168 Clark, Notes on the Laysan Finch. LApril 



specimen, so that I can say nothing as to their relative length. 

 There are nine secondaries in one wing but there seem to be ten 

 in the other; the wing is quintocubital. There are twelve rectrices. 

 While the resemblance to Pipilo is marked, except in the number 

 of the primaries, it should be noted that the differences in ptery- 

 losis between Telespiza and Loxioides or Psitiirostra are trivial 

 and of no significance. 



Metatarsus.— The covering of the leg in Telespiza is so nearly 

 like that of Loxioides, as given by Gadow, that no further descrip- 

 tion is necessary. 



Alimentary canal. — Here again the resemblance to Loxioides 

 is so great, no detailed account is worth while.. As the bird had 

 been kept in capitivity several weeks, the contents of the stomach 

 are of no importance. The crop-like dilatation of the lower end 

 of the oesophagus is marked but there is no real crop. The in- 

 testine is about 250 mm. long and is very narrow, its convolutions 

 resembling those of Loxioides so closely, that Gadow's figure 

 would do for either bird. 



Palatine region. — The bony palate of Telespiza, so far as could 

 be determined without a thorough cleaning, resembles that of 

 Loxioides, as figured by Gadow, but differs in having a longer 

 interpalatine bone, so that the anterior ends of the pterygoids are 

 separated from the posterior ends of the palatines by a space of 

 2 or 3 mm. 



It is fair to conclude from the sum of these characters that 

 Telespiza is, as has generally been supposed, closely related to 

 Loxioides, and except for the nostrils, it is more like that genus 

 than any other. In view of the restricted distribution of Loxioides 

 and the much wider range of Psittirostra, one would naturally 

 have expected the latter to be the nearest ally of Telespiza. How- 

 ever as the three genera have, together with Rhodacanihis, almost 

 certainly come from a single stock, the failure of the evidence to 

 fulfil this expectation is of no significance. 



Finally, I cannot refrain from expressing the opinion, based on 

 the study of Gadow's results in connection with these observations 

 on Telespiza, that the apparent resemblance to the Fringillidae is 

 superficial, and that those ornithologists are correct who look 

 elsewhere for the ancestry of the fringilliform birds of the Hawaiian 

 Islands. 



