Calculus › Section B · Limits › Topic B.4

Continuity and one-sided limits

Some functions you can draw without lifting your pen; others break. Continuity makes that idea exact, and one-sided limits are the precise way to talk about what happens at each side of a break.

Approaching from one side only

Until now, "x approaches \(a\)" meant from both directions at once. Sometimes only one direction makes sense: at the end of an interval, at the seam of a piecewise function, or where a formula is undefined on one side. The left-hand limit, written \(\lim_{x \to a^{-}} f(x)\), uses only inputs smaller than \(a\); the right-hand limit, \(\lim_{x \to a^{+}} f(x)\), uses only larger ones. Each is found exactly like an ordinary limit, by substitution or algebra, but sampling one side only. The greatest integer function \(\lfloor x \rfloor\), which gives the greatest integer less than or equal to \(x\), shows why we bother: its left and right limits disagree at every integer.

The existence test

One-sided limits give the cleanest verdict on the two-sided question: \(\lim_{x \to a} f(x) = L\) if and only if both one-sided limits exist and both equal \(L\). The two halves carry all the information. If either one-sided limit fails to exist, or they exist but disagree, the two-sided limit does not exist. This turns a fuzzy 'does it settle?' question into two concrete one-sided computations. At the seams of the real world, postage priced by weight, parking charged by the hour, calls billed by the minute, this test is the tool that says exactly where the breaks are and how big they are.

Worked example: both sides of a seam

Let \(f(x) = 2x + 1\) for \(x < 3\) and \(f(x) = 10 - x\) for \(x \ge 3\). Find the one-sided limits at the seam. From the left, use the first piece: \(\lim_{x \to 3^{-}}(2x + 1) = 7\). From the right, use the second: \(\lim_{x \to 3^{+}}(10 - x) = 7\). They agree, so \(\lim_{x \to 3} f(x) = 7\), and since \(f(3) = 10 - 3 = 7\) matches that limit, the function turns out to be seamless after all. The same check finds unknown constants: if the right piece were \(k - x\), the pieces would meet only when \(k - 3 = 7\), that is \(k = 10\); any other \(k\) leaves a jump.

Continuity at a point

A function \(f\) is continuous at \(x = a\) when three conditions all hold: \(f(a)\) is defined; \(\lim_{x \to a} f(x)\) exists; and the two are equal.

\lim_{x \to a} f(x) = f(a)

Continuity is the formal version of "the pen never leaves the paper." It also explains the strategy from the last topic: direct substitution worked on polynomials precisely because they are continuous everywhere, so for them the limit is the value. When a function misbehaves at a point, at least one of the three conditions has failed, and naming which one classifies the break.

Three kinds of discontinuity

A removable discontinuity is a hole: the limit exists, but \(f(a)\) is missing or defined as something else. The function \(\frac{x^2 - 4}{x - 2}\) has one at \(2\); redefining a single point repairs it. A jump discontinuity has two different one-sided limits, like the greatest integer function at every integer; no single-point fix can bridge it. An infinite discontinuity is a blow-up, like \(\frac{1}{x-2}\) at \(x = 2\), where at least one side shoots off without bound. Only the removable kind can be repaired, which is where the name comes from.

Worked example: classifying breaks

Classify the discontinuity of \(g(x) = \dfrac{x^2 - x - 2}{x - 2}\) at \(x = 2\). The numerator factors as \((x - 2)(x + 1)\), so for \(x \ne 2\) the function equals \(x + 1\): both one-sided limits are \(3\), but \(g(2)\) is undefined. A hole: removable, and setting \(g(2) = 3\) would repair it. Contrast \(h(x) = \dfrac{1}{x - 2}\) at the same point: as \(x\) approaches \(2\) from the right the outputs explode upward, from the left downward. Neither one-sided limit exists as a finite number, so this is an infinite discontinuity, the vertical asymptote you will study in B.5.

Which functions are continuous?

A stockpile of good behaviour, worth memorising once: polynomials are continuous everywhere. Rational functions are continuous wherever their denominators are not zero. Root, exponential, logarithmic and trigonometric functions are continuous at every interior point of their domains. And continuity survives the usual constructions: sums, differences, products and compositions of continuous functions are continuous, and quotients are too wherever the denominator stays nonzero. The practical effect is enormous. For almost any formula you can write down, the discontinuities are exactly the obvious suspects, points where a denominator vanishes, an even root meets a negative number, or a piecewise seam jumps.

Continuity on an interval

A function is continuous on an open interval \((a, b)\) when it is continuous at every point inside. For a closed interval \([a, b]\), we cannot ask for two-sided good behaviour at the endpoints, because the function may not exist outside, so we ask for continuity from the inside: right-continuous at \(a\) and left-continuous at \(b\). \(\sqrt{x}\) illustrates the point perfectly: at \(0\) only the right side exists, and \(\sqrt{x}\) is continuous on \([0, \infty)\) on exactly this technicality. Closed-interval continuity is the hypothesis the big 'guarantee' theorem next needs.

The Intermediate Value Theorem

If \(f\) is continuous on \([a, b]\), then \(f\) takes on every value between \(f(a)\) and \(f(b)\) somewhere inside the interval: for any \(N\) strictly between them, there is a \(c\) in \((a, b)\) with \(f(c) = N\). The pen-lifting picture is exactly right: a continuous path from one height to another crosses every intermediate height. The famous use is root-finding: if a continuous function is negative at one end and positive at the other, it must equal zero somewhere in between. The theorem promises existence without location, which is weaker and yet remarkably powerful.

Worked example: trapping a root

Show that \(f(x) = x^3 - 2x - 5\) has a root between \(2\) and \(3\). First, \(f\) is a polynomial, so it is continuous everywhere, including on \([2, 3]\). Then evaluate the ends: \(f(2) = 8 - 4 - 5 = -1\) and \(f(3) = 27 - 6 - 5 = 16\). Since \(0\) lies between \(-1\) and \(16\), the Intermediate Value Theorem gives a number \(c\) between \(2\) and \(3\) with \(f(c) = 0\). Narrow further by hand: \(f(2.1) = 9.261 - 4.2 - 5 = 0.061\), positive, so the root actually sits between \(2\) and \(2.1\). Repeating such checks, for instance halving the interval each time, is a primitive but reliable root-finder.

Common mistakes to avoid

The usual slips: checking only \(f(a)\) and declaring continuity, when the limit must exist and match too; assuming a function with a hole is 'not continuous anywhere', when one bad point is one bad point; applying the Intermediate Value Theorem to a discontinuous function, where it guarantees nothing; reading the theorem as telling you where the root is or that there is only one, when it says neither; and forgetting that quotients inherit a discontinuity wherever the denominator is zero, even when the numerator is perfectly tame. Finally, on a closed interval do not demand two-sided limits at the endpoints: one-sided continuity there is all that is required.

Key takeaways

One-sided limits sample from one side only, and the two-sided limit exists exactly when both exist and agree. Continuity at \(a\) needs three things at once: \(f(a)\) defined, the limit existing, and the two being equal. Discontinuities come in three kinds, removable holes, jumps and infinite blow-ups, and only holes are repairable. The familiar function families are continuous on their domains, and continuity survives sums, products, compositions and safe quotients. On a closed interval the endpoints get one-sided checks. The Intermediate Value Theorem converts continuity into guaranteed values and, above all, guaranteed roots.

Flashcards

Flashcard

x → a⁻

Approach from the left only

Sample inputs smaller than a.

Test your understanding

Question 1 of 5

f(x) = 2x + 1 for x < 3 and f(x) = 10 − x for x ≥ 3. What is the left-hand limit as x approaches 3?

Question 2 of 5

What are the one-sided limits of the greatest integer function ⌊x⌋ as x approaches 2?

Question 3 of 5

Which set of conditions makes f continuous at x = a?

Question 4 of 5

What kind of discontinuity does f(x) = (x² − 1)/(x − 1) have at x = 1?

Question 5 of 5

f(x) = x³ − 2x − 5 satisfies f(2) = −1 and f(3) = 16. What does the Intermediate Value Theorem guarantee?

Standards: 1

← Previous topicNext topic →