There is at present a need to develop a better technique for measuring the rate of icing on structures such as, for example, overhead transmission lines. For aircraft and helicopter icing, the most widely used method of measurement is the rotating cylinder. However, for measuring the icing of structures, this method is difficult to apply and also less accurate due to lower wind velocities. Different approaches are now being developed using fixed cylinders. Icing tests were conducted with fixed and rotating cylinders in a wind tunnel. The rate of icing was obtained through measurements of volume, accretion cross-section and time of deposition. Tests were made using five different liquid water contents and droplet diameter spectra, and four cylinder diameters, keeping the wind velocity and temperature constant. The rate of icing is presented as a function of the diameters of the fixed and rotating cylinders for each of the liquid water contents tested. Results indicate that at lower wind velocities the accretion rate is overestimated for the smaller rotating cylinders. This difference is probably due to the variation of the collection efficiency with diameter. From these results it is suggested that the rate of ice accretion on structures should be based on at least two fixed cylinders of different small sizes in order to take into account the effect of the collection efficiency.