218 THE CONDOR Vol. XXV 



from Osoyoos Lake, Okanagan Valley, taken May 6, 10, 11, 15, 16 (1922), is 

 of exceptional interest fromi the standpoint of the variation exhibited. Hpre, 

 from the same spot, are breeding birds exhibiting dark coloration exactly like 

 Alaskan specimens of morphna, others with the ruddy color of morphna as 

 seen on Vancouver Island, and still others, gray of color and with sparsely 

 streaked breast, that closely approach merrilli. 



There is a single bird at hand (no. 9693, IMus. Vert. Zool., adult 

 female), collected at the head of Marten Arm, Boca de Quadra, Alaska, June 

 10, 1909. This specimen, taken in a region where extremely dark-colored 

 song sparrows are the rule, exhibits the extreme of gray coloration that can be 

 seen in morphna from any part of its range. It matches exactly three speci- 

 mens (nos. 3168, 2791, 2773, Victoria Memorial Mus.) from Midway and Trail, 

 in extreme southeastern British Columbia. These three have been labeled 

 merrilli, and, indeed, iii appearance they and the Boca de Quadra skin ap- 

 proach, very closely to that subspecies. 



I have examined the type specim'en of Melospiza melodia inexspedata 

 Riley (Proc. Biol. Soc. Wash., xxi , 1911, p. 234), together with several 

 other specimens from the general region of the type locality and I cannot re- 

 gard that subspecies as worthy of recognition. It is indistinguishable from 

 morphna. Oberholser (Auk, xxxv, 1918, p. 187), urging the validity of the 

 race, says: "It is miost nearly allied to Melospiza melodia rufina, but is de- 

 cidedly smaller, particularly in so far as the bill and wing are concerned; and 

 is, in fact, a smaller, darker, somewhat heavily streaked edition of that bird." 



As regards color and markings, specimens of " inexspiectata" can be 

 duplicated froml the Alaskan coast. Furthermore, individual variation in the 

 rufina group is so extensive, as I have shown, that it is not practicable to 

 make the division indicated. As regards measurements, the appended table 

 shows the similarity of " inexspedata" of the interior and morphna of the 

 coast. 



In British Columbia there remains to be considered the status of 

 Melospiza melodia merrilli Brewst*. I cannot find that this name has been 

 published as pertaining to a British Colurribia bird, but I have examined 

 specimens in the Victoria Memorial Museum that were so labeled by different 

 authoriti(!s. These are apparently some of the same birds listed by Macoun and 

 Macoun (Catalogue of Canadian Birds, 1909, p. 540) as Melospiza dnerea 

 montana, taken at various points in southeastern British Columlbia. Speci- 

 mens from that section, froml Midway, Trail, and Pernie, for example, do cer- 

 tainly approach merrilli pretty closely, though I have seen none that are as 

 gray as the type of that subspecies. I have no grounds for discrediting the vali- 

 dity of merrilli as occurring elsewhere, but it is not practicable on the basis of 

 these variable specimens to give the race formal standing in a list of the birds 

 of British Columbia, with any definitely outlined range in that province. 



There are at hand seven specimens (including the type) of the type 

 series of Melospiza dnerea phaea Fisher (Condor, iv, 1902, p. 36) from the 

 coast of southern Oregon. These birds average of small size, but all things 

 considered, I do not regard them as representing a recognizable subspecies. 

 Such variation from the mode of morphna as is seen, both in color and 

 measuremjents, appears to me to be in the nature of intergradation toward 

 rleonensis, the adjoining subspecies to the southward. 



