(92) 

 174. Zenaida auriculata auriculata (Gay). 



Peristera auriculata Gay, Hist. Chile i. p. 3H1. pi. (5 (1847. — Chili : "en las provincias centrales de 

 la RepUblica"). 



Nos. 2106, 2104. ? ail, ? jnv., Goyaz, May 1906.— ? ad.: win<r, 142; tail 

 95 mm. 



I am unable to separate the Brazilian birds from the typical Chilian form. 

 The specimens from Para, Santarem, etc., however, constitute an easily recog- 

 nisable race, being characterised by its smaller size, mnch darker nnder-snrface, and 

 bnffy or greyish white (not pure white) tips to the rectrices. Its proper name is 

 Z. auriculata jessieae Ridgw.* 



175. Scardafella squammata squammata (Less.). 



Columba squamosa (neo Boanaterre, 1702) Temminck and Knip, Pir/. i. fam. Colombes, p. 127. 



pi. 50 (1808-11. — "dans le territoire de Buhin (sic!) au Bresil," rectius BaA/a — type in Paris 



Museum). 

 Columba squammata f Lesson, Traiti d'Orii. p. 474 (1831. — based on the above). 



Nos. 2065, 2067, 2068, 2081, 2085, 2086, 2087, 2088, 2089. <?<? ad., SS imm., 

 ? ?, Goyaz, April and May.— c?c? : wing 94—99; tail 95—102 ; bill 13—14 mm. 

 ? ? : wing 93, 95 J ; tail 96, 97 ; bill 131, 14 mm. 



This series agrees perfectly with other specimens from Bahia and Minas (jieraes 

 (Rio Jordao, prov. Aragnay) collected by Mous. Alphonse Robert. The ground- 

 colour of the upper parts is a dull greyish brown, the black edges to the feathers 

 of the pilenm, nape and back are rather narrow, the foreneck very pale vinaceous, 

 the bill short and slender. 



Birds from Venezuela (Cumana and Orinoco valley) and from Margarita 

 Island, of which 1 have twenty-three before me, diifer at a glance by the much 

 brighter, buffy or yellowish brown ground-colour of the upper parts, with much 

 broader black edgings to the feathers ; deeper vinaceous foreneck ; broader black 

 edges on the breast and abdomen ; and by having a thicker, generally also longer 

 bill. This very distinct form is to be called S. squammata ridgwaiji Richm. % PiS, 

 pointed out by Berlepsch and Hartert, there is no constant difference between 

 typical specimens from Margarita and those from the Venezuelan mainland. 



The range of the two forms is as follows : 



a. S. s. squammata (Less.). 



Campos of the interior of Brazil: Bahia (Hobert ; Wucherer) ; Fernambnco 

 (Forbes); Goyaz (Baer); Northern S. Paulo: Irisanga, Faz. do Jose Diaz, Sitio 

 do Bahu (Natterer), etc. 



N.B.— The locality " Pelotas, Rio grande do Snl " {Cat. B. Brit. Mus. xxi. 1893. 

 p. 465) is, no doubt, erroneous. 



Examined : 9 Goyaz, 1 Bahia, 2 Rio Jordao, Minas, all in the Tring Museum. 



• Zenaida jessieae Eidgwa.v, Ptoc. U.S. Mm. x. 1887. p. 527 (1888.— Santarem). 



f Les,son's name is perhaps merely a lapsus ealavii for C squamosa, but as this cannot be proved, 

 I think it must be accepted. If not, the Brazilian form requires a new name. 



t Scardafella rldgwayi Richmond, Proc. P.S. Mux. xviii. p. fiPO (1896.— Margarita Island). 



