The chromosome theory of inheritance was established during the three first decades of the 20th century. During the early stage of its substantiating, there were lots of puzzles and little evidence that could validate it. The cytological processes were obscure and several scientists maintained serious doubts concerning the existence of a connection between Mendel’s principles and the behaviour of chromosomes during cell division. It was vital to associate an external, observable characteristic of the organism to a specific chromosome, and this (...) was achieved when sex was connected to special chromosomes. At that time, however, some important scholars refused to accept or delayed acceptance of the conception that the hereditary factors (later called genes) were physical entities located along the chromosomes. Such was the case of Thomas Morgan (1866-1945) and William Bateson (1861-1926). Their attitudes could be explained by considering the doubtful ground of the hypothesis at that time. It is more difficult, however, to understand the attitude of Edmund Beecher Wilson (1856-1939). Being an expert in cytology he was acquainted with all the difficulties concerning the chromosome hypothesis. Nonetheless, from 1905 onward, he attributed little weight to the problems and dedicated a notable effort to obtaining evidence that could have grounded it. This paper analyses Wilson’s attitude focusing on his studies from 1900 to 1915 and the scientific context of this period. This study led to the conclusion that Wilson’s attitude could be explained in methodological terms by the adoption of an instrumentalist attitude, while Bateson and Morgan adopted a realistic perspective. (shrink)