The second-order polyspectrum is the Fourier transform of the second-order comulant function and is defined without requiring stationarity. It is the generalization of the power spectrum to the analysis of nonstationary processes. The second-order polyspectrum is capable of measuring, identifying, and characterizing the potential information concerning the second-order time-dependencies of a nonstationary process. In this paper, we apply the second-order polyspectrum. We use the second-order polyspectrum as a diagnosis and classification tool for simulated data, biological data, and sonar data. Finally, we investigate a beamforming procedure for nonstationary seismic data.