You're previewing the Bennett demo family. Numbers, schedule, and kids are sample data.Start your own
Loading page…
Every snow crystal that has ever fallen had six sides. Not five, not seven, not once. Something is enforcing that — and it is far too small to see.
Catch snow on a dark sleeve and look closely — properly closely, with the sun behind you.
Some are flat little plates. Some are stubby columns like tiny barrels. Some are the enormous feathery stars people draw on cards.
They look nothing like each other. And every single one of them has six of whatever it has. Six sides, six arms, six corners.
Nobody is checking. So what makes it come out six, every time, in every snowstorm on Earth?
A drop of water is made of tiny pieces called molecules. One piece of water is one oxygen with two hydrogens stuck to it — the famous H₂O.
The important bit is that those two hydrogens are not opposite each other. They sit off to one side, making a shape a bit like a wide V, or like a head with two ears.
While water is liquid, the molecules slide around each other and the shape doesn't show. Warm water is a crowd, shoving.
Cool it enough and the shoving stops. Each molecule reaches out and locks on to its neighbours — and because of that V shape, the neatest way for them to settle is in rings of six.
Ice is water molecules locked in six-sided rings. A snow crystal is that pattern, repeated so many times you can finally see it.
One ring is far too small for any eye. But a growing crystal doesn't build one ring. It builds rings on rings on rings, each new molecule dropping into the only slot that fits.
And a pattern made of six-sided rings can only ever grow into a six-sided thing. It has no way to produce a fifth corner: there is nowhere for that molecule to sit.
By the time a crystal is big enough to land on your sleeve it holds something like a million million million water molecules, all obeying that same rule. The shape you're looking at is not a picture of a snowflake. It is the shape of one water molecule, printed a million million million times.
Snow falls in Norway, in Japan, and on a mountain in Chile, on three different days. Every crystal in all three places comes out six-sided. Why can we be so sure of that in advance?
Because the six is the only thing the molecule fixes. Everything else is decided on the way down.
A crystal starts near the top of a cloud, usually around a speck of dust, and then falls — sometimes for half an hour, through air that changes as it goes. And the way it grows turns out to depend fiercely on the temperature it is growing in.
A scientist called Ukichiro Nakaya grew snow crystals in a laboratory in the 1930s, one at a time, at temperatures he controlled — and found something nobody expected. The shape doesn't change smoothly as it gets colder. It flips back and forth:
And on top of that, the wetter the air, the more wildly the arms branch.
So a crystal's shape is a record of its journey. Fell through a -15°C layer for a while? It grew arms. Dropped into drier air? The arms stopped and it thickened up. Wandered back into -5°C? A needle started somewhere on it.
Which raises the question everyone asks next: if the journey is so fiddly, why do the six arms on any one crystal match each other so beautifully?
Because the whole crystal is smaller than a grain of rice. All six arms are in the same place, at the same moment, in the same air. They get the same instructions, so they do the same thing. It isn't the arms copying each other — it's six arms having exactly the same weather.
Everything about a snow crystal comes from one of two places. Sort each one, and read the reason — a couple of these are not where they look.
Tap an item, then tap where it belongs
You have definitely been told this. It's worth doing properly, because the true version is better than the slogan.
First, the awkward fact. In 1988, a researcher called Nancy Knight was studying crystals collected from a storm over Wisconsin, and found two that looked identical — a matching pair of hollow columns, photographed side by side. Simple crystals can and do match.
So the flat claim isn't quite right. Here's what is.
A big branching crystal has something like a million million million molecules in it, and the number of ways to arrange all the tiny details of its arms is so enormous that no meaningful number describes it. Two of those matching exactly, all the way down, has never been seen and is not going to be.
The honest sentence is: simple crystals can match; complicated ones never do. And that is a more interesting thing to know than the slogan, because it tells you why.
Somebody says "no two snowflakes are alike — it's a scientific fact." What would a careful answer be?
You’re previewing as a parent — nothing here is recorded.
A water molecule is one oxygen bonded to two hydrogens with a bond angle of about 104.5°, which is why it is bent rather than straight. On freezing, hydrogen bonds link molecules into six-membered rings; ordinary ice is hexagonal (ice Ih), and this molecular hexagonal lattice is the origin of the six-fold symmetry of every snow crystal. (NOAA/NSSL; Kenneth Libbrecht, California Institute of Technology, SnowCrystals.com, 'The Snowflake Primer'; Libbrecht, 'The Physics of Snow Crystals', Reports on Progress in Physics, 2005.) Growth form depends on temperature and humidity — the relationship first mapped by Ukichiro Nakaya, who grew snow crystals under controlled conditions in the 1930s and produced the first artificial snow crystal in 1936. The sequence used here: thin plates near -2°C, needles near -5°C, hollow columns near -8 to -10°C, branching stellar dendrites near -15°C, and columns and plates again below about -22°C, with branching increasing as humidity rises. (Nakaya, 'Snow Crystals: Natural and Artificial', 1954; Libbrecht, SnowCrystals.com, 'The Snowflake Morphology Diagram'.) The six arms of one crystal resemble each other because the crystal is small enough that all six grow in effectively the same temperature and humidity at the same time — not because the arms communicate. (Libbrecht, 'The Snowflake Primer'.) In 1988 Nancy Knight of the National Center for Atmospheric Research identified and photographed two apparently identical snow crystals — hollow columns — collected from a storm over Wisconsin. Simple crystals can be alike; complex ones effectively never are, because of the number of possible arrangements of their fine detail. (NCAR/UCAR; widely reported in the atmospheric science literature; discussed by Libbrecht, 'No Two Alike?'.) A typical snow crystal contains on the order of 10^18 water molecules — written out in the episode as 'a million million million'. (Libbrecht, SnowCrystals.com.) Terminology note kept out of the episode for simplicity: strictly, a 'snow crystal' is a single ice crystal, while a 'snowflake' may be several crystals stuck together. The episode uses 'snow crystal' where the distinction matters.