STRUCTURE AND RELATIONSHIPS OF OPISTHOCGLIAN DINOSAURS. 245 
etc. The short, stout metapodials and blunted phalanges character- 
istic of the Opisthoccelia would be as ill adapted for propulsion in 
water or upon marsh lands as are those of the elephant. The reduc- 
tion in the number of claws offers further evidence along this line, 
and finds a parallel in the great ground sloths. In short, if the foot 
structure of these animals indicates anything, it indicates speciali- 
zation for terrestrial locomotion. 
In the genus Brachtosaurus there are very different proportions of 
fore and hind leg, of pelvis and thorax and tail, from those of the 
usual type of opisthoccelian dinosaurs. To harmonize with them we 
may expect that decided differences existed in the head and neck. 
The foot structure in A patosaurus and Dtiplodocus may, as before stated, 
be regarded as highly specialized, but the length and slenderness of 
limb, the deep thorax, the broad sacrum, the expanded ilium, and 
the abbreviated tail of Brachtosaurus all point to a greater agility 
and a much better adaptation to terrestrial habits than is found in. 
any other representative of the Opisthoccelhia. 
The varying structure of the dorsal vertebre in the larger dino- 
saurs offers some interesting problems as to the habits of the animals. 
In some genera a relation between the length of the dorsal spines and 
the breadth of the zygapophyses is noticeable. We may well 
assume, with other writers, that the heavier forms, such as Apato- 
saurus and Dtplodocus, which are provided with long spines in the 
sacral and posterior dorsal region, were adapted to rearing up on the 
hind legs as is represented in the conventional mounted skeleton of 
Megatherium. In these forms we find that the body is short and 
therefore well adapted to this habit. Morasaurus and Camarasaurus 
are less perfectly adapted to such a habit by reason of their shorter 
spines, but might well have been capable of assuming the upright 
position. The slender shaft of the ischium in the former genus could 
have hardly afforded sufficient support in the sitting posture. In all 
of these, however, it will be observed that the zygapophyses are 
strong and placed far apart, especially in the anterior dorsal region 
of Apatosaurus. Their separation is doubtless directly. due to the 
bifurcate spines, but that arrangement was a concomitant of the 
development of a median set of dorsal muscles whose function it was 
to elevate the anterior portion of the body. The unusually strong 
zygapophyses served to support and control the ponderous shoulders 
and neck when the upright position was assumed. All of this bespeaks 
great mobility of the anterior portion of the body. 
In Haplocanthosaurus we have noticed that the spines are short 
posteriorly, but reach their greatest length at the middle of the dorsal 
