If the # was first, you could use the techniques in Natural Sorting.

However, since that is not the case, I would probably create a calculated field in a query and sort on that:

SortField:  Right(SortString, Len(SortString) - InStr(SortString,"-"))

As Long as your data is always formatted with some character(s) before a single dash, then the # this would work. If you wanted sorted by the letter then the #:

SortField1:  Left(SortString, InStr(SortString,"-")-1)
SortField2:  Right(SortString, Len(SortString) - InStr(SortString,"-"))

Hi Gregory, Have you thought about  creating 2 fields for the data. One for the character string and the second for the numeric field. I don't know if you have this possibility, for you may be working with an established system that you cannot change.   John

With a similar situation I give everything a five character identifier. Rather than C-01 use C-001. If you think the need may arise for another character make it C-0001 then when you get up to the higher numbers, C-0101 or C-1200, they all just fall in order. If you think there may come a need for two alpha characters in front start with all getting two characters then go AA-, AB- then eventually CA-, CB-. The end result AB-1205, BA-0001, etc. all identifiers get the same number of characters.

Thanks for your suggestions folks. Using Donald Blackwells' suggest produced the same results, C-100 showed up right after C-10 and I'm afraid John Davy's suggestion would probably work but not feasible in this case. So the only solution that worked was to pad the numbers with zeros which I wanted to stay away from. Even tried to get some help from AI but so far that suggestion only produced 0's. Thanks again for your suggestions - much appreciated.    GD

Gregory G
In my suggestion for field 2, you could add clng( before the right function and then add an extra ) at the end.