DETERMINATION OF THE COMPLEX OF CELL WALL
SUBSTANCES IN PLANT PRODUCTS

L. Paloheimo and K. A: Vainio

Department of Animal Husbandry , University of Helsinki

Received September 30, 1965

In this department Paloheimo and Paloheimo (7) have developed a method
for the determination of the complex of vegetable cell wall substances. The main
principle of the method lies in boiling the sample with 0.05 N hydrochloric acid
(pH about 1.3). The residue is extracted with ethanol-benzene and the final residue
corrected for ash and protein. It was proved (7, p. 9) that in materials containing
a high percentage of cell wall substances mere boiling with water gives about the
same results as the boiling with 0.05 N acid. However, the filtration after boiling
with water is tedious and as the said authors endeavoured to develop a method
practicable even for starch containing materials, they preferred the boiling with
acid. The acid concentration was chosen to be just sufficient for hydrolysing the
starch to a filterable suspension. However, it appeared that, in spite of the very
low acid concentration, a considerable amount of the cell wall substances was
dissolved. For instance, of the pentosans in white clover grass about 40 % was
dissolved (7, p. 10) and according to Paloheimo et al. (8, p. 52) the pectins were
dissolved by the 0.05 N acid. It was, however, supposed that the easily hydrolys-
able membrane substances may be digested in the alimentary canal even in animals
with weak bacterial function. Thus the method proposed could be recommended
for an estimation of food value and it is at any rate biologically more satisfactory
than the Weende crude fibre determination or the 1-N-acid fibre determination.
Nor is the method too conventional: a doubling of the acid concentration or of the
boiling time has only a slight effect on the analysis results.

Salo (9) has developed a very useful and practical analysis system for the deter-
mination of carbohydrates and lignin in foods and faeces. Adding up the percent-
ages of cellulose, neutral sugar hemicellulose, uronic acid hemicellulose, and lignin
she obtains the percentage of total cell wall substances. Salo has even determined
the cell wall complex according to Paloheimo and Paloheimo (10, p. 132). In a com-

https://www.c-info.fi/en/info/?token=U9G6FrYd2aOq3Y30.ohjfpkrGUs6QAFULyumxlg.J2iSyUQARzIz_IExQJ_ByhUEBqFv4dp-iKedyoQ8pQXFGXMIVst_ODYPhJv56yVZPrsflMVEC-f8U9XNlpMwbM7BAIF2aPPpm73nMJ2ikyr5pNMUiOScjZZN6r1GK7zNZwkdciD2qXTJikIJufXiygOGOSQCMIZGKINUvyXTrYRoJa24L7IaH72Hnw


306

prehensive table (10, p. 128) it is evident that the relation between Sale’s and Palo-
heimot cell-wall-complex percentages differs rather widely. Only in a few cases has
Paloheimot method given higher results than that of Salo. In materials with a low
cell-wall-complex the former method usually gives considerably lower results than
the latter.

In the present paper we are describing a new method for the determination of
the plant cell wall complex which is built upon quite a different principle to the
method of Salo. However, the results obtained with these two methods are well in
accordance.

The principle of the new method

The material to be analysed is first extracted by boiling it in 80 % ethanol, then
by boiling in absolute ethanol and finally with water in room temperature. The
insoluble residue is weighed and the loss of ignition determined.The protein correction
is made and, if the material contains starch, even the correction for starch. The
boiling in 80 % ethanol extracts the sugars, most of the organic acids and their
salts, and the main part of the lipids and other substances, like chlorophyll, extract-
able with lipid solvents. The boiling in absolute ethanol completes the extraction
of the lipids etc. The extraction with water makes the residue free from fructosane
and some other cell enclosure substances which have not been extracted by ethanol.
The corrected residue is called by us the complex of cell wall substances or simply
the cell wall substances. In fact, plant cell walls contain also inorganic salts and
Si02 and thus the term organic cell wall substances would be more accurate.

As early as 1813 a similar principle was applied by Humphry Davy (1, p. 116).
In his book Elements of Agricultural Chemistry Davy presents a wood fibre determi-
nation in which the sample is alternately boiled in water and alcohol. No correction
for ash, protein or starch was used. Evidently the method was applied only to very
rough plant materials. Davy suggests that the composition of the wood fibre varies
in different plants and plant organs.

Details of the method

If the material is not to be extracted in the wet stage it is dried in vacuum
at 70° C. About 2 g is weighed and put into a 500 ml Erlenmeyer flask. 200 ml 80 %

ethanol is added and the flask is boiled under reflux for 2 hours. The contents are
then filtered using a Whatman No. 4 paper and a funnel heater. The residue is rinsed
into the original flask, 200 ml 80 % ethanol is added, and the flask boiled again
for 2 hours. The filtration etc. as above. Now follows the boiling with 200 ml absolute
ethanol. After the filtering the residue is washed with hot ethanol and rinsed with
the same solvent into the Erlenmeyer flask. The ethanol is evaporated out of the
flask on a water bath, 200 ml destilled water is added, and the flask is shaken for
2 hours. The suspension is filtered through a Whatman No. 4 paper the dry weight
of which is known. After drying, the paper with the residue is weighed and inciner-
ated. The final magnitude of the organic residue is obtained by subtractions. For



307

protein correction another residue is prepared and its N-content determined. If the
sample contains starch, the starch determination is made of a third residue. In
starch determination we have used the iodine colorimetric method of Paloheimo
(3, 4,5, 6). In this determination the wet residue is used together with filter paper.

Results and discussion

Table 1 shows the results obtained with the new method (a) compared with those
obtained with the method of Salo (b) and the 0.05 N acid method (c). The values
b and c are taken from Salo’s tables and the values a are results of our analyses made
of the same samples which Salo has used. The table shows that in many cases the
results a and b are very similar in spite of the difference in the principle of the
methods in question. However, with the materials 2, 3 and 4, which are very rich
in cell wall substances, the method of Salo (b) has given appreciably lower results

Table 1. The complex of cell wall substances in some plant materials determined with different methods.
(% of dry matter.)

abc
The new Salo's 0.05 N acid
method method method

1. Lichen (Cladonia rangiferina) 81.5 80.4 69.1
2. Marrow stem kale vascular bundles 76.0 67.9 74.2
3. Moss (Sphagnum recurvum) 71.1 59.4 67.3
4. Red clover, blooming, lower halves of stalks 66.1 60.4 64.5
5. Sugar beet pulp 57.9 66.4 37.8
6. Sedge (Carex gracilis) 57.8 58.3 52.4
7. Meadow grass (Poa pratensis), pasture stage 54.4 52.6 45.0
8. Red clover, blooming 48.9 50.1 38.9
9. Horsetail (Equisetum pratense) 46.9 49.5 32.2

10. Marrow stem kale, stalk rind 43.9 44.1 25.9
11. Wheat bran 43.3 41.7 20.0
12. Red clover heads 43.2 46.0 29.5
13. Fern (Dryopteris linnaeana) leaves 42.0 48.1 28.2
14. Birch leaves 38.2 41.9 23.5
15. Timothy, leave stage 37.9 36.7 30.9
16. Lucern grass 36.8 40.5 26.7
17. Aspen leaves (Populus tremula) 34.8 38.9 24.6
18. Marrow stem kale leaf blades 32.9 34.5 21.4
19. Chickweed (Stellaria media) 32.3 31.0 21.0
20. Swede leaf blades 31.0 35.1 18.2
21. Red clover leaves 29.1 36.2 16.6
22. Alder leaves (Alnus incana) 28.9 33.3 15.9
23. Sugar beet tops 24.8 26.2 15.5
24. Swedes, peeled 22.1 28.9 10.2

than the new method (a), and it is noteworthy that they are even lower than the
results obtained with the 0.05 N acid method (c). Evidently the said materials



308

contain some cell wall substances which have escaped in the processes of method b.
Salo (9, p. 58) points to the general untenability of the lignin determinations.
Especially materials 5 and 21 have given higher results when method b has been
used. This can be due to the fact that there are in plant materials uronic acid an-
hydrides and even other hemicellulose substances which are dissolved during the
80 % ethanol and cold water extractions of the method a, while these components
fall into the complex of cell wall substances when Sale’s principle is followed. In the
view of the authors no substance soluble in ethanol or cold water can be included
among the cell wall substances. It is also worth mentioning that Salo has operated
with crude lignin and therefore her lignin percentages are I—21 —2 units higher than
the corresponding true lignin figures.

As for the figures obtained with the 0.05 N acid method (c), most of them are
appreciably lower than those obtained with the new method (a). If we take it that
our new method gives a fairly reliable picture of the magnitude of the cell wall
complex, the conclusion seems justified that many plant materials contain in their
cell walls a considerable portion of substances which are converted in soluble form
by 0.05 N hydrochloric acid. Using the figures of Table 1 it is possible to calculate

r
this soluble fraction: • 100. It appears that in materials 2 and 4 the percentage

a
is only 3 but in materials 11 and 24 about 55.

In Table 1 the materials are arranged in order of declining a-values. It appears
that the order of c-values is not the same although the general tendency is similar.
The exceptional c-values, such as those for the materials 5 and 11, suggest that
possibly the 0.05 N acid method is more useful for an estimation of food values.
But for the plant chemist the new method (a) must be considered as the most
reliable one.

Table 2. Fractions dissolved by different extractions.
(% of dry matter)

80 % ethanol absol. cold
Ist extr. 2nd extr. ethanol water

Oat shoots 45.2 2.5 0.3 5.2
Red clover, blooming 33.2 1.7 0.2 3.8
Lucern grass 35.3 2.8 0.3 5.3
Red clover leaves 35.4 2.4 0.4 6.6
Swede leaf blades 40.2 0.3 8.6
Alder leaves 42.9 0.3 2.6
Birch leaves 36.5 2.2 0.5 3.6
Dryopteris linnaeana leaves 33.3 0.4 4.2
Equisetum pratense 24.3 2.8 0.4 8.1
Swedes, pleed 67.0 1.7 0.2 4.3

Table 2 presents some intermediary results which show that plant materials
may contain rather abundant quantities of substances soluble in 80 % ethanol.



309

The first extraction is very efficient and the two 80 % ethanol extractions together
seem to leave in the sample very few substances extractable withabsolute ethanol.
Evidently the bulk of lipids is extracted by 80 % ethanol. One can further see that
the cold water extraction is absolutely necessary. The amount of the water extract
varies considerably. In young timothy it was 9.8 % and in chickweed 11.7 % while
in Cladonia rangiferina the percentage was only 0.8.

Table 3. Ash and crude protein in different fractions, % of the fraction.

AU ethanol Water Final
extract extract residue

Ash Cr.prot. Ash Cr.prot Ash Cr.prot.

Oat shoots 11.3 18.2 56.7 10.5 5.8 24.6
Timothy, leaf stage 11.5 11.6 12.3 5.0 2.9 27.7
Lucern grass 18.8 25.5 36.6 11.4 4.2 29.4
Red clover, blooming 12.7 11.3 28.0 9.3 2.8 17.5
Red clover leaves 11.5 10.7 29.3 8.8 3.9 42.9
Red clover heads 8.1 18.3 34.7 12.1 3.3 21.1
Sugar beet tops 17.7 9.2 24.4 3 5 7.1 26.7
Chicken weed 36.7 27.8 47.4 3.6 7.7 33.6
Carex gracilis 13.7 14.2 41.4 9.5 2.5 18.8
Swedes, peeled 5.9 10.0 20.7 5.6 3.2 10.8
Potatoes, peeled 29.9 25.3 24.0 4.5 0.3 4.2
Sugar beet pulp 4.3 15.4 6.7 4.4 5.0 15.3
Brewers'grains 3.3 14.3 21.1 10.5 4.0 22.8
Wheat bran 8.0 17.4 38.6 12.7 4.7 15.9

Table 3 shows the ash and crude protein contents in different fractions. The
water extract is in all cases, except in the potatoes, the fraction richest in ash, while
the ash content in the final residue is the smallest. The latter circumstance is impor-
tant because it diminishes the source of error connected with the ash correction.
(It should be remembered that the ash is not an innate component in organisms or
their fractions but an artificial product of analytical procedures.) The crude protein
content is lowest in the water extract and highest, with a few exceptions, in the
final residue. Thus the protein correction in our method is rather large and, if the
coefficient used in the calculation of the crude protein content is not correct, this
correction can result in an erroneous percentage of the cell wall complex. However,
this risk is not very large because the residue after the extractions scarcely contains
other nitrogenous substances than true protein. In calculating the crude protein
we have used the conventional coefficient 6.25. Perhaps this is not the most valid
one for plant true protein.

From Table 4 it appears that usually the ethanol extracts contain more than
half of the total ash content. This indicates that the salts of the organic acids have
been extracted mainly with 80% ethanol. In sugar beet pulp and brewers’ grains,
which are leavings after water extraction processes, nearly 3/4 of the ash is found



310

Table 4. Partition of ash and crude protein into different fractions.

% of total ash % of total cr.protein
in in in in in in

ethanol water ethanol water
extracts extract residue extracts extract residue

Oat shoots 48.8 26.4 24.8 41.8 2.6 55.6
Timothy, leaf stage 59.7 17.4 22.9 21.0 2.5 76.5
Lucern grass 62.7 17.0 20.3 36.3 2.3 61.5
Red clover, blooming 61.7 14.7 23.6 26.3 2.3 71.4
Red clover leaves 51.5 22.8 25.6 14.3 2.0 83.7
Red clover heads 49.4 21.6 29.1 36.1 2.4 61.4
Sugar beet tops 69.6 11.7 18.7 33.5 1.6 64.9
Chicken weed 54.7 25.6 19.8 31.9 1.5 66.6
Carex gracilis 55.6 15.1 29.3 20.2 1.2 78.6
Swedes, peeled 70.2 15.3 14.5 69.6 2.4 28.0
Potatoes, peeled 82.8 11.4 5.8 50.2 1.4 48.4
Sugar beet pulp 12.7 14.7 72.5 16.3 3.5 80.2
Brewers'grains 13.8 14.7 71.5 12.5 1.5 86.0
Wheat bran 24.2 22.2 53.6 19.4 3.1 77.424.2

in the final residue. Further, the table shows that in most cases the bulk of the
crude protein is retained in the residue. Swedes and potatoes which are materials
poor in protein make the only exceptions. In most materials only 2—3 %of the total
ash is found in the water extract.

In the water extraction we have in some cases examined the effect of a longer
extraction time. It appeared that a 2 hours’ extraction was as efficient as a 24 hours’
treatment.

Excepting the potatoes, brewers’ grains, and wheat bran, the materials listed
in Tables I—41 —4 contain at most only traces of starch. The starch determination ac-
cording to the methods of Paloheimo involves the use of a standard solution prepared
of the product in question or, if wheat starch is used as standard, a knowledge of
the equivalency coefficient of the type of starch in question. Such coefficients are
given in the paper of Paloheimo.

The method described in this paper is not applicable to faeces analysis because
of the presence of bacteria and mucin material in the final residue. However, we
have developed a modification of the method which is valid for the investigation of
faeces. This method will be described in this journal in the near future.

Summary

The authors present a new method for the determination of the complex of
vegetable cell wall substances. The sample is extracted with boiling 80 % ethanol,
boiling absolute ethanol and cold water. The residue corrected for ash, protein, and,
if necessary, for starch, gives the amount of cell wall substances. Determinations
were made of the same samples of which Salo in this department, using quite a



311

different principle, has determined the cell wall complex. She determined separately
cellulose, neutral sugar hemicellulose, uronic acid hemicellulose, and lignin. Adding
up these items Salo obtained the total of the cell wall substances. The results obtained
with the new method are in most cases in agreement with the results of Salo (Table
1). The 80 % ethanol seems to be a very efficient solvent. In most cases more than
35 % of the dry matter of the sample was dissolved by it, while only about 0.3 %

was dissolved in the succeeding extraction with absolute ethanol (Table 2). I—l 21 —12 %

was dissolved by water.
The new method is compared also with the earlier method of Paloheimo in

which the sample is boiled in 0.05 N hydrochloric acid. It appeared that the results
obtained with the latter procedure are considerably lower than those obtained with
the new method. Evidently most plant materials contain cell wall substances which
are extractable with a very weak acid treatment.

REFERENCES

(1) Davy, Sir Humphry, 1814. Elemente der Agrikultur-Chemie. German translation by Fr. Wolff.
Berlin. 1 535.

(2) Honcamp, F. & Ries, F. 1914. Untersuchungen iiber verschiedenen Stroharten. Landw. Versuchs-
Stationen 84; 301.

(3) Paloheimo, Lauri. 1930. Zur Verwendbarkeit des jodkolorimetrischen Prinzips bei Stärkebestim-
mungen, Bioch. Z. 222: 150.

(4) 1948. Determination of starch according to the principle of iodine colorimetry. J. Sci.
Agric. Soc. Finland 20: 109.

(5) —& Antila, livo. 1931. fiber die Anwendung des Pulfrich-Photometers bei jodkolori-
metrischen Stärkebestimmungen. Bioch. Z. 238; 401.

(6) —& Paloheimo, Irja. 1931. Beiträge zur Jodkolorimetrie der Stärke nach der Methode
von Paloheimo, Bioch. Z.: 238: 391.

(7) —&Paloheimo, Irja. 1949. On the estimation of the total of vegetable membrane sub-
stances. J. Sci. Agric. Soc. Finland. 21: 1.

(8) Vainio, K. A., Kero, M.-L. & Herkola, Eine. 1961. Analyses of plant products in greater
detail. J. Sci. Agric. Soc. Finland 33; 51.

(9) Salo, Maija-Liisa. 1965. Determination of carbohydrate fractions in animal foods and faeces.
Acta agralia fennica 105: 1 —lO2.

(10) —1965. On the content of cell-wall constituents in various plant materials. J. Sci Agric.
Soc. Finland. 37; 127.

SELOSTUS:

SOLUSEINÄMÄAINEIDEN KOKONAISPALJOUDEN MÄÄRITTÄMINEN KASVITUOTTEISTA

L, Paloheimo ja K. A. Vainio

Yliopiston kotieläintieteen laitos, Helsinki

Kirjoittajat ovat kehittäneet uuden menetelmän soluseinämäkompleksin määrittämiseksi kasvi-
tuotteista. Yakuumissa 70°:eessa kuivattua ainesta keitetään 80-pros. etanolissa sekä sen jälkeen absol.
etanolissa. Kun saatu jäännös on vielä huiskutettu huonelämpöisessä vedessä, kuivataan ja punnitaan
se sekä määritetään hehkutuskevennys. Rinnakkaiskäsittelyllä saadusta jäännöksestä määritetään
G



312

raakaproteini. Jos näyte ei sisällä tärkkelystä, saadaan soluseinämäkolmpeksi vähentämällä hehkutus-
kevennyksestä raakaproteinin osuus. Jos taas näyte on tärkkelyspitoinen, on tärkkelys määritettävä
toisella rinnakkaiskäsittelyllä saadusta jäännöksestä ja hehkutuskevennyksestä vähennettävä myös
tärkkelyksen osuus. Analyysit suoritettiin samoista näytteistä, joista M.-L, Salo on aikaisemmin mää-
rittänyt soluseinämäkompleksin aivan toisenlaista periaatetta noudattaen. Useimmissa tapauksissa
uutta menetelmää käyttäen saadut tulokset ovat varsin yhtäpitäviä Salon esittämien tuloksien kanssa.
Sen sijaan uudella menetelmällä saadut tulokset ovat melkoisesti suurempia kuin Paloheimon aikaisem-
min esittämät 0.05-N-happomenetelmällä saadut. Ilmeisestikin kasvisolujen seinämissä, varsinkin jos
ne ovat hentoja, on runsaasti aineita, jotka liukenevat hyvin laimeaankinhappoliuokseen. Esimerkkeinä
uudella menetelmällä saaduista tuloksista mainittakoon eräiden kasvituotteiden soluseinämäaineiden
kokonaismäärät prosentteina kuiva-aineesta:

kukkivan puna-apilan varsien alaosa 66.1 koivun lehdet 38.2
sokeri]uurikasleike 57.9 timotei, lehtiaste 37.6
niittynurmikka, laidunaste 54.4 puna-apilan lehdet 29.1
kukkiva puna-apila 48.9 sokerijuurikkaan naatit 24.8
vehnän lese 43.3 kuoritut lantut 22.1