Cannizarro's Method of Atomic Weight Determination.
On September 3, 1860, because of the uncertainty about the determination of the atomic weights a conference was called at Karlsrule, Germany. This First International Chemical Congress was chaired by Jean Dumas (1800-1884) who had worked out an ingenious method for the precise determination of the densities of vapors. His method made possible, for the first time, the determination of gas densities of many substances normally liquid or solid at room temperature. The conference was attended by every important chemist in the world. The conference settled nothing. As the chemists prepared to leave in as great confusion of mind as when they had arrived,Angelo Pavesiof the University of Pavia stood at the door and handed out copies of a little pamphlet in Italian written by his friendStanislao Cannizzaro(1826-1916), professor of chemistry at the University of Genoa. The little pamphlet was entitledSketch of a Course of Chemical Philosophy(The English translation). It had originally appeared in theNuovo Cimentoin 1858 but had been reprinted at Pisa in 1859. A copy was given to the German chemistLothar Meyer. When Meyer returned home he read the pamphlet and, in his own words, "The scales seemed to fall from my eyes. Doubts disappeared and a feeling of quiet certainty took their place." He was so impressed that he made Cannizzaro's views the basis of his influential text book,Modernen Theorien der Chemiewhich was first published in 1864. Cannizzaro's method of atomic weight determination had cleared up the confusion.Cannizzaro's Interpretation of Avogadro's Hypothesis. He believed that acceptance of Avogadro's Hypothesis that equal volumes of gases, whether elements or compounds, contain equal number of particles was necessary for progress in the determination of atomic weights. He, however, drew a clear distinction between determining the relative molecular weights from gas density data using Avogadro's Hypothesis and determining relative atomic weights.
The essence of Cannizzaro's method of determining relative atomic weights of an element was to determine the relative molecular weights from the gas densities of as many gaseous or vaporizable compounds of that element as could be prepared. From the analysis of these compounds he found the smallest weight of the element contained in the molecular weight of the various gaseous compounds of the element. This smallest weight was accepted as the relative atomic weight of an atom of the element. The table below will illustrate the first part of Cannizzaro's method. Let us consider first just those substances containing hydrogen in Part A of the table. In column 1 there is listed the gas densities (grams per liter), all at the same temperature and pressure, of the various substances listed in first column. In column 2 are the percentage of hydrogen by weight in each of these gases, as determined by chemical analysis. Column 3 contains, for each substance, the product of gas density and percentage of hydrogen, that is, column 1 multiplied by column 2 (converting percentage to a decimal). This gives us in column 3 the weight of hydrogen present in one liter of the substance. Now an inspection of column 3 reveals that the numbers are integrally related so that if each number in column 3 is divided by the smallest number (that of hydrogen chloride), an integer is found. These are listed in column 4. Parts B and C of the table contain a similar analysis for oxygen and chlorine, repectively.

Cannizzaro's determination of atomic weights.
Having determined the number of atoms of an element per molecule of the element, he thus was able to determine the relative molecular weights of other compounds. The compound of water, for example, appears twice in the table, once in part A and once in part B. Its weight of hydrogen (Part A) is twice the least weight of that element, and its weight of oxygen (Part B) is equal to that element's least weight. According to Cannizzaro's interpretation, hydrogen and oxygen atoms must be in present in a 2:1 ratio in the water molecule. Using the known analytical result that the weight ratio of hydrogen to water is is nearly 1:8, he concluded that the oxyen atoms must therefore be 2 × 8 = 16 times heavier than hydrogen atoms. Thus the relative atomic weight of oxygen is 16.
The significance of Cannizzaro's Method.
Basic to Cannizzaro's entire scheme of atomic weight determination was the assumption that each list of compounds of a given element included at least one compound whose molecules contained only one atom of that element. If the list does not include such a compound, the weights in column 3 will still be multiples of some smallest weight. And the method may still be used for that element. The acceptance of Cannizzaro's proposal brought order out of the previously confusing assignment of atomic ratios and atomic weights.