CbapterIV

1701 ONWARDS
4.1 The Eighteenth Century

The development of analysis and its applications in various fields

formed the central aspect of the history of mathematics in the eighteenth
century. There was also important work in other areas.

There were several major texts in algebra published during the

century, including the works of Maclaurin and Euler. These books contain
some systematization of earlier material. Maclaurin's text contained a new
method of solving systems of linear equations. Usually, they referred to as
Cramer's Rule. Euler's book contained some details on various methods in
number theory. However, the central target of algebra, was the extension of the
equation solving techniques of Cardano (1501-1576) and Ferrari (1522-1565)
to polynomial equations of fifth degree and higher. Noone succeeded in this
effort. Towards the end of the century, Lagrange produced a detailed review of
the methods for solving cubic and quartic equations. He developed ideas,
which would be essential in dealing with higher degree equations.

Issac Newton (1642-1727)

Issac Newton was born on Christmas day of 1642, the year of

Galileo's death. The four Newtonian books that are best known today are the
Principia the 'Method of Fluxions' the Opticks and the Arithmetica universalis. The
last work was composed between 1673 and 1683, and it was first published in
1707. This treatise contains the formulas, usually known as 'Newton's
Identities' for the sums of the powers of the roots of a polynomial equation.
releal~IV
-

Figure 4.1 : Issac Newton

Cardano had known that the sum of the roots of

x" +a1x n-I + + an_Ix + an = 0 .

IS - a, . V·lete
, ha d £anned th i atlOns
e re ·

between the roots and coefficients. Girard in 1629 has shown how to find the
sum of the squares of the roots, or the sum of the cubes or of the fourth powers.

According to [10], Newton generalized this work to cover all

powers.

If k ~ n, the relationships

If k > n, the relationship

Sk + a1Sk_1 + ... + an_ISk_n+1 + a"Sk_" = 0 holds, where S; is the sum

of the i th powers ofthe roots.

U sing these relationships recursively the sums of the powers of

the roots can readily be found for any integral power. Also he introduced the
system of' literal indices' such as am and a" [24].

In the Arithmetica universalis there is another theorem generalizing

'Descartes Rule of Signs' to determine the number of imaginary roots of
polynomial. It is along with a rule for an upper bound for the positive roots
[10], [17],[74].

Make a series of fractions whose denominators are 1,2, 3...upto

the highest power in the equation, and whose numerators are the same numbers
in the reverse order. Divide the second by the fIrst, and the third by the second,
and so on, placing the results over the terms in the equation, beginning with the
second. If, under any term, if the square of the term, multiplied by the fraction,
exceeds the product of the terms on either side, place a '+' sign underneath;
otherwise, place a '-' sign. Under the fIrst and last terms place a '+'. The
number of alterations of signs, from '+' to '-' or from '-' to '+', will be the
minimum number of complex roots. He did not prove this rule.

Example 4.1 [77]

For the equation x 3 + px 2 + 3p2 X- q =0, with p, q assumed

positive, form the fractions~,.~). Dividing the second by the

123
fIrst, we obtain!; likewise, dividing the third by the second, we
3
obtain!. Comparing the squares of each term, multiplied by the
3
fraction above it, with the product of the terms on both side, and
writing a '+' if the square is greater than a '-' if the term is less,
we obtain:

1
-
3
+ 3p 2 X - q = 0
+ +

There are two alterations of signs; hence there are at least

two complex roots.

If there is a single missing term, it is considered as a 0 term.

~/UJ/WwIV

Example 4.2 [77]

Applying Newton's method to the equation x 4 - 6x 2 - 3x - 2 = 0

we determine

3 4 3
g 9 g
x
4
+ ** - 6x
2
- 3x - 2
+ + + +

where the missing term is indicated with a '**'. There are two
alterations in sign, and consequently at least two imaginary roots.

If there are two or more missing terms in a row, below the

missing terms should be placed a '-' then a '+', then a '-' alternatively. But
below the last missing term should always be '+'.

Example 4.3 [77]

Applying Newton's method tox 5 -1 =0, Newton's procedure

giVes

2 1 1 1 2
5 2 2 2 5
x +
5
** + ** + ** + ** - 1
+ + + +

There are four changes in sign, so there are at least four complex
roots.

In some cases, Newton merely states his theorems. The proofs

were not discovered for several years. For example, the rule for finding the
number of imaginary roots of an equation was not thoroughly understood till
1865 when Sylvester published a paper on the subject [74].

Newton showed that in equations with real coefficients imaginary

roots always occur in pairs [12],[17],[74].
rglw/devlV
-
It is gIven In [73] that, Newton has fIrst shown that in the
·
equat lon. x /I + alx /I-I + a 2x /1-2 + =
0 , - a is the sum of the roots, a the
J 2

sum of the product of two terms of both of the roots and so forth.

Approximate Solution of equations

Newton introduced a method for approximating the values of the

real roots of a numerical equation. This applies equally well to either an
algebraic or a transcendental equation. This method is modifIed and is known
as Newton's method, which is given in [17] as follows:

"If f(x) = 0 has only one root in the interval [a, b] and if neither

j'(x) nor f"(X) vanishes in this interval, and if Xo be chosen as that one of the

two numbers a and b for which f(x o) and f"(X O) have the same sign, then

Xl = Xo - f(x o ) is nearer to the root than is xo ."

f'(x o)

To construct a series for functions defmed implicitly, such as

l +a 2 y - 2a 3 + axy - x 3 = 0, Newton used a method of successive
approximations. He invented this. It is called the method of reversion. He
called such equation an 'affected equation' [77].

According to [77], to solve an equation y3 -2y-5 = 0 he fIrst

noted that 2 is an approximate solution. Let the actual solution be represented

by y = 2 + p, where p is a small quantity. Substituting this in the original

equation and simplifying, he got p3 + 6p2 + 1Ox -1 = 0 .

Since p is small, he neglected the higher terms of p, and got

10p-1=O

Then, p = 0.1 as the approximate solution to the second equation.

Then the actual solution should be p = 0.1 + q where q is another small number.
rg/w/Ibv IV
...
Substituting p = 0.1 + q where q is another small number. Substituting 0.1 + q
for p in the second equation, he got a third equation
q3 +6.3q2 + 11.23q + 0.061 = o.

Since q is small, he neglected q3 and q2.

Then he got 11.23q + 0.061 = 0

Then q = -0.0054 as a solution.

Again, since q = -0.0054 is an approximate solution, the actual

solution to this equation should be - 0.0054 + r , where r is small.

Then, he ignored the q3 term in the third equation, so the fourth

equation is obtained by substituting - 0.0054 + r into the equation
6.3q2 + 11.23q + 0.061 = 0

2
Ignoring the r term, solves this equation for r. Since the root is
2+ p + q + r + , were the values p, q, r.... were found, the solution

2.09455147 can be found. Then he tabulated his results. In many places,

Newton wrote only the coefficients of each polynomial.

+l +8 +12p +6i +/
-2y -4 -2p
y=2+p
-5 -5
Sum -1 +lOp +6i +/
p = 0.1 +q +l +0.001 +0.03q +O.3l +l
+6l +0.06 +1.2 +6.0
+10p +1 +10
-1 -1
..
Sum +0.061 +l1.23q +6.3l +l

+6.3l +0.000183708 -0.06804r +6.3r2

+11.23q -0.060642 +11.23

q=-0.0054 + r
+0.061 +0.061
Sum 0.000541708 +l1.16196r +6.3/
r=
-0.00004854+s

Newton's Method of Reversion for / - 2y - 5 = 0 [77].

Leibniz (1646-1716)

Gottfried Wilhelm Leibniz was primarily a diplomat. He was the

first European to use matrices to represent and solve a system of equations. On
April 28, 1693 he wrote to L 'Hospital that he found it convenient to let
numbers represent the coefficients in a system of equations. For a system with
two unknowns and three equations, he wrote as in [77]:

10+ llx + 12y = 0

20+21x+22y=0

30+31x+32y=0

where a number, such as 32, represents the coefficient of the

second variable in the third equation. He used first and second equations to
eliminate y. He multiplied the first equation by 22 and the second by 12. Then
subtracted, he got

10.22 + 11.22x -12.20 -12.21x =0

From it, he got the value of x.

The theory of determinants is usually said to have originated with

Leibniz. In 1693, he considered these forms in connections with a system of
-
simultaneous linear equations. Seki Kowa in Japan ten years earlier made
similar consideration [17], [69].

Figure 4.2 : Leibniz

The matrix algebra of Leibniz and the Fang Chang method of the
Chinese show close similarity. During his time there was interaction between
China and the West. Leibniz knew Claudio Filippo Grimaldi who was a Jesuit
mathematician appointed as a representative to the Chinese Imperial court.
However, there is no evidence that Grimaldi, or anyone else, ever
communicated to Leibniz any details of Chinese mathematics.

Maclaurin (1698-1746)

Maclaurin's work, 'Treatise of Algebra in Three Parts' was

published in 1748 although probably written in the 1730s. He defined algebra
as "A general method of computation by certain signs and symbols which have
been contrived for this purpose and found convenient. It is called a universal
Arithmetic and proceeds by operations and rules similar to those in common
arithmetic, founded upon the same principles" [36].
rglllJ/ikv IV
-

Figure 4.3 : Maclaurin

To Maclaurin, algebra is not 'abstract' but simpl generalized

arithmetic. He shows how the basic operations can apply to negati e
quantities.

In [36] it is given that, Maclaurin demonstrates how to deal 'tb

positive and negative quantities, To show the reason for the rule of igns in
multiplying such quantities, he notes that since + a - a = 0, also n ( +a - a = O.
The first term of this product, +na is positive. Therefore the second term m
be negative. Therefore, -a multiplied by +n is negative.

Similarly, since -n (+a - a) = 0, and the first teml r tbi

product is negative, the second term (-n) (-a) must be positive and qu t

ina.

In the fITst part of his work, Maclurian deals with su h t pi

manipulation with fractions, powers of binomials, roots of polyn mial an
sums of progressions.

For linear equations in more than one unknown

two and three equations in the same number of unknowns th
found by solving for one unknown in terms of the others and th n ub 'tuti .,
If there are more unknowns than equations, there may b an infinit num f
solutions. On the other hand, there may be no solution at all. He does not give
any examples of either case [36].

Maclaurin presents what he calls a 'general theorem' for

eliminating unknowns in a system of equations. Today this method is known
as 'Cramer's Rule', named after the Swiss mathematician Gabriel Cramer
(1704-1752). Cramer used it in a book on curves in 1750. A Chinese form of
lCramer's Rul~ existed nearly 2000 years before. For the system of equations,

ax+by =c

dx + ey = f , solving the fIrst equation for x and substituting, gets

y = af - de and a similar answer for x [10],[36],[77].

ae-db

A problem of the seventh chapter of Chiu Chang (Chapter 1 page

86) can be restated in terms of a system of equations in two unknowns x and y,
where x is the number of persons and y is the cost.

Then ax - ey = b and a' x - e' y = -b' (1)

Ifa = 9,a'= 6,b = 11,b'= 16,c = 1,c'= 1, then (1) becomes

9x - y = 11, 6x - y = -16 .. (2)

(1) and (2) can be expressed in matrix form as

Apply 'Cramer's Rule) for solving simultaneous equations.

-
be'+b' e ab'+a' b
X= ,y=---
ae'-a'e ae'-a'e

. 11 + 6 144 + 66
l.e., x=--=9, y= =70
9-6 9-6

When e = e' = 1, the result is identical to the solution given in the

Chiu Chang.

As early as in the beginning of the Christian era a variant of

tramer's Rule'" for solving two equations in two unknowns was known in
China. But there is nothing in either this treatise or any of the subsequent
commentaries that hints on the rule for three equations in three unknown or
generally nothing for p equations in p unknowns. However, we can find the
concept of a determinant in this method. But Chinese mathematicians did not
notice it. It was taken up by the Japanese Mathematician Seki Kowa in 1683.
Historians of Mathematics credit him with the discovery of determinants.

In Maclaurin's terminology, the 'orders' are the equation

coefficients; the 'opposite coefficients' are the coefficients of different
variables in different equations. Then <Cramer's RulJ is as described by the
Maclaurin:

"[The] numerator is the difference of the products of the opposite

coefficients in the orders in which y is not found, and the denominator is the
difference of the products of the opposite coefficients taken from the orders
that involve the two unknown quantities" [77].

In modem matrix terminology, the 'orders' correspond to the

column entries.

As in [10], [36], [77] consider the system ofthree equations,

ax+by+cz = m
-
dx+ey+ fz =n

gx+hy+kz = p

He dealt with by fIrst solving each equation for x. Then reduced

the problem to one in two unknowns. Then he applied the earlier rule to fmd

aep - ahn + dhm - dbp + gbn - gem

z= ---:;------"-------"-------'---
aek - ahf + dhc - dbk + gbf - gee

He notes the general rule that the numerator consists of the

various products of the coefficients of x and y as well as the constant terms,
each product consisting of one coefficient from each equation. The
denominator consists of products of the coefficients of all three unknowns.
Also, he explains how to determine the sign of each term. Then he solves for y
and x. Also he shows that the general rule determining each of these values is
similar to the one for z. Each of the three expressions has the same
denominator.

Maclaurin extends the rule to systems of four equations in four

unknowns. But he did not tell anything about any further generalizations [36].
The numerators and denominators involved in Maclaurin's
solution are known today as determinants. The use of such combinations of
coefficients as tools for solving systems of linear equations had appeared
somewhat earlier, Leibniz had suggested a similar idea in a letter to L'Hospital
in 1693. Japanese Mathematician Seki Kowa described the use of determinants
in 1683. He was showing, by the use of diagrams, how to decide whether a
given term is to be positive or negative [36].

Maclaurin tried to prove Newton's method of determining the

number of complex roots. After proving a number of introductory lemmas,
Maclaurin showed, "In a quadratic with two real roots, the square of the second
term is greater than four times the product of the first and third" [77].

Maclaurin gave a corresponding result for the cubic equation, "In

a cubic with three real roots, the square of the second term is greater than three
times the product of the fust and third."

Therefore, if the square of the second was not greater than three
times the product of the fust and third, the cubic equation must have two
complex roots. Maclaurin pointed out that this was equivalent to Newton's
rule.

The second part of Maclaurin's work is a treatise on solving

polynomial equations. This presents all that had been discovered up to his time
in well organized form. He includes ICardano's Ruld for solving cubics,
(Ferrari's Rule' for quartics, (Descarte's Rule of Sign~ and Newton's methods for
approximating numerically the solution to an equation. In it he shows that no
equation can have more roots than the degree of the highest power. Also,
"roots become impossible [complex] in pairs". Therefore, "an equation of odd
dimension has always one real root" [36].

Euler (1707-1783)

Euler gave a better introduction to algebra in his Vollstandige

Anleitung zur Algebra ('Complete Introduction to Algebra'). He defines algebra
as

The foundation of all the mathematical sciences must be

laid in a complete treatise on the sciences of numbers and in an
accurate examination of the different possible methods of
calculation. This fundamental part of mathematics is called
Analysis, or Algebra. In algebra, then, we consider only
numbers, which represent quantities, without regarding the
different kind of quantity [36].
-
Later in the text, he says Algebra is, "the science which teaches
how to determine unknown quantities by means of those that are known".

Figure 4.4 : Euler

Euler has a discussion of the algebra of positive and negative

quantities given in [36]. He says, "Let us begin by multiplying -a b T. ow

since, -a may be considered as a debt, it is evident that if we take that

three times, it must thus become three times greater and consequen
required product is -3d'. [Link] notes that -a times b will be -ba or -ab. H
says that -a times -b cannot be the same as -a times b or -abo
equal to +ab.

Euler introduces the concept of an imaginary number:

Since all numbers which it is possible to con i e

either greater or less than 0 or 0 itself, it is evident that e cannot
rank the square root of a negative number among t .
numbers, and we must therefore say that it is an im
quantity. In this manner we are led to the idea of numb
form their nature are impossible; and therefore th
called 'imaginary quantities', because they exist m
imagination. All such expressions as ~,A....... n
impossible or imaginary numbers, since they r p
negative quantities. But not withstanding these num
themselves to the mind; they exist in our imaginati n.
still have a sufficient idea of them; since we 1m th
is meant a number which, multiplied by itself pr u
this reason also, nothing, prevents as from makin
imaginary numbers, and employing them in cal ul ti n [
..
Euler does not realize that there may be problems in these

calculations. Although he has noted that ~ x ~ = 4, later he writes that the

general rule for multiplying square roots implies that ~ x ~ = .J4 = 2.

Although, Euler presents the determinant form of the solution for

a system of two equations in two unknowns, his standard method for larger
systems is that of solving for one variable in terms of the others. By this he
reduces the system to one of fewer equations in fewer unknowns.

Euler was a prolific writer on mathematics. His name is linked to

every branch of the subject. The following are some of the contributions to
notations:

f(x) for the functional notation,

e for the base of natural logarithms,

a, b, c for the sides of a triangle ABC,

S for the semi perimeter of triangle ABC,

2: for the summation sign,

i for the imaginary unit ~

'The Euler ¢ function' the Beta and Gamma functions of

calculus originated with Euler [15], [17].

A modem method for getting the formula for the solution of the
quadratic is as follows. This method uses determinants, given by Euler and
Etienne Be'zout (1730-1783) which was improved by Sylvester (1840) and
Hesse (1844)

Given x 2 + px + q = 0
-
x=u

3
Then X px qx: = 0 .. . .. .... .. .. . .. ....... ..... ...... ...... (1)

x - u . =0 (2)

x - u = 0 (3)

From (1 an ) eget

1 P q
o 1 - u =0
1 -(u -

Expandin~ - p u z)- u )- -q = 0

-p
Taking p=o then =

pz q)=O eget

- P 1 I
Then x = - ± -Vp 2 -4q [69].
2 2

Euler sho s that hen a root x + yH exists there is also one of

the form x - ~ so that there must be a factor of the form xx + px + q [73].

It is gi en in [73] that Euler proves three theorems:

(i) That an equation of odd degree has at least one root

-
(1Ill)

Be"zom: so 11$ SIIX-

~
in! BS IIDD.
"
wittlIn of
m " " "
0llI. In "Ai im'
1
inatreaUiseo 1 9

he mI to "'TfO"
-
imultaneo linear equations in 'n unknowns. He i known for an extension
of the e to a r em of equations in one or more unknowns in wbi h it is
required to fin the ondi'on on the coefficients nece sary for the equations 0

have common oIution.

Th ne ary ondition that the equations QI X hry C1 =0,

QzX bz.y= =0 °3 X b3 y c~
~
=0 have a common solution is that e

°1 bl
eliminant a b- ho d be equal to 0 [10].
-
~ ~ c3

Vande 0 e ( 735- 796)

The ep to ards the final answer to qu 0 of

soh abili of equations of any degree that could be sol e al ebrai . were
taken up exan Th pbile andennon e.

He res ed 0 th Paris ca e Reso 000

Equatio . in O. In it he analyze e of e al ebrai so 'ODS 0

polynomial [Link]. He an b o· (J that e so ability of qmlldratjc

equa ·on x - ° b).- ab=O d ded on e fa that: the roo a an b, 0 I

o the oeffi ients as !( a )

This a e 0 solutions depernllmg on wtIletb~~

positive or n . e square roo of a by -4ab [ [81].

andermo no asks ·onof~~n

can be sol similar expressions

1
Xu •••• XIII
n
 ... + n (PIn-I XI +.~.+ P/I n-I Xn )n]

where Pl'P2' .....,P n are the nth roots of unity. Today expressions like

PIXI + P2X2 + + PnXn are called "Lagrange resolvents". For cubic equations,
the method ofVandermonde and Lagrange leads atonce to the solution [81].

Joseph Louis Lagrange (1736-1813)

Joseph Louis Lagrange was born at Turin in 1736 and died at

Paris in 1813. In the eighteenth century, another aspect of algebra considered
was the solution of polynomial equations. Many mathematicians tried to
generalize the methods of Cardano and Ferrari to solve algebraically
polynomial equations of degree five and higher. But they were not successful.

Figure 4. 5 : Lagrange

It is given in [10),[36) that Lagrange, in his Re'f1exions sur la

the'arie alge'brique des e'quatians ('Reflections on the Algebraic Theory of
Equations') of 1770 began by undertaking a detailed review of these earlier
solutions to determine why the methods for cubics and quartics worked. He
was unable to fmd similar methods for higher degree equations. But he
sketched a new set of principles for dealing with these equations. This was to
lead him to successful theory of groups. Also to the proofs by Galois and Abel
ofthe insolvability of equations of degree greater than four.
rdha/[Link] IV

According to [36] [81] he started with a s stematic stud of the

methods of solution of the cubic equation x 3 + nx + p = 0 . This is with the

procedure of Cardooo. Taking x = y - ( ; ) y he transformed this equation into

the sixth degree equation y' + py3 - ( ~: ) ~ 0

Taking , =i this reduces to the quadratic equation

r' + pr - ( ~: ) =0

The quadratic has two roots '\ and '2 = _ (!!.)3 ~

3 '1

Lagrange knew that each equation i ='\ and i =', had three
roots. Thus there are six possible values for y.

They are ~, a~, a2~, ~, a~, a 2~ where a is a complex root of

3
x -1 = 0 or of x 2 + x + 1 = 0

Then he showed that the three distinct roots of original equation

x3 = a 2 V;; + aV;:;

As per [36] and [81], Lagrange next considered the solutions of

quartic equations. Ferrari's method of solving x 4 +nx 2 + px+q =0 was to add

2yx 2 + y2 to each side, rearrange, then he got the equation

 0 of ori .

eq on. Th

1 to only
1
Y, Y7l=- ~~ X_X.;

a degree ~(U3liIO

of m for soh7ing
I: and qualltic~ Lagran~~e til of
equatio of fonnx'" -I =0. If n .
cool of one of them. In . ular if, i

aitl is one of roo a fOT any 1<11

must be satisfied ertain fun tio of orig;inal equation"

functions wbi h take on onl aIue when the roo are pennuR~ :y all n!
possible permutations. H tried to :fin rul for g
such functions and the degree ofthe equation which they 0 I

Lagrange hoped to 01 e the general polyn .at f

degree n. He started with a symmetric fun tion of tb x'" .

It as unchanged under all n! permutations. Then he fin a fun Ii n v~

30
-
which take on 'r' different values under these permutations. Therefore 'v'
would be a root of an equation of degree 'r' with coefficients rational in the
original coefficients. After solving this, he could find a new function 'w',
which takes on, says values under the permutations which leave 'v' unchanged.
Therefore 'w' would satisfy an equation of degree's'. He continued in this way
until the function XI is reached.

He was unable to fmd a general method of determining these

indeterminate functions such that they were of a form, which could be solved,
by known methods. Also, he published a memoir on the approximation of roots
ofpolynomial equations by means of continued fractions in 1767.

4.2 Algebra in the 19th Century

In 1800, Algebra meant the solving of equations. Carl Friedrich

Gauss published the Disquisitiones Arithmeticae in the beginning of the 19th
century. In it he discussed the basics of number theory. Along with proving
the law of quadratic reciprocity he introduced various new concepts that
provided early examples of groups and matrices.

Gauss' study of the solutions of cyclotomic equations in the

Disquisitiones and Augustin-Louis Cauchy's study of permutations in 1815
helped on the problem of solving algebraic equations of degree higher than
four. Niels Henrik Abel fmally proved the impossibility of the solution of a
general equation of degree five or higher in terms of radicals in 1827. Then,
Evariste Galois formed the relationship between algebraic equations and
groups of permutations of the roots. Galois work was not published until 1846.
In 1854 Arthur Cayley gave the earliest definition of an abstract group. The
study of the 'numbers' determined by the solutions of algebraic equations led
to the definition of a field of numbers. Both Leopold Kronecker and Richard
Dedekind gave it. Soon after Heinrich Weber gave the abstract definition of a
field.
re/w/~IV

English mathematicians George Peacock and Augustus De

Morgan tried to axiomatize the basic ideas of algebra. Also, they tried to
determine exactly how much one could generalize the properties of the integers
to other types of quantities. This study led to the discovery of quaternion by
William Rowan Hamilton in 1843.

Modern algebra, the theory of matrices was developed in the mid

-nineteenth century. Determinants had been used as early as the seventeenth
century. But in 1850 James Joseph Sylvester related the term 'matrix' to refer
to a rectangular array of numbers. Then Cayley developed the algebra of
matrices. Cauchy began the study of eigen values early in the century in his
work on quadratic forms.

Fourier (1768-1830)

Jean Baptiste Joseph Fourier is best known for his development

of the Fourier series. Also he was interested in the theory of equations. He
made several contributions to the theory of equations. His most important
work was contained in his paper published in 1789. He revised it in 1820 as
'On the Use of the Theorem of Descartes in Finding the Limits of Roots' . This
c
is an extension of Descartes Rule of SignS:. He gave a new and simple proof of
it. The result is today known as Strurm's Theorem' (10], (77].

Figure 4.6 : Fourier

~/w/tbvIV

th
Let X is an m degree polynomial with positive first term.'
Consider X and its first m derivatives say X' ,X' ,X·.. .in the reverse order,

namely ...X·, X" ,X' ,X. Then he described his result:

If one substitutes the number 'a' into the sequence of

functions and writes the sign + or - of each result, one makes a
sequence of signs which are designate by 'a'; if one then
substitutes a number 'b'; greater than 'a', in the same sequence
of functions, and notes the resulting signs, a second sequence of
signs will be formed which we designate by 'fJ'. This being
done, determine how many times, in the first sequence of signs
, a " there is a change of sign from one term to the next, that is to
say how many times in this sequence two adjacent signs are +
and -. Then fmd how many times there is a change of sign in the
sequence 'f3'. From this method of comparing the two
sequences of signs ' a ' and ' f3 " we will first of all state and then
demonstrate the following consequences.

As per [77], Fourier's main results are

(i) If the number of sign changes in 'a' is equal to the number of sign

changes in ' f3 " then between the interval a and b there is no root.

(ii) There can not be more sign changes in ' fJ ' than those in 'a '

(iii) If' a' has one more change in sign than ' fJ " then there is exactly one

root between 'a' and 'b'.

(iv) If' a' has two more change in sign than ' fJ', then there are two real or

imaginary roots between a and b.

(v) If' a' has three more change in sign than' fJ' then there are one or

three real roots between a and b.

-
Fourier concluded that, there can not be more roots between a
and b than the excess of the number of sign changes in ' a ' over' f3 '.

Ruffini (1765-1822)

Paolo Ruffini was born in 1765. He was a student of Lagrange.

He published several papers to prove that the general equation of fifth degree
or higher cannot be solved by radicals [36], [81].

According to [49],[81] in 1798 in Bologna, he published his first

treatise, Opere matematiche di Paolo Ruffini. It is entitled Teoria generale delle
equazioni in cui si dimostra impossible la soluzione algebraica delle equazioni generale
di grado superire al quarto. This means, general theory of equations, in which it
is proved that the algebraic solution of the general equations of degree greater
than four is impossible.

Niels Henrik Abel (1802-1829) finally gave a complete proof of

the impossibility of such a solution [36], [49].

Ruffini's methods are those of Lagrange. He considered rational

functions of the roots of a general equation of degree n.

As per [81], Ruffini proved that the general quintic equation

couldn't be solved by radicals. His proof is not conclusive. It is based on the
hypothesis that these radicals can all be expressed as rational functions of the
roots.

Gauss (1777-1855)

Carl Friedrich Gauss was born in 1777 in Germany. He received

his doctorate from the University of Helmstadt in 1798. His thesis was
published in 1799 in Latin. It was entitled 'New Demonstration of the Theorem
That Every Rational Integral Algebraic Function in One Variable Can Be
Resolved into Real Factors of First or Second Degree'. Later Gauss modified
this statement as the 'Fundamental Theorem of Algebra' [10], [73], [77], [81].

Figure 4.7 : Gauss

In France, it is known as 'd'Alembert's Theorem. Gauss showed

that all previous attempts of demonstrations were inadequate.

According to [10] in the thesis, he proved that every polynomial

equation f(x)=O has at least one root, whether the coefficients are real or

lffiagmary.

Gauss' proof of the 'Fundamental Theorem of Algebra' is based

in part on geometrical considerations. In 1816 and in 1850, he published new
demonstrations to fmd a proof that is entirely algebraic.

In his first proof, he does not introduce complex numbers. He

proves the fundamental theorem in the following form:

"Every polynomial X with real coefficients can be factored into

linear and quadratic factors" [81].

Every algebraic equation of degree 'm' can be written as

x"' + Ax m- ' + Bx m - 2 + .... + M = 0

Gauss has given four proofs of the 'Fundamental Theorem of

)
Algebra [46].
relw/tWvlV

It is given in [39] and [81] that in Disquisitiones Arithmeticae, Gauss

discussed the solution of cyclotomic equations, equations of the form x n -1 =O.
He gave the complete solution by means of radicals.

According to [36] he also discussed the application of these

solutions to the construction of regular polygons.

Gauss knew the solutions of this equation in the form

Cos 2nk + iSin 2nk for k = 0,1,2, ..., n -1. He wanted to determine these solutions
n n
algebraically. Gauss was convinced that the general equation of degree higher
than four could not be solved by radicals.

4.3 Algebra in Twentieth Century

Twentieth century was a golden age for mathematics. It has been

estimated that more mathematical research has been done in the twentieth
century than in the whole of the previous history of mathematics. The subject
underwent tremendous growth. The crowning achievements of twentieth
century were the solution of three celebrated challenges: Fermat's Last
Theorem, the Four-Colour Problem and the Kepler Conjecture. Mathematics in
the twentieth century gave rise to two strands: pure mathematics and applied
mathematics. The most apparent trends in pure mathematics are those of
towards unity and abstraction. There is inter-relationship between the different
branches of modern pure mathematics. In the nineteenth century it was fairly
obvious whether one was working in algebra, geometry or analysis. But in the
twentieth century, mathematics does not easily divide into compartments.

The most unifying branch of modern mathematics is topology.

Though it was started in the nineteenth century, it has developed in the
twentieth. Now it exerts a major influence on all branches of pure mathematics.
The other major branches of pure mathematics are Abstract Algebra,
Functional Analysis and Graph Theory. The theory of infmite sets plays a
considerable part in both topology and functional analysis.

Algebra has been in connection with most branches of modem

mathematics. The growth of algebraic structure has been a powerful influence
in the twentieth century. Among various structures considered by
mathematicians, Group Theory remains the most fundamental. It is considered
the supreme art of mathematical abstraction. Evariste Galois (1811-1832) was
one of the great pioneers of this abstract theory. Galois launched Group Theory
for equations. On the last night of his life, he wrote a summary of his
discoveries in the theory of equations. Now it is known as Galois Theory. The
ancient problems like the duplication of the cube, the trisection of an angle, the
solution of the cubic and the biquadratic equations and the solution of an
algebraic equation of any degree are found in Galois Theory. His work was
published first in 1846 by Liouville in his Journal de Mathematiques.

The symmetries of an equilateral triangle, the permutation of

three objects and the solution of a cubic equation have isomorphic group
structure. Thus cases of geometry, the arithmetic arrangements and algebra are
covered by the same group. Group Theory is a unifying force of different
branches of mathematics. Another important aspect of Group Theory is its
study of 'transformations'. For example, an algebraic equation is transformed
when 'x' is replaced by 'y'. Also, we can transform a geometric figure by
projecting or stretching it on a different surface. Certain aspects of algebraic
equations of a geometric figure remain unchanged during transformations.
They are known as 'invariants'

According to [81], Galois proved: "If one of the roots of an

equation of arbitrary degree (with rational coefficients) is an immediately
periodic continued fraction, then another root is also a periodic continued
fraction, which one obtains by dividing -1 by the same continued fraction,
written in the inverse order."

Galois presented a first account of his work on the solution of

algebraic equations to the Academie des Sciences de Paris in May 1829. Eight
days later he presented another work on equations of prime degree.
Unfortunately, both papers were lost.

Figure 4.8: Galois

Galois' another theorem states that: "In order that an equation of

prime degree is solvable by radicals, it is necessary and sufficient that, if two of
its roots are known, the others can be expressed rationally". This theorem
reveals that the general equation of degree 5 cannot be solved by radicals.

Srinivasa Rarnanujan's (1887-1920) early experiences m

mathematics included the attempt to fmd a solution to the quintic equation. His
exposure to advanced mathematics came from the text, 'Synopsis of
Elementary Results in Pure Mathematics', by G.S. Carr. His command over
some branches of Analysis and the Theory of Numbers is not imaginable.
According to [36], he discovered the Bernoulli numbers. In 1911, he published
a short paper on this in the 'Journal of the Indian Mathematical Society'. Some
of his results are,

1+2+3+ ...+00 =-1112

-

Figure 4.9 : Srinivasa Ramanujan

He discovered jointly with G.H. Hardy the theorem dealing with

the 'partitions' of a number 'n' is considered to his greatest contribution. A
'partition' of a positive integer 'n' is a way of writing In' as a sum of positive
integers. For example, 5 may be partitioned in seven ways: 5, (4 +1), (3 + 2), (3
+ 1 +1), (2 + 2 + 1), (2 + 1 + 1 + 1), (1 + 1 + 1 + 1 + 1).

If pen) denotes the total number of partitions of 'n', then p(5) = 7.

Similarly p(1) = 1,

p(2) = 2,p(3) = 3,p(4) =5,p(6) = 11.

Computation shows that partition function pen) increases rapidly

with 'n'. For example, p(200) = 3,972,999,029,388.

As per Burton [11], in 1918, [Link] and Ramanujan proved

c,["
that for large 'n', the partition function satisfies the relation p(n):::: e r;:; ,
4nv3

where the constant c = IT(2/3Y/2 . For n = 200, the above formula gives 4.10 12

approximately which is close to the actual value of p(200).

Another example of Ramanujan's ability is his infmite series

representation for n. Computer scientists have exploited his series,

~ha/t4vIV

I
-=-LJ
.J8 ~ (4n)! (1103+26 390n)
4 4' , to calculate the value of 7t
..
to millIons of
1C 9801 n=O (n!) 396 n

decimal digits. Here each successive term in the series adds roughly eight

more correct digits. He discovered 14 other series for 2.. But he did not give
7r

any explanation about their origin. A remarkable one is

~= f(2n)3 42n+3
7! ~ n i 2n 4
+

There is a well known story that Hardy once visited Ramanujan

when he was hospitalized. Hardy arrived in a taxi with number 1729. When
Hardy said the dull number 1729, Ramanujan replied, "No, it is a very
interesting number; it is the smallest number expressible as the sum of two
cubes in two different ways."

That is 1729 = 13 + 12 3 = 93 + 103

Thus Hardy asked for the smallest number that is a sum of two
fourth powers in two different ways. Ramanujan replied that there was no
obvious example and the first such numbers must be very large. (The number
is 635,318,657 = 594 + 1584 = 133 4 + 1344) [11],[35].

According to [83] Ramanujan published a paper, "Note On a Set

of Simultaneous Equation" in the Journal of the Indian Mathematical
Society,IV, 1912,94-96 which is given as follows.

Consider the equations.

-

.........................................

where XI' X 2 , X 3 , ..•.. 'Xn and YI' Y2' Y3' ...•., Y n are 2n unknown

quantities.

Consider the expression,

...........(*)

Expand this in ascending powers of(), we get

............. (**)

By simplifying (*), we get numerator an expression of degree

(n -1) in () and its denominator an expression of the degree n in () I I

............. (***)

Equating the coefficients of like powers ofe, we have

 From these B" B2 , •••• ,Bn can easily be found and smce

Ai'~' ... ,A depend upon these values they also can be found.
n

Splitting (***) into partial fractions, we get

Comparing with (*), we get

Example 4.4 [83]

x+y+z+u+v=2

px + qy + rz + su + tv = 3
-

Where x,y,z,u, v,p,q,r,s,f are the unknowns. Proceeding as in

the general case, we get

x y Z U V
-+ + +--+--=
l-()p I-Oq I-Or I-Os I-Of

2 +3() + 1602 + 3103 + 1030 4 + 2350 5 + 6740 6 +1669()7 + 4526()8 +I159(l +....

2 +() + 3()2 + 2()3 + ()4

Then we can show that, 2 3 4 5 =
1- () - 5() +() + 3() - ()

Splitting this into partial fractions, we get the values of unknowns

as follows:

X=-
-3
p=-I
5
 .J~
~lwldevlV
-
18+vIS 3+-15
y= q=
10 2

18-vIS 3--15
Z= r=
10 2

8+ vIS -15 -1
u=- s=
2vIS 2

8- vIS -15+1
V= t=-
2vIS 2

The major characteristic of twentieth century applied

mathematics has been the diversity of its application. These applications were
restricted previously mainly to physical sciences. In this century, the field has
been steadily widened to non-physical sciences such as Economics, Biology
and Social Sciences. In any branch of human activity possessing some rational
element, a mathematical interpretation can be found. The inventions of
computers have revolutionised the possibilities of mathematical application.