Survey data for Norway spruce (Picea abies I.) trees at 12 different forest sites across Europe have been analysed. Three variables (dry weight/fresh weight ratio differences between current and two-year-old needles, ethylene emissions and violaxanthin/antheraxanthin ratios) were found which together show a significant relationship to tree damage in areas affected by forest decline and which are independent of site effects (P < 0.001). Results are also presented that support the contention that atmospheric pollution, especially tropospheric ozone, is the major cause of this problem. After a four-year fumigation experiment, 12-year-old Norway spruce trees showed similar cellular changes as those found in the 30-to 40-year-old trees of the survey affected by forest decline. In the fumigation experiment, these changes were most pronounced in those plants exposed to a combination of treatments including ozone, sulphur dioxide and acid rain. The pollutant concentrations used in the four-year study were representative of those experienced by trees growing in affected areas.