31 Doug Sponsler – The Risk of Pesticides to Honey Bees and Ecotoxicology

Speaker 1: From the Oregon State University Extension Service, this is Pollination, a podcast that tells the stories of researchers, land managers, and concerned citizens making bold strides to improve the health of pollinators. I'm your host, Dr. Adoni Melopoulos, assistant professor in pollinator health in the Department of Horticulture. Everybody knows that there are a number of pesticides out there that are really toxic to pollinating insects such as bees.

But just because something's toxic, does it mean it's going to have an impact on honeybee colony stocks or native pollinator communities? Well, to get at that question, I've invited Dr. Doug Sponsor to join us. Dr. Sponsor is currently a postdoctoral scholar at Penn State University Center for Pollinator Research. He's a native of Pennsylvania, but he did a lot of his work and his Ph.D., really exciting work at Ohio State University with Dr. Lee Johnson. Now what I really love about his work is he really thinks deeply about this area of ecotoxicology, about the actual mechanisms by which toxic pesticides might reach a pollinating insect and have an impact.

If you're like me and you scratch your head thinking about all the pieces that go into pesticide risk assessment and pollinators, this episode's for you. I hope you enjoy the show. For a few months now, I've been really looking forward to having a conversation with Dr. Doug Sponsor, and here's my chance. Welcome to pollination, Doug. Thanks, Antonia.

It's great to be here. Doug, I always get this question when I'm talking to pesticide applicators, but also to beekeepers. They want to know, using a specific pesticide to control a specific pest, how risky it is to honeybees.

I know there's this very complicated risk assessment that I'm hoping we'll get to talk about more in this episode, but I just want to know in broad strokes, what determines how risky a pesticide is to pollinators?

Speaker 2: The formal definition of risk used by regulatory agencies is the probability of adverse outcomes. It's something that's a question of likelihood. The adverse outcome is whatever you happen to be trying to avoid. It could be the death of individual bees or it could be something like pollination services, crop yield, maintaining biodiversity, and things like that.

It can be anything. But risk is the probability of that adverse outcome happening. And then the way that it's formalized in a quasi-mathematical way is that risk is the product of toxicity or hazard and exposure.

So hazard and toxicity can be thought of as basically interchangeable. It's the intrinsic capacity of a compound to cause an adverse outcome. And then exposure is the degree to which the receptor organism, in this case a bee, is actually exposed to that compound. And so it's the overlap between the hazard, the intrinsic toxicity of a compound, and exposure that creates the probability for adverse outcome, which is risk. So when it comes to the language that you see on the label, that language generally refers mainly to laboratory toxicity tests that really focus on that hazard component of risk, not so much on the exposure component. These are measures of acute lethal toxicity.

So acute means we're talking about short time frames, usually 24 to 48 hours. And lethal means that our endpoint is mortality, not some sub-lethal endpoint like navigational impairment or something like that. So this is a very narrow slice of what it means to be toxic, but it's the one that creates the most easily measured endpoint. Thus a compound that has very high toxicity was likely to be accompanied by stronger warning language on a label with respect to bees than something that has low toxicity to bees.

Speaker 1: Okay, super. So risk is a lot more than just toxicity. It has this element of exposure. Can you tell us a little bit more about what assumptions go into risk assessment with respect to exposure and why are we so focused on toxicity? Why doesn't exposure take a more predominant role?

Speaker 2: Well, the answer to that question is that toxicity is relatively easy and exposure is very, very hard. So the way that this is dealt with in a regulatory setting is with a tiered system of risk assessment. And so what this means is that there are so-called tier-one studies. So during the registration of a product, these tier-one studies have to be done and they're also called screening-level studies sometimes. And these are purely laboratory studies, usually just of acute lethal toxicity.

So the way these work, I've had a little involvement in this sort of thing. You take a cup of bees and you anesthetize them with CO2 or with cold. And then you take your pesticide and you put a little drop of it on each of those bees and you let them wake up and you leave them in their cup. And then you come back six hours later and count how many are dead.

And you come back 12 hours later and count how many are dead and 24 hours 36 hours, 48 hours, and so on. And that gives you what's called an LD50 for a certain timeframe. What LD50 stands for is a lethal dose that kills 50% of a treated population.

So if you treat 10 bees, the LD50 is the dose that results in five of those bees on average dying for whatever your timeframe is, say it's 24, 48 hours. So exposure is being completely controlled here. So we're taking exposure out of the equation. Now, if a compound demonstrates a concerning level of toxicity at these low-tier studies, then you move on to higher-tier studies. And the next step might be something like a flight cage study. Well, you have these foraging in a cage, usually with some attractive crop, say like oil seed rape that you treat in some controlled way with your pesticide in question. And then you let those bees actually forage on that plant, but forcing them to forage on the plant because they're flying in a cage. And so this incorporates some of the complexities of exposure, like, for example, the degree to which a pesticide applied to a plant gets incorporated into the nectar or into the pollen, or remains on the surface of the plant and is a risk for contact toxicity as bees land on a plant.

So it incorporates some of that mechanism. But of course, you're still forcing the bees to forage on that plant immediately after application. Then the highest tier studies are we actually put colonies out in the field, and let them forage freely, usually in close proximity to a treated crop. But in these situations, you can't force them to forage on the crop. So it simulates more closely what would happen if this were just a compound that was used by growers in the field in places where bees were foraging.

And as you go up these tiers, you get more and more realism, especially with respect to exposure. Because in that final case, you have the heterogeneity that occurs with bees foraging both on and off of a treated crop. And then you have the colony level complications that occur as the toxicants that bees encounter in the field are then processed and experienced by members of the colony, and maybe results or maybe don't result in effects like reduced honey storage or brood rearing or queen longevity or things like that. But as you go up those levels of complexity, it gets more and more expensive to perform, more and more logistically challenging to set up, and more and more difficult to interpret because you can't do a very statistically powerful study at a full field scale where you have lots and lots of replication because it's just logistically and financially impossible.

And so then the question is, if you don't see a result, does that actually mean that risk is negligible or does it just mean that your design failed to detect it? And then these create very, very controversial situations.

Speaker 1: Okay, so field experiments are real expensive and they're open to controversy. So there seems to be a real importance in being able to understand and model exposure. Now, we were talking a little bit earlier in the break that there are actually models of exposure that are used by regulatory agencies. Can you walk us through kind of what's entailed in those models? What are some of the assumptions?

Speaker 2: So the simplest way to model exposure for a Tier 1 level is to take some factor relating application rate to exposure and then compare that predicted exposure to the toxicity to see if that level of exposure is likely to cause an adverse outcome in the organism that is exposed to it. Traditional versions of this approach have been fairly simplistic relying on just a simple conversion factor from older datasets. Recently, the EPA has developed a more sophisticated model called the B-Rex model.

So B-E-E-R-E-X. It's a derivation of an older model called the T-Rex model, which stands for the terrestrial exposure model. Not the dinosaur. So the way the B-Rex model works, no, not the dinosaur, though I imagine a T-Rex could have an adverse, would in itself be an adverse outcome in those cases. The way the B-Rex model works is that it takes the application rate of your pesticides, so in something like kilograms per hectare or whatever, and it uses a body of empirical data for the rate at which applied pesticides get incorporated into plant tissues. And it uses those as a proxy for the rate at which they would get incorporated specifically nectar and pollen. So this is a leaf tissue dataset that is being used as a proxy for nectar and pollen that these would be exposed to. So as a protective proxy, what the agency chose was to use data for tall grasses, I believe, as being a... So by protective, I mean that you predict that the rate of incorporation into tall grass would be greater than the rate of incorporation into nectar and pollen. So in that sense, it's a conservative estimate. But using that as a proxy, you can convert an application rate into the expected concentration of pesticide in nectar and pollen, and then using data on the rate of consumption for different ages and casts of bees, you can turn that concentration in nectar and pollen into a dose B, and then that enables you to interpret that with respect to the LD50 data that you have from the laboratory to say whether that predicted dose would be expected to cause an adverse outcome.

Speaker 1: So what's wrong with that? And what's wrong with, for example, thinking about some of these T or 2 studies where you might provide a colony level dose, why are these not sufficient? Why can't we just use these current models? What are their limitations?

Speaker 2: So I want to be careful not to... I don't want to be critical much of things like the B-rex model, because I think they definitely have a place in regulatory decision support where you need a simple answer that will result in a yes or no to a decision about registering a pesticide, or about articulating the language of a label. But there's another side to this where we actually want to understand mechanistically what's happening. And B-rex is probably a pretty good risk assessment model, but it's not a mechanistic model. It does not get at the actual behavioral and chemical fate mechanics of exposure that go into actually bringing a B into the intersection with a pesticide. Okay.

Speaker 1: So in some ways, the mechanism isn't really well specified in these models. So if I'm thinking about this grower who's applying a pesticide in a field, how does the current model help us understand what the actual exposure that's taking place in that field is?

Speaker 2: Well, for example, in the B-rex model, so in all these risk assessment models, the goal is not to actually predict what bees will be exposed to in a field. The goal is to evaluate the risk associated with bees foraging on the treated crop. And so if we're interested in actually understanding how exposure happens, we have to acknowledge that the contamination in the environment is a heterogeneous thing. It's extraordinarily complex because you have both treated and untreated areas. And then even within treated areas, you have an uneven distribution of pesticide. So you have imagined that through time, this thing is changing as compounds move, as they degrade. And so you have these fluctuating spatial patterns of contamination.

One author is, I think, very aptly called a dynamic hazard surface. So you have this, imagine this fluctuating mosaic of contamination in the environment. And then you have bees moving, foraging, behaving in this environment, also in spatio-temporally complex ways. And so you do not have a simple situation of bees foraging on a uniformly treated crop and getting a uniform dose of pesticide.

Instead, you have a distributional exposure pattern. You have, especially with social bees, like honeybees, you have thousands of foragers moving through this complex environment, getting thousands of individual doses that are probably unique to those individual bees and then bringing them back as this broad distribution of levels of exposure. And that's something very different than just a worst-case scenario model of bees foraging solely on a treated crop. And so in that sense, these models are conservative, overly conservative, with respect to what's actually happening in the field. But there's another sense in which you can be under-conservative by failing to account for this distributional nature of exposure. Because if, for example, you measure pesticide residues in pollen or nectar or wax sampled from a colony, so you take a scoop of pollen from a pollen trap and you grind it up and you analyze it for pesticide residues, you're not getting that distribution. You're getting the average of that distribution.

So an example I like to give when I'm giving a talk is I ask people if I were passing around a box of chocolates through this room, and I assured you all that the average concentration of cyanide in these chocolates was quite safe for human consumption, would that make you feel safe to eat them? The answer, of course, is no, because knowing an average tells you nothing about the actual distribution of levels in there. It could be that all of them are completely clean and one of them has an absolutely lethal dose. In fact, that's what you'd expect to see many times in the field when bees are foraging because what shows up in the colony is going to be a combination of foraging on treated and untreated crops. And so the average that you get might not even exist in the field, which is a curious problem. If your prediction of exposure is literally impossible for an individual bee to experience, then it is a rather dubious utility for risk assessment. So instead, I think we need to fully appreciate this distributional nature of exposure and talk about exposure in terms of distributions rather than discrete values. Terrific.

Speaker 1: Well, let's take a break now. I don't want to come back and talk about some recent work that you've done with Dr. Reed Johnson sort of thinking about how all of this plays out in the context of a highly social bee, the honey bee, and how these models may have yet another layer of complexity when you enter the hive.

Okay, welcome back. Now, in the first part of the episode, we talked about how our capacity to predict the risk associated with a pesticide application to bees is really hampered by our lack of a mechanistic understanding of the determinants of exposure. I want to turn our attention to inside the colony. You've done some really interesting work with Dr. Reed Johnson kind of thinking about those returning bees and the pesticides they have and what when they come into this highly social environment in a honey bee colony, you characterize this as, you know, honey bees having in some sense a shock absorber against pesticide exposure. Can you explain what you mean by a shock absorber?

Speaker 2: Yeah, so the honey bee, it's a bit ironic that the most convenient organism for which to study toxicology from a logistical perspective is also the most problematic one for which to interpret toxicology. So the honey bee colony is a complex society. It consists of different behavioral casts, different ages, and specialization of labor.

It consists both of activities inside and outside the colony. And it's been compared to a multicellular organism. So this concept of the superorganism where you have closely connected individuals in a social colony are like the cells of a multicellular body. In the same way that being a multicellular organism enables your whole body to buffer effects on individual cells in your body, being a social colony buffers you against perturbations from the outside. So for example, if I don't eat for too long, my body can change its physiology so it starts to break down glycogen into glucose and enable metabolism to keep going. Similarly, if I am eating a lot, then it can start to store that excess glucose as glycogen.

Or if I am sick, I can be prompted to rest. So there are ways in which an organism can compensate for challenges that an individual cell cannot. In the same way, there are ways that a colony can compensate for challenges that an individual bee But not. So for example, if a pesticide kill suddenly reduces the foraging force of a colony, which is not uncommon. If you have an aerial spray, for example, during foraging, you have a lot of foragers die, a colony is not going to just lose the capacity to forage. What a honeybee colony does is accelerate the maturation of hive bees so that they forage at an earlier age it restores its foraging capacity. At some expense, of course, to its ability to do in-hive tasks, but in a way that the overall absorbs that shock of forager loss.

Speaker 1: So when we study just the parts, it may not translate to the whole because we don't really understand the ways in which honeybee colonies may compensate for an exposure incident. Absolutely.

Speaker 2: And yes, and this probably accounts for the difficulty of documenting the effects of pesticides in colonies in the field, because not only do you have all the complexities of exposure in the field that we already talked about, we have the added complexity of the buffering capacity of that colony. So that's important to remember that honeybees are unique in the degree of social organization that they have, probably unique in the degree of social buffering against pesticides that they have among bees. So things like bumblebees with a less sophisticated social colony structure are going to be less buffered and solitary bees will be at least buffered of all because they have no capacity for social buffering.

Speaker 1: Oh, okay. So that can explain how researchers have found, for example, solitary bees and bumblebees affected by pesticides, whereas honeybees might not. I'm thinking specifically that a really large-scale study in Sweden on oil seed rape, for example.

Speaker 2: Yeah, so that was a fantastic study by Mai Rundloff and her colleagues in Sweden. And they were looking at the neonicotinoid seed treatments in oil seed rape or canola. And you're right, they found they were able to find effects on solitary wild bees. They were looking at osmium, cavity-nesting, wild bee species, and bumblebees. But in that same study, they were not able to document effects on honeybees, which really suggests this social buffering capacity. In fact, there's some theoretical work out there saying that in general, in social insect colonies, one of the things that may drive the evolution of sociality in insects is this ability to buffer against outside perturbations, whether it be toxicological or weather events, thermoregulation challenges, things like that.

Speaker 1: The importance of different behavioral mechanisms within a honeybee colony really hit home to me when I was considering while reading your paper, the difference between pesticide exposure in pollen and nectar based on the different ways that bee colonies handle pollen and nectar. Can you explain that phenomenon to our listeners?

Speaker 2: Yeah, so I think with honeybees, one of the principle mechanisms of social buffering, and also just one of the fundamental mechanisms of how exposure works for honeybees, is their sharing of food that happens inside the colony. In this respect, honeybees really are unique because as far as I know, bumblebees do not engage in trophallactic food sharing, or mouth-to-mouth food sharing, the way that honeybees do.

I don't know whether the stingless bees and the tropics do, but I don't think they do it quite the way that honeybees do. For starters, it's just a primer on how honeybees transfer food. When a nectar forager, honeybees tend to forage either for nectar or pollen, sometimes for both. When a nectar forager comes back from a foraging trip, she does not consume that nectar herself, nor does she store it herself, but she immediately unloads that nectar via trophallaxis, a mouth-to-mouth transfer to what's called a receiver bee. There are bees whose job is in the hive, and this is an age-based thing. There are bees whose job it is to receive incoming nectar from forager bees so that they can go worry about the storage of the nectar while the forager bee can get right back out and forage. But it's not just a single forager bee. There's a whole population of these forager bees.

What is initiated when a forager returns is this cascade of trophallactic interactions where a forager will pass her nectar load to a receiver bee who may in turn pass it to several other bees, and you have this fanning out of this exchange of nectar. At some point in that process, it may be consumed by a hungry bee or may be stored. It's also during that process being dried down. So it's part of the process of converting nectar to honey is achieved by these trophallactic transfers that enable water to evaporate.

And if the bees have the space to store food and need to store food, they'll store it as honey, and that's the final destination prior to it being uncapped and eaten later on. So what this means for pesticide exposure is if you imagine, remember the basic temporal complexity of contamination, and a foraging creates this distribution of exposure levels that may range all the way from nothing at all to some very, very high dose. That distribution, so picture a bell curve or any shape distribution coming into the colony in these individual nectar loads, these are then going to get shared and mixed in this trophallactic cascade, which means that they're going to get diluted and gradually homogenized toward their means.

So that distribution is going to collapse as these trophallactic exchanges mix the nectar more and more thoroughly. And one very interesting thing, and this ties it back into this idea of social buffering, is that honeybees preferentially share food with bees of similar age. So the forager bees are the oldest bees in the colony. The receiver bees are somewhat younger than they are. And so the foragers are going to give it to receiver bees, but then those receiver bees are going to be prone to sharing food with bees of similar age. And so what you get is this gradual trickle down of field nectar from the oldest bees in the colony to the youngest bees in the colony and finally to the queen and the brood.

And so the queen and the brood are going to have nectar that's gone through a lot of cupbearers if you will. First of all, anything that was enough to outright kill a bee would not have been passed on because it would have died. But the highest doses from the field will have been largely diluted toward some average dose by the time it gets down to the youngest bees in the colony.

Speaker 1: So there is a way in which the bees that are most expendable, the foragers, that are towards the end of their life are perhaps experiencing those extremes. But when you get down into the more valuable bees like the nurses and the queens, there is this kind of, as you say, this shock absorbing that takes place.

Speaker 2: Yes. And I should add too that the brood and the queen are not just fed straight honey, they're fed the by the nursepies. So there's another filter, the nursepies are kind of the final filter. They're creating brood food and oil jelly, which is actually a secretion of their own body. So there's another chance to filter out any external toxicants that may have entered the hive. And so what? But then the case is very different for pollen.

Speaker 1: Pollen, good. Okay, I was going to ask you about that. Yeah.

Speaker 2: So this whole protective, tropholactic system does not exist at all for pollen. Pollen comes in as a solid material and instead of unloading to a receiver a pollen forager herself goes straight to the cell where the pollen is to be stored and drops the pollen there. And it forms like this patty and other bees will come along and press it down into like a little cake, add some saliva and maybe some regurgitated nectar to it forming what we'd eventually call bee bread. But you end up as pollen lows get deposited, and you get these stratified columns of caked pollen.

And so there's no mixing though. So this is they're going to retain whatever that distribution of exposure of pesticide levels that they had in the field is going to be retained in those pollen loads. And importantly, that pollen is not consumed by all the bees in the colony. It's consumed specifically by the nurse bees, which are the young workers who have the most resource-invested bees in the colony.

They have a lot of protein in fat stores. The reason they do this is their job to feed all the other bees in the colony with the secretions from their own body. So they eat that pollen. They use the protein in it to make brood food and Royal J. And that ultimately feeds not just the brood and the queen, but actually all the other bees in the colony get their protein needs from those nurse bees. So with pollen, any pesticide exposure that happens is going to happen directly to the most important cohort of bees in the colony. And it's going to happen in an undiluted way. So whatever the distribution of doses that occurred in the field was, that's going to be reflected in the pollen consumed by the nurse bees. Okay.

Speaker 1: So we've got two examples here of the way in which the food is coming in. And we talked about the other aspect of it as well, just earlier, that it's going to be coming in in varying amounts anyways. But when it hits the colony, we're going to see these complicated processes. And if it's coming in in pollen or if it's going to be coming in nectar, it could have quite different effects on the colony production in terms of, who are the most productive bees in the colony and how are they being affected by this exposure?

Speaker 2: Yes. This reinforces something that we've already known, but the formulation of pesticide matters immensely in the actual risk that's posed. So even if you take something, two formulations that have the same concentration of active ingredients, if one is formulated as a spray and the other as a dust or as micro-encapsulated particles, these dry formulations are going to be much more likely to be incorporated in the pollen and the nectar as well. Consequently, probably poses a greater toxicological risk to those nurse bees. Okay.

Speaker 1: And I guess that makes the argument even further for being able to understand those mechanisms and really bringing those mechanisms into our models of risk assessment. Yes.

Speaker 2: And I'll throw out there too that I think there's a really interesting basic biological question to ask here too because we must not forget that honeybees didn't just start experiencing toxic exposure when people started manufacturing synthetic pesticides on a large scale. So toxic exposure has always been a problem for honeybees. And I think that, the evolutionary history between honeybees and naturally occurring toxins in the environments, like plant secondary compounds and pollen and nectar, things like that.

I think, that those may have been instrumental in shaping this social buffering capacity of the honeybee colony. And so the question then becomes, in what cases is that system robust to these novel toxins that we're giving them? And in what cases do these novel toxins somehow evade this buffering system and cause damage to the colony? And I think, for example, insecticidal dust, maybe one case in which a novel toxin then evades the buffering capacity of the colony and damages that nurse bee population that might otherwise have been protected.

Speaker 1: So there are evolutionary ways to think about these current problems of pesticide exposure. That's fascinating. Okay, let's take a break and then we're going to come back and I'm going to ask you some questions that I ask all our guests about favorite books and tools and what your favorite bee is. Okay, so we're back. We've got these questions we ask our guests. And the first one I want to ask you about is a book. Do you have a book recommendation, a book that you want people to know about, or maybe a book that's been just really influential for you?

Speaker 2: Yeah, I think my favorite book on honeybee biology, maybe my favorite biology book overall is Tom Sealy's Wisdom of the Hive. Oh, yeah. So the structure of the book is there's an introduction to basic honeybee biology, but then it's really a commentary on a series of studies that he and colleagues did over more than a decade. And it's really, in addition to being just a beautifully compelling story of honeybee biology, it's a textbook of scientific methods, experimental design, hypothesis testing, and interpretation of data. And honestly, it's a book that I would use to teach a course in the philosophy of science or scientific method.

Speaker 1: It is really, it is an interesting book that way. It's been recommended a number of times, but nobody's quite put it that way. And I always loved it. I love the introduction, for example, like how are all these experiments bound together? You know, here's my process.

And I love that it is a kind of sustained set of questions and problems, as opposed to, you know, what you typically pick up in a book about a topic in biology, where it's kind of glued together, there's something kind of coherent about it. Yes.

Speaker 2: Yes. And it's accessible to do. I think that it could be read by someone without formal training in bee biology or someone with formal training in bee biology, and it would be equally useful to both.

Speaker 1: That's a great suggestion. So our next question is, is there a tool that you find indispensable? If you were on a desert island or something and you had to, you know, you wish you had something to do your work, what would be that tool?

Speaker 2: If I were on a desert island and needed to do science, I would probably leave my canteen of water and my last little package of food behind in order to take with me a computer with R loaded on it, because I had nothing to do in science.

It does not involve the use of R in one way or another. It's something I took the plunge with when I was in grad school, and it was probably the most practically useful decision I've made. So I use it for everything from graphics and data visualization to modeling, to statistical analysis, to just, you know, basic programming and data manipulation. And then I would say together with that, I've used the GIS program QGIS extensively. So I'm a big fan of free open-source software. QGIS is an excellent geographic information system that's free and open source, and there's a really active developer community behind it. And its functionality is really quite expansive at this point. And in fact, a little plug for QGIS 3.0, is coming out in a couple of weeks, I think. And I can't convey my excitement about that.

Speaker 1: It'll be big. And I think this really brings across as well that a lot of your work involves spatial ecology as well. You really do have a broad range of areas that you work in.

Speaker 2: And the other thing... I would hasten to add that toxicology is spatial ecology. I suppose, right? I think that that's often overlooked in traditional toxicology, that the spatial aspect of it is downplayed. One of the themes in my work is that toxicology is spatial ecology, and they're really two sides of the same coin. Okay.

Speaker 1: Well, I do have a picture of you on your island with your solar panel and your computer working out problems, which is wonderful. That image makes me happy. Okay. The last question we have is there a particular kind of pollinator or B that when you see, you know, just like, oh, or a memory of a bee or... Yeah. What's your favorite beer?

Speaker 2: Well, so we've talked really just about honeybees today because most of my toxicological thoughts have pertained mainly to honeybees, but I really am enamored with the diversity of wild bees.

And I actually do think quite a lot about them toxicologically too. It just hasn't come out in my writing yet. But as far as specific bees that really capture my imagination, I mean, the honeybee obviously does, and it does for so many reasons.

I don't want to downplay that, but I think that's obvious. So moving on from that, but maybe not too far right away, the honeybee species pastor soda has always really captured my imagination. It's called the giant honeybee. It's a very large bee.

It nests in these open combs off on a cliffside in Southeast Asia. And it's a beautiful bee, and it forms these beautiful colonies because the bees are all just layered on this open comb. So you see the whole nest right out in front of you. And they do this amazing rippling behavior that they use to ward off things that would land on the nest to try to steal honey or brood.

That creates this shimmering curtain-like appearance that is just mesmerizing. I've never seen these bees in the wild and I've never been to Asia, but it's a dream of mine to get to do that someday. And then finally, I would add that the other bees that have a special place in my heart are the earliest spring bees after a long cold Ohio or Pennsylvania winter. Those sweet little andrena that come out in the early spring and flit about the Claytonia blossoms, are special bees to me too.

Speaker 1: Well, spring is not that far away listeners, well it is, but just think that it's not. The adrena will be back. Well, Tuck, thanks for taking the time. I know you've got a busy schedule and we really appreciate you introducing us to the area of ecotoxicology.

Speaker 2: Thank you. It was great to be here, Antoni.

Speaker 1: Thanks so much for listening. Show notes with information discussed in each episode can be found at pollinationpodcast.oregonstate.edu. We'd also love to hear from you and there are several ways to connect. For one, you can visit our website to post an episode-specific comment, suggest a future guest or topic, or ask a question that could be featured in a future episode. You can also email us at pollinationpodcast.oregonstate.edu. Finally, you can tweet questions or comments or join our Facebook or Instagram communities. Just look us up at OSU Pollinator Health. If you like the show, consider letting iTunes know by leaving us a review or rating.

It makes us more visible, which helps others discover pollination. See you next week.