critter names

Doug Yanega dyanega at POP.UCR.EDU
Wed Oct 10 20:27:38 CDT 2001


Once again, a can of worms I've stirred up. ;-)

The point, lost amidst the idealism, is the enormous discrepancy
between what is desirable and what is *practical*. We're dealing with
over 1 million valid taxon names, and several million synonyms,
homonyms, lapsi, nomina nuda, etc. Consider ANY of your suggested
solutions applied to THAT mountain of information. For example,
Margaret wrote:

>A relatively simple way to deal with gender-based name changes in authority
>files (as I've done in my own research files) is to have an extra field
>that contains the stem of the specific epithet (for adjectival names that
>change with the gender of the genus in which they're placed) or the entire
>epithet (for names that don't change: patronyms and nouns in
>apposition).

That may be fine on the scale of your own research files, but do you
have the time to do this for several MILLION names, one at a time,
manually entered? I already have 65,000 species and 14,000 generic
names in my database, and trying to look up each species name to see
whether it's a noun or not, and each genus to determine its gender,
and then manually enter this data in another set of fields is NOT
practical at that scale. You CANNOT automate the process, because you
need to actually go to the original publication and see what the
author designated. No database program will *ever* be able to do that.
Look at Poole's Nomina Insecta Nearctica: in the Nomina, all epithets
listed are shown in their *original* orthography, so a substantial
portion of the binomials in the Nomina are in violation of the rules.
Does anyone here think they can write a program that can download the
data from the webpages, read all the binomials, and automatically
tell you which are correct and which are not? It simply can't be
done. It requires *human labor*, and none of the available electronic
resources have already invested this labor.

Una wrote:

>  >Consider that you can't teach a database program how to conjugate latin.
>
>Why can't you?  I think I could;  there aren't too many exceptions...

Because you need to know whether a word is a noun or an adjective,
and in *many* cases you can't determine this without referring to the
original description. What is the species epithet "sonora"? A
reference to the desert, or a variant of "sonorus"? I know some
authors, too, who have taken to declaring all of their species
epithets to be "nouns in apposition" - even epithets that are
latinized - simply to keep anyone from changing them should the genus
name change. How are you going to know that without referring to the
original publication?

>A good taxonomic database should provide automatic recognition that the
>epithets trivittatus and trivittata, or igniventris and igniventre, are
>variants of the same word.  Or, even better, it would include the full
>web of published taxonomic relationships, and *never* attempt to guess
>at the current correct name by pattern-matching on epithets.

By that definition, there are no good electronic taxonomic databases
available. The "full web" you refer to includes listings of all
genera under which a species has appeared, and that is information
which is rare to find even in *print* catalogs; NONE of the
electronic catalogs or checklists I've seen contain this information.
That means going to the library to manually compile the data for well
over half of those 1 million species. I don't think any of us has
that kind of time.

Let me point out one further problem (to those who have privately
accused me of magnifying a trivial issue); "searching" in the sense I
intended is more than just lookups. Suppose I've acquired an
electronic file of an order/family with, say, 25000 species. Now, a
new version comes out, with changes. It might not be an issue if the
*only* use I had for the data was for reference; I'd simply delete
the old file and use the new one. But I don't just use it for
reference; I run a museum, in which specimens have labels attached to
them, using old names. I have a specimen-level database, too, in
which I've already entered the genus and species names of thousands
of specimens, SOME of which may need to be updated. There are
individual specimen records involved, not just the authority file
itself, and those records need a fixed link back to the taxon record
in the authority file. But there are many types of changes possible;
there are some new taxa, some old taxa have been sunk, some have been
resurrected out of previous synonymy, and still others moved to other
genera.
        Without having taxon numbers, there's no simple way to
automate the process of determining which species fall into which
category; the best I can do is find all non-matches and then
personally examine each one to determine what each represents. That's
a LOT of work. In 1983, Ron Hodges gave a unique taxon number to each
NA lepidopteran. It'd be very easy to take a database of Hodges'
original list, and automate the comparison between that and a new,
updated list that ALSO uses those same numbers. There is a precedent,
for all you naysayers - and no one has EVER complained that Hodges
numbers are a nuisance. We could easily expand that system to all
insects, and these problems with automating taxon information would
go away. I do, in fact, assign each taxon in my database a unique
number (e.g., Boisea trivittata is 62434), so I *can* update the
specimen record taxonomic information automatically - there's no
other way, since the genus name is not stable, the epithet is not
stable, and epithet-author combinations are commonly not unique.
        Efficient cataloguing requires that there be unique and
permanent identifiers, and we presently do not have them. I'm not
advocating *replacing* conventional binomials! I simply believe that
if we're serious about computerizing (certainly not a concern when
the Codes were originally Codified) we need to either fix the Codes
so names never change, or come up with some supplementary system like
Hodges did, that accomplishes the same thing. I think a supplementary
system is both easier to implement and more likely to succeed than
asking the almighty Committees to change the almighty Codes - it'd be
a small investment, with worthwhile long-term payoffs.

Peace,
--

Doug Yanega        Dept. of Entomology         Entomology Research Museum
Univ. of California - Riverside, Riverside, CA 92521
phone: (909) 787-4315 (standard disclaimer: opinions are mine, not UCR's)
            http://entmuseum9.ucr.edu/staff/yanega.html
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82




More information about the Taxacom mailing list