In 1890 a young professor of Chemistry at the Faculty of Sciences of Toulouse tells, in a publication in the Annals of this faculty [1], his experience on the use of the classification of “corps simples” in education. To explain this, he develops an original approach by approaching this classification, graphically, from the periodicity of the curve representing, as a function of the atomic weights, the maximum valencies of the elements with respect to hydrogen, the hydrocarbon residues, the oxygen and chlorine alone or with a hydrocarbon residue.The comparison of the curve obtained with that representing the specific weights, always according to the atomic weights, allows him to arrive at a classification of the “corps simples” (elements) in natural families.More generally, the author speaks about the considerable progress made in teaching chemistry that such an approach has for students who appreciate its advantages.