34 



Mr. Godman in retaining the name given by Heineken for the present or the Madeiran species, 

 and that of laurivora for the species inhabiting the other islands, and which has the terminal 

 portion of the tail white. But little appears to be known respecting the present species, chiefly 

 owing to the fact that previous authors have overlooked it and considered it identical with 

 Columba laurivora ; and I have consequently no data respecting its habits, except the notes pub- 

 lished by Mr. Godman, who writes (Ibis, 1872, p. 216) as follows: — "Columba trocaz frequents 

 the high laurel forests of Madeira, where it breeds, only coming down occasionally to feed in the 

 cultivated lands. It is very shy, and not easily got at, even by the natives, who are acquainted 

 with its habits. I was out after them for two or three days without getting a shot, though I saw 

 several. The only plan is to go either early in the morning or late at night and sit under 

 the trees (Oreodaphne fastens) which they frequent, and take the chance of their settling above 

 your head without observing you. Even then they are not easily killed, as the foliage is so 

 thick, and the trees so high, as well as being usually placed on such a steep slope on the 

 mountain-side that, when shot, they sometimes fall more than 100 yards below you, smashing 

 themselves to pieces in the fall ; hence it is not easy to get good specimens for one's collection. 

 I, however, finally procured eleven fair skins. They feed on the fruit of the bay and till trees, 

 for which they search among the dead leaves upon the ground. The flesh has a strong flavour, 

 in consequence of the aromatic nature of the food." Mr. Godman (I. c.) enters fully into the 

 question of the synonymy of the present species and its allies, Columba laurivora and C. bollii ; 

 and it may be well to give a short recapitulation of what he says, the more so as after going 

 carefully into the question I fully agree with the views he adopts. It appears that there can be 

 no doubt that the name of trocaz is the correct one for the present species, as Dr. Heineken 

 (I. c.) describes it from the island of Madeira. Messrs. Webb and Berthelot, however, confuse 

 the Pigeon of the western group of the Canaries and that of Madeira under the name of Columba 

 laurivora, which name was intended to supplant the less classical one of C. trocaz. These 

 gentlemen figure, as the male, on the upper part of their plate, Columba trocaz of Heineken ; and 

 on the lower part of the plate is what they call the female, but which is the Pigeon of the 

 western group ; and as the Canaries are the first mentioned of the localities cited by Webb and 

 Berthelot, and this latter bird appears to be the Pigeon met with by them, I quite agree with 

 Mr. Godman that the name C. laurivora given by these gentlemen should be retained for it. 

 Respecting the synonymy subsequent to the publication of Webb and Berthelot's work, Mr. 

 Godman writes as follows : — " Bonaparte, in his ' Conspectus Avium,' ii. p. 45, considers that 

 both Heineken's bird and that of Webb and Berthelot belong to one species, which he calls 

 Trocaza trocaz, to which the locality ' Ins. Madeira' is assigned, the mention of the Canaries by 

 Webb and Berthelot being overlooked." This view, however, is altered in the ' Comptes Eendus,' 

 xliii. (1856) pp. 837, 948, where Prince Bonaparte justly considers that the birds figured in the 

 plate of the ' Ornithologie Canarienne' belong to two species. He applies the name Trocaza 

 bouvryi to the upper figure, retaining that of laurivora for the lower. The former name must 

 therefore be placed as a synonym of C. trocaz. Both the species are figured in the ' Iconographie 

 des Pigeons,' tt. 69, 70, C. laurivora being also figured by Knip, t. 43. 



Dr. Bolle, in his first paper (J. f. O. 1855, p. 171), appears to have followed previous authors 

 in confusing the Canarian with the Madeiran bird. He obtained no specimens, though he says 



