-K4 RKPORT METHODS FOR lO For the American Association of Official Agricultural Chemists. By R. C. KEDZIE, Reporter. LAXSIXG: D. D. Thorp, Printer and Binder. 1891. 6 the soil distilled water (amnioiiia-free) sufficient to moisten all the soil, and let the whole stand undisturbed for half an hour, then add more pure distilled water, and if the filtration is too slow use the filter-pump till a liter of filtrate is secured. If the soil extract is cloudy, filter through a plain filter. Each cubic centimeter of filtrate contains the w^ter-soluble materials of a gram of air-dry soil. 9. Soluble solids. — Evaporate 100 c. c. to dryness on the water- bath in a tared dish to determine the percentage of water-soluble materials in the soil ; each gram of residue representing a i)er cent of such materials. Test this dry residue for nitrates by pouring over it 10 c. c. of O.P. H» 8O4 holding in solution tliree or U>\xv milligrams of sulfate of brucia. 10. CJilorkles. — Titrate 100 c. c. with standard decinormal argen- tic nitrate with two drops of solution of KsCrO^ as indicator. Titrate in white porcelain dish and view the reaction through a "yellow glass plate of such tint as will eliminate the color of the chromic solution. The reaction will then be sharply defined. More than one part of soluble chlorides in one thousand of soil is injurious to agricultural plants. 11. Sulfates. — Precipitate the soluble sulfates in 100 c. c. with BaCU in presence of a few drops of HCl. and estimate the soluble sulfates. ■Reserve the rest of the water solution (8) for the estimation of nitrates (30). AciD-8uLU]{LE Materials. In the following scheme for soil analysis it is recommended to use the air-dry soil from the sample bottle for each separate in- vestigation. A determination made -once for all of hj'groscopic moisture and of water of combination on a separate specimen of air-dry soil will aftord corrections for all the other samples used. It is not desirable to ignite the soil before analysis, or to lieat it so as to change its chemical properties. In an agricultural chemical analysis the object is to find the kind and quantity of soil materials availaljle for the growing })lant. The reserve or inactive materials are not objects of immediate concern. Prof, llilgard takes C. P. IICI. of specific gravity 1. 11. '> as the solvent for soil materials, which may be supposed to fully represent the solvent action that may be secured by water and other solvents in the soil, and the at-tion of the roots of growing n& 4 W6 7 plants or other corroding agents. The same acid and of the same strength is here recommended, but with a modified form of using the same. Instead of digesting the soil with this acid in a covered beaker, with liability to continual variation of strength of the acid, it is proposed to use the acid with constant strength except so far as it may be neutralized by combination with the soil minerals. Instead of a porcelain beaker covered with a watch glass I pro- pose for the soil digestion a four-ounce vial of Bohemian glass, with a flat-topped ground glass stopper. The small steam bath is a copper vessel eight inches in diameter and five inches deep, with vertical sides ; the cover has four openings through which the vials nuiy readily pass down two and three-quarter inches to a perforated false bottom upon which the vials rest, and the space of two and a quarter inches below the false bottom serves for the hot water chamber of the steam bath. T'hrough the center of the cover and of the false bottom a vertical tube, open at both ends, extends nearly to the bottom of the vessel, for pouring in water to replenish the waste ; or a side tube near the bottom of the hot water chamber may be connected with a water reservoir to keep the water in the steam bath at a constant level. A Bun- sen burner serves to heat up the steam bath and to keep the body of the vials at the constant heat of boiling water. When such a bath is set to work the digestion can go forward day and night with very little care and attention. If the vials charged for diges- tion are placed in the apparatus when cold and then heated up to steam heat, no trouble is found from the vials breaking. AVhen the vials are proj)erly charged and the escape of acid prevented, thirty-six to forty hours of continuous digestion will be as effect- ive as five days of intermittent digestion in a covered beaker. 12. Acid digestion of the soil. — Weigh five grams of the air- dry soil into a four-ounce Bohemian vial, add 50 c. c. of C.P. HCl. sp. gr. 1.115, insert glass stopper, wire it securely, place in steam bath and digest for thirty-six to forty hours at the temperature of boiling water. Pour the contents of the vial into a small beaker, wash out the vial with distilled water, add the washings to the contents of the beaker, pour into this 2 c. c. of HNOs to peroxidize the iron and oxidize organic matter, and evaporate the contents of the beaker to complete dryness over the water bath. Cool the beaker, add 10c. c. of C.P. HCl. sp. gr. 1.115 and 50 c. c. of distilled water and heat to near boiling. Filter from sand and silica, 8 wash the filter with distiHetl water till tlie tiltrate sliows no reac- tion with sihx-r nitrate, and make the liltrate u]) to500e, c. (Solu- tion A). 1011 f. c. of which represents the soluhle materials from one gram of air-drv soil. 13. Siiiid (did silica. — Dry the filter and insoluble residue from A, transfer the residue to a tared platinum dish, burn the filter and add its ash to the dish, heat the dish and contents, at first gentl}' to avoid spurting of silica, then intensely to destroy organic matter, cool in desiccator and weigh. The increase of weight — minus the filter ash — represents the sand and silica. Boil this residue for fifteen minutes in oOc. c. of strong solution of sodic carboiuite. add 100 e.c. of boiling water, filter while still hot. aiul wash the filtei' and contents with boiling water till the sodic salfee are washed away. Dry the i-esidue. l)urn the filter aiid add its ash to the insoluble residue, heat this to redness, cool and weigh. Deduct the ash of filter and enter the balance as sand or insol- UHLE siiJ<'\TKS. The ditt'erence in weight between sand, and sand and silica, enter as simca. This sand and silica will res|)ect- ively represent the amount of these materials in five grams of air- dry soil, and these weights multiplied by twenty will give the ])er cent respectively of sand and silica in such soil. 14. Ferric oxide iiad alumina. — To 200 c. c. of Solution A (in an Erlenmeyer flask) add NILHO to alkaline reaction (avoiding excess), to precipitate ferric and aluminic oxides and phosphates. Expel excess of ammonia l)y boiling, let it settle, decant the clear solution through a filter ; add to the fiask 50 c. c. of hot distilleil water, boil, settle and decant as before. After pouring off all the clear solution possible, dissolve the residue with a few drops of IICl. with heat, add just enough N^H4lI0 to })recipitate the oxides. Wash by decantation with 50 c. c. of distilled water, and then transfer all the precipitate to the filter, and -wash witli hot distilled water till the filtrate becomes free from chlorides. (Savi' the filtrate and washings. Solution 15.) Dry tlu^ Hltei' and pre- cipitate in the air-bath at llO*^. transfer the pivcipitate to a tared platinum crucible, l)urn the filter and add the ash to the precip- itate, heat the whole red-hot, cool in desiccator and weigh. The increase of weight — minus the ash of lilter ami the phosphoric acid (found in a separate pi-ocess) — represents the weight of the ferri(! and aluminic oxides. 15. /•/■/■/•/'■ i>xiilf. — IMai'c the whole of the i with heat, and test the solution with excess of molybdate of aninioniuni at 70°. 23. Fofash and soda. — Evaporate the filtrate and washings (from H'l) to dryness, heat to low red heat to decompose oxalates and expel ammonia salts, dissolve in 25 c. c. of distilled water, filter and wash the precipitate, add to the filtrate and washings 10 c. c. of baryta water, and digest for an hour. Filter and wash precip- itate, add amnionic carbonate to the filtrate to complete precipi- tation of baryta, filter and wash this precipitate. Evaporate the filtrate and washings in a tared platinum dish, gently ignite the residue to expel ammonic salts, cool and weigh. The increase of weight represents the chlorides of potassium and sodium in two grams of air-dry soil. Separate and estimate the potassium chloride by platinic chlor- ide according to the official method of the Association of Agri- cultural Chemists. Subtract the weight of potassium chloride as thus found from the weight of potassium chloride and sodium chloride. The differ- ence represents sodium chloride. Alternate method. — For alternate method for alkalies, use J. Lawrence Smith's method as given in (Jrook's Select Methods, second edition, pp. 28 to 40. 24. Other alkali tnetah. — The salts of lithium, caesium aiul rubidium are occasionally found in very small amounts in soils. The agricultural uses of these salts are still in question, and their amount is too small to admit of quantitative estimation. A quali- tative examination may be made by the spectroscope with tlu' water-soluble materials (8) evaporated to dryness and dissolved with two or three drops of IICl. Test by spectroscope with plat- inum wire in Bunsen flame. 25. Xih'oi/t'N of the soil. — The com])iiic(l nitrogen in the soil and the state of combination in whic^h it is held are subjects of great importance to the agricultural chemist. The niti-ogen com- l)ounds in tlu! soil are usually placed in three classes: 1. The nitrogen combined with oxygen as nitrates oi' nitrites, existing as soluble salts in the soil. 2. The nitrogen combined with hydrogen as aimnoiiia. or organic nitrogen easily (jonA'ertible into ammonia. The annnoiiia 13 may exist as suits or be occluded by liydnited ferric or aluminic oxides and organic matter in the soil. 3. The inert nitrogen of the soil or the humose nitrogen. The nitrogen in the first and second classes is considered the active nitrogen of the soil so far as plant food is concerned, while the inert nitrogen is, for the time being, incapable of aifording sustenance to agricultural plants, and hence is properly jjlaced in a class by itself. But the exchanges between the first and second classes are well known to chemists ; the reduction of nitrates to ammonia, and the oxidation of ammonia to nitrates are familiar to agricultural chemists. It has also been a matter of discussion which of these forms is best fitted to nourish plant life. They seem to have equal agricultural activity, and their exchanges arc matters of almost daily occurrence. Why should they be separ- ately estimated in an agricultural chemical analysis? Why not class them together as active soil nitrogen and estimate their amount in one operation ? 26. Active soil nitrogen. — The material proposed for reducing the nitrates to ammonia, and at the same time to bring ammonia salts and organic nitrogen into condition for separation by distil- lation, is sodium amalgam. Liquid sodium amalgam may be readily prepared by placing 100 c. c. of mercury in a flask of half liter capacity, covering the warmed mercury with melted paraffine and dropping into the flask at short intervals metallic sodium the size of a large pea (taking care that the violence of the reaction does not^ project the contents from the flask), till 6.75 grams of sodium have combined with the mercury. This amal- gam contains one-half of one per cent of sodium, and may be preserved indefinitely under the covering of paraffine. The mer- cury is easily recovered at the close of the operation, and nothing of value is wasted except the sodium. To estimate the active soil nitrogeii, weigh fifty grams of air- dry soil and place it in a clean mortar. Take 200 c. c. of ammonia- free distilled water, rub up the soil with a part of the water to a smooth paste," transfer this to a flask of one liter capacity, wash- ing the last traces of the soil into the flask with the rest of the water. Add 25 c. c. of the liquid sodium amalgam, and shake the flask so as to break the sodium amalgam into small globules dis- tributed through the soil. Insert a stopper with a Kroonig valve and set aside in a cool place for twenty-four hours. Pour into the flask 50 c. c. of milk of litne aiul distil on a sand bath 100 c. c. 14 into a flask containing 20 c. c. of decinormal sulfnric acid, and titrate with decinormal soda solntion, nsing dimethyl orange as indicator. Estimate the nitrogen of tlie ammonia found as art ire soil nitrofjoi. If the ammonia produced is too small in amoujit to be readily estimated volumetrically, determine the ammonia by Nessleriz- ing the distillate. , 27. Estiination of nitrates in the soil. — When it is desired to estimate separately the nitrates in the soil the following modifica- tion of 20 nuiy be used : Evaporate 100 c. c. of the soil extract (8) to dryness on the water bath ; dissolve the soluble portion of the residue in lOOc. c. of amnionia-frce distilled water, filtering out any insoluble residue, place the solution in a llask ami add 10 c. c. of liquid sodium amalgam, insert stopper with Kroonig valve, set it aside to digest in a cool place for twenty-four hours, add 50 c. c. of milk of lime, distil and titrate as in 20, ami estimate the nitrogen as Ns 0.. Nesslerizing may be substituted for titration when the amount of nitrates is small. An approximate estimation of the amount of nitrates will be of value in determining which method of estimation to use. This may be done by evaporating a measured quantity of the soil extract (8) say See, more or less, on a porcelain cover on a steam bath or radiator, having first dissolved a minute fragment of pure sulfate of brucia in the soil extract. Wlien dry, pour over the residue concentrated sulfuric acid free from nitTates, and observe the color reactions j) rod need. If the nitrate (reckoned as KNOs) left upon evaporating the quantity of water taken cloes not exceed the two thousandths part of a milligram, only a pink color will be developed by adding the sulfuric acid; with the three thousandths part of a milligram, a pink with faint reddish lines; with the four thousandths part, a reddish (tolor; with the five thousandths part, a red color. By increasing or diminishing the amount of soil^extract evap- orated to setuire a color reaction of a certain intensity, an approx- imate estimate may l)e made of the amount of nitrates present. Blank experiments to test the acid, and the brucine will be required Ijcfore confidence can be placed in such estimation. 28. Total niirofjen of soils. — The total nitrogen of soils may be determined l)y the usual combustion with soda-h'mo, but this 15 process is often uusatisftictory, because of tlie large amount of material required when the organic matter or humus is in small amount. A modification of the KJeldahl method is more easy to carry out, and gives results equally satisfactory. Weigh out twenty grams of air-dry soil, place this in a Kjeldahl flask and pour in 20 c. c. of sulfuric acid (free from ammonia) holding in solution one gram of salicylic acid. (If the soil contains much lime or mag- nesia in the form of carbonate, enough more sulfuric acid must be added to secure a strongly acid condition of the contents of the flask.) Add gradually two grams of zinc dust, shaking the contents of the flask to secure intimate mixture. Place the flask in a sand bath and heat till the acid boils, and maintain the boil- ing for ten minutes. Add one gram of mercury and continue the boiling for one hour, adding 10 c. c. of sulfuric acid if the con- tents of the flask are likely to become solid. Cool the flask and wash out the sohible materials in the flask with 200 c. c. of pure water, leaving the heavy earthy materials in the Kjeldahl flask. Kinse the residue with 100 c. c. of water and add this to the first washings. Place this soluble acid extract in a liter digestion flask, add 35 c. c. of solution of potassium sulphide and shake the flask to secure intimate mixture of the contents. IntToduce a few fragments of granulated zinc, pour in 75 c. c. of saturated solu- tion of caustic soda, connect the flask with a condenser and dis- til 150 c. c. into a fl>isk containing 20 c. c. of decinormal sulfuric acid, and titrate with decinormal soda solution, using cochineal or dimethyl orange as indicator. Enter the nitrogen found in this operation as total soil nitroffcn. The difference between the total soil nitrogen and the active soil nitrogen will express the inert nitrogen of the soil. 2U. Acid soils. — Soils of good agricultural quality are usually neutral or slightly alkaline, but soils are found Avhicli give a deci- dedly acid reaction when blue litmus paper is pressed upon the moist surface. Swamp muck is often acid from the presence of humic acid. Drying the muck removes the acid quality by rendering the muck insoluble in water. If an acid soil becomes neutral by drying, and the water filtered through the dried soil is free from acidity, it is probable that the acid condition was caused by an organic acid of the humus class. But if the acid condition persists after drying tlie soil, the cause is to be sought in sulfates of some heavy metal, e. g., iron or copper, whose sulfates have an acid reaction. Ml^yniOl) 1011 ANALYSIS OF ASHES. PUEI'A RATIO X OF Asil. The material before combustion must be thoroughly cleaned from all foreign matters, especially from adhering soil: woods, barks, roots, etc., by brushing and dusting, wiping with a moist s[)onge, and linally by rublnng gently with a soft cotton cloth; seeds by placing on a line sieve and drenching them with distilled water with constant shaking till the water runs ofp clear, and linally rubbing the seeds between a soft cotton cloth. The ma- terial should then be dried to constant weight at the temperature of bailing water. Combustion of Organic Substances. The combustion should be carried on at a comparatively low temperature, never reaching a full red heat, because of danger of volatilizing alkaline chlorides, etc., nor in a strong draught of air lest the lighter parts of the ash, e. g. , silica, be carried away. C'ombustion is best carried on in a Hat platinum dish in a cast- iron muffle, eighteen inches long, three and a half inches high, and five inches wide at the bottom, the muffle resting on a fire- ])rick inside the furnace to moderate the bottom heat, and the fuel piled upon the top and sides of the muffle to burn the ma- terial by surface heat. When this " Lawes & Gilbert muffle" is not at command the ordinary assay furnace may be used by placing a fire-brick under the muffle, placing the platinum dish and material for combus- tion near the middle of the muffle, feeding the furnace with fuel on the top and sides of the muffle so- as to maintain it at a low red heat, and leaving the plug of the muffle so as to allow a very slow draught of air. When no muffle of any kind is available the substance may be burned to ash in a platinum dish properly guarded. In place of ■ a muffle use the sheet-iron dish commonly employed for a four- inch sand-bath. Place the empty sheet-iron dish on an iron tripod or other support, so that the gas flame from a Buusen 1? burner may cover the whole bottom of the dish. On this dish place a sheet-iron cone (of Kussia iron), six inches high, three and a half inches in diameter at the bottom, and one inch at the top. 8uch dish-and-eone-cover approximates the condition of a muffle for materials ])laced inside the cone. For the incineration use a tiat-bottomed platinum dish, three inches in diameter and one inch deep. Place the material for combustion in the platinum dish, jnit this in the empt}- sheet- ii'on dish, place over the platinum dish and inside the sheet-Iron dish the sheet-iron cone, and heat the sheet-iron dish to low red heat by gas flame. The cone should be made of Russia sheet- iron to avoid the danger of scales of iron rust falling into the ash during combustion. The cost of such combustion a^jparatus is snuill and the mani})- ulation simple, the platinum dish and contents will not be heated to volatilizing alkaline chlorides, there Avill not be sufficient draught of air to carry away any ash, yet the heat within the cone will slowly and securely incinerate the contents of the dish. With substances rich in silica and alkalies it is better to lirst char the substance. \\'ash with distilled water to remove soluble salts, then dry and incinerate the residue. Evaporate the watery extract and add this to the rest of the ash. With substances rich in phosphates, e. g., seeds and animal substances, char the material and remove salts by acetic acid, decant the acetic solution, wash with distilled Avater, and then complete the combustion. Add the acetic solution and washings to the final ash, evaporate to dryness, and gently ignite the whole to decompose the acetates. By this method seeds, etc., may be incinerated in eight to ten hours. In whatever way obtained the whole of the ash should be pn]- verized and intimately mixed before analysis: %* ^ A, ' '^ AxAiAsis OF Wood Ashes. Weigh out one hundred grams of air-dry ashes, and pass them through a smm- sieyey(1a v.eni^i r it^>Lub Lu Lliu in^jfe -)- to separate materials manifestlv/foreign, e. g., nails, broken glass and pot- tery, pebbles, ep»f\iind estimate the per cent of such accidental material&--''lnilverize any charcoal and semi-fused portions of rTremaining on the sieve, sift them and mix intimately with the sifted ashes, and preserve in stoppered bottles for analysis. y-^ Moisture. — ^^'eigh out five grams of these ashes in a tared platinum dish and lieat to llO*^ C, in air bath to constant weight. Cool in desiccator, and weigh. The loss of weight X ^0 = per cent of moisture in the ash. 2^ Carbon. — Heat this dried ash in platinum dish in the sheet- iron ani cone apparatus described for incinerating organic sub- stances till the ash is uniformly grayish-white and there is no further loss of weight; weigh and determine this loss of weight, which X 2*^ =^ P'^"" ^^^^ of charcoal in original ash. 3^ Sand and .silica. — Pla^e this ignited ash in a four-ounce glass- stoppered vial; measure out 50c. c. of HCl (sp. gr. 1.115) and pour on the ash cautiously to prevent loss of ash by spurting, and when all etfervescence has ceased, add the balance of the acid, insert glass stopper, wire it securely, and place in steam bath (described under soil analysis) for two hours; empty the vial into a })latiuum dish, wash the vial with distilled water, adding the washings to the ash solution, and evaporate the whole to dryness on water bath. A'dd 10 c. c, dilute HCl and 50 c. c. of distilled water to the contents of the platinum dish, transfer the contents to a Schleicher & Schuell filter, wash with distilled water till the last drops of filtrate are free from chlorides (when tested by solu- tion of Ag NOs), dry and ignite the precipitate and filter. If there are no grains of sand (revealed by grittiness when stirred with a gla?s rod), subtract the ash of the filter from the weight of tliis residue und estimate the balance as silica. If sand is present, boil the ignited and weighed residue in strong solution of Nas COs to dissolve silica, wash by decantation to remove all soda salts, dry and weigh the sand, the ditference between tlie weight of sand and silica + sand, will give the weight of silica, and this X ^^ ^ P^r cent of silica in the ash. /^# r '^° ^^'^•V^^'''^''i(' n<^id. — Evaporate the acid, filtrate and washings 4^^ A^fronTsTlicato 100c. cTTpTac^his in an Erlenmeyer flask of 250c. c. capacity, add NH4 HO till nearly neutralized, then add 30 c. c. of citro-magnesic* mixture, then 30 c. c. of cone. Nir4ll(), cork * The citro-iHagaesic mixture is prepared by*"dissolvuig-two liuudred and seventy grams of citric acid in 350 c. c. of warm water and adding, by degrees, twenty -seven grams of IMg COs. When effervescence ceases and the Hquid is cool add 400 c. c. of dilute (I to 10) ammonic hydrate, and dilute the whole to a liter. Preserve in a well-stopi)ered bottle. In the presence of a large excess, of NH3, 30 c. c. of this mixture will insure the ])re(ipitatiou of a decigram of P.j Ob even in the presence of ferric and ahuninic salts, unless tlieir <[uantity is excessive. See Suttcm's Volumetric Analysis, 5th edition, page 239. 19 the flask and sliake the flask and contents violently at inter- vals of a few minutes till crystallization is well established. Set the flask in a cool place for four hours, then filter out MgNH4P04 on a tared Gooch filter, wash the precipitate with dilute ammonia (1 to 3), dry the Gooch filter, ignite, at first gently and then intensely, to form pyrophosphate of magnesia. The increase of weight of the Gooch filter equals the pyrophosphate of magnesia from five grams of ash. This multi- plied by *7j7^^e|;X.6396 X ^== l^rff^, will give the percentage of Pz Oa^M ashes. '^y/^t In this method it is important to remove all the silica before precipitating the phosphoric acid. It is also essential to use so much of the citro-magnesic mixture as to prevent the precipita- tion of jDhosphate of iron or alumina. If the addition of the citro-magnesic mixture causes an immediate precipitation, the jsrecipitate is ferric or aluminic phosphate, and not enough of the citro-magnesic mixture was used. In this case the process must be renewed from the separation of silica, and the amount of citro-magnesic mixture increased till no precipitate forms im- mediately after its addition. In this case the addition of NH4HO in excess will cause the complete precipitation of phosphate of magnesia and ammonia after a time, while ferric and aluminic salts will be held in solution. Alternate metliod. — Molybdate of ammonia. The official method for analysis of insoluble phosphates as prescribed for determina- tion of total phosphoric acid is recommended as the alternate method. 5° Carhonic acid. — Heat four or five grams of ash in the sheet- iron and cone muffle till all charcoal is consumed; cool in a des- iccator, weigh out two grams of ash and transfer to a Schroetter alkalimeter. Fill one. chamber of the alkalimeter with HNOs (sp. gr. 1.2) and the other with cone. H2SO4 to dry the escap- ing CO2. Wipe the outside of the alkalimeter from every trace of dust and moisture, and weigh the apparatus. Open the stop- cock of the HN()3 chamber and permit the acid to flow so as to decompose the ash slowly, the COs bubbling up, a bubble at a time, through the H2SO4. AVhen effervescence ceases let the whole of HNOa flow into the reservoir below, attach a CaGla tube to the top of HN()3 chamber, heat the alkalimeter on sand bath to gentle ebullition and suck dry air through the apparatus till (lOi is removed. Set aside the alkaliiiieter till it becomes cold, and then weigh the apparatus. The loss of weight will rc^pre- sent the weight of C'Oi in two grams of ash. AUcrnalr iiii-fhniL — By Liebig's potash bulbs. The usual process of absorption by solution of KIIO, weighing, etc. 6'° C/tloriiic. — Pour out the nitric solution from the alkali- meter upon a filter, wash out the last traces of the solution,, pass the soluble matters through the filter and wash tlie insoluble res- idue with the water acidulktetl.witli IINOa. To mis filtrate add solution of AgXOs to complete precipitation of the chlorides, boil and stir with a glass rod till the silver chloride separates in rtocks, let it settle, decant the clear li(|uid upon a filter, add 100 c. c. of water acidulated with IIXOs and heat to boiling, again decant the clear liquid upon the filter and wash the precipitate with boiling distilled water; finally bring the precipitate upon the filter and wash with distilled water till the filtrate gives no reaction with dilute HCl. Dry the precipitate thoroughly and transfer to a tared porcelain crucible with cover, ignite the filter on the crucible cover, moisten the ash with a drop of HCl, evap- orate the excess of acid, place the lid on the crucible and heat the crucible till the silver chloride begins to melt around the edges. Cool the crucible and weigh. The increase of weight (minus the filter ash) multiplied by .1236 (i. e. ^^ X 50) will give the per cent of chlorine in the ash. This process should be carried on in the absence of direct sun- light. Alt I', mat)' inrlliiKl. — Boil ten grams of ash in 4()() c. c. of pure water for half an hour; transfer all to a measuring flask of 500 c. c. capacity, wash the beaker and add the washings to the flask, cool, make up the volume to 500 c. c. and mix intimately. Filter off through a dry filter 100 c. c, add a drop of solution of plienol- phthalein, and neutralize with dilute HXOs till only a faint pink color renuuns, add two drops of strong solution of Ka Cr ()« and titrate with standard decinormaj solution of Ag XOs (10.95G grams AgNOs @ 1000 c. c). Every c. c. of the standard silver solution equals .003540 grams of chlorine in two grams of ash, or c. c. X .1773 = per cent of chlorine in ash. In [jcrt'onning this titration, watch the reacticm through a plate of amber colored glass of such tint as will neutralize the 21 color of potassic ehromate. The reaction to form silver cliro- mate then becomes sharply defined. The reliability of this method will depend upon the accuracy with which neutralization by nitric acid has been made. The least trace of free acid or alkaline carbonate will vitiate the results. Second alternate metliod. — To 100 c. c. of the solution in the fore- going method add HNOa to strong acid reaction, then solution of AgNOs to complete precipitation, and then proceed as in the first method for estimating chlorine. These alternate methods are based on the assumption that boiling water will dissolve all the chlorides present in wood ashes. 7° Sulfuric acid. — Place five grams of ash in a digestion vial.- Measure out 50 c. c. of HCl. (sp. gr. 1.115) and cautiously pour the acid on the ash till efEervescence ceases, then pour in the rest of the acid, place the glass stopper in place and wire it securely, and place the vial in the digestion steam bath for two hourr. Pour the contents of the vial into a 250 c. c. measuring flask, wash out the vial and add tlie washings to the flask, cool, make up to 250 c. c. with distilled water and mix intimately. Filter through a dry filter 100 c. c. into a beaker, and evaporate on water bath till excess of acid is expelled; add 100 c. c. of distilled water, heat to boiling temperature and precipitate with BaCb in excess. Let it stand foi; twelve hours in a warm place, then decant the clear liquid through a filter, add 100 c. c. of boiling water to the precipi- tate, let it settle and then pour ofl: the clear liquid through the filter, repeating the process till the filtrate is free from chlorides; finally transfer the precipitate to the filter, wash this with dis- tilled water, dry the precipitate and transfer it to a tared cruci- ble, separating the precipitate from the filter as completely as possible, burn the filter separately, letting the ash fall into the crucible, heat this to low redness, cool and weigh. Subtract the filter ash from the increase in weight and multiply the remainder by .34335 for 8O3 in two grams of ash. (Preserve the filtrate and washings for 11°, Estimation of a/kalies-. ) 8° Oxide of iron. — Filter 100 c. c. of the original acid solution (for 7°) through a dry filter, nearly neutralize with ammonia water, then add a gram of sodic acetate and acetic acid till the odor of acetic acid is preceptil)le, boil to precipitate ferric phos- phate, filter while hot and wash precipitate with boiling distilled water till the filtrate is free from chlorides. Dissolve the ferric 22 precipitate on the filter with dilute HaSO* iuto a small Erleii- meyer flask, wash the filtrate, dry and ignite the same and add the ashes to the acid solution in the flask, reduce the ferric to ferrous salt by amalgamated zinc or by a coil of magnesium wire, till a drop of the solution gives no color, with NH* CyS. Pour off the solution of ferrous salt into a beaker, rinse the flask and add the rinsings to the beaker, add freshly-boiled distilled water to make 200 c. c. of the solution, add 2 c. c. of sulfuric acid, heat to 70'^, and titrate with standard solution of permanganate, and esti- mate the iron as ferric oxide. 9^ Lime. — Evaporate the filtrate and washings from ferric phosphate (8°) to 100 c. c. To the hot solution add 20 c. c. of con- centrated solution of ammonie chloride, and 40 c. c. of saturated solution of ammonie oxalate; boil the whole for ten minutes, and then let it stand in a warm place for six hours; decant the clear liquid upon a filter, wash the precipitate twice by decantation, then bring the precipitate upon the filter and wash it free from chlorides and oxalates, testing the washings by argentic nitrate. Puncture the point of the filter with a glass rod, wash the calcic oxalate into a 500 c. c. flask by a stream by the wash bottle, dissolve any oxalate adhering to the filter by dilute H2SO4 (1:10), add 20 c. c. of H2SO4 to the flask and make the volume to 300 c. c. with pure water, heat to 70° and titrate with standard solution of permanganate in which each cubic centimeter of permanganate is equivalent to .0063 grams of crystallized oxalic acid. Each cubic centimeter of the permanganate solution used will be equivalent to .0028 grams of ( 'a O. Alternate itHtlmd for linw. — Dry the washed precipitate of calcic oxalate; transfer the oxalate to a tared crucible, burn the filter on a platinitm wire, letting the ashes fall into the crucil)le, heat the crucible to low red heat, cool and moisten the contents witii a saturated solution of amnionic carbonate, dry and heat carefully to low red heat to expel amnionic salt, cool and weigh. Subtract the weight of filter ash, and estimate the increased weight as Ca C'Os. Second alternate method for lime. — Dry the washed preci]iitate (tf calcic oxalate, transfer to a tared platinum crucible, burn the filter and add the ash to the contents of the crucible. Overflow the calcic oxalate with cone. HaSO , heat gently to dryness, and then intensely to ex))el excess of lIaS04, cool in desiccator and 23 weigh. The increase of weight, minus filter ash, estimate as calcic sulfate, Ca SO4 X .41158 = Ca 0. 10'^ Magnesia. — Evaporate the filtrate and washings from calcic oxalate to 200 c. c, pour into a clean and unscratched Erlenmeyer flask of 500 c. c. capacity, add 30 c. c. of strong solution of (]SrH4)2HP04 and 50 c. c. of cone, ammonia hydrate, cork the flask and shake violently at intervals of a few minutes till crystallization is established, and then set aside for twelve hours in a cold place. When precipitated in this way crystals will seldom adhere to the sides of the flask, yet more perfect crystallization than when stirred with a glass rod. If crystals should form on the sides of the flask they are as readily detached by a rubber-tipped rod as in a beaker. Filter through a tared Gooch filter, wash the precipitate with amnionic hydrate, diluted with distilled water (1 to 3), till filtrate is free from phosphates (acidify a few drops of filtrate with nitric acid and test with molybdate of ammonia). Dry the pre- cipitate, ignite, at first very gently and then intensely, with blast lamp, to convert 2 Mg NH4 PO4 into Mga P2 Ot. Cool in desic- cator and weigh. The increase of weight X -30024: == Mg 0, in two grams of air-dry soil. 11° Estimation of alkalies. * — Concentrate the filtrate and wash- ings from (7°) to 100 c. c, add NH4 HO and (NH4)2C20* to com- plete precipitation of barium and calcium, filter, wash the pre- cipitate, evaporate the filtrate and washings to dryness in plat- inum dish and ignite gently. Add to the residue concentrated solu- tion of oxalate of ammonia, evaporate to dryness and ignite gently. Dissolve residue in distilled water, filter from insoluble Mg 0, acid- ify the filtrate with HCl, and evaporate to dryness in a tared plat- inum dish and ignite gently. The increase of weight represents the chlorides of potassium and sodium in two grams of ash. Separate and estimate potassium by PtCl4, in the usual way, and the sodium by difference. * Test the filtrate from (7°) for lithia by the spectroscope with a loop of platinum wire moistened with the iiltrate, held in a colorless Bunsen flame. The quantity of lithia is usually too small to be determined gravimetrically, but it may be estimated by diluting the solution with distilled water till the lithia line is on the point of disappearing- from the spectrum when a loop of clean platinum wire moistened with the solu- tion is placed in the Bunsen flame. The wire must be clean for each trial, and no concentration of the salt by repeatedly evaporating the solution on the loop of wire without cleaning it. One part of Li CI in 450.000 parts of water will show the lithia line in the spectrum. LIBRARY OF CONGRESS 002 781 560 J-^-' MdiKjtiiii'si'. — Manganese is not a usual constituent of wood aslies. Test the asiies for manganese by heating on platinum foil over a colorless Hunsen flame half a gram of ashes with a gram of sodic carbonate and a few grains of nitrate of potash. The green manganate of soda in the fused portion as it cools will show the presence of manganese. To estimate manganese, dissolve two grams of ash in lU'l. Evaporate excess of acid over the water bath, pour the whole on a filter and wash with distilled water to make 100 c. c, nearly neu- tralize with sodic carbonate, and then add half a gram of sodic acetate and 20 c. c. of strong bromine water and set the flask aside in a warm place for twenty-four hours, or until the bromine has nearly disappeared. Filter out the manganese oxide, wash thor- oughlj', transfer to tared crucible, heat gently, and then intensely, and estimate the residue as ^[na ()4. Mns ()< x .03013 = Mn 0. The manganese may be precipitated by passing a stream of chlorine through the solution till fully saturated, instead of using bromine water. The ashes of mineral coal contain only a small amount of alka- lies and phosphates, but a large amount of insoluble material, clay, etc. Their value depends mostly upon the sulfate of lime and phosphate pi'esent. They are often decomposed with diffi- culty. They should be ground to a fine powder, and five grams placed in the digestion vial, with 50 c. c. of IIC'l (specific gravity l.llo) and digested in the steam bath for six hours, and the sol- uble portion analyzed in the usual way. The aluminic material is in so large proportion that it is better to use the m<)lyl)dic method for estimating phosi)horic acid, after eliminating solul)le silica. LIBRARY OF CONGRESS DDD27aiSt.Dli