b
o’
J

HIHH‘I

1
1
I

HM!

Hi!

I
I

—
__—_d

137
112
THS

 

A COMPARATWE STUDY OF
'viSTHO-DS FOR THE QUANTITATIVE
RST’EMATION OF CELLLLOSE,
iflifiv‘li‘fiLL CLUSE AN :3? LiG-NZN

Tin-.5115 for: the Dtfil‘t"? .1 M. S-
‘5‘»...‘1 “flitflJA .N SirAzt‘J, C814 LECE

"9 g ’ 3
- all g.«.. -.
‘fl‘L‘NJ‘mu ll? .. 3t .DVJL}.\1‘~;

f‘. «‘3
:7“.{5

     

{P11}

.1.

5211.

WYlihf‘ "F:

~11 "
0- .

bay1l0M 0 Oh

..
1
1..

n
\o

DRE

, ..
1:1“!

I
‘.

 

IV

3
1.19 .31... 11.11.11. .

«may. .. -

.0113er a no. a? 5: .5-

 

U

 

..u........ 1 .. . .. 3.1. Lu.‘...#\1.n1.lt.1.11\.Q.... an“. .1.. . “an- ..¥.4q.aw.1..1u..\a 101:...»

Q. .F...\1b.‘i.n

‘1'. VI. . . .'

$

.I'.

v.\.

F

9

1...»...
.1L1V1... ‘.

nfi.
éhhWs

u- 1...-

#1.

ti» - .7. . . ...
“1.1.1.1....
. . 11.....11

A 00.913.qu mm or METHODS FOR
m qmnmrm nsmmmon
or CELLULOSE, mxmumsn
my new:

by
Benjamin Hartley Pringle

.A THESIS

submitted to the Graauate school of Michigan
State College of Agriculture and Applied
Science 1n.part1al fulfilment of the
requirements for the degree of

MASTER OF SCIENCE

Department of Chemistry
19%

 

3‘3?

\59 c

Acknowledgement

The author wishes to express his appreciation to
gr. O..L. Hoppert, for his enlightening under—
standing of the difficulties encountered in vorkb
ing with biological material, and for his interest
and kindly criticism throughout these studies and
in the preparation of this manuscript.

huh
.3
‘2
in
LA
~13

TABLE OF CONTLNTS

I Introduction

II Historical

III

IV.

A. Analysis of feeding stuffs
B. Cellulose

C. Hemicellulose

D. Lignin

E. Crude Fiber

F. Nitrogen-free extract
Description

A. Cellulose

B. Hemicellulose

C. Lignin

D. Nitrogen-free extract
E. Crude Fiber

Methods.1vailable

A. Cellulose

B. Hemicellulose

C. Lignin
Experimental

A. ScOpe of problem

1. Comparison of the available methods for the
determination of cellulose, hemicellulose, and

lignin

2. Relationship between the nitrogen content and
lignin value of the various residues

3. Effect of particle size and heat treatment on

the crude fiber value

10 -
12

12

13 -
lo -
18 -

20 —

25

2o -

27

27

27

10

12

15
18
20
22

23

25

2b

B.

C.

D.

F.

G.

Tana of Contents (Con't)

Samples
1. Composition 27 — 28
2. Treatment 28 — 29
a. drying
b. grinding

c. extraction

Nethods of Olmstead and Williams

1. Work leading up to the methods 29 - 31
2. Description of the methods 51 - 53
3. Data 33

Methods of Crampton and Maynard

1. Work leading up to the methods 55 - 39
2. Description of the methods 33 - 37
3. Data 37

Methods of Davis and Miller

1. Work leading up to the methods 37 - 39
2. Description of the methods 39 - 40
3. Data 40

Work on nitrogen-lignin value
1. Description of work 40 - 41
2. Data 41

Work.on Barticle size ani heat treatment on
the Crude Fiber value

1. Description of work 41 - 42

2. Data 42

Table of Contents (Con't)

VI Discussion

VII
VIII

IX

A. Cellulose determinations

B. Hemicellulose determinations

C. Lignin determinations

D. Nitrogen—Lignen values

E. Barticle size - heat treatment results
Conclusions

Bibliography

Tables

A..Tab1es on methods of Olmstead and Williams

B.

C.

D.

F.

G.

H.

Tables on kethods of Cranpton and haynard
Tables on methods of Davis and Miller

Table on Averages of all Cellulose
Determinations

Table on Averages of all Hemicellulose
Determinations

Table on Averages of all Lignin
Determinations

Tables on Nitrogen-Lignin Values

Table on Particle Size and Heat Treatment

45 - 44
44 - 44a
44a - 48
48 - 49
49
no - ol
1 - 5
Appendix
Tables I-V
Tables VI-K

Tables XI-XV

Table XVI

Table XVII

T3016 XVIII
Tables XIXRXXIII

Table XXIV

amusement

At present, the attention of research both in the purely theoretical
and also in the practical field seems to be turning more to the importance
of proper nutrition. In this connection we find research in the use of
different products or lay-products in the development of different feeds
either from an economic standpoint or because some substance essential to
a properly balanced ration may be gotten from certain of these products
or by—products. Much research is being done on sources of protein, essent-
ial elements, vitamins, fats and the carbohydrate complex. A large amount
of interest lies in the availability of these various feed constituents es
the animal.

Since the protein, fat, and less complex carbohydrates are quite
readily utilized by all animals, the availability here is not a question,
unless they are tied up in some manner with the less easily digested
portion of the feed. This less digestible complex usually consists of a
large portion of the so-called roughage. In herbivarous animals, adapt-
ations exist for utilization of feeds very high in the less digestible
complex. ls have such animals as the cow, deer, sheep, goat, camel and
others with the rumen as the first organ of the digestive system, while
in the horse we find the caecum and colon greatly enlarged. In many animals
including humans, there is no adaptation provided for utilization of these
difficultly digestible constituents.

Much work has been done towards finding a suitable means of determin-
ing the digestibility of feeds, but due to the unknown chemical structure
of these complex substances making up the less digestible portion, the
field of analysis seems to be in a constant state of change. Feeds have

been analyzed in a good many ways, as will be shown later in this paper.

2.
lFeeding stuffs have been analysed on the basis of protein, ash, crude fat
and crude fiber. .L combination of the nitrogenpfree extract and crude
fiber has also been used as a scheme of analysis. More recently, the trend
has been to divide the less digestible complex into substances vhich could
be more or less defined by chemical structure. This appears to give a.mmch
better basis for analysis than did former methods of dividing the complex
into indefinite and varying chemical fractions. The complex has been dividp
ed more recently into cellulose, hemicellulose, and lignin. The analysis
of these compounds in the feeds has formed the more recent work on methods
of feed analysis. There have been many methods brought forth for the anal-
ysis of cellulose, hemicellulose and lignin, but the purpose of this thesis
was to compare results by the three more prominent methods; namely, those of
Olmstsad and Iilliams (l), Crampton.and.laynard (23, and Davis and.liller
(3). In order to produce comparative results, the same samples were anal-
ysed by all three methods. This thesis was the result of a great deal of
interest concerning the feasibility of using one or a combination of these
methods in determining digestibility coefficients in some nutritional

investigations.

3.
HISTORICAL

We
The history of food stuff analysis is rather old and quite stagnant.

Generally, the methods of feed analysis may be divided into two groups:
namely, chemical and enzymatic or combinations of both.

The first significant method was the old hay-equivalents brought out
by Thaer in 1809. He estimated that ninety 1301111118. of dry hay or clover,
or vetch, or alfalfa, or sainfoin, or 200 pounds of potatoes or 266 pounds
of carrots, or 350 pounds of rutabegas with tops, or 1460 pounds of beets
with tops, or 525 pounds of radishes, or 600 pounds of white cabbage were
equal to 100 pounds of meadow hay in feeding value. The failure here to
consider the protein fraction led to inconsistent results. In 1831, Davy
suggested evaluating feeding stuffs by the amount of extract that was re-
moved by digestion with hot water. In 18%, Bouissinganlt preposed the
evaluation of feed stuffs upon the basis of their nitrogen content as
compared with ordinary hay of 1.34 per cent nitrogen.

The analysis of feeds proceeded more or less along the old hay-
equivalent methods until Hennenberg broke away from it entirely. Hennen-
berg divided his analyses into five parts: fats, proteins, carbohydrates,
ash, and moisture. The sum of these five determinations subtracted from
one hundred gave Hennenberg his ni trogen-free extract. Even at this early
date, Hennenberg saw possible discrepencies in his crude fiber determina-
tion.

In 1861+ Hennenberg and Stohmann proposed a weak acid and base digestion
which we know today as the [sands crude fiber method. This method has re-
mained essentially the same and has been adopted by the Association of
Official Agricultural Chemists with only slight modifications. Bedwell (h)

V and Levellin (5) have developed methods with only slight modifications of

Hennenberg's original one.

i Sherman (6) in 1896 made probably the earliest attempt in the United
States toward modifying Hennenberg's old scheme of analysis. He divided
his method into five parts. The soluble carbohydrates were removed by
heating with two and one-half per cent hydrochloric acid for two hours.

He next treated the residue with malt extract and treated it in the same
way as the first filtrate, to yield starch. The residue was then treated
with one to two and one-half per cent sulfuric acid and alcohol, made up

to two per cent sulfuric and boiled for six hours, after which reducing
power was determined to get the free pentosans. Ash and proteid was then
determined on the residue which had been previously weighed. Another sample
of residue was treated with one per cent sodium sulfite and heated. This
residue was weighed and corrected for proteid and ash and the loss was
equivalent to the lignin and allied substances. The residue from the above
minus proteid, if present, was considered to be cellulose.

In 1899, Brown and Beistle (7) modified Sherman's scheme during their
work on dried brewers' grains. Their scheme consisted of crude fat, mois-
ture, sugar, dextrin, starch, lignic acids, lignin, cellulose, pentosans,
protein, and ash. These gave a total of 96.01 per cent out of a‘ possible
one hundred, and it is one of the best results ever obtained with this type
of analysis.

In 190%, Simon and Lobrish (8) developed a new method for feed
analysis, namely, the use of strong potassium hydroxide and hydrogen perox-
ide. In the same year Konig (9) proposed the use of glycerol sulfuric acid.

In about 1910 Tollens and his pupils introduced a method for the
estimation of pentosans, as calculated from the yield of furfural.

Steigler in 1913 developed a method in feed stuffs analysis involving

the use of ten per cent hydrochloric acid and a stream of air. Some years

- c».

5.
later, (1930) Iellenberg used nitric and acetic acids in feed stuffs
analysis, which today has become (with modifications) the generally used
method for cellulose. In 1931 Scharr and Kurschner modified this somewhat
by using acetic acid, nitric acid, and trichloracetic. In this same year
Remy (10) introduced probably the first enzymatic method, in which he used
enzymes to isolate cellulose, lignin, and hemicellulose from the starch,
protein and fat. He compared his values with the crude fiber values, and
found a great loss of the indigestible residue. Iilliams and Olmstead (1)
in 1935 proposed the use of pancreatin digestion preliminary to their
determination of hemicellulose, cellulose, and lignin. In 1936 Horwitt,
Cogwill, and Mendel (11) first introduced the use of pepsin, diastase,
and trypsin digestion. In the same year Norman (12) extracted and hydrol-
ysed the sample before analysing the lignin. The method of Grampton and
Maynard (2) in. 1938 involved the use of pepsin before the determination of
lignin. Davis and Miller (3) in 1939 preposed a method which mds use of
pepsin, clarase, and trypsin digestions before the lignin was determined.

dangling

In 1857 Schulxe proposed the first method for the determination of
cellulose. His method depended on the removal of lignin and the other non;
cellulose carbohydrates by means of a mixture of nitric acid and potassium
chlorate. this method was modified by Hennenberg in 1868 and by Hoffmeister
in 1888. Many other modified methods were presented between the years of
1890 and 1910 by such chemists as Lange (1895), meson (1903), Muller
(1911). Ioning (1913).

In 1897 Buhler secured a phenolic lignification patent which was later
used by Kalb and Schoeller as a method for the determination of cellulose. I.

In about 1906 Gross and Bevan deveIOped a method which is commonly used to-

day, The method made use of chlorine. This method was modified by Dean

6.
i and Tower (13) in 1907, Schorger (it) in 1917, Bitter (15) in 1921;, and by
Jenkins (16) in 1930. The modifications have consisted mainly in the man-
ner in which chlorine is added to the determination. Opfermn (17) in
1921 proposed a method for the determination of alpha, beta, and gamma cel-
lulose. The insoluble portion in seventeen and one-half per cent sodium
hydroxide is filtered and washed and weighed as alpha cellulose, The alb—
line filtrate is acidified with acetic acid, precipitating out the beta
cellulose, which is dried and weighed. The gamma cellulose is obtained by
the difference between the total and the sum of the alpha and beta forms.
In 1923 Bray and Andrews (18) used the same method except that they dissolved
the alpha cellulose in seventy-two per cent sulfuric acid and titrated with
potassium dichromate. The alkaline filtrate was divided into two portions,
one of which was titrated with potassium dichromate, giving the beta and
gamma cellulose. The second portion was acidified with ten per cent sul-
furic to precipitate the beta cellulose, and the filtrate was then titrated
with dichromate giving the gamma cellulose.

Kalb and Schoeller (19) in the same year brought out their method based
on Duhlers phenolic delignification, in which the sample was treated with
dry phenol and hydrochloric acid for several hours. In 1923 Sieber (20)
modified the Gross and Bevans method by more carefully detecting the end
point, thus preventing the cellulose from being attacked. In 1926 Iohmeta
and Bakaguchi (21) modified the Gross and Bevans method by extracting with
two and one-half per cent potassium hydroxide previous to the determination.
laksman and Tommy (22) in 1927 presented a scheme of analyses for feed
stuffs in which the cellulose was precipitated from the acid filtrate with
alcohol. In the same year Kiesel and Seneganovski (23) proposed a method
in which eighty per cent sulfuric acid was used to dissolve the cellulose.

This mixture was diluted with water and heated for five hours. The

7.

glucose present was then determined by one of the conventional methods. m A
Kurschner and Hoffer (2)4) in 1931 proposed a new method for cellulose
which consisted of heating the sample with nitric acid and ninety-six per
cent ethanol. Schmidt, man-Chi Tang and Jandelbauer (25) in the same year
introduced a method which used chlorine dioxide and pyridine. This treat-
ment was mmposed to remove lignin and everything else except the cellulose.
Phillips (26) in 1932 introduced a new method using chlorine gas and sodium
sulfite to remove the lignin. The delignified product was bleached with
one-tenth per cent potassium dichromate and rendered colorless with sulphur-
our acid. It was then washed with amonia and water and weighed as cellulose.
A novel method based on an idea that all cellulose, regardless of source,
is equivalent colorimetrically was introduced by Paloheimo and Volavarra
(27) in 1933. The sample is treated with alkali to remove all non-cellulose
substances which are soluble in seventy per cent sulfuric. The residue is
now taken upin seventy per cent sulfuric which is diluted to fifty per cent
sulfuric and then a solution of potassium iodide-iodine, which gives an inn-
tense red color, is added. Olmstead and Iilliams in 1935 (1) introduced a
method for cellulose in which the sample was first digested in a buffer-bile
and pancreatin solution. The enzyme residue was treated with twenty-one and
four-tenths normal sulfuric acid, which is diluted. The mixture is hydrolysed
and the reducing sugars (both total and. non—fermentable) are determined from
which the cellulose is calculated. Acharya (28) in 1936 preposed a method
for cellulose involving the use of hypochlorous acid and sodium sulfite in
removing lignin, etc. The carbon and furfuraldehyde content is then deter-
mined in the residue from which the cellulose and qlan content are calcul-
ated. In the same year Ziska and Kohler (29) introduced the method which is
used as such or in a modified form in many cellulose methods today. The

method involved the prolonged heating of the sample with a mixture of one

hundred parts of ethyl alcohol to twenty of nitric acid ( d 1.29).

In 1937 Launer (30) introduced a simple volumetric procedure for the
determination of alpha, beta and gamma cellulose by oxidizing with potas-
sium dichomate. (A micro method was introduced by Strepkov (31) at about
the same time. The sample was treated with seventyatwo per cent sulfuric.
The solution was then neutralized and the glucose determined by the potas—
sium ferrocyanide method. In 1938 Ubaldini (32) introduced a method.using
ten parts of eighty per cent acetic to one part of nitric acid. This oxide
izes the xyloid and huminic lignites. The cellulose could be determined by
treating the residue with eighty per cent sulfuric at room temperature,
and then analysing for glucose. In 1938 Crampton and Maynard (2) introduc-
ed a method involving a slight modification of Kurschner and Hanak‘s (33)
method, using an acetic acid-nitric acid mixture. One year later, 1939,
Rise, Peterson and Harlow (3h) introduced the use of a mixture of ethanol-

amine, chlorine water and sodium sulfite for isolating the cellulose.

Eemigellulgge
.Although the literature contains many references to attempts at estim-

ating hemicellulose, there are practically no methods which are entirely
satisfactory. Much more attention has been given by research workers to
the separation into various fractions of the hemicellulose complex rather
than the quantitative estimation of the whole complex; Tollens suggested
probably the first method for the quantitative estimation of hemicellulose.
Although the method.was essentially an empirical one which requires adher-
ence to a strict set of conditions, it has been used up to the present day.
His method depends on the conversion of hemicellulose into furfural by
distilling with ten to twelve per cent hydrochloric acid and precipitating

it. The chances for error are obvious, because other plant constituents

9.
such as pectin and uronic acids also yield furfural. Krober in 1900 modif-
ied Tollens method slightly using phloroglucinol as the precipitating agent.
The method as outlined by Iorber is now used and called Tollens furfural
method. Much work has been done on the use of various precipitating agents
for furfural. Box and Plaisance (35) made an.extensive study on the use of
barbituric acids. 2,h - dinitrophenylhydrazine has also been used as a
precipitant. The serious objections to Tollens method or any of its modif—
ications seem to be that any hemicellulose under investigation may consist
of hexosans, pentosans, glucuronic acid, and galacturonic acid. The latter
three constituents of hemicellulose give different yields of furfural.
Hexosans, on the other hand, yield.practica11y no furfural when boiled with
ten to twelve per cent hydrochloric acid. Thusm the hemicellulose must be
rather simple in composition in order to be determined by the furfural meth-
od. Dore (36) in 1920 determined the loss of weight for woods when treated
with five per cent sodium hydroxide. Norman in 1929 described a method for
the determination of hemicellulosee in cereal straws, based on the furfur-
aldehyde method. Preece (37) in 1931 introduced a method in which the hemi-
cellulosee are actually isolated and weighed.‘ The method is rather tedious
and from all reports, apparently not too accurate. This was the first
attempt to isolate hemicellulosee directly. Many methods have been deveIOped
based on the principle that plant material freed from sugar and starch and
then hydrolyzed with dilute acid gives reducing sugars. The reducing sugars
in the hydrolysate are then determined by the usual methods. The result is
usually expressed in terms of glucose, although this sugar makes up only a
small part of the reducing compound. .A more serious objection is the fact
that the hydrolysate may contain hexoses, pentoses, uronic acids, and pos-
sibly other substances which.may reduce Fehlings solution. The reducing

value then is quite conflicting, a nd it is practically impossible to obtain

10.
a value which could be considered as representing hemicellulose. Olmstead
and.lilliams (1) have modified this method. They have determined the total
reducing sugars on the filtrate from their lignin determination and then
fermented out their hexoses and determined their non-fermentable reducing
sugars. The non-fermentable reducing sugars were interpreted as pentosans
and the difference which was composed of the fermentable sugars was used

for the calculation of cellulose.

ldznln

Numerous methods hare been described in the literature for the quant-
itative estimation of lignin. Much has been done in the field of wood.and
paper chemistry in separating lignin from its combination with other plant
constituents. Direct and indirect methods are employed, and the latter de-
pend upon the estimation of some characteristic group, such as methoxyl,
from which the per cent of lignin may be calculated. The indirect methods
have proven very unsatisfactory and have been generally discarded. In the
direct methods lignin is separated from the cellulose and other carbohydrates
associated with it and weighed.

The direct methods may be subdivided into the following classes: (1)
Those depending on dissolving all carbohydrate constituents and leaving
lignin as the residue, and (2) those that dissolve the lignin and thus sep-
arate it from the cellulose and other carbohydrate material present.
Treament with seventybtwo per cent sulfuric acid and fuming hydrochloric
may be considered as examples of the first class, whereas the method of
Mehta (38) may be considered.as an example of the second class, in which
the lignin is dissolved away from the rest of the sample.

In 1883 Flechsig (39) showed that seventy-two per cent sulfuric acid

will hydrolyze cellulose in the cold. In 1910 this fact led 0st and

ll.

Eilkening (M0) to a method for the quantitative estimation of lignin.
Their method has been modified by a great many people as to time, temperature
and strength of acid. In 1923 Klason (hi) modified the method by suggesting
the use of sixty-four or sixtybsix per cent sulfuric acid instead of seventy-
two per cent. laksman and dtevens (h2) have used eighty per cent sulfuric
for the isolation of lignin.

Fuming hydrochloric acid (d. 1.212 - 1.223 at 15°C.) was first used
by lillstdtter and Zechmeister in about 1913 (M3) for the isolation of lig-
nin. This method has been applied to the determination of lignin by many
workers in its original or somewhat modified form. In 191R Xonig and Rump
(MM) isolated lignin by heating the material with one per cent hydrochloric
acid under five atmospheres of pressure. lenzl (MS) in about 1923 used a
mdxture of phosphorous pentoxide and concentrated hydrochloric acid to re-
move polysaccharides, in determining lignin. The lignin isolated by the
sulfuric or fuming hydrochloric acid.methods contains some nitrogen and
inorganic matter. Correction should be made for these. laksman (h2) de-
termined the nitrogen in the crude lignin, and the weight thus obtained,
when multiplied by 6.25 is deducted from the weight of the crude lignin.
The assumption is that the nitrogen is in the form of protein, In 1922
Mahood and Cable (MG) presented what is known as the original seventyatwo
per cent sulfuric acid method, which is, perhaps, most generally used in
the estimation of lignin today. In the same year Phillips (h?) introduced
what is known as the original fuming hydrochloric acid method. This was
modified by Goes and Phillips in 1936 (#8). They compared both the orig-
inal and modified seventybtwo per cent sulfuric acid and fuming hydrochloric
acid methods and found that the modified fuming acid method probably gave
lignin results nearer the true lignin value. Olmstead and Williams (1)

in 1935, Davis and Miller in 1939 (3) and orampton and haynard in 1938 (2)

12.

, used slight modifications of the seventy-two per cent sulfuric acid method.

Crude Fipgr

The old Einhoff procedure originating in about 1860 and later used.by
Hennenberg and Stohmann in 186“ as part of their scheme of analysis of food
stuffs has never been changed as far as the method itself is concerned. The
method as used today is almost identical with the one first used by Hennen-
berg. In the literature only one reference to a change in the crude fiber
method was found, and that was by Ring (1:9) in which a round-bottom flask
was used under reflux instead of an Open flask. He avoided foaming by blow-
ing a stream of air through a constricted.gflass tube whose tip was about

one and a half or two centimeters above the level of the solution.

Nitro e -f t

The so-called nitrogen-free extract of foods and feeding stuffs has
been a subject of discussion among Agricultural Chemists since it was first
introduced by Hennenberg and Stohmann nearly eighty years ago. The term
represents something very complex and indefinite. Hennenberg and Stohman
secured its general adoption through their efforts to bring about improve-
ment of current methods of feed stuff evaluation at that time. Davy
attempted something similar at the beginning of the nineteenth century.
He extracted feed with hot water and supposed this extract to correspond
to what Hennenberg and Stohmann extracted with acid and alkali. Davy was
of the opinion.that the greater the amount extracted with hot water, the
more valuable were the feeding stuffs. The nitrogenpfree extract of
Hennenberg's was equal to the total dry organic materials minus the sum
of the crude fiber and the protein. There were many serious objections to
this procedure, so that little use is made of it today in the field of

nutrition.

13.
DESCRIPTION

Cellglggg

Cellulose is perhaps the most abundant organic compound occurring in
nature. Celluloses are the principal constituents of the cell walls of all
higher plants. .Although cellulose is most familiar to the average person
in the form of pure cotton or filter paper, this is about the only case in
which cellulose is found relatively pure. Cellulose is scarcely ever found
in a relatively pure condition, except innthe cases previously mentioned.
Bragg has said.that cellulose is preeminently the molecule of growth in the
vegetable world. The cellulose from the cotton boll on extraction with
alcohol and ether to remove fat, treatment with boiling dilute alkali and
washing gives a product which is ninetyanine and eight-tenths per cent
cellulose and five-hundreths per cent ash and is resistant to extraction
with seventeen and one-half per cent sodium hydroxide in the cold. This
product has been the starting point for all researches on the constitution
and.pr0perties of cellulose. However. in most cases in plants where cel-
lulose plays a structural role, the cellulose does not occur in a.pure or
easily purified state, but it is found in intimate association with other
cell wall constituents. Earlier it was believed that cellulose existed in
combination with cell wall substances such as lignin, pectin. cutin and
others to form lignocellulose, pectocellulose, and cutocellulose. This
theory was discarded, although some botany textbooks still list such com-
pounds. The chief point of support for the theory existed in the fact
they lignin could not be readily extracted from Jute fiber and thus it was
concluded that lignin must be combined chemically with cellulose. Bailey
and Nbrris (50) found the existence also of an amorphous cellulose in com-
bination with other polysaccharides in certain mucilages, but this appears

to be due to the physical proPerties of the polysaccharide. The modern

11+.
and generally accepted theory of steric structure of cellulose rules out the
possibilities of cellulose acting as a structural constituent in these cell
wall combinations. ‘Although the structural cellulose of the cell wall can-
not be combined in a true chemical sense with any other cell wall constituent,
its separation from them is exceedingly difficult. No single method will
accomplish this without attacking the cellulose also. This makes the deter-
mination of cellulose a difficult task.

Structural celluloses are largely composed of the same polysaccharide
as the cotton hair (alpha cellulose), although not necessarily in the same
state of molecular aggregation. The remainder is another polysaccharide or
mixture of polysaccharides firmly retained. The associated.polysaccharides
found in most structural cellulose is a xylan, although in the Gymnosperms
both mannan and xylem are found. .A means of describing this associated mat-
erial was suggested by Hawley and Norman (51) who proposed calling this
associated material 'cellulosan“.

Constitutionally, cellulose has generally been accepted as being com-
posed of a long chain (giving fiber strength) of l.h—beta glucose anhydride
formula. Some proofs of this structure are:

(1) Almost quantitative yield of glucose from cellulose on hydrolysis.

(2) Hydrolytic production of crystalline fragments of the cellulose

chain consisting of two, three and four glucose units.

(3) Formation of a triacetate.

(h) Formation of a well-defined crystallizable octacetate of cellobiose

by acetylation.

(5) Identification of beta glucosidic residues and a simple chain

structure by the study of rotations and kinetics of cleavage
products of cellulose hydrolyzed by strong acids at moderate temp-

eratures.

1'3.

Thus the cellulose molecule is considered to be a chain of considerable
length, the individual unit of which is anhydro beta glucose, the linkage
between the units being between the one and four carbon atoms, with the
terminal carbon lying alternately on one side of the chain and then the other.
There seems to be a great deal of discrepancy on chain length and molecular
weight. One group working on physical and chemical properties point to a
chain length of one-hundred fifty to two-hundred fifty units. .Another group
using viscosity measurements gives a chain length as seven to eight-hundred
units. Studies seem to indicate a maximum chain length of over twelve-
hundred units with a molecular weight (maximum) of over two-hundred thousand.
The micro structure of cellulose is in units called micelles, consisting of
about sixty cellulose chains or one-hundred to one-hundred twenty glucose units.

Cellulose is rather inert and has an exceptionally high tensile strength.
It is insoluble in water, and on treatment with cold seventeen and one-half
per cent sodium hydroxide, alpha cellulose is left as the residue. Cellulose
is soluble in seventy to eighty per cent sulfuric acid, and in fortybtwo per
cent hydrochloric acid. Certain inorganic salts such as zinc chloride and
hydrochloric acid will also dissolve cellulose. Cellulose swells in the up—
take of water, but it is not directional. Cellulose is decomposed by many
organisms, such as aerobic and anaerobic bacteria, mesophilic and thermo-
philic fungi, actinomyces and.even protozoa. The attack on cellulose is
thought to begin with an exo-enzyme, cellulase, which liberates the glucose
units from the cellulose chain with subsequent fermentation, the products

formed depending on the organism. .

W
If one considers the methods available for the determination of hemi-

cellulose, one is compelled to agree with Norman that “the chemistry of the

...-l

16.
hemicellulosee remains to be written.“ Their complex structure and.physica1
properties make examination very difficult. The generic term “hemicellulose“,
a very unfortunate one, originated in 1892 from the work of Schultze. These
substances are much more susceptible to acid hydrolysis than cellulose, but
were believed to be related in some way, probably as intermediate in its
formation, hence the name.

Various sugars (arabinose, xylose, and galactose) have been isolated from
the hydrolysate of hemicellulosee. The preparations usually contained more
than one sugar. Until the last decade, the hemicelluloses of Schultze were
believed to be true hexosans or pentosans or hexo-pentosans containing both
units. The presence of sugar acids of the uronic acid type has also been
demonstrated in hemicellulosee, and the true hexosan or pentosan, except
cellulosan, is believed to be very rare. Glucuronic and galacturonic acids
are commonly found in plant materials. Mannuronic acid has been isolated
in certain type of morine algae. The occurrence of mannuronic acid in plant
hemicelluloses is quite improbable, due to the hexose to pentose formation
through the formation of uronic acid being incomplete in the mannose series.
The pentose member of the mannose series would be d-lyxose which has never
been known to occur naturally. There are configurational groups of hemi-
celluloses: (l) the glucose series (dpglucose, dpglucuronic acid, and d-
zylose) and (2) the galactose series (degalactose, dpgalacturonic acid and
l-arabinose). Xylose usually predominates in the first group; galactose,
in the second. A,definition of hemicellulosee might be: those plant
constituents soluble in dilute alkali (cold four to ten.per cent macs),
much more easily hydrolyzed than cellulose, and appearing to be used.by
plants in growth as a reserve food. Herman (52) has given a concise def-

inition and differentiation:

17.

vCel oses

 

Not associated with Associated with
cellulosic fraction natural cellulose
Not containing Containing Not containing
uronic acid uronic acid uronic acid
Polypses Pol onides Cellulosans
l
Hexosans Encrusting Celluloic frame
Pentosans substances work substance
Hexapentosans Pentose and
uronic acid xylan (Mannan)
Reserves? Hexose and
uronic acid Glucosan?

Pentose and
hexose and
uronic acid

Much work has been done on the fractionation of hemicelluloses, but
due to imperfections of the extraction methods, the chemistry of the hemi-
celluloses has been complicated to a great extent.

Hemicelluloses as previously mentioned may be extracted from plant
tissues by alkali and are readily hydrolyzed by hot dilute mineral acids
to give reducing sugars and uronic acids. They are non-reducing and in
spite of the presence of uronic groups, they do not exhibit acid properties.
Many celluloses may be precipitated from alkiine solution by Fehlings sol-
ution, copper salts, lead acetate and other metallic salts. They may be
methylated and acetylated by the usual methods, although it is difficult
to carry these reactions to completion. With alkali and carbon disulfide,
a zanthate may be obtained. The majority of hemicelluloses give either
no color or a slightly greenish color with iodine. The hemicelluloses
are optically active and usually are more or less strongly lave-rotatory.

In young tissue cells only small amounts of hemicellulose are found,

combined.with about fifty per cent true cellulose and some pectic sub-

stances, but no lignin. In old tissue cells we find a great deal more

18.

hemicellulose, up to twenty per cent combined with about fifty per cent
true cellulosek up to twenty per cent lignin and practically no pectic
substances.

Hemicelluloses have an important influence on the biological decomp—
osition of the plant material in which it is found. They are readily fer-
mented.by mixed cultures of micro organisms and especially the fungi.
Evidence has been found by lenny and lhksman (53) that in aerobic rotting

of plants, the hemicelluloses are the first constituents removed.

lignin
Although lignin has been studied by chemists for about a centuryk

much of the chemistry of this substance is still imperfectly understood.
Much progress has been made in recent years in the studies concerning lig-
nin structure, and much still remains to be done. From time to time,
investigators have proposed probably structural formulae for lignin, but
most of them lack any definite chemical evidence to support them. This is
primarily due to the fact that a lignin isolation method has never been
deveIOped which would isolat e lignin in the pure state. Thus no means of
determining its purity is available. Lignin is an amorphous substance
ranging in color from light tan to a black, depending on the method used
for itsisolation. Evidently, lignin is so tied.up with the cell wall
encrustations that it is very difficult to separate in a very pure state.
The isolation methods employed have either not produced pure lignin or
they have altered the reactive groups present. The greater part of the
work on the structural formula, consequently, is of doubtful value. The
general assumption seems to be that lignin is homogeneous in nature, but
there is a large amount of evidence to show that this is not the case.

Lignin found in different species is not identical. Even the nature of

19.
lignin in one plant depends on the age and nature of the tissue.

Lignin may either be a complex misture of compounds with similar
properties but of unrelated chemical structure or may have the same basal
form varying in minor ways only, such as side-chains or substituted groups
present or in chain length or degree of polymerization. Due to the work
of Hilpert, a.third idea of structure may be added. This considers that
lignin has no real existence, but is merely an artifact or a reversion
product formed by the action of acid on certain of the methylated carbo-
.hydrates. The idea that lignin is a mixture of related compounds is most
generally accepted and has a certain amount of evidence in support of it.
Lignin is found chiefly in the middle lamella and.mainly in layers in the
secondary wall. The proportions of carbon, hydrogen, and oxygen found on
analysis of lignin are not those of a carbohydrate. The carbon content is
too high and the oxygen, too low. The methoxyl group is the only group of
the lignin molecule that is universally accepted as being present. Many
phenolic compounds have been isolated from lignin, supporting the exist-
ence of the aromatic ring in the structure of the lignin molecule. Klason
has attempted to show that lignin is a derivative of coniferyl alcohol by
polymerization and oxidation. Fuchs (5h) has suggested a very complex
aromatic ring structure with methoxyl groups substituted around the rings.
Freudenberg has suggested a quite simple lignin structure, somewhat like
the pattern of cellulose.

Lignin exhibits a notable resistance towards strong acids, while alk-
alies dissolve it more or less readily. Organic solvents have little
effect on it. Lignin is very easily oxidized by such agents as hypochlor-
ites, hydrogen peroxide, potassium permangante and ozone, giving lower
fatty acids and di-basic acids. Lignin may be very rapidly nitrated.with

nitric acid or nitrogen peroxide in the cold, but sulphonation requires

20.
more drastic conditions. In the presence of strong acids, lignin condenses
with aldehydes and ketones to give more or less insoluble products of the
phenol-aldehyde type. Lignin is unsaturated, as shown by its combination
with bromine. Lignin may be methylated fairly easily. It gives the Molisch
test, the phloroglucinol test, and the indol test. Considerable work has
been done in recent years on the microbial decomposition and digestibilities
of lignin in the animal body, but much of this work is very contradictory.
Many investigators claim that lignin is digested in part by certain animals,
while others deny this fact. Generally there is considerable evidence to
show that lignin in plant material is relatively resistant to microbial
attack, at least under anaerobic conditions, There is data indicating that
when alkali lignin is fed to cows and dogs, there is an increase in benzoic
acid (as hippuric acid) eliminated in the urine, the benzoic acid appar-
ently coming from the aromatic portion of the lignin. Csonka, Phillips
and Jones (55) have shown that the lignin was in part dimethoxylated in
these feeding experiments. Some of the higher fungi have been found to be
capable of using a part of the lignin from such sources as straw and timber.
Under anaerobic conditions, isolated acid lignin has been shown by Boruff
and Buswell (56) as probably having a bacteriostatic action, for on addit-
ion to an actively fermenting glucose substrate, gas production instantly
ceases and cannot be revived by further additions of glucose. ,Alkali
lignin similarly has a restrictive action. The only lignin derivative

examined, phenol-lignin, was found to be decomposed by a number of bacteria

and fungi.

r e - re tr
.As determined by the customary procedure for the analysis of feeding

stuffs, the nitrogen-free extract is the largest component of the rations

21.
of animals, representing forty to seventy per cent of the total dry matter.
The nitrogen-free extract serves as a source of energy for the body proc-
esses and for the deposition of fat. It is not the total amount or the
caloric content of the extract that makes it so important nutritionally,
but only that fraction of the gross energy which is available to the body.
This availability is controlled by digestibility and other metabolic fact-
ors, which are related to the chemical nature of the nitrogen-free extract.
Thus the nigrogen-free extract in order to be of real value has to be broks
en down into units which can give some answer as to the availability of the
feed. The term obviously does not represent a single constituent, but a
residiuum of numerous undetermined substances of variable nutritive value,
whose calculation by difference is multiplied by the errors in the methods
for determining fat, protein, ash, and especially crude fiber.

The nitrogen-free extract contains starch, hemicellulose, lignin, pect-
in, and various related substances. Starch as such is almost entirely
digested by the animal under the action of enzymes. On the other hand,
there are no enzymes secreted by animals which will digest cellulose, or
lignin, and this appears to hold true for most of the groups of compounds
making up the hemicellulose. Several investigators have failed to find a
pectin or an inulin digesting enzyme in animal tissues.

In assuming that the nitrogenafree extract comprises the more readily
digestible group of plant constituents, it was quite commonly believed that
during treatment with hot dilute acids and bases, only sugars, starches
and hemicelluloses were dissolved. However, it has been shown that var-
ious polysaccharides and polyuronides and varying amounts of lignin were
also extracted. Recent criticisms of the nitrogen-free extract have been
based on the fact that no consideration has been given to the lignin

fraction. No definite relationship between the hemicellulose and lignin

22.
; fractions in the nitrogen—free extract can be established, since their
distribution in plant materials varies with the plant and its degree of
maturity. The hemicellulose fraction of the extract consists not only of

the polymers of hexoses and pentoses, but also of mixtures of these.

ngdg Fibs:
That portion of feeding stuffs which is insoluble in hot one and

twenty-five hundreths per cent sulfuric acid and hot one and twentybfive
hundredths per cent sodium hydroxide after correction for ash is called
“crude fiber“. The original investigators knew that the crude fiber prod—
uct did not have a constant chemical composition as was shown by the varya
ing percentages of carbon, hydrogen, and oxygen. The indigestible residue,
(the portion of the feed not attacked.on passage through the digestive
system and consisting chiefly of cellulose, hemicellulose, and lignin)
contains according to many workers fifty percent more indigestible mater-
ial than did the crude fiber product. In the crude fiber product, pract-
ically all of the hemicellulose is lost and some of the lignin. Coleman
(57) and.nine colloborators analyzing bran found the crude fiber results
unsatisfactory. Remy (10) found in comparing his enzymatic method with
the crude fiber method that the latter caused a fifty per cent loss in

the indigestible residue.

This seems to indicate that the crude fiber product is defined by the
method used to isolate it. Hennenberg himself recognized, as previously
mentioned, that his product was quite variable, but he haped it would
serve a purpose in feed analysis methods for the time being, for he thought
that soon methods would be developed for determining the various constit-
‘uents of feeding stuffs. He used the term “crude fiber“ because he knew

ilt was of a variable nature. Ebw disappointed he would be today if he saw

23.
_that his hopes were yet far from realized.

It seems, therefore, at least from the standpoint of research, that
more specific methods for the determination of the complex carbohydrates
are needed. This is particularly true for the rations of herbivorous
animals. The major role which the higher carbohydrates play as energy
foods, especially for herbivore, and the fact that feeds measured by the
present methods which appear to have similar composition are found to be
quite different in feeding values emphasize the need for the deveIOpment

of more exact chemical and biOIOgical methods of analysis.

2h.
METHODS,AELILAELE

That the digestibility of the dietary carbohydrate does not follow
its partition into crude fiber and nitrogen-free extract with any marked
certainty, especially in the case of roughages, is evident from data col-
lected over a period of years. Published.literature contains ample evid-
ence that the crude fiber of forages may be as well digested as the protein,
as is shown by the work of Jones and Newlander, Mitchell and Hamilton, and
Morrison. It would seem, therefore, that if partition of the carbohydrate
fraction could be made into parts which were either biological or chemical
units, the usefulness of the feeding stuffs analysis in predicting probable
feeding value would be enhanced. Most workers on evolving their schemes of
analysis of the indigestible residue thought that a partition of the carbo-
hydrate fraction into useful biological or chemical units, would become
largely a consideration of the chief constituents of the cell wall carbo-

hydrates, lignin, hemicellulose, and cellulose.

mm
Two general groups of methods exist for the determination of cellulose:

(1) those depending on the removal of everything but cellulose in the plant
material, and (2) those depending on the hydrolysis of cellulose to gluc-
ose and its determination. Methods coming under the first class would.be
those of Cross and Bevan (chlorinating to remove all but the cellulose),
and Crampton and Maynard (removal of all encrustating substances by means
of nitric-acetic acid mixture). There are many methods which are on the
same idea as these examples mentioned, but these are the most general.

The method of Crampton and Maynard was used in the present work. Under the
second class any of the copper reduction methods may be used to determine

the glucose. The Shaffer-Somagyi method used by Olmstead and lilliams was

25.

also employed in these studies.

e i e e

There is no absolutely satisfactory method available at present for
the quantitative estimation of the hemicelluloses. The available methods
may be conveniently divided into three types. In the first type, the plant
material is freed from sugar and starch, hydrolyzed by dilute acid and the
reducing sugars determined by any of the available methods. The second
type of method depends on the measurement of furfural secured on boiling
the material with mineral acids. The third type depends upon the isolat-
ion of the hemicellulose and the weighing of it. ,All of the methods have
weaknesses. Olmstead and Williams modified the first method by determin-
ing the total sugars before and after fermentation with yeast. In this
manner, they secured both the cellulose and hemicellulose. This proced-
ure was used in the present studies.

Most investigators seem to stay away from determining hemicelluloses
because of their complex and variable make-up. They prefer to determine
the indigestible residue by an enzymatic method and the lignin and cellul-
ose by separate methods. By subtracting the sum of the cellulose and lig-
nin from the indigestible residue, they secure the hemicellulose values.
They have termed this difference, not hemicelluloses, but'bther carbohydrate
material“. Davis and Miller (3) have used this scheme, and it was also

'used in comparing it with other methods in these studies.

Mania
Numerous methods have been described.previously for the determination

of lignin. Generally they may be classified as direct and indirect. The
indirect methods have been discarded. The direct methods have consisted of

isolating lignin by removing other carbohydrates associated.with it. The

26.
methods available are divided into two general methods: (1) seventybtwo

per cent sulfuric acid, and (2) fuming hydrochloric acid method. These
both have been modified, so that we actually have four types available.
Olmstead and lilliams (1) used the sulfuric acid method but modified it

by using sixty per cent sulfuric acid instead of seventybtwo per cent.
Crampton and Maynard (1) used the seventyatwo per cent method, but mod-
ified it by treating with formaldehyde and.using a granulating mixture of
chloroform and acetic acid. Davis and Miller (3) used the seventy-two

per cent sulfuric acid but modified it by finally diluting thezicid
mixture out to three per cent sulfuric and.refluxing for two hours. .L
comparison of all three of these methods on the same materials was carried

out in the present investigation.

27.
EXPERIMENTAL

§902e of Prgblem

.L great deal of interest has been created in this department in some
nutritional work on the digestibility of certain foods. Much work has been
carried on and suitable methods for analysis were found to be quite essent-
ial. It was thought an interesting problem could be made in seeing how some
of these methods compared with each other on the same material. For a
scheme of analysis of the nutritional work, it was decided to divide it in-
to the newer units of cellulose, hemicellulose and lignin. The methods
chosen for comparison on these three constituents were those of Olmstead
and Williams (1) Crampton and Maynard (2) and Davis and Miller (3).

In working on the lignin determination, highly irregular results were
found, and it was thought that there might be a relationship between the
method of treating the sample before the lignin determination and the lignin
value. This idea was carried out in some work which will be described.later
in this paper.

In some nutritional work being carried on here, there arose the quest-
ion as to the effect of particle sizewand heat treatment on the analytical
results. Bran was chosen as the substance and crude fiber as the method,
and investigations were carried out, the description and summary of which

will follow later.

58.122119

l 3 Composifion

The samples used were fecal and food samples collected in the course
of a study of the coefficients of digestion of various foods, such as
apples, lettuce, cabbage, bran and others. The experiment consisted of

feeding a basal diet for one week, followed by the basal diet plus a

28.

~.supplement of some food, the digestibility of which was to be determined.
The subjects were fed weighed amounts of all constituents of the diet and
the samples used in this work were the fecal and food samples collected
during the experiment.

The basal diet was composed of the following foods: potatoes, meat,
cheese, milk, sugar, tea, bread, puddings (canstarch), jello, jam, crackers,
cookies, ice cream. The experiment lasted eight weeks, the plan being
given below.

First week -- basal
Second week -— basal and all bran
Third week -- basal
Fourth week - basal and prunes
m... C.T., L.O. (by three people)
basal and celery
B.P., 1.3., 5.0. (by three people)
Fifth week -- basal and all bran
Sixth week -- basal and lettuce
L.0.. H.P., 8.0. (by three people)
basal and cabbage
J.L.,.L.S., C.T. (by three people)
Seventh week - basal
Eighth week -- basal and oranges
J.L., E.P., 8.6. (by three peOple)
basal and.apples
C.T., L.O., A.S. (by three people)
2 - Treatment
The material was dried on steam baths in eight inch evaporating dishes

for about five to six days.

29.

The samples were ground in a Wiley mill and portions extracted with

ethyl ether for the various analyses.

gethods of Olmgtead and flilliamg
lggk leading up to the methodg
Methods of analysis of cellulose, hemicellulose, and lignin depend

first upon removal of protein, fats, resins, gums, and starch. Olmstead and
lilliams first tried Remy's enzymatic method which consisted of: pepsin-HUI,
neutral malt diastase, pancreatin-sodium carbonate. By this method, they
found the starch completely removed, also the proteins, fats, and resins were
adequately removed, but with a substantial loss (twelve per cent) in the
hemicellulose. Successive analysis of the steps indicated that the loss

was essentially in the malt diastase treatment. Leurs found in his study

of the malt enzyme, zytase, that it split hemicellulose. They then tried
another plant diastase, taka diastase, but found a much greater loss of
hemicellulose (thirty-one per cent). Finally, they found that animal di~
astase, pancreatin in neutral solution, removed starch without concurrent
loss of hemicellulose. The proteolytic enzymes of the pancreatin are quite
efficient and the amyloytic power is not greatly reduced at a pH of eight.
To obtain simultaneous digestion of protein and starch, the pH of the digest
was adjusted to eight and the hemicellulose recovery indicated that none of
it was lost in the feces and wheat bran. However, when the treatment was
applied to air-dried vegetables, as much as a forty per cent loss occurred.
In fatty materials, bile salts may be important in the pro-treatment. It
has already been indicated that cellulose is dissolved by sixty to seventyb
two per cent sulfuric acid and lignin is dissociated from it. Thus, it
would appear that the first treatment for the quantitative analysis of

lignin, cellulose, and hemicellulose is the treatment with strong sulfuric

30.
acid after the enzymatic digestion. Olmstead and Williams found seventy-
two per cent sulfuric acid at six to ten degrees had a tendency to char:
the pentosan recovery was low, and the lignin value was high. Jenkins
found that a lignin-like material was formed from pentose by treatment
with seventy-two per cent sulfuric acid. Seventyatwo per cent sulfuric
acid caused a loss of hemicellulose, but sixty per cent sulfuric acid did
not. The optimum for pentosans was fifty per cent sulfuric acid. lith
fifty per cent sulfuric acid, the cellulose is precipitated on being dil-
uted to four per cent. Sixty per cent sulfuric acid does not appreciably
decrease the yield of pentose and converts cellulose between ninetyasix
to ninetyaaight per cent: therefore, sixty per cent sulfuric acid was
adopted by Olmstead and flilliams.

In using strong sulfuric acid for dissolving cellulose and then diln
‘uting to hydrolyze the cellulose into glucose, Williams and Olmstead
found that the maximum reduction was obtained at the end of two and one-
half to six hours. Consequently, three hours was selected as the optimum
time. The higher the temperature, the less time was needed. Sixty per
cent sulfuric acid at six to ten degrees reached an Optimum between sixa
teen and thirtyafour hours.

To summarize, the principle of the analysis is this: Twenty-one and
four-tenths normal sulfuric acid.under controlled conditions dissolves the
cellulose and hemicellulose completely and dissociates them from the lig-
nin. The twenty-one and four-tenths normal acid is diluted to four per
cent and boiled for three hours. The lignin is precipitated quantitat-
ively, whereas the cellulose and hemicellulose are converted into their
constituent simple sugars, whiCh are soluble. The lignin is then filtered
and weighed. The cellulose and hemicellulose, now in the form of simple

sugars, are determined by cepper reduction. The non-fermentable reduction

31.
representing the hemicellulose, and the fermentable the cellulose and
mannans. Beduplication of results depends primarily on one's familiarity
with capper reduction technique. Iilliams and Olmstead also stress the
importance of adjustment of the sugar solutions to neutrality before

analysis.

of th

The reagents necessary are:

(l) Buffer-bile salt solution which consists of fifty cc. of .2M
potassium acid phosphate, twentyathree and four-tenths cc. of ,NN sodium
hydroxide, six and six-tenths cc. of water, and two grams of sodium tauro-
cholate.

(2) A.pancreatin—sodium chloride solution prepared daily and con-
sisting of one-hundred cc. of eight and one-half per cent sodium chloride,
and ten grams of pancreatin, the mixture being shaken for one-half hour
and filtered.

(3) A.sixty per cent sulfuric acid solution is made by diluting
six hundred cc. of G.P. ninety-five per cent sulfuric acid to one liter.

If the sample is dry, it may be weighed directly. In the case of
feces, if the fresh stools are to be used for analysis, the moist weight
is obtained and multiplied.by four, and this amount of water added. The
diluted sample is then ground in a ball mill for twenty minutes until it
passes through a twenty-mesh seive. Twenty-five cc. of this mixture is
then used in the determination. The sample is then transferred to a fifty
cc. glass stoppered bottle, steam sterilized at fifteen.pounds pressure
for thirty minutes and then cooled below fifty degrees. This kills the
spores and gelatinizes the starch. Twenty cc. of the buffer bile solution,

five cc. of the pancreatin-sodium chloride solution and a few drops of

32.
stoluene are added, and the sample incubated for three days at forty-five
degrees 0., with occasional shaking. The mixture is then filtered through
one-hundred twentyafive mesh silk cloth or centrifuged. The filtrate con-
taining the sugars, and nonpcarbohydrate material is discarded. The residue
containing the nonpsugars hemicellulose, cellulose, and lignin are washed
with two hundred cc. of water, fifty cc. of hot alcohol, twenty-five cc. of
hot benzene, and finally twenty-five cc. of ethyl ether. The residue is
transferred to a fifty cc. glass-stoppered container before the other is
completely evaporated. The residue is then dried in an oven at seventy
degrees C. for two hours or until residue is dry. Twenty cc. of chilled
(six to ten degrees C.) twenty-one and four-tenths normal sulfuric acid is
added, and the flask is briskly shaken at hourly intervals (particularly
during the first five hours) and is kept in the refrigerator for twenty-
four hours. The mixture is then rapidly diluted to three hundred cc. with
distilled water, refluxed for three hours, cooled to room temperature,
filtered through a loose layer of asbestos in a Gooch crucible and washed
successively with water, hot alcohol, benzene, and ether. The residue is
dried at one hundred and ten degrees, weighed, ignited, and reweighed, and
the loss of weight calculated as lignin. The filtrate containing the cel-
lulose, hemicellulose, and uronic acids as reducing sugars, is neutralized
with fifty per cent sodium hydroxide to phenol red. The filtrate is then
diluted to five hundred cc. Forty cc. aliquots are fermented by the
Somogy washed yeast procedure, and the reducing sugars are determined by
the ShafferbSomogy method. The total reducing sugars are determined with
aliquots of the unfermented filtrate by the Shaffer-Somogyi method also.
The non-fermentable reduction is interpreted on the xylose-arabinose curve
and multiplied.by the factor .88 to convert pentose to pentosan (hemi-

cellulose). The fermentable reduction is interpreted on the glucose curve

33-
m and multiplied by the factor .9 to convert it to cellulose.

Dale

The samples as described previously in this paper were analyzed for
cellulose, hemicellulose, and lignin by these methods outlined by Olmstead
and Williams. The results of these analyses will be found in tables I to

V, inclusive

Methgdg of sztop. and Lam” agd
wgrg leading up to the methods

.As a result of the work of such men as Norman and Jenkins, Goes,
Phillips, and others, Crampton and Maynard decided upon using some form of
the seventy-two per cent sulfuric acid method for lignin. The special
problem in the case of forage and animal feces lies in the removal of
protein without a simultaneous removal of part of the lignin. According to
the present information, lignin is soluble in varying degrees in dilute
alkali (hot or cold), boiling water, and dilute mineral acid (1.25 per cent
sulfuric acid at 100° 0.). Pre-treatment by enzyme digestion, however,
would seem to be a suitable possibility. In searching for a suitable meth-
od, they studied the work of Olmstead and lilliams, who in studies on human
diets and feces had.used a buffered pancreatin solution at a pH of eight.
In their proposal, cellulose and hemicellulose were determined on the
filtrate remaining on the removal of lignin. Inasmuch as a slight amount
of hemicellulose is removed by the alkaline pancreatin solution, a cor-
rection is made by analysing an additional sample of the diet, omitting
pancreatin. Lignin values are taken from the sample digested with pan-
creating. If lignin were isolated from both the enzyme and non-enzyme re-

sidues, any effect on the pre-treatment on the lignin would be seen. When

this treatment or procedure was tried with sheep and steer diets of grain

31+.
\vand hay, the results in every case showed a smaller lignin value in the
pancreatin treated sample. It seems probably that the long exposure
(seventy-two hours at fortyafive degrees C.) in a buffered solution at

a pH of eight, using sodium hudroxide as the alkali, might have dissolved
some of the lignin and thus resulted in lower lignin values. .Another dif-
ficulty was encountered in the use of the Hilliams and Olmstead procedure.
It was found impossible in the case of animal diets to get complete solut-
ion of the enzyme residue in the concentrated sulfuric acid. Subsequently,
when a lignin balance was attempted in a steer digestion trial, twenty-
five per cent more lignin was recovered from the feces than was consumed.

Pepsin, on the other hand, is active in an acid medium (pH 1-2) and
it has not been shown that lignin is soluble in dilute mineral acid at the
temperature employed. The effectiveness of this enzyme in removing proteins
was uncertain in view of studies by Horwitt in which not more than eighty-
nine per cent of the nitrogen was removable from spinach leaf by pepsin
digestion. There was the possibility, however, that the protein of the
materials would be reduced by pepsin digestion to a level no longer ser-
iously interferring with the lignin determination. If the hypothesis is
accepted, that lignin is not utilized by the animal, the usefulness of the
pepsin pre-treatment and subsequent analytical steps would be indicated
by a lignin balance trial.

The problem of rapidly dissolving the undigested residue in the
strong acid was solved by Poss and Hill, who found that lignified tissues
dissolved promptly (10-15 minutes) in seventyatwo per cent sulfuric acid
if first moistened with formalin. With acid in contact with the sample
just a short time, the chances of the formation of substances from the
carbohydrates (pentoses and hexoses) which might add to the lignin values,

was presumably largely avoided (Bitter). The use of a granulating reagent

35.
.4 (chloroform and acetic acid) in the precipitation of the lignin to hasten

the time necessary for filtration was preposed by Ross and Patter.

It seemed, therefore, that pepsin digestion<3f the ether extracted
sample, followed by solution of the residue in seventybtwo per cent sul-
furic acid and subsequent precipitation of lignin according to Ross and
Hill, and Bosstand.Patter procedures, could be successfully used for the

lignin determination in animal feed and feces.

D o e at

.L one gram sample of feces or feed is extracted with ether and dried
in an oven. It is then placed in a fifty cc. glass-steppered flask and
forty cc. of the two per cent solution of pepsin in .1 N hydrochloric
acid is added. This is digested at forty degrees for twelve hours with
frequent shaking, especially during the first feur or five hours. The
mixture is then filtered through bolting silk or centrifuged. The filtrate
containing sugars, protein and other substances is discarded.

The non-digestible residue containing cellulose, hemicellulose, and
lignin is then washed with hot alcohol, benzene, and ether and transferred
to a one-hundred cc. beaker and the last traces of other are removed with
mild heat. The residue is moistened with four cc. of forty per cent form-
aldehyde. Four cc. of seventy-two per cent sulfuric acid is added and.the
mixture allowed to penetrate the sample for two minutes. Six cc. of con-
centrated sulfuric acid is then added and the mixture is stirred vigorously
to dissolve the sampke, which would be complete in ten to fifteen minutes.
The beaker during this time is partially immersed in a water bath to keep
the temperature from rising above seventy degrees C. or so. lhen the
Sample has dissolved, thirty-five cc. of the granulating reagent (1:6

mixture by volume of chloroform and acetic acid) is stirred in and the

36.
the mixture poured into an eight hundred cc. beaker containing five hundred
cc. of distilled water. The mixture is boiled gently until all the chloro-
form has been driven off (15 minutes). The solution should be clear and the
lignin should settle in granular form. The mixture is then filtered through
a Gooch with suction. The filtrate containing the hemicellulose and cellul-
ose is discarded. The residue is washed with two-hundred cc. of five per
cent hydrochloric acid and dried, then weighed. It is then ignited and re-
weighed, and the loss in weight is calculated as lignin.

Kurschner and Hoffer have a procedure for the determination of cellulose
in which the sample is freed of non-cellulose, organic constituents by digest-
ion with alcoh 01 and nitric acid. Kurschner and Hanak had another method
in which they substituted acetic acid for the alcohol in the digestion re~
agent, and changed the time for boiling from two or three successive one hour
periods to twenty minutes of boiling. These gave practically the same re-
sults for feces, but lower values from certain feeds were secured from the
acetic-nitric acid reagent. A.possible explanation might lie in a differ-
ence in the resistance of the cellulose fractions of mature hay and grain
as compared to those of immature grasses. Certainly the acetic-nitric acid
mixture is the more powerful reagent. These results together with its
greater simplicity led Crampton and Maynard to use the acetic-nitric acid
reagent for their determination of cellulose. It was found, however, that
alcohol was preferable to water for the first washings to free the cellulose
from the digestion reagent. Centrifuging after each washing facilitated
the washing operation.

.L one gram air-dried, ether-extracted sample is placed in a one hundred
and fifty cc. round bottom flask. This flask must be wide necked and fitted
with a reflux condenser. Fifteen cc. of eighty per cent acetic and one and

one-half cc. of concentrated nitric acid are added and the mixture is boiled

37.

‘gently for twenty minutes. The mixture is then transferred to a fifty cc.
centrifuge tube and after twenty cc. of alcohol (ethyl) are added and the
tube shaken, it is centrifuged for ten minutes and the supernatent is dis-
carded. This washing is repeated and finally the residue is washed with a
stream of hot alcohol from a wash bottle into an alundum crucible. (A
carborundum product called lilfrax' and 'firefrax' was used.) The residue
in the crucible is then washed successively with hot benzene, hot alcohol,
and finally with ether, using suction during the filtering. The residue is
dried, weighed, ignited, and reweighed, and from the loss in weight the cel-

lulose is calculated.

his
The samples as were used in the Olmstead and Williams determinations

were again used.bere. They were analyzed according to the scheme of Crampton
and Maynard which has just been described. Tables VI-X give the results

obtained.

of D l

N r d e t

In certain digestibility studies on forage crops, a method was re-
quired which would yield information that could not be obtained by the
usual methods. In addition to giving the desired information, the method
should be sufficiently simple so as to be readily adapted to routine labor-
atory procedure. Davis and Miller tried a number of methods and suggested
several modifications which appear to have a number of advantages over
former procedures.

They prOposed a method combining both enzymatic and chemical proced.

ures. Previous dry ether extraction before enzymatic treatment was found

38.
a to give a more uniform enzymatic action. They also found it necessary to
autoclave the samples before adding the enzymes in order to prevent mold
growth. The methods proposed by lilliams and Olmstead, Horwitt, and by
Cagwill and Mendel, were used on a sample of Red Top grass out early in
Junel Davis and Miller found that the Williams and Olmstead method.gave
higher results for the undigested residue than the Horwitt, Cogwill and
Mendel method. Both methods gave approximately the same results when used
on samples of feces.

The effectiveness of various treatments in removing nitrogen from a
sample of Red Top was tested and the results of this study showed.that
enzymatic digestion was best. In subsequent work, it was adopted for pre-
liminary treatment of the sample. The Horwitt, Cogwill and Mendel methods
were modified by using a smaller volume and reducing the amount of pepsin
and trypsin. It was found.the amount of pepsin and trypsin did not have a
pronounced effect on the amount of nitrogen removed from the sample if more
than fifty milligrams per gram of sample was used. The trypsin extract was
filtered to prevent any undissolved material from increasing the weight of
the residue.

Davis and Miller also conducted tests on the fineness of the sample
and showed that best results were obtained when the entire sample passed
through a five-tenths millimeter screen. Apparently, the enzymatic action
is more uniform if the particles are small. The amount of material digest-
ed is greater in fine than in coarse samples.

After the enzymic digestion, the lignin is determined. Davis and.uil-
ler observed that the concentration of the acid is the important factor.
The results showed that the concentration must be above sixtyéfive per cent
for grasses. Below this, they found the results too high.

The time and temperature factors of the seventy-two per cent sulfuric

39.
_acid method were studied.by Ritter and Seborg and Mitchell. The results

they obtained using wood were confirmed in the present experiments on grass.
Formaldehyde was used by Crampton and Maynard to bring about quicker solut-

ion of the sample and to improve the rate of filtration. Davis and Miller,

however, found.that the use of formaldehyde increased the yield of lignin,

and consequently its use.

es 1 f e t d

The details of the Daria and Miller method are as follows: .A one gram
sample is extracted with anhydrous ether for sixteen hours. The sample is
transferred to a one-hundred cc. wide mouth, glass stoppered flask after
having been dried. The sample is moistened with water and autoclaved at
eighteen pounds pressure for one hour. Fifty cc. of .1 Nlhydrochloric acid
and one-tenth gram of pepsin (1:3000) are added and the mixture is filtered
and washed with water, after which it is returned to the flask by means of
sixty cc. of an aqueous extract of trypsin containing one-tenth gram of
trypsin.powder. The solution is made slightly alkaline with sodium hydrox-
ide and incubated for ninety six hours. The mixture is again filtered,
washed with water, alcohol, and ether. It is dried at one-hundred and ten
degrees c. for one to two hours and weighed; this is the undigested residue.
All filtrations are made through two-hundred mesh bolting silk, although
centrifugation was used throughout the present studies. The residue is
transferred to aluminum dishes for final drying and weighing.

The undigested residue is hydrolyzed with five per cent sulfuric acid
fer one hour. The mixture is filtered, washed, with hot water, and finally
with alcohol. The residue is transferred to a one-hundred cc. beaker.and
placed in a freezing bath and twenty cc. of seventy-two per cent sulfuric

acid is added with constant stirring. After fifteen minutes it is removed

Mo.
and the reaction is allowed to continue at room temperature for fortyafive
to sixty minutes. The mixture is stirred constantly during the reaction
period. The sample is then transferred to a one liter flask and diluted
to three per cent by weight of sulfuric acid and refluxed for two hours.
The mixture is filtered and the residue is washed.with hot water. The res.
idue is dried, weighed, ignited, and reweighed, and the loss in weight is
teported as lignin. The final filtration should be carried out within
thirty minutes after the completion of the refluxing. If the solution is
allowed to become cold, it is difficult to filter and wash, and if the anal-
ysis is completed, the results are invariably too high.

Cellulose was determined the Davis and Miller scheme by the acetic-
nitric acid method of Kurschner and Hanak, which also is the one used by
Crampton and Maynard.

Davis and Miller made no attempt to separate the remaining components
of the carbohydrate complex. These “other carbohydratbs“ can be obtained

by difference.

Etc
The samples previously used were subjected to the Davis and Miller

scheme of analysis. The sum of the lignin and cellulose values were sub-
tracted from the weighed undigested residue. This value was considered as
representing Iother carbohydrates“ probably mostly hemicellulose. The re-
sults of these samples as analyzed by the Davis and Miller methods are shown

in tables ZI-XY.

I rk o e ro e -li i v 'ue

‘223Qriptigg of the wggk
In analysing various samples by the Olmstead and William's method and

also by the Crampton.and Maynard methods, lignin values were obtained which

141.
\were quite inconsistent. The Olmstead and Williams values were much lower
than the Crampton and Maynard. These inconsistencies were experienced by
others in a comprehensive fiber study in 1939. It was therefore decided
to investigate the residue. This was done by determining the nitrogen con-
tent of the original samples before digestion. The same samples were di-
gested by the three enzymatic methods. The residues were collected,waahed,
and dried, and nitrogen determinations made by the micro Ejeldahl method.
The methods giving high lignin values were suspected of not having the
nitrOgen removed sufficiently by the particular enzymatic digestion proc-

ess employed.

‘Qata

The samples used previously were again selected for this phase of the
problem. The summary of results of these digestion studies in which the
relation of nitrogen present to lignin values was being determined is pre-

sented in tables XVI-XX.

Bartigle gigs and Heat Treatment
Degggiptigg of the nor;

In some recent nutritional work here, the question arose to whether or
not the methods of drying the samples as well as grinding might not be
influencing the analytical results. The samples had been dried (both food
and fecal) by heating five to six days on a steam bath. The grinding‘was
done in a liley mill, which gave a fairly fine particle size. It was de-
cided to take some All-Bran and moisten it to form a paste, and this was
placed on a steam bath and heated for_a.period of six days. .As the water
evaporated off, more water was added and stirred into the bran. .At the
end.of the heating period the sample was divided into two parts. one part

was ground in a.beart mill adjusted to give a grind of slightly larger

1+2.
particle size than the Wiley mill. The other part was finely ground in a
ball mill. Another portion of the bran which had not been treated with
heat was also divided into two parts, each part being ground in the same
manner as were the two parts of the heat-treated bran. These four samples,
two different grinds on two samples, one treated with heat and the other

not, were analyzed for crude fiber.

Data
The results of these studies on the relation.of particle size and heat

treatment to the crude fiber value are presented in Table XXI.

Cellulose Determinations:

In studying column one of tables I-V, VI-X, containing the results of
the cellulose determinations by the methods of Olmstead and Williams,
Crampton and Maynard, and.Davis and Miller respectively, one can see a
more or less definite variation between.the methods. Inasmuch as Davis and
Miller used the Crampton and Maynard method, the results were obviously the
same. The Olmstead and Williams method gave results on the fecal material
much lower than the method of Crampton and Maynard. The average percent
variation is about 25-50%. The Olmstead and Williams method offers numer-
ous places in which errors might enter in to give low results, if the
Crampton and Maynard method is to be taken as the correct value for the
per cent cellulose. First of all, there is an enzymatic digestion and
transfer in the Olmstead and Williams method. Secondly, there is treatment
with acid and hydrolysis. Thirdly, there is an indirect final step, in
using a cepper reduction method for the total sugars on one aliquot, while
fermenting the hexoses out and running another capper reduction procedure
for the non-fermentable sugars. These values are both read from curves and
the difference between the total and nonpfermentable sugar reduction is
taken as the cellulose value. Thus, one can easily see that with so many
transfers of the sample and stages in the determination, and the determinp
ing of the sugars by copper reduction and differences between two reductions,
that numerous errors might enter into the analyses to give the low results
as shown in tables I-XVIII when the Olmstead and Williams method is com-
pared with that of Crampton and Maynard. 0n the other hand, in the Crampton
and Maynard method, the ether extracted.sample is weighed out directly into
the reaction flask, the reagents are added and the contents heated for 20

minutes and filtered. The reagent takes out everything but the cellulose

1m.
and a small amount of ash which does not interfere, because after ignition
the ash is still present and the cellulose has been determined in a direct
manner, with no transfers except in filtering the final product. One can
see that when the analyses depend on finding the amount of a given substance
by determining one of its groups or parts or end products, a great number
of factors present themselves to be solved or regulated by definition, any
variation of which usually results in analytical error. The only place
that one may have to be cautious in the Crampton and Maynard method is to
be sure in rinsing the final product into the weighing crucible or centri-
fuge tube that all particles adhering to the sides are removed by rinsing
the tube with hot alcohol. The food samples, although a few are missing
and two or three other samples are somewhat out of agreement, have the same
trend: that is, the Crampton.and Maynard method gives higher results. Al-
though the samples analysed by the Olmstead and Williams method were done
in duplicates on the same filtrate (originating from a one gram sample)

the checks were not any closer and in most cases not as close as they were
in the Crampton and Maynard method, in which the duplicates were analysed

at different times on separate one gram samples.

figmigellploge determinations
.thhough Crampton and Maynard did not have any definite method for

the determination of hemicellulose in their scheme of analysis, they pre-
sented an alternate plan, that is, estimation by difference of the total
hemicelluloses plus any other carbohydrates not cellulose. The results
for hemicellulose by the method of Davis and Miller were calculated as
follows: The indigestible residue from the enzymatic treatment of the
sample was washed, dried and weighed. This residue included mostly cellup

lose, hemicellulose, and lignin, with perhaps a very small percentage of

MN.¢L.
protein plus carbohydrates which might not have been previously removed.
Davis and Miller suggested subtracting the sum of the cellulose and lignin
from the total indigestible residue and calling this difference "other
carbohydrates“, rather than hemicellulose. In studying column two in
tables I-V, XI-XV, and table XVII, one can notice how closely the so-called
“other carbohydrates” portion of the Davis and Miller scheme agrees with the
hemicellulose values as obtained by the method of Olmstead and Williams.
Both are indirect methods. The fact that the results by the method of Davis
and Miller check fairly well with the Olmstead and Williams procedure seems
to indicate that material involved in the former may represent largely the
hemicellulose fraction. The Davis and Miller method seems to give almost
consistently higher values than the Olmstead and Williams with both the
food and fecal samples. The average variation between Davis and Miller
values and Olmstead and Williams values is about .27. The method of
Olmstead and Williams seems to give better checks on duplicates because
they are all aliquots from the same filtrate, whereas the duplicates on
the Davis and Miller method are upon two different one gram samples and
were also obtained by difference, which would tend to cause a little more

variation between duplicates.

Lignin Determinatigns
The results of the lignin determinations, which are given in column

three of tables I-XE,show a definite variation. .A summary average is
shown in table XVIII. The method of Olmstead and Williams gives the low-
estvalues. whereas that of Crampton and Maynard gives the highest values.
The values by the Davis and Miller method are intermediate. The results
are fairly consistent for both the fecal and the food samples. The

Crampton and Maynard method gives an average lignin value range of from

1&5.
twenty-four to twenty-eight per cent on the food samples. The average
range for the lignin value by the Olmstead and Williams procedure is from
about four and one-half to eight per cent on the fecal samples and from
two-tenths to three-tenths per cent on the food samples.

The Davis and Miller procedure average range gives values of twelve
to eighteen per cent on the fecal material and from seven-tenths to three
and six-tenths per cent on the food samples. All of these methods use the
seventybtwo per cent sulfuric acid method with modifications either of time,
temperature or concentration of the sulfuric acid.

The Crampton and Maynard value seemed much too high as compared with
either of the other two values. This is probably due in part to the enzym-
atic treatment, because the nitrogen value of the residue was relatively
very'high.

The literature contains many references to the effects of the presence
of nitrogen, aldehydes and sugars in the enzymatic residue, as well as the
effect of time, temperature and acid concentration on the lignin value. It
is known that some carbohydrates, such as sugars, xylose, arabinose, fruct-
ose and sucrose, will in the presence of strong acids (72% sulfuric acid or
stronger), such as used in the lignin determinations by the sulfuric acid
method, give substances similar in physical and chemical properties to lig—
nin. It is also known that the presence of nitrogen in protein form gives
fission products which will condense with the lignin and thus give high
lignin values. Aldehydes also condense with lignin and this may account
for high values. Some of the higher sacdarides containing a:pentose yield
furfuraldehyde to some extent in the lignin determination, and this may
condense with the lignin. Fatty substances are also capable of increasing
the apparent lignin value, but due to their extraction with ether these

were reduced to a negligible amount in the above determinations. The method

116.
of Crampton and Maynard was found to give an enzymatic residue extremely
high in nitrogen, which may have been partly responsible for the high lig-
nin values. In the same method, formaldehyde was used, which has been
found by many workers to be a cause of high results. This would seem to
fit in with the idea that aldehydes may condense with the lignin to give
huminplike substances and subsequently high values. The use of seventy-
two per cent sulfuric acid has also been found by some workers to prod-
uce some charring of the sample and high lignin values may also result
from this. The Crampton and Maynard method makes use of this seventy-two
per cent sulfuric acid. The Olmstead and Williams method, as well as the
Davis and Miller method, has a more thorough enzymatic digestion procedure,
the Olmstead and Williams procedure making use of pancreatin, while the
Davis and Miller procedure makes use of pepsin, clarase and trypsin. These
procedures both take out more nitrogen than does the Crampton and Maynard
method. The Olmstead and Williams method makes use of sixty per cent sul-
furic acid and subsequent dilution and hydrolysis. The Davis and Miller
method uses seventy-two per cent sulfuric acid for forty-five to fifty
minutes, followed by subsequent hydrolysis after dilution to five per cent
sulfuric. The use of this strength acid and for a period as long as fifty
minutes may have given results slightly high as compared with the Olmstead
and Williams values. The methods and the reasons for the values may be

summed.up as follows:

ill! ' F‘II

IR!!! IIIU tul-

Gama“

“NIH

nun A?- !I.‘ blu-

....I to: Fills;

~Av

f' .Ll >1. ‘7

M7.

mfiwhfloaehc an eoeoaaom
owes mo conceaam on
magmas no

maepmaoQEop moon cocoa
on eoeoaaw convenes

pcmmoum

cowOApac mode 00»
doom no:

cofiumomad owpmeanco
ence

sandmasm &N~ mo own
ommmocfimsaoa mo own

mopncfis
om-m: you awesome»
camps op eokoaao
mamamw ecu ewes .ocoo

.comn macs sandmanm mNN

penance ca
one ewes cocoa and
onsmw can» canon :N

mopfimfia mHuoH hfiqo pompcoo a“
on op dokoafim mfimsmm and ones

mm on sass
no nonsense asses eosaaoaean
coapmowflo oaumeaamo doom

senses and:

ease .ocoo mo pompcoo mcwano
Aoooaumv easssnoaaop soaaaeo
mofipmomwo unwashuco doom
masaaoaean ens soapsaae

awe co assumes sameness mom

madame umomwfim

opmwooaaoqu

sesame awowoa

.n

.m

damsamz
was nonmaoao .m

amHHfiE
cud wfibmfi .N

masaaaaa
ens esomeHo .H

 

mamas» unsuna

ease spam gamma eons:
scenes «as an sauna

menace cammdfl Boa sham pmmfis
cones venues on» an mnopm

kw

mosfimb kc same

scene:

MS.

In a rough average, the Davis and Miller method seems to average three
times the Olmstead and Williams value, while the Crampton and Maynard values
average about five to six times the value by the Olmsteadnand Williams pro-
cedure. Some idea as to the reproducibility of results on the various meth-
ods may be gained from the fact that duplicates by the Crampton and Maynard
method had a grand average of .6h per cent difference between them, while
the methods of Davis and Miller and Olmstead and Williams had exactly the
same reproducibility, both having a grand average difference between dup-

licates of .06%.

.itr . -l i us

The purpose of carrying out this study has been discussed previously
in this paper. The results are shown in tables XIX-XXIII. With a few
exceptions, there seems to be a relationship between the amount of nitrogen
left in the residue after the enzymatic digestion and.the lignin value de-
rived from this residue. The resulting percentages of nitrogen are given
both on the basis of the digested residue and the original undigested
sample. The fecal samples all seem to hold true to the pattern, that
those residues containing the largest percentages of nitrogen also give
the highest lignin values. The food samples, with the exception of the
basal diet, do not rigidly follow this pattern. This may be due to pre~
sence of substances other than nitrogenous compounds in the residue after
enzymatic digestion. Substances such as carbohydrates contining pentoses
might yield furfuraldehyde in the lignin procedure and this subsequently
combined with the lignin to form humus-like condensation products, thus
giving rise to high lignin values without the nitrogen content of the res-
idue being correspondingly high. This may be the reason for irregularit-

ies of the lignin values of the food samples in columns two and three of

149.

table XXIII. In studies made last summer by other workers, it was found
that the residue obtained by the Crampton and Maynard method frequently

contained as much nitrogen as the original material.

Partigle-size and heat-tregtment results

The results of the study of the effect of particle size and heat
treatment on the crude fiber values are shown in table XXIV. They indic-
ate that in general particle size and heat treatment have little or no
effect on the crude fiber values. The ball mill grind on both heat-
treated and non-heat treated bran differed by only .19 per cent on four
sets of four samples each. The coffee mill grind on both the heat—treated
and nonpheat treated bran differed by only from .h to .5 per cent on four
sets of four samples each. The difference between the heat-treated coffee
mill and ball mill grind was only about .9 to 1.0 per cent. The differ-
ence between the non—heat treated coffee mill and ball mill grinds was
about .19 per cent. These variations are within the limits of accuracy

of the crude fiber determination.

2.

7.

9.

10.

50.

CONCLUSIONS

The Olmstead and Williams cellulose determination gives lower values
than does the Crampton and Maynard procedure.

The Olmstead and Williams method for lignin gives lower values (one-
fifth to one-sixth) than those of the Crampton and Maynard method and
one-third that of the Davis and Miller procedure.

The Davis and Miller "other carbohydrates" fraction has about the

same value as the hemicellulose fraction determined by the Olmstead
and Williams method.

The Crampton and Maynard enzymatic residue averages contain more
nitrogen than do the Olmstead and Williams or Davis and Miller.

The Olmstead and Williams enzymatic residue contains the least amount
of nitrogen.

The Olmstead and Williams enzymatic procedure seems superior because
it removes more nitrogen and is faster than that of Davis and Miller.
There appears to be a relationship between the percentage of nitrogen
remaining the enzymatic residue and the lignin value. In general, the
higher the percentage of nitrogen in the residue, the higher the lig—
nin value.

The Crampton and Maynard cellulose procedure seems to give better
checks on different samples than does the Olmstead and Williams method.
The Olmstead and Williams method for hemicellulose seems to give bet-
ter checks than the Davis and Miller determination of “other carbo-
hydrates“.

The procedure of Davis and Miller and Olmstead and Williams gives
duplicate checks of about the same differences for lignin. Both give

better checks than do the Crampton and Maynard procedure.

51.

11. The Crampton and Maynard method for cellulose seems to be a better
one than the method of Olmstead and Williams in that it is simpler
and more direct.

l2. Particle size does not seem to have any effect on the value of crude
fiber on the same sample of bran.

‘13. Heat treatment does not seem to have any effect on the crude fiber

value of the same sample of bran.

BIBLIOGRAPHY

(l) Olmstead, W. H. and Williams, R. D., a Bio-Chemical Method for Deter-
mining Indigestible Residue (crude fiber) in Feces: Lignin,
Cellulose, and Non-water Soluble Hemicelluloses, J. Biol. Chem.,

1_0§. 653 (1935)

(2) Crampton, s. w., and Maynard, L. A., The Relation of Cellulose and
Lignin Content to the Nutritive Value of Animal Feeds, J.
Nutrition, .15. 333, (1933)

(3) Davis, R. E., and Miller, C. O., tudies on the Partition of the Less
Easily Digested Carbohydrate Complex of Feeding Stuffs, U. S.
Dept. of Agriculture.

(h) Didwell, G. L., Report on Crude Fiber, J. Assn. Off. Agric. Chem.,
5, 1421, (1922)

(5)r1 Lawellin, S. J., Crude Fiber Determinations., J. Am. Assn. Cereal
Chem., 1, 208 (1922)

(6) Sherman, H. 0., The Insoluble Carbohydrates of Wheat, J. Am. Chem.
506., 2-29 2919 (1897)

(7) Browne, C..A., and Beistle, C. P., The Complete Analysis of Feeding
Stuffs, J. Am. Chem. Soc.,.gj, 229 (1901)

(8) Simon, 0., and Lobrish, H., Eine neue Methode der quantitativen
cellulose-bestimmung in Mahrungsmitteln and Faeces, z. Physiol.
Chem., 3g, 55 (19014)

(9) Konig, J., 2 ur kenntness der pflanzlichen zellmembran, J. Ber.

Chem. Ges., 33, 3561+ (1906)

(10) Remy, E., Experimentelle Studien zur Biochemie und BiolOgie der
Rohfaser, Biochem. z., _1_. 236 (1931)

(11) Horwitt, M. X., Cowgill, s. s., and Mendel, L. 13., Availability
of Carbohydrate and Lots of Green Leaf tagether with some
Observations on Crude Fiber, J. Nutrition,‘lg, 237 (1936)

(12) Norman,.A. G., The Composition of Forage Crops, I Rye fGrass,
Biochem. J., 19, 1391+ (1936)

(13)

(in)
(15)

(16)

(17)

(18)

(19)
(20)

(21)

(22)

(23)

(2h)

(25)

2.

Dean, A. L., and Tower, G. E., The Estimation of Cellulose in Wood
by the Chlorination Method, J. Am. Chem. Soc., 29, 1119 (1907)

Schorger, A. W., The Chemistry of Wood, II Discussion of Methods
and Results, J. Ind. Eng. Chem., 9, 561 (1917)

Bitter, G. J., Modified Method for Determining Cellulose in Wood,
J. Ind. Eng. Chem., l_6_, 9’47 (19214)

Jenkins, S. 11., Determination of Cellulose in Straws, Biochem. J.
.22. 1928 (1930)

Opferman, Determination of Alpha—beta, and gammapcellulose,
Paper, gg, 19 (1921)

Bray, 1!. L, and Anrews, T. 1.1., An Improved Method for the Determin-
ation of Alpha., Beta., and Gamer-Cellulose, Ind. Eng. Chem.,

.15.. 377 (1923)

Kalb, 1..., and Schoeller, V., Cellulose Determination by Phenol,
Cellulose Chemie., 3, 37 (1923)

Sieber, 11., Determination of Cellulose in Plant Material according
to the Method of Cross and Bevan, Zellstoff Papier, 3, 27 (1923)
Chem..Abstracts, 11, 171a (1923)

Kohmoto, T., and Sakaguchi, S., The Estimation of Cellulose in
Human Feces and the Digestion of food Cellulose. J. Biochem.
(Japan), 6, 61 (1926), Chem. Abstracts, _2_g, 31176 (1926)

Waksman, S. A., and Tenny, F. G., The Composition of Natural Org-
anic Materials and Their Decomposition in the $011,
I. Methods of Quantitative Analysis of Plant Materials,
Soil Science, 23;, 275 (1927)

Kiesel, A... and Semiganovski, N., The Determination of Cellulose
by Quantitative Saccharification, Ber., Q _B_, 333 (1927)
Che. Abstracts,.gl, 1706 (1927)

Kurschner, K., and Hoffer, A., A New Quantitative Cellulose
Determination, Chem. Ztg., 55, 161 (1931)

Schmidt, 31., Yuan-Chi Tong, and Jandel-‘bauer, W., Preparation of
Skeleton Substances from Encrusted Cell Walls by Chlorine
Dioxide in the One Step Procedure, Cellulose Chemie, 12
201 (1931), Che. Abstracts,‘g§, 1772 (1932)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(3h)

(35)

(36)

(37)

3.

Phillips, Max, The Quantitative Determination of Methoxyl, Lignin, and
Cellulose, in Plant materials, J..Assn. Off. Agri. Chem, $5, 119
(1932

Paloheimo, L., and Valovarra, V., Studies on Iodine Calorimetry of
Cellulose Dextrins and Foundation of an Iodocolorimetric Method
for the Determination of Cellulose, Biochm. z., 266, 301 (1933)
Chem..Abstracts,.g§, 891 (193k)

 

Achorya, C. N., Determination of Cellulose in the Soil, Proc. Soc.
Biol. Chem. India, 1, 8 (1936)

Zizka, J., and thler H., Direct Determination of Cellulose in Artif-
icial Leather, Tech. Elidka Koz., 12, 76 (1936), Chem. Abstracts,
32. 399 (1938)

Launer, H. F., Simplified Volumetric Determination of Alpha, Beta,
and Gamma Cellulose in Pulps and Papers, J. Research Natl.
Bur. StandardS..l§. 333 (1937).

Strepkov, L. M., The Micro Determination of Cellulose, Bot. Arch.,‘3§
M30 (1937)

Ubaldini, I., Oxidation of Coals with Nitric.Acid and the Determina-
tion of Cellulose in Lignites, Ann. Clim. applicata,‘g§, 199
(1939), Chem. Abstracts, 33, 891 (1939).

Kurschner, K., and Hanak,.A., Determination of Cellulose, Z. Unter-
such Legensm., 59, Man (1930) Cited, J. Nutrition,‘l§, 390, (1938)

Wise, L. E., Peterson, F. C., and Harlow, W. M., The.Action of
Ethanolamine on Woody Tissue, Ind. Eng. Chem. Anal. Ed.‘ll, 18
(1939)

Dox,.A. W., and Plaisance, G. P., A Comparison of Barbituric Acid,
Theobarbituric.Acid, and Malonylguandine as Quantitative Pre-
cipitants for Furfural, J. Am. Chem. Soc.,‘3§, 2156 (1916)

Dore, W. H., The Proximate Analysis of Conferous Woods, J. Ind.
Eng. Chem.‘1g, M76, (1920)

Preece, A. I., The Proximate.Ana1ysis of Box-Wood, and the Nature of
its Furfuraldehyde-yielding Constituents, Biochem. J.,‘gfi,
130“ (1931)

(38)

(39)

(MO)

(111)

(1+2)

(“3)

(1114)

(115)

(16)

(117)

(Its)

(119)

14.

Mehta, M. M., Biochemical and Histological Studies on Lignin,
Part I. The Nature of Lignin: its Physiological Significance
and its Estimation in Timber., Biochem. J., 19, 958 (1925)

Fleshsig, 5., Usher Dorstellung und Chemische Natur des Cellulo-
sezuckers, Z. Physiolog. Chem., ¥, 523 (1883), cited in
J. Assn. Off. Agr. Chem., 35, 124 (1932)

Ost, 3., and Wilkening, L., Saccharification of Cellulose, Chem. Ztg.
33;, 1461 (1910), cited in J. Assn. Off, igri, Chem., 15. 121,,
(1932)

Klason, P., Lignin Content of Spruce Wood, Cellulosechem., 3, 81
(1923), cited in J. Assn. Off. Agr. Chem. 15. 1214 (1932)

Waksman, S. S., and Stevens, K. R.,.A System of Proximate Chemical
Analysis of Plant Materials, Anal. Ed. J. Ind. Eng., _2_, 167
(1930)

Willstdtter,.R., and Zechmeister, L., Zur kenntnis der hydrolyse
von cellulose, Ber., 26, 2MOl (1913), cited in J..Assn. Off.
Agri. Chen. 15. 124 (1932)

Konig, J., and Rump, E., Chemie und Strukter der pflanzen-zell-
membran, Z. Nahr. Genussm.,‘g§, 188 (191M), cited in J. Assn.
Off. Agric. Chem., 15.. 1211 (1932)

Wenzl, H., The Quantitative Determination of Lignin in Wood Pulps,
Papier-Fab., 22, 101 (1929), cited in J.,Assn. Off. Agric.
Chem. , 15, 12F(1932)

Mahood, S. A., and Cable, D. E., The Chemistry of Wood, J. Ind.
Eng. Chem., 19 933 (1922)

_—’

Phillips, Max, The Quantitative Determination of Methoxyl, Lignin,
and Cellulose in Plant Materials, J..Assn. Off. Agri. Chem.,

15. 118. (1932)

Goss, M. J., and Phillips, 10x, Studies on the Quantitative Estim-
ation of Lignin, I. Factors Affecting the Fuming Hydrochlor-
ic Acid Method, J. Assn. Off. Agr. Chem., 19. 391 (1936)

Ring, 0., A.Modification in the Determination of Crude Fiber,
Sucmen Kemistilehti, 11;, 12 (1938), cited in Chem. Abstracts,
.32. 3616. (1935)

(50)

(51)

(52)

(53)

(5h)

(55)

(56)

(57)

5.

Bailey, K., and Norris, F. W., The Nature and Composition of the
Seed.of White Mustard, Biochem. J., gé, 1609 (1932)

Harley, L. F., and Norman, A. G., The Differentiation of Hemi-
celluloses, Ind. Eng. Chem. 25, 1190 (1932)

Norman, A. G., The Biochemistry of Cellulose, the Polyuronides,
Lignin, Etc., Clarendon Press, (1937)

Tenney, F. G., and Waksman, S. A., Composition of Natural Organic
Eaterials and their Decompositbn in the Soil, Soil Sci. §_§
55 (1929)

Fuschs, W., Investigations Concerning Phenol, Lignin, and Methoxyl
Glycol Lignin from Spruce Wood, J. Amer. Chem. Soc., 53, 673,
(1936)

Csonlca, F. A., Phillips, 111., and Jones, D. B. Studies on Lignin
Metabolism, J. Biol. Chem., £5, 65 (1929)

Boroff, C. S., Buswell, A. M., The Anaerobic Fermentation of Lignin,
J. Amer. Chem. Soc., 56, 886 (1934)

Coleman, D. A., Report on the Committee on Methods of Analysis,
Cereal Chem., 5, 269 (1928)

APPENDIX

 

 

 

 

 

 

 

3m. 1...... $8. $6 2...- $6 34 ii... 34 Hanna
395mm doom
sees the
RJ R; mom zone. and and sweep:
mm; :3 3.: m9». SA min mmé 8A. and a
mg; so; an: 3.; 3.... 3.4 mm.“ 3.... mm; on
8.: 3.: mm; mm. mm. a. dam mum $.m a
25m lean. Ram win 3d mmd no.” 84 no; m
a; 8.: 3.: mo.n 8% no.» and mad. :3 0
Rd min and 3. mm. m:. mo; me.” so; a
(“lead a N d m use a m H .n. dental m H
no $323 39.8» 395m mo $93.94 Canaan 3% mo 0.3.8.5. camaam 035.8 282
seesaw . cg sawmill
seen the

33.2: sec ascends

H 53

 

 

 

 

 

 

 

 

 

om. om. om. mg. mg. mg. pm.n m~.n m~.m nenmuaaq
ode—Em cook
Moon deacon
mm.m mm.m mm.HH ms.aa wa.m mm.m omence<
wo.m ma.m no.m sm.m Hm.oa ~m.m ms.m an.» an.» a
mo.m oo.m so.m mo.oa w».m mm.oa n~.m om.m am.m om
Hm.m om.m m~.m NH.Ha ~0.HH mH.HH :o.m mm.w oa.m a
os.m nm.m w:.m H~.NH ~m.ma m:.~a mmam mm.m -.m m
mm.m om.m ~m.m mm.na mz.na om.ma mw.m ow.m Hm.m o
mz.m H:.m om.m sm.aa mo.ma m~.HH mH.m oa.m o~.m a
twinge H m H. m anata m 7” «panama m. as
no 03.35% 393m magnum mo oweugd 395.6 cams—em no owunohd 3956 0358 each
snsnamnw anaamadnmdsmmmw unnnammadqmmwlluntn.
neon deacon

nadflfidh can uneven—Ho

Hands

 

humane

 

 

 

 

magnum
I I I I I I I I I magnum doom
sees senses
-.m mw.m mn.H ::.H ~m.m mm.m sensese
mm.m mm.m mm.w m~.m mw.m s~.m mm.s Ho.m :w.: A
Hm.m mw.m mm.m mH.H mH. mH. mw.H aw.H mm.H cm
m~.m ~a.m mm.w m~.H am.H mo.m oH.: mH.: :o.: a
s:.: Mm.: mm.: am. am. mm. oa.H m~.H mm.H m
mm.HH Hn.HH mm.OH -.m om.m :m.m :H.m wo.m om.m 0
we.» H:.m mm.n an. we. mg. mg.» mm.m HH.m m
m ens H, 7m H m.ene H m H (widen H m H
onam>4 mmemm Hmmsmm mo cmmambd magnum camemm mo ommao><_ magnum magnum mama
swemfig w mmoasaamuwacm mm 0 mu m

 

see: senses

mamfiaaah can encumeao

HHH mamds

 

 

 

 

mNm. mmm. 0mm. :mH. NmH. How. :m.H mm.H mm.H essence
mmm. mmm. mHm. H:o. mmo. mm.H cm.H Nm.H eeneecH eHnsem soon
see: new
:N.: mN.: mm. mm. mw.H mm.H smashes
Nm.m mm.m Hm.m mm.H Nm.H mm.H o:.: m:.: mm. H
mm.m :m.m om.m mo.H OH.H mo.H :m.H mm.H mm.H om
mo.: mo.: mo.: Hm. Nm. mm. mN.H :m.H mw.H a
mm.m Hm.m mm.m mm. mm. :m. mH.H mH.H mH.H m
mm.m Nm.m mm.m NH. NH. NH. mm. mm. :m. o
mm.m Nm.m Hm.m NH. NH. NH. om.H 0H.H Hm.H m
meneH in. H messH m H MsseH N H
.93 398.8 magnum mo mmeao>4 395m Sufism .Ho nuanced 395.8 magnum mesa
59mg R emoHsHHmoHsem m emodHHHmoimN

 

sees neeHm
eanHHHN ens eeeensHo

N; @1549

 

 

 

 

 

NHN. mom. mum. 0N0. :0w0. :mo. mH.H Hm.H mH.H ncmseno
mom. NHm. 00m. mmm. Hmm. mmm. HH.H HH.H HH.H -eeHnns
eHnsem sees
see» new
mm.: m:.: Nm. :m. mm.m Hm.m essence
Nm.m mm.m Hm.m Nm. Nm. NM. mN.m Nm.m mm.m H
m:.m m:.m mm.n NN.H mmHH mn.H :w.m Hm.m mm.m 0m
mm.m mm.: NM. Nm:. mN. mNH. mN.m mN.n ww.m m
0m.m HN.m mu.m 0:. 0:. 0:. mm.H mw.H H.m 0
mn.m m.m Nu.m mm. N0. 0m. 00.m 0m.m 0m.m m
awed a) H «and. m 1H imhdH m. a
mo .34 0353 0388 no 0335.4 395% 395.8 mo oweagd 033mm magnum mama
sHsmHIA w mnoquooHsem M cmogflmu &

 

secs seamen

28.3% .3 odd ddmumafio

>32

 

 

 

 

 

H0.0 00.0 00.0 - a . ~H.H mm.H nH.H mflmwmmawOom
m:.wm w:.~m : u . m~.m m0.m cwwump<
»
m0.Hm 0m.Hm 0m.0m . u - mz.0 00.0 H:.0 A
mm.mm 00.~m n.2m - 1 . am.m 0m.m mm.m om
:m.mm mm.mm 0~.mm . u s :0.m 00.m m0.m a
m:.mm 00.0m 00.0w . - - :0.m :0.N m0.m m
0m.:m Hm.mm 0H.mm - - - wo.m 0:.m H~.m 0
a u : u u - - g - 0
N04 m H mad m a mum \m AT
mo .mp4. magaam mamawm mo omduoh4_ magaww mamawm mo mwdhmhd mamaam mamaww osaz
mflqmfia w .mwdafiflaoofiammm mmoadaamo m

 

goo: pwufim

dhmnhwz.cqa nonmawuo

Hb mAm<H

 

 

 

 

 

hm.HH mm.HH mo.mH a s ww.H :w.H mm.H canmnfiaa
mamaam doom
mw.mm mmmmm . . ~m.mfi :0.NH 000.004
mm.~m mz.~m mm.~m - . nH.mH 0m.mH m~.HH A
:m.Hm 00.mm mm.am - . mm.MH 00.MH :a.ma om
mm.mm ~0.mm :0.0m - - -.HH 0w.HH :~.HH a
- n a u u . 00.HH : m
20.:m 00.nm mfi.:m - - 00.na 0H.MH mm.ma 0
mm.Hm m~.0m 0~.Hm - . mm.MH 0m.ma H0.MH m
N0H ‘m H1 d a mafl m a;
mo .mp4. mamawm camawm mo ommuobd. mamawm mamawm mo wmduo>4_ magadm mamadm mama
capmfiA‘m‘ oondHHoUHawmmw. mmoaflafloo mw11

 

x003 oncomm

vhdchwfi and nonmsmho

HHb mAmda

 

 

 

 

 

 

 

mm.: 00.: Am.: . s . NM.A mm.A 0m.A 000A00
00.0 00.m mm.m . u u :m.A mm.A 00.A 00000»
036.8 doom
mm.~m :0.~m - - . 00.m A0.m 00000»:
0A.mm Am.mm mw.Am . - - 0A.~ NA.» :A.~ A
:uém mmém :mfim .. .. .. +1.: NT: :3: um
m~.mm mm.um :w.mm n - u m0.: 00.: :0.: a
0:.00 mm.mm 0m.mm . s - :m.m mm.m ~m.m m
mm.Am m0.Am :~.Am - n - mm.: m0.m Aw.: 0
:~.:m 00.:0 00.:0 - - u Am.m mm.m wm.m m
:uxmmw,A m 1H. m 0000:. m, A m 0qq|d 01m A
mo omduobd 035$ magmm Mo mwambd. 393m magnum mo wwwnw>< mamawm 393m 93%
dwfig w 0003:004st mm 0003200 W
:00: :00000

dyad“: wad dongswuo

HHH> mama‘s

 

 

 

 

 

:0.0 00.0 00.0 u :0.0 00.0 0.0 000000A
00.: 0A.: :A.: . 00.A 00.0 00.A 0Ammwm0mwom
00.00 0A.00 - 00.0 00.0 000.000
0:.:0 :~.:0 0A.:0 - 00.0 00.0 00.0 A
00.00 00.00 0.00 - 00.0 00.0 00.0 00
00.:0 0.:0 00.00 a 0:.0 00. 00.0 a
00.00 00.00 00.:0 - 00.: A0.: 00.: 0
0m.A0 m0.A0 00.00 - A0.: 00.: m:.: 0
00.00 00.00 00.00 1 00.0 00.0 00.0 m
2.....- .0... ..... .000... .0... .0... 20...... .0-.- .3. :..
mammaa w. mmofisfidmofiemm m mmoasflfiou a ..

 

x00: npxfim
0002003 wad nonmawuo

NH mamdy

 

 

 

 

 

00.0 00.0 00.0 u n . 00.0 00.0 00.0 000000
00.0 00.0 00.0 . u t 00.0 00.0 00.0 0000000
000000 0000
:o.:m Omozm I I mm; SJ. 00.0.8.5
00.00 00.00 00.00 s a . 00.0 00.0 00.0 A
00.00 00.00 00.00 - u a 00.0 00.0 00.0 00
u n u u - - n a n a
00.00 00.20 00.00 . a a 02.0 00.0 00.0 0
00.00 00.00 00.00 . u n 00.0 00.0 00.0 0
00.00 00.00 00.00 : n - 00.0 00.0 00.0 0
.0 000 0, 0, 0 .0 000,0 0, H 0 000,0 0 *0
mo 000.234 camSmm 0350.0 .00 0.0.0.024 0.3000»... 0358 we ommnobd magnum magnum 0802
500.03%! 000330009000 00 00000.5ku 0%

 

0000 000000

0.0.0832 .08 003900.00

wag

 

 

 

 

 

ms. om. mg. n u n ~H.H mm.H MH.H Hawam
395mm dooh
mm.m~ mm.ma u u H~.m m~.m om.m omaam><
pm.mfl mm.mH mm.mH no.n : mo.m m:.m m:.m H:.m a
Hm.m ww.m :m.m m~.H - m~.H Hm.m om.m mm.m om
mm.HH HH.NH mw.HH Hm.m u Hm.m no.n mo.m mo.m a
~m.flfi mw.~a mw.Hfi :m.H . :w.a mm.m :m.m mm.m m
mm.~H mm.~H mm.~fl Fm.n a ~m.m wo.m mn.m H~.m o
u u a n a u n u a m
mad m m mafia N H mafia N A4
mo .23 398mm 038mm .3 mwwnm>< 395m madam mo omdaobd magnum 395mm 282
flag $1 mmod.AHmo§mm m 30338 R

 

goo: pmnfim
“mafia: cna mfipwa

HM as

 

 

 

 

 

mm.m ~m.n mm.m mm. mm. mm.a :m.a mm.H nahm-flfl4
mamswm doom
Hm.HH ~H.HH m:.oH ~m.mH :m.ma owaam><
~H.HH m:.HH om.OH om.oa om.oH ma.mfl om.mH m~.HH A
mm.HH mn.HH no.HH mm.m mn.m mm.na om.mH :H.ma om
-.Hfl Hm.HH mm.HH . : -.HH om.HH :~.HH a
I. II nl I. nl DI ' I. m
Hm.oH om.oH mn.OH - : mo.ma mH.mH mm.mH o
~.oH -.ofl m~.oH NH.HH NH.HH mm.ma cm.MH Hm.ma m
nuiqmu,fl 11m a \m1dand “w a ‘m,cqaxd N! H
Mo mwdambd mamawm mHmbm mo wwdhmbd mfimamw onEwm mo omanwk<v afimamm wamawm mama
‘dwdqubw «moanafluuaaom,m, mwoafiaflmu MW

 

x00; cuoomm
A332 can 323

HHM mAm4H

 

 

 

 

 

 

 

:N. m». m~. NH. mm. ma. mm.H mm.H on.“ mgmawo
mm. mm. mm. ma. mo. m. nw.a mm.” om.H mmnnpm
ofimeam doom
:o.wH mm.pfi ~m.m NH.N oo.m Ho.m omwnmuq
:~.mm wo.nm H:.mm - u s ma.» NH.~ :H.~ A
mw.mH m~.mfi m .NH 1 . mm. :~.: m~.: -.: om
:m.wH ~m.mfl Hw.wH : u mm.m mo.: 00.: 30.: a
:m.:H mo.mH :w.:H ::.m om.m mm.H :m.m mm.m ~m.m m
mm.flm H:.Hm mm.flm Hm.m m~.m Hm.m mm.: mo.m Hw.: o
mm.~fl hm.~fi mm.wa mz. Hm. gm. Hm.m mm.m mm.m m
mummy? imam 2? “Guam”? ”mam "as“ m “omufib‘ swam a? 03
1: qfiamfla m, AumoauHHmOHsmm ml madamMAmmuwwn.nuu.
game sphsom

hoHHflz .38 mfibwn

HHHM mag

 

 

 

 

 

 

pmHHHz and mean

b5" “38

oo.H mo.H mm. Hmo. . Hmo. :N.H Hm.H NN.H weapaoa
mo.H wo.H mo.H mm. - mm. mm.H oo.m om.H oHMMMmproH
0:.NH . mm.mH n m:.H Hw.m ~m.m owapm><
mm.oH ww.0H mm.0H hm.H - Hm.H o~.m ow.m mm.m H
H2.MH wN.MH mm.nH oc.m : oo.m mm.m om.m mm.m om
Hm.m 0m.m mz.m mm.H a mm.H mx.m om.m om.m a
Hm.:H mz.HH mm.:H mm. - mm. no.2 Hm.: no.3 m
mw.MH Ho.HH mH.MH : - : Hm.z ow.: m:.: o
0H.mH oz.mH om.HH :m. s :m. um.m mm.m mm.m m
Ncfifi 11m H NcfiH N H NucaH m a
no 93934 03:48 3936 mo mwapm>4 393m odmfimm .Ho owgobd camadmr 393mm oawz
“H53 w uwogHHSHSMR 333ku w -
goo: nanm

 

Hm. mm. ww. N30 I. NJ. mmoH MMoH 30H mcflmnmd.
ow. mh. mm. mho. : mHO. mm.H Hm.H mn.H mwmgwho

o Hmamm .m 0 oh

 

 

 

 

 

mm.mH HH.mH ' mfi. . ww.~ Hm.~ mmaumhq

om.mH mm.mH m~.mH mH. a mH. :~.m o~.w m~.w H
mm.:H mm.:H Hm.:H :o.H : no.H mz.m om.w mm.m om

a u - u x u n a u a
mm.:H HH.HH :m.:H mm. q mm. m:.w HmHm o~.w m
om.mH om.mH 0H.mH ow. . om. mH.~ mm.» oo.~ o
HH.wH Hm.wH m~.mH mm. u mm. :m.m Hm.m mm.m m
mowwfl.‘ 2ng a? mm W? 39mm 2? Hm Mm“? mama” 0an 03

quMHq m uuoHfiHHQOHgom m .lemmmaqwmwdwsunnnun.

 

Hoo=.n»anm
Hog; .93 £me

Baas

TABLE XVI

Average of all Samples on the various Methods for Cellulose

 

 

.Ave. % by .Ave. flfiby Ave. fiby
Samples Olmstead & Crampton and Davis and
Williams Maynard Miller
__Method Hethod Method
lst week (fecal) 2.h2 3.70 3.70
2nd week (fecal) 9.23 12.80 12.80
Nth week (fecal) 2.96 5.00 5.00
6th week (fecal) 1.92 5.87 5.87
8th week (fecal) 3.61 7.67 7.67
lst week (basal) 1.16 1.17 1.17
2nd week (bran) 3.27 1.88 1.88
km week (celery) - 1.214 1.211
Nth week (prunes - 1.32 1.32
6th week (cabbage) 1.6M 1.95 1.95
8th week (lettuce) 1.33 1.214 1.21+
8th week (oranges) 1.19 1.36 1.36
8th week (apples) 1.11 1.39 1.39

 

TABLE XVII

Average of all Samples on the various Methodsfbr Hemicellulose

 

.Ave. fiby Awe. fiby .Awe. % by
Olmstead and Crampton and Davis and

 

Samples Williams Maynard Miller
e_kethod _fiethoi, cuethcd
1st week (fecal) 2.05 - 2.71
2nd week (fecal) 11.51 - 10.h5
nth week (fecal) 1.M3 - 2.5“
6th week (fecal) .60 - 1.h2
8th week (fecal) .60 ~ .79
1st week (basal) .031 - -
2nd week (bran) .h5 ~ .65
hth.week (celery) - ~ .17
nth week (prunes) ~ ~ .19
6th week (cabbage) .ISM ~ .52
6th week (lettuce) .Ohl - .037
8th week (oranges) .070 - .073

8th week (apples) .353 - .u2

TABLE XVIII

.Averages of All Samples on the various Methods f or Lignin

 

 

ave. {by Ave. fly Ave. {by
Olmstead and Crampton and Davis and
Samples Williams Maynard Miller
ffiethod method Method
lst week (fecal) H.36 27.98 13.27
2nd week (fecal) 5.85 26.01; 11.19
nth week (fecal) 6.71 27.38 18.01
6th week (fecal) n.73 2M.u7 12.36
8th week (fecal) “.53 2M.12 15.22
lst week (basal) .218 6.01 .h9
2nd week (bran) .60 11.97 3.63
Nth week (celery) 0 n.23 .7M
hth week (prunes) - 3.02 .95
6th week (cabbage) .273 14.16 1.05
6th week (lettuce) .222 2.711 1.00
8th w eek (oranges) .217 7.86 .70
8th week (apples) .303 6.02 .91

 

 

 

 

mm.H memo. mo.m mm keep 2e: “mmamwvHeeea
afim mmmo. mm.m mm Moot dam mmfivdoooh
mm.w mmmo. mm.mH :m Moor pmnm mwmawmvfiooom
:m.m ommo. mw.: mm keep we: wmmmmm_eeea
02.: some. 0H.w mm aeee emmemmmwwm eeem
Him Name. one Hm. aeee emH emwmm eeerH
nomoupfin ammo pom madam ma unmfioa womfi sommm .oo .02 oflmewm manadm

mHneeHeau

ouoHE :1 woameem Hangmano no qmwouuaz

NHN mqmdy

 

“.0 :5

 

 

 

mm.z HH.N HHmo. om.m m4. aeee a»: eeHeeem Heeea
To :3

Hm.m mm.m 10.00. 15.: m4 Moo: dam mamaom. dmooh
“.0 .AV

3%: mod. :30. 23m :4 Mock umH moflmfiom Hwoom
n Ammadhmv

m:.mH mm.H mmmo. Hm.m 4. keep 2e: eHeeem eeee
. Anaemuaqv

om mm CUM demo. 36 m4 amok dam 393m doom
22.5

3.0 om; mnmo. m~.m 64. Mom: pea magnum doom

|uddeom
eeeeeeHeep eeeHeem eeeeemHn
do wmmam do dmmom madam HS page: com: .00 .oz onEwm 39.4mm
w7eeweeeHa zommm

 

neHeeeHeneueeeHav

umfidamom 953:; 65w cowumeHo no mmwoppwz

KNEE

 

A.o.av

 

 

 

o~.w mH.: :mzo. mw.mH mm Moo: A»: moamamm Hwooh
\ x m A+O.Av
Juno we); mono. wH.w m gee: Ham 3ng deoem
: 70:3
”3.: mm.m mfimo. mw.mH m some pea meamewm adooh
m Amoadhmv
oo.m mm.H momo. om.m m Meet a»: mamadm doom
. m Aquamaaddv
Nr.m Hm.m mamo. w~.m m #00: dam mamaom doom
. “Heeemv
m~ ma 00.2 womo. MH.~ Hm goo: pea oaaaem doom
0H6 mmmm
copmmmflezb oseamom eopmoMan
no dommm no oomom madam ma pnmfiom own .00 .02 onewm madmamm
maowoapwz ommm

 

neHHHeeHHenmaeeeHB

moseHmom waedhaa dam nemempu so newcupfiz

HNN wands

 

A.o.Av

 

 

 

 

mm.H memo. No.» \0 #wo3_g»: mofimemm Hooom
m.mH
m Aoflownv
Ho.m Nam. memo. wH.H o aeee eem eeHeeem Heeem
\. AOOOAV
0.0H on.m zomo. Nw.m :0 xeo: mma madmamm Hwomh
\ m Awmqsumv
n.2m :u.m :woo. mw. 0 Mock A»: onsem noom
. e 228:3
m.:H mm.H mHHo. ea. 0 keep eem eHeeem eeem
Aaomdmv
u I a u flo Moos pea oamfiam doom
umefimmm 1
umpmomaonb oscamem dopmom“n
no dommm do comdm madam a“ pnwfiow came .00 .oz oflmewm camawm
lmwmeeeeeaa :owmm
AeHeeeHenmueeeHav

meseHmmm mabwn cad nofiaaz so demoasz

HHNN HAm4H

 

 

 

mm.H mm.Hm m.mH mm.H 0m.Hm 0~.w wH.: 0m.0H m.: HH.m keen a»: eHeeem Heeea
m~.m 0e.0H H0.m ~00. 00.30 :m.m mm.: 0m.m Hm.m mm.m keen 00m eHeeem Heeea
mm.m 0m.~H 00.0H 0H.m 0m.:m He.: Nw.m 00.: 0H.: w0.m gee: eeH eHeeem Heeee
:m.m mm. m.:m am.w 00.n 00.0 m0.H . ma.mH 00.H aeee 0e: eHeaem eeee
02.: m~.m m.:H m~.H mm.HH m~.m Hm.m 0w. 0m.0m H~.m keen ewe eeHdeew eeee
H:.m 0m. . - No.0 m~.mH 00.: mHm. Hw.m 0m.H keep eeH eHeeem eeed
GSCHmmH aflofimmh 03M. mam."
039an .305 .mg .mfiwdd .mfle .andd .mfle
00am 0301.5 do do do 0 390%
.wHeo we eeeen eHeeemm eeeep .z & eHand m eeeen .zkm eHede a eeeep a a
eeeeeeHz m eeHHaakammdeed eeeeeea e eeeeeeeo eeeHHHHe e eeeamaaw

 

HHMN-NHN eeree do aeeeeem

HHHNH mama

.da 00009000 002

.00900900 pdodpfik wodaeaopmma

 

 

 

m0.0 m0.~ 00.0 :0.» eeee
000 eee eeH
mo 00.0034
0H.0 0a.» NH.» Hm.~ eweeeea
0H.0 H000.0 .Hm0H. 00.0 0000.0 ozmH. 0m.~ 0000.0 maeH. 00.0 0000.0 momH. 0 eHeeem
00.0 H000.0 0:0H. 00.0 H000.0 mmeH. H0.» 0000.0 000H. 00.0 m000.0 000H. 0 eHdeem
0H.0 0000.0 000H. 00.0 0000.0 mmmH. 00.0 ~000.0 0mzH. Hm.~ 0000.0 m00H. 0 eHeeem
00.0 0000.0 000H. 00.0 0000.0 0mmH. H~.0 0000.0 000H. 00.0 0000.0 000H. 0 eHeeem
«N .m ,
00.0 00.0 0 .0 mm.» emeeeeq
0m2.
00.0 0000.0 0H0H. 00.0 0H00.0 000H. mm.m 0000.0 mweH. 00H» 0H00.0 omzH. : eHdeem
now.
00.0 0000.0 000H. 00.0 0H00.0 0mmH. mm.~ 0H00.0 000H. :m.~ 0000.0 000H. 0 eHeeem
o .m
H0.0 0000.0 mm0H. .00.~ 0H00.0 0H0H. m~.~ 0000.0 momH. mm.~ 0H00.0 000H. 0 eHaeem
0 Jam
~H.0 0000.0 000H. 00.0 0000.0 0H0H. 0H.~ 0000.0 000H. 0m.» 0000.0 0NJH. H eHeeem
. dedHa eeeeu Hedda eeeeu. Hedda eeeeux eedHa eeeeo
R pdwaoa 06000 R pd00ma madam R udwwoa macaw * pdmama wedge wamEdm
eeHee HHHa-ee0000 eeHee HHfieuee00eo eeHem HHHauHHem 00000 HHHa-HHem
copdmapuedmm dwpdedpupdodndoz cepmmuwupdom dmpwoduupdomudoz
eeHm

eHeHeeem ee eeHeeHem 0H eepHm eeeeo

bHMM mam<y

 

 

veaua— -.,

I
l-
E.

i .
l

 

M

'7612.015 P957 127457

Pringle

T612.015 P957 127457
Pringle

’A comparative study of
methods for the quantitative

estimation of cellulose,
:Lhemicellulose,&11gnin.,_V:
“ w

v ‘.
'f

“RI/'5 ’45] -0

 

l
l,__ ___” ___ __ _ _H__,_r_- _ __ 0,---”
i

 

 

 

"7177171177 11! 7717! 71777777177 F7771?!“