BIRDS 319 



groups, but yet intergrading in their variations, should not be regarded 

 as true species, since these intergradations are evidently due to over- 

 laps in individual variation and not to interbreeding, on account of 

 their isolation on the different islands. This is a point that we do not 

 attempt to discuss, and simply follow the conventional canons of the 

 A. O. U. nomenclature. 



Geosfiza fortis as a species, including all thefortis varieties, has 

 its bill variation confined within the following limits in mature birds : 

 Length of culmen, 15 to 20; depth of bill at base, 11 to 16; width of 

 bill at base, 8 to 12. In both length of the culmen and width of the 

 bill at the base it intergrades with G. fuliginosa, but there is a con- 

 stant difference in the depth of the bill, the greatest basal depth in G. 

 fuUginosa being 9.5. 



Subspecies at the lower end of the G.fortis series have the bill 

 'shaped exactly like that of G. fuliginosa farvula; those at the upper 

 end have bills of a proportionally greater depth and with a rounded cul- 

 men. 



63«. GEOSPIZA FORTIS FORTIS (Gould). 



Geospiza fortis Gould, Proc. Zool. See. Lond., p. 5, 1837. — Zool. Voy. 



Beagle, ill, Birds, p. 101, pi. 38 (Charles Island). — Salvin, Trans. 



Zool. Soc. Lond., ix, p. 481, pi. 9, 1876. — Sharpe, Cat. Birds Brit. 



Mus., XII, p. 10 (in part). — Ridgway, Proc. U. S. Nat. Mus., xix, p. 



521, 1896; Bull. U. S. Nat. Mus., 50, Pt. I, p. 502, 1901. 

 Geospiza nebulosa Gould, Proc. Zool. Soc. Lond., p. 5, 1837. — Sharpe, 



Cat. Birds Brit. Mus., xii, p. 11 (in part). 

 Geospiza albemarlei Ridgway, Proc. U. S. Nat. Mus., xvii, p. 362, 1894; 



Bull. U. S. Nat. Mus., 50 Pt. i, p. 502, 1901. 

 Geospiza dubia albemarlei Rothschild and Hartert, Nov. Zool., vi, p. 



160, 1899. 

 Geospiza fortis fortis Koinscni'LT) and Hartert, Nov. ZooL, vi, p. 161, 



1899. 



Range. — Charles, Gardner (near Charles), Chatham, Indefatig- 

 able, Seymour, Duncan, Jervis, James, Albemarle (Tagus Cove) and 

 Narboro. 



This form presents the least departure from Geospiza fuliginosa 

 fuliginosa. In shape and size of the bill it is intermediate between 

 G.f. fuliginosa and G. f. dubia of the upper end of the G. fortis 

 series. Adults of G . fortis fortis oxiA G. fuliginosa ave always dis- 

 tinguishable by the larger bill of the former, but young birds of G. 

 fortis have bills almost exactly duplicating those of adults of G, fuli- 

 ginosa. We have examined young specimens which in fact could not 

 be definitely assigned to either species. 



Ridgway has described specimens from Albemarle as Geospiza 

 albemarlei. His specimens probably came from either Tagus Cove 



