Skip to content

Mathematics · Ch 9 — Applications of Derivatives

Rolle's Theorem or Rolle's Lemma

9.3.1

Rolle's Theorem or Rolle's Lemma

Rolle's Theorem (or Rolle's Lemma): if a real-valued function ff is continuous on the closed interval [a,b][a,b], differentiable on the open interval (a,b)(a,b), and f(a)=f(b)f(a)=f(b), then there exists at least one point cc in the open interval (a,b)(a,b) such that f′(c)=0f'(c)=0.

In words: any real-valued, differentiable function that takes the same value at two distinct points must have at least one stationary point (where the first derivative — the tangent's slope — is zero) somewhere strictly between them.

Geometrical significance. If f(x)f(x) is continuous on [a,b][a,b] (so its graph can be drawn without lifting the pen from x=ax=a to x=bx=b) and differentiable on (a,b)(a,b) (so the graph has a well-defined tangent at every interior point, with no sharp corners or vertical tangents), and if the graph starts and ends at the same height (f(a)=f(b)f(a)=f(b)), then the graph must rise and then fall (or fall and then rise) somewhere in between — and at the very top (or bottom) of that rise-and-fall, the tangent to the curve is momentarily horizontal, i.e. parallel to the X-axis, since f(a)=f(b)f(a)=f(b) forces the curve to "return" to its starting height.

Figure 2.3.1Fig. 2.3.1 — geometrical significance of Rolle's Theorem: two curves with f(a)=f(b); each has an interior point C(c, f(c)) where the tangent is horizontal, so f′(c)=0.
Fig. 2.3.1 — Fig. 2.3.1 — geometrical significance of Rolle's Theorem: two curves with f(a)=f(b); each has an interior point C(c, f(c)) where the tangent is horizontal, so f′(c)=0.

Drawn by us to help you understand the concept clearly, and verified to make sure it's accurate. For exams, practice from your textbook's own diagram.

What this figure shows. This figure illustrates Rolle's Theorem geometrically. A smooth curve y=f(x)y=f(x) is drawn over the closed interval [a,b][a,b], starting and ending at the same height since f(a)=f(b)f(a)=f(b). The curve rises and then falls (or dips and then rises) between the two endpoints, and the figure marks an interior point x=cx=c strictly between aa and bb at which the tangent line drawn to the curve is horizontal, i.e. parallel to the X-axis — visually showing that the curve must turn around somewhere in betw …

Worked Example 1 — Checking the hypotheses. (i) For f(x)=2x3−5x2+3x+2f(x)=2x^3-5x^2+3x+2 on [0,32]\left[0,\dfrac32\right]: this is a polynomial, so it is automatically continuous on the closed interval and differentiable on the open interval. Checking the endpoint values: f(0)=2f(0)=2 and f(32)=2(278)−5(94)+3(32)+2=54−90+368+2=2f\left(\dfrac32\right)=2\left(\dfrac{27}{8}\right)-5\left(\dfrac94\right)+3\left(\dfrac32\right)+2=\dfrac{54-90+36}{8}+2=2. Since f(0)=f(32)=2f(0)=f\left(\dfrac32\right)=2, all the conditions of Rolle's theorem are satisfied. (ii) For f(x)=x2−2x+3f(x)=x^2-2x+3 on [1,4][1,4]: again a polynomial (smooth everywhere), but f(1)=2f(1)=2 while f(4)=11f(4)=11; since f(1)≠f(4)f(1)\ne f(4), the conditions of Rolle's theorem are NOT satisfied for this function on this interval.

Worked Example 2 — Full verification, finding cc. Verify Rolle's theorem for f(x)=x2−4x+10f(x)=x^2-4x+10 on [0,4][0,4]. As a polynomial, ff is continuous and differentiable everywhere. f(0)=10f(0)=10 and f(4)=16−16+10=10f(4)=16-16+10=10, so f(0)=f(4)=10f(0)=f(4)=10: all conditions hold. Differentiating: f′(x)=2x−4=2(x−2)f'(x)=2x-4=2(x-2). Setting f′(c)=0f'(c)=0 gives c=2c=2, and indeed 2∈(0,4)2\in(0,4) — Rolle's theorem is verified.

Worked Example 3 — Finding bb given aa. Given that Rolle's theorem holds for f(x)=x3−2x2+3f(x)=x^3-2x^2+3 on some interval [a,b][a,b] with a=0a=0, find bb. Write f(x)=g(x)+3f(x)=g(x)+3 where g(x)=x3−2x2=x2(x−2)g(x)=x^3-2x^2=x^2(x-2), which is zero exactly at x=0x=0 and x=2x=2. So f(0)=g(0)+3=3f(0)=g(0)+3=3 and f(2)=g(2)+3=3f(2)=g(2)+3=3, i.e. f(0)=f(2)=3f(0)=f(2)=3. Since ff is a cubic polynomial (continuous and differentiable everywhere), all the conditions of Rolle's theorem are satisfied on [0,2][0,2], so b=2b=2. …