The number of living species is estimated to be between 5 million and 15 million, while the number of endangered species increases daily, writes Caroline Davis.

But we could be worrying about preserving species that do not exist, according to research from the University of California.

John Alroy from the National Center for Ecological Analysis and Synthesis analysed how many named species were duplicates or did not clearly refer to a single biological entity. He estimated that the current species count could be overinflated by almost half.

Professor Alroy modelled the process, by which newly named species are re-evaluated by taxonomists using historical data from a North American fossil mammals systematics database. Since mammals, especially the 4,861 fossil species on the database, have been studied disproportionately, it provides the next best thing to a complete taxonomic dataset. The database contains genus and species-level taxonomy and records all separate opinions on the status of each name.

ÖйúAƬ

Professor Alroy found that at any one time, up to 31 per cent of the names were invalid, leading him to conclude that diversity estimates were inflated by up to 44 per cent. And because mammals tended to receive more attention than other categories, diversity estimates for lesser studied species such as insects and fungi may be more inflated.

The problem is that the process of disqualifying invalid names is so slow that many invalid names remain to be flagged. "Validity is a matter of opinion, not proof," Professor Alroy explained. "Hence there is a steady flux of names back and forth between valid and invalid. What the method seeks to establish is the proportion of names that would be held invalid given an infinite amount of debate."

ÖйúAƬ

Surprisingly, the study also revealed that new names are no more robust than those coined more than 100 years ago. The research appeared in the Proceedings of the National Academy of Sciences.