1906.] OF THE TRACriEOPHON^E PA.SSERijlS. 149 



The autei'ior hiteral process in llylactes is of enormous length, 

 projecting far beyond the level of the end of the spina extern i. : 

 aloiig its free outer border are placed the articular surfaces for 

 the ribs. The spina externa is moderately long and slightly 

 pointed. The carina sterna is small. 



The sterna of all the remaining genera here discussed are of 

 the typical Passerine type, and present but few structural 

 variations : these, however, I will briefly indicate. 



Except in the case of the Philepittidte, all these sterna agree 

 in having a forked spina externa — in Philepitta it is unforked 

 and small, — a moderately long anterior-lateral process to which 

 alone the sternal ribs articulate, and but a single notch along 

 the posterior free boi-der, the processus intermedius being 

 wanting. In the climbing forms of the Formicariidas and Den- 

 di-ocolaptidfe the sternal plate is long and narrow, and the keel 

 shallow. In the Philepittidte and Pipridas the spina externa is 

 pierced on its dorsal aspect by a lai-ge pneumatic foramen, and 

 there is a second aperture in the sternum, immediately behind 

 the first. In some of the Synallaxinas the sternal foramen is 

 found, but the more anterior aperture is wanting. 



The coracoid grooves do not meet in the middle line, and have 

 prominent, sinuously curved, dorsal lips : the ventral lips are 

 thin and continued inwards on to the spina externa. These 

 grooves attain their maximum development in the Conopophagidse. 



The coracoid shaft is long and slender, but is apparently never 

 longer than the corpus sterjii. The procoracoid pi-ocess is never 

 large, and takes the form of a short, oblong, downwai-dly directed 

 plate; thus, though smaller than in the Ear3dsemidEe, it is larger 

 than in the higher Passeres. In the Hylactinse it extends upwards 

 to fuse with the inturned head of the acrocoracoid. The width of 

 the shaft at its base is increased by a short and nai-row flange of 

 bone along the outer border, but this never extends further forward 

 than the level of a line passing in front of the free end of the 

 spina externa. 



The scapula, except in its relation to the foramen triosseum, 

 which will be discussed presently, presents no feattu-es of sufiicient 

 interest to demand special notice. 



The ficrcida is of the typical U-shape, but varies as regards 

 the development of the hypocleideum and the form of its free 

 ends. The hypocleideum in Hylactes and in the Pipridse is vesti- 

 gial. In the Pittidfe it is linguiform and rises abruptly from its 

 base : in the Philepittidse it is almost quadrangular. It is 

 largest among the Dendrocolaptine, 8ynallaxine, and Formi- 

 cariine types, where it inclines to a cordiform shape. 



The natui-e of the articulations between the coracoid, scapida, 

 and fui'cula, where they meet to form the foramen triosseimi, are 

 interesting. They are of two kinds, one of which is peculiar 

 to the Hylactinas (text-fig. 51 e), while the other obtains more 

 or less exactly among all the other groups. 



In ffylactes—Conoipo'ph.a.gidss, — as I have already mentioned. 



