1890.] Dwight on the Horned Larks. I 39 



clearly described. This brings it down to a question of where to 

 draw the circumscribing lines, and allowing the already described 

 forms to stand as models, I have endeavored to draw them no 

 closer than those laid down by Mr. Henshaw. The excellence of 

 his work, based on 350 specimens, is attested by the fact that my 

 material does little more than support his conclusions, and where 

 I have arrived at different results, it is simply because I have 

 specimens that were not then obtained. In fact I have had too 

 many from certain localities, not enough from others. Breed- 

 ing examples of the Arctic forms alpestris and leucolcema are 

 few in number, and as for Mexico, there is next to nothing to 

 show what real chrysolcema is. Mr. Henshaw's material only 

 carried him as far south as Arizona and New Mexico, and he 

 naturally supposed that arenicola passed into chrysolcema, there 

 being nothing to show the existence of the well-marked desert race 

 adusta which intervenes. The naming of such intermediate 

 races as adusta and the dark race merrilli of eastern Oregon, 

 Washington and British Columbia, the only new ones except 

 pallida recognized in this paper, may be questionable, but they cer- 

 tainly are as well marked as already existing forms, and as mat- 

 ters stood arenicola was an intermediate race between leucolcema 

 and chrysolcema, north and south, and between praticola and 

 strigata, east and west. Among the surprises developed by the 

 material before me, is the existence of a form on the eastern slope 

 of the Sierra Nevada Mountains, referable directly to prat 1 'cola. 

 This is discussed under its proper head, and I now wish to call 

 particular attention to two facts which seem to have escaped gen- 

 eral notice, and are of great importance in understanding the plu- 

 mage of birds of this genus. One is the fact that but one moult 

 takes place in the year, the breeding plumage being the result 

 of the wearing away of the tips of all the feathers acquired during 

 the autumn moult. The effect is most striking on the black areas 

 which, clouded and obscured with yellowish tips in the autumn, 

 come out clear and defined in the spring. There is no evidence 

 whatsoever of a spring moult. The other is that young birds in 

 passing from first plumage, which is worn only a few weeks, 

 moult wings and tail as well as the feathers, usually moulted. 

 This fact and change of plumage without moult have not been 

 attributed to the Horned Larks, so far as I know, and only in a 

 general way is it known to be true of other species. 



