526 ON THE ANATOMY OP PASSERINE BIRDS [JunC 5, 



fied, except the last, which is developed into a three-way piece from 

 the presence of a bar running from before backwards at the middle 

 of the lower margin. The first bronchial half-ring is of the same 

 flattened and deep nature as the tracheal rings, and, like them also, 

 is not separated from the three-way piece by any interval. To its an- 

 terior end, on each side, as well as to the front of the three-way piece, 

 the intrinsic muscle is attached, which descends, broad and thin, down 

 the front of the lower part of the trachea, in contact with its fellow 

 of the opposite side, there to terminate (vide Plate LIII. fig. 8). 



The second bronchial semi-ring is not modified. It is separated by 

 a short interval from the first, and by a strikingly considerable one 

 from the third, which is the commencement of the normal bronchus. 

 I could not find that the muscles of the syrinx sent any fibres to 

 this second ring, as in Pachyrhamphus atricapillus, described by 

 Miiller, although otherwise this structure is almost identical in the 

 two birds. If they are present they must be extremely feeble ; and 

 the relative distances of the upper bronchial semi-rings favours the 

 >'iew that some special arrangement exists. 



The account, above given, of the voice-organs in the aberrant 

 Passeres in question, is entirely confirmatory of the results arrived 

 at by Johannes Miiller. Both Hylactes and Grallaria are com- 

 pletely tracheophone, as he predicted they would be found to be, 

 although they agree with one another, and differ from those pre- 

 viously described in having the syringeal end of the trachea less 

 abruptly distinguishable as being composed of two parts. Neither 

 Lipaugus, nor Heteropelma, nor Uadrostomus are far from the 

 mesomyodian types already known, as far as their voice-organs are 

 concerned, which structure clearly shows that our nomenclature is an 

 inefficient one when it places Hadrostomus as far from Pachyrham- 

 phus as either is from Tityra. 



Again, also, that the Pipridse and Cotingidse should be considered 

 to be different families is not borne out by the nature of the lower 

 larynx ; and it seems hardly possible to allow a difference in tarsal 

 scutellation to constitute a family difference, when not borne out by 

 more important points of internal structure. 



EXPLANATION OF PLATE LIII. 



Pig. 1. Syrinx of Grallaria yuatemalensis, front view. 



2. The same, back view. 



3. The same, lateral view, from within, showing the fusiform processits 



vocalis. 



4. Syrinx of Hylactes megapodius, front view. 



5. The same, back yiew. 



6. Syrinx of Lipaugus eineraceus, front view. 



7. The same, side view. 



8. Syrinx of Hadrostomus aglaia, front view. 



4 



