Hexacoordinate silicon has been claimed to occur in bis(8-dimethylamino-naphth-1-yl)silanes on the basis of X-ray data and NMR spectra. Very much the same effects are shown by a bis(8-dimethylamino-naphth-1-yl)carbinol, pointing to essentially the same bonding in the Si and C compounds. Faced with the choice that either the carbon in the carbinol is hexacoordinate, too, or that the silicon in the silanes is, in fact, tetracovalent, we prefer the latter conclusion. Chirality 10: 180-189, 1998. © 1998 KEY WORDS: hypercoordinate silicon in naphthylsilanes; X-ray structures; NMR spectra; reinterpretationIn an ever-increasing number of papers, it is claimed that hypercoordination at the ''higher'' elements of the periodic table is a much more widespread phenomenon than previously believed.1 Some details, however, are puzzling. For example, in (8-dimethylamino-naphth-1-yl)-(''DAN'' 2 ) silanes, the step from conventional tetracovalency to pentacoordination and subsequent steps from there to hexa-, hepta-, and octa-coordination 3 are claimed to occur all with the same ease though in the realm of preparative organophosphorus chemistry, four P-C bonds at one P atom are easily formed, but the fifth and sixth only with increasing difficulty 4 and a seventh not at all. 5 In organosilicon as well as organophosphorus chemistry, penta-and hexacovalency are accomplished most readily when electronegative substituents cause a low electron density at Si/P whereas the recent claims of hypervalency often concern compounds where such substituents are lacking.6 Germanium and tin are believed to be as prone to hypercoordination, 7 though their orbitals, due to the larger size of these atoms, are more diffuse and hence geometrically less favoured for significant overlap with the filled orbital at nitrogen of a dimethylamino group. Within the frame of molecular orbital theory, it is hard to see where such atoms take all the orbitals from to achieve hexa-, hepta-, and octacoordination.8 MO theory, though often taken as a dogma, is no more than a theory used to describe facts and not a fact in its own merit, 9 and it is, therefore, legal to disregard it, but then, tetracovalency of carbon can no more be accepted as a dogma either, and it is, then, permitted to apply the same methods in search of hypervalent carbon.10 If, on the other hand, the dogma of tetracovalent carbon is retained, compounds of the same type, but with carbon instead of P, Si, Ge, Sn, etc., would provide a conclusive counter check whether the arguments brought forward in favour of hypervalency at these elements are indeed an iron-cast proof or not.In the first part of a wider program devoted to these problems, we deal with the DAN-silanes 1a-e (see Scheme 2) for which Corriu et al.11,12 believed to have found evidence for hexacoordination at Si. 2, which is a close analogon of 1b and, particularly, of 1c, was readily prepared by peri-lithiation of di(methyl)(naphth-1-yl)amine 13 and subsequent reaction with methyl formiate (yield 43%). We obtained an X-ray stru...