Homework 1

Camilo Escobar Sarmiento

1. (a) Note that (1, 1) is a point on the conic x2 + y 2 = 2. Also, note that any line that goes through point (1, 1) has the

form y = t(x − 1) + 1. Now, we compute the intersection of one such line with the conic, we have the following:

((t(x − 1) + 1))2 + x2 − 2 = 0

t2 (x − 1)2 + 2t(x − 1) + 1 + x2 − 2 = 0

t2 (x − 1)2 + 2t(x − 1) + x2 − 1 = 0

t2 (x − 1)2 + 2t(x − 1) + (x − 1)(x + 1) = 0

(x − 1)(t2 x − t2 + 2t + x + 1) = 0

We have that t2 x − t2 + 2t + x + 1 = 0, therefore all rational points that lie on x2 + y 2 = 2 have coordinates
t2 − 2t − 1 t2 − 2t − 1
x= 2
and y = t( − 1) + 1 where t ∈ Q
t +1 t2 + 1

(b) I am going to prove that there are no rational points that lie on x2 + y 2 = 6. Note that this is equivalent to

proving that there are no integer solutions to x2 + y 2 = 6z 2 . Now, suppose that there exist (a, b, c) ∈ Z3 such

that a2 + b2 = 6z 2 and gcd(a, b, c) = 1 (If gcd(a, b, c) 6= 1 we can divide the equation by their common factor and

we would find another solution). We consider this equation mod 3 and given that the only squares mod (3) are 0

and 1 we conclude that a ≡ b ≡ 0 mod 3. Therefore 3 | a2 and 3 | b2 but this implies that 3 | a and 3 | b, which

implies in turn that 9 | a and that 9 | b, this in turn implies that 3 | c but this contradicts our assumption that

gcd(a, b, c) = 1, so we can conclude that there are no rational solutions to x2 + y 2 = 6.

(c) I will show that there are no rational solutions to 3x2 + 5y 2 = 4. Note that it suffices to show that there are no

integer solutions to the equation 3x2 + 5y 2 = 4z 2 . suppose that there exist (a, b, c) ∈ Z3 such that 3a2 + 5b2 = 4c2

and gcd(a, b, c) = 1 (If gcd(a, b, c) 6= 1 we can divide the equation by their common factor and we would find

another solution). We know that the only squares mod 3 are 0 and 1, therefore 3a2 ≡ 5b2 ≡ 4c2 ≡ 0 mod 3, which

implies that 3 | b2 and 3 | c2 which in turn implies that 3 | b and that 3 | c. But this means that 9 | b2 and that

9 | c2 , forcing 3 | a2 and 3 | a, but this contradicts our assumption that gcd(a, b, c) = 1, so we conclude that there

are no rational solutions to 3x2 + 5y 2 = 4z 2 .

in order to find another point Q = (x0 , y0 ) with rational coordinates in E(Q). We have that 2yy 0 = 3x2 which implies
27 y−5 27 27 31
that the derivative at point R is , therefore the equation of the line tangent to R is = or y = ( )x − ,
10 x−3 10 10 10
plugging this into E we obtain the following:
 2
27 31
x− = x3 − 2
10 10
272 2
x3 − x + f (x) = 0
100
Where f is a polynomial of degree 1. We know that x = 3 is a root of this cubic equation of multiplicity two, therefore,
272 129 27 129 31 383
by Vieta’s formulas we have that −6 = = x0 and y0 = · − 10 = . We have found two points in
100 100  10 100
 1000 
129 383 129 −383
E(Q) with rational, non-integer coordinates, namely Q = , and P = ,
100 1000 100 1000

3. (a) First, suppose that P 6= Q and let us define P ∗Q as the third point (different from P and Q) of intersection between

the line that passes through P and Q and our curve E(K). We compute the coordinates of P ∗ Q = (x3 , y3 ). We
y2 −y1 y2 −y1 y2 −y1
have that the equation of the line that passes through P and Q is y = x2 −x1 x + y1 − x2 −x1 . Let λ = x2 −x1 and

b = y1 − λx1 , we plug in y into the non-homogeneous equation for E and we have the following:

(λx + b)3 + x3 + 1 = 0

(λ3 + 1)x3 + 3λ2 bx2 + f (x) = 0

Where f is a polynomyal of degree 1. Now, using Vieta’s formulas we have that −(3λ2 b) − x1 − x2 = x3 and

y3 = λ(−3λ2 b − x1 − x2 ) + b. Now, note that the line that passes through P ∗ Q and P∞ is parallel to the line

y = −x therefore we define P + Q as the reflection of the point P ∗ Q with respect to the line y = x and we get

that P + Q = (y3 , x3 )

Now, if P = Q we compute the tangent line through P and find the point of intersection with E and we’ll name
 2
that point P ∗ P = (x0 , y0 ). Note that 3y 2 y 0 + 3x2 = 0 therefore the slope of the tangent line at P is − xy11 and
 2    2  
the equation of the tangent line is y = − xy11 x + xy11 x1 + y1 , let λ = − xy11 and b = xy11 x1 + y1 . Now we

plug y into E and we find the following:

(λx + b)3 + x3 + 1 = 0

(λ3 + 1)x3 + 3λ2 bx2 + f (x) = 0

Where f is a polynomyal of degree 1. Now, since x1 is a root of multiplicity two of the cubic equation, by Vieta’s

formulas we have that −3λ2 b − 2x1 = x0 and y0 = λ(−3λ2 b − 2x1 ) + b. Now, note that the line that passes through

P ∗ P and P∞ is parallel to the line y = −x therefore we define P + P as the reflection of the point P ∗ P with

respect to the line y = x and we get that P + P = (y0 , x0 )

(b) We want to find the Weierstrass normal form of the elliptic curve E : y 3 + x3 + 1 = 0. Consider the change of

variables x = u + v and y = u − v so our equation turns into

(u + v)3 + (u − v)3 + 1 = 0
 2u3 + 6uv 2 + 1 = 0

−6 36v
Dividing our equation by u3 and using the change of variables X1 = u and y1 = u we obtaing the following:
 v 2 −1
6 = −2
w u3
y12 x1
= 3 −2
63 6

y12 = x31 − 432

Therefore the Weierstrass normal form of E is y 2 = x3 − 432

4. (a) Attachment.

x1
(b) We define φ : Cns (K) → K such that φ(P∞ ) = 0 and φ(x1 , y1 ) = y1 .