Yes!  That's the ticket.

Here, then, is the complete query, embellished
with an order by so that the nodes with data
on the most volumes floats to the top.  I have
split the query across multiple lines for readability.
I have tested this query on my systems and it
is getting the results I expect:

select node_name, count (distinct volume_name) as Numvols
from volumeusage where stgpool_name = 'TAPEPOOL'
group by node_name order by Numvols desc

Thanks to both Bill and Thomas!

John Sorensen
SSSI