One point, one line, a perfect balance
In the previous guide you sliced a cone and watched the parabola fall out as the special cut, the one made by a plane tilted at exactly the same slope as the cone's side. That picture is true and beautiful, but it hides the curve's most useful secret. There is a second way to define a parabola that uses no cone at all — only a single point and a single line drawn flat on the page. Everything else about the curve flows from it.
Fix a point F, call it the focus, and a line d not through F, call it the directrix. Now play a game: which points P of the plane are exactly as far from F as they are from the line d? The distance to the line means the perpendicular distance — the shortest hop straight across to d. Demand |PF| = (distance from P to d), and the set of all winning P is the parabola. This is the focus-directrix property, and it is not a fact you prove about the parabola; for this rung it is the parabola's definition.
Because the rule speaks only of distances, the parabola is a textbook locus — the path traced by every point that obeys one distance condition, exactly as a circle is the locus of points a fixed distance from a centre. The circle balances one distance against a constant; the parabola balances one distance against another. That tiny change of partner is the whole difference, and it bends the closed circle open into an endless, sweeping curve.
From the balance to an equation
Now drop the picture onto coordinates and watch the definition turn into algebra. Place the focus at F = (0, p) on the y-axis and the directrix as the horizontal line y = -p, one step below the origin, with p > 0. The whole figure is now symmetric about the y-axis, which will keep the bookkeeping clean. Take any point P = (x, y) on the curve and write down the two distances the definition insists must be equal.
The distance from P to the focus is the ordinary distance formula: |PF| = sqrt(x^2 + (y - p)^2). The distance from P straight down to the line y = -p is just the vertical gap, |y - (-p)| = |y + p|. Setting them equal, squaring both sides to clear the roots, the messy y^2 and p^2 terms cancel in a small miracle, and what survives is startlingly simple. You are left with x^2 = 4 p y, or solved for y, the familiar y = x^2 / (4p).
Focus F = (0, p), directrix y = -p, p > 0
distance to focus = distance to directrix
sqrt(x^2 + (y - p)^2) = | y + p |
square both sides:
x^2 + y^2 - 2 p y + p^2 = y^2 + 2 p y + p^2
x^2 = 4 p y
=> y = x^2 / (4 p)So the parabola you doodled in algebra class — the graph of any y = a x^2 — was a focus-directrix locus all along. Matching a = 1 / (4p) tells you exactly where its hidden focus sits: at height p = 1 / (4a) above the vertex, with the directrix the same distance below. A wide, lazy parabola has its focus far away; a narrow, steep one hugs its focus close. The single number p, the focus-to-vertex distance, controls the whole shape.
Vertex, axis, and how wide it opens
Every parabola has one special point where it turns around, the vertex. In our setup it is the origin, and you can see why from the definition: the vertex is the point on the curve closest to the directrix, which must sit exactly halfway between focus and directrix. Halfway between F = (0, p) and the line y = -p is (0, 0). The line through the focus and vertex — here the y-axis — is the axis of symmetry: fold the page along it and the two arms of the curve land perfectly on each other.
There is one more measurement worth naming, because it tells you the curve's width near the focus at a glance. Draw the chord through the focus that runs perpendicular to the axis — parallel to the directrix. Its length is the latus rectum ('straight side'). Plug y = p into x^2 = 4 p y and you get x^2 = 4 p^2, so x = ±2p: the chord stretches from (-2p, p) to (2p, p), a total length of 4p — exactly four times the focus-to-vertex distance. That clean relationship is why the latus rectum is the surveyor's favourite handle on a parabola.
- Spot the focus and directrix (or read p from y = a x^2 via p = 1 / (4a)).
- The vertex is the midpoint between focus and directrix; the axis is the line through focus and vertex.
- The latus rectum has length 4p — mark points 2p to each side of the focus to fix the curve's width.
- Sketch the smooth arc through the vertex and those two endpoints, opening away from the directrix.
The mirror that points at the sky
Now for the property that turns this curve into a piece of engineering. Imagine the inside of the parabola silvered into a mirror. A ray of light arriving parallel to the axis — straight down from a star, say — strikes the curve and reflects. The reflective property says that every such parallel ray, no matter where it hits, bounces to pass through the one focus F. A whole shower of parallel light is gathered to a single bright point.
Run the light the other way and the same fact reads as a broadcast: a lamp placed at the focus sends out rays that, after one bounce, all travel parallel to the axis — a tight beam that does not spread. This is the engineering behind a satellite dish (incoming signal squeezed onto the receiver at the focus), a car headlight and a torch (bulb at the focus, beam thrown straight ahead), and a reflecting telescope. It is the single most lucrative consequence of the focus-directrix balance.
Honest edges and what comes next
A few traps worth naming. First, a parabola is not half of any ellipse and is not a U made of two circular arcs — it has its own shape, and the give-away is that it never closes and its arms keep widening forever without ever becoming parallel. Second, the focus is inside the bowl and the directrix is outside, never on the curve; if you ever find F sitting on the directrix, you have not drawn a parabola at all, only a single line. Third, 'p' in our formula means the focus-to-vertex distance, which is half the focus-to-directrix distance — different textbooks split that factor of two differently, so always check which one an equation is using.
Step back and notice the shape of what we did. We never needed the cone to define the parabola — the focus and directrix did all the work — yet the cone-slice and the focus-directrix curve are genuinely the same object, a fact made airtight by the Dandelin-sphere argument waiting at the end of this rung. Keeping the two pictures, the plane section and the locus, side by side is exactly how the rest of the conics will unfold.
The next guide keeps the same locus idea but changes the recipe in one decisive way: instead of balancing a point against a line, it balances a point against another point — two foci at once. Hold the two foci together and add a constant, and out come the ellipse and the hyperbola, the parabola's close cousins. The parabola, you will then see, is the delicate in-between case where one of those two foci has drifted infinitely far away.