79 [Vol. xxi. 



fluring tlie voyage of the ' Magenta' in being much larger 

 and in having a well-defined but hidden white patch, formed 

 b\r the bases of the feathers of the throat. 



" Looking through the first part of the ' Monograph of 

 the Petrels/ recently published by Dr. Godman, I noticed 

 that two specimens of Fregetta grallaria in the Turin 

 Museum, collected in the Pacific Ocean during the voyage 

 of the ^Magenta' (one between Callao and Valparaiso, and 

 the other at sea lat. 27° 53' S., long. 88° 04' W.), were 

 exactly like the plate and description in the Monograph. 

 The specimen from Tristan d'Acunha in the Turin Museum 

 was quite different, being much larger, with the upperparts 

 uniform brownish -black, with no white margins to the 

 feathers of the back, and with the base of the feathers of 

 the throat forming a defined but concealed white patch. 

 There is no trace of a dark shade on the sides of the lower 

 abdomen, and the black apical half of the under tail-coverts 

 is more sharply defined. 



" I have very little doubt that the bird from Tristan 

 d'Acunha in the Turin Museum belongs to an unnamed 

 species, which I propose to call 



" FrEGETTA * MELANOLEUCA, Sp. U. 



" Capite, collo, pectore summo, dorso et dimidio apicali sub- 

 caudalium nigro-fnliginosis, fere unicoloribus ; mar- 

 ginibus dorsi plumarum baud albidis ; supracaudalibus, 

 pectore imo, abdomine, guise macula vel plaga obtecta, 

 dimidio basali subcaudalium et subalaribus internis 

 albis; tibiis nigro-fuliginosis ; remigibus rectricibusque 

 nigris, sed harum basi obtecta alba ; tectricibus alarum 

 brunneis ; rostro et pedibus nigris. Long. tot. circa 

 210 mm., culm. 14, alse 165, caudse 78, tarsi 40. 

 "Had. Insula 'Tristan d'Acunha' dicta. 

 " It has already been stated that Fregetta grallaria also 

 occurs on Tristan d'Acunha, but the evidence is not satis- 

 factory. Sperling (' Ibis,' 1872, p. 75) says that on one 



* " It appears to me the generic name Freyetta, J3p., is sufficiently 

 distinct from Freyata, Briss., and should be used instead of C't/modroina, 

 Kid-w." 



