Time horizon: Middle Jurassic, Bathonian epoch (~167 mya)
Length: 16m (~52 ft.), perhaps more depending on maturity.
Probable mass: 12 tons, perhaps more depending on maturity.
A downright bizarre sauropod relegated to obscurity since is discovery by René Lavocat in the 1950s, with roots reaching far back in deep time, this animal may possibly be the most primitive titanosauriform yet discovered. This creature is so basal that it's sometimes classed as a non-neosauropod, well outside of the titanosauriformes and macronaria altogether. However the flattened phalanges are clearly a brachiosaur-like feature, found in Giraffatitan and its close relatives, and very unlike the rounded cylindrical phalanges found in camarasaurs, diplodocids, mamenchisaurs and other more basal sauropods. And the interlocking metacarpals are arranged in a tight circular "horseshoe" pattern, very much like brachiosaurs and titanosaurs, and rather different from non-macronarians. The primitive overall design of the hand with its huge scythe-like thumb claw carried off the ground, while not typical for brachiosaurs, points to a possible common ancestry between early brachiosaurs like Atlasaurus and their obscure cousins, the Klamelisauridae, which did have a similar basal hand design.
However, the strangest feature of Lavocat's sauropod is the wrist. There are no less than five carpal bones, arranged on three different horizontal planes, something apparently not known in any other eusauropod. Not only is the wrist extremely complicated, it also may have had great flexibility while at the same time having capacity to support a great deal of mass. The carpals can interlock, but most of them do not appear to be fixed positions. It may be that this specimen was immature and that the five carpals were a juvenile throwback to far more ancient ancestors, and would have later fused to leave only the two carpals commonly seen in most sauropods.
As if this wasn't crazy enough, this animal was, like Lapparentosaurus, among the many Madagascar sauropod specimens wrongly thrown into "Bothriospondylus". And it is clearly far more primitive than Lapparentosaurus, though the two coexisted in the same time and place. Due to this fact, Lavocat's sauropod is not the common ancestor of brachiosaurs and klamelisaurs, but probably the most primitive descendant of that common ancestor to survive as late as the Bathonian epoch. It's still waiting for an official name and phylogenetic classification.
According to the redescription by Lang and Goussard (2007) the specimen is "a nearly complete skeleton... that included a complete forelimb in articulation including the carpal bones". However only the hand, the carpals, and the radius were published in a photograph. How complete the specimen truly is may be debatable, so only the forelimb can be more or less accurately restored. The rest of the skeleton, in dark gray, is based on a mix of features from various basal brachiosaurs and their cousin Klamelisaurus. Due to the very basal features of the species and the scarcity of research done on it, Lavocat's sauropod will likely continue to be a very controversial dinosaur, both taxonomically and anatomically, for a long time.
Läng É. & Goussard F. 2007. — Redescription of the wrist and manus of ?Bothriospondylus madagascariensis: new data on carpus morphology in Sauropoda. Geodiversitas 29 (4) :549-560.
Lavocat R. 1955c. — Sur un membre antérieur du dinosaurien sauropode Bothriospondylus Owen, recueilli à Madagascar. Comptes Rendus de l’Académie des Sciences de Paris 240: 1795-1796.