1. The female-limited colour polymorphic damselfly Ischnura elegans has proven to be an interesting study organism both as an example of female sexual polymorphism, and in the context of the evolution of colour polymorphism. The study of colour polymorphism can also have broader applications as a model of speciation processes. 2. Previous research suggests that there exist correlations between colour morph and other phenotypic traits, and that the different female morphs in I. elegans may be pursuing alternative phenotypically integrated strategies. However, previous research on morphological differences in southern Swedish individuals of this species was only carried out on laboratory-raised offspring from a single population, leaving open the question of how widespread such differences are. 3. We therefore analysed multi-generational data from 12 populations, investigating morphological differences between the female morphs in the field, differences in the pattern of phenotypic integration between morphs, and quantified selection on morphological traits. 4. We found that consistent morphological differences did indeed exist between the morphs across all study populations, confirming that the previously observed differences were not simply a laboratory artefact. We also found, somewhat surprisingly, that despite the existence of sexual dimorphism in body size and shape, patterns of phenotypic integration differed most between the morphs and not between the sexes. Finally, linear selection gradients showed that female morphology affected fecundity differently between the morphs. 5. We discuss the relevance of these results to the male mimicry hypothesis and to the existence of potential ecological differences between the morphs.