AIMS/HYPOTHESIS: The aim of this study was to investigate the associations between components of adult height (leg length, trunk length, ratio of leg to trunk length) and Type II (non-insulin-dependent) diabetes mellitus and insulin resistance. METHODS: A cross sectional study was carried out on 4286 women of age 60 to 79 years from 23 towns across England, Scotland and Wales. RESULTS: Total height was weakly and inversely associated with diabetes but this masked differences in the association with leg and trunk length. Leg length was inversely associated with Type II diabetes [age adjusted odds ratio (95% CI) for diabetes for each standard deviation (4.3 cm) increase in leg length: 0.81 (0.73, 0.90)] whereas trunk length was not associated with diabetes [age adjusted odds ratio (95% CI) for diabetes for each standard deviation (3.6 cm) increase in trunk length: 1.05 (0.94, 1.18)]. Adjustment for potential confounding factors attenuated but did not remove the inverse association between leg length and the prevalence of diabetes: fully adjusted odds ratio (95% CI) per standard deviation increase in leg length was 0.87 (0.77, 0.98) and that per standard deviation increase in the ratio of leg to trunk length was 0.88 (0.78, 0.99). In non-diabetic women leg length was inversely associated with insulin resistance, whereas trunk length was positively associated with insulin resistance. CONCLUSION/INTERPRETATION: Leg length is an indicator of early childhood environmental circumstances, in particular of infant nutrition. These results suggest that poor infant nutrition is an important causal factor in the development of Type II diabetes and insulin resistance in later life.