This is a mock up off it.

The real object is in a cell over here, a red sphere above a glass sphere.

The idea is that lower sphere is denser.

The glass on the atmosphere is plastic.

It's a less than so.

These two sauces have different sizes, but the important thing is I have different densities.

So although it's the whole thing is neutrally buoyant.

Sphere, the top is being pulled up in the sphere of the bottom is being pulled down so that spring is being stretched.

First, we've got to make one of these on a smaller scale.

I had to make a family together a nightmare.

You've got a big sphere in the atmosphere, and you gotta connect them by a coy orbits vaguely spring shaped.

And you got some wine that round a rock like this.

And then you got very carefully Get the glue 11 in the glue on the other.

If you get too much glue, then it gets too heavy and it sinks.

You don't get enough to light.

When I press this switch, it just causes this loudspeaker to Lhasa late.

So this is going up, up and down.

About 50 hertz of the same frequency is your light bulbs, so it's very difficult to see under these lighting conditions.

But this surface of the speaker is going up and down, so it's accelerating alternately.

It's accelerating that way.

Then that way, rapidly backwards and forwards.

We're going to shake this object inside a liquid on.

Then the lighter sphere will move one way, and the densest fear will move the other on.

It will go like that, and similarly with this top sphere that gets propelled forwards at a higher acceleration than the sphere.

So what happens is that when the thing is accelerating upwards, it stretches like that.

So as you shake it up and down, this one wants to stay where it is that won't wants to move much more go with the flow, as you put it.

So what do we mean by swimming?

Well, what we mean by swimming is that you got no object in some liquid, perhaps, and in order to propel itself along this object, it has to the object has to deformed in some way has to move its body parts in some way.

It turns out that when for objects that are moving very slowly or very, very small objects in water such a bacterium if they in such a case, the viscosity of the water is very important.

And you find that if you do a stroke in which you so called reciprocal stroke in which the stroke looks the same, Ford's in times backwards in time.

So if I'm trying to stroke like that, and then if I may, you have to imagine I'm doing It's either very slowly, all I'm very small.

That would just propelled me for words when I do that, and then backwards again.

When I do that, we're gonna know where.

So that's a stroke in which, if you run the video forwards and back or backwards in time, you can't You can't tell just from looking at that stroke.

Whether the video is being run forwards or backwards in time on object, like a bacterium which is very small, which is living in a world dominated vibe by viscosity living a very low Reynolds number, it has to develop another trick in order to be able to swim so you can't use a stroke like this, which looks the same forwards or backwards in time.

It has to develop some other type of stroke.

So one way, for example, a bacterium could do it.

For example, an E.

Coli bacterium is that he's got a little whipped like tail on the back, and it rotates that round just like a corkscrew.

On that motion, you can tell it's not time symmetrical or other tricks as well.

I mean you, the bacteria can use a little silly, a little hairs that move along the edge of its body, and that could propel it through liquid.

So here, when we're if we if we are isolated at a small amplitude, it's a little like the bacterium it's living at Low Reynolds number because this is a time reversible motion, it's gonna go nowhere.

So what we've done is that we keep increasing the amplitude until the inertia of the liquid becomes important.

At that point, it starts to swim.

What essentially that the nurture of the liquid can rectify the motion, this oscillation in motion so it can turn oscillating motion into a jet of liquid that comes out of the bottom of this jet of liquid is what propels the spheres along.

It was cool, but it happened so quickly.

In real life, you have to go back and watch the video to see if it was actually interesting.

So when we went back and watched the video, we saw this huge jet being propelled out of the bottom sphere and a couple of water sees above the big spheres.

Oh, yeah, It became really interesting retrospectively.

I'm too clumsy to do this.

Mike did that.

Then you put it in the cell.

You then have to get all the liquid in here stationary.

You have to make sure the density of this is exactly right to get it to float in the middle.

I forgot about the bubbles.

Yeah, Bubba was right.

So if we Yes.

So this solution is so water, but we have to put it on the vacuum for awhile.

First, if there's air in the cell halfway through, a little bubble will appear on the top surface of this thing and it will rise to the top and you'll get a full three ding.

So you're constantly going against nature's desire to screw this experiment out.

The interesting question is, how does there's such an object?

Or any objects make a transition to swimming.

So at the Reynolds number of low amplitude of observation, it's not swimming.

Just like the bacterium that can't that can't swim with reciprocal motion, we'll keep increasing amplitude.

Suddenly it starts to swim and it goes off.

We get this jet of liquid underneath it or is it a smooth transition?

So does it started?

If we start very, very low speeds, does it gradually start moving with faster and faster speed?

And so what we seem to have found with this experiment is that there is indeed a transition between no swimming, and we reached a critical amplitude and it starts swimming.

This is one of these problems which mathematicians are worried about.

They don't know how to deal with this exactly on they write equations down, which worked for small velocities.

Or, you know, if you were doing this intrigue, but they don't correspond to the real world.

This has fallen through the gaps of mathematical analysis computer, you have to do this damned experiment.

I said that before.

People complain that Reese worry, but you have to do that in order to make sure you're not fooling yourself.

And they bashed together so many times that they almost come to a standstill.

When they made a little cluster, another one comes in and bashes it.

It makes a little cluster.

Because of the elasticity, the energy loss in these collisions, there's a a likelihood that they were plastic together and form little clusters.