Megalosaurus. At least that's what we've been told... Credit: Dorling Kindersley RF

One hundred and thirty-five years of questionable judgments, some driven by a lust for headlines, have left dinosaur nomenclature in disarray, according to two new studies.

The studies find that of 1,401 names given to dinosaurs species from 1824 to 2004, about 16 per cent of names were duplicates, and 32 per cent embodied errors of some other sort.

"It is a bit scary," says Michael Benton, a palaeontologist at the University of Bristol, England, who is author of the studies. A first discussion of the results is published today in the Royal Society's Biology Letters1, and a longer analysis will be published in December in Paleobiology2.

The high error rate is not just a problem for fossil hunters; it is a warning that scientists should take extra precautions when identifying new species as they assess modern biodiversity, too, says Benton. "We need to be sure species lists are properly vetted, so incorrect information isn't used in formulating policies."

Quest for glory

The fact that there are errors is not surprising. As Mark Goodwin, a palaeontologist at the University of California at Berkeley, says, "If you are in the trenches, dinosaur taxonomy and systematics will always be a work in progress." But the extent of the problems is a shock. "We knew there was an issue, but no one did the work of seeing how bad it was," says Peter Makovicky, a palaeontologist at the Field Museum in Chicago.

As more public money came to be used for exploration projects, Benton notes there was a growing risk that funding-agency and journal pressures might lead to unnecessary naming of genera or species. Concerned these factors "may be distorting" the field, Benton began making a catalogue about four years ago.

But despite these pressures, Benton noted that the scientific process has addressed the publication shortcomings through literature corrections, which he called "heartening."

These corrections eliminated duplications — called synonyms, or aliases — and other errors, such as a lack of sufficient fossil material, an undiagnostic description, or the use of a name already assigned to something else.

The hall of fame

For his studies, Benton completed a list of the top 30 dinosaur namers of all time, of whom half are still alive. For the top five, the percentage of names that have held up varies from 14 per cent to 64 per cent.

We knew there was an issue, but no one did the work of seeing how bad it was. Peter Makovicky , The Field Museum

At the top of the list is Othniel Marsh, who in the nineteenth century vied with Edward Drinker Cope to be America's king of the dinosaurs. Marsh named 80 dinosaurs from 1870 to 1899; 23 of those names are still valid, says Benton, giving a success rate of 29 per cent.

At number four on the list is Dong Zhiming, a Chinese scientist now semi-retired to Yunnan. He named 42 dinosaurs from 1973–2004, with 27 of those still valid (64 per cent).

The most prolific dinosaur namer still active is Xu Xing, who like Dong is associated with the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing of the Chinese Academy of Sciences. Xu sports a perfect record: 24 species named from 1999–2006, with all still valid. His rate of three namings per year is unmatched.

Much of Dong's early work was done when there was pressure on Chinese scientists to discover new species, and when international exchange and access to high-quality, peer-reviewed journals was limited. For all that, his 27 properly named species is the highest total ever achieved.

In contrast, Xu studied and collaborated at some of the finest US institutions, benefited from China's recent major research funding increases, and had ready access to top journals.