ON THE THEORY OF NORMAL GRAIN-GROWTH

The theory of normal grain growth that invokes random walk is further extended by allowing for biases in step direction. It is concluded that the predictions of this theory are in overall quantitative accord with experiment. However, the degree of agreement between theory and experiment as to the exact form of the grain size distribution function remains in doubt because of a lack of precision in the definition and measurement of grain size. Possibly, this difficulty may be surmounted by replacing measurements of grain size by those of the lengths of grain boundary edges as found in planar sections of polycrystals.