1412 functions of several variables

14.2 Limits and Continuity of Multivariable Functions

Just as with a function of a single variable, before we can investigate

differentiation we must consider limits and continuity. Although limits
of functions of two (or more) variables present some additional com-
You may want to spend some time now plications, for the most part the definitions and properties of limits for
reviewing Chapter 1 to refamiliarize
these functions closely resemble the definitions and properties given in
yourself with the main ideas and results
for limits of functions of one variable — Chapter 1 for functions of a single variable.
and rework a few limit problems.

Limits of Functions of Two Variables

One intuitive interpretation of the statement lim f ( x ) = L was that
x→a
“we can guarantee that f ( x ) is as close to L as we like by restricting x
to be sufficiently close (but not necessarily equal) to a.” The precise
definition of this concept said:

Given any e > 0 there exists a δ > 0 so that:

0 < | x − a| < δ ⇒ | f ( x ) − L| < e

When working with one variable, x could approach a in one of two

ways: from the left or from the right. Accordingly, we defined left
(x → a− ) and right (x → a+ ) limits and determined that when both of
these limits existed and were equal to the number L, the two-sided limit
existed and was equal to L. With functions of two variables, you can
approach a point ( x, y) not only from an infinite number of directions,
but along an even greater infinite number of paths.

x2 · y
Example 1. Investigate the behavior of f ( x, y) = near (0, 0).
x 2 + y2

Solution. Note first that f ( x, y) is undefined at (0, 0), as its formula

takes on the form “ 00 ”; this is not an issue if you are only concerned
about the behavior of the function near (0, 0). Constructing a table:
y 0.10 0.050 0.039 0.026 0.014 0.004 0.000 0.004 0.014 0.026 0.039 0.050
0.08 0.049 0.040 0.029 0.016 0.005 0.000 0.005 0.016 0.029 0.040 0.049
0.06 0.044 0.038 0.030 0.018 0.006 0.000 0.006 0.018 0.030 0.038 0.044
0.04 0.034 0.032 0.028 0.020 0.008 0.000 0.008 0.020 0.028 0.032 0.034
0.02 0.019 0.019 0.018 0.016 0.010 0.000 0.010 0.016 0.018 0.019 0.019
0.00 0.000 0.000 0.000 0.000 0.000 UND 0.000 0.000 0.000 0.000 0.000
-0.02 -0.019 -0.019 -0.018 -0.016 -0.010 0.000 -0.010 -0.016 -0.018 -0.019 -0.019
-0.04 -0.034 -0.032 -0.028 -0.020 -0.008 0.000 -0.008 -0.020 -0.028 -0.032 -0.034
-0.06 -0.044 -0.038 -0.030 -0.018 -0.006 0.000 -0.006 -0.018 -0.030 -0.038 -0.044
-0.08 -0.049 -0.040 -0.029 -0.016 -0.005 0.000 -0.005 -0.016 -0.029 -0.040 -0.049
-0.10 -0.050 -0.039 -0.026 -0.014 -0.004 0.000 -0.004 -0.014 -0.026 -0.039 -0.050
-0.10 -0.08 -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 x
 14.2 limits and continuity of multivariable functions 1413

indicates that the values of f ( x, y) generally appear to be getting closer

to 0 as the inputs ( x, y) get nearer to (0, 0).
A graph of this function (see margin) also seems to indicate the
values of f ( x, y) are all reasonably close to 0 for inputs near (0, 0).
Unfortunately, neither a table nor a graph proves that near (0, 0) all of
the function values are close to 0, but using some algebra:

x2 · y x2 x 2 + y2
0≤ = | y | · ≤ | y | · = |y|
x 2 + y2 x 2 + y2 x 2 + y2

As ( x, y) approaches (0, 0) along any path whatsoever, y → 0, so the

inequality above guarantees that f ( x, y) → 0 as well. J

x2
Example 2. Investigate the behavior of g( x, y) = near (0, 0).
x2 + y2

Solution. Again, g( x, y) is undefined at (0, 0), but this is not an issue

if our concern is what happens near (0, 0). If ( x, y) approaches (0, 0)
along the x-axis (where y = 0) we have:

x2 x2 x2
= = =1
x 2 + y2 x2 + 02 x2

for any x 6= 0. Along the y-axis (where x = 0), however:

x2 02 0
= = 2 =0
x 2 + y2 02 + y2 y

for y 6= 0. Because g( x, y) = 1 at infinitely many points arbitrarily

close to (0, 0) and g( x, y) = 0 at infinitely many other points arbitrarily
close to (0, 0), g( x, y) has no single limiting value as ( x, y) → (0, 0).
A contour map for g( x, y) (see margin) shows level curves for many
different levels all approaching (0, 0). (What does this tell you?) J

x 2 − y2
Practice 1. Investigate the behavior of F ( x, y) = near (0, 0).
x 2 + y2

xy
Example 3. Investigate the behavior of G ( x, y) = near (0, 0).
x 2 + y2

Solution. On the x-axis, G ( x, 0) = 0 for x 6= 0. Similarly, on the y-axis,

G (0, y) = 0 for y 6= 0, so 0 seems to be a good candidate for a limit.
Unfortunately, along the line y = x (see margin figure for a graph):

x·x x2 1
G ( x, x ) = = 2 =
x2+x 2 2x 2

for x 6= 0, so G ( x, y) has no single limiting value as ( x, y) → (0, 0). J

1414 functions of several variables

xy
Practice 2. Investigate the behavior of G ( x, y) = along the line
x 2 + y2
y = mx for an arbitrary slope m.

x2 · y
Example 4. Investigate the behavior of f ( x, y) = near (0, 0).
x 4 + y2
Solution. On the x-axis, f ( x, 0) = 0 for x 6= 0; on the y-axis, f (0, y) = 0
for y 6= 0. Along a generic line through the origin y = mx:

x2 · mx mx3 mx
f ( x, mx ) = = = 2 →0
x4 + (mx )2 x 2 ( x 2 + m2 ) x + m2

as x → 0, so f ( x, y) approaches 0 along any straight path approaching

(0, 0). The simplest non-linear curve is a parabola, and along y = x2 :

x2 · x2 x4 1
f ( x, x2 ) = 2
= 4
=
x4 + ( x2 ) 2x 2

Foiled again! (See margin figure for a graph.) J

x2 · y
Practice 3. Investigate the behavior of f ( x, y) = along a generic
x 4 + y2
parabola passing through the origin of the form y = Ax2 .

Clearly you cannot check every possible path to a limiting input

when attempting to determine a limiting value of a function, but the
preceding Examples help show that we need a precise definition of a
limit for this very reason.

Precise Definition of a Limit

The vector notation introduced at the end of Section 14.1 allows us
to state a formal definition of a limit for a function of two variables
that looks remarkably similar to the definition stated in Section 1.4 for
functions of a single variable.

Definition of a Limit: If, given any e > 0, there exists a δ > 0

so that:
0 < kx − ak < δ ⇒ | f (x) − L| < e
we write lim f (x) = L and say that “the limit of f (x), as x
x→a
approaches a, equals L.”

x2 y
Example 5. Show that lim = 0.
( x,y)→(0,0) x 2 + y2
Solution. From Example 1 we know that:

x2 · y x2 x 2 + y2
0≤ = | y | · ≤ | y | · = |y|
x 2 + y2 x 2 + y2 x 2 + y2
 14.2 limits and continuity of multivariable functions 1415

and furthermore we know that:

q q
|y| = y2 ≤ x2 + y2 = kh x, yi − h0, 0ik

Given any number e > 0, let δ = e. Then:

x2 · y
0 < kh x, yi − h0, 0ik < δ = e ⇒ − 0 ≤ |y| < e
x 2 + y2

which is what we need to show, according to the limit definition. J

Example 6. Show that lim x = a and lim y = b.

( x,y)→( a,b) ( x,y)→( a,b)

Solution. Given e > 0, let δ = e and note that:

q q
| x − a| = ( x − a) ≤ ( x − a)2 + (y − b)2 = kh x, yi − h a, bik
2

so that:
0 < kh x, yi − h a, bik < δ = e ⇒ | x − a| < e

as required. The other (very similar) result is left for you to prove. J

As you might suspect, the results of the Main Limit Theorem from
Section 1.2 carry over to limits of functions of two (or more) variables
(with very similar proofs, which we will omit here).

Main Limit Theorem:

If lim f (x) = L and limx→a g(x) = M
x→a

then (a) lim [ f (x) + g(x)] = L + M

x→a

(b) lim [ f (x) − g(x)] = L − M

x→a

(c) lim k · f (x) = k · L

x→a

(d) lim f (x) · g(x) = L · M

x→a

f (x) L
(e) lim = (if M 6= 0)
x→a g(x) M

(f) lim [ f (x)]n = Ln

x→a
q √ When n is an even integer in part (g) of
n n
(g) lim f (x) = L the Main Limit Theorem, we need L ≥ 0
x→a and f (x) ≥ 0 for x near a.

Combining the results of Example 6 with the Main Limit Theorem

allows us to compute a wide variety of limits very quickly.
√
Example 7. Compute lim 5x2 y.
( x,y)→(3,4)
1416 functions of several variables

Solution. Using results from the Main Limit Theorem and Example 6:

2√
 2 r √
lim 5x y = 5· lim x · lim y = 5 · 32 · 4 = 90
( x,y)→(3,4) ( x,y)→(3,4) ( x,y)→(3,4)

which agrees with substituting x = 3 and y = 4 into the expression. J

3x2 + 6xy − 7y2 + 12

Practice 4. Compute lim .
( x,y)→(2,−5) ( x + 2)2 + ( y − 5)2

Continuity
A function of one variable is continuous at x = a if lim f ( x ) = f ( a).
x→a
Graphically, this means that the graph of f is “connected” at the point
( a, f ( a)) and does not have a hole or break there (see margin). The
definition and meaning of continuity for functions of two (or more)
variables is quite similar.

Definition of Continuity: A function f of two (or more) variables

defined at a and for all points near a is continuous at a if:

lim f (x) = f (a)

x→a

Graphically, this definition means that the graph of f does not have
a hole or break at any point of continuity (see margin).
Just as we talked about a function of one variable being continuous
on an interval (or the entire real number line), we can talk about a
function of n variables being continuous on a region R in Rn .

Definition: A function f is continuous on a region R if it is

continuous at each point in R.

Most of the functions of two (or more) variables you will encounter
are continuous either everywhere (for example, at all points ( x, y) in the
plane for a function of two variables) or continuous everywhere except
at a “few” places. The results of Example 6 and the Main Limit Theorem
allow us to show quite easily that any polynomial function, such as
f ( x, y) = 3x3 − 5x2 y + 7xy2 + 9y4 − 17, is continuous everywhere, and
that any rational function is continuous everywhere except at points
where its denominator equals 0.

x2 y
Example 8. Where is f ( x, y) = continuous?
x2 + y2
Solution. Because f ( x, y) is a rational function, it is continuous every-
where except where the denominator is 0, and x2 + y2 = 0 only at (0, 0).
Because f ( x, y) is undefined at (0, 0), it is not continuous there. J
 14.2 limits and continuity of multivariable functions 1417

3x2 + 6xy − 7y2 + 12

Practice 5. Where is g( x, y) = continuous?
( x + 2)2 + ( y − 5)2

x2 y
Example 9. Define F ( x, y) = for ( x, y) 6= (0, 0) with F (0, 0) = 0.
x2+ y2
Where is F ( x, y) continuous?

Solution. At any point other than (0, 0), F ( x, y) is continuous because

it is a rational function with nonzero denominator. At (0, 0), we know
from Example 5 that lim( x,y)→(0,0) F ( x, y) = 0 = F (0, 0), so F ( x, y) is
continuous there as well, hence on all of R2 . J

x2 y
Practice 6. Define G ( x, y) = for ( x, y) 6= (0, 0) with G (0, 0) = 1.
x 2 + y2
Where is G ( x, y) continuous?

As with functions of a single variable, the Main Limit Theorem

allows us to conclude that sums, differences, products and quotients For quotients of continuous functions,
of continuous functions are also continuous. Furthermore, the result the usual caveat about the denominator
being nonzero applies.
from Section 1.3 about continuity of compositions of functions also
extends quite easily (with a similar proof) to situations where the
“inner” function is a function of two (or more) variables.

Composition of Continuous Functions:

If g(x) is continuous at x = a and

f (u) is continuous at u = g(a)
 
then lim f ( g(x)) = f lim g(x) = f ( g(a))
x→a x→a
so f ◦ g(x) = f ( g(x)) is continuous at a.

 
Example 10. Where is f ( x, y) = ln x2 + y2 − 3 continuous?

Solution. The function ln(u) is continuous for u > 0 and the function
x2 + y2 − 3 is continuous on all of R2 so we need:

x 2 + y2 − 3 > 0 ⇒ x 2 + y2 > 3

hence f ( x, y) = ln x2 + y2 − 3 is continuous at all points outside a

√
closed disk of radius 3 centered at (0, 0). J
 
Practice 7. Where is g( x, y) = arcsin x2 + y2 − 3 continuous?

Using Polar Coordinates to Investigate Limits

For functions of two variables, converting from rectangular coordinates
( x, y) to polar coordinates (r, θ ) can sometimes provide greater insight
1418 functions of several variables

into the behavior of a function near the origin. In Example 1, we could

have written:

x2 y (r cos(θ ))2 (r sin(θ )) r3 cos2 (θ ) sin(θ )

2 2
= 2 2
= = r cos2 (θ ) sin(θ )
x +y (r cos(θ )) + (r sin(θ )) r2
p
As ( x, y) → (0, 0), |r | = x2 + y2 → 0, so r cos2 (θ ) sin(θ ) ≤ |r | → 0,
showing that the limit, as ( x, y) → (0, 0), of the original expression
must be 0 regardless of the path involved. Applying the same method
to the expression in Example 2 yields:

x2 (r cos(θ ))2 r2 cos2 (θ )

= = = cos2 (θ )
x 2 + y2 (r cos(θ ))2 + (r sin(θ ))2 r2

so the value of this expression depends not on the distance of ( x, y)

from (0, 0) but rather on the path ( x, y) follows as it approaches (0, 0).

Practice 8. Redo Practice 1 using polar coordinates.

Practice 9. Redo Practice 2 using polar coordinates.

Polar coordinates can only help us investigate limits when the limit
point is (0, 0).
( x − 5)(y + 3)2
Example 11. Find lim .
( x,y)→(5,−3) ( x − 5)2 + ( y + 3)2
Solution. A change of variables with u = x − 5 and v = y + 3 yields:

( x − 5)(y + 3)2 u · v2
lim 2 2
= lim
( x,y)→(5,−3) ( x − 5) + ( y + 3) (u,v)→(0,0) u2 + v2

Now let u = r cos(θ ) and v = sin(θ ) so that the limit becomes:

r cos(θ ) · r2 sin2 (θ )
lim = lim r cos(θ ) sin2 (θ ) = 0
r →0+ r2 r →0+

no matter the value of θ. J

Limits and Continuity in Three (or More) Variables

Because of our use of the vector notation x → a, the definitions and
results about limits and continuity apply directly to functions with any
number of variables.

Example 12. Compute lim( x,y,z)→(7,3,π ) cos ( xyz).

Solution. The function cos(u) is everywhere continuous (on R), as

For limits of functions of three variables, is the polynomial function xyz (on R3 ), so we can conclude that
converting to spherical coordinates often
lim( x,y,z)→(7,3,π ) cos ( xyz) = cos(21π ) = −1. J
helps when the limit point is (0, 0, 0).
q
Practice 10. Where is f ( x, y, z) = x2 + y2 − z continuous?
 14.2 limits and continuity of multivariable functions 1419

When evaluating limits of multivariable functions, keep in mind that

showing that a limit does not exist merely requires finding two paths
that yield different limits, while showing that a limit does exist requires
an algebraic argument (possibly using polar or spherical coordinates).

14.2 Problems
In Problems 1–4, use the given level curves to determine each limit as best you can. (Assume each function
behaves “nicely” away from the given level curves.)

1. Several level curves of z = f ( x, y) appear below. 3. Several level curves of z = S( x, y) appear below.

(a) lim S( x, y) (b) lim S( x, y)

(a) lim f ( x, y) (b) lim f ( x, y) ( x,y)→(1,2) ( x,y)→(2,1)
( x,y)→(1,2) ( x,y)→(1,1)
(c) lim S( x, y) (d) lim S( x, y)
(c) lim f ( x, y) (d) lim f ( x, y) ( x,y)→(1,1) ( x,y)→(3,2)
( x,y)→(2,1) ( x,y)→(3,2)
4. Several level curves of z = T ( x, y) appear below.
2. Several level curves of z = g( x, y) appear below.

(a) lim T ( x, y) (b) lim T ( x, y)

(a) lim g( x, y) (b) lim g( x, y) ( x,y)→(3,3) ( x,y)→(2,2)
( x,y)→(2,2) ( x,y)→(2,1)
(c) lim T ( x, y) (d) lim T ( x, y)
(c) lim g( x, y) (d) lim g( x, y) ( x,y)→(1,2) ( x,y)→(4,1)
( x,y)→(1,2) ( x,y)→(3,2)
 1420 functions of several variables

5. A friend tells you that: x 2 + y2

27. lim p
( x,y)→(0,0) x 2 + y2 + 1 − 1
lim f ( x, 0) = lim f (0, y) = 5 p
x →0 y →0 x 2 + y2 + 1 − 1
28. lim
What can you conclude about lim f ( x, y) ? ( x,y)→(0,0) x 2 + y2
( x,y)→(0,0) xy − x
29. lim
6. A classmate tells you that: ( x,y)→(0,1) x2 + y2 − 2x + 2y + 2

lim g( x, 0) = 5 and lim g(0, y) = −5 x2 + y2 − 2x − 2y

x →0 y →0 30. lim
( x,y)→(1,−1) x2 + y2 − 2x + 2y + 2
What can you conclude about lim g( x, y) ? 31. Use the paths r1 (t) = ht, 0, 0i and r2 (t) = ht, t, ti
( x,y)→(0,0)
In Problems 7–30, compute the limit if it exists or to investigate:
show that the limit does not exist. x 2 y2 z2
h i lim
7. lim x2 y2 − 2xy5 + 3y ( x,y,z)→(0,0,0) x 2 + y2 + z2
( x,y)→(2,3)
h i What can you conclude about this limit?
8. lim x3 + 3x2 y2 − 5y3 + 1
( x,y)→(−3,4) 32. Use the paths r1 (t) = ht, 0, 0i and r2 (t) = ht, t, ti
x 2 y3 + x 3 y2 − 5 to investigate:
9. lim
( x,y)→(0,0) 2 − xy
x2
x2 + xy + y2 lim
10. lim ( x,y,z)→(0,0,0) x2 + y2 + z2
( x,y)→(−2,1) x 2 − y2
What can you conclude about this limit?
 
x+y
11. lim x · sin
( x,y)→(π,π ) 4
√ In Problems 33–38, compute the limit if it exists or
sin x2 + y2


12. lim e x+2y 13. lim
show that the limit does not exist.
( x,y)→(1,4) x 2 + y2
( x,y)→(0,0) xz2 − y2 z
33. lim
x−y ( x,y,z)→(1,2,3) xyz − 1
x 2 − y2 15. lim
14. lim ( x,y)→(0,0) x 2 + y2 34. lim [ x · e x + ln (2x − y)]
( x,y)→(0,0) x+y ( x,y,z)→(2,3,0)

x3 8x2 y2 x 2 − y2 − z2
16. lim 17. lim 35. lim
( x,y)→(0,0) x 3 + y3 ( x,y)→(0,0) x 4 + y4 ( x,y,z)→(0,0,0) x 2 + y2 + z2
xy + yz + xz
x3 + xy2 2xy 36. lim
18. lim 19. lim ( x,y,z)→(0,0,0) x 2 + y2 + z2
( x,y)→(0,0) x 2 + y2 ( x,y)→(0,0) x2 + y2
xy + yz2 + xz2
2 37. lim
x 2 + y2 + z2
p
( x + y) | xy| ( x,y,z)→(0,0,0)
20. lim 21. lim
x 2 + y2
p
( x,y)→(0,0) ( x,y)→(0,0) x2 + y2 x 2 y2 z2
38. lim
( x,y,z)→(0,0,0) x 2 + y2 + z2
2x2 + 3xy + 4y2
22. lim
( x,y)→(0,0) 3x2 + 5y2
39. The function f from Problem 1 is undefined at
xy + 1 (2, 1). Specify a value for f (2, 1) so that f will be
23. lim
( x,y)→(0,0) x + y2 + 1
2
continuous at (2, 1).
xy3 2x2 y
24. lim 25. lim 40. The function g from Problem 2 is undefined at
( x,y)→(0,0) x 2 + y6 ( x,y)→(0,0) x 4 + y2
(1, 2) and (3, 2). Can you specify values for g(1, 2)
x 3 y2 and g(3, 2) so that g will be continuous at each of
26. lim
( x,y)→(0,0) x 2 + y2 those points?
 14.2 limits and continuity of multivariable functions 1421

41. The function S from Problem 3 is undefined at x 2 + y2 + 1

45. F ( x, y) =
(1, 1), (1, 2) and (3, 2). Can you specify values x 2 + y2 − 1
for S(1, 1), S(1, 2) and S(3, 2) so that S will be x 6 + x 3 y3 + y6
46. G ( x, y) =
continuous at each of those points? x 3 + y3
p p
42. The function T from Problem 4 is undefined at 47. S( x, y) = x + y − x − y
(1, 2), (2, 2) and (4, 1). Can you specify values for 48. T ( x, y) = 2x tan(y)
T (1, 2), T (2, 2) and T (4, 1) so T is continuous at
49. ϕ( x, y) = b xyc
each of those points?
50. µ( x, y) = b x c + byc
In Problems 43–54, determine where the given func- 51. Φ( x, y) = b x + yc
tion is continuous.
52. Θ( x, y) = bcos( x + y)c
43. f ( x, y) = ln (2x + 3y) 53. f ( x, y, z) = x ln (yz)
xy
√
44. g( x, y) = e sin ( x + y) 54. g( x, y, z) = x + y x + z

Open and Closed Sets

The definitions for limits and continuity of a function of one variable
involve open intervals of the form | x − a| < δ, which we can express as
a − δ < x < a + δ. These intervals are called open because they contain
neither endpoint. An interval of the form a − δ ≤ x ≤ a + δ is closed
and one of the form a − δ < x ≤ a + δ is neither open nor closed.
For functions of two or more variables, the definitions in this section
involve sets of the form kx − ak < δ. In two dimensions, we can rewrite
this inequality in the form kh x, yi − h x0 , y0 ik < δ or:
q
( x − x0 )2 + ( y − y0 )2 < δ ⇒ ( x − x0 )2 + ( y − y0 )2 < δ2

which describes an open disk with center ( x0 , y0 ), sitting “inside” the

circle ( x − x0 )2 + (y − y0 )2 = δ2 . The disk is open because it does not
contain its boundary (the circle), just as an interval is open when it does
not contain its boundary (the left and right endpoints of the interval).
Similarly, in three dimensions our definitions in this section involved
an open ball of the form ( x − x0 )2 + (y − y0 )2 + (z − z0 )2 < δ2 sitting
inside the sphere ( x − x0 )2 + (y − y0 )2 + (z − z0 )2 = δ2 . Many regions
of interest are not intervals, disks or balls, however, so we will need a
definition of “open” and “closed” for more complicated sets.
In Rn , a point x is in the interior of a region R if there is a (possibly
very small) number δ > 0 so that a ball of radius δ centered at x
contains only points in R. A point x is on the boundary of R if any
ball centered at x (no matter how small the radius) contains both points
inside R and points outside R. A set R is open if it contains none of
its boundary points and closed if it contains all of its boundary points;
otherwise the set is “neither open nor closed.”
1422 functions of several variables

14.2 Practice Answers

1. If ( x, y) approaches (0, 0) along the x-axis, where y = 0:

x2 − 02 x2
F ( x, 0) = = =1
x2 + 02 x2
for x 6= 0. Along the y-axis, where x = 0:

02 − y 2 − y2
F (0, y) = = = −1
02 + y 2 y2
for y 6= 0, so F ( x, y) has no single limiting value as ( x, y) → (0, 0).

2. Substituting mx for y:

x · mx mx2 m
G ( x, mx ) = = =
x2 + (mx )2 (1 + m ) x 2 1 + m2

when x 6= 0. The limiting value of G ( x, y) depends on the line to

which we restrict the values of ( x, y).

3. Substituting Ax2 for y:

  x2 · Ax2 Ax4 A2
f x, Ax2 = 2
= =
x4 + ( Ax2 ) (1 + A2 ) x 4 1 + A2

when x 6= 0. The limiting value of f ( x, y) depends on the parabola

to which we restrict the values of ( x, y).

4. Applying the Main Limit Theorem and the results of Example 6:

3x2 + 6xy − 7y2 + 12 3 · 22 + 6 · 2(−5) − 7(−5)2 + 12

lim =
( x,y)→(2,−5) ( x + 2)2 + ( y − 5)2 (2 + 2)2 + (−5 − 5)2
211
which simplifies to − .
116
5. Because g( x, y) is a rational function, it is continuous for all ( x, y)
except where the denominator is 0, and that occurs only at (−2, 5).

6. Because G ( x, y) is a rational function, it is continuous everywhere

except at the points where its denominator equals 0, and this occurs
only at (0, 0). From Example 1 we know that lim( x,y)→(0,0) G ( x, y) =
0, but G (0, 0) = 1 6= 0, so G ( x, y) is not continuous at (0, 0).

7. Because x2 + y2 − 3 is a polynomial, it is continuous everywhere.

The domain of arcsin(u) is [−1, 1] and arcsin(u) is continuous on
(−1, 1), hence we need:

−1 < x 2 + y2 − 3 < 1 ⇒ 2 < x 2 + y2 < 4

so g( x, y) is continuous on the annulus consisting of all points inside

a circle of radius 2 centered at (0, 0) and outside a circle of radius
√
2 centered at (0, 0).
 14.2 limits and continuity of multivariable functions 1423

8. Substituting x = r cos(θ ) and y = r sin(θ ):

r2 cos2 (θ ) − r2 sin2 (θ )
F ( x, y) = F (r cos(θ ), r sin(θ )) = = cos(2θ )
r2 cos2 (θ ) + r2 sin2 (θ )

which shows that the limiting values of F ( x, y) as ( x, y) → (0, 0)

depend not on the distance of ( x, y) from (0, 0) but on the path along
which ( x, y) moves toward (0, 0).

9. Substituting x = r cos(θ ) and y = r sin(θ ):

r cos(θ ) · r sin(θ ) 1
G ( x, y) = F (r cos(θ ), r sin(θ )) = = sin(2θ )
r2 cos2 (θ ) + r2 sin2 (θ ) 2

which shows that the limiting values of G ( x, y) as ( x, y) → (0, 0)

depend not on the distance of ( x, y) from (0, 0) but on the path along
which ( x, y) moves toward (0, 0).

10. Because x2 + y2 − z is a polynomial, it is continuous everywhere.

√ √
The domain of u is [0, ∞) and u is continuous on (0, ∞), hence
we need 0 < x2 + y2 − z ⇒ z < x2 + y2 , so f ( x, y, z) is continuous
on the region in R3 below the paraboloid z = x2 + y2 .