[NHCOLL-L:3401] Re: Labels for wet collection

[Simmons, John E]

Dear Mr. Pentcheff,

I must say I was somewhat amazed to read your call for "evidence based
collections management" presented as if it was a new idea, considering
that you posted it on NHCOLL-L, the listserv run by SPNHC, the
professional society that has done the most to promote rigor in
collections management and rid the field of folklore and hearsay.  SPNHC
has been advocating the testing of preservation technologies since it
was founded, and the organization has been a leader in publishing this
sort of research.  Numerous SPNHC members, myself included, have been
advocating in print for more of these studies and have been engaged in
analysis of preservation technology for more than 20 years.

 

You have set the bar very high by demanding that studies be published in
peer-reviewed journals before procedures and techniques are used (by the
way, I presume that you actually meant to say that techniques should be
tested by controlled experiments, not just endorsed by peer review,
given that the peer reviewers can be just as ignorant as the authors of
studies?).  By following your standard for peer-review, most
collection-based research should not be published either because most of
it is based on specimens that were prepared by untested and unproven
folkoric techniques that can (and often do) have profound effects upon
the specimens that are not at all well understood.  Consider the use of
alcohol and other fluids, a preservation technology that has been
reported on anecdotally since 1662-I have reviewed more than 730
references on fluid preservation and found that although preservation in
fluids is a common practice, there is precious little proper scientific
basis for using one technique instead of another.  The literature is
full of anecdotal accounts of various researchers' favorite fixation or
preservation recommendations, very few of which have been subjected to
controlled experimental testing.  Are you seriously suggesting that we
cease fixing and preserving specimens until we rid the field of such
folklore?  Or do you advocate that we continue to produce science based
on spurious preservation techniques that are supported by mere anecdotal
evidence?

 

Demanding the "definitive study" is a convenient way to ignore
inconvenient facts (for example, the lack of "definitive studies" is the
justification the Bush administration is using to avoid dealing with
climate change).  I know a former NSF program officer who didn't want to
allocate grant funds to replace wooden drawers for study skins with
metal drawers because there is no definitive study demonstrating that
the off-gassing of acids from wood harms the specimens.  Does it really
require a definitive study to show that known acids affect known organic
materials that are in contact with them?  Or do you advocate that we
continue to place bird and mammal skins on raw plywood until such time
as definitive study says that we shouldn't do it?  It seems to me that
your standard might be a bit of a problem if we care at all about our
collections.

 

I agree with you that "the plural of 'anecdote' is not 'data,'" but on
the other hand, just because evidence is anecdotal does not mean it is
wrong.  When dozens of my colleagues try using laser printer generated
labels and all of them experience failures of the ink, I find this to be
sufficient evidence that there is a problem with laser printed labels in
fluid.  On the other hand, when I read your paper (Zala et al. 2005), I
find that you failed to test the variables suspected to cause ink
failure (abrasion by labels rubbing against the containers and against
specimens in the containers; the presence of specimens in the alcohol;
etc.).  You should know this because I presume you know that I read a
draft of the paper and I pointed out numerous shortcomings in the study
(which were not corrected) before it was published.  What Zala et al.
did was to demonstrate was that laser printer generated labels in very
small containers of clean alcohol held up for 14 years.  Personally, I
don't know of anyone collecting clean alcohol samples in small jars, so
I repeat my warning about extrapolating from these results for any
application in a collection of fluid preserved specimens.

 

I agree with you that anecdotal does not mean data, but then,
peer-reviewed does not mean correct, either.  Nor is peer review (or
even controlled experiments) the only way to acquire useful knowledge.
You are right that much more rigorous testing of preservation
technologies needs to be done, but your three assumptions as to why
anecdotal evidence is used to evaluate label printing are all incorrect.
What current practices reflect is that competent collections managers
recognize that research funds are limited, labs are few, most
collections care personnel don't have research time, and most scientists
who do collection based research are far happier with the Oral Tradition
they inherited from their beloved major professor than they are
interested in testing preservation technologies, so we must make
reasonable decisions based on the information we can evaluate, pending
the time when good controlled studies are done.

 

And by the way, you might want to be careful about how you interpret the
"evidence" you are anticipating from the manufacturer of the thermal
technology-I strongly suspect it does not meet your test of being
peer-reviewed.

 

--John

 

John E. Simmons

Collections Manager, Natural History Museum & Biodiversity Research
Center

and

Director, Museum Studies Program

University of Kansas

1345 Jayhawk Boulevard

Lawrence, Kansas 66045-7561

Telephone 785-864-4508

FAX 785-864-5335

jsimmons at ku.edu

www.nhm.ku.edu/herpetology

www.ku.edu/~museumst

 

 

 

-----Original Message-----
From: owner-nhcoll-l at lists.yale.edu
[mailto:owner-nhcoll-l at lists.yale.edu] On Behalf Of Dean Pentcheff
Sent: Thursday, April 19, 2007 5:29 PM
To: nhcoll-l at lists.yale.edu
Subject: [NHCOLL-L:3400] Re: Labels for wet collection

 

There have been excellent suggestions posted here to this list in the

last few days regarding labelling systems that have potential value

for wet collections.

 

What we haven't seen, though, is evidence.  As the trope has it: the

plural of "anecdote" is not "data".

 

It's clear that many collections workers have labelling approaches

that appear to work well for them.  But collections management is

supposed to be (or at least is promoting itself) a science in support

of science.

 

If the topic of durability of labels in wet collections is not

receiving published analysis in a reproducible, well-designed manner,

then I can think of the following possibilities:

 

1) Collections workers just don't think it's an important topic.

 

2) Good studies have been done but the workers have not chosen to

publish them in peer-reveiwed scientific publications.

 

3) Workers have done various quick-and-dirty tests that wouldn't stand

up to peer review, and base their opinions on those.

 

I dismiss (1).  In my experience collections professionals work hard

to permanently associate data with specimens, and think this topic is

important enough to have extremely strong opinions on it.

 

I'm not sure about number (2).  Is there any disincentive to

publishing well-done work?  The writeup is usually the least

time-consuming part of a good study.

 

I fear that (3) is the most likely cause.  It disturbs me.  It implies

that collection workers are willing to base important long-term

decisions on inadequate evidence and are uninterested in advancing

their colleagues' work.

 

We don't expect a drug to be prescribed just because our doctor has

tried that chemical out on a couple of her patients and it seemed to

work.  We expect proper double-blind placebo-controlled trials for

both safety and efficacy, with adequate sample size, published in the

peer-reviewed literature.

 

We've come to realize that that is the best way of gaining and

establishing knowledge.  Anything less is inadequate.  Then why do we

accept anecdote when making decisions about long-term labelling of

irreplaceable specimens?

 

Basing recommendations on personal experience puts collections

management in a pre-scientific guild-style mode of operation.  Lore is

acquired through experience and passed down in oral tradition.  I

don't think that's good enough to ensure the survival of labels, and I

don't think it's good for the future of collections management.

 

Obviously (or perhaps not, since I feel the need to state it) I have

strong respect for experience in this and other fields.  But I think

we should be striving to move further and turn that experience into

rigorous, independently-confirmable results.  If I'm in a new museum

in a remote location, I should be able to find literature that

explores and tests the major issues I will face in conservation

(including labels in wet collections).  I should be able to find tests

of things that worked, tests of things that failed, analysis of those

results, with predictions of what will work and directions for future

research.  That's how we'll all be able to profit from the solid work

of our colleagues.

 

That's what I'm arguing for: evidence-based collections management.

 

-Dean

P.S. In the way of evidence, Howard Newman of Alpha Systems has

promised to send me literature regarding the thermal transfer

technology described by Andy Bentley.  I'm looking forward to seeing

it.

-- 

Dean Pentcheff

dean at crustacea.nhm.org

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.yale.edu/mailman/private/nhcoll-l/attachments/20070420/4c67f45c/attachment.html