An interlaboratory study of a graphite furnace atomic absorption spectrophotometry (GFAAS) method for the determination of lead in wine was conducted. Seventeen laboratories from France, United States, and the United Kingdom, using a variety of GFAAS instruments, took part in the study. The method incorporated a novel matrix-matching procedure to minimize matrix effects between standards and samples. Six wine test materials were prepared and sent to participants as 12 blind duplicate or split level samples. There was good agreement between results obtained from participants and target values (24-279 micrograms/L) obtained with an inductively coupled plasma-mass spectrometry method. The precision of the GFAAS method was well within the range predicted by the Horwitz equation for the 6 test materials analyzed. Repeatability standard deviations ranged from 3 to 17%. Reproducibility standard deviations were in the range of 10 to 30%. The method is recommended for use for official purposes.