Seasonal fluctuations in organic nitrogen and nitrate nitrogen in the soil of the Sudan Gezira are described under conditions of extended fallowing in cotton rotations. Statistically, results show that soil organic nitrogen is independent of rotational phase. Cotton yields are not limited by the content of organic nitrogen in the soil, but by soil physical conditions which improve by extended fallowing. Evidence is presented to show that soil organic nitrogen and nitrate nitrogen in this soil do not function in close interdependence. This conclusion was drawn in other work previously published.IN a previous paper (Jones, 1957) the author described the changes in organic-nitrogen content that occurred under three-course rotational cropping in Sudan Gezira soil. The field experiment used for the investigation was one of the long-term rotation experiments at the Gezira Research Farm, Wad Medani, Sudan; but the practical value of the experiment in terms of the general agriculture of the area was limited. Although the frequent cropping involved in a three-course rotation might economically offer a better gross agricultural output than a rotation containing a longer fallow eriod, it was known that under the intense of better weed control through cessation of watering. What was not known, however, was the optimum length of fallow.In order to investigate this point a 'combined-rotations' experiment (Crowther and Cochran, 1942) was laid down, and the present paper describes a study of the changes in soil organic nitrogen and nitrate nitrogen that occur under extended fallowing on this experiment in the Gezira. The experiment was planned to provide a com arison between ing cotton together with the variations from inclusion of intervening dura (Sorghum oulgare) and lubia (Dolichos Zablab). Ten rotations were involved with a total of thirty phases. The different treatments were completely randomized in each of five blocks making a total of 150 sub-plots. Each sub-plot was 4 feddan" in area.Data for the dura and lubia variations are not included in the present paper, which is restricted to an examination of soil samples taken from all hases of the following rotations: A, continuous cotton; M, cotton an one ear fallow; G, cotton and two years fallow; K, cotton and three years fa low. The five rotation replicates per phase were sampled for * I feddan = 1.038 acres. heat of the Central Sudan P allowing was highly beneficial and permitted one-, two-, three-, and four-course rotations in terms o F fallows follow-