Motivated by the experimental study of Huang and Driscoll [Phys. Rev. Lett.
72, 2187 (1994)], we carry a systematic comparison between maximum entropy
and restricted minimum enstrophy vortices. We find that, when the angular
momentum constraint is reconsidered, the restricted minimum enstrophy princ
iple provides a good and useful approximation of the statistical theory, ex
cept for large energies. In the case investigated by Huang and Driscoll, th
is approximation even yields a better agreement with experiments than the f
ull statistical theory because the relaxation of the system toward the maxi
mum entropy state is not complete. However, we show that this agreement can
not hold in more general cases, due to physical inconsistencies of minimum
enstrophy principles. We also discuss the relevance of generalized thermody
namics (Tsallis entropy) to predict the equilibrium state. (C) 1999 America
n Institute of Physics. [S1070-6631(99)01710-9].