which lies between the extremes, the type element, silicon, foreshadows the members of its two series to an approximately equal extent.
As the tables which graphically portray the Periodic Law stand to-day, there is much which still remains to be cleared up. At the very outset we are met by the fact that we cannot tell in which group as familiar an element as hydrogen ought to be placed. It generally receives the first place in group one, but to some extent at least this position is based upon a misapprehension. Some years ago Pictet, of Geneva, was engaged in that work on the condensation and liquefaction of gases, which has rendered his name famous. On compressing hydrogen at a very low temperature, he obtained, on suddenly reducing the pressure, some heavy, steel-blue drops, much resembling mercury. This was erroneously supposed to be hydrogen in the liquid form. As a result, it seemed only natural to classify hydrogen with the metallic elements of the first group. Not only have later investigations shown that these drops were not liquid hydrogen, but quite recently Dewar has actually obtained this substance, which proves to be a colorless, limpid liquid, with the extraordinarily low specific gravity of about 0.07. As far then as physical properties go, there is no justification in classifying hydrogen with the metals of group one. Chemically, however, hydrogen is like the metals, electro-positive, though very weakly so, and it is possible that its position in the first group is less awkward than would be any other.
Of the other elements in the table with atomic weight below 100, all seem fairly well placed, though we have not as much knowledge of scandium as we could wish, and there is a difficulty that we shall soon notice in the eighth group. The first apparent blank space in the table is for an element with atomic weight of about 100. Such an element would be known as eka-manganese, and would possess properties which would to a considerable extent resemble those of manganese, but perhaps more closely those of ruthenium. Beyond this in the table we find many gaps, partly from the inadequacy of our chemical knowledge and partly from the likelihood that there exist rare elements which have not yet been discovered. Such elements probably occur in extremely small quantities, and may, for many years, perhaps forever, elude chemists. It seems improbable that there are undiscovered elements which exist in more than very small quantities; this is the testimony, not only of the chemical laboratory, but also of the spectroscope, that instrument which reveals to us the composition, not merely of substances in the laboratory, but also that of the sun and of the distant stars. From barium to tantalum few elements are known to which a definite place can be assigned, but here there is an indefinitely large number of what Crookes has called meta-clements, the rare earths. Their rela-
