Abstract
Analytical and simulation models were used to study the influence of the number of host genotype units (individual, independent units of host tissue that are genetically homogeneous) on the development of epidemics in pure and mixed (diverse) stands of crops. Analytical models indicated that the alloinfection/autoinfection ratio and, therefore, disease control derived from genetic diversification is greater for populations with a large number of host genotype units than for populations with a smaller number of host genotype units. For computer simulations of wheat stem rust [Puccinia graminis] over large geographic areas, the effectiveness of interfield diversification for disease control increased with increasing number of fields planted to wheat. The absolute severity of disease however, increased for both diversified and nondiversified regions when the number of fields was increased. Simulation studied of oat crown rust [P. coronata] for small plots showed that the number of host [Avena sativa] genotype units in a population is a more important determinant of the effectiveness of mixtures for disease control than is host genotype unit are (the ground area occupied by a host genotype unit). Results suggest that intraspecific or interspecific mixtures of large plants and the culture of alternating rows, swaths, or fields of different host genotypes may provide greater disease control than previously anticipated.