248 University of California Publications in Zoology ['Vol. 24 



actually of hybrid parentage in the usually accepted sense, in that 

 particular case with one parent Jimco caniceps, the other, Junco 

 mearnsi; and I am willing to admit (see postea) that individual hybrids 

 may occasionally occur between any species of Junco whose ranges 

 adjoin. To consider as hybrids, however, all the birds ( juncos in this 

 instance) of a whole region, despite the facts that they comprise the 

 only representative of the species in that region, that they all exhibit 

 the same combination of characters over a wide stretch of country, and 

 that they breed true, that is, transmit these same characters to their 

 young, is an utterly misleading use of the term. The junco of the 

 Stikine Eiver should not be regarded as a hybrid. 



As bearing upon the fact that it is the female of connectens that 

 shows variation from the hyemalis type, attention may be drawn to 

 the following statement by Dwight {loc. cit., p. 289) : "The variation 

 in females of the three species [i.e., hyemalis, oreganus, and mearnsi] 

 complicates the question still more, for the average females of the 

 three differ much less from one another than do the males, and the 

 hybrids between them would seem therefore to be much more num- 

 erous." I suppose the application of this to the present case (the 

 junco of the Stikine region) would lead to the conclusion that here 

 on a large scale is an example of sex-linked inheritance; that in the 

 hybridization of hyemalis and oreganus the female offspring only show 

 the oreganus characters. That such is not the case is shown by speci- 

 mens collected at Flood G-laeier (described farther on), where the race 

 here designated connecteTis comes into actual contact with oreganus. 



My own conclusions are as follows : The junco of the Stikine region 

 is a recognizable form, apparejttly the same to which Coues applied the 

 name Junco hiemalis connectens. It is most nearly related to Jutico 

 hyemalis hyemalis, and may be conceded to exhibit intergradation of 

 a sort, as between Junco hyemalis and Junoo oreganus. That is, the 

 peculiarities of this subspecies undoubtedly have some bearing upon 

 the relationship of those two aggregations of races. The intergrada- 

 tion exhibited, however, is apparently as between the two species Junco 

 hyemalis and Junco oreganus, rather than between two adjacent sub- 

 species, Junco hyemalis connectens and Junco oreganus oreganus. It 

 is not of the sort that is usually found between two subspecies of the 

 same species, and for the present it may well be disregarded as a sub- 

 specific criterion. It seems to me that this is a proper place for a 

 somewhat arbitrary division, and that Junco hyemalis and Junco 

 oreganus should still be regarded as separate specific groups. Then, 



