[Taxacom] formation of zoological names with Mc, Mac, etc.

[Francisco Welter-Schultes]

Paul is absolutely right that taxon name author strings are not 
truely unique. In practical life they are used as if they were, but 
in theory they are not. 
But of course it is not necessary to have 10 fields.

In AnimalBase we needed to add a 5th field "original spelling" 
(includes original combination, so I confirm Doug's assumption, 
usually the same subspecific names were established in different 
species) to create a unique entry (the program does not allow two 
identical entries). This was needed in estimated 200 out of 20,000 
cases, around 1 %, maybe a little more. In some 20 out of 20,000 
names (0.1 %) we in addition needed to add "page" (we simply add the 
page number in brackets to the "original spelling" field - but this 
is just a trick to avoid a 6th field) to get a unique name.

As so often, Linnaeus himself was the first to create the problems.
Papilio aglaja  Linnæus, 1758  p. 465  
Papilio aglaja  Linnæus, 1758  p. 481  

(there is a second example, Cardium muricatum, but Linnaeus 
saw and corrected that on p. 824. All other names were unique.)

Page alone is not sufficient as a 5th field, we have examples where 
identical subspecific names for different species were established 
on the same page (and here, Linneaus was not the first to create the 
problems...).

I don't know why the problem has never arised as such in connecting 
biodiversity related information, but it is possibly because these 
cases concern original combinations, the biodiversity databases work 
with currently used genus-species combinations.

I do not remember a case where a genus-species-author-year 
combination for a currently used name does not provide a unique 
identifier. Theoretically this is possible, the Commission can rule 
such things.

Author and year are important although they are rarely needed 
for providing a unique identifier in a currently used name (only when 
homonyms match weirdly, or the Commission allowed it), the most 
important thing is their function for the error control.

Different generic placements do indeed create problems, they must be 
treated as two different identifiers, two different species, they 
cannot be lumped automatically, this is one of the general problems 
of Linnean names. It can only be recommended not to change generic 
placements if not absolutely necessary.

Francisco


University of Goettingen, Germany
www.animalbase.org