DOCTORAL DISSERTATION SERIES

Chevmca.1 Siuclies O f The

TITLE

h / a v y Be an (Phaseolus Yul aaris )

AUTHOR

U rnoU C. O tt
A/*

u n iv e rs ity

DEGREE

/ «

DATE
(V

/

z 1 //

n icJn aan jfa te LoHeg P
T
PUBLICATION NO.

M

Mill

IIMMI

’l

^

111111' nmi

59/

IMIM

'2

UNIVERSITY MICROFILMS
A N N ARBOR

-

MICHIGAN

/9V3

SO M E C H E M I C A L S T U D I E S OF THE
N A V Y BEAN

(P E A S E G L U S VULGARIS)

Ar nold

C. Ott

A THESIS
Submitted, to the G r a d u a t e School of M i c h i g a n
S ta te Co ll eg e of A g r i c u l t u r e and A pp lie d
S c i e n c e in parti al f u l f i l m e n t of the
r e q u i r e m e n t s for the degree of

D O C T O R OF P H I L O S O P H Y
Department

of C h e m i s t r y

1943

have a p p r e c i a t e d

the h e l p f u l a s s i s t a n c e

of P r o f e s s o r G. D.

Ball

CONTENTS
I.

Introduction

. 1

II.

Quantitative Study of Navy Bean
Seed Coats .............. ........... .

6

A. Uronic Acids

6

................... .

1. Experimental .................... 9
B. "Crude Pentosans" ..................
.........

1. Experimental
C. "True Pentosans"

...................

D. Protein .........

13
13
16
18

III. Quantitative Determination of Crude
Lipids, Unsaponifiable Fraction, and
Sterols on Ground Dhole Beans ......... 23
A. Experimental ......
1. Determination of Crude Lipids ...
2. Determination of Unsaponifiable

IV.

V.

27
27
. 28

3. Determination of Sterols ........

29

Extraction of Total Protein from
Ground V/hole Navy Beans ...............

38

A. Experimental

38

....................

Extraction of Oil and Acquisition of
UnsaponiLiable Fraction from Ground
Beans
.........

46

A. Experimental ..............

46

1. Extraction of Oil

.............

46

2. Saponification of

theOil ...... 47

3. Separation otudies on Crude
Sterols ......................... 47
(a) Chromatographic Adsorption . 48
(b) Electrophoretic Separation . 53
(c) fractional Crystallization . 62
(1) Isolation of sterols . .. 62
E. Irradiation of Crude Navy Bean
Sterols .......
C.
Summary

61

Isolation of Irobable hydrocarbons . 73
.....

74

I. Introduction.
It is quite probable that within the last two
decades the scientific work done on soybeans has far
overshadowed, in scope at least, that done on any
other legume.

Certainly there are many factors or

characteristics inherent to the soybean which readily
account for the attention they have received.
On the other hand, the common navy bean has been
studied primarily from the standpoint of nutrition,
and this to no great extent,

oomewhat aside from

the nutritional phase, is the rather extensive work
of Ritthausen and Waterman, and others^^

on the

proteins of the navy bean.
A partial impetus for this work came from a
group representative of certain navy bean growers
of Michigan.

Their desire was to obtain a practical

result in a relatively short time in order to
alleviate in part the surplus cull beans as well as
older storage beans.

The fact Is known that older

navy beans (l+-b years) are much less permeable to
water than newly grown beans.

With this in mind it

appeared plausible to study in part the chemical
composition of the seed coats.

To date, little work

has been reported on the chemical compositon of the

na v y bean

seed coats.

S c h u l z e an d P f e n n i n g e r

r e p o r t e d h e m i c e l l u l o s e s to be a l a r g e
of the

seed coats,

the c o n t e n t

r e a c h i n g a maxiurn in the r i p e
c o n s i d e r a b l e w o r k h as
p e r m e a b i l i t y of
water and

salt

se e d

s ub s t a n c e s .

bean.

coats of the c o m m o n

coats,
of

important

the m o s t p a r t

p ro p e r l y ,

the s e e d coats,

of w a t e r

compounds present
cell wall- p e c t i c

b u i l d i n g u n i t s of th e se

of the u r o n i c a c i d

or m o r e

s or pti on .

c o n t e n t of

the p o l y u r o n i d e

Pentosans,

pectic

u r o n i c ac ids.

should lead toward

c o m p o n e n t s of

be an to

(3)

the c l o s e l y r e l a t e d

s u b s t a n c e s are for
A knowledge

However,

been p r e s e n t e d on the

s ol ut io ns.

Th e

constituent

of h e m i c e l l u l o s e s

A m o n g the m o s t h y d r o p h i l i c
in p l a n t s are

(2)

the

seed

content

an e x p l a n a t i o n

one of the m o s t

the n u m e r o u s

are believed

to be c l o s e l y r e l a t e d

furthermore,

th e i r d e t e r m i n a t i o n has

hemicelluloses,

to p o l y u r o n i d e s ;
be e n

grossly

c o n f u s e d w i t h the d e t e r m i n a t i o n of p o l y u r o n i d e s .
P r o t e i n s a re al so w e l l k n o w n
water

to be f u n c t i o n a l

in

sor pt i on .
The

quantitative

unsaponifiable
was done

d e t e r m i n a t i o n of cr u d e

f r ac ti on ,

on a m i x t u r e

and

lipids,

sterols respectively

of s t r a i n s of n a v y

bean s

grown

in M i c h i g a n ;

the

seed coat

the

crude

study.

Gritnme

(5)

same lot u s e d for

rerjorted v a l u e s

li p i d and u n s a p o n i f i a b l e m a t t e r

co mm on n a v y bean.
how ev er,

a pa rt of the

A

amount

Th e

for

of s te rol s or t h e i r

in the u n s a p o n i f i a b l e
common navy

of the

su r v e y of the li te ra tur e,

d o e s no t r e v e a l a n y ii g u r e s

approximate

for

the

characterization

fr act io n.
bean c o n t a i n s ab o ut

two-thirds

as m u c h p r o t e i n as does the c o m m o n v a r i e t y of
soybeans.

In thi s w o r k

it w a s d e s i r a b l e

tota l p r o t e i n

ex t r a c t

the

ma n n e r ,

and f u r t h e r a s c e r t a i n its a d a p t a b i l i t y to

plastic

f or m a t i o n .

Complete

effected with aqueous
n e a r l y com ple te
p r ot ei n,
with

white

in the m o s t

on l y to

e x t r a c t i o n h a s be en

s o d i u m sulfi te

in color,

r e s u l t s f r o m the

Thi s p r o d u c t

s u i ta bl e for p l a s t i c

up the m a j o r p o r t i o n

alfalfa

so i s o l a t e d

that

is

st er ol s m a k e

of the u n s a p o n i f i a b l e

fraction

seed oils.

seed o i l s of

different

treatment

f or ma tio n.

It is r a t h e r w i d e l y k n o w n

The

s o l u t i o n and

p r e c i p i t a t i o n of a g l o b u l i n - t y p e

s ulf ur d i ox id e.

of v a r i o u s

efficient

ste ro ls.

the v a r i o u s l e g u m e s

The u n s a p o n i f i a b l e

seed oil has

been

contain

portion

su ow n to c o n t a i n the

of

isomeric spina sterols, vvith o-C- and

- spina sterol

predominating in amount over the & - spinasterol. ^
Medium and red clover seed oil - unsaponif iable
(loc. cit.) contain relatively large amounts of
stigmasterol.

Atigmasterol occurs also as the

major sterol, accompanied by sitosterols, in the
(7) (8)
unsaponifiable fraction of soybean oil.
As previously mentioned no work has been found
in tbe literature on the characterization of any of
the sterols in the unsaponifiable fraction of the
navy bean-oil proper.

However, Likiernik

(9)

reported

a substance over a half centurv" ago, which he
isolated from the oil of the seed coats of Phaseolus
vulgaris.

This substance he called paraphytosterol.

The properties of this conipound (n. p. 149-150°,
Lr^Jp- 44.1°) are similar to-those of

Y

- sitosterol

and it is quite probable that these substances are
the same.
In view of the facts presented, the crude sterols
were isolated from the unsaponifiable fraction of
the bean oil and some progress was made in separating
individual compounds from the crude material.
Hydrocarbons are known to occur along with
sterols in varying amounts in all seed oils.

As a

general rule, however, the amounts found are small
in comparison to the amount of sterols.

Indeed,

their isolation from most seed oils for purposes
of .a thorough investigation hardly falls within
the province of an ordinary technical examination,
evidence here exists for the presence of possibly
two distinct hydrocarbons in navy bean oil.

6

II. Quantitative Study of Navy Bean Seed Coats.
A. Uronic Acids.
Uronic acids are widely distributed
throughout the plant kingdom.

They may arise as a

result of the oxidation o f ^ -D-glucose as
represented in Aquation I.
Equation I

H-C-OH"")
HO-C-H
0
H-C-OH I
H-C
1

«=<-D-Glucose

rn i
LLJ

>

h - c -q h
HO-C-H
H-C-OH
H-CCOOH

~<-D-Glucuronic acid

(10 )
D-Glucuronic and D-galacturonic acids are the
only uronic acids whicii have been isolated from
higher plants.

The latter occurs more abundantly

than the former.
Lannuronic acid has been isolated from various
( 11 )
(1 2 )
algae by Nelson and Cretcher
and by Bird and Iiaas.
An aldobionic acid, glucose-glucuronic acid has
been shown by Heidelberger and Goebel^-^to be the
fundamental building stone of the polysaccharide
derived from Type III pneumococcus and to be an
important constituent in the piieumococcus from

* H - sugar anhydride or some other residue

Friedlanaer1s Bacillus.
The work of Norman, Bonner, Meyer, and Mark,
Bchneider, et al^'^len ds evidence for the
polygalacturonide chain structure which is believed
to make up pectin.
Lef'evre and Tollens^1'3^ first demonstrated tha
uronic acids were decarboxylated by heating with
12/a hydrochloric acid for several hours, yielding
carbon dioxide along with a pentose, the latter
yielding furfural.

This is shown in Equation II.
Equation II
H

H

v-°

'o '- °

H-C-OH
HO-C-H
HO-Q-H
H-C-OH
C00H

12io HC1

D-Galacturonic acid

h

-6-o h

HO-C-H
h -6-o h

L-Arabinose

HC

CH

IlC

C-C-H

Furfural

This idea is the basis for the improved
quantitative method for the determination of uronic
acids as proposed later by Dickson, Otterson, and
Link.^1^

Fhillips, Goss, and Browne^

offer a

still further modification of the Dickson, Otterson,

and Link method.
A second method should be mentioned here; that
developed by Tollens,

(18 )

based on the color produced

when glucuronic acid was heated with naphthoresorcinol
in the presence of hydrochloric acid,

however, the

specificity of the naphthoresorcinol method was
challenged a number of years ago by kandel and
Neuberg.

( 19 )

many of the interferring substances,

such as the lower dicarboxylic acids, occur in plants;
therefore its use in the analysis of plant materials
would, be indeed subject to criticism.
Lot to be overlooked is the fact that carbon
dioxide is liberated from sources other than uronides.
khistler, kart in, and Harris

showed that aside

from carbon dioxide coming from uronic acids, it came
from cellulosic materials ana starchy substances
shown to be free of uronides.

This fact, the difference

in rate of carbon dioxide evolution between the
rapidly reacting uronides and the slower reacting
cellulose materials, gives access to a method for
determining uronic anhydrides in the presence of
large amounts of cellulosic materials.
( 21 )
Norman
in determining.uronic acids in
polysaccharides also points out that the rate of

carbon dioxide liberation may indicate the presence
or absence of uronic acids--the rate from
polysaccharides being much slower, but yet continuous.
(22)
Bowman and me Kinnis
'in studying the pentose
and uronic acid content of orange albedo worked out
a method whereby pentoses and uronic acids may be
determined simultaneously.

This method, they point

out, is supposed to do away with emperical constants.
(1) Experimental.
In this work the uronic anhydrides
were determined by the method of Dickson et a l l ^ )
The actual set-up patterned after the above authors
is shown in Figure I.
Figure I

Apparatus Used for the Determination
of Uronic Acids

The sample to be analyzed (about 1 gm.} is
placed in the reaction flask C with 1Q0 ml. of
12p hydrochloric acid.

Connections are made and

air is slowly drawn through the two soda-1ime
towers A and the water trap B containing sulfuric
acid in order to yield dry carbon dioxide-free air.
This operation is continued for 20 minutes, when the
temperature will have reached about 100° C, using an
oil bath on an electric hot plate.

At this point any

carbon dioxide in the system or that resulting from
carbonates in the sample will have been removed..

The

carbon dioxide liberated from the polyuronides as a
result of an increase in temperature passes through
condenser D and through bottle E, which contains
silver nitrate solution and is finally adsorbed in
tube F.

Tube F contains a known amount of standard

barium hydroxide introduced through funnel G-, this
funnel being washed with carbon dioxide-free water,
heating is continued for 4-5 hours at 140° C.

The

suction is then shut off with pinch clamp 2, and the
stopcock 1 on funnel G, which is fitted with a sodalime tube, is opened to permit the solution in tower F
to run into the flask.

The excess of barium hydroxide

is then titrated with standard hydrochloric acid using

phenolphthalien as an indicator.
To evaluate the accuracy of this apparatus
samples ofc< D-galacturonic acid monohydrate
(Eastman) were used.

The results are shown in

Table I.
Table I
Authentic
No.

-D-Galacturonic Acid feonohydrate (Eastman)
Weight of
Sample (gm)

Barium hydroxide
neutralized by
CO2 (me.)

% Uronic
Acid

1

0.1512

1.130

100.3

2

0.1216

1.148

100.1

In order to analyze the seed coats they were
first loosened from the beans by placing the latter
in distilled water for 15 minutes at 85° 0.

By use

of rubber hand rollers the seed coats were removed
from the beans.

Rapid drying with warm air made

final separations of the heavy bean from the seed
coats relatively easy by means of an air blast.
The seed coats were air dried, ground and passed
through a 20-mesh sieve.

It was desired to take

the beans just as they come from the market.

The

lot contained mostly pea beans, together with some

robust and blue pod strains.
The ground seed coats were divided into four
different lots.

Lot I was left untreated; Lot II

was treated with 1> acetic acid (after a commercial)
process) for ten hours at 00° G. ph of 2.8; Lot III
was treated with 0.8/» sodium hydroxide for 10 hours
at 60° c. pH of 11.5; while Lot IV was treated with
l/o sodium bicarbonate solution for 10 hours at 60° C.
and pE of 7.8.

All these lots were washed seven times

with distilled water and then air dried.
Lloisture determinations on the. above acquired
seed coats were made by the oven-dry method in
which the samples were dried to constant weight at
105-110° C.

These figures are listed in Table II.
Table II
Moisture in Navy Bean Seed Coats

Treatment

None

Average
moisture

8.07

1)0 CHrjCGOH

8.94

l/o iSlaRCO}

11.78

0.8;o NaOE

4.55

Table III gives the values for uronic acid
content of the seed coats expressed as uronic acid

anhydride or polyuronide,

mince one molecule of

CO2 is liberated from each polyuronide residue,
according' to liquation II, the percentage of C02
times 4 gives the percentage of polyuronide.
B. "Crude Pentosans."
Ey "crude pentosans" or "total pentosans"
is meant the pentosans equivalent to the total
furfural yielded from a given sample.

Each residue

weight of polyuronide yields one molecular weight of
( 2 ^)
furfural (Equation II) . However, Norris and Resch
found only about 42c
,o of the furfural expected when
uronic acids were determined in the presence of
common plant materials.

By reference to Nrober’s

(2/l)
tables' ^ the pentosans equivalent to this corrected
value of furfural given by the polyuronide, when
deducted from the "total pentosans", gives the socalled "true pentosans".
(1) experimental»
The same four lots of seed coats
used in the polyuronide determination were also
used for the determination of the "crude pentosan"
content.

The procedure for the determination of

pentosans was essentially that published in the

14
Table III
Polyuronide Content of Navy Bean Seed Goats
No.

'.'/eight of
sample
(gfii.)

Ba(GH)2
neutralized
by CO2 (me.)

,v CO2
(air dry
basis)

c
,o Uronide
(air dry
basis)

-/o Uronide
(oven dry
basis)

Untreated
.1

1.3331

3.101

4.45

17.80

19.36

2

1.2.492

2.510

4.42

17.68

19.23

3

1.3038

2.628

4.43

17.72

19.29

4

1,5707

3.254

4.56

18.24

19.85

5

1.5066

3.1.27

4.57

18.28

19.89

6

1.5090

2.999

4.37

17.48

19.02

Treated with 1% hc2h 3o2 for 10 Hours © 60°C . pH 2.8
1

1.3289

2 .646

4.38

17.52

19.24

2

1.4317

2.865

4.40

17.60

19.33

0

1.2159

2.435

4.41

17.64

19.37

4

1.4231

2.860

4 .42

17.68

19.42

Treated with O.8/0 NaOh for 10 Hours 4 60°C, pH 11.5
1

1.1245

2.150

4.21

16.84

16.93

2

2,1170

4.080

4.24

16.96

17.05

3

1.6308

3.175

4.28

17.12

17.20

4

1.0347

1.953

4.15

16.60

16.68

0
0
0
sO
1

Treated with l/o NaHCO2 for 10 Hours

PE 7.8

1

0.8964

1.749

4.29

17-16

19.49

2

1.9124

3.685

4.24

16.96

19.23

3

0.9124

1.749

4.22

16.88

19.14

4

1.2844

2.458

4.21

16.84

19.09

methods of Analysis of the A. 0. A. C.

; steam

distillation, however, was used for the recovery
of f u r f u r a l .

This method

involves

the a s s u m p t i o n

that the pentosans consist of equal amounts of
araban and xylan.. Hockett et al.

(25)

have shown that

even if D-lyxose and D-ribose were the furfural
yielding substances, in place of arabinose and xylose,
the error introduced would not be larger than that
associated as a consequence of the difference between
arabinose and xylose.
The procedure involves the treatment of the
seed coat material with 100 ml. of 12,0 hydrochloric
acid in 300 ml. distillation flask.

The sample

being of such a quantity as to yield not more than
0.300 gm, of phloroglucide.

The flask was placed

on a wire gauze connected to a v/ater-condenser.

The

distillate obtained by steam distillation was passed
through a small filter paper into a 500 ml. erlenmeyer
flask sitting in ice.

The acid boiled off was replaced

from time to time bv means of a separatory funnel in
top of the distilling flask.

Sealed into the stem

of the separatory funnel was a tube fixed so as to
spray the acid down the neck of the distilling flask.

Distillation was continued for several hours until
a negative test for furfural with aniline
hydrochloride paper is attained.

To the distillate

containing the furfural was added enough lyi
phoroglucinol in 12)o hydrochloric acid solution to
have approximately a two-fold excess of pliloroglucinol
over the furfural expected.

The phloroglucide

precipitate was allowed to form over night before
filtering through a weighed Gooch crucible.

The

precipitate was washed with 150 ml. of water and
dried for 4 hours at 100-105° 0., then cooled and
weighed in a weighing bottle.

The increase in

weight was assumed to be due to furfural phloroglucide.
To calculate the amount of pentosans the einperieal
tables of Krober were used.

The results are given

in Table IV.
C. "True lentosans".
As pointed out previously, the "true
pentosans” were calculated from the "total pentosans"
by correction for the furfural liberated from the
polyuronides.

For example, the percentage of

polyuronide in the untreated seed coats is 19.394>;
the furfural equivalent is 10.56c
,4; the corrected

17
T a b l e IV
Pentosan
No.

C o n t e n t of N a v y B e a n S e e d Go a t s

lveight of
sa m pl e

W e i g h t of
phloroglucide

(gru.)

(gm.)

Pentosans
( gm. )

°/o C r u d e
Pentosans
(a i r d r y
basis)

% Crude
Pentosans
(o v en d r y
basis)

Untreated

1

0.7080

0.2058

0.1867

26.4

28.7

2

0.7706

0.2314

0.2092

26.9

29.3

3

0.7674

0 .22.68

0.2053

26.7

29.1

Treated with Ip H C 2H 3Q 2 for 10 Hours @ 60°C. pH 2.8
1

0.9639

0.2872

0.2581

26.8

29.4

2

0.9440

0.2740

0.2465

26.2

28.8

3

1.0878

0.3303

0.2960

27.2

29.9

4

0.6476

0.1900

0.1729

26.7

29.3

Treated with 0.8% NaOH for 10 Hours © 60°C. pH 11.5
x

0

0.2080

0.1887

24.9

26.1

2

0.7213

0.1838

0.1676

23.3

24.4

3

0.6213

0.1697

0.1551

24.9

26.1

4

0.5357

0.1454

0.1355

25.3

26.5

Treated with Ip NaHCO-j for 10 Hours © 60°C. pH 7.8
1

0.8465

0.2417

0.2182

25.8

29.2

2

0.9446

0.2727

0.2454

25.9

29.4

3

0.5576

0.1597

0.1464

25.5

28.9

4

0.6422

0.1809

0.1651

25.8

29.2

value 4«44#>, which in turn is equivalent, according
to Ivrober’s tables, to 7,61-> pentosans.

This latter

figure deducted from the "total pentosans" gives
the "true pentosans".

These results are given in

Table V.
D. Protein.
The third factor of importance examined
in water sorbability, namely, the protein^,^ was
determined by a semi-micro Kjeldahl method for
nitrogen on each treated portion of the navy bean
seed coats.

For comparison the factor 6.25 was

used to convert nitrogen into protein.
are shown in Table VI.

The results

Table V
"True Pentosans”
(1 )

t 2)

Type of
Treatment

Average
"total
pentosans”
(dry basis)

• (3)

(4)

G
/o Pentosans
Furfural
equivalent equivalent
to uronide to uronideper 100 gm.
furfural
(dry basis)

(5)
% ”true
pentosans”

None

29.03

4.44

7.01

21.42

1-/6 CK3COOH

29.35

4.43

7.60

21.75

0.8/0 NaOH

25.78

3.86

6 .t>6

19.12

1> NaHCO^

29.20

4.41

7.56

21.64

Table VI
Percentage Protein

Treatment

Weight

of

sample
(gnu)

1.

2.

3.

4.

in S e e d C o a t s

> Nitrogen

(oven dry
basis)

Protein
(o v e n d r y
basis)

:/o

1.5350

0.734

4.90

"

1.3543

0.826

5.16

"

1.3676

0.825

5.16

CH^CQOK

1.6326

0,815

5.09

"

1.2042

0.822

5.13

1.2389

0.489

3.06

"

1.2911

0.475

3.05

"

1.2575

0.472

2.95

KaHC03

1.1515

0.739

4.62

"

1.4519

0,716

4.78

»

1.4093

0.758

4 •74

None

NaOH

Discussion
The total combined content of polyuronide
and '‘true pentosans” of the untreated seed coats
is in the neighborhood of 40>.
(2)

Pfenninger'

bchulze and

found a maximum of 48.65;
d of hemicellulose

in the ripened bean seed coats.

The uronide content

of the bean coats compares favorably with many of
the higher yielding hemicellulose preparations from
(2t))
various plants discussed by Norman.
A comparison of the moisture content of the
four lots of seed coats gives an indication of the
importance of the uronide and pentosans in the
water retention of the dried coats.

Treatment wdth

l/o acetic acid gave seed coats of slightly increased
moisture content, when air-dried, as compared with
the air-dried, untreated coats.

The sodium

bicarbonate treated samples showed the highest
moisture content, 1 1 .78/0, while the sodium hydroxide
treated samples gave the lowest moisture content,
namely 4.55>.

The sodium hydroxide samples also

showed greatest loss of uronide,
and protein.

"true pentosans”,

The sodium bicarbonate treatment

gave a small definite loss of protein and no

appreciable loss ot uronide or ’’true pentosan".

If

the water retention was correlated with the protein
content, one would expect a decrease in the water
retention of these bicarbonate treated samples; the
opposite condition, however, was observed.

III. Quantitative Determination of Crude Lipids,
Unsaponifiable Fraction, and bterols on
Ground Whole Beans
As pointed out before, Grimms

( 5)

reported the

value of the crude lipid content of the whole navy
bean as 1.32a
/o.

This crude lipid had a saponification

number of 189.2, iodine number of 135.7, fatty acids
87.51/0, and unsaponif iable matter 5-85
Alternate extractions of the navy bean flour
with ethyl alcohol and ethyl ether effected the
means of obtaining the total lipids.
originally rjroposed by Koch

This procedure

(27)
' and subsequently

used in a modified form by Kumagawa and Suto,
Sandol*''^ and D ill ^O)

(2 8 )

ps reported to extract a

maximum amount of liptid material.

It is well to

keep in mind in using this method that sterols often
occur in combination with sugars as characteristically
insoluble glycosides or phytosterol ins,^^ ^ ^ ^
which as a consequence may be only partially extracted
by the above procedure.

A procedure for complete

extraction of free and combined sterols should combine
acid and alkaline hydrolysis of the whole plant
tissue as preliminary steps to the extraction
procedure,

however,

such a method would be even

more involvedj furthermore, the amount of glycoside

present in most plant tissues is very small compared
to the total lipids.

Also to be noted is the fact

(35)
that the. calcium salt of phosphatidie' ' acid is
soluble in ether.
The separation of sterols from compounds of
similar solubility involves as the first step, the
separation of the glycerides which make up the bulk
of any lirjid extract.

This separation is accomplished

by saponification and extraction of the unsaponifiable
fraction with suitable solvents.

The success of this

separation offers only technical difficulties in
separating free sterols and those originally present
as esters.

Glycosides or phytosterolins are resistant

to ordinary alkaline hydrolyses, thus these combined
sterols would not be extractable to any degree.
After the unsaponifiable fraction has been freed
of glycerides, there are still present substances,
such as hydrocarbons, carotenoid pigments, aliphatic
alcohols and cyclic alcohols.

The separation of

sterols from these compounds is classically accomplished
by precipitation with

d i g i t o n i n .

^ T h e

use of

digitonin for the quantitative determination or
separation of plant sterol mixtures is subject to
criticism since sterol digitonides vary in solubility,

and, in some cases the digitonides may actually be
( 37)
more soluble than the sterol itself.
Also,
there is known a naturally occurring sterol, viz.,
(38)
calosterol
from the juice of Calotropis G-igantea,
which does not form a precipitate with digitonin.
This means the hydroxyl group at
the methyl group at C^q .

must be trans to

Furthermore, various lots

of digitonin may have variable amounts of companion
( 39)
saponins present. 1 Then, too, digitonin preparations
precipitate with certain alcohols other than sterols.^
Consequently, any use of digitonin as a quantitative
precipitating agent for plant sterols must be
established on a strictly emperical basis.
Numerous methods exist for the determination of
the main animal sterol (cholesterol).
In 19-10 Windaus^"^

suggested the precipitation

of cholesterol in alcoholic solution with an
alcoholic solution of digitonin with the subsequent
gravimetric determination of the cholesterol
digitonide, in which the chlesterol was precipitated
with digitonin and the digitonide determined by a
variation of the Lieberman-Burchard reaction.
Bernoulli^ 3 )

determined cholesterol by measuring

the red color developed when anhydrous zinc chloride

and acetyl chloride were added.

Bzent-Gyorgyi

(44)

used an oxidation method to determine micro amounts
of cholesterol digitonide, while Sobel, Drekter and
Natelson^^

determined cholesterol by precipitation

with pyridine sulfur trioxide.
All these methods are subject to a great deal
of criticism since each of them gives variable results.
Thornton* has examined the various color reactions
for a suitable method of determination of soybean
sterols.

His work indicated that most of the methods

now in use for determination of cholesterol are of
little value in determination of crude soybean sterols.
(L.3)
He made an extensive study of the Bernoulli reaction
and found that there was a characteristic transmission
maxima at 440 mu. for various sterols, but that the
extinction coefficient was different for the various
individual sterols.

The Bernoulli reaction

(43)

is

further complicated by the fact that it gives color
with carotenoid pigments and related substances and
hence may be used only on pure sterol solutions.
The method utilized in this work is a
modification of the Behoenheimer and Sperry

(42)

method.

This modification was developed as part of a
fellowship, held by L. C„ King and subsequently by

* Private communication

the author, granted by the Upjohn Farm Relief
Association.
A. Experimental.
(1)

Determination of crude lipids--Three

25 gm. samples of 80-mesh navy bean flour were each
transferred to 33 x 94 mm. paper thimbles.

Each

sample was covered with a cup-shaped filter paper,
made by pressing a filter paper over the open end of
a large test tube.

The thimbles were then placed

into continuous Doxhlet extractors.

Each thimble

was kept off the bottom of the chamber by means of
a 2 cm. length x S mm. glass tubing.

After 10 hours

of continuous extraction with 9 5?° ethanol, an 8-hour
extraction with dry ethyl ether was made,

following

this, a final 10-hour extraction was made with 95/0
ethanol.

These combined extracts were collected

into weighed 250 ml. beakers, and evaporated on a
steam bath, over which blew a swift current of air
from an electric fan.

This air current prevented

solvent-creeping, over-heating, and aided evaporation,
■dien all the solvents appeared to have evaporated
there was added to each beaker, with rotation 20 ml.
of absolute ethanol, the contents warmed and mixed
well.

Experience has shown these last steps to be

helpful in removing the last traces of water.
After drying the extracts, they were weighed to
constant weight (usually 2-3 weighings).

These

values so obtained are the total or crude lipids
and are shown in Table VIII.
(2)

Determination of Unsaponifiable-~To

the crude lipid in the weighed beakers were added
about 25 ml. of 95?o ethanol and 3-4 ml. of 10 N
sodium hydroxide.

The beaker and contents were

heated on the steam bath, with occasional rotation,
for 1 hour.

The cover glasses were removed and the

sides of the beakers washed down with a stream of
hot 10yo ethanol-water solution.

Heating was

continued for an additional 30 minutes.

The mixture

was transferred to 230 ml. separatory funnel with
the aid of hot water.

The beakers were rinsed with

about 50 ml. of "Skellysolve B ” and this then added
to the cooled mixture.

The soap solution was

extracted 5-6 times with ’’Skellysolve B ” using a
triangular funnel arrangement.

The stubborn emulsions

were broken by the addition of 10-12 ml. of 95yo
ethanol.

The combined petroleum ether extracts were

washed 3 times with 50 ml. portions of 10>0 ethanolwater solution.

The extracts containing the

unsaponifiable were returned, to dried 250 ml. beakers
and evaporated to dryness.

To each beaker were added

70-80 ml. of "Bkellysolve B".

The contents were

warmed, then filtered into dry, tared 150 ml. beakers,
and washed with warm "Skellysolve".

This step removed

some glycosides and other foreign material.

The

solvent was evaporated as before on a steam bath
using a moving air current.

The dried beakers and

contents were weighed to constant weight.

The weight

so obtained gives the value for the unsaponifiable
matter, the results of which are also listed in
Table VIII.
(3) Determination of Sterols--To each of
the beakers containing the unsaponifiable matter
were added 50 m l .'of 95> ethanol.

Cover glasses

were placed on the beakers and the latter were
warmed on the steam bath with occasional swirling.
The contents of the beakers were then quantitatively
transfered to 200 ml. volumetric flasks.
hor the determination of the sterols a 5 hi.
aliquot was taken from each flask and pipetted into
a 15 ml. centrifuge tube.

To each tube were added

2.5 ml. of digitonin solution (1 gm. in 1000 ml.
of water and the solution evaporated to 500 ml.}.

The tubes were warmed gently on the steam bath to
aid flocculation of the resulting digitonide.
After 1-hour of warming the tubes and contents
were allowed to stand over-night.

The digitonides

were concentrated under centrifugation for 15 minutes,
after which time the supernatant liquid was almost
completely separated from the settled digitonides.
This separation was accomplished by means of a
rubber bulb pipette.

The settled digitonides were

washed with 2 ml. of 10> ethanol-water solution and
re-centrifuged.

After again removing nearly all of

the wash solution above the digitonide, the latter
was dried at 80-90° G. for 1 hour.

If in certain

cases small amounts of digitonide adhered to the
side of the tube, it could be washed down with a
fine stream of hot glacial acetic acid or cold pyridine.
Following this the tubes were finally dried in the
vacuum oven for 1 hour at 55° and 25 inches of mercury.
Two ml. of dry glacial acetic acid were then added
and the loosely corked tubes heated on the steam bath
until all of the digitonides had gone into solution.
When the tubes had cooled to room temperature 2 ml.
of C. P. acetic anhydride were added in such a way
as to run down the sides of the tube.

The tubes

were corked, set in the refrigerator (8° G.) for
15 minutes, 0.1 ml. of concentrated G. P. sulfuric
acid carefully added to each tube, and the contents
then thoroughly mixed.

Immediately, the rack of

tubes was placed in a 45° oven for exactly 1 hour
after which time they were transferred to an 8°
refrigerator.

After 15 minutes the amount of light

in the region of 650 mu, transmitted by the reddishbrown sterol solution was measured, using a Fisher
"Electrophotometer".

The actual values for the amount

of sterols were read from a standard curve made up
from, known .amounts of crude soybean sterols.

These

values are also given in Table VIII.
The standard solutions were made from a stock
solution consisting of 225 mg. of recrystallized
crude soybean sterols dissolved in 95/“ ethanol made
up to 500 ml. at 25° C.

From this stock solution

proper aliquots were pipetted into seven 200 ml.
volumetric flasks so that when made up to the mark
with 95/o ethanol the flasks contained 10 mg,,
20 mg., 25 mg., 30 mg. > 35 mg., 40 rug., and 50 mg.
of crude soybean sterols respectively.

Prom each

one of these flasks containing the standard solutions
were taken a 5 ml. aliquot and treated exactly as

outlined above in the sterol determination.

The

resulting standard curve with the percent transmission
plotted against' the mgs. of crude soybean sterols
using a o$0 mu. filter in the 1'isher "Electrophotometer"
is shown in -figure II.

The readings, from which the

curve was plotted, are given in Table VII.

Table VII
Transmission Values for Standard Curve

Solution
Humber

Wt. of Crude
Soybean
Transmission at 1 Hour and 45°C. Filter 650 mu.
Sterol per
200 m l . of
sample
Set I Set II Set III Set IV Set V Set VI
Average

1

10 mg.

83.2

85.5

85.0

85.4

84.3

84.5

84.7

2

20 ”

70.1

72.6

72.6

-----

71.6

73.6

72.1

3

25 "

----

65.6

65.6

65.2

65.4

63.6

65.1

4

30 "

----

59.3

59.3

58.9

60.5

61.0

59.8

5

35 "

----

55.3

55.3

55.0

54.6

55.1

6

40 "

50.0

50.0

47.8

48.3

48.7

49.0

7

50 "

38.3

38.4

39.8

41.2

41.1

39.5

38.2

Standardization

Curve ,1 wour ,4 5 °c.

F igure

o o

by)

20

Fn,Te* SSOnw

30

40

50

60

TO

60

90

100

PERCENT TRANSMISSION
V*>

Table VIII
Crude Lipid , Unsaponifiable, and Sterols on Navy Lean Flour
(25.0 gm. samples)
Sample
No.

{jo moisture on oven dry basis * 8 .29)

Crude 70 Crude 70 Crude
UnsaponiLipid Lipid on Lipid on
fiable
fraction
(gin.) basis of basis of
seed
seed
(m g .)
(air dry (dry basis)
basis

% Unsaponi
fiable
on basis
of seed
(air dry
basis)

p Unsaponi­ 'P Trans­
Crude
jo Sterols
fiable
sterols on basis of
mission
on basis of at 1 Lour,
per
unsaponicrude lipid
200 cc.
fiable
45°C.
(mg.}
filter
650. rau.

1

0.6642

2.66

2.90

33.4

0.15

5.78

67.0

23.2

60.4

2

0.6572

2.63

2.87

33.7

0.15

5.83

67.7

22.7

58.7

3

0.6676

2.67

2.91

38.6

0.15

5.89

68.0

22.4

58.1

VjO
VTl

36

Discussion
The values obtained for the crude lipid agree
reasonably well considering the fact that the crude
extract is handled and weighed in relatively large
containers.
The results obtained, for the unsaponif iable
show better- agreement than is normally attained
on samples containing very large amounts of fatty
acids and a'small amount of unsaponifiable matter.
A reasonable agreement among such a set of values
lends evidence for the fact that 5 or 6 1-minute
extractions of the soaps with "ohellysolve" is
sufficient to remove the un sap on if ia ble matter.
The use of a small container for the-final weighing
of the un sap on i f ia b1e is also desirable.
f he va1ues for tne cruu e sterols cornpare
reasonably v.ell with tne figure obtained rrom the
large-lot extraction, subsequently to be reported,
j-r.e justification for the use of the standard, made
from crude soybean sterols, on tne determination of
the cruae sterols in' navy beans lies in the fact
that the sterol fractions of these two legumes are
ouite. similar.
The characteristics of the light transmission
of the reddish-brown colored solution used in the

37

sterol determination shows no characteristic absorption
maxima.

Consequently, no particular part of the

curve is ideally adopted for a color determination,
however, a portion was selected, which was conveniently
covered by a standard rilter (red filter o50 mu.)
supplied with the fisher "hlectrophotometer” , and
which gave a reasonable dispersion of transmission
values as a function of concentration of the sterol.
hithout doubt, much is to be desired in this
method.

However, if extreme care is taken to avoid

moisture-contamination of the dried, digitonide and
adherence to ''exact” duplication of steps in both
standard and. unknown samples, reasonable comparative
results can be obtained by an individual worker.

38

IV. Extraction of' Total Protein from Ground Whole
Navy Beans
A . Exp er inien t a 1.
Several solutions were tried for the
purpose of extracting the total protein from the
ground, air dry beans.

The total protein of the

bean flour calculated from the total nitrogen was
about 26/a.

In all cases 10 gru. of the SO-90 mesh

flour together with 150 ml. of extracting solution
were used in centrifuge bottles (250 ml.).

These

mixtures were allowed to stand for 1+ hours with frequent
stirring.

They were then centrifuged at 2000 r. p. m.

for 10 minutes, and 50 ml. portions of the clear
supernatant liquid were pipetted into Ejeldahl flasks
and the protein content determined,

uesults in

Table IX are given as protein (N x 6.25).
Table IX
Effectiveness of Extracting Solutions on
Navy Bean Protein
Extracting Solution

Average % Protein
Air dry Basis

1

Distilled water

10.7

2

0 . 4/e N a O H

22.7

3

IO.O53 NaCl

22.5

4

0.2/3 NaOII

22.3

5

0.55& N a oS0.

j

24.3

The use of sodium sulfite solution for
extracting the protein from soybeans is mentioned
by H o r v a t h . ^ 6 *
Aside from the results of Table IX favoring
0.5/0 sodium sulfite solution for extraction, there
is also obtained after precipitation with sulfur
dioxide, a more highly desirable product with regards
to plasticity and color.
In using the sodium sulfite solution for the
protein extraction, 66 gm. of 80-mesh bean flour
were added to one liter of 0.5% sodium sulfite
solution and this mixture was stirred with an
electric stirrer for 2g hours.
allowed to settle out.

The particles were

The decanted solution was

passed through the Oharpless super-centrifuge in
which the revolving bowl was lined with a celluloid
sheet to remove any particles not in solution.

In

order to precipitate the protein 100 ml. aliquots
were placed in 250 ml. centrifuge bottles, dilute
sodium hydroxide added until a slight amber color
appeared and then 50,5 sulfuric acid until precipitation
just started.

To another aliquot there was added

sodium hydroxide as above and sulfur dioxide jjassed
in until precipitation appeared complete.

Another

40

aliquot was treated with sodium hydroxide and precipitation
made with 50,o acetic acid; the same result was
accomplished with another aliquot with phosphoric acid.
All these solutions were centrifuged for fifteen
minutes and Kjeldahl determinations run on the mother
liquors.

The mother liquor from the aliquot treated

with sodium hydroxide and sulfur dioxide was clearest
in appearance and contained the least protein.

A

lijeldahl determination on a 50 ml. aliquot resulted
in a figure for the protein left in solution of
0.4 gm. per 100 ml. of supernatant liquid.

The sulfite

mother liquor previously contained 1.6 gm. of protein
per 100 ml., or in other words, 75^ of the protein was
precipitated.
In order to get a more definite picture of the
corresponding pK of the solution when maximum
precipitation was effected by the SO2 treatment, a
continuous stream of SO2 was passed into a solution
of protein extracted with sulfite and made alkaline
with 5 molar sodium hydroxide at pH of 11.5.
were removed,

Aliquots

then allowed to stand and the relative

precipitation noted.

The results are shown in Table X.

41

Table X
Corresponding; to max jura Precipitation
Time______
0

pH______

Result

min o

11.5

IS ruin.

11.3

X

23 min.

10.2

X

25 m i n .

6.3

23 m i n .

6.0

No precipitate

X X
X X X X X X

(clearest separation)
30 ruin.

4.9

X X X X X X

(mother liquor cloudy)
33 m i n .
35 mi n .

4.10

X X X X X X

(very fine ppt.)
2.00

X X X

X X X

(very fine ppt.)

The best textureb precipitate was obtained
between a pH of 5 and 5 (about 0.3-0.5‘
h with respect
to bO^).

It was later found better to precipitate

from a solution made slightly less basic with NaOH,
namely pH 10-11.
The protein obtained is a heavy, white, glossy
semi-solid which had very good plastic properties.
V.hen rubbed between the thumb and fore-finger the
protein became very smooth, transparent and dried
to a fine film.

A few attempts for plastic formation were
made with this protein,

For this purpose a special

brass mold was used through which live-steam could
be passed (90-95° C.).
Number 1.
Base

---------

Glutinizer

protein
lactic acid

Indurating agent - 20'p formaldehyde
Pressure — ■-— ---- 2,000 lb. / sq. in
T i m e ---------

—

15 minutes

T e m p e r a t u r e ------90° G.
The protein, of about 40><? moisture, and lactic
acid were kneaded in a mortar until the mixture
became tacky or rubbery, a little (1-1 formalin)
was then added and the mixture placed in a mold.
It was immediately pressed and held for 15 minutes.
The product after heating in a 50° oven for
1 week was araber in color, hard and brittle.
Number 2.
B a s e ------------- protein
Glutinizer

-- - lactic acid

Indurating agent - 30fb formaldehyde
P r e s s u r e --------- 4,000 lb. / sq. in,
T i m e ------------- 15 minutes
T e m p e r a t u r e ------90° C.

Tile protein was worked up as in ntl.

The product

had a somewhat grainy appearance, was rubbery, and
brown in color.
1m U L J . b e r

3

.
B a s e ------------- protein
G l u t i n i z e r ------— ortho phosphoric acid
Indurating agent - 30> formaldehyde
P r e s s u r e --------- 4,000 lb. / sq. in.
Time

------------ 13 minutes

T e m p e r a t u r e ------93° C.
The product had a high elasticity, i.e., bent
and snapped back quickly into original position,
higher pressure gave a clearer product.
Dumber 4.
Base ---------- ---- protein
Glutinizer

109> haOTI

Indurating agejit - 40/« formaldehyde
I r e s s u r e -------- - 2,000 lb. / sq. in.
Temperature ---- -- 90-95° C.
Time

------ ------- 15 minutes.

The protein was kneaded in a mortar with NaOH
until it became amber colored and rubber-like.

It

was then immersed in formalin, kneaded, and pressed
in the mold.

The product was clear, amber colored,

hard and brittle.
An inferior type of thermosetting plastic was
made from the ground beans without extracting any
of the protein.

To 15 gm. of 80-mesh navy bean

flour in a 500 ml. 3-necked flask was added 0.6 mole
of redistilled phenol, and 0.5 mole of trioxmethylene.
The mixture was stirred tor three hours at 80° C.
with an electric stirrer, during which time the
mixture became darker and more turbid.
transferred to a large watch gla S
oven at 50° C. for two weeks.

S

ciild

It was
placed in an

There resulted an

amber colored, glassy mass containing dispersed
particles.

The latter were probably due to insoluble

carbohydrate material and not protein since globulintype proteins are soluble in hot phenol.
A much better product was obtained when 15 gm.
of 80-mesh bean flour was mixed with a minimum of
water, ptyalin added, and the mixture then agitated
in a Waring blender.
for three hours.

This mixture was allowed to stand

It then was made slightly acid with

3P acetic acid and 0.3 mole of trioxmethylene was
added.

After removal from the blender the preparation

was heated to 80° C., 2 grn. of phenylisocyanate added,
and the maroon colored product poured on a watch glass.

45

After five days at 50 C. there resulted a hard, glossy,
thermosetting plastic with quite a homogeneous texture,
relatively insoluble in water.
Discussion
Navy bean proteins consist chiefly of the
(u n )

globulin type'^M

proteins which are soluble in

dilute neutral salt solutions.

According to Osborne^'^

about yto of the total navy bean proteins are alkaliinsoluble and nearly salt-insoluble, which may account
for the amount unextracted by the 0.5/° sodium sulfite
solution.
Since proteins are easily hydrolyzed by strong
alkaline solutions, thereby affecting unfavorably
the plasticity of the products obtained therefrom,
it is well to employ weak alkaline solutions having
reducing prox-jerties, such as a 0.y,o solution of
sodium sulfite.
bulfurous acid (SO2 ) has a triple advantage of
producing a precipitate of good plastic quality, of
bleaching the product, and of preventing its oxidation
by air.

V. Extraction of Oil and Acquisiton of Unsaponifiable
fraction from Ground Wavy Beans
As far as can be ascertained there is no other
account given for the extraction of the total lipids
{*•>)
from navy beans a,side from that of Grimme.
As
(9 )
pointed out on page 4, Likiernik
appeared to be
the only worker to have isolated any sterols from
any part of the navy bean.

Ke investigated only the

unsaponifiable fraction of the oil obtained from the
seed coats.
A. Exnerimental.
(1) Extraction, of oil— forty kilograms of
navy beans were finely ground in a hammer mill and
extracted six times with cold ethyl ether.

The

extraction was accomplished by the use of a 20-gallon
galvanized garbage can having a brass faucet located
about 2 inches from the bottom.

The flour was held

in a cloth sack which was fastened to rings inside
and near the top of the can.

The sack rested on two

test tube racks.
The extract was filtered and the last traces of
solvent distilled off in the presence of an atmosphere
of carbon dioxide.

The yield was 0.83 kilograms of

yellow oil, which represented 1.9/° of the original bean.

entire lot of oil, contained in a 12 liter round
bottom flask, were added 4 liters of ethyl alcohol
and 1 liter of 50> KCH (in hoO).

This mixture was

stirred on the steam bath for two hours,

following

saponification 4 liters of water were added and the
unsaponifiable was exhaustively extracted from the
soaxzs with "hkellysolve B".

The extracts were

concentrated to about 505*, washed free of alkali,
and the remainder of the solvent distilled off.

'The

yield was 48.7 gm. of a yellow colored serai-solid,
which represented 5*9h> unsaponifiable on basis of
crude lipid.
(3) Separation studies on crude sterols—
The entire amount of unsaponifiable matter was taken
up in 500 -tril. of x,etr°leuin ether and steam was jmssed
into the solution until near saturation occurred.
The solution was allowed to stand over-night.

A mass

of beautiful white crystals separated; yield of first
crop 22.5 gum; m.. p. 138-144°;
2.1 ml. chloroform, 1 = 1
reading).

dm.,

J -44° (35.4 mg. ,
-0.742°, average

The Liebermann-Burchard reaction was

distinctly positive.

The mother liquor was evaporated

to 300 ml. ana allowed to stand for 24 hours.

A

second crop of fine crystals was obtained; yield 7*2 g m. ;
n

m. p. 128-131 ;
1 = 1 dm.,

g"

r

^^

L ^ J j) -33° (38.6 mg., 2.1 ml. chloroform,
-0.698 average reading).

Again the

mother liquor was concentrated to 100 ml. and set aside
over-night.

A third fraction of soft, waxy-lifce

material was obtained; yield about 2 gm.

The total

amount of crude sterols acquired amounted to about
29 to 30 gms. or 59 to 02;0 sterols on basis of
unsaponifiable.
(a) Chromatographic adsorption of
crude sterol derivatives.
An attempt was made to separate the navy bean
sterols by chromatographic adsorption of the colored
esters

(19)
'' of the crude sterols.

The colored esters

v/ere formed by esterifying the crude sterols with the
colored azobenzene-p-carboxyl chloride.

The azoyl

chloride was prepared from the azobenzenemoncarboxylic
acid and thionyl chloride in the presence of anhydrous
of sodium carbonate.
Ireparation of azobenzenemoncarboxylic acid-This was accomplished by the method of A T w a y . ^ 0 ^
lara-nitrobenzoic acid (1 mole) was dissolved in
ethanol and mixed with about b mole of glacial acetic
acid.

To this solution with agitation were added

2 moles of zinc dust in small portions.

The

temperature of the mixture was kept between 5-10° C.
The cold yellow solution of p-hydroxylamino-benzoic
acid was poured into an aqueous 10c
p ferric chloride
solution and the mixture warmed to 45° and held at
this temperature for 1 hour.

The solid material was

filtered and washed well with water.

This solid

material was then extracted with warm ethanol.

The

p-nitrosobenzoic acid dissolved while a gray insoluble
substance remained.

from the dark green filtrate the

p-nitrosobenzoic acid separated upon evaporation of
the solvent.

This crude product was dissolved in a

minimum of warm glacial acid and to this were added
25 gm. of redistilled aniline.

The mixture is

stirred while heating for about three hours, after
which time a brown substance separated.

This was

filtered off, washed writh water, dissolved in alcohol,
horite added, and again filtered.
the crude brown acid separated.

Upon evaporation
following three

recrystallizations bright red crystals were obtained;
yield 10 g m ., m. p. 240-242° 0.
Preparation of the acid chloride of
azobenzenemonocarboxylic acid--To 2.7 gin. of the dry

50

colored acid were added 7 gm. of anhydrous sodium
carbonate and 33.8 ml. of thionyl chloride.

This

mixture was refluxed for lg hours and the product
taken up in low boiling petroleum ether.

The

sodium carbonate was removed by filtration and the
filtrate was then evaporated to dryness on the steam
bath.

kecrystallization of the residue from petroleum

ether at 0° gave bright red crystals which melted
at 93-94°.
Preparation of sterol esters of
azobenzenernonocarboxylic— Seven-tenths of a gram of
crude navy bean sterols (dried in a vacuum ■" 100°
for two hours) and G.o gm. of azobenzenernonocarboxyl
chloride were dissolved in 20 ml. of dry pyridine
contained in a 50 ml. glass-stoppered flask.

The

solution was heated on the steam bath for l| hours.
Water was added and the crude steryl azobenzenemonocarboxylate
esters, filtered, and washed.

The dried product was

recrystallized from absolute ethanol.

The violet

colored crystals melted at 133-135° C.
Chromatography of the esters--The adsorption
tube used was 2 cm. in diameter and 80 cm. in length
with a 10 cm. capillary attached to one end.

A small

pack of cotton, followed by 3 cm. of sand, and this
in turn followed by small portions of activated
anhydrous aluminum oxide (80-mesh).

Each portion

was packed quite firmly by tapping on the side of
the tube while slight suction was applied.

The

column was treated with 1-1 petroleum ether and
benzene until there was left about 1 cm. of solvent
over the alumina.

Two-tenths of a gram, of the

above described steryl esters were dissolved in
15 ml. of C. r, benzene and the resultant, deep red
solution poured into the tube and the suction turned
off.

A few milliliters of benzene were added to

wash down the last traces of the mixture.

A liter

separatory funnel was then stoppered tightly into
the tube, and pure high-boiling petroleum ether was
allowed to drop onto the column at the same rate as
the solvent was dropping from the capillary (10-20
dro£>s per minute).

The petroleum ether was added

only after the colored compounds had disappeared
from the top of the column.

The adsorbent was

never allowed to become dry during the experiment.
Following the addition of about .50 mis. of
petroleum ether a violet colored zone formed,

52

1+ cm, in length about 12 cm. from the top of the
column,

bipon washing the column with lg liters

of petroleum ether,

the colored zone had moved

down about 20 cms. more without any development of
a distinct chromatogram.

The zone was about 3 cm.

in length with a distribution of violet color toward
the top of the zone diffusing into a reddish-brown
near the bottom.

A further attempt to develop a

chromatogram was made by washing the column with
500 ml. of 1-1 petroleum ether and benzene.

This

effected nothing except to carry the colored zone
down beyond the-middle of the column.

A still

further attempt was made by 'washing the column with
300 ml. of 1-3-3 absolute ethanol, petroleum ether,
and benzene.

The violet and the redoish-browa

colored parts still persisted in staying together.

Discussion
■ ■ ,
raaenburg
et al (4Q' ) have shown that mixtures
of the azobenzenemonocarboxylic esters of the
following sterols were separated on adsorption
columns; cholesterol from stigmasterol, stigmasterol
from ergosterol, and cholesterol from ergosterol.
Cholesterol, stigmasterol, and ergosterol, as a
mixture, were each separated from one another.
However,

the ester o£~/£ -sitosterol could not be

separated to any degree of efficiency from
cholesterol,

stigmasterol or ergosterol.

It is now

conceded that the separability of sterols on alumina
columns depends primarily upon differences between
the number of double bonds in the molecule, and not
upon the position of the double bonds or the
structure of the aide chain.
from the results of the present work one would
conclude that the individual sterols in the navy
bean sterol mixture are of the same general type.
(b) Electrop)horetic separation of
(51)
crude navy bean sterols— According to wo ye r w '
particles of suspensions prepared by grinding
crystals of purified cholesterol with ice at -10°

54

show a remarkably uniform electrophoretic mobility.
xn fact, the mobility-pH curve of cholesterol (in
_4_ acetate buffers) suspension prepared as above is
150
in shape quite similar to that of a protein with an
isoelectric point at pH of 3.2.
Trial preparations of buffer mixtures— In this
work it was desired to have a buffer system which
would maintain the crude sterols in solution at

0.5° 0. and also have at this temperature a specific
cone! uotivi ty of a tout 10“^ m h o s .
Suffer 1.
nth an ol ( 9 5 w ) --------- --- 70 nils.
.lyricine (redis ti l le d )
acetic acid (glacial)

30 mis.

---- 30 mis.

hod iurn acetate. ------0.8 g m .
....
....
..-f
.o
ope on. a c o o n a u c ui v ny ---- p.ub x 10" H 0
P -

------ --------- - 4.7

In buffer y1 a 0.2p solution of crude navy
o
bean sterols crystallized within 24 hours ., 0 C*

55

buffer 2.
Ethanol (95/*)

---------- 60 mis.

lyridine (redistilled) --- 30 mis.
acetic acid (glacial)
-----

Chloroform

— -- 10 mis.
—

0.5 ml.

mpecii'ic conductivity -— -- 0.22 x 10 '

0

o

C

--------------- 4.4
In buffer/r2 a 0.2m solution of crude navy bean
sterols gave no crystal!j zation within 24 hours z 8° C.

Buffer 3®
Ethanol (957*) ------------ t>0 nils.
Pyridine (redistilled)
Acetic acid (glacial)
Chloroform

n th an o 1 am in e

__ _______
—.

--- .10 mis.

0.1 g m .
.-- _ 16 m l .

Specific c o n d u c t i v i t y
—

30 mis.

- -------- - 0 .5 m l .

modi urn a c e t a t e

-u

--

0.79 x 10“ ^ _ 0° C.

_____,.____ __ yo• J

In buffer y3 a 0.2 ; o solution of crude navy bean
sterols nave no crystallization within 24 hours z 0° C.

56

Buffer 4
hthanol (95L
h)

60 mis

lyridine (redistilled)

30 mis

Acetic acid (glacial)

---- 10 mis

Chloroform

0.5 ml

tiodium acetate

0.1 gm

NH-j bubbled through for 3 minutes
8.7
In buffer 44 a 0.2/o solution of crude navy bean

These experiments showed all buffers except 4l
to be satisfactory, however g'4 was chosen for its
slightly higher specific conductivity.
An electrophoretic analysis of the mixture was
attempted using the technique described by Tiseliusi
The navy bean crude sterols from fraction 1 and 2,
twice recrystallized, were dissolved in sufficient
buffer solution 44 to make a 0 .2m solution.

This

buffer has a pH of 8.7 and an ionic strength of
about 0.1.

The sample was dialyzea against frequent

changes of the same buffer until the specific
conductivity of the sample and buffer showed that
an equilibrium had been reached.

electrophoresis

was then carried out for 6,000 seconds at a
potential gradient of 4-2 volts per cm. in one
case, figure Ilia, and lor 12,000 seconds at a
potential gradient of 4*5 volts per cm. in another
case, i'igure Illb.

Photographic records were made

using the fongsv or th-ochlieren scanning technique.

(S1 i

figure Ilia shoves the electrophoretic patterns
obtained after electrophoresis for 6,000 seconds.
The descending pattern reveals a small peak migrating
as an anion.

The concentration of this material, as

determined from the relative areas of the two peaks,
is low.

The large peak shows no tendency to migrate

and remains electrophoretically homogeneous.

The

ascending pattern would seem to indicate the presence
of a small amount of material migrating toward the
cathode.

however, the separation from tne large amount

of electrophoretically inert material is incomplete.
The patterns obtained after electrophoresia of
the same sample for 12,000 seconds are shown in
i'igure Il lb.

The small peak migrating toward the

anode is no longer apparent.

This is probably due

to the concentration gradient, and therefore, the
refractive index gradient becomes so small that

58

the peak cannot be detected.

The ascending pattern

again shows the presence of a small amount of
migrating material.

However, the mobility of this

material is so low that it still has not separated
from the main components.

F ig u r e m

I

j 1— »

*

la

jH»

♦-I*1

a.

El e c t r o p h o r e t i c

Pa t t e r n s
vji

vO

Discussion
figure Ilia reveals a small peak, probably a
non-steroid impurity because of the low concentration,
migrating as an anion-negatively charged.

The rest

of the pattern shows no migrating material.

The

ascending side reveals some material (very low
mobility) migrating as a cation-positively charged,
kost of the material is inert.
i'igure Illb, descending pattern, does not
reveal any small peak (explained above) and
ascending pattern shows the same material as the
other ascending pattern.

Large peaks in each

pattern represent electrophoretically inert material,
ho actual separation was effected.
It is to be noted that in this work,9 unlike
v
■

that of m o y e r ’s, the sterols were in solution as
chemical individuals, or possibly associated, and
not as hydrated colloids.*

* Acknowledged credit is here given to Dr. C. K. Hardt
for the electrophoretic work and the developing of
the patterns.

B. Irradiation of Crude Navy Bean bterols.
Occurrence of vitamin-D activity, as a consequence
of irradiation of a given seed oil or the crude
sterols therefrom, is generally attributed to the
presence of small amounts of ergosterol.

However,

this does not exclude the possibility of other
activatable provitamin-D's, yet unisolated, being
present in the seed oils.
For this experiment a total of 0.11? gm. of
material, made up of equal parts from all the
fractions of the unsaponifiable portion, was
irradiated with ultra-violet light, in ethyl ether
for 5 minutes.

The irradiated solution was then

made up to 100 ml. with ethyl ether.

buitable

aliquots were taken from this stock solution for
assay by the standard biological method,,

The results

of this assay indicated an activity of 700 U. b. P.
units per gram of crude navy bean unsaponifiable
material.v

* The author wishes to thank Or. C. A. Hoppert for
the Vitamin-D assay.

(c) Isolation of St erols,from the
UnsaponifIable fraction of Navy
Bean Oil by fractional Crystallization.
Crystallizations of sterols or their derivatives
have resulted, in the isolation of numerous sterols
from plants.
Only parapiiytosterol

(0 J

has been reported, as

far as the writer can determine, as a sterol
constituent of a portion of the navy bean.
(1) axp er im en ta 1»
one and one-half grams of
crude sterols were taken from each of the first two
fractions (pages 47 and 48).
3 gm. of crude sterols,

To this combination of

in a 123 ml, erlenmeyer flask,

were added 60 ml. of acetic anhydride.
was heated for If Lours under reflux.

The mixture
The volume of

the solution was then decreased to the extent of
saturation.

After cooling for 1 hour at 12° C. the

crude acetates were filtered.
Bromination of the crude steryl acetates--Three
and one-tenth grams of crude acetates were dissolved
in 30 ml. of ethyl ether, and to this solution were
added 3S ml. of bromine-acetic acid solution (3 gm.

63

broruine in 100 r.il. ol' glacial acetic acid) following
the procedure of Yindaus and Lauth.

(34-)

After allowing

the solution to stand in the refrigerator,

the

insoluble tetrabromides were filtered off and washed
with cold ethyl ether.

Yield 1,4 g m ., m. p. 190-194?

This would correspond to about 25a sterol (on the
basis of stigmasterol)

in the crude mixture,

Six

recrystallizations of the tetrabromide from chloroformmethanol mixture gave a melting point of 194-196° C,
Stigmasteryl acetate— To a solution of 1.2 gm. of
the tetrabromide acetate in 12 m l . of glacial acid
were added 1.2 gm. of zinc dust.

This mixture was

refluxed for lg hours, filtered hot, diluted with
water and extracted with ethyl ether.

The ether

layer was washed with a dilute sodium sulfite solution
and then with water, and the ether finally removed
by evaporation on the steam bath.
acetate was $80 mg.

Yield of the crude

It was recrystallized 4 times

from ethanol ana twice lrum 2-1 methanol-chloroform
mixture; m. p. 139-140° C.;
2.1 ml. of chloroform, 1 dm.,
average rearning).

-54.0
^

(27.4 mg.,

-0.902 ,

Anal.5''

3.074 mg. gave 2.949 mg, HOE; 9.150 nig. CO2

Calc'd. for C31^ 50^2 }

C, 8I.I8 7 ;

H, 10.73>

C » 81*88A >

H > H.09>

^tigmasterol--i<'our~hundred milligrams of tlie
above acetate were hydrolyzed by refluxing for 1
hour with sufficient 10^ alcoholic potassium
hydroxide.

later was added to the mixture and the

latter extracted with ethyl ether.

The ether

solution was washed with a dilute sodium carbonate
solution,

then with v/atwr, and evaporated to near

dryness on a steam bath.

The residue w a s

recrystallized irom 95 m ethanol 3 times; im p.
168-109° C.;
chloroform, 1 - 1
Anal.

' -47.3° (28.0 mg., 2.1 ml. of
am.,

^ - u .64U ° )

3.121 ing. gave 3.153 mg. H 20 ; 9.552 C02
. 0 , 83 .47/0 ;

Calc ’d

C 09E).gO ,

C, 84• 40/o ;

H, 1 1 .70>o
E,11.30A

b t i ^ i a s t e 1 benzoate— About 300 mg. of the
free sterol were dissolved in 2 ml. of dry pyridine,
and to this was added 0.6 ml. of benzoyl chloride.
The imixture was heated on the steam bath for 2g hours
in a 25 m l . glass- stopper eel l'lask and the mixture

v ihcroaualyses are by Dr. T. 6 . Aa,

university of Chicago

then allowed to stand over-night.

The contents

were poured into an ice cold 5a sulfuric acid
solution.

The ethyl ether extract of the mixture

was washed with dilute sodium carbonate solution, and
finally with 10;o ethanol solution.

The contents were

evaporated to dryness on the steam bath, and then
recrystallized 3 times from ethanol and once from a
•iiethano1-chioroform mixture; m. p. 160.5-161.5°;
jj-24.5
1 - 1

dm.,

Anal.

(112.3 mg, , 2.1 ml, of chloroform,

cvCt4

-1.779°, average reading).

2.959 m g . gave 2.602 mg. H 20; 9.108 C02

Csl'e’d

C-^IIc^Op

0, 8 3.95o ;

Ii,

9.83'A

C, 33.65m ;

H, 10.15'A

Nearly all of the remainder of fractions
1 and 2 plus fractions 3 and 4 were combined and
converted into the acetates by dissolving in 250 ml.
of acetic anhydride.

The mixture was re fluxed for

2 hours arid allowed to stand over-night.

Floating

on the top of the solution was a wax-like cake,
yellow in color, and in the solution prjper crystals
were noted.

The flash and contents were warmed

slightly (about 40° C . ) ii: oruer to effect solution
of the crystals.

The mixture was filtered warm,

thus collecting the wax-liIce cake on the filter.
The residue (about 3 gm. ) was set aside.
The volume of the filtrate was reduced until
complete crystallization had nearly occurred.

The

contents were then cooled for 1 hour in the
refrigerator, alter which time the mixture was
filtered and washed with cold glacial acetic acid.
The mixture of supposed acetates (about 10 gm.)
was brominated, in the way previously described for
stigmasterol isolation in order to effect the removal
of the stigmasterol present.

The stigmasteryl

acetate tetrabromide was filtered off and the
derivatives in the filtrate were debrominated in
the usual manner with zinc dust.
mince

Y

-sitosterol,

(5t '
' the most .insoluble

of the sitosterol coii.pl ex, is also precipi tat able
as a bromide,

const,ant observation was made in

this work for Y

-sitosterol,

found for its presence.

but no evidence was

however,

its presence in

small quantities is not excluded.
The debrominated acetates ’were taken up in
sufficient 1-1 acetone-chlorolonu mixture to effect
what was considered to be a half-saturated so

The contents were allowed to stand over-night.
The solid acetates, designated as AC-1, were removed
by filtration.

The filtrate was subsequently

designated as AC-2.

The AC-1 fraction was subjected

to triangular fractionation using a 3-1 acetonemethanol mixture and later a 3-1 ethanol-benzene
mixture .

The separation of

is shown in figure IV.

-sitosteryl. acetate

The separation of a wax-

li'xe substance was also effected in this fractionation.

68

i'igure ±V

Ac /

Ac ‘

jccy
/Aasi

V ny

kU.

/ra-s-/-

-/A- «-T/fos/'<?>'}
’'t<C*/#/<T ~/'U££
/■'-<’>j-i /? < r r.

Separation o f - ^ -Sitosteryl Acetate
and a Wax-like substance

'~ / 3 -Sitosteryl acetate--Following the final
recrystallization from ethanol the compound, melted
at 123-1<4°;
chloroform, 1 - 1
Anal.

“38.5° (40.3 mg., 2.1 ml. of
dm. ,

-0 .744°, average reading).

2,970 mg. gave 2.912 H^O

;

8.886 mg. C02

C, 81. 594,
Calc 1d.

°3iH 52°2

H, 1 0 .9670

C ’ 81 *52A;

H, 1 1 .477°

-Sitosterol— The above acetate was saponified
with 10/a alcoholic potassium hydroxide and further
worked up as usual.

Three recrystallizations from

m e thanol-chloroform mixture yielded the free sterol;

m„ p. 130-138° C.;

-37.8° (19.2. mg,,

of chloroform, 1 = 1 dm.,

<^*^-0.346°,

Anal.

average reading)

2.929 mg. gave 2.957 mg. HgO ; 8.989 mg, C02
G ,

Ga le’d .

2.1 ml.

°2q ^50^

8 3 . 4 1 a

;

83*997°

H ,

1 1 . 2 5 a

H, 12.15/°

-y^7 -Sitos..t.eryl benzoate— -Fit ty milligrams of
free sterol were reacted with 0.2 ml. of benzoyl
chloride in 1,5 ml. of dry pyridine.

The mixture

was heated on the steam bath for ll hours and allowed
to stand over-night.

The contents were poured into

10 ml. of cold 5a sulfuric acid.
extracted with ethyl ether,

The mixture was

The fourth recrystallized

product from ethanol melted at 145-146° C.;
-13*5° (19-4 H g . , 2.1 ml. of chloroform,
1 = 1 dm.,

otC?) -0.125 , average reading).

The crystalline acetates of the AC-2 fraction
were obtained by evaporating the solvent on the
steam bath.

The residue was taken up in ethanol

and acetates hydrolyzed by addition of a 10yo solution
of potassium hydroxide.
as usual.

The solution was worked up

The crude sterols so obtained were dried

and then reacted with sufficient 3 ,5-cinitobenzoyl
chloride to yield the esters.

Triangular fractionation

was used again in an attempt to isolate one of the
three known n

-sitosterols" by taking advantage of

the fact, as did Vallis and Fernholz '

that the

corresponding 3,5-dinitrobenzoates differ greatly
*
±*i

4* V !’■' ■? -k»

uUoli

p.
1 t<i Vs ■{ *1 A 4* f
E
um
uuixilv •

m e

/

^- o01x4

—•*~1

pi

y J_‘~ui~

dinitrobenzoate is by far the most insoluble while
the <=*£-si to sterol derivative is the most soluble.
As a result of this attempt,

the most

insoluble

3,5-dinitrobenzoate obtained

had a in.

p. of 222-223° C.;

[<]^ X -1 7 .3° (23.6 mg. , 2.1 ml. of chloroform, 1 = 1 din.,
-0.1.95°, average reading).
melted at 155-157° C.;
of chloroform, 1 = 1 dm.,

The free sterol

-55.8°

(15.6 mg., 2.1ml.

-0.415°, average reading).

71

From tiie free stei*ol the acetate was made which
melted at 134-135°
reaction was positive.

Ihe Liebenuann-Burchard
bince all the derivatives of

the "c?<L-sitosterols" have a positive value for the
optical rotation witli a relatively large magnitude,
one is inclined to conclude from the above observations
that no " ^s^l-sitosterols'’ are ^resent in the navy,
bean oil.

The high, negative value of rotation for

the sterol is of interest si.ce it appears to be the
highest of the phytosterol group. L mycosterol,
ergosterol, has a rotation value of -132.0°.
Discussion
The navy bean oil herein studied, making up
about 2 .$/■!? of the bean, consists of 5.-4.Op
unsaponifiable.

The unsaponiiiabie consists of

about $$-00/0 crude sterols, of which about 23-25/0
consists of stigmasterol.

C,^>

-c -

otigmasterol

This percentage compares closely with that obtained,
from soybeans by Kraybill et a l . ^ * ^

They found about

one-half of the unsaponifiable to be crude sterols
and of this portion stigmasterol was present to
about 20-25P.
It is estimated thaly^ -sitosterol occurs to
the extent of 5-7jo of the navy bean crude sterols.

iW
C.H
\

h

C-—

H

^

h- h- i'-

R-— C-(—C-\— CU_h
wCW M'
\

-Si
Sitosterol
"Sitosterol", the common plant sterol, has
been resolved into three isomeric sitosterols
(CpcjH^jjO) .^ ^ ^

The order of their insolubility in

organic solvents is

and-sitosterol.

-Sitosterol has been isolated in the pure state
from cotton seed oil.

( 58 )

Cinchol, the sterol isolated

73

from carrots and cinchona bark is now considered
( 59)
to be identical w i t h ^ -sitosterol.
C* Isolation of Irobable hydrocarbons.
(1) Experimental.
Hie wax-like residue obtained from
on top of the acetylated mixture of crude sterols
mentioned on page 0$ was recrystallized from acetone
ana benzene eight times raising the melting point
from 55-080 to 72-73° C.

The white, waxy-like

material, designated as A, was insoluble in cold
concentrated sulfuric acid and did not decolorize
bromine-chloroform solution,

ho further

characterization was made.
A substance more soluble in acetic anhydride
was separated in the fractionation-of
page 08.

This was called substance B.

-sitosterol,
The results

of the qualitative tests were the same as those
designated for A above.

After.thirteen recrystallizations

from an ethanol-benzene mixture and from p-dioxane the
substance melted at 65-6o.5° C.
was made.

ho further study

Summary
From Tables III and V it is seen that the
polyuronide and the "true pentosans" content
of the navy bean seed, coats are about 19/° and
21/o, respectively, of their dry weight.
The protein is about o'-p of the dry weight of
the seed coats.
Treatment of the seed coats with 0.8> sodium
hydroxide lowers the content of both polyuronide
and "true pentosans", while 1)o acetic acid and
1/" sodium, bicarbonate have little effect,
rrote in of the seed coats is slightly lowered by
sodium bicarbonate treatment and noticeably by
treatment with sodium hydroxide.
There is an indication that polyuronides and
"true pentosans" are involved in water retention
in the dried seed coats.
from Table VIII it is seen that the crude lipid
and unsaponifiable are about 2.o5h,} end 0.15/'-’,
respectively, ol the air dry seeds.
A modified method based on the Liebermannburcliard reaction has been presented for the
determination of crude sterols.

75

(8) The crude sterols of the navy bean oil
unsaponifiable are about 59>.
(9) The proteins of the ground whole navy beans
have been effectively extracted with 0. 5>
sodium sulfite solution and precipitated
satisfactorily with sulfur dioxide.
(10) The protein is adaptable to plastic formation.
(11) Three sex>aration studies on crude sterols
separated from navy bean oil have been attempted,
namely, chromatographic, electrophoretic, and
triangular fractionation.

Only the latter was

successful»
(12) A vitamin-D activity of 700 U. h. 1. units per
gram of crude navy bean oil unsaponifiable matter
was observed.
(13) otigmasterol

-sitosterol have been Isolated

and characterized.
(14) fm unidentified sterol with a high negative
rotation has been isolated.
(15) Two different ptrobable saturated hydrocarbons
have been isolated from the navy bean oil.

Bibliography
(1) --inton, A, L. and A in ton, n. B., Structure and
Composition of Fo o d, Vol. II, John wiley
and Sons, Inc. Few fork, K. Y. p. 153.
(1935).
(2) ochulze, B. and i fenninger, U . , I. physiol.
Chem., 61, 279-351, (1909).
(3) miller, E. G., Plant Physiology, 2nd Ed.,
Pc G-raw-iiill Book Co., P. Y. , (19 38).
(4) Gortner, u. A., Outlines of Biochemistry, 2nd
E d . , John wiley and Sons, Inc., hew York,
P. Y., (1938).
(5)

Grimme, G., z. Bahr. Genussm, 21, 547, (1911).

(o ) B i n g , L . C .ana B a ll , G. jj., J . n m . wiem.
o4, 2488, (1943).

Soc . ,

(7) Aindaus, A. and kauth, A . , Per. 40, 3681,

(1907).

(8) sandquist, 11. and Gorton, J., B e r . 63, 1935, (1930)
(9) Likiernik, A., z. physiol. Chem., 1 5 , 42b,

(1891).

(10) Link, n. P., and nedaen, P.,I. Biol. Chem., 94,
307, (1931).
(11) kelson, A. L. ana Cretcher, L. H . , I. Am. Chem.
Soc., 51, 1914, (1929).
(12)

ni r d , G. iv-.and Baas, ± . , B io chem. J. 2 5 , 403,
(1931).

(13) Heidelberaer, m. and Goebel, P.. F. , J. Biol.
Ahem., 74, 613, (1927).
(14) Borman,

A. G . , Ann. Pev. niochem., 10, 65, (1940).

(15) Lefevre, *>•. I), and Pollens, B . , Ber. 40, 4513-4523,
(1907).
(16) nickson, A. D. , utterson, H . , and -^ink, k. P.,
J . .u.i. Ghem. .joc., 51, 775,779, (1930).

(I?) i-aillips, js-i* , ^oss, 1... <14, and B r o w n e , C. A * ,
J. nssoc. O ff ic ia l Agr.
16, 289,
( 1933).
~ '
(18)

T o l i e n s , b.,

ber.

44,

1788-1790,

(19) i-andel, J. A, and Heuberg,
148-151, (1.908).
(20)

(1908).

C . , ^iochem.

Z.,

13,

whistler, A,, Martin, a . R., and Harris, 1..,
Bu re au of st an d a r d s J. Research, 24,
13-25, (1940).

(21) H o m a n ,

A. 0., nature,

(22)

J. A.,

bowman,

aoc.,

1 4 3 , 284-285,

and be Ninnis,

(1939).

±x. 3., J . Am. Ch em

52, 1209, (1930)1

(23) Norris, 4. HA , and Huso h, 1. A.,
29, 1590-1596, (1935).

Biochem. O'.,

(24)

Association of Official Agricultural Chemists,
Official and Tentative Methods of Analysis
5th Ed. Washington, 0 . 641,J
) (1940).

(25)

hoc'ccett,A. C . , Out tag, A, , and -Midth,
1. riiii. Onem. boo. o 5 , 1-3, (1943).

Li.,

(26) Nor .uan, 2.. G. , The biochemistry of Cellulose,
the Polyuronides, -tAgiuin, etc . , Clarenden
rresfc, Oxf ord , ^1>•37) .
(27) j-'-och, V,., --xu, lour. Physiol. 11, 330-339,

(1904)

(28) iujiOi^n/a, k . and outo, N ., bio chem. Z. , 8_,
212-347, (1908).
(29) Bando, C. N . , Plant Ph ysiol. 3_> 155-184,

(1928).

(30)

Dill, -j .JA, jnaster ’s Thesis, b. b, C ,, (1942).

(31)

Tower, BA B. , and oalway, A. i;, ,JA Chem.
103, 399, (1913).

b oc .

(3 2) lantnen, i)., and Gohdes, 'A. , Diochem. Z. 272,
166, (19 34) •

78

(33) Thornton, n,. h. , Kraybill, H. R , , and Kitobeli, J. H . ,
JA, A m . Chem. hoc., 62, 2079, (1941).
(34) ding, L. C., Ph.D. Thesis, h.# S. C . } (1942).
(35) Steele, C. C . } In trod uc tion to Plant Biochemistry,
G. Bell and oons, London,' p. 57~i ("19347.
( 3b)

Windaus, A. Ber. 42, 238, (1909).

(37)

Aehoenheimer, R. and Data, H« , Z. ohvsiol. Chem.,
2 15 , 59, (1933).

(38) B a su , K.. P.. and ^ath, k. C. , Biochem. I. 2 8 ,
.1561/(1934).
(39) Sabotka, H. , The Chemistry of the Starids,
;villia a is and Wilkins, p . 12 5~ ( 19 3 8 ) ,
(40) Rindaus, A. and Reinhold, R., Z. nhvsiol. Chem.
126, 299, (1938) .
(41)

rindaus, A.,

(42)

bchoenheimer, R. and Sperry, W. L . , 1. Biol. Chem.,
106, 745, (1934).

Z .

physiol, Chem., 05, 13 0, (1910).

(43) Bernoulli, A. L . , C. A. 26, 3732, (1932).
(44) Bzent-Gyorgy i , A., Biochem. Z. 136, 107 and. 112,
(192 3).
(45) bobel, A. £>., nrekter, I. J., and Ratelson, S.,
J. Biol. Chejii.. 115 > 381, (19 3o).
(4 6 ) liorva t.b, A. A. , Hie 00 y he an Industry, Chemical
publishing Company of Hew York, Inc.,
Lev/ York, P. Y., p. 150, (1938).
(47) Aitthausen, n . , J. prakt, Chem,, 2£, 448, (1884).
(48) Osborne, T. b. , A . Am. Chem. Soc. 1_6, 633, 70 3,
757, (1894).
(4 9 ) Lm'idenburs, K . , Fernholz S., and Rail is, B. S.,
J . Org. Chem., 2> 294, (1938).

(50) -Iway, F. J,, Ber. 37, 344, (1904).
(51) noyer, L. A. , Biochem. Z. , 273, 122,
(52) Tiselius, A., Trans, x-araday o o c . ,

(1934).

22>

5^4, (1937)

(53) Congsworth, L. G . , O'. Am. Chem. ooc., bl, 529,
(1939).
(54) Aindaus, A . , and Eauth, A., Ber. 3 9 . 4379,

(1906).

(55; Anderson, r.. O'. et a l . , -J. Ain. Chew., 4 6 , 1450,
(3,924) , Ibid. 4 8 , 297 2 , (1926).
(56)

..ellis, A. S., ano lerniiolz, A., O'. Aw. Chew. Boc.
5 8 , 2446, (19 36),

(57) iaraybill, h. k. , Thornton, n, A., and Aldridge, E.
Ind. Brig. Oilera., 3_2, 113S, (1940).
(5o) Aallis, B. b. , ana Onakr&vortv, I . n., O'. Ore.
Chem., 2 , 335, (1937).
(59) 16 rscher1, ,.., Z . physio1. Che w., 257, 239, (1939)