42 MR. ST. GEORGE MIVART ON THE 
THE AXIAL SKELETON OF DINORNIS. 
The osteology of Dinornis has been in great part admirably described in the different 
memoirs of Professor Owen in the ‘ Transactions’ of the Zoological Society (see vol. iii. 
plate 19. figs. 1, 2, 3, plates 20 & 204. fig. 1, also plate 43. figs. 1, 2, 3; vol. iv. 
p. 17 and plate 4. figs. 1-4, p. 159 and plate 47; and vol. vii. p. 115, plates 7, 8, 9). 
Nevertheless it is not yet possible to give a complete description of its axial skeleton, 
although the magnificent specimen of D. didiformis above referred to affords an 
extensive amount of information on the subject. 
There appear to be fifteen cervical vertebrae and probably three cervico-dorsal, three 
dorsal, and two dorso-lumbar vertebree. There are also some eight or nine lumbar and 
lumbo-sacral vertebrae, and three sacral; but the number of caudal vertebre does not 
appear. 
THE VERTEBR. 
The cervical vertebre in size and proportion of parts present a close general resem- 
blance to those of Dromeus and Casuarius; but the fosse postaxiad to the neural spines 
appear to have been yet larger. The neural spine of the axis is lofty ; and those of the 
fourth, fifth, sixth, and seventh cervical vertebre are so grooved axially that they bifur- 
cate laterally. The hyperapophyses of the axis are about as large as those of Dromeus 
and Casuarius. 
Catapophyses appear already at the fourth vertebra, and continue on till, in the last 
cervical, they are replaced by a median hypapophysis. ‘The cervical styloid ribs were 
most probably developed much as they are in Caswarius. 
The dorsal and lumbar vertebre evidently had not equally elongated and much 
axially extended neural spines, as in Apteryx, but probably about as developed as in 
Droméus and Casuarius. 
The d/umbar and lumbo-sacral vertebre are broad; and the more postaxial of them 
(the lumbo-sacral) are devoid of transverse processes. The absence of these, occasions 
the presence of two large crescentic fossee in the pelvis, one being internal to each 
acetabulum. 
The sacral vertebre send transverse processes to abut against the postacetabular part 
of the ilium as usual; but they are not sharply differentiated from the more postaxial 
vertebree as they are in Apteryx and Struthio. 
The sacro-caudal vertebre are more dorsally situated with respect to the pelvis, as 
regards the acetabula, than in any of the Struthionide except Rhea and Apteryax, and 
closely resemble those of the last-named genus in this point, though they differ from 
them in not being ventrad to the postaxial portions of the ilia. These vertebrae send 
out transverse processes like those of the sacral vertebra, and are thus hardly to be 
