[Taxacom] Total number of name-bearing types

Thompson, Chris Chris.Thompson at ARS.USDA.GOV
Wed Jan 7 08:31:49 CST 2009


Yours is a quite reasonable question and one for which the answer should
be very simple GIVEN the appropriate biodiversity informatics
infrastructure. That is, a comprehensive nomenclator to scientific names
of organisms including all the information required by the appropriate
codes of nomenclature.

For question 1: "total number of name bearing types from which species
names are derived" = at least 1 type per species (taxon)

For question 2: "total number of binomial or trinomial combinations
derived from them" = in Zoological Nomenclature this is the total number
of available names (that is, valid names and available synonyms).

First before I go into more details about improving the answer to
question 1, let me editorialize first.

The problem that I see at least in Zoology is the focus has been on
SPECIES, not names. Despite the fact a Chinese philosopher declare more
two thousands years ago that all knowledge begins with names, we have
focused on SPECIES. So, for example, mammalogists do not list synonyms
as the assumption is once a synonym always a synonym. So, for example,
you find a monkey which is different (based on new character systems,
DNA) for the other currently recognized species in Brazil. So, you
declare you have a new species. Then because the populations of this
"new species" are minimal, you do not collect a specimen for a type.
That caused some arguments at the time, but a type, did exist. BUT what
was more amusing is that if mammalogists had done a proper job of it and
check the synonyms of the existing species they would have discovered
that the species had been described over 200 years ago and types existed
in a German museum.

Now we are trying to build a comprehensive Nomenclator for flies
(Diptera) which will contain all the necessary names and nomenclatural
details to answer questions like your exactly. BUT we after more than
two decades are not complete yet because no one really wants to fund
ancient history studies. The money in systematics goes to the new
glamour projects like the Barcodes of Life. Easy, simple, just take a
snip of flesh for everything and you are done. Oh, well ... some much
for the editorial.

There is at least one type for each species. So, given we know how many
species there are we can say there are minimally so many types. For
flies (Diptera), the BioSystematic Database of World Diptera (BDWD) now
documents 154,376 extant described species of flies. So, there are at
least as many types of flies.

For every known species there is at least one scientific name, but
frequently and unfortunately there are more than one. For each available
synonym (names for the same taxon = species) there is at least one type
with the exception of replacement names and emendations. Replacement
names and emendations having the same type as the name they replace. So,
for flies there are 186,653 available synonyms, minus 1,590 which are
replacement names or emendations. So, there is at least 185,063 types
for flies.

The types for available scientific names may be unique (single specimen,
such as holotype, lectotype or neotype) or a series of specimens
(syntypes). One can only determined the kind of type for each name by
critically examining the original and subsequent literataure about the
scientific name. Unfortunately, this process has not yet been completed
for flies. However, we can make an estimate based on what we know of the
name for which this process has been completed. 83% percentage of the
currently available names for flies for which there are types and their
status has been checked are unique (holotype, lectotype, neotype). Now
one needs to only estimate the average length of the syntypic series.
Unfortunately, we have no real data on which to base such an estimate as
we only record that a name is based on syntypes and assume some day
workers will select lectotypes, hence, the importance of the length of a
syntypic series is minimal. But we can say a syntypic series is at least
MORE than one, so add another estimated 31,732 "types" in sense of type

So, in summary, we can provide a fairly close estimation of the number
of types for flies. There are at least 216,795 type specimens which are
vouchers to 154,376 extant described species of Diptera.

HOWEVER, the real critical questions are:

of these types how many really exist today in Museum collection ? and
for how many to we have real data about them?

These are the important questions for Science and Society.

When types exist and their location is known, then scientist can
re-examine them and extract new knowledge based on the increased
knowledge of characters since their were originally described.

The other critical information associated with types is where they came
from. Types are individual organisms which at one point in time and at
one place were collected and/or observed. Hence, at a minimal we know
there is a point some where in the World that the species / taxon they
now belong to is known to have occurred. So for Wolfgang Lorenz GBIF
question, if we were to geo-code and database the type-localities (the
place where types occurred) then at a minimal we would have a spot on
the GBIF global view of the world for each species.

Conceptually simple and logical is the issue of geo-coding type
localities, but practically with numerous problems. However, all of
these problems = challenges can be addressed and resolved. The following
are a few example of them and their solution.

Then there is the problem of physical extant type / vouchers. Consider
the examples of:

1) Rare and obviously endangered species. Some one sees a new and
different Hawaiian honey-creeper, takes a picture, and then describes
the new species. The type is the specimen photographed and is extant at
least at the time the photograph was maded. But where is the specimen?
Today it has undoubtedly passed-on and has rotted away in the wild.
There are nomenclatural issues with this example, but the simple
challange remains, we have no voucher for which we can re-examine for
further information. Fortunately, however, we can determine the place
where the photograph was taken and geo-code that locality.

2) Old and poorly documented species. Linnaeus described Musca cellaris,
a species not recognized today, but obviously the species now called
Drosophila melanogaster or a closely related species. He gave its
distribution as only "Habitat in cellis vinariis aliisque" [=Wine
cellars]. What are we going to do about these type-localities? We can
make assumptions and agree that they are reasonable. Hence, we can
declare for common widespread species for which Linnaeus gave no
locality information or only broad locality information, but for which
the species does today occur in Sweden, then the proper geo-coding is
the location of Linnaeus' home, etc. In the case of Musca cellaris,
especially if Linnaeus had a wine cellar!

These are only two examples. The point is simply (at least for Zoology)
that there are already a number of solutions which are provided by the
International Code of Zoological Nomenclature to resolve the
type-locality of type of available Scientific names of animals. So, we
can if we want fix at least a single point on the GBIF world-view for
each and every valid species.

AND THE BOTTOMLINE is if we have good comprehensive NOMENCLATOR, then
there are many useful things we can do. So, why not support them and get
them finished properly

F. Christian Thompson
Systematic Entomology Lab., ARS, USDA
c/o Smithsonian Institution MRC-0169
PO Box 37012
Washington, D. C. 20013-7012
(202) 382-1800 voice
(202) 786-9422 fax
Chris.Thompson at ars.usda.gov 
www.diptera.org Diptera Website

-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu
[mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of David Remsen
Sent: Saturday, January 03, 2009 9:22 AM
To: taxacom at mailman.nhm.ku.edu
Subject: [Taxacom] Total number of name-bearing types

Dear Taxacomers,

Does anyone know of a source or can anyone venture a guess that might  
provide any sort of estimate as to the:

1. total number of name bearing types from which species names are  
derived, and
2. total number of binomial or trinomial combinations derived from them

Chapman and others provide an estimate of 1.8 million or so described  
species.   What material and nomenclature is circumscribed by all of  
this?  The subject came up at the ICZ meetings in Paris in August but  
the estimates were pure guesses and I wondered if there was anything  
with a rationale behind it.   Is it a crazy question to even ask?

David Remsen

David Remsen, Senior Programme Officer
Electronic Catalog of Names of Known Organisms
Global Biodiversity Information Facility Secretariat
Universitetsparken 15, DK-2100 Copenhagen, Denmark
Tel: +45-35321472   Fax: +45-35321480
Skype: dremsen


Taxacom Mailing List

Taxacom at mailman.nhm.ku.edu


The entire Taxacom Archive back to 1992 can be searched with either of
these methods:


Or use a Google search specified as:
site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here

More information about the Taxacom mailing list