The effect of finite size transducers on elastic parameters recovered from laboratory measurements has been studied in this paper. Ultrasonic transmission experiments in the physical modeling laboratory were used to acquire the first arrival traveltime data at different ray angles, to simulate walkaway VSP surveys in the field. The elastic parameters of the materials were then recovered by our developed inversion techniques, using velocities at different directions inverted from the first arrival traveltimes. In our laboratory experiments, point transducers and large transducers were used to carry out the transmission experiments. The finite size of the transducers introduced systematic errors in the first break picks along the survey line, and hence the determined velocity field was inaccurate. These systematic errors cannot be corrected by the inversion technique. Numerical simulations were also conducted to study the differences between experiments using large and point transducers. From our experimental measurements, accurate elastic parameter values are obtained when the ratio of the sample-thickness to transducer-diameter is greater than 36. Decreasing the size of the transducers is highly recommended when conducting laboratory experiments to recover elastic parameters. A method of offset correction is also implemented in an attempt to decrease the measurement errors. ¿ 2000 American Geophysical Union