The electronic structure of an unusual LiSn phase (computed using band structure calculations in the framework of the extended Huckel tight binding theory) is the starting point for a general analysis of the variation of electron-rich multicenter bonding across a period. The LiSn crystal structure of Muller and Schafer in question contains 2D slabs of Sn atoms arranged as microscopic stairs and intercalated by Li atoms. Discrepancies between an electron count derived from a recent extension of the Zintl-Klemm rules to electron-rich systems (5(2)/(3) electrons) and the experimental one (5 electrons for the Sn sublattice) and other failures of chemical "common sense" emerge in the analysis. The key for interpretation of a series of puzzling results was found in the comparative analysis of the Sn net with other main group element hypervalent slabs. Increasing s,p-mixing as one moves from the right to the left side of the same row of the periodic table is responsible for these effects. The result is that a lower electron count is found in the Sn slabs relative to the one expected from the extended Zintl-Klemm theory. The effect should also occur in discrete molecules. We also showed that the Li atoms have a role in the determination of the final structure, not only because of their small size but also through the degree of the electron transfer to the Sn sublattice.