1

r
1s 1
i
Mill

VIP
!!!!!! ,':::

1 Lyby
r
'
L : 1
1

IHP^
>

—i-
-"Id
t -

—
W^^rf
*£
.-

J
mr ..-..

1
1
t- -

^ ~""™
"ratfR
SSI
S^^M»
v
r
^r

.-_^_
1 /I <
I I*:-

m m
,

1> 1

1
/I
_ ^™ 1

i /j
^™^'
1 1
 Finite Sequences and Series

13-1 Arithmetic and Geometric Sequences

Objective To identify an arithmetic or geometric sequence and find a

formula for its nth term.

A sequence is a set of numbers, called terms, arranged in a particular order. We

begin with two of the simplest types of sequences: arithmetic and geometric.

Arithmetic Sequences
A sequence of numbers is called an arithmetic sequence if the difference of any
two consecutive terms is constant. This difference is called the common difference.
The following sequences are all arithmetic.

2, 6. 10. 14. 18 difference =4

17, 10, 3, -4, -11, -18, . . . difference = -7
a, a + d, a + 2d, a + 3d, a + Ad, . . . difference =d

Geometric Sequences
A sequence of numbers is called a geometric sequence if the ratio of any two
consecutive terms is constant. This ratio is called the common ratio. The following
sequences are all geometric.

1, 3. 9. 27, 81, . ratio = 3

64 -32. 16, - -8. 4.... ratio = . i

a. ar ar ,
2
, ar
3
, ai
4
ratio = /

The same notation can be used for all types of sequences. The first term of a
sequence is often denoted by fj, the second and third terms by t2 and f3 , and so on.
The /7th term of the sequence is then denoted by t
n . Some sequences can be defined
by particular rules or formulas. If you have a formula for r„ in terms of n, you can
find the value of any term of the sequence. For example, suppose a sequence has
the formula
2
tn = n + 1.

2
Then: t x
= l + 1 = 2

t2 =2 2
+ 1 =5
2
t3 = 3 + 1 = 10

r4 = A" + 1 =17. and so on

Note that this sequence is neither arithmetic nor geometric.

•^ In the building pictured at the left, the architect Harry Wolf was inspired by proportions
derived from the Fibonacci sequence, producing a harmonious effect. This sequence runs: 1,1,2,
3, 5, 8, 13, . . . Can you find how Wolf has used these values?

Sequences and Series 473

 Formulas for the nth term of general arithmetic and geometric sequences are
given below. Notice how similar these formulas are.

Arithmetic Sequence: h + (n - \)d

T T T
To get the start with and add the
nth term. the first difference
term /; — 1 times.

.(»-l)
Geometric Sequence:

To get the start with and multiply

nth term, the first by the ratio

term n — 1 times.

Definition of a Sequence
By now you probably have a good intuitive idea of what a sequence is, but can you
give a precise definition of a sequence? In mathematics, a sequence is usually
defined to be a function whose domain is the set of positive integers. For example,
the sequence with nlh term

t„ = 4n - 2

can be thought of as the function

t(n) = An — 2, where n is a positive integer.

Since a sequence is a function, to graph a sequence such as 1, 4, 7, 10, ... ,

you plot the points (1, 1), (2, 4), (3, 7), (4, 10). You can see how an arithmetic
sequence is related to its graph by completing the following activity.

Activii

For each sequence, (a) graph the sequence and (b) find the slope of the line
containing the graph of the sequence.

1. 0, 1. 2. 3. 2. 18, 14, 10, . . . 3. 5.3, 6, 6.7,

The activity above demonstrates that the graph of any arithmetic sequence
consists of discrete points on a line whose slope is the same as the common
difference of the sequence.

Example 1 a. Find a formula for the ;?th term of the arithmetic sequence

3, 5, 7,....

b. Sketch the graph of the sequence.

474 Chapter Thirteen

 .

Solution a. Use the formula tn = t x

+ (n- \)d. For
this sequence, t x
is 3 and the common dif-

ference d is 2.

tn = 3 + (n- l)-2
- In + 1
b. The equation t„ = 2n + 1 describes a line,
but f„ is defined only when n is a positive
integer. The graph of the sequence consists
of discrete points on the line t = 2/7 + 1

Example 2 a. Find a formula for the wth term of the geometric sequence

3, 4.5, 6.75, ....

b. Sketch the graph of the sequence.

(n_1)
Solution a. Use the formula tn = tx • r . In this

case, t\ is 3 and the common ratio r is 1.5.

(«-l)
tn = 3 1.5

1.5"
- 3
1.5'

(1.5)" = 2 -(1.5)"
1.5

b. The equation tn = 2 • (1.5)" is an exponen-

tial equation whose graph has an intercept

on the f-axis at 2. The graph of the sequence

consists of discrete points on the exponen-
tial curve t = 2(1.5)".

The graphs of all geometric sequences are similar to the one in Example 2.

The points lie along an exponential curve whose base is the same as the common
ratio of the sequence.
The graph of a sequence or its mh term can give you informa-
the formula for
tion about all of the terms of the sequence. Example 3 shows how you can find any
term of a geometric sequence if you know two of its terms.

Example 3 In a geometric sequence, f3 = 12 and t

6 = 96. Find t u .

Solution Substituting in the formula t

n
= t
x
' r ' gives
2 5
h=h ' >'
= 12 and f6 = ^i * r = 96.

Thus, t x
= —=—
;• r
, which gives r = 8 and /• = 2.

Since t 3, t u =t A
t 10 = 3-(2) 10
= 3072.

Sequences and Series 475

 CLASS EXERCISES

For Exercises 1-6, state whether the given sequence is arithmetic, geometric,
or neither. If arithmetic, give the common difference, and if geometric, give
the common ratio.

1. 3. 8, 13, 18,.. 2. 4, 8. 16, 32. . . .

3. 2, 5, 10, 17.. . 4. 23, 17, 11, 5,..

5. 27. -18, 12.

- 6. 1. -3,5, -7,..

For Exercises 7-10, state the First four terms of the specified sequence. Then
tell whether the sequence is arithmetic, geometric, or neither.

7. t„ = 5n + 2 8. tn = ±±±
" // + 2
3
9. t,
3" 10. t„ = n

11. Discussion Does the sequence 2. 2, 2, satisfy the definition of:

a. an arithmetic sequence? b. a geometric sequence?

12. Visual Thinking How does the graph of the sequence with nth term
t
n
= 3/; — 5 differ from the graph of the line y = 3.v — 5?

WRITTEN EXERCISES

Find the First four terms of the given sequence and state whether the sequence
is arithmetic, geometric, or neither.

1. t„ = 2n + 3 2. tn = n3 + 1

3. t„ = 3 • 2" 4. t„ = 3 - In

5. t„ = n + 6. /, (-2)"

7. t„ = (-!)"•« 8. t
n
=16-2 2/7

9. tR = sin 10. t„ = COS 11TT

f-

Find a formula for t„ and sketch the Mil

" *»T _ HI. Baa -
-
wm
graph of each arithmetic or geometric
sequence. 1 2 3 4

5 (i 7 8 9 10 11
11. 1, 4, 7, 10,...
12. 8, 6, 4, 2, . . .
12 U 14 15 16 17 18

13. 8, 4, 2, 1,... Iff 20 21 22 23 24 25

14. 0.3, 0.9, 2.7. 8.1,... 26 27 28 29 30 31

15. 30, 26, 22, 18,...

The sequence of numbers in each col-
16. 24, -12. 6, -3, .. . umn of a calendar is arithmetic.

476 Chapter Thirteen

 .

State whether the given sequence is arithmetic, geometric, or neither. Find a

formula for tn the nth term of each sequence. ,

17. 17, 21, 25, 29, . . . 18. 15, 7, -1, -9, . . .

19. 8, 12, 18, 27, . . . 20. 100, -50, 25, -12.5, . . .

1 2 3 4 ...
21. 1, 4. 9, 16, 22.
. . .

r 3' 4' 5'

_ 3 4 5 ...
23. 11, 101, 1001, 10001 24. 7*
4' 9' 16'
2 3 4
- - a a a
25. 2a 2b, 3a b, 4a, 5a + b, . . . 26. 9' 18' 36' 72'
'

2/3 5/3 8/3 11/3

27. 2 , 2 , 2 , 2 , ... 28. V2, </l, </l, ^2, . .

Find the indicated term of each arithmetic sequence.

29. t
x
= 15. t2 = 21, 20 = ? t 30. fj = 76, r3 = 70, r 101 - ?
31. r
3
- 8, f5 = 14, r 50 = ? 32. r8 = 25, t 20 = 61, r, = ?

Find the indicated term of each geometric sequence.

33. f, = 2" 4 2 = 2" 3 f 12 = , t , ? 34. f, = 2, t2 = 2 3/2 , f 13 = ?

35. f2 = 64, r5 = -8. 9 = ? r 36. fj = 81, t4 = 24, r7 = ?
37. How many terms are in the arithmetic sequence 18, 24, ... , 336?
38. How many terms are in the arithmetic sequence 178, 170, . . . , 2?

[j 39. Find the number of multiples of 7 between 30 and 300.

40. Find the number of multiples of 6 between 28 and 280.
41. How many 3-digit numbers are divisible by 4 and 6?
42. How many 4-digit numbers are not divisible by 11?
43. True or false? If the sequence a, b, c is arithmetic, so is the sequence sin a,
sin b, sin c.
a C
44. True or false? If the sequence a, b, c is arithmetic, then the sequence 2 , 2 , 2
is geometric.

45. Explain why the logarithms of the terms of the geometric sequence a, ar, ar",
at 3 , . . . form an arithmetic sequence.
46. Writing One of the principal ideas in the work of the English economist
Thomas Malthus (1766-1834) is often expressed as "food supply rises arith-
metically, but population increases geometrically." Write a paragraph or two
restating this idea in terms of sequences. If Malthus 's idea is correct, what
kind of consequences can be expected? Explain your answer.
47. Find x if the sequence 2, 8, 3.v + 5 is (a) arithmetic and (b) geometric.
3
48. Find x if the sequence 4, x, —x is (a) arithmetic and (b) geometric.

Sequences and Series All

 .

49. Find x and y if the sequence y. 2x + y. 7y. 20, . . . is arithmetic.

50. Find x and y if the sequence 2y, 2.vy, 2, 4?". ... is geometric.

51. Physics If the half-life of an element is 2 days, what fractional amount of the
element remains after 2 days? 4 days? 10 days? d days?
52. Consumer Economics If P dollars is invested at 8% interest compounded
quarterly, what is the value of the investment after 1 quarter? 2 quarters?

q quarters? n years?
53. Find an arithmetic sequence none of whose terms is divisible by 2, 3, or 7.

54. Find an arithmetic sequence all of whose terms are multiples of 2 and 3 but not
multiples of 4 or 5.

55. Consider the two arithmetic sequences:

A: 3, 14, 25 B: 2. 9, 16, . . .

Write the first five terms of sequence A that are also terms of sequence B.

56. a. Suppose you 18 and t 4 = 8. Find two different

are given the terms t
2
—
possible common ratios. Then
two sequences having these terms. find the
b. Given two terms of a geometric sequence, under what conditions will there
be two different common ratios that could be used to find two sequences
that have the given terms?
57. Find all right triangles having sides with integral lengths that form an arithme-
tic sequence.

58. Prove that there are no right triangles having sides with integral lengths that
form a geometric sequence.

For Exercises 59-64, use the following definitions as needed:

The arithmetic mean of the numbers a and b is

—— -

The geometric mean of the positive numbers a and b is Vab.

59. Find the arithmetic mean and the geometric mean of:

a. 4 and 9 b. 5 and 10
60. Assuming that a and b are positive, find x if the sequence a, x, b is:

a. arithmetic b. geometric

61. CD is the altitude to the hypotenuse of

the right triangle ABC.
a. Prove that AADC ~ AACB and.

therefore, that 7- = — that : is, b is

b c
the geometric mean of x and c.

b. Likewise it can be shown that a is

the geometric mean of _]_ and ? .

c. Use the result of parts (a) and (b) to

2 2 2
prove that a + b = c .

478 Chapter Thirteen

 62. Refer to the figure for Exercise 61. Use similar triangles to prove that h is the
geometric mean of x and y.

H 63. Prove that the arithmetic

geometric mean.
mean of two positive numbers is never less than their

64. Consider a circle with radius 6. Find A, the area of the inscribed equilateral
triangle, and B. the area of the circumscribed equilateral triangle. Show that
the geometric mean of A and B is the area of the inscribed regular hexagon.

13-2 Recursive Definitions

Objective To use sequences defined recursively to solve problems.

Sometimes a sequence is defined by giving the value of t„ in terms of the preceding

term. tn -\. For example, consider the sequence defined by the following formulas:

tn =2f„_, + 1

The second formula above states that the nth term is one more than twice the
(n — l)st term. Knowing that the sequence begins with t x
— 3. we can determine
the first few terms of the sequence as follows:

t
n
= 2tn _ x
+ 1 > t2 = 2t x + \ = 2(3) +1=7
t3 =2f2 + 1 = + = 15
2(7) 1

t4 = 2r 3 + 1 = 2(15) + = 31 1

The formulas t\ = 3 and t„ = 2t„-\ + 1 give a recursive definition for the se-
quence 3. 7. 15. 31 A recursive definition consists of two parts:

1. An initial condition that tells where the sequence starts.

2. A recursion equation (or recursion formula) that tells how any term in the

sequence is related to the preceding term.

In Section 13-1. you were given explicit definitions for sequences. To see the
contrast between a recursive definition and an explicit definition, consider the
arithmetic sequence

23, 20, 17, 14, ... .

Recursive definition: t\ = 23 Initial condition.

tn — tn - X
— 3 Recursion equation says that each
term is 3 less than the preceding term.

Explicit definition: t
n
= 26 — 3/; /„ is given explicitly in terms of //.

Sometimes it is possible to find an explicit definition for a sequence when

you are given its recursive definition, as the activity on the next page shows.

Sequences and Series 479

 Activity

t\ = 3; t„ = f„_, +4
a. Gi\ e the first five terms of this sequence. What kind of sequence is it?

b. Find a formula for /,, in terms of n.

2. t
x
= 64: t„ = 1

n-\

a. Give the first five terms of this sequence. What kind of sequence is it?

b. Find a formula for t„ in terms of n.

g Recursive definitions are important tools

shows. They are easier to
what is directly observed. Since
set

many
in modeling, as the example below
up than explicit definitions because they
repeated calculations of the same
reflect

formula can be done quickly using a loop in a computer program, recursion

formulas are very useful when using computers to solve problems.

Example The population of a certain country grows as a result of two conditions:

1. The annual population growth is 1% of those already in the country.

This growth rate equals the birth rate minus the death rate.

2. 20.000 people immigrate into the country each year.

If the population now is 5,000,000 people, what will the population be in

20 years?

Solution If you ignore the immigration for a moment and consider only the 1%
growth rate, then P„ = 1.01P,,-!. Since immigration adds another 20,000
persons each year, the recursion equation is

1.01/>„_ + 20.000.

In many practical situations, it is convenient to use a zero subscript to

indicate an initial observation. If we let P = 5.000.000 represent the pop-
ulation now. then P x
is the population at the end of one year, and P„ is the
population at the end of n years. To calculate the population for the first

few years, you could use a calculator and the recursion equation.

P = 5,000,000
P = ]
1.01 • 5,000,000 + 20,000 = 5,070,000
P2 = 1.01 • 5,070,000 + 20,000 = 5,140,700
P3 = 1.01 • 5.140,700 + 20,000 = 5,212,107

However, to calculate the population for each of the next 20 years, it is

computer program or use a spreadsheet program.

easier to write a simple
The BASIC program given on the next page uses the recursion equation
above to find the population.

480 Chapter Thirteen

 ID LET P = SDDDDDD < This is the initial population P .

ED FOR N = 1 TO ED
3D LET P = l.Dl*P + EDDDD < P„ = 1.01 • Pn -i + 20,000
14DNEXT N
SD PRINT 'THE POPULATION IN '
"iN-li" YEARS IS ''MNT(P)
bQ END

The program produces this output:

THE POPULATION IN ED YEARS IS b5m330

CLASS EXERCISES

Give the first four terms of each sequence.

1. t {
= 5; tH = f„_, +3 2. tx = 10; tn = fn _, + n
3. f, = 3; t„ = 2r„_i 4. r, = 4; f„ = 2f„_, - 1

5. Sometimes a recursive definition will tell how the (n + l)st term is related to
the /7th term. Give the first four terms of the sequence defined recursively by
t\ — 5 and tn+l - 2t„ + n.

6. Sometimes a recursive definition gives t„ in terms of more than one preceding

term. Give the first four terms of the sequence defined recursively by t x
—A,
t
2 = 16, and t„ = f„_j + t„- 2 .

WRITTEN EXERCISES

Find the third, fourth, and fifth terms of each sequence.

1. f, =6; t„ = t ll . {
+4 2. f, =9; r„ = ^„_,
3. r, = 1; t
n = 3r„_! - 1 4. / = 4;r„ = (fn _1 f- 10

5. r, = 1; tn = tn -i +2/7-1 6. r

7. f, =2; r2 = 4; fn = f„_, + r
H_2 8. r = 2; r2 = 4; r
;i
= r
;i
_, • r
/;
_2
9. fj =5; /
2 = 8; fn = (/„_, - r„_ 2 ) 10. r — 7; t2 — 3; t„ — r„_ 2r„_
n-2
11. Find an explicit definition for the sequence in Exercise 1

12. Find an explicit definition for the sequence in Exercise 2

Give a recursive definition for each sequence.

13. 9, 13, 17, 21,.. . 14. 81, 27, 9, 3,...

15. 1, 3, 7, 15, 31, 63 16. 1, 3, 7, 13, 21, 31,. .

17. 1, 3, 6, 10, 15, 21,... 18. 1, 2. 6, 24, 120, 720,

Sequences and Series 481

 i

19. a. Give the first eight terms of the sequence defined recursively by t\ = 3,

t
2
= 5. and tn = t
n _ x
— tn -2 .

b. Observing the pattern you get in part (a), tell what the 1000th term of the
sequence will be.
20. a. Give the first eight terms of the sequence defined recursively by t
{
=4,
and t„ = f—
« n—

b. Observing the pattern you get in part (a), tell what the 1000th term of the
sequence will be.
21. Geography Refer to the example on page 480. Suppose the population of
those living in the country grows 2% per year, and that an additional 50,000
people immigrate into the country every year.
a. Give a recursion equation for P„. the population in n years.
b. If the population now is 8.500.000. what will the population be in 5 years?

22. Chemistry Each day 89c of a quantity of radioactive iodine will decay.
a. Express this fact with a recursion equation.
b. Choosing a value for the initial amount of iodine, find the approximate
half-life of the iodine.

23. Visual Thinking Let S„ represent the number of dots in an n by n square

array. Pretend you have forgotten that S n = n Give a recursion equation that .

tells how 5„ +1 is related to S„ by reasoning how many extra dots are needed to

form the (// + l)st square array from the previous //th square array. Illustrate
your answer with a diagram of dots.
24. Suppose that everyone in a room shakes hands with everyone else exactly
once. Let Hn represent the number of handshakes if there are/? people (n > 2)
in the room. Give a recursion equation that tells how Hn is related to Hn -\.
{Hint: Suppose you H
know n — j. If another person enters the room, how many
additional handshakes will there be?)

25. Visual Thinking Let d„ represent the number of diagonals that can be drawn
in an //-sided polygon. The diagram at the left below shows that a hexagon has
9 diagonals, and so d 6 = 9.

a. Imagine pushing out one side of the hexagon so that a polygon of 7 sides is

formed. (See the diagram at the right above.) How many additional diago-
nals can be drawn?
b. Imagine pushing out one side of a polygon with // — 1 sides so that an
//-sided polygon is formed. Tell how many additional diagonals can be
drawn. Then write a recursion equation for dn .

482 Chapter Thirteen

26. Visual Thinking Let rn represent the number of regions formed when n lines
are drawn in a plane such that no two lines are parallel and no three are
concurrent. The diagrams below illustrate r r 2 and r 3 Give the recursion
x
, , .

equation for /-„.

r J
=2 regions r2 =4 regions /
3
= 7 regions

27. Visual Thinking Let /?„ represent the number of intersection points created
when n lines are drawn in a plane such that no two lines are parallel and no
three lines are concurrent. The diagrams in Exercise 26 show that p = 0, x

p 2 — 1, and /? 3 = 3.
a. Find p 4 and p 5 .

b. Give a recursion equation for pn .

28. Writing For what kind of problem would you be more likely to use a recur-

sive definition of a sequence than an explicit definition? Why? For what type
of problem would an explicit definition be more useful?

^yl For the parts of the following exercises designated in green, you will need
to use a computer or a programmable calculator.

29. Finance On the birth of their daughter, Mr. and Mrs. Swift began saving for
her college education by investing $5000 in an annuity account paying 10%
interest per year. Each year on their daughter's birthday they invested $2000
more in the account.

a. Let A„ represent the amount in the account on their daughter's /?th birthday.
Give a recursive definition for A n .

b. Find the amount that will be in the account on her 18th birthday.

30. Finance When Mr. Tallchief reaches

his retirement age of 65, he expects to
have a retirement account worth about
$400,000. One month after he retires,

and every month thereafter, he intends

to withdraw $4000 from the account.
The balance will be invested at 9% an-
nual interest compounded monthly.
a. Let A„ represent the amount in the
account n months after Mr.
Tallchief s retirement. Give a recur-
sive definition for A„.
b. When will there be no money left in the account?

Sequences and Series 483

 .

31. The Tower of Hanoi puzzle

consists of a block of wood
with three posts. A, B, and C.
On post A there are eight
disks of diminishing size
from bottom to top. The task
is to transfer all eight disks
from post A to one of the
other two posts given that:

1. only one disk can be

moved at a time:
2. no disk can be placed on
top of a smaller disk.

a. Let Mn represent the minimum number of moves needed to move n disks

from post A to one of the other posts. Obviously, M\ = 1 . What are M2 and
M 3? (To try this puzzle, you could use coins of different sizes and three
locations A, B, and G on a sheet of paper to represent the three posts.)
b. Suppose you know how to move (n — 1) disks or coins from post A to
another post, and that to do so requires M„_i moves. Explain why
Mn = 2M„_, + 1.

c. Use your answer to part (b) to check your values for M 2 and M 3. Then find
A/4 , M M6 M
5, ,
7, and M 8 .

d. Find an explicit definition for Mn .

e. According to legend, in the great Temple of Benares, there is an altar with

three diamond needles. At the beginning of time, 64 gold rings of decreas-
ing radius from bottom to top were placed on one of these "needles. Day and
night, priests sit before the altar transferring one gold ring per second in
accordance with the two rules given above. The legend also says that when
all64 rings have been transferred to one of the other diamond needles, the
world will come to an end. How long will it take the priests to transfer all
the rings? (You can use your answer to part (d), or you can find a computer
solution using the recursion equation given in part (b).)

32. Here famous problem posed in the thirteenth century by Leonardo de Pisa,
is a
better known as Fibonacci: Suppose we have one pair of newborn rabbits of
both genders. We assume that the following conditions are true.

1. It takes a newborn rabbit one month to become an adult.

2. A pair of adult rabbits of both genders will produce one pair of new-
born rabbits of both genders each month, beginning one month after
becoming adults.
3. The rabbits do not die.

How many rabbits will there be one year later? (To help with your solution to
this problem, write a recursion equation in which /„, the number of pairs of
rabbits n months from now, is expressed in terms of r„_ \ and rn - 2 . Remember
that r = 1

484 Chapter Thirteen

 33. Physics Newton's Law
of Cooling states that over equal time periods (for
example, one minute) the change in an object's temperature is proportional to
the difference between the temperature of the object at the beginning of the
time period and the room temperature. Thus, if t is the temperature of the
object when it is placed in a room whose temperature is /?, then t m the object's
temperature n minutes later, is given by the recursion equation

tn ~t„-i =k(tn - 1 -R),

where k is a proportionality constant.
a. Suppose that a cup of coffee at 98° C is placed in a room whose tempera-
ture is 18° C, and that 1 min later the coffee has cooled to 94° C. Find the
proportionality constant k, and explain the significance of its negative sign.
b. Show that the recursion equation given above can be rewritten as
t„ = 0.95r„_] +0.9.
c. Use a computer to print out a table of values for /„. When will the cup of
coffee cool to less than 75° C?
34. Medicine Suppose that in a closed community with population P, a flu epi-
demic begins and that the number of people newly exposed to the flu on a
given day is proportional to the number not yet exposed on the previous day.
a. If /„ represents the number of people exposed to the flu n days after it

begins, explain how the description above leads to the recursion equation

/„ — fn -\ — k{P — fn -i), where k is a proportionality constant.

b. Suppose in a college community of 2500 students, the flu begins with 100
students exposed to the flu; that is,/ = 100. On the next day,/\ = 220.
Find the value of k, and then show that/,, = 0.95/„_! + 125.
c. Use a computer to print out a table of values for/,,. About how long does it

take before the flu spreads through the whole student body?

Q 35. a. Ecology Suppose that a certain population grows from one generation to
the next in such a way that the population increase is proportional to the
population of the previous generation. Explain how this growth can be
modeled by the equation Pn - P„-\ = kP„- In your explanation, tell X
.

what P„ and Pn -\ represent.

b. Suppose that a population has a carrying capacity C. This is the maximum
size beyond which the population cannot grow. As the population gets
closer to C, overcrowding tends to reduce population increases. Sketch a
graph that models this type of population growth with overcrowding.
c. The population growth described in part (b) can be modeled by modifying
the equation in part (a) so that the new equation is

Pn -#>„_, =^,_ (C-P„_ 1 1 ).

Write a sentence or two explaining what this recursion equation says.

d. For a population of fish in an aquarium, let 500 and k C= — 0.002.
Choose a value for the initial population, and use a computer to print the
values given by the recursion equation in part (c). How well do the com-
puter results agree with your graph in part (b)?

Sequences and Series 485

 13-3 Arithmetic and Geometric Series and
Their Sums

Objective To find the sum of the first n terms of arithmetic or

geometric series.

The words sequence and series are often used interchangeably in everyday conver-
sation. For example, a person may refer to a sequence of events or to a series of
events. In mathematics, however, a distinction is made between a sequence of

numbers and a series of numbers. This distinction between a sequence and a series
can best be made by considering some examples.

Finite sequence: 2, 6, 10, 14

Finite series: 2 + 6+10+14

T <-
Infinite sequence: —1111
— — —
, , , ,
•

2 4 8 16

Infinite series: T2 + -r + 7r + TT +
4 8 16

As you can see from the examples above, a series is an indicated sum of the
terms of a sequence. The sum of the finite series 2 + 6 + 10 + 14 is 32. In Section

13-5 we will see that the sum of the infinite series —+—+ tt + t^ + " '
is 1.
2 4 8 16
However, not all infinite series have sums. In this section we will consider only
finite series and their sums.

If S„ represents the sum of n terms of a series, then Sn can be expressed either

explicitly or recursively as follows:

Explicit definition of S„: Sn = t\ + t

2 + /_-? + • • •
+ t
n

Recursive definition of S„: 5 =

Sn = Sn -i + t„ for n > 1

M I The recursive definition can be used as the basis for a computer program that
will find the sum of a series. Example 1 illustrates how this is done.

Example 1 Find the sum of the cubes of the first twenty positive integers.

3 3 3 3
Solution We want 5 20 = l + 2 + 3 + • •
+ 20 .

First find a formula for tn in terms of n. The mh term is t n — n ,

and is represented by T in line 30 of the computer program given

on the next page. The recursive definition of the sum Sn which , is

represented by S, occurs in lines 10 and 40.

486 Chapter Thirteen

 10 LET S = < Initial condition S =
2D FOR N = 1 TO ED
A
3D LET T = N 3
40 LET S = S + T< Recursion equation S n = Sn -\ + tn
SD NEXT N
bO PRINT ''THE SUM OF THE FIRST "iN-li" TERMS IS ' ' =, S
70 END

When the program is run, the computer gives this output:

~
O THE SUI1 OF THE FIRST 50 TERNS IS 44100 ©U
You can use the method of Example 1 to find the sum of any finite series.

However, if the series is arithmetic or geometric, you can also use the formulas
given in the following two theorems.

Sum of a Finite Arithmetic Series

The sum of the first n terms of an arithmetic series is

*n
2

Proof: Write the series for S twice, the second time with the order of the
fl

terms reversed. Then add the two equations, term by term:

Sn = t y
+ (f, + d) + fa + 2d) + • •
+ (r„ - d) + tn

Sn = t„ + (t„ - d) + (f„ - 2d) + • • •

+ (f ! + d) + t
x

2S„ - fa + tn ) + (t! + t n) + (r, + tn ) + • • •

+ fa + tn ) + (t l + tn )

Since there are n of these (t x + tn ) terms,

2Sn = n(ti + tn )

n(t x + tn )

Example 2 Find the sum of the first 25 terms of the arithmetic series

11 + 14 + 17 + 20 + •••.

Solution Step 1: First find the 25th term.

'25 =h + (n - \)d = 11 + (25 - 1)3 = 83

Step 2: 5 25 ^^4^^ 25(1

\
+83)
= 1175

Sequences and Series 487

 , \

Sum of a Finite Geometric Series

The sum of the first n terms of a geometric series is

= f,(1 -r")
S„
— r

where r is the common ratio and r # 1

Proof: Multiply the series for S„ by the common ratio and then subtract
the resulting series from the original one, as shown below.
n—2 n—
S„ = t\ + tyV /,/•" + *
+ +
i

t,r + , .

t\r
w-1
rS, t,r + til- + + r,r + rw + /,/'

S '"5, r, + + + •• + + t\r

S„(l -r) = tiQ -r")

ri(1
~ "

S = / )

r * 1

This formula is not defined for r — 1. If r does equal 1, however,

the geometric series is simply a series of repeated numbers, such
as a + a + + a, whose sum is na.

Example 3 Find the sum of the first 10 terms of the geometric series

2-6 + 18-54 + •••. i

10
_ r,(l-/ )
Solution 10
1 - r

2(1 -(-3) 10 ) __ 2(1 -3 10 )

-29,524
1 - (-3)

CLASS EXERCISES

1. Give an example of an arithmetic sequence and of an arithmetic series.

2. Which theorem can be used to find the sum of the finite series
1+2 + 4 + 8+ 16+ •••+ 1024? What is this sum?
3. a. Is the series 3 + 5 + 9+ 17 + --- + 32.769 arithmetic, geometric, or
neither?
b. Give a recursive definition of the series.
c. How would you modify the computer program in Example 1 to find the
sum of this series?

488 Chapter Thirteen

 WRITTEN EXERCISES

For each of the arithmetic series in Exercises 1-8, find the specified sum.

Q 1. S 10 h =: 3, t
w= 39 2. S 200 : r, = 18, t 200 = 472
3. S 50 5 : + 10+ 15 + •••
4. S 25 : 17 + + 33
25 + •••
5. S 12 :
hi
= 5 + 3« 6. S 40 : t x
=5, t3 = 11
7. 1 + 2 + 3 +•••+ 1000 8. 3 + 7 + +
1 1 + • • •
99

9. Find S 8 for the geometric series with t x

= 8 and (a) r = — , (b) r = —
y
10. Show that the sum of the first 10 terms of the geometric series

is twice the sum of the first 10 terms of the series

1 1+1 1
+
3 9 27

11. Show that 1 + 2 +4 + + 2"~ = 2" - • •

•
1

12. Show that 9 + 90 + 900 + + 9- 10 n— = 10"-

n/; •
1
1.

63V2
13. Show that \fl + 2 + 2\fi + • •
+ 64
V2- 1

14. For the series 1 + \^3 + \^ + ^3* + • •, show that

= 26 -42
+3 + 5 + 7=4
5,5 1/5
1

• • • •
n(n + 1)
15. Show that the sum of the first n positive integers is • • • •

• • • •
16. Let S„ be the sum of the first n positive odd integers.
a. Evaluate S , S2 , S3, and S4 .
• • • •
x

b. Suggest and prove a formula for S„. Ex. 16

Find a formula for tn , the wth term of the series. Then give a recursive defini-
tion for 5„, the sum of n terms of the series.

17. 2 + 4 + 6 + 8+--- 18. 1 + 3 + 9 + 27 + • •

•

19. 8 + 12 + 18 + 27 + 20. 50 + 47 + 44 + 41 +

TKJl For Exercises 21-26, you will need to use a computer.

Write a computer program, like the one given in Example 1, that will print the
sum of each series. Then run the program to find the sum.
4
21. 1 + 2 4 + 3 4 + ---+ 10 4 22. I + 2 2 + 3 3 +•••+ 15 15
1

23.
2 2
10 + 20 + 30 +
2
+ 1000 2 • •
24. VE + VlO + VT5 + + V200 •

Sequences and Series 489

 [3 25. a. Refer to the computer program given in Example 1. Delete line 60 and
insert the following line.

>4S PRINT NnS

Explain what happens to the computer's output.

b. Change lines 20 and 30 in the program so that the computer will print the

sums Si, S 2 S3 525 f° r me series

,

2" 1

+ 2~ 2 + 2~ 3 + •••
+ 2~ 25 .

c. Find S 25 using the formula given on page 488. (Your answer should be
approximately the same as the last number that the computer prints.)

d. Guess the sum of the infinite series

2" -3
' + 2~ 2 + 2 + • • •
+ 2~" +••-.

26. a. Modify the computer program in Exercise 25 and run it for the series

1 + JL2 + J. + ... + J-
25- 3
3 3 3 3

b. Find S25 using the formula given on page 488.

c. Guess the sum of the infinite series

1 + .L+ 1
33
+... + J. + ..,
3 32 y,

27. The sum of the first // terms of a series is Sn — n + An. Find t\, t 2 , and t3 .

28. The sum of the first /; terms of a series is Sn = In .

a. Find t\, t
2 . and f
3 .

b. Find S„ — Sn -\.
29. Find the sum of all multiples of 3 between 1 and 1000.
30. Find the sum of all positive 3-digit numbers divisible by 6.

31. Find the sum of all positive 3-digit numbers whose last digit is 3.

32. Find the sum of all positive odd numbers less than 400 that are divisible by 5.

33. Find the sum of the series 1—3 + 5 — 7 + 9— 11 H h 1001.

34. Find the sum of the series 1+2 + 4 + 5 + 7 + 8+ 10 + 1 1 + • • •

+ 299,
which is the sum of the integers except for multiples of 3.

35. The originator of a chain letter writes 5 letters instructing each recipient to
write 5 similar letters to additional people. Then these people each send 5
similar letters to other people. Determine the number of people who should
receive letters if the chain continues unbroken for 12 steps. Explain why the
process always fails. (There are laws forbidding chain letters that request

money.)
36. Value Appliance Store has radios that can be purchased on a daily installment
plan. For a particular type of radio, you pay only 1 cent the first day, 2 cents
the next day, 4 cents the next day, 8 cents the next day, and so on, for 14 days.
How much does the radio cost?

490 Chapter Thirteen

37. a. If you go back through ten generations, how many ancestors do you have?
Count your parents as the first generation back, your four grandparents as
the second generation, and so on. (Assume there are no duplications.)
b. How many generations back must you go in order to have more than one
million ancestors?

38. a. Finance Suppose a doctor earns $40,000 during the first year of practice.
Suppose also that each succeeding year the salary increases 10%. What is
the total of the doctor's salaries over the first 10 years?
b. How many years must the doctor work if the salary total is to exceed a
million dollars?

n(n +
39. The number T„ I +2 + 3 +-• + n = 1)
is
= 10

sometimes called a triangular number because it is pos-

sible to represent the number by a triangular array of
dots, as shown.
a. Evaluate T T2 T3 T4 and T5
x
, , , , .

b. Add any two consecutive triangular numbers. Then

make a conjecture and prove it.
40. Consider the series 1 +2' + 3" +••• + n . Evaluate S\, S2 , S3 , and S4 .

Then suggest a formula for S„. (Hint: See Exercise 39.)

41. The sum of every row, column, and diagonal of a magic square equals the
magic number M. For the 3-by-3 and 4-by-4 magic squares shown below, M is

15 and 34, respectively. Find the value of M for an n-by-n magic square that
contains consecutive integers starting with 1 . (Hint: How many numbers are in

an n-by-n square? What is their sum? Deduce the row sum M.) Check that

your formula gives M(3) = 15 and M(4) = 34.

M= 15 M= = 34

8 1 6 1 15 14 4

3 5 7 12 6 7 9

4 9 2 8 10 11 5

13 3 2 16

42. Finance If you invest $1000 per year every year for 10 years and if your

money compounded annually at 12%, how much money will you have at the
is

time you make your tenth investment?

43. Finance If you invest P dollars at an interest rate /• compounded annually,
n
after n years you will have P(\ + r) Suppose you invest P dollars
dollars.

every year for n years. Show that at the time you make your /?th investment,
n -
P[(\ +r)
you will have dollars.

Sequences and Series 491

 44. Finance The purpose of this exercise is to develop a formula for the monthly
payment P that is required to repay a loan of A dollars in n monthly install-

ments, with the interest on the unpaid balance equal to a monthly rate r. Let
A k = amount still owed after paying k installments. Then interest for the

(k + l)st month is rA k and the principal , paid off in the (k + l)st payment is

P - rA k Thus,.

A k+i = A k ~{P- ,A k ) = (1 + r)A k - P.

a. Use the equation above and the fact that AQ — A to find A u A 2 and A 3 , in

terms of A, r, and P.
b. Generalizing from A t , A 2 and , A$, you get

A„ = ( 1 + r)"A - [( 1 + r)""
1

+ ( 1 + r)"~
2
+ •• + ( 1 + r) + 1 ]P.

Since the bracketed quantity above is a geometric series, show that the

equation can be rewritten

(1+r) "" 1
A B = (l+r)^-[
r

c. Since A„ = (why?), use the last equation in part (b) to show that

Ar{\ + ,-)"

(1 +;•)"- r
V% 45. Finance A direct reduction loan is often used for buying cars and for mort-
gages on homes. You pay interest only on that portion of the loan that you
have not repaid (that is, the unpaid balance). Therefore the monthly payment
you make has two parts: /, which is the interest on the unpaid balance, and /?,

which is used to reduce the balance.

Suppose you take out a car loan for $10,000 at 12% annual interest,
calculated monthly, to be repaid in 5 years. Your monthly payment will be P.
a. Use the formula in Exercise 44(c) to show that P = $222.44. Remember
= monthly = 12% =
that r rate 0.01.
12
b. Study the table below. Then calculate the entries in the last row of the table.

n = payment A„ = amount outstanding /„ = interest R„ — loan reduction

number after payment on A„ = $222.44 - /„

10,000.00 100.00 122.44

1 9,877.56 98.78 123.66

2 ? ? ?

c. Express A n in terms of A,,-,.

d. Show that the values of R n form a geometric sequence with ratio 1.01,
which is (1 + monthly rate).

e. (Optional) Write a computer program that will print out the table above
for all 60 payments. Label the column headings N, A, I, and R.

492 Chapter Thirteen

 Infinite Sequences and Series

13-4 Limits of Infinite Sequences

Objective To find or estimate the limit of an infinite sequence or to

determine that the limit does not exist.

A sequence that does not have a last term is called infinite. Consider the infinite
geometric sequence

1111
2' 4' 8' 16'" " '
(\_
\2-

The terms of this sequence are surely getting smaller, but how small do they get?
With a calculator or logarithms, we can calculate that f 10 = (t) 10 ~ 0.001 and
1 oo = (t)
° ° ~ 0.00000000000000000000000000000 1
'
When ' you substitute .
1

larger and larger values of n. t„ = (\)" becomes a smaller and smaller positive
number. It never becomes zero, but we can make tn come as close to zero as we
like just by finding a large enough value for n.

The graph at the right illustrates this idea. No matter J

K
how small a positive number k we choose, we can always - •
1

2
make t„ be within k units of zero just by going far enough
3
to the right on the graph.
8
The preceding discussion can be summarized by the
1
following equation: 4~ •

lim 1
•
8
•
This k
is read "the limit of (\) as n goes to infinity is
i i i i
T • • h
O
i i i

zero." 4 n
l 2 3 5

As another illustration consider this sequence:

1+ l,-l 1+ l 1 +
(-1)"
,-l.
The graph of this sequence, shown below, illustrates that its limit is 1 . You can
make the terms of the sequence as close to 1 as you like just by going far enough to
the right on the graph.

lim 1 +
( \r\
= l
m
n I 2--

In everyday usage, the word "limit"

sometimes suggests a barrier, but in mathe- 1--
matical usage it is better to think of a limit
as a target. Thus, the limit 1 is a target
approached more and more closely by
O
(-1)"
1 + as n gets larger and larger.

Sequences and Series 493

 You can often estimate the limit of an infinite sequence by substituting large
values of n in the formula for the /?th term, as illustrated in the next example.

Example 1 Find: a. lim sin b. lim (0.99)"

Solution
a. When // = 100. sin (
— =
) sin (ttjtt) —0.01. As n gets larger, —

approaches and so does sin — . Thus, it seems that lim sin (

— ) =0.
/7 n-*o° \/;/

b. We can evaluate (0.99)" for large n with logarithms or a calculator. For

example.
10000 44
(0.99)
10 °
4.3 x 10 5
and (0.99) « 2.2 x 10 .

Thus, it seems that lim (0.99)" = 0.

We have just estimated that lim (0.99)" is 0, and earlier we saw that

lim ( —2/ ) =0. These two examples are special cases of the following theorem,
n—*« \

Theorem

In Example 1we estimated limits by considering t„ for large values of n. In

the next example, we show how to change the form of t„ to assist in finding a
limit.

/r+ 5/T + Vn
Example 2
1
Find: a. lim b. lim
h-xjc 2/1 3/7 »-»" 3« + 7

Solution In both parts (a) and (b), we divide numerator and denominator by the
highest power of n that occurs in the denominator.
a. Dividing numerator and denominator by n , we have:

ir + 1 n
2/i — 3«

Notice that when n is very large, — and — are very near 0. Therefore,

,r+ 1

z
2/7 - 3/7 L

when n is very large. For this reason, we say:

2
n + 1 1
lim
3/7

494 Chapter Thirteen

 I

b. Dividing numerator and denominator by n , we have:

2
A+ '

5/; + Vn n n
5/2

3 ii + 7 3 +

Notice that when n is very large, — , —rpr,

5'2
and —7 are very near 0.
" n n
2
5n + Vn +
=
Therefore,
3 „3 + 7 3 +
when n is very large. For this reason, we say:

5n
2
+ Vn
lim
"^ x 3/?"
1
+ 7

Situations in Which a Sequence Has No Limit

Not all sequences have limits. If the terms of a sequence do not "home in" on a
single value, we say that the sequence has no limit or that the limit of the sequence
does not exist. For example, the following sequence has no limit:

1-11-1
2 ' 3 '
4 ' 5 '
(-0"
n +
+1

1
'n
""
i'«
The diagram shows the graph of this se-

quence. Notice that the odd-numbered terms

form a sequence with limit 1. Similarly, the
even-numbered terms form a sequence with
~~ H h — 1 1 1
h
12
1

q 3 4 5 6 7 8 9
limit — 1. Nevertheless, there is no single lim-
iting number for all the terms of the sequence.
Thus, we say that this sequence has no limit.

Infinite Limits
Sometimes the terms of a sequence increase or decrease without bound. Here are
two examples:

(a) 3,7, 11, 15, ...An- I,... (b) -10, -100, -1000,. . . ,
-10",. . .

The terms in sequence (a) continue to increase as n becomes larger, but they do not
approach a fixed number as a target. No matter what fixed number is selected, the
terms of the sequence will eventually exceed it. For this reason, we say that the
limit of An — 1 as n increases without bound is infinity. We write this as

lim (4/? - 1) = 00.

n—»oo
Similarly, the terms in sequence (b) decrease without bound; that is, they are
farther and farther to the left of zero on the number line. We say that the limit of
— 10" as n increases without bound is negative infinity and write

lim (-10") = -00.

n—*° c

Sequences and Series 495

 ) —

It is important to realize that infinity is neither a number nor a place. When

we say that "n goes to infinity."' we mean that n increases without bound. When
we say that "the limit is infinity." we mean that the terms are increasing without
bound, not that they are approaching some number.

7/f
Example 3 Show that lim

Solution Dividing numerator and denominator by n~. the highest power of n in the

denominator, we get:

3
In- In
An - 5 4-A

When n is very large. -= is very near 0. Therefore.

7/r —— when n
Air - 5 A

is very large. Thus, since —In

increases without bound as n does,

In In
lim
n-»«
—4r = x and lim
«^=c - _
4/? 5

CLASS EXERCISES

Find the following limits.

1. lim - 2. lim =
„—>x n + 1

2
In + 8i/- 3«
3
3. hm
,. 1
4.
,.
lim
fl-MC 3/7 + 1 "-^ x
5/r + 7

5. lim cost — 6. lim log [cos

n-*oo V 77 / (J)
7. lim (0.999)" 8. lim (1.001)"
n—«
4 2
„ n + 9.999.999
9. lim n
u. um
H—»*: 2/7
Z/7 +
11. Discussion Do you think the sequence 1, 0, 1, 0. 1, 0, ... has a limit?
Why?

WRITTEN EXERCISES

In Exercises 1-12, find the given limit.

n + 5 77" + (-1)"
1. lim 2. lim
I

3. lim 1 +
rt-wx /? B-wo „ n—*cc

4. lim
4»-3 5. lim
3/?"" + 5/7
6. lim
2/7

n-»" 2/7 + 1 "^* 877- "-* " 677 + 7

496 Chapter Thirteen

 1

//
7. lim tan 8. lim sec 1 9. lim
n -»oo \nn „^.DC « + 1

2/3
-
10. lim
n-w
5«
6/2 —
8/1

1
11. lim log
n -f
12. lim log V 10
In Exercises 13-18, find the limit of the specified sequence or state that the
limit does not exist.

1 1 1 1
13. 14. 1, -4,9, -16,25, -36,. .

3' 9' 27' 81' 243'

3 4 5 6 7 8 2 3 4 5
15. 16
2' 3'4' 5' 6' 7' 10' io
2 '
10
3 '
10
4 '
10
5 '

nir \
17. L cos 18. tn = sin (nir)
2 )

In Exercises 19-30, evaluate the given limit or state that the limit does not
exist. If the sequence approaches » or — », so state.

19.
,
lim
»^ x
—
5//=
5/2
-7//
20. lim
1/2
5//

n + 10/? /? — -,
3

sin //
21. lim log (-) 22. lim

.. cos (hit)
23. lim i
24. lim e""
n—>ai

25. lim tan l-T" + nir 26. lim 2"

„-»oo \ 4 >

27. lim lo g sec 28. lim [(-1)" - 1]

*)]
V/7TT S/Sn 2 - 5n + 1
29. lim 30. lim
n-w» V/J — 1
^/j 2 + 7/7-3
n-\
31. a. If the mh term of a geometric series is tn — , show that

S '
=±2 lf
b. Find lim S„.

n-
32. Consider the formula for the sum of the series t + x
t x
r + ••• + t
{
r
'' '
and
then suggest a formula for the sum of the following infinite series:

t\ + t x
r + t x
r~ + '", where |r| < 1

33. Recall that e = lim

„-»oc \
( 1 +—
n)
) . Find lim
«-*«
( 1
V
+ —n )
I
by noting that

n/2
+ 1)" 1+ " +
i
nl t 1

Sequences and Series 497

 )

34. a. Show that lim ( 1 + — '

= e
3
. b. Evaluate lim ( 1 + -M '.

{Hint: See Exercise 33.)

35. The area A under the curve v =x between x = and x — 1 can be approxi-
mated by adding the areas of n rectangles as shown.

A = - •-+ - + +

3 3 3 3
-j(l + 2 + 3 + + n )
n

a. According to the result of Exercise 40 in Section 13-3,

3 n(n +
+ 3
+ 3
+ + n3 =
1

l 2 3

= n + In + 1
Use this formula to show that A„
Arr
b. As n becomes very large, what value is An approaching?
c. Find A by evaluating lim An .

W 36. Use the procedure in Exercise 35 to find the area under the curve y =
between and 1. You will need to know that

2 2 2 2 n(n + l)(2/i + 1)
l + 2 + 3 + + n =

1 1 1 1 CALCULATOR EXERCISES

1. a. Guess the value of lim ( V/i + 1 — Vn ) by evaluating Vn + 1 — vn for

several large values of n.

+ - V" + 1 + V
b. Multiply V/7 1 V7i by % .

Vn + 1 + V/i
c. Determine what happens to the expression in part (b) when n becomes very
large. Does your answer agree with your answer to part (a)?

498 Chapter Thirteen

 2. A sequence is defined recursively by the equations

^ +
t = 1, and t„ = J1L •
x

a. Find decimal approximations for the first five terms of the sequence.
b. Suggest a limit for this sequence.

3. a. Evaluate V 1 + Vl + Vl+ Vl + •
by considering its value to be
the limit of the following sequence as n approaches °°:

VTTVT, Vl + VT+ VI, Vl + Vl + VT~+ VT, and so on.

b. Compare your answer with the golden ratio, R = .

////COMPUTER EXERCISE

A student leaves home to go to the movies. Halfway there, the student remembers
some uncompleted homework and heads back home. Halfway back home the stu-
dent has a change of mind and heads back to the movies. You guessed it! Halfway
back to the movies, the student, overcome by an attack of conscience, heads back
to complete the homework. Suppose the student continues in this fashion. Write a
computer program to calculate how
far from home the student is after each of the

first twenty changes of mind. you think of the student as moving along a
(If

number line with home at and the movies at 1, does the student appear to be
approaching a limiting point on the number line?)

Manuel Sandoval Vallarta (1899-1977)

Manuel Sandoval Vallarta, a mathematician and

physicist, was a professor at the Massachusetts
Institute of Technology. He studied relativity
with Albert Einstein and electromagnetic theory
with Max Planck. In 1946 he returned to his na-
tive Mexico, where he founded the Instituto
Nacional de Energia Nuclear.
A pioneer in the field of cosmic-ray phys-
ics, and an advocate for the peaceful uses of
atomic energy, Sandoval Vallarta was president
of the Latin American Council on Cosmic Rays,
and represented Mexico in conferences of the
United Nations Atomic Energy Agency.

Sequences and Series 499

 \ .

13-5 Sums of Infinite Series

Objective To find the sum of an infinite geometric series.

The sum of an infinite series is very closely connected to the limit of an infinite
sequence. To see this, consider the infinite geometric series

Associated with this series is the following sequence of partial sums:

S,=4-

s, = 4- + 4-

c - I
+ I
+ I
-1

5 = + + + + "- +
" i i i Ti v2

using the formula S„ = '

—

-'.i

Since the sequence of partial sums \, j,3 7

. 1 — (4-)" has limit 1, we say that
the infinite series has limit 1 or has the sum 1.

In general, for any infinite series f] + t

2 +" + '
t„ +
Sn =h+ t2 + '-'
+ t„

is called the wth partial sum. If the sequence of partial sums S\, S2 , . . . , Sn , . .

has a finite limit S, then the infinite series is said to converge to the sum S. If the

sequence of partial sums approaches infinity or has no finite limit, the infinite

series is said to diverge.

Since we already have a formula for the /?th partial sum of a geometric series,

we can prove a theorem that tells when such series converge.

Sum of an Infinite Geometric Series

If \r\ < 1, the infinite geometric series

2
t] + t\r+ t x
r + + t x
r" +

converges to the sum S =

1 - r

If \r\ > 1 and r, ^ 0, then the series diverges.

500 Chapter Thirteen

 :

Proof: The /7th partial sum of the geometric series is

«. 'id - r")

1. Therefore, if Irl < 1:

~
—
'
r })
(1
r
hm Sc n = rhm 'i
;

—fid —-0)
1 /"
hm r n
since r
•

M—>»
= n when , . , .

r|
^
< ,
1

f|

1 - r

2. However, if \r\ > 1, /" increases or decreases without bound as n

approaches infinity. Thus, S„ becomes infinite and the series diverges.
3. If r = 1, the series becomes the divergent series

t1 + t1 + t1 +~-
4. If r = — 1, the series becomes the divergent series

Example 1 Find the sum of the infinite geometric series

9-6 + 4 .

Solution Since t x
— 9 and r = — 3'

27_
— ;• 5

Example 2 For what values of x does the following infinite series converge?

1 + (x - 2) + (a- - 2)
2
+ (x - 2)
3
+ •
•

Solution This is an infinite geometric series with r =x— 2. By the theorem on

page 500, the series converges when |r| < 1; that is, when \x — 2| < 1,
or

1< x < 3.

This interval 1 <x<3 for which the series converges is called the inter-
val of convergence for the series.

Our final example on the next page illustrates two important facts about
repeating decimals. First, they can be written as infinite geometric series, and
second, they represent rational numbers.

Sequences and Series 501

 .

Example 3 The infinite, repeating decimal 0.454545 . . . can be written as the infinite

series

0.45 + 0.0045 + 0.000045 + • •

What is the sum of this series?

Solution This is a geometric series with t

x
= 0.45 and r = 0.01. Therefore,

P 'i 0.45 0.45 _5_

1 - r 1 - 0.01 0.99 11

CLASS EXERCISES

For each infinite geometric series, find 5^ S 2 S 3 and 5 4

, , . Also, find the sum of
the series if it converges.

2.-L-JL-P4---L + ...
2 4 8 16

3. 1 + 3 + 9 + 27 + •
•

4. 1 + 0.1 + 0.01 + 0.001 +

5. Find the interval of convergence for each series.

a. 1 f A + a
2
+ a-
3 -
b. 1 h 2a + 4a 2 + 8a-
3
+
6. Express 0.3333 ... as an infinite geometric series. For this series, determine
(a) fj, (b) r, and (c) the sum.
7. Discussion Consider any infinite arithmetic series for which t\ # and
d 7^ 0. Explain why this series diverges.

8. Discussion Do you
J think the series 4- + |- + 'T + v5 H 1

tt h
2 3 4 n + 1

converges or diverges? Tell why.

WRITTEN EXERCISES

In Exercises 1-8, find the sum of the given infinite geometric series.

2. 1 1 + 1-J- +
3 9 27

3. 24 - 12 + 6 - 3 + • •
•
4. 1+ -L + -L + ...
4 16 64
5. 5 + 5" + 5~ 3 +•••
1

6. V27 + V9 + V3 + •
•

7. h+h+ ' '

' + tn + • '
', 8. /, + 2 + + t„ + •••,
t

= \-n
where t„ 8(5)"" where n — (— 2) t

9. Find the common ratio of an infinite geometric series with sum 8 and first

term 4.

10. Find the first three terms of an infinite geometric series with sum 81 and
common ratio -L.

502 Chapter Thirteen

 ?

11. What is the value of x if the series 1 + 2.x + 4x 2 + • •

converges to —3

12. What is the value of x if the series x —x +x — • • •

converges to ^-?

For each infinite geometric series, find (a) the interval of convergence and
(b) the sum, expressed in terms of *.
4
13. 1 + .v
2
+ .V + a- 6 + • •
•
14. 1 + 3a + 9a 2 + • •
•

15. 1 + (a - 3) + (a - 3) 2 + • •
•
16. 1 - (x - 1) + (x - l)
2
- •
•

17.1-f
+ A-£ + ...
18 .z_^ + z_...^
19. Show that the series sin x + sin x + sin x H converges to tan x if

-J. 77 _L

20. a. Show that the series tan x — tan x + tan x converges to sin x if

77 -. ^ 77

4 4
b. Are there other values of a for which the series converges?
21. Writing Explain why there is no infinite geometric series with first term 10
and sum 4.

22. Writing Explain why the sum of an infinite geometric series is positive if and
only if the first term is positive.

In Exercises 23-28, use the method of Example 3 to express the given repeat-
ing decimal as a rational number.

23. 0.777 ... 24. 0.636363 . . .

25. 44.444 ... 26. 5.363636 . . .

27. 0.142857142857 ... 28. 0.0123123123 . . .

For the series in Exercises 29-32, find the first four partial sums, Si, S 2 S 3 , ,

S 4 and suggest a formula for S„. Then find the sum of the infinite series by
,

evaluating lim S„. Note that since these series are not geometric, you cannot
use the sum formula given on page 500.

29 — h •

Pj-
• •
1 1

3-4
1 1

1-2 2-3 n(n+ 1)

30 !— + —
—1-3 3-5
—— — i

5-7
- i
1

(2//- l)(2n+ 1)
1

31 —!— H H
On - 2)(3/i +
1

1-4 4*7 7-10

1 1

1)

32
5 -3- + -5— + 7
+ • • •
+ 2n+l 2 + • •
•

'
1-4 4-9 9-16 n\n+\)

Sequences and Series 503

 33. Physics A ball is dropped from a height of 8 m. Each time it hits the ground, it

rebounds — of the distance it has fallen. In theory, how far will the ball travel

before coming to rest?

34. Repeat Exercise 33 for a ball dropped from a height of 10 m if it rebounds

95% of the distance it falls each time.

35. Each side of a square has length 12. The midpoints of the sides of the square
are joined to form another square, and the midpoints of this square are joined
to form still another square. If this process is continued indefinitely, find
(a) the sum of the areas of all the squares and (b) the sum of the perimeters.
36. Each side of an equilateral triangle has length 12. The midpoints of the sides of
the triangle are joined to form another equilateral triangle, and the midpoints
of this triangle are joined to form still another triangle. If this process is

continued indefinitely, find (a) the sum of the areas of all the triangles and
(b) the sum of the perimeters.

M
37. S„ is the /7th partial sum and S is the limit of the geometric series

'iHr + + --
What is the smallest value of n for which S - S„ < 0.0001?

38. Repeat Exercise 37 for the geometric series 1 + —2 + —4 + • • .

39. Here is an old paradox: Achilles races a turtle who has a 100-meter head start.

If Achilles runs 10 m/s and the turtle only 1 m/s, when will Achilles overtake

the turtle?

Erroneous Solution When Achilles covers the 100-meter head start, the turtle
has moved 10 m ahead. And when Achilles covers this 10 m, the turtle has
moved 1 m ahead. Every time Achilles runs to where the turtle was, the turtle
has moved ahead. Thus, Achilles can never catch the turtle.

Correct Solution: Let t

x
= the time for Achilles to cover the 100-meter head
start; let t
2
= the time to cover the next 10 m; let t 3 — the time for the next
1 m. and so on. Find the first three terms and the sum of the infinite series

t\ + t
2 + t3 H .

40. Writing Comment on the following paradox: You can never leave the room in

which you are sitting because in order to do so, you must first walk halfway to
the door. Then you must walk half the remaining distance to the door, and then
half the next remaining distance. Since you must continue to cover the halves
of these remaining distances an infinite number of times, you can never leave
the room.

Exercises 41-44 deal with sequences and series of complex numbers and
should be done sequentially.

41. In an arithmetic sequence of complex numbers, t\ = 1 + 3/ and t

2
= 3 + 4/.

Find the next three terms. Also find t 25 an d S 25 .

504 Chapter Thirteen

 42. In a geometric sequence of complex numbers, t x
= i and the common ratio is
r = 2/. Find the next four terms of the sequence and represent all five terms
graphically in an Argand diagram. Find the sum of these terms.

43. The formula for the sum of an infinite geometric series given on page 500
holds for complex numbers as well as for real numbers. Use this formula to
find the sum of the following series.
•4

1+ + + + - + 2
- 3
+
T T 4 1
27 9/ 3/ i

44. A bug leaves the origin and crawls 1 unit east, \ unit
j unit west, unit south, and so forth, as shown.
north, -A-

f
3 =~J+0i
Each segment of its journey can be considered as a 0-
complex number or vector. Hence, the bug's ultimate
'4 = CI ,

origin
destination can be considered as the sum
1 + 0/
S = /, + t2 + t3 + t4 H

of an infinite geometric series.

a. Find r for this series.

——
b. Use the formula S = - to show that S = — 2.
4- -^/. Thus, the ultimate
r 5 5

destination point is (
—5' —J.
5.
,

c. How far must the bug crawl to reach its ultimate destination?

Pi 45. Consider an infinite geometric series of positive terms that converges to 5.

What fractional part of 5 is the sum of the odd-numbered terms? the sum of the
even-numbered terms? (Give answers in terms of r, the common ratio of the

series.)

46. a. Prove that every rational number — can be expressed as a repeating deci-

mal. (For example, j- = 0.181818 .. . and j= 0.25000 ... or

0.24999 . . . .)

b. Prove that every repeating decimal represents a rational number.

Exercises 47 and 48 refer to the "snowflake curve"

defined as follows: The sides of an equilateral trian-
gle are trisected. A new equilateral triangle is placed
on the middle third of each trisection. The sides
common to the previous figure and the new triangles
are then removed. This process continues indefi-
nitely using the sides of the last figure obtained.

47. Find the area enclosed by the snowflake curve if

each side of the initial equilateral triangle is one

unit in length.

48. Show that the limit of the sequence of perimeters of

the snowflake curve is infinite.

Sequences and Series 505

 —

1 1 1 1 CALCULATOR EXERCISE

The infinite series —+ — + —— + — + • • •

converges to
1

0.3333

Determine how many terms of the series must be added to make the sum:
a. correct in at least the first 2 decimal places, that is 0.33.
b. correct in at least the first 3 decimal places, that is 0.333.
c. correct in at least the first 4 decimal places, that is 0.3333.

13-6 Sigma Notation

Objective] To represent series using sigma notation.

The Greek letter 2 (sigma) is often used in mathematics to express a series or its

100

sum in abbreviated form. For example. 2j k represents the series whose terms are
*=]

obtained by evaluating k first for A = 1. then for k — 2, then for k = 3, and so on

until k = 100. That is,

100

2 k
2
= l
2
+ 2
2
+ 3
2
+ •••+ 100
2
.

k=l

The symbol on the left above may be read as "the sum of k for values of k from 1

10

to 100." Similarly, the symbol Zj 3 A' may read as "the sum of 3k for values of k
k=5

from 5 to 10." This symbol represents the series whose terms are obtained by
evaluating 3 k first for k = 5, then for k = 6. and so on. to k— 10. That is,

10

2 3£ = 3-5 + 3-6 + 3-7 + 3-8 + 3-9 + 3-10= 135.

*=s

The symbol 3k is called the summand, the numbers 5 and 10 are called the limits
of summation, and the symbol k is called the index. Any letter can be used for the
index. For example:

i (- i)" = (- i)
/1=1
'
+
2
(- i) + (- i)
3
+ (- i)
4
+ (- i)
s

= _± + _L__L + J___L =
2 4 8 16 32 32

Sigma notation can also be used to represent an infinite series and its sum.
For example:

+-
y , 2 3
2(i) =(i)° + (|) + (i) + (i)
y=o
= l+l + l + l + --- = 2

506 Chapter Thirteen

The symbol Zj \-z) is read "the sum of (
— J
for values of j from to infinity"; it

j=0

represents both the infinite geometric series at the bottom of page 506 and its

sum, 2.

When you get used to using sigma notation, you will find it much easier than
manipulating series that are written in expanded form. The following properties are
consequences of the commutative, associative, and distributive properties of the
real and imaginary numbers. These properties are proved in Exercises 29 and 30.

Properties of Finite Sums

n n

1. 2 (flt + bt ) = 2 a, + 2 bt 2. 2*i ca, = cZj Qj

i=\ i=\ i=\ i=l

The properties above can be used together with previously derived sums to
derive the sums of many other series. Example 1 illustrates how these properties
may be used with the following known sums.

n(n + 1)
sum of integers: Zj k = (Proved in Section 13-3)
k=\ 2

sum of squares: 2j k
V 2
=
n(n+ l)(2w + 1)
(To be proved in Section 13-7)
A=l 6

n(n + 1)
sum of cubes: Z^ k = (To be proved in Section 13-7)
k=\

Example Express 1 • 2 + 3 • 4 + 5 • 6 + • •
+ 199 • 200 in sigma notation and
evaluate.
__^___ 100

Solution 1 • 2 + 3 • 4 + • • •
+ 199 • 200 = 2 (2k - \)(2k)

100

= 2 (4£
2
- 2k)
k=\

100 100

= 42 k
2
- 2 2 k
A=l k=\

100- 101 -201

-2 100- 101

= 1,353,400 - 10,100 = 1,343,300

Sequences and Series 507

 CLASS EXERCISES

Give each series in expanded form.

1. 2 5k 2. 2 k
2
3. 2(-l) 7 4.2
k=l k=3 7 =2 n=\

W
Express each of the following series using sigma notation.

5. 4 + 9 + 16 + 25 + 36
«-M +
7. The arithmetic series: 3 + 6 + 9 + • • •
+ 300

8. The infinite geometric series: ~T + ~^" + ^7 + ~£T +

WRITTEN EXERCISES
In Exercises 1-8, write the given series in expanded form.
10

1. 2
k=2
2. 2 5£ 2
*=1
4. S (4/; - 7)
/t=i n=3
6
""
5. S 3
1

6. S ;(-!)' 7. 2 4' 8. 2 1
13 - 3s|
M= t=-2 j=0

In Exercises 9-16, express the given series using sigma notation.

9. 4 + 8 + 12 + 16 + 20 10. 1+2 + 4 + 8+16 + 32

+ + +•••+ + 4 + + + +

M
11. 5 9 13 101 12. 2 6 8 • • •
200

-
13- 1+ { + { + i+- "•M + +
15. sin x + sin 2.x + sin 3.v H 16. 48 + 24 + 12 + 6 + •••

4 4

17. Show that 2 log f = log 24. 18. Show that 2 k log 2 = log 2
10
.

r=l

100 50

19. Evaluate: 2 cos £77 20. Evaluate: 2 sin (*

/
•
yj
\

A=l k-

8 i

V2 V2.
[j]
21. Evaluate: a. 2 (^ + ^»)
«=i
S
n=i

22. Evaluate: Zj (— + —i) (Hint: This is a geometric series with |/| < 1.)
n= 1

508 Chapter Thirteen

 8 100
4.' 6 '*
23. Show that 2 (-1)* log it = log I
V
2
1
'

• 3 • 5 • 7/)
24. Evaluate 2 LVa?J, where
*=1 t=]
L/:J is the greatest integer of k.

In Exercises 25-28, express the given series using sigma notation.

1-1 + 1 b -- 1-1-1
K 1 +
2~41-1-1
+ 1-1-1
25. a.
16 32 I"T6 + ^2
26. a. 27 - 9 + 3 - + 1 _ 1 b. -27 + 9 - 3 + - j +
^
1 1
3 9

27. 1 - 3 + 5 - 7 + • • •
- - 99 28. -2 + 4 - 6 + •••+ 100

Show how the commutative, associative, and distributive properties are used
to prove each of the following.

29. 2 ia
i=i
t
+ bt ) = 2 <H + 2
;=1
bt 30. 2
;=1
ccij = c 2
1=1
flj

Use the properties of finite sums to prove each of the following.

n n n n

31. 2 (a, + bj)

2
= 2 fl,
2
+ 2 2 tf,A + 2
;=1
^,
2

i=i

32. 2 (a.x
2
+ bx + c) = a 2 -v
2
+ b 2 x + C7?
A=l ,v=l A=l

Evaluate the following series using the method shown in the example.

33. 1 + 2-3 + 3-4 + •••+ 100- 101

-2
34. 1 -3 + 3 -5 + 5-7 + •• + 99- 101
35. 1 2 3 + 2 3 4 + 3 4 5 +
• • + 20• • • • • • • 21 • 22
36. 2 4 + 6 8 + 10 12 +
• •+ 398 400 • • • • •

V% 37. a. An 8-by-8 checkerboard has 64

little squares and many other
squares of various sizes. How
many squares does it have in all?

b. Find the total number of squares

in an /?-by-/? checkerboard.

Sequences and Series 509

 38. a. The 3-by-3-by-3 cube shown has
27 little cubes but also many other
cubes of various sizes. How many
cubes does it have in all?

b. Find the total number of cubes in

an n-by-n-by-n cube.

39. A stack of oranges is compactly ar-

ranged so the bottom layer consists of

oranges in an equilateral triangle with
n oranges on a side. The layer next to
the bottom consists of an equilateral
triangle of oranges with n — 1 oranges
on a side. This pattern continues up-
ward with one orange on the top. How
many oranges are there? Fourth laver

////COMPUTER EXERCISE
3C

The series 2 — diverges, but very slowly. Find how many terms are needed to
n=i

make the partial sum 1 + —+—+ • • •

H— greater than: (a) 3; (b) 4; (c) 10.

(The answer is surprising!)

13-7 Mathematical Induction

Objective To use mathematical induction to prove that a statement is

true.

Observe the pattern in the following statements:

1 -2 2

_ 2

111
1 !
1

1-2 2-3 3

_ 3
"
1-2 2-3 3-4 4

appears that

1
+ 1
+ 1
+•••+ l
= //

1-2 2-3 3-4 /?(/!+ 1) n + 1

Nevertheless, this appearance does not constitute a proof that the statement is true
for all positive integers n. One way to prove the statement true is to use a method
called mathematical induction.

510 Chapter Thirteen

 .

Proof by Mathematical Induction

Let S be a statement in terms of a positive integer n.

Step 1. Show that S is true for n = 1.

Step 2. Assume that S is true for n = k, where k is a positive integer,
and then prove that S must be true for n = k + 1

If you can do both steps (1) and (2) above, then you can conclude that S is
true for all positive integers. The reason is that once you know that S is true for

n = 1, step (2) tells you it is true for n = 1 + 1 = 2. Applying step (2) again, 5
must be true for n = 2 + 1 =3, then for n = 3 + 1 = 4, and so on.
You might think of mathematical in-
duction as something like setting up domi-
noes so they will all fall down. Set up the
dominoes so that if any one domino falls,
the next domino Then knock over
will fall.

the first domino, which will knock over the

second, which will knock over the third, and
so on. Thus, you can conclude that all the
dominoes will fall down.

Example 1 Prove that + 1

+ + 1
for all posi-
1 -2 2-3 3-4 n{n + 1) n + 1

tive integers n.

Solution

Step 1 Show that the statement is true for n = 1.

1
Yes, it is true.
1(1+ 1) + 1 1

Step 2 Assume that the statement is true for n — k, and then prove that it must be
true for n = k + 1.

Assume:
1
+ 1

+ + 1

1 -2 2-3 k(k+ 1) A + 1

Prove:
1

+ + + 1
+ 1 k+ 1
or
1 -2 2-3 k(k+ 1) (k+ 1)[(*+ 1) +1] (A- + 1) + r
1
+ + + 1
+ 1 k+ 1

1 -2 2-3 k(k +1) (k+ !)(£ + 2) k + 2

Basic strategy at this stage of proof: Show that the left side of the
"Prove" statement is equal to the right side. To do this, take the left side
and try to simplify it by using the commutative, associative, and distribu-
tive properties and the "Assume" statement. This is shown in the proof
on the next page.

Sequences and Series 511

 Proof:
+ + +
1 -2 2-3 k(k + D (k + \)(k + 2)

+ + + +
2-3 k(k + 1) {k + \)(k + 2)

Use the "Assume" statement.

4
k
+
k+ 1 (*+ l)(A

k(k + 2)
+ 1

(*+ t)(* + 2) a+ ua- + 2)

2
k + 2k + 1

(* + l)a- + 2)
)-
(^- + 1

a+ \)(k + 2)

k + 1

k + 2

Example 2 Prove that n + 2n is a multiple of 3 for all positive integers /;.

Solution

Step 1 Show that the statement is true for n — 1.

r + 2 • 1 = 3 is a multiple of 3. Yes. it is true.

Step 2 Assume that the statement is true for /; = k, and then prove that it

must be true far n = k + 1.

,

Assume: k' + 2k is a multiple of 3.

3
Prove: (k + l) + 2{k + 1) is a multiple of 3.

Proof: Use the same basic strategy as in Example 1.

3 3 2
(k + I) + 2(k + 1) = k + 3k + 3k + 1 + 2k + 2
- (A
3
+ 2*) + 3(A
2
+ *+ 1)

Use the
assumed
statement.

4
= a multiple of 3 + a multiple of 3

= a multiple of 3

In the last line of the proof, we have used the fact that the sum of
two multiples of 3, say 3/ and 3/, is another multiple of 3,

namely 3(/ + j).

•

512 Chapter Thirteen

WRITTEN EXERCISES

Use mathematical induction to prove the statements in these exercises. Note

that some of these statements can also be proved by other methods.

= n(n + 1)
-
1. 1 + 2 + + n 2. 1 + 3 + ••• + (In 1) = n

n n

3. Zj 2/ = //" + n 4. 2 2''
" ' = 2" - 1

i=i i=]

n{n + + "(" + M" +

5. 2
;=1
r =
\)(2n
6
1)
6.£,(, +
;=
D= ' 2)

"- )(2/?+1)
7. S(2/-l) 2 - " (2 1

3
8. (1 + a)" > 1 + /?.v. where x > -1

9. 2
;=1
(i -2' !

) = 1 +(«- l)-2" 10. 18" - 1 is a multiple of 17.

11. 1
1" — 4" is a multiple of 7. 12. //("" + 5) is a multiple of 6.

For Exercises 13 and 14, decide which

values of n will make the inequality true
and then prove your answer is correct.

13. 2" > In 14. n\ > 2"

15. If. in a room with n people (n > 2),

every person shakes hands once with

every other person, prove that there
2
n
are " „ handshakes.

16. Prove that the sum of the cubes of any

three consecutive positive integers is a
multiple of 9.

17. Use the triangle inequality (see Exercise 36 on page 99) to prove that

\
a \
+ a2 +" + ' a n\ — \
a \\ + l
fl
2! + ' ' " \a nl-

lS. Prove De Moivre's theorem (page 408) for every positive integer /;.

In Exercises 19-21, write out the first few terms of each sequence. Then
suggest and prove a formula in terms of n for the wth term, a n .

a ><-
10
19. a x
= 1
-, an =a„- - l -^j 20. a 1
= 1. a„ = 2a „ _ l
+ 1

21. a, - -. a„ = a„ _ + 1

1
(3w - 2)(3// + 1)

Sequences and Series 513

 In Exercises 22-24, write out the first few terms of each sequence. Then
suggest and prove a formula in terms of n for the wth partial sum, S n .

22. s„ = S ,-
3

23. S n = Zj (i ' '!) (Hint: Look at the sequence of factorials of integers.)

;=1

24. S„ = Zj /(/' + 1)(/ + 2) (Hint: See Exercises 1 and 6.)

25. Prove that a convex polygon with n sides (n > 3) has

2
—— diagonals.
_ ^

26. Prove that it is possible to pay any debt of $4, $5, $6. $7, . . .
, $n, and so on,
by using only $2 bills and $5 bills. For example, a debt of $1 1 can be paid with
three $2 bills and one $5 bill, or $1 1 = 3 • $2 + 1 • $5.

B Chapter Summary
1. A sequence is a function whose domain is the set of positive integers. A
sequence can be defined explicitly by a formula such as tn = 3n + 6 or recur-
sively by a pair of formulas, such as t
x
=6 and t
n
= 7 — tn _ x
.

2. A sequence is arithmetic if the difference d of any two consecutive terms is

constant. A sequence is geometric if the ratio r of any two consecutive terms is

constant.

3. A series is an indicated sum of the terms of a sequence. The formulas on pages

487 and 488 can be used to find the sums of finite arithmetic series and finite

geometric series, respectively.

4. If the /7th term of a sequence becomes arbitrarily close number L as /; gets
to a

larger and larger, we write lim t

n = L. If there is no single limiting number L
for all terms of the sequence, we say that the sequence has no limit. A se-
quence whose terms increase (or decrease) without bound is said to have a
limit of oc (or — oo).
5. Associated with any infinite series t
x
+ t2 + t
3 + • •
\ there is a sequence of
partial sums:
Si — t
{ ,
S2 — t\ + t2 , S3 = 11 + t<i + f3 , and so on.
If this sequence has a limit L, we say that the series converges to the sum L.
If the sequence of partial sums has no limit, the series diverges.

6. The infinite geometric series t\ + t\r +,t\r H converges to the sum

S = ^_
if |r| < 1. The series diverges if |;|
> 1 and /] ¥^ 0.

7. Any infinite repeating decimal can be written as an infinite geometric series

whose sum is a rational number.

514 Chapter Thirteen

 —

8. The Greek letter 2 (sigma) is often used to express a series or its sum. Some
properties of finite sums are given on page 507.
9. Mathematical induction, used to prove that a statement S is true for all positive
integers, involves two steps: (1) Show that S is true for n = 1, and (2) prove
that S is true for n = k + 1 whenever it is true for n — k.

Key vocabulary and ideas

sequence (p. 473, p. 474) sequence of partial sums (p. 500)
arithmetic and geometric sequences (p. 473) sum of an infinite series (p. 500)
recursive and explicit definitions (p. 479) converge, diverge (p. 500)
series, sum of a series (p. 486) sigma notation (p. 506)
limit of a sequence (p. 493) mathematical induction (p. 510)

rt Chapter Test
1. State whether each sequence is arithmetic, geometric, or neither, and 13-1
find a formula for t n in terms of n.

a. 10, 6, 2, -2,... b. 3,6, 11, 18, 27, ... c. -2, 6, -18, 54, . . .

2. In an arithmetic sequence, t2 — 6 and t

6 = 16. Find t
2 \-

9
3. In a geometric sequence, f3 = 9 and t 6 — —. Find /
)2 .

4. In a sequence, /] = 1, t2 = 4. and t„ — 2tn _ i

+ tn _ 2 - Find t 6 . 13-2

5. Give a recursive definition for the sequence 1, 5, 21, 85, ... .

6. For the sequence of positive 3-digit integers ending in 4, find an 13-3

explicit formula for (a) t„ and (b) S„.

7. Find the sum of the first 6 terms of the series 27 — 9 + 3 — 1 H .

8. Writing Briefly tell what the word ''limit" means in mathematics. 13-4

9. Find: a. lim
„^oc 4/T
}
n~
-
+
In
l

+
— 5
b. lim n cos n tt
"-^ x
c. lim (0.59)"
»^ x

10. Express 0.131313 ... as an infinite series and as a rational number. 13-5

11. Find (a) the interval of convergence and (b) the sum, expressed in

terms of x, for the series 1 + 3x

^j- H — 9x~
+

12. Express each of the following series using sigma notation. 13-6

a
i
!
"
>
+
i
i

49
1

+ b. 7 + 3-1-5-9-13-17
*
9 25 81

20

13. Evaluate 2 3k(k + 2).

k=l

14. Prove by mathematical induction: 13-7

n(5n -
2 + 7 + 12 + ••• + (5« - 3) = 1)

Sequences and Series 515