406 BULLETIN 133, UNITED STATES NATIONAL MUSEUM 



that there is no sharp dividing line. Phrygilus unicolor (d'Orbigny 

 and Lafresnaye), Avhich has been proposed as type of a genus 

 Geospizopsis Bonaparte/^ is identical in form with P. gayi and 

 must be considered a synonym of true Phrygilus. 



Haplospiza Cabanis/^ with Haplospiza. unicolor Cabanis as its 

 type, is closely similar to certain forms of Phrygilus unicolor in 

 general appearance. It differs structurally from true Phrygilus in 

 more rounded tail, more rounded wing tip, in having the ninth 

 (outermost) primary shorter than the third, about equal to the 

 second, more attenuate bill, and shorter lateral toes, with the fourth 

 toe without claw reaching barely beyond the base of the last joint 

 of the middle toe, and the second toe without claw barely extended 

 to that point. Haplospiza unicolor^ in addition to smaller size, is 

 distinguished at a glance from any Phrygilus in lacking any edging 

 of lighter color on the outer webs of the primaries. 



Near General Roca, Phrygilus fruticeti was common from Novem- 

 ber 25 to December 2 (four prepared as skins), and others were re- 

 corded at Zapala, Neuquen (one taken). The birds were not breed- 

 ing, but were gathered in little flocks that fed on the open ground 

 under shelter of desert shrubs of various species. The birds were 

 wary, and at any alarm rose and flew with a swiftly darting or 

 undulating flight for long distances before again dropping to cover. 

 In early morning they rested in low bushes in the sun. Their call 

 was a low, mewing note or a sharp pUck. On December 2 males were 

 singing a song that sounded like the effort of some icterid. 



On comparison of a good series that includes 10 skins fronj the 

 highlands of Peru, 5 from central Chile, and 11 from Rio Negro 

 and Neuquen, in Argentine Patagonia, it appears that Phrygilus 

 fruticeti pemviwruas Zimmer,^^ while not sharply differentiated, may 

 be distinguished from true fruticeti in adult males by the grayer 

 coloration above, with more sharply defined, blacker streaks, and 

 in females by heavier streaking above, especially on the head. The 

 white spots on middle and greater coverts are usually, but not always, 

 larger in peruvianus. In size the two forms, from the series at hand, 

 appear quite similar, except that in Peruvian birds the bill is 

 slightly larger. In five males from La Raya and one from Are- 

 •quipa, Peru, the wing ranges from 94 to 100.5 mm., the bill from 

 14.4 to 15.5 mm. ; in three males from Santiago and central Chile the 

 range is from 94 to 100.2 and 13-2 to 13.5 mm. ; while in five from Rio 

 Negro, Argentina (General Roca, Arroyo Cumallo, Arroyo Seco, 

 Paja Alta, and Nahuel Niyeu), it is 91 to 98 and 13 to 14.7 mm. 

 The length of wing, it will be noted, is practically identical. 



^ Compt. Rend., vol. 42, 1856, p. 955. 



^Mus. Hein., pt. 1, April, 1851, p. 147. 



13 Field Mus. Nat. Hist, Zool. Ser., vol. 12, Apr. 19, 1924, p. 63. (Matucana, Peru.) 



I 



