Information on the development and testing of a model of spatial diffusion and adoption of an agricultural innovation is presented. The data concern the practice of artificial insemination of cattle in southern Sweden. The model is intended to predict the time at which adoption occurred in terms of the time at which the innovation became available in the adopter's neighborhood, the time at which the adopter first heard about the innovation, and the adopter's set of personal acquaintances, age, and economic status. Eighteen critical variables are used in the model, including availability and cost of the innovation, projected benefits, existence of a support infrastructure, agricultural income of the farmer, and size of the farmer's herd. Test results show that the innovative farmers were consistently difficult to classify and that the linear relationship, which was broadly applicable to the early majority adopters, late majority adopters, and laggards, was inapplicable to them. The need for further research on models of this type is indicated. References are included. (Author/DB)