The paper is concerned with refining two well-known approximations to the Reed–Frost epidemic process. The first is the branching process approximation in the early stages of the epidemic; we extend its range of validity, and sharpen the estimates of the error incurred. The second is the normal approximation to the distribution of the final size of a large epidemic, which we complement with a detailed local limit approximation. The latter, in particular, is relevant if the approximations are to be used for statistical inference.