OBJECTIVE: The aim of the present study was to calculate the overall heritability of some routine biochemical analyses. Furthermore, as genetic and environmental influences might differ across various segments, genetic impact in the highest and lowest thirds of the distributions was estimated. METHODS: Ninety-six monozygotic and 120 dizygotic same-sex twin pairs aged 82 and older were tested. Structural equation modelling was used to estimate the genetic and environmental influences on serum levels of albumin, calcium, total cholesterol, HDL-cholesterol, GGT, potassium, sodium, creatinine, urea, urate, cobalamin, folate, homocysteine, free thyroxine and thyroid stimulating hormone (TSH). RESULTS: Additive genetic influence of between 66% and 28% of the variance was accounted for all values except creatinine, for which the genetic influence was marginal. The highest influence was found for homocysteine, cobalamin, folate and HDL-cholesterol. Genetic influence for the tests was mainly in congruence with previous findings in younger samples. When limited to the highest and lowest thirds of distribution, there were substantial differences in the proportion of genetic influence for some tests. CONCLUSION: For the majority of biochemical tests, the magnitude of genetic influence is considerable. Heritability estimates, however, should be considered in a broad context, with age, gender, morbidity and medication taken into account. Notably, for many test values, the genetic impact may differ considerably between the highest and the lowest range of the distribution.