In the present study we examined whether the diurnal pattern and the 24-hour integrated serum concentration of osteocalcin (S-OC(I] showed any variation due to age, sex, season, or smoking habits in 31 normal subjects, aged 23-47 years. Blood samples were drawn every 60 minutes from 4 p.m. until 6 p.m. the following day. Serum osteocalcin showed diurnal variation as described earlier but no influence of age or sex on the diurnal variation could be demonstrated. Ten subjects were smokers, all consuming more than 10 cigarettes daily, but the diurnal variation in OC with time was not different from an age-, sex-, and height-matched group of strictly nonsmokers. Sixteen of the subjects were initially examined in March and 15 in November. Eight subjects were examined during a period in both November and March, and 5 of these went through an additional study period in July. These seasonal studies showed that the diurnal time course pattern during different seasons of the year were statistically indistinguishable from each other. The S-OC(I) was significantly higher in subjects below 29 years of age compared with subjects aged 30-40 years (P less than 0.01). Men had higher values than women (P less than 0.05). Moreover, S-OC(I) was higher in November compared to March (P less than 0.05). There was no effect of smoking on S-OC(I). A significant correlation existed between S-OC(I) and age (r = 0.50, P less than 0.005). Multiple regression showed that among age, height, and weight, only age and height were significant determinants for S-OC(I) (r = 0.50, P = 0.02).(ABSTRACT TRUNCATED AT 250 WORDS)