>Use controls where possible.  Separate out results from different treatments
>so you can tell whether it was the essential oil or the FGMO which was the
>decisive factor.  Recent posts suggest that FGMO is not as good as was once
>thought but unfortunately this cannot be confirmed or denied from your
>results.  Use your own colonies for comparison: your neighbours probably will
>not have been smoked and disturbed as often as yours and this may have a
>bearing on the mite fall.

What do you use for control hives?
Untreated?
Treated with Apistan(R)?

With all of the additional factors that could enter into the experiment,
the sample size would have to be rather large to show any solid
indications. The advantage a queen breeder would have is that he has access
to a large number of queens from the same stock. Even so I would guess at
around 100 hives, 50 as control, to get solid numbers.

Any thoughts?