UNIVERSITY OF CALIFORNIA 
 
 COLLEGE OF AGRICULTURE 
 
 AGRICULTURAL EXPERIMENT STATION 
 
 CIRCLUAR No. 235 
 
 April, 1922 
 
 SOIL ANALYSIS AND SOIL AND PLANT 
 INTERRELATIONS 
 
 By D. E. HOAGLANDi 
 
 There exists a common impression that from a chemical analysis 
 of a sample of soil reliable indications may be obtained concerning 
 the productivity, fertilizer requirements, or crop adaptation of the 
 area which the sample is supposed to represent. It is the purpose of 
 this publication to show why this idea is erroneous and to describe in 
 brief outline certain modern views on the interrelations of soils and 
 plants. 
 
 It is true that Hilgard, as well as other soil chemists, was inclined 
 to attach considerable importance to soil analysis as applied to Cali- 
 fornia soils, but within recent years investigation on soils and plants 
 have progressed, and many of the older ideas with regard to their 
 relations are being discarded as a result of more recent researches. 
 In this respect soil and plant investigations do not differ from those 
 in any other field of science. New methods of experimentation, new 
 ideas gained from other sciences, new opportunities for critical study, 
 all enable the present-day student of plant nutrition to obtain a much 
 more accurate conception of the interrelationships of soils and plants 
 than was possible in the past. 
 
 According to the older teachings, which are still promulgated to a 
 considerable extent, it is very important to ascertain the total amounts 
 of potassium, nitrogen, phosphorus, calcium, etc., contained in an acre- 
 foot of soil, or if not the total quantities then the percentages of the 
 important elements soluble in some acid, such as hydrochloric acid, in 
 Hilgard 's method of soil examination. If the analysis of a soil showed 
 less than a certain percentage (a varying standard) of one of the 
 principal nutrient elements of the soil, then it was frequently thought 
 that the soil would respond to applications of fertilizers containing 
 that element. Another modification of this view is to the effect that 
 it is essential to determine the total quantities of important elements 
 
 i Acknowledgment is made of the review of the manuscript by Professors 
 Burd, Hibbard, Kelley, and Lipman, of the College of Agriculture. 
 
2 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION 
 
 present in the soil as a sort of inventory, the results of which are 
 thought to indicate the probable duration of the fertility of the soil. 
 The assumption is also made sometimes that high totals imply corre- 
 spondingly large quantities of plant foods available to the plant. 
 According to still other teachings, it is urged that plant foods 2 must 
 be restored to the soil in proportion to withdrawals by the crop. 
 
 In discussing these various points of view it may be convenient to 
 describe first certain experiments carried on during the past eight years 
 by the Division of Agricultural Chemistry (now part of the Division 
 of Plant Nutrition). At the beginning of the experiments large quan- 
 tities of different soils collected in various widely separated sections 
 of the state were assembled in Berkeley. Seven of these soils were 
 classified as fine sandy loams of the same soil series and seven as silty 
 clay loams of different soil series. A set of tanks, each of about one- 
 ton capacity, was installed, and after very thorough sieving and mixing 
 two tanks were filled with each soil, twenty-eight tanks in all. (One 
 soil was found to produce practically no crop and is not considered 
 in the present discussion.) During the first season all the tanks were 
 cropped with barley. In all subsequent seasons only one tank of each 
 soil was planted, the others remaining uncropped, but otherwise treated 
 similarly. Great care was taken to maintain uniform and favorable 
 moisture conditions in all tanks during the growing season. A repre- 
 sentative sample of each soil was analyzed for the total percentages 
 of all the important elements present. Determinations were also made 
 of the amounts of these elements soluble in strong hydrochloric acid 
 (Hilgard's method) and in citric acid (a weak acid). 
 
 One important part of the investigation consisted in making ex- 
 tracts with pure water of large samples of soil taken from the different 
 tanks and then analyzing these extracts. The analyses were made not 
 on one sample merely but samples were taken every two weeks during 
 the growing season and analyzed in the same way. 
 
 At the present time, after seven or eight years of intensive and 
 laborious work on these soils (many other experiments were performed 
 which are too technical for this discussion), certain general principles 
 have been suggested, which also receive support from a number of 
 other investigations and from theoretical considerations. Using the 
 results of the California experiments by way of illustration, we shall 
 now make a further inquiry into some of the common conceptions of 
 the soil in its relation to the plant. In the first place, the investiga- 
 
 2 This commonly used expression is not scientific;) I ly accurate, but is retained 
 Tor convenience. It is used to denote certain chemical elements in the soil which 
 are essential to plant growth, especially nitrogen, phosphorus, and potassium. 
 
Circular 235] sqi L analysis and plant interrelations 3 
 
 tions conducted in Berkeley show that there is no consistent relation 
 whatever between the crop yield in any year and the total quantity 
 of any element present in the soil. Neither was such a relation ob- 
 served when the percentages of the plant food elements soluble in 
 hydrochloric acid were considered. The data for the citric acid ex- 
 tracts were only slightly more significant. On the other hand, careful 
 study of the amounts of plant foods (especially nitrate) extracted by 
 water showed certain interesting relations to crop yields. No one 
 observation had any particular significance, but it was necessary to 
 take into consideration the data obtained on water extracts of samples 
 of soil taken during various periods of crop growth. 
 
 It is believed that these and other recent experiments emphasize a. 
 number of important principles involved in the interrelations of soils 
 and plants. The plant absorbs mineral nutrients from the soil solu- 
 tion. 3 Only after the required elements are dissolved in this solution 
 do they become of use to the plant. Now the amounts of the plant 
 foods found in the soil solution at any time are exceedingly small when 
 compared with the total quantities contained in the soil. Moreover, 
 the composition of the soil solution is by no means constant. In fact, 
 one of the outstanding characteristics of such a solution is that it is 
 changing even from day to day. For example, in the case of nitrate, 
 one of the most variable constituents, at the beginning of the growing- 
 season comparatively large amounts may be found in the soil mois- 
 ture, but after the crop has grown for some weeks or months scarcely 
 any nitrate may remain. This is equally true of both a very produc- 
 tive and a slightly productive soil. The quantity of water-soluble 
 potassium, calcium, magnesium, and other elements may also be mark- 
 edly decreased at this period. In many cases the total concentration 
 of plant foods dissolved in the soil solution may be very slight at the 
 time when the crop is making its greatest draughts. Not only does 
 the composition of the soil solution affect the growth of the plant, but 
 the growth of the plant reciprocally affects the soil solution. In- 
 creased quantities of certain elements may be brought into solution as 
 a result of absorption by the plant. For this reason the total quantity 
 of phosphorus found in a mature plant may be very much greater 
 than the amount present in the soil solution at any one time. 
 
 These and many other facts have led to the conclusion that the 
 really important consideration is the availability of plant foods at 
 the proper stages of plant growth. In other words, at those periods 
 when the plant needs to draw most heavily an the nutrients adequate 
 
 3 The term " soil solution" is used here to signify the soil moisture from 
 which the plant can absorb dissolved substances. 
 
4 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION 
 
 supplies must already be dissolved in the soil solution, or be capable of 
 entering into solution with sufficient rapidity. Thus the rate at which 
 the soil solution is replenished with plant foods as these are withdrawn 
 by the plant is the vital question. Unless we have some assurance that 
 these processes take place in such a manner as to meet the requirements 
 of the plant, it is of little interest to determine the total amounts of 
 the plant food elements present in the whole mass of soil. 
 
 Keferring again to the California experiments, it was found that 
 in these soils the total amounts of nitrogen in an acre-foot would be 
 equivalent to the amounts removed by large barley crops over a period 
 of 50 to 100 years, for potassium from 1000 to several thousand years, 
 and for phosphorus several hundred years. Yet the actual crop yields 
 from year to year showed that some of the soils were only one-third 
 as productive as others, and, as we have stated, the differences bore no 
 consistent relation to the total amounts of plant foods present, nor to 
 the amounts soluble in acids. After seven years' cropping, several 
 soils showed a decided decline in yield, while others maintained their 
 original production to a considerable degree. These changes also were 
 independent of the total or the acid-soluble plant foods. To illustrate, 
 after seven years' cropping, one soil produced a crop nearly three 
 times as large as that yielded by another soil of the same physical 
 character and originally containing an identical percentage of total 
 nitrogen and similar percentages of other elements. Since in these 
 experiments moisture and physical conditions were made as favorable 
 and uniform as possible, every opportunity was afforded to bring out 
 relations between the analyses of the soil and crop production, if such 
 relations existed. 
 
 It may, of course, be claimed that the total amount of any element 
 present in the soil represents a potential supply, which may be made 
 available by suitable management. But of what value is it to learn 
 that one soil contains in an acre-foot a total supply of potassium suffi- 
 cient for one thousand crops and another soil a supply for two thou- 
 sand crops when the information gives no clue to the availability of 
 this potassium during the next year or the next fifty years ? It is very 
 essential to learn just how the supply of potassium in the soil can be 
 made to dissolve in the soil solution at such rates or times as to meet 
 the requirements of the plant. No hint of the way in which this con- 
 dition can be brought about is contained in the figures indicating the 
 total quantity of potassium in the soil. Similar reasoning applies to 
 all the plant foods. 
 
 It is true that this series of thirteen soils did not include any soil 
 of extremely low productivity. It is certain that many highly infertile 
 
Circular 235] soil, analysis and plant interrelations 5 
 
 soils would also show a very low content of one or all of the important 
 plant foods. Such soils, however, are generally recognized without 
 any need of a chemical analysis. In California, at least, soils of this 
 character are seldom submitted to the chemist for examination. It 
 appears that many of the older ideas concerning the general utility of 
 soil analysis were based on comparisons of exceptionally poor soils and 
 highly productive soils. 
 
 If a soil analysis gives no indication whether or not plant foods 
 can be made available to the plant at the proper times, it necessarily 
 follows that such an analysis cannot be used to determine what method 
 of fertilization should be employed. When any soluble material is 
 added to a soil exceedingly complex chemical and biological changes 
 take place. The addition of one element may increase the availability 
 of other elements. For example, the addition of calcium to some soils 
 may increase the quantity of potassium in the soil solution. These 
 alterations in the chemical composition of the soil moisture may also 
 affect the life processes of the soil microorganisms, perhaps changing 
 the rate at which organic matter is broken down, or nitrates formed, 
 with very significant effects on the yield or quality of crops. 
 
 The conclusion is also clear that an estimate of the amounts of 
 plant food withdrawn by the crop furnishes no scientific basis for soil 
 treatment. Certain elements, such as potassium, may become ade- 
 quately available under proper conditions without any new supply 
 being added to the soil, while nitrogen in some cases may need to be 
 applied in even greater quantity than would be estimated from crop 
 withdrawals. The point to be emphasized is that neither the analysis 
 of the crop nor the analysis of the soil affords the essential information 
 concerning availability of plant food at various periods, and conse- 
 quently such data cannot serve as a reliable guide in soil management. 
 
 Turning now to another closely related phase of this subject, it is 
 sometimes claimed that different crops respond to different, although 
 very definite, fertilizer mixtures. Practically nothing is known of the 
 exact requirements of any crop, but even if such information were at 
 hand fertilizer mixtures cannot be prescribed, since the application 
 of such mixtures is likely to produce in different soils, or even in the 
 same soil under various treatments, different amounts of available 
 plant foods, and these effects are so complex that they cannot be pre- 
 dicted at the present time. 
 
 In any discussion of the relations of soils and plants reference 
 must be made to the importance of such factors as light, temperature, 
 humidity, total soil moisture, and soil aeration. If a maximum crop 
 is to be produced, there must be suitable adjustment between all the 
 
b UNIVERSITY OF CALIFORNIA EXPERIMENT STATION 
 
 factors involved. Under one set of seasonal influences a soil might 
 provide an environment in the soil moisture highly favorable to a 
 certain plant, while under other climatic conditions the availability 
 of the plant foods in the soil might be less fortunately adjusted to the 
 plant, resulting in a far smaller crop yield. Here, again, soil analysis 
 lends no aid. 
 
 There remains still another reason why soil analysis, as ordinarily 
 practiced, does not accomplish any- valuable purpose. Even if it were 
 possible to make a chemical analysis of a soil which would be capable 
 of any consistent interpretation, it would first be necessary to secure 
 a sample truly representative of the field or area in question. Soils 
 in general are known to be exceedingly variable and samples taken 
 only a few feet apart may give very different results. These conclu- 
 sions are drawn from extensive experiments initiated by the Division 
 of Soil Chemistry and Bacteriology (now part of the Division of Plant 
 Nutrition). The difficulty of obtaining samples of soil adequately 
 representing any area is so great that such sampling is impracticable 
 as a routine procedure. 
 
 No consideration has been given to alkali soils in this circular. 
 These require separate treatment, and the reader is referred to another 
 publication 1 of the Agricultural Experiment Station for a discussion 
 of soils containing excessive quantities of soluble salts. Certain chem- 
 ical tests are now being made on alkali soils when the samples are taken 
 under the supervision of the farm advisor. 
 
 Briefly restating the main point of the present discussion, we may 
 conclude that the soil should not be regarded simply as a storehouse 
 of plant foods, with a value to be assessed according to the total supply 
 of these elements present. A more accurate analogy is that of an 
 exceedingly complex and constantly changing chemical system in which 
 the chemical processes of the soil and of the plant have an intimate 
 relation. The soil miner als enter into solution at a rate which is de- 
 pendent on many factors, such as carbon dioxide production by micro- 
 organisms and by the plant, nitrate production, rates at which the 
 plant absorbs the various elements, temperature, moisture, etc. In 
 other words, the growth of the plant is determined by the nature of 
 the chemical changes taking place in the soil and plant, and not by 
 the percentage composition of the soil. No method of analysis appli- 
 cable to individual samples of soil can yield the kind of information 
 which it is essential to possess in order to draw any useful or reliable 4 
 conclusion concerning plant growth. 
 
 i The Present Status of Alkali, by W. P. Kelley, Circular 219. 
 
Circular 235] S0IL ANALYSIS AND PLANT INTERRELATIONS 7 
 
 Notwithstanding the exceedingly difficult nature of the problem, 
 progress has been made in the field of plant nutrition. The hope of 
 a better understanding of the principles of plant growth and of the 
 interrelations of soils and plants depends upon continued scientific 
 investigation. By this is meant much more than field trials of various 
 fertilizers. A real comprehension of the nature of the soil as a medium 
 for plant growth must be based on a knowledge of the laws of chemistry 
 and physics and biology as applied to the soil and to the plant. 
 
 Meanwhile practical advice concerning specific problems in the 
 management of soils must in general come from the farm advisor. 
 Because of the many factors, both agricultural and economic, which 
 must be taken into consideration, a knowledge of local conditions is 
 ordinarily indispensable to any plan for soil treatment.