1: \begin{abstract}

2: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3: We examine the Kosterlitz--Thouless universality class and show 

4: that essential scaling at this type of phase transition is not 

5: self--consistent unless multiplicative logarithmic corrections 

6: are included. In the case of specific heat these logarithmic 

7: corrections are identified analytically. To identify those 

8: corresponding to the susceptibility we set up a numerical method 

9: involving the finite--size scaling of Lee--Yang zeroes. We also 

10: study the density of zeroes and introduce a new concept called 

11: index scaling. We apply the method to the $XY$--model and the 

12: closely related step model in two dimensions. 

13: %Our method overcomes the hitherto encountered difficulties of 

14: %having to perform multi--parameter fits to extract the critical 

15: %coupling and critical indices from conventional thermodynamic 

16: %quantities. 

17: The critical parameters (including  logarithmic corrections) of 

18: the step model are compatable with those of the $XY$--model  

19: indicating  that both models belong to the same universality 

20: class. This result then raises questions over how a vortex binding 

21: scenario can be the driving mechanism for the phase transition. 

22: Furthermore, the logarithmic corrections identified numerically 

23: by our methods of fitting

24: are not in agreement with the renormalization group

25: predictions of Kosterlitz and Thouless.

26: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

27: \end{abstract}

28: