Let $f: U \subset \mathbb{C}$ be analytic and $\gamma_1, \gamma_2$ be two arbitrary closed paths. Prove that:

$ \int_{\gamma} f(z) \ dz = 0$ for any closed path $ \iff\int_{\gamma_1} f(z) \ dz = \int_{\gamma_2} f(z) \ dz $.

It's easy to see the $\implies$ direction, but I'm having a harder time with the other.

  • $\begingroup$ Take the $\gamma_1$ and $\gamma_2$ from the right side. Can you see that $\gamma_1 - \gamma_2$ forms a closed path? $\endgroup$ – Kezer Sep 14 '18 at 16:44
  • $\begingroup$ @Kezer, yes, but can any closed path be written like that? $\endgroup$ – Matheus Andrade Sep 14 '18 at 16:45
  • 1
    $\begingroup$ I'm sorry, I misread, I didn't see that $\gamma_1$ and $\gamma_2$ are arbitrary closed paths. I assume your $U$ is supposed to be open? Then, choose some circle with its disc fully contained in $U$ - Cauchy makes the contour integral $0$. It's a closed path, so all integrals for closed paths equal $0$. $\endgroup$ – Kezer Sep 14 '18 at 16:54
  • $\begingroup$ Thanks a lot, @Kezer! $\endgroup$ – Matheus Andrade Sep 14 '18 at 19:45

I am assuming that the question actually is: to prove that $\displaystyle\int_\gamma f(z)\,\mathrm dz=0$ for any closed path if and only if for any two paths $\gamma_1$ and $\gamma_2$ with the same initial points and the same final points, $\displaystyle\int_{\gamma_1}f(z)\,\mathrm dz=\int_{\gamma_2}f(z)\,\mathrm dz$.

If the second condition holds, let $\gamma$ be a closed path. And let $\gamma^\star\colon[0,1]\longrightarrow\mathbb C$ be the constant path such that $\gamma^\star(t)$ is equal to the initial (and final) point of $\gamma$. Then$$\int_\gamma f(z)\,\mathrm dz=\int_{\gamma^*}f(z)\,\mathrm dz=0.$$

  • $\begingroup$ Why am I allowed to take the path $\gamma^{*}$? $\endgroup$ – Matheus Andrade Sep 14 '18 at 21:32
  • $\begingroup$ @MatheusAndrade I don't understand your question. If $[a,b]$ is the domain of $\gamma$ and if $\gamma(a)=\gamma(b)=z$, then I defined $\gamma^\star (t)=z$. That's all. $\endgroup$ – José Carlos Santos Sep 14 '18 at 22:44
  • $\begingroup$ It was a stupid question. Thanks anyway. I see what I missed now. $\endgroup$ – Matheus Andrade Sep 14 '18 at 22:51

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.