Well, technically, doing experiments to gather data or distinguish between theories is science. This "experiment" isn't an experiment, it's a demonstration. A

Well, technically, doing experiments to gather data or distinguish between theories is science. This "experiment" isn't an experiment, it's a demonstration. And it is usually misinterpreted. If the interpretation is right, where does the oxygen go? In truth, the oxygen is replaced by CO2. Pretty much every oxygen molecule that's burned (i.e. combined with carbon) becomes one molecule of CO2. Therefore the amount of gas should remain unchanged. There is another effect at work. When you put the bowl over the top the air is warm, and being warmed, but the flame. Some of the air trapped is pushed out - if you do this you can see bubbles around the rim - and then the remaining air cools. It's that cooling that sucks in the water. The demonstration is, indeed, really poor.