On each of two horizontal beds in the Miocene sediments at Calvert Cliffs, Maryland, nine stations were placed in an equally spaced grid. At each station five samples were taken at random, and in each sample individuals were identified to species and counted. A list of species proportions or relative abundances constitutes a vector, p, with rank S = species richness. The statistical distribution of p is defined as the community structure. To identify community structure, species abundances are cumulated in successive samples and the species richness, S, the information function, H, and the evenness, E, are evaluated at each stage. The graph or curve of the cumulated frequency, the distribution function, is identifiable for various statistical distributions and allows for community structure identification. The entire procedure is called SHE analysis for community structure identification, SHECSI. For a log series distribution, the regression on the cumulated frequency lnS = a + blnE has a value of slope b = −1 and of intercept a = H. The results obtained from both beds at Calvert Cliffs are in agreement with the expectation of a log series distribution. Comparison of the results from these fossil populations with results from living and total populations indicates that a log series distribution is the expected distribution for the structure of living and fossil shallow-water foraminiferal communities.