A new sulfonated melamine-urea-formaldehyde (MUF) resin of relatively low melamine content, prepared according to a sequential formulation, has been shown to be highly effective when coupled with different natural vegetable tannins to produce leather with the same good characteristics of leather prepared with chrome salts. In particular, the antishrinkage effectiveness of the leather prepared according to the new approach is comparable to that obtained with chrome tanned leathers. The comparison of the traditional leather shrinkage temperatures test method with a new thermomechanical analysis (TMA) test method in tension yields thermograms presenting three major modulus of elasticity (MOE) peaks. These are closely connected to molecular level phenomena determining the shrinkage temperature of leather. The three determining parameters appear to be as follows: (1) The average value of the temperatures at which the three MOE peaks occur: the higher the value of this average, the lower is the shrinkage of leather. (2) The average of maximum MOE values of the TMA peaks: the higher this average is, the better is the leather in regard to antishrinkage effectiveness. This means the leather maximum MOE at each peak is a measure of the resistance to the contraction force induced by heat. (3) The relative intensity of the first TMA peak in relation to the second: the higher the value of the MOE for the first TMA peak is in relation to the second peak, the lower the leather shrinkage appears to be. However, it has not been possible to better define or quantify this latter effect. This new TMA test method in tension has also yielded a mathematical relationship correlating the thermogram peak temperatures and MOE averages with the traditional shrinkage temperature to a high degree of confidence. A previous TMA test method, in compression, has proven to yield more problematic and finally not very reliable results when one needs to apply it to a wide variety of different cases. (C) 2003 Wiley Periodicals, Inc.