On the Nestbuf o/Cassicus persicus, ^c. 301 



castiug of a disguising film ^ill not account for the second 

 operation. 



In Pyromelana the change of plumage is very slow; the 

 feathers daily gain in intensity, the pale buff of the under- 

 parts getting scarcely perceptibly deeper, until at length the 

 velvet-black and fiery orange in P. franciscana appear as 

 mere spots or shaft-streaks, which gradually expand fanvvise 

 towards the outer fringes of the feathers. This spotting, 

 however, is very uneven, some feathers being developed 

 in advance of others, so as to give the bird a very patchy 

 appearance. In the bright yellow and black plumage of 

 P. afra this is even more noticeable. 



At the change of plumage the flank-feathers and upper 

 tail- coverts are moulted out, being replaced by long soft 

 feathers, which droop over and almost hide the tail : but 

 none of the feathers of the head, back, breast, and belly are 

 lost; they simply undergo a gradual change of colour. 



If it is possible, and we know that it is, for the plumage 

 of birds to be seriously affected after death, there is no 

 reason for asserting that a perfect feather possesses no 

 vitality, and is therefore incapable of change of colour. 



Perhaps one of the most marked alterations in coloration 

 after death which 1 have noticed is that which takes place 

 on the breast of the male Gouldian Finch {Poephila 

 mirabilis). In life the breast is vivid ultramarine-blue or 

 very bright pansy-blue; after death the blue gradually 

 fades out of the feathers, leaving them of a dull lilac. 



XXVIII. — On the Nesting of Cassicus persicus, Cassidix 

 oryzivora, Gymnomystax melanicterus, and Todirostrum 

 maculatum. By Dr. Emil A. Goeldi, C.M.Z.S., Director 

 of the Museum in Parii.* 



1. Cassicus persicus and Cassidix ouyzivora. 

 The nests of most Brazilian birds are by no means easily 



* [For a previous article on a similar subject by I)r. Goeldi, see 'Ibis,' 

 1896, p. 299, and observe that the editorial I'ootnote at the couimencement 

 of that article is intended to refer to Koenig-Warthausen in J. f. O. 1868, 

 and not to Dr. Goeldi's excellent notes. — Edd.] 



SEU. VII. VOL. 111. 2C 



