Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

Why this matters

A courier charges £3 for a parcel of up to 2 kg, and £5 for a parcel over 2 kg and up to 10 kg. For a parcel of ww kilograms, where 0<w≤100 < w \le 10, the price in pounds is

P(w)={3if 0<w≤2,5if 2<w≤10.P(w) = \begin{cases} 3 & \text{if } 0 < w \le 2,\\ 5 & \text{if } 2 < w \le 10. \end{cases}

What does the price do near 2 kg? A parcel of 1.999 kg costs £3 and one of 2.001 kg costs £5, and however close to 2 kg the two weights are, the two prices stay £2 apart. So P(w)P(w) approaches no single number as ww approaches 2: lim⁡w→2P(w)\lim_{w \to 2} P(w) does not exist, as we prove in Example 2. But “does not exist” throws away what we know. Just below 2 kg the price is £3, and just above it is £5. We would like to say that P(w)P(w) approaches 3 as ww approaches 2 from the left, approaches 5 from the right, and so jumps by £2 at 2 kg.

A second situation has the same cure. The definition of the limit asks for ff to be defined on an open interval around aa, except possibly at aa. The square root x\sqrt{x} is defined for no x<0x < 0, because a negative number has no real square root (the square-root remark), so that definition says nothing about x\sqrt{x} at 0. Yet x\sqrt{x} is as small as we like when x>0x > 0 is small enough: 0.0001=0.01\sqrt{0.0001} = 0.01.

A one-sided limit looks at one side of aa only. It describes the jump of the price, and the square root at 0, and it gives a practical test for whether a limit exists: look at the two sides separately, then compare.

Limits from one side

Informally, lim⁡x→a+f(x)=L\lim_{x \to a^{+}} f(x) = L means that f(x)f(x) is as close to LL as we like for all xx close enough to aa and greater than aa. The window of the ε–δ game becomes the right half of the window of the definition of the limit: the punctured neighbourhood 0<∣x−a∣<δ0 < \abs{x - a} < \delta is the set (a−δ,a)∪(a,a+δ)(a - \delta, a) \cup (a, a + \delta), by part (e) of the proposition on distance inequalities, and a one-sided limit keeps only one of the two intervals.

In words. The right-hand limit is the definition of the limit with its window 0<∣x−a∣<δ0 < \abs{x - a} < \delta replaced by the right half, a<x<a+δa < x < a + \delta; the left-hand limit keeps the left half, a−δ<x<aa - \delta < x < a. The game is the same: a challenger names ε>0\eps > 0, you reply with δ>0\delta > 0, and you win the round if every xx in your half-window has f(x)f(x) within ε\eps of LL. Three points to notice:

Example. Let g(x)=xg(x) = x for x≤0x \le 0 and g(x)=x+1g(x) = x + 1 for x>0x > 0. Then lim⁡x→0+g(x)=1\lim_{x \to 0^{+}} g(x) = 1. Given ε>0\eps > 0, take δ=ε\delta = \eps. If 0<x<δ0 < x < \delta, then g(x)=x+1g(x) = x + 1, so ∣g(x)−1∣=∣x∣=x<ε\abs{g(x) - 1} = \abs{x} = x < \eps, where ∣x∣=x\abs{x} = x because x>0x > 0 (the definition of the absolute value).

Non-example. lim⁡x→0+g(x)\lim_{x \to 0^{+}} g(x) is not 0, although g(0)=0g(0) = 0. Take ε=12\eps = \frac12. Whatever δ>0\delta > 0 is offered, let d=12min⁡(δ,1)d = \frac12 \min\bigl(\delta, 1\bigr), which is positive. Since min⁡(δ,1)≤δ\min(\delta, 1) \le \delta, multiplying by the positive number 12\frac12 keeps the inequality, so d≤δ2<δd \le \frac{\delta}{2} < \delta, and the half-window 0<x<δ0 < x < \delta contains x=dx = d. There g(d)=d+1g(d) = d + 1, and adding 1 to d>0d > 0 keeps the inequality, d+1>1d + 1 > 1; so d+1d + 1 is positive, and ∣g(d)−0∣=d+1>1>12\abs{g(d) - 0} = d + 1 > 1 > \frac12 by the definition of the absolute value. So no δ\delta wins the round ε=12\eps = \frac12. (From the left, g(x)=xg(x) = x, and lim⁡x→0−g(x)=0\lim_{x \to 0^{-}} g(x) = 0, with δ=ε\delta = \eps as in the example.)

The figure plays the game of the definition of the limit for the parcel price of Why this matters, with the two-sided claim lim⁡w→2P(w)=5\lim_{w \to 2} P(w) = 5. It reports on the two sides of 2 separately, the left half-window and the right half-window, which is exactly what the one-sided limits do.

Figure 1:The ε–δ game for the claim that the parcel price has the limit 5 at 2 kg. Graph of the price for weights from 1 kg to 3 kg: height 3 to the left of 2 and height 5 to the right, with an open circle at (2,5)(2, 5), the claimed limit. Around it are a horizontal band of half-width ε\eps around y=5y = 5 (translucent, dashed edges) and a vertical window 0<∣x−2∣<δ0 < \abs{x - 2} < \delta (translucent, solid edges; the dotted line x=2x = 2 itself is left out, so the widget never uses the price at exactly 2 kg). A slider sets ε\eps from 0.1 to 2.5 in steps of 0.1, starting at 0.5. For each ε\eps the widget states, separately for the left and the right of 2, the largest δ\delta it found or that it observed none. For ε=0.5\eps = 0.5 it reports on the right that the values stayed in the band at every point it checked, out to the edge of the view (δ=1\delta = 1), and on the left that it observed no δ\delta: the values leave the band at points as close to 2 as it checks. A second slider sets your own δ\delta; points of the graph inside the window but outside the band are marked with crosses, and a sentence says whether your δ\delta passed the widget’s check at the points it tested, or names a point where it fails.

Try this:

  1. At ε=0.5\eps = 0.5, read what the widget reports on each side of 2. Which side’s values are within ε\eps of 5?

  2. Select the ε\eps slider and raise ε\eps one step at a time. At which ε\eps does the widget first find a δ\delta on the left? Why there, and not one step earlier?

  3. Now the widget finds a δ\delta on both sides. Does that make 5 the limit of P(w)P(w) as w→2w \to 2? (The definition asks for a δ\delta for every ε>0\eps > 0.)

The widget checks finitely many points, so what it reports are observations. The proofs are on this page: Example 2 proves that the right-hand limit is 5 and the left-hand limit is 3, so that lim⁡w→2P(w)\lim_{w \to 2} P(w) does not exist, and exercise Exercise 8 finds exactly the ε\eps for which some δ\delta works on the left.

Could a function have two different right-hand limits at the same point? No, by the argument of the uniqueness of limits, now with a point taken in the half-window.

So we may speak of the right-hand limit and the left-hand limit, when they exist.

Main results

The two half-windows together make up the punctured neighbourhood 0<∣x−a∣<δ0 < \abs{x - a} < \delta. So a limit should exist exactly when f(x)f(x) settles near the same number on both sides.

Proof (Rigorous track)

By part (e) of the proposition on distance inequalities, the window 0<∣x−a∣<δ0 < \abs{x - a} < \delta is made of the two half-windows a−δ<x<aa - \delta < x < a and a<x<a+δa < x < a + \delta. So a δ\delta that works for the limit works on each side, and conversely the smaller of a δ\delta for each side works for the limit.

The one-sided limits make sense. By hypothesis, ff is defined at every point of an open interval II containing aa, except possibly at aa. By the definition of an interval, II is (c,d)(c, d), (c,∞)(c, \infty), (−∞,d)(-\infty, d) or R\R, and since a∈Ia \in I, we have c<ac < a whenever II has a left endpoint cc, and a<da < d whenever it has a right endpoint dd. Let rr be the smaller of the numbers a−ca - c and d−ad - a that occur (one of them, if II has only one endpoint), and r=1r = 1 if I=RI = \R. Then r>0r > 0: adding −c-c to c<ac < a keeps the inequality, 0<a−c0 < a - c, and adding −a-a to a<da < d gives 0<d−a0 < d - a. Every xx with a−r<x<a+ra - r < x < a + r lies in II. Adding c−rc - r to r≤a−cr \le a - c keeps the inequality, c≤a−rc \le a - r, so c<xc < x by transitivity; and adding aa to r≤d−ar \le d - a gives a+r≤da + r \le d, so x<dx < d by transitivity. So ff is defined at every point of (a−r,a)(a - r, a) and of (a,a+r)(a, a + r), as parts (b) and (a) of Definition 1 require.

If the limit is LL, so are both one-sided limits. Let ε>0\eps > 0. Since lim⁡x→af(x)=L\lim_{x \to a} f(x) = L, there is a δ>0\delta > 0 such that ff is defined at every xx with 0<∣x−a∣<δ0 < \abs{x - a} < \delta, and ∣f(x)−L∣<ε\abs{f(x) - L} < \eps for each such xx. If a<x<a+δa < x < a + \delta, then 0<∣x−a∣<δ0 < \abs{x - a} < \delta, by part (e) of the proposition, read from right to left. So f(x)f(x) is defined and ∣f(x)−L∣<ε\abs{f(x) - L} < \eps: this δ\delta wins the round ε\eps for the right-hand limit. In the same way, a−δ<x<aa - \delta < x < a implies 0<∣x−a∣<δ0 < \abs{x - a} < \delta, so the same δ\delta wins the round for the left-hand limit. Since ε>0\eps > 0 was arbitrary, lim⁡x→a+f(x)=L\lim_{x \to a^{+}} f(x) = L and lim⁡x→a−f(x)=L\lim_{x \to a^{-}} f(x) = L.

If both one-sided limits are LL, so is the limit. Let ε>0\eps > 0. By part (b) of the definition there is a δ1>0\delta_1 > 0 such that ff is defined and ∣f(x)−L∣<ε\abs{f(x) - L} < \eps at every xx with a−δ1<x<aa - \delta_1 < x < a; by part (a) there is a δ2>0\delta_2 > 0 such that the same holds at every xx with a<x<a+δ2a < x < a + \delta_2. Let δ=min⁡(δ1,δ2)\delta = \min(\delta_1, \delta_2), which is positive, and let 0<∣x−a∣<δ0 < \abs{x - a} < \delta. By part (e) of the proposition, read from left to right, either a−δ<x<aa - \delta < x < a or a<x<a+δa < x < a + \delta.

  • In the first case, multiplying δ≤δ1\delta \le \delta_1 by the negative number −1{-1} reverses it, −δ1≤−δ-\delta_1 \le -\delta, and adding aa keeps it: a−δ1≤a−δ<x<aa - \delta_1 \le a - \delta < x < a. So xx is in the half-window of δ1\delta_1.

  • In the second case, adding aa to δ≤δ2\delta \le \delta_2 gives a<x<a+δ≤a+δ2a < x < a + \delta \le a + \delta_2, so xx is in the half-window of δ2\delta_2.

Either way, f(x)f(x) is defined and ∣f(x)−L∣<ε\abs{f(x) - L} < \eps. So δ\delta wins the round ε\eps for the limit; since ε>0\eps > 0 was arbitrary, lim⁡x→af(x)=L\lim_{x \to a} f(x) = L.

The theorem gives the limit when the two one-sided limits agree. Read the other way, it is the usual way to show that a limit does not exist.

How to use them. To decide whether lim⁡x→af(x)\lim_{x \to a} f(x) exists, find both one-sided limits and compare them. If they are equal, the limit is their common value, by Theorem 1. If they are different, or one of them does not exist, there is no limit, by part (a) or part (b) of Corollary 1.

Worked examples

Piecewise functions

A function given by different formulas on the two sides of aa is made for one-sided limits: each half-window sees only one formula. The value at aa itself, whichever formula or special value gives it, plays no part.

The jump of ∣x∣x\frac{\abs{x}}{x} at 0, the first of the three ways a limit can fail to exist in The Limit of a Function, fits the same pattern: its one-sided limits are −1{-1} and 1 (exercise Exercise 1), so part (a) of Corollary 1 gives again what the jump example proved there with the triangle inequality.

An endpoint

Common mistakes

Rigorous track

Summary

Exercises

Where this leads

These topics build on this one:

Found an error on this page? Report it (the form opens with this page filled in).

The theorem of this page and its corollary are the usual way to decide whether a limit exists: find the two one-sided limits and compare them.