The slope of a line measures constant rate of change. But most functions do not change at a constant rate—their graphs curve, steepen, and flatten. The derivative extends the idea of slope to curves, capturing how fast a function changes at a single point rather than over an interval.
The construction begins with the difference quotient: the slope of a secant line through two points on the graph. As the two points draw closer together, the secant line approaches the tangent line. The derivative is the limit of this process—the slope of the tangent line, defined precisely through a limiting operation.
This single concept—instantaneous rate of change via a limiting process—generates an entire framework. Rules convert the limit definition into efficient computation. Techniques extend differentiation to implicit and parametric settings. The derivative as a function reveals the shape of curves through sign analysis. Higher-order derivatives capture concavity and deeper layers of change.
Key Terms
The Derivative
Derivative— instantaneous rate of change: f′(a)=limh→0hf(a+h)−f(a)
Given a function f and a point x=a, the difference quotient measures the average rate of change of f over the interval from a to a+h:
The Difference Quotient and Its Limit
hf(a+h)−f(a)
Geometrically, this is the slope of the secant line connecting (a,f(a)) and (a+h,f(a+h)). As h→0, the secant line rotates toward the tangent line at x=a.
The derivative of f at a is defined as the limit of the difference quotient:
The Difference Quotient and Its Limit
f′(a)=h→0limhf(a+h)−f(a)
When this limit exists, f′(a) equals the slope of the tangent line to the graph of f at the point (a,f(a)). When the limit does not exist—due to a corner, cusp, vertical tangent, or discontinuity—the derivative is undefined at that point.
An equivalent form uses x approaching a directly:
The Difference Quotient and Its Limit
f′(a)=x→alimx−af(x)−f(a)
Both forms define the same quantity. The first emphasizes the increment h; the second emphasizes the point x approaching a.
Secant through P₁ and P₂: m = Δy/Δx = 1.93
The two marked points are a and a + h on the curve; the rise Δy over the run Δx is the difference quotient, drawn as the slope of the secant line through them, 1.93 here. Slide P₂ toward P₁ and the secant turns into the tangent while its slope settles on the derivative. Drag the second point in on the average rate of change visualizer.
The derivative is the number this slope approaches, and every later rule is a shortcut for computing it.
Notation
Several notation systems for derivatives coexist, each suited to different contexts.
Lagrange Notation
The symbol f′(x) denotes the derivative of f with respect to x. Higher derivatives use repeated primes: f′′(x), f′′′(x). For the nth derivative, parenthetical notation avoids stacking primes: f(n)(x). This notation is compact and emphasizes the function itself.
Leibniz Notation
The symbol dxdy denotes the derivative of y with respect to x. It suggests a ratio of infinitesimal changes—and while the derivative is defined as a limit rather than a literal fraction, this notation behaves like a fraction in the chain rule and in differentials. Higher derivatives follow the pattern dx2d2y, dx3d3y. The operator form dxd[f(x)] emphasizes differentiation as an operation applied to an expression.
Euler Notation
The operator D or Dx applied to a function: Df(x) means the same as f′(x). Higher orders use powers of the operator: D2f, Dnf. This notation appears in differential equations and operator theory.
Newton Notation
A dot above the variable: y˙ for the first derivative, y¨ for the second. Used almost exclusively where the independent variable is time. The notation signals that differentiation is with respect to t.
In LaTeX source, the Leibniz form is produced by \frac{dy}{dx}, the primes by f' directly, and the Newton dots by \dot{y} and \ddot{y}.
The table below collects the four systems side by side with their first-derivative and higher-derivative forms and the contexts where each one is most commonly used.
compact, function-centric; standard in textbook calculus
Leibniz
dy ⁄ dx
d2y ⁄ dx2, dny ⁄ dxn
behaves like a ratio; natural for chain rule and differentials
Euler
Df(x) or Dxf
D2f, Dnf
differential equations and operator theory
Newton
ẏ
ÿ, ⃛y
used almost exclusively when the independent variable is time
Derivative Graph Analysis
The sign of the derivative at a point reveals whether the function is increasing or decreasing there. Where f′(x)>0, the function rises; where f′(x)<0, the function falls. Points where f′(x)=0 or f′(x) is undefined are critical points—candidates for local maxima, minima, or neither.
The second derivative adds another layer. Where f′′(x)>0, the graph is concave up; where f′′(x)<0, the graph is concave down. Points where concavity changes are inflection points. Together, the first and second derivatives determine the complete shape of a curve: its direction, its turning points, and its bending.
Derivative graph analysis combines these tools into a systematic framework for understanding function behavior, from tangent line equations to optimization and curve sketching.
Horizontal tangent: slope = 0.00 at a local maximum
At the marked point the tangent line is level and the slope readout is exactly zero: the function has stopped rising and is about to fall, which is what a critical point looks like on the graph. To the left the tangents tilt up, to the right they tilt down, the sign change the first derivative test reads. Slide the point across the hilltop on the tangent line at a point tool.
The critical points are where the analysis starts; the sign of the derivative on either side finishes it.
The definition f′(a) produces a single number—the slope at one point. Replacing a with a variable x yields f′(x), a function that assigns a slope to every point in its domain. The derivative function has its own graph, its own properties, and can itself be differentiated.
The graph of f′ encodes the behavior of f: where f climbs, f′ is positive; where f descends, f′ is negative; where f has an extremum, f′ crosses zero. Reading one graph from the other—and reversing the process—is a core skill.
The derivative as a function develops this perspective, including the domain of f′, the relationship between the graphs of f and f′, and the interpretation of f′ as a standalone object.
Not every function has a derivative at every point. The limit definition requires the limit to exist and be finite, which fails at corners, cusps, vertical tangents, and discontinuities. Each failure mode has a distinct geometric signature on the graph.
THEOREM
Differentiability implies continuity—a function must be unbroken at a point to have a tangent line there.
The converse is false: f(x)=∣x∣ is continuous at x=0 but not differentiable, because the left-hand and right-hand slopes disagree.
Differentiability examines when derivatives exist, one-sided derivatives, differentiability on intervals, and pathological cases where continuity holds everywhere but differentiability fails everywhere.
Computing derivatives from the limit definition is tedious for all but the simplest functions. Differentiation rules replace the limit with algebraic procedures: the power rule handles xn, the product rule handles f⋅g, the quotient rule handles f/g, and the chain rule handles compositions f(g(x)).
Beyond computational rules, the Mean Value Theorem guarantees that every differentiable function achieves its average rate of change at some interior point. Rolle's Theorem and L'Hôpital's rule follow as consequences, connecting derivatives to existence results and limit evaluation.
Differentiation rules covers all standard rules, their derivations from the limit definition, and the major theorems that govern derivative behavior.
Standard rules apply directly when y is given explicitly as a function of x. Many relationships, however, are not written in explicit form. The equation x2+y2=25 defines y implicitly; implicit differentiation extracts dy/dx by differentiating both sides and solving.
Logarithmic differentiation handles products with many factors and expressions with variable exponents like xx. Parametric differentiation computes slopes of curves given as x=x(t), y=y(t). The inverse function derivative formula recovers the derivative of f−1 from the derivative of f.
Differentiation techniques develops each method with its underlying logic and the situations where it applies.
A small set of derivative formulas covers the vast majority of computation. The power rule gives dxd[xn]=nxn−1. Trigonometric derivatives follow a pattern: (sinx)′=cosx, (cosx)′=−sinx, (tanx)′=sec2x. The exponential ex is its own derivative, and (lnx)′=1/x.
These formulas, combined with differentiation rules, handle polynomials, rational functions, and all standard transcendental functions. Memorizing them eliminates the need to return to the limit definition.
Derivatives of common functions catalogs these results, derives key formulas from the limit definition, and organizes them into a reference framework.
Beyond the standard functions, several families have derivatives that appear frequently in advanced work. Inverse trigonometric functions produce algebraic derivatives: (arcsinx)′=1/1−x2, (arctanx)′=1/(1+x2). Hyperbolic functions parallel their trigonometric counterparts: (sinhx)′=coshx, (tanhx)′=sech2x.
Piecewise functions require checking differentiability at each boundary, and parametric curves use the formula dy/dx=(dy/dt)/(dx/dt).
Since f′ is a function, it can be differentiated again. The second derivative f′′(x) measures how the slope itself changes—geometrically, this is concavity. The third derivative and beyond capture progressively finer aspects of a function's behavior.
Certain functions exhibit patterns under repeated differentiation. Polynomials of degree n have (n+1)th derivative equal to zero. The exponential ex is unchanged at every order. Sine and cosine cycle through four derivatives before repeating.
Higher-order derivatives explores notation, derivative patterns, smoothness classes, and the connection to Taylor series.
f″(c) = 0.00 at an inflection point
The second derivative is read off the bend of the curve: positive where it cups upward, negative where it cups downward, and zero at the marked point where the bending switches sides. That switch is the inflection point, and f″ changes sign through it. Move the marker into either bend and watch f″(c) leave zero on the concavity and inflection points tool.
Each further derivative describes a finer feature of the graph, but the second is the last one with a shape of its own.
The derivative dy/dx is defined as a limit—a single object. Differentials separate it into two independent quantities: dx, a small change in x, and dy=f′(x)⋅dx, the corresponding change predicted by the tangent line.
The differential dy approximates the actual change Δy=f(x+dx)−f(x). For small dx, the approximation is accurate, making differentials a practical tool for linear approximation and error estimation.
Differentials develops the formalism, connects it to Leibniz notation, and applies it to approximation and error propagation.
The sections above describe each branch of the derivatives subtree in turn. The table below collects them as a single roadmap — grouping the nine sub-pages by category, naming what each one focuses on, and indicating the main payoff a reader should expect from following the link.
A derivative measures the instantaneous rate of change of a function. It is defined as the limit of the difference quotient: f′(a) = lim(h→0) [f(a+h) − f(a)]/h. Geometrically the difference quotient is the slope of a secant line, and as h shrinks toward zero that secant rotates into the tangent line — so f′(a) is the slope of the tangent at (a, f(a)).Read more →
Is dy/dx a fraction?
+
Not literally. The derivative is defined as a limit, not as a quotient of two numbers, and dy and dx carry no independent meaning inside that definition. But the notation is built to behave like a fraction, which is why the chain rule reads as though factors cancel and why differentials can treat dy and dx as separate quantities. Useful as an analogy, unsound as a proof.Read more →
What is the difference between dy/dx and f′(x)?
+
They denote the same derivative in different notations. Lagrange’s f′(x) names the derivative as a function and stacks primes for higher orders, switching to f⁽ⁿ⁾(x) once they pile up. Leibniz’s dy/dx names the two variables being related, which makes the chain rule and substitution transparent; higher orders become d²y/dx². Lagrange is compact, Leibniz is explicit about the variables.Read more →
What does a dot over a variable mean in calculus?
+
It is Newton’s notation for a derivative with respect to time: ẏ means dy/dt, and ÿ means the second derivative with respect to time. The dot is used almost exclusively when the independent variable is time, which is why it runs throughout physics and mechanics — velocity and acceleration appear as ẋ and ẍ — signalling the variable without naming it.Read more →