back to article Physicist unmasks 99-year-old mistake in English dictionaries

An Australian physics prof has discovered a 99-year-old error in the Oxford English Dictionary - repeated in most dictionaries worldwide - and is having it corrected. The error is in the definition of the noun "siphon", a tube used to draw fluid from a higher location to a lower one - as when emptying a vehicle fuel tank, an …


This topic is closed for new posts.
  1. brimful


    Why didn't I think of that. I could've been famous, won a Nobel Prize, been immortalised in science books to come, had huge amounts of money, and finally proven that you don't have to be a scientist to be intelligent. Right, I'm off to browse the dictionary to find other errors and have them corrected.

    Beer, cos I would like to use gravity to siphon it into my liver as the little ascii diagram below deomntrates


    Table ---

    | < pipe

    Me lying down

  2. Anonymous Coward
    Anonymous Coward

    Yes But.....

    Does it explain how to get the taste of petrol out of your mouth?

    1. Hardcastle


      How about dookie?

  3. Number6

    But would it work in a vacuum?

    Apart from a bit of surface-tension, atmospheric pressure is an important part of the process. because as the liquid travels down the long tube, it creates vacuum behind it. Air pressure on the liquid reservoir encourages the liquid up the short tube to fill the vaccum.

    Another interesting test is whether it is possible to siphon water up a hundred feet (that's 100ft above the surface of the reservoir), over a wall and down a thousand feet the other side. I bet it would manage to get up about 33ft.

    1. Destroy All Monsters Silver badge
      Thumb Up

      My thoughts exactly siphon on the Moon I would think.

    2. Anonymous Coward
      Anonymous Coward

      how about

      I start the siphon off with a pump instead? The liquid exitting into the lower (destination) reservior will have directly pulled more into the tube from the source without any requirement for air.

      I doubt the gigalitre siphon in South Australia has got some bloke sucking on a tube on the promise it's beer at the other end.

    3. Trevor 10
      Thumb Up

      I bet it would manage to get up about 33ft

      Lucky guess or did you do some research?

      The limiting factor in terms of how high you can raise the water from the upper reservior before the drop down the long lenght of tube is the vapor pressure for the liquid.

      The higher the siphon goes, the lower the pressure of the liquid in the tube at it's highest point, if the pressure gets low enough, bubbles form in the water and the siphoning effect is lost.

      For water, the vapor pressure point when at normal atmospheric air pressure is about 10meters/33ft.

      1. Number6

        No guess

        It's just one of those numbers I happen to remember. 33ft underwater the pressure is doubled, and I remember the old barometer stuff from when physics lessons had real physics in them.

    4. The Cube

      Work in a vacuum? - Yes and No

      The first question you should ask is "what is the vapour pressure of my liquid?" and this should tell you whether it would evaporate (or more likely freeze and then sublime) in the vacuum. Most liquids would be creating a localised "atmosphere" as they evaporated away into the (near) zero pressure of the vacuum which is likely to be below the vapour pressure of the fluid at whatever the local temperature is.

      The force driving the siphon is gravity, surface tension is more likely to be important in a capillary tube, these can achieve spectacular lift heights without any siphon at all and are popular in nature.

      The vapour pressure leads to the answer about how high a siphon tube you can use, the 100ft high pipe would be likely to reduce the local pressure of the liquid at the top to below the vapour pressure at which point the water would "boil" inside the tube. This would cause the siphon to fail by effectively air locking the syphon. The process is that the masses on each side of the trapped gas section would end up balanced against each other as if they were on a pair of scales. Only when the fluid passes over the top with sufficient density (i.e. generally not as a gas) does gravity do the job for us.

    5. Tom 35

      Water in a vacuum

      You can't have water in a vacuum, it would boil. Same for a 33' column of water, it would boil.

      Atmospheric pressure is an important part of the process, it keeps the liquid from boiling.

      1. Chemist

        Re : Water in a vacuum

        Separate the thermodynamics from the mechanism

        Gravity drives the thermodynamics - the potential energy lost is the product of the mass of liquid moved and the height difference between the reservoirs.

        The mechanism requires that the 'hump' over which the liquid has to climb is such that the liquid will not form a vapour bubble at the highest point - this point will depend on the atmospheric pressure. With a high-boiling liquid and a low hump the 'outside' pressure could be quite low. Practically this would require the entire app. to be a partial vaccum chamber

        In a normal environment, of course, atmospheric pressure is related to gravity.

    6. Anonymous Coward
      Anonymous Coward

      What if...

      ... you have a 1100ft flexible tube, start the siphoning with the bend just above liquid level, then slowly raise the bend up to 100ft?

      1. Tom 35

        What if... no luck

        When you got close to 33' the water would boil and the siphon would fail.

        Unless you were in a VERY big diving bell over 66' down...

  4. Popup


    Of course it requires atmospheric pressure!

    There is no way for the water in the downhill slope of the siphon to 'pull' water upwards in the initial up-hill part, and for that part you NEED the external atmospheric pressure to instigate a pressure gradient in the initial part of the siphon.

    Sure, it's gravity in the 'downhill' part of the siphon that exerts a partial vacuum in the upper part of the siphon, but without the external pressure it wouldn't work.

    The linked document is also full of rubbish. It insinuates that the cohesion required to 'pull' the water up the hill comes from hydrogen bonds! (snort!)

    It would be very easy to install a manometer in the topmost part of the siphon and show how the pressure changes.

    Also, it's childs play to show that you can't suck water more than 10m uphill, a figure that maybe just happens to coincide with the atmospheric pressure? Please!?!

    Even the wikipedia gets it right, unlike this Oz PhD. (Wonder what mail-order university he got his certificate from.)

    1. DPWDC


      Forgive my ignorance, but submerging a hose then just sticking your finger over the end until the hose is below the tank water level works just as well as sucking.

      Wouldn't that be a mechanical process rather than atmospheric?

      1. Boring Bob

        Nope, still atmospheric

        Nope, it's still atmospheric. When you lift up the tube you are applying a force to the tube to lift it, but you are not applying any force to the water inside the tube. It is atmospheric pressure that is lifting the water as you lift the tube.

        If there was no atmospheric pressure the water would just run out of the tube as you lift it.

        1. Anonymous Coward
          Anonymous Coward

          Don't see the connection

          A key point seems to be whether a liquid can act as a chain, i.e. can you pull something with it. It seems to me that you can. Note how a syringe is used to draw liquids from containers. Imagine two syringes fused end-to-end with quantity of liquid in each of them (forming a continuous column, i.e. no air bubbles). If you pull on one syringe out, the other will be sucked in (and vice-versa). If two people pulled at both ends, the stronger would pull the weaker forward. The enclosed liquid is acting just like a chain, and air pressure will have no bearing on this process whatsoever.

          The siphon is working in the same way. Gravity is pulling on both columns of water (like a syringe at both ends) and the highest column will have the greatest pull.

          I'm guessing there is some limitation, however - if the pull is great enough, presumably the liquid will change state to a gas. Comments?

          1. Rattus Rattus

            Two syringes

            Suction is an illusion caused by the push of atmospheric pressure. In your example, if you pull the end of one syringe out, that reduces the pressure inside the syringes, allowing atmospheric presssure to push the other end in. Do this in a vacuum and the other syringe end won't move.

            1. Anonymous Coward
              Anonymous Coward

              Re: Two syringes

              Quote: "Suction is an illusion caused by the push of atmospheric pressure [...] Do this in a vacuum and the other syringe end won't move"

              This will still work in a vacuum (assuming the syringe is strong enough to maintain its structural integrity). A syringe doesn't draw liquid because the atmosphere is pushing it. If you're in any doubt, part-fill one, put the end against your skin, and pull. What you will is suction from the syringe (pulling), not pressure from the atmosphere on the other side (pushing). Remember, atmospheric pressure will be constant on all sides of the syringe irrespective of movement, so it has no bearing on its action. Also, bear in mind that the atmosphere is a gas, which (unlike liquid) can readily expand or compress.

              1. Rattus Rattus

                Syringes and suction

                "put the end against your skin, and pull."

                Ah, but it's still pressure. This time, it's the pressure of the inside of your body pushing into the decreasing pressure inside the syringe. Your body's internal pressure is normally balanced against the pressure of the atmosphere outside. As you draw the air out of the syringe, it's internal pressure becomes less than this, so your skin is pushed in, by your own body.

                This effect can also be seen if you seal a container of hot air and pour cold water down it's outside. The container will constrict as the air inside has a lower pressure when it cools, so the air pressure outside is no longer balanced, pushing in on the container.

        2. Nexox Enigma


          """When you lift up the tube you are applying a force to the tube to lift it, but you are not applying any force to the water inside the tube."""

          So you're saying that you can move a container by applying a force to it, but that force doesn't transfer to the contents, a generic 'atmospheric' force just happens to cause all that mass to accelerate in exactly the same direction you move the container?

          Everyone seems to be an engineer today... Atmospheric pressure exerts force on open surfaces of a fluid, normal to the surface of that fluid (happens to be parallel to gravity in most cases.) Atmospheric pressure doesn't change much over distances that you could reasonably use a siphon (partial pressure of water would limit you to 10m, as mentioned above,) so the atmospheric forces on both bodies of water (one higher and one lower) is the same, and they cancel out.

          And hydrogen bond /are/ important (someone argued that point earlier,) otherwise water would have a significantly lower partial pressure, and likely not be a liquid at any sort of reasonable temperature.

          Now if you haven't taken a fluid statics class, please stop sharing your special versions of physics.

    2. Anonymous Coward
      Anonymous Coward


      You're talking about the mechanism for lifting water to the high point of the siphon prior to its operation, not the action of the siphon itself. If you are lifting the water by suction, atmospheric pressure pushes the water up the siphon as the pressure at the other end is reduced. But that isn't what the article is about.

      Note that is is not necessary to start a siphon in this way. If the column of water in the siphon is entirely submerged and then one end is moved away to a point lower to the surface of the liquid (keeping the column sealed), the siphon will start working irrespective of the height differential.

      And the reason the reason the siphon works - as this article is pointing out - is entirely due to gravitational force. In simple terms the mass of water in the column on one side exceed the mass of the water on the other side, and hydrostatic pressure at the top (caused by the gravitational force on both columns) will draw the water up the shorter column.

      BTW - Could we all aim to be a bit gentler in our postings?

  5. MnM


    as the credit expert gazillionaire would say, 'they know thaaat!'

  6. Raindancer

    Shorter OED seems OK

    My Shorter (two-volume) OED says:

    "A pipe or tube used for conveying liquid from one level to a lower level, using the liquid pressure differential to force a column of liquid... etc". It doesn't specifically say gravity, but that's what causes the pressure differential, right?

    It never occured to me that the Shorter would use different definitions for a word than the Compact...

    1. ajcarr
      Thumb Up

      New Oxford American Dictionary

      The NOAD (2nd edn), which is bundled with Mac OS X, has the following, correct, definition:

      siphon |ˈsʌɪf(ə)n| (also syphon)


      a pipe or tube used to convey liquid upward from a container and then down to a lower level by gravity, the liquid being made to enter the pipe by atmospheric pressure.

      • Zoology a tubular organ in an aquatic animal, esp. a mollusk, through which water is drawn in or expelled.

      verb [ trans. ]

      draw off or convey (liquid) by means of a siphon.

      • figurative draw off or transfer over a period of time, esp. illegally or unfairly : he's been siphoning money off the firm.

      1. Hardcastle
        Paris Hilton


        ...Paris could put the theory into practice...

  7. Graham Marsden

    Or perhaps...

    ... nobody else really gave a monkey's...!

  8. Richard Tobin

    You're oversimplifying

    If a siphon doesn't depend on atmospheric pressure, why can't you siphon water up more than 10m? Clearly gravity is the source of the energy, but it works by the weight of the long side reducing the pressure at the top of the siphon so that the liquid on the short side is pushed up by atmospheric pressure.

    It is reportedly possible to siphon in a vacuum if the liquid is thoroughly degassed, relying on the tensile strength of the liquid, but this is not the main mechanism in normal siphoning.

    1. Boring Bob

      A tissue

      "It is reportedly possible to siphon in a vacuum if the liquid is thoroughly degassed, relying on the tensile strength of the liquid".

      This is why you can siphon over small heights with a tissue. The tensile strength of water causes capillary actions between the fibres which will "suck" the water up a little bit, hopeful enough to get over the barrier and to allow siphoning to begin.

    2. edward wright

      @Richard Tobin

      Oh how I wish gravity was a source of energy, no need to burn nuclear or fossil fuels, it'd be just great!

  9. Roland the Headless Thompson Gunner

    isn't atmospheric pressure also necessary?

    Gravity makes the fluid flow out of the longer leg. But why doesn't a vacuum form inside the tube, above the level of the fluid in the upper container? Isn't that because of external atmospheric pressure?

  10. Anonymous Coward
    Anonymous Coward


    As the liquid falls out of the longer part of the tube, it creates a vacuum / lower pressure in the tube, and it is in fact the air pressure pushing the liquid into the tube from the top tank?

  11. jackharbringer

    Not strictly a dictionary...

    "An extensive check of online and offline dictionaries did not reveal a single dictionary that correctly referred to gravity being the operative force in a siphon,"

    ...but read all about it on WikiPedia:

    Also answers why 10m for water is the limit, under what circumstances a siphon will work in a vacuum and what role exactly gravity and atmospheric pressure play.

    I know it's in to bash the wikifiddlers around here, but give them some love when it's due. :)

  12. Chris Weatherby

    Bollocks Reg

    You're insulting engineers by publishing this.

  13. Anonymous Coward

    I don't know why

    ...but the word "gigalitre" makes me chuckle. It just sounds dirty.

  14. Anonymous Coward
    Anonymous Coward

    Pretty sure

    My physics teachers said it was atmospheric pressure too. Atmospheric pressure pushing the liquid up into the vacuum created by it falling away down the other side.

    it would be interesting to see if you could siphon a liquid out of a completely sealed container... I'm guessing not, but i don't have a handy experiment to check.

    Would it work without gravity? Probably not, but that doesn't mean that it's the main principle at work.

    1. Charles 9

      Excellent counterpoint...

      If atmospheric pressure was not a key factor in siphon action, then it would be possible to siphon out of a sealed container wherein the only possible inlet/outlet for the fluid would be the tube (example: a soda bottle with a tube running to the bottom, run through a hole in the bottle cap which is then caulked and screwed tight onto the bottle). Willing to wager a drink the siphon action doesn't work all the way because internal pressure decreases (because volume increases as the liquid flows out and no more air can come in) external air pressure becomes great enough to stop the flow.

      1. Nexox Enigma


        """If atmospheric pressure was not a key factor in siphon action, then it would be possible to siphon out of a sealed container wherein the only possible inlet/outlet for the fluid would be the tube"""

        Another non-engineer then?

        A siphon isn't going to be able to create a vacuum, as it isn't a pump. There's no way to remove the fluid from your container without replacing it or creating a vacuum.

        By adding a closed container, you actually cause atmospheric pressure to become relevant - by removing it's effect from one body of liquid. In your standard siphon, both bodies of liquid experience pressure from the atmosphere, and that cancels out, but when you seal the container, the atmosphere cannot exert force on the surface of the liquid, which means that it will not cancel out, and your siphon will break, as you expected.

        Has anyone even drawn a force body diagram of a siphon? I suggest you try and see just how nuts this atmospheric pressure suggestion really is.

  15. lukewarmdog


    Everyone knows Australia is upside down being on the opposite side of the world, of course their water works differently.

    And apparently the only thing keeping them from drifting off into space is alcohol.

  16. Chris 3

    Simple experiment...

    Fill a beer glass with water and invert it under the surface in a filled sink.

    Gradually lift the glass and marvel out how the level of the water in the glass is above that of the sink.

    Now tell me how atmospheric pressure has no impact on the siphoning process.

    1. BristolBachelor Gold badge

      I failed

      I got as far as "Fill a beer glass"

    2. The Cube

      Yes, this is an experiment to test vapour vs atmospheric pressure

      What the (quite revealing) beer glass experiment is doing is finding the equilibrium point in the system you establish.

      As you lift the beer glass you are lifting the contained water (that is above the main water level), this requires force proportional to the amount of water you lift (how far out of the water the glass is).

      Meanwhile we have the atmosphere pressing down on the exposed water surface (not the water in the pint glass though as the glass is in the way).

      As you lift the pint glass up you will find that any trapped gas in the top of the glass seems to expand (do this in hot water to see it better). This is due to the pressure reducing in the pint glass as some of the atmospheric pressure is offset by the weight of the lifted water trapped in the pint glass. The trapped gas is not just expanding as the pressure is reduced though, that air is a mixture of air and water vapour, as you lift the glass and reduce the pressure more of the water turns to vapour and your gas appears to further expand. Do this with a tall enough pint glass (10m or more) and you'll find that there is a point at which further lifting of the glass simply results in more gas at the top and no further rise in the water level.

  17. Rogerborg

    Typical lying Ozzie

    As any fool knows, the operative force is a chav sucking the petrol out of your tank.

  18. loopy lou

    PhDs all round for Reg readers

    Of course siphons need atmospheric pressure. The maximum height you can have bend above the inlet is even proportional to the atmospheric pressure. No atmosphere, no siphoning.

    As for the good doc, this extract from his paper demonstrates a truly mind boggling degree of stupidity:

    "The column of water acts like a chain with the water molecules pulling on each other via hydrogen bonds"

    Perhaps he should correct his own misconceptions before addressing everyone else's...

  19. Richard Conto

    The Map is Not The Territory

    Sadly, I can't get past the pay-wall to see the article, but the summary suggests that a "chain model" is the best way to mathematically model a siphon. Now, I can't comment on that - and for all I know, as a model of the flow rates, etc. of a siphon, it may be the best - suitable for video game simulations, etc..

    But I can say, from practical experience using old garden hoses to drain puddles at my childhood home, that atmospheric pressure isn't irrelevant. If you have a leaky hose or hose-coupling between the water level you are trying to drain and the top of the siphon,. you'll loose your siphon effect - and even if you have a leak on the lower end, you'll find a reduced siphon effect.

    Any physics explanation of the process that claims that atmospheric pressure is unimportant is flat-out bone-headed wrong. Sure, surface-tension (or at least the electro-static attraction between different sides of the H2O molecule) might help maintain a siphon in a vacuum - but as any pump maker/operator knows, there's a limit to the "drawing" power of a pump.

  20. Ash!


    ...isn't atmospheric pressure a side-effect of gravity?

    1. Joe H.


      ...atmospheric pressure comes from all the gas molecules that are piled near the surface of the earth as they are drawn towards the center of the planet by....oh, wait, never mind.

  21. Individual #6/42

    Atmospheric pressure

    The atmospheric pressure controls how high above the surface of the upper liquid the siphon tube can pass before it descends. I think you'd have trouble siphoning mercury above a metre wall.

  22. Stu


    It seems we have a bit of a connundrum here, nobody on these hallowed sci expert forums ;-) seems to be able to come to a concensus whether gravity is really involved or not.

    I call for a scientific experiment as part of El-Reg's PARIS project, entitled "Water siphoning experiment 1A in a microgravity environment - is air pressure or gravity the main culprit?"

    Just what is the cargo limit onboard the paper aeroplane experiment anyway? ;-)

    1. Number6

      You do need gravity

      You do need the gravity because the weight of the water in the long tube is what drives the process.

      1. Sooty


        in a gravity-less environment, you could substitute a ferrous liquid and a magnetic field.

  23. Paul Bruneau

    Side effects of gravity

    The earth, the sun, the earth's magnetic field, are all side effects of gravity. So should we say that gravity causes everything? I think not.

    Yeah, OK, gravity is creating the partial vacuum but by far the most interesting and noteworthy aspect of a siphon is the idea that it is atmospheric pressure pushing the liquid up the short leg. I'm fine with the so-called error in the dictionary.

  24. John Robson Silver badge

    Thought experiment

    Only 'cos I can't go there...

    Two open bowls on the surface of the moon. One (filled with mercury) on a table 1m above the other.

    Coil a hose in the top bowl, put a bung in the end and lift that end out

    - On earth it would stay full of mercury, even when put upright

    - On the moon I'm not so sure. I think it would stay level.

    On earth the height difference is the basis of a barometer / manometer, with zero pressure acting on the main surface of the liquid why would it object to zero pressure inside the tube?

    If you pumped the mercury up the tube and put the blocked end in the lower bowl...

    Then I expect that it would split at the top and each half would fall in the obvious direction. Liquids are not known for their tensile strength, and that is all (I can see) that would prevent this.

    Can we please ask the Myth Busters to visit NASA and have a play in one of their really big vacuum chambers?


    - Atmospheric pressure

    - Fluid

    - Tube diameter (capillary effect?)

    I think that height difference is not relevant to the discussion.

    1. edward wright

      @John Robson

      Sorry, Mercury would either boil or freeze on the Moon, liquids are a no-go without atmospheric pressure.

      So much as I'd like to vote for gravity, I go with atmospheric pressure, for without it you have no liquids, thus no syphon!

    2. Nexox Enigma


      Your hypothetical tube of mercury on the moon, if it didn't freeze, would indeed fall out of the sealed tube, because it would boil off under it's own weight and fill the void above the liquid with gas. This is the same effect that limits a terrestrial siphon to 10m height differential.

      And yes, that limit is derived from the atmospheric pressure, but saying that a condition limits a phenomenon is entirely different from saying that it drives said phenomenon.

      If you look at where the force to accelerate the fluid originates, it's always gravity.

      """I think that height difference is not relevant to the discussion."""

      And centuries of fluid statics would call you a dimwit. Until you hit some limiting condition, height differential is /all/ that matters (in a first order model, anyway - for more accuracy you'd need to look at density, viscosity, siphon geometry, siphon surface roughness, etc. You'd probably want to go full CFD at that point.)

    3. MarkBrown

      no need to go to the moon

      You don't need to go to the moon to test whether atmospheric pressure is responsible for siphoning. Just put a hole at the highest part of the hose and see if the siphon keeps working. It is atmospheric pressure (and a little surface tension, I guess) that stops the fluid in each side of the hose from separating and flowing back into its respective container.

      Bootnote: The fact that the pressure is (roughly) balanced at each end of the hose is not the whole point. The important part is that the pressure is not enough to counter gravity on the long side, so the fluid flows down, whereas it is more than enough to counter the gravity on the short side (which weighs less) so the fluid flows up.

  25. Muckminded

    Related stories

    I'm curious why a link titled "California school pulls 'oral sex' dictionary" is in the "Related stories" section at the butt end of the article. If oral sex is considered siphoning, then it throws the gravity theory out the window.

    1. Sarah Bee (Written by Reg staff)

      Re: Related stories


    2. John Dougald McCallum


      cos it refers to a Dictionary?????

  26. Eddy Ito
    Paris Hilton

    hypobaric chamber siphon

    So, does the El Reg PARIS team feel up to a few siphon experiments in the hypobaric chamber? You can use your choice of beverage! Come on, you know you want to.

  27. Robert Carnegie Silver badge

    No atmosphere, no liquid.

    No liquid, no siphon! sea level, immerse the whole of a hose in a tank of water so that the hose itself is filled; stopper one end of the hose; pull that end up out of the tank, drop it into the sea (I mentioned we are at sea level), and unstopper it. Water drains from the tank without anything being sucked or blown. It's gravity.

    Elsewhere, it may not be gasoline:

  28. bitten

    it is a complot

    I checked the internet and see that gravity stays unmentioned when describing drinking. Is gravity not doing all the work? And don't start about straws! Ignorance all over the place, it is time to save the world, where is the government when you need one?

  29. Steve X

    Needs both, surely?

    Try it on the moon (no atmosphere, gravity). Won't work.

    Try it in the ISS (atmosphere, no gravity). Won't work.

    1. edward wright

      @Steve X

      Try it within a centrifuge on the ISS and it'd work :)

      Might be a tad tricky to set up, mind.

      1. Charles 9

        A centrifuge would provide the gravity...

        ...nullifying the presumption that you can't run a siphon without gravity. I strongly suspect it doesn't work without gravity because gravity is a factor in producing atmospheric pressure (as in, the force of air pressure is generated by the acceleration due to weight of the atmosphere on the planet). So, for the siphon effect to work, you need atmospheric pressure, which requires both atmosphere (to provide the substance) and gravity (to allow the substance to exert a force).

  30. Anonymous Coward

    Why was he using the OED

    And not Oxford Dictionary of Physics, which probably has a much more detailed, and correct, discription.

  31. Mage


    They cheated by editing earlier today

    1. Martin 71 Silver badge

      Editing an editable encyclopedia to make it more accurate

      is hardly cheating :-)

      Yes, I know it was a joke. It's been that kind of day. Off for a beer

  32. John Dougald McCallum
    Thumb Up


    Elsewhere, it may not be gasoline:

    I think I know what that url refers to without going there Oh Crap mehahahahah

  33. Anonymous Coward
    Anonymous Coward

    There's no kind of atmosphere?

    Take two cylinders partly filled with fluid, and with pistons of negligible mass that move with minimal friction.

    Connect the fluid reservoirs in the two cylinders with a long tube.

    Fiddle with the apparatus until you remove any air or other gas by bleeding it off through a suitable valve.

    Push one piston so that almost all the fluid is in one cylinder and arrange the apparatus so that the tube goes up and then down (just like a siphon!)

    Raise the full cylinder above level of the other (even more like a siphon!).

    Question 1: will the fluid drain from the high reservoir to the lower? (ie in the manner of a siphon)

    Question 2: does "atmospheric pressure" play any role in any siphon-like action that might occur in this apparatus?

    Question 3: How might the fluid properties affect any siphon-like action?

    1. Nexox Enigma


      """Question 2: does "atmospheric pressure" play any role in any siphon-like action that might occur in this apparatus?"""

      Technically atmospheric pressure would be exerted on the back side of each piston, but as they'd be close together, it would be the same pressure, and would cancel out.

      If you stipulated that the whole contraption was enclosed in a vacuum, then there would be no atmospheric effect, and it would still work just fine.

    2. Rattus Rattus

      If I understand your setup correctly,

      ie, the fluid in the upper cylinder has to go a short way up the tube, above the level of the cylinder, before going down the rest of the tube to the lower cylinder, then no the fluid will not drain in the manner of a siphon. Whatever fluid is in the longer portion of the tube will fall into the lower cylinder, and any fluid in the shorter leg of the tube will fall back into the upper cylinder.

      Atmospheric pressure plays a role in that it's absence means there is nothing to push the fluid up the shorter leg and over the bend.

      1. Anonymous Coward
        Anonymous Coward

        and what, then, ...

        ... will fill the tube once the fluid on both sides has fallen away? Where did it come from and what conditions are require to generate/produce it?

        1. Rattus Rattus


          I thought he was talking about doing his experiment in vacuum?

        2. Rattus Rattus

          Oh, I see

          I see now, he wasn't talking about being in a vacuum. In that case, yes, the fluid will move as for a siphon. The fluid will fall in the longer part of the tube, reducing the pressure inside the tube. The atmosphere will push on the outside of the piston of the upper cylinder, thereby equalising the pressure and pushing the fluid up the shorter leg.

          1. Anonymous Coward
            Anonymous Coward


            ... of course I meant vacuum. Then compare the result with those under range of different atmospheric pressures.

            1. Rattus Rattus

              If you meant in vacuum

              then my earlier comment stands. The fluid won't siphon, it'll just fall out of the ends of the tube, with the gap in the middle of the tube filled with vapour from the fluid.

  34. gdeinsta

    Proven false by Hero of Alexandria in the 1st century AD

    "Liquid is, of course, drawn up the shorter limb of the siphon by the weight of that in the longer downward one:..."

    Hero of Alexandria disproved this in the first century AD. In the book "Pneumatics" he discusses an experiment using a siphon with a large diameter for the short leg and a small diameter for the long leg. Result, a greater weight of water in the short leg. Did the water travel up into the higher container (short leg), thereby generating free energy and saving us all from having to dig coal for a living? No it did not.

    It's not the weight, it's the pressure. Pressure does not act at all like a chain. You can't pull on pressure. And the tension in a chain only pulls in one direction, whereas the pressure in a liquid is the same in all directions - turn the pressure sensor in any direction and it registers the same pressure.

    Fail is for the grade he would get if he was a student in my Ancient Science and Technology course at Carleton University.

    1. Pablo


      I just did an experiment, to put this to the test.

      Materials: A plastic jug of water with two holes in it (not counting the top, wish is capped). Two lengths of tubing which fit through the holes tightly. Each tube extends below the water level.

      Result: By blowing air into either tube, I could make water come out the other. And it continued flowing as long as the exit of the water tube was lower than the bottle. The relative position of the air tube made no visible difference (i.e. it could be lower than the water tube).

      Interpretation: Though pressure is necessary, and there is an interplay of effects, gravity IS the driving force.

    2. The Cube

      Erm..... double entendres away

      The experiment you cite has water moving from the leg with the smaller weight to the leg with the larger weight, this would seem to vote in favour of gravity over pressure not against.

      The issue is in the wording, it is assumed in the basic phrase "Liquid is, of course, drawn up the shorter limb of the siphon by the weight of that in the longer downward one:..." that the siphon tube is of constant cross sectional area.

      The experiment you cite has a tube of varying cross sectional area and is the equivalent of putting a tall thin girl on one side of a see-saw and then putting a short fat munter (of greater overall mass) on the other side. Clearly the fat munter is going down (assuming there is a local gravity or simulated gravity through other acceleration). One would not expect the skinny girl to go down simply because she is taller.

      I leave it as an exercise for the student to determine the nature, position and net effect on the "siphon tube" between the tall pretty girl and the short munter under the relative "suction" applied by each.

      </Double Entendres>

    3. Nexox Enigma


      Not really sure what experiment you're talking about, but you're pretty much just wrong.

      """It's not the weight, it's the pressure."""

      pressure = force * area

      weight = force

      So under the influence of gravity, pressure and weight are linearly related, given a constant cross-section pipe. So your different leg diameters actually create different pressure, which, really, is all that matters in determining which direction water will flow in your hypothetical ancient siphon.

      """Pressure does not act at all like a chain. You can't pull on pressure."""

      Actually... Yes you can. I mean, that statement is really, extra wrong. You're clearly a history major, and not an engineer. You can 'pull' on a fluid in a rigid container until you lower it's pressure enough to cause it to boil. All sorts of hydraulic systems depend on this phenomenon.

      """turn the pressure sensor in any direction and it registers the same pressure."""

      For a static fluid, sure. If you measure pressure in a moving fluid, it does indeed depend on the direction (and the geometry of your measurement device.) That's still not really relevant to your chain theory, since in a chain you'd measure force, and in a liquid you'd measure pressure. In order to determine force in a liquid, you'd have to measure the pressure at multiple points, not in multiple orientations. And you can bet that for any static fluid pressure will vary linearly with height.

      """Fail is for the grade he would get if he was a student in my Ancient Science and Technology course at Carleton University."""

      And fail is the grade you'd get in any sort of fluid statics class.

    4. fred #257

      Right but irrelevant

      Respectfully, I think you're hairsplitting. Most commentards here seem to be thinking of a siphon as a hosepipe, in which case the diameters of both legs are the same and hence pressure is proportional to weight and the quote you object to is actually correct. It's certainly the weight of liquid in the longer leg (all right, manifested as a reduction of pressure at the top of the siphon, if you insist) that does the work. But in the argument of gravity-vs-atmospheric pressure as a power source, your point is kinda irrelevant.

      And it IS the 'weight' (i.e. gravitational force on the liquid) that 'powers' the siphon. You need external pressure (e.g. atmospheric) to keep it operating, but since that's equal at both ends of the tube then obviously it can't be providing the energy. In the case of 'suck-it-to-start' it's atmospheric pressure starts forcing the liquid up the tube, but even then the motive power is being supplied by the sucker, sucking. (And strictly speaking at that point you don't yet have a siphon operating).

      The maximum height of a siphon is determined by (external pressure minus vapour pressure of the liquid)**. Which is why the height gets less as the siphon liquid gets hotter (you can't siphon boiling water), just as it gets less as atmospheric pressure decreases. The Victorians had all this sort of thing worked out to a high degree of precision, since they used steam power.

      ** And by the density of the liquid, before someone goes all pedantic on me.

      (Beer, cos it's an interesting subject for experiments...)

    5. Ken Hagan Gold badge

      Re: Proven

      I can't believe that an experimental test gets down-voted.

      I can believe that someone in the 1st century was smart enough to realise there is a question worth asking here and smart enough to figure out a way of answering it. 'Tis a pity that the median intelligence of the human race hasn't changed much since then.

      Thanks to gdeinsta of Carleton and two hero points to Hero of Alexandria

  35. The Grump

    The trick is...

    to start the syphon without knocking over the pint glass on the bar. The length of the longer tube determines how quickly you can order another round. Science is fun - in a pub.

  36. PassingStrange

    He's wrong, pure and simple.

    "Liquid is, of course, drawn up the shorter limb of the siphon by the weight of that in the longer downward one: thus the operating force is gravity."

    That's an even bigger howler than the stock explanation, and the sort of thing you get if you only look at the maths, without thinking about the physics. Liquid isn't "drawn up" (as can be proved, as lots of people have pointed out, by raising the shorter limb above the column height that can be supported by atmospheric pressure - the flow stops, and a vacuum forms above the two limbs). For any flow to occur, liquid has to be *pushed* to the top of the shorter limb by atmospheric pressure. No pressure, no column of liquid; no column, no flow.

    So. Air pressure is pushing the liquid up in both limbs; the weight of the column of liquid in each limb is opposing that pressure. The column of liquid in the longer limb is greater than that in the shorter limb, so its weight cancels out more of the force due to air pressure at that limb, resulting in a net force (and resultant flow) from short limb to long.

    Atmospheric pressure at both ends is the same, so the *size* of the resultant force is, indeed, directly down to the difference in weight of the two columns of water. But. Without atmospheric pressure at the shorter limb, no column of water - and hence no siphon effect - is possible. Gravity does not *cause* the flow - it works on both sides to *oppose* it (and fails more badly at the shorter limb). The operating *force* is the unbalanced component of atmospheric pressure at the short limb.

  37. John Savard

    If Push and Not Pull

    As noted, "you can't pull on pressure". Hence, the weight of the fluid in the longer arm of the siphon does not pull the fluid up in the shorter arm. What, then, does cause the fluid to ascend in the shorter arm?

    It is air pressure after all; while the air pressure is higher at the lower level, the weight of the fluid in the longer arm more than counterbalances the difference in air pressures. So we have:

    push of air pressure at the top overpowers push of air pressure at the bottom minus the weight of the extra fluid in the longer arm.

    So instead of "air pressure" being wrong and "weight of fluid" being right, the right answer is both of them.

    1. Charles 9

      But what about the Hero experiment... which the short leg is much larger and thus actually heavier than the longer but thinner long leg? The siphon still works normally usually, but the restriction could cause the siphon to stop if the difference is too great and the flow restriction causes the liquid to cavitate ("boil") in the siphon. From what I've read, gravity and fluid density are the only things related to hydrostatic pressure. Still trying to see why siphons typically don't work in a vacuum in spite of this fact.

  38. Steven Jones

    The answer is...

    Whether atmospheric pressure is required or not depends on if this is an inverted siphon (like a trap on a sink) or a normal siphon. An inverted siphon doesn't require any atmospheric pressure. Of course water would boil in a vacuum, but and inverted siphon of mercury would work quite happily in such conditions.

    In the case of a normal siphon, there is a height limit (of about 33 feet for water) as it is necessary for the lower leg of the siphon to "suck" the water over the high point. Of course there is no such thing as "suck" - it's pushed by atmospheric pressure when a partial vacuum is at the upper end of the tube. However, once the column height reaches about 33 feet, that's it - normal atmospheric pressure won't push it any further, even with a perfect vacuum at the other end (and it can never be quite perfect due to the vapour pressure of the evaporating liquid).

    Of course, ultimately, it is the pressure difference between the upper and lower legs caused by the action of gravity on the water that provides the force to move the water. However, it simply would not work without enough atmospheric pressure to push the water over the "hump".

    Now this is not to say that it might not be possible to get a siphon flowing with a "hump" of more than 33 feet. Perhaps if it could be started by some other means (such as excess pressure on the higher side) it might be started and continue under momentum once flow started and the pressure was returned to atmospheric pressure, but I doubt it would give you much extra height and I'm feeling too tired to work out the physics of this at the moment.

  39. RodTheJedi


    Gravity is a myth.

    The Earth sucks...

  40. Valerion

    Based on today's events

    It appears they have also got the definition of "Democracy" wrong as well.

  41. Dave Taflin

    April Fools??

    Was the good doctor's report written April 1st, by any chance? Or maybe every day is April Fools' Day in OZ.

    1. Rattus Rattus

      re: every day is April Fools' Day in OZ

      No, not every day. Just the days that parliament sits.

  42. Jolyon Smith

    As ever, isn't the truth somewhere in the middle

    Gravity acts on the liquid in the long leg creating a force that acts in *conjunction* with the atmospheric pressure on the liquid in the vessel into which the short leg is immersed. This surely creates the imbalance in forces required to cause the liquid to flow. Gravity isn't pulling the water on it's own... gravity creates the partial vacuum which atmospheric pressure then fills by pushing liquid to fill that vacuum.

    This experiment should be recreated with beer. In a pub. All good theoretical science is done in pubs with beer.

  43. Vladimir Plouzhnikov

    C'mon people

    It's hydrostatic pressure (gravity related, of course, but nevertheless pressure) that explains the communicating vessels.

    in each vessel p = \rho g h,

    if h1 > h2 then p1 > p2 and the water flows from vessel 1 to vessel 2

    The atmospheric pressure does play a role but because the \rho of air is much lower than the \rho of water, the extra pressure differential created by the air will always be less than the differential between the water columns (the height of both water columns is measured against a reference point below the surface of the lowest column).

    In other words, it's the difference in potential energy that forces the water flow (just like the difference in potentials causes the electrical current flow).

  44. Antony King


    The trouble with using beer in a siphon is that after a while, all the bubbles would get trapped at the top and stop the process. Anyway, have you /tried/ drinking beer through a straw?

  45. The Evil One
    Gates Horns

    OED not entirely wrong

    It is atmospheric pressure that determines the height of the barrier over which the syphon can move the fluid. A syphon would not work in a vacuum. OED explanation is incomplete rather than completely wrong.

    1. Vladimir Plouzhnikov

      Water strength

      I think some (micro)siphons can be made to work in a vacuum due to the tensile strength of water.

  46. Robert Carnegie Silver badge, I'll get this...

    Yes, there's no such thing as suction really, just pressure without equal pressure on the other side. And yes, air pressure - in the typical example - pushes water up the siphon tube. But it does so only because with unequal amounts of water in the two arms of an inverted and asymmetric U - more like a J - gravity pulls unequally, more strongly, on water in the longer arm, and cancels the pressure on that side.

  47. Simon R. Bone

    90+ comments?!

    for a story about siphoning??

    Slow day at work guys?

  48. JC 2

    Not Atmospheric Pressure Needed

    By writing "atmospheric pressure" you would have to imply it is necessarily pressure from the atmosphere, which is obviously not needed! Instead, it only needs to be a sufficient pressure differential to overcome gravity, regardless of the cause of that pressure difference. For example, a vacuum container which within holds a sealed system with two cannisters at dissimilar pressure and a liquid in a tube between them.

    There is such thing as suction, writing out a definition as a way to explain it away is ignoring the whole POINT of words in a language, to have a term that eliminates the need to constantly write out a much longer definition using other words.

  49. Martin 77

    Water: a special case?

    Has anyone here actually read the article Dr Hughes wrote?

    He talks about water molecules behaving as chains - something he demonstrates with real chains.

    As he concentrates on just water molecules I do wonder if the same applies to other liquids with weaker bonds (or very heavy molecules, such as mercury). I can't help thinking that whilst gravity causes the liquid to flow downwards it is atmospheric pressure that stops the liquid draining from either side into the container beneath it and leaving a vacuum above.

  50. Dave92


    Popup got it exactly right on page 1.

    @Number 6: yes gravity is needed, but that's not the salient point. A drain needs gravity to work, so if a drain is working, gravity is present. If you take a working drain tube and lift the middle, at the point when the middle exceeds the surface level the drain becomes a siphon. Without atmospheric (or pump) pressure, water will stop flowing when the drain becomes a siphon. With (and only with) pressure sufficient to push the water up to the hump, water will continue to flow.

    To claim that the OED has a 99 year old mistake, and that atmospheric pressure is not the factor that causes a siphon (as opposed to a drain) to function is ridiculous, but it certainly got this Oz goofball his 15 minutes of fame.

  51. Aus_Eng

    OED is right

    It's not due to gravity, a siphon works INSPITE of gravity, if the gravity is to high a siphon will not work, but as long as there is air pressure (atmospheric on earth), a siphon will work. Gravity or not.

    If fact, a siphon will work perfectly well in zero gravity, the h or maximum height of the 'hump' would be infinate.

    Air pressure pushes on the surface of the liquid, that pressure if great enough will push the liquiq uphill, against gravity.

    On earth the air pressure is created by gravity, but the air pressure does not have to be supplied by gravity, it just is on earth.

    On the space station, where there is air pressure but zero gravity, a siphon will work just fine, again the h value would be infinate, if the available air volume, and liquid volume were also infinate.

    A the fluid moves out of the container the air pressure on the surface of the fluid will lower, on earth you would not see this effect because of the large amount of air available, and the flow from source to destination and destination to source for the air.

    But a siphon would work in zero gravity, if the air pressure was supplied by cylinders or compressed gas for example, or just the normal atmospheric pressure within the space station.

    The human circulation system relies on a mixture of pumping and siphoning effects to circulate blood through the body, but when astronaughts to go space, they dont die. The siphon effect works in zero gravity, and blood keeps circulating.

    Cooling system no the space station would most certainly rely upon siphons to move coolant.

    There is no 'vacuum', a vacuum is an absence of stuff, not a 'thing' in itself, if there is air in the siphon pipe it will not work because gas is compressible, and liquid for most part is not. thats why you have to fill the tube of a siphon will fluid before it will work.

    So this 'so called physist' is completely wrong and OED is right, it's odd that someone that calls themselves a physist would not simply revert to math, and look at the equation for a siphon,

    Bernoulli's Equation, and look at the terms that it uses to calculate a siphon, gravity is not a factor, density is, P is pressure, h is height

This topic is closed for new posts.