scope:

INTRODUCTION

Since 1945 [1] many scientists have advocated use of toxicity data in predicting potential damage to a receiving stream before damage occurred. Toxicity data have been used for predictive, comparative, and regulative purposes, and the acute toxicity test has become a "workhorse" in monitoring pollution effects for management decisions even though the reliability of such tests has been questioned [e.g., 2,3,4,5,6,7]. In January 1981, the U.S. Environmental Protection Agency (USEPA) distributed a draft document on the use of effluent toxicity testing in the second round of the National Pollutant Discharge Elimination System (NPDES) permit Issuance [8]. The USEPA proposed that a 24-hr screening test be conducted on a sample of 100% effluent. The proposed criterion for passing the screening test was 80% or greater survival of test organisms after a 24-hr exposure. If survival fell below 80%, a 48-hr definitive test was to be conducted.

Many Industries do not have toxicity testing capabilities, and they usually contract laboratories to conduct tests for them when they are required to do testing. Many different protocols are available for conducting acute toxicity tests [e.g., 9,10,11,12,13], Although attempts have been made to standardize test methods, differences still exist between methods, even though some are minor. In addition, possible differences exist in the interpretation and understanding of a test method by the people conducting a test. As a result, toxicity test results obtained for effluents possibly may not be comparable from test to test conducted within a laboratory and among laboratories.

The purpose of this research was twofold: (a) to compare the inter- and interlaboratory variation in toxicity test results using reference toxic ants and refinery effluent samples, and (b) to evaluate the utility of the USEPA proposed toxicity tests for industry. Duplicate effluent samples from five refineries and a synthetic mixture were sent to three laboratories for toxicity testing over a period of six weeks.