If the specific height of a gel is an issue, I have commonly used a layer of water-saturated butanol on top of the gel while it is polymerizing. It just has to be added with a bit of care then rinsed off thoroughly before going to the next step. When I run a protein gel that uses a stacking gel, I often use the butanol layer.
Roopali Rajput is absolutely right. If you want to prevent the aqueous phase formation you can overlay your unpolymerized gel mixture with a little bit of iso-butanol of iso-propanol to prevent oxygen access. This way the upper surface of your resolving gel will be smooth polymerization, and when you remove the organic phase and dry it, will be ready to pour the staking gel on top.
Roopali is right, but you can observe also another aspect of gel formation - when polymerization was occured we can observe some aqueous fraction becase of decreasing volume of polymer. Th aviod unpoymerized layer formation try to isolate top layer of the gel by adding mix of butyl alcohol with water on it.
This video is similar to the method I have always used. http://youtu.be/EDi_n_0NiF4
It's always helpful for me to see how something is carried out instead of just reading about it.
I'm not sure what you mean by an aqueous *layer* - and it may be because of the addition of alcohol - however, if you are asking why the gel is always slick and wet-looking, it is because the gel itself contains quite a bit of moisture.
I agree with Brian...I also use isoprapanol instead. In addition to preventing oxidation it also gives you a perfectly flat interface between stacking gell and resolving gel...which is also important in preventing differential migration just by virtue of the uneven surface (not nature of sample)