Black-throated Green Warbler (Setophaga virens)
Illustration: Louis Agassiz Fuertes (Wikimedia Commons)
This is part two of a three part series on hybridization in warblers (family Parulidae) called Infus'd Betwixt.
Humans have a knack for screwing up nature (needless to say). We’ve occupied almost every patch of productive habitat on the planet. We’ve not only occupied them, we’ve altered them too. One of the more obscure aspects of habitat alteration or outright destruction is the diminishment of isolating barriers that cause allopatric speciation. We don’t talk about habitat destruction in this way very often, but it has some grave consequences for certain species. Because humans are causing the ever more rapid destruction of geographic barriers between species, hybridization may very well be emerging as a more common phenomenon. Human-induced hybridization is yet another startling way in which we are managing to rid the Earth of its biodiversity. Golden-winged and Blue-winged Warblers in eastern North America seem to be falling victim to this kind of human-induced hybridization.
Determining the exact historical distribution of Golden-winged and Blue-winged Warblers is impossible but the notes left by 18th and 19th century ornithologists suggest that these two species occupied distinct ranges. They were mostly allopatric; Golden-winged Warblers to the north, Blue-winged Warblers to the south (with a little overlap here and there). By the late 1800’s a significant chunk of America’s eastern primeval forest had fallen to the axe. In its place were the indelible marks of human progress: pastures, fields, shublands and thickets. For millennia, dark and brooding forest which both species shunned in favour of more open habitats kept the breeding ranges of Golden-winged and Blue-winged warblers discrete. Each species followed their respective evolutionary trajectories with almost no chance to interact, almost no chance to interbreed. It was allopatric speciation at work and the forest separating the two species was crucial to seeing the process through to its end.
But as the axe and saw tore the forest barrier down, the Blue-winged Warbler began expanding its range into that of the Golden-winged Warbler. In many cases the Blue-wings occupied exactly the same habitats used by the Golden-wings: early successional shrubby habitats with adjacent forest edges. The scientific literature suggests that Golden-wings prefer slightly wetter habitats than their congeners, while Blue-wings prefer habitats that are slightly older and more ingrown. But these distinctions are not concrete and in many places both species occur in exactly the same habitat. Without the benefit of habitat differences to keep the species apart, the Blue-winged invasion was swift and not without consequences. In some places it took only a half century for Golden-winged Warblers to disappear, being replaced by their congener. Some areas have seen long periods of continued coexistence between the two species, but in general, following invasion by Blue-wings, the Golden-wings disappeared.
The decline of the Golden-winged Warblers continues at a rate of 15-18% annually throughout (most of) the remaining part of their range. Though not yet federally recognized as a species at risk in the United States, it is listed in Canada as threatened. The Golden-winged Warbler’s decline clearly has something to do with the Blue-winged Warbler, but what exactly? Was it the result of competitive exclusion of Golden-wings by more aggressive Blue-wings? It seems that in some interactions Blue-winged Warblers are the more aggressive species, but that isn’t always the case; some researchers have actually found that the Golden-winged Warbler to be the dominant species. In some parts of their shared range they respond to each others’ songs in aggressive ways. In others areas they don’t appear to consider each other competitors and overlap their territories without conflict. The dynamics of interspecific interactions during ecological invasions of this kind are complex, especially when invasions happen over a fairly large area. It can be hard to determine what leads to one species success at the expense of another. There has yet to be put forward a good explanation as to why the Blue-winged Warbler is managing so well, while the Golden-winged is in such dire straits. Understanding the interactions between these species becomes even more daunting when you consider they hybridize with impunity.
Wilson never saw a hybrid. Neither did Audubon. Neither did their contemporaries. The first Golden-winged × Blue-winged hybrid (a Brewster’s Warbler) wasn't discovered until 1874. The Brewster’s Warbler was a new thing in the late nineteenth century, and perhaps the first example in North America two species making contact across a geographical barrier because of human meddling. It marked a major milestone in the evolution of the Golden-winged Warbler; a milestone which may prove to be the last one on the Golden-wing's road.
Under normal circumstances hybrids are rare in nature, but in places where Golden-wings and Blue-wings co-occur, you can usually find hybrids flitting about without too much trouble. That’s a sure sign that interbreeding is happening often, perhaps more often than the casual observer realizes. While sometimes hybrids are obvious, other times they are not. That’s because a bird's outward appearance is not necessarily a true reflection of its genes. It takes some molecular detective work to diagnose some hybrids, thus the true extent of hybridization between Vermivora, one might say, is often hidden in plain sight.
Hybridization between Golden-wings and Blue-wings is so extensive that collectively the species have been called a hybrid swarm. A hybrid swarm describes a situation where members of a mixed-species population posses genes from other species, in addition to their own, as a result of hybridization and backcrossing. This means that if you take a Golden-wing from anywhere across its range and look at its genes you will find Golden-winged genes, but you will also find Blue-winged Warbler genes mixed in there too. Even birds that outwardly appear to be pure – they show no aberrant plumage – may very well be genetically impure. You may have a pair of Golden-winged Warblers breeding among the Northern Prickly-Ash (Zanthoxylum americanum) in your back forty, as I do. There may be no sign of Blue-winged Warblers in the area. That doesn’t mean your Golden-wings are pure. They could very well be carrying genetic material from both species. How did they get those Blue-wing genes if there are no Blue-winged Warblers around? The most likely explanation is that they were carried there by dispersing birds. Even a single impure Golden-wing, with origins from a place where hybridization is rampant, settling and successfully breeding on your farm is all it would take to pollute your local Golden-wings. The bottom line is that the only way to tell how genetically pure any bird is involves looking at its DNA.
Is anywhere safe from genetic mixing? There's some evidence that Manitoba and Quebec may harbour hybrid-free populations. Recent sightings of Brewster’s and Blue-winged Warblers in Manitoba means that the birds in that province are likely to start becoming polluted with Blue-winged Warbler genes. Until recently the relatively isolated Manitoba population remained out of reach of dispersing hybrids or Blue-winged Warblers. It was a safe haven for those precious pure Golden-wing genes. But how much longer that will last, who knows? Refuges of pure Golden-winged Warblers may exist elsewhere. Certainly not all Golden-winged Warblers in areas of sympatry show signs of hybridization; a 2009 study found less than 18% of the birds they sampled in some parts of the species’ range had polluted genes.
The pattern of gene flow between the two species is bidirectional, so typically-plumaged Blue-winged Warblers across their range (except in Kentucky, for some undetermined reason), have Golden-winged mtDNA. In exactly the same way that the Golden-winged Warbler’s genome is being polluted by the Blue-winged Warbler, the Blue-winged Warbler’s genome is being polluted by the Golden-winged Warbler. An eye for an eye. Or rather, a black-throat patch for a white one. If it’s genetic mixing we’re worried about, than the Blue-winged Warbler is at as much risk as the Golden-winged Warbler, but it doesn’t seem to get the same amount of attention.
One reason is probably because Blue-winged Warbler numbers are on the increase, and their range is expanding, while Golden-winged Warblers are suffering rapid declines with 64% of the global population of Golden-wings lost since the 1960’s. That’s speaking phenotypically of course. Are genetically pure Blue-winged Warblers actually on the increase? Early studies of the genetic interactions between Golden-wings and Blue-wings showed that gene flow was asymmetrical so that while Golden-wing genes became polluted, Blue-wing genomes remained relatively pure. We now know that’s not the case – gene flow is bidirectional. The genetic mixing flows both ways, leaving little chance genetically pure Blue-winged Warblers are faring much better than genetically pure Golden-wings.
The Vermivora situation is not the only one that has arisen due to human-caused environmental change. Another example can be found in a beautiful region of western Canada called the Peace Country. The Peace Country is a remarkable place because it’s the interphase of eastern and western flora and fauna. It’s here that species typically found west of the continental divide spill over the Rocky Mountain’s spine, mingling with eastern species they wouldn’t normally encounter elsewhere in their range. Pairs of closely related eastern and western warblers come into sympatry in the Peace Country, and that means that conditions are ripe for hybridization. Black-throated Green Warblers (Setophaga virens) meet Townsend’s Warblers (Setophaga townsendi). They hybridize. Myrtle Yellow-rumped Warblers (Setophaga coronata) meet Audubon’s Yellow-rumped Warblers. They hybridize too. Mourning Warblers meet MacGillivray’s Warblers (Geothlypis tolmiei). They also hybridize; and in their case it’s a relatively new fling, still fresh and passionate.
The Mourning × MacGillivray’s hybrid zone appears to be relatively new one. Mourning Warblers were apparently absent from northeastern British Columbia in the early part of the 1900’s. The first reports of this species came in the 1970’s. It was no accident, Mourning Warblers had some help accessing the Peace Country. Like the Vermivora warblers, Mourning and MacGillivray’s Warblers love shrubby open habitats, and shun the deep dark woods. Early regeneration of clear cut logging sites provides ideal habitats for these species. It was probably through range expansion into clear cut areas that Mourning Warbler reached the Peace County and came into contact with their closest relatives, MacGillvray’s Warbler. Human enterprise once again leads to the removal of a critical geographic barrier to hybridization, and biodiversity suffers.
There was a long running debate whether these two species actually hybridized. Apparent natural variation of traits within each species had led some biologists to doubt that birds which showed intermediate or mismatched phenotypic characteristics were hybrids. The best phenotypic trait for distinguishing the species is prominent white eye arcs in MacGillivary’s. Mournings are usually without eye arcs, but occasional birds throughout the species range may have narrow ones. Are these hybrids that have dispersed out of the hybrid zone? I’m not sure anyone can answer that. Immature birds can be really difficult to distinguish, even using formulae that take into account the relative length of tail and wing, an often reliable measurement for separating genetically pure adult Mourning and MacGillvray’s Warblers. So this situation was already complex long before the hybrid zone was confirmed using molecular techniques.
Hybridization and genetic mixing between Mourning and MacGillvray’s Warblers is really quite different from the Vermivora warblers. In this case the hybrid zone is estimated to be only about 130 km wide, and there’s little evidence to suggest that the genetic mixing generated there extends out into the wider range of both species. This is in sharp contrast to the Vermivora which hybridize across their highly overlapping ranges. It’s hard to say what keeps that hybrid zone so narrow.
East of the Rocky Mountains you'll find Morning Warbler, west MacGillvray’s Warbler. Generally speaking, you can recognize them by looking at their mtDNA or their Z chromosomes (analogous to the X chromosome in humans). Or for the more field-oriented naturalist, you may examine their plumage. Birds living east of the Rockies don’t have bold eye rings – the most important field mark for distinguishing these two species – west of the Rockies they’ve got more pronounced eye rings. The pattern also holds for song. Birds in the east have a song that is more or less distinctive from birds in the west, though it can require a little practice to tell them apart.
The seemingly convenient genetic and phenotypic separation of species falls apart within the hybrid zone. Genetic traits become mixed together, so that individual birds cannot be assigned to one or the other species based on their mtDNA or Z chromosome markers. There is a mix of complete, partial and absent eye arcs, and vocalizations become weird admixtures of Mourning and MacGillvray’s songs. Within the narrow hybrid zone it’s pretty much impossible to reliably identify pure birds and hybrids by plumage or song alone. The use of molecular techniques in combination with plumage characteristics could be the best way to pick out which birds are hybrids and which birds are not. Sometimes biology is far messier than field guides would lead us to believe.
Why don’t hybrids and backcrosses spread their mixed-species genes into the ranges of their genetically pure parent species? Are there evolutionary forces selecting against hybrids? No one is sure when it comes to Mourning and MacGillvray’s Warblers. But they are not the only ones in the Peace Country with a narrow and seemingly newly established hybrid zone.
Before the discovery of the Peace Country hybrid zone, only one putative Black-throated Green × Townsend’s Warbler hybrid was known. To look at its plumage any ornithologist would have called it a classic Townsend’s; genetically though, the story was different. It had the mtDNA of a Black-throated Green Warbler. At that time the two species were thought to have completely allopatric distributions, but given their plumage, song and behavioural similarities, as well as their recent divergence (about 1 million years ago or less), it was perhaps not surprising that these two species could interbreed if they ever came into contact.
Like Mourning Warbler, Black-throated Green Warbler may be new to British Columbia. The first bird was collected only in 1965. Since that time they have become locally common in some eastern parts of the province, including in the Peace Country where it forms a hybrid zone with its western counterpart, Townsend’s Warbler. It’s a narrow zone, seemingly only about 40 km wide on the east side of the Rocky Mountain crest. Similar to the Geothlypis warblers, Black-throated Greens east of the Peace County are distinctive from west-side Townsend’s. They differ in genetics, plumage and song. But in the contact zone there are individuals that sport intermediate genes and plumage, they’re fairly rare, but they’re there. Some birds in the contact zone can switch between the songs of both species and both species respond aggressively to the other’s vocalizations. Whether this is another sign of hybridization or simply just learning the wrong specie’s song is still an unanswered question. Also, many hybrid zone birds are intermediate for traits such as wing length, tail length and beak width. On average, birds in the allopatric populations would be separable based on such morphometrics.
Ornithologists have some reason to believe that the narrowness of the Black-throated Green × Townsend’s Warbler hybrid zone might be maintained by natural selection eliminating hybrids because their intermediate characteristics make them unfit. A number of things may be at work here, including that hybrids might have a disadvantage when it comes to migration and life on their wintering areas. Both species have very different migratory routes and distinct nonbreeding habitats. If hybrids have inherited a mixed bag of traits from their parent species they may be quickly weeded out of the population by taking an unsuitable route on migration or by attempting to winter in poor habitat. Inheriting maladaptive traits might be keeping the hybrid zone narrow. In the case of Vermivora hybrids, there apparently hasn’t been strong selection against them, so perhaps that's one reason why their hybrid zone has expanded. These two scenarios, narrow and broad hybrid zones, show that evolutionary history, geography and a little serendipity are all important in establishing patterns of hybridization among warblers. But the hybrid story doesn’t end there. At the other end of the Townsend’s Warbler’s range there’s something else going on. Something different and fascinating and worrying.
Townsend’s Warbler and Hermit Warbler (Setophaga occidentalis) diverged within the past 500,000 years. Diverged yes, but differentiated, hardly. By that I mean, although they took on different plumage characteristics in allopatry, their ecology and life history strategies remained almost identical. They’re so similar in fact, that when these species do coexist in the same patch of forest they maintain territories from which they exclude not only their cospecifics but from each other. They sing very similar or identical songs. They share similar breeding phenology and habitat preferences. And they hybridize.
Townsend’s and Hermit Warblers meet in the three distinct hybrid zones in the Olympic and Cascade Mountains of the Pacific northwest. Hybrids are more common than pure Townsend's or Hermits in Washington’s Olympic Mountains, occurring at a frequency of over 60%. In the Washington Cascades about 43% are hybrids and in the Oregon Cascades about 28% are hybrids. Why the difference? And is this difference meaningful in any biological way? It is unlikely that there are sufficiently different selection pressures at each of the three hybrid zones which can account for the differing proportions of hybrids in each respective location. For example there are no apparent differences between the vegetation in territories of Townsend’s, Hermits and hybrids across the three sites. No other ecological differences have been detected either. Hybrids must be rarer the further south you look for another reason.
The greater frequency of hybrids in the Cascades is probably the result of dispersal of Townsend’s Warblers out of the Rocky Mountains. By contrast, the more isolated Olympics, have received an almost total lack of Townsend’s Warbler immigration from the Rockies, thus resulting in fewer hybrids and more genetically pure Hermit Warblers. Genetic data bear out this hypothesis. The influence of an excessive influx of pure Townsend’s Warblers into the Washington Cascades can be seen in the genomes of birds in the contact zone. This is not the case in the Olympics. There are plenty of forested corridors with suitable Douglas-Fir (Pseudotsuga menziesii) habitat for Townsend’s Warblers to disperse between the Rockies and the Washington Cascades. The apparent level of dispersal between the Washington and Oregon Cascade hybrid zones is far lower and lower still is dispersal into the Olympics. At least among the Cascade hybrid zones, the discordance between the frequencies of hybrids in the three zones may also be linked to the age of the contact zone. The Washington Cascades zone has probably been in existence for at least several thousand years. The Oregon zone, is estimated to be far younger.
But the three hybrid zones we see today are not the same ones that have always existed. The present day hybrid zones have shifted south by about 2000 km from the site of initial contact between Townsend’s and Hermit Warblers, somewhere in British Columbia. Indeed Townsend’s Warbler has replaced Hermit Warbler over this vast swath of the Pacific Northwest. The displacement of Hermit Warbler was partly through the usual ecological processes of competitive exclusion, where the behaviourally dominant Townsend’s Warbler ousted the subordinate Hermit Warbler, but also involved a significant amount of hybridization. The hybridization left what researchers have termed a "wake" of Hermit Warbler mtDNA in the Townsend’s warblers that today occupy this zone.
These two species have such broad ecological overlap that they could occupy the exact same habitats if and when they co-occur. Nature however, has a way of sorting communities of organisms so that very often only a single species is able to occupy a particular niche or habitat, to the exclusion of all other similar species. In the case of our two warblers, there is an unevenness in competitive ability with Townsend’s Warbler males being more aggressive than Hermit Warbler males. When these species to come together, Townsend’s Warblers bully and harass Hermit Warblers. Overtime, Townsend’s Warblers outcompete and push out their congener counterparts. The Townsend’s Warbler’s competitive superiority doesn’t end there. Townsend’s Warblers seem to be more successful at attracting mates, of both their own species and Hermit Warblers. It’s a dreary state of affairs for Hermit Warblers when these two species meet.
It has been suggested that the hyper-aggressiveness of Townsend’s Warbler may be a result of its recent evolutionary past. Although fairly large at present, the breeding range of Townsend’s Warblers may have been severely reduced during the last glacial maximum, when it was confined to a relatively small and arid refugia in the continent’s interior. The restricted and stressful conditions there probably led to extreme competition for limited territories in this highly restrictive habitat. In such a scenario, only the most aggressive birds would establish territories and produce offspring, and hyper-aggressiveness would be selected for. The ice age breeding range of Hermit Warbler was more expansive, so perhaps competition would have been less intense in that species.
Aggressiveness in birds is influenced by a number of bio-chemical (and other) factors. Testosterone and other androgens play an important role when it comes to territorial defense. Androgens levels are highest in Townsend’s Warbler males. Hybrids males are next, with significantly higher androgen levels than pure Hermit males. It appears that differences in androgen levels may be inherited. It’s been found that hybrids which are phenotypically more Townsend’s-like have higher androgen levels than hybrids that are more Hermit-like.
In short, Townsend’s Warbler appears to be the better species. Star Trek enthusiasts may think of Townsend’s Warblers as an avian Khan, superior in every way to the feeble and passive Hermit Warblers. Compared to Hermit Warbler, in marginal habitat, Townsend’s Warbler is able to achieve better pairing success and is better able to maintain their territories, even in resource-stressed conditions. Townsend’s usurp male Hermit territories and produce fertile hybrid offspring after mating with Hermit females. Hermit Warbler’s even lose out to hybrids, which have better pairing success and maintain their territories more effectively than pure Hermits. But, Townsend’s warblers still maintain their place at the top; they have larger clutch sizes and greater reproductive success than hybrids.
Contained within the genome of Townsend’s Warblers are haplotypes which signal a rapid demographic expansion following a period of population bottleneck. That bottleneck, a period where the population was low and mating options were limited, probably occurred in the Townsend’s Warbler’s glacial refugium. With the receding of the glaciers the Townsend’s Warbler rapidly expanded, eventually moving into the range of the Hermit Warbler. Townsend’s Warblers that inhabit the Pacific coast rainforests are the product of hybridization, as they carry over 50% Hermit Warbler mtDNA. As coastal Townsend’s Warblers moved south they initiated further hybridization with Hermit Warblers. They also continued to push the Hermits out of previously occupied habitat. The result is that there may very well be a natural extinction in progress. Some researchers estimate that genetically pure Hermit Warblers may have a mere 5000 years left before being completely swamped by the behaviourally dominant Townsend’s Warbler. The remaining birds will look like Townsend’s Warblers, but coast populations will continue to maintain a signature of past introgression with Hermit Warblers.
In Part 3 of Infus'd Betwixt, I'll look hybridization from another angle, how it can lead to speciation in birds.
