STRUCTURE AND RELATIONSHIPS OF OPISTHOCELIAN DINOSAURS. 195 
ISCHIUM: M 
ME Gy lL 5 mem Oy Nn ee tT eile se URE a Sete eee SIU 1.170 
MememCeTONtitialcend £) tse aa ee ae a ae Seb 7 
IPRA TAURI OTIC cer cs Se ee SRA Se letond. fumed ees «350 
PRL SiTALG S20 an ls oes ey eo ers es ae Ps Be Ta 
PUBIS: 
(yah ORE SR as REE in AE: SE Se ates Ns EEN SE DY RNA Sy 2 1.210 
meeremoceagth at proximal end-se os. o ka at me age .860 
mem eacith ar, Gistal 6nd. S465 bed ee ee ee a ates A480 
MPP IEROGH OU GNA tan Ne" Ee ee SE OR eee ee ae Oe Ue TO 
RESTORATION. 
The accompanying figure of the skeleton of Apatosaurus (Plate 
Lill) is based upon the Museum specimen as figured in Plate Lu. The 
cervical series and the limbs are supplied according to Marsh’s figure, . 
with certain modifications where it is known to be in error. As the 
complete cervical series is not known, the number of vertebra and the 
character of the neural processes are more or less conjectural. The 
skull is regarded as essentially unknown. The figure of the Museum 
specimen is made with the greatest care from projection drawings of 
each bone made separately and assembled in the figure as a whole. 
The distortions not corrected in working out the specimen are copied 
exactly in the drawing. Supplied parts are indicated by dotted lines 
in the separate figures. 
The most noticeable feature brought out is the shortness of the 
body as compared with other proportions. ‘The crest of the dorsal 
arch stands just in front of the sacrum. The long spines in this region 
evidently served as the center-pole to which the muscles for lifting 
both the anterior part of the body and the tail were attached. In 
general outline a striking similarity will be noticed between this figure 
and Marsh’s first restoration of this genus. In fact, as pointed out in 
an earlier note, the original restoration represented almost the entire 
skeleton and was, in the main, correct. Later, Professor Marsh evi- 
dently became dissatisfied with its proportions and inserted three ad- 
ditional posterior dorsal vertebrz, a lumbar, and two cervicals.. The 
number of ribs was raised from ten to thirteen, the crest of the dorsal 
arch was carried forward to the middle, and a sufficient length of cos- 
tal cartilage was inserted to extend the thorax in proportion. These 
additions mark almost exactly the extent of his error. 
