Notes on scientific research. 115 



5 ccm. of phenylacetaldehyde are introduced into a 100 ccm. flask of special construction, 

 then 20 to 25 ccm. of solution of sodium bisulphite (containing about 30 percent. NaHS0 3 ); 

 the mixture is shaken vigorously for some minutes, and the whole is placed in a 

 water-bath. After some time, boiling water is gradually added until the solid compound 

 at first produced is completely dissolved. More water is then added, and the non- 

 aldehydes settle down in the graduated tube, where the volume can be read off when cold. 



As to the identification of nitrobenzene when present in bitter almond oil, see p. 6 

 of this Report. 



Re the estimation of vanillin according to Doherty see p. 100 of this Report. 



Physical Notes. 



As is generally known, the reason why the vapour tension curves of two sub- 

 stances do not run parallel to each other is to be found in the difference of the 

 changes of tension with identical changes of the temperature. The numerous curves 

 compiled by C. von Rechenberg 1 ) show that about 2000 pairs of curves cross each 

 other. When, in agreement with v. Rechenberg, the liquid and solid states are con- 

 sidered as chemical formations which on heating are decomposed into vapour mole- 

 cules, then the different stability of the molecular complexes is the cause of the varying 

 evolution of vapour, hence the latter is a measure for the former. When working out 

 the entire material v. Rechenberg was able to classify all the substances investigated 

 in relation to the increase in vapour pressure for the equal rise in temperature, and 

 this classification was clearly in accord with Rothmund's 2 ) solubility series. It may, 

 therefore, be concluded that in both cases the regularity is a result of the identical 

 cause mentioned above. 



Furthermore, C. von Rechenberg 3 ) utilized his excellent material of vapour tension 

 curves for testing Bancroft's rule which implies that all mutually soluble liquids of 

 which the vapour tension curves cross each other form "distinguished" mixtures with 

 either a minimum or maximum boiling point, v. Rechenberg was able to confirm the 

 general applicability of that rule, but with the restriction that such mixtures with 

 "distinguished" boiling points are limited to a certain range of pressure. 



With the object to investigate in the laboratory the separation of binary mixtures 

 by continuous distillation, A. F. Dufton 4 ) constructed a continuous still with a still-head 

 as had been devised by S. F. Dufton 5 ). The mixture to be separated was fed from a 

 small flask, where it was electrically pre-heated to its boiling point, into the middle 

 of the column. A siphon kept the level in the preheater constant. The lower part 

 of the still was also heated electrically by a wire coil and the current required was 

 measured by a watt-meter. To obviate any loss of heat from the lower part of the 

 column, a thick lagging of cotton-wool surrounded by a steam-jacket was employed. 

 Exact thermometers at the top of the still-head and at the bottom, of the column 

 indicated the purity of the separating products. 



Starting from a 50 per cent, mixture of benzene and toluene the author succeeded 

 in regulating the still to work automatically, yielding pure benzene and pure toluene 

 continuously. By measuring the head supply it was found that the separation of one 



x ) Zeitschr. f. physik. Chem. 99 (1921), 87. — 2 ) Ibidem 26 (1898), 489. — 3 ) Ibidem 99 (1921), 105. — 

 *; J&urn. chem. Hoc. 119 (1921), 1988. — 5 ) Journ. Sac. chem. Industry 38 (1919), 38 T; Bericht (German) 1920, 101. 



8* 



