lations involving an exact model of the valence: band of silicon [77] indicate that the anisotropy of the bands, while not important in the interpretation of mobility and resistivity data, plays an important role in the evaluation of the Hall factor. Our model averages out the anisotropy of the bandstand thus our values of the mass anisotropy factor, rA, are larger than an exact model would predict. A calculation based on the exact model, applicable over the entire range of interest of temperatures and dopant densities, is beyond the scope of this work. At the higher dopant densities there is fair agreement between the theoretical prediction and values of Hall factor deduced.from our experimental work and the data of Wolfstirn [15]. Data points from Wolfstirn's work showing ionization energies greater than 0.065 eV were not included in Figure 7.13 because of the high degree of compensation of these samples.