In message . com,
Raphanus writes

Stewart Robert Hinsley wrote:
In message .com,

...such processes include hybrid
sterility and hybrid breakdown.

Thanks. I think I understand the above. Now that I think about it,
this sort of had to be answer. The "mule" problem. I'm sure this is
Botany 101 and I appreciate you taking the time to give me a lesson.


No, unless you can find a convenient locus. Gene trees are not perfectly
correlated with species trees, and closely related species may not be
separated absolutely by DNA sequences.

I get some of this - but not all. I'll make a stab at asking a further
question. If closely related species are not separately absolutely by
DNA sequences - what are they separated by? I thought that with
respect to speciation DNA determined everything. E.g., the genome of
the Neanderthal and the genome of the human are different. I'm sure my
ignorance and confusion is showing. Could I be using "DNA analysis"
when I mean "genome determination?" Thanks in advance.


Only a few loci may be involved in speciation. In the Loxia curvirostra
complex (crossbills), speciation is associated with changes in beak
shape and size (adaptation to feeding on different types of conifer
seed) and calls (species identificiation). Speciation is so recent that
at other loci, including those typically used for studying phylogenies,
show pretty much the same variation in each species.

For a botanical example, ITS/ETS doesn't seem to fully resolve Sidalcea.

If you look at the whole genome then I'd guess that phylogenies would be
pretty much unambiguous, at least at the level of flowering plants. But
there'd still be the problem in identifying which populations represent
species.

Mimulus cupriphilus is restricted to the workings of a couple of
Californian copper mines. It appears to be a recent derivative of
Mimulus guttatus (IIR the parent species C). On a phylogeny, for most
loci, it's probably nested within M. guttatus, and just by looking at
the DNA sequence you couldn't tell it was a new species - you'd also
need to be able to predict the phenotype from the genome.

For another example of a potential pitfall, the populations of Gossypium
aridum found in one Mexican state have had their chloroplasts and about
3% of nuclear loci replaced by those from a relatively distantly related
species from Baja California. If you'd looked only at the cpDNA you'd
have drawn the wrong conclusions about the taxonomic affinities of those
populations.
--
Stewart Robert Hinsley