on the Genus Cocreba. 519 



Island, but judging from specimens obtained from neigbour- 

 ing islands, if these are really C, bartholemica, the presence 

 of a white wing-spot would appear to be an inconstant 

 character. In this connection it may be interesting to quote 

 Dr. Allen^s remarks on a splendid series of 72 specimens of 

 C. chloropyga (a wing-spotless species) from Chapada in the 

 Matto Grosso provin-^e of Brazil (Bull. Amer. Mas. N. H. 

 vol. iii. 1890-91, p. 348). He says : "■ abont one specimen in 

 ten of the adults shews a slight trace of while beyond the 

 primary wing-covert, but only in about one in twenty is it 

 distinct enough to readily attract attention. '^ Possibly 

 there is the same inconstant tendency in C bartolemicUy 

 and in the normal condition the species is without a wing- 

 spot. 



As I have before remarked, it is, with tlie material at 

 hand, somewhat difficult to define the limits of this species. 

 In two specimens from Anguilla, which I have examined, the 

 white superciliary stripe does not extend anterior to the eye ; 

 they both have conspicuous white frontal bands, and there 

 is a small white wing-spot just visible in both. I have little 

 doubt that Anguilla specimens can be referred to C. barto- 

 leniica. As regards examples from Antigua, which lies on a 

 different submarine plateau, I have examined 12 specimens 

 and they are all distinctly paler above and below than spe- 

 cimens from Dominica (19 specimens), Guadeloupe, and 

 !St. Kitts (C b. duininicana). The mantle is sooty grey ; 

 w ings conspicuously tipped with greyish ; pileum distinct 

 from mantle ; yellow of underparts paler and clearer than 

 in birds from Dominica ; and two specimens shew slight 

 signs of a wing-spot. 



Probably these Antiguan examples are referable to 

 ('. bartulemica, but should they be proved to be distinct I 

 propose the name C. atlantica for them. Specimens from 

 Barbuda in the British Museum and Tring collections also 

 agree with those from Antigua. Both Antiguan and Bar- 

 budau examples are distinguishable at a glance from 

 examples from Dominica. It should be noted that both 

 these islands are covered with limestone formations and 



