The primary purpose of this investigation was to determine whether or not a nonrandom distribution of sorocarps occurs in Polysphondylium pallidum, and to determine the factors that are instrumental in determining distribution. While a nonrandom distribution of centers does occur, the determining factor is probably not a substance present in the gaseous phase of the environment. There is a correlation between the density of sorocarps on the substrate and their spatial distribution. The clustered condition (sorocarps closer together than expected on the basis of chance) is associated with a high density of fruiting bodies, while a spaced distribution (sorocarps further apart than expected) is often found at low densities. Random distribution apparently can occur at almost any density. Center density depends upon the rate of center formation. This rate is markedly enhanced if the cells are in the stationary phase at the time of harvest and deposition, and are exposed to charcoal, mineral oil, or light. The latter factors are thought to remove or act upon a center suppressing substance in the environment. Cells taken from the logarithmic phase, incubated in the dark, and in the absence of charcoal and mineral oil show a reduced rate of center formation and a low center density. Three hypotheses are presented to account for the apparent relationship between the rate of center formation and center density, and the spatial distribution of centers: (1) A spacing substance surrounding the center may prohibit other centers from forming in the immediate area; (2) spacing may result from the removal of cells through aggregation in the area immediately surrounding the center; (3) spatial distribution may be determined by the action of acrasin. In all cases, the longer the interval between the appearance of one center and another, the greater the area controlled by the first center and the more spaced the distribution. The “acrasin” and “cell-depletion” hypotheses are favored for their equal acceptability as an explanation and because the postulation of an additional, unknown factor is not required. The dependence of sorocarp density and distribution upon cell pool size is considered. It is suggested that high cell number favors both randomization, by leaving enough cells after the first burst of aggregation to form secondary centers, and clustering, because aggregates that contain a large number of cells often split into closely associated secondary centers. Low cell number results in areas virtually free of myxamoebae surrounding newly formed centers, a condition which strongly favors a spaced distribution. The possible relationship between spacing in cellular slime aggregation and spacing in other developing, biological systems is considered.