I venture to lay before you certain considerations concerning the viscosity of liquids which may find a place in a discussion of the theory of liquids. They mostly pertain to experiment, but, after all, the purpose of physical experiment is to stimulate the theoretical physicist. I shall deal briefly with three aspects of viscosity: the viscosity of molten metals, which represent the simplest class of liquid; the influence of an external electric field on the viscosity of liquids; and the viscosity of liquids in the supercooled state. In the last two cases a positive effect has been found in the case of polar liquids only. As regards the viscosity of the simplest liquids, the difficulties in the way of a mathematically satisfactory theory, which shall lead to results that can be checked by experiment, are formidable, as will be agreed by anyone who has studied the papers of Born & Green, in which the greatest ingenuity and exceptional mathematical powers have not, as far as viscosity is concerned, produced formulae free from arbitrary constants for comparison with laboratory measurement. This being the case, a cruder picture that leads to figures in close agreement with experiment may perhaps be pardoned, until the pursuit of greater perfection brings something more precise and more firmly founded.