One of the most challenging questions in fluid dynamics is whether the threedimensional (3D) incompressible Navier-Stokes, 3D Euler and two-dimensional Quasi-Geostrophic (2D QG) equations can develop a finite-time singularity from smooth initial data. Recently, from a numerical point of view, Luo & Hou presented a class of potentially singular solutions to the Euler equations in a fluid with solid boundary Luo and Hou (2014a,b). Furthermore, in two recent papers Tao (2016a,b), Tao indicates a significant barrier to establishing global regularity for the 3D Euler and Navier-Stokes equations, in that any method for achieving this, must use the finer geometric structure of these equations. In this paper, we show that the singularity discovered by Luo & Hou which lies right on the boundary is not relevant in the case of the whole domain R 3. We reveal also that the translation and rotation invariance present in the Euler, Navier-Stokes and 2D QG equations are the key for the non blow-up in finite time of the solutions. The translation and rotation invariance of these equations combined with the anisotropic structure of regions of high vorticity allowed to establish a new geometric non blow-up criterion which yield us to the non blow-up of the solutions in all the Kerr's numerical experiments and to show that the potential mechanism of blow-up introduced in Brenner et al. (2016) cannot lead to the blow-up in finite time of solutions of Euler equations. Contents 1 Introduction 2 2 Some notations and definitions 8 3 Local regularity of the solutions 9 3.1 Local regularity for 3D Navier-Stokes or 3D Euler equations. . 10 3.