What is a species? How biologists draw the lines

A species is a group of living things that can breed with each other in nature and produce offspring that can breed in turn, and that do not normally breed with other groups. That is the biological species concept, the definition most textbooks start from. It works well for most animals, but not for bacteria, fossils or many plants, so biologists use several definitions side by side, and where one species ends and another begins is sometimes a judgment call.

In short

• The usual test of a species is reproductive: members interbreed and have fertile young, and don't interbreed with other groups.
• That test can't be applied to organisms that reproduce without sex, or to fossils, so other concepts based on shape, DNA and ancestry fill the gaps.
• About 2.5 million species have been named, but estimates of the total run from about 9 million to far more once microbes are counted.

What is the biological species concept?

The definition was set out by the biologist Ernst Mayr in 1942: species are groups of actually or potentially interbreeding natural populations, reproductively isolated from other such groups. OpenStax's biology textbook puts it as a group of organisms that interbreed and produce fertile, viable offspring. The University of California's Understanding Evolution project gives a useful way to picture it: a species is the biggest gene pool possible under natural conditions.

The power of the definition is that it says why species matter. Genes flow freely within a species and not between species, so each one evolves on its own track. Dogs look wildly different, from a chihuahua to a Great Dane, but they can all interbreed, so they are one species. Two birds can look nearly identical and still be separate species if they never interbreed, perhaps because their songs differ and neither responds to the other's.

The barriers that keep species apart come in two kinds. Some stop mating from happening at all: different breeding seasons, courtship, habitats or body parts that don't fit. Others act after mating: the hybrid embryo fails, or the hybrid lives but is sterile. Over time, as evolution pulls two separated populations apart, these barriers accumulate until the two can no longer merge again.

Where does the definition break down?

The hybrid cases show both its strength and its limits.

Mules. A mule is the offspring of a male donkey and a female horse. Horses have 64 chromosomes and donkeys 62, so a mule has 63, 32 from its mother and 31 from its father. With an odd number, the chromosomes can't pair up properly to make eggs or sperm, so mules are almost always sterile. Horse and donkey stay separate species: they can mate, but genes can't flow onward through the mule.

Ligers. A liger is the offspring of a male lion and a female tiger. Male ligers are sterile, but female ligers sometimes are not. Lions and tigers are still treated as separate species, because their ranges barely overlap today and they don't interbreed in the wild. The word "naturally" in the definition does real work here.

Bacteria. Most bacteria reproduce by splitting in two, without sex, and swap genes in ways that don't follow family lines. The interbreeding test can't be applied at all, so microbiologists compare DNA instead. A common rule of thumb compares whole genomes: two bacteria whose genomes are about 95% identical or more are usually treated as one species.

Fossils. Nobody can test whether two fossils could have interbred. Paleontologists group fossils by shape and, where a lineage changes gradually through rock layers, divide it into "chronospecies" at points that are partly a matter of convention.

Ring species. Around some barriers, such as a mountain range or a desert, a chain of populations circles the obstacle. Each population can interbreed with its neighbor, but the two ends of the chain, where they meet, cannot. Is it one species or two? The question has no clean answer, which is exactly what you would expect if species form gradually.

What other species concepts do biologists use?

Because no one definition fits every case, biologists use several. There are a lot of them: more than two dozen have been proposed and are still in use somewhere. The main ones are below.

Concept	A species is…	Works well for	Weak point
Biological	A group that interbreeds in nature and is cut off from others	Most animals and many plants	Can't be tested on asexual organisms or fossils
Morphological	A group that looks distinct in body form	Fossils, museum specimens, field guides	Look-alike species and very varied ones mislead it
Phylogenetic	The smallest group sharing a common ancestor and set apart by inherited traits	Any group with DNA or fossils	Tends to split groups into many more species
Genetic distance	A group whose genomes are similar above a set threshold	Bacteria and other microbes	The threshold is a convention
Ecological	A group that fills one niche, the role it plays in its habitat	Plants and microbes that hybridize	Niches are hard to measure exactly

In practice, these mostly agree, because a population that has been cut off long enough tends to look different, carry different genes and stop interbreeding all at once. They disagree at the edges, where a population is in the middle of splitting. That disagreement isn't a failure of biology. It is evidence that species are snapshots of an ongoing process.

How are species named and classified?

Every named species has a two-part scientific name, a system made standard by the Swedish naturalist Carl Linnaeus in the eighteenth century. The first part is the genus, a group of close relatives, written with a capital letter; the second picks out the species within it. Both are written in italics. Humans are Homo sapiens; the gray wolf is Canis lupus, which Linnaeus himself named. Names in Latin form mean that a biologist in Brazil and one in Japan know they are talking about the same animal, whatever it is called locally.

Species are then grouped into ever-larger ranks. Here is the gray wolf, from the broadest group down:

Rank	Gray wolf	What it groups together
Domain	Eukarya	All organisms whose cells have a nucleus
Kingdom	Animalia	Animals
Phylum	Chordata	Animals with a nerve cord along the back
Class	Mammalia	Mammals: hair, and milk for the young
Order	Carnivora	Cats, dogs, bears, seals and their relatives
Family	Canidae	Dogs, wolves, foxes, jackals
Genus	Canis	Wolves, coyotes, jackals and domestic dogs
Species	Canis lupus	Gray wolves and their domesticated form

The domestic dog is often treated as a subspecies of the gray wolf, Canis lupus familiaris, because dogs and wolves can interbreed and have fertile pups, though some catalogs list it as its own species, Canis familiaris. A subspecies is a population distinct enough to name but not cut off from the rest of its species.

How many species are there?

Nobody knows, and the honest answer is a range. The Catalogue of Life, a global list kept by taxonomists, counted about 2.46 million species in its extended release of 16 October 2026: about 1.8 million animals, 408,000 plants, 175,000 fungi and 76,000 in other kingdoms. That is the number named, not the number alive, and new ones are described every year.

For the total, the most cited estimate is a 2011 study in PLOS Biology by Camilo Mora and colleagues, which predicted about 8.7 million species of eukaryotes, give or take 1.3 million, of which about 2.2 million live in the ocean. On that estimate, about 86% of the species on Earth and 91% of those in the ocean have not yet been described.

Still argued

Microbes change the picture completely. Mora's 8.7 million counts only eukaryotes, organisms with a nucleus. Bacteria and archaea can't be counted the same way, and estimates for them depend on how a microbial species is defined. A 2016 study by Kenneth Locey and Jay Lennon, using scaling laws from about 35,000 sampling sites, put Earth's microbial species at upward of 1 trillion. A 2019 census by Stilianos Louca and colleagues, built from more than 1.7 billion DNA sequences from 492 studies, came to about 2.2 to 4.3 million kinds of bacteria and archaea, counted as clusters of near-identical marker genes, and said their data refuted the trillion-scale predictions. Much of the gap comes down to how finely DNA differences are split into species. The range of serious estimates spans several orders of magnitude.

The count matters for more than curiosity. Species are the units that conservation laws protect, and extinction rates, from the asteroid that killed the dinosaurs to today, are measured in species lost. And each species plays a part in its ecosystem, so knowing what lives in one is the first step to knowing how it works. Telling living species apart is also part of a wider question, what life is at all.

Check yourself

• Without looking back, explain why a horse and a donkey count as separate species even though they can mate.
• Why can't the biological species concept be used for bacteria or fossils, and what do biologists use instead?
• Why do species concepts disagree most about populations that are in the middle of splitting?

Sources

• 18.2 Formation of New Species — OpenStax, Biology 2e, 2018
• Defining a species — Understanding Evolution, University of California Museum of Paleontology
• How Many Species Are There on Earth and in the Ocean? — Mora and colleagues, PLOS Biology, 2011
• Catalogue of Life — Catalogue of Life, release COL26.10, 2026
• Scaling laws predict global microbial diversity — Locey and Lennon, PNAS, 2016
• A census-based estimate of Earth's bacterial and archaeal diversity — Louca and colleagues, PLOS Biology, 2019