152 



CHEMISTRY. (CHEMICAL ANALYSIS.) 



advantage of making possible the preparation of 

 pure metals free from aluminum. The process can 

 be applied simply as a source of heat for weld- 

 ing, soldering, etc., or, by utilizing the reducing 

 power of aluminum, in the preparation of pure 

 metals and of alloys. If only a moderate heat 

 is desired, the reacting mass may be diluted by 

 the addition of some inert mixture, as of alumi- 

 num and some cheap oxide, to which may be 

 added lime, magnesia, etc. Nearly all the metals 

 can be reduced by this method. 



Ferrocvanide of potassium is recommended by 

 M. Lucas as a test for the estimation of copper, 

 with the solution of which it gives a reddish- 

 brown coloration. If the solution is weak, the 

 precipitate will cool very slowly, and the solution 

 will be left clear but red. This coloration is rep- 

 resented to be ten times more sensitive than that 

 given by ammonia, and it can be utilized for the 

 estimation of small quantities of copper. 



Atomic Weights. The place of the newly 

 discovered elements of the atmosphere in the 

 periodic system has been studied by James Lewis 

 Howe, of Washington and Lee University. With- 

 out coming to a positive conclusion, the author 

 gives reasons for regarding the three elements 

 argon, neon, and helium as transition elements 

 between the halogens and the alkali metals, and 

 finds that there is sufficient space for them, as 

 well as for the other new elements, krypton, met- 

 argon, and xenon, in Group VIII the group of 

 iron, cobalt, and the platinum metals. An addi- 

 tional point in favor of the view that helium, 

 neon, and argon are true elements, implicitly 

 brought out by Ramsay in his address before the 

 Deutscher chemischen Gesellschaft, is furnished 

 by Dobereiner's law of triads. It was this idea 

 that led Ramsay to his search for neon. " Like 

 so many other groups of elements, the three ele- 

 ments helium, neon, and argon agree with the 

 relations pointed out so long ago by Dobereiner, 

 and this indicates that the elemental nature of 

 these gases is not different from that of other 

 elements." 



T. W. Richards and A. S. Cushman, after a 

 careful revision of the atomic weight of nickel 

 as obtained in different determinations, and com- 

 parison particularly with values found by Winkler 

 and Zimmermann, give the number as 58.706, and 

 as the average of their own determinations and 

 those of the two other authors 58.70. 



From experiments with the carbide and sul- 

 phide, M. Henri Gautier had determined the equiv- 

 alent of boron as very near 11. In subsequent 

 experiments, using the bromide and chloride, 

 which had been prepared in as perfect purity as 

 possible, and treating them with nitrate of sil- 

 ver, he obtained the number ll.01.fi. 



Mr. Harry C. Jones has called attention to the 

 agreement between the values of the atomic 

 weight of praseodymium and n2odymium found 

 by three experimenters Carl von Schele, Braun- 

 er, and himself working simultaneously but in- 

 dependently, and to the difference between their 

 results and those of Von Welsbach, which were 

 the only determinations on record to 1897. 

 Brauner and Jones used American material for 

 their determinations, and Schele monazite fur- 

 nished by Cleve. Thfe values obtained by Wels- 

 bach were 143.6 for the atomic weight of praseo- 

 dymium and 140.8 for that of neodymium. Those 

 obtained by Jones, with which those of Brauner 

 and Von Schele very nearly agree, were 140.5 for 

 praseodymium and 143.6 for neodymium. " The 

 atomic weight of praseodymium found by Von 

 Schele an/1 myself," says Mr. Jones, " is a little 

 lower than Von Welsbach's atomic weight of neo- 



dymium, but the atomic weight of neodymium 

 as found by Brauner and myself is identical with 

 Welsbach's atomic Weight of praseodymium. 

 These three pieces of work, then, make it cer- 

 tain that the original determinations of Von 

 Welsbach are not free from error; and, indeed, 

 they were published as only tentative, and are 

 not described in any detail." 



Chemical Analysis. Prof. J. Vertress calls 

 attention to some difficulties attending lighting 

 by acetylene, among which are the impurities in 

 the calcium carbide from which it is derived, and 

 the want of homogeneity of the substance, where- 

 by it is necessary to test and examine several 

 samples in order to get a mean value. Among 

 these impurities are sulphur, phosphorus, and 

 nitrogen, from which it results that the acetylene 

 \vill be contaminated with sulphureted hydrogen, 

 phosphureted hydrogen, and ammonia. It must 

 therefore be purified to the same extent as coal 

 gas, for fear that its use in closed places might 

 lead to accidents. Another drawback to the use 

 of acetylene is its liability, after burning a con- 

 siderable number of hours, to give a smoky flame. 

 This is caused by the burners attaining a tem- 

 perature higher than that of the decomposition 

 of acetylene, whereby the gas is decomposed into 

 carbon and hydrogen. The author has also ob- 

 served deposits of finely divided carbonlike soot 

 in acetylene-gas pipes and a liquid condensation 

 of carbides of hydrogen in the generators, \vhich 

 has to be removed by siphons. Still another in- 

 convenience is the formation of a fog in closed 

 places after a longer or shorter interval, which 

 is the result of decomposition of the gas, with 

 deposition of the carbon and the formation of 

 watery and ether vapors. As the presence of 

 hardly noticeable quantities of phosphureted hy- 

 drogen may make the acetylene spontaneously 

 inflammable, a rapid and exact method of gas 

 analysis is needed to estimate the quantity of that 

 gas that may be present. For this purpose 

 Walther Hempel and Leopold Kahl, after experi- 

 menting with a number of reagents, recommend 

 sulphate of copper in sulphuric solution as the 

 best. The gas is measured in a gas burette filled 

 with mercury, and is transferred thence to an- 

 other burette filled with mercury and 3 cubic 

 centimetres of cupric reagent saturated with 

 acetylene. After shaking for three minutes, the 

 volume of unabsorbed gas is measured; a quar- 

 ter of this volume represents the phosphureted 

 hydrogen present. 



The by-products obtained from the distillation 

 of coal in retort coke ovens, as described by Mr. 

 J. D. Pennock, are tar^ ammonia, and gas, from 

 w r hich benzine and cyanides may be recovered. 

 All the by-products vary in quality and quantity 

 with the composition of the coal, and are also 

 more or less affected by the temperature to which 

 the oven is heated. Generally speaking, the quan- 

 tity of tar and ammonia obtained is in direct 

 proportion to the percentage of volatile matter 

 in the coal. A bituminous coal containing 17 

 per cent, of volatile matter will yield l3i pounds 

 of ammonia, figured as sulphate, and 30 pounds 

 of tar, whereas one containing 37 per cent, of 

 volatile matter will yield 27 pounds of ammoni- 

 um sulphate and 110 pounds of tar. From the 

 large number of ultimate analyses which the 

 author has made of coals experimented upon in 

 ovens, he finds that coals having the highest per- 

 centage of hydrogen over and above that neces- 

 sary to combine with the oxygen are the ones 

 that make the best coke, and yield the best 

 quantity of by-products; in other words, a coal 

 containing a high percentage of oxygen will not 



