The edge of a series need not be the edge of its function
The geometric series one plus z plus z squared and so on converges when the absolute value of z is less than one. Its function, 1/(1 minus z), still makes sense beyond much of that circle; only z equal to one is singular. Analytic continuation means extending an analytic function through overlapping neighborhoods. Fabry gives conditions under which no boundary point permits such an extension: the circle becomes a natural boundary.
Missing terms alone do not build a wall
Consider one plus z squared plus z to the fourth and so on. Infinitely many powers are missing, yet its function is 1/(1 minus z squared). Only plus one and minus one are singular on the unit circle. This classroom identity shows why simply counting omitted coefficients is not enough. A useful theorem needs quantitative conditions on where nonzero coefficients occur and how large they remain relative to the convergence radius.
The original sparsity condition has a scale
One explicit condition in Fabry 1896 paper concerns increasing integer exponents c_v. The gap to the next exponent, divided by log c_v, must grow without bound, with the power series having finite positive radius and the other stated coefficient conditions. For exponents n squared, the successive gap is two n plus one, which outgrows log of n squared. The ratio, rather than an isolated large gap, is the relevant long-run comparison.
A singularity is not just a failed sum
A natural boundary forbids analytic continuation through every point on the circle. It does not merely say that a chosen partial-sum calculation becomes unstable, and it need not mean the series diverges at every boundary point. Fabry develops coefficient and sign-change arguments building on Hadamard. The paper ends with exploratory discussion of general coefficients; that discussion is not a proved probability-one statement or a theorem about every arbitrary series.
Make your agent test the easy counterexample first
Ask it to derive the two geometric-series identities and mark their singular points. Then compare exponent gaps for n, n squared and two to the nth power, clearly separating finite computed ratios from asymptotic reasoning. The task is to understand the conditions, not to announce a new analytic-continuation theorem from a plot. A later result with the same historical name must still be checked against the exact assumptions of the cited original.
Scientific scope
Infinitely many missing terms alone are not enough. The stated convergence-radius, coefficient and local sparsity assumptions matter; later formulations named Fabry's gap theorem are not all attributed to this original statement. The author's closing discussion of what happens for general arbitrary coefficients is explicitly exploratory, not a probability-one or category theorem. A natural boundary concerns analytic continuation, not necessarily divergence at every boundary point.
Read the original work
Sources and reading locations for this explanation
- Sur les points singuliers d’une fonction donnée par son développement en série (original1896)
1896paper,printed367–382for radius/sign-change setup;391explicitlog-gapcriterion;398–399exploratoryclosingdiscussion.
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Eugène Fabry
Original-paper scientific authors for this exact mathematical contribution. Independent papers grouped in one historical family are not a joint byline. Earlier theorems, supporting proof contributions and later generalizations remain separate. Individual effort shares, author nationality and unresolved cross-record identities are not inferred.
Research locations and affiliations
Documented mapping · Included in country comparison
Documented countries / regions: France
Original byline names the Faculty of Sciences of Montpellier. Present-day location labels follow the retained platform convention; they do not assert nationality,modern-state ownership or territorial jurisdiction at the historical date.