510 J. A. ALLEN ON BIRDS 



186, = T. Auduboni Baird, Review Am. Birds, 16), to the synonyms of T. Pallasi. My reason 

 for this is that a large series of specimens of T. Pallasi and T. fuscescens from the Eastern 

 States will usually afford those that in every respect match specimens of T. "nanus" and T. 

 "uslulalus," labelled as such at the Smithsonian Institution, with every possible grade between 

 these and the more common type. The variations in size and in the proportion of parts that 

 have been pointed out, it is easy to show are but irregular individual variations, which, as 

 already noted in the case of T. Swainsoni, do not accord with the variations in color; moreover, 

 they are only such variations, differing neither in character nor amount, as specimens of 

 almost any undoubted species, taken from a single locality, will present. 1 Hence the only 

 constant distinction is that of the degree of intensity of the rufous tint, which, we should 

 particularly note, varies alike throughout the plumage, being more or less intense above 

 as it is more or less intense below ; since the number and size of the spots which have been 

 alluded to as distinctive features are evidently of very little value, if of an}', — these, in birds 

 having this style of coloration, being the most inconstant of characters, as a reference to 

 the Dendrcecce and to the various genera of streaked sparrows so well shows. In accordance 

 with the degree of intensity of the rufous, we find the species in question naturally falling 

 .into two groups : the first composed of T. Pallasi Cab., T. fuscescens Steph. and T. Swainsoni 

 Cab., characterized by a very marked degree of rufous suffusion throughout the plumage ; 

 and the second, of T. nanus Aud., T. Auduboni Bd., T. astulatus Nutt. and T. Alicice Bd., and 

 characterized by their general almost entire absence of buff or rufous tints below, leaving 

 this aspect lighter or more ashy, and the purer olivaceous color of the dorsal aspect, due to 

 a like diminution of the ferruginous. In regard to an explanation of this difference in color, 

 it seems to some extent to be merely an irregular individual variation, but a careful general 

 examination of the subject shows that young birds are prone to a suffusion, sometimes of 

 rufous, sometimes of plumbeous, olivaceous, or yellow, the color varying in different species, 

 from which when fully mature they are generally nearly or quite free. As is well known, 

 it is a pretty constant feature of young birds in their first autumn, and even often serves 

 to distinguish specimens of the second and third years among the song-birds, and still later 

 among some of the marine and rapacious, from those of greater age. In the Oscines gener- 

 ally, we have good reason to believe that after a certain period birds marked by unusually 

 bright tints, as in Pinicola, Ourvirostra, Carpodacus, Pyranga, etc., the intensity of the color de- 

 clines with the increase of age, its maximum seeming to be when the birds first fully reached 

 maturity. This is farther borne out by actual observation in cage birds ; and it accords 

 with the general law in the animal kingdom of the waning of color with the decline of life. 

 We are hence led to believe that the comparatively small proportion of birds coming under 

 the second group are mainly very old birds, while specimens coming under the first, em- 

 bracing the great majority, represent the normal or usual condition of plumage. This view 

 is further borne out by the fact that young birds of T. fuscescens, raised in Massachusetts, 

 as I have repeatedly observed, as well as autumnal specimens of T. Swainsoni and T. Pallasi, 

 known from other characters to be such, present this rufous suffusion at its maximum of 



1 An authentic specimen of T. nanus (No. 45,904, Smith, iinens differ only in the rather larger bill of the latter, and its 



Inst. Cat.) collected at Sitka in May 18GG, by F. Bischoff, has more rufous plumage, while the genera! size of the T. Pallasi 



the wing measuring 3.55, the tail 3.00, the culinen .30, and specimen is less than that of the T. nanus, showing the in- 



the tarsus 1.10; while an authentic specimen of T. Pallasi constancy of the "smaller size" supposed distinctive of T. 



(No. 31,414, Smith Inst. Cat.) from Racine, Wis., has the nanus. 

 wing 3.45, the tail 2.80, culmen .35, and tarsus 1.10; the spec- 



