The distinction between primary and secondary qualities was not only motivated by the epistemological consequences of the corpuscular hypothesis. The way Boyle introduced the distinction for the first time in its modern sense had been largely determined by the details of his corpuscular theory of matter. This paper examines Boyle’s justification for his inferences from observables to unobservables, and the way methodological backgrounds of his arguments shaped the character of the distinction. [Projekat Ministarstva nauke Republike Srbije, br. ON179041: Dinamički sistemi u prirodi i društvu]