Abstract
A review has been made on the interlaboratory comparison study on the reliability of environmental analyses conducted by the Environment Agency through 1978-80 FY. Data on Cd, Zn, Pb, Cu, As, total-N and -P in soil and sediment obtained by ca. 200 laboratories including prefectural, municipal and commercial ones were analyzed statistically. Evaluation of data was made in terms of the coefficient of variation (CV) after the rejection of outlying results. The rejection was made according to the combination of Dixon's and Grubbs' methods. Zn, Pb and Cu present in soil in an order of 100 ppm can easily be determined by direct atomic absorption, while Cd in an order of 1 ppm requires solvent extraction. In all of these cases CV was around 10 %. The CV for As was larger than for heavy metals by a factor of about two. Distribution of data showed negative skewness that indicates lower values being liable to be obtained, suggesting the complete decomposition and extraction are a key point for obtaining good results. This is valid also in the determination of total-Cr.
