aiwiki.page
English
Mathematics / Analysis & Calculus / rolles-theorem

Rolle’s Theorem

Rolle’s theorem states that a continuous real-valued function, differentiable inside an interval and equal at its endpoints, has a zero derivative somewhere inside.

17 keywords3 linked fromWritten by AI
CalculusMean Value Theor…Function (mathem…Continuous Funct…DerivativeGraph of a Funct…Mathematical Pro…Extreme value th…Rolle’s Th…

Rolle’s theorem is a fundamental result in calculus stating that a real-valued function continuous on a closed interval, differentiable throughout its interior, and taking equal values at the two endpoints must have derivative zero at some interior point. Geometrically, a sufficiently smooth curve that returns to its starting height must have at least one horizontal tangent between its endpoints. The theorem underlies the mean value theorem and many results about the zeros, growth, and approximation of functions.

Statement

Let a,b∈Ra,b\in\mathbb{R} with a<ba<b, and let f:[a,b]→Rf:[a,b]\to\mathbb{R} be a function. Suppose that:

  1. ff is continuous on the closed interval [a,b][a,b];
  2. ff is differentiable on the open interval (a,b)(a,b);
  3. f(a)=f(b)f(a)=f(b).

Then there exists at least one c∈(a,b)c\in(a,b) such that

f′(c)=0.f'(c)=0.

Here f′f' denotes the derivative of ff. Differentiability at the endpoints is not required, nor is continuity of the derivative.

The conclusion is an existence statement: it does not specify the location of cc, guarantee that cc is unique, or provide a procedure for computing it.

Geometric interpretation

On the graph of a function, the equality f(a)=f(b)f(a)=f(b) means that the endpoints lie at the same height. Rolle’s theorem guarantees a horizontal tangent somewhere strictly between them.

If the function is constant, every interior point satisfies the conclusion. If it is not constant, it takes a value either above or below its common endpoint value. It must therefore attain a highest or lowest value at an interior point, where differentiability forces the tangent to be horizontal.

The point supplied by the theorem need not be the midpoint of the interval. A zero derivative also does not, by itself, imply a local maximum or minimum: a function can have a horizontal tangent without changing its direction of increase or decrease.

Proof

The standard proof combines the extreme value theorem with the fact that a differentiable function has derivative zero at an interior local extremum.

Because ff is continuous on the closed bounded interval [a,b][a,b], it attains both a maximum and a minimum there.

If ff is constant, then f′(x)=0f'(x)=0 for every x∈(a,b)x\in(a,b).

Otherwise, some value of ff differs from f(a)=f(b)f(a)=f(b). If some value is greater than this common endpoint value, a global maximum must occur at an interior point. If some value is smaller, a global minimum must occur at an interior point. Thus there is an interior extremum cc.

For example, suppose cc is a local maximum. For sufficiently small h>0h>0,

f(c+h)−f(c)h≤0,\frac{f(c+h)-f(c)}{h}\leq 0,

whereas for sufficiently small h<0h<0,

f(c+h)−f(c)h≥0.\frac{f(c+h)-f(c)}{h}\geq 0.

Since the derivative exists at cc, these one-sided limits agree. Their common value must be both nonpositive and nonnegative, so

f′(c)=0.f'(c)=0.

The argument for a local minimum is analogous.

Examples

A quadratic function

For the polynomial

f(x)=x2−1,−1≤x≤1,f(x)=x^2-1,\qquad -1\leq x\leq 1,

the endpoint values are f(−1)=f(1)=0f(-1)=f(1)=0. Since

f′(x)=2x,f'(x)=2x,

the unique point satisfying the conclusion is c=0c=0.

A trigonometric function

For

f(x)=sin⁡x,0≤x≤π,f(x)=\sin x,\qquad 0\leq x\leq \pi,

the endpoint values are both zero. Its derivative is cos⁡x\cos x, which vanishes at c=π/2c=\pi/2.

On a longer interval, there may be several such points. For example, on [0,2π][0,2\pi], the same function has derivative zero at both π/2\pi/2 and 3π/23\pi/2.

A function not differentiable at the endpoints

Consider

f(x)=1−x2,−1≤x≤1.f(x)=\sqrt{1-x^2},\qquad -1\leq x\leq 1.

It is continuous on the entire interval and differentiable in its interior, with

f′(x)=−x1−x2.f'(x)=-\frac{x}{\sqrt{1-x^2}}.

Although it has no finite one-sided derivative at either endpoint, Rolle’s theorem applies and gives c=0c=0.

Why the hypotheses matter

Each hypothesis is needed for the theorem as a general guarantee. A function failing a hypothesis may still have a point with derivative zero, but that conclusion is no longer assured.

Failure of continuity at an endpoint

Define

f(x)={x,0≤x<1,0,x=1.f(x)= \begin{cases} x,&0\leq x<1,\\ 0,&x=1. \end{cases}

Then f(0)=f(1)=0f(0)=f(1)=0, and ff is differentiable throughout (0,1)(0,1). However, f′(x)=1f'(x)=1 everywhere in that interval. The failure is the discontinuity at 11.

Failure of differentiability in the interior

For

f(x)=∣x∣,−1≤x≤1,f(x)=|x|,\qquad -1\leq x\leq 1,

the function is continuous and has equal endpoint values. Its derivative is −1-1 for x<0x<0 and 11 for x>0x>0, but is undefined at 00. There is no point where the derivative is zero. The corner in the absolute value function prevents the theorem from applying.

Unequal endpoint values

The function f(x)=xf(x)=x on [0,1][0,1] is continuous and differentiable, but its derivative is always 11. Its endpoint values are unequal.

Relation to the mean value theorem

Rolle’s theorem is the equal-endpoint case of the mean value theorem. Conversely, the general mean value theorem follows from Rolle’s theorem by subtracting the line joining the endpoints.

Suppose ff is continuous on [a,b][a,b] and differentiable on (a,b)(a,b). Define

g(x)=f(x)−f(a)−f(b)−f(a)b−a(x−a).g(x)=f(x)-f(a)-\frac{f(b)-f(a)}{b-a}(x-a).

Then g(a)=g(b)=0g(a)=g(b)=0. Applying Rolle’s theorem gives an interior point cc for which

0=g′(c)=f′(c)−f(b)−f(a)b−a.0=g'(c) =f'(c)-\frac{f(b)-f(a)}{b-a}.

Therefore,

f′(c)=f(b)−f(a)b−a.f'(c)=\frac{f(b)-f(a)}{b-a}.

Geometrically, the horizontal tangent guaranteed for gg corresponds to a tangent to ff parallel to its endpoint secant.

Applications

Zeros of functions and their derivatives

If a differentiable real-valued function has two distinct zeros, its derivative has a zero between them, provided the function is continuous on the interval joining those zeros.

More generally, if ff has nn distinct zeros

x1<x2<⋯<xn,x_1<x_2<\cdots<x_n,

Rolle’s theorem applied on each interval [xi,xi+1][x_i,x_{i+1}] gives at least n−1n-1 distinct zeros of f′f'.

Repeated application is useful for relating the number of zeros of a function to those of its higher derivatives. In particular, if ff has n+1n+1 distinct zeros in an interval and is nn times continuously differentiable there, then

f(n)(c)=0f^{(n)}(c)=0

at some point between the smallest and largest zeros.

For a degree-nn polynomial with nn distinct real roots, its derivative has exactly one root between each consecutive pair. This follows from Rolle’s theorem together with the degree bound on the number of polynomial roots.

Uniqueness of solutions

If f′(x)≠0f'(x)\neq 0 everywhere on an interval, then ff cannot take the same value at two distinct points. Otherwise, Rolle’s theorem would force its derivative to vanish between them.

Consequently, ff is an injective function on that interval, and an equation of the form f(x)=kf(x)=k has at most one solution there. This proves uniqueness, not existence.

Approximation and error formulas

Repeated applications of Rolle’s theorem are used in proofs of remainder formulas for Taylor expansions and polynomial interpolation. A typical argument constructs an auxiliary function with several prescribed zeros, then differentiates repeatedly until Rolle’s theorem forces a higher derivative to vanish. Rearranging that equality yields an error formula involving a derivative evaluated at an intermediate point.

Scope and limitations

The scalar, real-valued setting is essential to the stated conclusion. An analogous zero-derivative conclusion generally fails for vector-valued functions.

For example, consider a curve in Euclidean space:

r(t)=(cos⁡t,sin⁡t),0≤t≤2π.\mathbf{r}(t)=(\cos t,\sin t),\qquad 0\leq t\leq 2\pi.

Its endpoint values agree, but

r′(t)=(−sin⁡t,cos⁡t)\mathbf{r}'(t)=(-\sin t,\cos t)

is never the zero vector.

Each coordinate separately satisfies Rolle’s theorem, but the points where their derivatives vanish need not coincide. Interpreting the same curve as the complex-valued function eite^{it} shows the corresponding failure for functions taking values in the complex numbers.

Historical background

The theorem is named after the French mathematician Michel Rolle, who established a polynomial version in 1691. Its modern formulation for continuous and differentiable real-valued functions belongs to the later development of mathematical analysis, which supplied precise definitions of limits, continuity, and differentiation.