Habitat fragmentation can affect the morphological design of flying insects in different ways through changes in the costs and benefits of dispersal patterns. Larger wings should be favored if necessary resources are scattered across a fragmented landscape. The purpose of this research is to test this prediction in Bombus dahlbomii (Guérin), a native Chilean pollinator associated with the fragmented Maulino forest. We used hanging traps to collect individuals in the continuous native forest, in four small forest remnants and in two Pinus radiata plantations (matrix). We measured the body size (mm) and the length of the first and the second pairs of wings (mm) of each individual collected. In all, 280 individuals were captured. The body size and the length of both the first and the second pairs of wings were higher in the forest fragments and the matrix than in the continuous forest. A significant positive relationship between wing length and body size was observed for all habitats, but the percentage of variance explained was lower in the continuous forest than in the fragments and the matrix. These results suggest that the fragmentation of the Maulino forest would favor phenotypes with larger wings. It is probable that these phenotypes would be best at flying and obtaining scarce resources during foraging.