103 [Vol. xvi. 



(EsTEELATA wiLsoNi, Sharpe. 

 4 (J. 3?. Light phase. 

 S^ . 3 ? . Dark phase. 



" This species was just commencing to nest at the time 

 of our visit (January 4th) ; I only obtained one egg. The 

 dusky-breasted examples of this species are almost as dark 

 as G<j. trinitatis. 



" They have the tarsi and basal half of the toes very 

 dark brown, not flesh-coloured as in the lighter specimens. 

 I should not be surprised if this dark form proves 

 to be the young of the light form, as all the birds I saw 

 nesting had light breasts ; one of the specimens has a 

 nearly white breast, while the others grade into one 

 another. I fancy that if a large series of these Petrels 

 was examined links would be found to connect all these 

 three species, certainly CE. ivilsoni, and CE. arminjoniana. 



" The number of these Petrels on this island was extra- 

 ordinary. Prom the sea-shore to the summit of the island 

 they swarmed in countless multitudes." 



(EsTBELATA TEiNiTATis, Grigl. et Salvad. 



2 cJ . 1 ? . One in down. 



" This appears to be the least abundant Petrel on the 

 island, and was found much higher up the hillside than the 

 former species. It had young at the time of our visit," 



Pelecanoides dacunh^, sp. n. 



Nearest to P. urinatrix but smaller, especially as regards 

 the bill and feet ; flanks much less marked with grey ; the 

 throat and fore-neck also much whiter. 



Total length 8 inches, wing 4*3, culm. I'O, middle 

 toe 1"0, tarsus 1.0. 



Hah. Tristan da Cunha. Two females obtained. 



"I have not seen the Pelecanoides urinatrix {?) obtained at 

 G-ough Island by the ' Scotia,' but it is probably referable 

 to this new species." 



PuFFiNus GRAVIS, O'EeiUy. 



While lying-to, off Tristan da Cunha on account of the 



