Abstract
To investigate cycles or chaos caused by frequency-dependent selection, we relied on a model by Rice (2004), itself modified from previous models of frequency dependence in diploids (Altenberg 1991; Gavrilets and Hastings 1995; Cockerham et al. 2015), by letting selection coefficients (instead of fitnesses) be frequency dependent. The model focuses on a bi-allelic locus, where the relative fitnesses