Age differences in the androgyny score.

METHODS The sample examined consisted of 413 men and 406 women aged 20-90 years. This sample was obtained in two parts. The first contained 215 men and 272 women aged 62-90 years, who formed a simple random sample of the 27,000 older people living in a defined area of Edinburgh. The method of sampling these subjects, who are taking part in a longitudinal study of ageing persons, has been described elsewhere (Milne, Maule, and Williamson, 1971). The second part was composed of 198 men and 134 women aged 20-65 years, who were donors providing blood at the Regional Blood Transfusion Service in Edinburgh. I attended donor sessions and circulated a leaflet requesting volunteers for anthropometry. More people volunteered than could be measured in the time available. Bias was introduced by the Blood Transfusion Service policy of rejecting donors under 5 feet (1 5 m) in height or 8 stone (50 -8 kg) in weight. This sample, although not random, is from the population living in Edinburgh. The distribution of Landsteiner blood groups in the samples of older people and of blood donors, who were measured, did not differ significantly from that published for Edinburgh (Kopec, 1970). In this respect, therefore, the samples do not differ from the population of Edinburgh. The age distribution of the combined samples of older people and blood donors is given in Table I. Measurements of biacromial (BAD) and bi-iliac (BID) diameters were made from behind with the subjects standing, using the Harpenden anthropometer (Tanner and Whitehouse, 1957). Care was taken that the shoulders were in a relaxed position when measuring BAD, and that firm pressure was exerted with the branches of the anthropometer when measuring BID. The right suprailiac skinfold thickness was measured using the Harpenden caliper as described by Tanner (1959).