H.—34

(b) Magnesium Deficiency in Apple-trees.—Variations in the chemical composition of leaves at different distances along the new season's growth make it essential to sample leaf material judiciously. Magnesium has been found to be at a minimum in the lower leaves, particularly the fifth and sixth from the base, and there is a progressive increase to the tip leaves. Thus, with Cox's Orange, values in leaves from magnesium-deficient leaders range from 0-04 per cent, (as Mg) to 0-37 per cent., and with healthy leaders from 0-08 per cent, to 0-4-2 per cent. There is a marked inverse correlation between magnesium and potassium contents of the dry matter of apple-leaves. The respective values for potassium (as K) in the examples cited above are 2-5 per cent, to 0-5 per cent., and approximately 1-3 per cent, to slightly less than 1-0 per cent. In potassium-deficient leaders, however, the values are of quite a different order. Magnesium did not fall below 0-20 per cent., with an average of approximately 0-3 per cent., while potassium varied from below 0T per cent, to only 0-46 per , cent. These data strikingly emphasize the need for first obtaining precise information on mineral distribution in the various leaves before reliable sampling in deficiency studies can be achieved. Mineral Content of Phormium tenax. —Samples of leaves from the flax-nutrition experiments at Sergeant's Hill, Westport, have been analysed for nitrogen and mineral contents. Six different treatments were represented—namely, superphosphate, superphosphate plus nitrogen, superphosphate plus potash, superphosphate plus nitrogen plus potash, nitrogen plus potash, and untreated. On this pakihi soil a high deficiency of phosphate occurs and growth is exceedingly poor both on the untreated plot and that manured solely with potash and nitrogen. In the leaves on these two plots only 0-14 per cent, and 0-12 per cent. P 2 0 B respectively were found. The other samples gave 0-20 per cent, to 0'32 per cent. P 2 0 s . The use of phosphate was associated with an increase in both lime and magnesia contents. Potash and nitrogen contents were not directly correlated with the use of these fertilizers in the manurial treatment of the flax.

RESEARCH WORK AT AGRICULTURAL COLLEGES Grants were made by the Department during the year to Canterbury Agricultural College and Massey Agricultural College for a number of projects, which are reported on below. CANTERBURY AGRICULTURAL COLLEGE Subterranean Clover Investigation J. W. Caldee The sixth grazing season has been the most productive experienced in the course of the investigation. As a result of good autumn rainfall and a mild winter it was possible for the first time to winter the ewes and hoggets on herbage produced on the experimental area without the assistance of hay or other supplementary feed. Surplus growth in the spring was grazed by five hundred hoggets over a period of one month. The carrying-capacity, expressed in terms of ewes per acre, is as follows : —

The production in any one year is influenced by climatic conditions, especially rainfall. The most productive season was that of 1944, when 9'5 in. of rain fell in February, March, and April. This assured an early and dense re-establishment of the clover. The winter was mild. Over 5 in. of rain were recorded for August, September, and October, and there was almost a complete absence of hot, dry winds in the spring. This combination of favourable climatic conditions resulted in record production. The annual variation in production emphasizes the need for care in the management of a light-land farm where subterranean clover is extensively used. The generally accepted policy of avoiding overstocking and providing adequate reserves of feed must be adopted. The decision to carry a predetermined number of ewes during the ensuing season can be made in the autumn. By this time re-establishment of the clover has occurred and a fair estimate of the coming season's production can be made. Effect of Fertilizer Treatment on Production.—lt will be observed that treatment A, which started off below treatments B, C, and D in the first year, surpassed treatment B in the second year, and treatment C in the sixth year in which it approaches treatment D. It would appear— (i) That the relatively light but recurring applications of lime and super, (equivalent to £ cwt. super, and 2| cwt. of lime annually) is more effective than heavy annual applications of super, (treatment B —2 cwt. annually): (ii) That a heavy initial application of lime plus annual application of super, (treatment C) gives high production in the early years, but the effect of the lime diminishes, and by the sixth year production, while still higher than where no lime has been applied, is lower than where repeated light applications of lime and super, are applied (treatment A) : (iii) Potash (treatment D) has shown an advantage throughout, but here again the diminishing effect of the initial application of 1 ton of lime is apparent.

44

Ewes per Acre. First Second Third Fourth Fifth Sixth Year. Year. Year. Year. Year. Year. A treatment: 1 cwt. super, alternating with 5 cwt. lime .. 1-0 1*3 1-6 2-1 1-6 3-1 B treatment: 2 cwt. super, annually .. .. .. 1-3 1-2 1-2 1-4 1-7 2-3 C treatment: 1 ton lime initial, 2 ewt. super, annually .. 1-6 1-7 1-8 2-0 1-9 2-9 D treatment: .1 ton lime initial, 2 cwt. super, and J cwt. 1-6 2-1 2-3 2-3 1-8 3'2 potash annually Average wool-clip in pounds: Ewes .. .. .. 8-0 9'1 9-4 8-2 9-3 11-5