Nov., 1923 STATUS OF SOME NORTHWESTERN SONG SPARROWS 219 



The one outstanding fact brought out by an adequate series of specimens 

 of the rufina group is the relative uniformity of characters presented over a 

 tremendous area; and this despite the fact that there is a wide range of varia- 

 tion among birds from any one locality, for the same sort of variation crops up 

 almost everywhere. This uniformity in the British Columbian andi southeastern 

 Alaskan song sparrow is in marked contrast to the manner in which the species 

 breaks up into local races to the northward and to the southward. To the north 

 are the several more grayish-colored Alaskan races, exhibiting well-defined 

 characters of color and structure, and each occupying a definite and relatively 

 circumscribed habitat. To the southward, Melospiza melodia again breaks up 

 into rather sharply defined races, mostly of limited distribution. 



To some it may seem that the proper procedure in handling the rufina 

 group would be to recognize four subspecies where I have defined two. These, 

 centering about extremes that are admittedly appreciable, would be (1) the 

 ruddy bird {morphna) of Vancouver Island and the adjacent mainland; (2) 

 the grayish-toned merrilli extending into southeastern British Columbia; (3) 

 a darker-colored race {inexspectata) from the northern mainland and part of 

 the Sitkan district, Alaska; (4) a larger, dark-colored race (rufina), with 

 range as previously outlined in this paper. 



These, however, certainly are not definable races in the same sense 

 as caurina, kenaiensis, or sanaka, to the northward, or cleonensis, samuelis, 

 pusillula, and others to the southward. Their characters are hard to define, 

 and, as previously shown,' individual variation in nearly all sections is so great 

 that there would be a large proportion of specimens in any series that could 

 not be subspecifically allocated on any ground but the geographic situation of 

 the place of capture. 



It may be said that caurina and the related gray-colored subspecies to 

 the northward, the blackish California subspecies to the southward, and 

 what might be termed the melodia group to the southeastward, each represents 

 a section of the species comparable to what I have here called the rufina 

 group. Within each of the three first mentioned divisions, local differentia- 

 tion has proceeded to a point where several races, more or less sharply defined 

 and of restricted range, can readily be recognized. In the rufina group 

 we can see a strong tendency toward such subdivision, but, save between the 

 very extremes, it has not advanced far enough to permit of ready definition 

 of characters in the birds from different parts of the general range. 



It seems to val/d that in any study of a variable bird like Melospiza 

 melodda, stress should first be laid upon the major divisions of the species. In 

 the present^ case this would be on the reddish-colored rufina group in its 

 entirety, as compared with the group formed by the gray-colored A.laskan races 

 occurring abruptly to the northward, the brownish-colored melodia group (con- 

 nected through merrilli) to the southeastward, and the . blackish-colored 

 California subspecies (connected through phaea) to the southward. It seems 

 to me that it would be well to regard observed differences within each of these 

 larger divisions in a somewhat different light from those by which these 

 same divisions are separated from each other. 



In the mechanical handling of names, in our efforts to convey observed 

 facts, it seems to me that we are making a mistake when, for example, we 

 publish lists of names as the song sparrow names appear in our Check-List 



