I was weeding the mule garden about 15 feet away from the hive when the bees decided to swarm, but I missed the mass flight.  I had walked off to the barn to get another tool, and when I returned I heard a great buzzing overhead and to my left.  By the time I got my eyes turned around, the bees were out of sight.

Luckily, they hadn't flown far.  In fact, they'd settled into the honeysuckle on the exact same sassafras last year's absconded package settled on before flying away.  That means the bees would have been right beside our top-bar-hive-turned-swarm-trap...but I had talked Mark into helping me move that structure behind the barn the day before.  Oops.

We were woefully unprepared for a swarm, despite the fact that I've been blogging about little else all week.  I really should have known it was coming, not just because this is the swarming time of year, but also because the Warre hive's three boxes were getting pretty congested.

I took the top two photos up through the screened bottom the morning before the swarm left, catching the back (left) and front (right) of the hive, and it was evident that the cluster of bees was humongous at that time.  That's why the second photo is out of focus --- the bees were hanging all the way down to the floor.  The hive had been a bit like this (although not quite so extremely crowded) for a couple of weeks, and if I'd wanted to slow down swarming behavior, that would have been a good time to nadir a fourth body into my hive. 

However, I didn't have the spare equipment on hand because the place I ordered from is swamped and hadn't mailed my hive bodies two weeks after I paid for them.  (I haven't figured out how to make top bars yet, so I couldn't just make my own homemade hives.)  We do have a fourth hive body, but I'd been saving that box in case the package shows up before the new hive parts do.  Which is all a long explanation for why I let the bees get so congested that they swarmed before they need to have.  All of that said, Warre theory is that you want your bees to swarm to break the pest and disease cycle, so all I would have been doing with management is slowing down the swarm, not preventing it.

(As a side note, it's interesting to compare the inside of the original hive the morning before (top photos) and the afternoon after the swarm (bottom photos).  The comparison gives you an idea of how much of the hive's population leaves with the swarm and how much stays home to tend the worker brood and baby queens.  Generally, about 10,000 workers and drones (two-thirds of the population) leave, but they're soon replaced by workers hatching out in the old hive.)

Anyhow, mea culpa's aside, I had a swarm on my hands and I didn't want to lose it.  My first hope was that the bees would find my swarm trap even though the structure was now a couple of hundred feet away, but I soon discovered that the bees were considering another site instead.  How could I tell?  Well, I've spent the last week reading Honeybee Democracy (which I'll review here after finishing the last couple of chapters), and thus had discovered a lot about the behavior of bees in swarms.

It turns out that after a swarm leaves the hive, they always settle just like mine did, not far from the parent hive.  At that point, an astonishing exercise in consensus decision-making occurs.  Most of the bees hang out around the queen, but 300 to 500 workers peel off to survey the surrounding 30 square miles in search of nest sites.  If a worker finds a potential home, she returns to the swarm and performs a waggle dance (shown below) to inform her cohorts of the location.  Other bees go and check it out, dancing about the site if they like it or ignoring it if they don't, and eventually enough bees have settled on one location that the swarm figures they've found what they're looking for.  At that point, they all take off and relocate to their new home.

This all sounds a bit esoteric, but the truth is that once you have a swarm in front of you, waggle dances are extremely obvious and easy to decipher.  After only a few seconds of observation, I was able to find two workers dancing about the same location, which I videoed above.  The dance's angle away from vertical shows the angle the actual location is away from the sun, and the length of time the bee wiggles her butt shows how far away the site is.  Using this data, I was able to tell that our bees had only found one good site so far, and that it was two miles away to the southwest --- definitely not my swarm trap.

My first thought was to set up a potential hive on the other side of the yard, smear it with lemon balm, and hope my bees would find the site.  Then I changed my mind and figured I'd set up the potential hive much closer to the swarm tree to ensure the bees found it.  But half an hour of observation showed that only one worker checked out the empty hive, and she didn't even think it was worth crawling inside.  (Scouts looking for potential house sites go inside and measure the cavity volume to ensure the hive is what they're looking for.  When visiting a good site, they'll spend over half an hour buzzing around to check out the cavity from all directions.)

If we were going to lose the swarm otherwise, it seemed worthwhile to try to catch it.  Huckleberry told me to lay out a white sheet underneath the swarm and then cut down the tree to drive the swarm into a box (although I think our spoiled cat might have mostly been thinking about napping in the shade), but I got scared and called my beekeeping mentor (aka our movie-star neighbor).  More on the ensuing action in tomorrow's post....