Counterexamples to faudree and schelp's conjecture on hamiltonian-connected graphs

Faudree and Schelp conjectured that for any two vertices x, y in a Hamiltonian-connected graph G and for any integer k, where n/2 ? k ? n ? 1, G has a path of length k connecting x and y. However, we show in this paper that there are infinitely many exceptions to this conjecture and we comment on some problems on path length distribution raised by Faudree and Schelp.