A friend of mine, a secondary bio teacher, sent me through a couple of questions from students that she thought might be useful for a blog piece. And I agreed, so here they are:
How do scientists know when a species finishes and a new one is formed?
Are there different species in between the formation of a new species or does a, for example, Homo habilis have a Homo erectus child? The second question is actually quite similar to the first.
They’re good questions that show the student is giving thought to what it is that defines/delineates a species, & how we recognise a new species. The answer depends in part on what sort of speciation event you’re considering.
In some cases it’s pretty clear when you’ve got a new species. Polyploidy in plants, for example, can result in what’s effectively an instantaneous speciation event. And the new organism can be morphologically distinct from its parent species, making it relatively easy for biologists to recognise what’s happened. The first scientist to do so was Hugo de Vries, working with evening primrose (Oenothera lamarckiana): he noticed that some of his plants had much larger leaves & flowers than their neighbours, & that these differences were heritable. Subsequently, staining & microscopic examination of the plant cells shows that de Vries new ‘gigas‘ mutant had twice as many chromosomes as the ‘wild-type’ plants – it was the first description of polyploidy. And breeding experiments showed that the gigas individuals couldn’t breed with the original lamarckiana plants: a new species had formed.
But in other situations, it’s not quite so easy. We can usually tell when we’re dealing with a species: it’s a group of organisms which are capable of interbreeding to produce viable offspring. (Well, that’s the biological species definition. Won’t work for fossils, of course, so in this case we talk about the morphological species concept – structure is pretty much all we’ve got to go by. There are other concepts of ‘species’ – you can read about them here.) But decisions do tend to be retrospective, because we look backwards in time, comparing what we have now with what existed a while ago. This is particularly the case if you’re looking at anagenesis, where one species is replaced by a daughter species over some period of time. Cladogenesis, where one lineage branches into two, can perhaps be recognised more quickly. In fact, the beginnings of this have been observed in laboratory experiments using geneticists’ favourite lab animal, the fruit fly Drosophila melanogaster.
Anyway, in the example in that original question – no, you wouldn’t have a habilis mother giving birth to an erectus child. What you would have is an original population of habilis, somewhat isolated from other such populations. Over time, that population’s gene pool would gradually accumulate a different set of mutations from those in its parent population. Eventually members of that population could end up looking recognisably different from members of the original population, to the extent that a biologist looking at them would call them different species. (Voila! Homo erectus!) In living organisms we’d be looking for evidence of reproductive isolation.
And remember – populations evolve, not individuals.