by evaporation, and partly also by its destruction by a more profound oxidation. There was also imparted to the wine a perfume similar to that acquired by maturing; this increased by prolongation of the current. The passage of the current also assists its future preservation. The colouring matters are affected, but the author would not for the present draw any conclusions from his experiments. If by passage of an electric current the wine could be sterilised, so as to effect the complete removal of Bacterium aceti, this process would be of considerable technical importance. V. H. V.

Manuring Barley. By v. LIEBENBERG (Bied. Centr., 1887, 649— 651). The experiments were conducted at 12 stations, and the manures employed were (1) Chili saltpetre, (2) Chili saltpetre and soluble phosphate, and (3) Chili saltpetre, soluble phosphate and potassium sulphate. The spring of the year (1886) was dry, whilst June and July were wet. There were three failures; of the remaining nine stations, the nitrogen showed itself active in producing corn and straw at seven stations, whilst an influence on the straw only was manifested at two. In two cases, potash increased the corn and straw, whilst in one it acted on the straw only; the phosphate in five cases affected the corn and straw, and in one the straw only. It seems then that nitrogen is available for barley in all localities, whilst phosphates or potash are less so. E. W. P.

Manuring Oats. By v. LIEBENBERG (Bied. Centr., 1887, 651653). The experiments were conducted on the same principle and in the same manner as those on barley manuring. Nitrogen, and nitrogen combined with phosphate yielded an increase, but there is an uncertainty as to the gain produced by potash. Phosphates influence oats more than barley, because the former is slower and later in coming to maturity. E. W. P.

Comparative Experiments with Oats manured with Basic Slag on Moorlands. By BAESSLER (Bied. Centr., 1887, 653-655). -The soil on which the experiments were conducted contained 69 per cent. organic matter, 27 N, 0.29 P2O, and 13.66 per cent. ash. Kainite, basic slag, and superphosphates were used. Neither the super nor the slag produced any decided effects; kainite increased the yield of straw, but a general increase was remarked when the phosphates in either form were combined with kainite. The action of the slag was only 42 per cent. of that of the superphosphate, but this difference is compensated for by the lower price of the slag.

E. W. P.

Manuring Winter Wheat and Winter Rye. By v. LIEBENBERG (Bied. Centr., 1887, 656-658).—The manures were the same as those employed in the author's experiments on oats and barley. No decided nor especial advantage was remarked in the crop of wheat thus manured. As regards the rye, nitrogen alone produced no effect, but when combined with phosphates or kainite there was a very decided increase. The author accounts for the absence of effect

from the use of nitrogen by the fact that the land had previously received farmyard manure. E. W. P.

Manuring of Vines. By J. MORITZ and P. SEUCKER (Bied. Centr., 1887, 604-609).-The experiments were instituted to determine if artificial manures could not be used for vines instead of the expensive farmyard manure. The manures were those usually employed, namely, ammonia superphosphates and potash compounds, and peat. The analyses of the must as well as its total quantity show that farmyard manure can be replaced by other manures; mineral manures seem to prevent the usual premature ripening, and consequently rotting, of some of the berries on the vines, whilst farmyard manure induces an extra growth of weeds. These experiments have been carried on for nine years, and are still being continued.

E. W. P.

Condition of Potassium in Soils, Plants, and Moulds. By BERTHELOT and ANDRÉ (Compt. rend., 105, 833-840; 911-914).1 kilo. of the dried earth contained 235 grams organic carbon, 166 gram nitrogen at the beginning of the season and 1·73 gram. and the end, and 8.92 grams of potassium. The only methods available for the estimation of the total potassium are those in which the soil is completely decomposed by ammonium fluoride, calcium carbonate, or barium hydroxide.

Water percolating through the soil removed only 0.0029 gram of potassium per kilo., but if the soil was treated with successive quantities of water the amount dissolved increased to 0·143 gram per kilo. Heating the soil to dull redness had little effect on the solubility of the potassium.

Ethyl acetate and ammonia have practically no effect on the solubility, but water containing 2 per cent. of sugar or acetamide dissolves more potassium compound than pure water. Water saturated with carbonic anhydride dissolved 0198 gram per litre; water containing 2 per cent. of acetic acid dissolved 0.290 gram per litre; water with 2 per cent. of hydrochloric acid 0.404 gram; water with 2 per cent. nitric acid 0.296 gram. These results afford no definite evidence either as to the forms in which the potassium exists in the soil or as to the proportion assimilable by plants.

Concentrated hydrochloric and nitric acids at a higher temperature dissolve considerably more potassium, the exact quantity depending on the temperature and time of action. The whole of the potassium, however, is never removed. Calcination of the soil increases the proportion of the potassium soluble in acids, probably in consequence of some alteration of the silicates.

It is not possible to draw any sharp line of demarcation between the potassium which is soluble or insoluble, assimilable or nonassimilable.

In order to obtain similar information respecting the potassium present in living plants, Mercurialis annua was selected as an example. The dried plant contained per kilo. 19.35 grams of nitrogen and 27.87 grams of potassium oxide, and left 125 grams of ash. When treated with 10 times its weight of water for 24 hours,

18.92 grams of potassium per kilo., or two-thirds of the total amount, was dissolved. Hydrochloric acid of 2 to 3 per cent. dissolved 24:58 grams of potassium in 24 hours, or 5.66 grams more than water. The remaining 3.29 grams was present in some form insoluble in dilute acids.

A sample of vegetable mould which was examined in a similar way contained 324 per cent. of water and lost 45.9 per cent. when calcined in presence of air. The dried mould contained per kilo. 95'8 grams of organic carbon; 145 grams combined carbonic anhydride; 8.6 grams of nitrogen; and 11.65 grams of potassium. 2.96 grams or onequarter of the potassium was soluble in water; 5.84 grams was soluble in hydrochloric acid of 2 per cent., but even after incineration the quantity of potassium insoluble in acids was 2:14 grams per kilo. The proportion soluble in water was considerably less after incineration, owing doubtless to the action of the silicates on the potassium carbonate.

It is evident that the mould does not retain the whole of the potassium present in the original plant, a considerable proportion being removed by rain during the process of decay. The proportion remaining is, however, still much more considerable than in the soil, and mould is a truly complementary manure intermediate in its character between the organic and inorganic manures. C. H. B.

was

Best Time for Ploughing Yellow Lupines under. By BAESSLER (Bied. Centr., 1887, 615-618).—The periods into which the growth of the yellow lupine which was to be used as green manure divided, were (1) full bloom of main stem, (2) commencement of podding of the same, (3) full bloom of side shoots, (4) full ripeness of the pods of main stem. It was found that the plants should be ploughed under in the fourth period, for then they would give to the morgen N 1403 kilos., K2O 53′96, and P2O, 25·7, when 253,440 plants are allowed per morgen; at this period, the nitrogen and phosphoric acid are three times and the potash twice that found in the plant in the period of bloom. E. W. P.

Analysis of Rubbish-heaps employed to Improve Soils. By A. MAYER (Bied. Centr., 1887, 577-578).-The rubbish-heaps arc found at places of refuge which were frequented in times of flooding before Holland was protected by dykes; they consist largely of animal and vegetable remains. The analysis of one sample must here suffice: Organic matter, 60 per cent.; total nitrogen, 014; easily soluble nitrogen, 0·02; phosphoric acid, 0.78; potash, 034; calcium carbonate, 1·1. The material is richer in plant food than the various "muds" used for amelioration of poor land; in the former, the lowest quantity of phosphoric acid is higher than the highest percentage in the latter; the same may be said for potash and nitrogen, but the muds contain more calcium carbonate.

E. W P.

Analytical Chemistry.

Support for Funnels while Drying. By V. MEURER (Zeit. anal. Chem., 26, 614).-Two horizontal, parallel bars of glass tube are supported by pieces of glass rod, which have their ends bent upwards till they nearly meet, and thrust into the ends of the tubes which are bent downwards at right angles. A small hook of glass rod prevents the bars from springing apart under the weight of the funnels which rest between them. M. J. S.

Moisture remaining in a Gas after drying by Phosphoric Anhydride. By E. W. MORLEY (Amer. J. Sci., 34, 199-204).— The method employed for determining the amount of aqueous vapour left in a gas after drying with phosphoric anhydride, consists in drying the gas with that substance, and then passing it through a weighed apparatus in which the gas is first slightly moistened, then much. expanded, and, lastly, again dried by phosphoric anhydride. The decrease in weight of the apparatus is then due to the moisture left by phosphoric anhydride in that volume by which the gas passing out of the apparatus exceeds the gas entering it. In this way, it was found that the moisture left unabsorbed may be roughly estimated at a fourth of a milligram in 10,000 litres of gas. (For the result with sulphuric acid see Abstr., 1886, 278.) B. H. B.

Hygienic Air Analysis. By K. SONDÉN (Zeit. anal. Chem., 26, 592-598). The apparatus is essentially that of Pettersson (Abstr., 1887, 999; also 179 and 180), but with the inlet tube so arranged as to draw the sample of air for analysis from a pipette in which it has been collected, thus avoiding the necessity of performing the analysis in the locality from which the air is taken. The carbonic anhydride only is determined, the gases being saturated with moisture before each measurement. Results agreeing closely among themselves and fairly with determinations by Pettenkofer's method have been obtained with an apparatus containing only 18 c.c. of air.

M. J. S.

Estimation of Sulphurous Acid by Standard Iodine Solution. By J. VOLHARD (Annalen, 242, 93—113).—The drawback to Bunsen's volumetric method, involving the use of standard iodine solution and dilute sulphurous acid, is the fact that the acid solution will not be completely oxidised if it contains more than 0.04 per cent. of SO2. The incomplete oxidation of the sulphurous acid in stronger solutions is generally attributed to the action of the sulphuric acid on the hydriodic acid :-H2SO1 + 2H! = I2 + H2SO3 + H2O. The author finds that the true explanation is, that sulphurous acid is decomposed by a strong solution of hydriodic acid, yielding sulphur and iodine; the iodine at once oxidises sulphurous acid to sulphuric acid and is itself converted into hydriodic acid; this secondary action may be avoided by adding the moderately dilute sulphurous acid to the standard iodine solution. W. C. W.

Rapid Method of Determining the Total Acidity in Flue Gases from Vitriol Chambers, adapted for the use of Workmen. By W. YOUNGER (J. Soc. Chem. Ind., 6, 347-348).-The author recommends the use of Hurter's apparatus, which is described in detail in Wanklyn's book on "Gas Analysis." The method being intended only for the use of foremen, to serve as a guide in the working of chambers, no pretensions are made to any great degree of accuracy. The most useful property of the apparatus is that it indicates how far the combustion is carried on, what fuel is used, and how much is wasted. D. B.

Soda-lime Method for Determining Nitrogen. By W. O. ATWATER and C. D. WOODS (Amer. Chem. J., 9, 311-324).—The soda-lime is prepared by slaking 2 kilos. of lime in an iron kettle with a solution of 1 kilo. of sodium hydroxide, evaporating, and heating to fusion. It is ground whilst warm, and divided into two portions by sieves of mm. and 2 mm. This soda-lime is, on the whole, preferable to a more fusible one containing more soda, the complete decomposition of the substance being effected by the large surface presented by the coarse soda-lime with which the front part of the tube is filled. The same results are obtained whether the lime contains much or little soda, or even none, the ammonia probably being produced by the action of superheated steam. Combustions with sugar are not to be recommended for testing the purity of soda-lime, as even when recrystallised from alcohol it appears to contain a trace of nitrogen; stearic acid or oxalic acid is to be preferred. For the filling of the absorption-bulb, it is convenient to have a little thistle funnel blown on the end of the exit tube. If the tubes are well and closely packed with granular soda-lime, so as to leave no considerable open channel, and if a sufficiency of powdered soda-lime be well mixed with the substance, short tubes, 30 cm. long, may be used; cochineal is used as indicator.

When using the absolute method for the determination of nitrogen, a small correction should be made for the residual air and for the vapour-tension of the potash solution, namely, when at sp. gr. 1·36, 4 mm. at 9.5° and 6.5 mm. at 14.5°.

Although the results obtained by the two methods are almost the same, the soda-lime method is not quite so exact, but is much simpler, and sufficient for most purposes if the above precautions are observed. H. B.

Determination of Nitrogen by Kjeldahl's Method. By L. LENZ (Zeit. anal. Chem., 26, 590-592).—To ascertain whether the addition of permanganate is invariably necessary, comparative determinations were made on 11 nitrogenous substances, containing from 1-4 to 14 per cent. of nitrogen. In every case, the use of permanganate gave the higher result, the difference varying from 0.28 to 16 per cent. of the whole. The permanganate cannot, therefore, in any case be safely dispensed with. M. J. S.

Apparatus for Nitric Acid Determination. By KRATSCHMER (Zeit. anal. Chem., 26, 608-610).-A simple apparatus for Schlösing's method. A flask of 150 c.c. capacity is fitted with a caout

VOL. LIV.

0