The e.s.r. spectrum of the simple SiH3radical in a low temperature matrix was first reported by Gordy and coworkers1. They concluded that the large29Si splitting of 266 gauss is due to the strong deviation from planarity of the radical. This deviation from the simple methyl radical, which is known to be essentially planar, can be qualitatively explained by the electronegativity difference between silicon and carbon as suggested by Pauling2. Several other silyl radicals have since been reported3, all have3splittings relatively larger than the corresponding13C analogs. Recently several workers questioned the interpretation of the29Si splittings on the geometry of simple silyl radicals4,5. Recent evidence appears to show that the silyl radical is not as strongly pyramidal as previously supposed. Biddles and Hudson4pointed out that the29Si splittings are negative due to a negative magnetic moment, but the proton splittings can be either negative for a quasi-planar structure or positive for a pyramidal geometry. Satisfactory interpretation of the29Si splitting can be obtained from the INDO calculation, provided the proton splittings are negative. The large29Si splittingin SiH3calls for treatment of second and higher-order effects of the e.s.r. parameters6. A full calculation of SiH3is reported below.