BioEd Online

There are 8.7 million species on our planet — give or take 1.3 million. The latest biodiversity estimate, based on a new method of prediction, dramatically narrows the range of 'best guesses', which was previously between 3 million and 100 million. It means that a staggering 86% of land species and 91% of marine species remain undiscovered.

Camilo Mora, a marine ecologist at the University of Hawaii at Manoa, and his colleagues at Dalhousie University in Halifax, Canada, have identified a consistent scaling pattern among the different levels of the taxonomic classification system (order, genus, species and so on) that allows the total number of species to be predicted. The research is published in PLoS Biology¹ today.

Mora argues that knowing how many species there are on Earth is one of the most important questions in science. "Finding this number satisfies a basic scientific curiosity," he says.

Bob May, a zoologist at the University of Oxford, UK, who wrote a commentary on the work², agrees. "Knowing how many plants and animals there are on the planet is absolutely fundamental," he says. He also highlights the practical significance: "Without this knowledge, we cannot even begin to answer questions such as how much diversity we can lose while still maintaining the ecosystem services that humanity depends upon."

But the unstinting efforts of field taxonomists are not going to provide the number any time soon. In the more than 250 years since Swedish biologist Carl Linnaeus began the science of taxonomy, 1.2 million species have been identified and classified — less than 15% of Mora's new total. At this pace, May estimates that it will take another 480 years to complete the job of recording all species.

The catalogue of life

Instead, scientists have tried to predict the total number of species from the number already known. Some of the estimates amount to little more than educated guesses. "These predictions are unverifiable and experts change their mind," says Mora. Other approaches use assumptions that he describes as "unreliable and easy to break".

Mora's method is based on an analysis of the taxonomic classification for all 1.2 million species currently catalogued. Linnaeus's system forms a pyramid-like hierarchy — the lower the category, the more entities it contains. There are more species than genera, more genera than families, more families than orders and so on, right up to the top level, domain.

Mora and his colleagues show that a consistent numerical trend links the numbers in each category, and that this can be used to predict how many entities there should be in poorly catalogued levels, such as species, from the numbers in higher levels that are much more comprehensively described.

"The unique thing about this approach is that we are able to validate it," he says. "By testing the predictions against well catalogued groups such as mammals, birds, reptiles and amphibians, we were able to show that we could predict the correct number of species."

The analysis also reveals that some groups are much better known than others. For example, some 72% of the predicted 298,000 plant species on land have already been documented, in comparison with only 12% of predicted land animal species and 7% of predicted land fungi species.

May is impressed. "I like this approach. Not only is it imaginative and novel, but the number it comes up with is within the range of my own best estimate!"