Fully water-soluble, partially N-acetylated chitosans with fractions of N-acetylated units (F A) from 0.12 to 0.60 (degrees of acetylation from 12 to 60%) and known random distribution of acetylated and deacetylated units were degraded with human milk lysozyme at pH 4.5 and ionic strength 0.16 (M). The initial degradation rates (r) of human milk lysozyme were determined as a function of f A, and were found almost identical to the double logarithmic plot of r versus f A determined previously for hen egg white (HEW) lysozyme (Nordtveit et al., Carbohydr. Polym., 23, 253–260, 1994), suggesting that the substrate specificities of human and HEW lysozyme with respect to partially N-acetylated chitosans are indistinguishable. Human lysozyme degradation rates of two neutral-soluble chitosans (f A of 0.42 and 0.60) were compared at pH 4.5 and 7.0. The initial degradation rate at pH 4.5 was about five times higher than the rate at 7.0. However, the ratio between the r-values for the two chitosans at pH 4.5 and 7.0 were almost identical, indicating very similar substrate specifcities at the two pH-values. In a more detailed study with HEW lysozyme, the effect of pH on the rate of degradation was determined for three chemically different chitosans, showing similar pH-dependence with a rather broad optimum around pH 4 for all samples. The parallel behaviour of the three chitosans as substrates with relative degradation rates independent on pH indicate, as with human lysozyme, substrate specificities largely independent on pH. The substrate specificities of HEW and human lysozyme were shown to be quite independent of ionic strength, whereas the absolute rates of degradation for the chemically different chitosans increased with ionic strength up to 0.2 M, as expected for a positively charged enzyme attacking a positively charged substrate. Since the relative degradation rates of chitosans with widely different F A-values, and thereby charge densities, were independent both on pH and ionic strength, in a pH-range affecting the charge density of both enzyme and substrate, it follows that short-range interactions (i.e. hydrogen bonding, van der Waals interactions and electrostatic forces in the binding site) rather than long-range electrostatic interactions between the positively charged lysozyme and the positively charged chitosan substrate are of importance for determining substrate specificities. Our data suggests that tailor-made chitosans with a predetermined degradation rate in the human body can be made by simply controlling their F A-values.