So, I've read a proof for path-independence in the case of the existence of an anti-derivative of a continous (in the domain $D$) complex function $f$, being the anti-derivative $F$ an analytic funtion in $D$. The proof starts:

In the case of the existence of $F$ $$ F'(z)dz = u_xdx -v_xdy + i(u_xdy + v_xdx) = u_xdx + u_ydy + i(v_xdx + v_ydy) $$

Since $du = u_xdx + v_xdx$ and the same goes for $dv$ we have $F'(z)dz = du + idv$.

Now we can integrate this

$$ \int_{\Gamma} F'z(dz) = \int_{\Gamma} du + i\int_{\Gamma} dv = u(x_2, y_2) - u(x_1, y_1) + i(v(x_2, y_2) - v(x_1, y_1)) = F(z_2) - F(z_1) $$

First of all, I feel like this only proves the necessity of the path independency, and not the sufficiency, for which I can't find the proof.

Now, the object of my question was, if this proof is correct, isn't it enough to prove that for an analytical function $f$ in $D$ then

$$ \oint_C f(z)dz = 0 $$

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.