Axial-flow, impellers, like pitched-blade impellers, are being increasingly used for gas-liquid systems in stirred vessels. In this work we have used particle image velocimetry (PIV) and computational fluid dynamics (CFD) models to investigate gas-liquid flow generated by a downflow pitched-blade turbine. PIV measurements were carried out in a fully baffled stirred vessel (of 0.19 m diameter) with a dished bottom. Angle-resolved measurements of the flow field with and without gas dispersion were carried out. An attempt was made to capture key details of the trailing vortex, the accumulation of gas, and the flow around the impeller blades. A two-fluid model along with the standard k-epsilon turbulence model was used to simulate dispersed gas liquid flow in a stirred vessel. The computational snapshot approach was used to simulate impeller rotation and was implemented in the commercial CFD code, FLUENT 4.5 (of Fluent Inc., USA). The model predictions were verified by comparison with the PIV measurements and other available experimental data. The computational model and results discussed in this work are useful for better understanding and simulation of gas-liquid flow generated by axial impellers in stirred vessels.