When is it a subspecies? (3 Viewers)

[Accentor]

In some bird populations, there are consistent differences, some small, some moderate between two or more populations. Yet in many of these species, they are not considered subspecies. At what point does a population become a subspecies? What other criteria are used?

I can point to a recent population of Dark-Eye Junco. In the early 80's,a new population established itself in San Diego, Ca. In that time, the birds of this new population have gone through a few changes. Most notably, having less white in the tail, weighing slighly less, having shorter wings and tails, less aggressive behavior and so on. How long will it take for them to at least become a new subspecies and again, is it more than just genetic drift and developing new traits?

http://onlinelibrary.wiley.com/doi/...T+to+1200+BST;+Singapore+1700+SGT+to+1900+SGT.

http://www.bioone.org/doi/abs/10.1554/03-182?journalCode=evol

 

[njlarsen]

I think one good answer is to refer to this thread from earlier this year.

Niels

 

[Acrocephalus]

I can point to a recent population of Dark-Eye Junco. In the early 80's,a new population established itself in San Diego, Ca. In that time, the birds of this new population have gone through a few changes. Most notably, having less white in the tail, weighing slighly less, having shorter wings and tails, less aggressive behavior and so on.

In the case of introductions, the changes in measurements and body mass could be due to genetic drift caused by the founder effect; though the effect of selective evolution can not be ruled out. See this case from Helgoland:

Förschler M. I. & Geiter O. 2010. Morphological shifts in the House Sparrows Passer domesticus of Helgoland: insular syndrome or founder effects? Ornis Fennica 87:93–98. PDF

 

[njlarsen]

In the case of introductions, the changes in measurements and body mass could be due to genetic drift caused by the founder effect; though the effect of selective evolution can not be ruled out.

Forgive me for some hair-splitting, but genetic drift and founder effect are more separate than that.

In the case of a group of individuals being transferred to a new area (or choosing themselves to go there), these individuals may be genetically different from the average of the genes in the population they come from. Founder effect describes which genes are present in these founding individuals.

Genetic drift is about the changes in genetics that happen during the following generations, and which happens in every population that is small: one important aspect is that the changes are not caused by selection or genetic fitness. Imagine a group of 10 pairs, where one of the parents in one of the pairs carry a hidden (recessive) allele for something. If that pair happens to get more offspring than the other pairs (in humans that could be so simple as just loving kids more), then that recessive allele has increased its allele frequency in the following generation, and that is genetic drift. Unless there is a continued influx of additional immigrants, then the founder effect is done with generation one, and founder effect cannot cause genetic drift.

Niels

 

[Acrocephalus]

Forgive me for some hair-splitting, but genetic drift and founder effect are more separate than that.

In the case of a group of individuals being transferred to a new area (or choosing themselves to go there), these individuals may be genetically different from the average of the genes in the population they come from. Founder effect describes which genes are present in these founding individuals.

Genetic drift is about the changes in genetics that happen during the following generations, and which happens in every population that is small: one important aspect is that the changes are not caused by selection or genetic fitness. Imagine a group of 10 pairs, where one of the parents in one of the pairs carry a hidden (recessive) allele for something. If that pair happens to get more offspring than the other pairs (in humans that could be so simple as just loving kids more), then that recessive allele has increased its allele frequency in the following generation, and that is genetic drift. Unless there is a continued influx of additional immigrants, then the founder effect is done with generation one, and founder effect cannot cause genetic drift.

Niels

Thanks Niels for your clarifications. You are right that genetic drift and founder effect are two different things, but I think that in the case of the House Sparrow (and similar examples) the founder effect is the cause of the genetic drift in the following generations. A few House Sparrows founded the Helgoland population, so their genetic structure (allele frequency..) is transmitted to the following generations. With time, the founded population can be more like the founding individuals than the source population. So, a founder effect can cause genetic drift. Of course, genetic drift can be caused by reduction of population size in situ following for example a catastrophic event.

It should be noted that the authors of the article provided also an alternative explanation.

 

[mckaybailey]

Hi all,

I read some on this board from time to time. I haven't posted yet, but maybe I will if I see something that can easily be cleared up. Right now, I'm a grad student studying bird phylogeography and speciation.

Accentor: Ernst Mayr's famous definition of a subspecies was a population that differed taxonomically from other such populations. What does "differ taxonomically" mean? Well, nobody really knows, and it has been taken to mean any measurable differences among populations. Thus, there is nothing stopping someone from describing your recent junco population as a new subspecies. These days, however, there is a general push for more objectivity for subspecies. The result is people are beginning to demand some diagnostic criteria for subspecies. For example, a 70%, 95%, or 100% (phylogenetic species) criteria of diagnosability before a subspecies will be recognized.

Niels: Founder effects certainly increase genetic drift. Genetic drift is simply the change in allele frequencies that is expected in any population with a finite size. Drift is stronger in smaller populations, but it happens all the time in any real population (i.e. any population without an infinite number of individuals). If anything besides random chance is at work (e.g. love), then that is not drift. Founder events increase drift by reducing the population size. However, remember drift happens all the time in every real population. Most of the time, its effect is so small that we just ignore it. I hope this makes sense.

Bailey

 

[njlarsen]

Bailey,
welcome to birdforum.

Usually, in larger populations, the random changes are so small that genetic drift is not detectable (because the effect of someone loving kids a lot is balanced by another carrier going in the opposite direction). Notice that I do not disagree that such changes should occur in theory, just that in practical terms genetic drift is limited to small populations unless you look over very long time scales.

I strongly disagree that founder effect causes genetic drift. It is usually such that you talk about founder effect in small populations where the genetic composition is determined by the founding, and where the small population size then cause genetic drift. Less relevant to the above discussion are the cases where a point mutation cause presence of a gene (usually a disease allele) in a large population (just to mention one example, there are several genetic diseases that have been mapped in the Finnish (human) population using methods based on this model); with that type of founder effect, there is no increase in genetic drift.

Finally, who am I: I get my paycheck as a professor of human genetics at a medical school.

Niels

 

[mckaybailey]

Niels,

Loving your kid has nothing to do with genetic drift. Drift is just a random change in allele frequencies.

A founder event increases genetic drift by definition. It doesn't really cause drift because drift happens in all populations all the time anyway. Like you say, it usually has a small effect in larger populations but it still happens. By reducing a population size, a founding event increases genetic drift. There is absolutely no way around that. It's a truism.

If you're saying that it isn't the founder event per se, but the resulting small population size that results from a found event that causes drift, then I don't really see the distinction. A smaller population size is part of a founder event.

Bailey

 

[mckaybailey]

Niels,

I just read you bit about disease in Finnish population. Maybe we're talking past each other here. In what sense is the mutation of a new disease allele into a large population a founder effect?

A founder event (with associated effects on the founding population) is the establishment of a new population that is smaller than its parent population.

Bailey

 

[Accentor]

Thanks Niels for your clarifications. You are right that genetic drift and founder effect are two different things, but I think that in the case of the House Sparrow (and similar examples) the founder effect is the cause of the genetic drift in the following generations. A few House Sparrows founded the Helgoland population, so their genetic structure (allele frequency..) is transmitted to the following generations. With time, the founded population can be more like the founding individuals than the source population. So, a founder effect can cause genetic drift. Of course, genetic drift can be caused by reduction of population size in situ following for example a catastrophic event.

It should be noted that the authors of the article provided also an alternative explanation.

It's interesting that House Sparrows are being mentioned. Another example are the small regional differences seen in House Sparrows across North America; birds on the plains are slightly larger than other populations and those in the deserts brighter and perhaps smaller as well. With a continous range, would this be the founder effect or genetic drift?

 

[Accentor]

Hi all,

I read some on this board from time to time. I haven't posted yet, but maybe I will if I see something that can easily be cleared up. Right now, I'm a grad student studying bird phylogeography and speciation.

Accentor: Ernst Mayr's famous definition of a subspecies was a population that differed taxonomically from other such populations? What does "differ taxonomically" mean? Well, nobody really knows, and it has been taken to mean any measurable differences among populations. Thus, there is nothing stopping someone from describing your recent junco population as a new subspecies. These days, however, there is a general push for more objectivity for subspecies. The result is people are beginning to demand some diagnostic criteria for subspecies. For example, a 70%, 95%, or 100% (phylogenetic species) criteria of diagnosability before a subspecies will be recognized.

Niels: Founder effects certainly cause genetic drift. Genetic drift is simply the change in allele frequencies that is expected in any population with a finite size. Drift is stronger in smaller populations, but it happens all the time in any real population (i.e. any population without an infinite number of individuals). If anything besides random chance is at work (e.g. love), then that is not drift. Founder events increase drift by reducing the population size. However, remember drift happens all the time in every real population. Most of the time, its effect is so small that we just ignore it. I hope this makes sense.

Bailey

It does make sense. I guess until someone comes up with a more definitive system to categorize subspecies, we will just have to wait and see what populations are called that and which aren't.

 

[njlarsen]

It's interesting that House Sparrows are being mentioned. Another example are the small regional differences seen in House Sparrows across North America; birds on the plains are slightly larger than other populations and those in the deserts brighter and perhaps smaller as well. With a continous range, would this be the founder effect or genetic drift?

Most likely the second but equally possible is selection according to my personal opinion not based on any research into the populations.

Niels

 

[mckaybailey]

Accentor,

The desert sparrows fit a general trend in biology called Gloger's Rule, which states that darker forms tend to live in humid areas and lighter forms tend to live in drier areas. So, the pattern in sparrows is consistent with selection. Genetic drift is a very weak force and has almost no ability to affect directed change in large populations.

BTW, Gloger discovered this trend by studying birds.

Bailey

 

[fugl]

Well, whatever the mechanism—drift, selection or some combination of the two--the House Sparrow seems to be making itself thoroughly at home in North America. In view of this, I wonder how far local adaptation & differentiation will have to go before the sparrows are accepted as a legitimate part of our avifauna rather than as just a pest to be exterminated?

 

[hermmays]

Hi all! I'm a friend and research collaborator of Bailey's studying avian population genetics, phylogeography and systematics. I thought I would jump into the discussion.

Less relevant to the above discussion are the cases where a point mutation cause presence of a gene (usually a disease allele) in a large population (just to mention one example, there are several genetic diseases that have been mapped in the Finnish (human) population using methods based on this model); with that type of founder effect, there is no increase in genetic drift.

Mutation, as far as I'm aware, is not a type of founder effect. Founder events of course occur at the population level and mutation at the level of individual loci.

BTW are you talking about Kari Stefansson's deCODE genetics group in Iceland or some other studies in Finland? I know there are large GWA studies in Iceland.

Herm

 

[hermmays]

Loving your kid has nothing to do with genetic drift. Drift is just a random change in allele frequencies.

I think all he means is that variation in how much one loves their kids would occur at random with respect to the survival of a particular allele subject to genetic drift. Individuals have all sorts of things that influence their survival but as long as all those those things occur at random with respect to a particular allele that allele will be subject to drift. Of course it isn't either/or and alleles can be subject to both drift and selection.

Herm

 

[Accentor]

Accentor,

The desert sparrows fit a general trend in biology called Gloger's Rule, which states that darker forms tend to live in humid areas and lighter forms tend to live in drier areas. So, the pattern in sparrows is consistent with selection. Genetic drift is a very weak force and has almost no ability to affect directed change in large populations.

BTW, Gloger discovered this trend by studying birds.

Bailey

Right, but in some cases, examples of Gloger's rule (as well as Bergman's and Allen's Rules) often coincide with subspecific classification. The Song Sparrow is a good example of that. Can these House Sparrows be on the same level?

 

[mckaybailey]

Yes, the Song Sparrow is a great example. The problem with defining subspecies within a species like the Song Sparrow is that, from a population genetics standpoint, there is massive amounts of gene flow across its range. The forces of gene flow, which works to homogenize populations, and selection, which works to differentiate populations, act against one another and this results in large stretches of the range where one character state gradually changes to another state. For example, from a dark color to a lighter color. We call these areas of gradual change "clines."

This gradual change makes it impossible to draw an objective boundary between subspecies with clinal differences. To make matters worse, different characters can have different cline distributions. For example, following Gloger's rule, plumage color might change from areas of high precipitation to areas of low precipitation. This cline would follow precipitation patterns. However, body size, following Allen's rule, might change from north to south. This cline would follow latitude patterns. So even if we could agree on optimal subspecies boundaries using plumage color, there would be different optimal boundaries if we looked at body size or other characters.

The arbitrary nature of these kinds of subspecies wouldn't be such a big deal except for the fact that non-systematists seem to take it for granted that they are real. For example, the US endangered species act protects subspecies. What makes the matter all the worse is that there are really distinct subspecies mixed in among all the arbitrary ones. For example, Myrtle and Audubon's Warbler have a sharp boundary that is concordant in many characters (e.g. plumage, genetics, song, etc). To a non-specialist looking at a list, on paper the subspecies of Yellow-rumped Warbler look just like the subspecies of Song Sparrow. In fact, they are very different entities; some can be talked about objectively and planned for in a conservation context, and some can't.

So to answer your question. I don't think there is any real reason that we would want to call desert forms of the Song Sparrow subspecies and not do the same for House Sparrows. Some might argue that House Sparrows in America should be divided into multiple subspecies. I would argue, based on my reasoning above, that it would be better to not recognize the desert forms of either species as subspecies. We can still acknowledge that there is geographic variation within these species (much like we do House Sparrows now), but this kind of geographic variation simply isn't amenable to having a name stamped on it.

Bailey

 

[Accentor]

So to answer your question. I don't think there is any real reason that we would want to call desert forms of the Song Sparrow subspecies and not do the same for House Sparrows. Some might argue that House Sparrows in America should be divided into multiple subspecies. I would argue, based on my reasoning above, that it would be better to not recognize the desert forms of either species as subspecies. We can still acknowledge that there is geographic variation within these species (much like we do House Sparrows now), but this kind of geographic variation simply isn't amenable to having a name stamped on it.

Excellent post Bailey. The thing is, the Song Sparrows are recognized as subspecies; the desert form being saltonus and the from in the Pacific Northwest being morphna and most of the coastal Ca. forms falling under heermanni

At some point, these differences in Song Sparrows warranted subspecific status but similar ones among House Sparrows did not.

 

[mckaybailey]

The difference is that some systematist argued strongly for the Song Sparrow subspecies and no one has done this same for the House Sparrow. You might say this is because the geographic variation within the Song Sparrow is older and it is certainly more extreme (the differences among Song Sparrow populations are greater than the differences among House Sparrow populations). However, there are plenty of bird subspecies with no greater differences among them than the differences we find within American House Sparrows (e.g. Dendroica dominica stoddardi off the top of my head). So I think the reason the House Sparrow doesn't have subspecies is more historical (no one has made the case for them) than conceptual (there's nothing fundamentally different about geographic variation within House Sparrow versus other bird species that do have named subspecies).

Like I said before, I think this is inconsistent and wrong. My solution, however, would not be to start carving the American House Sparrow in subspecies, but to remove clinal subspecies from other bird species such as the Song Sparrow. Such a classification would be more consistent, more rigorous, and less confusing (for both you and me!).