Open discussion

Matt Skaggs had the audacity to ask, and follow up on, a paleobotany/evolution question involving the California flora, which I really should know more about than I do, instead of talking about stupid stuff like football. You can add to that discussion of course (copied below), one that ranges taxonomically from cypresses to sequoias to serpentine-tolerant mustards, and conceptually from Darwin to Wright to Goldschmidt, but I figured I’d better make a post allowing for questions/discussions on miscellaneous topics of interest, for putting up links to interesting articles, and so forth. The abrupt transition from bluegrass song lyrics to the evolutionary origins of serpentine endemics in California is well known to mess with peoples’ senses of flow and thus probably should be avoided. [Editor’s note: the term “bluegrass”, as used above, does not refer to Poa pratensis except in a very historically indirect way–apologies for any taxonomic confusion.]

In other thrilling news, I messed around with some other WordPress “Theme Options” to see if I could get the content and comments to span the screen better, but had no luck and figured I’d better leave well enough alone before I broke something. I was able to force replies to comments to nest to a max of one indentation level to aid the cause however.

Below is copied the discussion by Matt and I, for your literary pleasure. Note that this will likely bring a barrage of comments, so try to get yours in early.

I just read your background for the first time, and with your connections to California, thought I would offer up a paleobotany topic. I have always been fascinated by the distribution of Cupressus in California. Much of the global diversity resides there, but many of the species/subspecies are very rare. I’ve often thought that if one could truly understand the distribution and genetics of this group, the persistent mysteries of speciation- at least in plants – would be resolved. And if that were to happen, it would connect Darwin to Goldschmidt via the work of Sewall Wright. And there’s more: by the time we are in High School we know about the megafauna in the La Brea Tar Pits, but few are aware that the tar pits existed in a cypress forest that was once widespread but is now essentially gone. Understanding the change in forest composition could reveal what really happened to the megafauna. Paleobotany indicates that the Sierra also once had widespread cypress forests but of low diversity. My rather extensive research on Cupressus all occurred in the 1990s, and cladistics has improved our understanding a lot since then. Did you take in interest in these sorts of questions when you were in California?

[my response]
Uh oh, I’m in trouble here!

The cypresses are indeed interesting, biogeographically, (and ecologically) like so many taxa in California. I’m no expert on them, that’s for sure, but their distribution exemplifies a familiar pattern across a number of important plant taxa. The one I have the most experience with is Arctostaphylos (the manzanitas) because I worked on one of the rare ones for a while. And there is the classic poster-child case–Giant Sequoia. And there are other examples of such disjunct distributions, like Ceanothus spp. (CA lilacs), and similar kinds of evidence even within individual species, such as (off the top of my head) Brewer spruce (Picea breweri), red fir (Abies magnifica), and foothill pine (Pinus sabiniana) all of which show major subspecies/race disjunctions, some of which are very hard to impossible to explain with existing data, although I’m unsure on how much genetic analysis has been done for each.

Arctotaphylos and Ceanothus are better analogues for the Cupressus situation, in terms of distribution pattern, as all involve many clearly different species scattered and disjunct over a large area. On the other hand, the sequoia and other tree examples mentioned above are better analogues, in the sense that they are wind-pollinated coniferous trees, like the cypresses (whereas the manzanitas and lilacs are smaller (i.e. shrubs), but more importantly, are insect pollinated). And these differences are likely reflected in the enormous numbers of species of both genera–where else will you find 40-60 species of shrubs in a single genus in an area about the size of California? And boatloads of similar examples can be found among the herbaceous species, especially the insect pollinated ones (e.g. Penstemon, Clarkia).

And then again, Cupressus and Arctostaphylos/Ceanothus share a distinguishing similarity, in that they originated from the “Madro-Tertiary geoflora”, so they share a long term ecological/environmental history, i.e. having originated in hot and dry, fire-adapted vegetation groups from the Mediterranean and similar xeric climates of North America, i.e. what are now Mexico and California. And certain life history traits that originated from that experience (e.g. latent bud sprouting after fire, long seed dormancies and cone closures broken only at heat thresholds), very likely influence their subsequent evolutionary history.

So, perhaps all that just to say that the cypresses really are an interesting case, as you state.

As for process, giant sequoia is probably the classic case of a taxon known to formerly have covered a much large area earlier in the Cenozoic which has since shrunk and split to its present scattered distribution. There is almost certainly an environmental tolerance component to this, since existing sequoia groves are restricted to +/- definable environmental ranges, and known, late Cenozoic geological events such as the rise of the Sierra Nevada (and consequent drying of the Great Basin) were very likely involved in the range reduction. But, how do we explain sequoia absence in the many Sierran locations currently having environmental conditions very similar to those in existing sequoia groves, but yet lacking sequoias?

But I’m pretty sure I’ve not answered your question about what Cupressus teaches us about plant evolution, or connected Darwin with Goldschmidt via Sewell Wright, much less in relating anything to megafaunal extinctions. Can you elaborate on what you were after there?

[Matt’s response]

Thanks for taking the time to respond, you make some interesting points. I’m afraid I scattershot a few too many topics. Cypress distribution and diversity is interesting primarily because it challenges notions of genome plasticity.

My favorite hobby is an odd one, serpentine endemism and rarity, I have never been able to do much with the manzanitas because they are so darn hard to tell apart in the field. I did spend considerable time studying two oddball genus, Streptanthus and Hesperolinon, that are scattered on disjunct serpentine exposures in central California. Obligate self-fertilization is common in these groups, usually on serpentine and usually involving rarity. The central California serpentine/peridotite exposures are thought to all be young, almost certainly having been exhumed during the rise of the coast ranges and the Sierra Nevada. I like how that collapses the cause tree, these plants have had little time to diversify and so are almost certainly neo-endemics. The paradigm is that the rare serpentine plants evolved from more widespread and generalist species. It is really hard to imagine evolving selfing and serpentine tolerance sequentially for a single, rare, neo-endemic species; it is inconceivable that it would happen multiple times in the same genus.

A mutant seed plant that has developed serpentine tolerance faces a major hurdle. It is surrounded by closely related individuals that lack the tolerance, and the lineage is subject to loss of tolerance due to back-crossing. Borrowing ideas from Sewall Wright, the facilitation of speciation requires both the migration into a new niche (in this case serpentine) and reproductive isolation. These plants seem to have isolated in place by obligate selfing. So what do we have? We have multiple examples within a single genus in which the evidence strongly supports a simultaneous major metabolic change (immunity to heavy metal toxicity) and a major change in reproductive strategy. What plausible scenarios remain other than a Goldschmidtian homeotic mutation to a single founder plant? Heresy!


19 thoughts on “Open discussion

  1. Matt, this is even more interesting, because the manzanita that I studied (A. hispidula) was in fact serpentine tolererant, and also pretty rare (Siskiyou mountains), although not strictly serpentine obligate because you’d sometimes find it on apparently normal soil as well (but it never competed well in those situations).

    “What plausible scenarios remain other than a Goldschmidtian homeotic mutation to a single founder plant? Heresy!”

    Well I’ll throw out an idea off the top of my head for sake of discussion, not claiming anything beyond hand waving. I don’t know whether any existing chromosome count and/or C-index data support this for Streptanthus, but as far as the reproductive isolation goes, I wonder if you could maybe have an instantaneous reproductive isolation produced by allo-polyploidization, as is so common in plants. If this were followed by a strong selection pressure for obligate self-pollination, driven by the need to avoid pollen waste (assuming hybrids with either parent species are +/- sterile, and hence a reproductive waste), then I can see a potential isolating process other than homeotic point mutation. How the metal tolerance could arise I have no idea though, but I thought I remembered hearing that polyploids often show a greater stress tolerance than do their diploid parents. Arthur Kruckeberg might well have delved into these issues, I can’t remember.

    The Evolutionary Consequences of Polyploidy is a good review article discussing these and other issues.

    • Jim,
      I have spent a few vacations botanizing the Siskiyous, in fact my first trip there was to the type location for Cupressus modocensis. That was West Fork Seiad Creek, which follows a fault line between cypress on peridotite and Douglas Fir on the country rock. I saw the same pattern you saw. Cypress seedlings would appear in the margin of the fir forest, but I was unable to locate healthy saplings or mature trees in there. They were quickly shaded out. In contrast, I was never able to find a single individual Lewisia cotyledon on anything but ultramafics, even above tree line where shade was not a factor. The same was true for the other obligate serpentine forbs, although they can certainly grow in glacial till in my garden north of Seattle if I protect them from winter moisture.
      My ideas about founder plants on serpentine involve numerous concepts, some of which are somewhat new in the literature. Others date back to the classic works of Kruckeberg (I have all his books and I have even met him!). My basic claim is that the specific circumstances of these plant groups make alternative explanations extremely unlikely to the point of being implausible. Thus an explanation such as allo-polyploidy can adequately explain a particular aspect of the circumstances, but ultimately becomes untenable or unparsimonious based upon other factors. Developing an argument this way is a big hill to climb and requires a very unlikely set of circumstances. In this case the circumstances are two traits that lack viable sympatric pathways, heavy metal tolerance and obligate selfing.
      I should add enough detail about the concepts to justify my bold conclusion. I will start with the concept of reproductive isolation as a facultative adaptive tool for establishment on inhospitable substrates. (Note that this is a bit more generalized than the claims I am making about Streptanthus.) This paper (not paywalled) summarizes recent work in this area and describes how the mechanism works in plants with reproductive plasticity:
      There is, however, an important difference. The plants described in the paper show reproductive plasticity within variable populations, including individual plants with different types of flowers. The species in Streptanthus that interest me instead have closed carpels on all of the flowers on all of the individuals. So while the concept of exploiting reproductive isolation applies, the bit about competitive selection between phenotypes with differing reproductive strategies does not.
      I will touch again on Goldschmidt, but I wanted to make one point here. When Goldschmidt developed his ideas, the understanding of the homeotic processes was in its infancy. I’m not sure whether the specific homeotic genes had been identified, I would have to go back and read his book again. At any rate, I give him the benefit of the doubt and assume that when he invoked “homeotic,” he was referring to any mechanism that could have a homeotic effect. This could perhaps include any mutation that results in profound changes in gene expression. I do not know enough to say whether that limits the possible mechanisms to point mutations in the homeotic genes or not.

    • Anybody who spends their vacations botanizing in the Siskiyous ranks very high on the list of people I’d want to talk to Matt. But now you’ve done it, you’ve got me reading Kruckeberg’s old papers, not to mention Antonovics’, and mating system evolution, and the Jepson Manual, and the geologic origins and chemistry of the CA ultramafics, and associated mercury mining during the gold rush, among other things. In other words, exactly what I needed 🙂

      Anyway, I agree with your points about likely explanations for this phenomenon, was just throwing an idea out there. Have you seen this one?

      With respect to Cupressus (or Hesperocyparis) and your original question, you’re wondering how a non-selfing conifer genus could display some of the same biogeographic traits as herbaceous annuals like Streptanthus, particularly with respect to serpentine endemism, yet have no ability to protect a favorable genotype by high selfing rates, is that the basic question?

      And which Streptanthus and Hesperolinon were you looking at btw?

  2. Jim,
    Glad to hear that you find these topics stimulating. I am looking forward to revisiting the literature in the coming months, I have a lot of catching up to do. Starting with cladistics! I actually had to go read the Wiki page to remind myself about the new names. The taxon of interest in Seiad creek was of course Hesperocyparis bakeri. I will have to be more careful about things I think I remember.
    I suppose the group is now “false cypress,” but if you don’t mind I will continue to use the common and established epithet because it is shorter to type.

    Yes, you summed up my interest in California cypress distribution quite succinctly. Cladistics supports more distinct species of Hesperocyparis in California than species of true cypress throughout the entire Old World. Most of the California species are rare, obligate or nearly so on serpentine, or both. In short, the same pattern we see with selfing forbs. What is not quite clear is the rate of neo-endemism. The odd distribution led Elna Bakker (a popular writer rather than an academic, but an authority nevertheless) to hypothesize an “age of Cypresses,” which is plausible based on paleo work but supports a conclusion that the extant populations are paleo-endemics. If they are neo-endemics, then there is the problem of isolating the serpentine-adapted mutation in a wind-pollinated conifer, as you state. But it is harder still to make them work as paleo-endemics. It would be plausible to assume that a species could be bodenvag at some time in the past but became obligate to serpentine later due to competition…but so many species following exactly the same path?

    The 2005 paper you linked with Kruckeberg as co-author is new to me and definitely rekindles my interest in the general topics. One part in particular comes very close to my own ideas:

    “Boyd&Martens (1998b) offer three theories to explain the “preadaptive” nature of nonserpentine populations to serpentine conditions: (a) high rates of gene flow from serpentine to nonserpentine populations bring serpentine tolerance alleles into the latter population, (b) a constitutive serpentine adaptive trait presents little or no cost to a plant, or (c) a serpentine adaptive trait is adaptive for more than one function.”

    I have always struggled with Kruckeberg’s ideas about pre-adaptation as a blanket explanation because I am not sure that it qualifies as a “trait” distinguishable from genome plasticity, but I will set that aside for the time being. What really intrigues me is (c), although I am looking at it differently than the authors. After years of reading on these topics, I developed an idea that I called “the law of two” (although it is more precisely stated “the law of two or more”). This is the idea that a mutation that results in an adaptive change will not result in speciation, only variation, but a mutation that results in two or more adaptive changes can. As I delved deeper into speciation, I discovered that this basic concept is very similar to Sewall Wright’s ideas about the adaptive landscape, and combines those ideas with Goldschmidt’s ideas about the differences between variation within a species, and the appearance of new species. This leads directly to the following hypothesis: the explosion of neo-endemism on serpentine is directly attributable to a speciation pathway in which serpentine tolerance and an additional adaptive trait co-occur. In selfing forbs, the second adaptive trait is instantaneous reproductive isolation (I realize that reproductive isolation via closing the carpel stretches the definition of “adaptive,” but it would be equally facultative if the flower morphology changed such that a different pollinator took over).

    Rather than point you to specific species in the selfing forbs, I suggest that you skim the Jepson manual on Hesperolinon. Note how much of the diversity is obligate on serpentine and rare. My next post will discuss this genus and explain why I think these selfing plants are our best hope for applying reductionist logic to speciation pathways.

  3. Matt, thanks for that and I hope to get time to continue this. I read the Dietrich paper and it was helpful.

    Based on various lines of evidence, I really think that adaptation to serpentine does not likely involve any homeotic-type mutations, in probably any CA taxon. The ability to tolerate serpentine conditions (high Mg++, Ni++, low CA++, etc), and/or various genes known from molecular analysis to be involved in cell cation exclusion, or metal binding processes, are too widely distributed across too many very unrelated taxa (ferns to angios to gymnos, at least), indicating an ancient origin. I think it’s much more likely a gene selection/expression and/or pleiotropy issue, and that a Goldschmidt-type process is not relevant here, and almost certainly not responsible for both serpentine tolerance and conversion to selfing simulataneously.

  4. Jim,
    This is a very reasonable challenge to an idea that would find very little mainstream acceptance. As I wrote earlier, I have to be careful about genetics, I have spent far more time on population dynamics. Specifically, I am not sure how a pleiotropic mutation would be substantially different from a homeotic mutation in this case. For clarity I will insert a basic definition of homeotic: “relating to, caused by, or being a gene producing a usually major shift in the developmental fate of an organ or body part.”
    In obligate selfing plants with a closed carpel, it is very likely that they evolved from plants with an open carpel. Since the purpose of a flower is to attract a pollinator, the flower structure development has basically collapsed (I can envision no gradual closing of the carpel via competitive selection). This seems to fit the basic definition of “homeotic” above.
    Cell cation exclusion/metal binding is the other half of the question. What sort of mutation can cause a major change in that? I’m not sure science has answered that question. Many serpentine plants (but not all, as with most characteristics) show decreased vigor both on and off serpentine when compared to closely related taxa. It seems to me that the collapse of a significant but non-essential metabolic process could both reduce vigor and also result in cation exclusion (just picking one for simplicity). So now we have two choices: either the two changes (closed carpel and cation exclusion) occurred independently, in which case we have a very perplexing problem of population dynamics in our carefully chosen group, or they occurred simultaneously. If they occurred simultaneously, it is either one mutation or two independent mutations. If it is one mutation that causes a collapse in both flower structure and a metabolic process…that sounds homeotic to me. If it is two independent mutations that co-occurred spontaneously, then you truly have a hopeful monster and near certainty that it happened in a single founder plant!
    BTW, since the meristem is the basic reproductive unit in a plant, it is conceivable that a mutation in a meristem could produce a small population of identical mutant plants that cross the serpentine boundary. With obligate selfing this does not change much about the origin except that the founder plant (or more accurately the founder meristem) could occur adjacent to serpentine rather than on it.

    • You were right Matt, this topic is fascinating–I’d give it 100% attention if I could–and it’s forcing me to remember things long forgotten, which is a good thing. Thanks for the defns and other focusings of the discussion so as to keep from talking past each other.

      I don’t think, based on my current understanding, that any major mutation would be required to confer metal and/or cation tolerance, but rather just changes in selection intensity and/or gene regulation of existing genes/alleles. For example, over-expression and slight modification of zinc finger genes, which bind zinc. But I have more learning to do on that. We might be able to make use of the fact that Streptanthus, being a mustard, is thereby related to Arabidopsis, the plant mobio superhero with it’s fully sequenced genome.

      Hang on.

  5. Jim wrote:
    “I don’t think, based on my current understanding, that any major mutation would be required to confer metal and/or cation tolerance, but rather just changes in selection intensity and/or gene regulation of existing genes/alleles.”
    That is the counterargument, but the challenge is to make the population dynamics work that way. The snapshot in time that we get is of plants that are fully adapted to serpentine and unable to hybridize due to obligate selfing. When you refer to selection intensity, you can’t mean on serpentine, since the tolerance would already be fixed in all phenotypes. You can’t mean entirely off serpentine either, as there is no selection pressure to adapt to heavy metals. So you are left with colonization across the boundary. (As an aside, I know of no observed situations where plants appear to be gradually colonizing across an abrupt serpentine boundary, nor are there instances where plants appear to be gradually developing tolerance on soils that are partly derived from serpentine, but that is a bit tangential to this argument.) As we discussed earlier, obligate selfing is very unlikely to develop in a species that is already fully adapted to serpentine. So when an individual plant crosses the boundary, it can no longer be subject to selection pressure since the genome is already both adequately adapted to heavy metals and fixed by obligate selfing.
    I would even be content with an arm-waving explanation, but I just don’t see how this works under any selection pressure.
    I like your idea about Arabidopsis, this can never really be resolved without an understanding of the genetics, and that is probably beyond me. This paper describes work on Caulanthus, which is very closely related to Streptanthus:
    And of course, I like this quote:
    “…the acquisition of nickel tolerance may involve genetic changes in only a few genes that have major phenotypic effects and supports our hypothesis based on phylogenetic relationships that the evolution of serpentine tolerance in C. amplexicaulis var. barbarae and related Streptanthus species may have been rapid and involved critical changes in only a small number of genes (Pepper and Norwood, 2001).”

  6. “When you refer to selection intensity, you can’t mean on serpentine, since the tolerance would already be fixed in all phenotypes.”

    Yes, selection on serpentine soil. Tolerance thereof is almost certainly a quantitative trait (either inherently multi-genic, or a single gene but subject to high regulation of expression, or both), and moreover I’d guess that tolerance of any single trait of the serpentine tolerance syndrome (i.e. tolerance of low Ca++ and soil water, and high Ni++, Mg++, temperature, etc) is also by itself, quantitative. Probably don’t want to think in terms of simple inheritance patterns on this.

    The other question, of how and when selfing would have originated relative to soil tolerance, is a totally open question to me, but I can’t conceive of how this would originate before the soil tolerance is in place, because we always have to remember that going to an obligate selfing system is highly problematic due to the load of deleterious mutations scattered throughout the genome. There has to be some decided selective advantage that outweighs the inbreeding depression that will typically result.

  7. Goldschmidt argued that variation within a species can lead to selection that improves fitness and facilitates adaptation to a changing environment, but that this mechanism does not lead to speciation. These neo-endemic plant species do show considerable genetic and phenotypic variation on serpentine, even to the extent that obligate selfing is never quite obligate at the population level. In Goldschmidt’s rather radical interpretation, fruit fly experiments can tell you plenty about this type of intra-specific selection and the refinement of traits, but it can tell you nothing about speciation. One species does not gradually diverge into two via sympatric selection. FWIW, I think Goldschmidt went too far, and so does nearly everyone else, but I suspect he may have been more right than wrong. One of Stephen Jay Gould’s later essays was an attempt to rehabilitate Goldschmidt in the technical literature based upon precisely this point.
    My primary hypothesis regarding these plants is that the only plausible explanation for how the new species form is from a single founder plant, either a mutant plant/meristem off serpentine that produces a single crop of tolerant seeds, or a single mutant seed on serpentine. This is the only plausible mechanism because I cannot work out a competitive selection process for tolerance, nor can I work through the establishment of obligate selfing that way. This is based on considerable work on population dynamics and speciation theory, not so much on genetics (so genetics becomes the proving ground unless a hole can be found in my logic). Because of my ignorance of genetics, I cannot really sustain an argument about homeotic mutation any more than I have already, and I am more interested in the founder effect anyway, which is equally controversial.
    So I guess my point is that the selection you are talking about on serpentine is entirely reasonable for improving fitness, but does not answer the question about whether the species originated from a selection process or a founder plant. To argue that speciation occurs without selection, as Goldschmidt did and I am doing, remains highly heretical.

  8. There are several uncertainties for me here. One goes back to your earlier statement that we have a “snapshot in time” of species fully adapted to serpentine. I’m not sure what you feel is the evidence for that–paleobotanical of some type? To me, how fast serpentine tolerance might have evolved in various taxa in the past, and also whether they are paleo- or neo-endemic, is completely unknown, but I’m no expert on the topic either.

    I think there are some pieces of evidence that argue against the importance of obligate selfing in serpentine tolerance. The most obvious takes us back to your original mention of Cupressus, in which that’s not even possible (and same for Calocedrus and Pinus jeffreyi, which also show tolerance). But there are some angios on serpentine that are not obligate selfers either.

    Selfing helps the most when an advantageous, mutant allele is recessive, and in direct proportion to how far below 0.25 the frequency of that allele is in the population, by increasing the number of homozygous recessives in the population. A dominant advantageous allele is not helped as much by selfing, because heterozygotes will always express the advantageous trait, and hence be selected for. But I don’t know anything about the expression pattern of the alleles of the genes for the several traits involved in serpentine tolerance, which at least in California where it’s hot and dry for 5 months, must involve both chemical and heat/drought tolerance, at least for perennial plants that have to withstand the summer. [The annuals can avoid the heat phenologically, but they still have to tolerate the metals and the Mg/Ca ratio.]

    My complete hand wave would be that advantageous alleles might well be dominant though, in the sense that they cause an over-expression of some protein already necessary for other functions, such as e.g. over-expression of zinc finger genes to bind up nickel, and perhaps something similar for chromium and iron (chlorophyll a/b binding protein?). A large part of adaptation and selection likely involves slight modifications of protein structure and changes in expression levels, to fulfill some new function, or even just outright pleiotropy without modification. And so it also might not take much to over-express heat shock proteins in many CA taxa, which very likely have such genes in their genomes. There should then be a relatively quick selection for individuals having such over-expressed alleles, which would thereby quickly increase their frequency in the population on serpentine, leading towards fixation in the population, without the need for increased selfing necessarily arising.

    But there may be cases where an advantageous allele is in fact recessive, and if accompanied by a mutant that enforces (or just encourages) self pollination somehow (petals never opening or whatever), that might also occur and be an explanation in some cases. Could very well be multiple mechanisms operating across various taxa here. Or even within taxa.

  9. I think my last comment disappeared. I will move on.
    “…how fast serpentine tolerance might have evolved in various taxa in the past, and also whether they are paleo- or neo-endemic, is completely unknown”
    It seems largely unknowable for most tolerant plants, although genetic analysis may eventually shed some light. But the distribution of rare, selfing, tolerant annuals, combined with the recent emergence of their habitat, strongly suggests neo-endemism. Kruckeberg agrees.
    “I think there are some pieces of evidence that argue against the importance of obligate selfing in serpentine tolerance”
    I agree, I do not consider it important beyond enabling reproductive isolation. Isolation suffices to fix a trait no matter how it occurs. But I do think selfing helps tightly constrain the possible pathways to tolerance when it does occur. I should have been more clear about that. My next comment will fully flesh this out.
    Your point about dominant and recessive genes is an important insight that I had not considered. I need to chew on that.
    Let’s look at selfing and tolerance in terms of adaptation, competition, and plasticity. First selfing. There is a perfectly acceptable model as to how it helps fix an advantageous allele, and you laid that out very nicely. If “adaptive” in plants means improving fitness/fecundity in response to changes in prevailing conditions, then selfing is clearly not adaptive. Selfing plants suffer from mutational loading, as you mentioned earlier, and also have reduced ability for further adaptation. Selfing is also not competitive in that it does not improve a plant’s ability to compete for light, space, moisture, or nutrients. It is an example of plasticity, and can be thought of as “enabling.”
    Now tolerance. I argue that the development of heavy metal tolerance is not adaptive in the neo-Darwinian, survival-of-the-fittest sense. While it fits our general idea of what it means to adapt, it is conceptually distinct from adapting to a change in the prevailing conditions (off serpentine in this case) such as an increase in herbivory. It is adapting to an entirely different environment. The development of tolerance is also not competitive, again looking at light, moisture, etc. As with selfing, it increases population size not by outcompeting others, but by establishment in an environment essentially free from competition. This all fits with the well-documented evidence of tolerant plants being generally less vigorous in direct comparison to a similar taxon off serpentine. Tolerance, like selfing, is actually a manifestation of genomic plasticity and can also be considered enabling.
    These descriptions hint at why I think these traits are important for constraining the population dynamics, which will be in my next post. I will address your comments about gene frequency in that post, at least in broad strokes.

    • Just a couple points here for now.
      1. I completely forgot about the relative recent-ness of the serpentine habitat, geologically speaking, which does argue for neo-endemism (but not necessarily for neo-serpentine tolerance).
      2. I made a mistake in my previous comment; the critical population-level value where an advantageous recessive allele exerts no selective pressure towards development of selfing would be 0.50, not 0.25 (assuming a diploid species and thus disomic inheritance). [0.25 would be the critical value of homozygous recessive genotypes in the population]. But the larger point remains, in that the selection pressure to become a selfer is driven by the expression patterns of the alleles at all of the genes in the genome, not just one or a few, so unless a recessive nickel tolerance (or other serp. tolerance trait) allele has an absolute whopper of a positive effect, it’s not likely to drive a move towards selfing, not by itself anyway. And moreover, I have a really hard time conceiving of how favorable but recessive alleles could arise, molecularly. That’s a tough nut to crack.
      3. However, I forgot about the other selective force driving a selfing system–the purely demographic one, allowing a single founder plant to establish a population. Given the sparseness of serpentine communities, the low competition levels could permit survival of otherwise really poor genotypes that could never compete off serpentine, and which then purge their genome-wide genetic load steadily over generations, getting steadily more fit.
      4. That Caulanthus nickel tolerance study you linked to earlier might be really helpful here.

    • Been down with the flu…

      I promised a summation, although at this point I think you have worked through everything yourself. It is gratifying to see that this still an active field of inquiry after I was away for so many years. I suspect others sense, as I do, that if we are ever to understand speciation, the transition to serpentine tolerance is perhaps the most promising topic because of the possible pathways that can be excluded.

      I want to mention one more thing. In David Raup’s book Extinction, he discusses the Gambler’s Ruin model in which extinction occurs when a graph of population vs. time bumps into the zero line, at which point the species is extinct. This got me to thinking about this simple graph. Despite many decades of work, I doubt that we could make a reasonable guess as to what an entire graph of this type would look like for any species, even our own! While we can surmise about bottlenecks and things like that, we really know very little about what the beginning of the curve looks like.

      On to the summation:


      Rare, selfing, tolerant species on serpentine in central California are neoendemics.
      Tolerant, selfing species on serpentine evolved from outcrossing, intolerant species off serpentine.
      There is no accepted model for the development of selfing in an outcrossing species other than as a facilitation for genetic isolation, and selfing is strongly associated with edaphic specialization, therefore selfing and tolerance are linked.
      Selfing reduces general fitness due to mutational loading and other factors.
      As a corollary to 3. and again invoking extant population distributions, selfing, tolerance and speciation are linked.

      How can this situation be fitted to the neo-Darwinian synthesis of selection via survival of the fittest? As you correctly parsed out, there is a plausible mechanism for selection, but the selection process is focused entirely on genetic isolation, not competition or adaptation (that is, not adaptation to the conditions experienced by the parent plant off serpentine). The selection mechanism calls for plants of variable tolerance and flower structure. Outcrossing (functional) flowers face back-crossing pressure, while flowers of reduced functionality can better produce tolerant offspring. With time, the functional flowers disappear from the population, changes accumulate as the selfing rate increases, and eventually an obligate selfing, fully tolerant, distinct species forms.

      The problems with this are not fatal, but they are powerful. First of all, the mechanism is not parsimonious because a single founder plant that is both tolerant and has nonfunctional flowers can create a new species simply due to cessation of competition. But the biggest problem arises in trying to place a number other than “1” at the beginning of a population vs. time graph for our newly evolved tolerant, selfing species. The selection mechanism described above requires a continual seeding of semi-tolerant or fully tolerant individuals across the boundary. During this process, obligate selfing must arise. In order for that to happen, obligate selfers must arise that are more closely related to outcrossing individuals off serpentine than to nearby obligate selfing individuals on serpentine. This would mean that a population of selfing plants has multiple, distinct, selfing lineages, each of which is more closely related to an outcrossing parent than they are to each other. I just can’t see how that makes any sense.

      OK, that is the hypothesis in a nutshell. I have left out some of the alternate pathways to speciation via selection that I have worked through to what I think are dead ends, but I am interested in any arguments to the contrary. Now that I have laid this out with single-minded determination, I intend to comment on some of your comments. I find No. 2 in your last comment fascinating but I have to get smarter on genetics to fully comprehend it.

  10. In his book “Geology and Plant Life” Kruckeberg favorably reviews the papers of Bateson and DiMichele as describing plausible mechanisms for abrupt saltational speciation. Unfortunately, both papers are in books that I do not have access to, maybe you do. The papers are:

    Bateman, R.M., DiMichele, W.A., 1994. Saltational evolution of form in vascular plants: a
    neoGoldschmidtian synthesis. In: Ingram, D.S., Hudson, A. (Eds.), Shape and Form in Plants and
    Fungi. Academic Press, London, pp. 63–102.
    Bateman, R.M., DiMichele, W.A., 2002. Generating and filtering major phenotypic novelties:
    neoGoldschmidtian saltation revisited. In: Cronk, Q.C.B., Bateman, R.M., Hawkins, J.A. (Eds.),
    Developmental Genetics and Plant Evolution. Taylor & Francis, London, pp. 109–159.

Have at it

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s