If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Variation in a species

Natural selection relies on genetic variation within a species. Mutations alone aren't sufficient for creating diversity; sexual reproduction plays a crucial role as well. By combining genetic material from two parents, offspring inherit a unique mix of traits, resulting in vast potential for variation. This process allows for more efficient adaptation and survival, making sexual reproduction a widespread biological strategy. Created by Sal Khan.

Want to join the conversation?

  • duskpin ultimate style avatar for user Emily Taylor
    With so many possible combinations of chromosomes, it seems unlikely that siblings would look similar. How does this happen?
    (20 votes)
    Default Khan Academy avatar avatar for user
    • leaf yellow style avatar for user Aaron Yip
      Each of those genes govern one small characteristic (that's similar to either from your mom or your dad). On average, you'll be getting about half from each. So let's use a hypothetical, really simplified example: there's 4 genes that control hair for curliness, color, texture, and density. Your mom's hair is very curly, brown, smooth, and thick. Your dad's hair is not curly, blonde, sandy, and thin. Your hair could be very curly, blonde smooth, and thin. Your sister's hair could be very curly, blonde, sandy, and thick. That sounds pretty similar.

      Secondly, dominant traits also come into play. Some visible physical traits are much more common even if you're genetically different on the inside.

      And most importantly, you guys are both taking genes from the same two pools. There are countless more variants for curliness, hair colors, textures, and thickness and all the other genes that you won't have any chance of inheriting if your parents don't have them. So compared to all the other people outside the family, siblings usually look pretty similar. I mean, from an elephant's perspective, all humans probably look exactly alike. ;)
      (19 votes)
  • mr pink red style avatar for user Paul Grieselhuber
    Was there sexual reproduction (or some other version of DNA transfer and recombination) before the advent of male & female 1.4 billion years ago?
    (8 votes)
    Default Khan Academy avatar avatar for user
  • aqualine ultimate style avatar for user ITPro4470
    I noticed a correction as in the video. The correction says 'Sal says "gender" but means "sex"'
    What is the difference between gender and sex?
    (7 votes)
    Default Khan Academy avatar avatar for user
    • piceratops ultimate style avatar for user Just Keith
      In this context, "sex" refers to the physical traits about being male or female or otherwise -- such things as genetic distinctions, differences in reproductive organs, and sexual dimorphism (in species where that occurs).

      "Gender" refers to the psychological, social, and/or culture aspects about sexual identity and roles.
      (16 votes)
  • aqualine ultimate style avatar for user Sarvesh Parakh
    I have another question. He said there are 3 billion codes for the DNA. This is very complex. But if humans could interpret them with some high technology computers and change these DNA codes using some biogenetic stuff or chemical combinations even before the baby is born, is it possible to bring out desired results? And if during a test-tube baby, the sperm is genetically changed to have some other characters instead of what it has, and them the ova is fertilized under observation where necessary changes are brought somehow, then can the baby have that particular character? For eg., if the doc makes changes in the DNA structure of the sperm and inserts the gene for high IQ removing the original gene of low IQ, then is it possible that after fertilization if the sperm's genes remain dominant in the IQ subject, that the baby born will have high IQ? If this is possible, we can have a a new studying stream called baby engineering!!
    (7 votes)
    Default Khan Academy avatar avatar for user
  • aqualine ultimate style avatar for user Shrey Mahey
    Do we have the same sets of chromosomes in every cell? I mean, just one cell can't identify whether you have curly or straight hair, can it? In each cell, if there are 23 chromosome pairs, are they the same in every cell or does every cell have different pairs that "code" something else in the body (because 23 pairs of chromosomes can't really make you up, even if the same type of pairs are in every cell)? I'm a bit confused, and would like some help.
    (6 votes)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user hadley.justin
      Yes, we have the same chromosomes in every cell with the exception of our reproductive cells. In most cells in the human body there are 23 chromosome pairs, and in reproductive cells there are 23 single chromosomes. Each cell in your body has the genetic information for the makeup of your entire body. So even the cells at the end of your big toe have the encoded information for the curly, or straight hair on the top of your head. So yes, 23 pairs of chromosomes really can contain the information for the makeup of your entire body. In fact there is a significant amount information encoded in your DNA that is essentially switched off, because it is no longer needed. So those 23 chromosomes not only include the information for your entire body in each and every cell, they contain some information about your ancestry as well.
      (9 votes)
  • leafers seed style avatar for user christina.marie.loew
    Ok, so hypothetical question. Say you and your spouse have the maximum number of combinations of kids. If you have one more kid, then it would be a repeat of a combination. Does that mean this kid would be exactly the same as the other kid who had the same combination? Or would they still be different?
    (6 votes)
    Default Khan Academy avatar avatar for user
    • leafers tree style avatar for user Alva Wei
      Even if two of my however many quadrillion children happened to be genetically identical at fertilization, they would not physically identical. They wouldn't even be genetically identical for long after fertilization. This question could also be asked about identical twins. Since one zygote divides into two embryos in the case of identical twins, shouldn't they be, well, identical? But during embryonic development, when the cells are dividing and their DNA is almost constantly being replicated, hundreds of mutations will occur and accumulate. Also, environmental differences will cause changes in each individual's epigenetics - changes in gene expression not controlled by mutations or other changes in the DNA itself. The mutations and epigenetic changes will result in two different individuals. They will only be slightly different physically, but they will still be different.
      (7 votes)
  • piceratops ultimate style avatar for user Evan
    How did variations originally come about: i.e. how could humans develop black hair if thousands of years ago lets say all humans had brown hair. How could a gene for black hair develop?
    (7 votes)
    Default Khan Academy avatar avatar for user
  • starky ultimate style avatar for user Demster
    If sexual reproduction can only use current and existing genes from both parents, how do we account for "new" traits, and not just novel gene combinations?
    (5 votes)
    Default Khan Academy avatar avatar for user
    • piceratops ultimate style avatar for user Just Keith
      New genetic traits occurring that are not the result of novel gene combinations happen by mutations. These mutations happen as the gametes are forming. This is not all that unusual, most gametes have a few mutations. Most of the time the mutation has little or no effect on the phenotype. When there is an effect that matters, it is usually harmful. But, once in a while, the mutation is beneficial.

      There are many kinds of mutations. Let me (greatly over-simplifying) mention one class of mutation that is more likely (not always of course) to lead to a beneficial change in traits. This is gene duplication.

      In gene duplication the offspring gets an extra copy of a gene, perhaps even on a completely different chromosome. Sometimes this is catastrophic, but a lot of times it can enhance an existing trait or simply just provide a backup copy of an existing gene. Gene duplication is a very common mutation and there are many genes we have extra copies of.

      Now, if we have a spare copy of a gene, then some of the time that allows a major mutation to occur in one of the copies without causing a catastrophic problem. Now, in particular, if one of the copies undergoes what is called a frameshift mutation, then that gene will now code for a completely different protein. This might do very little, might be harmful, or might actually cause a new trait to appear rather suddenly. There are documented cases of this kind of mutation.
      (7 votes)
  • blobby green style avatar for user Rich
    If my cells contain chromosomes that contain the information passed down from both of my parents, is it possible for me to pass down a characteristic that I did not inherent? For example, I inherited black hair from my mother but my father has blonde hair. Since my chromosomes contain that information, can my offspring inherit the blonde hair trait?
    (7 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Luis Aragon
    how does variation work
    (5 votes)
    Default Khan Academy avatar avatar for user
    • leaf orange style avatar for user meilibatsy
      The main source of almost all variation is mutations, which are random copying errors in DNA.
      In sexually reproducing species, the formation of offspring requires a set of genes from each parent, so the parents' genes (including mutations) are mixed up, leading to more variation.
      (4 votes)

Video transcript

The whole process of natural selection is to some degree dependent on the idea of variation, that within any population of a species, you have some genetic variation. So, for example, let's say I have a bunch of-- well, this is a circle species, and one guy is that color, and then I got a bunch more, maybe some are that color-- oh, that's the same color-- that one, and that one, and that one. And for whatever reason, sometimes there are no environmental factors that will predispose one of these guys to be able to survive and reproduce over the other, but every now and then, there might be some environmental factor, and it makes maybe, all of a sudden, this guy more fit to reproduce. And so for whatever reason, this guy is able to reproduce more frequently and these guys less frequently. And some of them get killed, or whatever, or eaten by birds, or whatever, or they're just not able to reproduce for whatever reason, and then maybe these guys are something in between. So over time, the frequency of the different traits you see in this population will change. And if they are drastic enough, maybe these guys start becoming dominant and start not liking these guys, because they're so different or whatever else. We could see a lot of different reasons. This could eventually turn into a different species. Now, the obvious question is what leads to this variation? In a population what leads to this-- in fact, even in our population, what leads to one person having dirty blonde hair, one person having brown hair, one person having black hair, and we have the spectrum of skin complexions and heights is pretty much infinite. What causes that? And then one thing that I kind of point to, we talked about this a little bit in the DNA video, is this notion of mutations. DNA, we learned, is just a sequence of these bases. So adenine, guanine, let's say I've got some thymine going. I have some more adenine, some cytosine. And that these code, if you have enough of these in a row, maybe you have a few hundred or a few thousand of these, these code for proteins or they code for things that control other proteins, but maybe you have a change in one of them. Maybe this cytosine for whatever reason becomes a guanine randomly, or maybe these got deleted, and that would change the DNA. But you could imagine, if I went to someone's computer code and just randomly started changing letters and randomly started inserting letters without really knowing what I'm doing, most of the time, I'm going to break the computer program. Most of the time, the great majority of the time, this is going to go nowhere. It'll either do nothing, for example, if I go into someone's computer program and if I just add a couple of spaces or something, that might not change the computer program, but if I start getting rid of semicolons and start changing numbers and all that, it'll probably make the computer program break. So it'll either do nothing or it'll actually kill the organisms most of the time. Mutations: sometimes, they might make the actual cell kind of run amok, and we'll do a whole maybe series of videos on cancer, and that itself obviously would hurt the organism as well as a whole, although if it occurs after the organism has reproduced, it might not be something that selects against the organism and it also wouldn't be passed on. But anyway, I won't go detailed into that. But the whole point is that mutations don't seem to be a satisfying source of variation. They could be a source or kind of contribute on the margin, but there must be something more profound than mutations that's creating the diversity even within, or maybe I should call it the variation, even within a population. And the answer here is really it's kind of right in front of us. It really addresses kind of one of the most fundamental things about biology, and it's so fundamental that a lot of people never even question why it is the way it is. And that is sexual reproduction. And when I mean sexual reproduction, it's this notion that you have, and pretty much if you look at all organisms that have nucleuses-- and we call those eukaroytes. Maybe I'll do a whole video on eukaryotes versus prokaryotes, but it's the notion that if you look universally all the way from plants-- not universally, but if you look at cells that have nucleuses, they almost universally have this phenomenon that you have males and you have females. In some organisms, an organism can be both a male and a female, but the common idea here is that all organisms kind of produce versions of their genetic material that mix with other organisms' version of their genetic material. If mutations were the only source of variation, then I could just bud off other Sals. Maybe just other Sals would just bud off from me, and then randomly one Sal might be a little bit different and whatever else. But that would, as we already talked about, most of the time, we would have very little change, very little variation, and whatever variation does occur because of any kind of noise being introduced into this kind of budding process where I just replicate myself identically, most of the time it'll be negative. Most of the time, it'll break the organism. Now, when you have sexual reproduction, what happens? Well, you keep mixing and matching every possible combination of DNA in a kind of species pool of DNA. So let me make this a little bit more concrete for you. So let me erase this horrible drawing I just did. So we all have-- let me stick to humans because that's what we are. We have 23 pairs of chromosomes, and in each pair, we have one chromosome from our mother and one chromosome from our father. So let me draw that. So I'll draw my father's chromosomes in blue. So I have 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15-- I'm running out of space. Let me do more here-- 16, 17, 18, 19, 20, 21, 22, and then I'll throw another one here that looks a little bit different. I'll throw one here that looks like a Y, and we'll talk more about the X's and the Y chromosomes. Then I have 23 chromosomes from my mother. And not to be stereotypical, but maybe I'll do that in a more feminine color. Let's see, so I have 23 chromosomes from my mother. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23. So what's going on here? I have 23 from my mother. I have 23 from my father. Now, each of these chromosomes, and I made them right next to each other. So let me zoom in on one pair of these. So let's say we look at chromosome number-- I'll just call this chromosome number 3. So let me zoom in on chromosome number 3. I have one from my mother right here. Actually, maybe I'll do it this way. Remember, chromosome is just a big-- if you take the DNA, the DNA just keeps wrapping around, and it actually wraps around all these proteins, and it creates this structure, but it's just a big-- when you see it like that, you're like, oh, maybe the DNA-- no, but this could have millions of base pair, so maybe it'll look something like that. It's a densely wrapped version of-- well, it's a long string of DNA, and when it's normally drawn like this, which is not always the way it is, and we'll talk more about that, they draw it as densely packed like that. So let's say that's from my mother and that's from my father. Now, let's call this chromosome 3. They're both chromosome 3. And what the idea here is that I'm getting different traits from my father and from my mother. And I'm doing a gross oversimplification here, but this is really just to give you the idea of what's going on. This chromosome 3, maybe it contains this trait for hair color. And maybe my father had-- and I'll use my actual example. My father had very straight hair. So someplace on this chromosome, there's a gene for hair straightness. Let's say it's a little thing right there. And remember, that gene could be thousands of base pairs, but let's say this is hair straightness. So my father's version of that gene, he had the allele for straightness. And remember, an allele is just a version of a gene, so I'll call it the allele straight for straight hair. Now, this other chromosome that my mother gave me, this essentially, and there are exceptions, but for the most part, it codes for the same genes, and that's why I put them next to each other. So this will also have the gene for hair straightness or curlyness, but my mom does happen to actually have curly hair. So she has the gene right there for curly hair. The version of the gene here is allele curly. The gene just says, look, this is the gene for whether or not your hair is curly. Each version of the gene is called an allele. Allele curly. Now, when I got both of these in my body or in my cells, and this is in every cell of my body, every cell of my body except for, and we'll talk in a few seconds about my germ cells, but every cells other than the ones that I use for reproduction have this complete set of chromosomes in it, which I find amazing. But only certain chromosomes-- for example, these genes will be completely useless in my fingernails, because all of a sudden, the straight and the curly don't matter that much. And I'm simplifying. Maybe they will on some other dimension. But let's say for simplicity, they won't matter in certain places. So certain genes are expressed in certain parts of the body, but every one of your body cells, and we call those somatic cells, and we'll separate those from the sex sells or the germs that we'll talk about later. So this is my body cells. So this is the great majority of your cells, and this is opposed to your germ cells. And the germ cells-- I'll just write it here, just so you get a clear-- for a male, that's the sperm cells, and for female that's the egg cells, or the ova. But most of my cells have a complete collection of these, and what I want to give you the idea is that for every trait, I essentially have two versions: one from my mother and one from my father. Now, these right here are called homologous chromosomes. What that means is every time you see this prefix homologous or if you see like Homo sapiens or even the word homosexual or homogeneous, it means same, right? You see that all the time. So homologous means that they're almost the same. They're coding for the most part the same set of genes, but they're not identical. They actually might code for slightly different versions of the same gene. So depending on what versions I get, what is actually expressed for me, so my genotype-- let me introduce another word, and I'm overwhelming you with words here. So my genotype is exactly what alleles I have, what versions of the gene. So I got like the fifth version of the curly allele. There could be multiple versions of the curly allele in our gene pool. And maybe I got some version of the straight allele. That is my genotype. My phenotype is what my hair really looks like. So, for example, two people could have different genotypes with the same-- they might code for hair that looks pretty much the same, so it might have a very similar phenotype. So one phenotype can be represented by multiple genotypes. So that's just one thing to think about, and we'll talk a lot about that in the future, but I just wanted to introduce you to that there. Now, I entered this whole discussion because I wanted to talk about variation. So how does variation happen? Well, what's going to happen when I-- well, let me put it this way. What's going to happen when I reproduce? And I have. I have a son. Well, my contribution to my son is going to be a random collection of half of these genes. For each homologous pair, I'm either going to contribute the one that I got from my mother or the one that I got from my father, right? So let's say that the sperm cell that went on to fertilize my wife's egg, let's say it happened to have that one, that one, or I could just pick one from each of these 23 sets. And you say, well, how many combinations are there? Well, for every set, I could pick one of the two homologous chromosomes, and I'm going to do that 23 times. 2 times 2 times 2, so that's 2 to the twenty third. So 2 to the 23 different versions that I can contribute to any son or daughter that I might have. We'll talk about how that happens when we talk about meiosis or mitosis, that when I generate my sperm cells, sperm cells essentially takes one-- instead of having 23 pairs of chromosomes in sperm, you only have 23 chromosomes. So, for example, I'll take one from each of those, and through the process of meiosis, which we'll go into, I'll generate a bunch of sperm cells. And each sperm cell will have one from each of these pairs, one version from each of those pairs. So maybe for this chromosome I get it from my dad, from the next chromosome, I get it from my mom. Then I donate a couple more from-- I should've drawn them next to each other. I donate a couple more from my mom. Then for chromosome number 5, it comes from my dad, and so on and so forth. But there's 2 to the twenty-third combinations here, because there are 23 pairs that I'm collecting from. Now, my wife's egg is going to have the same situation. There are 2 to the 23 different combinations of DNA that she can contribute just based on which of the homologous pairs she will contribute. So the possible combinations that just one couple can produce, and I'm using my life as an example, but this applies to everything. This applies to every species that experiences sexual reproduction. So if I can give 2 to twenty-third combinations of DNA and my wife can give 2 to the 23 combinations of DNA, then we can produce 2 to the forty-sixth combinations. Now, just to give an idea of how large of a number this is, this is roughly 12,000 times the number of human beings on the planet today. So there's a huge amount of variation that even one couple can produce. And if you thought that even that isn't enough, it turns out that amongst these homologous pairs, and we'll talk about when this happens in meiosis, you can actually have DNA recombination. And all that means is when these homologous pairs during meiosis line up near each other, you can have this thing called crossover, where all of this DNA here crosses over and touches over here, and all this DNA crosses over and touches over there. So all of this goes there and all of this goes there. What you end up with after the crossover is that one DNA, the one that came from my mom, or that I thought came from my mom, now has a chunk that came from my dad, and the chunk that came from my dad, now has a chunk that came from my mom. Let me do that in the right color. It came from my mom like that. And so that even increases the amount of variety even more. So you can almost now, instead of talking about the different chromosomes that you're contributing where the chromosomes are each of these collections of DNA, you're now talking about-- you can almost go to the different combinations at the gene level, and now you can think about it in almost infinite form of variation. You can think about all of the variation that might emerge when you start mixing and matching different versions of the same gene in a population. And you don't just look at one gene. I mean, the reality is that genes by themselves very seldom code for a specific-- you can very seldom look for one gene and say, oh, that is brown hair, or look for one gene and say, oh, that's intelligence, or that is how likable someone is. It's usually a whole set of genes interacting in an incredibly complicated way. You know, hair might be coded for by this whole set of genes on multiple chromosomes and this might be coded for a whole set of genes on multiple chromosomes. And so then you can start thinking about all of the different combinations. And then all of a sudden, maybe some combination that never existed before all of a sudden emerges, and that's very successful. But I'll leave you to think about it because maybe that combination might be passed on, or it may not be passed on because of this recombination. But we'll talk more about that in the future. But I wanted to introduce this idea of sexual reproduction to you, because this really is the main source of variation within a population. To me, it's kind of a philosophical idea, because we almost take the idea of having males and females for granted because it's this universal idea. But I did a little reading on it, and it turns out that this actually only emerged about 1.4 billion years ago, that this is almost a useful trait, because once you introduce this level of variation, the natural selection can start-- you can kind of say that when you have this more powerful form of variation than just pure mutations, and maybe you might have some primitive form of crossover before, but now that you have this sexual reproduction and you have this variation, natural selection can occur in a more efficient way. So that species that were able to reproduce and essentially recombine their DNA and mix and match it in this way were able to produce more variety and were able to essentially be selected for their environment in a more efficient way so they started to essentially outnumber the ones that couldn't, so it became a kind of very universal trait. But you could have imagined a world, and there are science fiction books written about this, where you have three genders, where you have gender one, two, three. You could have 10 genders. It just happens to be that on Earth, this notion of having two genders turned out to be a very efficient and stable way of introducing variation into a population. So, hopefully, you found that interesting. In the next video, I'll go more into the detail of how exactly meiosis and mitosis works.