What is a Species, anyway?

Once considered a distinct species (Melithreptus laetior), the Golden-backed Honeyeater is now lumped with the Black-chinned Honeyeater (Melithreptus gularis)

Over the past few weeks I have been able to add several new species to my life list mostly because I have begun to pay extremely close attention to the identification of some difficult birds.  Since some of the species I have been looking for are so incredibly difficult to identify, I have begun to wonder about the definition of a species and why do we have this compulsion to classify animals.

Modern taxonomy started with Linnaeus in the 18th century and today we continue to expand on his systematic classification of plants and animals.  Taxonomy is the process of classifying a species within the context of other related species.  Closely related species for example are placed in the same genera while closely related genera are placed in the same family.  On and on it goes up through orders, classes, phyla, and eventually kingdoms.

Could this desire to classify and place each animal in a defined category be motivated by our attempt to seek order out of chaos?   This classification does provide evolutionary context for a species, which can valuable for scientists trying to study any particular species.  First, let’s back up a few steps to the very beginning…

What is a species?

I think the most basic definition of a species is a group of organisms that can interbreed and produce viable offspring.  Simple enough, right?  If they can interbreed and their offspring can breed successfully, then bingo, we’ve got a species.  It’s nice and simple at first glance, but let’s dig a little deeper.

The Endangered Cape Sable Seaside Sparrow is now classified as a subspecies of the Seaside Sparrow (Ammodramus maritimus mirabilis).

First, a group of organisms must be able to interbreed.  Seems simple enough.  Obviously some animals just aren’t compatible.  After all, a warbler and an eagle aren’t going to be breeding anytime soon.  Furthermore, throughout the animal and plant world, we find an almost infinite ways that individuals distinguish between individuals of a closely related species and their own.  One way to ensure that breeding only happens with individuals of your own species is to separate breeding by time, space, or behavior.  For example, if one species of plant blooms and releases its pollen in March, another closely related species that attracts the same pollinators would benefit from blooming in April.  In fact, both plants would benefit, as the pollinators wouldn’t go from a bloom on species A to a bloom on species B.  This separation in time ensures that these two species don’t interbreed, but does it mean that they can’t interbreed? These are two very different ideas.

What happens if a scientist comes to the forest in March and collects pollen from species A and stores it until species B blooms in April.  She then hand pollinates species B with species A pollen.  A few months later you have fruit on species B and if you germinate those seeds they grow to be full grown hybrids.  What does this mean?  When will this plant bloom? March? April? Somewhere in between?  How does this impact our definition of a species?  The species don’t breed in nature but they are capable of interbreeding under artificial conditions.  Does this mean they are still separate species or not?  How do we handle the idea of artificial interference?

Let’s look at the second half of our earlier definition; the offspring of a species must be viable.  This second condition means that some species will be capable of breeding but that their offspring will be sterile.  The most common example is a mule, a hybrid between a horse and a donkey that takes on characteristics of both but is sterile.  If you want more mules, you have to breed more horses and donkeys.  That is a clear cut example of how this works, but what if we have species that do hybridize and produce viable hybrids?

The Mottled Duck (Anas fulvigula) is rapidly declining in Florida and hybridization with the ubiquitous Mallard (Anas platyrhynchos)

The obvious example to a birder in eastern North America is the Blue-winged and Golden-winged Warblers.  These two closely related yet very distinctive wood warblers hybridize regularly where there ranges overlap.  Not only do these birds produce viable offspring, at one time these offspring were thought to be a distinct species, the Brewster’s Warbler.  Today, we know that both Brewster’s and Lawrence’s Warblers are in fact hybrids of the Blue-winged and Golden-winged Warblers.  But if our rule was that a species couldn’t produce viable offspring with another species then how do we consider these warblers?  Why don’t we simply classify them as a single species with two very different subspecies or color morphs with a zone of overlap?

The reasoning stems from the frequency of hybridization.  If these birds were indeed the same species, interbreeding should occur rather frequently but in fact this hybridization occurs far less than would be predicted by chance encounters and matings.   Further evidence comes from mitochondrial DNA comparisons.  And here, we enter a world that Linneaus could never have fathomed.

The Kalkadoon Grasswren (Amytornis ballarae) was recently split from the more widespread Dusky Grasswren (Amytornis textilis).

Using DNA or mitochondrial DNA comparisons means that we can classify animals based solely on their genetic code, even when morphological differences aren’t readily visible.  Recent work in birds has resulted in the splitting of numerous species and I believe this trend will continue.  I must admit that when I read papers describing the differences in base pairs and other factors involved in genetic comparisons, I am a bit lost.  My background is in ecology and natural history, not genetics so I have a lot of catch up to get current with this research.

However, these genetic comparisons bring to mind a very big question… Why?  Why do we care that two populations are two specific species even though they are indistinguishable in the field?  What impact could this have on us anyway?  Even more baffling, what impact could this have on these two populations if we call them a single or a separate species?  I will address these questions and a few more in next week’s Sunday Essay.  In the meantime, leave your thoughts on why we, as scientists, have a compulsion to classify species even if it requires extremely advanced techniques and machines to distinguish one from the other.

Be Sociable, Share!


  1. Hey Drew,
    Another great article. I consider genetic classification to be the flat screen tv of current science: it is the newest and “greatest” technology.

    The same thing has been occurring in human subjects research lately. We scroll through myriad DNA abnormalities with the hope of finding some meaningful change that results in a physiologic change. The problem is that genetic differences (genotypes, the base pair differences you mention above) will generally translate to very small changes on the whole (phenotype).

    Then you get into mitochondrial DNA. The idea is that mtDNA is less variable than genomic DNA. So this gives you a better trend of the true ancestry of the bird, but you can imagine that birds with similar mtDNA could have very different genomic DNA leading to very distinct phenotypic traits…

    Although I really like the history that the Linnaean taxonomy brings about, but perhaps there is something to be said to using mtDNA, at least in some fashion?

Speak Your Mind