Introducing indigenous plants into cultivation and breeding programs in order to obtain new taxa can create the undesirable effects and pose a real threat to natural populations. The main problem of our study was to assess whether the phenotypic characters might allow distinguishing cultivated Convallaria majalis from plants occurring in natural habitats. The range of phenotype variability of Convallaria majalis L. shoots from forests and gardens growing in different edaphic factors was studied in 2005–2006 in north (Gdańsk), central (Warsaw) and south (Cracow) regions of Poland. Ten or eleven forest locations and the same number of garden locations were designated in each region. The soil at the forest sites was poorer in phosphorus and calcium and more acidic than at the garden sites from which the cultivated plants were obtained. Thirty flowering shoots were collected from each study site. The numbers as well as size of flowers, size of racemes and leaves were investigated. The following hypotheses were formulated: (a) due to conscious and unconscious selection, cultivated plants should have more numerous and larger flowers and a longer inflorescence (these characters are most significant for the ornamental appearance of the plant) than plants growing in forests; (b) cultivated plants exhibit less phenotype variability than do natural populations. The forest plants from all the three regions together had, in comparison with garden plants, significantly shorter racemes with fewer flowers, a narrower and shorter perianth, and a longer lower leaf with a narrower and longer leaf blade. In the forest plants raceme length exhibited the highest variability while the lowest was noted in perianth length. More morphological differences were noted between the forest and garden plants than among those from the three different regions. The results obtained indicate that the level of variability in the studied characters was similar in forest and cultivated plants. A high level of variability in some morphological characters of lily of the valley probably stems from the high phenotype plasticity of this plant and the diversified biotope conditions in which it occurs. The significant differences between forest and garden plants could have been related to differences in habitat conditions.