The results of two sets of soil nitrate measurements on the three-course rotation plots at the Gezira Research Farm are described. The first set is a long-continued series taken over 21 years; it shows for the first foot samples that there is a distinct tendency for nitrates to be depressed during the rainy season, this tendency being more marked in the more open rotations. This is ascribed to washing down of nitrates. The second set of measurements shows accumulations in the second foot and a distribution consistent with downward movement.

In the Sudan Gezira the yields of irrigated cotton are positively correlated with the amount of rainfall in the 6 weeks before sowing. It follows from the work described here that this is not due to increased nitrate in the top foot of soil at sowing time.

The sampling of this experiment for nitrate was started by my predecessor, H. Greene, and continued by O. W. Snow. The field experiment itself has been conducted by the Plant Physiology Section of the Research Division of the Ministry of Agriculture.

The difference between the power to accumulate nitrates in the soil after various rotations has been determined by incubating the soils at the termination of these rotations, and measuring the nitrate formed between January and May, a period of 4 months. The test has been repeated during 8 successive years, and the mean result is presented. The previous treatments, each of which lasted 3 years, consisted of (a) a ley grazed off each season by sheep, (b) a crop of lucerne which furnished two or three crops of hay each yer, and (c) and (d) an arable rotation of potatoes, wheat, and either a hoed crop of kale or a mixed ryegrass and clover crop of hay.

In each case, the ammonia capable of distillation with magnesia was reduced during the incubation to an amount which differed little in all the soils, while the nitrate made an almost steady increase—most regular, however, after the previous cropping with lucerne—at least for a period of 4 months. The soil after the previous grass and clover hay crop started at a lower level of nitrate than the others and, even after the longest incubation period, never quite reached the accumulation which occurred in the other cases. Comparing the accumulation which occurred after the two 3-year ley crops (grazed ley and hayed lucerne) with that after the arable cropping, there was very little difference.

Precise confidence limits of expected monthly rainfall must be regarded as fundamental to the specification of the climatic regions of an agricultural region. The science of weather forecasting is necessarily based on extensive statistics of past experience. However, unless account is taken of such factors as skewness, the means and standard deviations calculated from such statistics will lack the necessary precision and may often be misleading.

This paper shows that frequency distributions of monthly rainfall, which were demonstrably skew, may be suitably transformed so that on the new scale they approximate closely to theoretical normal distributions. As a result precise fiducial or confidence limits—shown in fact to fit the data satisfactorily—may be estimated and afterwards reconverted to the original units.

Practical applications of these confidence limits include the more accurate representation of rainfall patterns, valuable guides to the vital question of optimum time of planting, and finally, better estimates of the range of mean monthly rainfall where data are available only from a limited number of years.

Application of these limits to some special cases is discussed in some detail.

Six pairs of 2-year-old monozygous twin heifers were used to determine the effect of underfeeding for 10 weeks prior to calving. Those which had been well fed before calving produced more milk than their sisters which had been poorly fed.

There was no effect on milk or butterfat composition, probably because none of the animals lost much weight after calving.

The effect of the treatment differed markedly between twin pairs, the maximum difference in milk yield being 2866 lb., the minimum 135 lb., and the mean 1057 lb. for 270 days' lactation.

The ratios of between-twin-pair to within-twin-pair variance for butterfat, total protein and lactose percentages of the milk were as follows:

The rates of nitrification and the extent to which a number of nitrogenous materials are nitrified in soil have been studied. Of the substances used ammonium sulphate was most rapidly and most completely nitrified under these conditions. A reputedly genuine sample of Peruvian guano was almost as rapid as ammonium sulphate. At the other end of the scale was horn meal, which was nitrified most slowly and in which the percentage of nitrogen converted was lowest. On a basis of nitrification alone there appears to be no justification for the traditional view that hoof and horn is slow- and that dried blood is quick-acting. Two samples of bone meal showed very different rates of decomposition. This, and the anomalous behaviour of hoof and horn separately as compared with mixed hoof and horn, suggest that differences occur in the behaviour of different samples of what are nominally the same materials. Some of the commercial implications of these findings are discussed.

1. Laboratory experiments in five soils to which hoof and horn materials were added at different concentrations have shown that the percentages nitrified throughout the transformation were largely independent of the level of application. The ranges of concentration used were 90–360 p.p.m., 150–350 p.p.m. and 100–300 p.p.m. nitrogen.

Similar results were obtained with dried blood in two soils.

2. Experiments in two soils with ammonium sulphate and in one soil with glycine, urea and guano showed a different response to level of application, the percentages nitrified at the different concentrations being widely different throughout the period of active transformation.

3. An explanation of the results is given on the basis of the relative rates of formation of ammonia and of its transformation to nitrate.

4. The importance of ammonia determinations in the interpretation of nitrification studies is emphasized.