Let {B_t} be a standard one-dimensional Brownian motion and cp a twice continuously differentiable strictly increasing function of (inf, inf) onto (a, b). Then {Xj :_ {phi(B_t)} is a time-homogeneous Markov process with continuous sample paths. Take phi(x) = e^(x^3). Given phi is a continuous strictly monotone function on (a, b) onto (c, d), can we compute the transition probability density for process {X_t = e^(B_t^3)}