For example, take a water bottle. Fill it with water and then turn it upside down. Instead of flowing steadily downward, it gulps down in parts. Why?

  • 4
    $\begingroup$ if you poke a big enough hole in the bottom of the bottle, it won't happen anymore. $\endgroup$
    – endolith
    Commented Mar 16, 2014 at 19:29

4 Answers 4


The gulping you describe is due to air being sucked into the bottle and temporarily halting the flow through the nozzle. When the bottle is filled with water, it is at a particular pressure. When you turn it over and some water leaves, the pressure is now lower in the bottle.

Once the pressure in the bottle is lower than atmospheric pressure, air forces it's way back into the bottle. This equalizes the pressure and water flows again. Then the pressure drops, air gets sucked in, and so on. Eventually all the water is gone and the bottle is filled with air at the same pressure as the atmosphere.

  • 3
    $\begingroup$ Note that you can both demonstrate this and overcome it, in many cases, by swinging the bottle so its contents swirl. If you can get them spinning fast enough, momentum will pull them against the outside edge of the bottle and the bottle neck, opening an air channel up the center, and this "tornado" will break the vacuum and permit the liquid to flow out quickly. ....Or you can just punch a hole in the bottom of the bottle to admit air, and see that this restores rapid flow. Note that this is why we normally punch two holes, not one, before pouring liquid from a can. $\endgroup$
    – keshlam
    Commented Mar 16, 2014 at 19:53
  • 1
    $\begingroup$ This is a good answer, but it feels incomplete to me. It doesn't explain why the same thing doesn't happen in a larger opening even when you try to pour pretty fast. I suspect there is some kind of hydraulic-jump like effect when the contents try to flow to too rapidly through the opening that prevents the establishment of a steady-state. However, I'm in over my head on this. $\endgroup$ Commented Mar 16, 2014 at 21:44
  • 1
    $\begingroup$ @tpg2114 Alienating answers require alien language, i.e. jargon. There are complicated ways to answer anything. But ELIF (ExplainLikeImFive) answers take the talent of simplicity & straightforwardness. $\endgroup$
    – Daz C
    Commented Mar 17, 2014 at 0:32
  • 3
    $\begingroup$ @Dazc - perhaps ExplainLikeImFive should be a valid tag in physics stackexchange? I'd believe that there are many valid, useful questions here that need exactly that. $\endgroup$
    – Peteris
    Commented Mar 17, 2014 at 1:32
  • 1
    $\begingroup$ +1 This is also the reason why they put a second smaller hole in coffee cup lids (and why that increases the flow). Also, did you ever drink water from one of those 2.5 gallon plastic containers with the spout on them (e.g. when camping)? Over time the flow gets cut off because the air pressure in the container is too low; the only way to relieve it is to turn the container upside down to let new air in, or to poke a hole in the top. Same with the occasional big "bloop" in water cooler tanks. $\endgroup$
    – Jason C
    Commented Mar 17, 2014 at 4:24

As water flows out of the bottle, air will flow in to replace it. If you just flip the bottle straight upside down, then because the bottleneck is so narrow, then at any given time, sometimes water will block the flow of air into the bottle, or air will block the flow of water out of the bottle. Essentially they "take turns" going in or out, which creates the "gulping" effect.

You can prevent this, if you can reserve some of the space in the bottleneck for water and some of it for air. One way to do this is to tip the bottle more slowly, so that air flows in the top of the bottleneck while water flows out the bottom. If you must hold the bottle completely upside down, you can twirl it to create a whirlpool effect (also called a vortex): in this case, air flows into the center of the bottleneck while water flows out the sides. Either way will allow air to flow in at the same time water flows out, so that everything goes more smoothly.


There is an important point that the previous answers have missed. And this phenomenon is not widely known -- although water flowing out of an overturned bottle is!

The glug-glug and the air bubbles rising up in the water are due to a fluid dynamical instability called the Rayleigh-Taylor instability (wiki). This instability occurs when a heavy fluid (water in the bottle, in this case) is suspended above a light fluid (the air outside the bottle). The heavy-over-light configuration is inherently unstable, and it only takes a small perturbation to upset the balance.

In an overturned bottle, minuscule capillary waves at the surface (due to vibrations for e.g.) are amplified into larger amplitude waves until the point when the air in the bottom becomes a bubble, is cut off by the water, and rises inside the bottle as a "glug".

Limiting cases of this instability are also responsible for the poor performance of fusion reactions, for the necklace shapes observed in supernova explosions, patterns found in volcanic eruptions, the mushroom shape in nuclear explosions, and the mammatus clouds formed on earth!

If you are interested in reading more about the science behind this, read about it in the Journal of Fluid Mechanics here.


  1. There is often enough air pressure to support a column of water inside a bottle. So water does not flow out of an overturned bottle merely because of the air pressure!
  2. The real reason why water flows out of an overturned bottle is the presence of an instability called the Rayleigh Taylor instability that amplifies the capillary waves and causes the air to bubble through the water.
  • $\begingroup$ Excellent answer. Welcome here! $\endgroup$
    – peterh
    Commented Sep 1, 2016 at 22:04

The "gulping" is due to the sucked-in air forming a bubble on its way up and by this briefly blocking the bottleneck. The shape of the bottle is important here. You wont see any "gulping" using a top sealed tube.

  • $\begingroup$ "sucked-in air"? This is a notion that unsettles me on a physics site. $\endgroup$
    – Beska
    Commented Mar 17, 2014 at 14:52
  • $\begingroup$ @Beska True, out-flow of water and inflow of air go hand-in-hand. No sucking and no pushing involved. Was just replicating the wording from above without thinking. $\endgroup$ Commented Mar 19, 2014 at 12:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.