Apologies as this is VERY LONG but I think a significant step forward in talking about the status of the Northwestern Crow. A second short email that follows this is a short summary with advice for birders (who don't want to read this stuff).
In a previous message I mentioned the talk I saw James and Renee Ha (at REI Flagship store on Thursday, February 17, 2011). James presented pretty good mitochodrial DNA evidence for the Northwestern Crow being a distinct species. I recorded this talk for my own use. Those interested can download a copy from my DropBox. It's a 39MB (quite large) 80 minute AAC audio file that will play on iPod/iPhone/iTunes and other audio players. The portion of interest for the distribution of the Northwestern Crow is from 31 minutes to 50 minutes though the whole talk is worth listening to. Click on this link to get a copy (it's just a URL shortener link to the file). http://bit.ly/KP9TYt I've since exchanged emails with James Ha and he told me the paper (Ha, R.R., Walsh, H., Diniz, F.M., Bentzen, P. The biogeography of northamerican crows based on control region mtDNA) is expected to be submitted for publication this summer. I summarize James and Renee Ha's argument below. My comments are in square brackets [ ... ]. Dr James Ha and his father (an animal behaviorist with experience recording sound on sabatical) one summer recorded crow vocalizations at multiple locations around the Salish Sea (i.e. in the Georgia Basin/Puget Trough) to answer the question "is there a statistically significant difference in Pacific-northwest Crows vocalizations in different locations" The locations selected are: Northern/Coastal population (Nanaimo)
Olympic Peninsula population (Sequim State Park)
Whidbey Island population (Lake O'Neil Campgrounds? Cape Disappointment State Park)
Southern/Inland population (the Seattle UW campus) Evening assembly calls were recorded at 5m to 15m [that 3 times range means a potential 5.2dB difference in level depending on range differences or 3x in amplitude] They looked at:
Peak frequency
Amplitude at certain frequencies (5kHz, 6kHz and 7kHz) - 6kHz for the reported results
Call length in milliseconds The question: Is there a difference in any of these measures between these populations of crows. At 95% confidence limits (p=0.05) there is. Look for clusters in the variables on two (or more) dimensions e.g. amplitude at 6kHz versus call length is presented where the difference is largest. The Nanaimo and Seattle UW campus are clearly distinguished. The Nanaimo crows have less amplitude at 6kHz and shorter calls than the Seattle crows. This is what birders hear (lower, hoarser, shorter vocalizations as one travels north in the Salish Sea). The birds at intermediate locations have intermediate amplitude at 6kHz and call length compared to the crows in the north and south.
There is a cline in call parameters. There is no sharp boundary between these "two groups" of crows. So what about the birds in SE Alaska (outside of the Salish Sea region). Is there a difference in calls between them and the Nanaimo crows? They didn't have the money to go up there and do the same measurements there. [That's a shame]. The would have liked to do recordings in Juneau and up the Alaskan panhandle to see if there are changes there. So there is a north-south cline in the vocalizations in the Salish Sea [at least at this resolution of measurement] How is this cline explained? There could be American Crows in Seattle and Northwestern Crows in Nanaimo and hybrid/intergrade Northwestern x American Crows in between. Or there could be American Crows in Seattle and Northwestern Crows in Nanaimo and a discrete mixture of Northwestern Crows and America Crows at locations in between. Using thier current data they couldn't distinguish these [Actually I think you might be able to sort this out by looking at the shape of the clusters but it might be that the difference is too small to distinguish between them. This is like curve fitting two peaks. But with better data this might become clear (if you can reduce the "noise" in the data). How big is the variance in one group compared to the two end points.] Conclusion of the vocalization study: there is a significant difference between the vocalizations of in the north of the Georgia Basin and the south of the Puget Trough and there is a cline from hybridization or overlap of call in between those locations. Hence the need for phylogenetic work. The Burke Museum funded a postdoc to work with Renee Ha initially to answer the question "does the genetics of the crows in the Puget Sound [and Georgia Basin?] differ from those in Alaska". This study expanded (over 10 years or so) into looking at the phylogenetics of crows across the whole of the USA and Canada: including the American, Northwestern and Fish Crow (but excluding the Tamulipas Crow). Using automated sequencing mitochondrial DNA of 121 tissue samples from museum samples across the US [this is what the dead birds you hand in]. They found a phylogenetic tree that grouped all American Crows (those from the east and the west) share a single mtDNA sequence with some single basepair changes together (in a clade). Another grouping (clade) is of "Alaskan Crows" (i.e. from samples from crows in Alaska). Samples from crows on the Olympic Peninsula [and elsewhere around the Salish Sea] are intermediate between these two groups. The Fish Crows appear as another clade. As these Fish Crow is an undisputed species it was included as a "out group" to see if it appeared in the phylogenetic analysis of these mtDNA sequences as a clade to show that these are actually useful differences. In their initial work (reported by Renne Ha at the WOS meeting a few years back) there was a problem: the Fish Crows didn't come out as a distinct clade (species) which was a problem. This was from an insufficient number of samples from Fish Crow. When the number of Fish Crow samples was increased and the phylogentic tree was generated (in 2011) the Fish Crow appears species. They also include a cline in the tarsus length of crows sampled along in the Georgia Basin/Puget Sound: tarsus length increases as the move to the south.
James and Renee Ha conclude that the Northwestern and American Crow seem to be separate species in their endpoints [i.e. in Alaska and outside the Pacific Northwest] with as good standing as the Fish Crow.
The Northwestern clade split off from the American clade sometime in the Pleistocene era [I asked Renee later for an approximate date and, if I remember correctly she said 100,000 before present]. The crows around the Salish Sea are intermediate between the two clades. This could be because they American Crows are invading an initial Northwestern crow population in the Puget Trough and generating a hybrid zone. Or possibly the two species could be separating out in this region (given the north to south cline) [e.g. sympatrically if they breed assortively perhaps based on call or behavior? or parapatrically or peripatrically by separation into different habitats or niches?]. They currently can't answer this question. There are plenty of problems: The paper isn't published so it hasn't been peer reviewed yet. Perhaps the referees will find issues with the mtDNA work and the phylogenetics. I suspect not. There is no nuclear DNA (nDNA) phylogenetics work. Perhaps this has been done in the last year or perhaps they've decided to publish without nDNA work. This did worry Renee Ha last time I talked to her. They didn't do any vocalization work in Alaska or south of Seattle or more inland of Seattle. Perhaps there is a small cline continuing say along I90 to the east (after all Eastern birders often comment on how the local birds sound different) and I5 to the south. This is the sort of work an amateur bird sound recordist could do (in fact the data probably already exists in libraries if one could get it in raw form). This answers some of Eugene Hunns questions: On Jun 19, 2012, at 2:33 PM, Eugene Hunn wrote: > In the case of the “Northwestern Crow,” while crows from the Kenai Peninsula to the outer Olympic Peninsula may vary little, there is no clear boundary between those crows and the mass of crows throughout western Washington. There is a cline down the Salish Sea basin but the lack of a boundary doesn't mean the end points don't exist.
> The ecological preference for beach foraging could simply be due to where the available crow foods are to be found along the densely forested outer coasts. If there are really two crow species in western Washington there should (or perhaps “must”) be a zone of contact, of sympatry, where one could demonstrate the coexistence of two distinct species populations. The "zone of contact" is all through the Georgia Basin/Puget Trough. Of course this also means (as Eugene and others have said) is that there are no "pure" Northwestern Crows in Western Washington or perhaps even BC. Perhaps the Northwestern Crow should really be called the "Alaskan Crow"? > A few mitochondrial DNA differences proves nothing about how the crows throughout western Washington interact, with each other and with the environment. That's true but phylogenetics using mtDNA and nDNA and the phylogenetics are how species are defined today. You may regret that but it is how most biologists work today.
This is not a new issue but one that appeared when cladistics started to be used to generate trees of fossils algorithmically from descriptions features rather than having a taxonomist make his educated best guess. The book "In Search of Deep Time" outlines the history of cladistics and documents this fight (and it was a fight starting in the 1970s) rather well with results like "there are no fish" (i.e. "the fish" do not form a clade).
Phylogenetics just uses similar techniques to get to the underlying order carried in the DNA and seems to ruffle the same feathers. > Has anyone done a comprehensive study of vocal patterns (of breeding birds) throughout the region. If so, can they demonstrate that there are two sympatric vocal systems in play? Or, would such a study show a clinal pattern of variation linking the extreme outer coastal call types with the calls further south along the coast and/or inland, as with the “Western Flycatchers”? Again just because there is a cline doesn't mean there aren't distinct species at the endpoints (ring species are the extreme example of this and of course drive "tickers" up the wall). The counter example is our other favorite hybrid around Western Washington: the Glaucous-winged x Western Gull. Huge hybridization zone with perhaps 40% of the birds in the southern Puget Trough being intergrades (the number varies depending on where you are sampling). They breed assortivley it seems. Does that make the Glaucous-winged Gull and Western Gull not true species but just one large Larid species with two different subspecies? Thanks for reading this all the way through :-)
