14 Florida Agricultural Experiment Station ages of protein, calcium, magnesium and iron than the outer cabbage leaves. It is obvious that a relatively small difference in the point of division of plant parts selected for analysis could invalidate comparison of data from different sources. Coleman and Ruprecht (4) obtained values that were in gen- eral somewhat higher for Florida vegetables than comparative data herein reported. They suggested that their data on iron analysis may have been high as a result of the use of an iron mill for sample preparation. Effect of Soils on Plant Composition Certain differences in plant composition appear to be attribut- able to major soil characteristics. Differential response of plant varieties is not considered in this report because they appear to be of minor importance. Moderate differences in fertilization within the same produc- ing area on similar soil types did not appear consistently to influence plant composition. There was some indication that differences in practices between areas may have been a factor. Cabbage grown in the Hastings area received relatively more potassium in the fertilizer and the crop was found to be rela- tively high in this constituent. Soil types were found to have considerable effect, but the chemical composition of the soil did not always correlate with plant analyses. These data are given in Tables 2 to 6, inclusive. The soil analyses as given in pounds of each element per acre are thought to be the most logical for correlation with plant analysis. The data for peats and mucks have been adjusted to compensate for the differences in apparent volume weights of soil. Soil data are given also on the direct weight-ratio basis in the appendix for those readers interested in a more detailed study of the data from this stand- point. The differences in exchangeable bases and weak acid-soluble phosphorus in the soils at the time of harvest appeared to be due more to soil type characteristics than to residual accumu- lation from differences in fertilization. Relatively higher potash concentration in the marl soils under tomato culture (Table 5) as compared with bean culture (Table 3) indicated differential residual effect of different treatment on a given soil. There is also some indication of similar residual effect on the Belle Glade area soils.