ABSTRACT

THE USE OF ULTRAVIOLET SPECTROMETRIC
PROCEDURES IN FORAGE EVALUATION AND
IN DETERMINATION OF DIFFERENCES
IN LIGNIN STRUCTURE

by James L. McCampbell

Twenty-one forage lignins (l3 grass and 8 legume) and
their respective fecal lignins were analyzed by ultraviolet
spectrometric procedures and quantitative color reactions to
determine if the lignin structure changed during the di—
gestion process. Twenty forage lignins (6 grass and lb
legumes) were analyzed by these same procedures to determine
if there were differences in lignin structures among the
forages. The values obtained from the ultraviolet spectro-
metric procedures and color reactions for the twenty forage
lignins were used to develop prediction equations estimating
digestibility and digestible dry matter intake/cwt. in sheep.

The lignin molecule did undergo a change during the di-
gestion process and the change was greater in grasses than
in legumes. This change was either: a reduction in the
amount of phenolic hydroxyl and methoxyl groups attached to
the benzene ring, but with a proportionally greater re-
duction in the phenolic hydroxyl content; or alternatively

the phenolic hydroxyl groups were replaced on the benzene

James L. McCampbell

ring by methoxyl groups. This is a minor change considering
all of the possible changes that could have taken place.

Lignins from legumes and lignins from grasses differed
in the chemical make up of the lignin molecules. The aromatic
compounds in the lignin from grass molecule had a greater
proportion of methoxyl groups and/or less phenolic hydroxyl
groups than did the aromatic compounds in the lignin from
legumes.

The methoxyl content increased with maturity of the
alfalfa hay, and lignin in forages with high in vitro dry
matter digestibility contained less methoxyl groups than did
forage lignin with low in vitro dry matter digestibility.

Alfalfa lignin had a greater phenolic hydroxyl content
than did birdsfoot trefoil lignin and siberian reed canary
grass contained more methoxyl groups than did the reed
canary grass lignin.

The prediction equations using data for combined grasses
and legumes were developed to estimate digestible dry matter
and digestible dry matter intake/cwt. These prediction
equations had low squared multiple correlation coefficients
(.52 and .80 respectively) and complicated mathematics which
would limit their usefulness. The four prediction equations
for grasses and legumes when taken separately have squared

multiple correlation coefficients (.93 and .96 for legumes and

.90 and .99 for grasses) sufficiently great to be of practical

James L. McCampbell

importance. The prediction equation for digestible dry matter
intake/cwt. in grasses was 3 = 3.6M — .7AX3 — 6.03X9. The
independent variable phenol (X3) was a quantitative color
reaction for phenol-like compounds. The independent variable

2u3rml (X9) was the absorbtivity at 2A3 mu on the difference

spectrum.

THE USE OF ULTRAVIOLET SPECTROMETRIC PROCEDURES
IN FORAGE EVAULATION AND IN DETERMINATION OF
DIFFERENCES IN LIGNIN STRUCTURE

By

James L. McCampbell

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Dairy

1969

ACKNOWLEDGEMENTS

Sincere appreciation is extended to Dr. J. w. Thomas for
his guidance during the course of this study and in the pre—
paration of this manuscript. Thanks are due to Dr. J. L. Gill,
Dr. H. M. Sell, Dr. M. B. Tesar, and Dr. D. Penner for their
many suggestions and for their critical reading of this
manuscript. Gratitude is extended to Dr. C. A. Lassiter,
Chairman of the Dairy Department, for providing financial
aid in the form of a research assistantship. The author is
especially indebted to his parents whose many sacrifices and
constant interest made graduate study possible. Finally,
words seem inadequate to convey the gratitude which is due
to my wife, Lexa, for her unyielding assistance in the
preparation of this thesis and for her patience, under-
standing, and moral support throughout the course of this

study.

ii

TABLE OF CONTENTS

Page
LIST OF TABLES ................................... v
LIST OF FIGURES ............................... ... vii
APPENDIX TABLE .................................. viii
INTRODUCTION ..................................... 1
REVIEW OF LITERATURE ............................. 3
Chemical Properties of Lignin
Definition ................................. 3
Lignin Complex ............................. A
Chemical Composition ....................... 5
Nitrogen Content of Lignin ................. 9
Molecular Weight ...... . .............. ..... lO
Methoxyl and Hydroxyl Groups in Lignin ..... ll
Physical Properties of Lignin
Color and Solubility ....................... 12
Color Reactions of Lignin .................. l2
Ultraviolet Characteristics ....... ......... l3
Infrared Characteristics ................... l8
Lignin Digestibility and Its Value as a
Predictor of Forage Quality ................ 19
MATERIALS AND METHODS ............................ 27
Chemical Analysis
Lignin Extraction .......................... 3O
Ultraviolet Spectrometry Analysis .......... 3A
Guaiacol Analysis .......................... 35
Phenol Analysis ............................ 36
Statistical Analysis ....................... 37
RESULTS AND DISCUSSION ........................... 38
Change in Lignin Structure During Digestion ... 38
Grasses .................................... 38
Legumes .................................... A5
Combined Legumes and Grasses ............... A6

iii

Page

Differences Between Grasses and Legumes in the

Lignin
Ultraviolet Characteristics ................ 48
Alfalfa Hay Early Cut vs. Late Cut ............ 57
Alfalfa Hay First Cutting vs. Second Cutting .. 62
Birdsfoot Trefoil Hay First Cutting vs. Second
Cutting .................................... 62
Alfalfa Hay vs. Birdsfoot Trefoil Hay ......... 63
Alfalfa Silage vs. Alfalfa Hay ................ 65
Alfalfa Silage vs. Alfalfa Silage Formic Acid
Treated ......................... ...... ..... 66
Alfalfa Variety Vernal vs. Variety DuPuits .... 66
Alfalfa Breeder Line 01 vs. Breeder Line 03 ... 66
Brome Hay First Cutting vs. Second Cutting .... 67
Reed Canary Grass Hay First Cutting vs. Second
Cutting .... ................................ 68
Siberian Reed Canary Grass Hay First Cutting vs
Second Cutting ............................. 68
Reed Canary Grass Hay vs. Siberian Reed Canary
Grass Hay .................................. 68
Bromegrass Hay vs. Reed Canary Grass Hays ..... 69
Lignin Determination .......................... 69
The Relationship of the Extracted Material to
the Lignin Content in the Forage ........... 72
Relationship Between the Lignin Characteristics,
Animal Data and Lignin Content ............. 72
Prediction Equations for Digestible Dry Matter
and Digestible Dry Matter Intake/cwt ....... 78
SUMMARY .......................................... 86
BIBLIOGRAPHY ............................... ...... 89
APPENDIX . ........................................ 95

iv

Table

10

ll

l2

13

LIST OF TABLES

Page
The Lignin Ultraviolet Characteristics and
Their Designations ............................ 29
Forage Descriptions and Their Code Numbers .... 31

The Effects of Digestion on Changes in Lignin
U.V. Characteristics of the Legumes and Grasses
Combined and Separate ......................... 39

Differences in Lignin Ultraviolet
Characteristics Between Legumes and Grasses ... A9

Differences in Lignin Ultraviolet
Characteristics for 20 Samples of Forages ..... 58

Differences in Lignin Ultraviolet
Characteristics Between Alfalfa and Birdsfoot
Trefoil ....................................... 6A

Differences in Lignin Ultraviolet
Characteristics Between Reed Canary Grass and
Siberian Reed Canary Grass .................... 7O

Differences in Lignin Ultraviolet
Characteristics Between Reed Canary Grasses
(Common and Siberian) and Bromegrass .......... 70

Comparison of Two Methods for the Determination
of Lignin Content in Forages .................. 71

Comparison of the Amount of Extracted Material
to the mg of Lignin in the Forage ............. 73

Correlations Between the Lignin Characteristics
of the Legumes and Grasses Combined, Animal
Data and % Lignin ............................. 75

Correlations Between the Lignin Characteristics
of Legumes, Animal Data and % Lignin .......... 76

Correlations Between the Lignin Characteristics
of Grasses, Animal Data and % Lignin .......... 77

Table
1A

15

l6

17

Prediction Equations on Combined Legume and
Grass Forages for Digestible Dry Matter (DDM)
and Digestible Dry Matter Intake per Hundred

Pounds of Live Body Weight (DDM/cwt) ..........

Prediction Equations for DDM and DDM/cwt on

Legume Forages ................................

Prediction Equations for DDM and DDM/cwt on

Legume Forages ..... . ............. .............

Prediction Equations for DDM and DDM/cwt on

Grass Forages .................................

vi

Page

Figure

LIST OF FIGURES

A Model Structure of Lignin Showing Three
Different Phenylpropane Units [(1) p-
hydroxycinnamic acid, (2) coniferyl alcohol,
and (3) sinapyl alcohol] and the Type of
Bonds that Can Be Formed Between these Units

Differences and Direct Absorption Spectra of
a Grass Lignin Preparation. These Spectra
Illustrate the Type of Data Obtained from

Forage Lignins ...............................

Schematic Diagram of the Lignin Extraction

Procedure ........................... . ........

The Difference Spectra of a Sample Forage
Lignin Preparation and a Sample Feces Lignin

Preparation ............................. .....

The Direct Spectra of a Sample Forage Lignin
Preparation and a Sample Feces Lignin

Preparation ....... ...... ......... ..... . ......

The Difference Spectra of Lignin from a Grass

and a Legume .................................

The Direct Spectra of Lignin from a Grass and

a Legume .....................................

vii

Page

16

32

Al

AA

52

55

APPENDIX TABLE

Table Page
1 Lignin Concentration, % Digestible Dry Matter

Maximum and Digestible Dry Matter Intake/cwt
for the Forages Studied ......... .. ...... ..... 95

viii

INTRODUCTION

Proper evaluation of forage is of academic importance and
there is still a practical need for quick and accurate
evaluation of forage samples. Attempts have been made to use
lignin content of the forage as an indication of forage quality
but the results have been variable and of limited usefulness
in estimating the nutritive value of a forage. Current
lignin analytical procedures do not take into account dif—
ferences in the structure of lignin molecules. Differences
in the basic units of the lignin molecules might influence
dry matter digestibility of the plant to a greater extent
than does the total amount of lignin.

If the basic units are as or more important than the
total lignin content, more knowledge should be obtained about
the structures of the lignin units, the type of compounds
present, the type of groups attached to these compounds, and
chemical linkages among the compounds. Ultraviolet spectro—
metry methods which are specific measures of certain lignin
characteristics were developed by wood chemists to provide
information about the lignin structure of wood. These lignin

ultraviolet characteristics were analyzed in this thesis on

forage and fecal lignin to determine if the lignin structure
changed during digestion of the forage. The same lignin ultra—
violet characteristics were used to determine if lignin from
grasses differed from that of legumes, if stage of maturity or
treatments had an effect on the lignin structure, if the
structure of lignin from first and second cutting hay dif-
fered, or if species of grasses or legumes differed in lignin
structure. Prediction equations were also developed using
these lignin ultraviolet characteristics to predict forage

digestibility and intake by sheep.

REVIEW OF LITERATURE

Extensive reviews on chemistry of wood lignin can be
found in the literature (Brauns, 1952), (Brauns and Brauns,
I960), (Hachihama and Jyodai, l9A6), (Harborne, 196A),
(Schubert, 1965), (Gould, 1966) and (Pearl, 1967). Since
the chemistry of wood lignin is adequately reviewed in the
above books and articles, the literature review in this thesis
will discuss lignin as it applies to forage plants, such as
legumes and grasses, and only refer to wood lignins where

pertinent.

Chemical Properties of Lignin

 

Definition. The term lignin has different meanings to

 

different people in various disciplines. The botanists,
agronomists, and animal nutritionists consider lignin as a
metabolite of the growing plant or a structural element con-
tributing strength to the mature plant and as an undigestible
feed ingredient. The microbiologists and soil chemists con—
sider lignin a residue of decay, while the enzymologists and
plant biochemists think of it as an end product of the

enzymatic dehydrogenation of specific phenylprOpane monomers.

Organic chemists consider lignin as a complex polymer that
challenges their fundamental interest in chemical structure.

Brauns (1952) defined lignin as that encrusting material
of the plant which: (1) is built up mainly, if not entirely,
of phenylprOpane building stones; (2) carries the major part
of the methoxyl content of the wood; (3) is unhydrolyzable
with acids, readily oxidized, and soluble in hot alkali and
bisulfate, and (A) readily condenses with phenols and thio
compounds. Brauns and Brauns (1960) updated this definition
to include the production of aromatic aldehydes upon the
oxidation of lignin with alkaline nitrobenzene and the
production of "Hibbert monomers" (vanillin, ethoxypropio-
vanillone, and vanilloyl methyl ketone) upon subjection to
ethanolysis.

The structure or structures of lignin in their native
state have not been determined to date principally because of
inability to isolate lignin in its native state unchanged
chemically or physically.

Lignin Complex. There is a bond between lignin and

 

cell wall carbohydrates but little is known about this
linkage. Siegel (1956) has shown that when cellulose, in
the form of filter paper, is added to a system of eugenol,
peroxidase, and hydrogen peroxide there is deposited on the
paper a lignin—like polymer similar to that observed in the
presence of tissue sections. The author suggests that ether

linkages may be essential for lignification. Bolker (1963)

used high—resolution differential infrared spectroscopy to
study the existence of a bond between lignin and cell wall
carbohydrates. He concluded that an acetal or hemiacetal
bond exists between the carbonyl groups of lignin and the
hydroxyl groups of some portion of the holocellulose and on
cleavage liberated a ketone group in the lignin. Freudenberg
proposed a model structure of lignin, as reported by Pearl
(1964), that contained a disaccharide attached to a carbonyl
group on one of the monomeric units.

Chemical Composition. The actual structure of lignin.

 

is not known, but research workers have determined either
theoretically or by actual experiments some important aspects
of the lignin structure. Pearl (1967) hypothesized that
coniferous lignin was a polymerized product derived from a
building unit R-C-C-C, in which R is the A-hydroxy—3-
methoxyphenyl group. Most recently lignin chemists have
referred to the A—hydroxy—3—methoxyphenyl groups as the
guaiacyl group and the phenylpropane structure as a C6—C3
unit. Alkaline oxidation with nitrobenzene (Creighton 33

al., 19A1) and with cupric oxide (Pearl, 19A2), giving

vanillin in yields above 25%, has been accepted as proof of

the guaiacyl moiety existing in lignin. Harris 33 a1. (1938)
subjected isolated lignin to high pressure hydrogenation and
obtained a combined yield of over 50% of A-n-propylcyclohexanol,
A-n-propylcyclohexane-l,2-diol and 3-(A-hydroxycycloheXyl)-

l—propanol. These high yields of hydrogenation products with

cyclohexylpropane structures proved that lignin was essentially
aromatic and demonstrated that for the most part it was a
condensation polymer of guaiacylpropane monomers.

Kefford (1958) concluded that linkages between the
building units involved the phenolic hydroxyl group and the
three carbon side chain but there is variation in the actual
linkages formed. A model of a lignin molecule is shown in
Figure 1, this model demonstrates the type of compounds that
can be present in the lignin molecule and the type of bonds
that can be formed between the basic units. Three aromatic
compounds are represented in the model: compound 1 is p-
hydroxycinnamic acid; compound 2 is coniferyl alcohol; and
compound 3 is sinapyl alcohol. Other compounds may be present
in lignin. Two types of bonds can be formed according to
this model and these bonds are ether and carbon-carbon bonds.
The ether bonds can be formed between side chains or between
a phenolic hydroxyl group and a side chain. A carbon-carbon
bond can be formed between any two carbon atoms.

Lignin has been isolated by many different procedures and
the isolated lignin takes its name from the isolation procedure.
Examples of this naming procedure are: ethanol lignin ex—
tracted by means of ethyl alcohol in the presence of small
amounts of mineral acid; and sulfuric acid lignin extracted
by concentrated sulfuric acid from the plant material.

In studies of lignin isolated from the stalk and leaves

of normal maize and brown—midrib mutants Gee gt a1. (1968)

Figure l.

 

HOC=O
CH
H
CH
:/\\
H .} (l)
HCOH ‘\.’/
HC~e o
I
noon
/4"\\
H 8 R
’l
H$OH '? (2)
HC <3 ,/ OCHB
I a"
HG~———__O
i
. L
-\ ‘ -) OCH3
6____

(3)

A model structure of lignin showing three dif-

ferent phenylpropane units [(1) p-hydroxy—

cinnamic acid,

(2) coniferyl alcohol, and

(3) sinapyl alcohol] and the type of bonds
that can be formed between these units.

found that the elemental analysis of normal dimethylformamide
(DMF) lignin was C, 62.23%; H, 5.51%; N, 0.77%; O, 29.61%;
and DMF lignin from brown—midrib mutants contained C, 60.72%;
H, 5.A3%; N, 1.2A%; 0, 30.10%. The phenolic acids in DMF
lignins were isolated by thin layer chromatography after
liberation by alkaline hydrolysis and extraction from lignin
with anhydrous ether. Ferulic and p—coumaric acids were the
predominant products liberated. There was no difference in
the amount of erulic acid liberated from normal or mutant
lignin, but the amount of p—coumaric acid liberated from the
mutant lignin was approximately 50% less than that from
normal lignin. Alkaline nitrobenzene oxidation of the lignins
resulted in lower yields of syringaldehyde, vanillin, and p-
hydroxybenzaldehyde in the brown-midrib mutant lignin when
compared to normal maize lignin. The authors have suggested
that when the mutant gene is present the lignin core has
fewer sites at which p—hydroxycinnamic acid can be esterified
thus changing the lignin formed by the mutant plants. This
research indicates that mutant plant might be found that have
higher digestibility due to a change in the lignin structure.
Kato gt a1. (1967) found that the elemental analysis of
tobacco milled wood lignin from the stalk was C, 59.28%; H,
6.16%; O, 3A.56% and from the midrib C, 62.01%; H, 5.86%;

O, 32.13%. The yields of p—hydroxybenzaldehyde, vanillin,
and syringaldehyde from alkaline nitrobenzene oxidation of the

stalk was 2.19%, 15.7%, and 5.88% respectively, and from the

midrib 3.7A%, 10.63% and 0.0% respectively. The change in
the alkaline nitrobenzene oxidation products as well as the
elemental analysis indicates that the lignin composition is
different in different locations of the plant.

Nitrogen Content of Lignin. Wood lignin is practically

 

free from nitrogen. The very small quantities of nitrogen
(0.2-0.3%) found in wood lignin are considered by Brauns
(1952) to be accidntal impurities caused by protein materials
hydrolyzable only with difficulty. Bondi and Meyer (1948)
found that the nitrogen content of lignin prepared from green
plants is due neither to accidental contamination with protein
nor to the presence of lignin—protein complexes. They found
that grass lignin contains about 1.5% nitrogen and legume
lignin contains about 3.0% nitrogen.

In studies comparing lignin from the stalk and leaves of
normal maize with that of brown—midrib mutants of maize Gee
gt gt. (1968) found that the brown-midrib mutants contained
almost twice as much nitrogen as did normal maize, 3.22 and
1.70, respectively. Ely gt gt. (1953) found that the nitrogen
content of lignin isolated from orchardgrass hay decreased
with maturity of the plant. The immature plant lignin con-
tained A.25% nitrogen and the most mature plant lignin con-
tained 2.41%. Lignin isolated from corresponding feces had
an average nitrogen content of 5.19 and 2.19 percent,
respectively.

Czerkawski (1967) characterized the nitrogenous impurities

in the purified grass lignin. He hydrolyzed the lignin three

10

times with 6 N HCl for 16 hours of 1050 and the free a-

amino groups were estimated by the ninhydrin method with
alanine as a standard. The nitrogen content left in the
residue was determined by the Kjeldahl method. More than
half of the resistant nitrogen could be removed as a-

amino nitrogen presumably in the form of amino acids and the
rest of the nitrogen remained in the residue. This leads one
to conclude that part of the nitrogen in lignin is protein and
part is nitrogen attached in some way to the lignin molecule.
This nitrogen could possibly be amine groups that were not
cleaved from the amino acids L-tyrosine or L-phenylalanine
when these monomers were used in the formation of lignin.
Brown (1961) proposed a lignification pathway that included
the synthesis of these two amino acids and then the formation
of the monomeric units of lignin from them. Since the
structure of these amino acids have the basic C6—C3 unit it
is possible that the nitrogen in grasses and legumes is

part of the structure of lignin.

Molecular Weight. Some of the physical methods that

 

have been used for the determination of the molecular weight
of lignin are diffusion, depression of freezing point by
B-naphthol, osmotic pressure methods and the ultracentrifuge.
'Ihe values vary between 250 and 11,000 (Brauns, 1949). The
xnolecular weight can be determined by the introduction of
cleavage of known groups, such as methoxyl, methanol, or

fornmldehyde by quantitative reactions that measure the

ll

amount of groups cleaved or introduced. The molecular weight
when determined this way is approximately 800.

Brown gt gt. (1967) used selective decay of wood by fungi
to determine the molecular size of lignin in different parts
of the plant. They concluded that lignin was a polymer of
finite rather than infinite size and that it differed at least
in molecular size according to its location within wood cell
walls. The larger molecular weight lignins were located close
to the center of the cell.

Methoxyl and Hydroxyl Groups in Lignin. The methoxyl
content of different wood lignins vary from 14 to 20% and is
higher in hard wood lignins than in soft wood lignins.
According to Phillips gt gt. (1939) and Bondi and Meyer (1943)
the methoxyl content of lignin isolated from green plants is
lower than that of spruce wood lignin, and increases with the
age of the plant. Bondi and Meyer (1948) found that lignin
from grass contains 9-10% methoxyl groups while lignin from
legumes contains approximately one half of this amount. They
also found that the methoxyl content of the lignin isolated
from feces differs very little from that of the corresponding
plant lignin therefore there is no noticeable rupture of ether
linkages during digestion in the animal. Ely gt gt. (1953)
found similar results when working with orchardgrass lignin.

Data concerning the phenolic-hydroxyl content of lignins
is small in volume. Gee gt gt. (1968) found that the phenolic—

hydroxyl content of dimethylformamide lignin was 10.41 x 10.}-1

12

moles per gram in normal maize and 6.23 x 10’“ moles per gram
in the mutant. Brauns and Bruans (1960) reported that the
total hydroxyl content of milled wood lignin prepared in air
was 10-11% and of native spruce lignin was 10.2%. The
phenolic hydroxyl content was 4.6% for milled wood lignin

and 3.4% for native spruce lignin.

Physical Properties of Lignin

 

Color and Solubility. The color of various lignins

 

extracted by 0.5 N NaOH is light to dark brown. The lignins
are readily soluble in dilute NaOH, ethanol, and in acetone.

Up to 70% of the lignin is soluble in 90% acetic acid. Phenol
and ELnaphtol dissolve up to 15% of their weight of lignin when
heated about 20°C above their melting points. Lignins are not
soluble in other common solvents. Brauns (1949) concluded

that the color of lignin and lignin derivatives depend upon

the mode of preparation. He also stated that there was evidence
that protolignin was almost as white as cellulose and the color
of wood and other lignified materials were caused by extraneous
coloring matter rather than by lignin.

Color Reactions of Lignin. Certain color reactions become

 

important in the study of lignin because they are specific for
definite chemical structures or groups in the lignin molecule
Gierer (1954) found that quinonemonochloroimide reacts with

lignin in a weak alkaline solution to yield a blue "dyestuff".

He investigated this reaction using 80 model compounds

l3

chemically related to lignins and found that the reaction

was specific for the free p-hydroxybenzylalcohol group in the
lignin preparation. A positive Wiesner color reaction (Adler
gt gt., 1948) indicates the presence of a coniferylaldehyde
group in the lignin, possibly as an end group, by the pro-
duction of a red purple color. A positive Cross—Bevan color
reaction (Migita gt gt., 1955) indicates the presence of a
syringyl group at the terminal end of the lignin molecule by
the production of a red—violet color. The Maule color re-
action (Spearin and Dresler, 1954) was developed to dif-
ferentiate between hardwood and softwood lignins. A dark-
purple color indicates the presence of hardwood-like lignin
and a brown color indicates the presence of softwood—like
lignin.

Ultraviolet Characteristics. The use of ultraviolet

 

spectra for characterization of lignin preparations was

develOped in the past decade and it has been an invaluable
aid to the lignin chemist. Patterson and Hibbert (1943) found
that the ultraviolet absorption spectra of ethanol lignins in—
dicate that lignin is aromatic in nature and that a carbonyl
group or an ethylenic double bond is present in conjugation
with the aromatic nucleus. Aulin-Erdtman (1949) reported that
the pH value of the solvent influenced the ultraviolet absorp-
tion spectrum of the lignin preparations. The minimum shifted
toward higher absorbances and longer wavelengths in alkaline

solutions than in neutral solutions. Aulin—Erdtman (1949)

14

also reported that the shift in wavelengths from the peak on
the neutral spectrum to the peak on the alkaline spectrum was
roughly prOportional to the content of phenolic hydroxyls pre—
sent. In the case of lignin preparations studied by Aulin—
Erdtman (1949) the position of the maximum was practically
unchanged in all cases; therefore, they probably contained
very little phenolic hydroxyls. Figure 2 shows an example of
a difference and direct spectra. Wexler (1964) employed
ultraviolet spectrometry for interpreting the characterization
of lignosulfonic acids. He found the direct spectrograms use-
ful in establishing the gross features of the material,
whereas the difference spectrogram was of value in determining
the aromatic hydroxyl content besides being a characteristic
physico-chemical property of the material. The difference
spectrogram (spectrum) is the difference in the absorbance
between the neutral and alkaline spectrograms. Goldschmid
(1954) found two maximums in the difference spectra; one at
250 mu and the other at 300 mp. These maximums are charac-
teristic for the absorption of the phenolate ion of simple
substituted aromatic hydroxyl compounds. The 250 mu maximum
is common to carbonyl and phenolic hydroxyl groups while the
300 mu maximum of the difference curve is characteristic for
phenolic hydroxyl groups only. He also found that the precise
structure of the side chain had very small effects on the

ultraviolet absorption characteristics of model compounds.

15

Figure 2. Difference and direct absorption spectra of a
grass lignin preparation. These spectra
illustrate the type of data obtained from
forage lignins.

Difference spectrum ’—_“fi*‘“"*

C)
)

Sodium hydroxide direct spectrum "

Phosphate buffer direct spectrum *———**———*

16

Figure 2

Agsorbance

mo.

OH.

O
(\l

A

 

l
.L- 1

ozm omm

AJEV mecmam>w3

owm owm

1J5

/

/.

02m 0mm

17

To date most of the ultraviolet spectrometry work has
been done with wood lignin, but Stafford (1960) used the
ultraviolet absorption spectra to see if there was a dif-_
ference between young green timothy shoots and mature
timothy hay. She concluded that mature hay had greater
absorbances in the difference curve than did young green
shoots at the same wavelength. The peaks of the difference
curve were at 250, 300, and 350 mu with an area in the
region of 270 to 280 mu where the optical density of the
alkaline spectrum which makes the difference spectrum go
below zero absorbance. In further research Stafford (1962)
found that lignin isolated from the sclerenchyma cell of
the leaf blade differed from the lignin of the rest of the
plant. The lignin from the other parts of the plant had a
greater absorbance in the difference spectrum at 350 mu than
at 250 or 300 mu but the lignin from the blade had a greater
absorbance at 250rmi than at 350 mu. Therefore she concluded
that lignin isolated from different parts of the plant dif-
fered in structure. Gee gt gt. (1968) used ultraviolet
spectroscopy to determine the amount of phenolic hydroxyl
present in lignin isolated from maize. These authors
calculated the percent phenolic hydroxyl in maize lignin by
taking the absorptivity at 300 mu on the difference spectra
times 0.414.

Allinson (1966) studied lignin extracted from three

varieties of alfalfa. Each variety had two sample clones,

18

one with high in vitro digestibility and the other with low
in vitro digestibility. Optical density peaks occurred on
the difference spectrum at 250, 300, and 350 mp and with an
area in the region of 270 to 285 mu where the Optical density
of the neutral spectrum was greater than the optical density
of the alkaline spectrum which makes the difference spectrum
go below zero. The observed spectra were different not only
in Spectral conformation but also quantitatively. These
differences were present especially between clones grouped
on the basis of nutritive value. This work suggests that
ultraviolet spectrometry values could be used as a method of
estimating the relative nutritive value of forages or as one
or several parameters in a regression equation related to
nutritive value.

Infrared Characteristics. During the last twenty years

 

infrared absorption spectra have been developed to assist in
characterization of lignin. Data on infrared absorption
curves can be used to determine whether lignins from dif-
ferent species or isolation procedures differ in gross
characteristics or structure, such as fewer hydroxyl or
carboxyl groups. Sell gt gt. (1961) studied the infrared
spectra of lignins isolated from alfalfa stems, timothy
stems, and aspen wood by the potassium hydroxide pellet
method. They concluded that the alfalfa stem lignin and

timothy stem lignin showed some differences, but were more

alike than when compared to aspen wood lignin. Infrared

l9

characteristics were not investigated in this thesis project.

Lignin Digestibility and Its Value as a Predictor of

 

Forage Quality. As a forage plant matures the composition

 

changes. Protein and the highly digestible carbohydrate
fractions decrease in amount while the lignin and cellulose
fractions increase. Phillips gt gt. (1939) found this to be
the case in the oat plant. Miller gt gt. (1967) analyzed
baled hay that was allowed to dry to four different moisture
contents ranging from 26 to 58%. There was a greater percent
ash, cell wall constituents, cellulose, acid detergent fiber,
and lignin in the hays baled at the higher moisture content
than those baled at the lower moisture content. The in-
crease in cell wall constituents and fiber which decreased
the nutritive value was probably due to the extent of the
temperature rise which was caused by the amount of moisture
in the hay at the time of baling. The higher the moisture
content the higher the temperature rose resulting in greater
oxidation of nutrients which lowered the nutritive value in
the final product. Bechtel gt gt. (1945) observed an in—
creased percentage of lignin, ash, and crude fiber in brown
and black hays because other organic constituents had been
lost in the process leaving indestructible lignin as a
greater percent of the material remaining. Pigden and
Heinrich (1957) studied the lignin content of the leaves

and stems of six clonal lines of wheatgrass and notes that

there were significant differences in lignin content between

20

clones for both leaf and stem. They also found a highly
significant difference in lignin content between years and
in percent leaf between lines. This suggests that the
amount of lignin formed in the plant is not constant from
year to year and may be influenced by the environment.
Besides variation due to the amount of lignin formed by
the plant, there is a possibility of forming artifact lignin
during the laboratory analysis procedure of the sample. Van
Soest (1964) found that when wet forage samples were heated
above 50°C during drying the apparent lignin content in—
creased by as much as 300%. He also reported that negative
estimates of digestibilities can occur due to the formation
of artifact lignin in processing the feed and feces. This
occurs when more artifact lignin is formed in heating feces
than feed. In the same study, Van Soest found that the
lignin content of forages increases as the digestibility
of the dry matter decreased. The correlation coefficient
between lignin and digestibility was -.82 for grasses,
-.74 for alfalfa, and -.40 when all were combined. In
later research Van Soest (1965) found that heat-drying of
forages at temperatures above 50°C showed analytically
significant increases in yields of lignin and fiber. The
increase in acid-detergent fiber was largely accounted for
by the production of artifact lignin via the non—enzymatic
browning reaction.

Stallcup gt gt. (1955) found that the lignin content

21

of hays influenced the digestibility of forages and de-
creased the passage of nutrients through the rumen. Digest-
ibilities of hays were negatively related to their lignin
content and there were more ingesta remaining in the rumen
12 hours after feeding when sheep were fed hays high in
lignin as compared to hays low in lignin.

The digestibility of cell-wall carbohydrates is affected
by the extent of lignification and this mode of action is
by physical barrier or by some other means. Dehority gt gt.
(1962) used four stages of maturity of timothy, three of
alfalfa, and two of orchardgrass as substrates to study the
rate and extent of hemicellulose fermentation decreased as
the plant matured but when forage particle size was reduced
by ball-milling, to remove any physical barrier by lignin,
the extent of hemicellulose fermentation was increased.
Additionally, the rate and extent of pectin digestion was
estimated with the three stages of maturity of alfalfa. Both
the rate and extent of pectin digestion decreased as the
alfalfa matured and the effect of maturity was lessened con-
siderably after reduction of particle size by ball-milling.
These data indicate that both hemicellulose and pectin diges-
tion are influenced by maturity of the forage and suggest
that this is the result of lignin forming a physical barrier
between the plant hemicellulose or pectin. This then results
in a decrease in the ability of the forage to be degraded

by rumen bacteria. Goering and Van Soest (1968) reported that

22

when sodium chlorite was ensiled with forages the estimated
true digestibility increases as the level of sodium chlorite
increased up to 3-5%. They found that true digestibility of
orchardgrass increased from 86% with 0% NaClO2 and that the
lignin content decreased from 5.9% to 2.6%, respectively,
after being ensiled for 15 days. This shows that the sodium
chlorite is chemically removing the physical or chemical
barrier that protects the cell-wall constituents from attack.

Results concerning the digestibility of lignin are
varied but most research workers agree that lignin from
grasses and legumes is undigestible. Bondi and Meyer (1948)
found that the solubility, the apparent molecular weight,
the methoxyl content, and the nitrogen content of the lignin
remained unchanged by digestion. On the other hand the
disappearance of the aldehyde fraction from the degradation
products of fecal lignin showed that a change in side chains
had occurred and that the nonphenolic hydroxide groups had
disappeared. Crampton and Maynard (1938) found that 97.8%
and 99.3% of the dietary lignin was recovered in the feces of
rabbits and steers, respectively. Ellis gt gt. (1946) con-
cluded that lignin from sudangrass hay was not digested by
the cow, sheep, or rabbit. The percent recovery of lignin
in the feces was 102%, 100%, and 97% for the cows, sheep, and
rabbits, respectively.

Forbes and Garrigus (1948) found that the average re-

covery of lignin was 102:7 percent on bluegrass pasture that

23

was cut each day in a series of digestion trials with steers.
With sheep, the average recovery of lignin was 105.4 percent
on orchardgrass and ladino clover. From this study dry matter
digestibility and total digestible nutrient content of the
various forages were found to vary inversely with the lignin
content of the forage. Kane gt gt. (1950) concluded that
there was no loss of lignin during the process of digestion
when they were able to recover 98.8% of the lignin in

alfalfa hay fed to three cows.

Ely gt gt. (1953) found that the apparent digestion
coefficients of lignin in dairy cow rations containing
orchardgrass hay cut at different stages of maturity ranged
from 3.6 to 16.0 percent. Davis gt gt. (1947) found that
when four yearling and two year-old Hampshire ewes were fed
dehydrated pea and lima bean vines, the digestion coefficients
for lignin were 16.2% and 10.6%, respectively. Hale gt gt.
(1947) reported that the digestion of lignin in the rumen
was about 3.1% but the fecal digestion of lignin in the hay
was about 30%, indicating that most of the digestion of
lignin occurred after the forage left the rumen.

The wide variation in the recorded estimates of lignin
digestibility is not surprising especially since a variety
of determination or isolation procedures and techniques were
used to obtain these estimates.

Tomlin gt gt. (1965) used orchardgrass, bromegrass,

timothy, reed canary grass, alfalfa, red clover, and

24

birdsfoot trefoil at three stages of maturity to study the
relationship of lignification to in vitro cellulose digest-
ibility in a 12 hour incubation system. Lignin content was
found to be negatively correlated with in vitro cellulose
digestibility for grasses and legumes, and the regression
equations for the two groups were significantly different.
Lignification was linearly and negatively related to cellulose
digestibility as the grasses matured, but this relationship
did not exist for alfalfa. Patton (1943) studied the lignin
and cellulose content of nine Montana grasses. In general,
the increase in lignin and cellulose content during growth
appeared to have been similar in the various species with
a coefficient of correlation between lignin and cellulose
greater than +0.9 in the 123 samples analyzed. Patton and
Gieseker (1942) determined lignin and cellulose in five
species of grasses, at five different stages of maturity,
and in two localities in Montana. Lignin increased with
advancing season, from about 5% in May to 18% in September,
with considerable species differences. They found that the
lignin content closely paralleled the cellulose content, both
increasing to maturity. They believed lignin to be of definite
value in predicting the feeding value of forage plants.

Van Soest (1967) concluded from his research that the
amount of lignification did not affect the digestibility of
the cellular contents. The digestibility of the structural

carbohydrates of the cell wall declined with the maturity of

25

the forage and these components tended to form an increasing
proportion of the dry matter of the plant with age. Digest-
ibility of the cell wall carbohydrates were highly negatively
correlated with lignification. This was also reported by
Van Soest and Marcus (1964). Van Soest (1965) found that the
lignin was highly correlated with cell wall constituents,
+0.73; acid-detergent fiber, +0.83; and cellulose, +0.71.
These correlations held true within species but not between
species. Colburn gt gt. (1968) carried out digestion trials
on 17 forages (l6 orchardgrass, l alfalfa) with growing
Jersey steers. The digestible dry matter of first cutting
orchardgrass hay declined at the rate of 0.4 percent for

each day's delay in harvest from May 11 to June 3. The pre-
diction of percent digestible dry matter by chemical
composition was eXpressed by the equation y = 59.2 - 2.75Xl

— 0.24X2 + 0.66X3 + 0.50Xu, where X1, X2, X3, X)4 represent
percent lignin, hemicellulose, crude protein, and cellulose
respectively. The coefficient of variation was 3% and the
coefficient of correlation was 0.80.

Lignin has been used as an indigestible marker to
determine the coefficient of digestibility of other feed
constituents. Ellis gt gt. (1946) developed the lignin ratio
method for determing the digestibility of other feed

constituents. The formula is shown below.

26

loo—loo E .n in feces
Z n in feed

y = percent digestibility of a specific nutrient, n

 

n = percent of a specific nutrient in either feed or
in feces

x = percent lignin in feed

2 percent lignin in feces

Advantages of this method are that the digestibility of feeds
may be determined without measuring total feed intake and
total fecal output. This technique also permits studies of
the digestibility of pasture forages under grazing conditions.
The differences between the digestion coefficients when
measured by this method and by the conventional method were
not significantly different.

One disadvantage of the lignin ratio method is the
difficulty of obtaining a representative sample of pasture
forage. A second disadvantage is the daily variation in
lignin content of the feces. Because of the two reasons

stated above it may be hard to get repeatability using the

lignin ratio method for pasture studies.

MATERIALS AND METHODS

The experiment was designed to answer three major
questions. 1) Has the ultraviolet characteristics of lignin
changed during digestion? 2) Are the ultraviolet charac-
teristics and specific color reactions of lignin different
in legumes and grasses, between species of grasses or species
of legumes, or between different stages of maturity? 3) Are
these lignin characteristics related to sheep intake or
digestibility and can prediction equations be developed to
predict the digestibility and the digestible dry matter
intake of a pOpulation of forages?

Several sheep digestion trials have been run on forages
at Michigan State University over the last six years.
Representative samples of the dried forages used in these
trials and their feces were collected and stored in a
refrigerated room for future use. There were different
species of grasses and different species of legumes in these
samples which provided a good sample of the population.

Among these forages were samples of the same forage but with
different harvesting treatments or additives and other of the

same forage but harvested at different stages of maturity.

27

28

A representative sample of a difference spectrum and
direct spectra is shown in Figure 2 and will help to explain
how the lignin ultraviolet characteristics were obtained. Any
of the numerous values from the spectra could have been used
in this study. Those selected are defined and described
in the tOp twelve lines of Table 1. Most of the selected
observations have been related to the content of specific
functional groups of wood lignin preparations. Their sig—
nificance will become clearer when discussed in the results
and discussion section.

To provide information concerning the first question
the lignin from 21 forage (l3 grass and 8 legume) samples
and their respective feces samples (combined from 3 or 4
sheep) were analyzed by ultraviolet spectrometry. The dif-
ference spectrum and direct spectra were obtained for each
forage and fecal sample. Data on nine of the fourteen lignin
ultraviolet characteristics listed on Table l were used in
the analysis of the forage and fecal samples. The nine
values or characteristics used were 243 mp, 265 mp, 300 mu,
and 337 mu, % POH, Ratio O.D., A mu, and mu shift.

To provide information concerning the second and third
questions twenty forage (6 grass and 14 legume) samples were
analyzed by ultraviolet spectrometry and values for the
fourteen lignin ultraviolet characteristics listed in Table l
were used. These fourteen characteristics were analyzed to

determine if differences in lignin structures occur between

29

Table l. The lignin ultraviolet characteristics and their

 

 

designations.

Designation Definition of Characteristic

243 mu absorbtivity<a> at 243 mu on the difference
spectrum b

265 mu absorbtivity at 265 mu on the difference
spectrum

270 mu absorbtivity at 270 mu on the difference
spectrum

300 mu absorbtivity at 300 mu on the difference
spectrum

337 mu absorbtivity at 337 mu on the difference
spectrum

% POH the % phenolic hydroxyl groups in lignin based
on the percent increase in absorbtivity of the
neutral peak to the alkaline peak on the direct
spectra< ) X 0.21

Ratio O.D. the ratio of Optical density of the minor peak
to the maximum peak on the difference spectrum

A mu the distance in mu between the minor peak and
the maximum peak on the difference spectrum

mu shift the distance in mu between the neutral peak
and the alkaline peak of the direct spectra

300 POH the % phenolic hydroxyl groups in lignin, based
on the absorbtivity at 300 mu on the difference
spectrum X 0.414

250 POH the % phenolic hydroxyl groups in lignin based
on the absorbtivity at 250 mu on the difference
spectrum X 0.192

250 g POH the % phenolic hydroxyl groups in lignin based
on the absorbtivity at 250 mu corrected for
base line(3 x 0.192

Guaiacol mg of guaiacol like compounds/gm lignin

Phenol mg of phenolic acids/gm lignin

(a)absorbtivity = absorbance

 

(concentration) (cell length)

(b) (C) (d) See Figure 2.

30

grasses and legumes, between treatments of the same forage,
between stages of maturity of the same forage, or between
species of grasses or legumes. Values for these same lignin
characteristics were also related to animal intake and
digestibility as single predictors or in multiple prediction
equations to determine their value in estimating the nutritive
value of forages. A list of twenty forages and their descrip-

tions is presented in Table 2.

Chemical Analysis

 

Lignin Extraction

 

The lignin was extracted from the forage in two steps
as shown by Figure 3. The forage was first extracted with
a detergent-acid mixture to obtain a lignino-cellulose (acid
detergent fiber) preparation free from proteins, fats,
pigments, pectins, sugars, starch and hemicellulose. The
acid detergent fiber procedure (Van Soest, 1963) dissolves
all plant components except for lignin and cellulose
thereby leaving a lignino—cellulose (acid detergent fiber)
residue. The lignin was extracted from the acid detergent
fiber by dilute sodium hydroxide when the mixture was heated.
Lignin is soluble in hot alkaline solutions but cellulose is
not. The lignin extraction procedure used was a modification
of the procedure described by Stafford (1960). A slightly
different extraction procedure was used for the first part of

the study than used for the second part. Details for both

31

Table 2. Forage descriptions and their code number.

 

Sample No.(°) Description
6401-01 Alfalfa hay first cutting
6401-02 Alfalfa hay second cutting
6502-01 Alfalfa hay early cut
6502-02 Alfalfa hay late cut
6701-21 Alfalfa silage control
6701-22 Alfalfa silage formic acid treated
6701—01 Alfalfa hay (control to 6701 silage study)
6201—11 Birdsfoot trefoil first cutting
6201-12 Birdsfoot trefoil second cutting
6702-OV Alfalfa hay var. Vernal
6702—OD Alfalfa hay var. DuPuits
6703-02 Alfalfa hay breeder line
6703-03 Alfalfa hay breeder line
6704-01 Alfalfa hay breeder line
6401-51 Bromegrass first cutting
6401-52 Bromegrass second cutting
6401-41 Reed canary grass first cutting
6401-42 Reed canary grass second cutting
6401—31 Siberian reed canary grass first cutting
6401—32 Siberian reed canary grass second cutting

 

(a)

the number,

Year harvested is indicated by the first two digits of
sequential sheep trial by fourth digit,

forage species by fifth digit, and cutting number or
other information by last digit.

32

 

 

 

 

FORAGE
I
' i
I
l !
l 1
Acid detergent All other
fiber plant
material
Residue Lignin

Figure 3. Schematic diagram of the lignin extraction
procedure.

33

these procedures are given.
1. Forage vs Feces

A 0.10 gm sample of the acid detergent fiber was added
to 10 ml of 0.5 N NaOH in a stoppered test tube and the mixture
set in an oven for 16 hours at 70°C. At the end of this
extraction time the test tube was centrifuged at 2000 rpm for
10 minutes and the supernatant was then transferred to a
beaker or similar type vessel. The residue was washed twice
using 5 ml of distilled water each time, centrifuged at 2000
rpm and the supernatants transferred to the same beaker. The
supernatant and washes were combined for analysis.

The residue was re-extracted using 10 ml of 0.5 N NaOH
for 24 hours at 70°C. At the end of this period the ex-
traction mixture was treated as before. The first and second
extracts were analyzed separately in the laboratory then the
values combined in all analysis of results.

2. Forage Analysis

The lignin extraction was the same as described above
except that the initial extraction time was 48 hours instead
of 16. It was thought that with the extra length of the
initial extraction there would be no need to run a second
extraction. The second extract did contain lignin but in

very small quantities.

34

Ultraviolet Spectrometry Analysis

 

The procedure used was a modification of the procedure
described by Stafford (1960). Each sample of extracted lignin
was neutralized to about pH 8.5 to 9.0 then the total volume
was brought up to 50 ml with distilled water. The ultraviolet
spectrometry analysis was made on aliquots of the sample.

One 1 ml aliquot was added to a test tube containing 10 m1 of
0.05 M phosphate buffer at pH 7.0. The other 1 ml aliquot
was added to a test tube containing 10 ml of 0.05 N NaOH at a
pH of about 12.0.

The Beckman model DK-2A ratio recording spectrophotometer
was used to determine the ultraviolet absorption spectra on
both the neutral and basic solutions. The range of wavelengths
scanned was 220 to 340 mu. The neutral solution was placed
in the reference cell and the basic solution was placed in the
sample cell to determine the difference spectrum. Distilled
water was placed in the reference cell and either the neutral
or basic sample was placed in the sample cell to obtain the
direct spectra of the forage. The difference spectrum measures
the difference between the neutral and basic direct spectra
and the direct spectra measure the absorbance of the neutral
or basic solution over the wavelengths scanned. Twelve of
the fourteen lignin ultraviolet characteristics come from the
difference and direct spectra (10 from the difference spectrum
and 2 from the direct Spectra). The lignin preparations had

to be analyzed within about 5 hours after the pH was

35

standardized to 8.5-9.0 because the added acid tended to
change the lignin ultraviolet properties of the lignin
preparations. This time factor limits the number of samples

that could be analyzed in one day.

Guaiacol Analysis

 

The lignin preparations were analyzed to determine the
amount of free guaiacol like compounds present in the pre-
parations. The procedure used was similar to the one described
by Stafford (1960). A 4 m1 aliquot of the extracted lignin
solution was added to a test tube that contained 3 ml of 0.5
M tris (hydroxymethyl) amino—methane buffer at pH 9.0 and 0.5
ml of a freshly prepared alcoholic solution containing 25 Ag
of 2,6-dichloroquinonechlorimide. The compounds were mixed
and allowed to stand at room temperature for one hour.
Absorbance readings were made at the end of the hour at 610 mu
with a Beckman DU spectrOphotometer, using guaiacol as a
standard. The guaiacol was purchased from Matheson, Coleman,
and Bell. The absorbance readings were converted to mg of
guaiacol-like compounds per gm of lignin in the original
forage.

Stafford (1960) developed a formula for the conversion
of mg of guaiacol to mg of lignin (mg guaiacol x 32 = mg lignin)
in the forage. This formula was used to estimate the amount of
lignin in the forage and these results were compared to total

lignin found by the method of Van Soest (1963).

36

thenol Anatysis

 

The extracted lignin solutions were analyzed to determine
the amount of free phenol—like compounds present in the pre-
parations. The procedure used was similar to the one described
by Swaim and Hillis (1959). A 0.5 ml aliquot of the lignin
preparation was added to a test tube containing 10.5 ml
distilled water. Next 1.25 ml of Folin—Denis reagent was
added and then, after about one minute, 1.25 ml of saturated
Na2 CO3 was added. The compounds were mixed and allowed to
stand for one half hour at room temperature. If the test
tube became cloudy or a precipitate formed the solution was
filtered or centrifuged before the absorbance readings were
taken. After one—half hour absorbance readings were made
at 760 mu with a Beckman DU spectrophotometer, using phenol
as the standard. The phenol analysis was determined at the
same time as the guaiacol analysis and absorbances read 0.5
hour apart.

The Folin-Denis reagent was made by adding 25 mg sodium
tungstate, 5 gm phosphomolybdic acid, and 12.5 ml of
phosphoric acid to 187.5 ml distilled water. This mixture
was refluxed for two hours, cooled overnight and diluted
to a volume of 250 ml with distilled water. The solution
was then stored in a cool dark place until used.

The saturated sodium carbonate solution was made by
adding 35 mg of anhydrous sodium carbonate to 100 ml distilled

water. All of the sodium carbonate went into solution when

37

the solution was heated to 80°C. The solution was then allowed
to cool overnight. This super-saturated solution was seeded
with crystals of sodium carbonate and the supernatant

saturated solution was removed as needed.

Statistical Analysis

 

Standard procedures for analysis of variance, orthogonal
contrasts, correlation coefficients, and multiple linear
regressions as outlined by Steel and Torrie (1960) were

employed.

RESULTS AND DISCUSSION

Change in Lignin Structure During Digestion

 

Differences were found in the lignin ultraviolet charac-
teristics between forages and their respective feces which
indicates that the lignin molecule had undergone changes
during the digestion process. These changes will be dis-
cussed in some detail, first for grasses and legumes
separately and then for the combined grasses and legumes.

Grasses. The lignin extracted from grasses differed
significantly from the lignin extracted from the corresponding
feces in five of the nine lignin characteristics studied
(Table 3). The absorbtivities of the lignin at 265 and 270
ml were significantly less in the forage lignin than in the
fecal lignin. Wexler (1964) indicated that the difference
spectra of phenolic substances may reveal negative absorption
in regions where the neutral form absorbs more strongly than
the ionized or phenolate form. Both the forage and fecal
lignin had negative absorption in the range of 280 to 260 mu
which indicated that the difference spectra were similar in
shape, but not in the amount of absorbtivity (Figure 4).

The difference in absorbtivities between the forage and fecal

38

39

mo.v m at mooom Eoph
Ho.v m pm mooom EOL%

.a manme Lon mm oEMm soapmcwflmmo

mummgfio omnpom Ayv
mummmflo ommgom

oflpmflgopompmso AUV

 

 

ooNszCm mommmgm ma u 2 nos

omm%amcm mmEdmmH w n z ADV

ooNszcw mmwmgog Hm n 2 Amy

:>.NH mm.m~ mm.am mo.ww ma.zw ww.mm Fm.mw Fm.mw mo.mw 18 4 .m

mo.oH mm.o mm.o 0H.OH mm.o m©.o mo.on mm.o m:.o .Q.o oapmm .m

mH.oa mw.o wH.H m:.ow mm.m om.m :m.ow mm.a m©.H mom R .N

mm.ou,mm.mfl onzm.am ma.ma o.ma on o.mm mo.an a>.ma Amy m.mm pmanm :8 .m

no.0“ mm.© onmm.m mm.oa mm.a mm.m mw.oa mm.: mm.w mmm .m

H:.oa mm.m onmo.m mm.oH mo.m mw.m mm.oH om.m Amvwm.: com .2

mm.oa Hw.al onmm.ml 52.0“ Hm.OI Amvzm.al mm.ow :H.HI Amvmm.ml omm .m

gm.oH :H.H- Amvza.m- mm.oa am.o- gm.H- sm.ow mm.o- Amvga.m- mom .m

mm.ow mm.m Ho.m m:.ow Hm.a mm.o 03.0“ mm.m wm.a mam .H
mm mooom ommpom mm mmomm mmmgom mm mmomm mmmtom

onmommmgw

onmoEswmq vooQHQEoo

Am

onoflpmflgopomtmco

 

.Amm H cmmEV mumpmdmm ocm UosHQEoo mommmpw pew mmEson map Mo
.>.D cflcwfia CH momcmno co mmmoOLd Coflpmmmfio mo poommm one .m magma

moflpmfltopompmgo

40

Figure 4. The difference spectra of a sample forage lignin
preparation and a sample feces lignin preparation.

 

Forage lignin %~ =

Fecal lignin x

 

ma—i’

41

Figure 4

.1.

of-

0mm

oom

AJEV cpmcmfim>m3

0mm

.

OJN
_

 

 

ty

0.3...

Absorbtivi

o.mHl

O
O
("\l

1

 

 

 

 

omm

..Ill." 1. |..II."IUPI.. 1|-..

42

lignins in this region must be due to the fact that the
difference between the neutral spectrum and alkaline spectrum
of the forage lignin is greater than that of the fecal lignin
(Figure 5). These differences indicated that the forage
lignin contained more ionizable groups on the benzene ring
than did the fecal lignin.

The absorbtivity of the lignin at 300 mu was signif-
icantly greater in the forage lignin than in the fecal
lignin. Goldschmid (1954) found that the 300 mu maximum
of the difference spectra is a characteristic of phenolic
hydroxyl groups. This suggests that the observed differences
between the forage and the fecal lignins were due to dif—
ferences in amount of phenolic hydroxyl groups present in the
lignins. However, the direct spectra data showed that there
was no significant difference between the forage lignin and
the fecal lignin in amount of phenolic hydroxyl groups.

The absorbtivity of the extracted lignin at 337 ml was
significantly greater in the forage lignin than in the fecal
lignin. Wexler (1964) found that the interference of
vanillin, syringealdehyde and similar conjugated phenol
compounds can be detected by the appearance of a peak at
about 340rm1 on the difference spectrum. The difference
spectrum of the forage in Figure 4 showed a peak at about
340 mu which suggests that in the forage lignins there
were more conjugated phenol compounds present than in the

fecal lignins.

43

Figure 5. The direct spectra of a sample
preparation and a sample feces

Forage direct spectra
in sodium hydroxide
in phosphate buffer

Feces direct spectra
in sodium hydroxide
in phosphate buffer

forage lignin
lignin preparation.

44

Anew npwcmdm>m3

 

 

 

02m omm com 0mm com a cam omm
O . a. A ~ . .
.m ..(x

.3. //
V...
Momnfilx /fl
m XX .. / \*\\*I

w . .. \ Jr
.D M . .
APON; X /\ 9r .

J» \x. /./ it.)
{ Il'li
.mm ..
i (f

w . /./
m. .. /
m .om- Jr //

 

45

There was a significant decrease in the mu shift (defined
in Table 1) from the forage lignin to the fecal lignin. Wexler
(1964) found that methoxyl groups tended to stabilize the
absorption at 280 mu and the mu shift decreased as the pro-
portion of hydroxyl to methoxyl and other ether groups
decreased. It would appear that either hydroxyl groups were
lost or methoxyl groups were formed during digestion but since
the % POH was not different this would indicate that methoxyl
groups were formed during the digestion process.

There were no significant differences between forage
lignin and fecal lignin in the absorbtivity at 243 mu. There
are no differentially ionizable groups on the lignin molecule
that absorb at the 243 mu wavelength. If a difference had
been present,it would have been due to something other than
ionizable groups on the benzene ring.

The ratio OD and A mu characteristics were not signif-
cantly different between forage and fecal lignins indicating
similarity in the difference spectra of lignin from these two
sources.

Legumes. The forage lignins significantly differed from
their respective fecal lignins in two of the nine lignin
characteristics studied (Table 3). The absorbtivity on the
difference spectrum at 270 mu was less in the forage lignins
than in the fecal lignins which was similar to that found for
grasses, only the difference was not as great in legumes as

in grasses. The standard error of the mean was higher in

46

the legumes and the variation was greater between samples,
causing the difference to be less than that observed in the
grasses. There was a significant decrease in the mu shift
for the fecal lignin as compared to the forage lignin which
indicates the loss of hydroxyl groups or the formation of
methoxyl groups in the lignin molecule. Since there was no
difference between the forage and feces in the absorbtivity
on the difference spectrum at 300 mu or in the phenolic
hydroxyl content it can be assumed that there was a formation
of methoxyl groups on the lignin mOlecule without the loss of
a significant amount of phenolic hydroxyl groups. The
absence of any other differences indicates that the general
form of the spectra, both direct and difference, were similar
in shape and size for both lignins. Therefore the lignin
molecule does not undergo as much change during digestion in
legumes as in grasses.

Combined Legumes and Grasses. When data for legumes and

 

grasses were combined the forage lignin significantly dif-
fered from the fecal lignin in four of the nine lignin
characteristics (Table 3). Differences in these four char-
acteristics were also significant in grasses and two of
these four were significant in the legumes. The combined
analysis showed no difference on the difference spectrum at
337 mu. The absorbtivity of the forage lignin was signif—
icantly less at 265 and 270 mu than was the fecal lignin

because the difference between the neutral spectrum and

 

.vl:y
I'm.

n-n\

l .

f
I
{l‘l

1r.-

'hm

‘ w

r.
'7‘

...,

L‘

 

47

alkaline spectrum of the forage lignin was greater than that
of the fecal lignin.

Decreased values for the absorbtivity at 300 mu on the
difference spectrum, my shift, and phenolic hydroxyl content
were obtained from fecal lignin when compared to forage lignin.
The absorbtivity of the forage lignin at 300 m1 was signif—
icantly greater than that of fecal lignin which indicated
that the amount of phenolic hydroxyl groups decreased during
digestion. The decreased mu shift indicated that the ratio
of phenolic hydroxyl groups to methoxyl groups was less in
the fecal lignin than in the forage lignin.

These data indicate that there was a change in the
lignin molecule during the digestion process and that the
change was greater in grass lignins than in legume lignins.
The change might be a reduction in the amount of phenolic
hydroxyl and methoxyl groups attached to the benzene ring
but with a proportionately greater reduction in the amount
of the phenolic hydroxyl groups than in the amount of methoxyl
groups. Another possible explanation for this change is
that a phenolic hydroxyl group is replaced on the ring by a
methoxyl group as shown below (Brewster and McEwen, 1961).
There are pH conditions in the gastro-intestinal tract that

would allow this to take place if a methyl donor was present.

 

2m
‘3‘“

7‘5
...“

:"fi c
I...“ I.
'. u
‘2‘ ‘
N4 |
“.'-~..‘
V
y
’3: n
V..b u
r: P‘
5‘“. u,
2" :”y
‘ ~V‘
"f‘lA‘

48

 

HZCOH HZCOH
CH H
H I
CH H
methyl > [::]
+
donor

OH OCH
3

Differences Between Grasses and Legumes in the Lignin
Ultraviolet Characteristics

 

 

There were significant differences between the grasses
and legumes in twelve of the fourteen lignin ultraviolet
characteristics studied (Table 4). The grass lignins had
significantly greater absorbtivities at 243, 300, and 337
mu and significantly greater negative absorbtivities at 265
and 270 mu. The peak at about 340 mu in the difference
spectrum of grass lignins indicated the presences of vanillin
and syringealdehyde or similar conjugated phenols in the
lignin preparations but no such prominent peak was noted in
the difference spectrum of legumes lignins. The conjugated
phenols in the lignin preparations tend to increase the
absorbtivities at 243 and 300 mu but probably not enough to
account for all of the differences noted between the legume
and grass lignins. The grass lignin preparations have three
absorbtivity peaks on the difference spectrum,a 240, 300, and

340 mu; whereas the legume lignin preparations have only two

49

Table 4. Differences in lignin ultraviolet characteristics

between legumes and grasses (mean t SE).

 

 

 

(b) Legumes differ

from grasses at P <.01.

Characteristics<a> Legume Grass SE
243 mu —1.55E:; 8.95 i 0.94
265 mu -8.05 —15.57 i 0.48
270 mu —6.55(b) —15.62 i 0.51
300 mu 8.04(°) 15.29 i 0.90
337 mu 5.63(b) 25.41 i 0.44
Guaiacol 6'5O(§i 14.33 i 0.59
Phenol 213.8 487.3 1 5.2
Ratio 0.0. 0.62 0.35 i 0.10
mu shift 27.3 (b) 23.2 i 1.4
A mu 46.3 (b) 99.2 i 1.4
300 POH 0.59 0.62 i 0.10
250 POH -0.08(b) 0.08 i 0.06
250 c POH 0.12(b) 0.25 i 0.02
% POH 7.27(b) 3.49 i 1.24

(a) Characteristic designations same as for Table 1.

 

AKL
c K
0

9A”
1

my.

he

NI 6

F-
..‘_

50

absorbtivity peaks on the difference spectrum, at 300 and
possibly at 340 NJ. Lignins from legumes had a negative
absorbtivity over a wider range of wavelengths (280—235 mu)
than did the grasses (280-250 mu). The grass lignins had a
greater negative absortivity in the region of 280 to 260 mu
than did the legume lignins {—6.25 vs -3.25). The difference
spectra of grass lignin differed from that of the legume
lignin in both the absolute values for absorbtivity at a
given wavelength and the shape of the respective spectro-
grams (Figure 6).

The grass lignin preparations contained more free guaiacol-
like compounds than did the legume lignin preparations (14.33
vs 6.5 mg/gm lignin). The grass lignin preparations also had
more free phenol—like compounds present than did the legume
lignin preparations (487 vs 214 mg/gm lignin). These dif—
ferences are similar to those observed between the difference
spectra of the forage lignins. This indicates that the
legume lignin is either more tightly bound and resistant to
chemical breakdown or has different monomeric units than does
the grass lignin.

The distance in my between the maximum peak and minor
peak on the difference spectrum was greater for the grass
lignins than for the legume lignins. The maximum peak in
the grass lignin was found at approximately 340 mu whereas
the maximum peak in the legume lignin was at approximately

300 mu. The minor peak of the grass lignin was distinct and

51

Figure 6. The difference spectra

of lignin from a grass
and a legume.

Legume

 

0

Grass

1!

52

07m

9m:

omm

oom

AJEV zpwsmam>m3

omm

n

omm

~

 

Absorbtgvity
u;

0.0:

o.mH1

 

Figure 6

O

O

(\I
\

53

occurred at approximately 240rm1 while the minor peak in

the legume lignin was indistinct and found at approximately
340 mLL The difference in the number of peaks and the
position of the maximum peak shows that there were different
compounds ionizing in the grass than in the legume samples.
This indicates a difference in basic structure.

The direct spectra of the grass and legume lignins had
similar differences to those found on the difference spectra.
Also, the absolute absorbtivity of the grass lignin was
greater than the absolute absorbtivity of the legume lignin
throughout the wavelengths scanned (Figure 7). This shows
that the lignin molecule of grasses contained a greater
proportion of aromatic compounds which had a greater absolute
absorbtivity than the aromatic compounds that made up the
legume lignin molecule. The shift in mu also indicated that
the basic structures are different in that the ratio of
phenolic hydroxyl groups to methoxyl groups is greater in the
legume lignins than in the grass lignins. There are four
methods for estimating the phenolic hydroxyl content of the
lignin. These were developed using ligno—sulfonate pre-
parations to determine the conversion factors to multiply
times a specific ultraviolet measurement. These methods
showed varying results in the data on forages. The 300 POH
method showed no difference in the percent phenolic hydroxyl
groups between the grass and legume lignins. The 250 POH

and 250 C POH methods showed that the grass lignins contain

54

Figure 7. The direct spectra of lignin from a grass and a
legume.

Legume direct spectra
in sodium hydroxide so
in phOSphate buffer .94

 

 

Grass direct spectra
in sodium hydroxide K
in phosphate buffer -—m—a&—_——

 

55

Figure 7

Absorbtivit

0.0m-

y

O

O

m
1

O
O
:1-

0.0mL

 

 

omm

owm

AJEV mnpmcoam>m3

com

4

 

 

 

omm

.‘ll‘llu ‘J'I

 

omm

Ill:

56

a significantly greater percent phenolic hydroxyl groups than
did legume lignins. The % POH method showed that the legume
lignins contain a significantly greater percent phenolic
hydroxyl groups than did grass lignins. Wexler (1964) con—
Cluded that the phenolic hydroxyl values obtained from the
difference spectrum were more accurate than the value obtained
from the direct spectra when using none pure compounds.
Therefore there are more phenolic hydroxyl groups in lignin
from grasses than in lignin from legumes and also more
methoxyl groups because the ratio of phenolic hydroxyl groups
to methoxyl groups is less in the lignin from grasses.

There was a difference between legume lignin preparations
and grass lignin preparations in the chemical make up of the
lignin molecule. The grass lignin molecule had a greater
proportion of aromatic compounds with a greater specific
absorbtivity than the aromatic compounds which make up the
legume lignin molecule. These aromatic compounds seem to
have more methoxyl groups and/or less phenolic hydroxyl groups
attached to the benzene ring. The percentage of this aromatic
compound present in the lignin sample could be a factor re-

lated to the digestibility of the forage.

57

Alfalfa Hay Early Cut (6502—01) vs Late Cut (6502-02)

 

Three of the fourteen lignin characteristics were signif-
icantly different between the early cut alfalfa hay and the late
cut hay (Table 5). The late cut alfalfa hay had a significantly
greater negative absorbtivity than did the early cut hay at
265 mu while the 250 POH and 250 C POH lignin values showed
that the early cut alfalfa hay had more phenolic hydroxyl
groups than did the late cut hay. The negative absorbtivity
on the difference spectra at 265 mm is due in part to the
absorption by the non-ionized phenolic hydroxyl group at
this wavelength compared to no absorption by the ionized
phenolic hydroxyl group. Greater values for this difference
(or the greater the negative absorbtivity) indicate increased
amounts of phenolic hydroxyl groups present. This is dif-
ferent than the observed difference at the 250 POH and 250
C POH lignin characteristics but similar to the trends
observed in the 300 POH and % POH lignin characteristics. If
there had been a difference in the amount of phenolic hydroxyl
groups present then there would have to have been a difference
in the methoxyl content since there were no observed dif-
ferences between the ratios of phenolic hydroxyl group to
methoxyl groups. Therefore as the alfalfa plant matures there
is an increase in the phenolic hydroxyl content and methoxyl
content. Either or both might affect the digestibility of

the hay.

58

Table 5. Differences in lignin ultraviolet characteristics
for 20 samples of forages (mean i standard error).

 

Lignin Ultraviolet Characteristics(a)
Forage Sample(b) 243 mu 265 mg_ 270 mu 300 mp 337 mu

 

6401-01 3.02 — 3.44f — 3.81f 9.27 7.10
6401—02 7.63 - 0.20e — 0.768 11.08 8.09
6502—01 3.37 - 3.51G — 3.84 6.81 5.24
6502—02 — 2.04 - 7.59H — 6.32 7.73 4.64
6701—21 - 8.06 -13.47Kl —10.90k 12.25 6.96k
6701-22 — 5.75 -10.16KJ — 9.01k 9.70 8.13k
6701—01 - 5.66 — 8.40I - 6.501 7.69 5.041
6702-OV - 6.55 -12.34P -10.15 9.90 5.78
6702—00 - 2.92 — 8.99O — 7.58 8.26 6.35
6703—02 — 4.14 - 8.76 - 7.18 6.98 3.35-
6703—03 — 4.11 — 9.27 — 7.92 6.22 3.81
6704—01 — 4.83 - 9.24 — 8.04 6.19 3.59
6201—11 2.31 - 5.96 - 5.47 5.58 4.91N
6201-12 5.99 - 3.27 — 4.17 4.83 5.88M
6401-51 9.88 -19.13R -18.52R 19.38Q 28.39Q
6401—52 4.37 —10.88Q -10.09Q 9.53R 12.71R
6401-41 5.72 —17.79 -17.17 15.82 24.97
6401-42 8.05 -l4.96 —l4.96 14.04 24.16
6401-31 15.45 —17.30T —18.17t 19.78S 36.46S
6401-32 10.24 -13.34S —14.74S 13.19T 25.76T
Std. error 10.94 10.48 :0.51 £0.90 £0.44

 

(a) Characteristic designations are same as for Table 1.

(b) Forage sample descriptionsare in Table 2.

59

Table 5 (continued).

 

Lignin Ultraviolet Characteristicsfal

Forage Sample(b) Guaiacol Phenol Ratio OD mu shift A mu

 

6401—01 9.11 283.2 0.43 29.5 47.0
6401-02 8.03 294.7 0.78 29.0 47.0
6502—01 7.51 204.4 0.52 29.0 47.0
6502—02 6.07 171.1 0.62 30.5 45.5
6701—21 «6.00 270.3 0.62 26.0 46.0J
6701—22 6.93 273.1 0.86 27.5 40.5L
6701—01 4.61 184.1 0.74 26.5 46.5
6702—0v 8.91 226.9 0.61 28.5 46.0
6702—00 8.49 216.0 0.83 28.0 45.5
6703—02 5.08 170.7 0.48 27.5 47.5
6703-03 6.35 167.4 0.50 24.5P 44.0
6704-01 5.64 160.4 0.46 30.50 45.0
6201—11 3.43 175.0 0.42n 23.0 58.5M
6201—12 4.86 196.1 0.76m 22.5 42.0N
6401—51 19.78Q 515.6q 0.34 23.5 99.5R
6401—52 9.50R 302.1? 0.35 23.5 103.5Q
6401—41 11.34 493.4 0.25 26.0 99.5
6401-52 14.75 509.9 0.33 24.0 97.0
6401—31 19.18S 693.4S 0.42 23.0 98.5
6401—32 11.42T 412.8t 0.41 19.0 97.0
Std. error :0.59 :35.2 :0.10 11.4 il.4

g

(a) Characteristic designations are same as for Table l.

(b) Forage sample descriptions are in Table 2.

60

Table 5 (continued).

 

Lignin Ultraviolet Characteristics<a>

 

Forage Samp1e<b) 300 POH , 250 POH 250 c POH % POH
6401—01 0.63F 0.10F 0.20F 5.27e
6401—02 1.07E 0.43E 0.43E 4.49f
6502—01 0.58 0.12G 0.23G ’6.66
6502—02 0.71 —0.19H 0.08H 7.79
6701—21 0.66 —0.27 0.04 5.30
6701—22 0.57 —0 19 0.09 4.92
6701-01 0.52 —0.22 0.05 5.63
6702-0v 0.52 —0.22 0.03 10.66
6702-00 0.48 -0.13 0.04 9.31
6703—02 0.59 —0.24 0.03 10.34
6703-03 0.48 —0.23 0.03 10.63
6704-10 0.53 —0.29 0.03 11.26
6201-11 0.52 —0.02n 0.14N 4.97
6201-12 0.44 0.18m 0.24M 4.50
6401—51 0.65 0.07 0.25Q 3.28
6401—52 0.52 0.04 0.17B 3.81
6401—41 0.60 —0.03 0.20P 3.11
6401—42 0.68 0.05 0.280 4.08
6401—31 0.65 0.18 0.32 3.32
6401-32 0.60 0.13 0.29 3.35
Std. error iO.lO i0.06 :0.02 :1.24

 

(a) Characteristic designations are same as for Table l.

(b) Forage sample descriptiomsare in Table 2.

Table 5

G >H
I >K

i >k

J>l

at

at

at

at

at

(continued).

P <.01
P <.01

P <.05

61

>N
>n
>p
>P
>R
>I‘
>T

>t

at

at

at

at

at

at

at

.01
.05
.05
.01
.Ol
.05
.Ol

.05

62

Alfalfa Hay First Cutting (6401-01) vs
Second Cutting (6401—02)

 

 

Six of the fourteen lignin characteristics were signife
icantly different between the first cutting of alfalfa hay
and the second cutting hay (Table 5). All six of these
values, 265 and 270 mp, 300 POH, 250 POH, 250 C POH, and
% POH, are affected by the amount of phenolic hydroxyl groups
Fpresent in the lignin. The absorbtivities at 265 and 270 mu
and the % POH indicate that the first cutting hays have more
phenolic hydroxyl groups present than the second cutting hays,
but the 300 POH, 250 POH and 250 C POH shows that the second
cutting hays contain a greater percentage of phenolic hydroxyl
groups than the first cutting hays. Since the ratio of
phenolic hydroxyl groups to methoxyl groups is not signif-
icantly different then there is probably little or no dif—
ference in the structure of the lignin molecule between first

and second cutting alfalfa hay.

Birdsfoot Trefoil Hay First Cutting (6201-11) vs
Second Cutting (6201—12)

 

 

Five of the lignin characteristics were significantly
different between the first cutting of birdsfoot trefoil hay
and the second cutting hay (Table 5). The absorbtivity at
337 mu was significantly greater in the second cutting than
in the first cutting trefoil hay which indicates that the

second cutting lignin preparations contained more conjugated

 

63

phenol compounds, such as vanillin, than did the first cutting
lignin preparations. The greater amount of conjugated phenol
compounds in the second cutting hays caused the ratio O.D. to
be significantly different and it also significantly reduced
the m1 difference between the maximum peak and minor peak on
the difference spectra. The 250 POH and 250 C POH values
showed the second cutting hay contained significantly more
phenolic hydroxyl group than did the first cutting but the 300
POH and % POH values showed no difference between the cuttings.
Therefore there was a greater amount of phenolic hydroxyl
groups in the second cutting hay, which might have been due
to the presence of more conjugated phenol compounds present

in the second cutting hay.

Alfalfa Hay vs Birdsfoot Trefoil Hay

 

Ten of the fourteen lignin characteristics were signif-
icantly different between alfalfa hay and trefoil hay
(Table 6). The absorbtivities at 265, 270, 300 and 337 mu
were significantly greater in the alfalfa lignin samples
than in the trefoil lignin samples. The alfalfa hay had a
greater ratio of phenolic hydroxyl groups to methoxyl groups
than did the trefoil hay and the trefoil hay had a greater mu
difference between the maximum and minor peaks than did the
alfalfa hay. Three of the methods for determining amount of
phenolic hydroxyl groups present showed that the alfalfa hay

contained more phenolic hydroxyl groups than did the trefoil

 

64

Table 6. Differences in lignin ultraviolet characteristics
between alfalfa and birdsfoot trefoil (mean t SE).

 

 

Characteristic<a> Alfalfa Birdsfoot Trefoil SE
265 mu -1.82(b) —4.61 10.48
270 mu —2.28(°) —4.82 :0.51
300 mu 10.17(°) 5.20 10.90
337 mu 7.59(°) 5.39 :o.44
Guaiacol 8.57(c) 4.14 $0.59
mu shift 29.2 (C) 22.7 :1.4
A mu 47.0 (b) 50.2 11.4
300 POH 0.85(°) 0.48 :0.10
250 POH _ 0.26(°) 0.10 10.06
250 c POH 0.31(C) 0.19 :0.02

 

(a) Characteristic designations same as for Table l.
(b) Alfalfa differs from birdsfoot trefoil at P <.05

(c) Alfalfa differs from birdsfoot trefoil at P <.01

65

hay and the other method showed no difference between the
species.

Apparently alfalfa lignins have a greater proportion of
aromatic compounds that contain more phenolic hydroxyl groups
than do the aromatic compounds that make up trefoil lignins.
This would explain the difference in the absorbtivities at
specific wavelengths on the difference curve, the difference
in the amount of guaiacol—like compounds in the lignin pre-

paration, the difference in the mu shift (or ratio of phenolic

 

hydroxyl groups to methoxyl groups), the difference in the mu
distance between the maximum and minor peaks on the difference

spectrum and the greater phenolic hydroxyl content.

Alfalfa Silage (6701-21 & 22) vs
Alfalfa Hay (6701-01)

The only difference between the silages and hay in the
lignin characteristic studied were the absorbtivities at 265,
270 and 337 mu (Table 5). The silages contained significantly
more conjugated phenol compounds than did the hay. The silages
had significantly greater negative absorbtivities at 265 and
270 mu than did the hay. Therefore there is very little dif-
ference in the lignin molecules of alfalfa silage and alfalfa

hay.

66

Alfalfa Silage (6701-21) vs Alfalfa Silage
Treated with Formic Acid (6701—22)

The alfalfa silage had a greater negative absorbtivity
at 265 mp than did the formic acid treated silage and a greater
mu difference between the maximum peak and minor peak on the
difference spectrum than did the treated silage (Table 5).
This shows that the differences in the lignins in these two

silages were very small.

Alfalfa Variety Vernal (6702-OV) vs
Variety DuPuits (6702-OD)

The only difference in the lignin characteristics between
the two varieties of alfalfa was that Vernal had a greater
negative absorbtivity at 265 mu than did DuPuits (Table 5).
Therefore, there is no difference in the lignin structures
between the two varieties of alfalfa. The forages were
harvested on the same day but the variety DuPuits was more

mature than the variety Vernal.

Alfalfa Breeder Line (6704-01) vs
Breeder Line (6703—03)

The ratio of phenolic hydroxyl groups to methoxyl groups
was the only characteristic that showed a significant dif-
ference between the breeder lines (Table 5). Breeder line

6704-01 had a greater ratio than did breeder line 6703—03

67

while the phenolic hydroxyl content was similar. Therefore,
the difference must be due to the fact that the breeder line
6704-01 had fewer methoxyl groups than the other line. This
greater methoxyl content might be contributing to the low

digestible dry matter content of the 6703-03 line.

Bromegrass Hgy First Cutting (6401-51) vs
Second Cutting (6401—52)

 

 

First cutting bromegrass hay had significantly greater
negative absorbtivities at 265 and 270 mu and had signif—
icantly greater absorbtivities at 300 and 337 m1 than did
the second cutting hay (Table 5). The first cutting hay
also had more free guaiacol-like and phenol-like compounds
in the lignin preparation than did the second cutting hay.
The second cutting hay had a greater difference in mu
between the maximum and minor peaks than did the first
cutting hay. These observations indicate that there is a'
small difference in structure of the compounds analyzed,
such as position of the groups on the benzene ring, even
though there was no difference in the ratios of phenolic

hydroxyl groups to methoxyl groups.

 

68

Reed Canary Grass HaygFirst Cutting (6401—41) vs
Sgcond Cutting (6401-42)

There were significantly more phenolic hydroxyl groups
present in the second cutting hay than in the first cutting
hay by the 250 C POH method. There is a difference, but just
what this difference is cannot be identified at present since
the methods for determining phenolic hydroxyl content were
based on wood lignin and give variable values for forage

lignin.

Siberian Reed Canary Grass Hay First Cutting
(6401-31) vs Second Cutting (6401-32)

The first cutting hay had significantly greater negative
absorbtivities at 265 and 270 m1 and had significantly
greater absorbtivities at 300 and 337 mu than did the second
cutting hay (Table 5). The first cutting hay also had more
free guaiacol-like and phenol-like compounds in the lignin
preparations than did the second cutting hay. These
observations indicate that there were differences in the

structure of the lignins between the two different cuttings.

Reed Canary Grass Hay vs Siberian Reed
Canary Grass Hay

 

Siberian reed canary grass hay had significantly greater

absorbtivities at 243 and 337 mu than did reed canary grass hay

69

which indicates that there are more conjugated phenol compounds,
such as vanillin, present in Siberian reed canary grass lignin
preparation (Table 7). Reed canary grass hay had a greater
ratio of phenolic hydroxyl group to methoxyl groups than did
Siberian reed canary grass hay. Siberian reed canary grass

hay contains more methoxyl groups than does reed canary grass
hay which might be a factor since reed canary grass is more

digestible than Siberian reed canary grass hay.

Bromegrass Hay vs Reed Canary Grass and
Siberian Reed Canarthrass Hays

Reed canary grass hays had a significantly greater
negative absorbtivity at 270 m1 and had a significantly
greater absorbtivity at 337 mu than did bromegrass hay which
indicates that there were more conjugated phenol compounds
present in the reed canary grass lignin preparations (Table
8). There was a significantly greater phenolic hydroxyl
content in the reed canary grass hays than in the bromegrass
hay with the 250 C POH method, and the reed canary grass hays

had the most conjugated phenol compounds present.
Lignin Determination
The correlation between Stafford's method (1960) for

lignin content and the 72% sulfuric acid method was not

significantly different from zero (Table 9). Stafford's

70

Table 7.

canary grass (mean t SE).

Differences in lignin ultraviolet characteristics
between reed canary grass and Siberian reed

 

( ) Reed Canary
Characteristics 9

Siberian Reed

 

 

Grass Canary Grass SE
243 6 88(F) 12.84 10.94
337 24.56(D) 31.11 10.44
mu shift 25.0 (b) 21.0 11.4
250 POH 0 01(C) 0.15 10.06
250 c POH 0 24(9) 0.30 10.02
(a) Characteristic designations same as for Table l.
(b) Reed canary grass differs from Siberian reed canary grass
at P <.01.
(C) Reed canary grass differs from Siberian reed canary grass
at P <.05.
Table 8. Differences in lignin ultraviolet characteristics

between the reed canary grasses (common and

Siberian) and bromegrass (mane i SE).

 

Reed Canary

(a)

Characteristics

 

Grasses Bromegrass SE
270 16.26(°) 14.30 0.51
337 27.84(°) 20.55 0.44
250 c POH 0.28(C) 0.21 0.02

 

(a)

Characteristic designations same as for Table l.

(b) Reed canary grasses differ from bromegrass at P <.O5.
(C) Reed canary grasses differ from bromegrass at P <.01.

71

Table 9. Comparison of two methods for the determination of
lignin content in forages.

 

 

Lignin Lignin
by Stafford's by 72% Sulfuric
Forage Method Acid Method
% %

6401—01 9.20 6.31
6401-02 13.91 9.54
6502-01 10.15 8.03
6502—02 8.87 10.81
6701-21 6.16 6.33
6701-22 7.54 6.78
6701-01 7.42 8.96
6201-11 4.85 8.73
6201-12 6.96 8.50
6702—OV 7.13 5.00
6702-OD 8.40 6.11
6703-02 6.57 8.08
6703-03 8.02 7.89
6704—01 6.26 6.93
6401—51 9.02 2.85
6401-52 6.26 3.86
6401—41 5.62 3.12
6401—42 6.48 2.88
6401-31 8.94 2.74
6401-32 7.99 4.14

 

r = 0.25; P ‘<l5

72

suggested method is based on the mg of guaiacol-like

compounds present in lignin. This indicates that the amount
of free guaiacol-like compounds in the lignin preparation is
not related to the amount of lignin in that preparation and
should not be used to estimate the amount of lignin present

in the forage.

The Relationship of the Extracted Material
to the Lignin Content in the Forage

The correlation of the amount of extracted material and
the mg of the lignin in the original ADF residue was deter-
mined on legumes and grasses separately (Table 10). Neither
correlation significantly differed from zero which means that
a varying amount of cellulose was extracted. Some cellulose
may have been so tightly bound to the lignin that both were
extracted simultaneously. The extracted cellulose does not
affect the spectrograms obtained because carbohydrates in
neutral or alkaline solutions do not absorb in the ultra-

violet region scanned (Aulin-Erdtman, 1949).

Relationship Between the Lignin Characteristics
Animal Data and Lignin Content

 

Combined Legumes and Grasses. Correlation coefficients
were determined between the values per unit of acid detergent

fiber for the fourteen lignin characteristics and three

73

Table 10. Comparison of the amount of extracted material to
the mg of lignin in the forage.

 

 

 

 

mg_extracted material mg lignin

0.1 gm ADF 0.1 gm ADF

Forage by weight loss by analysis

E5111:
6401—01 31.8 18.9
6401—02 29.4 24.9
6502-01 32.6 24.2
6502-02 27.4 22.0
6701-21 32.7 15.5
6701-22 33.6 17.2
6701-01 38.4 22.6
6201-11 46.1 26.9
6201-12 34.0 27.5
6702—OV 33.7 15.6
6702-OD 34.7 17.8
6703-02 34.3 25.3
6703—03 34.7 23.6
6704-01 32.0 24.2
Grass

6401—51 27.5 8.1
6401—52 26.2 12.6
6401-41 28.8 9.6
6401—42 20.4 8.7
6401-31 28.2 8.1
6401-32 32.4 12.9

 

Legumes r 0.25; P <.20

Grasses r 0.42; P <.22

74

measures of animal response as well as lignin content

(Table 11). Most of the correlations were not significantly
different from zero and those that were significantly dif-
ferent from zero were of low magnitude and not useful as
single predictors of animal performance or lignin content.

Legumes. Correlation coefficients were determined for
the legume forages between the lignin characteristics
(expressed as per unit of acid detergent fiber and per gm
lignin) and animal data and lignin content (Table 12). Most
of the correlations were not significantly different from
zero and those that did differ from zero were not of such
magnitude to be important biologically as single predictors
of animal preformance or lignin content.

Grasses. Correlation coefficients for the grass hays
were determined between the lignin characteristics (expressed
as per unit of acid detergent fiber and per gm lignin) animal
data and lignin content (Table 13). Several of the correlation
coefficients were not significantly different from zero or
were of such low magnitude as to be unimportant biologically
as single predictors of animal performance or lignin content.
The correlation coefficient between the phenol content and
digestible dry matter/cwt was -.98 and the correlation
coefficient between the absorbance at 337 mu on the dif-
ference spectra and digestible dry matter/cwt was -.96.

These two correlations with digestible dry matter/cwt were

of sufficient significance that they could be used as single

75

.m.o.< Em H.o pom ohm

.H mapme ca mm meow coapmcwfiwma

omoze ADV

on

 

 

mo.v m mo>Hm 33.0 A h zC¢
mm.u so. 1H. Ho.- as a
5:. ma. ma. Hm. peach :8
om. om.- me.n Ho.u no oapom
He. om.u om.- mo. mom a
mm.: mm. mm. :m.. mom 0 omm
oH.- mm. :1. ma.. mom 0mm
OH. mo.u ma. mm.: mom oom
mm.l HH.I mo. :m.l Apvwmm
Na. ON.I mo.l m:.l Aflvoom
mm. we. as. 10.- onosm
OH. am. am. so.. onmom
mm.u em. we. mH.: o mam
Ha. NH. me. Hm.- AQW%owosm
so. Ha. mm. Hm.u ona ooaoso

cacmfiq R p30\.2.o p30\.z.m xmz goppwz mam Amvmoflpmfipopompmco
oaoapmowao oaoaomomao

 

.cflcwfifl & ocw mpmo Hmeficm .ooQHnEoo mommmpm out
moEson map mo moflpmfigopomhmzo cficwfla map cooZpon mQOHpmHoggoo

.HH mant

76

.m.m.< Em H.o Hog 09m omone ADV

.H oHnmE CH mm meow mQOHpmcmeoQ

Amv

.mo.v m mo>Hm mm.A m Hc<

 

 

om. mm. mm. mo.- 15 a
mo. mo. mm. mo. puHsa as
mo. om.- m:.- mo. no oHosm
Hm.1m.u Hm.n mo. mom H
mm. me. me. mm.u mom 0 omm
.mm. mm. Hm. oH.- mom omm
mm. mo.- Hm. m:.n mom oom
mm.- mm. mH. mm. oo.- om. mH. Hm.- smm
1m.- om. mm. mo.u mm.- mo. mo. 11.- com
om. Hm. mm. as. mo. em. mm. 1 HH.- osm
Hm. om. am. am. as. so. mm. a HH.- mom
11. mT mo. mo. OH. mo. HH.- NH.- mam
1T . mT mH. me. H0. H1. oH. mm.- Hostsm
Hm.- mo. mo. gm. :H. as. mH. om.- HoooHoso
gov gov on on
chqu H pso\zQ p30\zm xmz Loppmz Ham AmvmoHpmHgopompmco
oHoHomowHo oHoHohomHo

 

«moESmoH mo mOprHpopowgmno chwHH 02p coozpmo mCOHpmHogmoo

.chwHH H out sumo HmEHcm

.mH mHQMB

77

.m.o.< Em H.o god who omoce ADV

.H oaose 2H

.Ho.v m mo>Hm mm.oa p mcm ocm

mm 65mm mQOHpmcmeoQ Amv

mo.v m mo>Hm Hm.OA p Hc<

 

 

om. so. to. o:.u as a
mo.. om. Hm. Hm.u smash :5
Hm. oo.u mm.u mo.u mo oHpom
so. me. am. mo.- mom H
mm.. mm.- om.u s:.u mom 0 omm
Ho.n 1H.u ms.u Hm.- mom 0mm
os.n mm.n we.» 1H. mom oom
Ho.u HH. os.a om.u mm.n om.u mm. mo.u Hmm
up. mo. m:.u ms.u mm.| sm.n em. w:.u oom
H». H:.| we. om. Hm. mm. mo.| ow. osm
mu. sm.n mm. mm. mm. mo. em. as. mom
01.: :H. om.u Hm.- mm.: mm.. mm.. os.u mam
om.: Ho.n mm.n mm.u ss.u mm.- om.u mm.- Hososa
ms.a mm.: mm.- om.- 11.- om.u em.u mm.u Hooonso
on on on on
chmHA m pzo\zm p3o\zm xmz Houpmz HMO VmoHpmHuopomgmco
oHoHpmome oHoHomomHo

 

.chwHH R ocm dump HmEHcm
amommmpw mo moHpmHLopomgmzo QHcmHH map cooZHoo mCOHpmHoLpoo

.MH meme

78

predictors of digestible dry matter/cwt.

Prediction quations for Digestible Dry Matter
and Digestible Dry Matter Intake/th

 

Multiple linear regression analysis was used to develop
prediction equations for digestible dry matter and digestible
dry matter intake/cwt. Six different equations were developed;
two for combined legumes and grasses, four for legumes (two
per unit A.D.F. and two per gm lignin), and two for grasses.
The importance of these equations is discussed below.

Combined Legumes and Grasses. The digestible dry matter

 

prediction equation that was developed using data for the
combined legumes and grasses had no biological importance due
to a low multiple correlation coefficient (R2) of .51 at
P <.28 (Table 14). This equation included eight independent
variables (lignin ultraviolet characteristics) and with this
many variables the mathematics would be too long and involved
to use as a practical predictor. The standard partial
regression coefficients indicate that the variables 250 C POH,
mu shift, and 243 mu were more important than the other five
variables in the equation.

A prediction equation for digestible dry matter/cwt was
deve1Oped using nine of the independent variables which had
a R2 of .80 at P <.01 (Table 14). The standard partial
regression coefficients were of about equal size indicating

that they were of equal importance in the equation. This

79

Table 14. Prediction equations on combined legume and grass
forages for digestible dry matter (DDM) and
digestible dry matter intake per hundred pounds
of live body weight (DDM/cwt).

DDM
9 = 46.11 - 220.30xl - 54°25X4 + 1.57x6 — 103.19x7 +
109.46X8 — lu3oO7X9 - 265.97Xll - lolqu2

R2 = .51

Standard partial regression coefficients.

.bf = —.17 bg = .20
on = -.73 b; = -.68
= * = _
6g .46 611 .15
b; = -.13 b§2 = —.26

Independent variables (see Table l for details).

x1 = 300 mu X5 = 337 mu xlO = 265 mu
X2 = 300 POH X6 = mu shift X11 = 270 mu
X3 = Phenol X7 = Guaiacol X12 = % POH
X“ = 250 C POH X8 = 250 POH X13 = A mu

x9 = 243 mu X14 = Ratio 00

DDM intake/cwt

y = 4.32 + 49.57XlO — 56.06Xll - 12.83X9 + 6.57X8 -
1‘03X14 + .05X6 - .50X3 — 18.40X7 - 7:06X5
R2 = .80
Standard partial regression coefficients.
b* = .46 b* = .22 b* = —.22
10 8 3
* = _. * = _. * = —ou0
bll 52 bl4 31 b7
* = -. = .28 b* = —.18
b9 27 bg 5

80

R2 value is not of sufficient size for precise prediction

of digestible dry matter intake/cwt. The low predictability
and the complicated mathematics of using several variables
in an equation would limit the usefulness and application of
these relationships although the variables could easily be
determined.

Legumes per unit A.D.F. The independent variables

 

used were good predictors of digestible dry matter and
digestible dry matter intake/cwt for legumes. Ten independent
variables were used in the prediction equation for digest—
ible dry matter and gave an R2 of .93 at P <.15 (Table 15).
The standard partial regression coefficients indicate that
the variables phenol, 250 C POH, 265 mu, mu shift were more
important than the other six variables in the equation. The
independent variables are easily obtained from the direct

and difference spectra and from chemical analysis for phenol.
The only difficulty would be using calculations involving ten
variables.

A prediction equation for digestible dry matter intake/
cwt was determined using seven independent variables which
gave an R2 of .96 at P <.001 (Table 15). The prediction
equation with a multiple correlation coefficient of this
magnitude would be a good predictor of digestible dry matter
intake/cwt even if the calculations would be difficult. The
standard partial regression coefficients were of the same

magnitude and of equal importance in the equation. The

81

Table 15. Prediction equations for DDM and DDM intake/cwt
on legume forages.

DDM
9 = -18.07 + 2078.09xl — 136.37X2 - 30.80x3 — 260 14Xu +
682.37X5 - 2304.43xll + 243.69x9 + 1963.81xlo —
15.89xlu + 2.08x6

R2 = .93

Standard partial regression coefficients.

bi = .64 bIl = -.93
bfi = -.48 b; = .21
b§ = —.80 bIO = .92
bfi = -.92 bi“ = -.51
b% = .50 b% = .84.

DDM intake/cwt

A

y = 5.05 + 49.74X10 — 4.48X + 1.98X + .22X -

8 3 l2
22.66X5 - 42.25X7 - 17.39Xl

R2 = .96

Standard partial regression coefficient.

*= *=_
blo .93 b5 .88
bg = -.85 b; = -.92
bg = .93 bi = —.91
b* = .89

12

 

82

difference in probability levels between the two R2's is due
to the amount of degrees of freedom available for error.

Per gm Lignin. Nine independent variables were used in

 

the prediction equation for digestible dry matter and gave
an R2 of .98 at P <.01 and the standard partial regression

coefficients indicate that the variables are of equal a3

... J F“

importance in the equation (Table 16). The prediction
equation for digestible dry matter intake/cwt was determined

using six variables which gave an R2 of .76 at P <.03

 

 

(Table 16). The standard partial regression coefficients L
show that the variables 265 and 270 mu were more important
than the other four variables in the equation.

Grass per unit A.D.F. The prediction equation for di—
gestible dry matter had an R2 of .90 at P <.15 with only
three independent variables (Table 17). The standard partial
regression coefficients showed that the variable mu shift was
more important than the other two in the equation. Such a
prediction equation is of biological importance due to its
high multiple correlation coefficient and ease with which
calculations can be made using only three variables.

The prediction equation for digestible dry matter intake/
cwt had an R2 of .99 at P <.005 with two independent variables
(Table 17). The standard partial regression coefficients were
of the same magnitude and of equal importance in developing
the equation. This prediction equation is of practical and

biological importance due to its very high multiple

83

Table 16. Prediction equations for DDM and DDM intake/cwt
on legume forages.

DDM 11

10 4 - 4.17X7 -

+ 181.17X8 - 4.70Xl)4 + 2.31X6

9 = 40.22 — 8'32X1 + 5.12X — 222.87X

1

1.91X9 + 0.85X3

R2 = .98

 

1r

Standard partial regression coefficients.

bil = -.97 b; = -.98 6% = .98
610 = .94 6% = —.88 bi, = -.72.
4:- if: =

bu .98 b3 .89 6g .98
DDM intake/cwt

9 = 0.18 + 0.7lx10 - 0.74xll - 3 23X4 + 0.83xl3 +

0.13xl _ 0.86x5
R2 = .76

Standard partial regression coefficients.

bio = .78 63 = —.50 bi = .53
b* = -.76 t = .60 b* = -.24

H*
u;
U1

84

Table 17. Prediction equations for DDM and DDM intake/cwt
on grass forages.

DDM

§= 15.60 + 1.83x6 + 26.09x1, - 2.74x3

R2 = .90

Standard partial regression coefficients

b = 1.83 b; = —.27

*
6
bi“ = .26

DDM intake/cwt

y = 3.64 - .74X3 — 6.03X9

R2 = .99

Standard partial regression coefficients

*=—. *=-.6O
b3 73 b9

85

correlation coefficient and ease of which calculations can
be made with only two variables.

The independent variable phenol content of ADF was used
in all of the prediction equations. Therefore it was
relatively important in all the equations, judged by the
magnitude of the standard partial regression coefficients,
and strongly suggests its importance as a determinant of
digestibility.

Any prediction equation is useful only if the independent
variables are easy to obtain in the laboratory and the R2
or predictability is sufficiently high to be of practical
importance. The prediction equations for grasses and legumes
are useful to get an estimation of the digestible dry matter
and digestible dry matter intake/cwt. The prediction
equations should not be used when the predictor values fall

outside of the ranges used to deveIOp the equations.

SUMMARY

The lignin molecule underwent a change during the
digestion process and this change was greater in grass
lignins than in legume lignins. The change might have been
a reduction in the amount of phenolic hydroxyl and methoxyl
groups attached to the benzene ring, but with a proportionally
greater reduction in the phenolic hydroxyl content. Another
possible explanation for this change is that a phenolic
hydroxyl group was replaced on the benzene ring by a
methoxyl group.

There was a difference between legume lignin pre-
parations and grass lignin preparations and this difference
was in the chemical make up of the lignin molecules. The
grass lignin molecule had a greater proportion of aromatic
compounds with a greater specific absorbtivity than the
aromatic compounds which make up the legume lignin molecule.
These aromatic compounds have more methoxyl groups and/or
less phenolic hydroxyl groups attached to the benzene ring
and can account for the difference in specific absorbtivity.

The phenolic hydroxyl and methoxyl content tended to

increase with maturity of the alfalfa hay. This is

86

 

87

consistent with the data reported by Phillips _t gt. (1939).

An alfalfa breeder line with high in vitro and in vivo dry
matter digestibility had a lower methoxyl content than did an
alfalfa breeder line with low in vitro and in vivo dry matter
digestibility. This indicates that forage having a lignin
molecule that is high in methoxyl content will have low dry
matter digestibility.

Alfalfa hay lignin contains a greater proportion of
aromatic compounds that contain more phenolic hydroxyl groups
than do the aromatic compounds that make up trefoil lignin.

The lignin from Siberian reed canary grass hay contained
more methoxyl groups than did the reed canary grass hay
lignin. This might be related to the observation that this
sample of reed canary grass hay was more digestible than the
Siberian reed canary grass hay.

Other comparisons were made among the forages studied
but there were no differences found in the structure of
lignin in these comparisons.

Lignin content was determined by a procedure that was
based on the amount of guaiacol-like compounds present in
lignin. These lignin values were compared to values obtained
from a standard laboratory method and found that the content
of guaiacol-like compounds cannot be used to determine the
amount of lignin present in the forage.

Prediction equations for estimating sheep digestibility

and digestible dry matter intake/cwt were develOped for

 

88

grasses and legumes separately and combined. The prediction
equations for the combined legumes and grasses had low
multiple correlation coefficients and complicated mathematics
that would limit the usefulness of these equations. The
prediction equations for grasses and legumes when taken
separately have squared multiple correlation coefficients
high enough to be of practical importance in estimating the
digestible dry matter and digestible dry matter intake/cwt

in sheep (.90 and .99 respectively for grasses and .93 and

 

”magnum-5A.; "1' '«l ‘-
. a

.96 respectively for legumes). The prediction equation for
digestible dry matter in grasses was 9 - 15.60 + 1.83X6 +
26'09X14 - 2.74X3. The independent variables used were mil
shift, ratio O.D., and phenol respectively. The miishift
variable (X6) was a measurement of the distance in mu between
the peak on the neutral spectrum and the peak on the alkaline
spectrum. The phenol independent variable (X3) was a
quantitative color reaction for phenol—like compounds. The
ratio O.D. (X14) was the ratio of the optical density of

the maximum peak to the Optical density of the minor peak on
the difference spectrum. This is an example of the type of

prediction equations that are presented in the text.

BIBLIOGRAPHY

Adler, E., K. J. Bjorkquist and S. Haggroth. 1948. Uber
die Ursache der Farbreaktionen des Holzes. Acta Chem.
Scand. 2:93.

Allison, D. W. 1966. Factors Affecting the Nutritive Value
of Forages. Ph.D. Thesis, Department of Crop Science,
Michigan State University.

Aulin-Erdtman, G. 1949. Ultraviolet SpectrOSCOpy of Lignin
and Lignin Derivatives. TAPPI 32:160.

Bechtel, H. E., A. 0. Shaw, and F. W. Atkeson. 1945. Brown
Alfalfa Hay. Its Chemical Composition and Nutritive
Value in Dairy Cattle Rations. J. Dairy Sci. 28:35.

Bolker, H. I. 1963. A Lignin Carbohydrate Bond as Revealed
by Infra-red Spectrosc0py. Nature 197:489.

Bolker, H. I. and N. G. Somerville. 1962. Ultraviolet
Spectroscopic Studies of Lignins in the Solid State.
1. Isolated Lignin Preparations. TAPPI 45:826.

Bondi, A. H. and H. Meyer. 1943. On the Chemical Nature and
Digestibility of Roughage Carbohydrates. J.
Agricultural Sci. 33:123.

Brauns, F. E. 1949. The Physical Properties of Lignin.
TAPPI 32:157.

Brauns, F. E. 1952. The Chemistty of Lignin. Academic Press,
New York, New York.

 

Brauns, F. E. and D. A. Brauns. The Chemistry of Lignin.
Supplemental Volume. Academic Press, New York, New York.

 

Brauns, F. E. and H. Seiler. 1952. Reactions with Homogenized
Spruce Wood. TAPPI 35:67.

Brewster, R. Q. and W. E. McEwen. 1961, Chemical Properties
of Phenol. Organic Chemistry. Prentice-Hall, Inc.,
Englewood Cliffs, N. J., p. 602.

 

89

 

90

Brown, Stewart A. 1961. Chemistry of Lignin. Science
134:305.

Brown, W., S. I. Falkehag, and E. B. Cowling. 1967. Molecular
Size Distribution of Lignin Wood. Nature 214:410.

Colburn, M. W., J. L. Evans, and C. H. Ramage. 1968. Apparent
and True Digestibility of Forage Nutrients by Ruminant
Animals. J. Dairy Sci. 51:1450.

Crampton, E. W. and L. H. Maynard. 1938. The Relation of
Cellulose and Lignin to the Nutritive Value of Animal
Feeds. J. Nutrition 15:383.

Creighton, R. H. J., J. L. McCarthy and H. Hilbert. 1941.
Aromatic Aldehydes from Spruce and Maple Woods. J.
Am. Chem. Soc. 63:312.

Czerkawski, J. W. 1967. The Determination of Lignin.
British Journal of Nutrition 21:325.

Davis, R. E., C. 0. Miller and I. L. Lindahl. 1947.
Apparent Digestibility by Sheep of Lignin in Pea and
Lima Bean Vines. J. of Agric. Res. 47:285.

Dehority, B. A., R. R. Johnson and H. R. Conrad. 1962.
Digestibility of Forage Hemicellulose and Pectin by
Rumen Bacteria In Vitro and the Effect of
Lignification Thereon. J. Dairy Sci. 45:508.

Ellis, G. H., G. Matrone and L. A. Maynard. 1946. A 72%
Sulfuric Acid Method for the Determination of Lignin
and Its Use in Animal Nutrition Studies. J. Animal
Sci. 5:285.

Ely, R. E., E. A. Kane, W. C. Jacobson and L. A. Moore.
1953. Studies on the Composition of Lignin Isolated
from Orchardgrass Hay Cut at Four Stages of Maturity
and from Corresponding Feces. J. Dairy Sci. 36:346.

Forbes, R. M. and W. P. Garrigus. 1948. Application of a
Lignin Ratio Technique to the Determination of the
Nutritive Intake of Grazing Animals. J. Animal Sci.

7:373.

Gee, M. S., O. E. Nelson, and J. Kuc. 1968. Abnormal Lignins
Produced by the Brown-Midrib Mutants of Maize. II.
Comparative Studies on Normal and Brown-Midrib -l
Dimethylformamide Lignins. Archives of Biochemistry
and Biophysics 123:403.

 

91

Gierer, J. 1954. Die Reaktion van Chinonmonochlorimid mit
Lignin. Acta Chemica Scandinavica 8:1319.

Goering, H. K. and P. J. Van Soest. 1968. tg Vitro
Digestibility of Lignified Materials Ensiled with
Sodium Chlorite. J. Dairy Sci. 51:974.

Goldschmid, O. 1954. Determination of Phenolic Hydroxyl
Content of Lignin Preparations by Ultraviolet Spectro-
photometry. Analytical Chemistry. Vol. 26, pp. 1421.

Gould, R. F. 1966. Lignin Structure and Reactions. American.
Chemical Society, Washington, D.C.

 

Hachihama, Y. and S. Jyodai. 1946. The Chemistry of Lignin.
Nippon Hyoronsha, Tokyo, Japan.

 

Hale, E. E., C. W. Duncan and C. F. Huffman. 1947. Rumen
Digestion Studies. II. Studies in the Chemistry of
Rumen Digestion. J. Nutrition 34:747.

Harborne, J. B. 1964. Biochemistry of Phenolic Compounds.
Academic Press, New York, New York.

 

Harris, E. E., J. D'Ianni, and H. Adkins. 1938. Reaction
of Hardwood Lignin with Hydrogen. J. Am. Chem. Soc.
60:1467.

Kane, E. A., W. C. Jacobson, and L. A. Moore. 1950. A
Comparison of Techniques Used in Digestibility Studies
with Dairy Cattle. J. Nutrition 41:583.

Kato, K., J. Nakano, and N. Migita. 1967. Studies on
Tobacco Lignin. Part I. Some Characteristics of
Tobacco Milled Wood Lignin. Agricultural and Biological
Chemistry 31:1010.

Kefford, N. P. 1958. Lignification of Plants in Relation
to Ruminant Nutrition. Journal of the Australian
Institute of Agricultural Science 24:297.

Kuc, J. and O. E. Nelson. 1964. The Abnormal Lignins
Produced by the Brown-Midrib Mutants of Maize. Archives
of Biochemistry and Biophysics 105:103.

Migita, N., J. Nakano, T. Okada, J. Ohashi, and M. Noguchi.
1955. Lignin XXIII. Cross-Bevan Color Reaction of
Hardwood Lignin. J. Japan Forestry Soc. 37:26.

92

Miller, L. G., D. C. Clanton, L. F. Nelson, and O. E. Hoehne.
1967. Nutritive Value of Hay Baled at Various Moisture
Contents. J. Animal Sci. 26:1369.

Patterson, R. F. and H. Hibbert. 1943. Studies on Lignin
and Related Compounds LXXIII. The Ultraviolet Absorption
Spectra of Ethanol Lignins. J. Am. Chem. Soc. 65:1869.

Patton, A. R. 1943. Seasonal Changes in the Lignin and
Cellulose Content of Some Montana Grasses. J. Animal
Sci. 2:59.

Patton, A. R. and L. Gieseker. 1942. Seasonal Changes in the
Lignin and Cellulose Content of Some Montana Grasses.
J. Animal Sci. 1:22.

Pearl, 1. A. 1942. Vanillin from Lignin Materials. J. Am.
Chem. Soc. 64:1429.

Pearl, 1. A. 1964. Century-Old Puzzle. Chemistry and
Engineering News, p. 81.

Pearl, 1. A. 1967. The Chemistry of Lignin. Marcel Dekker,
Inc., New York.

 

Phillips, M., M. J. Goss, B. L. Davis, and H. Stevens. 1939.
Composition of the Various Parts of the Oat Plant at
Successive Stages of Growth with Special Reference to
the Formation of Lignin. Journal of Agricultural
Research 59:319.

Pigden, W. J. and D. H. Heinrich. 1957. The Lignin Content
of Six Clonal Line of Intermediate Wheat Grass.
Canadian Journal of Plant Science 37:128.

Schubert, W. J. 1965. Lignin Biochemistry. Academic Press,
New York.

 

Sell, H. M., C. W. Duncan, and L. D. Brown. 1961. Factors
Affecting the Utilization of Feed by Ruminants. Annual
Report of Cooperative Regional Project NC-25.

Siegel, S. M. 1956. The Biosynthesis of Lignin: Evidence
for the Participation of Celluloses as Sites for
Oxidative Polymerization of Eugenol. J. Am. Chem.
Soc. 78:1753.

Spearin, W. E. and R. C. Dresler. 1954. Stain Technique for
Differentiating Hardwood and Softwood Mechanical Pulps.
TAPPI 37:39.

93

Stafford, H. A. 1960. Differences between Lignin-Like
Polymers Formed by Peroxidation of Eugenol and Ferulic
Acid in Leaf Sections of Phleum. Plant Physiology 35:108.

Stafford, H. A. 1962. Histochemical and Biochemical Dif-
ferences Between Lignin-Like Materials in Phleum
pratense L. Plant Physiology 37:643.

Stallcup, 0. T., J. L. Cason, and B. J. Walker. 1955.
Influence of the Lignin Content of Hay on the Passage
of Nutrients Through the Rumen. Arkansas Experiment
Station Bulletin #572, pp. 3—16.

Swain, T. and W. E. Hillis. 1959. The Phenolic Constituents
of Prunus Domestica. I. The Quantitative Analysis of
Phenolic Constituents. J. of the Science of Food and
Agriculture 10:63.

 

Thompson, S. O. 1967. Ph.D. Thesis, University of Wisconsin,
Madison, 1965, as reviewed by I. A. Pearl in The Chemistry

 

 

of Lignin. Marcel Dekker, Inc., New York.

 

Tomlin, D. C., R. R. Johnson, and B. A. Dehority. 1965.
Relationship of Lignification to In Vitro Cellulose
Digestibility of Grasses and Legumes. J. Animal Sci.
24:161.

Van Soest, P. J. 1963. Use of Detergents in the Analysis
of Fibrous Feeds. II. A Rapid Method for the
Determination of Fiber and Lignin. J. of The
Association of Official Agricultural Chemists 46:829.

Van Soest, P. J. 1964. Symposium on Nutrition and Forage
and Pastures: New Chemical Procedures for Evaluating
Forages. J. Animal Sci. 23:838.

Van Soest, P. J. 1965. Symposium on Factors Influencing
the Voluntary Intake of Herbage by Ruminants: Voluntary
Intake in Relation to Chemical Composition and Di-
gestibility. J. Animal Sci. 24:834.

Van Soest, P. J. 1965. Use of Detergents in Analysis of
Fibrous Feeds. III. Study of Effects of Heating and
Drying on Yields of Fiber and Lignin in Forages. J.
0% tge Association of Official Agricultural Chemists
4 :7 5.

Van Soest, P. J. 1967. Development of a Comprehensive System
of Feed Analysis and Its Application to Forages. J.
Animal Sci. 26:119.

94

Van Soest, P. J. and W. C. Marcus. 1964. A Method for the

Determination of Cell Wall Constituents in Forages Using
Detergent and the Relationship Between this Fraction

and Voluntary Intake and Digestibility. Paper 81
presented at the annual meeting of the American Dairy
Science Association, Tucson, Arizona.

Van Soest, P. J. and R. H. Wine. 1965. Development and Use

of Chemical Methods for Determining the Nutritive Value
of Forages. NE-24 Annual Report.

Wexler, A. S. 1964. Characterization of Lingosulfonates by

U.V. Spectrometry. (Direct and Diff. Spectrograms).
Analytical Chemistry 36:213.

.u. EEE‘Lrhb/UEEEEE/CEE C Errkrt/rk/CCL

95

Appendix Table I. Lignin concentration, % digestible dry
matter, maximum and digestible dry matter
intake/cwt for the forages studied.

 

Max. intake

 

% DDM cwt DDM/cwt

Legumes Lignin % lbs. lbs.
6401-01 6.31 63.15 4.18 2.44
6401—02 9.54 54.68 3.16 1.66
6502-01 8.03 62.28 3.69 1.86
6502-02 10.81 55.70 3.49 1.45
6701-21 6.33 62.54 2.10 1.30
6701—22 6.78 63.70 2.38 1.51
6701-01 8.96 57.91 2.41 1.40
6201-11 8.73 61.71 3.20 1.98
6201-12 8.50 62.58 3.34 2.09
6702-OV 5.00 63.41 2.86 1.68
6702-OD 6.11 66.33 2.15 1.42
6703—02 8.08 61.16 3.12 1.66
6703—03 7.89 53.78 2.58 1.15
6704-01 6.93 63.39 2.78 1.76
Grasses

6401-51 2.85 63.86 3.36 1.85
6401-52 3.86 61.65 3.61 2.08
6401—41 3.12 63.17 3.23 1.71
6401-42 2.88 61.02 3.08 1.84.
6401—31 2.74 60.95 2.35 1.22-
6401-32 4.14 53.62 2.82. 1.27