[Taxacom] Taxon concepts and published names

Ginzbarg, Steve sginzbar at biology.as.ua.edu
Mon Jun 26 15:09:31 CDT 2006


I have recently added a field 'AccordingTo' (sec) to our database where
I can record (currently in free text, a drop-down list would be better)
the reference used to make a determination. I have used it a couple
times when I wanted to show that I had used a revision, not just a
flora. It would be a good idea to start using it consistently.

I looked at the Taxonomic Concept Transfer Schema,
http://tdwg.napier.ac.uk, voted a TDWG standard in 2005. I agree that we
should try to follow it. However I think we need short range goals as
well as long range goals. As Theodore Roosevelt said, "Keep your eyes on
the stars but keep your feet on the ground." Before we can use a
synonymy based on concepts we will need to agree on how to record
concepts consistently when we database specimen determinations. Mapping
the relationships of concepts to one another (includes, overlaps, etc.)
is no small task. Major changes in concepts usually happen in taxonomic
revisions. Floras follow these treatments. However, specimens may not
key out in the same way when using a flora following a revision as when
using the revision itself. Should the concepts be considered equivalent
or not? 

This brings me to the Ginzbarg Uncertainty Principal of taxonomy
analagous to the Heisenberg Uncertainty Principal of quantum mechanics.
Just as it is impossible to know both the exact position of a particle
and the speed that it is moving, so it is impossible to know both the
current distribution of a taxon and the taxon concept represented by the
distribution. If we want to see the distribution of a particular taxon
concept we can look at a published monograph. The monograph has the
disadvantage, though, of being a snapshot in time while our databases
can be continuously updated with new records. I think that GBIF should
include a disclaimer to this effect.

Very few specimen determinations made to date tell what reference was
used. Our immediate need is a synonymy that can synthesize the
information that is available now, i.e. data from specimens identified
with published names only, the taxon concept not specified. The synonymy
should be able to answer "If the determination is published name A, can
we say that the accepted published name is B or is it necessary to
re-examine the specimen to determine what the accepted published name
is?". Do we have such synonymies in botany? We have human-readable
synonymies but not synonymies that would allow a computer to answer this

The following database design would allow a computer to answer the

Determination 00 -- 1 Published Names 1 -- 00 Synonym (Published Name [+
"some information"]) 00 -- 1 Accepted Name

(If a published name is not in the list of synonyms it is accepted.)

Suppose a specimen has been determined as Species A. There are two
concepts of species A listed in the synonymy, each having a different
accepted name. Since the determinator only provided the published name
and not a taxon concept, I don't know which concept was meant and have
no basis for choosing one or the other. An accepted name cannot be
determined. The specimen must be re-examined. On the other had there may
be other synonymies which are unambiguous and an accepted name can be

Here is a design using the current BONAP database.

Determination 00 -- 1 Synonyms (Published Name [including "some
information"]) & Accepted Names 00 -- 1 Accepted Name 

With two important modifications the BONAP database design could meet
this need. 

1.) A list of all published names treated is needed to which published
names of determinations could be matched and from which published names
could be selected when recording a determination in a specimen database.
Selecting from a list containing a mixture of published names and taxon
concepts causes inconsistency in the recording of the names. A published
name should be listed if a taxon concept of that name is treated as a
synonym. It should also be listed if the name alone is listed, either as
a synonym or an accepted name. The list needs to be separate from the
list of synonyms and their accepted names.

2.) As recommended in the Taxonomic Concept Transfer Schema the
'AccordingTo' or more generally 'some kind of definition', e.g. "p.p."
should be recorded separately from the published name. Currently BONAP
records the 'some kind of definition' in the same field as the authors
of the name. At other times the "some kind of definition" replaces the
authors, e.g. Lespedeza intermedia sensu Clewell, 1966 rather than
Lespedeza intermedia (S. Wats.) Britt. sensu Clewell, 1966. Separating
them would allow published names to be linked to their taxonomic

A design using the Taxonomic Concept Transfer Schema is a refinement of
the design proposed above.

Determination concept 00 -- 1 Taxon Concepts 1 -- 00 Synonym concept 00
-- 1 Accepted concept
(If a taxon concept is not in the list of synonym concepts it is

-Steve Ginzbarg
Steve Ginzbarg, Collections Manager
Herbarium (UNA)
Department of Biological Sciences
Box 870345
The University of Alabama
Tuscaloosa, AL 35487-0345
(205) 348-1829, FAX: (205) 348-6460
sginzbar at biology.as.ua.edu

> -----Original Message-----
> From: Robert K. Peet [mailto:peet at unc.edu] 
> Sent: Sunday, June 25, 2006 9:42 AM
> To: Ginzbarg, Steve
> Cc: taxacom at mailman.nhm.ku.edu
> Subject: Re: [Taxacom] a Standard for Taxonomic Verification 
> Qualifiers
> Steve,
> If you are going to go through all the work of checking the 
> determinations of the specimens in your collection, you 
> should be sure to record in your database not just the names, 
> but the taxonomic concepts (name as used by some authority).  
> For example, record not just Andropogon virginicus L., but 
> Andropogon virginicus L. sec Radford et al 1968, or sec 
> Godfrey & Wooten 1979, or sec FNA 2003, or sec Hitchcock & 
> Chase 1950, or sec Bloomquest 1948, or sec Small 1933 --- 
> they are all different concepts of A. virginicus L. with 
> different sets of specimens circumscribed.
> TDWG has adopted standards for concepts, and they are in the 
> design plans for future releases of database systems 
> associated with PLANTS, ITIS, GBIF, and SPECIFY.  They will 
> soon impact you directly. If you don't do the determinations 
> by concept now, you are likely to deeply regret it in a few 
> years when you have to do the work over again.  Take a look 
> at the county range maps on PLANTS for Andropogon virginicus 
> kin in NC-SC and you will see the problem.  They have lots of 
> dots for Andropogon virginicus, but these are not based on 
> the concepts used by PLANTS, as evidence by such taxa as A. 
> glaucopsis, A capillipes, A. glomeratus and A. gyrans being 
> listed for NC-SC with no county records.  In short, we cannot 
> trust any of those A. virginicus records in the maps because 
> all these other taxa were lumped in by Radford et al, which 
> is where the dots came from. 
> If future determinations are made to concept, we will know 
> where there are uncertainties, and where datasets (such as 
> for maps) can be merged and where not.
> Bob Peet
> ======================================================================
>       Robert K. Peet, Professor & Chair         Phone:  919-962-6942
>       Curriculum in Ecology, CB#3275            Fax:    919-962-6930
>       University of North Carolina              Cell:   919-368-4971
>       Chapel Hill, NC  27599-3275  USA          Email:  peet at unc.edu
>                     http://www.unc.edu/depts/ecology/
>                   http://www.bio.unc.edu/faculty/peet/
> ======================================================================
> On Fri, 16 Jun 2006, Ginzbarg, Steve wrote:
> > At UNA we're are examining our AL specimens for the AL 
> vascular plant 
> > checklist and atlas. When the current determination has 
> been confirmed 
> > by herbarium staff, a pencil check mark is written on the 
> collection 
> > or determination label. A check box in the record for the 
> > determination in our database is also checked. (We don't 
> have the time 
> > to annotate all the specimens where we concur with the 
> current det.) 
> > Material deemed inadequate to make a positive determination, e.g. a 
> > sterile specimen, does not receive a check and is not considered a 
> > voucher for the checklist and atlas. We have a program that 
> displays 
> > the AL county records. If the determination of a specimen from a 
> > county has been confirmed by the staff of the institution 
> where it is 
> > housed a black dot is shown in that county. I there is a 
> specimen with 
> > a current det but the det has not been confirmed by staff 
> it shows as 
> > a grey dot. If there were an element VerifiedByInstitution in 
> > DarwinCore ("Y" if the determination has been confirmed by 
> the staff 
> > of the institution in which it is housed) we could supply 
> GBIF with this information.
> >
> > Some institutions do not provide the most recent 
> determination to GBIF 
> > but rather the most recent determination that the herbarium staff 
> > accepts (determined by an expert). While I personally would 
> want to be 
> > very confident in my own determination before disagreeing with an 
> > expert, I would not want to judge which determinations UNA should 
> > provide to GBIF on the basis of name recognition of the determiner.
> > Experts can make mistakes.
> >
> >> -----Original Message-----
> >> From: taxacom-bounces at mailman.nhm.ku.edu
> >> [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of 
> >> taxacom2 at achapman.org
> >> Sent: Thursday, June 15, 2006 5:26 PM
> >> To: taxacom at mailman.nhm.ku.edu
> >> Subject: [Taxacom] a Standard for Taxonomic Verification Qualifiers
> >>
> >> I am currently examing the possibility of establishing a TDWG 
> >> Standard for Taxonomic Verification Qualifiers and am seeking 
> >> information from anyone currently using such qualifiers.  
> If you are 
> >> using one (or have suggestions for one), I would 
> appreciate a copy, 
> >> which can be sent to me off-line.
> >>
> >> Later, I hope that we can start a WIKI discussion on the tdwg Web 
> >> site (http://www.tdwg.org), and a presentation is proposed for the 
> >> TDWG meeting in St Louis in October.
> >>
> >> I am currently aware of three standards that are in use 
> (see attached 
> >> file):
> >>
> >> 1. from Herbarium Information Standards and Protocols for the 
> >> Interchange of Data (HISPID) Vers. 3
> >>
> >> 2. A similar one based on this and used by the Botanic Gardens 
> >> community International Transfer Format for Botanic Gardens Plant 
> >> Records (ITF) Vers. 2.0
> >>
> >> 3. One used by the Australian National Fish Collection
> >>
> >> In the Data Quality Document I prepared for GBIF lat year, I 
> >> suggested a two level standard that I would like to see some 
> >> discussion on. I believe that none are entirely suitable, and 
> >> possibly an incorporation of all four would be the best.
> >>
> >> Chapman A.D. (2005). Principles of Data Quality. Report for the 
> >> Global Biodiversity Information Facility 2005. 61pp.
> >> Copenhagen: GBIF.
> >>
> >> The reasons for such a standard I see as
> >>
> >> 1. The need for improved documentation of quality with the 
> increasing 
> >> distribution of primary species data 2. The introduction 
> of privacy 
> >> legislation in many countries that is beginning to restrict the 
> >> distribution of people's names, including the names of 
> determiners of 
> >> specimens.  If we cannot exchange the name of the 
> determiner, we need 
> >> some other method to reliably document the confidence we 
> have in the 
> >> identification.
> >>
> >> I look forward to your responses.
> >>
> >> regards
> >>
> >> Arthur D. Chapman
> >> Australian Biodiversity Information Services Toowoomba, Australia
> >>
> >
> > _______________________________________________
> > Taxacom mailing list
> > Taxacom at mailman.nhm.ku.edu
> > http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
> >
>   ====================================================================
>   Robert K. Peet, Professor              Phone:  919-962-6942
>   Department of Biology, CB#3280         Fax:    919-962-6930
>   University of North Carolina           Cell:   919-368-4971
>   Chapel Hill, NC  27599-3280  USA       Email:  peet at unc.edu
>               http://www.bio.unc.edu/faculty/peet/
>   ====================================================================

More information about the Taxacom mailing list