[NHCOLL-L:345] RE: Further on the accession by committeequestion

[Arthur H. Harris]

Dear Beth--

Some good points--responding to the original correspondent, I was
set off (sensitive spot) by the "one example of blobbus maximus"
being as good as any other.  In general, I agree that what one looks
for is a sample size (and spacing of samples in time and geographic
space) sufficient for reasonable statistical analysis, and where
space and other resources are limiting factors, lines necessarily
must be drawn.  My concern tends to be over whether lines are being
drawn logically or arbitrarily.  The field as a whole hasn't done a
particularly good job of communicating what is necessary, what is
desirable, and what is luxurious.

Although not sanguine enough to believe that natural history
curators are necessarily going to be logical, one possible approach
(from a theoretical point of view) would be to go to a negotiated
piece of the resource pie but allow the curators rein within that
piece (a sort of "we can allocate resources for up to 200 bird
specimens over the next year--which 200 is up to you").

You made an important point in "Yet none of the curators could ever
articulate guidelines for deciding when there was "enough" of a
given category of thing." Where resources are being stressed, it
seems a reasonable enough demand from the administration that
curators set up guidelines for at least routine acquisitions.

"Elizabeth E. Merritt" wrote:
> 
> Dear Art -
> 
> To your "differences in culture, differences in perception" I would add,
> "failure of communication."  I come from a science background, with
> training in both organismal and cellular biology.  I found myself in the
> position of administering a large museum collections department, and
> reporting to a CEO from a non-science background.  When he asked "when do
> you have enough biological specimens" even I had to admit the explanations
> you repeat ("sufficiently large enough samples to be able to characterize
> populations and that we need population data from throughout the geographic
> and chronologic ranges of a taxon") don't justify making a study skin of
> every dead bird that comes through the door.  Yet none of the curators
> could ever articulate guidelines for deciding when there was "enough" of a
> given category of thing. So it would lead the the inevitable stand-off: the
> administration fuming that there was limited space, time and money, and
> that the collections policy had to help limit collecting to fit those
> resources, and the curators claiming that since each specimen is unique,
> each added precious information to the collection, and there was no basis
> for saying "400 cardinals from Cincinnati is enough."
> 
> I, rationalist that I am, always felt that the key lies in your wording -
> "sufficiently large samples", with the key being a statistical analysis of
> how large those samples would have to be.  In fact, the uniqueness of the
> specimens is a red herring. The question is when does an additional
> specimen stop adding a statistically significant amount of information to
> the pool.
> 
> I would love to hear more discussion of this, if people want to weigh in.
> 
> Beth Merritt
> American Association of Museums


-- 
Laboratory for Environmental Biology
Centennial Museum (Natural and Cultural History)
University of Texas at El Paso, El Paso, TX  79968-0915
phone (915)747-6985; fax (915)747-5808; aharris at utep.edu
http://www.utep.edu/leb     http://www.utep.edu/museum