474 



SCIENCE 



[N. S. Vol. XXIX. No. 742 



cbanges and adjustments it is easily and quickly 

 adapted for use as an optical bench or as an inter- 

 ferometer for electromagnetic radiation. The 

 wave-lengths preferred are from 10 to 15 cm., the 

 parabolic mirrors of 35 cm. aperture, lenses and 

 prism 22 cm. high, prism-table 26 cm. diameter, 

 and the length over all 225 cm. It is made of oak, 

 and provided with four graduated circles for read- 

 ing the angles through which different parts of 

 the apparatus are rotated when in use. 



A Method of Determining the Electrode Potentials 



of the Alkali Metals: Gilbebt N. Lewis and 



Chaeles a. Keaus, Massachusetts Institute of 



Technology. 



The electrode potentials of the metals of the 

 alkalies and the alkaline earths, notwithstanding 

 their great importance, have never been deter- 

 mined, because of the extreme reactivity of these 

 metals. The method now adopted, which has 

 proved entirely successful in the case of the so- 

 dium electrode, consists in measuring the electro- 

 motive force between the metal and its dilute 

 amalgam in mercury, with an electrolyte consist- 

 ing of a solution of a salt of the metal in liquid 

 ethyl amine. The electromotive force so obtained 

 can readily be shown to be independent of the 

 electrolyte and the solvent. It is, therefore, the 

 same as would be obtained if the electromotive 

 force between the metal and amalgam could be 

 measured in an aqueous solution. The potential 

 of the amalgam, against a normal aqueous solu- 

 tion of a salt of the metal may with certain pre- 

 cautions be measured directly against a normal 

 electrode. Adding the electromotive force so ob- 

 tained to the electromotive force between the 

 metal and amalgam gives directly the potential 

 of the metal in a normal solution of its ion in 

 water (potential of the normal electrode taken as 

 zero ) . In the case of sodium, this method has 

 made it possible to determine the electrode poten- 

 tial within a few tenths of a millivolt. The value 

 obtained is about half a volt higher than that 

 which has been previously assumed for the sodium 

 electrode. 

 Non-Newtonian Mechanics and the Principle of 



Relativity: Gilbert N. Lewis and Richaed C. 



ToLMAN, Massachusetts Institute of Technology. 



The laws of non-Newtonian mechanics previ- 

 ously derived by one of the authors from the 

 fundamental conservation laws and from a simple 

 assumption in regard to the nature of light are 

 identical with those which Einstein has obtained 

 from the principle of relativity and the laws of 



electro-dynamics. In this paper it is shown that 

 the same equations may be obtained without the 

 aid of the electro-magnetic theory from the prin- 

 ciple of relativity and the conservation laws. 



On the Influence of Temperature and Transverse 

 Magnetization upon the Resistance of Bismuth 

 and Nickel: F. C. Blake, Ohio State University. 

 The resistance of nickel and bismuth was in- 

 vestigated over a range of temperature from 

 — 192° C. to +183° C, and for all field- 

 strengths between and 36.6 kilogauss. For 

 measuring temperature flat spirals of fine pla- 

 tinum wire were attached to the mica supports 

 of the bismuth and nickel spirals. The apparatus- 

 was that previously used by duBois and Wills 

 {Verh. d. D. Phys. Ges., I., p. 169, 1899). 



At liquid air temperatures no such high values 

 of the resistance of bismuth were obtained as had 

 been obtained by Dewar and Fleming and by 

 duBois and Wills. Instead, a maximum of resist- 

 ance was found between — 160° and — 180° C. 

 for fields greater than 30 kilogauss. The higher 

 the field the higher the temperature at which this 

 maximum appeared. 



If R' is the resistance in the field H at the 

 temperature T, and R^ the resistance without the 

 field at 0° C, then R'/R„ = f(T, H). A set of 

 isothermal curves, R'/Ra^f{H) and another set 

 of isopedal curves, R'/R^ = f(T) were experi- 

 mentally determined. 



For nickel a set of isothermal curves, 

 {R' — R) /Ro = f(E), where R is the resistance 

 of the nickel out of the field at temperature T. 

 For all temperatures investigated the fraction 

 [R' — R)/R(, was negative for fields greater than 

 2,500 gauss, and its value was greater for tho 

 higher temperatures. For fields greater than 10 

 kilogauss it increased with increasing field except 

 at liquid air temperatures; at — 190° C. it was 

 a maximum at 8 kilogauss, decreasing slowly for 

 higher fields. For fields less than 2,500 gauss this 

 fraction was positive and it was thought that part 

 or all of this increase in resistance for low fields 

 could be explained by longitudinal magnetization, 

 whose presence could not be wholly avoided. 



A New Form of Standard Resistance: Edwabd B. 



Rosa, Bureau of Standards, Washington. 



The new form of resistance standard, which has 

 been developed at the Bureau of Standards during 

 the past two years, difi'ers from the Reichsanstalt 

 form in being smaller and having the resistance 

 coil sealed air tight in a case that is filled with 

 pure oil, insuring protection for the resistance- 



