LAUSR

We revisit the topic of shape evolution during the spherical collapse of an $N$-body system. Our main objective is to investigate the critical particle number below which, during a gravitational collapse, the amplification of triaxiality from initial fluctuations is effective, and above which it is ineffective. To this aim, we develop the Lin-Mestel-Shu theory for a system of particles initially with isotropic velocity dispersion and with a simple power-law density profile. We first determine, for an unstable cloud, two radii corresponding to the balance of two opposing forces and their fluctuations: such radii fix the sizes of the non-collapsing region and the triaxial seed from density fluctuations. We hypothesize that the triaxial degree of the final state depends on which radius is dominant prior to the collapse phase leading to a different scheme of the self-consistent shape evolution of the core and the rest of the system. The condition where the two radii are equal therefore identifies the critical particle number, which can be expressed as the function of the parameters of initial state. In numerical work, we can pinpoint such a critical number by comparing the virialized flattening with the initial flattening. The difference between these two quantities agrees with the theoretical predictions only for the power-law density profiles with an exponent in the range $[0,0.25]$. For higher exponents, results suggest that the critical number is above the range of simulated $N$. We speculate that there is an additional mechanism, related to strong density gradients that increases further the flattening, requiring higher $N$ to further weaken the initial fluctuations.