Unexpected fouling in condensers on Central Electricity Generating Board power stations operating on the river Trent in the UK in the 1970s prompted an extensive investigation of the phenomenon. Fouling was caused by deposition of calcium phosphate rich scale on the cooling water side. A fouling test rig was specially constructed to study the performance of different chemical treatments. The rig ran two tubes in parallel, one with undosed water, and the latter's data sets from over 100 "control" tests were analyzed to determine the mechanism and rate of fouling. The trends could not be explained in terms of conventional scaling indices. Insight into the mechanism was provided by separating precipitation (loss from solution) and deposition (adhesion of some of the precipitated mass to the heated surface): scaling was found to be caused by the deposition of particulates in the recirculating cooling water, driven by the change in pH as the water went through the evaporative cooling stage. Quantitative expressions for fouling were generated from tests on 1.83 m long tubes and were found to give a reasonable prediction of the fouling behavior observed on a full sized (18.3 m long) single condenser tube on an operating condenser. The results indicated that recirculating water systems need to be approached very differently to once-through systems, and the use of scaling indices for such systems is not recommended.