Neighbor-preserving homomorphisms between projective ring planes (Q1060415)

Let R,R' be associative rings of stable rank 2, \(P_ 2(R)\), \(P_ 2(R')\) projective ring planes as defined by the second author [Geom. Dedicata 11, 285-308 (1981; Zbl 0468.51001)] and \(\psi:P_ 2(R)\to P_ 2(R')\) a full neighbour preserving homomorphism (i.e. for any two points x,y of \(P_ 2(R)\) there is a line \(\ell\) in \(P_ 2(R)\) such that \(\psi(\ell)\) is not neighbouring \(\psi(x)\) and \(\psi(y)\)). Then it is possible to coordinatize the planes in such a way that for \(0=(1,0,0)R\), \(p_ 1=(0,1,0)R\), \(p_ 2=(0,0,1)R\), \(e=(1,1,1)R\) one has \(\psi (0)=0\), \(\psi (p_ i)=p_ i\), \(\psi (e)=e.\) The main theorem states that then there exist a subring S of R and a ring homomorphism \(\psi\) :S\(\to R'\) such that (i) ker \(\phi\) \(\subseteq rad S\) (Jacobson radical of S), (ii) \(\phi (S)^*=R'{}^*\cap \phi (S)\) \((R^*=\) the set of the units of R), (iii) every point (resp. line) in \(P_ 2(R)\) can be represented as \((\alpha_ 0,\alpha_ 1,\alpha_ 2)R\) (resp. \(R(\alpha_ 0,\alpha_ 1,\alpha_ 2))\) with \(\alpha_ i\in S\), and (iv) \(\psi ((\alpha_ 0,\alpha_ 1,\alpha_ 2)R)=(\phi (\alpha_ 0),\phi (\alpha_ 1),\phi (\alpha_ 2))R\), \(\psi (R(\alpha_ 0,\alpha_ 1,\alpha_ 2))=R(\phi (\alpha_ 0),\phi (\alpha_ 1),\phi (\alpha_ 2))\).