570 INlr. R. C. L. Perkins — Intruduction to 



mention other purposes. In Oreomyza, on the other hand, 

 the tongue is much degraded from its normal structure, 

 while only two of the species, and those but on the rarest 

 of occasions, have been seen to suck honey, and then only 

 from the shallow " lehua " flowers. The genus is almost 

 entirely insectivorous and feeds chiefly on exposed caterpillars, 

 spiders, and moths. 



6. The Thick-biUed Species uf the Second Group. 



There still remain to be considered the thick-billed species 

 of the second division of the Drepanines which have no 

 similar forms in the first. 



There are seven such forms, distributed in no less than 

 six genera, one of the latter {Psittacirostra), with its single 

 unmodified species, ranging over the whole group of forest- 

 bearing islands. One species of a peculiar genus {Telespiza) 

 is restricted to the outlying island of Laysan ; another, also 

 forming a peculiar genus [Pseudonestor), is found only on 

 the mountain of Haleakala in Maui ; while three peculiar 

 genera with four species are confined to the large island of 

 Hawaii, namely, Rhodacanthis with two species and Lo.vioides 

 and Chloridops each with one. It is now generally conceded 

 that all these forms are only extreme modifications of the 

 more normal Drepanines. In my published notes it is true 

 that I placed this section under the Fringillidse, but I did so 

 merely in deference to the opinions of systematic workers, 

 Messrs. Wilson and Evans and Rothschild, and more par- 

 ticularly to those of Dr. Gadow, who had availed himself of 

 the opportunity of carefully studying the different forms side 

 by side, whereas at that time I had secured no such facilities. 

 Personally I was convinced that all belonged to one family — 

 Avhether called Drepanididse, Fringillidae, or otherwise, — and 

 always maintained this in my cori'espondence against general 

 opposition, and that too at a time when Mr. Rothschild 

 himself was setting forth descriptions of the Drepanines 

 under such diverse families as Fringillidse and Meliphagidse ! 

 Although biological considerations first suggested to me the 

 common origin of all the present family — honey-suckers and 



