The formation of crevice corrosion on type 304L stainless steel when immersed in 0.05 M ferric chloride solution has been used to investigate electrochemical potential noise measurements. The surface activity of the stainless steel was simultaneously studied using a scanning electrode technique to provide corroborating evidence of crevice corrosion. The spontaneous potential noise fluctuations were recorded in a freely corroding system with respect to a reference electrode. Power spectral densities calculated by fast Fourier transforms and a stochastic technique were used for the analysis of the time records, in order to reveal fundamental characteristics of the fluctuations resulting from the initiation and propagation of crevice corrosion. A stochastic analysis tool based on a Poisson-process test has been developed and evaluated using data generated by computer simulation of both stochastic and deterministic processes, before applying the analysis technique to real corrosion processes. There appears to be agreement between the two analysis methods; both revealed the presence within the time series of stochastic and deterministic features. Using a combination of noise analysis techniques it was possible to obtain evidence of the different corrosion processes occurring on the 304L stainless steel which were metastable pitting, propagation and termination of localized corrosion events, and the development of crevice corrosion.