Parametric Equations and Curves

Parametric equations describe a curve by expressing each coordinate as a function of an independent third variable (the parameter, usually

t

x = f (t)

y = g (t)

. Key result: any motion or curve in the plane can be traced by letting

t

vary, capturing not just the shape but the direction and timing of traversal — information lost in ordinary

y = F (x)

form. Applies to: projectile motion, circles and ellipses, cycloids, orbital paths, computer graphics, and any curve that loops back on itself. Who/when: rooted in the analytic geometry of Descartes and Fermat (1630s), formalized through the calculus of motion by Newton, Euler, and others into the 18th century.

A parametric curve is the set of points $(x, y)$ obtained as a parameter $t$ ranges over an interval, where $x = f (t)$ and $y = g (t)$ are functions of $t$ . This matters because the ordinary function form $y = F (x)$ can only describe curves that pass the vertical line test — one $y$ per $x$ — which excludes circles, spirals, self-intersecting loops, and the paths of moving objects that double back. The parametric viewpoint frees us from that restriction and, crucially, encodes how a curve is drawn: its starting point, direction, and speed. The surprising fact is that the same geometric curve can be produced by infinitely many different parametrizations, each describing a different "trip" along the same road — and choosing the right one can make a hard problem trivial.

The Core Insight

Stop thinking of a curve as a static relationship between $x$ and $y$ . Instead, picture a point in motion: at each "instant" $t$ , the moving point sits at coordinates $(f (t), g (t))$ . As time flows, the point sweeps out a path. The parameter $t$ acts like a clock controlling a pen — the curve is the trail, and $t$ tells you when and where the pen was at every moment. This single shift turns geometry into kinematics and unlocks curves no single-valued function can express.

Formal Statement

Parametric Equations of a Plane Curve

A parametric curve in the $x y$ -plane is given by a pair of functions

x = f (t), y = g (t), t \in [a, b]

⟳This is a definition, not a derived result —

t

is the independent parameter, and

f, g

map each value of

t

to a coordinate pair.

where:

$t$ = the parameter, an independent real variable (often interpreted as time);
$[a, b]$ = the parameter interval over which $t$ ranges;
$f, g$ = real-valued functions defining the horizontal and vertical positions;
the point $(f (t), g (t))$ = the location of the moving particle at parameter value $t$ .

The collection of all such points is the trace (the geometric curve). The function $t \mapsto (f (t), g (t))$ together with its direction of increasing $t$ is the parametrization (the trip).

Orientation

The orientation (or direction) of a parametric curve is the order in which points are traced as $t$ increases from $a$ to $b$ . Two parametrizations producing the same trace may have opposite orientations.

A standard parametrization of the circle of radius $r$ centred at the origin is:

x = r cos t, y = r sin t, t \in [0, 2 π]

⟳From the right-triangle definitions

cos t = x / r

and

sin t = y / r

; verify by checking

x^{2} + y^{2} = r^{2} cos^{2} t + r^{2} sin^{2} t = r^{2}

To recover a Cartesian relation, eliminate the parameter using the Pythagorean identity:

x^{2} + y^{2} = r^{2} cos^{2} t + r^{2} sin^{2} t = r^{2} (cos^{2} t + sin^{2} t) = r^{2}

⟳Square both equations in [2] and add, then apply

cos^{2} t + sin^{2} t = 1

Slope of a Parametric Curve

If $x = f (t)$ and $y = g (t)$ are differentiable and $\frac{d x}{d t} \neq = 0$ , then the slope of the tangent line to the curve is

\frac{d y}{d x} = \frac{d y / d t}{d x / d t} = \frac{g ^{'} ( t )}{f ^{'} ( t )}

⟳Apply the chain rule

\frac{d y}{d t} = \frac{d y}{d x} \cdot \frac{d x}{d t}

and solve for

d y / d x

The curve has a horizontal tangent where $d y / d t = 0$ (and $d x / d t \neq = 0$ ), and a vertical tangent where $d x / d t = 0$ (and $d y / d t \neq = 0$ ).

The arc length of a smooth parametric curve traced once over $[a, b]$ is

L = \int_{a}^{b} (\frac{d x}{d t})^{2} + (\frac{d y}{d t})^{2} d t

⟳A small step has length

d s = (d x)^{2} + (d y)^{2}

; factor out

d t

to get

d s = (d x / d t)^{2} + (d y / d t)^{2} d t

and integrate.

Why It Holds / Derivation

The Tangent Slope Formula

We derive equation [4], the slope of a parametric curve.

Step 1. Suppose that, near a point of interest, the curve can locally be written as

y = F (x)

for some differentiable function

F

(guaranteed when

d x / d t \neq = 0

, so that

x

moves monotonically there). Step 2. Since

y

depends on

x

and

x

depends on

t

, the chain rule gives

\frac{d y}{d t} = \frac{d y}{d x} \cdot \frac{d x}{d t} .

Step 3. This expresses how fast the vertical position changes as a product of the slope and how fast the horizontal position changes. Step 4. Provided

\frac{d x}{d t} \neq = 0

, divide both sides by it:

\frac{d y}{d x} = \frac{d y / d t}{d x / d t} .

Step 5. Writing

d y / d t = g^{'} (t)

and

d x / d t = f^{'} (t)

yields equation [4]. The slope is a ratio of two velocities, not a derivative taken directly with respect to

t

■

The Arc Length Formula

Step 1. Partition

[a, b]

into small subintervals of width

Δ t

. Over one subinterval the point moves from

(f (t), g (t))

(f (t + Δ t), g (t + Δ t))

. Step 2. Approximate this short piece by a straight segment. By the distance formula its length is

Δ s \approx (Δ x)^{2} + (Δ y)^{2} .

Step 3. Factor

Δ t

out of the radical:

Δ s \approx (\frac{Δ x}{Δ t})^{2} + (\frac{Δ y}{Δ t})^{2} Δ t .

Step 4. Sum over all subintervals and let

Δ t \to 0

. The Riemann sum becomes the integral [5], with

Δ x /Δ t \to d x / d t

and

Δ y /Δ t \to d y / d t

■

Leonhard Euler | 1707–1783

Though analytic geometry began with Descartes and Fermat, it was Leonhard Euler who systematically developed the calculus of curves expressed through a parameter, especially in his Introductio in analysin infinitorum (1748) and his studies of mechanics. Euler treated the coordinates of a moving point as functions of a single variable and freely differentiated and integrated them — laying the groundwork for the modern parametric treatment of curves, the cycloid, and the geometry of motion. His notation and habit of thinking of curves as paths of motion remain the standard pedagogical approach today.

Building Intuition

Imagine driving a car along a winding road at night with the headlights off, but with a GPS that records your position every second. The road is the trace — the fixed geometric curve. The log of timestamped positions is the parametrization. Two different drivers on the same road produce the same trace but different logs: one might speed up on the straights, another might reverse over a section, a third might go the opposite direction entirely.

The function $y = F (x)$ is like a road that never crosses itself and never doubles back east-to-west — a strong, artificial restriction. Parametric equations remove the restriction by handing control to the clock $t$ . This is why a circle, which fails the vertical line test, parametrizes effortlessly: the moving point simply goes around, no contradiction. And the slope $d y / d x = (d y / d t) / (d x / d t)$ is just rate of climb divided by rate of forward progress — exactly how steep the road feels.

A parametric curve $(\cos t,\ \sin 2t)$: a self-intersecting figure-eight that no $y=F(x)$ could describe

Worked Examples

Eliminating the Parameter — A Projectile's Path

Given: A ball is launched from the origin with horizontal speed

20 m/s

and initial vertical speed

30 m/s

. Using

g = 10 m/s^{2}

, its position is

x = 20 t, y = 30 t - 5 t^{2} (t in seconds) .

Find: (a) the Cartesian equation

y = F (x)

of the trajectory; (b) the maximum height. Step 1 — Solve for $t$ from the $x$ -equation:

t = \frac{x}{20} .

Step 2 — Substitute into the $y$ -equation:

y = 30 (\frac{x}{20}) - 5 (\frac{x}{20})^{2} = \frac{3 x}{2} - \frac{5 x ^{2}}{400} = \frac{3 x}{2} - \frac{x ^{2}}{80} .

This is a downward parabola — the trace of the projectile.

Step 3 — Max height via the parametric vertical velocity:

\frac{d y}{d t} = 30 - 10 t = 0 \Rightarrow t = 3 s

. Step 4 — Evaluate height:

y (3) = 30 (3) - 5 (9) = 90 - 45 = 45 m

y = \frac{3 x}{2} - \frac{x ^{2}}{80}, y_{m a x} = 45 m at t = 3 s

Tangent Line and Arc Length of a Circle Arc

Given: The curve

x = 4 cos t

y = 4 sin t

for

t \in [0, \frac{π}{2}]

(a quarter circle of radius

4

). Find: (a) the slope at

t = \frac{π}{4}

; (b) the arc length of the quarter. Step 1 — Derivatives:

\frac{d x}{d t} = - 4 sin t

\frac{d y}{d t} = 4 cos t

. Step 2 — Slope from [4]:

\frac{d y}{d x} = \frac{4 cos t}{- 4 sin t} = - cot t .

t = \frac{π}{4}

- cot (\frac{π}{4}) = - 1

. Step 3 — Arc length integrand from [5]:

(- 4 sin t)^{2} + (4 cos t)^{2} = 16 sin^{2} t + 16 cos^{2} t = 16 = 4.

Step 4 — Integrate:

L = \int_{0}^{π /2} 4 d t = 4 \cdot \frac{π}{2} = 2 π .

(Check: full circumference $= 2 π r = 8 π$ ; a quarter is $2 π$ . ✓)

slope = - 1, L = 2 π \approx 6.28 units

Common Misconceptions

Pitfall 1 — Treating $d y / d x$ as $d y / d t$ over $d t$ wrongly. The slope is

\frac{d y / d t}{d x / d t}

, not

d y / d t

alone and not

\frac{d}{d t} (y / x)

. You must divide the two separate

t

-derivatives. Pitfall 2 — Forgetting the parameter interval changes the curve.

x = cos t, y = sin t

over

[0, 2 π]

is a full circle; over

[0, π]

it is only the upper semicircle. The interval is part of the definition. Pitfall 3 — Assuming eliminating $t$ gives the whole story. When you convert

x = t^{2}

y = t^{4}

into

y = x^{2}

, the Cartesian form

y = x^{2}

allows all

x

, but the parametrization only produces

x = t^{2} \geq 0

— the curve is only the right half of the parabola. Pitfall 4 — Ignoring orientation.

x = cos t, y = sin t

goes counterclockwise;

x = cos t, y = - sin t

traces the same circle clockwise. Same trace, different parametrization.

Same Trace, Wildly Different Parametrizations

Consider these three parametrizations, all over suitable intervals, of pieces of the line $y = x$ :

$x = t, y = t$ — moves at constant speed, rightward.
$x = t^{3}, y = t^{3}$ — same line, but the point accelerates; it covers ground slowly near $t = 0$ and rapidly for large $∣ t ∣$ .
$x = - t, y = - t$ — same line, opposite direction.

All three have the identical trace (the line $y = x$ ), yet their parametric "trips" differ in speed and direction. This proves that a parametrization carries strictly more information than the trace: the curve does not determine its parametrization. Conversely, $x = t^{2}, y = t^{2}$ traces only the ray $y = x$ with $x \geq 0$ — a different trace entirely, despite the deceptively similar equation.

Connections

Parametric equations sit at the crossroads of several big ideas:

They generalize ordinary functions: every $y = F (x)$ is the trivial parametrization $x = t, y = F (t)$ . See Functions and Their Graphs.
They are the natural language of Vectors in the Plane: a parametrization is exactly a vector-valued function $r (t) = ⟨ f (t), g (t)⟩$ , with velocity $r^{'} (t)$ .
The slope and arc length formulas rely entirely on The Chain Rule and the Definite Integral.
Polar Coordinates are a special parametric system: $x = r (θ) cos θ, y = r (θ) sin θ$ parametrizes polar curves by the angle $θ$ .
In physics they realize Projectile Motion and orbital mechanics directly, since position as a function of time is a parametrization.

Real-World Applications

Projectile and ballistic motion — gun trajectories, basketball shots, and water-fountain arcs are naturally given as $x (t), y (t)$ , with the parabola emerging only after eliminating time.
Computer graphics & CAD — Bézier curves and B-splines that define fonts (TrueType), car body panels, and animation paths are parametric polynomials $r (t)$ controlled by anchor points.
Orbital mechanics — planetary and satellite positions are parametrized by time; Kepler's equation relates the parameter (mean anomaly) to position along an elliptical orbit.
Robotics & CNC machining — a tool head follows a programmed path $(x (t), y (t), z (t))$ ; the parameter governs feed rate, controlling cutting speed.
Cycloids and gear design — the curve traced by a point on a rolling wheel, $x = r (t - sin t), y = r (1 - cos t)$ , defines the cycloidal gear teeth used in clocks and precise mechanisms.
Roller coaster and track design — engineers specify track centerlines parametrically to control curvature and acceleration smoothly.
Lissajous figures in electronics — oscilloscope patterns $x = A sin (a t + δ), y = B sin (b t)$ reveal frequency ratios and phase between two AC signals.

Historical Development

c. 1630s — Pierre de Fermat and René Descartes independently found analytic geometry, linking curves to equations in $x$ and $y$ — the soil from which parametric thinking grew.
1637 — Descartes' La Géométrie published as an appendix to the Discourse on Method, establishing coordinate methods.
1658–1659 — Christiaan Huygens studies the cycloid, proving it is the tautochrone (the curve of equal descent time) — a result naturally expressed parametrically.
1696 — Johann Bernoulli poses the brachistochrone problem; the solution, the cycloid, is again a parametric curve, igniting the calculus of variations.
1687 — Isaac Newton's Principia treats planetary orbits as paths traced over time, an implicitly parametric kinematic view.
1748 — Leonhard Euler's Introductio in analysin infinitorum systematically uses a single auxiliary variable to describe curves, formalizing the parametric method.
1750s–1780s — Euler and Lagrange develop mechanics in which position-versus-time is parametrization; arc length and curvature formulas mature.
1815 — Jules Antoine Lissajous (born; work c. 1857) later studies the harmonic curves bearing his name, vivid parametric figures.
1962 — Pierre Bézier (and Paul de Casteljau independently) develop parametric Bézier curves at Renault and Citroën, launching modern computer-aided geometric design.

Subtleties & Limits

Several subtle issues lie just beyond the introductory treatment:

Smoothness vs. corners. A curve can be parametrized by differentiable functions yet still show a sharp cusp — for instance the cycloid has cusps where $d x / d t = d y / d t = 0$ simultaneously. Such points need careful handling because formula [4] becomes the indeterminate $0/0$ . Distinguishing genuine geometric corners from mere "stalls" in the parametrization is a real subtlety.
Reparametrization invariance. Quantities like arc length and curvature are properties of the trace and should not depend on the parametrization; proving this invariance rigorously requires the change-of-variables theorem and the notion of an arc-length parametrization (where $∣ r^{'} (s) ∣ = 1$ ). Constructing that natural parametrization explicitly is often impossible in closed form, even for simple-looking curves like the ellipse, whose arc length leads to the non-elementary elliptic integrals — a famous "unsolvable in elementary terms" problem.
Space-filling curves. Continuous parametrizations can be pathological: in 1890 Giuseppe Peano exhibited a continuous map of an interval onto an entire square — a "curve" with no well-defined tangent anywhere — forcing mathematicians to define precisely what smoothness and rectifiability mean.