NOVITATES ZOOLOGICAE XXV. 1 918. 349 



was certainly hi error, since nudirostris is not identical and several other forms 

 have now been separated. 



Reichenow (Vog. Afr. i. pp. 394-397, 1901, and iii. p. 806, 1905) got very 

 much nearer the truth, for he separated " Vinago calva " and " Vinago calva 

 nudirostris," but his distribution cannot be right, and he appears to have taken 

 some females and other specimens of other forms of calva for nudirostris, though 

 he correctly recognised the latter as different. In 1902 and 1905 he further 

 separated the West African forms mto two, distinguishing as a new subspecies 

 Vinago calva sharpei" from Sierra Leone to Calabar. Mr. C. H. B. Grant, 

 Ibis, 1915, pp. 36 and 37, returned to the older less correct notions, separating 

 only " Vinago calva calva " and " Vinago calva salvadorii," a form described in 

 1897 and synonymized by Reichenow with nudirostris ; Mr. Grant said that 

 he " could not see any difference between Columha calva Jemm. and Vinago 

 nudirostris Swains., and that Vinago calva sharpei was " a pure synonym of 

 V . nudirostris," which is by no means correct. 



We have purposely united Osmotreron (or Dendrophassa as it should appar- 

 ently be called), Vinago, and Treron. The extent of the naked cere or base of 

 bill is merely a specific character. Treron teysmanni, though placed in " Osmo- 

 treron," connects the latter and typical Treron m colour and has the rhamphoteca 

 extending to the feathers on the forehead. " Vinago " australis from Mada- 

 gascar, though in colour a ty^jical " Vinago," has the bill of an " Osmotreron." 



The two prmcipal characters by which the various forms of T. calva can be 

 sej)arated, are : 



1. The extent of the naked space ("cere") on the forehead. This is a 

 sharp division, and one is tempted even to regard it as a specific character, but 

 we do not think that forms with a short and a wider naked space occur together, 

 and it is m single instances possible to mistake a female of calva for a male of 

 nudirostris, as generally, though not quite constantly, the females have the 

 naked space less developed. The colour of the under tail-covcrts is the same 

 in both sexes. 



2. The sharply defined, more pure lavender grey, or less defined, duller, 

 more greenish nuchal collar, as well as generally brighter or duller coloration. 



Length of wing is rather variable, as in most Pigeons ; it can, therefore, only 

 be of importance in cases where large series have been measured. Females have 

 shorter wings, but many specimens in collections are, as usual, wrongly sexed. 



We distinguish the following forms : 



1. Treron calva calva (Temm.). 



Columba Caira Temmiiick, in Temm. and Kn\p, Pigeons, i. p. 35. pi. 7 (1808— from a collection 

 made on the coasts of " Loango and Angola." Restricted terra typica : Loango. The figure 

 IS called " bad " by Salvadori, but it is, in fact, very good, showing the dull nuchal coUar and 

 general dark coloration to perfection). 



Colours dark and dull, head more olivaceous green than yellowish, nuchal 

 collar greyish green to greenish grey, and not sharply defined. Back some- 

 times, apparently chiefly m fresh plumage, with grey tinge. Base of bill naked 

 for about li the length of the rhaniijhotheca and nearly or quite to a line con- 

 nectmg the fore-edge of the eye. Wings 20 <J, 155-172, generally 160-165, 

 measures above 170 rare, § 148-153 mm. Base of bill and forehead magentf 

 or bright red. 



