1909 ] Brewster, Barrow's GoMen-eye in Massachusetts. 159 



by the female of this species. The bill of americana, when viewed 

 from above, has a very different appearance; being much more 

 typically duck-like in shape. It rarely, if ever, shows any pure 

 yellow except near the tip where there is sometimes a narrow bar 

 of this color on the culmen, just behind the nail, with perhaps 

 some indication of a corresponding marking on the lower mandi- 

 ble, also. The dark bar on the wing is much less often lacking 

 in islandica than in americana but as it is not infrequently quite as 

 conspicuous and perfect in the latter as in the former it possesses 

 no great value as a diagnostic character. 



Many writers have asserted that islandica is the larger of the two 

 forms, especially with respect to its wing measurements. There is 

 perhaps some average difference of this kind although the smallest 

 bird of either kind in my collection is an adult female of islandica 

 taken in June among the mountains of British Columbia. As to 

 the difference in the width of the nail at the tip of the bill, to which 

 Mr. Ridgway has called attention, I am unable to verify it. 



If I were asked to restate the characters just formulated, placing 

 them in the order of their relative importance, I should arrange 

 them thus: (1) Shape and proportions of bill; (2) coloring of head 

 and neck; (3) coloring of bill; (4) presence or absence of continuous 

 dark band across white wing patch. When all the marks of dis- 

 tinction which I have attributed to one or the other species are 

 possessed in combination by a single bird the identity of the speci- 

 men is open to no doubt, but unfortunately there is perhaps no one 

 of them all which is invariably confined to the form of which it is 

 ordinarily characteristic. Indeed, one cannot handle any consid- 

 erable number of female Golden-eyes killed in winter in New 

 England without coming upon specimens which are far from 

 typical, while some of these are likely to be so nearly "half-way" 

 intermediates between americana and islandica that their definite 

 reference to either form is impracticable, except on purely arbitrary 

 grounds. I used to suspect that such birds might be of hybrid 

 origin but I now incline to the opinion that they represent nothing 

 more nor less than a curiously one-sided transfer or borrowing of 

 external characters which are not always constant. They fail, 

 however, as far as I have observed, to furnish series perfectly con- 

 necting americana with islandica. Oddly enough the unfilled gap 



