If a fluid body has cooled to the point at which it begins to become solid, for example, sulphur, in a round vessel, a crust of sulphur is not formed upon the surface of the cooled vessel, and another crust upon the surface of the sulphur itself, as might be expected; on the contrary, if a crystal be formed upon a point of the inner surface of the vessel, the crystal enlarges by growing in the direction of its axis, and the mass which surrounds the crystal remains liquid, and sometimes cools, without the molecules arranging themselves in the same manner as the crystal already formed. On examining the cooled mass, we observe that it shews a lamellar texture where the crystal was formed, and that the mass which surrounded it does not shew this texture in the same degree. This explains how veins of large-granular granite traverse a small-granular granite, as well as other phenomena of the same nature. This observation also affords an explanation of another phenomenon. If the half of the liquid mass has become solid, and if the fluid part be poured off, we obtain isolated crystals, which have been formed in the fluid mass. If the fluid part be not poured off, and be permitted to cool slowly, it contracts, as is the case with most bodies, and the contraction produces the same effect as the decantation; small cavities will be formed, and these will be traversed and covered over with distinct crystals. We also observe this phenomenon in the geodes of primitive and volcanic mountains, in which the crystals they contain are of the same minerals as those of which the mountains themselves are composed. NOTE E, p. 23. ON THE DISTRIBUTION OF BOULDER STONES IN SCOTLAND, HOLLAND, GERMANY, SWITZERLAND, AND AMERICA. Numerous large blocks are met with in almost every country of Europe, and frequently far removed from their original situations. This is frequently the case in Scotland: thus, in the Edinburgh district, we have numerous blocks of primitive rocks, of which no fixed rocks occur nearer than in our Highland mountains.