ORNITHOLOGICAL EXPLORATIONS. 305 



mary. The latter two distinctions are formulated by Seebohra (Cat. 

 Brit. Mus., V, p. 38) thus : " Exposed portion of bastard primary 0.3 to 

 0.35 inch [7.6 to 8.9™°'] in adults; 0.4 to 0.45 [10 to 11.4"™] in birds of the 

 year" in bo realis, against "0.5 to 0.6 inch" [12.7 to 15.2""] in xanthodryas, 

 and " length of wing 2.7 to 2.4 inches " [69 to 61™"] in borealis, against 

 " 2.85 to 2.6 inches " [72 to 66™"'] in xanthodryas. 



Comparing these measurements with those of borealis given by me in 

 the table under the head of the latter species, it is evident that the 

 individual variation, even in the same sex, is as great as the difference 

 between the maximum of the larger and the minimum of the smaller spe- 

 cies, as given by Seebohm. It will furthermore be seen that the average 

 measurements of the first primary in my series is equivalent to the min- 

 imum length of the corresponding feather of the young bird, as stated 

 by Seebohm, and that only one or two, out of eleven, have a bastard pri- 

 mary as small as the maximum given by the latter author. It is there- 

 fore evident that, for instance, No. 89158, in regard to the dimensions, 

 should rank as Ph. xanthodryas. It agrees, however, so closely with the 

 others of the same series in color that there can be no thoughfof sep- 

 arating it from these. 



Below are given the dimensions of two genuine Ph. xanthodryas col- 

 lected by Mr. Jouy in Japan, one a bird in summer plumage, the other 

 one shot in the autumn. They are both males, as are all my specimens 

 of borealis. A comparison of their dimensions gives the same results as 

 above; they differ in no way, the largest of the two being even a trifle 

 smaller than the largest borealis. The first primary is, however, in both, 

 longer than the average of borealis, although in the smaller specimen 

 it is a little shorter than the longest first primary in borealis. But the 

 color is decidedly much more yellowish than in the eleven specimens 

 belonging to the latter species. 



It may here be well to remark that not only are the two Japanese 

 specimens completely uniform in color interne, but there is not the slight- 

 est difference between the specimen No. 92556 of borealis in autumnal 

 plumage and the other ten obtained during the spring and summer 

 months.* 



Finally it may be stated that there is no difference in color whatever 



'That the relative length of the second and sixth primaries is of little value is evi- 

 dent from the fact that in No. 75416 from Alaska the sixth ig longer than the second. 



15861 Bull. 29 20 



