November 24, 1905.] 



SCIENCE. 



651 



atoms, either negative or positive, and that 

 material atoms are able to combine with 

 them — potassium with one of the positive 

 kind, zinc with two, chlorine with a nega- 

 tive one, — and so transport them in elec- 

 trolysis. 



The third great step was made by the 

 study of light, a physical property again. 

 Bunsen and Kirchhoff found that, heated 

 in the gaseous state, every atom emits a 

 definite set of light waves, producing a 

 characteristic line-spectrum which is yet 

 the sharpest test of the kind of atoms one 

 is dealing with, and which so became the 

 most fruitful guide in the detection of new 

 kinds. 



The last generalization that I have to 

 mention, and which we owe to Newlands, 

 Mendeleeft' and Lothar Meyer, includes 

 physical- properties in general, and asserts 

 that they vary with increasing atomic 

 weight in a periodic way. This shows it- 

 self most sharply in the atomic volume, 

 which passes through maximum values in 

 lithium (7), sodium (23), potassium (39), 

 rubidium (85) and caesium (133). A 

 corresponding periodicity is observed in 

 other properties, as, for example, that of 

 combining with electrical atoms or valency, 

 which in the said elements passes through 

 unity. Analogous behavior is exhibited 

 by the melting points and boiling points, 

 which for these metals are exceptionally 

 low. 



If my program did not to a certain ex- 

 tent exclude quite recent investigations, 

 confining me to a view of past history, I 

 should like to consider one more physical 

 property, tliat of radioactivity, which also 

 seems to be a property of atoms. I can 

 only insist on the fact that it was physical 

 properties again, the making the air con- 

 ductive for electricity, and the spectrum, 

 which revealed radium. 



Properties of Molecules. — Turning to 

 molecules, I have three predominant gen- 



eralizations to outline. The first is 

 Mitscherlich 's discovery of the fact that 

 analogous molecular constitution corre- 

 sponds to analogous outer crystalline form, 

 to so-called isomorphism. Let me add 

 that there is hardly any more satisfactory 

 proof of the soundness of our concept of 

 the internal structure of matter than, e. g., 

 the identity of the crystalline forms of the 

 alums, which we consider to have corre- 

 sponding internal structure. 



A second step, to a certain extent a sim- 

 ilar one, was made by Pasteur when he de- 

 duced disymmetry of molecular constitu- 

 tion from disymmetry in behavior, optically 

 as well as crystallographically. For in- 

 stance, the dextrorotatory ordinary tartaric 

 acid and its l^vorotatory antipode showed 

 this disymmetry both in optical rotation 

 and in the particular so-called enantio- 

 morphous crystalline form. The molecules 

 were supposed to have analogous structures 

 differing from each other as the right hand 

 from the left. As is well known, it was 

 only later that the probable molecular 

 structure was sharply defined, and stereo- 

 chemistry was founded. 



The third great step was the opening 

 of a way to determine the molecular 

 weights of dissolved substances. It was 

 chiefly the application of Avogadro's law 

 to osmotic pressures, in connection with 

 Raoult's measurements of freezing points 

 and vapor pressures, that opened the way. 

 We may now assert that the liquid state 

 is not characterized by high molecular com- 

 plexity. But the great innovation, intro- 

 duced by Arrhenius and immediately 

 brought into relation with the achievement 

 in question, was the admission of the ex- 

 istence of ions in electrolytes — for ex- 

 ample, the presence of negatively charged 

 chlorine atoms and positively charged 

 sodium atoms in an ordinary salt solution. 

 Once more it was a, physical property, the 



