CRYSTALLITE  . In

media which, on account of their viscosity, offer considerable resistance to those molecular movements which are necessary for the building and growth of crystals, rudimentary or imperfect forms of crystallization very frequently occur . Such media are the volcanic rocks when they are rapidly cooled, producing various kinds of pitchstone, obsidian, &c . When examined under the microscope these rocks consist largely of a perfectly amorphous or glassy base, through which are scattered great numbers of very minute crystals (microliths), and other bodies, termed crystallites, which seem to be stages in the formation of crystals . Crystallites may also be produced by allowing a solution of sulphur in carbon disulphide mixed with Canada balsam to evaporate slowly, and their development may be watched on a microscopic slide . Small globules appear (globulites), spherical and non-crystalline (so far as can be ascertained) . They may coalesce or may arrange themselves into rows like strings of beads—margarites—(Gr. papyapIrgs, a pearl) or into groups with a somewhat radiate arrangement—globospherites . Occasionally they take elongated shapes—longulites and baculites (Lat. baculus, a staff) . The largest may become crystalline, changing suddenly into polyhedral bodies with evident double refraction and the optical properties belonging to crystals . Others become long and thread-like—trichites (Gr . Opii , rptx6r, hair)—and these are often curved, and a group of them may be implanted on the surface of a small crystal . All these forms are found in vitreous igneous rocks .

H . P . J . Vogelsang, who was the first to

direct much attention to them, believes that the globulites are pre. liminary stages in the formation of crystals . Microliths, as distinguished from crystallites, have crystalline properties, and evidently belong to definite minerals or salts . When sufficiently large they are often recognizable, but usually they are so small, so opaque, or so densely crowded together that this is impossible . In igneous rocks they are usually felspar, augite, enstatite, and iron oxides, and are found in abundance only where there is much uncrystallized glassy base; in contact-altered sediments, slags, &c., microlithic forms of garnet, spinel, sillimanite, cordierite, various lime silicates, and many other substances have been observed . Their form varies greatly, e.g. thin fibres (sillimanite, augite), short prisms or rods (felspar, enstatite, cordierite), or equidimensional grains (augite, spinet, magnetite) . Occasionally they are perfectly shaped though minute crystals; more frequently they appear rounded (magnetite, &c.), or have brush-like terminations (augite, felspar, &c.) . The larger microliths may contain enclosures of glass, and it is very common to find that the prisms have hollow, funnel-shaped ends, which are filled with vitreous material . These microliths, under the influence of crystalline forces, may rank themselves .side by side to make up skeleton crystals and networks, or feathery and arborescent forms, which obey more or less closely the laws of crystallization of the substance to which they belong . They bear a very close resemblance to the arborescent frost flowers seen on window panes in winter, and to the stellate snow crystals .

In magnetite the growths follow three axes at right angles to one another; in augite this is nearly, though not exactly, the

case; in hornblende an angle of 57° may frequently be observed, corresponding to the prism angle of the fully-developed crystal . The interstices of the network may be partly filled up by a later growth . In other cases the crystalline arrangement of the microliths is less perfect, and branching, arborescent or feathery groupings are produced (e.g. felspar, augite, hornblende) . Spherulites may be regarded as radiate aggregates of such microliths (mostly felspar mixed with quartz or tridymite) . If larger porphyritic crystals occur in the rock, the microliths of the vitreous base frequently grow outwards from their faces; in some cases a definite parallelism exists between the two, but more frequently the early crystal has served merely as a centre, or nucleus, from which the microliths and spherulites have spread in all directions . (J . S .