The type II emissions of August 7--8, 1972, are reexamined on the basis of two proposed production mechanisms, one in which the emissions are produced in the ambient plasma and one in which the emissions are produced in the shocked plasma. We first develop a new model for the radial dependence of the solar wind density in which the density for large distances goes as the inverse square of distance, on the basis of the conservation of particles and the observed lack of radial dependence of the solar wind velocity. We note that the density model derived from type III bursts may not apply directly to type II bursts because of solar wind density structures due to stream interactions. We show that the model in which the emissions are produced in the shocked plasma provides the more satisfactory explanation of the data.