A remark on elimination of singularities for mappings of 4-manifolds into 3-manifolds. (Q1398190)

Suppose one wants to find a map between given manifolds with only `mild' singularities. There are natural obstructions to the existence of such a map, namely the so-called Thom polynomials of the not-allowed singularities. These are cohomology classes Poincaré dual to singularity submanifolds. In some cases the vanishing of these natural homological obstructions are sufficient for the existence of a map with mild singularities. However, \textit{O. Saeki} found examples where the vanishing of Thom polynomials is not enough, e.g., he proved -- using Rokhlin's theorem about the signature of 4-manifolds -- that the complex projective plane does not admit a fold embedding into \(\mathbb{R}^3\). The paper under review gives an elementary proof of this theorem.