Ever since the age-long committal of Earth Sciences to the hunt for natural resources turned into a largely non-profitable activity and the completion of the rather brutal metamorphosis of plate tectonics gave birth to a more mature and steady management of research in this field, Geochemistry has been undergoing a profound change. With all the excuses and the turmoil gone, the objectives of Geochemistry now join those of all other modern scientific fields: aside from a more accurate description of the world, either past or present, a set of quantitative concepts and rules must be built that will permit the outcome of geological processes and the future of geological systems to be predicted. Behind these vague terms hide such enormous challenges as the prediction of volcanic eruptions, the safety of drinking water, the evolution of the greenhouse effect, just to mention a few. The meaning of the quantitative approach in natural sciences is still blurred by confusion: there is no more quantification in plotting a few concentrations or isotopic ratios from a table in a geochemical diagram than coloring a geological map. Obviously, a wealth of high-quality information is still to be gathered through the description and comparison of geochemical measurements, but no more than can be gained from expert field work or from the intelligent practice of the optical microscope. In most cases, observations can be expressed in numbers that we call measurements, while the processes and causes invoked can be parameterized with yet more numbers. However, only a quantitative approach can test whether these processes and causes, parameterized in the most appropriate way, can reproduce the observations.