That brown adipose tissue is activated when mammals need to defend their body temperature is well established; also that prolonged exposure to cold leads to a recruitment of the tissue (i.e. increased amounts of the uncoupling protein UCP1). It is also generally accepted that it is this process that constitutes nonshivering thermogenesis, and that no other organ, protein or process is involved in nonshivering thermogenesis. Concerning the response of mammals to overnutrition, the situation is unsettled. The discussion centers on the phenomenon of “diet-induced thermogenesis”; it is, however, not established what this thermogenesis experimentally entails. In all animals, eating leads to a metabolic increase due to the cost of the digestive process as such (“obligatory diet-induced thermogenesis”). However, an additional component, occurring as a direct or indirect effect of overnutrition, is what is referred to as diet-induced thermogenesis (or better: “facultative diet-induced thermogenesis”). Although discussed for more than 30 years, the existence of this thermogenesis has repeatedly been questioned. The methodological shift from rats to mice, caused by the development of transgenic techniques, has had the generally unrealized effect that the metabolism of the experimental animal (i.e. of mice) is mainly governed by the need for the animal to defend its body temperature: mice under normal animal house conditions are in reality cold-exposed. Therefore, only in mice at thermoneutrality can the issue of the existence of metaboloregulatory thermogenesis be settled. We find that under such conditions, feeding mice an obesogenic diet (i.e. a high-fat diet or a cafeteria diet) does induce a recruitment of brown adipose tissue, accompanied by an increase in the metabolic response to norepinephrine. This response is fully dependent on UCP1. Moreover, mice that lack UCP1 become obese on a normal diet – but even more obese on a high-fat diet. Thus, it would seem that diet-induced thermogenesis does exist, and that in its absence, obesity easily develops. This must also mean that there must be signalling processes that, based either on food amount or quality – or on the effect of overfeeding, i.e. on obesity – govern the recruitment and activity of brown adipose tissue. Probably the mediatory signal is leptin. The blood levels of leptin increase not only with increasing lipid stores in the body but also as an acute response to feeding, and all leptin-induced thermogenesis is derived from brown adipose tissue. Mice without leptin are unable to develop diet-induced thermogenesis – whereas exposure of such mice to cold still leads to (some) classical nonshivering thermogenesis. Why this energy-wasting response to overnutrition has developed is still unknown. The idea that it may allow animals to “extract” e.g. protein from a poor quality diet without making them suffer from problems of overweight is appealing – but experimental evidence for this is still scarce. The present insight that also adult humans possess brown adipose tissue expands the interest in a metabolic defence against overnutrition from being an animal adaptation to being a possible factor in human health.