Last Friday we spent the afternoon working in the apiary.  At the end of February we had split the Salvia colony into our (then empty) Rosemary hive.  As a month had since passed, we expected we should see evidence of a new Queen by now, so we needed to see how this colony was progressing.  There was also rain forecast over the weekend, and as yet we had not managed to split our Lavender hive this season.  As it seems bees prefer to swarm on the first warm sunny day after a rain, we decided we really should split before the weather changed again.

A month ago we divided part of the Salvia colony into the Rosemary hive, hoping they'll make a new Queen

Things got off to a bumpy start in the apiary though.  Mr. Curbstone was standing a least 12 feet to the side of the Lavender hive when he was investigated, and promptly stung right on the top of his head, while preparing the smoker.

With the bees on high alert, there was no question of using the smoker

This was seemingly unprovoked, but the bees seemed a lot more defensive than the last time we manipulated the hives.  After removing the stinger, and quickly donning his veil and gloves, we went ahead and set to work.

Splitting Lavender

Recent activity at the entrance to the Lavender hive suggested they were building up dramatically.  Like Salvia, the Lavender colony began to build up in January, and the population increase at the entrance over the last few weeks was impressive. Mr. Curbstone’s sting may simply have been a result of the Lavender hive having a surplus of guard bees wary of any activity close to the entrance.

The entrance to the Lavender hive has been increasingly busy in the last few weeks

Lavender was an afterswarm captured from the same feral colony as Salvia last April.  This colony has never been quite as robust as Salvia, but has been a very strong colony nonetheless (Salvia seems to be the exception when it comes to producing hoards of bees).  We know the Lavender Queen is only one year old, as she was a virgin Queen when she arrived, and it was a number of weeks before she began laying eggs.

Our goal on Friday was to see how Lavender was building up so far this season, as we hadn’t checked inside this hive, except to feed, since late last fall.  Providing there was ample population, we’d then divide the colony in half, much as we did with our early season split of Salvia in February.

As hive entrance activity was booming, we weren’t surprised to see this when we opened the lid.

The Lavender bees had almost consumed the pollen patties, and fondant we left for them in late winter

Plenty of bees on the Vivaldi board, devouring pollen, and finishing up the last of a fondant block.

The upper hive body, number 5, was primarily honey, with a few undrawn frames, that we transferred after the original Rosemary colony failed in January, and there was clearly plenty of activity.

The upper hive body was crowded with bees

The bees had been busy drawing out the comb on this frame, with lots of new white wax.

Drawing out wax on this frame gives them more room for nectar stores in spring

Our overall impression was that Lavender wasn’t quite as overpopulated as Salvia had been when we split that colony in February.  Lavender in comparison had much less burr comb built up between frames.  Salvia had been packing drone comb in anywhere it would fit, but there was less burr comb to remove in Lavender, which made spring cleanup a little easier.

Unlike Salvia, Lavender only had a few drone pupae between the frames, like this strawberry-eye stage drone

We started by looking for eggs, as eggs are necessary for the colony to raise a new Queen.  Eggs also suggest the colony still has a functioning Queen.  When scanning for eggs, sometimes it’s easier to find younger uncapped larvae first, and work outwards until eggs are located.  Lavender had no shortage of young larvae and eggs.  This Queen has definitely been busy.

Larvae from the Lavender colony

As we worked through the hive, frames containing larvae and eggs were transferred to the empty Chamomile hive that was robbed out in October.  We also moved frames of pollen and honey stores.  We tried to be more aggressive with this split than we were with Salvia, dividing the colony more cleanly in half.  We somewhat under-split Salvia, as you’ll see later in this post.

There's no question Lavender is building population

With as busy as this colony seemed, we were surprised we hadn’t found any Queen cups.  It is swarm season after all.  Just as we thought we were about done with this split though, toward the bottom of the brood nest, in the lower corner of a frame, we found this…

Larger than an empty Queen cup, this was clearly being drawn into a Queen cell...but was it empty?

…a Queen cell!  Not just an empty Queen cup, as we had found in Salvia in February.  This was an uncapped Queen cell, with a Queen larvae inside.  Between the sun shining in my face, and trying to focus while looking through my veil, I didn’t manage to get a clear shot of the larvae, but you can see the end of this Queen cell is not yet sealed.  There was clearly a white Queen larva inside (just not clear in the image, for which I apologise).

This cell contained a larva, and was close to being capped

Conventional wisdom is that once these cells are capped, no matter what the beekeeper does, the colony is committed to swarming.  This was the only Queen cell we found, so hopefully we caught this colony just in time.  We’d prefer not to lose the Lavender Queen if we can help it as this colony has done so well for us over winter, and she’s still a relatively young Queen.

When encountering Queen cells during a split, it is usually recommended to move the existing Queen, along with some brood, and bees, to the new hive, and leave the Queen cell, and remaining colony intact in the parent hive.   The belief is that this simulates a swarm more closely.  Like Salvia though, Lavender is such a large colony for this early in the season that locating the Queen was like looking for a dwarf’s needle in a giant’s haystack, and we didn’t want to leave the hive open for as long as it may take for us to find her.  Instead, we moved the uncapped Queen cell to the new Chamomile hive, and were very careful to scan each frame being transferred into the new hive to avoid transferring the Queen.

After being empty since October, the Chamomile hive has now been repopulated

With the Lavender colony divided more or less evenly to create the new Chamomile colony, now we wait.  Just as we did with Rosemary.  We’ll keep an eye on entrance activity, and if everything seems normal, we’ll recheck Chamomile in a month.  We hope the Queen cell will give them a slight advantage, and help them establish this colony more quickly.

Rechecking Rosemary

We split Rosemary from the Salvia colony a month ago.  It takes 16 days to produce a new Queen, who then has to embark on her mating flight sometime during the following few weeks, return to the hive safely, and hopefully soon thereafter begin laying eggs.

Development times for workers, Queens, and drones, are different

Before checking on the progress of a split, it helps to know what to expect.  In the split-development timeline, some things are relatively constant, including the time it takes to produce a Queen, and the time of development of a worker bee from a new egg, to hatch.  The variable part is in the middle.  Once a Queen hatches, it can be some time before she takes off on her mating flight.  Whether or not the Queen can fly is largely dictated by the weather.  We’d had a dry start to the year, but we had some significant rainfall events during the month of March, that potentially could have delayed her mating flight.

Considering the timeline to produce a new laying Queen, and factoring in the weather, we hoped at a minimum to see eggs in the Rosemary colony by now, but knew it may still be too early to see capped brood.

The green hive, Rosemary, was split from the adjacent Salvia hive

Before opening the hive, we observed the hive entrance for activity.  Compared to the parent colony, bees exiting and entering Rosemary were relatively few and far between, as you can see in this video that was taken one week after we split out this colony.  Salvia on the far left of the hive stand was the parent colony, with Rosemary just to the right.

However, this weekend there were bees entering the Rosemary hive with pollen, which was an encouraging sign, and suggestive that brood may now be present.

We opened the hive, and were pleased to still see a modest number of bees in residence.  We’d witnessed a few small orientation flights in front of Rosemary soon after the split, as the capped brood we’d transferred hatched.  However, once all remaining brood hatches, orientation flights cease until new brood is produced by a new Queen.

The bees had clearly been consuming some of the honey stores we left with them at the time of the split, but were also storing fresh nectar.  However, the first brood frame we encountered, we found our first evidence of chilled brood.  This is why I said we under-split Salvia.  We obviously didn’t provide enough nurse bees at the outset to cover the amount of brood we transferred to Rosemary.

The arrows point to black, dessicated remains of larvae that failed to survive

This was an early season split, and our first split.  In hindsight, we should have shaken in more nurse bees to help cover the brood that was transferred during the split.  The parent colony has a laying Queen, and can easily produce more nurse bees as needed, but a Queenless split has to make do with whatever population the beekeeper gives them at the time of the split, either as adult bees, or soon-to-emerge capped brood.  Rosemary wouldn’t be able to produce any new bees on her own for at least 5-6 weeks (16 days to make the Queen + time for mating flight and egg laying + 20 days until the new workers hatch).  It’s a balance transferring in enough capped brood to see them through, and providing enough existing bees to care for the brood until they hatch.  With the Lavender split above, we were much more aggressive, to ensure there are enough bees to keep the entire brood nest warm.

We continued to work our way through the hive, trying to be efficient, and as minimally disruptive as possible.  A number of the remaining frames were empty, and we were getting quite concerned.  However, just as we were contemplating splitting Salvia for a second time, we finally found larvae!

It was a relief to finally find some larvae in this hive, approximately 7-8 days old

These larvae are approximately 7-8 days old, and if we reopened the hive today, this brood should now be capped.

Because these cells weren’t capped at the time of our inspection, it’s difficult to say for certain whether these larvae are destined to be drones or workers, and the distinction is important.

Capped worker brood is flat (yellow arrow); capped drone brood is domed (red arrow). Worker brood can only be produced by a laying Queen.

If it’s worker brood, that’s confirmation we have a laying mated Queen, as only she can produce worker brood.  If it’s drone brood, this colony may still be Queenless, and the brood being produced is the result of an unmated laying worker.

Clues the colony may have a laying worker are that laying workers tend to scatter brood through the hive haphazardly, not together in a central location.  The pattern in this hive was nice and even, so that was encouraging.

The larvae were clustered together, suggesting we likely have a laying Queen

We did find one anomaly though.  Multiple eggs in a single cell.

The lighting isn't ideal, but if you look closely (click to enlarge image), the arrow is pointing to a cell that contains TWO eggs

Multiple eggs in one cell can occur either due to a new Queen who is still learning the ropes, or a laying worker.  New Queens usually don’t lay more than two eggs per cell, but laying workers may lay 3 or more eggs per cell.

Another clue to a laying worker is they often lay eggs on the side wall of the cells, not on the cell floor.  This is because workers have shorter abdomens than Queens, which aren’t long enough to reach the cell floor.

Here however, we have only a single egg in a cell. All the eggs we observed were laid on the floor, not on the cell walls

If the colony has a laying worker, the only practical solution will be to shake the bees out, as a laying worker colony usually won’t accept a new Queen, even if one is introduced.

New Queens laying extra eggs in a cell usually stop in a few days with a little experience.   We were a little surprised to see day 8 larvae, and still find two eggs being laid in a cell, but a few laying errors can still be forgiven at this early stage, so we’re not too concerned. We’ll have to wait and see how the brood looks next time.

Next time we check Rosemary, we hope to see brood that looks like this

Our instinct is that these larvae will prove to be workers, not drones.  We noted the frame position of the brood so we can quickly recheck the hive next weekend.  We’ll be looking for smooth flat cappings over the cells, as confirmation that we do indeed have a laying Queen.  Fingers crossed!


Although we lost two colonies over winter, our apiary is now back to four colonies.  We’re crossing our fingers that Rosemary will prove to have a well-mated Queen, and that Chamomile will soon be a thriving colony.  In the meantime, if you keep bees, the second annual National Winter Loss and Management Surveys are available between March 30th and April 20th, 2012.  Participation is entirely voluntary, but the data you provide about the health of your hives, and management practices, may help researchers to better understand winter colony losses, and help other beekeepers to better manage their own hives.

If you’re interested in the results of the 2010-2011 survey, they were recently published here. (PDF)