Our knowledge of the behaviour of radio-elements in precipitation reactions is mainly due to Fajans and his co-workers, who investigated the relation existing between the completeness of the removal of the radio-element from solution and the solubility of its salt. They found that, when a filterable quantity of the precipitate of a common element was brought down in a solution containing an infinitesimal quantity of a radio-element, the radio-element itself was precipitated, if its corresponding compound was also insoluble. This general conclusion was later investigated by Fajans and Richter (
loc. cit
.) with respect to thorium B, and it was further established that the degree of removal of the radio-element was dependent on the solubility of the precipitate of the ordinary element. Thus with very insoluble precipitates such as bismuth sulphide and barium sulphate, thorium B was completely precipitated, while less insoluble precipitates like silver chloride carried down only part of the radio-element. It was also suggested that when the precipitate was the insoluble salt of an isotope, the removal of the radio-element was to be ascribed to its solid solution in the isotopic precipitate, whereas removal by precipitates of dissimilar elements was an adsorption effect. This last fact, the connection between the adsorption of radio-elements by various substances and the solubility of the corresponding active compounds, was the subject of detailed researches by Paneth, and by Horovitz and Paneth. As a result, it was proved that the connection between the magnitude of the completeness of removal of the radio-element and the solubility of the analogous radio-active compound, held for a solid precipitate added to the solution as well as for the case of a precipitate brought down in the solution. Experiments have also been conducted in this field by Ebler and van Rhyn and others.