386 
divergence, they are intermediate between Dinornis ingens and Dd. robustus. The 
extremes of sternal modifications and the intermediate gradations in the Dinornithide 
are exemplified in the cut, fig. 35, p. 419. . 
The pelvis of Dinornis gravis includes a sacrum of seventeen vertebrae, of which the 
two anterior ones support movable ribs. 
The axis vertebra has a low antero-posteriorly extended neural spine, and a somewhat 
more produced but thinner and axially shorter hypapophysis from the hinder part of the 
under surface of the centrum. 
In the third cervical the neural spine is bifid, and the hypapophysis is reduced in 
size. This process disappears in the fourth cervical; and processes for muscular attach- 
ments from the under surface of the centrum begin to be developed from the attached 
base of the parapophysis at the sixth cervical. They form a pair of processes which, 
us the cervical vertebrae recede in position, approximate and seem to be represented in 
the fourteenth cervical by a single low hypapophysis in the centre of the under surface 
of the vertebral body. ‘The serial homologue of this process recedes in position and 
gains in length in the last cervical, here resuming the character it manifests in the 
second and third of the cervical series. The hypapophysis gains in length both longi- 
tudinally and vertically in the first dorsal; in the fourth the base of the process is co- 
extensive with the under surface of the centrum; in the fifth dorsal it subsides midway, 
and on the sixth the hypapophysis disappears. 
The length of the dorso-sacral series of vertebrae, which represent the vertebra of the 
trunk, equals that of the combined tibia and metatarse. In Dinornis casuarinus and 
Dinornis gracilis the trunk-series is shorter than those limb-bones: in Dinornis ele- 
phantopus and Dinornis erassus the reverse proportions obtain. 
Since the above notes were written I have been made acquainted, through the kind- 
ness of Capt. F. W. Hutton, C.M.Z.S., with the discovery of additional evidences of the 
present species of Dinornis in a swamp near Hamilton, province of Otago, South Island. 
Of these remains, now in the Museum of Natural History, Otago, Capt. Hutton 
writes:—“ D. gravis also appears to me to be a good species, although the tibia 
approaches very closely to that of D. casuarinus, but is more robust, the length being 
only about three and a half times the circumference of the middle of the shaft, while 
in D. casuarinus it is more than four times the circumference. The measurements, 
however, given by Dr. Haast (Trans. N. Z. Inst. i. p- 86. no, 13) of bones found én situ 
in the Glenmark Swamp appear to connect D, gravis with D. crassus. 
In the swamp at 
Hamilton the bones we 
re so confusedly mixed together that in the whole collection I 
have only two leg-bones that I am absolutely certain belonged to the same bird. They 
belong, I consider, to D. gravis, although larger than those figured by Professor Owen ; 
their dimensions are as follows :-— 
