Many view with alarm the reported demise of approximately one farm in four, or about one million farms, during the 1960s. Kyle (p. 1) reports that a group of fairly conservative agricultural economists . . . expects a continued increase in the concentration of production on fewer and fewer In 25 years, this could result in a nearly complete demise of typical family farms in the units classified as commercial full-time farms. Recent data, however, suggest that such fears are premature. According to official census counts, the total number of commercial farms decreased from 2.42 million in 1959 to 2.17 million in 1964, and further, to 1.73 and 1.70 million farms, respectively, in 1969 and 1974.1 These losses represent 10.3%, 20.3% and 1.7% for the periods 1959-64, 1964-69, and 1969-74. One might conclude, therefore, that the number of farms has fallen in recent years but that the rest of decline fell sharply 196974. Two factors suggest that such a conclusion may be erroneous. The 1969 and 1974 censuses counted only those farm units which had gross sales exceeding $2,500, whereas the 1964 and earlier ones included units that had sales above $50. Since the major analytical purpose of census data is to describe various characteristics of a common set of farm units, some definitional change is desirable when commodity price changes have a significant impact on sales, the principal qualifying criterion. But the definitional change, as well as the farm product price escalation of the early 1970s, may have introduced serious bias in census counts of farm numbers. One way of appraising the impact of the definitional change is to assume that the 1969 definition applied also in 1959 and 1964. If so, the reduction in farm numbers 1959-74 would have been only 370,000 instead of the 720,000 actually reported, since many farms in the earlier years did not have $2,500 of gross sales. On the other hand, the number of farms qualifying in 1974 was substantially expanded because of the product price escalation that occurred in the early 1970s. The price escalation, therefore, had the effect of understating losses in numbers of We are, therefore, left in a quandary. The one factor suggests that 720,000 farms lost is too high; the other, too low. Perhaps the only safe conclusion is that the farm numbers game is a dangerous one to play. Similarly, escalated prices in the 1970s also preclude the observation that the rate of decline is diminishing in recent years. The index of prices received by farmers for all products stood at 96, 95, 107, and 192 for the years 1959, 1964, 1969, and 1974, respectively (U.S. Department of Agriculture 1976, p. 455). These facts imply that price effects may have been particularly significant in understating farm losses in the 1974 census.