A split-plot field experiment was conducted to investigate the influence of planting dates on the formation of nodules in a variety of cowpea (Paraguay 6). Results indicated that the principal factor influencing both nodulation and yield was day-length, the highest number of effective nodules being produced from the June planting.

Where ancient agricultural sites are brought back into cultivation, using new forms of irrigation, cotton often grows badly in the first few seasons. A theory is developed to account for these failures, and it is shown that heavy irrigations before planting not only produce better crops but may also economise in water.

Repeatability estimates of milk yield and length of lactation of local, ½ Anglo-Nubian × local and ¾ Anglo-Nubian × local cross-bred goats were calculated. There were highly significant differences in the effect of dams on milk yields for the third, fourth and fifth lactations for local goats, when considered together. No significant differences were found for the first two lactations. For the ½ Anglo-Nubian cross-bred goats significant differences were also found for the first and second lactations, but not for the third lactation. No significant differences were found for the two lactations in the case of the ¾ Anglo-Nubian cross-bred goats. The effect of lactation periods were not significant for the three types of goats. The mean repeatability estimates for local goats were 0·47, 0·42 for ½ Anglo-Nubian and 0·54 for ¾ Anglo-Nubian cross-bred goats, which were comparable to those reported for cattle. The between-dam effect was highly significant when three lengths of lactations were combined in local goats, and the interval between lactations were also significant, but these two effects were not significant for the ½ Anglo-Nubian and ¾ Anglo-Nubian cross-bred goats. The estimates of repeatability for length of lactation averaged 0·29 for local goats, 0·17 for ½ Anglo-Nubian and 0·53 for ¾ Anglo-Nubian cross-bred goats. The correlation coefficients between milk yield and length of lactation were positive and statistically significant for all the three types of goats, ranging from 0·59 to 0·93. The similarity of the genetics of goats to those of cattle and sheep is mentioned.

A breeding programme involving the progeny testing of small East African Zebu cattle for improved growth rates is described and an assessment made of environmental and genetic improvements which have occurred over a ten-year period. Evidence is presented to justify continuing the programme, but suggestions are made to modify it in order to speed up the testing period and also to make more use of genetically superior stock.

The productivity of a herd of Mashona cattle is described, where hand-feeding was practised in winter to prevent loss of body weight. Calves weighed about 50 lb at birth, and gained weight at the rate of 1.27 lb a day to six months of age, compared widi 1.56 and 0.87 lb a day for calves at two other sites. Young animals gained an average of ½ lb a day from weaning to maturity. Mashona steers responded to pen-feeding and produced carcasses which were satisfactory for the Smithfield market. The mature weight of breeding cows was just under 700 lb and the calving rate varied from 72 to 92 per cent. Cows which failed to calve and those whose calves were early-weaned gained almost 100 lb liveweight by the end of die grazing season. Early-weaned calves fed a complete diet were 40 lb heavier than their contemporaries at 28 weeks of age. The need to develop die breed is discussed.

The productivity of climax and serai grassland in South Africa was compared at three stocking rates with steers. On climax grassland, over nine years, gains per head at one steer per morgen (= 2·12 acres = 0·86 hectare) were not significantly greater than at two steers per morgen. Liveweight gains per morgen were 70 per cent higher at the heavier stocking rate and were 162 lb and 273 lb at one and two steers respectively. On serai grassland, over three seasons stocked at two steers and three steers per morgen, the liveweight gains per steer were significantly less at the higher stocking rate but there was no significant difference in the gains per morgen, which were 366 lb and 318 lb at two and three steers respectively. The liveweight gains on the seral grassland were significantly higher than on the climax grassland at the same stocking rate Calculation of the total digestible nutrients (T.D.N.) has shown that with heavier stocking more T.D.N. per morgen may be harvested on both grassland types. However, with heavy grazing the efficiency of conversion of T.D.N. into liveweight was less. The quality of the sward, measured in per cent crude protein, P2O5 and crude fibre was better at the heavier stocking owing to delayed maturity of the grass when it is continually grazed short.

In connection with the processing of potatoes grown at Tarland, Aberdeenshire, certain field data were obtained over a period of eight years. It was found that the yield per acre was inversely correlated with rainfall and positively correlated with sunshine for the period May to September; the indications were that the period May to July was more important in this respect, high rainfall and low sunshine during this period depressing the crop more than high rainfall in the August to September period. It has been possible, statistically, to separate the effect of rainfall from that of sunshine (and temperature), and an analysis shows that sunshine (or temperature) is die dominant factor. The number of tubers produced was greatest in those years when sunshine was high and rainfall low. The dry matter content of die tubers was also highest when the sunshine was high and the rainfall low, the closest inverse correlation being with the rainfall for the month of August. The combination of high yield of potatoes and high dry matter content meant that the actual food produced was highest in the drier, more sunny, years.

The main nutrient deficiency in Malawi is a chronic lack of nitrogen, which is found on all but a few soils. Phosphorus also becomes limiting quite rapidly on certain soil types as a result of permanent cultivation, but there has so far been little response to potash or lime. Incipient sulphur shortage is automatically corrected by the use of sulphate of ammonia and single superphosphate to correct the major deficiencies. Trace elements have not been widely tested. Samples of organic manure vary widely in moisture and nutrient content, and their nitrogen status is frequently low, so it is safest to consider manure as primarily a source of P, K and minor nutrients, and to supplement it with fertilizer nitrogen. It should be possible, with good cultivation of synthetic varieties and improved selections, to obtain maize yields of 15–20 bags/acre with fertilizer dressings of up to 80 lb/acre of the limiting nutrients.

Response to potassium fertilization was tested, over a two-year period, in 24 experimental fields, predominantly in coarse-textured soils. Only two fields showed significant yield increases. Potassium availability was estimated on the basis of (a) the free energy (ΔF) of potassium exchange for calcium and magnesium, (b) the proportion of potassium in total exchangeable cations, and (c) soluble potassium extraction in correlation with the potassium content of the plants. Woodruff's method (ΔF) yielded the best agreement (correlation coefficient r = 0·77). ΔF values showed a sufficient supply of available potassium, which may have accounted for the lack of response to fertilization.

Plant population studies were carried out in 1963 and 1964, including row spacing of 75 cm at different within-row densities, for purposes of comparison with the common practice of drilling at 15 cm spacing. Planting in spaced rows 75 cm apart, with increasing intra-row density from 55 to 175 seeds per metre of row, increased seed yield compared with drilling and a high intra-row density was better than low intra-row density for seed yield. The increase in seed yield was due to a higher number of stems, flowers and pods per unit area, and to a higher number of flowers forming mature pods. Seed set was better in spaced rows than at drill-spacing. Flowering took place over a shorter period, and the peak was attained earlier, at higher within-row density. Plant competition influenced plant branching as early as sixteen days after emergence. The relation between seedling development and seed yield is discussed.

Four depth of tillage treatments were given by ploughing to 7 cm. depth by bullock-drawn plough and 14, 21 and 28 cm. depth with a tractor-drawn mouldboard plough in 1962–63. A significant decrease in bulk density was observed with the 21 and 28 cm. ploughing, negatively correlated with dry matter of crop roots and yield of grain. The effectiveness of 28 cm. tillage in reducing the population and dry matter of weeds was significant when compared to tillage at 7 and 14 cm. and there were significant relations between dry matter of weeds and yield of grain. Crop roots under tractor tillage were significantly greater than with 7 cm. tillage. Grain yield was maximum under 28 cm. tillage followed by 21, 14 and 7 cm. deep tillage in that order. The higher grain yields were associated with a greater number of fertile tillers per metre of row and higher yields per tiller.

Methods of recording experimental data in the laboratory and field are discussed. Use of specially designed record sheets, punch cards and punched paper tape in conjunction with an electronic computer extends the scope of experimental work and leads to improved efficiency and accuracy.

The use of cultivation techniques to conserve rainfall, by preventing runoff, and to improve the soil surface infiltration rate, has produced spectacular increases in the yield of cotton on the Loess Plain soils of Northern Nigeria, where yields of the order of 2000 lb seed cotton per acre can now be produced regularly irrespective of the rainfall pattern. As yields of this level had not been obtained previously, it has been concluded that lack of soil moisture and poor soil aeration have in the past set a ceiling to seed cotton yields. The effects of these cultivation treatments on the two other major crops of the area, sorghum and groundnuts, have now been examined, and results from a series of field experiments over six seasons indicate that adequate soil moisture is essential to the production of high yields, but that the other soil conditions which are alleviated by these cultivations are not so critical for these two crops as they are for cotton. Reasons are suggested why spectacular increases in yield in response to the treatments are not generally to be expected with sorghum and groundnuts.

An investigation was carried out in 1964 to determine the effect of seed size on oil content and seedling emergence in safflower (Carthamus tinctorius L.). Small, medium and large seeds yielded 37·0, 36·6 and 35·7 per cent oil respectively, and this inverse relationship between seed size and oil percentage (r = −0·9635) was attributed to the increase in proportion of hull in large seeds. In a pot experiment, plants from large seeds emerged earlier than those from small seeds when sown at both 1 and 3 inches depth. Further, plants from large seeds had greater dry weights and leaf areas than the others. The same sort of correlation was also obtained between seed size and early emergence in a field experiment using small, medium and large seeds.

Propagules of five tall fescue genotypes, planted monthly, from April to December, made the greatest growth in April and September and the least in December. Despite the diverse geographic origin of the genotypes, there was no indication of interaction between genotypes and daylength. Propagules grew more vigorously when cut just above the basal leaf sheath then when cut to two or three inches in length. The complete removal of roots, which would facilitate planting, did not reduce growth and appeared to enhance the development of new roots. Propagules consisting of two or three culms grew more vigorously than those consisting of a single culm. Accordingly, if propagules were fragile or if survival was poor, two-culm propagules would become established more readily. Genotypes differed in growth rate even when planting weights were kept uniform. Results of this study could be used to reduce the time and cost of vegetative propagation of tall fescue.