Purpose: To compare the dimensional measurements and surface topography of stereolithographic models generated from computerized tomography (CT), magnetic resonance imaging (MRI), and laser scanning (LS) data with the same subjects' natural ears and ear casts. Materials and Methods: Stereolithographic models were manufactured from images of the subjects' ears and ear casts recorded by CT, MRI, and LS, and dimensional measurements were compared. In the second part of the study, all stereolithographic models were CT scanned and reconstructed in an STL file format. A comparison of the surface topography of the CT, MRI, and LS model ears was made. Results: Intraclass correlation coefficients indicated that dimensions could be reliably measured on the CT, MRI, and LS stereolithographic models. A 2-way analysis of variance revealed no statistical differences between the various sources of data (P = .991). The smallest differences of surface topography were observed on the MRI/CT superimpositions. Conclusion: The dimensional measurements on the stereolithographic models were similar to those from the original source. Only small differences were apparent between the surface topography of the CT, MRI, and LS models. MRI may be particularly appropriate to fabricate a prosthesis because it involves no radiation for the patient and internal form can be reproduced. The use of this technique in clinical practice requires further study.