>>Look at the ds.Tables[0].Select() method. it allows you to build an sql like expression to evaluate on the dataset. The other, to return a distinct list, you use dt.Select(condition, columns[]) where condition is the expression to evaluate, and columns[] is an array of columns which you want to use to determine the distinct.
>>
>>Hope that helps, if you want more, I can give you specific code tomorrow.... Let me know..
>
>Thanks, I will take a look at that by the weekend. I will let you know on the results.

BTW, Kevin Goff's latest code magazine article deals with this particular kind of data munging. You might want to take a look at it. Its an excellent read.