Power spectra of cyclical lacustrine strata. Fourier analysis seeks to reproduce an irregular curve (such as a depth-rank curve) using a series of sine and cosine functions. The power spectrum gives the relative contribution of the individual sine and cosine functions to the overall curve. The output gives a series a peaks which are the predominant cycles within the stratigraphic section. For example, in the Lockatong and lower Passaic Formations, the predominant cycles are 3.9, 20, and 74 m thick, which yield the typical 1:5:20 Milankovitch ratios. In the uppermost Passaic Formation, the predominant cycles are 5.4, 28, and 112 m thick, also yielding the 1:5:20 ratio. This is the some of the strongest evidence that the lake-level cycles in the Newark basin result from Milankovitch cycles. Modified from Olsen et al. (1996).