What springs can teach us about speciation

A question I often find myself asking when I look at springs is: why are discharge springs SO diverse and such hot spots of endemism?

This question is a central question in biology in general; why do some areas seem to foster higher rates of diversification? What conditions are required to ‘create’ new species? G.A.B. springs present a great model system for thinking about these questions and I’m going to present a short ramble along the lines I have followed whilst pottering around in their ankle deep water.


Discharge springs are ridiculously diverse

One of the remarkable things about discharge springs is the high proportion of endemic species (see https://spinelessecology.wordpress.com/2014/08/25/springs-101/). Whilst other freshwater bodies in the aridzone are characterised by ‘cosmopolitan’ species (in other words, species that are found broadly across a wide variety of fresh water systems), springs generally host these species PLUS a suite of locals that are never found in anything other than a spring, and are rarely even found outside of a single locality.

To give you a general picture of how impressive this is, consider the following. Thanks to the efforts of a dedicated few [1] we have a general idea of the kinds of things we find in some of the bigger regional spring clusters (or supergroups)[2]. Whilst the total number of species varies considerably across supergroups, the majority of the species found are insects (the orange parts of the pies in Figure 1), few of which are endemic. In contrast, within the snails (one of the other major representatives) a large portion are local endemics, and in most places about half of the species are found nowhere else.


This pattern of local diversity is repeated at a smaller scale. Take Edgbaston for example, one of the most diverse spring complexes in Queensland to remain in good condition [3]. This complex sits within the Barcaldine supergroup (number 4 in Figure 1), home to the widest diversity of snails in Australian springs (23 species). Within just one locality in this group (Edgbaston – an area of about 8,000 hectares containing about 200 springs) we find fifteen of those 23 species, nine of the fifteen endemics in the region (Figure 2).

Endemic snails

We know that the majority of these species have radiated locally rather than on the coast where the majority of their relatives live (see my other post about the origins of spring species). So, that leaves me asking, why have snails radiated so heavily in springs but nowhere else? And what is so special about Edgbaston?


How to make a species

Generally there is a lot of debate about how you get the formation of new species. Rather than engage in that here I’ll briefly highlight the conditions I feel are needed and why.

The first thing you need is isolation of a population– most species concepts acknowledge this. If you have two populations still in sympatry [4] any mutation is likely to either spread across the whole species or dissolve in the broader population pool. You need to get a little pocket of individuals trapped somewhere in allopatry [5] before there is the potential for innovation to spread, fix and create a new species.

Second, you need that population to be small. Now this one is open for debate still, but I feel we can at least all agree that the probability of getting fixation [6] in any trait is easier with a small population.

Now it gets tricky. Third, you need some form of force fostering change. Now technically this doesn’t have to be directional, especially in small populations drift can play a huge role [7] – drift is a neutral process that sees a trait that may offer no particular advantage fix across a population just because there are not many individuals. It can also be adaptive, for example the place a population is stuck may also experience some form of environmental change (i.e. get warmer, get wetter, the vegetation may change etc.) that the species can only endure with some innovative adaptation. If you have the same environment you are unlikely to get changes that are adaptive (you might get drift). And if you have a selective force (that is not so strong that it kills everything) you are likely to get changes that reach fixation quicker.

Finally, and this is the cherry buried at the back of the fridge somewhere that you really need to finish your cake, you have to get changes to the mating system [8]. This is probably going to require a lot of generations apart, or some particular selective force on the way they reproduce (maybe the pheromones they used to use are interfered with, maybe the nesting site no longer available, maybe the display behaviour is no longer effective enough). What this means is, most of the time, if a species satisfies steps one through three, you might end up with something that looks different, but if/when this population comes back in to sympatry you will get introgression and their special traits may dissolve back into the more common pool or, if some mechanism limits this introgression, persevere as a local race. Something is not a new species until its mating recognition system keeps its population separate from the population it diverged from (Figure 3).


Permanent impermanence

So, how do these steps associate with snails in springs? And how is it different for insects? In my mind, springs create the perfect environment for speciation in snails due to their permanent impermanence. Locally, a set of springs are permanent – there is always a spring or two found somewhere in the locality even if one particular spring dries up. However, each individual spring is impermanent – it may dry, it may change its flow path, it may be invaded by a new plant species etc.

For a snail that must be in the water at all times, each spring is like an island of wetland in a sea of aridity. There is no way to easily traverse between springs so each population in each spring is, for all intents and purposes, isolated (step 1). Now population estimates at a whole spring level are not yet possible, but from my preliminary data you’re probably looking at least a couple of 1000 individuals in your average sized spring for each species, which is pretty small (step 2).

And finally, you have the force, which in this situation creates the perfect conditions due to that permanent impermanence. Species are trapped in isolated populations, which means change can come about just by drift spreading some freak mutation across the whole population. I already see this in some springs where the prevalence of a black morph of one species is increasing in one small spring.

What you also get is some strong selective forces creating changes. If a species is trapped in a spring that begins changing – its getting smaller, its getting shallower, its getting hotter – its’ options are to adapt to that change or perish (Figure 4). If it adapts, that trait will quickly spread across the individuals due to the small population and frequent breeding (it looks like most species breed all year around). So, as long as they stay isolated for long enough that some form of change happens to the reproductive system (in springs snails it looks like a lot of this is minor changes to the reproductive organs, even those that look almost exactly the same have different reproductive organs [9]) they will then become a new species. When this new species later disperses, perhaps back into allopatry with its close relative, it will not be able to interbreed and its new traits will shape its new ecology. Early work on species of spring snails in South Australia present a great example of something like this happening – two species with close phylogenies now co-inhabit some springs but have very different thermal tolerances so persist in different areas (i.e. the more high temperature tolerant one sticks to the hot bits)[10].


The awesome thing about springs however, is that the risk you take in being stuck in a changing environment is not likely to cause extinction. If a population perishes in one locality in the face of environmental change, it’s not the end of the world. Thanks to the local permanence of water it is very probable that somewhere close by there is another population, chugging along nicely, acting as a backup that can, when it gets wet and the springs all connect up, can re-invade that area and start the whole process again. It’s kind of like you’ve got these little risks being taken all the time that are no real threat to the species as a whole because you’ve got a backup hidden away somewhere else.

Why just the snails?

You’ll recall that, whilst proportion of species of snails in springs that have radiated locally in this way is high, there are other organisms that do not have the same patterns of endemism. The insects, for example, are the most specious in the system but share numerous species across supergroups and can be found broadly across all springs (Figure 1). Why no local radiation events making new endemic species of insects? This all comes down to the way the steps to speciation interact differently with the traits of most insects.

The majority of insect species in springs, whilst aquatic at either one stage of their life (e.g. soldier flies, caddis flies) or their whole life (e.g. diving beetles) are able to tolerate desiccation at some point. This point is usually the most dispersive phase (i.e. the adult, with wings and a nice hard carapace to keep it moist inside). This means that, on the path to speciation, they don’t even get to step 1. As soon as spring starts changing, or becomes inhospitable in any way, the adults fly away and persist somewhere else where life is a bit easier.

This is a general pattern with endemism in springs – the species that have broader dispersive capabilities have fewer endemics than those that have limited ability to move between springs. In some way, the disadvantage of not being able to flee a changing spring is the catalyst for creating new species. Some springs researchers are taking this so far as to argue for the recognition of evolutionarily distinct lineages in some spring endemics [11]. Since it seems inevitable that this combination of factors will continue to create species in the future, what we should perhaps protect is not the species we recognise now but their potential to become species in the future.


1) I have the work of Winston Ponder of the Australian museum to thank for most of my work, though there are numerous others.

2) Keep in mind, the majority of these animals are invertebrates and the few specialists are unable to describe the vast number of discovered species fast enough (so this is a conservative estimate)

3) We actually expect that many other complexes were comparably diverse back in the day – for example the Eulo complex, a few hundred kilometres further south of Barcaldine, would have been almost as diverse before it suffered broad-scale habitat alteration. See Fairfax, R. J. and R. J. Fensham (2003). “Great Artesian Basin springs in Southern Queensland 1911-2000.” Memoirs of the Queensland Museum 49(1): 285-293.

4) Sympatry i.e. still living in the same locale (http://en.wikipedia.org/wiki/Sympatry) <- though this article is HEAVILY influenced by someone arguing for sympatric speciation – something I am not adhering to here

5) Allopatry i.e. isolated, in different locales, opposite of sympatry (http://en.wikipedia.org/wiki/Allopatric_speciation)

6) Fixation i.e. when a particular allele or gene is spread across the whole population to the point all individuals posses it (http://en.wikipedia.org/wiki/Fixation_(population_genetics))

7) This one is important to me because I adhere to a definition of species that believes they are generally under stabilising selection on the conditions and behaviours that get them breeding; so making new species is a tricky affair. See xxxxxx

8) It’s hard to find an in-depth summary of Patterson’s concept online that is freely available. This is a great book, if you are interested: http://www.amazon.com/Speciation-Recognition-Concept-Theory-Application/dp/0801847419

9) Winston Ponder is responsible for pulling apart all these species, the ones I am talking about here are in this paper: Ponder, W. F. and G. A. Clark (1990). “A radiation of hydrobiid Snails in threatened artesian springs in western Queensland ” Records of the Australian Museum 42(3): 301-363.

10) Again, Ponder, this time with Colgan: Colgan, D. J. and W. F. Ponder (2000). “Incipient speciation in aquatic snails in an arid-zone spring complex.” Biological Journal of the Linnean Society 71(4): 625-641.

11) Nick Murphy and Michelle Guzik are pulling this line with their work on Amphipods in South Australian springs, seen here: Murphy, N. P., M. Adams, M. T. Guzik and A. D. Austin (2013). “Extraordinary micro-endemism in Australian desert spring amphipods.” Molecular Phylogenetics and Evolution 66(3): 645-653.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s