Abstract Graphical displays were used early in geophysics and crystallography, mineralogy, petrology and structural geology by the early 1800s, but nineteenth-century geology obstinately remained mainly descriptive. Charles Lyell's quantitative classification of the Tertiary Sub-Era in 1828 was a notable exception. Nevertheless, by 1920 the quantitative approach had become established. W. C. Krumbein, who introduced the computer into geology in 1958, encouraged use of probabilistic sampling and process-response models. Early work focused on databases, statistical data analysis and display. By the 1970s, stochastic simulation, deterministic modelling and spatial 'geostatistics' (pioneered by Matheron and his co-workers), were of growing importance. The introduction of the personal computer and the graphical user interface in the 1980s brought well-proven quantitative methods out of the research environment onto the workbench and into the field. Since the mid-1980s, the analysis, display and modelling of behaviour in three dimensions, underpinned by spatial statistics, computational fluid-flow, visualization technology, etc., has proved of economic benefit to mining, petroleum geology and hydrogeology. Other, computationally intensive, methods likely to be of importance in the Earth sciences are the application of 'robust' statistical methods, increasing use of Bayesian methods in uncertainty (risk) estimation (as a result of a renewed interest in statistical intervals and forecasting), and computational mineralogy.