Erdős Problem #1063
Editorial depth: Authored explanatory DRAFT — NOT editorially accepted.
Snapshot classification: open (not automatically the live result).
Subject areas: number theory.
Read this status correctly
The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.
UNACCEPTED DRAFT — mathematical editorial review pending
This material is a provisional, source-qualified manuscript published for line-by-line scrutiny and real HTML/LaTeX testing. It has NOT received separate GCL editorial acceptance or independent mathematical certification. Its presence on this website is not evidence of a solution, reviewed coverage, or final publication.
Review the underlying draft on GitHub.
The question in plain language
Take the binomial coefficient
For \(k\ge2\), look at the \(k\) consecutive integers \(n,n-1,\ldots,n-k+1\). Erdős and Selfridge observed that, when \(n\ge2k\), at least one of these integers fails to divide \(\binom nk\).
Define \(n_k\) as the smallest \(n\ge2k\) for which exactly one of those \(k\) divisibility tests fails. Erdős Problem #1063 asks how quickly \(n_k\) grows with \(k\).
A small example
For \(k=4\) and \(n=8\),
Both \(8\) and \(6\) fail to divide \(70\), so \(n=8\) does not qualify. For \(n=9\), however,
Among \(9,8,7,6\), only \(8\) fails to divide \(126\). Therefore \(n_4=9\), since \(n\ge2k=8\) rules out any smaller candidate.
Independent exact-integer enumeration gives
The sequence is not monotone across these small values. This is a finite observation, not a general growth theorem.
What is known
Erdős and Selfridge posed the question as Problem 6447 in the American Mathematical Monthly (1983), page 710. Jean-Marie Monier's published solution (1985), pp. 435–436, records an upper bound \(n_k\le k!\) for \(k\ge3\). Stijn Cambie's later observation improves this to
and the standard asymptotic for the least common multiple yields an exponential upper bound of roughly \(\exp((1+o(1))k)\). These bounds do not determine the true order of growth.
The maintained catalogue remains open. The full original 1983 and 1985 proofs have not yet been independently inspected in GCL's source audit.
GCL's current state
Source audit #2334 compares the published formulations with the frozen formal provider. That file includes formal proofs of the first four values and unproved placeholders in some more general statements.
Our local enumeration checks the first eight values directly from integer binomial coefficients. No deep upper-bound argument or later external formal development has been independently kernel-replayed by GCL.
What remains
A true resolution would determine the growth of \(n_k\) much more precisely. Near-term source work should inspect the original papers and prove the least-common-multiple upper bound from first principles, preserving the exact “one exceptional divisor” quantifier. Larger exact searches and proof replay are useful tests but cannot establish the asymptotic order by themselves.
Sources and claim boundaries
- Erdős Problem #1063, statement and reported bounds.
- P. Erdős and J. L. Selfridge, Problem 6447, American Mathematical Monthly (1983), p. 710.
- J.-M. Monier, solution to Problem 6447, American Mathematical Monthly (1985), pp. 435–436.
- Frozen formal-conjectures file.
- Authenticated source audit #2334 and finite calculation script; partial source verification only, no mathematical proof certification or editorial admission.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #2334 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: PROTECTED_READY_BACKLOG.
- Formal-statement coverage: A provider statement was included in the pinned snapshot. Presence does not mean the statement or proof has been checked.
- External mathematical status: consult current problem page and its dated bibliography.