Of the water used, four liters are found in the elutriating vessels and nine liters in the receiving vessel No. 5. The apparatus is left standing for an hour until the liquid in the elutriators is clear and the portions in each vessel are received on weighed filters dried at 125°, and the weight of each portion determined.

It is recommended that the loss on ignition of each part be also determined. The separated particles thus secured are classified as follows:

No. 1. Débris and gravel. No. 2. Coarse sand. No. 3. Fine sand. No. 4. Clayey sand. No. 5. Finest parts or clay.

Although the method of Nöbel has been much used, the results which it gives are entirely misleading. The convection currents produced in the conical vessels by the passing water and the flocculation of the soil particles prevent any sharp separation into classes of distinct hydraulic value. The process may be useful for a qualitative test, but its chief claim to a place in this manual is in its historic interest arising from its use in the first attempts at silt analysis.

215. Method of Dietrich.^[148]—The difficulties attending the silt separation by the Nöbel method, led Dietrich to construct an apparatus in which the sides of the elutriating vessels were parallel, but these vessels, with the exception of the first, were not set in an upright position.

The apparatus (Fig. 29) consists of a series of cylindrical vessels connected by rubber tubing.

The elutriators are of the following dimensions:

No. 1. Seventeen centimeters long, 2.8 centimeters in diameter, position upright.

No. 2. Thirty-four centimeters long, four centimeters in diameter, inclined 67°.5.

No. 3. Fifty-one centimeters long, 5.2 centimeters in diameter, inclined 45°.

No. 4. Sixty-eight centimeters long, 6.4 centimeters in diameter, inclined 22°.5.

The rubber tubes passing from one vessel to the other are furnished with pinch-cocks so that each one of the elutriating vessels can be shut off from the others and independently removed from the circuit.

The stream of water is made to pass through the apparatus under a constant pressure of one meter.

Only the fine earth, boiled with water or hydrochloric acid, is to be placed in the apparatus. The part coming through a sieve with a mesh 0.67 millimeters is to be used and placed in No. 1. About thirty grams of soil, are employed for each elutriation. Before adding the soil, the air is completely removed from all parts of the apparatus by connecting it with the water supply and allowing it to be filled with water.

The rate of flow is controlled by the orifice of the last effluent tube and the analyst is directed to continue the operation until the effluent water collected in the beaker glass (5) is clear. The particles then remaining in each of the vessels are collected separately.

The author of the method claims that in respect of likeness of particles the results are especially gratifying and that duplicate analyses give results fully comparable. The process, however, has not commended itself to analysts, but it marks a distinct progress toward the principles of later investigators. Had each of the elutriating vessels been placed upright and the rate of flow determined, the apparatus of Dietrich would have served, to a certain extent, for the more rigid investigations of his successors.

216. Method of Masure.—The sifted earth, from ten to fifteen grams, is carefully mixed with 200 cubic centimeters of water. It is then introduced into a doubly conical elutriator B, Fig. 30, of about 250 cubic centimeters capacity. A current of distilled water is allowed to flow from a Mariotte’s bottle, A, which secures a regular and constant flow. The bottle A is joined to the elutriator B by means of a rubber tube and the vertical glass tube D, the top of which is expanded into a funnel for the purpose of receiving the water from the Mariotte flask. The current of water flowing upward through the elutriator B carries in suspension the most finely divided particles of clay, and these are collected with the emergent water in the receiver C. The sand and coarser particles of clay remain in the elutriator. The water flows out by the tube F, the diameter of which should be less than that of D. When the emergent water becomes limpid the operation is terminated. After the apparatus is disconnected, the water is decanted from the sand in the elutriator, and the whole residue is weighed after drying for two hours at 110°.

The fine soil collected in C may also be separated and weighed, for control, after drying as above.

The pebbles and coarse sand separated by the sieves should also be weighed. By this process the soil is separated into four portions; viz.,

(1) Pebbles. (2) Coarse sand. (3) Fine sand and other materials not carried off by the current of water. (4) Fine soil, carried into the receiver C.

217. Method of Schöne.—The method of Schöne^[149] is based on the combination of a cylindrical and conical separatory tube through which the flow of water is regulated by a piezometer.

If, in the process of silt separation, the water move perpendicularly upward with a given velocity, e.g. = v the separation is dependent:

(1) On the volume of the silt particles, (2) On their specific gravity, and, (3) On their state of disintegration.

If it be assumed that the silt particle is a sphere with a diameter = d, then according to Newton’s law of gravity, the following formula would be applied: d = v² (3Z
4g (S − 1)).

In the above formula Z = a coefficient which depends on the condition of the surface against which the hydraulic pressure or resistance works, in this case a sphere; g = the acceleration of gravity equivalent to 9.81 meter; and S = the specific gravity of the particle.

This expression signifies that in a given case, the velocity of the current in the apparatus is just sufficient to counteract the tendency of a given particle to sink. All particles of a smaller diameter, in such a case, will be carried on by the current, while all of a greater diameter would separate by sedimentation. These theoretical conditions are not met with in practice where silt particles of all shapes and degrees of aggregation abound. These particles, whatever their shape, may be said to have the same hydraulic value when carried by the same current. It is necessary, therefore, to secure some uniform standard of expression to assume a normal form of particle and a normal specific gravity. For the form, a sphere is evidently the normal which must be considered and for specific gravity that of quartz is taken; viz., 2.65. The mean coefficient for Z may also be placed at 0.55, although slightly different values are ascribed to it. Substituting these values in the formula, it is reduced to the expression; d = v² × 0.0000255 millimeters. It can, therefore, be said that by this or that velocity of the current, silt particles will be removed of this or that diameter, it being understood that all particles of equal hydraulic value to spherules of quartz of the given diameter are included in each class. In order to have the theoretical formula agree with the results of analysis it is necessary to modify it empirically to read d = v^⁷⁄₁₁ × 0.0314 millimeters.

This formula is found to agree well with the results obtained for all velocities between 0.1 millimeter and twelve millimeters per second, the ordinary limits of silt separation.

The Apparatus.—The conic-cylindrical elutriating vessel A, B, C, D, E, F, G, Fig. 31, is of glass. The part B, C, is cylindrical, ten centimeters in length and as nearly as possible five centimeters in diameter.

The conical part C, D, is fifty centimeters in length. Its inner diameter at D must not be greater than five centimeters nor smaller than four centimeters.

The bend, D, E, F, should have the same diameter; viz., four to five centimeters.

The part A, B, C, D, and D, E, F, G, may be made of separate parts and joined by a rubber tube.

Outflow Tube and Piezometer.—The outflow tube and piezometer, H, J, K, L, is constructed as shown in Fig. 32. It should be made of barometer tubing having an internal diameter of about three millimeters. The tube is bent at J at an angle of forty to forty-five degrees. The knee J must be as acute as possible not to interfere with the inner diameter. The form and especially the magnitude of the outlet are of great importance. It must be circular and nearly 1.5 millimeter in diameter. It must not be larger than 1.67 millimeter nor smaller than 1.5 millimeter. The opening should be so made as to direct the stream of outflowing liquid in the direction shown by the arrow.

The piezometer L, K is parallel to the arm H, J, of the delivery tube. Its graduation has its zero point in the center of the outlet K. It commences with the one centimeter mark. From one to five centimeters it is divided into millimeters, from five to ten centimeters into one-fourth centimeter, from ten to fifty centimeters into one-half centimeter, and from fifty to 100 centimeters into centimeters. The dimensions given are those required for ordinary soils and for velocities ranging from two-tenths millimeter to four millimeters per second.

For greater velocities, a delivery tube with a larger outlet must be used and the piezometer must be of greater internal diameter than indicated.

Arrangement of the Apparatus.—The apparatus is conveniently mounted as shown in Fig. 33, giving front and side views of all parts of apparatus in position ready for use. When numerous analyses are to be made much time is saved by having a number of apparatus arranged en batterie.

The Sieve.—The soil, before being subjected to elutriation, should be passed through a sieve of which the meshes are 0.2 millimeter square.

The Process.—To measure the diameter of the cylinder, two marks are made with a diamond upon the glass which are distant from each other a certain space, for instance, h centimeters. The space between these two marks is filled with water exactly measured. Suppose that a cubic centimeters were used, then the diameter is determined by the formula:

In order to determine that the elutriating cylinder is strictly comparable in all its parts this measurement should be made upon several parts thereof.

The apparatus should now be tested in regard to the quantity of liquid which it will deliver under a given pressure in the piezometer. By means of the stop-cock H the flow of water is so regulated that the outflow at c can be measured at a given height of the water in the piezometer. Suppose that a cubic centimeters of water flow in t seconds, then the quantity which would flow in one second is determined by the formula, Q = a
t cubic centimeters. Since according to the law of hydraulic outflow the quantities are proportional to the square root of the height of the column it is easy to compute from any given height the quantity which will flow from any other one desired. For the retardation due to capillary attraction, it is sufficient, in general, to take it in a constant quantity; if this constant quantity be represented by C, the observed height of the water in the piezometer by h, and the quantity of water flowing out by Q, the data required for any given velocity can be calculated from the following proportion:

It is necessary to compute the magnitude of this constant C which is to be subtracted. This is accomplished by measuring the quantity of water which flows out at two different heights of the column in the piezometer. From the foregoing proportion, the value of C is as follows:

The value of C can be the more exactly determined as h₁ is greater and h₂ smaller. It is best to choose the lowest height from which an exact reading can be made; that is, by which the regular rise and fall of the level of the water in the piezometer (in consequence of the formation of drops) just begins to disappear. This usually takes place when h₂ = 1.5 centimeter to 1.7 centimeter. For the higher value h₁ it is best to take about 100 centimeters. Suppose, for example, the following results are obtained:

The same quantity of water which flows out in a unit of time passes also at the same time over a cross section of the elutriating cylinder. The diameter of this cylinder being D the equation is derived

Since the velocity in the elutriating cylinder v is directly as the quantity of water overflowing so is

is obtained from the means of a number of estimations; for example as illustrated in the following data:

Then are obtained the following values of hₙ and vₙ:

In order to be able easily and rapidly to judge under what pressure the outflow has taken place in any particular instance, a larger number of values are computed with the help of the formula given and placed together in tabular form. As an example the following table may serve which was computed for one of the apparatus used. Usually it will be sufficient to test the apparatus for four different heights and then to interpolate the values for all the others. The numbers marked with a star in the table are those which were determined by experiment; the others were calculated.

Suppose the problem is by means of the apparatus tested above, to separate into a number of groups a mixture of silt particles, whose hydraulic values are found between the following diameters: 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07 millimeters. The table will show at once under what pressure of water the piezometer must be placed in order to give the values; viz., 1.4, 2.8, 7.0, 15.0, 29.0, 53.0, and 83.0 centimeters respectively.

The apparatus described above, is adapted for velocities in the elutriating cylinder varying from two-tenths millimeter to four millimeters per second. The largest silt particles which can be separated by the velocities given above, have approximately a diameter of 0.08 millimeter. For the separation of larger particles a sieve can take the place of the silt apparatus. If, however, it be desired to subject larger particles to silt analysis, the dimensions of the elutriating cylinder and of the outlet of the delivery tube must be changed accordingly.

Preparation of Sample.—The conduct of silt analysis of natural soils must, in certain cases, be preceded by a special treatment of the sample. If the latter be rich in humus the organic substance must previously be separated as completely as possible. With sandy soils this can be accomplished by ignition. With clayey soils, on the contrary, it is to be performed by boiling the soils at least one hour with water which contains from one to two per cent of free alkali. Soils which contain lime must also be subjected to treatment with dilute hydrochloric acid, and the hydrochloric acid must be as carefully removed, as possible before the sample is subjected to elutriation; afterward follows the boiling of the sample in the ordinary way with water. This, of course, can be omitted when it has already been treated with boiling dilute alkali. It is also important to remove the larger particles by a sieve before the elutriation begins. It is well to pass a sample through a sieve after it has been boiled, by which all particles of a larger diameter than 0.2 millimeter are removed. This will usually require about one liter of water and this water should be allowed to rest from one to two hours and poured off with the suspended material which it contains. Only what subsides should be brought into the apparatus. In rinsing the sample as much water must be used as will fill the apparatus up to its cylindrical portion.

After the sample has been placed in the apparatus, the water is allowed slowly to enter, being careful to avoid reaching more than the lowest required velocity, until the outflow begins. The water then is so regulated by the stop-cock as to bring it to the desired height in the piezometer. This being accomplished, the different velocities which have been decided upon for separating the particles of silt are used one after the other, as soon as all the silt which can be removed at each given velocity, has been secured. From three to five liters of water will be required for the separation of each class of particles. Sometimes the reading of the height of the water in the piezometer is difficult; as, for instance, when foam or bubbles accumulate therein. These bubbles can be removed by simply blowing into the tube, or dropping into it a little ether. The outflow of water can be received in vessels, beaker glasses, or cylinders, in which it is allowed to subside. The finest particles which remain in suspension in the water are best determined by difference. If it be desired to weigh them directly, the water can be treated with ammonium bicarbonate until it contains from one to two per cent thereof. The precipitation then takes place in a few hours.

The collection and weighing of silt particles are accomplished in the usual way. That which finally remains in the elutriating vessel is taken out after the end of the operation by closing the stop-cock, removing the stoppers with the piezometer tube, pouring the contents of the elutriating vessel into a beaker glass and rinsing out carefully all adhering particles. Examples of the working of the apparatus follow:

The soil was taken from the Imperial Russian Agricultural Experimental Institute at Gorki. It was a fine clay sand and was carefully treated with hydrochloric acid. The results of the analysis are given in the following table:

Holthof modifies the apparatus of Schöne by putting into the lower mouth of the elutriator a little mercury so that the particles of earth are deposited upon its surface and are thus better agitated and washed by the current of water.

218. Mayer’s Modification of Schöne’s Method.—An improvement of Schöne’s apparatus in the direction of greater simplicity has been tested by Mayer^[150] with satisfactory results:

The apparatus, (Fig. 34), consists of a glass vessel having a glass stop-cock at the bottom for admitting the water. For a distance of twenty centimeters the sides of the tube are parallel and the diameter about one centimeter. Next for a distance of fifty centimeters the tube is conical expanding at a regular rate until the internal diameter reaches five centimeters. For a distance of ten centimeters the vessel is again strictly cylindrical and it is in this cylindrical portion that the separation of the different constituents takes place. The vessel is then rapidly narrowed until it carries the stopper A two centimeters in diameter. This stopper carries two glass tubes, one F bent downward to conduct the overflow into the receiving vessels, and one H for the purpose of regulating the rate of overflow by the height of the column of water therein. The orifice of the overflow tube F should be so regulated that with a pressure of five centimeters water in H, one liter shall pass over in ten minutes.

If the separation be conducted in an apparatus thus mounted and graduated with a pressure of two centimeters in H all that portion of the soil which can properly be called clay will pass over. The fine earth, that is, earth in which all coarse particles have been removed by proper sifting, is used in ten-gram lots for each experiment. The residue, after the separation is complete, consists of pure sand or at least pure sand mixed with humus. Before the fine earth is placed in the apparatus, the calcium carbonate therein is removed with hydrochloric acid. The treatment with hydrochloric acid, however, is not to be recommended in soil containing many undecomposed particles of calcium carbonate or dolomite for then large additions to the silt output might be made from these particles, which could not be regarded as coming from the soil as it actually exists. For alluvial soil, however, previous treatment with hydrochloric acid is recommended unconditionally.

219. Schöne’s Method as Practiced by Osborne.—The apparatus used by Osborne^[151] was obtained from Germany and was similar to that described by Schöne in his original paper, except that it was furnished with a second elutriating tube as suggested by Orth. The modification made by Orth consists, essentially, of a second elutriating tube with straight sides into which the bulk of the soil is introduced, only the final part being carried over into the Schöne’s tube proper. Water is supplied to the apparatus under constant pressure by means of a Mariotte’s bottle.

The preliminary treatment recommended by Schöne is omitted, as these steps have been shown to be undesirable, on account of affecting the accuracy of the results.

Twenty grams of the air-dried soil are passed, under water, through a sieve of one-fourth millimeter mesh. That part of the soil which remains in suspension after being sifted is placed at once in the Schöne’s tube of the apparatus, the coarser portion being rinsed into the Orth tube. The current is regulated so that the largest particles of quartz carried off have an average diameter of 0.01 millimeter. When all is carried off that can be removed at this rate the current is increased until the largest quartz grains passing off have a diameter of 0.05 millimeter.

As noticed by Hilgard with Schulze’s apparatus, secondary currents are formed during the process of elutriation which descend along the walls of the conical portion of the Schöne’s tube and some distance along the sides of the cylindrical portion. The tendency of these currents is to produce globular aggregates of particles which fall to the bottom. They are broken up from time to time by increasing the velocity of the current but even this method fails to disintegrate a considerable quantity of them.

220. Statement Of Results.—Two samples of soil from the garden of the experiment station analyzed by Schöne’s method gave the following proportions of sediment.

In the table the term clay is used to designate all that part of the soil which has diameters less than 0.01 millimeter and which remains suspended after twenty-four hours standing in water having a depth of 200 millimeters.

The last column C represents the average of three direct beaker elutriations according to the method of Osborne.

The differences which these figures show are found to be due to imperfect separation of the finer grades from the coarser and even when the various fractions separated by the Schöne method are subjected to beaker elutriation and the portions separated from them added to the grades to which they properly belong the Schöne elutriator was found to effect far less exact separations than the beaker method.

Samples of prairie soil from Mercer County, Ill., not boiled, were examined by the two methods with the following results:

In this case it is seen that Schöne’s method varies considerably from the beaker method and if the beaker method be regarded as correct the Schöne method is evidently less reliable.

In the next table are given the data of the examination of brick clay from North Haven, Conn., by the two methods.

The failure of the Schöne method to give the results obtained by the beaker method is ascribed to the fact that it is impossible for the current of the strength used to disintegrate the clay and further that the particles after they are once separated tend to coalesce by the currents produced by the elutriating process.

221. The Berlin-Schöne Method.—Osborne has also made a study of the Schöne method as modified by the Bodenlaboratorium of Berlin. The directions for the analysis by this laboratory method are as follows:

Five hundred grams of the soil are sifted through a sieve with circular holes two millimeters in diameter. Of the earth passing the sieve from 30 to 100 grams are boiled in water with constant stirring from one-half to one hour or longer, according to the character of the soil. The finer the texture of the soil the smaller the quantity taken and the longer the time of boiling. Treatment with acids or alkalies is not practiced.

The finer portion of the soil remaining suspended in the water, after boiling, is poured into the Schöne tube, the remaining coarse part is rinsed into the Orth tube. The clay, together with the finest sand, is collected in a separate vessel, the water in which it is suspended is evaporated and the residue after drying in the air is weighed. The rest of the operation is carried out as previously described except that the products of elutriation are not ignited but weighed air dried, in order that they may be further examined, chemically if desired. By proceeding in this manner the following results were obtained:

For the sake of comparing the mechanical separation attainable by this procedure with those yielded by other methods, the air-dried products were ignited and again weighed and examined.

By subtracting from the ignited portion above 0.05 millimeter, 49.37 per cent, the amount of this soil that remained on a 0.25 millimeter sieve, the fraction between 0.25 millimeter and 0.05 millimeter is found, and the separations in this analysis may be compared with those previously obtained by the beaker method as follows:

Osborne concludes from the above facts that the Berlin-Schöne method, while showing close agreement with the beaker method, does not give results which are identical with that method. On subjecting portions separated by the Berlin-Schöne method to the beaker analysis additional separations were secured. In the case of heavy loams the inability of the Berlin-Schöne method to effect even a rough or approximate separation of the several grades becomes very conspicuous.

222. Method of Hilgard.—Two important principles lie at the foundation of this method; viz., 1, the use only of separating vessels of true cylindrical shape and 2, the employment of a mechanical stirrer to break up the floccules formed during the process of separation. The points in the apparatus to be considered are uniformity of the cross section of the elutriator at every point, exact perpendicularity of position, careful control of the rate of flow and continuous operation of the mechanical stirrer. According to Hilgard’s observations the stirring due to the current of water alone is not sufficient to break up the floccules unavoidably formed during the separation, while any inclination of the sides of the elutriating vessel from the perpendicular due either to a conical shape or false position favors in the highest degree the formation of floccules due to reflex currents formed in the body of the liquid.

In order to carry out the idea suggested by Türschmidt of substituting for the accidental and indefinite products usually appearing in the statements of silt analyses sediments of known and definite hydraulic value a constant head of water is used, secured by means of a Mariotte’s bottle connecting with the tube delivering the current through a cock provided with an arm moving on a graduated arc.

According to Hilgard the separation of sediments by the method of subsidence does not possess the analytical accuracy of the moving liquid method, especially when the latter is combined with mechanical stirring. The subsidence method requires close and continuous attention and in the case of fine sediments tending to flocculation the difficulties of the method are greatly increased. The views of Hilgard in respect of the laboriousness of the subsidence method lose, however, some of their force since the modifications of Osborne have come into use. The simplicity and cheapness of the apparatus required for subsidence give it at the start many advantages over the more elaborate process with a churn elutriator. For rigid scientific investigation, however, the method of Hilgard is commended as a standard of comparison in all cases.