Let $q$ and $p$ be coprime. And without loss of generality, as $p$ and $q$ are interchangeable, let $p>q$, $p=q+d$.

If $p$ and $q$ are coprime, the fraction cannot be simplified. Therefore, we can rewrite $p/q$ as $(q+d)/q$, and we obtain $1+d/q$. As the fraction for $p/q$ cannot be simplified, $d/q$ can also not be simplified, therefore $d$ is also prime to $q$. [We can do same arguement for $p$ too]. Is this correct?

  • 1
    $\begingroup$ Yes but perhaps you should explain why: If $d/q$ $could $ be simplified to $d'/q'$ with $|q'|<|q|$, then $p/q=(q'+d')/q' $ is a simplification of $p/q$.... In general if $p,q$ are co-prime then $p, \pm (pn+q)$ are co-prime for any $n\in \Bbb Z.$ $\endgroup$ – DanielWainfleet Aug 6 '18 at 20:27

Yes, if $d$ and $q$ had a common factor, it would also be a factor of $d+q=p$. This is the heart of the Euclidean algorithm for greatest common divisor.

  • $\begingroup$ Alright, thanks :) $\endgroup$ – Dean Yang Aug 6 '18 at 18:20

Of course it is. If $q$ and $d$ have a common factor $f$:

$q = Qf$
$d = Df$


$p = q + d = (Q +D)f$

and more generally

$kq\pm md = (kQ \pm mD)f$

will have that factor, too.


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.