[Taxacom] Animaltaxacom...

[Curtis Clark]

...might as well be the name of this list, based on some recent posts 
about species boundaries.

It is not uncommon among angiosperms for morphologically dissimilar 
forms to interbreed when sympatric, and form fertile offspring. In some 
cases, all the plants in one locality may have inferred (or in some 
cases demonstrated) hybrid origin, but in another locality the parent 
species may be sympatric with few hybrids, and when allopatric they are 
true to diagnosis.

Some botanists have tried to shoehorn this into the BSC (perhaps most 
notable being Verne Grant). In the BSC view, all the interbreeding forms 
are "semispecies", and the whole group, a "syngameon", is the biological 
species.

Nevertheless, these "semispecies" are, exclusive of their hybrids, often 
easily distinguished by morphology and site preference. If the hybrids 
were not perceived (*independent of whether they could actually form*), 
no one would question that the parents are separate species. And in some 
cases the parent species and even the hybrids have fossil records 
extending back tens of millions of years.

To subsume these vast collections into single biological species is 
effectively to say that phenotype is unimportant. And if in fact there 
substantial genetic interchange, and yet phenotypes remain distinct, 
this would certainly be a noteworthy mystery of evolution.

What appears to be happening, though (based on admittedly limited 
studies), is that the hybrids are most often evolutionary dead ends; 
that there is little if any gene flow between the parent species.

The broadest (and weakest) evidence is that the parent species maintain 
their diagnostic features over time. Were there wholesale gene flow 
between the parents, this would imply that natural selection can 
maintain a phenotype independent of its genetic basis.

In my research, I delved deeper into a single genus (/Encelia/ of the 
Asteraceae), where there are two independent lines of evidence that show 
little or no gene flow between the parents:

1) it is not unusual for two species to form F1 hybrids in areas of 
sympatry, but backcrosses (as judged by morphology) are rare. Progeny 
tests of hybrid individuals in the same area show a full range of 
backcrosses and F2s. None of this variation shows up as mature plants. 
Our conclusion was that strong selection eliminated these individuals.

2) In broad pattern, several lines of molecular evidence give the same 
cladogram as morphology. The genus appears to be recently evolved, so 
it's not surprising that the molecular tree is inconclusive out toward 
the tips, but hybridization occurs even between major clades. Most 
salient is that two diploid species appear to have arisen from hybrids, 
and in both cases their ribosomal internal transcribed spacers (ITS) are 
chimeric, but with these species removed from the analysis, the rest 
don't show clear evidence of hybridization. Thus, we would expect to see 
a molecular signature of widespread hybridization, but we don't.

Any astute observer would say that this doesn't negate the BSC; that the 
selection against backcrosses is just another breeding barrier. I don't 
disagree. But it renders unreliable any of the commonly used "tests of 
sympatry": getting fertile offspring is uninformative.

I have a lot more to say about speciation, but since the bulk of current 
taxacom denizens seem to believe that species aren't real, I don't want 
to push my mythology.


-- 
Curtis Clark                  http://www.csupomona.edu/~jcclark/
Director, I&IT Web Development                   +1 909 979 6371
University Web Coordinator, Cal Poly Pomona