What you say about river boundaries is true, or at least partly
true. Boundaries have shifted with changing river courses. But this
usually leads to a recognition that using the river as the boundary
is not a good thing and the boundary is then defined by some other,
less variable, means.
I think nearly all boundary treaties specify that the boundary is to
be marked on the ground by a set of commissioners from each party.
Their agreement and the subsequent acceptance of the marked boundary
by the interested governments locks it in "as marked".
An interesting example is the Mass.-R.I. boundary near where I live.
The description of the starting point in the original RI charter says
it is 3 miles south of Populatic Pond, the southernmost point of the
Charles River. However the actual marked boundary is over 6 miles
south of that point (I think I know what went wrong). Nevertheless,
the as-marked boundary has been upheld in the US Supreme Court twice,
on the basis that the marked boundary overrides any written
description.
No doubt some California counties have had their boundaries kinked by
earthquakes unless, of course, the boundary followed the fault line
(not unreasonable since this might create a topological feature.)
BJB
--- In BoundaryPoint@y..., Anton Sherwood <bronto@p...> wrote:
> concerning
> > > http://hotwired.lycos.com/netizen/96/53/special4a.html
>
> (My thought when I first read of the anomaly was that the US
boundary is
> "49th parallel as of 1846" and the Wa boundary is "49th parallel as
of
> 1889". Seems I was almost on the right track.)
>
>
> bjbutler@b... wrote:
> > The written description of the boundary does not matter once it is
> > marked on the ground and agreed to by both parties. This is a
basic
> > tenet of boundaryhood,
>
> ...not consistently followed where rivers shift gradually.
>
> Say, are there any prominent boundaries that have been kinked by
> earthquakes?
>
> > often overlooked by those unfamiliar with such
> > issues, as apparently the lawyers and other combatants mentioned
in
> > the article are. The Canada/Washington boundary is coincident
with
> > the Canada/US boundary and whether this falls exactly on the 49th
> > parallel is irrelevant.
>
> Surely that depends on the wording of the laws/treaties.
> If it was known early on (as the article says) that the US/Can
boundary
> is not accurately 49 degrees, then it's unreasonable to suppose that
> when Washington's constitution forty years later said "49 degrees"
it
> really meant the other line. I'd have expected it to say explicitly
> that Washington's line follows the 1846 line, but evidently it does
not.
>
>
> > Besides, 49th parallel according to what
> > datum? Surely no one would expect boundaries to change just
because
> > the world's ellipsoid is refined.
>
> Agreed, though it seems to me there might be some valid exceptions
to
> that doctrine.
>
> --
> Anton Sherwood