In his Complex Variables text, Conway defines a triangular path as follows:
A closed path $T$ is said to be triangular if it is polygonal and has three sides.
I am having some confusions regarding his definition. First of all, why does he specify that the path is polygonal ? Does this mean some general version of a triangular path? My point is, if the path has three sides and is closed, then why does he need to specify that it is polygonal? What else could it be other than a triangle?
Additionally, I have another confusion regarding his proof of Morera's Theorem.
Morera's Theorem. Let $G$ be a region and let $f:G\to\mathbb C$ be a continuous function such that $\int_{T}f=0$ for every triangular path $T$ in $G$; then $f$ is analytic in $G$.
In his proof (which according to this question is wrong?), where does Conway make use of the hypothesis that $\int_{T}f=0$ for every triangular path $T$? TIA.
Polygonal refers to the fact the sides are straight and not curved. It is also oriented. Another term often used for this in maths is PL, meaning piecewise-linear.
Regarding the integration, let $T$ be the oriented triangle formed by $a, z, z_0$, so $T = [z_0,z] + [z,a] + [a,z_0]$, and note $0 = \int_T f = \int_{[z_0,z]} f + \int_{[z,a]} f + \int_{[a,z_0]} f = \int_{[z_0,z]} f - \int_{[a,z]} f + \int_{[a,z_0]} f$, where we have reversed the orientation on $[z,a]$ and thus the sign of the integral. We rearrange to get $F(z) := \int_{[a,z]}f = \int_{[z_0,z]} f + \int_{[a,z_0]} f$.