Cypsclidfe, Caprimulgidyej and Podargidie. 367 



species. I should be quite williug to do the same, were it 

 not for some specimens in the British Museum which stand 

 somewhat between the two. There is no constant difference 

 except in the paler colour. I have also recognized the 

 eastern bird as a subspecies, calling it pekinensis ; but I 

 must admit that it is by no means so distinct as M. murinus, 

 and that it is very difficult, and often impossible, to say to 

 which of these forms a given specimen should be referred. 

 Nevertheless, I think it was just as well to accept the eastern 

 paler form as a subspecies, as such a recognition draws 

 general attention to it, and may better tend to further 

 enlightenment on the subject, than if it had been passed 

 over in silence. 



I have no evidence that M. murinus ever breeds south of 

 the Mediterranean. Mr. Whitaker (above, p. 97) says that 

 M. apus breeds ''at Tunis and other towns in the north of 

 the Regency^^; but he kindly informs me {in litt.) that speci- 

 mens were not shot, or at least not preserved, and therefore 

 his statement is open to doubt. I have seen specimens from 

 the north coast of Tunis, shot during the breeding-time, 

 which were all M. murinus, and it is not likely that both 

 breed in the same places. 



The white-rumped Swifts of Tunis, which were recorded 

 as M. ciffinis by Koenig, are described as a new species 

 {M. koenigi) by Reichenow. They are, however, in my 

 opinion, not specifically different, but may be regarded as 

 an extreme M. yalilejensis, Antin., if that form is kept sub- 

 specifically distinct (see Cat, B. xvi. pp. 454,455). Peruvian 

 examples of M. andecola have been separated as M. a, par- 

 vulus by Berl. & Stolzm. (P. Z. S. 1892, p. 384), on the 

 ground that they are smaller ; but I am not convinced that 

 the small size is peculiar to Peruvian birds, for the skins 

 in the British Museum do not agree with that theory. 



M. niansee, Rchw., which I had not seen when writing 

 the Catalogue, is a good species, and may be described as a 

 very small M. (Bquatorialis. 



The above-mentioned M. willsi, described in ' Novitates 

 Zoologicse,^ iii. pt. ii., shows a similar relation to M. melba, 



