Level density dependence on parity for sup(238)U

It has been shown that the distribution of gGITAsub(n)sup(1)p resonances of uranium-238 can be in a good agreement with the Porter-Thomas theoretical distribution only if it is assumed that a considerable omission of p levels is observed in experimental data. The level density of uranium-238 with a spin of 1/2 appears to depend on parity. Also the level density of p resonances exceeds approximately 4.7 times (and not three times) the number of s resonances in the same energy range. It has been possible to put into agreement experimental and theoretical distributions of gGITAsub(n)sup(1) by assuming the spin dependence of a strength function