Tumors in vivo usually exhibit deceleratory rather than exponential growth kinetics. Deceleration results from a negative feedback inhibition of growth the effectiveness of which increases progressively with tumor size. A major obstacle to understanding the fine mechanisms of deceleratory growth has been the lack of a tissue culture system proven to exhibit these kinetics. Rat C6 glioma cells in culture enter a prolonged phase of deceleratory growth about 72 hr after subculture. As with tumor growth in vivo, the kinetics of the C6 deceleratory phase can be described with high accuracy by Gompertz, logistic, inverse cube root, and power functions. The least-squares correlation coefficients for the linearized forms of the rate equations for these functions were typically in the range of 0.75 to 0.90. During the deceleratory phase, the equilibrium growth rate of C6 cells is a monotonically decreasing function of population density at all observable densities, whether sub- or supraconfluent. Any change in density produces a compensatory change of opposite direction in the growth rate. To account for these relationships, a density-equilibrium hypothesis of C6 growth regulation is presented. The fine mechanisms responsible for this regulation do not involve medium or substratum depletion, production of conditioned medium factors, or the extracellular matrix. The development of growth deceleration appears to correlate with the extent of contact interactions between cells and their neighbors, indicating that deceleratory growth regulation is probably mediated by contact interactions. However, these interactions are fundamentally different from those postulated by the traditional contact inhibition theory. Confluency is irrelevant to the contact modulation which produces C6 growth deceleration. Any degree of contact, even at very sparse subconfluent densities, appears capable of exerting some degree of growth inhibition.