Summary Munk & Hassan have ruled out the possibility of exciting the Chandler wobble by electromagnetic coupling of the mantle to the core. However their argument is based on geomagnetic calculations which are not directly applicable to the problem. Electromagnetic damping of the wobble has been suggested but never quantitatively investigated. In this paper the equatorial components of electromagnetic torque available to act on the mantle are rigorously calculated using a model first developed by Bullard to explain the westward drift of the secular variation, and subsequently successfully extended to account for the irregular ms/decade changes in the length of day. These expressions for the torque components are then used to calculate the response of the mantle, in the form of wobble at the Chandler frequency, to disturbance of the magnetic field in the core. It is shown that electromagnetic core-mantle coupling fails by several orders of magnitude either to generate or to damp the Chandler wobble.