This page assumes you have seen none of the notation in Parametric surfaces — tangent planes, surface area. We build every symbol from the ground up, in an order where each piece leans only on the pieces before it.
Recall Two everyday tools we will lean on (quick review)
The x–y–z axes. Three number lines, all meeting at one corner (the origin) and all at right angles to each other. The first, x, runs right; the second, y, runs forward; the third, z, runs up. Any place in space is reached by walking some amount along each. That's all "3D coordinates" means.
Sine and cosine of an angle θ. Draw a right-angled triangle. For one of its non-right angles θ: cosθ is (side next toθ) ÷ (longest side), and sinθ is (side oppositeθ) ÷ (longest side). Two facts we use: cos0∘=1 and sin0∘=0 (a squashed-flat triangle), while cos90∘=0 and sin90∘=1 (a fully "opened-out" right angle). Read cosθ as "how aligned" two directions are and sinθ as "how spread apart" — that intuition is all we need below.
Before anything else: a point in space needs three numbers — how far right (x), how far forward (y), how far up (z).
The picture: start at the origin (the corner where the three axes meet) and step x along the first axis, y along the second, z along the third. The tip is your point (x,y,z); the whole journey-arrow is ⟨x,y,z⟩.
Figure s01 — A point in 3D. The blue segment steps x to the right along the x-axis, the yellow segment steps y forward along the y-axis, the green segment steps z up along the z-axis; the red dot is the finished point (3,2,2.5) and the thin white arrow from the origin to it is the vector ⟨3,2,2.5⟩. Alt text: three coloured steps building up to one red point in a 3D coordinate box.
Why the topic needs it: a surface lives in 3D, so every point on it, and every arrow we draw on it, is one of these triples.
Here is the single most important idea to picture.
The picture: on the left, a flat piece of graph paper — the (u,v) plane, a region we call D. On the right, 3D space. The map r picks up each dot of the flat paper and glues it somewhere in space. The whole sheet, once glued, is the surface.
Figure s02 — Left: the flat parameter region D, ruled with blue u-lines and yellow v-lines; every crossing is one input pair (u,v). Right: the map r lifts and bends that grid into 3D, so the straight blue and yellow lines become curved lines drawn on the surface S. Alt text: a flat coloured grid on the left turning into a wavy coloured grid on a surface on the right.
SymbolD
the flat region on the parameter paper that we are allowed to feed in. Symbol∈ means "belongs to", so (u,v)∈D reads "the pair (u,v) is a point of the region D."
On plain graph paper there are two families of lines: the horizontal ones (fix v, let u change) and the vertical ones (fix u, let v change). When we glue the paper into space, those straight grid lines become curved lines on the surface.
The picture: latitude and longitude lines on a globe. Longitude lines are one family, latitude lines the other; every point sits where one of each crosses.
Why the topic needs it: the directions these grid curves head in are the two arrows from which the entire theory (planes and area) is built. That is the next section.
If you walk along a curve, at each instant you are pointing somewhere. That instantaneous direction-and-speed is the tangent vector. We build this notion right here, from scratch, before using it.
The picture: a bead sliding along a bent wire; the little arrow taped to the bead, always aimed straight ahead along the wire, never sideways off it. That arrow lies flat against the curve.
Figure s03 — The blue path is the curve r(t); the red dot is the bead at t0; the yellow arrow is its tangent (velocity) vector, aimed straight along the path in the direction of travel. Alt text: a bead on a curved blue path with a yellow arrow pointing along the curve.
This idea is developed further in Tangent Lines and Velocity Vectors, and the length of such a moving arrow, added up, is Arc Length — but you need nothing from those pages to follow this one. Why the topic needs it: the two grid curves each have a tangent arrow, and those two arrows will pin down the tangent plane.
The limit above — "chord over step, as the step shrinks to zero" — is exactly the tool called the derivative.
Now, our map r(u,v) has two dials. When we differentiate we must say which dial we're nudging while holding the other still. That is exactly what the curly ∂ marks.
Why ∂ and not the ordinary d? The plain dtd is for a function of one dial (like the curve above). When there are two or more, we need to declare which one moves — the curly ∂ is that declaration. It answers the precise question "rate of change in the u-direction alone."
Two arrows starting at the same point, as long as they don't lie along the same line, sweep out a flat sheet: a plane. The u-tangent and v-tangent do exactly this.
Why the topic needs it: "a point lies in the plane" will be phrased as "an arrow inside the plane is perpendicular to the normal," and perpendicular = dot product zero. That single equation is the tangent-plane equation.
This is the tool the whole topic hinges on. It takes two arrows and produces a third arrow.
The picture below shows all three gifts at once: the flat parallelogram, the arrow rising perpendicular from it, and the right-hand twist from a to b.
Figure s04 — The blue arrow a and yellow arrow b fence off the green parallelogram; its area equals the length ∣a×b∣. The red arrow is a×b, rising perpendicular to the green sheet — its direction set by curling the right hand from a toward b. Alt text: two arrows spanning a shaded parallelogram with a third arrow standing up perpendicular to it.
Full details of computing it live in Cross Product. Note the length bars vs no bars: ru×rv (no bars) is the arrow — used as the normal; ∣ru×rv∣ (with bars) is its length — used as the area factor.
The final symbol: to get total area we chop the surface into countless tiny patches, measure each, and add them all up. That infinite sum is an integral.
Why the topic needs it: the area of the surface S (the glued sheet from §1) is
A(S)=∬D∣ru×rv∣dA=∬D∣ru×rv∣dudv
— every symbol is now something you have met: A(S) (the area of the finished surface S), ∬ (add up patches), D (parameter region), ∣×∣ (area of one stretched patch), dA=dudv (size of one tile on the flat paper). And thanks to §7 the integrand is nonzero wherever the surface is regular.