Abstract
Mark Shegelski, Ross Niebergall and Mark Walton are right to say that it is not obvious why a curling stone curls (June p19). However, they are mistaken in suggesting that Jack Allen's description is incorrect (March p17). If a left–right asymmetry exists in the coefficient of friction, as Allen suggests, the resulting transverse force on the stone is non-zero. This is because the frictional force on one side of the stone (the right, for a clockwise rotation) is greater than that on the other side. The amount of left-right asymmetry need not be large; my calculations show that 0.3% is enough to explain the observed effect. What puzzled me, until I read Allen's explanation, is why the coefficient of friction should exhibit any asymmetry at all.