Two bits worth on the "single record" issue: a single good (i.e.
unequivocally accepted) record of a species far from its home
territory establishes the _possibility_ of its occurrence in the location
in question, which surely has two important implications:

- It will lend credence to subsequent one-person or less clearly
identified sightings, which might otherwise be dismissed, giving
a false picture of the vagrancy pattern of the bird in question;

- Let's not kid ourselves that every vagrant that makes it to a given
region is seen. Each record is only a statistical event, posited
the number of actual vagrants in the area and the probability of
their being observed. There are obviously quite a few Black-throated
Blue Warblers in Washington for every one that actually shows up in
the field of vision of the minute proportion of the population that
knows a warbler from a canary. This makes single records inherently
significant.

On Thu, 1 Jun 1995, Peter Rauch wrote:

> >Date: Thu, 1 Jun 1995 13:54:21 -0800
> >From: dpaulson at ups.edu (Dennis Paulson)
>
> >[Maybe what Byron]
> >meant is that a single such record is meaningless,
> > It's only when such records occur in some numbers that meaning
> >emerges.
>
> Perhaps a single record _is_ meaningless ornithologically, but on the
> other hand, I'd say "it depends." It depends on what the questions
> are. If, for example, one is ever to have a second record, then a first
> [only, meaningless?] record must be recorded first. If, for example, we
> were to ask questions about how many individual birds of a species must
> arrive at a new, barren Hawaii or Galapagos to speciate into all those
> marvelous variations, then having records, however disparate the
> species, of out-of-range individuals has some value (e.g., it might
> indicate the propensity for some species to occasionally go far out of
> range, other rarely ever, other "never", ...).
>
> One of the problems in _counting_ out-of-range incidents is that we forget
> to count the observations we've made, rather than just the positive
> incidents; if I go out to a particular area repeatedly and "observe" that
> I see no X-us y-us, but one time I see an X-us y-us, have I observed
> once or many times? Does the answer to this question make any difference
> in the value of that one positive record?
>
> But, again, like I just said earlier today, I've got a cellar full of
> stuff that might be useful someday....
> Peter
>

_________________________________________________________________________
JAMES WEST|University of Washington DP-32, Seattle, WA 98195|206-543-4892