[Taxacom] Diagnosing species

[Bob Mesibov]

Richard Zander's recent TAXACOM posts have pointed to regarding species as 
diagnosable entities for taxonomic purposes, rather than as historical 
entities. By "historical" I mean "phylogenetic" as understood by most 
TAXACOM readers: a species as a single branch on a dichotomous tree of life. 
Zander's diagnosis would include a range of characters not commonly thought 
of as species-defining.

At first glance this suggestion seems to be an abandonment of the Darwinian 
goal of a taxonomic system which accurately reflects the genealogical 
history of life.

At second glance, it does no such thing. Botanists like Zander are aware 
that a very large proportion of plants have hybrid origins, and they have no 
reason to think that hybridisation hasn't been going on for a very long 
time. Even if hybridisation was less frequent among the ancestors of today's 
plants than it was in recent times, the multitude of hybrid ancestors that 
must have existed should encourage botanists to throw strictly dichotomous 
trees on the No Longer Useful heap. The genealogical history of plants is a 
network, and "reconciling" incongruent gene trees into a dichotomous 
structure is not unlike a Ptolemian trying to "save the circles" to keep the 
Earth at the centre of the Universe. It perpetuates a falsehood.

Defining genera as historical entities makes just as little sense. The 
reality is that depending on which gene tree you look at, a plant species 
could logically belong to several historical genera at one and the same 
time. Genera also need diagnosing.

Microbiologists are very comfortable with this approach to taxonomy, even if 
many botanists are still not. Most zoologists don't want to think this way 
at all, even though they would admit they come from a hybrid ancestor, i.e. 
the first eukaryotic cell. They prefer to think that animal evolution has 
been largely dichotomous since the earliest metazoans appeared, and if gene 
exchange happens at all, it occurs mainly during slow sympatric or 
parapatric speciation, and less frequently afterwards, as reproductive 
isolation gradually becomes a no-exception policy. (I'm thinking here of 
speciation not driven by chromosomal rearrangements. These isolate animal 
lineages immediately.)

However, I wonder how many zoologists actually test their molecular 
phylogenies for evidence of hybridisation or introgression, as botanists do 
more and more? If you don't look for signs of hybridisation you won't find 
it, and a successful hybrid species looks just like a (excuse me!) real one.

Getting back to seeing species as diagnosable entities, that seems an 
eminently sensible approach to dealing with a Network of Life. Arguments 
about adhering to strict monophyly vs. allowing some paraphyly could be 
restricted to those bits of the Network where they have some logical 
relevance. (I'm thinking here of higher taxonomic divisions, although I read 
that monophyly has run into serious trouble at the Domain level in recent 
years.)

As for the argument that phylogenetic software doesn't (yet) deal adequately 
with reticulate evolution, so until it does we should continue to believe 
that animal evolution is strictly dichotomous... Hmmmm. Hey, Procrustes, 
maybe you should learn more about your guests before settling them down in 
that bed of yours?
---
Dr Robert Mesibov
Honorary Research Associate, Queen Victoria Museum and Art Gallery
and School of Zoology, University of Tasmania
Home contact: PO Box 101, Penguin, Tasmania, Australia 7316
(03) 64371195; 61 3 64371195

Australian Millipedes Checklist
http://www.qvmag.tas.gov.au/zoology/millipedes/index.html
Tasmanian Multipedes
http://www.qvmag.tas.gov.au/zoology/multipedes/mulintro.html
Spatial data basics for Tasmania
http://www.utas.edu.au/spatial/locations/index.html
Biodiversity salvage blog
http://biodiversitysalvage.blogspot.com/
---