xiv FroceediiKjs of the Royal Society of Victoria. 



dating its birth from the publication of their own text- 

 books, Fiom my present point of view, I would ask 

 you to go back to the year 1863 as the date of the 

 beginning of new things — to the year when Newlands 

 first gave us the true principle for the classification of the 

 elements — a principle afterwards extended by Lothar Meyer, 

 Mendeleiefi", and others, and now so well known as the 

 Periodic Law. This is by far the grandest and most fruitlul 

 and far-reaching generalization of the past quarter century. 

 It underlies much of the most important work of the present 

 and of the future. 



If the known elements be arrano-ed in the order of their 

 atomic weights, from hydrogen (1) to uranium (239), along 

 a horizontal line, that line may be cut up into sections and 

 these sections may be placed each below the last, in such 

 manner that the elements which naturally resemble one 

 another — which form a natural group — always fall in the 

 same vertical line. To put it in another way — and a better 

 way — we can construct a curve of which the magnitudes of 

 the atomic weights are the abscissae and those of any pro- 

 perty capable of exact measurement are the ordinates ; and 

 we find that the curve, representing variation in the intensity 

 of the property, is of a periodic character, and that similar 

 elements occupy similar positions in the different periods. 



The properties of the elements (including the properties 

 of their compounds) are therefore functions of the atomic 

 weights. It follows from this, tliat it is highly desirable 

 that our knowledge of the atomic weights of all known 

 elements should be as exact as possible. 



The unit generally adopted as the standard for atomic 

 weights is H = 1. But almost all actual determinations of 

 atomic weights of other elements involve the previous know- 

 ledge of the value of 0, or of the ratio O : H. This may be 

 called the fundamental ratio, which underlies all others. If 

 our knowledge of that ratio be inexact, then all deduced 

 ratios, all our atomic weights, will be inexact in proportion, 

 and the greater the atomic weight the greater will be the 



