Fractions are among the most difficult mathematical concepts for elementary school students to master (Behr, Harel, Post, & Lesh, 1992; Bezuk & Cramer, 1989; Moss & Case, 1999). Research indicates that manipulatives (e.g. fractions circles, fractions strips) positively impact students' conceptual and procedural understanding of fractions without impeding their ability to complete algorithmic procedures involving fractions (Cramer & Henry, 2002). Unfortunately, a variety of practical and pedagogical challenges associated with using manipulatives during instruction make it difficult for teachers to implement them effectively in classrooms. As a result, students receive far less exposure to manipulatives than the National Council of Teachers of Mathematics (NCTM) recommends for students in grades K-8 (Char, 1991; Hatfield, 1994; Hodge & Brumbaugh, 2003; National Council of Teachers of Mathematics, 2000). This study examines the following research questions: (1) Are there differences in students' knowledge of fraction magnitude when they are taught basic fraction concepts using virtual manipulatives compared to when they are taught basic fraction concepts using physical manipulatives?; and (2) Are students able to complete more practice exercises and/or more games using virtual manipulatives than physical manipulatives? The results of the post-assessment data collected in this study support Clements' (1999, 1996) hypothesis that computers can provide students with virtual representations of mathematical concepts that are just as meaningful as physical manipulatives. The results of the analyses of the data collected in the practice logs overall (i.e. across both weeks of the intervention) and during the second week of the intervention provide quantitative evidence that virtual manipulatives are more time-efficient than physical manipulatives. It appears that gender is not a strong predictor of the outcomes associated with manipulative-based instruction. (Contains 1 table.)