Scientists group living organisms into categories known as species. There is no universally agreed upon definition of species but a popular idea is that species are groups of organisms that can successfully interbreed. So because a Human can’t get a Chimpanzee pregnant, we can clearly state they’re different species.

This concept becomes problematic when we talk about ring species. That’s when one animal can’t interbreed with another, but a third animal can interbreed with both. What species does the third animal belong to?

But ring species are rare, so most of the time, reproductive isolation works well for neatly classifying animals into nice discrete categories with strong genetic walls between them.

The problem begins when we want to speak of species across time. For example, wikipedia claims Homo Erectus lived from 2 to 0.1 million years ago (mya) before evolving into our species about 0.35 mya. In order for these numbers to fit the reproductive definition of species, we’d have to believe that all Erectus from 2 to 0.1 mya could interbreed with each other and yet none of them could interbreed with even the first Homo Sapiens.

But if evolution is gradual, we’d expect the last Erectus to be more similar (and thus more reproductively compatible) with the earliest Sapiens than with the earliest Erectus and thus the boundary between the two is wholly arbitrary.

Of course even if we reject reproductive compatibility as our criterion and instead go with morphological similarity, the same problem ensues. One species blends into another.

Richard Dawkins makes the point:

But what if evolution isn’t gradual? According to Punctuated Equilibrium theory by Niles Eldridge and Stephen Jay Gould, the bulk of evolution occurs in relatively rapid jumps, followed by long periods of stasis.

As we can see from the above pictures, if evolution is gradual, where we draw the line between species 1, 2 and 3 is subjective because we’re artificially imposing discrete categories on a continuous process, but if evolution is punctuated, then the new categories occur naturally, whenever evolution plateaus.

Of course not all plateaus signify a new species. Some might merely indicate a new race or ethnic group. Others might indicate an entirely new kingdom of life (i.e. plants vs animals). It should all depend how big the morphological jump is before the plateau. For example we might say a 10% jump in overall phenotype indicates a new race while a 40% jump indicates a new species etc. A population might have to become a new race four times before it qualifies as a new species.

You might be asking what does it matter when one species stops and another one starts. It matters a lot when deciding what species has survived the longest, conquered the most environments, or has the greatest biomass. None of those accomplishments are meaningful if the very identification of species is arbitrary.

So how old is our own species? From about 1997 to 2017, the consensus was that our species was about 0.2 mya since that was the approximate age of both the Mitochondrial Eve and the oldest anatomically modern human skulls (found in Ethiopia). A more recent find in Morocco threatens to push the date back to over 0.3 mya, however Richard Klein argues that it’s a proto-modern human, not a modern one. So perhaps the Moroccan skull does not belong to any species. It might be from a period of rapid evolution predating evolutionary stasis.