erally off the farm, and from the laboring and white-collar categories into higher business positions.
Further, just as the 1952 business leaders are drawn from somewhat lower occupational groups than the 1928 business leaders, so on the whole their grandfathers occupied the lower occupational brackets in greater numbers than did the grandfathers of the 1928 business leaders. The conclusion of a trend to increased occupational mobility, drawn from the earlier analysis, is given further support by the data for the longer time period. Comparisons between the grandfathers and the fathers of the 1928 and 1952 business elites show that, in general, there is little difference in the kind and direction of occupational sequences. However, there have been noticeable increases in recruitment from lower occupational categories over three generations during the last decades.
Mobility data in this study, and in most other studies, are presented in the form of tables which cross-classify father's occupation by son's occupation. These tables indicate the frequency of movement from any position to any other position in the occupational structure. The usual procedure in previous mobility research has been to consider primarily the diagonal cells, where father's and son's occupational position coincide. The frequency, or usually the proportion of cases in these cells, is primarily used as the measure of occupational inheritance or immobility. The remaining proportion of cases, those where the son's occupational position differs from that of his father, is taken as the amount of mobility experienced by the population. Thus, if 60 per cent of the sons of professional fathers are professionals, the remaining 40 per cent have experienced mobility; while if only 20 per cent of the sons of clerical workers are themselves clerks, 80 per cent of this group have been mobile. Therefore, the description runs, sons of clerks are twice as mobile as sons of professionals.
The inadequacy of these procedures has been demonstrated for systematic mobility research, especially for studies which compare social structures that differ from one another in place or time.1
The source of the difficulty is that no account is taken of the total number of positions available in each occupational class. Movement into and out of each occupational class needs to be considered in relation to this availability or demand factor. For example, this study will show that in the period from 1905 to 1912, in the community studied, the number of positions available in unskilled work outnumbered the positions