In an interlaboratory study, 8 French laboratories were tested for their proficiency in using the AOAC mouse bioassay for paralytic shellfish poisoning (PSP). Each laboratory received 1 saxitoxin (STX) standard solution, 1 STX acidified water solution for determination of the titer, 1 noncontaminated shellfish sample, 1 naturally contaminated shellfish sample, and 2 shellfish samples spiked, respectively, at low (152.8 microg STX/100 g meat) and moderate (334.7 microg STX/100 g meat) levels. All samples were analyzed in duplicate. Mean recoveries were 35.1% for the low level and 46.6% for the moderate level. Relative standard deviations (RSD) for within-laboratory variations (repeatability) ranged from 5.4 to 9.8%; RSD for between-laboratory variations (reproducibility) varied from 7.8 to 39.6%, depending on STX level. On the basis of overall performance, all 8 participating laboratories were proficient in their use of the AOAC mouse bioassay.