This paper investigates the robust stabilization problem of semilinear Markovian jump distributed parameter systems (which are modeled by parabolic partial differential equations) with time-varying delay and incomplete transition probabilities. Based on Takagi-Sugeno (T-S) fuzzy theory, a T-S fuzzy model is obtained to describe the nonlinear systems. Furthermore, by constructing a novel Lyapunov functional candidate, several sufficient delay-dependent conditions, which ensure the considered systems stochastically stable and strictly dissipative, are established in terms of linear matrix inequalities. Finally, a fuzzy controller design approach is proposed, and two examples are presented to demonstrate the effectiveness of the designed controller.