school children increasingly trace the identity of materials through their transformational history (e.g., sawdust comes from grinding up wood, so it must still be the same kind of stuff with some of its properties). This move can lead them to “hyperconservation of material kind”—a commitment to thinking that the identity of material is generally preserved which prevents them from being able to engage with the idea of chemical change. For example, they may see chemical changes as involving simply the mixture of substances whose identities are maintained during the process. Yet attending to transformation history can spawn productive insights in other contexts. For example, it allows them to think of materials as underlying constituents that maintain some core properties and to explain the properties of large-scale objects in terms of the materials of which they are composed. This move may be quite helpful to them in constructing an initial understanding of density as an intensive characteristic of materials.

Ultimately, however, in developing an understanding of atomic-molecular theory, students will need to reconsider the relation between properties that characterize entities at macro and micro levels and the ways assumptions about entities at the micro level can be used to explain observable phenomena. For example, although some macro-level properties are explained in decompositional terms (e.g., the weight and mass of an object is a function of the weight and mass of the atoms or molecules of which it is composed), other macro-level properties are emergent characteristics explained in terms of interactions among entities at the micro level. For example, objects are solid not because they have solid atoms, but because of bonding patterns among atoms and molecules. Thus, another major area of difficulty concerns linking up micro-level processes and entities with macro-level phenomena (Ben-Zvi, Silberstein, and Mamlok, 1989). Thus, elementary schoolchildren often have difficulty seeing how micro-level entities are related to macro-level ones, sometimes thinking that everything must appear the same at all levels of analysis (Nakhleh and Samarapungavan, 1999).

Unfortunately, an understanding of the distinction and linkages between macro and micro levels is often obscured by current teaching approaches that do not engage students with thinking through these issues and that have not systematically developed students’ epistemological understanding of the nature of models and theories. Students may be introduced to atoms and molecules through thought experiments about dividing materials into little pieces. This approach encourages students to think of atoms and molecules as just little pieces of materials that inherit all of their macroscopic properties. They then may not recognize that atoms/molecules are preexisting entities with distinct properties and characteristics (Pfundt, 1981). Students may be taught about the atomic-molecular theory as a “rhetoric of conclusions” or list of facts, rather than being engaged in model-based reasoning and exploring how to explain and make sense of a wide range of