The following answer was selected and edited by New Scientist staffThe questioner here is correct in thinking this was just an illusion.Firstly, 1 litre of air at a pressure of 1 atmosphere weighs approximately 1.25 grams. So assuming the person was using a 500-millilitre bottle, the empty bottle on the surface would contain roughly 0.625 grams of air.Boyle's law states that gas pressure multiplied by volume is constant. At 30 metres' depth in salt water you are under about 4 atmospheres. Filling the bottle at this depth would take four times as much air as it would at the surface. On bringing it to the surface, assuming it didn't explode or leak from the increased pressure - which is likely - you would now have a bottle pressurised to 4 atmospheres containing half a litre of gas.This simply means that the original weight of the gas at surface pressure is now multiplied by four, giving a figure of 2.5 grams of air in the bottle. The bottle's mass itself remains constant - the only change is the weight of gas it contains. Therefore, the bottle would actually be 1.875 grams heavier on the surface than one filled at normal atmospheric pressure. The bigger the bottle the more gas it holds, so the more noticeable the effect would be. For example a 1.5-litre bottle filled at the surface would contain 1.875 grams of gas, while filling it at 30 metres' depth would give 7.5 grams and so on.For another real-world example look no further than the cylinder the divers were wearing at the time. These are constant-volume pressurised tanks. Typically these hold 2700 or so litres of air at approximately 230 atmospheres pressure. If containers really did get lighter when pressurised, then these tanks would weigh next to nothing when full. The reality is they weigh approximately 3.4 kilograms more than when empty.As for how people are taken in by the suggestion, try holding two objects that differ in weight by only 2 grams and see if you can tell the difference. You can't. Sometimes one just believes what one is told.Richard Whitcombe, Ebbw Vale, Gwent, UK

The following answer was selected and edited by New Scientist staffOne possible explanation is that the instructor deceived everyone by using helium, not oxygen, to fill the bottle. Helium has a molecular weight of 4, compared with an average of about 28.8 for air - which is a mixture of about 79 per cent nitrogen and 21 per cent oxygen. Therefore up to a pressure of about 7 atmospheres (a water depth of about 60 metres), a volume of helium weighs less than the same volume of air at atmospheric pressure. This could explain why the bottle seemed lighter.Simon Iveson, Mayfield East, New South Wales, Australia

The following answer was selected and edited by New Scientist staffWe are conditioned from birth to make certain assumptions about the world around us and I think that this is a case of toying with a preconceived perception.In the interests of research, and being a touch bored, I fitted a car tyre valve into the cap of an empty 2-litre fizzy drink bottle. I hasten to add that nobody else should try this.After many destruction tests, and explanations to my now hard-of-hearing neighbours, I pressurised a bottle to a "safe" 120 pounds per square inch, the equivalent of filling a bottle with air at a depth of 74 metres before surfacing. The bottle was as tight as a drum and did feel light in comparison with an identical control. After carefully releasing the pressure the apparent weight returned.I suspect that if an object is rigid and unyielding we expect it to be heavy, and so falsely overestimate its lightness when it isn't. My tests also showed that if you tried this at a depth of 100 metres, the bottle would explode at or close to the surface, with dangerous and possibly fatal results. So, as I said, don't try any of this.Anthony Higham, Edenbridge, Kent, UKWe can only add to Anthony Higham's warnings. Do not attempt any of his experiments yourself, and also read the post below from Michael Clark - Ed

The following answer was selected and edited by New Scientist staffThis question raises important points. First, nitrogen narcosis sets in for some individuals at a depth of 30 metres, and an inexperienced diver breathing air at this depth would experience an altered perception of reality. In any case, I do not know of any method of extracting air from an oxygen tank at a depth of 30 metres.Also, the possibility of confusing air and oxygen tanks here is extremely dangerous. Any attempt to use pure oxygen as a breathing gas at a depth of 30 metres would involve breathing oxygen at a partial pressure of 4 atmospheres. The accepted safe limit is 1.4 atmospheres. Any partial pressure over 2 atmospheres would inevitably lead to convulsions, which have a tendency to shorten one's life expectancy if experienced underwater.Neither of the world's leading diver training agencies, the Professional Association of Diving Instructors and the British Sub Aqua Club, approve of inexperienced divers descending to a depth of 30 metres.Michael Clark, Newark, Nottinghamshire, UK

Bottle's volume is bigger after immersion cause it had been increased during the ascension process. I mean after the immersion the bottle contains bigger air mass amount. That implies the bottle is heavier after the immersion than before.