We analyze in details the statistical significance of the claim by Bird <2002> of a power law distribution of plate areas covering the Earth and confirm that the power law with exponent 0.25 ¿ 0.05 is the most robust and parsimonious model for all plates, including the very largest plates, when taking into account the constraint that the plates areas must sum up to 4 π steradians. We propose a general class of fragmentation models that rationalize this observation and discuss the implications for the Earth dynamics and the general self-organization of tectonic deformations at multiple scales.