CHEMISTRY. (CHEMICAL ANALYSIS.) 



153 



produce good coke, nor will it yield much tar 

 and ammonia. If the analysis shows water of 

 constitution 5 to 10 per cent.", indicating the char- 

 acter of a lignite, the yield of by-products will 

 be small and the coke of poor quality; or, having 

 at hand the proximate analysis of a coal found 

 in the bituminous regions, one may judge very 

 clearly the yield of by-products per 2,000 pounds 

 of coal by referring to a chart subjoined to the 

 author's paper. 



In a paper contributed to the Chemical Society 

 of England, reporting progress in his investiga- 

 tions of Moorland waters and bearing upon their 

 acidity, William Ackroyd spoke of the impor- 

 tance of that quality, which has been associated 

 with plumbo-solvent action, inasmuch as moor- 

 land waters are consumed by more than 5,250,000 

 people in England. The author differentiated be- 

 tween organic and inorganic acidity, and de- 

 scribed a method of ascertaining the ammonia 

 in the water. This method consists of titration 



with alkali, using phenolphthalein as indi- 

 cator; and aspiration of air free from carbon 

 dioxide through another portion, and a second 

 titration for the residual acidity. Data were 

 given showing the loss of acidity due to diffusion 

 in observations extending over one hundred and 

 eighty-one days. Comparisons were made prov- 

 ing that the differences in acidity in waters from 

 various gathering grounds were due to differ- 

 ences of gradient. It was finally shown that the 

 organic acid in solution, always assumed to be 

 humic acid, is of lower equivalent than that of 

 the lowest humic acid known. In the discus- 

 sion of the paper Prof. Dewar said that the 

 plumbo-solvent action of the water supplied to 

 Sheffield, Bradford, and Huddersfield had been 

 overcome by the addition of finely precipitated 

 carbonate of lime to the reservoirs, with the re- 

 sult that the quality of the water was improved, 

 both by the removal of the acidity and also indi- 

 rectly by precipitation of the dissolved peaty 

 matter. 



From a long series of researches on specimens 

 of air taken from high mountain regions and 

 over the sea, M. Armand Gautier finds that free 

 hydrogen is present in the proportion of about 

 1.5 volume to 10,000 volumes of air. The pro- 

 portion of hydrogen is considerably greater in 

 the specimens of pure air than in specimens taken 

 from the neighborhood of towns, while the pro- 

 portion of combustible carbon compounds is con- 

 siderably less. 



The incandescent mantles for gas burners are 

 composed almost entirely of oxide of thorium, 

 ThO 2 , to which has been added a small quantity 

 of oxide of cerium, varying from 0.5 to 2 per 

 cent. With the exception of small quantities of 

 lime, they contain only traces of impurities, such 

 as nitrate of thorium and oxides of zirconium, 

 neodymium, lanthanum, and yttrium. By a series 

 of experiments with photometric measurements, 

 of which he gives the results in detail, E. Hintz 

 has satisfied himself that these impurities have 

 no appreciable influence on the luminous power 

 of the mantle, and that, therefore, not much no- 

 tice need be taken of them in the analysis of 

 incandescent bodies. 



Certain mineral waters exist, though they are 

 very rare, which are relatively very rich in fluor- 

 ine. The most interesting example is that of 

 the waters of Gerez, in the north of Portugal, 

 which are perhaps unique in Europe. An analy- 

 sis by Charles Lepierre determines the proportion 

 of alkaline fluorides (probably of sodium) as 

 from 22 to 25 milligrammes per litre, which is 



equivalent to from \Q to 12 milligrammes of com- 

 bined fluorine. The existence of fluorides in the 

 residue can be discovered positively by evaporat- 

 ing half a litre of Gerez water in a platinum dish. 

 The analytical detection of fluorine can be made 

 by the ordinary method of heating the residue 

 from the water with sulphuric acid and showing 

 the corrosion of glass; by Lannes's or Carnot's 

 method; if silica is present, by the formation of 

 SiF 4 ; or by the formation of crystals of fluosili- 

 cate of sodium, the little hexagonal prisms of 

 which are characteristic. The last reaction is 

 very sensitive. 



The committee of the British Association on 

 a uniform system of recording the results of the 

 chemical and bacterial examination of water and 

 sewage recommend in their report that results of 

 analysis be expressed in units per 100,000 except 

 in the case of dissolved gases, when these should 

 be stated as cubic centimetres of gas at C. and 

 760 millimetres in 1 litre of water. This is in 

 accord with the determination of a former com- 

 mittee of conference with the American Associa- 

 tion. The committee suggest that the results 

 in the case of all nitrogen compounds be ex- 

 pressed as parts of nitrogen over 100,000, includ- 

 ing the ammonia expelled on boiling with alka- 

 line permanganate, which should be termed al- 

 buminoid nitrogen. The nitrogen will therefore 

 be returned as ammoniacal nitrogen from free 

 and saline ammonia; nitrous nitrogen from ni- 

 trites; nitric nitrogen from nitrates; organic 

 nitrogen (either by Kjeldahl or by combustion, 

 but the process used shall be stated) ; albuminoid 

 nitrogen. The total nitrogen of all kinds will 

 be the sum of the first four determinations. The 

 committee are of opinion that the percentage of 

 nitrogen oxidized that is, the ratio of 2 and 

 3 to 1 and 4 gives sometimes a useful measure 

 of the stage of purification of a particular sample. 

 In raw sewage and in effluents containing sus- 

 pended matter it is also desirable to determine 

 how much of the organic nitrogen is present in the 

 suspended matter. The report further gives sug- 

 gestions concerning the manner of collecting 

 samples. 



M. Balland has communicated to the French 

 Academy of Sciences the results of his investi- 

 gations of the chemical composition and alimen- 

 tary value of the principal fruits, including 

 grapes, oranges, filberts, pomegranates, currants, 

 walnuts, figs, bananas, olives, dates, apricots, al- 

 monds, cherries, quinces, strawberries, raspberries, 

 medlars, peaches, pears, apples, and plums. All 

 these fruits contain when ripe from 72 to 92 

 per cent, of water. In the dried or partly dried 

 fruits of commerce the proportion of water rarely 

 exceeds 35 per cent., and is frequently below 10 

 per cent, in almonds and nuts. In pulpy fruits 

 the nitrogen represented by vegetable albumen 

 ranges from 0.25 per cent, in the pear to 1.45 per 

 cent, in the banana; in the seed fruits (almonds 

 and nuts) it is higher, amounting to from 15 to 

 20 per cent, in the dried nut. The fatty matters, 

 with all the products soluble in ether (essential 

 oils, resinous and coloring substances), are gen- 

 erally in smaller proportion than the nitrogenous 

 substances; the only exceptions are olives, al- 

 monds, and nuts, in which the oil predominates 

 (58 to 68 per cent.) when they are dry. The 

 ashes, some of which, as those of figs, pears, 

 and plums, show traces of manganese, are in- 

 considerable in quantity. So also is the inert 

 cellulose, which is in notable proportions only 

 in quinces and medlars. Acidity reaches its maxi- 

 mum in currants and raspberries (1.25 per cent.). 

 Sugar and the substances called extractive, such 



