1899.J CRAT!fIAL OSTEOLOGY OF THE PAEEOTS. 21 



with a very short free extremitv, and the fossa for tlie temporalis 

 muscle is seen to extend npwards behind it, instead of being 

 merely overhung by it as in Psittacus. The squamosal process is 

 well developed, rather long and pointed at the end ; it is somewhat 

 shorter and broader in Lorius than in the others. The posterior 

 ramus of the prefrontal is well developed and extends behind the 

 middle of the orbit ; but it does nor. create a suborbital ring, 

 though, especially in Trichoglossus and Kos, it may come very near 

 to the squamosal. 



The posterior wall of the tymipanic cavity is formed after the 

 fashion of Fsittacux, but leaves an aperture of an apparently 

 ditferent shape, by reason of the greater forward growth of its 

 middle portion, so that the cresceutic form of the aperture, or 

 rather the development of a conspicuous notch below and another 

 above and posteriorly, is better marked. The latter or upper 

 notch is just below and behind the suprameatal tubercle. The 

 anterior wall of the tympanum shows (in all four genera) an 

 ascending bar or splinter of bone that walls off from the tympanic 

 cavity tlie articulation of the quadrate. This is a little point o£ 

 resemblaoce to Nestor, but it is the only one I can detect, and 

 unsupported it goes for nothing. The groove or area in front of 

 the suprameatal process is well-marked ; it is very much more 

 extensive than in Psittacus, for it extends into an excavated sur- 

 face on the squamosal process, reaching well in front of the 

 glenoid notch. The jugular foramen is exceedingly small, and the 

 posterior recess of the tympanum not large. 



The quadrate has two deeply separated heads ; the inner one 

 is verv small and bent inwards almost perpendicularly to the shaft. 

 The two sockets on the squamosal and prootic elements are dis- 

 tinctly and rather widely separate, and the latter is a small deep 

 hollow. 



The mandibular fontanelle is a minute orifice placed much 

 further back than in Psittacus. The basitemporal ridges are better 

 marked than in Psittacus, but they distinctly terminate below the 

 foramen for the vagus, and are separated by a notch from the 

 succeeding ridge which marks the under border of the parocci- 

 pital. It is true that both in Nestor and Psittacus there is at the 

 same point a slight change of direction and appearance of discon- 

 tinuity, but, especially in the former skull, the ridges are nearly 

 continuous. 



Family Cacatttid^. 



The Cockatoos possess certain cranial characters in common 

 and their skulls are easily to be recognized, but there are many 

 variations within the faniily and even within the restricted g-nus 

 Cacatua. 



The orbital ring is complete by union of the prefrontal and 

 postfrontal bones, and from' the hinder part of the suborbital bar 

 thus formed a strong process runs backwards to fuse, in most cases 

 though not in all, with the squamosal process, and thus (as has been 



