CHEMISTRY. 



127 



posed being very different at different places ; 

 and the fact that absorption} etric exchange had 

 not gone far enough to reproduce equilibrium 

 would account for the few cases in which the 

 dissolved gases exceeded the amount calculated. 



W. N. Hartley has made experiments in the 

 application of spectrum photography to quan- 

 titative analysis in the case of a number of 

 metals. The sensitiveness of the spectrum re- 

 action, he reports, is practically unlimited when 

 applied to magnesium compounds dissolved in 

 water. With a given length of spark, A60 ^ 0()60 

 of a milligramme of that metal could easily be 

 detected. When the strength of the spark was 

 greatly increased, but the striking distance be- 

 tween the electrodes was left unaltered, the 

 sensitiveness was increased ten thousand fold. 

 One part of magnesium was detected in 10,- 

 000,000,000 parts of water. The spectrum re- 

 action of arsenic is the weakest, and those of 

 antimony and tellurium are weak, while that 

 of bismuth is not strong. Evidence is afforded 

 in the case of the aluminum spectrum that it 

 is not invariably the longest or strongest line 

 that is most persistent. Even the longest lines 

 are shortened by great dilution of the solutions, 

 but there is a pair of lines in the spectrum of 

 copper which become greatly attenuated, yet 

 remain long lines till they disappear. 



Alfred Smetham has reported the results of 

 an analysis of ensilage taken from the lower 

 part of a silo about nine feet in depth, some 

 ten days after the grass was put in. Samples 

 were taken from between one foot and one 

 foot six inches, and from between one foot six 

 inches and two feet from the bottom. A sample 

 from the bottom layer was also examined, to 

 determine to what extent the excess of water 

 had influenced the quality and the nature of the 

 fermentation. An increase of mineral matters 

 was remarked, which was due to the salt that 

 had been used in preparing the silo. A loss of 

 about six per cent, of nitrogenous compounds 

 was observed as compared with the grass, but 

 a slight gain in nitrogen as compared with the 

 same grass made into hay. It was made clear 

 that the conversion of indigestible fiber into 

 soluble compounds which has been claimed for 

 ensilage has been overestimated, and that the 

 loss of substance by fermentation has, by many, 

 been exaggerated. Among the more impor- 

 tant results of the investigation is the direct 

 proof that in ensilage a considerable portion of 

 the insoluble albuminoids is rendered soluble. 



Charles W. Dabney, Jr., and B. von Herff, 

 of the North Carolina Experiment Station, re- 

 port upon the comparative value of the " Ruffle 

 method " and of the " copper oxide " or " abso- 

 lute " method for the determination of nitro- 

 gen, as ascertained in their experiments. They 

 have worked the Ruffle method very nearly as 

 described by its author, always using Bohe- 

 mian glass tubing, but have not tried it with- 

 out soda or without hypophosphite, as had been 

 suggested ; this method has the advantage of 

 being simpler, cheaper, and somewhat quicker 



than the copper-oxide method. There are 

 fewer risks to run, fewer combustion-tubes 

 break, and for several reasons a much larger 

 percentage of complete analyses are obtained by 

 it. In using the copper-oxide method, the au- 

 thors followed in the main the course described 

 by Prof. Johnson, but without finding it neces- 

 sary to pass oxygen ; so they could omit the 

 chlorate of potash from the end of the tube. In 

 analyzing compounds rich in nitrogen, and es- 

 pecially those containing much nitrate, the addi- 

 tion of charcoal-powder to the substance causes 

 the nitrogen to come off more regularly, and 

 gives generally better results. Excellent re- 

 sults were obtained in jetting the air before 

 combustion, and the nitrogen afterward, out 

 of the tube, by using carbonic dioxide, without 

 a pump, though that required more time than 

 when a good pump was used. Magnesite or 

 carbonate of manganese, put in the back end ot 

 the tube, are the best sources of carbon dioxide 

 for this purpose. Bicarbonate of soda can be 

 used also. A combination of this method with 

 a good pump is effective and quickens the pro- 

 cess. The authors have devised an improved 

 pump for the purpose. Their analyses as a 

 whole show that in fertilizers containing small 

 amounts of nitrogen, the Ruffle method and the 

 copper-oxide method give equally good results. 

 By the copper-oxide method the errors are apt 

 to be in the direction of too much, by the Ruf- 

 fle method in the direction of too little. 



For the determination of very small quanti- 

 ties of silver, Carl Friedrich Fohr uses a com- 

 bination of the ordinary dry process and the 

 blowpipe test. His method does not require 

 any more time than an ordinary assay, and the 

 determination by it may be completed in four 

 and a half or five hours. It is accurate for 

 portions of silver that only form fractions of a 

 thousandth per cent., and is especially suitable 

 for very poor silicates. 



Agricultural Chemistry. The origin of the dark- 

 green circles of grass alternating in some sea- 

 sons with fungi, which are called "fairy rings," 

 after having been a subject of inquiry since it 

 was brought under attention in the Royal Soci- 

 ety's "Transactions," has received a new expla- 

 nation from the studies of Messrs. Lawes, Gil- 

 bert, and Warrington, of Rothamsted. Believing 

 that the growth of the fungi was owing to some 

 extraordinary power they had of assimilating 

 nitrogen either from the atmosphere or from 

 the soil, the authors tried direct experiments to 

 determine the question. Samples of soil were 

 taken from within a fairy ring, from immediate- 

 ly upon it, and from outside of it. Of these 

 specimens the soil from within the ring yielded 

 upon analysis the lowest percentage of nitrogen, 

 that from the ring itself a higher percentage, 

 and that from outside of the ring a higher per- 

 centage still. The soil had therefore lost nitro- 

 gen by the growth of the fungi, and the obvious 

 conclusion was that the fungi possess a greater 

 power than the grasses of abstracting nitrogen 

 from the soil. The analyses of the various 



