Where the environment is more diverse, experiments tend to be widespread
over a large area. Also, as the research methodology is new and at times still
in the making, field staff, collaborating students, etc. are more prone to errors
than under a controlled environment. In our case, for example, the trial on
one of the sites nearly failed, because some of the stakes were too short
(about 1.8 m above ground, instead of 2 m) and were browsed to death. In
the other trial, the 2 month interval treatments were not monitored closely
enough, until a great number of plants were about to die. Had the assistants
been more closely supervised by researchers, these mistakes might have been
avoided.