This chapter discusses some of the basic concepts used in probability theory which are of particular importance for the applications. A stochastic variable which can only assume a finite number of values within a limited interval is discrete. A discrete stochastic variable is specified by its probability function. The normal distribution is a continuous stochastic variable. The normal distribution is symmetrical around its mean and unlimited both downwards and upwards. The normal approximation is obtained by calculating the mean and the variance of the binomial distribution, which are then used as parameters in the normal distribution. The Poisson distribution is limited downward by 0 but unlimited upward and is consequently asymmetric. In epidemiology the Poisson distribution is used in connection with person years rather than persons. The number of person years is regarded as a constant and as an interval of time during which a new case of disease can occur at any time.