Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Grokking evolution

  1. Jul 4, 2016 #1


    User Avatar
    Gold Member

    I have read some amount of literature on evolution and natural selection over the years and, while I certainly learn a lot, it never seems to quite help me grok the breadth of change over deep time.

    So I have some rather naive thoughts that I'll phrase in the form of incredulity. They are simplified, for the sake of brevity; I understand that it's a nuanced topic.

    I don't for a second doubt that it's the case, I just have to ... trust in the science.

    1] Dogs are bred for different things. (Not the best starter topic, since this involves artificial selection) Terriers are bred for digging. If we assume that this is an instinctive trait (not learned from a parent), then there is something in the wiring of the brain that tells all these dogs to dig. Their DNA, the only thing a breed has in-common with its kin, must encode the genes to build the neural pathways that cause it to want to dig. That, and ten thousand other subtle behaviors must be encoded in the genes, which cause proteins to make pathways, which survive the embryonic process to birth, which cause it to ... what? dig? That subtlety of encoding is virtually inconceivable to me.

    2] Complex mammals dream; cats do; dogs do. I suppose rodents do too. I'm not sure about non-mammals. That means that the process for dreaming must have come from their common ancestor - that shrew that lived alongside the Tyrannosaur 80 million years ago. If more distant animals - such as any of the mammals' classificationary cousins - also dream, that means dreaming was well-established even before mammals split off. Unless we are going for convergent evolution here...

    3] Cows have tufted tails, the better to swat flies away with, and over-large eyelashes that help keep flies out of their eyes (pick any other subtle traits). They evolved these things as an evolutionary advantage (granted, this is oversimplified). Does this mean that of the one gajillion cowoids that went before it, a statistically significant number of them died or were unable to breed effectively because they had short eyelashes and broom-like tails? I should find a better example but, in essence, does it not hold true that a critter has some subtle trait because .9999 gajillion of its ancestors dies whereas of those that di not have that trait a whole gajillion died? I am aware that many traits evolved alongside other traits, without any apparent advantage. Still, the net outcome is that advantageous traits - even ridiculously subtle ones - were selected for.

    4] I imagine an idealized critter, with only two traits, say, a blob that has a skin-texture and a leg-length.
    The smooth-skinned has a slight advantage over pebbly-skin, but long legs has an advantage over short legs. (Let's say they live on the surface of fast-running water or some such). Smooth skin prevails over pebbly-skin, but in doing-so, wipes out a bunch of long-legged, pebbly-skinned varieties. Ultimately, the two traits evolve independent of each other. Skin gets smoother, legs get longer, even though many of both kinds of being wiped out.

    I guess I can see it at the scale of just a few traits, but the number of individual traits of any complex critter is virtually uncountable. This critter has wider tubes for its lymph nodes, that critter has striations in its liver, another critter has eyes nearer the top of its head, yet another has smaller blood corpuscles. It is virtually unfathomable that these traits can evolve essentially independent of each other in a population, when the traits are essentially competing with each other to "help" the critters survive.

    Some day I want to write a program like this. An idealized critter, with a hundred traits, all with some sort of (unknown!) advantage or disadvantage, or both - (traits that, in the real world, might to allow it to - or prevent it from - breeding just one more time than its neighbor). I want to run it against an idealized "challenging landscape". The challenges are just idealized properties as well; just randomly-invoked events - one knocks these traits up and those ones down, another knocks this set up and that set down. Of course, the sets of traits affected will strongly overlap - in a sense the traits themselves are competing. I want to watch and see if, after 10,000 generations, the critters remaining will have selected traits that equate to advantages. Someday...
  2. jcsd
  3. Jul 4, 2016 #2


    User Avatar

    Staff: Mentor

    Last edited by a moderator: Jul 4, 2016
  4. Jul 4, 2016 #3
    I'm not really sure what your question is. It is what it is, what survives is what survives, and what dies..doesn't not survive. Not all survived traits are independent, many are coupled, e.g., the development of the eye and the visual cortex among many others.

    Basically, yes. But it doesn't necessarily mean that the short eyelashes and broom tails caused their demise per se, it just means that another cow cam along that didn't have those traits that had another trait not related to those that allowed it to out-compete the short eyelashed cowoids. You know, like some massive udders or something like that :oldsmile:

    There is studied hierarchy of behaviors that run down from mammals to lower vertebrates such as reptiles, amphibians, and fish. It's a science called "neuroethology."


    There are so-called instinctual neural assemblies that are indeed, encoded by the DNA that lay down neural pathways that, say, "code" for a digging behavior. But these need to be reinforced through training or "imprinting" typically to reach full expression in a process
  5. Jul 4, 2016 #4
    Wow, I thought all my effort writing a response to this thread was wasted as Evo closed the thread "in write." I'm glad it's back up and my text wasn't deleted. Hope someone gets something from it :-p

    Edit: Oops, re-reading it, it looks like my response got cut off mid way for some reason..and unfortunately, those were the best parts..:redface:
    Last edited: Jul 4, 2016
  6. Jul 4, 2016 #5


    User Avatar
    Staff Emeritus
    Science Advisor

    Here are a couple of links to programs that simulate evolution, though they may not have 100 traits.

  7. Jul 5, 2016 #6
    DaveC, on the matter of dreaming noted in your point 2, I have no knowledge of the science and history of dreaming, but it may have evolved long ago. Michael Graziano postulates the development of a complex attentional processing system for neural signal enhancement which laid the foundation for what he describes as the "attention schema - an internal model of attention. This model constitutes what we might call awareness or consciousness. If this idea is anywhere close to right, and dreaming uses that process (which I assume it might given the similarities between conscious awareness and dreaming), then dreaming may have evolved upwards of several hundreds of million years ago... I rather like Graziano's proposal.

    http://www.princeton.edu/~graziano/Graziano_JCN_2014.pdf (Journal of Cognitive Neuroscience)

    Speaking of neuroethology, Graziano also has an interesting paper about ethological action maps:

    http://www.princeton.edu/~graziano/graziano_2015.pdf (Trends in Cognitive Sciences)
  8. Jul 5, 2016 #7

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    My understanding is that describing evolution in terms of 'survival fitness' only has meaning in terms of large-scale population dynamics; applying 'survival fitness' to a single member of a population (or even a single species) is generally problematic.

    Dogs actually present a few subtle challenges: first, they are intentionally bred and cross-bred, and the particular trait selected for breeding is an entirely artificial construct- many breeds suffer from specific disorders (cataracts, glaucoma, hip dysplasia...) that would cause problems in the absence of domestication. Second, dogs, of all species, have the largest variability in physical size. The extensive history of domestication and breeding provides an interesting 'laboratory' for canine genetics:


    As for dreaming, I don't know that the function is understood; assigning an evolutionary status may not make sense unless it is posited as a necessary function of the central nervous system. Currently, existence of REM sleep is a proxy for dreaming, which is about the best we can currently do.


    Regarding cow physiology, tufted tails are not unique to cows. Perhaps more interestingly, lions are the only members of Panthera with a tufted tail, and male lions are the only members with a mane. Asking 'why is this?' has no obvious answer- evolution-as-survival fitness cannot plausibly respond, but that's because it's not the right question. Similarly: if walking on two legs confers a significant survival advantage, why are there only a few living mammalian species (AFAIK) that exclusively walk upright? Evolution is not a directed process- it does not 'act' to solve a problem, and it is not an intermittent process.

    Personally, I'm interested in comparing the genome of a modern alligator (or Ginkgo) to one that lived 100 Myr (250 Myr) ago- what genomic changes have occurred in the interim?

    "Evolution" type algorithms exist in a variety of contexts to solve a variety of problems, likewise experiments.
  9. Jul 5, 2016 #8


    User Avatar
    Staff Emeritus
    Science Advisor

    Dave, I'll ask my boss if he's in today about some of this and see what he says. He was a biologist and we've talked about evolution before.
  10. Jul 5, 2016 #9


    User Avatar
    Science Advisor

    If you can't show that the trait is an adaptation (eg. increases survivability of the organism), the question of natural selection does not arise.

    Furthermore, one should be aware that natural selection is not as fundamental a concept as evolution.
  11. Jul 5, 2016 #10


    User Avatar
    2017 Award

    Staff: Mentor

    Is that true in general? It is true for modern humans, but I see no evidence that it is true in general. Humans are quite slow compared to many animals of similar size, for example.
    We inherit more than the DNA. We also get the epigenome.

    For your specific example, I don't know if terriers dig (more than other dogs) if they don't learn it from others. The larger concept "explore the world around you" is certainly inherited, if you do that exploration more below the ground, at the ground, on trees or whatever is a different question.
  12. Jul 5, 2016 #11

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    Since hominds are one of the few bipedal mammals, post-hoc rationale is generally restricted to primates. It's commonly asserted as being beneficial, also the lack of rationale as to why it occurred for hominids.

    http://elucy.org/Main/WhyBipedalism.html [Broken]

    I could not find similar excuses for kangaroos (bipedal) or pandas (quadrupedal), which also have opposable thumbs and so could hold tools with all the supposed benefits accruing unto.
    Last edited by a moderator: May 8, 2017
  13. Jul 5, 2016 #12


    User Avatar
    Gold Member

    Slow, but... we still were able to maintain predation over most of the animal population with our legs using a persistence hunting strategy. But this strategy is used by furry four legged animals like wolves too, so to what extent two legs are better I don't know. I think one thing that helps us a lot (and maybe one reason other apes don't persistence hunt) is sweating (a low energy process) instead of panting (a high energy activity) makes cooling during persistence hunting more efficient.
  14. Jul 5, 2016 #13


    User Avatar
    Gold Member

    But it does, ultimately.

    Whether because it couldn't breed due to its short-eyelashe or because some one long-eyelasher bred more, using up resources, the outcome is the same.

    Same thing. In a turbulent world of competition, the cows with short eyelashes lost to cows with long eyelashes. The massive udder thing should not factor in.

    Say you've got four cows:
    short eyelashes, normal udders
    short eyelashes, massive udders
    long eyelashes, normal udders
    long eyelashes, massive udders

    The udder trait will evolve independently of the eyelash trait. (again, this is not always so, but mostly traits have degrees of freedom not constrained by other traits)
    Which means that, when all is said and done, the long eyelashers outcompeted the short-eyelashers.
  15. Jul 5, 2016 #14


    User Avatar
    Gold Member

    [/PLAIN] [Broken]
    Do you mean that digging is an instinctive trait passed down to descendants of all mammals?

    Except that terriers naturally dig. To what extent, I'm not concerned, the point is, they have that trait to a level noticeably higher than other dogs.
    Last edited by a moderator: May 8, 2017
  16. Jul 5, 2016 #15


    User Avatar
    Gold Member

    Yes. The problem is, that - ultimately - that is what is happening.

    Our population models involve large populations, but in reality, the variations in species is actually occurring on an individual basis, as each critter breeds or doesn't.
  17. Jul 5, 2016 #16


    User Avatar
    Gold Member

    This is an interesting point, methinks.

    If my understanding of evolution is correct, there are few general advantages. The proof (of survival) is in the pudding (of actually surviving). Humans survived because of a large number of traits in conjunction with its upright stance - for example good binocular vison, developed hands that can be used once freed from locomotion, large brain to make tools. Some other critter with upright stance but none of these other traits would find themselves at a disdvantage. Too much of their energy budget would be being wasted on a vertical stance if they have poor eyes or clumsy hands.

    The suvival happens on an individual basis, since the "landscape" (both literal and virtual) of drivers (distant mastodons, thick fur, wide plains, clever competing predators, etc.) is nearly limitless. In other words, there is no "formula" that works generally, just de facto successes.

    Most of the common traits that disparate species share are not independent at all - they were handed down from a common ancestor.

    In my opinion.
  18. Jul 5, 2016 #17
    Yeah, that's probably correct. There's no real reason to think that eyelash length and udder size are linked-traits (in cows at least, if you're a human female stripper, maybe :rolleyes:)

    The point is that, so what? What's your point? Eyelash length may confer an advantage to keep dust or bugs out of your eyes or serve as a secondary sexual characteristic. In either case, it's a trait subject to natural selection as is udder size, and they may be selected for independently or partially independently. That kind of goes without saying. What's your quandary?
  19. Jul 5, 2016 #18

    jim mcnamara

    User Avatar

    Staff: Mentor

    I think Dave's problem is assumptions, one is: there is meaning or direction or intention or any kind of logic to Natural Selection. It is random. No logic or predestination whatsoever.

    A meteor falls on a population of cattle. They all had short eyelashes. That gene now becomes rare and whatever selection pressure there was in the past to preserve the trait is no longer there. Long eyelashes "win" for no reason at all( to put it in a really bad sentence with wrong meaning).

    Think Yoda: There is no win - only live to reproduce. And. Reproduce more than the next cow.

    You cannot focus on traits, only environment and maybe trends in the environment.

    Does that help?
  20. Jul 5, 2016 #19


    User Avatar
    Gold Member

    Yeah, it's just weird to think that it happens to individuals in the population. Literally, more short-eyelashed cows died (from short-eyelash problems) before they could reproduce (otherwise, that is not the trait that would be selected-for).
  21. Jul 5, 2016 #20


    User Avatar
    Gold Member

    Mm. No I'm not assuming that.

    To be clear, I'm adequately well-read in evolution and natural selection, and there's no doubt of its veracity. Its just when I start looking closely - a at human scale, of individuals, rather than the overarching million years of populations - that it gets hard to fit in my head.

    Random events like this would balance out in the long-term. Next century, a flashflood wipes out a population of long-eyelashed cattle.

    That's one of the things I'm getting at.

    Our models deal with populations.
    But the actual mechanism of selection acts upon individuals.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Discussions: Grokking evolution
  1. Evolution Is True? (Replies: 11)

  2. Evolution of organisms (Replies: 30)

  3. Organs and evolution (Replies: 5)

  4. Divergent evolution (Replies: 2)

  5. Virus and evolution (Replies: 7)