Vibration testing of nuclear reactors is discussed as a part of the determination of the response of such systems to earthquakes. The basic theory of vibration testing is presented along with a comparison of impulse, ambient, and steady-state testing. Steady-state tests provide a method of obtaining the complete dynamic characteristics of a system and of selectively studying each of the components of the system; e.g., containment, steam generator, pressure vessel, instrumentation, etc. Generally, both impulse and ambient studies do not provide as much detailed information while being less time consuming and creating less interference with other operations. A series of tests performed on the UCLA research reactor, the Carolinas-Virginia Tube Reactor, and the Experimental Gas-cooled Reactor at Oak Ridge are used to illustrate results obtained with steady-state tests. These illustrate the effect of the vibrations on instrumentation as well as the response of the reactor cores, fuel elements, biological shielding, steam generators, exhaust stacks, and the containment structures. The tests of the UCLA reactor included tests with the reactor at full power. The examples illustrate the complexity of the soil-structure-reactor system while also indicating the nature of the results which may be obtained with vibration tests.