Does there exist infinitely many coprime pairs of integers x,d such that x, x+d, x+2d, x+4d are all square numbers?

One example would be 49,169,289,529. This is the only example I have found so far and I believe it may be the only one.


The answer is yes. Here is another example, $$ x=2021231^2, \qquad d=82153503191760. $$ Then $$ x+d=9286489^2, \qquad x+2d=12976609^2, \qquad x+4d=18240049^2. $$

Basically, write $x=X^2$, $x+d=Y^2$, $x+2d=Z^2$, and $x+4d=W^2$. Eliminating $x$ and $d$ reduces to the two equations: $$ X^2-2Y^2+Z^2=0, \qquad 3X^2-4Y^2+W^2=0. $$ This pair of equations defines a smooth curve $C$ of genus one in $\mathbb{P}^3$, which we can put into Weierstrass form by sending the point $(1,1,-1,1)$ (say) to the point at infinity. The elliptic curve you obtain is $$ y^2 = x^3 - x^2 - 9x + 9 $$ with Cremona reference 192A, and Mordell--Weil rank $1$. The point $(5,8)$ has infinite order, and corresponds to your solution (up to changes in sign). Taking multiples of this point and mapping back to $C$ gives infinitely many coprime pairs $x$, $d$ such that $x$, $x+d$, $x+2d$, $x+4d$ are squares.

| cite | improve this answer | |
  • 1
    $\begingroup$ Maybe explicitly state there are infinitely many solutions? $\endgroup$ – joro Nov 21 '13 at 14:30
  • $\begingroup$ Not sure if this should be another question but is there a way to generate pairs sequences of rationals of this form with the same common difference? $\endgroup$ – David Cushing Nov 21 '13 at 14:34
  • $\begingroup$ I guess the letters in the two equations for $C$ should be capital. And there may be a typo in the first equation ? $\endgroup$ – Chris Wuthrich Nov 21 '13 at 15:15
  • $\begingroup$ I still think you meant $X^2$ instead of $Y^2$ in the first equation ... (sorry to be picky). $\endgroup$ – Chris Wuthrich Nov 21 '13 at 16:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.