Regarding the question about checking in misnumbered/misdated issues:

We don't have "automated" serials check-in; what we have is an online record
which has free-text fields for entering receipt data (we're using "classic"
NOTIS, without the LSER module).  When encountering erroneous data on
issues, we try to determine the correct data, and enclose them in brackets;
we also add a note regarding the mistake.  For example:

     v.10:no.1-<12>(1994:Jan.-Dec.)(12 mn 11)

(We have a strange way of twisting around the ANSI holdings standard in
order records, which is necessary because of the limited space in our
receipt fields.)

I would say that this probably isn't much of a problem most of the time.
However, some things get checked in as misnumbered when it was really
checker error:  either a mistake in checking in the issue in hand, or
logically deducing a misnumbering because of a mistake in checking in the
previous issue.

Determining the correct number/date isn't usually very difficult.  However,
there are some publications (especially in our Special Collections and
Africana departments) which are so irregularly published or so totally
messed up in numbering that we ignore the numbering altogether and just
check in by date (hoping that *that* is correct).

A problem that crops up from time to time is the checker failing to notice
that a misnumbering has gone on for a long time with little hope of the
publisher correcting it.  In those cases we would much rather consider it a
skipped number than a misnumbering--and hope that the publisher doesn't make
the problem even worse by belatedly correcting the mistake!

Kevin M. Randall
Head, Serials Cataloging Section
Northwestern University Library
Evanston, IL   60208-2300

internet: kmr@nwu.edu
   phone: (708) 491-2939
     fax: (708) 491-7637