188;).] TROtHILlD.li, CAI'KIMULGID^, AND CYPSKLID.E. 897 



hvpohyals quite long, their cartilaginous ends curling well up 

 behind the skull. Minute cerato-hyals may ossify in Chordediles ; 

 but I have not observed them to do so in P. nuttalli. 



This anterior portion of the apparatus has the arrow-head form 

 seen among the class generally where that part is short, and the 

 glossohyal is not produced far forwards. 



Of the Ctipseline Skull. — Upon a lateral view of the skull of 

 Panyptila saxatilis the external narial aperture is seen to be very 

 large and elliptical in outline. The maxillary process of the nasal 

 descends in a str^iight line to join the maxillary below, wliicli latter 

 bone, like the preinaxillary, is very delicately constructed. I have 

 failed to find a free lachrymal in the Swifts, and if it does occur it is 

 very rudimentary. It is just possible that it may be incorporated 

 with the pars plana, as it is in the Swallows and some other 

 passerine birds. 



Owing to the narrowness of the mid-frontal region, the orbital 

 cavity, though verv capacious, is not relatively so deep as in 

 Chordediles, but lias the same general aspect it presents in the 

 Swallows. The interorbital septum usually contains one or two 

 extensive vacuities in it, and the greater part of the tract for the 

 olfactory nerve and its cranial exit is deficient in bonv support. 



The })ars plana is of a quadrate outline, very large and, I believe, 

 in all true Cypseline birds, of a tuberous conformation, as it here is 

 in the Rock-Swift. 



Panyptila has a quadrate-bone in many respects like Chordediles ; 

 the orbital process, however, is very much better developed, though 

 not quite so well as it is in the Swallows, where it wears more the 

 character of the Passeres. The two facets upon the mandibular foot 

 of the bone are almost exactly as we find them in the Night-hawks, 

 while the antero- posterior compression of the body is equally well 

 marked. Two narrow facets are found upon the summit of its 

 mastoidal head. 



The quadrato-jugal rod is slender and retains a uniform calibre 

 to the maxillo-jugal junction, when it becomes laterally compressed 

 before arriving at the expanded portion. 



Seen from above, the general contour of the skull of the Rock- 

 Swift closely resembles that of a Swallow ; in the former, however, 

 we find the median portion of the premaxilla constructed upon the 

 same plan as I described it for the Goatsuckers, being reduced to a 

 thread-like rod of bone between the insertion of the nasals behind, 

 and where it merges into the mandibular tip in front. 



This gradually expands after it passes the former point to make 

 rather a broad insertion as it abuts against the cranio-facial region, 

 leaving, as it does so, a conspicuous triangular vacuity on either hand, 

 between it and the nasals. 



The cranio-facial region is somewhat concaved, while tiie inter- 

 orbital space, upon this aspect of the skull, is narrow, in both of 

 which particulars the Swifts agree with the Hirundinidop. 



The vault of the cranium is smooth and uniformly convex, being 

 barely marked in the median line by a longitudinal crease. All of 



