Redpolls (1 Viewer)

Say they were species, how would they have originated? Say that the lack of genetic differentiation is indeed due to there not being enough time for sorting, that implies that what caused the redpolls to speciate occurred recently. What would that event be?

Allopatry obtained during the last glaciation is not a good candidate. Although still quite a long period, perhaps a lack of structuring can be related to a large effective population size of both redpolls. But than there is still another problem: Lesser must have comprised two allopatric populations, one restricted to the Britisch Isles and the other to the Alps. So if the last glaciation caused the origin of Greater and Lesser that would mean the Lesser evolved twice independently (note that this problem also applies to the remaining redpolls). So they must be older.

So what was it? The last interglacial period? Wouldn't you expect Greater and Lesser to be less isolated at that time, just as they are in the current interglacial? They would already have to behave like species at that time were they not to mix. So they must still be older. The glaciation before that? Perhaps then there was a single Lesser population? But then you are dealing with divergence time that you would expect to be reflected by the genetic data right?

Any other ideas for a speciation scenario?

Ben, some of your questions probably are valid even if these two are only subspecies - how much exchange is there between British and continental Lesser?

Niels

I would say that if you apply the PSC or ESC as seems what Knox et al 2001 like better, you have to make the case that at least one of the two has a monophyletic origin (actually both but let's be lenient) right? The authors seem to say the two look/behave different so it must be o.k. with monophyly. However, it seems difficult to make that case (which would be in line with that they are not species). Sure you have to look at other data than just genetics, but the interpretation must be comprehensive. So than you have the BSC left (which I like better as more specifically focuses on the question why they would behave in a way that restricts gene flow).

Stoddart 2011

Another redpoll article...

• Stoddart 2011. Identification: Redpolls: Photo Guide. Birdwatch 234: 41-46. [18 photos.]

ABA Blog

Andy Boyce, The ABA Blog, 23 Jan 2013: Open Mic: Hoary Redpoll; A False Dichotomy?

[See also: Back to basics: Redpolls.]

Richard Klim said:

Andy Boyce, The ABA Blog, 23 Jan 2013: Open Mic: Hoary Redpoll; A False Dichotomy?

[See also: Back to basics: Redpolls.]

Click to expand...

I had a look at both of these. Interesting and quite convincing to a simple lad like me.

One thing - if we drop to one species of redpoll the word "Common" is redundant, leaving us with just "Redpoll" as the English name.

John

Farnboro John said:

I had a look at both of these. Interesting and quite convincing to a simple lad like me.

One thing - if we drop to one species of redpoll the word "Common" is redundant, leaving us with just "Redpoll" as the English name.

John

Click to expand...

"So, what am I trying to say with all this? I guess the take home message is that if you believe in the current scientific evidence, you have to acknowledge that we have no idea what or if a Hoary Redpoll is. If you don’t believe in the scientific literature but you do believe in the power of statistics, then you have to acknowledge that even if you see a pale redpoll in Colorado, there is no reason to believe that it is a Hoary. In fact, the statistics tell us that it is almost certainly NOT a Hoary Redpoll."

I assume this is not true for non vagrants however. A big white thing in the Arctic region is a Hoary if we decide that that is what they are, based on morphology alone. There may be loads of intermediates and they may be all one "species", but we can give names to identifiable forms whatever the genetics. In redpolls, many can be "identified" to accord with these field identifiable forms, even if there is no (as yet observed) genetic difference.

And don't tell me that big white thing I saw in Suffolk wasn't a big white thing from the Arctic because it was more likely a variant Lesser. Bayesian stats? pah.

cheers, alan

“If it looks like what we call Hoary (or Coues’ Arctic) Redpoll, it probably is not because what we do not call Hoary Redpoll may look the same (which we'd acknowledge if it wasn't for the fact that we call these birds Hoary (or Coues’ Arctic) Redpoll) etc. etc.”

Lumping them all as Redpoll is wholly uninspiring, even if it is correct. I’d still want to know where the birds might be coming from.
Moreover, hornemanni may not be different genetically, but at least there appears to be no gradual change into something else [not even on Iceland].

Yeah I am not sure I quite understand or believe the statistics part of the argument, although I am skeptical on whether we need 2 or 3 species of Redpoll.

Also, +1 to Farnboro John...if we lump it should be just called Redpoll...I am not aware of anything not in the Redpoll complex that shares the name Redpoll, so I can't imagine any form of confusion involved in simplifying it.

I prefer to look at morphology too than simply taking genetics as being the distinguishing marker, as we know it can take centuries for genetic divergence to occur, even when assortative breeding occurs (as with Arctic/Common Redpoll).

People may ridicule me, but this is why I also keep a UK400 list (morphology list for me), it is not about obtaining a bigger list but about recording birds that appear different (redpolls, geese etc). Personally, i think it would be a backward step to lump them all under Redpoll.

Is there any evidence for assortative mating in redpolls?

Assortative mating between flammea and cabaret found by Lifjeld & Bjerke (1996).
http://www.avibirds.com/pdf/B/Barmsijs1.pdf

Here's a very short summary of Molau 1985: Gråsiskkomplexet i Sverige. Vår Fågelvärld 44(1):5-20. The study was mostly based on trapping the birds at Torne lappmark, northern Sweden during the breeding time. About 1100 birds were ringed during 13 years of study. The author found a distinct correlation between plumage characters and biometrics, many plumage characters to distinguish between Common and Arctic Redpolls, differences in habitat and diet, differences in movements and their timing and differences in vocalisations. He trapped only about six intermediate looking individuals. He also commented that the species are easier to distinguish during the summer than in winter. He never saw a mixed pair.

The author also kept 10 Arctic and 6 Common Redpolls in cages over moult. He noted that a couple of young, difficult looking Arctics changed to paler, easy to identify birds. Different sexes of Arctic and Common were not particularly interested to each other when in the same cage, but when a male and female of the same species were kept together, they quickly started to behave as paired. When four Arctic and two Common were released together, they separated to two pure species flocks.

If somebody can find the paper and read Swedish (the English summary is quite short), it is worth reading.

Lifjeld & Bjerke 1996

mb1848 said:

Assortative mating between flammea and cabaret found by Lifjeld & Bjerke (1996).

Click to expand...

This thread is now rather long and convoluted...

See post #31: Lifjeld & Bjerke 1996.

[This was the basis of BOURC's BSC split of cabaret. Parkin & Knox 2010 acknowledge that "There are anecdotal reports of weakness in this analysis" [by the principal author, no less!!], but nevertheless conclude that "no refutation has been published, and we retain C. cabaret as a third species of redpoll".]

Molau’s article sounds interesting!
Now someone should try to find out if the situation in North America is similar to that in Sweden, or if there is a gradual transition instead.

I don’t think that assortive mating in a bird group that is very good at evolving local dialects and subtle differences (cf. Chaffinch, Bullfinch, Crossbill) is very informative about speciation...

From Richard: ABA Blog
________________________________________
Andy Boyce, The ABA Blog, 23 Jan 2013: Open Mic: Hoary Redpoll; A False Dichotomy?


Andy Boyce the blog entry author responded to criticism on Montana birding listserv.
“John and Mob,

I think a lot of people misunderstand the probability argument I am making, and that it is totally independent of the genetic argument. The point is this; We have a pair of species (I don't buy this, but for the sake of argument we will make that a given) where pale individuals cannot be identified with 100% certainty. I am being generous and saying that 99% of pale Redpolls are Hoaries (over the species entire range). That means that 1% of pale Redpolls are Commons. Because Common Redpoll is up to 1000x more numerous in Montana, it means that here, there are many more pale Common Redpolls than we would expect. I miscalculated in the original blog post, but I re-did the math and with the numbers presented above, it comes out to 8% of pale Redpolls (that is Redpolls that appear to have Hoary characteristics) are actually Hoaries, while the remaining 92% of pale Redpolls are pale Commons. Please understand that this is a hypothetical scenario with numbers drawn from estimates, but this concept applies anytime there is a possibility of misidentifying a rare species by mistaking it for a common one, or in cases where the plumage/morphology of two species overlap. I am NOT saying that there are NO Hoary Redpolls in MT, I am simply saying that it if we accept that there are far more Common Redpolls than Hoaries in MT, and that they overlap in their traits, there may be far fewer Hoaries than we think. This is a purely statistical argument, and I know that it rubs some folks the wrong way, but that doesn't make it invalid.

If you like math, check out this wikipedia page for a better explanation of what I'm doing:

http://en.wikipedia.org/wiki/Bayes's_theorem

Happy Redpoll'ing

Andy Boyce
Missoula, MT”

Dan,

Because Hoary Redpolls (as now described) breed much farther north than Common Redpolls, regardless of anything else, Common Redpoll is far more expected to show up in Montana than Hoary Redpoll. Let me put it this way..

Let's say we are in a museum with representative Redpoll specimens collected uniformly from across the worldwide range of Redpolls. In such a situation, we expect to correctly ID a Hoary Redpoll based on paleness 99% of the time. However, when we are at the southern edge of their distribution, i.e. Montana or Colorado, we expect the southern species (or form) to be far more common. Suddenly, that 1% of pale Redpolls that are in fact Common Redpoll become a much larger proportion of the population because Commons are disproportionallu abundant in this location. We can even flip this scenario. Let's say 99% of dark Redpolls are Common, and 1% are Hoary, overall. If we travel way up to the shrubby tundra in the high arctic, and we see a dark Redpoll amongst a ton of pale birds, the probability that it is one of the 1% (dark Hoaries) is in fact much higher than 1% because of the location we are in and the fact that Hoary Redpoll is far more numerous than Common at that location.

Hope that helps.

Andy”

I like this Maryland/DC bird committee skins workshop report it has a lot of photos of dead redpolls:
http://www.mdbirds.org/mddcrc/pdf/rcskins2007.pdf .

In other words...

• In Arctic tundra, pale redpolls are the common form of the locally common 'species', and are therefore almost certainly Acanthis hornemanni.
  
  
• But further south, pale redpolls could quite realistically be either the common form of the locally rare 'species' (A hornemanni) or the rare form of the locally common 'species' (A flammea).

Richard Klim said:

In other words...

• In Arctic tundra, pale redpolls are the common form of the locally common species, and are therefore almost certainly Acanthis hornemanni.
  
• But further south, pale redpolls could quite realistically be either the common form of the locally rare species (A hornemanni) or the rare form of the locally common species (A flammea).

Click to expand...

Thanks, Richard. It's precisely this concept that many field ornthologists keep in mind when studying or censusing bird taxa on their breeding grounds (eg Pierre Yésou when researching the high-latitude large white-headed gulls, or Evgeniy Panov when researching shrikes and wheatears in remote areas of Russian Asia). I would hypothesise it applies to many other groups, such as the yellow/citrine and the white wagtail complexes, the rosefinches, the hierofalcons, the peregrine complex and the crossbills. It's quite difficult to convince birders that a vagrant doesn't have to be a clearly identifiable segment of a population, but that statistically the odds may favour it being a non-representative individual or from an intermediate population...:eek!:
MJB

Statistics are a great way to think about the probabilities of rare birds occurring, but the numbers used in the formula have to be correct.

For one thing, Andy's numbers depend on there being complete (even if rare) overlap between the 'species'. If birds at the extremes are 100% identifiable, then there is 0% chance of misidentification with such birds. Thus if Common Redpolls never have a combination of white rump, limited, fine streaks on the flanks, no streaks on the undertail coverts, and a short bill, then a redpoll with those characters is certainly a Hoary, wherever it is.

Further, most Hoaries in areas where they are rare (and perhaps even where they are common) are probably misidentified as Commons; I suspect only the most obvious adult males are called Hoaries by most birders. This would lead us to underestimate the ratio of Hoaries: Common, and overestimate the number of Hoary-like birds which are actually Common.

alytothrix said:

If birds at the extremes are 100% identifiable, then there is 0% chance of misidentification with such birds. Thus if Common Redpolls never have a combination of white rump, limited, fine streaks on the flanks, no streaks on the undertail coverts, and a short bill, then a redpoll with those characters is certainly a Hoary, wherever it is.

Click to expand...

Yes, agreed, where a supposedly diagnostic suite of characters has been confirmed. But many casual reports of out-of-range Hoary/Arctic are perhaps based only upon overall paleness, which is arguably unsafe.