Every extreme has a name

Fill a 375 millilitre soft drink can with water and it weighs 375 grams. Fill the same can with osmium and it weighs about 8.5 kilograms: near enough to a nine litre bucket of water, shrunk down to something you can close one hand around. Fill it with hydrogen gas instead and you get about 34 milligrams, a couple of grains of rice. That is the full span of everyday matter, and osmium sits at the top of it.

Its weight comes with hardness. Alloyed with iridium it tips fountain pen nibs and the pivots inside instruments, where a soft point would wear flat in a year.

The same soft drink can, filled with lithium, comes to about 200 grams: an apple. Water is 1.0 on this scale, so lithium floats, and it floats high, riding with close to half of itself standing clear of the surface. An ice cube manages about eight per cent. Drop lithium in and it fizzes as it goes, shedding hydrogen into the air.

That lightness is the whole reason your phone lasts a day. A battery has to carry its charge carrier around with it, and gram for gram lithium carries more charge than any other metal.

The hottest setting on a kitchen oven is around 250 degrees Celsius. Tungsten does not soften until about fourteen times that number. Steel puddles into a bright orange pool at roughly 1,500 degrees; at that heat a tungsten bar is still a bar, and stays one for another 1,900 degrees.

Which is why a coiled tungsten wire can sit inside a glass globe glowing white and simply keep glowing, and why the black tip on a masonry drill bit is tungsten locked to carbon.

The coldest air ever recorded at a weather station on Earth was about 89 degrees Celsius below zero, at Vostok in Antarctica. Helium boils about 180 degrees colder again, a bare four degrees above absolute zero, the bottom of the temperature scale that the third law of thermodynamics puts out of reach in any finite number of cooling steps.

Keep cooling it and nothing else happens. At ordinary air pressure helium stays liquid all the way down, because its atoms keep jiggling even at absolute zero and that jiggle is bigger than the weak pull holding them together. To make it hold still you have to squeeze it as well, to roughly 25 times the pressure of the air around you. It is the only element that behaves this way. Hospitals lean on that: the magnet inside an MRI scanner sits in a bath of liquid helium so it can carry current without heating up.

When two atoms share electrons, one of them usually pulls harder. The Pauling scale rates that pull, built from the strength of the bonds atoms actually make, and it runs from about 0.7 at the meek end to 3.98 at fluorine, which tops it. Fluorine takes electrons off almost anything it meets, and it is fluorine chemistry that etches frosted patterns into glass. Score the same pull a different way, from the energy of the outer electrons themselves, and neon and helium come out above it: the record travels with the scale you pick up, which is worth knowing before you quote it.

Tie that pull down inside a stable chain and it turns useful and calm: the slippery grey coating on a frypan is carbon wrapped in fluorine, and the fluoride in toothpaste locks fluorine into tooth enamel to harden it.

At the other end sits francium at about 0.70, though that figure is a calculated estimate est. rather than a bench measurement. Caesium at 0.79 is the lowest bench-measured value.

Every atom holds its own electrons at some grip strength, and that strength can be measured as energy. Take the same count of atoms of each, which works out to a bit over four grams of helium and about 133 grams of caesium. Freeing one electron from every helium atom takes about four kettles' worth of energy, taking a kettle as a full 1.7 litres going from tap to boil, about 570 kilojoules. The same job on the caesium takes about two thirds of a kettle.

Caesium's loose hold is what makes it useful. Shine light on it and electrons come away, which is how some light sensors work; and the wobble of its outer electron is so steady that caesium clocks define the length of a second, which is what keeps phone networks and bank transfers in step.

Francium is usually expected to beat caesium here, and it does not: its measured value sits about four and a half per cent higher, so caesium keeps the record.

A picometre is a millionth of a millionth of a metre, so both ends of this record are past imagining until you scale them up. Blow a helium atom out to the size of a twenty cent coin, 28.5 millimetres across, and a caesium atom on that same scale comes out about 248 millimetres wide: a dinner plate beside a coin.

Back at real size, each atom takes up two of those reaches, so a hundred million caesium atoms shoulder to shoulder would stretch about 49 millimetres, roughly the length of your thumb. Francium is calculated to be wider still at about 260 picometres est., but nobody has ever gathered enough of it to check.

Siemens per metre is simply a rating of how easily a metre of something passes a current. Silver tops it at about 63 million, with copper just behind at about 59.6 million: put them side by side and a silver wire carries what a copper wire with six per cent more metal in its cross-section would, which is about three per cent fatter. It leads on heat too: a silver teaspoon warms along its handle faster than any other metal.

Six per cent is a real margin and still not worth the price, which is why the wiring in your walls is copper. Silver goes where only a few grams are needed and the join has to stay clean: the fine grey lines printed across the face of a rooftop solar panel, and the contacts inside a light switch that must not pit after twenty thousand flicks.

Weigh out a tonne of the ordinary matter in the universe and about 750 kilograms of it is hydrogen, another 230 is helium, and everything else in existence shares the last twenty. Count atoms instead of weighing them and it is starker still: scoop up a hundred atoms at random and about 93 are hydrogen and 7 are helium. You would need to scoop about 1,300 before expecting a single atom of oxygen.

It is close to hand as well as far away. Two atoms in every three in the water you drink are hydrogen.

Load a ute tray with a tonne of ordinary rock and you are carrying about 461 kilograms of oxygen, none of it breathable. It is locked into the minerals, bound tight to silicon and to metals.

Look at what that becomes. Beach sand is silicon with two oxygens on it. Window glass is the same pairing melted and cooled. The concrete in the driveway is oxygen bound to calcium, silicon and aluminium. Nearly everything solid you stand on or build with is mostly oxygen by weight, which is a strange thing to know about a gas.

Astatine forms in trace amounts as heavier elements break down, and it leaves almost as fast as it arrives. Reported estimates of how much exists in the entire crust at any one moment range from under a gram to a few dozen grams est.: on the low reading, every rock on the planet between them holds less astatine than would sit on the tip of a teaspoon. The standard reference tables give no crust figure for it at all, and none for francium either.

Nobody has seen a visible piece. Gather enough in one place and its own radioactive heat would boil it away before you could look at it. Its bulk numbers, the density and the melting point, are all calculated rather than weighed. One isotope, astatine-211, is being studied as a cancer treatment that would carry radiation to a tumour cell by cell.

In November 2016 the names nihonium, moscovium, tennessine and oganesson were confirmed for elements 113, 115, 117 and 118, closing the last four gaps in the bottom row of the table. They are made an atom at a time by firing one beam of nuclei at another and waiting, sometimes for months, for a single fusion to land.

They do not stay. Oganesson's known form has a half-life under a millisecond, and a blink of your eye takes around 300 milliseconds: hundreds of half-lives pass while your eyelid is down. Their densities, melting points and atom sizes are all calculated, because the total ever made would not cover a pinhead.