36 



SCIENCE 



[N. S. Vol. LIV. No. 1384 



scribed as an ortho and a meta compound. It has 

 been supposed to be different from the meta com- 

 pound prepared by KafEka from m-sulpho benzal- 

 dehyde by the Perkin synthesis. We have shown 

 that it is identicalj and also that on oxidation with 

 permanganate it yields meta sulpho benzoic acid. 

 It can be easily characterized by the melting points 

 of its aniline and toluidine salts. 



Separation of aromatic primary and secondary 

 amines: I. N. Hultman and H. T. Clarke. The 

 separation of primary and secondary amines can in 

 most cases be carried out satisfactorily by taking 

 advantage of the solubility in alkali of the ben- 

 zenesulfonyl or toluenesulfonyl derivatives of the 

 primary amines. The recovery of the base from 

 such derivatives is however apt to be troublesome, 

 and in certain instances difficulties are met with 

 in the alkali treatment owing to the ease with 

 which the sodium salts are hydrolyzed. Thus they 

 can be extracted from their solution in alkali 

 merely by shaking with ether; and in cases where 

 only a small quantity of primary base is present, 

 the sulfonyl derivative of the secondary amine 

 acts in the same way as an extracting solvent. 

 This was found to be the case particularly with 

 p-toluidine and its monomethyl derivative. This 

 mixture moreover can not be separated by treat- 

 ment with zinc chloride solution, as is possible in 

 the case of aniline and methylaniline, since methyl- 

 p-toluidine appears to form an insoluble zinci- 

 chloride. Advantage can therefore be taken of 

 the observation that primary aromatic amines, 

 on heating to 160-180° TOth urea are converted 

 into symmetrical diarylureas, ■while secondary 

 bases, typified by methylaniline, do not react at 

 all with urea. It is thus merely necessary to heat 

 the mixed bases with urea in slight excess over the 

 calculated amount, and treat the product with 

 dilute acid. 



Potassium derivatives of the aXkyl amines: Ed- 

 ward C. Franklin. By the action of metallic 

 potassium on methyl amine, ethyl amine and di- 

 methyl amine in the presence of platiaimi black, 

 the compounds represented by the formulas 

 CH3NHK, C,H„NK and (0H3),NE: have been 

 prepared. 



The existence and reactions of positive halo- 

 gens attached to carton in aromatic compounds: 

 Ben H. Nicolet. W. A. Noyes, J. Stieglitz, L. 

 W. Jones and others have interpreted and con- 

 tributed a large amount of data on the positive 

 nature of halogen attached to nitrogen. HoweU 



and Noyes have shown the I of Cjl. to be positive. 

 It is now shown that a number of compounds, 

 most typically p-iodoaniline and 3-iodo 4-toluidine, 

 can be hydrolyzed with acids in such a way that 

 the halogen is replaced by hydrogen, and that the 

 halogen resubstitutes to give di-halogenated prod- 

 ucts — two proofs of positivity. Such halogens ap- 

 pear to be readily removed (replaced by hydro- 

 gen) by heating with SnOl. and HCl. 



Diphenyl-^-naphthylemethyl : M. Gomberq and 

 F. W. Sullivan, Jr. Diphenyl-j3-naphthylmethyl 

 was prepared and found to exhibit the character- 

 istic chemical reactions and physical properties of 

 free radicals. Determinations by the cryoscopic 

 method showed that dissociation increased with 

 rising temperature and with dilution. Contrary 

 to the theory that the color of free radical solu- 

 tions is due to dissociation of the dimolecular to 

 the monomolecular form, investigation by colori- 

 metric methods showed that the increase in color 

 intensity was quite independent of changes in dis- 

 sociation. It was also found that not more than 

 one third of the monomolecular free radical was 

 in the colored form. From these facts we believe 

 that the color of solutions of free radicals is best 

 explained on the assumption that the dimolecular 

 form dissociates to the mono-molecular form in 

 which equilibrium exists betweeen the colorless 

 benzenoid and the colored quinonoid tautometer. 



Contribution to the structure of benzidine: For- 

 mation of rings through the m and p positions of 

 benzene: Roger Adams and W. C. Wilson. 

 Benzidine condenses readily with aromatic di- 

 aldehydes or ketones of the type represented by 

 terephthalaldehyde and resodiacetophenone, to 

 give condensation products which analyze for one 

 molecule of benzidine plus one molecule of dialde- 

 hyde with the elimination of two molecules of 

 water. It is barely possible that these substances 

 consist of two molecules of benzidine with two 

 molecules of dialdehyde, with the elimination of 

 four molecules of water. The substances are too 

 insoluble to allow the molecular weight to be 

 obtained. If these substances are of the simpler 

 type, it is difficult to see how Kaufler's formula 

 for benzidine can be accepted, since his structure 

 assumes that the amine groups are as close to- 

 gether as the amine groups in ophenyleuediamine. 

 Benzidine also condenses with the monazine of 

 terephthalaldehyde to produce a product which 

 analyzes for one molecule of each with the elimi- 

 nation of two molecules of water. It is apparent 



