prof. from Colorado is wrong
> Air "weighs" about 14.7 pounds per square inch of area on which it rests, including the surface of a liquid; this pressurizes the liquid to this amount.
Air pressure acting on the liquid in the reservoirs on *both sides* is equal. Actually, it's slightly higher on the lower one. You could even submerge both ends of the tube in the liquid and it would still work, though you will get reduced flow since the outflow tube will now be fighting against water pressure in the lower reservoir.
The original pedant/boffin had the right idea in trying to explain water as a chain. They're maybe not his exact words, but it makes sense if you think about it like this... A previous poster here (Mike Bell) suggests an experiment where you have a physical chain dangling out of a drawer. Once you have enough of the chain dangling over the edge, it will pull the rest of the chain with it. Water is pretty much exactly like a chain, except that at lower temperatures it loses its chain-like nature and becomes a gas, so the individual molecules don't form chains. Once the pressure drops again, and the water condenses, at re-attaches to neighbouring molecules and you've got your chain links back again. It's slightly more complicated than that because there's an energy gap that needs to be crossed in changing state from liquid to gas
The prof from Colorado is completely wrong if he thinks (and it would seem that he does) that atmospheric pressure is the operative force. See my example at the start to disprove that (ie, air pressure at the lower reservoir is *greater*, and the siphon works when both ends are submerged). Also, he completely discounts the much greater *water* pressure which operates on the higher reservoir.
I've rambled on a bit more than I wanted. In summary, though, this is simply a case of hydrostatic equilibrium. You have to take into account all the forces:
* gravity affects atmospheric pressure on both reservoirs
* it also affects water pressure at the inlet and outlet (the weight of water above the tube openings)
* it also affects the water in each leg of the tube
* in all cases where gravity is in effect, you have to add the weight of all the air/water above; in each reservoir the "weight" of the water above it is simply proportional to the depth (ignoring any compression), while within the tube the higher up you go, the less pressure there is
* if and only if the pressure at the top of the tube is not so low that a vapour lock forms, then gravity will naturally continue to draw fluid from the higher to the lower reservoir (aka, "water finds its level").
I'll be heretical here and say that is follows from this that a siphon *can* actually work in a vacuum, provided there is gravity. It would need to be set up so that there was some liquid in both reservoirs to begin with. Depending on the rate of boil-off from the surface of the reservoirs, you would still get a siphoning effect, which would work until there wasn't enough *water pressure* (everyone seems to have completely forgotten about water pressure: it's much more important than atmospheric pressure!) in the upper reservoir to prevent an air lock from forming at the top of the tube, or there is no water pressure at all at the tube's ingress or egress (ie, the tube is just at the water level, which allows all the water on that side to flow down back into the reservoir). You could easily verify this by setting up weighing scales on both the reservoirs, and running the experiment once without the siphon being opened, the other with it open. If the weight of water in the lower reservoir increases or simply decreases at a slower rate (balanced by an increased rate of loss at the higher reservoir) then the experiment would prove that you *can* operate a siphon in a vacuum.
I guess people don't study applied maths in schools these days.