A New Class of Faults and their Bearing on Continental Drift

T and half-shears. Many geologists1 have maintained that movements of the Earth's crust are concentrated in mobile belts, which may take the form of mountains, mid-ocean ridges or major faults with large horizontal movements. These features and the seismic activity along them often appear to end abruptly, which is puzzling. The problem has been difficult to investigate because most terminations lie in ocean basins. This article suggests that these features are not isolated, that few come to dead ends, but that they are connected into a continuous network of mobile belts about the Earth which divide the surface into several large rigid plates (Fig. I). Any feature at its apparent termination may be transformed into another feature of one of the other two types. For example, a fault may be transformed into a mid-ocean ridge as illustrated in Fig. 2a. At the point of transformation the horizontal shear motion along the fault ends abruptly by being changed into an expanding tensional motion across the ridge or rift with a change in seismicity. A junction where one feature changes into another is here called a transform. This type and two others illustrated in Figs. 2b and c may also be termed half-shears (a name suggested in conversation by Prof. J. D. Bernal). Twice as many types of half-shears involve mountains as ridges, because mountains are asymmetrical whereas ridgos have bilateral symmetry. This way of abruptly ending large horizontal shear motions is offered as an explanation of what has long been recognized as a puzzling feature of large faults like the San Andreas. Another type of transform whereby a mountain is transformed into a mid-ocean ridge was suggested by S. W. Carey when he proposed that the Pyrenees Mountains were compressed because of the rifting open of the Bay of Biscay (presumably by the formation of a midocean ridge a.long its axis). The types illustrated are all dextra.l, but equivalent sinistral types exist. In this article the term 'ridge' will be used to mean midocean ridge and also rise (where that term has been used meaning mid-ocean ridge, as by Menard" in the Pacific basin). The terms mountains and mountain system may include island arcs. An arc is described as being convex or concave depending on which face is first reached when proceeding in the direction indicated by an arrow depicting relative motion (Figs. 2 and 3). The word fault may mean a system of several closely related faults. Transform faults. Faults in which the displacement suddenly stops or changes form and direction are not true transcurrent faults. It is proposed that a separate class of horizontal shear faults exists which terminate abruptly at both ends, but which nevertheless may show great displacements. Each may be thought of as a pair of half. shears joined end to end. Any combination of pairs of the three dextral half-shears may be joined giving rise to the six types illustrated in Fig. 3. Another six sinistral forms can also exist. The name transform fault is proposed for the class, and members may be described in terms of the features which they connect (for example, dextral transform fault, ridge--convex arc type). The distinctions between types might appear trivial until the variation in the habits of growth of the different types is considered as is shown in Fig. 4. These distinctions are that ridges expand to produce new crust, thus leaving residual inactive traces in the topography of their former positions. On the other hand oceanic crust moves down under island arcs absorbing old crust so that they leave no traces of past positions. The convex sides of arcs thus advance. For these reasons transform faults of types a, b and d in Fig. 4 grow in total width, type f diminishes and the behaviour of types c and e is indeterminate. It is significant that the direction of motion on trarn,form faults of the type shown in Fig. 3a is the reverse of that required to offset the ridge. This is a fundamental difference between transform and transcurrent faulting.