November 23, 1917] 



SCIENCE 



499 



existence of such an electromer could be 

 found. 



In an article^" published in the Jouriml 

 of the American Chemical Society, I pre- 

 sented evidence, which I believe to be con- 

 clusive, to show that the certain derivatives 

 of hj'droxylamine, prepared by Meisen- 

 heimer, represent the first known cases of 

 electromers, viz., compounds identical in 

 structural formulas, but dissimilar in chem- 

 ical and physical properties by virtue of a 

 different arrangement of valence electrons, 

 and the concomitant differences in force 

 fields within the molecules. 



A consideration of the properties of hy- 

 droxj'lamine, and its various derivatives 

 and, in particular, the peculiar behavior of 

 the hydroxyl group in such compounds, led 

 me to conclude that this hydroxyl group 

 could not be regarded as identical with 

 negative hydroxyl, — OH. This opinion 

 was expressed by Stieglitz.^^ 



The similar behavior of hydroslamine and hal- 

 ogen amines, of hydroxylamine and hydrogen per- 

 oxide, stiU more the fundamental similarity exist- 

 ing between hydroxylamine and ammonia, and 

 between their salts, and above all, the fact that, as 

 far as the writer is aware, no hydroxylamine de- 

 rivative has been found to exchange hydroxyl for 

 halogen by treatment with halogen acids, or phos- 

 phorus halides, are facts upon which the writer's 

 views are based. (According to a later private 

 communication from Dr. Jones, he has now reached 

 the same conclusion in this question and has dis- 

 covered further evidence supporting it.) 



These facts all imply that the hydroxyl 



group in hvdroxvlamine may be positive, 



H-0- + H, 



— + H 



N l-H 



hOH. 



Compounds of the formula R3N(0H)X" 

 are found among the products which result 



10 L. W. Jones, J. Am. Chem. Soc, 36, 1268 

 (1914). 



1-/. Am. Ch. Soc, 36, 288 (1914). 



IS Dunstan and Goulding, J. Chem. Soc, 69, 839 

 (1896); 75,1005 (1899). 



when hydroxlamine is treated with halogen 

 alkyls. Moist silver oxide converts these 

 substances into hydrated amine oxides, 

 R3N(0H)„, H^O. The same hydrated 

 amine oxides may be prepared by the action 

 of hydrogen peroxide" upon tertiary 

 amines, RjN. In fact, Hantzsch and Hil- 

 lard^" suggested that hydrogen peroxide 

 might react with tertiary amines by addi- 

 tion and that the reaction may be revers- 

 ible. 



(CH3)3N + HOOH ^ (CH3)3N(0H),. 

 By careful dehydration of hydrated amine 

 in vacuo, amine oxides, R3NO, are formed. 



These amine oxides and their hydrates 

 are oxidizing agents, and in this property 

 show a striking resemblance to hydrogen 

 peroxide. In fact, Dunstan and Gaulding, 

 in summing up their behavior, say : 



We conclude that the oxygen is in an "active" 

 condition analogous to the oxygen atom ia hy- 

 drogen peroxide. 



Thus, trimethylamine oxide rapidly de- 

 composes in two ways : 



2(CH3)3NO = 2(CH3)3N + O,, 

 (CH3)3NO= (CH3),NH + CHjO. 



These changes, looked at from the point of 

 view presented by the electron theory, 

 would lead to the inference that the oxygen 

 atom in amine oxides should be represented 

 as follows : R3N Z ; and that the hydrated 

 amine oxides, or their salts, should receive 



OH. 



the following formulas: R,N .' 



X 



If the hydroxyl group in hydroxylamine 

 be regarded as positive, and if this condi- 

 tion of the hydroxyl group be retained in 

 the alkyl (aryl) derivatives, two inferences 

 concerning the behavior of substituted 

 hydroxylamines must follow logically. 



In the first place, compounds containing 

 in their formula the group N — OH, or the 



isMerling, Ber., 25, 123 (1892); Wernick Wolf- 

 fenstein, Ber., 31, 1553 (1898); Mainlock and 

 Wolff enstein, Ber., 33, 159 (1900). 



20 Hantzsch and Hillard, Ber., 31, 2058 (1898). 



