Disease resistance is a sought-after trait in plant breeding programmes. One strategy to make resistance more durable is to increase the number of resistance genes, thereby increasing the number of pathotypes withstood. One of the most important diseases on roses is powdery mildew (PM) (Podosphaera pannosa). Recent studies show that pathotypes of PM and different types of resistances in roses exist. The results of this study aim to contribute to PM resistance in roses by the development of pathotype-specific markers on a genetic map. A diploid rose population (90 genotypes) derived from a cross between Rosa wichurana and Rosa `Yesterday' was used to construct a genetic linkage map encompassing 20 AFLP primer combinations, 43 SSR, and 2 morphological markers. By applying the F1 pseudo test cross population strategy, two parental linkage maps were constructed (parent `Yesterday' 536 cM; parent R. wichurana 526 cM). Both parental maps consisted of seven linkage groups with an average length of 70 cM (Kosambi) corresponding to the seven haploid rose chromosomes. These new maps were used to identify QTLs controlling disease resistance. The offspring population was screened for resistance to two PM pathotypes, R-E and R-P. QTLs for controlling pathotype-specific disease resistance were mapped by applying Kruskal-Wallis rank-sum tests and simple interval mapping. With two pathotypes analysed, nine QTL loci were detected on linkage groups 2, 3, 5 and 6, explaining 15-73% of the phenotypic variance for pathotype-specific disease response. The genetic maps developed here will be useful for future rose breeding, pathotype-specific resistance research and development of a consensus map for roses.