The Belgian chemist, Stas, who had been associated with Dumas in a classical determination of the atomic weight of carbon, set himself to determine, with the highest degree of precision then possible, the atomic weights of about a dozen of the elements, with a view of ascertaining (1) whether an atomic weight is a definite and constant quantity, or whether, as suggested by Marignac, and subsequently by Crookes, an atomic weight represents “a mean value around which the actual weights of the atoms vary within certain narrow limits”; (2) whether, if the atomic weights of the elements are respectively definite and invariable, the numbers are commensurable as alleged by Prout and Dumas; and (3) if it should turn out that the numbers are severally fixed and commensurable, whether this necessarily indicates that the elements are built up of a primordial matter, the πρώτη ιλη of the ancients, referred to by Prout in 1816. Stas devoted many years to the solution of these questions, working on a scale and with an accuracy and manipulative skill previously unapproached. The main results of his labour appeared in 1865. He concluded (1) that the atomic weights of the elements are absolutely constant values, and are not affected by the nature of the compounds in which they occur, or the physical conditions of their existence; (2) that the numbers so obtained are not commensurable: to quote his own words: “_On doit considérer la loi de Prout comme une pure illusion._” Hence the elements must, on the basis of Stas’s experimental evidence, be regarded as “_individualités à part_,” as he expressed it—each a primordial and unalterable substance. The appearance of this monumental work, which will ever remain one of the classics of chemistry, created a great impression. Its effect persists to this day. It constituted a model and furnished a standard which each successive worker has striven to emulate, with the result that atomic weights to-day are among the best ascertained of physical constants. Space will not permit of any detailed account of the work done in connection with atomic weights during the forty-five years which have elapsed since the publication of Stas’s memoirs; and the reader who desires fuller information must be referred to the special treatises on the subject, such as the _Constants of Nature_ of F. W. Clarke, or the monographs of Meyer and Seubert, Becker, Sebelien, and Van der Plaats. Reference, however, must be made to the determinations by Lord Rayleigh, Leduc, Morley, Noyes, Guye, Dixon, and Edgar of values which, like those of oxygen, hydrogen, nitrogen, silver, and the halogens, are largely made use of as fiduciary values in atomic-weight work.