It seems with the current discussion about what to do with the recent
odd change of volume numbering for _Popular photography_, we've forgotten
a couple of important factors:  1) what will patrons need in order to
use the bib and holdings records when they are trying to find issues of
this serial, and 2) what's the basis for the LC policy that calls for
creation of a new record for this situation.  I suspect the two factors
are inter-related.

The problem I can see occuring if we don't make some sort  of change
in control of the serial is this:  patrons will look for issues with the
new volume numbering and become confused as to which issue they need --
v. 57, no. 1 dated 1993, or the first vol. 57, no. 1 dated years ago.  Not
to mention that if the publisher keeps with the new situation, the potential
confusion increases with every issue.  The next question is, "Patrons can
just look at the dates,can't they?"  Well, yes, but most cases where patrons
go looking for an issue of a serial with a volume number in hand, they
have found that volume number from a citation, in either an index or a
footnote within an article or book.  You'd be amazed how many patrons
don't write down the complete citation.  Many times, they don't even
write enough of the title to enable them to find the call number, much less
both volume number AND date.  And you'd be surprised how many indexes
don't give both enumeration and chronology.  And you'd be surprised how
many indexes have mistakes in them (especially electronic indexes).

If the publisher does take the hint from the people at Texas A & M, and
adds the designation "new volume" or "new series" to the issues, that'd
solve everything.  All we need that would be a little out of the
ordinary in cases with "new series" would be a note in the bib record
explaining that the first issue with the new numbering constitutes the
beginning of the new series.

If the publisher doesn't take the hint, I'd suggest we at least need to
change the last digit of the Cutter number.  That way, we'll provide
a distinction between the old numbering and the new.  It might be possible,
with careful manipulations of holdings records, to keep the same bib
record, make appropriate notes explaining the change of numbering and the
accompanying change of call number. But that seems complicated.  It'd
probably be easier to follow the LCRI and create a new bib record, too,
particularly since that constitutes the closest thing to a national
standard that most of us follow.

If your library doesn't classify serials, I guess you've got a different
situation.  In that case, without a call number to manipulate, it seems
the olny way to reflect the change of numbering is with a new record.

Our staff usually try to avoid "making up" an explanation to put in
a bib or holdings record, or finding any sort of "creative responses" to
wierd cases like this numbering change.  Our public service staff find
it much easier to use records when helping patrons if the records
reflect what really happened, rather than a complicated "revision" of it.

Steve Savage
Head, Serials Dept.
University of Kentucky Libraries