22 NOKTH AilERICAX BIRDS. 



Wing, 1 1.70 - 12.00 ; tail, 4.80 - 5.20 ; tarsus, 2.40 - 2.75. T.iil more even, 

 and lighter colored ; the dark bars nairower, and more .sharply dofniod. 

 Colors generally paler, and more grayish, llab. South Ami-rica (Brazil, 

 etc.) \-Sl\: perlataA 



Wing, 12.00-13.50; tail, 5.G0 - 6.00 ; culnien, .85-.'J5; tarsus, 2.70-2.85; 

 middle toe, 1.45- l.GO. Colors as in var. perlala, but secondaries and tail 

 nearly white, m abrupt contrast to the adjacent parts; tail usually without 

 bars. Ilab. West Indies (Cuba and Jamaica, Mus. S. I.) . . var. /urea /a.' 



Wing, 11.00; tail, 5.00; culmen, about .85; tarsus, 2.05-2.45; middle 



shafts. The tail has four rather distinct grayish bands. The facial circle is ochraccous, some- 

 wliat darker across the foreneck ; the face white, with the ante-orbital spot claret-brown. Entire 

 lower parts imraaeulate pure white. Wing-formula, 2, 1 = 3. Wiug, 12.30 ; tail, 5.30 ; cul- 

 men, .70 ; tarsus, 2.75 ; middle toe, 1.45. 



No. 24,283, Nicaragua, (Captain .1. Jl. Dow, ) is like the specimen just described, in the unifonu 

 dark wash of the upjKT parts, but this is deeper ; the lower parts, however, are quite different, 

 being ochraceous-orangc, instead of pure white. 



The remaining fivit specimens (from San .Salvador, Costa liica, and Nicai-agua) are alike, and 

 differ from northern birds in the deeper dark mottling of the upper parts ; the white specks 

 very conspicuous, and usually sagittate. The facial circle deep black where it crosses the fore- 

 neck. The lower parts vary in color from nearly pure white to deep orange-rufous ; the dark 

 markings of the lower surface are larger, more angular, and more transverse than m true i>ra- 

 tincola. The wing measures 11.30-13.00; tai-su.s, 2.55-2.95. The northern form varies 

 from 12.50-13.00 (wing) and 2.50-2.85 (tarsus). It is thus seen tliat while these southern 

 birds average .smaller in general dimensions;, they have actually larger feet, the average length of 

 the tai-sus being 2. SO in the Central American series, and only 2. CO in the northern series. This 

 exactly coincides with the case of Sturncl/a, the S. mic{/na var. mcxicana of the same region being 

 smaller bodied and shorter winged than var. magna of the United States, but with much 

 larger feet, se<t p. 



' Slrix flammea, var. pcrlaia. Strix pcrlaia, LuilT. Veiv.. Doubl. 59, 1823. — TscnuDi, Av. 

 Consp. Wiegm. Archiv. 267, 1844. — Haiht. Syst. Ind. Azara. p. 3. —Max. Beitr. Ill, 263 

 (excl. syn.). — Strickl. Orn. Syn. 1, 1855, 180 (excl. syn.). Strir flammea, U.vrwk, Zoo\. 

 Bciig. 34. — ScHOMB. Verz. Faun. Brit. Guian. p. 732. — Spix, Av. Bras. I, 21. 



This is a .still further differentiated or more appreciably modified race. It differs in smaller 

 size (wing, 11.70-12.50 ; tail, 4.80-5.20 ; tai-sus, 2.40-2.75) and more square tail, while the 

 colore also present constant dilferences. The tail is much lighter compared with the wings, the 

 bands narrower and more sharply defined, though the same in number. 



- St rut flammea, var. fitratla. Sln\e/ii)ratn, Te.mm. PI. Col. 432, 1S3S. — D'Okd. Hist. Nat. 

 Cuba Ois. p. 34. Nab. West Indies (Cuba and Jamaica). 



This fiirui is the most distinctly characterized of all the races o( S. flammea which we have 

 examined. It has the general plumage decidedly lighter and less rufous, while the secomlarics 

 and tail are abruptly lighter than the adjacent parts, and usually free from bands, though there 

 are sometimes traces of them. 



All the American races of Slrix flammea differ very decidedly fjoni the European form (var. 

 flammen) in much larger size. The differences in color are not so appreciable, and there is 

 hardly any certain difference in this respect. The extreme phases, however, appear to be darker 

 in the var. flnmmca than in the var. jmitincoja. The supposed differences in the character 

 of the feathers fringing the oiiereulum, insisted on by MacGillivray (History of British Birds, 

 111, 1S40, p. 473), 1 am unable to appreciate, for I cannot find that they differ in the least in 

 the two rares. That excellent ornitliologist states that in the American " species " the feathei-s 

 of the opereulum are induced to a simi)le tube, having neither filaments nor shaft, while in the 

 European bird they are perfect feathers, with all their parts complete. Though this may have 

 been the case with the one or more specimens of ]>ratmcola examined by Mr. JlacGillivray, I 

 have yet to see an American specimen which has not the feathers of the operculum just as per- 

 fectly developed as in European examples. 



