T violation should be tested independently of CP violation. Besides K system, B meson decays provide another good place to study T violation. In the Standard Model, T violation in $B^0 \rightleftharpoons \bar{B^0}$ oscillation is expected to be small. The angular distribution of $B\to VV$ decay permits one to extract the T-odd correlation. In the absence of final state interaction, T violation in $B\to J/��(l^+ l^-) K^*(K_S��^0)$ decay can reach $4-7%$ via $B^0-\bar{B^0}$ mixing.

Latex, 11 pages, revised version will appear in PLB