150 BULLETIN 5 0, UNITED STATES NATIONAL MUSEUM 



ACCIPITER GENTIUS ATRICAPILLUS (Wilson) 



American Goshawk ^ 



Adult male. — Upperparts varying from deep gull gray (rarely) 

 to slate-gray and deep neutral gray, becoming darker on the inter- 

 scapulars, nape, occiput, and crown, the top of the head often being 

 blackish slate to slate black and occasionally even a little blacker, 

 the nape and sides of the crown so heavily mottled with white as to 

 be more white than slate; upper wing coverts like the back, often 

 tinged with mouse gray and hair brown; remiges fuscous, above 

 indistinctly broadly barred with darker and fuscous on the inner 

 webs; below broadly barred and mottled with whitish on the inner 

 webs; five outermost prim.aries emarginated on their inner webs; 

 fourth and fifth primaries the longest; the second (from the outside) 

 between the sixth and seventh in length; rump and upper tail coverts 

 like the back; tail similar but the rectrices tipped with whitish and 

 crossed by four broad fuscous bands, the fuscous bands narrower, 

 however, than the slate gray interspaces; tail slightly rounded, the 

 lateral rectrices 18-26 mm., shorter than the median ones; lores, 

 cheeks, and auriculars varying from dirty white streaked with black- 

 ish to nearly pure black, the superciliary stripes white more or less 

 mottled with blackish slate; forehead occasionally whitish; entire 

 underparts white thicldy mottled, barred, or vermiculated with mouse 

 gray to dark mouse gray, the feathers of the chin, breast, and abdomen 

 with blackish shaft streaks, which, like the vermiculations, are very 

 variable in width, the shaft streaks usually absent in the feathers 

 of the sides, flanks, thighs, and the under tail coverts; under tail 



« The variations of the goshawk cannot be considered understood in a final 

 sense until adequate breeding material is studied from all parts of its range. In 

 the case of no other North American hawk is our present combined museum 

 material so incomplete, over 80 percent of the specimens being winter birds or 

 migrants. While the treatment here accorded the species is all that I can do 

 with any assurance, it may be pointed out that there may be a large, relatively 

 pale form in northwestern Canada and Alaska (not in coastal, southern Alaska) 

 which appears to migrate southeastward in autumn. Similarly, the status of the 

 supposed western race, striatulus, is still to be definitely settled. Taverner's 

 action in considering it synonymous with atricapillus is borne out by most of the 

 facts, but it is just barely possible that part, at least, of the original type series 

 of striatulus may be migrants or strays of the Queen Charlotte Island race, 

 laingi. I agree with Taverner in "sinking" striatulus at least for the present and 

 do not see much ground for thinking its status is apt to be changed. The fact 

 remains, however, that the whole matter is not as clear cut as it might be. The 

 only change I can foresee is that possibly the name striatulus may be considered 

 an earlier one for laingi. 



