Non-Euclidean geometry and volumes of hyperbolic polyhedra (Q1360240)

Two polygons \(P\), \(Q\) in the Euclidean, hyperbolic or elliptic plane are ``equivalent by dissection'', if there exists a dissection into polygons \(P = P_1 \cup \dots \cup P_n\), \(Q = Q_1 \cup \dots\cup Q_n\) and isometries \(\Phi_i\) (obviously in the sense of the peculiar geometry) with \(\Phi_i(P_i) = Q_i\). An elementary proof of the theorem, that polygons of equal area are also equivalent by dissection (W. Bolyai, 1832) is given by the author. The analogous theorem formulated in Euclidean 3-space is not true: Polyhedra of equal volume are not necessarily equal by dissection. This theorem, well-known as ``Hilbert's third problem'', was proved by M. Dehn in 1900 with the help of the so-called ``Dehn-invariant'' \(\psi(P)\), which is invariant under all isometries of \(P\). J.-P. Sydler (1965) gave the following characterisation: Polyhedra \(P\), \(Q\) in Euclidean 3-space are equivalent by dissection iff \(\text{vol}(P) = \text{vol}(Q)\) and \(\psi(P) = \psi(Q)\). For a polyhedron in the hyperbolic or elliptic 3-space the Dehn-invariant can de defined analogous to the Euclidean case. It is not yet certain, if Sydler's theorem is valid in these two geometries. In the last section of this paper the author concentrates on the 3-dimensional hyperbolic case. The volume of an orthoscheme (special tetrahedron, 3-dimensional generalisation of an right-angled triangle) can be computed by a formula given by N. I. Lobaschevskij. The hypothesis that any orthoscheme is equal by dissection to certain asymptotic orthoschemes , i.e. orthoschemes with two vertices at infinity, remains unproven. This is one of the main obstacles on the way to the solution of Hilbert's third problem in hyperbolic 3-space.