The use of chloramphenicol in veterinary medicine was banned in the EU in 1994. As the Community Reference Laboratory for antibiotic residues in food of animal origin, one of our functions is to organize inter-laboratory studies. A first inter-laboratory study for the analysis of chloramphenicol (CAP) in milk by ELISA kits was organized in 2001 and a second one for the detection of CAP in pig muscle by ELISA kits in 2002. These studies were intended to allow participants to control their CAP ELISA methods when used routinely and also to compare the performances of various ELISA kits for the detection of chloramphenicol in milk (commercial or in-house kits). In 2001, 15 participants received ten randomly coded frozen milk samples (four blank samples and six spiked milk samples from 0.5 to 5.0 μg/l). In 2002, 20 participants received eight randomly coded frozen muscle samples (two blank samples and six incurred muscle samples from 2.1 to 6.5 μg/kg). They were asked to analyse each sample in triplicate with the ELISA kit of their choice. Different kits from different suppliers were used and compared in the two studies. The results of the two inter-laboratory studies on ELISA kits were satisfactory regarding qualitative results. The global rates of false compliant results of 2.2% for milk and 0.0% for pig muscle samples were lower than 5% whichever kit was used. The global rates of false non-compliant results (16.7% and 10% for milk and muscle respectively) were also satisfactory. The distribution of false non-compliant results depends on the kind of kit used and on the detection limit for milk as well as for muscle. Moreover the sample preparation was very important to avoid false non-compliant results. Finally, this study demonstrates that ELISA kits for chloramphenicol in milk and muscle globally show good repeatability and accuracy. So these kits could be considered as suitable tests for screening purposes.