Technical note A case of vertebrate fossil forgery from Madagascar Bernhard Zipfel * , Celeste Yates & Adam M. Yates Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, WITS, 2050 Johannesburg, South Africa Received 19 July 2010. Accepted 3 December 2010 INTRODUCTION The high value of rare fossil specimens results in some fossil dealers and collectors purchasing these specimens from dubious and even illegal sources. Complete verte- brate fossils in particular are rare, and are therefore more easily sold at relatively high prices. As a result there is a demand for the production of fake fossils particularly in developing countries where the trade in fossils represents a means to economic survival (Mateus et al. 2008). Most dealers, however, have little or no scientific knowledge on the fossils they purchase and may therefore inadvertently purchase fake fossils. Both China and Morocco, for example, are known to produce both genuine and fake fossils (Dalton 2000, 2004a, b; Milner et al. 2001; Padin 2000). As a result, China, among many other countries, including South Africa, has instituted very strict legislation regarding the trade and export of fossils. Fraudulent fossils do not just affect dealers and collectors, but have also embarrass- ingly deceived scientists. The best known of these is the famous ‘Piltdown Man’ from England, a forgery merging the cranium of a modern human and the mandible of an orangutan. This forgery was put forward as an early human ancestor that confused the scientific community for decades (Weiner 1955), whilst the first genuine early hominin, the now famous ‘Taung Skull’, holotype of Australopithecus africanus from South Africa (Dart 1925, 1929; Dart & Craig 1959; Hrdlicka 1925), was forced to take a ‘back seat’ for many years. More recently, the famous Archaeoraptor specimen from the Lianoing Province of China received coverage by a number of publications including National Geographic and Nature (Sloan 1999; Rowe et al. 2001) but in reality the specimen represents at least two and perhaps up to five separate individuals of two or more different species fraudulently merged into a single specimen (Zhou et al. 2002). Invertebrates from Morocco, for example trilobites have been skillfully carved out of rock and sold as genuine fossils. As most trilobites from Morocco are genuine, and the trade in these invertebrates from that country is legal, it is understandable that a fossil trader may be deceived. In other cases, a genuine fossil of a common living species, such as the tooth of a great white shark (Carcharodon carcharias), may be sold as a representative of a similar, but extinct species such as a megalodon ( Carcharocles megalodon). Fossil frauds are therefore committed not only for profit, but also for publicity (Mateus et al. 2008). Mateus et al. (2008) suggest a number of methods of fraud recognition and describe three kinds of hoaxes: 1) Those that contain no original fossil material, such as shapes carved in rock; 2) Those that contain original fossil material, but are en- tirely or partially altered in order to give the appear- ance of a more complete specimen, for example, a skull carved from a limb bone. 3) Those that are true fossils but a combination of multi- ple individuals, mostly from the same species. Here we report on a specimen that was brought to the Bernard Price Institute for Palaeontological Research, University of the Witwatersrand by a fossil dealer for identification. The specimen had been obtained illegally in Madagascar by the fossil dealer, reportedly from the vicinity of known dinosaur localities in the Cretaceous Maevarano Formation of the Mahajanga Basin (Depéret 1896; Besairie 1936, 1972), with the intention of having it prepared in South Africa. The fauna of this formation is well known and is the subject of ongoing research programmes (e.g. Forster et al. 1998; Krause & Hartman 1996; Sampson et al. 2001; Krause et al. 1999; Buckley et al. 2000; Curry-Rogers & Forster 2001; Rogers 2005; Fanti & Therrien 2007). DESCRIPTION The specimen initially appeared to consist of two articu- lated vertebral centra in a nodule of matrix. It measured approximately 23 cm by 15 cm. A superficial resemblance to a small skull could be seen. After removal of the ‘matrix’ it became apparent that the specimen was a forgery using genuine pieces of dinosaur bone that attempted the construction of a vertebrate skull (Fig. 1). The ‘matrix’ proved to be crushed rock, probably from the Maevarano Formation mixed with commercial cement covering inner layers of resin and plaster of Paris (Fig. 2). The construc- tion of the skull consisted of an ‘orbit’, a pig-like ‘snout’, crude ‘teeth’ a lower ‘jaw’ and a ‘post-orbital’ region. The snout and post-orbital region were constructed from genuine fossils of dinosaur vertebral centra (Fig. 2). These are most probably from a titanosaur sauropod; however, this is purely speculative as there is not enough material to be diagnostic. The teeth were constructed with a series of roughly shaped stones cemented to the upper and lower ‘jaws’ (Fig. 2). DISCUSSION Clearly, this forgery was a poor attempt at creating what seemed to have been intended to resemble a vertebrate skull. As the ‘skull’ was bound by matrix, it appears that the intent of the forgery was probably not to convince anyone that this was a genuine fossil assemblage, but rather to give a vague impression that the matrix block was indeed fossiliferous and superficially gave the impression that it potentially contained what seemed to be a vertebrate skull. This resulted in the sale of the specimen before the complete contents of the matrix could be deter- mined by the purchaser. According to the classification of fossil forgeries by Mateus et al. (2008), this forgery would ISSN 0078-8554 Palaeont. afr. (December 2010) 45: 29–31 29 *Author for correspondence. E-mail: bernhard.zipfel@wits.ac.za