4
$\begingroup$

This is related to How many sequences of rational squares are there, all of whose differences are also rational squares?

Are there infinite sequences $a_n$ of rational cubes whose first differences are positive squares?

Fixing $a_n$ leads to finding points on the elliptic curve $y^2=x^3-a_n^3$. If it is of positive rank there will be infinitely many choices for $a_{n+1}$.

Is there an explicit construction that avoids finding points on curves?

What about infinite sequence of integer cubes -- in this case fixing $a_n$ leads to finding integral points on elliptic curves and these are finite.

A possible generalization might be:

Are there infinite sequences of numbers of type $X$ (e.g. triangular, Fibonacci) whose first differences are positive squares?

I am mainly interested in explicit constructions like the case for $X=\square$.

$\endgroup$
5
  • 1
    $\begingroup$ Careful with the generalization - there is an infinite sequence of numbers of the form $n(n+1)(2n+1)/6$ whose first differences are positive squares. $\endgroup$ Dec 2, 2011 at 10:12
  • $\begingroup$ First differences of Fibonacci numbers are Fibonacci numbers, and J.H.E. Cohn has proved earlier on that the only square Fibonacci numbers are 1 and 144. Bugeaud-Mignotte-Siksek generalized this recently by proving that the only perfect powers in the Fibonacci sequence are 1, 8 and 144. $\endgroup$ Dec 5, 2011 at 12:36
  • $\begingroup$ Sorry, I misread your last question, so my comment doesn't seem really helpful. $\endgroup$ Dec 5, 2011 at 12:37
  • 1
    $\begingroup$ François, they need not be consecutive, e.g. $F_{13}-F_{11}=2^{4} \cdot 3^{2}$ $\endgroup$
    – joro
    Dec 5, 2011 at 12:54
  • $\begingroup$ For Fibonacci numbers, see mathoverflow.net/questions/84797 $\endgroup$ Dec 8, 2013 at 19:02

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.