OBJECTIVE: Differences in the metabolic response to overfeeding and starvation may confer susceptibility or resistance to obesity in humans. To further examine this hypothesis, we assessed the changes in 24 h energy metabolism in response to shortterm overfeeding and fasting in Caucasians (C) and Pima Indians (I), a population with a very high propensity for obesity. METHODS: We measured the changes in 24 h energy expenditure (24 -EE) and 24 h respiratory quotient (24-RQ) in response to 48 h of mixed diet overfeeding (100% above energy requirements) and fasting in a whole body respiratory chamber in 14 male subjects (7 Ca7 I, age 30 AE 6 y, mean AE s.d.). Results were compared to a 24 h measurement under eucaloric conditions. RESULTS: Mean 24-EE increased in response to overfeeding and decreased in response to fasting (all changes P`0.01), with no differences between C ( 9.1% and À9.1%) and I ( 8.6% and À9.6%). Similarly, mean 24-RQ increasedadecreased in response to overfeedingafasting, respectively (all changes P`0.01), again with no differences between C ( 0.06 and À0.05) and I ( 0.05 and À0.05). The changes in 24-EE in response to overfeeding and fasting were positively correlated (r 0.70, P`0.01), whereas those in 24-RQ were not (r 0.40, NS). CONCLUSIONS: Pima Indians do not appear to have an impaired metabolic response to short-term overfeeding and fasting that could explain their propensity for obesity. Individuals with a large increase in energy expenditure in response to overfeeding appear to have a small decrease in energy expenditure in response to starvation (spendthrift phenotype) and vice versa (thrifty phenotype).