Camarasaurus is an unappreciated sauropod. It wasn't the heaviest or longest of the earth-shaking dinosaurs, but the blunted skull and large teeth of the Jurassic sauropod indicate that it had a different lifestyle than the more famous Diplodocus and Apatosaurus. In 1920, paleontologists at the AMNH created a skeletal and muscular reconstruction of the dinosaur, Camarasaurus being proportionally bulkier for its size than other sauropods then known.
The reconstruction of a model of Camarasaurus had another purpose, as well. In 1908 Oliver Hay published a paper advocating a sprawling, crocodile-like posture for Diplodocus, a posture inferred for all large sauropods. Osborn and others did not think such a pose reasonable, though, and the Camarasaurus restoration served to illustrate that sauropods carried their legs underneath their bodies rather than out to the side, although they still were a little knock-kneed. From a 1920 overview of the model by W.K. Gregory;
The muscular restoration of Camarasaurus was preceded by careful studies of the consecutive poses of the skeleton ofApatosaurus (Brontosaurus) in three phases of the stride. These poses were worked out in a minature model of the complete skeleton, with flexible joints. The conclusion reached is that Apatosaurus was a walking and not a crawling quadruped. As compared with living crocodiles the body was well raised off the ground in the stride. This is contrary to the ideas of Tornier, Hay and others, who advocated a crawling stride. The limbs, however, had not the fore-and-aft pendulum action of the perfected mammalian quadrupeds; the stride had a wide, lateral swing.
Unfortunately no illustrations of the life model were included with the paper, but the text makes it clear that the AMNH reconstruction was at odds with the concept of sprawling sauropods. Camarasaurus was not an aquatic creature like Osborn's vision of Diplodocus, though. The overall conformation of its features seemed to be more consistent with an animal that spent most of its time on land, at best wading into the shallows;
Camarasaurus might well have been an efficient wader. It was positively devoid of special adaptations for swimming; the pectoral and pelvic arches, and the limbs and backbone were adapted to support the great weight of the body presumably on land, as well as in wading; the tail was relatively small and feeble. There was a high ratio of limb power to weight.
[Actually, this is something of a revision of the habits of Camarasaurus. In an 1898 paper, Osborn thought it was both a wader and an apt swimmer, but apparently 20 years later Camarasaurus would not longer be using its tail to propel itself through freshwater lakes.]
The Camarasaurus restoration was not based upon newly-exhumed material, either. In 1877, O.W. Lucas contacted E.D. Cope about some fossil bones found in Canyon City, Colorado, which were dug up and brought back east for description. Among the bones recovered were those that Cope used to describe Camarasaurus that same year (thus having priority over Marsh's "Morosaurus"). The bones were then used by John Ryder to make a life-size restoration of the animal (I assume on paper and not as an actual mount), one of the first full-scale reconstructions of a sauropod dinosaur. Ultimately, however, Cope's collection was obtained by AMNH president Morris K. Jesup in 1902, allowing H.F. Osborn and his cadre of paleontologists a chance to reanalyze Cope's collection.
Gregory, W.K. (1920) "Restoration of Camarasaurus and Life Model." Proceedings of the National Academy of Science, Vol. 6, pp. 15
Osborn, H.F. (1898) "Additional characters of the great herbivorous dinosaur Camarasaurus." Bulletin of the AMNH, Vol. 10, pp. 219-233
Osborn, H.F.; Mook, C.G. (1919) "Camarasaurus, Amphicoelias, and Other Sauropods of Cope." Bulletin of the Geological Society of America, Vol. 30, pp. 379-388
Osborn, H.F.; Mook, C.G. (1920) "Reconstruction of the Skeleton of the Sauropod Dinosaur Camarasaurus Cope (Morosaurus Marsh)." Proceedings of the National Academy of Science, Vol. 6, pp. 16-17