Demonstration for Fermat’s Last Theorem and Beal’s Conjecture

José William Porras Ferreira

[email protected]

Abstract
Fermat’s Last theorem (F LT ), 1637, states that if n is an integer greater than 2, then it is impossible to
find theree natural numbers x, y and z in xn + y n = z n where such equality is met being (x, y) = 1. Beal’s
conjeture (BC), 1993, states that Ax + B y = C z equation, where (A, B, C, x, y, z) ∈ Z+ and (x, y, z) > 2
are different exponents, then (A, B, C) must have a prime factor, for integer solutions, but if (A, B) = 1 and
the exponents (x, y, z) > 2 are different, there are not integer solutions.

The present proof contains three theorems that finally allow us to demonstrate the Beal’s Conjecture,
transforming the equation of Beal’s conjecture into the form of Fermat’s Last Theorem equation. Since there
are no solutions in integer numbers for Fermat’s Last Theorem equation, then the Beal’s conjecture does not
have solution in integer numbers for unequal exponents or with two equal exponents, but all greater than
two, being two of their bases coprimes.

Keywords: Pythagorean Theorem, Fermat’s Last Theorem, Beal Conjecture.

AMS Classification: 11D41

1 Introduction
1.1 Fermat’s Last Theorem(FLT)
Fermat’s last theorem (FLT) or Fermat-Wile’s theorem is one of the most famous theorems in the history of
mathematics [1]-[2]-[3]. During the last three centuries, the challenge before solving FLT, allowed the develop-
ment of several areas such as numerical analysis, modular algebra and some conjectures.
Before the test performed by Andrew Wiles (1995), [4]-[5]-[6], FLT was one of the “most difficult mathematical
problems”.
Using modern notation, Fermat’s last theorem can be stated as follows:
If n is an integer number greater than 2, then it can’t be found three positive integer number x, y and z such
that the equality is met being (x, y) = 1 and:

xn + y n = z n

Some of the mathematicians who achieved partial test were Pierre de Fermat (1667)[7], for n = 4, other al-
ternative test for n = 4 were made later [8]; Leonard Euler (1735) for n = 3, [9]-[10]-[11]. Sophie Germain.
[12], showed special cases as if n and 2n + 1 are both prime, then Fermat’s conjecture for power n, meant that
one of x, y or z would be divisible by n. Germain checked for n < 100 and Legendre extended it to n < 197.
Dirichlet and Legendre, (1823-1825), [13]-[14], went from n = 3 to n = 5. Lame (1840), [15]-[16], did so for
n = 7. Likewise the test has been extended to exponent n = [6, 10, 14].
Andrew Wiles in 1995 [6] finally attained a general proof for all the exponents greater than 2, Wiles could prove
Fermat’s last theorem from the connection, outlined by Frey, and demonstrated by Ken Ribet in 1985 [17]-[18],
that a demonstration of the so-called Taniyama-Shimura conjecture [19]-[20], would directly lead to a demon-
stration of Fermat’s last theorem . In short, the Taniyama-Shimura conjecture states that every elliptic curve
may be uniquely associated with a mathematical object called a modular. If the FLT is false, then there would
be an elliptic curve such that cannot be associated with any modular form, and therefore the Taniyama-Shimura
conjecture solution would be false, i.e, the Taniyam-Shimura conjecture solution would demonstrate the FLT.
The demonstration of the Taniyama-Shimura conjecture was already on a challenge of the most importance,
because that was one of the points of the socalled Langland’s program, whose goal is to unify areas of math-
ematics which apparently have no unrelated. Wiles spent 8 years following the demonstration of Ribet in
complete isolation working on the problem and only relying on his wife, which is a way of working unusual
in mathematics, where it is common to methematicians from around the world to share their ideas often. To
not raise suspicion, Wiles was publishing articles periodically , as any mathematician of any University in the
world would. Their initial study meant the first significant advance in the theory of Galois prior to an attempt
to extend the Iwasawa theory with an inductive argument (1990-1991). When it seemed that it stagnated, he

1
sought other directions. In the summer of 1991, he seeks in Iwasawa theory solutions but also seemed not to
reach the central themes to solve the FLT. During his research, Wiles sought help from some of his colleagues
on new techniques and / or developments, discovering a new one about an Euler system that had recently been
developed by Victor Kolyvagin and Matthias Flach and that fit the inductive part of his proof. Wiles studied
and expanded this approach in his proof and in January 1993 asked his Princeton colleague, Nick Katz, to help
this reasoning. For Wiles at the time, his developments and reasoning fit right, but he wanted someone else to
check it out. Its conclusion at the time was that the techniques used by Wiles seemed to work properly. ([2]
pp 209-232) but had subtle errors that Wiles finally corrected and successfully completed its demonstration in
1995 [21]-[22]-[23].
Because Wiles used more than 100 pages and modern mathematical techniques, is in practice impossible that
this demonstration is the same one that hinted at Fermat. (Fermat had a copy of the “Arithmetical of Dio-
phantus” on whose banks scoring reflections that were emerging him. In one of these margins it enunciated
the theorem and wrote in latin:“Cuius rei demonstrationem mirabilem sane detexi. Hanc marginis
exiguitas non caperet”, whose tranlation is: “I have a truly marvelous demonstration for this fact,but this
margin is too narrow to contain it”. Although Fermat in 1667, proved the case n = 4, using the method of
infinite descent; it is likely that him had deceived to believe that he had a proof for the general case. It can be
even that will have noticed his error further: their marginal notes were for personal use, and therefore Fermat
would have not had to backtrack with his comments.
On October 7, 2013 I was invited to Paris by WASET (World Academy of Science Engineering and Technology)
to give a lecture on the demonstration of the Fermat’s last Theorem, by using mathematical tools that existed
in the 17TH century, which was published by WASET [24]. Here is an even more condensed solution and
was presented at the invitation of the Department of Physics of the Universidad del Valle Cali, Colombia on
December 4, 2014, [25].

1.2 Beal’s Conjecture(BC)

Andy Beal, (1993) [26]-[27]-[28]-[29]-[30] stated that Ax + B y = C z (note that x, y and z are unique exponents)
where (A, B, C, x, y, z) ∈ Z+ and (x, y, z) > 2 are different exponents would have no solution using (A, B, C)
coprime bases. While working on Fermat’s Last Theorem, Andy Beal studied equations with independent
exponents. He worked on several algorithms to generate solution sets, but the nature of the algorithms he
developed required a common factor in the bases. He suspected that using coprime bases might be impossible
and set out to test his hypothesis. With the help of computers and a colleague, Andy Beal tested this for all
variable values up to 99. Many solutions were found, and all had a common factor in the bases, but not with
coprime bases.
During the period after the discovery by Andy Beal in 1993, he tried to check if there was something similar
already, writing to scientific journals and teachers in number theory. One of them Dr. Harold Edwards, from
the Department of Mathematics at New York University, who is well-known for being the author of “Fermat’s
Last Theorem, a genetic Introduction to Algebraic Numerical Theory”, [31] sent the paper to Dr. Earl Taft,
also a mathematician at Rutgers University who received it and sent it to his colleague Jarell Tunnell, an expert
on the Fermat’s last theorem. All of them confirmed that there was no known antecedent of this fact being
unknown before 1993, when Beal discovered and postulated that there could be a possibly finite number of
solutions between integers.1
The solution to the Beal conjecture necessarily requires in its bases three integers coprimes, all of which must
have same integer exponents, and greater than 2, that is to say if there were different exponente in their bases,
can’t be considered as a counter example to Beal’s conjecture, as this example:

2713 + 23 35 733 = 9193

34 293 893 + 73 113 1673 = 27 54 3533

Here also is confirmed that Beal’s Conjecture (BC) not only met with prime common bases, but also with
composite common bases if one of the exponents is less than three and after eliminating the common bases can
have solutions in Z+ and finally, when the common bases from BC are eliminates, that is, if all the exponents are
different or two exponents are equal but all exponents are greater than two, shows that there are not solutions
in Z+ .
The proof of this conjecture is based on Fermat’s Last Theorem demonstrated in this manuscript, exponent
algebraic properties and reduction ad absurdum.

1 summarize from:http://bealconjecture.com

2
2 Theorems
2.1 Theorem 1
.
Let x, y natural numbers, a, b, and c be positive real numbers, which satisfy the equations:

x+y =a
x2 + y 2 = b2
xn + y n = cn
Then it is true c < b < a for all n ∈ Z+ , n > 2.
Demonstration.

Let’s try initially with a > b.

Indeed:

a2 = a · a
= (x + y)(x + y)
= x2 + y 2 + 2xy
> x2 + y 2
> b2

Consequently, a > b.

On the other hand:

bn = b2 · bn−2
p
= (x2 + y 2 )( x2 + y 2 )n−2
p p
= x2 ( x2 + y 2 )n−2 + y 2 ( x2 + y 2 )n−2
> x2 xn−2 + y 2 y n−2
> xn + y n
> cn

Then, b > c, which completes the demonstration that a > b > c.

2.2 Theorem 2: Solution of Fermat’s Last Theorem

.
The equation z n = xn + y n for n > 2, has no solutions in positive integers (Z+ ).

Demonstration.

The method of reduction ad absurdum will be applied to prove this theorem where (x, y) = 1 so there is
minimun solution between the integers.
In fact, they are (x, y, z) coprime positive integers, which satisfy the equation:

z n = xn + y n

for some, n ∈ Z+ , n > 2, and x < y < z.

Let us assume that U 2 = x2 + y 2 , where U is a positive or an irrational integer given that (x, y) = 1.
By Theorem 1, we have that U > z, from which we can say that there exists k ∈ R+ such that U/z = k. Let’s
see this, p
U x2 + y 2
= √
z n
xn + y n
Raising both members to 2n results:
U 2n (x2 + y 2 )n
= = k 2n
z 2n xn + y n

3
 s
U (x2 + y 2 )n
2n
So that z = k, with k =
(xn + y n )2
In accordance with the above, it can be stated that k can not be a positive integer.
In effect, since k > 1, it has the following:

ky < kz, (1)

2 2 2
Then k y <U (2)
2 2 2 2
Therefore k y <y +y (3)
2 2
So (k − 2)y < 0. (4)
√
It follows that 1 < k < Uy < 2. So things, k is rational or irrational number.
In order to complete the demonstration, consider the different cases for U and K mentioned above:
Case 1 U ∈ Z+ . √
U
Suppose that k ∈ Q, where 1 < k = ab 6= y+l < Uy < 2, (a, b) = 1 and l ∈ Z+ , where y < y + l < U

U2
z2 =
k2
x2 + y 2
=
k2
= υ + w2
2

xb yb √
Where υ = ,w= since (x, y) = 1. If a | x then υ 2 ∈ Z+ and a - y, then w2 ∈ Q and z = υ 2 + w2 ∈ / Z+ .
a a √
If a - x then υ 2 ∈ Q and a | y, then w2 ∈ Z+ and z = υ 2 + w2 ∈ / Z+ . If a - x then υ 2 ∈ Q and a - y, then
2
w ∈ Q.

All possible integer solutions of z = (y + l) < U and z n = (y + l)n where l ∈ Z+ and υ 2 , w2 ∈ Q must not be
considered because these type of solutions are not with integers, have not a minimun integer solution and lead
to an absurd.
Let see: √
Suppose that k ∈ Q, where 1 < k = ab = y+l U
< Uy < 2, (a, b) = 1 and l ∈ Z+ , where y < z = y + l < U
Ub U x2 (y+l)2 y 2 (y+l)2
(if a - U , z ∈
/ Z), if a | U then a = (y + l) < U and k = y+l , therefore z 2 = υ 2 + w2 = U2 + U2 =
2 2
(y + l)2 ( x U+y 2
= (y + l) , which means that for the same value z = y + l can be found with any x and y, which
2 )
leads to uncertainty.

To eliminate this uncertainty, let’s take the minimun value of l and verify if there can be a minimun value
of (x, y) that corresponds to an integer solution of z. The primitive Pythagorean triples2 (U, y, x), where
2
−m2
U = y + m, y = x 2m , x > m ∈ Z+ , the minimun value of (x, y) that corresponds to an integer solution of
2
U is with m = 1, then U = y + 1, where y = x 2−1 , for any x = 2s + 1 and s ∈ Z+ . But since z < U , then
z = (y + l) < U = y + 1, therefore 0 < l < 1, can’t be integer and there are no minimun value of z ∈ Z+ , and
because the good order of the integers, requires that to have an integer solution in a equation with integers,
there must be a minimun integer value, and here it has been proven that it does no exist taking solution of
z with fractional numbers, therefore it is an absurd that z ∈ Z+ , under the assumption that k ∈ Q. The
above is obvious, since the only possible integer solution with (x, y) ∈ Z+ is with the primitive Pythagorean
triples (U, x, y) and there are not solutions with integer numbers less that the same U ∈ Z+ solution in right tri-
angles, similar and smaller to the right triangle (U, x, y), because it would go against the Pythagorean’s theorem.

On the other hand, if now k ∈ I, since U/z = k, it implies trivially that z ∈ I, because U ∈ Z+ , contradicting
once again the fact that z ∈ Z+ .

Case 2
Trivially z ∈ I, if k ∈ Q, contradicting the fact that z ∈ Z+ .
U
On the order hand, if (k 6= y+l ) ∈ I, where l ∈ Z+ we affirm that z ∈ I. In effect, 1 ≤ x and y ≥ 2, with which
√ U
√ U
(U ∈ I) ≥ 5. Since 1 < ((k 6= y+l )I < Uy < 2) and z = , it would always be the quotient of two different
k
/ Z+ . √
irrationals, therefore z ∈
U
If (1 < (k = y+l ∈ I) < Uy < 2), cannot be considered, because the solution of z and z n do not meet with
2 https://www.researchgate.net/publication/322255294PythagoreanTriples

4
 U
integer numbers and it will be a similar to Case 1, when k 6= y+l ∈ Q. This is obvious, since the only possible
integer solution is with the primitive Pythagorean triples, and can’t be one integer solution with small similar
rigth triangle to rigth triangle (U, x, y).

In this way Fermat’s Last Theorem is proved by reduction ad absurdum and maybe we have followed the
same path of Fermat when he said: “Cuius rei demostration mirabilem sane detexi. Hanc marginis
exiguitas non caperet”, and it can be concluded that in any rigth triangle (U, x, y) with (x, y) = 1 any integer
z in its hypotenuse less than U , must be z n 6= xn + y n for n ≥ 2.

Corollary 1. In equation z n = xn + y n for n > 2 if x, y or z they are irrational in the form z = a(f /d)
or x = a(h/g) or y = c(j/i) , with (a, b, c) ∈ Z+ , and [(d, f ), (g, h), (i, j)] ∈ Z+ , where (d, f ) = 1, (h, g) = 1 and
n n n
(f, i) = 1, where (f > d, h > g, j > i) andp(d, g, i) > 2 then the equation √ z = x + y for n > 2 has no solution
U
in the integer numbers, since z < U = x2 + y 2 , and 1 < k < y < 2, would be a case similar to what has
√
been demonstrated in Theorem 2, that is, if U is an integer or an irrational greater than 2, and 1 < k < 2,
then z ∈/ Z+ , and if z ∈ I, by definition z ∈
/ Z+ .

2.3 Theorem 3. Solution of Beal’s conjecture

Theorem In equation Ax + B y = C z , where (A, B, C, x, y, z) ∈ Z+ and (x 6= y 6= z) > 2 then (A, B, C) must
have a prime factor pn where p is a prime and n ≥ 1, to have integer solutions with the exponents (x, y, z)
greater than 2.
In other words: Ax = pn ar , B y = pn bs , and C z = pn ct If pn is eliminated and ar + bs = ct with (a, b) = 1 and
exponents (r, s, t) are different and greater than 2, there are no solutions in Z+ .
Demostration through Theorem 2 (FLT) and reduction ad absurdum.

1. In Equation Ax + B y = C z , where (A, B, C, x, y, z) ∈ Z+ and (x 6= y 6= z) > 2 then (A, B, C) must have

a prime factor pn where p is a prime and n ≥ 1, to have integer solutions with the exponents (x, y, z)
greater than 2.
In other words: Ax = pn ar , B y = pn bs , and C z = pn ct If pn is eliminated and ar + bs = ct with (a, b) = 1
and exponents (r, s, t) are different and greater than 2, there are no solutions in Z+ .

2. Let proof with a numerical examples when ar + bs = ct with (a, b) = 1 and the exponents (r, s, t) are
different and one of the exponents is less than 2, there are solutions in Z+ .
The equation Ax + B y = C z where Ax = 36 , B y = 183 , C z = 38 have solution in Z+ , but after eliminating
the prime factor pn = 36 , the equation remains ar + bs = ct with (a, b) = 1, but one of the exponent is
less than 2. See table 1.
Table 1. Example of a solution of the equation Ax + B y = C z with (x 6= y 6= z) > 2, where after
eliminating the prime factor pn , the equation ar + bs = ct has a solution in the integers but one the
exponents r, s or t is less than 3.

Ax By Cz prime factor ar bs ct
36 183 38 pn = 36 1 23 32

Beal’s Conjecture also compiles when the common factor is a composite number, if after the elimination
of the composite factor one of the exponents is less than 3 can have solutions in Z+ . For example, the
Equation a2 + b4 = c3 has integer solutions and can be converted to Beal’s Conejecture, with generating
numbers (a, b, c)[35] of the form3

a = (3m4 + 4n4 )(9m8 − 408m4 n4 + 16n8 )

b = 6mn(3m4 − 4n4 )
c = 9m8 + 168m4 n4 + 16n8

This form has infinite solutions in Z+ , (see Table 2)

Table 2. Some examples with (m = 1, n = 1) and (m = 3, n = 2) where a12k for k ≥ 1 can be a common
factor of Ax , B y and C z with Ax = a2+12k , B y = (a3k b)4 , C z = (a4k c)3 and then (A, B, C, x, y, z) ∈ Z+ .
3 Rafael Parra Machio. ECUACIONES DIOFÁNTICAS, pp 22. Web: http://hojamat.es/parra/diofanticas.pdf

5
 A14 B4 C3 Composite a2 b4 c3
numbers
268114 (26813 · 6)4 (26814 · 193)3 268112 26812 64 1933
14294626114 (1429462613 · 6444)4 (1429462614 · 280873)3 14294626112 1429462612 64444 2808733

To clarify, Figure shows how it is possible to go from Beal’s Conjecture Ax + B y = C z to ar + bs = ct

equation or from ar + bs = ct to Ax + B y = C z with solutions in Z+ − 0 if one of the exponents (r, s, t) is
less than 3. Fig. 1. How is it possible to go from BC, Ax + B y = C z , to the ar + bs = ct equation with
solutions in ar + bs = ct if one of the exponents (r, s, t) is less than 3.

3. As we already know that there exist integer solutions in Ax + B y = C z with common factors and after
eliminating them the resulting equation ar + bs = ct with (a, b, c) ∈ Z+ and (a, b) = 1, has at leas one of
the exponents (r, s, t) less than 3, we must continue the demonstration assuming: integers solution in the
equation ar + bs = ct with, (a, b) = 1 and (r 6= s 6= t) > 2

Figure 1: How is it possible to go from BC to the ar + bs = ct equation whit solutions in Z+ or from ar + bs = ct

to BC with solutions in Z+ if one of the exponents (r, s, t) is less than 3

4. For this part of the demonstration the method of reduction ad absurdum will be applied.

Indeed they are (a, b, c) ∈ Z+ with (a, b) = 1, which satisfy the equation:

ar + bs = ct

For some (r 6= s 6= t) > 2 =⇒ (a, b, c) ∈ Z+ .

assuming that r is the smaller exponent and greater than 2 then:

s t
(ar + bs = ct ) ⇔ (ar + (b r )r = (c r )r ) ⇒ (exponent property)
s t
There are two cases to be analyzed for b r and c r
Case 1: r | s and r | t or (b = bkr kr +
1 and r - s) or (c = c1 and r - t), (b1 , c1 , k) ∈ Z then:

bs/r = u, ct/r = v −→ (u, v) ∈ Z+

and
(ar + bs = ct ) ⇔ (ar + ur = v r ) ⇒ r > 2
Which leads to an absurdity, since according to Theorem 2, this equation has no solution in the integer numbers.

Case 2: r - s or r - t and (b 6= bkr kr +

1 or c 6= c1 , (b1 , c1 , k) ∈ Z ); therefore b
s/r
or ct/r or both must not
be fractional numbers:
Proof :

6
 d f
Assuming u = bs/r = e or v = ct/r = g or both are fractional numbers (d, e, f, g) ∈ Z+ ⇒ (d, e) = 1 and
(f, g) = 1 :
 r
d
ur = = bs
e
or  r
r f
v = = ct
g
then u or v or both can’t be fractional numbers. It is impossible that an integer be a fractional number; there-
fore: u = bs/r or v = ct/r or both must be irrational numbers.
For this case: If any u = bs/r or v = ct/r or both bases are irrational numbers, then the equation:

(ar + bs = ct ) ⇔ ar + (bs/r )r = (ct/r )r ] ⇔ [(ar + ur = v r )]

is not resolving in Z+ . (Corollary 1 of Theorem 2).
The demonstration scheme is the same for any smaller exponents and it is not necessary to expand the demon-
stration for each of the others exponents (s, t) if one of them is the smallest, and there are not solutions in Z+
when (r 6= s 6= t) > 2.

Corollary 1: The equation ar + bs = ct , has no solution in the integers, if all the exponents (r, s, t) > 2
and two of them are equal.
It is obvious that if there is no solution with exponents (r 6= s 6= t) > 2, neither is there if two of its exponents
are equal, vastly follow the same procedure from point 4 of Theorem 3,i.e if the exponents r = s or r = t or
s = t, there are no solutions in integers.
In conclusion it is absurd to consider that the equation ar + bs = ct , (a, b) = 1 and (r 6= s 6= t) > 2 or two of
the exponents are equal, can have solution in integer numbers, being demonstrated in all its forms the Beal’s
Conjecture.

3 Conclusion
A simple demo using 17TH-century mathematical tools such as stated by Fermat in the margin of the “Arit-
methical of Diophantus” writting his notes can exist. Taking into account that Fermat was who introduced the
principle of infinite descent, which was used on his show for n = 4k in the FLT, it wouldn‘t be strange that Fer-
mat did think that he had a general solution of his last theorem or used a similar procedure to the described here.

The Theorem doesn’t have a major application, but to be considered the most difficult problem in the world,
for 360 years, the search for its solution, allowed the advancement of mathematical science during the last four
centuries and where great mathematicians such as Euler (1707-1783), Lagrange (1736-1813), Germain (1776-
1831), Gauss (1777-1855), Cauchy (1789-1857), Lamé (1795-1870), Dirichlet (1805-1859), Liouville (1809-1882),
Kumer (1810-1893), Vaudiver (1882-1973), Taniyama (1927-1958), Shimura (1930-), Wiles (1953-) and many
other mathematicians who contributed to the advancement of the sciences and number theory in search of their
show.

Using the demonstration of Fermat‘s Last Theorem as a base, mathematical transformations with exponent prop-
erties and reduction ad absurdum, it was possible to confirm Beal’s Conjecture i.e. when equation Ax +B y = C z ,
where (A, B, C, x, y, z) ∈ Z+ ⇒ (x, y, z) > 2, has common bases, (prime or composite numbers) there are integer
solutions if after eliminating the common bases, one of the exponents is less than three. If the resulting equation
have all the exponents (r, s, t) greater than two (all different, or two exponent equal), with ((a, b) = 1), confirms
that the equation:

(ar + bs = ct ) and (r, s, t) > 2 has not solution in Z+ , solving Beal’s conjecture.

References
[1] Cox D.A.Introduction to Fermat’s last theorem. Amer. Math. Monthly 101 (1), pp 3-14. 1994.
[2] Singh S.Fermat’s Last Theorem. London. ISBN 1-85702-521-0. 1997.
[3] Ribenboim P.The history of Fermat’s last theorem (Portuguese). Bol. Soc. Paran. Mat. (2) 5 (1), pp
14-32. 1984.

7
 [4] Van der Poorten A.Remark on Fermat’s Last Theorem, Austral.Math. Soc. Paran. Bol. Soc. Paran.
Gaz.21 (5), pp 150-159. 1994.
[5] De Castro Korgi R.The proof of Fermat’s Last Theorem has been announced in Cambridge, England
(Spanish), Lect. Mat. 14, pp 1-3. 1993.

[6] Wiles A.Modular elliptic curves and Fermat’s Last Theorem (PDF). Annals of Mathematics
141(3):pp.443-531. Doi:10.2307/211855,May 1995.
[7] Heath-Brown D.RThe first case of Fermat’s Last Theorem. Math. Intelligencer7 (4),pp.40-47. 1985.
[8] Cox D.A.Introduction to Fermat’s last theorem. Amer. Math. Monthly 101 (1), pp 44-45. 1994.

[9] Barlow P.An elementary Investigation of Theory of Numbers. St. Paul’s Church-Yard, London: J. John-
son. pp. 144-145. 1811.
[10] Gautschi, W.Leonard Euler: His Life, the Man, and His Work. SIAM Review 50 (1): pp 3-33. 2008.
[11] Macys J.-j.On Euler’s hypothetical proof. Mathematical Notes 82 (3-4). pp 352-356. 2007.

[12] Del Centina, A.Unpublished manuscripts of Sophie Germain and a revaluation of the work on Fermat’s
Last Theorem. Archive for History of Exact Sciences 62.4 (2008): pp 349-392. Web. September 2009.
[13] Carmichael, R.-D.The Theory of numbers and Diophantine Analysis. Dover N.Y. 1959..
[14] Legendre, A.M.Research on some analysis of unknown objects, particularly on Fermat’s theorem (in
French) Mem. Acad. Roy. Sci. Institut France 6:1-60.1823.
[15] Lamé, G.Memory on Fermat’s Last Theorem (in French). C. R. Acad. Sci. Paris 9. pp.45-46.1839.
[16] Lamé, G.Memory on the undeternined analysis demonstrating that the equation x7 + y 7 = z 7 is impossible
in integer numbers (in French).
J. Math. Pures
Appl. 5: pp.195-211.1840.
[17] Ribet, K.A. Galois representation and modular forms. Bulletin AMS 32, pp 375-402. 1995.
[18] Ribet, K.A.“From the Taniyama-Shimura Conjecture to Fermat’s Last Theorem.” Ann. Fac. Sci. Toulouse
Math. 11, pp. 116-139,1990.

[19] Serge L.Some History of the Shimura-Taniyama Conjecture. Notices of the AMS. Volume 42, Number
11, pp 301-1307. November 1995.
[20] Gerhard F.Links between stable elliptic curve and certain Diophantine equations. Annales Universitatis
saraviensis. Series Mathematicae 1 (1) iv+40. ISSN 0933-8268, MR 853387. 1986eonard Euler: His Life,
the Man, and His Work. SIAM Review 50 (1): pp 3-33. 2008.
[21] Gerd F.The proof of Fermat’s last Theorem R. Taylor and A. Wiles. Notices of the AMS. Volume 42
Number 7, pp 743-746 July 1995.
[22] Darmon H., Deamond F. and Taylor R. Fermat’s Last Theorem. Current Developments in Math.
International Press. Cambridge MA, pp 1-107. 1995.

[23] Kleiner I.From Fermat’s to Wiles: Fermat’s Last Theorem Becomes a Theorem. Elem. Math. 55, pp.
19-37 doi: 10.1007/PL00000079.2000.
[24] Porras Ferreria J.W.Fermat’s Last Theorem A Simple Demonstration. In-
ternational Journal of Mathematical Science. Vol 7 No.10, pp 51-60. 21013.
Avalaible online: http://www.waset.org/Publications/?path=Publicationsp=82 or
https://www.researchgate.net/publication/284899338 Fermat27s Last Theorem a Simple Demonstration.
[25] Avalaible online: http://coloquiofisica.correounivalle.edu.co/
programacion/periodo-2014-ii-1.
[26] Goldfeld D.Beyond the Last Theorem. Math: Horizons volume 34, pp 26-31.MSeptember 1996.

[27] Breen M. and Emerson A.Beal Conjecture Prize Increased to 1 million America Mathematical Society.
Bull. Jun 6,2013.

8
[28] Abramson A.Billionary offers 1 million to solve Math. Problem. abcNews. Jun 6,2013.
[29] Waldschmidt M.On the abc Conjecture and some its consequences. 6th World conference on 21th Century
Mathematics. Abdus Salan School of Mathematical Sciences (ASSMS) Lahore (Pakistan). Pp 1.79. March
7, 2013.

[30] Mauldin R.D.A Generalization of Fermat’s Last Theorem: The Beal’s Conjecture and Prize Problem.
Notices of the AMS, volume 44, number 11. pp 1436-1439. December 1997.
[31] Harold M. Edward’s.A Genetic Introduction to Algebraic Number Theory. Graduate Texts in Mathetics
50. New York Springer. ISBN 0-387-95002-8. 2000.

[32] Eckert, Ernest. Primitive Pythagorean triples, The College Mathematics Journal, Mathematical Asso-
ciation of America, 23(5): 413-7, 1992. Doi: 10.2307/2686417, JSTOR 2686417.
[33] B. Berggren,“Pytagoreiska trianglar” (in Swedish), (English: Pythagoran triangles), Elementa: Tid-
skrift for elementar matematik, fysok och kemi 17 (1934), pp 129-139- (English: Elemental: Journal of
Elementary Mathematics, Physics and Chemistry 17 (1934), pp 129-139.

[34] Darmon H. and A. Granville. On the equations z m = F (x, y) and Axp + By q = Cz r

Bull, London, Math. Soc. Volume 27, pp 513-543. 1995.

Caption List
Fig. 1. How is it possible to go from BC to the ar + bs = ct equation whit solutions in Z+ or from ar + bs = ct
to BC with solutions in Z+ if one of the exponents (r, s, t) is less than 3.
Table. 1. Example of a solution of the equation Ax + B y = C z with (r 6= s 6= t) > 2, where after eliminating
the prime factor pn , the equation ar + bs = ct has a solution in the integers but one of the exponents r, s or t
is less than 3.
Table. 2. Some examples with (m = 1, n = 1) and (m = 3, n = 2), where a12k for k ≥ 1 can be a common
factor of Ax , B y and C z with Ax = a2+12k , B y = (a3k b)4 , C z = (a4k c)3 and then (A, B, C, x, y, z) ∈ Z+ .

Acknowledgments To all my professors from the Escuela de Cadetes in Cartagena, Colombia and the U.S.
Naval Postgraduate School, in Monterrey, California, USA. Special thanks to PhD Rafael Galeano and MsC
Pedro P. Ortega Palencia, Professors of Department of Mathematics of University of Cartagena, Colombia,
and MsC William Caballero Professor of Mathematics of the Escuela Naval Almirante Padilla in Cartagena,
Colombia for their valuable comments to the drafts of this manuscript.

Conflict of interest The authors certify that they have no affiliations with or involvement in any orga-
nization or entity with any financial interest (such as honoraria; educational grants; participation in speakers
membership, employment, consultancies, stock ownership, or ther equity interest; and expert testimony or
patent arrangements),or non (such as personal or professional relationships, affiliations, knowledge or beliefs)
in the subject matter or materials discussed in this manuscript. Potencial conflicts of interest related to indi-
vidual author’s commitments. Potential conflicts of interest related to commitments of editors, journal staff, or
reviewers.