## Theorem Statement
If $f$ is continuous on $[a, b]$ and $F(x) = \int_a^x f(t)\, dt$, then:
$F'(x) = f(x) \quad \text{for all } x \in (a, b)$
Or in Leibniz notation:
$\frac{d}{dx}\int_a^x f(t)\, dt = f(x)$
### What It Means
* Integration and differentiation are inverse operations.
* Every continuous function has an antiderivative.
* The rate of change of accumulated area equals the height of the function.
---
## Rigorous Proof
### Step 1: Set Up the Difference Quotient
For $h > 0$:
$\frac{F(x+h) - F(x)}{h} = \frac{1}{h}\left[\int_a^{x+h} f(t)\, dt - \int_a^x f(t)\, dt\right]$
### Step 2: Apply Interval Additivity
$\int_a^{x+h} f(t)\, dt = \int_a^x f(t)\, dt + \int_x^{x+h} f(t)\, dt$
Therefore:
$\frac{F(x+h) - F(x)}{h} = \frac{1}{h}\int_x^{x+h} f(t)\, dt$
> **Note:** This is the average value of $f$ on $[x, x+h]$.
### Step 3: Use Extreme Value Theorem
Since $f$ is continuous on $[x, x+h]$, it attains:
* Minimum $m_h = f(c_h)$ at some $c_h \in [x, x+h]$
* Maximum $M_h = f(C_h)$ at some $C_h \in [x, x+h]$
Thus:
$m_h \leq f(t) \leq M_h \quad \text{for all } t \in [x, x+h]$
### Step 4: Bound the Integral
$m_h \cdot h \leq \int_x^{x+h} f(t)\, dt \leq M_h \cdot h$
Dividing by $h$:
$m_h \leq \frac{1}{h}\int_x^{x+h} f(t)\, dt \leq M_h$
### Step 5: Apply Squeeze Theorem
As $h \to 0$:
* $c_h \to x$ and $C_h \to x$ (both trapped in $[x, x+h]$)
* By continuity of $f$: $\lim_{h \to 0} m_h = \lim_{h \to 0} M_h = f(x)$
Therefore:
$\lim_{h \to 0} \frac{F(x+h) - F(x)}{h} = f(x)$
$\boxed{F'(x) = f(x)}$
---
## Why the Chain Rule Is Needed for $u(x)$
### The Problem
FTC2 requires the upper limit to be exactly $x$. For:
$G(x) = \int_a^{x^2} f(t)\, dt$
The upper limit is $u(x) = x^2$, not $x$.
### Where the Proof Breaks
Form the difference quotient:
$\frac{G(x+h) - G(x)}{h} = \frac{1}{h}\int_{x^2}^{(x+h)^2} f(t)\, dt$
Critical observation: The interval width is:
$(x+h)^2 - x^2 = 2xh + h^2 \neq h$
### The Interval Width Problem
| Feature | Naked $x$ | Composite $u(x) = x^2$ |
| :--- | :--- | :--- |
| **Width** | $h$ | $2xh + h^2$ |
| **Ratio** $\left(\frac{\text{width}}{h}\right)$ | $1$ | $2x + h \to 2x$ |
### Extracting the Derivative
Rewrite:
$\frac{1}{h}\int_{x^2}^{(x+h)^2} f(t)\, dt = \frac{(x+h)^2 - x^2}{h} \cdot \frac{1}{(x+h)^2 - x^2}\int_{x^2}^{(x+h)^2} f(t)\, dt$
$= (2x + h) \cdot (\text{average value of } f \text{ on } [x^2, (x+h)^2])$
Taking the limit:
* Average value $\to f(x^2)$ by continuity
* $(2x + h) \to 2x = u'(x)$
Result:
$\lim_{h \to 0} \frac{G(x+h) - G(x)}{h} = f(x^2) \cdot 2x$
### General Formula
$\frac{d}{dx}\int_a^{u(x)} f(t)\, dt = f(u(x)) \cdot u'(x)$
---
## Variable in Lower Bound: Where the Negative Sign Emerges
### The Setup
Define:
$G(x) = \int_x^a f(t)\, dt$
### Forming the Difference Quotient
$G(x+h) - G(x) = \int_{x+h}^a f(t)\, dt - \int_x^a f(t)\, dt$
### The Critical Step: Interval Additivity
Apply the defining property:
$\int_x^a f(t)\, dt = \int_x^{x+h} f(t)\, dt + \int_{x+h}^a f(t)\, dt$
Solving for the difference:
$\int_{x+h}^a f(t)\, dt - \int_x^a f(t)\, dt = -\int_x^{x+h} f(t)\, dt$
**The necessity:** The algebra forces the negative sign because we are removing area from the left side of the interval.
### Completing the Proof
$\frac{G(x+h) - G(x)}{h} = -\frac{1}{h}\int_x^{x+h} f(t)\, dt$
Taking the limit:
$\lim_{h \to 0} \frac{G(x+h) - G(x)}{h} = -\lim_{h \to 0}\frac{1}{h}\int_x^{x+h} f(t)\, dt = -f(x)$
$\boxed{\frac{d}{dx}\int_x^a f(t)\, dt = -f(x)}$
### Where the Necessity Emerges
| Step | Upper Bound Variable | Lower Bound Variable |
| :--- | :--- | :--- |
| **Difference** | $\int_x^{x+h} f(t)\, dt$ | $-\int_x^{x+h} f(t)\, dt$ |
| **Divide by $h$** | $\frac{1}{h}\int_x^{x+h} f(t)\, dt$ | $-\frac{1}{h}\int_x^{x+h} f(t)\, dt$ |
| **Limit as $h \to 0$** | $f(x)$ | $-f(x)$ |
---
## The Deep Reason: Orientation
The negative sign is unavoidable due to the algebraic structure of oriented integrals:
$\int_b^a f(t)\, dt = -\int_a^b f(t)\, dt$
This is required by:
$\int_a^b f + \int_b^a f = \int_a^a f = 0$
When solving $\int_{x+h}^a f - \int_x^a f$, the algebra demands:
$= \left(-\int_a^{x+h} f\right) - \left(-\int_a^x f\right) = -\int_a^{x+h} f + \int_a^x f = -\left(\int_a^{x+h} f - \int_a^x f\right) = -\int_x^{x+h} f$
---
## General Case: Both Bounds Vary
Define:
$H(x) = \int_{v(x)}^{u(x)} f(t)\, dt$
Split using constant reference point $a$:
$H(x) = \int_{v(x)}^a f(t)\, dt + \int_a^{u(x)} f(t)\, dt = -\int_a^{v(x)} f(t)\, dt + \int_a^{u(x)} f(t)\, dt$
Differentiate:
$\frac{dH}{dx} = -f(v(x)) \cdot v'(x) + f(u(x)) \cdot u'(x)$
### General Leibniz Rule:
$\boxed{\frac{d}{dx}\int_{v(x)}^{u(x)} f(t)\, dt = f(u(x)) \cdot u'(x) - f(v(x)) \cdot v'(x)}$
---
## Summaries & Key Takeaways
### Integral Derivative Rules
| Integral | Derivative | Intuition |
| :--- | :--- | :--- |
| $\int_a^x f(t)\, dt$ | $f(x)$ | Upper bound moves right, adding area |
| $\int_x^a f(t)\, dt$ | $-f(x)$ | Lower bound moves right, removing area |
| $\int_a^{u(x)} f(t)\, dt$ | $f(u(x)) \cdot u'(x)$ | Chain rule for upper bound |
| $\int_{v(x)}^a f(t)\, dt$ | $-f(v(x)) \cdot v'(x)$ | Chain rule for lower bound |
| $\int_{v(x)}^{u(x)} f(t)\, dt$ | $f(u)u' - f(v)v'$ | Net rate of change |
### Key Principles
* The derivative equals (value at upper bound) × (speed of upper bound) minus (value at lower bound) × (speed of lower bound).
* The negative sign for the lower bound is algebraically unavoidable—it emerges from the orientation property $\int_b^a = -\int_a^b$.
### Connection to First FTC
| Feature | First FTC | Second FTC |
| :--- | :--- | :--- |
| **Statement** | $\int_a^b f(x)\, dx = F(b) - F(a)$ | $\frac{d}{dx}\int_a^x f(t)\, dt = f(x)$ |
| **Role** | Practical: evaluate integrals | Theoretical: guarantees antiderivatives exist |
| **Dependency** | Uses Second FTC to ensure $F$ exists | Independent foundation |
> **Final Note:** The Second FTC provides the theoretical bedrock that makes the First FTC possible.