[Taxacom] Diagnosing species

[Richard Pyle]

>      I agree to an extent.  SUBspeciation in particular is 
> probably usually a long gradual process (the early stages 
> that can eventually lead to full reproductive isolation).  
> However, I also suspect that full speciation is *fairly 
> often* a relatively short-term (even violent) occurrence due 
> to rapid climatic change, geological disasters, disease, or 
> other events that separate or even bottleneck two subspecies 
> into geographically isolated remnants.  The founder effect 
> can thrust one or both remnants into a fast-track toward 
> reproductive isolation from each other. 

I certainly agree that it *can* happen quickly (technically, I suppose,
within a single reproductive event where siblings are able to cross, but are
otherwise unable to reproduce with others).  But I wonder: how often does it
happen via these sorts of "abrupt" mechanisms? And, when we say "abrupt",
how many generations are we talking about?  

Let's look at founder effect.  Sure, you get a genetic bottleneck, but how
many generations do you reckon are needed before barriers to reproduction
creep in? Suppose male individual "A" and female individual "B" are
vicariantly isolated from the rest of their population, and they get
together and produce offspring.  That first generation of offspring would
presumably be the same species as the original population, just as they
would have been in the event "A" and "B" had reproduced in the absence of a
vicariant event.  Perhaps in-breeding can amplify the effect such that it
would only take a few more generations before the entire new population is
reproductively isolated from the entire old population.  

But I wonder how often this rather extreme founder effect really happens
across the scope of biodiversity.  I suspect in the majority of a
founder-populations begin with much more than two individuals (my thinking
here is that the average probability of extinction in such cases is rather
high, otherwise).  

> If the recovering populations are able to come back together 
> fast enough, gene flow might easily resume (speciation hasn't 
> occurred), and it might take a second Ice Age to finish the 
> job of speciation for good.  I think animals tend to speciate 
> easier and more cleanly.  Plants tend to reestablish gene 
> flow more easily, thus making speciation a "messier" 
> fuzzier process.  

I guess it depends on which animals you're talking about.  Corals seem to be
more like plants.  I have a hunch (not much more than that, at this point)
that we'll eventually come to recognize this sort of "messiness" among reef
fishes.  But of course, vertebrates are the exception, not the rule.  In the
world of animals, it seems to be really all about insects, when it comes to
speciation.

>     Anyway, the younger of the two former subspecies (farther 
> from where the species originated) is probably more likely to 
> change more, and it can be regarded as a daughter species, 
> just like a population that colonizes an island.  

Hmmm....not sure I understand your use of "younger" in this context.  If
subpobulation "A" fractions off of major population "B", then they are
equally removed from each other in terms of time.  The fractional population
may be more apt to fix mutations due to lower genetic "inertia", but this
isn't a function of "age" ("younger"); it's a function of population size.
The genomes of the individual organisms in the fractioned population are
exactly as old as the genomes in the original population -- and, indeed,
exactly as old as essentially all life on Earth -- about 3-4 billion years,
give or take.

> True 
> dichotomies producing "sister" species are probably rare 
> events.  

Hmmm...how are you defining "sister" species here?  In terms of relative
population sizes at the initial isolating event?

> That's one reason I maintain that speciation and 
> evolution in general is mainly a paraphyletic process.  In 
> summary, there is a very wide variation in the fuzziness of 
> species, depending on their internal cohesiveness (very 
> strong in some plants) and the vagaries of external forces.  
> But Darwinian gradualism (population expansion and 
> subspeciation) is a longer process that operates during the 
> equilibrium phases of punctuated-equilibria.  Speciations 
> tend to mainly occur in bunches after punctuations, even 
> among the fuzziest of plant species.  Therefore, those 
> punctuations would be the singular events we can mark on a 
> historical timeline, whether that event was instantaneous 
> (massive asteroid impact) or climate changes that take 
> decades or centuries to cut a species into two remnants.  
> Even the latter is geologically fast.

Well, you may be right -- and perhaps, if we're lucky, we'll be able to
extract information concerning these punctuations at the species level from
whole- (or large-) genome sequencing.  I'm still skeptical, however, that we
will ever be at the point where we can ascertain whether or not two
individual organisms (or populations of organisms) have crossed the magic
threshold into species-dom in more than a trivial number of cases (let along
majority of cases, and certainly not all of the cases).  Thus, species will
*still* be what consumers of nomenclature say they are (by definition); it's
just that we may find ourselves arguing about it less and less.

Aloha,
Rich