The aim of this work was to characterise macroporosity of soil preparations by means of image analysis. Preparations were made from the samples taken from 0-20 cm layer of four types of soil: typical soil lessive developed from loess, proper rendzina developed from cretaceous marl, chernozemic rendzina developed from chalk and mud-gley soil developed from clayey silt. The values of porosity by image analysis, PK, (pores of the diameter greater than 200 µm) were compared with the values of field air capacity, (PPP. pores greater than 20 µm) and total porosity, determined by classical methods used in soil science. For most of the samples porosity obtained by computer image analysis, PK. was lower than the values of field air capacity, PPP., as expected. For the rest of standards, the values obtained morphometrically, PK. were higher than PPP. it probably resulted from unequal pore sizes in the dried and moistened soil material and innate attributes of the method of soil water desorption. Statistical analyses revealed strong, statistically valid dependencies, with high values of the correlation coefficients (0.883-0.951) between porosity, PK total porosity, PO, and field air capacity, PPP. Moreover, correlation indices calculated for Po and Ppp confirmed the already known relation.