It’s a little early for a Hallowe’en tale, but yesterday we made a rather grisly discovery in the apiary.  After only 5 months, the Chamomile colony, despite our efforts to help, is completely dead.

Dead bees are all that remains of the Chamomile hive (click any image to enlarge...if you dare)

If we weren’t paying attention to our hives, we might have been shocked to find an empty hive.  Blamed it on ‘CCD’, or any number of other potential, intangible causes.  Sometimes though, why a colony collapses isn’t inexplicable.  We’ve been watching these hives all season, and carefully recording our findings as we go.  Putting all the pieces of the puzzle together, it isn’t really a mystery why this colony has collapsed.

Each hive is photographed, and each inspection is logged, so when things go awry, we can look at overall trends in health, and population

Part of beekeeping, rather than bee-having, is to provide for the bees if and when they need assistance.  To prevent colony starvation, we’ve been feeding all the bees in the apiary during the fall nectar dearth, which can be quite pronounced in this area.  I had promised a post about how, and why we’re feeding, but I’ll postpone that until our next bee post, and for now just focus on the reasons for the demise of this hive.  Rather than throw our hands up, and say ‘oh well, they’re gone’, we feel it’s important to critically evaluate any loss, and try to learn from the experience.

This colony was acquired as a purchased package colony in May, one of two colonies installed that day.  If you’ve been following our beekeeping exploits this year, you may recall that this hive has always been the straggler in our apiary.

As the season progessed, Chamomile, our smallest colony, had less than half the population of our strongest colony

Chamomile’s neighbor to the left, Rosemary, set off to a roaring start, and built up very quickly, much like her feral neighbor, Salvia.  Chamomile however was slower to build up population, and stores, than any of the colonies.  Initially, being new beekeepers, we ascribed a large part of the difference to individual colony variation, weather, and the lateness of the acquisition of this colony. In hindsight though, there was more to it than that.

At first our concern was whether Chamomile’s Queen was well mated.

Chamomile Queen (yellow arrow)

Queen breeders in California suffered setbacks this spring when foul weather hampered the production of spring Queens.  Soon after she arrived though, the Queen was laying eggs, and producing what appeared to be healthy worker brood.  All signs pointed to a good hive, with a well mated Queen.

The first capped worker brood looked excellent (the holes are where the wire is in the foundation)

Toward the end of summer though it was clear that Chamomile was struggling to store much honey.  It hasn’t been a stellar honey year for many in California.  Early in the season even commercial beekeepers were lamenting the failures of the both the citrus and the black sage honey crops, due in part to the protracted, unusually wet spring.  Even so, Chamomile’s neighbors, although not accumulating enough reserves to share with us this season, were at least managing to provide for themselves.

As the summer bee populations drop off as we slide into fall, lurking issues in a hive can become unmasked. By mid-August it was clear that we had an escalating Varroa problem.

After staying clean all of summer, Chamomile suddenly found itself overwhelmed with Varroa

Until this point, the Chamomile hive had been relatively clean, with an all-time high mite drop count of a mere 4 mites.  Certainly not enough to necessarily treat the colony.

One of our feral colonies, however, had been battling a high mite count for a few weeks, so in mid-August we chose to treat all four colonies with Thymol.

By the end of the four-week treatment period however, Chamomile went from our lowest, to our highest mite-burdened colony (relative to total population).

CSI: Chamomile Hive

Looking back through the inspection logs, photographs from through the season, along with the findings today, the most notable differences in this early failed hive, compared to other colonies have been that Chamomile had:

1) Low population throughout the season. A lack of strong build up even during the months with lots of natural nectar and pollen availability.

2) A lack of drones all season.  Colonies that are weak, don’t tend to produce drones.  Producing drones is energetically expensive, both for the Queen, and for the colony at large.  Drones consume reserves, but don’t contribute to the function of the hive, and they don’t forage.  It’s like having a house full of guests that don’t contribute to the rent, and don’t grocery shop.  They clean out the fridge, and don’t pick up after themselves.  If you’re well off, maybe you don’t mind a few extra lazy house guests, but if you’re barely making ends meet, it’s not an expense you can afford.

During our last inspection, we left Chamomile's drone frame in place, because it only contained worker brood, and this colony needed all the population it could muster

All of our other colonies, during the peak season were producing plenty of drones, suggesting they were doing well enough they could afford that expense.  Our interim Varroa control measure prior to treatment, was drone trapping, as drone cells are where mites prefer to reproduce. As such, we were closely monitoring drone production in all the hives, as it’s critical to remove the frames at the right time, or you simply breed more mites.  Chamomile, however, only ever placed worker brood or nectar in her drone frames. During our last complete inspection, Chamomile was the only hive we left the drone frame in situ, because during that inspection it was still filled with capped worker brood.

3) Poor honey storage.  Even at the end of the last nectar flow, Chamomile only had a few frames of honey stored.  Far less than the other colonies.  Likely a function of low forager population, and possibly an early indication of robbing.

Here you can see what should be capped honey, but the honey is missing, with the caps still in place

As of today’s inspection, it was obvious that robbers had cleaned out what remained of Chamomile’s stores now the colony is dead.

Turning the frame over, it was clear robbers had taken the honey through the other side of the frame, chewing though the foundation layer, rather than removing the caps

4) Extreme Varroa mite burden.  This colony experienced an early fall, overwhelming, exponential increase in Varroa drop counts, more than twice that of the neighboring colonies.  This wasn’t unmasked until toward the end of treatment.

During treatment, Chamomile's mite count climbed 3300%

As summer bee populations naturally dwindle in the fall, weak colonies can rapidly become overwhelmed with mites, as they don’t have the population of bees to cope with the level of infestation, and any population losses due to Varroa directly, or indirectly as a consequence of vectored diseases like Deformed Wing Virus (DWV), are much more significant.

Chamomile’s Last Days

After the robbing screens were installed on all of the hives in August, Chamomile was the only hive in the apiary where we noticed yellow jackets defeating the robbing screens.  Just before the recent rains, I was watching the hive entrance to assess how much pollen was going into the hives, and to watch for signs of robbing bees, and saw three yellow jackets, in succession, exiting the hive.

Every hive inspection starts with looking at the ground in front of the hives. Yellow jackets have been a persistent problem this fall.

I knew this was very very bad, but as the neighboring hive, Lavender, was actively defending their robbing screen from intruders, I was reluctant to open Chamomile to see what was going on, for fear of setting off a robbing frenzy in the apiary.  I could only guess how much damage these Yellow Jackets were doing to the hives, as they were no doubt stealing brood from the brood nest, and attacking workers within the colony.

Yesterday, when going up to feed all the colonies, I found a disproportionate number of dead bees at the entrance of the Chamomile hive.

Despite robbing screens, and entrance reducers, these bees were dead at the entrance of Chamomile's hive. Even as I photographed this, Yellow Jackets were exiting the hive.

More evidence something was very wrong. Opening the lid, I could see the ants had suddenly made themselves at home, and only two or three bees where in the area where the food was.  (I also discovered that bee suits are NOT ant proof!)  Most notable was that the pollen we’d put in before the recent storms, was still relatively untouched.  The other colonies have been devouring theirs.  This hive suddenly seemed eerily quiet.

Our other colonies are devouring their pollen patties, but Chamomile's were mostly untouched

Removing the feeder, I could see straight down to the screened bottom board to a small pile of dead bees at the bottom of the hive.  A few bees were wandering about, but it was clear the hive was completely dead.

This emerging worker starved before it hatched

During the inspection of the empty hive this morning, the most notable finding, other than starved, partially emerged brood, was Queen cups, and supersedure cells.  It’s apparent that for whatever reason, the Queen in Chamomile either absconded, was failing, died, or was perhaps even killed by the intruding yellow jackets.

I tore this supersedure cell as I removed the adjacent frame. It contained a desiccated pupa.

This colony was clearly trying to replace their Queen.

I accidentally flattened this supersedure cell while examining the other side of the frame, but this one was still sealed.

Our last complete inspection on September 18 showed no signs of supersedure cells, or Queen cups.

Opening the sealed supersedure cell revealed this dead pupa

Multiple empty Queen cups were also present on a number of the frames.

Empty Queen cups (red arrows), and partially drawn Queen cups (green arrows)

Whether the original Queen absconded, or was killed, we don’t know, but in just over three weeks Chamomile went from a weak, but functioning hive, with eggs, capped and uncapped brood, and descended into panic mode desperately trying to replace their Queen, and failed.

In late September there were still eggs and larvae in this hive

However, even if the larvae in the supersedure cells had hatched as Queens, they likely would still have been doomed to fail.  Drone production ceased here by mid-late August, in all of our hives.  We haven’t seen a drone in the apiary since early September.  Attempting to replace a Queen this late in the season, after the drones are gone, is most likely to result in a failed mating, and the demise of the colony.  The only way to replace the Queen at this time of year, with any assurance, is for the beekeeper to replace her with a known, mated, purchased Queen.

That neither of the capped Queen cells had hatched suggests that the attempt to replace her was too little, too late.  It’s likely the hive population dwindled too much, and the few remaining bees simply abandoned the hive, leaving the remaining brood, and supersedure cells, to starve.

Hive Post Mortem

The good news in all this is there are no overt signs of significant disease in the Chamomile hive.  Not all diseases however leave much evidence behind, especially when there are relatively few bees remaining.  Overall, putting all the pieces of the puzzle together though, it appears that the demise of Chamomile was the result of a perfect storm.  A weak colony, a high Varroa burden, and relentless robbing from yellow jackets, and failed supersedures, all pushed this colony beyond the brink.

After the recent rain, the additional clues that the colony was near failure, was that a number of the bees exiting the hive were no longer the classically light blond bees we were used to seeing.  There were just too many dark bees in the hive, suggesting that the neighboring colonies were taking advantage of undefended stores, and simply cleaning out the hive pantry.

Bees that rob the stores of other colonies can easily overwhelm a weak hive

Two other observations, that were somewhat more subtle, were that in the last week we had noticed that the few bees being mobbed on the neighboring hive’s robbing screen, were in fact little blond Chamomile-like bees.  We questioned why when we first observed this, that bees from a weak colony would try to be gaining entrance to a strong hive. Now it’s evident this was likely a final act of desperation on the part of any Chamomile survivors, as their hive no longer was hospitable to them.

One additional observation was that this last weekend we noticed what appeared to be a number of scout bees investigating the eaves of the house and workshop.  We’d seen this a lot during spring swarm season, but it seemed strange to see these bees looking for a new place to live in mid-October.  Not that we want to catch a late swarm, as they’re often very difficult to overwinter, we went ahead and set out a nuc box in case there was a colony looking for a new home, not realizing they may actually have been some of the few last worker bees from Chamomile.  Now Chamomile is completely dead, the ‘scouts’ have gone.

Lessons Learned?

Perhaps the biggest lessons learned from our experience with Chamomile, are when to be concerned, and when to intervene.  It’s a fine dance between inspecting, and trying to provide everything we can for the bees, versus interfering, and causing disruption in the hive.

Every colony is different, but we’ve learned that even a subtle perceived weakness in a colony can quickly escalate to collapse.

From our reading over the last year, and attending numerous meetings, we knew that ‘weak Queens should be replaced’ in the fall.  The question was, what constitutes a ‘weak Queen’?  Is it always obvious?  Obviously a hive full of drones suggests a Queen is failing, or has been replaced by a laying worker.  The solution is obvious.  It’s time to re-queen.

Next year, as summer winds down, we'll be more proactive about requeening any weak colonies

However, with a Queen that only lays worker brood, with a good laying pattern, why requeen her?  In Chamomile’s case she should have been requeened in late summer when it was clear she had never built up as well as the other hives.  Brood patterns aside, by sheer numbers alone, the colony was weak, and our instincts were right.  We should have replaced her, rather than give her the benefit of the doubt.  The signs were more subtle than a Queen that had overtly failed, but clearly Chamomile hadn’t proved her strength.

The lessons in regards to Varroa are two fold.  One is, never assume because mite counts are low, that they can’t explode at a moment’s notice, especially in the fall.  We proved that during treatment. We knew numbers could climb rapidly, but we didn’t know just how rapidly until September.

What this graph from our last post really showed, was that Chamomile was completely overwhelmed with mites

Even with all the homework we did in regards to Varroa, we were still not prepared for what we found in the Chamomile hive after treatment.  That said, it’s now clear that perhaps some of the aberrant escalation in count was due to yellow jackets opening and robbing out cells containing larva, exposing many mites at once, rather than the gradual exposure that happens with natural brood hatching.  Again, a shortcoming of natural fall mite counts.

As for treatment, overall, we consider the Thymol treatment to have been a wholesale failure, throughout the apiary this fall.  Starvation, and Varroa are still the greatest threats to our remaining colonies.  During our last inspection we made a conscious effort to ensure that all the colonies were still laying eggs, and had capped and uncapped brood.  We did find eggs in Chamomile, but when we found eggs, we stopped digging through the remainder of the hive, to keep the inspection as brief as possible during robbing season.  In hindsight, perhaps we should have looked for the Queens in all of our hives, maybe we would have been alerted to the magnitude of the problem sooner. At the very least we could have combined this hive with another before it completely failed, although, perhaps with Chamomile’s mite count, it was probably best we didn’t.

Any treatment, even ‘organic’ treatments have the potential to harm the Queen.  Even if the Queen survives however, treatment can still interfere with egg-laying. Perhaps Chamomile stopped laying soon after the last inspection, and the colony simply couldn’t sustain itself as the remainder of the summer bees died off?  It’s a double edged sword.  Treating mites may harm a colony, but leaving mite numbers unchecked will KILL a colony.

Varroa Mites on Drone Brood

Having a better understanding of the nectar flows in this area, next season we’ll also pay even closer attention to stores, and the need to step in and feed sooner.  In hindsight, we probably should have been feeding a little earlier in the season, before it was evident that stores were rapidly starting to decrease, to maintain what the colonies had, rather than trying to catch up as stores dwindled.  We weren’t expecting the nectar dearth to be as early, or as protracted as it’s been.

Lessons learned, especially in regards to timing of feeding, and managing Varroa, we hope will make us better beekeepers.  We’ll trap for yellow jacket Queens early next Spring, to help prevent overwhelming numbers in the fall.  We’ll be more astute to robbing (which isn’t always as frenzied as some may suggest).  We’ll be more aggressive with Varroa control starting in early spring by dividing surviving colonies, forcing breaks in the brood cycle, doing regular mite monitoring, drone trapping, and treating early, and keeping the hives balanced in population to decrease the stimulus for robbing at the end of summer.  If all the hives are of similar strength by the end of summer, as the remaining hives are now, the robbers simply seem to give up.

In the meantime, this morning while we were going over the Chamomile hive with a fine-toothed comb, we removed all of the frames that contained pollen, and this is why.  Wax moth.  One more bane of the beekeeper’s existence.

We found two wax moths inside the empty Chamomile hive this morning. It took no time at all before they moved in.

Now with no bees to defend the hive, the wax moths are circling like vultures, trying to move in and wreak their own havoc in the remains of Chamomile’s hive.  They can completely destroy drawn combs in very little time.  Drawn combs are valuable to beekeepers, and we’d like to keep Chamomile’s combs for our spring hive splits.  The pollen will be cleaned out of the frames, and the frames frozen before being stored over winter.  This will take care of any potential wax moth eggs on the comb, to prevent the combs from being destroyed over winter.

With Chamomile gone, we’re now crossing our fingers that our remaining colonies survive, although all of them are still short on reserves for winter, so I’m off to make more feed.  We’ll get into more detail about feeding our colonies in our next post.