In each of two seasons, undisturbed lysimeters 0.8 m in diameter and 1.05 m
in length taken from five soil types were cropped with winter wheat. They
received autumn applications of the pesticides isoproturon and linuron as w
ell as a bromide tracer and spring applications of dimethoate and MCPA. Lea
chate was collected at regular intervals and concentrations of the various
solutes determined. Rainfall from December to March was 290 and 191 mm in t
he first and second seasons, respectively. Both springs were exceptionally
dry with less than 50% of the mean April-to-June rainfall of 138 mm. Total
flow from the lysimeters ranged from 335 to 477 mm (and from 0.78 to 3.95 p
ore volumes) over the two seasons. Leaching to drainage of bromide highligh
ted soils where preferential flow was influential with total losses ranging
from 24% of applied for a strongly structured, alluvial clay loam to 79% f
or an unstructured sand. Leaching to drainage of isoproturon (K-oc approxim
ate to 100 mlg(-1)) was observed from all but a peat soil with losses great
er (0.31-1.01% of applied) from the clay loam and a deep medium loam, where
patterns of leaching clearly indicated preferential flow mechanisms, than
from the sand and a light loam over gravel (0.04-0.18% of applied) where a
broad breakthrough curve indicated that matrix flow was more important. Lin
uron (K-oc approximate to 500 mlg(-1)) was detected in occasional samples o
f leachate from the clay loam, the light loam over gravel and the medium lo
am during the first season only (maximum loss 0.12% of applied). The sandy
soil, often considered most vulnerable to leaching, gave the smallest total
losses of pesticide of the four mineral soils, whilst significant preferen
tial flow in the deep, medium loam was believed to result from a compacted
topsoil. Neither of the spring-applied pesticides was detected in the leach
ate, as flow following application was very small and relatively slow. (C)
2000 Society of Chemical Industry.