A RAPID IMMUNOASSAY METHOD FOR QUANTIFYING MOLD IN HARVESTED WINE GRAPES

A rapid immunoassay developed by the University of California, Davis, and E. I. Dupont de Nemours was compared to the visual inspection assa y presently used by the California Department of Agriculture to determ ine the amount of mold in harvested wine grapes (Vitis vinifera L.). O ver 1000 trucks were sampled with both assays during the 1991 and 1992 harvests from inspection stations in the Central Coast, Livingston, W oodbridge, Ripen, and North Coast wine grape production areas in Calif ornia. The Pearson correlation coefficient (R) between the average of the visual assays and immunoassay was 0.99 and 0.97 for 1991 and 1992, respectively. Analysis of variance of the similarity of the two assay s indicated that it was significantly affected by method of harvest (m achine or hand) (p < 0.01), but not significantly affected by color, s ite, variety, or time of harvest. In 1992, when enough data were obtai ned for analysis, the correlation between the two assays was significa nt (p < 0.01) at all sites but North Coast (R = 0.23). This was probab ly due to the very low level of mold detected at the North Coast site with the visual assay (0.24%). At the other sites, R ranged from 0.66 to 0.76. The lowest correlation of the two assays was with machine-har vested black grapes (R = 0.23). The R for hand-harvested white and bla ck grapes was 0.72 and 0.73, respectively, and 0.73 for machine harves ted white grapes. This was concluded to be due to the difficulty of vi sually detecting diseased berries in machine harvested black grapes. F requency distribution analyses indicated that the visual assay agreed closely with the immunoassay, usually resulting in a variance to mean ratio of less than one.