Effect of some magnesium fertilisers on mineral composition of pasture on Horotiu sandy loam

Publisher Website
Google Scholar
Abstract
In a 5-year experiment on a yellow-brown loam of medium Mg status, finely ground dolomite, serpentine, and talc-magnesite, from Nelson, were compared as Mg sources when topdressed once on pasture at rates providing 110, 220, and 440 kg total Mg/ha. Calcined magnesite (MgO) at the highest rate of Mg and limestone (L) at 4200 kg/ha were also included. Dry matter yields were determined at grazing height, and the experiment was conducted under the technique of mowing with return of clippings. In a second experiment managed under a total removal system, with replacements of P and S (in superphosphate) and K (in KCI) only, dolomite and MgO were compared at rates providing 31.5, 63, and 126 kg Mg/ha, repeated after 21 months, and measured over 4 years. Samples from all yield cuts were analysed for Mg, Ca, K, Na, P, and total N. In both experiments responses in total yields were very small. Overall, MgO increased herbage Mg concentrations more than did dolomite. It had its maximum effect on herbage Mg in the first cut after application, but dolomite required from about 6 months at the lowest rate to 18 months at the highest rate to achieve maximum effects on soil Mg and Ca, on pH, and on herbage Mg concentrations. L increased, and MgO depressed, herbage Ca concentrations, but dolomite (CaC03. MgC03) had no effect. No clear effects of treatments on herbage P, K, or Na levels were detected. Ground serpentine and talc-magnesite failed to change soil pH, and even at the highest rate of application had relatively small effects on soil and herbage Mg levels. In Exp. 2, maximum recovery of Mg in herbage removed was only 11.4% in the first 21 months and 6.0% in the next 27 months. In Exp. 1, in 1967 and 1969, the correlation coefficient between herbage Mg in late winter to early spring and logarithm of soil Mg levels was 0.97. This good relationship for the one uniform site contrasts with the earlier very poor correlations obtained for pooled data from different sites. Results indicate that Mg fertilisers at low rates cannot be expected to raise herbage Mg concentrations appreciably when the soil Mg status is high.