PROGNE. 281 



the back ; chin, throat, jugulum, and sides of head (below the centre of teh 

 eje) neck and body, with inside of wings, grayisli-brown, without gloss or 

 lustre (as in Cotyle riparia), rather lighter along the median line. Rest of 

 under parts dull white, not very sharply defined, passing behind into pure 

 white on the anal region and crissum — the latter having the shafts of the 

 longer feathers dusky, in contrast witli the snowy white of the plume. A 

 concealed white stripe on the sides under the wings, as in other Prague. Tibia 

 gray, the feathers tipped with whitish. 



The female (30,717) is quite similar, with much less gloss above, the white 

 of the belly apparently passing further forward, and still less sharply defined; 

 the throat a little lighter. 



(No. 30,718, -^j Guatemala.) Total length, 6.30; wing, 5.10; tail, 2.70; 

 perpendicular depth of fork, .53 ; ditferenee between 1st and 9th primary, 

 2.55 ; length of bill from forehead, .50, from nostril, .28, along gape, .84 ; 

 width of gape, .62; tarsus, .49 ; middle toe and claw, .73; claw alone, .23; 

 hind toe and claw, .46 ; claw alone, .22. 



This Mexican and Central American species has generally been 

 considered to be identical with the West Indian P. dominicensis, 

 but a comparison of large series of specimens shows considerable 

 differences. It is decidedly smaller, and the depth of fork of tail 

 only two-thirds as great. As to color, none of the specimens before 

 me exhibit any trace, on the under surface of the body, of the 

 glossy steel blue of the back, found in males of dominiceni^is ; in 

 this respect resembling females and immature males of the latter 

 species. From these, however, it may be distinguished by smaller 

 size — even in the males — and by having the shafts of the longer 

 crissal feathers dusky, instead of being pure white. The white of 

 the belly is less abruptly defined against the gray of the breast 

 (which is darker also), and apparently occupies a wider space. 



The male bird (No. 30,718), the measurements of which I have 

 given, appears to have a disproportionately short foot. In No. 

 30,*7n, supposed female, from Duehas, the dimensions of the leg are 

 as follows: Tarsus, .56 ; middle toe and claw, .81 ; claw alone, .21 ; 

 hind toe and claw, .51 ; claw alone, .26, or nearly the size of P. 

 subis with longer middle toe and claw. Other specimens are inter- 

 mediate somewhat in this respect. It may be that the differences 

 indicate a second species, but I cannot define it from the materials 

 at my command. The only other difference I note is a greener or 

 less purple gloss to the back. 



It is barely possible that fully adult males of this species may 

 have steel blue throats, as in true dominicen.ns, although I find no 

 allusion to the fact in any description. Even in this case, however, 

 the smaller size, less deeply forked tail, and dusky shafts of the longer 

 crissal feathers will distinguish them. 



