scientific omissions and published negative evidence

Harvey E. Ballard, Jr. ballardh at OAK.CATS.OHIOU.EDU
Wed Oct 1 08:49:18 CDT 1997

Getting on this other track I see opening up (to veer off the=
 Science-Religion debate), and onto scientists admitting error or at least=
 reporting "negative evidence"...

I have talked personally with dozens of investigators who found little=
 phylogenetically useful sequence variation in certain gene regions that=
 have been touted to do so for certain groups in published papers.  I am=
 thinking specifically of the Internal Transcribed Spacer region of nuclear=
 ribosomal DNA, and a number of chloroplast introns and intergenic spacer=
 regions (notably the much touted trnT-trnF region).  "Negative" evidence=
 that these regions show insufficient levels of divergence within a great=
 many genera has yet to be captured in a collaborative paper to warn=
 researchers against investing substantial funds for sequence data for=
 numerous taxa in an all-out attempt.  My impression from talking with folks=
 whose research spans a diversity of vascular plant genera, is that regions=
 routinely mentioned in the recent literature as having great phylogenetic=
 utility in fact don't work in the majority of attempts, but that=
 these--very costly--exercises have yet to be published in a composite=
 manuscript.  Where variation has been low, a few researchers have used a=
 number of gene regions in combination to get sufficient variation to=
 discern reasonably well supported clades.  But more than a few researchers=
 I know have trusted the PUBLISHED literature as indicative of a=
 "fool-proof" region for phylogenetic reconstruction and have spent=
 thousands of dollars in an all-out attempt with many taxa, holding a=
 handful of synapomorphies to show for the effort.

I would be interested to hear from editors of nationally recognized journals=
 about whether, generally, they would accept manuscripts detailing negative=
 evidence of this sort--"lack" of phylogenetic information and poor=
 systematic utility of some now well known gene regions for many vascular=
 plant groups.  Such "warning" papers would, I believe, caution some of the=
 many over-zealous researchers who only see papers published on the UTILITY=
 of certain types of data and not others balancing this perspective.  (The=
 only recent example of presenting negative evidence--non-utility of the ITS=
 region--that I can think of is a 1993 paper by P. Soltis and Kruzoff in=
 Molecular Phylogenetics and Evolution (vol. 2), showing no=
 intra-populational variation and virtually nothing useful between the two=
 (morphologically very different) species of Lomatium.

I vaguely recall a thread like this several months ago, but I didn't catch=
 later whether anyone followed up on collating responses on such "negative=
 results" from folks.

Harvey Ballard


Harvey E. Ballard, Jr., Assistant Professor

Department of Environmental and Plant Biology

Porter Hall

Ohio University

Athens, OH 45701

(614) 593-4659 (office & lab phone)

(614) 593-1130 (fax)

<bold>NOTE: After 7 November, area code =3D 740

</bold>email: ballardh at

More information about the Taxacom mailing list