H.—29

to protective work on the bad bend at the lower end of the property, necessitating practically a complete renewal of the fascining of the bank with willows, in addition to re-erecting the stock fence further back, owing to bank-erosion. A proposed cut through the low-lying ground on the opposite bank would give relief of a constantly-recurring expense in renewing protective work and prevent a decided menace to established work down-stream. Field Experiments. The total number of field experiments conducted by officers of the Fields Division was 984, compared with 955 carried out during the previous year. Most of these experiments were on the properties of selected farmers, while the remainder were located on experimental or demonstrational farms. Many of the projects were carried out in collaboration with the Plant Research Bureau or with the Wheat Research Institute. A close degree of co-operation with these organizations is necessary so that the results of fundamental research work may be followed up by field experiments, and this is mainly brought about through various committees of research workers and extension officers. DESCRIPTION AND PROGRESS OF EXPERIMENTS. 1. Grassland. (a) Yield Trials under Mowing and Grazing Technique. —At Marton Experimental Area six experiments are being conducted under this heading, and the scope of these was referred to in the last annual report. One of the experiments referred to previously —viz., a trial with different cocksfoot strains —has been finalized recently. At the Ruakura Farm of Instruction three mowing and grazing trials, one on each of the three major soil types of the middle Waikato basin, are being continued, and these investigate the effects from lime, phosphate, and potash respectively. The response to date from superphosphate on limed ground has been just about double that from super on unlimed ground. As the experiments were laid down on newly-sown pasture and the results quoted above are not substantiated by observational trials on old pasture, a further experiment to investigate the effect of lime on newly cultivated soil as compared with undisturbed soil is now in progress. The general technique of mowing and grazing trials, particularly the alleged disturbing influence of transference of fertility from one plot to another, is to be tested in a new experiment. In addition to the above, a modification of the mowing and grazing technique has been adopted to measure the influence of cultivation on a sod-bound sward of paspalum. fb) Observational Top-dressing Experiments. —The survey of fertilizer responses on grassland is being continued, and in those districts which have been covered by the soil surveys plots are being established on each major soil type. Although the effect of phosphate, potash, and lime is the chief object of these investigations, most of the experiments are designed to compare various forms of phosphate in addition. It has not so far been possible to include " Heskett " slag on account of supplies of this material not being available. (c) Demonstrations and Trials of Grass and Clover Strains. —These trials, which are carried out in collaboration with the Grasslands Division of the Plant Research Bureau, continue to provide valuable demonstration areas to farmers, and they also serve as experiments to investigate the behaviour of the new pedigree lines of grasses and clovers under various climatic conditions. (d) Grazing Trials. —Of the six grazing trials carried out, two in Taranaki are designed to compare production from fields treated with potash as against fields not receiving potash. The remainder investigate production of rye-grass strains under grazing. One of the latter, established at Winton Experimental Area, compares production of wool and lamb from fields sown with better-type Southland seed with production from certified rye-grass fields. The outstanding feature of the first year's results was the higher carrying-capacity of the certified rye-grass areas, and this represented a monetary gain over the Southland rye-grass of at least 33 per cent, from wool and lamb actually sold. (e) Clover-inoculation Trials.—Trials carried out in collaboration with the Field Mycologist of the Plant Research Bureau investigate the value of inoculating white clover and also test out different strains of culture in various parts of New Zealand. (/) Subterranean Clover Experiments.—Eighty-one experiments are being carried out with subterranean clover to determine the utility of various strains and also to investigate the possibilities of surface-sowing of this species on hill country. The failure of the Dwalganup strain in all districts where it has been tried so far has been one of the features of the experiments. It has also been indicated that in Central Otago subterranean clover is unable to thrive under extreme dry and cold conditions, but further experiments which have been laid down will enable the exact climatic range of this species to be determined. (g) Investigations into Effect of Pastures on Feed Flavour. —This work has been consolidated at Ruakura, although some of the Morrinsville pastures on which experiments were previously carried out are being kept under observation. The past season was not generally conducive to feed flavours arising from pastures, but valuable data has been secured from the grazing experiment at Ruakura in regard to the effects of pasture management and manuring on changes in sward composition. Following work conducted by the Dairy Research Institute, which suggested that feed flavours can be much reduced by removing the herd from grazing three hours before milking in the afternoon, information has been sought regarding the effect of this procedure on production. At the Waimate West Demonstration Farm an experiment conducted with two herds, one being removed three hours before milking and the other normally treated, indicated that no loss in production need be anticipated by adopting the procedure suggested.

37