A minimum sample size of 10 000 tomato or pepper seeds
is recommended by the International Seed Federation (ISF)
[http://www.worldseed.org/isf/ishi_vegetable.html] with a
maximum subsample size of 10 000 seeds. This sample size
should allow detection of down to 0.03% contamination
with 95% confidence. Smaller subsamples (e.g. 5 9 2000
seeds) are recommended if high populations of saprophytic
bacteria are likely to mask the potential presence of Xanthomonas.
Smaller sample sizes can be tested with this protocol
but with a lower confidence of detection, which
should be defined for each sample size used. For example,
for protected cultivations or high-class hybrid seeds, the
minimum number of seeds can be lowered to a single sample
of 2000 seeds with 98% confidence of detecting down
to 0.2% contamination. The tests described have not been
validated on treated or pelleted seeds.