The whole numbers have a hidden architecture — primes are the atoms, and a clock is a number system all its own.
You have spent the whole curriculum learning what numbers do — they add, scale, slope, integrate. Number theory asks a quieter question: what are the whole numbers made of? It turns out the integers have a hidden architecture. A handful of numbers — the primes — are the indivisible atoms, and every other whole number is built from them in exactly one way. With nothing but division and remainders you can find the largest common factor of two giant numbers in a few lines, and you can build a brand-new arithmetic that loops around like the face of a clock. These ideas look like a charming curiosity. They are also the machinery that keeps your online life private — the security behind every padlock in your browser leans on primes and clock arithmetic. Here is a first, honest tour.
Start with the most basic relationship between two whole numbers: does one go evenly into the other? We write b ∣ a — read "b divides a" — to mean there is a whole number k with a = b · k. No remainder left over. So 3 ∣ 12 because 12 = 3 · 4, but 3 ∤ 13, because the closest we get is 13 = 3 · 4 + 1, with a stubborn 1 left behind.
The honest picture is a rectangle of dots. Lay out n dots in d columns. If the columns come out perfectly even, the rectangle is full — and that is exactly what d ∣ n means. If the last row is ragged, the gap you see is the remainder r. Every division of whole numbers can be written once, cleanly:
a = q · b + r, with 0 ≤ r < b
This is the division algorithm, and the remainder r is the hero of this whole lesson. Divisibility is simply the case r = 0.
Divide: 30 = 4 · 7 + 2. The quotient is 4, the remainder is 2. Since r = 2 ≠ 0, 7 ∤ 30. The full list of numbers that do divide 30 — its divisors — is 1, 2, 3, 5, 6, 10, 15, 30.
Some numbers refuse to be split into a rectangle wider than one column. A prime is a whole number greater than 1 whose only divisors are 1 and itself: 2, 3, 5, 7, 11, 13, … These are the atoms. A number with more divisors — 12, 30, 100 — is composite: it can be broken into smaller factors.
To hunt primes by hand, the ancient Greek sieve of Eratosthenes is unbeatable. Write the numbers in a grid, circle 2 (prime), then cross out every later multiple of 2; circle the next survivor, 3, and cross out its multiples; and so on. Whatever survives is prime. The crossing-out is just divisibility, applied over and over.
Now the deep fact. Take any whole number bigger than 1 and keep splitting it into smaller factors until you can split no further — every leaf of the resulting factor tree is prime. Do it a different way and you get the same atoms, possibly reshuffled. This is the Fundamental Theorem of Arithmetic: every integer greater than 1 factors into primes in exactly one way (up to order).
12 = 2² · 3, 60 = 2² · 3 · 5, 360 = 2³ · 3² · 5
This uniqueness is why primes deserve the name "atoms." It is the bedrock that makes fractions reduce to lowest terms in one definite way, and it is the reason every later result in this lesson holds together.
It is tempting to call 1 the smallest prime. Don't. A prime has exactly two divisors; 1 has only one (itself), so it sits in a class of its own — a "unit." If we let 1 be prime, uniqueness would shatter: 12 = 2²·3 = 2²·3·1 = 2²·3·1·1 = … There has to be one factorization, so 1 is deliberately left out.
Two numbers usually share some factors. The largest one they share is the greatest common divisor, written gcd(a, b). You could factor both numbers and compare — but for big numbers that is slow, and factoring is genuinely hard. There is a stunningly fast shortcut, over two thousand years old: Euclid's algorithm.
The idea rests on a single observation. Any number dividing both a and b must also divide the remainder r = a − q·b. So the common divisors of (a, b) are exactly the common divisors of (b, r) — a smaller pair with the same gcd. Replace the larger number by the remainder and repeat. The numbers shrink fast; when the remainder finally hits 0, the last nonzero remainder is the gcd.
48 = 2·18 + 12 → 18 = 1·12 + 6 → 12 = 2·6 + 0.
The last nonzero remainder is 6, so gcd(48, 18) = 6. Check: 48 = 6·8 and 18 = 6·3, and 8 and 3 share nothing — 6 really is the largest common factor.
Here is where number theory opens a door into a whole new world. A clock face has only the hours 0 through 11; add 5 hours to 9 o'clock and you do not get 14 — you wrap around to 2. The clock has quietly thrown away everything except the remainder after dividing by 12. That is the entire idea of congruence.
Fix a modulus m. Two numbers are congruent mod m — written a ≡ b (mod m) — when they leave the same remainder on division by m; equivalently, when m ∣ (a − b). Every integer is congruent to exactly one of 0, 1, 2, …, m−1: its residue. The whole number line collapses onto the dial.
17 ≡ 5 (mod 12), 29 ≡ 5 (mod 12), −7 ≡ 5 (mod 12)
All three land on the same spot — they are the same "time of day." Congruence is not a vague approximation of equality; it is exact equality of remainders.
What makes congruence powerful is that it behaves like ordinary arithmetic for adding and multiplying. If a ≡ a′ and b ≡ b′ (mod m), then
a + b ≡ a′ + b′ and a · b ≡ a′ · b′ (mod m).
That single rule does real work. Last digit of a big power? The last digit is just the value mod 10. Since 7 ≡ 7, the powers of 7 cycle 7, 9, 3, 1 and repeat with period 4 — so the last digit of 7100 is the last digit of 74, which is 1. Day of the week? Work mod 7. Divisible by 9? A number is congruent to the sum of its digits mod 9 — that is why "casting out nines" catches arithmetic slips.
The days repeat with period 7, so reduce mod 7: 60 = 8·7 + 4, i.e. 60 ≡ 4 (mod 7). Count 4 days on from Wednesday → Thursday, Friday, Saturday, Sunday. We never had to count all 60 days.
Adding and multiplying are safe; cancelling a common factor is not. Mod 6 we have 2·4 = 8 ≡ 2 and 2·1 = 2, so 2·4 ≡ 2·1 (mod 6) — yet "dividing by 2" would claim 4 ≡ 1, which is false (4 and 1 are different residues mod 6). Cancellation only works when the factor shares no common divisor with the modulus. Keep that one trap in mind and clock arithmetic is wonderfully reliable.
| Idea | What it says | The honest test |
|---|---|---|
| Divisibility | b ∣ a means a = b·k, no remainder | write a = q·b + r; check r = 0 |
| Primes & FTA | every n > 1 is a unique product of primes | factor tree; 12 = 2²·3 (and 1 is not prime) |
| gcd | the largest shared factor of a, b | Euclid: last nonzero remainder, gcd(48,18)=6 |
| Congruence | a ≡ b (mod m): same remainder, m ∣ (a−b) | the residue, 17 ≡ 5 (mod 12) |
| Mod arithmetic | + and × respect congruence (but not ÷) | last digit, day of week, casting out nines |
Six questions to lock it in. Tap the answer you think is right.
This lesson is enrichment beyond the standard high-school sequence — a guided glimpse at number theory, the gateway to competition mathematics and to modern cryptography (RSA encryption rests directly on primes and modular arithmetic). The seeds were planted in grade school with factors, multiples, and primes (factors, multiples & primes). Four ideas anchor it: divisibility through the division algorithm a = q·b + r; unique prime factorization (the Fundamental Theorem of Arithmetic, with the deliberate convention that 1 is not prime); the Euclidean algorithm for the gcd; and congruence as arithmetic on a clock. The central caution to reinforce is that congruences add and multiply freely but cannot be divided. From here the road forks toward abstract algebra and the theory of numbers — a quiet door swinging open onto a vast field.