confidence and checklists

[Barbara Ertter]

Another tangent of the "confidence" question concerning which I have my
two-bits to toss in, based on a project I've been doing, is that of
compiling a checklist from multiple sources.  While the argument is
sometimes made that all records should be supported by vouchers, it seems
foolish to exclude the vast amount of unvouchered distributional data that
exists (e.g., species lists in environmental impact reports) beyond what is
even feasible to voucher, especially considering the uphill battle that
natural history collections are currently fighting to justify the expense
of properly maintaining what they already have.  Of course, many such
unvouchered records are totally erroneous; then again, so are the
identifications on more vouchered specimens than we probably like to admit
(the advantage is that one can more easily double-check the identify of a
vouchered specimen, though with global-positioning capabilities this might
become a more viable option with unvouchered populations as well, at least
those that remain extant).  In any event, the "quick-and-dirty" solution
I've been using is a four-level coding:

1 = record backed by vouchered specimen
2 = record based on personal observation with a high degree of confidence
(e.g., I saw lots of poison-oak when I walked thru the area)
3 = record based on reliable secondary source
4 = record based on any other source (i.e., it might be there, but take
with as large a grain of salt as you want)

I'm also beginning to develop some rules of thumb to prioritize what needs
to be vouchered:  a) populations of taxa that are difficult enough to
identify or otherwise potentially questionable such that future
verification is likely to be desirable;  b) those that are uncommon enough
that future researchers would have a difficult time relocating material
without a precise location.  I don't deny that arguments can be made for
the vouchering of populations of common, taxonomically well-defined
species, which I continue to do occasionally; the point is determining
priorities given limited time and resources.

Barbara Ertter
University and Jepson Herbaria
University of California, Berkeley
ertter at ucjeps.herb.berkeley.edu