Abstract
Central to many issues surrounding reduction in science is the relation between a physical system and its components. In this article we examine how thermodynamic theory relates properties of whole systems to properties of their components. In order to keep the analysis general, we focus our study on universal properties like volume, heat capacity, energy and temperature. In the cases examined we find that scientific explanation requires appeal to properties of components that are spatially as extensive as the whole system. We discuss some implications of our study for the purported paradigmatic reductions of heat and temperature to molecular motion. We conclude that while macro systems reduce ontologically to micro components, epistemologically the reduction of theoretical concepts in general fails.