Skip to content

Erdős Problem #1081

Editorial depth: Authored explanatory DRAFT — NOT editorially accepted.
Snapshot classification: disproved (not automatically the live result).
Subject areas: number theory, powerful.

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

A positive integer is squarefull when every prime factor occurs to exponent at least two. Thus \(4\), \(8\), \(9\) and \(36\) qualify, but \(12=2^2\cdot3\) does not. The number \(1\) qualifies vacuously.

Erdős Problem #1081 asks how many distinct integers at most \(x\) can be expressed as a sum of two positive squarefull integers. Writing this number as \(A(x)\), Erdős guessed

\[ A(x)\sim c\frac{x}{\sqrt{\log x}} \]

for some positive constant \(c\). That guess is false.

A small example

The first examples include \(2=1+1\), \(5=1+4\), \(8=4+4\), \(9=1+8\), and \(10=1+9\).

Exact trial factorization finds 25 squarefull positive integers through \(250\), including \(1\), and 151 distinct sums of two such numbers at most \(250\). Therefore \(A(250)=151\) under this explicit convention. Counting multiple representations of the same integer would answer a different question.

What is known

Odoni (1981) disproved the proposed asymptotic. Later work sharpened the counting estimates. A published article by Valentin Blomer, Binary Quadratic Forms with Large Discriminants and Sums of Two Squareful Numbers II, Journal of the London Mathematical Society 71 (2005), 69–84, states in its author abstract that

\[ A(x)=x(\log x)^{-\alpha+o(1)}, \qquad \alpha=1-2^{-1/3}\approx0.2063. \]

The maintained catalogue attributes a more refined logarithmic-factor estimate to Blomer and Granville (2006). The correct exponent is not \(1/2\).

GCL's current state

Authenticated source audit #2352 checked the published author abstract, distinguished the 2005 Blomer result from the 2006 Blomer–Granville refinement, and independently enumerated \(A(250)\).

GCL has not replayed the original analytic proof. No finite table proves a logarithmic asymptotic.

What remains

Review Odoni's original theorem and both later papers, including the correct powers of \(\log\log x\), the convention concerning \(1\), and whether the source counts distinct integers or representations. Extend finite enumeration only as a reproducibility check.

Sources and claim boundaries

Erdős Problem #1081; Odoni (1981), Acta Arithmetica, pp. 145–162; Blomer (2005), DOI; Blomer–Granville (2006), Duke Mathematical Journal, pp. 261–302; GCL return #2352. The published disproof is external, and this manuscript remains unaccepted.

Catalogue and evidence provenance

  • Pinned catalogue source: teorth/erdosproblems at 536408716724.
  • GCL issue: #2352 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
  • GCL triage: OUTSIDE_STRICT_OPEN_TRIAGE.
  • Formal-statement coverage: No statement was confirmed in GCL's pinned formal-provider snapshot. Presence does not mean the statement or proof has been checked.
  • External mathematical status: consult current problem page and its dated bibliography.

Back to catalogue · Compendium introduction