64 Director' s Annual Report. 



whistle of five or six clear notes, which I was only partially able 

 to reproduce. The bird, however, came nearer and finally lit in 

 the thick branches of a tree fift_y feet or so below me. In ni}- haste 

 to get down out of the tree I was in, some small limbs gave way. 

 and I came clattering down quite a distance before I was able to 

 regain my footing. That the disturbance had frightened the Hoa 

 I felt quite certain, and was therefore agreeably surprised to find 

 that through it all it had not moved from the tree where it had 

 first lit. It seemed to be willing to be studied, an opportunity for 

 which I was exceedingly grateful. 



This bird, as well as the two former specimens, confined its 

 range to the undergrowtli, several times coming down to within 

 three or four feet of the ground. At no time did it make a long 

 flight or alight on the top of the trees. As it was raining all the 

 while the bird was especially atftive in preening and shaking its 

 feathers. The trees and vines w'ere everywhere covered with thick 

 wet moss, and although the bird hopped about from branch to 

 branch, carefully inspe(5ling each limb, I did not see it catch any 

 insects, or even probe into the moss. Hopping from tree to tree, 

 it worked its way around the head of the little side valley, up 

 which it had come in answer to my call, to where a large purple- 

 flowered lobelia was in profuse blossom, and began to feed. The 

 ease and grace with which the feat was acomplished was indeed 

 interesting, and left no doubt in my mind as to one of the probable 

 cau.ses of the remarkable development of the tongue and bill. 

 The tongue was inserted with great precision, up to the nostrils, 

 in the flower, while the bird balanced itself on the branches, as- 

 suming almost every imaginable attitude in its operations. In all 

 three of the birds secured, the crown was smeared with the sticky 

 purplish white pollen of this lobelia. I had a preconceived idea 

 that the bird would also feed on the flowers of the wild banana. 

 This conie(5lure I was not able to prove or disprove by my ob- 

 servations, further than that in each case no bananas were to be 

 seen in the valleys below or anywhere in the vicinity w^here the 

 birds were secured. 



The third specimen, like the two preceding it proved to be a 

 male, in perfect plumage. The bodies of the last two and the 

 stomach of the first one — its bodv being badly mangled — are pre- 



[154] 



