Recoveries were measured by adding pure standards to strawberry and black currant samples prior to S. HaÈkkinen et al. / Food Research International 32 (1999) 345±353 347 extraction and hydrolysis. The recoveries were good or tolerable (70±96%) for most of the compounds studied (for details, see HaÈ kkinen et al., 1998). In the calculation of the final results, the recoveries of the pure standards were not taken into account. To study repeatability (within-day precision) of the HPLC method, a hydrolyzed black currant sample was analyzed 6 times within 1 day. The coefficients of variation (CV) were between 5.4 and 7.3% (HaÈ kkinen et al., 1998). Within-laboratory reproducibility (day-to-day precision) of the whole method (including sample extraction and hydrolysis plus HPLC analysis) was studied by duplicate analyses of a freeze-dried black currant sample on 6 days during a period of 6 weeks. The CVs were between 7.7 and 19.5% for the main phenolics present in black currant (HaÈ kkinen et al.).