Intrinsic viscosities and molecular weights were evaluated to obtain Kuhn-Mark-Houwink-Sakurada (KMHS) relationships for some selected polymers by means of a non-linear regression analysis. The proposed method yields unbiased values of the KMHS parameters a and K , improves the fitting and shows that KMHS relationship must be considered as a simple correlation with two interdependent parameters, a and K .

The errors associated with the determinations of number average molecular weights through osmotic measurements and intrinsic viscosities through viscometric measurements are calculated by statistical computing. They are compared with the results previously obtained by means of graphical extrapolations, showing the advantages of the proposed method. ZUSAMMENFASSUNG:Die mit der Bestimmung von zahlenmittleren Molekulargewichten durch Osmometrie und von Grenzviskositaten durch Viskosimetrie verbundenen Fehler werden durch statistische Rechnungen bestimmt. Sie werden mit den friiher durch graphische Extrapolationen erhaltenen Resultaten verglichen und zeigen die Vorteile der vorgeschlagenen Methode.