Several series of flume experiments were performed in order to investigate the current response and orientation of models and shells over different substrata. Eighteen different models, of simple symmetrical shapes, and eight species of modern shells, six pelecypods and two gastropods, were used on two different sediment bases, medium sand and mud. All movements of objects were recorded photographically and final measurements were made from the photographs. The experiments were begun with model sets in random orientation, all objects being in the concave-up attitude. From the experiments it was concluded that most objects take up a preferred final orientation of longest axes across the current. Only exceptionally, for example, in the gastropod, Turritella, and the very light weight pelecypod, Cultellus, do current parallel orientations result. Some of the models were loaded with lead to produce an eccentric centre of gravity and to simulate the effect of the thickened umbonal regions of shells. These loaded models indicate the tendency for an up-current preferred orientation of centre of gravity (= load). Some non-loaded models produced unidirectional patterns, reflecting the relative ease of transport in certain positions. By analysing the behaviour of the objects during the experiments it was possible to break down the final orientation patterns into their component parts. This operation demonstrated that the objects are most perfectly oriented during transport and that finally, this pattern can become more or less dispersed due to impedence between objects, irregularities on the substrata and inversion of objects to the convex-up attitude. Experiments on the two substrata indicate that objects can be moved more easily and for greater distances, and are less likely to be inverted on sand than on mud. This is considered to be due, at least in part, to the relative mobilities of the two sediments: sand sized particles are more easily eroded at a given current velocity than are mud sized particles. Observations on the burial of objects indicate that burial begins at lower velocities on sand, and that, by scour, objects can become buried in bizarre attitudes to the sedimentary interface. It is restated that objects are most stable in the convex-up position. They are soon inverted to this position on mud and because of this it seems difficult to develop drifted assemblages on this substratum.