[Taxacom] DNA only descriptions

[Stephen Thorpe]

>In my opinion, the crucial aspect we should not forget is that we use morphology 
>as a proxy for genetic differences

IT DEPENDS WHAT ONE IS DOING! Descriptive taxonomy, or "systematics"? 
Descriptive taxonomists would still describe morphology even if systematists 
knew EVERYTHING about the genetics ...



________________________________
From: Kim van der Linde <kim at kimvdlinde.com>
To: Kenneth Kinman <kennethkinman at webtv.net>
Cc: taxacom at mailman.nhm.ku.edu
Sent: Sat, 21 August, 2010 3:16:00 AM
Subject: Re: [Taxacom] DNA only descriptions

Hi Ken,

I agree with most what you said. There is one item I would like to
elaborate on.

On 8/18/2010 11:55 PM, Kenneth Kinman wrote:
> For many taxa, an appropriate weighing "from all possible sources" is
> still a work in progress, and molecular data is perhaps still often
> given too much weight when such data is still far from complete.
> What you regard as "far more [molecular] data" may not impress
> morphologists of some taxa.  As you said, it has to be weighed on a
> case to case basis, and therefore it is unwise to make any
> generalizations one way or another without reference to the
> particular taxa being discussed.

I think the worst 'solutions' are the cases in which molecular and
morphological data are lumped in a single analysis. Those generally tend
to be dominated by the molecular data as it is often numerically
outweighing the morphological data.

In the drosophilid taxonomy, we have two competing morphological studies
resulting in substantially different trees. The first phylogeny
(Throckmorton) is primarily based on internal morphological
characteristics, while the second (Grimaldi) scored a huge number of
external traits. Now, 20+ molecular studies and 12 genomes later, we
know rather precise the phylogeny of the genus and 12+ included genera.
  The Throckmorton phylogeny was rather close to the molecular tree,
while the Grimaldi phylogeny was substantially further off. So, there is
an obvious conflict between the Grimaldi tree at one hand, and the
molecular and Throckmorton's trees at the other hand. So, what to do
with the Grimaldi tree? Generally, it is given limited weight, 
especially after it became clear that the cladistic analysis was only 
partial and reanalysis of the data in a more thorough way did not result 
in the same answer.

It is really easy to sit down with a bunch of people, find the worst 
offending molecular articles and use that to dismiss everything. It is 
easy to do the opposite as well.

In my opinion, the crucial aspect we should not forget is that we use 
morphology as a proxy for genetic differences. Species delineation is 
one thing, but taxonomists also group species in genera, tribes, 
families etc. to roughly reflect genetic relatedness (yes, I know, 
paraphyletic taxa, but let's keep the discussion focused). And because 
of that, if there is a conflict between morphology and DNA, the only 
sensible question to ask is what causes this difference. The more we 
study the evolution of traits at the molecular level, the more examples 
we get where large morphological differences are determined by minimal 
genetic differences, like a single point mutation in a regulatory 
element. Do I have to give such a morphological characteristic a lot of 
weight? Of course not.

Similar issues arise with molecular data. We routinely use genes for 
phylogenetic reconstructions, but once those genes are subject to 
directional selection, you can fall in the same pitfalls as with 
morphological traits.

Really, the whole morphology vs molecular discussion is so passe. Sure, 
individuals can keep sitting on their little islands, but reality is 
that the world moves forward. Just producing a molecular phylogeny 
becomes less and less important; many of them are produced in studies 
examining the (molecular) evolution of morphological traits. In most of 
those studies, there is no dispute about what to use as a basis, the 
molecular tree is used to map the morphological traits over, not the 
other way around (sure, I suspect there will be some exceptions, I just 
haven't seen them).

Knowing all this, I do not have a problem with DNA-only descriptions. 
Just as we have to determine the variability in morphological traits 
before we can determine the boundaries between species, we have to do 
the same with DNA circumscriptions.

But I think the most important aspect is that we forget that we make 
*proposals* when we publish either species or a phylogeny. Only when it 
is accepted by our peers, and used accordingly, we know that out 
*proposal* has merits as long as nothing better comes along. Many are 
far too obsessed with wanting to have the final answer, but reality is, 
we are always limited in what we actually know and hence, even the best 
supported hypothesis could one day be superseded by a better one. So, 
instead of fighting about morphology vs molecular, let's all do our job 
with the tools we have at hand and let's keep an open mind to what other 
have to offer as well, even if that means that you might be wrong. In an 
upcoming era in which a genome can be sequenced in one day for less than 
$100, you bet that it is going to dominate the field of systematics. The 
question we should ask ourselves is how do we integrate that the best way.

Kim

_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of these 
methods:

(1) http://taxacom.markmail.org

Or (2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom  
your search terms here