The variation in the content and composition of Calendula officinalis essential oils was studied using supercritical CO 2 extraction followed by GC-MS. Samples of marigold were harvested at four different vegetation stages. A total of 43 different compounds were identified in the essential oils. The identified components were represented mainly by monoterpenes, sesquiterpenes and sesquiterpene alcohols. The qualitative and quantitative composition of the essential oils of C. officinalis varied during the studied vegetation stages. To determine the significance of changes in the identified compounds, to classify the obtained plant samples into groups and to find the similarity measure between them, several statistical data mining techniques were applied, namely hypothesis testing, principal component analysis and hierarchical cluster analysis. a-Thujene, a-pinene, c-cadinene and d-cadinene were predominant in all vegetation phases. Development of buds and the end of the flowering stages were distinguished by the highest content of essential oils, while full-flowering stage was distinguished by sesquiterpenols (1,10-di-epi-cubenol, 1-epi-cubenol, s-and a-cadinols, b-selinelol constituted 11.6% of the oils).