Search Knowledge

© 2026 LIBREUNI PROJECT

Introduction to Calculus/Limits and derivatives
Page 3 of 12

Mean Value Theorem: The Guarantee of Speed

Rigorous proofs and applications of Rolle's Theorem, the Mean Value Theorem, Cauchy's Generalization, and L'Hôpital's Rule.

Lesson briefLimits and derivatives · 3 h · 3 outcomes

Required knowledge

  • Differentiability, continuity on closed intervals, and the Extreme Value Theorem
  • Chain rule and local extrema conditions (Fermat theorem)

Learning outcomes

  • Prove Rolle theorem and Lagrange Mean Value Theorem from the Extreme Value Theorem
  • Use derivative sign and MVT bounds to prove monotonicity, inequalities, and root uniqueness
  • Apply Cauchy Mean Value Theorem and L’Hôpital’s Rule while verifying indeterminate-form hypotheses

Assessment target

Prove a functional inequality or root-uniqueness claim using the Mean Value Theorem and evaluate indeterminate limits without circular differentiation.

Scope boundary

  • Multivariable Taylor remainder forms and vector-valued mean value inequalities.

The Mean Value Theorem (MVT) is one of the most critical theoretical tools in calculus. It establishes a direct connection between the average rate of change over an interval (global behavior) and the instantaneous rate of change at a point within that interval (local behavior).

Rolle’s Theorem: The Foundation

Before proving the general Mean Value Theorem, we must prove its special case, Rolle’s Theorem.

Theorem 1.3.1Rolle's Theorem

Builds on:Theorem 1.1.6

Let f:[a,b]→Rf: [a, b] \to \mathbb{R} be a function that satisfies:

  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.

ProofExpand proof

By the Extreme Value Theorem (Theorem 1.1.6), the continuous function ff on the compact interval [a,b][a, b] attains a global maximum MM and a global minimum mm

∃ xmin⁡,xmax⁡∈[a,b]such thatm=f(xmin⁡)≤f(x)≤f(xmax⁡)=M∀x∈[a,b].\exists\, x_{\min}, x_{\max} \in [a, b] \quad \text{such that} \quad m = f(x_{\min}) \le f(x) \le f(x_{\max}) = M \quad \forall x \in [a, b].

Case 1: Constant function M=mM = m

If M=mM = m then ff is constant throughout [a,b][a, b] Therefore:

f′(x)=0for all x∈(a,b),f'(x) = 0 \quad \text{for all } x \in (a, b),

and every interior point c∈(a,b)c \in (a, b) satisfies the theorem.

Case 2: Non-trivial extremum in the interior M≠mM \neq m

Because f(a)=f(b)f(a) = f(b) at least one of the extrema must occur at an interior point c∈(a,b)c \in (a, b) Without loss of generality, suppose the maximum occurs in the interior: f(c)=Mf(c) = M for some c∈(a,b)c \in (a, b)

Since f(c)≥f(x)f(c) \ge f(x) for all x∈[a,b]x \in [a, b] the numerator f(x)−f(c)≤0f(x) - f(c) \le 0 everywhere. We evaluate the one-sided difference quotients:

  1. Right-hand derivative x>cx > c Here x−c>0x - c > 0 so dividing non-positive by positive yields a non-positive ratio:
f(x)−f(c)x−c≤0  ⟹  f+′(c)=lim⁡x→c+f(x)−f(c)x−c≤0.\frac{f(x) - f(c)}{x - c} \le 0 \implies f'_+(c) = \lim_{x \to c^+} \frac{f(x) - f(c)}{x - c} \le 0.
  1. Left-hand derivative x<cx < c Here x−c<0x - c < 0 so dividing non-positive by negative reverses the inequality:
f(x)−f(c)x−c≥0  ⟹  f−′(c)=lim⁡x→c−f(x)−f(c)x−c≥0.\frac{f(x) - f(c)}{x - c} \ge 0 \implies f'_-(c) = \lim_{x \to c^-} \frac{f(x) - f(c)}{x - c} \ge 0.

Since ff is differentiable at cc both one-sided limits must agree with the two-sided derivative:

f′(c)=f+′(c)=f−′(c).f'(c) = f'_+(c) = f'_-(c).

The only real value that satisfies both f′(c)≤0f'(c) \le 0 and f′(c)≥0f'(c) \ge 0 is:

f′(c)=0.f'(c) = 0.
∎
Rolle's theorem: $f(-\sqrt{3}) = f(\sqrt{3}) = 0$ implies $f'(c) = 0$ at both critical points $c = -1$ and $c = 1$. Rolle's theorem on the cubic polynomial x cubed minus 3x between roots minus square root of 3 and square root of 3 with horizontal tangents at x = plus and minus 1. c=-1 c=1
Figure 1.3.1: Rolle's theorem: f(−3)=f(3)=0f(-\sqrt{3}) = f(\sqrt{3}) = 0 implies f′(c)=0f'(c) = 0 at both critical points c=−1c = -1 and c=1c = 1.

The Mean Value Theorem

We generalize Rolle’s Theorem to functions where f(a)≠f(b)f(a) \neq f(b) by tilting the coordinate system.

Theorem 1.3.2Mean Value Theorem (MVT)

Builds on:Theorem 1.3.1

Let f:[a,b]→Rf: [a, b] \to \mathbb{R} be a function satisfying:

  1. ff is continuous on [a,b][a, b]
  2. ff is differentiable on (a,b)(a, b)

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

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

The tangent at $c$ is parallel to the chord: $f'(c) = \frac{f(2) - f(0)}{2} = 4$. Graph of y = x cubed on [0, 2] with the chord from (0, 0) to (2, 8) and the parallel tangent line at x = c = 2 over square root of 3. c
Figure 1.3.2: The tangent at cc is parallel to the chord: f′(c)=f(2)−f(0)2=4f'(c) = \frac{f(2) - f(0)}{2} = 4.

Pause & Predict: Does the MVT extend to Vector-Valued Functions? Suppose f:[a,b]→R2\mathbf{f}: [a, b] \to \mathbb{R}^2 is smooth and continuous. Does there necessarily exist c∈(a,b)c \in (a, b) such that f′(c)=f(b)−f(a)b−a\mathbf{f}'(c) = \frac{\mathbf{f}(b) - \mathbf{f}(a)}{b - a} Think of a particle moving along a circular path before opening the analysis below.

Show hintExpand

Consider a particle travelling in a full circle f(t)=(cos⁡t,sin⁡t)\mathbf{f}(t) = (\cos t, \sin t) on [0,2π][0, 2\pi]. What is the average velocity vector, and does the particle ever come to a complete stop?

Reveal analysisExpand

Analysis: No! The Mean Value Theorem fails for vector-valued functions f:R→Rn\mathbf{f}: \mathbb{R} \to \mathbb{R}^n n≥2n \ge 2 Consider f(t)=(cos⁡t,sin⁡t)\mathbf{f}(t) = (\cos t, \sin t) on [0,2π][0, 2\pi] The average velocity over one full orbit is: f(2π)−f(0)2π−0=(1,0)−(1,0)2π=(0,0).\frac{\mathbf{f}(2\pi) - \mathbf{f}(0)}{2\pi - 0} = \frac{(1, 0) - (1, 0)}{2\pi} = (0, 0). However, the velocity vector is f′(t)=(−sin⁡t,cos⁡t)\mathbf{f}'(t) = (-\sin t, \cos t) whose magnitude is ∥f′(t)∥=(−sin⁡t)2+(cos⁡t)2=1\|\mathbf{f}'(t)\| = \sqrt{(-\sin t)^2 + (\cos t)^2} = 1 for all tt The instantaneous velocity never vanishes! In higher dimensions, one can only obtain the Mean Value Inequality: ∥f(b)−f(a)∥≤(b−a)sup⁡t∈(a,b)∥f′(t)∥\|\mathbf{f}(b) - \mathbf{f}(a)\| \le (b - a) \sup_{t \in (a, b)} \|\mathbf{f}'(t)\|

ProofExpand proof

We construct an auxiliary function g(x)g(x) measuring the vertical difference between the curve y=f(x)y = f(x) and the secant chord passing through (a,f(a))(a, f(a)) and (b,f(b))(b, f(b))

g(x)=f(x)−[f(a)+f(b)−f(a)b−a(x−a)].g(x) = f(x) - \left[ f(a) + \frac{f(b) - f(a)}{b - a}(x - a) \right].

We check the three hypotheses of Rolle’s Theorem (Theorem 1.3.1) on [a,b][a, b]

  1. Continuity: gg is continuous on [a,b][a, b] because ff is continuous and the linear chord is continuous.
  2. Differentiability: gg is differentiable on (a,b)(a, b) because ff is differentiable and linear terms are smooth.
  3. Boundary conditions: Evaluating gg at both endpoints:
g(a)=f(a)−[f(a)+0]=0,g(a) = f(a) - \left[ f(a) + 0 \right] = 0,g(b)=f(b)−[f(a)+f(b)−f(a)b−a(b−a)]=f(b)−f(b)=0.g(b) = f(b) - \left[ f(a) + \frac{f(b) - f(a)}{b - a}(b - a) \right] = f(b) - f(b) = 0.

Since g(a)=g(b)=0g(a) = g(b) = 0 Rolle’s Theorem guarantees that there exists at least one c∈(a,b)c \in (a, b) such that:

g′(c)=0.g'(c) = 0.

Differentiating g(x)g(x) directly:

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

Evaluating at x=cx = c

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

The theorem identifies a point where the tangent slope equals the secant slope; the interactive Python visualization below lets us inspect that relationship directly. Let us write a Python script to verify the MVT for f(x)=x3−3xf(x) = x^3 - 3x on [0,2][0, 2] The secant slope is (f(2)−f(0))/(2−0)=(2−0)/2=1.0(f(2)-f(0))/(2-0) = (2-0)/2 = 1.0 We find c∈(0,2)c \in (0, 2) where f′(c)=3c2−3=1.0  ⟹  3c2=4  ⟹  c=4/3≈1.1547f'(c) = 3c^2 - 3 = 1.0 \implies 3c^2 = 4 \implies c = \sqrt{4/3} \approx 1.1547

python

Interactive Lab

1import numpy as np
2import matplotlib.pyplot as plt
3 
4def f(x): return x**3 - 3*x
5def df(x): return 3*x**2 - 3
6 
7a, b = 0.0, 2.0
8secant_slope = (f(b) - f(a)) / (b - a)
9 
10# Find c in (a, b) such that df(c) == secant_slope
11c = np.sqrt(4.0 / 3.0)
12 
13x = np.linspace(-0.2, 2.2, 200)
14y = f(x)
15 
16plt.plot(x, y, label='f(x) = x^3 - 3x', color='#3b82f6', linewidth=2)
17plt.plot([a, b], [f(a), f(b)], '--', color='red', label='Secant Line')
18plt.scatter([c], [f(c)], color='black', zorder=5)
19 
20# Tangent line at c
21tangent_line = f(c) + df(c) * (x - c)
22plt.plot(x, tangent_line, ':', color='green', label=f'Tangent at c ≈ {c:.4f}')
23 
24plt.legend()
25plt.title("Mean Value Theorem Visualization")
26plt.grid(True)
27plt.show()
28 
29print(f"Secant Slope: {secant_slope}")
30print(f"Derivative at c: {df(c)}")
31 

Critical Corollaries of the MVT

The MVT allows us to transition from local derivative properties to global function properties.

Corollary 1.3.3Zero Derivative Implies Constant

Builds on:Theorem 1.3.2

If f′(x)=0f'(x) = 0 for all xx in an open interval II then ff is constant on II

ProofExpand proof

Pick any two points x1,x2∈Ix_1, x_2 \in I with x1<x2x_1 < x_2 Since ff is differentiable on II it is continuous on [x1,x2][x_1, x_2] and differentiable on (x1,x2)(x_1, x_2) By the MVT, there exists c∈(x1,x2)c \in (x_1, x_2) such that:

f′(c)=f(x2)−f(x1)x2−x1.f'(c) = \frac{f(x_2) - f(x_1)}{x_2 - x_1}.

Since f′(x)=0f'(x) = 0 for all x∈Ix \in I we must have f′(c)=0f'(c) = 0 Therefore,

f(x2)−f(x1)x2−x1=0  ⟹  f(x2)−f(x1)=0  ⟹  f(x2)=f(x1).\frac{f(x_2) - f(x_1)}{x_2 - x_1} = 0 \implies f(x_2) - f(x_1) = 0 \implies f(x_2) = f(x_1).

Since this holds for any x1,x2∈Ix_1, x_2 \in I ff is constant on II

∎

Cauchy’s Generalized Mean Value Theorem

We can generalize the MVT to compare two functions parametrically.

Theorem 1.3.4Cauchy's Mean Value Theorem

Builds on:Theorem 1.3.1

Let f,g:[a,b]→Rf, g: [a, b] \to \mathbb{R} be continuous on [a,b][a, b] and differentiable on (a,b)(a, b) If g′(x)≠0g'(x) \neq 0 for all x∈(a,b)x \in (a, b) then there exists some c∈(a,b)c \in (a, b) such that:

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

ProofExpand proof

First, note g(b)−g(a)≠0g(b) - g(a) \neq 0 because if g(b)=g(a)g(b) = g(a) Rolle’s Theorem would imply there is a point where g′(c)=0g'(c) = 0 violating the hypothesis. We define:

h(x)=[f(b)−f(a)]g(x)−[g(b)−g(a)]f(x).h(x) = [f(b) - f(a)]g(x) - [g(b) - g(a)]f(x).

  1. hh is continuous on [a,b][a, b] and differentiable on (a,b)(a, b)
  2. Compute h(a)=f(b)g(a)−g(b)f(a)h(a) = f(b)g(a) - g(b)f(a) and h(b)=f(b)g(a)−g(b)f(a)h(b) = f(b)g(a) - g(b)f(a) Thus h(a)=h(b)h(a) = h(b)
  3. By Rolle’s Theorem, there exists c∈(a,b)c \in (a, b) such that h′(c)=0h'(c) = 0

h′(c)=[f(b)−f(a)]g′(c)−[g(b)−g(a)]f′(c)=0.h'(c) = [f(b) - f(a)]g'(c) - [g(b) - g(a)]f'(c) = 0.

Dividing by g′(c)[g(b)−g(a)]g'(c)[g(b) - g(a)] yields:

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

∎

L’Hôpital’s Rule

Cauchy’s MVT is the fundamental tool used to prove L’Hôpital’s Rule for indeterminate limits of the form 0/00/0 or ∞/∞\infty/\infty

Theorem 1.3.5L'Hôpital's Rule

Builds on:Theorem 1.3.4

Let ff and gg be differentiable on an open interval containing cc (except possibly at cc If lim⁡x→cf(x)=lim⁡x→cg(x)=0\lim_{x \to c} f(x) = \lim_{x \to c} g(x) = 0 and g′(x)≠0g'(x) \neq 0 for x≠cx \neq c then:

lim⁡x→cf(x)g(x)=lim⁡x→cf′(x)g′(x),\lim_{x \to c} \frac{f(x)}{g(x)} = \lim_{x \to c} \frac{f'(x)}{g'(x)},

provided the limit on the right exists.

Worked Trace: Cauchy’s Mean Value Theorem Calculation

To see Cauchy’s MVT in practice, consider f(x)=x2f(x) = x^2 and g(x)=x3g(x) = x^3 on the interval [1,2][1, 2]

  1. Verify Hypotheses: Both ff and gg are polynomials, so they are continuous on [1,2][1, 2] and differentiable on (1,2)(1, 2) Moreover, g′(x)=3x2>0g'(x) = 3x^2 > 0 for all x∈(1,2)x \in (1, 2) satisfying g′(x)≠0g'(x) \neq 0

  2. Compute Secant Quotient: f(2)−f(1)g(2)−g(1)=22−1223−13=4−18−1=37.\frac{f(2) - f(1)}{g(2) - g(1)} = \frac{2^2 - 1^2}{2^3 - 1^3} = \frac{4 - 1}{8 - 1} = \frac{3}{7}.

  3. Compute Derivative Ratio: f′(c)g′(c)=2c3c2=23c.\frac{f'(c)}{g'(c)} = \frac{2c}{3c^2} = \frac{2}{3c}.

  4. Solve for cc: 23c=37  ⟹  9c=14  ⟹  c=149≈1.5556.\frac{2}{3c} = \frac{3}{7} \implies 9c = 14 \implies c = \frac{14}{9} \approx 1.5556.

  5. Interior Verification: Since 1<149<21 < \frac{14}{9} < 2 the solution lies strictly in the open interior (1,2)(1, 2) exactly as guaranteed by the theorem.

Counterexample & Failure Mode: Boundary Breakdown Under Weakened Hypotheses

The Mean Value Theorem requires both continuity on the closed interval [a,b][a, b] and differentiability on the open interval (a,b)(a, b) Violating either condition causes the theorem to collapse immediately.

  1. Failure of Differentiability (Interior Cusp): Consider f(x)=∣x∣f(x) = |x| on [−1,1][-1, 1]

    • Continuity holds everywhere on [−1,1][-1, 1]
    • At x=0x = 0 the left-derivative is −1-1 and right-derivative is +1+1 so ff is not differentiable at x=0x = 0
    • Secant slope: f(1)−f(−1)1−(−1)=1−12=0\frac{f(1) - f(-1)}{1 - (-1)} = \frac{1 - 1}{2} = 0
    • Instantaneous derivative: f′(x)=−1f'(x) = -1 for x<0x < 0 and f′(x)=+1f'(x) = +1 for x>0x > 0
    • Result: There is no point c∈(−1,1)c \in (-1, 1) where f′(c)=0f'(c) = 0 The conclusion fails completely.
  2. Failure of Continuity (Endpoint Jump): Consider f(x)={0x=01x∈(0,1]f(x) = \begin{cases} 0 & x = 0 \\ 1 & x \in (0, 1] \end{cases} on [0,1][0, 1]

    • ff is differentiable on (0,1)(0, 1) with f′(x)=0f'(x) = 0 everywhere.
    • Secant slope: f(1)−f(0)1−0=1−01=1\frac{f(1) - f(0)}{1 - 0} = \frac{1 - 0}{1} = 1
    • Result: f′(c)=0≠1f'(c) = 0 \neq 1 for all c∈(0,1)c \in (0, 1) The failure of continuity at even a single boundary point invalidates the guarantee.

Concept Check: Boundary Differentiability

Verification Check

0 / 4 answered

Consider f(x)=x2/3f(x) = x^{2/3} on [−1,1][-1, 1] with f(−1)=f(1)=1f(-1) = f(1) = 1. Evaluate each statement regarding Rolle's Theorem and the derivative:

Classify each option as correct or incorrect, or leave uncertain options unanswered (“I don't know”) and verify at any time.

1
Rolle's Theorem is guaranteed to hold because the function is continuous on the closed interval [−1,1][-1, 1].
2
No point c∈(−1,1)c \in (-1, 1) satisfies f′(c)=0f'(c) = 0; the derivative is nonzero wherever it is defined.
3
Rolle's Theorem fails to apply because ff is not differentiable at the interior point x=0x = 0.
4
The function fails to satisfy Rolle's Theorem because its graph is discontinuous at the origin x=0x = 0.
Failure of the Mean Value Theorem when differentiability fails at an interior point: $f(x) = x^{2/3}$ has chord slope $0$ on $[-1, 1]$, but $f'(x) \ne 0$ for all $x \ne 0$. Failure of the Mean Value Theorem for the cusp function x to the two thirds on [-1, 1], where the chord is horizontal but no horizontal tangent exists.
Figure 1.3.3: Failure of the Mean Value Theorem when differentiability fails at an interior point: f(x)=x2/3f(x) = x^{2/3} has chord slope 00 on [−1,1][-1, 1], but f′(x)≠0f'(x) \ne 0 for all x≠0x \ne 0.
Section Exercises

Deep Practice

Apply the Mean Value Theorem and its structural corollaries to establish root bounds, functional inequalities, and indeterminate limits.

Exercise 1.3.1: Rolle's Theorem Root Verification

Verificationapplication

Verify that Theorem 1.3.1 applies to f(x)=x3−4xf(x) = x^3 - 4x on the interval [0,2][0, 2] and find the exact value of c∈(0,2)c \in (0, 2) where f′(c)=0f'(c) = 0

HintExpand

Check the three hypotheses of Rolle's Theorem: continuity on [0,2][0, 2], differentiability on (0,2)(0, 2), and endpoint equality f(0)=f(2)f(0) = f(2).

Worked solutionExpand
  1. Hypothesis 1: ff is a polynomial, hence continuous on [0,2][0, 2]
  2. Hypothesis 2: ff is differentiable on (0,2)(0, 2) with f′(x)=3x2−4f'(x) = 3x^2 - 4
  3. Hypothesis 3: f(0)=03−4(0)=0f(0) = 0^3 - 4(0) = 0 and f(2)=23−4(2)=8−8=0f(2) = 2^3 - 4(2) = 8 - 8 = 0 So f(0)=f(2)=0f(0) = f(2) = 0
  4. Locating cc By Theorem 1.3.1, there exists c∈(0,2)c \in (0, 2) such that f′(c)=0f'(c) = 0 3c2−4=0  ⟹  c2=43  ⟹  c=23=233.3c^2 - 4 = 0 \implies c^2 = \frac{4}{3} \implies c = \frac{2}{\sqrt{3}} = \frac{2\sqrt{3}}{3}. Since 3≈1.732\sqrt{3} \approx 1.732 c≈1.155∈(0,2)c \approx 1.155 \in (0, 2)

Verification criteria. Requires verifying all three hypotheses and solving for the exact positive root 2/32/\sqrt{3}

Exercise 1.3.2: Mean Value Tangent Point Construction

Constructionapplication

Let f(x)=x2f(x) = x^2 on [a,b][a, b] Find the exact value of c∈(a,b)c \in (a, b) guaranteed by Theorem 1.3.2. What geometric relationship does cc have with the interval endpoints?

HintExpand

Compute the secant slope f(b)−f(a)b−a\frac{f(b)-f(a)}{b-a} and set it equal to f′(c)=2cf'(c) = 2c.

Worked solutionExpand
  1. The secant slope is: f(b)−f(a)b−a=b2−a2b−a=(b−a)(b+a)b−a=b+a=a+b.\frac{f(b) - f(a)}{b - a} = \frac{b^2 - a^2}{b - a} = \frac{(b - a)(b + a)}{b - a} = b + a = a + b.
  2. The derivative is f′(x)=2xf'(x) = 2x
  3. By Theorem 1.3.2, set f′(c)=f(b)−f(a)b−af'(c) = \frac{f(b)-f(a)}{b-a} 2c=a+b  ⟹  c=a+b2.2c = a + b \implies c = \frac{a + b}{2}.
  4. Geometric property: For any quadratic parabola f(x)=Ax2+Bx+Cf(x) = Ax^2 + Bx + C the tangent line parallel to the secant line occurs exactly at the midpoint c=(a+b)/2c = (a + b)/2 of the interval.

Verification criteria. A complete derivation requires setting secant slope equal to 2c2c and concluding c=(a+b)/2c = (a+b)/2

Exercise 1.3.3: Boundary Discontinuity Invalidating MVT Guarantee

Example / Impossibletransfer

Consider f(x)=x2/3f(x) = x^{2/3} on [−1,1][-1, 1] Does Theorem 1.3.2 guarantee a point c∈(−1,1)c \in (-1, 1) where the tangent slope equals the secant slope? Explain why or why not.

HintExpand

Compute the derivative f′(x)f'(x) and check where it is undefined on the open interval (−1,1)(-1, 1).

Worked solutionExpand
Show cusp diagram(Click to reveal)
Failure of the Mean Value Theorem when differentiability fails at an interior point: $f(x) = x^{2/3}$ has chord slope $0$ on $[-1, 1]$, but $f'(x) \ne 0$ for all $x \ne 0$. Failure of the Mean Value Theorem for the cusp function x to the two thirds on [-1, 1], where the chord is horizontal but no horizontal tangent exists.
Figure 1.3.3: Failure of the Mean Value Theorem when differentiability fails at an interior point: f(x)=x2/3f(x) = x^{2/3} has chord slope 00 on [−1,1][-1, 1], but f′(x)≠0f'(x) \ne 0 for all x≠0x \ne 0.
  1. The average rate of change on [−1,1][-1, 1] is: f(1)−f(−1)1−(−1)=12/3−(−1)2/32=1−12=0.\frac{f(1) - f(-1)}{1 - (-1)} = \frac{1^{2/3} - (-1)^{2/3}}{2} = \frac{1 - 1}{2} = 0.
  2. The derivative is: f′(x)=23x−1/3=23x3.f'(x) = \frac{2}{3}x^{-1/3} = \frac{2}{3\sqrt[3]{x}}.
  3. At x=0∈(−1,1)x = 0 \in (-1, 1) f′(0)f'(0) is undefined (the graph has a vertical cusp).
  4. For all x≠0x \ne 0 f′(x)≠0f'(x) \ne 0 because the numerator is 2≠02 \ne 0
  5. Therefore, there is no point c∈(−1,1)c \in (-1, 1) with f′(c)=0f'(c) = 0 even though the secant slope is 00
  6. Theorem 1.3.2 does not apply because ff is not differentiable throughout the entire open interval (−1,1)(-1, 1)

Verification criteria. Requires computing secant slope =0= 0 showing f′f' is undefined at 00 and never zero elsewhere, and explicitly identifying the violation of open-interval differentiability.

Exercise 1.3.4: Root Uniqueness via Strict Monotonicity

Prooftransfer

Prove that the polynomial p(x)=x5+4x−7p(x) = x^5 + 4x - 7 has exactly one real root. State explicitly which structural theorems guarantee (a) existence and (b) uniqueness.

HintExpand

Use Theorem 1.1.5 to establish existence on a compact interval, and Theorem 1.3.1 (or sign of p′p' to prove uniqueness by contradiction.

Worked solutionExpand

Part 1: Existence (via Theorem 1.1.5) p(x)p(x) is a polynomial, hence continuous on R\mathbb{R} Evaluate at endpoints x=1x = 1 and x=2x = 2 p(1)=15+4(1)−7=1+4−7=−2<0.p(1) = 1^5 + 4(1) - 7 = 1 + 4 - 7 = -2 < 0. p(2)=25+4(2)−7=32+8−7=33>0.p(2) = 2^5 + 4(2) - 7 = 32 + 8 - 7 = 33 > 0. Since p(1)<0<p(2)p(1) < 0 < p(2) by the Intermediate Value Theorem (Theorem 1.1.5), there exists at least one r∈(1,2)r \in (1, 2) such that p(r)=0p(r) = 0

Part 2: Uniqueness (via Theorem 1.3.1) Suppose for contradiction that p(x)p(x) has two distinct real roots r1<r2r_1 < r_2 Then p(r1)=p(r2)=0p(r_1) = p(r_2) = 0 Since pp is continuous on [r1,r2][r_1, r_2] and differentiable on (r1,r2)(r_1, r_2) by Rolle’s Theorem (Theorem 1.3.1), there must exist some c∈(r1,r2)c \in (r_1, r_2) such that p′(c)=0p'(c) = 0 Now compute the derivative: p′(x)=5x4+4.p'(x) = 5x^4 + 4. Since x4≥0x^4 \ge 0 for all real xx 5x4+4≥4>05x^4 + 4 \ge 4 > 0 for all x∈Rx \in \mathbb{R} Thus p′(x)p'(x) is strictly positive everywhere and can never equal 00 This contradicts the existence of cc Therefore, p(x)p(x) cannot have two or more real roots; the root is strictly unique.

Verification criteria. Requires: (a) invoking Theorem 1.1.5 with explicit negative and positive values; (b) assuming two roots for contradiction; (c) invoking Theorem 1.3.1; (d) proving p′(x)≥4>0p'(x) \ge 4 > 0 to reach contradiction.

Exercise 1.3.5: Proving Functional Inequalities: Sine Bound

Proofsynthesis

Use Theorem 1.3.2 to prove that sin⁡x<x\sin x < x for all x>0x > 0

HintExpand

Apply the Mean Value Theorem to f(t)=sin⁡tf(t) = \sin t on the interval [0,x][0, x] for x>0x > 0.

Worked solutionExpand
  1. Let x>0x > 0 Consider f(t)=sin⁡tf(t) = \sin t on the closed interval [0,x][0, x]
  2. ff is continuous on [0,x][0, x] and differentiable on (0,x)(0, x) with f′(t)=cos⁡tf'(t) = \cos t
  3. By Theorem 1.3.2, there exists c∈(0,x)c \in (0, x) such that: sin⁡x−sin⁡0x−0=cos⁡c  ⟹  sin⁡xx=cos⁡c.\frac{\sin x - \sin 0}{x - 0} = \cos c \implies \frac{\sin x}{x} = \cos c.
  4. For any c∈(0,x)c \in (0, x) we know that cos⁡c≤1\cos c \le 1 Furthermore, since c∈(0,x)c \in (0, x) and c>0c > 0 cos⁡c=1\cos c = 1 can only occur if cc is an integer multiple of 2π2\pi But at c=0c = 0 cos⁡0=1\cos 0 = 1 for all c∈(0,2π)c \in (0, 2\pi) cos⁡c<1\cos c < 1 In general, for any c>0c > 0 the inequality cos⁡c<1\cos c < 1 holds almost everywhere, and specifically on (0,min⁡(x,2π))(0, \min(x, 2\pi)) cos⁡c<1\cos c < 1 For x≥1x \ge 1 sin⁡x≤1≤x\sin x \le 1 \le x and since sin⁡x≤1<x\sin x \le 1 < x for x>1x > 1 the claim holds. For 0<x≤10 < x \le 1 0<c<x≤1<π/20 < c < x \le 1 < \pi/2 where cosine is strictly decreasing from 11 so cos⁡c<cos⁡0=1\cos c < \cos 0 = 1
  5. Therefore: sin⁡xx=cos⁡c<1  ⟹  sin⁡x<x.\frac{\sin x}{x} = \cos c < 1 \implies \sin x < x.

Verification criteria. Requires setting up the interval [0,x][0, x] applying Theorem 1.3.2, and justifying the strict inequality cos⁡c<1\cos c < 1

Exercise 1.3.6: Cauchy MVT Ratio Evaluation

Constructionapplication

Let f(x)=x3f(x) = x^3 and g(x)=x2+1g(x) = x^2 + 1 on [1,3][1, 3] Find the exact value of c∈(1,3)c \in (1, 3) guaranteed by Theorem 1.3.4.

HintExpand

Compute the ratio f(3)−f(1)g(3)−g(1)\frac{f(3)-f(1)}{g(3)-g(1)} and set it equal to f′(c)g′(c)\frac{f'(c)}{g'(c)}.

Worked solutionExpand
  1. Compute function values: f(3)=27,f(1)=1  ⟹  f(3)−f(1)=26f(3) = 27, f(1) = 1 \implies f(3) - f(1) = 26 g(3)=10,g(1)=2  ⟹  g(3)−g(1)=8g(3) = 10, g(1) = 2 \implies g(3) - g(1) = 8
  2. The secant ratio is: f(3)−f(1)g(3)−g(1)=268=134.\frac{f(3) - f(1)}{g(3) - g(1)} = \frac{26}{8} = \frac{13}{4}.
  3. The derivatives are f′(x)=3x2f'(x) = 3x^2 and g′(x)=2xg'(x) = 2x Note g′(x)=2x≠0g'(x) = 2x \ne 0 on (1,3)(1, 3)
  4. By Theorem 1.3.4, set: f′(c)g′(c)=3c22c=3c2=134.\frac{f'(c)}{g'(c)} = \frac{3c^2}{2c} = \frac{3c}{2} = \frac{13}{4}.
  5. Solve for cc 3c=132  ⟹  c=136≈2.167.3c = \frac{13}{2} \implies c = \frac{13}{6} \approx 2.167. Since 1<136<31 < \frac{13}{6} < 3 c∈(1,3)c \in (1, 3) as guaranteed.

Verification criteria. Requires computing both differences, setting 3c/2=13/43c/2 = 13/4 and finding c=13/6c = 13/6

Exercise 1.3.7: L'Hôpital Evaluation of Quadratic Cosine Indeterminacy

Verificationapplication

Evaluate lim⁡x→01−cos⁡xx2\lim_{x \to 0} \frac{1 - \cos x}{x^2} using Theorem 1.3.5, verifying hypotheses at every step.

Fill-in Exercise

1 blank to complete

Second-step L'Hôpital quotient

Applying L'Hôpital's Rule once to lim⁡x→01−cos⁡xx2\lim_{x \to 0} \frac{1 - \cos x}{x^2} gives lim⁡x→0sin⁡x2x\lim_{x \to 0} \frac{\sin x}{2x}, which is still of the indeterminate form 0/00/0.
Applying L'Hôpital's Rule a second time yields lim⁡x→0cos⁡x2=\lim_{x \to 0} \frac{\cos x}{2} =
HintExpand

Check that both numerator and denominator evaluate to 00 before each derivative step.

Worked solutionExpand
  1. Step 1 (Check initial form): lim⁡x→0(1−cos⁡x)=1−1=0\lim_{x \to 0} (1 - \cos x) = 1 - 1 = 0 lim⁡x→0x2=0\lim_{x \to 0} x^2 = 0 Denominator derivative: ddx[x2]=2x≠0\frac{d}{dx}[x^2] = 2x \ne 0 for x≠0x \ne 0 The form is indeterminate 0/00/0
  2. Apply Theorem 1.3.5 once: lim⁡x→01−cos⁡xx2=lim⁡x→0sin⁡x2x.\lim_{x \to 0} \frac{1 - \cos x}{x^2} = \lim_{x \to 0} \frac{\sin x}{2x}.
  3. Step 2 (Check second form): lim⁡x→0sin⁡x=0\lim_{x \to 0} \sin x = 0 and lim⁡x→02x=0\lim_{x \to 0} 2x = 0 Denominator derivative: ddx[2x]=2≠0\frac{d}{dx}[2x] = 2 \ne 0 The form is again 0/00/0
  4. Apply Theorem 1.3.5 a second time: lim⁡x→0sin⁡x2x=lim⁡x→0cos⁡x2.\lim_{x \to 0} \frac{\sin x}{2x} = \lim_{x \to 0} \frac{\cos x}{2}.
  5. Evaluate the direct limit: lim⁡x→0cos⁡x2=cos⁡02=12.\lim_{x \to 0} \frac{\cos x}{2} = \frac{\cos 0}{2} = \frac{1}{2}. Therefore, the limit is 1/21/2

Verification criteria. A complete derivation requires verifying the 0/00/0 condition prior to each differentiation step and obtaining 1/21/2

References & Further Reading

  • Spivak, M. (2008). Calculus (4th ed.), Chapter 11 (“Significance of the Derivative”). Publish or Perish.
  • Rudin, W. (1976). Principles of Mathematical Analysis (3rd ed.), Chapter 5 (“Differentiation”) — includes Darboux’s theorem on intermediate values of derivatives (Theorem 5.12), L’Hôpital’s rule (Theorem 5.13), and vector-valued mean value inequalities (Theorem 5.19). McGraw-Hill.
  • Apostol, T. M. (1967). Calculus, Volume 1: One-Variable Calculus, with an Introduction to Linear Algebra (2nd ed.), Chapter 4 (“The Mean-Value Theorem and Its Applications”). Wiley.