A faceted stone is always cut from a single crystal, and not from an ordinary lump of the mineral, which is generally a mass of crystals. The chief reason why jewels are cut from natural crystals is that these, by virtue of their crystalline nature, are remarkably homogeneous, and therefore clear and limpid when free from cracks and flaws. A stone which is not homogeneous can never have the purity and limpid brilliancy of a single crystal, for at every point of contact of one part with another reflection takes place. Among minerals used as precious stones which are not crystals may be mentioned the opal. The opal probably owes its peculiar beauty to the very fact that it is filled with minute cracks or cavities, each of which contributes some tint of color by reason of its extreme thinness, just as the colors of the soap bubble are due to the thinness of its film. Or take the agate. Here the stone consists of layers of different materials differently colored. Its beauty is of a different nature from that of clear crystals, which it can never rival in brilliancy. Stones like the agate are generally classed apart as semi-precious stones, and their interest depends upon beauty of structure or color, or possibly to a large extent upon their rarity. The turquois, for example, is a very rare stone, which is apparently absolutely uncrystallized, but possesses great beauty of color, and is therefore much prized. The same is true of carnelian. On the present occasion we are not concerned with those opaque or curiously structured minerals whose beauty resides almost solely in their color. Those who have had no practical acquaintance with minerals have little idea how variable and accidental are their colors. They may scarcely realize that the ruby and the sapphire are the same mineral, and that this mineral also occurs, and is used in jewelry, absolutely colorless, when it is known as lux sapphire, green as the so-called Oriental emerald, and yellow as the so-called Oriental topaz; that topaz itself may be yellow, brown, blue, or colorless; that zircons range from colorless through almost all conceivable shades of brown and green, and that even diamond has been found green, red and blue. When we come to consider the properties by which precious stones are recognized, I shall say little or nothing about color, for it is of little value as a criterion. There are, for example, certain red stones which the most skillful experts cannot by their color alone refer with certainty to ruby, garnet or spinel. It might be expected that a noteworthy difference in chemical composition would accompany this difference of color, or that the pigment could be ascertained by analysis. In reality this is scarcely ever the case. It is fairly certain that the emerald owes its color to the presence of chromium, but the variation in the analyses of precious stones cannot generally be attributed to anything indicated by the variation of color.