Yearly, a larger percentage of government regulations are stated in st
atistical terms. More than simple techniques are sometimes necessary t
o comply with these regulations, both in analyzing and designing exper
iments and surveys. The problem considered in this paper is how to eff
iciently sample a mining site to show that the site has been successfu
lly reclaimed. Different sequential analyses are suggested and a two-s
tage plan by Stein is recommended and illustrated. This approach allow
s us to sample with a predetermined probability of detecting a success
ful reclamation. It is pointed out that the usual sample size formulas
are not directly applicable to this problem.