Metals And Alloys From A Colloid-Chemical Viewpoint.

ZAY JEFFRIES,* Cleveland, Ohio (written discussion?).-The microscope has, indeed, proved a mighty tool in the study of the structures of substances. It is limited in its resolving power to the wave length of the light used for illumination. A little in resolving power can be gained by using ultraviolet rays, which can only be detected by the photographic plate. Even the highest powered microscope only permits the study of aggregates of millions of atoms. The x-ray spectrometer, on the other hand, makes it possible to study the positions of the atoms in a crystal. The positions of the atoms in over fifty crystals have now been ascertained. But in amorphous substances the x-ray spectrometer is only able to tell us that the atoms seem to have no regular arrangement; the actual arrangement and method of grouping is still a mystery. In most perfectly formed crystals and in so-called "grains" in metals, the x-ray spectrometer teaches us that colloids seem to play no role inasmuch as the atoms are arranged in regular layers rather than in random clusters. In the boundary region between crystalline grains of like as well as unlike substances and in amorphous substances like glass, it seems probable that the random clusters, or colloids, play an important role. There still remains the application of wave lengths connecting the ultraviolet and x-rays both by direct reflection or diffraction and by the ultramicroscope. In the study of substances like glass, for example, it may be possible to identify the clusters or molecules by one wave length and to study the positions of the atoms in a cluster by using a smaller wave length. Any light that can be thrown on this important unknown subject and its relation to the properties of substances is therefore very welcome, and Mr. Alexander's paper is so written as to stimulate thought and experimentation in this direction. JEROME ALEXANDER (author's reply to discussion ?).-Col. Bancroft states' that in the case of alloys, the mass is always crystalline. While this generally may be so in the case of alloys, as ordinarily met with, before the alloy reaches this state or condition, its constituent particles must be in a still more finely dispersed state. In alloys, the speed' of transition between the finely dispersed state where the phases are amorphous, and the crystalline state where the phases are of known composi-