Similarly, plots of colony scores for PCs 1 and 2 made evident that body size differed significantly between sampling periods, with the November bees clustered around the lowest values of PC1 and the bees from May grouped at the other extreme (Fig. 3).