212 On neiv and little-known Birds from Formosa. 



brown, with chestnut centres ; the rest o£ the underparts are 

 black barred with white ; there are traces of white spots on 

 the basal third of the outer webs of the primaries, and im- 

 perfect white bars across the inner webs of both the primaries 

 and secondaries. Length of wing 5*1 inches, tail 2'2, culmen 

 1'15, tarsus 1-75, middle toe and claw 1-7. The first primary 

 is about as long as the secondaries and about 1"25 inch 

 shorter than the fourth, which is the longest. 



Upper mandible dark grey, lower mandible bluish grey, 

 shading into yellowish green on the gonys; legs and feet 

 greyish black ; orbital ring and irides yellow. 



No species of Rallina has been known to occur in China 

 or Japan, but a species has been described from one of the 

 Loo-choo Islands under the name of Euryzona sepiaria 

 (Stejneger, Proc. United States Nat. Mus. 1887, p. 395). 

 This appears to be a larger bird with a paler crown, and 

 much less white on the wings. The next nearest species, 

 geographically, is Rallina euryzonoides, from the Philippine 

 Islands. This species is about the same size as the Formosan 

 bird, and has about the same amount of white on the quills ; 

 but this character varies considerably, as some examples 

 from Luzon have white on both webs, like the Formosan 

 bird, whilst others have the white restricted to the inner 

 webs. The birds from the Philippine Islands have a chestnut 

 head when adult, and it is possible that such may be the 

 case with the Formosan bird, though there is not the slightest 

 appearance of anything of the kind. Compai'cd with imma- 

 ture examples of Rallina euryzonoides or Rallina superciliaris 

 from Malacca, the upper parts of the Formosan bird are 

 much darker. 



I am afraid the genus Rallina is a very bad one, and ought 

 not to be regarded as distinct from the genus HypotcBnidia. 

 The latest authority on the subject has certainly made a 

 curious mistake in the characters which are supposed to 

 separate them (Sharpe, ' Catalogue of the Birds in the British 

 Museum,' xxiii. pp. 2 & 3). In both these supposed genera 

 the culmen is decidedly shorter than the middle toe and 

 claw. 



