Are ant social parasites and host ants examples of sympatric speciation?
Whew, this is quite the question!
The short answer is: I don't know, but probably not.
To give a longer and somewhat more complete answer (though still not one that might completely satisfy you) I'm going to back up a little bit and get our other readers up to speed on some of the different strategies ants use to exploit each other--the different flavors of nest parasitism. I also think it might be useful to define sympatric speciation.
There are many species of ants, such as the members of the (Diploroptrum group within the genus Solenopsis, aka 'thief ants') that are thought to make their living from living very close to larger ant nests and stealing stored food and/or eggs and other brood. At the nest level, these ants are technically parasites (I've heard myrmecologists refer to them as kleptoparasites, although this term is usually reserved for organisms that steal food or other resources from other organisms, not baby-eaters). The exact behavioral ecology and the extent to which these relationships are specific is not known for many of these groups, but the hosts of thief ants tend to be larger ants from different genera, so it's unlikely that members of this variety of nest parasitism are examples of sympatric speciation.
While "thief ants" tend to build their nests just on the outskirts of a host nest, there are other ants that actually live inside ant nests. Animals (ants and otherwise) that live in other social insect nests are said to be "inquiline." There are many kinds of arthropods and other organisms that have observed to be inquilines in different types of ant nests, and there are ants that are inquiline in termite nests (for example, members of the genus Metapone). Either case could, I suppose, be considered an "ant social parasite," but for the purposes of this blog post, I'll focus on ants that are inquiline within other ant nests. Much more detail about the many flavors of inquilinism can be found in this excellent review (Buschinger 2009), but rather than summarize that article, I'm going to emphasize that inquilines are the category of parasites that might be most promising group in which to look for examples of sympatric speciation. I will return to this idea (finally!) in just a few more paragraphs.
Another parasitic strategy that some ants employ is stealing pupae from another nest, bringing them back to their own nests, and making those ants do all the work. These ants have sometimes been referred to as slave-raiders (see a great discussion here as to whether or not to refer to use this terminology); Polyergus is the textbook example. This genus is exclusively comprised of ants that steal pupae from members of the genus Formica. There are several species of Polyergus, but some of them seem to have non-overlapping geographic ranges. For example, Polyergus samurai is found in Japan, whereas other species are found in Mongolia, Northwestern Europe, or the American Southwest. These distinct geographic ranges suggest that allopatric speciation might be most likely mechanism of diversification in this genus, Although it is intriguing that they come out sister to their host ant, Formica, in the ant phylogeny by Moreau et al. (2006).
Speaking of allopatric speciation: what is that? And what is sympatric speciation?
Saying that speciation is allopatric is basically just saying that a pair of populations speciated because they were geographically isolated. Speciation is said to be sympatric if reproductive isolation occurs in populations whose ranges overlap substantially (it doesn't count if the populations speciate while they're separated and then later come back into contact, which can make distinguishing between these two modes tricky). The classic examples of sympatric speciation involve host shifts in parasites: if the ranges of the hosts overlap, and speciation occurs because of a host-shift, then many would argue that this is an example of sympatric speciation. Others would actually classify this as "micro-allopatric," because at small scales, the ranges of the new species never overlap, i.e., they're "never" on the same plant. I think this distinction is a little silly, though. To me, allopatric speciation implies the existence of a (usually geologic) barrier to dispersal between populations, like a river or a mountain range. To say populations are in allopatry because they are behaviorally inclined to eat different plants is just...silly. Stephen Stearns at Yale has a great lecture on speciation here if you want more depth on this somewhat daunting subject.
Anyway, this definition suggests two intriguing ways in which sympatric speciation might be important for the diversification of ant social parasites: new species might arise in lineages that are already parasitic because of a host switch, and parasitism might arise within a population of non-parasitic ants. The former case is straightforward in principle: in order to better dominate/fool their hosts, parasitic ants need to adjust their smell (cuticular hydrocarbon profile), which is also important in mate recognition; thus, if you've been parasitizing different species, you might not recognize each other as mates. There are several species of ants that are known to parasitize a few different hosts, and these might be interesting candidates to look for cryptic, sympatric speciation within (cryptic species are genetically isolated, but difficult or impossible to tell apart under the microscope). I don't know of any documented examples of this speciation-by-host-shift mode, however, which may be due in part to how risky the nest-raiding lifestyle is. It may also be due to my incomplete knowledge of this subject, and the generally incomplete state of natural history and genetic knowledge of ants. In general, though, the inquiline lifestyle is risky, and host ants are patchily distributed across landscapes. Host specialization is a good strategy for herbivores with abundant food sources (whose only defenses are chemicals; plants won't try to bite your body in half), but if there are examples of speciation-by-host-switching in ants, I suspect those species would have a very high extinction probability, so at any given time they would be exceedingly rare.
Sympatric speciation due to a shift from a free-living lifestyle to parasitizing your relatives is difficult to rule out. There are many instances of social insect parasites that seem to be closely related to their hosts (this pattern actually has a name: Emery's rule). There are numerous documented examples of intra-specific parasitism: newly mated queens attempting (and sometimes succeeding) in joining pre-existing colonies (so: intra-specific inquilines), and of same-species raids in which pupae are stolen and made to work in the new colony. It seems like a simple logical sequence to move from within-species parasitism to sister-species parasitism in a perfectly sympatric way. There is some unpublished evidence of multiple origins of inquilinism in the genus Tapinoma in North America, which is consistent with sympatric speciation, but these data are very preliminary, and cannot rule out allopatric speciation, as I'll discuss below.
The difficulty I have with the sympatric scenario is that there is still no mechanism for reproductive isolation: what prevents the "parasites" from mating with the "hosts" if they haven't already speciated? To me, it seems more likely that Emery's rule arises from the fact that sister species are pre-adapted to foiling each other's nestmate recognition abilities. The smells ants use to recognize each other have a hereditary component, and so closely-related species are more likely to stumble upon the perfect disguise, as it were. My scenario requires that:
1) Populations of a single species were isolated,
2) These populations speciate allopatrically,
3) They return to contact with one another, and
4) One evolves the ability to exploit the other.
This may seem like an unlikely scenario, but the first three stages are thought to have happened repeatedly in insect species, especially over the past few million years following the glaciation cycles in the northern hemisphere. (Siepielski et al. describe the ecological repercussions of this process in damselfly communities in the Northeastern United States here. A pattern that would be consistent with my scenario would be a higher proportion of close-relative parasitisms in areas that are thought to have undergone more recent, repeated range shifts and expansions, such as the Northern Eurasia and North America, and the continental islands of Indonesia, as opposed to areas that have been more stable, like the Amazon Rainforest and perhaps some of the Southern hemisphere temperate areas.
To me, this seems like a more likely scenario than either the host-switching or the rise of parasitism scenarios discussed above (note that Buschinger 2009, linked above, definitely disagrees with me). Some evolutionary biologists believe that, in general, we should assume speciation has been allopatric unless there is compelling evidence otherwise (for a more full discussion, see a 2002 review of sypatric speciation by Berlocher and Feder here). While my scenario is essentially a "just so story," I think these scenarios are important to consider when evaluating the likelihood of different mechanisms in ecology and evolution.
However, there are many areas of the world where new ants have been introduced, and it would be theoretically possible to catch a host-switch sympatric speciation event in action. Just as the most clear-cut example of sympatric speciation occurred during a host-switch from one plant to an introduced one (Rhagoletis flies switching from native hawthorn trees to introduced apple trees in North America), it might be very advantageous for inquiline ants in Europe or Japan to be able to exploit, say, the invasive Argentine ant (Linepithema humile). This might require a specialized smell disguise, which in turn could lead to a mutual lack of attraction between otherwise compatible mates.
This and many other areas of social insect parasitism are completely fascinating to me, but these ants are so rare that much of our knowledge is very fragmentary. When you look for ants, keep your eyes out for inquilines, but don't be disappointed if you don't find any!
I hope this helps!
Jesse Czekanski-Moir & the AntAsk Team