Size and surface morphology of pollen has been studied in 87 twit varieties of 10 fruit species during the period of 1990-1995. No preceding work of that type came to our knowledge, yet. The samples comprised a wide variety of cultivars included male sterile, self-incompatible, partially self-fertile stone fruits, diploid and hexaploid plums, diploid and triploid apples. The large number of species and varieties facilitated the comparison of items within and between the respective species. It was stated that the size, shape and surface morphology of pollen is genetically determined and those data, combined with other variety characters, are suitable for the classification and distinction of varieties. In assessment of pollen size and shape, their moisture content is crucial. The major diameter of the swollen pollen as well as the length and width of the dry grains are characteristic to species and/or to variety. The width and shape changes largely with moisture content. Large grains are proper to quince, apricot, peach and almond, medium sizes are found in apple, sweet cherry, sour cherry, European plum, whereas small size is typical to Japanese plums. The low number of varieties studied does not allow conclusions concerning differences within pears, quinces and almonds as species. In the rest of species, valid differences have been registered as between varieties. Within species, as apple and plum, the effect of ploidy (i.e. number of chromosomes) was expressed in the size of their pollen. In stone fruit species, the correlation between size. of anthers and size of pollen grains was positive. Genetic relations between the self-fertile sour cherry varieties of the Pándy type (Debreceni bőtermő, Kántorjánosi, Újfehértói fürtös) as well as the self-incompatible apricots of "giant" fruit size are supposed to be analysed by pollen studies but there did not turn out any decisive conclusion, yet. Other characters also should be considered. The assembly of pollen characters is decisive in the determination of the variety. The ratio of empty pollen grains, the grain size and the density as well as the size of the pits on the surface are best suited to distinguish pollen lots.