VEGETABLE PHYSIOLOGY AND XQRICULTURE. 67 Chemistry of Vegetable Physiology and Agriculture. Action of Micro-organisms on cert -tin Colouring Matters. By J. RAULIN (Corn@. rend., 107, 4B.;-447).-A.~1eergillus niger will grow in cultivation liquids containing ammonium nitrate, but not in similar liquids containing salts of aniline, rosaniline or indigo- carmine. Slightly acidified yeast solution, beer wort, and sugar solution tinted with indigocarmine, slowly decolorise in presence of air arid in absence of all organisms, but this change is due to oxidation and does not take place in an atmosphere of carbonic anhydride. Certain aikobic organisms retard or prevent decolorisation by pre- venting the access of oxygen. Active beer-y east also decolorises indigocarmine in absence of ;)xygen, but decolorisation is due to reduction and the colour returns if the liquid ir, exposed to air.I n this case, reduction is accompanied by the development of micro-organisms similar in ap- pearance to the lactic ferment. The change takes place most rapidly i f the yeast solution has been exposed to the air a t 2 4 O for several days ; it then acquires a putrid odour and is full of bacteria. The rate of decolorisation increases with an increase ill the number of orpanisms, and the addition of an antiseptic, or any other cause which destroys the organisms, prevents the changes. Reduction is due to changes connected with the vital processes of the organisms, and is not due to the liberation of hydrogen. It is not analogous to the reduction of indigocarmine by an alkaline solution of glucose, since in that case the presence of organisms is not essential.Similar reducing actions were observed with logwood, orchil, cochineal, safranin, and several artificial colouring matters. C. H. B, Decolorisation of Tincture of Turnesole in closed Vessels. By R. DUBOIS (Bull. Xoc. Chew., 49, 963-964).-Tbe decolorisation of turnesole kept in a closed vessel is entirely due to the action of wrnis, as when the solution is sterilised the blue colour is permanent. The colourless solution contains only one species of liring organisms ; this is a very small, completely spherical micrococcus, which can be cultivated in slightly alkaline peptonised gelatin. The liquid de- colorised by these micro-organisms a t once regains its blue colour on exposure to the air.Formation of Starch from Various Substances. Bg T. BOKORNY (Chem. Cenir., 1888, 858-4339, from Bey. deut. bot. Ges., 6, 116--120).-Since earlier experiments had proved that a l p can be fed with methylal, the author has shown, by means of further experi- ments, that starch is formed from methylal. Spirogyra were used as the material for the investigation. Whilst in the absence of light, the formation of starch could not be observed, its formation could be readily detected after the spirogyra had lain in 1-0-1 per cent. I?. S. K. f 268 ABSTRACTS OF CHEMICAL PAPERS. methylal in the sunlight. Spirogyra fed on solution of methvl alcohol of' the same strength, namely, 1-0.1 per cent., showed a t the end of 6 to 24 hours a very considerable new formation of starch.The author finds that glycol and glyeerol, like mannitol, are also able to form starch. J. W. L. By H. MOLISCH (Ann. Agronom., 14, 334--335).-1t is known that roots excrete an acid juice capable of attacking minerals. The author finds that the liquid has much more extensive powers, namely, it has both reducing and oxidising pro- perties; turns tiiicture of guaiacum blue ; oxidises tannins and humic substances, and consequmtly promotes the decomposition of hunius : transforms cane-sugar into reducing sugar, and acts feebly like diastase ; corrodes a plate of ivory ; and modifies the organic matter of soil. The root membranes are not simply permeated with this juice, it may sometimes 'be seen to exude in dnoplets. Matter Excreted by Roots.J. M. H. M. Occurrence of Solid Hydrocarbons in 'the Vegetable King- dom. By H. GUTZEIT (Rer., 21, 2881-2882. Compare Abbot and Trirnble, Abstr., 1888, 1329).-The author points out that he has already described solid hydrocarbons which were obtained from the fruit of Heracleum giganteurn, hfirt., Eeracleum spondylium, L., and Pastinnca sativa, L. (Beitrage zur Pflmueirchemie, Jena, 1879), and that others have already proved the presence of such compounds in the vegetable kingdom. F. S. K. Constituents of Bark of Rhamnus Frangula and R. Purshiana. By P. SCHWABE (Arch. P h ~ m . [3], 26, 569-594).- The coarsely powdered bark is freed from f a t by means of ether, and extracted with 98 per cent. alcohol ; the extract is mixed with several times its volume of water, and is shaken up by portions with ether.The first ethereal solutions are dark coloured, but on repeating the operation 10 or 1'2 times the ether remains colourless. The unit)ecl ethereal solutions are distilled, when a light-yellow deposit forms in thin layei s on the side of the vessel. The deeply coloured mother-liquor is filtered after remaining 24 hours. The residue on the filter is repeatedly washed with alcohol and ether, and finally crjstallised several times from boiling alcohol, until the microscope shows distinct crystallisation, undoubtedly due to frangulin. The mother-liquor filtered from frangulin was brcught to dryness, taken up with a little alcoliol, mixed with several times its weight of water and again shaken up with ether, but only once or twice ; in this way the beauti- fully crystalline body emodin (trihydroxjmethylanthraquinone) was srpalated.On distilling off the ether, tLe residue is readily seen under a lens to be permeated with crystals. This is heated with glacial acetic acid, in which it readily dissolves, and on cooling emodin crptallises out. Recrystallisation yields emodin as a light red crptalline mass, which melts at 254". The yield was fran- gulin 0.06 per cent., and emodin 0.10 per cent,. Fresh bark gaye no frai:gulin, thus confirming an observation previously made by Castelmann ; whilst bark a year and a-half old gave frangulin 0.04 perVEQETABLE PHTSIOLOOE. AND AQRICULTURE. 69 cent., and emodin 0.10 per cent. as before. Emodin, C15H1005 + HzO, gives red to brown-red amorphous precipitates with the alkaline earths and with lead, copper, and mercury salts.It dissolves in dilute alkalis to a splendid dark cherry-red colour, but gives no crystals on evaporation. A solution in alcoholic potash, heated at 100" in a sealed glass tube, showed fine needle-like aggregates after standing nilopened for 24 hours. FranguZin melts a t 228" to 230°, whilst, Casselmann puts it a t 249" and Faust a t 226", the differences being due to impurity in the last two cases. It is almost insoluble in water and ether, more easily in chloroform, benzene, and alcohol, and very soluble in liot acetic acid. When dry, it forms a beautiful, light- yellow, brittle mass, with somewhat silky lustre. Its composition is C,,H,,09. Four to five hours' boiling with 20 per cent.sulphnric acid converts it into emodin and glucose. If anywllere, frangulinic acid should make its appearance here, but the compound previously described under this name is identical with emodin. The coarsely powdered root of RhanawuspurshiaJna (Cascara sagrada) was extracted with ether, and then with 98 per cent. alcohol. After the addition of water, the ether extract' was shaken repeatedly with light petroleum until the oily extracts became almost colour- less. On removing the petroleum, the dark-coloured mother-liquor gave an immediate brown-red, flocculo-crystalline precipitate. As in the case of Rhamnus frangda bark, the alcoholic extract shaken up with ether after the addition of water afforded a crystalline product. The petroleum product which proved to be identical with this, was found to be emodin.Fraiigulin was not present, although i t may possibly occur in older bark. J. T. Japanese Tobaccos. By M. PESCA and H. IMAI ( B i d Cenfr., 1888,629-637).-The authors analysed the soils and the tobaccos produced in a locality in Opmada, and also one other Japanese tobacco. Permeability and a certain amount of humus are far more important as regards the soil than the amount of plant food, as the latter can be supplied by manuring. The best tobacco soil exhibited only a moderate absorptive power for bases. The manures applied are human and animal excreta, wood and straw ashes, and bath-water. As regards analysis as a means of judging the quality of a tobacco, the authors came to the following conclusions :-(1.) Just as an alco- holic drink should contain a certain percentage of alcohol, so a tobacco should contain a certain percentage of nicotine ; but that the quality of a tobacco depends on the percentage of the nicotine has not yet beenproved.(2.) Nitric acid is not contained in well-fermented tobaccos. ( 3 . ) The ammonia in the older analyses was too high, as it included some that came from amides. The ammonia, which only comes to some tenths of a per cent., cannot account for lowered quality. (4.) The amount of albumino'ids, reckoned as it formerly was, without allowing for the amide-iiitrogen present, affords no standard for judging tobacco. The worst tobacco analysed had the lowest total nitrogen, yet owing to the nicotine and amides being very low, it had the highest amount of albuminoids.( 5 . ) Arnide nitrogeii for the most part indicates the presence of compounds which do not injure and perhaps even enhance70 ABsTRACTS OF CHEMICAL PAPERS. the quality of the tobacco. One of the most important duties of the fermentation is to change albumino'ids into amides. (6.) The deter- mination of the amount of substances that ether extracts is of bnt little use. (7.) Carbohydrates, with the exception of cellulose, should not be present in well-fermented tobacco. A study of their decomposition, and of the formation and decomposition of organic acids and of amides should prove useful in determining the quality and guiding the culture and treatment of tobacco. (8.) Dis- tinctions as to quality can only be drawn when the differences in the amounts of the substances estimated are considerable, and they can only be made safely when the whole composition proves quite satis- factory or quite the reverse ; they can rarely be drawn from differences in the amounts of single substacces.Inferiority can be more safely inferred than superiority. In very bad tobacco, the albuminojids, the sulphuric acid, the clilorine and generally the mineral acids are high, whilst the amide nitrogen and the potash are low. (9.) To indicate good quality and especially combustibility, there should be a medium amount of bases, especially of potash and lime. Within certain limits these bases appear capable of replacing one another. The percentage of one or the other must fall very low before it is to be regarded as a bad indication.Magnesia, if exceptionally high, appears to injure the combustibility. (10.) Mineral acids if high indicate bad combustibility ; but they must be exceptionally high t'o surely indicate low value. Apart from silica, phosphoric acid appeal s to be the least injurious, chlorine considerably more, and sulphuric acid the most injurious to the quality and combustibility. (11.) 'l'hc bases soluble in water and either free or existing as carbonates, appear to have no important influence, but, the amount of carbonic anhydride up to a certain point indicates increased combustibility, and a large amount of carbonates in proportion to mineral salts indicates good value. (12.) A high proportion of bases in the ash, provided it is not caused by magnesia or iron, points especially to good combusti- bility.H. H. R. The remaining conclusions relate to the ash constituents. Formation of Nitrates in Soils of different Degrees of Fertility. By P. P. DEH~RATN (Ann. Agronom., 14, 289-320).-1t appears probable, from the researches of Lawes and Gilbert, War- ington, and others, that a soil exhausted by cropping contains only nitrogenous organic matter difficult to nitrify, and that the relative sterility which is produced by a number of successive crops without manure, is due not only to a decrease in the total amount of nitroqen, but also to the residual nitrogenous matter being less apt to nitrify than that in a fertile soil. To obtain confirmation of this, the author has studied the rate of nitrification in different soils, fertile and exhausted, manured and unmanured, under different conditions of humidity, temperature, division, &c.A few only of the results, most of which are provisional and require further elucidation, are given below. A saturated atmosphere is ultimately unfavourable to nitrifi- cation ; probably because monlds are encouraged which destroy the nitrate. Soil very 6nely sifted, placed in a funnel, and submitted toVEGETABLE PHYSIOLOGY AND AGRICULTURE. 71 frequent waterings, so as to alternate periods of comparative dryness and moistlure, is very favourably circumstanced for nitrification ; for example, in the 189 days from 1 7 t h May to 22nd November, there was formed per 1000 kilos. of soil, 880 grams nitric acid: reckoning the weight of the soil at 3600 tonnes per hectare, this would give 819.8 kilos.nitrogen per hectare nitrified in that space of time, a quantity infinitely superior to the requirements of the most exhausting crop. The waterings in this experiment were equivalent to five times the normal rainfall. The accumulation of nitrate i i i the soil, at any rate t o the extent of 700 mgrrns. per kilo., does not retard the rate of nitrification. The organic matter was more nitrifiable at the com- mencement of the experiment than afterwards, since much more nitrate was formed in the first 27 days than in any subsequent similar period, notwithstanding the lower temperature of this first month as compared with the two following. Again, in the December to January period more nitrogen was nitrified per diem than in the October +,o November period, in spite of the lower temperature ; from this the author infers that the organic matter of the soil is subject from time to time to changes rendering it more or less easy of nitrification.Trituration of the soil and elevation of temperature were both found to greatly accelerate nitrification. The addition of sodium nitrate t o the soil in quantities of 0.06 and 0.60 per cent. almost prevented nitri- fication for the first 40 days, and greatly retarded it during a subsequent like period, although eventually nitrification at something Iike the normal rate occurred; the addition of 1 : 1000 of common salt to the soil exercises little or no effect, 2.5 : 1000 is injurioua, and 5 : 1000 prevents nitrification. Three soils, long unmanured and poor in nitrogen, developed very little nitrate for some weeks after being placed in circumstances the most favourable for nitrification, but afterwards fairly rapid nitrifi- cation set in, which the author attributes to changes undergone by the organic matter under the new conditions.A fourth poor soil, unma.nured and equally poor in nit,rogen with the other three, proved, however, very nitrifiable, it developed from the first more nitrate than a fertile soil placed in the same conditions. J. M. H. M. Loss and Gain of Nitrogen in Agriculture. By B. FRANK (Bied. Cerh., 1888, 610-617) .-Among the sources of loss is the volatilisation of animonia from the soil. In some experiments in which ammonium sulphate solution was added to samples of soil, the author found that the ammonia thus added soon disappeared, being to a small extent converted into nitrates whilst the greater part volatilised.A light and pure sandy soil does not expel the ammonia and has only a feeble nitrifying power, so that the ammonium salt is retained nearly undiminished for a long time. Of the individual constituents, quartz grains and clay are inactive, whilst calcium carbonate causes both slow nitrification and also partial liberation of ammonia. Another source of loss was mentioned by Boussingault, namely, that when nitrates were given to plants growing in the dark, there was a libera- tion of free nitrogen, which be attributed to the action of some organic substance excreted by the roots. The author experimented72 ABSTRACTS OF CHEMICAL PAPERS.on the point, growing bean seeds in the dark in nutritive solutions, both with and without nitrogen compounds. As in every case there was a loss of nitrogen, it could not be due to reduction of nitrates, and he attributes it to the loss of nitrogen consequent on the decay of those parts of the seed not made use of by the germinating plant. Next, treating of the gain of nitrogen, he combats Hellriegel's view that the root nodules of the Leguminosae are concerned in rendering free nitrogen available to the plant. His own experiments lead him to the coriclusion that the land gains in combined nitrogen in some way besides that caused by lightning discharges, which at present is the only one undisputed. He found that the presence of vegetation raised the amount of this gain, that this could not be accounted for by the ammonia of the air, and that it must be derived from the free nitrogen of the air.The gain rises with increased plant development, and both the kind of soil and the species of plant have an influence. Lupines are very effective as compared with non-leguminous plants ; but the difference in the powers of increasing the combined nitrosen is one of degree, not one of kind ; hence it cannot be ascribed to the nodules. Further, nodules did not occur in lupines grown in sterilised soil, yet the plants developed better than in a pardlel experiment with unsterilised soil where the plants had nodules, and besides the nodules the legumes have no other organs to siipply nitrogen which other plants do not also possess.The next question is how soil unoccupied by a crop gains in combined nitrogen. This gain takes the form chieHy or entirely of organic nitrogen compounds, thus agreeing with Ber- thelot's experiments, and is explained by the growth in the soil of the cryptoganiic plants, algae containing chlorophyll and allied forms which the author discovered there. He next discusses the question whether these plants avail themselves of free nitrogen or of nitrogen oxidised in the soil by an inorganic process. He asserts that under the influence of calcium or magnesium carbonates free nitrogen can be oxidised to nitric acid, the action being distinct a t lOO", still apparent a t 45-50', and no longer apparent a t 15-22", but thinks that in the German climate this action can only very rarely occur.He con- cludes that it is not an inorganic process that makes the free nitrogen available, but that the combined nitrogen is due to a development of plant cells containing albumino'ids which is not to be connected with any process occurring in the soil. This power of assimilating free nitrogen is very different for different plants; the result is smallest in fallow land, where only the lower plant forms are at work; it is larger with hjghsr plant forms, and among these the lupines and probably other Leguminosae produce the greatest result. H. H. R. Action of Superphosphate on Nitrates. By A. DWARDA (Chew. Ceiztr., 1888, 899--900).-Experirnents by the author go to show that, at the ordinary temperature, the phosphoric acid, hydro- flouric acid, and readily decomposable organic compounds, as well as ferrous salts present in superphosphates, cause no loss of nitrogen from nitrates which may have been mixed with them.In those cases, however, where the mixtures are exposed to a moderately high temperature, the loss of nitrogen is considerable, although in thisANALYTICAL CHENIST RY. 73 case the ferrous compounds take no part in the reaction. The author also found that iii mixtures of superphosphate and nitrates the soluble phosphate becomes insoluble much more quickly than when the nitrate is omitted, and the mixing of nitrate with superphos- phate for any length of time is therefore not to be recommended. J. W. L.VEGETABLE PHYSIOLOGY AND XQRICULTURE. 67Chemistry of Vegetable Physiology and Agriculture.Action of Micro-organisms on cert -tin Colouring Matters.By J.RAULIN (Corn@. rend., 107, 4B.;-447).-A.~1eergillus nigerwill grow in cultivation liquids containing ammonium nitrate, but notin similar liquids containing salts of aniline, rosaniline or indigo-carmine.Slightly acidified yeast solution, beer wort, and sugar solutiontinted with indigocarmine, slowly decolorise in presence of air aridin absence of all organisms, but this change is due to oxidation anddoes not take place in an atmosphere of carbonic anhydride.Certain aikobic organisms retard or prevent decolorisation by pre-venting the access of oxygen.Active beer-y east also decolorises indigocarmine in absence of;)xygen, but decolorisation is due to reduction and the colourreturns if the liquid ir, exposed to air.I n this case, reduction isaccompanied by the development of micro-organisms similar in ap-pearance to the lactic ferment. The change takes place most rapidlyi f the yeast solution has been exposed to the air a t 2 4 O for severaldays ; it then acquires a putrid odour and is full of bacteria. Therate of decolorisation increases with an increase ill the number oforpanisms, and the addition of an antiseptic, or any other causewhich destroys the organisms, prevents the changes. Reduction isdue to changes connected with the vital processes of the organisms,and is not due to the liberation of hydrogen. It is not analogous tothe reduction of indigocarmine by an alkaline solution of glucose,since in that case the presence of organisms is not essential.Similar reducing actions were observed with logwood, orchil,cochineal, safranin, and several artificial colouring matters.C.H. B,Decolorisation of Tincture of Turnesole in closed Vessels.By R. DUBOIS (Bull. Xoc. Chew., 49, 963-964).-Tbe decolorisationof turnesole kept in a closed vessel is entirely due to the action ofwrnis, as when the solution is sterilised the blue colour is permanent.The colourless solution contains only one species of liring organisms ;this is a very small, completely spherical micrococcus, which can becultivated in slightly alkaline peptonised gelatin. The liquid de-colorised by these micro-organisms a t once regains its blue colour onexposure to the air.Formation of Starch from Various Substances.Bg T.BOKORNY (Chem. Cenir., 1888, 858-4339, from Bey. deut. bot. Ges.,6, 116--120).-Since earlier experiments had proved that a l p can befed with methylal, the author has shown, by means of further experi-ments, that starch is formed from methylal. Spirogyra were used asthe material for the investigation. Whilst in the absence of light,the formation of starch could not be observed, its formation could bereadily detected after the spirogyra had lain in 1-0-1 per cent.I?. S. K.f 68 ABSTRACTS OF CHEMICAL PAPERS.methylal in the sunlight. Spirogyra fed on solution of methvl alcoholof' the same strength, namely, 1-0.1 per cent., showed a t the end of6 to 24 hours a very considerable new formation of starch.Theauthor finds that glycol and glyeerol, like mannitol, are also able toform starch. J. W. L.By H. MOLISCH (Ann. Agronom., 14,334--335).-1t is known that roots excrete an acid juice capable ofattacking minerals. The author finds that the liquid has much moreextensive powers, namely, it has both reducing and oxidising pro-perties; turns tiiicture of guaiacum blue ; oxidises tannins and humicsubstances, and consequmtly promotes the decomposition of hunius :transforms cane-sugar into reducing sugar, and acts feebly likediastase ; corrodes a plate of ivory ; and modifies the organic matterof soil. The root membranes are not simply permeated with thisjuice, it may sometimes 'be seen to exude in dnoplets.Matter Excreted by Roots.J.M. H. M.Occurrence of Solid Hydrocarbons in 'the Vegetable King-dom. By H. GUTZEIT (Rer., 21, 2881-2882. Compare Abbot andTrirnble, Abstr., 1888, 1329).-The author points out that he hasalready described solid hydrocarbons which were obtained from thefruit of Heracleum giganteurn, hfirt., Eeracleum spondylium, L., andPastinnca sativa, L. (Beitrage zur Pflmueirchemie, Jena, 1879), and thatothers have already proved the presence of such compounds in thevegetable kingdom. F. S. K.Constituents of Bark of Rhamnus Frangula and R.Purshiana. By P. SCHWABE (Arch. P h ~ m . [3], 26, 569-594).-The coarsely powdered bark is freed from f a t by means of ether, andextracted with 98 per cent. alcohol ; the extract is mixed with severaltimes its volume of water, and is shaken up by portions with ether.The first ethereal solutions are dark coloured, but on repeating theoperation 10 or 1'2 times the ether remains colourless.The unit)eclethereal solutions are distilled, when a light-yellow deposit forms inthin layei s on the side of the vessel. The deeply coloured mother-liquoris filtered after remaining 24 hours. The residue on the filter isrepeatedly washed with alcohol and ether, and finally crjstallisedseveral times from boiling alcohol, until the microscope shows distinctcrystallisation, undoubtedly due to frangulin. The mother-liquorfiltered from frangulin was brcught to dryness, taken up with a littlealcoliol, mixed with several times its weight of water and againshaken up with ether, but only once or twice ; in this way the beauti-fully crystalline body emodin (trihydroxjmethylanthraquinone) wassrpalated. On distilling off the ether, tLe residue is readily seenunder a lens to be permeated with crystals.This is heated withglacial acetic acid, in which it readily dissolves, and on coolingemodin crptallises out. Recrystallisation yields emodin as a lightred crptalline mass, which melts at 254". The yield was fran-gulin 0.06 per cent., and emodin 0.10 per cent,. Fresh bark gayeno frai:gulin, thus confirming an observation previously made byCastelmann ; whilst bark a year and a-half old gave frangulin 0.04 peVEQETABLE PHTSIOLOOE. AND AQRICULTURE. 69cent., and emodin 0.10 per cent. as before. Emodin, C15H1005 + HzO,gives red to brown-red amorphous precipitates with the alkaline earthsand with lead, copper, and mercury salts.It dissolves in dilutealkalis to a splendid dark cherry-red colour, but gives no crystals onevaporation. A solution in alcoholic potash, heated at 100" in asealed glass tube, showed fine needle-like aggregates after standingnilopened for 24 hours. FranguZin melts a t 228" to 230°, whilst,Casselmann puts it a t 249" and Faust a t 226", the differences being dueto impurity in the last two cases. It is almost insoluble in waterand ether, more easily in chloroform, benzene, and alcohol, and verysoluble in liot acetic acid. When dry, it forms a beautiful, light-yellow, brittle mass, with somewhat silky lustre. Its composition isC,,H,,09.Four to five hours' boiling with 20 per cent. sulphnricacid converts it into emodin and glucose. If anywllere, frangulinicacid should make its appearance here, but the compound previouslydescribed under this name is identical with emodin.The coarsely powdered root of RhanawuspurshiaJna (Cascara sagrada)was extracted with ether, and then with 98 per cent. alcohol. Afterthe addition of water, the ether extract' was shaken repeatedlywith light petroleum until the oily extracts became almost colour-less. On removing the petroleum, the dark-coloured mother-liquorgave an immediate brown-red, flocculo-crystalline precipitate. Asin the case of Rhamnus frangda bark, the alcoholic extract shaken upwith ether after the addition of water afforded a crystalline product.The petroleum product which proved to be identical with this, wasfound to be emodin.Fraiigulin was not present, although i t maypossibly occur in older bark. J. T.Japanese Tobaccos. By M. PESCA and H. IMAI ( B i d Cenfr.,1888,629-637).-The authors analysed the soils and the tobaccosproduced in a locality in Opmada, and also one other Japanesetobacco. Permeability and a certain amount of humus are far moreimportant as regards the soil than the amount of plant food, as thelatter can be supplied by manuring. The best tobacco soil exhibitedonly a moderate absorptive power for bases. The manures appliedare human and animal excreta, wood and straw ashes, and bath-water.As regards analysis as a means of judging the quality of a tobacco,the authors came to the following conclusions :-(1.) Just as an alco-holic drink should contain a certain percentage of alcohol, so a tobaccoshould contain a certain percentage of nicotine ; but that the qualityof a tobacco depends on the percentage of the nicotine has not yetbeenproved.(2.) Nitric acid is not contained in well-fermented tobaccos.( 3 . ) The ammonia in the older analyses was too high, as it included somethat came from amides. The ammonia, which only comes to some tenthsof a per cent., cannot account for lowered quality. (4.) The amount ofalbumino'ids, reckoned as it formerly was, without allowing for theamide-iiitrogen present, affords no standard for judging tobacco.The worst tobacco analysed had the lowest total nitrogen, yet owingto the nicotine and amides being very low, it had the highest amountof albuminoids.( 5 . ) Arnide nitrogeii for the most part indicates thepresence of compounds which do not injure and perhaps even enhanc70 ABsTRACTS OF CHEMICAL PAPERS.the quality of the tobacco. One of the most important duties of thefermentation is to change albumino'ids into amides. (6.) The deter-mination of the amount of substances that ether extracts is of bntlittle use. (7.) Carbohydrates, with the exception of cellulose,should not be present in well-fermented tobacco. A study of theirdecomposition, and of the formation and decomposition of organicacids and of amides should prove useful in determining thequality and guiding the culture and treatment of tobacco. (8.) Dis-tinctions as to quality can only be drawn when the differences in theamounts of the substances estimated are considerable, and they canonly be made safely when the whole composition proves quite satis-factory or quite the reverse ; they can rarely be drawn from differencesin the amounts of single substacces.Inferiority can be more safelyinferred than superiority. In very bad tobacco, the albuminojids, thesulphuric acid, the clilorine and generally the mineral acids are high,whilst the amide nitrogen and the potash are low.(9.) Toindicate good quality and especially combustibility, there should be amedium amount of bases, especially of potash and lime. Withincertain limits these bases appear capable of replacing one another.The percentage of one or the other must fall very low before it isto be regarded as a bad indication.Magnesia, if exceptionally high,appears to injure the combustibility. (10.) Mineral acids if highindicate bad combustibility ; but they must be exceptionally high t'osurely indicate low value. Apart from silica, phosphoric acid appeal sto be the least injurious, chlorine considerably more, and sulphuricacid the most injurious to the quality and combustibility. (11.) 'l'hcbases soluble in water and either free or existing as carbonates,appear to have no important influence, but, the amount of carbonicanhydride up to a certain point indicates increased combustibility, anda large amount of carbonates in proportion to mineral salts indicatesgood value.(12.) A high proportion of bases in the ash, provided itis not caused by magnesia or iron, points especially to good combusti-bility. H. H. R.The remaining conclusions relate to the ash constituents.Formation of Nitrates in Soils of different Degrees ofFertility. By P. P. DEH~RATN (Ann. Agronom., 14, 289-320).-1tappears probable, from the researches of Lawes and Gilbert, War-ington, and others, that a soil exhausted by cropping contains onlynitrogenous organic matter difficult to nitrify, and that the relativesterility which is produced by a number of successive crops withoutmanure, is due not only to a decrease in the total amount of nitroqen,but also to the residual nitrogenous matter being less apt to nitrifythan that in a fertile soil.To obtain confirmation of this, the authorhas studied the rate of nitrification in different soils, fertile andexhausted, manured and unmanured, under different conditions ofhumidity, temperature, division, &c. A few only of the results, mostof which are provisional and require further elucidation, are givenbelow. A saturated atmosphere is ultimately unfavourable to nitrifi-cation ; probably because monlds are encouraged which destroy thenitrate. Soil very 6nely sifted, placed in a funnel, and submitted tVEGETABLE PHYSIOLOGY AND AGRICULTURE. 71frequent waterings, so as to alternate periods of comparative drynessand moistlure, is very favourably circumstanced for nitrification ; forexample, in the 189 days from 1 7 t h May to 22nd November, there wasformed per 1000 kilos.of soil, 880 grams nitric acid: reckoning theweight of the soil at 3600 tonnes per hectare, this would give 819.8kilos. nitrogen per hectare nitrified in that space of time, a quantityinfinitely superior to the requirements of the most exhausting crop.The waterings in this experiment were equivalent to five times thenormal rainfall. The accumulation of nitrate i i i the soil, at any ratet o the extent of 700 mgrrns. per kilo., does not retard the rate ofnitrification. The organic matter was more nitrifiable at the com-mencement of the experiment than afterwards, since much morenitrate was formed in the first 27 days than in any subsequentsimilar period, notwithstanding the lower temperature of this firstmonth as compared with the two following.Again, in the Decemberto January period more nitrogen was nitrified per diem than in theOctober +,o November period, in spite of the lower temperature ; fromthis the author infers that the organic matter of the soil is subject fromtime to time to changes rendering it more or less easy of nitrification.Trituration of the soil and elevation of temperature were both foundto greatly accelerate nitrification. The addition of sodium nitrate t othe soil in quantities of 0.06 and 0.60 per cent. almost prevented nitri-fication for the first 40 days, and greatly retarded it during asubsequent like period, although eventually nitrification at somethingIike the normal rate occurred; the addition of 1 : 1000 of commonsalt to the soil exercises little or no effect, 2.5 : 1000 is injurioua, and5 : 1000 prevents nitrification.Three soils, long unmanured and poor in nitrogen, developed verylittle nitrate for some weeks after being placed in circumstances themost favourable for nitrification, but afterwards fairly rapid nitrifi-cation set in, which the author attributes to changes undergone bythe organic matter under the new conditions.A fourth poor soil,unma.nured and equally poor in nit,rogen with the other three, proved,however, very nitrifiable, it developed from the first more nitrate thana fertile soil placed in the same conditions. J. M. H. M.Loss and Gain of Nitrogen in Agriculture. By B. FRANK(Bied. Cerh., 1888, 610-617) .-Among the sources of loss is thevolatilisation of animonia from the soil.In some experiments in whichammonium sulphate solution was added to samples of soil, the authorfound that the ammonia thus added soon disappeared, being to a smallextent converted into nitrates whilst the greater part volatilised. Alight and pure sandy soil does not expel the ammonia and has only afeeble nitrifying power, so that the ammonium salt is retained nearlyundiminished for a long time. Of the individual constituents, quartzgrains and clay are inactive, whilst calcium carbonate causes bothslow nitrification and also partial liberation of ammonia. Anothersource of loss was mentioned by Boussingault, namely, that whennitrates were given to plants growing in the dark, there was a libera-tion of free nitrogen, which be attributed to the action of someorganic substance excreted by the roots.The author experimente72 ABSTRACTS OF CHEMICAL PAPERS.on the point, growing bean seeds in the dark in nutritive solutions,both with and without nitrogen compounds. As in every case therewas a loss of nitrogen, it could not be due to reduction of nitrates,and he attributes it to the loss of nitrogen consequent on the decayof those parts of the seed not made use of by the germinating plant.Next, treating of the gain of nitrogen, he combats Hellriegel's viewthat the root nodules of the Leguminosae are concerned in renderingfree nitrogen available to the plant. His own experiments lead himto the coriclusion that the land gains in combined nitrogen in someway besides that caused by lightning discharges, which at present isthe only one undisputed.He found that the presence of vegetationraised the amount of this gain, that this could not be accounted forby the ammonia of the air, and that it must be derived from the freenitrogen of the air. The gain rises with increased plant development,and both the kind of soil and the species of plant have an influence.Lupines are very effective as compared with non-leguminous plants ;but the difference in the powers of increasing the combined nitrosenis one of degree, not one of kind ; hence it cannot be ascribed to thenodules. Further, nodules did not occur in lupines grown in sterilisedsoil, yet the plants developed better than in a pardlel experiment withunsterilised soil where the plants had nodules, and besides the nodulesthe legumes have no other organs to siipply nitrogen which other plantsdo not also possess.The next question is how soil unoccupied by acrop gains in combined nitrogen. This gain takes the form chieHyor entirely of organic nitrogen compounds, thus agreeing with Ber-thelot's experiments, and is explained by the growth in the soil of thecryptoganiic plants, algae containing chlorophyll and allied formswhich the author discovered there. He next discusses the questionwhether these plants avail themselves of free nitrogen or of nitrogenoxidised in the soil by an inorganic process. He asserts that underthe influence of calcium or magnesium carbonates free nitrogen can beoxidised to nitric acid, the action being distinct a t lOO", still apparenta t 45-50', and no longer apparent a t 15-22", but thinks that in theGerman climate this action can only very rarely occur. He con-cludes that it is not an inorganic process that makes the free nitrogenavailable, but that the combined nitrogen is due to a developmentof plant cells containing albumino'ids which is not to be connectedwith any process occurring in the soil. This power of assimilatingfree nitrogen is very different for different plants; the result issmallest in fallow land, where only the lower plant forms are atwork; it is larger with hjghsr plant forms, and among these thelupines and probably other Leguminosae produce the greatest result.H. H. R.Action of Superphosphate on Nitrates. By A. DWARDA(Chew. Ceiztr., 1888, 899--900).-Experirnents by the author go toshow that, at the ordinary temperature, the phosphoric acid, hydro-flouric acid, and readily decomposable organic compounds, as well asferrous salts present in superphosphates, cause no loss of nitrogenfrom nitrates which may have been mixed with them. In thosecases, however, where the mixtures are exposed to a moderately hightemperature, the loss of nitrogen is considerable, although in thiANALYTICAL CHENIST RY. 73case the ferrous compounds take no part in the reaction. The authoralso found that iii mixtures of superphosphate and nitrates thesoluble phosphate becomes insoluble much more quickly than whenthe nitrate is omitted, and the mixing of nitrate with superphos-phate for any length of time is therefore not to be recommended.J. W. L