10G PROCEEDINGS OF THE ACADEMY OF 



b. Melanistic j^hase. 



Adult : Prevailing color plain blackish-brown ; the tibite, lining 

 of the wings, and sometimes the breast, inclining, more or less, 

 to rufous. Crissum usually white, sometimes immaculate, gener- 

 ally barred with rusty or blackish ; occasionally with dusky and 

 whitish bars of equal width. No white on the throat, or else but 

 little of it. Young: Brownish black, variegated with ochraceous 

 spotting, in amount varying with the individual. 



Remarks. The above diagnosis is based upon upwards of one 

 hundred specimens, and ma}' be considered as covering every 

 phase, seasonal, sexual, and individual, assumed by the species. 

 So great is the extent of individual variation, that, taking a series 

 of specimens of either plumage, scarcely two can be found which 

 are alike in all their markings, especially those on the lower sur- 

 face of the body. 



The plumage of the upper surface, however, is so constant, that, 

 taking ever so large a series of adults, and turning them on their 

 bellies so that the backs of all are presented to view at once, a 

 specimen in the least degree aberrant is very rare, while one very 

 noticeably different can scarcely be found ; and when such a one 

 is detected, it is generally found that the difference is a seasonal, 

 rather than an individual, one. Thus, in specimens which have 

 just completed the new moult, some of the longer scapulars are 

 seen to be slightly variegated with oehraceous on their edges ; 

 while in those in which the feathers have become old, these paler 

 mottlings and edges are " worn" off, causing a very uniform 

 shade of dark grayish-brown to prevail continuously over the 

 upper parts, though each individual feather fades gradually in 

 depth of color on its borders. As a rule, the brown of the upper 

 parts is a shade lighter and more grayish in the males than in the 

 females, and there is apt to be a slight rufous edging to the 

 feathers of the neck, as well as a tinge of the same on the upper 

 tail-coverts ; while not unfrequently the longer scapulars have 

 their exterior edges tinged with rufous. But these differences 

 between the sexes are so exceedingly slight as to be appreciated 

 only upon actual comparison of specimens, and even then do not 

 prove sufficiently constant to be considered of any importance. 



Reversing the position of the specimens, so as to present their 

 lower surfaces to view, they can be immediately placed into two 



