Erdős Problem #668
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: open (not automatically the live result).
Subject areas: geometry, distances.
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.
The question in plain language
Among all sets of \(n\) distinct points in the plane, some sets determine as many unit-distance pairs as possible. How many geometrically different maximizing configurations exist?
Erdős Problem #668 asks whether the number of pairwise noncongruent maximizers tends to infinity as \(n\) grows. A subsidiary question asks whether there are always more than one for \(n>3\). That second assertion is false already for \(n=4\), while the asymptotic first question remains OPEN.
Two configurations are congruent if a Euclidean distance-preserving motion, possibly including reflection, carries one to the other. Counting only nonisomorphic graphs of unit-distance contacts is not equivalent to counting noncongruent point configurations.
A small example
Two equilateral triangles of side length one, on opposite sides of a shared side, use four points. There are five unit-distance edges: the shared base and the four edges joining its endpoints to the two opposite apexes. The distance between the two apexes is \(\sqrt3\), so the sixth pair is not at unit distance.
This example is uniquely optimal at \(n=4\). All six pairs cannot be at distance one, because four mutually equidistant points cannot lie in the Euclidean plane. Thus the maximum is exactly five. Any four-point configuration with five unit pairs has unit-distance graph \(K_4\) with one edge missing. The two vertices of that missing edge are each at distance one from both remaining vertices. The intersections of two unit circles around the remaining base endpoints are the two opposite equilateral apexes. Their positions are forced, up to a rigid motion or reflection.
Hence the number of noncongruent maximizers at \(n=4\) is
$$
M(4)=1.
$$
This independently checked elementary argument refutes only the subsidiary every \(n>3\) assertion. One exceptional small \(n\) cannot refute a claim that \(M(n)\to\infty\).
What is known
Let \(U(n)\) be the largest number of unordered pairs at unit distance among \(n\) distinct planar points. Let \(M(n)\) be the number of congruence classes of configurations attaining \(U(n)\). The principal question is \(M(n)\to\infty\).
The maintained source refers to computational results by Engel, Hammond-Lee, Su, Varga and Zsámboki, and by Alexeev, Mixon and Parshall, suggesting uniqueness for several sizes between five and twenty-one. These computations largely addressed unit-distance graph isomorphism. Different Euclidean realizations can sometimes share a graph, so a graph count must not be silently presented as a classification of geometric congruence classes.
Problem #90 concerns the separate value of \(U(n)\) itself, whereas #668 concerns the multiplicity of extremizers. Even perfect knowledge of the value \(U(n)\) would not automatically enumerate all maximizers.
GCL's current state
The original authenticated RESULT/1 source return has historical disposition PARTIAL. This status and the original research exceptions are preserved, not silently converted into a theorem verdict. The maintained Erdős source and source-specific mathematical definitions were reviewed, and this article is editorially reviewed as a source-exception-annotated explanation under EC-BULK-005-R10, effective only after the exact-hash protected admission merges.
Source interpretation and unresolved mathematical qualification: Unit-distance graph isomorphism does not imply Euclidean congruence of extremal point configurations; eventual multiplicity is open.
The explanation is accepted for publication of established background, examples, and explicitly attributed current claims. It does not establish or formally certify the disputed original theorem, external model-assisted proof, historical interpretation, or any unreplayed Lean kernel file. Original source-fidelity and proof-review research holds remain.
What remains
A bounded useful successor would select one small \(n\) and enumerate both: the graphs that can realize \(U(n)\) edges, and the geometric embeddings of each graph, counted modulo Euclidean isometries. It must rule out flexible continuously many realizations when claiming finiteness or uniqueness, and must record whether reflected copies are identified.
Finite exhaustive calculations can settle individual sizes if genuinely complete, but cannot alone prove or disprove \(M(n)\to\infty\). Historical source checking, critical editorial review, protected publication and public mathematics rendering checks also remain.
Sources and claim boundaries
-
Maintained Erdős Problem #668, OPEN principal question and explicitly reported \(n=4\) exception.
-
GCL original-issue source report, exact \(n=4\) proof and distinctions.
-
Original Erdős [Er97f] and 2025 computational studies [EHSVZ25], [AMP25]: precise original pages/certificates not independently inspected.
Boundary: exact finite result \(M(4)=1\), without a result on asymptotic growth, independent full source review, Lean certification, protected publication, public MathJax readback or V2 admission.
Qualified editorial critical review: R10 source-exception dossier. The source-semantic exception is visible and its original research hold remains. Publication classification: REVIEWED EXPLANATORY ARTICLE WITH UNRESOLVED SOURCE/PROOF QUALIFICATIONS. Independent theorem proof or Lean certification: NOT CLAIMED.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1939 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: STATEMENT_CAPTURE_REQUIRED.
- 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.