We address issues of investigations in plant ecology with a time span exceeding the range of ordinary research projects. According to the findings in the papers of this special issue of Community Ecology and other related publications, long-term monitoring is faced with specific methodological challenges:

1. Since the system to be monitored as well as the objectives of monitoring may change over time, the spatial, temporal and thematic scales should be changed considerably.

2. Experience shows that monitoring results are often used for different purposes. However, there exists no sampling design that would offer a true multi-purpose application. The selection of the sampling design will therefore restrict the future use of data.

3. While the selection of sampling units is random and therefore statistical, the selection of variables is usually preferential. This may seriously hamper the results thematically and statistically.

It is concluded that combining precise ground measures with remote sensing data by appropriate mathematical models will be a most promising approach in future monitoring projects.