[Taxacom] Total number of name-bearing types

David Remsen (GBIF) dremsen at gbif.org
Fri Jan 16 08:47:33 CST 2009


Dear Tony

I was motivated to ask the original question for a couple of reasons.   
It was really intended to understand the scope of Karl Magnacca's  
"theoretical holotypes."   Of course, I think the adjacent questions  
of how many physical types exist and their distribution into the  
various type categories is of value for a range of additional reasons  
outside the intent of my question.

My motiviation, hopefully not too off-base was, to systematically  
reconcile the different degrees of cardinality between taxon names and  
taxa to understand the basis by which they might be rationalized.

So for example, within the GBIF index of species names, collectively  
representing some 160 million plus individual specimen and  
observational records, there are approximately 5 million distinct  
strings purported to refer to species that fit the general pattern of  
a binomial or trinomial.  I want to understand what this really  
means.  Is it a single species name spelled incorrectly 5 million  
different ways?  Hope not.  Is it 5 misspellings for 1 million  
correctly spelled names?   What forms the basis for a name anyway?   
What counts as one?

So starting with taxa we know that there is at least 1 type per  
species as Chris Thompson asserts.

And we know there is at least 1 name per type.

We also know that additional names can subsequently be created that  
also refer to the type.

This includes replaced or emended names which I believe are governed  
by all the major codes.  It also includes re-combinations that result  
from assigning the type to a new genus.  In botany I understand this  
is also a code-governed act, but in zoology it does not.   In either  
case. all available (zoological code) or valid (botanical) names are  
linked to a type.  The only exceptions are names that either once had  
nomenclatural standing and lost it such as many pre-1980 prokaryote  
names, or they never had it in the first place.

All subsequent orthographies can be resolved to one of these binomial  
combinations which are tied to an original nomenclatural act and this  
act is tied to a type.   But I'm not really sure what constitutes one  
"name" in this sense.  If zoology only counts the first usage as "the"  
name then a subsequent combination is just another orthography of the  
same name.

So I guess what really matters to me is the number of distinct  
orthographies that can be tied to a single type.   And so the total  
number of types would be a useful figure to know.  That's the number  
that we have to resolve all orthographies to.

Then we need to tie all those types to taxa I suppose.  In this way we  
would really have a complete catalogue of life.

I know it's a lot of work but I do agree that if we have a good  
comprehensive nomenclator there would be many useful things we could do.

Cheers,

David


----------------------------------------------------------------------------
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
----------------------------------------------------------------------------


On Jan 15, 2009, at 4:55 AM, <Tony.Rees at csiro.au> <Tony.Rees at csiro.au>  
wrote:

> Dear all,
>
> Actually I would appreciate some more background from David Remsen  
> regarding the motivation for his original question - my feeling  
> (which may be quite unwarranted of course) was that he was seeking  
> an independent way to calculate known biodiversity to see if the  
> often-quoted figure of 1.7-1.8 million described species stacks up -  
> i.e., if the numbers of type specimens could be established from  
> museum records or literature, then some reduction factor employed to  
> account for multiple types and synonymized taxa, then you could see  
> how close to the 1.7-1.8 million figure was the answer. What appears  
> to be happening may not be contributing to this, as it is  
> extrapolating the number of types from the estimated number of  
> described taxa, therefore circular reasoning if my original  
> assumption is correct - David, would you care to comment here?
>
> Regards - Tony
>
> ________________________________________
> From: taxacom-bounces at mailman.nhm.ku.edu [taxacom-bounces at mailman.nhm.ku.edu 
> ] On Behalf Of Kenneth Kinman [kennethkinman at webtv.net]
> Sent: Thursday, 15 January 2009 1:40 PM
> To: taxacom at mailman.nhm.ku.edu
> Subject: [Taxacom] Total number of name-bearing types
>
> Dear All,
>       I just realized I made a mathematical error in my last post.  At
> 1,700,00 accepted species times an estimated overall minimum ratio of
> 1.5, the minimum number of type specimens would actually be 2,550,000.
>       In any case, I still suspect the number actually exceeds three
> million.  But as I said, the most critical information is probably an
> estimated ratio for angiosperms, because of the large number of  
> species
> (although an estimate for molluscs would also add to the precision any
> estimate).
>          -------------Ken Kinman
>
>
> _______________________________________________
>
> Taxacom Mailing List
>
> Taxacom at mailman.nhm.ku.edu
>
> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
>
> The entire Taxacom Archive back to 1992 can be searched with either  
> of these methods:
>
> http://taxacom.markmail.org
>
> Or use a Google search specified as:  site:mailman.nhm.ku.edu/ 
> pipermail/taxacom  your search terms here
> _______________________________________________
>
> Taxacom Mailing List
>
> Taxacom at mailman.nhm.ku.edu
>
> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
>
> The entire Taxacom Archive back to 1992 can be searched with either  
> of these methods:
>
> http://taxacom.markmail.org
>
> 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