On the Four-Dimensional Diluted Ising Model

8 we have found critical exponents very similar to those of the pure Ising model. We have found that the value of the critical exponents show that for lattices up to V = 32 4 the system, for p = 0:3, is not described by the mean eld theory, as one might have believed. Moreover the critical exponents that we have found are very near to those of the pure percolation. A possible explanation would be that the crossover from the percolation to the pure Ising is quite small, however we do not see any indications which point in this direction. These results suggest the existence of a new xed point, which can be reached only starting with strong disorder. It would be very interesting to investigate analytically the properties of this xed point. It may be possible that replica techniques may be useful here. Acknowledgments J. J. Ruiz-Lorenzo is supported by a MEC grant (Spain). It is a great pleasure for us to acknowledge interesting discussions with E. 9 the mean value we use whose obtaining with the whole set of hyperplane-hyperplane correlations. Using the c obtained in the susceptibility ts we calculate the exponent of the correlation length in a two parameter t, the result is: (11) with a 2 =d:o:f = 0:86. The largest value of that we have used in the previous t is max = 4:69(5). Taking account the error bars on c in (11) we report the nal value: = 0:7(1) (12) In Figure 2 we show the data for the non-connected susceptibility (Fig 2. lower), the connected one (Fig. 2. middle) and the inverse of the correlation length (Fig. 2. upper) along with our best ts for these observables. Also, we plot in Figure 1. the speciic heat (Fig 1. upper).