Small spider pulls big empty shell up into a bush to make itself a home, and I haz questions

November 24, 2021 • 1:30 pm

This is one of the most amazing pieces of spider behavior I’ve ever seen (filmed, of course, by the BBC and narrated by Attenborough). You have to watch yourself it as it’s too complex to describe.

There are several questions that arise, and I have no answers:

a.) Does every member of the spider species do this, or is this a behavior evinced by just one individual? (Nobody knows.)

b.) If the latter, how the hell did that spider figure out what to do? If it’s not species-wide, it probably isn’t genetically encoded in the brain, and this behavior would have to be figured out! I don’t think that spiders have that kind of savvy, though they can spin very intricate webs or build trapdoors. Those however, are species-wide evolutionarily derived behaviors.

c.) How does the process of affixing one strand after another to the shell lift it up? The spider isn’t strong enough to haul the shell up, nor does it seem to be using the silk as a pulley, which wouldn’t work anyway

If readers can answer any of these questions, be my guest!

Happy Thanksgiving to all. I’m taking a tiny break tomorrow, so although there will be posts, don’t expect many. Enjoy your noms instead!

h/t: Jim

29 thoughts on “Small spider pulls big empty shell up into a bush to make itself a home, and I haz questions

  1. The answer to a) would appear to be at least two, since one spider suspends a shell that is severely affected by the wind and Attenborough declares that “this one shows how it should be done” before showing a perfect suspension.

    1. Or it would be serendipity that they filmed a spider shell acquisition from start to end and with such a fine photography!

      My guess is that someone noted the behavior in that desert, and that it evolved in the species (or in an ancestor of several species).

  2. Regarding spider hauls, I think we had this general question here a long time back. This is common for prey among web spiders and, if memory serves, it has been described in terms of pulleys [ https://www.wired.com/2016/12/pull-car-ditch-super-strength-physics/ , https://courses.lumenlearning.com/physics/chapter/9-5-simple-machines/ ].

    Spiders do not need to be strong, but they do need to use many strands in order to haul longer distances or higher heights.

  3. I’m no spider expert but … Spiders regularly hang food items from their webs as a sort of larder, don’t they? Didn’t know a spider was equipped to eat a snail in its shell but I guess it must be. Spider silk normally shrinks a little after it has been extruded. This is how they can make webs that maintain their structure without turnbuckles and other tensioning devices.

      1. You have it! Once the habit of hoisting up dried leaves as shelters has evolved, spider mutations that promote similar behavior towards terrestrial gastropod shells (initially when leaves are absent but afterwards specifically with regard to snail shells if these provide superior protection, are not too heavy, and are sufficiently available) will be selected and become characteristic of the spider species.

  4. Well, the video directly states that each strand attached is shorter than the previous one, and cites that as the way the shell is lifted. It makes sense, since the spider can stretch a short strand to make contact, which will put it under tension. Enough such strands under tension, and the shell will be lifted.

    As for the nature of this behavior, I’d say it has to be evolved. Spider brains just aren’t big enough to figure that kind of thing out. It likely started with the spiders just taking shelter, then using a web to facilitate getting back up the plant from the shell, to eventually actually hauling the shell up to avoid being on the ground at all. As long as each step improves the spider’s odds to live long enough to find a mate, it’ll get selected over time.

  5. Not convinced by the “each strand shorter than the previous one” explanation. As that would mean that at any given time, all the weight is being carried by the single – shortest – strand. Try it yourself – with some string and any object.

  6. c) Say you have a weight suspended from a rope. If you attach a 2nd rope to the weight and pull it in order to tie it to the ceiling at a point a little offset to the point that the 1st rope is tied this will require force as little as just over half the weight ( https://demonstrations.wolfram.com/WeightSuspendedByTwoCables/ – the smaller the angle between the 2 ropes, the smaller the linear force on each with F=W/2 the limit for a=0 ). The nice thing however is that you can cheat a lot by not doing the process in a static but in a dynamic way: swing the weight (requires very little force) and, at the right moment, tie the rope. Then repeat the process. Also, take advantage of this being a 3D problem: for attaching the 3rd rope, swing the weight by applying a force that has a component perpendicular to the plane of the 3 points you have thus far (2 on the ceiling and 1 on the weight).

    1. That makes sense – I can see that working. But watching the video, it looks more like a single point at the top and multiple along the circumference of the shell. But, that would still work – right? Essentially rolling the shell a little bit with every new shorter strand at a different place along the circumference of the shell. You can see that the shell is gradually turning as it is being hoisted.
      I guess that’s also why we don’t see any slack strands – which should appear as the initial, longest, strands became longer than the distance between the top point and the lowest point on the shell. The shell must be rolling over continuously – sort of like rolling a tape roll to get the tape back on the roll.

  7. As for c), and surely this is essentially the same with simplification as #3-Torbjorn’s above:

    Suppose you have a heavy uniform steel rod whose weight is more than you can lift, but half of it isn’t more. You want to get it up on a roof (onto which you can climb, not carrying anything). You tie a long rope to one rod end, another to the other end.
    Note that you are strong enough to lift an end of the rod when the other is supported (by the ground—except that this time you’ll lift successively each rope, with, after the first lift, the other rope supporting it. (The spider uses more than two of course.)
    So you somehow carry the free rope ends up onto the roof.
    From there, pull one up some ways, not so far as to lift the rod off the ground, which you’re too weak to do anyway; attach it tightly. Now do the same with the other, except it’s not the ground supporting the other end, it’s the other taut rope.
    Just keep alternating as before till it’s lifted up onto the roof.

    A mechanically clever or experienced person (definitely not me!) would set up pulleys of course; e.g. lifting the Internal Combustion Engine engine out of a (soon out-of-date, with EVs!) ICE car.

    The strength of spider web ropes is extraordinary, or so I am given to believe.

    EDIT: Didn’t renew, and when written only the first six or so showed—I think #12 is really the same!

  8. I am a bit late to this but want to throw my 2 cents in. The shell does not roll up into the initial silks as it rises, so that doesn’t seem to explain where the length goes as the spider attaches shorter silks. Using fiber ropes or steel cable as a model has limitations. Silks both expand and contract, unlike rope or cable. My guess it that the initial silks contract so that the entire series of silks remains in tension. It is also sticky, which may contribute to the silks working together as a system. I think the solution has to do with these characteristics of the silk rather than the mechanics of attachment. The mechanics of the attachment to the shell and the supporting plant are important for balance, but not for lift.

    1. Look at the video at about 00:54, the shell is facing opening down. At about 00:57 it has rotated about 90 d anti-clockwise. And by 01:07 it has rotated almost 180 d.

      1. The shell does not appear to rotating around itself, like someone rolling tape back onto the roll. It appears to change its angle due to the location the silks are placed on the shell.

    2. From my #13 it seems clear that I largely disagree, mainly because the role of the silk ‘as a bungee chord’, and the other factors you mention, may play some role, but I think it is minor.

      The fundamental factor, one which Archimedes ‘almost’ knew, is that if an upwards pull of say 3 inches once raises the centre of gravity of the object by a single inch, then the average force needed in the lift is exactly 1/3 the force needed to directly lift the entire object that 1 inch. And 3 becomes instead 2 in my example of lifting a uniform rod. Some kind of a flat tripod (like an upper case Y) is closer to the 3. A good example of 4 would be approximated by one of those old tire irons for loosening a car wheel. For young people (under age 70 maybe) who’ve not experienced changing their own flat tires, this is a nearly perfectly symmetrical cross, the slight differences mainly slightly different sockets at the ends (e.g. I own one whose sockets are the old fashioned USian inches—IIRC they deal with wheel nuts which are 5/8, 11/16, 3/4 or 13/16). If you imagine a giant one weighing 100 kg., then as long as you can lift somewhat more than 25kg (=~55lb.= a quarter the weight) and have 4 light strong cords, you could probably figure out some maneuvers to lift that up onto your roof.

      The basic idea works for other objects, not essentially 1-dimensional like my pole(=dipod), tripod and quadrupod, and is conservation of energy (or “work against gravity” here), that being
      (distance lifted) times (weight lifted), with distance just following the centre of gravity of the object.

      A civil engineer could explain this better (but not an uncivil engineer, because they drink too much beer—that motivates creating some doggerel, maybe with the word ‘peer’ or ‘leer’).

      My pet spider is named Archimedes of course.

      1. “The fundamental factor, one which Archimedes ‘almost’ knew, is that if an upwards pull of say 3 inches once raises the centre of gravity of the object by a single inch, then the average force needed in the lift is exactly 1/3 the force needed to directly lift the entire object that 1 inch.”

        Is there a concise statement of this?
        A formula perhaps?

        1. The 3rd last paragraph, “The basic idea…” seems to be what you want.

          But to be more specific, I think it is as follows below. [Check that the sides both are in units: mass x (distance squared) /(time squared), same as units of energy, so the equation is the latter’s conservation] :

          fd = f’d’

          where d’ is how far the centre of gravity rises, f’ is the gravitational force needed to directly do that, d is the actual distance upwards you push or pull upwards using your ‘mechanical advantage’ device, and in this case f is solved for to give the lesser force actually needed with the use of the device.

          The famous quote here is Archimedes’ leverage one about giving him a long enough lever and he can move the entire globe—but he needs to move his end of the lever a hell of a long ways before one could notice any distance he has moved the earth !

          The everyday check here is pushing a door near the edge opposite the hinges in the usual way, versus pushing it much nearer the hinge. In the latter case you must push, say, 3 times as hard but only 1/3 as far if you try it that hard way and 2/3 of the way over towards the hinge.

          Within the […] above, I would recall the units for energy by remembering either of (Newton’s?)
          1/2 mass times (velocity squared), or Einstein’s

          mass x (velocity of light squared),

          and the units of force by Newton’s famous

          mass x acceleration.

          A real physicist has all those units hardwired into her brain.

      2. Your description seems to require a type of pulley, as you mention in 13. Or it would require puling the ropes up farther and farther with each lift, so as to shorten them relative to the ground. The issue is not the strength of the silk, which we agree is astonishing for its thinness, but what draws it upward with each shorter silk. A bungee is like a cable or a rope: it cannot contract like a silk. I am willing to imagine a wholly mechanical explanation is correct, instead of mine, which relies on specific biological characteristics of spider’s silk. I just haven’t heard a convincing one yet. What we are trying to do is find a way for the entire entanglement to constantly remain in tension, even though each thread is longer than the previous.

        1. Would the spider not untie the thread to be ‘pulled’, then tie it again when finished, where ‘tie’ is surely more like ‘stick’?
          My person on the roof would be doing that, keeping tautness. (Spellingwise, taughtiness is next to naughtiness?).

          1. I am now picturing a spider tying its silk into intricate bows and fancy knots and it is making me smile. Happy belated Thanksgiving!

Leave a Reply