EVOPEDIA

← all systems · see it in The System

Limit (mathematics): the value a sequence or function approaches, and the practice of approaching, bounding and proving it

In mathematics, a limit is the value that a function (or sequence) approaches as the argument (or index) approaches some value.

Properties published scienceour proposal

Sourced fields, and our own readings.

aliasesLimits
born atlate 5th century BC (Antiphon) to 1647 (Saint-Vincent), contested
born precisioncentury to year; contested (C-2658)
boundaryInside: the practice of approaching a value by an endless process and bounding the error (polygons, tables, iterates, difference quotients; reading, writing, saying, gesturing and computing it; teaching it) and the structure that practice works on (the epsilon-N and epsilon-delta definitions, uniqueness, one-sided and infinite limits, Cauchy sequences, convergence in metric and topological spaces, the standard-part reading). Outside: what the limit is used to define (continuity, derivative, integral, series sums: their own walks), the numbers it lands in (real numbers), the infinite it handles (infinity), and the physical systems it stands in for.
nested levelA part inside mathematics (a concept analysis is built on); used inside calculus to define its derivative and integral; runs on writing, on computers and on people, who teach, learn and extend it.
scalesocial

Measures (17) published science

characteristic clock: from Gregoire de Saint-Vincent's terminus (1647) to Weierstrass's epsilon-delta form (1861)214 years (derived from two dates the sources give)
clock: from Eudoxus's rigorous exhaustion (c. 408-355 BC) to Weierstrass (1861)about 2,200 years (derived; Eudoxus's dates are approximate)
response lag: The Analyst (1734) to Maclaurin's Treatise of Fluxions (1742)8 years (derived)
Bolzano's work unknown to other mathematicians after his death30 years
Archimedes' bounds on pi from 96-sided polygons3 + 10/71 < pi < 3 + 10/70 (a range of 1/497) dimensionless
terms discarded to get within 0.001 of 1/3 with the decimals 0, 0.3, 0.33, ...3 terms
Basel sum, a limit of partial sumspi^2/6 dimensionless
Newton's method near a simple rootorder at least 2: correct digits roughly double per step order of convergence
machine epsilon, double precisionabout 2.2e-16 relative error
step size when nothing is known of fsqrt(eps) x (about 1.5e-8 x in double precision, derived) step
quantifiers saved by the hyperreal definition of limit (Keisler)2 quantifiers
freshmen tested on the basic limit theorem251 (17 female), three cohorts students
invalid-to-valid pathway ratio among correct answersabout 2 to 1 ratio
cued dialogues needed to fix a limitation7.7-9.3 question-response dialogues (average)
GeoGebra study pretest means (control, experimental)4.23 (SD 1.14), 4.25 (SD 1.15); t = .059, df = 48 score
citations to Misra and Sudarshan 1977 (quantum Zeno) by October 2006about 535 citations
arrow-below notation: invented and popularised1905 (Leathem), 1908 (Hardy) year

Open contentions (9) open contention

Where sources or readings disagree. Both stay on the map.

Partners and parts published science

Built from / builds (9)

Sits inside / holds (3)

Trades with (valence) (16)

expressed by (9)

floor filled by (14)

Kinds (6)

Notes published science

Layers

The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.

a limit: the value a sequence or function approaches, met as a process, a table, a graph, words or a written lim, and fixed by an epsilon-style definition (reader's view)

  1. L1: An approach in the world or on the page: polygons filling a circle, a runner closing a gap, the decimals 0.3, 0.33, 0.333 closing on 1/3, a function at points nearer and nearer to c.
  2. L2: Seeing or hearing the approach: a graph with a moving point, a table, the words "approaches" and "tends to".
  3. L3: Codes: lim, epsilon, delta, N, the arrow; held by a number-and-space network in experts rather than by language areas.
  4. L4: Approaching and bounding: shrink the gap, bound the error, discard finitely many terms; or step h on a machine.
  5. L5: Learning and control: building the concept (APOS), choosing heuristics, fixing misconceptions with cues.
  6. L6: The definition: epsilon-N and epsilon-delta, uniqueness, Cauchy sequences and completeness, the standard part.
  7. L7: Use: continuity, derivative, integral, series, closed sets, numerical analysis, limits of physical theories.

the boardroom (shared floors)

  1. L3: Codes: the limit held as notation (lim, epsilon, delta, N), a table, a graph or words. In professional mathematicians, analysis statements, like algebra, geometry and topology, recruit a bilateral prefrontal, parietal and inferior temporal network also used for numbers, with minimal use of language areas. Cauchy wrote epsilon for error but an infinitesimal alpha in his continuity definition.
  2. L4: Approaching: shrink the distance and bound the error. Pick epsilon, find N (or delta), discard the finitely many terms outside; simplify before substituting ((x^2 - 1)/(x - 1) = x + 1 away from 1); squeeze between bounds; subdivide an interval and keep the half with a sign change (Cauchy's IVT proof); on a machine, a difference quotient at step h. Thought as fictive motion; or, in Robinson's reading, two stages: evaluate at an infinite index, take the standard part.
  3. L5: Learning and control: the concept built from actions to processes to objects to schemas (APOS); choosing the heuristic and applying it; misconceptions from prior knowledge (about two invalid pathways for each valid one among 251 freshmen); feeling (enjoyment when made visible). Which conception helps is contested.
  4. L6: The structure: for every epsilon > 0 there is N (or delta) with every later term (or nearby value) within epsilon; unique on the reals and in Hausdorff spaces; one-sided and infinite limits; Cauchy sequences, which converge in the reals but not in the rationals (completeness); pointwise and uniform limits; the equivalent sequential definition; the standard part in nonstandard analysis (C-997).
  5. L7: Use: continuity, the derivative, the integral and series defined by limits; closed sets by limit points; long-term behaviour by limit sets; numerical analysis by order of convergence; physics by limits of theories (thermodynamic limit, ray optics); computability by the limit lemma.

way in: A machine's channel: the limit taken with finite-precision numbers

  1. L1: A machine cannot let h reach 0: numbers are stored with a relative rounding error of about the machine epsilon (2.2e-16 in double precision), so x + h is rounded and (x + h) - x need not equal h.
  2. L2: The step chosen: h about sqrt(eps) times x, small without a large rounding error.
  3. L3: The difference quotient computed: both cardinal sins committed, adding a small number to a big one and subtracting nearly equal numbers.

way in: Emotion: the feel of learning the limit (enjoyment, frustration, the critic's scorn)

  1. L1: Learners report learning the limit as interesting, motivating and enjoyable when it is made visible; others name what held them back.

way in: Hearing: the words of approach ("approaches", "tends to", "gets close to")

  1. L1: The limit read aloud: "the limit of f of x as x approaches c equals L"; "f of x tends to L as x tends to c". The phrases carry a motion the formal definition does not.

way in: Sight: a graph with a moving point (the graphics view)

  1. L1: The graphs of f, |x - 2| and |3x + 1 - 7| drawn together, with points moved by a slider.
  2. L2: Seeing the two distances shrink together: students watch the points and graphs as the slider moves.

way in: Sight: a table of values closing in (f(0.9), f(0.99), f(0.999) ...)

  1. L1: A table of f(x) = (x^2 - 1)/(x - 1) at 0.9, 0.99, 0.999 and 1.001, 1.01, 1.1: 1.900 ... 2.100, undefined at 1; or the decimals 0, 0.3, 0.33, 0.333 closing on 1/3.

way in: Sight: the written limit (lim with the arrow below, epsilon, delta, N)

  1. L1: The notation: lim with x -> c written below, or f(x) -> L as x -> c; the arrow-below form from Leathem (1905) and Hardy (1908).

way in: The world approached: shapes filled by polygons, a runner closing a gap, a rate at an instant

  1. L1: Processes that never finish in finitely many steps: polygons filling a circle (exhaustion), Achilles closing on the tortoise (Zeno), the rate of change at an instant.
  2. L2: The paradox as taken in: an infinite number of steps seems to stand between the runner and the goal.

way in: Thought: the approach held in mind (potential infinity, fictive motion)

  1. L1: An endless process held as a thought: Aristotle's potential rather than actual infinity; experts think limits and continuity as motion, even after static definitions. Where a thought comes from is contested (C-413).
  2. L2: The motion simulated: a fictive-motion metaphor brings the motor system in to ground the concept.

way in: Touch and movement: dragging a slider in dynamic software

  1. L1: The learner moves x toward 2 with a slider and watches f(x) move toward 7; entering functions by keyboard takes skill some learners lack.

way out: Computing: the limit handed to a machine (iterate, difference quotient, automatic differentiation)

  1. L1: The numbers produced, read back as a value or a graph.
  2. L4: An iteration or a difference quotient programmed; or the derivative computed exactly by automatic differentiation.

way out: Modelling the world: a limit put in place of a large or fast system

  1. L1: The prediction about the world: a gas's average internal energy from its pressure and temperature.
  2. L7: A theory taken to a limit (N -> infinity) to deduce novel, robust behaviour.

way out: Speech and gesture: explaining the limit aloud, with hands that move

  1. L1: The words and the gestures: mathematicians gesture motion while proving.
  2. L5: The teacher or tutor chooses what to say and which cue to give (or skips the link to the intuitive notion).

way out: Writing: the limit and its epsilon proof on paper or board

  1. L1: The marks: lim with the arrow below, read back by the eye.
  2. L4: The bound worked out on the page: |3x + 1 - 7| = 3|x - 2|.
  3. L6: The definition stated: for every epsilon, an N.

Sources (29) published science

Limit (mathematics) (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01)
Newton's method (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Nigus WH, Sbhatu DB, Tesfay BG (2023) Freshman STEM students' misconceptions in a basic limit theorem of lim x→0 (sin x/x). Heliyon 9(12):e22359. doi:10.1016/j.heliyon.2023.e22359 (CC BY-NC-ND, PMC10750035) (behind its links)
Gebeyehu BM, Atnafu AM, Ejigu WT (2021) Implementing GeoGebra integrated with multi-teaching approaches guided by the APOS theory to enhance students' conceptual understanding of limit in Ethiopian Universities. Heliyon 7:e07012. doi:10.1016/j.heliyon.2021.e07012 (PMC8141776) (behind its links)
The Analyst (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Bair J, Blaszczyk P, Ely R, Henry V, Kanovei V, Katz KU, Katz MG, et al. (2017) Cauchy, infinitesimals and ghosts of departed quantifiers. arXiv:1712.00226 (behind its links)
Calculus (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Series (mathematics) (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Limit of a function (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01; the (ε, δ)-definition article redirects here) (behind its links)
History of calculus (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Method of exhaustion (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Numerical differentiation (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Itano WM (2006) Perspectives on the quantum Zeno paradox. arXiv:quant-ph/0612187 (abstract) (behind its links)
Thermodynamic limit (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Limit of a sequence (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Amalric M, Dehaene S (2016) Origins of the brain networks for advanced mathematics in expert mathematicians. PNAS 113:4909-4917. doi:10.1073/pnas.1603205113 (behind its links)
Baydin AG, Pearlmutter BA, Radul AA, Siskind JM (2018) Automatic differentiation in machine learning: a survey. JMLR 18(153). arXiv:1502.05767 (behind its links)
Blaszczyk P, Katz MG, Sherry D (2012) Ten misconceptions from the history of analysis and their debunking. Foundations of Science; arXiv:1202.4153 (behind its links)
Real number (Wikipedia) (behind its links)
Infinity (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Zeno's paradoxes (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Marghetis T, Nunez R (2013) The motion behind the symbols: a vital role for dynamism in the conceptualization of limits and continuity in expert mathematics. Top Cogn Sci 5(2):299-316. doi:10.1111/tops.12013 (PubMed abstract; full text not reached) (behind its links)
Khatin-Zadeh O, Farsani D, Eskandari Z, et al. (2022) The roles of motion, gesture, and embodied action in the processing of mathematical concepts. Front Psychol. doi:10.3389/fpsyg.2022.969341 (CC BY, PMC9616004) (behind its links)
Butterfield J (2011) Less is different: emergence and reduction reconciled. Foundations of Physics 41:1065-1135 (preprint arXiv:1106.0702) (behind its links)
Continuous function (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Nonstandard analysis (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Katz KU, Katz MG (2011) Cauchy's continuum. arXiv:1108.4201 (abstract) (behind its links)
Bottazzi E, Katz MG (2021) Infinitesimals via Cauchy sequences: refining the classical equivalence. arXiv:2106.00229 (abstract) (behind its links)
Bascelli T, Blaszczyk P, Borovik A, Kanovei V, et al. (2017) Cauchy's infinitesimals, his sum theorem, and foundational paradigms. arXiv:1704.07723 (abstract) (behind its links)