

position, whence the most perfectly operculated Palmeria, Vestiaria, Drepanis, and 

 Hemignathus can be traced. On the other hand, Psittacirostra has still a small but 

 distinct opercular flap, then follows Loxioides, and lastly Pseudonestor, with no operculum 

 whatever and with round nostrils embedded in soft surroundings. Bhodacanthis and 

 Chloridops make a side departure from the last two genera ; there is no operculum, 

 but the nostrils are long-oval, embedded in soft surroundings ; the soft portion bordering 

 the lower margin of the nostrils lies in a deeper level than the dorsal border of the 

 nostril, and seems to be the remnant of the inner lower flap which is so common 

 in many of the Drepanididae, see Part II., PI. III. fig. 37. Anyhow there is no detailed 

 resemblance of the narial arrangement of other birds with any of the Fringillidae, while 

 Phodacanthis and Chloridops run surprisingly close to the Tanagrine Pitylus (not to 

 the Fringilline Pheucticus, Paroaria, or Chloris), and still greater is the resemblance 

 between Psittacirostra and Tanagra, e. g. bonariensis. An absolute distinction 

 between any of the thick-billed Hawaiian birds and the Fringillidae seems to be 

 that in the latter the generally open and roundish nostril is blocked, so to speak, 

 from the inside by the anterior little concha naris, which projects into the fundus of 

 the nostril. 



We have here remarkable cases of collateral or convergent development, as exhibited 

 by the numerous members of the Drepanididse. The long- and slender-billed forms 

 have developed features which make them very similar to certain Meliphagidae 

 (cf. Arachnothera and Hemignathus), and in the case of Drepanis pacifica and Acrulo- 

 cercus the resemblance has become startling. Equally near or puzzling has become the 

 approach to various Fringiilidae and Tanagridae by the thick-billed Hawaiian birds. 



The analogies with Fringiilidae extend even to some unexpected details of structure 

 as well as habits. Mr. Perkins says (' Aves Hawaiienses,' Part VI.) that Pseudonestor 

 is " in its movements Parrot-like in the extreme, especially in the varied hanging 

 attitudes that it assumes, while the similarity is still further increased by the shape of 



its beak generally clinging to the under sides of the thin branches or twigs, the 



head raised above the upper surface." Does all this not remind us at once of Loxia 

 pityopsittacus % 



Curiously enough, there exists another still more striking analogy between the 

 Crossbills and some Hawaiian birds, namely with Loxops inch Chrysomitridops. As 

 already known to Cabanis, when he established the genus Loxops, the under jaws of 

 these little birds are not symmetrical — the distal half of the under jaw is twisted either 

 to the right or to the left. It is interesting to note that the amount of twisting varies 

 individually, right- and left-billed specimens occurring in equal numbers, and that it 

 is smallest in young birds. There is not the slightest doubt that this asymmetry is 

 acquired individually by their twisting open husks or seeds, or cracks of bark, in search 

 of their food. 



It would be a case of great rashness to look upon the twisted bills of Loxops and 

 the pendulous attitudes of Pseudonestor as confirmatory indications of their Fringilline 

 affinities. Now it so happens that Loxops and Chrysomitridops combine with their 



