In the forest management unit Radovljica the continuous forest inventory was launched in 1979. In the year 2009 the third repetition of measurments was done in the form of systematic sampling net of 200 x 200 m. The objectives of our work was to assess the positional accuracy of the sample plots and to check what kind of measurement errors have been done ten years ago, when the previous remeasurment was done. We have remeasured 326 permanent sample plots,228 of them were located using the GPS system. The survey showed that 60% of sampling plots centers had different theoretical and actual location for more than 20 m. We have find out that consumption of time has influence on the amount of measurement errors that appear on sampling plots. The survey also showed that estimated number of trees per hectare and configuration of terrain has influence on the amount of errors on sampling plots.