H.—34

Linen Flax. —In association with the Botany Division an investigation was made into the relationship between soil background and quality and yield of linen-flax fibre. This work, which concerned Geraldine County principally, was completed for the 1943-44 season and a report prepared. Great Barrier Island. —A survey of the soils of the Great Barrier Island and a report prepared for the Minister of Rehabilitation. The island was found to contain approximately 8,000 acres of ploughable land which could, be developed for dairying, 20,000 acres which could be developed for grazing, and 30,000 acres which should be preserved in forest. At present about 1,500 acres are used for dairying, 24,000 acres for grazing, while the remainder is in forest, scrub, and swamp vegetation. Many of the soils of the island are eroded, this being due mainly to frequent scrub fires. Most of the soils require hea.vy applications of lime. There is no limestone on the island. Agricultural Development, Southland. —A general survey of 150 square miles of the Southland Plains to the east of Invercargill was made at the request of the Public Works Department. This survey covered only the broad soil pattern and delineated soils suitable for development. Irrigation, Ashburton County.—ln the Ashburton County six border-dyked paddocks which will be irrigated next summer have been carefully sampled to determine their fertility status. This will give a reliable basis for observations on chemical and fertility changes under the irrigation water. Saline Soils.—Further samples have been analysed from the Kaipara Harbour reclamation scheme and Timaru Airport. In both cases there has been little improvement in the more saline patches. An area near the Waimakariri, Canterbury, was investigated for the Disabled. Soldiers' Re-establishment League. The salt content of the soil is too high to enable basket willows to be grown successfully. Kerikeri Citrus Orchards. —Complete soil and leaf analyses were made from citrus areas at Kerikeri. The soils showed low lime, while the effects of top-dressing were reflected in lowered potash and magnesium compared with roadside samples. The magnesium figure was low enough to suggest the trial of magnesium compounds as top-dressings. Leaf analyses also indicated that magnesium should be tested, but none of the data was conclusive enough to be certain that this was a primary cause of lowered production. In co-operation with the Plant Diseases Division and the Horticulture Division, experiments were laid down on two orchards. These were designed to gauge the response of orange and lemon trees to magnesia, manganese, zinc, and molybdenum. Pacific Islands Soils. —Advice has been given to other Departments on the plant-food content of soils received from Pacific islands. An article has gone forward for publication on the laterite soils of New Caledonia. Dunging and Urine Plots, Lincoln. —In co-operation with the Grasslands Division the soil laboratory is following the changes induced by the return and non-return of dung and urine to pasture at Lincoln. Glasshouse Soils. —At the request of the Horticulture Division an investigation was made of glasshouses where yields were decreasing. Analyses indicated that potash had been used greater than optimum amounts. Shortage of staff precludes expansion on problems such as these, but there is no doubt that much work of a detailed nature could profitably be done in this field. Probably nowhere else does man change the nature of the soil so quickly and in different ways. Each soil has to be considered individually. Compost. —At the request of the Internal Marketing Division we have investigated the large-scale composting of vegetable wastes, particularly cabbage. This material is difficult to handle, in that it is attractive to flies and easily waterlogs in the bins because of the high water content (90 per cent.). From pilot experiments suggestions have been made for the large scale bins. Free drainage and good aeration are called, for, and if this is supplied cabbage is capable of breaking down very quickly. Pot Work. —Pot work has continued during the year in using Virginia stock to evaluate the fertility of soils sent from different parts of New Zealand. A search for a better phosphate-indicator has led to cress being used on large-scale experiments, and data from these are now being assembled. In collaboration with Miss Cocks Johnston, exhibits have been prepared for the Health Department mobile shows. Pot exhibits have illustrated the need for lime and phosphate, and the difference in natural fertility of the various soils in New Zealand. Soil Tilth. —An investigation has commenced to evaluate in physical terms good and poor tilth. A deteriorated farm in South Canterbury has provided a good starting-point for testing methods of measuring tilth and pore space. Such data is likely to be of fundamental importance in assessing methods of bringing back to good tilth soil that has deteriorated through overcropping. Parallel with the physical investigation a start has been made to fractionate the humus from good and poor phases of the same soil type. Fluorine. —Analyses for fluorine from the main soil types has continued. By comparison with American data, New Zealand soils would appear to have an average content of fluorine. Soil Mechanics. —Among the investigations undertaken this year have been— (1) High Bank, Ashburton. —An embankment had failed on saturation with water due to a high air voids content. After investigation a better grading of material of high density was suggested : (2) Nihotupu Dam. —Partly as the result of the investigations mentioned in last year's report, a tender was accepted for the construction of a rolled fill earth dam. The specifications for the contract include soil mechanics data and represent a big step forward in the application of laboratory data for engineering construction. Work is in progress in deciding how far deviation from the specifications can be allowed, for safety : (3) Turakina Tunnel. —Among the possible methods of stabilizing running sand behind the tunnel is chemical stabilization. This involves forming an artificial silicate in the sand through pumping chemicals into it. Small-scale experiments were in line with experience overseas that running sand could be stabilized this way, and while the method was not put into actual practice data is now available should the occasion arise again : (4) Waipori Dam. —Investigations were made of borrow-pit material for an earthen causeway : (5) Lake Kimihia. —Tests have been made at the request of a private contractor on the site of a proposed opencast mine. Advice was wanted on a safe batter for the cut to the coal and the likely permeability of the stop-bank material that was to hold back the lake from the workings :

35