Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passaudited 2026-10-02
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.

Residues of exact differentials vanish

Statement

Assume the Axiom of Choice as inherited from the completion and coefficient-field/residue suppliers. Let k be a field, C a smooth integral curve over k, and p a closed point whose residue field κ(p) is finite separable over k. For every f∈k(C), the exact differential df has residue zero at p: res⁡p(df)=0. Consequently the residue functional kills the image of d ⁣:k(C)→Ωk(C)/k1, and adding an exact differential does not change the residue at any point where the coefficient-trace residue is defined. This holds in every characteristic.

Facts & Assumptions

Given: a field k, a smooth integral curve C over k, a closed point p with κ(p) finite separable over k, the local coefficient-trace residue of Residue of a rational differential at a separable closed point at p, and a rational function f∈K=k(C).

[F1]

The local construction of Residue of a rational differential at a separable closed point applies at p and uses only the local ring OC,p: choosing a uniformizer t of the discrete valuation ring OC,p gives a k-compatible coefficient field κ(p)↪O^C,p and an isomorphism O^C,p≅κ(p)[ ⁣[t] ⁣], so that K=Frac⁡(OC,p) embeds in κ(p)((t)) and every rational differential ω is written ω=a dt with a∈K whose Laurent expansion in κ(p)((t)) has a residue of its −1-coefficient; the residue is res⁡p(ω)=Tr⁡κ(p)/k([t−1]a)∈k, it is k-linear in ω, and it depends only on the local data at p (Residue of a rational differential at a separable closed point, Local rings at closed points of smooth curves are discrete valuation rings).

[F2]

The differential dt is a K-basis of ΩK/k1=Ωk(C)/k1, so every rational differential has a unique expression ω=a dt with a∈K (A uniformizer differential generates the module of differentials).

[F3]

Formal calculus in κ(p)((t)) (Formal Laurent series K((x)), their order, derivative, and residue): the formal derivative D is defined coefficientwise by D(tn)=ntn−1, so D(∑nantn)=∑nnantn−1 with no division by integers; the formal residue is res⁡t(g)=[t−1]g. The coefficient field κ(p) is embedded compatibly with k by Residue of a rational differential at a separable closed point, so D kills the image of k and the composition D∘jt is a k-derivation of K into κ(p)((t)), where jt:K↪κ(p)((t)) is the completion embedding. This is the derivation passed through the universal property in [F6]; no differential-module identification for the completed field is asserted.

[F4]

The residue res⁡p(ω) of [F1] does not depend on the choice of uniformizer t (The residue is independent of the uniformizer).

[F5]

For the finite extension κ(p)/k the field trace Tr⁡κ(p)/k ⁣:κ(p)→k is k-linear and Tr⁡κ(p)/k(0)=0 (The norm NK/F and trace Tr⁡K/F of a finite field extension).

[F6]

The universal property of algebraic Kähler differentials (Derivations are maps out of Ω): for a ring map k→K and a K-module M, every k-derivation δ:K→M is uniquely of the form Δ∘d for a K-linear map Δ:ΩK/k1→M.

[F7]

The Axiom of Choice is The Axiom of Choice.

Proof

Proof technique: direct; expand f as a formal Laurent series in a uniformizer and compute the −1-coefficient of its formal derivative.

1.1F1F2given

(Set-up.) Let t be a uniformizer of OC,p, let jt:K↪κ(p)((t)) be the completion embedding, and write jt(f)=∑n≥Nantn. By [F2] the exact differential is df=a dt for a unique a∈K, and by [F1] its residue is res⁡p(df)=Tr⁡κ(p)/k([t−1]jt(a)).

2.1F2F3F6step 1.1

(The derivative is formal through the universal property.) Give M=κ(p)((t)) its K-module structure through jt. By [F3], δt:=D∘jt:K→M is a k-derivation; [F6] therefore gives a unique K-linear map Δt:ΩK/k1→M with Δt(dg)=D(jt(g)) for every g∈K. Since jt(t)=t, one has Δt(dt)=1. By [F2], df=a dt, so Δt(df)=jt(a) and also Δt(df)=D(jt(f)). Thus the Laurent expansion of the algebraic coefficient a is exactly D(jt(f)). This comparison uses only the universal property for ΩK/k1 and makes no claim that Ωκ(p)((t))/k1=κ(p)((t)) dt.

3.1F3step 2.1algebra

(The critical coefficient.) The coefficient of t−1 in D(jt(f))=∑nnantn−1 receives a contribution only from n=0, and that contribution is 0⋅a0=0; every other index contributes to a different power. By step 2.1, [t−1]jt(a)=0.

4.1F1F5step 1.1step 3.1

(Residue of the exact differential.) Applying the trace of [F1] to the vanishing coefficient just found and using [F5], res⁡p(df)=Tr⁡κ(p)/k(0)=0; this is the asserted vanishing at p.

5.1F3F4step 4.1

(Independence of the uniformizer, all characteristics.) The vanishing just proved was computed in the uniformizer t, and by [F4] the residue does not depend on that choice, so res⁡p(df)=0 for every uniformizer of OC,p; the computation of the formal derivative used the integer coefficients n literally, with no division by an integer at any point, so it is valid in every characteristic, including characteristics dividing an exponent occurring in f.

6.1F1F7step 4.1step 5.1∎

(Consequences.) Since f∈K was arbitrary, the residue functional kills the image of d ⁣:K→ΩK/k1; if a rational differential is changed by an exact differential, ω′=ω+df, then res⁡p(ω′)=res⁡p(ω)+res⁡p(df)=res⁡p(ω) by the k-linearity of res⁡p in [F1]. This includes f=0 and f∈k, for which df=0. The Axiom of Choice [F7] is inherited from the completion and power-series suppliers used in [F1] and [F3].

Depends on

Used by

Dependency tree · two levels

53 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