732 



NA TURE 



[December 2, 19: 



Letters to the Editor. 



[The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 din he undertake to return, or to cot respond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 token of anonymous communications .] 



The Isotopes of Antimony. 



Owing to the kindness of Prof. G. T. Morgan, who 

 prepared a specimen of pure antimony trimethyl for 

 this purpose, I have now been able to obtain the 

 mass-spectrum of antimony. The element is char- 

 .1 1 11 ed by two lines of nearly equal strength at 121, 

 123. The first is the more intense by perhaps 10 to 

 20 per cent. If sufficient exposure is given two faint 

 companions are visible at 122, 124, but the general 

 evidence suggests that these are due to hydrogen 

 addition products. The isotopic nature of the lines 

 121, 123 is amply confirmed by the appearance of 

 similar pairs 15 and 30 units higher, due to molecules 

 of their monomethides and dimethides. The most 

 trustworthy measurements show that the masses of 

 the isotopes of antimony are most probably less than 

 whole numbers by one to two parts in a thousand. 



These results show that the chemical atomic weight 

 120-2 at present accepted is certainly too low. They 

 are, however, in excellent agreement with the value 

 121-77 recently obtained in America by Willard and 

 M' Alpine. F. W. Aston. 



Cavendish Laboratory, 

 Cambridge, November 16. 



Experiments on the Theory of Soil -acidity. 



In a recent paper " On the Adsorption of Ions 

 (Phil. Mag. (VI.) 44, 321) the origin of soil-acidity 

 has been discussed (pp. 338-45, especially pp. 343-45). 

 In the following a short account is given of some 

 experiments carried out with Mr. Kamalacharan 

 Bhattacharya and Mr. Bankim Chandra Roy. 



It was suggested that the acidity is due to the 

 adsorption by the gels (of silicic acid, aluminium 

 oxide, and ferric oxide) of the anions of acids. The 

 adsorption is so strong that the adsorbed substance 

 cannot be washed out by water, and the aqueous 

 extract is neutral. The anions are adsorbed on the 

 surface by " chemical forces " whereas an equivalent 

 number of cations forms the mobile second sheet of 

 the double layer. If the cations consist in part 

 of H' ions, in treating with excess of a neutral salt 

 (KC1) solution there is a displacement of the cations 

 of the second layer by the cations of the neutral 

 salt, as the latter is present in relatively enormous 

 concentration, and as the forces acting on the cations 

 of the mobile second sheet of the double layer are 

 mainly electrical in nature. 



Experiments have been carried out with powdered 

 precipitates of silica, ferric oxide, and alumina. Of 

 these, silica has been found to adsorb appreciable 

 quantities of acids, e.g. acetic, citric, hydrochloric, 

 and nitric. The adsorption is so strong that on 

 repeated washing the adsorbed substance cannot be 

 removed so that the aqueous extract soon becomes 

 perfectly neutral. On now shaking the precipitate 

 with KC1 — which is tested with indicators to be 

 perfei tly neutral — the aqueous extract (free from 

 particles of the precipitate) is found to be distinctly 

 acid. The amount of the acid depends on the amount 

 of tin- precipitate. In the extract with the neutral 

 salt solution, acetates or citrates could not be detected. 



The formation of insoluble salts of alkali metals 

 or replacement of hydrogen ions by metallic ions in 



NO. 2770, VOL. I IO] 



complex silicic acids is evidently out of the question. 

 It is unanimously agreed that silica is an acid, and 

 the probability of forming definite complex acids 

 with acetic acid is very remote. 



It might be argued that the acids are adsorbed 

 as such, that is, the entire molecule is adsorbed. This 

 point has been settled by simultaneous experiments 

 on electro-osmosis. The apparatus used was a modifica- 

 tion of that used by Briggs (Joitrn. Phys. Client. 22, 

 1918, 256), which the writer found was employed by 

 I >r. Ishikawa in the Physical Chemistry Laboratory of 

 Prof. F. G. Donnan, University College, London. 

 The sample of pure precipitated silica (British Drug 

 Stores, Ltd.) we are using shows a marked negative 

 charge in pure water. On treating with acetate the 

 charge increases as shown by the rate of motion. 

 ili. results are accurate within about 10 per cent. 



Velocity in cm. per min. 



Pure water 2 cm. ; 2 '05 cm. 



N/rooo sodium acetate . . 3-9 cm. 



\ '2000 acetic acid . . . 27 cm. ; 27 cm. 



N/1000 sodium hydroxide . 3-4 cm. ; 3-2 cm. 



N/100 acetic acid . . . i-- cm. 



N/1000 hydrochloric acid . 23 cm. 



N/1000 potassium chloride . 3-1 cm. 



It will be seen that in the presence of sodium acetate 

 and potassium chloride the negative charges increase 

 100 and 50 per cent, respectively. The experiments 

 have been carried out under identical conditions. 

 In the case of sodium acetate the presence of hydroxyl 

 ions have to be taken into account. It will be seen, 

 however, that acetanions are adsorbed to a greater 

 extent than hydroxidions, and it is well known that 

 the more strongly adsorbed substance largely dis- 

 places the other which is not so strongly adsorbed. 

 In the case of potassium chloride the question of 

 hydrolysis does not arise. 



As is to be expected from the greater mobility of 

 the hydrogen ions and the views of the writer (Far. 

 Soc. Disc, Oct. 192 1, Phil. Mag. (VI.) 44, 330-37), 

 the acids of the same concentrations show a smaller 

 charge than their salts. The charge is, however, 

 undoubtedly greater than that with pure water, so 

 that there is unmistakable evidence of the adsorption 

 of anions, but owing to the effect of hydrogen ions the 

 charge indicates a smaller adsorption than is really 

 the fact. 



We are at present engaged in working with the 

 gels (which are likely to have greater specific surface) 

 and with chemically pure silica or silicic acid gel. 

 It appears that the electro-osmotic apparatus is also 

 capable of further improvements. 



These experiments clearly show that we are really 

 dealing with the kinetic exchange of ions (hydrogen 

 or Al" in the case of soil acidity) in the second sheet 

 of the double layer or present as electrically adsorbed, 

 as suggested by the writer. It is not necessary to 

 assume the hydrolysis of potassium chloride into 

 alkali and acid in water, or the displacement of 

 hvdrochloric acid from alkali chlorides by humus 

 acid, or the formation of insoluble salts of alkali 

 metals, as has been done in the past. 



[here is other corroborative evidence in support 

 of this point of view. J. N. Mukherjee. 



Physical Chemistry Department, 



University College of Science, Calcutta, 

 September 20. 



New Spectra of Water Vapour, Air, and Hydrogen 

 in the extreme Ultra-violet. 



After reading of the excellent work of Prof. Wood 

 on the extension of the Balmer series of hydrogen, 

 I decided to investigate the Lyman series of hydrogen 



