Tải bản đầy đủ - 0 (trang)
Chapter 2. The Soil Organic Fraction

Chapter 2. The Soil Organic Fraction

Tải bản đầy đủ - 0trang



matter literature into a coherent whole, much practical information on

the subject has been accumulated, and its importance in relation to soil

fertility and physical condition is widely recognized. Many of the

“what ? ” questions have been answered, although most of the “hows?”

and “whys?” remain to be explained. It is known that the organic

fraction has a profound effect upon the structure of soils and that the

deterioration of structure which accompanies intensive tillage is usually

less rapid in soils of relatively high organic content. The absorption

and retention of water, the reserves of exchangeable bases, the capacity

to supply nitrogen, phosphorus, and some of the minor elements to growing plants, and the adequacy of aeration and other properties of soils are

all dependent in some degree upon the organic fraction.

In recent years there has been a tendency in some quarters to attribute very remarkable properties to soil organic matter, and claims have

been made for its effects on plant growth which border on the fantastic.

For .the most part these claims are without basis in fact and are not supported by the sort of experimental evidence which can bear close scrutiny: Since the small amounts of organic matter in mineral soils exert

an influence on chemical and physical properties far out of proportion

to their percentage by weight, there is ample justification for emphasis

on the importance of this fraction in practical agronomy as well as in

agronomic research, without attributing to it the properties of a panacea

for all causes of poor plant growth,



As a component of soils the organic fraction is the resultant of a

number of complex formation processes characteristic of the environment in which it is found. Residues of the vegetative cover might be

termed the parent material from which it is formed. By means of two

processes, both of which are due chiefly to the activities of microorganisms, organic matter characteristic of a given soil is produced. The first

of these processes is decomposition of plant residues with resultant modification of their chemical composition and properties; the second is

syfithesis of new microbial cells which in tu rn die and are decomposed

by other microorganisms. That the type of plant residues returned to

the soil varies greatly from one place to another is obvious to everyone;

what is not so obvious is that the microbial population of soils varies not

only with geographical location but also with position in the profile.

For example, Gray and McMaster (1933) found biological activity in

the leached horizons of certain Podzols to be only about 4 per cent of

that in the organic horizons, and Gray and Taylor (1935) were able to



demonstrate marked contrasts in biological activity a t different levels

between Podzols and a virgin clay soil. Comparison of the potential

activity of populations in profiles of several soil types led Newman and

Norman (1941) to conclude that the soil population is directly a characteristic of its environment and not readily subject to change. The

data of Vandecaveye and Katznelson (1938, 1940) support this conclusion.

Plant materials from which soil organic matter is formed have some

similarity in their major structural constituents and in the rates a t which

they decompose. Numerous investigators have confirmed Hebert’s

(1892) early observation that lignin and protein are the fractions which

accumulate after the other components are largely decomposed, although

it is improbable that these are unchanged themselves during the decomposition process. Alterations are effected which may make the lignin

and protein quite unlike their counterparts in undecomposed plant


During breakdown of plant residues in soil large numbers of microbial cells of various kinds are elaborated which exist for only a short

time before being themselves subject to decomposition by other microorganisms. The proportion of the soil organic fraction consisting of

living microbial cells is very difficult to determine, but Russell’s (1950)

estimate of 1-2 per cent of the organic matter is probably conservative.

Certainly a much larger fraction of the total is microbially derived and

somewhat different from plant residues in its chemistry.

I n view of the diversity of the organic materials which find their way

into the soil and the variety of soil populations which also contribute

to what is commonly called humus, it would be surprising indeed if the

soil organic fraction everywhere had a common composition. The fragmentary nature of soil organic matter literature is no doubt due in part

to the fact that this fraction in a Podzol, for example, may be very different from that in a Prairie Soil or a Sierozem, even though called by

the same name in each case.



1 . W i t h Respect t o Climate m d Vegeta,tion

The quantity of organic matter in soils as a function of climate has

been studied by Jenny (1930, 1941), and the subject has been reviewed

recently by Ensminger and Pearson (1950) in Volume I1 of this series.

The relationships expressed by Jenny are applicable to loamy grassland

soils of the central United States; in general terms the Jenny formula

indicates that a t constant temperature soil nitrogen increases logarith-



mically with increasing moisture, whereas if moisture is kept constant,

soil nitrogen declines exponentially as the temperature rises. Jenny

(1941) points out that such soil-climate functions enable one to visualize

approximate major trends at the expense of local details. When applied

to specific situations, the theory frequently does not hold; in general, it

is not applicable where the dominant influence of climate on plant growth

and organic matter decomposition is not clear-cut, as, for example, when

soils are brought into cultivation and the soil-climate equilibrium is disturbed. Soils long in cultivation eventually attain a new equilibrium

level of organic matter which is usually, though not always, lower than

that which existed in the virgin state. In Europe, where most of the

arable soils have been in cultivation for a long time, the climate-organic

matter relationships are not at all apparent. Soil drainage is also an

important consideration in any local situation. Where drainage is poor,

organic matter accumulates regardless of climatic effects, owing to retardation of the decomposition processes.

Many tropical soils in particular do not have the very low organic

matter levels which would be predicted by southward extrapolation of

the soil-climate relationships found in the United States. Jenny et al.

(1949) observed that many Colombian and Costa Rican soils are rich in

nitrogen and organic matter, particularly in comparison with California

mountain soils. They found that the annual production of organic matter in the form of leaves and twigs was much greater in tropical than

in California forests ; however, decomposition rates of Colombian forest

soils were calculated to be much more rapid. In a later publication

Jenny (1950) attempted to reconcile these observations with his earlier

work on United States soils by explaining that, whereas decomposition

of surface litter in tropical soils proceeds rapidly, that of organic material which becomes incorporated in the mineral soil appears to be slow.

This, combined with the abundant production of vegetation, would account for organic matter accumulation within the profile. Jenny suggests calcium deficiency as a possible explanation for the inhibiting

effect on decomposition. Ensminger and Pearson (1950) have suggested

that the available phosphorus content of these soils may be too low to

support a n active microbial population, It is difficult to conceive of a

situation in which nutrient elements in surface litter are adequate to

permit rapid decomposition, while in the soil directly underneath, from

which the leaves and twigs originally obtained their nutrients and to

which the nutrients are returned after decomposition of the litter, there

is an acute shortage of these nutrients. The fact that these soils can

produce luxuriant vegetation is incompatible with the view that microorganisms are inhibited by nutrient deficiency.



Recently Smith et al. (1951) have called attention to the relatively

high organic matter contents of Puerto Rican soils, which compare favorably in this respect with the best soils of the temperate regions. The

high organic matter levels can be maintained even under intensive land

use. These authors believe that the absence of killing frost is chiefly

responsible for the observed organic matter levels, since this favors production of plant material more than microbiological decomposition processes. This appears to be the most plausible explanation so f a r advanced

to account for organic matter accumulation in tropical soils.

2. With Respect to Horizon within the Profile

The quantity and nature of organic matter occurring a t various

depths in soils are dependent upon a number of environmental factors.

Of these the most clearly related are probably rainfall and type of vegetative cover. Where the annual addition to the soil organic fraction



5ot '














FIQ.1. Distribution of organic matter in profiles representing several great soil


1. Byera e t al. (1935) 2. Brown and Thorp (1942) 3. Byers et al. (1935)

4. Hopper e t al. (1931) 5. Brown and Byers (1935)

takes the form of leaf fall the layers of accumulation are a t and near

the surface, with a very sharp decrease below the shallow surface layer,

indicating that downward movement is slight. Such compounds as are

sufficiently soluble to be leached down the profile in humid region soils

may also be readily assimilated by soil organisms and decomposed, and

on this account do not accumulate to any extent. On the other hand,

under grassland vegetation most of the subsoil organic matter is formed



in place from root residues. I n consequence organic matter in quantity

is found a t greater depths and the rate of decrease down the profile is

gradual. Figure 1, taken from the data of several investigators, shows

the organic matter distribution i n a number of soil profiles. The data

used in the figure were obtained from carbon contents and use of a single

conversion factor, and on that account do not give an entirely accurate

estimate of the vertical distribution of total organic matter, since the

carbon content of the organic fraction usually decreases with increasing

depth; however, they are useful for purposes of comparison.

The work of Jenny (1941) dealing with the effect of rainfall on nitrogen-depth relations demonstrated that the greater production of vegetation in moist climates is reflected through the entire profile.


1. Polysaccharides

As precursors of soil organic matter, the constituents of plant tissue

may be divided into three general groups which together comprise the

major portion of mature plants: (1) polysaccharides, (2) lignin, and

(3) protein. Of these the polysaccharides are by far the most abundant,

cellulose alone usually amounting to about half the dry weight of tissue.

The other major polysaccharides are the hemicelluloses, including polyuronides. Norman (1931, 1933) has shown that the hemicelluloses suffer

very early and extensive loss during microbial decomposition, followed

by increasingly rapid removal of cellulose. After the initial stages in

decomposition most of the weight loss is due to breakdown of cellulose.

On the basis of their susceptibility to microbial attack, the principal

structural polysaccharides of plants are therefore not likely to accumulate in soils. It is difficult to estimate how much of the organic fraction

in soils consists of cellulose, hemicelluloses, and simpler polysaccharides

since suitable analytical methods are not available for the purpose.

Waksman and Stevens (1930) applied to soil systems a proximate

method of analysis which had been used a great deal in the study of

plant residue decomposition and obtained figures of the order of 10 per

cent of total organic matter as polysaccharides. By applying this type

of analysis to some of the soils of northeast Scotland, Forsyth (1948)

obtained figures of the order of 0-5 per cent cellulose, 5-20 per cent

hemicelluloses, and 5-15 per cent uronic anyhydride. Although the fractions determined by these empirical methods undoubtedly do not give

a good separation of components, it is obvious that soil organic matter

differs considerably from plant material with respect to polysaccharides.



I n view of the susceptibility to decomposition of the major polysaccharide constituents of plants, which appear to function chiefly as an energy

source €or microorganisms, Forsyth (1948) believes that any direct contribution of plants to soil polysaccharides must come through the polyuronide hemicelluloses.

The characterization of soil polysaecharides is far from complete, but

it appears that the polyuronides occupy an important position among

them. Uronic acid units apparently are less susceptible to decomposition

than are pentosans and hexosans associated with them in the hemicellulose group, as shown by Waksman and Reuszer (1932). Norman and

Bartholomew (1943) observed that 10-15 per cent of the organic carbon

of surface soils they analyzed was present in uronide groupings and that

the proportion of uronic to total carbon increased with depth. Why

these sugar acids should accumulate in soil is not clear, since polyuronide

gums are readily subject to microbial attack (Norman and Bartholomew,

1940). These workers suggested that uronic acids are stabilized by combination with some other grouping to give a resistant complex. It may

well be that the situation is analogous to the accumulation of organic

nitrogen compounds in soil, which are not only stabilized by association

with clays, possibly with lignin, but are also continually being synthesized in microbial tissue. Fuller (1946, 1947) has adduced evidence of

the microbial origin of soil uronides, although this is indirect, since it is

based on similarities between decarboxylation rate curves of soil organic

matter and bacterial gums.

Estimates of the quantity of uronic carbon in soils have been criticized by Bremner (1950a) in a recent review on grounds that the method

of determination is not specific for the uronic carboxyl grouping, giving

results which he considers to be impossibly high. His contention is that

the conventional method, involving prolonged boiling with 12 per cent

hydrochloric acid, splits off carboxyl groups of nonuronic origin, although these do not interfere seriously when the method is applied to

plant materials. The objections raised by Bremner should be considered

in any attempt to assess the quantitative importance of uronic acids in

soils, but on the other hand, the accumulation of data indicating the presence of relatively large amounts of uronides cannot be overlooked. One

piece of circumstantial evidence which has not been emphasized previously is the fact that the carbon content of organic matter decreases with

depth, whereas the uronide content increases. Since the carbon content

of the organic matter of most surface soils is in the vicinity of 50-52

per cent, whereas that of pure uronide is 40.9 per cent, a n increase in

the proportion of uronide would lower the carbon content of the organic

fraction. Coupled with this is the fact that the exchange capacity of




organic matter also increases with depth, indicating the presence of more

acidic groups.

Forsyth (1948, 1950) and Stevenson et ccl. (1952) have been able to

isolate and identify uronic acids in the hydrolyzates of soil organic matter extracts by means of paper chromatography, thus providing unequivocal evidence of their presence. Forsyth (1950) determined the

constituent sugar units in two soluble polysaccharides obtained from a

“fulvic” fraction representing 1-2 per cent of the total organic matter.

Of these polysaccharides 15.8 and 16.9 per cent, respectively, were present as. uronic anhydride, which would account for 0.15-0.34 per cent of

the soil organic fraction. However, Forsyth regarded his yields as minimal.

The positive identification of uronides in soil organic matter, together

with the isolation by Forsyth and Webley (1949) of a considerable

number of soil bacteria capable of synthesizing polysaccharides containing uronic acid units, gives further support to the belief that this group

constitutes an important fraction in soil.

The carbohydrate content of soil microorganisms is subject to considerable variation, depending on the age of the cells, the amount and

nature of available substrate, and other factors. Cellulose appears not

to be a major structural constituent, although its presence has been

demonstrated in the cell wall of a few bacteria of the acetic acid group

(Brown, 1886; Beijerinck, 1898) and in several species of fungi (DeBary, 1887; Mangin, 1899 ; Thomas, 1928). Chitin, a polymer of N-acetylglucosamine which is analogous to cellulose in many respects, is a

structural component of many of the filamentous fungi. Schmidt (1936)

isolated chitin from fifteen species of fungi in yields up to 4 per cent of

the dry weight of mycelial tissue. Norman and Peterson (1932) reported a nitrogen content of only 3 per cent in the alkali-resistant fraction of Aspergillus fisherii as compared with 6.9 per cent in pure chitin,

which would indicate the presence of other polysaccharides linked to

chitin or occurring as infiltrating substances in the structural fabric.

The presence of amino sugars in soil, observed by Bremner (1949a), confirms the occurrence of chitin o r its degradation products.

A large number of microorganisms have the capacity to synthesize

polysaccharides, the formation of which may be endocellular, capsular,

or exocellular, and frequently may be a major metabolic product, according to Evans and Hibbert (1946). Forsyth and Webley (1949) found

that bacteria capable of polysaccharide synthesis were present in various agricultural, moorland, and forest soils, and estimated that these

may form 5-16 per cent of the viable bacterial population. Chemical

examination of the polysaccharides produced distinguished four types,



in three of which uronic acids occurred i n combination with other types

of sugar units, presumably in the form of mixed polymers. Fungal

polysaccharides have not been extensively studied, but it is probable

that they contribute substantially to the soil organic fraction i n view

of their known synthetic abilities and of the considerable proportion of

fungal tissue in relation to other forms. A piece of fungal mycelium

5p in diameter and 1 em. long would be roughly equivalent in weight to

a million small bacteria of the type found in the indigenous soil population.

2. Lignin-Derived Fraction

In the decomposition of plant residues in soil, cellulose and hemicelluloses are readily utilized by a wide variety of soil microorganisms as a

source of energy, whereas lignin is broken down only slowly. As a consequence the relative proportion of lignin in the residue increases ; this

fact is one of the chief supports for the belief tha*t a large proportion

of soil organic matter is either lignin or lignin-derived. Further support is provided by the presence in soil of a fraction which, like lignin,

may be dissolved in alkali and subsequently precipitated by the addition

of excess acid, is resistant to hydrolysis by strong mineral acid, contains

methoxyl and phenolic hydroxyl groups, and is attacked by relatively

mild oxidizing agents. Gottlieb and Hendricks (1945) attempted to

obtain more direct evidence of a lignin fraction in soil organic matter

through the application of alkaline nitrobenzene oxidation and highpressure hydrogenation, techniques which have been very helpful in

elucidating the structure of wood lignin. They were unable to isolate

definitely characterizable products from soil organic matter preparations,

in contrast to the situation with wood lignin, from which propyl benzene

derivatives can be obtained in good yield. This indicated drastic alterations in plant lignin during the course of decomposition in soil. On the

basis of similarities between alkali lignin and soil organic matter preparations in their behavior toward hydrogenolysis, these workers hypothesized a condensation of demethoxylated lignin molecules with the

production of fused ring structures. In relation to this point it may be

noted that Scheffer and Welte (1950) observed marked similarities in

the ultraviolet absorption spectra of alkali lignin from several sources

and soil humic acid preparations.

In addition to the probable conjugation of ring structures in lignin

during decomposition, there is ample evidence of other changes in the

molecule which differentiate the lignin-derived fraction in soil from the

unaltered constituent in plants. Methyl groups are split off, as shown

by decrease in methoxyl content (Waksman and Smith, 1934). Cation



exchange capacity increases (Millar et al., 1936 ; Bartlett, 1939), probably owing to oxidation of side chains to carboxyl and exposure of phenolic hydroxyl groups by demethylation.

Unlike the polysaccharides, lignin is not synthesized by soil bacteria,

but Thom and Phillips (1932) have reported the presence i n fungi of

large amounts of a fraction resembling lignin in its resistance to strong

acid hydrolysis. This fraction, varying from 2.65 to 54.08 per cent in the

species they analyzed, differs from lignin in higher plants in that the

methoxyl content is negligible. However, this does not exclude the possibility that the fungal constituent is structurally similar to lignin.

More recently Pinck and Allison (1944) reported synthesis of ligninlike

complexes by several species of fungi.

It appears, then, that the resistant portion of soil organic matter is

somewhat unlike lignin, though resembling it in several respects. The

carbon content alone serves to give an indication of this, since wood

lignin usually contains more than 60 per cent, whereas organic matter

of surface soils rarely contains more than 52 per cent, and the value may

be much less in subsurface layers. A fresh approach is needed in which

the microbial origin of a large part of the soil organic fraction is given

full consideration. This is not to say that techniques and methods developed in the study of plant lignin need be rejected, although past

experience indicates that these frequently cannot be used without modification. Rather this is an argument that the field of microbial chemistry

is one to which soil scientists might profitably give more attention. The

extrapolation of plant chemistry to cover this field has been useful in

organic matter research, but interpolation between the properties of

plants and of microorganisms might come closer to the true situation.

3. Organic Nitrogen Praction

Nitrogen accounts for something like 5 per cent of soil organic matter

and must occur as an integral part of many of the compounds present,

since it is extremely difficult to obtain any sort of soil extract free from

nitrogen. It occurs in fairly constant proportions in the organic matter

of soils of very diverse character ;indeed, Read and Ridge11 (1922) found

the nitrogen content to be more constant than that of carbon. Much of

the organic nitrogen is unquestionably protein-derived since proteins are

present in plant residues and microbial protoplasm, although adequate

experimental verification for this inference has been obtained only recently. Kojima (1947a, b ) was able to account for 37 per cent of the

total nitrogen in a muck soil as alpha-amino nitrogen and isolated several common amino acids in good yield. However, she concluded that

not more than 66-75 per cent of the organic nitrogen could be consid-



ered as protein. Comparative uniformity in amino acid composition

was indicated by the work of Bremner (1950b), who found the tame

twenty amino acids in hydrolyzates of six soils. No free amino acids

were detected before hydrolysis. Bremner (1949a) considered one-third

of the total nitrogen in soils in proteinlike combination to be a minimal

figure. If generalization may be permitted on the basis of the work

cited, roughly between one-third and two-thirds of the soil organic nitrogen occurs in the protein-derived fraction. Information concerning the

manner of combination of the remainder is very meager, but some of it

occurs in heterocyclic ring compounds such as the purines and pyrimidines. Bremner (1951) estimates that not more than 10 per cent of soil

nitrogen occurs as nucleic acids. Nitrogen containing rings must also

form a part of the fused ring structures of the lignin-derived fraction.

Flaig (1950) has suggested the similarity to soil humic acids of melanin,

which appears to form a chain structure through conjugation of 5, 6dioxyindole rings. The presence of glucosamine in soil hydrolyzates

indicates that part of the nitrogen is present in the carbohydrate fraction.

The question of a ligno-protein complex in soil has not yet been resolved, but so f ar the evidence in support of the existence of such a

complex is only presumptive at best. The idea was advanced to account

for the apparent resistance of soil nitrogen to microbial decomposition,

since free proteins are readily attacked. I n view of the accumulation of

lignin due to its relatively unreactive character, it seemed reasonable to

postulate a masking effect of resistant lignin upon easily decomposed

protein, Certain superficial similarities were observed between a synthetic ligno-protein (Waksman and Iyer, 1932) and soil humus. The

postulated mechanism of combination does not adequately explain the

resistance to microbial attack, as pointed out by Norman (1942), nor the

reactivity toward hydrogen peroxide of soil organic matter (McLean,

1931). Alternative explanations have been offered by Ensminger and

Gieseking (1939, 1942), who found proteolysis to be inhibited in the

presence of clays, and by Broadbent and Norman (1946), who obtained

evidence that some of the soil organic nitrogen is quite readily mineralized. I n view of the evidence that the acid-resistant fraction in soils

is unlike plant lignin in many of its properties and that a large portion,

perhaps more than half in some cases, of the soil nitrogen is of nonprotein nature, it would seem that the lignin-protein idea in its original

form is now obsolete. Mattson and Koutler-Andersson (1943) suggest

that some of the stable nitrogen compounds in soil might be produced

by interaction of oxidized lignins and ammonia or possibly aromatic

amines a t the sites of phenolic hydroxyl groups to form amidophenols,



which upon oxidation and condensation would form a polymer containing ring nitrogen. The proposed reaction seems tenable and is partially

substantiated by the finding of Bennett (1949) that when hydroxyl

groups in oxidized lignin were blocked by methylation, very little nitrogen was assimilated into the molecule upon treatment with ammonia,

whereas more than 7 per cent nitrogen could be “fixed” by an oxidized

commercial lignin.


Fractions Obtained by Empirical Fractionation Procedures

One of the properties of soil organic matter which makes its characterization difficult is its inherent insolubility. No solvent is known which

dissolves the complex completely, Strong alkali brings a considerable

part of the material into solution and on that account has been used

extensively in research dealing with so-called humic and fulvic acids,

which are, respectively, the precipitate and filtrate resulting from the

addition of excess acid to an alkali extract. All extractants with an

alkaline reaction suffer the disadvantage that the material is concurrently oxidized and probably altered in other ways, so that the preparations thereby obtained are t o some extent artefacts of the method.

A procedure popular in Germany divides soil organic matter into

two fractions, depending on solubility or insolubility in acetyl bromide.

However, Springer (1943) and Siege1 (1941) reported inability to obtain

clear relationships between the relative amounts of organic material insoluble in this reagent, presumed to be the true humus, and other soil

properties. Other separations have been based on peptization in sodium

chloride (Sowden and Atkinson, 1949) and on density (Henin and Turc,

1950)) but whether these represent an improvement over other empirical metlods has not yet been clearly demonstrated.

In an attempt to get away from the difficulties of the classical extraction with strong alkali and the attendant drastic effect on soil organic

matter, Bremner and Lees (1949) investigated the extracting ability of

a number of inorganic and organic salts of sodium. Of these the pyrophosphate proved t o be the most satisfactory, having an extracting efficiency in tenth-molar solution comparable to that of fifth-molar sodium

carbonate. The chief disadvantage of this reagent appears to be that

it removes only a small proportion of the total organic matter, on the

order of 10 per cent in three of the four soils Bremner and Lees investigated, though the percentage was somewhat higher in the fourth. However, the more drastic extraction by strong alkali is by no means complete, so that the advantage of obtaining relatively unchanged material

through the use of pyrophosphate probably outweighs the disadvantages

of low yield and selective solution, Neutral extractants will probably

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Chapter 2. The Soil Organic Fraction

Tải bản đầy đủ ngay(0 tr)