This report describes the results of the international line scale comparison Nano3, which was carried out between 2000 and 2003 and which was accepted as supplementary comparison CCL-S3. This comparison was initiated by the BIPM working group on nanometrology as one of five international comparisons in the field of dimensional nanometrology. Two high quality line scales, one made of Zerodur and one made of fused silica (quartz), with 280 mm main graduation length and additional smaller graduations of only a few mm were chosen as transfer standards. These scales were produced using advanced and optimized lithography and processing technologies by the Dr Johannes Heidenhain GmbH, Germany. A considerable number of characterizations of the graduations were performed in order to ensure an optimized line edge quality of the scales used in the comparison. Moreover, it was decided to have long gauge blocks manufactured out of the same piece of substrate material as was used for the scales. In this way, it was possible to independently determine important substrate material parameters like thermal expansion, compressibility and to investigate the long-term stability of the substrate materials.The Zerodur line scale standard revealed a small length reduction of about (−7 ± 4)×10-8/a, which was confirmed by the measurements on the long gauge blocks. This length change of the Zerodur line scale could be taken into account for the comparison of participant's data by the application of a linear drift model. On the quartz samples and linescales no comparable effects were observed.The line scales were measured by 13 national metrology institutes from four different metrology regions. Two institutes decided to withdraw from Nano3 after the measurements were performed, but before Draft A was circulated. The measurement uncertainties that were evaluated by the participants over the 280 mm length of the graduations showed a variation from about 300 nm down to 30 nm.The good line edge quality of both scales allowed a meaningful separation of the length-dependent and length-independent deviations from the weighted mean values. Therefore a meaningful comparison of these deviations with the evaluated uncertainty contributions of the participating institutes was possible.For the most important measurand of this comparison, namely the position deviations of the line structures on the 280 mm main graduation, three results out of the 11 data sets provided had to be excluded by application of the En-criterion for the quartz and Zerodur scales respectively. Investigations of the reasons for the deviations have already been started by the respective institutes, including bilateral follow-up line scale comparisons with the Nano3 pilot laboratory.Main text.To reach the main text of this paper, click on Final Report.Note that this text is that which appears in Appendix B of the BIPM key comparison database kcdb.bipm.org/.The final report has been peer-reviewed and approved for publication by the CCL, according to theprovisions of the Mutual Recognition Arrangement (MRA).