The ecological species concept claims that species are groups or populations that share the exact same ecological niche. The definition gets a little tricky with organisms that change niches over their development. For examples, some organisms live in the water as larvae, and on land as adults. It's sometimes expanded to include a set of niches.
The phenetic species concept defines species as groups of organisms that are similar in some defining trait(s) chosen by taxonomists, like eyebrow density, tastiness, or affinity for staring off into the distance. They'd probably pick something more scientific, but we'd be all for running a phenetic species taste test.
The phylogenetic species concept is similar, but with a slight twist: it states that a species is the smallest set of lineages or populations that can be recognized by a unique combination of different traits.
Is it just us, or do none of these concepts seem right? In a world of imperfect species concepts, allow us to introduce the victor—the biological species concept, or BSC. Is it foolproof? No. Does it allow for horse-dogs? No. Do we love it anyway? You'd better believe it.
Championed by Ernst Mayr, the BSC puts the emphasis where it belongs: on reproduction. In 1963, Mayr defined a species as follows, "species are groups of interbreeding natural populations that are reproductively isolated from other such groups."
The fact the a species must be some 'natural' designation means that, regardless of what sort of transgenic weirdness scientists can conjure up in a lab, we cannot claim to make new species. Even when we're creating organisms that share genes with distantly-related organisms (ahem, Fish Tomato anyone?).
Species are limited to reproduction in natural populations, not lab mice. The 'reproductively isolated' bit is taken to mean the following: if there is a chance that genes from one population could end up in the offspring of another population, then they belong to the same reproductive group. The two populations are reproductively isolated if there is no way they can share genes without alien intervention. We added in that last part, but it seems important somehow.
The BSC is useful because it emphasizes interbreeding, and so a species becomes synonymous with its gene pool. As long as two organisms belong to the same gene pool, they're the same species. This makes things easier, because scientists have a lot of tools for tracking genes in a population, measuring gene flow, and testing genetic similarity between related species. Seems simple. Almost too simple. As much as we love the BSC, there are still lots of cases where it starts to fall apart.
One of the more interesting scenarios in which the BSC doesn't quite work is ring species. Ring species are groups of closely related species whose ranges partially overlap to form an imperfect ring. Each species can hybridize and pass on genes with the species on either side of it, but the kicker is that the last species in line is so different from the first species, that they cannot hybridize.
A classic example of ring species.
This is like evolution's version of telephone. Usually it comes out totally distorted and hilarious. Ring species are less hilarious, but fascinating from the perspective of the BSC.
Here's why they're so fascinating: the first and last species in the ring cannot hybridize with each other. According to the BSC, they're separate species…right? Nope. Because of the hybridization that occurs in all the in-between species, it is theoretically possible that a gene from the first species could get passed on through all the others and ends up in the last species. That's a pretty successful game of telephone. Under the BSC, it makes them the same species.
In fact, any time there's hybridization between two species—even if it's only the tiniest little smidgeon—the BSC says they're the same species. This gets to be a little troublesome, as there are lots of examples of two populations that can hybridize, but rarely do. Maybe they have totally different geographic ranges or seasonality, or slightly different mating preferences. If we look at their population genetics we see that they belong to separate gene pools. The pools spring a rare leak every once in a while.
What do we do if the BSC says two entities are the same species, but our intuition says it just isn't so? We invent a new taxonomic level. Enter the subspecies. A subspecies is a population or group of organisms that can interbreed with individuals from another group or population, but usually don't. Now we have a way of designating potentially interbreeding groups from actually interbreeding groups, and we don't have to abandon the BSC. Have you thanked a taxonomist today?
The truth is, there are plenty of weird situations where the BSC isn't foolproof. For the time being it's the best we've got, and it's a pretty decent working definition. By focusing on interbreeding and gene flow, we can shift our focus from what a species is and get to the good stuff: why are there so many different species, and how did they arise?