1906.] OF TUE TRACHEOPHOXE PASSERES. 155 



the forms now under discussion. These trochlese are not only 

 all in the same pla,ne, but they are also all of the same length ; in 

 their relative sizes, however, they show not inconsiderable 

 diflerences. 



The Oonopophagin«! show the least specialised condition of 

 these parts, in that all trochlene are large and placed wide apai't. 

 The II. is turned inwards. In the Pteroptochinse all are small : 

 and they ai-e still further, relatively, reduced in the Pipridie, 

 especially in so far as lY. is concerned. In the Pipridaj III. is the 

 largest, IV. the smallest. 



Among the Formicariine and Dendroeolaptine types there 

 appears to be a marked tendency to reduce the IV., and this is well 

 seen in Homorus^ where trochlese II. and III. are large, and IV. 

 extremely small. The same is true of Pseudocolaptes and Baiara, 

 though not to so marked an extent. In Devdrocolajdes the 

 trochleje are wide apart, and III. is deeply cleft in the middle. 

 Xiphorhynchus is peculiar in having a rather small II. trochlea 

 which is turned inwards, while III. is deeply cleft as in Dendro- 

 colaptes. 



The depression for Mc. I. is generally well marked, especially 

 so in Homorus. 



The Formicariine, Dendroeolaptine, and Synallaxine types agree 

 with the Conopophagidfe in that Ph. I. of D. IV. is only about half 

 the length of the same phalanx in D. II. III. In Pseudocolaptes 

 all the phalanges of this row are remarkably short, the length 

 decreasing from within outwards; and in X i'phorhynchus this 

 abbreviation has been carried to excess, all the phalanges of this 

 vow being subequal and extremely short, while in both genera 

 they lie closely pressed together. In Dendrocolaptes these 

 phalanges are also short and subequal. 



ix. Summary. 



Miiller (6) was the first to utilise the structural characters of 

 the syrinx for systematic purposes. He it was a\1)0 coined the 

 term " Tracheophone Passeres " and brigaded together the forms 

 possessing this type of windpipe. But his arrangement of the 

 Passeres as a whole was unsatisfactory, inasmuch as he failed to 

 discriminate between what we now regai'd as Passeres and the 

 outlying forms which go to make up the " Coraciifonnes." 



Huxley appears to have been the first to difl;erentiate between 

 the Passeriform and Coraciiform types : while the further subdi- 

 vision of the Passeres seems to have been first placed on a satis- 

 factory footing by Sclater and Salvin (10). These authors adopted 

 Miiller 's " Tracheophonea?," but for the sub-division of the group 

 which they found necessary they employed such characters as 

 were aflTorded by the scutellation of the tarsus, the shape of the 

 tail, and so on. Garrod (6) improved on this, and it will probably 

 materially aid those who may read this paper if Garrod's scheme 

 is given here. It is as follows : — 



