1906.] OF THE TRACHEOPHOXE PASSERES. 143 



parasphenoidal rostvum and passes over, and beyond, the hinder 

 ends of the palatine for some considei'able distance, so that what 

 Appears to be the hemipterygoid lies immediately under the pedate 

 expansion of the main shaft of the pterygoid. 



The differences which obtain among these groups in the 

 matter of the palato-pterygoid articulation are of sufficient 

 interest to demand some further notice. I have shown that the 

 Piprida?, Pteroptochin?e, Formicariinse, Dendrocolaptinfe, and 

 Synallaxinas all agree in common with the smaller Cotingid?e, 

 while the Philepittidte markedly differ therefrom. 



Now among the Cotingida^ this palato-pteiygoid articulation 

 appears under three forms. In what is doubtless the most 

 primitive the shaft of the pterygoid is continued forwards, 

 precisely as in the Philepittidse (p. 142) ; and this is well seen in 

 Chasmorhynchus for example, and, in a slightly more specialised 

 form, in Calyj^tomena among the Eurylsemid*. In Cotingids, such 

 as Aulia or Lathria, we have apparently the intermediate stage 

 between that seen in Philejntta and Chasmorliynchus on the one 

 hand, and that of the smaller Cotingids, such as Hadrostomus, and 

 the Formicariinpe — and the types associated therewith in this 

 connection — on the other. In this intermediate stage the ptery- 

 goid impinges against the parasphenoidal rostiaim, and is then 

 continued forward in the form of a pedate plate, articulating with 

 what is evidently a large hemipterygoid fused with the palatine, 

 by means of a sinuous joint. In the smaller Cotingids, and in 

 types such as ThcmnnojjMlus and Pteroptochus^ this joint has become 

 ■obliterated by anchylosis. 



I cannot recall a similar fusion between pterygoid and palatine 

 in any of the higher Passerines, but it occurs with some frequency 

 among the Coraciiform birds, as, for example, in the Capitonidaj 

 and Bucconidfe. 



The 2}cdatine has a scroll-shaped body turned with its concavity 

 inwards, and, fusing with the hemipterygoid, forms an articular 

 surface whereby it is enabled to glide backwards and forwaids 

 along the parasphenoidal rostrum. The ventral free edge of the 

 scroll forms, with that of the opposite palatine, a deep channel 

 along the roof of the palate immediately behind the vomer. 

 From the outer convex surface of this sci'oll there is given off, 

 near its anterior border, a narrow horizontal lamina, which sends 

 a short spur backwards to form the transpalatine process (W. K. 

 Parker) and a long, slender rod forwards to fuse with the palatine 

 process of the premaxilla. Except in the Pittidse, the maxillo- 

 palatine processes lie so far back that their free ends touch, or 

 almost touch, the antero-inferior angle of this palatine scroll ; but 

 in the group in question these processes have shifted forwards so 

 that their free ends are far removed from any relation with this 

 portion of the palatine. The transpalatine spur is especially long 

 in Xiphorhynchus (text -fig. 50,/, p. 140), and this is apparently 

 •correlated with the remarkable form of the beak. 



The Quadrate. — As in the Eurylfemidfe, though not in so marked 



