COMPARISON OF BINOMIAL AND IWAO TYPE SEQU ENTIAL SAMPLING PLANS FOR MONITORING ONION THRIPS (THRIPS TABACI) [THYSANOPTERA, THRIPIDAE] IN ONIONS

Transient differences in onion thrips (Thrips tabaci) densities were o bserved between the margins and the centre of some onion (Allium cepa) fields. Onion thrips populations consist in aggregates contagiously d ispersed within onion fields. Two types of sequential sampling plans w ere established for T. tabaci in onions: a binomial plan based on the presence of 5 thrips plant(-1), and one of the Iwao type, requiring co unts of all thrips present. Acceptation boundaries for these plans wer e calculated for economic thresholds of 0.9 and 2.2 thrips leaf(-1). P lans were validated in fields with thrips populations varying between 0.01 and 32.33 thrips leaf(-1). Binomial sequential sampling plans wer e found to be as reliable as plans requiring counts of all thrips, wit h an error rate not exceeding 4%, a level below the theoretical error rates of 5 and 10% used in the calculations. Correct decisions were re ached in 90.4 and 83.6% of the situations with plans requiring counts of all thrips, and in 84.8 and 82.4% of the situations with binomial p lans for economic thresholds of 0.9 and 2.2 thrips leaf(-1), respectiv ely. The average number of samples needed to reach a decision varied b etween 10 and 14 plants according to the plan used. Use of binomial pl ans can save up to 62 and 88% of the counting effort for thresholds of 0.9 and 2.2 thrips leaf(-1), respectively.