The authors describe an experiment involving the solution of the initial value problem for systems of ordinary differential equations using extrapolation in a parallel processor implementation. Each of $p$ processors is used to solve the system numerically using a different (fixed) mesh size: $h\sb 1>h\sb 2>\dots>h\sb p$. As a processor completes its assigned integration at the required mesh points (note that Processor 1 is done before Processor $p$) it can be used in the calculation of the extrapolation table. Examples are given along with values observed for the speed-up.
J.A.Crow (Corvallis)