The literature on market segmentation published during the last 10 years is very rich, but unfortunately all of the published studies treat market segmentation as a cross-sectional analysis at one particular point in time. What is lacking is a serious study and description, also of the dynamic aspects of market segmentation. These aspects can be illustrated by the following questions : 1. How do the relative and absolute size of different market segments change over time?; 2. How does market behaviour - expressed in probabilistic terms - change within a given market segment as time goes by?; 3. Is the market segmentation structure which was found to be optimal (according to certain statistical criteria and available data) three years ago still the best available structure for describing and predicting consumer behaviour in 1975? In this paper the first two questions are briefly discussed, while the last two questions are treated in more depth. Questions 3 and 4 are illustrated with data derived from three consecutive cross-sectional analyses undertaken by the author in 1963 1966 and 1971 using the same measurement instrument. The paper ends suggesting to you a model for fully dynamic studies of the market place based on segmentation concepts.