Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-generatedprecheck passjudge pass (z-ai/glm-5.2)audited 2026-07-27
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

If f≤g≤h near c and f and h have the same limit at c, then so does g

Statement

Let A⊆R, let c be a limit point of A (Limit point, isolated point, adherent point, derived set, and dense subset of R) and let f,g,h:A→R. Suppose there is a real η>0 with

f(x)≤g(x)≤h(x)for every x∈A with 0<∣x−c∣<η,

and suppose the limits of f and of h at c exist and are equal, say lim⁡x→cf(x)=lim⁡x→ch(x)=L (The ε-δ limit lim⁡x→cf(x)=L of f:A→R at a limit point c of A). Then the limit of g at c exists, and

lim⁡x→cg(x)  =  lim⁡x→cf(x)  =  lim⁡x→ch(x)  =  L.

This is the one result on this page that produces a limit rather than computing one. No hypothesis whatever is placed on g beyond the two inequalities: g may be wildly irregular, as x↦x ψ(1/x) on the companion page is, and the theorem still delivers its limit at c.

The proof is a direct ε-δ argument and uses no choice principle.

Facts & Assumptions

Given: A set A⊆R, a limit point c of A, functions f,g,h:A→R, a real η>0 with f(x)≤g(x)≤h(x) for every x∈A satisfying 0<∣x−c∣<η, and a real L with lim⁡x→cf(x)=L and lim⁡x→ch(x)=L (The ε-δ limit lim⁡x→cf(x)=L of f:A→R at a limit point c of A, Limit point, isolated point, adherent point, derived set, and dense subset of R).

[L1]

The limit condition: for every real ε>0 there is a real δ>0 such that every x∈A with 0<∣x−c∣<δ satisfies ∣f(x)−L∣<ε, and likewise for h (The ε-δ limit lim⁡x→cf(x)=L of f:A→R at a limit point c of A).

[L2]

Absolute value: for t>0, ∣u∣<t is equivalent to −t<u<t (Basic properties of the absolute value).

[L3]

Order arithmetic in R: the order is transitive, and mixed chains compose, so u<v≤w gives u<w and u≤v<w gives u<w; adding a constant to an inequality (Order is preserved by adding a constant and by adding inequalities); of finitely many positive reals the smallest is positive, the order being total (Ordered field). Order is preserved by adding a constant and by adding inequalities states its moves in their STRICT forms only; the non-strict forms used below follow by adjoining the equality case, in which the two sides coincide, the order being total (Ordered field).

[L4]

Neighbourhoods: Nδ∗(c)={ y:0<∣y−c∣<δ }, and a smaller radius gives a smaller punctured neighbourhood (The ε-neighbourhood and the punctured ε-neighbourhood of a point of R).

Proof

technique · direct
1.1

Let ε>0 be an arbitrary real. By [L1] fix reals δ1,δ2>0 such that every x∈A with 0<∣x−c∣<δ1 satisfies ∣f(x)−L∣<ε and every x∈A with 0<∣x−c∣<δ2 satisfies ∣h(x)−L∣<ε; let δ be the smallest of δ1, δ2 and η, so δ>0.

L1L3L4choose
2.1

Let x∈A with 0<∣x−c∣<δ. Then 0<∣x−c∣<δ1 gives L−ε<f(x), and 0<∣x−c∣<δ2 gives h(x)<L+ε, while 0<∣x−c∣<η gives f(x)≤g(x)≤h(x).

step 1.1L2L3L4
3.1

Chaining those four inequalities, L−ε<f(x)≤g(x)≤h(x)<L+ε, hence L−ε<g(x)<L+ε, that is −ε<g(x)−L<ε, that is ∣g(x)−L∣<ε.

step 2.1L2L3
4.1

So for every real ε>0 a real δ>0 has been produced with ∣g(x)−L∣<ε for every x∈A satisfying 0<∣x−c∣<δ: the limit of g at c exists and equals L.

step 3.1L1∎

Remarks

  • Where the three hypotheses are spent. The inequality f≤g is used only for the lower estimate and g≤h only for the upper one; the equality of the two outer limits is what makes the two estimates close on the same number L. Drop it and the argument gives only lim⁡f≤lim inf⁡-style information, which this page does not develop.

  • The order hypothesis is local. It is imposed only on A∩Nη∗(c), so the theorem is insensitive to the behaviour of the three functions far from c, and to their values at c; that is The limit at c depends only on the restriction of f to a punctured neighbourhood of c, and passes to any subset of the domain having c as a limit point in action.

  • Typical use. To prove that a bounded oscillating factor is killed by a factor tending to 0: if ∣u(x)∣≤B near c then −B∣x−c∣≤(x−c)u(x)≤B∣x−c∣ near c, and both outer functions tend to 0. That is exactly how x ψ(1/x)→0 is proved on the companion page.

  • The sequential analogue is The squeeze theorem.

Depends on

Used by

Dependency tree · two levels

17 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources