Hi Lachlan,
The bond valence sums are also calculated automatically by the latest
version of ICSD-for-WWW ( http://www.ill.fr/dif/icsd ) - a more recent
version than you have (see http://barns.ill.fr/dif/icsd/news.html ).
ICSD uses the table from Brown & Altermatt, and calculates the valence
sum for all bonds that you ask it to draw (it is actually xtal-3d that does
the work). Since the Ro parameter depends on the particular valence state,
the lowest valence state is initially assumed. The result is compared with
this assumption, and the calculation repeated if a higher valence state is
indicated. This works well, and is quite automatic, but alternatively you
can specify your own Ro value if you don't like the one chosen for you.
Incidentally, this kind of calculation was the basis for ILL's most cited paper,
a 1990 Rietveld refinement with Bell labs on charge transfer in high Tc
superconductors (see: http://www.ill.fr/dif/citations ).
Alan H.
Alan Hewat, ILL Grenoble, FRANCE fax (33) 476.48.39.06 hewat@ill.fr
ftp://ftp.ill.fr/pub/dif http://www.ill.fr/dif/welcome.html