( 158) 



From Wenman Island we have also one single male only, but it is unite black. 

 It measures : — 



We slionld most certainly have expected a new species from Wenman Island, 

 bnt we can see no reason whatever to separate our specimen from typical Geospiza 

 strenua. 



While the differences in size between the 7n(tles and/emales are very small, the 

 young birds are mucli smaller than old ones. The accompanying figures will 

 illustrate that these differences occur alike in the various islands, and also that they 

 cannot satisfactorily be separated. 



3. Geospiza darwini sp. nov. 



This is perhaps a form originally evolved from the following species, G. coni- 

 rostris, but it has a much larger and heavier beak, and thus stands somewhat 

 between G. strenua and G. conirostris. 



Adult 6. Intensely black, feathers on breast, abdomen, and back slightly edged 

 with olive ; it differs conspicuously from the other large Geospizae, G. maynirostns, 

 G. strenua and G. conirostris, by the olive rump, a character more or less apjiarent 

 in the small Geospizae only ; under tail-coverts whitish buff, secondaries ti])ped 

 slightly witli butfy white, outer edge of primaries olive. Bill comj)resscd and 

 rounded, like in G. conirostris, but, unlike the other species of Geospizae, alunptly 

 narrowed 3 millimetres from the tip and elongated sharply to the point. 



Aditlt ?. Head, neck and throat black, slightly edge<l on each feather with 

 olive buff, rest of body blackish, broadly variegated with olive buff, wings brown 

 edged with dark buff. 4 c? J, 1 ? measure as follows : — 



Hab. Culpepper Island, Galapagos. 



4. Geospiza conirostris conirostris Ridgw. 



G. crmirostrii, Ridgway in /'. C.S. Xuf. .!/««. XII. p. loC, Fig. 2 (1890) ;.Kidgway in 1'. U.S. .\<it. 



Mus. XIX. p. 516 (1896), 

 G. media, I. sujn-a c. p. 107, Fig. 3 (1890) ; Ridgway, /. infra c. p. 577, PI. LVII. Fig. 13 (1896). 



Ridgway has separated two species of big-billed finches from Hood Island, 

 0. conirostris and G. media, and with only eight of the former and four of the latter 

 supposed species before him, this might have seemed very jilausible, but with our 



