The pyrolysis of 1-ethyl-3,3-dimethylpyrrolidine-1-oxide gives 1-hydroxy-3,3-dimethylpyrrolidine, which, on dehydrogenation, 3,3-dimethyl-1-pyrroline-1-oxide. The identification of the nitrone is described. The cyclo-additions of styrene to this nitrone and t o 5,5-dimethyl-1-pyrroline-1-oxide were investigated, and the direction of addition was deduced from the n.m.r. spectra of the adducts.In connection with other work (1) it was desired to prepare pyrroline ring systems v ith geminal substitution a t a p-position, and to this end attenlpts were made to syntliesize one or more of the di~nethyl compounds I-IV.Three approaches were tried. The first followed the nitroalkane synthesis (2) and involved the attempted conjugate addition of nitromethane to senecialdehyde, this being the first step on the route to I1 and IV. Under a variety of reaction conditions, no absorption due to an unconjugated carbonyl group was found in the infrared spectrum of the product, and it was concluded that 1,4-addition had not occurred to an appreciable extent. This result is presumably due to the steric effect of the substituents a t the pposition of the unsaturated aldehyde, and it is pertinent that, whereas nitronlethane adds readily across the conjugated system of acrolein (3), in the reactioil of crotonaldehyde with nitroinethane, conjugate addition is not the main path.The second approach, which suffered from the disadvantage that it would be expected to yield a mixture of isomers 111 and IV, was the direct oxidation of the corresponding secondary ainine to the cyclic hydroxylamine (ref. 4, cf. 5 ) followed by dehydrogenation to the nitrone.As others have recently reported (6, 7), some difficulty was experienced with the direct oxidation with hydrogen peroxide, a step which gave erratic and low yields. The best yield 'For Par1 I V see R. Bonnell. J. Cltem. SOG. 2313 (1966.