I | | | | | || l| il | | tlh | Be ee cee ee ne eae ee en cree ies 9 > STUDIES ON THE METHOD OF | QUANTITATIVE DETERMINATION OF THE SOIL COLLOIDS : sna MITCHAEL IVANOVITCH WOLKQFF eB Os) ‘ t eat etd Peete ar ireal he arelt avai Rede St Bk 4 , Saeiees --ssehasan irish Soeas Re IN ot eae ach : THESIS sluagQlaes ON THE BVETHPQLD AVE QUANTITATIVE DETEBYINATIQON QE THE SQIL COYLLOQIDS. Thesis for Degree of M. S. Michael Ivanovitch Wolkoff 1916 Tht FSIS TABLE OF CONTENTS. Pase. DOPIN1L1ONs ce cccecccvceccsccccccccceecerescrsecccesveccessenr General Review of Literature. Historical.....sseccceeeed The Forms of Colloids in Soils..ccccccsccsvccccveseesee2 OPiginerceccercectececcecctcececccecseccsetcccescecseced Review of Methods of Retimating the Soil Colloids.....14 The Theoretical Basis for the Method. .ccccrcceccsesesedd BXPCPIMeNnbal. cccovccvevecvevscccvcvesevsessscsccceseseedy. Directions for the Methodecccccccccccsccccreccecessese dd SUMMALYo scree cccccccccccvcccccccccsscccsessssesssceses OO Literature CitedeccccccccccccccccccccccccccccccceceeseQs wane jin, Ne’ ppd pee yA tw re 4 ILLUSTRATIONS, Figure Page 1. 20 4. 8. Classification of Dispors0ids.ccccccccsscccveccsccvece ld Showing the Relation between the Length of Shaking ang the Amount of Colloids Obtained... cccccccccvceece 08 Showing the Minimum Blectrolyte Requirement for Solutions from Different Types of Soils..sceseseeee ee 42a Showing the Relation between the Law of Mass Action and the Flocculation of Clay Colloidal Solution......49a Showing the Relation between the Law of Mass Action and the Flocculation of Muok Colloidal Solution......49b Showing the Resistance of Colloidal Gel and Solution on Adding of Different Amounts of N/5 HClecccccccce ee 529 Showing the Resistance of Colloidal Gel and Solution on Adding of Different Amounts of N/5, ALK(S04,)9...2.052b Showing the Resistance of the Dialysed Clay Colloidal Solution and the Water outside of the DialyseresceeeeeD/8 Showing the Resistance of the Dialysed Muck Colloidal Solution and the Water outside of the Dialyser.......57b STUDIES ON THE METHOD OF QUANTITATIVE DETERMINATION OF THE SOIL COLLOIDS. DEFINITION. According to the present conception, the colloid is defined (1) as matter so tinely divided that the sur- face energy vecomes a predominant factor. Since von Weimarn (2) advanced the crystalloidal theory, which he has supported with data of his numerous experiments, col- loids are almost universally considered as being midway between the suspensions on one side and the true solu- tions on the other. The accompanying diagram adopted from Ostwald (3) is suggestive of the relation of colloids to the suspensions and the true solutions. GENERAL REVIEW OF LITERATURE. HISTORICAL. Although Graham(4) is considered to be the father of colloidal chemistry, a number of investigators during the last half of the 18th and first part of the 19th cene turies studied and described certain substances which we now call colloidal. Thus, Bergman (5) in 1779 described the colloidity of alkali silicates in excess of acids. Ruhland (6) in 1812 suggested that certain sols of metals are the metals in "a very fine state of division". Later @ similar view was advanced by Poggendorf (1848) (7). Berzelius (8) in 1848 observed the suspensions of arsenius sulphide. Wackenroder wrote (9) in 1846 on the floccula- - ta - Figure 1. CLASSIFICATION OF DISPRRSOIDS. After to. Ostwald. DISPERSOIDS Tonic Dispersotus. Folecular Dispersoids. ——_ _—/ ~~ Colloidal Magnitude of the phase portion Solutions. is about 1 ww and the smaller. Magnitude = Ody -leuwu. Proper or coarse ? PS Decreasing degree of colloidality Dispersion. Suspensions, emulsions, > etc. Increasing degree of dispersion Magnitude of the phase portion is greater than O.1 yy. ~2- ting action of frost in the case of suspended sulphur. Sobreso and Selmi (10) (1850) studied the coagulation of sulphur suspensions by salts. Schmidt (11) discovered that gum could be purified through the animal membrane. Finally Graham in (4) 1861 published the results of his Classical experiments, dividing matter into colloidal and crystalloidal, and proposing a method of separation of the colloids from the substances in the true solutions. Since that time the colloids have attracted the attention of many investigators and writers. Among the most promin- ent ones could be mentioned Ostwald, von Benmmelen, Hardy, Schulze, von Weimarn, Bancroft, Rohland, etc. Notwith- standing this fact, the chemistry of colloids, especially the chemistry of s0il colloids, is rightly regarded as being more or less ina state of infancy, promising to yield gratifying results to the investigator. THE FORMS OF COLLOIDS IN SOILS. Because of the enormous specific surface of a col- loidal particle the most important property of colloids is its surface energy, and due to this surface energy they are considered of so great importance in the soil management. Perhaps there is not a single chemical, physical, or biological process takes place in the soil which is not influenced by the presence of its colloidal portion. In this connection we must mention, however, that we do not agree with Sjollema (12) who thinks that practically all soil constituents, except quartz grains and the un- decomposed mineral fragments, are colloidal, for the - 3 simple reason that they adsorb the organic dyes. Soil colloids can be classified after Taylor (13) as follows: 1. Humus and decayed organisms. 2. Hydroxides of iron, aluminum, etc. 3. Amorphous silicates, resulting trom wea- thering of the crystalline silicates. 4. Bacteria. The humus as a whole can not be classified with the colloids because some humus particles are very coarse Besides, it is doubtful altogether, as pointed out by Sohngen’ (14) that there was ever prepared a colloidal s0o- lution of carbon. Muravlinski (15), on the other hand, working with the alkali extracts of the Russian chernosem soils found that an alkali extract, when tested with a defractometer gave a more pronounced cone and a greater number of sub-microns in the ultramicroscope than the water extracts. Since the mineral constituents have been found to be predominant in the water extracts and mostly organic in the alkali extracts, he concluded that the soil colloids generally belong to organic matter. It is very probably that not the hydrates of iron and alumina are present in the soil, but mostly oxides, as regarded by Russel (16) and Cameron and Bell (17) for alum inum hydrate is of rare occurance in the ordinary soils. This was found by Liebrich (18) and also by Cameron and Bell (loc.cit.). Kahlenberg and Lincoln (19) have shown that the hydrolysis is practically complete if the silicate is pre- e 4 @ sent in a dilute solution. In such a case the silica must be present as colloidal silicate and not as silicic acid. | Most of the colloids ina soil, being mixed with a large body of solids and only a small portion of water, necessarily come in contact with the solid coarse soil constituents and become adsorbed forming a coating around them (Russel - loc. cit.). Gans (20) thought that there is a constant ratio between AloQz and Si0g in clay soils but this was later disproved by Stremme (21). ORIGIN. The origin of the inorganic soil solloids lies in the mother rock. The weathering is considered by Cornu (22) as a process of a gel-forming from the rocks by means of the atmospheric agencies in combination with the gradually formed humic acids. If humic acids are lacking, the minerals may hydrolize (Luz (23)) forming calcium, magnesium, and sodium silicates, clay silicates and col- loidal ferric hydroxide. Clay silicates may turther de- compose into hydrated colloidal clay and colloidal sili- cic acid. In the case of orthoclase, for instance, the following steps are suggested by Lyon, Fippin and Buckman (24) :<- KA181i30g # HoO = HA18i30g $ KOH KOH # COp = KpCOz 4 HoO HA1Si30g = HA1Si0g 4 2Si0o, which is either quartz or colloidal silica, or changes into a complex hy-= drated silicate. - 5 « The colloids resulting from the weathering of rocks are mostly in the gel forms and it is hardly possi- ble, as Taylor (25) thinks that some of them, being form- ed as sols of different electric charges, precipitate one another. There is hardly any evidence showing the soil colloids to be positively charged, which is in accord with Codn' s (26) theoretical consideration that all soil colloids should be negatively charged, since they are solids and their dialectric constant is necessarily less than that of water. The humic colloids, of course, are resulted from the plant and animal life. Besides, there are some minor factors influencing the available amount of colloids pre-e sent in the soil ata given time. Fertilizers may in- crease or decrease the colloidal formation; urine which is a part of the applied manure, contains colloids (27). Climate also may modify the colloidal content of a given soil, as shown by Lipman and Waynick (28) ina recently published article. During the last few years a considerable interest has been taken in the question of soil colloids. A great deal has been written on the subject, especially in other countries. The contributions of Rohland (29), Ramann (30), and Niklas (31) have very good discussions on the soil colloids. The possible functions of colloids in the soils are liberally discussed, many hypothesis are advanc- ed, and experimental data are gradually being accumulated Therefore, it seems advisable to present at this time a short review of the subject, limiting it, however, mainly - 6 - to the experimental phases. The most important role of colloids in the soil is, perhaps, that of adsorption and naturally this ques- tion has been studied more than any other one. The stu- dies, however, deal mostly with pure colloids and applied to soils only from analogy. Ostwald (32) demonstrated that colloidal ferric hydroxide adsorbs electrolytes. Cornu and Lazarevic (33) showed that hydrogels either simple or mixtures resulting from the mutually precipitated colloids, adsorb crystal- loids. Sohngen (34) found that colloids adsorb nitrogen and oxygen, thus providing a better condition for the growth of bacteria. In the culture media containing starch, inorganic salts and water, colloidal silicic acid and humus favored, while ferric hydrpxide and aluminum hydroxide retarded the cleavage of starch by B. ocraceus, although they had no influence upon the growth of the ore ganism. The cleavage in urea by bacteria is favored by the presence of colloids. Sokolovski (35) found that the adsorption in. the soil is the greatest at the surface and showed that it is dependent upon the specific surface of the soil. Assuming that most of the colloids are formed at the surface and thereby increase the effective surface of the soil, a specific relation was pointed out to exist between the adsorption and the soil colloids. Besides the specific surface, thé adsorption depends upon the mass of the soil. The adsorbing property of a soil changes with the change of its hygroscopic property. It diminishes with the heating of the soil. Parker (36) noticed that the smaller the soil particles the greater the selective a adsorption of potassium from potassium chloride in solue- tion. Rakovsky (37) noted that the adsorption by starch of sodium hydroxide increased in the presence of salts of potassium and sodium, the increase of 0.1% of salt causing often the increase in adsorption as much as 10 - 12% Sime ilar influence was noticed upon barium hydroxide. Findley (38) found that the solubility of carbon dioxide in colloidal solution of ferric hydroxide was greater than in pure water. The different soils possess a different adsorbing property; thus, Rohland (39) admits that, in the opposition to regularity of the rate of adsorption by clay or clay colloids no such regularity is noticed in the case of muck or marshy soils. From results of determining the freezing point of soils at different moisture contents, Bouyoucos (40) forwarded the hypothesis that a part of the soil moisture either adsorbed by the colloidal portion of the soil or held by it in the chemical combination. From theoretical consideration Anon (41) sees an analogy be- tween the clay suspensions and the negatively charged col- loidal solutions in their power of adsorption. Closely relating to the question of adsorption in the soil is its acidity, and this is one of the most im- portant questions with respect to soil management. Baumann and Gully (42) attribute the acid reaction of peat moss not to the free acid, since such an acid does not exist, but to the adsorption of bases by the colloids, which leave the acid radicle or free mineral acid behind. This view is supported by Czapeck (43), Wieler (44), Harris (45), et. al. Some investigators oppose this ex- planation. Tacke and Suchting (46) think that the evi- dences are strong for the existence of free humic acids in the soils they have studied. Tacke, Deutsch and Arnd (47), Rindel, Oden, and Ehrinburg and Barr (48) for dif- ferent reasons support their views. Thayer (49) prepared humic acids from different soils and considered these acids to be colloidal in their nature. In fact, no one seems to deny that the humic acids are colloidal. Thus, it appears that the colloids of the muck soils, no matter in what form they exist, are attributed to be largely re- sponsible for the acidity of those soils. . | Yariloff (50) suggests that the "ripening" of the soil in the spring, i.e. the condition when the soil is the most suitable for the spring cultivation, is brought about by the bacterial action, resulting in the increase of the formation of gases, and to the increase of the col- loidal content. Vernadsky (51) considers the colloids to be the seat of action of soil gases in reduction, oxi- dation, and hydration processes that take place in the soil. Findley and Williame (52) and later Findley and Howell (53) have found that the solubility of COs and Nod in colloidal solution of ferric hydroxide is higher than in pure water, and increases with the concentration of the colloid. Lynde and Dupre (54) as a result of their stu-— dies on osmosis in soils, hold that the colloidal portion of the soil acts as a semipermiable membrane. If this is the case, the colloids in the soil should modify the rate of capillary movement of water in a soil and, finally, to - 9- affect its evaporation. This later process was, in fact, suggested by Keen (55) as being modified by the presence of colloids. He has found that water evaporates slower from the soil than from sand, but ignited soil behaves like sand. That this characteristic of a soil is not due to the humus content was shown by the fact that when hue mus was removed by 2% solution of caustic soda, the soil acted g@s before. His conclusion was that the colloide are responsible for this property of the soils. Impermeability of a soil is sometimes associated with the occurance of the colloidal silica- (56), or with aluminum silicate (57), although the influence of the latter compound is doubted by Cameron and Bell (58). Bouyoucoes (50) has observed that on increase of the ten- perature of some soils the percolation of water in those soile had decreased, after attaining the maximum point at some 30 or 409°C. The diminished rate of percolation of water he attributed to the swelling of the present colloi- dal gels. Rohland (loc. cit.) contends that the soil con- taining humus and the peat soils are rich in colloids. Such soile have a much greater water holding capacity, and the greater power to absorb the water vapor. 6&0 Attenberg (60) thinks that in the analysis of a soil there should be madé a test of the physical properties, euch as hygroscopicity, pore space, capillarity, water holding capacity, relation to the root hair, floccula- tion, Brownian movement, etc. of different grades. In- deed, K8nig, Hasenbaumer and Krénig (61) by the use of -~10-<- four mixtures of CHBr3z and CéHe of different densities and with the aid of the centrifuge separated the soil particles and found the lightest portion to contain the most of the available plant food. Takadora (62) examin- ed some soils for their property to swell on wetting and found the order of the degree of swelling for different soils to be:- Mineral acid soil> humus soily clay s0il > sandy soil. They have tested several other proper- ties, but the results were variable. Sharp (63) ina preliminary paper announced that by the application of sodium salts to the soil and then by leaching them out the soil becomes impervious to water. The leachings from so treated soils contain a larger amount of suspend- ed material than those of the untreated soils. There have been several attempts made to determine the more direct influence of colloids upon the crop pro- duction. Thue, Voelcker (64) found that aluminum silicate caused a large increase in crop of tares and mustard. The increase due to the application of sodium silicate was somewhat smaller, while kaolin did not affect the crop yield at all. The increase in the crop production was at- tributed to the improvement of the physical condition of the soil resulting in the larger retention of moisture. Lyon, Fippin and Buckman (65) state it was found thet the roots of the growing plante, leaving an acid residue, coagulate some colloids, causing the salts adsorbed by the colloids to diffuse out and be available for the use by the root hairs. This, they think, accounts for the fact that the plant is able to obtain more nutrient material from the soil than is possible to dissolve with the sole - lle vents ordinarily present in it. Giles and Carrero (66) have met with entirely different success. They grew rice in water cultures and found that the presence of colloid- al iron decreased its growth. Gregoire (67) reported some results on growing the barley in solution cultures with some colloids. Gedroitz (68) after determining the amounts of colloids in a large number of soils, using the dialyser as a method of separation, came to the conclusion that the colloidal sol is not important in the soil at all since there is not an appreciable amount of it in any given eoil. The gel, according to this investigator, is the important portion of the soil for all the physical changes in the soil which are brought about by liming, heat and frost are due to their presence. ° There remains another question involving the soil cOlloide, which was studied quite extensively. The ques- tion is that of flocculation of the soil colloidal solu- tions or soil suspensions. In 1866, or only a few years after the publica- tion of Graham's classical investigations on colloidal substances, Schulze (69) recorded some of his results on the calcium and magnesium salt requirements for floccula- tion of clay suspensions. Later Schloessing (70) worked along the same line. Durham (71) made an interesting dis- covery that although it requires a very small amount of sulphuric acid to flocculate the suspension of white clay (kaolin?), on further additions of sulphuric acid he reach- ed the point when suspension did not clearify for a long - 12 « time. Now, if to this mixture of clay suspension and sulphuric acid he added either more acid or some water the suspension clearified quickly. Evidently, there is an equilibrium between the ions of true solution and the solid particles of clay. The flocculating action of sodium care bonate, on the other hand, continued to increase with the increase in concentration. While working on the method of mechanical analysis, Hilgard (72) noticed that clay sus- pension coagulated on passing through the narrow glass tube and flocculation is approximately inversely propor- tional to the size of the particles. A moderate increase in temperature decreased the flocculation in his case. He (73) also studied the effect of lime on the texture of clays. | Brewer (74) found the different clay suspensions to be of different stability. In tact, some suspensions settle within a few days, while others remain turbid at the end of seven years, when kept at nearly the same ten- perature and in a quiet place. The acids he found to flocculate more quickly than the salte. Barus (75) in 1888 observed that non-electrolytes retard the clearing of suspensions. Later (76) he tested the hypothesis that the hydration of clay or kaolin particles is responsible for keeping their particles in suspension, and came to the conclusion that such is not the case. He determined the densities of tripoli and bole in both water and ether and found them to be the same in both liquids. Since tripoli has practically the same density as quarts, and bole ap- proaches that of Kaolin, he justified his conclusion on | - 13 = these grounds. Spring (77) noticed that the Clearing pow- er of salt depends upon the valence of the salt and the cation of the electrolyte, confirming in part the quanti- tative formula of Schulze (78) that the coagulating power of trivalent cation: divalent: monovalent as 350:20:1. Bodlauder (79) also measured the power of differ- ent salts for clearing the Kaolin suspensions. Quincke (80) from his studies on pure colloids and kaolin suspen- sions advanced a theory on coagulation which in short im- plies the change in surface tension between the liquid and the oily substances. He claims to have observed oily films around the solid particles. Hall and Morison (81) while studying the efficiency of electrolytes in floccu- lating the kaolin suspensions found that the order of efficiency of acids to be HCl HNO3g >» Ho60,. In the case of cations of the salte it is Al> Ca> K- > Na. Acids are better congulante than their salts. Exceptiony are Alo (804)3 which is equal to Ho80,4, but does not exe ceed it. Maschhaagpt (82) found that NaOH stabilizes soil suspensions sat low concentratione, while if present above -0O15 N, it causes flocculation. Similar results were ob- tained with NagCOz in which case the coagulation begins above 0.16 N. Oden (83) in rather extensive studies with peat colloidal solutions used NaCl for the flocculation. He had to saturate hie colloidal solution with the above salt and allow it to stand for 24 hours in order to bring about flocculation. McGeorge (84) working with suspensions of Hawaiian clays obtained results similar to those of - 14 - Hall and Morison with the exceptions that he found Alo(804)3 to be the best flocculant among both salts and acids and the order of efficiency of strong acids was HNO3> HCl > HoSO4. This brief and not at all an exhaustive review of the literature on the soil colloids reveals a fact that, notwithstanding the large amount of energy that has been spent on the question of the soil colloids, our knowledge regarding this very important and interesting branch of ecience is yet very fragmentary. The results of the sing- ular experiments indicate that this field of the investi- gation is able to yield very good results. But until now the most fundamental problem in the soil colloids is not solved. There is no adequate method for determining the quantity of colloids in soil, while this knowledge is of prime importance in order to understand certain of the phenomends™ in the soil which are influenced by colloids in their process, The necessity of the method for esti- mation of colloids in the soil was recognized a long time ago and a great many attempts have been made to solve this problem. The following is a brief review of the proposed methods of estimating the soil colloids with a few remarke on their applicability. REVIEW OF THE NETHODS OF ESTIMATING THE SOIL COLLOIDS. As early as 1856 Schmidt (85) has discovered that the gum could be rarefied through the animal membrane. This was the beginning of colloidal chemistry and at the same time an unconsoious method of separation of colloids from crystalloids. But Thomas Graham is considered as the - 15 - father of colloidal chemistry. Among the series of class- ical experiments performed by Graham we find the first method of determining the colloids in the sol forms. Following are some of his experiments on dialyzing:- 1. 10 gms. Nacl and 2 gm. Japanese gelatin was dissolved in hot water and diluted to 100 c.c., cooled to become a firm jelly. Then 100 c.c. of solution of 2% gelatin was poured on the top of it. Left for 8 days and analyzed for NaCl at different depths. The following amounts of NaCl found at the end of 8 days in different layers, beginning trom the top, No. of stratum 1 2 38 4 § 6 &8 9 Nacl gms. -015 .015 .026 .035 .082 .130 a2 ~55 486 No. of stratum 10 11 #12 #13 14 15 & 16 Nacl gms. 630 _ 996 1.172 1.19 1.2 5245 Total 9.992 gm. of 10 gm. In another instance NaCl was let to dialyze through the parchment paper and the results obtained are given below - 100 c.c. of #01 % 2 gm. of Nacl - Nacl diffused out 86% 50 c.o. ®* * $2 8° *® " " " “ 92 25 c.ce * *e e#2 0° * . ° ” * 96 The dialyzing continued for 24 hours at 109 - 120°C, The natural conclusion was that the salts in solu- tion have the power to pass through the colloidal layer whether it is in the form of jelly or as parchment paper. Graham worked with an astonishingly large number of substances and classified them according to their pro- perty to dialyze through either animal or vegetable parch- ment paper. In order to give an idea of such power in different substances the following table may be cited: 100 c.c. 10% solution was used in each case at 10 - 15°c for 24 hours employing the loop dialyzer. Substance used. Gms. passed. Relative diffusion Gum Arabic 0.029 004 Starch sugar 2.00 «266 Cane " 1.607 e214 Milk 0 1.387 ~185 VYannite 2.621 - 349 Glycerine 35.300 - 440 Alcohol 3.57 476 Nacl 7.50 1.000 The substances whose dialyzing power is similar to that of gum arabic are considered colloidal, and according to Graham they could be separated from the non-colloids by dialyzing. Gedroitz (86) proposed to use this method for de- termining the amount of colloids present in the soil. He separated the colloidal solution from the soil mechanical- ly vy shaking in water for three minutes and dialyzing the solution for some 3 - 4 months. Such method has, it seems, very little value for either practical work or for the theoretical investigations. In the first place it is impossible to thoroughly separate colloids from cyretalloids by dialysis, as shown by Graham's work. Crystalloids do not diffuse after a cer- tain dilution is obtained. Also, the small amount of col- loids does diffuse and is lost. Oden (87), for instance, has found the dialyzing to be not an efficient method of purification of colloidal solution obtained from humus, because some of the colloid material goes through the mem- brane. Kahlenberg (88) points out that it is even possi- ble in some cases to have colloidal particles to pass out leaving some crystalloidal substances behind. K@8nig (89) -~ 17 ~- has shown that oxidation may take place during the dialyz- ing. Leaving the solution for a considerable period of time, which is required for the dialyzing, some colloidal material will go into solution and pass through the dialyz- ing membrane. If we consider now the difficulty of controlling the bacterial action during the process we may rightly conclude that the dialyzing in its present form is not ap- plicable for determining the colloids in soils even if we are able to separate the colloidal solution from the soil proper. The time required is another prohibitive factor for application of the method for practical purposes. Up to the present time the dialysis, with some mod- ifications from the original Graham apparatus, is used very extensively, especially by biologists. Only very re- cently the principle discovered by Graham is being develop- ed and the time required for dialyzing is shortened. Martin (90) has discovered that colloidal particles cannot ve forced through the colloidal or gelatinous membrane. With thie fact as a basis he put a layer of gelatin or of gelatinous silicic acid upon the Chamberland's filter ana forced the solution containing colloids through it. With this device he was able (by applying high pressure - 30 - 100 atmos.) to obtain a clear solution of salts, but free from proteins. Bechhold (1906) (91) discovered that the concen- tration of gelatin plays an important role in passing through the membrane. Using collodium, eising, gelatinous formaldehyde, or other similar substance, upon some sup- - 18 « port he forced the solution through the filter. The most important of his results are summarized in the following table:-= Dispersoid Concentration of used. membrane which Remarks. holds back the colloid sol. Sol of Platinum 2% Size of particle 244 mu Cell Fe on) 3) a Casein wilt 2.9 Colloidal el 3e Size 40 4M 1% Haemcglolain sol 4 1% Gelatin sol 4 | Serum Albumen 4<- 4.5 Mol wt. 15,000 - 3,000 Silicic acid (coll) 4.5 Neuter Alvbumosen A 8 Mol. wt. 2,400 " B&cC 10 Traces passed pextrin 10 Passed little - Mol. wt, 965 All crystalloids : oo Passed all Later, Schoep (92) constructed an apparatus, with which he could use very high pressure without breaking the membrane. The objection to this form of separation ie the absorption of a disperse phase by es material through which @® solution passes. | The adsorption in the soil is varied roughly with the fineness of the soil grains and this property was used by several soil investigators for estimating the soil cole loids. Thus, wiven Mitscherlich (93) opened the subject by proposing the méhod of estimating the hygroscopicity of soile. But since hygroscopicity isa the property especial- ly pronounced in colloids this was a step toward the esti- mation of colloids in soils. The method consists of further drying the air-dry soil in a thin layer over phosphorus pentoxide. Then it was pleced in a desiccator over 10% solution of sulfuric acid for 24 hours, where the condensation of water in the - 19 - soil takes place. The process may be hastened by provid- ing the partial vacuum in the desiccator. Then the per- centage of the water adsorbed is determined. Since there is a danger of coagulation of some colloids by drying and thereby reducing its power of ad- sorption, Ehrenberg (94) suggested to reverse the pro- cess, i.e. first adsorption and then drying. Lipman and Sharp (95) found that for the best results the layer of #0i1 should be about 1 mm. thick. They also found that the absorption of water is increased with the increase in temperature and vice versa. Sjollema (1905) (96) first to my knowledge pro- posed to use the dyeing material for adsorption by soil, and he struck the right cord, if his success in populariz- ing the method can be judged by the number of his follow- ers. - His method consisted in treating the small quanti- ty of soil with a dye solution. Since not ell the soil particles adsorb a certain one dye he used several of then. Then the water was decanted and the soil was analyzed under microscope, comparing different soils. Endel (1902) (97) went further and advanced the following method: A portion of soil is mounted in Canada balsam, colored in the cold solution of fuchsin, and the picture is taken, magnifying the examined sample 280 times. The colored portions of a picture are cut out and their weight relative to the total weight of the picture gives the percentage of the colloids ina soil. Aschley (98) (1909) made a thorough study of clays and proposed to use malachite green for comparative esti- - 20- mation of colloidal material in clays. This method is also based upon the adsorptive pow- er of colloidal particles and consists of treating 20 gms. of clay with 400 c.c. of water containing from 1 to 3 gm. of dye. This was vigorously agitated tor an hour, and permitted to settle over night. The sample of the solu-e tion is taken out with the pipette and compared with the standard solution of the same dye. Two very striking features are brought forth by Aschley's proposal:- (1) Ease of manipulation, and (2) the inadequacy of the method itself. Since that time a considerable number of men either proposed a new method or modified a given one, suggesting some new detail for improvement of existing method. Thus, KOnig (99) and his associates (1911) advo- cated Methyl violet. Gorski (1912) (100) proposed crystal violet and Pelet-Jolivet (101) suggested the methylin blue. The described methods, based on adsorption of dyes by colloids, even at their best are open to serious criticism. Van der Leeden and Schneider (102) compared three methods; von Bemmelen's, which will be given later, Mitscherlich's and Pelet-Jolivet, and came to the conclus- ion that in order to rely upon these methods one should keep in mind the following points - (1) Many silicates (wholly non-colloidal) adsorb dyes; (2) The mixture of gels, containing an abundance of Alo0z and Feo03 in com- parison with 8109 are not colored with methyline blue; (3) The presence of the capillary water influences the ad- sorption of dye; (4) The presence of the electrolytes modi- - 21-6 fies the adsorption of dye also. This point was very strongly brought out by Gedroitz (103) in his experiments. The same soil adsorbs different amounts of dye according to the treatment with one salt or another. The dyes used by Gedroitz were methyl violet and crystal violet; and (5) Uncertainty, whether clay does not adsorb the dye. Another class of men tried to determine the cole- loidal portion of a soil by purely chemical means - chem- ical analyses. The first representative of this school was von Bemmelen (104), who was one of the first to intro- duce the colloidal chemistry to the soils. Von Bemmelen suggested that the soil, besides the insoluble particles, contains (A) amorphous colloid silicates, and (B) kaoline silicates. Treating the soil first with HCl and then with — Ho804 one can approximately separate these two groups from each other and from the soil. Method:- Treat the soil with boiling HCl (sp. grav. 1.19) and silicate (A) will be diesolved. Al, Fe, etc. - bases go in solution. Remaining silica goes down as ppt. and can be extracted with an alkali. A free colloidal ferric oxide and the bases bound with the humic substances go into solution aleo. Colloidal SiOo could also be ob- tained if treated at this stage with the strong alkali. The remaining soil is then treated with the concentrated Ho80,4 and the bases go in solution, while silica goes down asa precipitate. Von Bemmelen did not add anything new by advancing this method. It was known long before his time and used by geologists for determining the so-called "Zeolites" in k, ~ 22 - soils. The method perxjse can hardly stand criticism. Concentrated H2804 or HCl are capable of dissolving a great many substances in addition to the hydrated sili- cates. FRAPS «< (105) (1914) proposed a method of determin- ing the amounts of the ammonia soluble inorganic soil colloids. His method is as follows :=- Digest 100 gms. of eoil with 2000 c.c. of N/5 HCl at room temperature for 24 hours. Filter and wash thoroughly. Wash back into the bottle with 2000 c.c. of 4% NHz and let digest at room temperature for 24 hours, shaking every half hour for 4 hours. Filter ona large folded filter getting as much of the soil as possible on the filter and continue to pour back the filtrate until it comes through clear. Discard the residue. Take 1500 c.c. of the filtrate, coagulate with ammonium carbonate (and heat, if necessary), let settle, collect on the ash free filter, ignite and weigh. Fuse the precipitate with sodium and potassium carbonate, dissolve in HCl and evaporate to render silica insoluble. Filter off and weigh silica, if pure; if cone taminated with iron, purify. ‘Precipitate iron and alumina in the filtrate with ammonia, ignite and weigh the pre- cipitate. Fuse with potassium acid sulfate and dissolve, reduce iron with zinc, and filtrate with permanganate. The author confesses, however, that the outlined method is intended only for soils low in lime; in the case with soils of high lime content several extractions should be made and much stronger HCl should be used. - 25= The method is not supposed to estimate the total colloidal constituents of the soil. Pence (106) suggests a possible method for deter- mining the hydrated silicic acid in clay consisting of the following:- To 5 gms. of clay ina casserol add 120 C.c., 5% N@aoCOz- Boil 10 minutes over the free Bunsen flame with the rotaty motion to prevent bushing. Let settle. Decant through a hardened filter paper. Repeat twice. Tranefer theclay to filter paper and wash with hot diluted NaoCO3. Determine 8102 in the filtrate. The method is not very delicate, not being able to detect as much as .2% of the colloidal silica, and depends on a "compensation of errors". Hilgard (107) has simplified the problem consider= ably, proposing the method of separating the colloids from soile by first loosening them by either prolonged, gentie kneading of the wet clay, by prolonged digestion in hot water, or by the boiling for a short time. Then the col- loids are thrown down by some electrolyte, such as Nacl, which can be washed out of the coagulated colloida. Dupont (108) went somewhat further, when he ad- vanced a new method of mechanical analysis, the part of which was a separation of colloids. If this portion be taken out of the entire method, the determinations of colloids in soils would consist in breaking the soil ag- gregates by means of oxalic acid and boiling for WX min- utes on water bath. Filter and separate from clay by centrifuging for 12 minutes at 800 or even better at 1000 - 1200 revolutions per minute. The resulting col- loidal solution is treated with ammonium carbonate, which = 24 « will coagulate the mineral colloids. The coagulated m- terial is collected, evaporated and burned; weighed be- fore and after the burning. After a careful study of the literature bearing on different methods the only rational and the most promising method appeared to be the one proposed by Hilgard and mode ified by Dupont, i.e. one based upon the mechanical forces for the separation of the colloidal portion of the soil from the soil proper. The metMod, however, is not free from objections. The employing of oxalic acid and the boiling are the features open to serious criticism, since some of the colloidal material, thus treated, will go into solution and be lost, introducing an error in the col- loidal determinations. Again, is it possible to get all the particles from the soil which are capable to stay in suspension? What is the bases for the centrifuging for that length of time and at a given speed? What is the stability of the resultant solution? Is ammonium carbon- ate the best and the most convenient coagulant possible of employment? How much of a given electrolyte to add for the vest results? Can not the dialysis answer the same purpose as an electrolyte? These and other questions may be properly raised. It seemed advisable, therefore, to study this method experimentally and, if possible, to suggest a desirable improvement. THE THEORETICAL BASIS FOR THE METHOD. This method of separation of the colloidal por- tion of the soil proper is based upon the undisputable fact that the attraction between the solid soil particle for = 25 « water is greater than that of a solid particle for another eolid particle. Since this is true, it is possible to gradually loosen the individual soil particles and finally separate them one from another. The process of separation is hastened if the mechanical force of agitation of the eoil mass in water is applied to it. It is true there will be an unavoidable error due to the increase of solubility of the soil material during the agitation, bus this error is very small in comparison with other factors involved. The similar error, of course, is repeated in every other proposed method. BXPERIMENTAL. The experimental work was naturally divided into two parts; namely; (1) the separation of the colloidal por- tion of the soil from the soil mass, and (II) the separation of the colloids from the crystalloids. I. The Separation of the Colloids from the Solid Meas. In the first place it wae decided to ascertain how much of the separated material from a given soil by shaking and centrifuging could be separated again from the quartz sand after it had been added to it and dried. Experiment 1. Separation of colloidal clay and col- loidal muck solutions from quartz sand to which they were previously added and dried. To a given portion of acid washed quartz was added a colloidal clay or muck material, either in the form of sol or asp a precipitate. The colloidal solutions were obtained by shaking a brick yard clay and a muck eoil with about 10 times their weight of distilled water for 4 hours and cen- - 26 - trifuging at the rate of 2000 revolutions per minute for 15 minutes. The resultant solutions could be kept for sev- eral weeks without an appreciable sedimentation on the bote tom of the vessel. The precipitates were obtained either by adding a small quantity of HCl N/5 or by passing the solu- tion through a Chamberland's filter and collecting the res- idue. After the precipitated material was added to the quartz the whole mass was well mixed and allowed to dry. If the solution was added, it was permitted to evaporate, stirring the contents occasionally. Before the mixture be- came dry it was thoroughly mixed. In each case an adequate portion was taken, water was added as in mechanical analysis shaken for 4 hours and centrifuged at the rate of 2000 revo- lutions per minute for 15 minutes. The upper liquid was de- canted, to the sediment more distilled water was added, thoroughly stirred and centrifuged as before. This process was repeated until the solution became clear. Then the 200 c.C. portion of the obtained solution was evaporated and the total dry material was calculated. A sample of the untreated quartz was run as a check and the dry weight of it was substracted from the total weight of the treated samples. The results are given below. No. Wt. of Nature In what Wt. of Wt. of % of quartz of form colloid colloid colloid taken. colloid, added. added. obtained. recovered 1 25 gms. clay ppt. HCl. .1484 1416 95.41 2 256 * # « ” 03198 ° 2827 88.48 3 25 * " # “ 2 3079 2537 82.41 4 25 * “ “ Cham. .9957 «7209 72.40 5 25 *® " " ° ~9409 7482 79.53 6 25 * # ° ” 4169 03212 77.94 7 25 * # solution «1250 01126 90.00 8 25 * muck " - 1250 01033 82.64 9 25 * ° " ~ 1250 ©1000 80.00 - 27 « The results presented in this table show that it was impossible to recover all the fine material from the quartz after drying, no matter in what form the colloidal material was added. The results also indicate that the gel obtained by means of hydrochloric acid is just as reversi- ble (or even more) as that obtained by a Chamberland's fil- ter or by drying. The attempt was made to repeat the experiment with the artificially prepardd colloid added to the quartz sand. For this purpose a solution of the colloidal ferric hydro- xide was prepared by Graham's method (91), the dialyzing being continued with the daily change of water for 14 days. The colloidal solution contained .701 gms. of dry matter per 100 c.c. of solution. 100 and 200 c.c. portions of this solution were added to 100 gms. of quartz and evaporat- ed to dryness. But when 25 gms. samples were taken, shak- en with water for 4 hours and centrifuged in the usual mane ner, the resultant solution was found to be perfectly clear, practically none of the iron hydroxide being able to stay in suspension. Evidently, the drying of the colloidal mat- erial with quartz was detrimental to its stability. Briggs, Martin and Pearce (109) found that the even-drying even reduced the percentage of clay obtained in the mechane- ical analysis of soils. In order to ascertain to what ex- tent the similar drying affects the amount of dry material which is possible to obtain from a field soil the following experiment was performed - Experiment 2. Effect of Drying upon the Quantity of the Colloidal Material Obtained. A quantity of fresh silt loam was sifted through a - 28 « two millimeter mesh and divided into two portions. One half was allowed to dry at the room temperature, while the other one was kept in the atmosphere saturated with the water vapor. After 10 days the samples from each portion were taken, shaken with water, and the separation of tne colloidal solution was brought about by means of the cen- trifuge. The percentage of moisture of each portion was determined and the per cent of the fine material obtained was calculated, basing on the weight of the oven-dry soil, TABLE II. Sample. % HoQ. Time of Amn't Amn't 4% ob- Aver. % shaking. Taken. obtain- tained. obtained. l.pried 1.81 4 hours. 4.9111 .2268 4.62 2. " 41.81 4 © 4.9111 .2350 4.78 eee 3.Fresh 22.934 " 6.694 .6235 10.94 10.59 4. © 22.934 * 5.694 .5832 10.24 5.Dried 1.8115 * 4.9111 .3943 8.03 6. " 1.8115 *® 4.9111 .3734 17.60 782 7.Presh 22.93 15 “ 6.694 .6542 11.49 8. © 22.9315 " 5.694 .6500 11.42 nha Sample Relation between then. l. Dried 44.4. Re ed Se Freeh 100.00 4. 0 5. Dried 68.2 6. ° 7. Fresh 100.00 8. " It is an undisputable fact, as revealed by the re- = 29 « sults presented in this table, that the drying of the soil makes a large portion of the colloidal material highly ire reversible. In the case with the 4-hour shaking more than twice as much of the material was obtained from theffeshn eoil than from the dried sample of the same soil. When the shaking continued for 15 hours, the direction of the magni- tude of the obtained material was the same, but the ratio between them was not as wide as in the samples of 4-hour shaking. Evidently, the amount of agitation to which a giv- en 90il1 is subjected has a considerable influence on the amount of the colloidal matter. This point was tried and the results are presented in the table of the next experi- ment. Experiment III. Effect of length of shaking of the soil in water upon the quantity of colloidal material ob- tained in the mechanical separation. In this experiment the samples of the air-dry bricke- yard clay and muck were taken and shaken in water from 3 minutes to 36 hours. The rest of the procedure remained the same, i.e. the solutions were centrifuged at the rate of 2000 rev. per minute for 15 minutes. The results follow. TABLE III. CLAY SOIL. Ave. of Time Amount of Dry. Wt. of % of colloids No. of of Clay taken. colloids obtained. trials. shaking. obtained. 1 3 min. 20 gms. 4.2341 gms. 21.17 3 1 hour 20 (°* 4.5110 * 22-56 3 4 " 8s 20 * 4.7154 * 25258 2 16 * 5 * 1.2582 * 25.16 2 26 = 5 * 1.2513 * 25.03 2 36 «= 5 * 1.2669 * 25.34 TABLE III a. MUCK SOIL. 2 4 hours 5 gms. 0553 gms. 1.11 2 16 ° 5 * 0739 1.48 2 26 ° 5 * -0853 * 191 2 56 “ 5 * -0900 * 1.80 ~ - 30 «= It is evident that the length of shaking has a con- siderable influence upon the total colloid material which is possible to bring in a disperse state. These results are in accord with those obtained by Briggs, Martin and Pearce (110) for the finest clays, and show that the greater the period of time of agitation to which the sample was sub-e jected the more solid material was brought in suspension until it reached the practically constant value in the case with clay. In muck the constant was not reached, but the increase due to 10 hours agitation between 26 and 36 was not as great as between 16 and 26 hours, showing this value is being approaching the constant. The accompanied chart il- lustrates this point. The next factor influencing the amount of the collot+ dal material as well as the quality of the same is the speed of the centrifuging. In order to throw some light upon tne question the following experiment was performed. Experiment 4. Effect of speed of the centrifuging on the amount of the colloidal material obtained and on the stability of the resultant hydrosol. 10 gm. portions of clay were brought in suspension by agitating in water for 4 hours. Then, two samples were subjected to 15 minutes centrifuging of 1000 revolutions per minute; other two samples were centrifuged at the rate of 2000 revol. per min., and the remaining two were let to run at 2800 revol. per min. The centrifuging was repeated until the resultant solution in each case was nearly clear. The 400 c.c. portions were set aside for 10 days in bottles in which the colum of the solution reached 13 inches. At Hof Colloids = of ek le ids Showing the Relator _ MICHIGAN | eon RES A, oe — —-—— “Figure: Be Anount of ‘Colléida, Obtained. em ee eee soe 2 ee eee ee i | 2: — : ceaaal i 0 4 . 26 = 36 Se, :Ro © a ¥ 8 ae be eet ; a er q DErARTMENT OF MATHEMATICS - 31 - the end of the 10 day period the upper 12 inches of the so- lution was carefully sxphoned off and the dry weight of the material in voth the upper 12 inches and the lowest inch was determined. The results are summarized in the next table. TABLE IV. Speed of No. of aAmn't § Average %ob- % settled in centri- trials. taken. Dry Wt. tained. 10 days. aver. fuging. obtained. of 2 trials. 1000 4 10 gms. 3.1023 31.02 53.25 2000 4 10 * 2.3620 23.62 27.90 2800 4 10 (* 2.1201 21.20 20.34 As one should expect, both the amount of the solid material and the stability of the obtained solution vary with the variation of the speed employed. The greater the force developed in a given length of time the more ot the g0lid material will settle, the easiest sedimentation being that of the coarsest particles, as one notices in the figure of the right hand column. More than half of the solid mate ter settled in 10 days in the samples subjected to the low- est speed and nearly 80% remained in suspension in the case when the highest speed was employed. Possibly, vesides the size of the particles the stability of different solutions was affected by the mass action, since the solutions re- sulted from the lowest speed of the centrifuging were more concentrated than those resulted from the higner speeds, There is still another factor remains which influ- ences both the quantity and the quality of the suspension. Time has undoubtedly the same influence as the speed of the centrifuging and it seemed advisable to determine just to what extent it modifies the suspension, Experiment 5. Effect of time of the centrifuging on the amount of the colloidal material obtained and on the - 32 = stability of the resultant solution. In this experiment the usual procedure was followed with the exception of the time, which was varied, and the speed of 2000 revolutions per minute remained the same throughout the experiment. As in the preceding experiment, 400 c.c. of the solutions was set aside for 10 days without disturbing and the percentage of the settled material was determined. | TABLE V. Time of Trial. Total sol- Dry ma- Dry ma- # in low- centri- id materi- terial terial est inch fuging. al obtain- in upper in lowe- after 10 ed. 12" aft- est 1" days. er 10 after 10 days. days. 10 minutes 1 1.2928 gms. .O800gms. .1120gms. 58.33 2 1.2424 * 20790 * -1000 * 55.87 Ave. 1.2676 * . 57.10 15 0 1 1.0563 * 01125 ~0525 31.82 2 «9677 21077 20470 30.38 Avee 1.01658 31.10 20 * i «1100 0388 26.07 2 ©1120 00385 25.57 AVG. 25.82 350 " a 9102 1178 0210 15.13 2 «8434 -~1126 ~0385 15.07 Ave. 8768 - 15.10 45 0 a 5775 ©1225 00233 15.91 2 «5570 1168 ~0125 11.72 Ave. «5673 13.82 60 ” 1 24995 «1030 20105 9.25 2 - 5087 e 1045 «0120 10.30 Ave. «5041 9.78 The percent of the solid material in the lowest inch of the solution varies from 58 6 a little over 9. If there were no sedimentation at all and the particles were uniformly distributed throughout tne liquid, the per- cent of the solid material in the lowest inch of the solu- tion would be some 7.7. Since such a solution does not ree main on standing of uniform concentration in all its depths = 33 « one will notice that the solution resulted from the cen- trifuging for 60 minutes has settled very little indeed. The stability gradually decreased with the decrease of time of the centrifuging. In connection with the above ex- periments it should be mentioned that when the artificially prepared colloidal ferric hydroxide was centrifuged at the speed of 2800 revolutions per minute for 45 minutes, there was no sediment left on the bottom of the tube. As a summary of the results of the first part of the problem the following points should be mentioned:- 1. The soil colloids on drying even at the room temperature become highly irreversible. The examination of the soil for the colloidal content, therefore should be made before the soil is dried out after it is taken from the field. 2e The length of shaking of the soil with water has an influence upon the quantity of the suspended material obtained. Care should be taken to thoroughly agitate the soil in order to break all the aggregations and to wash the colloidal particles off tne coarse soil fragments. Se The speed of the centrifuging effects both the quantity and the quality of the colloidal solution. If but &® low speed is applied only a very coarse suspension of low stability is obtaind, though the amount of the solid mater- ial is comparatively great. Thus, a considerable speed should be employed in order to obtain the solution that would nearest approach the hydrosols of the artificially prepared colloids. 4. Time has similar effect on the resultant solu- = 34 « tion: the longer the solution is centrifuged the less the material stays in suspension and the better the stability becomes, and vice versa. Therefore, a reasonable time should be allowed for the centrifuging. Time, as it seems. from our experiments, may take place of the speed, i.e. the longer the solution is centrifuged the less speed it requires for bringing its stability to a desirable degree. II. SEPARATION OF THE COLLOIDS FROM THE TRUE SOLUTION After an extensive review of the literature the most practical methods of the separation of colloids from the true solution appeared to be (a) coagulation or floccu- lation by means of an electrolyte and (b) dialyzing, as described by Graham (loc. cit.). In the first question, namely, the coagulation of the colloidal particles by means of the electrolytes, it was decided to undertake several additional experiments in selecting the most convenient electrolytes for the purpose. For the next few experiments on flocculation the soil colloidal solutions were prepared by adding to a bulk of fresh soil about 10 times its weight of distilled water, shaken well, and allowed to stand over night. Then, tne upper portion was siphoned off and centrifuged at the rate of 2000 revolutions per minute for 15 minutes. The result- ant solution would stand for several weeks without appreci- able sedimentation. In most of the experiments recorded the same solutions were employed. The exceptions will be men- tioned later on. - 35 «= NATURE OF SUSPENSIONS USED. N. s0il Reaction of Dry matter Freezing point used. soil to lite per 100 depression of mus paper. c.c. Of solution. suspension. Ll. Brickyard Clay (subsoil) neutral 3633 2003 2. Miami Silt Loam “ 20700 © ~002 3- Clyde silt * “ ~0913 -003 4. Muck " 20274 «002 5. Brickyard Clay (6011) " 8098 6. Peaty Muck “ 093538 7. Kaolin ~0247 The bacterial action in the colloidal solutions during the experiment #+Sie not controlled. The acid, salt and alkali solutions were N/5 in strength and were the same throughout the experiments. Experiment 6. Qualitative Test of Electrolytes for Flocculation of Colloidal Solutions. In this experiment to 5 c.ca. of suspension was add- ed 5 c.c. of N/5 electrolyte, shaken vigorously for a short time and let stand over night. The best flocculated solu- tion was recorded with five positive signs (xxxxx). Next in apparent efficiency was marked four and so on until a negative sign was used, if no precipitate appeared at the pottom of the test tube in 24 hours. Duplicate determina- tions were made in all of the experiments. TABLE VI. EFFICIENCY OF ELECTROLYTES IN FLOCCULATING SOIL COLLOIDAL SOLUTIONS. 5 c.c. of 1 2 3 4 Electrolyte Clay Miami Clyde n/5. (subsoil) Silt Loam. Silt Loam Muck. 1. HCl XXXXX XXXXX XXAXX Le yvaci XXAAK - = Se KCl XXXXX XXXX - 4. NH,gCl XXXXX XXX - 5. BaCho XXXXX XXXXX XXXXX § c.c. of Flectrolyte n/5 11.NaNno3 12.KNO3 13.NH m0°) 14.Ca(NO 16.AgNO3 17.Po(NO3)o0 18 .H2804 19 .KHSO 20. . (3H4) 2804 22. K28207 vee uS04 25.FeS04 26.¥Fe (804) % 27 Ko 28.Naso; 29 .NaHso 50 .Na28003 $1-A1K (804 ) 2 56.KOH 58. MgO 59.Cad 40.H 41 -NaHPS, 42. Cal (Po jo 4B. Kon 46.K9C0 47 tem 48 .NaHCO 49.Caco 50. 3ugcOsMe (OH) o 51.FeCco 52. paCO8 53.CH2COOH - 36 e 1 Clay (subsoil) Table Vi. (continued) 2 Miami Silt Loam. 3 Clyde Silt Loam. XXKXKN XAKAK XXX XAKIOK hope Muck. pe PERRR HE * Heed i - 37 « fable Vl. (continued) 5 c.c. of 1 2 Electrolyte Clay Miami n/5 (subsoil) Silt Loam 55. 56.Ca(CoH302) Xxxxxx XXXXXK S7ePb(CoH300)5 XXXxx XXXXK 58.C2H204 XXX XXX 59.C6Hg07 XXKHKX XAAXX 60. Gleinic C17H33CO0H - - 61.A820 - = 62. (tH } gogo XXXXXK XXXXX 63.NaKt gf XXXXXK - 64 .XB XI XXXXX 65.XI XXXK XXXXX 66.KSCN XXXXX XXXXX 67.Pb05 - - 68. KoCT207 XXXXKX XXX 5 6 Clay Peaty (s0il) Muck 1. HCl XARXX XXXXX ze Nacl XXXXX - Se KCl XXAXXX - 4. NH4Cl XXXXX - 5. Baclo XAXXX XXXKH 6. CaClo XXXXX ‘XXXX Ve HgClo XXX - 8. MgClo XXXXX x 9. sncl4 XXXXX XXXX 10.HNO XKXIK XXXXX 11.Nanod3 XXXXX - 12.KNOz XXXX - 13.NH XXXXX - 15. Migios ) XXXXX XXXXX 15 Hg (Seayo XXAXX XXXXX “AghO3 XARXAXK XXXXX 7. “Pb(N03) XXX XXXAXX 18.H9804 XXKK XXXXX 19 .KHSO XXXKK XXX 20. (WH4} 0804 _XXXXX - 21.Ko804 XXXXX - 22-K28007 XAXXX XXX 24.CuS04 XXAKXX XXX 25.Feso XXKXXK XXXX 26.Fe2(904) 5 XAXXX XXAXXX 27 .Ko8 AXAXXK - 28.Naso XXX - 29 .NaHSOs XXHXHKHK - 30 .Wa 28202 XXXX o 31 ALK (50 jo = XXXXx XXXXAX 52.Fe mia} 804) AXXXXX XKXAX 33.Fe8 - - 5 Clyde Silt Loam 4 Muck, - 38 « fable VI (continued) 7 5 5 c.c. of Clay Peaty Kaolin . nf/b (soil) Muck Rlectrolyte 34. Zzns - 55.Na0H XXXXX 36.KOH XXXXX 37.Ba(OH)o XXXXX ° XXX 39.Cad XXXXX 40 .H3P04 XXXAXKX 41.NaHPO4 XXXKX 42. Cait, (PO Je x 43.085(704) 2 - 44 .KH XXXAX 45.Ko 4 XXXXX 46.K5CO XXXXX 47.Naocbs XXXX 48.NaHCos XXX 49.Caco - 50. 3MgCO3Me (0H) o XXXXKX 51.Feco - 52. . (WH4}oCO XXXXX 53.CH boen XOXO 54 sad a XXXXX XXXXX XXXXX XXXXX XXXXKX XXXXX XXXXX XXXXX XXXXX 2H302 56. Ca(C2H302)2 57. Pb(CoH305) 0 58.CoHD04 59.C 60. ofetnic C197H33CO0H 61 eAB80C 62. (NH4)0C204 63.NakC 4H40¢ 64.KBr 65.KI 66.K8CN XX XXX 67.Pb05 - 68 .KoCro0y XXXXX f g ‘LL EP HEEB Bee The results presented in the above table show that besides the familiar difference in efficiency of different electrolytes with the same colloidal solution, the same electrolyte does not act alike with different suspensions, the easiest solutions to flocculate being those of clay and kaolin, followed by loams and, finally, mucks. This question ia almost entirely overlooked by many soil investi- - 39 = gators. As far as we were able to determine, no studies of any importance have been recorded in soil literature on the flocculation of other suspensions but those of different clays and kaolin. As a result, the conclusions regarding this (as well as perhaps others) process have been based upon the results obtained from studies with a limited num- ber of soils. But such conclueione, judging from the re- sults presented in the above table, may be erroneous, due undoubtedly to the fact that soile differ one from another in many reepects, namely - chemically, physically, and bi- Ologically. The soils in our experiment have different origin and different history with respect to their manage- ment. Being from the same locality, they have only one factor in common, namely, the climate. Very probably, a given type of soil, if exposed to different climatic cone Gitions, would behave differently with the same electrolyte. Strong acids are very good coagulants but they are not always better than some of their salts. This point is especially well brought up by the next experiment. The salts of heavy metals seem to have a much stronger floccu- lating power than those of lighter ones. The trivalent cation is more efficient than divalent one and this latter is better than a monovalent cation. Yet, the tetravalent stannic chloride does not seem to do as efficient work as the divalent barium chloride or calcium chloride. Contrary to the prevalent opinion, bases flocculate when used in this concentration. Only muck resists monovalent bases and b yields fairly easily to divalent both carium hydroxide and calcium hydroxide. Experiment 7. Baturated) Gms. 1 -0955 2 -0967 4 -0978 8 1030 16 24 32 gms. -0950 -0958 -0980 -0992 gms. «0820 «1028 01112 01138 Ems. ~0950 -0985 -0983 ~0997 200 c.c. Of original solution evaporated = .1015 gms. MUCK SUSPENSION. 1 2 ‘3 TABLE XII b. electrolyte 2 N/5 added. = gms. gms. 2 e0821 ~Q802 4 «0845 -0813 6 8 -0855 ©0815 12 0876 -0819 gms. ~0800 -0885 e 1004 01023 gms. -0807 20912 e 1080 e 1056 Pb (NO3)o 1 2 gms. e 1045 1089 - 1096 ©1123 - 55 « The results presented in the table (XII) leave no doubt regarding the influence of the amount of electrolyte added. The adsorption of the salt sometimes becomes so gred that the dry weight is larger than that of the original so- lution. Similar to the results of the preceding table, they show that the best results are obtained when only enough of the electrolyte is added to cause a complete coag- ulation of the suspended material. If any excess of the salt, or acid, or base is added the adsorption of that salt, or acid, or base may cause the error, which would be leger than if the dry matter of the original solution was deter- mined by the evaporation. Thewashing of the excess of salt would remove some of the reversible colloids that would pass through the filter paper. The general conclusion from the results obtained in the last two experiments is as follows:- The flocculation of the soil colloidal solution with N/5 hydrochloric acid, aluminum potassium sulphate, ferric sulphate and, perhaps, lead nitrate and calcium hydroxide can be employed as a means of separation of the colloidal material from the sub- stances in the true solution, providing that only enough of the electrolyte is added to cause a complete flocculation. At this point the amount of the electrolyte adsorbed is very emall and can be disregarded in the determination of the gel Although only two soils were studied for this purpose, yet these soils may be considered as representatives of two ex- treme cases. The clay soil is a splendid type of the miner- @l soil, while the muck soil similarly represents the organ- ic type. Considering that most of the common soils lie bee tween these two extremes, it is velieved, that the general - 56 «= conclusion presented above is justified on these grounds. Dialysis, It seemed advisable to compare the dialyzing with the flocculation of the soil colloidal solution as a means of their purification from the crystalloids, Experiment XIII. Efficiency of Dialyzging in Puri- fication of the Soil Colloidal Solutions. A series of dialyzers were prepared with a veget- able parchment paper for the membrane. The clay, muck and ferric hydroxide colloidal solutions were dialyzed, chang- ing the distilled water daily. The purity of the colloidal solutions was determined by means of measuring its electrica resistance. Since there was no apparent chemical change in the colloidal solution during the dialyzing and, therefore, no marked change in the structure of the particles occurred, it is supposed that the influence of the solid particles on the electric resistance was practically the same throughout the experiment. All readings were made at 25°C. After reading the colloidal solution was returned to the dialyzera thus leaving the volume of the dialyzed solution the same. The readings of both the colloidal solution and the outside distilled water was made in order to determine the differ- ence in their resistance after 24 hours, and in few of the latest determinations after several days of standing. TABLE XIII. ELECTRICAL RESISTANCE AT 25°C. WITH THE SAMR ELECTRODES. - 57 « Table X1I11I (continued) Ferric hydroxide. Days of Clay colloidal sol. collet asl Dialyz- Ho0 outside Colloidal H20 out- ing. solution. ohms. solution side ohms. ohme. ohms. 0 2,228 - 2,587 i 3,145 3,917 3,623 6,500 2 5,085 6,139 4,168 8,405 3 8,151 13,054 4,674 11,203 4 12,629 25,655 §,004 13,383 6 17,220 27,620 5,514 16,655 8 23,765 54,625 6,219 27 5425 13 57, 850 74,655 7,155 33, 390 15 44,725 101,915 7,754 47,650 1? 47,900 107 ,910 8,534 50,455 20 §2,825 114,565 9,184 56,410 <8 58,225 106,100 11,685 93,775 35 67,845 150,315 17,540 134, 300 45 71,080 24,710 55 88, 420 153,410 gelatinized TABLE XIV. FLECTRICAL RESISTANCE AT 25°C. WITH THE SAMB ELECTRODES. Muck colloidal solution. Days of Colloidal Ho0 outside dialyzing. solution. ohms. ohire. _ CG 6,820 1 14,080 28,275 2 21,180 56, 640 3 24,765 79,3525 § 29,825 89,170 7 33,140 133,215 13 40 ,090 113,880 17 41,660 165,750 23 42,535 115,320 35 59 , 820 115,790 The distilled water used in these three dialyzers had the electrical resistance between 173,760 and 186,100 ohms. It will be seen from the figures presented in the above tables and the illustration by the charts that the dialysing is an extremely slow process of purification of the colloidal solution. Yet, the dialyzing of cley and muck ___M.CHIGAN® dete ee cst me rat Figure Se": : a ‘ = 1 + c = Se eee ' . ‘ ' a eee ! 1 : vs y i : : pe a eee mrs op pa ae Sos. ho ; ‘ r ee ae oe : ; : t | ; + i i 100\. : eet ester webiste = mac : 1 a : po. gt he | : Ra eter eek es eer a |. ae ood ae = n . 4 me ° , | i led : a Ts ‘i sm * nig : . oO 2 oa oe ee = ; : i oI : =f 4 g : + so {| ee ‘ z | a cS 2 & I ee ' aH ky. - a ep ,—J =. Resistance oe oe — ee oe we ee ee wget ; DeraR TM EG Obs MOFemas cs ysed Muck Colloidal outside of the Dialyser. Ba eo. nye +h Ht MICHIGAN Aj PROS | e Resistance of =the i 4a he oS3 ts af a or tn ; ato eieepeen rs 10 Jal Ss pOlUUsonN Showing th 4 ’ ; a! bbe tr sts . _ — 2 = — . Ss *suyg jo puesnoyy, ut | eouezstsey DEPARTMENT OF MATHEMAT.CS @- 58 e suspensions is much faster than that of the prepared ferric hydroxide. Since it ie probable that some of the colloidal particles go through the parchment paper, taking the factor of time into an account, and considering the fact that some of the solid particles atick to the membrance and dry out on it, making it almost impossible of removing, the dialyz- ing in its present form can hardly be used advantageously for the quantitative determination of the soil colloids. DISCUSSION. From the foregoing experiments the following points outetand as having an important bearing on the method of mechanical determination of the soil colloida: - le. The resulta of experiments 1 and 2 emphasize the importence of having the samples to be analyzed in the fresh state, since the drying makes the largest portion of colloids either irreversible or, if reversible, very slowly 80. 2. The shaking of the sample should be continued for a considerable length of time. 15 hours is an appro- priate period to be set for a minimun. 4. The centrifuging should be both fast and pro- longed. The length of the centrifuging takes place of its epeed and vice versa. In order to obtain the soil colliddal solution of the high stability the centrifuging should be done at the rate of at least 2000 revolutions per minute for 30 minutes. 4. In the quantitative determination of the colloid al portion of the soil the centrifuging should be repeated until the solution after centrifuging is fairly clear. From 8 to 15 repetitions is usually sufficient, depending - 59 « upon the nature of the soil and the weight of the sample taken for the analysis. It should be mentioned that it is almost impossible to reach the point when the solution after centrifuging ia perfectly clear, especially in the case of very heavy soils. But as a matter of fact the amount of solid material present in such a semi-turbid so- lution is very amall in comparison with the solutions ob» tained by the first few centrifugings. An experiment was performed, though not recorded in this paper, in which 14 repetitions of the centrifuging was made and the dry weight of each portion was determined. It does not seem necessary to record the entire experiment here, but it will suffice to say that by the first centrifuging 22% of the total weight of dry matter was obtained, and only 1% by the 14th. By the first seven centrifuginge over 75% of the total was obtained. Thus, one notices that the existing error, is not large and should not diminish the value of the method. 5. In separation of colloids from the substances in the true solution one of several electrolytes can be em- ployed. Among the best ones are aluminum potassium sul- phate, ferric sulphate, hydrochloric acid, and, to some ex- tent, lead nitrate and calcium hydroxide. 6. The minimum amount of the electrolyte employed to cause a complete flocculation should be added. If the electrolyte is added in excess, the increase in the adsorbed coagulating ions may give erroneous results. Directions for the Method. ‘ From the study of the method proposed by Dupont, the following modification is offered which, it is believed, has several advantages over the former one by being based - 60 « upon the experimental evidences:- A sample of soil taken directly from the field is sifted through a two millimeter mesh, mixed well and the weighed portions are taken for both the colloidal and the moisture content determinations. 10 grams, or any other appropriate amount, of the fresh soil is placed in the shaking bottle which is used in mechanical analysis, filled with distilled water up to 54 oz. merk, few drops of ammonia added and shaken for 15 hours or more in the shaking machine, When the particles are disintegrated, transfer the contents of the bottle into the centrifuge tubes dividing the solid portion equally be- tween two or four tubes. Add enough water to fill the tubes and centrifuge for 30 minutes at the rate of 2000 revolutions per minute. Decant the Piquid off, transfer the settled gel into the evaporating dish, evaporate, cool and weigh, Knowing the total volume of the solution obtained, the volume taken for the flocculation, and the weight of the oven-dry soil, calculate the percentage of the colloidel material. The duplicates should easily check within 0.5%. SUMMARY. 1. The drying of the soil even at the room temper- ature decreases the amount of the colloids present in solution. It makes the colloidal particles either irrever- sible or, if reversible, very slowly so. ze Shaking of the sample of soil in water influences the amount of colloidal material to be brought in suspension Se The time and the speed of the centrifuging modi- fies both the quantity of the colloids in suspension and the - 61 « stability of the resultant hydrosol. The stability of the solution increases with the speed and the time of centri- fuging, while the amount of the solution decreases with it. 4. Besides the fact that the flocculating efficien= cy of different electrolytes with the same colloidal solu- tion is different, the results show that the efficiency of the same electrolyte with the solutions from different soils varies considerably depending largely upon the chemical com- position of the soil. 5. Schulze'ts valency law does not hold true with the soil colloidal solutions studied. 6. Humus material hinders the coagulating power of the electrolytes. 7. It takes a greeter amount of an electrolyte for the flocculation of the more concentrated soil colloidal so- lution than that for the less concentrated one. 8..In the flocculation of the soil colloidal solu- tions by the electrolytes the reaction obeys, within the experimental error, the law of mass action. 9. The adsorption of the flocoulating substance is increased with the increase of the amount of the substance present in solution at the time of flocculation, 10. The solid particles of the gel resulting from the soil colloidal solutions studied hinder the electricel conductivity of these solutions. Jl. The dialyzing is not an adequate method for the quantitative determination of colloids in the crystalloidal solution. 1. 6. - 62 « LITERATURE CITED. Bancroft, W.D. 1914. The theory of colloidal chemistry. In Jour. Phys. Chem., Ve 18, De 949. Cameron, F. K. 1915. Soil colloids and the soil solution. In Jour. Phys. Chem.,v. 19,p.1-13. Bigelow,s. L. 1912. Theoretical and Physical Chemistry, p.245. Kassuto, L. 1915. General Colloidal Chemistry (Russian edition),p.10. Ostwald, Wo. 1915. A Hand Book of Colloid Chemistry, pp.4-12. Taylor, WW. 1915. Chemistry of Colloids, p.2-3. von feimarn,P.P. 1913. New Systematics of Aggregated Condition of Matter and the Basic Law of Dispersoidology (Petrograd,1913). Ostwald, Wo. 1915. A Handbook of Colloid Chemistry, p.43. Graham, T. 1861. Liquid diffusion applied to analysis. In Phyl. Trans., ve1i51, pp.183-224. Bergman,T. 1779. Cited by Walden,P.,in "The history of colloidal silicic acid,Zeits. Chem. Ind. Koll.,v.6,p.234-5; Chem. Abs., ve4,(1910} p.2078. Ruhadand. 1912. Cited by Taylor in Chemistry of Colloids, p.12. Possendorf. 1848. Cited by Taylor (loc. cit.). Berzelius. 1844. Cited by Taylor (loc. cit.). 9. 10. 11. 12. 13. 14. 16. 17. 18. 19. 20. 21. 226 -~ 63 ~ Wackenroder. 1846. Cited by Taylor (loc. oit.). Selmi. 1850. In Ann. Chim. Phys., v.28, p.210-215; Taylor, Chem.Coll.,p.14. Schmidt, #. 1856. In Poggendorf's Ann., (1856),p.337; Ostwald's A Handbook of Colloid Chemistry, p.264. Sjollema, B. 1905. Anwendung von Farbstoffen bei Boden-Untersuchungen. In Jour. f. Landw., v.53, p.67. Taylor, W. W. 1915. Chemistry of Colloids, p. 290. Sohngen, N. L. 1914. In Chem. fleekblad, v.11, p.593-6; Chem. Abs.,9,(1915)p.23. Muravlinski, S. M. 1914. Ultramicroscopy of soil extrasts. In Pochvovedenie, vel5, No.2-3,p.23-27; Exp. Sta. Reo.,30,(1914) p.516. Russell, &. J. Soil Condition and Plant Growth (London). Cameron, F. K.; and Bell, J. M. 1905. The mineral constituents of soil solution. U. S. Dept. Aor. Bur. Soils Bul.30, p.22. Lisbrich. 1897. In Zeit. prakt. Geo.,(1897) p.212; cited by Cameron and Bell ( los. cit.). Kahlenberg, and Lincoln, 1898. In Jour. Phys. Chem.,v.@, p. 88. Gans,R. 1913. in Centralblatt (1913),p. 22-23. Stremme, H. 1914. UWoer die Physikalische Natur der Kolloiden wasser- haltigen Tonerdesilikate. In Centralb. f. Min. usw., (1914) p.80; Biedermann's Zentr. f. Agrik.-chem., ve 44,(1915) p. 66. Cornu, F. 1909. The modern theory of weathering in the light of colloidal ~ 64 - Chemistry. In zeits. Chem. Ind. Koll.,v.4,p.291-5; Chem. Abs., ve5, (1911). p. 851. 24. Luz, A. 1914. Laterit, seine Betrachtung in Lichte der Kolloid-chemie. In Kolloid-Zeits., 14,p.80; Biedermann's Zentr.,v.43, (1914) p. 720. 24. Lyon,T.L.y Fippin,#.0., and Buckman, H.U. 1915. Soils. Their Properties and Management, p.22. 25. Taylor,W.W.. 1915. Chemistry of Colloids, p.292. 26. Coehn. 1898. In Wied. Ann.,64,p.217; oited by Bogue in Jour. Phys. Chem, v. 19, (1915) p.666. 27. Baduel, C. 1910. Formation of an emulsion of urine and ether in the Jacquemet-Testevin test and the presence of a colloidax in the urine. In Rif. med, 61-67; Biochem. Centr.,v.9, 115-6: Chem. Abs.,v.4, (1910) p.1979. 28. Lipman,C.R., and Wawnink,t.N. 1916. A detailed study of effects of climate on the important properties of soils. In Soil Soience,v.1,p.5-49. 29.Rohland, P. 1909. Colloidal substaaces in clays. In Zeits. Blektroah., 15, 540-2; Chom. Abs.,v.3,(1909) p.2865. 1910. Colloidal matter of olay. In Sprechsaal, 42, 655-7; Cham, Aha,, v.4, (1910) p.505. 1910. Collonide in slaw and adsorption phenomena. In Landw. Yahrb.,39, 369-72; Chem. Abs.,v.4, (1910)p.3270. 1914. Importance of colloidal substances in the soil from acricultural stendpoint.In Monatsh. Landw., 6,p.264-5. 1914. Colloidal matter in clay. In Int. Mitt. Bodenk., v.3, 487-94: lixp.Sta.Rec.,22,p712; v.29,p.817; v.30,p.718. 1914. Soil colloids and their adsorption phenomena. In Landw. Yahrb., 44,259-47; Chem Abs., v.9, (1915) p.2279. 40. Ramann, &. 1912. Colloidal studies in soil work. In Kolloide Chem. Beih., 2,289-503; Chem.Abs.,v.6, (1912) re1197, 41. 426 530 38. 39-6 41. 42. ~ 65 - Niklas, d. 1914. Colloidal Chemistry and its importance in soils, Geology, and mineralogy. In Int. Mitt. Bodenk.,34,3843-403; Exp. Sta. Rec.,30, (1914) p.513. Ostwald, iio. 1911. Adsorption of electrolytes by salts. In von Bemmelen Gedenboek, 264-74; Chem.Abs.,5. (1911) p.3367. Cornu, F., and Lazarevick, M. 1911. Adsorption compounds in minerals. In Zeits. Chem. Ind. Koll. ,4,295-8; Chem Abs., v.59 (1911) p.851. S6hngen, N.L. 1914. Influence of colloids on microbiologic processes. In Centr. Bakt. Parasit., II.abt.48, 621-47; Chem.Abs.,8,497. Sokolovski, A.M. : 1914. From the sphere of phenomena of adsorption iff soil. {n Jour. Opytn. Aégr.,v.15, p.67=-117. Parker, heGe 1914. Selective adsorption by soils. In Jour. Agr. Res.,1,3,179. Rakovski, A. 1914. To the teaching concerning adsorption. In Jour. Opytn. Asron., 14, pe/-12. Findley, A. 1908. The influence of colloids and colloid suspensions on the solubility of carbon dioxide in water. In Chen. News, v.98, p.163. rohlana,P. 1914. Die Rodenkolloide und ihre Adsorptionf&higkeit. In Landw. Yahrb., 47,2, pe 259-247. Bouyoucos,a.eJ., and McCool,M.M. 1915. The freezing point method as a new means of measuring the concentrataon of the soil solution directly in the soil. Mich. Agr.Cobl. Bul, 24, p.27-41. Anon. 1911. The significance of adsorption in the pricipitation of suspension colloids. In Sprechsaal,44,560-2; Chem Abs., ve5, (1911) p.170. Baumann,A., and Gully,i. 44. 44, 4D. 46. -~ 66 - 1910. The free humic acids of upland moors. In Mitt. Bayr. Moorkulturanstalt, 31-156; Chem Abs., v.5, p.952. Czapeck. Cited by Skene in Ann. Bot.,v.29, (1915) p.71. Wieler. Cited by Skene in Ann. Bot., v.29,p.71.(1915). Ty = 7 Tai Harris, dete 1914. Soil Acidity. Mich. Asr.coll.Tech.Bul.19, Tacke,B.,and Stichtins. Humas acids. In Landw. Yahrb.,41,717-54; Chem.Abs.6, 2275. 47.lVacke,B.,Deutch,and Arnd. 48. a9. D356 1914. Humic acid. In Landw. Yahrb.,45, 95-265; Chem.Abs.,8,1179. Skene, 1915. He cites the authors in Ann. Bot., v.29,p.71. Uhayer, i. 131%. Snlloide) chenicel studies of humus from lemed and and unlimed soil. In Jour. Landw.,60, p.1-18; Chem. Abs., ve6 (1912), p.1797. Yariloff,A. 1915. Ripening of the soil. In Jour. Upytn. Agron., v.16,pel. Vernadski,V.l. 1914. Chemical composition of soils. In Pochvovedenie, v.15, (1914) No.2-3, p.1-21; Exp.Stea.Rec.,30,(1914) p.513. Findley,A.,and Williams,T. 1913. Influence of colloids and fine suspensions on the solubility of gases in water. In Jour. Chem. Soc.,1034, ‘ p.-645-45; Chem. Abs.,v.7, (1913) p.2502. Findley, A., and Howell, 0.k. 1914. Influence of colloids and fine suspensions on the solubility of gases in water. In Jour.Chem.Soc. ,104, p.636-45; Chem.Abs.,v.7,(1913)p.2502. Lynde, and Dupre, 1915. Osmosis in soils.In Jour.Amer.Soc.Agr.,v.7,p.15-29. Keen,B.A. 1914. The evaporation of water from soil.In Jour.Agr.Sci., Ve 6, p.456-476. 60. 61. 62. 63. 64. 67. - 67 - Kellermann, K.¥. 1911. The relation of colloidal silica to certain Impermiable soils. In Science,v.33, p.189-90; Chem.Abs.,v.5,p.1655. Cushman. 1905. in U.S.Dept.Agr. Bur.chem. Bul. 40. Cameron, F.K.,and Rell,J.\. 1905. The mineral constituents of the soil solution.U.5S. Dept.Agr.Bul 30,p.25. Bouyoucos, Ged. 1915. Effect of temperature on some of the most important physical processes of soils. Mich.Agr.Coll.Bul.22,p.39. Attenberg,A. 1914. The properties of soil grains and the plasticity of soils. In Koll.-chen.Beih., 6,p.55-85; 8.5.8.2, 32. (1919), 617. Kbnig, J., Hasenbdumer, J., andkr&nig,cC. 1914. Separation of soil particles according tp the specific gravity, and the relationship between plants and soil. In Landw. Yahrb.,v.46,p.169-251. Takadoro,T. 1914. Colloid properties of acid soils of Japan. II. In Jour. Coll. Agr.,6,p.117-29; Chem.Abs.,9 (1915),p.1815. Sharp,L.T. 1916. Salts, soil-colloids and soils.In Sci.Amer.Suppl., 81, 2101, 234. Voelcker,J.A. 1909. Pot cultures experiments.- Effect of colloid substances on productivitiness in soils. In Jour.Roy.Agr.Soc. London, v.70, p. 388-495. Lyon, T.L.,fippin,#.0.,and Buckman, HU. 1915. Soils.Their properties and manasement, p.411-412. Giles,P.,and Carrero,Jd. 1914. Assimilation of colloidal iron by rice. In Jour. Agr. Res.,v.4, p.205-10. ~ arepoire, A. 1912. The action of some hydrolizable salts and some colloids on plants. In Ann.Sta.Agr.etat. & Gembloux, 58; Biedermann's Zentr.,42,p.334-40; Chem.Abs.,v.8,(1914)p.1137. 68. 69. 70. 71. 72. 74. 76. 7/e 78. 796 80. = 68 - Gedroitz,K.kK. 1912. Colloidal chemistry in soils.In Jour.Opytn.Agr. ,13, p. 463-420. Schulze,F. 1866. In Pogsendorff's Ann.,p.366; cited by Quincke, in Chem. News,v.84,(1901) p.174. Schloessing,M.C. 1870. Sur la precipitation des limons par des solutions salines tres-etendues. In Comp.Rend., 1xx,p.1345; cited by Quincke, Chem.News, 84 (1901),p.174. Durhan, W. 1876. Suspension, solution,and chenical combination. In Chem. News,v.37,po4/-~ Hilgard,b.l. 1879. Un the flocculation of particles, and its physical and technical bearings.In Amer.Jour.Sci.,v.17 (4),p.205. Hilgard,#.a. 1877. Cited by Johnson, Un the mechanical effects of tillage. In 11th Ann.Rep.Conn.Board Agr.,(1877) p.136. Brewer, W.H. 18385. Un the suspension and the sedimentation of clays. In Amer. Jour.Sci.,v.29,(4) p.1. Barus, OC. 188. In Phys.Beibl.,xii,p.563; oited by Quincke, Chem News, v.84 (1901) p.174. Barus,C. 1889. The subsidence of fine particles in liquids. Jn Amer. Jour. Sci.,v.3/7,(45)p.122. Spring. 1900. In Rec. Trav.Chim.Pays.Bac.,ve19,p.204; cited by Kassuto, General Colloidal Chemistry (Russ.),(1915) p.116. Schulze, 1882. Cited by Kassuto, (see cit. 77.). Bodldnder. 1894. In Gott.Nachr.,p.267; cited by Quincke in Chem. News, 84,174. Quincke, a. 1901. The clearing of turbid solutions and the movement of - 69 - small suspended particles by the influence of light. Chem. News, v.84, p.174. 81. Hall,A.L.;and Morison,C.a.T. 1907. The flocculation of turbid liquids by salts. In Rothamsted Memoir, 8,p.244. 82. Maschhaupt,dJ.G. 19145. Remarks on Rohland's "the action of hydroxyl ions on clays and clay soils in marling". In Landw.Vers.Sta., v.84, p.467-70; Chem.Abs.,v.8,(1914) p.2214. &4.Uden,58. 1914. Colloid chemistry of humus. In Kolloid-Z.,v.14, p.123-40; Chem. Abs.,v.8, (1914)p.2768. 84, McSseorge, if. 1915. #ffect of fertilizers on the phhsical properties of Hawalian soils. Haw.Agr.ixp.Sta.Bul. 48, p.19-22. 2 AN TA ~a.4 7 fs “4F O56 sehmia lg lie 1656. See cit.1l. 86. Gedro@itz, Kk. 1912. Colloidal chemistry of soils. In Jour.Opytn.Agr.,13,p.364. 87. Jden,s. TOLloil sasaiscry of Tatas. tn Kelleoid-@.,14, p.124-40; ChemeAbs. (1914) p.2768. 88. Kahlenberg,L. igfO, Outlines of Chemistry, p.295. 39. Kbnig,d. 1911. Uber die Ausvendung der Dialyse und die Bestimung der Oxydationkraft fllr die Bentreiluns des Bodens. In Landw. Vers.Sta.,Bd. 79-80, p.491. 90. Martin, ow. 1896. In Jour.Physiol.,20,p.364; cited by Ostwald in & Hand- book of Colloid-Chemistry, p.264. 91. Bechhold, H. 1906. in Z.f.Blektroch.,12,777; cited by Ustwald in A Hand- book of Colloid-Chemistry, p.264. 92. Schoep,A. 1910. Bull. Soc.Chim.Belg., 24,354; cited by Ostwald in A WO 01 e \O ON e 996 100. 104. -~ 70 - Handbook of Colloid-Chemistry, p.266. Rodewald,H. and Mitscherlich,A.h. 1904. Die Bestimung der Hydroskopizitadt. In Landw. Vers.Sta., Rd.59,p. 455-441. Hhrenbers,P. 1911. Beitrage zur Physikalische Bodenuntersuchung, Zeit. fF. “orst.= und Jagdwesen, Bd.44,8.435-47. Cited by Lyc., Fippin,and Buckman in Soils, p.168. Lipman,C.B., and Sharp,L.T. 1912. The hy£roscopic Moisture of soils.In Jour.Phys.Chem., 15, pe 709-22; Chem.Abs.,v.6,(1912) p.402. Sjollema,B. 1905. Anwendung von Farbtstoffen bei Boden-Untersuchungen. In Jour. Landw.,v.33,p.607; cited by Gedroitsz in Jour. Opytn. Agr.(1905) No.6,p. 7064. andel. 1909. In Koll.Zeits.,5,p.244; cited in Jour. Upytn.Agr.,12,p.142. Ashley, Hit. 1909. The colloid matter of clay and its measurements. U.S.Geol.survey Bul.4388,p.15. KOnig,J.,Hasenbaumer,J.,and Hassler,C. 1911. In Landw. Vers.Sta.,8d.5-6, p.377-441; Jour. Exp.Sta. (Russ. ) ve6,(1905) p.764. sorski,J.M. 1912. Beitrage zur Bestimung der Kolloide in Ackerboden. In Ztschr. f.Landw.Vers.in Oesterr,v.15 (1912)p.1201; Jour.Upytn.Acr.,v.14,(1914) p.101. Pelet-Jolivet,L. Citea by Gedroitz in Jour. Upytn.Agr.,v.15. (1914)¢.186-217. van der Lecden,R.,and Schneider,F. Int. Mitt.f.Bodenk.,B8d.2,p.81-109. Sedrobitz, kek. 1914. In Jour. Opytn.Agr.,v.15, p.186-217, von Bemmelen. 18e8. Die Adsorptbionswerbingungen und Adsorptionsveriuigten 107. 10%. -71- der Ackererde. Landw.Vers.Sta.,Bd.35,p.69-136; Adsorption, by van Bemmelen,(1910) p.115. Fraps,G.s. 1914. Ammonia-soluble inorganic soil colloids.Tex.®xp.Sta.B.165. Pence. The determination of hydrated silicic acid in clay. In Trans.Amer.Ceram.Soc.,v.12, p.45=53. Hilgard,h.W. 1906. Soils,p.61. Dupont,d. 1911. A new method of physical analysis of the soil. In Comp. Rend.,v.1945, p.889-891. 109, Brigss,L.J., Martin,#.0., and Pearce,J.R. 110. 111. 112. 113. 114. 1904. The centrifugal method of mechanical soil analysis. U.S.Dept. Agr.Bur.Soils Bul.24,p.19-20. Same publication as cit. No.109,page 20-21. Schulze. 1882. Sited by Kassuto, General Colloidal Chemistry,p.101. Keppeler,G., and Spangenbersg, A. 1097. Notice on the flocculating action of colloids on clay suspension. In Jour. Landw.,v.55,pe299=-300; Cheme Abs., v2, (1908) p.1025. Lyon,T.L., Fippin,E&.0., and Buckman,H.0. 1915. Soils, p.159. Whitney, M. 1891. Soil investigations. In 4th Ann. Rep. Md. Exp.Sta.,p.258. Quincke, G. 1901. The clearing of turbid solutions and the movement of small suspended particles by the influence of light. In Chem. News,v.84,p.174. 116. Bary,B. 1911. The manner of solution of colloidal substances. In Comp.Rend.,v.152,p.1386-7; Chem.Abs.,v.5,(1911)p.2770. 117.Bancroft, W.D. 1915. The coagulation of albumen by electrolytes. In Jour. 118.. 119. 120. 121. 122. - 72 - Phys.Chem.,v.19, p. 349-460. Diblaux, J. 1911. Colloids considered as electrolytes. In Zeits.Chen. Ind. Koll.,7,73-81; Chem. Abs.,v.5, (1911)p.1226. Jordis, i. 1908. Siliocates and the chemistry of colloids. In Zeits. Blektr.,13,525; Chem. Abs.,v.2,(1908)p.1225. Ashley, H.E. 1909. The colloid matter of clay and its measurements. U.S.Geol.Survey Bul.588,p.15. Arrhenius,s. 1908. On agglutination and coagulation. In Jour Amer. Chem.Soa., v.40, p.1382. Beams,H.T., and Hastlack,H.&. 1915. The electrical Synthesis of colloids. In Jour. Amer.Chem.So0c., v.37, pe 2667-835. _ rr L-.” ont c a re. vpee — _ Fat at — iT, me *. * eg s TE - $ ee a ROOM USE ONLY MICHIGAN STATE UNIVERSITY LIBRARIES iii 4386 Se nes i i “Til Lo 93 031 a® sr) a