22.4. C1 FUNCTIONS 469

=n

∑i=1

∫ 1

0

(∂f

∂xi

(x+

n

∑k=i+1

vkek + tviei

)vi −

∂f

∂xi(x)vi

)dt

+n

∑i=1

∫ 1

0

∂f

∂xi(x)vidt =

n

∑i=1

∂f

∂xi(x)vi

+n

∑i=1

∫ 1

0

(∂f

∂xi

(x+

n

∑k=i+1

vkek + tviei

)− ∂f

∂xi(x)

)vidt =

n

∑i=1

∂f

∂xi(x)vi +o(v)

and this shows f is differentiable at x.Some explanation of the step to the last line is in order. The messy thing at the end is

o(v) because of the continuity of the partial derivatives. To see this, consider one term. ByProposition 16.2.2, ∣∣∣∣∣

∫ 1

0

(∂f

∂xi

(x+

n

∑k=i+1

vkek + tviei

)− ∂f

∂xi(x)

)vidt

∣∣∣∣∣≤ √

p∫ 1

0

∣∣∣∣∣∂f∂xi

(x+

n

∑k=i+1

vkek + tviei

)− ∂f

∂xi(x)

∣∣∣∣∣dt |v|

Thus, dividing by |v| and taking a limit as |v| → 0, this converges to 0 due to continuityof the partial derivatives of f . The messy term is thus a finite sum of o(v) terms and istherefore o(v).

Here is an example to illustrate.

Example 22.4.3 Let f (x,y) =(

x2y+ y2

y3x

). Find Df (x,y).

From Theorem 22.4.2 this function is differentiable because all possible partial deriva-tives are continuous. Thus

Df (x,y) =(

2xy x2 +2yy3 3y2x

).

In particular,

Df (1,2) =(

4 58 12

).

Here is another example.

Example 22.4.4 Let f (x1,x2,x3) =

 x21x2 + x2

2x2x1 + x3

sin(x1x2x3)

. Find Df (x1,x2,x3).

All possible partial derivatives are continuous, so the function is differentiable. Thematrix for this derivative is therefore the following 3×3 matrix 2x1x2 x2

1 +2x2 0x2 x1 1

x2x3 cos(x1x2x3) x1x3 cos(x1x2x3) x1x2 cos(x1x2x3)

Example 22.4.5 Suppose f (x,y,z) = xy+ z2. Find D f (1,2,3).

22.4. C! FUNCTIONS 469-y/ (3 (e+ 3 y nective) vi - oo) dt7]k=i+1fl of _y?f ey.+Y/ Bx Vit = Lae (a) v;+E (52 (= E nevtmes) sf (ae) )) va -y 5 az) vj +o(v)i+land this shows f is differentiable at x.Some explanation of the step to the last line is in order. The messy thing at the end iso(v) because of the continuity of the partial derivatives. To see this, consider one term. ByProposition 16.2.2,\/of Of[ (3 [+3 y VEEk mes) _— ax; ) v;dtk=it+11 n< ve [ Ee (++ y nective) — vd (x)k=i+1dt|v|Thus, dividing by |v| and taking a limit as |v| — 0, this converges to 0 due to continuityof the partial derivatives of f. The messy term is thus a finite sum of o(v) terms and istherefore o(v). WfHere is an example to illustrate.2 2Example 22.4.3 Let f (x,y) = ( * oe ) Find Df (x,y).From Theorem 22.4.2 this function is differentiable because all possible partial deriva-tives are continuous. ThusQxy x7 42Df (x,y) = ( es 32 ).In particular,4 5priy=(4 §).Here is another example.XTX? +25Example 22.4.4 Let f (x1,x2,x3) = xXx; +.x3 . Find Df (x1,x2,x3).sin (x1x2%3)All possible partial derivatives are continuous, so the function is differentiable. Thematrix for this derivative is therefore the following 3 x 3 matrix2x1 X2 xt + 2x2 0X2 XxX] 1X2X3 COS (x142.x3) X1X3 COS (x1.%2x3) X 1X2 COS (x1 X2%3)Example 22.4.5 Suppose f (x,y,z) =xy +2. Find Df (1,2,3).