1906.] OF THE TRACHEOPHONE PASSERES. 139' 



Synallaxinfe and Xenicida? it forms a large quadrangular plate 

 slightly hollowed along its external bordei\ 



The floor of the olfactoiy chamber appears only in a few cases 

 to be partially ossified. Thus, in Xiphocolaiites cdhicollis for 

 example, and possibly in other species of this genus, the anterior 

 portion of the cliamber is partly closed in, when the skull is seen 

 from below, by the inward extension of the palatal processes of 

 the premaxilla. 



The Premaxilla and JSfasal. 



Excej)t in the Dendrocolaptidse and Formicariidse, the 

 premaxilla presents no really distinctive characters, being but a 

 small, pointed, triradiate bone. In the Dendrocolaptidfe, 

 however, it is often produced, as in XipJiorhynchus^ into a long, 

 decurved probe. In the species in question this portion of the 

 beak is of considerable length, subcircular, and rod-like. As 

 a consequence of this modification, the anterior nai^es have shifted 

 backwards so as to lie immediately in front of the lachrymo-nasal 

 fossa, and have become i-educed to small oval apertures. 



The nasals in such genera as I have been enabled to examine 

 of the Furnariinfe, Synallaxinfe, and Philydorinfe, of Dr. Sharpe's 

 'Hand-list,' are of the schizorhinal type (text-fig. 50 g, p. 150). 

 And on this account, as well as on the evidence of the skeleton 

 generally, I suspect that it will be found advisable, on further 

 work, to merge all three in a single subfamily, and with these 

 will probably follow also the Sclerurinse, Margarornithinpe, and 

 GlyjDhoi-hynchinfe. 



The Maxillo-Jugal Arch. 



The maxilla, as usual, is completely fused with the premaxilla, 

 but its approximate size may be measured, generally, by the 

 antero- posterior extent of the maxiilo-palatine processes. These 

 differ somewhat markedly in the different groups here associated. 

 They will be found in what is probably their most primitive form 

 in the Pittidse, Conopophagida?, Formicariida?, and Dendrocolaptina?. 

 Herein the maxiilo-palatine process is triangular in form, the 

 apex of the triangle rising gently from a broad base which sweeps 

 round poster'iorly into the quadrato-jugal bar. The apex, it should 

 be mentioned, underlies the free end of the vomer. In the 

 Formicariidse the maxilla, at the point from which the palatine 

 process leaves, is pierced by pneumatic foramina. 



In the Pipridfe these processes are strap-shaped, with a pedate 

 free end, and stand out almost at right angles to the long axis of 

 the maxilla. Into the base of this bone there opens a pneumatic 

 foramen conspicuous when the skull is seen from below. The 

 form and arrangement of these elements recall the maxillo- 

 palatines of the Cotingidge, indeed the differences between the two 

 are a negligible quantity. 



In the Conopophagida3 and the Philepittida? these processes 

 are of considerable length and slenderness. They spring, in the 



