“…It is significant, as pointed out by all researchers in this field, for the greatest part of the variance to be ascribed to the geochemical data variance or geochemical (spatial) variation; otherwise, a contoured geochemical distribution map cannot be produced (Garrett, 1969;Howarth, 1983;Sharp, 1987). In the latter case, where geochemical data show no spatial persistence, element concentrations may be plotted at the sample sites as variable-size dots (Bølviken et al, 1986;Bj€ orklund and Gustavsson, 1987;Lahermo et al, 1990), symbols (De Vos et al, 1996), exploratory data analysis symbols (Englund and Sparks, 1988;O'Connor et al, 1988;Demetriades, 1990), or recording on maps the sample site analytical data (Van der Sluys et al, 1997). However, final decisions about geochemical distribution map plotting may be decided upon following a thorough geostatistical structural analysis of the data (Journel and Huijbregts, 1978;Rendu, 1978;Clark, 1979;Isaaks and Srivastava, 1989;Clark and Harper, 2007a, 2007b.…”