How could you distinguish between a species in which there is a lot of variation and two separate species?

There are no clear-cut answers. The big problem is there is no universally accepted definition of “species.” The most restrictive is Ernst Mayr’s “biological” definition that hinges on fertility. At the other end of the spectrum is the identification of the Eastern Meadowlark (Sturnella magna) and the Western Meadowlark (Sturnella neglecta) as separate species because they are physically identical and are perfectly capable of interbreeding – and the offspring of one species become the other species if raised by the other parents – they do not interbreed because they sing different songs. Yet Orcinus orca is a single species even though three separate populations with very different dentition share territory in the Pacific and have not interbred in thousands of generations.

Members of a single diverse species can successfully interbreed and produce fertile offspring. Members of two separate species can’t. If you can’t get them to reproduce, you trust the scientists.

There’s still some controversy over the subject. But the most accepted definition for speciation is the inability to interbreed.

It’s a thin line.

In Origin of Species, Darwin admitted that he himself didn’t know.