346 MESSES. A. AND E. NEWTON ON THE OSTEOLOGY OE THE SOLITAIRE. 
and the ungual segment (Plate XX. figs. 108, 109) proportionally longer. In the inner 
toe the proximal phalanx (Plate XX. figs. 128-130) is the longest in the whole foot, 
whereas in Didus it would appear to he about equal in length to that of the middle toe. 
This (Plate XX. figs. 116-118), again, is relatively and jn large (male!) specimens 
absolutely longer and especially thicker than that of Didus , and the next two phalanges 
of this toe (Plate XX. figs. 119-124) are in proportion to it, but the ungual segment 
(Plate XX. figs. 125-127) is longer than the third. Of the outer toe the first (Plate 
XX. figs. 110-112) is the shortest of the proximal phalanges of the whole foot. The 
other segments of this toe, except the ungual (Plate XX. figs. 113-115), are missing. 
This is also, like the rest of the ungual phalanges, larger and especially longer than 
in the single corresponding example of Didus. 
§ 9. Skull. 
Of the cranial portion of the skull (Plates XXI. figs. 139-147, XXII. figs. 149, 150) 
the collection includes five specimens, more or less imperfect, but all containing the 
foramen magnum, which, as in Didus , but not to the same extent, varies not a little in 
shape, one specimen (of a male'?) having it much wider below than above, but in the 
others the upper and lower curves are more nearly alike. So also the occipital condyle 
is to some extent inconstant. In the (male!) specimen just mentioned, and in two at 
least of the smaller ones, it slopes gradually from its base ; but in an example of median 
size it rises abruptly, and on one side (the left) slightly overhangs its base, so that 
regarding it from that aspect it looks as if it were pedunculate. 
Three of the specimens (Plate XXI. figs. 141, 144, 147) are sufficiently perfect to 
show the supraoccipital orifice, the existence of which in Pezophaps was surmised by 
Dr. Melville (op. cit. p. 114). The other small foramina and protuberances correspond 
singularly with those in Didus , exhibiting the same amount of individual variation. 
On the other hand the cranium of Pezophaps differs remarkably from that of Didus in 
several important features, some of which have beep, already pointed out by Dr. Melville 
( loc . cit.), while others will be rendered more apparent by an inspection of the accom- 
panying figures. It is narrower and longer, entirely destitute of the peculiar frontal 
protuberance, and generally more compressed and less ventricose, while the orbital cham- 
bers are larger. In one respect, however, that accurate osteologist has been misled. 
He attributes the apparent depression of the central tract of the upper surface entirely 
to the stalagmitic deposit on its anterior and posterior parts in the only specimen then 
known. In the specimens in the present collection, which are wholly free from that 
deposit, this apparent depression is equally to be observed. It is due, however, to the 
elevation of the anterior and particularly of the posterior parts into two bony ridges of 
the cancellous structure which in Didus occupies the whole surface, and indeed is most 
prominently developed on its central portion (Trans. Zool. Soc. vi. pi. 23. fig. 1), but in 
Pezophaps does not rise in the middle to any considerable degree, and consequently 
leaves the exterior surface almost perfectly flat, while the posterior ridge, being continued 
