AbstractThe friction coefficient of linear macromolecules is calculated for the correlated chain model using KIRKWOOD's approximation. This model (proposed by SANCHEZ and FRANKENBERG) incorporates a certain long‐range correlation between the segments of a random chain, and contains two parameters: CF, which depends on this correlation (CF = 1 for a random coil and CF = 0 for a fully extended chain), and the radius of gyration RG. In the limit of long chains, the friction coefficient B is a function of these two parameters and the ratio B/6πη0RG (η0 = solvent viscosity) decreases with CF. This means that the correlation between segments has a greater influence on RG than on the equivalent STOKES radius of the macromolecule. These results are used to interpret the influence of excluded volume on B. The validity of the model is ascertained by comparing the theory with the light scattering envelopes and friction coefficients recently reported by MIJNLIEFF et al. for poly(α‐methylstyrene) in good and poor solvents.