1: \begin{abstract}

2: We consider the Euler equation

3: for an incompressible fluid on a three dimensional torus,

4: and the construction of its solution as a power series

5: in time. We point out some general facts on this subject,

6: from convergence issues for the power series

7: to the role of symmetries of the initial datum. We then turn

8: the attention to a paper by Behr, Ne$\check{\mbox{c}}$as and Wu

9: \cite{Nec}; here, the authors chose a very simple

10: Fourier polynomial as an initial datum for the Euler equation

11: and analyzed the power series in time for the solution, determining the first 35

12: terms by computer algebra. Their calculations

13: suggested for the  series a finite convergence

14: radius $\tau_3$ in the $H^3$ Sobolev space,

15: with $0.32 < \tau_3 < 0.35$; they

16: regarded this as an indication that

17: the solution of the Euler equation blows up. \par

18: We have repeated the calculations of \cite{Nec}, using

19: again computer algebra; the order has been increased

20: from $35$ to $52$, using the symmetries of the initial

21: datum to speed up computations. As for $\tau_3$, our results

22: agree with the original computations of \cite{Nec} (yielding

23: in fact to conjecture that $0.32 < \tau_3 < 0.33$). Moreover,

24: our analysis supports the following conclusions: \parn

25: (a) The finiteness of $\tau_3$ is not at all an indication

26: of a possible blow-up. \parn

27: (b) There is a strong indication

28: that the solution of the Euler equation does not blow up

29: at a time close to $\tau_3$. In fact,

30: the solution is likely

31: to exist, at least, up to a time $\theta_3 > 0.47$. \parn

32: (c) \Pade analysis gives a rather weak indication

33: that the solution might blow up at a later time.

34: \end{abstract}

35: