Biological crusts (BCs) are soil ecological engineers critical in restoration yet their occurrence, and relationship with vegetation in a primary succession are poorly understood. The objectives of the study were: (1) to determine BC cover changes on a chronosequence and toposequence of tailings dumps and their correlation with vegetation structural and substrate attributes and (2) to determine vegetation community changes along a BC cover gradient. We hypothesized that: (1) BC cover significantly increased with tailings dump age and that it was positively correlated to vegetation cover, species richness, density, and (2) vegetation communities significantly varied along a BC cover gradient. Occurrence patterns of BCs and their relationship with vegetation attributes and substrate properties were assessed on six gold tailings of different ages (10, 17, 30, 82, 92 and 110 years). BC cover and vegetation attributes were studied in 1-m2 quadrats. Texture, electrical conductivity (EC), total nitrogen (N), organic carbon (OC) and available metals (Zn, Ni and Mn) for surface tailings in patches with and without BCs were determined for each dump. Data were analyzed using two-way ANOVA, ordination analysis and Spearman correlation analysis. Two-way analysis of variance shows that age and topographic position had significant effect on BC covers. BC cover significantly increased with age as hypothesized, to a peak at 92 years and declined thereafter. Cyanobacteria and algae significantly decreased with age, while lichens and mosses increased significantly. A canonical correlation analysis (CCA) revealed a positive correlation between BC and vegetation cover, species richness and density on the four youngest dumps (10–82 years) which was however negative on the two oldest dumps (92 and 110 years). Areas of different BC cover classes (high = 70–100%, medium = 30–69%, low = 6–29% and very low ≤5%) were associated with different vegetation communities, with high and medium classes forming a similar vegetation community. A principal component analysis (PCA) based on tailings physicochemical properties showed that BC plots on the three younger dumps separated from those on the three older dumps, mainly based on variation in silt and total nitrogen content. The results point to the existence of a succession stage dependent relationship between BCs and vegetation. This suggest existence of a potential mutual relationship between BCs and vegetation on mine tailings, thus BCs can be used as proxy indicators of vegetation development. Canonical correspondence analysis, Chronosequence, Correlation, Cyanobacteria, Lichens, Nitrogen, Organic carbon, Species richness, Vegetation cover