Douglas Futuyma, NYSbirds, 21 Nov 2015...
Andrew Rush, NBHC ID-FRONTIERS, 22 Nov 2015...Western Flycatcher: two species or one?
Having enjoyed viewing the "Western" Flycatcher in Central Park today, I thought it might be useful to read the most relevant information on the question of whether or not the "Pacific-slope" and "Cordilleran" Flycatchers are distinct species. Some birders who have seen (or, hopefully, will soon see) this bird may be interested.
First, regarding identification, the account by Lowther in Birds of North America notes that "the commonly heard call, especially from migrant or overwintering birds, is a sharp seet! that is NOT DIAGNOSTIC for either species." [Emphasis added.] And, "identification by call of migrants and vagrants outside the known breeding range is problematic." Lowther discusses discrimination by measurements of flight feathers at some length, and notes that there is much overlap: only about 60% of individuals of KNOWN identity (by breeding location) can be discriminated.
The split of Western Flycatcher (Empidonax difficilis) into two species was recognized by the AOU in 1989, on the basis of a paper by Johnson and Marten in 1988. Since then there has been controversy about their "species'" status, based on phenotypically intermediate birds, especially in interior southern Canada (British Columbia and Alberta), where the ranges of the two forms come into contact. Johnson and Marten reported data from a contact zone in California, but not the Canadian zone.
In 2002, Johnson and Cicero reported that the two forms had different mitochondrial DNA (mtDNA) sequences. I should note that mtDNA is probably the DNA data one should LEAST rely on for discriminating species, because it is not at all representative of most of the genome, for several reasons.
The only subsequent DNA-based study -- and it is a good one-- is by A.C. Rush et al. 2009, Analysis of multilocus DNA reveals hybridization in a contact zone between Empidonax flycatchers, Journal of Avian Biology 40:612-624. They analyzed both mtDNA and chromosomal DNA for 48 specimens, including birds from 8 sites in the Canadian contact zone, plus "pure" Pacific-slope and Cordilleran populations from California and the Rocky Mountains in the U.S. The chromosomal DNA was not gene sequences, but a much cruder kind of data (AFLP), based on a method that reveals single small changes within any of a number of very short DNA sequences scattered throughout the chromosomes.
Their first important finding is that none of the chromosomal DNA markers completely separates Pacific-Slope from Cordilleran Flycatchers, even those distant from the contact zone. The "species" differ only in the percentages of different versions of each marker (like different proportions of people with different eye colors). Second, they found that birds from the contact zone have the same mtDNA as Pacific-slope Flycatchers. However, the chromosomal DNA markers showed a broad band of hybridization: over a span of 400 kilometers, birds show evidence of mixed ancestry.
Their data conform to a scenario in which an ancestral flycatcher was divided into two populations, presumably in the Pacific and Rocky Mountain regions, for a long time. They became somewhat genetically different, mostly by a process of random genetic change known as genetic drift, in which most of the genetic changes have absolutely no effect on the characteristics of the birds (plumage, voice, behavior, ecology). Any two separated populations will undergo this process of random genetic divergence -- all it needs is time. In this scenario, the two populations expanded their ranges northward and came into contact recently, perhaps less than 10,000 years ago, after the most recent glaciation receded. They have been freely interbreeding in southern Canada, and the Pacific-slope mtDNA (the authors suggest) has been spreading into the contact zone faster than the chromosomal DNA because the density of Pacific-slope Flycatcher populations is higher (so they flood the zone) and because they arrive earlier in the spring, so they may get more mates and father more hybrid offspring.
The bottom line: there is extensive hybridization and gene flow between these forms. The authors comment about the AOU decision to split: "We now wonder whether, given the present evidence, the decision to formally split..would have been made." If it were up to me to hand down AOU decisions, these forms would be retired, and we would go back to a single species, Western Flycatcher. I hope the AOU ultimately comes around to that view.
Stony Brok, NY
Farnsworth & Lebbin 2004 (HBW 9):Re: Western Flycatcher: two species or one?
Because I am the primary author on the most recent genetic analysis of these species (mentioned in the post by Douglas Futuyma cited by Peter Post), I thought that I could add a little to this discussion. I recently finished my dissertation research, most of which focused on these two species. While it is true that the two species are admixed in their DNA over a large part of the West, the Pacific-slope populations west of the crest of the Sierra, Cascades, and Coast Ranges (i.e., the Pacific Slope) remain genetically and phenotypically distinct. We know that gene flow from interior populations to the west slope Pacific-slope populations occurs to some extent, but it does not result in widespread genetic mixing like it does on the east slope. So, it is a little more complicated than two species just merging (back) into one. Pacific-slope seems to be merging more into Cordilleran than Cordilleran is merging into Pacific-slope. I’m not sure taxonomists will take this nuance into consideration when deciding what to do with these species, but from an evolutionary perspective, it is interesting. I will have at least a couple of more papers on this out soon.
As soon as you cross the crest of the Pacific Slope to the east side, you encounter mostly genetically intermediate birds with intermediate songs or calls. There is some proportion of admixed birds in populations all the way to the Black Hills and to northern Utah and Colorado. On the other hand, you almost never encounter birds with intermediate songs or calls on the west side and almost no birds are mixed in their DNA…and these are limited to areas like Mt. Shasta in California, which is very close to admixed populations.
So, if you have seen a Cordilleran Flycatcher in southern Colorado, New Mexico, or Arizona, you are probably safe…in terms of listing. If your Cordilleran Flycatcher is from Alberta, Idaho, Montana, Wyoming, or the eastern parts of the coastal states, you have a higher likelihood of having seen an admixed bird.
One last thing in terms of identifying these species. I have not formally analyzed the position notes yet (i.e., ‘pee-o-weet’ and ‘weet-seet’) but it seems that these change in a slightly different way than the songs geographically. I.e., you can encounter birds whose position note is more purely Cordilleran that has a more intermediate song type and a more intermediate genotype.
I hope this is interesting to some of you.