1: \begin{abstract}

2: Let $f $ be a germ of an analytic function at infinity that can be analytically

3: continued along any path in the complex plane deprived of a finite set of

4: points, $f \in\mathcal{A}(\overline{\C} \setminus A), \quad \sharp A <

5: \infty $. J.~Nuttall has put forward the important relation between the

6: \textit{maximal domain}  of $f$ where the function has a

7: single-valued branch and the \emph{domain of convergence} of the diagonal Pad\'e

8: approximants for $f$. The Pad\'e approximants, which are rational functions and

9: thus single-valued, approximate a  holomorphic branch of $f$ in

10: the domain of their convergence. At the same time most of their poles tend to

11: the boundary of the domain of convergence and the support of their limiting

12: distribution models the system of cuts that makes the function $f$

13: single-valued. Nuttall has conjectured (and proved for many important special

14: cases) that this system of cuts has \emph{minimal logarithmic capacity} among

15: all other systems converting the function $f$ to a single-valued branch. Thus

16: the domain of convergence corresponds to the \textit{maximal} (in the sense of

17: \textit{minimal} boundary) domain of single-valued holomorphy for the analytic

18: function $f \in\mathcal{A}(\overline{\C} \setminus A)$. The complete

19: proof of Nuttall's conjecture (even in a more general setting where the set $A$

20: has logarithmic capacity $0$) was obtained by H.~Stahl. In this work, we derive

21: \textit{strong asymptotics} for the denominators of the diagonal Pad\'e approximants for

22: this problem in a rather general setting. We assume that $A$ is a finite set of

23: branch points of $f$ which have the \emph{algebro-logarithmic character} and

24: which are placed in a \emph{generic position}. The last restriction means that

25: we exclude from our consideration some degenerated ``constellations'' of the

26: branch points.

27: 

28: \end{abstract}

29: