This report is an outgrowth of the MURA Users' Conference of June 1959. At that conference the group on beam separators discussed the problem of whether particle separation could be achieved at the machine energies under consideration. A preliminary version of the scheme outlined here was given at the conference. Later, after flaws were discovered, it was modified. The attempt is not to show that this is the way it should be done, but to show with reasonable certainty that there is at least one way it can be done.