Hi, the built-in aggregate() function is --I'm sure-- a very powerful tool to summarize and analyze data efficiently. However, I think it's also quite difficult to grasp. I'm reading its function reference right now. To be concrete, from what I understand the column with the counts will be somewhere in the middle of the returned matrix. But the position of that column is not so easy to derive. It somehow must have to do with the combinatorial possibilities of the list members passed in as the second argument. (Of course this applies only to the case of passing lists rather than individual series.) Perhaps the documentation could be more explicit about the index numbers there, providing a concrete formula? Or I wouldn't mind an alternative interface, either. For example, perhaps accept an optional fourth argument as a pointer to a matrix, and this gets filled with the counts. That way it wouldn't be necessary to juggle position indices around. just an idea, thanks sven