Objective. To examine mortality from cirrhosis of the liver and primary liver cancer among first generation migrants to England and Wales. Design. Comparison of standardised mortality ratios (SMRs) for cirrhosis of the liver and primary liver cancer in men and women aged 20-69, by country of birth for the five year period 1988-1992. Setting. England and Wales. Results. There was a statistically significant two-fold excess of mortality from cirrhosis of the liver among male migrants from East Africa (SMR 286), India (SMR 261) and Bangladesh (SMR 254) as well as men born in Scotland (SMR 253) and Ireland (SMR252). Among women, only those born in Scotland (SMR 254) and Ireland (SMR 237) showed significant excess mortality. For liver cancer, significant excess mortality occurred among men born in the Caribbean (SMR 312), Bangladesh (910) and the African Commonwealth other than East Africa (1014), with Scottish and Irish born men showing more moderate excesses (136 and 170, respectively). SMRs were elevated also in all groups of foreign-born women but, probably owing to the small numbers of deaths, none of the findings reached statistical significance. Conclusions. Of public health concern is the excess mortality from cirrhosis in first generation immigrants to England and Wales from Scotland and Ireland (men and women) and in male migrants from India, Bangladesh and East Africa. Of equal concern is increased mortality from liver cancer in all foreign-born groups of both sexes, particularly among Bangladeshis, and African-Caribbeans. As well as promoting sensible drinking among immigrant men, specific preventive measures for those of Bangladeshi, African-Caribbean origin may include selective screening for hepatitis B and C and other tumour markers. Screening for liver cancer using imaging techniques needs further investigation. The benefit/cost ratio should be assessed by the Screening Committees of the UK Departments of Health. At local level, variation in incidence and prevalence of hepatic disease and feasible prevention programmes should be assessed within developing health improvement programmes.