Nucleophilic displacements which occur with retention of configuration have beell designated SNi reactions (1). The most exteilsively studied reaction of this type is that between thionyl chloride and secoildary alcohols. Recent worlc in this field has led to the coilclusion that SNi reactioils possess considerable ionic characteristics, with carboilium ions or ion-pairs suggested as intermediates (2-5). Rearrailgeineilts typical of intermediate carboilium ions have been observed by Cram (2) for the reactioil of thioilyl chloride with the stereoisomers of 3-phenyl-2-butailol, 2-phenyl-3-pentanol, and 3-phenyl-2-pentanol. The present coininunicatioil reports the forination of a rearranged product from the reaction between thionyl chloride and 3-methyl-2-butanol.3-Methyl-2-butanol was added dropwise, with stirring, to a 100% excess of purified thionyl chloride a t -35' C. Stirring was continued for 2 hours a t -35' C. The hydrogen chloride produced was pumped off a t about 25 111111 pressure. The reaction iuixture was then allowed to stand a t room teiuperature for 1.5 hours before the excess thioilyl chloride was distilled a t 29-31' C a t 25-30 mrn. The resulting chlorosulphite was decomposed by heating at 40-50' C for 1.5 hours. Evolution of sulphur dioxide was noted during the decomposition. On raising the temperature, the crude product distilled. The distillate was washed with water and with dilute sodium carbonate solution until the washing reinailled basic. The washed product was dissolved in ether and dried over phosphorus pentoxide. After reinoval of the solvent, the residue was distilled to give about 20y0 yields of alkyl chloride, b.p. 84-85" C (lit. (Ga) for f-amyl chloride, b.p. 86' C), which gave an iinmediate precipitate with alcoholic silver nitrate a t room temperature. The infrared absorption spectruill of this product showed all the absorption peaks as l;no~~rn t-arnyl chloride together with an extra, weal; and broad band in the 1726 to 1750 cm-I region. These results appear to indicate that the reactioil product may be t-amyl chloride contaminatcd with a small amount of un1;nown inzpurity.T o coilfirin the identity of the reaction product as chiefly t-amyl chloride, it was converted to the Grigilard reagent followed by reaction with carboil dioxide to give dimethylethylacetic acid (7,8) with an over-all yield of 35%. The low yield was to be expected since it is difficult to prepare Grignard reagents froill t-alkyl halides. The dimethylethylacetic acid ~v a s identified by conversion to the p-phen~~lphenacyl ester, ~vllich iuelted alone and on adinixture with an authentic sample a t 84' C (lit. (Gb) 1n.p. 86.5' C). Considering these results, it is reasonable to conclude that t-amyl chloride was obtained from the reaction of thionyl chloride with 3-i11ethyl-2-butanol. A 1,2-hydride shift inust have occurred in the reaction. A siinilar type of 1,2-shift has been noted in the reaction of thioilyl chloride with 2-butanol-1-C14, resultiilg in the forinatioil of some 2-butyl-4-C14 chloride. T h e rear...