Correction: An earlier version of this article misstated the number of acres in a hectare. A hectare covers 2.47 acres, not 100 acres. This version has been corrected.

For centuries scientists have pondered a central question: How many species exist on Earth? Now, a group of researchers has offered an answer: 8.7 million.

Although the number is still an estimate, it represents the most rigorous mathematical analysis yet of what we know — and don’t know — about life on land and in the sea. The authors of the paper, published Tuesday evening by the scientific journal PLoS Biology, suggest that 86 percent of all terrestrial species and 91 percent of all marine species have yet to be discovered, described and catalogued.

The new analysis is significant not only because it gives more detail on a fundamental scientific mystery but because it helps capture the complexity of a natural system that is in danger of losing species at an unprecedented rate.

Marine biologist Boris Worm of Canada’s Dalhousie University, one of the paper’s co-authors, compared the planet to a machine with 8.7 million parts, all of which perform a valuable function.

“If you think of the planet as a life-support system for our species, you want to look at how complex that life-support system is,” Worm said. “We’re tinkering with that machine because we’re throwing out parts all the time.”

He noted that the International Union for Conservation of Nature produces the most sophisticated assessment of species on Earth, a third of which it estimates are in danger of extinction, but its survey monitors less than 1 percent of the world’s species.

For more than 250 years, scientists have classified species according to a system established by Swedish scientist Carl Linnaeus, which orders forms of life in a pyramid of groupings that move from very broad — the animal kingdom, for example — to specific species, such as the monarch butterfly.

Until now, estimates of the world’s species ranged from 3 million to 100 million. Five academics from Dalhousie University refined the number by compiling taxonomic data for roughly 1.2 million known species and identifying numerical patterns. They saw that within the best-known groups, such as mammals, there was a predictable ratio of species to broader categories. They applied these numerical patterns to all five major kingdoms of life, which exclude microorganisms and virus types.

The researchers predicted there are about 7.77 million species of animals, 298,000 of plants, 611,000 of fungi, 36,400 of protozoa and 27,500 of chromists (which include various algae and water molds). Only a fraction of these species have been identified, including just 7 percent of fungi and 12 percent of animals, compared with 72 percent of plants.

“The numbers are astounding,” said Jesse Ausubel, who is vice president of the Alfred P. Sloan Foundation and co-founder of the Census of Marine Life and the Encyclopedia of Life. “There are 2.2 million ways of making a living in the ocean. There are half a million ways to be a mushroom. That’s amazing to me.”

Angelika Brandt, a professor at the University of Hamburg’s Zoological Museum who discovered multiple species in Antarctica, called the paper “very significant,” adding that “they really try to find the gaps” in current scientific knowledge.

Brandt, who has uncovered crustaceans and other creatures buried in the sea floor during three expeditions to Antarctica, said the study’s estimate that 91 percent of marine species are still elusive matched her own experience of discovery. “That is exactly what we found in the Southern Ocean deep sea,” Brandt said. “The Southern Ocean deep sea is almost untouched, biologically.”

Researchers are still pushing to launch a series of ambitious expeditions to catalogue marine life over the next decade, including a group of Chilean scientists who hope to investigate the eastern Pacific and a separate group of Indonesian researchers who would probe their region’s waters.

One of the reasons so many species have yet to be catalogued is that describing and cataloguing them in the scientific literature is a painstaking process, and the number of professional taxonomists is dwindling.

Smithsonian Institution curator Terry Erwin, a research entomologist, said fewer financial resources and a shift toward genetic analysis has cut the number of professional taxonomists at work. Erwin noted that when he started at the Smithsonian in 1970 there were 12 research entomologists, and now there are six.

“Unfortunately, taxonomy is not what cutting-edge scientists feel is important,” Erwin said.

In a companion essay in PLoS Biology, Oxford University zoologist Robert M. May wrote that identifying species is more than a “stamp collecting” pastime, to which a Victorian physicist once compared it. He noted that crossing conventional rice with a new variety of wild rice in the 1970s made rice farming 30 percent more efficient.

“To the contrary, we increasingly recognise that such knowledge is important for full understanding of the ecological and evolutionary processes which created, and which are struggling to maintain, the diverse biological riches we are heir to,” he wrote. “It is a remarkable testament to humanity’s narcissism that we know the number of books in the U.S. Library of Congress on 1 February 2011 was 22,194,656, but cannot tell you — to within an order of magnitude — how many different species of plants and animals we share our world with.”

Erwin said researchers would continue to search for the best way to quantify global diversity beyond the new method. Erwin himself has been using a biodegradable insecticide since 1972 to fog trees in the Amazon and kill massive amounts of insects, which he and his colleagues have classified. Based on such sampling, Erwin posited in 1982 that there were roughly 30 million species of terrestrial arthropods — insects and their relatives — worldwide.

Extrapolating from that sample to determine a global total, he said, was a “mistake, one which others have repeated,” he said.

Erwin added he still thinks counting actual specimens is the best route, noting he and his students determined in 2005 that there are more than 100,000 species of insects in a single hectare (or 2.5 acres) of the Amazon. Noting that insects and their relatives count for 85 percent of life on Earth, he wondered why there’s such a fuss about counting the rest of the planet’s inhabitants: “Nothing else counts.”