On square roots of a rotation (Q1184885)

The article starts off with an erroneous statement: the existence of square or \(m^{th}\) roots of matrices is seemingly completely understood in theory [see \textit{R. A. Horn} and \textit{C. R. Johnson}, Topics in matrix analysis (Cambridge 1991; Zbl 0729.15001), Chapter 6.4 or \textit{J.-C. Evard} and the reviewer, Linear Algebra Appl. 162-164, 447-519 (1992)] with references from 1857 on. This error does not affect the paper, however, since its aim is to find a real square root of a rotation which does not have the eigenvalue -1 and is a rotation itself. The method proposed is Jacobi-like and iterative, low dimensional computed examples are included, but the stability of the algorithm itself is not tested.