Science & Technology Q&A for Kids (and others) [Part 132]

Hello everyone. Welcome to another episode of Science and Technology. Q and A for kids and others. And I see a variety of questions that are built up here. Let me see. Let's maybe start off with one from M. Rudo. What would a bio computer look like? So, you know, the computers that we have today that we are used to using are electronic computers. They work by having electrical signals going through lots of tiny wires and switches, billions of them in microprocessors. The what would it mean to have a bio computer? It would mean that the material that was being used for the computer wasn't electronics. With solid state, a traditional microprocessor is made mostly of silicon etched to make these little wires and so on. It's really all a solid and there are electrons that are going through these little tracks in the solid. What one would presumably mean by a biocomputer is something which, instead of being made from solid state materials with etching and all this kind of thing, it's made from biological materials. Now, in some sense, we humans and all the other biological critters of the earth, so to speak, are examples of biocomputers in the sense that there are. What is a computer? Well, a computer is in the end something where you can run lots of different possible computations on the machine. You can give different inputs and the machine can do follow different rules to compute. The answer. It's, I mean, the big and non trivial fact about computing is that universal computing is possible in the sense that you can have a particular piece of computer hardware, a particular device, and then by feeding it different programs, you, you can get it to do different computations. And that's why in some sense you can expect to have the same, I don't know, word processor or something, running on computers of different kinds with different computer hardware, so to speak. Well, so then the question is, could you take that thing that's running on a piece of traditional computer hardware, I.e. solid state electronics, and could you somehow deploy that on something that is made of biological materials? And although the things that we compute are a little different from the things that computers tend to compute, in some sense we biological organisms are doing sort of computations in the way that they just through living. And all the things that are involved in living, they're effectively doing computations. Now, for example, in our brains we like existing electronic computers. We use electrical signals in the nerve cells in our brains to be the things that sort of carry information around our brains. In most of the rest of the way that we operate as biological organisms, we're not using electrical signals. That is, we're not using streams of electrons, effectively an electrical charge and so on, to carry information. Instead, we're transporting molecules around. We're actually moving. Maybe it's individual atoms of sodium or potassium or something, or maybe it's actual whole molecules, proteins that are made of lots of atoms and so on. We are computing things by actually moving sort of material, molecules and so on around. And one of the things that's become clear in molecular biology in the last few decades is that it's a very orchestrated process. What happens, you might have the picture that biology is just that we're all just sort of pots of chemicals and that the things that happen inside us are just the result of chemical reactions from molecules kind of in a liquid, for example, bouncing around randomly. And when two molecules that can react happen to hit each other, then you get a chemical reaction. Something happens. It seems decreasingly likely that that's what's going on most of the time in biology. It seems that biology is full of kind of elaborate orchestration where molecules are being specifically moved along little fibers where molecules have particular shapes, where they orient themselves so that they will expose a particular part of the molecule so that some other molecule can fit into the thing that's exposed and so on. It really feels like a process where there's. There's a sort of giant piece of choreography of molecules that's leading to the processes of life. And so it really. It looks very much like sort of a computation that might happen in a existing electronic computer, except for the fact that it's happening with moving around molecules and things like this through perhaps things that look a lot like wires. They're little fibers in the skeleton of cells and things like this. And different things happening on the surface of a cell and ways in which, for example, different pieces of molecules are exposed on the surface of a cell so that they are arranged so that other molecules can come notice that they're there and all those kinds of things. So that's sort of the overall architecture of what seems to be happening in living systems is something that is, in a sense, very computational. It's all about these molecules moving around in very orchestrated ways. Now, how do you program these things? Well, we don't yet know. And it's pretty tricky because the thing that seems to be the most important about these molecules is basically their shape. One of the things that's also kind of a big idea of biology is the notion that you can kind of make a molecule of essentially any shape just out Of a protein. What is a protein? Well, the ultimate thing that we know about biology is we store genetic information on DNA. For us humans, it's 6 billion base pairs that are sort of different segments of this DNA molecule. For a human, if you stretched out the DNA molecules that you'll find in every single cell in our bodies, it will be a meter long molecule if it's sort of stretched all out. But on that meter are a billion little segments that could be A's or T's or C's or G's that essentially spell out the genetic code for a particular one of us, so to speak. Well, what's done with that genetic code? There are sections of it that correspond to so called genes. And when at there are processes that cause those sections of the DNA to be transcribed. And what are they transcribed into? In the end, every triple, every three base pairs turn into a single amino acid and those amino acids are strung together. It all is done eventually by this thing called the ribosome, which is a little machine like molecule that kind of in which pieces of. Well, it's actually rna, which is one stage removed from DNA. But the pieces of rna, it's a big long strand of RNA which contains the genetic code, kind of goes into the ribosome and extruded from another part of the ribosome is a string of protein. And the protein consists of sequences of amino acids. Each triple of base pairs corresponds to one unit of amino acid. The protein is then the sequence of amino acids strung together. And it could be 100amino acids, it could be 100,000amino acids. But the cool thing in biology is that it turns out when you have that string of amino acids that the thing will fold itself up basically because, well, for example, there are sections of that protein which tend to be hydrophobic, they tend to stay away from water. And the ones that are hydrophilic, they tend to kind of expose themselves to water. And this whole molecule will be very floppy, but it will tend to organize itself so that it kind of has the minimum energy configuration, roughly a configuration where kind of the pieces of the molecule that are most attracted to each other can kind of get to be stuck together. And the pieces that are repel from each other. Repel from each other. But the thing will fold itself up into this kind of potentially a blob that can be any shape. And so in our bodies there might be 30,000 different kinds of proteins, maybe a little bit more, and they have all kinds of shapes. Like the actin protein that forms our muscles is this kind of triple helix, that's a long, stringy thing. The protein hemoglobin that is in our red blood cells, there's a little cage that just has a little space in it about the right size for an iron atom that can be used to kind of bind oxygen and transport oxygen around our bodies. And all the different proteins that we have, they're all different shapes. They're sometimes globular, really a blob. Sometimes they have these things called alpha helices, which are these kind of spiral regions that are kind of long and straight. And so, for example, proteins that stick through the membrane of cells might have an alpha hel. That is the actual sort of tube that goes through the wall of the cell. But any case, the big point is that it turns out that so far as we can tell, and this isn't completely understood, the kind of space of possible shapes that you can make with proteins is at least very broad. It might not in some sense be every possible shape, but it's a very wide range of different shapes. And those shapes can be used as sort of puzzle pieces that fit together to do all kinds of things. And in a sense, what one imagines is that sort of a biocomputer is all about being able to have the right puzzle pieces to fit them together, to do the things that correspond to following rules, that correspond to the computations one wants. And by the way, there's some more details of this, about how electrons are transported on these molecules and so on, but more, I think the key point is this sort of shape thing that allows you to make these different shapes of puzzle pieces and then sort of see how they'll fit together. So I think the big difference between sort of a current, sort of molecular scale computer that's a biological one, and an electronic computer is probably that in the biological one, one tends to be transporting molecules around, whereas in the electronic computer, one's transporting electrons around. Now, by the way, when people talk about building quantum computers, there's a whole story about the relationship between quantum mechanics and those computers, which I'm a bit skeptical about. But the one thing that is true about those computers is the things that they use to kind of store information are different from the things that traditional electronic computers use. Electronic computers are using the presence or absence of electrons. There will be quantum computers that use ions, kind of pieces of atoms, that use kind of little loops of magnetic flux, different kinds of things that are used to kind of store information. Well, in the sort of bio case, I think the main thing that's being used to store information is probably the presence of molecules of different shapes and so on. So I think that that's kind of the picture of sort of what a biocomputer will tend to look like. Now we have examples of things that are very computational. Like the immune system seems to be very computational and it seems to consist of all these different, for example, antibodies and T cells and so on that all have sort of different shapes and that all fit together with each other and with sort of antigens that come in and invade our bodies and so on. They fit together in this kind of shape kind of way because they have these proteins on them that have different sequences that cause them to have different shapes. And so the immune system is probably a good example of a place where something very, very much like traditional computation is happening, but it's happening in a way that. Is not going through electrical signals. It's going through, oh, we've got a T cell that has this shape of receptor on the side and it's going to interact with the T cell with that shape of receptor and that's going to lead to this. And it's all this kind of sort of puzzle piece, cascade of puzzle pieces kind of story. Now people have tried to, well, a variety of kinds of things for molecular computing. One that was popular for a while, I haven't seen so much of it recently is DNA based computing where what one's effectively doing is saying one has a piece of DNA and it has this sequence of base pairs on it and that piece of DNA will be, it will be able to, let's say you have a small section of DNA, it will, with particular base pairs, it will be able to sort of puzzle piece into another piece of DNA that has the kind of anti collection of those base pairs on it. And so what you can think about doing is kind of using little pieces of DNA as pieces of kind of a construction kit to make different shapes of things. So people have made things like nested patterns and so on out of little kind of DNA tiles, where what's exposed on the edges are these kind of DNA sequences and so on. Actual biology doesn't seem to do that. Actual biology makes proteins and it's the proteins that fit together in puzzle pieces. But it's rather easy to know kind of what sequence you're getting with DNA. You can just literally have a machine that will essentially print DNA at least up to some length with a particular sequence. And by making these kind of DNA tiles, you can potentially have something which fits together with particular Matching rules. And that's sort of a way to think about doing computation. But I think that the real story here is computation with molecules and with shape recognition, rather than computation with electrical signals and pure gates and so on. Because in an electronic circuit, in an electronic computer, you just have electrons there, and every electron is the same as every other electron. And you don't have this possibility of saying, oh, the actual things that are carrying information, they themselves have kind of particular shapes or ways of connecting to other things. So I think that's sort of the key idea that biology brings and that we will eventually, I suspect, have technology that kind of matches. And it is sort of nice to be able to kind of leverage existing biology and think about actual ribosomes that evolved over hundreds of millions of years or something back a few billion years ago, and actual sort of the actual things that are used in the different kinds of the 20amino acids that commonly exist in the critters of the Earth and so on, to actually use all of that apparatus that biological evolution has developed for us. But it's undoubtedly the case that that's not the only way to make sort of molecular scale puzzle pieces and so on. And my guess would be that there will be more robust ways to do that that don't have some of the potential fragility and sort of operating constraints of traditional biological kinds of materials. Because, remember, biology, you know, biological evolution is only evolved to deal with organisms that live in this very narrow region on the surface of the Earth and so on and in the oceans and things like this. All right, let's see. Well, there's a question here from rbs. Can you grow life from a computer program? Depends what you mean by life. You know, I have to say my own view of what life is has changed a bit. I think that in ancient Greek times, people would say anything that can move itself must be alive. But then there were steam engines later on, people said anything that can have these chemical effects must be alive. But then there were just explicit chemical reactions discovered that produced those chemicals and so on. Then people talked about self reproduction, the ability to sort of make a copy of the thing you had as being characteristic of life. But then one realized that in computational systems, it's rather easy to make copies of things, and computer worms make copies of themselves all the time. But I think I'm still not really confident in my current view of what life is. But I think that the thing that is pretty unique about life as it exists now is this kind of orchestration of molecular processes. There Are plenty of systems like any old liquid, where there are molecules kind of running around in all sorts of somewhat random and somewhat organized ways. But this, this very kind of structured kind of wire like way in which things are arranged in biological systems seems somewhat unique. Now, is it something where there really has to be sort of explicit orchestration? I don't think so. In fact, the thing that I realized, oh, sometime last year actually was this thing that I call the mechanoidal phase. This thing where you can have a very simple computational system like a cellular automaton made of a bunch of identical kind of idealized cells. And you can just have a particular rule for that. And when you look at what the pattern that's produced by running that rule, what that looks like, sometimes it'll look very simple, sometimes it'll look very random, and sometimes it'll have this look that it has a mechanism that it really means something that even though it just got made from this rule, when we look at it, we say that really looks like somebody on purpose tried to put all these pieces together. And I think that kind of mechanoidal phase is something which is similar to what we're seeing in an actual life, biological life. And I don't fully understand yet kind of how, how to characterize that and how to think about the way in which kind of this orchestration is necessary for things like self reproduction. You know, I do tend to think that perhaps it's confusing when we see these individual organisms that, you know, that then produce another organism and so on. I mean, there is some sense in which life is really just one structure that is this giant tree of life of all the organisms that have ever lived, all connected together, all sort of giving rise to other organisms. And that what one should really think about is the sort of giant sort of graph of all of those processes in all of those organisms. It's something that ranges in scale from the kind of the individual molecules that existed in creatures, you know, a billion years ago, to the things that exist in, you know, whole herds of, of animals running around in the current, in the current world. That, that sort of multi scale thing is the story that there really is only one thing that is life on Earth. It's all part of the same object and there's just been one of it. Now, you know, that kind of raises questions about how did that start? How did one get that thing? How did it get produced? You know, were there sort of some false starts and then this one that really took off. We don't know the Answer to that. My guess is there were false starts, and there was one that really took off, and it's been going for a few billion years. And, well, we'll see. Hopefully it'll keep going for a while longer. So let's see, another biology question here from Chriso. Why are the different colors of flowers but not trees? Why? Questions like that in biology are funny in the way that biologists tend to think about things. The answer is always, it's that way because it's evolutionarily good, because it makes the organism more successful, because it makes the organism have more progeny, which makes more of that kind of organism, and so on. That isn't always the explanation. Sometimes there are physical processes that essentially make a biological thing that make the shell of a mollusk or something. And those physical processes have what we might think of as side effects. Oh, it happens to also make this pattern on the shell. But what's strange about biology is that biology has this tremendous habit of recruiting side effects. If it so happens that when you make this bone shorter, this happens, and eventually that means the creature can stand on two legs and it can reach higher on the tree or something, that thing, which was sort of a side effect of some physics of how bones grew, suddenly gets recruited by biology to be something that's really useful for evolution. So this kind of question of why do things happen and whether did they happen that way because they make the organism fitter and they make it sort of more successful in the kind of struggle for life and evolution? Or did it happen that way just because it's sort of a side effect of other things that are going on? Now, when it comes to the colors of flowers and things like that, it is interesting that there is a very wide range of colors in flowers, and it's. There are. The actual way in which those pigments are made is, you know, there's physics behind what allows those different colors to be made. But there's sort of a question of why have all those different colors? And by the way, the colors extend a bit into the ultraviolet and so on. And when we talk about colors, we humans, when there's an object, light falls on the object. What will happen is different frequencies of light, different wavelengths of light, effectively different colors of light will be emitted from the object. Something like a metal is often mirror, like, so that any frequency of light, any color of light that falls on the mirror, will be reflected from the mirror. But other kinds of objects. Well, an object that is like a blue flower, for example, what happens is white light falls on the flower, white light is a mixture of all different colors of light. But if it's a blue flower, what that means is only the blue light is reflected and the other colors of light are absorbed. But when we humans perceive color, we just break all possible frequencies of light into, oh, that's a red section, that's a green section, that's a blue section. There's actually a whole range of different possible frequencies of light that we humans don't particularly perceive. We only perceive the amount that's in the red section, the amount in the green section, the amount in the blue section. So different flowers, and I don't know the details of this, will have different whole spectra of kind of frequencies of light that they reflect. So we say it's a blue flower, but actually it's a flower with a particular distribution of frequencies. Okay, well, so, you know, so what? Well, presumably what's happening in biology is that flowers. Well, flowers, kind of flowering plants that date from the Cretaceous period, same time as the end of the dinosaurs type thing. Last. Last period of the dinosaurs. Flowering plants have sexual reproduction, and they need a thing like a bee, for example, to go and take a piece of a cell from one plant and go take it to another plant and. Or most of the time they need that. Plants have different mechanisms for getting fertilized, but that's a very common thing. You want a bee to go from this plant to some other plant. And so I think one of the things that tends to happen is you want the bee to go try and fertilize another plant of the same species. Because by definition, if you try and if you take that cell from the. From the first plant of one species, Species are defined by the fact that they can't fertilize. They can only fertilize other members of the species and not other species. That's kind of the definition of what different species are. And so you want the bee to go from a plant of one species to a plant of the same species. And presumably, perhaps, I don't know, the bee is kind of programmed to say, oh, that was a blue flower there that looked like this. Let me. You know, it's color coded. Let me go fertilize the other color of fl. The same color of flower over there. I don't know if that's right, but that seems like a potential explanation for why there will be a diversity of colors, because you're kind of always trying to code in a particular region. You always want the kind of. The flowers of the bees to Go from flowers of the same color, meaning flowers of the same species. Now it's kind of a fun question. If you have a whole bunch of different colors of flowers and you know, if there's a blue flower here and a blue flower, you know, where I am in Massachusetts and there's a blue flower in the tip of South America, doesn't really matter because no bees gonna fly from one to the other. But the question is, if you have a whole sort of field of flowers and some are blue, some are red, some are green, some are yellow, can you, is it, under what circumstances can you sort of arrange the field so that you have kind of, mostly kind of the, the right colors of flowers in a place so that the bee will tend to not get confused. It's kind of a similar problem to if you have a map of the earth and you're trying to fill in colors for countries, you want to figure out how to not have two countries of the same color next to each other. And sort of a result in math that was finally proved in the 1960s, although long, long thought to be true, is a so called four color map theorem that says given a, a map of the Earth on the surface of a sphere, you can, it's sufficient to use only four colors. You can always color any map in such a way that no two adjacent colors, no two adjacent countries have the same color by using at most four colors. And so I kind of wonder what the analogous result is for flowers in kind of a field that will be pollinated by bees. Now I think in the case of trees, the mechanisms of fertilization are a bit different. They're things like they'll have some piece of fruit that falls on the ground, gets eaten by some animal and it excretes them somewhere else, and that will fertilize another tree. I mean, a thing to realize is trees have to be a certain distance apart. And so their kind of methods of getting fertilized will tend to be things that want to have their children not living too nearby. So they'll want things like they'll either be blown by the wind some distance away, they'll have an animal that's going to walk around for a while before it excretes the seed, those, those types of things. And I think, yeah, I should have explained also that the, the way that the plants get sort of fertilized, that flowering plants, they have pollen, pollen has these kind of weird shapes pollen is, is made from. It has a kind of a hard, it's, it's proteins. But Then it has these. Gosh, what are they made of? I think they're aluminosilicates, maybe. I'm not sure. There's kind of these cages that get made in pollen that are very funky shapes. I mean, sometimes they're polyhedral, sometimes they have spiky pieces and so on. And again, that's. That's all, I think, part of the mechanism to make sure that it's sort of only pollen of the right species that gets used in the other plant. So, gosh, let's see. So I think my answer is, I'm guessing that colors of flowers have to do with sort of getting things like bees to be able to recognize which ones to pollinate and hummingbirds and who knows what else, whereas with trees, it doesn't really matter. It's a different story. Now, an interesting question that I absolutely don't know the answer to, and I've wondered about for a long time, is the shapes of leaves on trees. They're quite diverse from an oak leaf to a maple leaf to a pine to a fir tree, a. Et cetera, et cetera, et cetera. There are all kinds of different shapes of leaves. Why are there those shapes? Nobody knows. A little bit is known. There's sort of a biological evolution type theory for the fact that in tropical places it's, you know, when a lot of moisture accumulates on a leaf, it's useful to have a little drip tip at the end where the moisture kind of, kind of gets gets pulled there and then drops off the leaf. So you don't have your leaf covered with water. But beyond that, the shapes of leaves, I don't think anything is really known about why leaves are the shapes they are. Now, that why? Question again, reflects back on sort of evolutionary biology. Why does an oak tree need leaves of that particular shape? Why does that make it more successful? Well, maybe that isn't what's going on. Maybe it's just a side effect of some other feature of how oak trees work, that they tend to have leaves of that shape. That's my guess, actually. And that it really doesn't matter much what the shape of leaves are. It's just kind of a reflection, a side effect of other features of how oak trees work, that they have leaves of that shape. It's something where I've long been curious whether if you look at all the different shapes of leaves that exist on plants of the Earth, whether all possible shapes as generated by some particular method for sort of generating these shapes will be realized or not. I don't know the answer. I've always, I've thought for ages about, about setting up a kind of global leaf project that might be done by, by young kids actually go collect a bunch of leaves. And I suppose I should rethink about this because, you know, I last thought about it probably a decade ago and you know, young kids with smartphones was not so much of a thing. It's now pretty firmly a thing. And it's a question of, you know, you get a leaf, you pick it up, you photograph it and you feed that all into some database somewhere and then you try and sort of pick out the shapes of leaves. And where do these different kinds of leaves grow? What is the distribution of shapes of leaves? I mean, it's something that isn't known and we don't know what would cause it. And we certainly don't know the answer to the why question about why one leaf is one shape and one is another. So that's a, you know, that's sort of a boundary to knowledge. Now I must say that in the case of flowers, I think somebody at our high school summer program maybe a couple years ago took a bunch of pictures of flowers project. I suggested take a bunch of pictures of flowers and just try to see how the different colors of those flowers are distributed in color space. And I'm trying to remember what she concluded. One has to look it up. I think my impression is that it was not uniform in color space. Now what would it mean to be uniform in color space? Because the way we assign colors to things, and this is what I was sort of explaining earlier, is very human centric. We say how much red, how much green, how much blue. Actually, to get a more uniform color space we usually use coordinates xyz, which are essentially the amount of excitation of our red, green and blue color cone cells in our eyes. It's slightly more complicated thing, but it's roughly the amount of red, green or blue. But it is not obvious that a uniform distribution in red, green, blue space corresponds in any particular way to a uniform distribution in the space of colors as seen by bees. For example. Bees have a different visual system. Bees see some distance into the ultraviolet. I'm not sure how many kinds of color receptors bees have. I'm kind of guessing it's more than our three, but I don't know how many. The all time winner I believe is the mantis shrimp, which has like a dozen or 15 different kinds of color receptors, which presumably is important for recognizing different kinds of plant and animal life. You know, in the ocean that the shrimp finds important for what it's going to eat and things like this. Okay, I'm seeing lots of questions here about things like this. Well, I see Paula asking what causes a four leaf clover? Why are they so rare? Well, okay, so there's this whole phenomenon of so called phyllotaxis, just Greek for kind of the making of leaves and so on. It's when a plant grows it. So well, different kinds of plants grow in slightly different ways, but there is a region of kind of active growth. So okay, big difference between plants and animals. Plants have rigid cell walls, animals do not. So once you've laid down a cell in a plant, it's got. Oh my gosh, what is it called? Begins with C. Oh my gosh, forgot the word. The, the kind of scaffolding of plant cells. Cellulose is not the thing. Well anyway, there are, there are rigid cell walls for plants. Animal cells don't have rigid cell walls. If you have rigid cell walls, that means that whenever you grow you can never kind of squidge around what you already grew. So for example, that means growing in a tree like shape is a good deal because you can grow one little piece of the tree. You grow out the branches, you grow out more branches, you grow out more branches. You can fill a lot of space with that. In an animal you can sort of start growing and you can. It happens very commonly it happens even in the very beginning of all of us animals when from a fertilized egg cell just starts growing more cells and they just keep on sort of coming in and they keep on sort of squidging around and they eventually make this kind of sphere of cells and so on. And a new cell, cells divide, a new cell is produced, but everything is soft. So the cells can kind of push on each other and just make a whole sphere. That doesn't work with plants. I guess it is cellular cell walls. Once a, a plant cell is there, it's rigid. So you have to kind of in your strategy for growth, you have to take account of that. So the typical way a plant grows is there'll be some growing tip of the, of the, of the stem of a plant. It's called the apical meristem. It's a particular collection of cells that are dividing at the tip and they're where how the plant grows. And for example, I think if you. The heart of palm that is the, that's on a palm tree. It kind of just grows up the trunk and it's always just sprouting leaves at the top. That region at the top has this apical meristem, which is a very kind of. Very kind of rich thing, sort of full of nutrient that you can get in little cans and eat. But that's the sort of special cells that have this division capability in plants, okay? So when there is this area where they're dividing cells, that's where the plant organs, things like leaves, they come out from that area. And so what's happening is when the plant is growing, the clover is growing, whatever else, eventually it will grow this. It'll be filling in cells, and they'll be making these rigid walls. And then there'll be this moment where it decides for some reason that it will kind of make a sprout. You know, it'll make a leaf or something. Well, the question of how does it decide where to make that leaf? Well, it decides based on plant hormones. Auxins. Boy, I'm having a difficult day with words here. Plant auxins, which are the analog of our kind of growth hormones. And what happens is. And for example, if you do tree surgery and things, you can inject hormones into a tree to get it to produce a sprout in a particular place, to produce a branch in a particular place. But what's basically happening is the decision of where the sprouts will come out is determined by the concentration of this plant hormone. And so what happens is, where there is lots of hormone, you'll get a sprout. But once the sprout has been produced, it uses a bunch of hormone to produce that sprout. And so the hormone will be depleted in that place. And so there won't be another sprout immediately in that area. So then if you look at a kind of a growing stem, there'll be sort of around the rim of the stem, there'll be a certain density, certain amount of plant hormone in different places around this rim. And as the plant kind of grows, it will kind of get transported up vertically, up sort of the stem of the plant. Okay? Every time a sprout comes out, you deplete the plant hormone on that side. And then what happens is the next sprout, and maybe you. You're building up more and more plant hormone. And when the plant hormone gets above some critical value, then another sprout comes out. So an interesting math fact which I rediscovered in the 1990s, I guess it's been discovered. This fact has been discovered, like, oh, probably 30 times since probably the early part of the 20th century. It turns out that in this kind of setup, where you have this kind of ring of concentration of something like plant hormone. And you keep on removing a piece of plant hormone on one side. And then you ask, well, where does the plant hormone build up? It turns out that the way it works is it will on average, the place where the next sprout will come out is 137.5 degrees away on the circle from where the previous sprout came out. And that's a fact that you can deduce from just doing a bit of math. And it's related to the golden ratio and so on. But what happens is that's kind of the typical angle between successive sprouts on a stem of a plant. So, for example, if you look at a palm tree, you'll see that the leaf scars on the trunk of the palm tree, you can see them going around in kind of a spiral. And that spiral will have the feature that if you look at sort of where does the next one show up? Kind of around the trunk, it's 137 degrees away. And the same thing happens if you look at a daisy, for example, and you look at the little stamens in the center, little yellow things in the center. You kind of follow them back to where they were produced from the stem of the plant. You'll see the same exact phenomenon. This 137.5 degrees. And that fact, that golden ratio thing means that when you start counting where, how many of these things there are in a particular region, you get Fibonacci numbers. 1, 1, 2, 3, 5, 8 numbers where a particular Fibonacci number is the sum of the two previous numbers in the sequence. And that. And the ratio of successive numbers eventually becomes the golden ratio, about 1.618 square root of five plus one over two. The. And that, that's all related to the angles in these plants. But the, the cause of it is this phenomenon of you just have sort of plant hormone building up, getting, getting depleted on one side. And then you kind of are looking at the effect of many different depletions. And quite quickly, that sort of, it, it goes to a steady state where the kind of. The next place where there's the least depletion is 137.5 degrees away. If you look at different kind, whether it's a pine cone, whether it's a strawberry, all these kinds of things, you'll find the same angle all over the place. Okay, so that's the most typical form of phyllotaxis is this Fibonacci spiral. Phyllotaxis but there are other forms of phyllotaxis that happen. There are some plants where you have the sprout of the plant and it will just produce leaves on opposite sides. By the way, people make a big thing out of the fact that if you have this 137.5 degree thing, that means that the higher leaves, well, the lower down leaves are minimally shaded by the higher leaves. But it's one of these things that's very complicated in biology because. Yes, that's true, but the math that leads to that is identical to the math that leads to the statement about how depletion of hormone works and so on. So the question is, what came first? And my strong guess would be this depletion of hormone thing came first. That's the mechanism, so to speak. But then the plant kind of said, goody, that means my lower leaves don't get as shaded from the sun. So it's all good. And there's no big change in how the plant actually grows. But so the thing that presumably happens in a clover, for example, is that you're building up plant hormone and at some point the way the thing works is it produces some of these sprouts. Now, I think it's complicated because this little place where there's this little growing tip tends to be a polyhedral shape and it tends to actually have cells of particular shapes and they divide in a particular way and they make. And so my guess is that in a typical clover that there's actually some actual shape of cells in the little growing tip that divides there. And my assumption would be that it is that what happens is most of the time the total amount of plant hormone that is needed to get any sort of leaves to come out. When it gets above that threshold, it produces, you know, three leaves. But just sometimes it has enough extra that another leaf gets produced. There's enough plant hormone to produce another leaf. My guess is that's how it works. Now, could one compute what the fraction of sort of four leaf clovers should be? Probably in some first approximation. Probably, yes. By looking at kind of the. Well, I'm not sure, but looking at features of this plant hormone, I mean, it's worth remembering that in biology, sort of everything is approximate in some sense. And biology has this feature that it often tries to turn the continuous into the discrete. So for example, let's say you're making a zebra and it's going to have black and white stripes. Well, what happens is that there's a, in the growing zebra and the Zebra embryo. There is particular chemicals that diffuse through the zebra embryo and they also react with other chemicals and those produce chemical waves. The chemical waves go up smoothly and down smoothly and so on. But there's a particular mechanism in the growing zebra that switches on black pigment or switches it off. It's usually these things called Hox Genesis that are particular just switch basically kind of protein that acts as a switch where it says if there's more than this concentration of this hormone like substance, then start producing protein. And if there's less than that, then don't produce protein. That's a pretty common mechanism in biology. Above a threshold, something discretely starts to happen. Below a threshold that doesn't. And so that's the kind of thing that leads to, oh, you know, by the time you have this amount of something to make something happen, usually it'll make three of it, but just sometimes it'll make four of it. It's same thing with fingers, for example. Most people end up with five, some end up with six. And it's rare, but it happens. And it happens because when there's this kind of switching that's going on, it is the switching. Something was a little different in the concentrations. You know, the concentration got a little bit higher before the switching started to happen or something like that. And so you ended up with something which wasn't exactly five, it was only approximately five. Okay, let's see. Oh my gosh, there's. Kephala says, says the mantis shrimp apparently has 12 types of cone cells in Thai. Do I have an intuition about what the space of all these colors occupy in the brain of the animal? It's a very interesting question, you know, imagining what we can't imagine, so to speak. Imagine what? Imagining what we have never experienced is really hard. For example, you know, it's perfectly possible to have infrared cameras. It's perfectly possible to sense distributions of light frequencies beyond what we normally pay attention to of red, green, blue. And you know, with the coming generation of augmented reality, virtual reality kind of headsets and so on, we'll have every opportunity to take what is usually in our visual field with usual colors and add in, oh, we've got a sensor that can also tell us about the infrared, the kind of, the heat, for example, coming off some particular object. But the question is, how do we, how do we project that into our brains? How do we, how do we, you know, we're using up our brains, know about these three colors, rgb basically, if we've got more colors to tell our brains about how do we show that to our brains? Because our brains have, you know, in our primary visual cortex, we've got cells that are responding to red and green and blue and so on, starting from the cells in our retina that respond to those colors of light. And the question is, you know, how do we feed in another seven colors, for example, you know, what would we do? I don't know. And I think as we go to higher levels in visual processing, it's kind of like what does it seem like to us now at some level, what we're doing in our visual processing is always to try and sort of reduce information. I mean, when we get that first image on our Retina, we've got 10 million different sort of points of light on our retina, yet we don't. As we look around, we're not saying, oh my gosh, there's 10 million things coming in here. We're saying, there's an object, it's a camera looking at me. You know, whatever it is, it's an object. I'm not distinguishing all of the individual pixels that are part of that camera. My brain is taking this kind of higher level object and that's what I'm kind of taking out of it. And I suspect for the mantis shrimp, it's kind of the same thing. It says there's a fish of a particular kind. You know, that's a fish I can eat, that's a fish I can't eat, that kind of thing. And it's kind of the brain of the mantis shrimp is probably taking all those colors and kind of, you know, pulling it up into this kind of summary. That's a summary of objects. Now, you know, the interior design of the mantis shrimp house, so to speak. That's a different question. You know, what goes with what color for the mantis shrimp? What is the kind of, what's the kind of almost subconscious reaction to colors that it would have? Same with us. I mean, again, I think it's a very interesting question how we take additional sensory data, even like infrared or ultraviolet, and kind of mix it with the data that we already have. And how do we build up a model, a sort of perception model that incorporates those additional kinds of elements of input. Let's see, comment from memes here. There's been a lot of cool research in regard to photosynthesis recently. Do I have something to say about that? Photosynthesis. So photosynthesis is the process of chloroplasts and plants, and particularly the protein chlorophyll Being able to take light from the sun and being able to essentially produce energy from that, or more particularly produce kind of, it has a electron cascade that it produces. That is what leads to kind of chemical reactions that can produce energy for the plant. People didn't understand for a long time how chlorophyll could possibly work and what seems to be the case, and I don't know the details of this, but what seems to be the case is that there are subtle sort of quantum mechanical effects which are leading to the possibility of, of photons of light producing the effects that they have on chlorophyll. And my impression is that what's happening is there is an atom of, oh gosh, I should know. Some metal, I think. Which one is it? Not cobalt, I think maybe manganese perhaps. Anyway, some metal that's held in this little tiny cage inside the protein that is the chlorophyll and that somehow that has some kind of very sort of amplified interaction with the photons of light. I think it's a little bit like the way that a, an oled organic light emitting diode displays work where they are kind of exciting atoms that are held in kind of a, in that case in a solid state, kind of in a little kind of trap produced by sort of the way the semiconductor crystal works. I think it's a sort of the inverse kind of thing in some way in chlorophyll that it's kind of the, the photon that's coming in kind of has a sort of a outsized kind of resonant effect on, on this particular atom. And that, that's sort of a key part of, of how chlorophyll works. And it's something where sort of the most obvious calculation would say it shouldn't be able to work. Now, you know, interesting question. Can we sort of use chlorophyll or something which has the idea of chlorophyll to make a better photovoltaic? A photovoltaic is something that's using these days, semiconductors to have, you know, photon of light comes in, electron goes out, that makes an electric current that, that allows you to produce electricity from light. Is there a way to use some idea that comes from the way that chlorophyll works to make a better one of those? I think the general thought is that there may very well be, but we don't know what it is yet. All right, let's see. Gosh, Prabh is asking what's the difference in a species and A variety. How do you know if something is the main species or its variety? I think, you know, what is the species? Well, the traditional definition of a species is you can mate if you're the same species. You can't mate if you're different species. But it's very fuzzy, like for example, horse and donkey meat to produce a mule, but the mule can't then mate with anything, is then not capable of having mule babies. That's right on the edge that horse and donkey are kind of close enough together that they can still successfully mate. And I don't know to what extent. It's really a question of like with a pollen and so on, of what fits in what type thing and to what extent it's a, you know, at a macroscopic, a large scale or at a molecular scale how that works. But that's sort of the traditional definition of species is that they are sort of reproductively isolated, that the, you know, within the species they're all mating with each other, but they can't mate across different species. Now varieties tend to be our breeds and so on tend to be sort of subparts of the species where you get sort of different characteristics of the species, but mating is still possible. So for example, for dogs, you can breed dogs. So you get a giant St. Bernard or a tiny poodle. And now bad things happen. If you mate a male Saint Bernard with a female poodle, I think the results are not good because I think the growing fetus just sort of gets too big and it's bad news. So. But it does manage to produce a, you know, a growing little dog fetus. It just doesn't sort of mechanically work. But I think the, the thing that, that's again, different breeds of dog have been sort of separated off. They have different characteristics, but they still have enough. They still have the same genetics, they still have the same things necessary to. Because one of the things that has to happen in reproduction is that the DNA of the kind of the mother and father organisms have to be kind of matched up. And if that doesn't work, then it's kind of game over in terms of making the sort of the, the, the new organism. Because it's going to, it's going to use some stretches of DNA from the mother, some stretches of DNA from the father on every chromosome. There's maybe only about five crossovers. So it's big stretch from mother from father and so on. And if it's different species where the DNA has been kind of mixed up, then that's just not going to work, you're just not going to have the, you're going to miss, you know, a big chunk of DNA that was the thing necessary to make a tail or something like this. And so that's, I think the reason that doesn't work there. But of course the St. Bernard vs. Poodle, that's also encoded on the DNA, but somehow the DNA hasn't sort of moved around enough that it can't get matched up to make a viable organism, I suppose. I mean, it's also the case that in, You know, varieties, breeds and so on can be very different in appearance, yet their sort of core genetics is somehow similar enough to allow this kind of reproduction to work. And I guess, you know, the question would be at what point do you, when you breed things separately, at what point do you get critters that form different species? And that's not a well understood thing. And I think, I'm sure there are by now experiments on that. But I don't think it's, I don't think there's a good sort of theory of how that works. And it's like, well, how many, you know, how many changes do you have to have before you have that? Now again, there can be these purely sort of mechanical things like the Saint Bernard is big and the poodle is small and that doesn't work for mating very well. And so there can be things like that that lead to kind of reproductive isolation independent of the fact that the DNA doesn't match up. But I think, Yeah, I think this whole question about when's it a different species and when's it a merely a, a different breed, different variety, it's a tricky thing. I mean, for plants, for example, there can be very fixed genetic kind of, you know, very fixed genetics across a very large number of plants. They're sort of cuttings that can go from one plant to another. And for example, for us humans, we can even detect that rather well because, for example, I don't know, if you look at something like apples, you know, a given variety of apple, you know, a Golden Delicious, I like those ones is a very specific genetic sequence. And they're all kind of identical. All those apple trees are identical enough that they basically are making this using the same genetic sequence. If you change the genetic sequence, we, our taste system will detect that and we'll say that's not a Golden Delicious apple that tastes different. And so that's a. And that causes all kinds of issues. Like with bananas, that's causes all kinds of issues because there's kind of this very specific clone, this very specific thing that got infected with various kinds of diseases and parasites and so on, and then all those ones died off. And it's a whole, whole long tale of whoa. But it's interesting that we sort of, our taste system is sensitive enough to pick up those kinds of things. Okay, I think I need to get going here. There's one last question from Kephalis here. Could it be possible to disable some kind of cone cell in our eye and thus lead someone to perceive super color? Something that activates the other two types of cone cells while not activating the other type in a way that is not physically possible. The problem is we've got three kinds of cone cells. If one of them isn't working, then you are, for example, red, green, colorblind. It turns out the females of the species sometimes have two different kinds of blue cone cells which can lead to kind of super perception of color. Because if their brains, if their brains will wire up for this because that is carried on the X chromosome and females have two X chromosomes, that means that there can be two different kinds of blue cone cells and that can lead to this kind of super super perception. Normally if you remove a type of cone cell, you'll just have some kind of colorblindness. Now, interesting question. There are gene therapies that try to add back in retinal cells to treat various kinds of genetic diseases. So it's a very amusing question whether we could have a genetic something that is put in there by, let's say, a retrovirus or by a gene editing CRISPR Cas9 mechanism or something like this that is an additional kind of cone cell and where we could sort of be, you know, be just growing some new kind of cone cell in our eyes. Problem is, you have to have the whole system be sort of updated for that to be there. Because you've got the fibers in the optic nerve and they're going to the primary visual cortex. And it's not obvious that even if you have a new kind of cone cell, it's not going to get wired up in the adult brain. It's going to be really difficult to wire that up. Maybe, for example, below age of 5 or 6 or something. Even if, for example, you get amblyopia where one eye has not been operating properly and the brain only programs itself for the eye that is working properly. If you patch the eye that, if you fix the optics of the eye and then patch the eye that was working properly, the brain at that age will relearn how to see with both eyes, so to speak. And so probably at that age, if you were to inject kind of the additional cone cell gene editing kind of thing, you might be able to successfully wire up the brain to get kind of super perception of color. And. But I think, you know, already this phenomenon happens with, I don't know, what fraction of the females of the species, so to speak. And, you know, the main consequence of that tends to be sort of heightened color awareness. And perhaps, you know, when it comes to human language, people say, oh, that's a blue, that's a green, that's a whatever. You know, an interesting question would be, you know, is there a kind of additional sort of way of describing colors that could be communicated between folks who have, you know, those additional cone cells? And now the whole story of blue and green and the history of different color words for blue and green and the fact that those were not distinguished in, for example, ancient Greek and so on is a whole different story. And one that would be interesting to trace. And perhaps one would find if there was something where there was a language that was mostly spoken by women, like the hiragana script in Japanese, for example, mostly used by women, maybe one would find some bizarre phenomenon where there are additional color words associated with the fact that many people in that population were able to perceive more detailed colors. All right, you've gotten me to speculate on lots of kinds of things, and I have to go because I'm late for something else. But thank you for all those interesting questions and look forward to talking to you another time. Bye for now.