NO. 1116. rROCEEDrXGS OF THE NATIONAL MUSEUM. 509 



between ^^Cactornis''^ hrevirostris and Geosjnzafratercula, or between C. 

 projiinqua and G. conirostris. Furthermore, if tliis group be recognized 

 to consist of two or more genera instead of one, then, to be consistent, 

 Camarhynchus must also be divided into three genera. 



A very careful consideration of all the facts in the case, as they ap- 

 pear to me, compels me to conclude that the safest course is to regard 

 the species of '' Gactoruis^^ as simply more slender-billed Geosjnzci', the 

 degree of departure from the typical (Tcospizine bill being largely a 

 si)ecitic character. This is a conclusion which J regret having to adopt, 

 for I regard large genera as a nuisance, and wonld be glad to have the 

 slightest excuse for keeping Geospisa to its old limits. 



As a matter of convenience, I have separated the species into two 

 groups, which correspond m tlicir bmits with Gactoruis and Geospiza 

 as usually recognized;' but how slight and unsatisfactory a basis this 

 division rests upon may be seen by reference to the characters given in 

 the first part of the following " key to the species." 



Owing to the gradual transition from one form to another, and the 

 almost perfect resemblance betw een them in coloration, I have found 

 it impossible to construct an analytical "key" to the species after the 

 usual plan, but have drawn up ihe foHowing as an aid to their more 

 ready identiticatiou. 1 am prevented from making the "key" more sat- 

 isfactory by the circumstance that I have no specimens of G. assimilis 

 and G. bar nngtoiii for comparison with the other so-called Cactorni, and 

 am therefore unable to give comparative measurements of these forms 

 alongside of G.fatigata and G. abiiujdoni. 



KEY It) THE .STECIKS OK GEOSPIZA. 



rt'. Width of inaudible at base (across chin) nearly or quite ecjual to, sometimes 



iii-eater than, length of gonys Subgenus Geonpiza. 



a-. Width of mandible at base (across chin) much less than length of gonys. 



Snl>genus Cactornis. 

 Subgenus GEOSPIZA. 

 « '. Depth of bill at base greater than length of maxilla from nostril. 



h '. Bill very short and thick, its depth at base very much greater than length 

 of maxilla from nostril, 

 c '. C4onys more than 0.50 of an inch. 



d\ Wing 3.55-3.70, tail 2-2.10. (Charles and Chatham Islands.) 



1. (i. iiuujiiirosiris (p. 512). 

 d-. Wing less than 3..50, tail less than 2. 



e'. Depth of bill at base morr than 0.80; widtli of mandible at base 

 (across chin) more than 0.(55. (Tower Islaiul.) 



2. G. parhiirhynclia (p. 516). 

 eK Depth of bill at base less than 0.80; width of mandible at base 

 (across chin ) less than 0.65. (Abingdon, Biudloe, James, Jervis, 

 Indefatigable. Chatham, and Charles islands.) 



3. G. strenua (p. 514). 



'With this difference: Cactornis pallida, Sclater and Salvin, I have transferred to 

 Camarhynchus, to which it is certainly far more nearly related (see p. 544). 



