Not so terrible

Jun 24th 2009
From The Economist print edition

Prehistoric monsters may not have been as monstrous as once believed

DINOSAURS—terrible lizards, to translate their name into English—were big. Of course, there were many small ones, too. But what caught the public imagination in the 19th century, when the first of them were dug up, and has not let go of it since, is the idea that some dinosaurs were unfeasibly huge; the largest animals, by a long chalk, ever to walk the face of the Earth.

Gary Packard of Colorado State University and his colleagues do not deny that creatures like Apatosaurus, Brachiosaurus and Diplodocus were big—and bigger than any terrestrial mammal. But they do question just how big. In a study published this week in the Journal of Zoology, they examine the method traditionally used to estimate the weight of extinct animals from their fossilised bones and find it wanting. Their revised version has slimmed some of the giants by 50%.

A specimen of Apatosaurus (better known to many laymen as Brontosaurus) that had been examined using the previous method is now thought to have weighed 18 tonnes rather than 38 tonnes—the equivalent of only about three fully grown African elephants, instead of seven. A Brachiosaurus slimmed down from 32 tonnes to 16. Even a slender Diplodocus, once believed to have tipped the scales at five and a half tonnes when alive, has lost a tonne and a half from its weight.

Working out an animal’s mass from the size of its bones is not, of course, an exact science. But it is possible to use the weights and bone sizes of existing species to make mathematical models of the relationship. The problem comes when these models are used to extrapolate beyond the range of reliable data (ie, of living animals whose weight is known). Since the largest dinosaurs have skeletons a lot larger than any living terrestrial species, such extrapolation is fraught with difficulty.

Dr Packard and his colleagues argue that the most widely used model has got the extrapolation wrong. It uses what are known as logarithmic transformations to cope with changes in the relationship between bone size and weight. This works well enough for smaller animals, but it predicts that an elephant should weigh one and a half times as much as it actually does, and that deviation is likely to be much larger for a Brachiosaurus, Apatosaurus or Diplodocus.

Dr Packard’s model gets round this by abandoning simple logarithmic transformation in favour of a more complicated, so-called nonlinear model. This fits the data for existing large species better and so, he concludes, may provide a better estimate of the weights of large extinct ones.

Of course, no one can know for sure. Despite the confident depiction of them in films such as “Jurassic Park”, the reconstruction of dinosaurs is subject to constant re-interpretation. But even at a mere 18 tonnes, Apatosaurus would still have been an awesome sight.

http://www.economist.com/sciencetechnology/displayStory.cfm?story_id=13240208