THE clash of the publishing titans continues. Encyclopaedia Britannica has hit back at a study by Nature published in December 2005 claiming that its accuracy is only slightly better than the free online resource Wikipedia.

According to the survey, in 42 matching pairs of science entries on topics from Agent Orange to the West Nile virus, Wikipedia made 162 errors, which "comes close" to its competitor's 123.

Britannica angrily disputed the results on its website, and has run half-page adverts in major newspapers in the US and the UK this week. "We rebutted the Nature study because it was done badly," says Tom Panelas, head of corporate communications at Britannica.

Nature stands by its findings. "We reject those accusations," says Jim Giles, co-author of the report, "and are confident our comparison was fair."

One of the biggest threats to printed encyclopedias is the availability of information online. With over a million articles, compared with Britannica's 65,000 in its print version and 120,000 online, Wikipedia eclipses its competitor in terms of sheer number of entries. It has, for example, an entry on Encyclopaedia Britannica. The gesture has not been reciprocated.