There are some videos going around that claim to show that recent snowfalls are somehow "fake", because when people try to melt the snow with a lighter or blowtorch, they are surprised by the results. Typical story with links to several videos: http://www.adguk-blog.com/2014/01/fake-snow-reported-in-multiple-states.html?m=1 What's happening here is: A) The snow is melting, but the very loose fluffy structure of the snow wicks away the water, turning dry snow into wet snow, and eventually turning the snow into slush. B) The snow is blackened when a lighter is held underneath it because of the soot from the lighter (the products of incomplete combustion). It's not burning. C) The smell is fumes from the lighter (also from incomplete combustion) and/or people briefly burning nearby objects like gloves. D) It is not sublimation. At these temperatures and pressures, and over this timeframe, any sublimation is not going to be measurable. See end of this post for proof. Heres a typical setup. Snow is more air than water. And the colder it is, the more air (and less water) it has in it: http://www.nrcs.usda.gov/wps/portal/nrcs/detail/or/snow/?cid=nrcs142p2_046155 He then gets a blowtorch out, which just speeds up the process. With the blowtorch there's also going to be more evaporation of the water, which will help things keep dry. However you can see there's also a lot of liquid water here (in the form of slush). I don't think the evaporation is a big factor. Unfortunately he stops right there. All he has done is partially melted the snow, the remaining snow has soaked up the water created, and you get slush. If he continued, then you'd get a little puddle of water. Notice also that the snow does not get black. The blowtorch is cleaner burning and far away from the snow. This proves the black coloration was coming from the lighter when held close. He then compares this to an ice cube: The problem with this comparison is that snow and ice are radically different. Ice is solid water. Snow is mostly air, a fluffy mass of lots of tiny snowflakes, which are ideally shaped to absorb water. So when bits of the snow melt, it just gets absorbed into the surrounding snow, making slush. When the ice cube melts, the water has nowhere to go, so you get a puddle. So what these videos are demonstrating is little more than the fact that using a lighter is not a very good way of melting snow. But it does melt. If you just left it there you'd get a puddle. If you kept melting it with the lighter it would eventually stop transitioning to slush, and start pooling water. Same thing with the blowtorch, just quicker. And you could really speed up the process by putting it in a microwave. The next video, titled " Burning snow!!!!!! FAKE POISON SNOW! Make viral!!!!! " again shows normal snow, and the experiment is repeated. Again we see the snow melting normally where small bits of it fall on the surface, and you see the snow melting and being wicked away when a flame is applied to the larger mass: It's getting black because the lighter in this case is being held under the snow, the snow gets a coating of soot. If you put a lighter under anything, you'll get soot on that thing. Like this glass slide and piece of ice. (The soot comes from incomplete combustion when you hold the flame too close to an object and restrict the oxygen supply. If you lower the flame half an inch, then there is very little blackening.) And a more detailed (35 second) video explanation I made: Unfortunately, the maker of the video interprets this as the snow "burning". Clearly it's not though. The snow does not catch fire, it just get a little sooty. Here's another example: This debunking video by Joshua Barry demonstrates the same effects with a butane lighter, but when held over a large source of heat (the stove), the snow much more rapidly turns slushy along the bottom edge, and then starts to drip. And in this second video he demonstrates in more detail what is going on: Joshua notes that the smell comes when the blackening is happening, but he does not understand why. This is because of incomplete combustion. When butane (in the lighter) burns with a good supply of oxygen then it combusts fairly cleanly into carbon dioxide and water. But if there's a lack of oxygen flow (like it's too close to some object), then there's much more leftover carbon (in the form of the soot) and toxic carbon monoxide. What you are smelling is basically the carbon, and the unburnt butane gas. In this video I demonstrate clean (full combustion) and dirty (partial combustion) burning of butane. There's also been some discussion of sublimation (where ice transitions directly from solid to gas). This has been suggested as an explanation as to why the snow seems to not produce liquid. But it's a false and needless explanation. Sublimation happens mostly at low pressure. If there is any water loss here it's because the ice crystals rapidly melt, and then just as rapidly evaporate. I think though that the majority of the snow turns to liquid. This could be verified by weighing it on a sensitive scale as it is melted with a flame. [UPDATE] Someone has done this experiment, and demonstrated zero detectable sublimation or evaporation. It's just melting. See video: So what's going on with these bogus "fake snow" videos? One possibility is that it just a hoax - someone might have made it as a joke, the other people followed suit. But it's quite possible that the makers of these videos actually believe what they are saying. It's possible that the recent cold weather has brought snow to places that do not normally have it, and these people are simply are unfamiliar with the properties of snow, and how it melts. Most people will find this story rather silly. And yet I just spent half an hour writing the above explanation (and more time now expanding and editing it). Why waste time over a silly story? I do it because people get genuinely concerned about such things, and it's good to explain it for them, so they can stop worrying. I also do it because the more bunk is out there, the less time people will have for real issues. I'm improving the signal to noise ratio.