22.2. THE DERIVATIVE 465

It follows easily the last term at the end is o(h,k, l) and so the derivative of this function isthe linear transformation coming from multiplication by the matrix

((z2 + y

),x,2zx

)and

so this is the derivative. It follows from this and the description of the derivative in termsof partial derivatives that

∂ f∂x

(x,y,z) = z2 + y,∂ f∂y

(x,y,z) = x,∂ f∂ z

(x,y,z) = 2xz.

Of course you could compute these partial derivatives directly.Given a function of many variables, how can you tell if it is differentiable? In other

words, when you make the linear approximation, how can you tell easily that what is leftover is o(v). Sometimes you have to go directly to the definition and verify it is differen-tiable from the definition. For example, here is an interesting example of a function of onevariable.

Example 22.2.9 Let f (x) ={

x2 sin( 1

x

)if x ̸= 0

0 if x = 0. Find D f (0).

f (h)− f (0) = 0h+h2 sin(

1h

)= o(h) ,

and so D f (0) = 0. If you find the derivative for x ̸= 0, it is totally useless information ifwhat you want is D f (0). This is because the derivative turns out to be discontinuous. Tryit. Find the derivative for x ̸= 0 and try to obtain D f (0) from it. You see, in this exampleyou had to revert to the definition to find the derivative.

It isn’t really too hard to use the definition even for more ordinary examples.

Example 22.2.10 Let f (x,y) =(

x2y+ y2

y3x

). Find Df (1,2).

First of all, note that the thing you are after is a 2×2 matrix.

f (1,2) =(

68

).

Then

f (1+h1,2+h2)−f (1,2) =(

(1+h1)2 (2+h2)+(2+h2)

2

(2+h2)3 (1+h1)

)−(

68

)after some simplification,

=

(4 58 12

)(h1h2

)+

(2h1h2 +2h2

1 +h21h2 +h2

212h1h2 +6h2

2 +6h22h1 +h3

2 +h32h1

)

=

(4 58 12

)(h1h2

)+o(h) .

Therefore, the matrix of the derivative is(

4 58 12

). You let the o(h) terms be the higher

order terms, polynomials in the components of h which have higher degree than 1.

22.2. THE DERIVATIVE 465It follows easily the last term at the end is 0 (h,k,/) and so the derivative of this function isthe linear transformation coming from multiplication by the matrix (2 + y) x, 2zx) andso this is the derivative. It follows from this and the description of the derivative in termsof partial derivatives thatof _2,, of __ of _Ox (x,y, 2) =Z +y, oy (x,y, 2) =X, Oz (x,y, 2) ~~ 2Xxz.Of course you could compute these partial derivatives directly.Given a function of many variables, how can you tell if it is differentiable? In otherwords, when you make the linear approximation, how can you tell easily that what is leftover is o(v). Sometimes you have to go directly to the definition and verify it is differen-tiable from the definition. For example, here is an interesting example of a function of onevariable.Example 22.2.9 Let f (x) = { nants) PXF0 rind D f (0).f (A) — f (0) = Oh+A? sin (=) =o(h),and so Df (0) = 0. If you find the derivative for x 4 0, it is totally useless information ifwhat you want is Df (0). This is because the derivative turns out to be discontinuous. Tryit. Find the derivative for x 4 0 and try to obtain Df (0) from it. You see, in this exampleyou had to revert to the definition to find the derivative.It isn’t really too hard to use the definition even for more ordinary examples.xyExample 22.2.10 Let f (x,y) = ( y2id ). Fina Df (1,2)First of all, note that the thing you are after is a 2 x 2 matrix.fi.a=(§ ).2 2Fivehaehey-seiay= (MEE! (8)Thenafter some simplification,(4°55 mn), 2hy hy + 2hj + Aj hy +3~\ sg 12 hy 12h hy + 6h5 + Oh5h, +h3 + h3hy=(§ 5 ) ( n ) rotn).Therefore, the matrix of the derivative is . You let the o (h) terms be the higher4 58 12order terms, polynomials in the components of h which have higher degree than 1.