Nutrient Cycling in Agroecosystems 54: 267–277, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands. 267 The effect of long-term phosphatic fertiliser applications on the amounts and forms of cadmium in soils under pasture in New Zealand C.W. Gray 1,∗ , R.G. McLaren 1 , A.H.C. Roberts 2 & L.M. Condron 1 1 Department of Soil and Physical Sciences, Lincoln University, P.O. Box 84, Canterbury, New Zealand; 2 AgResearch, Ruakura Agricultural Research Centre, Private Bag 3123, Hamilton, New Zealand ( ∗ Corresponding author) Received 23 June 1998; accepted in revised form 3 November 1998 Key words: cadmium, phosphate fertiliser, leaching, residence time Abstract This paper presents the results of an investigation into the rate of Cd accumulation and changes in forms of Cd in a soil that has been subjected to long-term superphosphate fertiliser application. Results indicate that there had been a significant accumulation of Cd in the soil during the past 44 years. On the high fertiliser treatment (376 kg superphosphate ha −1 yr −1 ), Cd was estimated to have accumulated at a rate of 7.8 g ha −1 yr −1 . During the course of the trial, there was an increase in the proportion of Cd associated with exchangeable and soil organic matter fractions on the fertilised plots, which was related to a corresponding increase in soil organic carbon levels. Results also indicate that although there was a large proportion of added Cd associated with the organic fraction, the concentration of applied Cd occurring in the residual fraction was also substantial (i.e >25% of added Cd). In addition, an investigation into the effects of residence time of Cd in the soil indicated that there was a redistribution of Cd into less soluble forms with time (i.e. residual Cd), along with a decrease in total soil Cd concentrations. There was also evidence of movement of Cd down the soil profile in this irrigated soil. The implications of these results for Cd phytoavailability are discussed. Introduction Reports in the literature have indicated that soil Cd concentrations are related to the amount of superphos- phate fertiliser added to soil (Williams and David, 1973, 1976; Rothbaum et al., 1986). The Cd content of phosphate fertilisers will vary depending upon the source of the raw rock phosphate used in its manu- facture. In New Zealand, up until the mid-80’s, the main source of rock phosphate used in phosphate fer- tiliser manufacture was from oceanic sedimentary and guano-based deposits from the Christmas Islands and Nauru, with Cd concentrations ranging between 34 and 69 mg kg −1 (Rothbaum et al., 1986). On an inter- national scale, these are considered high Cd phosphate rocks, with Cd concentrations in other phosphate rocks being significantly lower e.g. Israel (Arad) 12 mg kg −1 (Syers et al., 1986) and Jordan 6 mg kg −1 (Bramely, 1990). A recent national survey examining the Cd status of pastoral (fertilised) versus native (unfertil- ised) soil sites indicated that on average, total soil Cd concentrations at pastoral sites were double those of their native counterparts, with total soil Cd highly cor- related with total soil phosphorus (P) (Roberts et al., 1994). The accumulation of Cd in soils from the long- term application of phosphatic fertilisers has been well documented in many different countries (Mulla et al., 1980; Mordvedt et al., 1981; Isermann, 1982; Kofoed and Klausen, 1983; Rothbaum et al., 1986; Baerug and Singh, 1990; Andersson and Siman, 1991; Singh, 1991). However, in New Zealand only limited data of this nature is available (Rothbaum et al., 1986; Loganathan et al., 1995; Loganathan et al., 1997). Recent work by Hamon et al., (1997) with an Aus- tralian soil suggests that a substantial proportion of the Cd added to soil in fertiliser may be immobil- ised to plant unavailable forms. Gray et al. (1998) also provide evidence of Cd ‘immobilisation’ in New