Hi John

That was interesting but still not there. Yes it overcomes the data type
issue but does not count the nulls and I do need the last used row which
includes nulls rather than a count.

It was also interesting because I continued testing MATCH to see how data
type affected results in a number of other wbooks. I found that using MATCH
(99^99 etc worked correctly on numeric fields and returned N/A on text and
vice versa when using MATCH("ZZZ" etc. - not unexpectedly I might add now.
In the case where a number (not N/A) was returned it proved to be the last
used row in that column which is what I'm after. If I can get MATCH to read
both types all I have to do is loop through all fields of the wbook to get
the maximun row number.

What is confusing the whole investigation is the wbook with the huge bloated
UsedRange coincidently has a text first field. I thought it was the misuse
of the data to MATCH, 99^99 or "ZZZ", that was creating the infinite loop.
However COUNTA also causes the same problem in this same wbook.

Thinking it might be the UsedRange I then tried MATCh on another misaligned
UsedRange which also had a first field as text. It worked correctly on that.
COUNTA didn't bother it either.

The puzzle therefore is why does this one wbook (up to now) have this
affect. If I do open it, it processes normally. I must resolve this.

And my original question still stands also - how can I create a MATCH
function which reads both text and numeric fields.

Geoff

john wrote:

sorry if first suggestion along wrong lines.

not tested but does doing this solve text / numeric problem?

=COUNTA('C:\Path\[File.xls]Sheet1'!A:A)
--
jb