;colony/science  / Mathematics, Made Visual  / Why does 0.999… equal exactly 1?
Mathematics, Made Visual

Why does 0.999… equal exactly 1?

0.999… and 1 are the very same point on the number line, written two different ways. Once you see why, there is no gap left to shrink.

Plate 155 — The vanishing gap geometric series · gap = 10⁻ⁿ → 0
Add nines one at a time and watch the gap to 1 collapse.
Predict firstAs you add more nines, does the gap to 1 settle at some tiny fixed size, or keep heading toward zero?
the gap to 1 shrinks ×10 with every nine0.9… gap to 1 = 10⁻¹ = 0.1 0.90 1.0010.9 the gap9⁄10 + 9⁄100 + 9⁄1000 + … → 1
PLATE 155 · THE VANISHING GAP
Number of nines 0.9…
Add nines and watch the gap to 1 collapse by a factor of ten each time.
Gap to 1
10⁻¹→ 0
Nines so far
1partial sum 0.9
Here's the clean proof: one third written as a decimal is 0.333…, going on forever. Multiply both sides by three. The left side is 1. The right side is 0.999…. So 0.999… and 1 are two names for the same number. The slider shows why: every nine you add shrinks the gap to 1 by ten. There's no last nine and no smallest leftover gap — push it as far as you like and the gap heads to nothing at all.
Try with the plate
  • Add nines until the gap to 1 reads 10⁻⁹ or smaller
  • Use the bar to confirm the partial sums 9/10 + 9/100 + … are climbing toward exactly 1

0.999… equals exactly 1 because a repeating decimal is the limit of its partial sums, and those sums, 1 − 10⁻ⁿ, head to 1. The gap shrinks by a factor of ten with every nine and reaches zero, so no number lies between 0.999… and 1. The quickest proof: 1/3 = 0.333…, and multiplying by three gives 1 = 0.999….

The short answer

Almost everyone's gut reaction is that 0.999…, with nines going on forever, must be just a tiny bit less than 1. It feels like it's always falling a hair short. But it isn't short at all — it is exactly 1, the same number wearing a different costume. Here's the quickest way to believe it. You already accept that one third is 0.333…, with threes repeating forever. Multiply both sides by three. Three times one third is 1. Three times 0.333… is 0.999…. So 0.999… and 1 have to be the same thing. The slider shows what's really going on: each nine you add shrinks the gap to 1 by a factor of ten — to a tenth, a hundredth, a thousandth. There's no last nine, so there's no smallest leftover gap. The distance to 1 doesn't stop at some tiny value; it heads all the way to nothing.

The common mix-up

Most people think 0.999… is a number forever creeping up on 1 but always falling a hair short. In fact it is the limit of its partial sums, that limit is exactly 1, and no number lies between them.

What's actually happening

Few facts in mathematics provoke as much stubborn disbelief as this one: 0.999…, with the nines repeating forever, equals 1 exactly. Not approximately, not in the limit as a figure of speech, but precisely and literally the same number. People argue about it in pubs and online for hours. The resistance comes from a very natural picture in our heads: we imagine 0.999… as a process, a number that keeps adding nines and forever creeps toward 1 without quite arriving. If it's always still climbing, surely it never gets there. The trouble is that this picture quietly misunderstands what an infinite decimal actually means.

Start with the proof that convinces most people on the spot, because it uses something they already accept. We are all happy to write one third as 0.333…, threes marching on forever. Now simply multiply both sides of that by three. On the left, three lots of one third is plainly 1. On the right, three lots of 0.333… is 0.999…. The two sides were equal before we multiplied, so they're equal after: 0.999… = 1. There's a second, equally clean argument. Call the number x = 0.999…. Multiply by ten to get 10x = 9.999…. Subtract the first equation from the second and the endless tail of nines cancels exactly, leaving 9x = 9, so x = 1. No sleight of hand, no rounding.

The deepest way to see it, though, is to be honest about what an infinite decimal is. It is not a number with infinitely many digits already laid down; it is the limit that its finite pieces approach. The pieces here are 0.9, 0.99, 0.999, and so on, each equal to 1 minus a tenth, a hundredth, a thousandth. That leftover gap, 10 to the minus n, shrinks by a factor of ten with every nine you add and races toward zero. Crucially, there is no final nine, so there is no smallest possible gap — any gap you propose is beaten by adding one more nine. In the real numbers, if two values have nothing strictly between them, they are the same value, and nothing fits between 0.999… and 1. The usual protest, that some infinitely small sliver must remain, fails for a simple reason: the ordinary number line contains no nonzero infinitesimals to be that sliver. The gap doesn't get tiny. It is gone. The simulator lets you watch it vanish, nine by nine.

Remember this

0.999… equals 1 because the gap between them shrinks to zero, leaving no number in between — they are one point with two names.

Try it at home Hunt for the gap
  1. 1Try to write down a number that sits strictly between 0.999… and 1. Whatever you propose, 0.999… already has more nines than your number can beat.
  2. 2Now do the subtraction by hand: write 1.000… above 0.999… and subtract. The differences in every column race to zero, with no smallest nonzero remainder anywhere.
  3. 3Conclude that since nothing fits between them and their difference is zero, the two are the same number. The slider's shrinking gap is the same idea made visible.

Common questions

Isn't 0.999… just infinitely close to 1, not actually equal?

No. 'Infinitely close' would mean a nonzero infinitesimal gap remains, but the ordinary real number line contains no nonzero infinitesimals. The gap 10⁻ⁿ genuinely reaches zero, so the two values are equal, not merely close.

What's the simplest proof?

Everyone accepts that 1/3 = 0.333…. Multiply both sides by three: the left side is 1 and the right side is 0.999…, so 0.999… = 1. Alternatively, let x = 0.999…, then 10x = 9.999…, and subtracting gives 9x = 9, so x = 1.

Does this mean every terminating decimal has two names?

Yes. Any decimal that ends can also be written with a trailing string of nines — for example 1 = 0.999…, 0.5 = 0.4999…, and 2.7 = 2.6999…. This double representation is a normal feature of the decimal system, not a flaw.

Built & checked by Nilesh Singh · how this is made · last updated June 2026