The purpose of this study was to investigate whether physicality (actual and active touch of concrete material), as such, is a necessity for science experimentation learning at the kindergarten level. We compared the effects of student experimentation with Physical Manipulatives (PM) and Virtual Manipulatives (VM) on kindergarten students' understanding of concepts related to the use of a beam balance as a means to compare and differentiate materials according to their mass. A pre-post comparison study design was used that involved four conditions, with 20 kindergarten students in each condition. The first and second conditions included students who had correct prior knowledge of what a balance beam does and who were given PM and VM, respectively, to use for the study's purposes. The third and fourth conditions included students who had an incorrect prior knowledge of what a balance beam does and who were also given PM and VM, respectively, to use. All conditions followed the same learning tasks, which was a series of experiments. The learning process occurred in the context of a clinical interview, which was used as a means to collect data before, during, and after the learning process. Findings revealed that kindergarteners learned more from experimentation in both PM conditions and the VM condition whose participants had correct prior knowledge of what a balance beam does than in the VM condition whose participants had an incorrect knowledge of what a balance beam does. Physicality appears to be a prerequisite for students' understanding of concepts concerning the use of a beam balance, as a means to compare and differentiate materials according to their mass, only when the students have incorrect prior knowledge of what a beam balance does. (Contains 6 tables and 2 figures.)