In an environment with a wide seasonal range of radiation and vapour-pressure deficit, the energy and aerodynamic components of Penman's evaporation formula were calculated daily for a year, and the potential evaporation was determined from several sources of data and by alternate modes of computation. Reference values for the energy component were obtained from measured net radiation over a well-watered crop. Reference values for the aerodynamic component were calculated from wet- and dry-bulb temperatures and windspeed using a relationship defined by Penman. Total radiation and relative duration of sunshine were found to be effective alternative sources of data for the energy component. When calculating the effective terrestrial radiation, it was found that equally satisfactory results could be obtained from a relationship based on proposals by Swinbank as by Penman's method which includes a vapour pressure term. Daily Piche and tank evaporimeter data were found to be satisfactory alternatives to the aerodynamic component. The conclusions reached are: ( 1) The minimal instrumentation needed to determine potential evaporation is a net radiometer, a Piche evaporimeter, and maximum and minimum thermometers, provided a suitable relation between the Piche and aerodynamic component is available. If a solarimeter is used in place of a net radiometer or sunshine recorder, a reliable albedo coefficient is necessary. When a sunshine recorder is used, a satisfactory relation between n N and Q t Q e is also required. is also required. ( 2) The use of inappropriate constants in the Penman formula is probably a greater source of error when determining potential evaporation than the deficiencies inherent in a particular type of instrument, e.g., a sunshine recorder. The constants appropriate to the particular climatic environment must be used.