How to compare or evaluate Clusters from a same data set, not labelled ??

I have a data set of (n x m). I did clustering and get 10 clusters out of that. Then after some modification in the same data set, I get a different data set (n' x m'). Again I got same numbers of cluster i.e. 10. So , now I have two different 10-10 clusters from the data sets ( n x m) and (n' x m'). I have to evaluate these two sets of 10 and 10 clusters is there any way.??? any toolbox or something.

Can i apply Purity measures?? But I guess that can be applied only to a labelled data set..