Hybrid zones have been long lionised as "windows on the evolutionary process", principally because one can dissect how different traits can prevent gene flow and thereby contribute to species formation. The Bombina hybrid zone has been at the forefront of this research for the last 15 years, and I've been exploring the effects of mating behaviour and habitat preference on the strength of assortative mating. My collaborators on this work have been Beate Nurnberger and Sonja Kohler of LMU in Munich, and my PhD supervisor Nick Barton in Edinburgh. We also had an amazing amount of help from Ioan Ghira of Babes-Bolyai University when we were working in Romania.
What are Bombina?
A Bombina orientalis stuck on a stem. This species is the pin-up of the Bombina genus.
For a start, they're not bumble bees, as their latin name is Bombus. The common name for the genus Bombina is fire-bellied toad, and one can often find them in pet stores or in Sea Life centres and the like. However, the most commonly kept species Bombina orientalis isn't very interesting for speciation research, as it occurs in China and doesn't (as far as I know) hybridise with anything else.
The two species from Eastern Europe are far more interesting. Bombina variegata (the yellow-bellied toad) is the more western species, and it is a classic puddle-dwelling amphibian. It has a thick warty skin and long legs, both of which adapt it to crawling between puddles once they dry out. They are typically found in upland and mountainous areas, perhaps because puddles last longer at high altitudes in the hot Central European summer.
Their bellies are mostly yellow; they roll over and arch their back to expose their underside to would-be predators to show that they're poisonous. This is known as the 'Unken reflex' after the German name for Bombina. If you get Bombina skin secretions in your mouth or up your nose you can get 'Unkenschnupfen': an uncontrallable sneezing fit.
Bombina variegata from above and below. Note the extensive yellow patches on the belly and the warty back.
B. variegata also have a quiet and fairly rapid mating call. To listen click here. B. variegata don't have vocal sacs, and their calls can therefore only be heard over 20-30 metres. You can find large numbers of males together in a shallow puddles after rain in spring and early summer, and the males typically sing faster and faster as the puddles get warmer through the day. The males sing to attract females and then fight to determine who gets to mate. The victorious male ends up clasping the female around the waist (males have special leathery pads on their forearms for this purpose), and they swim around together for several hours. The females deposit eggs onto stems and leaves in the water and the male then fertilises them.
Bombina bombina, the fire-bellied toad, is a lowland pond-dwelling species found in Eastern Europe and the Balkans. They have shorter limbs than B. variegata, and a thinner skin. Their backs are smooth, and their bellies have only a few red spots. They also have an 'Unken reflex', but they're generally much harder to catch than B. variegata.
Bombina bombina in various poses. Note the vocal sacs and smooth dorsal skin. They tend to look happier than B. variegata.
The mating system of B. bombina differs markedly from B. variegata: the males have vocal sacs, and a loud chorus can be heard over several kilometres. To listen, click here. The males defend small territories on the surface of their pond, and the females swim about deciding who to mate with. The males also signal to each other by making ripples. As with B. variegata, the male grasps the female around the waist and they swim around together laying eggs.
Bombina hybrid zones
Despite all the differences listed above, B. variegata and B. bombina still interbreed with each other whereever they meet. They typically come together where floodplains meet ranges of hills throughout Eastern Europe. More surprisingly still, their hybrids survive to breed, and the hybrid zones contain a variety of hybrid types, ranging from almost pure individuals to toads exactly intermediate between the parental species.
Distribution of B. variegata and B. bombina in Europe. The lines show the hypothesised colonisation routes after the last ice age. The numbers indicate well studied areas of the hybrid zone: 1. Krakow 2. Przemysl 3. Pescenica 4. Apahida
(Modified from a map by JW Arntzen)
The hybrid zones in Krakow and Przemysl are classic tension zones, which means that there is a smooth transition between populations containing mostly B. bombina individuals through intermediate populations to sites containing only B. variegata. The frequency of allozyme alleles diagnostic for either species also follows a smooth cline.
A pie map of the Krakow tension zone. The black portion of the pie shows the proportion of B. variegata alleles in that site. Note the smooth transition as one moves north-south. From Barton & Szymura (1986)
The zone is probably maintained by selection against the hybrid toads, acts to steepen the transition between the two species, and dispersal, which widens it. The width of the zone therefore represents a balance between these two forces, and we can use this fact to make many useful inferences. The intermediate sites contain mainly hybrids, but some pure individuals disperse into these sites from outside the zone. This dispersal creates associations between the allozyme alleles (=linkage disequilibrium), and these associations can tell us about the rate of dispersal.
The Krakow tension zone in one dimension. The zone appears as an abrupt step because the y axis is on a logit scale. Note the close fit of the points to the line. From Barton & Szymura (1991)
Hybrids don't survive as well as pure individuals, perhaps because alleles from B. bombina conflict with B. variegata alleles when they come together in a hybrid (=endogenous selection). Habitat differences don't seem to play a role in these two hybrid zones, as there is no obvious connection between habitat type and the toads found there.
Mosaic hybrid zones in Bombina
The hybrid zones in Pescenica and Apahida are structurally quite different to the Krakow and Przemysl zones, because here the toads show a preference for different habitat types. The pattern of genotypes on the landscape then partially reflects the distribution of the two habitat types. In Pescenica, this results in a patchwork pattern of hybrids and purer types in the centre of the hybrid zone, but overall the shape is still similar to a tension zone. Zones where the two species interbreed in a patchwork are known as 'mosaic' hybrid zones.
The mosaic hybrid zone in Pescenica. Note the patchwork structure at the centre. From MacCallum et al. 1998.
The hybrid zone in Pescenica. Because B. bombina and bombina-like hybrids prefer ponds and B. variegata and variegata-like hybrids prefer puddles, there is a consistent difference in allele frequency between the two habitat types. Modified from data in MacCallum et al. 1998.
This mosaic structure probably arises because adult toads move to their preferred habitat type. However, hybrids might also be unfit in either ponds or puddles because they carry alleles adapting them to the other habitat type (=exogenous selection).
In the Apahida hybrid zone, there is no overall transition from B. bombina to B. variegata, and the structure of the zone is determined entirely by the distribution of habitat. The toads here also have a active preference for their different habitats, although selection against tadpoles in the wrong habitat type is probably also important.
The mosaic hybrid zone in Apahida. Here, the distribution of ponds and puddles on the landscape determines the struture of the zone.
My Ph.D. work (available here) focused on how this mosaic structure could be maintained. I found that the rate of dispersal between ponds and puddles was quite high (c. 20% of toads in a site are immigrants), and thus extremely strong selection is required to maintain the differences between B. bombina and B. variegata in this area. This implies that the Apahida hybrid zone is collapsing into a hybrid swarm.
Assortative mating and habitat preference
Considering that B. bombina and B. variegata have different mating calls, can they tell the difference when choosing a mate? Studying Bombina mating patterns in the lab is quite hard, because habitat type also determines how likely a female is to mate there. We therefore opted to use molecular markers to indirectly infer the mating pattern from eggs sampled from natural sites containing both species.
It turns out that there is almost no assortative mating, but breeding habitat choice appears to be very important. To explore the latter further, we aim to quantify how the distribution of adult toads among different habitat types translates into the distribution of offspring. If breeding habitat choice is much stricter than resting habitat choice, the eggs should only be found in the most pond- and puddle-like habitats, whereas toads should be found everywhere. More detailed results should appear here soon.