Differential Calculus

Differential calculus is the study of how functions change, captured by the derivative — the instantaneous rate of change, or the slope of the tangent line. The central object is the derivative

f^{'} (x) = lim_{h \to 0} \frac{f ( x + h ) - f ( x )}{h}

, which converts a function into its rate-of-change function. It applies anywhere quantities vary: velocity from position, marginal cost in economics, optimization, and curve sketching. Developed independently by Newton and Leibniz in the late 17th century.

Differential calculus is the branch of mathematics that quantifies instantaneous change. Given a function $f (x)$ , its derivative $f^{'} (x)$ measures how fast $f$ changes at each point — geometrically, the slope of the tangent line to the graph at that point. It matters because almost every dynamic process in science — motion, growth, decay, flow, optimization — is expressed in the language of rates. The surprising fact is that an instantaneous rate, which naively involves dividing by zero, is made perfectly rigorous through the concept of a limit, allowing us to evaluate change "at a single instant" rather than over an interval.

The Core Insight

The slope between two nearby points on a curve is an average rate of change. By sliding the second point ever closer to the first — taking a limit as the separation shrinks to zero — the average rate becomes an instantaneous rate. The derivative is precisely this limit, and it transforms the impossible idea of "slope at a single point" into a well-defined number.

Formal Statement

The Derivative

The derivative of a function $f$ at a point $x$ is defined as the limit of the difference quotient:

f^{'} (x) = h \to 0 lim \frac{f ( x + h ) - f ( x )}{h}

⟳The fraction is the slope of the secant line through

(x, f (x))

and

(x + h, f (x + h))

; letting

h \to 0

collapses the secant onto the tangent.

Symbols:

$f^{'} (x)$ — the derivative (Lagrange notation); also written $\frac{d y}{d x}$ (Leibniz), $\overset{y}{˙}$ (Newton), or $D f$ .
$h$ — the small horizontal increment, approaching zero.
$f (x + h) - f (x)$ — the change in output (rise).
$h$ in the denominator — the change in input (run).

A function is differentiable at $x$ if this limit exists. The derivative function $f^{'}$ assigns to each $x$ the slope of the tangent there.

Equivalent Point Form

An alternative form fixes a point $a$ and lets the variable $x$ approach it:

f^{'} (a) = x \to a lim \frac{f ( x ) - f ( a )}{x - a}

⟳Substitute

x = a + h

into [1]; as

x \to a

h \to 0

, and the two definitions coincide.

Differentiability Implies Continuity

If $f$ is differentiable at $a$ , then $f$ is continuous at $a$ . The converse is false.

x \to a lim f (x) = f (a) whenever f^{'} (a) exists

⟳Write

f (x) - f (a) = \frac{f ( x ) - f ( a )}{x - a} \cdot (x - a)

; both factors have finite limits (

f^{'} (a)

and

0

), so the product

\to 0

Why It Holds / Derivation

The Power Rule from First Principles

We derive $\frac{d}{d x} x^{n} = n x^{n - 1}$ for a positive integer $n$ .

Step 1. Apply the definition [1] to

f (x) = x^{n}

f^{'} (x) = h \to 0 lim \frac{( x + h ) ^{n} - x ^{n}}{h}

⟳Direct substitution of

x^{n}

into the difference quotient.

Step 2. Expand

(x + h)^{n}

with the binomial theorem:

(x + h)^{n} = x^{n} + n x^{n - 1} h + (2 n) x^{n - 2} h^{2} + \dots + h^{n}

⟳The binomial theorem

(a + b)^{n} = \sum_{k = 0}^{n} (k n) a^{n - k} b^{k}

with

a = x

b = h

Step 3. Subtract

x^{n}

and divide by

h

. The

x^{n}

terms cancel, and every remaining term has at least one factor of

h

\frac{( x + h ) ^{n} - x ^{n}}{h} = n x^{n - 1} + (2 n) x^{n - 2} h + \dots + h^{n - 1}

⟳Factor

h

from the numerator (after the

x^{n}

cancellation) and cancel with the denominator.

Step 4. Take the limit

h \to 0

. Every term containing

h

vanishes, leaving:

f^{'} (x) = n x^{n - 1}

⟳All terms after the first carry a positive power of

h

, hence

\to 0

∎

Gottfried Wilhelm Leibniz | 1646–1716

A German polymath — philosopher, diplomat, and mathematician — Leibniz developed his version of calculus in the 1670s, publishing in 1684. His genius lay in notation: the symbols $d y / d x$ and $\int$ were so well-designed that they survive virtually unchanged today, encoding the operations they represent and suggesting valid manipulations almost automatically. His emphasis on differentials (infinitesimally small quantities $d x$ , $d y$ ) gave calculus a flexible algebraic feel. A bitter priority dispute with Newton, fueled by nationalist rivalries, embittered Leibniz's later years, though both men are now credited as independent co-discoverers.

Building Intuition

Picture driving a car. The average speed over a trip is total distance divided by total time — a secant slope. But the speedometer reads your instantaneous speed: the slope of the tangent at one moment. The derivative is the mathematical speedometer. Now imagine zooming into a smooth curve under a microscope: the more you magnify, the straighter it looks. At sufficient magnification, any differentiable curve becomes indistinguishable from a straight line — and the slope of that line is the derivative. This is the principle of local linearity: differentiable functions are "locally straight."

Secant lines converging to the tangent as $h \to 0$

<svg viewBox="0 0 420 300" xmlns="http://www.w3.org/2000/svg">
<rect width="420" height="300" fill="#0a0e17"/>
<line x1="40" y1="260" x2="400" y2="260" stroke="#94a3b8" stroke-width="1.5"/>
<line x1="40" y1="20" x2="40" y2="260" stroke="#94a3b8" stroke-width="1.5"/>
<text x="405" y="264" fill="#e2e8f0" font-size="12">x</text>
<text x="28" y="20" fill="#e2e8f0" font-size="12">y</text>
<path d="M 60 250 Q 200 250 360 60" stroke="#d4a853" stroke-width="2.5" fill="none"/>
<text x="330" y="55" fill="#d4a853" font-size="12">y=f(x)</text>
<circle cx="160" cy="218" r="4" fill="#e2e8f0"/>
<text x="140" y="238" fill="#e2e8f0" font-size="11">(x, f(x))</text>
<circle cx="300" cy="120" r="4" fill="#94a3b8"/>
<text x="285" y="112" fill="#e2e8f0" font-size="11">(x+h, f(x+h))</text>
<line x1="120" y1="248" x2="340" y2="80" stroke="#94a3b8" stroke-width="1.2" stroke-dasharray="5 4"/>
<text x="250" y="180" fill="#94a3b8" font-size="10">secant</text>
<line x1="100" y1="245" x2="260" y2="180" stroke="#e2e8f0" stroke-width="1.8"/>
<text x="105" y="240" fill="#e2e8f0" font-size="10">tangent (limit)</text>
<line x1="160" y1="218" x2="300" y2="218" stroke="#94a3b8" stroke-width="1" stroke-dasharray="3 3"/>
<line x1="300" y1="218" x2="300" y2="120" stroke="#94a3b8" stroke-width="1" stroke-dasharray="3 3"/>
<text x="220" y="232" fill="#94a3b8" font-size="10">h</text>
</svg>

Worked Examples

Instantaneous Velocity of a Falling Object

Given: A ball dropped from rest has position

s (t) = 4.9 t^{2}

metres after

t

seconds. Find its velocity at

t = 3

s. Step 1. Velocity is the derivative of position:

v (t) = s^{'} (t)

. Step 2. Apply the power rule [7] (with the constant multiple rule):

\frac{d}{d t} (4.9 t^{2}) = 4.9 \cdot 2 t = 9.8 t

. Step 3. Evaluate at

t = 3

v (3) = 9.8 \times 3 = 29.4

v (3) = 29.4 m/s

This matches free-fall under gravity, $g \approx 9.8 m/s^{2}$ , accumulated over 3 seconds.

Maximizing Area of a Rectangular Pen

Given: A farmer has 40 m of fencing to enclose a rectangular pen. What dimensions maximize the area? Step 1. Let width

= x

. Perimeter gives

2 x + 2 y = 40

, so

y = 20 - x

. Step 2. Area

A (x) = x (20 - x) = 20 x - x^{2}

. Step 3. Differentiate:

A^{'} (x) = 20 - 2 x

. Step 4. Set

A^{'} (x) = 0

for a critical point:

20 - 2 x = 0 \Rightarrow x = 10

. Step 5. Since

A^{''} (x) = - 2 < 0

, this is a maximum. Then

y = 20 - 10 = 10

x = y = 10 m, A_{m a x} = 100 m^{2}

The optimal rectangle is a square — a recurring theme in optimization.

Common Misconceptions

The derivative is not a fraction — though

\frac{d y}{d x}

looks like one. It is the limit of a fraction. While Leibniz notation can be manipulated as if

d y

and

d x

were separate quantities (and this often works), treating them as genuine numbers without justification can lead to error. Also beware:

f^{'} (x) = 0

does not guarantee a maximum or minimum — it could be a horizontal inflection point (e.g.

f (x) = x^{3}

x = 0

Continuous but Not Differentiable

The function $f (x) = ∣ x ∣$ is continuous everywhere but not differentiable at $x = 0$ . From the left, the slope is $- 1$ ; from the right, it is $+ 1$ . The two one-sided limits of the difference quotient disagree:

h \to 0^{-} lim \frac{∣ h ∣}{h} = - 1 \neq = + 1 = h \to 0^{+} lim \frac{∣ h ∣}{h}

⟳For

h < 0

∣ h ∣ = - h

so the quotient is

- 1

; for

h > 0

∣ h ∣ = h

giving

+ 1

This proves the converse of [3] fails: continuity does not imply differentiability. A sharp corner has no single tangent line.

Connections

Differential calculus is one half of a profound unity. Its partner, Integral Calculus, reverses differentiation — and the Fundamental Theorem of Calculus reveals that these inverse operations are bound together, linking the slope of a curve to the area beneath it. The machinery of derivatives rests entirely on Limits and is intimately tied to Continuity. The derivative's geometric meaning depends on the Tangent Line, while its applications to maxima and minima form the subject of Optimization. Extending derivatives to functions of several variables opens the door to Partial Derivatives and the gradient.

Real-World Applications

Physics — Kinematics: Velocity is the derivative of position; acceleration is the derivative of velocity. Newton's second law $F = m \frac{d v}{d t}$ is a derivative statement.
Engineering — Optimization: Minimizing material in a soda can, maximizing the strength of a beam, or finding least-cost designs all use $f^{'} (x) = 0$ .
Economics — Marginal Analysis: Marginal cost and marginal revenue are derivatives of total cost and revenue; profit is maximized where marginal cost equals marginal revenue.
Medicine — Pharmacokinetics: The rate at which a drug concentration changes in the bloodstream, $\frac{d C}{d t}$ , governs dosing schedules.
Machine Learning — Gradient Descent: Neural networks are trained by following the negative derivative (gradient) of a loss function to minimize error.
Biology — Population Dynamics: Growth rates $\frac{d P}{d t}$ model bacterial colonies and ecosystems.
Computer Graphics — Curve Smoothness: Derivatives ensure smooth Bézier curves and animation paths with no jarring velocity jumps.

Historical Development

c. 250 BCE — Archimedes uses the "method of exhaustion" to find tangents and areas, anticipating limit ideas without formalizing them.
1629 — Pierre de Fermat develops adequality, a method for finding maxima, minima, and tangents that foreshadows the derivative.
1665–1666 — Isaac Newton, during the plague years at Woolsthorpe, develops his "method of fluxions," treating quantities as flowing in time.
1675–1684 — Gottfried Leibniz independently invents his calculus with the notation $d y / d x$ and $\int$ , publishing in Acta Eruditorum (1684).
1687 — Newton publishes the Principia, using calculus-based reasoning (though geometrically disguised) to derive the laws of motion and gravitation.
1711–1716 — The Priority Dispute between Newton and Leibniz erupts publicly, dividing British and Continental mathematicians for a century.
1734 — Bishop George Berkeley's The Analyst attacks the logical foundations, mocking infinitesimals as "ghosts of departed quantities."
1821 — Augustin-Louis Cauchy rigorously grounds calculus in the concept of the limit, dispensing with vague infinitesimals.
1872 — Karl Weierstrass provides the precise $ε$ – $δ$ definition of the limit, completing the rigorous foundation of differential calculus.

Subtleties & Limits

While differential calculus is fully rigorous, deep subtleties remain at its boundaries. Weierstrass (1872) stunned mathematicians by constructing a function that is continuous everywhere but differentiable nowhere — a curve so jagged it has no tangent at any point, anticipating fractals. This raises the question: how "typical" is differentiability? In a precise sense, most continuous functions are nowhere differentiable. Furthermore, non-standard analysis (Abraham Robinson, 1960s) rehabilitated Leibniz's infinitesimals as legitimate numbers in an extended number system, showing his intuitive approach was logically sound after all. Modern research extends differentiation to fractional orders (e.g. a "half-derivative" $\frac{d ^{1/2} f}{d x ^{1/2}}$ ) and to fractal sets where ordinary calculus fails.