552 



SCIENCE 



[N. S. Vol. XL VI. No. 1197 



not assumed, two different types of oxida- 

 tion reactions become necessary, and these 

 two types would be contradictory. This 

 was pointed out several years ago.^ 



Direct evidence for the polar nature of 

 valences involved in the Grignard reactions 

 is given by some recent experiments.' 

 Without going into details, these results 

 may be quoted. 



resistance in ohms. (ordinary conductlvltsf 

 apparatus) 



Ether above 1 X IOt 



Ether containing ethyl bromide. . above 1 X 10' 

 Ether containing 1.2 gm. magnes- 

 ium as Grignard reagent 



(MgCjHjBr) per 100 c.c 7. 1 X 10' 



Same with 0.3 mg. magnesium. . 1. X lOis 



1/50 M KCl aqueous solution 1.26 X 102 



A cell constructed with magnesium and 

 platinum electrodes, and a dry ethereal so- 

 lution of ethyl bromide containing a small 

 amount of previously prepared Grignard 

 reagent as solution gave electromotive 

 forces of from 0.5 to 1.5 volts. 



These results are of the greatest signifi- 

 cance with regard to the question of polar 

 and non-polar valence and indicate that the 

 valence or linkings of organic compounds 

 are of the same character as those of inor- 

 ganic compounds. They bear out the ex- 

 planation of the Grignard reaction on the 

 basis of the electron conception of valence 

 published several years ago, and in addi- 

 tion will unquestionably throw light on the 

 processes operating in solutions, aqueous 

 and otherwise. 



"With regard to Professor Jones's work 

 on electromerism, some interesting develop- 

 ments may be presented. As we under- 

 stand the term, electromerism means elec- 

 tronic tautomerism and includes substances 

 structiirally identical, but mutually trans- 

 formable by an exchange of negative elec- 

 trons between atoms composing the mole- 



2 Falk and Nelson, Jour. Amer. Chem. Soc, 36, 

 209 (1914). 



3 Nelson and Evans, ibid., 39 (1917) (January). 



cules. Thus ammonium nitrate, NH^NOj, 

 and hydroxylamine nitrite, NH3OH NO,, 

 while mutually transformable by a suitable 

 exchange of negative electrons, since as far 

 as the charges on the atoms are concerned 

 they differ only in the valence of the ni- 

 trate and nitrite nitrogen atoms, are not 

 structurally identical and would not, there- 

 fore, be classed as electromers. 



Professor Jones in his paper considered 

 electromeric nitrogen compounds. In elec- 

 tromers, the states of oxidation of certain 

 atoms in the structural isomers are differ- 

 ent. A number of years ago we showed 

 that the explanation of the isomerism of a 

 number of structurally identical organic 

 compounds may be referred to the state of 

 oxidation or the valence of certain atoms. 

 The compounds referred to are generally 

 known as geometrical or cis-trans isomers. 

 Direct evidence based upon the ionization 

 constants of organic acids'* showed that the 

 isomerism of maleic and fumaric acids is 

 due to phenomena now included under 

 electromerism while addition reactions of 

 unsaturated carbon compounds lead to 

 similar conclusions.'^ The evidence in de- 

 tail is given in the published papers and 

 need not be repeated here. It is possible 

 to go somewhat farther. "Werner and 

 Pfeiffer" have placed in parallel the so- 

 called geometrical isomerism of double 

 bonded carbon atoms and the isomerism due 

 to plane configuration of certain cobalt, 

 chromium and platinum compounds: 



C. C2 Pt Pt 



r/' "^x x/ \r R-^ ^X x-/ ^R 



(Pt(NH3)2CI.&c) 



4 Falk, Jour. Amer. Chem. Soc, 33, 1140 (1911). 



5 Nelson and Ealk, School of Mines Quarterly, 

 30, 179 (1909) ; Talk and Nelson, Jour. Amer. 

 Chem. Soc., 32, 1637 (1910). 



« Werner, ' ' Neuere Anschauungen auf dem 

 Gebiete der anorganischen Chemie" (1913), pp. 

 343, 345; Pfeiffer, Ztschr. physilc. Chem., 48, 40 

 (1904). 



