What Phil put in his reply that someone said about it not being necessary to install it at each node is untrue in the Bitnet topology. If you mix RSCSV1 and RSCSV2 sites, you will lose all benefits of the list processor. The product version of the processor is not what is in use inside of IBM. What happens is you somehow send RSCS a list of people to send it to, and it converts it into NJE dataset headers. RSCSV2 and JESx support the idea of multiple dataset headers per job, and they will just peel off the copies as they arrive. HOWEVER, RSCSV1 does not support this, and as a result if he receives this file with all the dataset headers, he will spool it into n different copies, one per remaining recipients. There isn't really anything one can do, except modify the DMTNJI line driver in RSCSV1 to support this, and I'm not sure how trivial it would be. Bill