There have been various attempts to argue from the success of certain aspects of scientific practice to the existence of necessary connections between distinct events. Many of these arguments are based on an inference to the best explanation (IBE). I will start by reviewing some of these IBE arguments as well as their recent critique by Helen Beebee. Beebee’s critique is convincing with regard to the particular arguments she criticises. I will, however, present other aspects of scientific practice that need to be explained. More particularly I will argue that the failure of certain manipulatory practices can best be explained in terms of necessary connections. Beebee’s critique does not apply to this IBE-argumen