356 NOVITATES ZOOLOQICAE XXV. 1918. 



Treron bicincta. 



There are some misprints in Novitates Zoologicae, 1910, p. 193. The 

 wings of T. b. bicincta measure 161-186, those of T. c. leggei 142-148 mm. 



Treron curvirostra hainana subsp. nov. 



Nearest to T. curvirostra nipalensis from Northern India, but generally 

 larger, bill as a rule deeper, wings longer, the grey of the crown not reaching 

 quite so far backwards, the nape dull green without any ashy-grey wash, whUe 

 in T. c. nipalensis there is a more or less distinct indication of a grey band. 

 Wings (J 145-157, ? 147-151 mm. 



Hab. Hainan. 



Type: cj Mt. Wuchi, Haman, 5. iv. 1903. Katsumata leg. 



We cannot help adopting, as Oberholser has done (Smithson, Miscell. Coll. 

 Ix. No. 7. 1912. J). 3), the name curvirostra of Gmelin, although the figure on 

 which it is based does not show tlie grey crown, nor can we at present discuss 

 the other subsj)ecies (of which nasica must be one), for want of sufficient material. 



The forms of Columba guinea. 



In the Cat. B. Brit. Mus. xxi. pp. 266-268, Count Salvadori recognized two 

 species, C. guinea and C. phaeonota. The latter, according to modern ideas, is 

 a subspecies of guinea, and was first recognized as such by the senior writer in 

 1891, Kat. Vogels. Senckenherg . Mus. p. 186, under the name of Columba guinea 

 trigonigera, a name which though comprising both guinea and phaeonota, chiefly 

 refers to the former. 



Reichenow, Orn. Monatsber. 1898. p. 82, described CoZ«?n6o guinea uhehensis 

 from Iringa in Uhehe, and in 1901, Vog. Ajr. i. p. 402, added C. guinea longipennis 

 from E. Africa, from the Victoria Nyanza to Ugogo, while he lets C. guinea guinea 

 range from West Africa to N.E. Africa, where it is found from 16° lat. to the 

 Victoria Nyanza. C. phaeonota he treats as a species, inhabiting South Africa. 



In 1905, Journ. f. Orn. p. 113, Erlanger identifies both longipennis and 

 uhehensis with phaeonota, which he says ranges from S. to E. Africa. In this 

 judgment he must undoubtedly have been wrong, and it is probably due to 

 some confused notes or faulty memory. About longipennis there is no question 

 that it is a form of guinea, but not phaeonota, while uhehensis cannot either be 

 phaeonota, as it has the rump light grey as in guinea, though otherwise it 

 appears to be nearer phaeonota . 



In Ibis 1915, pp. 37-39, Mr. C. H. B. Grant reviews the forms of Columba 

 guinea, from an examination of the material in the British JIuseum. He recog- 

 nized three subspecies : 



C. guinea guinea : West Africa. 



C. guinea longipennis : Abyssinia and " Sudan " to East Africa. 



G. guinea phaeonota : South Africa. 



This cannot be correct, because, if the Abyssmian form is the same as the 

 East African one, it must be called : C. guinea dilloni (Bp.). Moreover, Abys- 

 sinian and East African specimens differ as much, or, in fact, even more from 

 each other, than E. African ones differ from the \\'est African form. According 

 to Mr. Grant's measurements males from W. Africa have the wings 212-231, 



