160 Brewster, Barroiv's Golden-eye in Massachusetts. [ April 



lies not midway between the two species, as might be expected, 

 but much the nearer to islandica. In other words americana 

 seems to approach islandica very closely through birds possessing 

 certain characteristics of the latter species, whereas islandica ex- 

 hibits little or no tendency to appropriate any of the characters of 

 americana. Or, to put the case still more definitely, if somewhat 

 figuratively, americana may be said to have forged towards islandica 

 a closely-welded chain, quite continuous up to the point where it 

 abruptly terminates, just before reaching the narrowly circum- 

 scribed limits of the area occupied by islandica, a comparatively 

 stable and immutable form. For although the birds which supply 

 the links of this chain grade perfectly into typical americana on 

 the one hand they do not seem ever to pass a definitely fixed point 

 in their approaches to islandica on the other. Yet collectively 

 they exhibit, more or less unmistakably, nearly all the characteristics 

 of ultra-typical specimens of islandica. Because of these condi- 

 tions it has been my custom, when identifying female Golden-eyes 

 taken in America, to refer all specimens not typical — or nearly so — 

 of islandica to americana. This practise may be somewhat arbi- 

 trary but it is at least consistent with the facts in the case, as I 

 understand them. It is possible, of course, that my evidence is 

 incomplete and that the missing links in the chain of approaching 

 females to which I have called attention may yet be found. But if, 

 as I am inclined to believe, they do not exist, how can their absence 

 be explained ? Before attempting to answer this question it may 

 be well for me to say a few words about the variations that I have 

 noted in male Golden-eyes of both kinds. 



There is never any difficulty in separating the adult males of the 

 two species. They are, indeed, so strikingly unlike that one can 

 distinguish them almost at a glance, without direct comparison. 

 The male of islandica seems subject to remarkably little variation 

 of any obvious kind. The .male of americana is less uniformly 

 characterized. It occasionally has a bill shaped much like that of 

 islandica or white cheek markings so elongated vertically as to 

 somewhat resemble those of that species. Dr. Gilpin asserts that 

 "both males have the violet wash in the green of the head" but I 

 have never known it to be shown conspicuously by americana, nor 

 to be other than conspicuous and widespread on the head of 



