Calculus Volume 3 · Chapter 4 · Section 4.3
A surface has a slope in every direction. Freeze all variables but one, and the ordinary derivative gives the slope along that one.
Colour key surface along : along : tangent, secant key idea · 3D figures turn when dragged · page links open the textbook
A hill , drawn as a contour map (Figure 4.22). You stand at , height 2.
Step east: the ground drops per step. Step north: it drops 1.
A surface has no single slope. Name a direction first. The two easiest directions give the two partial derivatives.
In , the terms , and are constants, so they vanish; is times .
Cut the hill with the plane . The cut is a curve, ; its slope at is .
Cut with instead: , slope (p333).
A partial derivative is the slope of a slice.
Only moves.
, , : move one of , freeze the other two.
Rise between two level curves over the run between them: a secant slope.
The limit is the one-variable derivative of ; that is why every familiar rule applies.
Mixed partials measure how the -slope changes as you step in — and that is the same as how the -slope changes as you step in .
A partial differential equation links an unknown function of several variables to its partial derivatives. On a wire, has the solutions ((4.22)).
To check a solution, compute each partial and substitute; nothing more.
Try: the saddle at — along the slice is a valley, along a ridge, both flat at the centre. The slope depends on the point and on the direction; the 2D graph below is the slice laid flat.
Try: the next level curve (the book's estimate), then shrink . A contour map gives a secant; only the limit gives the tangent.
Try: the third function with the corner at the origin. x-first and y-first disagree there: the mixed partials exist but are not continuous, so Clairaut does not apply.
Try: mode 3 against mode 1. A wiggly profile has a large , so makes it fade fast.
| Mistake | Result | Fix |
|---|---|---|
| Differentiating the frozen variable | of written as | a term without is a constant: 0 |
| Dropping the inner derivative | chain rule: times | |
| read as "y first" | order swapped when it matters | subscripts left to right; right to left |
| Contour estimate taken as exact | for | it is a secant; differentiate for the tangent |
| Clairaut used without continuity | "proved" for the counterexample | check the hypothesis (p338) |
| Quotient rule in the wrong order | sign error in Example 4.18a |
Exercises from p344; the book's key: p862, p867–868.
| # | task | answer |
|---|---|---|
| 4.12 | , by limits | , |
| 4.13 | , | |
| 4.14 | , at from contours | ; exact |
| 4.15 | , , | |
| 4.16 | ; , in the key | |
| 4.17 | ; ; | |
| 4.18 | solves | , (not in the key) |
| # | task | answer | key step |
|---|---|---|---|
| 113 | , , by limits | ||
| 115, 117 | signs from the bowl's graph | ; | slope of the slice |
| 119 | , | is a constant | |
| 121 | , | chain rule | |
| 125 | , | inner derivatives 2, −1 | |
| 127 | at | ||
| 129 | of at | — | |
| 133 | cylinder | ; ; | circumference × height; base area |
| 135 | , | ||
| 143 | , | one variable at a time | |
| 145 | , | , | , then |
| 147 | , | ||
| 151 | solves | the key omits the minus sign | |
| 159 | at | , | |
| 161 | at | , | , |
Fix and let . Then (p331)
A partial derivative is an ordinary derivative of a one-variable function, so the power, product, quotient and chain rules come for free.
Take , . From the limits, and , so
Away from the origin both mixed partials equal
which has no limit at . Clairaut's continuity hypothesis is exactly what fails.
For : and , so (Figure 4.24).
With , at mode 1 keeps of its height; mode 3 keeps about . Decay grows with : after a short time only the smoothest mode is left — the reason Kelvin needed only the first terms.
The two tangent lines at have directions and . They span a plane:
If the surface is smooth, this is its tangent plane — §4.4.
Access for free at openstax.org · CC BY-NC-SA 4.0
Derivatives of a Function of Two Variablesbridgedefinition
Hill ; stand at , height 2.
Step east ( fixed): ground drops per step.
Step north ( fixed): drops per step.
Derivatives of a Function of Two VariablesalgebraExample 4.14 · Checkpoint 4.12
Same with :
By the limit:
Derivatives of a Function of Two VariablesalgebraExample 4.15a
: is a number
→ 0; →
: is a number
→ 0; →
Derivatives of a Function of Two VariablesalgebraExample 4.15b · Checkpoint 4.13
Derivatives of a Function of Two Variablesgeometry
Cut the hill with the plane …
… the cut is the curve .
Tangent at : slope .
Derivatives of a Function of Two VariablesbridgeFigure 4.21
: slope
: slope
: slope
Derivatives of a Function of Two Variablesgeometry
Plane : slope
Plane : slope
Derivatives of a Function of Two VariablesbridgeExample 4.16
: at ?
Along :
at , at
Derivatives of a Function of Two VariablesbridgeCheckpoint 4.14
, point
Along : at , at
Exact: →
Functions of More Than Two VariablesalgebraExample 4.17
( fixed)
( fixed)
( fixed)
Functions of More Than Two VariablesalgebraCheckpoint 4.15 · Example 4.18 · Checkpoint 4.16
,
4.18a :
4.18b :
Higher-Order Partial Derivativesalgebra
mixed:
Higher-Order Partial DerivativesalgebraExample 4.19 · Checkpoint 4.17
Higher-Order Partial DerivativesbridgeTheorem 4.5 · Clairaut
first: how changes stepping up in .
first: how changes stepping across in .
Both come from the same corner sum .
Higher-Order Partial Derivativesalgebraproof · Theorem 4.5
So . As both points → ; continuity ⇒ .
Higher-Order Partial Derivativesalgebracounterexample
,
⇒
⇒
The mixed partials have no limit at the origin.
Partial Differential Equationsalgebra(4.17)–(4.19)
| Equation | Name | Unknown |
|---|---|---|
| heat equation (2D) | ||
| wave equation (2D) | ||
| Laplace's equation (2D) |
: space · : time · : a constant of the material
Partial Differential EquationsalgebraExample 4.20
, ;
Partial Differential EquationsbridgeCheckpoint 4.18 · Figure 4.23
· peaks shrink
✓
Partial Differential Equationsbridge(4.22) · Figure 4.24
:
⇒ ✓
Partial Differential Equationsbridgestudent project · Kelvin
,
: keeps , keeps
: against
Kelvin needed only the first terms of his series.
§4.3wrap-up
| Objective | You can now |
|---|---|
| 4.3.1 two variables | by the limit or by freezing; read them as slice slopes |
| 4.3.2 more variables | : freeze all but one |
| 4.3.3 higher order | ; when continuous |
| 4.3.4 PDEs | name heat, wave, Laplace; verify a solution by substitution |