Why Not Use Relevant Data in Soil‐Plant Relationships?

Chemical data in soil-plant relationships are relevant when they are based on the same volume from which plant roots obtain nutrients and water. For example, if plants are grown in pots containing 1,000 cm or 1 cubic decimeter (dm) of soil, chemical data are relevant when they are also determined and calculated on the basis of 1 dm. Similarly, chemical data are relevant when they represent the sampling depth and volume under field conditions. Under any of these or related conditions, chemical data obtained on a volume basis will be appropriately representative. Data determined and calculated on a weight basis will be equivalent to the volume and depth relationships when weight/ volume (wt/vol) or bulk density (BD) is approximately 1 g/ cm. In all other cases chemical characterization regresses in relation to volume by the factor (1/g/cm) and with respect to depth of sample by the factor [(1/g/cm) X dm] where dm is the depth desired. The magnitude of the relevance of chemical data reported on a weight basis cannot be judged by the editor or reader because investigators do not generally provide weight/volume measurements. A rare exception was that of Novais and Kamprath (1978) who reported wt/vol and expressed extractable P on a volume basis. The first three soils in Table 1 have a wt/vol of 1.68 (Norfolk), 1.59 (Goldsboro), and 1.23 (Portsmouth) and the respective Bray 1 P of 158, 141, and 71 /ig/cm from the above-cited publication, except that the authors used ppm in place of the appropriate /ig/cm. All other soil data in the table are from unpublished results for wt/vol (Mehlich 1972, 1973) and P by a new extractant (Mehlich, 1978). By reporting wt/vol on disturbed soil or bulk density on undisturbed soil, the information is in hand to calculate the degree of variability of chemical data expressed on a weight basis. The data in the accompanying table show that all values are unity when wt/vol = 1. However, with increasing g/cm the volume/unit weight (1/g/cm), depth, and volume of soil on a field basis decrease while extractable P similarly decreases. These properties conversely increase, when wt/vol decreases below 1. The maximum deviation of 2 million kg soil/ha decreased in depth from 2 dm (20 cm) to 1.2 dm (12 cm) and increased from 2 dm to 15 dm; the corresponding extremes in wt/vol on a field basis were 1.2 X 10" dm? and 15.4 X 10" dm. The comparable deviations for extractable P were 60 and 770% for Norfolk sandy loam and Canadian peat, respectively. Evidently data reported on a weight basis do not accurately represent the factual status of the soil-root nutrient environment. Hence, if advances in this branch of soil science are to be made, notably in soil test calibration with plant nutrient uptake and the development of mathematical models, it is imperative to use data based on volume. In greenhouse experiments, pot size quantities of soils in excess of 0.1 dm are, in general, not accurately measured by volume. It is therefore expeditious to weigh the soil in quantities calculated from wt/ vol determined in the laboratory to fill the choice of pot size. For example, if wt/vol was 1.3 g/cm, weigh out 1.3 kg soil per dm of selected pot size. Report soil chemical analyses in jug/ cm rather than in ppm. Express plant analysis data in mg of element in total plant mass per pot and convert this to uptake in mg element/dm (mg element/pot divided by dm soil/pot). Determination and reporting of chemicals on a volume basis would be independent of differences in wt/vol. Conversion to a field basis would be influenced by the choice of depth of soil depending on soil properties and plant characteristics. Valid