Social interactions are common for all living organisms. In animal breeding, these interactions are of interest as they are often a source of indirect genetic effects (IGEs). An IGE is a heritable effect of an individual on the trait value of another individual. In aquaculture populations and some plants, social interactions have an additional consequence -interactions in the form of competition inflate variability of trait values among individuals. The phenotypic variability of a genotype has been studied as a quantitative trait in itself, and is often referred to as inherited variability. The main objective of this thesis was to study the genetics of inherited variability, with a focus on the relationship between competition (i.e., IGEs) and variability. In the thesis, we used Nile tilapia as a model species. We found that variability of body weight and body size traits in Nile tilapia is heritable, and shows a large genetic coefficient of variation, which offers good opportunities for improvement of uniformity by means of genetic selection. To study the genetic relationship between social interactions and variability, we developed a quantitative genetic model that integrates both phenomena. In this model, interactions between social partners lead to divergence (competition) or convergence (cooperation) of their phenotypes (e.g., body weight) over their life time. The effects of social interaction in the model are heritable and can evolve. These effects comprise direct genetic effect of the focal individual and IGE of its social partner. With a simulation study we showed that the model yields increased variability of body weight with increase of competition, similar to what is observed in real aquaculture populations. Selection for cooperation will therefore lead to decreased variability. These findings suggest that IGEs may be creating an entire level of genetic variation in variability, that has so far been overlooked. Using existing statistical models, we show that direct genetic effects of competition on variability could be captured with a direct model of inherited variability, and similarly, IGEs of competition could be captured with an indirect model of inherited variability. According to kin selection theory individuals should show better social behavior, i.e., less competition, towards relatives, which should be reflected in their body weight and the variability thereof. We tested this hypothesis by comparing two treatments in an experiment, in which tilapia were reared in either kin or in non-kin groups. Individuals had significantly higher body weight in kin groups, however, there was no difference in variability of body weight between the two treatments.Findings of this thesis demonstrate that variability of body weight in tilapia is heritable and that genetic variation in variability may comprise not only direct genetic effects but also IGEs. Studies focusing on evolution of variability/uniformity, therefore, should consider IGEs.