At the beginning of the 1990s, a simple genetic model that explained flower development was presented based on Arabidopsis thaliana and Antirrhinum majus floral homeotic mutants. According to this model, which is a milestone in plant development studies, flower development can be explained by three classes of genes (A, B and C), each one controlling the identity of organs in two adjacent whorls. Intriguingly, more than 20 years later, there are still some unanswered questions, in particular regarding the universality of the class A-function genes. Class A genes are well characterised in A. thaliana, but so far no A mutants have been described in other plant species nor in Antirrhinum majus. Here, we retrace the story that led to the proposal of the ABC model focusing on the contribution of A. majus to this model. Although fewer groups are still using A. majus as a model system, this plant was a master contributor to our comprehension of the molecular networks controlling flower development.