The aim of this study was to present a method for assessing the connectivity of forest areas and the rules for generalisation of forest patches as the spatial planning units. The analyses were carried out for the entire area of Poland. Topographic maps (vector data) as well as software for spatial analysis (QGiS) and database analyses were used. Method of generalisation of forest patches involved two stages. The first one included separation of continuous forest areas consisting in the identification of patches located at a maximum distance of 50 m from each other. Patches generalised into continues forest areas were distributed among eight area classes: over 50 000 ha, 25 000,1−50 000 ha, 500,1−25 000 ha, 200,1−500 ha, 25,1−200 ha, 5,1−25 ha, 0,6−5,0 ha and up to 0,5 ha. The other step included separation of isolated continuous forest patches, which means patches that were out of the range of continuous forest areas in higher area classes. Ranges of continuous forest areas were calculated using 500 m buffer. Results of this study shows that forest patches cover 33,9% of Polish land area. Methods used in this study allowed to reduce input number of patches (764,850) by over 50% up to 338,682 ones. Generalisation of data caused changes in the landscape metrics: mean patch size more than doubled, patch density decreased by more than half, and the largest patch index increased very strongly (forty five times). Based on the literature review and results obtained in this research, it was found that in Poland afforestation intended to conservation of specialized forest species should focus on creating ecological corridors between distinguished forest patches (continuous forest areas). With the current forest cover and forest spatial structure in Poland, afforestation intended to conservation of specialized forest species by reducing edge effect in existing forest patches remains also important, but rather is a secondary need.