Swedish inactivated polio vaccines have contained per single human dose a mean amount of viral antigen equivalent to 1 x 10(7.5) CCID50 of type 1, 1 x 10(7.4) of type 2 and 1 x 10(7.8) of type 3 produced on primary monkey kidney cells. Potency tests were made in comparison with an equivalent amount of live virus suspension of all three types. Validation of tests has been based on the response to type 1 only. Based on clinical experience with vaccine lots from 1957 and the establishment of the second live reference virus suspension in 1966, the minimum limit of immune response in guinea-pigs--expressed in extinction values--was decided as 1.5 for type 1 and type 3, and 1.0 for type 2. The potency test method used since 1959 in Sweden was two subcutaneous injections 2 weeks apart using 10 guinea-pigs per dilution and blood collected 1 week thereafter. Potency tests made according to European Pharmacopoeia revealed a somewhat lower value for type 2. D-antigen content in Swedish vaccines was low, however, the Swedish vaccine has protected against many episodes or outbreaks of wild virus in Sweden and immunized individuals elsewhere in the world. For the Swedish population a clear-cut clinical motivation for requiring a higher potency for type 2 as required in the European Pharmacopoeia or increased levels of D-antigen in the final product has not been presented. It was concluded that the European Pharmacopoeia method did not distinguish between doses of 0.5-1.0 ml. The minimum limit extinction value for type 2, i.e. 2.0 seemed to high.(ABSTRACT TRUNCATED AT 250 WORDS)