[guided]We start from condition (1): assume $f \in C^1(\Omega)$ and $\partial_{\bar{z}_j} f = 0$ for all $j = 1, \dots, n$. The goal is to show that $f$ is locally representable by a convergent [power series](/page/Power%20Series).
**Why does $C^1$ plus Cauchy--Riemann give separate holomorphicity?** Fix $a \in \Omega$ and choose $r = (r_1, \dots, r_n)$ with $r_j > 0$ such that $\overline{\mathbb{D}^n(a, r)} \subset \Omega$. For each fixed $(z_2, \dots, z_n)$, consider the function $z_1 \mapsto f(z_1, z_2, \dots, z_n)$. This is a $C^1$ function of one complex variable satisfying $\partial_{\bar{z}_1} f = 0$. In one complex variable, the classical Cauchy--Riemann theorem tells us that a $C^1$ function satisfying $\partial_{\bar{z}} f = 0$ is holomorphic --- the $C^1$ regularity is what allows us to write $f$ as the integral of its differential form and apply Green's theorem to establish the [Cauchy integral formula](/theorems/345). Since the same argument applies to each variable $z_j$ with the other variables held fixed, $f$ is separately holomorphic on $\Omega$.
**Why do we need local boundedness, and how do we get it?** To pass from separate holomorphicity to a [power series](/page/Power%20Series) representation, we need to iterate the one-variable Cauchy formula across all $n$ variables. This iteration requires that $f$ be locally bounded (so that the iterated integrals converge and [Fubini's theorem](/theorems/2961) applies). Since $f \in C^1(\Omega)$, it is in particular continuous, hence bounded on the compact set $\overline{\mathbb{D}^n(a, r)}$.
**Obtaining the [power series](/page/Power%20Series).** With separate holomorphicity and local boundedness in hand, we apply [Osgood's Lemma](/theorems/3379) (or equivalently, iterate the one-variable [Cauchy integral formula](/theorems/345) directly as in the proof of the [Cauchy Integral Formula on Polydiscs](/theorems/3398)). This gives the iterated Cauchy integral representation of $f$ on the polydisc $\mathbb{D}^n(a, r)$. We then expand the Cauchy kernel $\frac{1}{\zeta_j - z_j}$ as a geometric series $\sum_{\alpha_j = 0}^\infty \frac{(z_j - a_j)^{\alpha_j}}{(\zeta_j - a_j)^{\alpha_j + 1}}$ in each variable. Since the geometric series converges uniformly on $\overline{\mathbb{D}^n(a, r)}$ (because $|z_j - a_j| < r_j \leq |\zeta_j - a_j|$ for $\zeta_j$ on the boundary), we may interchange summation and integration to obtain the convergent [power series](/page/Power%20Series)
\begin{align*}
f(z) = \sum_{\alpha \in \mathbb{N}_0^n} c_\alpha (z - a)^\alpha
\end{align*}
on $\mathbb{D}^n(a, r)$, where the coefficients $c_\alpha$ are given by the iterated Cauchy integrals. This establishes condition (2).
Alternatively, one can bypass [Osgood's Lemma](/theorems/3379) and derive the [Cauchy integral formula](/theorems/345) directly: the $C^1$ condition and Cauchy--Riemann equations allow an application of [Stokes' theorem](/theorems/1530) to pass from an integral over the polydisc boundary to the iterated contour integral. The result is the same.[/guided]