Subspecies of Great Gray Owl proposed based on genetic "measurements." (1 Viewer)

[jmorlan]

Range-wide genetic differentiation among North American great gray owls (Strix nebulosa) reveals a distinct lineage restricted to the Sierra Nevada, California.

Molecular Phylogenetics and Evolution 56 (2010) 212–221

Abstract: http://www.ncbi.nlm.nih.gov/pubmed/20193768

Proposes Strix nebulosa yosemitensis.

"The American Ornithologists Union (AOU) suggests that “subspecies should represent geographically discrete breeding populations that are diagnosable from other populations on the basis of plumage and/or measurements, but are not yet reproductively isolated” (AOU, 2010). Our data indicate that the Sierra Nevada population of great gray owls is clearly diagnosable with either mitochondrial or microsatellite loci, and therefore is effectively an independent, isolated lineage. Thus, our data support subspecies status for the Sierra Nevada great gray owl population under the AOU criterion. "

I think it's a stretch to claim that genetic data alone constitutes either a plumage or a "measurement." The traditional intended meaning of measurement is a physical measurement of the bird itself, not a genetic measurement. The AOU definition discusses examples where populations differ phenotypically but not genetically and states that such subspecies may be acceptable. It says nothing about the reverse which appears to be the case here.

 

[screaming piha]

Why not? If a population looked the same as others but could not reproduce it would be a cryptic species, so why not a cryptic subspecies?

 

[njlarsen]

The AOU criteria seems a little limited, how about voice differences?

Niels

 

[Richard Klim]

And as discussed earlier, why do the (numerous) authors merely 'suggest' a named subspecies, but refrain from formally describing it?
http://www.birdforum.net/showthread.php?t=164835

Richard

 

[jmorlan]

The AOU criteria seems a little limited, how about voice differences?

Niels

As far as I know, voice differences have never been used to delineate subspecies. For example the White-crowned Sparrow is famous for numerous well-defined regional song dialects. To allow these dialects to be used to describe subspecies would lead to a huge number of named subspecies that could never be diagnosed in museum trays.

Subspecies have to differ "taxonomically" which means they have to diagnosable based on populations differences visible in museum specimens. However some features fade or disappear in museum specimens (e.g. eye color) so soft-part colors are often noted on specimen tags. But I've never personally seen specimen tags note anything about vocalizations.

Richard, Sorry I missed your earlier post based on the prepublication version. I agree completely that without a designated type and an actual description of the bird, this cannot be considered a valid subspecies. I would have thought this would have been addressed in the peer review process.

 

[njlarsen]

I am not sure I understand why subspecies should be different from species. For example, SACC recently split the warbling doradito into two species based on voice in spite of the following statement: "I have examined and measured the type specimen in New York (LandbeckĠs specimen), and it is essentially like flaviventris but a bit longer-winged, perhaps brighter yellow below. and with a stronger cinnamon tone on the crown. There is essentially no reliable way to separate citreola and flaviventris based on specimens other than the longer wing, although doubtless there will be overlap in a larger series. The real difference is the voice."

Niels

 

[jmorlan]

Voice can be used to diagnose biological species if there is evidence that vocalizations are an "isolating mechanism." If song playback indicates that birds with different voices do not respond to playback of the other population's song, but strong territorial reaction to playback of its own population's song, then we would conclude that the birds do not recognize the other song as the same species.

Also if song is widely known to be an isolating mechanism between other congeners, it could be used in argument for species distinctness.

Although there was one controversial study to the contrary, it seems clear that song-dialects in White-crowned Sparrow (where song is known to be learned) are not an isolating mechanism.

Subspecies are, by definition, not reproductively isolated, so song differences cannot be used to diagnose them. Subspecies are merely a way of using nomenclature to describe geographic variation within species.

 

[Mysticete]

Usually people defer on naming subspecies/species because the author's formal training or interest isn't in taxonomy. This occurs often (far too often) in the marine mammal literature.

I Personally don't see much a need to recognize subspecies within the Gray Owl, but then I think subspecies are not a terribly useful concept to begin with.

 

[njlarsen]

Voice can be used to diagnose biological species if there is evidence that vocalizations are an "isolating mechanism." If song playback indicates that birds with different voices do not respond to playback of the other population's song, but strong territorial reaction to playback of its own population's song, then we would conclude that the birds do not recognize the other song as the same species.

Also if song is widely known to be an isolating mechanism between other congeners, it could be used in argument for species distinctness.

Although there was one controversial study to the contrary, it seems clear that song-dialects in White-crowned Sparrow (where song is known to be learned) are not an isolating mechanism.

Subspecies are, by definition, not reproductively isolated, so song differences cannot be used to diagnose them. Subspecies are merely a way of using nomenclature to describe geographic variation within species.

You might very well be right with the sparrow (where song probably is learned to some extent anyway?), but then think about two non-overlapping populations of an antbird. In voice they differ in two characteristics, but a third characteristic has not been detected. If three differences existed, then they would be good enough for a split according to the operational definition applied in much of what SACC does (quote: "the split standard of three diagnosably different vocal variables in songs"). So because they are not different enough to be full species, the difference in voice is to be completely ignored? Sorry, but that does not make sense to me.

Now if you like Mysticete think that subspecies is a classification that is useless anyway, then that is a different story

Niels

 

[jmorlan]

This is getting off the original topic of naming a subspecies based entirely on genetic data. There is nothing in the article suggesting that Great Gray Owls in the Sierra have different vocalizations. My point is that voice alone cannot be used to diagnose subspecies, but that is not a failing of the subspecies concept. Unlike species, subspecies are not necessarily an evolutionary unit

I think the subspecies concept is a neat way to describe intraspecific geographic variation and I agree with the AOU's position which I interpret to mean that subspecies must be phenotypically diagnosable.

 

[njlarsen]

This is getting off the original topic

Agree, let us close this discussion

Niels

 

[ntbirdman]

This is getting off the original topic of naming a subspecies based entirely on genetic data. There is nothing in the article suggesting that Great Gray Owls in the Sierra have different vocalizations. My point is that voice alone cannot be used to diagnose subspecies, but that is not a failing of the subspecies concept. Unlike species, subspecies are not necessarily an evolutionary unit

Quoting from the paper:

No differences in plumage or morphology have been observed across this barrier; however, the Sierra Nevada population differs from northern populations in migratory behavior, nest site selection, and prey preference (Bull and Duncan, 1993)

So there are characters to diagnose the subspecies beyond genetic data, if you think behavioral and ecological characters can be taxonomically meaningful.

 

[jmorlan]

It could be argued that differences in diet, etc. are simply a consequence of the owls occupying different regions with different available niches and different resources.

If behavioral and ecological characters can be shown to be acting as isolating mechanisms or part of an isolating mechanism, then I think they can be taxonomically meaningful at the species level.

But subspecies, by traditional definition require differences in plumage or morphology which can be diagnosed by examining of museum specimens. I think going beyond that requires revamping the definition of subspecies.

 

[Mysticete]

In regards to basing subspecies on morphology, what happens if different characters delimit different units. For instance if you use one character (or related set of characters), you might create different subspecific unit than if you use another set).

I am actually fine with using vocalizations for "subspecies" divisions (or clear cut ecological distinctions. I am skeptical of using only molecular diagnoses at this level simply because different markers may show different patterns, and that with enough markers you could probably delimit an infinite amount of "subspecies" which doesn't seem useful

 

[jmorlan]

In regards to basing subspecies on morphology, what happens if different characters delimit different units. For instance if you use one character (or related set of characters), you might create different subspecific unit than if you use another set).

I agree that it's more art than science. In fact the more characters you throw into the mix, the more subspecies you might be able to name. Look at Oberholser. He could see "decided" differences in color or even size that nobody else could ever see. Unitt wrote that about half of all named subspecies are probably invalid.

It's partly for this reason that the subspecies concept has been under attack. There are supposedly rules such as the famous 75% rule, but following such rules is rather tricky. The bottom line is that there is no unambiguous test that we can use to determine the validity of any claimed subspecies. It's all just informed opinion, experience and judgment.

I am actually fine with using vocalizations for "subspecies" divisions (or clear cut ecological distinctions.

On what grounds? Has your proposal been used in practice and published somewhere. I dread the day I'm trying to match an unknown specimen with known subspecies in a museum collection in which the features I need to know are the type of nest it built or what it ate. It's just not practical.

Besides I thought you said "...subspecies are not a terribly useful concept to begin with."

I am skeptical of using only molecular diagnoses at this level simply because different markers may show different patterns, and that with enough markers you could probably delimit an infinite amount of "subspecies" which doesn't seem useful

Agreed.

 

[Mysticete]

I don't like subspecies...i think most subspecies either are invalid to begin with, or are probably distinct enough for species treatment. At best they are useful place holders. I would be more comfortable with "informal" names for regional variants.

I am not aware of any subspecies designations solely based on call. There are full species where a specimen in hand is less useful than a recording of it's call (flycatchers). There are also distinctive forms recognized mostly on call (see crossbills).

your question of how you would match an unknown specimen with a species/subspecies, when the species/subspecies isn't diagnosable by morphology, already occurs. Slimy Salamanders are mostly indistinguisable from one another, and there is something like 9 recognized species. More confusing are some species of snakes, such as rat and king snakes, where the traditional subspecies/species boundaries don't fit well the phylogenetic pattern. This situation occurs as well with chipmunks, deer mice, fence lizards, etc. You could take the devil's advocate position and argue if these are not insurmountable problems for identification of cryptic species, than they are not insurmountable problems for cryptic subspecies.

Again mostly arguing from the devil's advocate position...

 

[jmorlan]

Devil's Advocate?

Okay, but just to be clear I think there's a logical fallacy in your arguments. That fallacy is that subspecies concepts have something to do with species concepts.

They don't.

Subspecies are about variation and species are about diversity.

 

[Kirk Roth]

I don't like subspecies...i think most subspecies either are invalid to begin with, or are probably distinct enough for species treatment. At best they are useful place holders. I would be more comfortable with "informal" names for regional variants.

Its hard to let this pass. Without formal names for subspecies (or anything for that matter), we would be inviting a lot of colloquialism and confusion. Avoidance of such is the whole point of Linnean taxonomy. I don't want to have to write papers in which my topic is "Melospiza melodia, you know - the one with lightish brown medium colored spots from west Texas." I believe that subspecies should be used (in birds) the way they traditionally have been - geographic morphological distinctions which do not contribute to reproductive isolation. And if we don't name those distinctions formally, we'll be that much less effective at communicating about them.

I agree with Joe that we are too late in the taxonomic game to shake up the definition of subspecies to include genetic or behavioral distance. However, that is not to say that we just ignore the differences altogether. Red Crossbills are a good example - they are considered "Call Types," not subspecies. It seems to me that the Sierra Nevada Gray Owls are possibly a different "Behavioral Type" (aren't most birds?) or "Genetic Variety" (aren't we all?), but I'm very hesitant to say that they fit the definition of a subspecies. I haven't read the paper, but I wonder if the Sierra Nevada owls would show the behavioral difference if they were dropped down in Canada and vice/versa.

Slimy Salamanders are mostly indistinguisable from one another, and there is something like 9 recognized species. More confusing are some species of snakes, such as rat and king snakes, where the traditional subspecies/species boundaries don't fit well the phylogenetic pattern. This situation occurs as well with chipmunks, deer mice, fence lizards, etc. You could take the devil's advocate position and argue if these are not insurmountable problems for identification of cryptic species, than they are not insurmountable problems for cryptic subspecies.

Or try goldenrods (Solidago), in which some species are identical except for chromosome number. The unspoken issue here is that different taxa need different criteria for classification. The American herpetologists seem content to name many species on the basis of genetic difference, but at the same time, Appalachian salamanders are creatures of very limited mobility. Not as much opportunity for integration, compared to a Northern Flicker. Plants are classified as varieties more often than subspecies, to reflect morphotypes that are often genetically though not phenotypically plastic. Again, this would not make much sense with birds. Bacteria, to use a more extreme example, have been shoved into generic and specific naming with much difficulty, but "strains" and "lineages" are utilized to be more informative.

I'm not ready to muck up our traditional definitions of bird subspecies - whether by changing that definition or by dropping it altogether. From a descriptive standpoint, I think it works out alright.

 

[Mysticete]

Subspecies are traditionally considered stepping stones to species in the biological species concept. Forms with some level of distinction from one another but no evidence of any degree of reproductive isolation. Organisms on the pathway to becoming species but not deserving that distinction. This was the definition Mayr and other BSC proponents used. However even they, decades ago, argued that the use of subspecies was greatly greatly flawed. They didn't hold that characters showing clinal variation should be used for subspecies definition, and many traditional subspecies were named off of insignificant sample size. In addition to that genetic studies have show that many times subspecies show no genetic distinction from one another, which suggests that many forms (see song sparrow) probably have not reached a point where subspecific recognition is useful.

I think many ornithologists are moving away from the subspecies concept. How often do you see major studies revising the subspecies classification of birds? Even the most conservative subspecies supporters would argue that our current subspecific classification of birds is profoundly flawed

 

[jmorlan]

Subspecies are traditionally considered stepping stones to species in the biological species concept.

Not at all true. You cite Mayr who made perfectly clear that subspecies are not stepping stones to species. Geographic isolates may be stepping stones, and they may (or may not) also happen to be subspecies, but subspecies in general are not evolutionary units.

Forms with some level of distinction from one another but no evidence of any degree of reproductive isolation. Organisms on the pathway to becoming species but not deserving that distinction. This was the definition Mayr and other BSC proponents used.

Mayr never used that definition for subspecies in any of his writings that I have read. Mayr did not regard subspecies as being on the pathway to becoming species unless they also happened to be geographic isolates. In those cases it was the geographic isolation which provided the pathway to an independent evolutionary future. That pathway is not available to non-geographically isolated subspecies which are connected by hybrid zones. Subspecies are merely "Intraspecific populations which differ taxonomically from other such populations." There is nothing about pathways or stepping stones in the standard definition.

I'm not sure if you are playing "Devil's Advocate" again, but I'll challenge you to find a quote from Mayr saying that subspecies are stepping stones to species.