Epsilon-Delta Definition of Limits

⚡ Key Takeaways

One-line statement: $\lim_{x \to c} f(x) = L$ means that for every $\varepsilon > 0$ there exists $\delta > 0$ such that $0 < |x - c| < \delta$ implies $|f(x) - L| < \varepsilon$ . Key result in words: A limit exists when the output of a function can be forced arbitrarily close to a target value by restricting the input sufficiently close to (but not equal to) a point. Where it applies: Real analysis, calculus, the rigorous foundation of continuity, derivatives, and integrals. Who/when: Bernard Bolzano (1817) and Augustin-Louis Cauchy (1820s) pioneered the idea; Karl Weierstrass (1860s) gave the modern static formulation.

💡 The Core Insight

A limit is a promise under challenge. An adversary picks a tolerance $\varepsilon$ , daring you to keep $f(x)$ within $\varepsilon$ of $L$ . You respond with a neighborhood radius $\delta$ . If you can always respond successfully — for every conceivable $\varepsilon$ , no matter how tiny — then the limit equals $L$ . The order of quantifiers is sacred: $\varepsilon$ comes first (the adversary moves first), and your $\delta$ is allowed to depend on $\varepsilon$ .

DefinitionEpsilon-Delta Definition of a Limit

Let $f$ be a real-valued function defined on some open interval containing $c$ , except possibly at $c$ itself. We say the limit of $f(x)$ as $x$ approaches $c$ is $L$ , written $\lim_{x \to c} f(x) = L$ , if:

1

\quad \forall \varepsilon > 0,\ \exists \delta > 0\ :\ \forall x,\ \bigl(0 < |x - c| < \delta \implies |f(x) - L| < \varepsilon\bigr)

Derivation hint: This is the direct symbolic transcription of "for every output tolerance, there is an input tolerance making the implication hold"; the quantifier order $\forall\varepsilon\,\exists\delta$ encodes that $\delta$ may depend on $\varepsilon$ .

Symbol glossary:

$f$ — the function under study, defined on a punctured neighborhood of $c$ .
$c$ — the limit point of the input ( $c$ need not be in the domain).
$L$ — the candidate limit value (the target output).
$\varepsilon$ (epsilon) — an arbitrary positive real measuring tolerance on the output.
$\delta$ (delta) — a positive real measuring tolerance on the input; chosen in response to $\varepsilon$ .
$|x - c|$ — distance from $x$ to $c$ ; the condition $0 < |x-c|$ excludes $x = c$ .
$|f(x) - L|$ — distance from the output to the target.

The two key inequalities defining the "tubes" are:

2

\quad 0 < |x - c| < \delta \quad\Longleftrightarrow\quad x \in (c - \delta,\, c) \cup (c,\, c + \delta)

Derivation hint: $|x-c| < \delta$ is the open interval of radius $\delta$ about $c$ ; the strict lower bound $0 < |x-c|$ removes the single point $x=c$ , leaving the punctured interval.

3

\quad |f(x) - L| < \varepsilon \quad\Longleftrightarrow\quad f(x) \in (L - \varepsilon,\, L + \varepsilon)

Derivation hint: This is the definition of absolute value as distance; $f(x)$ lies in the open horizontal band of half-width $\varepsilon$ centered at $L$ .

TheoremUniqueness of Limits

If $\lim_{x \to c} f(x) = L_1$ and $\lim_{x \to c} f(x) = L_2$ , then $L_1 = L_2$ . A function approaching a point can have at most one limit there.

ProofUniqueness of Limits

We argue by contradiction. Suppose $L_1 \neq L_2$ , and set

4

\quad \varepsilon = \frac{|L_1 - L_2|}{2} > 0

Derivation hint: Since $L_1 \neq L_2$ , the distance $|L_1 - L_2|$ is positive, so half of it is a legitimate positive $\varepsilon$ we may feed to both limit definitions.

Step 1. By $\lim_{x\to c} f(x) = L_1$ , there is $\delta_1 > 0$ with $0 < |x-c| < \delta_1 \implies |f(x) - L_1| < \varepsilon$ .

Step 2. By $\lim_{x\to c} f(x) = L_2$ , there is $\delta_2 > 0$ with $0 < |x-c| < \delta_2 \implies |f(x) - L_2| < \varepsilon$ .

Step 3. Let $\delta = \min(\delta_1, \delta_2)$ . Since $c$ is a limit point of the domain, choose any $x$ with $0 < |x - c| < \delta$ . Then both inequalities hold simultaneously.

Step 4. Apply the triangle inequality:

5

\quad |L_1 - L_2| \le |L_1 - f(x)| + |f(x) - L_2| < \varepsilon + \varepsilon = |L_1 - L_2|

Derivation hint: Insert and subtract $f(x)$ inside $|L_1 - L_2| = |(L_1 - f(x)) + (f(x) - L_2)|$ , then bound each term by $\varepsilon$ from Steps 1–2; the sum $2\varepsilon$ equals $|L_1 - L_2|$ by our choice in [4]. This yields $|L_1 - L_2| < |L_1 - L_2|$ , a contradiction. Hence $L_1 = L_2$ . $\blacksquare$

∎

mathematician · Karl Weierstrass | 1815–1897

Often called the "father of modern analysis," Weierstrass crystallized the epsilon-delta formulation in his Berlin lectures during the 1860s, though it never appeared in a single landmark publication of his — it spread through the notes of students such as Eduard Heine and Georg Cantor. He insisted on arithmetization: every concept of analysis should reduce to inequalities among real numbers, banishing the loose language of "infinitely small quantities" and "approaching motion." His static $\forall\varepsilon\,\exists\delta$ formulation made rigorous proofs possible and exposed pathologies — most famously his everywhere-continuous, nowhere-differentiable function (1872) — that earlier intuitive calculus could not even articulate.

🧠 Intuition

Picture the graph of $f$ near $x = c$ . Draw a horizontal band $(L - \varepsilon, L + \varepsilon)$ — this is the "target zone." The limit claim says: I can find a vertical strip $(c - \delta, c + \delta)$ (with the line $x=c$ itself poked out) so narrow that the entire graph over that strip stays trapped inside the horizontal band. The adversary keeps shrinking the band; you keep narrowing your strip in response. The limit exists and equals $L$ precisely when you can win this game indefinitely. The crucial mental shift is from dynamic to static. There is no "moving point" sliding toward $c$ . Instead, there is a logical statement: a guarantee that closeness of input forces closeness of output. The smaller $\varepsilon$ you must satisfy, the smaller $\delta$ you typically need — and the functional relationship $\delta(\varepsilon)$ encodes how "steeply" the function bunches up near $L$ .

Figure

The epsilon-delta tube: the graph over

(c-\delta, c+\delta)

stays inside the band

(L-\varepsilon, L+\varepsilon)

ExampleA linear function:

\lim_{x \to 3}(2x + 1) = 7

Given: $f(x) = 2x + 1$ , $c = 3$ , claimed limit $L = 7$ .

Goal: For arbitrary $\varepsilon > 0$ , produce $\delta > 0$ making the implication in [1] true.

Step 1 — Analyze the target. We need $|f(x) - 7| < \varepsilon$ , i.e.

6

\quad |(2x + 1) - 7| = |2x - 6| = 2|x - 3| < \varepsilon

Derivation hint: Substitute $f(x)=2x+1$ and $L=7$ , simplify the difference, and factor out the coefficient $2$ to reveal a multiple of $|x-3|$ .

Step 2 — Solve for the input constraint. From [6], $2|x-3| < \varepsilon \iff |x - 3| < \dfrac{\varepsilon}{2}$ .

Step 3 — Choose $\delta$ . Let $\delta = \dfrac{\varepsilon}{2}$ .

Step 4 — Verify. If $0 < |x - 3| < \delta = \varepsilon/2$ , then $|f(x) - 7| = 2|x-3| < 2\cdot \tfrac{\varepsilon}{2} = \varepsilon$ . ✓

\boxed{\ \delta = \frac{\varepsilon}{2}\ \text{(dimensionless ratio of tolerances)}\ }

ExampleA quadratic:

\lim_{x \to 2} x^2 = 4

Given: $f(x) = x^2$ , $c = 2$ , $L = 4$ .

Step 1 — Factor the output difference.

7

\quad |x^2 - 4| = |x - 2|\,|x + 2|

Derivation hint: Use the difference of squares $x^2 - 4 = (x-2)(x+2)$ and the multiplicativity of absolute value $|ab| = |a||b|$ .

Step 2 — Tame the wild factor. The term $|x+2|$ varies with $x$ , so we restrict in advance. Insist $\delta \le 1$ , so $|x - 2| < 1 \implies 1 < x < 3 \implies |x + 2| < 5$ .

Step 3 — Bound the product. Under that restriction,

8

\quad |x^2 - 4| = |x-2|\,|x+2| < 5\,|x - 2|

Derivation hint: Replace the bounded factor $|x+2|$ by its upper bound $5$ (valid only because we pre-restricted $\delta\le 1$ ), turning the product into a constant multiple of $|x-2|$ .

Step 4 — Force below $\varepsilon$ . We need $5|x-2| < \varepsilon$ , i.e. $|x-2| < \varepsilon/5$ . Combine both requirements:

9

\quad \delta = \min\!\left(1,\ \frac{\varepsilon}{5}\right)

Derivation hint: Take the minimum so that both the pre-restriction ( $\delta\le1$ ) used in Step 2 and the final tolerance ( $\delta\le\varepsilon/5$ ) from Step 4 hold simultaneously.

Step 5 — Verify. If $0 < |x-2| < \delta$ , then $\delta \le 1$ gives $|x+2| < 5$ , and $\delta \le \varepsilon/5$ gives $|x^2 - 4| < 5\cdot \tfrac{\varepsilon}{5} = \varepsilon$ . ✓

\boxed{\ \delta = \min\!\left(1,\ \tfrac{\varepsilon}{5}\right)\ }

⚠️ Order of quantifiers is non-negotiable

The statement $\forall \varepsilon\, \exists \delta$ is completely different from $\exists \delta\, \forall \varepsilon$ . Swapping them produces a stronger, usually false condition: it would demand a single $\delta$ working for all $\varepsilon$ , forcing $f$ to be constant near $c$ . Always read left to right: the adversary picks $\varepsilon$ first, then you pick $\delta$ in response. Your $\delta$ legitimately depends on $\varepsilon$ (and on $c$ ).

⚠️ The value

f(c)

is irrelevant to the limit

The clause $0 < |x - c|$ deliberately excludes $x = c$ . The limit describes behavior in the punctured neighborhood. A function can have $\lim_{x\to c}f(x) = L$ while $f(c)$ is undefined, or equals something entirely different from $L$ . Equality $\lim_{x\to c}f(x) = f(c)$ is an extra condition — it is precisely the definition of Continuity.

CounterexampleA function with no limit:

\sin(1/x)

at

0

Consider $f(x) = \sin(1/x)$ for $x \neq 0$ . As $x \to 0$ , the argument $1/x$ blows up, so $f$ oscillates between $-1$ and $1$ infinitely often in every punctured neighborhood of $0$ . No candidate $L$ works: fix $\varepsilon = 1/2$ ; every $\delta$ -interval $(0,\delta)$ contains points where $f = 1$ and points where $f = -1$ , so the outputs cannot all lie within $1/2$ of any single $L$ .

10

\quad f\!\left(\frac{1}{2\pi n + \pi/2}\right) = 1, \qquad f\!\left(\frac{1}{2\pi n - \pi/2}\right) = -1, \quad \text{for arbitrarily large } n

Derivation hint: Set $1/x = 2\pi n \pm \pi/2$ where $\sin$ attains $\pm1$ ; solving for $x$ gives points arbitrarily close to $0$ as $n\to\infty$ , so both values $\pm1$ occur in every $(0,\delta)$ . Hence $\lim_{x\to 0}\sin(1/x)$ does not exist, even though $f$ is bounded and continuous everywhere except $0$ .

connection

The epsilon-delta definition is the seed from which much of analysis grows.

<a class="wiki-link" href="/wiki/ndli-isc-12-mathematics-continuity">Continuity</a> is the special case $L = f(c)$ : $f$ is continuous at $c$ iff $\lim_{x\to c}f(x) = f(c)$ , removing the puncture.
<a class="wiki-link" href="/wiki/math-analysis-one-sided-limits">One-Sided Limits</a> restrict $x$ to $c < x < c+\delta$ (right) or $c-\delta < x < c$ (left); the two-sided limit exists iff both one-sided limits exist and agree.
Sequential Criterion for Limits (Heine's theorem) shows $\lim_{x\to c}f(x)=L$ iff $f(x_n)\to L$ for every sequence $x_n \to c$ with $x_n \neq c$ , bridging function limits and Limits of Sequences.
The Derivative is itself a limit, $f'(c) = \lim_{h\to 0}\frac{f(c+h)-f(c)}{h}$ , so every differentiation argument silently invokes epsilon-delta.
Generalizing $|x-c|$ to a metric $d(x,c)$ yields the definition of limits in <a class="wiki-link" href="/wiki/math-topgeo-metric-spaces">Metric Spaces</a>, and replacing balls with open sets gives Topological Limits.

🔬 Where this appears

⏳ History

- c. 250 BCE — Archimedes uses the method of exhaustion, bounding areas between approximations differing by less than any prescribed amount — a proto-epsilon argument.

- 1660s–1680s — Newton and Leibniz develop calculus using fluxions and infinitesimals; powerful but logically unfounded ("ghosts of departed quantities," per Berkeley, 1734).

- 1734 — Bishop George Berkeley publishes The Analyst, attacking the incoherence of infinitesimals and demanding rigor.

- 1817 — Bernard Bolzano, in Rein analytischer Beweis, gives an essentially modern definition of continuity using inequalities, anticipating epsilon-delta but largely unread.

- 1821 — Cauchy's Cours d'analyse defines limits and continuity with the language "as small as one wishes," moving toward rigor but still partly verbal/dynamic.

- 1823 — Cauchy uses explicit

\varepsilon

and

\delta

symbols in his lectures on the calculus, the first appearance of the now-iconic letters.

- 1860s — Weierstrass, in Berlin lectures, delivers the fully static, quantifier-explicit

\forall\varepsilon\,\exists\delta

definition; disseminated via student notes (Heine 1872).

- 1872 — Heine publishes the modern definition in print and proves uniform continuity results; Weierstrass exhibits a continuous nowhere-differentiable function.

- 1960s — Abraham Robinson's non-standard analysis rehabilitates infinitesimals on rigorous logical footing, showing the epsilon-delta approach is one of two equivalent rigorous foundations.

open-problem · Beyond the standard definition

Epsilon-delta is not the end of the story but a gateway to subtle distinctions and ongoing pedagogical and foundational questions.

Pointwise vs. uniform. The definition gives pointwise limits; demanding a single $\delta(\varepsilon)$ valid across an entire domain yields uniform continuity and uniform convergence, whose failure ( $\delta$ shrinking unboundedly as $c$ varies) drives deep phenomena like the non-interchangeability of limits and integrals.
Choice of foundation. Whether limits are "really" about epsilon-delta (Weierstrass) or genuine infinitesimals (Robinson's non-standard analysis) is a live philosophical and pedagogical debate; the two are provably equivalent yet cognitively distinct.
Constructive content. In constructive/computable analysis, one demands an explicit, computable function $\delta(\varepsilon)$ — a modulus of continuity. Not every classically continuous function admits a computable modulus, raising questions at the analysis–logic interface.
Pedagogy. Empirical mathematics-education research repeatedly documents that the nested-quantifier structure is among the hardest conceptual hurdles undergraduates face; the optimal way to teach it remains genuinely unsettled.

Formal Statement

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also