A model was used to calculate the rate of carbon exchange by Fucus spiralis L. in the field. The model combines laboratory measurements of the effect of photon irradiance, desiccation and temperature on photosynthesis and respiration in air and water, with a simulation of changes in tidal level over a lunar month, received photon irradiance for different coastal water types and fieldmeasured coefficients of water loss (D). The model predicts field growth rates of 1.3 and 2.1 % (m01 C mol-' C d-l) in April and 2.1 and 4.8 % in July for the highest and lowest values of D, respectively. These values are in general agreement with published estimates of growth rates for F. spiralis in the field and laboratory. The net production per tide is very variable, even under the regular diurnal changes in photon irradiance used in the model. This is a result mainly of the length of time spent in light and dark per tide, but partly a response to whether F. spiralis is in air or water around midday, and whether maximal production is possible in air or water, which will depend largely on the value of D. The value of D was the major influence on total net production, and the relative contribution of air and water Under Jerlov's 'coastal 7' water type, net production when D was the lowest measured in the field was 1.6 and 2.3 times greater than that when D was the highest measured in the field for April and July respectively, but only 0.63 for Apnl, and 0.53 for July, of that possible if desiccation was absent. In April, water contributed between 49 and 55 % of the total net production for Jerlov's 'coastal 13' and 'coastal 1' water types when D was the lowest, and 82 and 84 % for these water types when D was the highest measured in the field. Similar values for July are 23 and 28 % when D was low and 56 and 63 "h when D was high. The effect of &fferent coastal water types on total production was fairly small. Under field values of D, spring tldes were between 1.85 and 2.3 times more productive than neap tides for both months. Production was calculated assuming F. spiralis was present at higher and lower positions on the shore. Under all conditions, except for the extreme case of no water loss, production decreased markedly higher up the shore, and increased down the shore, with maximal production between 1 and 3 m lower depending on conditions. These calculations reinforce the suggestions that the position of F. spiralis on the shore is not determined by carbon balance per se.