V. p. p.— 90. S.— W. 
 
 U. S. DEPARTMENT OF AGRICULTURE. 
 
 Report No. 71. 
 
 SOME MUTUAL RELATIONS BETWEEN ALKALI 
 SOILS AND VEGETATION. 
 
 THOMAS H. KEARNEY, Assistant Physiologist, 
 
 Ihri.sioii of V(()fUtlih' Phijisjjbimiuiii<l }''ithoi(i'jy. 
 
 FRANK K. CAMERON, Son, Chkmist, 
 
 T>iristoii, of Soils. 
 
 WASHINGTON: 
 
 GOVERNMENT PRINTING OFFICE. 
 1902. 
 
Book^_^ 
 
V. p. p.— 90. S.- 
 
 U. S. DEPARTMENT OF AGRICULTURE. 
 
 Report No. 71. 
 
 SOME MUTUAL RELATIONS BETWEEN ALKALI 
 SOILS AND VEGETATION. 
 
 ^ 
 
 ^r, 
 
 THOMAS H. KEARNEY. Assistant Physiouxjist, 
 
 Du'kion of Vcgetahle I'lnixiohxiii mid I'alJio/oijy, 
 0^ AND 
 
 FRANK K. CAMERON. Soil Chemist, 
 
 Division of tSoils. 
 
 WASHINGTON: 
 
 GOVERNMENT PRINTING OFFICE. 
 1902. 
 
 C>h^ ^ 
 
 a 
 
T} 
 
 
 19 i909 
 
 Da 91 Qt , 
 
LETTltR OF TRANSMITTAL. 
 
 U. S. Department of Agriculture, 
 Division of Ve(;etable Physiology and Pathology, 
 
 Wash In (/fan, D. C, June 24, 1001. 
 Stk: I have tlie lionor lo ti-aiisinit licrewith several papei's on Some 
 ^lulual Relations Ix^tween Alkali Soils and Vej>etation, prepared by 
 Mr. Thomas II. Kearney, of lids Division, who was detailed to the 
 work at the reipiest of the Chief of the Division of Soils, and Dr. 
 Frank K. Cameron, of that Division, and respectfully recommend 
 their publication as Report No. 71 of the Department. The studies 
 so far made in connection with the work discussed in these papers 
 have bi"OU,i>ht to light some important facts and have opened lines of 
 inquiry which promise to develop methods of dealing with alkali soils 
 that will reduce their injurious effects on crops grown where such 
 soils exist. 
 
 Respectfully, 
 
 Albert F. Woods, 
 
 Chief of Division. 
 Hon. James AVilson, 
 
 Secrefanj of Agriculfiire. 
 
 3 
 
.ETTER OE SUBMITTAL 
 
 U. S. Department of Agriculture, 
 
 Division of Soils, 
 Washington, D. C, June 3^, 1001. 
 Sir: I respectfully siibiuit herewith the manuscript of a report pre- 
 pared by Mr. Thomas H. Kearney and Dr. Frank K. Cameron to throw 
 light upon problems encountered by the field parties of the Division 
 of Soils in the soil survey of certain areas in the West affected with 
 alkali. As this is treated largxdy from the physiological side, it seems 
 proper that it should be transmitted by you for publication. 
 Respectfully, 
 
 Milton Whitney, 
 
 Chief of Division. 
 Mr. Albert F. Woods, 
 
 Chief, Division of Vegetable Physiology and Pathology. 
 4 
 
CONTENTS. 
 
 Page. 
 The Effect upon Seedling Plunts of Certain Components of Alkali Soils. By 
 Thomas H. Kearney and Frank K. Cameron: 
 
 Introduction . - - - - 7 
 
 Methods of experiment - - - - - 9 
 
 Salts emploj-ed .. --■ 9 
 
 Plants selected for experiment 10 
 
 Details of manipulation — . - 13 
 
 Determination of the limit of endurance . - 1") 
 
 Results with pure solutions - 19 
 
 Concentration maximum permitting survival of the roots 19 
 
 Concentration minimum prohibiting elongation of roots 24 
 
 Results with less soluble salts - 25 
 
 Results with mixed solutions 27 
 
 Magnesium sulphate in mixtures 29 
 
 Magnesium chloride in mixtures . 32 
 
 Sodium carbonate in mixtures. - - 33 
 
 Sodium sulphate in mixtures - - 35 
 
 Sodium chloride in mixtures . . - _ ...--. 36 
 
 Calcium chloride in mixtures - 37 
 
 Sodium bicarbonate in mixtures 37 
 
 Calcium sulphate and calcium carbonate in mixtures . _ 37 
 
 General significance of results with mixed solutions 40 
 
 Stimulating effect of dilute solutions - 47 
 
 Economic importance of the results — -- 52 
 
 Summary _ - 54 
 
 Conclusion - - - - •'j-^ 
 
 Bibliography - - - ^^ 
 
 Formation of Sodium Carbonate, or Black Alkali, by Plants. By Frank K. 
 Cameron: 
 
 Introduction - - - 00 
 
 Creosote bush - - 61 
 
 Greasewood . . - 03 
 
 Absorption of mineral constituents by the plant. — - 05 
 
 Comparison of analyses - _ . _ 08 
 
 Summary ... 09 
 
 Resistance to Black Alkali by Certain Plants. By Frank K. Cameron: 
 
 Introduction '^^ 
 
 Method of examination _'_ ''^ 
 
 Distichlis spicata "^1 
 
 Isolation and identification of acid exudation 72 
 
 Hydroscopic salt on the plant surface ''^3 
 
 Selective absorption of soil constituents 73 
 
 Function of the acid exudation 74 
 
 Phosphorus in the plant - 74 
 
 Ash analyses . _ . — - 75 
 
 Suaeda intermedia and Atriplex bracteosa 70 
 
 Summary 77 
 
 5 
 
SOME MUTUAL RELATIONS BETWEEN ALKALI SOILS 
 AND VEIIETATION. 
 
 THE EFFECT UPON SEEDLING PLANTS OF CERTAIN COMPONENTS 
 
 OF ALKALI SOILS. 
 
 By Thomas H. Kearney aud Frank K. Cameron. 
 INTRODUCTION. 
 
 Everyone avIio is fjuniliur witli alkali soils knows that their charac 
 ter varies i^reatly in different loealities, one salt or combination of 
 salts predominating over othcn-s which may l)e i)resent.' Sometimes 
 sodium carbonate, the dreaded '' l)lack alkali,'' is relatively abuiuhint 
 as compared with the other soluble soil components. In other cases 
 this salt may be entirely absent or pres(uit mei'ely as a trace, while 
 one or more of the "white alkali" salts, e. g., sodium chloride oi- 
 sodium sulphate, plays the most important pai't. 
 
 It is also known that these salts are not all <H|ually injurious to veg- 
 etation. Sodium carbonate, for instanc.e, is generally believed to be 
 much more harmful than any other salt of common occurrence, owing 
 probal)ly to its pronounced corrosive action on the plant tissues. 
 Gypsum, or the dihydrate of calcium sulphate, on the other hand, is 
 harmless and even beneficial in ordinary cases. Exi)eriments with 
 solutions of chemically equivalent strength show ^'ery marked differ- 
 ences in the action of different salts upon plant gi'owth. Hence the 
 question whether the salt forming the greater part of the soluble com- 
 ponents of a given soil is, to take a concrete case, the vei-y injurious 
 sodium carbonate or the relatively harmless sodium chloi-ide, may 
 often determine whether that soil is utterly useless or (piile valuable 
 to th<' fai'mer. 
 
 It becomes, therefore, a question of great importance to everyone 
 who is concerned with soils which contain an appreciable amount of 
 alkali to know definitely the relative harmfulness of the salts both 
 severally and in mixtures, since the latter is the condition under 
 which they almost invariabl}^ occur in nature. 
 
 Field observations will give some idea of how the soluble salt com- 
 ponents comj)are in this regard. l>ut the conclusions are necessarily 
 somewhat vague and unsatisfactoiy; for in the field and under the 
 conditions that are found in nature it is practically impossible to study 
 
 'See Bulletin No. 17, Division of Soils, U. S, Department of Agriculture (1901). 
 
 7 
 
the effect of any one soil component. It is rare indeed that the 
 "alkali" is composed of but one salt or chemical individual. And, 
 as will be brought ^ut later, it is entirely impossible to predicate 
 anything definite as to the action of a mixture of salts upon a plant 
 from a previous knowledge of the effects produced by each single salt. 
 Conversely, it is equally impossible to draw conclusions as to the action 
 of any one of a mixture of salts from observations of the effects pro- 
 duced by the mixture itself. 
 
 The more exact methods of the laboratory are necessary in order to 
 give us precise knowledge, and with this end in view the present inves- 
 tigation was undertaken. It is not claimed that the results so far 
 obtained are in all respects conclusive. The fact that only two species 
 of plants were employed in these first experiments is sufficient indi- 
 cation that they are not. In physiological research nothing is more 
 dangerous than generalization from the behavior of one or a few 
 species of plants to that of plant life as a whole. It is a well-estab- 
 lished fact that species differ widely in their reaction to a given 
 chemical or physical condition. Witness the fact that seaweeds will 
 thrive in water containing 1.5 to 3 per cent of sodium chloride, and 
 that salt marshes, wliose soil is saturated with water containing nearly 
 or quite as much of this salt, often support a luxuriant vegetation, 
 while the average crop is killed l)y a much more dilute solution of 
 sodium chloride. Certain plants show a marked aversion to limestone 
 soils, while other species are almost entirely limited to soils having 
 a high content of lime.^ But it is needless to multiply illustrations 
 of so familiar a phenomenon. 
 
 That a similar diversity is manifested by different cultivated crops 
 in their sensitiveness to various mineral salts when present in the soil 
 solutions is well known. Therefore we can not safely j)redict, until 
 experiments with many different plants have been made, that the 
 order of harmf ulness of the alkali salts here established for two plants 
 will be found to hold for all or even many of those which are com- 
 monly cultivated in the alkali regions. But, as it is obviously essen- 
 tial to the satisfactory prosecution of alkali soil work that a definite 
 standard for comparison of the salts be established, there need be no 
 further apology for the presentation of these first results of what it is 
 hoped will become an exhaustive investigation. 
 
 In the progress of the work numerous data were accumulated which 
 appeared to possess a more than ordinary degree of scientific interest, 
 especially as relating to the chemical theory of the dissociation of 
 electrolytes in solution and to the recently published hypothesis that 
 various salts, or rather their dissociated ions, enter into compounds 
 
 ' The interesting subject of "lime-loving" and '■ lime-avoiding" plants has been 
 much discussed by European botanists. It is synoptically treated by Drude 
 (Handbuch der Pflanzengeographie. p. 51) and by Schimper (Pflanzengeographie, 
 p. 105). The latter author gives an extensive bibliography. 
 
9 
 
 with the proteids of the protoplasm of plants and animals, whicli "ion 
 proteids " play a highlj^ important part in life processes and phenomena. 
 This aspect of the subject will be treated particnlai-ly in discnssinji' 
 the significance of the experiments with mixed solutions. 
 
 METHODS OF EXPERIMENT. 
 SALTS EMPLOYED. 
 
 In the selection of a series of salts for investigation the exj)erience 
 of members of the Division of Soils in field and laboratory served as 
 a guide. Salts were used which have been determined as forming 
 definitely injurious components of alkali soils and as occurring in 
 sufficient quantity to be of practical importance. In about the order 
 of their general abundance in the Western ITnited States these are 
 sodium chloride (NaCl), sodium sulphate (NaoS04), sodium carbonate 
 (NagCOg), sodium bicarbonate (NallCOg), magnesium chloride (MgCU), 
 magnesium sulphate (MgS()4), and calcium chloride (CaCL,). Inci- 
 dentally, experiments were made with gypsum (CaS0,2II.,0), calcium 
 carbonate (CaCO.5), calcium bicarbonate [Ca(HC()3)o], and with 
 magnesium carbonate (MgCOg), and bicarbonate [Mg(IIC03)2], as 
 well as with an aqueous solution of carbon dioxide (COo), the last in 
 order to test a theory that suggested itself during the experiments 
 with carbonates and bicarbonates. 
 
 In preparing and standardizing the solutions much assistance was 
 rendered by Mr. Seidell, of the Division of Soils. 
 
 The solutions were invariably made with salts manufactured by 
 Baker & Adamson, and found to be practically chemically pure, dis- 
 solved in distilled water. ' They were made up in each case on the 
 basis of a normal solution — i. e., of a gram-equivalent per 1,000 c. c. of 
 
 ' The water used in all experiments was distilled through a tin worm and was 
 collected and stored in Winchester quart bottles of practically insoluble glass. 
 A. conductivity test showed this to be an unusually pure water, but in order to 
 establish this point beyond doubt, a portion of this same water was redistilled 
 from glass, the first and last portions being of course discarded. A. test of the 
 distillate showed it to possess about twice as great conductivity as that which 
 tad been distilled only once from the tin. A comparison of cultures of lupines 
 in the water which had been only once distilled with that which was redistilled 
 showed practically no difference in the amount of growth made by the roots. As 
 Galeotti has lately shown [Biol. Centralbl., 21, 321I (1901)], the oligodynamic action 
 of relatively concentrated • • colloidal " solutions of metals disappears in the presence 
 of weak solutions of electrolytes. Thus a solution of copper containing 1 gram- 
 atom of metal per 126,000 liters of water produced no effect upon Spirogyra in 
 the presence of a 0,01 per cent solution of sodium chloride, and a solution of 1 
 gram-atom of copper per 63,000 liters of water acted only after twenty-four hours. 
 althou.;h in the absence of the electrolyte the toxic effect of the colloidal copper 
 solution is manifested at a dilution of 1 gram-atom of copper per 126,000,000 liters 
 of water. (See footnote, p. 50.) Hence it is practically certain that in the 
 experiments described in this report no complications were to be feared from the 
 possible presence of a trace of metals in the water used. 
 
10 
 
 solution. In other words, in the case of monovalent comj)ounds, one 
 gram-molecule was contained in a liter of solution, while in the case of 
 bivalent compounds, a half gram-molecule was present.' In this way 
 only is a really instructive and fair comparison of the effects of different 
 salts obtainable. Many experiments made in times past in which com- 
 parisons were l)ased upon simple percentages of solute to solvent l)y 
 weight are for this reason of far less value than if normal solutions 
 had been employed. In ordei' to studj^ comparatively those effects 
 produced by different electrolytes which are not dependent upon 
 their respective chemical natures, but which are common to tliem all 
 and due only to their active masses (such, for instance, as effects due 
 to the osmotic pressure existing in the solution), it is obviously neces- 
 sary to take into consideration the number of reacting weights of the 
 electrolyte introduced and the amount of electrolytic dissociation 
 which takes place. That is to say, one must consider the concentra- 
 tion of the solution with respect to the number of reacting chemical 
 equivalents, molecules, or ions which may be present. Moreover, 
 attempts to study comparatively the effects produced by different 
 kinds of ions in the solution can only be made bj^ approaching the 
 subject in this manner. But in all statements in this report of the 
 concentration of a given solution ])oth fractions of a normal solution 
 and parts of salt to 100,000 of solution ai'e given in order that the 
 results may be readil}' intelligible to readers who are familiar with 
 one or tlie other method, as the case may be. 
 
 The method pursued in these experiments was to make and care- 
 fully standardize a large volume of a normal solution of each salt 
 and then dilute to the required strength as occasion demanded. 
 
 In beginning the experiments the limit for each salt as determined 
 by investigators in the field was first tried, but immediately showed 
 itself to be too high. So lower and lower concentrations had to be 
 tested until the critical one was reached. 
 
 PLANTS SELECTED FOR EXPERIMENT. 
 
 For a variety of reasons the white lupine {Lupinus dlhus) was 
 emploj^ed in nearly all the experiments, although subse(iuently alfalfa 
 {Medicago sativa) was introduced for comparison. The lupine has a 
 seed of good size, averaging 10 to 12 mm. in greatest diameter. As an 
 abtmdant stipply of nutritive material is stored in the thick seed 
 leaves, there is no danger of starvation of the seedlings in experi- 
 ments of short duration such as those here described. The lupine 
 seeds germinate readily, sending out a vigorous radicle with clean, 
 
 'Dandeno [Bot. Gazette 32, 229 (1901)] has recently called attention to a certain 
 amount of confusion which has existed among both chemists and physiologists 
 as to the preparation of a normal solution, and it has seemed wise to describe in 
 detail the procedure followed in this investigation. 
 
11 
 
 bright, white surface. If the seeds are germinated in a proper medium 
 (spliagnum or peat moss saturated witli water was actually employed) 
 the root is usually straight or nearly- so. These cluiracteristics are 
 important, as they permit the easy and accurate measurement which 
 is essential to a. determination of the amount of gi'owtli made during 
 a given pei'iod. The white lupine has tlie further advantage of l)eing 
 a favorite subject for expei'iment with plant ])hysi()logists, so that 
 numerous data for com[»arisoirare available. 
 
 In one series of expei'iments lupine plants were used whicli had 
 been grown for eleven days in a prepared culture solution, and had 
 not only developed a considerable root system, but had unfolded two 
 or thi-ee leaves in addition to the seed leaves. In these plants all the 
 processes essential to the life of 'a mature imlividual were undoul)t- 
 edly in full activity. As a rule, however, a much earlier stage of 
 gi'owth was preferred, as cleai'ly affording a more sensitive index of 
 the effect of solutions. Experiments with older plants indicated that 
 they are less delicate registers of toxic elfect. An additional advan- 
 tage in using verj" young plants is that they are practically independ- 
 ent of the substratum so far as food sujiply (that is, the mineral 
 ash constituents) is concerned, that stored in the thick cotyledons 
 answering all purposes. Consequently the confusion which would 
 unavoidably arise if a culture solution of sevei-al salts c(mtaining 
 the necessary elements of plant food wer<^ introduced is avoided by 
 the emi)loyment of seedlings. 
 
 Lupine seedlings were transferred directly from the sphagnum, in 
 Avhich they had germinated twenty-four to forty-eight hours pre- 
 viously, to the solution in which the experiment was to be made. In 
 this stage of growth the seed leaves are still closely ai)pressed one 
 to another, and ai'c pale yellow in color. The initial root is 3 to 
 o cm. long, and shows as yet no indication of the app<'arance of lat- 
 eral branches. Care was taken to keep the moss so wet as to j)reclude 
 a normal development of root hairs; and in this respect the result 
 would be the same if the i-adicles had been immersed in water imme- 
 diately after germination. It was desired to render as slight as possi- 
 bl(^ the change of conditions in transferring fiom oiu' medium to the 
 other. There is every reason to believe that under these circum- 
 stances the amount of injury sustained by the plants as a result of the 
 change of substratum was reduced to a minimum.^ 
 
 ' Wolf demonstrated [Landwirtbsch.Versudvst.,6, '2(i:'.. (ISfU)] that pi ants whicli 
 iial been grown in soil until a considerable root system was developed and then 
 shifted to an aqueous solution (as in the experiments of Dt' Saiissure and others) 
 CO lid not be depended upon to give as satisfactory results as plants which had 
 been cultivated from the moment of germination in aqueous solutions. But in 
 the case of seedlings transferred from loose wet sphagnum to water before any lat- 
 eral roots had appeared no difficulty of this sort need be apprehended. 
 
12 
 
 From the experiments of others with plants cultivated in salt solu- 
 tions it would appear that Lupinus albus agrees pretty closely in 
 point of sensitiveness with other large-seeded Leguminosse, e. g., peas 
 {Piswm satiinmi), beans {Phaseolus vulgains), and, at least in some 
 eases, with the horse bean (Viciafaha).^ 
 
 In order to determine how closely plants of the same familj'^ corre- 
 spond in their resistance to toxic effect, and at the same time to 
 obtain data as to the behavior of a plant whose economic importance 
 in arid regions is inestinmble, a number of experiments were made 
 with alfalfa {Medicayo sativa). Here we have to deal with a plant 
 whose seeds are many times smaller tlian those of tlie white lupine 
 (1.5 to 3 mm. in greatest diameter). The radicle of the alfalfa seed- 
 ling is correspondingly small and delicate, and hence requires more 
 careful manii)ulation than does that of Lupinus. Alfalfa seeds were 
 germinated in wet blotting paper, and were transferred to the solu- 
 tions when the radicles were 1 to 2 cm. long. 
 
 A basis for comparison of the effects of toxic solutions upon plants 
 of very different character and relationship is afforded by lleald's 
 investigations of the action of extremely dilute solutions of hydro- 
 chloric acid upon seedling peas, pumpkins, and maize.^ Tliis author 
 calculates that while one part of hydrogen ions (liberated by dissocia- 
 tion) in G,400;{)00 parts of water killed the root tips of the pea {Pisum 
 sativiun),'-^ one part in 3,200,000 was required to produce a similar 
 effect upon the pumpkin {Cucurhita pepo) and one part in only 
 1,000,000 to destroy the root tips of maize {Zea mays). In other 
 words, maize offers four times and the pumpkin twice as much resist- 
 ance to the toxic effect of hydrochloric acid as do peas and lupines. 
 
 These results emphasize the importance of extending the present 
 investigations to other plants of as widely different botanical relation- 
 ship as possible. It is also of great moment that experiments be made 
 with different stages of growth of the same plant, from the germinat- 
 ing seed to some point near maturity. It is as certain that the same 
 kind of plant at various periods of development differs in its reaction 
 to a given salt solution as that the reaction of the same i^lant to the 
 same solution will be affected by variations of temperature and, per- 
 haps, of illumination.* 
 
 'But not always, for True [Annals of Botany, 9, 373, (1895)] found the white lupine 
 "more strongly affected by a 0.25 per cent solution than is Viciafaba by one of 
 1 per cent KNO3 content." He finds Pisum likewise more sensitive than Vicia 
 faba. 
 
 "Bot. Gazette, 22, 136 (1898). 
 
 ^The white lupine appears to be about equally sensitive to H-ions, for Kahlen- 
 berg and True [Bot. Gazette, 22, 91 (1896)] determined its limit of endurance in a 
 solution of HCl to be b^Iq^ normal, while later Kahlenberg and Austin [Journ. 
 Physical Chem., 4, 557 (1900)] fixed upon ^^Vo normal as a more accurate limit. 
 
 ^Storp [Landwirthsch-Versuchsst. , 13, 76 (1884)] found zinc sulphate to be 
 extremely injurious to germinating seedlings when exposed to the light, but 
 harmless, or nearly so, in the dark. 
 
13 
 
 The practical value of such a development of these studies is indi- 
 cated by certain conditions to which agriculture in alkali regions is 
 subject. It is well known that while at the beginning of the season 
 the salt components are often pretty equally disti-il)uted through a 
 considerable depth of soil and are in consequence comparatively harm- 
 less, the increased evaporation which accompanies increased temper- 
 atures and decreased atmospheric moisture as the season advances 
 draws these salts to the surface of the soil, wh-ere they often effloresce 
 and form "crusts" (especially in the case of sodium carbonate and 
 sodium sulphate). Hence older plants are frequently exposed to the 
 action of much more concentrated solutions than the same individuals 
 when younger had to contend with. Furthermore, the accidents of 
 irrigation may materially alter the alkali content of a soil in the midst 
 of the growing season of a crop. It is therefore to be hoped that this 
 important extension of the investigation may soon receive attention. 
 
 DETAILS OF MANIPULATION. 
 
 The manner of preparing the solutions and tlie plants to be culti- 
 vated has already been described. A few words about methods and 
 details followed in the experiments are in order. 
 
 To contain the solutions, glass vials nearly 3 cm. in diameter and 
 holding about 70 c. c. of liquid were used. In the experiments with 
 lupines, only one plant was suspended in each vial by means of a hole 
 bored through a close-fitting thin cork stopper, the aperture being 
 entirely closed by means of cotton batting. Protection against undue 
 evaporation from the upper portion of the plant was secured l\y plac- 
 ing several vials in a glass jar containing a little water and inverting 
 another jar over the whole. The plant was so adjusted in the cork 
 that 1 to 3 cm. of the terminal portion of the radicle was immei-sed in 
 the solution, the uppermost portion of the radicle extending through 
 the vapor-saturated space between solution and stopper, while the 
 hypocotylary section was invested with moist cotton. 
 
 In the case of alfalfa five or six plants were inserted in each vial in the 
 following manner: A piece of aluminum wire was passed through the 
 cork stopper in such a way as to allow it to be raised or lowered at discre- 
 tion. On the portion of the wire included in the vial five or six small 
 loops were made of proper size to hold in place each a seedling plant, with 
 its seed leaves resting on the loop and its root immersed in the solution. 
 
 The duration of the culture in the salt solution was generally lim- 
 ited to twenty-four hours, as it was usually possible at the end of that 
 period to determine accurately whether the root tip had been killed 
 or not. Frequently, however, the plants were returned to the solution 
 for a second period of equal dui-atiou in order to remove all doubt 
 upon this point. ^ If at the end of that period no growth had taken 
 
 ' In this particular, as in others, the experimental metiiods outlined by Kahlen- 
 berg and True [Bot. Gazette. 22, 87, 90 (1896)] have been followed, as it was 
 desirable to make as close comparison as possible with their results. 
 
14 
 
 place since the first examination, it was regarded as reasonably cer- 
 tain that the root tips had perished, and a less concentrated solution 
 was tried. To obviate the i^ossibility of mistaking a temporary condi- 
 tion of plasmolysis for final loss of vitality the roots were in earlier 
 expei-iments transferred, after twenty-four hours, from the salt solu- 
 tion to distilled water; but this precaution soon proved to be need- 
 less. In all the experiments a control culture in distilled water was 
 maintained under conditions of temperature and iliumiufltion iden- 
 tical with tho.se of the salt cultures. As a matter of course, the growth 
 of the roots- is by no means as rapid in distilled water as in ordinary 
 river water or in a prepared culture solution. 
 
 It was sought to keep the external conditions as nearly as possible 
 uniform during the entire series of experiments and a temperature of 
 19° to 21° C. was maintained in the laboratory.^ 
 
 The rate of growth during the period of experiment was ascertained 
 bj^ marking the radicle with India ink just before placing it in the solu- 
 tion. The mark, which was made as fine as was compatible wdth per- 
 manency, was placed at a distance of 15 mm. from the root tip in the ease 
 of the lupines and 10 nun. in the case of alfalfa (Medicago) so as safely 
 to include the entire zone of active growth in the primary root.- This 
 method of measuring the growth of roots was employed by Sachs in 
 his classical studies upon the growth of primaiy and lateral roots,'^ and 
 has been widely adopted b}' plant physiologists.' l^y comparison of 
 the marked root with a ruled sui-face the amount of growth during 
 any given period can be determined with all the accuracy necessary 
 in experiments of the kind hei'e described."' 
 
 By using a considerable number of individual plants in each experi- 
 ment with each solution (usually five in case of LuiDinus and ten or 
 twelve in case of Medicago) it is believed that the variant due to 
 individual differences in vigor has been practically eliminated.** 
 
 'In this connection Klemm [Jahrb. f. wiss. Botanik, 28, 659 (1895)] calls atten- 
 tion to the great variability exhibited by plants us to their limit of endurance in 
 solutions of acids of definite concentration if other external conditions be varied. 
 Askenasy [Ber. deutsch, bot. Gesellsch., 8, Gl (1890)] describes the effect upon the 
 growth of roots produced by different temperatures or by a variation of temper- 
 ature during a limited period of time. 
 
 ■Sachs determined the length of the grovfing portion, in the case of roots of 
 other Leguminos;v, to be 8 to 10 mm. for Viciafaba and 3.5 to 6.5 mm. in Pisvvi 
 sativum. [Arb. d. bot. Inst. Wixrzburg, 1, 413 to 419 (1873) : Gesammelte Abhandl., 
 2, 803 (1893)] 
 
 •'Gesammelte Abhandl.. 2. 778. 
 
 ■* For example, Kahlenberg and True use this method in all their experiments 
 with plants in solutions of toxic substances. [See Bot. Gazette. 22, 88 (1896)] 
 
 •"•Askenasy [Ber. d. deutsch. bot. Gesellsch., 8, 64. (1890)] shows that this method of 
 marking causes a retardation of growth during the first hour thereafter, but that 
 this is overcome after two hours. Consequently the method could be used with- 
 out hesitation in these experiments, although it is sometimes attended l)y disad- 
 vantages when the phenomena of growth itself are studied. 
 
 ''More than 3,500 seedlings of Lupiims albus and 700 of Medicago saliva were 
 employed in the whole series of experiments. 
 
15 
 
 Indeed, that this was the case was pretty effectually shown by several 
 repetitions of the experiments with most of the solutions. It is also 
 indicated by the general regularity witli which toxic effect is shown 
 to increase with every increase in concentration of the solution of 
 each salt. By several times repeating experiments with solutions 
 of approximately tlic ci-itical strength the aljove-mentioned source of 
 error due to fluctuations in temperature, etc., was likewise reduced to 
 a minimum. 
 
 DETERMINATION OF THE LIMIT OF ENDURANCE. 
 
 In ascertaining the degree of concentration of a given salt solution 
 which will just permit the root tips to retain their vitality during the 
 period of exiieriment, one must of course be able to determine also 
 the point at which death definitely occurs. The death point is evi- 
 dently to be sought far below the degree of concentration which- per- 
 mits no elongation wliatever to occur during the iieriod of experi- 
 ment, for often radicles, of which the marked zone liad increased in 
 length several millimeters (even 6) 'at some time during the experi- 
 ment, were indubitably dead at the end of twenty-four hours. ^ The 
 mere fact of elongation, irrespective of the time in wliich it has taken 
 place, does not therefore determine the concentration of a salt solu- 
 tion in which roots will survive, although sometimes useful in ascer- 
 taining whether the root is absolutel}^ dead at the end of a given 
 period. It is to the general condition of the apical portion of the 
 root that we must look for a criterion. While it is sometimes difficult 
 to describe those symptoms which denote the death of the root tip, 
 it is comj)aratively easy to recognize them after one has acquired 
 sufficient experience with the behavior of plants grown in toxic 
 solutions. 
 
 One of the most easily detected of the i^henomena accompanying 
 death in plants is final loss of turgor due to excessive plasmolj'sis. 
 In other words, the tissues lose their water, and are unable to make 
 good the loss, even wlien restored to normal conditions. This is due 
 primarily-to a change in the osmotic equilibrium of the plant cells. 
 Ordinarily, through the controlling activity of the protoplasm, a suffi- 
 cient osmotic pressure is maintained in the sap cavity of the cell to 
 
 ' Experiments were made with solutions of a strength known to be fatal, yet 
 permitting some elongation during twentj'-four hours. Sodium sulphate (0.05 
 normal), sodium carbonate (0.02 normal), and magnesium chloride (0.05 normal) 
 were selected, and in every case it was found that elongation ceased entirely after 
 three to five hours. In a water control, on the other hand, growth was still pro- 
 gressing at the end of six hours, and an examination at the end of twenty-four 
 hours showed that it had been pretty equally distributed throughout the entire 
 period. These results as to toxic action correspond with Sachs's statement [Land- 
 wirthsch. Versuchsst., 1, 219 (1859)]; Gesammelte Abhandl., 1, 430 (1892)] that 
 "roots appear to lose more and more the power of absorbing water containing 
 salt the longer they are in contact with it. " 
 
16 
 
 retain the necessary minimum of water. But through various influ- 
 ences, such as exposure of the tissues to a salt solution whose concen- 
 tration exceeds a certain limit, this power of adjustment may be 
 temporarily lost. In such cases a considerable proportion of the cell 
 water diosmoses through the ectoplasm, and the protoplast in conse- 
 quence shrinks away from the cell walls, to which it is normally 
 closely applied. If the unfavorable condition persists, this tem- 
 porary plasmolysis may become permanent, and the cell is killed 
 outright. 
 
 Such disorganization due to extreme plasmolysis can usually be 
 detected immediately by an examination of the plant tissues with the 
 microscope, and is one of the best indications of death. ^ Roughly, 
 however, injury of this nature is sufficiently indicated after a certain 
 lapse of time by loss of rigidity and elasticity in the plant or part of 
 a plant affected ; in other words, it becomes flaccid. If, for example, 
 a root thus rendered flaccid by culture in a salt solution fails to regain 
 its turgor after being transferred to water or to a nutritive solution, 
 it may safely be considt3red as injured beyond recovery. This was 
 found to be the most satisfactory test of death employed. ^ 
 
 The color of the tissues is often a useful symptom of destructive 
 changes. Thus all the sodium salts emploj^ed, when given in suffi- 
 
 ' " The only externally iierceptible change [indicating death] is in many cases 
 collapse, a more or less strong, irregular recession of the protoplast from the cell 
 wall, which does not, however, accompany by any means all reactions of sub- 
 stances which occasion death." [Klemm, Desorganisations-erscheinungen der 
 Zelle. Jahrb. fiir wiss. Botanik, 28, p. 657 (189.1).] 
 
 -Sachs [Arb. bot. Inst. Wtirzburg, 1, 386; Gesammelte Abhandl., 2, 774] men- 
 tions as an indication of the approaching death of the root tip the disorganiza- 
 tion of the cells of the root cap, which becomes mucilaginous. This was noted in 
 many cases, but was not found to be a practical test of complete loss of vitality. 
 Another indication of injury to the apical portion of the root is a sharp bend 
 near the tip, which is very different from the normal gentle curvatures. This 
 usually appears where loss of turgor from plasmolysis is not manifested. While 
 indicating injury, this symptom by no means necessarily implies complete loss of 
 vitality and, therefore, does not serve our purpose as a symptom of death. Solu- 
 tions of a certain concentration of magnesium sulphate, magnesium chloride, and 
 calcium bicarbonate were found to produce this phenomenon in a marked degree. 
 In the case of the salt last mentioned the roots continued to grow slowly in dis- 
 tilled water, during a second period of twenty-four hours. True [Ann. of 
 Botany, 9,377, (1895)] alludes to these " sharp curves characteristic of injury." 
 
 Another means of detecting loss of vitality in protoplasm, to which, however, 
 recourse was not had in the progress of this work, is its coloration when dead by 
 means of nigrosin, Vv^hich does not color and does not injure living protoplasm. 
 See Pfeffer [Ueber Aufnahme von Anilinfarben in lebende Zellen. Unters. aus d. 
 bot. Inst. Tubingen, 2, 268, 269], who found in experiments with roots of duck- 
 weed (Lemna) and with Spirogyra that nigrosin is not absorbed by cells while 
 alive. Living root hairs exposed for three days to a 0.5 per cent solution of this 
 stain assumed no coloration whatever, while hairs after death when similarly 
 treated readily absorbed it. 
 
17 
 
 cieiit amount, decolorized the tissues of the apical portion of the 
 root. This lost its normal brilliant white appearance ^ and assumed 
 a lurid-whitish color. In the case of sodium carbonate (NajCOg) 
 and of sodium bicarbonate (NaHCOg) there occurred a marked clear- 
 ing of the tissues similar to that produced by the hydrates of potas- 
 sium and sodium, the root tips becoming nearly transparent. This 
 change is completed long before any loss of turgor is apparent. Mag- 
 nesium salts (chloride and sulphate) discolored the surface of the 
 roots, producing brownish spots which gradually spread over the 
 whole surface. 2 The difference in character of i)hysiological effect 
 produced by salts of the same acid in the case of sodium on the one 
 hand, and of magnesium on the other, is very great when gauged by 
 these external appearances. 
 
 Another effect produced by some of these salts is an irregular 
 enlargement of a portion of the root. This is very marked in the 
 case of calcium chloride, in a solution of 0.3 normal or thereabouts. 
 The root just above tlu^ tip develops a fusiform swelling of which the 
 greatest transverse diameter (2 to 3 mm. ) lies 5 to 10 mm. from the 
 apex of the root. A less marked formation of this sort is .sometimes 
 produced by magnesium chloride, and even by other salts.-^ 
 
 It is well to emphasize once more the fact that the death of the tip 
 of the primary root, and not that of the plant as a whole or even of 
 the entire root, was taken in these experiments as the indicator of the 
 toxic action of solutions. The condition of the distal 10 to 20 mm. 
 
 ' The " shining white opaque appearaiace which is characteristic of all healthy- 
 roots and which is due to air contained in the Intercellular spaces." (Sachs, Land- 
 wirthsch. Versuchsst., 1, 216; Gesammelte Abhandl.. 1,427). 
 
 'Mettenius [quoted by "Wolf in Landw. Versuchsst., 7, 202, (1865)] found that 
 these spots, which appear on the roots of both the bean and maize when placed 
 In solutions of magnesium salts, are due to a coagulation of the contents of the 
 epidermal cells, which he did not, however, further describe. Wolf remarks that 
 they do not appear upon plant roots in magnesium salt solutions if a salt of jiotas- 
 sium, ammonium, or calcium be present. 
 
 ^ Sachs (Arb. bot. Inst. Wlirzburg, 1, 411, 412: Gesammelte Abhandl., 2, 800) 
 describes swellings of apparently similar character which developed upon roots 
 grown in moist air and watered at long intervals. Wolf [Landw. Versuchsst., 6, 
 218 (1864)] found that a concentrated solution of potassium sulphate acted in the 
 same manner. ''The root tips soon swell in the solution; the form of the root 
 finally resembles that of the root of a tuber-bearing plant. Such swellings arise 
 in particular abundance where lateral roots will break through." The action of a 
 one-fourth per cent solution of potassium nitrate upon roots of Lupinus albus as 
 described by True [Ann. of Botany, 9, 374 (1895)] is exactly similar to that of cal- 
 cium chloride. '• Swellings appeared near the tips and the ends tapered suddenly 
 to sharp points. On the other hand, the growth in thickness was much greater 
 than normal, the radicles above the swellings reaching the size of large radicles 
 of Vieia fabcr of the same length . " 
 
 8287— No. 71—02 2 
 
18 
 
 onl}' was necessarily involved.^ In the more dilute solutions which 
 are still considered toxic, because destroying the root tip, the proxi- 
 mal portion of the root and the upper part of the plant are often not 
 conspicuously injured by twenty-four or forty-eight hours' exposure. 
 After a certain lapse of time lateral roots are sometimes put forth 
 and grow vigorously in a solution (especially of calcium chloride) 
 which had killed the apical portion of the primary root. 
 
 This power of gradual accommodation on the part of the plant to a 
 solution which at first checked its growth and even destroyed the 
 sensitive tissues of the root tip has often been remarked. It is but a 
 step from this to the well-known fact that by gradually increasing the 
 strength of a salt solution in which plants are cultivated they can be 
 made to endure a degree of concentration which would soon be fatal 
 if administered directly.^ It follows that the limits of endurance here 
 recorded for Lupinus albus are merely those of its root tip, selected 
 as being the most sensitive indicator, and are in soine cases lower than 
 the limits which would denote death of the plant as a whole. Further- 
 more, the limit of endurance for the entire plant could undoubtedly be 
 still further elevated by gradually increasing the strength of solution 
 in which the plants are cultivated. 
 
 But our present investigation aims merely at a comparison of the 
 relative toxicitj^ of the various "alkali" salts, to attain which the 
 simplest and readiest means are to be preferred. A standard for 
 further comparisons, rather than a thorough investigation of the 
 problem in all its ramifications, is the end of the present paper. 
 
 ' This was likewise the objective of the experiments of Kahlenberg and True 
 [Bot. Gazette, 22, 88, (189())]. In order to obtain results closely comparable with 
 theirs, especially as bearing upon the hypothesis of electrolytic dissociation, their 
 mode of procedure has been closely followed in this as in other details. In advocacy 
 of this method of determining toxic action, Professor True writes: ''Repeated 
 experiments for years have convinced me that the method used gives the most deli- 
 cate and easily managed test that I know of for bulky objects like Lupinus roots." 
 
 Coupin [Rev. Gen. de Botanique, 10, 177 (1898)] criticises the work of Kahlenberg 
 and True, previously quoted, to the effect that it Is impossible to accurately deter- 
 mine the toxic limit of a solution in the short period of experiment (twenty-four 
 hours) allowed by those authors. However, as Professor True observes, it was not 
 the point at which the whole plant succumbs, but that which marks the death of 
 the zone of growth in the primary root, which formed the objective of his experi- 
 ments. Coupin "s method was to grow his plants for several days in the solutions 
 to be tested, taking the strongest solution in which the plant as a whole continued 
 to grow after the first few hours as mai-king the limit of endurance ("equivalent 
 toxique'). It is obvious, therefore, that no direct comparison is possible between 
 the results obtained by Coupin on the one hand and by Kahlenberg and True, as 
 well as those here recorded, on the other, Coupin's limits of endurance being 
 necessarily much higher. 
 
 - Thus Stange [Bot. Zeitung. 50, 292 (1892)] ionnd that root tipao^Lnpimis albus 
 and Phaseolus viiJgdris soon died if exposed directly to a 0.5 per cent solution of 
 potassium nitrate, but by gradually increasing the concentration they could be 
 made to endure nearly 1 per cent without death of the protoplasm. 
 
19 
 
 RESULTS WITH PURE SOLUTIONS. 
 
 CONCENTRATION MAXIMUM PERMITTING SURVIVAL OF THE ROOTS. 
 
 By applying the methods and tests outlined above it was possible 
 to determine with a r<*asouable degree of accuracy tli<^ limit of concen- 
 tration for each of the salts in i^ure solution in which the root tips of 
 young seedlings of white lupine could just survive. It is believed 
 that, like conditions being maintained and the same plant in the 
 same stage of development being used, the limits will not be materially 
 altered by further experiment. Moreover, it is regarded as not improb- 
 able that the salts will be found toxic in about the order stated below 
 if other plants or other stages of growth of the same plant be tested 
 with them. The limit of endurance in a solution of each particular 
 salt will doubtless be higher or lower for different objects, but the 
 general sequence of harmfulness should remain ])ractically unaltered, 
 so far as the higher plants are concerned. Experience alone can 
 demonstrate the correctness of this assumi^tion. 
 
 The limit of concentration permitting roots of white lupine to 
 retain their vitality during twenty-four hours is, for each of the more 
 important readily soluble "alkali" salts, as follows, the limit being 
 stated both in parts of salt per 100,000 of solution and in fractions of a 
 normal solution : 
 
 Table I. — Results of experiments with pure solutions. 
 
 Name of fait. 
 
 Magnesium sulphate . . . 
 Mafrnesium chloride ( 1 ) 
 
 Sodium carbonate 
 
 Sodium sulphate (3) 
 
 Sodium chloride (3) 
 
 Sodium bicarbonate (4) 
 Calcium chloride 
 
 Degree of concen- 
 tration. 
 
 Parts per Fractions 
 100,(XK» i of a nor- 
 of solu- mal solu- 
 tion, tion. 
 
 7 
 
 13 
 26 
 
 116 
 
 167 
 
 1,377 
 
 0.0013.5 
 .0035 
 .005 
 .0075 
 .03 
 .02 
 .35 
 
 Notes.— (1) With magnesium chloride the limit of endurance (for the whole plant), as deter- 
 mined by Coupin [Rev. Gen. deBot., 10, IHK ( lK!t8)],is0.8 per cent, while with magnesium sulphate 
 the limit is 1 per cent, thus reversing the order of toxicity for the two salts as given above. 
 Wolfe (Landw. Versuchsst., 6, p. 211) notes the strongly toxic eflfect of magnesium solutions upon 
 roots of bean and maize. The brown coloration of the surface of the radicle, induced by these 
 salts, appeared a few hours after immersion. Wolf's suggestion that the very poisonous effect 
 of magnesium sulphate may be iluo to the decomposition of the salt by excretions of the roots 
 can not be regarded as possessing great probability. His experiments, which were designed 
 primarily to ascertain the volume of water absorbed by the plant from solutions of various salts 
 of different concentration, are considered by him to iiidicate that the cell wall [ectoplasm] is 
 less permeable to sulphates than to other salts (1. c, p. 217). Loew (Bui. No. is, Div. Veg. 
 Phys. and Path., p. 42) found that Spirogyradied after four or five days of immersion in a 0. 1 per 
 cent solution of magnesium sulpliate, but remained alive for a long period in equivalent solutions 
 of sulphates of potassium, sodium, and calcium. Similarly a 1 per cent solution of magnesium 
 nitrate killed a smaller Spirogyra in six to twelve hours, while the nitrates of potassium, sodium, 
 and calcium, in solutions of corresponding strength, did not destroy the plant. The peculiarly 
 poisonous action of salts of magnesium described by Loew is explained by him on the hypothesis 
 that calcium forms intimate compounds with proteids. and that these are essential to the organ- 
 ization and life of the cell-nuclei and chloroplasts of the higher plants. Consequently, if mag- 
 nesium is supplied without calcium to plants, especially in the form of readily soluble salts, such 
 as chloride, nitrate, and sulphate, the acids of the magnesium salts would be attracted by the 
 calcium which formed part of the nuclear proteid compounds. The latter would consequently 
 be disorganized, magnesium being unable to take the place of calcium in proteid compounds 
 withoutfatal disturbances of eciuilibrium inthecell. Asevidencefor thishypothesisisadduced 
 the corrective effect of the addition of lime to either soils or culture solutions in which plants 
 are suffering from magnesium poisoning, and the further fact that plants suffer less in culture 
 solutions from which both calcium and magnesium are absent than in such as contain magne- 
 ium but no calcium. It must bo observed, however, that the chemical rationale of this theory 
 
20 
 
 rests upon the assumption that calcium is a stronger base than magnesium, and will exert a 
 greater attractive force upon acids, while it ignores the application of the mass law to the dis- 
 tribution of an acid between two bases, which itself accounts very satisfactorily for the facts 
 obssr vGcl • 
 
 (3) Of sodium sulphate Wolf (Laudw.Versuchsst., 6, pp. 310,213) indicates that solutions of more 
 than 0.05 per cent are toxic to roots of the bean (Phaseotus vulgaris). Loeb [Am. Journ. Physiol- 
 ogy 3, 393 (1900)], found sodium sulphate to be more poisonous than sodium chloride to eggs 
 of a-fish (Fnndulus heteroclitus). This he attributes to a ijreeipitation of calcium from its ion 
 proteid compounds in the protoplasm, a reaction effected through the sulphions dissociated by 
 sodium sulphate. , , . , , , , ^, ^, , 
 
 (3) The minimum toxic concentration for sodium chloride, the same plant and the same methods 
 being used, is placed about three times as high (one-sixteenth normal) by True [Amer. Journ. 
 Scl,ser. 4.9, 187 (1901)]. As the experiments with sodium chloride here described were repeated 
 several times, without variation in the result, no explanation for this discrepancy is apparent. 
 
 Many experiments have been made with sodium chloride as to its effects upon v)lants. It may 
 be of interest to refer to some of those in which limits of endurance have been determined, 
 especially as these are in all cases much higher than that given above for root tips of Lupinu« 
 albus. Storp fBiedermann's Centralbl., 13, 7(i ( 188-t)] found that sodium chloride in a solution of 
 greater concentration than 0.01 per cent retarded the germination of seeds. Eschenhagen [Ueher 
 den Einfluss von Losungen verschiedener Concentrationeii auf den Wachsthum der Schimmel- 
 pilze (1889)], quoted by Stange in Bot. Zeitung (1893, p. 255), gives the following limits for the 
 active growth of fungi in solutions of sodium chloride and of sodium nitrate: 
 
 Fungus. 
 
 A.spergillus 
 Penicillium 
 Botry tis — 
 
 Richter [Uebor die Anpassung der Siisswasseralgen an Kochsalzlosungen Flora, 75, 4 (1892)] 
 found that Zyyni'iiin xtellinum qenninum lived two months in a 6 per cent solution of sodium 
 chloride added to a culture solution, and more than a year when the sodium chloride solution 
 was 2 j)er cent or weaker. De Freitag [Archiv fiir Hygiene, 11, 08 (189i))] is authority for the 
 statement that fidrUlns tHberrnlosis lived three months, and the typhus Bacillus six months 
 in a saturated .solution of sodium chloride. Coupin [Revue G-en. de Botanique, 10, ITT (1898)] 
 obtained the following limits for various plants in solutions of sodium chloride: 
 
 Plant. 
 
 Per cent 
 limit of 
 endur- 
 ance. 
 
 Wheat : - 1-8 
 
 Peas 13 
 
 White lupine ^ 1-2 
 
 Maize I I-* 
 
 Vetch 11 
 
 0.5 
 .35 
 
 According to W. Sigmund [Landw. Versuchsst., 47, 1 (189t))] the maximum concentratifm of 
 NaCl solutions endurable by germinating seeds of cereals is 0.5 per cent, of legumes 0.3 per 
 cent, of rape 0.1 per cent Loew [Bui. 18, Div. Veg. Phys. and Path., p. 19] found that Spirogyra 
 suffers in a solution containing 0.5 per cent of sodium chloride. 
 
 (+) Carbonic acid (HCO3) is here regarded as a monovalent acid, so that a gram molecule 
 (instead of one-half of a gram molecule) to the liter has been used in making up normal solu- 
 tions of sodium bi<'arbonate. To prevent inversion to the normal carbonate (Na.2C03) |see 
 Cameron and Briggs. Bui. 18, Div. of Soils, UWO; also Jour. Physical Chem., 5, .537 (1901)] solutions 
 of the bicarbonate were always well charged with carbon dioxide and were tested for 
 hydroxyl with phenolphthaleine before being used in culture experiments, and again at the end 
 of the experiment. It is quite possible, of course, that a small error was thus introduced, as tin; 
 carbonic acid formed l)y the dilution of carbon dioxide in water may have retarded somewhat 
 the dissociation or ionization of the sodium hydrogen carbonate. It is improbable that sodium 
 hydrogen carbonate, unaccompanied by the normal carbonate, would ever occur in nature 
 except in the presence of an excess of carbon dioxide, which fact is a further justification of 
 the procedure here described. 
 
 In order to demonstrate that this excess of carbon dioxide was not in itself injurious to the 
 roots ot white lupine, the following simple check experiment was made: Carbon dioxide was 
 forced into distilled water until a saturated solution was obtained. Plants were then entered in 
 this solution, which was protected as comjjletely from less of carbon dioxide as circumstances 
 would permit. After twenty-four hours the solution was tested with barium hydrate, and the 
 heaviness of the resulting precipitate of barium carbonate showed that very much more carbon 
 dioxide still remained than is present in ordinary water. During this period the roots grew 
 nearly as well as in water containing only the normal quantity of carbon dioxide. It might be 
 supposed that a solution of carbon dioxide in water and presumaljly containing the hypothet 
 ical carboiiicacid must needs be itself quite toxic, as it would be expected to yield the hydrogen 
 ion which recent investigations have shown to be excessively toxic. In this connection some 
 work of Pfeiffer [Ann. Cliein. (2),23,ti35 ( isyi)] will prove interesting. This investigation showed 
 that a solution of carbon dioxide is an exceedingly poor conductor: that in fact the highest con- 
 ductivity observed in such solutions was only about a thousandth of that which Kohlrausch's 
 work showed it should possess. See also Knox [Ann. Phys. Chem., 54, H (1895)] and Walker and 
 Cormack (Journ. Chem. Soc, 77, 5 (1900)]. 
 
 It would seem rational, therefore, to consider that carbonic acid does not exist itself, or <at 
 least in only minute quantities in solutions of carbon dioxide, but is potetitialhj present in its 
 constituents and only forms in the presence of some added influence, su(;h as a base. And that, 
 therefore, even a concentrated aqueous solution of carbon dioxide would contain no hydrogen 
 ions, or so very small a quantity as to be ineffective against so delicate an indicator as a plant 
 root. 
 
21 
 
 Expei'inients to Jiseertain llic liinitof eiuhiranco in \>uvo solutions 
 were also made with seedlings of alfalfa {3f('(lir(i(/o sufira). Although 
 absolute limits foi- this plant have not, as yet, been determined, they 
 appear to be somewhat lower for every salt tlian 'withe aiso, ot Ltipnius 
 albtis, hut more than one-half as high. Thus for magnesium sulphate 
 the limit appears to lie between 0.00()(J25 and O.OOl^o normal, while for 
 magnesium ehloride the limit will be found between O.OOIl'o and 0.0025 
 normal. 
 
 A glance at the pi-eeeding table shows very clearly that it is the 
 basic rather than the acid radicle of the salts nsed which chieflj' deter- 
 mines theii- relative toxicity. Tn other words, the cathions derived 
 from these salts are vcM-y much moi'e active in their effect upon plant 
 tissues than are the anions. This is strikingly brought out by a 
 comparison among tluMiiselves of the three chlorides of magnesium, 
 sodium, and calcium, on tin; one hand, and of the chlorides and 
 snlphat(!s of magnesium and sodium, respectively, on the other. Tn 
 the f()i'mer(;ase, although the anions (CI) ai'c identical in kind we find 
 nmgnesium chloi'ide <Mght times as toxic as sodium chloride^ and one 
 hundred tinu's as toxic as calcium chloride. Tn the latter case, mag- 
 nesium sulphate is only twice as toxic as magiuvsium chloi-ide, while 
 sodium sulphates is little moi-e than twoand one-half tinu's as injui-ious 
 as the ('(U'respoiuling chloride. 
 
 The results with salts of magnesium, as compared with thos<; of 
 sodium, confirm the i-esults obtained by W. Wolf, Loew, and others 
 as to the sti'ongly i)oisonous cpialities of the former base. 
 
 All four of the salts of scxlium with which ('X[)eriments W(M'(; 
 nuuh' ar<^ widely distributed and often very abundant in the alkali 
 r(\gions of the W(\stern Tnited States. As was t-o be expectc^d, sodium 
 carbonate or black alkali was found to be the most harmful of tln^se, 
 but it is not much more injurious than sodium sulphate. That the 
 latter is much more poisonous than sodium chloride is a result not 
 altogether anticiiiated at the beginning of thi^ investigation.' As was 
 predicted, sodium bicarbonate proved to be somewhat less toxic than 
 sodium chloride.- 
 
 As a matter of fact, the limit of endurance in a solution of sodium 
 bicarbonate is not mnch higher in pai'ts of salt per 100,000 of water 
 
 'Stewart [Ninth Ann. Rep. Utah Agr. Exp. Sta. p. 26 (1898)] found sodium 
 chloride more injurious than sodium sulphate to germinating seeds of legumes and 
 cereals. 
 
 -Very different results from these here recorded as to the relative toxicity of the 
 carbonate and bicarbonate of sodium were obtained by Coupin [Rev. G-en. de 
 Botanique, 12, 180 (1900) J. Experimenting with seedlings of wheat, this author 
 found that the least concentrated fatal solution ( • • equivalent toxi(iue'" ) is 1 . 1 grams 
 per 100 of water for sodium carbonate, while for the bicarbonate it is O.G gram. 
 Hence the latter would be twice instead of one-fourth as poisonous as the former. 
 Sigmund [Landw. Versuchsst., 47, 2 (1896)] found that while Na.,C().,al. a concen- 
 tration of 0.5 per cent killed germinatmg seedlings of vetcli and rape and retarded 
 the development of wheat seedlings, NaHCO^ at the same concentration was 
 harmless. 
 
22 
 
 and is no higher in fractions of the reacting weight than it is for 
 sodium chloride. It should be mentioned, however, that the plants 
 survive in a solution of the bicarbonate of the strength given in the 
 table in much better condition than in the corresponding concentra- 
 tion of the chloride, so that the latter must be regarded as the more 
 harmful of the two salts. The wide distribution of sodium bicarl)on- 
 ate and its abundance as a component of many alkali soils renders 
 the'^demonstration of its marked poisonous effects upon vegetation, 
 even when present in comparatively dilute solutions, a matter of no 
 little importance. Although much less injurious than is the normal 
 carbonate or "black alkali," the presence of this salt can not be 
 neglected in future estimations of the value of western soils. 
 
 An explanation of the harmful action of sodium bicarbonate which 
 at first suggested itself was that by its dissociation free hydrogen ions 
 are liberated, though the weight of evidence on chemical grounds is 
 rather against this view.' It has been shown b,y recent investigators^ 
 that it is probably the hj^drogen ions dissociated by certain acids 
 (especially the strong mineral acids) which make them so injurious to 
 orgaiiisms, even in t^xtremely dilute solutions. If this were the reason 
 for the toxicity of sodium bicarbonate it would follow that water heavily 
 charged with carbon dioxide, as in the check experiment described 
 above (p. 20), would prove similarly injurious to plant roots bj^ reason 
 of the dissociation of hydrogen ions by the carbonic acid (IICO.5), 
 which is supposed to be formed when carbon dioxide is dissolved in 
 water. But, as has already been noted, no toxic effect was obtained 
 with an aqueous solution of carbon dioxide.'' 
 
 ' Walker and Cormack, Joiirn. Chem. Soc, 47, 5 (1900) and Bodlander, Zeit. fur 
 physik. Cliem., 35, 25 (1900). 
 
 •Kahlenberg and True, Bot. Gazette, 22, 87 (1896); Heald, 1. c, p. 134; Loeb, 
 Pfluger's Archiv f. die gesammte Physiologie, 69, 4 to 9 (1898); Kahlenberg and 
 Austin, Journ. of Physical Chem., 4, 553 (1900); True, Amer. Journ. Sci. ser. 4, 9, 
 183 (1900). 
 
 'There exists among plant physiologists some diversity of opinion as to the direct 
 effect of large quantities of carbon dioxide upon the growth of roots. The sub- 
 ject is evidently one which needs a more thoroughgoing investigation, not only 
 from a scientific standpoint, but from economic reasons also, as it is intimately 
 connected with tillage and drainage problems. For an extended discussion of this 
 question see Lopriore in Jahrb. fiir wiss. Botanik, 28, 531 (1895). The author men- 
 tions that Boehm found roots of the bean {Phaseolns vulgaris), when exposed to 
 an excess of carbon dioxide, to be shorter, and the lateral roots fewer, than is 
 ordinarily the case. Jentys [Bui. Internat. Acad. Sci. Cracovie, 1.S92, 306 (1893) J, 
 found that by passing atmospheric- air to which had been added 4 to 12 per cent of 
 carbon dioxide through the soil of culture pots, an injurious effect upon the roots 
 of the bean and the yellow lupine could be detected, although the injury was 
 less than in Boehm "s experiments. On the other hand, wheat was practically 
 unharmed. Lopriore (1. c.p. 623), concludes that carbon dioxide in excess has a 
 hindering but not a permanently injurious influence upon the functions of proto- 
 plasm. This effect is not ascribable to the absence of oxygen, but is specific. 
 Plant cells can gradually accommodate themselves to a quantity of carbon dioxide, 
 which, it applied directly, would injure them. Lopriore "s experiments were made 
 chiefly with Mucor, yeast, and pollen grains and tubes. 
 
23 
 
 As there is probably but a small difference in the amount of sodium 
 ions yielded by sodium chloride and by sodium hydrogen carbonate, 
 at the dilutions here involved, the difference in their toxicity observed 
 must in all probability be ascribed mainly to the anions. 
 
 It is likely that the great toxicity of normal sodium carbonate is 
 largely due to the hydroxyl ions resulting from the hydrolysis of 
 this salt. In the case of the bicarbonate of sodium in all the experi- 
 ments involving its use, and described in this paper, hydrolysis was 
 avoided by dissolving carbon dioxide in the solution in amounts suf- 
 ficient to prevent any inversion to the normal carbonate, a reaction 
 which would necessarily result were hydrolysis of the bicarbonate 
 permitted.^ Since it seems reasonablj^ certain that HCO3 ^^^^^ ^^'^ ^^^^ 
 toxic, the toxic influence of the sodium bicarbonat(^ solutions could 
 be safely attributed to the sodium ion alone were it not for the fact 
 that toxic solutions of this salt produce the peculiar " clearing " effect 
 upon plant tissues which is well known in the case of the normal car- 
 bonate of sodium and of the hydrates of potassium and of sodium. 
 This effect is very different from that caused by other salts of sodium, 
 e. g., the sulphate and the chloride. 
 
 Calcium chloride was found to be ten times less injurious tlian is 
 sodium chloride. For this reason, and because it rarely i)redominates 
 in areas of any considerable size, this salt can not be regarded as, 
 under ordinary (urcumstances, a dangerous compontnit of alkali soils. 
 As we shall in-esently see, there is reason to believe that it can in 
 many cases be a highly beneficial component of the soil. 
 
 Attention should be directed to the fact that the figures given in 
 the above table represent oidy ai)proximate results, the determination 
 of the absolute limit for each salt depending tlieorotically upon the 
 testing of an almost infinite number of concentrations. Thus, as a 
 rule, solutions of a concentration of 0.2, 0.15, 0.1, 0.075, 0.050, etc., 
 normal were employed, although more numerous intermediate concen- 
 trations, e. g., of 0s2000, 0.1825, 0.1750, etc., normal could have been 
 tested. However, it is doubtful whether the reaction ujion plant 
 tissues of finer differences could be detected, and it is believed that 
 for all practical purposes a sufficient number of concentrations was 
 used. As has already been noted, the limits of endurance in the case 
 of different salts are not of precisely equal value, the roots not sur- 
 viving in all in exactly the same condition. Thus roots which survived 
 after twenty-four hours in a 0.005 normal solution of sodium carbonate 
 presented a perfect appearance and grew vigorously in distilled water 
 during a subsequent period of twenty-four hours. On tlie other hand, 
 roots which endured a 0.25 normal solution of calcium chloride pre- 
 sented a markedly abnormal aspect at the end of twenty-four hours, and 
 made little subsequent growth when transferred to water. Likewise 
 
 'See paper on Equilibrium between Normal Carbonates and Bicarbonates in 
 Aqueous Solutions, Cameron and Briggs. Bui. 18, Div. Soils, U. S. Department 
 Of Agriculture (1901); Jour. Physical Chem., 5, 537 (1901). 
 
24 
 
 roots survived in better condition in 0. 0075 normal sodium sulphate than 
 in 0.02 normal sodium chloride solution. It was found much easier to 
 determine sharply the limit of endurance for sodium carbonate and 
 sodium bicarbonate than for other salts, as in 0.005 and 0.02 normal 
 solutions of the two carbonates, respectively, all, or nearly all, roots 
 survived in apparently perfect condition, while in 0.0075 and 0.025 
 normal, respectively, all roots were killed and symptoms of advanced 
 disorganization were apparent after twenty-four hours. 
 
 CONCENTRATION MINIMUM PROHIBITING ELONGATION OF ROOTS. 
 
 A comparison of the seven salts above enumerated in regard to the 
 degree of concentration of each in which absolutely no elongation of 
 the roots occurred during twenty-four hours is interesting, as illus- 
 trating how far this ijoint is removed from that of the minimum con- 
 centration which is still toxic. It will be seen that the position of the 
 salts in this scale does not at all correspond with their sequence in 
 the table of limits of endurance. In many cases, especially when the 
 solution was still more concentrated, not only no increase of length 
 but a positive shrinkage of 0.5 to 2 mm.^ was detected. 
 
 Table II. — Concentrations which absolutely prevent growth. 
 
 Concentration of 
 solution. 
 
 Name of salt. 
 
 Sodium carbonate . . . 
 Sodium bicarbonate . 
 Magnesium chloride . 
 
 Sodium chloride a 
 
 Sodium sulphate 
 
 Calcium chloride 
 
 Magnesium sulphate 
 
 Parts per 
 100,000 of 
 solution. 
 
 260 
 417 
 960 
 1,160 
 1,410 
 1,652 
 1,680 
 
 Normal. 
 
 0.05 
 .05 
 .2 
 .2 
 .2 
 .3 
 .3 
 
 aAccordiug to Pfeffer (Pflanzenphysiologie, Ed. 2, 1, 414) a culture solution to which enough 
 potassium nitrate or sodium chloride is added to render it isosmotio with a 3 percent potassium 
 nitrate or 1.7 per cent sodium chloride solution causes a cessation of growth lu ordinary plants 
 while an increase to 3 per cent is necessary to prevent growth in halophytes. 
 
 It is impossible to reconcile this sequence, as compared with that of 
 Table I, with the notion, which still appears to find advocates, that the 
 injurious effect of these salt solutions is merely a function of their 
 osmotic pressures. If any fresh evidence were needed to disprove this 
 assumption it is afforded by the fact, very clearly brought out in the 
 present investigations, that marked toxic effects frequently appear 
 long before loss of turgor has manifested itself or cessation of growth 
 has occurred. It is certain that no useful conclusions as to the degree 
 of toxicity of a solution can be drawn from its osmotic pressure. 
 
 True [Bot. Gazette, 26, 407 (1898)] calls attention to the difficulty of 
 distinguishing the purely chemical from the merely osmotic (plasmo- 
 
 ' In some solutions this loss of length due to plasm aly sis was as great as that found 
 by Sachs in roots which were exposed for thirty minutes to the dry air of a room. 
 (Arb. bot. Inst. Wtirzburg, 1, 396; Gesammelte Abhandl., 2, 784, 785.) 
 
. 25 
 
 lyzing) effect of a salt solution. He experimented with SpirogjTa in 
 order to obtain means of making such distinction, comparing its 
 behavior in a solution of cane sugar, which is believed to possess no 
 chemically toxic properties, with that in solutions of sodium chloride 
 and ijotassiura nitrate. The maximum concentration of the sugar 
 solution in which life could be maintained was determined to be 0.75 
 normal. Allowing for differences of dissociation, 0.46 normal should 
 then be the maximum endurable concentration of a sodium chloride 
 solution if only its osmotic pressure were involved. In fact, however, 
 0.1 normal was found to be the actual limit, so that a definite toxic 
 action of sodium chloride must be admitted (loc, cit., p. 410).' 
 
 Were the injurious action of these solutions attributable to plas- 
 molysis alone, an approximately equal amount of elongation should 
 take place in solutions of different salts, if each solution contain an 
 equal fraction of a gram equivalent to a given amount of water, grant- 
 ing tliat the dissociation of each salt w^as equally great a1 the given 
 concentration, as would be approximately true foi' stiong electrolj^tes 
 at the concentrations here used; for elongation and growth in general 
 are intimately connected with the turgor conditions of the tissues,^ 
 which, in turn, depend upon the osmotic force exerted by the sur- 
 rounding solution. That force being equal for each of two solu- 
 tions, the turgor and the amount of elongation of the roots immersed 
 in each should also be equal if osmosis were the only factor involved. 
 That this is not the case is sufficientlj^ established by the figures given 
 in Table II. 
 
 RESULTS W^ITH LESS SOLUBLE SALTS. 
 
 Besides the easily soluble alkali salts a few othei-s were used in 
 experiments, i. e., calcium sulphate [CaS04] calcium carl>onate 
 [CaCOg], calcium bicarbonate [Ca(HC03)2], and the carbonate and 
 bicarbonate of magnesium [MgCOg and Mg(HC03)o]. These Avere 
 found to be either toxic in a very slight degree, indifferent, or posi- 
 tively stimulating to growth. 
 
 ' From True's results it is clear that at the concentrations involved in our experi- 
 ments with pure solutions the toxic effect observed must in every case be referred 
 to action of a chemical rather than a purely physical nature. In some of the 
 mixed solutions, such as the very concentrated ones containing calcium sulphate, 
 it may be that their osmotic pressure determines the limit of endurance of the 
 plant roots. 
 
 - For example that, except perhaps in rare instances, growth can not be resumed 
 after an interruption (such as is occasioned by transference of plants from one 
 medium to another) unless the turgor of the plant or the organ concerned is nearly 
 or quite normal, was shown by Curtis [Bui. Torr. Bot. Club, 27, 1 (1900)] in the 
 case of mycelia of Mucor, Botrytis. and Penicillium, grown in a plasmolyzing 
 solution (4 per cent potassium nitrate). As this author expresses it, " there is a 
 necessity of a certain turgor force before growth is possible, and growth can not 
 occur until a turgor pressure has been reached which is normal to the plant grow- 
 ing in the given solution.' (Loc. cit., p. 11.) 
 
26 
 
 lu a (necessarily dilute) solution of i>yi)sum, which contained a con- 
 siderable quantity of the undissolved salt in suspension, the plants 
 grew decidedly more vigorously than in pure water.' 
 
 In a saturated, but necessarily very dilute, solution of normal cal- 
 cium carbonate [CaCOg], roots of Lupinus elongated nearly twice as 
 much and remained in decidedly better condition during twenty-fou?* 
 hours than in distilled water. This solution gave a faint reaction for 
 hydroxyl (with phenolphthaleine) at the beginning of the experi- 
 ment, but none at the end of tw^enty-four hours, doubtless because of 
 the production of carbonic acid through the excretion of carbon 
 dioxide by the roots. But a solution of calcium bicarbonate 
 [Ca(HC03)2], made b.y saturating a portion of the same calcium car- 
 bonate solution with carbon dioxide, permitted only about one-third 
 as much growth of the roots as took place in distilled Avater. Their 
 condition was decidedly abnormal at the end of twenty-four hours, 
 even the turgor being poor.^ 
 
 Magnesium carbonate [MgCOg], in a solution w^hich gave a strong 
 hydroxyl reaction with phenolphthaleine, allowed the roots to grow 
 about as rapidly as in distilled water and to remain in about normal 
 condition. On the other hand, a portion of the same solution to 
 which an excess of carbon dioxide was added and in Avhich no free 
 hydroxyl could be detected (either before or after the experiment) 
 exerted a strongly toxic action upon the roots. These made practi- 
 cally no growth during twenty- four hours; their turgor became 
 decidedly inferior, and there occurred a marked discoloration of 
 brownish spots, such as is produced by the readily soluble mag- 
 nesium salts. Here it is obviously a case of a greater amount of 
 magnesium in solution, owing to the i)resence of carbon dioxide in 
 
 ' The stimulating effect which lime often exercises iii)on the growth of plants is 
 too well known to require illustration. The presence of calcium salts in consid- 
 erable quantity leads to a more vigorous production of root hairs than is normally 
 the case, as can easily be demonstrated by culture experiments, in which only the 
 tip of the root is immersed in the calcium salt solution. On the surface of the root 
 above the solution a great number of unusually long root hairs appear. To this 
 effect of the presence of lime, and the consequent readier absorption of potassium 
 and ammonium salts from the soil, Loew attributes in part the benefits obtained 
 by liming. {Bui. No. IS, Div. Veg. Phys. and Path., U. S, Department of Agricul- 
 ture, p. 43. ) That calcium salts directly stimulate growth, apart from the produc- 
 tion of root hairs, is. however, shown by cultures with the root entirely immersed 
 in an aqueous solution, thus precluding any important development of these organs. 
 
 ^ Schloesing's investigations [Comptesrendus, 74, 1553 (1872)] showed that 100,000 
 parts of pure water, i. e, , free from dissolved carbon dioxide, would dissolve about 
 1.3 parts of calcium carbonate. Treadwell and Renter [Zeit. fiir anorg. Chem., 
 35,38 (1900)] showed that by increasing the pressure of the carbon dioxide in 
 the gas phase in contact with the solution until it was one atmosphere, the solu- 
 bility was increased so that 100,000 parts of water would dissolve IIG parts of 
 calcium carbonate. Even at this extreme solubility there would be but 46 parts 
 of calcium per 100,000 of water, as against about 60 parts in a saturated solution of 
 calcium sulphate, in which plants thrive well. 
 
excess.' Wliy the coiTespondiiig calciviiii solution slioiild also hinder 
 growth can not be satisfactorily explained at present. 
 
 RESULTS WITH MIXED SOLUTIONS. 
 
 Upon comparing the limits of endurance for lupine roots in pure 
 solutions of th(i "alkali" salts with the limits detei-nuned by the 
 methods employ(Ml in a held survey, it became obvious that the for- 
 mer were vastly lower than the lattcM-; and that furthei-more the order 
 of toxicity of the several salts as tixed by investigators in the field 
 differed greatly from that obtained by expiuiments in th(^ laboratory. 
 This was strikingly the case with magnesium sulpliate, which is 
 decidedly the most toxic of the scn^^n salts when alon<^ in a pure 
 aqueous solution, but which is regarded as the h^ast- injurious by 
 students of alkali soils. Uut- it was recalled that iioiu^ oi" these salts 
 usually occurs in any notable <|uantity in tlie soil sa\'e in the [)i'es- 
 ence of one or several others, both of the readily solubki salts and of 
 the comparatively insoluble magnesium carbonate, calcium carbonate, 
 and (vilcium sulphate. 'I'lie key to the <liscrepancy appeared there- 
 fore^ to lie in mixtures of the various salts, and the study of these 
 became, logically, the lU'xt ste]) in the investigation. 
 
 In exj)erimenting with mix(Ml solutions it was i)laiined to test every 
 possi]>le combination of two of th<^ readily soluble salts with which 
 experinu'uts were made in pure solutions. Another line of experi- 
 ments, from which were obtained results which are believed to be of 
 considerable scientific interest and from aii economic point of view 
 to indicate one of the possible solutions of the alkali Soil problem, con- 
 sisted in combining each of the readily soluble salts with each of 
 three difficultly soluble ones — calcium sulphate, calcium carbonate, 
 and magnesium carbonate. The only triple mixtures so far tried are 
 ttiose of each readily soluble salt (except calcium chloride) with cal- 
 cium sidphate and calcium carbonate. Sodium bicarbonate was 
 tested only in this triple mixture. 
 
 Although the work with mixtures of salts is by no means com- 
 
 1 Treadwell and Reuter [Zeit. fiir anorg. Chem., 17, 199 (1898)] showed that at 
 15° C. and under a xiartial pressure of carljon dioxide in the vapor phase equal to 
 zero, pure water dissolves about Go jmrts of magnesium carbonate per 100,000. 
 With a partial pressure of carbon dioxide In the vapor phase equal to one atmos- 
 phere there was dissolved about 1,211 parts magnesium hydrogen carbonate, 
 equivalent to (598 parts of the normal carbonate per 100,000 of solution. It is thus 
 seen that the solubility is enormously increased by the presence of carbon dioxide. 
 Cameron and Briggs [Bui. 18, p. 32, Division of Soils, U. S. Department of Agricul- 
 ture (1900) J showed that a solution of magnesium carbonate in solution in equi- 
 librium with ordinary air contained about 18 parts of magnesium in 100,000 of 
 solution, which might have been expected to be enough to prohibit growth in view 
 of the toxicity of solutions of magnesium chloride and magnesium sulphate. It 
 should be further noted that it was shown that 6 parts of the dissolved magnesium 
 was combined as the normal carbonate, so that the solution contained more than 
 appreciable-amounts of OH ions, resulting from the hydrolysis of this latter salt. 
 
28 
 
 pleted, the data thus far obtained throAV so much light on the whole 
 subject of alkali soils, and go so far to account for the fact tliat the 
 limits of endurance of plants in pure solutions of the various salts 
 are low as compared with those determined from the observations 
 of -survey parties in the field, that it seems advisable to present 
 them here. 
 
 In the case of mixtures of two readily soluble salts, solutions of 
 each, of twice the desired concentration, were mixed in equal vol- 
 umes. Where one of the salts is a comparatively insoluble one, it 
 was added in solid form to a solution of definite concentration of the 
 soluble one, and tlie mixture was tlien diluted to the required con- 
 centration, as though the more soluble salt alone were present. (The 
 source of error incurred by this method was considered so slight as 
 to be practically negligible.) Tlie mixture was then allowed to 
 stand for a week or ten days with freqiu'ut shaking, in order to l)ring 
 it to equilibrium before using. In all mixtui-es of magnesium (car- 
 bonate and of calcium carbonate alone witli other salts, the undis- 
 solved residue was removed by filti-ation. Likewise in earlier experi- 
 ments witli calcium sulphate added to other salts, the residue was 
 removed, but in tliose upon which are based the limits given in tlie 
 tables it was retained. In all cases wliere l)otli calcium sulpliate and 
 calcium carbonate were added, the undissolved residue i-eniained 
 during the culture. The difference in limil due to the presence or 
 absence of a solid excess was, liowever, usually imperceptible, and 
 always slight. 
 
 In every case the object was to ascertain how far the limit of endur- 
 ance for the roots in the presence of the more toxic salt could be 
 raised b}' addition of one that is less injurious. Although the con- 
 centration of solution of the latter is invariabl}' stated, if it be a 
 readily soluble salt, it is the concentration of solution of the more 
 poisonous salt as denoting a corresponding limit of endurance ifo 
 which attention is chiefly directed. It is interesting that in cases 
 where both of the salts mixed are readily soluble ones the less toxic 
 salt a.ppears usually to be more effective in neutralizing the more 
 toxic one when added in concentration somewhat above rather than 
 below that in which plant roots will endure it when alone. Thus, in a 
 mixture containing equal volumes of 0.0075 normal sodium cai'bonate 
 and 0.01 normal sodium sulphate, roots of two plants survived, but all 
 died when the mixture contained 0.0075 normal sodium carbonate and 
 only 0.005 normal sodium sulphate. Also a majority of the roots 
 could retain their vitality in a mixture containing equal volumes 
 of 0.0025 normal magnesium sulphate and of 0.01 normal sodium 
 sulphate, but not in 0.0025 normal magnesium sulphate plus 0.005 
 normal sodium sulphate. Similar results were obtained by adding 
 sodium sulphate to magnesium chloride and sodium chloride to mag- 
 nesium sulphate. The reverse was true, however, in the mixtures of 
 magnesium chloride and sodium chloride, the less concentrated solu- 
 tion of the latter proving more beneficial. 
 
29 
 
 The coiiceiilvations are stated, as in preceding; tal)les, botli in parts 
 of salt to 100,000 of solution and in fractions of a normal solution. 
 
 In the following- tables of the effects of mixtures each of the more 
 soluble alkali salts (excepting sodium bicarbonate) is taken up in 
 succession in the order of its toxicity in pure solution. The neutral- 
 izing effect is expressed in terms of the greatest concentration of the 
 m<»i-e toxic salt endurable in the presence of the U'ss toxic one. As 
 the determination of tlie value of a less injurious salt in neutralizing 
 a more toxic one was the o1)jective of all expei'iments with mixtures, 
 it follows tluit the numbei- of added salts decreases successively from 
 table to table. For comparison, the limit of endui-ance for tlie more 
 toxic salt in pure solution is stated at the head of ^the table. The 
 details of neutralizing etfect upon each salt are taken up in connec- 
 tion with its res[)ective table, while a discussion of the general sig- 
 nificance of the whole series of experiments with mixtures of two 
 solutions is appended. 
 
 Tlie results embodied in Tables III to IX wcie obtained from experi- 
 ments with Lupin us alhus only. In Table X, however, the limits 
 are given for both Lupin us filhus and j\Ii'(lica(j(> saf/iva (alfalfa) in 
 solutions of each readily soluble salt (excepting calcium chloride) in 
 the presence of an excess of calcium sulphate and calcium carbonate 
 together. 
 
 MAGNESIUM SULPHATE IN MIXTURES. 
 
 The following table shows the results of exi)eriments with Lupinus 
 in solutions of magnesium sulphate with other salts added: 
 
 Table III. —Limits forin(tijiu'sinm snlpliufe in viixtitres. 
 
 Name of salt added. 
 
 Greatest endurable 
 coiicontration o f 
 magnesium s u 1 - 
 phate. 
 
 In f rac 
 
 tions of a 
 
 normal 
 
 solution. 
 
 None 
 
 Magnesium cbloride 
 
 Sodium carbonate 
 
 Sodium sulphate 
 
 Sodium chloride ._ 
 
 ( 'alciu m chloride 
 
 Magnesium carbonate 
 
 Calcium carbonate 
 
 Calcium sulphate 
 
 Calcium sulphate and calcium carbonate 
 
 . 00125 
 . 000625 
 . 00125 
 .00375 
 . 0075 
 
 '.01 
 .02 
 
 In parts 
 per 100,000 
 of solution. 
 
 3.5 
 
 21 
 
 42 
 120 
 
 56 
 112 
 360 
 240 
 
 Concentration of the 
 salts added. 
 
 In fractions 
 
 of a normal 
 
 solution. 
 
 0. 0025 
 
 .0025 
 
 .01 
 
 .015 
 
 2 
 
 Saturated. 
 
 Saturated. 
 
 Saturated. 
 
 Saturated. 
 
 In parts per 
 100,(KK) of 
 solution. 
 
 12 
 13 
 80 
 H7 
 1,1(11 
 
 Saturated. 
 
 Saturated. 
 
 Saturated, 
 
 Saturated. 
 
 In the light of figures given above, the enormous disci'epancy between 
 the results obtained by exi)eriments with this salt in [)ure solution and 
 the limit determined by field survey is completely obliterated. For 
 in alkali lands magnesium sulphate is rarelj^ if ever, found in any 
 quantity except in the presence of calcium sulphate; and it is com- 
 monly accompanied by both sodium and calcium sulphate (the Billings, 
 
80 
 
 Mont., type of alkali soiP). Addition of sodium sulphate, which is 
 itself so injurious in a pure solution, raises the limit for magnesium 
 sulphate three times, while the presence of calcium sulphate allows a 
 small proportion of the roots to barely survive during twenty-four 
 hours in a solution of magnesium sulphate 480 times as concentrated 
 as that which, in pure solution, represents the limit of endurance. A 
 careful comparison was made between 0.3 and 0.4 normal solutions of 
 magnesium sulphate, botli in the absence andtlu^ presence of an excess 
 of calcium sulphate, five individuals of Lupuuisalbus being cultivated 
 for 48 hours in each of the four solutions. The following table gives 
 the results: 
 
 Table IV.—^Iagnesium suljyhate with atidicithout calcium suljjhate. 
 
 Solutions. 
 
 Average 
 elongation 
 
 of the 
 marked por- 
 tion of the 
 root. 
 
 General condition of the roots. 
 
 
 Millinii'ters. 
 0.7 
 
 10.3 
 
 .3 
 13.0 
 
 Extremely flaccid, and discolored 
 with brownish blotches; extreme- 
 ly plasmolyzed. 
 
 Turgor normal; plasmolysis none; 
 but all roots quite badly discol- 
 ored. 
 
 Magnesium sulphate (0.3 normal) + calcium 
 sulphate. 
 
 Magnesium sulphate (0. 4 normal) 
 
 Magnesium sulphate (0.4 normal) + calcium 
 sulphate. 
 
 Turgor normal; plasmolysis none; 
 all but one root quite badly dis- 
 colored. 
 
 In both pure solutions the protoplasm of the nearly isodiametric 
 cells of the periblem was completely withdrawn from the cell wall and 
 collected with the nucleus in a compact mass near the center of the cell; 
 while in both solutions to which calcium sulphate had been added 
 no trace of plasmolyzing action could be detected in the cells of the 
 periblem, the protoplasm being closely applied to the wall, with large 
 vacuoles in the older cells, and the nucleus usually peripheral. Pre- 
 cautions were taken while preparing the sections to keep the tissues 
 immersed in the culture solution, and the absence of plasmolysis in 
 the roots taken from the solutions containing calcium suljihate is 
 sufficient evidence that the pure solutions had produced this effect 
 during the period of culture rather tlum after withdrawal.^ 
 
 'See Whitney and Means, Bui. 14, Div. Soils, U. S. Department of Atjcriculture 
 (1898), and Cameron, Bui. IT, p. 32, Div. Soils, U. S. Dei)artmeiit of Agriculture 
 (1901). 
 
 '^Wolf's observation (see footnote, p. 40) that hoth Ca (NO.,)., and Mg (NO,,).^ 
 are readily absorbed by plant roots when mixed together, while neither is readily 
 absorbed from a pure sohition, renders it highlj' probable that in this case of a 
 mixture of MgSO^ and CaSO^ it is the rapid endosmosis of the salts into the cells 
 of the iilant roots which prevents plasmolysis of the latter. In short this mixture 
 is to be compared with those substances described by Overton [Vierteljahrsschr. 
 Naturf. Gessells ch. Zurich 40, 1 (1895)] which produce only transient plasmolysis, 
 owing to their more or less rapid passage through the ectoplasm into the cell sap. 
 As determined by De Vries [Jahrb. fhr wiss. Butanik. 14, .537 (1884)], a 1.8 per 
 cent solution of magnesium sulphate (which would correspond to our 0.3 normal 
 
31 
 
 In the presence of both calcium sulphate and calcium carbonate 
 added in excess to a solution of magnesium sulphate, the limit of 
 endurance is only two-thirds as high as when calcium sulphate alone 
 is added. 
 
 Calcium as the chloride lias also a powerful effect in neutralizing the 
 toxicity of magnesium sulphate. But here the addition of a new 
 anion (CI), besides the added cathion (Ca), seems to diminish the 
 beneficial effect of the latter, since the chloride, although a readily 
 soluble salt, raises the limit for magnesium sulphate only one-third as 
 much as does the little- soluble calcium sulphate. In a mixture of 
 calcium chloride and iiuignesium sulphate a crystalline precipitate of 
 calcium sulphate separates slowly or rajiidly in proportion to the con- 
 centration of the solutions, so that the case becomes that of the 
 contact of solid calcium sulphate with a solution of magnesium chlo- 
 ride. As would be expected, the limit of endurance for magnesium 
 sulphate plus calcium chloride is the same as that for magnesium 
 chloride plus calcium sulphate (Table V). 
 
 Sodium salts are very much less effective in jieutralizing magnesium 
 sulphate than are salts of calcium. In the case of sodium salts it is 
 the chloride which is most effective, so that here we seem to have a 
 beneficial effect of the anion as well as of the cathion. Yet the absence 
 of any neutralizing effect when magnesium chloride is added to mag- 
 nesium sulphate shows thai the CI ions alone are ineffective. 
 
 In one case the addition of a salt with a common basic ion — i. e., 
 magnesium carbonate — raises the limit of endurance in magnesium 
 sulphate eiglit times.' By a simple process of elimination, since 
 magnesium ions are ineffective in the form of magnesium chloride 
 when added to magnesium sulphate, although chlorine ions appear 
 to have in themselves some neutralizing value when added as sodium 
 chloride, we are comi^elled to attribute the beneficial influence of 
 magnesium carbonate to COg, or more probably HCO.5, ions, a point to 
 which we will return in discussing the stimulating effect of dilute 
 solutions of sodium carbonate and sodium bicai-])()nate. Noteworth}^ 
 is the fact that calcium carbonate, although much less soluble than 
 the corresponding salt of magnesium, is twice as effective an anti- 
 dote for magnesium snlphale. This affords another striking proof of 
 the great efficacy of calcium as a remedy for magnesium poisoning. 
 
 solution), is the isotonic equivalent of a 0.1 normal solution of porassiom nitrate, 
 which is usually taken as the unit in measurements of osniotic pressure of solu- 
 tions. True [Bot. Gazette, 26, p. 410 (189G)] found that plasmolysis of Spirogyral 
 cells in a KNO., solution first appeared at a. concentration of 0.25 normal. De 
 Vries' results seem to indicate that the osmotic value of each component in a 
 mixed solution (of two or three salts) is equal to that of the respective compo- 
 nent when present alone at th-e given concentration, a point not in accord with 
 well-established facts. 
 
 'The roots barely survive in poor condition in a 0.01 normal magnesium sul- 
 phate solution plus an excess of magnesium carbonate: but in 0.005 normal magne- 
 sium sulphate solution plus magnesium carbonate some of the roots were perfectly 
 normal after a twenty-four hours' culture. 
 
32 
 
 MAGNESIUM CHLORIDE IN MIXTURES. 
 
 The results of experiments with magnesium chloride in mixtures 
 with other salts are shown in the following table: 
 
 Table V. — Limits foi' magnesium chloHde in mixtures. 
 
 Greatest endurable 
 cone entration of 
 magnesium chloride. 
 
 Name of salt added. 
 
 None 
 
 Sodium carbonate 
 
 Sodium sulphate- 
 
 Sodium chloride. 
 
 Calcium chloride 
 
 Magnesium carbonate 
 
 Calcium carbonate 
 
 Calcium sulphate 
 
 Calcium sulphate and calcium carbonate 
 
 In frac- 
 tions of a 
 normal 
 solution. 
 
 In parts 
 per 100,000 
 of solu- 
 tion. 
 
 0.0025 
 .0025 
 .01 
 .01 
 .1 
 
 .0025 
 .04 
 .2 
 .2 
 
 12 
 12 
 
 48 
 48 
 480 
 12 
 192 
 960 
 960 
 
 Concentration of the 
 salts added. 
 
 In fractions 
 
 of a normal 
 
 solution. 
 
 0.00375 
 
 .01 
 
 .02 
 
 .15 
 Saturated. 
 Saturated. 
 Saturated. 
 Saturated. 
 
 In parts per 
 100,000 of 
 solution. 
 
 19.5 
 
 80 
 116 
 726 
 
 Saturated. 
 Saturated. 
 Saturated. 
 Saturated. 
 
 In the alkali soils of the Western United States magnesium chloride 
 rarely occurs in such large quantities as to be regarded as more than 
 secondaiy in importance;^ 
 
 As in the case of magnesium sulphate, calcium is found to be much 
 more effective than sodium in neutralizing magnesium, but here cal- 
 cium chloride is relatively more effective than with magnesium sul- 
 phate, raising the limit for magnesium chloride one-half (instead of 
 only one-third) as far as does calcium sulphate. But calcium is much 
 less effective with the chloride than with the sulphate of magnesium, 
 as is evident from the relative efficacy of calcium sulphate in raising 
 the limits of endurance of the two magnesium salts. Hence we have 
 here another indication (as in the case of calcium chloride added to 
 magnesium sulphate) that chlorine ions by their ijresence lower the 
 neutralizing efficacy of calcium; although in the absence of the latter 
 base, magnesium chloride is only one-half as toxic as is magnesium 
 sulphate. While the beneficial effect of calcium sulphate upon mag- 
 nesium sulphate is decreased bj^ the addition of an excess of calcium 
 carbonate, the presence of the carbonate does not affect the value of 
 calcium sulphate as an antidote to magnesium chloride. 
 
 While calcium carbonate is equally effective in raising the limits 
 of the two soluble magnesium salts (sixteen times), magnesium car- 
 bonate, which raised the limit of magnesium sulphate eight times, 
 
 ' Magnesium rarely makes its specific effects upon plant life felt in the " alkali " 
 soils, owing to the omnipresence there of considerable calcium salts. In certain 
 areas of the Eastern States, notably in the so-called " serpentine barrens" of Penn- 
 sylvania and Maryland, it appears to be relatively more important, probably 
 because it is there present in excess over calcium, although the actual amount of 
 both, which may be present in the soil solutions at any given time, must be 
 extremely small. 
 
83 
 
 has no effect upon magnesium cliloride. As it appears 1,o be neces- 
 sary to regard the IICO., ions as the effective element in the former 
 combination, we must conclude that tliese act beneficially in the 
 presence of Mg and SOi, but are powerless in tlie presence of Mg and 
 CI. But as calcium carbonate is equally effective as an antidote to 
 magnesium chloride and to magnesium sulphate, it would follow that 
 the power of CI ions to hinder the effect of IICO.. ions disappears in 
 the presence of Ca ions, while, as already noted, CI -ions appear to 
 diminish the value of Ca ions as an agency foi' counteracting Mg ions. 
 Comparisons such as these show how dilficult it is to attempt an 
 interpretation of toxicological i)henomena in the liglit of current 
 chemical and pliysiological ideas. Possil)ly d(^terminations of the 
 solubility and degree of dissociation of thesis dilfercMit salts in mix- 
 tures may afford somc^ clue to the numerous anonuilies. On tlu^ othei' 
 hand, it is diftieult to see any justification for using th(^ reactions of 
 organisms in determining the dissociation constants of electrolytes. 
 The many nonconcordant results recently described in the literature 
 can hardly be regarded as throwing discredit upon the dissociation 
 hypothesis, but rather as demonstrating the- unsatisfactory nature of 
 the method employed for the investigation in hand. 
 
 Sodium sulphate and sodium chloride are eciuall}' elfective in rais- 
 ing the limit of magnesium chloride (four times). Tlu^ former is more 
 effective, and the latter decidedly less so, tlian in the case of magne- 
 sium sulphate, sotliat the anions (CI, SO,) and not alone the cathious 
 (Mg, Na) appear to make their influence felt in these cases. 
 
 SODIUM CARBONATE IN MIXTURES. 
 
 Table VI shows results of experiments with mixtures of other salts 
 with sodium carbonate: 
 
 Table VI. — Limits for sodimn citrboiutte in iiii.vtiires. 
 
 Name of salt added. 
 
 Greatest endurable 
 concentration of 
 sodium carbonate. 
 
 Concentration of the 
 salts added. 
 
 In frac- 
 tions of a 
 
 normal 
 solution. 
 
 In parts ' In fractions 
 perI(IO,(KIO of a normal 
 solution. solution. 
 
 In ]iarts per 
 1110,(101) of 
 solution. 
 
 None 
 
 Sodium sulphate 
 
 Sodium chloride 
 
 Calcium chloride 
 
 Magnesium carbonate 
 
 Calcium carbonate 
 
 Calcium sulphate _ _ 
 
 Calcium sulphate and calcium carbonate 
 
 0.005 
 .007.5 
 .0025 
 .25 
 .01 
 .0075 
 .03 
 .03 
 
 36 
 39 
 13 
 1,3(X) 
 53 
 39 
 156 
 1.56 
 
 0.01 
 .01 
 
 Saturated. 
 Saturated. 
 Saturated. 
 Saturated. 
 
 80 
 
 58. 
 
 1.377 
 
 Saturated. 
 
 Satuiated. 
 
 Saturated. 
 
 Saturated. 
 
 As the above table shows, sodium chloride is ineffective as an anti- 
 dote to sodium carlionate; calcium carbonate barely raises the limit 
 
 8287— No. 71—02 3 
 
34 
 
 of endurance, while magnesium carbonate merely doubles it. It is 
 interesting tliat magnesium carbonate should liere be more effective 
 than the corresponding salt of calcium, since in all other cases the 
 latter is the more beneficial.^ Sodium sulphate is likewise of very 
 littl<j neutralizing value, and the st)luble salts of magnesium possess 
 none so far as was ascertained. Calcium sulpliate raises tlie limit 
 only six times, the presence or absence of an excess of calcium car- 
 bonate not affecting tlie value of the sulphate. This comparatively 
 slight efficacy of calcium sulphate in neutralizing "black alkali" is 
 rather sui-prising in view of the accepted ideas of students of alkali 
 soils in regard to the curative value of gypsum.- The comparative 
 inefificacy of calcium sulphate in this case contrasts strikingly with 
 its power to neutralize sodium in the forms of sulphate (Table VII) 
 and chloride (Table VIII). 
 
 Calcium chloride is the only salt found to be very effective in 
 neutralizing sodium carbonate, raising the limit of endurance for the 
 latter fifty times. A mixture of solutions of the two salts causes an 
 immediate lieavy precipitate of calcium carbonate, to which fact the 
 efficacy of the added salt must be largely ascribed. We should be 
 dealing in this ease with a solution of sodium chloride containing a 
 large excess of calcium carbonate. Yet by direct addition of solid 
 calcium carbonate to a solution of sodium chloride, the limit of endur- 
 ance for the latter can be raised only three times, i. e., to 0.0(3 normal 
 (see Table VIII). Here again chemistry appeal's to be i)owerless to 
 afford an explanation of a phenomenon which, in the j)resent state 
 of our knowledge, must be regarded as paradoxical.^ 
 
 A very noteworthy result was obtained by experiments with sodium 
 carbonate, as well as with sodium bicarbonate, in the presence of an 
 excess of calcium sulphate and calcium carbonate. In solutions of 
 critical concentration of both of these mixtures a majority of the 
 roots of the lupine plants were completely destroyed with pronounced 
 
 ' The probability of the formation of a double carbonate, with a consequent 
 lowering of the active mass, as well as a probable change of nature of the ions, 
 suggests itself very forcibly in this connection. It is hoped that time and oppor- 
 tunity will be found in the near future to test this supposition in the laboratory. 
 
 -' Hilgard. Bui. 128, Agr. Exp. Sta.. Univ. Calif., pp. 16 to 18 (1900). It should be 
 stated, however, that Hilgard recommends the application of gypsum under phys- 
 ical conditions which would not proliahly be considered analogous to those under 
 which the experiments here described were performed. 
 
 ^In marked contrast with this anomalous case is that of the mixture of magne- 
 sium sulphate and calcium chloride, in which a precipitate of calcium sulphate is 
 formed and which is therefore to be regarded as a solution of magnesium chloride 
 containnig a solid excess of calcium sulphate. Here the limit of endurance is the 
 same as when solid calcium sulphate is added directly to a solution of magnesium 
 chloride. The same thing is true of a mixture of sodium sulphate and calcium 
 chloride in which the limit of endurance is the same as for sodium chloride plus 
 calcium sulphate. 
 
35 
 
 corrosion due presumably to the action of hydroxy! ions, while a 
 smaller number survived in apparently perfect condition, and this 
 occurred in i-epeated experiments. In critical solutions of other salts 
 and mixtures of salts, however, there was rarely such sharp differ- 
 ence in appearance between roots which survived and those which 
 died during- twenty-four hours' culture; as a rule none of the roots 
 were in normal condition at the end of the experiment. In other 
 words, the difference of individual plants in their powei" to resist toxic 
 action appears to be more pronounced in the case of the two carbonates 
 of sodium than of other "alkali salts." This would indicate that the 
 selection of plants foi- resistance to "black alkali" olfers a simpler 
 problem than where i-esistance to other components of alkali soils is 
 to be soui^ht. 
 
 SODUIM SULPHATE IN MIXTURES. 
 
 Experiments with sodium suljihate in mixtures with other salts 
 show the following results: 
 
 Table VII. — Limits for sodium sulphate in mixtures. 
 
 Name of salt added. 
 
 None 
 
 Sodium chloride 
 
 Calcium chloride 
 
 Magnesium (carbonate 
 
 Calcium carbonate 
 
 Calcium sulphate 
 
 Calcium sulphate and calcium carbonate a 
 
 Greatest endurable 
 concentration of 
 sodium sulphate. 
 
 In frac- 
 tions of a 
 
 normal 
 solution. 
 
 0.(1075 
 .00375 
 2 
 
 !03 
 .04 
 .5 
 .3 
 
 In parts 
 perl00,(KK) 
 ofsolution. 
 
 53 
 
 26.5 
 
 1,372 
 
 212 
 
 281 
 
 3,5;i0 
 
 1,908 
 
 Concentration of the 
 salts added. 
 
 In fractions 
 
 of a normal 
 
 solution. 
 
 0.01 
 .2 
 Saturated. 
 Saturated. 
 Saturated. 
 Saturated. 
 
 In parts per 
 100,000 of 
 solution. 
 
 58 
 1,103 
 Saturated. 
 Saturated. 
 Saturated. 
 Saturated. 
 
 a See Bui. 17., p. 2'i et soq.. Division of Soils. U. S. Department of Agriculture, 1901. 
 
 Sodium sulphate is very abundant in "alkali" soils, often occurring 
 in contact Avith each or several of the other salts. In the Billings, 
 Mont., type, for example, it is accompanied by the sulphates of mag- 
 nesium and calcium, while in the Fresno, Cal., type it is in contact with 
 sodium carbonate. 
 
 Most effective for neutralizing this salt is calcium sulphate, which 
 raises the limit more than sixty times when added alone. In the 
 presence of an excess of calcium carbonate, however, calcium sul- 
 phate can increase the limit of endurance for sodium sulphate only 
 about forty times. This is probablj^ due to a forcing back of the dis- 
 sociation and decrease of solubility of the calcium sulphate by the 
 calcium carbonate, although, as Cameron and Seidell' have shown, 
 either salt is rather soluble in dilute solutions of sodium sulphate. 
 
 ' Solution Studies of Salts Occurring in Alkali Soils. Bui. 18, Division of Soils, 
 U. S. Department of Agriculture. 
 
36 
 
 As in the case of all tlie other salts excei)t sodium carbonate, cal 
 cinm chloride is less effective than calcium sulphate, raisinij: the limit 
 only twenty-seven times. In this case, as in the mixture of calcium 
 chloride with magnesium sulphate, a crystalline precipitate of calcium 
 sulphate is formed, so that it actually becomes a case of the contact 
 of calcium suli)hate with a solution of sodium chloride, and the same 
 limit was obtained by a direct test of this latter mixture. Calcium 
 carbonate is much less effective in neutralizing sodium sulphate than 
 in counteracting the toxic action of magnesium salts. 
 
 Calcium sulphate is more effective as an antidote to sodium sulphate 
 than to any other salt tried except magnesium sulphate. In both 
 these cases the anion of the added salt is the same as that of the more 
 toxic one; hence the cathions alone seem to operate. Possibly a dou- 
 ble salt of sodium and calcium is formed in this mixture. Since cal- 
 cium sulphate is much less efificacious in neutralizing the chlorides of 
 magnesium and sodium than the corresponding sulphates, wiiile it is 
 generally more beneficial than is calcium chloride, it seems almost 
 certain that tin; beneficial action of Ca ions is in some way hindered 
 by the presence of CI ions. That CI ions are in themselves less toxic 
 than are SO^ ions would appear from the fact that the chlorides of 
 magnesium and of sodium are less injurious in pure solution than are 
 the corresponding sulphates. 
 
 SODIUM CHLORIDE IN MIXTURES. 
 
 Experiments with sodium chloride in mixtures with other salts 
 3'ielded the results shown in the following table: 
 
 Table Vill. — Limits for sodium chluride in mixtures. 
 
 
 Greatest endurable 
 cone entration of 
 sodium chloride. 
 
 Concentration of the 
 salts added. 
 
 Name of salt added. 
 
 In frac- 
 tions of a 
 
 normal 
 solution. 
 
 In parts 
 per KKt.OUO 
 of solution. 
 
 In fractions 
 
 of a normal 
 
 solution. 
 
 In parts per 
 100,0(X) of 
 solution. 
 
 
 0. 02 
 
 !04 
 .06 
 
 .2 
 
 116 
 
 1.160 
 
 233 
 
 848 
 
 
 
 
 0.2 
 Saturated. 
 Saturated. 
 
 1,101 
 
 
 Saturated. 
 
 
 Saturated. 
 
 
 1.160 
 1,160 
 
 Saturated. 
 Saturated. 
 
 Saturated. 
 
 Caicium sulphate and calcium carbonate 
 
 Saturated. 
 
 Sodium chloride is probably the most widely distributed and gen- 
 erally abundant of the soluble components of alkali soils, occurring 
 practically wherever the land is notably impregnated with these 
 noxious salts.' As the above table shows, calcium sulphate and cal- 
 cium chloride are equall}^ effective in neutralizing the toxicity of 
 sodium chloride, although in the case of sodium sulphate and the 
 
 ' A notable exception is the Billings area in Montana. 
 
87 
 
 soluble salts of inagnesiuin, the former is decidedly more beneficial 
 than the latter. 
 
 As in the case of magnesium chloride, the presence of calcium 
 carbonate does not affect the neutralizing value of calcium sulphate, 
 although decidedly diminishing it in the case of magnesium sulphate 
 and sodium sulphate. It has been shown by Cameron and Seidell' 
 that an excess of solid calcium carbonate has Imt a very slight effect 
 on the solul)ility of calcium sulphate in sodium chloride solutions at 
 the concentrations here involved. And from the general resemblance 
 between the phenomena presented by calcium sulphate in contact 
 with sodium chloi'ide and magnesium chloride solutions, it is proba- 
 ble that calcium carbonate would have a like effect in the latter cases. 
 It is to be regretted that laboratory investigations of the solubility of 
 solid calcium cai'bonate and calcium sulphate in contact with solu- 
 tions of soluble sulphates have not yet been made. 
 
 CALCIUM CHLORIDE IX MIXTURES. 
 
 In the experiments with calcium chloride in mixtures with other 
 salts the following i-esults were obtained: 
 
 Table IX. — Liviits fur calcium cldoride in mixtures. 
 
 Name of salt addfil. 
 
 None 
 
 Magnesium carbonate. 
 
 Calcium carbonate 
 
 Calcium sulphate 
 
 Greatest endurable 
 concentration of 
 calcium chloride. 
 
 In frac- 
 tions of a 
 normal 
 Solution. 
 
 In parts 
 per 1(X),0(H) 
 of solution. 
 
 .2,5 
 
 1.377 
 1,:{77 
 1,:577 
 1,652 
 
 Concentration of the 
 salts added. 
 
 In fractions 
 
 of a normal 
 
 solution. 
 
 In parts 
 per 100,000 
 of solution. 
 
 Saturated. Saturated. 
 Saturated. Saturated. 
 Saturated. Saturated. 
 
 (I About. 
 
 Calcium chloride is quite generally distributed in alkali soils, being 
 usually present in small patches wherever sodium chloride is abun- 
 dant. As wouhl be expected from the relatively very high concentra- 
 tion of the pure solution of this salt endured by roots of the white 
 lui)ine, the limit can not be materially raised by the addition of other 
 salts. 
 
 SODIUM BICARBONATE IN MIXTURES. 
 
 Sodium 1)icarbonate was tested in mixture only with calcium sul- 
 ])hate and calcium carbonate together, which raised its limit of endur- 
 ance two and one-half times (see Table X). It usually occurs in 
 nature in contact Avith the normal sodium carbonate. 
 
 CALCIUM sulphate AND CALCIUM CARBONATE IN MIXTURES. 
 
 In the tables of limits of endurance in mixtures, as in those of pure 
 solutions, the figures do not perfectly state the case. For example, 
 
 Bui. 18, Division of Soils, IT. S. Department of AiiTlcnltiire (l'.)Ol). 
 
38 
 
 although the limit of endurance in sodium carbonate plus magnesium 
 carbonate is 0.01 normal, while for sodium suli)liate plus magnesium 
 carbonate it is 0.03 normal, the proportion of individual seedlings 
 whose roots survive in good condition is decidedly greater in the latter 
 mixture than in the former. 
 
 None of the readily soluble salts occurs abundantly in alkali soils 
 save in the presence of calcium sulpliate or calcium carbonate, and 
 most often in contact witli both. Hence it follows that the limits of 
 endurance for the more soluble salts in the presence of these two salts 
 of calcium, as obtained by means of water-culture experiments, 
 should agree closely witli the limits determined })y soil investigators. 
 This proved to be the case, due allowance being made for the influ- 
 ence of the pliysical pro[>erties of a soil as compared witli an ai^ueous 
 solution. 
 
 The following ta1)le sei'ves to bring together, for ready comparison, 
 the limit of endurance for roots of both white lupine and alfalfa in 
 solutions of six of the easily soluble salts to which a solid excess of 
 both calcium sulphate and calcium carlionate was added, the mixtures 
 being l)rought to equilil)rium before using. 
 
 Table X. — Results with mi.vtures containing two calciiivi salts. 
 
 Name of salt. 
 
 Limits for lupine 
 (Lupinus albus). 
 
 Limits for alfalfa 
 (Medicago sativa).rt 
 
 Parts per 
 
 100, 000 of 
 solution. 
 
 Magnesium sulphate 
 Magnesium chloride- . 
 
 Sodium carbonate 
 
 Sodium sulphate. 
 
 Sodium chloride 
 
 Sodium bicarbtniate . . 
 
 2,240 
 
 960 
 
 156 
 
 2,160 
 
 1, 160 
 
 417 
 
 Normal 
 solution. 
 
 Parts per 
 100, 000 of 
 solution. 
 
 0.4 
 
 .03 
 .3 
 
 1,960 
 
 960 
 
 104 
 
 2,160 
 
 1,160 
 
 667 
 
 Normal 
 solution. 
 
 0.35 
 .2 
 .02 
 .3 
 
 a In the case of alfalfa a few roots barely survive in O.f 5 and in 0.3 normal magnesium sulphate, 
 while in 0.35 normal they make a noteworthy amount of growth during forty-eight hours. In 
 0.2 normal sodium sulphate they make a decidedly better growth, and in 0.1 normal sodium 
 chloride two and one-half times as much growth as in the water control. 
 
 Th«^ close correspondence between the white lupine and alfalfa in 
 their resistance to the effects of these mixed solutions is worthy of 
 note, especially as alfalfa appears to be more sensitive than the lupine 
 to pure solutions. ''J'he only sei-ious discrepancy occurs in the mix- 
 ture of sodium bicarl)onate, calcium sulphate, and calcium carbonate, 
 to which alfalfa roots appear to be nearl}' twice as resistant as are 
 those of white lupine. 
 
 That in the neutralizing effect wpon more toxic salts which these 
 two relatively insoluble salts exert calcium sulphate plays a much 
 more important part than does calcium carbonate is obvious from a 
 comparison of the limits of endurance in solutions to which either 
 calcium sulphate or calcium carbonate alone has been added. Indeed, 
 in the case of magnesium sulphate and of sodium sulphate the limit 
 of endurance is decidedly lower in the presence of both calcium salts 
 
89 
 
 than when caleiiun sulphate alone is added. On the oilier liand, the 
 presence or absence of calcium carbonate appears to liave no effect 
 upon the neutralizing value of calcium sulphate when added to mag- 
 nesium chloride, sodium chloride, or sodium carbonate. 
 
 An interesting comparison is that of the soluble salts, one with 
 another, in respect to their degree of toxicity in pure solution on the 
 one hand, and in the presence of an excess of calcium sulphate and 
 calcium carbonate on the other. It will be observed that the sequence 
 in the first column is verj'- different from tliat in the second. The 
 most toxic salt or mixture is placed at the head of each column. 
 
 Table XI. — Order of toxicity irith (uid iritJionf ciilriion salts. 
 
 In imre solution: 
 
 Magrnesium sulphate. 
 Magnesium fhloride. 
 Sodium carbonate. 
 Sodium sulphiite. 
 Sodium chloride. 
 Sodium bicarbonate. 
 Calcium chloride. 
 
 In jiresence of an excess of C<aS04 and CaCOa: a 
 S(^dium carbonate. • 
 
 Sodium l)ii-arbi>nate. 
 Matjuesium chloride.?) 
 Sodium chloride b 
 Calcium chloride. 
 Sodium sulphate. 
 Magnesium sulphate. 
 
 «It has already been suggested that the limit in some of these highly concentrated solutions 
 containing an e.xcess of calcium salts may bear some relation to the osmotic pi-essure of the 
 solutit)n. It is therefore not a mere coincidence, perhaps, that the sequen "e in this column is 
 almost identical with that in Table II (of concentrations ])i'ecluding any growth during the 
 culture). 
 
 6 These two salts are equally toxic in mixtures if reacting weights be compared, while magne- 
 sium chloride is the more toxic of the two in parts of salt per 1(KI,(KKI of solution. 
 
 The interest and importance of the results obtained from the exper- 
 iments made with mixed solutions show the great desirability of 
 extending further this line of investigation. In fact, no aspect of the 
 work promises more substantial returns. An interesting problem 
 among the mam' which suggest themselves in this connection is that 
 of a possible relation between the degree of toxicity of a salt, alone or 
 in mixture, and the i*eadiness with which it is taken up by the plant 
 from a solution. The occasion seems opportune to redirect attention 
 to a series of experiments made long ago by Wilhelm W(df ^ as to the 
 
 'Landw. Versuchsst., 7, 198 (1865). The studies were made with a series of 
 solutions, each of which contained two salts in equal amount. Combinations 
 were made with (1) ?alts of the same acid, but of different bases; (2) salts of 
 the same base, but of different acids; (3) with both base and acid different. In 
 each culture 200 c. c. of solution was employed, and after one-half of this volume 
 had l)een absorbed by the plant (allowance being made for the small quantity of 
 water evaporated directly from the solution) the amount of each salt taken up 
 with the water was estimated by analy.sis of. the residual 100 c. c. of solution. 
 Young beans and maize were used in the experiments. Some of the results 
 obtained were as follows: 
 
 From three ini.\'tures, each containing O.Ol grams of each two salts, the plants 
 absorbed in percentages of the original (quantity of each salt supplied: 
 
 From ammonium nitrate plus calcium nitrate. 92 per cent of the former and 9t 
 per cent of the latter. 
 
 From ammonium nitrate plus magnesium nitrate, 92 per cent of the former and 
 86 per cent of the latter. 
 
 From magnesium nitrate plus calcium nitrate, 74 per cent of each. 
 
 Potassium nitrate was taken up from all combinations with other nitrates 
 
40 
 
 amount of each salt absorlied l)y a plant from a mixed sohition. 
 Especially interesting, as compared with the toxicological phenomena 
 of pure and mixed solutions, respectively, are Wolfs results as to 
 the effect of calcium sulphate in stimulating the absorption of other 
 sulphates. 
 
 GENERAL SIGNIFICANCE OF RESULTS WITH MIXED SOLUTIONS. 
 
 To enter into a discussion, from the purely chemical point of view, 
 of the widely accepted hypothesis of the dissociation of electrolytes 
 in solution would be to exceed the proper limits of this paper.' It 
 is sufficient to say that salts such as those with which we are here 
 dealing are held to dissociate in dilute solutions, more or less com- 
 
 (Na, NH^, Mg, Ca) in absolutely greater amount than from a simple solution. 
 From a solution containing 0.0'25 gram eacli of potassium nitrate and calcium 
 nitrate the plants absorbed 100 per cent of the former and 88 per cent of the latter. 
 From an equivalent solution of potassium nitrate plus magnesium nitrate, 100 per 
 cent of the former and 88 per cent of the latter. The stimulation of the plant by 
 the presence of calcium to take up greater quantities of potash is referred by 
 Loew (1. c.,*p. 44) to the increased development of root hairs induced by thecalcium. 
 But if the presence of magnesium has exactly the same effect, as would appear 
 from the experiment just quoted, we must look further for an explanation. Absorp- 
 tion of ammonium nitrate is decreased by the presence of other nitrates, irhile 
 that of calcuun and of magneaiinii nitrates is stiniiilated tlterebij. It is remarkable 
 that while neither of these last tico salts is readily absorbed from a, simple solution, 
 both are easily absorbed when mixed together. 
 
 Plants could be grown in mixtures of potassium and calcium sulphate 
 (KjSO^ + CaSOj and of calcium and magnesium sulphate (CaSO^ + MgSOJ, but 
 never in mixtures of sulphates of potassium and sodium ( K^SO^ + Na^.SO J , of 
 potassium andammonium (K.^SO^ + (NHJ^ SOJ, nor of potassium and magnesium 
 KjSO^ + MgSO^), even when the solutions were very dilute. 
 
 Potassium and ammonium salts were taken up much more readily in the pres- 
 ence of a calcium salt than from a pure solution. This was notably the case with 
 the sulphates, which are absorbed with difficult}' from unmixed solutions, (xypsum 
 (calcium sulphate) is absorbed in very small (luantity in the presence of a potas- 
 sium salt, but greatly stimulates absorption of the latter. From a mixture of 
 calcium and magnesium sulphate little of either salt is taken up, but the presence 
 of magnesium nitrate considerably increases the amount of calcium sulphate 
 drawn from a solution. 
 
 From mixtures of a sulphate and a phosphate, the latter is always taken up in 
 greater quantity. Ma.;nesium sulphate is taken up in greater quantity in the 
 presence of a phosphate than are other sulphates. 
 
 De Saussure's principle of the absorption of salts in solution by plant roots— 
 i. e., that the salt is taken up in smaller proportion to the water absorbed than it 
 occurs in the culture solution; in other words, that the residual solution becomes 
 more concentrated — applies to the absorption of sodium chloride in the presence 
 of a nitrate (KNO3, NH^NOj, Ca(N(),)2), but does not hold as to the absorption 
 of the nitrate itself. 
 
 ' For the presentation of the subject in simple terms the reader is referred to 
 a former publication by one of us. (Rep. No. 64, U. S. Department of Agricul- 
 ture, p. 144, 1900.) 
 
41 
 
 pletely at a givoii concentration according to the specific properties of 
 the particular salt. The result is a lil)erati()n of ions — atoms or atomic 
 groups carrying or in som(^ way associated with an electric charge. 
 Cathions, tliose furnished by the basic radicle, carry positive elec- 
 tricity, while anions, derived from tlie acid radicle of the salt, are 
 negatively charged. Ions possess a much greater velocity ^ than do 
 undissociated molecules, and it is now believed by many pliysiologists 
 that salts owe to the pi-operties of their ions rather than to their entire 
 molecules the toxic and other action which they exert upon organisms.^ 
 
 It is believed that the results of the present investigation tend to 
 confirm this view, althoug]i it must be admitted that serious anomalies 
 exist, to some of Avhich attention has already been directed. 
 
 Pure solutions of the salts dealt with are shown to be generally 
 injui-ious to plants, and this largely by virtue of the catliions which 
 the}" yield, as a comparison of the position of tlie several salts in tlie 
 table of toxicity in pure solutions sliows conclusively. Tlius magne- 
 sium salts, irrespective of the character of their anions, are much 
 more injurious tlian is any sodium salt, while the three clilorides (of 
 magnesium, sodium, and calcium) dift'ei- enormously in toxicity, 
 regardless of the fact that tliey yield a common anion. 
 
 An inspection of the tables of limits in mixed solution given above 
 makes it clear that the addition of a second, less toxic salt in most 
 cases increases the concentration of solution of the more harmful one 
 ill which root tips can I'etain their vitality. It is also demonstrated 
 that addition of a second sal: of the same base, ]ien(;e furnishing a 
 different kind of anion only, is usually much less efficacious in raising 
 the limit than is the admixture of a salt of a different base. Thus 
 magnesium chloride is ineffective as an antidote to magnesium sul- 
 phate, sodium chlorid(^ to sodium carbonate or to sodium suliDhate, 
 and calcium carbonate to calcium chloride. 
 
 If the assumption be granted tliat in the dilutions here involved 
 the magnesium salts nve practically completely dissociated and that 
 the anions do not have a toxic effect, then a 0.00125 normal solution 
 of magnesium will be tlie limit wlien the metal is combined as the sul- 
 phate and a 0.0025 normal solution when combined as the chloride, but 
 about a 0.002 normal solution when both chloride and sulphate are pres- 
 ent, with two eipiivalents of the former to one of the latter. The same 
 line of reasoning holds for the other cases cited, and from these facts it 
 is evident that the anions have a part in determining the toxic effect of a 
 
 ' That the physiological action of ions may be in some sort a function of their 
 specific velocities is indicated by Loeb's comparison of the effects of hydrogen and 
 hydroxyl ions, as well as of various basic cathions, upon the absorption of water 
 by a muscle. [Pfliigers Archiv.. 69, 21. (1898).] 
 
 '^Of a rapidly growing literature on this subject the papers of Kahlenberg and 
 True and of Kahlenberg and Austin, dealing with plants, and those of Loeb, 
 Garrey. Anne Moore, Kahlenberg. Clark and others, treating ion action upon ani- 
 mals, may be cited as of great importance. (See the Bil)liography, p. 5G.) 
 
42 
 
 salt, although a niiich smaller one in general than have the cathlons. 
 Furthermore, these views are in harmony with Loeb's idea that SO4 
 ions are more toxic than CI ions, because they tend to precipitate 
 calcium from its proteid compounds. 
 
 In other cases, however, addition of a salt which furnishes new 
 anions, but not new cathions, to the mixture is effective in raising the 
 endurable limit of concentration for the more toxic salt. A striking 
 case is the elevation of the limit for magnesium sulphate eight times 
 by the addition of magnesium carbonate. Here it would appear that 
 the HCO3 anions alone can be the effective agency. Sodium snlphate 
 slightl}' raises the limit of sodium carbonate, and a relatively nnim- 
 portant increase of the concentration of a calcinm chloride, solution 
 in which lupine roots can survive is obtained by addition of calcium 
 sulphate. But in these last two cases the effect is so small as to be 
 almost negligible, and is perhaps entirely attributable to the forcing 
 back of the dissociation of the more toxic salt i-ather than to any 
 direct physiological action of the new anions. 
 
 The superior efficacy of cathions over anions in neutralizing the 
 toxic effect of other cathions is illustrated by the discovery that 
 sodium is eciually effective as an antidote to magnesium chloride, 
 whether it be added as sulphate (Na2S04) or as chloride (NaCl).^ A 
 much more striking illustration is afforded by the fact that calcium, 
 when added to a solution of magnesium sulphate or of sodium sul- 
 phate, is very much more efficacious when furnished as the relatively 
 insolulile sulphate than as the readily soluble chloride. In other 
 words, the presence of chlorine anions actually hinders the full exer- 
 tion of the physiological effect of calcium cathions, unless we are to 
 believe that the superior efficacy of calcium sulphate is due merely to 
 its greater influence in retarding the dissociation of the sulphates of 
 magnesium and of sodium. 
 
 If we turn now to the effect of mixtures in which two kinds of 
 cathions are present we find that these are almost invariably much 
 less i)oisonous than is the pure solution of tlie more toxic salt. Even 
 the addition of a sodium salt (sulphate or chloride) to one of magne- 
 sium (sulphate or chloride) raises the limit of endurance for the latter 
 three to six times. Still more remarkable is the effect of magnesium 
 carbonate as an antidote to salts of sodium (carbonate, sulphate, 
 chloride), raising their limits two to four times. But these effects are 
 trivial as compared with the extraordinary efficacy of calcium in 
 counteracting the toxic effects of other bases (magnesium, sodium). 
 
 Even when added as the but slightl}^ soluble carbonate, calcium 
 raises the limit of magnesium sulphate and of magnesium chloride 
 sixteen times, of sodium suljihate more than five times, and of sodium 
 
 ' On the other hand, sodium chloride is twice as effective as sodium sulphate in 
 neutralizing magnesium sulphate. 
 
43 
 
 chloride three times. Calcium chloride^ mixed with an equal volume 
 of a magnesium or a sodium salt raises the limit of the latter as fol- 
 lows: Magnesium sulphate, one luindred and sixty times; magnesium 
 chloride, forty times; sodium carbonate, fifty times;- sodium sulphate, 
 twenty-seven times, and sodium chloride, ten times. 
 
 The most effective of the calcium salts tried was, however, calcium 
 sulphate. This, when added alone in solid excess, increases the 
 maxima of concentration endurable by the roots as follows: Magne- 
 sium sulphate, four hundred and eighty times; magnesium chloride, 
 eighty times; sodium carbonate, six times; sodium sulphate, sixty-six 
 times, and sodium chloride, ten times. Here we have probably the 
 greatest effect of one kind of ion in neutralizing the effect of another 
 kind that has yet been obtained in experiments with plants. 
 
 It is noteworthy that the effect of the calcium ions upon different 
 salts having a common basic ion differs greatly. Thus plant roots 
 can endure three times the concentration of a solution containing 
 magnesium cathions and sulph-anions to wliich calcium sulphate is 
 added than tliey can of a solution containing magnesium cathions and 
 chlor-anions plus calcium sulphate. Yet tlie former solution in the 
 absence of calcium salts is endui'able in concentration only one-half as 
 great as is the latter without a calcium salt. Here the effect may be 
 partly due to differences of dissociation in the two solutions. But it 
 appears necessary to attribut<' the greater part of it to an adverse 
 influence, presumably exerted by chlorine ions, upon the physiological 
 action of calcium ions in the presence of magnesium ions. Similar 
 problems are suggested by the wide differences in the degree to which 
 calcium sulphate can neutralize the toxic action of each of the three 
 sodium salts. 
 
 That the iihenomena exhibited by the ]-oots of plants in their reac- 
 tion to these various mixed salt solutions are not to be regarded as 
 mere functions of chemical changes in the solution itself is patent. 
 The problem is undoubtedly a much more intricate one, involving 
 chemical reactions of great comj)lexity in the protoplasm of the plant 
 itself. In this connection it is important to call attention to the 
 strikingly similar results ol)tained by Loel) ^ as to the relative toxic 
 effect upon animals of jiure and of mixed solutions. 
 
 A pure salt solution, e. g., of sodium chloride, was found to be 
 
 ' Loew (Bui. No. 18. Div. Veg. Phys. and Path., U. S. Department of Agriciilture. 
 p. :5;3), referring to an experiment made by Boehm, appears to doubt the value of 
 calcium in the form of the chloride as a plant nutrient, owing to the formation 
 of hydrochloric acid in the assimilation of calcium by the plant. Here is another 
 suggestion as to the reason for the inferiority of calcium chloride to calcium sul- 
 phate in neutralizing the toxic action of salts of other bases. 
 
 ■As has already been noted (under Table VI). a heavy precipitate of calcium 
 carbonate is formed in this mixture, so that it becomes in great part a solution 
 of sodium chloride i)lus a solid excess of calcium carbonate. 
 
 "Seethe papers by this author cited in the Bibliography (p. 58). 
 
44 
 
 strongly poisonous to marine animals in varions stages of develop- 
 ment — i. e., ai\H]i{F(indulushete7'0clitns), a Jellylish(Gonionemus sp.), 
 and a sea urchin (Arbacia sp.). But the addition in small quantity 
 of a salt yielding another kind of cathion, such as magnesium, potas- 
 sium, and calcium, more or less neutralized this toxic effect, although 
 each of these salts was itself toxic in pure solution. As in the case of 
 plants, calcium was particularly^ effective. That it is the cathions 
 rather than the anions added to the solution which are chiefly effective 
 as counter agents is evident from the fact that of each base the chloride 
 only was used. Moreover, in only one mixture of three chlorides 
 could fertilized eggs of the sea urchin be brought to an advanced 
 stage of development, but sodium bromide could be successfully sub- 
 stituted for sodium chloride in the mixture.^ It is clear, therefore, 
 that the anions play a very subordinate part in the physiological 
 action of such mixtures. 
 
 Loeb suggests that the physiological effect of a pure solution, 
 whether toxic or stimulating, is attril)utable to a reaction whereby 
 various cathions which are assumed to enter into combination with 
 the proteids of the organism are replaced ])y tlie cathion of the sur- 
 rounding solution, in accordance with the law of mass. Thus, in 
 case of ail animal or organ immersed in a solution of sodium chloride, 
 ions of calcium and of potassium would be forced from their organic 
 compounds and sodium ions would be substituted for them. This 
 would cause a disturbance of equilibrium and finally a cessation of 
 irritability in the tissues. Such effect can be prevented, or, if it has 
 not proceeded to the point of disorganization, counteracted by the 
 addition to the solution of salts containing the corresponding cathions, 
 i. e., potassium and calcium. Hence the author derives his concep- 
 tion of a " pliysioh)gically balanced salt solution," examples of which 
 are sea water, the blood of animals, and a mixture of definite concen- 
 trations of the chlorides of potassium, sodium, and calcium. The 
 chief function of such a solution is regarded as the maintenance of 
 "a certain physical condition, a certain labile eijuilibrium, of the 
 protoplasm or the colloids.""^ 
 
 From considerations such as these, and fi-om the discovery that a 
 close analogy as to absorption of water exists between the belwivior of 
 a frog's gastrocnemius immersed in a solution of a i^otassium, sodium, 
 or calcium salt and that of potassium, sodium, and calcium soaps,^ 
 the development of Loeb's theory of the existence and function of 
 "ion-proteid compounds" was logically inevitable. The hypothesis 
 is stated as follows: "Salts or electrolytes in general do not exist in 
 living tissues as such exclusively, but are partly in combination with 
 
 'Amer.Journ. Physiology. 3, 442 (1900). 
 
 •-'Ibid., 3,445 (1900). 
 
 «See Pfliiger-s Archiv, 75, :508 (1899). 
 
45 
 
 proleids. Tlie salt or cloctrolyte molecules do not enter into this 
 combination as a whole, but through their ions. The great impor- 
 tant^ of these ion-pi-oteid compounds lies in the fact that by the 
 substitution of one ion for another the physical properties of the j)ro- 
 teid compounds change. We thus possess in these ion-proteid com- 
 pounds essential constituents of living matter wliich can be modified 
 at desire, and hence enable us to vary and control the life phenomena 
 themselves. * * * jf j^ Ij^ ^j.^^g ^_|^^^ nf^ phenomena depend ui^on 
 the presence of a number of various metal j)roteids (Na, Oa, K, and 
 Mg) in definite proportions, it follows that sohdions ii-liidt amfdin 
 only one class of metal ions must act as a poison. The reason for 
 this is that tlie one class of mi^tal ions will gj-achudly take the place 
 of the other metal ions in tln^ ion-i)roteids of the tissues. Even a 
 pure NaC'l solution must thus l>e poisonous, althougii this salt i)erme- 
 ates all our tissues and is tlie main constituent of the [solu])le] inor- 
 ganic matter of the ocean." ^ 
 
 Pauli,- who pul)lished the same hypothesis almost simultaneously, 
 states his views with greater positiveness. "The general disti-ibu- 
 tion of the ion-proteid compound in the living organism can not ])e 
 doubted; indeed, we have sti-ong reasons for the assumption that all 
 the proteids of the protoplasm exist there only in combination with 
 ions." And again, "Not salts, l)ut salt-ions, are indispensable to the 
 organism." '•'• 
 
 Loeb's experiments show that to the same ions or mixtures of ions 
 different animals or different organs or stages of development of the 
 same animal may react in a different manner. Thisw^as noted in the 
 case of embryonic as compared with fully developed tissue and with 
 myogenic as compai-ed with nuerogenic contractions. Thus in pure 
 solution magnesium chloride is more favorable to the development of 
 fertilized eggs of the sea urchin than is sodium chloride, although the 
 latter causes while the former prevents i-hytlimical muscular conti-ac- 
 tion. On the other hand, as the predominant salt in a triple mixture 
 of chlorides (potassium and calcium being present in much smaller 
 quantity), sodium chloride favors, while magnesium chloride pre- 
 vents, the development of fertilized sea-urchin eggs.^ Calcium ions 
 prevent rhythmical muscular contraction, l)ut allow the muscle to 
 retain its irritability mucli longer than is possible in a solution from 
 
 ' Amer. Journ. Physiology, 3, 337 (1900). 
 
 ■Ueber physikalisch-chemische Methode uud Probleme in der Medizin, 19, 
 Wien (1900). 
 
 ■'Loew, although attempting no such extensive generalization, has touched upon 
 the question of ion proteids and their relation to vital phenomena in his discus- 
 sion of the harmfulness to plants exhibited by magnesium salts in the absence of 
 calcium. (See Bui. No. 18, Div. Veg. Phys. and Path., U. S. Department of 
 Agriculture, p. 42, 1899. ) 
 
 ^Amer. Journ. Physiology, 3, 439 (1900). 
 
4(^ 
 
 which they are absent.^ Results similar to those obtained T)y Loeb 
 have recently been recorded by other investigators.' 
 
 That tlie converse case may also occur is indicated by Loel)'s inves- 
 tigations: "Different combinations of ions may exist which all have 
 the same effect. It seems as if the physical condition of the colloids 
 were the essential point and that this might be affected by various 
 ion combinations in the same way."^ 
 
 It is not to l)e doubted that many peculiarities in relation to ions 
 will likewise be discovered in plants as compared with animals. A 
 case in point is that of magnesium chloride, which in pure solution is 
 eight times as toxic as sodium chloride to roots of the white lupine 
 and of alfalfa wliile the two salts are about equally toxic when cal- 
 cium is i)reseut . Hence lupine roots react toward these two salts in a 
 wholly different mauner than do sea-urchin eggs. Furthermore, a 
 comparison of different plants, one with another, or of different organs 
 or stages of development in the same plant, will surely reveal numer- 
 ous dissimilarities. 
 
 The importance of the ion-j)roteid theoiy as an aid to the study of 
 the effects, both toxic and lieneficial, which solutions of electrolytes 
 induce in organisms, can hardly be overestimated. It is to be regarded 
 as the only really scientific explanation of this class of phenomena 
 which has yet been attempted. Incomplete as the theory is in its 
 present form, and many as are the anomalies needing further study, 
 we can not but Avelcome it as a most promising instrument wherewith 
 to attack the vast problem of the physical properties and energies of 
 protoplasm.^ 
 
 Meanwhile it is highly desirable that the study of ion action upon 
 plants be extended. Experiments should be made with a larger num- 
 ber of different ions, and with mixtures containing more than two 
 kinds of cathions."^ It is most essential that many species of plants 
 be tested in order that we maj' determine what classes of reaction to 
 ions are peculiar to certain groups of organisms and what, if any, may 
 
 1 Festschrift fur Adolf Fick, p. HI (lb99). 
 
 -See the papers of Garrey, Anne Moore, Gushing, Lillie, and Stiles cited in the 
 Bibliography, (p. 56). True has lately experimented with Cladophora gracilis 
 grown in various synthetic solutions resembling sea water, and has made the 
 highly interesting discovery that an indefinite prolongation of life could be 
 obtained only when a solution equivalent to sea water in its other components, 
 but containing much more NaCl, was employed. Addition of calcium and potas- 
 sium salts was found necessary in order to neutralize effectively the toxic action 
 of a sodium salt solution. 
 
 ' Amer. .Journ. Physiology. 3, 443 (1900). 
 
 ^For certain limitations of the theory as now formulated reference should be 
 made to the very important paper of Kahlenberg [Journ. Physical Chem., 5, 339 
 (1901)]. 
 
 ■• Loeb's discovery that fertilized eggs of the sea urchin could be developed to the 
 pluteus t^tage in mixtures of three, but not of two chlorides, indicates that much 
 is to be expected from such an extension of these investigations. See Amer. 
 Journ. Physiology, 3, 441, (1900). 
 
47 
 
 be re^nrded as generic jjropert ies of protoplasni. No less important, 
 as Loeb's work with animals lias conclusively shown, will be the com- 
 parative stnd y of different organs and functions and stages of growth 
 in the same i)lant, as to their different reactions to the same ions and 
 combinations of ions. 
 
 From the point of view of agriculture the ion-proteid theory will 
 doul)tless throw light u[)on much that is now ol)scur(^ and ev^en para- 
 doxical in the relation between the plant and the soluble components 
 of the soil. Nothing is more certain, in the light of such observations 
 as are recorded in this paper, than the inadequacy of soil i)hysics and 
 soil chemistry alone to explain many details of this relation. The 
 chemist ry of protoplasm and its proteid compounds must surely be 
 taken into account before we may hope to get to the bottom of the 
 subject. 
 
 STIMULATING EFFECT OF DILUTE SOLUTIONS. 
 
 As an incident of these investigations it was demonsti'ated that in 
 the case of certain salts, when plant roots are exposed to pure solu- 
 tions which are much too dilute to produce any toxic effect, there 
 occurred a decidedly stimulating effect upon gi-owth, as compared 
 with that in the distilled-water control during a corresponding period. 
 As would be expected, this was shown to l)e the case for salts of cal- 
 cium, both the chloride and the sulphate acting as stimuli. Here, 
 however, we have to do with salts which contain valuable elements 
 of plant food. 
 
 But a marked stimulating action occui's in i>ure solutions of sodium 
 carbonate (slight in 0.002 normal, marked in 0.00125 normal and of 
 sodium bicarbonate o.Ol normal). The most i)i'onounced effect was 
 obtained in a 0.00125 normal solution of sodium carbonate, the average 
 elongation of the roots in that solution being one and one-half times 
 as great as in distilled water during the same period. In the case of 
 the two carbonates of sodium it seems necessary to i-egard the effect as 
 one of chemical stimulus, pure and simple. That this is not due to the 
 sodium ions is evident from the fact that very dilute solutions of other 
 sodium salts (sul[)hate, chloride) gave purely negative results. It was 
 at first thought that the physiological effects of sodium carbonate 
 (NajCOg) were attributable to the presence of hydroxylions in the solu- 
 tion, since the cori-osive, clearing action of more concentrated solutions 
 of this salt is precisely similar to that produced bj^ potassium hydrate 
 and sodium hydrate. But toxic, as well as stimulating reactions of 
 exactly the same character were obtained with solutions of the bicar- 
 bonate (NallCOg), in which a large excess of carbon dioxide was dis- 
 solved, and which gave no reaction with phenolphthaleine, even at the 
 end of the experiment. ^ In this case the consideration of free hydroxyl 
 
 ' Solutions of sodium carbonate which were many times too dilute to produce a 
 stimulating elfect, yet gave a strong reaction with phenolphthaleine. 
 
48 
 
 ious must T)e exclnded. Hence the eonclusiou seems iinavoi(lMl)le that 
 the carbonic acid (HCO3) ions produce the stimulating effect, improb- 
 able as this M^ould appear. To what agency should be ascribed the 
 characteristic toxic action (so different in kind from that of sodium 
 sulphate and sodium chloride) of stronger solutions of sodium bicar- 
 bonate, in which no free hydroxyl could be detected, is a question to 
 which no answer can at present be given. ^ 
 
 None of the other salts with which experiments were made in pure 
 solution were shown to stimulate elongation of tlie roots, although 
 the possibility is not excluded that solutions still more dilute than 
 those emjiloyed will give positive results. Magnesium sulphate was 
 found to be indifferent (neither toxic Jior stimulating) at 0.0003125 
 normal, magnesium chloride at 0.000()25 normal, sodium sulphate 
 (nearly) at 0.002 normal, and sodium chloride ~ (apijroximately) at 
 0.005 normal. 
 
 These observations accord with the well-known i^rinciple that many 
 violent poisons, if given in sufficientl}" minute doses, serve as benefi- 
 cial stimuli. Familiar examples are the action of arsenic, mercury, 
 
 'In experiments with sodium carbonate and sodium bicarbonate as to their 
 effect upon animals, Loeb encountered a very similar anomaly. The stimulating 
 effect of various hydrates upon the absorption of water by a muscle immersed in 
 a sodium chloride solution was shown to be clearly due to the hydroxyl ions, 
 being equal in amount when equivalent solutions of hydrates were used, irre- 
 spective of the character of the basic ions [see Pfluger's Archiv, 69, 10 (1898)], 
 The similar effect produced by carbonates of sodium and potassium was ascribed 
 to the same factor, the hydroxyl ions (1. c.p. 20). On the other hand, the effect of 
 sodium carbonate (NajCO^) in stimulating skeleton formation in the pluteus of a 
 sea urchin appears to be due to the carbonic acid (HCO3) ions, since sodium in 
 other forms, as well as hj'droxyl in the form of potassium hydrate, gave negative 
 results [Am. Journ. Physiology, 443, (1900)]. 
 
 - Pfeffer [Pflanzenphysiologie, Ed. 2, 1, 425] observes that possibly chlorides (e.g., 
 sodium chloride), like so many other substances, act in dilute solutions as chem- 
 ical stimuli. Storp [Biedermann's Centralbl. , 13, 76 ( 18S4) ] obtained a stimulating 
 effect upon the germination of seeds by immersing them in a 0.01 per cent solution 
 of sodium chloride. Jarius [Landw, Versuchsst., 32, 149 (lS8f3)] found that 
 even a 0.4 per cent solution of sodium chloride stimulated the germination of seeds 
 of v»^heat, rye, rape, maize, beaDs.and vetches. Jones and Orton (Bui. Vermont 
 Agric. Exp. Station No. 56, p. 13) observed, as a consequence of the application of 
 sodium chloride to land in order to exterminate the weed known as Orange Hawk- 
 weed (Hieracium aurantinciiin) ,a marked stimulating effect upon the growth of 
 grass in the same field. Peligot [Comptes rendus, 73, 1078 (1871)] suggests that 
 the stimulating effect upon field crops which is sometimes obtained with sodium 
 chloride may be due to its facilitating the decomposition of calcium phosphate 
 and thus increasing the amount of phosphoric acid at the disposal of the plant. 
 Kellner [Landw. V'ersuchsst., 32, 365 (1886)] attributes to a similar liberation of 
 phosphoric acid the stimulating effects of iron sulphate upon plant growth recorded 
 by Koenig and by Griffiths (see p. 49). Reveil [De Taction des poisons sur les 
 piantes, p. 41 (1865)] found that sodium hypochlorite in a solution of 0.1 per cent 
 stimulates germination and growth, but is injurious, especially to herbaceous 
 plants, when applied in greater concentration. 
 
49 
 
 strychnine, digitaline, etc., npon animals. Numerous investigators 
 have obtained similar effects with plants by supplyiug tliem with very 
 small quantities of various substances wliich can not be regarded as 
 sources of plant food, such as the extremely toxic salts of some of the 
 heavy metals. In i:)ractically all such cases, however, it is very prob- 
 able that considerable hj^drolysis had taken place and that the stimu- 
 lation might well be attributed to the hydroxyl ions thus introduced 
 into the solution. 
 
 Raulin experimented extensively with the fungus Aspergillus as to 
 the effect of various metallic salts in stimulating or liindering its 
 growth, his being among the first considerable work in this line.^ 
 
 The Avell-known observations of Frank and Kriiger- indicate tliat 
 copper in small quantities (furnished by spraying with Bordeaux mix- 
 ture) stimulates the growth of the potato, acting favorably upon almost 
 every organ and function, although this metal is well known to be 
 
 1 Ann. Sci. Nat., ser. 5, 11, 243 (1869). — The sulphates of zmc and of iron were 
 found to produce marked stimulating effect, the former increasing the dry weight 
 of the fungus two to three or even seven times, tlie latter about twice. In order 
 to show that the acid radicle was not responsible for the results, a corresponding 
 amount (0.06 gram of salt per 1,000 grams of culture solution) of ammonium sul- 
 phate was tried, but no stimulation was obtained. To demonstrate still more com- 
 pletely that basic radicles are here chiefly concerned other salts (nitrates of iron 
 and of zinc, zinc acetate, iron citrate) were tried and yielded stimulative effects 
 similar to those of sulphates. In cases where both iron and zinc were added to 
 the same culture solution (e. g., zinc nitrate plus ferric citrate, or ferric sulphate 
 plus zinc acetate, or zinc acetate plus ferric citrate) the stimulating effect was 
 decidedly more marked than when only one base was used. When sulphates of 
 both zinc and iron were present the effect was nearly twice as great as in the 
 absence of the former, and was e.xactly twice as great as in the absence of the lat- 
 ter. The diminution of the stimulating effect was almost as great if instead of 
 merely withdrawing one or the other base an equal portion of the second base was 
 substituted for the first; in other words, when two parts of zinc (or of iron) were 
 substituted for one part each of zinc and of iron. The stimulating effect of the 
 different salts of zinc expresses itself in a crop from two to four and six-tenths 
 times, that of iron in a crop one and four-tenths to two and seven-tenths times as 
 great as in the pure culture solution. 
 
 Manganese was found to i>rodnce effects similar to those of iron and of zinc, but 
 "less constant, less appreciable." Silica (as silicates of potassium and of sodium) 
 when added to the culture solution increased the dry weight of Aspergillus in the 
 ratio of 1. '3 or 1.4 to 1. 
 
 Raulin wrongly concluded that zinc and silica are indispensable to this fungus, 
 but justly emphasizes "" this influence of infinitely small <iuantities of substances 
 iipon vegetation "' (1. c, p. 253). 
 
 J. Koenig [Landw. Jahrb., 12, 837 (1883)] and Griffiths [Journ. Chem. Soc, 
 1884, p. 71, and 1885, p. 46] obtained evidence of a stimulating effect of iron 
 sulphate upon the growth of p'.aiits by watering soils used in culture experiments 
 with a solution of this salt. On the otlier hand Kellner [Landw. Verauchsst., 
 32, 365 (1886)], following the same method of experiment, obtained only negative 
 results. 
 
 ^Ber.d.deutsch.bot. Gesellsch., 12, 8 (1894). 
 
 8287— No. 71—02 -i 
 
50 
 
 extraordiuarily poisonous to plauts.^ Miani^ records the interesting 
 observation that in a vapor-saturated chamber the mere presence of 
 copper in the neighljorhood of but not in contact with a lianging drop 
 of water containing spores of Ustilago and poHen grains of various 
 phxnts stimulated the germination of tlie latter. II. Schulz'^ found 
 that alcoholic fermentation is accelerated by the i)resence of a small 
 quantit}^ of mercuric chloride and of other substances. The devel- 
 opuKMit of Aspergillus and of Penicillium in glycerol cultures was 
 stimulated, according to Pfeffer/ by the presence of small quantities 
 of zinc, manganese, cobalt, etc. Subsequently numerous experiments 
 as to chemical stimulation were made by Richards upon fungi. ^ 
 
 ' For a classical discussion of the toxic effect of exceedingly dilute solutions of 
 metallic salts upon the alga Spirogyra, seeNageli, Neue Denkschr. schweizerischen 
 Gesellsch. f. gesaramt. Natnrw., 33 (1893). Copper in a solution of 1 part to 
 1,00!),000,000 of water was found to be fatal! (I.e.. p. 23.) Attention has lately 
 been redirected to the extreme toxicity of copper by Deherain et Demoussy 
 fComptes rendus, 132, 533 (1901)] and by H. Devaux (1, c, p. 717). The latter's 
 observation that protoplasm absorbs less copper when exposed during several 
 hours to a large (luantit}- of a running very dilute solution (e. g.. of 1 part copper 
 to 400,000,000 parts water) than when placed for a short time in a single drop of a 
 much more concentrated solution (1 part copper to 30,000 parts water) leaves wholly 
 unexplained the negative results as to the extraordinary toxicity of this substance 
 recorded by Miani [Ber. deutsch. hot. Gesellsch.. 19, 461 (1901)], who immersed 
 his subjects for a long period in a single drop of solution. Nageli's experiments 
 have been more recently repeated ( upon Spirogyra and other organisms ) by Israel 
 und Kiingmann [Virchow's Archiv, 147, "393 (1897)], who made careful studies of 
 the ••oligodynamic "' effects produced by extremely dilute solutions of copper. A 
 noteworthy contribution to this subject by G-aleotti has lately appeared [Biol. 
 Centralbl., 21, 321 (1901)], in which the effect produced by a '•colloidal " solution 
 of copper [prepared after the electrical method recently described by Bredig and 
 Muller in Zeitschr. fur physik. Chemie, 31, 258 (1899)] is compared with that of an 
 "ionic"' solution of copper sulphate containing an equivalent amount of copper. 
 This author found that the former (colloidal) solution plasmolyzed the protoplasm 
 of Spirogyra in a dilution ( 1 gram-atom copper in 12.600,000 to 126,000,000 liters of 
 water) at which the ionic solution (of copper sulphate) produced no effect what- 
 ever. He therefore concludes that the action of the colloidal solution is a catalytic 
 one, closely analogous to the catalyzing action of such colloidal solutions (of cop- 
 per and other metals) upon hydrogen superoxide. 
 
 -Ber. deutsch. bot. Gesellsch., 19. 4()1 (1901). 
 
 •Pfiiigers Archiv f . die gesammte Physiol.. 42, 517 (1888). 
 
 ^ Jahrb. fiir wiss. Botanik, 28, 238 (1895). 
 
 "Ibid., 30, 665 (1897). Richards experimented with sulphate of iron and with 
 salts of zinc, cobalt, nickel, and manganese, as well as other substances, using as 
 subjects one species each of Aspergillus, Botrytis, and Penicillium. The estimation 
 of the amount of stimulus obtained was based upon the increase in dry weight of 
 the whole mass of mycelium in the culture as compared with that in a control free 
 from the stimulating substance. Zinc sulphate was found to be the most power- 
 ful stimulant, while important results were also obtained with sulphates of iron, 
 cobalt, and nickel. Salts of lithium were likewise very effective. It was found, 
 however, that acceleration of the development of the mycelium was accompanied 
 by an unfavorable influence upon the production of conidia, when salts of zinc or 
 of iron, amygdaline, or morphine were added to the culture solution. In other 
 words, a stimulation of one function or phase of development does not neces- 
 sarily imply stimulation of the organism as to all its functions. 
 
51 
 
 Recently ;iu iinporttint paper upon the effect of certain clieniical 
 stinnili npon fnngi and alga? has been published by OnoJ 
 
 Similar results as to the stimulation of life processes afforded by the 
 presence of small quantities of various non-nutritive substances have 
 been obtained in experiments witli animals. L(»eb- found this to be 
 true of certain acids, liyd rates, and mineral salts, the accelerating 
 effect i^roduced by the solutions upon tlie absori)tion of water by 
 muscles, the rhythmical contraction of muscles and the segmentation 
 of eggs being attributed to hydrogen ions, hydroxyl ions, or different 
 basic cathions, as the case nuiy be/^ 
 
 The as yet obscure pi-pblem of the mode of action of chemical 
 stimuli as regards i)lants has been discussed by Pfeffer,' from whom 
 it may be i^ermissible to <xuote at some length : 
 
 "In the regulation of activity chemical stimuli certainly play a 
 very extensive i^art. It is obviously a matter of chemical stimula- 
 tion that the seeds of Orobanche and of Latlnwa germinate only 
 upon the roots of host plants, and prol)ably the same occurs with 
 fungi. In the case of initiatory or only regulatory stimuli, theie may 
 be i^artly involved substances which the organism does not neces- 
 sarily require. In fact, under certain circumstances very different 
 substances can cause an acceleration of activity. * * * These 
 and similar plienomena obviously arise from different causes. Partly 
 it may be a matter of pliysiological counter reactions, which can also, 
 for example, occasion an increase of respiration, of circulation of the 
 protoplasm, etc., in response to injurious or other action. In other 
 cases a more simple chemical acceleration of reaction may be con- 
 cerned, as in catalytic action." 
 
 ' Journ. Coll. Sci. Univ. Tokyo, 13, 141 (1900). This author experimented with 
 various species of algie and fungi in order to determine their reaction to minute 
 quantities of the sulphates of zinc, nickel, iron, and cobalt as well as to sodium 
 fluoride, lithium nitrate, and potassium arsenate. He found a marked increase in 
 the total amount of vegetable matter produced in the presence of any one of these 
 substances, the increase in the c?se of algae being due, however, to the stimulation 
 of vegetative reproduction rather than to any marked increase in the size of indi- 
 vidual plants. The optimum dose for algse is considerably smaller than that 
 for fungi, 0.0001 gram molecule in most cases proving toxic to the former. Zinc 
 sulphate exerts the greatest stimulating effect. These salts (especially ZnSO^ 
 and NaFl) tend to hinder the development of spores in fungi. Copper sulphate 
 and mercuric chloride stimulate the growth of fungi, but not of algit. 
 
 "See all the papers of tliis author cited in the Bibliography. 
 
 ''It is probably vi^orth while at this point to call attention to the fact that in 
 nearly every case where this stimulative effect has been observed, electrolytes 
 have been used which are known to show marked hydrolysis, with the formation 
 either of hydroxyl ions, or more generally, as in the case of salts of the heavy 
 metals, of hydrogen ions. And it may well be that, as in the studies of Loeb, the 
 stimulating effects observed by former investigators may rightly be ascribed in 
 the majority of cases to the presence of these ions. 
 
 ^ Jahrb. fur wiss. Botanik, 28, 238, 239 (1895). 
 
52 
 
 111 allot her work^ Pfeffer emphasizes tlie idea of coiinter reactions, 
 suggestinj^ that "one has to do in this accelerating stimulation with 
 one of the manifold reactions which serve, through more intensive 
 activity, to counteract as far as possible an injurious influence or to 
 compensate injuries." And again :*" Probably this [stimulating] 
 effect results from a general reaction of the organism against injurious 
 substances, since similar effects are induced by ether, alkaloids, etc., 
 effects which also find expression in an increase of fermentation and 
 resi)iration. * *. * It is easy to understand * * * that further- 
 more such substances as are poisonous only in higher concentration 
 generally occasion no obvious [stimnlating] effect." 
 
 If it can be shown tliat such slJniulaling effect- is sufticicntly 
 permanent to express itself in a marked increase in the yield of a 
 crop, its economic importance would be obvious. That the pres- 
 ence of a certain amount of calcium salts in the soil may really act 
 as a chemical stimulant to growth, apart from the value of the salts 
 as plant food, or in rendering soluble other nutritive soil components 
 there appears to be some reason to believe. It is not impossible that 
 other substances, even perhaps those salts of magnesium and of sodium 
 which constitute the most noxious components of alkali soils, when 
 present in quantity too small to be harmful, may be activel}^ stimula- 
 tive rather than merely indifferent to jilants. That several of them 
 are likewise valuable as sources of nutritive material is well known. 
 Whether, after all, the distinction between the chemically stimulating 
 effect and the utility as food of certain mineral salts be always as sharp 
 as is commonh' supposed, is a question which can not yet be regarded 
 as decisively answered. 
 
 ECONOMIC IMPORTANCE OF THE RESULTS. 
 
 Some of the facts ascertained by these experiments with salt solu- 
 tions in their effect upon plants have a direct practical bearing upon 
 agricultural conditions and methods in regions where alkali salts are 
 frequent. Attention is particularl}^ directed to the effects obtained 
 by the addition of lime salts to others. Each of the common soluble 
 alkali salts is found to be very injurious when alone, but usually 
 much less harmful when two are mixed, esi^ecially when a salt of lime 
 is one component of the mixture. This is strikingly the case with sul- 
 phates of magnesium and of sodium, the noxious effects of these salts 
 being enormously lessened by the application of lime, particnilarly in 
 the form of gypsum or land plaster (the dihydrate of calcium sul- 
 phate). Contraiy to the general imjiression, this corrective effect was 
 found in w^ater-culture experiments to be considerablj^ less for "black 
 alkali" (sodium carbonate) than for any of the "white alkali" salts, 
 although even the harmfulness of black alkali can certainly be greatly 
 diminished by the use of gypsum. 
 
 'Pflanzenphysiologie (Zweite Auflage), 1, 374 (1897). 
 Ubid., p. 409. 
 
53 
 
 The soluble cliloride of lime could apparently also be used to advan- 
 tage upon a soil which is strongly impregnated with alkali. With 
 this salt the best effects would be anticipated when it is used as a I'eui- 
 edy for black alkali, although it is likewise a powerful antidote for the 
 chloride and sulphate of soda and of magnesia. l>ut, except in rare 
 instances, the use of chloride of lime upon a large scale is hardly i)rac- 
 ticable. The little-soluble carbonate of lime is likewise more or less 
 beneficial in all cases except that of black alkali, but it is a much 
 less powerful remedy than* is land jjlaster (calcium sulphate). 
 
 Much economic value should attach to an extension of these experi- 
 ments by using mixtures of more than two salts. It would thus be 
 possible to imitate nu^re closely the conditions whi(Oi obtain in alkali 
 soils, where several or all of these salts usuall}' occur together. 
 Furthermore, other kinds of plants should l)e tried in order to deter- 
 mine to what extent plants diffei- one from another in their power to 
 resist the effect of various combinations of alkali salts. In Ihis con- 
 nection experiments should be nu^ide with wheat, barley, sugar beets, 
 and oilier important crops of the region, as it may be found that one 
 crop is better adapted than another to withstand the effects of this oi" 
 that type of alkali soil. 
 
 This leads to the possibility of selecting alkali-resistant breeds of 
 each of the leading crops. By observation of a stand of wheat or of 
 alfalfa which has been injui-ed by the "rise of alkali" or by the use of 
 alkaline irrigating watei', it is usually possible to find here and there 
 individual plants which have succeeded in surviving the injui-ious 
 effects of the salts. ►Similar differences in the power of individuals 
 to resist the action of alkali salts was detected in th<^ culture exper- 
 iments. l>y continued selection of the seed of such resistant individ- 
 uals, sowing it season after season in alkali soil, thei'e is reason to 
 hope that in time a race could be developed and fixed which would 
 flourish in soils containing a greater amount of alkali than can be 
 endured by the ordinaiy agi'icultural varieties.^ It will likewise be 
 very interesting to determine whether a race bred to resist black alkali, 
 for example, will also pro\"e to be propoi'tionately resistant to white 
 alkali, or whether it will l)e possible and desirable to develop differ' 
 ent races to suit different types of alkali soil. An observation already 
 cited (seep. 34) would indicate that the different power of resistance 
 possessed bj^ individuals of the same species of plant is brought out 
 
 ' Observations made by Roos, Rousseaiix, and Dugast [Ann. de la Science Agron. , 
 ser. 2. Gieme annee, 2, 336 (1900)] indicate snch diilerences among the grapes culti- 
 vated in Algeria. It was found that of different varieties growing in the same 
 soil the fruit of some absorbed less sodium chloride f rtmi the soil than was taken 
 up by others. As the sale of wine containing too high a content of sodium 
 chloride is prohibited by law in France, the econonaic importance of this discovery 
 is obvious. Although the problem here involved is somewhat different from that 
 of the power of resistance to the poisonous effects of a salt upon the plant, it serves 
 to illustrate the general principle that different individuals or races show marked 
 dissimilarity in their behavior in the presence of a given soil component. 
 
54 
 
 more sharply in the presence of the carbonates of soda than wlien 
 other "alkali" salts are concerned. 
 
 So great appears to l)e the j)ronnse of results to be obtained l)y breed- 
 ing alkali-resistant races of the more important field crops of the far 
 western United States, that the Department of Agriculture has already 
 undertaken work on this line. During the past season experiments 
 with this end in view were begun under the direction of Mr. Webber, 
 of the Plant-BreediuiT Laboratory, Division of Vegetable Physiology 
 and Pathology. It is hoped that they will demonstrate the practical 
 value of this method of approaching the problem. 
 
 SUMMARY. 
 
 As the result of these i^reliminary studies, the following facts can 
 be regarded as established: 
 
 (1) Those readily solul)le salts of magnesium and of sodium which 
 are characteristic components of alkali soils are exceedingly injurious 
 to plants when exposed to pure solutions of them of concentration 
 above a minimum which is specific for each. 
 
 (2) They are toxic in the following sequence, beginning with the 
 most harmful : Magnesium sulphate, magnesium chloride, sodium car- 
 bonate, sodium sulphate^ sodium chloride, and sodium bicarbonate. 
 
 (3) Calcium chloride in pure solution is ten times less injurious 
 than sodium chloride, and two hundred times less injurious than 
 magnesium sulphate, if chemically equivalent solutions are considered. 
 
 (4) Magnesium carbonate in a saturated solution is not markedly 
 injurious, while magnesium bicarbonate in saturated solution acts as 
 a strong jwison. Calcium carbonate and calcium sulphate are posi- 
 tively stimulating in saturated solutions, while calcium bicarbonate 
 appears to be decidedly injurious. 
 
 (5) The toxic effect of the injurious salts is due very much more to 
 the influence. of the cathions (derived from the basic radicle) than to 
 the anions (fui-nished by the acid radicle). 
 
 (6) By mixture of equal volumes of two readily soluble salts, or by 
 the addition of a solid excess of a relatively insoluble to a solution 
 of an easily soluble salt, the toxic effect of the more harmful compo- 
 nent can in a majority of cases be diminished, or the concentration of 
 the more toxic salt endurable by the roots of plants can be increased. 
 
 (7) This increase is much greater in cases where a different kind of 
 cathion is added to the sohition than when a new anion only is 
 introduced. 
 
 (8) Addition of sodium ions to a solution containing magnesium 
 ions in most instances markedly weakens the toxic action of the latter. 
 
 (9) Addition of calcium, ions to solutions containing either sodium 
 or magnesium ions nearly always counteracts to an extraordinary 
 degree the injurious effect of the sodium or magnesium ions, this 
 beneficial influence being usually much more marked when calcium 
 is furnished as the sulphate than when the chloride is added. 
 
55 
 
 (10) Tlie aiiielioratinii' effect of ealciniu sulphate is much more 
 luai'ked when it is added to sulphates of maguesiuni and of sodium 
 than when it is mixed with the coi-respondini^- chloride. It raises the 
 concentration limit endurable by plant roots in magnesium sulphate 
 four hundred and eighty times, in sodiuui sulphate more than sixty 
 times. 
 
 (11) Even plasmolysis, although sui:)]30sedly a reaction to purely 
 physical stimuli, can apparently be completely i^revented by altering 
 the chemical naturt* of a solution without materially diminishing its 
 osmotic pressure. At any rate, plasuiolysis was not detected in cases 
 whei'e a solid excess of calcium sulphate had been added to a 0.3 or 
 even 0.4 normal solution of magnesium sulphate, although a pure 
 solution of magnesium STdjjhate is very strongh" i^lasmolyzing at 
 the concentrations named. 
 
 (12) Calcium chloride appears to be peculiarly effective in neutral- 
 izing the effect of sodium cai'bonate. 
 
 (13) The effect of one kind of ion in counteracting the physio- 
 logical action of another kind can not be entirely explained by a study 
 of the chemisti-y of the solution itself, l)ut must in part be referred 
 to complicate*! changes in the proto[)lasm of the organisms. The 
 theory that ions furnished by the dissociation of electi-olytes form 
 intinuite combinations with the proteids of protoplasm, and that 
 their mutually antagonistic effect expresses itself in a replacement 
 of one kind of ion b,y another as a result of change in the composition 
 of the surrounding solution, would appear to aft'ord the key to this 
 l)roblem. 
 
 (14) At a certain degree of dilution all of these salts become 
 indifferent (i. e., neither toxic nor stimulating) in their action upon 
 plants tissues. The maximum concentration of the indifferent solu- 
 tion is likewise specific for each salt. 
 
 (15) At a still greater dilution some of them, as the salts of calcium 
 and the two carbonates of sodium, produce a j)ositively stimulating 
 elfect upon the gi'owth of roots. 
 
 (10) Indi\i(lual plants show a marked dissimilarity in their power 
 of resistance to the toxic action of the alkali salts. Such individual 
 differences are strikingly accentuated in solutions of sodium carbon- 
 ate and of sodium bicarbonate of the maximum conccuitration which 
 will permit any of the roots to retain their vitality. 
 
 CONCLUSION. 
 
 Too great stress can not be laid upon the fact tliat the experiments 
 upon wliich the i^resent rejiort is based are merely preliminar3\ 
 Furthermore, they were designed prim'arily to afford a standard for 
 comparison of the salts involved. It is not to be expected— indeed, it 
 is assuredly not true^ — that in the soils containing these salts the con- 
 ditions are quite comparable to those maintained in the laboratory in 
 the course of these investigations. The physical nature of the soil, 
 as well as the presence of various other soluble substances, renders it 
 
56 
 
 certain that nowhere in the field will these salts be found to have 
 anything like the poisonous eflCect wliich they severall}' exert upon 
 the roots of plants immersed in water solutions. Nevertheless it is 
 only from such experiments, conducted under simplified conditions, 
 that we can draw conclusions as to the actual eft'ect of the components 
 of alkali soils upon plant growth. 
 
 It is very desirable that this line of investigation be continued and 
 extended. Further combinations, perhai^s of more than two salts, 
 should be tested; an attempt should be made to imitate as closely as 
 possible natural soil conditions; plants in different stages of growth 
 should be tried, for in irrigated regions it often happens that a stand- 
 ing crop is exposed to a varying soil content of soluble salts at differ- 
 ent periods of its development. Finally, it is highly important that 
 the experinn^nts be repeated with other plants of widely' different 
 relationship and, as far as possible, of actual agricultural importance 
 in the regioii concerned. For while we may assume for the present 
 that the same sequence of harmfulness of thes<>veral salts will obtain 
 in the case of most or all ordinarily cultivated plants, this is open to 
 doubt, and it is (piile certain that the actual limits of endurance differ 
 in the case of different plants. 
 
 BIBLIOGRAPHV. 
 
 ASKENASY. E. — Ueber einige Beziehiingen z.wischen Wachsthiim unci Teuiperatiir. 
 Ber. deutsch. bot. Gesellsch., 8, 61 (1890). 
 
 BoDLANDER, G. — Ueber die Liisliclikeit der Erdalkalikarbonate in Kohlensfmre- 
 haltigem Wasser. Zeit. fiir physik. Chem., 35, 25 (1900). 
 
 Bredig & MuLLER. — Ueber anorganische Fermente. I. Ueber Platinkatalyse 
 Tind die chemische Dynamik des Wasserstoffssiiperoxyd. Zeit. fiir physik. 
 Chem., 31, 258 (1899). 
 
 Cameron, F. K. — Application of the theory of solution to the study of soils. 
 Report 64, U. S. Department of Agriculture, pp. 141 to 172 (1900). 
 
 Cameron, F. K.~Soil solutions. Bulletin No. 17, Division of Soils, U. S. Depart- 
 ment of Agriculture (1901). 
 
 Cameron, F. K.— Solubility of gypsum in aqueoiis solutions of sodium chloride. 
 Bulletin No. 18, Division of Soils, U. S. Department of Agriculture; Journ. 
 Physical Chem., 5, 556 (1901). 
 
 Cameron & Briggs. — Ecniilibrium between carbonates and bicarbonates in 
 aqueous solution. Bulletin No. 18, Division of Soils, U. S. Department of 
 Agriculture (1901); journ. Physical Chem., 5,537 (1901). 
 
 Cameron & Seidell. — Solubility of gypsum in aqueous solutions of certain elec- 
 trolytes; solubility of calcium carbonate in aqueous solutions of certain 
 electrolytes in equilil)rium with atmospheric air. Bulletin No. 18, Division 
 of Soils, U. S. Department of Agriculture (1901). 
 
 Clark, J. F. — Electrolytic dissociation and toxic effect. Journ. Physical Chem., 
 3,263 (1899). 
 
 Clark, J. F. — On the Toxic Value of Mercuric Chloride and its Double Salts. 
 Journ. Physical Chem., 5, 289 (1901). 
 
 CouPiN, H. — Sur latoxicite du chlorure de sodium et de I'eau de mera I'egarddes 
 vegetaux. Rev. Gen. de Botanique, 10, 177 (1898). 
 
 CouPiN, H. — Sur la toxicite des composes du sodium, du potassium et de Tammo- 
 nium Ji regard des vegetaux superieurs. Rev.Gen.de Eotanique, 12,177 
 (1900). 
 
57 
 
 CouPiN, H. — Sur la resistance aux agents chimiques du protoplasma a Tetat de 
 
 vie ralentie. Comptes reudus Soc. Biol., 53, 541 (1901 ). 
 CouPiN, H. — Snr la sensibilite des vegetaiix superieurs a des doses tres failjles de 
 
 substances toxiques. Comptes rendus Acad. Paris, 32, Gl") (1901). 
 CouPiN, H. — Sur la sensibilite des vegetaux snperienrs a Taction utile des sels de 
 
 potassium. Ibid., 1582. 
 CusiiiNG, H.^ — Concerning the poisonous effect of pure sodium chloride .solutions 
 
 upon the nerve-muscle preparation. Amer. Journ. Physiology. 6, 77 (1901). 
 Curtis, C. C— Turgidity in mycelia. Bui. Torr. Bot. Club, 27, 1 (1900). 
 Dandeno, J. B. — The application of normal solutions to biological problems. 
 
 Bot. Gazette, 32, 229 (1901). 
 Deherain & Demoussy. — Sur la germiiiati(tn dans Teau distille. Comptes rendiis 
 
 Acad. Paris, 132, 523 (1901). 
 Devaux, H. — De I'adsorption des poisons metalli([ues tres dilues par les cellules 
 
 vegetales. Ibid., p. 717. 
 DiiUDE. O.— Handbuch der Pflanzengeographie. Stuttgart (1S90). 
 ENtiET., M. R. — Sur la dissolution du carbonate de magnesie parl'acide carbonicjue. 
 
 Comptes rendus Acad. Paris, 100, 444 (1885). 
 Engel, M. R. — Sur la formation de riiydrocarbonate de magnesie. Ibid., p. 911. 
 EscHENHAGEX. — Ueber den Einfluss von Losungen verschiedener Concentration 
 
 auf das Wachsthum der Schimmelpilze. Stolp (1889). 
 Frank und Kri'ger. — Ueber den Reiz. welchen die Behandlnng niit Kupfer auf 
 
 die Kartiiffelpflanze hervurbringt. Ber. deutsch. bot. Cxesellsch., 12, 8 
 
 (1894). 
 FREiTA<i, C. J. DE, — Ueber die Einwirkung concentrirter Kochsalzl<")sungen auf 
 
 das Leben von Bakterien. Archiv fiir Hygiene. 11, 00 (1890). 
 Galeotti, G.— Ueber die Wirkung kolloidaler und elektrolytisch dissoziirter 
 
 Metalllosungen auf die Zellen. Biol. Centralbl., 21, 321 (1901). 
 Garrry, W. E. — The effects of ions upon the aggregation of flagellated infusoria. 
 
 Amer. Journ. Physiology, 3, 291 (1900). 
 Griffiths, A. B. — Researches on the growth of plants under special conditions. 
 
 Chem. News, 47, 27 (1883). 
 Griffiths, A. B. — Experimental investigations on the value of iron sulphate as 
 
 a manure for certain crops. Journ. Chem. Soc. London. 45, 71 (1884). 
 Griffiths, A. B.— On the application of iron sulphate in agriculture and its 
 
 value as a jdant food. Ibid. 47, 4G (1885). 
 Heaij), F. D. — On the toxic effect of dilute solutions of acids and salts upon ])lants. 
 
 Bot. Gazette, 22, 125, t. 7 (189G). 
 Hiloard, E. W.— Nature, value, and utilization of alkali lands. Bui. No. 128, 
 
 Agr. Exp. Sta. Univ. of California (1900). 
 Israel & KLiNGMANN.—Oligodynamische Er.scheinungen (v. Niigeli) an i)flanz- 
 
 lichen und tierischen Zellen. Virchow's Archiv, 147, 293 (1897). 
 Jarius, M. — Ueber die Einwirkung von Salzh'isnngen auf den Keimungsprocess 
 
 der Samen einiger einheimischen Culturgewachse. Landw. Versuchsst. , 
 
 32, 149 (1886) 
 Jentys, S. — Sur Tinfluence de la pression partiale de I'acide carboni(iue dans I'air 
 
 souterrain sur la vegetation. Bui. Internat. Acad. Sci., Cracovie (1892), 300 
 
 (1893). 
 JoNKs & Orton.— Orange Hawkweed or "Paint Brush." Bui. No. 50, Vermont 
 
 Agric. Exp. Sta. (1897). 
 Kahlenberg, L. — The theory of electrolytic dissociation as viewed in the light 
 
 of facts recently ascertained. Journ. Physical Chem., 5, 339 (1901). 
 Kahlenberg and Austin. — Toxic action of acid s dium salts on Lupinus albus. 
 
 Journ. Physical Chem., 4, 553 (1900). 
 Kahlenberg and Mehl. — Toxic action of elei-trolytes upon fishes. Journ. 
 
 Physical Chem., 5, 113 (1901). 
 
58 
 
 Kahlenberg and True. — On the toxic action of dissolved salts and their electro- 
 lytic dissociation. Bot. Gazette, 22, 81 (1896). 
 
 Kellner, O. — Untersuchnngen iiber die Wirkung des Eisenoxyduls auf die Vege- 
 tation. Landw. Versuchsst., 32, ;>6.") (1886). 
 
 Klemm, p. — Desorganisationserscheinungen der Zelle. Jahrb. fnrwiss. Botanik, 
 28. 637. tt.8,9 (1895). 
 
 Knox, W. F. — Ueber das Leitungsvermogen wasserigen Losungen der Kohlen- 
 saure. Ann. Phys. Chem., 54, 44 (1895). 
 
 LiLLiE, R. — On differences in the effects of various salt solutions ou ciliary and 
 on muscular movements in Arenicola larvoe. Amer. Journ. Physiology, 5, 
 56 (1901). 
 
 LoEB. J. — Physiologische Untersuchnngen nber lonenwirkungen. Mitth. 1. 
 Pfliiger's Archiv f lir die gesammte Physiologie. 69, 1 (1898) ; Mitth. 2. 1. c. . 71, 
 457 (1898). 
 
 LoEB, J.— Ueber die Aehnlichkeit der Flussigkeits-re.sorption in Muskeln und in 
 Seifen. Ibid., 75, 303 (1899). 
 
 LoEB, J. — Ueber die Bedeutung der Ca- und K-Ionen fiir die Herzthatigkeit. Ibid., 
 80, 239 (1900). 
 
 LoEB, J. — Ueber lonen welche rhythmische Zuckuugen der Skelett-muskeln 
 hervorrufen. Festschrift fiir Adolf Fick 101. Braunschweig (1899). 
 
 LoEB, J. — On the nature of the process of fertilization and the artificial produc- 
 tion of normal larva^ (plutei) from the unfertilized eggs of the sea urchin. 
 Amer. Journ. Physiology, 3, 135 (18;)9). 
 
 LoEB, J. — On ion-proteid compounds and their role in the mechanics of life phe- 
 nomena. 1. The poisonous character of a pure NaOl solution. Ibid.. 837 
 (1900). 
 
 LoEB, .J. — On the different effect of ions upon myogenic and neurogenic rhythmi- 
 cal contractions and upon embryonic and muscular tissue. Ibid. , 383 ( 1900) . 
 
 LoEB, J.— On the artificial production of normal larvae from the unfertilized eggs 
 of the sea-urchin (Arbacia), Ibid., 434 (1900). 
 
 LoEB, J. — Further experiments on artificial parthenogenesis and the nature of the 
 process of fertilization. Ibid.. 4, 178 (1900). 
 
 LoEB, J. — Experiments on artificial i)arthenogenesis in annelids (Chtetopterus) 
 and the nature of the i>rocess of fertilization. Ibid., 433 (1901). 
 
 LoEB, J. — On an apparently new form of muscular irritability produced by solu- 
 tions of salts (preferably sodium salts) whose anions are liable to form 
 insoluble calcium compounds. Ibid., 5, 363 (1901). 
 
 LOEW, O. — The physiological role of mineral nutrients. Bui. No. 18, Div. Veg. 
 Phys. and Path., U. S. Dept. Agric. (1899). 
 
 LOPRIORE, G. — Ueber die Einwirkung der Kohlensiiure auf das Profcoplasma der 
 lebenden Pflanzenzelle. Jahr. fiir wiss. Botanik, 28, 531, tt. 6, 7 (1895). 
 
 MiANi, D.— Ueber die Einwirkung von Kupfer auf das Wachsthum lebender 
 Pflanzenzellen. Ber. deutsch. bot. Gesellsch., 19, 461 (1901). 
 
 Moore, Anne.— The poisonous action of saline solutions. Amer. Journ. Physi- 
 ology, 4, 386 (1900). 
 
 Moore, Anne. — The effect of ions on the contraction of the lymph hearts of the 
 frog. Ibid., 5, 87 (1901). 
 
 Nageli, C— Ueber oligodynamische Erscheinungen in lebenden Zellen. Neue 
 Denkschr. d. schweizerischen Gesellsch. fiir die gesammten Naturwiss., 33, 
 51 pp. (1893). 
 
 Ono. N. — Ueber die Wachsthumsbeschleunigung einiger Algen und Pilze durch 
 chemische Reize. Journ. Coll. Sci. Imp. Univ. Tokyo, 13, 141, t. 13 (19)0). 
 
 Overton, E. — Ueber die osmotischen Eigenschaften der lebenden Pflanzen- und 
 • Tierzelle. Vierteljahrsschr. Naturf. Gesells ch. Ziirich 40, 1 (1895). 
 
59 
 
 Pauli, W. — Ueber physikalisch-chemische Methode und Probleme in der Medizin. 
 
 Wien (lyOO). 
 Peligot. E. — Sur la repartition de la potasse et de la sonde dans les vegetaux. 
 
 Comptes rendus Acad. Paris. 73, 1072 (18T1 ). 
 Pfepfer, W. — Ueber Aufnahme von Anilinfarben in lebende Zellen. Unters. 
 
 aus der bot. Institut Tubingen. 2, 179 (1886). 
 Pfeffer, W. — Ueber Election organischer Ncihrstoffe. Jahrb. f i'lr wiss. Botauik, 
 
 28,205 (1890). 
 Pfeffer, W. — Pflanzenphysiologie. Ein Handbnch der Lehre voni Stoffwechsel 
 
 iind Kraftwechsel in der Pflanze. Zweite Auflage, erster Band (1897). 
 Pfeiffer. E. — Ueber die electrische Leitungsfiihigkeit des kohlensanren Wassers 
 
 und eine Methode. Fliissigkeitswiderstande nnter Imhen Drncken zu messen. 
 
 Ann.Phys. Chem.. 23, 625 (1884). 
 Raulin, J. — Etudes chimiques sur la vegetation. Ann. des Sci. Nat., ser. 5. 11, 
 
 9:3 (1869). 
 Reveil. — Recherclies de physiologie vegetale. De Taction des poisons sur les 
 
 plantes. Paris (1865). ' 
 Richards, H. M. — Die Beeindussung des Wachsthiims einiger Pilze durch 
 
 chemische Reize. Jalirb. fiir wiss. Botanik, 30, 665 (1897). 
 Richter, a. — Ueber die Anpassung der Siisswasseralgen an Kochsalzlrisungen. 
 
 Flora, 75, 4, tt. 1, 3 (1892). 
 Roos, RoussEAUX et Dugast. — Rapport sur les vins des terrains sales de I'Alge- 
 
 rie. Ann. de la Sci. Agronom.. ser. 2, (iieme annee, 2, 276 (1900). 
 Sachs, J.— Ueber den Einfluss der chemischen und physikalischen Beschatt'enheit 
 des Bodens auf die Transspiration der Pflanzen. Landw. Versxichsst. , 
 1, 203 (1859); Gesammelte Abhandl.. 1, 4.17 (1892). 
 Sachs, J. — Ueber das Wachsthum der Haupt- und Nebenwurzeln. Arbeiten des 
 bot. Instituts Wiirzburg 1, ;;85, 584 (1873-74); Gesammelte Abhandl., 2, 
 773 (1893). 
 ScHiMPER, A. F. W. — Pflanzengeographie auf physiologischer Grundlage. Jena 
 
 (1898). 
 ScHLOESiNG, Th. — Sur la dissolution du carbonate de chaux par Tacide car- 
 
 bonique. Comptes rendus Acad. Paris, 74, 1552 (1872). 
 ScHULZ. H. — Ueber Hefegifte. Pflligers Archiv fiir die gesainmte Physiologie, 
 
 42, 517 (1888). 
 SiGMUND, W. — Ueber die Einwirkung Chemischer Agentien auf die Keininng. 
 
 Landw. Versuchsst. 47, 1 (1896). 
 Stange, B. — Beziehungen zwischen Substratconcentrationen, Turgor und Wachs- 
 thum bei einigen phanerogamen Pflanzen. Bot. Zeitung, 50, 253 (1892). 
 Stewart, John. — Effect of alkali on seed germination. Ninth Ann. Rep. Utah 
 
 Agr. Exp. Sta. p. 26 (1898). 
 Stiles. P. J. — On the rhythmic activity of the (esophagus and the influence upon 
 
 it of various media. Amer. Jouru. Physiology, 5, 338 (1901). 
 Storp, Konig u. a. — Ueber den Einfluss von Koclisalz-und Zinksulfathiiltigem 
 
 Wasser auf Boden und Pflanzen. Biederm. Central bl., 13, 76 (1884). 
 Treadwell & Reuter. — Uber die Lr)slichkeit der Bikarbonate des Calciums und 
 
 Magnesiums. Zeitschr. fiir anorgan. Chemie, 17, 170 (1898). 
 True, R. H. — On the influence of sudden changes of turgor and of temperature on 
 
 growth. Ann. of Botany, 9, 3<)5 (1S95). 
 True, R. H. — The physiological action of certain plasmolyzing agents. Bot. 
 
 Gazette, 26, 407 (1898). 
 
 ' An extensive bibliography of the earlier literature of tiie subject is given by 
 this author (pp. 169 to 176). 
 
True, R.H.— The toxic action of a series of acids and of their sodium salts on 
 
 Lupinus albus. Amer. Joiirn. Sci., ser. 4, 9, 183 ( 1900) . 
 Vries, H. de.— Eine Methode zur Analyze der Turgor. Jahrb. fur wiss. Botanik, 
 
 14, 427 (1884). 
 Walker & Cormack.— Dissociation constants of very weak acids. Journ. Chem. 
 
 Soc. 77, 5 (1900). 
 Whitney & Means.— Alkali soils of the Yellowstone Valley. Bui. 14. Div. Soils. 
 
 U.S.Dept.Agric. (1898). 
 Wolf. W.— Die Saussure'chen Gesetze der Aufsaugung von einfachen Salzlo- 
 
 sungen durch die Wurzeln der Pflanzen. Landw. Versuchsst.. 6, 203 (18G4.) 
 Wolf, W. — Chemische Untersuchungen iiber das Verhalten von Pflanzen in der 
 
 Aufnahme von Salzen aus Salzlusungen, welche zwei Salze gelust enthalten. 
 
 Ibid., 7, 193 (1805). 
 
FORMATION OF SODIUM CARBONATE, OR BLACK ALKALI, BY 
 
 PLANTS. 
 
 By Frank K, Cameron. 
 
 INTRODUCTION. 
 
 Considerable attention has been jmid witliin the past few years to 
 the possibility of jj;rowing- valuable forage erops on some of the alkali 
 soils of the ai-id West. This subject Avas first taken up in California.' 
 The great value of saltbushes for certain soil conditions and for cer- 
 tain kinds of cattle feeding seems to be well estal)lislie(l, ])ut as botli 
 Hilgard and Goss^ have i)()inted out there is an element of danger, 
 expressed in tlie prevalent belief that most of t hes(^ plants, including 
 the grease wood, cliico, and other indigenous phints, convert the less 
 harmful neutral salts, such as sodium chloride^ and sodium sulphate, 
 into alkali carbonates — tluit is to say, the less hai-mful '' white alkali" 
 is converted into the more noxious "l)lack alkali,'' as has been 
 shown by the presence of sodium carbonate immediately under such 
 plants, whereas no trace of it exists some distance away. It may be 
 possible that the plants with their enormous root systems actually 
 gather up minute ti-aces of sodium carbonate, which may be present 
 in lower depths of soil, gradually causing an accumulation at the 
 surface on the decay of their roots and branches. I>ut the generally 
 accepted hypothesis of the conversion of the neutral salts appears 
 more probable, as will be seen in the course of this paper. It would 
 seem probable that plants growing in ]»unches or mats would be more 
 effective in producing these localized black-alkali spots, but some of 
 the most striking illusti-ations of this phenomenon have been observed 
 in connection with more upright species, sucli as Sarcohatns rcnni- 
 ctdatu.s, the common "greasewood" of the West. 
 
 In the study of the alkali soils of the arid regions the field parties 
 of the Division of Soils have found the local flora of great value in 
 indicating the character of the particular soils where they are found. 
 This apparent relation between the plant and the salts present in the 
 soil became of interest in this connection and was referred to the 
 
 ' University of California, Agricultural Experiment Station, Bui. No. 12.5 (1809). 
 2 New Mexico College of Agriculture and Mechanical Arts, Agricultural Experi- 
 ment Station. Bui. No. 22, p. 41 (1897). 
 
 Gi 
 
62 
 
 laboratory foi- consideration. The results of some preliminary inves- 
 tigations have proved of such interest as to warrant immediate 
 Dublieation. 
 
 CREOSOTE BUSH. 
 
 A specimen of the creosote busli ^ ( Corilleii frideukifa) was examined. 
 This, wliile a desert i)lant, is said to shun soils wliere there is much 
 water-soluble salts. Mr. Means states that its presence can be taken 
 as a sure indication of land free from injurious quantities of alkali. 
 It is found in dry, well-drained upland soils. 
 
 The material was thoroughly air dried. The leaves and_stems were 
 then carefully separated, and both of the separated samples were 
 ground fine in an agate mortar. A portion of each sami)le was burned 
 to ash. The finely ground air-dried material and the ash were each 
 carefully leached with successive small portions of water until the 
 leachings ceased to show the presence of chlorides. The leachings in 
 each case were then brought together and made up to a volume of 500 
 cubic centimeters, and the various determinations were made with 
 100 cubic centimeter portions. Tlie carbonates- were determined 
 by titrating with a twentieth normal (N 20) solution of hydrogen 
 potassium sulphate until loss of color, using phenolphthaleine as indi- 
 cator. So soon as tlie color had disappeared a drop oi two of a solu- 
 tion of potassium cliromate was added and the chlorine determined 
 by titrating with a tenth normal (N/10) solution of silver nitrate. 
 The sulphates, when determined, were estimated gi-avimetrically as 
 barium suli)hat-e in tlie usual manner. For convenience the acids thus 
 found to be present are stated as the corresponding sodium salts. 
 This procedure seemed to be justified by a subsequent determination 
 of the amount of sodium present in the solution. It is a well estab- 
 lished fact, and a familiar one to chemists, that when a salt of an 
 alkali metal is burned down with cliarcoal or other organic matter a 
 j)art of the mineral acid is volatilized and driven oflf, the alkali base 
 forming a carbonate, which is a stable compound even at quite high 
 temperatures. Nevertheless this is a point often overlooked in tlie 
 discussion of ash analyses. In obtaining the ashes the examinations 
 of which are described in this paper, ver}^ great care was exercised to 
 reduce the amount of this loss of the mineral acid as far as possible, 
 and the burning was done at as low a temperature as possilde. In 
 some cases the large amount of fused salt in the burning ash coated 
 the charred organic matter in such a way as to render further com- 
 bustion at a comparatively low tempei-ature (|uite impossilile. In these 
 cases the combustion was stopped, the fused salts leached out with 
 water, and the residue reburned. It seems probable, as will appear 
 from the results which will be presented, that the loss of mineral acids 
 
 ' Collected by Mr. Thos. H. Means near Tempe, Ariz. ; kindly identified for us 
 by Mr. F. V. Coville. 
 
 ■Report 64, Division of Soils. U. S. Dept. Agr. (1900); Amer. Chem. Jour., 23, 
 571 (1900). Bui. 18, p. 77, Division of Soils, U. S. Department of Agriculture (1901). 
 
68 
 
 ill t\u' bnriiinu' of llie plant lo asli was kept down to a v<'i'y sinall por- 
 eeiitag'e by followinji; the j)r()cediir(^ described. 
 
 The data obtained on examination of the ashes from the creosote 
 bush are presented in the following table : 
 
 Table XII. — Analysis of the ash of the creosote hi(s]i. 
 
 Weight of sample, grams 
 
 Weight of ash, grams 
 
 Asli.per cent of plant 
 
 Na-C't ):i. per cent of ash 
 
 NaCl.per cent of ash. 
 
 NaoCOs, per cent of airdried i)lant. 
 NaCl.per cent of air-dried plant ._- 
 
 Leaves, 
 
 8. 5G59 
 .X-2H2 
 
 9. tit! 
 
 8. ,91) 
 
 5.71 
 .«ti 
 
 Leaves 
 
 and small 
 
 stems. 
 
 3. 6943 
 . 3795 
 10. 37 
 13. 18 
 
 5.37 
 
 l.:i5 
 
 Stems. 
 
 I). 6445 
 .3710 
 
 5. 58 
 17. 73 
 
 3. .55 
 .99 
 .19 
 
 The dry leaves, which had been ground fine in a inoi'tar, were 
 extracted with distilled water at the room temperature in the manner 
 described above. The exti-act failed to show the presence of either 
 sodium carbonate or sodium chloride, l)ut appeared to be slightly acid. 
 An extract made by boiling the leaves witli water also failed to show 
 any cldorides oi* cai'boiuites. 
 
 From the facts which have been ])resent(Hl it would appear that 
 while the i>lant does conlain chlorine tlierc is no sodium chloi-ide 
 l)resent as such, and therefore^ it is })r()bable that the chlorine is in 
 oi-ganic combination although nothing is definitely known of the 
 preseiu-e of such combinations in plants. The sodium is lai'gely in 
 excess of the amount rec^uii-ed to balance the chlorine as sodium chlo- 
 ride. This fact was shown by an actual determination of the sodium.^ 
 It would se<nn, therefore, that at least a large part of the sodium in the 
 plant is in oi-ganic combination, jjossibiy with some organic acid, and, 
 on coml>ustioii or ultimate decay of the plant tissues, much sodium 
 carbonate would be formed, as was found to be the case when the plant 
 was reduced to ash in the laboratory. 
 
 It is interesting to note that tihe mineral constituents, as shown by 
 the asli analyses, had accumulated in the leaves to about twice the 
 amount in which they were held by tht; stems. The difference is very 
 much h'ss, however, if we consider only the water-soluble constitu- 
 ents in the ashes. Assuming, for the sake of argument, that the base 
 in combination with the carl)onic acid and chlorine as determined was 
 entirely sodium, its distribution is shown l)y the following table: 
 
 Table XIII. — Distribution of sodium in leaves and stems. 
 
 Part of plant. 
 
 Percentages calculated for ash. 
 
 Percentages calculated for air- 
 dried plant. 
 
 From 
 NaoCO,. 
 
 From Totni From 
 NaCl. 1 ^"^'*1- NaoCOs. 
 
 From 
 NaCl. 
 
 Total. 
 
 Leaves ' 3.86 
 
 Stems 7.69 
 
 3.25 6.11 ' 0.37 
 1.40 a09 .43 
 
 ! 
 
 0.22 
 .07 
 
 0.59 
 .50 
 
 ' Unfortunately it was not anticipated at the time this determination was made 
 that the exact fij^ure would be retiuired in this discussion, and' the data were not 
 entered in the laboratory notebook and have been mi&laid. 
 
64 
 
 It appears tliat in the leaves there was about 2.7 times as much 
 sodium as was necessary to l)alance tlic chhiriiie, wliile in the stems 
 tliere was more tiian seven times as much of tlie base as the acid would 
 require. This suggests the possibility that the chlorine was being 
 eliminated through the leaves, probalily in the form of some volatile 
 compound, whicli may l)e the source of the odor from the plant. This 
 idea is brought out somewhat more strikingly, perhaps, b}^ noting that 
 the analytical figures given above indicate that the total arhount of 
 water-soluble mineral constituents in the leaves is 1.19 times the 
 amount in the stems, but that the amount of chlorine in the leaves is 
 2.75 times that found in the stems; from which it would appear that the 
 chlorine was being concentrated in the leaves and, as has been pointed 
 out, was there present, in all probability, in organic coml)ination. This 
 is a point which merits furtlier attention, and it is hoped that it will 
 be the subject of a more thorough investigation in the future. 
 
 (iREASEWOOD. 
 
 A more thorough examination of a specimen of grcasewoo<U {Sarco- 
 hatus rermiculafus) was nuide. This is a typical "alkali plant," its 
 presence being usually regarded as a good indication of much water- 
 soluble material in the soil. Mr. Means reports that whenever he has 
 observed it the soil generally shows the presence of sodium carbonate, 
 the only exception being in Montana, where the soluble salts are 
 entirely sulphates. Iti would appear that this latter statement war- 
 rants further examination of the locality mentioned. 
 
 The analytical results obtained from examination of the ashes 
 follow : 
 
 Table XlY.—^lnalynis of the iiaJi of the greasewood plant. 
 
 Leaves ' 
 and blos- 
 soms (1). 
 
 Leaves 
 and blos- 
 soms (2). 
 
 Stems. 
 
 Weight of sample, grams 
 
 Weight of ash, grams . 
 
 Ash, per cent of plant 
 
 NaoCUs, per cen t of ash 
 
 NaCl, per cent of ash . _ 
 
 Na2S04, per cent of ash 
 
 Na2C08, per cent of air-dried plant - 
 NaCI, per cent of air-dried plant . .. 
 Na2S04, per cent of air dried plant- . 
 
 6395 
 
 750.') 
 
 85 
 
 93 
 
 47 
 
 97 
 
 43 
 
 29 
 
 (Hj 
 
 5. 0000 
 1.1730 
 
 3:5.47 
 
 57.90 
 
 22.24 
 
 13. 1)9 
 
 lOs (5817 
 
 .5374 
 
 4.94 
 
 39.46 
 
 14.31 
 
 3.69 
 
 1.45 
 
 .71 
 
 1.18 
 
 1 Owing to the relativel.v largo amount of fnsed salts which coated the carbon or other organic 
 matter, this latter conld not be completely burned off when reducing such a large samislo to ash. 
 
 Five grams of leaves and blossoms, by successive leachings with 
 distilled water at room temperature until the leachings aggregated 2 
 liters, gave (1) 5. SI and (2) 5. 08 per cent of sodium chloride. In both 
 experiments the washings showed no trace of soluble carl>onates, but 
 were slightly acid. The residue from (1) after ignition gave a trace 
 of sodium chloride and 0.04 per cent of sodium carbonate. 
 
 Collected by Mr. Frank D, Gardner near Salt Lake, Utah. 
 
65 
 
 Bv the method of Carius — that is, heating in sealed tubes with fuiu- 
 ing nitric acid and silver nitrate — 
 
 (1) 0.2327 gram of leaves and blossoms gave U.031U gram AgCl, 
 equivalent to 5.43 pei- cent of sodium chloride. 
 
 (2) 1.0750 grams of leaves and blossoms gave <).1432 gram AgCl, 
 equivalent to 5.43 per cent of sodium chloride. 
 
 From these results it would appear that the plant contains chlorine, 
 l)ut, within the limits of experimental error, all the chlorine is present 
 as sodium chloride, whicli can be leached out with water at ordinary 
 temperatures. This is probably true of the major pai-t of the sul- 
 phates also, although this was not shown quantitatively. A striking 
 feature is the much larger amount of ash from the leaves and blossoms 
 than from the stems and tlu^ markedly larger percentage of the alkali 
 salts in the ash of the former. The idea suggests itself that possibly 
 this plant takes up and stores the salts and holds them as such 
 until it is ready to use such part of them as it needs.' On the other 
 hand, it may be, for all that we now know, that these salts are 
 l)resent as described only l)ecause the plants can not prevent their 
 accumulation, and, so far from being an inheriMit feature of the plant's 
 econoni}', it may be a most undesirable accident due to tlieir peculiar 
 environment, but an accident in spite of which these i)articular plants 
 are able to survive.-' But, as Schimper • has pointed out, this can not 
 be true in all cases, as eviden<'ed by the fact that halophilous plaiits 
 show a tendency to take up more salts than nonhalophilous si)ecies, 
 even when grown in nonsaline soils. 
 
 None of the chlorine, apparently, was in organic combination, this 
 
 'Schimper [Indomalayiscbe Strandflora. p. 12 (1891); Pflanzen-Geographie, 
 p, 99 (1898)] has expressed the opinion that halophytes thrive on salty soils 
 because of a peculiar physiolojjjical structm-e which enables them to reduce to a 
 minimum the evaporation from their leaves and, in consequence, the absorption 
 of the salt solutions in the soil throuyh their roots. The salt content of their sap 
 is thus kept below a certain concentration, although this concentration may, 
 and often does, greatly exceed that which would be determined by "osmotic 
 equilibrium." 
 
 Stahl [Bot. Zeitung, p. i:59 (1891)] observes that only a few species, such aa 
 Rcaumnrid hirtelln, described by Volkens [Die Flora der Aegyptisch-Arabischen 
 Wiiste, p. 27 (1887)]. are known to be able to free themselves from the salt. 
 
 Diels [Jahrb. fiir wiss. Botani(iue, 32, 810 (1898)] objects that Stahl experimented 
 with cultivated plants and that the retarded root action noted by Schimper does 
 not tvike place under natural conditions, and that, as a matter of fact, and probably 
 through the agency of malic acid, most, if not all. the halophytes rid themselves of 
 an excess of chloride. DieFs methods of experiment, as well as the conclusions 
 which he draws from his own premises, are criticized by W. Beneke. Jahrb. fur 
 wiss. Botanique 36, 179 (1901). 
 
 Directly bearing upon this hypothesis is an observation by Detmer [Bot. Zeitung, 
 42, 791 (1884)] that "organic acids under the condition prevailing for the vegetable 
 organism are in a position to decompose chlorides with a formation of free hydro- 
 chloric acid." See also, Osborne, Report Conn. Ag. Ex. St., liiOO, p. 141. 
 
 -Contejean, Geog. Bot., p. 71. 
 
 ■Schimper, Pflanzengeogra[)hie. p. 101 (1898). 
 
 8287— No. 71—02 5 
 
66 
 
 plant being in wtrikiuii- contrast in this respoot to tlie CovUlea tri- 
 dei)tata examined above. 
 
 Another interesting' point is that the k^aehings of the air-dried 
 leaves and blossoms must have contained abont tliree times as inncli 
 sodium as was necessary to l)ahinee the hj'drochloric and sulphuric 
 acids ]3resent in tlie plaiit. The total amount of sodium calculated 
 from the ash analysis would l)e 8.. 32 per cent. A direct determination 
 of the sodium made on an aliquot part of the leachings gave 8.55 per 
 cent, while the amount calculated as necessary to balance tlie liydro- 
 chloric and sulphuric acids, as determined by the ash analysis, is 2.68 
 per cent. The residue after leaching contained practically no chlo- 
 rine, sulj)hates, or carbonates. Tt would appear that in the burning 
 of the plant or in its decay the sodium, which is probablj' present in 
 organic combination, juelds sodium carbonate as a decomposition prod- 
 uct, and this in turn is found in the ash or debris. It seems j)robable 
 that a large part of the chlorine whicli was originally taken up or at 
 least held by the plant in the form of sodium chloride has been thrown 
 off by the plant i]i some manner, the sodium being retained in organic 
 combination. 
 
 ABSORPTION OF MINERAL CONSTITUENTS BY THE PLANT. 
 
 Inspection of the analyses of the ashes of plants in genei'al, whether 
 leaves, stems, or in fact any part of the plant tissues, shows that there 
 arc more than enough base-forming elements to counterbalance the 
 possible inorganic acids which the results indicate to be present. 
 Moi'cover, the ashes are alkaline. Tt is still an open question as to 
 how these bases, which appear in excess, or, nu^re generally, how all 
 the l)ases, are taken up and assimilated by the plants and what 
 becomes of the acid radicals. While it is possible that some of the 
 alkaline materials may have Ix^en absorbed by the plant in the form 
 of carbonates as sucli, the amount thus absorbed will be relatively 
 very little, for by ()l>vi(>us metathetical reactions or doul)le decomposi- 
 tions there would be formed carbonates of the alkaline metals. These 
 latter would be hydrolized in water to some extent, giving caustic 
 solutions which would nndoubtedly coi'rode the tissues of the plants. 
 The question as to the disposition of the acid residues is then perti- 
 nent. Several possible explanations suggest themselves, which seem 
 worthy of attention in this connection. 
 
 It is possible that chlorine, for example, which may have been in 
 the acid radical, lias been changed by the plant in such a way as 
 to form organic substances, and that these organic substances may 
 be exhaled by the i)lant as odors or exuded by the leaves or roots. 
 Against the latter suggestion the experiments of Diels ^ indicate that 
 the excretion of such substances by the roots is very improbable. On 
 the other hand, the chlorine or sulphur may be retained in the plant 
 tissues in organic combination in such form that they more or less 
 
 Jahrb. fiir wiss. Botanique, 32, 316 (1898). 
 
67 
 
 completely disappeai' on combustion, the organic coml)ination volatil- 
 izing as such, or by decomposition yielding volatile products contain- 
 ing the chlorine or sulphur.^ In evidence against this view are the 
 results obtained in the examination of the sample of SarcoJxtfus 
 vennieidafus, where it was found that the total amount of chlorine in 
 the plant, as determined by the C'arius method, in which there Avas 
 afforded no opportunity for any of the chlorine to escape, was the 
 same as the amount leached out of the ashes by water, within the 
 limits of experimental error. 
 
 Another idea that presents itself is that the bases and acids are 
 taken up by the plant in the form of salt solutions; that the plant 
 selects and retains the bases and excretes the acid radicals in some 
 maniun- as acids. It is noteworthy, in this connection, that it has 
 been observed generally in the cases of water cultures that the nutri- 
 ent solutions gradually become acid unless special conditions are 
 intnjduced to prevent it. Occasionallj^ however, cases have been 
 found where the culture solutions actually become alkaline.- The 
 point of special importance in this connection is that either a base or 
 an acid radical, more often the latter, is either rejected oi- ejected by 
 the plant. 
 
 It seems to have been generally supposed that tlie acidity of these 
 solutions was due to organic acids formed and excreted by the plant, 
 but no satisfactory proof for this view has been adduced. The weiglit 
 of (evidence is now decidedly against this view. It is not at all dil'tl- 
 cidt, fi'om the point of view of the chemist, to construct a prol)able 
 "mechanism" for the phenomena presentecl, supposing that the plant 
 has selectively retaine* I the basic constituents and excreted the acids, 
 and that the acidity of the culture solutions is due to the free mineral 
 acids. Diels's'^ investigations in this direction are particularly interest- 
 ing. He found that certain halophilous plants, when placed in distilled 
 water, steadily lost the sodium chloride they contained. Tie showed 
 that the salt was not excreted as such,^ and offers as a probable expla- 
 nation that the greater anu)unts of malic acid — the formation of which 
 is shown to be a usual accompaniment of growth in succulent plants, 
 such as nu)st of the halophytes are — decomposes the sodium chloride, 
 forming sodium malate and hydrochloric acid, and this latter is possi- 
 bly excreted by the roots."' The solutions become acid, but, on account 
 of the experimental difficulties, it was not definitely proved that the 
 
 ' It is not intended to imply that chlorine and sulphur may not play very differ- 
 ent parts in the plant economy, but the general considerations advanced might 
 be true for either of these or other elements. 
 
 - Witness the classical investigations of Stohmann. Sachs, and Knop, described 
 by Johnson in How Crops Grow, p. ISO. 
 
 ■'Log. cit. See also Kearney, Contributions from U. S. National Herbarium, 
 Vol. V. No. :>, p. 277 (1900); and Benecke, Jahrb. fiir. wiss. Botaniciue. 36, 17!) ( 1901 ). 
 
 ■•This point was established as early as 1865 liy Wolf, Landw. Versuchstt., 7 i^p. 
 20, 211 (18G5). 
 
 ^*^ee reference to Benecke on p. <)-!. 
 
68 
 
 acidity was due to the presence of hydrochloric acid. It is intended 
 that some experiments in this direction shall soon l)e made in the 
 laboratory. 
 
 A. somewhat simpler explanation than the one just described may 
 be offered — simpler because it does not require that the plant must 
 first take up the acid radical and then go through the reverse process 
 of exuding it again. It is known with reasonable certainty that a 
 certain amount of hydrolysis takes place in aqueous salt solutions, 
 although the absolute amount may be, and with ordinary strong elec- 
 trolj^tes usually is, very small indeed; nevertheless, it does take 
 j)lace to some extent, and it seems not impossible that the plants 
 might show their selective properties in the solution, taking iip the 
 base more rapidly than the acid, the latter in consequence being left 
 in greater proportion in the culture or soil solution. Of classical 
 importance in this connection is tlie Avork of Kulm,^ who found t\\ai 
 when maize was grown in a solution containing ammonium chloride, 
 the ammonium residue was partly taken up by the plant and hydro- 
 chloric acid remained in the solution. In fact, there does not seem 
 to be any inherent difficulty in supposing that the plant might selec- 
 tively absorb anj' ion for wliicli it might have a special predilection. 
 As soon as this ion is removed fi-om tlie solution the corresponding 
 ion with its opposite charge of electricity must either be removed 
 from the solution by precipitation or volatilization, for example, or it 
 at once reacts witli the water. Supposing the ion removed by the 
 plant to be a base, the action of the remaining acid ion on the water 
 must necessarih' be accompanied b}" the liberation of oxygen from 
 the water of the solution. Whether or not any observation of this 
 kind has been made I do not know, but tlie liberation of the oxygen 
 njight very well take place so slowly as to escape detection. The 
 question as to what becomes of the electrical energy on the ion wliich 
 the plant al)sorl)s will be answered in a consideration of the work 
 energy, heat energy, or other equivalent forms of energy involved in 
 the mechanism of the absorption process, and does not necessarily 
 demand further consideration at this point.- 
 
 It must be admitted in all frankness that the known facts in our 
 possession are not sufficient to justify a positive opinion as to the 
 views just presented. They seem, however, to be founded on a 
 rational basis and are put forward tentatively as suggestive of ])ossi- 
 blt' lines of investigation and the justitication for formulating them 
 here will be found in the results of future work. Whatevei' may be 
 the bearing of this work on the ideas here presented, it can not fail 
 to be of the utmost importance in throwing light upon the difficult 
 problem of plant nutrition. 
 
 ' Hftnneberg"s Jonrnal, pp. 116 and 135 (1864). 
 
 '■ hese views are not intended to imply that salts can not be taken up as such, 
 ev n i>y nonhalophllous plants, nnder certain conditions. Wolf (loc. cit. ) has 
 lonj since shown that this may be done, and that, moreover, in such cases the 
 process can not be a simple ditfusion phenomenon. 
 
69 
 
 From i\w data presented above it is evident that in the decay of 
 wood or leaves' or, in general, of plant tissues, alkaline carbonates 
 are fni-nished to the soil. It may be that the j)rocesses of decay will 
 furnish at the same time organic acids stronger than carbonic acid 
 and in sufficient quantity to combine with all the bases and prevent 
 an alkaline reaction. As has been shown in this laboratory carbonic 
 acid itself may be formed in sufficient amounts to convert all the 
 carbonates to the form of bicarbonates and thus prevent an alkaline 
 reaction. There is not sufficient evidence to justify a i^ositive state- 
 ment, but it would seem pi-obable that this can not be always the case 
 and that in fact there is alkali formed by the decay of plant tissues. 
 In humid regions tlie alkali thus formed is removed by leaching or 
 similar processes and by chemical reactions with the other soil com- 
 ponents, for which reactions water is necessary. 
 
 In the ai-id regions, such as ai"e found in the western part of the 
 United States, peculiar phenomena, due to the si3ecial conditions 
 ther(^ existing, have been observed. The indigenous plants which are 
 found on the alkali lands are comparatively few in number, both as 
 to species and as to individuals; others have been artificially intro- 
 duced. They all have the property of absorbing more oi" less large 
 amounts of water-sohible minei'al salts and on analysis all show 
 characteristically large j^ercentages of bases. When the leaves or 
 debris from these plants hav<.' decomposed there is often found greater 
 or less accumulation of carl)onates, althotigh before the plant was cul- 
 tivated that particular region may have been quite free fi'om soluble 
 carbonates. The decay of any organic matter with the accompany- 
 ing formation of carbonic acid in a soil containing solul)le salts of the 
 alkali metals must be expected to result in the formation of soluble 
 carbonates, partly by dissolving lime or magnesium compounds, fol- 
 lowed by subsequent metathetical reactions ordotible decompositions 
 with the alkali salts; more slowly and in lesser degree, perhaps, but 
 nevertheless surely, if the fornuition of car])on dioxide is continued, 
 by a distribution of the base between the two acids. This last proc- 
 ess, liowever, is probably of decidedl}^ minor importance in the phe- 
 nomejia nnder consideration. 
 
 Owing to the conditions of climate and drainage existing in the 
 arid regions these carbonates when formed ai'e not leached awa3% as 
 in the humid regions, and gradually accumulate to the more serious 
 detriment of the soil. 
 
 COMPARISON OF ANALYSES. 
 
 For the purpose of comparison, two analj'ses of grease wood (Sdrroba- 
 tus verinieulatiis) ash are here quoted, the first i)ublished by llilgard,^ 
 and the other by Goss and Griffin.' 
 
 I University of California, Report of Agr. Exp. Sta., p. 142 (ISDO). 
 
 -New Mexico College of Agr. and Mech. Arts, Agr. Exp. Sta., Bu'. 22, p. -41 
 
 (1897). 
 
70 
 
 Table XV. — Txro analyses of ash of greaseicood i^lant . 
 
 Constituents. 
 
 First 
 analysis. 
 
 Second 
 analysis. 
 
 
 Percent. 
 13.03 
 
 Percent. 
 13. 12 
 
 
 
 
 
 SiO., 
 
 11.81 
 
 18.53 
 
 39.45 
 
 1.36 
 
 1.09 
 
 3.00 
 
 KoO - 
 
 23.06 
 
 NiioO . 
 
 23.89 
 
 CaO 
 
 6.53 
 
 MgO - 
 
 1.35 
 
 MnO" 
 
 Trace. 
 
 FeoOa 
 
 1 r.06 
 
 3.51 
 4.93 
 0.46 
 15. 04 
 
 14.73 
 
 AUO3 - 
 
 p„o- . .. --- 
 
 4.12 
 
 SOi 
 
 4.33 
 
 cot 
 
 23.80 
 
 CI . 
 
 8.01 
 
 
 
 
 
 103.04 
 3.25 
 
 101.81 
 1.81 
 
 
 
 
 
 99.79 
 
 100.00 
 
 1 By difference. 
 
 While these analyses differ considerably in details they indicate the 
 same general conclusions; that is, the asli or decomposition products 
 of the plant will yield a very large amount of alkali in the form of 
 carbonates. The figures in Hilgard's analj^sis, he states, indicate the 
 l^resence of aV)out 25 per cent of sodium chloride; about 8 per cent of 
 Glauber's salt (Na2SO4l0H2O), and about 30 per cent sodium carbon- 
 ate. Combining the figures of Goss and Griffin's analysis in the con- 
 ventional way, we find about 13 per cent sodium chloride and 29 per 
 cent sodium carbonate. The figures are misleading, for they depend 
 upon an arbitrary calculation of the data as salts, and the effect of the 
 other constituents can not properly be ignored. Similarly, but a quali- 
 tative comparison can be made from the data obtained by us. If it 
 may be assumed that the leaves and stems are of equal mass in the 
 individual plants wiien air dried, our results compare quite well with 
 the analyses just cited. 
 
 Acknowledgments are due Messrs. F. D. Gardner and Atherton Sei- 
 dell for assistance in the exiierimental work described. 
 
 SUMMARY. 
 
 It would seem as a result of the experiments described in this paper 
 that in certain cases at least a transformation of neutral salts to the 
 corresponding carbonates through the agency of plant growth is pos- 
 sible and even probable, and tliat this factor must be taken under con- 
 sideration in determining the value and use of such plants. Some 
 tentative suggestions are offered as to the disposition of the mineral 
 salts in plant economy, which it is hoped will lead to more exhaustive 
 investigations. 
 
RESISTANCE TO BLACK ALKALI BY CERTAIN PLANTS. 
 
 By Frank K. Cameron. 
 
 INTRODUCTION. 
 
 While working- in the San Joaqnin ^'alley, California, (luring this 
 past summer one of the field parties of the Division of Soils observed 
 three species of plants which appeared to be characteristic growths on 
 soils oontaitiing much " black alkali" or sodium carbonate. Super- 
 ficial examination in tlie field l)rought out the fact that the stems and 
 leaves of these three plants were quite acid, in some cases very 
 markedly so. A possible connection was suggested between this fact 
 and the one first noted — that these plants were all found on soils con- 
 taining much sodium carbonate. Specimens were collected and sent 
 in to the laboratory for further examination. Thej^ were kindly 
 identified by Mr. Kearney, of the Division of Vegetable Physiology 
 and Pathology. They consisted of three sami^les of DistichI is sp icafa , 
 numbered I, II, and III; one sample of Suaeda iiifermedia, which was 
 separated into two portions, the first numbered IV, consisting of the 
 stems alone, and the second numbered V, being composed of leaves 
 alone; one sample of Airiph.r bracfeo.sa, which was also separated into 
 a portion numbered VI, consisting of stems alone, and a poi'tion num- 
 bered VII, consisting of leaves alone. 
 
 Samples I, II, and III were thorouglily air dried by being allowed to 
 remain for about two montlis in the sacks in which the}^ were received 
 at the laboratory. It should be stated tliat a rough determination of 
 the acidity they displayed was made as soon as they were received in 
 the laboratory, and tlie results agreed fairly well with those obtained 
 by the more careful examination subse(piently made. 
 
 Samples IV, Y, VI, and VII were found to be very wet and in seri- 
 ous danger of fermenting when received at the laboratory. The)' were 
 therefore jilaced in a hot-air oven and dried for several days at from 
 105° to 110°C. In each case the material was then cut into small 
 pieces and kept in carefully covered beakers, to which, however, the 
 air had free access. 
 
 METHOD OF EXAMINATION. 
 
 The method of examination was in ali cases to steep the sample, 
 which had been cut into small pieces about a centimeter in length, 
 overnight or for about twenty hours in a convenient amount of dis- 
 
 71 
 
72 ■ 
 
 tilled water. About GOO cubic centimeters of the siipernatant solution 
 was then decanted through a folded filter, and the analytical details 
 carried out with 100 cubic centimeter portions of the filtered liquid. 
 It was thought probable that this procedure would give a close approxi- 
 mation to the soluble salts on the plant or held in its tissues in the 
 form of inorganic salts. The acid material on the surface of the plant 
 was evidently quite soluble in water. It was concluded, as will l)e 
 shown later, that it was an organic acid, and that in all probability 
 considerable quantities of its sodium or other salts, as well as the acid 
 itself, were on the surface of the jdant and dissolved in the water. 
 The amount of free acid was determined by titrating with a solution 
 of potassium hydroxide, which in tui-n had been carefully standard- 
 ized by titration against a twentieth noi-inal (N/20) soluti(Ui of acid 
 potassium^ sulphate. The other determinalions were made in the con- 
 ventiourd wa}'. 
 
 DISTICHLIS SPIC'ATA. 
 Table XVI. — Distichlis npicaia. 
 
 Grams of material 
 
 Ciiljic centimeters of leachings 
 
 Percentage (mineral matter) leached 
 out 
 
 Culiic centimeters N/3() acid equiva- 
 lent to 1 gram substance 
 
 Samjjle I. 
 
 Sample II. 
 
 Sample III. 
 
 13.13 
 
 750 
 
 35.35 
 1.350 
 
 39.48 
 1,500 
 
 4.53 
 
 5.13 
 
 5.73 
 
 3^5. 03 
 
 I.IG 
 
 3.4S 
 
 Percent- 
 age dis 
 tribu- 
 tion. 
 
 Ca 
 
 Mg .-. 
 Na... 
 
 K 
 
 SO4... 
 CI 
 
 CaS04 
 CaClo . 
 
 MgCl". 
 KCl.:. 
 NaCI , 
 Xa.... 
 
 5. 63 
 
 1 8(3 
 
 41.30 
 
 13. 40 
 
 4.38 
 
 33.44 
 
 100.00 
 
 34 
 
 Percent Percent- 
 in air- j age dis- 
 dried ' tribu- 
 
 material.! tion. 
 
 0.354 
 .084 
 
 1.867 
 .60a 
 .198 
 
 1.511 
 
 4. 530 
 
 .380 
 .474 
 .338 
 
 1.154 
 .688 
 
 1.598 
 
 4. 530 
 
 3.89 
 1.17 
 
 3:1 13 
 9. 70 
 4.69 
 
 48. 43 
 
 Per cent 
 in air- 
 dried 
 
 material. 
 
 Percent- 
 age dis- 
 tribu- 
 tion. 
 
 100. 00 
 
 6.64 
 3.58 
 4.57 
 18. 49 
 57 08 
 10. 64 
 
 100. 00 
 
 0.148 
 .060 
 
 1.696 
 .497 
 .340 
 
 3.480 
 
 3.35 
 
 3.10 
 
 37.34 
 
 13.33 
 
 3.73 
 
 53. 46 
 
 5.130 
 
 .340 
 .133 
 .334 
 .947 
 3.933 
 .545 
 
 5.130 
 
 100. 00 
 
 3.83 
 5.90 
 8.30 
 33.37 
 53.03 
 6.77 
 
 100.00 
 
 Per cent 
 in air- 
 dried 
 
 material. 
 
 0. 166 
 .130 
 
 1.561 
 .701 
 .1.56 
 
 3.006 
 
 5. 730 
 
 .319 
 .338 
 .470 
 1.333 
 3.981 
 .388 
 
 730 
 
 Table XVII. — Soil {()-12inches) in n-ltich Scunple I of DisticJilis sjiicata iras found. 
 
 Ca .. 
 Mg... 
 Na . . 
 K ... 
 SO4. 
 CI... 
 CO., . 
 
 HCO; 
 
 Per cent. 
 
 IIK). 00 r Percentage soluble, 1 gram soil to 30 
 I cubic centimeters water 
 
 1.60 
 
 . 75 
 
 5. .50 
 
 3.47 
 
 10.73 
 
 39.96 
 
 38.00 
 
 100.00 
 
73 
 
 The nnalytieal data obtained from an examination of tlio Distichlis 
 spicufa — Samples I, II, and III — are uiven in Tal)le XXI. Tlie most 
 striking point bronglit out is the very large amount of acid shown to 
 be on Sample I, amounting foi" 1 gram of the aii"-dried material to 
 the equivalent of 23 cubic centimeters of a twentieth normal (X/20) 
 acid. This substance was unquestionably an organic acid and a 
 fairly strong one. It did not api^ear 1o ad on crystals of calcite very 
 readily. This might have been due, howevcu-, to the fornuition of a 
 slightly soluble lime salt, which would pi-otect the calcite from the 
 solvent. The acid very readily decomposed the alkali carbonates and 
 neutralized not only ammonium hydrate but potassium or sodium 
 hydroxide in the pi-esence of cochineal or phenolphthalein as indi- 
 cator. It will be seen -by referring to the analytical figures that a 
 large amount of sodium is left after balancing the acids by the bases 
 found. Tliis would seem to find its i-eadiest explanation in supposing 
 that there was a much greater (juantity of the orgainc acid on the 
 plant than indicated by tlie equivalent of 23 cubic c(^ntimeters of 
 twentieth normal acid, but present in the form c)f the sodium or other 
 salts. Ky I'eferring to the analysis (Tal)le XVII) of the soil fi-om which 
 this Sample I of Di-sficlilis spicafd was taken, it will be seen that there 
 Avas relatively a A'ery lai'ge amount of sohil>lc carboiuit(\s jiresent, 
 about 2 per cent of the soil being comi)osed of these substances — an 
 anu)unt which would absolutely prohibit the growth of any ordinary 
 l)lant, even though much of th<^ salt was in llic form of l)icarl)onate. 
 Much of this material probably came in contact with the grass leaves, 
 in the form of dust or otherwise, with the result that tlie acid decom- 
 posed the car])()nates with the foi-mation of salts of the oi'gaiiic acid. 
 These same views seem to hold for Samples II and III as well, but to 
 a lesser extent, as is shown by the quantitative measurements given. 
 
 It would api^ear fi-om what could be learned in the field that this 
 grass, in the locality from which Samples II and III wei-e taken, often 
 carries as much of the acid material as Sample I shows, oi- even more. 
 Unfortunately for this investigation the most favorable season for 
 securing samples had passed before Samples II and III were gathered 
 and sent in. This subject will receive more careful attenlion during 
 iinothei- field season. 
 
 ISOLATION AND IDENTIFICATION OF ACtD EXUDATION. 
 
 Careful attempts were made to isolate ov at least to identify this 
 organic acid, but the atteini^ts proved unavailing for several reasons. 
 But very little material was at command when the investigation was 
 taken iq). The relatively large amounts of inorganic salts obtained 
 in the water extracts could not be well sei^arated and presented great 
 analytical difficulties in the attempts to isolate so small a (piautity of 
 the acid as was at our disposal. Attempts to ci-ystallize the material 
 from solution, either as the acid itself or as a. salt, proved disastrous 
 
74 
 
 on account of tlie rapid and abundant growtli of funj^i in the solution 
 when evaporation of the solvent at ordinary temi^eratures %vas 
 attenuated. Tlie solutions of the material failed to give any reactions 
 b}^ which it could be identified as one of the simpler and better-known 
 organic acids. For these reasons efforts to identify it were abandoned 
 temporarily and further work on it posti)oned until a time when a 
 larger amount of the material could l)e obtained. It is confidently 
 believed tliat the experience thus far gained will iusure a successful 
 issue to the next attempt in this direction. 
 
 HYDROSCOPIC SALT ON THE I'l.ANT SURFACE. 
 
 The analytical results would indicate that calcium clilorido as sucli 
 was on the grass, but if present no signs of it were observed on the 
 air-dried material. The samples were all thoroughly dry and not the 
 least evidence of any deliquescent substance on the surface was 
 apparent. It should l)e remembered, however, that the evidence 
 obtained in the examination of the organic acid indicated that the 
 calcium salt was much less soluble than tlie sodium or potassium salt. 
 In all probability the greater part of tlie calcium in combination in 
 the solid phase and not in the form of calcium sulphate was present 
 as the calcium salt of the organic acid; and the greater part of the 
 sodium which was assumed above to be in combination with the 
 organic acid was in reality' in combination with the chlorine, which 
 the analysis as stated assumes to be combined with calcium. 
 
 On the other hand, it has been noticed that this grass when grow- 
 ing in the field is frequently covered with a moist, stick}' substance, 
 which there is reason to l)elieve is caused by moisture absorbed from 
 the air by the salts, but only in sufficient quantity to partially dis- 
 solve them, making a paste or gummy mixture. So that it is not so 
 improbable tliat calcium chloride is sometimes formed and is to be 
 found as such on the living plant . 
 
 SELECTIVE ABSORPTION OF SOIL CONSTITUENTS. 
 
 Another point biought- out very strikingly by an examination of the 
 analyses is the relatively large amounts of both calcium and potassium 
 found in the leachings from the plants, when the proportion of these 
 elements in the water-soluble portion of the soil is considered. These 
 facts might possibly find an explanation in part in the lesser solu- 
 bility of the calcium and potassium salts of the organic acid and the 
 accumulation of sncli salts formed by contact of dust from the soil 
 with the acid, lint such reasoning does not afford an explanation of 
 the enormously increased ratio betAveen the chlor ions and the sulph 
 ions found in the plant leachings as compai'ed with the ratio of these 
 substances in the soil. The relative abs()ri)tive powers of the plant 
 for these various constituents are probably the controlling factors. 
 
75 
 
 It would seem desirable to give earnest attention to this sul)ject 
 Avitli plants grown nnder careftil supervision in the lield or laboratory, 
 as the evidence here presented indicates that the removal or cropping 
 of these plants for any purpose would result in taking from the soil 
 enoi'mous quantities of desirable plant food and t he consequent raising 
 of the proportion of undesirable elemejits in tlie soil. 
 
 FUNCTION' OF THE ACU) EXUDATION. 
 
 AVhen the lai'ge amounts of soluble carbonates found in the soils 
 upon which these plants gi-ow are considered, and avIkmi tlie disas- 
 trous corrosive action of this substance is remembered, the produc- 
 tion of the strong organic acid b}' the plant seems a wise protective 
 measure of natui-e. The tendency of sodium carbonate to outstrip 
 other salts in accumulating in the very top layers or crusts of a soil 
 and tliere corroding the root crowns of plants has ])een frequently 
 noted by all investigators of alkali proldems. It would seem that 
 this organic acid is produced by tlie plant in the manner most favor- 
 able to its being brought into contact with the surface sodium car- 
 bonate, partly converting this latter 1o the sodium salt of the acid 
 and pai'tly, in all probability, to sodium l)icarbonale, which, there is 
 strong reason for believing, is not itself so harmful to plant growth 
 as the noi'mal carbonate.' 
 
 PHOSPHORUS IX THE PLANT. 
 
 In the attempts to identifj^ the organic acid on Sample I, DisficJdis 
 sj^icata, some leachings were obtained which contained a small a-mount 
 of organic matter mechanically suspended, as well as some in solution. 
 They were allowed to stand for several daj's in an Erlcnm(\yer flask, 
 the mouth of whicli was covered with an inverted beaker. A rai)id 
 and voluminous growth of fungi was observed. On filtering off a 
 small portion of the sohition after it had been standing a day or two 
 a decided though small amount of phosphoric acid was shown to be 
 present. No trace of this substance was found in freshl}' prepared 
 leachings of the plant. It would seem pi'obable that it was formed 
 as a result of the action of organisms either upon dissolved organic 
 
 ' The especially pernicious effect on plants of carbonate of sodium is in all prob- 
 ability due to the fact that this salt readily hydrolizes in water witli the formation 
 of considerable amounts of sodium hydroxide', and it i this latter substance which 
 is in reality responsible tor its great destructive power. Sodium bicarbonate or 
 hydrogen carbonate, Na-HCOj, might be expected to hydrolize to some extent 
 also, being composed of a strong base in combination with a weak acid: this would 
 be equivalent to a partial inversion to the normal carbonate. But in the i)resence 
 of so much carbon dioxide as is present in soils this inversion to the normal car- 
 bonate would be greatly retarded or altogether prevented. The normal dissocia- 
 tion of the hydrogen carbonate would then be very small indeed and any chemical 
 activity of the compouiad depending on the formation of ions wouul be corre- 
 spondingly small. 
 
76 
 
 matter in the leachings or perhaps upon the organic matter mechan- 
 ically suspended in the solution. From lack of material it was not 
 possible to determine the amount of phosphorus in the plant, but the 
 qualitative observations cited would indicate that it was present in 
 considerable amount. It should be observed that it was a constituent 
 of the readilj^ Avater-soluble portion of the soil from which the plant 
 was taken in very small amounts, if, indeed, it were present at all. 
 Attempts to detect it by the phosphomolybdate method failed to show 
 a trace. The remarkable ability of this plant to take from the soil 
 solutions the mineral constituents it needed, in the presence of the 
 enormous excess of other readily soluble substances, is bi'ought out 
 very strikingly in this connection. 
 
 For the reasons here presented it would seem that this plant is 
 worthy of the serious consideration of the botanist and physiologist, 
 and is undoubtedly of very great economic importance. 
 
 ASH ANALYSES. 
 
 Ash analyses of all the plants considered in this paper were made, 
 in the hope that some conclusions might be drawn as to the inorganic 
 materials in the plants themselves, and as to how much, relatively-, 
 was capable of being removed by leaching. These analyses will not, 
 however, be presented, for it is ol)vious that they have no value what- 
 ever for the purposes here indicated. The mixture of salts on the 
 plants w^as so large in amount, and fused at so low a temperature, that 
 it quickly coated the organic matter, so that it was necessary to heat 
 to a very high tempei'ature and thoroughly stir the mixture to obtain 
 anj'thing like a thorough combustion of the organic material. This 
 resulted in a very great loss of the salts b}^ volatilization, sodium 
 chloride and potassium chloride being especially important in this 
 connection. Further, tlie burning of either sulphates or chlorides of 
 the alkalies with organic materials necessarily means the more or less 
 complete volatilization of the sulphur and chlorine, respectivel}', and 
 the formati(ni of the corresponding alkali carbonates, a point often 
 overlooked in the considei'ation of ash analyses. As a consequence 
 of these factors, the results obtained would certainly be misleading. 
 It would appear, from the analyses of the ashes of the plants we are 
 considering, that much more of these soluble mineral constituents 
 can be leached fi-om the plants than the plants ever contained, 
 which is an obvious absurdity. For this reason it does not seem Avorth 
 while to give these ash analj'ses any further consideration. 
 
77 
 
 SUAEDA INTERMEDIA AND ATRIPLEX BRACTEOSA. 
 Table XVlll.^Snaeda intermedia. 
 
 
 Sample IV— 
 Stems. 
 
 Sample V— 
 Leaves. 
 
 
 7.61 
 
 75(1 
 14.73 
 
 .84 
 
 14 35 
 
 Cubic centimeters of leaobings . 
 
 750 
 
 Percentage (mineral matter) leached out _ 
 
 Cultic centimeters N/30 acid, equivalent to 1 gram sub- 
 stance 
 
 33.80 
 
 Ca . 
 
 Mg. 
 Na. 
 K .. 
 SO4. 
 01... 
 
 CaS04 . 
 CaClo .. 
 MgCi. . 
 MgS04. 
 KCl.... 
 NaCl... 
 Na 
 
 Per- 
 ' centage 
 I distribu- 
 tion. 
 
 0.48 
 .75 
 
 43. 3(i 
 7.30 
 1.43 
 
 47.78 
 
 Percent- p 
 
 afi?dried' ^^^'^^^ 
 l,,K ! distribu- 
 
 stance. I "on. 
 
 0.071 
 .111 
 
 6. S3!) 
 
 l.OCiO 
 .211 
 
 7.038 
 
 100.00 
 
 1.63 
 
 .00 
 
 3.66 
 
 .35 
 
 13.71 
 
 64.84 
 
 16.83 
 
 100. 00 
 
 14. 730 
 
 0.15 
 
 56. 18 
 4.53 
 
 100. 00 
 
 .339 
 
 . 000 
 
 .393 
 
 . 057 
 
 3.015 
 
 9. 551 
 
 3.476 
 
 14. 730 
 
 ..51 
 
 .(X) 
 
 3. 35 
 
 .64 
 
 8.61 
 
 53.33 
 
 35. 56 
 
 100. 00 
 
 Percent- 
 age in 
 
 air dried 
 
 sub- 
 stance. 
 
 0. 036 
 
 .170 
 
 13.375 
 
 1.075 
 
 .306 
 
 8. 938 
 
 33.800 
 
 ~.l3i 
 . 000 
 . 559 
 
 .152 
 2.049 
 13. 456 
 8.463 
 
 33. 800 
 
 Tablp: XIX. — Afriplc.r hracteosa. 
 
 Grams of substance 
 
 CuIhc centimeters of leacbiugs _. 
 
 Percentage (mineral matter) leached out 
 
 Cubic centimeters N/30 acid, equivalent to 1 gram sub- 
 stance 
 
 Sample VI- 
 Stems. 
 
 31.49 
 750 
 4. 48 
 
 Sample VII- 
 Leaves. 
 
 17.79 
 7.511 
 10. 34 
 
 3. 61 
 
 Per- 
 
 Percent- 
 
 centage i^fg«,!'^^ 
 
 distribu 
 tion. 
 
 sub- 
 stance. 
 
 Per- 
 centage 
 distribu- 
 tion. 
 
 Percent- 
 age in 
 
 air-dried 
 
 sub- 
 stance. 
 
 Ca 
 
 3.33 
 3. 05 
 
 37.79 
 7.90 
 4. 10 
 
 45.84 
 
 0. 104 
 .093 
 
 1.693 
 .3.54 
 .184 
 
 2.05:3 
 
 0.39 
 
 3.m 
 
 41.:30 
 5.39 
 6.60 
 
 42.96 
 
 0. 040 
 
 Mg 
 
 .;344 
 
 nI.. : ::::::::::;::::::: 
 
 4. 329 
 
 K. . 
 
 . 5.52 
 
 SO4 
 
 .676 
 
 CI 
 
 4.399 
 
 
 
 
 100.00 1 
 
 4.480 
 
 100.00 
 
 10.240 
 
 CaS04 
 
 5.80 
 1.74 
 8.03 
 .00 
 15. 03 
 53. 18 
 17.33 
 
 .2ti0 
 .078 
 .3t)0 
 .000 
 .673 
 3.3:38 
 .771 
 
 1.3:3 
 
 .00 
 
 7. .57 
 
 7.08 
 
 10. 37 
 
 53. .56 
 
 30. 19 
 
 .136 
 
 Cai.li 
 
 .000 
 
 MgCl., 
 
 MgSOi 
 
 .775 
 
 .735 
 
 KCl .! :.::::::::::;::::.:.:::. 
 
 1. 0.53 
 
 NaCl 
 
 5. 485 
 
 Na 
 
 2. 067 
 
 
 
 
 KM). 00 
 
 ■4. 480 
 
 1(X). m 
 
 10. 340 
 
 111 Tables XVIII and XIX are found tlie data obtained from an 
 examination of the Stiaeda Interined la und Atriplex hracteosa, respec- 
 
78 
 
 tively. For the purj)oses of this paper they may very well be discussed 
 together. Both analyses show the j)roductiou by the plants of an 
 organic acid or acids strong enough to decompose alkali carbonates; 
 and that to some extent salts of this acid or acids, as well as the acids 
 themselves, accumulate on the plants. In both plants this acid organic 
 material is accumulated on the leaves rather than the stems, a situa- 
 tion which would seem more favorable for its being brought into con- 
 tact with the alkali carbonates-on the surface of the ground. 
 
 The accumulation of considerable amounts of i)otassium is again a 
 noteworthy feature with these species. In both cases there is an 
 apparently greater proportion of potassium in the water-soluble por- 
 tions of the stems than in the leaves, when considered in relation to 
 the other elements present; but when considered in relation with the 
 air-dried material as a whole, the amount of potash is about the same 
 in both leaves and stems for each of these plants. 
 
 In the case of the stems of AtripJex hracteosa the conventional state- 
 ment of the analytical results as salts would indicate the presence of 
 calcium chloride as sucli, and on the stems and leaves of both the 
 Suaeda intermedia and Airiplex hracteosa considerable amounts of 
 magnesium chloride are indicated. That these salts were actually 
 present as such, however, is negatived by the fact that the air-dried 
 samples did not in any case show the presence of au}^ notably deli- 
 quescent substance on their respective surfaces after being dried in 
 the oven. The conventional method of statement is again misleading, 
 as in the case of the DistichHs spiccda, discussed above. 
 
 SUMMARV. 
 
 From the facts which have been presented in tliis paper the follow- 
 ing conclusions seem justified:^ 
 
 1. That the plant sfjccies here considered can make a satisfac- 
 tory growth on soils containing relatively large amounts of soluble 
 carbonates. 
 
 2. That this satisfactory growth is i^robablj^ due, in large measure 
 at least, to the i:)roduction and exudation by these plants of consider- 
 able amounts of soluble organic acids capable of decomposing soluble 
 carbonates, and thus protecting the root crowns from the corrosive 
 action of hj'drolized alkalies. 
 
 o. That it appears certain that large quantities of the most valuable 
 plant foods are removed from the soil by these plants, and that in anj^ 
 contemplated use of them, involving their cropi:»ing or removal from 
 the soil, this factor merits earnest consideration. 
 
 'Acknowledgment iw due Mr. Athertoi Seidell for assistance in making: the 
 analyses presented in this ijaper. 
 
 o 
 
/ 
 
 EJe'lS