( ly ) 



Bahia skins (E. fuliyinosn). A series from (Jniuaua is also not ilifferent. This 

 species is not included in Chapman's paper on the birds of Trinidad, thongli the 

 type oi P. f/imvsn has lieeu collceted on that island. 



44. Oryzobonis augolensis (Linn.). 



Loxia migiilensU Linnaeus, Si/nt. Xnl. xii. i. (ITOfi) p. .'30:1 (ex Edwards ; " Aiignl.i " — prrnve ! 



We substitute Eaut Brazil as typical locality). 

 L. Iiirridu Scopoli, Ami. Nat. Wxt, i. (1769) p. 140. 



2 <i<S and 3 ? ? from ( "ai)aro, March and April; S S <S from Scelet, April ; 

 and 1 S each from Laventille, December, and Chagnaramas, Jannary. This series 

 agrees e.xactly with specimens from Surinam, British Guiana, and Cnmami. I have 

 3") adnlt males from different localities before me, and none of them shows any 

 trace of a chestnut cross-band on the throat. Therefore it seems very probable 

 that 0. speciihiris Finsch,* if not au accidental variety, may yet turn out to lie a 

 different race. 



45. Oryzoborus crassirostris (Gm.). 



Loxia crussirnslris Gmelin, Si/xt. A'tit. I. ii. (I7SM) p. Hi'd [ex Latham : PTab. ign.— We substitute 

 Cai/fime.'] 



Two (?c? ad. from Seelet, March. 



They agree with specimens from the Orinoco (Qniribana de C'aicara) and Upper 

 Rio Negro (Lamalonga and Marabitanas) in having rather large and strong bills. 

 The birds found in British (iniana have mostly smaller bills, but this is not quite 

 constant. 



40. Ostinops decumanus (Pall.) 



XaulliurmiK ihcumunns Pallas, Spicil. fasc. vi. (1769) p. 1 (Surinam). 



Oatinops deatmanus insuhii-is Dalmas, Mini. Snc. Zool. France xiii. (1900) p. I.'i7 (Tobago). 



Two c? and S ¥ ? from Caparo, April, and Seelet, April. 



Count Dalmas separated the Tobago birds as a subspecies on account of their 

 smaller size and paler castaneous rump, but there must be some mistake here. 



On comparing 4 cJcJ and 3 ? ? from Tobago with a very large series from 

 different localities on the mainland, I cannot find the slightest difference, either in 

 size or in colour. The females are always considerably smaller than the males, 

 the crest is much shorter, and the rump usually paler castaneous. These arc just tlie 

 characters by which 0. d . hi.'^iildiiti is said to differ from the typical form, and I 

 strongly suspect that ( 'ount Dalmas was misled by wrongly sexed specimens. 



Measurements : 



• /'. /.. .i. 1870, p. 5S3. 



