Mark Dixon suggests (Forum, Vol. 50, 137) that nobody thus far has attempted to quantify the errors from tilt that arise while observing with the marine sextant. The issue in fact, with the related problem of what exactly is the axis about which the sextant is rotated whilst being ‘rocked’ (to define the vertical), was the subject of a lively controversy in the first two volumes of this Journal some fifty years ago. Since the consensus of opinion seems to have been that the maximum error does not necessarily occur at 45 degrees, whereas Dixon's table suggests that it does, some reiteration of the arguments may be in order.