• Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.

    Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
ZEISS DTI thermal imaging cameras. For more discoveries at night, and during the day.

Crossbills (1 Viewer)

In relation to this point: "Kroodsma's research is also instrumental in demonstrating that tyrant-flycatchers and finches (among other birds) have very different methods of song development. Tyrant-flycatchers have innate songs, whereas finches learn their songs. In other words, this is an apples to oranges comparison. I think a more fruitful (so to speak) approach would be to compare crossbill differentiation with tit/chickadee differentiation, although that is a developing field as well. Speaking of apples and oranges, it should go without saying that crossbill and human speciation are not comparable."

I do appreciate all that. However, with finches, there is well documented geographical variation in plumage which you can compare to vocal variation. For crossbills, there are studies showing some average (i.e. not diagnosable) differences in bill size or shape but not much more.

Conventional wisdom is that oscines learn their songs but suboscines have innate songs. However, I don't know if anyone has ever suggested that if you play warbler songs at a crow pullus, it will end up singing like a warbler. Also, some oscine calls may be more innate in nature than their more complex songs. Even with closely related species, bill size and shape (note: relevant to crossbills) and gape muscle strength constrain vocal ability. As a result, I do think that studies of this nature would be a very helpful step in resolving what is going on with these difficult birds.
 
thomasdonegan said:
for crossbills, other diagnosable differences between populations are so elusive
Click to expand...

Call differences are, quite frankly, also subjective and elusive.

Only a handful of crossbill researchers claim they can distinguish call types, and only using sonagrams. I never seen a blind test when to two humans independently validated call types from random sets of recordings.

I would expect to see objective definitions what differs in call types and where the borders cross into different call types. I would also expect that when some objective characteristics of the sound (eg. frequency plus change) were plotted, then calls of randomly chosen set of individuals would fall into clear clusters corresponding to call types.

And yes, finches are famous for learning sounds - also to learn and change songs during life. I would expect experiments when captive crossbills were housed for several years with different call types, or in presence of recordings, and they kept their calls.
 
I think it's been said before on a similar thread, but it really doesn't matter how easily humans can identify a species, it only matters how well the bird itself can

Although I agree that much more work needs to be done before I would feel comfortable labeling the different types as different species.

Also I am not aware of any subspecies which have non fixed distributions during the breeding season. Of course, in winter there is plenty of mixing.
 
lewis20126 said:
Is this necessarily the case with all animals? - certainly not the case with the use of the "subspecies" concept in plants.

cheers, alan
Click to expand...

It certainly is the case!! I believe you are thinking of plant varieties, which do not (necessarily) have a geographic component.

To illustrate with a popular example, Cannabis sativa has several subspecies, depending on the authority. C. s. sativa is the northern subspecies is the northern form, preferred for hemp rope, while C. s. indica is the southern subspecies... cherished for other purposes.

C. s. sativa can be further subdivided into two varieties: C. s. sativa var. sativa may grow tall and has relatively low resin levels. C. s. sativa var. spontanea is shorter and bushier. I'll not get into all the varieties of indica! But these varieties, while having a strong genetic component, are not necessarily geographically divergent, although each may fare better in specific microhabitats.

But the point is that the botanical community has, for over a century, had a system for genetic but nongeographical variation - perhaps ornithologists should become more comfortable with this as well.

I feel that there is a good case for some biological species within the Red Crossbill complex, but at our current level of understanding and for those "taxa" that may not yet have achieved species definition, I feel that we are dealing with different "varieties" of crossbills, not subspecies.
 
I do appreciate all that. However, with finches, there is well documented geographical variation in plumage which you can compare to vocal variation. For crossbills, there are studies showing some average (i.e. not diagnosable) differences in bill size or shape but not much more.
Click to expand...

And yes, finches are famous for learning sounds - also to learn and change songs during life. I would expect experiments when captive crossbills were housed for several years with different call types, or in presence of recordings, and they kept their calls.
Click to expand...

??? I'm confused by these comments. You guys do realize that crossbills ARE finches?

I would have the opposite expectation in the proposed experiment - that crossbills learn their songs just like other finches. However, your result would be more interesting, and would lend evidence that perhaps call note specificity had developed for the benefit of crossbills to tell the difference between good and bad potential mates.

Which brings me around again to the point I was trying to make above, but apparently failed at: species are defined by who breeds with whom, not who sounds like whom. Frankly, I wouldn't care what sounds crossbills make, except that there seems to be a correlation between exclusive breeding populations and variability in calls. But the populations are important here, and whether they tell each other apart by plumage, call type, smell, or ESP, it shouldn't matter so long as the differentiation is real.

And while I'm at it, I think its an exaggeration to suggest that the call types can only be distinguished by an elite few researchers. I can only speak for the American birds, but upslurs vs. downslurs vs. even notes are fairly easy to distinguish. The highness vs. lowness seems a bit less easy to my untrained ears, but not impossible. And some, such as the Type 4, are quite distinctive. Here in the eastern U.S., I've heard Types 1, 2, and 4, and these are distinguishable. And I'm no Loxiologist by any means...
 
There was a typo on one of the emails above which seems to have been propagated. For "finch" read "chickadee" or "tit".

I suppose my point is, if voice is learned, there are no genetic differences and no morphological differences between populations, then how are humans deciding which populations are or are not interbreeding with which other populations? And how do we not know that any observed non-interbreeding is maintained by preferences for "social groupings" rather than incapacity to interbreed and produce fertile offspring? (Maybe this discussion also then goes on to species debates and splitters vs. lumpers which there is plenty of material on already elsewhere.)

I would not want to respond on the other points, as I agree with the last posting. Some of the papers concerning the geographical variation in different calls of these birds based on studies of the European forms are also quite convincing.
 
Mysticete said:
I think it's been said before on a similar thread, but it really doesn't matter how easily humans can identify a species, it only matters how well the bird itself can

Although I agree that much more work needs to be done before I would feel comfortable labeling the different types as different species.

Also I am not aware of any subspecies which have non fixed distributions during the breeding season. Of course, in winter there is plenty of mixing.
Click to expand...
Personally I think that the only question in the Crossbill species debate is "where exactly do we draw the line when defining two separate species?".

There is one very good example of a species that has 'sub groups' with clearly different vocalisations, and which can now be found co-habiting the same areas throughout the year - often with very limited interbreeding between the different 'groups'.
I realise that it is generally considered 'politically incorrect' to even suggest that the same subspecies/race concept that we apply to other animals can be applied to humans (I think partly because the term 'subspecies' can be taken as an implication that one or more races are 'below' others), but I do think that comparisons can be made.

So the crossbill scenario (as I understand it) is that there are groups that can be separated by different vocalisations but (in most cases) have been found to be practically indistinguishable genetically, and physical differences are so slight that they are of no practical use. The idea that these should be treated as different species is apparently supported by the fact that they can be found breeding in the same areas with limited interbreeding between groups with different vocalisations.

The human comparison: There are groups that can be separated by vocalisations (I can recognise several languages even though I may not understand them, and there are easily distinguished regional differences even within a country). Genetic differences are apparently slight, but physical differences can sometimes be obvious - although not necessarily linked to differences in vocalisation. Finally (and this may be disputed by some), although humans from all parts of the world can be found in the same cities, there is relatively little interbreeding between groups with different vocalisations (languages). I doubt that anyone can disagree that wherever there are immigrants into a country, communities of people with the language and culture become established. There is obviously some degree of mixing between people from different countries, but I think that it is fair to say that this is clearly restricted if one person does not speak the language understood by the other.

Returning to the crossbills - perhaps there is mixing, and interbreeding, between different 'vocal groups'. If this were the case, surely it would be most likely when a bird from one group learnt the calls/song used by another group and then started using these when associating with birds that use these different calls? Exactly the same as humans learning, and using, another language to allow them to communicate and mix with another group.

It may well be that crossbills are best treated as a number of different species, personally though I'm yet to be convinced, and until a lot more research is available I think that it is better to be conservative with regard to the number of species (I'm not even convinced that Scottish Crossbill is really a valid species!;)).
 
I suppose my point is, if voice is learned, there are no genetic differences and no morphological differences between populations, then how are humans deciding which populations are or are not interbreeding with which other populations?
Click to expand...

So the crossbill scenario (as I understand it) is that there are groups that can be separated by different vocalisations but (in most cases) have been found to be practically indistinguishable genetically, and physical differences are so slight that they are of no practical use.
Click to expand...

Please read some of the articles listed above, or here:

http://www.birdforum.net/showthread.php?t=135285&highlight=crossbills

There are morphological differences. Granted, these "ornithologists' differences" and not "birders' differences," but they are statistically significant and consistently found. Otherwise... why would this body of research even exist?

I'm sure I'd agree that if the ONLY thing different about these birds were their voices, it would be a fairly minor distinction, but that simply isn't true.
 
I have read all these papers.

Let's look at the Condor "new species" description from last year.

Diagnosis: "Very similar to other Red Crossbills in North America but larger (body mass and bill depth, but upper mandible relatively short) on average than other Red Crossbills currently found commonly north of Mexico (Table 1) and with distinctive vocalizations (see below; Smith and Benkman 2007). As in other Red Crossbills, males are reddish in body coloration whereas females are greenish-gray and smaller than males (Table 1; see below)."

Conclusion:
Voice - different, but see above discussion.
Plumage - the same.
Biometrics - "average differences" - i.e. not diagnosable. If you look at the table of biometric data in the paper, you can see some highly unimpressive divergence in biometrics. Data on range of measurements for males for the variables mentioned in the species diagnosis section is below:

Mass:
Type 2 28.8–38.6
Type 5 28.6–38.6
New species 29.2–39.4

Bill depth:
Type 2: 8.97–10.38
Type 5: 8.77–10.14
New species: 9.05–10.56

Bill length from nostril:
Type 2: 13.79–18.87
Type 5: 13.95–16.87
New species: 13.40–17.30

These are not diagnostic differences. With enough data, one might be able to generate a positive t-test, showing statistically significant average differences, but based on these sorts of data you would expect at least 80-90% of individuals not to be capable of being identified based on biometrics.

Which takes me back to the original posting. If voice is all there is to go on in terms of real diagnosable differences, and this can shift, who is to say this is not a learned character?

The paper states: "We found one instance of a female of call type 2 shifting its call to match that of its mate, a South Hills Crossbill, in successive years, but because this shift occurred only in the year following this pair’s successful breeding such call shifts should affect our estimates of the frequency of hybridization only minimally (Keenan and Benkman 2008; see also Summers et al. 2007)." That is one possible interpretation but one could see people arguing for others in terms of delimiting species based on (i) actual observed hybridisation between "species"; and (ii) learning by one "species" of the other "species" voice.

Going back again to the original posting, I do think this and other papers make up a fantastic piece of research but it seems a bit premature to be describing sympatric "species" without diagnosable differences except for in characters that may be learned. AOU's NACC seemed to have a similar view.
 
It's A Crossbill Jim, but not as we know it.

I can do all the UK occuring loxia calls by ear. Do I get a prize ?;) Sonagrams show the finer details and variations between "types" which may partly be responsible in taking the splitting madness too far. I prefer to use my ear first, then sonagram to catalogue the call.

Scottish Crossbill does exist, it's just that it might not be the one you think it is.....or there might be more than one !

Some pretty earth shattering stuff re- Scottish Crossbill calls in the pipeline. Keep tuned to Loxia Fantastica for updates.
 
I am a tad surprised at the discussion on learned vocalizations. Fringillids are well known to learn song, not only that they learn calls which is highly unusual in passerines. On the other hand they do some weird stuff like call matching between pairs, and seem to sort each other out into single call type flocks based on these learned calls. It is complex, and this is why crossbills have not been a shoo-in for consideration as different species. The calls are learned but at the same time they are used for identification/segregation of groups, this tendency to mix with those of your own call type may be culturally facilitating specialization. The problem is that you need only a few to mix with the wrong gang in populations to keep gene flow alive.
To get a handle on the crossbill issue it would be good to study some other related finches that show geographic variation in call type. Three in North America are Pine Grosbeak, Evening Grosbeak, and Purple Finch....I wouldn't be surprised if Pine Siskin also fits the bill (ahem). The three mentioned above have identifiable call type differences but they stay within their own separate geographic areas, at least most of the time. It would be interesting to do molecular as well as vocal playback work on these simpler systems to understand fully the role of call type in differentiation. Then these results can be applied to better understand or resolve which populations or call types in the Red Crossbill could in fact have reached species thresholds, versus some NEW type of within species variation that is yet unnamed (subspecies does not work in this case as noted above). Alvaro.
 
Evening Grosbeak

mb1848 said:
Per Evening Grosbeak, vocalization and morphology matching (if not molecular) Earbirding has a timely post:
http://earbirding.com/blog/
Click to expand...
Re the potential reinstatement of Grinnell's five subspecies, Gillihan & Byers 2001 (BNA Online) mentions six:

  • vespertinus
  • brooksi (incl californicus, warreni)
  • montanus (incl mexicanus)
Which five did Grinnell recognise?

Richard
 
Biometrics - "average differences" - i.e. not diagnosable.
Click to expand...

All the quoted biometrics are statistically significant, which (to a scientist) is what defines a "real" difference. If your argument is with the overlap in morphology of a minority of individuals in a museum tray, than I guess the issue is with statistics rather than biology - and frankly I'm too tired to argue about that! I'll end by maintaining 1) that overlapping morphology is a problem with several plant species (Solidago, and other composites, especially), some flies, and probably many others. And that is without getting into Hugh Patterson's "Recognition Species" concept. In a taxa with such complex behavior as birds, I'd be amazed if they are exempt from this sort of crypsis. and 2) I still maintain that breeding behavior, not morphology, is a better basis for species, (or "Type") definition.

To get a handle on the crossbill issue it would be good to study some other related finches that show geographic variation in call type. Three in North America are Pine Grosbeak, Evening Grosbeak, and Purple Finch
Click to expand...

Very good idea. To be more challenging, I'd add redpolls to the list. However, I'd caution that one key aspect of crossbill biology and the assumed driver of population differentiation is a specialized and ephemeral food source. All of our other finches seem to be moderately general feeders (okay, compared to crossbills anyway). But they would be very good models for vocalization studies.

I don't suppose there are any good data regarding Hispaniolan vs. Two-barred Crossbill vocalizations? While those have more geographic isolation, I'd wonder if vocalizations vary in a similar way to Red Crossbills.
 
Kirk Roth said:
All the quoted biometrics are statistically significant, which (to a scientist) is what defines a "real" difference. If your argument is with the overlap in morphology of a minority of individuals in a museum tray, than I guess the issue is with statistics rather than biology - and frankly I'm too tired to argue about that! I'll end by maintaining 1) that overlapping morphology is a problem with several plant species (Solidago, and other composites, especially), some flies, and probably many others. And that is without getting into Hugh Patterson's "Recognition Species" concept. In a taxa with such complex behavior as birds, I'd be amazed if they are exempt from this sort of crypsis. and 2) I still maintain that breeding behavior, not morphology, is a better basis for species, (or "Type") definition.
Click to expand...

In all fairness, the data presented in the above post included some examples, such as :
Bill depth:
Type 2: 8.97–10.38
Type 5: 8.77–10.14
New species: 9.05–10.56
Click to expand...

With these differences, it would be more than half of the birds that would not be diagnosable by bill depth (I think). In the monograph on subspecies that we had a long thread about earlier, one or more of the authors argued that if two populations differ by some few percent in one average measure, then you just had to increase the sample size enough and you would end up with a significant statistical difference that would not be biologically meaningful. As I understand it, that is the cause for the statistics called diagnisability which basically measures how large a percentage of the population cannot be id'ed to that population by whatever measure is used.

However, I do agree that the most important thing is to look at whether the birds themselves think that another bird belongs to the same species or not; the problem is to be able to ask the birds that question and be able to understand the answer ;)

Niels
 
Kirk Roth said:
All the quoted biometrics are statistically significant, which (to a scientist) is what defines a "real" difference. If your argument is with the overlap in morphology of a minority of individuals in a museum tray, than I guess the issue is with statistics rather than biology - and frankly I'm too tired to argue about that! I'll end by maintaining 1) that overlapping morphology is a problem with several plant species (Solidago, and other composites, especially), some flies, and probably many others. And that is without getting into Hugh Patterson's "Recognition Species" concept. In a taxa with such complex behavior as birds, I'd be amazed if they are exempt from this sort of crypsis. and 2) I still maintain that breeding behavior, not morphology, is a better basis for species, (or "Type") definition.
Click to expand...

Sure but with overlap of about 80-90% in all 3 measurements it becomes useless in everything but the extremes. Under what theory are measurements with an overlap like that 'statistically significant' to the level that they are an argument for different species? It's fine when there are other strong arguments (like in the tapaculos Thomas Donegan has worked with) but if I understand this right there aren't yet for the crossbills. The voices are perhaps learned in them and the describers mention that they have observed voice shifts. If this is the case the breeding behavior can change. A crossbill with medium measurements could be one 'species' in one years. Shift voice and be another 'species' in another year. The genetic difference they reported in the description of the new crossbill is not impressive. Perhaps the South Hills Crossbill is a species but to my eyes evidence that can't be questioned is presently missing.

This is about birds and the established requirements for taxonomy of other groups are often different and this make them irrelevant to the discussion. The possibility of asexual reproduction in plants alone make any direct comparison with birds impossible.
 
Kirk Roth said:
All the quoted biometrics are statistically significant, which (to a scientist) is what defines a "real" difference. If your argument is with the overlap in morphology of a minority of individuals in a museum tray, than I guess the issue is with statistics rather than biology - and frankly I'm too tired to argue about that!
Click to expand...

I agree that science is all about measuring and quantifying differences, but I am not a fan of describing Crossbill species solely on "mean" bill depths, due to the massive variation or "range" of depths that can be present - birds in the overlap could potentially be of several types. Certainly here in Scotland the "overlap" is not restricted to a minority of specimens, quite the opposte ! In fact the overlap can be pretty extensive both sides of the mean which is what causes the problems with Common, Scottish and Parrot Crossbill. If you use calls to diagnose 'speciation' then you have some Parrot Crossbills with minimum bill depths of 11.9mm, well within the range of Scottish Crossbill. If you hadn't used a call, then it would probably be Scottish on historical and published biometrics.

My position is that data sets and slide rules are not the best components of Crossbill 'speciation' (if they are even species). Let's call it 'classification' - that would be cultural aspects (calls), breeding behaviour and ecological factors (where they are found, what they feed on etc) with biometric data factored in, though I don't think the 'minimum bill depth is a brilliant measurement of bill 'fitness' from my experience in Scotland.

Another thing - a Crossbill 'type' may be specialized to feed on a particular cone/seed, but that doesn't mean it can't survive on others....and whi ch is why a given population has variation in the first place.

THIS http://pinemuncher.blogspot.com/2009/09/thats-life.htmldiscussed the 'Scottish problem' and natural variation in that population a year ago with similar content to what is being discussed on here.
 

Similar threads

J
Messy Taxonomy
Replies
15
Views
3K
James Lowther
J
wolfbirder
Southern Israel - Key Site Information
Replies
12
Views
3K
frimp
F
wolfbirder
Category:Israel
Replies
0
Views
2K
wolfbirder
wolfbirder
wolfbirder
Trying to sweep up in Southern Israel 21-26 February 2020
2 3 4
Replies
78
Views
13K
wolfbirder
wolfbirder
Richard Klim
Birdwatch - Listcheck
Replies
2
Views
2K
Mysticete
Mysticete

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top