272 FLORIDA GEOLOGICAL SJRVEY-I3TH ANNUAL REPORT

 The tables for white and negro farmers separately present many interesting features which it would take too long to discuss, but most of them can be picked out readily enough with the aid of the bold-face and italic figures. Generally speaking, the negroes are most efficient where they are least numerous, and those on the east coast seem to have nearly as high standards as the whites in some other regions (as already indicated by the illiteracy figures).
 The census tells little about the foreign white farmers except their numbers, but by doing a little adding, subtracting and dividing we can ascertain that of those in central Florida in 1910, 90.2% owned their farms, 7.3% were managers, and 2-5% tenants: while the corresponding figures for native white farmers were 84.1, 6.5, and 9.4. This agrees very well with the showing with respect to illiteracy of the rural white and foreign population brought out in an earlier chapter. The nationality or foreign farmers is not given by counties, but a little more than half of the.foreign white farmers in Florida in 1910 were in central Florida, and the leading nationalities among them in the whole State were English, German, Canadian, Swedish, Irish, Scotch, and Danish.
 The State census of 1915 dealt with population and manufacturing only, but for some years past the State agricultural department has been taking a census of crops, etc., every two years, going into much more detail than the federal censuses; and two of these State censuses have been used in the foregoing pages in determining the relative importance of different crops in each region. The nurniber of acres in cultivation in each county has been given in the last few biennial crop censuses, and the report for 1917-18 gave the number and acreage of farms, but nothing about the color and tenure of farmers, the value of farm property, or the expenses of farming. On account of the limited funds available for these crop censuses the work has to be done rather hastily, and the results are further vitiated by typographical errors, so that it is not safe to use them for statistical work involving ratios and percentages.