In all the work reported, five grams were used, as the soils contained so much humus that this amount gave enough humus for good work in the final weighings. The results obtained so far appear in the following tables:

Note.—Numbers in parentheses indicate results, generally the earliest ones, which the authors do not consider strictly comparable with the rest of the work. They are given solely for the purpose of exhibiting all the work that has been done to date. When a mean is included in parentheses it indicates that it is calculated from all the results obtained, including those not considered strictly comparable. Bogus is a name given to a peaty soil which is very sterile.

313. Summary of Results.—1. The modified method gives much higher results than the original method of Grandeau.

2. In the Grandeau method marked irregularities follow a change in the strength of the ammonia solution. These differences in results bear no relation to the strength of the solution used. They seem to be errors due to the difficulty of securing uniform and complete washing of the soil by the ammonia solution.

In the modified method the change in the strength of the ammonia solution makes practically no difference in the amount of humus extracted, except in the case of the peat soil where two per cent ammonia failed to extract all the humus. But the results show no considerable increase when the strength is increased to over four per cent.

3. The factor of time has not been fully investigated, but the results so far obtained indicate that the time exerts less influence in the modified than in the Grandeau method.

4. Table III shows that considerable quantities of the peat soil are still passing into solution in the Grandeau method at the end of ten days. With ordinary soils this is not true; but in the case of soil No. 5, a black soil, the solutions were colored at the end of a week. On the peat soil the modified method extracted from ten to fifty per cent more than the Grandeau, and on the ordinary soil from two to three times as much humus.

5. In comparing duplicate results by both methods it is found that with soil No. 3, peat soil, the following differences appear calculated to percentage of the total amount involved in the determination:

Special attention was paid to this point in case of soil No. 7, an ordinary soil; taking all results into consideration the greatest difference in percentage of total amount involved was, by the modified method, nineteen per cent, and by the Grandeau, thirty per cent. In the set of six special determinations made by both methods to test this point and which are strictly comparable with each other, the maximum range was by the modified method 7.8 per cent and by the Grandeau 8.3 per cent of the total amount involved in the determination. From which it appears that the modified method is on the whole capable of yielding rather more concordant results than the Grandeau.

314. Estimation of Free Humic Acids.—This process, due to Müntz^[201] is essentially that of Huston and McBride. Twenty grams of the soil are reduced to a fine powder and saturated with fifty cubic centimeters of concentrated ammonia and allowed to digest two or three days in a warm place. The volume is then made up to one liter with water, well shaken, and set aside for one day in order to permit the subsidence of the solid matter. At the end of this time 500 cubic centimeters of the supernatant liquor are taken and acidified with hydrochloric acid in order to precipitate the humic bodies. The humus is collected on a filter, dried and weighed. It is then ignited and the weight of ash deducted from the first weight thus giving the actual weight of the humus obtained, free from mineral matter. This process gives the free humic acids. By previous treatment of the sample with hydrochloric acid as in the process of Huston and McBride, the total humus is obtained. The estimation of the free humic acids is of importance in determining the quantity of lime or marl which should be added to acid lands.

315. Humus Method of Von Bemmelén.^[202]—Von Bemmelén obtains the content of humus by the multiplication of the content of carbon in the soil by the factor of Wolff; viz., 1.724. The estimation of carbon, water, and of the loss on ignition is conducted in combustion tubes in a current of oxygen. The nitrogen estimation is carried on according to the method of Dumas.

In soils containing calcium carbonate the carbon content is derived from the carbon dioxid taken up by the potash bulbs during combustion (a); from other carbonates not decomposed on ignition and which are subsequently determined in the residue by treatment with hydrochloric acid in a carbon dioxid apparatus, (b) and the total carbon dioxid derived from the carbonates in the soil (c).

For each estimation from three to five grams of the soil are taken, because with smaller quantities the errors of analysis too strongly influence the results. The carbon is then calculated according to the formula:

The Carbon Dioxid of Carbonates.—It is necessary to expel the carbon dioxid at ordinary temperatures, because on heating to boiling, carbon dioxid would be formed from the humus. In a flask, as small as possible, the soil is treated at ordinary temperature, with dilute sulfuric or citric acid, the escaping gas dried over sulfuric acid and taken up with soda-lime. Behind the soda-lime is a small tube filled with pieces of glass and moistened with sulfuric acid, which retains any moisture taken out of the soda-lime. A stream of about one liter of air, free from carbon dioxid, is sufficient to drive out all of the carbon dioxid when the estimation is made at ordinary temperatures.

A volcanic earth from Deli, which contained five per cent of humus, gave, at a temperature plus or minus 15°, 0.01 per cent CO₂. At boiling temperature two analyses gave 0.54 and 0.56 CO₂. This soil contained no carbonate, and the carbon dioxid found at the boiling temperature, must have come from the humus substances under the influence of the dilute acids.

A heavy clay containing 6.9 per cent of humus gave, at plus or minus 15°, 3.60 per cent CO₂; at 100° without boiling, it gave an additional 0.53 per cent, and with boiling an additional 0.11 per cent, or a total of 4.24 per cent CO₂. A light clay containing 3.2 per cent of humus, gave, at 15°, 5.09 per cent CO₂; at a boiling temperature an additional 0.43 per cent, and by continued boiling an additional 0.27 per cent.

316. Estimation of Humus by the German Method.—The German experiment stations follow the method of Loges,^[203] depending on the oxidation of the humic bodies with copper oxid after evaporation of the sample with phosphoric acid. The object of the preliminary evaporation is to set the humic acids free in order that they may be better and more easily oxidized than when burned in the combined state.

The sample of soil is placed in a Hoffmeister dish (Schälchen), moistened with dilute phosphoric acid and evaporated to complete dryness. The dish and its contents are rubbed up with pulverized copper oxid and placed in a combustion tube of sixty centimeters in length, open at both ends. There is then placed in the tube, and held in place by asbestos plugs, granular copper oxid to a length of twenty centimeters.

The combustion tube is placed in a proper furnace and one end connected with two washing-flasks, the first containing potash lye, and the other a solution of barium hydroxid. These flasks are to free the aspirated air from carbon dioxid. The other end of the combustion tube is connected with an appropriate apparatus for absorbing the carbon dioxid. Loges recommends the Pettenkofer absorption tube and a Fresenius drying cylinder.

Between the absorption apparatus and the aspirator, is also placed a washing-flask containing barium hydroxid solution, serving to detect any unabsorbed carbon dioxid. The layer of granular copper oxid is first heated, the air being slowly aspirated through the apparatus meanwhile, but not through the absorption bulbs. All the carbon dioxid is thus removed from the apparatus.

The absorption system being connected, the tube is heated slowly from the front, backwards, and after the tube is well heated a slow current of air is drawn through and continued until the combustion is complete, which is usually in about three-quarters of an hour.

After the tube is cool the powdered copper oxid and residue of combustion are removed, and for this reason the tube is stopped with a cork at both ends instead of being drawn out and sealed at one end. The tube can thus be refilled without disturbing the granular layer of copper oxid.

The drying cylinder used between the combustion tube and the absorption system has its upper part filled with cotton to avoid the deleterious effects of the nitric oxid produced in the combustion. With this arrangement the use of metallic copper in the combustion tube to reduce the nitric oxid can be dispensed with, the moist cotton holding back the acid fumes. The per cent of humus is obtained by multiplying the per cent of carbon found by 1.724.

317. Method of Raulin for the Estimation of Humus.^[204]—The volumetric estimation of humus in soil by a solution of potassium permanganate would be convenient and practical if the combustion of the organic matter were complete, and if the browning of the liquor did not render the end of the reaction uncertain. The process of Schmidt, modified as below, has given satisfactory results.

In a small flask, with flat bottom, containing about 250 cubic centimeters, are introduced ten cubic centimeters of a solution of manganese sulfate containing sixteen grams of the anhydrous salt per liter, and ten cubic centimeters of a ten per cent solution of potassium permanganate. The solution is heated for a few minutes, the liquor is decolorized and manganese bronze is precipitated. One hundred cubic centimeters of water are added, and four cubic centimeters of sulfuric acid containing 150 cubic centimeters of monohydrated acid per liter. There is now added an exactly measured volume of the humic liquid properly prepared, so that in oxidizing completely it destroys at most only half of the manganese dioxid. The mixture is submitted to gentle ebullition for eight hours, the water being kept at a constant volume. The excess of manganese dioxid remaining is dissolved hot by a measured portion of decinormal oxalic acid in slight excess, and the excess of oxalic acid is removed by a solution of potassium permanganate containing one gram per liter. The volume of oxalic acid not destroyed by manganese dioxid is calculated from the amount of permanganate consumed. The volume of oxalic acid, which corresponds to the same quantity of dioxid as the introduced humus, is also calculated by taking the difference between the volume of oxalic acid necessary to destroy all the dioxid formed by ten cubic centimeters of the ten per cent permanganate solution, and the volume of the oxalic acid which has destroyed the dioxid remaining after the action of the humus. The first volume of oxalic acid, that is to say, that which destroys the dioxid formed by ten cubic centimeters of ten per cent permanganate is determined in a preliminary titration.

In regard to the humic liquor, it is prepared by treating ten grams of earth with soda solution in the usual manner. It will be easy to calculate the volume of the oxalic solution equivalent to the total volume of the humic solution, of which a determined fraction has been assayed, and consequently the volume of oxalic solution equivalent to the humus in ten grams of the dry earth. This number of cubic centimeters of the decinormal oxalic solution multiplied by 0.8 will express in milligrams the weight of oxygen necessary to burn the humus from ten grams of dry earth. Humus not being a definite compound, but a residue of complex organic matters partially oxidized, it will require as much more oxygen to complete the combustion as the previous oxidation has been less pronounced. This weight of oxygen necessary to burn the humus from ten grams of dry earth may serve to detect the total value as well as the weight of the humus itself. However, if we wish to have directly the weight of the humus, resource can be had to a table which, without being rigorous, can be regarded as sufficiently exact when the variability of the constitution of humus is taken into account.

318. Pasturel’s Method.—According to Pasturel^[205] the process of Raulin does not furnish figures that are rigorously exact only with soil of which the humus contains forty-five per cent of carbon. When the richness in organic carbon is less, the results of the estimation are too high. Pasturel modifies the process as follows:

Manganese Sulfate.—Dissolve sixteen grams of the pure anhydrous manganese sulfate in distilled water and make the solution up to one liter.

Potassium Permanganate.—Make a solution of ten grams of potassium permanganate in one liter of water; 100 cubic centimeters of the liquor just mentioned are diluted to one liter and constitute the potassium permanganate solution one to ten.

Oxalic and Sulfuric Acids.—A solution of oxalic acid is prepared containing 6.3 grams of the acid in one liter of water, and a dilute solution of sulfuric acid, by dissolving 150 grams of the monohydrated acid in one liter of water.

Humus Solution.—The solution of humus is prepared by the following process: Ten grams of fine earth are freed from all their carbonates by dilute hydrochloric acid. After washing, the filter is broken and the dirt is washed into a small flask. Not more than twenty or thirty cubic centimeters of water should be employed for this purpose. Twenty cubic centimeters of a liquor containing two grams of caustic soda are added, and the flask is placed upon a sand-bath and maintained at a boiling temperature for six hours. It is then diluted with water, filtered and washed as long as the waters are colored. The liquor is treated with dilute sulfuric acid until almost the whole of the soda is saturated. It is indispensable, however, to maintain a slight alkalinity in order that the organic matter may rest totally dissolved. The precipitation of silica which is almost always produced is without inconvenience. Afterward the volume is completed to 500 cubic centimeters and the humus solution is then ready for use.

Estimation of the Humus.—Ten cubic centimeters of the manganese sulfate are placed in a flask and ten cubic centimeters of the permanganate added, and the whole is then slightly heated, and afterward 100 cubic centimeters of water and four cubic centimeters of sulfuric acid are added. The humic liquor is now introduced in such proportion that the humus which it contains dissolves at the greatest, a half of the precipitated manganese and the rest of the process is continued as described by Raulin.

319. Estimation of Carbonates in Arable Soil.—The principle of the determination depends on the liberation of the carbon dioxid from its compounds in the soil by acting on them with strong acid, and the desiccation, absorption, and weighing of the evolved gas. Any of the ordinary forms of apparatus for estimating carbon dioxid may be used in this determination.

The apparatus of Knorr^[206] has been used with satisfaction for many years in the laboratory of the Department of Agriculture.

The apparatus consists of a flask A, Fig. 65, in which the carbon dioxid in the soil is liberated. A condenser, D, fits by means of a ground-glass joint into the neck of the flask in which the liberated gas, together with any air or aqueous vapor which may be carried forward, is cooled. This prevents any excess of vapor of water from entering the absorbing bulbs, which could easily happen at the end of the experiments when the contents of A are raised to the boiling point. The bulb B contains the acid, usually hydrochloric, which is employed for decomposing the carbonates. It is provided with a guard bulb-tube, C, which serves to absorb any carbon dioxid which might enter the apparatus with the air during aspiration at the close of the determination. The carbon dioxid is dried in the bulb-tube, E, in oil of vitriol, and absorbed in the potash solution in F. It is advisable to aspirate a slow current of air through the apparatus by means of the tube G during the whole of the operation. The quantity of the sample to be taken depends on its richness in carbonates. Many soils are so poor in carbonates as to render any attempt at exact determination nugatory. On the other hand, a comparatively small sample of marls will be sufficient. A preliminary qualitative test will indicate, in a general way, the quantity of the sample to be taken. The sample of soil, five to fifty grams, having been transferred to A, which should be perfectly dry, is made into a batter with freshly boiled distilled water. When all the parts of the apparatus are properly connected gas-tight, the cock between B and A is slowly opened and the hydrochloric (nitric) acid in B allowed to flow into A at such a rate as will secure a moderate evolution of gas.

When the carbonate is entirely decomposed, a lamp is brought under A and its contents gradually raised to the boiling point. The aspiration of air, free from carbon dioxid, is meanwhile continued until all the liberated gas has been absorbed in F. Usually about fifteen minutes will be sufficient to accomplish this purpose.

320. Bernard’s Calcimeter.—For a rapid and approximately accurate method of determining the amount of carbonate in the soil, estimated as calcium carbonate, Bernard makes use of the well-known method of the volumetric estimation of carbon dioxid. The sample to be examined should not be powdered in any way. The sample in a natural state, but well air-dried, is gently broken up by the fingers and passed through a sieve having ten meshes to the centimeter. Of the fine earth thus obtained, one gram is taken, for the determination. If the percentage of carbonate in the soil exceeds fifty then only half a gram is taken.

The apparatus employed is one well known. The small erlenmeyer C is fitted with a rubber stopper carrying an exit tube for the gas and a small thermometer. This flask is connected by means of a rubber tube and small glass tube to the measuring burette B. This burette is graduated from 0 to 100 cubic centimeters. Below, by means of a rubber tube, it is connected with the open bulb A, which, by means of a cord about its neck, can be suspended by the hook as shown in the figure. The measuring tube is filled with water through A until the level of the liquid in B is slightly above the zero mark. Meanwhile the one gram of earth has been placed in C, together with the tube D three-fourths filled with an equal mixture of water and strong hydrochloric acid. The greatest care must be taken that no part of the acid be spilled.

The rubber stopper is now forced into C until the level of the water in B is just at the zero mark. Grasping C in the right hand and A in the left, the operator inclines C until the contents of D are emptied. Meanwhile as the gas is evolved, A is lowered at such a rate as to always keep the level of the water in B and A on the same plane. In a few moments the evolution of gas is complete, and the volume given off is read at once without correction. This volume multiplied by 0.4 gives the percentage of carbonate in the sample examined. It is understood that the determination is made at ordinary temperatures; viz., 17° to 22°. Example:

One gram of a soil treated as above, gave of carbon dioxid (uncorrected) 65 cubic centimeters. 65 × 0.4 = 26.00 = per cent calcium carbonate in sample.

The above method is useful in the classification of soils and in determining approximately the quantity of calcium carbonate which they contain. The practical use of this method is of great value in determining the character of fertilizer to be applied. It is well to know the percentage of carbonate in selecting mineral fertilizers.

321. Soils Deficient in Carbonates.—When a soil contains but a small quantity of carbonates, Müller^[207] has called attention to the fact that the carbon dioxid absorbed by the water in which the soil is rubbed up may vitiate the result. Instead of water a titrated solution of sodium carbonate is employed. The apparatus is composed of a flask containing the mixture of the sodium carbonate and the soil on which the hydrochloric acid is to act. The hydrochloric acid is contained in a small tube, as in Scheibler’s apparatus. The gas is received in a rubber tube 1.5 meters long and three to four millimeters interior diameter, and connected with a burette, the open mouth of which dips into the water of a cylinder of proper length. The volume of gas is read when the burette is raised or lowered in the cylinder until the liquid within and without stands at the same level.

During the action of the acid on the carbonates the flask is constantly shaken.

Several readings of the volume of gas are made, the evolution flask being vigorously shaken before each one. Finally, in order to allow for the variations in temperature and pressure of the exterior air which may take place between the beginning and the end of the reaction, a second flask containing air is placed by the side of the evolution flask and communicating with a narrow ᥩ tube half filled with water. Any variations in the volume of the air in the flask will be shown by variations in the height of the liquid in the two arms of the ᥩ tube, and the volume of the variation can be easily determined by having the ᥩ tube calibrated.

If now a equals the volume per cent of carbon dioxid in the atmosphere of the evolution flask at the end of the reaction, v the volume of gas disengaged, and V the volume of the atmosphere in the evolution flask, the per cent of carbon dioxid contained in a given length of the rubber tube will be equal to a
2. This arises from the fact that the first gas which passes into the rubber tube is composed solely of air, while the last contains a per cent of carbon dioxid. By reason of the shaking of the flask the mean richness of the contents of the tube in carbon dioxid, will be sensibly a
2.

From the above data the following equations are derived:

1. va
2 + Va = v.