Alcohol drinking behavior is usually studied from two perspectives, factors leading to drinking behavior and the behavioral effects of alcohol drinking. Many detailed medicolegal, pharmacological and psychiatric studies have been conducted on the behavioral effects of alcohol drinking. Few studies have considered the biological aspects of a desire for alcohol. However, these issues can not be ignored. The role of genetic, environmental and nutritional factors in alcohol preference has extensively debated. Recently, biochemical, physiological and pharmacological studies have also been performed to elucidate the mechanism of the desire for alcohol. In this study, the biological aspects of drinking behavior and alcohol preference have been studied using inbred strains of mice as an animal model on alcoholism. In addition, factors affecting drinking behavior of human beings are discussed based on the results obtained from a medico-legal study of alcohol-related cases. 1. Alcohol preference in several animal species The alcohol preference expressed as a ratio (%) of the volume taken (water and 10% (v/v) alcohol solution), was not constant in several animal species. The preference ratio was observed to be 2.7 +/- 0.7, 3.7 +/- 0.8, 22.0 +/- 19.8 and 39.6 +/- 5.4 in male inbred strains of SAMP2, DBA/2cr, B10.Br/Sg and C57BL/6J mice respectively, and 10.7 +/- 7.6, 15.9 +/- 13.6, 31.3 +/- 22.6 and 32.4 +/- 16.7 in male Donryu, DA, Wistar and Buffalo rats respectively, and 91.3 +/- 9.1 in male Golden hamster, and 2.1 +/- 0.3 in Hartley guinea pigs. Rabbits and Japanese monkeys do not demonstrate high alcohol preference. By comparison, the alcohol preference of a Japanese people was estimated to be approximately 11-35% on the basis of data obtained by questionnaire. 2. Development of alcohol dependence and withdrawal by voluntary alcohol intake in mice Eight strains of male mice, C57BL, C3H, SWM, SW, KK, KSB, KR and DBA, were offered a choice of water or 10% sake solution (sake containing 10% alcohol). Both young (3 months of age) and old (8 months of age) groups were studied simultaneously. The degree of intoxication was measured by recording the drinking behavior on a pulse recorder, by measuring gas-chromatographically the blood alcohol concentration, by taking depth electroencephalogram readings and so on. Intoxication, shown by lack of coordination such as grossly impaired gait, was observed only in the older mice of a strain with a moderate natural alcohol preference such as C3H, SWM, SW, KK and KSB.(ABSTRACT TRUNCATED AT 250 WORDS)