Non-target-site resistance (NTSR) comprises a set of mechanisms conferring resistance to multiple modes of action. Investigation of the number of loci involved in NTSR will aid in the understanding of these resistance mechanisms. Therefore, six different multiple herbicide-resistant Alopecurus myosuroides plants with different herbicide history were crossed in two generations with a susceptible wild type. Seeds from the backcrossing generation were studied for their segregation rate for resistance to five herbicides with four different modes of action (HRAC groups C 2 , A, B and K 3 ). Taking into account that NTSR is a set of quantitative traits, the numbers of loci controlling NTSR were estimated using a normal mixture model fitted by the NLMIXED procedure of SAS. Each herbicide was controlled by a different number of loci comparing the six plants. In most of the cases, chlorotoluron resistance was controlled by one locus, whereas resistance to fenoxaprop-P-ethyl needed one or two loci. Resistance to pinoxaden was in all plants conferred by two loci. Cross-resistance of fenoxaprop-P-ethyl and pinoxaden was found in all backcrossings, indicating that at least one of the two loci is responsible for both resistances. Resistance to mesosulfuron + iodosulfuron was conferred by a minimum of two loci. Results indicated that a minimum of five different loci can be involved in a multiple NTSR plant. Furthermore, the plant-specific accumulation of NTSR loci was demonstrated. Such behaviour should be taken into account when evaluating the development and further spread of herbicide resistance.