In the “Claosaurus” post earlier this week, I temporarily(?) lost my mind when I said that the Hadrosaurus Benjamin Waterhouse Hawkins assembled for the Academy of Natural Sciences was 100% plaster reconstruction. Thanks to John Sime, among others, for pointing out that this was incorrect. As usual, the truth is more complicated, and therefore much more interesting.
The Hadrosaurus project began when Hakwins was commissioned to create a series of life-sized prehistoric animals for display in New York City’s central park, under the direction of Joseph Leidy. The exhibition was cancelled when Hawkins’ on-site workshop was burned down by vandals, but he was able to salvage the Hadrosaurus skeleton for display at the Academy in Philadelphia. This reconstruction was based on little more than two limbs and a handful of vertebrae. It was a well-reasoned attempt – and it drew huge crowds – but it wasn’t long before new dinosaur finds rendered it obsolete. In 1901, Charles Beecher wrote that the Hadrosaurus mount had “long since ceased to have any value or interest except as a historical attempt.” No longer considered informative, the original Hadrosaurus was probably dismantled around the start of the 20th century. At least three plaster copies were distributed to other museums, but these were also discarded long ago.
There is no question that Hawkins’ reconstruction doesn’t reflect our present understanding of this animal, so in that sense it isn’t “real.” Still, it is of historic interest whether Hawkins used the handful of original Hadrosaurus fossils in the mount itself, or whether the entire display was fabricated. There is precedent for both posibilities: John Peale mounted an original mastodon skeleton in 1802, but the Smithsonian’s first attempts at Basilosaurus and Triceratops (1895 and 1900 respectively) included no real fossils. This question was actually up for discussion as early as 1926. Responding to an inquiry from Peabody Museum paleontologist Richard Lull, Academy of Natural Sciences curator Witmer Stone wrote that the Hadrosaurus mount was a complete reproduction. When Lull followed up with William Matthew of the American Museum of Natural History, however, Matthew recalled that “some or all of the original bones were used.”
The two letters reproduced below are in the collection of the Department of Vertebrate Paleontology Archives at the Peabody Museum of Natural History, and are shared with permission.

Letter from Witmer Stone to Richard Lull, January 26, 1925. Courtesy of the Dept. of Vertebrate Paleontology Archives, Yale Peabody Museum of Natural History.

Letter from William Matthew to Richard Lull, January 30, 1925. Courtesy of the Dept. of Vertebrate Paleontology Archives, Yale Peabody Museum of Natural History.
A look at the original Hadrosaurus fossils, now cataloged as ANSP 10005, suggests that Matthew was correct. At least a couple of the bones appear to bear drill holes, a tell-tale sign that they were once fastened to an armature. Likewise, in a photograph of the mount in Hawkins workshop, the elements that were actually recovered – the left leg*, part of the pelvis, and a scattering of vertebrae – appear to be darker in color. This suggests that these are the real bones, and the rest of the skeleton is plaster…unless Hawkins painted plaster casts to demonstrate which elements had been found.
*Note that the image below has been flipped horizontally for some reason. In the original, the left side of the skeleton is facing the camera.
The answer to this little conundrum can be found in the official guidebook to the Academy of Natural Sciences, published in 1879. Apparently there were two versions of Hadrosaurus on display. The original 1868 mount did include the original fossils, but when the museum moved to a larger facility in 1876 (in part because of the spike in visitation caused by the Hadrosaurus exhibit) the mount was remade. The bones were not faring well in open air and were rapidly deteriorating, so they were retired to the collections and replaced with casts. Anyone who saw the Hadrosaurus before 1876 saw the fossils incorporated into the mount, and anyone who visited later saw a complete facsimile. Still, I’m pretty sure William Matthew was remembering incorrectly. He was born in 1871, so unless he was carefully observing the composition of the mount at age 5, he shouldn’t have seen the original version!
References
Beecher, C.E. 1901. The reconstruction of a Cretaceous dinosaur, Claosaurus annectens Marsh. Transactions of the Connecticut Academy of Arts and Sciences. Vol. 11, pp. 311-324.
Carpenter, K., Madsen, J.H. and Lewis, L. 1994. Mounting of Fossil Vertebrate Skeletons. Vertebrate Paleontological Techniques, Vol. 1. Cambridge, UK: Cambridge University Press.