(- 401 ) 



small, aud others tliat, wore interinediate. The hype o^ rompress/ro.sfn'.'fha,& the base 

 of the upi)er bill shot away, but we measure it as follows: culmen 15, nostril 

 to tip 10-75, depth at base 10, width at base 8. A ? psittacula (No. 353 

 Webster-Harris expedition) from ludefiitigable Island : culmen 14, nostril to 

 tip 10-5, depth 10, width at base 8 mm. 



Geospiza psittacula affinis Ridgw. 



Grospiza affinis Rothsch. & Hart., Nov. ZooL. VI, 18',)9, p. 168. — Camnrhi/ni'hns affinis Ridgw., 

 Birds Xorth it Middlf Ama: I, 1901, p. 481. 



We have reconsidered the position of this form, and as we have two specimens 

 which are barely distinguishable in size from psittacula, we are bound to reduce it 

 to the rank of a subspecies. The only differences are a slightly smaller bill and 

 shorter wing. (See measurements Nov. Zool. VI p. 178, and Ridgway, I.e.) 



Geospiza incerta Ridgw. 



NiiV. ZoiiL. VI, 1899, p. 168. — Camarhynchus i/irertus Ridgw., liiifls Xorth i(- .Wddle Amer. I, 

 1901, p. 482. 



Unfortunately we have no new material of this form, but we must say that we 

 should not be astonished if a larger series should prove this form to be identical 

 with true G. psittacula. 



Geospiza habeli (Scl. & Salv.) 



Nov. Zool. VI, 1899, p. 168.— Oi ma rIii/ncJit,s tud«li Ridgw., Birds Xnrlhi- Middir Amer. I, 1901, 

 p. 480. 



We have received some fresh material from Mr. Beck, but it does not call for 

 any remarks. We may only say that this species differs considerably more from 

 G. psittacula than affinis and iiwerta do (while compressirostris is a synonym ai 

 G. psittacula), therefore we cannot understand why Mr. Ridgway places it between 

 psittacula aud affinis, which, in his arrangement, is followed by affinis and com- 

 pressirostris. Mr. Beck writes : 



" Geospiza habeli is not very rare in the higher portions of Abingdon. We 

 found some nests in a strip of thick green bushes and trees at the foot of a steep 

 slope on the top of the island. The nests were similar to those of Geospiza 

 crassirostris, and stood about 8 to 20 ft. from the ground. The eggs are similar in 

 colouration to those of other Geospizae, but one very heavily incubated set was dotted 

 with fine pale spots all over, resembling a pale set of our ' horned lark' {Otoro)'i/s 

 a.lpesf/-i.f)." 



No. 803, Abingdon 15. iv. lOol, ? (No. 958) shot. Nest situated in a green 

 thorn tree 12 ft. from the ground, in the middle of a thick patch of wood and under- 

 brush. (Composed of stems of grass, fine grass, moss and puffs of cotton. Four 

 fresh eggs. 22-8 x 1.5-4, 235 x 15-7, 23 x 1.5-0, 23-5 x 15-(i. 



Geospiza paupera (Ridgw.) 



Nov. Zool. VI, 1899, p. 169. — C'amnrhi/nrJius pauper Ridgw., Birds Xorth <f; Middh Amer. I, 

 1901, p. 48.?. 



We have nothing to add, having received only one new specimen from Green, 

 which, however, is the most adult male we possess. 



26 



