The capability of clonal plants to reproduce both sexually and vegetatively and their resulting hierarchical organisation into ramets and genets pose problems for the determination of fitness. We develop ramet-based measures of clonal plant fitness including both modes of reproduction. To this end we use simple population-dynamic equations accounting for limited space, limited dispersal, and disturbance. Neglecting interactions among ramets (r-selection regime) we derive an expression for initial growth rate r as a fitness measure. At higher densities interactions among ramets lead to density control (K-selection regime) and competitive exclusion of genotypes or species may occur. In this case we apply an invasibility criterion to derive an abundance fitness measure: competitiveness C. The optimum of C corresponds to an evolutionary stable strategy. C increases with the proportion of reproductive adults, with inverse module mortality, and with the sum of sexual and vegetative reproduction. The latter are defined as the product of the rate of module production and two correction factors accounting for juvenile mortality before establishment and for the efficiency of space exploitation by propagules. Thus C is equivalent to potential life-time offspring production, in contrast to realized life-time offspring production R₀. Because the correction factor for space exploitation cannot be expressed analytically we obtain it from complementary spatially-explicit individual-based simulations. To illustrate an application of the fitness measure C we predict optimal allocation to vegetative and sexual reproduction. In a homogeneous habitat an intermediate allocation may maximize fitness only if there is a non-linear trade-off between the modes of reproduction. However even if this trade-off is linear, spatially heterogenous disturbances can lead to an intermediate optimum of allocation because seed dispersal with subsequent vegetative spread improves the utilization of available space for recruitment if the spatial extent of single disturbances is much larger than the distance of vegetative dispersal. Thus, our study underlines the importance of spatial features for fitness measures especially of clonal plants.