THE work on long-run changes in industrial concentration inaugurated by the classic I956 article of Professors Hart and Prais and continued by them and by others has reached generally agreed conclusions about the direction and magnitude of change. Concentration rose in the first three decades of the century, was stagnant or declined in the following two decades and began to rise again in the I950S and I96os. Professor Hart is correct in pointing to a number of areas of real agreement, and we take this opportunity to reiterate our faith in these basic conclusions. We find some recent selective and polemical 'refutations' of their results wholly unpersuasive.