In the problem of designing installations in which vessels or pipes of mild steel are heated electrically to a temperature not greater than the Curie point, either by conducting into or by inducing in them alternating current at supply frequency, the electrical engineer appears to find little practical assistance in published work. The paper gives some details of actual installations and the empirical formulae developed by the author from numerous tests and used in the design of new installations.In each example the required design values are given explicitly in terms of the data of the installation.