Cartesian groups not belonging to topological projective planes (Q804950)

In Abh. Math. Semin. Univ. Hamb. 60, 257-264 (1990; Zbl 0721.51016) the author constructed topological affine planes with non-continuous parallelism; consequently these planes have no topological projective extensions. The author provides examples of Cartesian groups such that the parallelism of the corresponding topological affine plane is continuous but does not yield a topological projective plane. This contrasts the situation for locally compact ternary field with certain coherence conditions which ensure a topological projective extension, e.g. \textit{Th. Grundhöfer} [Abh. Math. Semin. Univ. Hamb. 57, 87-101 (1986; Zbl 0652.51017)]. Explicitly, the construction uses an ordered skewfield \((K,+,\cdot;\leq)\) and two distinct strictly increasing families \((c_ i)_{i\in Z}\) and \((d_ i)_{i\in {\mathbb{Z}}}\) from \({\mathbb{Z}}\) into K whose supremum is \(+\infty\) and whose infimum is -\(\infty\) and which are normalized such that \(c_ 0=d_ 0=0\), \(c_ 1=d_ 1=1\). Let \(\delta\) be the strictly increasing bijection of K obtained by linearly extending \(\delta (c_ i)=d_ i\). Then a new multiplication * on K is defined by \[ x*y=x\cdot y\quad if\quad x\geq 0,\quad x*y=x\cdot \delta (y),\quad if\quad x<0 \] and \(C=(K,+,*)\) is a Cartesian group that coordinatizes a generalized Moulton plane. Evidently, the subgeometry induced on the parallel strip \(]c_{i-1},c_{i+1}[\times K\) is isomorphic to a subgeometry of a usual Moulton plane over K (i.e. \(\delta (x)=x\) for \(x\geq 0\) and \(\delta (x)=kx\) for \(x<0\) for some \(k>0)\); thus the plane coordinatized by C can be viewed as the collection of these Moulton subgeometries glued together thus obtaining a plane with infinitely many vertical lines at which non- vertical lines are bent. It is shown that C is a topological Cartesian group with respect to a Hausdorff topology \(\tau\) of K if and only if \(\tau\) is finer than the order topology of K and K is a topological skew field with respect to \(\tau\). This corresponds to results of \textit{P. Hartmann} [Geom. Dedicata 31, No.3, 321-332 (1989; Zbl 0687.51003)] who determined all topologies which make a projective Moulton plane constructed from an ordered skewfield a topological projective plane.