Having reached this conclusion it is natural to seek for confirmation by the investigation of the densities of the rocks concerned in the phenomena. As will appear by a table given further on, the density of the Henry Mountain trachyte has been determined to be 2.61.; but the densities of the erupted lavas of the Plateaus are not yet known. There can be no doubt however that the latter are heavier. Von Cotta in his Lithology gives 2.9 to 3.1 as the density of basalt, and 2.6 to 2.9 as the density of the more basic trachytes. And in general, it is well established that where the state of aggregation is the same, basic igneous rocks are always heavier than acidic. But in order that the laccolitic structure should have been determined by density, the acidic rock of the laccolites must have been heavier in its _molten_ condition than the more basic rocks of the neighboring volcanoes; and since in the _crystalline_ condition the acidic is the lighter, it follows that it has gained less density in cooling than the basic. If the amount of contraction of the several rocks in passing from their natural molten condition to the crystalline condition could be determined experimentally, a crucial test would be applied to our conclusions as to the origin of laccolites. The matter is however beset with difficulties. Bischof attempted by melting eruptive rocks in clay crucibles to obtain their ratios of expansion and contraction, but his method involved so many sources of error that his results have been generally distrusted. He concluded that the contraction in passing from the molten to the crystalline state is greater in acidic than in basic rocks. Delesse by an extended series of experiments in which crystalline rocks were melted and afterward cooled to glasses, showed that acidic rocks increase in volume from 9 to 11 per cent. in passing from the crystalline state to the vitreous, while basic increase only 6 to 9 per cent. Mallet concluded from some experiments of his own that the contraction of rocks in cooling from the molten condition is never more than 6 per cent., and that it is greater with basic than with acidic rocks; but considering that the substances which he treated were artificial and not natural products, that his methods were not uniform, and that he ignored the distinction between the vitreous and the crystalline, of which Delesse had demonstrated the importance, no weight can be given to his results.