We have examined the indirect effect of the herbicide glyphosate on the spider Lepthyphantes tenuis in field margins. Glyphosate was applied to a randomised block design field experiment comprising 360, 720 and 1440 g glyphosate AE ha-1 treatments and an unsprayed control. Spiders were sampled in each month from June to October 1998. Spider abundance was significantly lower in all the treatments than in the unsprayed control. Abundance was also significantly lower in the 720 and 1440 g treatments than in the 360 g treatment. No significant difference could be detected between the 720 and 1440 g treatments. Poisson regression models showed that patterns of decline in L tenuis were related to increasing dead vegetation and decreasing vegetation height. Glyphosate applications only had a within-season indirect habitat effect on L tenuis as field margins sprayed 16 months after an application of 360 g glyphosate ha-1 showed no detrimental effect.