H.—34.

(3) Survey of Limestones in the South Island. The results obtained in the Southland investigations suggested the desirability of obtaining figures for the cobalt contents of all commercial limestones in the South Island. A considerable number of samples has been received from the different lime-works, and they are now being analysed. The full data are not yet available, but variations in cobalt content from 6-1 p.p.m. to 0-2 p.p.in. have been found. The majority of the samples examined showed a cobalt content ranging between 1 p.p.m. and 2-5 p.p.m. One interesting feature of the examinations which have been carried out is the association of high cobalt soils in a number of cases with limestones carrying very small amounts of cobalt. In several cases the residual soils contain 26 p.p.m. of cobalt, while the limestones contain only 0-2 p.p.m. (4) Cobalt in Fertilizers. As recommended by the Cobalt Committee, samples of cobaltized fertilizer have been obtained from time to time from the different fertilizer companies and examined for cobalt. In general, the commercial samples of cobaltized fertilizer have shown a satisfactory cobalt content. After a period of eighteen months' storage, there has been no change in the amount of water-soluble cobalt contained in a prepared cobaltized superphosphate. (5) Use of Cobaltized Fertilizers and other Cobalt-containing Materials. In this section of the cobalt investigations trials have been made of the effect of cobalt phosphate and cobaltized superphosphate on the cobalt content of pastures. Cobalt phosphate used at the rate of 5 lb. per acre still shows, after two complete years from the date of application, a greatly enhanced cobalt content of the pasture. Several series of tests with cobaltized superphosphate used at rates varying from 2 oz. to I lb. of cobalt chloride equivalent per acre have been made on two soil types in the Nelson District. Certain of these experiments are now finalized, and a considerable amount of data is available concerning the duration of effect of small applications of cobalt salts on the cobalt content of the pasture. The applications of cobaltized superphosphate have been made both in the autumn and spring. For a March application, a dressing supplying 1 lb. of cobalt chloride equivalent did not prove more effective in duration of effect than a lb. dressing. In each case there was no increase in the cobalt content of the pasture over the control by the end of the following December. In the case of August applications, a dressing supplying £ lb. of cobalt chloride equivalent per acre was ineffective after the end of December ; in the case of the | lb. and I lb. dressings, some increase in cobalt content of pasture was shown up to the end of the following June. On this occasion, pasture from the 1 lb. treatment was showing 0-16 p.p.m. cobalt, and the J lb. 0-07 p.p.m., while the control figure was 0-06 p.p.m. It must be remarked, however, that the effect of the larger dressings of cobalt salts on the cobalt content of the pastures is pronounced for a relatively short period of three to four months, when the cobalt contents with the different dressings tend to converge. Experiments dealing with the value of high cobalt soils derived from serpentine rocks have been established in the Sherry Valley, and preliminary results indicate substantial improvement in the cobalt content of the pastures by top-dressing with these soils. ••■■ ■ (6) Animal Experiments. Animal tests in connection with the use of cobalt compounds have been continued at Glenhope, Sherry Valley, the Rainy River district, Westport, and Southland. At Glenhope sheep have been maintained in perfect health on pastures top-dressed with cobalt phosphate. During the present season an attempt has been made to correlate health of sheep with the cobalt content of the animal organs. For this purpose a line of twenty-five hoggets has been run on an unhealthy pasture, a few animals ■ being drafted out periodically for slaughter, so that a fairly complete picture should be available of the downward trend of cobalt in the animal as well as an upward trend after the administration of cobalt salts. The experiment dealing with the use of cobalt licks for dairy cows in the Sherry Valley has been finalized. No difference in milk yield resulted from the use of cobalt licks. At Atapo, Rainy River district, an experiment with sheep on the value of cobaltized fertilizers has been carried out, but so far no differences have resulted from the use of the cobaltized fertilizer. It must be noted that this experiment is being conducted on the Moutere Hills type of soil, which has a rather low cobalt content, but not as low as that of the granite soils. At Westport, further drench experiments have been made with calves, in order to ascertain whether copper deficiency in addition to cobalt is present on certain of the pakihi soils. The results of the drench experiment showed that the best result was obtained from the use of a combined coppercobalt drench. The results with cobalt alone and copper alone were inferior to those obtained from the Combined cobalt-copper drench. This result is in keeping with pasture analyses, which have shown low figures for both cobalt and copper on certain of the pakihi soils. In Southland, difficulty has been experienced in obtaining sickness on the control plots. In the 1938-39 season no sickness developed in the case of any of the experimental sheep, but the cobalt figures for livers in the case of the control sheep were very low. This point is of interest, indicating that sickness does not always occur if the cobalt content of the liver is low. During the present season an attempt is being made to induce sickness in the control hoggets which have been carried over from the previous season, and to obtain information from a new set of

43