It has already been mentioned that the rise of pressure of argon with rise of temperature has been carefully measured by Drs. Randall and Kuenen, and that it is quite normal; no sign of splitting has been observed. But the range of temperature was not great (it was only from 0° to 280°), and it is quite possible that the change, if there was one, was so minute as to have escaped detection. Again, a more delicate method of detecting such a change is in the measurement of the ratio of the specific heats. The most trustworthy number obtained was 1·659 for the ratio, instead of 1·667, the theoretical figure. A mixture of 5 per cent of diatomic molecules should have reduced this ratio to 1·648. Here the evidence is, however, inconclusive. But on the whole, the presumption is against the hypothesis that argon is a mixture of monatomic with diatomic molecules. It still remains for us, therefore, to account for the fact that in the periodic table there is no place for argon, provided it be insisted on that the elements must follow each other in the numerical order of their atomic weights. If the numbers in the table actually showed regular intervals, or if there were any regularity to be detected in their differences, argon might be regarded as of wholly exceptional behaviour. But this is not so. Argon is an extreme instance of divergence, but similar divergences, though not of equal magnitude, are common. In attempting to offer an explanation of such anomalies, it must be remembered that the question is in itself a far-reaching one; and that although argon has served to direct attention anew to the anomalies of the periodic table, yet these anomalies existed before argon was discovered. It is necessary above all things to be clear as to what is under discussion. We speak of “atomic weights,” or “atomic masses.” What is meant precisely by these expressions? By mass, we understand that property of a body, in virtue of which, when acted on by a certain force for a certain time, it acquires a certain velocity. The product of the mass into half the velocity squared, or ½MV^2 (where M and V stand respectively for mass and for velocity), is what is termed kinetic energy. If the mass chosen be 1 gram, and the velocity 1 centimetre per second, the unit of energy is the product; it is termed an erg. The same unit of energy, the erg, is derived by the action of unit force, termed 1 dyne, through unit length, 1 centimetre. We have thus two equations, where F and L represent force and length, Kinetic Energy = ½(MV)^2 and Linear Energy = FL.