Fluctuating asymmetry as an index of fitness: causality or statistical artifact?

During the last decade, the study of fluctuating asymmetry (FA) in relation to different fitness aspects has become a popular issue in evolutionary bi ology, FA is suggested to be important as an index of selective advantage i n both sexual and natural selection. Evidence for these hypotheses is obtai ned from field studies in which FA is found to be negatively correlated wit h various fitness components. In this paper, we demonstrate that the appare nt significance of FA may depend on assumptions of linearity between size o f a trait and FA, and between trait size and fitness, By means of numerical examples, me show that even small, and probably common, deviations from th e assumed linearity can bias the statistical analyses to such an extent tha t it casts serious doubt on the validity of the postulated causal relations hip between FA and fitness. We recommend reanalyses of existing data and em phatically suggest that the underlying statistical assumptions are critical ly evaluated in future studies of FA, especially concerning linear relation ships between the variables.