Sex-limited colour polymorphism occurs in several animal taxa and is usually explained in the context of sexual selection. Specifically, for polymorphism restricted to the female sex, multiple phenotypes may have evolved in response to male harassment. Such male harassment is generally considered to entail differential costs to female morphs, which may ultimately result in fitness differences. However, contrary to this prediction, most previous studies do not support that female morphs (andromorphs and heteromorphs) differ in measures of quality and (or) fitness components. In this study, we evaluate quality and fitness differences between mated female morphs of the damselfly Enallagma cyathigerum. We suggest that many earlier studies may have failed to observe morph differences in quality or fitness because selection by male harassment was weak. Here, we selected a study population for which our expectation was that levels of per female capita male harassment were high. Nevertheless, also in this population mated female morphs did not differ in body size or condition (body mass/body length). However, mated female morphs did differ in levels of developmental instability: heteromorphs consistently showed a higher level of fluctuating asymmetry than andromorphs. Also, mated female morphs differed in fecundity: andromorphs had a lower clutch size than heteromorphs. In addition, larger females contained more eggs, but the slope of this relationship was steeper in heteromorphs. In conclusion, mated female morphs of the damselfly E. cyathigerum at our study site clearly differed in one quality estimate (developmental instability) and in our measure of fitness (fecundity).