Fractures and other non-welded discontinuities are important mechanical and hydraulic features of rock masses. Their effect on wave propagation can be modelled as a boundary condition by displacement discontinuity methods. For small-amplitude wave incidence, such as an ultrasonic waves, the magnitude of the stress wave is too small to mobilize non-linear fracture deformation, and so linear models are adopted in these studies to describe fracture deformational behaviour. J.G. Cai and J. Zhao (Int. J. Rock Min, Sci., 2000, 37(4), 661–682.) used the method of characteristics to examine P-wave attenuation across linear deformable fractures by considering interfracture wave reflections. In the present study, a series of laboratory tests were carried out to verify the theoretical solutions obtained by Cai and Zhao. These included ultrasonic tests across an aluminium specimen (calibration tests), intact cement mortar specimens, single-fracture specimens, and two-fracture specimens. During the tests, ultrasonic waves propagate normally across artificial fractures cast by cement mortar under different static stresses. Transmitted pulses are captured and subsequently are compared with the theoretical predictions. Generally, the experimental results agreed well with the theoretical predictions. In addition, the experimental study also provided further understanding of wave propagation across discontinuities (e.g. the applicability of displacement discontinuity methods in the dynamic problems).