Five macroscopic boundary parameters can be extracted from three-dimensional orientation maps. Serial sectioning, which includes consecutive steps of material removal, and electron backscatter diffraction (EBSD) measurement were employed to extract a stack of two-dimensional sections of a pure nickel sample. The EBSD patterns were collected from large millimetre scale areas and mechanical polishing was applied to prepare the sections. The three-dimensional microstructure was then reconstructed from these sections. A new alignment algorithm based on the minimization of misorientation between two adjacent sections has been developed to accurately align the sections. Differently from the conventional alignment methods, the new algorithm corrects not only the translational misalignment but also rotational and plane parallelity misalignments. The aligned three-dimensional microstructure exhibits smooth grain boundary planes and continuous orientation gradients inside the grains as experimental scatter induced by misalignment was largely removed. Grain boundaries were reconstructed from the aligned three-dimensional map, and the distribution of boundaries in the domain of five macroscopic boundary parameters was computed using kernel density estimation. Methods for estimating the reliability of the distributions are demonstrated. This distribution is compared with the distributions obtained previously for other face-centered cubic materials, including a different pure nickel sample.