The use of behaviour genetic heritability analysis to study demographic behaviour is fraught with problems. We explain the concepts and methods used by behaviour geneticists, which are based on Fisher (1918) and Jinks and Fulker (1970), point out their deficiencies, and show that the basic assumptions of the behaviour genetic model do not hold. A behavioural trait should be analysed not by using heritability but by using the coefficient of intensity of inheritance. Confusion between statistical concepts and heritability abounds. Fertility differs from other behavioural traits in many respects. It is affected by many known environmental factors. Male and female fertility are affected by different factors and should be studied using different techniques. Galton’s 19th century idea of nature-nurture or Fisher’s early 20th century genetics have little use in the genomic era. We need new concepts. One of these could be the species value of a gene, another is regulatory genes i.e. + or – genes that regulate a behavioural trait. The latter poses a serious challenge to the Fisherian concept of additive genes and this concept has to be discarded. Molecular genetics is the key to the understanding of human and animal behaviour.