[Taxacom] DNA only descriptions

[Sergio Vargas]

Hi,

> Well, DNA barcoding is exactly developed for doing just that, with the added advantage that you do not need to know all your species.

yes, provided the barcoding system works as it do for (some) insects.
This is an issue in it self for some groups...

> I think this whole discussion is focusing too much on mor. vs mol.

I think the whole discussion is focused on whether one should discard
or omit information. I think most taxonomic researchers are nowadays
more than happy to have DNA sequences backing up (or even in conflict)
with their works. Perhaps I'm too optimistic, but there is no
justification for simply omitting information (molecular,
morphological, behavioral, ecological, etc.) about the organism. That
would be, I think, enough "compelling evidence" to exclude DNA only
descriptions.

>The reality is that collecting DNA sequences is REALLY easy nowadays, while collecting phenotypic data remains a bitch to do
> because most of the time, we cannot automate it (the main problem of genotype-phenotype mapping). So, instead of hanging on to > morphology as the key to identification, we can help this field dramatically forward by adopting DNA for identification purposes and
> spare the limited time of those knowing morphology to the many many morphology based questions beyond species recognition.

I partially agree. You say, collecting DNA sequences is REALLY easy
nowadays. Collecting DNA sequences is REALLY HARD for the majority of
the world. Importing the sequencing chemistry in most "third world"
countries is ridiculously expensive, not to mention blockaded
countries. The same argument could be made in the case of microscopes,
etc. but one microscope is a one time (almost) purchase (investment?)
for sequencing you need the sequencer + an appropriate lab. + the
chemistry (+ PCR, +DNA) every time you sequence, if (a big if
sometimes) you can sequence directly after PCR!

On identification: DNA is useless in some (strikingly) simple cases, a
fish, bird or plant census for instance or coral and coral reef
monitoring. In both cases you cannot or just don't want/need to
collect specimens for the job. Many common species of trees in a
tropical rain forest are easily determined to species level after
rather short training (1 undergrad semester), and some workers at many
biological stations (e.g. La Selva, Costa Rica) can get to know the
flora of a particular area better than most botanists. So... yes, DNA
for identification is in some cases faster and easier and in some
others (I would say a big majority) amazingly slow and expensive
(collect a leaf of a 50m tall tree in the middle of nowhere for DNA
can be really expensive). In addition, identification is not
description, and if wrongly implemented or applied may not even lead
to the discovery of putative new species.

> I am waiting for the day that we have hand-held sequencers that can be taken into the field to do a on the spot DNA identification of > your plant, insect, etc.

keep waiting, perhaps I am too pessimistic this time. I am still
waiting for the day "water-proof" telemetry devices don't get ruined
after (just) 6 hours of (real) rain or the day that the newly super
cool GPS actually works under 60m tall trees.

cheers

sergio

-- 
Sergio Vargas R., M.Sc.
Molecular Geo- and Palaeobiology Lab.
Dept. of Geo- & Environmental Sciences, Palaeontology and Geobiology
LMU München
Richard-Wagner Str. 10
80333 Munich, Germany
tel. +49 89 2180 17929
s.vargas at lrz.uni-muenchen.de
sergio.vargasr at ecci.ucr.ac.cr