I "^S CiiADBOURNE, Spring Pluiuagc of the Bobolink. [April 



taken in June 1895, escaped by tearing a hole in tlie cloth top of 

 the cage. However, through the kindness of Messrs. Allen, 

 Batchelder, Brewster, Merriam, and others, I have been able to 

 compare a series of more than 175 male Bobolinks, probably 

 covering about the same localities and dates as the material used 

 by Mr. Chapman in the preparation of his articles on this 

 subject.' Among the skins loaned by the American Museum of 

 Natural History was the specimen from Corumbd, Brazil, on which 

 the hypothesis was originally based that the black and 7vhite dress 

 was entirely due to ' moulting ' luithout any ' color-change ' in the 

 individual feathers. The assumption being that ' moulting ' is 

 conclusive proof of the absence of any color-change, — a supposi- 

 tion which has been frequently used by writers on this subject, 

 but one which is absolutely without proof of its correctness as yet, 

 while it is shown to be untenable by examination of the CorumbA 

 bird, as will appear later. 



First as to my pet Bobolink, which was kept from January until 

 the breeding plumage was complete : — The bird always seemed 

 well and strong, and the color-change teas not accompanied by any 

 increase in feather-loss, i. e., not greater than during the winter and 

 often for several days in succession there were no cast-off feathers 

 at all to be found. The total during the three weeks that the 

 change was in progress was thirteen, — namely two broken 

 rectrices and eleven contour feathers. It is hardly possible that 

 any stray specimens were unnoticed, for even had they fallen 

 outside of the cage they would have been found in the room, and 

 a wire netting protected the window. ' Pin-feathers ' could hardly 

 have been overlooked, if present ; for I often held the bird in my 

 hand and carefully examined it, blowing back the plumage until 

 the skin could be seen. It is also safe to say, doubtless, that the 

 cast-off feathers were not eaten by the bird itself. Hence it 

 follows that unless the previous plumage was made up of only two 

 tail and eleven body feathers, both of the former on the same 

 side, — which was certainly not the case, — my Bobolink was 

 unquestionably an instance of color-cha?ige in the phunage ivithout 

 '■moult.' 



' Cf. Auk, VII, 1890, pp. 120-124 ; also ifiid., X, 1S93, pp. 339-341. 



