Although holding no university appointment, and with a laboratory of the most modest dimensions and character, Berzelius, exercised great influence as a teacher. Some of the most notable chemists of the last century, such as Heinrich and Gustav Rose, Dulong, Mitscherlich, Wöhler, Chr. Gmelin, and Mosander, were among his pupils; and many of them have testified to his stimulating power as an investigator of nature, and to his merits as a worthy, genial man. The reasonableness of Dalton’s conjecture received further support from the discovery by Gay Lussac in 1808, that gases always combine in simple proportions by volume, and that the volume of the gaseous product formed, when measured under comparable conditions of temperature and pressure, stands in a simple relation to the volumes of the constituents. The law of pressure discovered by Boyle, that of thermal expansion by Dalton, and of volumes by Gay Lussac (which, it ought to be stated, was previously and independently made by Dalton), are explained on the assumption that equal numbers of the particles—either as simple particles or as compound particles—are present in the same volume of the gas. This method of explanation was first clearly stated by the Italian physicist =Avogadro= in 1811, but its significance, as will be seen subsequently, was not appreciated until half a century later. As the values for the atomic weights gradually became more exact, speculations arose as to the significance of the numerical relations which were observed to exist among them. In 1815 =William Prout= threw out the supposition that the atomic weights of the gaseous elements are multiples by whole numbers of that of hydrogen. Extended into a generalisation, this might be held to indicate that all kinds of matter are so many forms of a primordial substance. Subsequent inquiry showed that Prout’s “Law,” as it is sometimes called, was not tenable in its original form. Certain elements, it was conclusively proved, had atomic weights which were not whole numbers. Dumas subsequently modified the law, after a redetermination of a large number of atomic weights, by assuming that the substance common to the so-called elements had a lower atomic weight than unity. Although there are a considerable number of elements whose atomic weights, based upon the most accurate determinations, are remarkably close to whole numbers, the investigations of Stas and others afford no valid reason for believing that Prout’s hypothesis, and the underlying supposition to which it has been held to point, are justified by experimental evidence. CHAPTER X THE BEGINNINGS OF ELECTRO-CHEMISTRY