Johnston] NAVAHO POPULATION 143 



were supplied to me by these researchers consisted of single-year 

 age distributions, by sex, for each year from 1880 to 1948. These 

 data were collected by means of lengthy and repeated interviews 

 conducted, usually in the Navaho language, with a number of Ramah 

 Navaho respondents, in an atmosphere conducive to the highest 

 degree of mutual confidence and cooperation. Although it must 

 be recognized that, even under these ideal circumstances, some 

 individuals were probably "forgotten" by these respondents, it is 

 highly unlikely that the number of such omissions would suffice to 

 seriously bias the resultant distributions.^^ 



Secondly, the median ages for the Ramah community prior to 1920 

 are not much lower than those observed in other newly established 

 communities. Such communities typically contain a relatively 

 high proportion of young children and yomig adults. Under these 

 circumstances, median ages below 16 are not surprising. These 

 data do suggest, however, that the demographic characteristics of 

 the Ramah community, for all their accuracy, cannot be assumed 

 to be representative of the Navaho population as a whole. 



Three of the remaining four sets of Navaho median ages obtained 

 from data prior to 1920 are also extremely low, ranging from 14.39 to 

 16.48 years. The fourth set, for the Canoncito and Puertocito bands, 

 averages approximately 19 years (for both sexes combined). Here, 

 as with the Ramah, the small size of the population reduces the 

 significance of the medians obtained. It is noteworthy, however, 

 that the three sets of low medians pertam to the 1910 census of the 

 Navaho population and to two agency estimates; the Fort Defiance 

 Agency in 1915 and the San Juan Agency in 1916. All three of these 

 populations are more likely to have been underenumerated (or under- 

 estimated) than the much smaller and relatively well defined Puerto- 

 cito and Canoncito bands.^^ 



The remaining 19 sets of median ages derived from Navaho age 

 distributions can best be interpreted by giving first consideration to 

 the extreme values observed. The lowest medians in this group are 

 found among the Navaho population in 1936-38, the Fort Defiance 

 subagency population in 1956-57, and among the Indians in the 

 Navaho area as enumerated in 1960. In the former two instances, 

 available information regarding the procedures of data collection that 



^ The most likely omissions would be children who died shortly after birth. However, 

 the infant-child ratios for these Ramah populations, as shown in table 32, do not suggest 

 that such omissions were significant in number. More generally, it can be argued that 

 with any genealogical record, the probability of recall of ancestors is a function of the 

 number of their descendants. In the case of the Ramah community, however, it should be 

 noted that the community was founded within the living memory of its oldest inhabitants, 

 so that these differential probabilities would scarcely have an important influence upon 

 the recollections of its present inhabitants. 



•2 The total of 680 given in table 31, p. 158, for the San Juan Agency In 1916 represents 

 a sample of approximately 10 percent selected by me from the original rolls for that 

 agency. The actual population total for that agency would therefore be close to 6,800. 



