26 September 2020 | New Scientist | 43

F

OR most of history, we have had little

trouble defining species. There was a

general assumption that a finite number

of distinct forms of life had existed

unchanged since creation, each sitting in a

clearly defined pigeonhole: human, housefly,

hawthorn and so on. Within the past few

centuries, and particularly after Darwin,

evolutionary theory has emerged as a

more satisfactory way to explain how species

came into existence. Yet in doing so, it has

made species far harder to define.

There are several aspects to the problem.

One is that if we accept the idea of species

evolving from other species, then we must

allow that an ancestral species can gradually

morph into one or more descendants. We

would still like to place organisms in discrete

categories, but doing so is difficult if species

blur into one another through time. “As we

have come to terms with evolution, it has

highlighted a problem with the machinery in

our heads we use for classifying,” says Frank

Zachos at the Natural History Museum of

Vienna in Austria.

For Jody Hey at Temple University in

Philadelphia, the more important problem is

that biologists often have two objectives

in mind when they define species: one

is the traditional desire to divide nature

into easily recognisable packages; the

second is to explain, in evolutionary terms,

how those species came into existence.

“Humans have conflicting motivations

towards species,” he says.

Some researchers argue that these

two objectives can never be achieved

simultaneously. Down the decades,

biologists have come up with a few dozen

clever ways to define species. Some make

it easy to classify the organisms we

encounter – by their physical appearance,

for example – but tell us little about the

evolutionary process itself (see “Sadistic

cladistics”, page 49). Other definitions get to

the heart of how species come to exist, but

can be difficult to use in the real world.

Hybrid bonanza

In principle, advances in genetic sequencing

could have helped by indicating how

genetically distinct different groups of

organisms are and how long ago lineages

diverged. But sequencing has arguably

made the problem worse by revealing that

interbreeding – more technically,

introgression – between closely related

“species” is common across the tree of

life. “It does seem to be the rule, not the

exception,” says Michael Arnold at the

University of Georgia in Athens. Indeed,

evidence of introgression stretches right to

our front door: our ancestors interbred with

various ancient hominins that might, in the

eyes of some, count as distinct species.

Another problem is that looking at genes

rather than observable features makes it

easier to find new species, leading to what

some researchers have called taxonomic

anarchy. For instance, a biologist can argue

that a previously recognised species should

really be split into two or more “new” species,

as happened when genetic analysis of the

African elephant led to its being separated

into savannah and forest-dwelling species.

To help add more rigour to the business of

defining new species, earlier this year Zachos

and other biologists proposed establishing

the first single authoritative list of the world’s

species. “Species” itself will remain a slippery

concept, but at least we could all agree

on where to draw the lines. Colin Barras

SPECIES DON’T REALLY EXIST

Taxonomic anarchy

5

The African elephant is now

seen as two species not one

WI

M^ V

AN

DE

N^ H

EE

VE

R/N

AT

UR

EP

L.C

OM