A secondary fault is defined as a fracture which arises as a direct result of movement on a master transcurrent fault. Some previous approaches to the study of secondary faulting are discussed, and fallacies in the arguments of McKinstry (1953) and Moody and Hill (1956) are pointed out. The effect of movement on a fault is to reduce the initial shear stress everywhere except in the vicinity of the ends of the fault, where it causes complex additional stresses (see first paper in this series on the theoretical aspects of secondary faulting). Thus it is proposed that secondary faulting is an end effect of a master shear movement, and on this basis six major modes of secondary faulting, labelled types A to F, are described. The usefulness of these results in the analysis of fault systems is illustrated by applying them to the Alpine, San Andreas, and Mac Donald faults. In each case it is possible to predict or explain the curvature, location, and sense of the secondary faults in the area. In addition, the development of the master fault may be traced by locating the ends of the shear zone at various times in the past.