Tom Barrett said:

> There are another 15 hives in this apiary. I must presume that
> varroa has struck every hive.

> I will check all of the hives as soon as the honey is taken off.

Yes, you should check.  Sadly, you are doomed to check all
your hives on a regular basis until further notice.

You may check sooner than "when honey is taken off".  You
may want to try "sugar rolls" or "ether rolls" while the supers
are on to see if any hives are infested badly enough to prompt
immediate attention.  Ignoring the mite problem "until the supers
are off" is, in my view, the sole cause of hives that die out soon
after supers are removed, an often-documented scenario.

> But presumably the infestation in some hives will be greater than
> in others. And indeed some hives may show no varroa on the inserts -
> but does this mean that they do not have varroa or merely that no varroa
> were killed by the Bayvarol?.

It is hard to say.  It may be better to view ALL hives as "infested with varroa"
to one extent or another, and simply view the tests as indicators of when
to treat any one hive.  I do regular "sugar rolls" on my hives, as the impact
on the bees is nil, and I also do regular "powdered sugar dusting" to keep
varroa levels down when the supers are on (and the varroa population can
get out of hand).  But I DO NOT "sugar dust" any hives that test as "clean".
(Of course, I have no hives that have tested as "clean" during July an August,
so I do end up sugar-dusting 100% of my hives...)

Why treat anything that does not exist?  While one must admit that any/all
tests may miss a low-level infestation, a low-level infestation is not "worth"
treating.

> Should I treat only those hives with major infestations (and with a lot of brood),
> and not treat those with light infestations or no apparent infestations?

Yes, but the decision of what level of mite drop implies the need for treatment
is not "standardized" in any way.  There are no absolutes in this little war
we are fighting against the varroa.

I would submit that the non-toxic approach of using "sugar dusting" to knock
down the mite population during the critical summer months when supers are
on allows one to establish a much lower tolerance in the decision-making
about what level of infestation implies the need for treatment.  When one has
a low-cost non-toxic treatment that does not require one to remove supers,
one can afford to treat at lower "mite-drop" levels.

> And what constitutes a major infestation and what constitutes a light infestation?

There are no standards in this area, because the tests vary so much (even if using
Bayvarol or Apistan, there are simply too many variables from hive to hive, such as
temperature, humidity level, amount of hive ventilation, and so on).

> I would imagine that a three mite drop is a light infestation.

Agreed.  But how much longer until you pull supers?  Let's assume that you
got 100% of the emerged mites (those not in cells reproducing) in your test.
If one had 3 mites "on the bees", it is reasonable to assume an equal number
"in the cells".

a)  As bees emerge, so do the offspring of any mites that are in specific cells.
     In about a week, the offspring are ready to reproduce, and move into cells
     containing bee larvae.  Female mites can live 2 - 3 months in summer, and
     5 - 8 months in the fall/winter, and can reproduce many times, each time
     laying 4 - 6 eggs.

b)  So, if you detected 3 mites, and assume that this is 100% of "mites on bees",
     you may have 3 mites "in cells", and may soon have 10 to 15 mites (male mites
     die after mating, so you never would get a full 6 mite increase from every mating).
     From there, the numbers start to get scary, not matter what ratio of male to female
     mites you assume out of each "generation".  Even a minimal "powers of 3" (where
     3 reproducing offspring per female are created) rate leads to:

        3, 9, 27, 81, 243, 729, 2187, 6561, 19683, 59049, 177147

Read up on the "screened bottom board" (sometimes called the "Beltsville board"), and
consider adding them to every hive.  Thorne even sells the 8-mesh screen one needs
to build one.  Since some number of mites fall off bees without any treatment at all, this is,
in my view, the best method the detect infestation.  One can count "mite drop per week",
which is a much better "sample" than a 48-hour test, and can be done at any time, even
when supers are on.  One really need not "count" the mites, since a glance is all one needs
to triage a hive into "OK", "Treatment Needed Soon", or "Critical Condition - Treat At Once"
categories.

I tend to view mites as "weeds in the garden", and apply the same rationale to both
"weeding" and sugar-dusting.  My gardens will never be 100% weed-free, and I don't
expect my hives to be 100% mite-free.  But ignoring either means that I will soon
have a mess.  The good news is that I have never needed to sugar dust more than
once a month for any one hive.  I wish I only needed to weed once per month.  :)

        jim

        farmageddon