( 883 ) 



J c? iul. and 1 ? ad. from Igarape-Assi'i, 21. i., 5. iii. 04. Noe. 1025, 2015. 

 Iris : S, " ronge f'once " ; ? , " bruu ronge." Collected by Mous. A. Robert. 



This well-marked species is still verj' rare in collections, and hitherto only 

 known from Para and Borba, Rio Madeira. It differs from L. melanops in having 

 the toj) of the head and the mantle slaty black, only the bases of the feathers being 

 white, and in having no visible white spots on the lower back and upper wing- 

 coverts. The well-defined white superciliary stripe is another point of distinction. 



The juvenile plumage of this species seems to be unknown. 



Its jirojier name is the one used above, L. ktikli having undoubted priority over 

 L. Itaupi. Bonaparte's diagnosis, though very bad, cannot be referred to any other 

 sjjecies. 



Wing 20.), 210, 210 ; tail 130, 13s, 138 mm. 



IH. Columba speciosa Gm. 



Culiiinha sjicriiixu Gmelin, Sysl. Xnt. 1. ii. (1788) p. 783 (ex Buffoii, ct D'Aubonton, I'l. nit. 213 — 



Cayenne) ; Sclater & Salviu, /'. Z. S. IKG7, p. .5'.I0 (Pani). 

 Lepklociias speciuxa Pelzeln, Ziii- Orii. linisil. iii. (1800) p. "274 (Para). 



1 S ad. in beautifnl plumage, 23. xi. iio. No. Iy3. Iris "brownish red"'; 

 bill " cardinal red." 



Identical with sjK'cimcus from British Guiana. 



11 -J. Columba plumbea togoteusis Berl. k Leverk. (an subsp. dist. ?) 



Clihirueiiaa plnmbca Vieill., subsp. uov. hoijotauis Berlcpsch and Levcrkiibn, (Jrnis vi. (1890), p. 32 



(Bogota coll.) 

 C'ltumba rimicm (nee Temminck !) Sclater & Salvin, P.Z.S. 18("i7. p. 580 (Capim River). 

 C. phimhca (nee Vieillot!) Goeldi, Ibis 1903, p. 409 (Rio Gapim). 



1 c? ad. from Prata, 11. xii. 05. No. 317. " Iris white." Wing 175 ; tail 135 ; 

 bill l(i^ mm. 



It diifers from an adnlt bird from Pina, Nortln'rii Peru, which we may consider 

 to represent G. p. bogotensis, by having the throat almost of the same lilac tint as 

 the Ibreueck, while in the Peruvian skin there is a distinct vinaceous bufi' throat- 

 patch. This is also very well pronounced in a good series from Merida and N.AV. 

 Ecuador ; they differ, however, a little in coloration from the Peruvian and Para 

 examples. 



"While it is quite jiossible that the birds from northern South America may 

 be divisible into several geographical races, there can be iitt the slightest doubt 

 that they are altogether different from (J. p. plumhca Vieill., whicli is confined in 

 its range to Eastern Brazil, from Bahia to Rio Grande do Sul. I have now examined 

 2(j specimens of the tyjiical, and about a dozen of the northern form, and found 

 them always easily sej)arable. In C. p. bogotensis the head and nape, as well 

 as the nnderparts, are always strongly washed with vinaceous or lilac, while in 

 C p. plumbea they are pure light grey, with an almost imperceptible lilac hue. 



Berlepsch and Hartert* have already shown that C. piirpureotincta Ridgw., 

 erroneously united with C. plumbea by Salvador!, had nothing to do with this 

 species, but lieloiiged to the grouji of ('. iiigriroilris, which is characterised by its 

 short, thickish bill, rufescent blown (not greyish) under wing-coverts, etc., etc. 



• ^V«r. iSir./. ix. (1002) p. 117. 



2i) 



