The age at menarche in Assam and Burma.

Most of the reports from India on the age at menarche have been based on recollected ages of commencement of the periods and the accuracy of these is doubtful. Examples of such ages vary from 13.42 years (Israel, 1959) based on replies from 2,227 women from different parts of India to 14.88 years (Shah, 1958) based on questionnaire forms completed by 2,391 Hindu girl entrants to Guj-rati University, Ahmedabad, in West Central India. Two other series are quoted by Shah from the same area of India-989 girls of Maharaja Sayajirao University during 1950, 1951 and 1952, had an average recollected age at menarche of 13.8 years, and during 1954, 2,370 girls in high schools in Poona had an average recollected age at menarche of 14.06 years. The technique adopted in this study was to ask the girls their age and whether or not they had started menstruating. From the data obtained (Tables 1 and 2) the average age of onset of menstruation was calculated by the statistical technique of probit analysis. In Assam this average age was found to be 13 21 0 1 1. In a study in Burma by Foll (1958) a menarcheal age of 14-40 was reported; unfortunately the calculation was in error, and the corrected value is 13 25 i 0.08. In both these series the probit technique gave an excellent fit to the data. The standard deviation in Assam was 1 .62 years and in Burma 1 12 years. The 1,150 girls in the Digboi, Assam, series were mostly Assamese, Bengalis and Nepalis with a few Central and Southern Indians, and the majority came from Hindu families, although there were some Buddhists, Sikhs, Christians and Muslims. In the Chauk, Burma series, the 702 girls were predominantly Burmese Buddhists, with a few Chinese, Indians or Anglo-Indians. If the method of probit analysis is to be used with success it is essential that the girls being interviewed should have an accurate knowledge of their ages, and ages in the tropics are notoriously unreliable. It is only in communities where accurate birth dates are required for making personal horoscopes (e.g. in Burma where Birth Notices in the Press often include the exact time, to the nearest second, of the birth) or where there is a settled community having adequate birth records, e.g. in some missions or amongst an industrial company's employees that any accuracy may be expected. Studies by probit analysis have been reported by Wilson and Sutherland (1953), and include a series by Ellis (1950) in Southern Nigeria (Thomson 1952), and their own work in Nigeria, Ceylon and England. Their results are summarized in Table 3 and compared with the Assamese and Burmese findings.