Statistically well-documented sequences of fossil species are rare, as has been observed, for example, by Eldredge and Gould (1972), Boucot (1978) and Dodson and Hallam (1977). Rapid advances in theoretical paleontology made over the last few years have, however, made it imperative that the many new and exciting theories we now have at our disposal be tested against real data (Maynard Smith, 1981). Although suitable data should be available from several sources, it seems that the most likely area for providing good study material is that of micropaleontology. The present paper summarizes the results of mainly genetically oriented statistical analyses of borehole samples of the Cretaceous bolivinid foraminifer Afrobolivina afra Reyment, a species which has been well studied in several connections (Reyment, 1959, 1966, 1980; Castelain et al., 1962; Campbell and Reyment, 1978; Campbell, 1979; Reyment and Reyment, 1979; Campbell and Reyment, 1980). Although these contributions are of significance for the present account, the results given here are new. The quantitative-genetic analysis of the present data owes much to the research of Lande (1976, 1979, 1980b), who has demonstrated how the principles of Falconer (1960) can, in some cases, be successfully adapted for models framed in terms of the evolution of the average phenotype in a population by natural selection and genetic drift. The intent of the present paper is to attempt to apply quantitative-genetic reasoning to the interpretation of morphologic changes in an evolutionary series and to search for eventual interactions between the average phenotype of the organism and the paleoenvironment. It is shown that phenotypic variation in characters of the foraminiferal test is closely associated with fluctuations in the paleoenvironment, as evidenced in the properties of the fossil substrate. These environmental variations seem to have been caused by epicontinental transgressions and regressions. The analysis points to regionally valid weak selective effects as having brought about the morphological changes in the foraminifer, the influence of which seems to have been heightened by the ecophenotypic reaction of the test to the environment; random genetic drift does not appear to have been of first-order importance. Variation in the size of the proloculus of megalospheric individuals is shown to be connected with population density and certain chemical elements. Obviously, quantitative-genetic principles can only be applied to fossil material in a tentative manner and it is important that the limitations of this approach be recognized. The present study should therefore be seen as one possible interpretation of the observed sequence of morphologic changes, and not the only one.