SECTION .

| Integration Using Tables and Computer Algebra Systems

EXAMPLE | Use the Table of Integrals to nd y xsx 2 1 2x 1 4 dx.

SOLUTION Since the table gives forms involving sa 2 1 x 2 , sa 2 2 x 2 , and

sx 2 2 a 2 , but not sax 2 1 bx 1 c , we rst complete the square:

x 2 1 2x 1 4 − sx 1 1d2 1 3

If we make the substitution u − x 1 1 (so x − u 2 1), the integrand will involve the
pattern sa 2 1 u 2 :

y xsx 2 1 2x 1 4 dx − y su 2 1d su 2 1 3 du
− y u su 2 1 3 du 2 y su 2 1 3 du

The rst integral is evaluated using the substitution t − u 2 1 3:

y u su 2 1 3 du − 12 y st dt − 12 ? 23 t 3y2 − 13 su 2 1 3d3y2
u
. y sa 2 1 u 2 du − 2 sa 2 1 u 2 For the second integral we use Formula 21 with a − s3 :
u
1
a2
2
ln su 1 sa 2 1 u 2 d 1 C y su 2 1 3 du − 2
su 2 1 3 1 32 lnsu 1 su 2 1 3 d

Thus

y xsx 2 1 2x 1 4 dx
x11
− 13 sx 2 1 2x 1 4d3y2 2 sx 2 1 2x 1 4 2 32 lns x 1 1 1 sx 2 1 2x 1 4 d 1 C
2
■

■ Computer Algebra Systems

We have seen that the use of tables involves matching the form of the given inte-
grand with the forms of the integrands in the tables. Computers are particularly good
at matching patterns. And just as we used substitutions in conjunction with tables, a
CAS can perform substitutions that transform a given integral into one that occurs in its
stored formulas. So it isn’t surprising that computer algebra systems excel at integration.
That doesn’t mean that integration by hand is an obsolete skill. We will see that a hand
computation sometimes produces an inde nite integral in a form that is more convenient
than a machine answer.
To begin, let’s see what happens when we ask a machine to integrate the relatively
simple function y − 1ys3x 2 2d. Using the substitution u − 3x 2 2, an easy calcula-
tion by hand gives
1
y 3x 2 2 dx − 13 ln 3x 2 2 1 C
| |
whereas Derive, Mathematica, and Maple all return the answer
1
3 lns3x 2 2d

The rst thing to notice is that computer algebra systems omit the constant of integra-
tion. In other words, they produce a particular antiderivative, not the most general one.
Therefore, when making use of a machine integration, we might have to add a constant.
 CHAPTER | Integrals

Second, the absolute value signs are omitted in the machine answer. That is ne if
our problem is concerned only with values of x greater than 23 . But if we are interested in
other values of x, then we need to insert the absolute value symbol.

EXAMPLE | Use a CAS to evaluate y xsx 2 1 5d8 dx.

SOLUTION Maple and Mathematica give the same answer:

1
18 x 18 1 52 x 16 1 50x 14 1 1750 12 10 8
3 x 1 4375x 1 21875x 1
218750 6
3 x 1 156250x 4 1 390625
2 x2

It’s clear that both systems must have expanded sx 2 1 5d8 by the Binomial Theorem
and then integrated each term.
If we integrate by hand instead, using the substitution u − x 2 1 5, we get
Derive and the TI-89 and TI-92 also
give this answer. y x sx 2 1 5d8 dx − 181 sx 2 1 5d9 1 C
For most purposes, this is a more convenient form of the answer. ■

EXAMPLE | Use a CAS to nd y sin 5x cos 2x dx.

SOLUTION Derive and Maple report the answer

8
2 71 sin 4x cos 3x 2 35
4
sin 2x cos 3x 2 105 cos 3x

whereas Mathematica produces

2645 cos x 2 192

1 3
cos 3x 1 320 1
cos 5x 2 448 cos 7x

We suspect that there are trigonometric identities that we could use to show that these
answers are equivalent. Indeed, if we ask Derive, Maple, and Mathematica to simplify
their expressions using trigonometric identities, they ultimately produce the same form
of the answer:

y sin5x cos2x dx − 213 cos 3x 1 25 cos 5x 2 17 cos7x ■

■ Can We Integrate All Continuous Functions?

The question arises: Will our basic integration formulas, together with the Substitution
Rule, integration by parts, tables of integrals, and computer algebra systems, enable us to
nd the integral of every continuous function? In particular, can we use these techniques
to evaluate y e x dx? The answer is No, at least not in terms of the functions that we are
2

familiar with.
Most of the functions that we have been dealing with in this book are called elemen-
tary functions. These are the polynomials, rational functions, power functions sx a d,
exponential functions sa x d, logarithmic functions, trigonometric and inverse trigono-
metric functions, and all functions that can be obtained from these by the ve opera-
tions of addition, subtraction, multiplication, division, and composition. For instance,
the function

f sxd − Î x2 2 1
x 3 1 2x 2 1
1 lnscos xd 2 xe sin 2x

is an elementary function.
 SECTION . | Integration Using Tables and Computer Algebra Systems

If f is an elementary function, then f 9 is an elementary function but y f sxd dx need

2
not be an elementary function. Consider f sxd − e x . Since f is continuous, its integral
exists, and if we de ne the function F by

y0 e t
x 2
Fsxd − dt

then we know from Part 1 of the Fundamental Theorem of Calculus that

2
F9sxd − e x
2
Thus f sxd − e x has an antiderivative F, but it has been proved that F is not an elemen-
tary function. This means that no matter how hard we try, we will never succeed in
evaluating y e x dx in terms of the functions we know. The same can be said of the fol-
2

lowing integrals:
ex
y x
dx y sinsx 2 d dx y cosse x d dx
1 sin x
y sx 3 1 1 dx y ln x
dx y x
dx

In fact, the majority of elementary functions don’t have elementary antiderivatives.

EXERCISES .
– Use the Table of Integrals on Reference Pages 6–10 to . Verify Formula 53 in the Table of Integrals (a) by differ-
evaluate the integral. entiation and (b) by using the substitution t − a 1 bu.

. y tan 3 s xd dx . y e2 sin 3 d . Verify Formula 31 (a) by differentiation and (b) by substi-

tuting u − a sin .
dx 1
y y2
3
. . dx – Use a computer algebra system to evaluate the inte-
x 2 s4x 2 1 9 x 2 s4x 2 2 7 CAS
gral. Compare the answer with the result of using tables. If the
s2y 2 2 3 answers are not the same, show that they are equivalent.
. y e 2 x arctanse x d dx . y dy
y2 . y sec 4x dx . y x 2 s1 1 x 3d 4 dx
dx
. y0 x 3 sin x dx . y 2x 3 2 3x 2 dx
. y x s1 1 2x dx . y e x s3e x 1 2d
tan 3s1yzd
. y z2
dz . y x sinsx 2 d coss3x 2 d dx
. y tan 5x dx . y sin 4 x dx
. y sin 2x cos x lnssin xd dx
1
sin 2
. y s1 1 s
dx
y x
3
. d
s5 2 sin
ex
y y0 x 4e2x dx
1
. dx . CAS . Computer algebra systems sometimes need a helping hand
3 2 e 2x from human beings. Try to evaluate
x 4 dx sec 2 tan 2
. y sx 10
22
. y s9 2 tan 2
d y s1 1 ln xd s1 1 sx ln xd 2 dx
s4 1 sln xd 2 with a computer algebra system. If it doesn’t return an
. y x
dx . y e t sins t 2 3d dt answer, make a substitution that changes the integral into
one that the CAS can evaluate.
 CHAPTER | Integrals

. Improper Integrals
In de ning a de nite integral yab f sxd dx we dealt with a function f de ned on a nite
interval fa, bg. In this section we extend the concept of a de nite integral to the case
where the interval is in nite. In this case the integral is called an improper integral. One
of the most important applications of this idea, probability distributions, will be studied
in Chapter 12.
y Consider the in nite region S that lies under the curve y − 1yx 2, above the x-axis, and
1 to the right of the line x − 1. You might think that, since S is in nite in extent, its area
y=
≈
must be in nite, but let’s take a closer look. The area of the part of S that lies to the left
1 of the line x − b (shaded in Figure 1) is
=1
area=1-
b

G
x=1 b
1 1 1
y1
b
0 1 b x Asbd − dx − 2 −12
x2 x 1 b
FIGURE
Notice that Asbd , 1 no matter how large b is chosen.
We also observe that

lim Asbd − lim

bl` bl`
S D
12
1
b
−1

The area of the shaded region approaches 1 as b l ` (see Figure 2), so we say that the
area of the in nite region S is equal to 1 and we write
1 b 1
y1 dx − lim y 2 dx − 1
`

x2 bl` 1 x

y y y y

1 2 4
area= 2 area= 3 area= 5 area=1

0 1 2 x 0 1 3 x 0 1 5 x 0 1 x

FIGURE
Using this example as a guide, we de ne the integral of f (not necessarily a positive
function) over an in nite interval as the limit of integrals over nite intervals.

(1) Definition of an Improper Integral If yab f sxd dx exists for every number
b > a, then
ya y
` b
f sxd dx − lim f sxd dx
bl` a

provided this limit exists (as a nite number). An improper integral is called con-
vergent if the corresponding limit exists and divergent if the limit does not exist.

The improper integral in De nition 1 can be interpreted as an area provided that f

is a positive function. For instance, if f sxd > 0 and the integral ya` f sxd dx is convergent,
|
then we de ne the area of the region S − hsx, yd x > a, 0 < y < f sxdj in Figure 3 to be

AsS d − y f sxd dx
`

a
 SECTION . | Improper Integrals

This is appropriate because ya f sxd dx is the limit as b l ` of the area under the graph
`

of f from a to b.

y=ƒ

FIGURE 0 a x

EXAMPLE | Determine whether the integral y1` s1yxd dx is convergent or divergent.

SOLUTION According to De nition 1, we have

1 b 1
y1 dx − lim y | |g
` b
dx − lim ln x 1
x bl` 1 x bl`

− lim sln b 2 ln 1d − lim ln b − `

bl` bl`

The limit does not exist as a nite number and so the improper integral y1 s1yxd dx is
`

divergent. ■

Let’s compare the result of Example 1 with the example given at the beginning of this
section:
` 1 ` 1
y1 x 2 dx converges y1 x dx diverges
Geometrically, this says that although the curves y − 1yx 2 and y − 1yx look very simi-
lar for x . 0, the region under y − 1yx 2 to the right of x − 1 (the shaded region in
Figure 4) has nite area whereas the corresponding region under y − 1yx (in Figure 5)
has in nite area. Note that both 1yx 2 and 1yx approach 0 as x l ` but 1yx 2 approaches
0 faster than 1yx. The values of 1yx don’t decrease fast enough for its integral to have a
nite value.

y y

1 1
y= y= x
≈

finite area infinite area

0 1 x 0 1 x

FIGURE FIGURE
j1 (1/≈) dx converges. j1 (1/x) dx diverges.
` `

EXAMPLE | Salicylic acid pharmacokinetics In Exercises 5.2.14 and

5.5.33 we used the function Cstd − 11.4te 2t to model the concentration of SA in the
 CHAPTER | Integrals

bloodstream of volunteers, where t is measured in hours and C is measured in mgymL.

Calculate y0` Cstd dt and interpret it.

SOLUTION The de nition of an improper integral says that

y0 y y
` b b
Cstd dt − lim Cstd dt − 11.4 lim te 2t dt
bl` 0 bl` 0

We integrate by parts with u − t, dv − e2t dt, so du − dt and v − 2e 2t:

y0 te2t dt − 2te 2tg 0 2 y0 s2e 2t d dt

b b b

TEC In Module 5.8 you can investi-

gate visually and numerically whether − 2be 2b 2 0 2 e 2t g b
0
− 2be 2b 2 e 2b 1 1
several improper integrals are conver-
gent or divergent.
We know that e 2b l 0 as b l `, and by l’Hospital’s Rule we have

b 1
lim be 2b − lim − lim b
bl` bl` eb bl` e

− lim e 2b − 0
bl`

Therefore
C

y0
`
C=11.4te_t Cstd dt − 11.4 lim s2be 2b 2 e 2b 1 1d
bl`
3
− 11.4s20 2 0 1 1d − 11.4 smgymLd ? h

We have previously interpreted y01 Cstd dt in terms of the “availability” of the drug
0 t during the rst hour, with units of concentration times time. Similarly, y04 Cstd dt
5
measures the availability over the rst four hours. The improper integral y0` Cstd dt (the
FIGURE total area under the concentration curve in Figure 6) measures the availability for all
y0 Cstd dt − 11.4 smgymLd ? h time. In other words, it measures the long-term availability of SA.
`
■

We next de ne an improper integral over an interval that extends in nitely far in the
negative direction.

(2) Definition If yab f sxd dx exists for every number a < b, then

y2` f sxd dx − a lim y

b b
f sxd dx
l2 ` a

provided this limit exists (as a nite number).

EXAMPLE | Evaluate y2`

0
e x dx, if it is convergent.

SOLUTION Using De nition 2, we have

y2` e x dx − a lim
0
y e x dx − a lim
l2` a
0

l2`
fe xg a 0

− lim s1 2 e a d − 1
a l2`

a
because lim a l 2` e − 0. ■
 SECTION . | Improper Integrals

Finally, we de ne an integral over the entire real line by splitting the real line into
separate parts.

c
(3) Definition If both yc` f sxd dx and y2` f sxd dx are convergent, then we de ne

y2` f sxd dx − y f sxd dx 1 y f sxd dx

` c `

2` c

where c is any real number. (See Exercise 30).

1
| Evaluate y
`
EXAMPLE dx.
2` 1 1 x2
SOLUTION It’s convenient to choose c − 0 in De nition 3:

1 1 1
y2` 1 1 x 2 dx − y2` 1 1 x 2 dx 1 y0
` 0 `
dx
1 1 x2

We must now evaluate the integrals on the right side separately:

1 dx
y0 dx − lim y g
` b b
− lim tan21 x
1 1 x2 bl` 0 1 1 x2 bl` 0

− lim stan 21 b 2 tan21 0d − lim tan21 b −

bl` bl` 2
1 dx
y2` 1 1 x 2 dx − a lim
0
y
l2` a
0

1 1 x2
− lim tan21 x
a l2`
g 0
a

− lim stan 21 0 2 tan 21 ad

al 2`

−02 2 S D 2
−
2

Since both of these integrals are convergent, the given integral is convergent and
y
1
1
y2` 1 1 x 2 dx −
`
y= area=π 1 −
1+≈ 2 2

0 x Since 1ys1 1 x 2 d . 0, the given improper integral can be interpreted as the area of
the in nite region that lies under the curve y − 1ys1 1 x 2 d and above the x-axis (see
FIGURE Figure 7). ■

Improper integrals are used in various areas of biology, a few of which are illustrated
We will study continuous probability in Example 2 and Exercises 25–28. In the study of continuous probability distributions
distributions in Chapter 12. many improper integrals arise. Probability density functions f have the property that

y2` f sxd dx − 1
`

and a basic fact for normal distributions is that

y2` e 2x
`
dx − s
2

(The proof of this fact requires methods that are beyond the scope of this book.)
 CHAPTER | Integrals

Improper integrals also occur with great frequency in the mathematics required for
medical imaging. In particular, functions de ned in terms of improper integrals have
become the basic tools needed to develop the theory of the CT scan, which creates
images from X-ray data.

EXERCISES .
. Find the area under the curve y − 1yx 3 from x − 1 to x − b . Drug pharmacokinetics The plasma drug concentration
and evaluate it for b − 10, 100, and 1000. Then nd the of a new drug was modeled by the function Cstd − 23te22 t,
total area under this curve for x > 1. where t is measured in hours and C in mgymL.
(a) What is the maximum drug concentration and when did
; . (a) Graph the functions f sxd − 1yx 1.1 and tsxd − 1yx 0.9 in it occur?
the viewing rectangles f0, 10g by f0, 1g and f0, 100g (b) Calculate y0 Cstd dt and explain its signi cance.
`

by f0, 1g.
(b) Find the areas under the graphs of f and t from x − 1 BB . Spread of drug use In a study of the spread of illicit drug
to x − b and evaluate for b − 10, 100, 10 4, 10 6, 10 10, use from an enthusiastic user to a population of N users, the
and 10 20. authors model the number of expected new users by the
(c) Find the total area under each curve for x > 1, if it equation
exists.
cNs1 2 e 2kt d 2 t
−y
`
e dt
0 k
– Determine whether each integral is convergent or diver-
gent. Evaluate those that are convergent.
where c, k, and are positive constants. Evaluate this
1 1
y3 y0 integral to express in terms of c, N, k, and .
` `
. dx . dx
sx 2 2d3y2 s
4
11x Source: Adapted from F. C. Hoppensteadt et al., “Threshold Analysis of a
Drug Use Epidemic Model,” Mathematical Biosciences 53 (1981): 79–87.
1 x
y2` y0
21 `
. dw . dx
s2 2 w sx 2 1 2d 2 . Photosynthesis Much of the earth’s photosynthesis
occurs in the oceans. The rate of primary production
y4 y2` e
` 21
. e 2yy2
dy . 22t
dt
depends on light intensity, measured as the ux of photons
(that is, number of photons per unit area per unit time). For
y2 y2` s y 3 2 3y 2 d dy
` `
. sin d . monochromatic light, intensity decreases with water depth
according to Beer’s Law, which states that Isxd − e 2kx,
e2sx where x is water depth. A simple model for the relationship
y2` xe2x y1
` 2 `
. dx . dx
between rate of photosynthesis and light intensity is
sx
PsId − aI, where a is a constant and P is measured as a
x11
y1 y2` cos
` `
. dx . t dt mass of carbon xed per volume of water, per unit time.
x 2 1 2x Calculate y0 PsIsxdd dx and interpret it.
`

y0 y2` re ry3 dr
` 6 Source: Adapted from A. Jassby et al., “Mathematical Formulation of the
. se 25s ds .
Relationship between Photosynthesis and Light for Phytoplankton,” Limnol-
ogy and Oceanography 21 (1976): 540–7.
ln x
y1 y2` x 3e2x
` ` 4
. dx . dx
x . Dialysis treatment removes urea and other waste products
x2 ln x from a patient’s blood by diverting some of the blood ow
y2` y1
` `
. dx . dx externally through a machine called a dialyzer. The rate at
9 1 x6 x3
which urea is removed from the blood (in mgymin) is often
1 ex
ye y0
` `
. dx . dx well described by the equation
xsln xd3 e 132x

K
cstd − c0 e 2Kty V
V
– Sketch the region and nd its area (if the area is nite).
where K is the rate of ow of blood through the dialyzer (in
. S − hsx, yd | x < 1, 0 < y < e j x
mLymin), V is the volume of the patient’s blood (in mL),
. S − hsx, yd | x > 22, 0 < y < e 2xy2
j and c0 is the amount of urea in the blood (in mg) at time
t − 0. Evaluate the integral y0` cstd dt and interpret it.
 CHAPTER | Review

. A manufacturer of lightbulbs wants to produce bulbs that This shows that we can’t de ne
last about 700 hours but, of course, some bulbs burn out
y2` f sxd dx − tlim y f sxd dx
` t
faster than others. Let Fstd be the fraction of the company’s l ` 2t
bulbs that burn out before t hours, so Fstd always lies
between 0 and 1. . For what values of p is the integral
(a) Make a rough sketch of what you think the graph of F 1
y1
`
might look like. dx
xp
(b) What is the meaning of the derivative rstd − F9std?
(c) What is the value of y0 rstd dt? Why?
` convergent? Evaluate the integral for those values of p.

. If y2`
`
f sxd dx is convergent and a and b are real numbers, – Evaluate the integral, given that
show that
y0
`
e 2x dx − 12 s
2

y2` ya y2` yb
a ` b `
f sxd dx 1 f sxd dx − f sxd dx 1 f sxd dx

y0 e 2x y2 dx y0
` 2 ` 2
. . x 2e 2x dx
. (a) Show that y2`
`
x dx is divergent.
(b) Show that
y0
`
. sx e 2x dx
y
t
lim x dx − 0
t l ` 2t

Chapter REVIEW
CONCEPT CHECK

. (a) Write an expression for a Riemann sum of a function f. . If rstd is the rate of growth of a population at time t, where t
Explain the meaning of the notation that you use. is measured in months, what does y610 rstd dt represent?
(b) If f sxd > 0, what is the geometric interpretation of a
. (a) Explain the meaning of the inde nite integral y f sxd dx.
Riemann sum? Illustrate with a diagram.
(b) What is the connection between the de nite integral
(c) If f sxd takes on both positive and negative values, what
yab f sxd dx and the inde nite integral y f sxd dx?
is the geometric interpretation of a Riemann sum? Illus-
trate with a diagram. . State both parts of the Fundamental Theorem of Calculus.
. (a) State the Substitution Rule. In practice, how do you use
. (a) Write the de nition of the de nite integral of a continu-
it?
ous function from a to b.
(b) State the rule for integration by parts. In practice, how
(b) What is the geometric interpretation of yab f sxd dx if
do you use it?
f sxd > 0?
(c) What is the geometric interpretation of yab f sxd dx if f sxd . De ne the following improper integrals.
takes on both positive and negative values? Illustrate
ya y2` f sxd dx y2` f sxd dx
` b `
(a) f sxd dx (b) (c)
with a diagram.
. Explain exactly what is meant by the statement that
. State the Midpoint Rule. “differentiation and integration are inverse processes.”
. (a) State the Evaluation Theorem. Answers to the Concept Check can be found on the back
(b) State the Net Change Theorem. endpapers.

TRUE-FALSE QUIZ

Determine whether the statement is true or false. If it is true, . If f and t are continuous on fa, bg, then
explain why. If it is false, explain why or give an example that
disproves the statement.
. If f and t are continuous on fa, bg, then
b
S
ya f f sxd tsxdg dx − ya f sxd dx ya tsxd dx
b
DS b
D
. If f is continuous on fa, bg, then

ya f f sxd 1 tsxdg dx − ya f sxd dx 1 ya tsxd dx ya 5f sxd dx − 5 ya f sxd dx

b b b b b
 CHAPTER | Integrals

. If f is continuous on fa, bg, then . y0 sx 2 x 3 d dx represents the area under the curve
2

y − x 2 x 3 from 0 to 2.
ya x f sxd dx − x ya f sxd dx
b b

. All continuous functions have antiderivatives.

. If f is continuous on fa, bg and f sxd > 0, then
. All continuous functions have derivatives.

ya sf sxd dx −
b
Îy b

a
f sxd dx . If f is continuous, then y2` f sxd dx − lim y2t
` t
f sxd dx.
tl`

. If f is a continuous, decreasing function on f1, `d and

. If f 9 is continuous on f1, 3g, then y f 9svd dv − f s3d 2 f s1d.
3

lim f sxd − 0, then y1 f sxd dx is convergent.

`
1
tl`
. If f and t are continuous and f sxd > tsxd for a < x < b,
. If ya` f sxd dx and ya tsxd dx are both convergent, then
`
then
ya` f f sxd 1 tsxdg dx is convergent.
ya f sxd dx > ya tsxd dx
b b

. If f and t are differentiable and f sxd > tsxd for a , x , b, . If ya f sxd dx and ya tsxd dx are both divergent, then
` `

then f 9sxd > t9sxd for a , x , b. ya f f sxd 1 tsxdg dx is divergent.

`

. y21
1
S x 5 2 6x 9 1
sin x
s1 1 x 4 d2
D dx − 0 . If f is continuous on fa, bg, then

. y25 sax 2 1 bx 1 cd dx − 2 y0 sax 2 1 cd dx

5 5 d
dx
Sy a
b
D
f sxd dx − f sxd

EXERCISES

. Use the given graph of f to nd the Riemann sum with six (c) Use the Evaluation Theorem to check your answer to
subintervals. Take the sample points to be (a) left endpoints part (b).
and (b) midpoints. In each case draw a diagram and explain (d) Draw a diagram to explain the geometric meaning of the
what the Riemann sum represents. integral in part (b).
y . Evaluate

y0 s x 1 s1 2 x 2 d dx
1

y=ƒ
2 by interpreting it in terms of areas.
. Express
0 2 6 x n
lim o sin x i Dx
n l ` i−1

as a de nite integral on the interval f0, g and then evaluate

the integral.
. (a) Evaluate the Riemann sum for . If y06 f sxd dx − 10 and y04 f sxd dx − 7, nd y46 f sxd dx.
f sxd − x 2 2 x 0<x<2
. (a) Write y03 e 2xy 2 dx as a limit of Riemann sums, taking the
with four subintervals, taking the sample points to be sample points to be right endpoints.
right endpoints. Explain, with the aid of a diagram, what (b) Use the Midpoint Rule with six subintervals to estimate
the Riemann sum represents. the value of the integral in part (a). State your answer
(b) Use the de nition of a de nite integral (with right end- correct to three decimal places.
points) to calculate the value of the integral (c) Use the Fundamental Theorem to evaluate y03 e 2xy2 dx.
Round your answer to three decimal places and compare
y0 sx 2 2 xd dx
2

with your estimate in part (b).

 CHAPTER | Review

. The following gure shows the graphs of f, f 9, and ; . Use a graph to give a rough estimate of the area of the
y0x f std dt. Identify each graph, and explain your choices. region that lies under the curve y − x sx , 0 < x < 4.
Then nd the exact area.
y
b
– Find the derivative of the function.
c
t2
. Fsxd − y
x
dt
0 1 1 t3
x 1 2 t2
. tsxd − y
sin x
dt
a 1 1 1 t4

– Use the Table of Integrals on the Reference Pages to

. Evaluate: evaluate the integral.
d arctan x d
y0 y0 e arctan x dx y csc 5t dt y e x s1 2 e 2 x
1 1
(a) se d dx (b) . . dx
dx dx
d
y0 e arctan t dt cot x
x
(c)
dx . y s1 1 2 sin x
dx

– Evaluate the integral.

y1 s8x 3 1 3x 2 d dx y0 . Use Property 8 of integrals (page 338) to estimate the value

2 T
. . sx 4 2 8x 1 7d dx
of
y1 sx 2 1 3 dx
3

y0 y0
1 1
. s1 2 x 9 d dx . s1 2 xd9 dx

S D
. Use the properties of integrals to verify that
2

y
12x
y0 ssu 1 1d
1
. .
y0 x 4 cos x dx < 0.2
2
dx 4
du 0<
1
x

x csc 2 x –
y0 y Evaluate the integral or show that it is divergent.
1
. dx
2 . dx
x 11 1 1 cot x 1 ln x
y1 y0
` `
. dx . dx
s2x 1 1d3 x4
y0 v 2 coss v 3d dv y0 sins3
1 1
. . td dt
y2` e 22x dx
0
.
1
y0 e y1
1 2
. t
dt . dx
2 2 3x
x12 . The speedometer reading v on a car was observed at
y sx 2 1 4x y1 x 3 ln x dx
2
. dx . one-minute intervals and recorded in the chart. Use the
Midpoint Rule to estimate the distance traveled by the car.
x
y0 y0 ye 20.6y dy
5 5
. dx . t (min) v smiyhd t (min) v smiyhd
x 1 10
0 40 6 56
t 4 tan t dt
y2 y4 y1 1 42 7 57
y4 4
. dt .
2 1 cos t s2t 1 1d 3 2 45 8 57
3 49 9 55
y1 x 3y 2 ln x dx y sin x cosscos xd dx
4
. . 4 52 10 56
5 54
y esx dx y tan21 x dx
3
. .
. Let rstd be the rate at which the world’s oil is consumed,
ex where t is measured in years starting at t − 0 on January 1,
sec tan
y0
1
. y 1 1 sec
d . 1 1 e 2x
dx 2000, and rstd is measured in barrels per year. What does
y015 rstd dt represent?
 CHAPTER | Integrals

. A population of honeybees increased at a rate of rstd bees . Environmental pollutants In Section 10.3 a model for
per week, where the graph of r is as shown. Use the Mid- the transport of environmental pollutants between three
point Rule with six subintervals to estimate the increase in lakes is analyzed. It is shown that, for certain parameter
the bee population during the rst 24 weeks. values, the concentration of pollutant in one of the lakes as
a function of time is given by an equation of the form
r
12000 xstd − k 2 ke 2at cos bt
The environmental impact of the pollutant is a function
8000 of both its concentration and the duration of time that it
persists. The integral of the concentration over time is a
summary measure of this impact. Calculate this measure of
4000
impact over the rst unit of time.
. Niche overlap The extent to which species compete for
0 4 8 12 16 20 24 t resources is often measured by the niche overlap. If the
(weeks) horizontal axis represents a continuum of different resource
types (for example, seed sizes for certain bird species),
. An oil leak from a well is causing pollution at a rate of
then a plot of the degree of preference for these resources
rstd − 90e 20.12t gallons per month. If the leak is never xed,
is called a species’ niche. The degree of overlap of two spe-
what is the total amount of oil that will be spilled?
cies’ niches is then a measure of the extent to which they
. Antibiotic pharmacokinetics An antibiotic tablet is compete for resources. The niche overlap for a species is the
taken and t hours later the concentration in the bloodstream fraction of the area under its preference curve that is also
is under the other species’ curve. Many species’ niches are
Cstd − 3se 20.8t 2 e 21.2t d best modeled by a function that has a peak at some inter-
mediate resource type and decreases to 0 asymptotically.
where C is measured in mgymL. The niches displayed in the gure are given by
(a) What is the maximum concentration of the antibiotic
and when does it occur? n1sxd − e 2| x28 | n 2 sxd − e 2| x212 |
(b) Calculate y02 Cstd dt and interpret your result. for species 1 and 2, respectively. Calculate the niche overlap
(c) Calculate y0` Cstd dt and explain its meaning. for species 1.

. Population dynamics Suppose that the birth and death y

rates in a population change through time according to the
functions bstd and dstd. The net rate of change is de ned as n¡ n™
rstd − bstd 2 dstd.
(a) Find an expression for the net change in population size
between times t − a and t − b in terms of rstd. 0 10 x
(b) Use Property 4 of integrals (page 336) to show that your
answer to part (a) can also be expressed in terms of the . If f is a continuous function such that
total number of births and the total number of deaths
y0 f std dt − xe 2x 1 y0 e 2t f std dt
x x
over this period.
. Angiotensin-converting enzyme inhibitors are medica-
tions that reduce blood pressure by dilating blood vessels. for all x, nd an explicit formula for f sxd.
The rate of change of blood pressure with respect to dosage . Find a function f and a value of the constant a such that
is given by the equation
2 y f std dt − 2 sin x 2 1
x

8 l vR9sdd a
P9sdd − 2
Rsdd 3 . If f 9 is continuous on fa, bg, show that
where v is blood velocity, is blood viscosity, l is the length
2 y f sxd f 9sxd dx − f f sbdg 2 2 f f sadg 2
b
of the blood vessel, and Rsdd is the radius of the vessel as a a
function of the dose d. Use a substitution to integrate P9sdd
and show that you obtain Poiseuille’s Law: . If f 9 is continuous on f0, `d and lim x l ` f sxd − 0, show
that
4 lv
y0
`
Psdd − f 9sxd dx − 2f s0d
Rsdd 2
 CASE STUDY c | Kill Curves and Antibiotic Effectiveness

CASE STUDY c Kill Curves and Antibiotic Effectiveness

Recall that in this case study we are exploring the relationship between
the magnitude of antibiotic treatment and the effectiveness of the treat-
ment. One of the most important components of our analysis is the antibiotic
concentration pro le, which is a plot of the antibiotic concentration as a function of time.
In the simple model of Case Studies 1a and 1b we modeled a single dose of antibiotic
using the equation
dc
() − 2kc
dt
for some positive constant k. From this we saw that the concentration as a function of
time is

( ) cstd − c0 e 2kt

where c0 is the concentration at t − 0. (See Figure 1.)

( g/mL)
1.5
Concentration

1.0

0.5
FIGURE
Antibiotic concentration pro le
0 10 20 30 t (hours)
modeled by the function cstd − c0 e 2kt
with c0 − 1.2 mgymL and k − 0.175 Time

Three of the most common measures of the magnitude of antibiotic treatment are
(1) the peak antibiotic concentration divided by MIC, denoted by ; (2) the duration of
time for which the antibiotic concentration remains above MIC, denoted by ; and (3) the
area under the antibiotic concentration pro le divided by MIC, denoted by . These are
shown in Figure 2. In Case Study 1a you derived expressions for the rst two.
( g/mL)

∏=cmax/MIC
cmax
Concentration

å=area A/MIC

A
MIC
FIGURE t
Three measures of the magnitude
of antibiotic treatment Time (hours)

. Find an expression for in terms of k, c0, and MIC, using Equation 2. The area
under the concentration pro le should be calculated from t − 0 to `.

We saw in Case Study 1a that, for a given antibiotic and bacteria species (in other
words, for a given value of k and MIC ), all three quantities , , and increase with one
another. In other words, it is not possible to have a high value of without also having
high values of and . Here you will show that we can break this dependency if we
 CHAPTER | Integrals

divide the total amount of antibiotic given c0 into multiple smaller doses. This is referred
to as dose fractionation.
In the simplest case, suppose that instead of giving a total amount of c0 mgymL
of antibiotic at t − 0, we instead give c0 y2 at t − 0 and another dose of c0 y2 at time
t − t^ . The time t^ is called the interdose interval. Furthermore, suppose that at each
dose the concentration instantly increases by c0 y2, and otherwise it decays according to
Equation 1.
. Find an equation for the concentration as a function of time. Figure 3 plots this
function for a speci c choice of constants, along with the concentration pro le
when a single dose of c0 mgymL is given at t − 0.
c(t) ( g/mL)
1.2

Concentration
FIGURE 0.8
Red curve is the concentration
pro le modeled by the function from 0.4
Problem 2 with c0 − 1.2 mgymL,
t^ − 10, and k − 0.175. Blue curve is
0 10 20 30 t
the concentration pro le when all (hours)
the antibiotic is given at t − 0. Time

. Use your answer to Problem 2 to nd an expression for , the peak concentration

divided by MIC.
. Use your answer to Problem 2 to show that is the same under dose fraction-
ation as it is for the single dose case in Problem 1.
. Using your answers from Problems 3 and 4, explain how it is possible to use
dose fractionation to increase without also increasing .

One of the reasons different drug doses and interdose intervals are used for different
infections is to achieve different values of , , and .
Applications of Integrals 6

Arteries of the human hand are

shown in a colorized X ray. In
Section 6.3 we use an integral to
calculate the flux in an artery (the
volume of blood that passes a
cross-section of an artery
per unit time).
GJLP / CNRI / SPL / Science Source

. Areas Between Curves

PROJECT: Disease Progression and Immunity
PROJECT: The Gini Index
. Average Values
. Further Applications to Biology
. Volumes
CASE STUDY d: Kill Curves and Antibiotic Effectiveness
CASE STUDY b: Hosts, Parasites, and Time-Travel

387
 CHAPTER | Applications of Integrals

W
E HAVE ALREADY SEEN some of the applications of integrals in Chap-
ter 5: drug pharmacokinetics, measles pathogenesis, bacterial growth, basal
metabolism, breathing cycles, dialysis treatment, tumor growth, photosynthe-
sis, medical imaging, bioavailability, the sterile insect technique, the rumen microbial
ecosystem, and the spread of drug use.
In this chapter we explore some additional applications of integration: areas between
curves, cerebral blood ow, disease progresssion, average values, survival and renewal,
cardiac output, and volumes.

. Areas Between Curves

y In Chapter 5 we de ned and calculated areas of regions that lie under the graphs of
y=ƒ functions. Here we use integrals to nd areas of regions that lie between the graphs of
two functions and then we show how this idea occurs in cerebral blood ow. As well, in
the project after this section we see how the study of measles pathogenesis involves areas
S
between curves.
Consider the region S that lies between two curves y − f sxd and y − tsxd and be-
tween the vertical lines x − a and x − b, where f and t are continuous functions and
0 a b x
f sxd > tsxd for all x in fa, bg. (See Figure 1.)
y=© Just as we did for areas under curves in Section 5.1, we divide S into n strips of equal
width and then we approximate the ith strip by a rectangle with base Dx and height
f sx*i d 2 tsx*i d. (See Figure 2. If we like, we could take all of the sample points to be
FIGURE
right endpoints, in which case x*i − x i.) The Riemann sum
S=s(x, y) | a¯x¯b, ©¯y¯ƒ d
n

o f f sx*i d 2 tsx*i dg Dx
i−1

is therefore an approximation to what we intuitively think of as the area of S.

y y

f (x *i )
f (x *i )-g(x *i )

0 a x 0 a x
b b
_ g(x *i )
x *i
Îx

FIGURE (a) Typical rectangle (b) Approximating rectangles

This approximation appears to become better and better as n l `. Therefore we

de ne the area A of the region S as the limiting value of the sum of the areas of these
approximating rectangles.

n
() A − lim
n l ` i−1
o f f sx*i d 2 tsx*i dg Dx
 SECTION . | Areas Between Curves

We recognize the limit in (1) as the de nite integral of f 2 t. Therefore we have the
following formula for area.

(2) The area A of the region bounded by the curves y − f sxd, y − tsxd, and the
lines x − a, x − b, where f and t are continuous and f sxd > tsxd for all x in
fa, bg, is
A − y f f sxd 2 tsxdg dx
b

y
Notice that in the special case where tsxd − 0, S is the region under the graph of f
y=ƒ and our general de nition of area (1) reduces to our previous de nition (De nition 5.1.2).
S In the case where both f and t are positive, you can see from Figure 3 why (2) is true:
y=©
A − farea under y − f sxdg 2 farea under y − tsxdg
0 a b x
− y f sxd dx 2 y tsxd dx − y f f sxd 2 tsxdg dx
b b b

a a a
FIGURE

A=j ƒ dx-j © dx
b b

a a EXAMPLE | Find the area of the region bounded above by y − e x, bounded below
by y − x, and bounded on the sides by x − 0 and x − 1.
y
SOLUTION The region is shown in Figure 4. The upper boundary curve is y − e x
and the lower boundary curve is y − x. So we use the area formula (2) with f sxd − e x,
y=´ tsxd − x, a − 0, and b − 1:
x=1

1 A − y se x 2 xd dx − e x 2 12 x 2
1

0
g 1
0

y=x Îx − e 2 12 2 1 − e 2 1.5 ■
0 1 x

In Figure 4 we drew a typical approximating rectangle with width Dx as a reminder

FIGURE of the procedure by which the area is de ned in (1). In general, when we set up an inte-
gral for an area, it’s helpful to sketch the region to identify the top curve yT , the bottom
y curve yB, and a typical approximating rectangle as in Figure 5. Then the area of a typical
yT rectangle is s yT 2 yBd Dx and the equation
yT-yB n

o syT 2 yBd Dx − ya syT 2 yBd dx

b
A − lim
n l ` i−1

yB
Îx summarizes the procedure of adding (in a limiting sense) the areas of all the typical
rectangles.
0 a b x
Notice that in Figure 5 the left-hand boundary reduces to a point, whereas in Figure 3
the right-hand boundary reduces to a point. In the next example both of the side bound-
FIGURE aries reduce to a point, so the rst step is to nd a and b.

EXAMPLE | Find the area of the region enclosed by the parabolas y − x 2 and
2
y − 2x 2 x .

SOLUTION We rst nd the points of intersection of the parabolas by solving

their equations simultaneously. This gives x 2 − 2x 2 x 2, or 2x 2 2 2x − 0. Thus
2xsx 2 1d − 0, so x − 0 or 1. The points of intersection are s0, 0d and s1, 1d.
 CHAPTER | Applications of Integrals

y yT=2x-≈ We see from Figure 6 that the top and bottom boundaries are

(1, 1) yT − 2x 2 x 2 and yB − x 2

The area of a typical rectangle is

yB=≈
syT 2 yBd Dx − s2x 2 x 2 2 x 2 d Dx − s2x 2 2x 2 d Dx
Îx
(0, 0) x
and the region lies between x − 0 and x − 1. So the total area is

A − y s2x 2 2x 2 d dx − 2 y sx 2 x 2 d dx
1 1
FIGURE
0 0

−2 F x2
2
2
x3
3
G S D
1

0
−2
1
2
2
1
3
−
1
3
■

√ (mi/ h) EXAMPLE | Interpreting the area between velocity curves Figure 7

shows velocity curves for two cars, A and B, that start side by side and move along the
60
A
same road. What does the area between the curves represent? Use the Midpoint Rule to
50 estimate it.
40
30 B SOLUTION We know from Section 5.3 that the area under the velocity curve A rep-
20 resents the distance traveled by car A during the rst 16 seconds. Similarly, the area
under curve B is the distance traveled by car B during that time period. So the area
10
between these curves, which is the difference of the areas under the curves, is the
0 2 4 6 8 10 12 14 16 t distance between the cars after 16 seconds. We read the velocities from the graph and
(seconds) convert them to feet per second s1 miyh − 5280
3600 ftysd.

FIGURE
t 0 2 4 6 8 10 12 14 16

vA 0 34 54 67 76 84 89 92 95
vB 0 21 34 44 51 56 60 63 65

vA 2 vB 0 13 20 23 25 28 29 29 30

We use the Midpoint Rule with n − 4 intervals, so that Dt − 4. The midpoints of

the intervals are t1 − 2, t2 − 6, t3 − 10, and t4 − 14. We estimate the distance between
the cars after 16 seconds as follows:

y0
16
svA 2 vB d dt < Dt f13 1 23 1 28 1 29g

− 4s93d − 372 ft ■

■ Cerebral Blood Flow

In a paper 1 published in 1948, Seymour Kety and Carl Schmidt described a method for
measuring cerebral blood ow in which the patient inhales a mixture of gases including
a tracer of 15% nitrous oxide. Let Astd be the arterial concentration of N2O measured as
blood enters the brain and Vstd the venous concentration of N2O in blood owing out of
the brain in the jugular vein. Figure 8 shows typical graphs of Astd and Vstd, which are
measured in units of mL of N2O per mL of blood. Although Astd . Vstd, you can see

1. S. Kety et al., “The Nitrous Oxide Method for the Quantitative Determination of Cerebral Blood Flow in
Man: Theory, Procedure and Normal Values,” Journal of Clinical Investigation 27 (1948): 476–83.
 SECTION . | Areas Between Curves

that after about 10 minutes Astd and Vstd are almost the same because the brain is becom-
ing saturated with nitrous oxide.

A
0.04

mL of N2O per mL of blood

V

0.02

FIGURE
Arterial and venous
concentrations of N2O 0 2 4 6 8 10 t (min)

The area between the curves A and V, shaded in Figure 8, plays a key role in cal-
culating the cerebral blood ow F, which we measure in mLymin. We rst consider
y010 Astd dt. If we divide the interval f0, 10g into subintervals of equal length Dt, then
the volume of N2O that ows past a point in the artery during such a subinterval from
t − ti21 to t − ti is approximately

sconcentration of N2O in bloodd ? svolume of bloodd − Asti dsFDtd

Assuming that F remains constant, we see that the total volume of N2O that enters the
brain during the rst 10 minutes is approximately
n n

o Asti d F Dt − F i−1
i−1
o Asti d Dt
If we now let n l `, we get the total quantity of N2O brought to the brain during the
rst 10 minutes:
F y Astd dt
10

A similar argument shows that the quantity of N2O that leaves the brain during this time
period is
F y Vstd dt
10

The difference of these quantities

F y fAstd 2 Vstdg dt
10

is therefore the quantity of N2O that is taken up by the whole brain during the 10 minutes
of inhalation. Let’s call this quantity QBs10d. Then

QBs10d − F y fAstd 2 Vstdg dt

10

0
 CHAPTER | Applications of Integrals

and therefore
QBs10d
( ) F−
y010 fAstd 2 Vstdg dt

It turns out that QBs10d can be found by other methods, so if the area between the
curves is known, Equation 3 can then be used to calculate the cerebral blood ow.

EXAMPLE | Cerebral blood ow

(a) Use the Midpoint Rule with ve subintervals to estimate the area between the
curves A and V in Figure 8.
(b) If it is known that the amount of N2O absorbed by the brain is QBs10d − 60 mL,
determine the cerebral blood ow.

SOLUTION
(a) We divide the interval f0, 10g into ve subintervals, whose midpoints are t − 1, 3,
5, 7, and 9. Then we use the graphs in Figure 8 to estimate the values of Astd and Vstd
at these midpoints and calculate their differences:

t Astd Vstd Astd 2 Vstd

1 0.029 0.007 0.022

3 0.038 0.027 0.011

5 0.040 0.033 0.007

7 0.042 0.037 0.005

9 0.043 0.041 0.002

Using the Midpoint Rule with Dt − 2 min, we get the following estimate:

y0
10
fAstd 2 Vstdg dt < 2f0.022 1 0.011 1 0.007 1 0.005 1 0.002g − 0.094

So the area between the curves A and V is approximately 0.094 smLymLd ? min.
(b) With QBs10d − 60 mL and our result from part (a), we use Equation 3 to calculate
the cerebral blood ow:
QBs10d 60
F− < < 640 mLymin ■
y010 fAstd2 Vstdg dt 0.094

EXERCISES .
– Find the area of the shaded region. – Sketch the region enclosed by the given curves. Draw a typi-
cal approximating rectangle and label its height and width. Then
. y y=5x-≈ . y
nd the area of the region.
y=œ„„„„
x+2
. y − e x, y − x 2 2 1, x − 21, x−1
(4, 4) x=2
. y − ln x, xy − 4, x − 1, x−3

y=x . y − x 2, y2 − x
1 x
y=
x+1
x . y − x 2 2 2x, y − x 1 4
 SECTION . | Areas Between Curves

– Sketch the region enclosed by the given curves and nd . A cicada wing is shown. Estimate its area using the
its area. Midpoint Rule with six subintervals.
. y − 12 2 x 2, y − x2 2 6
y (cm)
. y − x 2, y − 4x 2 x 2

. y − e x, y − xe x, x−0 0.5

. y − cos x, y − 2 2 cos x, 0<x<2

0 1 2 x (cm)
© Tropper2000 / Shutterstock.com

; – Use a graph to nd approximate x-coordinates of the

points of intersection of the given curves. Then nd (approxi- . A cross-section of an airplane wing is shown. Measure-
mately) the area of the region bounded by the curves. ments of the thickness of the wing, in centimeters, at
20-centimeter intervals are 5.8, 20.3, 26.7, 29.0, 27.6, 27.3,
. y − x sinsx 2 d, y − x4
23.8, 20.5, 15.1, 8.7, and 2.8. Use the Midpoint Rule to
. y − x cos x, y − x 10 estimate the area of the wing’s cross-section.

. Sketch the region that lies between the curves y − cos x

and y − sin 2x and between x − 0 and x − y2. Notice
that the region consists of two separate parts. Find the area 200 cm
of this region.
. The widths (in meters) of a kidney-shaped swimming pool
. Sketch the curves y − cos x and y − 1 2 cos x, were measured at 2-meter intervals as indicated in the
0 < x < , and observe that the region between them gure. Use the Midpoint Rule to estimate the area of the
consists of two separate parts. Find the area of this region. pool.

. Sometimes it’s easier to nd an area by regarding x as a

function of y instead of y as a function of x. To illustrate
this idea, let S be the region enclosed by the line y − x 2 1 5.6 5.0 4.8
6.8 4.8
and the parabola y 2 − 2x 1 6. 7.2
6.2
(a) By sketching S, observe that if you want to integrate
with respect to x you have to split S into two parts with
different boundary curves.
(b) If you integrate with respect to y, observe that there is a
left boundary curve and a right boundary curve. . Cerebral blood flow The table shows measurements of
(c) Find the area of S using the method of either part (a) or Astd, the concentration of N2O owing into a patient’s
part (b). brain, and Vstd, the concentration of N2O owing out of the
brain, where t is measured in minutes and Astd and Vstd are
. Find the area of the region enclosed by the curves y − x measured in mL of N2O per mL of blood.
and 4x 1 y 2 − 12.

. A laurel leaf is shown. Estimate its area using the Mid- t Astd Vstd
point Rule with six subintervals.
1 0.031 0.008
y (cm) 3 0.041 0.029
5 0.042 0.035
3 7 0.044 0.042
© Vasilius / Shutterstock.com

9 0.045 0.044
2

1 (a) Use the Midpoint Rule to estimate y010 fAstd 2 Vstdg dt.
(b) If the volume of N2O absorbed by the brain in the rst
0 1 2 3 4 5 6 x (cm) 10 minutes is 64 mL, calculate the cerebral blood ow.
 CHAPTER | Applications of Integrals

. Cerebral blood flow Models for the arterial and venous (b) What is the meaning of the area of the shaded region?
concentration functions in Figure 8 are given by (c) Which car is ahead after two minutes? Explain.
0.05t 2 0.05t 2 (d) Estimate the time at which the cars are again side by
Astd − Vstd − side.
t2 1 1 t2 1 7
(a) Find the area between the graphs of A and V for √
0 < t < 10.
(b) If the volume of N2O absorbed by the brain in the rst 10
minutes is 60 mL, determine the cerebral blood ow.
. Racing cars driven by Chris and Kelly are side by side at the A
start of a race. The table shows the velocities of each car (in
miles per hour) during the rst 10 seconds of the race. Use B
the Midpoint Rule to estimate how much farther Kelly 0 1 2 t (min)
travels than Chris does during the rst 10 seconds.
. Birth and death rates If the birth rate of a population
t vC vK t vC vK
is bstd − 2200e 0.024t people per year and the death rate is
0 0 0 6 69 80 dstd − 1460e0.018t people per year, nd the area between
1 20 22 7 75 86 these curves for 0 < t < 10. What does this area represent?
2 32 37 8 81 93
3 46 52 9 86 98 . Find the number a such that the line x − a bisects the area
under the curve y − 1yx 2, 1 < x < 4.
4 54 61 10 90 102
5 62 71 . Find the values of c such that the area of the region bounded
by the parabolas y − x 2 2 c 2 and y − c 2 2 x 2 is 576.
. Two cars, A and B, start side by side and accelerate from
rest. The gure shows the graphs of their velocity functions. . Find the area of the region bounded by the parabola y − x 2,
(a) Which car is ahead after one minute? Explain. the tangent line to this parabola at s1, 1d, and the x-axis.

■ PROJECT Disease Progression and Immunity

In Section 5.1 we considered the progression of measles in a patient with no immunity

and graphed the measles pathogenesis curve N − f std (page 325), which we modeled
with the function
f std − 2tst 2 21dst 1 1d

We saw that the total amount of infection by time t (measured by the area under the
pathogenesis curve up to time t) played an important role in determining whether indi-
viduals develop symptoms. In particular, symptoms appear only after the total amount of
infection exceeds 7848 scellsymLd 3 days, which occurs at day 12 of the infection for
individuals with no immunity, that is,

y0
12
f std dt − 7848 scellsymLd 3 days

; . Plot the curves cf std for c − 0.9, 0.85, 0.8, 0.6, and 0.4. These resemble curves
for patients that have increasing levels of immunity against the virus at the time
of the infection.
. Some of the patients in Problem 1 will develop symptoms and some will not.
Find the areas under the curve from t − 0 to t − 21 for each value of c and
 APPLIED PROJECT | The Gini Index

compare them with the value 7848 that is needed to display symptoms. Which
patients will become symptomatic at some point during their infection?

The term infectiousness refers to the extent to which the disease is transmitted between
individuals. For patients without immunity, we saw in Section 5.1 that infectiousness
begins around day t1 − 10 and ends around day t 2 − 18. Infectiousness begins on day
10 because the concentration of virus in the plasma after 10 days [that is, the value
f s10d] is the threshold concentration required before any transmission can occur. Fur-
ther, infectiousness ends on day 18 because the immune system manages to prevent
further transmission from this point onward.

. Plot the points P1 − st1, f st1dd and P2 − st 2, f st 2dd on the graph of f. These points
show the values of f at the beginning and end of the infectious period. Draw a
line between the points. What is the slope of this line? Find an equation of this
line.

. Given that L − f st1d is the threshold concentration of the virus required for trans-
mission to begin, plot the point P3 on the curve N − 0.9 f std where 0.9 f std − L.
The value of t satisfying this equation is the time at which infectiousness begins
for a patient with c − 0.9. It has been shown that the time at which infectious-
ness ends for such patients can again be determined by drawing a line through
P3 with the same slope as that in Problem 3 and then determining the time t 4 at
which it intersects the curve N − 0.9 f std. Draw this line on the graph and deter-
mine t 4.

. Repeat Problem 4 for cf std with c − 0.85, 0.8, 0.6, and 0.4. Note that some
N
patients may not have a point corresponding to P3.
P¡ . Find the area between the graph of f and the line in Problem 3. This area repre-
P™ sents the total level of infectiousness of an infected person. (See Figure 1.)
f
. Find the areas enclosed by the curves N − cf std and their corresponding
L intersecting lines for c − 0.9, 0.85, 0.8, 0.6, and 0.4. (Note that this will not be
feasible for patients that have no infectious period.) Compare these areas to the
one found in Problem 6.)

0 10 18 t . Which patients are

(days) (a) symptomatic and infectious?
(b) symptomatic and noninfectious?
FIGURE
(c) asymptomatic and noninfectious?

■ PROJECT The Gini Index

How is it possible to measure the distribution of income among the inhabitants of a given
country? One such measure is the Gini index, named after the Italian economist Corrado
Gini, who rst devised the index in 1912.
We rst rank all households in a country by income and then we compute the per-
centage of households whose income is at most a given percentage of the country’s total
income. We de ne a Lorenz curve y − Lsxd on the interval f0, 1g by plotting the point
say100, by100d on the curve if the bottom a% of households receive at most b% of the
 CHAPTER | Applications of Integrals
y
1 total income. For instance, in Figure 1 the point s0.4, 0.12d is on the Lorenz curve for the
United States in 2010 because the poorest 40% of the population received just 12% of
the total income. Likewise, the bottom 80% of the population received 50% of the total
income, so the point s0.8, 0.5d lies on the Lorenz curve. (The Lorenz curve is named after
(0.8, 0.5) the American economist Max Lorenz.)
Figure 2 shows some typical Lorenz curves. They all pass through the points s0, 0d
and s1, 1d and are concave upward. In the extreme case Lsxd − x, society is perfectly
egalitarian: The poorest a% of the population receives a% of the total income and so
(0.4, 0.12)
everybody receives the same income. The area between a Lorenz curve y − Lsxd and the
line y − x measures how much the income distribution differs from absolute equality.
0 0.2 0.4 0.6 0.8 1 x The Gini index (sometimes called the Gini coef cient or the coef cient of inequality)
is the area between the Lorenz curve and the line y − x (shaded in Figure 3) divided by
FIGURE the area under y − x.
Lorenz curve for the United States
in 2010 y y
(1, 1)
1

y= x y= x
income
fraction

y= L (x )

0 1 x 0 x
population fraction 1

FIGURE FIGURE

. (a) Show that the Gini index G is twice the area between the Lorenz curve and
the line y − x, that is,
G − 2 y fx 2 Lsxdg dx
1

(b) What is the value of G for a perfectly egalitarian society (everybody has the
same income)? What is the value of G for a perfectly totalitarian society (a
single person receives all the income)?

. The following table (derived from data supplied by the US Census Bureau)
shows values of the Lorenz function for income distribution in the United States
for the year 2010.

x 0.0 0.2 0.4 0.6 0.8 1.0

Lsxd 0.000 0.034 0.120 0.266 0.498 1.000

(a) What percentage of the total US income was received by the richest 20% of
the population in 2010?
(b) Use a calculator or computer to t a quadratic function to the data in the
table. Graph the data points and the quadratic function. Is the quadratic
model a reasonable t?
(c) Use the quadratic model for the Lorenz function to estimate the Gini index
for the United States in 2010.
 SECTION . | Average Values

. The following table gives values for the Lorenz function in the years 1970,
1980, 1990, and 2000. Use the method of Problem 2 to estimate the Gini index
for the United States for those years and compare with your answer to Problem
2(c). Do you notice a trend?

x 0.0 0.2 0.4 0.6 0.8 1.0

1970 0.000 0.041 0.149 0.323 0.568 1.000
1980 0.000 0.042 0.144 0.312 0.559 1.000
1990 0.000 0.038 0.134 0.293 0.530 1.000
2000 0.000 0.036 0.125 0.273 0.503 1.000

CAS . A power model often provides a more accurate t than a quadratic model for
a Lorenz function. If you have a computer with Maple or Mathematica, t a
power function s y − ax k d to the data in Problem 2 and use it to estimate the
Gini index for the United States in 2010. Compare with your answer to parts (b)
and (c) of Problem 2.

. Average Values
It is easy to calculate the average value of nitely many numbers y1, y2, . . . , yn:
y1 1 y2 1 ∙ ∙ ∙ 1 yn
yave −
n
T But how do we compute the average temperature during a day if in nitely many tem-
perature readings are possible? Figure 1 shows the graph of a temperature function Tstd,
15
where t is measured in hours and T in °C, and a guess at the average temperature, Tave.
10 In general, let’s try to compute the average value of a function y − f sxd, a < x < b.
We start by dividing the interval fa, bg into n equal subintervals, each with length
5
Dx − sb 2 adyn. Then we choose points x1*, . . . , x n* in successive subintervals and cal-
6 Tave
culate the average of the numbers f sx1*d, . . . , f sx n*d:
0 12 18 24 t

f sx1*d 1 ∙ ∙ ∙ 1 f sx *n d
FIGURE n
(For example, if f represents a temperature function and n − 24, this means that we take
temperature readings every hour and then average them.) Since Dx − sb 2 adyn, we can
write n − sb 2 adyDx and the average value becomes

f sx 1*d 1 ∙ ∙ ∙ 1 f sx n*d 1
− f f sx1*d Dx 1 ∙ ∙ ∙ 1 f sx n*d Dxg
b2a b2a
Dx
n
1
−
b2a
o f sx i*d Dx
i−1

If we let n increase, we would be computing the average value of a large number of

closely spaced values. (For example, we would be averaging temperature readings taken
 CHAPTER | Applications of Integrals

every minute or even every second.) The limiting value is

n
1 1
o f sx *i d Dx − ya f sxd dx
b
lim
nl` b2a i−1 b2a
by the de nition of a de nite integral.
Therefore we de ne the average value of f on the interval fa, bg as
For a positive function, we can think of
this de nition as saying 1
ya f sxd dx
b
fave −
area b2a
− average height
width

EXAMPLE | Find the average value of the function f sxd − 1 1 x 2 on the

interval f21, 2g.

SOLUTION With a − 21 and b − 2 we have

1 1
ya f sxd dx − y21 s1 1 x 2 d dx
b 2
fave −
b2a 2 2 s21d

−
1
3
F G
x1
x3
3
2

21
−2 ■

EXAMPLE | World population In Chapter 1 we modeled the size of the

human population of the world with the exponential function

Pstd − s1.43653 3 10 9 d ? s1.01395d t

where t is measured in years and t − 0 corresponds to the year 1900. What was the
average population in the 20th century?

SOLUTION The average population for 0 < t < 100 was

1
y0
100
Pave − s1.43653 3 10 9 d ? s1.01395d t dt
100

−
1
100
? s1.43653 3 10 9 d ? F s1.01395d t
lns1.01395d
G t−100

t−0

s1.01395d 2 1
100
− 1.43653 3 10 7 ?
lns1.01395d
< 310 3 10 7 − 3.1 3 10 9

So the average world population in the 20th century was about 3.1 billion. ■

If Tstd is the temperature at time t, we might wonder if there is a speci c time when
the temperature is the same as the average temperature. For the temperature function
graphed in Figure 1, we see that there are two such times—just before noon and just
before midnight. In general, is there a number c at which the value of a function f is
exactly equal to the average value of the function, that is, f scd − fave? The following
theorem says that this is true for continuous functions.
 SECTION . | Average Values

The Mean Value Theorem for Integrals If f is continuous on fa, bg, then
there exists a number c in fa, bg such that
y
1
ya f sxd dx
b
y=ƒ f scd − fave −
b2a

ya f sxd dx − f scdsb 2 ad
b
that is,
f(c)=fave

0 x
The Mean Value Theorem for Integrals is a consequence of the Mean Value Theorem
a c b
for derivatives and the Fundamental Theorem of Calculus. The proof is outlined in Exer-
FIGURE cise 23.
The geometric interpretation of the Mean Value Theorem for Integrals is that, for pos-
You can always chop off the top of itive functions f, there is a number c such that the rectangle with base fa, bg and height
a (two-dimensional) mountain at a f scd has the same area as the region under the graph of f from a to b. (See Figure 2 and the
certain height and use it to ll in the
more picturesque interpretation in the margin note.)
valleys so that the mountain becomes
completely at.
EXAMPLE | Since f sxd − 1 1 x 2 is continuous on the interval f21, 2g, the Mean
Value Theorem for Integrals says there is a number c in f21, 2g such that

y21 s1 1 x 2 d dx − f scdf2 2 s21dg

2
y
(2, 5)
y=1+≈
In this particular case we can nd c explicitly. From Example 1 we know that fave − 2,
so the value of c satis es

f scd − fave − 2
(_1, 2)

Therefore 1 1 c2 − 2 so c2 − 1
fave=2
So in this case there happen to be two numbers c − 61 in the interval f21, 2g that
_1 0 1 2 x work in the Mean Value Theorem for Integrals. ■

FIGURE Examples 1 and 3 are illustrated by Figure 3.

EXERCISES .
– Find the average value of the function on the given interval. (c) Sketch the graph of f and a rectangle whose area is the
same as the area under the graph of f.
. f sxd − 4x 2 x , f0, 4g
2

. f sxd − sx 2 3d2, f2, 5g . f sxd − ln x, f1, 3g

. f sxd − sin 4x, f2 , g
; . f sxd − 2 sin x 2 sin 2x, f0, g
. tsxd − s
3
x , f1, 8g
; . f sxd − 2xys1 1 x 2 d 2, f0, 2g
. f s d − sec2s y2d, f0, y2g
. hsxd − cos 4 x sin x, f0, g
. Find the average value of f on f0, 8g.
. hsud − s3 2 2ud , f21, 1g 21
y

–
1
(a) Find the average value of f on the given interval.
(b) Find c such that fave − f scd.
0 2 4 6 x
 CHAPTER | Applications of Integrals

. The velocity graph of an accelerating car is shown. . Measles pathogenesis In Section 5.1 we modeled the
(a) Use the Midpoint rule to estimate the average velocity infection level of the measles virus in a patient by the
of the car during the rst 12 seconds. function
(b) At what time was the instantaneous velocity equal to the f std − 2t st 2 21dst 1 1d
average velocity?
where t is measured in days and f std is measured in the
√ number of infected cells per mL of blood plasma. Over the
(km/h)
60
course of the 21-day infection, what is the average level of
infection?
40 . Length of a fish For a sh that starts life with a length
20
of 1 cm and has a maximum length of 30 cm, the von
Bertalanffy growth model predicts that the growth rate is
29e 2a cmyyear. What is the average length of the sh over
0 4 8 12 t (seconds) its rst ve years?

. In a certain city the temperature (in °F) t hours after 9 . Breathing is cyclic and a full respiratory cycle from the
was modeled by the function beginning of inhalation to the end of exhalation takes about
5 s. The maximum rate of air ow into the lungs is about
t
Tstd − 50 1 14 sin 0.5 Lys. This explains, in part, why the function
12
Find the average temperature during the period from 9 1 2 t
f std − sin
to 9 . 2 5
. If a cup of coffee has temperature 95°C in a room where the has often been used to model the rate of air ow into the
temperature is 20°C, then, according to Newton’s Law of lungs. If inhalation occurs during the interval 0 < t < 2.5,
Cooling, the temperature of the coffee after t minutes is what is the average rate of air ow during inhalation?
Tstd − 20 1 75e 2ty50. What is the average temperature of . Blood flow The velocity v of blood that ows in a blood
the coffee during the rst half hour? vessel with radius R and length l at a distance r from the
. The population of Indonesia from 1950 to 2000 has been central axis is
modeled with the function P
vsrd − sR 2 2 r 2 d
Pstd − 83e 0.18t 4 l

where P is measured in millions and t is measured in where P is the pressure difference between the ends of the
years with t − 0 in the year 1950. What was the average vessel and is the viscosity of the blood (see Example
population of Indonesia in the second half of the 20th 3.3.9). Find the average velocity (with respect to r) over the
century? interval 0 < r < R. Compare the average velocity with the
. Blood alcohol concentration In Section 3.1 we modeled maximum velocity.
. If f is continuous and y1 f sxd dx − 8, show that f takes on
3
the BAC of male adult subjects after rapid consumption of
15 mL of ethanol (corresponding to one alcoholic drink) by the value 4 at least once on the interval f1, 3g.
the concentration function
. Find the numbers b such that the average value of
Cstd − 0.0225te 20.0467t f sxd − 2 1 6x 2 3x 2 on the interval f0, bg is equal to 3.
where t is measured in minutes after consumption and Cstd . Prove the Mean Value Theorem for Integrals by applying
is measured in mgymL. What was the average BAC during the Mean Value Theorem for derivatives (see Section 4.2) to
x
the rst hour? the function Fsxd − ya f std dt.

. Further Applications to Biology

In Chapter 5 and in Sections 6.1 and 6.2 we presented several applications of integration
to biology. In this section we consider additional examples of how integrals are used in
biological settings: survival and renewal of populations, blood ow in veins and arteries,
and cardiac output. Other applications are explored in the exercises.
 SECTION . | Further Applications to Biology

■ Survival and Renewal

A population may be continually adding members while some of the existing members
die. If we can model how these changes occur with suitable functions, we can predict the
population size at any point in the future.
Suppose we start with an initial population P0 and new members are added at the rate
Rstd, where t is the number of years from now. We call R a renewal function. In addi-
tion, the proportion of the population that survives at least t years from now is given by
a survival function S. [So if Ss5d − 0.8, 80% of the current population remains after
5 years.]
To predict the population in T years, we rst note that Sstd ? P0 members of the cur-
rent population survive. To account for the newly added members, we divide the time
interval f0, Tg into n subintervals, each of length Dt − Tyn, and let ti be the right end-
point of the ith subinterval. During this time interval, approximately Rstid Dt members
are added, and the proportion of them that survive until time T is given by SsT 2 tid.
Thus the remaining members of those added during this time interval is

sproportion survivingdsnumber of membersd − SsT 2 tid Rstid Dt

Then the total number of new members to the population who survive after T years is
approximately
n

o SsT 2 tid Rstid Dt

i−1

If we let n l `, this Riemann sum approaches the integral

y0
T
SsT 2 td Rstd dt

By adding this integral to the number of initial members who survived, we get the total
population after T years.

A population begins with P0 members and members are added at a rate given by
the renewal function Rstd, where t is measured in years. The proportion of the
population that remains after t years is given by the survival function Sstd. Then
the population T years from now is given by
Equation 1 is also valid if t represents
any other unit of time, such as weeks
y0
T
or months.
() PsT d − SsT d ? P0 1 SsT 2 tdRstd dt

EXAMPLE | Predicting a future population There are currently 5600 trout

in a lake and the trout are reproducing at the rate Rstd − 720e 0.1t shyyear. However,
pollution is killing many of the trout; the proportion that survive after t years is given
by Sstd − e 20.2t. How many trout will there be in the lake in 10 years?

SOLUTION We have P0 − 5600 and T − 10, so by Formula 1, the population in ten

years is
Ps10d − Ss10d ? 5600 1 y Ss10 2 td Rstd dt
10

− 5600e20.2s10d 1 y e 20.2s102td ? 720e 0.1t dt

10

− 5600e 22 1 720 y e 0.3t22 dt

10

0
 CHAPTER | Applications of Integrals

Writing e 0.3t22 as e 0.3te 22 gives

Ps10d − 5600e 22 1 720e 22 y e 0.3t dt

10

− 5600e 22 1 720e 22
e 0.3t
0.3
G 10

720 22 3
− 5600e 22 1 e se 2 e 0 d
0.3

− 5600e 22 1 2400 se 2 e 22 d < 6956.95

Thus we predict that there will be about 6960 trout in the lake ten years from now. ■

Although we presented Formula 1 in the context of populations, it applies to other

settings as well, such as administering a drug over time as the body works to eliminate
the drug. The exercises investigate additional applications.

■ Blood Flow
In Example 3.3.9 we discussed the law of laminar ow:
P
vsrd − sR 2 2 r 2 d
4 l
which gives the velocity v of blood that ows along a blood vessel with radius R and
length l at a distance r from the central axis, where P is the pressure difference between
the ends of the vessel and is the viscosity of the blood (see Figure 1).

R r

FIGURE
Blood ow in an artery l

In order to compute the rate of blood ow, or ux (volume per unit time), we consider
smaller, equally spaced radii r1, r2 , . . . . The approximate area of the ring with inner
Îr radius ri21 and outer radius ri is

ri 2 ri Dr where Dr − ri 2 ri21

(See Figure 2.) If Dr is small, then the velocity is almost constant throughout this ring
and can be approximated by vsri d. Thus the volume of blood per unit time that ows
across the ring is approximately
FIGURE
s2 ri Drd vsri d − 2 ri vsri d Dr
and the total volume of blood that ows across a cross-section per unit time is about
n

o2
i−1
ri vsri d Dr

This approximation is illustrated in Figure 3. Notice that the velocity (and hence the
volume per unit time) increases toward the center of the blood vessel. The approxima-
FIGURE tion gets better as n increases. When we take the limit as n l ` we get an integral that
gives the exact value of the ux (or discharge), which is the volume of blood that passes
 SECTION . | Further Applications to Biology

a cross-section per unit time:

F − y 2 r vsrd dr
R

P
−y 2 r
R
sR 2 2 r 2 d dr
0 4 l

F G
r−R
P P r2 r4
y0
R
− sR 2r 2 r 3 d dr − R2 2
2 l 2 l 2 4 r−0

−
P
2 l
F R4
2
2
R4
4
G −
PR 4
8 l

The resulting equation

PR 4
( ) F−
8 l
is called Poiseuille’s Law; it shows that the ux is proportional to the fourth power of
the radius of the blood vessel.
In Exercise 10 you are asked to investigate the effect on blood pressure if the radius
of an artery is reduced to three-fourths of its normal value.

■ Cardiac Output
aorta Figure 4 shows the human heart and associated blood vessels. Blood returns from the
vein pulmonary body through the veins, enters the right atrium of the heart, and is pumped to the lungs
arteries
through the pulmonary arteries for oxygenation. It then ows back into the left atrium
pulmonary pulmonary
arteries veins through the pulmonary veins and then out to the rest of the body through the aorta. The
cardiac output of the heart is the volume of blood pumped by the heart per unit time,
that is, the rate of ow into the aorta.
The dye dilution method is used to measure the cardiac output. Dye is injected into the
right left
atrium atrium right atrium and ows through the heart into the aorta. A probe inserted into the aorta
measures the concentration of the dye leaving the heart at equally spaced times over a time
pulmonary interval f0, T g until the dye has cleared. Let cstd be the concentration of the dye at time t.
veins If we divide f0, T g into subintervals of equal length Dt, then the amount of dye that ows
past the measuring point during the subinterval from t − ti21 to t − ti is approximately
vein sconcentration of dye in blooddsvolume of bloodd − csti dsF Dtd

FIGURE where F is the rate of ow that we are trying to determine. Thus the total amount of
dye is approximately
cst 1dFDt 1 cst 2dFDt 1 ∙ ∙ ∙ 1 cstndFDt
and, letting n l `, we nd that the amount of dye is

y0 cstdF dt − F y cstd dt
T T
A−
0

Thus the cardiac output is given by

A
( ) F−
y0
T
cstd dt

where the amount of dye A is known and the integral can be approximated from the
concentration readings.
 CHAPTER | Applications of Integrals

t cstd t cstd EXAMPLE | Cardiac output A 5-mg dose (called a bolus) of dye is injected
into a right atrium. The concentration of the dye (in milligrams per liter) is measured
0 0 6 6.1 in the aorta at one-second intervals as shown in the chart. Estimate the cardiac output.
1 0.4 7 4.0
2 2.8 8 2.3 SOLUTION Here A − 5 and T − 10. We can use the Midpoint Rule with n − 5
3 6.5 9 1.1 subdivisions to approximate the integral of the concentration. Then Dt − 2 and
4 9.8 10 0
y0
10
5 8.9 cstd dt < fcs1d 1 cs3d 1 cs5d 1 cs7d 1 cs9dg Dt

− f0.4 1 6.5 1 8.9 1 4.0 1 1.1g s2d

− 41.8
Thus Formula 3 gives the cardiac output to be
A 5
F− < < 0.12 Lys − 7.2 Lymin ■
y0 41.8
10
cstd dt

EXERCISES .
. Animal survival and renewal An animal population bloodstream over time so that the fraction e 20.2t remains after
currently has 7400 members and is reproducing at the rate t hours. The patient currently has 80 mg of the drug present
Rstd − 2240 1 60t membersyyear. The proportion of mem- in the bloodstream. How much will be present in 24 hours?
bers that survive after t years is given by Sstd − 1yst 1 1d.
. Water pollution A contaminant is leaking into a lake at a
(a) How many of the original members survive four years?
rate of
(b) How many new members are added during the next
four years? Rstd − 1600e 0.06t gallonsyh
(c) Explain why the animal population four years from now
is not the same as the sum of your answers from parts (a) Enzymes have been added to the lake that neutralize the
and (b). contaminant over time so that after t hours the fraction of
the contaminant that remains is Sstd − e 20.32t. If there are
. City population A city currently has 36,000 residents and
currently 10,000 gallons of the contaminant in the lake, how
is adding new residents steadily at the rate of 1600 per year.
many gallons are present in the lake 18 hours from now?
If the proportion of residents that remain after t years is
given by Sstd − 1yst 1 1d, what is the population of the city . Insect survival and renewal Sterile fruit ies are used
seven years from now? in an experiment where the proportion that survive at least
t days is given by e 20.15t. If the experiment begins with
. Insect survival and renewal A population of insects
200 fruit ies, and ies are added at the rate of 5 per hour,
currently numbers 22,500 and is increasing at a rate of
how many ies are present 14 days after the start of the
Rstd − 1225e 0.14t insectsyweek. If the survival function for
experiment?
the insects is Sstd − e 20.2t, where t is measured in weeks,
how many insects are there after 12 weeks? . Blood flow Use Poiseuille’s Law to calculate the rate
of ow in a small human artery where we can take
. Animal survival and renewal There are currently 3800
− 0.027 dyn ? sycm 2, R − 0.008 cm, l − 2 cm, and
birds of a particular species in a national park and their
P − 4000 dynycm 2.
number is increasing at a rate of Rstd − 525e 0.05t birdsyyear.
If the proportion of birds that survive t years is given by . Blood flow High blood pressure results from constriction
Sstd − e 20.1t, what do you predict the bird population will be of the arteries. To maintain a normal ow rate ( ux), the
10 years from now? heart has to pump harder, thus increasing the blood pressure.
Use Poiseuille’s Law to show that if R0 and P0 are normal
. Drug concentration A drug is administered intravenously
values of the radius and pressure in an artery and the
to a patient at the rate of 12 mgyh. The patient’s body elim-
constricted values are R and P, then for the ux to remain
inates the drug over time so that after t hours the proportion
constant, P and R are related by the equation
that remains is e 20.25t. If the patient currently has 50 mg of
the drug in her bloodstream, how much of the drug is present
eight hours from now?
P
P0
− SDR0
R
4

. Drug concentration A patient receives a drug at a con-

stant rate of 30 mgyh. The drug is eliminated from the Deduce that if the radius of an artery is reduced to three-
 SECTION . | Volumes

fourths of its former value, then the pressure is more than . Cardiac output The graph of the concentration function
tripled. cstd is shown after a 7-mg injection of dye into a heart. Use
the Midpoint Rule to estimate the cardiac output.
. Cardiac output The dye dilution method is used to mea-
sure cardiac output with 6 mg of dye. The dye concentra-
tions, in mgyL, are modeled by cstd − 20te20.6t, 0 < t < 10, y
where t is measured in seconds. Find the cardiac output. (mg/ L)
[Hint: Integration by parts is required.]
6
. Cardiac output After an 8-mg injection of dye, the
readings of dye concentration, in mgyL, at two-second 4
intervals are as shown in the table. Use the Midpoint Rule to
estimate the cardiac output. 2

t cstd t cstd 0 2 4 6 8 10 12 14 t (seconds)

0 0 12 3.9
2 2.4 14 2.3
4 5.1 16 1.6 . Drug administration A patient is continually receiving a
6 7.8 18 0.7 drug. If the drug is eliminated from the body over time so
that the fraction that remains after t hours is e 20.4t, at what
8 7.6 20 0
constant rate should the drug be administered to maintain a
10 5.4
steady level of the drug in the bloodstream?

. Volumes
In trying to nd the volume of a solid we face the same type of problem as in nding
areas. We have an intuitive idea of what volume means, but we must make this idea pre-
cise by using calculus to give an exact de nition of volume.
We start with a simple type of solid called a cylinder (or, more precisely, a right cylin-
der). As illustrated in Figure 1(a), a cylinder is bounded by a plane region B1, called the
base, and a congruent region B2 in a parallel plane. The cylinder consists of all points on
line segments that are perpendicular to the base and join B1 to B2. If the area of the base is
A and the height of the cylinder (the distance from B1 to B2) is h, then the volume V of the
cylinder is de ned as
V − Ah
In particular, if the base is a circle with radius r, then the cylinder is a circular cylinder
with volume V − r 2h [see Figure 1(b)], and if the base is a rectangle with length l and
width w, then the cylinder is a rectangular box (also called a rectangular parallelepiped)
with volume V − lwh [see Figure 1(c)].

B
h
h h
r w
B∞
l
(a) Cylinder V=Ah (b) Circular cylinder V=πr@h (c) Rectangular box V=lwh
FIGURE
For a solid S that isn’t a cylinder we rst “cut” S into pieces and approximate each
piece by a cylinder. We estimate the volume of S by adding the volumes of the cylinders.
We arrive at the exact volume of S through a limiting process in which the number of
pieces becomes large.
 CHAPTER | Applications of Integrals

We start by intersecting S with a plane and obtaining a plane region that is called
a cross-section of S. Let Asxd be the area of the cross-section of S in a plane Px perpen-
dicular to the x-axis and passing through the point x, where a < x < b. (See Fig-
ure 2. Think of slicing S with a knife through x and computing the area of this slice.) The
cross-sectional area Asxd will vary as x increases from a to b.

Px

A(a) A(x)
A(b)

0 a x b x
FIGURE

Let’s divide S into n “slabs” of equal width Dx by using the planes Px , Px , . . . to slice
1 2

the solid. (Think of slicing a loaf of bread.) If we choose sample points x*i in fx i21, x i g,
we can approximate the ith slab Si (the part of S that lies between the planes Px and Px ) i21 i

by a cylinder with base area Asx*di and “height” Dx. (See Figure 3.)

y y
Îx

0 a b x 0 a=x¸ ⁄ x™ ‹ x¢ x∞ xß x¶=b x
xi-1 x*i xi

FIGURE
The volume of this cylinder is Asx*i d Dx, so an approximation to our intuitive concep-
tion of the volume of the ith slab Si is

VsSi d < Asx*i d Dx

Adding the volumes of these slabs, we get an approximation to the total volume (that is,
what we think of intuitively as the volume):

n
V< o Asx*i d Dx
i−1

This approximation appears to become better and better as n l `. (Think of the slices
as becoming thinner and thinner.) Therefore we de ne the volume as the limit of these
 SECTION . | Volumes

sums as n l `. But we recognize the limit of Riemann sums as a de nite integral and
so we have the following de nition.

It can be proved that this de nition is Definition of Volume Let S be a solid that lies between x − a and x − b. If
independent of how S is situated with the cross-sectional area of S in the plane Px, through x and perpendicular to the
respect to the x-axis. In other words, x-axis, is Asxd, where A is a continuous function, then the volume of S is
no matter how we slice S with parallel
n
planes, we always get the same answer
o Asx*i d Dx − ya Asxd dx
b
for V. V − lim
n l ` i−1

When we use the volume formula V − ya Asxd dx, it is important to remember that
b

Asxd is the area of a moving cross-section obtained by slicing through x perpendicular

to the x-axis.
Notice that, for a cylinder, the cross-sectional area is constant: Asxd − A for all x. So
our de nition of volume gives V − yab A dx − Asb 2 ad; this agrees with the formula
V − Ah.

EXAMPLE | Volume of a liver A CAT scan produces equally spaced cross-

sectional views of a human organ that provide information about the organ otherwise
obtained only by surgery. For example, such measurements of liver volume can be
related to diseases such as cirrhosis.1 Suppose that a CAT scan of a human liver shows
MedicalRF.com / Getty Images

cross-sections spaced 2 cm apart. The liver is 20 cm long and the cross-sectional areas,
in square centimeters, are 0, 39, 63, 128, 117, 106, 94, 79, 58, 18, and 0. Use the Mid-
point Rule to estimate the volume of the liver.

SOLUTION In using the Midpoint Rule we will use n − 5 subintervals; their mid-
points are 2, 6, 10, 14, and 18. If Asxd is the area of the cross-section of the liver at a
distance of x centimeters from one end, then the volume is

V − y Asxd dx
20

20 2 0
− fAs2d 1 As6d 1 As10d 1 As14d 1 As18dg
5
− 4 f39 1 128 1 106 1 79 1 18g

− 4 ? 370 − 1480

y The volume of the liver is approximately 1480 cm 3. ■

EXAMPLE | Show that the volume of a sphere of radius r is V − 43 r 3.

y
r SOLUTION If we place the sphere so that its center is at the origin (see Figure 4),
then the plane Px intersects the sphere in a circle whose radius (from the Pythagorean
_r 0 r x Theorem) is y − sr 2 2 x 2 . So the cross-sectional area is
x
Asxd − y2 − sr 2 2 x 2 d

1. J.-Y. Zhu et al., “Measurement of Liver Volume and Its Clinical Signi cance in Cirrhotic Portal Hyperten-
FIGURE sive Patients,” World Journal of Gastroenterology 5 (1999): 525–26.
 CHAPTER | Applications of Integrals

Using the de nition of volume with a − 2r and b − r, we have

V − y Asxd dx − y
r r
sr 2 2 x 2 d dx
2r 2r

y0 sr 2 2 x 2 d dx
r
−2 (The integrand is even.)

F G S D
r
x32 r3
−2 r x2 −2 r3 2
3 0
3

− 43 r 3 ■

Figure 5 illustrates the de nition of volume when the solid is a sphere with radius
r − 1. From the result of Example 2, we know that the volume of the sphere is 43 , which
is approximately 4.18879. Here the slabs are circular cylinders, or disks, and the three
parts of Figure 5 show the geometric interpretations of the Riemann sums
n n

o Asxi d Dx − i−1
i−1
o s12 2 x i2 d Dx

when n − 5, 10, and 20 if we choose the sample points x*i to be the midpoints xi. Notice
TEC Visual 6.4A shows an animation that as we increase the number of approximating cylinders, the corresponding Riemann
of Figure 5. sums become closer to the true volume.

(a) Using 5 disks, VÅ4.2726 (b) Using 10 disks, VÅ4.2097 (c) Using 20 disks, VÅ4.1940

FIGURE Approximating the volume of a sphere with radius 1

You have probably seen the formula V − 43 r 3 for the volume of a sphere before. But
calculus is required to prove it, so this may be the rst time you have have seen it proved.
© Lightspring / Shutterstock.com

This formula is useful in biology because the shape of a tumor is often modeled by a
sphere. (See, for instance, Example 3.5.14.) And the shape of a bacterium can often be
modeled as a circular cylinder capped by two hemispheres as shown in the margin.

EXAMPLE | Find the volume of the solid obtained by rotating about the x-axis the
region under the curve y − sx from 0 to 1. Illustrate the de nition of volume by
sketching a typical approximating cylinder.

SOLUTION The region is shown in Figure 6(a). If we rotate about the x-axis, we get
the solid shown in Figure 6(b). When we slice through the point x, we get a disk with
radius sx . The area of this cross-section is

ssx d 2 −
© Sebastian Tomus / Shutterstock.com
Asxd − x
 SECTION . | Volumes

and the volume of the approximating cylinder (a disk with thickness Dx) is
Asxd Dx − x Dx
The solid lies between x − 0 and x − 1, so its volume is

G
1
x2
y0 Asxd dx − y
1 1
V− x dx − − ■
0 2 0
2

y y
y=œ„x

Did we get a reasonable answer in

Example 3? As a check on our work,
let’s replace the given region by a
œ„x
square with base f0, 1g and height 1. If
we rotate this square, we get a cylinder
with radius 1, height 1, and volume
0 x 1 x 0 1 x
? 12 ? 1 − . We computed that the
given solid has half this volume. That
seems about right.

Îx

FIGURE (a) (b)

EXAMPLE | The region enclosed by the curves y − x and y − x 2 is rotated about

the x-axis. Find the volume of the resulting solid.

SOLUTION The curves y − x and y − x 2 intersect at the points s0, 0d and s1, 1d. The
region between them, the solid of rotation, and a cross-section perpendicular to the
x-axis are shown in Figure 7. A cross-section in the plane Px has the shape of a washer
(an annular ring) with inner radius x 2 and outer radius x, so we nd the cross-sectional
area by subtracting the area of the inner circle from the area of the outer circle:
Asxd − x2 2 sx 2 d2 − sx 2 2 x 4 d
TEC Visual 6.4B shows how solids of Therefore we have

F G
revolution are formed. 1
x3 x5 2
V − y Asxd dx − y
1 1
sx 2 x d dx −
2 4
2 −
0 0 3 5 0
15

y y

(1, 1)
y=x A(x)
y=≈
≈

(0, 0) x 0 x x

FIGURE (a) ( b) (c) ■

 CHAPTER | Applications of Integrals

The solids in Examples 2–4 are all called solids of revolution because they are
obtained by revolving a region about a line. In general, we calculate the volume of a
solid of revolution by using the basic de ning formula

V − y Asxd dx
b

and we nd the cross-sectional area Asxd in one of the following ways:

■ If the cross-section is a disk (as in Examples 2 and 3), we nd the radius of the
disk (in terms of x) and use
A− sradiusd2

■ If the cross-section is a washer (as in Example 4), we nd the inner radius r in and
outer radius rout from a sketch (as in Figures 7 and 8) and compute the area of the
washer by subtracting the area of the inner disk from the area of the outer disk:

A− souter radiusd2 2 sinner radiusd2

rin
rout

FIGURE

We now nd the volume of a solid that is not a solid of revolution.

TEC Visual 6.4C shows how the solid EXAMPLE | Figure 9 shows a solid with a circular base of radius 1. Parallel
in Figure 9 is generated. cross-sections perpendicular to the base are equilateral triangles. Find the volume of
the solid.

SOLUTION Let’s take the circle to be x 2 1 y 2 − 1. The solid, its base, and a typical
cross-section at a distance x from the origin are shown in Figure 10.

y
y B(x, y)
≈
y=œ„„„„„„ C
C
y

B
0 x x
œ 3y
œ„
y _1 0 1 x
A 60° 60°
A A B
y y

x (a) The solid (b) Its base (c) A cross-section

FIGURE FIGURE
Computer-generated picture
of the solid in Example 5 Since B lies on the circle, we have y − s1 2 x 2 and so the base of the triangle
| |
ABC is AB − 2s1 2 x 2 . Since the triangle is equilateral, we see from Figure 10(c)
 SECTION . | Volumes

that its height is s3 y − s3 s1 2 x 2 . The cross-sectional area is therefore

Asxd − 12 ? 2 s1 2 x 2 ? s3 s1 2 x 2 − s3 s1 2 x 2 d

and the volume of the solid is

V − y Asxd dx − y s3 s1 2 x 2 d dx
1 1

21 21

F G
1
x3 4 s3
− 2 y s3 s1 2 x 2 d dx − 2 s3 x 2
1
− ■
0 3 0
3

EXERCISES .
– Find the volume of the solid obtained by rotating the . A log 10 m long is cut at 1-meter intervals and its cross-
region bounded by the given curves about the x-axis. Sketch sectional areas A (at a distance x from the end of the log)
the region, the solid, and a typical disk or washer. are listed in the table. Use the Midpoint Rule with n − 5 to
estimate the volume of the log.
. y − 2 2 12 x, y − 0, x − 1, x − 2
. y − 1 2 x 2, y − 0 x smd A sm 2 d x smd A sm 2 d
. y − sx 2 1 , y − 0, x − 5 0 0.68 6 0.53
. y − s25 2 x 2 , y − 0, x − 2, x − 4 1 0.65 7 0.55
. y − x 3, y − x, x > 0 2 0.64 8 0.52
3 0.61 9 0.50
. y − 14 x 2, y − 5 2 x 2 4 0.58 10 0.48
5 0.59
– Here we rotate about the y-axis instead of the x-axis. Find
the volume of the solid obtained by rotating the region bounded . (a) If the region shown in the gure is rotated about the
by the given curves about the y-axis. Sketch the region, the x-axis to form a solid, use the Midpoint Rule with
solid, and a typical disk. n − 4 to estimate the volume of the solid.
. x − 2sy , x − 0, y − 9 y
. y − ln x, y − 1, y − 2, x − 0 4

2
. Volume of a pancreas A CAT scan of a human pancreas
shows cross-sections spaced 1 cm apart. The pancreas is
12 cm long and the cross-sectional areas, in square centi- 0 2 4 6 8 10 x
meters, are 0, 7.7, 15.2, 18.0, 10.3, 10.8, 9.7, 8.7, 7.7, 5.5,
4.0, 2.7, and 0. Use the Midpoint Rule to estimate the (b) Estimate the volume if the region is rotated about the
volume of the pancreas. y-axis. Again use the Midpoint Rule with n − 4.

. Volume of a bird’s egg

(a) A model for the shape of a bird’s egg is obtained by
© Sebastian Kaulitzki / Shutterstock.com

CAS
rotating about the x-axis the region under the graph of

f sxd − sax 3 1 bx 2 1 cx 1 dds1 2 x 2

Use a CAS to nd the volume of such an egg.

(b) For a red-throated loon, a − 20.06, b − 0.04,
c − 0.1, and d − 0.54. Graph f and nd the volume of
an egg of this species.
 CHAPTER | Applications of Integrals

– Find the volume of the described solid S. . Find the volume common to two spheres, each with radius
r, if the center of each sphere lies on the surface of the other
. S is a right circular cone with height h and base radius r
sphere.
. The base of S is a circular disk with radius r. Parallel cross-
sections perpendicular to the base are squares. . Find the volume common to two circular cylinders, each
with radius r, if the axes of the cylinders intersect at right
. The base of S is an elliptical region with boundary curve angles.
9x 2 1 4y 2 − 36. Cross-sections perpendicular to the x-axis
are isosceles right triangles with hypotenuse in the base.

. The base of S is a circular disk with radius r. Parallel cross-

sections perpendicular to the base are isosceles triangles
with height h and unequal side in the base.
(a) Set up an integral for the volume of S.
(b) By interpreting the integral as an area, nd the volume
of S.

Chapter REVIEW
CONCEPT CHECK

. Draw two typical curves y − f sxd and y − tsxd, where . If we have survival and renewal functions for a population,
f sxd > tsxd for a < x < b. Show how to approximate the how do we predict the size of the population T years from
area between these curves by a Riemann sum and sketch the now?
corresponding approximating rectangles. Then write an
. (a) What is the cardiac output of the heart?
expression for the exact area.
(b) Explain how the cardiac output can be measured by the
dye dilution method.
. Suppose that Sue runs faster than Kathy throughout a
1500-meter race. What is the physical meaning of the area . (a) Suppose S is a solid with known cross-sectional areas.
between their velocity curves for the rst minute of the race? Explain how to approximate the volume of S by a
Riemann sum. Then write an expression for the exact
. (a) What is the average value of a function f on an volume.
interval fa, bg? (b) If S is a solid of revolution, how do you nd the cross-
(b) What does the Mean Value Theorem for Integrals say? sectional areas?
What is its geometric interpretation? Answers to the Concept Check can be found on the back
endpapers.

EXERCISES

– Find the area of the region bounded by the given curves. . y − 1 2 2x 2, y− x| |

2 2
. y−x , y − 4x 2 x
. x 1 y − 0, x − y 2 1 3y
2
. y − 1yx, y − x , y − 0, x − e
 CHAPTER | Review

. MRI brain scan Shown is a cross-section of a human that remain after t months is Sstd − e 20.09t, how many sh
brain obtained with an MRI. Use the Midpoint Rule to will be in the lake in three years?
estimate the area of the cross-section.
. Cardiac output After a 6-mg injection of dye into a heart,
y (cm) the readings of dye concentration, in mgyL, at two-second
intervals are as shown in the table. Use the Midpoint Rule to
15
estimate the cardiac output.
© Allison Herreid / Shutterstock.com

10 t Cstd t Cstd
0 0 14 4.7
2 1.9 16 3.3
5 4 3.3 18 2.1
6 5.1 20 1.1
8 7.6 22 0.5
0 5 10 15 x (cm) 10 7.1 24 0
12 5.8
. Birth and death rates The birth rate of a population is
bstd − 1240e 0.0197t people per year and the death rate is . Find the volume of the solid obtained by rotating about
dstd − 682e 0.008t people per year. Find the area between the x-axis the region bounded by the curves y − e 22 x,
these curves for 0 < t < 20. What does this area represent? y − 1 1 x, and x − 1.

. Find the average value of the function f std − t sinst 2 d on . Let 5 be the region bounded by the curves y − tansx 2 d,
the interval f0, 10g. x − 1, and y − 0. Use the Midpoint Rule with n − 4 to
. Find the average value of the function f sxd − x 2 s1 1 x 3 estimate the following quantities.
on the interval f0, 2g. (a) The area of 5
(b) The volume obtained by rotating 5 about the x-axis
. Antibiotic pharmacokinetics When an antibiotic tablet
is taken, the concentration of the antibiotic in the blood- . Let 5 be the region in the rst quadrant bounded by the
stream is modeled by the function curves y − x 3 and y − 2x 2 x 2. Calculate the following
quantities.
Cstd − 8se 20.4t 2 e 20.6t d (a) The area of 5
(b) The volume obtained by rotating 5 about the x-axis
where the time t is measured in hours and C is measured in
mgymL. What is the average concentration of the antibiotic . Find the volumes of the solids obtained by rotating the
during the rst two hours? region bounded by the curves y − x and y − x 2 about the
following lines.
. Salicylic acid pharmacokinetics In a study of the effects
(a) The x-axis (b) The y-axis
of aspirin, salicylic acid was formed and its concentration
was modeled by the function ; . Let 5 be the region bounded by the curves y − 1 2 x 2 and
y − x 6 2 x 1 1. Estimate the following quantities.
Cstd − 11.4te 2t
(a) The x-coordinates of the points of intersection of the
where the time t is measured in hours and C is measured in curves
mgymL. What is the average concentration of the salicylic (b) The area of 5
acid during the rst four hours? (c) The volume generated when 5 is rotated about the
x-axis
. Survival and renewal Suppose a city’s population is
currently 75,000 and the renewal function is . The base of a solid is a circular disk with radius 3. Find the
0.05t volume of the solid if parallel cross-sections perpendicular
Rstd − 3200e
to the base are isosceles right triangles with hypotenuse
If the survival function is Sstd − e 20.1 t, predict the popula- lying along the base.
tion in 10 years.
. The height of a monument is 20 m. A horizontal cross-
. Animal survival and renewal The sh population in section at a distance x meters from the top is an equilateral
a lake is currently 3400 and is increasing at a rate of triangle with side 14 x meters. Find the volume of the
Rstd − 650e 0.04t sh per month. If the proportion of sh monument.
 CHAPTER | Applications of Integrals

CASE STUDY d Kill Curves and Antibiotic Effectiveness

In this case study we have explored the relationship between the magni-
tude of antibiotic treatment and the effectiveness of the treatment. To do so,
in Case Study 1b we showed that a suitable model for the size of the bacteria
population Pstd (in CFUymL) as a function of time t (in hours) is given by the piecewise
de ned function

Recall that MIC is a constant referred

( a) Pstd − H6e ty3
12
if t , 2.08
if t > 2.08
to as the minimum inhibitory concen-
tration of the antibiotic and c0 is the
if c0 , MIC, where MIC − 0.013 m gymL. On the other hand, if c0 . MIC
concentration at time t − 0.
6e 2ty20 if t , a
( b) Pstd − 6Ae ty3 if a < t , b
12 if t > b

where the constants a, b, and A are de ned by a − 5.7 lns77c0d, b − 6.6 ln s77c0d 1 2.08,
and A − s77c0d22.2.
In Case Study 1a we used these so-called kill curves to plot the relationship between
(a measure of the magnitude of antibiotic treatment) and two different measures of
the killing effectiveness, denoted by D and T. The quantity D is the drop in population
size before the population rebound occurs, and T is the time taken to reduce the bacteria
population size to 90% of its initial size. (Refer to Figure 1.)

P(t) (CFU/mL)
12

10
Bacteria population size

8 Area=I

6
6§ 0.9=5.4
FIGURE 4 Î
Three measures of killing effectiveness. 2
The blue curve is bacteria population
size in the absence of antibiotic. The 0 T 10 20 30 t (hours)
red curve is bacteria population size in
the presence of antibiotic. Time

Now that we have a model that works reasonably well, we can use it to make predic-
tions about other patterns, and then compare these with available data. As an example,
another measure of the killing effectiveness of an antibiotic is the area I between the
population size curve in the absence of antibiotic, and the kill curve in the presence of the
antibiotic as shown in Figure 1. In many cases this measure might be preferable because
it incorporates both the drop in bacteria population size, and the length of time for which
this reduced population size is maintained. Let’s see what our model predicts about the
relationship between I and the magnitude of antibiotic treatment .
. Suppose that a . 2.08 [that is, 5.7 lns77c0d . 2.08]. Find an expression for I in
the modeled populations in terms of c0. You should assume that c0 . MIC.
 CASE STUDY d | Kill Curves and Antibiotic Effectiveness

. The result from Problem 1 should give you a function of the form I − tsc0d for
some function t. Substitute the values k − 0.175 and MIC − 0.013 into the
expression for obtained in Case Study 1c. This will give − hsc0d for some
function h.
. Using the concept of a function’s inverse, explain how to obtain an expression
giving I as a function of in terms of t and h 21. Find an explicit expression for
this function.
. Plot the function obtained in Problem 3.

The experimental kill curves shown in Figure 2 have also been used to quantify the
relationship between I and .1 In other words, the values of I and have been calculated
for each experimental kill curve in Figure 2. If we overlay these data points on the plot
from Problem 4, we obtain Figure 3. You can see that, again, our model predicts the
observed data reasonably well.
(CFU/mL)
0 0.019 0.038 0.075 0.15 0.3 0.6 1.2
12
(Colony-forming units/mL)

10
Bacteria population size

4
FIGURE
The kill curves of cipro oxacin for 2
E. coli when measured in a growth
chamber. The concentration of
0 10 20 30
cipro oxacin at t − 0 is indicated
above each curve (in mgymL). Time (hours)

I
300

200

FIGURE
100
Predicted relationship between
I and , along with experimental
observations obtained using the kill
curve data in Figure 2. 0 100 200 300 400 500

1. Adapted from A. Firsov et al., “Parameters of Bacterial Killing and Regrowth Kinetics and Antimicrobial
Effect in Terms of Area Under the Concentration-Time Curve Relationships: Action of Cipro oxacin against
Escherichia coli in an In Vitro Dynamic Model.” Antimicrobial Agents and Chemotherapy 41 (1997): 1281.
 CHAPTER | Applications of Integrals

CASE STUDY b Hosts, Parasites, and Time-Travel

In Case Study 2c you will derive a model for the dynamics of the geno-
types of Daphnia and its parasite. Recall that we are modeling a situation
involving two possible host genotypes (A and a) and two possible parasite
genotypes (B and b). Parasites of type B can infect only hosts of type A, while parasites
of type b can infect only hosts of type a. You will then derive an explicit solution of a
simpli ed version of the model in Case Study 2d. This will give the frequency of host
genotype A and parasite genotype B as functions of time. These functions are

qstd − 12 1 Mq cossct 2 q d
()
pstd − 12 1 Mp cossct 2 p d

where qstd is the predicted frequency of host genotype A at time t and pstd is the pre-
dicted frequency of the parasite genotype B at time t. In these equations q , p , and c are
positive constants, and Mq and Mp are positive constants that are strictly less than 12 (the
biological signi cance of these constants is explored in Case Study 2a).
In this part of the case study you will use Equations 1 to make predictions from the
model that can be compared with data from the experiments.
Recall that, in the experiment, a host from a xed layer of the sediment core was chal-
lenged with infection by a parasite from either the same layer, a layer above the xed
layer (that is, from its future), or a layer below it (that is, from its past). We can view dif-
ferent depths in the sediment core as representing different points of time in the history
of the Daphnia-parasite interaction (see Figure 1). In this way Equations 1 can equally
be viewed as specifying the frequency of the host and parasite genotypes as functions of
location in the sediment core. Increasing values of t correspond to shallower points in the
core as shown in Figure 2.

Shallow Young

Depth
Deep Shallow

1.0 p(t)
Depth

Time

0 q(t)

Time
core Old Young
(increasing t)

Deep Old

FIGURE FIGURE
 CASE STUDY b | Hosts, Parasites, and Time-Travel

In the experiment introduced in Case Study 2 on page xlvii, researchers chose a xed
depth and extracted a layer of sediment of width W centered around this depth. This
layer is shown in Figure 3.
Depth
Deep Shallow
W W
1.0 p(t)

0 q(t)
+D
D
Time
FIGURE Old Young

After the contents of this layer were completely mixed, hosts and parasites were
extracted at random from the mixture. Researchers also took deeper and shallower layers
(that represent the past and the future for hosts located in the layer at ) and completely
mixed each. The center of these layers was a distance D from the center of the focal layer
(see Figure 3). They then challenged hosts from the layer at with parasites from their
past (that is, from the layer with D , 0), present (the layer at ), and future (the layer
with D . 0). For each challenge experiment the fraction of hosts becoming infected was
measured.
We can use our model to predict the fraction of hosts infected. To do so, we rst need
to know the predicted frequency of hosts of type A in the layer at as well as the fre-
quency of the parasites of type B in the layer at 1 D.

. Consider a focal layer at location with width W as shown in Figure 4.

Depth
Deep Shallow
W

q(t)

W W
– –– + ––
2 2
Time
FIGURE Old Young

The frequency of host type A will vary across the depth of this layer as speci-
ed by the function qstd. Show that, when this layer is completely mixed, the
frequency of A in the mixture is given by

1 2 sin ( 12 cW)
qave s d − 1 q cossc 2 q d
2 cW

Hint: You might want to use the trigonometric identity

sinsx 1 yd 2 sinsx 2 yd − 2 cos x sin y

CHAPTER | Applications of Integrals

. Consider another layer at a location a distance D from , again with thickness W,

as shown in Figure 5. The frequency of parasite type B will vary across depth in
this layer, as speci ed by the function pstd. Show that, when this layer is com-
pletely mixed, the frequency of B in the mixture is given by
1 2 sin ( 12 cW)
pave s 1 Dd − 1 Mp cosscs 1 Dd 2 p d
2 cW

Depth
Deep Shallow
W

p(t)

W +D W
+D-–– +D+––
2 2
Time
FIGURE Old Young

. Suppose that hosts from the layer at are challenged with parasites from the
layer at 1 D. Use the facts that only B parasites can infect A hosts and only b
parasites can infect a hosts to explain why the fraction of hosts infected in this
challenge experiment is predicted to be

I s d − pave s 1 Ddqave s d 1 f1 2 pave s 1 Ddgf1 2 qave s dg

The nal step is to recognize that the experiment was actually conducted with several
different, randomly chosen depths . Therefore we need to average Is d in Problem 3
over the possible depths. Because Is d is periodic, we need only average over one period
of its cycle. Its average is therefore

1
y0
T
F− Is d d
T

where T − 2 yc is the period of I s d.

. Show that

4 sin 2 ( 12 cW)
( ) FsDd − 12 1 Mp Mq cosscD 2 *d
c 2W 2

where * − p 2 q.
Hint: You might want to use the trigonometric identity

cos x cos y − 12 scossx 1 yd 1 cossx 2 ydd

Equation 2 was used in Case Study 2a to predict the experimental outcome expected
under different conditions.