516 PROFESSOR OWEN ON THE 
Museum of the Royal College of Surgeons, which served as the subject of his pencil ” 
(op. cit., N., note, p. 332), seems to have been sent to press without due consideration. 
With respect to the pelvis of Didus, Mr. Erxleben drew no more of the pubic bones 
than the specimens at that time warranted. It was at my suggestion that this portion 
of bone, originally detached, was brought into contact with the ischium at two points, 
as it is in Didunculus. The more perfect specimens of the long and slender pelvic 
heemapophysis, since obtained and recognized, seem to show that the second junction 
does not take place, but that the pubis extends freely backward, with a graceful down- 
ward curve, and for an extent corresponding with the characters of the same bone in 
Pezophaps. (Compare Pls. LXIV. and LXV. of the present Memoir with fig. 70, pl. 18, 
N., and the restoration in dotted outlines in fig. 179, pl. 24, N.) Nevertheless a pelvis 
with the whole extent and entire lower border of the ischium seems still to be a deside- 
ratum in the collections of the bones of both Didus and Pezophaps which have as yet 
reached England. ‘The better-preserved sacral elements of the pelvis permitted sixteen 
vertebre to be counted in that extensive anchylosed mass of bone-segments. Messrs. 
Newton state that one specimen of pelvis of Pezophaps, complete in its posterior half, 
‘has eighteen coalesced sacral vertebre.” It is to be regretted that this specimen is 
not figured ; the subjects, at least, of figs. 66, 68, 69, & 70, in their paper, are plainly 
mutilated behind. The two “ perfect examples” [of sacrum ?] “of Didus ineptus show 
only sixteen (vertebra), which is probably the normal number in that species.” Op. cit. 
N., p. 334. 
The essential characters of the pelvis show a close correspondence in Didus and 
Pezophaps. ‘The articular surface of the centrum of the last dorsal” [first ‘ sacral’ by 
the character of confluence] “is in Pezophaps almost exactly as in Didus”'. Other pelvic 
correspondences are seen in the general shape and disposition of the ilia, which, how- 
ever, are not developed behind in Pezophaps so as to give the flatness and breadth to 
the posterior half of the pelvis which seem to specifically characterize the Dodo. The 
position of the skeleton in Pl. LXV. has been selected to exemplify this peculiarity. 
Other particulars, especially the more essential ones, such as the length, curvature, 
and movable articulation of the ribs of the first sacral vertebra*—the confluence, short- 
ness, and straightness of the pleurapophyses of the next three sacrals—the suppression 
of the rib-elements in the three succeeding vertebra, and their reappearance in the 
eighth and sometimes in the ninth sacral as strong abutments against the ilia above and’ 
N., p. 334. 
? In this, as in my former paper, I adhere to the usual characters of the sacrum afforded by coalescence. 
Messrs. Newton are influenced by its extent—and where it leaves the ribs free, reckon such yertebre as 
“dorsal.” Accordingly my “first sacral’ is their “last dorsal.” Anchylosis, like most of the characters of 
the classes of vertebra in anthropotomy, is an artificial one, and might justify the ascription to the Colwmbacei 
or “Gemitores” of four sacrums, viz, ‘ caudal,’ “ pelvic,” “ lumbar,” and “ dorsal ;” for the vertebrae answering 
to the lumbar and anterior caudals in Mammals and Reptiles are massed with the interacetabular or proper 
pelyic yertebree into one extensive and complex bone. 
