138 



NATURE 



[February 2, 1922 



March, 1788, he tried the experiment of fitting a 

 retort containing caustic volatile alkali to a gun-barrel 

 filled with crushed pyrolusite (maganese dioxide), and 

 heating the latter to redness, whilst the retort was 

 also heated. Signs of nitrous acid and nitrous air 

 soon made themselves manifest, and by continuing 

 long enough nitrous gas was obtained. The experi- 

 ment was repeated many times ; its success depended 

 on the nature of the pyrolusite, the temperature of 

 the furnace, and the patience of the experimenter. 

 Full details are given as to the best way of carrying 

 out the experiment. It frequently happens that the 

 ammonia passes over unchanged. Red lead was 

 found, unexpectedly, not to be active, but green vitriol 

 burnt white gave better results. 



The changes are correctly explained by Milner as 

 due to oxidation. With burnt alum he obtained the 

 curious result of the evolution of a large amount of 

 inflaminable gas mixed with hepatic air (sulphuretted 

 hydrogen) and sulphur, whilst sulphur remained in 

 the gun-barrel. It is, therefore, not sufficient merely 

 to bring the volatile alkali in contact with a substance 

 containing dephlogisticated air, but another substance 

 is also necessary which has a strong attraction for 

 the combustible substance. 



It is also noteworthy that Black in his " Lectures 

 on Chemistry " (edited by John Robison, Edinburgh, 

 1803) states that "our newspapers inform us that 

 the French chemists procured saltpetre for the Army 

 by blowing alkaline gas, and even putrid steams, 

 through red-hot substances which readily yield 

 oxygen " (vol. 2, p. 245); and there is a statement that 

 " Mr. Milner of Oxford {sic) published a paper in the 

 79th volume of the Philosophical Transactions . . . 

 but he did not attempt to ascertain how much of the 

 nitrous acid might be produced from a limited and 

 known quantity of the volatile alkali " (vol. 2, p. 455). 

 Black {ibid.) gives a clear explanation of the process ; 

 the ammonia " is a compound of hydrogen and azote, 

 we need only suppose that part of it is totally decom- 

 posed and destroyed by the action of the oxygen con- 

 tained in the manganese. Part of it, uniting with 

 the hydrogen, forms water or watery vapour; and 

 part, uniting with the azote, forms vapours of nitrous 

 acid." I have not traced the reference to the "news- 

 papers," but a footnote on the same page (455) 

 reads: '' Jamiary, 1796. There is a rumour that the 

 French have manufactured saltpetre, during a part 

 of the war, by obtaining nitrous acid from the 

 vapours of volatile alkali, forced to pass through red 

 hot manganese. Author." 



Many strange names have of late been given to the 

 process of ammonia oxidation ; we have heard of the 

 " Ostwald-Mittasch process " and others. The first 

 use of platinum as a catalyst appears to be due to 

 Kuhlmann. of Lille, in 1839. J. R. Partington. 



45 Kensington Gardens Square, W.2. 



A Specimen of Wrought-iron Currency from the Kisi 

 Country, Sierra Leone Protectorate, West Africa. 



A SPECIMEN of iron currency from the Kisi country 

 was_ obtained by one of us' (E. R. M.) while on 

 service in West Africa in 1915 through the agency 

 of his servant, Ali Badara, the son of a chief in the 

 adjoining Momo-Fullah country ; and a description 

 of it may be of interest to readers of Nature. 



As this form of currency ceased to be used after 

 the establishment of the British Protectorate in 1787, 

 the age of the specimen may be estimated at not less 

 than 130 years, and probably more. 



The "coin " (Fig. i) is of rough workmanship, and 

 consists of a strip of roughly forged rectangular sec- 

 NO. 2727, VOL. 109] 



tion, one half being twisted and the ends hammered 

 out into thin blade-like projections, the broad end 

 serving to prevent the "coin" slipping through the 

 belt in which it is carried. 



Analysis shows the metal to be wrought-iron of 

 good quality, probably made by the direct process of 

 reducing an oxide ore by carbon in presence of a 

 basic slag containing much iron oxide to prevent car- 

 bur isation of the iron, most of the slag then being 



k 



B A 



Fig. I.- -Photomicrograph X 1/5. 



expressed by hammering the pasty mixture of iron 

 and slag. The percentage composition is as follows : 

 Carbon, 0-095; silicon, 0103; manganese, nil; sul- 

 phur, 0024 ; and phosphorus, 0-046. For the analysis 

 drillings were taken from the wider part of the speci- 

 men and fragments from the narrow end. These 

 were washed in benzene to remove the coating of 

 black grease from the surface of the metal. 



The metal is extremely soft and easily bent, the 

 Brinell hardness at the point A being 121 (using a 



Fig. a. — Photomiciograph X 275. 



ball I mm. diameter and a load of 10 kilograms). A 

 small fragment was cut from the point B, em- 

 bedded in solder, polished, etched with 2 per cent, 

 nitric acid in alcohol, and photomicrographed. The 

 photomicrograph (Fig. 2) shows the typical crystalline 

 structure of a pure iron, together with elongated 

 inclusions of slag. 



R. C. Gale. 



E. R. Macpherson. 

 Chemistry and Metallurgy Branch, 

 Artillery College, Woolwich. 



Molecular Structure of Amorphous Solids. 



, A QUESTION of fundamental impoitance in the theory 

 of the solid state is the nature of the arrangement 

 of the ultimate particles in amorphous or vitreous 

 bodies, of which glass is the most familiar example. 

 Is it to be supposed that the moiiecules are packed 



