New Zealand has a long history of intensive horticulture and many of the greenfield sites designated for residential development have previously been extensively used for horticulture. Currently there is limited information on the potential levels and likely effects of agrichemical residues in horticultural soils available to regulatory authorities with a mandate to manage contaminants in soil. A series of investigations were undertaken to identify the key contaminants present in horticultural soils, to determine the availability and toxicity of contaminants to plants and earthworms, and to estimate likely human exposures to selected contaminants in residential settings. Trace element and selected organochlorine pesticide concentrations were measured in soil samples collected from horticultural and grazing properties in three regions of New Zealand (Auckland, Tasman and Waikato). Elevated levels of arsenic ( <2-58 mg kg- 1 ), cadmium (<0.1-1.5 mg kg- 1 ), copper (5-523 mg kg-1), lead (5-243 mg kt1 ) and LDDT (<0.03-34.5 mg kt1 ) were detected in horticultural soils from all three regions. With the exception of cadmium and zinc, significantly higher levels of contaminants were generally detected in horticultural soils than in grazing soils (p<0.05). Concentrations ofLDDT, arsenic, cadmium, copper, and lead in some soils and particularly orchard soils exceeded soil criteria for the protection of ecological receptors and/or human health. Additional analyses undertaken on the Auckland region samples infrequently detected organophosphorus and organonitrogen pesticides and at levels generally less than 1 mg kt1 in horticultural soils. Acidic herbicides were not detected in the Auckland samples. p,p '-DDE and p,p '-DDT were the predominant DDT residues measured in horticultural soils. The p,p'-DDE:p,p'-DDT ratios in the measured horticultural soils ranged from 0.4 to 9.7. Significant negative correlations were found between the p,p '-DDE:p,p '-DDT ratios and copper in the Auckland (p<0.001) and Waikato (p<0.02) orchard soils. These results suggest that copper may be a contributing factor to inhibited degradation of p,p'-DDT to p,p'-DDE in orchard soils and that the ratio of p,p '-DDE:p,p '-DDT in horticultural soils, especially orchard soils, should not be used as an indicator ofrecent use ofp,p'-DDT. A persulfate oxidation method was used to estimate the fraction of p,p '-DDT in orchard soils available for microbial degradation. The proportion of p,p '-DDT oxidised by the persulfate ranged from 4 to 3 7% indicating that a significant proportion of the p,p '-DDT in the orchard soils (up to 96% in some cases) may not be available for degradation by soil micro-organisms. The availability and toxicity of aged LDDT and trace element residues in orchard and grazing soils to the endogenous earthworm Aporrectodea caliginosa (Lumbricidae) was assessed using a 28 day laboratory assay. The worms bioaccumulated the aged DDT residues by up to a factor of 3 with concentrations of LDDT in worm tissue increasing with increasing soil concentration (p<0.001). Worm tissue concentrations ofLDDT correlated with the amount ofLDDT desorbed from soil by two biomimetic extraction techniques, Tenax resin and C-18 disks, indicating that these methods may be suitable for determining the availability of aged LDDT residues to A. caliginosa. Concentrations of arsenic, copper and lead in worm tissue increased with increasing soil concentration of these trace elements (p<0.01). Cocoon production decreased with increasing soil and earthworm tissue copper concentrations (p<0.05). A glasshouse study was undertaken to assess uptake of LDDT, arsenic, cadmium, copper and lead into lettuce (Lactuca sativa) and radish (Raphanus sativus) from 10 soils containing typical contaminant concentrations. LDDT concentrations in the assayed soils ranged from 0.02 to 12 mg kg· 1 • The maximum plant tissue LDDT concentrations (DW) followed the order radish hypocotyl (190 µg kg- 1 ) >radish leaf (77 µg kg- 1 ) >lettuce (28 µg kg-1 ) and there were significant correlations between all plant tissue types and the soil concentrationofLDDT andp,p'-DDE (p<0.01). Concentrations of cadmium, copper, lead and zinc in lettuce, copper in radish hypocotyls, and copper and zinc in radish leaves increased with increasing soil trace element concentration (p<0.02). Lettuce cadmium concentrations for plants grown on four out of ten assayed soils were equivalent to or exceeded the NZ food standard for leafy vegetables of 0.1 mg kg"1 FW. Lead concentrations in radish hypocotyls grown in three out of eight soils were equivalent to or exceeded half of the NZ food standard for lead in root vegetables of 0.1 mg kg·1 FW. Phytotoxicity was determined by measuring the dry mass yield of lettuce and radish, and by a five day seedling emergence and root elongation assay for lettuce and ryegrass (Lolium perenne). The radish leaf yield and the root length of lettuce seedlings grown in orchard soils for five days decreased with increasing soil copper concentration (total and/or neutral salt extractable). A simulated gastric extraction method was used to determine the bioaccessible fraction of arsenic, cadmium and lead in soil. The maximum percent bioaccessible fraction for the trace elements followed the order arsenic (45%)<lead (83%)<cadmium(100%). Likely human exposures to LDDT, arsenic, cadmium and lead in residential settings under two home grown produce consumption scenarios (10 and 50%) were estimated using methodology generally consistent with current NZ government policy. Estimated daily intakes of LDDT, arsenic, cadmium and lead for lifetime exposures in residential settings did not exceed the WHO tolerable intakes under the 10% homegrown produce scenario. Similarly, the estimated LDDT and arsenic intakes for the 50% homegrown produce scenario did not exceed their respective WHO tolerable intakes. Estimated cadmium daily intakes for the 50% home grown produce consumption scenario exceeded the WHO tolerable intake for cadmium at soil concentrations greater than 0.75 mg kg- 1 for a child and 1 mg kt1 for the lifetime exposure. Estimated child lead intakes exceeded the WHO tolerable intake at soil concentrations greater than 450 mg kg- 1. The results presented in this thesis have implications for on going applications of agrichemicals and soil amendments to horticultural land as well as residential subdivision of former horticultural land. Trace element concentrations on some· properties have already reached concentrations where negative effects on terrestrial organisms have been demonstrated to occur. The results in this thesis suggest that a precautionary approach should be adopted to subdividing former horticultural sites. Horticultural land being subdivided in particular for lifestyle blocks should be assessed prior to subdivision to ensure that thresholds for acceptable human exposures are not exceeded.