 I am Anvesh Mazumdar, National Coordinator for Science, Olympiad in India, and it's my pleasure and honor to invite you all to this Infosys Award function for the Olympiad Medalist of 2016. The award function will start later. At the beginning, as is the tradition, we have two lectures, one after the other, and then we'll have a break, and then the award function will follow. So at the outset, I request all of you to please put your mobile phones in silent mode. So the first speaker today is Professor Sanjay Sanay. Professor Sanay obtained a BSc in physics, chemistry, and mathematics from St. Stephen's College and University of Delhi, and then a master's in physics from University of Pune, with specialization in astrophysics and nonlinear dynamics. However, he was fascinated with insects from his childhood, and this drew him to study the aerodynamics of insect flight as a PhD student at the University of California, Berkeley. After finishing his PhD, he served as a post-doctoral research fellow at the University of Washington, Seattle. And then he joined the National Center for Biological Sciences, NCBS Bangalore, a part of TIFR, where he is currently an associate professor. So, Professor Sanay, it's a great pleasure to welcome you. And to give you a talk, just for the students, as a remark, I mean, his career will show you that he started off as a physicist, and now he's doing completely different things, maybe which has a lot of physics in it also, but it has a lot of biology. So it's not that what you start off in your early career, you're stuck with that forever. So please keep that in mind in your careers as well. You can always change. Professor Sanay. Thank you very much for inviting me here to your presentation. I'm so glad to speak to a room full of such potential. What I'm going to talk about today is some work that has evolved, as was mentioned earlier, my main day job, so to speak, is to study how insects fly. But this project evolved out of a curiosity, almost like a hobby project, and it has grown and grown, and I want to share some of the excitement of what we've been finding in this lecture, but also to draw a question that science is in the backyards, that I would really like to emphasize, there's nothing else that you take away from this talk. I'd especially like you to take home the fact that science is right there in your backyards, but I realize this is Bombay, maybe backyards are not common. Bombay still has a third of its geographies, just a jungle, you know. The Sanjay Gandhi National Park is a fabulous place to be, and you're going to find a lot of these sorts of examples out there and also wherever you can look for them. So let me start with something that you might be more familiar with. So this is human architecture. We understand this. We are instinctively aware of this. We stay in structures that were built by humans. Now the oldest structure here in Cane de Bernanese is fairly sophisticated even at this stage. This is a building that's set to have been built in 4850 BC, and this is Manhattan, low Manhattan as it looks today, and we have a good 6,000 to 7,000 years of architecture to look at, and we're proud of them. We often look at this history of 6,000 to 7,000 years and think that architecture is the purview of human beings, and that we view the ultimate knowledge of all this. But it goes back much further than that. So I'm going to give you two examples. These are both from mammals. At your left here is a beaver dam somewhere near Alberta, Canada. That little rodent there, it's not little, it's actually the biggest rodent that there is, is what builds these structures. And it's fascinating, since it's been building these structures for nearly 20 million years, and in the course of this, actually changing how rivers move, changing certain rates, making very fundamental impacts on their ecosystem, and the reason it does this, at least so far as we know, is that it needs to build a cave in here, right in the middle, which is protected from wolves, coyotes, and all sorts of predators. It needs to ensure that there's enough water surrounding it, and so if the stream is somewhat shallow, what it does is it puts together a huge structure here, the beaver dam, that then accumulates these waters, and in the course of that, actually changes everything that happens downstream. On your right here is a pearly dog nest, another rodent, and this is just a picture of it, but it has to be a picture, because from the outside, all you see are little holes, and they're very well conceived, and pearly dogs are remarkable at being able to engineer these underground structures, in which house they're young, they're old, they're food, and all sorts of things. And they do something even more interesting than that, which is not very evident here, but you can see this little mound on the hole, that actually allows ventilation through the hole, so some of these holes don't have mounds, some of these have mounds, and then air moves over it. It naturally flows through, because of a great pressure difference, and you'll see this theme again in some other structures. Birds are also architects, I'm sure a lot of you have seen the Bayaa Weaver Birds, which exists all around us, and this marvelous structure it's a nest which houses the eggs of the pear, the Bayaa pear, and it's really a masterpiece of weaving and application, it's done by two birds, with nothing but the bees in the internet. And if you thought that a structure like this evolves, but once in a, you know, as a freak incident, and that's not true either. Completely independently of the Bayaa, in South America, in the New World, we have this structure that is built by this bird, this is the Oropendala, and if you were to go to places like Panama or Costa Rica, you would find these hanging there, they were completely independently revolved. Now, what's impressive about this structure is that they're able to withstand tremendous winds, cyclones, storms, all sorts of environmental pressures that leave it completely unbattered. We can go back further in evolution, here are marine mammals, marine animals, sorry, this is a worm, it's called a lab worm, I don't know, and it's a solitary worm, what it does is it burrows through the sand, and just like the prairie dog, what it does is it leaves some people matter here, and you can see that mount right here, but leaves the other hole at ground level, so there's a difference between the heights here, and when water flows over this, some of it flows through the tunnel, just because of the pressure difference, and this is a fit of feeders, so what it does is it filters the water and eats as the water is flowing through, and of course, the coral reef is a remarkable structure, some of these are kilometers long, and as you might be aware, they house all sorts of species, and in fact, they are the backbone of many, many marine ecosystems, and these are all engineered by cousins of jellyfishes, jellyfishes are considered by us, to be some of the oldest metazone, multicellular animals, and yet here we have structures that are pretty remarkable. So, this sort of engineering, and you know, I could go on and on about principles on which they're built, and what they teach us about architecture, and about how we should engineer our buildings, but you don't have to go to Panama or to South America to see these structures, they are all around us, in fact, they're right here, this is, I'm going to just show you four examples, or five examples, from my backyard in Bangalore, and there we see these structures, so here's the high of a honey bee, I think you're all aware of this, this is a pistachio, it's a pretty nasty bee, it's an Asian honey bee, and this structure is built literally overnight, by a large number of bees, they are all sort of huddling here, and this superstructure is built to make this coordinated activity between them, that structure houses their brood, houses the queen, it is also the store where the store honey, and because it's a structure of this nature that really is the place where all of the colony exists, it must have certain properties, and these are sort of the general things that one sees in any structures which house a lot of individuals, one property is that it should be relatively infactional, so you can't afford for an infection to come in, and then spread through the hive, and then you end up with a dead colony, that won't work, so the hive has antibiotic properties, this is something that we all know instinctively, we take honey, we chew on the hive, wax it's quite tasty, here is the nest of a beagle rat, this is again something ants that we see in our backyard, these ants through coordinated activity are able to go on plants, take leaves, turn them around, as you can see a lot of them have to work together to be able to do that, and then what they do is they use the larvae, some almost like a tooth clips, if they squeeze the larvae or they stimulate it, the larvae produces a filament, and they use this filament as the binding threads that put together this hive, and then this hive of course, that's a food that they bring, their family was, they will catch other insects, dead insects as well and bring it here to feed the young, these are paper wasps, and these again they're large nests made out of paper, and they are in fact able to chew on sometimes the plant material, they're garbitated, they turn it into paper, these extraordinarily large hives are actually very, very large, and they're able to house again the fruit and the entire colony. A little less dramatic perhaps is a carpenter bee, a very large black bee that often gets called a bumblebee, it's not a bumblebee, we don't have bumblebees below Himalayas in India, it's this bee, it's the carpenter bee, and what it does is it looks for dead wood, and in the dead wood it'll make a neat little hole with its proboscis, small part, and it's a nice circular hole, and if you were to look inside the hole, there are neat chambers in which they deposit their larvae, and they periodically go out, hunt, bring in food, and feed their larvae. Now, I'm showing you structures that are products of a long process, and in general I want to talk about science also, as a process, not so much as a product. The actual way in which this insect is able to do this is quite remarkable, and it is something that any of you would be able to record with your cell phone cameras. I'm just going to show you a video that I put out of YouTube, but this is a really remarkable video, this process. This is a part of us, and what it is doing is making a nest, and this nest is shaped like water, and you can see that in doing so, it's using all its body parts, the antennae, extraordinary mechanosensors, we study these in their old flight, but they're equally important here, and they keep bringing these balls of clay, and over a period of a few hours, they're able to build this remarkable structure now, right after this comes part of really, that's my favorite part. This insect is a perfectionist, and what it's going to do, having built this, and make sure that it has the right structural properties, it's got the right strength, and it's going to be able to withstand whatever load that is going to have to carry, once it's ensured that this insect now lays eggs, those eggs will hatch into larvae, and the insect will then go and periodically bring it food to eat, in fact, it does something quite remarkable, it goes out, it strains a caterpillar, a large caterpillar or a moth or something like that. The caterpillar is alive but parallel, it's like a zombie, it's then brought here, put in, and the insect larvae will feed on this caterpillar, without killing it, because if it's dead then it's actually a hazard for the nest, so it's kept alive, and the insect continues to feed both the caterpillar and the larvae, and that's how this whole structure works. This, of course, all of you are familiar with, this remarkable structure is made by Spider, and Spider actually goes to a fair number of iterations to be able to make this structure, and it serves not only as the place where Spider hangs out, but also as a way of cutting animals, which then eat. It also serves as a way of harvesting water, collecting dew drops in the morning, and it's made out of material that is remarkably, it's about as strong as tea, not stronger, and it's able to withstand tremendous loads, for instance, of an insect coming and pitting into this. It doesn't make the, and to convince you that this structure is a product of extraordinary mental ability, you can actually experiment with something like this, or you can do, is feed the Spider various kinds of cortex. So here's a normal Spider, here's Spider and Marijuana, benzodiazepine, caffeine, so a lot of you identified with that this morning, so I will hydrate, these are remarkable, and what these chemicals do is interfere with the coordination of the mental coordination of how business are built, and I should point out that that coordination operates both in space and in time, so they have to do things together at the same time while also being able to keep track of what they've achieved and how much further they need to go. This is really something that requires tremendous amounts of mental ability and insects have. So these are the sorts of things that we were gossiping about for a long time in my lab, up until a few years ago, when we decided finally to do something about it. We gossiped about something for a long enough time, it's a good idea to just take it up as a career, and that's sort of what we did, and specifically what we were interested in was of these structures, these are termite mumps, and they existed in our backyard, I'm sure many of you have seen this. This particular structure, this is the student who works with this, his name is Amritanj, he's about six feet tall, and the structure is slightly taller than that, and there's as much of it above the ground as there is underneath it. So what the termites are doing is excavating material, putting it on top, and then shaping it into this remarkably fluted structure, you can see these folds in the structure, and this entire structure houses its solid. Amritanj trying to understand what it means, what is this structure all about, what is the function of this structure, why is this structure required by the termites, how is it that certain termites can survive without it, certain other species of termites, and many questions like that. Now, underneath this structure are some remarkable things. So, first of all, there are millions of termites, and these termites are segregated in the past. There are soldiers, there are workers, two types, major and minor workers, and then there are these termites called aleids, aleids, some of them were healer meaning wing bearer, and the aleids for their swing. So, these are the only termites that can actually fly, and you might have seen them in large numbers after the first rains, they get all over the lights, and they make a lot of mess. Now, in many colonies, termites cannot actually, the termites eat good material, plant material, but many of the termites can't actually digest this material, some can, some can. Those who can have certain bacteria and their guts, which help them digest it, but those who cannot, and particularly these mom-breeding termites form a fungus with it that grows only in this mom, it can't grow outside, it has to grow inside the mom, it is formed, it is carefully tended to, it is kept in many different chambers, and in fact, if you were to look at the metabolic of this mom, a large part of it is accounted for by the fungus, not the termites. You might actually, if you were, somebody who works on fungus, turn the question around and say, hey, you know, what's, maybe it's the fungus mom, and the termites are just being recruited to help the fungus out. Inside this mom, there's a complex maze of tunnels and bridges and all sorts of crazy architecture that we do not understand. It's topologically complicated, and it's mechanically, of course, really, really hard to decide. But let me just show you a quick film. This was shot by a cell phone camera by my student, followed by the sitting sleeves. So this is NCBS at night, go down in, there is a busy, busy mom, a busy, busy nest of termites. And you can see just how much activity is going on there. Look hard, you might not see any traffic jams in there, and I'll come to that in a minute. So let's briefly think about what termites are, what these moms are. Termites are actually cockroaches. They are cockroaches that have become social. And they tend to, so if you are looking from the point of view of an evolutionary biologist, they are very, very similar. And so these moms are built by these termites. And when it rains, or at the first sign of increasing humidity in the environment, there are little chambers. And these chambers are somewhere at the side of the mom where termites will come and they'll open this chamber and they release this special class of termites or the allates. As you can see, all of them have wings. They also have eyes, and I'll tell you a little later that most other termites don't have eyes. So these are released one by one. They make a hole just large enough for these termites to go up, or these allates to go up, and then they go out and they mate. After they've gone, the hole is closed and the termites go back in. So these will mate, okay? They'll mate, and then they will get rid of their wings immediately, and then they begin to grow, okay? So this is now a female, and she's about as big as my thumb. A termite is about a few millimeters in size. This female is about as big as my thumb, okay? And most of her is just ovaries. Nothing but eggs, it's just a big egg machine. And her body is that big, it's this little, sort of vestigial kind of body. You can see that she had eyes, because she was an alley, little legs, but she's not going to be able to move around. She is the one who actually controls the model, and she ensures that all of these soldiers and workers are working in their, you know, she controls the rate at which she produces these individuals, and she can go on and on for about a few decades, thanks to 20 years, all along she has a mate that continues to mate with her. And that mound is this incredible structure that I just mentioned, and we've been trying to understand what its function is. This is my collaborator, Rupert Soar, he's an architect from the UK, and he came up with a method of trying to figure out what the internal structure of the mound is. What are the different mazes and tunnels and so on inside? And what he developed was a very nice method, where what we, that's off a little bit of the mound and pours in it, gypsum plaster of Paris fills it, and then at the end of the day, he sort of allows it to hide and washes off all the mud, but what you get then is a negative image of the mound, negative 3D image of the mound, and that's what he's probably setting next to it. It's a really wonderful structure, and that allows us then to reconstruct, you know, what the internal structure of the mound might look like, and as you can see a lot of it, because this is failed, and it's a negative image, a lot of it is just empty space As to the function of the mound, there are many, many ideas, and I'm just going to play you a short video by my hero, David Attenborough, as he is talking about this month, and I think he does a fabulous job of explaining this, so I thought it's just better to share that with you from life and your video. So termites can take refuge from the heat, although brown, it's cool and relatively stable. But two million insects living below the ground create a difference. It's like a fortress, it keeps the termites safe from ants, and so, and other predators. But in addition to that, we have been wondering about what is it that the mound does for the termites? And there are few ideas, again as was also mentioned in the video earlier, but these are by no means something that we know for a fact, and these are developing and we're still trying to figure out much of this. One idea is that the termite mound serves a thermoregulatory purpose, meaning that it is required to keep the temperatures inside the mound at some constant level. The second idea is that we serve as some sort of chimneys to allow the stale air, the carbon dioxide, that's generated by the column to go out in periodic intervals. And these are structures that then can be manipulated as and when the stale air increases termites, both in terms of ventilating the mound by opening or closing these chimneys. These are all big questions. We really don't know the answers to this. And much as there's been a lot of theories on this for nearly 100 years, a lot of them actually have been turned out to be dead end. And so we continue to work on this. What it would be especially interesting is how is that single termites which don't carry blueprints and in fact, which don't even have eyes, as you can see, this is the head of a termite, there are no eyes, have this impressive antennae, these mandibles that you saw were being used in fighting, that they're also used in hunting materials, but they don't have eyes. And how do they know what to do? And what are the cues that are used by these termites to put together, and they're busy. They continually modifying and remodeling the mound structure and trying to make sure that integrity is maintained. They can actually increase its size, depending on the moisture levels and what they're doing is constantly collecting, if you concentrate on any one termite in this video, you can see that what they're doing is really sort of taking materials and depositing or taking materials from one place, depositing it in another. And in fact, what they do is they ingest the soil and they mix it with lignocellulose, which is sort of digested plant material, and that's what makes this glue work so well. So the main questions that we are tackling is what is the function of the mud? How do termites know where to build? How do they manipulate the building material and if we remind you that the building material is soil, but that soil changes from place to place. The same termites might build a mound in the western parts with very different soil than it does in Bangalore, with a more claying soil. And also what sensory cues guide individual termites in this building does. How do they sense their environment and how are they responding to it? But before we even ask questions like this and that's sort of just the way we work, we decided to not take any, make any assumptions. We say, let us just start with the null hypothesis that the termite mound has no function. And the idea is that if it is true that the termite mound has no function and that it's just a side product and it has to pull out material, so this is just a way of depositing the material. If that is true, then it shouldn't care about any injury to the mound, okay? And so that's the first hypothesis that we decided to test. The idea is very simple. What you do is you go in, you make a hole in the mound and you ask, what will the termites do? They respond to this hole. And so that's what we've done here. My student, Sri Krishna Verma Raja made a hole about two centimeters in size, in diameter, and we wait. And in about 23 minutes after the first termite shows up, the hole is filled, it's filled with a scab. And let me just show you the very list, okay? So that's the video here. The plot here tells you just what fraction of the hole has been filled, 100 being a fully filled hole and zero being a fully opened. What we do then is we analyze the images to look at what that fraction is. And so as you can see, if you do this, and if you do this again and again, you keep getting the same form of the termite. The termite is a sigmoid. And this is sort of the ratio of the cover to the total area. And we keep doing this experiment over and over again and we keep getting the same result, except that the sigmoid might look slightly different. It might be shallower or steeper, but it's always a sigmoid. And that's suggested to us that there must be an exponential sort of recruitment process and an exponential D recruitment process that's ongoing. And so we then decided to sort of take the line when the room was attested in some theoretical aspects of this, we thought, yeah, we could build a model. So we have to make a few assumptions. One of the assumptions was that the rate of building is a good proxy for the number of times. In other words, the rate at which the building is happening indicates how many termites are at work. A corollary of that is of course that each termite is working at some constant rate. So then you can actually use this assumption. How might such recruitment work? It could be chemically mediated. So it might be that the termite shows up at a place, finds a hole there, we don't know how, but then lays down a chemical. And then that chemical attracts other termites, which then from there, they find the hole, they lay down the chemical, and that could lead to a recruitment process that's exponential. Because the number of termites that are recruited are directly proportional to the number of termites that are on this point. It could also be mediated by sound. We know just from observations in the lab that termites can vibrate their heads very, very rapidly. These vibrations actually lead to a sound that you can hear. The de-recruitment could be mechanically mediated because termites are proud, and then they stimulate each other to stop work, sort of like local termites. There could be other means, which is just some grace. So how do you actually go about testing even some of these assumptions? My student, Amrit Amsh, came up with a very nice method. What he did was he, instead of just making a hole, he put a funnel there. Now, what the termites do is they come out and they start building over the funnel. That's sort of what it looks like when you take a picture from above, and now you can see the termites actually on the surface, and you can measure the rate at which they're building, and so you can ask, is the number of termites a good proxy for the rate of building? Oh sorry, the other way around. Is the rate of building a good proxy for the number of termites? And that turns out to be a pretty nice assumption. So, here is the area of birth, here is the number of termites, there are many different trials here, about 10 of them, but he's just one of the trials. And you can see that this regression is pretty high. We know that indeed this is a good assumption. So we can then put this assumption into a larger kind of a mathematical model. I want to go through the details of this model except to say that you can put these assumptions in, build a model, and what you get out of it is a differential equation which once you solve, you get a sigma, and that's not surprising. You sort of get what you put in. But this model is useful because you can play around with the constants and then start asking questions that go back to the biology. You can ask for instance, how do you change the rates at which termites are building? How do you make a sigmoid shallow or steep, or something like that. But let's go to the question of the sensory cues that are being used. So that we developed a variation of the final assay, which is a parallel plate assay. What you do is instead of putting a funnel there, you put a parallel, set a parallel plate, and they start building on the parallel plates and then you can fill from the sides. And you get time-lapse images that allow you to see the rate at which the termites are building. And so that's what it looks like. This is the hole. That's where the parallel plates are kept. And we use this then to ask, how is the rates at which they're building different? So this is an experiment that was done through the day. It starts at eight in the morning, ends at about 10 in the night, and as you can see in the morning hours, these rates are very steep, but as they go towards evening, they begin to get shallow. And this was something that prompted us to ask the question, could it be that light isn't there? Could also be temperature, but we first thought it was to ask about light. And for that, we came up with two sets of experiments. One is this mound in the shade. So we have a mound in the shade, the temperature is more or less constant at both places. What we do is we make two holes at the same height. And in certain those holes, tubes that are either transparent or covered with black paper. There's something interesting about this. In the tubes that are transparent, the termites fill the tube entirely with mud. It's completely clear. In those which have a black paper on top, they only fill up the walls, but not the center of the pore. And you can wave these muds and see that there's significant difference between these two methods. So what this suggested to us was that the light has a sensory cue, which was very surprising because as I mentioned, these termites don't have eyes. So we wanted to double check, and so here's another assay. What we do is this time is have a parallel plate assay, but the core of it is an aluminum plate, which doesn't have a light, doesn't go past it. We keep one side in light and the other side in dark, and we ask how is the rate of lighting different? And as you can see, this is one side, and this is the other, and there's a difference amount of lighting in these two places. And these are multiple trials. The black bar is the light part and the dark part is the red part, and you can see that always the light is more building than the dark. So we knew then that they can sense light, and that light is a cue. When you make a hole in the mom, it brings in light, and that light activates the termites to come and start building over it to minimize that light. It's in some sort of negative feedback loop. So I'll leave this here, and I'll go to another question that we've been pursuing, which is about termite mounds and moisture. And we know that when it rains, there is a lot of humidity, but this is something that has been known for many, many years. There are, these are just two Sanskrit couplets, which were pointed to me by a student who knows a little bit about this. And she mentioned that there are these two couplets in which there's this indication that wherever you see a termite mound is some indicator of where there is a water source underneath. So the word Valmiko, Valmiko is the Sanskrit equivalent of the termite mound. It's actually literally an ankle, but it's not an ankle, it's a termite mound. And so here it says that if there's an ankle near the Jambu tree, always towards its east, you will find a good water source at two Kuru Shahs, the length of a roomed body from there. And here, two other trees, so you have tar and other peary surrounded by ankles. There should be an aquifer towards the west at the distance of four Kuru Shahs, the sixth hospital. So when you talk to people, they mentioned that there is some relation between where the mounds are and where there are water sources. We were interested in this question because we know that soil moisture is very important. But how do termites use soil moisture? What effect does soil moisture have on the building process? You can see that there is a new mound right here, there's the old mound, this happened right after the, that this is a more moist soil in this. So we again use the parallel plate asset for this purpose. And what we did was have a series of five plates in which we put completely dry soil first and then variable amounts of moisture. So we know exactly how much moisture is in there. And then we drop in 30 termites in each and then come back 45 minutes later and you see that there's a differential mound in the building because it's a multiple number of times.