44 Negro Migration 



lowi ng distribution of farms by size__ groups from 1860 to 

 1910 gives a clear picture of the extent to which the pre- 

 dominating type of farm has shifted from the large planta- 

 tion to the smaller owner or_ tenant farm. 10 



Unfortunately the census does not subdivide the farms 

 over 1,000 acres in size. The rapid decline of the average 

 size of farms indicates the effect of these very large plan- 

 tations in the grouping. This rapid decline in the average 

 has been taken to mean the disappearance of large farms 

 between 1860 and 1880. But the number of farms of over 

 500 acres decreased by only 176 during the 20-year period 

 It is therefore evident that the average is not an exact index 

 of the subdivision. A better idea is gained from the differ- 

 ence of this arithmetic average, which is affected by very 

 large farms, and the median, which is the size of the farm 

 above which half of the number of farms are found and 

 below which half are found. Every large farm is evi- 

 dently balanced against one small farm in determining the 

 median. It will be noted that while the average declined 

 from 430 to 188 acres during the period from 1860 to 1880, 

 the median declined only from 98 to 92 acres. 



During the period, 1860 to 1880, plantations of over 1,000 

 acres held their own in number but were being reduced 

 nearer the 1,000 acre type by subdivision into tenant farms, 



10 United States Census of 1910, Agriculture, Vol. VI, p. 320, 

 1890; Agriculture, p. 116. The figures for 1860 are estimates 

 based on the census. As the census of 1860 enumerated 

 only Improved Acreage in farms, the size published is too small 

 to be comparable with the later years which included all acreage. 

 This reduced the number of large farms tremendously. The 

 estimate of Banks, Economics of Land Tenure, p. 20-21, was 

 therefore accepted and applied to the 1860 figures. The estimate 

 is close enough for purposes of comparison. The figures for 

 1870 are omitted because they were enumerated on the same 

 basis of improved acreage only, and because the notorious in- 

 accuracies of this census made an estimate based upon them of 

 little value. 



