Cattle activity or grazing s.l. can be subdivided into three components: dung deposition, herbage removal (foraging or grazing s.s.) and trampling. All these actions modify vegetation. At medium or large scale, the pattern of cattle activity is generally described only as the foraging behaviour. Such a description implicitly consider grazing as the principal behaviour of the three primary activities. Our purpose was first to determine in an observational study the medium-scale distributions of dung-pat density, trampling effect and herbage removal in a mountain wooded pasture. These distributions were related to 'natural structures', such as slope, vegetation openness, cover of trees, shrubs and rock outcrops, fodder potential, and 'management-induced structures', such as distance to fence or to the nearest watering place. Results showed that the three variables describing cattle activity exhibited significantly different spatio-temporal patterns. Moreover, the relative influence of environmental factors was different for each activity. Secondly, in an exclosure experiment we simulated the fine scale effects of these factors, separately or in combination, and compared them with cattle grazing over a one-year period. Multivariate analyses of vegetation data in the first year showed an overwhelming seasonal shift and significant differences induced by treatments. Thus, grazing alone appears to be an unrealistic indicator of cattle activity and it might be necessary to consider dunging, trampling and grazing separately in spatially explicit models of vegetation dynamics.