of the two central feathers. In the wings 

 the lesser wing-coverts had a broad transverse band of metallic violet, 

 the median coverts a similar band of dark brown, and the greater 

 coverts a longitudinal band of olive-yellow, which becomes brown 

 in a completely adult male. The wing-quills themselves were 

 greyish brown with olive edges. In the female the tail was dark 

 brown M'ith an edging of olive-yellow. The wing-quills and wing- 

 coverts were ashy grey with longitudinal or transverse bands of 

 olive-yellow distributed in the same way as in the male. In the 

 female the general contour-feathers had a broad transverse band 

 of olive-yellow, while in the male most of these feathers bad been 

 replaced by others with transverse metallic bands. It must be 

 noted that in all these cases the olive-yellow part of the feather 

 has a peculiar looseness of structure visible even to the unaided 

 eye. It should also be observed that the yellow edging to the 

 quills is seen both in the rectrices and remiges of the female in 

 this as in numerous other Sun-birds, while in the male the edging 

 is replaced in the case of the rectrices by a metallic band. The 

 respective distributions of longitudinal and transverse bands should 

 be especially noticed, as showing how very closely the nature of 

 the stripe depends upon the nature of the feather, that is upon its 

 elongation. 



From the above description it is obvious that some sort of relation 

 exists between the olive-yellow margins of the feathers of the 

 female and the metallic margins of the feathers of the male. 

 In general, we may say that there is a tendency for the feathers 

 with olive margins in the female to be replaced by feathers with 

 metallic margins in the male. It will be noticed that the change 

 is associated with increased pigmentation in the male; iu the 

 median wing-coverts there is only slightly increased pigmentation 

 without metallic colour. Such a tendency is very widely spread 

 in the family, but the extent of replacement differs greatly. Thus 

 in the species described above the olive edging of the greater 

 wing-coverts and wing-quills is not replaced by a metallic edging 

 in the male. In Nectarinia famosa an olive edging in the same 

 feathers in the female is replaced in the male by a metallic edging. 

 In Anilirobaphes violacea the tail-coverts are edged with yellow in 

 the female and in most males : according to Shelley, some males as 

 an individual variation have this edging metallic. It is, however, 

 needless to multiply examples. Enough has been said to justify 



19* 



