Statistical methods are applied in order to characterize R–R interval sequences of the ECG. By way of example, the interval distribution and the joint probability density function of pairs of adjacent intervals are treated in detail for data on one normal testee and three patients. The results are compared with the properties of the respective cardiotachograms. These methods of display, if used with discretion, provide a compact overall view of the degree of irregularity of intervals, and of the form of interdependency between the durations of successive intervals.