I tried to send an attachment, It would not let me. Email me if you want a word version with pictures.
Charles
Death to varroa
First off, let me start by apologizing in advance. I normally try to write interesting stories with a timeline that have a beginning, middle and end. This article is hard for me as it’s a collection of facts and random thoughts, all headed towards a goal. Please forgive the apparent randomness, and read carefully, as I attempt to present a lot of information in no particular order. If it seems disjointed, I warned you.
I have to thank Randy Oliver, a friend who has been working and writing on this regularly, for prompting me to write this. His work at finding some controls for varroa has encouraged me to share this publicly. While I offered it up as proprietary info, he is insistent that it’s shared with everyone, in the hopes someone else can use it and build on it.
As many of you know there has been a fairly new treatment concept floating around for a year or two now. The basis of the treatment is oxalic acid (OA) mixed with glycerin (Gly) and placed in the hive. I believe the original work came from South America, with them using a solution of 10 g of OA mixed with 20 mL of glycerin soaked into cardboard and placed in the brood nest. Randy has been working on it, and many others have been trialing variations of it for a while now.
Note: use of experimental methods on a hive requires a special permit and a lot of caution. My test have been done as bench test on mites alone.
So far the results have been mixed at best. The method holds promise, but we are missing some key elements. The current system is outlined well in Randy’s articles, but is basically OA/Gly soaked on blue shop towels and placed in the hive in the middle of the broodnest. While I will go into more detail, the main point of this article is not that method, but things I learned to help further it.
After several apparent failures and a whole lot of discussion on what was going on, I decided it was time to start applying some engineering principles to the problem. I think it’s important to understand a bit of the differences in science vs. engineering. Most science is about a hypothesis and testing it to see if it is valid. Engineering is different in that it takes what we know and tries to build on it, as opposed to understating what we don’t know better. My background is engineering, so for this one I decided to apply basic problem solving principles.
In Engineering we would look at the towel issue a bit differently. Successful problem solving is about identifying the problem. In this case I broke the towel issues down to two things--the pesticide (poison) or the delivery. The towel is both these things, and sometimes worked and sometimes doesn’t. So for my thinking the goal is twofold: 1) to determine if the poison is the issue or 2) or whether the problem was the delivery system. This question was based on having run several different trials that varied elements: some used more OA, or increased the water in the solution; others tried cardboard for the shop towels. Basically, whatever options may spring to mind, I would call it firefighting. Throwing things at it hoping one works. My goal was to refine that a bit.
Towards that I have to add I have been using vaporized OA for several years quite successfully. While it has its limitations, in my area the bees are broodless in November, so it is perfect for a super knockback of mites. Because of my faith in this treatment, I had done quite a bit of research into how oxalic actually works. Unfortunately, there is not much information out there. The basic concept is that 2 grams of OA vaporized in a hive leaves micro crystals spread throughout the hive, and covering the bees. These tiny crystals are the key to mite death. Many discussions have been had about the mode of action, and to date there are two theories that I have heard that actually make sense. Both were posted in discussions with some German researchers over the last couple of years.
1. the crystals somehow burn the feet of the mites and then they are unable to hold on to the bees
2. The mites absorb some of this acid thru the feet and die from that exposure.
Some other theories about OA accumulating in bee hemolymph or fat bodies have been explored and may also have some merit, but are unlikely as so far vaporization shows no OA level change in the hemolymph/fat body of brood. This is in contrast to the dribble method, which has been shown to elevate OA levels in bees.
I would like to add a comment here: we have assumed that varroa is a tick type creature feeding on hemolymph like a tick would, but it turns out mites are more like a spider, feeding on the fat bodies of the bee. One thing I did observe in the microscope that has been hard to catch in a photo is that some varroa are in fact quite engorged. A quick look in the microscope shows they are thicker and look a lot like an over inflated water balloon, all the seams stretched, this dissipates with time and after a day or so you don’t notice it. This engorgement does not appear to affect the mite’s ability to move or cling to bees at all, but may be a method of survival until a host is found, or may be from the host, to enable it to survive until it enters a capped cell. I am unsure when this engorgement takes place, but I speculate by color and hardness of the carapace of the mite that it is newly emerged mites with this feature.
So back to the problem at hand: There have been spotty results with OA on towels and I disagreed a bit with some of the experimental directions as they seemed completely random. So in talking to Randy the one thing that kept sticking out was the fact we didn't know how OA/Gly worked, ingestion, contact or anything else.
I got this bright idea to test some mites. The concept seemed simple. Catch some mites and expose them to OA for different amounts of time, and see the effects.
While this sounds easy, trying to capture enough mites for experiments proved trickier than I thought. Early season hives had been treated and mite counts were low. Running around trying to "capture phoretic mites” took several hours. I painstakingly searched bees looking for mites to capture.
The first round ended up being tricky. I had no real plan, just a goal to see how long the mites needed to be in contact and what happened to them. So after playing around a bit, I settled on a method. For those who have never captured a mite, they are very active, trying to pin one down for any length of time is about impossible. Right after capture they are super active and cling to anything touched. After much trial and effort I settled on Elmer’s non toxic glue. A tiny dab placed on the mite’s carapace and then allowed to semi dry; I could then touch them with the head of a pin, pick them up and flip them over. Once the glue has dried be very careful with the mite, since if the mite is allowed to touch anything, it will break loose from the glue cleanly.
Once flipped over I handed the mites OA crystals. Pure OA like I use in my vaporizer. The crystals are larger, and finer than a salt crystal, but relative to a mite’s legs, the crystals are huge. The visual equivalent is one crystal is like a basketball in humans hands. What’s interesting is when you hand one to a mite; she rolls around like basketball pro getting ready to make a slam dunk. The mite will roll the crystal around while she tries to "walk".
Mite Feet
I need to add some observations here: first off, mite feet. You may be aware bees have both tarsal claws as well as a sticky padlike arolium. This part of the foot can be selected on honeybees, they have a choice of which part they use, and they switch back and forth, which is why sometimes bees can walk on glass and other times not. The arolium is like a wet pad that adheres to a surface. It seems mites have not much of a claw but a proportionally large foot called the empodium.
Figure 1 Close-up of varroa empodium
I would say it’s as big as a bee’s, which when you realize a mite is much smaller, it’s out of proportion. It’s much like a big, wet floppy glove that sticks to things. The second interesting observation to me is that mites have eight legs and two pedipalps(mouth parts), but even though the front two legs are visually the same as the other six, they don’t seem to use them at all for mobility. Only the rear six were used to manipulate the OA crystals. I watched closely to be sure the mites did not touch the OA with their mouthparts.
The first trials with OA were set at 5 minute and 15 minute exposure times, 5 mites in each group and a control. So each mite was held on its back and allowed to handle a crystal OA for the allotted time period and then wait, it turns out that after handling OA the mites are all dead within about 6 hours, where as the control group survive about 30 hours or more.
I also tested the theory that they burned their feet by placing them back on bees. It seems the mites have no trouble holding on right up until death. So the theory of burned feet causing problems adhering to a bee is incorrect. There is a method of lethality, but it appears to be through absorption through the feet. It should be noted that the amount absorbed is not noticeable via a dissection microscope. I half expected to be able to see “footprints” or small crystals disappear, like a child with a sucker, but there is no visual indication of any change to the crystal visible, even with as long as 30 min exposure.
Figure 2 Varroa mite rolling an OA crystal with its feet. Notice the pin it’s glued to in the backdrop. The crystal is blurry as the mite never stops moving it.
This first trial was pretty interesting to me, as it turned out that there was no noticeable difference in the death rate and survival time between the 5 min group and the 15 min group. This leads me to the need for more experiment for shorter periods to see what the bottom line of exposure may actually be. But they real question is how does that relate to in hive exposure? As of yet, this remains a huge unknown and I will address a bit later. Bottom line is, how long do the mites actually contact the OA via feet? There is no apparent grooming behavior or avoidance indicating the mites are not “adverse” to the acid contact and visually doesn’t seem harmed by it. While I am not sure if they are even capable of any grooming, they also don’t seem to attempt to discard the crystals.
Figure 3 Shell of glue after mite is dislodged
In discussion of these results, Randy asked about OA contact with the carapace. I assumed that would be a non-issue, but tried it anyway, and as suspected it was a complete non-issue. No change in expected life span when dry OA was applied to the carapace. Dry crystals anywhere but the feet seem to be no problem for varroa. Randy informed afterwards that this is also what Papežíková observed.
Additional Experiments
The next step really was to dial down the time, but curiosity being what it is, I decided to switch to the OA/GLY solution and see how that worked out. My plan was to keep the exposure to the 5 min level and see how it compared in lethality. In due course mites and time was procured, and I ran into problem one. It’s practically impossible to create a droplet small enough for the mite to handle, or to limit their contact. If you try to touch a foot, in a millisecond they are completely engulfed in the droplet due to leg movement and surface tensions of the solution. So this presented a huge problem, the test to compare dry crystals to wet OA was not going to work.
Sometime while pondering how to overcome this problem, I decided to soak a towel and let the mite walk on it for 5 minutes. Turns out, it doesn’t work. No early deaths, so I upped the time to 15 min, and no noticeable change either. In thinking about it and looking close at the towel through the microscope, it actually makes sense. The towels absorb the OA into the fibers, and the mites are actually walking on top of the fibers, much like a spider on a web. The capillary attraction of the towel keeps the solution off the feet much like a doormat does in our homes, helping keep the mite and those feet up and away from the active ingredient.
The Winning lottery ticket
This problem led to more thought and conversation, and the plan to test the solution on the carapace. I have to admit, I was not real keen on this test as I assumed it was a complete waste of time. The mite’s carapace looks much like that of crabs, and seems to be an impenetrable protective shell, but at this point I didn’t have much of a new plan.
A new batch of mites, which is no easy task, but getting easier with practice, and I was on my way. As it turns out, it’s actually pretty simple to place a microdot of the OA/Gly on the carapace. No need to restrain the mites at all just set the timer and wait. Turns out they are also pretty easy to clean off, just touch with a piece of paper towel and tada! Pretty darn dry. Within 30 minutes a batch of 5 mites were done, but here was the winning lottery ticket. As I finished up the last one, I took a look back at the first mite I had treated. Within 30 minutes the mites are pretty well paralyzed. They sit up on their back legs and tremble. This is quite different than anything I observed with foot contact (I did go back and rerun that test to see if I missed it) and within the same 6 hour time frame, the mites treated with OA/Gly were all dead. While the end result was the same, the partial apparent paralysis was astounding to say the least.
Within about 30 minutes mites treated with OA/Gly on the carapace for as little as 5 minutes were effectively dead. The significance of this is astounding. While with feet I questioned how long it would be on them, I am reasonably confident that the solution anywhere on the body would be there for a while.
A note from Randy: A German researcher, Saskia Schneider, also found that OA in glycerin adheres to bees’ bodies for many days, thus exposing varroa to extended exposure. The glycerin acts as a humectant, which apparently increases the penetration of OA through the mite’s cuticle. Charlie’s findings suggest that the simple act of a treated bee drawing its contaminated leg across the carapace of a mite might be enough to expose it to a lethal dose. This explains a plausible mechanism of delivery from the towels to the mites
Mites grooming skills are fairly nonexistent; they are completely unable to reach their carapace to clean it, meaning it’s going to be there a while. Now I am comfortable with that 5-min exposure.
Figure 4 Droplet of OA/Gly on mite
Experimental Controls
I need to divert here a bit and describe something I know I don’t have the proper words for. In this experiment I always had controls of non-exposure, but with the last trial I added glycerin only as an additional control. There is some lethality in mites treated with GLY only, but without the paralysis. Mortality with 5 min exposure was around 40% in the same 6 hour window, so significantly less than with the OA/Gly combo. The sample size on the control lots was 10 mites so this test needs to be rerun with a larger group.
In this I observed that the glycerin alone cleaned off the carapace completely, BUT the OA/Gly solution seems to have some surfactant value associated with it. While it does clean off, it leaves a small residue, much like drying a wet dog. The majority is good, but you can observe that slightly wet clump here and there. It may be from the acid etching the carapace and hairs on the mite’s back or some other detail I can’t yet explain, but removal is not 100%. I would say 95% as an estimate, but I lack the skills and tools to define or refine it. I do realize glycerin is not known as a surfactant, but the OA/Gly mix seems to have less surface tension and the ability to make smaller droplets.
Figure 5 Mite with OA/Gly residue
Back to the Drawing Board
At this point I was too astounded, and had to regroup my thinking a bit. So I headed back to the thought processes so far. Everything I had seen about OA/Gly trials seemed to be attempting to jam in more OA and so raise the level of OA in the towels. At this point it seemed, based on what I just learned, it was intuitively obvious to me, and this was the wrong direction. Dry OA is harder to move around the hive, and increasing the ratio made the towels harder to work with.
A note from Randy: I found that increasing the ratio made them easier to work with. But the main reason is that in my preliminary experiments, bees appeared to be repelled by high-glycerin towels. I reduced the amount of glycerin in order to get the bees to touch the towels.
The total amount of oxalic applied to the hive also appeared to make a difference—which is why we went from a single towel to 1.5 towels per dose.
It seemed to me it was critical to determine if it was the dose or the application method that’s giving us inconsistent results. So while it’s not chemically correct, I decided to play a bit with what I had. I wondered at what level the acidity of the solution started to change. OA in itself is a very strong acid, so I took the 80% solution I had been using and tested the pH level. I reduced the OA levels and tried some new tests. It turns out that you can drop to a 5% OA solution before you start to see much of a change in the pH level. It is a crude indication that the 5% solution was as acidic as the 80% solution. As it turns out, it seems the 5% solution works just fine. I know from a chemistry point this is not related to the saturation level, a bit more work in this area is warranted.
Even at the 5% OA level the mites are paralyzed in about 30 minutes, and dead in the same 6-hour time frame!
The implications of this are actually pretty cool. It means we can refocus a bit. It’s seemingly not about how many grams of OA are in the towels, but how we get the solution onto the mites. While increasing the volume may have an effect, it seems to me, fussing about the ratio seems to have very little real effect. I believe this will allow is to focus on delivery methods and not worry so much about the OA levels.
Mode of Delivery
The next question in my head is delivery, so I set out to track how this solution moves around the hive. I ordered some UV fingerprint dye, and mixed up a 20% solution (no science, just a stab) and soaked a towel in it. The towel was placed on a hive with the goal of tracking the dye thru the hive. I must caution you; this is not as easy as it sounds. It turns out honey and pollen glow under UV light, and quite a wide spectrum. Not as easy as picking out one color that’s different. As of right now I don’t have anything to share, it turned cold and I dint learn anything valuable. Randy mentioned that other researchers are using different tracers and CAT scans to explore this also.
I did in this experiment collect some mites that seem to have died from their exposure to the towel. I have to say the image is quite startling. There are some startling differences in the appearance of dead mites from a treated vs. an untreated hive. This is very preliminary and turns out to be really hard to photograph. It seems that digital images process UV differently than human eyes, so images that were super cool and surreal in the microscope and in the hive, did not turn out and photograph well enough to share. I will keep working on that. If you have any tricks on UV digital photos I would love to learn them.
At this point the need to treat for mites and winter weather has set in. I have had to suspend testing for the time being due to cold and trouble procuring mites. I am very eager to start up again come spring!
Mite collection
For these experiments I played around with many methods of mite collection. It’s harder than one would think to collect 30 live healthy mites at one time. First off I had to wait until later in the season when counts were climbing, a mite level of 1 or 2 per 300 bees takes forever to collect enough mites. I ended up waiting until mid September to find a few with really high counts that had not been treated.
My first attempts were visual. Taking frames of bees and watching for quite a while to observe that phoretic mite on a bee, then placing that bee in a collection jar. As it turns out, spotting mites is akin to seeing deer in the woods. Even with cheater glasses for grafting they are hard to spot. They are there, but seldom do we actually see them. Even hives with super high numbers do not produce a lot of visually obvious mites. The second attempt was thru a friend, he had purchased a new Swienty unit that uses CO2 to knock out the bees and the mites, and a perforated disk below the bees for mites to fall thru and capture. The system itself was fairly slick it did a fair job of knocking out the bees and mites, although mites seemed drunk from the CO2 they were not out long. The big problem was the CO2 gun was it didn’t seal well and uses a special screw neck cartridge. Unfortunately the cartridge ended up being a one and done deal as it was leaked empty by the next day. Cartridges were unavailable locally so this plan was abandoned.
The last method was to use the Swienty shaker with powdered sugar. This method seemed to work much better. A few spoonfuls of powdered sugar and allow the bees to frolic in it over and over and a fair mite drop was achieved. Mites in powdered sugar are seemingly stuck in quicksand. They are unable to move around much but try like crazy. It was fairly simple to find them and place them on a towel for cleaning.
Figure 6 new batch of mites pulled from the sugar awaiting cleaning
It sounds simple to clean sugar off a mite, but it’s not as easy as it looks. My first trial was with a wet paper towel. I figured the water would absorb the sugar quickly. Turns out mites can and do drown very fast on a wet paper towel one batch of mites completely wasted. I need to add here that I met with a representative from a Finnish company in CA this fall, and he showed images of them rinsing mites under a sprinkler head in a really fine honey sieve of the metal type. I have not tried it yet, but he swore it worked fine. Randy mentioned that this method has a high premature mortality rate via other researchers.
In the end I found just letting them walk on a towel and then flipping them over a few times cleaned them off well enough to work for my test.
Figure 7 a mite fresh from the sugar
What’s next?
The $64,000 question is where do we go from here? As an Engineer, my thought process is to be able to quantify how much OA the mites are actually getting in then via contact methods. It seems to me with our capabilities we should be able to determine PPM or some measurement. If we can do that then it seems to me we can start to seriously evaluate the different delivery methods without long drawn out trials. If we can measure PPM of exposure in a given time frame, and compare one method against the other in a quicker time frame, that should allow us to dial into the best delivery method. Basic problem solving requires a measurement system. This is something that’s been lacking as far as I see it. If we can master this we can evaluate several delivery changes with just small samples instead of waiting till seasons end to determine success.
So far I have not found a lab that is able to help me determine PPM, any suggestions would be appreciated please feel free to email me.
As far as experiments, I have contacted some friends who are also doing towel trials; this next season will see us testing lower OA levels percentage wise head to head with the standard 80% levels to see if there is a difference. The goal is field confirmation of if it’s the delivery or the dosage that’s in need of fine tuning.
While I suspect that lower OA levels will make it easier to focus, there is also a possible downside. Right now we suspect chewing and removal is at least a small portion of the delivery, lower levels may reduce the taste enough that bees actually ingest more OA directly which could be bad. I have been advised that bees do not like the taste of glycerin, and they cannot be forced to eat it even directly applied, but It is a big unknown to me if the level of acid to glycerin ration may affect they process.
I will also be doing a lot more work with UV dyes. Hopefully I or others will be able to get a better handle on how and why the OA/Gly solution makes its way around the hive and onto varroa, which after all is the key to future success.
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
|