A Quarter of Numbers Below 100 Are Prime, and an Eighth Below 10 000
Gepubliceerd op 16-11-2026 · 3 min leestijd · Dagelijkse calculators
Antonio Delgado — Redacteur tradities en tijdverdrijf bij OneKitly
Symboliek · Kalenders
Getoetst aan 2 bronnen
Count the primes below a limit and the density falls steadily: 25 out of the first 100, so one in four; 168 out of the first thousand, one in six; 1229 out of ten thousand, one in eight. The thinning follows a rule found by Gauss as a teenager and proved a century later — the number of primes below n is close to n divided by the natural logarithm of n. That estimate gives 145 for a thousand against the true 168, and 1086 for ten thousand against 1229, closing the gap in percentage terms as the numbers grow. Euclid had already proved the list never ends, so the primes get rarer forever without ever running out.
25 primes below 100, 168 below 1000, 1229 below 10 000. They thin out, they never stop, and the rate of thinning has a formula.
Euclid's proof in three sentences
Suppose the primes stopped somewhere. Multiply them all together and add one; the result leaves a remainder of one when divided by any prime on the list, so none of them divides it. Either the new number is itself prime or it has a prime factor outside the list, and in both cases the list was incomplete. Two thousand three hundred years later, no shorter argument for an infinite set has been written, and it needs nothing beyond multiplication and the idea of a remainder.
Deserts of arbitrary length
Thinning on average does not mean thinning smoothly. It is easy to build a run of a hundred consecutive numbers with no prime among them: take 101 factorial and list the numbers from two more than it to 101 more than it, each of which is divisible by the amount added. The same construction produces a gap of any length you ask for, somewhere far out on the number line. Yet twin primes — pairs two apart — keep appearing as far as anybody has looked, and whether they do so forever is still open.
| Limit | Primes | Estimate n/ln n |
|---|---|---|
| 100 | 25 | 22 |
| 1000 | 168 | 145 |
| 10 000 | 1229 | 1086 |
Uitgewerkt met onze eigen calculator
Lijst van priemgetallen
Gegeven
- Bovengrens (2–100000)
- 50
Resultaat
- Hoeveel priemgetallen
- 15
- Som
- 328
Deze cijfers komen uit de calculator hieronder, ze zijn niet met de hand ingetypt — ze worden opnieuw berekend telkens als de tool verandert.
Reken het door met je eigen cijfers →Veelgestelde vragen
- Why does the logarithm show up at all?
- Because the chance that a number near n survives division by every prime below its square root falls roughly like one over the logarithm of n. The heuristic is not a proof and it lands very close: multiplying together the probabilities of dodging 2, then 3, then 5, and so on gives a product that behaves like 1/ln n. The proper theorem needed complex analysis and took until 1896, and the estimate a schoolboy could have guessed turned out to be the right one.
- Is the estimate always low?
- The simple n over ln n version is, throughout the range anybody computes by hand. A refined version using an integral rather than a division overshoots instead, and the question of where it first switches from over to under was settled only by showing such a crossing must exist somewhere beyond 10 to the 300 — without anyone finding it. Number theory is full of statements verified for every number ever checked and still unproven, which is why checking is not the same as knowing.
Artikelen die je misschien interesseren
Alle gidsen →Vergelijkbare tools
Bronnen
Een fout in dit artikel gezien?