Checking whether one number divides another by performing the full division works — but it is slow, especially for large numbers. For the most common divisors, patterns in the decimal digits provide an answer instantly. Each pattern is a consequence of how modular arithmetic interacts with the base-10 place value system, compressed into a rule simple enough to apply mentally.
What Are Divisibility Rules?
A divisibility rule is a shortcut that determines whether a number is divisible by a given divisor without carrying out the division. Instead of computing 7,836÷3 and checking for a remainder, the rule says: add the digits 7+8+3+6=24, and check whether that sum is divisible by 3. It is, so 3∣7836.
Every rule exploits the structure of base-10 representation. A number like 5,432 is shorthand for 5⋅103+4⋅102+3⋅10+2. The behavior of the powers of 10 under modulo determines which digit patterns reveal divisibility — and different divisors respond to different patterns.
The rules are shortcuts, not definitions. The definition of divisibility is a∣b when b=a⋅k for some integer — equivalently, when bmoda=0. The rules provide fast paths to that same conclusion.
Divisibility by 2
A number is divisible by 2 if its last digit is even: 0,2,4,6, or 8.
Since 10 is divisible by 2, every power of 10 is also divisible by 2. The contribution of each digit beyond the ones place — d⋅10k for k≥1 — is automatically a multiple of 2. Only the final digit d0 remains outside that guarantee, so d0 alone determines divisibility.
The number 1,746 ends in 6 (even), so 2∣1746.
The number 3,821 ends in 1 (odd), so 2∤3821.
The even numbers — …,−4,−2,0,2,4,6,… — are precisely the integers divisible by 2. The odd numbers are those that leave a remainder of 1.
Divisibility by 3
A number is divisible by 3 if the sum of its digits is divisible by 3.
The reason traces to a single congruence: 10≡1(mod3). Every power of 10 is also congruent to 1 modulo 3: 102≡1, 103≡1, and so on. A digit d in the kth place contributes d⋅10k≡d⋅1=d(mod3). The place value drops away entirely, and only the digit itself matters. The number is congruent to the sum of its digits modulo 3.
The number 729: 7+2+9=18, and 1+8=9. Since 3∣9, the number is divisible by 3.
The number 1,234: 1+2+3+4=10, and 1+0=1. Since 3∤1, the number is not divisible by 3.
The process can be repeated: if the digit sum is large, sum its digits again. The chain always terminates at a single digit, and that digit determines divisibility.
Divisibility by 4
A number is divisible by 4 if its last two digits form a number divisible by 4.
The number 100 is divisible by 4, and so is every higher power of 10. Every digit beyond the tens place contributes a multiple of 100 — automatically a multiple of 4. Only the last two digits affect the remainder when dividing by 4.
The number 3,516: the last two digits form 16, and 16÷4=4 exactly. So 4∣3516.
The number 7,230: the last two digits form 30, and 30÷4=7.5. So 4∤7230.
For quick mental checks, it helps to know which two-digit endings are divisible by 4: 00,04,08,12,16,20,24,28,32,36,40,44,… — every fourth number from 00 onward.
Divisibility by 5
A number is divisible by 5 if its last digit is 0 or 5.
The logic is the same as for 2: since 10 is divisible by 5, every digit beyond the ones place contributes a multiple of 5. Only the last digit matters, and the only single digits divisible by 5 are 0 and 5.
The number 4,835 ends in 5, so 5∣4835.
The number 7,832 ends in 2, so 5∤7832.
This is the second simplest rule after divisibility by 10 — a single-digit check with only two possible passing values.
Divisibility by 6
A number is divisible by 6 if it is divisible by both 2 and 3.
Since 6=2×3 and gcd(2,3)=1, divisibility by 6 requires passing both individual tests: the last digit must be even, and the digit sum must be divisible by 3.
The number 1,254: last digit 4 (even — passes the 2 test), digit sum 1+2+5+4=12 (divisible by 3 — passes the 3 test). Both conditions hold, so 6∣1254.
The number 2,135: last digit 5 (odd — fails the 2 test). No need to check the digit sum. The number is not divisible by 6.
This approach generalizes. For any composite divisor n=a×b where a and b are coprime, divisibility by n is equivalent to divisibility by both a and b.
Divisibility by 8
A number is divisible by 8 if its last three digits form a number divisible by 8.
The number 1,000 is divisible by 8 (1000=8×125), and so is every higher power of 10. Digits beyond the hundreds place contribute multiples of 1,000 — all divisible by 8. Only the last three digits affect the outcome.
The number 53,104: the last three digits form 104, and 104÷8=13 exactly. So 8∣53104.
The number 12,345: the last three digits form 345, and 345÷8=43.125. So 8∤12345.
This rule is harder to apply mentally than the rules for 2 or 5, since it requires dividing a three-digit number by 8. For quick estimates, checking whether the three-digit number is close to a known multiple of 8 (8,16,24,…,992,1000) can help.
Divisibility by 9
A number is divisible by 9 if the sum of its digits is divisible by 9.
The reasoning is identical to the rule for 3: 10≡1(mod9), so every power of 10 is congruent to 1 modulo 9, and each digit contributes only its face value. The number is congruent to its digit sum modulo 9.
The number 8,127: 8+1+2+7=18, and 18÷9=2 exactly. So 9∣8127.
The number 4,825: 4+8+2+5=19, and 19÷9=2 remainder 1. So 9∤4825.
Since 9=32, every number divisible by 9 is automatically divisible by 3 — but not the other way around. The number 729 is divisible by both; the number 12 is divisible by 3 but not by 9.
Divisibility by 10
A number is divisible by 10 if its last digit is 0.
This is the simplest rule of all. The base of the number system is 10, so a number has no remainder when divided by 10 exactly when the ones place is empty.
The number 5,230 ends in 0, so 10∣5230.
The number 5,235 ends in 5, so 10∤5235.
Since 10=2×5, divisibility by 10 requires divisibility by both 2 and 5. The only digit that is simultaneously even and a multiple of 5 is 0 — which is why the rule reduces to a single condition.
Divisibility by 11
A number is divisible by 11 if the alternating sum of its digits (from right to left: add, subtract, add, subtract, …) is divisible by 11.
The congruence behind the rule is 10≡−1(mod11). Successive powers of 10 alternate: 100=1, 101≡−1, 102≡1, 103≡−1. Each digit's contribution to the remainder alternates in sign depending on its position.
The number 9,273: alternating sum from the right gives 3−7+2−9=−11. Since 11∣(−11), the number is divisible by 11.
The number 5,831: alternating sum gives 1−3+8−5=1. Since 11∤1, the number is not divisible by 11.
If the alternating sum is negative, that is fine — a negative multiple of 11 still indicates divisibility. The result −11, −22, or 0 all confirm the number is divisible by 11.
Combining Rules
Divisibility by a composite number can be tested by checking its coprime factors separately.
Is 2,340 divisible by 12? Since 12=4×3 and gcd(4,3)=1, check each factor independently. Last two digits: 40, and 40÷4=10 exactly — divisible by 4. Digit sum: 2+3+4+0=9, divisible by 3. Both conditions hold, so 12∣2340.
Is 8,910 divisible by 15? Since 15=5×3 and gcd(5,3)=1, check both. Last digit: 0 — divisible by 5. Digit sum: 8+9+1+0=18 — divisible by 3. Both pass, so 15∣8910.
The coprimality condition is essential. Testing divisibility by 12 as "divisible by 2 and divisible by 6" is not sufficient, because gcd(2,6)=2=1. The factorization must split 12 into coprime components — 4 and 3 — for the combined test to work.
Which of 2,3,4,5,6,8,9,10 divide 4,320? Last digit 0: divisible by 2, 5, and 10. Digit sum 4+3+2+0=9: divisible by 3 and 9. Since divisible by both 2 and 3: divisible by 6. Last two digits 20: 20÷4=5 — divisible by 4. Last three digits 320: 320÷8=40 — divisible by 8. The number passes every test.
Is 7,623 divisible by 3? Digit sum: 7+6+2+3=18. Since 3∣18, yes.
Is 7,623 divisible by 9? Same digit sum: 18. Since 9∣18, yes.
Is 7,623 divisible by 6? Last digit 3 is odd, so 2∤7623. Therefore 6∤7623, regardless of the digit sum.
Is 31,724 divisible by 11? Alternating sum from the right: 4−2+7−1+3=11. Since 11∣11, yes.
Common error: testing divisibility by 12 by checking 2 and 6 instead of 4 and 3. The number 18 is divisible by both 2 and 6, but 12∤18. Coprime factorization is required.
Why These Rules Work
Every divisibility rule is a consequence of how the base 10 behaves under modular arithmetic.
A number written in decimal is dk⋅10k+dk−1⋅10k−1+⋯+d1⋅10+d0. Reducing this expression modulo a divisor n replaces each 10j with its remainder modulo n, and the pattern of those remainders determines the rule.
For n=2 or n=5: 10≡0, so all terms except d0 vanish. Only the last digit matters.
For n=4: 100≡0(mod4), so all terms beyond d1⋅10+d0 vanish. The last two digits determine divisibility.
For n=8: 1000≡0(mod8), extending the same logic to the last three digits.
For n=3 or n=9: 10≡1, so 10k≡1 for all k. Every digit contributes its face value and nothing more. The number reduces to its digit sum.
For n=11: 10≡−1, so 10k alternates between 1 and −1. Digits contribute with alternating signs, producing the alternating-sum rule.
Each rule is a human-readable compression of a modular identity. The modular arithmetic page develops these identities in full.