omental. A equação de Petroski (1995) atendeu de modo satisfatório aos critérios de aplicação utilizados e mostrou ser adequada para a população em questão. AbstractObjective: To describe anthropometric profile and body composition of elderly by anthropometry and evaluate the applicability of Petroski (1995) and Durnin & Womerseley (1974) protocols. Methods: We evaluated 37 elderly individuals of both genders. The anthropometric measurements were performed according to the guidelines of the International Society for the Advancement of Kineanthropometry and established the variables weight, height, eight skinfolds and seven muscle girths. It was calculated body mass index (BMI), waist-to-hip ratio (WHR), sum of five skinfolds and body density by Petroski (1995) and Durnin & Womersley (1974) elderly protocols with the conversion to percentage of body fat (%BF) by Siri (1961) protocol. For further discussion, it was applied the Student t test (p<0.05). Results: Age, weight, height and BMI did not differ between genders. The sum of five skinfolds resulted in normal range for both genders. The WHR indicated "moderate risk" for men, and "high risk" for women to develop metabolic complications. The Durnin & Womersley (1974) protocol underestimated in male subjects and overestimated in female subjects the %BF obtained by the equation of Petroski (1995), respectively. For both protocols, female subjects showed higher %BF than male subjects. Conclusions: We conclude that individuals of both genders have high body fat, especially in the abdominal region. The Petroski (1995) protocol answered satisfactorily to the applying criteria used and showed to be adequate for this population.