Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Working with the famous Baker-Davenport system of simultaneous Pell equations \begin{align} 3x^2-2 &= y^2, & 8x^2-7 &= z^2, \qquad(\star) \end{align} I am left, after a series of substitutions and reductions, with the equation $$ u(u+1)\left(96u(u+1)+11\right) = v(v+1). \qquad(\dagger) $$ The point is now to prove that the only solutions with $u \ge 0$ are $u=0$ and $u=5$, yielding, respectively, the two known (and only) solutions $x=1$ and $x=11$ in ($\star$).

Are there any ways of attacking the formulation ($\dagger$), different from those which would naturally be applied to ($\star$), which might lead to a relatively straight-forward proof?


EDIT: For example, by unique factorization of integers, we can write $(v,v+1)=(abc,rst)$ for postive integers $a,b,c,r,s,t$ with $gcd(abc,rst)=gcd(v,v+1)=1$. Now, with appropriate permutation, we can write \begin{align} u &= ar, & u+1 &= bs, & 96u(u+1)+11=ct. \end{align} But all methods and manipulations I've tried from that point seem to lead to dead ends.

share|improve this question
Noting for documentation sake that OP posted, and then deleted, another question about these equation recently, mathoverflow.net/questions/143301/… –  Gerry Myerson Sep 29 '13 at 5:36
@GerryMyerson: I found soon after I posted that the result was trivial, so I didn't think it worthy of remaining on MO. I did not know (1) that deleted posts are still visible, nor (2) that such documentation was required or expected. As you can tell from my reputation points and/or profile, I'm fairly new to MO and MSE — are these rules and expectations listed somewhere? Sorry, Kieren. –  Kieren MacMillan Sep 29 '13 at 11:22
Deleted posts are visible to those with enough points. I don't know that documentation is required, but speaking for myself, I like people posting here to put all their cards on the table. –  Gerry Myerson Sep 29 '13 at 12:18
@GerryMyerson: I'm certainly not trying to hide anything. I just didn't think that deleted observation was worth MO bandwidth, once I saw how trivial it was. –  Kieren MacMillan Sep 29 '13 at 14:17
add comment

1 Answer

Sorry, it's probably no easier: both are looking for integer points on a genus-$1$ curve. In fact it's not hard to get from Baker-Davenport to your equation: multiplying the two equations in $(\star)$ yields $(3x^2-2)(8x^2-7) = (yz)^2$, and both $x$ and $yz$ must be odd, so we can write $(x,yz) = (2u+1,2v+1)$, and then subtracting $1$ and dividing by $4$ yields $(\dagger)$.

share|improve this answer
Yes, that's my "series of substitutions and reductions". I was (am) hoping that there would be (e.g.) a factorization approach applicable to ($\dagger$) that isn't applicable to ($\star$) directly. –  Kieren MacMillan Sep 29 '13 at 11:25
[See edit of original post for more thoughts.] –  Kieren MacMillan Sep 29 '13 at 11:36
add comment

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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