Thank you very much.

It's great to be here.

You all for coming.

We're going to have a.

So we're talking about insects today in
particular how to stay dry to stay alive.

I'm going to talk about it on a couple of
my favorite model resumes for looking at

watermelon see this is a for example this
is a raft built by ads to talk about that.

We'll talk a little bit about
how flying insects repel water.

And if there's time the end.

Will get to my wife's ex boyfriends
Valentine's gift to her.

Which was this poodle wishes and
Good Morning America last year.

So.

Let's I collaborated with in particular
with Craig Tovey this this this is

an image of the academic a lot of you have
to climb in order to get it in my lab.

It's arduous to reach the top.

With views very nice.

Michael and I did this and giraffe
model I'll discuss Nate's and Andrew

aren't here I don't think they did the
they did most of the experimental work.

And on the higher number of
undergraduates some in particular to take

the photos that you'll see
throughout this talk in the videos.

Most of which are our labs.

People use they ask me
who funds this stuff.

And as a physics of living systems and
N.S.F.

biology in particular physiological and
structural systems.

So you can check out.

You can check out my abstracts
on the net of online.

If you want to see more about my funding.

Let's begin with the fundamentals of

what you're seeing here is the mating
dance or combat of the water Strider.

It's on the water surface.

So this this animal is basically
somersaulting jumping wrestling.

It's going to there's a head lock
on the right hand side here.

It's doing all this on the water surface.

All this without following
through which is just amazing.

It's impossible to our scale but
on that way.

It's you know it's not not very difficult.

So what support in the way to these
insects is surface tension which

is a force per unit length or
equivalent energy producing area.

As a result water services tend to resist
defamation and they act like trampolines

which prevent them from objects from
sinking through if they're very small.

I realize a lot of people who probably
wouldn't research is not exactly

organism biomechanics So

I would have a I'm going to talk about
surface tension for your own benefit.

So when a drop of water is placed on
a solid There's basically two regimes that

can occur.

One of which is biologically favorable
such as this which old codgers hydrophilic

hydrophobic and then one hundred filling.

The idea is that if you're hydrophilic
water tends to form a very thin flat layer

and completely coat the surface.

That's a circular with a con angle
of contact of the zero degrees.

If it completely touches the surface.

What organisms One is they want
to avoid contact of the water so

they generally try to have these
contact angles to be as close to

one hundred eighty degrees as possible.

One hundred eighty degrees and
correspond to a perfect sphere.

Now if you don't know
if you were in water.

PELLANT pants.

The way it works is through a combination
of chemistry and texture chemistry

acts to make the surface hydrophobic
in one hundred eighty as possible.

And texture basically acts to
make the water see a composite

surface basically a combination of air and
pant material.

So particular.

There are two laws that
describe how that works.

One is the called the Castle Law and
it says The There's this context and

all the drop on the macroscopic scale and
it's a weighted average of the contact

angle on the air and on the solid and
if for example which a lot

of these surveys are all show you if it's
mostly air the contango becomes very high.

If a job water hits the surface
with sufficient speed it can judge

you know once the surface where instead
of being supported on an array of A.

Air pillars it actually penetrates and
what's the service.

So this is a kind of thing to be avoided.

Now one of the reasons water problems
has been continuously studied for

the last hundred years is
because of his to resus

which is that you know experimentalists
is a terrible word because it means

if you measure something once it may not
be the same if you measure that again.

In particular contact Ingles have
a range of values in nature and

this is due to that on the microscopic
scale there's tiny of mountains in

Vail valleys that trap the contact line.

It's not raining today but
last week when it was raining.

You might have noticed water
drops will stick to your windows.

And from the theory I showed previously

that should be possible because
the contacting all should be the same.

The top and bottom the drug.

But because of history says
there is a minimum angle and

a maximum data receiving and
data advancing.

And it is all the drop looks like kind
of like my will my wife is pregnant.

Right now.
So

they might pregnant wife bill they were
going to give birth in about a month and

her belly is even larger than this but

as a result there's a there's a you can
see there's a vertical force a shoulder

with these different angles here and
that's what supports drops on windowsills

if any of your experiments
involve small bits of water.

You should continue to read.

There is a great book by three French men
very little mathematics that talks about

capital or phenomenon.

Now.

One of so in my field.

Capillarity the Holy Grail was based
on ten years ago when they discovered

something called the sacred Lotus
there's a social with Buddhism but

basically it's a particular plant.

Law and nine when I said published
this ten years ago it has two

thousand sites this paper
has two thousand citations.

And the planet has a Cassie Baxter layer.

So.
That most of this planet is actually

covered in air and is due to very
small very small pillars on the plant.

There is that one.

Drops form they form perfect spheres and
moreover when they roll off they carry

detritus and things with them so that
each time a range the plane gets cleaner.

Since then there's been you can imagine
there's a ton of material scientists

trying to publish on this but there's
still a lot of work that has to be done.

In particular water through fabrics.

There's still a lot not understood why
they're so fragile and expensive to make.

There's still going to be a long it's
going to be a long time before we see

this throughout our daily lives.

So they're going to talk about
three mechanisms I think might

play a role in making water
repellent seem more robust.

So let's start with what you're
most familiar with as Prius said

you might feel a little creepy and crawly
so you might want to finish the sandwich

part of your your meal but we'll start
with ants in particular fire ants.

So these are these are far

from the the wetlands of Brazil
a place where it floods twice daily.

This is composed entirely of
ants on the water surface.

Nothing else.

No glue this and holding onto themselves.

You know.

So these ants have learnt
it here together and

over generations they've learned to
do this in a very organized way.

This is the topic of my students Ph D.

thesis and I only talk about the rats
today but they're capable of building

bridges in case they want to cross
long obstacles and they also

kill people building big works in which
to store their colonies overnight.

And these new acts they can even open and

close the holes in them to reduce
water loss and maintain temperature.

Now if you want to build
something you need a glue and

the and school who is basically their own
claws and mandibles with it because it can

link together strongly enough that they
can form these long chains in the lab.

Let's get this first.

One of the answer very
strong this is one and

holding up twenty of its neighbors as
I said these are S.C.M. images of.

Mandibles and
the claws gripping each other.

And these these these answer very
strong and because there are so

many they act like a continuum sort of
like between of solid and of fluid.

You can estimate a elastic modulus based
on the strength of an ant its size and

its size L. and basically the length
of defamation of its arms and

this is basically it will be
something similar to silly putty.

Using the micro C.T.

That was actually available in this
building you can visualize what inside

the structure looks like the particular
it's actually like the lotus leaf.

It's a structure of mostly air.

It's only twenty percent and
eighty percent air.

And what makes the virus so

strong is that each end has six legs and
a large number of neighbors

these red points show you where it's
connected to the neighboring ants.

And you can see these neighbors
connect all all across the body.

So we see that they can aggregate but why.

Well that's so this is a this is
a typical experiment to do in our lab.

This is a drop of water and here's an ant
and you see it has very poor properties.

If it were to be a good word to try to
swim in particular it's contact angles.

So it sticks to water it hasn't he can as
a result if you are trying to swim and

if you're and you basically
stick to water with every step.

It's a terrible thing.

Moreover there's a high has to resus
basically if you want to remove

your arm from the water you have to
deformed the contact angle a great deal.

All of these are terrible for swimming.

But luckily these ants live in colonies
and they learn to cooperate and so

what they've learned to do is build text
textured surfaces by looking together

at the exact spacing that's required
to prevent water from penetrating.

If I want to I I was going to ask
the Georgia Tech some team to

see if they could build
a raft like this but

of course that you know they
wouldn't be water repellent.

But because they're in the life
skill on a very small length scale

water will tend not to penetrate
these pores made between the ends.

And by the casting Baxter
law I stated earlier

if they have half the area
fraction air and half of it and

their contact angle will become much
greater than it was if it were slowly and.

You don't have to believe me
you can ask answer selves.

So this is this what happens when
you try to drown these ants.

On You Tube.

Everyone wants to drown us drown the ends.

So I was disappointed
when we can drown them.

This as I believe this is one
million hits on You Tube It's quite

quite a large number of middle
schoolers interested in our research.

But you see when you
try to push them down.

They basically form a form
bubbles between them.

This is all air pockets and that prevents
water from penetrating the raft.

It's very non-intuitive.

And you push the wash even further and
further in depth.

You see they form this air lair that
that allows them to them to stay dry.

Even when under water.

This air there has a limit of course
because as you go deeper and deeper.

There's going to be increasing
pressure that will eventually push

out these bubbles.

What maintains they are there.

This is a common principle in nature.

This is.

In the water appellants litter.

There's always it's always known there's
a hierarchy of life scales that prevents

water from penetrating.

And particularly for the rats there's
the spacing between the ants.

There's the space in between the legs and

on the legs themselves
there's these tiny hairs.

For water to try to wet this very complex
surface has got to do a lot of do a lot of

work in particular you can predict
how deep you can put each of these

structures in the water before water will
penetrate those holes of varying size.

They say to fill the hole pores between
the body you have to go one centimeter but

to fill the holes between the hairs
you have to go almost a metre so

this makes the ant very robust.

This is a picture taken of an ant
under water that's floating

floating gently upwards and you see
it's actually not touching the water.

If you look very closely.

There's a thin layer of air
that surrounds its body.

Biologist call this a plaster on or
an artificial deal.

Basically the inward diffusion of oxygen
is equal to the S. metabolic rate and

this means you can survive indefinitely
under water simply by breathing the air

absorbed into the air layer some engineers
a couple years ago trying to build

a version that's fit for humans on a large
scale but it's still having problems.

So we've seen these ants have great
properties when they're digging there but

how useful is this during flash floods.

They have to be able to build
a raft quickly enough to search for

everyone to survive for
example let's say This whole of the we

had a fire drill and we all had to go out
on the water service and build a raft.

Everyone in this building how
quickly would that have and

why everyone has to carry their
belongings we have our laptops and

our centrifuges our pipes and
things like that.

No books and

we're basically trampling upon each
other to try to get away from the water.

Basically we have those that are be.

Upon And
then there was those of the travelers.

How to do this an organized way.

So I'm going to show you a video of
an experiment where we take in and

ball you do that by taking the ants and
swirling them up like this.

In a cup.

And that turns them into
a perfect sphere and

then we put them on the water surface and
watch what happens.

So this is a time lapse video of about

about two hundred seconds and
you see the rat the ants

know intuitively to transform from a
sphere to a pancake they build build build

this is about eight thousand and so
this is about this probably the entire

This is four times that mechanical
engineering class but a thousand and.

They build build build link
together at the edge and

eventually they form a structure
that has an equilibrium.

It looks a lot like a drop
of spreading fluid.

Here's a smaller number of and
so you can see a little bit.

What's happening on the edge.

They're all running toward
the edge franticly.

And then getting to this equilibrium
structure and then of course

you have the students that are the black
sheep they're just doing their own thing.

Everyone thinks they can start
their own colony there they are.

And here's a side view.

Where you get you see it goes
from a sphere to a pancake shape.

There's some interesting
stuff going on here.

I think enough for a whole nother Ph D.

thesis these ants can actually
walk under water on what surfaces.

It's partly because they actually
extrude a drop of hydrophilic and

hydrophobic fluid a composite drop
that allows them to stick to almost

anything even if it's weight.

So you see there are two layers.

There's basically the like I said there's
the underclassman those are the base

the on the bottom that are wet and
there's upper classman that are on top.

That are dry.

How does how does this happen.

So I want to talk a little bit modeling.

I know we're I know we're an Ivy So
there's not the math department so

I'll try to talk about
an intuitive way but

their ideas base is a prick to predict how
quickly this droplet spreads over time.

And we basically parameterize
that by the area of this raft.

We can simply record that from a video and

from the area basically from every square
centimeter we know the density of and so

you know it's it's it's it's thirty and
spur.

Basically in the area of my pinky now.

So we're trying to predict the number of
ants on the bottom layer of the raft as

a function of time it starts out
as the number involved in involved

in the sphere and extends all the way to
what I'll show is about half the colony.

Model inputs we basically have the total
number of ants This is basically

the weight of the ball each and
one milligram so.

One gram of ants is a thousand and
the whole

the time scale the problem set by
the walking speed of the ends.

Basically the answer walk.

They basically run like hell
to get out of there and

if they don't twice as fast the ball's
rafters built twice as quickly and

each and walks a pinky with him and
a second.

There is a number of layers
of the raft and equilibrium.

So that's something we can predict I think
that's based on how strong the RAF needs

to be to be to survive waves and
things like that in experiments from one

thousand to ten thousand and that's always
consistent with a two point five and.

So there and
have a local rule that they basically said

we want to build a raft
that's two hands high.

So here's a model in a nutshell.

Imagine you're a hoarder

a horde of people what you want to do is
get out of there as quickly as you can.

Well if you're starting at a random
position within the horde you basically

want to run as quickly as you can to the
edge of the Horde and that's what they do

they basically take a bee line for
the edge in a random direction.

Once they get to one edge.

They try to find the scape
of course there is none.

So they turn around and then.

And basically run to another it and

from one edge they keep on running to
another edge in another edge until.

They Basic get trapped.

So every time they get to an edge.

There's actually some small chance that
their friend reaches out and grabs them.

If their friend grabs them they
become part of the raft and

that's how the raft expands.

So that we can actually
measure the probability that

they grab with a probability P. and

if this is a unit rock that means
it's the one center in diameter.

We can predict by integrating over all
possible angles the green and the peg.

And so that give us an expected travel
distance on a raft of unit with.

And that's actually the entire model
because now you know what one and.

How long it has to travel to get to the
edge and how long before the Iraq expands.

The rest of model is assuming
the following things.

Well every end basically
does something similar.

We can predict what one end does but

over ten thousand and
these averages are very very good.

So the number of ants that basically add
to the side is proportional to number and

running around on top divided by
the travel time it takes to get here.

Now the number ends rolling on top.

We call that low end of the travel
time is basically this number.

Dimension lies by the radius and
divided by this.

And speed.

We can convert radius to the number
of ants using this density.

Number I mentioned the number of ants
in a pinky which in the pinky area.

And so

that number of ants is collecting at the
side of the raft every unit time and the.

Forming a lair H. thick.

So this gives us a different equation for
the raft of growth and

relies on the random walk lands and

there's basically two regimes you
have to take into consideration

one in the very beginning where
the top layer is saturated with ants.

And that's happening the beginning.

So here you have the whole this is
whole the top is filled with and

they're all going in random directions but
because the RAF starts out small

the number of and some top a small and
so the growth is slow.

That's why it's very
slow in the beginning.

But as the raft expands you get a larger
number of constructors and those and so

you see this left starts
increase in speed.

The rate increases rate increases but
as this thing gets bigger.

Eventually you can run out ants and that's
what happens the very end because you

basically have a diminishing number of
ants on top and as a result the speed

slows down and the you have this
basically this very non-linear behavior.

You know this is a log log scale.

So it's very non-linear in comparison.

If you took a drop of shampoo and
put it on the water surface it spread out

according to a power law so
that this would be straight.

But here it's nonlinear.

So that's that's the model of
how quickly the ends can do it.

And this all happens within
two hundred seconds.

So they're very very good at it and
as a result in the videos I showed you can

have ten thousand and and zero will drown
you have literally zero casualties.

Now these refs aren't always circular For
example

this is a example of what happens when
the raft reaches land and actually docks.

So this is a very Now as to a cell.

I would call it like an amoeba which
reaches out the flagellum essentially

these ants have local information
they see the wall here.

And as a result they walk towards it and
this thing extends extends and

extends until it gets to the edge.

This work is still ongoing but

I think that it will have
representations in the swarm literature.

There's still a lot of fundamental
questions about how quickly you can

build things.

In particular there's now new devices
have you seen this is the Kubler modular

robot that can basically construct larger

robots by by connecting together the Grand
Challenge of robotics in two thousand and

nine is based on take a bucket
of robot parts dump it out and

have this bucket of parts
build bigger robots.

So we're hoping to collaborate
with bodices to do that.

And as you who are do biology know so
Woman together for

example on this amoeba like Dick to
tell them to form a spore slug and

a stalk all that is due to
this kind of swarm behavior.

OK So
that's the four story I wanted to tell.

It's about cooperation.

Now the next story I want to tell is about
about insects the insects that work alone.

So I want to talk a little bit about
how insects fly in the rain so.

So this is a housefly and this is
a water drop that hits the housefly

if you're trying to design an insecticide
this is probably very worst thing that

can happen because it means all your
insecticide gets on the ground and

none of it gets on the fly.

And part of the reason is because
the like like the ant rafts

these the flying insects are actually
highly hydrophobic as a result people had

to build very carefully choose droplets
sizes that were actually touched in six.

This is a scanning electron microscope
picture of Aircel insecticides which

are only twenty micro meters so one fifth
of your human hair that actually stick to

the insects by inertial impaction but
anything larger will basically bounce off.

So I want to talk about how
mosquitoes fly in the rain.

So there's a mosquito is a funny animal
there are they like human environments but

no one's really asking how they can
fly in these conditions that are wet

like rain and fog.

So they get you situated.

Let's think about the two competitors
we've got rate a range up here and

a mosquito they're basically roughly
the same size a couple millimeters.

But the mosquito is mostly its
body is very thin so as a result.

It's mass and the mass of the raindrop
is two to fifty times a mosquito.

So it's basically equivalent of us
being trapped in these the bus.

They're very very heavy.

Over their speed is very is is
very high compared to the mosquito

mosquito probably travels at
most one metre per second.

This far in a second but
these drops at terminal velocity.

If you ever try to see them on high
speed video they're quite fast

five to nine meters per second.

The drops because of their high speed they
carry a great deal momentum that's that's

basically thrown laterally and
he time it hits a solid surface.

So this is a drop in
hitting a bird's wing.

And.

The momentum so base of the month on
the drop the mass times velocity.

And the time scale of the impact
which is a about a millisecond

they give you an idea of
the forces that are involved.

Upon this impact and
is based on ten thousand mosquito weights.

So an extraordinary amount of force.

So how do these mosquitoes survive.

So.

First of all these drops to hit the misuse
actually quite frequently in a rainstorm.

You can calculate this based on
the mosquitoes impact the area

of potential impact that
of its wings on its legs.

Which is about thirty millimeter
squared and a very heavy rain.

It will rain about fifty
millimeters an hour.

That's basically heights if you feel
a bucket for the millions hour.

That's basically the rain intensity

come out of these to get the frequency
of impact upon the area of the mosquito.

And that's basically you get an impact
one every twenty five seconds.

So they'll be quite quite frequent
if they're in the rain storm.

Moreover there's there's basically
two ridges impact that are important.

The direct impacts are worse
than the red area and

the glancing impacts which you
can see as the majority area.

The purple in the green and so
that if you look at the areas involved

the glass impacts are probably
about three times more likely and

that corresponds to our experiments.

So this is a we basically
make a simulated drops and

this is a mosquito able to recover
easily after perturbation by the drop

which now the drops on
the show you in the following

experiments are all slower than terminal
velocity about two meters per second and

you can cover the force
applied on the wing.

That is causing the torque
on the in seconds.

Only about two mosquito waves.

So this is easily easily survivable.

Like I said the next kind

of impact they can occur is that when it
hits direct hits the body directly and and

partly if the insignia flies at the with
the vertical body position the drops

will simply glance off plants off
the back even though they hit directly.

You can see the hydro for business is very
important here and preventing the drop

from forming greatly and I'll talk more
about this defamation in the second.

But if insects on lucky enough to
basically be flying horizontally.

If we get him back to directly.

They'll basically form
a long mast with a drug.

So this is a direct hit with a drop.

You can see when they get hit both of
them go together and mechanics is called

In the last impact where you have two
objects that stick together upon impact.

I'll talk about how we can
predict the momentum precisely

after impact based on the relative masses.

In all these cases these insects
can free themselves in this case.

We don't exactly know why we think
it has to do with the head high drag

on the legs and
wings that allows it to separate.

So to make progress in
this on this problem.

It's actually very hard to do
that experiment with insects.

So we choose mimics basically

small Styrofoam balls that have
the same mass of the insects and.

We think the mass is actually the most
important property in determining

the impact phenomenon.

So like I said if you have two
bodies that impact each other.

The month before impact is going to
equal the momentum after impact.

So as a result the ratio of the speed
after impact the speed before impact is

going to the ratio of the masses involved
in particular the mass of the drop to

the combined mass of the drop and
now miscues are so lightweight that

they actually the drops is almost Low them
momentum upon impact it may be only two to

seven percent seventeen percent on
contrast if you take a dragon fly it will

lose about ninety percent because that's
actually several times a drop weight.

And because they lose very little mental
as I'll show you the force is very low.

So this is very non-intuitive it's very
different from the impact on a bird's wing

where the drop is spread here.

The force and the mentum

afterward depend strictly on the weight of
the mosquito compared to that of the draw.

There are some cases here where
the theory here doesn't fit very well and

that's the case if the drop hits the side
of the mosquito move back and spins it.

There you have some dissipation and
it is all you've got lower final speed.

So these are two mimics of the same size
compared drops a very different size

particular Here's what would be drizzle
and here's would be a typical rain.

And as you see what you would expect
on a solid surface is that of a drop.

It's that it should deform greatly
if it's large but in this case.

What determines the force is
determined by the mass of the mimics so

the small drops happen to form or
other small drops to form or

it's the large jobs that actually
have the largest number of G.'s.

So the number of G.'s involved here
is around is around one hundred.

Now I'll get into that in a second here.

That's OK.

So to predict how

many G.'s there are I first have to show
to prove that the drops don't splash and

I'm not going to go into this in
great detail but basically what keeps

the drop together is the surface tension
and any time they drop the forms.

It generates increased
pressure inside the drop.

The the so there is a this
is the rest of the equation

basically the momentum of the impact
which is the change in the drop speed.

Over the time of impact of the last
pressure due to the increase in

surface tension and
gravity which is noticeable in this case.

The defamation time based
the amount of time for

the drop to the fore in can be estimated
can be estimated by the change

in speed over how much the drop the forms
in radius and the pressure gradient can

be estimated by the change in pressure due
to changing from a sphere to an ellipse.

Using these together we can base a project
how much of this drop will change

in radius.

As a function of the masses
of the two items.

And if it's almost precisely our data.

Stating that basis we can predict
how much the drop the forms.

Depending on the size of the insect.

In particular the large the larger
the small drops will to do for

more as shown up in the left.

So we see that drops to
form rather than splashing

that means you can assume they're in a
lasting impact that they can go together.

And this is the number G.'s.

That is measured from these
experiments that I showed.

So it's actually quite high.

The number of G.'s.

Let's let's calibrate this for you.

The the human tolerance
was taken on a rocket sled

this guy basic took this rocket sled at
a high speed and stopped it suddenly.

He he suffered he suffered
about twenty G.'s which

is known to be the human limit and
they called it.

They call it eyeballs out because
I was detached from his retinas So

he kind of had trouble
seeing the rest of life.

So that's about twenty G.'s
how much you can withstand and

when I get to this dog work.

I'll show these a dog.

These are hairy mammals actually
shake much higher than that.

And so they have to close their eyes
to their eyeballs from popping out.

But I'll get to that in seconds.

But the so depending on the size
of the drop size the drop you

can basically get between fifty and three
hundred G.'s honest on a single mosquito.

That's it may seem like a lot but

you have to keep in mind these
mosquitoes a very strong exoskeleton.

And three hundred G.'s when you're
a very lightweight mosquito is

only about two point six grams of force.

So that's essentially like
the weight of a feather that

that's falling on the mosquito.

So it's actually very low.

We've done basically
impaction experiment where

you basically squeeze a mosquito on
a analytical balance and you can see.

Five base in order of
magnitude higher than that.

So so they basically have no
problem surviving depending on.

Depending even on many
large range of drops items.

That's so I think I'm going to so

this is a this is a video of
a mosquito being hit by a jet.

There's still a lot of
things we don't know for

example if it's by a series of drops
how the mosquito will behave here.

The number of G.'s is only one hundred.

So even though it's hitting being hit by
a jet that's going ten meters per second.

The number of Jesus actually
is still quite survivable.

I think going to get some of this stuff.

So OK

OK It's OK.

So we've seen two methods of water
pellets the one based on construction

of the animals with with each
other the other the animal I'd

like a particle is so small that base of
the drop just carried itself with it and

as a drop in the Splash
an animal survived.

Now you one thing sort of tension is
useful on the larger scale but turns out

drops for example the living animal have
to depart from small bundles of hair so

surface tension as I'll show is
important in determining how quickly and

I will have to remove water and

as Lord knows I got I a cook I got
I approval to do this right or

so on and the last ten minutes let's talk
about how and will shake to remove water.

So it's actually harder than you would
think if you go to the if you just shut

the door of your office and go on all
fours as I do on do on occasion you try.

To do this is actually quite difficult.

And the reason is well
you have a four legs

the front legs basically turn clockwise
the back legs turn counterclockwise and

you've got to do a self wringing motion so

that your body rings itself out and
spins at high frequency.

The result is quite spectacular as
I show they do it at various speeds

the number of gravity's involved
which we can calculate.

Is basically doing ten and seventy.

So it's for these animals is a lot
larger than the human tolerance and

if you shake your head.

Very quickly from left to right.

It does it does hurt.

Moreover it's actually quite effective.

So I like to compare these things
the washing machines even though

they're very different processes but

your water machine as you know will draw
your socks within about ten minutes.

But this gives the animal seventy
percent dry in about one seconds.

We can measure the weight
of the animal before and

after determine how much how dry it is and
for rats.

They basically got almost all
the water off instantaneously.

So it's very very good.

So I'm going to talk
about how it happens and

how it changes depending on animal size.

First of all why is it even
there in the first place.

It seems like it's just really
cute little phenomenon.

But in actual In actuality it's a matter
of life and death for these animals.

There's a great article one
hundred twenty eight by J.B.S.

Haldane he was a geneticist
called on being the right size.

It's a great article for your classes to
talk about the consequences of scaling

up with article says so if you come out
of the bath tub and you step on a scale.

You'll find that you have weighed
a pound heavier due to the water on you.

The water is basically forming a very
thin layer a fifty of an inch thick and

that altogether weighs about a pound.

Which is nothing.

But as you get smaller and
smaller this weight increases.

So a webmaster it turns out has to carry
about its own weight of water when it

comes out of a bath.

And a fly has to carry
many times its own weight.

You can see the problems of losing
this water become increasingly

difficult as you get smaller and smaller.

This is coupled to the problem that
water takes a lot of energy to evaporate

especially in cold weather.

For example one pound of water takes
around two hundred fifty calories

to evaporate from the body.

That means an average sized dog which
weighs about sixty pounds with a pound of

water nerds for

one can calculate that it takes about
twenty percent of its daily calories.

Just to draw that water off.

So this is a very big
deal in cold climates.

This is a typical experiment our lab.

We take a pink strong and
we do for particle tracking so.

This is a troll that gently taped
to the back of the dog and.

You can see that's the actually
the back of the dog.

So the bear back of the dog is going
ninety degrees in the left and

nine of us the right.

The dog skin is very very loose.

If those of you who've
taken a freshman physics.

You might remember this is
a simple harmonic motion.

With an amplitude of
around ninety degrees.

If you take the dog skin and
move it yourself.

You see it's very nice it is very loose
and it can move about sixty degrees on

its own which means the backbone of
the dog is only moving thirty degrees

of poultry poultry amount so
the back of the so

the berry back of the dog the vertebrae
is very moving a very small angle.

This means that if animals had tight skin

their philosophies would
be one one third as high.

If they had loose skin and

as I'll show later the centripetal
force will be only one tenth the size.

So it's critical these animals have loose
skin in order to get the high speeds

associated with this large amplitude you

can verify that they're
the amplitude of the back.

It's small and these high speed X.
ray videos done at young huge bangs live

you can see the background doesn't move
doesn't move very much as they shake.

OK now that one of the in

one of the most interesting about biology
is that diversity is the spice of life.

So these animals have to get they've got
to be asleep go from a very small size

the size of a mouse all the way to
a grizzly bear and still figure out how

to get the shaking mechanism to work
it all these different sizes so

I try to design a washing machine
that works in the pocket and

works as big as your car and
let's see how that how they do that.

This is the smallest shaker in the world.

It's a mouse and it shakes quite fast.

Thirty times per second.

So in a single second it will shake thirty
times and you see as effective at removing

water you get a little
larger This is a rat is a.

Shakes about twenty times per second.

So it's slower than the mouse but
still still capable of removing water.

You know it's the mice and

the rat they tend to lift their
front paws up off the surface.

The other animals keep their
four paws in the ground.

We don't know why that is the other
thing you'll notice is they keep

their eyes closed.

I like to call the shaky bacon.

This animal.

It's a pig it shakes at ten herds so
ten times per second.

So this is becoming in the realm
of what is reasonable.

You see is also pretty good at
getting rid of some of these

hay particles and things like that.

We were

surprised at how good the sheep
was sheep of lanolin in there and

their fur which makes
a pretty water appeal.

But you see it's actually
quite quite good shaking

surprisingly good.

So that's only six times per second.

So we're getting slower and slower and

as and
the animals are getting more ferocious.

So we went to the zoo this is
another animal from the zoo we.

And by we I mean my grad student and hos.

We filmed this animal this is a bear
as about four times per second.

Of course the most
ferocious animal of all.

Humans.

So you can go you can try this at home
you can only shake your head at most

two to three times per second.

Is that you know no she she didn't want
to be a social with our experiments.

So there's this trend this clear and

obvious trend that the small
animal shaking faster.

Why why is that.

Now my answer is that all these animals.

If you're a drop in an animal you don't
know if you're on a grizzly bear or

you're a mouse.

So the drops have the same
physics associate with them.

And in particular and
the animals hair transplants or form.

Bottles for these animals to
lose this water equally easily.

They basically have to have the same
central force to fight the surface

tension of the draw.

Because the animals gets smaller
you get a smaller radius with

the with decreasing size they
have to compensate by increasing

their frequency to increase
the central force.

So let's see if this this
hypothesis matches the experiment

with the first is that
this this this is a fact

found by McMahon how basically the mass
of animals changes with their crossing.

The radius.

Basically is smaller and lighter animals
have a smaller radius as you expect.

The second is found by Tate he
was a pharmacist actually he.

He actually invented I.V. dripping.

But basically he found that if you have
well he didn't actually use a tuft of hair

but he used a tube of varying radius.

If you have the radius to Tube you can
predict precisely how heavy this job is

going to be one of the jacks.

And the force balance that
dictates the size of this drop

is the central force which depends on the.

The velocity squared or the frequency and
amplitude of the shake.

And the radius of course.

And the surface tension which is basically

the circumference of the drop to
pod times the radius of the drop.

Here we use the radius of the hair clump

because that's where it's
contacting the solid.

There is not a function here to the fact
that some of the drop is left behind

as objects.

But this relation between the center
before the surface and in states

shows you how frequency of the animal
should scale with radius of the body and

we substitute radius for
mass because it's easier to measure.

So this day this prediction States as the
animal frequency as the animal mass goes

up the frequency should go
down to the negative exponent.

There is how the experiment prediction
matches with the experiments

we studied about thirty five animals
sixteen species and eight dog breeds.

And they all fall along
this pretty nice fit.

It's pretty good.

If you see this is five orders of
magnitude in mass it's rare rare you can

have such a large X. axis the Y.

axis goes between it was three to
about thirty times per second.

The best fit to this line
is has an exponent of

point two two where
the theory is point one nine.

So it's quite close.

Considering that we didn't take
anything of the animal into account.

We only took the physical.

Drop.

Some you might wonder well maybe this is
just do the larger animals tend to go.

They tend to go slower just
because they're size but

scaling from the editor for
example on the trot gallop transition.

They see how quickly
they can move their legs.

When they transition from trouble from
a trot to a gallop they have a very

different exponent point one four
compared to point one zero point two two.

So it's not simply due to the fact
that they're larger they actually

believe they're actually
tuned into these drop sizes.

So we believe this finding in the future
will be useful for autonomous robots.

There is some nice examples of autonomous
robots for example this Mars rover.

There are basically lose their
capabilities because of exposure to dust.

In this case the power of lost on these
sensitive solar panels due to collection

of dust and only regain power once a wind
swept through and blew the dust free.

So for a robot to be truly
autonomous it really has to learn to

remove particles from itself.

Another inside this study
provides is that this is a this

mechanism is really terrible for
the fluids of different properties.

I call this an engine mechanism
ill suited for modern times.

Basically it is very mammals they evolve
loose skin to aid with locomotion

shivering eventually turn to
shaking with a combination of for

this all the only ways to get rid of
water are dense oily hair like these

otters do or to have this inertial
shaking at a very high frequencies

where centripetal forces
beat surface tension.

But you can you can check pretty easily
that this doesn't work if the if the fluid

has a very long time scale for
example crude oil drips very slowly and

that is all any animal that's coated in
a high viscosity fluid will not be able to

do this and that's a large reason why
these oil spills are such a problem.

So I'm going to try to wrap up.

I talk about three very different topics.

But they all dealt with surface tension.

We saw that basically synthetic
fabrics and plants they

basically have a combination of chemistry
and bumpiness that acts to repel water.

And animals are because
they're able to move and

have all these different
properties from plants.

They've learned to of how water in very
different ways the ants they can build

services by linking
together these mosquitoes.

Their mass dictates the impact phenomenon.

And these mammals they do what we
call active water repellent see where

this is a new idea in the literature
you use energy to remove drops.

And in terms of waiting for
evaporation it's actually very effective.

So I will leave you with this thought of
active water policy which is involving

construction or use of energy compared to
passive water policy involving structure.

A lot of the reason I was able to do this
work is because my joint position in

biology in which I basically teach a new
class every two years based on and six.

So that I taught a class two years
ago on and on and it was called and

lab this year were teaching
a class on insect flight.

They'll be a poster session based
on their work in about two weeks.

If you're interested in learning more and

another thought I'd like to leave you
with is that I think these images and

these problems are a great way to
teach physics to young audiences.

This is this is one of the articles
that was written on our work and

see these these cute little ants.

I think it's I think kids
really like this stuff.

So I'm with that I'm happy to
open the floor for any questions.

Thank

God.

That's a good question.

I get asked that sometimes that can't.

The answer so

going to cooperate in can they actually
like road themselves all together.

It seems that hasn't happened yet.

In part because so

they had to build a raft very
quickly within two hundred seconds.

They built that you imagine
have this flash for

They've got to build it almost
instantaneously as well into the answers

randomly oriented so each of them
is facing a different direction.

Some of them are maybe even sideways and
so their legs are all pointed in different

directions so they can't seem to I think
if they're pointed all in the same

direction maybe they might
be to row together but

because they're all pointed
in opposite directions.

They don't seem to be able to coordinate.

It is very the end raft I mean if you play
with it in the lab it's surprisingly.

Drag drag it has very low drag.

So you just blow it.

This thing will just simply it will
simply just glide very easily and

because it's mostly air.

So it is it is quite possible if they
could just basically coordinate themselves

but they just they just
only are oriented properly.

Yeah.

It takes it takes them longer to
disassociate but they do have to do that

especially when they get to land when they
basically get to to land they only have

a couple seconds to realise lands there
and then they have to grab on to land.

So they do have to sense land is there but
it's not as quick partly.

A lot of the so

one basic thing about the swarm systems
is that they only have local information.

So only the ants touching
land will know that there.

They have to get there.

Whereas in the rat.

They're all just trying to get the hell
out of there as quickly as they can.

So they're all running
in random directions.

So they don't really need to have
this directionality associated so

it is slower but
they do have they do disassociate.

Thanks to.

The There's one thing that's always
struck me and a lot of these experiments

are actually quite difficult to do because
any small traces of Sir fact in these for

example you're if you're close you
go like this with your clothes.

They're small but you detergent anything
any of it is small resist residues

will cause these rafts to sink
immediately and everyone to drown

Basin the plaster on layer is dissipated
if you have lower surface tension and

they basically can't breathe
underwater any longer.

So in the lab it's almost
it's almost impossible

to do unless one is very very careful.

When the while there's
a whole bunch of biofilms and

other things on the surface that
would also think that would rule.

Lower the surface tension.

These animal spent a great deal of
time cleaning themselves in the wild

they'll clean themselves basically
half the time so I think partly

it's a deal with those kinds of things but
it's it's a good question.

I think is a question of chemistry and
but I think it's it is pertinent but

we don't really know
the answer to that when

the water is only three percent lower
surface tension than fresh water which

is a high surf attention of of
water combine with anything else so

they can flow in salt water and
often they'll actually they'll float for

weeks some because they're on
this in the panels of Brazil so

sometimes they get swept out of the ocean
so that someone is off the float for

weeks before hitting land and
now if they start eating shell there.

And that's been documented.

Yeah.

In the back.

So I always I'm when I describe
the AFT I always think that it's worse

to be on the bottom because
that's where it's wet and cold.

And there is turnover for
example we try to take a picture one.

Of what the raft bottom looks
like is kind of a Zen thing.

And what we did was pick
off the mass of the top.

But what it turns out is that they ask the
bottom will only stay there if they're And

from the top.

So as you pick the ones on
top the whole thing shrinks.

And which states the end sort of they and

this is been documented before
they have little sensors.

For example.

They'll stay in these configurations only
if they feel feet on top of them as soon

as you don't get enough feet on top of
them they'll release their neighbors.

But also requires a lot of
coordination we don't understand.

For example if you're on the bottom.

You've got you've got five or
six neighbors and

the links are one way one someone grabs
you or you grab somebody but somehow.

If they want to escape they can
they can sort of tell their

friends I want to escape
when they like go.

So it's it's.

It's not clear how they do that but
they they do exchange they do exchange.

If there's less ants on top and

sometimes they'll just exchange randomly
if there's no one walking on them.

You'll see a hole in the raft and
eventually they'll get filled.

Yeah.

The

question is why what sets
the equilibrium height of the raft.

Well.

I think.

I think part like I said if I think
partly there are some strength issues.

If it becomes a single layer it's actually
not very strong and they have this rule

that the they have to be underneath other
and so there was not just escape but

there's I mean there's also evolutionary
pressures and things like that if there

were a third layer there basically
an extra safe there because they were

basically birds picking the ass off from
the top and fishing and from the bottom.

So I think if there were a third
there everyone would want to be

sandwiched in the middle.

But if there are basically people
on the top there to being in.

On the bottom being in
that is sort of more fair.

So I think somehow that's what ended up.

So.

There's I think there's a probably written
Point reason why there's not more than

two layers.

We document they build these towers and
things like the big arcs there's many

layers in the bivouacs So they're
capable of building very many layers.

But they just don't for the RAF's I
think is do this because of these

these these pressures and
things like that.

Yeah.

I just reviewed a paper
on this last week and.

These these fire ants.

They're very territorial so
one colony will basically.

Basically is very seldomly aligned
with another if you have two rats

from different colonies they'll
just totally be like pirate ships.

They'll totally at their base and
link together to form this mound and

then kill each other and
everyone will die.

There is no survivors.

So you don't want to rouse
them touching each other.

If you do that then I think it's OK
as long as they're from the same

they they can sense
almost instantaneously.

If there is someone from the same or
different colony.

Yeah.
In the back.

I think.

I actually I don't know
to be tough to study.

First of all but.

I do think you are somewhat protected if
you close them but maybe if they also get

dizzy if they keep their eyes open and
look at the world going that quickly.

But they they do they all tended
they all tend to do that.

So it's it is a quite a large number
of G.'s I mean if you can imagine

moving thirty times for a second.

That's it that's quite fast.

So there's not really much President for
animals dealing with that many G.'s.

Maybe if there were we could see
if they also close our eyes.

Yeah.

Yes yes that's true.

Our model base is just consider
the base of the the top of the back.

That's the only place you
can measure the amplitude.

And I as I said we measured how the rats
are after the shaking and the base and

thirty percent the water remaining I think
a lot of that is on places that they're

basically not they're not be able to
move like on their hind quarters and

things like that they generally
they stay somewhat but

those are sort of away from
the core regions of their body.

So maybe that's less less
per person per day to them.

I mean sometimes you'll see dogs.

They'll sort of whip their tail afterward
to get rid of the watermen tail but

they definitely don't stand on their
front legs and shake their back as well.

So there.

They've basically I think they've done
as best they can but you're right there

are some parts that are probably remain
sopping softening but the thing about

capillary is if you have parts on the dogs
that are wet they'll sort of seep sort

of equilibrium sort of seep in me and
make the whole dog medium leeway.

So I think if they wait a little while.

Eventually they'll be to shake again and
that water will be a bit come off.

Well thanks for all your questions.

I hope you enjoyed the talk.

Thank you.