In this paper a systematic method for the design of efficient parallel
algorithms for the dynamic evaluation of computation trees and/or exp
ressions is presented. This method involves the use of uniform closure
properties of certain certain classes of unary functions. Using this
method, optimal parallel algorithms are given for many computation tre
e problems which are important in parallel algebraic and numerical com
putation, and parallel code generation on exclusive read and exclusive
write parallel random access machines. Our algorithmic result is comp
lemented by a P-complete tree problem.