Silphidae on Tree of Life
jgrehan at SCIENCEBUFF.ORG
Fri Feb 11 22:43:02 CST 2005
I did respond on the question of geographic information and there was a response posted - basically to the effect that geography was not a priority for the research.
On the molecular-morphology question, I guess I've staked out my positon so there is not really anything I could have added to the posting. Molecular analysis is popular because it is easy, and because there is the widespread assumtion that it is sufficient on its own to delineate phylogeny, particuarly if one takes the view that it is cladistic in the same sense of morphology (in which case molecular trees will be shown to be better supported since there are more characters as Zander points out, and that is why glaring anomolies in morphology such as the human-orangutan connection can simply be ignored).
From: Taxacom Discussion List on behalf of Steve Manning
Sent: Fri 2/11/2005 5:17 PM
To: TAXACOM at LISTSERV.NHM.KU.EDU
Subject: Re: [TAXACOM] Silphidae on Tree of Life
Thanks for the clarification. I am mildly surprised that Derek's post
didn't elicit a response from John Grehan since he is the one who requested
the geographical inputs in the first place and has been strong in his posts
about favoring morphology at least as much as molecular phylogenies!
P. S. Time AND money (and equipment)(and motivation)(and
expertise)? Wow! What a concept!
At 09:32 AM 2/11/2005 -0600, Richard.Zander at MOBOT.ORG wrote:
>As I understand it, Steve, certain genes and other molecular groups have
>particular rates of mutation. Some genes are very variable and are useful at
>lower taxonomic levels, and others are not, and are used to analyze
>phylogenetic relationships at higher levels.
>The sampling has not been great, but from what has been done, particular
>genes have the same sequence, for, say, all samples of a species. So single
>samples are commonly used as representative of particular taxa. This is not
>great but because of the observed regularity, funds and time have been
>focused on large scale analyses with exemplar taxa. If you have the funds
>and time to determine more exactly how well single samples represent entire
>taxa, there are a whole lot of us out here who are interested.
>Yes, molecular study is informed by morphological study. The fact that the
>results of molecular study and morphological study generally concur is no
>longer amazing, so molecular study is now expected (by me, anyway) to solve
>knotty problems and resolve contradictions, disappointing so far. In spite
>of the paper by Lee in Taxon 53(4): 1019-1022, 2005, morphological traits
>are too few to be impressive statistically, and this is one of the reasons
>molecularists tend to discount morphological studies when they contradict
>well-supported molecular results. If you have the funds and time to
>investigate how to interpret morphological characters such that cladograms
>based on morphology generate high bootstrap or posterior probabilities,
>there are a whole lot of us out here who are interested.
>Oh, the Lee paper . . . the argument is that morphological data already has
>all non-relevant data deleted, so there are as many traits (about) useful
>for phylogenetic analysis as in molecular studies. True, but probably not a
>relevant argument, since the number of morphological traits needed to
>uniquely define 10 taxa is 11. The branch length is not 10.
>Convergence afflicts morphology, as it does, doubtless, exons. This hasn't
>been dealt with well, and if you have the funds and time . . . . . . etc.
>Richard H. Zander
>Missouri Botanical Garden
>PO Box 299
>St. Louis, MO 63166-0299
>richard.zander at mobot.org <mailto:richard.zander at mobot.org>
>Bryophyte Volumes of Flora of North America:
>Shipping address for UPS, etc.:
>Missouri Botanical Garden
>4344 Shaw Blvd.
>St. Louis, MO 63110
>From: Steve Manning [mailto:sdmanning at ASUB.EDU]
>Sent: Thursday, February 10, 2005 5:14 PM
>To: TAXACOM at LISTSERV.NHM.KU.EDU
>Subject: Re: [TAXACOM] Silphidae on Tree of Life
>I was surprised by the latter part of this. Are you saying that the
>samples taken for molecular phylogenies are not from specimens identified
>as to geographic location from which the specimen was taken? Presumably
>the alpha taxonomy could have been done in the past rather than along with
>the sampling (?) I would hope that molecular phylogenies are based on a
>reasonably thorough sampling of the variation, including geographic
>localities, within each taxon(?)
>In other words I guess what I am really trying to say is that alpha
>taxonomy results in geographic distribution information only after the
>info. is gathered and the classification is made, and I don't see why
>molecular phylogenies should be fundamentally different in this regard. In
>either case, incomplete geographic info. seems better than none at all,
>assuming it is relatively easy to upgrade as new info. is brought to your
>Are molecular phylogenies accepted for publication in the absence of
>reference to voucher specimens, including localities, from which the
>molecules are extracted?
Dr. Steve Manning
Arkansas State University--Beebe
Mathematics and Science
Professor of Biology
P.O. Box 1000
Beebe, AR 72012
More information about the Taxacom