Wyell corrected me on this - according to him, the problem is not Jumbo datasets, but large subfiles. Sounds reasonable, and since I haven't been in the application side of the house in the 9 years I've been here, I'm sure that explanation to me was the simplistic though effective approach. * To join/leave the list, search archives, change list settings, * * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *