Introduction: Body composition assessment is essential in studies in medical, anthropological and sports science. The most commonly used methods are indirect (field) methods including bio-electrical impedance analysis (BIA), caliperometry, and ultrasound scans (US). Aim: To study reliability of body composition estimates using ultrasound scanner and Russian bioimpedance equipment in adults. Methods: A total of 191 adults living in Moscow, aged 18-74 years, were recruited. Among them were 135 women, aged 18-67 years, and 56 men, aged 18-74 years. The examination program included measurement of height and weight, waist and hips circumferences, determination of body composition using the BodyMetrix™ ultrasound scanner (IntelaMetrix, USA) and bioimpedance analyzer (ABC-02 Medass, NTC Medass, Russia). Reliability analysis was performed using Bland-Altman method. Results: More than 37 % of the individuals were overweight and obese. In the group of females, no significant differences in the amount of body fat mass (BIA 19,7 kg, US 18,7 kg) and fat-free mass (BIA 43,9 kg, US 43,6 kg) as well as the percentage of body fat (BIA 30,1 %, US 30,2 %) estimated by these methods were found. Similarly in the group of males, body fat mass (BIA 12,1 kg, US 12,5 kg) and fat-free mass (BIA 61,2 kg, US 61,6 kg) as well as the percentage of body fat (BIA 16,5 %, US 17,3 %). We however revealed a systematic bias, which increases with an increase in the value of the estimated parameters. Conclusion: Our findings suggest moderate reliability of the absolute fat mass in the group of females. For all other estimates the reliability was low suggesting that ABC-02 "Medass" equipment does not seem to produce systematically valid results. Therefore, ultrasound scanners should be preferred in studies requiring high quality assessment of body composition.