92 



CHEMISTRY. 



vantages and disadvantages of the processes in 

 common use, viz., the combustion process, the 

 ammonia process, and the oxygen process. The 

 combustion process is declared to u yield abso- 

 lutely untrustworthy evidence on which to 

 found an opinion as to the probable source of 

 the organic matter." The ammonia process 

 furnishes results which are not delicate enough 

 to allow the recognition of the finer grades of 

 purity or impurity. But of the oxygen pro- 

 cess the author says that its results are con- 

 stant and extremely delicate, and that it draws 

 a sharp line between putrescent and probably 

 harmless organic matter (a point of the first 

 importance here). By it bad water could never 

 be passed as good. But it is only when the 

 process is properly carried out that such pre- 

 cise results are to be expected. The proper 

 plan of using the oxygen process is as follows: 

 In two carefully cleaned twenty-ounce flasks place 

 500 septems (a septem = 7 grains, or yoVo gallon) of 

 the water ; to each add 20 septems of dilute (1 in 3) 

 sulphuric acid, and 20 septems of solution of potas- 

 sic permanganate (2 grains per 1,000 septems). At 

 the same time treat two similar quantities of dis- 

 tilled water in precisely the same manner, and note 

 the exact time at which the permanganate solution 

 was added. The oxygen used up by the water must 

 be determined at the end of one hour and of three 

 hours. To the flasks, after standing the appointed 

 time, add a sufficiency of potassic iodide (1 in 10), 

 and then a standard solution of sodic hyposulphite 

 5'4 grains per 1,000 septems), until all of the free 

 iodine is removed (to be determined by adding a few 

 drops of starch solution). By deducting the quantity 

 of oxygen equivalent to the hypo-solution used from 

 that in the quantity of permanganate originally add- 

 ed, we obtain the quantity of oxygen used by the 

 water. The blank experiments with distilled water 

 give the value of the hypo-solution. It is obvious 

 that the samples of water must have a pink tint at 

 the end of the one hour or of the three hours ; other- 

 wise fresh experiments are to be made with larger 

 doses of permanganate. The only important errors 

 that can arise would be due to the presence of fer- 

 rous salts, sulphuretted hydrogen, and nitrites. But 

 the first two would be discovered in the analysis, 

 and by their taste and smell ; as for the nitrites, 

 they act immediately on the permanganate solution, 

 and any decolorization taking place during the first 

 five minutes must be due to nitrites and allowed for. 

 It is admitted that permanganate fails to oxidize some 

 substances, as urea; but nevertheless the quantity 

 of oxygen used gives evidence of the relative quan- 

 tity of matter in the water which is likely to be in- 

 jurious; and this is precisely the one great object 

 of water- analysis. The quantity of oxygen used dur- 

 ing the first hour as compared with that used in the 

 first three hours gives valuable information as to the 

 relative quantities of putrescent, easily oxidizable 

 matter, and of non-putrescent and less easily oxidiz- 

 able matters. 



The Temperature of Flames. Rosetti has, 

 with the aid of his very ingenious calorimeter, 

 investigated the temperature of different flames, 

 and finds the maximum temperature of a Bun- 

 sen flame to be 1,360 C., resulting from a com- 

 bustion of 1 volume of gas and 2| volumes of 

 air. The admission of a greater or less quan- 

 tity of air reduces the temperature. Changes 

 in pressure have but slight influence on the 

 temperature. The flame given by gas diluted 

 with the same volume of nitrogen shows a 



temperature of 1,180, and diluted with 3 vol- 

 umes of nitrogen, 1,040. The same degrees 

 of dilution with carbonic acid show respec- 

 tively 1,100 and V80. Among other tempera- 

 tures noted were the following: 



Locatelli lamp 920 



Stearin e candle 940 



Petroleum lamp with chimney 1,030 



The same without chimney : 



Illuminating- part 920 



Booty envelope 780 



Alcohol lamp (alcohol 0'912) 1,170 



(alcohol 0-822) 1,180 



The slight difference in heating power result- 

 ing from widely varying percentages of water 

 in the alcohol is worthy of remark. 



A New Test for Carbolic Acid, A very deli- 

 cate test for carbolic acid has been discovered 

 by Dr. E. W. Davy in molybdic acid dissolved 

 in strong sulphuric acid. When a drop or two 

 of a dilute aqueous solution of carbonic acid 

 is brought in contact with a few drops of the 

 molybdic solution, there is immediately pro- 

 duced a light yellow or yellowish-brown tint, 

 which, passing to a maroon or reddish brown, 

 soon develops a beautiful purple coloration. 

 The application of a gentle heat will hasten 

 the development of the purple reaction ; but 

 it will take place, though more slowly, at the 

 ordinary temperature. It is the production of 

 this purple under the circumstances described 

 that constitutes Dr. Davy's test for carbolic 

 acid. The molybdic solution is made by dis- 

 solving, with the assistance of a gentle heat, 1 

 part of molybdic acid in 10 parts by weight of 

 pure concentrated sulphuric acid. The mode 

 of using this reagent is simply to add three or 

 four drops of it to one or two of the liquid 

 under examination, placed on any white por- 

 celain or white delf surface. In carrying out 

 this test, however, it will be found most con- 

 venient to use a small white porcelain capsule 

 having a handle, which will admit of the ap- 

 plication of heat when it may be desirable to 

 hasten the reaction of that agent. So delicate 

 is this test that one small drop of an aqueous 

 solution of carbolic acid (1 per 1,000 by weight), 

 when mixed with three or four drops of the 

 molybdic solution, immediately produces the 

 yellowish-brown effect, which soon passes into 

 a very distinct purple. Nor is this the extreme 

 limit of its application, for the solution affords 

 a certain test even when five times more dilute 

 than above. 



Determination of the Heat Value of Fuel. 

 With regard to the question of the heat value 

 of fuel, it has been proved that conclusions from 

 the results of elementary analysis are very 

 uncertain, and also that little reliance can be 

 placed on direct e\ 7 aporation experiments. The 

 faults of these methods are pointed out by 

 Weyl, who in " Die Chemische Industrie " 

 recommends, as preferable, decomposition of 

 the fuel by dry distillation, and analytical de- 

 termination of the solid, liquid, and gaseous 

 products of decomposition. In this method 

 the accident of too small a sample being used 

 is avoided, as also too great pulverization and 



