Community level ecology is considered to support significantly the recognition of the ecological status of plant taxa and the identification of plant ecogroups, thus it generally provides extended data sets on the spatial and temporal changes of ecological factors. Since research based exclusively on plant structure and their supposed adaptation to the environment is now considered inadequate, statistical methods can be used. Assuming that co-occurrence of plant fossils on a single hand specimen in the case of autochthonous or paraautochthonous floras is the result of their growth in the same phytocenosis, quantitative ecological analysis on Mesozoic materials would yield significant insights. In this paper statistical and multivariate quantitative analyses of Early Jurassic plant fossil records from the Steierdorf Formation in Anina (South Carpathians, Romania) are presented. Four palaeoecological groups of taxa were distinguished by Principal Component Analysis (PCA) and interpreted as plant assemblages of various palaeobiotopes associated with the sedimentary facies of the enclosing formation. A group of samples was analyzed using the Principal Coordinate (PCO) method and the statistical significance (p {less than or equal to} 0.05) (p ≤ 0.05) of individual binary responses of taxa along the first two PCO ordination axes was tested by General Linear Model (GLM). They revealed putative palaeoecological gradients: axis 1 – disturbance caused by water level fluctuations, axis 2 – temperature, corresponding with the already assumed environmental and climatic change at the Hettangian/Sinemurian boundary. Multivariate analyses enabled the identification of palaeoecological groups and thus inferring palaeogeographical conditions based on Mesozoic materials.