This paper presents the results of an experimental investigation on the rocking behavior of rigid blocks. Two types of test specimens have been tested, namely M and C types. Nine blocks of the M type and two blocks of the C type with different aspect ratios were tested with varying initial rotational amplitudes and with different materials at the contact interface, namely concrete, timber, steel, and rubber. The results showed that the interface material has significant influence on the free rocking performance of the blocks. Blocks tested on rubber had the fastest energy dissipation followed by concrete and timber bases, respectively. Analysis of the test results has shown that the energy dissipation in the case of tests on a rubber base is a continuous mechanism whereas in the case of tests on rigid bases, i.e. timber and concrete, energy dissipation is a discrete function. Finally, the rocking characteristics of the blocks were calculated using piecewise equations of motion and numerical analysis. It was possible to predict the correct free rocking amplitude response when a reliable value for the coefficient of restitution was used.the piecewise equations of motions [3]. Pena et al. [6] used complex coupled rocking rotations and discrete element methods to predict the rocking response of four specimens. Both methods are extremely sensitive to the rocking parameters. Finally, it was found that repeatability of the rocking tests under random vibration does not exist.Shenton and Jones [7] showed that under horizontal ground excitation a rigid block has five modes of response namely rest, rocking, sliding, rocking-sliding, and free-flight. Shenton [8] analytically derived the criteria that govern the initiation of these modes in the static frictionpeak ground acceleration parameter space. Oliveto et al. [9] derived analytical expressions for the minimum acceleration impulses for uplift and overturning. They showed that the minimum overturning impulse of a flexible system is always smaller than the corresponding impulse for the rigid system. Taniguchi [10] showed that the effect of the vertical ground acceleration component on the response criteria is extremely important and went on to derive the criteria for initiation of rocking, sliding, and rocking-sliding.The rocking problem is stiff and highly nonlinear in nature, resulting in a variety of rocking responses even for relatively simple harmonic excitation. For an undamped rocking system, Jeong et al. [11] found that quasi periodic and chaotic motions dominated the response. For a damped rocking system, Wong and Tso [12] showed that out-of-phase harmonic and sub-harmonic responses can be stable, and that all in-phase steady-state rocking responses were unstable. When in-phase periodic vertical acceleration was added to a horizontal periodic excitation, the response changed to quasi periodic and chaotic [11]. The stability regions of harmonic/sub-harmonic responses as well as possible chaotic responses were determined using a discrete mapping technique by Hogan ...