In this paper I take a look at what I take to be the best argument for dispositions. According to this argument we need dispositions in order to understand certain features of scientific practice. I point out that these dispositions have to be continuously manifestable. Furthermore I will argue that dispositions are not the causes of their manifestations. However, dispositions and causation are closely connected. What it is to be a cause can best be understood in terms of counterfactuals that are based on dispositions.