starlarvae

As near as I can figure, gene regulatory networks and epigenetic mechanisms both regulate when and were and to what extent genes get expressed. But what is the relationship between GRNs and epigenetics? Are epigenetic mechanisms a subset of GRNs? If so, what distinguishes epigenetic mechanisms from other aspect of GRNs? Or are GRNs and epigenetic mechanisms distinct cellular mechanisms that work in tandem? Any clarification regarding these terms will be appreciated!

Guilt by association? More importantly, what do you make of his assessment?

Commentary from the dark side, but what do you think? Is this a fair criticism of the conclusions drawn by the researchers? Or, where does this critic go wrong? http://www.evolutionnews.org/2015/03/resurrected_fla094081.html

Does it strain credulity to assert that the necessary two mutations, which repurpose a distantly related protein, occur predictably in each independent strain? That's one heck of a coincidence. Lucky flagellates.

Speciation event is really speciation "event." Sort of an event, but not really. More like an ongoing process. (The literal event occurs when the taxonomists say it does.) Natural selection is really natural "selection." Sort of like selection, but not really. More like wait and see what happens (i.e., see how distribution of reproductive success yields trait distributions among offspring) Selection pressure is really selection "pressure." Sort of like pressure, but not really. More like wait and see what happens. . . . Why is evolution theory so mealy-mouthed, so reliant on metaphors? Why can't it just say what it means, literally?

Hmmm . . . . I thought "speciation event" was standard terminology. Speciation in space colonies? Never thought of that!

On origins of gender dimorphism, see recent paper on origins of pregnancy at http://www.cell.com/cell-reports/fulltext/S2211-1247%2814%2901105-X . Comment from Brig Klyce at http://www.panspermia.org/whatsnew80.htm :::: "Thousands of ancient genes simultaneously repurposed and re-regulated by invasive TEs? Such complex, coordinated steps must have relied on powerful genetic programming. It strains credulity to suggest that the programming was suddenly, incrementally composed by darwinian trial-and-error. The programming must have been available already." In other words, this seems to be an amazing case of "pre-adaptation."

If they're of distinct lineages, then the differences might be due to plasticity PLUS genomic variation. The variation referenced in the OP was intraspecific variation. How far back into ancestry do you go to determine "distinct" lineages, anyway? Second cousin twice removed? "postzygotic reproductive incompatibility" ? I don't think I said or implied that. On the contrary, I assume that with artificial insemination, any two dog breeds could be interbred. I still would like to see an example of a species naturally as plastic as dogs --- an example that does not cite castes or life-cycle stages or genders as distinct phenotypes within a species, because such variants cannot be reproductively isolated. Arm waving? If anybody knows of a study that compares species by degree of plasticity, quantified, I would like to see. Mimulus, so far as I can determine, is a genus designation, and so we would expect it to encompass broader phenotypic variability than would a species. I think you're right about typical usage of "plasticity", but if you substitute "phenotypic variability" or "morphological variability" or something like that, is doesn't change the point made in the OP.

Are you kidding? The dramatic within-species variability shown in the dog chart in the OP is the result of selective breeding. I think this thread already has established that the "species boundary" in a given case is decreed by a panel of taxonomists. Phenotypic variation alone won't do it. You need reproductively isolated phenotypic variants. As the variants become more and more varied, at some point the taxonomists step in and pass judgement. I'm not sure how important genomic/genotypic differences are, since the environment can select based only on phenotypes. It would be useful if some official panel of taxonomists made public their criteria for decreeing a new species (degrees of genetic difference, phenotypic difference, interbreeding, behavioral repertoire, etc. If you know of such, tip me off. How much stage setting is required before the play begins? I guess we'll have to wait for the taxonomists to take their seats, first. The Wiki entries are interesting, but we have to be careful not to let phenotypically distinct castes (as with social insects) or life-cycle stages (as with metamorphosing creatures) count as phenotypical variants in the context of this thread, because such intraspecific variants, unlike dog breeds, cannot become reproductively isolated from one another.

Recency of gene flow from wolves, etc. So what? Look at some of the goofier dog breeds. They've pretty much had any wolfiness bred out of them. What constitutes an "evolutionary history"? If length of time is the missing ingredient, then what exactly is it that time would accomplish? Reproductively isolate distinct sub-species? But that's exactly what the breeders have done. Unless you're talking about phenotypically distinct sub-species (or castes) that can become reproductively isolated from one another, as dog breeds can, your example is not relevant. Subjective judgements? Phenotypes are empirical. You can measure parameters. Height, length, weight, skull shape, leg shape, tail shape, fur thickness, color, pattern --- it's endless. Where in nature will you find such phenotypic variability as you will find in the dog chart? BTW, are the frog variants reproductively isolated from one another? If not, then the example is not relevant. And your phrase, "independent evolutionary history" is not clear. Is it just another way of talking about reproductive isolation?

The distinctions you draw among and between species and sub-species seem awfully arbitrary. Dog breeding is about removing variation from a sub-species, you say. Well, maybe it's about creating novel sub-species. Look again at the chart in the OP. The enormous variability within the species -- that is, from sub-species to sub-species -- is the result of selective breeding. Within a sub-species, variability will be relatively small -- IF the sub-species is reproductively isolated. That's how breeders create sub-species. Consider the flies in your example -- the two types became reproductively isolated (or, at least became genetically distinguishable) because of their preferences for the different kinds of trees. The reproductive isolation of variants is all we're talking about. Each kind breeds with its own kind. If the flies display less phenotypic variability than do the breeds of dogs (do they differ phenotypically in ways other than feeding behavior?), then why the fly speciation but no dog speciation? (You don't actually say that there was a speciation event with the flies, but you seem to imply it.) As for breeding out differences from dogs, look again at the OP chart. Don't you see lots of differences among the breeds? Non-argument? I'm just noticing a case of reproductive isolation of sub-species (aka, breeds) in which the sub-species vary dramatically from one another in their phenotypes. That is supposed to set the stage for speciation. What's the missing ingredient?

How is your first reference relevant to contemporary dog breeds? The breeds in the chart in the OP haven't been engaged in gene flow behaviors with wolves or wild dogs. Breeders keep the breeds reproductively isolated. That's the point. Social insects: If each caste could reproduce on its own, I suspect speciation would follow. But to cite non-reproducing castes in a species of insect misses the point. You can't get each caste to be reproductively isolated. The bird example shows color variation. Do these variants cohabit in the same niche? If they've been subjected to the same selection pressures, why did they turn out so differently colored? Or, do the colors indicate gender within the species? If the latter, then this is a bad example, because neither gender could become reproductively isolated from the other. The frogs: variation in color, but morphologically, allometrically not a whole lot of difference, nothing like that shown in the dog breed poster. Phenotypic variation within a species PLUS reproductive isolation of each variant sets the stage for speciation in the standard understanding of speciation by selection.

Look again at the dog chart in the OP. Where in nature will you find that degree of variation within a species? It matters if you argue that phenotypic variation correlates with variable reproductive success. If big variation among reproductively isolated subgroups of a species doesn't create new species (as is the case with dog breeds), then small variation won't do it.

yeah, natural selection can at most tune the piano, It can't compose the melody. Because natural selection will never have variants among which to choose that will be as dissimilar as those produced by breeders. If the breeders' variants don't speciate (is that a word?), there's no way nature can speciate based on selection among variants. Nature's primary source of new species must be something other than natural selection.

Natural selection can't be the primary mechanism of evolution.

The "it" that occurs in nature is speciation. The theory behind the process is weak where it hinges on variation within a species, because nature can't match breeders' ability to produce morphological variants from a founding population. The dog breeds tell us that variation within a species can't have much to do with speciation. You can say that speciation via natural selection takes a long time, but what does all that time do besides reproductively isolate variant phenotypes?

The intent of the breeders, or lack of it in nature, has no bearing on the process of speciation. Reproductive isolation of subgroups, the various subgroups being subjected to differing selection pressures, should do it. How much morphological variation among reproductively isolated subgroups does speciation require? Even more than the breeders have produced? That would never happen in nature.

Admittedly, the critique of the definition of "species" took this thread far afield. The OP really had to do with reproductive isolation of sup-populations, each isolated group being subject to HIGHLY selective selection pressures at the hands of the dog breeders. The point in the OP is that, given the morphological variation among dog breeds--much more extreme than would occur within a population in nature--why has there not yet been bred a non-dog? How does nature pull new species out of the much less pronounced variation that occurs in natural populations? If the evolutionary account of speciation leans mainly on variation plus selection, then you can't contrive a situation more ripe for generating new species than that of dog breeding, yet no non-dog has emerged.

Well said. Leave it to financial interests distinguish species? why not? Who needs a panel of taxonomists?

Pick & choose from among a menu of definitions according to my purposes? That's weird science. This link: http://en.wikipedia.org/wiki/Species#Definitions_of_species really is an eye opener. Is there any term outside of the biological sciences that is so problematic, any term whose definition is so wildly elastic?

That hunk of rock [Pluto] originated whenever and however it did. At some point it was deemed to be a planet. Then not. It became a planet only when it was granted planethood, and ceased to be one when the definition of "planet" was changed. If the definition of "rock" gets changed, then the object might cease to be a hunk of rock. Point is, anyone can look up the definition of planet and see if Pluto qualifies. It's not clear that anyone can look up the definition of species and see if Great Dane qualifies. You want to have it both ways. You wants speciation events to occur in nature and you want taxonomists to decide what counts as such an event. SO, prior to the taxonomists' verdict being rendered, how do you know whether or not any of those events out there in nature are speciation events? If the pros in their fields have specific thresholds of, say, reproductive isolation, or phenotypic (or genotypic) variability, or other parameters that determine whether two critters are of the same or distinct species, then what are those parameters and what are the thresholds? I would like to see (and it might shut me up if I did) the formal description of the features collectively that distinguish some group of organisms as a species, distinct from all others. If such parameters with their thresholds and attendant formal descriptions of features are not to be had, then I suppose that an accusation of "arbitrary and capricious" is to a degree justified.

How do we know how much variation is enough? We have to wait for the panel of taxonomists to declare a new species, I guess. It's hard (impossible?) to imagine a species producing as much variability in nature as the breeders have been able to pull out of the dog genome -- and reproductively isolate. No, I cannot imagine that it would occur in the wild. I think it's fair to say that the specialized breeds ARE reproductively isolated. The breeders make sure of it. Then I guess that the origin of species occurs when the taxonomists decree that a new species has originated. Speciation events occur at meetings of taxonomists, not in the wild. That seems to be the implication of what you guys are saying.

Right. "Species" does not name a natural kind. It's just a convention of terminology. If the taxonomic panel perceives a lack of family resemblance, then two species. So, the actual origin of species does not take place in nature or at the kennel club, but in the tally of votes among the panelists. Makes sense to me.

Why would it matter to their taxonomic status whether their current state is the result of natural or artificial selection? If a person wanted to create a new species from a founding population of dogs (without manipulating genes in a laboratory), that person would do what dog breeders have done -- namely, divide the founding population into inbreeding subgroups that are reproductively isolated from one another and breed within each group selectively for morphologies that diverge from those of the other groups. Normal evolution theory says that this, effectively, is what nature does, but with much less capacity to render morphological variation from a given species genome -- that is, generating much less variability in a population for selection to work on, as compared to dog breeders. The dog breeder's chart included in the OP can be read as a depiction of dog morphospace, and it's pretty well filled up, but still no new species. (If you want people to take seriously an assertion that Great Danes and Chihuahuas are distinct species, I think you'll have an uphill battle.) (I don't suppose there's much interbreeding between Eskimos and Tutsis, but would anyone say that these ethnicities represent different species? Have biologists quantified how much interbreeding would have to go on before two groups of whatever would be designated as the same or distinct species?) On your characterization of "species" as being a useful, if fuzzy, convention for tracking relatedness and breeding status of populations ---- all kinds of scientific terms are fuzzy (e.g., do the planets include Pluto or not?), so I don't pick on biology in particular. But one can find references to speciation events, and it seems to be a standard term. But given the fuzz around "species," such events would seen to consist of a panel of judges (whoever they are) deciding that some critical number of criteria are met. But it seems pretty arbitrary which criteria are considered, how much they are weighted relative to one another, and what the thresholds are for deciding species or nonspecies in any particular case. As an exercise in truth-in-advertising maybe we need to rename Darwin's book, "On the Origin of Pragmatically Useful but Admittedly Only Loosely Defined Categories" Again, my point in the OP is that dog phenotypes vary much more than would be found in any natural population of some given species. And phenotypic diversity is supposed to be fertile ground for speciation, yet we find no non-dogs resulting from dog breeding. (& who gets to decide dog/non-dog? The American Kennel Club? {!} )

I can't imagine a dog breeder thinking, "Hmmm, I'd better be careful. My breeding project might produce a non-dog." A dog breeder wouldn't last long as a breeder if she produced dogs that were not successful at passing their genes along. Whether breeders select or nature selects, both will factor in prospects for reproductive success. If natural selection produces new species by working on the variation in a population, then, with all the variation among dog morphologies, why yet has no non-dog been bred? The link in post #5, above, goes to a web page that makes clear the ambiguity of the term "species." And I like your anthropomorphizing of nature as an agent who "considers" before she "selects." My comment referred to category names, like "species" and "niche". I suppose you could say that those terms are like "flat" in your example, insofar as they ought to be retired or superceded.