Well, what Neal had in his office I think really seals the deal. Apparently Steve Emlen, from Cornell, collected a series of Jacanas from western Panama in the mid-1990s. These jacanas clearly show mixed assemblages of the facial ornamentation of the two species (along with the plumage typical of the Northern Jacana, but drabber, and with much less distinction between the brown and black regions that one would find in a proper Northern Jacana).
I couldn't get big fotos on this blog, so I'll redirect you to my website for the photos. In the first two photos, you can see a Northern Jacana compared to a Wattled Jacana. Note the three lobes in the Northern versus the two lobes and the long rictal wattles (handlebars). In the second set of photos you can see hybrid jacanas compared to a proper Wattled Jacana. Note that the rictal wattles are reduced, and the specimen label notes that the shield had tinges of blue throughout a very palid red (rather than the bright red of a Wattled Jacana). Also, note that the shields are reduced and some are tri-lobate (like Northern Jacanas) rather than the bi-lobate form of Wattled Jacana. In the final picture, you can see the variable nature of the brown plumage on the underparts, and you might find that it's easier to see the reduction of the rictal wattles.
Hybridization in the eight species of jacanas (the remaining six species are found in the Old World) is unknown except for Wattled and Northern. Likewise, Wattled and Northern Jacanas have often been classified as conspecific. Below is a a small phylogeographic tree of the two species. The data come from the DNA barcode (COI) region, a small section of mitochondrial DNA.
The first thing that you should note is that there is consistent genetic differentiation between Wattled (J. jacana) and Northerns (J. spinosa). This is what you would expect to see if the two forms were really separate species, though this doesn't PROVE that they are distinct species. (But with the hybridization data, it makes a strong case, more on that below). Secondly, the is little genetic variation in either species across their sampled ranges, and none of the variation is geographically structured. That suggests that each species is able to maintain gene flow over long distances, and that maybe the species are relatively young. The final thing that you should notice is that the degree of genetic differentiation between the two species is small (about 1.5%), which confirms that they split recently, and falls within the range that we expect hybridization to be possible (several studies have show that there are genetic incompatibilities that make even successful hybrids inviable after about 7-8% differentiation).
The fact that Wattled Jacanas from Cocle have Wattled Jacana mtDNA and Northern Jacanas from Chiriqui have Northern mtDNA suggests that the species boundary is relatively robust between these two forms. Ornithologists used to think that if there was hybridization, then "speciation was not complete", and therefore called the two forms subspecies. We now know that a narrow hybrid zone where genes come in but don't leak beyond the narrow hybrid zone is a fairly common stage at the tail end of the speciation process, and that for most of the individuals in each species, species-specific genetic cohesion is maintained. So I think that it's fair to say that the hybridization actually substantially bolsters the argument that we've got two reproductively-isolated species here.
But that final part is where things get interesting, and where I hope to have a student working on this question next year. First, we need to improve our sample of mtDNA to demonstrate that outside the contact zone that species-specific genetic cohesion is maintained (sample of one and three don't exactly pass muster in peer review). Secondly, what is the genetic composition of the hybrids? Do they always Northern mtDNA, which would suggest a mating of a Northern female with a Wattled male, since mtDNA is transmitted only by mothers, or do they only have Wattled mtDNA, which would suggest the reverse? Often we find unidirectional hybridization, that is either because one species is demographically much larger than the other, or because one species prefers mates of the other. Such is the case in the Manacus hybrid zone in Bocas del Toro, where candei females prefer the golden plumage of vitellinus males over the white collars of the males of their own species.
One of the interesting aspects of jacana biology is that they have a curious polyandrous mating system. Females have harems of males. Each male receives a clutch, but it may or may not be its offspring that he cares for (This comes from Steve Emlen's paper in Proceeding of the Royal Society published in the late 90s; their study population was in Gamboa: there is a nice PBS movie about this research). From the female perspective there is little disadvantage to hetero-specific mating (i.e. across species lines), compared to most bird species where the female gets only one clutch per season and so therefore much be much choosier about her mate. This may facilitate the hybridization. So my student will need to do some behavioral ecology to get at this question too.
Finally, what limits the gene exchange among jacana species? There must be something going on because Jacanas don't make a hybrid swarm where the two species are completely blurred and we can't draw limits to species.
Because mitochondrial genes are passed through the female line, and in birds, unlike mammals, females are heterogametic (i.e. have the two different classes of sex chromosomes) mitochondrial DNA might have a much harder time crossing the species boundary compared to nuclear DNA. This is because of the phenomenon of Haldane's Rule, which observes in hybridization the heterogamic sex suffers the consequences of hybridization more severely than the other sex....thus we observe only female mules, because any male offspring of a horse X donkey mating are inviable and are aborted. If Haldane's Rule is operative, and there is no reason to think that it isn't, we should observe that most hybrids are male (which we would need to do during the preparation of museum specimens, since sex organs are inside birds), and perhaps more interestingly, we should also observe that nuclear DNA is moving quite handsomely between the two species since hybrid males may successfully be mating and backcrossing, bringing a half a genome's worth of heterospecific DNA into the backcross.
However, there are two ways that the breeding system might actually put the stops to the introgression of heterospecific genes. First, imagine that a male rears his clutch, if it is clearly conspecific, he doesn't know whether or not the clutch is his. But if it is a hybrid clutch he can be pretty sure that it isn't his, and perhaps his care is reduced, or the clutch is abandoned, or perhaps even killed. Again, this is where observations are important. Secondly, we've seen that the secondary sexual traits of hybrids are under-developed (the facial shields and wattles). Perhaps hybrid males are "unlucky in love" as it were, and since it is likely hybrid males that are the engines of backcrossing gene flow (backcrossing being the means of getting heterospecific DNA into the parental population), then if they don't mate often, they are "socially sterilized" rather than intrinsically sterile.
There's a lot of interesting biology yet to be discovered in the jacana hybrid system, but the first thing that we need to do is map the hybrid zone. So, when you are out birdwatching, I beg you to pay some attention to the flocks of Jacanas that you see. Both kind of data points are considerably useful: if you only see Wattled or Northern that is as useful as if you see mixed flocks. If you can email me (millerma AT si DOT edu) with your sighting, I'll update this posting with a map of your points. This is a great opportunity to do some citizen science here in Panama!
