A minor error in Howison et al. (J. Eng. Math. 48:321-337, 2004) obscured the fact that the points at which the free surface turns over in the solution of the Wagner model for the oblique impact of a two-dimensional body are directly related to the turnover points in the equivalent normal impact problem. This note corrects some of the earlier results given in Howison et al. (J. Eng. Math. 48:321-337, 2004) and discusses the implications for the applicability of the Wagner model. © 2012 Springer Science+Business Media B.V.