Backwards airflow in a Venturi tube When air is flowing through a Venturi tube, it's speed increases at the narrow section. If the airflow is arranged by moving the Venturi tube through calm air, this means air will momentarily flow backwards (against the motion of the tube).
Obviously this backward acceleration is caused by pressure differences, but I'm having trouble coming up with a layman's answer for where the force causing this comes from.
So, where does the backward force acting on air in a forward moving venturi tube come from?
Here's a picture describing the situation http://i.imgur.com/tiIjxZq.png
On both cases the situation should be identical, only difference is the chosen (inertial) reference frame.
 A: The Venturi effect is a phenomenon internal to a Venturi tube and is unaffected by the surrounding air. The effect is only meaningful with respect to the frame of reference of the Venturi tube itself.
Let's say that, instead of the Venturi tube being set on a stationary lab table, that it is instead strapped to the wing of an airplane flying eastward at 200km/h. You are also strapped to the wing in order to take measurements on the system. You compare the velocity of the air in the bell to the velocity of the air in the neck and find that the latter is in fact higher, with respect to the Venturi. However, taking an actual velocity measurement, you would find that neither the bell nor neck wind speed would be 200km/h with respect to the tube; and if you placed your hand behind the apparatus, you would not feel a jet of air blowing even faster than the wind that you feel due to being strapped to the wing of an airplane. The Venturi 'scoops up' the air and drags it along.
In other words, while it is technically true to say that the air traveling through the neck of an eastward-traveling Venturi has a higher westward velocity than the air in the bell, that is only because the former westward velocity is less negative than the latter; both streams of air are still moving eastward with respect to the surrounding atmosphere.
A: It's an interesting question, but I think you're making some incorrect assumptions.  The classical Venturi effect is the increase in velocity (and decrease in pressure) in an air stream moving through a constriction in a closed pipe.  You can prove that the velocity increases in direct proportion to the change in cross-section of the pipe.  You can also prove that the pressure decreases by noting that, for a particle to accelerate, it must have more force on one side than the other.  Since force in a gas equals pressure, the pressure must decrease as the particles accelerate.
But this is only true for a closed system with constant energy in the flow.  If you put a Venturi out in the open air as shown in your drawing, the back pressure from the constriction will reduce the amount of gas flowing into the pipe.  So the gas flowing into the entrance of your Venturi will be slower than the free air stream.  It's true that whatever gas does make it into the front aperture will subsequently speed up in proportion to the change in cross section.  But the physics conspires such that you will never get a flow in the negative direction (with respect to the outside air) as you suggest--sorry.
I'm editing my answer based on your follow-up question, with the following image:

Note that the air on the top of the wing does indeed go faster than the air on the bottom.  It also appears that the air about half a wing-cord above the wing is actually moving faster (to the right) than the free stream. So your premise is correct for an airfoil, which casts reasonable doubt on my Venturi answer.  I still think the backpressure in a Venturi would prevent backwards flow, but there may be some configuration that gets just enough increase in velocity to allow it.  I'm not sure it's the sort of thing you can prove mathematically.  It will take an experiment, or a good finite element simulation program like the one that generated this image.
