Non-universality for two-dimensional frustrated lattices?

Abstract
An exact relationship between the eight vertex (8V) model and the fully frustrated square Ising model (FFSI) is established. On the basis of this it is shown that the T=0 point of the FFSI model corresponds to the F model limit of the 8V model. The critical exponent eta =1/4 for the FFSI model is obtained. The corresponding value of eta for the fully frustrated triangular lattice was shown by Stephenson (1964, 1966) to be 1/2. Possible reasons for this apparent violation of universality are discussed.