…and this is a 
Earlier today, Jerry mentioned to me that he noticed my earlier blog posts on the meeting, and thought I wasn't being critical enough. So I think that means I'm supposed to let my inner beast out for this one. (Nah, actually, it's because I'm in note-taking transcription mode while listening to these talks. I have to digest them for a bit before I can do any synthesis.)
What is the biogeography of speciation? Can one species split into two while splitting into two? Allopatric speciation: no gene exchange; Parapatric: limited exchange; Sympatric: free gene exchange. Allopatric is sort of the dogma of evolutionary biology. Everybody assumes gene flow and biogeography are the same thing, but they really aren't.
Nobody contests whether allopatric speciation happens, the question is simply how often it happens. Species concept Coyne uses: groups of interbreeding populations that show substantial reproductive isolation from other forms.
Why is there a controversy about biogeography? Darwin's concept was largely sympatric. The existence of species in the same area implies that they arose in the same area (clearly not necessarily true). The environment is regarded as important. There haven't been enough opportunities for allopatric speciation -- not that many barriers in the history of the world. Speciation is relatively difficulty with gene flow. Biologists opinions about geography have been conditioned by their own histories.
Can we estimate the frequencies of these different kinds of speciation? We have so many indubitable examples of allopatric speciation. Conditions are present everywhere.
Parapatric speciation: conditions are fairly easy, but data from nature is sparse and hard to get. Need evidence of clinal differentiation or evidence that allopatry never happened (which would be very hard to do). One example given: cave salamanders, two species, that abut a surface species, enabling a path for gene flow. Can't entirely rule out the possibility of an allopatric speciation event in their history, however.
It's easier to find evidence for the most controversial pattern, sympatric speciation. Theoretically supportable, and there are also experiments that demonstrate in the lab (with unlikely requirements, such as the complete lethality of intermediates).
- Complete or substantial sympatry
- True sister taxa not based on hybridization
- Substantial reproductive isolation
- History of taxa must make allopatry unlikely
Criteria for verifying sympatric speciation:
Special examples:
- Polyploidy. Up to 70% of angiosperm species have a polyploidization event somewhere in their ancestry. May still require allopatry to keep hybrids from being diluted out.
- Homoploid hybrid speciation.
- Parasitic indigobirds imprint on the song of the father, so laying eggs in different host yields individuals that only breed with individuals with similar stepparents.
- Palms on Lord Howe Island: very small island, so necessarily sympatric. Reproductive isolation by flowering time depending on soil type offers an alternative explanation, though: it reduces gene flow
Coyne doesn't regard this as true sympatric speciation because there was some kind of trickery that set up a reproductive barrier.
What about cases that satisfy the case of gene flow while speciation occurs? Under these stringent criteria, Coyne thinks 5 cases satisfy. The best cases are cichlids in crater lakes in Cameroon and Nicaragua. Littorina, banded molluscs that live in different tidal zones. Rhagoletis, the apple maggot fly, may not be the best case; they eclose at different times depending on the fruit on which they are laid, which represents a reproductive barrier.
Conclusion:
- There is no doubt that allopatric and peripatrica speciation occur
- Parapatric speciation may occur, but evidence is hard to come by
- Sympatric speciation is theoretically feasible, but…
- The few studies that suggest sympatic speciation occurs suggest that it only occurs rarely
Life Science
Comments
Posted by: mikecbraun
|
October 30, 2009 4:17 PM
Where do subspecies fit into this biogeography discussion? If there are polytypic species whose subspecies occupy the same region and, perhaps, later give rise to new species, is this allopatry or sympatry? I'm unclear on this.
Posted by: Glen Davidson
|
October 30, 2009 4:18 PM
Not a word about magical speciation? And it's almost Halloween, too.
Couldn't Behe wear his astrologer's uniform and explain how "poof" makes new species?
Glen D
http://tinyurl.com/mxaa3p
Posted by: taranaki
|
October 30, 2009 4:22 PM
Serious question - not attempted kiss up.
How do you get such a good summary of a session done so quickly? Coyne was scheduled to talk until 2:55. Your post was up 15 minutes later. I assume you are writing as he is speaking - but still it takes some time to get it together, write it, check for typos, etc. This post has 607 words - and no typos (did not check the scientific words that Word did not know). Same for your other posts. When you have some time, I would hope you can describe your process.
Note taking transcription is impressive enough - especially when it is done well. Any synthesis and critical examination could be used as evidence of a god and/or human cloning. Or a couple of assistants.
Posted by: Squiddhartha | October 30, 2009 4:27 PM
taranaki, I can't speak for PZ, but I've developed (evolved?) my techniques for live note-taking on my laptop, very much along the lines of what PZ's done for this meeting; it's a survival skill when one goes to as many meetings as I do and doesn't have a hope of remembering everything. Good touch-typing skills are the key.
The downside is that everybody now wants me to share my notes as the meeting minutes. :)
Posted by: MadScientist
|
October 30, 2009 4:28 PM
So, with sympatric speciation the cases offered essentially do not have enough data to rule out the sympatric scheme nor enough to say it's got to be one of the others?
Posted by: Antiochus Epiphanes | October 30, 2009 4:33 PM
Sympatric speciation through chromosomal translocations has been well documented in a number of flowering plants, and like PZ said, homoploid and polyploid hybrid speciation is rampant. Sympatric homoploid speciation has ben documented in Agrostis tenuis (Bradshaw, I think like 1968?) and a number of other edaphic endemics. I don't think botanists regard this as all that rare. I guess its good that Coyne has finally come around to the possibility that sympatric speciation occurs....more than Mayr did anyway.
Knowing that Jerry Coyne reads this blog, I must now apologize for being flippant. Its just that vertebrates are terrible models for studying speciation.
Posted by: SciencePundit
|
October 30, 2009 4:36 PM
PZ,
I have to agree with Coyne: you've been too accomodating so far. ;-)
Posted by: Antiochus Epiphanes | October 30, 2009 4:41 PM
Rift! Deep Rift! Myers is an accomodater!
Posted by: dave souza
|
October 30, 2009 4:45 PM
This seems to me to misrepresent Darwin. He knew about allopatric speciation from the Galapagos, but had a bigger puzzle with the two species of rheas having overlapping territory in South America. It was not until the 1850s that he found his answer to that puzzle, in a complex economy of nature providing more opportunities for different species diversifying from original species. He also worried that large populations were needed to provide enough variation as the raw material for natural selection, so focussed his arguments on that as the main cause without dismissing the rarer cases of isolation leading to diversifying species.
We now know that, if anything, it's the opposite in terms of frequency, but a lot of speciation still appears in large populations, as in cichlids in lakes, where we might claim that it's allopatric speciation due to each variety finding an isolated circumstance or territory, but in Darwin's terms that would be sympatric speciation with each variety finding a new niche in the economy of nature.
Posted by: Antiochus Epiphanes | October 30, 2009 4:52 PM
I always found the distinction between allopatry and sympatry to be arbitrary. As long as a reproductive isolation mechanisms is in place, physical location means bugger-all.
Posted by: theshortearedowl | October 30, 2009 4:55 PM
There has to be a reproductive barrier sooner or later... otherwise they're not speciated! (Well, by some definitions.) I thought sympatry/allopatry was supposed to be geographical?
Posted by: Shaggy Maniac
|
October 30, 2009 4:59 PM
I also posted this on the Schemske thread in response to another comment, but it seems relevant here as well:
Rapid Evolution of Reproductive Isolation in the Wild: Evidence from Introduced Salmon
Andrew P. Hendry,1* John K. Wenburg,2 Paul Bentzen,2, 3 Eric C. Volk,4 Thomas P. Quinn3
Science 20 October 2000: Vol. 290. no. 5491, pp. 516 - 518
http://www.sciencemag.org/cgi/content/abstract/290/5491/516
It's one of my favorite speciation articles and it deals with arguably sympatric populations.
Posted by: Kel, OM | October 30, 2009 5:18 PM
I was wondering about the cichlids in Lake Victoria, the amazing diversity there from a single species floored me.
What about the three-spined sticklebacks in British Columbia lakes?
And while I'm answering questions, does peripartic speciation come into it? Or does it fall under allopatric speceiation?
Posted by: Thanny
|
October 30, 2009 5:58 PM
No typos?
"Speciation is relatively difficulty with gene flow."
PZ probably uses a spell checker, which won't find typos like the one above.
Still, pretty good overall. I've seen published books with a higher rate of typographical errors. One of my curses is that I see such errors instantly when reading, but not when writing. The error PZ made above is one that's particularly easy to make, as one's brain debates which way to phrase something.
Posted by: qbsmd
|
October 30, 2009 6:43 PM
Why aren't ring species considered evidence for sympatric speciation? Or sympatric speciation in progress, since an extinction event in the middle of the ring would leave two separate species?
Posted by: David | October 30, 2009 6:57 PM
theshortearedowl #11
my understanding of all four biogeographical models of speciation: allo-, peri-, para-, and sym-patry, is that geography, or rather the species' range, is more a part of the descriptive term for each (the Greek root πατρίς or "patris" meaning "homeland"), not that geography necessarily caused each- one could argue that a river, or a mountain (allo-) or a subpopulation in search of food (peri-, para-) or shelter (cave organisms perhaps?) contributed to speciation but not that it did in a sympatric context- in that event, some other condition must be the cause- possibly differentiated fitness in certain environmental conditions, or morphology divergence where populations at either extreme of someting like a beak shape (which might alter bird call) becomes a barrier to gene transfer...
Posted by: David | October 30, 2009 7:17 PM
and then there is the case for Wollbachia, a genus of bacteria which infects arthropod species, (16% of neotropical insects, possibly up to 70%) and which is notable for altering the reproductive capabilities of their hosts
Posted by: Copernicus | October 30, 2009 7:25 PM
Thanny #15
but with a Levenshtein distance of only one, the typo you highlight can be forgiven... similar of course to your grammatical error when placing the comma between "make" and "as" in your last sentence!
Posted by: Maggie Moo | October 30, 2009 7:42 PM
Ha ha ha!
Nicklaus, good example of Skitt's Law (or Hartman’s Law of Prescriptivist Retaliation) for Thanny @#11!
“any article or statement about correct grammar, punctuation, or spelling is bound to contain at least one eror”
Posted by: Kristine
|
October 30, 2009 8:56 PM
qbsmd asked my question (and helped me articulate for myself what I was trying to ask). Thanks, qbsmd! I also wonder where sexual selection fits into this.
I view a species as a group of individuals who "hover" around an abstract "point" (not unlike a range of measurements around a certain temperature), who can interbreed, but because all (except for certain ants, wasps, etc.) are genetic individuals (I think creationists forget that), a subgroup could start interbreeding and genetically move away from the larger group or another subgroup, filling a unique niche but not necessarily being geographically or genetically isolated from the others. (Isn't that what we're talking about?)
Is it therefore possible to have a "species for all intents and purposes," which doesn't breed with another species, but could? In other words, if we discovered that what we thought were two distinct species actually can successfully interbreed, but don't for whatever reason, would we still consider them different species, or one? (Am I making sense?)
Posted by: David Marjanović, OM | October 30, 2009 8:57 PM
As far as I can tell from the post, this talk wasn't about speciation at all.
It was about cladogenesis.
Under some species concepts, that's the same as speciation; but under most it isn't.
Posted by: David Marjanović, OM | October 30, 2009 9:01 PM
Of course you are. You've just described the two Biological Species Concepts™. There are 145 more definitions of "species" out there...
Depending on the species concept, there are from 101 to 249 endemic bird species in Mexico.
Posted by: David | October 30, 2009 9:17 PM
David
Or if articulated by an entomologist, most mammals are in the same genus!
Posted by: Copernicus | October 30, 2009 9:21 PM
Of course, if you happen to use Facebook, you could always try this species model approach:
What Species of Pathogenic Bacteria Are You?
(can't claim credit for it but I am determined to be closest to Salmonella typhi!)
Posted by: Rorschach | October 30, 2009 10:00 PM
Reminds me of this :
Facebook Song
Posted by: Midnight Rambler
|
October 31, 2009 1:59 AM
Part of the problem is that it depends on how you define "sympatric", and the scale you measure it at. If you have a host-specific organism, especially one that actively seeks out its host, a host-switching event can lead to the newly-adapted form almost never encountering the ancestral one, and diverging from it with almost no gene flow. Likewise with organisms that are restricted to particular microclimates, like beetles found in moss clumps vs. bark.
There are also many plants with species pairs where one is adapted to edges and openings, and one to deep forest. If speciation occurs strictly on the margin, one might call that peripatric, but if it occurs throughout a forest mottled with treefall openings, it would be seen as sympatric (easily done with little gene flow if a flowering period change occurs at some point in the process).
There's just too many gray areas, and I think you can either say sympatric speciation is common, or define it so that sympatric speciation doesn't exist (with the palm example Coyne seems to be doing the latter).
Posted by: mythusmage
|
October 31, 2009 4:41 AM
There is a situation where a man's sperm is unable to penetrate and fertilize a woman's egg. The egg is in effect impervious to the sperm. Which means that man and that woman cannot breed. Given time might such a thing lead to speciation, and has it in the past?
Posted by: Copernicus | October 31, 2009 9:18 AM
mythusmage #28
but then we are talking about the biogeographic model of prophelaxipatry (προφύλαξις, πατρίς: prophulaktikos + patris)
Posted by: David Marjanović, OM | October 31, 2009 9:19 AM
This is what the final, decisive, defining, irreversible step of speciation is if you use Mayr's species concept.
Posted by: Copernicus | October 31, 2009 9:56 AM
David M.
But surely the inability of a human sperm to penetrate the egg membrane, or the egg's inability to accept a sperm, or perhaps some in transit issue to do with capacitation, doesn't lead to speciation per se, but is more an issue to do with natural selection where whichever gender is unable to produce offspring, by definition cannot perpetuate the species?
Posted by: Antiochus Epiphanes | October 31, 2009 2:12 PM
Speciation happens, but there really is no real entity that is a "species", which is to say that no amount of experiment would allow one to discover what a species is. Rather, a species is a taxonomic decision made by a taxonomist; organisms are grouped based on some idea of how it is useful to group them. Although evolutionary biologists will fight over species concepts, these are rarely used in practice when species are circumscribed. Rather, taxonomists name species based on subjective but explicit criteria.
Posted by: John Scanlon, FCD | October 31, 2009 11:14 PM
I sometimes like the idea that species barriers are the 'real' entities that come into existence and persist (or later dissolve), while the 'species' themselves are just the bulk stuff that flows through the channels so defined. That is the way to make sense of ring species.
It's a figure-ground thing, I don't think either is objectively more true. For those interested in what individual organisms look and behave like, including most of biology and evolution as far as I'm concerned, named 'species' are quite a handy method of keeping track of phylogeny and adaptation. Widespread lateral transfer erodes both views equally, so I don't have an answer for microbes (I'm an animal chauvinist).