Agrosaurus: the Australian Dinosaur that Wasn't

In 1891, H.G.Seeley published a description of a prosauropod dinosaur, which he named Agrosaurus macgillivrayi. It was based on a few limb bones, a claw and a tail vertebra that had been "discovered" in the Natural History Museum in London, bearing the tag "Fly, 1844. Jn. Macgillivray, from the N.E. coast of Australia".

John Macgillivray was a naturalist with the 1844 expedition that consisted of four British ships, led by the Fly, to construct a beacon on Raine Island off the eastern coast of the Cape York Peninsular in Queensland, in the far north of the Australian continent. For four and a half months, while the beacon was being constructed, the ships' crews thoroughly explored the eastern side of the Cape York Peninsular, north of Cape Grenville.

John Macgillivray's task during the expedition was to collect animal specimens for Edward Smith Stanley, the 13th Earl of Derby. Unfortunately his note books have never been found, and it appears that he failed to mentioned collecting any fossil specimens in a book he wrote a few years later, during another expedition in the same area. A book was published by the geologist J. Bette Jukes in 1847 documenting the results of the 1844 Fly expedition, but this also failed to mention anything about the collection of fossil material.

Since the exact location of this "Australian" dinosaur was not known, in 1995 a group of ten palaeontologists and volunteers set about searching the rocks around Captain Billy Landing in the area in which the Fly was known to have been, following reconnaissance flights in 1993 that indicated where outcrops of Mesozoic age rocks occured in the region. It was hoped that, if the original fossil bearing rocks which yielded Agrosaurus could be found, more material may be waiting to be discovered. Also, it would give palaeontologists a means of determining the exact age of the Agrosaurus material. Prosauropods were only common during the Late Triassic and Early Jurassic periods, so an approximate age could be guessed, but not with much precision.

After two days the rocks of Captain Billy Landing proved devoid of fossils, at least that could be seen from the surface. The next two days were spent searching the rocks further northward, again with no results. It was decided at noon on the fourth day to head back to Captain Billy Landing for a second, more intensive survey, during which time two scraps of what could have been fossil bone were discovered. The fifth day was spent breaking rocks in the area where the possible fossils scraps were found, however no further material was discovered.

Later analysis seemed to suggest that the two scraps were indeed fossil bone, however they did not compare at all with the Agrosaurus material. The fossil scraps were solid, their internal spaces filled with minerals, whereas the spongiosa of the Agrosaurus fossils were open. The matrix rock surrounding the Agrosaurus material had been broken down with acetic acid during preparation, indicating that it had originally been cemented together with carbonate minerals. The rock of the Helby Beds at Captain Billy Landing had been thoroughly leached of all of its carbonate. The evidence strongly suggested that Agrosaurus had been originally discovered somewhere other than in the Helby Beds of Cape York. No other rock outcrops of the right age are known anywhere along the coast of north eastern Australia.

In 1906 Von Huene described the rock matrix from which the Agrosaurus fossils had been prepared as "a grey breccia full of bone fragments", and being "extremely reminiscent of the bone breccia at Durdham Downs near Bristol". When the residue of the acid preparation technique to recover the bones was examined, jaw fragments of a sphenodontid reptile were recovered. They were identicle to a sphenodontian called Diphyodontosaurus avonis, a lizard-like creature that is commonly found in the Triassic rocks of the Bristol area in England, an area from which the prosauropod Thecodontosaurus antiquus has also been discovered. The similarity between Agrosaurus and Thecodontosaurus had long been acknowledged, to the degree that at least one researcher had re-assigned the Agrosaurus macgillivrayi material to Thecodontosaurus macgillivrayi.

It was decided to test what remained of the matrix adhering to the prosauropod bones to see whether it was more similar chemically to the Cape York rocks, or those of Durdham Downs in England. The abundances of seven trace elements were plotted for the Agrosaurus material, samples of Durdham Downs rocks, and the two fossil scraps found on Cape York. An analysis of the Uranium-Thorium and Thorium-Lead ratios were also conducted for the same four samples. In both tests the Agrosaurus material compared more closely with the Durdham Downs rocks from England than to those collected from Cape York. The evidence strongly suggested that Agrosaurus had been mislabeled somewhere along the way, and that the bones were actually attributable to Thecodontosaurus antiquus, having been found in England rather than in Australia. Suddenly it looked like Australia's oldest dinosaur was no longer actually from Australia.

So, how did the label end up on the wrong material? According to Seeley, the material had passed through several hands before finding its way to the Natural History Museum in London. Prior to 1879 it was in the hands of a Mr Samuel Long Waring Esq. of "The Oaks" in Norwood. Following his death that year, they passed into the hands of Edward Charlesworth, who then sold the material to the museum. Seeley eventually found the material, with the wrong label, twelve years later - in the mammal hall of the museum. The registry details for the material bought in July 1879 make mention of the name "Mr Macgillivray", so it would seem that the label was already associated with the fossils when the museum acquired them. Its history of ownership before 1879 is anyone's guess.

Australia's oldest dinosaur is now tied between the Queensland sauropod Rhoetosaurus brownei and the Western Australian theropod Ozraptor subutaii, both dated to the Bajocian in the Middle Jurassic (175-170 million years ago). Thankfully at least these two dinosaurs are definitely known to have come from Australia, their discoveries having been well documented.


Long, J.A. 1998 Dinosaurs of Australia and New Zealand: and other animals of the Mesozoic. University of New South Wales Press, Sydney.

Vickers-Rich, P. and T.H.Rich 1999 Wildlife of Gondwana. Dinosaurs and Other Vertebrates from the Ancient Supercontinent. Second Edition. Indiana University Press.

Vickers-Rich, P., T.H.Rich, G.C.McNamara and A.Milner 1999 Agrosaurus: Australia's Oldest Dinosaur? Records of the Western Australian Museum Suppliment No.57: 191-200

Main Index