Fig. 4: No evidence for deleterious pleiotropy constraining wing shape evolution.
From: Macroevolution along developmental lines of least resistance in fly wings
a, Loadings of fitness components measured in adults and offspring on PC1 all have the same sign, suggesting that different fitness components are correlated, which indicates a main axis of genetic quality. b, PC1 explained more variation than expected by chance (the red dashed line indicates the observed variance; the grey bars indicate the null distribution). c, Despite significant genetic variation in wing shape and fitness, we found no relationship between wing shape and the five fitness components investigated (two-block PLS analysis). d, There was also no relationship between the isofemale lines’ fitness (here indicated as scores on PC1) and their multivariate distance to the average wing shape, as expected if wing shape was under stabilizing selection.
