Thursday, September 10, 2015


Earlier this summer, we noticed our new package of bees wasn't building up as quickly as it should, and the brood pattern wasn't full and solid like it would be in a healthy colony.  We started looking closer at the combs, and noticed a few things.  Most of the uncapped cells all had eggs or larvae in them, so the queen was laying fine.  But many of the uncapped larvae didn't look quite right--some were sort of discolored and laying at the back of the cells; others started to extend out toward the front of the cells, but were slightly twisted, as if they had a stomach ache. In short, a lot of our brood frames looked like these and these.  Sounds like trouble.  Not trusting our limited beekeeping experience to officially diagnose the problem, we decided to send a comb sample to the USDA bee lab, where they test it for free and tell you what disease your bees have (if any).  A few weeks later, the results came back: our bees had European Foulbrood (EFB).  Oh no!

Here's the old brood comb, with the sample we cut out to send to the Bee Lab.  Although they only ask for a 2" x 2" square, we cut out a section to fit a small USPS box instead.  Don't want them to wish they had more to test!

So what are our options for treatment?  While EFB is no walk in the park to get rid of, it's not quite as bad as its evil sibling American Foulbrood (AFB) in that it doesn't form spores.  For AFB, the only treatment is to kill the bees and burn them along with the frames and comb (hive boxes and other equipment the bees have come into contact with can sometimes be sterilized by high heat or bleach).  For EFB, there are a few more options (and here):
  1. Do nothing, since in a strong hive, symptoms will often clear up when the nectar flow is strong.
  2. Treat with the antibiotic Oxytetracycline (trade name Terramycin).
  3. Re-queen, since a disruption in the brood cycle will often clear up symptoms.
  4. Use the "shook swarm" method to restart the colony on uninfected combs in a new hive (this also disrupts the brood cycle).
  5. Kill the bees and burn or sterilize everything, same as for treating AFB.
Of these options, we had already been trying #1 for a while, and it clearly wasn't clearing up.  We're trying to minimize chemical treatments of our bees, so #2 wasn't a great option, either.  #3 would be ok, but it seems like we could get the same effect by #4, which was preferable to us since our queen seemed to be laying well and we already had everything we needed.  #5 is obviously a last resort, and we weren't there yet!

So, shook swarm it is!

The general operation is simple, and shown here: we're just taking all the bees from the old hive and shaking them into the new hive.  A few tricks that aren't shown in the video: the new hive should have a queen excluder on the bottom (actually functioning as a queen includer) so that she doesn't have the opportunity to decide she doesn't like the new digs. Also, the chances are good that she'll end up in the new hive if we're efficient in our shaking operation, but we can increase the odds by spraying the frames down with sugar water to reduce flying bees before shaking them, or if we wanted to be really sure, we could find the queen and catch her, and then install her in the new hive once the rest of the bees are in there.

Also, in our case, since we're trying to clear up EFB, the new hive should just have foundation and not already-drawn comb so that the bees have a chance to sort of purge their system before they have to feed new brood. Some sources recommend holding off on feeding them for a few days for the same reason.

After a couple weeks in the new hive, they've started drawing out comb and generally looking healthier.  Can you find the queen in this picture?  Her name is Waldo.

If we blow a little smoke on them to clear the bees away, we can see that the brood pattern looks a lot more full than it did in the old hive.  Time will tell if they can build up enough to make it through the winter, but they've got a better chance now than they did before!
Stay tuned to find out what happened to all the old frames, comb, and the honey they had stored.

Have you dealt with EFB before?  Have you done the shook swarm method with your bees?  How did it go?  Let us know in the comments section below!

Thursday, September 3, 2015

What Selenium?

A couple months ago, we noted that our soil had non-alarming levels of all the heavy metals tested for, except selenium.  The selenium level was 40x higher than background for our area, and possibly in the range that we would want to do something about it.  So we did some more digging (ha!) to verify if we did indeed have such high levels of selenium, find out if it is dangerous, and find out where it came from.

First, verification.  We sent soil samples to a second lab (more local, and cheaper for single-element heavy metal testing).  One sample taken with our fancy soil sampler tool, the other with a standard spade-dug hole from areas we knew we wanted to dig up.  The results came back: < 1 ppm selenium in both samples!


That was unexpected!  But, it seems we have a bona fide controversy!  Get some popcorn, ye lovers of analytical chemistry-themed drama!

We should preface this discussion by saying that, since it wasn't exactly the same sample analyzed by both labs, it's possible that we hit a pocket of high-selenium soil in the first sample and missed it in the second sample.  That is, both labs could technically be correct.  But...we think the 30 ppm number is erroneously high, as we'll explain below.

On the analytical side of things, the two labs used slightly different techniques.  Both labs digested the sample using nitric acid and hydrogen peroxide (to render it soluble and fully oxidize the selenium), and fed it into an inductively-coupled plasma (ICP).  But the output of the plasma was analyzed by atomic emission spectroscopy (AES) at the first lab, and by mass spectrometry (MS) at the second lab.  Both techniques suffer from plenty of spectral interferences (see here and here) in addition to the instrument-specific quirks that crop up with any piece of analytical equipment. There are ways of compensating for the spectral interferences, however, none of the known interferences are supposed to be able to increase instrument response from < 1 ppm to ~30 ppm.

Of the two labs, the second reported extensive quality control data, including blank runs (to make sure that no selenium is detected in a sample that's not supposed to have any), multilevel sample matrix spikes (where they add a known amount of selenium to our dirt and make sure the level increases by the amount they added), and standard control runs (where they run other materials with known selenium content to make sure selenium is detected at the expected level).  The first lab did not, but responded by email to say that they run two standard control samples per of which came back almost three times as high as it should have, and said that other samples in the same batch also showed elevated selenium levels.    But if it were random jumps in apparent selenium, it's weird that both of our samples were so close to each other.  Did the other samples show up to 30 ppm selenium? We don't know. We asked enough questions that the first lab finally said to just send them another sample and they'd run it for free (not including shipping, of course).  We haven't taken them up on the offer yet.  In any case, we were leaning toward higher confidence in the second lab's (low selenium) results anyway.

Nevertheless, if the first lab was wrong, we wanted to know why.  So, we took to the interwebs to search for possible interferences from other components in our dirt that might not have been accounted for .  The first lab mentioned that they used the 203.985 nm emission line to measure the selenium, and we found an awesome tool from NIST that conveniently allows one to see what other elements might emit in the same range.  The general fertility test for our soil said we had really high levels of Mg, Ca, and P, so we started there.  But Ca has nothing in the right wavelength range, and Mg and P aren't supposed to affect that the selenium measurement at that wavelength very much.  According to the first lab, molybdenum is the only known interference at that wavelength, but we couldn't find anywhere that confirmed that, or to what degree it interferes.

What the ICP-AES spectrum of our soil might look like in a perfect world, with equal concentrations of Mo, Se, and Ca, an instrument resolution of 0.035 nm, and no bcvkground noise.  If anyone wants to buy us an actual ICP, we'll gladly replace this figure with a real graph!

Unfortunately, there aren't a lot of qualitative biological indicators that could help us differentiate 30 ppm selenium from 1 ppm selenium, either.  Sometimes selenium hyperaccumulators, such as some species of Astragalus and Stanleya, can be used to gauge if a soil is high in selenium, but our yard didn't have any of the hyperaccumulator species growing when we moved in, and selenium toxicity in most plants doesn't usually present itself until much higher selenium levels than 30 ppm. (Unfortunately, the numbers cited for selenium toxicity in wheat and peas in that article don't actually appear in the paper it cited!)  Similarly, we couldn't find any data suggesting that we'd be able to notice defects in our soil invertebrates or other wildlife at 30 ppm.

Second, would it be dangerous anyway, even if we were at 30 ppm?  The first lab said possibly, quoting an EPA document that says 20 ppm is the threshold level at which they start to dig deeper into things like bioavailability (which depends on soil pH, soil sulfate content, the form of selenium in the original source, and other factors).  On the other hand, A&L Eastern Labs says not to worry about concentrations less than 50 ppm.  What about a local office?  We emailed the Colorado extension service, who said that the main risks in this area are forage plants that hyperaccumulate selenium, leading to toxicity in grazing livestock, but not normally vegetables grown for human consumption.  If it were accumulated to dangerous levels in our veggies, we would expect to see symptoms like brittle fingernails and hair, and we haven't yet.

Third, where could it have come from? Most soil selenium comes from weathering seleniferous rocks, volcanic eruptions, coal burning, and metal refining.  We wondered if it might have been from spilled chicken feed, since we were dragging our chicken tractor around the yard with our extremely messy broilers last summer, and the chicken feed they spilled was fortified with selenium.  However, it turns out the FDA only allows 0.3 ppm selenium in poultry feed, which means that, unless our feed is mixed by scofflaws, we'd have to have our soil essentially made out of composted chicken feed with very little selenium transport for that to be the main contributor.  The broilers were messy, but not messy enough to build up 8" of soil throughout the yard from spilled feed!

Final conclusions? The 30 ppm was a false positive.  Also, that was a lot of research hours spent to decide there's nothing to worry about.

Time to get back to the garden!

The garden in August: some late tomato blight, some powdery mildew, lots of raccoon, squirrel, and chicken damage, but no dangerous levels of selenium!