Experiment
Experiment
First Edition
1951
New York
Toronto
London
Receive9.t 17-'1'.3_3
~()ok No.
[Link]~MENTS
..
IN BIOCHEMISTRY
EXPERIMENTS IN BIOCHEMISTRY
Copyright, 1951, by the McGraw-Hili Book Company, Inc. Printed in the
United States of Amerita. All rights reserved. This book, or parts thereof,
may not be reproduced in any form without permission of the publishers.
Second Printing
PREFACE
Experiments in Biochemistry is designed as a laboratory guide and reference book for students in
beginning biochemistry. Experiments of wide scope and uniform quality have been provided. Opportunities
are presented to become skilled in the use of common and special biochemical apparatus, to become
proficient in the quantitative determination of biological substances in a variety of naturally occurring
materials, and to gain experience in th~ synthesis, isolation, purification, and analysis of typical
biological compounds. The student's understanding of, and appreciation for, the 'development, use,
and application of biochemical procegures is furthered by detailed explanations and discussions of the
principles underlying chemical reactions and techniques and of background and related current investigations. The more than 900 references to the original and the review literature should aid in increasing
the student's comprehension of biochemical subject matter.
Among the new experiments which have been found to interest students are the, synthesis of an
aromatic amino acid, synthesis of a peptide, asymmetric-enzymatic synthesis of an optically amino acid
derivative, isolation and analysis of an amino acid amide, isolation of a sterol, polarimetric analyses of
amino acids and carbohydrates, separation and identification of p~ines and pyrimidines by paper chromatography, identification of amino acids in a mixture of amino acids, identification of carbohydrates in
a mixture of carbohydrates, determination of nicotinic acid by microbiological assay with a lactobacillus,
determination of choline by microbiological assay with a mutant strain of Neurospora crass a, and determination of vitamin D by the chick bone-ash method.
If it is to be expected that the student will put forth his best efforts, the results of his work should
be evaluated by objective standards. For this reason the authors have employed analytically pure unknowns
in the quantitative experiments, have applied the recovery principle in experiments on the determination of
biological substances in naturally-occurring materials, and have estimated the precision of analyses aJld
the purity of preparations on the basis of the stu~ent's analytical data.
Precise directions for the preparation of special reagents, complete'lists of apparatus and reagents,
precautions in the conduct of experiments, discussions of biochemical principles and techniques, and a
, summary of the results obtained by students in th_e authors' classes over a period of more than fifteen
years are given in the Appendix.
The authors have the Jtope and belief that Experiments in Biochemistry will be stimulating to
students and useful to instructors.
'lIAX S. DUNN
NILLIAM DRELL
CONTENTS
Experiment
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Title of Experiment
Determination of the Equivalent Weight of an Unknown Organic Acid
Synthesis of Glycine
Preparation of Glycine Methyl Ester Hydrochloride
Preparation of Glycine Anhydride (Diketopiperazine)
Preparation of Glycylglycine Hydrochloride Monohydrate
'Preparation of G lycylglycine
Synthesis of Hippuric Acid
Synthesis of Phenylalanine
Kjeldahl Analysis of an Amino Acid or Amino Acid Derivative
Analysis of an Amino Acid by the Van Slyke Volumetric Nitrous Acid Method
Volhard Analysis of Amino Acid or Peptide Hydrochloride
IlSolation of L-Tyrosine from Silk or Casein Hydrolysate
Photometric Analysis of Tyrosine Based on th~ Millon Reaction
Photometric Determination of Tyrosine in Casein
Preparation of Tyramine Hydrochloride
Isolation of L-Cystine from Human Hair
Polarimetric Analysis of L-Cystine
Photometric Analysis of L-Cystine Based on the Reaction of Cysteine
with Phosphotungstic Acid
Isolation of Asparagine Monohydrate from Lupine S_prouts
Amide-Nitrogen Analysis of Asparagine Monohydrate
Identification of Amino Acids in a Mixture of Amino Acids
. Preparation of Casein from Milk
lDetermination of Phosphorus in Purified Casein v
Determination of Acid Number of Unknown Lipid -../
Determination of Saponification Number of Unknown Lipid '-'
'Determination of Iodine Number of Unknown Lipid"V"
Isolation of Cholesterol from Beef Spinal Cords ../
Preparatlon of Starch from Potatoes
Determination of Starch Factor and Starch in a Starch-Protein Mixture
Isolation of D-Xylose from Corncobs
IsolSltion of L-Arabinose from Mesquite Gum
Isolation of D-Mannose from Ivory-nut Waste
Isolation of Cl-D-Lactose Monohydrate from Whey Powder
Polarimetric Analysis of a-D-Lactose Monohydrate
Determination of Free and Combined Glucose in Corn Sirup
Identification of Carbohydrates in a Mixture of Carbohydrates
Separation and Identification of Purines and Pyrimidines by Paper
Chroma tography
Determination of Vitamin C in Grapefruit Juice
Determination of Nicotinic Acid by Microbiological Assay with
Lactobacillu s arabindsus 17-5
vii
Pas.e
1
2
4
6
7
8
9
10
13
15
19
20
22
23
24
25
27
28
29
31
32
45
47
49
50
51
53
55
58
60
62
64
66
68
69
71
84
88
91
Experiment
number
40
41
42
43
44
45
46
Title of Experiment
Determination of Choline in Egg Yolk by Microbiological Assay with
Cholineless Neurospora crassa
Determination of Vitamin D by. the Chick Bone-ash Method
Determination of the, Activity of Salivary Amylase and the
Concentration of Unknown Starch Solution
Determination of Alkaline Phosphatase Activity in Blood Serum
Asymmetric Enzymatic Synthesis of Acet,YI-L-phenylalanine
p-Toluidide and Isolation of L- arid D,.Phenylalanines
Determination of Cholesterol in Blood ~rum
Determination of Nitrogen Compounds in Urine
Page
95
98
102
104
106
108
110
117
157
Appendix
Bibliography
Index
Data Sheets
183
viii
E X'P E RIM E N T
DETERMINATION OF THE EQUIVALENT WEIGHT OF AN UNKNOWN ORGANIC ACID
This experiment is to test the ability of the student in performing simple quantitative manipulations and calculations.
EXPERIMENTAL
Obtain approximately 1.0 g. of an unknown solid organic acid. Transfer the sample to a dry weighing bottle. Transfer about 1.0 g. (weighed to 1 mg.) of the solid acid to a 100:-mI. volumetric flas~
Add about 50 ml. of distilled water, rotate the flask until the solid dissolves, and add distilled water
to the mark. Stopper the flask and thoroughly mix the liquids.
Pipette a 25.O-ml. aliquot of this solution into each of three 125-ml. conical flasks. Add 2 drops
of phenolphthalein indicator solution to each flask and titrate each solution with standard, approximately 0.1 N, sodium hydroxide delivered from a burette.
-1-
EXPERIMENT
SYNTHESIS OF GLYCINE
Glycine (sugar of gelatin), the simplest amino acid, was first isolated from an acid hydrolysate
of gelatin by Braconnot(1820, 1827). It tastes more swe~t than other amino acids but less so than
sucrose, some other sugars, and compounds such as dulCin ([Link] OCJ1 4NHCONH.) and sacc;harin
(<;::J!.SOoNHYO) (Cameron, 1947).
Glycine was first synthesized in about 20 per cent yield from bromoacetic acid and ammonia
(Perkin and Duppa, 1858; Cahours, 1858, 1859). It was crystallized as the copper salt because separation (in the free form) was difficult from the ammonium salts of iminodiacetic acid (called diacid) and
trimethyleneaminetricarboxylic acid (called triacid) (see equations below).
[Link] + [Link] + 3NH,
[Link]
->
->
Kraut (1890, 1891) increased the yield of glycine to about 50 per cent by increasing the molar
ratio of ammonia to halogen acid to 20: 1. At a molar ratio of 220: 1 Robertson (1927) obtained nearly
the theoretical yield of glycine. Excess ammonia and chloride ions were removed by treating the crude
product with silver oxide. Orten and Hill (1931) found that glycine could be purified more conveniently
by crystallizing it from 85 per cent (by volume) aqueous methanol.
Purified glycine has been prepared in 55 to 60 per cent yield by the reaction at 60 to 65 0 of a
mixture containing ammonia, carbonate, and chloroacetic acid in an equivalent ratio of [Link] 1 (Chadwick
and Pacsu, 1941; Cheronis and Spitzmueler, 1941; Dunn, Butler, and Frieden:, 1941). Robertson obtained the same yield at rOOm temperature with a 60: 1 molar ratio of ammonia to chloroacetic acid. The
effectiveness of small amounts of carbonate in depressing the producti~n of the di- and triacids and in
increasing the yield of glycine has been explained by the conversion of glycinate ions to alkali-stable
carbaminoglycinate ions.
CO. + 2NH,
->
[Link]'" + NH4 +
-> -O~CNHCHaCOO-
+ NH 4CI
Various methods for the synthesis ~f., glycine have been discussed by Clark (1943), Carter and
Hooper (1944), Dunn (1943), Dunn and Rockland (1947), and Block (1946). The synthesisof glycine
labeled with radioactive (C14) carbon has been described by Sakami, Evans, and Gurin (1947), Loftfield
(1947), and Ostwald (1948).
Animals convert glycine to creatine (Bloch and Schoenheimer, 1941), uric acid (Shemin and Rittenberg, 1947), serine (Sakami, 1949), and the protoporphyrin (C'4H'40 ..l\l4) mOiety of hemoglobin
(Wittenberg and Shemin, 1949). According to Shemin (1945) the nitrogen, as well as the carboxyl and
';djacent carbons; of serine are utilized in the in vivo synthesis of glycine. In animals glycme is incorporated into the proteins of intestinal tissue, liver, and spleen (Friedberg, Winnick, and Greenberg,
1947; Winnick, Friedberg, and Greenberg, 1947). The yeast Torulopsis utilis transforms glycine into
serine (Ehrensvard, Sperber, Saluste, Reio, and Stjemhelm, 1947), proline (Ehrensvard et a/., 1947),
and purines (Abrams, Hammarsten, and Shemin, 1948). The conjugation of glycine with benzoic acid
and other aromatic acids to form hippuric acid (benzoyl glycine) and analogous derivatives has been
reviewed by You,ng (1939), Stekol (1941), Lewis (1941), Quick (1944), and Handler and Perlzweig (1945).
Glycine is indispensable for growth of the chick, but not of the rat and other animals. It has ,been employed clinically for the alleviation of muscular disfunctions although its curative powers have been.
questioned (Burman, 1943).
-2-
Glycine is to be synthesized by the reaction of chloroacetic acid and ammonia (CICH 2 COOH +
2NHs ... NH 2CH 2 COOH +[Link]). It is readily crystallized from 80 per cent aqueous methanol, since it
i.. only slightly soluble while [Link] is almost entirely soluble in this solvent.
EXPERIMENTAL
Mark a saO-mI. round-bottomed pyrex flask at the 100-ml. level with a wax pencil. Place 162 g.
(1.3 moles) of technical ammonium,carbonate,l (see end of experiment for footnotes) 190 ml. of distilled
water, and 75 ml. (1.1 moles) of 15 N ammonium hydroxide in the flask. Stir the mixture thoroughly and
heat it on a water bath at 60 until the solid dissolves. Prepare a solution containing 35 g. (0.37 mole)
of chloroacetic acid (m.p. 62 to 64 0) in 60 ml. of distilled water, and add this solution slowly, with
stirring, to the ammoniacal solution in the flask. Label the flask, cover the neck of the flask with an
inverted beaker, and place the covered flask in an oven at 60 to 65 0 for 24 hr. or longer. The yield and
quality of the glycine are not altered by heating the mixture longer than 24 hr.
Set up th~ reduced-pressure distillation apparatus shown in Fig. 7 (see Distillation in the Appendix).
Close the screw clamp on the tubing leading to the safety flask. Turn on the valve controlling the
flow of water from the water pump. Open the screw clamp sufficiently to permit rapid evolution of gas,
from the liquid in the distiiling flask, and as the gas evolution diminishes, heat the water bath sufficiently to maintain rapid distillation. Continue the heating and distillation under reduced pressure until
the volume of the solution has been reduced to about 100 ml. Add 100 mI. of d~stilled water to the flask
and repeat the distillation process until the volume of the solution has been reduced to about 100 mI.
Transfer the residual solution to a I-liter beaker and add 500 ml. of methanol. Stir the mixture
thoroughly, cover the beaker with a watch glass, label the beaker, and place the covered beaker in a
refrigerator overnight or for longer time. Transfer the supernatant liquid to a Buchner funnel and filter
it with suction. Add 100 mI. of methanol to the beaker, stir the mixture for about 3 min., and transfer
the alcoholic suspension of crystals to a Buchner funnel. Apply suction until the crystalline precipitate is nearly dry. Resuspend the glycine in 75 mI. of methanol, stir the mixture "for about 3 min., transfer the suspension to a Buchner funnel and apply suction until the crystals are nearly dry. Transfer the
alcoholic filtrates to the container provided for that purpose to permit recovery of the alcohol. Transf 7 r
the crystals to an evaporating dish, label and cover the dish, and place the covered dish overnight in an
0
oven at 50 Transfer the dry product to a labeled, tared (weighed) sample bottle, and submit the (stoppered) bottle to the instructor. Calculate the percentage yield of glycine and submit the calculations to
the instructor.
NOTE
1. Technical ammonium carbonate is a mixture of anhydrous and monohydrated ammonium carbonate, ammonium bicarbonate, and ammonium carbamate (N1400CNH2)' It has an apparent equivalent weight of 60 to 65,
and it is assumed to be principally ammonium carbonate monohydrate.
-3-
EXPERIMENT 3
PREPARATION OF GLYCINE METHYL ESTER HYDROCHLORIDE
Glycine ester is utilized for the pr~paration of ethyl hydantoate (Harries and Weiss, 1903),
glycine anhydride (Fischer, 1906; Dickinson and Marshall, .1929), pofyglycine esters (Frankel and
Katchalski, 1942; Fraenkel-Conrat, Cooper, and Olcott, 1945) and optically active peptides (Bergmann
and Fruton, 1941; Sheehan and Frank, 1949).
Glycine methyl ester hydrdchloride is to ,be prepared by passing dry hydrogen chloride into a
suspension of glycine in anhydrous methanol. This procedure, first described by Curtius and Goebe]
(1888), is common1y employed for the preparation of the ester hydrochlorides of amino acids and peptides. The free esters are readily liberated with base and separated by ether extraction.
A
B
C
D
AND PREPARATION
GLYCINE METHYL ESTER HYDROCHLORIDE.
H
I
J
K
L
M
N
o-
Rubber tubing
125-ml. flask (trap)
200-ml. r';und-bottomed fl ask
600-ml. beaker
GI ass funnel
lOO-ml. 6 N sodium hydroxide solution
T-tube, I arge bore
Glass rod
EXPERIMENTAL
Set up the apparatus shown in Fig. 1.
Place 15 g. of purified glycine and 75 m!. of anhydrous methanol in the 200-m!. round-bottomed
flask. Admit a small stream of concentrated sulfuric acid from the dropping funnel on to the mixture of
sodium Fhloride and concentrated hydrochloric acid in the SOO-m!' conical flask. The sulfuric acid
-4-
should be added at such a rate that the liberated hydrogen chloride bubbles in a steady stream thr,ougb
the sulfuric acid in the widemouthed bottle into the reaction mixture in the 200-ml. flask. If any hydrogen chloride escapes into the room, reconstruct the apparatus. Manipulate the stirring rod in the tube
to prevent clogging of the delivery tube with solid glycine methyl ester hydrochloride. Immerse the
reaction flask at intervals in an ice-water bath to prevent the temperature from rising above the boiling
point of the alcoholic suspension. The temperatute of the suspension should be maintained high enough
to prevent crystallization of the product. A clear solution should form in about 30 min.
Transfer this solution to a beaker and rinse the flask with two lO-ml. portions of hot methanol.
Immerse the beaker for 30 min. in an ice-water bath. Filter the suspension 'on a Buchner funnel. Redissolve the precipitate in the minimum volume (about 6 mI. per g.) of hot methanol; if undissolved
material is present, filter the suspension. Transfer the solution (or filtrate) to a beaker, cover and
label the beaker_, and allow the covered beaker to stand 30 min. in an ice-water bath or overnight in the
refri~erator.
Filter the suspension on a Buchner funnel, Cover the funnel with a watch glass, and apply suction
(for 10 min.) until nearly all the methanol is remo\red. Transfer the crystals to an evaporating dish,
cover and label '#;he dish, and allow the covered dish to stand overnight in an oPen at 50. Submit the
product to the instructor in'.a tared, labeled sample bottle. Calculate th_e percentage yield and submit
the calculations to the instructor.
-5-
EXPER'IMENT 4
PREPARATION 0 F GLYCINE ANHYDRIDE (DlKETOP1PERAZ1NE)
Glycine anhydride is formed by tho:: spontaneous decomposition of glycine ester in water (Fischer,
1906), although it is prepared more conveniently in abo,ut the same yield (45 per cent) by dehydrating
and condensing glycine in hot glycerol (Blanchetiere, 1927; Maillard, 1914a, 1914b; Morrow and Sandstrom, 1935) or'ethylene glycol (Sanni~, 1942). A modification of Sannie's me{bod isto be employed in
this experiment.
EXPERIMENTAL
Place 80 mI. of technical ethylene glycol in a 125-ml. conical flask, suspend a 360 0 thermometer
in the flask, and heat the flask in a hydrogenated-cottonseed-oil bath until the temperature of the ethylene glycol reaches 185. Add 15 g. (0.20 mole) of technical glycine, initially in I-g. portions to avoid
frothing and loss of the ethylene glycol mixture by sudden liberation of water vapor: Maintain the
mixture at about 180 and stir it until the glycine dissolves. Maintain a temperature of about 175 for
45 min. while stirring the solution at intervals. Remove the flask from the bath and cool it in air or
under running water. Stopper and label the flask and allow it to stand 30 min. in an ice-water bath or
overnight in the refrigerator.
Decant (and discard) as much of the supernatant liquid as possible. Add 5 ml. of methanol to the
suspension and filter the mixture oh a Buchner funnel. Disconnect the pressure tubing leading to the
filter flask, add 10 mI. of methanol to the dark-brown precipitate in the funnel, and stir the suspension
thoroughly without disturbing the filter paper. Attach the pressure tubing and apply suction until the
precipitate is nearly dry. Press the precipitate with an inverted glass stopper until as much solvent as
possible is removed. Repeat this process, using a second 10-ml. portion of methanol.
Dissolve the precipitate in the minimum volume (10 mI. per g.) of boiling distilled water, and in
undissolved solid is present, filter the mixture. Transfer the filtrate to a flask and cool the flask for
30 min. in an ice-water bath or overnig..'1t in the refrigerator.
Filter the suspension on a Buchner funnel and was_h the precipitate on the funnel with distilled
water. Transfer the preCipitate to a 12S-ml. flask and dissolve it in the minimum volume of boiling distilled water. Add 1 g. of decolorizing carbon, heat the mixture to boiling, and continue heating for 5 min.
Filter the hot suspension on a Buchner fkn~l which has been thoroughly heated by immersing it in hot
water. Triturate the carbon residue with 10 ml. of hot distilled water. If the filtrate is [Link], repeat
the decolorizing procedure using 0.5 g. of carbon. Transfer the filtrates and any crystalline material in
the filter flask to a 125-ml. conical flask. Use distilled water to wash traces of crystals from the flask.
Stopper and label the flask and allow it to stand for 30 min. in an ice-water bath or overnight in the
refrigerator. Filter the !;uspension of crystals and wash the crystals as previously described, first with
11) mi. of 50 per cent methanol and then with 10 ml. of methano}.l Transfer the precipit13te to an evaporating dish, cover and label the dish, and allow the covered dish to stand overnight in an oven at 50.
Submit the product to th~ instructor in a tared, labeled sample bottle. This product is sufficiently pure
for the preparation of glycylglycine hydrochloride, but recrystallization from water would be required to
yield an analytically pure product. Calculate the percentage yield and submit the calculations to the
instructor.
NOTE
1. An additional crop (0.5 g.) may be obtained by evaporating the combined filtrate and washings in vacuo to
about 25 pl. and allOWing the residual1iquid to stand overnight in the refrigerator.
- 6-
EXPERIMENT
Glycylglycine hydrochloride monohydrate is teadily prepared by acid hydrolysis of glycine anhydride under the conditions described by Fischer alld Fourneau (1901) and Schott, Larkin, Rockland, and
Dunn (1947). The prescribed time of heating shOUld be observed closely, since both peptide bonds
would be split and glycine, rather than glycylglycine, formed by prolonged heating.
EXPERIMENTAL
Heat concentrated hydrochloric acid (6 ml. per g. of glycine anhydride) to boiling, add the
purified glycine anhydride (as rapidly as possible but in smail portions to avoid foaming) prepared in Exp. 4, and maintain the solution at the boiling point for 90 sec. Immediately immerse the
flask in an ice-water bath. After 15 min. tran,sfer the flask to the refrigerator and allow it'to stand
overnight. Filter the suspension on a Buchner funnel. Recrystallize the product from the minimum
volume (about 35 mI. per g.) of hot 80 per cent ethanol. Dry the product in air 1 and submit it to
the instructor in a tared, labeled sample bottle. 'the product should be nearly analytically pure.
Calculate the percentage yield and submit the calculations .to the instructor.
NOTE
1. Glycylglycine hydrochloride monohydrate is stal)le at 70 0
~~
-7-
EXPERIMENT
PREPARATION OF GL,YCYLGLYCINE
Glycylglycine may be prepared from glycylglycine hydrochloride monohydrate by neutralizing the
hydrochloride with base (Fischer, 1906). Silver oxide is a satisfactory base for laboratory preparations,
but it is too expensive for large-scale work. Sodium h~droxide is unsatisfactoiy, because glycylglycin~
and sodium chloride are difficult to separate by fractional crystallization from water or other solvents.
Ammonium hydroxide is satisfactory because ammonium chloride is more soluble in methanol than glycylglycine.
EXPERIMENTAL
Dissolve the glycylg1ycine hydrochloride monohydrate prepared in Exp. 5 in the minimum volume
'(4.5 rnl. per g.) of distilled water in a small beaker. Add sufficient concentrated ammonium hydroxide
(1 mI. per g. of glycylglycine hydrochloride monohydrate) to bring the solution to pH 7.0. Test the pH
of the solution with nitrazine paper, which changes from yellow-green to blue color over the pH range
.6.0 to 7.5. Determine the shade of color to be expected by testing a buffer solution at pH 7.0 or by
reference to a color chart.
Add methanol (1.3 mI. per ml. of solution) and immerse the beaker in an ice-water bath for 25 min.
Filter the suspension of crystals on a Buchner funnel and wash the precipitate with two lO-ml. portions
of methanol. Transfer the crystals to an evaporating dish and allow the product to dry in air for 30 min.
Cover and label the dish and allow it to stand overnight in an oven at 50 0 Submit
the product to the
I
instructor in a tared, labeled sample bottle. Calculate the percentage yield and submit the calculations
to the instructor.
-8-
EXPERIMENT
N-Benzoylglycine was named hippuric acid from the Greek words "hippos" (horse) and "ouron"
(urine) because of its abundance in horse urine. It is present in relatively high concentration in the
urine of herbivorous animals, and 0.6 g. is excreted daily in the ~rine of an adult person. Hippuric
acid is synthesized in the liver and the kidney at the rate of about 1.5 g. perl hr. on a constant input of
sodium benzoate. Hippuric acid formation is considered to be a process for the detoxification of benzoic
acid. Benzoid' acid in the body is derived primarily from foods (such as cranberries) in which it occurs
naturally or to w~ich its sodium salt has been added as preservative. Most of the glycine utilized for
conjugation with benzoic acid is derived from body protein or is synthesized in vivo, since only a small
patt of the isotopic (N 15) niuogen present in ingested "tagged" glycine was found in the urinary
hippuric acid (Rittenberg and Schoenheimer, 1939).
Hippuric acid is to be synthesized by tbe reaction of benzoyl chloride and glycine (Schotten, 1888;
Baumann, 1886). The reaction mixture is kept cold. and it is maintained at pH 9 to 11. Under these
conditions hydrolysis of benzoyl chloride is minimized, sodium chloride rather than hydrochloric acid is
formed, and benzoylation of the glycine is nearly complete.
EXPERIMENTAL
Add 15 g. (0.2 mole) of purified glycine and 100 ml. of distilled water to a I-liter three-necked
flask. Rotate the flask until the glycine dissolves. Add 6 N sodium hydroxide (about 35 ml.) to pH 9
to 11 as shown by test (blue color) with thymolphthalein indicator. Add ice to the solution until the
temperature is reduced to about 2 0 and immerse the flask in an ice-water bath. Stir the mixture vigorously (motor stirrer) while adding 24.5 mI. (0.21 mole) of benzoyl chloride from' a dropping funnel at the
rate of about 1 ml. (20 drops) per min. At 3-min. intervals check the pH of the 'solution and aad 6 N
sodium hydroxide as required to maintain the solution at pH 9 to 11. Add ice to tlia solution if necessary to keep the temperature below 50. When all the benzoyl chloride has been added, continue stirring for an additional 20 min.
Filter the solution to remove any suspended particles and add concentrated hydrochloric acid
from a dropping funnel to the filtrate with stirring until the solution is at about pH 3, as shown by test
(blue color) with congo red paper. Filter the suspension of hippuric acid and wash the precipitate four
times with IS-mI. portions of ice-cold distilled water to remove sodium chloride. Wash the precipitate
three times with IS-mI. portions of isopropyl ether to remove benzoic acid. Dry the product in air in a
tared, covere4 evaporating dish and show it to the instructor before starting the next experiment. Calculate the percentage yield and submit the calculations to the instructor.
-9-
The Perkin reaction (condensation of an aldehyde with an active methylene group) was first
adapted to the synthesis of phenylalanine by PlOchl (1884), but the most extensive 'early iflvestigations
were made by Erlenmeyer (l886-1904b). Sodium amalgam was first employed to reduce the azlactone,
but hydriodic acid and red phosphorus were utilized,by later work~rs (Harington and McCartney, 1927;
Lamb and Robson, 1931) to effect the reduction and ~ydrolysis of the azlactone simultaneously. Gillespie and Snyder (1943) have described the laboratory preparation of phenyl~lanine. According to
Breslow and Hauser (1939) acetic anhydride acts as a dehydrating agent and sodium acetate as a base
causing enolization of the hippuric acid. The synthesis of phenylalanine qy other metho~s has been
discussed by Block (1946), Carter and Hooper (1944), Dunn (1943), and Dunn and 'Rockland (1947).
[Link] is an essential dietary nutrjent (Rose, 1938), and normally it is ox!dized in the
body. Some. individuals whose metabolic processes are deranged excrete relatively large amounts of
abnormal products such as p-hydroxyphenylpyruvic acid ("tyrosinuria"), 2,S,-dihydroxyphenylacetic acid or
homogentisic acid ("alkaptonuria "), and phenylpyruv~c acid ("phenylpyruvic oligophrenia "). Garrod
(1923) has discussed these inborn errors of metabolism. The biochemical error in phenylpyruvic oligophrenials believed to result from a block in the hydroxylation of phenylalanine to tyrosine. Hall,
Sydenstricker, and Rawls (1948) and Lanyar (1942) have investigated the formation by alkaptonurics of
homogentisic acid from the optical isomers of phenylalanine and tyrosine. Cataracts and granulocytes
are produced in animals on phenylalanine-deficient diets (Hall, Bowles, Sydenstricker, and Schmidt,
1948). Analogues of phenylalanine inhibit the growth of vaccinia virus (Thompson and Wilkins, 1948),
Neurospora crass a mutants (Mitchell and Niemann, 1947), lactic acid bacteria (Beerstecher and Shive,
1947), other organisms (Dittmer, 1949), and the rat (Ferger and du Vigneau~, 1949).
Phenylalanine is to be synthesized by (a) condensing benzaldehyde ,*,ith hippuric acid in the
presence of acetic anhydride and sodium acetate and (b) reducing and hydrolyzing the resulting
azlactone (lactimide) by heating it with concentrated hydriodic acid and red phosphorus (equations below).
(0 OH
1
GH2
~H
I
/0:_
(ow)
NoAc
9'
11____ - ' ,~
GH,
~H ' .......O==:;,'c-H
I
I1
0-
/ 0 0-
-o~
GH-GH
If
~
~H
1
~H
l-O~
If
~
G=G
(-H20) ..
AczO
6 6 6
~H ~
-,
~ip_puric _acid
GOOH
~H-GH2-O'
H~
...
NH z
HI+ red P
phenylalanine
O-COOH
4- benzilidine2 -phenyl-5-oxazolone,
or azloclone
HP03
- 10-
NaT
EXPERIMENTAL
4-Benzylidine-2-phenyl-S-oxazolone. Place 16.4 g. (0.2 mole of anhydrous sodium acetate, 35.8 g.
(0.2 mole) of hippuric acid, 20 mI. (0.2 mole) of ben~aldehyde, and 72 mI. (0.77 mole) of acetic anhydride in a I-liter three-necked flask. Fit the flask with a 110 0 thermometer, a 3-ft. reflux condenser
with an attached calcium chloride drying ,tube, and fJ. Hershberg (1936) stirrer, as shown in Fig. 2.
F _ 110 0 thermometer
G-3-ft. condenser to which
a calcium
chloride drying
tube is attached
Heat the mixture on the water bath with vigorous stirring until the temperature rises rapidly to
75 0 owing to the heat of the reaction. Keep the temperature below 90 0 by cooling (ice water, if necessary) the flask. Note the formation of an orange solution and the deposition qf solid material on the
walls of the flask. Break up this cake with a stirring rod so that it will not insu~te the solution and
delay cooling. After precipitation appears to be complete, replace the ice bath with a hot;;water bath,
heat the bath to boiling, and continue heating at thi::; temperature for 1 hr. Allow the mixture to cool to
room temperature. Add distilled water slowly with stirring to decompose the excess acetic anhydride.
A violent reaction may ensue if the water is added too rapidly or in too large portions. Allow the mixture to cool to room temperature while stirring it at intervals, filter the suspension of yellow needlelike crystals, and wash the precipitate on the Buchner funnel with three sO-ml. portions of boiling
distilled water. Allow the solid to dry overnight at 50 0 in a tared, covered evaporating dish. Show the
product to the instructor. Calculate the percentage yield.
Phenylalanine. Place 35 g. (0.14 mole) of 4-benzilidine-2-phenyl-S-oxazolone, 20 g. (0.64 mole)
of red phosphorus, and 180 ml. (1.37 moles) of constant-boiling (57 per cent) hydriodic acid in a I-liter
round-bottomed flask. Fit the flask. with a 3-ft. reflux condenser to which is attached a rubber tube
leading to the sink. The purpose of this tube is to carry away any phosphine (PH 3 ) which may be
formed. Phosphine is poisonous; it is readily ignited; and when mixed with air, it is explosive.
Heat the mixture slowly and carefully to the boiling point with a gas flame, and allow the mixture
to reflux for about an hour until all the solid oxazolone dissolves to form a green solution. Remove the
flame and allow the solution to cool to room temperature. Add an additional 17.5 g. (0.70 mole) of the
,oxazolone slowly, ~o avoid excess heating which would cause the solution to boil over. Allow the
mixture to reflux continuously for 6 hr. Discontinue heating and add immediately, but slowly and
cautiously, 0.5 g. of solid carbon dioxide through the condenser. Add a total of about 10 g. of the
-ll-
carbon dioxide while observing these precautions. Sufficient gaseous catbon dioxide is liberated to
flush all the phosphine from the flask. Add 300 ml. of distilled "Yater, filter the suspension, and wash
the precipit~te of benzoic ac;id and (excess) ph?sphorus on the Buchner funnel with a total of 150 ml.
of distilled water.
Distill the combined filtrate and washings to- dryness, using th~ reduced-pre~sure distillation
apparatus shown in Fig. 7 (see Distillation in the Appendix). Aqd 50 ml. qf distilled water to the flask,
stir or heat the,suspension until the solid dissolves, add 5 g. of decolorizing' carbon, stir the mixture
for 5 min., and filter the suspension. Add 15 N ammonium hydroxide slowly with stirring to the filtrate
until [Link] is about 6, as tested (yellow color) with methyl red ineVcator. Allow the mixture to stand
overnight.
Filter the suspension on a Buchner funnel and apply suction until the p~oduct is nearly dry.
Dissolve the 'solid in boiling distilled water (20 ml. per g.), and add 5 g. of decolorizing carbon to the
solution (or suspension, if suspended material is present). Stir the mixt~re at the boiling temperature
for 5 min. and filter the suspension on a Buchner funnel which has been preh~ated by immersion in hot
water. Add slowly an equal volume of 95 per cent ethanol to the filtrate and allow the solution to cool
spontaneously to room temperature with frequent shaking. Place the [Link] in the ~efrigerator and
allow it to stand overnight or longer. Filter the suspension and wash the crystals twice on the Buchner
funnel with IS-mt. portions of 95 per cent ethanol. Allow the product to dry overnight in a covered
evaporating dish at 50. Submit the product to the instructor in a tared, labeled sample bottle. Calculate the percentage yield based on (a) hippuric acid and (b) the azlactone. $ubmit the calculations to
the instructor.
- 12-
EXPERIMENT
EXPERIMENTAL
Obtain three approximately I-g. samples of an unknown amino acid or amino acid derivative.
Obtain assignment to a Kjeldahl digestioh apparatus and a Kjeldahl distillation apparatus. Proceed
simultaneously with the analysis of the three samples.
Digestion of Sample. Transfer 1 g. (weighed to l mg.) of the amino acid sample quantitatively
to a 100-ml. volumetric flask, add about 25 ml. of 0.5 N sulfuric acid, and rotate the flask until the
sample dissolves. Add 0.5 N sulfuric acid to the mam and mix the liquids thoroughly. Pipette a
25.O-ml. aliquot of the amino acid solution into each of three 800-ml. Kjeldahl flasks. Pipette a 25-ml.
aliquot of the 0.5 N sulfuric acid into each of two 800-ml. Kjeldahl flasks. 1 Add' (from a graduate) to
each Kjeldahl flask 20 ml. of the digestion reagent (cupric sulfate and sodium 'sulfate in 18 N sulfuric
acid). Place the flasks on a Kjeldahl digestion rack and heat the mixture in each flask to Qoiling.
(Care should be taken that the fume hood connected with the digestion apparatus is operating efficient.
ly, since sulfur dioxide, irritating to the eye~ and throat, is evolved during the digestion process.)
Maintain gentle, but constant, boiling until the water is driven off, the brown color first formed disap-
- 13-
pears, and dense white fumes of sulfur trioxide appear in the flask. Continue gentle boiling for at
least 1 hr., at which time the solution should be colorless or colored pale blue-green. 2
Cool each flask in air and then to about 40 0 under running water. Add 350 mI. of distilled water
to the warm sblJltion in each flask to prevent caking of the white sodium sulfate which precipitates, and
rotate each flask until the liquids are thoroughly mixed. Allow each solution to cool to room temperature.
Oi sti !lotion of the Ammonia. Clean the distillation apparatus by distilling water from a Kjeldahl
flask until about 250 ml. of distillate have collected in the receiver. (This precaution may be omitted
if it is certain that the apparatus has been thoroughly steamed previously.)
,
Pipette 50.0 ml. of 0.1 N sulfuric acid 3 into each of the required number of 500-ml. filter flasks,
each of which is fitted with a short piece of rubber tubing on the side arm. Insert the side arm of each
filter fla'sk into the neck of a rubber balloon and tie the neck of the balloon to the side arm with stout
thread. 4 Add 4 to 6 drops of methyl red and p-nitrophenol mixed indicators's,'Olution to each filter flask.
Connect each flask to a delivery tube of the distillation apparatus by means' of a tightly fitting rubber
stopper. Dry each flask on the outside and dry 1 in. on the inside in order that the rubber stopper may
fit tightly. Open the3aucet which controls the flow of water to the condenser to permit a small, but
steady, stream of water to flow from the condenser. Tum on the switch which controls the electricity
to the heating coils (or light the gas burners).
Add ~bout 0.5 g. of granular zinc so that hydrogen liberated by reaction with the alkali will promote quiet boiling. Incline each flask in turn'at an [Link] of about 30 0 from the horizontal position.
Add 30 ml. of 18 N (saturated), carbonate-free sodium hydroxide from a graduate in such a manner that
the alkali does not come in contact with the inch of dry surface inside the flask. Hold the flask in this
position until the alkali runs down the inner wall of the flask and collects as a layer beneath the liquid. Connect_each flask in tum to the distillation apparatus and make certain that the flask fits tightly
on the rubber stopper. Rotate the flask gently until the acidic and basic layers mix smoothly and form
a homogeneous solution. Lower each flask until it rests on its heating unit.
When about half of the liquid has distilled, tum off the electricity (or gas) and allow 'the flask to
cool (5 to 10 min.). Remove the flask and discard the solution. Remove the receivers from the distillation apparatus and titrate each solution in tum with standard 0.1 N sodium hydroxide measured from
a SO-ml. burette.
NOTES
1. Omit this step if the blank value is furnished by the instru9tor. Only one blank determination is required
for a particular set of reagents.
2. Unknowns will be limited to amino acids which digest within 2 hr. If digestion is continued for a longer
time, additional (S to 10 mI.) of digestion, reagent' iiIay be required to replace the sulfuric acid (as sulfur dioxide)
and water lost by vaporization.
3. The same 0.1 N sulfuric acid should be employed throughout the experiment. It is unnecessary that the
exact normality of this acid be known, since the ammonia in the sample is calculated from the difference 'in volumes of standard base used to titrate the excess acid in the blank and unknown solutions.
4. In conventional Kjelciahl distillation apparatus the delivery tube extends into the acid in the receiver.
At the start of distillation small quantities of ammonia are trapped in bubbles of air which pass through the solution. For this reason ammonia is lost, and the analysis js inaccurate. Bradley (1942) has .eliminated this error by
attaching a rubber balloon to the side arm of the receiver. It was observed that the pressure [Link] by distilling into a closed system was insufficient to burst the balloon and that there was no water condensate on thq' walls
of the balloon which would absorb ammonia. Under these conditions 99.9 per cent of the ammonia was reccjvered.
Analytical results of comparable accuracy are attainable by distilling the ammonia from an all-glass appar~tds by
a micro or semimicro procedure.
5. Solutions of mixed indicators are advantageous in titrating ammonia with a strong acid, sinc;e the color
change at the end point is sharper and more distinct than that obtained with a single indicator. Mixed indicators
consisting of methyl red and methylene blue, methyl red and guinea red, and other combinations have been described by johnson and Green (1930).
-14 -
EXPERIMENT
10
ANALYSTS OF AN AMINO ACID BY THE VAN SLYKE VOLUMETRIC NITROUS ACID METHOD
Nitrous acid procedures for the determination of amino nitrogen by volumetric,mac;ro- and microand manometric methods have been described by Van Slyke (1911, 1912. 1913-1914, 1915b, 192%).
The simple aliphatic alpha-amino acids react quantitatively with nitrous acid in about 4 min. at
room temperature. The time for, complete reaction of beta-, gamma-, delta-, and epsilon-amino acids
increases'directly with the distance of the amino from the carboxyl group (Dunn and Schmidt, 1922).
Ab~ut 30 min. is required for the complete reaction with lysine. Arginine, tystine, glyci~e, histidine,
tryptophan, an,d serine yield higher than the theoretical amount of nitrogen, particularly at elevated
temperatures (Schmidt, 1929). These amino acids give extra nitrogen because nitrous acid is reduced
by intermediate products as well as by the original amino acid.
Kendrick and Hanke (1937, 1940) reported that these abnormalities were, prevented by introducing
pota!:?sium iodide into the deaminizing mixture (manometric apparatus). Preferential oxidation by iodine
rather than by nitrous acid was assumed to explain the iodide effect, but Dunn and Porush (1939) proposed that an unreactive mercuric iodide complex was formed. Compounds such as ammonia, urea,
asparagine, creatinine, and purines react slowly with nitrous acid.
An unknown amino acid is to be analyzed by the Van Slyke volumetric macro nitrous acid procedure. Nitrous acid, formed by mixing sodium nitrite solution and 3 N hydrochloric acid, r~acts with the
amine solution in the deaminizing chamber to form a hydroxy compound, nitric oxide, water, and nitrogen:
NaNO, + HCI _. NaCI + HNO,
RNHa + HN0 2
--
ROH + Na + H 20
The gaseous products are shaken with alkaline permanganate solution, which oxidizes the nitric oxide
to potassium nitrate.
NO + KMn0 4
-+
KNO. + MnO a
The volume of residual nitrogen is measured in a gas burette at atmospheric pressure and room tempet'dr"
ture. The weight of nitrogen in the sample is calculated from these data by applying gas-law principles
or, more conveniently, by employing the conversion factor in the table of factors prepared by Van Slyke.
EXPERIMENTAL
Obtain approximately 0.5 g. of an unknown amino acid and assignment to a volumetric macro\Van
Slyke apparatus (shown in Fig. 3). Transfer 0.5 g. (weighed to 1 mg.) of the sample quantitatively to
a 25-ml. volumetric flask, add 10 ml. of 0.5 N sulfuric acid, ~d rotate the flask until the sample dissolves. Add distilled water to the mark and mix the liquids thorougJIly.
Adjustment of the Apparatus. The glass parts should be clean [Link] tightly with heavywalled soft-rubber tubing. The stopcocks should be well lubricated, turn easily, and be attached to the
apparatus with rubber or wire. The lower bulb and the tube leading to the upper bulb of the Hempel
Pipette should be filled with alkaline potassium permanganate solution saturated with nitrogen (shaken
- 15-
with air for about a minute). The upper bulb should contain only a few milliliters of the permanganate
'reagent. The gas burette should be filled and the level bulb about one-third filled with 0.5 N sulfuric
acid.
h
a
2.
Fig. 3. VAN SL YKE MACRO VO._l..UMETRIC AMINO NITROGEN APPARATUS.
Place bulb E in position 2, open'stopcock e to the Hempel pipette H, and allow the permanganate
solution'to fUJ the bore of stopcock e. 1 Close stopcock e, raise b-qlb E'to position 1, tum stopcock d
so that it connects with exit tube g, open stopcock e to connect with stopcock d, and allow the sulfuric::
acid to flow until it fills the bore of stopcock d. Close stopcock e. Make certain that no air bubbles
are present in the capillary tubes.,
Removal of Air from Reagents and Deaminizing Chamber. Tum stopcock d to connect the deaminizing chamber B with exit tube g. Close stopcocks a, b, ,and c. Add sufficient (about 8 m!.) 3 N
hydrochloric acid to fill funnel A to mark h. Open stopcock a and admit the aciq solution into the deaminizing chamber B. Close stopcock a and add 30 per cent. sodium nitrite solution sufficient (about
33 ml.) to fill funnel A to mark i. Open stopcock a, allow all but 1 or 2 m!. of the sodium nitrite
solution to run from funnel A into deaminizing chamber B, and close stopcock d.
Agitate the deaminizing chamber with the motor at a uniform rate such that nitric oxide is generated rapidly with no strain on the apparatus. Although the rate of shaking is unimportant, the same
rate should be employed for all determinations. Continue the shaking until the liquid is forced down to
mark i, stop the motor, and open stopcock d to permit the gas '(nitric oxide and residual air) to escape
through g. When nearly all the solution has run from A into B, close stopcock d. Tum on the motor
- 16-
and allow the deaminizing chamber to shake until the liquid has been forced down to i. Close stopcock
a. Turn stopcocks d and e so that the deaminizing chamber B is connected to gas burette D.
Reaction of Amino Acid with Nitrous Acid. Transfer quantitatively 5.00 mI. of the unknown amino
'acid solution from the 25-ml. volumetric flask to burette C by means of a 5-ml. pipette. Place bulb E
in position 2. Turn stopcock c so that burette C is open into B. Admit all but about O.2_ml. of the
amino acid solution into B. Close stopcock c and rinse burette C with appro'ximately 4 mI. of distilled
water delivered from a wash bottle. Turn' stopcock c so that all but about 0.2 ml. of the liquid is admitted to B. Close stopcock c and add I ml. of distilled water to C. Turn stopcock c so that all but
about 0.2 ml. of the l~quid is,admitted to B. Turn on the motor and allow the deaminizing chamber to
shake for 5 min. at the rate employed previously. Open stopcock a and allow the liquid in A tQ run into
B until all the gas above the liquid in B has been forced into burette D. Close stopcock e. 2
.
Removal of Nitric Oxide and Mea'surement of Nitrogen. Place bulb E 1n position 1. Open stopcock
e so that qurette D is connected with the Hempel pipette H. When all the gas has been forced into the
low7 r bulb'of H, close stopcock e. Agitate the Hempel pipette H slowly for 5 min. (with the motor or by
hand). Lower puJb E to position 2, turn stopcock c to connect Hand D, and/when the unabsorbed gas
in H has been forced past the bore of stopcock e into D, close stopcock e. Allow the burette to drain
for J min. Lowe): bulb E to a position close to burette D such that the meniscus of the liquid in the
~urette and the surface of the liquid in bulb E are at the same height. Read the volume of the gas in
burette D to 0.01 ml. Return the gas in D to H by the technique previously described, agitate the
Hempel pipette for 1 min., return the gas to D, arid read the voll,lme of the gas in D. If the observed
volumes do not differ by more than 0.02 mI., it can be assumed that the nitric oxide has lleen completely
absorbed. if the volumes differ by more than this amount, repeat the .absorption procedure.
Read the thermometer and the barometer. Record the temperature and the corrected pressure (see
Pressure in the Ap~ndix).
When the analysis has been completed, open stopcock b and allow the liquid to drain from B.
Rinse A, B, and C with distilled water. Analyze a second 5-ml. aliquot of the unknown amino acid
solution. The volumes of nitrogen gas should agree within 0.05 ml. If necessary, perform additional
analyses until the volumes of nitrogen in two consecutive analyses agree within 0.05 mI.
Determination of Nitrogen from Reagents. Repeat the described procedures with S.O-mt. volumes
of the 0.5 N sulfuric acid until the volumes of nitrogen gas agree within 0.05 ml. Reagents suitable
for the analysis of amino acids should give not more than 0.4 mI. of residual gas.
Conversion Factor. See the accompanying table for factor with which to convert the corrected
volume of nitrogen to weight of nitrogen.
MILLIGRAMS OF AMINO NITROGEN CORRESPONDING TO ONE MILLILITER OF NITROGEN GAS
Corrected pressure, mm. Hg
Ternperature,
730
732
734
736
738
740
742
744
16
17
18
19
20
0.5570
0.5540
0.5515
0.5490
0.5460
0.5585
0.5555
0.5530
0.5505
0.5475
0.5600
0.5575
0.5545
0.5520
0.5495
0.5615
0.5590
0.5560
0.5535
0.5510
0.5630
0.5605
0.5580
0.5550
0.5525
0.5645
0.5620
0.5595
0.5565
0.5540
0.5660
0.5635
0.5610
0.5580
0.5555
0.5675
0.5650
0.5625
0.5595
0.-5570
21
22
23
24
25
0.5435
0.5410
0.5380
0.5350
0.5325
0.5450
0.5425
0.5395
0.5365
0.5340
0.5465
0.5440
0.5410
0.5380
0.5355
0.5480
0.5455
0.5425
0.5400
0.5370
0.5495
0.5470
0.5440
0.5415
0.5385
0.5510
0.5485
0.5455
0.5430
0.5400
0.5525
0.5500
0.5470
0.5445
0.5415
0.5540
0.5515
0.5485
0.5460
0.~430
26
27
28
29
30
0.5295
0.5'265
0.5235
0.5210
0.5175
0.5310
0.5280
0.5250
,0.5220
0.5190
0.5325
0.5295
0.5265
0.5235
0.520?
0.5340
0.5310
0.5280
0.5250
0.5220
0.5355
0.5325
0.5295
0.5265
0.5235
0.5370
0.5340
0.5310
0.5280
0.5250
0.5365
0.5355
0.5325
0.5295
0.5265
0.5400
0.5370
0.5340
0.5310
0.5280
-17 -
748
750
752
754
756
758
760
16
17
18
19
20
0.5690
0.5665
0.5640
0.5610
0.5585
0.5710
0.5680
0.5655
0.5630
0.5600
0.5725
0.5695
0.5670
0.5645
0.5615
0.5740
0.5710
0.5685
0.5660
0.5630
0.5755
0.5730
0.5700
0.5675
0.5645
0.5770
0.5745
0.5715
0.5690
0.5660
0.5785
0.5760
0.5730
0.5705
0.5675
21
22
23
24
25
0t'\555
0.5530
0.5500
0.5475
0.5445
0.5575
0.5545
0.5515
0.5490
0.5460
0.5590
0.5560
0.5530
0.5505
0.5475
0.5665
0.5575
0.5545
0.5520
0.5490
0.5620
0.5590
0.5560
0.5535
0.5505
0.5635
0.560;;
0.5575
0.5550
0.5520
0.5650
0.5620
0.5595
0.5565
OJ5535
0.5665
0.5635
0.5610
0.5580
0.5550
26
27
28
29
30
0.5415
0.5385
0.5355
0.5325
0.5295
0.5430
0.5400
0.5370
0.5340
0.5310
0.5445
0.5415
0.5385
0.5355
0.5325
0.5460
0.5430
0.5400
0.5370
0.5340
0.5475
0.5445
0.5415
0.5385
0.5355
0.5490
0.5460
0.5430
0.5400
0.5370
0.5505
0.5475
0.5445
0.5415
0.5385
0.5520
0.5490
0.5460
0.5430
0.5400
0.5800
0.5775
0.5745
0.5720
0.5690
a The figures in the table are the milligrams of nitrogen per milliliter of nitrogen gas liberated from the
amino acid. The weight of nitrogen actually liberated i'l twice that indicated, since half the gas is derived
from the amino acid and half from the nitrous acid.
NOTES
1. The alkaline permanganate solution should be in contact wijll stopcock e, as short a time as possible,
since it is attacked by the alkali and discolored by the manganese dioxide. When the apparatus is not in use, force
the permanganate solution into the Hempel bulbs. As a further safeguard, run the 0.5 N sulfuric acid solution through
the bore of stopcock e. The efficiency of the permanganate solution as an oxidant for nitric oxide diminishes with
use, as may be noted by the accumulation of manganese dioxide in the Hempel pip~tte. For this reason, record
the number of determinations and, after 10 runs, drain the permanganate solution from the Hempel pipette, rinse
I
the pipette and connecting tubes with 0.5 N sulfuric acid, and rinse them finally with distilled water. Refill the
Hempel pipette with the permanganate solution.
2. If more than 100 ml. of gas collects in the ga~ burette D, place bulb E in position 1 to prevent displacement of all the liquid in D, turn stopcock e to connect D with the Hempel pipette, and permit about 20 mI. of the
gas to enter the Hempel pipette. Turn stopcock to connect D with the deaminizing chamber B, and lower bulb E
to position 2. These manipulations should,l:)e performed as quickly as possible to minimize the danger and loss
of gas caused by the pressure developed in B while the latter is a closed system.
-18 -
EXPERIMENT
11
EXPERIMENTAL
Obtain an approximately I-g. sample of each of three unknown amino acid or peptide hydrochlorides. Transfer 1 g. (weighed to 1 mg.) of one of the samples quantitatively to a 100-ml. volumetric
flask. Add 50 ml. of distilled water, rotate the flask until the sample dissolves, add 2 mi. of concentrated nitric acid, add distilled water to the mark, mix the liquids'"thoroughly, and stopper the flask.
Prepare solutions of the other samples similarly.
Pipette three 25.O-ml. aliquots of One of the unknown solutions into separate 10O-ml. glass-stoppered bottles (or flasks). Add 5 ml. of reagent-grade nitrobenzene to each bottle. Add standard, ap_
proximately 0.1 N, silver nitrate solution from a 50-ml. burette until no further precipitation of silver
chloride is observed. Stopper and rotate the bottle to coagulate the silver chloride suspension, allow
the suspension to settle, and add 1 drop of the standard silver nitrate solution. If no additional precipitation occurs, add 2.00 mI. of the standard nitrate solution. Stopper the bottle and shake it vigorously for 30 sec. Rinse the stopper and inside neck of the bottle with distilled water. Add 1 ml. of
ferric alum indicator solution. Titrate the excess silver ions with standard, approximately 0.05 N, potassium thiocyanate solution to a faint-pink color which is permanent for at least 30 sec. Repeat this
procedure using the other unknown solutions.
- 19-
EXPERIMENT
12
Tyrosine was first isolated fro.m casein by Liebig (1846, 1847). Since the alkali-fusion procedure
employed for the liberation of tyrosine was not satisfactory, proteins were ,hydrolyzed with hydrochloric
acid by Bopp (1849) and with s';llfuric ~cid by Hint~l"berger (1849). Later investigators have liberated
tyrosine from proteins by the ,action of microorganisms (Emmerling, 1897) and by enzymatic digestion
with pepsin (Borkel, 1903), trypsin (Cohn, 1895), and p~pain (Emmerling, 1902; Bergmann and Niemann,
1938).
Crude tyrosine can be readily isolated from~roaln hydrolysates because of its low solubility at
the isoelectric point, pH 5.7. It may be purified b crystallization from water, although large volumes
of water and large containers are required. Dilu ethanol was employed by Sopp, but this solvent is
unsatisfactory. Extraction of the leucine from f e crude tyrosine was proposed by Habermann and
Ehrenfeld (1902), and the purification of tyrosine by precipitating it as its phosphotungstate (Sjollema
and Rinkes, 1911), ethyl ester hydrochloride (Plimmer, 1913), or mercuric chloride complex (Hanke,
1925) has been suggested.
The best source of tyrosine is silk, whichcontains from 15 to 30 per cent of a gummy protein,
sericin, and from 70 to 85 per cent of a fibrous protein, silk fibroin. The tyrosine content of 'silk fibroin
from Chinese, Indian, Italian, Japanese, and Russian sources varies from 3.5 to 13 per cent. Casein 6
contains about 6 per cent tyrosine.
L-Tyrosine i; to be isolated by precipitating it at the isoelectric point from the hydroch-loric acid
hydrolysate of casein or silk.
EXPERIMENTAL
Add 100 g. of silk waste or commercial casein and 300 ml. of 8 N hydrochloric acid to a 2-liter
round-bottomed flask. Place the flask in a sand bath and attach a reflux ,condenser. Heat the bath
sufficiently to dissolve the protein a:nd cause the acid solution to boil gently. Continue gentle boiling
for 12 hr.
Filter the hot, dark-colored hy~rolyspte on a Buchner funnel with the aid of celite. Wash the
precipitate on the funnel with three 20-mf. portions of boiling distilled water. Transfer the filtrate and
washings to a I-liter round-bottomed flask and set up the reduced-pressure distillation apparatus shown
in Fig. 7 (see Distillation in the Appendix). Distill the solution under reduced pressute on a water bath
to a sirup. Add 100 ml. of hot distilled water and distill the solution under reduced pressure on a water
bath to a sirup. Immediately, add 250 ml. of hot distilled water and stir the mixture until it is homogeneous. Heat the solution to boiling, remove the flame, add 10 g. of decolOrizing carbon in small
portions ,with occasional stirring. Add 5 g. of celite and filter the suspension on a Buchner funnel.
Wash the precipitate on the funnel with two 25-ml. portions of boiling distilled water.
Place the combined filtrate and washings in a l.5-liter beaker-and add sufficient (about 375 ml.)
3 N sodium hydroxide to bring the pH to 5.7. Determine the pH of the solution by test with methyl red
indicator, which changes color from red to yellow over the pH range 4.4 to 6.0. Determine the shade of
color to be expected by testing a buffer solution at pH 5.7. Allow the beaker (covered and labeled) to
stand overnight in the refrigerator. Filter the suspension on a Buchner funnel and wash the precipitate
on the funnel with two 25-ml. portions of distilled water.
Transfer the preCipitate to a 600-ml. beaker and add 300 mI. of distilled water. Add 3 N sodium
hydroxide slowly from a dropping funnel with stirring until the solid dissolves. Filter the solution on.a
-'20 -
Buchner funnel to remove any suspended material. Transfer the filtrate to a 600-ml. beaker and add
sufficient {volume same as that of .the 3 N sodium hydroxide added previously) 3 N hydrochloric acid to
bring the solution to pH 5.7. Immerse the beaker for 30 min. in an ice-w;lter bath or allow it to stand
... ~emight in the refrigerator. Filter the suspension on a Buchner funnel and wash the precipitate on the
.funnel with 25 ml. of distilled water.
Dissolve the wet product in the minimum volume (200 ml. per g.) of boiling distilled water in a
large beaker. The ra~e of solution may be IS low . If suspended material is present, filter the solution
apd allow the filtrate to stand overnight 'in a beaker (covered and labeled) in the refrigerator. Filter
the suspension and wash the crystals on the funnel with 25 mI. of distilled water and two 2S-ml. portions
of methanol. D-ry the product in an evaporating dish (covered and labeled) at 50. Submit the product 1
to the instructor in a tared, labeled sample bottle.
NOTE
1. An additional crop (0.5 g.) may be obtained by evaporating the combined filtrate and washings to a smal1
volume and recrystallizing the resulting crude tyrosine from the minimum volume of distilled water.
- 21-
EXPERIMENT 13
PHOTOMETRIC ANAL YSIS OF TYROSINE BASED
The Millon procedure, first described {?y Weiss (1919), has been employed by Folin and Marenzi
(1929a) and other investigators. Tyrosine has also been determined by an iodometric bromination
method (Plimmer and Phillips, 1924) and by the photometric analysis of the blue-colored complex
fonned with a phosphotungstic acid-phosphomolybdic acid reagent (Looney, 1926) and of the red- to
yellow-colored complex with diazotized sulfanilic acid (FUrth and Fleishmann, 1922; Hanke, 1928).
The purity of the tyrosine prepared in Exp. 12 is to be determined by a modification of Arnow's
(1937) photometric Millon method. The constr~cya6and operation of a visual colorimeter are described
in the Appendix (1""'" Phn~ometry) and will be rmonstrated by the instructor.
EXPERIMENTAL
Obtain three L-tyrosine unknowns and assignment to a visual colorimeter. Transfer approximately
0.25 g. (weighed to 1 mg.) of each sample quantitatively to a SOQ..ml. volumetric flask. Add 100 ml.
of 0.5 N sulfuric acid to each flask and rotate each flask until the sample dissolves. Add 0.5 N sulfuric
acid to the mark and mix the liquids in each flask thoroughly.
Pipette I-mI. aliquots of the t4ree unknown solutions and the standard solution, containing approximately 0.5 mg. per mI. of purified L-tyrosine, to four separate 6-in. test tubes. Add 5 ml. of 15 per
cent mercuric sulfate in 6 N sulfuric acid to each tube with the aid of a, l()..ml. graduate. Mix the
solutions in each tube and immerse the tubes for 10 min. in a bOiling water bath. Remove the tubes and
allow them to cool to room temperature. Place a small funnel in a 25-ml. volumetric fla,sk and transfer
one of the solutions quantitatively to the flask. Rinse the funnel with distilled water, add 5 ml. of 0.2
per cent sodium nitrite solution, add distilled water to the mark, and mix the liquids thoroughly (flask
should be stoppered the minimum time, since nitric oxide is evolved). Transfer each of the other'
solutions in turn to a 25-ml. volumetric flask as described and treat each'solution Similarly. After the
solutions have stood for 5 min., compare their color intensities in turn in a visual colorimeter with that
of the standard, which has been developed simultaneously.
- 22-
EXPERIMENT
14
The reaction of phenols with a mixture of mercurous and mercuric nitrites to give red-colored
products was discovered by Millon (1849) and was first applied to tyrosine by Hoffman (1853). Gibbs
(1926, 1927) and Lugg (1938) have discussed this test.
Values ranging from 4.5 to 6.5 per cent tyrosine in casein have been reported. Fischer (1901)
found 4.5 per cent by isolation, and Osborne and Guest (1911) considered this percentage to be the most
reliable. Folin and coworkers obtained values from 5.3 to 6.4 per cent by photometric determination
with their phenol (phosphotungstic acid-phosphomolybdic acid) reagent (Folin and Denis, 1912) and with
their Millon's reagent (Folin and Ciocalteu, 1927; Folin and Looney, 1922). Values ranging from 5.5 to
6.0 per cent have been found more recently by photometric (Beach et al., 1941; Williamson, 1944) and
microbiological (Gunness, Dwyer, and Stokes, 1946) assays. Tyrosine has also been determined by
spectrophotometri'c (Goodwin and Morton, 1946), chromatographic (Moore and, Stein, 1949; Stein and
Moore, 1949; Tristram, 1946)~> and L-amino acid decarboxylase (Hanke, 1948) methods.
Tyrosine in casein is to be determined by Block and Bolling's (1945) modification of the MillonFolin method. The protein is hydrolyzed with sodium hydroxide rather than acid, which partially
destroys tyrosine. Tryptophan is removed as its insoluble mercuric sulfate complex, since it gives an
interfering colored product with Millon's reagent. The construction and operation of a visual colorimeter are described in the Appendix (see Photometry) and will be demoqstrated by the instructor.
EXPERIMENTAL
Obtain approximately a O.3-g. sample of an unknown casein and assignment to a visual colorimeter. Transfer 0.3 g. (!"eighed to lmg.) of the casein quantitatively to a 50-ml. conical flask, add
10 ml. of 5 N sodium hydroxide,' attach a reflux condenser, place the flask in a melted hydrogenatedcottonseed-oil bath, heat the bath to 120 0 , and maintain this temperature for 5 hr.
Remove the flask and cool it to room temperature. Add 10 ml. of 5 N sulfiuic acid and transfer
the mixture quantitatively to a 500-ml. trolumetric flask. Add distilled water to' the mark and mix the
liquids thoroughly. Pipette 10.0 ml. of this solution into a 50-ml. centrifuge tube (:4) and 5.00 ml. into
a 50-ml. centrifuge tube (B). Submit B to the instructor, who will add 5.00 ml. of a standard solution
ot pure L-tyrosine. Pipette into separate 8-in. test tubes 5.00 ml. of standard L-tyrosine solutions containing 0.500 mg. (C) and 0.800 mg. CD) of L-tyrosine, respectively. Add 5.0 ml. of distilled water to each tube.
Add (from a 10-ml. graduate) to each of the four tubes 10 ml. of distilled water and 5.0 ml. of
Folin's reagent (mercuric sulfate in sulfuric acid). Place the tubes in a boiling water bath. After 10
min. remove the tubes and cool them to room temperature. Add about 15 mg. of filter aid to each of the
centrifuge tubes A and B, and centrifuge the tubes at moderate speed for 5 min. to separate the insoluble tryptophan-mercuric sulfate complex. Carefully transfer the supernatant liquids containing the
soluble tyrosine-mercuric sulfate complex to separate 50-ml. volumetric flasks. Wash each precipitate
as follows: Add 5 ml. of a solution of 1.5 per cent mercuric sulfate in 6 N sulfuric acid to each centrifuge tube, stir the precipitates thoroughly, and centrifuge the suspensions for 5 min. Add each supernatant solution to its corresponding volumetric flask. Repeat the washing process using 3 ml. of the
mercuric sulfate-sulfuric acid reagent. Add each wash liquid to its respective volumetric flask. Transfer the contents of tubes C and D to separate 50-mt. volumetric flasks and add 8 ml. of the mercuric
sulfate-sulfuric acid reagent to each flask.
Pipette 1.0 ml. of 0.8 per cent sodium nitrate solution into each of the four volumetric flasks and add
distilled water to the marks. Mix the liquids in each flask (stoppered the minimum time, since nitric oxide is
evolved), and allow the solutions to stand for 10 min. Compare (within 30 min.)the intensities of the
colors in the unknown solutions with that of the standard tyrosine which matches the unknown more closely.
- 23-
EXPERIMENT 15
PREPARATiON OF TYRAMINE HYDROCHLORIDE
Tyramine causes a rise in blood pressure, and it increases the force of the heart beat, although
its action is less pronounced than that of adrenaline. That tyramine may originate in animal tissues
from bacterial action is indicated by the observation that tyrosine is decarboxylated by Escherichia
coli (Sasaki, 1914) and Proteus vulgaris (Kawai, 1916). Tyrosine decarboxylases are present in milk,
blood, urine, yeast, and microorganisms (Gale an~s, 1945).
Tyramine is to be prepared by thermal decarboxylation of tyrosine, essentially by Johnson and
Daschavsky's (1924-1925) modification of Graziani's (1915) method. Tyrosine is heated in a mixture
of high-boiling solvents until the evolution of carbon dioxide ceases. Solvents which have been
employed, in addition to diphenylamine (Abderhalden and Gebelein, 1926) and diphenylmethane, include
glycerol (MaIllard, 1914a), paraffin, anthracene, and a-naphthyl amine (Johnson and Daschavsky, 19241925). Satisfactory analogous procedures for other amines have not been reported.
EXPERIMENTAL
Place 10.0 g. of technical tyrosine, 60 g. of t~chnical diphenylmethane, and 60 g. of technical
diphenylamine in a 250-mY. conical flask. Place the flask in a sand bath and heat the bath while
stirring the mixture (carefully, to avoid breaking the thermometer) with a 360 0 thermometer: Carbon
dioxide is evolved at about 210,. Continue heating and stirring (for about 15 -min.) until the evolution
of gas ceases. Remove the flask and allow it to cool to.,room temperature. Add 100 ml. of diethyl
ether, stir the mixture thoroughly, and filter the suspension of tyramine on a Buchner funnel. Transfer
the filtrate to a bottle provjded for that purpose. Repeat this extraction process twice, using 50-mI.
portions of ether. Transfer the tan-colored precipitate to a 250-ml. conical flask, add 75 mI. of absolute
ethanol, and heat the mixture on a water bath until the solid dissolves. Remove the flask and add 5 g.
of decolorizing carbon. Place the flask on the ~ater bath and heat the mixture for 5 min. yvhile stirring
it at intervals. Filter the hot suspension on a'Buchner funnel and repeat the decolotizing process until
the filtrate is colored light tan. Trans'(er th~ filtrate to a beaker, add 5 mI. of 12 N hydrochloric acid,
stir the mixture thoroughly, and add 75 ml. of diethyt" ether. Stir the mixture thoroughly and allow it to
stand for 5 min. (on longer stan~ing a blue or blue-green impurity forms [Link] difficult to remove).
Filter the suspension of nearly white crystalline tyramine hydrochloride on a Buchner funnel and wash
the precipitate twice with 10-ml. portions of diethyl ether. Redissolve the product in the minimum
volume (about 35 mI.) of hot absolute ethanol, cool the solution for 30 min. in an ice-water bath, filter
the suspension on a Buchner funnel, and wash the precipitate on the funnel with 15 ml. of diethyl ether.
Transfer the product to a watch glass and allow it to dry in air. Submit the product to the instructor in
a tared, labeled sample bottle.
- 24-
EXPERIMENT 16
ISOLATION OF L-CYSTINE FROM HUMAN HAIR
L-Cystine has been isolated from hair and other keratins as its mercuric sulfate comple:(Hopkins and Cole, 1901b), copper salt (Harris,1922-1923), phosphotungstate (Winterstein, 1901-1902),
hydrochloride (Toennies and Bennett,'193S-1936), and cuprous mercaptide derivative (Vickery and
White, 1932; Lucas and Beveridge, 1940). Tyrosine, the principal impurity, has been removed from
crude cystine by precipitating it at its isoelectric point (PH S.7) (Greenbaum, 1935) and by extracting
tyrOSine ethyl ester from the est~rified mixture (Plimmer, 1913).
The cystine content of keratins from different animals was determined by Wilson and Lewis (1927),
who precipitated cystine at its isoelectric point and determined cystine in the crude product by the
Folin and Looney (1972) photometric method and on the basis of the total sulfur content. Values by
the two methods were in good agreement. The percentages of cystine found were human hair, 15.6 to
21.2; sheep'wOoI, 8.0 to 19.0; feathers, 7.05 to 12.2; rabbit hair, 11.9 to 14.0; tortoise shell, 6.4 to 8.1;
rat hair, 14.1; cat hair, 13.1; and dog hair, 19-.0. Cystine has been isolated from hog hair (Greenbaum,
1935) and horsehair (Okabe, 1928). According to Clay, Cook, a~d Routh (1940), there is no constant
relation between the age of the individual and the percentage of cystine in the hair. Dark hair was
found to contain more cystine than light hair and hair of males more than that of females.
Vickery and Leavenworth (1929) obtained more than 10 per cent of analytically pure L-cystine
from human hair, but yields usually are only about 5 per cent: Lucas and Beveridge (1940) reported
14.8 per cent, but the purity of the product was not established. Unless the product is carefully purified, considerable amounts of sulfate, sulfinic and sulfonic acids, DL-cystine, and mesocystine may
be present.
L-Cystine is to be isolated from the hydrochloric acid hydrolysate of purified human hair,
essentially by the method first described by Morner (1899) and modified by Folin (1910), Gortner and
Hoffman (1941), and Vickery and Leavenworth (1929). L-Cystine is precipitated at (or near) its iSDelectric point (pH 5.0) and is purified by precipitating it with ethanol at the isoelectric point. The
yield of analytically pure product is about 5 per cent of the weight of the dry purified hair. Because of
its low solubility (libout 0.1 g. per 100 mI. at 100) it is impracticable to purify L-cystine by crystallization from water. Tyrosine may be removed by trituration with hot water, but this procedure is not
entirely satisfactory since the solubility of tyrosine is only about five times that of cyst~ne.
EXPERIMENTAL
Obtain about 600 g. of men's hair, remove the matches and other debris, and suspend the picked
hair in 1.5 liters of commercial cleaning solvent (or petroleum ether, b.p. 60 to 100). Stir the hair
thoroughly at intervals for 30 min. Place a layer of cheesecloth in a funnel, transfer the hair to the
funnel, and allow the solvent to drain. Resuspend the hair in 1.5 liters of the solvent, stir the hair at
intervals for 10 min., and drain as before. Repeat this process until the hair appears to be thoroughly
clean. Spread the clean hair on paper, preferably in the sun, until the solvent has evaporated.
Weigh 400 g. of the clean, dry hair and transfer the hair to a 2-liter round-bottomed flask. Add
1 liter of 8 N hydrochloriC acid, attach a reflux condenser, and place the flask in a sand bath: Heat the
sand sufficiently to cause gentle boiling and maintain this temperature for 8 hr. Longer heating may
reduce the yield of L-cystine. Fit a Buchner funnel with filter cloth (Saran, Dow Chemical Co. product
is satisfactory), add a layer of celite and filter the hot suspension of hu~in. Wash the precipitate on
the funnel with two 1000ml. portiol}s of boiling distilled ,water.
- 25-
Transfer the combined filtrate and washings to a 2-liter round-bottomed flask, set up the reducedpressure apparatus shown in Fig. 7 (see Distillation in the Appendix), and distill the solution to a
sirup under reduced pressure on a water bath. Add 700 ml. of boiling distilled water, stir the mixture,
heat the solution to boiling, remove the flame, add 20 g. of decoloriz'ing carbon in small portions while
stirring, and allow the suspe~sion to stand---fortS-min. Filter the suspension on a Buchner funnel and
wash the carbon residue on the funnel with two 20-ml. portions of bojling distilled water. If the filtrate
is dark colored, repeat the decolorizing process, using ~O g. of decolorizing carbon. The final color of
the filtrate should not be darker than light amber.
Transfer the filtrate to a beaker and add 6 N sodium hydroxide dropwise with constant stirring to
pH 3.5 1 (tyrosine crystallization is minimum at this low pH). Determine the pH with congo red paper,
which changes color from blue to purple over the pH rahge 3 to 4. Determine the shade of color to be
expected by testing a buffer solution at pH 3.5. Allow the solution to stand fpr 30 min. in an ice-water
bath or overnight in the refrigerator.
Filter the suspension and wash the gray-colored precipitate on the Buchner funnel with two 50-mt.
portions of boiling distilled water. Discard the washings. Suspend the crude cystine in 100 ml. of
boilin-g distil1~d water and add 12 N hydrochloric acid (about 40 ml.) until the precipitate dissolves.
Heat the solution to boiling, remove the flame, add 15 g. of decolorizing carbon in small portions while
stirring the mixture, and allow the suspension to stand for 10 min. Filter the hot suspension on a
Buchner funnel and wash the carbon residue on the funnel with two 2s..ml. portions of hot N hydrochloric
acid. Filter the combined filtrate and washings to remove traces of carbon. Transfer the filtrate to a
beaker and ad4_ 6 N sodium hydroxide dropwise to pH 3.5, determined as previously described. Allow
the solution to stand for 30 min. in an ice-water bath or overnight in the refrigerator.
Filter the crystalline suspension on a Buchner funnel and wash the precipitate on the funnel with
two 25-ml. portions of boiling distilled water. Test the final wash liquid with Millon's reagent (see
Exp. 21, Step XI, Test 5), -and if the test for tyrosine- is positive, continue washing with additional
portions of boiling distilled water until the test is negative.
Resuspend the cystine in 100 mI. of boiling distilled water and add 12 l'{ hydrochloric acid dropwise with stirring until the solid dissolves. If the solution is colored, decolorize it with 10 g. of
decolorizing carbon as previously described. Filter the suspension and wash the precipitate on the
funnel with 25 mI. of hot N hydrochloric acid. Add 6 N sodium hydroxide dropwise to the combined
filtrate and washings to pH 4.5 (tested as previously described), and add an equal volume of 95 per,
cent ethanol. Place the solution in an ice-water bath for 30 min. or overnight in the re{rigerator.
Filter the suspension on a Buchner funnel and wash the crystals with distilled water unt_il the
test for chloride ion (white pre,cipitate of silver chloride) is negative (1 ml. of the AgNO.-HNO. reagent
added to 1 mI. of the filtrate). Wash the crystals with two ~ml. portions of,95 per cent ethanol and
/'
apply suction until the product is nearly\ dr~.. -Dry' the product in a covered evaporating dish overnight
0
at 50 Submit the product to the instructor in a tared, labeled sample bottle. Calculate the percentage yield and submit the calculations to the instructor.
NOTE
1. The pH should not increase beyond 7 at any time, since cystine is decomposed even in slightly alkaline solution. Sodium acetate, rather than sodium hydroxide, has been employed to avoid this danger.
- 26-
EXPERIMENT
17
The optical rotation of L-cystine prepared in Exp. 16 is to be determined, and the purity of the
sample is to be calculated from the specific rotation,l which is a dependable criterion of purity for
optically active amino acids. The construction and operation of a polarimeter are described in the
Appendix (see Polarimetry) and will be demonstrated by the instructor. The beasurement of optical
rotation cqnsist~ essentially in determining the number of angular degrees to which the plane of polarized light is rotated by a solution of an optically active substance.
EXPERIMENT AL
Qbtain from the instructor three approximately O.S-g. samples of L-cystine and one of analytically
pure L-cystine. Obtain assignment to a polarimeter. Transfer 0.5 g. (weighed to 1 mg.) of each
sample quantitatively to a SO-ml. volumetric flask, add 1.00 N hydrochloric acid to each flask, rotate
the flasks until the samples dissolve, and add 1.00 N hydrochloric acid to the marks.
Turn on the sodium light and allow it to warm up for 10 min. Thoroughly rinse the polarimeter
tube with distilled water and fill the tube with distilled water in such a manner that ,air bubbles are
removed: Place the tube in the instrument, focus the eyepiece, and rotate the scale until the dark fields
match. Record the scale reading to three figures. Repeat this matching process several times 'While
approaching the matching point from both directions.
Empty the tube, rinse it thoroughly with the minimum volume of one of the unknown L-cystine
solutions, and fill it with this solution. Match the fieldsas described and record the scale readings.
Place a thermometer in the L-cystine solution and record the temperature. Repeat these operations
using each ofthe remaining L-cystine solutions in turn.
NOTE
1. Specific rotation is calculated from the equation
[a]t = av
D
where
wI
a=
Specific rotation and temperature are related (for L-cystine)by the equation [a]t
derived by Toennies and Lavine (1930).
D
- 27-
= [(2.061t) _
264.84]0,
EXPERIMENT 18
PHOTOMETRIC ANALYSIS OF CYSTINE BASED ON THE REACTION
OF CYSTEINE WITH PHOSPHOTUNGSTIC ACID
The reaction of cysteine with phosphotungstic acid, first observed by Winterstein (1901-1902),
was applied by Folin and Looney (1922) to the quantitative determination of' cystine. Because the
reagent was reduced by tyrosine ~d other substante~, more specific reagents were devised by Folin
and Marenzi (192%) and Folin (1934). Further improvements were effected py substituting NaHCO, for
Na2C03 to prevent the turbidity caused by sodium phosphotungstate (Tompsett, 1931), introducing a
buffer solution to maintain the pH at 5.0 to 6.5, over which range the colored complex is stable (Lugg,
1932a, 1932b; Schoberl and Rambacher, 1937-1938), and measuring the color intensities with a photoelectric, rather than a visual, colorimeter (Balint, 1938; Block and Bolling, 1940; Kassell and Brand,
1938).
Cysteine (formed by reduction of cystine) has been determined by (a) titration with 'standard
iodine in the presence of starch (Morner", 1901-1902), (b) titration with standard iodate in the presence
of iodide and starch (Okuda, 1925), (c) reaction withstandard iodine in potassium iodide solution and
titration of the excess iodine with standard thiosulfate (Virtue and Lewis, 1934), (d) photometric
determination of the colored complex formed with 1,2-naphthoquinone-4-sulfonate (Andrews and Andrews,
1~37; Bushill et ai., 1934; Hess and Sullivan, 1943; Lugg, 1933; Sullivan et aI., 1942) or with sodium
nitroprusside (Shinohara and Kilpatrick, 1934), (e) determination of nitrogen or sulfur in cuprous
mercaptide (Lucas and Beveridge, 1940; Vickery and White, 1932; Zittle and O'Dell, 1941), and (f)
reaction with iodine and gasometric determination of the nitrogen formed by reaction of the excess
iodine with hydrazine (Hess, 1933).
Cystine has been determined by (a) oxidation and gravimetric analysis of sulfate as barium
sulfate (Callan and Toennies, 1941) and (1: polarographic analysis (Callan and Toennies, 1941; Stern
e~ al., 1939).
.
The purity of unknown cystine samples is to be determined by reducing the cystine to cysteine
and analyzing photometrically the blue-colored complex formed by reaction of the cysteine with
..
phospho-l8-tungstic acid. According ~o Clarke (1~32) one molecule of cystine is reduced by sulfite to
one molecule of cysteine sodium mercaptide [NaSCH 2 CH(NHJCOOH] and one molecule of sodium
,S-cysteinesulfonate [HOOCCH(NH2)~H2S0,Na). Zinc, tin, s"odium cyanide, sodium amalgam, cuprous
'chloride, and cuprous oxide have been "employed as reducing agents.
EXP ER IMENT AL
bbtain three L-cystine unknowns and assignment to a visual polorimeter. Transfer quantitatively
approximately 0.25 g. (weighed to 1 mg.) of each sample to a 25O-ml. volumetric flask. Add 100 mI.
of 0.5 N sulfuric acid to each flask and rotate the flasks until the solids dissolve. Add 0.5 N sulfuric
acid to the marks and mix the liquids in each flask thoroughly.
Pipette" 2.00-mI. aliquots of the three unknown solutions and the standard solution, containing
1.00 mg. per mI. of purified L-cystine, to four separate 100-ml. volumetric flasks. Add to ea9h flask
10 mI. of saturated sodium bicarbonate solution and 2.0 ml. of Folin's improved phospho-l8-tungstic
acid reagent. Mix the liquids thoroughly and allow the mixtures to stand for 10 min. Dilute each
solution to the mark with distilled water and mix the liquids in each flask thoroughly. Compare the
color intensities of the unknown solutions with the standard which has been developed simultaneously.
- 28-
EXPERIMENT 19
,
(SOLATION OF ASPARAGINE MONOHYDRATE FROM LUPINE SPROUTS
Asparagine, the beta amide of aspartic acid, is a metabolic product of sprouted lupines (Lupinus),
vetches (Vicia), beans (Phaseolus), soybean (Glycine hispida), and other members of the legume family.
It is widely used as an essential 'component of media on which bacteria and yeasts are cultured.
Asparagine was first isolated from asparagus sprouts by Robiquet (1805) and Vacquelin and
Robiquet (1806), but the first practicable isolation procedure was reported by C;hibnall (1924). Asparagin~ has been isolated most commonly from lupine sprouts by modifications of the method of Piria
(1848). The early literature has been revieweci'by Schul?e and Winterstein 0910), Vickery and Schmidt
(1931), Mumeek (1935), [Link] ai. (1937), and Chlbnall (1939).
Asparagine is to be isolated from lupine sprouts by a modification of Vickery and Pucher's (1943)
method. Lupine seeds are grown in the dark in tap water for 10 to 12 days (Lupinus angustifoiius) or
15 to 20 days (Lupinus albus). The highest yield has been obtained from Lupinus albus. The etiolated
seedlings are ground and extracted with water; the extract is heated to coagulate the proteins, the
suspension is filtered, the filtrate is evaporated, and asparagine monohydrate is allowed to crystallize.
E XP ER IMENTAL
Germination of Seeds. Obtain 100 g. of Lupinus albus seedsand assignment to a tray (or part of
a tray) about 2 ft. square and approximately 6 in. deep which is provided with a drain tube at the bottom
and near the side of the tray. Attach a rubber tube to the drain and close the tube with a screw clamp.
Place the tray in a well-ventilated, lightproof room or cabin~t. Fill the tray nearly full with clean sand,
run in tap water sufficient to cover the sand, and allow the water to drain. Punch holes about 1/2 in.
deep and as closely spaced as possible in the moist sand. Place one seed in each hole and cover the
holes with sand. Resaturate the sand with tap water at 48-hr. intervals. After the sprouts appear, allow
them to grow for an additional 18 or 19 days.
Extraction of Asparagine. Remove the entire mass of sprouts, roots, and residual seed substance.
Remove the adhering sand by immersing and agitating [Link] mass in a pan of ~ater. Blot the roots
with cheesecloth to remove excess water. Fill the glass (or metal) bowl of a Waring Blendor about
three-fourths full with the moist sprouts. Add 150 ml. of boiling distilled water, place the cover on the
bowl, and turn on the motor. When the material has been thoroughly macerated, .transfer the suspension
to a 2-liter flask containing 500 mI. of boiling distilled water, and place the fla=:;k in a boiling water
bath. Macerate the remainder of the sprouts similarly, add the suspensions to the 2-liter flask, and
continue to heat the bath\for 15 min.
Place a filter paper on a 6-in. Buchner funnel, add a suspension of filt~r aid, and apply suction
until about 1/2 in. of well-packed filter aid has deposited. Place a perforated filter paper on the filter
aid, disconnect the tube to the water pump, and discard the liquid in the filter flask. Transfer the hot
suspension to the Buchner funnel, [Link] tube to the water pump, and apply suction until the bulk
of the water has been removed and the residual material on the funnel is nearly dry. Transfer enough
residual material to fill halfway the bowl of the Waring Blendor, ?dd 150 ml. of boiling distilled water,
stir the material until it is thoroughly macerated, and transfer the mixture to a 2-liter flask' containing
200 mI. of boiling distilled water. Place the flask on a boiling water bath. Repeat this procedure until
all the residual material has been extracted. Filter the suspension on a Buchner funnel and wash the
residual material on the funnel with 200 ml. of boiling distilled water. Combine the filtrate and washings. This solution should be preserved in the refrigerator in a stoppered, labeled flask if it is allowed
- 29"-
to stand overnight. A few milliliters of toluene should be added as preservative if the solution is
allowed to stand for a longer time.
Isolation of Asparagine Monohydrate. Transfer the solution t5' a 2-liter beaker and heat it on a wire
gauze over a free flame until the volume is reduced (from about 1,200 mI.) fo 500 mI. Remove the flame,
add 2 g. of decolorizing carbon slowly to avoid foaming, stir the mixture at intervals for 15 min. and
filter tIre
s~spe.'}sio..'}. 'rransfer th'e filtrate to a l-liter round-bottomed flask, Set up the reduced pressure
apparatus shown in Fig, 7 (see Distillation in the Appendix), and distill the liquid 'under reduced-pressure on a water bath to a thin sirup (about 40 mI.). Transfer the sirup to a 100-mi. beaker and rinse the
sirup from the flask with two IS-ml. portions of boil!ng ,distilled water. Lab~l the container, cover it
with a watch glass, and allow it to stand in the refrigerator until the next laboratory period (but not
longer than 48 hr. becallse of the danger of bacterial 'or mold contamination).
Decant the [Link] liquid to a Hirsch or small Buchner funnel, and &fter the liquid has filtered,
add the crystalline mat~rial. Filter the mixture and wash the crystals on the funnel with 5 mI. of icecold distilled water. :I'ransfer the crystals to a IOO-ml. beaker, add 30 ml. o( distilled water, and heat
the mixture on a water l)ath until the crystals dissolve. Add 0.5 g. of decolorizing carbon, heat the
mixture for 5 min. on a water bath with frequent sUrring, and filter the hot sUspension on a Hirsch funnel.
Reheat and refilter the filtrate if carbon particles are observed. Allow the filtrate to cool slowly to room
temperature with frequent shaking. Place the container in the refrigerator and allow it to st~nd overnight.
Filter the suspension and wash the crystals on the funnel with'two 10-ml. portions of 95 per cent ethanol.
Allow the crystals to dty in air in a covered, labeled evaporating dish. Submit the product 1 to the instructor in a tared, labeled sample bottle. Calculate the percentage yield of [Link] monohydrate
based on tne weight of the seeds and submit the calculations to the instructor.
NOTE
1. A second =op may be obtained by evaporating the combined flItrates and Washings to a small volume
and recrystallizing the crl.\de\ product from water.
- 30-
EXPERIMENT 20
AMIDE-NITROGEN ANALYSIS OF ASPARAGINE MONOHYDRATE
EXPERIMENTAL
'Obtain a sample of asparagine monohydrate and assignment to a pH meter. Transfer about 0.25 g.
(weighed to 1 mg.) of the sample quantitatively to a 100-m!. volumetric flask. Add 25 mI. of distilled
water and rotate the flask until the sample dissolves. Add distilled water to the mark and mix the
liquids thoroughly. Pipette 25.0 m!. of the solution to a 50-m!. volumetric flask and submit the flask to
the instructor, who will add a known quantity of pure asparagine monohydrate.
Standard Curve Relating pH and Ammonia Cancentration of Ammonium Borate-Boric Acid Solutions. Pipette 10.0 m!. of 4.0 per cent boric acid solution into a 2S0-m!' beaker. Add 150 m!. of
distilled water2 from a graduate. Mix the solutions and measure the pH with a pH meter. Pipette into
this mixture 1.00 m!. of a standard, approximately 0.01 M, ammonium hydroxide solution,3 mix the liquids
thoroughly, and determine the pH. Continue adding 1.00-m!. volumes of the standard ammonium hydroxide s~lution and measuring the pH until there is a total volume of 10 mI. of ammonium hydroxide.
Calculate the equivalents of ammonia in eacn mixture and plot on coordinate paper the points
relating the ammonia and corresponding pH values. Draw a smooth curve, label the ordinates, and title
the graph.
Determination of Amide Ammonia in Asparagine Monohydrate. Pipette 5.00 mI. of the unknown
asparagine monohydrate solutiol!)nto 'a 500-m!. distilling flask. Add 75 m!. of distilled water and a few
boiling stones. Attach a Liebig condenser to the side arm of the flask. Pipette'.10.0 ml. of 4.0 per cent
boric acid solution into a 125-ml. conical flask and add about 25 mI. of distilled water. Insert the end
of the condenser into the flask and [Link] the apparatus so that the tip of the condenser is below the
surface of the boric acid solution. A~d 10 m!. of 6 N sodium hydroxide to the distilling flask with the
aid of a pipette, in such a manner that the alkali solution is introduced directly into the solution without coming in contact with the wall of the flask. Immediately stopper the distilling flask and heat it on
a wire gauze over a small flame to gentle boiling. Continue heating for 20 min., remove the receiver,
remove the burner, and rinse any boric acid solution from the condenser to the receiver. Transfer the
boric acid solution quantitatively to a 2S0-ml. gt'aduate, add distilled water to the mark, and mix the
liquids thoroughly. Transfer the solution to a beaker and determine its pH with the pH meter. Estimate
the equivalents of ammonia corresponding to the pH value by interpolating the standard curve.
Analyze the asparagine monohydrate recovery solution by the described procedure.
NOTES
1. By the Wagner method the a~lInonia in the boric acid solution is titrated directly in the presence of
methyl red indicator. Taylor and Smith titrated with the aid of the glass electrode. The present modification is
more convenient, although somewhat less accurate, than electrometric titration.
2. Distilled water of uniform quality obtained from the container on the side shelf should be used throughout
this experiment.
3. Determine the titer of the ammonium hydroxide solution on the day of the experiment by pipetting an
aliquot into an excess of standard acid and back-titrating with standard sodium hydroxide solution using the mixed
indicator employed in Exp. 9.
- 31-
EX P
ER 1M EN T
21
The amino acids in an unknown mixture of amino acids are to be identified by applying the group
separations and specific tests described below.
'SCHEME OF ANALYSIS
Step I. Detection of Amino Acids. Ninhydrin Test.
Step II. Detection of Pyrrolidine-containing Amino Acids. Proline and Hydroxyproline. Lead
Dioxjde Oxidation Test.
Step III. Detection of Sulfur-containing Amino Acids. Cystine and Methionine. Sodium Fusion
Test.
Step IV. Detection of Aromatic Amino Acids. Tryptophan, Tyrosine, and Phenylalanine.
Xanthoproteic Test.
Step V. Solubility Separation.
1. Proline. Extraction with Absolute Ethanol.
2. Cystine and Tyrosine. Separation from Water-soluble Amino Acids.
Step VI. Detection of Basic Amino Acids. Arginine, Histidine, and Lysine. Phosphotungstic
Acid Precipitation Test.
Step VII. Detection and Separation of Acidic Amino Acids. Glutamic Acid and Aspartic Acid.
Lime-Ethanol Precipitation Test.
.
Step VIII. Separation of Basic Amino Acids and Tryptophan.
1. Arginine, Histidine, and Lysine P hosphotungstates Precipitation.
2. Tryptophan Phosphotungstate Precipitation.
3. Phosphotungstic Acid Removal~
Step IX. Detection of Amino Acids Containing Adjacent Amino and Hydroxy Groups. Serine and
Threonine. Periodic Acid Oxidation Test.
Step X. [Link] of Ketone-forming Amino Acids. Valine, Leucine, and Isoleucine. Permanganate Oxidation Test.
Step XI. Detection of Individual Amino Acid$.
1. Proline. [satin Test.
\
2. Hydroxyproline. pyrrole-tsatin Test.
3. Cystine. {3-Naphthoquinonesulfonic Acid Test.
4. Methionine. Sodium Nitroprusside Test.
5. Tyrosine. Mercuric Nitrite Test.
6. Tryptophan. Glyoxylic Acid Test.
7. Phenylalanine. Nitric Acid-Hydroxylamine Test.
8. Arginine. Hypobromite-a-Naphthol Test.
9. Histidine. Bromidazole-Ammonia Test.
10. Lysine. Picrate Precipitation and Bromine-Phosphomolybdic Acid-Phosphotungstic
Acid Tests.
11. Glutamic Acid. Hydrochloride Precipitation Test.
12 Aspartic Acid.. Glyoxal-2,4-Dinitrophenylosazone and C-rfpric Aspartate Hydrate
Precipitation Tests.
13. Threonine. Periodic Acid-Acetaldehyde-p-Hydroxydiphenyl Test.
14. Serine, Periodic Acid-Formaldehyde-Chromotropic Acid Test.
- 32-
EXPERIMENTAL
Obtain a homogeneous mixture of unknown amino acids and identify each amino acid. Solid
materials are to be weighed on a horn-pan balance or measured by comparison with the volum:es of
weighed s"ample in test tubes on the reagent shelf.
Make an accurate record of all experimental observations. If any result is inconclusive, repeat
the test on a sample of the pure amino acid obtained from the instructor. After the group tests have
been completed, prepare a flow sheet outlining the individual tests of Step XI applicable to the unknown
mixture. Submit a written report summarizing the results of each test and the conclusions.
Step I. Detection of Amino Acids. Ninhydrin Test. 1 Transfer 5 mg. of the unknown to a 3-in.
test tube. Add 3 mI. of distilled water and 1 ml. of 0.1 per cent ninhydrin solution. Heat tlie mixture
in a boiling wate~ bath for 3 min. Note the color of the solution. Add 1 drop of glacial acetic acid and
1 ml. of chloroform. Shake the tube vigorously. The appearance of a blue to violet or red color after
heating and of an orange color entirely concentrated in the chloroform layer after acidification indicates
the presenc~ of at least one amino acid in the unknown.
Step II. Detection of Pyrrolidine-containing Amino Acids. Proline and Hydroxyproline. 2 Lead
Dioxide Oxidation Test. Transfer 20 mg. of the unknown to a 125-ml. conical flask, add 5 mI. of distilled water, and stir the mixture until all, or almost all, the solid dissolves. Add 10 ml. of phosphate
(Na2HP04) buffer solution to maintain the pH at 8.7 and add 0.2 g. of lead dioxide. Stir the mixture,
fit a Hopkins condenser into the neck of the flask, and heat the flask in a boiling water both for 20 min.
Filter the suspension and transfer 5 mI. of the filtrate to a 6-in. test tube. Add 1 mI. of a 4 per cent
solution of p-dimethylaminobenzaldehyde in 95 per cent ethanol and add 5 ml. of N hydrochloric acid.
Mix the liquids thoroughly and heat the tube in a boiling water bath for 1 min. The immediate appearance of a red color indicates the presence of proline, hydroxyproline, or both these amino acids.
Step III. Detection of Sulfur-containing Amino Acids. Cyst'ine and Methionine. Sodium Fusion
Test. Place a clean 3-mm. cube of sodium metal (obtained from the instructor) in a clean, dry 3-in.
test tube and add 15 mg. of the unknown. Suspend the tube through an asbestos board which is supported by an iron ring. Heat the tube with a flame until sodium vapors are visible in the lower part of
the tube. Add 10 mg. of the unknown (Cautiously) to the tube. Heat the tube until the glass is red and
continue heating for 1 min. Allow the tube to cool and add 0.5 mI. of 75 per cent ethanol to destroy the
unreacted sodium. Remove the tube from the board, heat it with a flame, and immerse the hot tube in a
50-ml. beaker containing 10 mI. of distilled water. Shatter the cracked tube, pulverize (with glass rod)
the glass and solid material, and filter the suspension. Transfer 5 mI. of the filtrate to a 6-in. test tube
and add 6 N acetic acid until the solution is neutral to litmus paper. Add 1 drop of 2 N lead acetate
solution, heat the mixture to boiling, and continue boiling for 2 min. to dissolve any lead chloride. The
formation of a black precipitate of lead sulfide indicates the presence of cystine or methionine or both
these amino acids.
Step IV. Detection of Aromatic Amino Acids. Tryptophan. Tyrosine. and Phenylalanine. Xanthoproteic Test. 3 Transfer 20 mg. of the unknown to a 4-in. test tube. Add 1 ml. of concentrated nitric
acid and 1 drop of concentrated sulfuric acid. Heat-the tube for 5 min. in a boiling water bath. Cool
the tube in running water and note the color of the solution. Add (Carefully) 2 ml. of saturated sodium
hydroxide, mix the liquids, and note the color of the solution. A y~l1ow-colored acid solution and an
orange to orange-red alkaline solution indicates the presence of an aromatic amino acid. Disregard any
white precipitate (sodium chloride) in the alkaline solution. Phenylalanine may give only a faint color
in acid solution, but it yields a definite orange color in basic solution.
Step V. Solubil ity 'Separations. 1. Proline. Extraction with Absolute Ethanol.4 Transfer all the
unknown to a 6-in. test tube, add 10 mI. of absolute ethanol, and stopper and shake the tube vigorously
for 2 min. Allow the mixture to stand, and decant most of the supernatant liquid to a dry gravity filter.
Collect and preserve the filtrate in a 3-in. evaporating dish. Add 10 mI. of absolute ethanol to the test
-33 -
tube, sto~thetube, and shake it vigorously for 1 min. Transfer the mixture to the funnel and collect
the filtrate.
aporate the combined filtrates on a water bath and preserve the dry residue (principally
proline) for the c nfirmatory test (Step XI, Test 1).
Transfer thb dry residue in the test tube to a dry fi.1ter, using 25-ml. portions of diethyl ether as
rinse fluid. Pour the ether filtrate into the sink and flush it into the drain with water. Preserve the
residual material for use in Step V2.
2. Cystine and Tyrosine. 5_eparation from Water-soluble Amino Acids. s If cystine and tyrosine
are absent (shown in Steps III and IV), transfer the proline-free residual material prepared in Step V-l
to a 50-ml. conical flask, add 20 ml. of distilled water, and rotate the flask (or hea,t it in a water bath)
until the solid dissolves. Proceed to Step VI.
If undissolved solid remains (cystine or tyros!ne present), continue to heat the flask in the water
bath for 15 min. at 80 0 Cool the flask for 15 min. i~ an ice-water bath, filter the mixture on a dry
gravity filter, and transfer the filtrate to a labeled flask. When the solution is not in use, preserve it
in the refrigerator for use in Step VI. If additional precipitate forms, filter the suspension and utilize
the filtrate in Step VI. Wash the precipitate withtwo S-ml. portions of methanol, dry it in air, and
preserve it for the confirmatory cystine (Step XI, Test 3) and tyrosine (Step XI, Test 5) tests.
Step VI. Detection of Basic Amino Acids. Arginine, Histidine, and Lysine. Phosphotungstic
Acid Precipitation Test. 6 Transfer 0.5 mI. of the solution of the soluble amino acids prepared in Step
V-2 to a 6-in. test tube. Add 7 mJ. of distilled water and 0.5 mI. of a 20 per cent solution of phosphotungstic acid. Stir the mixture vigorously for 1 min. The immediate appearance of a white precipitate
indicates the presence of at least one of the basic amino acids.
Step VII. Detection and Separation of Acidic Amino Acids. Glutamic Acid and Aspartic Acid.
Lime-Ethanol Precip it ation Test. 7 Heat the remaining volume of the solution of the soluble amino
acids prepared in Step V-2, -remove the flame when the solution starts to boil, and add 0.7 g. of
powdered, carbonate-free calcium hydroxide. Stopper the flask and shake it vigorollly for 1 min. Cool
the flask to 0 0 in an ice-water bath and shake it vigorously. Cool 50 mI. of 95 per cent ethanol to QO
in a 12S-ml. flask, add the amino acid-lime mixture slowly while rotating the flask. stopper the flask,
and shake it vigorously. Filter the mixture as quickly as possible, or centrifuge it if filtration is slow.
Preserve the precipitate for later use in this test. Acidify the filtrate (or supernatant liquid) with 0.5 N
sulfuric acid to pH 4 (test with congo red paper, which changes color from blue to purple over the pH
range 3 to 4). Filter the calci~m sulfate suspenSion, wash the solid on the funnel with 5 mI. of distilled
water, ~nd evaporate the combined filtrate and washing to 15 mI. Preserv~ this solution for use in Step
VIII.
Transfer the lime-ethanol precipitate to a SO-ml. flask, add 10 ml. of boiling distilled water,
stopper the flask, and shake it vigorously for 1 min. to dissolve the calcium glutamate and calcium
aspartate. Heat the suspension to boiling and fi1t~r (or centrifuge) it to remove the excess calcium
hydroxide and any calcium carbonate which m~y h~ve formed. Add the filtrate to 30 mI. of 95 per cent
ethanol while rotating the flask and ,\001 the mixture in an ice-water bath for 15 min. with frequent
stirring. If no precipitate forms in 15 'min., allow the mixture to stand overnight in the refrigerator. If
a precipitate forms, filter the suspension and preserve the precipitate for the glutamic ~cid (Step XI,
Test 11) and aspartic acid (Step XI, Test 12) confirmatory tests.
Step VIII. Separation of Basic Amino Acids and Tryptophan. Test the solution of the soluble
amino acids prepared in Step VII for tryptophan by the procedure given in Step XI, Test 6. If both the
basic amino acids and tryptophan are absent, omit Step VIII. If basic amino acids are absent and
tryptophan is present, omit Step VIII-I. If basic amino acids are present and tryptophan is absent,
omit Step VIII-2.
1. Precipitation of Arginine, Histidine, and Lysine Phosphotungstates The object of this procedure is to precipitate the basic amino acids but not tryptophan, which might interfere with the confirmatory tests for the basic amino acids. The quantity of phosphotungstic acid required to accomplish
this purpose is variable, depending upon the number of basic amino acids present in the unknown.
Transfer the amino acid solution prepared in Step VII to a 125-ml. conical flask, add 37 mI. of a
20 per cent solution of phosphotungstic acid, add 1 mI. of 6 N hydrochloric acid, and stir the mixture
at intervals for 5 min. Immediately filter the suspension with suction and preserve the precipitate
- 34-
(A-I). Add 1 mI. of the phosphotungstic acid reagent to 5 mL of the filtrate. If a precipitate forms, combine this suspension with the remaining filtrate and add 22 mI. of the phosphotungstic acid reagent. Stir
the mixture at intervals for 10 min. and filter the suspension with suction. Preserve the precipitate (A-2).
If tryptophan was present in the unknown, test the filtrate (from which A-2 was removed) for tryp~
tophan by the procedure given in Step XI, Test 6, to determine if tryptophan was removed in precipItate
A-2. If this test is negative, preserve precipitate A-2, test A-2 for tryptophan, and (if the test is negative) omit Step VIII-2. If the test is positive, combine precipitates A-I and A-2. Preserve the phosphotungstic acid precipitate (A-lor A-I plus A-2) for the removE' 1 of phosphotungstic acid in Step VIII-3.
Preserve the filtrate from precipitate A-I (if tryptophan was absent from the filtrate of A-2) or from A-I
plus A-2 for the precipitation of tryptophan phosphotungstate (Step VIlI-2) (if tryptophan was present)
or for the removal of phosphotungstic acid (Step VIII-3) (if tryptophan was absent).
2. Precipitation 0/ Tryptophan Phosphotungstate. Basic amino acids absent. Add 10 mI. of the
phosphotungstic acid reagent and 0.5 mI. of 6 N hydrochloric acid to the solution preserved from Step
VII, stir the mixture, and place it in an ice-water bath for 1 hr. or in the refrigerator overnight. Filter
the suspension and preserve the precipitate (containing some glycine if glycine is in the unknown) for
use in 'Step VIII-3. If the filtrate is turbid, add 0.5 g. of decolorizing carbon, stir the mixture, and
filter the suspension, using two thicknesses of filter paper. If necessary, repeat this process until the
filtrate is clear. Preserve the filtrate for use in Step VIII-3.
Basic amino acids removed. Evaporate the filtrate from. Step VIII-1 to 10 ml. and carry out the
procedure described in this section.
3. Removal 0/ Phosphotungstic Acid. Transfer the phosphotungstic acid precipitate (from Step
VIII-I) to a separatory funnel. Add 10 mI. of distilled water, 5.0 ml. of N hydrochloric acid, and 25 mI.
of a 1: 1 mixture of amyl alcohol and diethyl ether.s Shake the funnel, while holding the stopper and
stopcock firmly with the funnel in an inverted position, until the precipitate dissolves (turn the stopcock at intervals to release the pressure). Allow the layers to separate and draw off'the lower, heavier
water layer containing the amino acids. If the water layer does not settle, add additional 10-ml. portions of distilled water with shaking until the layers' separate. Reextract the aqueous solution with 10
ml. of the alcohol-ether mixture to remove traces of phosphotungstic acid. Preserve the aqueous
solution for the confirmatory tests in Step XI for arginine (Test 8), histidine (Test 9), and lysine (Test
10). Place the alcohol-ether solution of phosphotungstic acid in the container provided for that purpose.
Transfer the phosphotungstic acid filtrate from Step VIII-1 (or Step VIII-2) to a separatory funnel,
add 20 mI. of the amyl alcohol-ether mixture, and shake the funnel as described. Remove the lower
water layer and reextract it with 10 mI. of this solvent. Add N sodium hydroxide to pH 4 (test with
congo red indicator, which changes color from blue to purple over the pH range 3 to 4), and preserve the
solution for use in Step IX. Place the alcohol-ether solution of phosphotungstic.,acid in the special
container.
Step IX. Detection of Amino Acids Containing Adjacent Amino and Hydro><}' Groups. Serine and
Threonine. Periodic Acid Oxidation Test. 9 Obtain the aeration assembly shown in Fig. 5 (see Aeration
in the Appendix). Put 0.1 ml. of the unknown amino acid solution preserved in Step VII or VIII into the
reaction tube; add 3 ml. of concentrated (50 g. per 100 ml.) potassium carbonate solution and 2 drops of
arltifoam agent (turkey red oil). Apply suction and aerate for 2 min. Disconnect the receiver, add 5 ml.
of Nessler reagent (Hgl2 in an alkaline solution of KI) to the receiver, and add 2 ml. of 0.5 M periodic
acid to the reaction tube. Immediately reassemble the apparatus and aerate for 5 min. The appearance
of a red color in the Nessler solution indicates the presence of ammonia in the solution and of serine,
threonine, or both these amino acids in the unknown.
Step X. Detection of Ketone-forming Amino Acids. Valine, Leucine, and Isoleucine. Permanganate Oxidation Test. 10 Transfer 4 ml. of the solution prepared in Step VII or VIII to a 50-ml. conical
flask. Add 1 ml. of 6 N sulfuric acid and 1 rnl. of 30 per cent aqueous sodium nitrite solution. Heat
the flask for 15 min. in a boiling-water bath to dearninize the amilio acids and destroy the excess
nitrous acid. Add N sodium hydroxide to pH 4 (test with congo red indicator). Add 10 ml. of pH 6.8
buffer solution (KH aP0 4 and K 2 HP0 4 ), 1 g. of solid potassium perrnanganate, and a boiling stone. Insert
a stopper fitted with a glass tube leading downward into a test tube containing 5 mI. of water. Immerse
the receiver in an ice-water bath. Heat the flask with a small flame and continue heating until the vol- 35-
ume of the liquid has been reduced to about 3 m!. Immediately remove the receiver to prevent the liquid
from being sucked back as the reaction tube cools. Transfer 3 m!. of the solution in the receiver to
another test tube. Add 3 m!. of water, 4 ml. of 95 per cent ethanol, and (Cautiously) 4 m!. of concentrated sulfuric acid. Cool the mixture to room temperature under rlVlning water. Add 2 m!. of a freshly
prepared 6 per cent alcoholic solution of salicylaldehyde, and place the tube in a water bath at 60 for
20 min. The appearance of a red color indicates the presence of a ketone-forming amino acid. Preserve
the remainder of the solution in the receiver for the leucine and valine (Step XI, Test 15) and isoleucine
(Step XI, Test 16) confirmatory tests.
Step XI. Detection of Individual Amino Acids. 1. Proline. [satin Test. n Transfer 20 mg. of the
solid isolated in Step V-l to a 6-in. test tube. Add 5 mI. of glacial acetic acid and 1 mg. of isatin.
Heat the test tube for 5 min. in a boiling-water bath., The appearance of a blue color confirms the presence of proline.
2. Hydroxyproline. Pyrrole-lsatin Test. 12 Transfer 4 drops of the solution prepared in Step vm
to a test tube. Add 1 m!. of water, 1 m!. of 0.01 M cupric sulfate solution, 1 m!. of 3 N sodium hydroxide, and 1 ml. of 6 per cent hYclrogen peroxide. A~low the solution to stand for 5 min. with frequent
shaking. Heat the solution in a boiling-water bath for 5 min. and cool it to room temperature. Add N
hydrochloric acid dropwise until the deep-blue color of cupric sulfate has been discharged. Add 0.3 mt.
of N hydrochloric acid. Transfer 2 m!. of the solution to a 4-in. test tube and add 2 mI. of a freshly
prepared 1 per cent aqueous solution of isatin. Add 2 mI. of N hydrochloric acid. Heat the tube for 5
min. in a boiling-water bath. The appearance of a red color co}1firms the presence of hydroxyproline.
3. Cystine. S-Naphthoquinonesulfonate TestY Transfer 1 mg. of the precipitate obtained in
Step V-2 to a 6-in. test tube. Add 5 m!. of 0.1 N sodium hydroxide, stir the mixture until the solid
dissolves, and add 1 m!. of a freshly prepared 5 per cent solution of NaCN in N sodium hydroxide
(obtain this solution from the instructor). Allow the solution to stand for 10 min., add 1 mI. of a freshly
prepared 0.5 per cent solution of 1,2-naphthoquinone-4-sulfonate, and stir the mixture. Add immediately 5 ml. of a 10 per cent solution of sodium sulfite in 0.5 N sodium hydroxide. Stir the mixture, and
after 10 min., add 1 mI. of a 2 per cent solution of sodium hydrosulfite in 0.5 N sodium hydroxide. If
the deep-red color first formed persists after the addition of the hydrosulfite, the presence of cystine
is confirmed. If the red color is discharged, cystine is absent.
4. Methionine. Sodium Nitroprusside Test. 14 Transfer 0.5 mI. of the solution prepared in Step
VII or VIII to a 6-in. test tube. Add 2 m!. of water and 3 mI. of 3 N sodium hydroxide. Stir the mixture,
add 1 ml. of a freshly prepared 1 per cent solution of sodium nitroprusside, and mix the solutions.
(Caution: Sodium nitroprusside is highly poisonous.) Heat the tube for 5 'min. in a water bath at 40.
Cool the tube in an ice-water bath for 5 min. and add 3 m!. of 6 N hydrochloric acid slowly, while
shaking the cold solution. The appearance of a red color confirms the presence of methionine.
5. Tyrosine. Mercuric Nitrite Test. 15 Transfer 1 mg. of the precipitate obtained in Step V-2 to a
3-in. test tube. Add 1 m!. of water apd 2 drops of Millon's reagent. Stir the mixture and heat the tube
for 1 min. in a bolling-water bath. The appearance of a bright-red color confirms the presence of
tyrosine.
6. Tryptophan. Glyoxylic Acid Test. 16 Transfer 1 drop of the filtrate from Step VII to a pyrex
test tube. Add 2 m!. of distilled water, ad~ 2 drops of the Hopkins-Cole-Benedict reagent (magnesium
glyoxylate solution), and mix the liquids. Incline the tube and allow 2 mI. of concentrated sulfuric
acid to flow down the side of the tube and form a layer beneath the aqueous solution. Heat the tube
for 2 min. in a boiling-water bath. The appearance of a purple to black ring at the interface of the
liquids confirms the presence of tryptophan. If the layers are mixed (Cautiously), the solution will be
colored uniformly purple.
7. Phenylalanine. NitricAcid-Hydroxylamine Test,17 Transfer 2 m!. of the solution prepared in
Step VII or VIII to a 6-in. pyrex test tube and evaporate the solution to dryness in a boiling-water bath.
Cool the tube, add 2 m!. of the nitrating agent (KNO, in concentrated sulfuric acid), and heat the tube
in a boiling-water bath for 20 min. with occasional (Cautious) stirring. Remove the tube, cool it under
running water, and transfer the contents (Cautiously) to a 50-m!. flask containing 3 ml. of di~tilled
water. Stir the mixture and cool the flask in an ice-water bath for 5 min. Add 4 mI. of 20 per cent
hydroxylamine hydrochloride solution, allow the flask to stand for 1 min., and add (Cautiollsly) in small
- 36-
portions 10 mI. of concentrated ammonium hydroxide. Remove the flask from the bath. The appearance
of a violet color confirms the presence of phenylalanine.
8. Arginine. Hypobromite-a-Naphthol Test. iS Transfer 0.1 ml. of the solution prepared in Step
VIII-3 to a 6-in. test tube containing 10 mt of distilled water. Mix the solutions thoroughly and transfer
0.1 ml. to a ().in. test tube. Add 5 mI. of distilled water, 1 m1. of 6 N sodium hydroxide, and 1 mI. of a
0.02 per cent solution of a-naphthol in 20 per cent ethanol. Mix the liquids, place the tube in an icewater bath for 3 min., add 2 (not more) drops of" sodium hypobromite reagent (bromine in 1.5 N sodium
hydroxide), and stir the liquids. The appearance of a red color confirms the presence of arginine.
9. Histidine. Bromimidazole-Ammonia Test.1!l Transfer 0.1 mI. of the solution prepared in Step
VIII-3 to ~ 3-in. test tube, add 2 ml. of distilled water, and add a solution of bromine in 33 per cent
acetic acid dropwise until a light-yellow color persists. Allow the solution to stand for 5 min . Add 2
ml. of an ammonia-ammonium carbonate solution and heat the tube for 5 min. in a [Link] bath.
The appearance of a dark blue-violet color confirms the presence of histidine.
10. Lysine. Picrate Precipitation Test. 20 Transfer 10 ml. of the solution of basic amino a~ids
prepared in Step VIII-3 to a flask. Add 15 ml. of a saturated solution of picric acid (1.4 g. per 100 ml.
of water). Stopper the flask and shake it vigorously. The appearance of yellow crystals of lysine
picrate confirms the presence of lysine. If no crystals lIPpear within a few minutes, allow the flask
to stand overnight in the refrigerator. Determine the melting' point of the precipitate.
Bromine-Phosphomolybdic Acid-Pho#Jhotungstic Acid Test. 2i Transfer 0.5 ml. of the solution of
basic amino acids prepared in Step VIII-3 (if histidine was absent) or 20 mg. of the lysine picrate preparation to a 6-in. test tube. Add 1 mI. of saturated bromine,water, add 2 drops of 6 N hydrochloric acid,
and allow the solution to stand for 5 min. Add a 5 per cent solution of sodium arsenite dropwise until
the yellow bromine color is discharged. Add sodium carbonate solution to a basic teaction to litmus
paper. Heat the solution for 2 min. in a boiling-water bath, add 0.5 ml. of the Folin-Ciocalteu reagent
(phosphomolybdic acid-phosphotungstic acid), and heat the mixture. for 5 min. in a boiling-water bath.
Lysine gives a green color which changes to a deep blue or blue green within 1 to 2 min. Histidine gives a purple color on the addition of sodium carbonate and obscures the color formed by lysine.
Arginine yields a green color which deepens to blue on standing 20 min. Lysine and lysine picrate
behave similarly, except that the yellow picric acid alters the shade of color to a pronounced greenish
blue.
11. Glutamic Acid. Hydrochloride Precipitation Test. 22 Transfer the precipitate obtained in Step
VII to a 125-ml. flask. Add 15 ml. of distilled water, rotate the flask until the solid dissolves, and add
slowly, with stirring, M oxalic acid solution to pH 4 as tested with congo red indicator. Heat the suspension of calcium oxalate in a boiling-water bath for 15 min. with occasional ~tirring. Add additional
oxalic acid solution until the pH is 4 and digest the suspension in a boiling-water bath for 10 min.
Repeat this process until the pH is not lowered during the digestion process.
Filter the suspension and wash the precipitate thoroughly on the funnel with 10 ml. of cold water.
Evaporate the clear filtrate and wash liquid on a boiling-water bath to 10 mI. Cool the solution -in an
ice-water bath and saturate it with hydrogen chloride from a generator in the fume hood. Stopper the
flask tightly and allow it to stand in the refrigerator until the next laboratory period. The formation' of
colorless needle-like crystals indicates the presence of glutamic acid hydrochloride. Centrifuge the
suspension (or solution, if glutamic acid is absent) and preserve the centrifugate for use in Test 12.
Dry the precipitate in air and determine its melting point.
12. Aspartic Acid. Glyoxal-2,4-dinitrophenylosazone Test. 23 Transfer 2 mI. of the solution
prepared in Test 11 to a 50-ml. flask and heat the flask for 5 min. in a boiling-water bath to remove
excess hydrochloric acid. Add 10 ml. of water, 1 mI. of concentrated sulfuric acid (Caution), and 1 ml.
saturated bromine water. Heat the solution for 5 min. in a boiling-water bath. Cool the solution to
room temperature. Add 1 mI. of M potassium bromide solution and 2.5 mI. of 1.5 N potassium permanganate solution. Stir the liquids and allow the mixture to stand for 10 min. at room temperature.
Add 6 per cent hydrogen peroxide dropwise until the solution is decolorized. Attach an aeration
assembly as shown in Fig. 5 (see Aeration in the Appendix), and add to the receiving tube cooled in an
ice-water bath 5 ml. of a clear (filter, if necessary) 0.5 per cent solution of 2,4-dinitrophenylhydrazine
in 2.4 N hydrochloric acid. Immerse the reaction flask in a boiling-water bath and aerate the solution
-'37 -
until the volume has decreased to 8 ml. Filter the orange-colored suspension (formed if aspartic add
was present) on a Hirsch funnel and wash the precipitate on the funnel with 2 mI. of water. Dissolve
the precipitate by filtering 3 mI. of pyridine repeatedly through the funnel using suction. Add 20 mI. of
distill~ water to the pyridine solution, add 2 ml. df 6 N sodium hydroxide, and stir the mixture. The
appearance of a blue color confirms the presence of aspartic acid. I
Cupric Aspartate Hydrate Precipitation Tf!"St. 24 Evaporate the remainder of the solution prepared
in Test 11 to a sirup on a boiling':'water bath. Add 5 ml. of disti1lrd water and repeat this pr~cess. Add
5 ml. of distilled water to the sirup and heat the mixture on a boiling-water bath un,til the sirup dissolves. Continue heating and add 0.3 g. of solid CuCO! in small portions with stirring. Filter the mixture after 5 min. and allow the filtrate to stand in the refrigerator until the next laboratory period. The
formation of fine blue needles with a violet-colored'tinge indicates the pres~nce of aspartic acid.
13. Threonine. Periodic Acid-Acetaldehyde-p!Hydroxydiphenyl Test. 2S Transfer 0.1 mI. of the
solution prepared in Step VII or VIII to a 6-in. test tube. ,Add 2 drops of an 'antifoam agent (turkey red
oil), 5 mI. of 0.5 M sodium bicarbonate solution, and 7 ml. of 0.1 N sodium arsenite in 2 per cent sodium
bicarbonate solution. Mix these liquids and add 1 ml. of 0.5 M periodic acid. Immediately conbect the
aeration assembly-shown in Fig. 5 (see Aeration in the Appendix) to the receiving tube, to which 10 mI.
of concentrated sulfuric acid and 10 mg. of powdered p-hydroxydiphenyl have been added. Aerate
for 20 min. The appearaI!ce of a violet color in the sulfuric acid confirms the presence of threonine.
Preserve the periodic acid-oxidized solution in the reaction tube for use in Test 14.
26
14. Serine. Periodic Acid-Formaldehyde-Cbromotropic AcidTest. Add 2 drops of methyl red
indicator solution to the reaction tube (preserved from Test 13) containing the periodic acid-oxidized
solution. Add 6 N acetic acid dropwise to an acid reaction (red color). Add 10 ml. of distilled water
to a clean receiving tube and attach the aerating assembly shown in Fig. 5 (see Aeration in the Appendix). Place the reaction tube and the trap in.a boiling-water bath. Place the receiver in an ice-water
bath and aerate until the 'volume of solution in the reaction tube is reduced to 5 mI. (about 20 min.).
Transfer 3 ml. of this solution to a 6-in. test tube. Add 5 ml. of water, 3 ml. of concentrated sulfuric
acid (Caution), and 10 mg. of solid chromotropic acid. Stir the mixture and heat it in a boiling-water
bath. The appearance (within 20 min.) of a violet-rose color confirms the presence of serine.
15. Leucine and Valine. Acetone-$alicylaldehyde Test. 27 Transfer 1 mI. of the solution prepared
in Step X to a 6-in. test tube. Add 10 ml. of 6 N sodium hydroxide and 1 ml. of a 6 per cent solution of
salicylaldehyde in ethanol. Heat the tube for 15 min. in a boiling-water bath. The appearance of a red
color confirms the presence of leucine, valine, or both these amino acids.
16. Isoleucine. Methyl Ethyl Ketone-'SalicylaldehydeTest.'JJ3 Transfer 5 mI. of the solution prepared in Step X to a 25()..ml. flask. Add 100 ml. of distilled water,S ml. of concentrated sulfuric acid,
and 35 mI. of Deniges reagent (15 per cent HgSO. ip 6 N sulfuric ,acid). Attach a condenser and reflux
the solution for 30 min. Cool and filter the suspension of the acetone-mercuric sulfate complex, using
a Hirsch funnel and a smootl;l filter p~er., Tr~nsfer a S-ml. portion of the filtrate to a 6-in. test tube.
Add 3 mI. of 95 per cent ethanol, 2 mb bf concentrated sulfuric acid (Caution), and 1 ml. of a 6 per cen1
solution of salicylaldehyde in ethanol. Heat the tube i'n a boiling-water bl;lth. The appearance (within
15 min.) of a red color confirms the presence of methyl ethyl ketone in the solution and of isoleucine in
the unknown.
17. Alanine. Ninhydrin-Acetaldehyde-p-HydroxydlphenyITest. 29 Transfer 0.5 of the solution
prepared in Step VII or VlII to a 6-in. test tube. Add 3 mI. of water and add 0.1 N sodium hydroxide to
pH 7 (test with bromthymol blue indicator). Add 2 mI. of a 1 per cent solution of ninhydrin and heat the
tube for 3 min. in a boiling-water bath. Cool the solution and attach the aerating assembly shown in
Fig.s (see Aeration in the Appendix). Add 5 mI. of concentrated sulfuric acid and 10 mg. of p-hydroxydiphenyl to the receiving tube. Allow the solution to aerate for 15 min. The appearance of a pink to
violet color confirms the presence of alanine. Preserve the solution in the reaction tube for use in the
confirmatory test for glycine.
30
18. Glycine. Ninhydrin-Formaldehyde-Chromotropic AcidTest. Attach the reaction tube (containing the ninhydrin-reaction products preserved from 1'est 17) to the aeration assembly shown in Fig.
S (see Aeration in the Appendix). Place the reaction tube and the trap in a boiling-water bath and the
receiving tube containing 10 ml. of water in an ice-water bath. Aerate for 20 min. Transfer 3 ml. of the
- 38-
solution in the receiver to a 6-in. test tube. Add 5 ml. of water and (Caution) 3 mI. of concentrated
sulfuric acid. Add 10 mg. of chromotropic acid, stir the mixture, and heat the tube in a boiling-water
,.-.bath. The appearance of a deep-violet color confirms the presence of glycine.
o-Phthalaldehyde Test. 31 Transfer J ml. of the solution prepared in Step VII or VIII to a test
tube. Add 2 ml. of a pH 8.0 buffer solution (Na2HP04 and NaH 2P0 4) and -s ml. of the o-phthalaldehyde
reagent. Mix the solutions [Link] allow the/mixture to stand for 2 min. Add 5 mI. of a freshly prepared
and cooled mixture of 6 ml. of ethanol and 1 ml. of concentrated sulfuric acid. Shake the tube vigorously
for 30 sec. and add 15 mI. of chloroform. The appearance of a green color in the chloroform layer indicates the presence of glycine.
NOTES
1. The observation of Ruhemann (1910) that a blue-colored solution is formed/by boiling an aqueous solution of ninhydrin and an alpha-amino acid was explained (Ruhemann, 1911) by the following postulated reactions:
CX
Ii
II
c",-/OH
from
C/\H
CCC"
I
reducmg
~
substance,.
~~ino,
aCid
"
~ H
\
OH
II
II
"'
/
CX
II
C/\NH
II
(A) ninhydrin
( triketohydrindine
hydrate)
1,3-diketohydrindole
(8)~3-diketohydrindamine
(A)
(8)
diketohydrindylidene
diketohydrindamine
Retinger (1917) has presented objections to this mechanism.
Ammonium salts (Harding and Warneford, 1916) and amines (Harding and MacLean, 1916) give reddish-violet
to blue colors with ninhydrin, but proline, hydroxyproline, diketopiperazine, urea, hippuric acid, purines, and
(3-, Y-, and 8-amino acids give little or no color.
The sensitivity of nine amino acids tested by Abderhalden apd
Schmidt (1911, 1913) ranged from 1 part of histidine in 79,000 parts of water to 1 part of valine in 16,000 parts of
water. Harding and MacLean (1915) found that 1 part of alanine and other amino acids could be detected in
1,500,000 parts of water.
The ninhydrin reaction has been applied to the analysisof amino acids (MacFadyen, 1944; Schott et at., 1944) and to
the determination of amino acids in proteins (Stein and Moore,1949), blood (Hamilton, 1945; Hamilton and Van Slyke,
1943; MacFadyen, 1942; Van Slyke and Dillon, 1938; Van Slyke, Dillon, MacFadyen, and Hamilton, 1941; Van
Slyke, MacFadyen, and Hamilton, 1942), and urine (Van Slyke, MacFadyen, and Hamilton, 1943). Analytical procedures based on the reaction of ninhydrin with alpha-amino acids to give ammonia (MacFadyen, 1944; Sobel,
Hirschman, and Besman, 1945), carbon dioxide (Hamilton, 1945), and volatile aldehydes (see Step XI, Tests 17
and 18) have been reported. Virtanen and Rautanen (1947) have shown that of the amino acids in proteins only
alanine, valine, leucine, isoleucine, phenylalanine, and methionine form volatile aldehydes. Moore and Stein
(1949) have investigated the color yields from reactions of ninhydrin with amino acids.
- 39-
Alpha-amino acids react with ,B-naphthoquinonesulfonate to give red-colored solutions (Folin, 1922a) and
with p-nitrobenzoy.l chloride and alkali to give 4-isobutyl-2-p-nitro~henYI-S-oxazolone (Waser and Brauchli, 1924),.
which forms a deep-violet-blue color with alkali salts (Karrer and Keller, 1943).
2. Proline, hydroxyproline, or both amino acids are determined usually by the reaction of isatin or p-dimethylaminobenzaldehyde with pyrrole or related product formed by o~dati:on with PqOa.(Guest, 1939), NaaOa(Morse,
1933), CUS04 and NaaOa (McFarlane and Guest, 193~), or NaOCI (Lang, 1933; Waldscnmidt-Leitz and Akabori, 1934).
According to Pratesi (1933), the isatin_-pyrrole condensation product has the following structure:
H-C-_
II
Ic, _
~)N
H-C-C
C",
(X
/C=O -
3. Salkowski (1888) observed that yellow to red solutions resulted when some proteins and amino acids
were heated with concentrated nitric acid. Neither phenylacetic acid nor phenylpropionic acid gave a color.
According to Johnson and Kohmann (191Sa, 1915b) tyrosine reacts with concentrated nitric acid to give principally 3-nitrotyrosine and with a mixture of concentrated nitric and sulfuric acids to give 3,S-dinitrotyrosine. The
color reaction probably depends upon the nitration of the benzene and indole rings to form nitro derivatives which
have resonating electronic structures. Tryptophan'may undergo partial oxidation as well as nitration.
4. The solubility of L-proline is 1.5 g. per 100 g. of absolute ethanol (Kapfhammer and Eck, 1927). Other
amino acids are only slightly soluble in this solvent.
5. The solubility of L-cystine is abouJ: 0.01 g. and of L-tyrosine about [Link]. per 100 g. of water at 25 0 All
other amino acids are much more soluble. Tryptophan, while reasonably soluble, dissolves slowly.
6. This test is performed here, rather than .after Step VII, to avoid possible interference from calcium
phosphotungstate, which is only slightly soluble and might precipitate. If tryptoPban and a basic amino acid are
present, the precipitate may be colored pink or brown. If tryptophan alone is present"a precipitate may form which
changes rapidly to a brown scum. According to Van Styke, Hiller, and Dillon (1942) the solubilitie;; at 22 in
millimoles per liter of 0.25 N hydrochloric acid of the following .amino acid phosphotungstates are lysine, 0.055;
[Link], 0.390; arginine, 0.69; c!stine, 0.507; tryptophan, 4.43; proline, 21.1, and glycine, 32.8. Since cystine '
was removed previously, only lysine, histidine, and ~rginine would precipitate under the described experimental
conditions. The ratios of amino acid to phosphotungstic acid found by Van SIyke et ai. for the complexes were
cystine, 1: 1; arginine, histidine, and lysi~e, ,1: 5: 1; and glycine, proline, and tryptophan, 3: 1.
..Tungstic acid and phosphoric acid f~rm a series of complex acids of the general formula PaO!' nW03 mHaO or
H 3P04n/2(WOs)m'H20. According to Wu (1920) the complex acids are stable only if the values for n are 18 or
larger. The degree of hydration varies, but the values for m (or m') commonly fall between 14 and 25. Commercial
phosphotungstic acid consists principally of PaOs 24WOs mHaO but may contain as much as 10 per cent of the
"18" acid. The structure of p]losphotungstic acid purified by ether extraction was reported by Van Slyke et ai. to
be PaOs 24W03 17HaO.
7. The p'recipitatlon of glutamic acid and aspartic acid as their calcium salts was first described by Ritthausen (1869a). Ritthausen's procedure has been modified by Foreman (1,914), Chibnall et ai. (1940), and Bailey
et ai. (1943). Barium hydroxide has been used in place of calcium hydroxide (Dakin, 1920; Jones and Moeller~
1928). The acidic amino acids 'have been separated from other amino acids by adsorption on a synthetic anionexchange (polyamine-formaldehyde) resin, Amberlite (Cannan, 1944).
8. The relatively high solubility of phosphotungstic acid in a mixture of amyl alcohol and diethyl ether was
reported by Van Slyke (1915a).
9. Malaprade (1934) observed that carbon chains with adjacent hydroxyls are split by periodic acid, and
Nicolet and Shinn (1939) and Shinn and Nicolet (1941) found that amino acids with adjacent amino and hydroxyl
groups are split similarly. The rate of reduction of l'eriod;'c acid by nonhydroxylated amino acids is only onethousandth that by serine. The products ammonia, aldehydes, and iodate are yielded in the oxidation of
- 40-
hydroxyamino acids. Glyoxylic acid (CHOCOOH) and formaldehyde are formed from serine; glyoxylic acid and
acetaldehyde from threonine. Van Slyke, Hiller, and MacFadyen (1941) and Van Slyke, Hiller, MacFadyen, Hast.. ings, and Klemperer (1940) analyzed serine, threonine, and hydroxylysine quantitatively by means of the periodic
acid reaction.
10. Fromageot and Heitz (1939) reported that leucine and valine could be determined by deamination, oxidation of the hydroxy acids to acetone, and ph/otometric snalyj>is of the acetone-salicylaldehyde condensation product.
Block, Bolling, and Kondritzer (1940) found that acetone (from valine and leucine) and methyl ethyl ketone (from
isoleucine) could be !\eparated by precipitAting acetone as its HgS04 complex. Both acetone and methyl ethyl
ketone react with salicylaldehyde in acid solution to give colored products, but only acetone gives a colored
product in alkaline solution. According to Fabinyi (1901>a, 1900b) the colored complex (A) is formed from two
molecules of salicylaldehyde and one of acetone and (8) from two molecules of each substance. Analogous compounds are formed from methyl ethyl ketone.
6-oH
6-oH ~~l~oC~
CH=CH-C==CH
CH==CH--CO--CH==CH
(Al
H0-6
di-o-hydroxydibenzalacetone
(8)
11. Pyridine, dihydropyrrole, and'pyrrolidine give no color with isatin (Fromm, 1935). According to Grassman and von Arnim (1934, 1935) is at in (A) condenses with proline to yield the blue-colored complex (B) and with
hydroxyproline to yield an analogous pr6duct. Only a red-colored solution has been observed with hydroxyproline
under the authors' conditions of this test.
(X
,
o
II
C
CX)c~o
)c==o
C
;N--H
H-C-C
H-C--C
(X
isotin
(Al
(8)
II.
C",/C=O
N
12. Proline is oxidized to give a colored product when tested with p-dimethylaminobenzaldehyde but not with
isatin, according to Guest (1939). The yields of pyrrole from proline and hydroxyproline vary Widely under different
conditions (Guest and McFarlane, 1939).
13. This test for cystine (Sullivan, 1929a, 1929b; Sullivan and Hess, 1929) depends upon the reduction of
cystine with sodium cyanide
R-S-S-R+NaCN-R-S-Na
cystine
sodium
cysteinate
R-SCN
S-cyanocysteine
and upon the condensation of cysteine with sodium 1,[Link]-4-sulfonate to give a red-colored solution.
Sodium hydrosulfite (Na2S20~ is added to intensify the color. Tin or zinc and hydrochloric acid, titanium chloride, sodium
- 41-
amalgam, and other reducing agents may be used, although the technique with sodium cyanide is most simple.
Since reduction with sodium amalgam yields 2 moles of cysteine per mole of cystine, both cystine and cystein
can be determined by reducing different samples with NaCN and Na-Hg. The test for cysteine is highly specific
since three functional groups ( - SH, - NH2, and - COOH) are required.
14. McCarthy and Sullivan (1941) have shown that a red-colored solution is formed when methionine is heated
in an alkaline solution of sodium [Link] [Na2(NO)Fe(CN)s1 and the mixture is acidified. The sensitivity of
the reaction is 20 to 50 parts of methionine per 1,000,000 parts of solution. Many common amino acids give no
color, but histidine, histamine, and carnosine (.B-alanylhistidine) give methionine-like reac~ons. Tryptophan
yields a reddish-brown color which, unlike the red colors given by methionine and histidine, is extractable 'l'!ith
butyl alcohol. Glycine has been added to the reagent to eliminate the interfering histidine color. There are no
interfering substances in the present procedure, since both histidine and tryptophan have been removed previously
in the solution tested. Modifications of this test have been. proposed by Csonk, and Denton (1946) and by Horn,
Jones, and Blum (1946).
15. Millon (1849) observed that a red solution or precipitate was formed when proteins were heated with a
solution of mercury in nitric acid, and Hoffmann (1853) found that tyrosine and other phenols gave this te,st. The
red-colored complex probably is the mercury salt of nitrotyrosine (Calvery, 1938). Chlorides and alkali interfere
by precipitating the mercury.
16. Tryptophan forms a pink to violet color when it is treated with concentrated sulfuric acid and glyoxylic
acid or fotmaldehyde by the Adamkeiwiez (1874), the Hopkins-Cole (Cole, 1903; Hopkins and Cole, 1901a, 1901b,
1903), or the Acree-Rosenheim (Acree, 1906-1907; Rosenheim, 1906a) procedure. The historical development of
this test has been discussed by Harvey, Miller, and Robson (1941), who suggested that a compound (A) first
formed is oxidized to a blue pigment.
eOOH
NH
J3 -carboline-4-carboxylic acid
Rydon (1948) has studied the analogous color reactions of methyl-substituted derivatives of tryptophan.
17. Kapeller-Adler (1932) proposed that the colored compound (A) is formed by the indicated reactions:
Q
eH 2
I
eHN~
I
eOOH
HN0 3
Q
eOOH
4-nilrobenzoic
ocid
NO
HN0 3
Q-NO'
N~OH
-NO'
eOOH
eOOH
3,4 -dinitrobenzoic
ocid
nilronilro50benzoic
ocid
phenylolonine
NH40H
c7
NOONH 4
eOONH 4
Support for this mechanism was presented by Block and Bolling (1939), who prepared diaci-3,4-dihydrodinitrophenylalanine by these reactions. Modified procedures have been described by Kuhn and DesnueUe (1937) and
Blode and Bolling (1945). Of the amino acids yielding interfering colored nitration products, Kapeller-Adler
removed tryptophan by acid hydrolysis, tyrosine by oxidation with permanganate, and histidine by precipitation
with phosphotungstic acid. Knight and Stanley (1941) corrected for tryptophan, while Brown (1944) precipitated
tryptophan with mercuric sulfate. In the present procedure these interfering amino acids are removed before
applying the phenylalanine test.
18. Sakaguchi (1925a, 1925b) observed that arginine reacts in alkaline solution with sodium hypochlorite
and a-naphthol to give a red-colored product assumed to have the structure shown on following page.
- 42-
NH-C-N-O
" 0I
CH2
NH
~H2
~H2
lvU
CHNH 2
COOH
Sakaguchi and later workers (Poller, 1926; Weber, 1930) found that glycocyamine ~uanidinoacetic acid), monomethylguanidine, dimethylguanidine, and trimethylguanidine gave colored solutions with the Sakaguchi reagent,
while guanidine, nitroarginine, creatinine, methylguanidinoacetic acid, and asymmetrical di-and trimethylguanidines
did not. Substitution of hypooromite for hypochlorite simplifies the preparation of the teagent without reducing its
specificity (Weber, 1930). The color formed by arginine is detectable in 1:2,500,000 dilution. In order to remove
interfering substances, arginine has been separated by electrodialysis (Macpherson, 1942) and by adsorption on
Permutit (Dubnoff, 1941).
19. Knoop (1908) observed that histidine and histamine but not imidazo1eacetic acid, imidazolepropionic,
acid, or imidazolelactic acid gave a positive test with this reagent. According to Kapeller-Adler (1934) little or
no color is produced with carnosine (S-alanylhistidine), a..amino-N-methylimidazolepropionic acid, or a-methylamino-S-imidazolepropionlc acid. It has been reported that two (plimmer and Phillips, 1924) or more (Lieben and
MiiUer, 1928) bromine atoms react per molecule of histidine, but the nature of the reaction is unknown. Tryptophan,
tyrosine, glycine, and some other amino acids alter the shade of color (Woolley and Peterson, 1937-1938), but these
effects. are minimized in the present test.
20. Kossel (1898) first identified lysine by precipitation as its picrate. Lysine picrate precipitates readily
because of its low solubility [340 mg. at 0 (Tristram, 1939) and 540 mg. at 21 to 22 (Lawrow, 1899) per 100 mI.
0
of water]. According to Block and Bolling (1945) lysine picrate decomposes (with a slight explosion) at 250 if
impure and at 266 if pure. Picric acid melts at 121 to 122 0.
21. This proced~re is a modification of that given by Boulet, Nelson, and McFarlane (1947). These authors
eliminated interfering amino acids by adsorbing lysine and arginipe (which gives no color with the reagent) on an
ion-exchange column (Decalso) and eluting them with sodium carbonate solution.
22. Glutamic acid hydrochloride (m.p. 210) was first identified by Hlasiwetz and Habermann (1873). Waelsch
and Prescott (1945) have detected glutamic acid by oxidizing it with ninhydrin to formylpropionic acid and coupling
this product with 2,4-dinitrophenylhydrazine. On the addition of alkali a red-brown color is formed. Of the other
amino acids tested, only aspartic acid gave an oxidation product which formed a [Link]. Glutamic acid has been
determined by oxidation with chloramine-T to succinic acid, enzymatic oxidation of the latter to camon dioxide,
and estimation of the carbon dioxide (Cohen, 1939).
23. Pucher, Vickery, and Wakeman (1934) observed that aspartic acid could be determined as a blue pigment,
which was formed, according to Suomalainen and Arhimo (1947), by the indicated reactions:
N02
COOH
COOH
Br
~H=N-NH-b-NO'
I
CHNH 2
I
CH 2
I
COOH
I
CHOH
I
CH 2
I
COOH
I
CHBr
I
CHO
aspartic
acid
malic
acid
dibromoacetaldehyde
CH=N-NH
OH-~ blue
pigment
NO,
NO~
glyoxal-2,4-dinitrophenylosazone
Arhimo (1939) found that aspartic acid could be oxidized directly with bromine without prior treatment with nitrous
acid and that tyrosine and dihydroxyphenylalanine were the only interfering amino acids.
- 43-
24. Ritthausen (1869b) was the first to observe tha~ aspartic' actd may be readily isolated as its cupric
hydrate. It has been reported that the hydrate contains two (Bergmann ang Niemann, 1937), four and one-half
(Bailey et aZ., 1943; Ritthausen, 1869b), and five (Abderhalden and Weil, 1911) molecules of water of hydration.
25. According to Shinn and Nicolet (1941), "the acetaldehyde which is produced almost immediately from the
[Link] of periodate on threonine can be quantitatively carried over (without important contamination by formaldehyde) by aeration at pH close to 7 and preferably at pH 7.,0 to 7.2." Although Eegriwe (1933), Barker and Summerson (1941), and Miller and Muntz (1938) have shown that propionaldehyde, glyceric aldehyde, pyruvic acid,
methylglyoxal, dihydroxyacetone, and other aldehydes and ketones react with p-hydroxydiphenyl in concentrated
sulfuric acid to give red-purple colors, interfering substances of these types are not pr!:lsent. Formaldehyde gives
a blue-green color which would obscure the color given by'acetaldehyde.
26. Eegriwe (19~7) observed that formaldehyde, but not other aldehydes, reacts with chromotropic acid
8-dihydroxynaphthalene-3,6-disulfonic acid) to give a violet-l'ose color detectable at a dilution of 1:360,000.
~uis test was employed by Boyd and Logan (1942) for the quantitative determination of serine.
27. According to Block and Bolling (1945) methyl ethyl ketone (formed from isoleucine) gives no color with
salicylaldehyde in alkaline solution.
28. The composition of the acetone-mercuric sulfate complex has been reported as 2HgS043HgO(C'H3)2CO
(Denig~s, 1898a, 1898b) and 3HgS045HgO2(CH,)2CO (Van Slyke, 1917). Methyl ethyl ketone is not precipitated
in the concentrations present (Block and Bolling, 1945).
29. Alexander and Seligman (1945) determined alanine by oxidizing if with ninhydrin to acetaldehyde and
estimating the latter by photometric analysis of its p-hydroxydiphenyl complex. Acetaldehyde may also be determined as its bisulfite complex (Virtanen and Rautanen, 1947). Only the volatile aldehydes from leucine,
norleucine, ,and norvaline intetfere with the former procedure, and their color intensities may be decreased by
forming the' chromo gens at a higher -temperature {37"}. Aspartic acid yields acetaldehyde at pH 4 but does not react
"It pH 5.5. Alanine is quantitatively converted to acetalde~yde on boiling it with ninhydrin for 1 hr. at pH 5.5
(Alexander and Seligman, 1945). Glycine is qua~titatively converted to formaldehyde by boiling it with ninhydrin
for les~ than 1 hr. at pH 5.5 (Alexander, Landwehr, and Seligman, 1945) and 1.5 hr. at pH 1 (MacFadyen, 1'945).
'-The aeration procedure employed in the test (see Note 25) minimizes interference from glycine.
30. MacFadyen (1945) and Alexander, Landwehr, and Seligman (1945) determined glycine by oxidizing it with
ninhydrin, condensing the formaldehyde with chromotropic'acid, and estimating thJ color'intensity of the product
photometrically.
31. Accordingto Zimmerman (1930) and Klein and Linser (1932a) the o-phthalaldehyde test is given by
glycine, tryptophan, cystine, arginine, alanine, asparagine, and ammonium salts, although only the colored
products from glycine, tryptophan, and ammonium ion are extracted by chloroform. Since tryptophan, cystine; and
arginine have been removed, they do not interfere in the present test. This color reaction has been utilized by
Patton (1935) for the determination of glycine in'proteins.
- 44-
EXPERIMENT 22
PREPARATION OF CASEIN FROM MILK
Casein, first recognized by Scheele (1780) as a protein in milk, has been most commonly prepared
by Hammarsten's (1873-1814) method. _Dilute acetic acid is added to skim milk, the precipitate is"dissolved in dilute sodium hydroxide, this process is repeated several times, ~nd the product is washed
with ethanol and dried. Van Slyke and Bosworth (1913) modified the Hammarsten procedure by dissolving the casein in ammonium hydroxide and adding ammonium oxalate to precipitate traces of calcium as
calcium oxalate. Van Slyke and Baker (1918) avoided possible decomposition by alkali by precipitating
casein with lactic acid and centrifuging at pH 7.0 to remove calcium and magnesium phosphates. To
prevent denaturation of the casein, Cohn and Hendry (1943) avoided the use of hot organic solvents and
a pH more alkaline than 6.3. Clark, Zoller, and Dahlberg (1920), Zoller (1921), and Northrop (1923)
precipitated casein at pH 4.6 with, a mixture of hydrochloric anicitric acids. According to Van, Slyke
and Carpenter (1924) a high-purity product free from inorganic salts is obtained by electrodialyzing
casein suspensions.
Most casein preparations are mixtures of proteins, according to the solubility studies of Linderstrom-Lang and Kodoma (1925) and the electrophoretic experiments of Warner (1944). Casein is used
widely for animal and microbiological experimentation. The industrial importance of casein in the manufacture of glue, adhesives, plastics, textile fibers, and coated papers has been reviewed by Sutermeister and Brown (1939).
Casein is to be prepared essentially by Dunn's (1949) method. Hydrochloric acid is added to s}dm
milk to pH 4.6 (isoelectric point), the suspension is centrifuged, and the precipitate is washed w~th
water, ethanol, and. ether. The product is not an entity,
... but it has reasonably constant composition and
properties.
- 45-
NOTE
1. Contamination of the milk or the casein with dirt or other insolli>le impurities should be avoided, since
they will be retained in the final product. The casein [Link] nit. be allowed' to dry at any time prior to the final
step, since it will turn to a dark-colored, hard, lumpy tna$s. If any of the ,solutions are stored during the course-of
the experiment, they should be preserved in the refrigerator, but not longer than 24 hr. to avoid bacterial or mold
contamination.
,EXPERIMENT 23
DETERMINATION OF PHOSPHORUS IN PURIFIED CASEIN
The literature on methods for the gravimetric. titrimetric. and photometric determination of phosphorus has been reviewed by Peters and Van Slyke (1946). Phosphorus is determined gravimetrically
as strychnine phosphomolybdate. ammonium phosphomolybdate. magnesium ammonium phosphate. or
magnesium pyrophosphate and volumetrically by titrating these compounds with standard acid. alkali;
or permanganate. Phosphate in urine and fertilizers is determined by precipitating phosphate as
Ur 20 3(P0 4)2 with an excess of a standard solution of a soluble uranium salt'. Photometric methods have
been described based on the color intensities of'molybdivanadophosphate (Kitson and Mellon, 1944a.
1944b) and other molybdenum complexes.
Hammarsten (1883) reported that casein contained from 0.83 to 0.88 'Per cent of phosphorus, and,
30 years later. values ranging from 0.71 to 0.85 per cent were found by Van Slyke and Bosworth (1913)
and Van Slyke and Baker (1918). It has become evident, however, from the work of Berggren (1932),
Cherbuliez and Schneider (1932), and LinderstrolIl-'Lang (1928, 1929) that purified casein fractions
'
contain from 0.15 to 2.3 per cent of phosphorus.
The phosphorus content of purified casein, prepared in Exp. 22 and ashed by the wet method of
Neumann (1902-1903), as modified by King (1932), is to be determined by Brigg's (1922) modification
of the Bell-Doisy (1920) photometric method.
EXPERIMENTAL
Obtain approximately 0.5 g. of purified casein and assignment to a visual colorimeter.
Transfer 0.5 g. (weighed to 1 mg.) of the casein quantitatively to an 8-in. pyrex test tube, using
the irunimum volume of distilled water as rinse fluid. Place a glass bead in the tube and in one additional tube to be used only for reagents. Transfer to each tube from a 10-mI. graduate 5 mI. of a 1: 1
mixture of concentrated sulfuric acid and nitric acid. Heat each tube in a fume hood (or under an inverted funnel connected to a water aspirator) with a microburner (Tirrill burner with the barrel top
removed) until the brown fumes of nitrogen oxides initially formed are replaced by the dense fumes of
sulfur trioxide. (Heating time required is about 20 min.) Heat each tube for an additional 10 min.,
regulating the flame so that the mixtures boil smoothly. Cool the tubes (unknown solution is browncolored) under running water. Add 5 drops (Cautiously) of 60 per cent perchloric acid to each tube.
Heat each tube until white fumes appear and for an additional 5 min. Both solutions 'should be colorless at this point. If not, repeat the perchloric acid treatment.
Cool the tubes to room temperature. Add to each tube 10 ml. of distilled water, 3 drops of phenolphthalein indicator solution, and sufficient concentrated ammonium hydroxide (about 12 ml.) to produce
a faint-pink color. Add 6 N sulfuric acid dropwise to each tube until the color is discharged.
Transfer the contents of each tube quantitatively to separate 250-ml. volumetric flasks and fill
each flask to the mark with distilled water. Mix the contents of each flask thoroughly. Pipette into
separate SO-ml. volumetric flasks (a) 10 ml. of the blank solution, (b) 10 mI. of the unknown solution,
(c) 5 ml. of the unknown solution, (d> 10 ml. of a standard phosphate solution containing 0.015 mg. per
mI. of phosphorus, and (e) 10 ml. of a standard phosphate solution containing 0.025 mg. of phosphorus per
milliliter. Submit flask (c) to the instructor, who will add a known amount of phosphate.
Pipette into each of the five 50-mI. volumetric flasks 5 ml. of each of the following solutions in
the order given: molybdic acid-sulfuric acid reagent, 1 per cent hydroquinone solution, and freshly
- 47-
pre~ared 20 per cent sodium sulfite solution. Fill each flask to the mark with distilled water, mix the
contents of each flask thoroughly, and allow the solutions to stand for 30 min. Compare the color intensity of each blue-colored unknown ~olution in tu..rn (visual colorimeter) with that of the standard
phosphate solution which matches the unknown more closely. The,blank solution should not be colored
detectably.
~XPERIMENT 24
DETERMINATION OF ACID NUMBpR OF UNKNOWN LIPID
"Acid number" is defined as the milligrams of potassium hydroxide required to titrate the free
fatty acids in 1.00 g.' of lipid. Acid numbers provide a basis for estimating the quality of fats and oils,
ev~n though they vary widely for samples differing in source, age, and storage conditions. According to
Lewkowitsch (1922), the acid numbers of 27 types of semidrying oils (see the table facing page 246 in
Lewkowitsch) varied from 0.53 to 57.4 and of 15 fish oils (see the table facing page 435inLewkowitsch)
(from 0.18 to 21.6.
Increase iI~ acid number and rancidity may occur simultaneously, owing to the liberation of butyric, oleic, and other fatty licids from the triglycerides and aldehydes, ketones, and other types of
products by oxiClation. Aromatic amines, cysteine, ascorbic acid, pyrogallol, carotene, a-tocopherol,
and other antioxidants have been employed-to retard oxidation of fats in lard, milk, and other food
products.
The acid number of an unknown lipid is to be determined in this experiment.
EXPERIMENTAL
Obtain approximately 8 g. of an unknown lipid (oil). 'Fransfer the oil to a sample bottle fitted
with a bored cork and medicine dropper. Weigh the bottle and cork ass~mbly to 1 mg. Transfer about
2 g. (80 drops) of the oil to a 125-ml. conical flask, replace the cork assembly, and weigh the sample
bottle with cork assembly. Transfer a 2-g. quantity of the oil to each of two additional flasks in the
manner described.
. Pipette a_ 25-ml. aliquot of 95 per cent etlfanol to each of two clean, dry 125-ml. conical flasks
and to each of the three flasks containing the unknown oil. Add 8 drops of phenolphthal~in indicator
solution (from a dropping bottle) to one of the five flasks, and heat the mixture to boiling over a wire
gauze, using a small flame. Titrate the hot solution with standard approximately 0..1 N potassiu,m
hydroxide delivered from a 10-ml. microburette. During the titration rotate the flask held by means of a
clamp. The end point is a faint-pink color which persists for 30 sec. Titrate the solution in each of
the four remaining flasks similarly.
- 49-
EXPERIMENT 25
DETERMINATION OF SAPONIFICATION NUMBER OF UNKNOWN,LIPID
EXPERIMENTAL
Obtain approximately 2 g. of an unknown lipid. Transfer about 0.4 g. (weighed to 1 mg.) (about
16 drops) to each of three clean, dry conical flasks by the technique described in Exp. 24. Pipette
into each flask 2S ml. of 95 per cent ethanol and 25.0 ml. of standard, approximately 0.5 N potassium
hydroxide in 9S per cent ethanol.
Clamp each flask in a water bath, ins~rt a Hopkins condenser (see Distillation in the Appendix)
in each flask, heat the bath sufficierltly to reflux the alcohol gently, and continue heating for 30 min.
\
Remove the flasks and rinse the alcoholic liquid from the condensers into their respective flasks with
distilled water from a wash bottle. Add 3 drops of 0.1 per cent phenolphthalein indicator solution
(from a dropping bottle) to each flask, and titrate the excess alkali in each flask with standard approximately 0.5 N sulfuric acid delivered from a SO-ml. burette.
EXPERIMENT 26
DETERMINATION OF IODINE NUMBER OF UNKNOWN LIPID
The "iodine number" of a lipid is defined as the grams of iodine absorbed per 100 g. of the lipid.
Historical studies on the determination of degree of un saturation of lipids have been reviewed by Lewkowitsch (1921). In 1857 Cailletet added a solution of the lipid in --alcohol to a mixture of bromine and
potassium hydroxide until the bromine color was discharged. Since this solution was unstable, Mills
and coworkers added an excess of a solution of bromine in carbon tetrachloride and determined the
excess bromine by adding iodide and titrating the liberated iodine with thioi;ulfate in the presence of
starch. Since bromine was substituted, as well as added, under these conditions, it was necessary to
correct for the former.
The now-classical procedures were proposed by Hubl (1884), Wijs (1898), and Hanus (1901). By
the H\ibl ptocedute a solution of ilie lipid in chlo~ootm is mixed with a solution of iodine and HgC12 in
ethanol, the mixture is' allowed to stand for 2 hr. in the dark, iodide is added, and the liberated iodine
is titrated with thiosulfate in the presence of starch. According to the Wijs method a solutIon of the
lipid in chloroform or carhon tetrachloride is mixed with a solution of iodine and an equivalent quantity of
chlorine in glacial acetic acid, the mixture is allowed to stand for 10 min. in the dark, iodide is added,
and the liberated iodine is titrated with thiosulfate in the presence of starch. The Hanus m,ethod is
essentially the same as that of Wijs except that the iodinating solution contains bromine in place of
chlorine.
Nearly theoretical iodine numbers of pure unsaturated fatty acids are obtained by each of these
methods. On the other hand the iodine numbers found for lipids vary considerably, as shown by reports
referred to by Lewkowitsch (1921) and Elsdon (1926). Iodine numbers by the Hanus method usually are
from 2 to 4 per cent lower than by the Wijs method. The Wijs method has been widely used in England
and Europe, and ,it is the official method of the American Oil Chemists' Society. The Hanus method is
used extensively in the United States, and it is the official method of the American Association of
Official Agricultural Chemists.
Iodine numbers vary from 120 (tobacco seed) to 200 (linseed) for drying lipids, 95 (radish seed)
to 140 (grape seed) for semidrying lipids, 70 (paradise nut) to 115 (cherry seed)'Jor nondrying oils, 35
(porpoise) to 180 (cod liver) for marine-animal oils, 70 (horse's foot) to 130 (chrysalis) for terrestrial-animal oils, 5 (myrtle wax) to 95 (chaulmoogra) for vegetable lipids, and 25 (butter) to 110 (rattlesnake)
for animal fats, according to data given by Lewkowitsch (1901, 1921), Elsdon (1926), and Jamieson
(1943).
According to Scotti (1938) iodination proceeds by the reactions shown in the equations below. ;'
212 + 2Hg(OAc)2 + 2H 20 - HglOI + HIO + 3HOAc + HgIOAc
-CH=CH- + HglOAc __ -
CHI-CHHgOAc-
The Winkler and the Aschmann modified iodine methods, the Ka~fman bromine method, and the
Rosemund and Kuhnhenn pyridine sulfate dibromide method have been discussed by Elsdon (1926) and
Jamieson (1943).
- 51-
The iodine number of an unknown lipid is to .be determined in the present experiment by the Norris
and Buswell (1943) rapid modification of the Hanus (1901) method. Equations for the reactions are
given below. The probable mechanism of the catalytic action of mercuric acetate is shown on page 51.
la + Bra -+ 2IBr
Hanus solution
EXPERIMENTAL
Obtain about1.2 g. of an unknown lipid (oil). Transfer 0.2 g. (weighed to l mg.) (about 8 drops)
to each of three thoroughly dried 25().ml. glass-stoppered flasks by the technique described in Exp. 24.
Add 10 nil. of chloroform to each of the three flasks and to two other (empty) flasks. Rotate the flasks
until the samples dissolve. Draw up Hanus solution (iodine and bromine in glacial acetic acid) above
the mark in a 25-ml. pipette, using the house vacuum or a water aspirator (Caution: never by mouth,
since the liquid and vapors are exgemely corrosive). Disconnect the rubber tubing attached to the
pipette, allow the solution to fall to the mark, and deliver 25.0 ml. to one of the five glass-stoppered
flasks. Pipette 10.0 ml. of a 2.5 per cent solution of mercuric acetate in glacial acetic acid into the
flask by the described technique. Stopper and shake the flask. ,Allow the ~olution to stand for 3 min.
During this time fill a 50-ml. burette with standard 0.1 N sodium thiosulfate solution. Immediately
transfer 10 ml. (10-ml. graduate) of 15 per cent potassium iodide solution and 100 mI. (100-ml. graduate)
of distil1~ water to the flask. Immediately (since potassium iodide is appreciably oxidized in a few
minutes by dissolved oxygen) titrate the liberattfd iodine with the standard thiosulfate solution to a
pale yellow-colored end point, add 2 mI. of O.S per cent soluble starch solution, and continue the titration to the disappearance of the blue color. Near the end of the titration, stopper and shake the flask
between additions of thiosulfate.
EXPERIMENT 27
ISOLATION OF CHOLESTEROL FROM BEEF SPINAL CORDS
Accounts of the discovery, the early history, the chemistry, and the physiological properties of
cholesterol and other sterols have been given by Bills (1935), Heilbron (1936), Butenandt (1936a-1936c),
Callow (1938), Strain (1943), Fieser (1949), Sobotka (1938), and Friedmann (1937).
The sterols occur free and combined as fatty acid esters in animals (zGosterols), plants (phytosterols), and yeasts and fungi (mycosterols). Cholesterol, originally called "cholestrine" from the Gr~k
words ""chole",(bile) and "stereos" (solid), was first isolated from gallstones by Poulletier de la Salle
in 1782. Cholesterol is abundant in brain and nerves, where it may have a special function. Okey (1945)
has shown that egg yolk ,and brain are the only common foods containing more than 1 per cent cholesterol.
Both plants and animals synthesize cholesterol, but, according to Schoenheimer (1932), mammals utilize
plant 'sterols with difficulty, if at all.
Examples of sterols occurring in natural products are coprosterol (C 27 H4I O) in feces, ergosterol
(C U H44 0) in yeast, and stigmaste'rol (Ca9H480) in soybean. Types of natural sterols are the bile acids,
se~ hormones, toad poisons, cardiac glycosides, sapogenins, and adren~ hormones. Examples are chollc
acid (C1~400S) in bile, testosterone (C 19H2IOa) in urine of males, estrone or theelin (C lI HaaO,,) in urine
of pregnant females, bufotalin (CUHa706) in the parotid secretion of the toad, digitoxigenin (C a3 Hu O,,) in
cardiac glycoside from foxglove (Digitalis purpurea), digitogenin (C 27H440, ) in noncardiac glycoside
from foxglove, and corticosterone (C2oH2904) in the adrenal cortex. Cholesterol may be the parent substance of some, if not all, of these sterols.
The sterols are derivatives of the hydrocarbon cyc1opentanoperhydrophenanthrene. The structural
formula of choles"terol (C U H460) containing this nucleus is sh_own below.
I
C-CH2-CH2-CH2-C'
20
lsi
CH3
25 (
23
22
24
CH3
?I--_---"'I'[Link]
21
27
HO
Cholesterol is readily prepared crystalline from organic-solvent extracts of brain from sheep or
other animals (Baumstarck, 1885; Rosenheim, 1906a; Tebb, 1906), althoug~ it is prepared commercially
from beef spinal cords. Ether was first employed as the solvent, but acetone and ethylene dichloride
(Porsche, 1945) are used commonly because lecithin and other phospholipids are less soluble in these
solvents than in. ether.
In the present experiment cholesterol is to be isolated from acetone extracts of beef spinal cords
essentially by the method of Giesy (1920).
EXP ERIMENTAL
Transfer about 400 g. of beef spinal cords, preserved in the refrigerator, to a meat grinder. Grind
this inaterial as fine as possible. Transfer the ground material (including the adhering material) to a
tared evaporating dish or pyrex plate. Add 400 g. of filter aid and knead. the mixture to a homogeneous,
- 53-
nearly dry powder. Dry the powder at 75 to 80 for 40 hr. or longer. Cool the dish (and contents) to
room temperature and weigh the dish (and contents).
Transfer the dry material to a 2"liter beaker. Add 750 mi. of acetone and stir the mixture at intervals for 10 min. Decant and filter the supernatant liquid. Collect the filtrate in a 2-liter round-bottomed
flask. Add 500 mI. of acetone to the residual material in the beaker, stir the mixture for 10 min., and
filter the supernatant liquid. Repeat the extraction process using '500 mI. of acetone. 2
Set up the distillation apparatus shown in Fig. 6 (see Distillation in the Appendix) and distill the
combined acetl,me filtrates until 1 Hter of acetone distillate has collected. Repeat the described extraction process twice, using 500-ml. portions of the recovered acetone as solvent. Di-still the combined
acetone filtrates from a 2-liter round-bottomed flask until crystals appear in the flask. Immediately
transfer the hot acetone suspension to a beaker, allow the mixture to cool to room temperature, and
filter the suspension on a Buchner funnel. Distill th~ filtrate until crystals appear in the hot liquid and
collect the (second) crop of crystals as previously described. Transfer the acetone filtrate to the special
container.
Transfer the first and second crops of crystals to a 250-ml. flask. Add 50 mi. of 0.5 N KOf! in 95
per cent ethanol, attach a reflux condenser, and reflux the mixture (water bath) while adding sufficient
ethanol through the condenser to keep the solid in solution. Continue refluxing for 40 min. to saponify
the cholesterol palmitate present as impurity. Remove the condenser, allow the solution to cool, and add
glacial acetic acid (about 1.5 mI.) dropwise until the solution is neutral to litmus paper. Allow the mixture to cool for 30 min. in an ice-water bath, filter .the !Suspension on a Buchner funnel, and recrystallize
the moist product from the minimum volume (about 10 ml. per g. of crude cholesterol) of 95 pet cent
ethanol. Filter the suspension on a Buchner funnel and apply suction until the solid is nearly dry.
Transfer the product to a watch glass, cover the watch glass with a filter paper, and dry the product .
overnight at 50. Submit the product to the instructor in a tared, labeled sample bottle. Determine the
melting point (see Melting Point in the Appendix) of the samples furnished by the instructor.
NOTES
1. The authors are indebted to Dr. Sven Lassen, Van Camp Laboratories, Terminal Island, Calif., for helpful
suggestions.
2. Dry the residual solid in air and place the dry product in the container provided for that purpose.
EXPERIMENT 28
PREPARATION OF STARCH FROM POTATOES
The early history of starch has been reviewed by Herstein (1911). The constitution and properties
of starch have been discussed by Meyer (1942), Hassid (1943, 1945a, 1945b), Brimhall and Hixon (1943),
Hixon and Rundle (1944), Haworth (1946), and Kermack (1946).
Ancient people used starch in cosmetics, in medicines, and for stiffening fabrics. The preparation
of starch fr<:>m grain was described by Cato in 184 B.C. In the nineteenth century it was found that starch
is converted to sugar (glucose) during malting and by acid hydrolysis. Amylbse and amylopectin were
recognized as constituents of starch by Nageli in 1858 and were separated by Maquenne and Roux (1903).
By current procedures the relatively soluble amylose is extracted from starch with hot water and is precipitated from the solution with methanol (McCready and Hassid, 1943), butanol (Schock, 1942), or thymol (Haworth, Peat, and Sagrett, 1946). Starches prepared from tapioca, rice, corn, potato, wheat, bean,
and Easter lily contain from 20 to 30 per cent amylose (Bates, French, and Rundle, 1943). Waxy-corn
starch consists entirely of amylopectin (Schopmeyer, 1945).
It was first proposed by Meyer (1942) that the amylose fraction of starch consists of linear molecules, while amylopectin is made up of branched molecules. Amylose contains from 250 to 500 and
amylopectin about 20 to 30 gl~cose residues per chain, as determined by the "end-group" assay method.
By this procedure the starch is completely methylated, the methylated product is hydrolyzed, and the
resulting 2,3,6-trimethylglucose and 2,3,4,6-tetramethylglucose are separated. Since the glucose residue
at the nonreducing end of the chain has four hydroxyl groups which can form stable methyl derivatives,
compared with three such groups within or at the reducing end of the chain, the length of the chain (or
branch) can be estimated from the p'ercentage of the tetramethylglucose obtained. Haworth has shown
that starch chains consist of a-glucopyranose units joined by an oxygen atom (glucosidic linkage)
through the first carbon atom (the reducing group) and the fourth carbon atom of the adjacent glucose
unit. Formulas for starch and for these derivatives are shown below (Hassid, 1945b).
GH20Me
Gf-WMe
GH20Me
r:~
0 C~ o---o-QMe
~~~ ~~~
OMe
GH20Me
GHi)Me
~ ~ QH+
n
Me
OMe
2,3,4,6tetramethylglucose
HOMe
2,3,6-trimethylglucose
GH30H
0
HOMe
2,3,6-trimethylglucose
According to Meyer's modification of Nageli's micellar theory of starch structure, starch granules
are built up of crystalline dendritic units (trichites) consisting of alternate layers of amylose and amylopectin (Sjostrom, 1936). Starch granules are striated concentrically, owing to the nonuniform distri- 55-
,
bution of water in the layers. Canna starch -granules are oyster-shaped and are among the largest known.
_The chemical and physical properties of starches from more ,than 30 different plants have been investi:
gated (Barham, Wagoner, Campbell, and Harclerode, 1946; Bear, 1942; Kreger, 1946; Speich, 1942;
Harris and Jesperson, 1946a, 1946b; Briant, Personius, and Casse~~ 1945; Armitage, 1943; and Lampitt,
Fuller, and Goldenberg, 1948).
The influence of place, season, soil, variety of plant"and other factors on the starch content of
wheat (Barham, Kramer, and Reed, 1943; McCalla and Corns, 1943), sweet potatoes (Boswell et al., 1944),
and other l,llants has been studied. Starch has been manufactured from sweet potatdes (Kimbrough, 1942),
wheat (Dimler, Davis, Rist, and Hilbert, 1944), corn (Kelling, 1944), peas (Hilbert pnd MacMasters, 1946),
Easter lily bulbs (Stuart and Brimhall, 1943), water chestnut (Shafee and Sarin, 1937), and other plants.
Waxy (glutinous) starch (stains iodine red brown rather than the usual blue) is obtained from rice, sorghum, millet, bar~ey, an..d corn (Hixon and Sprague, 1942; MacMasters and Hilbert, 1944).
Starc;;h contains about 0.1 per cent of phosphorus compounds (probably amylose phosphate) (Briggs
and Hanig, 1946). Fatty acids embedded in the starch layers are difficult to remove with fat solvents
(ether) but are easily extracted with hydrophilic solvents (methanol). Fatty materials interfere with the
fractionation of starch, and they modify the iodine absorption, gelling power, and otlier properties.
Thick-boiling and thin-boiling starches, oxidized starch, gelatinized starch, and other, products
are manufactured from corn, kaffir, rice, potato, tapioca, and other plants for use in coating paper
(Bullard et aI., 1947), impregnating textiles, color printing, cold-water pastes, adhesives, and other
uses. Large quantities of starch are converted to glucose for use as food and as a 'source of polyalcohols and organIc acids (Kennedy, 1946).
Starch is to be prepared by macerating potatoes, mixing the macerated product with water, draining
the suspension, and allowing the starch particles
to settle.
,
EXPERIMENTAL
\
Preparation of Purified Starch. Wash, peel, and slice about 600 g. of potato tubers. Thoroughly
macerate 500 g. of the slices by grating them with a metal grater. Collect, all the product including the
expressed liquid as well as the pUlp. Thoroughly mix the ground material in a suitable container
(enameled pan) with 500 ml. of distilled water. Drain the suspension on a d0!1ble layer of cheese~loth,
remove as much as possible of the fluid, and repeat this extraction process until nearly all the starch
has been removed. Use separate beakers for the extracts. Discard the in~oluble residue.
Allow the separate beakers to stand (for! hr.) until the finely divided starch particles have
settled. Carefully pour the supernatant fluids)nto separate containers and preserve them for the tests
described below. Collect the severa\ starch precipitates in one beaker, using 100 mi. of distilled water
as rinse flUid. Allow the'starch to settIe, pour the supernatant liquid into a beaker, and preserve this
liquid for the tests described below. Wash the precipitate with two additional portions of distilled
water and preserve the wash liquids separately for later tests.
Color Changes in Potatoes and Aqueous Extracts from Potatoes. 1 Macerate about 20 g. of sliced
potatoes and permit one half of this material to stand in air. Place the second halfin a beaker, add 50
ml. of distilled water, and heat the mixture for 15 min. at the boiling point. Allow the mixture to stand
in air. Observe the color changes occurring over 24 hr. in the supernatant liquids and products obtained
in this and the preceding section.
Reducing Substances in Potatoes. Test all the extracts and wash liquids obtained in the two
preceding sections by the following method: Heat 5 ml. of Benedict's qualitative sugar re,agent to
boiling in a test tube. No change in color and no precipitation should occur. Add 5 drops of one of -the
extracts, heat the mixture to boiling, and maintain boiling for 1 min. Allow the mixture to cool to room
temperature. Observe any changes. Discard the solutions containing the reducing substances, since
they, will putrefy if allowed to stand.
Nitrogenous Suhstances in Potatoes. Heat to boiling 100 mi. of the first aqueous extract obtained
in the first 'experimental section. Add 50 ml. o( 6 N acetic acid and allow the mixture to cool to room
temperature. Filter (or centifuge) the coagulum, wash the precipitate with 95' per cent ethanol and di- 56-
ethyl ether, and dry the precipitate in air. Weigh the dried material. Decompose it and a sample of
......purified starch by the procedure given in Exp. 21, Step III, using about 0.5 g. of test material.
Perform the following test for nitrogen: Add to the filtrate 2 ml. of potassium fluoride solution
and 1 mi. of saturated ferrous sulfate solution. Allow the mixture to stand for 5 min. and then add
concentrated hydrochloric acid dropwise until the mixture is slightly acid to litmus paper. The appearance 9f a brilliant-blue color or a blue pre~ipitate of Prussian blue {Fe4[Fe(CN)J3} confirms the pres'ence of nitrogen.
NOTE
1. A dark-colored product, called melanin, is formed by the action of tyrosinase on tyrosine and the condensation or'the intermediate qui~ones (Duli~re and Raper, 1930; Raper, 1927).
EXPERIMENT 29
DETERMINATION OF STARCH FACTOR AND STARCH IN A STARCH-PROTEIN MIXTURE
The determination of starch is of interest to sci~ntific workers and agencies concerned with
botanical, horticultural, and other types of problems. Analyses are complicated by the presence of
cellulose, hemicellu10ses, pentosans, gums, mucilages, glucosides, pectins, tannins, and varying
proportions of amylose and amylopectin.
It has been proposed (Krocker, 1846; Polit and Dhar, 1925) that starch may be determined by
hydrolyzing it to sugar, fermenting the sugar, and measuring the loss in weight due to carbon dioxide.
A more practicable procedure is to solubilize the starch with hydrochloric acid, precipitate the starch
with ethanol, and weigh the (dried) product (Sullivan, 1935a). Solubilizing and extracting agents
e~p10yed by other workers include MgC1 2 Qirak, 1935), CaC1 2 (Mannich and Lenz, 1920), malt amylase
(Hanes, 1936), salivary amylase (Hassid, McCready, and Rosenfe1s, 1940), takadiastase (Denny, 1934a,
1934b), NaClO (Balch, 1941), and HCl0 4 (Nielsen and Gleason, 1945).
Litner (1907) and other workers (Mannich and Lenz, 1920; Hopkins, 1934; Clendenning and Wright,
1945) have determined starch on the basis of its optical rotation. Although it has been assumed that
all starches have a specific rotation of 200 (Association of Official Agricultural Chemists, 1945, page
251), Earle and Milner (1944) h~ve shown that it varies from 201 to 204 for different plants.
Several types of procedures have been proposed for the determination of starch as its starchiodide complex (Bates, French, and Rundle, 1943; Kerr and Trubell, 1943; Pucher, Leavenworth, and
Vickery, 1948). Plant tissues are extracted first with dilute ethanol, to remove glucosides and other
substances which yield colored products with iodine, and then with cold hydrochloric acid or hot
calcium chloride solution. Starch, but not other polysaccharides, is preCipitated with iodine, and the
concentration of the resulting blue starch-iodide complex held in colloidal solution is determine~
spectrophotometrically using purified potato starch as standard (see Exp. 36, Note 3, for discussion of
the starch-iodide reaction).
Starch has been determined most commonly on the basis of the glucose liberated on hydrolysis.
The literature on the starch factor published prior to 1896 has been reviewed by Fresenius and Griinhut
(1896). The value 0.936 was established by N9yes, Crawford, Juniper, Flory, and Arnold (1904) and
Pucher and Vickery (1936) confirmed th).s factor. Factors ranging from 0.92 to 1.01 have been reported
(Earle and Milner, 1944; Etheredge, 194'( Sullivan, 1935a). As shown by Sullivan (1935b) and indicated
in Fig. 4, the theoretical starch factor [(C&H1oOs)n -H 2 0 / nC6 H12 0J approaches the limiting value 0.900
as the number of anhydrog1ucose residues per molecule of starch increases. This approximate value
would be found experimentally in the absence of any decomposItion during hydrolysis, since most
starches contain about 80 per cent of branched-chain amylopectin (molecular weight about 500,000) and
20 per cent of straight-chain amylose (molecular weight about 60,000) (Stuart and Brimhall, 1943).
Modifications oLFehling (1849, 1858) procedure have been employed by Fraps (1932), Hartmann
and Hillig (l931),Bish (1929), Hassid, McCready, and Rosenfels (1940), and Yemm (1935) for the determination of glucose in hydrolyzed starch. Whitmore (1934) combined the ferricyanide procedure
of Gentele (1859), as modified by Hagedorn and Jensen (1923a), with the titrimetric determination of
ferro cyanide by ceric sulfate. The last reaction was shown to be quantitative by Furman and Evans
(1929). Because the end point in the titration was not sharp, Whitmoyer employed the internal indicator
CYrzurine G, which is extremely sensitive in acid solution to amounts of ceric sulfate. Hassid (1936,1937)
substituted the internal indicators o-phenanthroline, ferrous sulfate, and setopaline C. The latter indicator is especially useful in the analysis of plant extracts. Hassids method has been found applicable
to quantities of reducing sugars ranging from 0.3 to 3.5 mg.
- 58-
The starch factor is the number by which glucose is multiplied to give the quantity of starch from
IJhich the glucose wa:; derived. Starch is to be determined in this experiment by hydrolyzing it with
lydrochloric acid, oxidizing the liberated glucose with alkaline ferricyanide, and oxidizing the resulting
'errocyanide with standard eerie 'sulfate ,solution.
CeH120e +
Fe(CN)e3-~
Fe(CN)e
4-
Ce4+~
Fe(CN)e4 - +
Fe(CN2e
~0.93
S-
oxid~tion
products
+ Ce3+
STARCH- Q_LC09::
COfWERSON FACTCRS
~
l..L
0.91
10
30
ANHYDROGLUCOSE UNITS
FER STARCH MOLECULE
Fig. 4.
EXPERIMENTAL
Obtain a O.S-g. sample of purified potato starch and a O.S-g. sample of a mixture containing
purified potato starch and protein. Transfer 0.4 g. (weighed to l mg.) of the purified starch and 0.4 g.
(weighed to l mg.) of the starch-protein mixture to separate 250-ml. flasks. Pipette 50 mI. of N hydrochloric acid into each flask, attach a reflux condenser (lO-in.) to each flask, and immerse both flasks
in the same boiling-water bath for 180 min.
Remove the flasks and immerse them in an [Link] bath. Rinse each condenser with distille1
water into its respective flask. When the solutions have cooled to room temperature,1 transfer the contents of each flask quantitatively to separate I-liter volumetric flasks. Add N sodium hydroxide to each
flask until the solution is neutral to bromthymol blue indicator (olive-green color). Mix the contents of
eac)1 flask thoroughly, add distilled water to the marks, mix the solutions thoroughly, and transfer a
lOO-ml. aliquot of each solution to separate conical flasks. Label and stopper the flasks. Store them
in the refrigerator. As soon as possible, but no later than 24 hr., analyze the hydrolysates for glucose.
Pipette 5.00 ml. of one of the solutions into a 50-ml. conical flask. Add 5 ml. of the special
alkaline potassium ferrieyanide solution (preserved in a dark bottle in the refrigerator), mix the solutions, place the flask in a boiling-water bath, and maintain this temperature for exactly 15 min. Cool
the flask to roo~ temperature, add 5 ml. of 5 N sulfuric acid, and mix the solutions. Add 7 drops of
setopaiine C indicator solution (freshly prepared each day) and titrate the solution with standard, approximately 0.02 M eerie sulfate solution (10-ml. microburette). The solution turns a golden-brown
color at the end point.
Repeat this process with additional aliquots of the same solution until consecutive titrations
agree within 0.01 ml-. Repeat these manipulations using 5-mI. aliquots of the other starch solution.
NOTE
1. At this point the solutions may be stored for several days if desired.
- 59-
EXPERIMENT 30
ISOLATION OF [Link] FROM CORNCOBS
The history of xylose has been given by Harding (1923a). D-Xylose, first isolated from wood by
Koch (1886), has been obtained frqm straw, corncobs, flax, coconut 'shells, apricot shells, cottonseed
hulls, and other materials listed by Hudson and Harding (1917, 1918). Early workers obtained xylose
from the polymer xylan; however, Bertrand (1891) foru\d that xylose could be isolated directly from
plants without preliminary separation of xylan.
About 3 per cent yield of xylose was obtained from corncobs (Zea mays) by Stone imd Lotz (1891),
who employed the classical ammonia extraction procedure. LaForge and Hudson (1918) increased the'
yield to 5 per cent by extracting corncobs with hot water and hydrolyzing the xylan with sulfuric acid.
From 10 to 12 per cent yield was obtained by Hudson and Harding (1918), who hydrolyzed corncobs
directly with 7 per cef!t sulfuric acid. Modified methods have been described by Monroe (1919), Ling
and N anji (1923), and Dunning and Lathrop (1945). Firstenberger (1943) has described the large-scale
preparation of D-xylose from cornstalks.
Methods have been reported for the quantitative determination of xylose by selective fermentation
with the microorganism Hansenula suaveolens (Wise and Appling, 1945) and by precipitation as its
benzylidine dimethylacetal derivative (Breddy and Jones, 1945) (see Exp. 36, Test 13, for the latter
test).
D-Xylose is not readily assimilated by man (Blatherwick et ai., 1936), but it is an essential
nutrient for some microorganisms (Carnien, Dunn, and Salle, 1947). Administered to rats, D-xylose
leads to increase in nonfermentable reducing substances in the liver and other organs (Blatherwick
et al., 1936)" and to cataracts (Darby and Day, 1940). Fermentation of xylose by Clostridium acetobutylicum yields butanol, acetone, and ethano!" in quantities varying from 20, 26, and 46 per cent,
respectively, in the early stages to 60, 30, and 10 per cent after 118 hr. (Underkofler, Christensen, and
Fulmer, 1936).
D-Xylose is to be isolated by crystallizing it from a 'sulfuric acid hydrolysate of corncobs.
EXPERI~ENTAL
Weigh 100 g. of corncobs, break ,he cobs into small pieces, and grind them to a coarse powder in
a food grinder. Transfer the powder to 'a 2-liter round-bottomed flask and add 600 ml. of N sulfuric acid.
'Reflux the mixture on a sand bath for 2 hr.
'Allow the suspension to settle, decant the orange-colored supernatant liquid to a Buchner funnel,
transfer the precipitate to the funnel, and wash the precipitate on the funnel with 100 rol. of distilled
water. Transfer the filtrate to a flask and add 85 g. (0.54 equivalent) of powdered Ba(OH)28H20 in 10-g.
portions, with"irequent vigorous stirring to minimize coating the barium hydroxide particles wi~h barium
sulfate. Place the flask in a boiling-water bath. Add 15 g. (0.15 equivalent) of powdered barium carbonate
slowly enough to prevent foaming 1 while stirring and heating the mixture. Continue heating and stirring
for 15 min. Transfer 1 drop of the mixture to a spot plate containing congo red ind~cator solution. If
the mixture is acid (blue color), add 10 g. of the barium carbonate powder and repeat the described
procedure. Continue this process until the indicator solution is colored red (pH 4 to 5).
Filter the hot suspension on a [Link] funnel fitted with a filter paper covered with 1/2 in. of
filter aid. Wash the precipitate thoroughly with two 100-ml. portions of boiling distilled water. Add N
sulfuric acid slowly to the filtrate until it appears that all the bari~m ions have been converted to insoluble barium sulfate. Allow the precipitate to settle, transfer 10 ml. of the supernatant liquid to a
- 60-
test tube, and add N sulfuric acid dropwise. If a precipitate is formed, return the test\solution to the
flask and add a volume of N sulfuric acid calculated to precipitate the barium,ions in the entire solution. Test the supernatant liquid as described and continue this treatment until all the barium ions
have been removed. Test a drop of the final supernatant liquid with congo red indicator. The color
should be red, indicating that the solution has not been acidified beyond pH 4.
Add 1 mi. of glacial acetic acid and 15 g. of decolorizing 'carbon to the sUl;;pension and allow the
mixture to stand at room temperature for,15 min. with frequent stirring. Filter the suspension, wash the
preCipitate with hot distilled water, an~ transfer the filtrate and washings to a 2-liter flask. Set up the
reduced-pressure apparatus 'shown in Fig. 7 (see Distillation in the Appendix) and distill the pale-yell owcolored filtrate 2 under reduced pressure on a boiling-water bath to a thin sirup (about 50-m!. volume).
Pour the sirup with continuous stirl:ing into 3 volumes of methanol and allow the mixture to stand in the
refrigerator overnight or until the next laboratory period. Filter the mjxture to remove any impurities
which may have preCipitated.
Distill the filtrate 3 under reduced pressure on a boiling-water bath to a thick sirup (10 to 15 ml.).
Add .30 ml. of hot methanol to the warm sirup with stirring, transfer the mixture to a beaker, and add
several seed crystals of D-xylose. Allow the (covered) beaker to stand in ~he refrigerator overnight or
until the next laboratory period. Filter Jhe suspension and wash the precipitate with 20 m!. of 7S per
cent methanol.
Dissolve the crude D-xylose in distilled water (0.5 mI. per ~.) and add a drop of glacial acetic
acid and 0.5 g. of decolorizing carbon. Allow the mixture to stand for 10 min. with frequent stirring.
Filter the suspension and pour the clear filtrate into 3 volumes of methanol with vigorous stirring.
Seed the solution with a few crystals of D-xylose and allow it to stand in the refrigerator overnight or
until the next laboratory period.
Filter the suspension and wash the crystals with 7S per cent methanol and with absolute methanol.
Allow the recrystallized product 4 to dry in air in a covered evaporating dish. Submit the product to the
instructor in a tared, labeled sample bottle. Calculate the percentage yield and submit the calculations
to the instructor.
NOTES
1. A few drops of butyl alcohol may aid in disper!Sing the foam.
2. Repeat the decolorizmg procedure with 5 g. of decolorizing carbon if the color of the filtrate is darker
than pale yellow.
3. If the liquid is dark-colored when the volume has been reduced to 40 mI., decolorize it with 1 g. of decolorizing carbon.
0
4. a-D-Xylose melts at 145 and [a~ = +93.6 deg. in 4 per cent aqueous sofutio_n. The equilibrium value
after mutarotation is +18.8 deg. (Bates and associates, 1942).
.
EXPERIMENT 31
ISOLATION OF L-ARABINOSE FROM MESQUITE GUM
L-Arabinose has been isolated from beet pulp, cherry gum, wheat, rye bran, peach gum, [Link] black wattle gum, and other plant materials (Bates and associates, 1942; Harding, 1922). Meg..
quite gum is the most convenient source, and the methods described by Anderson and Sands (1925,
1926, 1941), Anderson. Sands, and Sturgis (1925), and Isbell (referred to by Bates and associates,
1942) are most commonly employed. According to Anderson and coauthors, mesquite gum is the exudate
of Pro sop is juliflora and other species of the mesquite tree native to various parts of Texas, New Mexico, Arizona, and northern Mexico.
Mesquite gum exudes ftom the stem and branches of the tree in irregular pieces weighing 5 to
25 g. When the gum first appears, it is soft and sticky and may run do~'n the branch, but it gradually
dries to a hard, brittle mass. The gum is a pectin-like material which, according to Ehrlich (1917),
is the calcium salt of anhydrogalactomethoxytetragalacturonic acid. Anderson and Otis (1930, concluded that the organic acid consists of arabinose, galactose, and 3-methoxyglucuronic acid
(CHOCHOHCHOCH~CHOHCHOHCOOH) residues in the ratio [Link], respectively. White (1946, 1947)
has suggested that the ratio probably is [Link] 1. He has shown that Ul.e uronic acid component and part
of the arabinose occupy terminal positions in the complex, that arabinose occurs in the furanose configuration, Hnked in the polysaccharide by glycosidic oxygen bridges at the first and second carbon
atoms, and that galactose occurs in the pyranose configuration, linked by oxygen bIidge~ at the first,
third, and sixth carbon atoms, It appears that mesquite gum bas a branched-chain structure concerned
only with galactose units, that the four arabinose units form a branch or tail attached to the remainder
of the p'Olysaccharide and terminated by a residue of arahofuranose, and that the araban fraction of the
gum is attached at the three position of a gaiactose anhydride unit. The position of the methoxy group
In the giucuronic acid and the exact structural relations of the various monosaccharide units have not
been determined.
L-Arabinose is an essential nutrient for some microorganisms (Carnien, Dunn, and Salle, 1947)
and L- and D-arabinose are utilized equally well by some microorganisms (Nicolle, 1944) and the rabbit
(Corley, 1929). Clostridium acetobutylicum ferments arabinose to butanol, acetone, and ethanol in the
ratio [Link] 10, respectively (Undetkcfler and Hunter, 1938).
L-Arabinose is to be isolated by crystallizing it from a sulfuric acid hydrolysate of mesquite gum.
The' gum is partially hydrolyzed by this treatment to L-atabinose and a residual polysaccharide
(gaiactose-meiliox:yglucurollic acid), since the latter contains etber linkages mote resistant to hydrolysis
than the glycosidic linkages of araban (polymer of anhydroarabinose).
EXPERIMENT AL
Weigh 100 g. of coarsely powdered mesquite gum and suspend this material in 600 mI. of distilled
[Link]. Allow the suspension to stand at room temperature with occasional stirring until the gum dissolves completely except fot inso1uble impurities. It may be necessary to allow the mixture to stand
overnight or longer time to dissolve the gum. Strain the suspension using several thicknesses of cl,eesecloth, add to the strained Jiquid 30 mI. of 18 N sulfuric acid, slowly with stirring, and heat the mixtqre
for 3 hr. in a water bath at about 90, with occasional stirring.
Add to the cooled solution 80 g. (0.5 equivalent) of powdered Ba(OH)1'8H 20 in l~g. portions, with
frequent vigOEouS stirring to avoid incomplete reaction due to coating of the barium hydrOXide particles
with barium sulfate. Place the flask in a boiling-water bath and add 10 g. (0.1 equivaLent) of powdered
barium carbonate slowly to minimize foaming 1 while stirring and heating the mixture. Continue heating
- 62-
and stirring for 10 min. Test a drop of the mixture on a spot plate wHh congo red indicator solution. If
the solution is add (blue color). add 10 g. of barium carbonate and repeat the described process. Continue to add barium carbonate in this manner until the acid is neutralized (pH 4 to 5), as shown by the
coior (red) of the indicator.
Allow the precipitate to settle and decant the hot supernatant liquid to a Buchner funnel fitted with
a filter paper covered with a 1/2-in. layer of filter aid. Wash the precipitate thoroughly with two loo-ml.
portions 0: boiling distilled water. Precipitate the excess barium ions ftorn the combined filttate and
washings as described in Exp. 30 (the final pH may be somewhat less than 4). Add 30 g. of decolorizing
carbon to the suspension, heat the mixture in a water hath at 80 0 for 30 min., with occasional stirring,
and filter the sus?ension. Refilter the filtrate if necessary to remove traces of carbon.
Transfer the filtrate to a 2-liter round-bottomed flask, add about 1 g. of antifoam agent (turkey red
oU), and distill the solution to a volume of about 130 mI.. using the apparatus shown in Fig. 6 (see
Distillation in the Appendix) and a water bath. Add 400 ml. of 95 per cent ethanol (heated nearly to
boiling on a water bath) and mix the liquids thoroughly. Allow the mixture to stand until the gummy
material has settled and a clear supernatant liquid forms. Decant the latter and add 200 ml. of methanol
(heated to 60 0 on a water bath) to the residual material. Allow the gum to settle and decant the clear
supernatant liquid. Extract the gum with another 200-ml. portion of hot methanol. Distill the combined
ethanol and methanol extracts to a sirup (volume about 40 mi.), add 2 mI. of distilled water, and stir the
mixture thoroughly. Add several seed crystals of Larabinose and allow the mixture to stand in the refrigerator overnight en [Link] next laboratory period.
Filter the suspension, wash the L-arabinose crystals with methanoi, and apply suction unW the
crystals are nearly dry. 2 Weigh the crude product.
Dissolve the crude product in [Link] water (0.3 rol. per g.), [Link] 95 per cent ethanol (2.5 rol. per
g.) heated nearly to boiling on a water bath, and add decolorizing carbon (0.05 g. per g.). Stir the mixture thoroughly, filter the hot suspension, and wash the carbon on the funnel with hot 95 per cent ethanol.
Allow the filtrate to cool spontaneously to room temperature without shakin-e; or seedin g. Decant the
upper layer, add hot methanol to the sirupy lower layer, stir the mixture thoroughly, and allow it to stand
until the suspended material settles. Decant the supernatant liquid, reextract the lower layer with hol
metilanol, combine the methanol extracts, and distill the solution in vacuo to about 6 ml., using the
reduced-pressure apparatus shown in Fig. 7 (see Distillation in the Appendix). Seed the solution with
a few crystals of L-arabinose and place the mixture in the refrigeratot overnig.~t or until the next
h3boratoc}' period.
Filter the slJspension and wash tile crystals tv,'ic-e witb .lfcethanol. Allow the crystals to dry in air
in a covered evaporating dish. Submit the product to the instructor in a tared, labeled sample h()tt~e.
Calculate the percentage yields of the crude and the purified L-arahinose 3 ani submit the calculations
to the instructor.
NOTES
1. A few drops of butyl alcohol may aid in dispersing the foam.
2. An additional crop of Cl'ystals may be obtained by evaporating the filtrate and treating .i.t as described.
0
3. !J..L-Arahinose melts at 155 to 157 (Anderson artd Sands, 1941) and (a]M = +190.6 deg. in 4 per cent
aqueous solution (Bates and associates, t942). The mutarotation [Link] is +104.5 deg."(Bates and associates,
1942).
- 63-
EXPERIMENT 32
[SOLATION OF D-MANNOSE FROM IVORY-NUT WASTE
The most convenient source of D-mannose i~ vegetable-ivory (endosperm of the s_eed of the tagua
palm, Phytelephas macrocarpa) turnings from butt<>n factories, although it occurs in the fruit of the snowberry, Sympboricarpus racemosus (von Lippmann, 1921), the seedsof Daubentonia drummondii (shrub of
the pea family which grows in the .coastal plain frPm Florida to Texas) (Curl and Nelson, 1944), and
other plant an~ animal materials Q;lailey and Roe, 1944; Fischer and Hirschberger, 1888,' 1889a, 1889b).
D-Mannose has been isolated by crystallization from the sulfuric acid hydrolysate of ivory-nut
waste (Bates and associates, 1942; Clark, 1922; Horton, 1921; Hudson and Sawyer, 1917; Isbell, 194i;
Levene, 1923, 1924, 1935) and through the intermediate phenylhydrazone (Fischer and Hirschberger,
1888, 1889a, 1889b) and methylmannoside (Hudsor1 and Jackson, 1934; Sheehan and Freudenberg, 1942).
It has been reported thai a-D-mannose, which first crystallizes, is transformed on recrystallization to
B-D-mannose. Wise, Ratliff, and Browning (1948) have determined man nose in hardwoods.
Mannose is absorbed slowly from the intestillal tract of the rabbit (Bailey and Roe, 1944), and it
is le&o.i\y iennenteo. by yeasts \Bai\ey and Roe, 1~1,) aim '.ac,ic ",cit. 1;:rdct~lia \CCt'ITii~:T1, Dolni., <IU
Salle, 1947). The history of mannose has been given by Harding (1923b).
D-Mannose is to be isolated by crystallizing it from a sulfuric acid hydrolysate of ivory-nut meal.
EXPERIMENTAL
Transfer 100 g. of ivory-nut meal (ivory-nut waste sieved through a 2O-mesh sieve) to a SOO-ml.
flask and ad~. (1.44 equivalents) of 24 N sulfuric acid. Stir the mixture thoroughly and allo,," it
to stand at room te;P~mture overnight or until the next laboratory period. :Rinse the mass from the
flask to a 2-liter beaker, add distilled water to a total volume of solution of 1 liter, and stir the mixture
thoroughly. Filter the mixture on a Buchner funnel fitted with-several thicknesses of cheesecloth,
allow the liquid to drain, add 50 mI. of distilled water, and allow the liquid'to drain as completely as
possible while pressing the mass wi~ an inverted glass stopper.
,
Transfer the suspension to a 2-htet beaker 1;lnd add 25 g. (0.25 equivalent) of powdered barium
carbonate slowly to minimize foaming,l while stirring and heating the mixture. Continue heating and
stirring for 15 min. Test a drop of the mixture on a spot plate with congo red indicator solution. If the
mixture is acid (blue color), add 10 g. of powdered. barium carbonate and repeat the described procedure.
Continue this process until the solution is neutral (red color) to congo red (pH 4 to S)JJJ
Filter the hot suspension on a Buchner funnel fitted with a filter paper and 1/2~. of filter aid.
Wash the precipitate thorQughly with two 100~ml. I?0rtions of boiling distilled water. Add N sulfuric
acid to the combined filt,&ate and washings as described in Exp. 31 until all the barium ions have been
converted to'insoluble barium sulfate. Add 30 g. of decolorizing carbon and stir the mixture thoroughly
at intervals for IS min. Filter the suspension, and if particles of carbon are observed, refilter the
fil1!ate, using a fluted gravity filter. Transfer the filtrate to a 2-liter round-bottomed flask and distill
it to a thick sirup (20 to 25 mI.) under reduced pressure, using the apparatus shown in Fig. 7 (see
Distillation in the Appendix) and a water bath. 2
Add 100 mI. of methanol (heated to 60j whHe continuously stirring the mixture. Add 200 ml. of
isopropyl alcohol while stirring and allow the mixture to stand until the gummy impurities settle. Decant the supernatant liquid, add 30 mI. o(hot methanol to the gummy residue with stirring, and add 60
ml. of isopropyl alcohol with stirring. Allow the g;um to settle, decant the alcoholic supernatant
liquid, and reextract the gummy residue with 30 ml. of hot methanol. Distill the combined splutions
_ 64-
,under reduced pressure on a water bath to a thin sirup (25 to 30 q11.). Add 25 ml. of glacial acetic acid,
stir the mixture, add several crystals of D-mannose, and allow the mixture to stand overnight or for
longer time at room temperature until crystals are observed. Place the mixture in a refrigerator for 24
hr. and filt~r the suspension (on longer standing the crystals may become gummy). Wash the crystals
with a solution containing 1 part of methanol and 3 parts of 95 per cent ethanol. Apply suction until
excess solvent has been removed (10 min.). Weigh the product.
Recrystallize the crude mannose using the indicated quanti ties of materials per 5 g. of mannose.
Dissolve the product'in water (5 mI.), add glacial acetic acid (1/2 drop) and decolorizing carbon (0.25
g.), and allow the mixture to stand for 15 min. in a boiling-water bath with frequent stirring. Add
methanol (10 mI.), stir the mixture, and add isopropyl alcohol (20 mI.). Allow the gummy impurities to
settle and decant the supernatant liquid to a Buchner funnel. Add a few crystals of D-mannose to the
filtrate and stir the mixture vigorously at intervals until crystals are observed. Place the mixture in
the refrigerator for 1 day. Filter the suspension and wash the crystals with' 95'per cent ethanol until
the washings are neutral to litmus paper. 3 Wash the crystals with 10 mI. of absolute ethanol and allow
the crystals to dry in air in a covered evaporating dish. Submit the product to the instructor in a tared,
labeled sample bottle. Calculate the percentage yields of the crude and the purified mannose 4 and submit the calculations to the instructor.
NOTES
1. A few drops of butanol may aid in dispersing the foam.
2. If the filtrate is dark-colored when the volume has been reduced to 200 mI., decolorize it with 5 g. of
decolorizing carbon.
3. An additional crop of crystals may be obtained by evaporating the filtrate and treating it as described.
4. CX-D-Mannose melts at 131 to 1320 (Sheehan and Freudenberg, 1942). [o:]~ = +29.3 deg. in 4 per cent
aqueous solution, and the equilibrium value after mutarotation is +14.2 deg. (Bates and associates, 1942).
EXPERIMENT 33
[SOLATION OF [Link] MONQHYDRATE FROM WHEY POWDER
The history of lactose has been reviewed by Whittier (1925-1926, 1944). Lactose was manufactured from whey as early as the seventeenth century because of its supposed value as a remedy for
gout, distemper, and other disorders. It occurs in the milk of all mammals (except, possibly, the whale),
although the percentages range from about 1.8 per cent for the rabbit to 8 per cent for the human. Cow's
milk contains about 4.5 per cent lactose.
The large quantities of lactose manufactured at the present time in connection with cheese production find use in medicinal tablets, infant food, and bacteriological media. Lactose is usually prepared
by crystallization from extracts of whey powder (Leviton! 1949; Leviton and Leighton, 1938; Pederson,
1913; Webb and Ramsdell, 1944).
It has been reported that,excess lactose in the diet changes the intestinal flora from the putrefactive to the fermentative type CRettger and Cheplin, 1923; Kulp and Rettger, 1924). On the other hand it
has been found that excess lactose causes rats to develop alopecia (loss of hair), fail to grow, and'die
in 3 to 17 days (Ershoff, 1946; Handler, 1947a; Riggs and Beaty, 1947). Lactose is employed to differentiate the lactose-fermenting colon group of organisms from the nonfermenting typhoid-dysentaty bacteria.
a-D-Lactose monohydrate is to be isolated by crystallizing it from an aqueous-ethanol extract of
whey powder essentially by the procedure of Leviton and Leighton (1938).
EXPERIMENTAL
Obtain 25 g. of finely powdered, spray-dried, "sweet" whey powder 1 and assignment to a Waring
Blend or. Transfer the powder and 300 ml. of 95 pet cent ethanol to the bowl of the Blendor. Add 155
ml. of hot (90 to 95) distilled water and stktlfe mixture for 10 (not more) min. Add 5 g. of filter aid
and stir for 10 sec. Immediately filte'r the suspension on a dty Buchner funnel as rapidly as possible,
since the lactose may start to crystallize. Transfer the filtrate to a 2-liter round-bottomed flask, add
5 ml. of concentrated hydrochloric acid and 150 inl. of 95 per cent ethanol, and mix the solutions
thoroughly. The reduction in pH (from about 6 to 4) increases the solubility 01 the proteins, while the
increased concentration of alcohol decreases the solubility 'of lactose. Add 'Several crystals of
a-D-Iactose monohydrate and place the mixture in the refrigerator until ctystallization appears to be
complete (about 4 days). Shake the flask at intervals to accelerate crystallization. Filter the suspension on a Buchner funnel, wash the crystals with 25 ml. of 95 per cent ethanol, apply suction until the
product is nearly dry (10 min.), and weigh the crude material.
Suspend the crude lact?se monohydrate in distilled water (1.0 ml. per g.) at about 90 0 and stir the
mixture until most of the solid dissolves (5 min.): If necessary add additional distilled water (1 ml.
portions) and heat as directed. If the mixture is colored, add decolorizing carbon (0.05 g. per g.) and
heat the mixture for 10 min. in a boiling-water bath with occasional stirring. Filter the hot suspension
on a Buchner funnel. If traces of carbon are observed, filter the filtrate on a fluted filter. Add several crystals of a-D-lactose monohydrate, stir the mixture, and place it in the refrigerator overnight or
until the next laboratory period. Filter the suspension on a Buchner funnel and wash the crystals twice
with 20-ml. portions of 9S per cent ethanol. Allow the product to dry in air in Ii!- covered evaporating dish.
Suqmit the product to the instructor in a tared, labeled sample bottle. Calculate the percentage yields
of the crude and the purified a-D-Iactose monohydrate2 and submit the calculations to the instructor.
- 66-
NOTES
1. "Sweet" whey is the product obtained after rennet coagulation of the casein in skim milk during the
manufacture of cheese. It contains about 6d to 70 per cent lactose, 10 to 15 per cent protein (lactalbumin and
lactoglobulin), and 5 to 10 per cent minerals.
2. a-D-Lactose monohydrate gives [o:]~ = +85.0 deg. in 7.6 per cent aqueous solution, and the equilibrium
value after mutarotation is +52.6 deg. (Bates and associates, 1942).
EXPERIMENT 34
POLARIMETRIC ANALYSIS OF CX-D-LACTOSE MONOHYDRATE
The optical rotation of the CX-D-Iactose monohydrate prepared in Exp. 33 is to be determined, and
the purity of the sample is to be calculated on the basis of its specific rotation. The procedure to be
followed is essentially the same as that described in Exp. 17.
EXPERIMENTAL
Obtain from the instructor three 2.5-g. samples of a-D-Iactose monohydrate. Weigh each 2.5-g.
sample quantitati vely to 1 mg. and transfer each sample quantitatively to a clean, dry 50-mI. volumetric
flask.
Turn on the sodium lamp and allow it to warm up for about 10 min. Observe the zero reading of the
scale with the polarimeter tube filled with distilled water, as described in Exp. 17. The following
manipulations must be performed rapidly if the analytical results are to be satisfactory.
Simultaneously add about 40 mI. of distilled water (at room temperature) to one of the flasks and
start the time clock. Immediately stopper the flask and shake it vigorously. Record the time (less than
2 min.) required to dissolve most of the lactose as indicated by a pronounced decrease in turbidity of
the suspension. Continue to shake the flask until all but traces of the lactose dissolve. Add distilled
water rapidly to the mark, stopper the flas"k, and mix the contents of the flask thoroughly. Rinse the
polarimeter tube with two small portions of the lactose solution and fill the tube with this solution.
Determine the rotation of the solution and record the time of this observation. Determine the rotation
at I-min. intervals, recording the time of each observation, until 10 readings have been made. The
time required for the 10 readings should be approximately the same'as that which elapsed prior to the
first reading. Record the final temperature of "the solution. Repeat this procedure using the second
and third samples.
"
Correct each rotation value for the zero reading and determine the logarithm of each corrected
rotation. Plot the logarithmic values -as ordinates and time In minutes as abscissas on coordinate
paper. Draw the best straight line through (of as close as possible to) the points and extrapolate this
line to the time at which most of the t~etose dissolved. Estimate the observed rotation at this time
and calculate the specific rotation [a]~ for each sample. Calculate the percentage purity of each
sample from the specific rotation determined experimentally and that of pure a-D-Iactose monohydrate
(+85.0 deg. at 20 C.). The temperature coefficient is small and may be neglected.
- 68-
EXPERIMENT 35
DETERMINATION OF FREE AND COMBINED GLUCOSE IN CORN SIRUP
Corn (maize) was first cultivated extensively by the natives of the Americas, and'today abbut 70
per cent of the world's corn is produced in the United States. Com 'contains (dry basis) 67 per cent
starch, 13 per cent fiber, 8 per cent protein, 7 per cent' soluble materials (protein and minerals), 3.5 per
cent oil, and 1.5 per cent ash and miscellaneous substances.
I
Com sirup is prepared by digesting starch with dilute hydrochloric acid, neutralizing the excess
acid with sodium carbonate, ad'sorbing the colored and other impurities on carbon, and evaporating the
eluate to the desired consistency. Approximately 2 billion pounds o( corn sirup are consumed annually
in the United States as table sirup and in making candy and bakery products. Corn sirups contain, 22 to
43 per cent glucose, 21 to 34 per cent maltose, 0 to 7.5 per cent levulose, and 6 to 20 per cent higher
sugars (Bishop, 1944).
Free and combined glucose in com sirup are to be determined by Somogyi's (1945) method. Glucose (free or liberated by hydrolysis of maltose, dextrin, and other combined forms) is oxidized with
cupric ion to gluconic acid, and the resulting cuprous ion is converted to a colored complex with
Nelson's _(1944) chromogenic (arsenomolybdic acid) reagent. The intensity of the color is determined
by comparing it with that formed by a standard glucose solution.
EXPERIMENTAL
Obtain a sample of commercial com sirup. Transfer about 0.5 g. of the sirup to a sOO-ml. volumetric flask by the technique described in Exp. 24. Add about 300-ml. of distilled water, thoroughly
mix the liquids, fill the flask to the mark with gistilled water, and thoroughly mix the liquids. Pipette
50.0 mI. of this solution (A) into a 250-ml. volumetric flask. Add 100-ml. of di~til1ed water, mix the
liquids thoroughly, fill the flask to the mark with distilled water, and mix the liquids thoroughly
(solution B). Pipette 25.0 ml. of solution A into a 250-ml. volumetric flask and, submit the flask to the
instructor, who will add a known quantity of pure glucose. Add 200 ml. of distilled water to the flask,
mix the liquids thoroughly, fill the flask to the mark with distilled water, and mix the liquids thoroughly
(s,olution C).
Pipette 25.0 ml. of solution A and 25.0 mI. of 1.0 N hydrochloric acid into a 12S-ml. flask. Attach
a small condenser (Hopkins or 100in. Liebi'g) to the flask, immerse the flask in a boiling-water bath, and
maintain boiling for 2.5 hr. Remove the flask and detach and rinse the condenser into the flask. Add
25.0 mI. of 1.0 N sodium hydroxide 1 and transfer the mixture to a 250-ml. vot'umetric flask: Add 150 mI.
of distilled water, mix the liquids thoroughly, fill the flask to the mark with distilled water, and mix the
liquids thoroughly (solution D).
Determine the apparent free glucose in the unhydrolyzed com sirup (solutions B and C) by the
following procedure: Transfer (lO-ml. graduate) 2.0 mI. of the cupric-ion reagent (Na2HP0 4 , NaOH,
Rochelle salt, cupric sulfate, and sodium sulfate) to each of six 6-in. test tubes.2 Pipette into each
tube 2.00 ml. of one of the following solutions: (a) unknown sugar solution, (b) unkown sugar solution
containing added pure glucose, (c) standard solution containing 0.035 mg. per ml. of glucose, (d)
standard solution containing 0.060 mg. per mI. of glucose, (e) standard solution containing 0.10 mg. per
ml. of glucose, and (f) standard solution containing 0.15 mg. per ml. of glucose. Cover the mouth of
each tube with a marble. Immerse the tubes in a boiling-water bath for 10 min., remove the tubes, and~
cool them under running water. Transfer (IO-ml. graduate) 1.0 ml. of the chromogenic reagent (ammonium
molybdate, sulfuric acid, and disodium arsenate) to each of the tubes. Transfer quantitatively the
- 69-
colored solution in one of the tubes to a 2S-ml. volumetric flask, rinse the tube into the volumetric fl'ask,
add distilled water to the mark, and mix the liquids thoroughly. Repeat this procedure with each of the
other five s~lutions. Compare (visual colorimeter)<the intensity of the color of the unknown solution
with that of the standard which matches it most closely.
Determine the glucose in the hydrolyzed corn sirup (solutions D and E) by the described procedure,
'
using freshly prepared standard solutions.
NOTES
1. Do not add the alkali until ready to carry out the analysis, since a neutral solution of glucose may be
decomposed by bacteria or molds within a few hours.
2. If a photoelectric colorimeter is to be used, add 2 mI. of the cupric-ion reagent to an additional test tube
containing 2.0 mI. of distilled water.
'EXPERIMENT 36
-71-
6.
7.
8.
9.
10.
EXPERIMENTAL
J
Obtain from the instructor an unknown sample of a dry homogenous mixture containing two or more
carbohydrates. Determine which carbohydrates are present on the basis of the group tests (Steps I to
IX), prepare a flow sheet indicating the procedures to be followed in separating and identifying the carbohydrates in the unknown, and submit the flow sheet for the approval of the instructor. When the confirmatory tests have been completed, submit a written report summarizing the experimental results and
conclusions. If any of the results are incooclusive, repeat the test on a sample of pure carbohydrate
obtained from the instructor.
Step I. Detection of Carbohydrates. 1 Thymol Test. Transfer 5 mg. 2 of the unknown to a 3-in. test
tube, add 2 ml. of distilled water, add 3 drops of 5 per cent thymol in 95 per cent ethanol, and stir the
mixture. Incline the tube and introduce concentrated sulfuric acid in a manner such that the acid forms
a layer below the aqueous solution. If a carbohydrate is present a rose-colored ring will appear at the
interfate of the liquids in a few seconds. The color will deepen on standing and will spread throughout
the solution when the liquids are mixed.
Step II. Detection of Glucopolysaccharides. Dextrin, Glycogen, and Starch. Iodine Test. 3 Transfer 30 mg. of the unknown to a 3-in. test tube, add 2 mI. of water, add 2 drops of 0.01 N iodine in 0.01 N
potassium iodide solution, stopper the tube, and shake it vigorously for 10 sec. The appearance of a
color ranging from deep blue to magenta or reddish brown indicates the presence of at least one glueDpolysaccharide. Other saccharides give no reaction.
Step III. Detection of Monosaccharides and Oligosaccharides. Monosaccharides: Arabinose,
Fructose, Galactose, Glucose, Mannose, Rhamnose, and Xylose. Oliogosaccharides: Lactose, Maltose,
Raffinose, and Sucrose. Nitrochromic Acid Test. 4 Transfer 30 mg. of the unknown to a 6-in. test tube,
add 2 ml. of distilled water, and stir the mixture. Add 4 ml. of concentrated nitric aCid, add 5 drops of
5 per cent potassium chromate solution, and stir the mixture. The appearance of a blue color within a
minute or two indicates the presence of at least one low molecular-weight carbohydrate.
Step IV. Detection of Ketosaccharides_., Fructose, Inulin, Raffinose, and Sucrose. Phosphoric
Acid Test. s Transfer 20 mg. of the ui\known to a dry 3-in. test tube, add 1 ml. of 85 per cent phosphoric
acid, and stir the mixture. Heat the tube in a boiling-water bath for 1 min., remove the tube immediately,
and observe the color. A brown or brownishwblack solution indicates the presence of fructose or of a
saccharide containing fructose. Other sugars give little or no color under these conditions.
Step V. Detection of Reducing Saccharides. Arabinose, Fructose, Galactose, Glucose, Lactose,
Maltose, Mannose, Rhamnose, and Xylose. Basic Cupric Citrate Test. 6 Transfer 50 mg. of the unknown
to a 6-in. test tube, add 5 ml. of Benedict's reagent (cupric sulfate, sodium citrate; and sodium carbonate),
and stir the mixture. Heat the tube for 3 min. in a boiling-water bath. Allow the tube to cool to room
temperature. The formation of a red-rust precipitate indicates the presence of at least one reducing
sugar in the unknown. Traces of nonsugar impurities may form a small amount of cuprous oxide and
color the solution green.
Step VI. Detection of Reducing Disaccharides. Lactose and Maltose. Methylamine Test. 7 Transfer 30 mg. of the unknown to a 6-in. pyrex test tube, add 4 mI. of distilled water, and stir the mixture.
Add 3 drops of 5 per cent aqueous methylamine hydrochloride solution, heat the mixture to boiling over
a free flame, and continue b9i1ing for 30 sec. Remove the tube and add immediately 3 drops of 6 N
sodium hydroxide. The appearance of a yellow color which changes on standing to a carmine red indicates the presence of at least one reducing disaccharide. Fructose gives a brown color and other sugars
a yellow color.
-72-
- 73-
been reduced to 10 ml. Add 0.5 ml. of carbon disultide to this solution (,5.5) and preserve the solution
for the tests in Step XI. Pass air over the residue (P-5) on the filter and preserve it for use in Step X-6.
4. Lactose. Precipitation by 85 Per Cent Ethanol. (a) Poly~accharides absent. Dissolve the
unknown (P-1, P-2, or 0-3) in 10 ml. of distilled water. Add 0.5 mi. of carbon disulfide to 7 mi. of this
solution (S-5) for use in Step XI. Add 30 ml. of 95 per cent ethanol to the remaining 3 mi. of S-5, shake
the mixture thoroughly, and allow the solution to stand at room temperature for 30 min. or longer.
When precipitation of lactose (if present) appears to be complete, decant the supernatant liquid,
evaporate it to dcyness on a boiling-water bath, and preserve the residue (P-6) for later use. Wash the
lactose precipitate twice with 5-mI. portions of 95 per cent ethanol and discard the washings. Dry the
lactose by passing air through the flask and collect ~he lactose (P-7) by scraping the sides of the flask
with a rubber policeman. Preserve the lactose (p-7) for use in Step XI, Tests 8, 10, and 11.
(b) Polysaccharides present. Add 30 ml. of 95 per cent ethanQl to the solution to 3 ml. S-4 prepared in Step X-3, 'shake the mixture vigorously until the precipitate (if any forms) coagulates, filter
the suspension on a dcy gravity filter, collect the filtrate in a dry SO-ml. flask, and allow the flask to
stand for 30 min. or longer; If lactose is present, treat it as described in part (a).
5. Inulin. Extraction with Water. Transfer the solid P-4 from Step X-3, consisting principally of
starch and inulin, to a 6-in. test ,tube, add 10 ml. of boiling distilled water, stir the mixture thoroughly,
and place the tube in a hot-water bath. Allow the solid to 'settle and filter the supernatant liquid with
suction. Transfer the residual material to the filter and apply suction until the solid is nearly dry. Preserve the solid (P-8) for use in Step XI, Test 1. Cool the filtrate in an ice-water bath until a precipitate 14 forms (10 to 30 min.), filter the suspension with suction, preserve the filtrate S-6 for use in Step
XI, Test 2, wash the precipitate (P-9) with 3 ml. of distilled water and with 5 mI. of 95 per cent ethanol,
apply suction until the precipitate is nearly dcy, and preserve the product for use in Step XI, Test 2.
6. Glycogen. Precipitation by Saturated Ammonium Sulfate Solution. Transfer the precipitate p-s
prepared in Step X-3 to a 50-mi. conical flask. Add 10 mi. of distilled water and stir the mixture until
the solid dissolves. If any undissolved saccharide remains, filter the suspension and reject the precipitate. Add 7.5 g. of powdered ammonium sulfate and shake the mixture vigorously at intervals for 15
min. Filter the mixture, preserve the filtrate S-7 for use in Step XI, Test 1, wash the precipitate (P-I0)
with 5 ml. of 95 per cent ethanol, allow the solid to dry in air, and preserve it for use in Step XI, Test 1.
Starch and glycogen, but not dextrin, are precipitated by saturating the solution with ammonium sulfate.
Starch, but not dextrin or glycogen, is precipitated by half saturating the solution with ammonium sulfate.
Step XI. Detection of Individual Carbohydrates. 1s 1. Starch, Dextrin and Glycogen. Iodine
Test. Transfer to a 3-in. test tube 10 mg. of the precipitate P-4 prepared in Step X-3 or p-8 prepared
in Step X-5. Transfer 20 mg. of the precipitate p-l0 prepared in Step X-6 to another 3-in. test tube.
Add 2 mI. of distilled water to each tube and stir the mixtures. Transfer 2 ml. of each of the solutions
S-3, from Step X-3, and S-7, from Step\)C-o, to separate 3-in. test tubes. Add 2 ml. of distilled water to
a fifth 3-in. test tube. -Add 2 drops of 0.01 N iodine (in 0.01 N potassium iodide solution) to each tube
and note the colors formed in the five tubes. Add 0.5 g. of sodium chloride to the tube containing the
precipitate P-l0 and to the tube containing the solution S-7. Stir the mixtures until the salt dissolves.
Note any increase in the intenSity of the iodine color. Any opalescence of either solution prior to the
addition of iodine is caused by glycogen. Neither dextrin nor glycogen interferes with the formation of
starch iodine blue color. In the absence of other polysaccharides, dextrjn gives a blue-violet or violetred color with iodine, while glycogen gives a brown-red color, which is intensified by sodium chloride.
2. Inulin, Raffinose, Sucrose, and Fructose. Resorcinol Test. 16 Transfer to separate 6-in. test
tubes 0.5 ml. of the solution S_117 prepared in Step X-I, 5 mg. of the solid P-9 or 0.5 ml. of the solution
S-6 prepared in Step X-S, and 0.2 ml. of the sohiion s-5 prepared. in Step X-3 (or X-4 or Step XI, Note 15).
Add to each tube 1 ml. of distilled water and 5 mI. of an ethanol-sulfuric acid mixture (40 mI. of acid in
ISO ml. of 95 per cent ethanol) and stir the mixtures. Add to each tube 5 mI. of a 0.2 per cent solution
of resorcinol in 95 per cent ethanol and heat the tubes for 2 min. in a boiling-water bath. The appearance
of an intense ,red color confirms the presence of fructose or fructose-containing saccharide.
3. Raffinose and Sucrose. Basic Cupric Citrate Test. Omit this test if raffinose and sucrose were
shown to be absent by the resorcinol test (Test 2). If reducing sugars are pre~ent, it is advantageous to
destroy them with Benedict's reagent before testing for raffinose and sucrose.
Transfer to separate 6-in. test tubes 1.0 mI. of the solution 5-1 prepared in Step X-I and 0.5 mI.
of the solution S-5 prepared in Step X-3. Treat each solution as follows: Add 5 mI. of Benedict's
reagent, stir the mixture, and heat the tube in a boiling-water oath for 5 min. Filter the suspension of
cuprous oxide, and if it appears that the cupric ion in the solution has been almost depleted, repeat
this procedure. Remove the excess cuptic ions by saturating the solution with hydrogen sulfide (fume
hood) and filtering the suspension of cupric sulfide. Add 6 N sulfuric acid dropwise to the filtrate until
the solution is acid to litmus paper and heat the mixture until the volume has been reduced to 4 ml. Test
0.5 mI. of this solution for raffinose and sucrose by the resorcinol test (Test 2) and preserve the remaining 3.5 mI. of solution for the diazouracil test (Test 4).
4. Raffinose and Sucrose. Diazouracil Test. ls Omit this test if raffinose and sucrose were shown
to be absent by the resorcinol test (Test 2).
Transfer to separate 4-in. test tubes 1.0 mI. of the solution S-1 prepared in Step X-I and 0.5 mI. of
the solution S-5 prepared in Step X-3. Dilute each solution with distilled water to a final volume of 3.5
ml. Transfer the 3.5 mI. of solution prepared in Test 3 and 3.5 mI. of distilled water to separate 4-in.
test tubes. Add 1 mI. of 2 N sodium hydroxide to each tube, stir each mixture, and cool the tubes to 50
in an ice-water bath. Add 7 mg. of diazouracil (obtain from the instructor) to each tube and shake each
mixture until a clear solution is formed. Allow each solution to stand at 5 to 10 0 for 20 min., add 2
drops of M magnesium chloride solution, and shake each mixture. The appearance of a green to bluegreen to brown-green color within 20 min., followed by intensification of the color adsorbed on the
precipitated magnesium hydroxide confirms the presence of raffinose, sucrose, or both these sugars.
Sucrose is more sensitive to the test than raffinose and gives a more intense color. Other saccharides
form a yellow to brown-red color which interferes markedly with the test.
5. Rhamnose. Periodic Acid-p-Hydroxydiphenyl Test. Transfer 1.0 ml. of the solution S-1 prepared in Step X-I to an 8-in. test tube and carry out the procedure described in Step IX. The appearance
of a violet color in the sulfuric acid confirms the presence of rhamnose.
6. Fructose. Cobaltous Chloride Test. 19 Transfer 0.5 mI. of the solution S-1 prepared in Step X-I
to a 3-in. test tube, add 2 mI. of a 1 per cent aqueous solution of cobaltous chloride, and stir the mixture.
Heat the tube in a boiling-water bath for 2 min., cool the tube to room temperature, add 4 drops of concentrated ammonium hydroxide, and allow the mixture to stand for 20 min. The appearance of a violet
color confirms the presence of fructose. Green cobaltous hydroxide precipitates in the absence of
fructose. In the presence of another carbohydrate, fructose may form a violet-colored supernatant
liquid discernible after the cobaltous hydrOXide has settled.
7. Arabinose and Xylose. Benzidine Test. 2o Transfer 1 drop of the solut'ion prepared in Step X-2
to a 3-in. test tube to which 0.5 mI. of a 4 per cent solution of benzidine in glacial acetic acid has been
added. Heat the mixture to vigorous boiling over a flame and immerse the tube in an ice-water bath. The
appearance of a red color confirms the presence of a pentose. Hexoses give yellow to brown colors
which do not interfere with the test unless present in relatively high concentrations.
8. Lactose and Maltose. Acidic Cupric Acetate Test. Omit this test if reducing disaccharides
(Step VI) and monosaccharides (Step VII) were shown to be absent.
Transfer to separate 6-in. test tubes 1 mI. of the solution S-5 prepared in Step X-4 and (if lactose
was isolated) one-third of the solid p-6 prepared in Step X-4. Add 5 ml. of Barfoed's reagent to each
tube, heat the tube in a boiling-water bath for 5 min., and cool the tube to room temperature. If little or
no brick-red precipitate forms, the absence of reducing monosaccharides is confirmed. If an abundant
precipitate of cuprous oxide forms, filter the suspension. If the copper reagent has been used up, as
shown by the color of the filtrate, repeat the described procedure until reducing monosaccharides have
been destroyed. Saturate the filtrate with hydrogen sulfide, filter the suspension of cupric sulfide, and
add 6 N sulfuric acid dropwise to the filtrate until it is just acid to litmus paper. Heat the solution for
15 min. in a boiling-water bath to remove excess hydrogen sulfide, evaporate the solution to 1 mI., and
preserve it for use in Test.9.
9. Lactose. Fermentation Test. 21 Transfer 1 g. of active dehydrated yeast to a mortar, add 10
mI. of distilled water, and grind the mixture to a uniform suspension. Transfer the suspension to a
centrifuge tube, centrifuge the mixture, decant the supernatant liquid, resuspend the residual material
in 10 mI. of distilled water, centrifuge the mixture, decant the supernatant liquid, and suspend the
residual yeast in 30 mI. of distilled water.22
- 75-
Add to separate Einhorn saccharometers (fermentation tubes) 20 mg. of lactose dissolved in 1 ml.
of distilled water, 20 mg. of maltose dissolved in 1 mI. of water, (}.5 mI. o the solution 5-5 prepared
in Step X-4, and 1 mI. of the solution prepared in Test 8. Add 10 mI. of the prepared yeast suspension
to each tube, stir the mixture in each tube to a uniform suspension, and fill the long arm of each saccharometer with the yeast suspension. Incubate the tubes at room temperature .overnight or at 35 for
6 hr. Compare the volumes of carbon dioxide formed. The production of gas in amounts significantly
greater than that from the blank (lactose) and compara~le to the control (maltose) indicates the presence
of at least one yeast-fermentable saccharide (glucose, mannose, sucrose, or maltose).
10. Galactose, Lactose, and Raffinose. Mucid Acid Test. 23 Transfer to separate 50-ml. flasks 3
ml. of the solution 5-1 prepared in Step X-I, 100 mg. of the precipitate p-7 prepared in Step X-4, and
one-third of the solid p-6 prepared in Step X-4. Dilute eac1;t solution to 10 mI. with distilled water~ add
4 mI. of concentrated nitric acid to each flask, stir each mixture, ana heat the flask in a boiling-water
bath for a sufficient time to reduce the volume in each flask to about 5 mI. (about 3 hr.). Add 3 ml. of
distilled water to each flask and allow the mixture to stand overnight (or lonser) at room temperature in
in a stoppered flask. If any crystals form, e~amine them under a microscope and compare them with
photomicrographs of crystals provided by the instructor. Determine the melting point of the water-washec
and dried crystals. The formation of mucic acid (m.pv213 to 214) confirms the presence of [Link]
or galactose-containing compound (raffinose or lactose).
11. Reducing Sugars. Phenylhydrazine Test. 2~ Transfer 0.2 g. of phenylhydrazine hydrochloride and
0.3 g. of powdered sodium-acetate to a test tube, add 2 mI. of distilled water, and stir the mixture until
the solid dissolves .. Only freshly prepared reagent should be used.
Transfer to a 4-in. test tube 0.5 mI. of the solution 5-5 prepared in Step X-3 Jor Step X-4 or Step
XI, Note 15}, add 2 mI. of distilled water, and cool the tube in an ice-water bath. Add 2 mI. 9f (freshly
prepared) phenylhydrazine reagent, stir the mixture, and llllow it to stand in a,n ice-water bath for 1 hr.
with occasional stirring. If a precipitate forms, filter the suspension with suction and. wash the solid
with two 3-ml. portions of distilled water.25 Discafd the washings. The formation of mannose phenylhydrazone (m.p. 188) confirms the presence of mannose.
Add 2 mI. of the phenylhydrazine reagent to the filtrate (or 0.5 ml. of solution '5-5, if no mannose
phenylhydrazone formed), stir the mixture, and heat the tube in a bo~ling-water bath for 30' min. Filte~
the hot suspension on a small suction filter which has been heated by im,mersion in hot distilled water:Wash the precipitate with two 2-ml. portions of hot distilled water and discard the washings. 25
Allow the filtrate to cool to room temperature, and if a precipitate forms, filter the suspension.
Add 5 ml. of acetone to the precipitate on the filter paper, return the filtrate t_o the paper, and continue thi
process until no more precipitate dissolv~s. Evaporate the acetone on a boiling-water bath and recrystallize the residue from hot wat~f?S Recrystallize the precipitate remaining on the filter paper.'25
Add 2 ml. of the phenylhydrazine reagent to 0.5 ml. of the solution 5-1 prepared in Step X-I, heat th.
solution to boiling in a water bath, allow the mixture to cool to room temperature, filter the suspension,
wash the precipitate with distilled water, and discard the washing. 25 This test is confirmatory for
rhamnose or fructose but not both sugars.
Although sepatation of individual osazones from a mixture of osazones is not readily accomplished,
advantage may be taken of the following solubility relations: The osazones o~ the monosaccharides,
(except gal~ctose)are slightly soluble, while those of the disaccharides are readily soluble in hot water.
Maltosazone, but not lactosazone, is soluble in acetone.
12. Arabinose. Benzhydrazide Test. 26 Transfer 2 mI. of the solution 5-5 prepared in Step X-3 (or
[Link] X-4 or Step XI, Note 15) to a 6-in. test tube, add 10 ml. of a solution of [Link] in 57 per
'cent ethanol, stopper the tube, and shake it vigorously. Allow the tube to stand at room temperature
with occasional shaking until the next laboratory period. If no crystals form, allow the t,ube to stand
until the following period. Filter the suspension, wash the crystals with 5 mI. of ice-cold 95 per cent
ethanol, and recrystallize the product from the minimum volume of hot 95 per cent ethanol. The formatior
of arabinose benzhydrazone (m.p. 210 to 212) confirms the presence of arabinose.
13. Xylose. Benzaldehyde-Methanolic Hydrochloric Acid Test. 27 Transfer 3 ml. of the solution
S-5 prepared in Step X-3 (or Step X-4 or Step XI, Note 15) to a 50-mI. conical flask and evaporate the
solution to dryness (thick sirup) on a boiling-water bath. Add 1 ml. of absoiute methanol, stir the mixture.
- 76-
and heat the solution until the solvent evaporates. Place the flask in an oven at 100 and allow it to
stand until moisture has been removed (15 to 20 min.). Add 6 mI. of absolute methanol, 1 ml of 1.4 N
methanolic hydrochloric acid, and 2 mI. of redistilled benzaldehyde. Stir the mixture thoroughly, add a
seed crystal of dibenzylidinexylose dimethyl acetal, stopper the flask, and allow the flask to stand at
room temperature for a week with occasional vigorous shaking. Fi1t~r the su&pension (fine intertwining
needles), wash the precipitate with 5 mI. of water and with 5 mI. of methanol, and dry the product in air.
R_ecrystallize the derivative by dissolving it in the minimum volume of chloroform and adding petroleum
ether (b.p. 30 to 600 ) to a faintly cloudy solution. Add sufficient chloroform to form a clear solution,
allow the solution to stand in the tefrigemtor overnight or longer, filter the suspension, and dry the
!
product in air. The forrration of dibenzylidinexylose dimethyl acetal (m.p. 211 0 ) confirms the presence
of xylose.
NOTES
1. Ihi and Pechmann (1884), Ihl (1885, 1887), and Molisch (1886a, 188B: discovered that sugars (treated
with concentrated sulfuric or hydrochloric acid) react with phenols (types shown below) to give ~hromogenic con-
OH
OH
OH
OH
cotechol
pyrogollol
OH
HO-6-00
cresol
guaiacol
phloroglucinol
OH
HO
CH3"
OH
CR
WaH
'CH3
resorcinol
orcinol
thymol
jJ - naphthol
a-naphthol
densation products. The specific absorption spectra of the hydrolysis and dehydration products formed by the reaction of carbohydrates and sulfuric acid have been determined by Ikawa and Niemann (1949), and the absorption
spectra of the products ,formed by the condensation of acid-treated carbohydrates with phenols have been reported
by Foulger (1931), Militzer (1946), and Pinof! (1905). These products include furfural from arabinose and xylose
(pentoses) (equation below),
r.--'
r.--'
C--:OH'
:HOr-C
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'H__ JJ
HCI
)10
H-C
C-CHO
\1
HI--C-H
~--~
H-C-C-H
I
I
C-CHO
oct=! ~-_-_-_~-_-QBJ
pentose
furfural
S-methylfurfural from rhamnose (methylpentose), and 5-(w-hydroxymethyl) furfural in 20 to 2S per cent yield from
fructose (ketohexose). Glucose, galactose, and mannose (aldohexoses) give only about 1 per cent of 5-(CtJ-hydroxymethyl) furfural, since this substance is converted nearly quantitatively to levulinic acid and formic acid (equations
on following page) (Blanksma, 1909; van Ekenstein and Blanksma, 1910; Kiermayer, 1895; Wheeler and Tollens, 1889.
- 77-
Diill, 1895; Pummerer and Gump, 1923; and Pummerer, Guyot, and Birkofer, 1935).
H-, ~-~-H
Hexose
HO~C-C,
/C-CHO
5-(hydroxymethyl) furfural
I I
H-C-C-H
I C-CHO
I
I I
o 0
HOH2C-C
-HCOOH'5-hydroxylevulinaldehyde
levulinic acid
v. Udr~nkszky (1888) was the first to consider the a-naphthol (Molisch) reaction to be a condensation with
'furfural. Bredereck (1931, 1932) obtained a condensation product from 5-(hydroxymethyl) furfural and a-naphthol
which was readily sulfonated (once at 0 0 and twice at 30 0 ) by concentrated sulfuric acid (equations below).
H-~-~-H
HOCH2- C
C-CHO
~
cx.- naphthol
5 -(hydroxymet hyl)
furfural
OH
H-C-C-H~
I C
I ---C-H
HOCH 2- C
H-C-C-H
H2S04
I C
I ---C
,. HOCH2- \
SO#i
_OH
OH
a-Naphthol has been widely used for the Molisch test, although thymol is more stable and does not become
colored on long standing (Levine, 1929-1930). The test is not specific since aldehydes, ketones, organic acids,
and compounds which yield carbohydrates on hydrolysis (glycoproteins, glycolipids, and nucleic acids) give chromogenic products on reaction with sulfuric acid. For this reason, sugar in urine cannot be detected by the Molisch
reaction. Reagents and apparatus free from impurities must be used, owing to the high sensitivity of t'he Molisch
test (glucose and starch, 0.001 per cent; sucrose, 0.0001 per cent; and raffinose, 0.00005 per cent). Quantitative
procedures utilizing the Molisch reaction (Yamafuji, Yoshida, and Fukuura, 1941; Yamafuji and Yoshida, 1939) and
the thymol test (Bollinger, 1944) for the determination of sugar have been described.
2. Weigh the material in this (or subsequent) test on a horn-pan balance, or estimate the weight by comparison with the volumes of weighed samples in test tubes on the reagent shelf.
3. The formation of a blue-colored substance by the action of iodine was observed in 1814, soon after the
discovery of iodine, according to Barger (1930), who has reviewed the literature prior to 1930. The stable blue
-78 -
color is formed by reaction with amylose, while amylopectin gives a less stable red complex (Bates, french, and
Rundle, 1943). Branching of chains and [Link] affinity are related, since amylopectin from potato starch has less
affinity for iodine than the less highly branched amylopectin from cornstarch (Meyer, Wertheim, and Bernfeld, 1941).
Glycogen, which is even more highly branched than amylopectin from potato starch, forms an unstable iodine complex
(Bates; French, and Rundle, ,1943). MoleculBl," size and iodine affinity are also related, since amylodextrin has a
lower iodine affinity than amylose. Concentrations of iodine as low as 10""'6 N give a d~tectable color with 0.06
per cent starch solution and 0.01 to 0.1 per cent potassium iodide. The sensitivity depends also on the temperature, pH, salts, type of illumination, and nature of the starch (Koreman, 1934; .Woodard, 1934).
Although the nature of the bonding between starch and iodine is not known, it ~as been suggested that a
compound or solid solution is formed. It has been proposed (Rundle, Foster, and Baldwin, 1944; Stein and Rundle,
1948) that iodine molecules lie within 'the amylose matrix of the complex. An81ogous complex formation between
starch and organic substances (such as alcohols, ketones, and,organic acids) (Bear, 1944; Whistler and Hilbert,
1944) and between iodine (and other halogen's) and organic polymers (such as polyvinyl alcohols and polyamides)
has been observed, although these complexes do not absorb in the visible spectrum (West, 1947). Iodine forms
blue-colored complexes with colloidal metallic hydroxides and basic acetates as well as with compounds containing the <x-pyrone or ')I-pyrone ring (Barger, 1930).
'(-pyrone
a-pyrone
According to Turner (1943) iodine is not required to maintain the blue color once it is formed from iodine
and starch.
4. Agulhon's (1911) nitrochromic acid test, as modified by Fearon and Mitchell (1932), is given by all
organic compounds containing the - CHOH group, but the reaction is less rapid with polysaccharides than with
simpler sugars. Tertiary alcohols, aldehydes (other than formaldehyde), ketones, amino acids, and fatty acids
give no reaction. Nitrites, peroxides, and hypohalites also form a blue-colored pigment, which is thought to be
the anhydride of nitrous acid.
S. Only fructose solutions darken in 24 hr. at room temperature. The higher specificity of ketoses and
ketosaccharides in acid solution was reported by Dehn, Jackson, and Ballard (1932): It has been observed that
glucose, maltose, starch, and glycogen do not form a colored product when heated with ~s per c~nt phosphoric
acid for 30 min. (Hestrin and Mager, 1947).
6. In the copper reagent, introduced by Fehling (1849, 1858), cupric ion is maintained in alkaline solution
through complex formation with Rochelle salt (sodium potassium tartrate). The copper solution and the base are
mixed shortly before use, owing to the instability of the reagent. Copper complex-forming substances which have
been employed in place of Rochelle salt include glycerol (Lowe, 1870; Haines, Pond, and Webster, 1920); phosphate
(Folin and McElroy, 1918), citric acid (Rosenthaler, 1904; Benedict, 1908-1909), ammonia (pavy, 1880), potassium
cyanide (Gerrard, 1893), and potassium sulfocyanide (Bang, 1906; Rudisch and CeIler, 1907).
The reactions of carbohydrates in basic solution are exceedingly complex, according to Nef (1908, 1914),
who isolated 93 different products. Ev~ns (1929, 1942) has investigated the types and proportions of the substances formed by enolization, epimerization, polymerization, rearrangement, and oxidative degradation. Fischer
and Hooker (1918 suggested that the yellow and green colors which are observed at low levels of glucose and in
the intermediate stages of the reaction are explained by the colloidal properties of th~ cuprous oxide. By substituting sodium carbonate for potassium hydroxide and citrate for tartrate, Benedict (1907, 1908-1909) prepared a
reagent which was more stable, more sensitive, and more suitable for the detection of glucose in urine (since it
was not reduced by uric acid) than Fehling's solution.
7. Lactose, maltose, other disaccharides containing the 1,4-glycosidic linkage, and low molecular-weight
dextrins respond to this test (Fearon, 1942; Malpress and Morrison, 1949). Methylamine, ethyl amine, and ethanolamines (but not higher homologues) are satisfactory reagents. A red color is formed which turns yellow on acidification but is restored in basic solution. Color formation is inhibited by an excess of a hexose or amine. The
reagent is sensitive to 0.05 per cent lactose.
- 79-
!S. ~he observation in 1815 that cuprous oxide is produced by the oxidation of honey with cupric acetate
solution formed the basis of methods for the detection'of reducing sugars with this reagent (Bates and associa'tes,
1942). Barfoed (1873) found that glucose, but not dextrin, was reduced by cupric acetate in 1 per cent acetic ac)d.
It was noted later- that the ratfl of oxidation of reducing ..sugars decreased as the acidity increased and that monosaccharides, but not dfsaccharides (Hinkel and Sherman, 1907; McGuigan, 1907), are oxidized (fructose> galactose
> glucose). All disaccharides give the test after sufficient reaction time (Bunzel, 1908; Mathews and I\!cGuigan,
1907). von Fellenberg (1947) reported the following activities of the fisted disaccharides compared t9 glucose:
sucrose, 1:75; lactose, 1:45; and maltose, 1:26. Products which have been isolate~ after cupric acetate oxidation
of fructose, glucose, and galactose include glucosone and formic, carbohic, glyoxylic, oxalic, and glycolic acids
(Evans, Nicoll, Strouse, and Waring, 1928).
By replacing the acetic acid o/ith lactic acid, Tauber and Kleiner (1932-1933) found that the cuprous oxide
was proportional to the glucose concentration and that [Link] and disaccharides could be distinguished
on the basis of their rates of reactions. A precipitate is iproduced by 0.1 per cent or higher concentration of a
monosaccharide in the presence of 2 per cent maltose, 4 per cent lactose, or 10 per cent sucrose. Chlorides interfere in tests by the modified as well as the original procedures (Welker, 1915).
9. This procedure, introduced by Bial (1902), has been employed virtually unchanged for the qualitative and
quantitative determination of pentoses. Militzer (1946) observed that the purity of the hydrochloric aciQ and the
concentration of the ferric chloride are critical factors. He recommended as a confirmatory procedure that the absorption spectrum of the pigment in butyl alcohol be examined. Orcinol condenses with furfural (formed by decomposition of the pentose) to give the pigment. Procedures have been proposed for the quantitative determination of
pentoses (Brown, 1946; Drury, 1948; McRary and Slattery, 1945) and of hexoses (in the absence of pentoses)
(Briickner, 1943; S\il'rensen and Haugaard, 1933).
10. Periodic acid preferentially oxidizes compounds with adjacent hydroxy or hydroxy and amino groups (see
Exp. 21, Step IX) as illustrated by the reactions (equations below) for a pentose and a methylpentose.
+ 4HCPOH + 4HIO.
Nicolet and Shinn (1941) determined acetaldehyde in the presence of formaldehyde in developing a procedure for
the quantitative estimation of methylpentoses in the presence of other sugars.
I
11. Proceed with this section only after your flow sheet has been approved by the instructor.
12. The approximate volumes (in milliliters) of absolute methanol required to dissolve 1 g. of each of the
listed carbohydrates (shaken (or 2 min. at room temperature) are rhamnose, 10; fructose, 25; raffinose, 25; xylose,
130; maltose, 160j and arabinose, 320. The other sugars are appreciably less soluble. It should be not'ed \that the
solubility of each of these sugars is increased significantly in the presence of other [Link], that the methanol
solution may contain appreciable quantities of sugars other than the three more soluble ones, and that these overlapping solubilities should cause no difficulties in any of the tests.
13. Test 30 mg. of the solid (P-2) for reducing dlsaccharides by the methylamine test (Step VI) if the latter
was uncertain.
14. If no precipitate forms, preserv;e ),he solution (S-6) to confirm the absence of inulin (Step XI, Test 2).
Wpen inulin is heated in water, glycerol, 'or glycol and is precipitated from these solutions with ethanol, it may be
converted to a cold-water-soluble form (Bates and associates, 1942).
15. If the separations given in Steps X-3 and X-4 have been omitted in your procedure, dissolve the unknown
(P.1, P-2, or P-3) in 10 mI. of distilled water and designate this solution as S5. Add 0.5 mI. of carbon disulfide
as preservative.
16. A red pigment soluble'in ethanol or other alcohols is formed when resorcinol and fructose (or a fructosecontaining saccharide) are heated with hydrochloric acid. The specificity of this reaction was first observed by
Seliwanoff (1887). Rosin (1903) found that the amyl alcohol extract of the neutralized acid miXture was colored
light yellow, exhibited a green fluorescence, and had a characteristic spectrum. All ketohexoses and ketopentoses
tested responded similarly. A red-colored solution is formed directly, using the ethanol-sulfuric acid reagent described by Pinof! (1905). Although other carbohydrates give no color, they diminish the sensitivity of the test.
A colored product which has the same absprption spectrum is formed when glucose and glucose-containing oligosaccharides are heated for a relatively long time. According to Koenigsfeld (1912) this is explained by the formation of fructose from glucose by epimerization.
- 80-
Sen and Sinha (1923) have postulated that resorcinol condenses with
"H-C-C-H
H-CII C-CHO
"
V
HOVOH
furfural
f~ural
(-H20)
resorcinol
H0-fYYJr0H
HO--CX'Y't-0H
C~"OH
(0)
~C~
R/ . . . . .H
"
R/
xanthene stage
fluorene stage
H0-fYYFO
0=CXYJr0H
~C~
"
I
R
quinoid stage
C~
I
17. Heat this solution (and similar ones) in a boiling-water bath until the odor of. carbon disulfide has disappeared.
18. Raybin (1933) discovered that a blue-colored solution is formed by the reaction of diazouracil with an
alkaline solution of raffinose or sucrose.
"N
N=C-O
O=C
C-N/
II
H-N-C-H
diozourocil
A compound which was salted out from neutral solution had indicator properties. The reaction is specific for the
glucose-fructose linkage of sucrose and sucrose-containing oligosaccharides (Raybin, 1937).
19. The specificity of this reagent toward fructose was first observed by Dehn, Ja~kson, and Ballard (1932).
20. The benzidine test for the detection of pentoses was developed by Tauber (1937-1938). The sensitivity
NHz
benzidine
of the test is 0.01 mg. of a pentose. It may be used for the detection of pentoses in urine, since other urinary
constituents do not interfere. Riboflavin, nucleic acids, and other pentose-containing compounds which are hydrolyzed by acetic acid give the test.
21. Monosaccharides (glucose, mannose, and fructose) are fermented aerobically by brewer's yeast to ethanol
and carbon dioxide by the action of enzymes known collectively as zymase.
Disaccharides are not fermented unless first hydrolyzed to monosaccharides by hydrolases. Since brewer's yeast
does not contain lactase, lactose may be determined quantitatively after other sugars have been fermented (Baker and
-81-
Hulton, 1910). Products other than ethanol and carbon dioxide are formed during anaerobic dissimilation of a
monosaccharide, but anaerobic fermentation may be inhibited by sodium azide (Winzler, 1944). Galactose has been
determined by selective fermentation with different strains of yeast (Wise and Appling, 1944).
\
22 Water-soluble carbohydrates which tend to give a large amount of carbon dioxide in the blank are remdved
by this procedure.
,
23. Kent and Tollens (1884, 1885) first reported that galactose is f1 source of mucic acid, although it had
been observed earlier that lactose may be converted to mucic acid by the action of nitric acid. According to Acree
(1921) impure nitric acid should be used, since the oxidation of galactose to mucic acid is accelerated by oxide_s
of nitrogen. All monosaccharides are oxidized by nitric acid to dicarbo,xylic acids, although only mucic acid precipitates. Saccharic acid (from glucose) is isolated as its sparingly soluble silver or mot;lopotassium salt (Morrow
Bnd Sandstrom, 1935).
24. Emil Fischer (1884, 1887) discovered that reducing sugars react with phenylhydrazine to form crystalline
derivatives called phenylhydrazones and phenylosazones. ,Weygand (1940) has presented evidence for the following
reactions:
CHO
I
CHOH
I
R
0-NHNH2
CH=NNH~
I
~CHOH
I
R
sugar
CH2-NHNH~
~-
C=O
hydrazone
CH2NHNH-Q
C=NHNH-Q'
I
R
Lf\_
\~-
NH)
/
CH=NH
b=NNH~
I
. \J_
CH=[Link]~
CH=NNH-Q'
or
C=NH
C=NNH-Q'
I
R
osazone
According to Fieser and Fieser (1944) cessation of the reaction after two molecules of phenylhydrazine have
become attached is explained by the formation of tautomeric chelate rings (shown below) which, through resonance,
are stabilized structures.
--
Q")
Fischer and Hirschberger (1888, 1889a) observed that mannose forms an insoluble phenylhrdrazone in almost
quantitative yield, and this principle was employed by Bourquelot and H~rissey (1899) for the quantitative determination of mannose i~ the presence of other sugars. Quantitative yields of osazones are obtained with 2,4-dinitrophenylhydrazine, but not with phenylhydrazine (Neuberg and Strauss, 1946). Numerous derivatives oLphenylhydrazine have been utilized for the separation and identification of carbohydrates (van der Haar, 1920). Hann and
Hudson (1944) prepared crystalline phenylosotriazoles by the reaction of osazones with copper sulfate.
r~NNH-D
CH=N
C=NNHJ\
C=N
"'=1-
phenylosazone
>-D O-NH2
+
R
phenylosotriazole
I
25. Dissolve the precipitate in the minimum volume of hot distilled water and allow the solution to cool
spontaneously to room temperature. Examine a drop of the suspension under the low power of a compourtd microscope and compare the crystals with the photomicrographs supplied by the instructor.
26. Militzer (1941) reported that benzhydrazide (C6HS CONHNH 2 ) and arabinose react to form a slightly soluble
hydrazone. This reaction has been employed for the quantitative determination of arabinose in plant gums (Hirst,
Jones, and Woods, 1947). Glucose, galactose, mannose, and pentoses other than arabinose form readily soluble
benzhydrazones. Rhamnose, when present in quantities larger than 300 mg., interferes with the preparation of
arabinose benzhydrazone.
27. According to Breddy and Jones (1945) xylose is the only sugar which forms a crystalline derivative when
treated under anhydrous conditions with benzaldehyde and methanol in the presence of acid. A possible structure
of the derivative is shown below.
Cadmium bromide and cadmium xylonate-form a characteristic double salt when the-sugar' is oxidized with bromine
in the presence of cadmium carbonate (Bertrand, 1891). Procedures have been reported for the quantitative determination of xylose with the aid of these derivatives.
- 83-
EXPERIMENT 37
/
j-H I
J
;)c-H
16
H-C2
113
5C---N
N-C
purine
pyrimidine
Both types of nucleic acids give adenine (6-aminopurine), guanine (2-amino-6-hydroxypurine), and cytosine
(2-hydroxy-6-aminopyrimidine)~ In addition, pentose nucleic acid yields uracil (2,6:-dihydroxypyrimidine)
and desoxype~tose nucleic acid thymine (2,6-dihydroxy-5-methylpyrimidine). Xanthine (2,6-dihydroxypurine) and hypoxanthine (6-hydroxypurine) are not present in nucleic acids. They 'are widely distributed
in nature and may be formed by enzymatic deamination of guanine and adenine. These topics have been
reviewed by Davidson (1949, Levene and Bass (1931), Cold Spring Harbor Symposia on Quantitative
Biology (1947), and Society for Experimental Biology (1947).
Johnson and Coghill (1925) reported that 5-methylcytosine was a component of tubercle bacilli,
but this observation has not beell confirmed (Vi scher, Zamenhof, and Chargaff, 1949). The methylated
purine derivatives (purine alkaloids) theophylline (1.. 3-dimethylxanthine), theobromine (3,7-dimethylxanthine), and caffeine (1,3,7-trimethylxanthine) are of interest because of their occurrence in tea,
cocoa, and coffee.
Chrbmatographic methods, first employed by Tswett (1906) to separate the pigments of green leaves
on a column of calcium carbonate, have been used to separate and characterize many types of biological
substances. Substances have been separat~d by (a) adsorption on columns of starch, cellulose, activated
carbon, silica. gel, alumina, and oth~r ads6ibents, (b) ion exchange with columns of synthetic zeolites,
synthetic r~sins, and other anionic- and cationic-exchange materials, and (c) partition on strips, cylinders,
and piles of filter paper. This topic has been reviewed by Cassidy (12 collaborators) (1948), Williams
and Synge (1950), Strain (1942), Zechmeister (1950), and Zechmeister and Cholnoky (1941).
Methods have been described for the separation and identification of nuc1eosides (adenosine,
guanosine, inosine, xanthosine, uridine, and cytidine), nuc1eotides (adenylic acid, guanylic acid,
uridylic acid, and cytidylic acid), purines, and pyrimidines derived from pentose nucleic acids (yeast
and pancreas) and from desoxypentose nucleic acids (yeast, thymus, spleen, and tubercle bacilli) by
partition chromatography on starch (Reichard, 1948; Edman, Hammarsten, Low, and Reichard, 1949;
Daly and Mirsky, 1949), countercurrent distribution (Tinker and Brown, 1948), ion exchange (Carter and
Cohn, 1949; Cohn, 1950), and paper chromatography (Carter, 1950; Hotchkiss, 1948; Markham ana Smith,
1949; Vischer and Chargaff, 1948a). Individual compounds on chromatograms are identified by their
ultraviolet absorption characteristics at wave lengths (region of 260 mil) corresponding to their absorption
maxima and are determined from their extinction values. Chromatograms on paper strips are located by
forming their uranyl salts (converted to brown-colored uranyl ferro cyanide) (Vischer, Magasanik, and
Chargaff, 1949) or mercury complexes (converted to black mercuric sulfide) (Vischer and Chargaff, 1948a)
-84-
as well as by areas of fluorescence induced by a mercury-vapor lamp (Markham and Smith, 1949; Carter,
1950; Holiday and Johnson, 1949) and by contact prints of images obtained with filtered ultraviolet
light on sensitized paper (Markham and Smith, 1949).
Purines and pyrimidines in a mixture of these substances. are to be separated and identWed by the
chr~fl1atographic methods described in the experimental part.
EXPERIMENTALl
Obtain approx~mately 0.1 g. of an unknowlJ mixture containing ribonucleic aeld and one or more of
the following compounds: adenine, guanine, hypoxanthine, xanthine, caffeine, theobromine, theophylline,
cytosine, S-methylcytosine, thymine, and uracil.
Hydrolysis of Ribonucleic Acid. 2 Transfer 0.10 g. of the unknown to a 4-in. test tube, add 2 mI.
(graduate) 01 2 N sulfuric acid, and shake the suspension to dissolve most of the solid. Heat the tube,
with occasional shaking, for 1 hr. in a boiling-water bath. Allow the tube tol cool to room temperature
and any sediment to settle. Use the supernatant solution for the experiments described below.
Rp Values of Purines and Pyrimidines.3 Place 5 strips of filter paper of convenient dimensions
(2 1/4 by 12 in.) on a clean sheet of waxed paper and draw on each a horizontal pencil line about 3/4 in.
from one end. (The strips should be handled with forceps, since grease or [Link] spots interfere with the
development of the chromatograms.) Using small glass tubes drawn to capillary tips, apply each solution
to be tested ori a 1/2-in. section of the pencil line. A maximum of three spots should be applied to each
strip in such a manner that the spots are separated and lie away from the edges of the paper. Before developing the chromatograms, allow the spots to dry in air.
Apply each of the following components 4 separately to paper strips: (a) cytosine and uracil, (b)
guanine, thymine, and xanthine, (c) adenine and hypoxanthine, (d) caffeine, theobromine, and theophylline, and (e) adenine, guanine, and 5-methylcytosine.
Suspend the strips, in the vessel which is provided, in such a manner that the- strips do not touch
each other or the sides of the container and that the bottom edges dip about 1/8 in. below the surface of
the solvent (1 part 2 N sulfuric acid and 3 parts n-propanol).5 Allow the 'chromatograms to develop until
the solvent front travels about 10 in. above the pencil line (16 to 20 hr.). Remove the strips, notcJt the
edges at the position of the solvent fronts, hang/the strips, and allow them to dry in air. Carry out the
indicated tests {described below) on the following strips: (a) iron-bromine-ammonia test, (b) diazotized
sulfanilic acid test, (c) silver-salt test, and (q) and (e) potassium-bismuth iodide test.
Color Tests for Purines and Pyrimidines. Since the reagents are toxic and corrosive, they should
be stored at all times in separate, labeled glass atomizers in'the fume hood, and 'they should be applied
to the paper strips only in the fume hood. The stril?s should be sprayed uniformly, since excess reagent
causes "running" of the solution. Since some colors fade rapidly, the position of each colored spot
should be marked as soon as possible by notching the' edges of the papers at positions corresponding to
the centers of the spots. 6
lron-Bromine-Ammonia Test.7 Spray a paper strip with FeS0 4-HCI solution, suspend the wet strip
in bromine vapor (generated from liquid bromine in the bottom of the container) for 2 to 4 min., remove
the strips, and allow the excess bromine to evaporate in the fume hood. Suspend the strip in ammonia
vapor (generated from concentrated ammonium hydroxide in the bottom of a container)"for about 3_ min.
Remove the strip and allow the amm;nia to evaporate. Uracil forms a pink color rapidly on exposure to
'
ammonia. Cytosine forms a fainter color whic~ appears only after evaporating the ammonia.
Diazotized Sulfanilic Acid Test. s Spray a paper strip with diazobenzenesulfonic acid reagent
(freshly prepared). A stable deep-red color is given by xanthine and guanine, while thymine yields a
transient red color. The o'ther purines and pyrimidines do not react.
Si Iver-salt Test. 9 Spray a paper strip with silver nitrate solution, immerse the wet strip twice in
[Link] ammonium hydroxide, allowing the excess liquid to drain after each immersion, and expose the
wet strip to hydrogen sulfide (generated by adding several drops of strong acid to a concentrated solution of sodium sulfide in the bottom of the container) until spots develop (10 to 20 min.). Adenine,
guanine,-hypoxantliine, and xanthine give a positive reaction.
- 85-
Potassium-Bismuth Iodide Test. tO Spray a paper strip with the potassium-bismuth iodide reagent,
allow the strip to dry in air for 10 to 15 min., immerse the 'strip in distilled water for 5 sec., remove fue
strip, and mark the spots which develop. Adenin~ and caffeine (at relatively high concentration) give
red or orange-red coiored spots before treatment with water. Guanine develops this color immediately
on immersion. Uracil, xanthine, and thYmine do not react. The other nucleic acid components appear
as red or orange spots on removing the strip from water. The colors produced by hypoxanthine, cytosine,
and 5-methylcytosine fade rapidly. Excessive hydrolysis (blackening) may obscure some of the colors
if the strip is permitted to remain too long in the water.
Identification of Purines and Pyrimidines in Unknown Mixture. Place four paper strips on waxed
paper and mark the pencil line as described. Apply the unknown containing the hydrolyzed yeast nucleic
acid to a 1/2-in. section of the pencil line at the center,of each strip, using a capillary glass tube as
described. Apply solution A (mixture of 1_typoxanthine, cytosine, 5-methylcytosine, thymine, and uracil,
each at the concentration given in Note 4) and solution B (mixture of adenine, caffeine, guanine, theobromine, theophylline, and xanthine, each at the concentration given in Note 4) to each strip at either
side of the unknown. Dry the strips in air and allow them to develop in the solvent mixture. Apply the
four tests separately to the developed, dried strips, using only one test per strip. Mark the solvent
fronts and the positions of the spots.
Submit a written report to the instructor giving the experimental RF values, the results of the
tests, and purines and pyrimidines identified in the ugknown.
NOTES
1. Mitchell, H. K., and Drell, William (unpublished data). The authors are indebted to Dr. H. K. Mitchell,
California Institute of Technology, for permission to use these procedures before publication.
2. Purines and pyrimidine nucleotides are readily liberated by mild acid treatment of nucleic acids. On the
other hand pyrimidine nucleotides are resistant to hydrolysis. They are commonly hydrolyzed w~th partial de0
struction by treatment at 175 for several hours with 20 per cent hydrochloric acid (Hunter and Hyenka, 1937), 25
per cent sulfuric acid (Levene, 1903) or 98 to 100 per cent formic acid (Vischer and Chargaff, 1948b). Under these
conditions the purines are destroye~, and cytosine may be, to a large extent, deaminated to uracil by hydrochloric
acid and sulfuric acid, but not formic acid, treatment. Uridylic acid is partially resistant to formic acid hydrolysis,
while cytidylic acid, which is hydrolyzed by formic acid under optimum conditions, gives recoveries of cytosine
not higher than 80 per cent (Vischer and Chargaff, 1948b). Bacher and Allen (1950) have shown that pyrimidine
0
nucleotides may be hydrolyzed to nucleosides in 95 per cent yield by treatment for 2 hr. at 100 with La(N03)3 at
pH 10.
3. In filter-paper chromatography the R F value 'is defined as the ratio of tpe length of travel of a component
(measured from the center of maximum concentr~tion) to the length of travel of the solvent front (Consden, Gordon,
and Martin, 1944). The R F value of a gl~en compound is constant for a particular solvent, and for mixed solvents
(for example, n-butanol and water) it reflects the partition coefficient of the solute between the liquids. The separation of the component substances in a mixture probably is dependent in part upon the degree of adsorption by
the filter paper.
4. The concentrations of the purine and pyrimidine solutions are 5 mg. per ml. except xanthine, which is
2.5 mg. per mI.
5. At higher proportions of n-propanol, unsubstituted purines separate mae widely but methylated purines
travel too close to the solvent front. Increasing the proportion of acid gives better separation of the methylated
purines but poorer resolution of unsubstituted purines.
6. The positions of the purine and pyrimidine spots may be detected before color development by examining
the air-dried strips under an ultraviolet (254-mp., Mineralite) lamp (Carter, 1950). Under the influence of the absorbed light the compounds appear as dark zones. Guanine fluoresces and is somewhat difficult to detect. Since
ultraviolet rays are harmful to the eyes, exposed strips should be examined only through a glass pane.
7. This test is based on the Wheeler and Johnson (1907) procedure for uracil and cytosine, which yield
purple precipitates after treatment in aqueous barium hydroxide lIYith excess bromine water or hypochlorous acid
(Johnson, 1943). The pyrimidines are converted to the colored barium salt of dialuric acid through the intermediates shown on the following page.
_ Rfl-
H-N-C=O
H-N-C=O
O=C
O=C
I CI/Br
I I""Br
H-N-C-H
H-N-C=O
I CI/H
I I""OH
I/OH
O=C
I I
""OH
H-N-C-H
H-N-C=O
I
OH
OH
dialuric acid
isadialuric acid
dibromoxyhydrouracil
~Substituted uracil derivatives (such as uridine), thymine, and the purines do, not respond to this test. Phenylhydrazine converts dibromoxyhydrouracil through isobarbituric acid (A) to the phenylhydrazone (B). Osazones (C)
are formed from ~substituted dibromoxyhydrouracil throue:h the intermediate isodialuric acid (Levene, 1925).
H-N-C=O
I C-OH
I
I I
H-N-C-H
O=C
H-N-C=O
o=l
H-N-C=O
O~t
l=NHNH-D
\J-
l-NHNH__f\
I I
H-N-C-H
(A)
R-N-C=NHNH-D
(8)
(e)
The blue color produced by treating dibromoxyhydrouracil (or analogous uri dine derivative) with urea, cyanide, and lithium arsenophosphotungstate is the basis of the quantitative method employed to determine uracil,
cytosine, uridine, and cytidine in nucleic acid hydrolysates (Pircio and Cerecedo, 1948; Soodak, Pircio, and
Cerecedo, 1949). It is of interest that no color is produced from uracil or cytosine on paper strips treated with
bromine and ammonia unless the strips are pretr~ated with iron.
8. Adenine, guanine, hypoxanthine, xanthine, and theophylline couple with diazobenzenesulfonic acid to
give intensely colored compounds (Burian, 1904, 1907), which were shown by the following series of reactions
(equations below) to be diazo compounds substituted in the 8 position (Fischer, 1909):
H-N-C=O
H-N-C=O
H-i-f=O
-o-CL
H_l_~_?-N=N f_'
O=C
C- -H
xanthine diazadichlorobenzene
....It:!4
CL
I C-N-H
I
I I )C-NH
H-N-C-N
8-aminoxanthine
I C-N-H
I
I I ""'C=O
H-N-C-?H
O=C
O=C
uric acid
In contrast to guanine and xanthine, adenine and other purines do not couple in dilute solution or in the presence
of excess base (Burian, 1904).
Pyrimidines react with diazobenzenesulfonic acid to give red-colored compounds. Thymine yields a more
intense color than uracil or cytosine, and in general the 4-methyl and the 5-methyl analogues are more deeply
colored than the unsubstituted pyr~midines (Johnson and Clapp, 1908-1909). Coupling is inhibited by substituents
in the 3 position of pyrimidines and the 7 position of purines. A specific test for thymine, based on this reaction
(Hunter, 1936), has been adapted to the quantitative estimation of thymine (Woodhouse, 1949).
9. Adenine, guanine, hypoxanthine, and xanthine form silver salts which are only slightly soluble in dilute
ammonium hydroxide. Black silver sulfide, formed after the excess silver is removed, indicates the position of
these purines. Under suitable conditions the purines and pyrimidines form silver, lead, mercury, and other metalli<
salts (Levene and Bass, 1931; Myers, 1909-1910). Purines ~re commonly determined as their silver salts, but this
method has shortcomings (Hitchings and Fiske, 1941). Mercuric salts (Vi scher and Chargaff, 1948a) and uranyl
salts (Vi scher, Magasanik, and Chargaff, 1949) have been used to identify purines and pyrimidines on paper strips.
10. Adenine forms a slightly soluble, red crystalline derivative of the following composition:
C,HsNs .m2BiI3 H 20
(Bruhns, 1890). Guanine forms a similar derivative:C s Hs Ns O. HI .2Bil3 2Hp (Levene and Bass, 1931). It has
been reported (Levene and Bass, 1931) that cytosine f~rms a red precipitate with this reagent. This reagent, prepared with strong nitric acid, has been used as a precipitant to identify alkaloids (Stephenson and Parker, 1921).
87 -
EXPERIMENT 3&
DETERMINATION OF VITAMIN C IN GRAPEFRUIT JUICE
Vitamin C, the lactone of L-threo-3-ketohe~uronic acid, is also known as cevitamic acid, ascorbic
acid, and the antiscorbutic vitamin. The last two names denote the sp~cific action of vitamin C in preventing scurvy in primates and the guinea pig. Other animals are not subject to this disease because of
bacterial flora in their intestinal tracts which synthesize vitamin C. Scurvy has been known for ce~turies
as an epidemic disease,occurring especially among soldiers and sailors deprived for long periods of time
of fresh fruit juices and'vegetables. The disease is characterized by pathological changes in the teeth,
bleeding gums, decreased resistance to infections, sore and swollen joints, anemia, hemorrhagic lesions
of the skin 'and eyes, susceptibility to bone fracture, and delayed healing of wounds.
Szent';-Gyorgyi (1928) isolated vitamin C from oranges, ~abbage, paprika, and the adrenal gland. The
constitution of vitamin C was determined through the work of Hirst (1933), Haworth (1933), and Haworth
and Hirst (1933). Vitamin C was synthesized by Reichstein, Griissner, and Oppenhauer (1933). The
synthetic methods lead to vitamin C through the lactonization of 2-keto-L-gulonic acid (prepared from
D-glucose through D-sorbitol and L-sorbose) and of 3-keto-L-gulonic acid (prepared from L-xylose).
COOH
COOH
I
I
HO-C-H
I
I
C=O
I
H-C-OH
I .
HO-C-H
I
CH 0H
H-C-OH
C=O
H-C-OH
HO-C-H
CH20H
2- ketogulonic acid
3-ketogulonic acid
Vitamin C has been determined in blood, urine, plant extracts, and other materials by biological,
chemical, and physical methods. The biological methods depend upon the determination of the smallest
amount of vitamin C which will prevent decrease in weight, th(( appearance of signs of scurvy (Sherman,
LaMer, and Campbell, 1922), or histological changes in the teeth (Reid, 1947) of guinea pigs. Biochemical procedure;:; are based on the preferential oxidation of vitamin C by ascorbic acid oxidase' (Tauber
and Kleiner, 1935). Because this reaction was nonspecific, Stewart and Sharp (1945) oxidized all reducing substances with an excess of the oxidase, selectively reduced the dehydroascorbic acid with
Escherichia coli, and determined the ascorbic acid by titration with 2,6-dichlorophenol-indopqenol.
Vitamin C has been determined directly by titration with iodine (Adams, Acker, and Frediani, 1947),
2,6-dichlorophenol-indophenol (Tillmans, Hirscn, and Reinshager, 1928), methylene blue (Gal, 1936;
Lund and Lieck, 1936), and other oxidizing agents. Physical methods include determination of the
intensity of the absorption spectrum at 265 mJl (Robertson, 1934) and of the oxidation-reduction
potential with the polarograph (page and Waller, 1946).
Biological methods ale time-consuming, expensive, and lac1dng.l.n precision but have the advantage
of specificity. Chemical procedures often are not highly reliable because of'incomplete extraction of
vitamin C, susceptibility of vitamin C to oxidation by air, and interference of oJher reducing substances.
-88-
,.....The direct visual titration _of vitamin C with 2,6-dichlorophenol-indophenol by the methods of Bessey
(1938a, 1938b) and Bessey and King (1933) have been u~ed widely, but numerous modifications have
been suggested to permit more accurate determination by photometric and potentiometric procedures
(Association of Vitamin Chemists, 1947; Bolomey and Kemmerer, 1947; King, 1936, 1941; Mills and
Roe, 1947; Rosenberg .. 1942; Satherfield, '1947).
It has been found (American Can Company, 1943) that citrus juices, peppers, strawberries, broccoli, beet greens, horse-radish, and kohlrabi are [Link] sources of vitamin C. Good sources are
tomatoes, ,cabbage, apples, grapes, and other fruits and vegetables, while poor sources include celery,
asparagus, bananas, green peas and beans, rhubarb, and potatoes. The vitamin C contents (given in
parent~eses as milligrams per 100 g. of product) of representative foods [Link] (20 to 1,80), fresh
lemon juice (53 to 56), fresh tomatoes (10 to 60), raw cabbage (18 to 180), fresh bananas (2 to 15), and
celery (1 to 10). Significant studies have been made of vitamin C in canned foods, different types of
foods under varying enviro_nmental conditions, and various body tissues ~ndlfluids in health and disease.
The daily allowance of vitamin,C recommended by the Committee on Foods and Nutrition of the Natio~l
Research Council varies from 30 mg. for children under one ,Year to 150 mg. for lactating, mothers (Nl:jtional
Research Council, 1943).
Vitamin C is to be determined in grapefruit juice by a modification of the method of Ramsey and
Colichman (1942). Total reducing ~ubstances are determined by reaction with iodate, vitamin C in another
aliquot is oxidized with 2,6-dichlorophenol-indophenol, and the remaining reducing suostances in the
dye-oxidized aliquot ar determined by reaction with iodate. Equations for these reactions are shown
below:
0=1
KI03
OH
i=:J
I
"
o=i=-l
c=o
I
I
0
----
H-C
HO-C-H
Ascorbic acid
KI
CH20H
ascorbic acid
6HCI
HO-C-H
3H20
H-C
CH20H
5KI
r:J
I
dehydroascorbic acid
- - 6KCI
HO~
~O
~
HO~
CI~N~
CI---l0LN~
rJr-OH
dehydroascorbic acid
Fresh grapefruit juice contains an average of about 41 mg. of ascorbic acid per 100 g. compated
with 33 mg. per 100 g. of the canned juice.
- 89-
EXPERIMENTAL
I
Obtain about 210 ml. of grapefruit juice which has been protected from air by preservation with
solid carbon dioxide. Pipette two 50-mI. aliquots into a 250-ml. volumetric flask, add 50 ml. of N hydr()chloric acid and 1 ml. of a 5 per cent solution of 8-hydroxyquinotine sulfate as an inhibitor of atmospheric
oxidation (Highet and West, 1942), mix the solutions, and dilute the mixture to the mark with distilled
water. Mix the solutions and preserve the mixture in a stoppered flask labeled solution A.
Pipette one sO-ml. aliquot of the grapefruit jui-;:e into a se~ond 2s0-ml. volumetric flask. Add 50
mI. of N hydrochloric acid, 1 ml. of a 5 per cent solJltion of 8-hydroxyquinoline sulfate, and an aliquot
of a standard solution of pure ascorbic acid (the latter is to be added by the instructor). Mix the solutions, dilute the mixture to the mark with distilled water, mix the solutions, and preserve the mixture in
a stoppered flask labeled solution B.
Pipette a 2S-ml. sample from flask A into a 125-ml. conical flask. Add 1 mI. of iodine-free 0.1 N
potassium iodide solution, 2 mI. of a 1 per cent starch indicator solution, and a piece of solid carbon
dioxide about 1/2 in. in diameter. Titrate the solution with standard 0.001 M (0.006 N) KIO, solution
(lO-ml. burette). The end point is reached when the solution turns blue, owing to the formation of the
starch-iodine complex.
Pipette a second 25-ml. sample from flask A into a 12s-ml. conical flask. Maintain the temperature
at 20 to 25, protect the solution with solid carbon dioxide, and titrate the solution with a 0.1 per cent
solution of 2,6-dichlorophenol-indophenol (sodium salt)delivered from a 50-mI. burette. The end point
is reached when a definite pink color appears, indicating a slight excess .of the red dye. Add 2 mI. of 0.1
N potassium iodide solution and 2 ml. of starch indicator solution. Continue the titration by adding
standard 0.001 M KIa, solution dropwise, allowing 5-sec. intervals between drops.
Pipette a 25-ml. aliquot from flask B into a 125-ml. conical flask and repeat the described titration
procedure. Repeat this process using a second 2S-ml. aliquot. Repeat these titrations until the volumes
of standard KIa, solution used agree within 0.05 mI.
- 90-
EXPE;RIMENT 39
DETERMINATION OF NICOTINIC ACID BY MICROBIOLOGICAL ASSAY
WITH LACTOBACILLUS ARABINOSUS 175
Nicotinic acid was first prepared by Huber (1867) as an oxidation product of nicotine (equation
below); it was first isolated from a natural material (rice) by Suzuki, Shinamura, and Odake (1912); and
H
I H2-C
~C,C-C-H
I
H-C
I
II
H-C~N/C-H
~N"//
IIc
C-H2
C-H2
H-C-P 'C-COOH
H-C,
-~N/
C-H
CH 3
nicotinic acid
nicotine
it was first determined to have vitamin activity by Elvehjem, Madden, Strong, and Woolley (1937).
Nicotinic acid is widely distributed in natural materials, but liver, wheat bran, yeast, and peanut are
among the best sources (Krehl. de la Huerga, Elvehjem, and Hart, 1946). Data are available on the
nicotinic acid content of dehydrated foods (Heberlein and Clifcorn, 1944), canned foods (American Can
Company, 1943), prepared foods (Sarett, Bennett, Riggs, and Cheldelin, 1946), meat (Greenwood, Kraybill, Feaster, and Jackson, 1944), feeds (Hale, Davis, and Baldwin, 1942), and germinating grains
(Klatzkin, Norris, and Wokes, 1948).
The classical investigations of Goldberger and coworkers of the U. S. Public Health Service led
to the discovery that nicotinic acid deficiency results in pellagra (dermatitis, redness of the tongue,
and other symptoms) in humans, blacktongue in dogs, atrophy of the bone marrow in rats, and other
manifestations in these and other animals. Sebrell (1934) has reviewed Goldberger's studies. Dieta~y
nicotinic acid is required because it is concerned with glycolysis and respiration as a functional part
of coenzymes I and II. The daily allowances of nicotinic acid recommended by. the Food and Nutrition
Board of the Nationa~ Research Council (National Research Council, 1943) are 4 to 12 mg. for children
up to twelve years and about 20 mg. for older children and adults.
Chemical and microbiological methods have been proposed for the quantitative determination of
nicotinic acid, but there appears to be no .satisfactory animal assay procedure (Dann, 1947). The chemical methods are based on the observation of Konig (1904), who found that pyridine reacts with cyanogen
bromide and aniline, or other aromatic amine, to yield a yellow-colored compound.
Nicotinic acid has been determined microbiologically with Proteus (Lwoff, 1938) and with Shigella
paradysentenae (Dorfman, Horwith, Koser, and Saunders, 1939) but most commonly with Lactobacillus
casei (Snell and Wright, 1941; Roberts and Snell, 1946). Nicotinic acid, nicotinamide, and nicotinuric
acid (nicotinylglycine) have been determined microbiologically in urine with Lactobacillus arabinosus,
and Leuconostoc mesenteroides (Johnson, 1945; Johnson, Hamilton, and Mitchell, 1945b) and Nl-methylnicotinamide has been determined by a fluorometric method (Perlzweig, 1947). In Neurospora crassa
mutants, hydroxyanthranilic acid is a precursor of nicotinic acid (Mitchell and Nyc, 1948).
Microbes were first observed by van Leeuwenhoek in 1685 (Harris, 1921), and they were shown to
be the causative agents of fermentation, putrefaction, and disease by Pasteur about 1860. Investigations
concerning the nutritional requirements of microorganisms by Uschinsky (1893), Henneberg (1903), OdaJensen (1919), Fred, Peterson, and Anderson (1921a, 1921b), Koser and Rettger (1919), Wood, Geiger,
and Werkman (1939-1940), Mueller (1922, 1935a, 1935b), and later workers led to the discovery that
some vitamins are essential growth substances for lactic acid bacteria. The first practicable micro- 91-
biological assay procedure for the determination of a vitamin was that proposed for riboflavin by Snet)
and Strong (1939). Since that date analogous methods have been described for many vitamins and
amino acids.
'
Microbiological methods are based on the premise that growth of the organism is directly proportional to the concentration of any essential nutrient. Growth produced in response to graduated amounts
of the standard (pure nutrient) and the test sample is measured in terms of titratable acid (or density of
cell suspensions), response curves are drawn, and the content of essential substance in the test sample
is estimated by interpolating the -standard curve.
Microbiological methods generally are reasonably satisfactory, although they vary widely in precision and accuracy depending upon the vitamin or amino acid to be determined, the material to be
assayed, the types and amounts of interfering stimtUatory or inhibitory substances in the test sample,
the microorganism employed, the number of levels and the number of replicate tubes at each level of
standard and unknown, the composition of the basal medium, the skill of the microbiologi_st, at}d other
factors. Assay results are considered to be dependable if the mean deviation from the mean of the
titrations at different levels of sample is low (2 or 3 per cent), if added pure substance is recovered
nearly quantitatively (usually 100 5 per cent), and if the values found agree closely with.. those
obtained by chemical or other methods known to be reliable.
The chemistry, physiological activity, microbiological assay, and clinical uses of nicotinic acid
have been reviewed by Rosenberg (1942), Association of Vitamin Chemists (1947), Merck and Company
(1939), Snell (1947), Dann 1947), Perlzweig (1947), and Gyorgy (1950).
'..
EXPERIMENTAL
Obtain a commercial vitamin tablet 1 containing nicotinic acid (niacin) or nicotinamide (niacinamide). Weigh the tablet to 1 mg. and grind it with 0.5 mI. of distilled water in a clean mortar. Add
5 mI. of distilled water, stir the suspension, and transfer the supernatant solution quantitatively ~o a
volumetric flask of such size that, when filled to the mark, it will contain approximately 0.050 pg
(1 p.g equals 0.001 mg.) of nicotinic acid (from the tablet) per milliliter. Add 5 mI. of distilled water to
the residue in the mortar, stir the mixture, and transfer the suspension quantitatively to the volumetric
flask. Add distilled water to the mark and mix the liquids (and any suspended material) thoroughly.
Pipette 50.0 mI. of the diluted solution into a 100-ml. volumetric flask; submit the flask to the
instructor, who will add a known quantity of pure nicotinic acid. Add distilled water to the mark and
mix the liquids thoroughly.
Pipette into (duplicate) clean, dry 4-in. test tubes distilled water, standard, sample, recovery
sample, and basal medium (in the order given) according to the schedule shown in the accompanying
table. A graduated pipette or a pipetting machine may be used. Plac.e the 40 tubes in a rack, mark
the position of each tube on a diagram of the rack, cover the rack with a double-layered cap of clean
Boote toweling, constructed so as to fit snugly, and autoclave the rack of tubes for 15 min. at 15
p.s.i. pressure (about 121C.). Allow the tubes to cool to room temperature and store them in the
refrigerator.
Prepare an inoculum of Lactobacillus arabinosus 17-5 (American Type Culture Collection
No. 8014), which has been carried on'a yeast-glucose-agar stab preserved in the refrigerator, by the
following procedure: While holding in the right hand a Nichrome wire attached to a nonconducting
handle, pass the wire through a blue gas flame until all parts of the wire have been heated to redness.
In this and all subsequent operations avoid touching the wire to any object. Allow the wire to cool in
air for about 15 sec. Hold the tube containing the L. arabinosus stab culture and a 50-mI. centrifuge
tube containing 20 mI. of sterile inoculum medium in the left hand, grasp the two cotton plugs between
the three smallest fingers of the right hand with the palm held upward, remove ~he cotton plugs and
rotate the tubes for about 10 sec. while holding the mouths in a blue gas flame. Insert the sterile wire
- 92-
into the agar stab containing the culture, place the wire tip in the liquid inoculum medium, and stir the
liquid for a few seconds. Repeat this inoculating process once without resterilizing the needle. Flame
the mouths of the tubes and replace the cotton plugs. Turn down the ed~es of the cotton plug over the
edge of the centrifuge tube and place a rubber band over the cotton edges. Incubate the inoculum for
24 hr. at 3S 0 and preserve the inoculum (not longer than 24 hr.) in the refrigerator.
Centrifuge the inoculum culture, remov~ the plug, flame the mouth of the tube, decant the supernatant liquid, add (aseptically) 10 mI. of sterile saline (0.9 per cent sodium chloride solution), using a
sterile pipette 2 or automatic pipette with sterile syringe assembly, stir the mixture with a sterile wire,
flame the mouth of the tube, replace the plug, centrifuge the suspension, decant the supernatant liquid,
add 40 mI. of sterile saline, and stir the mixture. Pipette 0.1 ml. (2 drops) of the inoculum suspension
into each of the forty 4-in. test tubes (while pipetting, uncover only about half of the tubes at a time)
and incubate the rack of (covered) tubes for 72 hr. at 35 o. Steam the rack of tubes for 10 min. (about
100C.) in the autoclave to stop acid production (by destroying most of the/organisms) and to remove
carbon dioxide.
Transfer the contents of each tube quantitatively to a 50-mI. conical flask, using two S-I1lI. portions of distilled water as rinse fluid. Add 2 drops of bromthymol blue indi!=ator solution to each tube
and titrate the acid in each tube with standard, approximately 0.030 N sodium hydroxide delivered from"
a SO-ml. burette. A blue-green color appears at the end poinf (PH 6.8). Titrate the acid in the remaining
tubes to the same end-point color (preserve the solution first titrated, and each fifth one thereafter, for
comparison).
Plot the experimental data on millimeter coordinate paper, label the ordinate Volume of
N
sodium hydroxide and the abscissa Micrograms of nicotinic acid per tube, draw a smooth curve, and interpolate this standard curve.
TYPES AND VOLUMES IN MILLILITERS OF SOLUTIONS
TO BE PIPETTED INTO 4-INCH TEST TUBES
Solution
Distilled water
Standards. .. . " ..
Basal medium.
Tube number
1
2.50
0
0.50
2.25
0.25
0.50
2.00
0.50
0.50
1.75
0.75
0.50
1.50
1.00
0.50
1.25
1.25
0.50
1.00
1.50
0.50
0.75
1:75
0.50
0.50
2.00
0.50
0
2.50
0.50
Tube number
Solution
Distilled water
Sampleb
Recovery sample b
Basal medium
10
11
12
2.00
0.50
0
0.50
1.50
1.00
0
0.50
13
1.00
1.50
0
0.50
14
15
16
17
18
19
0.50
2.00
0
0.50
0
2.50
0
0.50
2.00
0
0.50
0.50
1.50
0
1.00
0.50
1.00
0
1.50
0.50
0.50
0
2.00
0.50
20
0
0
2.50
0.50
Constituent
. . . ..
. . ... . . . . ...
. . . . ..
. .... . . ... . . , ..
. . . .
... . . . ... . . . .. . ... .
.. ... .. . . . '".. .
. . . .. ...
. ...
. ... .. . . . ..
.. . ... . . . ...
. . . .. . ...
. . .. . . . . ...
........ ........ .......
... ... . . . .,. .
... . . ... . . .. .... .
.. . . . . ...
. .
.. .. . . .. ....
..
. . . .. . . ... . . ..
. . .. . .
. . .... . .
"
22.5
6.0
3.6
5.0
m1.
g.
g.
mi.
0.75
2.0
1.2
2x lO-a
5x 10""'"
IX 10-1
.
5.0 m1.
.
.
.
2.5 mi.
.
.
.
1.5 mi.
IX 10-4
IX 10-4
2x 10"""
2x 10-4
IX 10-.'5
5x 10-'
1 X 10""
1 X 10""'"
IX 10...3
.'.
.
1.5 ml.
.
.
.
5X 10-2
5x 10-2
2X 10-2
-
IX 10""'"
IX 10-..3
8 The specified quantities of the indicated substances and solutions (preparation described in the Appendix)
are dissolved in 45 mI. of distilled water, N sodium hydroxide solution is added dropwise to pH 6.8 (olive-green
color of bromthymol blue indicator solution tested on a Spot plate), and the mixture is diluted to 50.0 mI. with
distilled water.
It is convenient to prepare a dry ball-mill mixture of the components (except glucose and casein hydrolysate),
as described by Carnien and Dunn (1948), and to prepare the basal medium as follows: Suspend 0.49 g. of the dry
ball-mill mixture in 20 mI. of distilled water, add 6.0 g. of glucose and 22.5 mI. of the casein hydrolysate, stir and
heat the mixture until the solids dissolve, neutr'alize the solution to pH 6.8, and dilute the neutral solution to 50.0
mI. This solution should be preserved '[Link] toluene.
b The values refer to the concentrations in the final 3-ml. volumes per tube prepared as described in the preceding table.
NOTES
1. If a tablet containing yeast or other food material is to be assayed, suspend it in 10 parts of 3 N sulfuric
acid and autoclave the suspension for 30 min. at 15 p.s.i. pressure.
2. Remove one of the sterile pipettes carefully from the can of pipettes provided by the instructor. The pipette
will contain a piece of cotton inserted about 1/2 in. into its mouth. Sterilization is effected by wrapping' each
pipette in a towel and autoclaving for 15 min. or by placing the pipettes in a metal can and heating the can for
1 1/2 hr. in an oven at 160 to 180 o.
EXPERIMENT 40
DETERMINATION OF CHOLINE IN EGG YOLK BY MICROBIOLOGICAL ASSAY
WITH CHOLINELESS NEUROSPORA CRASSA
I
choline
neurine
lecithin
muscarine
acetylcholine
On diets deficient in choline, fat is deposited in the liver and hemorrhages occur in the kidneys
(Best, Hershey, and Huntsman, 1932; Griffith and Wade, 1940). Disorders ari~ing from choline deficiency include disturbance of the blood lipid levels (Paul, Daum, and Kemp, 1947), decreased erythropoieSis (Nutr. Rev., 1948a), and formation of neoplasms in the liver and other organs (Copeland and
Salmon, 1946). It has been found that choline is an antiperosis (perOSis, slipped tendon) factor for
chicks (Almquist, 1946). About 2 mg: of choline are excreted daily in the urine of adults (Borglin, 1947)
and 3 to 15 p.g per 100 ml. of sweat (Johnson, Hamilton, and Mitchell, 1945a). The lethal dose of choline (for the rat) is 29 to 75 p.g of choline chloride per 100 g. of body weight, depending upon the mode
of administration (Hodge, 1944).
Choline is synthesized in the body, but the rate is too slow to provide for all of the metabolic
needs (Stetten, 1942). The in vivo precursors of choline are ethanolamine (NH 2CH 2CH 2 0H) and methyl
groups, the latter derived from methionine, sarcosine (CH l NHCH 2COOH), and betaine [(CH,)3-+NCH 2COO-]. The need for choline and other methyl donors is accentuated on diets high in nicotinic
- 95-
acid, owing to the formation and excretion of methylated compounds 'such as N I- methylnicotinamide,
nicotinuric acid, and trigonelline (methylbetaine of nicotinic acid) (Perl zweig and Huff, 1945). Methionine,
the active methylating agent in the synthesis of creatine from' glycocyamine, may be formed from choline
and homocystine (Almquist, 1946). Choline may function other than as a methyl donor, since nonmethylated analogues are effective lipotropic agents (McArthur, 1946). The role of choline in transmethylation
(du Vigneaud, 1941, 1942-1943),and as a lipotropic factor (Nutr.,Rev., 1948b) has been reviewed.
Choline has been estimated biologically by determining the level of test material required to protect ag~inst fatty infiltration of the liver (Griffith, 1941) or kidney hemorrhage (Engel, 1942) and by
determining the acetylcholine (formed from choline), assayed with the aid of isolated rabbit intestine
(Fletcher, Best, and Solandt, 1935). The liberati0ll of trimethylamine by oxidative deamination is the
basis of a quantitative method employed by Klein abd Linser (1932b, 193,3). The precipitation of the
periodate (Roman, 1930), the reineckat~ (Paal, 1929), and the phosphotungstate (Street, Kenyon, .and
Watson, 1946) has been utilized for the determination of choline. The most satisfactory method probably
is microbiological assay with cholineless Neurospora crassa, as described by Horowitz and Beadle
(1943) .. This procedure has been modified by Luecke and Pearson (1944a, 1944b), Siegel (1945), and
Hodson (1945). Methods for the determination of choline have been reviewed by Handler (1947b),
The choline content of egg yolk is to be determined by microbiological assay with cholineless
(strain No. 34486) Neurospora crassa, essentially by the method of Horowitz and Beadle (1943), as
modified by Hodson (1945).
EXPERIMENTAL
Prepare a stock culture of cholineless Neurospora crassa as follows: Obtain an agar slant inoculated with the organism and incubated for at least 3 days at room temperature. Using the aseptic
technique described in Exp. 39, streak the surface of a sterile agar slant lightly with conidia (spores)
from the tinted part of the growth. Allow the inoculated tube to stand for 5 to 7 days at room temperature.
Obtain an egg (chicken, duck, turkey, or other type), boil the egg for 10 min., separate the yolk,
and weigh (to 1 mg.) about 200 mg. of the yolk on weighing paper. Transfer the weighed quantity of
yolk quantitatively to a 50-ml. conical flask, add 3 mI. of N sulfuric acid, break up the particles with
a stirring rod, add 17 mI. of N sulfuric acid, stir the mixture, rinse the stirring rod, plug the flask with
cotton, and autoclave the mixture for 3 hr. at 15 p.s.i. pressure. Cool the flask to room temperature,
transfer the acid solution quantitatively to a 1,OOO-ml. volumetric flask, add about 800 mI. of distilled
water, and add sufficient N sodium hydroxide (20 mI.) to bring the pH to 7 (olive-green c~ior of bromthymol blue indicator solution tested, on a spot plate). Add distilled water to the mark and mix the
solutions thoroughly.
Pipette 50.0 mI. of the solution into a 100-ml. volumetric flask, add distilled water to the mark,
and mix the solutions thoroughly. Pipette 25.0 ml. of the egg solution into a lOO-ml. volumetric flask
and submit the flask to the instructor, who will add a known quantity of choline. Add distilled water tc
the mark and mix the solutions thoroughly.
Pipette distilled water, standard solution, and basal medium (see the accompanying table) into.
one series of clean, dry, numbered (with a pencil on the frosted spot) 50-mI. conical flasks (or 2-oz.
widemouthed bottles), according to the schedule shown in the accompanying table. Similarly, pipette
distilled water, sample solution, and basal medium into a second series of flasks, and distilled water,
recovery-sample solution, and basal medium into a third series of flasks. Use a 10-ml. graduated
pipette for these transfers. Pipette 10 mI. of distilled water to two additional 50-mI. conical
flasks.
Plug each flask with cotton. Autoclave the flasks, a 1-ml. graduated pipette, and a metal spatull
at 15 p.s.i. pressure for 15 min. Allow these articles to cool 'to room temperature. Using aseptic tech
nique, transfer with the aid of the spatUla several portions of conidia from the growth on the N eurospor.
slant to a flask containing 50 mI. of sterile distilled water, and prepare a uniform suspension by rotatil
- 96-
the tlask. Pipette 1 drop of the suspension to each of the 16 flasks and allow the inoculated solutions
to stand at room temperature for 5 to 7 days.
Swirl the mixture in each flask to collect the growth as a single mass, transfer the contents of
each flask to a Hirsch funnel, apply suction until nearly all the liquid has been removed, wash the
solid mat of mycelium once with S mI. of distilled water, roll the mat into a compact mass, and transfer
it to a 4-in. test tube which is numbered 'and labeled. Treat the other samples similarly. Dry the unstoppered tubes overnight at 90. Weigh each mycelium mat to 0.2 mg. (a Roller-Smith precision torsion
balance is convenient-for this purpose).
Plot the experimental values obtained from the standard solutions on millimeter coordinate paper,
label the ordinate Milligrams of mycelium and the abscissa Micrograms of choline, and draw a smooth
curve. Estimate the ch-;line in the test and recovery samples by interpolating the standard curve.
Solution
1
Distilled water
Standard solution B
Sampleb
Recovery sample b
Basal medium.
10
11
12
13
14
15
16
. . . . . . . . .. . .
. . .. .. .. .. .. .. . . . . . . . .
Constituent
Sucrose . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Biotin solution
Salts solution . . . , . . . . . . . , . . . .
NH400C(CHOH)aCOON14
NH4NOs '
4.0 g.
1.0 mI.
50.0 mI.
Copper , . . . . . . . . . . . . . . . . . . . . . . . . . . .
Manganese
Molybdenum .
Boron ....
2.0
5 X 10-7
0.5
0.1
0.1
0.05
0.01
0.01
K 2HP04
MgS04 7H20
NaCl... " . . . . . . . . . . . . . .
CaCI2 6H20
Microconstituents solution
Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Iron -
10.0 ml.
2 X 10-4
2 X 10-5
1 X 10-5
2 X 10"'11
2 X 10-<1
1 X 10-15
B The specified quantities of the indicated substances and solutions are dissolved in distilled water, and the
solution is diluted to 100 mL in a volumetric flask. Check the pH with a pH meter and, if necessary, adjust the pH to
5.5. Preserve the solution under toluene. If any suspended material is present, shake the mixture thoroughly before removing an aliquot.
b The values refer to the concentrations in the final 1()..m1. volume per flask as described in the preceding
table.
- 97-
EXPERIMENT 41
DETERMINATioN OF VITAMIN D BY THE CHICK BONE-ASH METHOD
The chemistry and physiology of vitamin D ,have been reviewed by Rosenberg (1942) and Bills
(1935). The discovery by Mellanby (1919) that aniptat fats promote normal calcification of bones was
followed by the postulate of McCollum, Simmonds, and Becker (1922) that. a vitamin was responsible for
this effect.
It was shown later by Windaus, Rosenheim, Hess, Steenbock, and others that cholesterol irradiated
with ultraviolet light had antirachitic activity and that ergosterol, present as an impurity, was the substance activated. The irradiated products include calciferol (vitamin D2) and the inactive or relatively
toxic substances lumisterol, tachysterol, toxisterol, and suprasterols I and II. Crystalline vitamins D2
and D3 have been isolated from tuna-liver oil (Broc~man, 1936; Brockman and Busse, 1938), and vitamil1
0 3 has been synthesized from cholesterol (Windaus, Lettre, and Schenck, 1935; Schenck, 1937). 7-Ketocholesterol, formed by the oxida,tion of cholesterol, was reduced to 7-hydroxycholesterol, and the latter
(as its dibenzoate) was decomposed by heat to 7-dehydrocholesterol. Vitamin D3 was isolated from the
irradiated products of 7-dehydrocholesterol. Formulas are given below for vitamin D 2, vitamin D 3, and
related compounds (see Exp. 27 for cholesterol and the numbers of the carbon atoms).
CH3
CH3
I /
CH3
H-C--C=C-C-CH
CH3
HO
CH3
CH3
ergosterol
CH3
I /
I I I \
CH
3
H-C--C=C-C-CH
CH3
CH3
OH
calciferol (vitamin D2)
7 - dehydrocholesterol
(provitamin D3)
- 98-
I I I
/CH
H-C--C-C-C-C-H
. 11 1
~3
OH
vitamin D3
Milk, eggs, and other common foods are relatively poor sources of vitamin D, but the liver oils
from the cod and other fish (especially tuna and other members of the Percomorphi are excellent sources
(Bills, 1935). It has been reported that the liver oil of the California bluefin tuna contains 46,000 Intetnational Units of vitamin D per gram and Japanese bluefin tuna 61,000 International Units per gram.
Fish synthesize vitamin D, but higher animals depend upon exogenous sources, although some antirachitic substance is formed in the skin by the action of ultraviolet light. Vitamin D is stored by
animals as long as 12 weeks, and it is transferred from the mother to the fetus (Embleton and Collings,
1947). The Committee on Foods and Nutrition of the National Research Council (1943) has recommend..
ed a daily allowance of 400 to 800 International UnHs of vitamin D for all individuals. The toxic ~ffects
of massive doses of vitamin D2 and D! on dogs, rats, humans, and other animals have been 1eviewed by
Morgan (1943), Paul (1946), and Ziskin, Gibson, Skarka, and Bellows (1943). Chicks deficient in vitamin D develop black pigmentation of the feathers (Decker and McGinnis, 1947).
The rat line test for the bioassay of vitamin 0 was devised by Shipley, Park, McCollum, Simmond;;;,
and Parson (1920-1921) and McCollum, Simmonds, Shipley, and Park (1922), who stated that "this test
depends on the power of a given substance to cause the reappearance of a provisional zone of calcification in epiphyseal cartilages of animals with very severe rickets whose cartilages had been rendered
calcium free by faulty diets." A chick bone-ash method proposed by Hart, Kline, and Keenan (1931)
was the outgrowth of earlier observations made by lBethke, Steenbock, and Nelson (1923-1924). The rat
line test has been discussed by Bills (1947) and the chick bone-ash procedure by Waddell and Kennedy
(1947b). The line test cannot be used with chicks since a line of calcification does not form during the
healing process. The rachitogenic diet proposed by Hart, Kline, and Keenan (1931) has been modified
by later workers, and numerous comparative studies have been made of methods for the determination of
vitamin D. Willgeroth, Halpin, Halloran, and Fritz (1944) have described a method for the determination
of vitamin D with turkeys.
The International Unit of Vitamin D was defined originally as the vitamin D activity of 1 mg. of the
international standard solution (olive oil) of ergosterol irradiated under defined conditions. Later it W$S
stated that the International Unit was the vitamin D activity of 1 mg. of the international standard solution of irradiated ergosterol which had been found equal in potency to 0.025 p.g of crystalline vitamin D2
(calciferol). One unit had a potency such that, given to a rachitic rat for 8 successive days, a wide
line of calcium deposit was produced in the metaphysis of the proximal ends of the tibiae and of the
distal ends of the radii.
The United States Pharmacopeia unit is equal in antirachitic potency for the ;;t to 1 International
Unit, but cod-liver oil is employed as the reference standard. The U.S.P. reference cod-liver oil has
been adopted as the standard for the bioassay of vitamin D oils by the chick bone-ash method, where
the chick unit was equal to 1 unit of vitamin D in the U.S.P. reference cod-liver oil. The first referende
oil contained 95 units per gram and the second 115 units per gram. Crystalline vitamin D! was assigI'ed
a potency of 40 million units per gram, but Waddell and Kennedy (1947a) have reported that it contains
50 million units per gram.
Vitamin D2 and vitamin D! have approximately equal potency for the rat, but vitamin D2 has almost
no chick activity. Hence the response of chicks to vitamin oils varies, depending upon the proportions
- 99-
of the various forms of vitamin D present. The livers of the white sea bass, the dogfish, and other
species of fish contain forms of vitamin D which give more than 100 per cent activity when tested by
the chick method (Jukes and Sanford, 1939).
'
Vitamin I) is to be determined in an un,1mown fish-liver oil by the chick bone-ash procedure deI
scribed in the experimental part.
EXPERIMENTAL1
Obtain a sample of vitamin D oil (standard or test) containing 0 to 20 A.O.A.C. units per gram.
Prepare 3 kg. of the A.O.A.C. rachitic basal ration2 as follows: Transfer the weighed. quantity
of each food material to a large dish, knead the foods until they are thoroughly mixed, and sieve the
mixture through the screen provided for that purposJ. Place the vitamin D oil in a I-oz. sample ,bottle
fitted with a cork and a medicine dropper, weigh the bottle with its assembly to 10 mg., and transfer
20.0 g. of the oil to a 100-ml. beaker. Add 60 mI. of petroleum ether (b.p. 30 to 60 0 ) and stir the mixture
until the oil dissolves. Transfer the ether solution to the food in the dish, using 15 mI. of ether as
rinse fluid, mix the ether solution with the feed, and sieve the material twice. Remix any large particles
of corn with the main part of the feed. Spread the mixture in a thin layer on paper and allow it to stand
(2 hr.) until the ether evaporates. Store the mixture in a tightly capped tin container or amber bottle.
Obtain assignment to a thermostatically controlled electric [Link] which is located in a room
free from draft. The room should be dark except for a 15- to 25-watt light bulb suspended near the food
trough in each brooder compartment. Chicks will not eat properly if the illustration is inadequate.
Observe and record the temperature at intervals. If necessary, adjust the thermoregulator to maintain
the tempetature at 95 1 0
Obtain a group (8 to 10) of day-old chicks which have been starved or have been fed a basal ration containing little, or no, vitamin D since birth. Weigh the group of chicks and place them in a
brooder compartment. Fill the special water container and the special feeder. Refill these containers
as required but not less often than once a day. If necessary, clean the feeders before adding water or
feed. Remove and clean the dropping pans twice weekly. At the expiration of 7 days adjust the thermoregulator so that the temperature of the brooder is 90 1. After 3 days adjust the thermoregulator to
give a temperature of 85 10.
At the end of 21 days remove and weigh the group of chicks. Decapitate each chick (if desired,
after anesthetizing them in a closed container with ether). Remove and keep separate the tibiae from
the right and left legs. In preparing the tibiae for extraction, immerse the legs to a point above the hip
joint in boiling water for about 1 min. Remove the legs and defeather them. Lay each chick on its
back, bend the leg forward until the skin over the hip joint is taut, cut with a sharp knife through the
skin and flesh at the hip joint, and complete!y sever the leg at the hip joint. Take care' not to cut the
bony part of the tibiae.
Bend each leg backward at the knee joint until the skin is taut, cut the skin and halfway through
the flesh at the knee joint, and remove the skin from the tibia by pulling on the claws while holding
the tibia firmly. Place the tibiae in boiling water for 5 min. and dissect all the flesh from the bones,
being careful not to remove any of the cartilage attached at the end of the tibiae.
Wrap each tibia (from left and right legs) separately in a circ1~ of filter paper. Crush the ends of
each bone by a shilFP blow from a hammer. Tie all of the wrapped and crushed tibiae in a bundle labeled
approphately with india ink. Place the bundle in a Soxhlet apparatus (preferably one large enough to
hold several students' samples) fitted with an electric heating mantle and extract the bones for 20 hr.
with 95 per cent ethanol or isopropyl alcohol. Preserve the solvent and observe any material which
deposits on standing. Reextract the bones in the Soxhlet for 20 hr. with diethyl ether, remove the
bundles, and allow them to stand in air until the ether evaporates.
Label a No.2 porcelain crucible with a marking pencil, heat the crucible to redness, and continue
to heat at this temperature for 30 min., using a blue gas flame. Place the hot crucible in a -desiccator
(before closing. the lid completely allow the hot expanding air to escape for 30 sec. through a slight
opening) over anhydrous calcium chloride~ allow the crucible to cool to room temperature, and, as
rapidly' as pOSSible, weigh the crucible to 1 mg.
- 100-
Untie the bundle of bones, unwrap each bone, place all the bones in the weighed crucible, and
crush the bones in crucible, using a pestle. Avoid loss of material during these manipulations. Heat
, ... the bones in the crucible for 70 hr. at 100 in an electric oven. Allow the crucible to cool to room
temperature (at least 3 hr.) in a desiccator over anhydrous calcium chloride. Open the desiccator
slowly in a manner such that the inrushing air will not ;weep the powdered sample ftom the crucible.
Weigh the crucible to 1 mg., recording the weight found exactly 5 min. after the crucible has been
removed from the-desiccator.
Heat the crucible for 2 hr. at 800 (red heat) i~ a muffle furnace, allow the crucible to cool to room
temperature in a desiccator over anhydrous calcium chloride, and weigh the crucible to 1 mg., recording
the weight found exactly 3 min. after the crucible has been removed from the desiccator.
Plot the per cent bone ash found for the various levels of vitamin D (supplied by instruCtor) on cqordinate paper, label the ordinates, and draw a smooth standard curve. Interpolate the standard curve
to determine the number of units of vitamin D in the unknown oil.
Optional Procedure. Sever the upper joint (first exposed joint from the foot) of the middle toe of
the left leg (toes from right legs are to be treated similarly) by stretching the toe and cutting between
the ends of the bones. Be careful not to cut the cartilage. Wipe the middle toes of all the left legs
with a clean cloth and place them in a crucible prepared as described previously.
Heat the crucible for 70 hr. at 100, cool it in a desiccator for 3 hr., and weigh it as described.
Heat the crucible in a muffle furnace at 800 for 1 to 2 hr., cool the crucible in a desi_ccator as described, and weigh it to 1 mg. as described.
I
NOTES
L The authors are indebted to Dr. Sven Lassen, Van Camp Laboratories, Terminal Island, California, for
helpful suggestions.
2. Ground yellow corn, 58; wheat-flour middlings or wheat gray shorts, 25; crude domestic precipitated
casein, 12; precipitated calcium phosphate, 2; iodized salt (0.02 per cent potassium iodide), 1; nonirradiated
yeast (7 per cent minimum nitrogen), 1; and MnS04 '4H 20, 0.02 (Association of Official Agricultural Chemists,
1945).
A ration of the following composition has been used satisfactorily: ground yellow corn, 43; ground whole
wheat, 13; ground whole oats, 13; dried skim milk, 11; precipitated casein, 8; dehydrated alfalfa-leaf meal, 8;
steamed bone meal, 2.7; and U.S.P. sodium chloride, 1.1. During the experimental period (21 days) each chick
will consume (including spillage) from 350 to 500 g. of this ration.
EXPERIMENT 42
DETERMINATION OF THE ACTIVITY OF SAL/VARY AMYLASE AND
THE CONCENTRATION OF UNKNOWN STARCH SOLUTION
The chemistry, action, and function of amylases h~[Link] been reviewed by Hopkins (1946), Caldwell
and Adams (1946), Hanes (1937), and Tauber (1949). Amylases are enzymes which catalyze the hydrol..
ySis of starch and glycogen to dextrins, maltose, and glucose. They occur in nearly all animal~, seeds
and germinating sprouts, bacteria, molds and fungi, blood (plasma and leucocytes), and organs (especially the pancreas) of higher animals. In a few species [man, pig, rat, and mouse but not dog, cat,
sheep, and goat (Cohn and Brookes, 1936)] they are found in high concentration in the saliva.
The significance of wheat amylases in milling and baking technology has been discussed by
Geddes (1946). Wheat amylase was discovered by Kirchoff (Sumner and Somers, 1947) in 1811 and malt
amylase by Payen and Persoz (1833). a-Amylase (dextrinizing enzyme), the predominant form in bacteria, fungi, and animal fluids, rapidly cleaves starch to products (achrodextrin, maltose, and glucose)
which give no color with iodine. a-Amylase, liberated from the bound form during germination, effects
internal disintegration of starch by fission of the 1,4-glycosidic linkages. S-Amylase (saccharifying
enzyme), free in ungerminated cereal grains, is inactive toward raw starch but splits mildly degraded
starch to liberate terminal maltose units from the starch molecule. The amylase of sweet potatoes is
largely in the beta form (Balls, Walden, and Thompson, 1948).
The methods employed for the isolation of amylases depend upon the destruction of a-amylase
with acid and S-amylase by heat (Kneen and Beckord, 1946). Purified amylases have been prepared
from pancreas (van Thoai and Bernere-Silhol, 1946), soybean (Newton, H~xon, and Naylor, .1943), wheat
flour (BallolJ and Luck, 1941), sweet potatoes (Balls, Thompson, and Walden, 1946), saliva (Stark, 1942),
b~teria (Kneen and Beckord, 1946), and fungi (Aspergillus oryzae) (Volz and Caldwell, 1947) by fractionation of materials extracted with organic solvents. It has been proposed that crystalline a-amylase
is a "pure" protein with a molecular weight of 20,000 (Meyer, Fischer, and ,Bernfeld, 1947) or 45,000
(Caldwell, Booher, and Sherman, 1931). According to Williams, Schlenk, and Eppright (1944) a sample
of purified amylase contained per gram 4.1 mg. of inositol and lesser amounts of nicotinic acid, pantothenic acid, biotin, and folic acid. It has been reported that free amino groups, free -SH, and tyrosine
are essential for amylase activity (Caldwell, Weill, and Weil, 1945) and that amylase action is inhibited
by vitamin C (Seshagirirao and Giri, 1942) and proteins isolated from wheat (Militzer, Ikeda, and Kneen,
1946a, 1946b).
Methods for the quantitative determination of amylases are based on the time of hydrolysis to
substances which yield a blue or reddish color with iodine in potassium iodide (Redfern, 1947), the
quantity of maltose formed in a definite time (Willstatter, Waldschmidt-Leitz, and Hesse, 1923), the
decrease in viscosity of a starch substrate, and the total redUCing substances formed from starch
(Andersch, 1946; Somogyi, 1938). Blood amylases in diseases of the pancreas have been discussed by
Myers (1943).
Saccharogenic time has been defined as the time in minutes required at 40 to hydrolyze a buffered
(pH 6.8) solution containing 0.5 per cent (by volume) saliva, 0.25 per cent soluble starch, and 0.09 per
cent sodium chloride. According to Bauer and Martin (1948) saccharogenic activity is highest after
breakfast and lowest at night, decreases markedly in saliva which has stood for 1 hr. at room temperature, and decreases slowly thereafter up to 1 week. Saliva preserved with toluene decreases
little in potency under these conditions. Saccharogenic time varies from 1 to 6 min. for normal
individuals, but in individuals with skin disorders it may be as long as 1 .hr.
The activity of salivary amylase (ptyalin) and the concentration of an unknown starch solution ar~
- 102-
to be determined by an adaptation of the Wohlgemuth (1908) method. The time required to dige:;;t whole
starch solutions of known and un~nown concentrations, from the colloidal opalescent state (blue color
,......with iodine in potassium iodide) through the erythrodextrin stage (red color) to the achrodextrin end
point (no color), is to be measured. These data are to be used in making the required calculations.
EXPERIMENTAL
Obtain (from the instructor) two pieces of paraffin and two unknown starch solutions (prepared froln
intact, not soluble, potato starch). Shake these solutions well before removing aliquots.
Chew a piece of paraffin and expectorate into a dry filter paper. Collect about 2 mI. of the filtered
saliva. Pipette 1.00 mI. of the filtered saliva into a 100-ml. graduate, add water to the 20-ml. mark, and
mix the liquids thoroughly. Pipette into separate 6-in. test tubes (in such a manner that none of the
liquids comes into contact with the waUsof the tubes)S.O ml. of each uriknown starch solution and 5.0 mI.
of a standard 0.5 per cent starch solution. Add to each tube (lO-ml. graduate) 4.0 mI. of a sodium chloridephosphate buffer solution (pH 6.8). Pipette 1.00 mI. of the diluted saliva into the tube containing the 0.5
per cent standard starch solution. Mix the liquids quickly and immediately place the tube in a 40 water
bath. Record the time (the aid of another student is permitted). At I-min. intervals transfer (medicine
dropper) 1 drop of the mixture to a spot-plate depreSSion containing 2 drops of 0.01 M iodine in 0.01 M
potassium iodide sol~tion (this solution should be used within 1 min. after dispensing since it fails to
give a satisfactory color on longer standing). Near the end point (disappearance of red color) perform
this test at IS-sec. intervals. Record the time required to reach the end point. Repeat this procedure
while testing the unknown starch solutions simultaneously.
Calculate the dilution of saliva required to digest the 0.5 per cent starch solution, assuming an
inverse prop'ortionality between concentration of saliva and end-point time. Calculate the approximate
end-point times for the unknown starch solutions.
Collect 2 mI. (or larger volume up to 20 mI., if the amylase activity was found to be low) of the
filtered saliva, pipette 1.00 ml. into a lOa-mI. graduate, dilute the saliva to the desired (calculated)
volume with distilled water, and mix the liquids thoroughly. Pipette into separate 6-in. test tubes 5.0
ml. of the unknown starch solutions and standard starch solutions of the following concentrations: 0.20,
0.50, 1.0, 1.5, 2.0, and 2.5 per cent. Add 4.0 mI. (graduate) of the sodium chloride-phosphate buffer
solution to each tube. Pipette 1.00 ml. of the diluted saliva into the tube containing the 0.2 per cent
starch solution. Repeat the procedure described previously and record the end-point time.
Test each of the other starch solutions similarly and simultaneously. The color tests should be
made at the stated intervals, although it is unnecessary to begin testing a [Link] starch solution
until the digestion of that of the next lower concentration is complete. Plot-the data relating time and
starch concentration on coordinate paper, label the ordinates, and draw a smooth' curve. Estimate the
concentration of each unknown starch solution by interpolating the standard curve.
Determine the activity (units) of the amylase preparation by calculating the milliliters of 1 per
cent starch solution which would be digested to the achromic point in 30 min. at 40 by 1.0 ml. of
undiluted saliva.
- 103-
EXPEI-<IMENT 43
DETERMINATION OF ALKALINE PHOSPHATASE ACTIVITY IN BLOOD SERUM
The determination, functions, and clinical significance of phosphatases have been discussed by
Barker (1949), Roche (1946), Sunderman (1942), Bodansky and Jaffe (1934), Baur (1949), and the Paul
Lewis Laboratories (1948). Normal ranges of alkaline phosphatase per 100 mI. of serum are 1.5 to 4.0
Bodansky units for adults and 5.0 to 13.0 units for cl\ildren. Alkaline phosphatase activity,is redu~ed
in (asting and on fat-free diets. It is increased on high-fat diets, and values up to 190 units have been
observed in rickets, bone malignancy, obstructive jaundice, and other diseases. The phosphatase
activity of rat blood serum is approximately 80 Bodansky units per 100 mI. in the presence 01 magnesium
ions (added to activate the enzyme), and it is about 2S per cent lower in the absence of magnesium ions.
The urinary excretion of alkaline phosphatase by. men and women is scanty and irregular (Burgen, 1947).
The normal function of alkaline phosphatase may be to stimulate protein synthesis in the development of bone-marrow cells, teeth, hair, mammary gland, and epithelial cells of the smail intestine.
Since phosphatase activity is intense in embryonic tis~ue, it may be concerned with the formation of
nucleic acids and protein in the cytoplasm and the nucleus of cells. Nucleic acids may act as phosphate
donors to "funnel" ~nergy into the protein-synthesis mechanism during peptide-bond formation.
Th'e increase in alkaline phosphatase in blood serum in bone disease may result because of (a)
overproduction of the enzyme from the injured bone tissue or (b) the increased capacity of bone for
cellular activity in the absence of bone synthesis.
Acid phosphatase (optimum pH, 5.S) was first observed in the urine of subjects with carcinoma of
the prostate, and it is known to increase in blood serum with the onset of "letastases. Females of all
ages excrete about 50 units of acid phosphatase per day and males variable amounts up to 350 units per
day after puberty (Burgen, 1947; Kirk, 1948). The acid phosphatases of the prostate. gland and of the
red cell appear to be different enzymes (Abul-Fadl and King, 1949). Gottschalk (1945) has proposed a
micromethod for the determination of acid phosphatase.
The alkaline phosphatase of rat blood serum is to be determined by the method of Bodansky (1933),
as modified by Flock and Bollman (1948). Inorganic phosphate (corrected for preformed inorganic phosphate), liberated from 13- glycerophosphate under sta?dardized conditions, is to be estimated by the
photometric method of Fiske and Subbarow (1925), essentially as modified by King (1932). According to
Bodansky each phosph?te unit is equi~alent tothe "lib,eration of 1 mg. of phosphorus as phosphate ion
during the first hour of incubation at 37 'and at pH 8.6, with a substrate containing sodium B-glycerophosphate, hydrolysis not exceeding 10 per cent of the substrate."
EXPERIMENTALl
Place a rat and a wad of cotton saturated with ether under a bell jar. When the rat has been
place it on its back on a board and tie its legs to nails at the comers of the board. Pick
up the skin on the left ventral side adjacent to the heart, make a longitudinal incision with sharp
scissors through the skin and body wall, cut and remove a section of the body wall of a size sufficient
to expose the heart, and carefully remove the heart from the thoracic cavity, with as little damage as
possible to the connecting blood vessels. Hold the animal in such a position that its exposed heart is
directly over the mouth of a IS-mI. centrifuge tube, make an incision in the heart with scissors, and
collect as much blood as possible by pressing on the animal. Allow the blood to stand for 15 min.
Wrap the carcass in a paper towel and discard it in the waste container. Centrifuge the blood-clot
suspension, pour the clear ;;;upernatant serum into a test tube, and stopper the tube. Label the tube,
preserve the serum in the refrigerator, and retain serum for use in Exp. 46.
ane~thetized,
- 104-
Submit a labeled 6-in. test tube to the instructor, who will add 0.40 mI. of an unknown phosphate
solution. Transfer approximately 17 mi. of the sodium {3-glycerophosphate reagent (preserved,in the
refrigerator) and allow it to warm to 37 in a water bath. Add to the test tube 0.40 ml. of ~istilled
water, S~O mI. (lO-ml. graduate) of the warmed glycerophosphate reagent, and 0.100 ml: of the rat blood
serum (I-mi. graduated pipette). Immediately stopper the tube with a clean, dry rubber stopper, invert
the tube several times, and place the stoppered tube in a water bath at 37 o. Record the time. Carry out
the described procedure, adding 0.40 mi. of distilled water, 5.0 ml. of the glycerophosphate reagent, and
0.100 mI. of the serum to a second tube, and 0.50 mL of distilled water and 5.0 mi. of the glycerophosphate reagent to a third tube. During the incubation period, add to a fourth tube 0.50 mI. of the serum
and 5.0 mI. of distilled water for use in determining "preformed" phosphate.
After 60 min. incubation at 37 remove each tube in turn from the bath and add immediately to each
tube 4.5 mI. pf 10 per cent trichloroacetic acid to stop enzymatic action and to precipitate the blood
serum proteins. Stopper each tube, shake each tube vigorously, and filter each mixture on dry filter
paper of low ash content. Pipette 5.0 mI. of each filtrate into separate test, tubes calibrated to contain
10.D-ml. volumes (or into 10-ml. glass-stoppered graduated cylinders). The filtrates are stable if preserved in the refrigerator.
Pipette into separate calibrated test tubes (or 10-ml. graduated cylinders) 1.00 ml., 2.00 mI., and
3.00 mI., respectively, of a standard phosphate solution containing approximately 0.02 mg. per mI. of
phosphorus. Add to each tube (by means of a 10-ml. grad\lated pipette using the house vacuum or water
aspirator but not the mouth) 2.25 ml. of 10 per cent trichloroacetic acid and sufficient distilled water to
bring the volumE! to 5.0 mI. Add in turn to each of the seven tubes 1.0 ml. of 5 per cent ammonium
molybdate in 7 N sulfuric acid, 0.50 mI. of l,2,4-aminonaphtholsulfonic acid reage~t, and sufficient
distilled water to bring the volume to 10.0 mI. Stopper each tube, invert each tube several times, and
allow the tubes to stand for 10 min.
Using a visual colorimeter, determine the relative color intensity of each unknown solution compared with that of the standard (set at 20 mm.) of nearest color intensity. If any color is observed in the
control tube, compare its intensity with that of the most dilute standard (set at 10 mm.). If the intensity
is measurable, calculate the concentration of phosphorus liberated from the glycerophosphate reagent
and apply this correction to the unknown solutions.
Optional Procedure. Prepare the solutions described previously and, in addition, a solution containing 3.0 mI. of distilled water, 2.25 ml. of 10 per cent trichloroacetic acid, 1.0 mI. of the ammonium
molybdate reagent, 0.50 mI. of the aminonaphtholsuifonic acid reagent, and sufficient distilled water to
bring the volume to 10.0 mI. Adjust the photoelectric colorimeter [Lumetron. Model 400A, equipped with
a red (650-mp.) filter] to give an optical density reading of 0 with this solution. 4'JIeasure the color intensities of the unknown, standard, and blank, solutions. Plot the data from the standard, draw a smooth
curve with Optical Density as the ordinate and Phosphorus Concentration as [Link], and estimate
the values of the unknown and blank solutions by interpolating the standard curve. Correct the
phosphorus values of the unknowns {or the phosphorus liberated in the blank solution.
NOTE
1. If the total phosphorus approaches 60 mg. per cent, the determination must be repeated at lower concentration or shorter incubation time, since the enzyme is [Link] these conditions.
-105 -
EXPERIMENT 44
ASYMMETRIC ENZYMATIC SYNTHESIS OF ACETYL-L-P/fENYLALANINE [Link]
AND ISOLATION OF L- AND D-PHENYLALANINES
L- and D-phenylalanines have been prepared by (a) asymllletric enzymatic synthesis, with papain,
of acetyl-L-phenylalanine phenyl hydrazide (private communication from Dr. Carl Niemann, California
Institute of Technology, Pasadena, California) and of acetyl-L-phenylalanylglycine anilide (Behrens,
Doherty, and Bergmann, 1940), (b) asymmetric enzymatic hydrolysis, with chymotrypsin, of acetyl..Lphenylalanine methyl ester (private communication, Dr. Carl Niemann), (c) resolution of DL-phenylalanine through the brucine salts of the formyl derivatives (Fischer and Schoeller, 1907; du Vigneaud
and Meyer, 1932; du Vigneaud and Irish, 1938) and the cinchonine salts of the benzoyl derivatives
(Fischer and Mouneyrat, 1900), and (d) the destruction of Lphenylalanine in DL-phenylalanine by the
action of yeast (Ehrlich, 1908) and L-amino acid oxidase (Stumpf and Green, 1944).
Papain (Balls, Thompson, and Jones, 1940; Lewis a~d Woodward, 1948; Sumner and Somers, 1947)
is the dried albuminous exudate of the fruit of Carica p1lpaya L., a member of the Caricaceae family of
tropical palms native to Latin America and the West Indies. The latex is obtained by scarifying the
green fruit from trees which grow from 20 to 40 ft. high. About 5 lb. of fresh latex is required to give
lIb. of dried latex. Each tree produces from 3 to 8 oz. of dry latex per year, each acre of trees yields
from 80 to 175 lb. per year, and the production is maximum when the trees are 12 to 14 months old.
Dried papain was first made in 1878 by Henri Widmark of Berlin but was first produced commerCially in
Ceylon. At the present time papain is marketed from most of the tropical countries, but it ~omes principally from Ceylon and Tanganyika (British protectorate on the East African Coast).
In 1935 more than 300,000 lb. of papain was imported into the United States for the preparation of
protein hydrolysates, the tenderization of meat, and other uses. Commercial papain varies in color from
light yellow to chocolate brown, but there is no relation between color and activity. The proteolytic
activity of papain is inhibited or destroyed by oxidation and is increased by hydrogen sulfide, hydrogen
cyanide, glutathione, cysteine, and other reducing agents (Scott and Sandstrom, 1942). The relative
proteolytic activity of papain preparations is measured by determining the 'extent to which a protein
such as fibrin (Heyl, Caryl, and Staley, 1914), casein (Balls, Swenson, and Stuart, 1935) beef powder
(Maher and Wirth, 1946) or hemoglobin (Anson, 1931) is digested under stan9ard conditions. Crystalline
papain has been prepared as thin needles of molecuia! weight 27,000 2,000 (Balls and Lineweaver,
1939; Lineweaver and Schwimmer, 19,\1).
Acetyl-DL-phenylalanine, prepared by acetylation of DL-phenylalanine, is treated at pH 4.5 with
papain, cysteine, and toluidine. The acetyl-L-phenylalanine p-toluidide, which separates from the
solution as soon as it is formed, is hydrolyzed, and L-phenylalanine is isolated from the hydrolysate.
The unreacted acetyl-D-phenylalanine is hydrolyzed, and D-phenylalanine is isolated. The described
experimental directions are essentially those kindly provided by Dr. Carl Niemann.
EXPERIMENTAL
Acetyl-DL-phenylalanine. Add 10 g. (0.06 mole) of DL-phenylalanine (prepared in Exp. 8), 21 mI.
(0.13 mole) of 6.0 N sodium hydrOXide, and 40 mI. of distilled water to a 250-ml. conical flask. Stir the
mixture until the phenylalanine dissolves. Immerse the flask in an ice-water bath, cool the mixture to
5, add 17 mI. (0.18 mole) of acetic 'anhydride (99 to 100 per cent) from a dropping funnel at the rate of
about 1 mI. (20 drops) per minute while rotating the flask at frequent intervals. Allow the mixture to
stand for 1 hr. in an ice-water bath. Add sufficient concentrated hydrochloric acid (3 to 4 ml.) drop wise
- 106-
with stirring to reduce the pH from'" to 2, as determined with universal indicator paper~ Scra'tch the
sides of the vessel and allow the mixture (turbid if crystallization has begun) to stand for 1 hr. or longer
in the ice-water bath. Filter the suspension on a Buchner funnel, wash the crystals on the funnel twice
0
with 20-ml. portions of ice-cold distilled water) and dry the product overnight at 50 or in air
a covelled
evaporating dish. Weigh the product, retain 10 g., and submit the excess to the instructor in a tared,
labeled sample bottle.
Acetyl-L-phenylalanine p-Toluidide. Add 2 go of commercial papain powder and 20 mI. of distilled
water to a 125-ml. conical flask. Stopper the flask, immerse the flask in an ice-water bath, and shake it
vigorously at intervals for 30 min. Centrifuge the s,uspension 15 min. Add to a 250-mi. conical flask
10 g. (0.05 mole) of acetyl-DL-phenylalanine, 180 mI. of pH 4.5 buffer (acetic acid and sodium acetate)
solution, and 0.6 g. of L-cysteine hydrochloride. Stir and heat (not above 500 ) the mixture until the
solids dissolve. Add 5 g. (0.05 mole) of p-toluidine (m.p. 42 to 43)\and the nearly clear papain solution.
place the flask immediately
(to avoid precipiStopper the flask, shake it until the solid dissolves, and
o
.
tation of the acetyl-DL-phenyialanine) in an oven at 40. Crystallization of the acetyl-L-phenylalanine
p-toluidid~ usually begins within 1 hr. Allow the flask to stand at this temperature for at least 5 days,
immerse the flask for 20 min. in an ice-water bath while rotating it at intervals, filter the suspension on
a Buchner funnel, and wash the crystals twice on the funnel,with 25-ml. portions of distilled water.l
Suspend the crystals in 80 mI. of absolute ethanol, heat the mixture in a bOiling-water bath until
the crystals dissolve, and filter the hot solution through a heated Buchner funnel. Transfer the hot
solution to a beaker, cool the beaker for 30 min. in an ice-water bath (or allow it to stand overnight in
0
the refrigerator), filter the suspension on a Buchner funnel, and dry the precipitate for an hour at 50
or overnight in air. Weigh the product, retain 5 g., and submit the excess to the instructor in a tared,
labeled sample bottle.
L-Phenylalanine. 2 Add 5 g. of the acetyl-L-phenylalanine p-toluidide and 65 mI. of 6 N hydrochloricacid to a 200-ml. round-bottomed flask, attach a reflux condenser, and heat the mixture until the
solvent refluxes vigorously. Continue vigorous refluxing for 8 hr. Set up the reduced-pressure distillation apparatus shown in Fig. 7 (see Distillation in the Appendix) and distill the clear solution in vacuo.
Dissolve the residual material in 15 mI. of distilled water. Add 30 mI. of concentrated ammonium hydroxide and note that a copious precipitate of p-toluidine forms almost immediately. Cool the mixture
for 30 min. in an ice-water bath, filter the suspension, wash the crystals on the funnel with 5 ml. of
distilled water, and extract the combined filtrate and washing twice with 20-ml. portions of chloroform,
to remove the dissolved p-toluidine. Remove the excess ammonia by evaporating the aqueous phase to
about 30 mI., cool the residual liquid for 30 min. in an ice-water bath, filter the suspension, wash the
crystals of L-phenylalanine successively with 5-ml. portions of ice-cold distilled water and 95 per cent
o
ethanol, and dry the product at 50 or in air. Submit the product to the instruct?r in a tared, labeled
sample bottle.
.
2
D-Phenylalanine. Evaporate in vacuo the filtrate preserved for this purpose to half its volume.
add concentrated hydrochloric acid to pH 2 (tested with universal indicator paper), scratch the sides of
the vessel, and cool the mixture for 1 hr. in an ice-water bath. Filter the suspension and crystallize
the product (acetyl-D-phenylalanine3) from water. Hydrolyze the product by refluxing it with 6 N hydro;.
chloric acid. Isolate the D-phenylalanine. Submit the product to the instructor in a tared, labeled
sample bottle.
in
NOTES
1. Allow the filtrate to stand for severa! days until nearly all the L-phenyla1anine derivative has precipitated,
filter the suspension, and retain the precipitate for ,the isolation of L-pheny1alanine and the filtrate for the isolation
of D-phenylalanine.
2. According to Fisher and Schoeller (1907) the specific rotation in water of L-phenylalanine is
[ct]~ = -35.1 deg., and that of D-pheny1alanine is [(d~ = +35.0 deg.
3. According to du Vigneaud and Meyer (1932) the specific rotation of acetyl.D-phenylalanine in absolute
ethanol is (0:]26 ==-51 deg. Knoop and Blanco (1925) found (0:]]) =-51.8 deg.
])
-107 -
EXPERIMENT 45
DETERMINATION OF CHOLESTEROL IN [3LOOD SERUM
The chemistry and physiology of cholesterol have been discussed by Bills (1935), Bloor (1943),
and Peters and -Van Slyke (1946). Cholesterol was first isolated from gallstones by workers in tl~e
seventeenth century. It has been shown subsequently to be a constituent of bile, blood, intestinal concretions, cysts, eggs, tubercle bacilli, milk, and nurllerous other types of biological materials. There
is a close relationship between cholesterol and other sterols including the\bile acids, vitamins D2 and
D 3, male and female sex hormones, carcinogenic hydrocarbons, and hormones of the suprarenal cortex.
Nearly all the cholesterol of bile and red blood cells, and from 20 to 40 per cent of that in blood
plasma, is in the free form. The concentration of total cholesterol in human blood plasma ranges from
150 to 275 mg. per 100 ml. under normal conditions to as much as 800 mg. per 100 mI. in uncontrol1ed
diabetes mellitus, lipoid nephrosis, obstructive jaundice, hypothyroidism, and hemorrhagic and aplastic
anemias. Plasma cholesterol levels as low as 50 mg. per 100 mI. are associated with hyperthyroidism,
pernicious anemia, hemolytic jaundice, acute'liver diseases, infections, and malnutrition. Cholesterol
is excreted into the intestine, where it is converted by bacterial action into its saturated isomers
coprosterol and dihydrocholesterol. It has been suggested that cholesterol may be a precursor of vitamin D and that it may playa role in the absorption and transportation of fat, regulation of cell permeability and membrane equilibria, neutralization of bacterial toxins, and growth of normal and malignant
cells. Steiner (1948) has discussed the significance of cholesterol in coronary arteriosclerosis.
Free cholesterol is determined quantitatively by gravimetric [Link] its slightly soluble
digitonide and by colorimetric analysis of the colored complex formed with the Liebermann-Burchard
reagent (acetic anhydride and sulfuric acid). Total cholesterol is determined similarly after saponification of cholesterol esters. Procedures for the quantitative determination' of cholesterol and data on
the cholesterol content of blood and other tissues have been reported by Bloor (1943), Clarke and
Marney (1945), Foldes and Murphy (1946a, 1946b), Hepburn and Kotlikoff (1943), Hodges, Sperry, and
Anderson (1943), Hoffmeyer (1944), Lamb, Mueller, and Beach (1946), Leiboff (1942), Levin (1945)
Marquardt (1947), Popjak (1943, 1946), Saifer and Kammerer (1946), Sobel and Mayer (1945), Teed
(1944), and other workers.
Total cholesterol (free cholesterol and cholesterol esters) is to be determined in rat blood serum.
by photometric. analysis of the colored,complex formed with the Liebermann-Burchard reagent, essentiall~
by the procedure of Kingsley and Schat(ert (1949). According to these authors the color intensity of
hydrolyzed serum is about 84 per cent that of unhydrolyzed serum, owil1g to the greater color intensity of
cholesterol esters compared with free cholesterol. Monn\er, Farchadi, and Maulbetsch (1941) found that
the cholesterol content of rat blood serum averages 80 mg. per cent.
EXPERIMENTAL
Obtain the rat blood serum preserved in Exp. 43. Soften six No. 7 corks with a rotary press and
shake them for 15 min. with 75 ml. of chloroform in a stoppered flask. Submit a clea.n, dry, labeled
6-in. pyrex test tube to the instructor, who will add 10.0 mI. of a chloroform solution containing a known
quantity of pure cholesterol. Add (I-mI. graduated pip,ette) to this recovery solution 0.20 ml. of the
serum and 0.20 mI. of distilled water. Add to a second tube 10.0 mI. of anhydrous C.P. chloroform and
0.40 mi. of the serum. Add to a third tube 10.0 mI. of the chloroform and 0.40 mI. of distilled water. If
a check on the completeness of the extraction procedure is desired, set up an additional tube containing
7.0 mI. of chloroform, 0.40 mI. of distilled water, and 3.0 ml. of the standard cholesterol solution.
-108 -
Obtain assignment to a Kahn shaker and a centrifuge before proceeding with the next steps, which
must be carried out without long (more than 1 or 2 min.) delay. Add to each tube in turn 1.5 g. of powdered
anhydrpus magnesium sulfate. Stopper each tube and shake it (by hand) for about 10 sec. to prevent the
salt from caking. Agitate the tubes in the shaker for S min., add O.S g. of fuller's earth to each tube,
'shake the (stoppered) tubes for 10 sec. by hand, and centrifuge the tubes at 3,000 r.p.m. for 10 min.
Obtain assignment to a photoelectric colorimeter. Cautiously transfer (graduate) 16.0 mI. of cold
C.P. acetic anhydride (preserved in the refrigerator) to a 12S~ml. conical flask and add 4.0 mI. of cold
C.P. concentrated sulfuric acid (stored in the refrigerator). Stopper the flask, immerse it in an ic~
water bath, and (Cautiously) rotate the flask to mix the solutions. Pipette into separate clean, dry,
6-in. test tubes S.O-ml. portions of the serum, the blank solution, and the serum plus recovery solution.
A':dd to each tube (by means of a l~ml. graduated pipette, using the house vacuum or the water aspirator
but not the mouth) 2.0 mI. of the cold, freshly prepared acetic anhydride-sulfuric acid reagent. Stir
each mixtme with a clean, dry stirring rod. Stopper the tubes and allow them to stand for 10 min. in a
water bath at 20 0.5.
Transfer the solutions to matched test tubes and determine (within 15 min.) the optical densities,
using a photoelectric colorimeter [Lumetron, Model 400A, equipped with a r~d (6S~m(l) filter] adjusted
so that the optical density reading of the blank is zero. While the colors are developing, pipette 0,
1.00, 2.00, 3.00, and 5.00 mI. of the standard (0.12 mg. per ml,) choiesterol solution into separate test
tubes. Dilute each solution to 5.0 mI. with chloroform, add 2 mI. of the freshly prepared acetic anhydridesulfuric acid reagent, stir each mixture, stopper the tubes, and immerse the tubes for 10 min. in a water
bath at 20 0.5. Determine the optical densities with a photoelectric colorimeter adjusted as described.
Plot the dataobtained with the standard solutions using Optical Density as the ordinate and
Cholesterol Concentration as the abscissa, draw a smooth curve, and estimate the values of the unknown solutions by interpolating the standard curve.
EXPERIMENT 46
DETERMINATION OF NITROGEN COMPOUNDS IN URINE
This topic has been reviewed by Hawk, Oser, and Summerson (1947), Peters and Van Slyke (1946),
Harrison (1947), Todd and Sanford (1948), Cantarow and Trumper (1945), Kleiner (1945), Marshall (1926),
and Bodansky and Bodansky (1940).
Normal and abnormal constituents in urine include (a) inorganic cations: ammonium, sodium, potassium, calcium, and magnesium; (b) inorganic anions: carbonate, chloride, sulfate, phosphate, and nitrate;
(c) organic acids: oxalic, ethereal sulfuric, hippuric, benzoic, lactic, acetoacetic, ,B-hydroxybutyric,
phenaceturic, homogentisic, phenylpyruvic, phenyllactic, p-hydroxyphenylpyruvic, and urocanic acids;
(d) nitrogen compounds: amino acids, urea, creatinine, creatine, allantoin, uric acid, purines, pyrimidines,
peptides, and proteins; (e) carbohydrates: glucose, fructose, galactose, and pentoses; and (f) miscellaneous: taurine, thiocyanates, glutathione, ergothioneine, acetone, fat, blood, bile, pus, pigments,
hormones, enzymes, and vitamins.
Urine is formed by filtration of the diffusible constituents of plasma through the glomerulus. Water,
chloride, bicarbonate, potassium, phosphate, uric acid, urea, sulfates, and glucose are partially reabsod>ed as the filtrate passes along the tubules. Chloride is preferentially reabsorbed in the distal tubule
and glucose in the proximal tubule. Although the total volume of the glomerular filtrate may be as much
as 150 liters per 24 hr., only from 1 to 2 liters of ur,ine are excreted. The volume of urine is decreased
by high temperature, physical exercise, acute nephritis, diarrhea, diseases of the heart and lungs, and
other disorders. It is increased in diabetes mellitus, diabetes inSipidus, and degeneration of the kidney
and by drugs such as calomel, digitalis, acetates, and salicylates.
Normally urine is colored amber yellow, although it may be nearly colorless, milky, dark yellow,
brownish red, brownish black, greenish yellow, or orange in pathological condition:;. Urine which is
clear when voided may become turbid on standing, owing to the formation of agglomerates (nucleoprotein
and epithelial cells) or the precipitation of phosphates. The odor normally is faintly aromatic, but it
may become ammoniacal because of bacterial decomposition. The pH varies from 5 to 8 depending upon
the relative amounts of monohydrogen and dihydrogen phosphates excreted by the kidney as required to
maintain a nearly constant (about 7.4) pH of the blood. The pH, as well as the total acid, of urine
varies with the time of day, and usually acidity is iI\creased in pathological states. Alkaline urine
results from the ingestion of sodium bicarbonate or alkali-forming salts of organic acids (such as
tartaric and citric acids) found in citrus juices and other foods. Acid urines are excreted following the
ingestion of acid-forming foods such as bread, cereals, and m~at. The specific gravity of urine usually
varies from 1.015 to 1.025, although it may fall to about 1.00 after excessive water ingestion and may
rise to 1.04 in chronic nephritis or after excessive sweating.
The average values found for the volume, specific gravity, total nitrogen, urea nitrogen, ammonia
nitrogen, uric acid nitrogen, creatinine nitrogen, and undetermined nitrogen of 400 urine specimens
were obtained by Beard (1935) and are shown in the accompanying table. According to Schmidt and Allen
(1938) total-nitrogen values may fall as low as 3.6 g. for low-protein diets and rise as high as 23.0 g.
on high-protein diets. It has been observed that creatinine and uric acid excretion is practically constant on diets varying widely in amounts of nitrogenous components.
Proteins, redUCing substances, and acetone are to be determined qualitatively and total nitrogen,
ammonia nitrogen, urea nitrogen, creatinine nitrogen, and uric acid nitrogen are to be determined quantitatively in 24-hr. urine samples collected on low- and high-protein diets. Knowledge of the composition
of urine aids in understanding the metabolic processes of the body and is indisp~nsable for clinical
purposes. For the latter purpose it is also ne~essary to have complete information on the subject's diet
and clinical history.
-110 -
Volume, mI
Specific gravity... ~ .
Total nitrogen, g ....
Urea nitrogen, g. / . . . . . . . . .
Ammonia nitrogen, g. . . . . . . .
Uric acid nitrogen, g ...
Creatinine nitrogen, g
Undetermined nitrogen, g
11
Low
High
protein
protein
1,472
1.023
1,408
1.019
15.3
13.2
0.68
0.21
0.66
0.55
Normal
1,364
1.022
11.2
g.S
0.54
0.19
8.0
6.2
0.43
0.18
0.62
0.63
0.57
0.34
EXPERIMENTAL
Select from Tables of Food Composition (see Experiments in the Appendix) a high- or low-protein
approved by the instructor. Restrict the food intake to this diet for 2 days prior to and during the
collection of a 24-hr. urine sample. In collecting the urine, empty the bladder at 6 or 7 A.M. and discard
this urine. Collect all urine voided during the succeeding 24 hr. in a clean 2.S-liter screw-capped amber
bottle containing 10 mI. of toluene as preservative. Measure the total volume (graduate), dilute the urine
to the 2-liter mark with distilled water, and thoroughly mix the liquids. Transfer about 200 mI. of the
diluted urine to a glass-stoppered bottle, add S mI. of toluene, label the bottle, and store it in the refrigerator. Discard the remainder of the original urine. Pipette two 25.O-ml. portions of the diluted urine
(free from toluene) into separate SO-ml. volumetric flasks. Label these flasks, one for qualitative and
one for quantitative recovery, and submit them to the instructor at the beginning of the laboratory period.
The instructor will add known quantities of the pure substances to be determined. Tsansfer these solutions to glass-stoppered bottles, label, add S ml. of toluene to each bottle, and store in the refrigerator.
~t
QUALITATIVE TESTS
1
Proteins. 1. Coagulation Test. Transfer about S mI. of the urine to a test tube and heat the
upper third of the solution to boiling. Add 5 drops of N acetic acid and mix the solutions thoroughly.
A precipitate which forms on heating but which dissolves when the urine is acidified with acetic acid
is calcium Of magnesium phosphate or a mixture of these substances. A flocculent precipitate which
does not dissolye in the acetic acid solution is protein, usually albumin.
2. Heller's Nitric Acid Ring Test. Place 5 mI. of concentrated nitric acid in a test tube. Incline
the tube and pour S mI. of the urine carefully down the side of the tube. Place the tube in a vertical
position and allow it to stand for about 1 min. A ring of white precipitate appearing at the interface of
layers usually indicates a protein.
Reducing Substances. 2 If the test for protein was positive, add 5 drops of 6 N acetic acid to 5 ml.
of urine in a test tube and heat the mixture to boiling. Filter the suspended coagulum and add S drops
of 6 N sodium hydroxide to the filtrate. }-Ieat 5 mI. of Benedict's qualitative reagent (cupric sulfate,
sodium sulfate, and sodium carbonate) to boiling in a test tube, add 0.5 mI. of the protein-free urine
filtrate, and heat the mixture for S min. in a boiling-water bath. Remove the tube from the bath and
"bserve any color or precipitQte. which may form immediately and after 5 min.
OH
Acetone. 3 Behre's (1928) Salicylaldehyde Test. Add 3
drops of a 10 per cent solution of salicylaldehyde ip 95 percent ethanol to 5.0 mI. of the urine. Add a 1/2-in. piece of
CH=CH
solid sodium hydroxide. A deep-red color (30 min.) indicates
/
O=G
the presence of dihydroxybepzalacetone (formula at right).
\-()\
-b
f \
CH=CH
~_
OH
-lli -
QUANTITATIVE TESTS 4
I
"\
Total Nitrogen.s Pipette 5.O-ml. aliquots of the urine and the urine recovery samples into separate
10~ml. volumetric flasks, add distilled water to the marks, and mix the liquids thoroughly. Pipette
1.00 mi. of eaci). urine sample into separate 8-in. pyrex test tubes. 'Pipette (using the house vacuum or
water aspirator but not the mouth) 1.0 ml. of 18 N sulfuric acid (S-ml. graduated pipette) into each tube.
Heat the tubes with micro burners (Tirrill burner with barrel top removed) until dense white fumes of
sulfur trioxide fill the tubes. Cooi each tube for 1 min., add 2 drops of 30 per cent hydrogen pe!oxide
from a dropping bottle, and heat until white fumes appear. If the solution is discolored (brown or yellow),
add 1 drop of 30 per cent hydrogen perOXide and reheat the solution until white fumes appear. If necessary, repeat this procedure until the solution is colorless. Heat each solution to boiling and continue
gentle boiling for 5 min. Cool each .solution to room temperature.
Transfer each solution quantitatively to separate 10~ml. volumetric flasks, add sufficient distilled
water to bring the total volumes to 75 mI., and pipette (using the house vacuum or water aspirator but
not the mouth) 15.0 mI. (25-ml. graduated pipette) of Nessler reagent 6 (potassium mercuric iodide and
sodium hydroxide) into each flask.
Pipette into separate 10~ml. volumetric flasks 5.0 mI. and 10.0 ml. of a standard ammonium
sulfate solution containing 0.080 mg. of nitrogen per milliliter. Pipette into each flask 1.0 mI. of 18 N
sulfuric acid (see precaution described preViously), add sufficient distilled water to bring the volume of
solution to 75 mI., and while rotating the flask, [Link] each flask (see precaution described preViously) 15.0 mI. of Nessler reagent. Add distilied water to the mark and mix the solutions in each flask
thoroughly. Using a visual colorimeter, compare the color intensity of each unknown solution with that
of the standard solution which matches it more closely.
Ammonia. 7 Pipette 10.~ml. aliquots of the urine and the urine recovery samples into separate
10~ml. volumetric flasks, add distilled water to the mark, and mix the liquids thoroughly. Transfer
2.~g. portions of dry, activated Permutit 8 (treated with 0.3 N acetic acid, washed with water, and dried
in air) to separate clean, dry 10~ml. volumetric flasks. Pipette 10.0 mI. of each diluted urine into one
of the flasks in such a manner that it falls directly on the Permutit without coming in contact with the
w,alls of the flask. Rotate the suspension gently but continuously for 5 min. Allow the suspension to
settle for 1 min. Tie a piece of filter paper over the tip of a 2.00-ml. pipette and pipette 2.00 mI. of the
supernatant liquid into a 25~ml. volumetric flask which has been rinsed with two l~ml. portions of
0.1 M potassium iodide solution and finally with distilled water, to remove any traces of mercur,y compounds iphibitory to the action of urease. Preserve these solutions for the determination of urea.
Add 25 mI. of distilled water to each Qf the volumetric flasks containing the Permutit, rotate the
flask gently for about 10 sec., allow the suspension to settle for 1 min., and decant as much supernatant
liquid as possible without disturbing the Permutit. Repeat the described procedure using a second
'25-ml. portion of distilled water. Add 10 ml.?6f 0.5 N sodium hydroxide (giaduated pipette), rotate the
flask for 20 sec., and add distilled water to bring the volume to about 75 mI.
Pipette into separate 100-ml. volumetric flasks 5.0 ml. and 10.0 mI. of a standard ammonium sulfate solution contatining 0.080 mg. of nitrogen per milliliter. Add 10 ml. of 0.5 N sodium hydroxide to
each flask (graduated pipette) and sufficient distilled water to bring the total volume to 75 mI. While
rotating each flask (unknown and standard) in turn, add (graduated pipette, using described technique)
10.0 mi. of Nessler reagent. Immediately dilute the solution in each flask to the mark with distilled
water, mix the solutions in each flask thoroughly, and ~lowthe flasks to stand for 10 min. or longer.
Compare the color intensities in a visual colorimeter, using the standard solution which matches the
unknown more closely.
Urea. 9 Place one commercial urease tablet (containing phosphate buffer salts) on a sheet of smooth
paper, crush the tablet, and transfer the powder to one of the 250-ml. volumetric flasks (prepared as described, under Ammonia) containing 2.00 mI. of the supernatant liquid preserved for the determination of
urea. Add a (powdered) urease tablet fo the second 25~ml. flask (15 to 20 mg. of urease powder and
1 ml. of pH 6.8 phosphate buffer may be used instead), and heat both flasks for 5 min. at 45 to 50 in a
water bath. Remove the flasks, add 200 mI. of distilled water, and mix the solutions thoroughly.
\
<
-112 -
Pipette into separate 250-ml. volumetric flasks 5.0 mI. and 10.0 mI. of a standarcfammonium
sulfate solution containing 0.25 mg. of nitrogen per milliliter. Add 200 mI. of distilled water to f1ach
flask and mix the solutions in each flask thoroughly. Add (graduated pipette with described precautions)
"'15.0 mI. of Nessler reagent to each flask (while rotating the flask), immediately dilute the solution in
each flask to the mark with distilled water, and mix the solutions in each flask thoroughly. Compare
the color intensities, using a visual colorimeter and the standard solution which matches the unknown
more closely.
Creati!1ine. lO Pipette 1.00-ml. aliquots of the urine and the urine recovery samples into separate
100-ml. volumetric flasks. Pipette into separate 100-ml. volumetric flasks 1.00 mI. and 2.00 mI. of a
standard creatinine ,s01ution containing 0.50 mg. of creatinine per milliliter. Pipette into each flask
20.0 mI. of a saturated solution of purified picric acid and 1.0 ml. of 3.0 N sodium hydroxide. Rotate
each flask until the solutions are mixed and allow the flasks to stand for 10 min. Add distilled water
to the mark, mix [Link] in each flask thoroughly, and compare the color intensities, using the
standard solution which matches the 'unknown more closely.
Uric Acid. ll Pipette 2.00-ml. aliquots of the urine'lind the urine recovery samples into ;;eparate
100-ml. volumetric flasks, add distilled water to the mark, and mix the solutions in each flask thoroughly.
Pipette into separate 50-ml. volumetric flasks 5.00 mI. of each diluted (1:50) urine sample and 5.00 mI.,
10.0 mI., and 15.0 ml. of a standard (freshly prepared) uric acid solution containing 0.0040 mg. of uric
acid per milliliter. (Caution: The sodium cyanide solution and the uric acid reagent described below
are very poisonous. These solutions should never be pipetted by mouth. Discard them directly into the
drain, rinse the container thoroughly, and flush the drain with running water.) Transfer (graduate) into
each of the flasks 10 mI. of a 12 per cent (freshly prepared) solution of sodium cyanide, 10 ml. of a SO
per cent solution of urea, and 8 mI. of uric acid reagent (phosphotungstic acid). Allow the flasks to
stand for 20 min., add distilled water to the mark, mix the solutions in each flask thoroughly, and compare the color intensities of the solutions, using the standard which matches each unknown most closely.
NOTES
1. Other protein-precipitation tests include those of Roberts (nitric aci~ and magnesium sulfate), Exton and
Osgood-Haskins (sulfosalicylic acid and sodium sulfate) and Purdy (acetic acid and sodium chloride). Relative
quantities of albumin observed clinically are estimated as trace (visible ring or cloudiness), small (distinct granular cloud which settles to about one-tenth the height of the urine column and is equivalent to about 0.1 per cent
albumin), moderate (dense cloud with marked flocculation indicating 0.2 to 0.3 per cent albumin), and large (heavy,
curdy precipitate tending to solidify and indicative of more than 0.5 per cent albumin). '.
2. Other tests include (a) reduction of copper (Fehling and Haines), bismuth (Nylander), mercury, iron, and
silver ions in alkaline solution, (b) formation of a crystalline osazone with phenylhydr~ine (Kowarsky-Blumel),
(c) optical rotation, and (d) fermentation with yeast. Reducing substances are recorded clinically as absent (no
precipitate and original blue color unchanged), reducing substances equivalent to 0.1 to 0.5 per cent glucose
(solution green or greenish opalescent), and reducing substances equivalent to 1 to 2 per cent glucose (orangebrown precipitate and blue supernatant fluid or heavy, bright-red precipitate with faintly blue supernatant liquid).
3. Acetone appears in the urine in diabetes, fevers, gastrointestinal disturbances, nervous disorders,
eclampsia, pernicious vomiting of pregnancy, and after chloroform anesthesia. Other tests are those of Liebin
(iodoform) and Rothera and Lange (nitroprusside). Acetone is formed from acetoacetic acid and ,8-hydroxybutyric
acid, primary products of fatty-acid oxidation. Acetoacetic acid is detected by Gerhardt's (ferric chloride) and
Lindemann's (iodine) tests and [Link] acid by the Black and Bart (oxidation to acetone by hydrogen
peroxide) and Osterberg and Helmotz (1934) (sodium nitroprusside and ammonium hydroxide) tests.
4. The quantitative procedures described belcw are applied as required in clinIcal laboratories:
Acetone. Acetone is determined by iodometric titration (Behre and Benedict, 1926), gravimetric (Denig~s,
1898a, 1898b) or iodometric (Van Slyke, 1917, 1929a) analysis of its slightly soluble basic mercuric ,sulfate complex, and colorimetric determination of its salicylaldehyde complex (Behre and Benedict, 1926; Behre, 1940;
Sumner, 1921). About 20 mg. of total acetone bodies is excreted daily by the normal adult. In severe diabetes the
daily excretion of acetoacetic acid and acetone may be as much as 6 g. and of ,[Link] acid 100 g.
Allantoin. A1lanto,in is determined by colorimetric analysis of the complex formed ,with Folin's ammoniacal
copper and phosphomolybdate reagent (Larson, 1932; Young and Conway, 1942; Young, MacPherson, Wentworth,
and Hawkins, 1944). Allantoin is the principal purine excreted by all mammals except man, the anthropoid ape,
and the Dalmatian coach dog.
113 _
Amino Acids. Most of the amino acids have been determined in urine by microbiological assay (Dunn,
Carnien, Shankman, and Block, 1947; Eckhart, Cooper, Fatoon, and Davidson, 1948; Frankl and Dunn, 1947; Woo~
son, Hier, Solomon, and Bergheim, 1948; Dunn, Akawie, Yeh, and Martin; 1950; Wheeler and Gyorgy, 1948; Yeh,
Frankl, Dunn, Parker, Hughes, and GyCitgy, 1947; Frankl, Martin, aljld'Dunn, 1947). Studies have been made of the
renal clearance of essential amino acids and urinary excrC'ltion of methionine and other amino acids in liver disease
and in cystinuria. The normal 24-hr. excretion varies from 1 mg. oc. apparent free aspartic acid to 190 mg. of apparent
free histidine and from about 10 mg. of total (free plus combined) methionine, to 350 mg. of total glutamic acid.
Total amino acids in urine have been determined by manometric measurement of the carbon dioxide formed
from reaction with ninhydrin (Hlimilton and Van Slyke, 1943; Van Slyke, Dillon, MacFadyen, and Hamilton, 1941;
Van Slyke and Folch, 1940; Van Slyke, MacFadyen, and Hamilton, 1943), formol titration (Henriques and Soreq_sen,
1910; Nathrop, 1926), colorimetric determination of the naphthoquinone sulfonic acid complex (Folin, 1922a, 1922b),
and iodometric determination of copper in the filtrate separAted from the precipitate formed with copper phosphate
(Albanese and Irby, 1944). The daily excretion of alpha-amino nitrogen is 100 to 150 mg. or about 1 per cent of the total
nitrogen. The excretion is increased in typhoid fever, acidosis, atrophy of the liver, and other wasting diseases.
Hippuric Acid. Hippuric acid is determined by extracting urine with ether, destroying urea in the extract with
bromine and hypobromite, and estim ating nitrogen in the residual material byKjeldahl analysis (Griffith, 1926; Q}lick, 1926).
Quick (1940)has employed hippuric acid excretion as a test of liver function and a measure of hepati_c damage. The
average daily excretion of hippuric acid normally is about 0.7 g. The excretion is increased after ingesting plums,
prunes, cranberries, and other foods containing hippuric acid or its precurl?Q's.
Lactic Acid. Lactic [Link] determined by oxidizing it to acetaldehyde, forming acetaldehyde bisulfite, and
titrating the bound bisulfite iodometrically (Friedemann and Graeser, 1933). The normal daily excretion (about 10
mg.) of lactic acid is increased after exercise and in pneumonia, eclampsia, atrophy, and other pathological conditions where tissue oxidation is diminished.
Organic Acids. Urine, freed from carbonates and phosphates, is titrated to pH 2.7 (Van Slyke and Palmer,
1920). The normal 24-hr. excretion, equivalent to 400 to 750 mI. of 0.1 N acid, is increased markedly in diabetes.
Protein. Protein is determined by biuret (Hiller, Greif, and Beckman, 1948) or Kjeldahl (Folin and Denis,
1914) analysis of the precipitate formed with trichloroacetic acid or heat, by measuring the volume of the protein
precipitated with Esbach's reagent (picric acid and citric:: acid) or with the reagent of Shevky and Stafford (1923)
(alcoholic solution of phosphotungstic acid and hydrochlaic acid), and by estimating the concentration of the
suspension formed with sulfosalicylic acid (Kingsbury, Clark, Williams, and Post, 1926). Although albumin is not
detectable, normal urine probably contains traces of this protein. Clinically, albumin values must be carelated
with other diagnostic tests.
Purines. Uric acid and other purines are precipitated as their copper salts, uric acid is separated as its
hydrochloride, and nitrogen in uric acid and remaining purines is determined by the Kjeldahl method (Hitchings
and Fiske, 1941). The daily excretion of purines, normally 16 to 60 mg., is increased in leukemia.
Reducing Substances. Reducing substances are dMermined by oxidation with ,cUPric ion and determination
of excess cupric ion through formation of white cuprous thiocyanate (Benedict, 1911), iodometric analysis (Shaffer
and Hartmann, 1921; Shaffer and Somogyi, 1933); and colorimetric analysis of the .colored complex formed with
dinitrosalicylic acid (Sumner, 1921, 1925) or phosphomolylSdic acid (Folin and Berglund, 1922; Folin and Svedberg,
1926; Benedict, 1931). Other methods include oxid,ation with ferricyanide and iodometric determination of the
excess of this reagent (Hagedorn and Jensen, 1923a, 1923b; Hanes, 1929), time required to decolorize a standard
ferricyanide solution (Hawkins, 1929; Hawkms and Van Slyke, 1929: Hoffman, 1937), and colorimetric analysis of
the picric' acid complex (Benedict, 1918; Benedict and Osterberg, 1918; Lewis and Benedict, 1915).
Relatively high concentrations of reducing substances (mainly glucose) are found in pathological conditions
(diabetes), especially following the administration of a standard dose (SO g.) of pure glucose. Concentrations may
rise as high as 10 per cent in severe diabetes.
Titratable Acidity. The urine is titrated to pH 8.5 or 9.0 by F9lin's (1903) method or to pH 7.4 by the procedure of Henderson and Palmer (1914). Although the titratable acidity varies widely, since it is dependent upon
the diet, the average 24-hr. excretion is equivalent to ab<>ut 350 mt.. of 0.1 N acid.
Miscellaneous. Procedures are available for the quantitative determination of indican, phenols, urobilinogen,
oxalic acid, sulfur, ethereal sulfates, inorganic sulfate, phosphate, chloride, fixed bases (sodium, potassium,
calcium, and magnesium), iron, iodine, heavy metals (lead, mercury, arsenic, and zinc), inulin, citric acid, vitamin
C, the B-complex vitamins, and numerous other constituents of normal and pathological urines.
5. The procedure to be used in this experiment is essentially that of Koch and McMeekin (1924). Nitrogen
may be determined by other methods, including those described in Exps. 9 and 20. Urea contributes about 80 per
cent of the daily nitrogen excretion, which averages 12 to 18 g. Relatively little nitrogen is excreted in "the feces.
Nitrogen equilibrium (nitrogen balance) is defined as the state of a normal adult in which the nitrogen of the ingested food and that excreted are approximately equaL
-114 -
6. The reaction of Nessler reagent with ammonia may be repJ:.esented as NHPH + 2HgI.. -- +SOH - -+ NHg 21 +
71- + 4H 20. The colored product, present in the colloidal state, is precipitated at higher concentrations of ammQnia
(0.1 N or greater). Chloride ion interferes with color formation.
7. The procedure to be employed is essentially that of Folin and Bell (1917). Ammonia may be determined
by aerating it into boric acid and titrating the mixture with standard acid using ,bromcresol green indicator (Van
'Slyke and Cullen, 1914). The average daily output of atn,nlonia is about 0.7 g., which is equivalent to 2.5 to 4.5
per cent of the total nitrogen. Ammonia excretion is increased in acidosis of starvation or diabetes and by the in.
"
ingestion of acids or acid-forming foods. It is decreased in alkalosis and by the ingestion of alkali-forming foods.
Amino acids and glutamine are probable precursors of ammonia.
8. Permutit is a synthetic aluminum silicate (zeotite) of the approximate composition (2Si0 2A1 2 03Na20 6H20).
Ammonium ion is selectively adsorbed by ion exchange rteaction on the Permutit displacing an equivalent amount of
sodium (or hydrogen) ion. At pH 7 ammonium ion is adsorbed by Permutit in the presence of sodium ion and in
alkaline solution ammonium ion is displaced by sodium ion. Permutit may be reactivated by treatment with acetiq acid.
9. The procedure to be employed is essentially that of Folin and Youngberg(1919). Ammonia may be determined by aerating it into boric acid and back-titrating with standard sulfuric acid (Van Slyke and Cullen, 1914)
or by allowing the ammonia to diffuse into standard acid in an adjoining compartment of the Conway micro-diffusion
apparatus and determining the ammonia by titrating the solution or by colorimetry (Cor/way, 1933; Conway and
Byrne, 1933). The average daily excretion of ureais 25 to 35 g., which is 80 to 90 per cent of the total nitrogen.
Urea excretion is increased in fevers and is decreased in kidney and liver diseases.
10. The procedure tp be employed is essentially that of Folin (1905, 1914). ,Creatinine may be determined by
photometric analysis of its red-colored picrate (Greenwald, 1925, 1928; Greenwald and Gross, 1924; Bonsnes and
Taussky, 1945; Peters, 1942) or 3,5-dinitrobenzoate (Benedict and Bebre, 1936; Langley and Evans, 1936). The
average excretion of creatinine is 1.0 to 1.8 g. daily, although considerably higher values have been reported
(Hobson,1939). Except for dietary fluctuations of creatinine intake, the daily creatirune excretion is nearly constant
for a given individual. Most of the body creatinine is found in muscle tissues. Creatinine appears to be a waste
product not utilized by the body. Creatinine excretion is increased in fevers and decreased in muscular disordersl
11. The procedure to be employed is essentially thjlt of Falin (1933, 1934). Uric acid may be determined by
photometric analysis of the colored complex formed with arsenophosphotungstic acid and sodium cyanide (Benedi<;:t
and Franke, 1922; Buchanan, Block, and Christman, 1945; Christman and Ravwitch, 1932), phosphotungstic acid
and alkali (Benedict, 1915; Benedict and llitchcock, 1915; Folin and Denis, 1913a, 1913b; Folin and Wu, 1919), or
2,6-dichloroquinone chloroimide (Fearon, 1944). Other methods include precipitation of ammonium urate and
titration of the latter with standard permanganate solutioJ). (Folin and Shaffer; quoted by Mathews, 1925).
- 11.c; _
APPENDIX
Topic
Page
Apparatus.
118
General
118
Individual
........................... .
118
119
Special .
. ..............
122
Instructions to Students . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
122
Suggestions to Instructor . . .
122
First Aid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
128
. , ............................. .
129
Experiments.
129
Shelf .
129
Special
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
129
Techniques
147
Aeration
147
Distillation ....... -.
147
Filtration .
. . . . . . . . . . . . . .. . . . . . . . .. . . . . . . . .
149
Cleaning Glassware.
149
149
pH .....
...... ......................... .
150
Phot,ometry
.................................. .
150
Polarimetry
.................................. .
154
..................................... .
Bibliography ..................................... .
Pressure.
-ill-
155
157
APPARATUS
General (available at all times in the laboratory)
Aspirators, water, each with tmp
Balances, quantitative, sensitivity, 0.1 mg.
Balances, triple-beam, capacity, 610 g., extra kilogram weight
Centrifuge, clinical
Centrifuge, International
Cork borers
Cork-borer sharpener
Cork press
Corks
Dryer, air, gas-heated
Files, triangular, S-in.
Grease.. stopcock
Hood, fume
Icebox and ice
Labels
Oven, electric with [Link], 35 to 1800C.
Paper, roll, 6-in. diameter
Refrigerator, electric, +soC.
Stoppers, rubber
Tubing, soft-glass
Water, distilled
Water, tap
Individual (in each student's locker)
Beakers, 50-mI., lOO-ml., 600-ml., l,OOO-mI., l,SOO-ml., 2,OOO-mL
Bottles, sample, 2-oz., widemouthed (6)
Bottles, weighing (3)
Brush, test-tube
Burette, sO-ml.
\
Burner, Thrill, 30 in. of rubber tubing and wing top
Clamp holders (4)
Clamps, extension, equipped With rubber fittings (2)
Clamps, screw (2)
Clamps, universal or Fisher Casta loy with three-pronged grip (2)
Condenser, Liebig, 3-ft. with two 30-in. pieces of rubber tubing
Dish, evaporating, 100-mm. diameter
Filter paper, 7-cm. diameter
Flasks, conical, SO~ml., 125-1lI1., 2S0-ml., SOO-ml.
Flask, filter, sOO-ml.
Flask, Florence, 1,OOO-ml.
Flask, round-bottomed, 2,000-ml.
Flask support, cork or wood
Flasks, volumetric, glass-stoppered, 50-mI., IOO-ml., 2S0-ml.
Funnel, Buchner, 71-mm. insiqe diameter
Funnel, glass (60 deg.), 50-mI{). top diameter
Graduate, 10-ml. cylinder
-118 -
L~ck, combination
Matches, box
Medicine dropper
Pan, enamel, 2-qt.
Pipettes, volumetric, I-mI., S-m1., 10-ml., 25-ml.
Plate, spot
Policeman, rubber (2)
Rings, iron extension, 3- and S-in. diameters
Rods, stirring (2)
Rubber tubing, pressure, 3 ft.
Spatula, metal
Sponge
Test tubes, soft-glass, 6-in. (6)
Test-tube holder
Test-tube rack
Thermometer, 110C.
Towel
Tripod, iron, 6-in. outside diameter
Tweezers, metal
Watch glass, 125-mm. diameter
Wire gauze, galvanized-iron, 16-mesh, 6-in. square (2)
,ecial (issued by the storekeeper or instructor). More than one piece is required in some experiments.
Experiment
Apparatus
number
21,36
Aeration assembly
10
Amino nitrogen, Van Slyke macro volumetric
21,28
Asbestos board with hole for 3-in. test tube
37
Atomizer, all glass with rubber bulb (DeVilbiss No. 31)
39,40
Autoclave, steam-heated
21,26
Balance, horn-pan
41
Balance, platform, with weights, 2-kg. capacity
40
Balance, Roller-Smith torsion
9
Balloon, rubber
10
Barometer
4,14
Ba th, h ydrogenated-c ottonseed -oil
12,15,30,32
Bath, sand
43
Bell jar
43
Board, with nails
46
Bottle, amber, screw-capped, narrow-mouthed, 2:5-liter
41
Bottle, amber, screw-capped, widemouthed, 2.5-liter
41
Brooder, electric with thermostat
41
Bucket, enamel, S-gal.
24,29,38
Burette, H}ml.
Cabinet, lightproof, equipped with 4 trays, sand, air vents, and drain,
19
32 by 47 by 58 in.
22
Centrifuge bottle, 200-ml.
43
Centrifuge tube, IS-mI.
14,21,36,39,44
Centrifuge tube, 50-ml.
16,19,28,31,32
Cheesecloth
Chromatographic container. Glass jar, 4 3/4 by 9 7/8 by 14 in., with
wood cover (5 by 10 in.) containing 8 slits, 4 evenly spaced per
side. Slits and cover are sealed with scotch tape after paper
37
strips are inserted in slits
-119 -
Apparatus
Clock, time
Colorimeter, photoelectric, Lumetron, Model 400A, Klett-Summerson, or other
Colorimeter, visual, Duboscq
Condenser, Hopkins
Condenser, Liebig,_lO-in.
Crucible, porcelain, No.2
Cylinder, graduated, glass-stopper~d, 10-mi.
Desiccator with-cover and crucible plate, lOb-mm. inside diameter
Distillation apparatus (see Techniques, Appendix)
E"{Jerim,ent \
number
341
35,43,45
1~,14,18,23,35,
43,46
2l,25,3S
14,29
41
43
41
2,4,8,12,16,19~
30,31,32,44
Distillation apparatus (see Techniques, Appendix)
Drying tube, 3-in.
Filter flask (see Flask, filter)
Filter paper, 15-cm. diameter
Filter-paper strips, 2 1/4 by 12 in.
Flask, conical, I-liter
Flask, conical, 2-liter
Flask, distilling, SOO-ml.
Flask, filter, 12S-ml.
Flask, filter, SOO-ml.
Flask', Florence; 1.5-liter
Flask, glass-stoppered, 2SO-mL
Flask, Kjeldahl, 800-ml.
Flask, pyrex round-bottomed, 200-ml.
Flask, pyrex round-bottomed, 500-ml.
Flask, pyrex round-bottomed, I-liter
Flask, pyrex three-necked, I-liter
Flask, volumetric, 25-ml.
Flask, volumetric, lOO-mi.
Flask, volumetric, 500-ml.
Flask, volumetric, 1,000-ml.
Funnel, Buchner, 1S1-mm. inside dial}leter at plate
Funnel, glass (60 deg.), 150i mm. top diameter
Funnel, glass (60 deg.), 300-mm. top diameter
Funnel, Hirsch, [Link]. diameter of perforated plate
Furnace, electric muffle, 1000oC.
Graduate, 2SO-ml. cylinder
Graduate, 2-liter cylinder
Grinder, food
Hammer
Incubator, 25 to 70C.
Inoculating needle with holder (resistance wire, 3 in., B. & S. gauge No. 26)
Kjeldahl digestion apparatus (macro)
Kjeldahl distillation apparatus (macro)
Knife
Lamp, ultraviolet (Mineralite), 2S4 mJl peak emission
Mantle, electric (for heating Soxhlet extraction apparatus)
Melting-point apparatus with Dow-Corning silicon fluid No. 702
Microscope, compound
Mortar with pestle, 100-mm. outside diameter
-120 -
27
8
31,32,41
37
30
30,32
20
21
9
21
26
9
3,44
2
8,12,19
8
10,13,35
11,18,40,46
13,14,35
29,40
32
31,32
16
19,21,36,40
41
20
46
27,28,30
4
36,39,40
39,40
9
9
41
37
41
21,27,36
36
36,39
Experiment
number
Apparatus
Pe~cil, wax
pfL_meter, Beckman (Model M equipped with 4990-42 glass electrode), or
other make
Pipette, graduated, I-mI.
Pipette, graduated, 1Q,.ml.
Pipette, graduated, 25-ml.
Pipette, heavy-walled capillary, 2-mm. inside diameter
Pipette, volumetric, 2-ml.
Pipette, volumetric, SO-rol.
Pipetting machine, Brewer automatic (Baltimore Biological Company), or
other make
Plate, pie, heat-resistant-glass, 12-in. diame~er
Polarimeter, sensitivity 0.05 deg.
Rods, stirring
Saccnarometer, Einnorn
"'Scissors
Shaker, Kahn
Sieve, 20-mesh
Sieve, 10-mesh
Slide, glass
Soxhlet extractor
Stirrer, Hershberg
Stirrer, motor
Test tubes, pyrex, 8-in.
Test tube, soft-glass, 3-in.
Test tubes, soft-glass, 4-in.
Test tubes, soft-glass, 8-in.
Test-tube rack to hold 4-in. test tubes
Thermometer, 360C.
Tubing, glass, capillary
Waring Blendor
Weigh~s, calibrated quantitqtive, 1 mg. to 50 g.
121-
28
20,22,40
40,43,45
39,40,43
46
37
18,35,46
9;35,40
39
27
17,30,31,32,
33,34
45
%
43
45
32
41
36
41
8
7
23
28,36
36,37,39,40
36
39
4,14,15,21,27,
36
37
19,33
1,9,11,13,14,17,
18,20,23,24,
25,26,34,39,
40,41
EXPERIMENTS
The student's experimental data should be recorded immediately, legibly, and completely in a
tightly bound notebook to be submitted at intervals for grading. The final data are to be submitted on
the data sheets or cards, except in certain experimJnts where special reports are required.
It is assumed that the student is familiar with the principles of orgariic and quantitative chemistry
and [Link] has had adequate training in laboratory techniques. In most of the quantitative experiments
the highest degree of precision and accuracy is not attainable, owing to uncontrollabie variables inherent in the determination of biological substances. On the other hand, precision and accuracy within
about 1 per cent are usually possible and often realized by the best qualified students. It has been
found desirable, therefore, to furnish quantitative balances of O.l mg. sensitivity, calibrated weights,
and high-grade (Ex ax) but uncalibrated volumetric apparatus. In most of the experiments the quantities
of substances have been selected so that weighings need be made only to 1 mg.
Instructions to Students
Construct a wash bottle during first laboratory period.
Work carefully with a minimum of noise and cQnfusion.
Keep flames away from inflammable chemicals.
Follow directions closely to avoid explosions or other accidents.
Conserve chemicals.
Keep your -desk clean.
Handle special (ex pen si ve) apparatus carefully.
Store all unused equipment (cleaned and set to drain) in the locker.
Label and cover all containers placed in the refrigerator, incubator, oven, and other special places.
Do not store solutions in volumetric flasks.
Do not autoclave volumetric apparatus (except graduated pipettes).
Record experimental data completely and neatly in a tightly bound notebook.
Study the directions for each experiment before the assigned laboratory period.
Study the explanatory material given in the manual.
Submit reports of preparations on 3... by S-in. cards according to instructions given by the instructor.
Special reports are to be made on Exp. 21, 36, and 37. The report on each of the other experiments is to
be made using only the data-sheet form provided (or a verbatim copy).
Preparations are to be submitted in sample bottles each labeled with the student's name and the
date, the tare (weight, empty) of the bottle, and the name and yield of the product.
Suggestions to Instructor. Double or larger quantities of chemicals and solutions specified under
Reagents are used by students in most of the experiments. It has been found advantageous in determining the purity of students' preparations to issue thr~e preparations to each student. By this means the
analytical skill of the student, as well as the purity of the preparations, may be evaluated with reasonable reliability. In grading the students' results, the authors' practice has been to assign 10 per cent
to yield, 40 per cent to purity, and 50 per cent to analytical skill.
Students' laboratory observations and calculations, recorded in bound notebooks, are graded on
the basis of completeness, neatness, and accuracy. Students' reports are limited in most experiments
to the data and calculations recorded on the data sheets provided. The results of students' preparations
are submitted on cards of uniform (3... by S-in.) size. Special reports ate submitted on the procedures
followed and the results obtained in Exps. 21 (identification of amino acids), 36 (identification of carbohydrates), and 37 (separation and identification of purines and pyrimidines).
-122 -
-123 -
lit, 97.6 to 87.1 for castor, and 115.5 to 112.5 for mustard. Satisfactory unknowns hav.e been prepared
from combinations of any two of these o]ls.
27. The students' yields of recrystallized cholesterol have averaged 4.9 (1.0 to 8.7per cent of dry
spinal cords and 2.8 (0.8 to 4.6) per cent of moist spinal cords. The melting points of students' preparations have averaged 144 (139 to 148.5), corr~cted.
28. Th~ students' yields of purified starch have averaged 12 (6 to 18) per cent of moist potatoes.
29. Unknowns containing starch and protein (purified gelatin) have been issued. Recoveries of
glucose added to purified starch ha,;:e ranged from 98.1 to 103 per cent when experimentally determined
starch factors varying from 0.96 to 1.06 were employed.
30. The students' yields of recrystallized D-xylose have averaged 5 (0 to 11) per cent of air-dried
corncobs.
31. The students' yields of recrystallized L-arabinose have averaged 111 (0.5 to 20) per cent of
commercial mesquite gum. Yields of the crude carbohydrate have averaged 21 (3 to 36) per cent.
32. The students' yields of crude D-mannose have averaged 12 (0.3 to 23) per cent of the ivorynut meal.
33. The students' yields of recrystallized D-lactose have averaged 9.7 (0.8 to 28) per cent of :nhey
powder. The purity (polarimetric analysis) of these products has averaged 94.5 (88 to 98) per cent.
34. Students' preparations of D-lactose have been analyzed.
35. The" percentages of reducing substances (ca4:ulated as glucose) found in colorless corn sirup
have averaged _31 (23 to 40) per cent before hydrolYSis and 73 (48 to 85) per cent after hydrolYSis, as
determined photometrically with a visual colorimeter. The percentages averaged 31.5 (24 to 40) per
cent before hydrolysis and 74 (59 to 116) after hydrolYSis, as determined photometrically with a photoelectric colorimeter. Recoveries of glucose added to corn sirup have averaged 105 (71 to 137) per cent
by visual photometric analysis of unhydrolyzed corn sirup and 96 (56 to 147) per cent by analysis of
hydrolyzed corn sirup. Recoveries have averaged 96 (82 to 127) per cent by phot~lectric photometric
analysis of unhydrolyzed corn sirup and 95 (57 to 116) per cent by analysis of hydrolyzed corn sirup.
The quantities of pure glucose added to students' solutions estimated to contain about the same amount
of glucose were 8 to 16 mg. (8 to 16 ml. of a solution containing 1.0 mg. of glucose per milliliter) for
unhydrolyzed sirup and 1.5 to 3.8 mg. (10 to 25 mI. of a solution containing 0.15 mg. of glucose per
milliliter) for hydrolyzed sirup. On this basis the quantities of glucose added per 2-ml. aliquot (see
data sheet) were 0.064 to 0.128 mg. and 0.060 to 0.15 mg., respectively.
36. Carbohydrates which have been issued in unknown mixtures (two to folir carbohydrates) include
starch, glycogen, dextrin, inulin, raffinose, lactose, maltose, fructose, galactose, glucose, mannose,
rhamnose, arabinose, and xylose. Of the carbohydrates issued, 74 per cent were identified, 26 per cent
were present but not reported, and 35 per cent were reported although not present. All the carbohydrates
in 32 per cent of the 220 unknown mixtures issued were identified. The accuracy of the reports increased
as the number of carbohydrates per mixture decreased from four to two and in cases where a preliminary
test was made of an unknown consisting of only one carbohydrate. The number of carbohydrates identified increased if the number of carbohydrates present was made known to the student. Each sample
consisted of a homogeneous mixture prepared from 1 g. of each component (0.5 g. of starch, dextrin,
or glycogen). Glucose was omitted in mixtures containing other reducing saccharides except rhamnose
and fructose.
Whole (not soluble) starch should be used. Most commercial preparations of dextrin have been
found unsatisfactory because of contaminating reducing substances (probably maltose), Dextrins may
be purified by precipitating them from a 0.5 per cent aqueous solution with sufficient ethanol to give a
70 per cent solution. Most commercial glycogens are unsatisfactory, since they give little or no color
with iodine. Glycogen prepared from clam muscle or rabbit liver has been found satisfactory. Inulin
prepared from dahlia bulbs has been found more satisfactory than the commcerical preparations tested.
37. Unknown mixtures are prepared by mixing 10 mg. of each purine or pyrimidine with sufficient
ribonucleic acid to give a total weight of 100 mg. Unknowns (one to three components) have been
issued containing adenine sulfate dihydrate, caffeine, guanine hydrochloride dihydrate, hypoxanthine,
theobromine, theophylline, xanthine, cytosine, 5-methylcytosine, thymine, and uracil. Of the substances issued (40 unknowns) 93 per cent were identified, 7 per cent were present but not reported,
and 10 per cent were reported, although not present. All the components in 3.0 unknown mixtures were identified.
- 125-
38. Students have reported canned grapefruit juice t6 contain from 36 (27 to 43) to 45 (35 to 48)
mg. of ascorbic acid per 100 mI. Recoveries of ascorbic acid added to grapefruit juice have averaged
97 (80 to 115) per cent. The quantities of ascorbic acid adde? t6 students' solutions estimated to contain about the same amount of ascorbic acid were 15 to 30 mg. (15 to 30 ml. of a solution containing
1 mg. of ascorbic acid per milliliter). On this basis the quantities of added ascorbic acid per 25-ml.
aliquot (see data sheet) were 1.5 to 3.0 mg.
The ascorbic acid (U.S.P.) utilized in recovery experiments has been found to be 97 to 98 per
cent pure when titrated with standard KIO, solution. Ascorbic acid solutions should be protected from
atmospheric oxidation immediately after preparation by adding 8-hydroxyquinoline sulfate (10 mg. per
100 ml.) and by storing the solution in a nearly filled, tightly stoppered container. Under these conditions the reducing power of the ascorbic acid decreases nearly linearly at the rate of 0.1 per cent per
hr. up to 48 hr. and probably longer. Correction factors, if established accurately, may be used satisfactorily, although the correction is negligible provided the solution is used within an hour or two after
preparation.
39. Students have reported that commercial tablets (stated to contain 100 mg. of nicotinic acid
per tablet) contained 104 (82 to 130) mg. of nicotinic acid per tablet. The volumes of standard (0.03 N)
base required to titrate the acid formed per tube have ranged from 0.8 to 4.0 mI. (standard solutions) at
o/lg of nicotinic acid and from 8.0 to 11.0 at 0.15 /lg-'Of nicotinic acid. Recoveries of nicotinic acid
added to students' solutions estimated to contain about the same amount of nicotinic acid have averaged
93 (44 to 131) per cent. The quantities of nicotinic acid added to students' solutions were 1.0 to 2.0 flg
(10 to 20 ml. of a solution-containing 0.1 flg of nicotinic acid per milliliter). Natural products to be
analyzed for nicotinic acid should be autoclaved to ensure complete liberation of the active vitamin. A
dey ball-mill mixture may be used in place of the amino acids, vitamins, AGU solution, and salts solutions.
In preparing this mixture, transfer the indicated quantities of components to a dry, sealed ball-mill jar,
rotate the mill overnight, and store the mixture in a tightly stoppered amber bottle. The quantities of
substances to be used are L-cysteine hydrochloride,S g.; DL-tryptophan, 2 g.; thiamine chloride, 20 mg.;
calcium d-pantothenate, 20 mg.; pyridoxine hydrochloride, 40 mg.; riboflavin, 40 mg.; p-aminobenzoic
acid, 2 mg.; biotin, 0.1 mg.; adenine sulfate dihydrate, 200 mg.; guanine hydrochloride dihydrate, 200 mg.;
uracil, 200 mg.; K2HP0 4 , 10 g.; KH 2P0 4, 10 g.; MgS0 4 (anhydrous), 2.2 g.; FeS0 47H 20, 0.2 g. ;and MnS044H20,
0.15 g. To prepare 50 ml. of the basal medium, mix 0.44 g. of the ball-mill mi~ture, 0.3 g. of (NH4)2S04'
and the indicated (see Exp. 39) quantities of casein hydrolysate, glucose, and sodium acetate.
~O. Students have reported that boiled egg yolk contained 1.25 (1.10 to 1.56) mg. per cent of choline.
The milligrams of mycelium obtained at different levels of choline (standa,rd) ranged from 2.5 to 2.8 mg. at
0.6 flg of choline and from 23 to 28 mg. at 6.0 /lg of choline per tube. Recoveries of choline added to
students' solutions estimated to contain about the same amount of choline have averaged 108 (87 to 128)
per cent. The quantities of choline iuided to students' solutions were 0.15 to 1.5 flg (10 to 25 mI. of
solutions containing 0.015 or 0.060 p.g of choline per milliliter. These amounts are equivalent to 0.17
and 0.068 flg of choline chloride).
41. The students are divided conveniently into groups of two, each group caring for 8 to 15 chicks.
The right tibiae should be utilized by one student and the left tibiae by the other student of each group.
It is desirable that the chicks be obtained from eggs of hens maintained on a diet relatively low in
vitamin D. The feed (Exp. 41, Note 1) for each group of chicks is to be prepared separately by each
group of students, and it should be stored in 5-gal. widemouthed metal containe~s which are kept tightly
closed when not in use. The food components may be obtained readily from grain milling companies,
feed supply stores, and chemical supply houses. -U.S.P. standard vitamin D oil for use in chick (not
rat) assays is obtainable from the U. S. Reference Standards, 4738 Kingsessing Ave., Philadelphia 43,
Pennsylvania.
Relatively smooth response curves are obtained over the range of 0 to 15 units (A.O.A.C.) per 100
g. of feed, although the most useful range is 5 to 15 units. Students have reported per cent ash ranging
from 34 (0 units of vitamin D) to 46 (15 units of vitamin D) per cent for tibiae and 8.5 (8 units of vitamin
D) to 10.7 (16 units of vitamin D) per cent for toes.
Day-old starved chicks average about 40 g. in weight. The average weiiht on the twenty-second
day is about 180 g. for chicks with adequate vitamin D and about 140 g. for chicks on vitamin D-deficient
diets. Both groups consume about the same quantity of food.
~
- 126-
At the start of the ashing procedure place the covered crucible at the front of the muffle furnace
After 15 min. remove the cover, move the crucible to the rear of the furnace, and continue
eating for 60 min. with the furnace door slightly open.
Chick assay values of commercial Vitamin D oils usually are at least 50 per cent below the stated
potencies because chicks do not respond to vitamiJl D 2 , which, in addition to D3I is present in the sample.
42. Students have reported 515 (165 to 1,110) units for the activity of salivary amylase. Absolute
errors found for the unknown starch solutions have averaged 5 (0 to 21) per cent.
43. The quantities of phosphorus added to recovery solutions estimated to contain about the same
amount of phosphorus (as phosphate) were 0.02 to 0.04 mg. of phosphorus (0.088 to 0.176 mg. of KH 2 P0 4 )
dissolved in 0.3 mI. of solution.
44. The students' yields of acetylphenylalanifle have averaged 84 (70 to 95) per cent, [Link] p-toluidide 90 per cent, and L-phenylalanine ([0:]: = +35 to +36 deg.) 30 per cent~of the theoretical amounts.
45. Cholesterol recoveries (largely previous determinations by Bloor's method) have averaged 53
(27 to 85) per cent. Values for rat serum cholesterol have averaged 60 to 110 mg. per 100 mi.
46. Of the qualitative unknowns issued, 92 pet cent were identified, 7 per cent were present but
not reported, and 14 per cent were reported although not present. In preparing the recovery solutions
for the qualitative tests, bovine plasma albumin (Armour's) or egg albumin scales (5 ml. of a 2.5 per
cent solution), glucose (0.5 to 2.0 ml. at the low level and 5.0 to 10.0 ml. at the high level of a 10 per
cent'solution), and acetone (0.5 ml.) were added to urine aliquots.
Nitrogen values reported for 24-hr. urine sampies from 10w- and high-protein diets are distributed
generally according to the data shown in the table accompanying Exp. 46. Recoveries of substances
added to students' urines have averaged 98 (93 to 105) per cent for total nitrogen, 99 (77 to 120) per
cent for urea nitrogen, 84 (63 to 108) per cent for ammonia nitrogen, 115 (70 to 154) per cent for creatinine
nitrogen, and 102 (56 to 133) per yent for uric acid nitrogen. The quantities of substances added, estimated to be about the same as those in the urine aliquot, were 200 to 600 mg. of ammonium sulfate
[2 to 6 ml. of a solution containing 10:6 mg. of (NB4)2S04 per milliliter], 10 to 30 mg. of creatinine
(2 to 6 ml. of a solution containing 5 mg. of creatinine per milliliter), and 4 to 12 mg. of uric ac.;id
(4 to 12 mI.. of a solution containing 1 mg. of uric add per milliliter).
Consult Tables of Food Composition, Bureau of Human Nutrition and Home Economics, U. S. Department of Agriculture, Miscellaneous Publication No. 572, DecemhE;!r, 1949, for the percentages of proteins
in a large variety of foods.
mooc.).
- 127-
FIRST AID
Burns, Acid or Base. Flood face, eyes, or other parts with running water from the tap. Neutralize acid burns with saturated sodium bicarbonate solution, base bums with saturated boric acid solution, and oxidizing agents with sodium thiosulfate solution.
I
Burns, Fire. Apply analgesic, antiseptic ointment and bandage with sterile gauze.
Fire, Chemicals. Smother with sand, carbon dioxide gas from pressure cylinder, or with both
materials. Use water from fire hose or other soutce, except for burning oil.
Fire, Clothes. Extinguish with water (shower, if available), wrap in fire blanket, or use both
methods.
Call Instructor Immediately.
I
_; 128 -
Unless specified otherwise, all reagents are prepared from high-quality (analyzed, C.P., or reagentgrade) chemicals.
Acids and Bases. Common concentrated, C.P. acids and bases have the, following specifications:
Acid or base(s)
Specific
gravity
1.051
0.899
1.180
1.422
1.689
1.525
1:836
.. ..... . . . .. . . . .. . .
.............. .... .
...................
Per
cent
99.5
28.9{b)
36.0
70.3
85.0
50.0
96.0
Normality
(approximate)
I
17
15
12
16
44
19'
36
Molecylar
weight
60.1
17.0(b)
36.5
63.0
98.0
40.0
98.1
[Link] specific gravity and per cent of acid or base for commercial products often vary from the values for highly
purified samples given in the table.
b NH3 .
Shelf. Concentrated and 6 N solutions of the acids and bases listed in the table are available.
Directions for the preparation of saturated sodium hydroxide solutions are given below under SpeCial.
I
A saturated solution of (technical) sodium dichromate with which to prepare cleaning solution (see
Cleaning Glassware, under Techniques) is provided. To prepare this solution, dissolve 3,050 g. of
N a2Cr20,'2H20 in 1 liter of distilled water with stirring and moderflte heating.
Special. Special reagents are available on the side shelf as required for the designated experiments. The chemicals are obtainable from concerns listed below and other companies given in the
Chemical Guide Book (Chemical,Markets, 25 Spruce St., New York City), Green Book, Buyers Directory
(Schnell Publishing Company, 59 John St., New York City), and Buyers Guidebook Number (Chemical
Industries, 522 Fifth Ave., New York City).
Al!lend Drug and Chemical Company, 117 E. 24th St., New York City (ivory-nut nieal)
American Type Culture Collection, 2029 M St., Washington, D. C. (test-tube and lyophilized cultures of
bacteria, fungi, yeasts, algae, protozoa, and bacteriophages)
Arthur H. Thomas' Company, West Washington Square, Philadelphia, Pennsylvania
Baker and Adamson (General Chemical Company), 40 Rector St., New York City (inorganic and organic
chemicals)
Carbide and Carbon Chemicals Company, 30 E. 42d St., New York City (inorganic and organic chemicals)
Cheney Brothers, Manchester, Connecticut (silk waste)
Coleman and Bell Company, Norwood, Ohio (urease tablets)
Commercial Solvents Corporation, 17 E. 42d St., New York City (organic solvents)
Difco Laboratories, Detroit, Michigan (microbiological culture media)
Dougherty Chemical Company, 87-34 134th St., Richmond Hill 18, New York (cytosine and other nucleic
acid derivatives)
,
Dow Chemical Company, Midland, Michigan (inorganic and organic chemicals)
E. 1. du Pont de Nemours, Wilmington, Delaware (inorganic and organic chemicals)
Eastman Kodak Company, Rochester, New York (organic chemicals)
Eimer and Amend, 635 Greenwich St., New York City (bacteriological culture media, indicators, indicator paper, and papain)
.
Fleischmann Yeast Company, New York City
General Biochemicals, 677 Laboratory Park, Chagrin Falls, Ohio (amino acids, proteins, and vitamins)
- 129-
G. Frederick Smith Chemical Company, 867 McKinley Ave., Columbus, Ohio (acids and salts of ceri~Il?'
iodine, and chlorine)
H.~. Hastings Company, Atlanta, Georgia (Lupinus albus seeds)
H.M. Chemical Company, 144 N. Hayworth Ave., Los Angeles, California (amip.o acids, microbiotogical
assay media, and vitamins)
Hoffman-LaRoche, Nutley, New Jersey (amino acidS and vitamins)
Huron Milling Company, 9 Park Place, New York City (glutamic acid and other amino acids)
International Miaerals and Chemical Company, 20 N. Wacker Drive, Chicago, Illinois (glutamic acid and
other amino acids)
J. T. Baker Chemical Company, Phillipsburg, New Jersey (analyzed chemicl,lls)
Kraft Foods Company, 500 Peshtigo Court, Chicago,lIllinois (sweet whey powder "Krafen")
Mallinckrodt Chemical Works, St. Louis, Missouri (inorganic chemicals)
Martin Drug Company, Tucson, Arizona (mesquite gum)
Meat packers (spinal cords, fresh beeO
Merck and Company, Rahway, New Jersey (amino acids, inorgan1c and organic chemicals, and vitamins)
Nutritional Biochemicals Corporation, Cleveland, Ohio (amino acids, enzymes, nucleic acids and
derivatives, and vitamins)
Paragon Division, Matheson Company, East Rutherford, New Jersey (organic chemicals)
Pfanstiehl Chemical Company, Waukegan, Illinois (amino acids and carbohydrates)
Schwarz Laboratories, 202 E. 44th St:, New York City (diazouracil, nucleic acids, and derivatives)
Ultra-Violet Products, 5205 Santa Monica Blvd., Los Angeles, California (ultraviolet lamp Mineralite)
Van Camp Laboratories, Terminal Island, California (amino acids, vitamins, and vitamin D concentrates)
Wallerstein Company, 180 Madison Ave., New York City (papain)
Winthrop'Chemical Company, 170 Varick St., New York City (amino acids)
Unless otherwise specified, all solutions listed below as per cent are prepared by dissolving the
required amount of substance in distilled water and diluting the solution to 100 ml. Each special
reagent bottle (or other container) should be labeled with the name of the reagent, the experiment number,
the date on which the reagent was prepared, the date on which the reagent (if unstable) is to be discarded,
and the exact concentration (normality or molality, if a standard solution). Reagents marked with asterisk
are to be dispensed from dropoing bottles.
Required per
Experiment
Reagent
student
number
Acetic anhydride (purity 99-100 per cent)
16-72 ml.
8,44,45
27,36,46
0.5-1,750 ml.
Acetone (b.p. 55.5-55.8 0 )
30 ml.
Acid, boric, 4 per cent solution
20
Acid, chromotropic (1,8-dihydroxynaphthalene-3,6-disulfonic
21
20 mg.
acid), technical
8
180 ml.
Acid, hydriodic, constant-boiling (57 per cent)
22
56
mI.
Acid, hydrochloric, 0.1 N
5-100
mI.
20,21
Acid, hydrochloric, 0.1 N (standard)
20-100 ml.
16,21,29,
Acid, hydrochloric, N
35,38
Acid, hydrochloric, 1.0 N (standard)
17
200 ml.
10,12
40-50 ml.
Acid, hydrochloric, 3 N
12,16
Acid, hydrochloric, 8 N
300-1,000 ml.
Acid, oxalic, M. Dissolve 12.6 g. of H2 C2 0 4 '2H 2 0 in
21
1 ml.
distilled water and dilute the solution to 1 liter.
23
1 mI.
*Acid, perchloric, 60 per cent solution
Acid, periodic, 0.5 M. Dissolve 114.0 g. of HI04 '2H 2 0 in
1-4 ml.
21,36
distilled water and dilute to 1 liter.
1 mI.
Acid, phosphoric, 85 per cent
36
-130 -
Reagent
Experiment
number
21
Required per
student
1-50 mI.
18,46
8-40 mI.
21,46_
15-80 m!.
9
9,10,13,
18,21
25
30-32,40
37
14,29
31,46
32
43
150-250 m!.
1-1,500 ml.
37
50 m!.
20-1,000 m!.
2 m!.
10-30 m!.
2-30 mI.
60 mI.
20 mI.
0.1 mI.
40
1 tube
Reagent
Stock T/1edium
Experiment
mumber
Re quired pe~
student
39
1 tube
39
0.7 m!.
36
46
9,10,21
39
0.3 g.
100 mg.
43
3.5 m!.
2
20
21
162 g.
10 mI.
2 mI.
43
7 m!.
22
2 mI.
7.6 g.
30 mI.
5 g.
4 g.
1 g.
0.5 g.
0.1 g.
0.1 g.
0.02 g.
2.5 ml.
5 g.
5 g.
-132 -
36
46
1.25 mI.
Reagent
Experiment
number
46
Required per
student
15 m!.
21
31
38
39
65 m!.
1 mg.
0.1 mg.
36
10-20 mI.
30-32
30-32
25-40 g.
80-220 g.
28,36,46
5-15 m!.
8,36
36
2-20 m!.
10 mI.
36
0.5 mg.
25 m!.
36
3 ml.
40
37
21
1 mI.
1 mI.
1 ml.
21
,1
m!.
Reagent
Experiment
number
Required pe'r
student
16
1 mI.
16,44
1-180 ml.
12
1 mI.
21,46
2-24 mI.
42
50 mI.
1 mI.
21
2 mI.
21
10 mI.
30-32,36
37
1-10 mI.
0.1 ml.
8,41
21
36
4,8,12,15
16,19,21
8,38
12
39
1 g.
0.7 g.
0.5-55 g.
1-10 g.
100 g.
5.6 mI.
Reagent
in 1,500 mI. of distilled water, arid adjust the pH of the
solution to 3.0 with 18 N NaOH. ,A~d 20 g. of decolorizing
carbon, stir the suspension for 10 min., filter the suspension,
and dilute the filtrate to 2 liters. Preserve the final solution
in a glass-stoppered bottle stored in the refrigerator. The
solution is stable under these conditions for 3 months or
longer.
Cerie sulfate, 0.02 M (standatd). Since commercial eerie sulfate
varies in purity, an entire lot 'should be tested and preserved.
Dissdlve 6 g. of eerie sulfate [Ce(HSO.. )J in a solution prepared by mixing (Cautiously) 30 mI. of concentrated sulfuric
acid with 500 ml. of distilled water. Dilute to 1 liter, mix
thoroughly" and store the solution in a glass-stoppered bottle.
Standardization of eerie Sulfate So(ution. Prepare 0.02
M (standard) solution of ferrous ammonium sulfate (Mohr's
salt) [Fe(NH.. )2(SO.. h" 6H z0] as follows: Dissolve 0.78 g.
(weighed to 1 mg.) of Mohr's salt in 50 mI. of a solution containing 1.5 ml. of concentrated H2S04 in a 100-ml. volumetric
flask. Dilute to the mark and mix thoroughly. Pipette
(Caution) 5.0 ml. of the eerie sulfate solution into a 125-ml.
conical flask, add 50 ml. of distilled water, and add 1.5 ml
of concentrated H2S04, Titrate with the standard Mohr's salt
solution until the color of the eerie sulfate is nearly discharged. Add 8 drops of setopaline indicator solution and
continue the titration until the color of the solution changes
from a golden brown to a light yellow. Before using this
reagent, it must be standardized against pure D-glucose by
the procedure described in Exp. 29. 1 mI. of 0.02 N eerie
sulfate is equivalent (approximately) to 1 mg. of D-glucose.
Chicks (day-old), starved
Chick diet (see Experiments, Appendix)
Chloroform, U.S.P.
Chloroform (dry)
Chloroacetic acid (m.p. 62-64)
Cholesterol (m.p. 148)
Cholesterol (m.p. 148), standard solution (0.12 mg. per ml.)
Choline, standard solution (1.20 p.g per ml.). Dissolve 0.69 g.
(weighed rapidly to 1 mg.) of choline chloride (dried in
vacuo over-a desiccant) in distilled water and dilute to
1 liter in a volumetric flask. Prepare a -standard solution by diluting 2.00 mI. to 1 liter in a volumetric flask.
Chromogenic (arsenomolybdate) reagent. Dissolve 25 g. of
ammonium molybdate [(NH.. hMo044H20] in"450 mI. of
distilled water. Add 21 mI. of concentrated H2 S04 , mix
thoroughly, and add a solution of 3 g. of N a2HAsO.7H20
in 25 mI. of distilled water. Mix thoroughly, incubate
the mixture for 24 hr. at 37, and filter if any suspended
material is visible. Store the solution in a brown bottle.
Chromotropic acid (see Acid, chromotropic)
Cobaltous' chloride (CoCI 26H2 0), 1 per cent solution
Congo red indicator solution (see Indicator, congo red solution)
- 135-
Experiment
number
Required per
student
29
50 mI.
41
41
21,36,44
26,45
2
45
45
40
4-8
1-40 mI.
50-75 mI.
35 g.
1 mg.
11 mI.
15.5 mI.
35
6 mI.
36
2 mI.
Reagent
Corncobs, air-dried
Com sirup, commercial
Cotton
Creatinine, C.P., standard solution (0.5 mg. per mI.)
Cupric carbonate [CuCOsCu(OH)21
Cupric-ion reagent. Dissolve 2.8 g. of Na2 HPO. and
4.0 g. of Rochelle salt (KNaC4H40 64H aO) in 70 mI.
of distilled water. Add 10 ml. of N !'{aOH anej a
solution of 0.8 g. of CuS04SH 20 in 10 mI. of distilled water. Dilute to 100 mI. and allow the sdlution
to stand for 24 hr. Filter if suspended material is visible.
Cupric sulfate, 0.01 M solution. Dissolve 2.50 g. of CuSO.-sH 2 0
in distilled water and dilute to 1 liter.
L-Cysteine hydrochloride
L-Cystine, standard solution (1 mg. per ml. in 0.5 N H2 S04)
Cytosine, 0.5 per cent 'solution in 2 N H2S04, Dissolve 50 mg.
in 10 mI. of 2 N sulfuric acid.
Deniges reagent. Dissolve 150 g. of HgS04 in 6 N H2S0 4 ,
dilute to 1 liter with 6 N H2 S04, and mix thoroughly.
Diazobenzenesulfonic acid reagent. Mix 4 g. of powdered
sulfanilic acid, 6 mI. ef distilled water, and 4 mI. of
concentrated HCI. Stir to a uniform suspension. Add 4 mI.
(in O.S-ml. portions over 2 min.) of a 50 per cent solution
of NaNO a with stirring and cooling in a water bath at room
temperature. Allow the mixture to stand for 5 min., filter
the suspended white, crystalline diazo derivative, and wash
the precipitate on the funnel with 5 ml. of distilled water.
App~y suction for 5 min. Weigh the moist solid and dissolve
ft in a volume of 4 N NaOH sufficient to make a 10 per cent
solution (stable for 6 hr.): The diazo compound should not
be allowed to dry completely, since the dry material is unstable to light and may explode when touched.
Diazouracil. Test with sucrose before issuing.
Dibenzylidine-xylose-dimethylacetal (m.p. 211). Add 9 mI. of
the benzaldehyde-methanol-HCI mixture'(preparation described in Exp. 36) to 1 g. of n-iyIose in a 6-in. test tube.
Shake vigorously and allow it to stand overnight. Filter the
suspension of needle crystals and wash the crystals first
with 200 mI. of distilled water and then with 50 ml. of
methanol. Dry the product at 50 for 1 hr. and store it in a
dark bottle. The yield is about 1 g.
2,6-Dichlorophenolindophenol (sodium salt), 0.1 per cent solution
1,8-Dihydroxynaphthalene-3,6-disulfonic acid (see Acid, chromotropic)
p-Dimethylaminobenzaldehyde (m.p. 72-73) 4 per cent solution
in 95 per cent ethanol
2,4-Dinitrophenylhydrazine (m.p. 199-200) 0.5 per cent
solution in 2.4 N HCI. Heat the mi~ture to boiling to
dissolve the solid, cool, and filter if a residue remains.
Diphenylamine, technical
Diphenylmethane, technical
- 136-
Experiment
number
30
35
39,40,43
46
21
35
Required per
student
100 g.
0.5 g.
1-2 plugs
3 mI.
0.3 g.
12 ml.
21
1 mI.
44
37
0.6 g.
2 mI.
0.1 mI.
21
3S mi.
37
1 mI.
36
3S mg.
36
0.1 g.
38
200 ml.
21
1 mI.
21
S mI.
18
15
60 g.
15
60 g.
Experiment
number
40
Reagent
Egg (chicken)
Ethanol, 95 per cent
8,16,19,21,
24,25,27,
Required per
student
1
5-715 mI.
28,31-33,
36,44
15,21,32,36,
Ethanol, absolute
1-125 mI.
44
Ethanol-sulfuric acid reagent. Dissolve 40 mI. of concentrated H2S04 in 150 mI. of 85 per cent ethanol.
Ether, diethyl
Ether, isopropyl (b.p. 67-69)
Ether, petroleum (b.p. 60-1000)
Ethylene glycol, technical
Ferric alum indicator solution (see Indicator, ferric alum
solution)
Ferrous sulfate, saturated solution. Dissolve 300 g. of
FeS0 4'7H 20 in a solution of 25 mI. of 6 N H 2 S04 in
750 mi. of distilled water. Heat to boiling, remove
the flame, and add (during the course of 'an hour)
three very small portions (excess will cause frothing)
of powdered iron to reduce any ferric ion. Heat the
'solution to boiling between each addition of iron.
Add 250 mI. oftlistilled water, allow the solution to
cool overnight, and filter.
Ferrous sulfate-hydrochloric acid reagent. Dissolve 0.5 g.
of FeS0 4'7H 2 0 in 100 ml. of N HCl. The solution is
stable for a week.
Filter aid (Celite 503)
36
5 mI.
15,21,28,41,
43
7
I
16,36,40
1-500 mI.
45 mI.
1.5-50 mI.
80 ml.
28
1 mI.
37
1 mI.
12,14.16.19.
1-400 g.
27,30:-33
Folin's improved reagent [see Acid, phosphotungstic
(Folin's improved reagent)]
Fuller's earth
Gelatin, purified
a-D-Glucose, anhydrous
a-D-Gluc6se, anhydrous, standard solution (0.035 mg. per mI.)
a-D-Glucose, anhydrous, standard solution (0.060 mg. per ml.)
a-O-Glucose, anhydrous, standard solution'(0.10 mg. per ml.)
a-O-Glucose, anhydrous, standard solution (0.15 mg. per ml.)
/3-Glycerophosphate (substrate) reagent. Dissolve 2.5 g. of
sodium ,B-glycerophosphate, 2.12 g. of monosodium diethylbarbiturate (veronal sodium, barbital sodium), and 1 g.
of MgCl z'6H zO in distilled water. Dilute to SOD mI., transfer to a glass-stoppered bottle, add sufficient petroleum
ether (b.p. 30-36) to form a 2-cm. layer, and preserve
in the refrigerator.
Glycine, purified
Glycine, technical
Grapefruit juice, commercial canned
Guanine, 0.5 per cent solution in 2 N H2 S04 , Dis~olve 50 mg. of
guanine hydrochloride (C s HsNO HCl'2H 2 0) in 10 ml. of 2
N sulfuric acid.
-137 -
45
g.
29
.5 g.
39,46
-1.5 g.
ml.
mI.
mI.
mI.
7 mI.
35
3S
35
35
43
3
4,7
38
37
15 g.
15 g.
210 mI.
0.1 mI.
Reagent
Hair (crude), men's
Hanus solution. Suspend 13.2 g. of iodine in 800 ml. of
glacial acetic acid. Stir and heat until the iodine
dissolves. Allow the solution to cool (in air or under
running water) to room temperature. Aqd 3.0 ml. (9 g.)
of bromine to 200 mI. of glacial acetic acid. Mix
thoroughly. Mix the iodine and bromine solutions.
Hippuric acid (benzoylglycine)
I
Hopkins-Cole-Benedict reagent. Place 10 g. of magnesium
powder in a 2-liter flask. Add enough distilled water
to cover the powder and add 'slowly (while cooling the
flask under running water) 250 ml. of a cold s:aturated
solution of oxalic acid (H2C20,,' 2H 2 0). Stir, filter if
any suspended material is visible, acidify with acetic
acid, and dilute to 1 liter with distilled water.
Hydriodic acid (see Acid, hydriodic)
Hydrochloric acid (see ACid, hydrochloric)
Hydrogen peroxide, 6 per cent solution
Hydrogen peroxide, 30 per cent 'solution
Hy*ogen sulfide (from house line or generator)
Hydroquinone, (m.p. 168-169), 1 per cent solution. Add
sulfuric acid to acid reaction with litmus paper.
p-Hydroxydiphenyl (m.p. 164-165)
Hydroxylamine hydrochloride (m.p. 153-155), 20 per cent
solution
8-Hydroxyquinoline sulfate,S per cent solution
*Indicator, bromthymol blue, 0.5 per cent solution. Dissolve
0.5 g. of bromthymol blue (dibromthymolsulfonphthalein)
in 33 ml. of 95 per cent ethanol, add 67 mI. of distilled
water, and mix thoroughly.
Indicator, congo red paper
*Indicator, congo red, 0.04 per cent solution. Dissolve 0.04 g.
of congo red (sodium tetrazodiphenylnapl}thionate) in 20 mI.
of 95 per cent ethanol, add 80 mi. 'of distilled water, and
mix thoroughly.
*Indicator, ferric alum, 10 per cent solution. Dissolve 10 g. of
ferric ammonium sulfate [Fe(NH,,)S04kl2H20] in 95 ml. of
distilled water and dilute to 100 ml. with 16 N HNO,.
Mix thoroughly.
Indicator, litmus paper
Indicator, methyl red paper
*Indicator, methyl red, 0.1 per cent solution. Dissolve 0.1 g.
of methyl red (dimethylaminoazobenzene-o-carboxylic
acid) in 50 mI. of 95 per cent ethanol and dilute to 100
ml. Mix thoroughly.
*Indicator, mixed (methyl red and p-nitrophenol) 0.02 per cent
solution of methyl red and 0.1 per cent solution of
p-nitrophenol. Dissolve 0.02 g. of methyl red in 20 mI.
of 95 per cent ethanol and dilute to 100 ml. with distilled
water. Dissolve 0.10 g. of p-nitrophenol (m.p. 112-113)
in the solution. Mix thoroughly.
-138 -
Experiment
number
16
26
Required per
student
600 g.
125 ml.
8
21
36 g.
1 ml.
21
46
36,37
23
1 mI.
21,36
21
20 mg.
4 mI.
38
21,29,39,40
4 mI.
1 mI.
7,16,21
30-32
10 'strips
1 mI.
11
1 mi.
21,27,32,36
8,12
21
10 strips
10 strips
1 ml.
9,20
1 mi.
1 mI.
Gas
25 ml.
Experiment
number
Reagent
- 139-
Required peT
studeTlt
5 'Strips
1,23-25
1 ml.
29
7
1 ml.
1 mI.
44
40
39
5 strips
0.1 mI.
10 ml.
36
1 mI.
42
15 mI.
21
21
32
9
1 mg.
2 mI.
100 g.
125 mI.
30
1 culture
33,34,36
21
0.02-2.5 g.
1 ml.
21
19
21
0.2 g.
36
1 mI.
~5
9 g.
36
32
26
14
13,14
31
2,4-6,12,22,
30-32,36
3,30,36
20 mg.
20 mg.
SO ml.
35 mi.
20 ml.
100 g.
16-975 ml.
100 g.
0.1 g.
1-100 ml.
Reagent
Experiment
number
36
30
32
36
37
Required per
student
1 ml.
25 mI.
.t m!.
1 mi.
0.1 mI.
22
16,21
100 m!.
1 m!.
23
25 m!.
21
1 ml.
21
1 ml.
21,46,
5-90 mI.
40
1 culture
39
39
21
21
21
0.1 g.
11.S m!.
1 m!.
2 ml.
2 mI.
11
50 m!.
Reagent
Experiment
number
37
1
Required per
student
0.1 g.
44
42
2 g.
10 g.
46
4 g.
44
36
23
10 g_.
0.6 g.
10 mI.
43
6 mI.
20 g.
0.5 mI.
21
5 mI.
Reagent
add to the residual solution 21.3 g. of Na,[Link]
and 300 ml. of distilled water. Distill the solution
and collect the first 300 mi. which comes over. Store
the reagent in an amber bottle. It is stable almost
indefinitely. [Patton and Foreman, Science, 109, 339
(1949).] The method of Chaudhuri [J. Am. Chern. ,Soc.,
64,315 (1942)] has been found less satisfactory.
Wawzonek and KarIl [J. ,Am. Chem. Soc. , 70, 1666
(1948)] have described a synthesis of o-tetrabromoxylene.
Picric acid (see Acid, picric)
Potassium bismuth iodide reagent. Suspend 2.4 g. of
Bi(NOs),5HaO in 50 mi. of distilled water. Ada 7.2 g.
of KI and stir the mixture until the resulting black precipitate dissolves to a 'Clear bright-red solution.
Potassium carbonate solution (50 g. per 100 mi.)
*Potassium chromate, 5 per cent solution
Potassium ferricyanide (alkaline) solution. Dissolve 8.25 g.
of potassium ferricyanide [Ks Fe(CN)6) and 10.6 g. of
anhydrous sodium carbonate in distilled water and
dilute to 1 liter. Transfer to a dark bottle and preserve
in the refrigerator (stable for 6 weeks).
Potassium fluoride (KF), 10 per cent solution in water
Potassium hydroxide, 0.1 N (standard) solution
Potassium hydroxide, 0.5 N solution in 95 pe_r cent ethanol
Potassium hydroxide, 0.5 N (standard) solution in 95 per
cent ethanol
Potassium iodate, 0.001 M (standard) solution. Dissolve
0.21 g. of KI0 3 (weighed to l mg.) in water and dilute
to 1 liter in a volumetric flask.
Potassium iodide (iodine-free), 0.1 N solution. Dissolve
16.6 g. of KI in 1 liter of water.
Potassium iodide (iodine-free), 15 per cent solution in water
Potassium permanganate
Potassium permanganate (alkaline) solution, 4 per cent in
0.5 N NaOH
Potassium thiocyanate, 0.05 N (standard) solution in water.
Dissolve 4.86 g. of KCNS (weig):[Link] to 1 mg.) in water
and dilute to 1 liter in a volumetric flask.
Potato (irish), fresh
Purines (see Experiments, Appendix)
Pyridine (b.p. 113.5-115.5)
Pyrimidines (see Experiments, Appendix)
Rats
Resorcinol, 0.2 per cent solution in 95 per cent ethanol
Salicylafdehyde (m.p. 1_2), 6 per cent in absolute ethanol.
Prepare fresh each week.
Salicylaldehyde (m.p. 1_2), 10 per cent in 95 per cent
ethanol. Prepare fresh each week.
Salts (microconstituents) solution. Dissolve in 5 liters of
distilled water 6.0 mg. of H,BO" 4.0 mg. of
(NH4)61\1070Z44HzO, 90 mg. of Fez(SO.k6H zO, 80 mg. of
- 142-
Experiment
number
Required per
student
37
1 ml.
21
36
29
6 mi.
1 mi.
30 mI.
28
24
27
25
2 ml.
50 mI.
50 ml.
75 mi.
38
150 mI.
38,42
4-20 mi.
26
21
10
50 mi.
1 g.
150 ml.
11
250 mI.
28
37
21
37
43,45
36
21
600 g.
1
10 ml.
7 mi.
46
1 mI.
40
10 ml.
3 mI.
Reagent
CuS0 45H 20, 7 mg. of MnS04'H20, and 880 mg. of
ZnS047H20. Shake the mixture before using to
resuspend any insoluble material fonned.
Salts solution. Prepare 2 liters of solution containing
2 per cent ammonium tartrate [(NH4)2C4H40J, 0.4
per cent ammonium nitrate (NH4NO~), 0.4 per cent
K2HP0 4, 0.2 per cent MgS0 47Hp, 0.04 per cent
NaCI, and 0.08 per cent CaCI26H20. Sterilize (by
autoc!aving) this solution in aliquots of a volume
which will be utilized in a day or so to avoid
bacterial contamination of solutioITS exposed to
air for longer time.
Salts A solution, 10 per cent K2HP0 4 and 10 per cent
KH 2P04 solution in water
Salts B solution. Prepare 100 mI. of a solution containing
4 per cent of MgS0 47H 20, 0.2 per cent of FeS047H2 0,
and 0.2 per cent of MnS0 44H 20, in 0.1 N HCl.
Setopaline C indicator solution (see Indicator, setopaline C
solution)
Silk waste
Silver nitrate, 1 per cent solution in water
Silver nitrate,S per cent solution in 2 N nitric acid
Silver nitrate, 0.1 N (standard) solution in water
Sodium acetate, anhydrous
Sodium arsenite, 0.1 N solution in 2 per cent sodium bicarbonate solution. Dissolve 2.6 g. of sodium arsenite
(NaAs0 2) in 100 mI. of 2 per cent NaHCO, solution.
Sodium arsenite,S per cent solution
Sodium bicarbonate, 0.5 M solution in water. Dissolve 42 g.
of N aHC0 3 in distilled water and dilute to 1 liter.
Sodium bicarbonate, N solution in water. Dissolve 84 g. of
NaHC0 3 in distilled water and dilute to 1 liter.
Sodium bicarbonate, saturated solution. Dissolve 103 g. of
NaHCO s in a liter of distilled water.
Sodium .carbonate
Sodium chloride
Sodium chloride, 0.9 per cent solution
Sodium chloride buffer (pH 6.8) solution [See Buffer
(sodium chloride-phosphate) solution]
Sodium cyanide,S per cent solution in N NaOH. Prepare
fresh each week.
Sodium cyanide, 12 per cent solution in water. Prepare
fresh each week.
Sodium hydrosuifite, 2 per cent solution of Na2SP42H20 in
water. Prepare fresh each week.
Sodium hydroxide, pellets
Sodium hydroxide, 0.03 N (standard) solution in water
Sodium hydroxide, 0.1 N solution in water
Sodium hydroxide, 0.1 N (standard) solution in water
Sodium hydroxide, 0.5 N solution in water
Sodium hydroxide, N solution in water
- 143-
Experiment
number
Required per
student
40
50 ml.
39
0.38 mI.
39
0.38 mI.
12
37
16
11
8,36,39
21,36
100 g.
1 mI.
1-5 mI.
450 mI.
0.3-17 g.
17 ml.
21
21
1 ml.
5 ml.
36
10 mI.
18
40 mI.
21
36
39
0.2 g.
0.5 g.
50 mI.
21
1 mI.
46
50 mI.
21
1 mI.
46
39
21
1,9,20
46
21,29,35,40
0.5 g.
500 mI.
1 ml.
100-300 mI.
20 ml.
20-25 mI.
Reagent
Experiment
number
36
12,21,46
14
9,21
21
21,28
Required per
student
5 mI.
4-400 mi.
10 mI.
4-150 mI.
1 ml.
50 mg.
13
14
10,21
21
20 mI.
4 mI.
1-175 mI.
1 mI.
36
21
20 mI.
5 mI.
18
8 ml.
26
37
125 mI.
0.1 mI.
37
0.1 mi.
16
27
28,29
42
1.5 liters
400 g.
0.5 g.
5 ml. each
26.38
8-10 ml.
20,21
40
4 g.
23
4 g.
10
Reagent
Sulfuric acid-n-propanol reagent (1 pt. 2 N H2 S0 4 to 3 pt.
n-propanol)
Theobromine, O.S per cent solution in 2 N H2 S0 4 , Dissolve
50 mg. in 10 ml. of 2 N H2 SO4 ,
Theophylline, 0.5 per cent solution in 2 N H2 S04 , Dissolve
50 mg. in 10 mI. of 2 N H2S04 ,
Thread, stout
Thymine, 0.5 per cent solution in 2 N H 2 S0 4 , Dissolve 50
mg. in 10 ml. of 2 N H 2 S04 ,
*Thymol, 5 per cent solution in 95 per cent ethanol
Thymolphthalein,_ indicator solution (see Indicator, thymolphthalein solution)
Toluene
p-Toluidine (m.p. 42-43)
Toweling (Boote), test-tube-rack cover
Trichloroacetic acid (see Acid, trichloroacetic)
Turkey red oil (sulfonated castor-bean oil)
L-Tyrosine, standard solution (0.5 mg. per mI. in 0.5 N
H2S04 )
L- Tyrosine, technical
Uracil, 0.5 per cent solution in 2 N H2 S04 , Dissolve 50 mg.
in 10 mI. of 2 N H2 S04 ,
Urea, 50 per cent solution in water
Urease, commercial powder
Urease, tablets (containing phosphate buffer salts)
Uric acid reagent [see Acid, phosphotungstic (Folin's
improved reagent)]
Uric acid (purified), standard solution (0.004 mg. per mI.).
Stock Solution (1 Mg. per MI.). Transfer 0.25 g. (weighed to
l mg.) of purified uric acid to a 250-ml. volumetric flask.
Dissolve' 0.30 g. of lithium car~onate in 35 mI. of distilled
water, warm the solution to 60, warm the volumetric flask,
transfer the warm lithium carbonate 'solution to the flask,
and shake the flask until the uric acid dissolves (5 min.).
Cool under running water. Add 4 mL of 40 per cent formaldehyde solution (formalin), 70 mI. of distilled water, and
(gradually) 10 mI. of N H 2S04 , Dilute to the mark, mix
thoroughly, and store in a glass-stoppered bottle in the
dark. Dilute 1 mI. of the stock solution to 250 mI. with
water in a volumetric flask and mix thoroughly.
Vitamin-D-deficient diet (see Chick diet)
Vitamin D oil, commercial (with approximate concentration of
vitamin D known)
Vitamin D oil, standard (see Experiments, Appendix)
Vitamin solution. Dissolve the following quantities of vitamins [or corresponding aliquots of stock vitamin solutions
(in 50 per cent ethanol)] in 50 mI. of water and dilute to
100 mI. (in a gmduate) with 95 per cent ethanol: thiamine
chloride, 6 mg.; calcium d-pantothenate, 6 mg.; pyridoxine
hydrochloride, 12 mg.; riboflavin, 12 mg.; p-aminobenzoic
acid, 0.6 mg.; and biotin, 0.03 mg. Resuspend the solid,
which separates, before using.
- 145-
Experiment
number
37
Required per
student
100 mI.
37
0.1 mI.
37
0.1 mI.
9
37
1 piece
0.1 mI.
36
1 mL
19,39,40,46
44
39
5-20 mI.
5 g.
1
21,31
13
5 mI.
1 mI.
15
37
0.1 mI.
10 g.
46
46
46
50 mI.
40 mg.
2 tablets
46
30 mI.
41
20 g.
41
39
1 g.
1.25 mI.
Reagent
Experiment
number
39
20
33
37
Required per
student,
1
500 mI.
25 g.
0.1 mI.
30,36
36
9
0.1 g.
1 g.
3 g.
H 2 S0 4 ,
D-Xylose, crystalline
Yeast, baker's, active dehydrated
Zinc, granular
-146 -
TECHNIQUES
Aeration. See Fig. 5 for the aeration assembly used in Exps. 21 and 36.
A
E
Fig. 5.
Distillation. See Figs. 6 and 7 for the apparatus to be used in distilling liquids under atmospheric
Fig. 6.
A-Burner
F-Condenser, water-cooled
B-Tripod
C-Woter bath
-147 -
H
A-Flask, SOD-mi. round.
bottomed
.a-Water bath
C-Foam trap (glass), 1-1 iter
D-Condenser, Graham
E-Flask, SOO-ml. filter
F_Flask, l-liter roundbottomed (trap for water
aspirator)
G_ Tube leading to water
aspirator
H -Rubber stopper (i nverted
position)
I -Rubber stopper
J -Pressure tubing
Fig. 7.
D E
A-Conical flask, 125-ml.
B-Test tube, 6-in. (with protruding ring of glass
as shown)
C-Rubber stopper
D-Glass tube, inlet
E-Glass tube, outlet
Fig.8.
-148 -
pressure and under reduced pressure. The Hopkins condenser shown in Fig. 8 is to be used in Exps.
21, '25, and 35 for refluxing liquids. Distilling apparatus should be set up properly especially to avoid
closed systems which would explode during distillation. In distilling under reduced pressure, turn on
the water aspirator before applying heat, disconnect the tubing leading to the wafer a~l>irator before
removing the source of heat when discontinuing distillation, and employ only round-bottomed flasks.
Inflammable solvents (diethyl ether, acetone, and petroleum ether) should be distilled at atmospheric pressure, using steam from the house line or hot water heated some distance away but never with
an open flame. Other solvents (benzene, alcohols, and higher boiling ethers) may be distilled safely by
means of an'electric hot plate or mantle with covered heating elements or with a flame-heated.. water bath.
Fi Itration. Relatively large amounts of suspensions may be filtered m9st conveniently under reduced pressure on a Buchner funnel and relatively small amounts on a Hirsch funnel. A precipitate on
the funnel should be washed by disconnecting the tube leading from the filter flask to the water aspirator, adding the wash liquid to the funnel, stirring the precipitate and wash liquid in such a manner that
no fibers ar~ detached from the filter paper, and reapplying suction.
Filtration of colloidal materials, which filter slowly because of ~logging of the pores of the filter
paper, may be facilitated by depositing a layer (by filtering a suspension) of filter aid on the filter paper
before introducing the colloidal product. Colloidal material may also be filtered effectively by gravity
on a fluted filter (Robertson, 1943).
Cleaning Glassware. All glassware, especially that in special apparatus, is expensive and difficult to replace. It should be handled with great care to avoid inconvenience to the class and the assessment of breakage bills.
In cleaning- glassware, first remove as much of the foreign material as possible with the aid of
soap solution and a brush or other mechanical means. Following this preliminary cleaning, dissolve
remaining deposits in tap water, acetone or methanol, tap water, and distilled water, in the order given.
Allow the washed apparatus to drain and dry. If desired: traces of moisture may be removed quickly by
passing _hot air from a dryer (gas-heated, with compressed-air attachment) through the apparatus.
If the apparatus is not clean after the water has drained, add sufficient (avoid excess) stock
cleaning solution. to wet the entire inner surface of the flask, pour any excess cleaning solution into
the stock bottle, add 1 to S mI. of water to the apparatus, rotate the apparatus -so that the (now hot)
cleaning liquid will come into contact with the entire surface, and allow the cleaning liquid to stand in
the apparatus for 5 min. Rinse the apparatus thoroughly with tap water and then with distilled water.
Discard the spent (reduced) cleaning solution, which has changed color from the initial reddish brown
to a dark green.
Stock cleaning solution is prepared by mixing (Cautiously) 200 ml. of concentrated sulfuric acHi
with 7 ml. of saturated sodium dichromate solution. It is extremely corrosive and ~hould be handled
with great- caution. Rags soaked with cleaning solution may ignite spontaneously and should not be
stored. in the locker.
Melting Point. The purity of the substance prepared in Exp. 27 and the identity of the substances
investigated in Exps. 21 and 36 may be determined by means of their-melting points.
Meltjng-point tubes are prepared by heating a test tube or thin-walled glass tubing uniformly over
a small area until the glass is thoroughly softened, removing the heated area from the flame, immediately
drawing the glass until the heated part is drawn to the desired diameter, cutting the tube to the proper
length, and sealing one end in the edge of the flame. Use an oxygen or compressed-air gas flame to
soften pyrex glass.
A quantity of the dried material sufficient to form a layer 6 to 8 mm. deep is placed in a 1 to
2-mm. capillary tube (prepared as described) 8 to 10 cm. in length. It is convenient to pack the powder
by drawing a triangular fil~ across the tube and by tapping the tube sharply on the desk. Attach the
tube to a thermometer -with the sample directly opposite the bulb by means of a small rubber band or a
helix constructed from resistance wire (B. & S. gauge No. 28).
The student's thermometer should be checked against a calibrated thermometer by the following
procedure: Immerse both thermometers to the same depth in a suitable liquid heated at least 10 above
the temperature at which the substance melts, allow the bath to cool in air, and observe Simultaneously
the temperatures recorded on the two thermometers. Prepare a table of these data, which should be
- 149-
used as the basis of temperature corrections to be applied to the student's thermometer. Fixed points
for thermometer calibration are given in chemical handbooks.
pH. pH defined by Sorensen in 1909 as the negative logarithm of the [Link] concentration
(- log [H+1, or log 1/ [H +]), is commonly determined by photometlic and electrometric methods. The
approximate pH may be detennined photometrically by comparing the colors (intensities and tints) of an
unknown and standard buffer solutions of known pH values to which the same small volumes of an indicator solution have been added. Bromthymol blue, which changes color from yellow to blue over the
pH range 6.0 to 7.6, is an example of an indicator used for this purpose.
pH is detennined electrometrically with the hydrogen electrode, the quinhydrone electrode, and
the glass electrode. The principle. of all electrometric methods is the determination (with a voltmeter)
of the potential difference between two half cells joined by a salt bridge. One half cell is a standard
cell of constant potential, usually the calomel cell (saturated HgCl 2 solution in contact with metallic
mercury by means of a KCI bridge). The other half cell is the solution of unknown pH (that is, potential).
pH is calculated from the equation, pH = (E - 0.243)/0.059, where E is volts measured as ptltential
difference between the two half cells.
The glass electrode consists of a glass bulb which (a) contains a fixed solution of constant pH
and a metal- metal salt (such as Ag-AgCI) electrode of constant potential and which (b) is immersed
in a solution of unknown pH with liquid-junction contact to a calomel (or other cell). Sensitive vacuumtube potentiometers are required to measure the potential, since glass bulbs of relatively high resistance
are commonly used. The accuracy of the measurement is limited to about 0.02 pH unit, primarily because
of variations in the liquid-junction potential. It is also necessaty to shield the glass membranes from
the effects of induced currents. The glass electrode method is widely used, since it is convenient,
rapid, and adaptable to colored, turbid, and viscous solutions. pH values .below 0 and above 10 cannot
be-'detennined accurately with the glass electrode because of the variable hydration of the hydrogen ion
on the one hand and the vary-ing permeability of the glass for metal ions on the other. The operation of
the pH meter to be used in the present experiments will be explained and demonstrated by the instructor.
Photometry. For discussions of the principles and applications of photometry see reviews by
Strafford (1936), Drabkin in Glasser (1944), Gibb (1942), Snell and Snell (1948), Yoe (1928), Summerson
(1939), MUller (1939), and Vredenburg (1950).
Photometry is the technique employed in determining the concentration of a substance dissolved
in water or other solvent by means of the ultraviolet (200 to 400 mp.l), infrared (800 to 1,500 mp.) or
visible (410 to 810 mil) light which is transmitted; In photometric analysiS, light (preferably nearly
monochromatic of a particular wave length) is passed through a 'solution of a colored substance, the
amount of absorbed light is compared, with that absorbed by a known concentration of the pure substance
in a standard solution, and the concentration of the substance in the unknown solution is calculated
from these data. Visual photometric analysis is perfonned with a visual colorimeter (named, more correctly, comparator), such as that shown in Fig. 9 (shown on following page), by detennining the depth
(thickness) of the unknown colored solution required to give the same color intensity as that of a fixed
depth of a standard solution. This is accomplished by lengthening or shortening the light path (lowering
or raising the glass plunger in the colorimeter cup) and matching the color fields (one half from the unknown and the other from the 'standard solution).
The absorption and transmission of light by an absorbing medium are expressed by the Beer-Lambert equation, log 10-/1 = kcl, where 10 is the incident light, 1 is the transmitted light, k is the extinction
coefficient, c is the concentration of the absorbing material, and I is the thickness (in centimeters) of
the absorbing solution. Some of the light which is passed into an absorbing layer is absorbed and a
fraction (l/x) is transmitted. The intensity of the light transmitted is lo(l/x) when the absorbing layer
is of unit thickness, 10(I/x) (l/x) for two layers, lo(l/x 3 ) for three layers, and lo(l/xl) for 1 layers.
Taking logx of both sides of the equation 1 = lo(l/xl) gives I = logx 10/1. Since the logarithm of any
number to the base 10 is related to the logarithm of this number to another base by a constant,
lo~o 10/1 = k logx 10/1, and 10g,_0 Iell = kl. The last expression is the Beer-Lambert relation at unit
concentration.
11 mil (millimicron)
= 10
- 150 -
I'
L-Lens system
B-Biprism
R-Rhomboid prism
----t---p
P-Glass plunger
C-Cup
G
Fig. 9.
For solutions of different concentrations, light transmissions may be made equal by adjusting the
length of the light paths (depths of the solutions). Under these conditions
loglo 10 11 1 = logloIo'I12 = k 1 c l l l = k2c212'
Since kl k2 (for the same solute), cll l C2/2 This relation (Beer's law) is commonly employed in
visual photometric procedures, but it holds strictly only when the light source is monochromatic (or
composed of narrow spectral band obtained with a filter). The relationship is invalidated also for
solutes which associate, dissociate, or shift in equilibrium on dilution. At similar concentrations
(where 11 and 12 differ by only about 5 mm.) of solute in the unknown and standard solutions, the relationship holds approximately. The concentration of solute in the unknown may be determined by determining the color intensities of different concentrations of the pure solute in a series of solutions,
plotting these experimental values, drawing a smooth curve, and interpolating this standard curve.
A diagram of the Duboscq visual colorimeter is shown in Fig. 9. In carry,ing out a photometric
analysis with a visual colorimeter, attention should be given to the following points:
1. Wash the cups and glass plungers with distilled water and dry them carefully with lens paper
to avoid scratching the glass surfaces. Be careful not to crack or ,chip these glass parts.
2. Place the cups in position in the instrument and raise each cup by rotating its control wheel
until the plunger makes gentle but firm contact with the botto~ of the cup. The reading on each scale
should be zero. If not, consult the instructor, who will adjust the scales.
3. Observe the circular field through the eyepiece. The entire field should be nearly colorless
and of uniform appearance. If one half of the field differs perceptibly from the other, change the angle
of the mirror, the focus of the lenses by adjusting the eyepiece, or the position of the colorimeter with
reference to the source of light.
=;;
=;;
-151 -
4. Lower the cups to their original positions, remove ,the cups, and rinse them with the standard
solution. Half fill each cup with the standard solution and place the cups in position in the instrument.
5. Raise the left cup until the scale reading is 20.0 mm. (thickness of the solution from the bottom of the cup to the bottom of the plunger). Make certain that all air bubbles have been displaced.
6. Raise the right cup until the color intensity of one half fie~d, observed through the eyepiece,
appears to match exactly that of the other half field. Note that thecolor of the rigM half field reflects
the solution in the left cup, and vice versa. Record the scale reading. Repeat these' manipulations four
times and record each scale re?ding. Raise the rigllt cup until the scale reading is 1 or 2 mm. less than
that observed when the color intensities appeared to be matched. Lower the cup until the color intensities of the two half fields appear to match exactly. R~peat these manipulations four times and record
"-the scale readings. The averages of the two sets of scale readings should agree within 0.1 mm., and
the average value should differ from the scale reading of the left cup by not more than 0.1 mm. If the
values disagree more than this. amount, repeat the described procedures.
7. Lower the right cup and discard the solution. Rinse the cup twice with the unknown solution
and half fill the cup with this solution. Adjust the right cup until the half fields of the unknown and the
standard match. Record and average the readings approached from above and below the matching point
as described. If the average scale reading of the unknown differs from that of the standard by more than
5 mm., repeat the determination using a standard solution of different concentration.
8. Discard the solutions, rinse the cups and plungers with distilled water, and wipe the cups and
plungers with lens paper and the metal parts of the instrument with a towel. Replace the instrument in
its container.
Photoelectric colorimeters (see Figs. 10 and 11) are usually graduated to give optical density
(absorbance) amd per cent transmission (transmittance). the light, rendered essentially monochromatic
p
12.0v
200
Ohms
AC
Fig. 10.
M-Microammeter {calibrated as optical density
and per cent transmission)
'f-Filter
P-Push button (on and off SWitch)
C-Photovoltaic cell
S -Semple tube
B-Blank tube
L-Projector lamp
by passage through a filter, is passed through water ot a blank solution containing all unknown components except the chromogenic substance, and the instrument'is adjusted for zero optical density, or
100 per cent transmission. Solutions containing the chromogen at various concentrations are placed ir
tum in the light path, and the optical denSity is read from the scale. Measurements are made usually
with light of the wave length at which the unknown colored compound exhibits maximum absorption.
Optical density is defined as log}o/} and per cent transmission~ as 100}/}.
Since log I 0 /1 = kcl,
0
c = (l/kl)/rog 1/1), and a plot of c against log 1/1 on millimeter coordinate paper gives a straight lit
of slope llkl. The light-path length is kept constant by use of matched glass containers Ccuvettes) in
the form of test tubes Or rectangular glass cells. In measurements of per cent t;ansmission (100//[0)'
C '" - Ilk! (log 100[I/ ),and a straight line results if c is plotted against 1001// on semilogarithmic
o
0
-152 -
120v AC
<C5>
:0
I
L e -Lens
F-Filter
L.
S -Sample or blank
~#$~F
:I
I
I
SW1-Light switch
Swz-Galvanometer switch (when closed galvanometer
is shunted from circuit)
+
400 Ohms
1000 Ohms
Fig. 11.
paper (transmission as the vertical logarithmic axis). Optical density and per cent transmission may he
compared at any concentration of a chromogen since
1
10
1
I
- log - = - -log - =:
kl
1
kl
10
Optical density
or
I 1)
1(
I',)
colorimeters 2
Single-photocell
are designed so that light passing through a colored solution and
impinging upon a photoelectric cell generates a current proportional to the intensity of the light. The
current is measured by means of a microammeter. The difference in currents produced by two photocells
(one receiving light directly from the source and the other after it has passed through the colored solution) is measured by means of the two-photocell instruments.3
Two principal types of photocells, the barrier-layer or photovoltaic cell and the high-vacuum cell,
are used in the two-cell photometers. The former consists of a photosensitive surface (cuprous oxide or
selenium semiconducting layer on a plate of copper or iron covered by a light-transmitting layer of metal)
which, on exposure to light, liberates electrons and produces current. This cell requires a simple circuit
for measurement of the current, which is nearly proportional to the intensity of the light. The latter type
of photocell consists of a highly evacuated cell with a composite cesium cathode. When a pot~ntial is
applied, the cathode becomes photosensitive and a current is generated which is propo_!1:ional to the inintensity of the light received by the cell. This current (in contrast to that generated by barrier-layer
2
Lumetron Model 400A, Cenco-Sheard, Sanford, Evelyn, Fisher, Hellige-Diller, Kromatrol, beitz, Pfaltz and
Bauer, and Y oe and Crumpler.
3 Aminco
-153 -
cells) may be amplified, making possible accurate measurements at low optical densities. More'comL""jcated circuits are required for operation of the high-vacuum ,cell.
The following points should be observed when using the Lum7tron Model 400A single-cell photometer:
1. Obtain at least two matched 6-in. test tubes which are free from scratches. pean and dry the
tubes.
2. Plug the power cord of the instrument into the constant-voltage transformer and the power cord of
the transformer into a 105- to 125-volt, 6O-cyc1e, a-c line. If necessary, the instructor will adjust the
(lower scale) zero setting of the microammeter.
3. Insert the proper filter in the light path.
4. Fill a matched tube with the blank solution ~nd place it in the opening (marked B in Fig. 10) so
that the mark near the lip of the tube faces you (the tUbes were matched in this position). Similarly, fill
a second matched tube with the standard or sample to be analyzed (usually that of lowest intensity) and
place the tube in the opening' (marked S) as described.
5. Move the holder to bring the blank in the light path. Tum the fine-control (lower) knob to a ~
central position. Press the push button and tum first the coarse-control knob and then the fine-control
knob until the needle rests on the line of the scale indicating zero optical density.
6. Release the push button, move the holder to bring the sample in the light path, press the button,
and observe the reading.
7. Repeat steps 5 and 6 until the readings are cl6sely agreeing.
8. Remove the sample tube, rinse the tube twice with distilled water and once with the sample
solution of next higher intensity, or rinse t)1e tube twice with distilled water and dry the tube, and fill
the tube with the sample solution.
9. Rinse the matched tubes and allow them to drain. Disconnect the power cord of the instrument.
The Klett-Summerson photometer is designed so that the current of one photocell receiving light
directly from the source is balanced (by a null-point galvanometer) against the current produced a second
photocell receiving light which has passed through the sample solution. Balance is obtained by means
of a potentiometer which indicates on a scale the light absorption of the sample. The scale is graduated
in optical density (multiplied by 500) units; matched 5-in. test tubes (14-mm. outside diameter) are used;
and three filters are employed (No. 42, blue range (400 to 465 mji); No. 54, green range (500 to 570 mIL);
and No. 66, red range (640 to 700 mil)].
Observe the following points when using the Klett-Summerson photometer.
1. Clean and dry 5-in. matched test tubes.
2. Plug the power cord of the instrument into a 110- to 120-volt, [Link], a-c or d-c power line.
3. Insert the proper light filter into the holder and place the latter in the space provided between
the lamp housing and the tube holder.
4. Adjust the plJinter (by means ~f the knob directly above) until it coincides with the line (only
with the lamp off;.
./
5. Tum the large-scale knob until the scale reading is zero. Tum on the light switch and place
a matched tube containing the blank solution in the tube holder with the mark on the tube facing you.
Open the galvanometer switch (right side of instrument), adjust the pointer (which swings when the
switch is on) to coincide with the line by means of the knob (left side of test-tube holder), and, after
waiting several minutes, readjust the pointer.
6. Replace the tube with one containing a standard or unknown solution, turn the large-scale knob
until the pointer coincides with the line, and record the scale reading. Adjust the instrument again, using the
blank and unknown solutions to obtain check readings, which- should agree within 1 per cent.
7. Remove the sample tube and replace it with one containing solution of next higher intensity
and repeat the described manipulations.
8. Rinse the matched tubes and tum off the switches to the lamp and the galvanometer.
The concentrations of colored substances are to be determined photometrically in Exps. 13 and 14
(tyrosine), 18 (cystine), 23 and 43 (phosphate), 35 (glucose), 45 (cholesterol), and 46 (ammonia).
Polarimetry. The principles and applications of polarimetry have been reviewed by Bates and
associates (1942), Brown and Zerban (1941), Weissbel,"ger (1946), and Forrest in Glasser (1944).
- 154 -
Polarimetry is the technique used in measuring the number of angular deg~ees to which the plane
of polarized light is rotated by a solution of an optically active substance. Polarized light (vibrates in
a single plane) is produced by passing white light through a nicol prism [two wedge-shaped pieces of
calcite (crystalline calcium carbonate) cemented together with canada balsam]. The pl~e of polarization may be rotated to the right (dextro) or the left (levo) depending upon the structure and spatial configuration of the optically active substance. The rotations of the antipodes-of an optically active sub- ,
stance are equal in magnitude but opposite in sign.
Optical rotation is measured with a polarimeter under arbitrary conditions, and the specific rotation,
characteristic of each optically active substance, is calculated from the equation
[a] t
D
= ax
100 =
lxpxd
a x 100
lxc
p x d)
The principle of the polarimeter is shown in Fig. 12. The light (L) entering the polarimeter is
divided into two rays; one (shown) emerges from the polarizer (P) polarized, and the other (not -shown)
OS
Fig.12.
A-Analyzer
O-Objective
C-Condensing I ens
P -Polarizer
E-Eyepiece
Q-Quartz plate"
S -Scale
T -Tube
lamp, electric}
is bent, reflected, and absorbed. The phase of the plane-polarized light is altered half a wave length
by the quartz plate (Q) which covers half the field in causing the field to be half shadowed.
The Spencer polarimeter with Lippich half-shade polarizer (Spencer Lens Company, Buffalo, New
York) has been found satisfactory for use by students under the conditions described in this manual.
The analyzer and polarizer consist of Polaroid disks in place of the usual nicol prisms. The polarizer
is made of two pieces of Polaroid sheet, one covering the entire aperture and the other (narrow strip)'
covering the central part. There is produced by this means a central dark field with which the two adjacent fields (produced by light passing through the solution) are matched when the end point is reached.
Polaroid, a polarizing material in sheet form, consists of microscopic needles of iodoquinone embedded
in cellulose nitrate or other matrix. All the needles are orientated in the same direction.
Pressure. Atmospheric-pressure values of use in Exp. 10 are obtained by reading a barometer in
the laboratory. In determining the corrected reading, subtract the appropriate value interpolated from
the following list of corrections at the indicated temperatures: 150 (2 mm.), 200 (2.5 mm.), 250 (3.0
mm.), and 300 (3.5 mm.). The corrections are to reduce readings of a mercurial barometer with a brass
scale to OOC.
-155 -
BIBLIOGRAPHY
Abderhalden, E., and Gebelein, F. 1926. Uber DecarboxylierungvonAminosauren unterDildung der entsprechenden
Amine und tiber die Darstellung der> Enolform von, 2,5-Dioxo-piperazinen. Z. physioL Chem., 152, 125.
- - and Schmidt, H. 1911. Uber die Verwendung von Triketohydrindenhydrat zum Nachweis von Eiweiss-stoffen
wid deren Abbaustufen. Z. physiol. Chern. 72, 37.
'
- - and - - . 1913. Einige Beobachtungen und Versuche mit Triketohydrindenhydrat (Ruhemann). Z. pbysioL
Chem., 85, 143.
- - and Weil, A. 1911. Uber den Gehalt iigyptischer Mumien an Eiweiss und Eiwe,issabbauprodukten. Z. physiol.
Chem.,72, 15.
Abrams, R., Ham~arsten, E., and Shemin, D. 1948. Glycine as a precursor of purines in yeast. J. BioI. Chern.,
173, 429.
Abul-Fadl, M. A. M., and King, E. J. 1949. Properties of the acid phosphatases of erythrocytes and of the human
prostate gland. Biochem. J 45. 51.
Acree, S. F., 1906-1907. On the detection of formaldehyde in milk. J. Bio!. Chem., 2, 145.
- - . 1921. Mucic acid, etc. Brit. Pat. 160,777 (1921). Chern. Abstracts, 15, 2545.
Adamkiewicz, A. 1874. Farbenreaktionen des Albumin. Arch. ges. Pbysiol., 9, 156.
Adams, J., Acker, M., and Frediani, H. S. 1947. Iodimetric titrimetric determination of ascorbic acid. J. Am.
Ph arm. Assoc., 36, 170.
Agulhon, H. 1911. Recherche colorimetrique de l'alcool en presence de l'acetone. Reactions colorees de certains
groupements organiques en presence d'acides mineraux et de bichromate de potassium. Bull. soc. chim. (4) 9,
881.
Albanese, A. A., and Irby, V. 1944. Determination of urinary amino nitrogen by the copper .method. J. Bioi. Chem.,
153, 583.
Alexander, B., Landwehr, G., and Seligman, A. M. 1945. A specific micromethod for the colorimetric determination
of glycine in blood and urine. J. BioI. Chern.~ 160, 51.
- - and Seligman, A. M. 1945. A colorimetic method for the microdetermination of a-alanine in blood. J. Bioi.
Chem., 159, 9.
Almquist, H.J. 1946. Interrelations between choline, betaine, and methionine. Science, 103, 722.
American Can Company. 1943. The Canned Food Reference Manual, 2d ed. American Can Company, New York.
Andersch, M. A. 1946. The determination of serum amylase, with particular reference to the use of /3 -amylose as
the substrate. J. BioI. Chem., 166, 70S.
Anderson, E., and Otis, L. 1930. The composition and structure of mesquite gum. J. Am. Chern. Soc., 52, 4461.
- - and Sands, L. 1925. Preparation of I-arabinose from mesquite gum. Ind. Eng. Chem., 17, 1257.
- - and - - . 1926. The composition of mesquite gum; the isolation of d-galactose and I-arabinose. J. Am.
Chem. Soc., 48, 3172.
- _ and--. 1941. I-Arabinose. Org. Syntheses, ColI. Vol. 1, 67.
- - , - - , and Sturgis, N. 1925. Some plant gums of the Southwestern United States. Am. J. Pharrn., 97, 589.
Andrews, J.C., and Andrews, K. C. 1937. Substances which inhibit color development in the Sullivan method for
cystine. J. Bioi. Chem., 118, 555.
Anson, M. L. 1937. Estimation of papain with hemoglobin. J. Gen. Physio/., 20, 561.
Arhimo, A. A. 1939. The determination of small amounts of aspartic acid by the malic acid method of Pucher.
Suomen Kemistilehti, 12B, 6.
Armitage, F. D. 1943. A systematic identification of starches. Inti. Chemist, 19, 383, 398.
Arnow, L. E. 1937. Colorimetric determination of the components of 3,4.-dihydroxyphenylalanine-tyrosine mixtures.
J. BioI. Chem., 118, 531.
Association of Official Agricultural Chemists. 1945. Official and Tentative Methods of Analysis, 6th ed. Washington, D.C.
Association of Vitamin Chemists. 1947. Methods of Vitamin Assay, Interscience Publishers, New York.
Bacher, J. E., and Allen, F. W. 1950. Mild dephosphorylation of pyri~dine nucleotides. Federation Proc., 9, 148.
Bailey, K., Chibnall, A. C., Rees, M. W., and Williams, E. F. 1943. Critique of the Foreman method for the estimation of the dicarboxylic acids in protein hydrolysates. Biochem. J., 37, 360.
- 157-
- 158-
Bell, R. D., and Daisy, E. A. 1920. Rapid colorimetric methods for the determination of phosphorus in urine and
blood. j. Bioi. Chem.. 44. 55.
Benedict, S. R. 1907. The detection anl esti!TIation of reducing sugars. j. Bioi. Chern 3. 101.
1908-1909. A reagent for the detection of reducing sugars. j. Bioi. Ch'em., 5, 485.
1911. The detection and estimation of glucose in urine. j. Am. Med. Assoc.. 57. 1193.
1915. On the colorimetric determination of uric acid in blood. J. Bioi. Chern 20, 629.
1918. A modification Qf the Lewis-Benedict method for the determination of sugar in the blood. J. BioI.
Chern 34. 203.
- - . 1931. The analysis of whole blood. II. The determination of sugar and of saccharoids (non-fermentable
copper-reducing substances). J. Bioi. Chern .. 92. 141.
and Behre, J. A. 1936. Some applications of a new color reaction for creatinine. J. BioL Chern., 114. 515.
and Franke, E. 1922. A method for the direct determination of uric acid in u~ine. J. Bioi. Chem.. 52, 387.
and Hitchcock, E. H. 1915. On the colorimetric estimation of uric acid in urine. J. BioI. Chem.1 20, 619.
and Osterberg, E. 1918. A method for the determination of sugar in normal urine. J. Biol~ Chem.. 34. 195.
Berggren, R. E. L. 1932. The application of the Fiske-Subbarow colorimetric method to the determination of phosphorus in casein. j. Bioi. Chem.. 95. 461.
Bergmann, M., and Fruton, J. S. 1941. The specificity of the proteinases. Advances in Enzymol 1. 63.
_ _ and Niemann, C. 1937. On the structure of proteins: cattle hemoglobin, egg albumin, cattle fibrin, and gelatin. J. 8iol. Chem.. 118. 301.
.
- - and - - . 1938. On the structure of sUk fibroin. J. BioI. Chern 122, 577.
Bertrand, G. 1891. Recherches sur quelques derives du xylose. BulL soc. chim. (3), 5. 554.
Bessey, O. A. 1938a. Vitamin C. Methods of assay and dietary sources. J. Am. Med. Assoc 111, 1290.
- - . 1938b. A method for the determination of small quantities of ascoroic acid and dehydroascorbic acid in
turbid and colored solutions in the presence of other reducing substances. j. BioI. Chem., 126, 771.
--and King, C. G. 1933. The distribution of vitamin C in plant and animal tissues, and its determination.
J. BioI. Chem., 103. 687.
Best, C. H., Hershey, J. M., and Huntsman, M. E. 1932. The effect of lecithin on fat deposition in the liver of
the normal rat. J. Physiol., 75, 56.
Bethke, R. M., Steenbock, H., and Nelson, M. T. 1923-1924. Fat-soluble vitamins. XV. Calcium and phosphorus
relations to growth and composition of blood and bone with varying vitamin intake. J. BioI. Chem., 58, 71.
Bial, M. 1902. Die Diagnosie der Pentosurie. Deut. med. Wochschr., 28, 253.
Bills, C. E. 1935. Physiology of the sterols, including vitamin D. Physiol. Revs., 15, 1.
- - . 1947. Vitamin D assay - line test and chemical methods. Bioi. Symposia, 12, 409.
Bish, E. J. B. 1929. The determination of small quantities of starch in vegetable tissue;, Biochem. J. 23, 31.
Bishop, J. P. 1944. Extending the use of products from corn in postwar processed foods. Proc. Ins!. Food.
[Link].,5th conr., 125.
Blancheti~re. A. 1927. S~paration et dosage des 2,5-diacipiperazines en presence des amino-acides et des
peptides. Bull. soc. chim. (4), 41, 101.
Blanksma, J. J. 1909. Over de constitutie van het oxymethylfurfural. Chem. Weelthlad; 6, 1047; Cent. 1910, 1, 539.
Blatherwick. N. R., Bradshaw, P. J., Cullimore, O. S Ewing, M. E., Larson, H. W., and Sawyer, S. D. 1936. The
metabolism of d-xylose. J. BioI. Chern., 113, 405.
Dloch, K., and Schoenheimer, R. 1941. The biological precursors of creatine. J. BioI. Chem., 138, 167.
Block, R. J. 1946. The isolation and synthesis of the naturally occurring a-amino acids. Chern. Revs., 38, 501.
_ _ and Bolling. D. 1939. Chemical metabolic studies on phenylalanine. I. The nitration of phenylalanine.
J. Bioi. Chem., 129, 1.
and - - . 1940. The Determination of the Amino Acids. Burgess Publishing Company, Minneapolis, Minn.
_ _ and _ _ . 1945. The Amino Acid Composition of Proteins and Foods. Charles C. Thomas, Publisher, Springfield, Ill.
- - , - - , and Kondritzer, A. A. 1940. Microestimation of leucine, isoleucine, and valine. Proc. Soc. Ex{ltl.
Bioi. Med., 45, 289.
Bloor, W. R. 1943. Biochemistry of the Fatty Acids. Reinhold Publishing Corporation, New York.
Bodansky, A. 1933. Phosphatase studies. II. Determination of serum phosphatase. Factors influencing the
accuracy of the determination. J. BioI. Chern., 101, 93.
- - and Jaffe, H. L. 1934. Phosphatase. Serum phosphatase in diseases of the bone: interpretation and significance. Arch. Internal Med., 54, 88.
Bodansky, M., and Bodansky, O. 1940. Biochemistry of Disease. The Macmillan Company, New York.
Bollinger, A. J. 1944. Color reactions for sugars. I. Identification and determination of monosaccharides with
thymol, HCI and FeC13 Proc. Roy. Soc., N. S., Wales, 77, 109.
Bolomey, R. A., and Kemmerer, A. R. 1947. Spectrophotometric studies of the Roe method for the determination
of dehydroascorbic acid. J. BioI. Chem., 167. 781.
-159 -
Bonsnes, R. W., and Taussky, H. H. 1945. On the colorimetr '. determinaHon of creatinine by the Jaffe reactiol
J. Bioi. Chem., 158, 581.
Bopp, F. 1849. Einiges iiber Albumin, Casein, und Fibri.n. Ann '69, 16.
Borglin, N. E. 1947. Excretion of choline in urine. Acta Pharmaco/. Toxico/ .. 3, Suppl. 1.
Borkel, C. 1903. Uber Pepsin - Fibrinpepton. Z. physiol. Chern 38. 289.
Boswell, V. R.. Deonier. M. T., Carolus, R. L., Edmond, J. B., Garrison, O. B., Cochran, H. L., WQodard, 0., Anderson, W. S., Miller, J. C., and Wright, R. E. 1944. Place and season effects on yields and star_ch content of 38
kinns of sweet potatoes. U. S. Dept. Agr. Circ. 714.
Boulet, M., Nelson, J. A., and McFarlane, W. D. 1947. A rapid colorimetric method for the determination of lysine
in protein hydrolysates. Can. J. Research, 25B; 540.
Bourquelot, E., and H~rissey, H. 1899. Sur la dosage du mannose melange ~ d'autres sucres. Compt. rend. .129.
3~
Boyd, M.
J.,
J.
Braco'lllot~ H. 1820. Sur la conversion des mati~res animales en [Link] par Ie moyen de 1'acide
sulfurique. Ann. chirn. et phys., (2). 13, 113.
_ _ . 1827. Recherches sur la fermentation du framage, et sur l'oxide caseeux et l'acide caseique. Ann. chiin.
et phys. (2), 36, 159.
Bradl;y, J. A. 1942. Kjeldahl distillation without absorbing acid. Ind. Eng. Chern., Anal. Ed 14. 705.
Bradstreet, R. B. 1940. A review of the Kjeldahl determination of organic n!trogen. Chern. Revs., 27, 331.
Breddy, L.J., and Jones, J.K. N. 1945. The quantitative estimation of xylose. J. Chern. Soc., 738.
Brederick, H. 1931. Zur Molisch-Reaktion. Ber., 64B. 2856.
_ _ . 1932. Zur Molisch-Reaktion (II. Mitteil.). Ber., 65.3, 1110.
Breslow, D. S., and Hauser, C. R. 1939. Condensations bro~ght about by bases. V. Condensation of the anhydride
with the aldehyde in the Perkin synthesis. I. Am. Chern. Soc 61, 786.
Briant, A. M., Personius,- C. J., and Cassel, E. G. 1945. Physical properties of starch from potatoes of different
culinary quality. Food Research, 10, 437.
Brigg, A. P. 1922. A modification of the Bell-Doisy phosphate method. J. BioI. Chern., 53, 13.
Briggs, D. R., and Hanig, M. 1946. Chemical state of phosphorus in starch as indicated by titration curves on
electrodialyzed starches. Cereal Chern 23. 277.
Brimhall, B., and Hixon, R. M. 1943. Two types of ~olecules in starch. Wallerstein Labs. Co rnmuns. , 6, 95.
Brockmann, H. 1936. Die Isolierung des antirachitischen Vitamins aus Thunfischleberol. Z. physiol. Chem.,
241, 104.
_ _ and Busse, A. 1938. Die Konstitution des antirachitischen Vitamins der Thunfischleber. Z. physiol. Chern.,
256, 252.
Brown, A. H. 1946. Determination of pentose in the presence of large quantities of glucose. Arch. Biochern." 11,
269.
Brown, C. A., and Zerban, F. W. 1943. Physical and Cherpical Methods of Sugar Anl\lysis, 3d ed. John Wiley &
Sons, Inc., !'f ew York.
Brown, W. L. 1944. The determination of pJtenylalaiUne in proteins. J. BioL Chern., 155, 277.
BrUckner,J. 1943. Uber Erkennung uod Be\timmung vetschiedener Hexosen nebeneinander. Z. physiol. Chern.,
277, 181.
Bruhns, G. 1890. {Tber Adenin und Hypoxanthin. Z. physiol. Chern., 14, 533.
Buchanan, O. H., Block, W. D., and Christman, A. A. 1945. The metabolism of the methylated purines. I. The
enzymatic determination of urinary uric acid. J. BioI. Chern., 157, 181.
Bullard, J. G. and others. 1947. Starch ~or Paper Coating, Monograph No.3. Tech. Assoc. Pulp Paper Ind.
Bunzel, H. H. 1908. The rate of oxidation of sugars in an acid medium. Arn. I. Physiol., 21, 23.
Burgen, A. S. V. 1947. Urinary excretion of phosphatases in man. Lancet, 252, 329.
Burian, R. 1904. Diazoaminoverbindungen der Imidazole und der Purinsubstanzen. Ber., 37, 696.
_ _ 1907. Weitere Beitragezur Kenntnis der Diazoaminoverbindungen der Purinbasen. Z. physioL Chern., 51,
425.
Burman, M. 1943. Muscular dysfunctions. Merck Rept., 52, 11.
Bushill, J. H., Lampitt, L. H., and Baker, L. C. 1934. The determination of cystine. Use of the Zeiss photometer.
Biochern. I., 28, 1293.
Butenandt, A. 1936a. Biochemistry of the sterol group. Lecture I. Chemistry of sterols, bile acids, and neutral
saponins. I. Soc. Chern. Ind., 55, 753.
1936b. Biochemistry of the sterol group. Lecture II. Cardil!c poisons and vitamin D. I. Soc. Chern. Ind.,
55, 891.
- 160-
- - ' 1936c. Biochemistry of the sterol group. Lecture III. The sex hormone group. J. Soc. Chern. Ind., 55,990.
Cahours, A. 1858. Recherches sur les acides amid~s. Compt. Tend., 46, 1044.
_ _ 1859. Untersuchungen liber die Aminsauren. Ann., 109, 10.
Caldwell, J. R., and Moyer, H. V. 1935. Determination of chloride. Ind. Eng. Chem., Anal. Ed., 7, 38.
Caldwell, M. L., and Adams, M., in Anderson, J. A. 1946. Enzymes and their Role in Wheat Technology, Chapter
IL Interscience Publishers, New York. '
_ _ , Booher, L. E., and Sherman, H. C. 1931. ~,ystalline amylase. Science, 74, 37.
_ _ , ,Weill, C. E., and Weil, R: S. 1945. Further studies of the essential groups of pancreatic amylase. ,. Am.
Chern. Soc., 67, 1079.
Callan, T. P., and Toennies, G. 1941. Determination of sulfur in organic compounds. Ind. Eng. Chern., Anal. Ed.,
13, 450.
Callow, R. K. 1938. Steroids. Ann. Repts. on ProgTess Chern. (Chern. Soc. London), 35, 281.
Calvery, H. 0., in Schmidt, C. L. A. 1938. Methods of analysis and reactions of the amino acids and proteins.
Chemistry of the Am;no Acids and Proteins. Charles C. Thomas, Publisher, Sprfngfield, Ill.
Cameron, A. T. 1947. The taste sense and the relative sweetness of sugars and other sweet substances. SugaT
Research Foundation, N. Y., Sci. Rept. Ser. No.9.
Carnien, M. N., and Dunn, M. S. 1948. A composite basal medium for the microbiological assay of leucine. J.
BioI. Chern., 173, 137.
_ _ , _ _ , and Salle, A. J. 1947. The effect of carbohydrates on acid production by [Link] acid
bacteria. J. BioI. Cbern., 168, 33.
Cannan, R. K. 1944. The estimation of the dicarboxylic amino acids in protein hydrolysates. J. BioI. Chem., 152,
401.
Cantarow, A., and Trumper, M. 1945. Clinical Biochemistry, 3d ed. W. B. Saun,ders Company, Philadelphia.
Carter, C. E. 1950. Paper chromatography of purine and pyrimidine derivatives of yeast nucleic [Link]. J. Am.
Chern. Soc., 72, 1466.
Carter, H. E., and Hooper, J. R., in Sahyun, M. 1944. Synthesis and isolation of'certain amino acids. Outline of
the Amino Acids and Proteins. Reinhold Publishing Corporation, New York, 94 pp.
Cassidy, H. G., (and twelve collaborators). 1948. Chromatography. Ann. N. Y. Acad. Sci., 49, 141.
Chadwick, A. F., and Pacsu, E. 1941. The rates of ammonolysis of some <X-halogen acids and <X-hal~!;lJl acyl
peptides. J. Am. Chern. Soc., 63, 2427.
Cherbuliez, E., and Schneider, M. L. 1932. Recherches sur la caseine. I. La caseine n'est pas un corps homogene. Etude de son fractionement par Ie chlorure d'ammonium. He Iv. Chim. Acta, 15, 597.
Cheronis, N. D., and Spitzmueler, K. H. 1941. Studies in ammonolysis. L The ammonolysis of halogen fatty a'ids
and preparation of a-amino acids. ,. OTg. Chern., 6, 349.
""
Chlbnall, A. C. 1924. Investigations on the nitrogenous metabolism of the higher plants. Part VL The role of
asparagine in the metabolism of the mature plant. Biochem. J., 18, 395.
- - . 1939. Protein Metabolism in the Plant. Yale University Press, New Haven.
- - , Rees, M. W., and Williams, E. F. 1943. The total nitrogen content of egg albumi,n and other proteins.
Biochem. J., 37, 354.
,
- _ , _ _ , _ _ , and Boyla~d, E. 1940. The glutamic acid of normal and malignant tissue proteins. Biochem.
J., 34, 285.
Christman, A. A., and Ravwitch, S. 1932. The determination of uric acid in human urine. J. BioL Chern., 95,115.
Clark, E. P. 1922. Note on the preparation of mannose. J. BioI. Chem., 51, 1.
Clark, W. M., Zoller, H. F., Dahlberg, A.O., and Weimar, A. C. 1920. Studies on technical casein. n. Grain-curd
casein. Ind. ~ng. Chem., 12, 1163.
Clarke, D. H., and Marney, A. F. 1945. The determination of the free and total cholesterol of plasma with photoelectric colorimeter. J; Lab. Clin. Med., 30, 615.
Clarke, H. T. 1932. The action of sulfite upon cystine. J. BioI. Chern., 97, 235.
_ _ in Gilman, H. 1943. Organic Chemistry, an Advanced Treatise, Vol. II, 2d ed. John Wiley & Sons, Inc!"; New
York, 1079.
Clay, R. C., Cook, K., and Routh, J. L 1940. Studies in composition of human hair. J. Am. Chern. Soc., 62,2709.
Clendenning, K. A., and Wright, D. E. 1945. Composition and specific rotary power oli- starches in relation to
source and. type. Can. J. ReserUch, 23B, 131.
Cohen, P. P. 1939. Microdetermination of glutamic acid. Biocbem. j., 33, 551.
Cohn, E. J., and Hendry, J. L. 1943. Casein. OTg. Syntheses, ColI. Vol. II, 120.
Cohn, E. W., and Brookes, M. H. 1936. The diastatic activity of rat saliva. J. Bioi. Chern., 114, 139.
Cohn, R. 1895. Zur Kenntniss des bei der Pancreasverdauung entstehenden Leucins. Z. physiol. Chern., 20, 203.
- 161 -
Cohn, W. 1950. The anion-exchange separation of ribonucleotides., J. Am. Chem. .soc., 72, 1471.
Cold Spring Harbor Symposia on Quantitative Biology. 1947. Nucl~ic Acids-and Nucleoproteins. Cold Spring
Harbor Symposia Quant. BioI., Vol. 12
Cole, S. W. 1903. On certain colour reactions ,of proteid due to tryptophan. J. Physiol., 30, 311.
Consden, R., Gordon, A. H., and Martin, A.J.P. 1944. Qualitative analysis of proteins: a partition chromatographic method using paper. Biochem. J., '38, 224.
Conway, E. J. 1933. An absorption apparatus for the micro-determination of certain volatile 'substances. n. The
determination of urea and ammonia in body fluids. Biochem. J., 27, 430.
_ _ and Byrne, A. 1933.,' An absorption apparatus for th~ microdetermination of certain volatile substances.
I. The micro-determination of ammonia. Biochem. J., 27~ 419.
Copeland, D. H.! and Salmon, W. D. 1946. The occurrence of neoplasms in the liver, lungs, and other tissues of
rats as a result of prolonged choline deficiency. Am. J. Path., 22, 1059.
Corley, R. C. 1929. Pentose metabolism. Ill. A comparison of the rates of disposal of d-arabinose and I-arabinose
in the rsQbit. J. BioI. Chem.. 82. 269.
Csonka, F. A., and Denton, C. A. 1946. Methionine determination in proteins and foods. J. Bioi. Chem.. 163.329.
Curl, A. L., and Nelson, E. K. 1944. A water-soluble mannan from the seeds of Daubentonia drummondii. J. Am.
Chern. Soc 66. 1227.
Curtius, T., and Goebel, F. 1888. Darstellung von salzsaurem Glycinather aus salzsaurem Glycocoll. J. prakt.
Chem. (2), 37, '159.
Dakin, H. D. 1920. Amino-acids of gelatin. J. Bioi. Chem.. 44, 499.
Daly, M. M., and Mirsky, A. E. 1949. Chromatography of purines and pyrimidines on starch columns. J. Bioi.
.
Chem., 179, 981.
Dann, W. J. t947. The bioassay fol' nicotinic acid. BioI. Symposia, 12. 181.
Darby, W. J., and Day, P. L. 1940. Blood sugar levels in rats receiving the cataractogenic sugars galactose and
xylose. J. Bioi. Chem.. 133, 503.
Davidson, J. N. 1949. Nucleoproteins, nucleic acids and derived substances. Ann. Rev. Biochem.. 18. 155.
Decker, A., and McGinnis, J. 1947. Black pigmentation in feathers of Buff Orpington chicks caused by vitamin D
deficiency. Proc. Soc. Exptl. Bioi. Med.. 66. 224.
Dehn, W. M., Jackson, K. E., and Ballard, D. A. 1932. Identification of common carbohydrates. Ind. Eng. Chem.,
Anal. Ed 4. 413.
Denig~s, G. 1898a. Sur les functions pouvant se combiner au sulfate mercurique. Cas des acetones. Compt.
Tend.. 126. 1868.
_ _ 189Bb. Combinaison, recherche et dosage de l'acetone ordinaire avec Ie sulfate mercurique. Compt. Tend.,
127, 963.
Denny, F. E. 1934a. Improvements in methods of determining starch in plant tissues. Contribs. Boyce Thompson
Inst.. 6, 129.
_ _ 1934b. Starch-determination methods involving solubility in acids., Contribs. Boyce Thompson Inst., 6, 381.
Dickinson, W. P., and Marshall, P. G. 192~. T))e isomeric monohydroxyphenylalanines. Part I. A new synthesis
of the 0- and m-isomerides and a comparison of their properties with those of tyrosine. J. Chern. Soc., 1495.
Dimler, R. J., Davis, H. A., Rist, C. E., and Hilbert, G. E. 1944. Production of starch from wheat and other
cereal flours. Cereal Chern 21. 430.
Dittmer, K. 1949. The synthesis and microbiological properties of ,8-3-thienylalanine, a new anti-phenylalanine.
J. Am. Chem. Soc., 71, 1205.
Dorfman, A., Horwith, M. K., Koser, S. A., and Saunders, F. 1939. The use of the dysentery organism for the
quantitative determination of nicotinic acid. J. BioI. Chem., 128, xx.
Drechsel, E. 1877. Zur Volhard'schen Silberbestimmungsmethode. J. prakt. Chem 15. 191.
Drury, H. F. 1948. Identification and estimation of pentoses in the presence of glucose. Arch. Biochem.. 19. 455.
Dubnoff, J. W. 1941. A micromethod for the determination of arginine. J. Bioi. Chem 141. 711.
Duli~e, W. L., and Raper, H. S. 1930. The tyrosinase-tyrosine reaction. VIL The action of tyrosinase on certain
sul;>stances related to tyrosine. Biochem. J 24. 239.
Dilll, G. 1895. Die Einwirkung von Oxalsaure auf Iqulin. Chem. Ztg 19. 216; Cent., 1895. I, 534.
Dunn, M. S., 1943. The constitution and syntheses of the amino acids. In Schmidt, C. L. A. Chemistry of Amino
Acids and Proteins. Charles C. Thomas, Publisher, Springfield, ill.
_ _ 1949. Casein. Biochem. Prep I, 22.
_ _ , Akawie, S., Yeh, H. L., and Martin, H. 1950. Urinary excretion of amino acids in liver disease. J. Clin.
Invest., 29. 302.
_ _ , Butler, A. W., amd Frieden, E. H. 1941. Quantitative investigations of amino acids and peptides. VI. The
function of carbonate in the synthesis of glycine from ammonium carbonate. J. Phys. Chem., 45. 1123.
- 162-
, _ , Carnien, M. N., Shankman, ;:;., and Block, H. 1947. Urinary excretion of twelve amino acids by normal
male and female subjects measured microbiologically. Arch. Biochern., 13, 207.
_ _ and Porush, I. 1939. Quantitative investigations of amino acids and peptides. V. The function of iodine in
amino nitrogen analyses by the nitrous acid method. J. BioI. Chern., 127, 261.
_ _ and Rockland, L. B. 1947. The preparation and criteria of purity of the amino acids. Advances in Protein
Chern., .3, 295.
_ _ and Schmidt, C. L. A. 1922. The influence of position and of temperature upon the reaction of aliphatic
amino nitrogen with nitrous acid. J. BioI. Chern., 5.3, 401.
Dunning, J. W., and Lathrop, E. C. 1945. The saccharification of agricultural residues. A'continuous process.
Inri. Eng. Chern., .37, 24.
Earle, J. R., and Milner, R. T. 1944. Improvements in the determination of starch in corn and wheat. Cereal
Chern., 21, 567.
Eckhardt, R. D., Cooper, A. M., Faloon, W. W., and Davidson, C. S. 1948. The urinary excretion of am,ino acids
in man. Trans. N. Y. A cad. Sci 10, 284.
I
Edman, P., Hammarsten, E., Low, B., and Reichard, P. 1949. Partition chromatographic separation of adenine
and guanine. J. BioI. Chern., 178, 395.
Eegriwe, E. 1933. Reaktionen und Reagenzien zum Nachweis organischer Verbitldungen. II. Z. anal. Chern., 95,
323.
_ _ 1937. Reaktionen und Reagenzien zum Nachweis organischer Verbindungen. IV. Z anal. Chern., 110, 22.
Ehrensvard, G., Sperber, E., Salu1ste, E., Reio, L., and Stjernholm, R. 1947. Metabolic connection between
proline and glycine in the amino acid utilization of Torulopsis utilis. J. Bioi. Chern., 169, 759.
Ehrlich, F. 1908. Uber die Spaltung racemischer Aminosauren mittels Hefe. II. Mitteilung. Biochem. Z, 8, 438.
_ _ 1917. Die Pektinstoffe! ihre Konstitution und Bedeutung. Chern. Ztg., 41, 197.
Eisner, A., and Wagner, E. C. 1934. Stability of aqueous solutionsof boric acid used in the Kjeldahl method.
Inri. Eng. Chern., Anal. Eri., 6,'473.
van Ekenstein, W. A., and Blanksma, J. J. 1910. Ther das (J)-Oxymethyl-furfurol als Ursache einiger Farbreaktionen der Hexosen. Ber., 4.3, 2355.
Elsdon, G. D. 1926. Edible Oils and Fats. Ernest Benn, Ltd., London.
Elvehjem, C. A., Madden, R. J., Strong, F. M., and Woolley, D. W. 1937. Relation of nicotinic acid and nicotinic
acid amide to canine black tongue. J. Am. Chern. Soc., 59, 1767.
Embleton, J., and Collings, A. J. 1947. Transference of vitamin D from the female rat to her young. Nature, 159.
341.
Emmerling, O. 1897. Die Zersetzung von Fibrin durch Streptococcen. Ber., .30. 1863.
_ _ 1902. lIber die Eiweiss-spaltung durch Papyotin. Ber .35. 695.
Engel, R. W. 1942. Modified methods for the chemical am biological determination of choline. J. BioI. Chern
144, 701.
'.
Erlenmeyer, E., Jr. 1886. Uber die PhenYlglycidsaure. Ber., 19, 2576.
1887. Zur Kenntniss der Phenyl-a- und der Phenyl-CX,,B-oxypropionsaure. Ber., 20, 2465.
1892. Zur Kenntniss der Phenylbrenztraubensaure und der Phenylglycidsaure. Ann 271. 137.
1893. Uber die Condensation der Hippursaure mit phtalsaureanhydrid und mit Benzaldehyd. Anp., 275, 1.
1897. Ther eine Anscheinend allgemeine Reaction der CX-aminosauren von der Formel RCH(NH2)COOH.
Ber., .3P. 2896.
1899. Zur Kenntniss der CX-Amidosauren. Anno, .307, 70.
_ _ 1904a. Zur Kenntniss der ~Amidosauren. Ann., .3.37, 205.
_ _ 1904b. Uber Azlactone und ihre Unwandlungsprodukte. Ann., .3.37, 265.
Ershoff, B. H. 1946. Studies on the nutritive value of lactose and galactose with the single-food choice method.
Am. J. Physioi.,. 147, 13.
Etheredge, M. P. 1944. Report on starch in raw and baked cereals. J. Assoc. Olfic. Agr. Chemists, 27, 404.
Evans, W. L. 1929. The mechanism of carbohydrate oxidation. Chern. Revs., 6, 281.
- - . 1942. Less familiar aspects of carbohydrate chemistry. Chern. Revs!, .31, 537.
_ _ , Nicoll, W. D., Strouse, G. C., and Waring, C. E. 1928. The mechanism of carbohydrate oxidation. IX. The
action of copper acetate solutions on glucose, fructose, and galactose. ). Am. Chern. Soc., 50, 2267.
Fabinyi, R. 1900a. Verfahren zur Darstellung von Farbstoffen aus Salicylaldehyd und Ketonverbindungen. Deut.
Reich Pat. 110,520. Cent., II, 301.
- - . 19001>. Verfahren zur Darstellung eines neuen Seidenfarbstoffes. Deut. Reich Pat. 110,521. Cent., II, 302.
Fearon, W. R. 1942. Detection of lactose and maltose by means of methylamine. Analyst, 67. 130.
_ _ 1944. The detection and estimation of uric acid by 2,6-dichloroquinone-chloromide. Biochern. ,., .38, 399.
- 163-
_ _ , and Mitchell, D. M. 1932 The nitro chromic acid reaction for the detection of primary and secondary alc~
hols with speCial reference to saccharides. Analyst. 57, 372
Fehling, H. 1849. Die quantitative Bestimmung von Zucker und Stiirkmehl mittelst Kupfervitriol. Ann., 72, 106.
1858. Die quantitative Bestimmung von Zucker. Ann., 106, 75.
von Fellenberg, T. '1947. Separation of sugars. Mitt. Gebiete LebenSln. flyg 38. 265; Che>n. Abstracts, 42.
3288 (1948).
Ferger, I\L A., and du Vigneaud, V. lQ_49. The antiphenylalanine effect of ,B-2-thienylalanine for the rat. }. BioI.
Cbem.. 179. 61.
Fieser, L. F., and Fieser, M. 1949. Natural Products Related to Phenanthrene, 3d ed. Reinhold Publishing
Corporation, New York.
_ _ and _ _ 1944. Organic Chemistry. D. C. Heath and Company, Boston, 351.
Firstenberger, B. G. 1943. The data, design, and specifiCjations for a plant to produce xylose from corn stalks.
Iowa State Coli. }. Sci., 18. 27.
Fischer, E. 1884. Verbindungen des Phenylhydrazins mit den Zubkerarten. Ber 17. 579.
1887. Verbindungen des Phenylhydrazins mit den Zuckerarten. n. Ber 20. 821.
_ _ 190L Uber die Hydrolyse des Caseins durch Salzsiiure. Z. physiol. Cbem, 33, 151.
_ _ 1906. Synthese von Pol,ypeptiden. XV. Ber 39, 2893.
_ _ and Foumeau, E. 190L tiber einige Derivate des Glykocolls. Ber.. 34. 2868.
_ _ and Hirschberger, J. 1888. Uber Mannose. I. Ber., 21, 1805.
_ _ and _ _ 1889a. Uber Mannose. IL Ber., 22, 365.
_ _ and _ _ 1889b., Uber Mannose, IV. Ber 22, 3218.
_ _ and Mouneyrat, A. 1900. Spaltung einiger racemischer Aminosauren in die optisch aktiven Komponenten. IV.
Ber., 33, 238~.
_ _ and Schoeller, W. 1907. Synthese von Polypeptiden. XXIL Derivate des I~Phenylalanin. Ann.. 357, 1.
Fischer, lL 1909. Zur Frage der Bindung der Purinbasen 1m Nucleinsanremolekiil. Z. physiol. Chern., 60, 69.
Fischer, M. M., and Hooker, M. O. 1918. A note on the colloid chemistry of Fehling's sugar test. }. Lab. Clin.
Med., 3, 368.
Fiske, C. H., and Subbarow, Y. 1925. The colorimetric determination of phosphorus. }. Bioi. Chem., 66, 375.
Fletcher, J. P., Best, C. H., and Solandt, O. M. 1935. The distribution of choline. Bioche>n. J.. 29. 2278.
Flock, E. V., and Bollman, J. L. 1948. Alkaline phosphatase in the intestinally~ph of the rat. }. Bioi. Chem..
175. 439.
Foldes, F. F and Murphy, A. J. 1946a. Distribution of cholesterol, cholesterol esters and phospholipid phosphorus in normal blood. Proc. Soc; Exptl. Bioi. Med., 62, 215.
_ _ and _ _ 1946b. Distribution of cholesterol. cholesterol esters and phospholipid phosphorus in blood in
thyroid disease. Pm c. Soc. Exp tl. BioI. Med., 62, 218.
Folin, O. 1903. The acidity of urine. Am. J. Physiol., 9, 25.
1905. Approximately complete analyses of thirty "normal" urines. Am. J. Pbysiol., ]3, 45.
1910. On the preparation of cystin. J. BioI. Chern., 8. 9.
1914. On the determination of creatinine ansi creatine in urine. ]. Bioi. Chem., 17. 469.
1922a. A system of blood analysi~ Supp.
A new colorimetric method for the determination of the
amino-acid nitrogen in blood. ]. BioI. <sEern., 51. 377.
_ _ 1922b. A colorimetric determination of the amino-acid nitrogen in normal urine. J. BioI. Chem., 51, 393.
_ _ 1933. Standardized methods for the determination of uric acid in unlaked blood and in urine. ]. Bioi.
Chem., 101, 111.
_ _ . 1934. The preparation of sodium tungstate free from molybdate, together with a simplified process for the
preparation of a correct uric acid reagent (and some comments). }. Bioi. Chern 106, 311.
_ _ and Bell, R. D. 1917. Applications of a new reagent for the separation of ammonia. ]. Bioi. Cbem., 29. 329.
_ _ and Berglund, H. 1922. A colorimetric method for the determination of sugars in normal human urine. J.
BioI. Chern 51, 209.
_ _ and Ciocalteu, V. 1927. On tyrosine and tryptophane determinations in proteins. J. Bioi. Chern., 73, 627.
__,_and Denis, W. 1912 Tyrosine in proteins as determined by a new colorimetric method. J. BioI. Cbem..12. 245.
_ _ and _ _ 1913a. A new (colorimetric) method for the determination of uric acid in blood. J. BioI. 'Chern.,
13. 469.
_ _ and _ _ 1913b. On the colorimetric determination of uric acid in urine. J. Biol. Chern., 14. 95.
_ _ and _ _ 1914. The quantitative determination of albumin in urine. J. BioI. Chem., 18. 273.
_ _ and Looney, J. M. 1922. Colorimetric methods for the separate determination of tyrosine, tryptophane, and
cystine in proteins. }. Bioi. Chern 51, 421.
III
-164 -
_ _ and Marenzi, A. D. 1929a. Tyrosine and tryptophane determinations in one--tenth gram of protein. }. BioI.
Chern., 83, 89.
_ _ and _ _ 1929b. An improved colorimetric method for the determination of cystine in proteins. J: Bioi.
-165 -
Gottschalk, R. G.
1945. Microdetermination of acid phosphatase. A study of the reaction of King and Armstrong.
Biochirn. et Biophys. Ada. 2, 582.
\
Grassman, W., and von Amim, K. 1934. liber die Reaktic:m des Nihhydrins und Isatins mit Prolin und Oxyprolin.
_ _ and _ _ 1935. Uber neue Farbreaktionen des pyrrolidins und Prolins. n. Ann., 519, 192
Graziani, F. 1915. Anhydrides and arnines of Q',.amino acids. Aui accad. nazi. Lincei, 24. I, 822, 936; Chern.
Abstracts, 9, 3065.
Greenbaum, F. R. 1935. Method of separation of large amounts of tyrosine from cystine. Am, ]. Phafm., 107, 162.
Greenwald, L 1925. Chemistry of Jaffe's reaction for creatinine. IL Effect of substitution in the creatinine
I
molecule and a possible formula for the red tautomer. ]. Arn. Chern. Soc., 47, 1443.
_ _ 1928. The chemistry of Jaffe's reaction for creatinine. V. 'The isolation of the red compound. ]. Bioi.
Chern., 80, 103.
.
I
_ _ and Gross, J. 1924. The chemistry of Jaffe's reaction for creatinine. A red tautomer of creatinine picrate.
J. Bioi. Chern .. 59, 601.
Greenwood, D. A., Kraybi11, lL R., Feaster, J. F., and Jackson, J. M. 1944. Vitamin retention in processed meat.
lnd. Eng. Chern., .36, 922.
Griffith, W. lL 1926. Benzoy1ated amino acids in the animal organism. III A method for the determination of
hippuric acid and a study of the synthesis of hippuric acid in rabbits. J. Bioi. Chern., 69, 197.
_ _ 1941. The nutritional importance of choline. ]. Nutrition, 22. 239.
_ _ and Wade, N. J. 1940. Choline metabolism. Il The interrelationship of choline, cystine, and methionine in
the occurrence and prevention of hemorrhagic degeneration in young rats. ]. Bio'f. Chem., 1.32, 627.
Guest, G. lL 1939. A note on the colorimetric determination of proline: Can.]. Research, 17B, 143.
_ _ and MacFarlane, W. D. 1939. A study of the colorimetric determination of pyrrole With isatin and the application of the method to biological materials. Can. J. Research, 17B, 133.
Gunness, M., Dwyer, L M., and Stokes, J. L. 1946. Microbiological methods for the determination of amino acids.
m. Extension of the uniform assay method for the ten essential amino acids to include tyrosine. J. Bioi.
Chem., 16.3, 159.
Gyorgy, P. 1950. Vitamin Methods, Vol. I. Academic Press, New York.
van der Haar, A. W. 1920. Nachweis zur Trennung und Bestimmung der reinen und aus Glukosiden usw. erhaltenen
Monosaccharide und Aldehydsauren. Verlagsbuchhandlung Gebriider Borntraeger, Berlin.
Habermann, J., and Ehrenfeld, R. 1902. Eine quantitative Methode zur Trennung des Leucins und Tyrosins. 2.
physiol. Chern., .37, 18.
Hagedorn, H. C., and Jensen, B. N. 1923a. 2ur Mikrobestimmung des Blutzuckers mittels Ferricyanid. Biochern.
2., 135,46.
_ _ and _ _ 1923b. Die Ferricyanidmethode zur Blutzuckerbestimmung. II. Biochem. 2., 137, 92.
Haines, W. S., Pond, G. P., and Webster, R. W. 1920. An improved test for the detection of glucose, especially
in urine. J. Am. Med. Assoc., 74, 301.
Hale, E. B., Davis, G. K., and Baldwin, H. R. 1942 The distribution of nicotinic acid in feeds. ]. Biol. Chem.,
146, 565.
Hall, W. K., Bowles, L. L., Sydenstricker,\V"P., and Schmidt, H. L., Jr. 1948. Cataracts due to deficiencies of
phenylalanine and of histidine in the rat. A comparison with other types of cataracts. J. Nutrition, 36, 277.
_ _ , Sydenstricker, V. P., and Rawls, K. 1948. Factors affecting the excretion of metabolites of phenylalanine
and tyto sine in alkaptonuria. Federation Proc., 7. 289.
Hamilton, P. B. 1945. Gasometric determination of glutamine amino acid carboxyl nitrogen in plasma and tissue
filtrates by the ninhydrin-carbon dioxide method. ]. Bioi. Chern., 158, 375.
_ _ and Van Slyke, D. D. 1943. The gasometric determination of free amino acids in blood filtrates by the
ninhydrin-carbon dioxide method. ]. Bioi. Chern., 150, 231.
Hammarsten, O. 1873-1874. Upsala Liikareforen. Forh. 9, 363. Quoted by [Link], 0., and Hedin, S. G.,
Textbook of Physiological Chemistry, 7th ed. trans. by J. A. Mandel, Jolul Wiley & Sons, Inc., New York (1914).
_ _ 1883. 2ur Frage, ob des Caseinein einheitlicher Stoff sei. 2. physiol. Chem .. 7, 227.
Handler, P. 1947a. The biochemical defect underlying the nutritional failure of young rats on diets containing
excessive quantities of lactose or galactose. ]. Nutrition, 33, 221.
_ _ 1947b. The determination of choline in biological materi~. Bioi. Symposia. 12, 361.
_ _ and Perlzweig, W. A. 1945. Detoxication mechanisms. Ann. Rev. Biochem., 14, 617.
Hanes, C. S. 1929. An application of the method of Hagedorn and Jensen to the determination of larger quantities
of r~ducing sugars. Biochem. ]., 23, 99.
- - . 1936. The determination of starch in plant tissue, with particular reference to the apple fruit. Biochem.
J.
30, 168.
- 166-
_ _ 1937. The action of amylases in relation to the structure of starch and its metaboli~m in the pI apt . Parts
I-Ill. New Phytologist, 36, 101.
.... Hanke, M. E. I!? 48. Conditions necessary for complete decarboxylation of pure L-Iysine and L-tyrosine by amino
acid decarboxylases. Federation Proc., 7, 158.
Hanke, M. T. 1925. The quantitative estimation of tyrosine and histidine in protein. A method for estimating
tyramine in protein-containing mixtures.
BioI. Chern 66. 475.
1928. Determination of the tyrosine content of protein. J. BioI. Chern.. 79, 587.
Hann, R. M., and Hudson, C. S. 1944. The action of copper sulfate on phenylosazones of the sugars. Phenyl-Dglucosotriazole. }. Arn. Chern. Soc., 66. 735.
Hanus. J. 1901. Die Anwendung von lodmonobromid bei der Analyse von Fetten und Olen. Z. Untersuch. Nahr.
U. Genussm., 4, 913.
Harding, T. S. 1922 The discovery and production of arabinose. Sugar, 24, 656.
- - ' 1923a. History of xylose. its discovery and methods of preparation. Sugar. 25, 124.
_ _ 1923b. History of mannose. ,its discovery and method of preparation. Sugar. 25, 583.
Harding, V. J., and MacLean, R. M. 1915. A colorimetric method for the estimation of amino acid <X-nitrogen.
J. BioL Chern 20. 217.
I
_ _ and _ _ 1916. The ninhydrin reaction with amines and amides. J. BioI. Chern., 25, 337.
Harding, V. S., aM Warneford, F. H. S. 1916. The ninhydrin reaction with amino acids and ammonium salts. }.
Bioi. Chern., 25, 319..
Harington, C. R., and McCartney, W. 1927. Note on the Erlenmeyer amino-acid synthesis. Biochern. I., 21. 852.
Harries, C., and Weiss, M. 1903. Uber das Hydantoin und Isomerien bei den Methrlhydantoinen. Ann., 327, 355.
Harris, L. J. 1922-1923. On the exis!ence of an unidentified sulphur grouping in the protein molecule. 'Part IL
On the estimation of cystine in certain proteins. Proc. Roy. Soc. (London), 94B, 44L
Harris, R. H.. and Jesperson, E. 1946a. A study of the effect of various factors on the swelling of certain cereal
starches. I. Colloid Sci., 1, 479.
_ _ and _ _ 1946b. Factors affecting some physicochemical properties of starch. Food Research. 11. 216.
Harris, W. F. 192L Anthony von Leeuwenhoek, the first bacteriologist. Sci. Monthly. 12, 150.
Harrison, G. A 1947. Chemical Methods in Clinical Medicine, 3d ed. J. and A. Churchill Ltd., London.
Hart. E. B . Kline. O. L . and Keenan, J. A. 1931. A ration for the production of rickets in chicks. Science, 73.
.f.
710.
Hartmann, R G., and Hillig. F. 193L The determination of starch in flour by diastase-acid hydrolysis. J. Assoc.
Ollie. Agr. Chemists, 14. 112.
Harvey, D. G. Miller, E. J . and [Link]. W. 1941. The Adamkiewicz, Hopkins and Cole, and Rosenheim tests
for tryptophan. J. Chern. Soc 1,53.
Hassid, W. Z. 1936. Determination of reducing sugars and sucrose in plant materials. Ind. Eng. Chern., Anal.
Ed., 8, 138.
1937. Determination of sugars in plants. Ind. Eng. Chem., Anal. Ed., 9, 228. "
1943. The molecular constitution of starch and the mechanism of its formation. Quart. Rev. BioI., 18, 311.
1945a. The molecular constitution of starch. Wallerstein Labs. Cornrnuns 8, ~4.
1945b. Recent advances in the molecular constitution of starch and glycogen. Federation Proc., 4, 227.
_ _ , McCready, R. M., and Rosenfels, R. S. 1940. Determination of starch in plants. Ind. Eng. Chern Anal.
Ed., 12, 142
Hawk, P. B., Oser, B. L., and Summerson, W. lL 1947. Practical Physiological Chemistry, 12th ed. The Blakiston Company, Philadelphia.
Hawkins, J. A. 1929. A micro time method for determination of reducing sugars, and its application to analysis
of blood and urine. I. Bioi. Chern., 84, 69.
_ _ and Van Slyke, D. D. 1929. A time method for determination of reducing sugars, and its application to
analysis of blood and urine. }. Bioi. Chern . 81, 459.
Haworth, W. N. 1933. The constitution of ascorbic acid. }. Soc. Chern. Ind., 52, 482
_ _ 1946. Starch. J. Chern. Soc., 543.
and Hirst, E. L. 1933. Synthesis of ascorbic acid.' J. Soc. Chern. Ind., 52, 645.
_ _ , Peat, 8., and Sagrott, P. E. 1946. A new method for the separation of the amylose and amylopectin components of starch. Nature, 157, 19.
Heberlein, D. G., and Clifcorn, L. E. 1944. Vitamin content of dehydrated foods. Effect of packaging and storage.
Ind. Eng. Chern., 36, 912
Heilbron, I. M. 1936. Recent developments in the chemistry of the sterols and related compounds. ,. Soc. Chern.
Ind., 55, 129T.
Henderson, L. J., and Palmer, W. W. 1914. On the several factors of acid excretion. J. Bioi. Chem.. 17, 305.
-167 -
Henileberg, W.
1903. I Zur KelUltniss der Milchsiiuren Bakt;rien. Paul and Parey, Berlin.
Henriques, V" and SOrensen, S. P. L. 1910. Uber die quantitative Bestimmung der Aminosiiuren, Polypeptide und
der Hippursiiure im 'Harne durch FormolUtration. Z. physiol. Chem., 64, 120.
Hepburn, J. S., and Kotlikoff, R. 1943. Comparative study of certain methods for the determination of serum
cholesterol. Rev. Gastroenterol., 10, 170.
Hershberg, E. B. 1936. A new type of laboratory stirrer. Ind. Eng. Chem Anal. Ed., 8, 313
Herstein, B. 1911. The centenary of glucose and the early history 01 starch. Ind. Eng. Chern.. 3, ISS.
Hess, W. C. 1933. The gasometric determination of cysteine and cystine. J. Biol. Chern., 103, 449.
_ _ and Sullivan, M. X. 1943. The cysteine, cystine, and methionine content of proteins. J. Bioi. Chem., 151,
635.
Hestrin, S., and Mager, J. 1947. Color test for identification of glucose. Ind. Eng. Chem., Anal. Ed., 19, 1032.
Heyl, F. W., Caryl, C. R., and Staley, J. F .. 1914. Standardization of commercial papain., Am.}. Pbarm., 86, 542.
Highet, D. M., and West, E. S. 1942. A procedure for the determination of ascorbic acid based upon the use of a
'standardized solution of 2,6-dichlorophenol indophenol in xylene. J. BioI. Chem., 146, 655.
Hilbert, G. E., and MacMasters, M. M. 1946. Pea starch, a starch of high amylose content. }. Bioi. Chem., 162,
229.
Hiller, A., Greif, R. L., and Beckman, W. W. 1948. Determination of protein in urine by the biuret method. J.
Bioi. Chem., 176. 1421.
Hinkel, F. C., and Sherman, H. C. 1907. Experiments upon Barfoed's acid cupric acetate solution as a means of
distinguishing glucose from maltose, lactose and sucrose. J. Am. Chern. Soc., 29. 1744.
Hinterberger, F. 1849. Untersuchung des Ochsenhorns. Ann., 71, 70.
Hirst, E. L. 1933. The structure of ascorbic acid. J. Soc. Chern. Ind., 52, 221.
_ _ , Jones, J. K. N., and Woods, E. A. 1947. The quantitative determination of galactose, mannose, arabinose,
and rhamnose. J. Chern. Soc 1048.
Hitchings, G. H., and Fisk~, C. H. 1941. The determination of the purines. J. Bioi. Chern., 140, 491.
Hixon, R. M., and Rundle, R. E. 1944. Recent advances in starch chemistry. In Alexander, J., Colloid Chemistry,
Theoretical and Applied, Reinhold Publishing Corporation, New York, 5, 667.
_ _ and Sprague, G. F. 1942. Waxy starch of maize and other cereals. Ind. Eng. Chem., 34, 959.
Hlasiwetz, H., and Habermann, J. 1873. Uber die Proteinstoffe. Ann.. 169, 150.
Hobson. W. 1939. Urinary output of creatine and creatinine associated with physical exercise, and its relationship to carbohydrate metabolism. Biochem. J. 33, 1425.
Hodge, H. C. 1944. Acute toxicity of choline hydrochloride administered intraperitoneaUy to rats. Proc. Soc.
Exptl. Bioi. Med., 57, 26.
Hodges, R. C., Sperry, W. M., 'and Andersen. D. H. 1943. Serum cholesterol values for ~nfants and children. Am.
J. Diseases Child7en, 65, 8~8.
Hodson, A. Z. 1945. The use of Neurospora for the determination of choline and biotin in milk products. }. Bioi.
Chem 157, 383.
Hoffman, W., S. 1937. A rapid photoelectric method for the determination of glucose in blood and 'lJ'ine. J. Bioi.
Chern., 120, 51.
Hoffman, R. 1853. Reaktion auf Leucin und ~yrosin. Ann., 87, 123.
Hoffmeyer. J; 1944. Cholesterol content in rabbit serum. Acta Physiol. Scand.. 7. 313.
Holiday, E. R., and Johnson, E. A. 1949. Location of paper chromatogram spots of purine and pyrimidine derivatives in ultra-violet light. Nature. 163. 216.
Hopkins, C. Y. 1934. Polarimetric estimation of starch. Can. J. Re_search, 11, 751.
Hopkins. F. G., and Cole, S. W. 1901a. Adamkiewicz's proteid reaction, chemistry of glyoxylic acid. Proc. Roy.
Soc. (London;, 68, 21.
- - and _ _ 190 lb. A contribution to the chemistry of proteids. Part L J. Physiol., 27, 418.
_ _ and _ _ 1903. A contribution to the chemistry of proteids. Part IL J. Physiol., 29. 451.
Hopkins, R. H. 1946. The action of the amylases. Advances in Enzymol., 6, 389.
Hom, M. J., Jones, D. B., and Blum, A. E. 1946. Colorimetric determination of methionine in proteins and foods.
J. Bioi. Cbem., 166, 313.
Horowitz, N. H., and Beadle, G: W. 1943. A microbiological method for the determination of choline by use of a
mutant of Neurospora. J. Biol. Chern 150. 325.
Horton, P. M. 1921. Preparation of mannose from ivory-nut shavings. Ind. Eng. Chem., 13, 1040.
Hotchkiss, R. D. 1948. The quantitative separation of purines, pyrimidines and nucleosides by paper chromatography. J. Bioi. Chem., 175. 315.
Huber, C. 1867. VorIiiufige Notiz iiber einige Derivate des Nicotins. Ann 141, 271.
Hubi. B. 1884. Eine allgernein anwendbare Methode zur Untersuchung der Fette. Dinglers poly tech. J 253, 281.
_ 168 _
Hlldson, C. S., and Harding, T. S. 1917. 1}le preparation of xylose. J. Am. Chern. Soc., 39, 1038
... _ _ and _ _ . 1918. The preparation of xylose from corn cobs. J. Am. Cbern. Soc., 40, 1601.
_ _ and Jackson, E. L. 1934. Improvements in the preparation of crystalline d-mannose. J. Am. Cbern. Soc.,
56, 958.
_ _ and Sawyer, H. L. 1917. The preparation of pure crystalline mannose and a study of its mutarotation. J.
Am. Cbern. Soc.. 39. 470.
Hunter, G. 1936. A test for thymine, with observations on the keto-enolic type of diazo-test. Biocbem. J., 30,745.
_ _ and Hlynka. I. 1937. Note on the preparation of purines and pyrimidines from nucleic acid. Biocbern. J., 3}.
486.
Ihl, A. 1885. Phenole als Reagenzien fiir Kohlehydrate. Chern. Ztg., 9. 281; Cent., 1885, 761.
_ _ 1887. Farbenreaktionen des Riibenzuckers. Chem. Ztg., 11, 2.
_ _ and Pechmann, A. 1884. Phenol und Diphenylamin als Reagenzien fiir Kohlehydrate. Ber. [Link]. Ges. z.
Ford. d. chm. Ind., 106; Cent., 1885, 761.
[kawa, Mo, and Niemann, C. 1949. A spectrophotometric study of the behavior of carbohydrates in seventy-nine
per cent sulfuric acid. J. BioI. Chern.. 180, 923.
Isbell, H. S. 1941. Preparation of d-mannose. J. Research, Natl. Bur. Standards. 26. 47.
Jamieson, G. S. ~943. Vegetable Fats and Oils, 2d ed., Reinhold Publishing Corporation, New York.
Jirak, L. W. 1935. Control of the washing-out process in potato starch manufacture. Z. Spiritusind., 58, 81;
Chern. Abstracts, 30. 2791 (1936).
Johnson, A. H., and Green, J. R. 1930, Modified methyl red and sodium alizarin sulfonate indicators. Ind. Eng.
Chern., Anal. Ed., 4 2.
Johnson, B. C. 1945. The microbiological determination of nicotinic acid, nicotinamide, and nicotinuric acid.
J. 'BioI. Chern 159, 227.
_ _ , Hamilton, T. -5., and Mitchell, H. H. 1945a. The effect of choline intake and environmental temperature on
the excretion of choline from the human body. J. Bioi. Chern., 159, 5.
_ _ , _ _ , and _ _ 1945b. The excretion of nicotinic acid, nicotinamide, nicotinuric acid. and N '-methylnicotinamide by normal individuals. J. Bioi. Chern .. 159, 231.
Johnson, T. Eo 1943. A sensitive color reaction for the detection of uracil and cytosine. Science, 98. 90.
_ _ and Clapp, S. H. 1908-1909. Researches on pyrimidines: the action of diazobenzene sulfonic acid on thymin,
uracil and cytosine. J. Bioi. Chem., 5, 163.
_ _ and Coghill, R. D. 1925. Researches on pyrimidines. The discovery of 5-methylcytosine in tuberculinic
acid, the nucleic acid of the tubercle bacillus. J. Arn. Chern. Soc., 47, 2838.
_ _ and Daschavsky, P. G. 1924-1925. Researches on amines. X. The formation of tyramine by decarboxylation
of tyrosine produced from silk. J. BioI. Chem., 62, 725.
_ _ and Kohmann, E. F. 1915a. L Determination of the structure of nitrotyrosine. J. Am. Chem. Soc., 37, 1863.
_ _ and _ _ 1915b. IL Synthesis of 3,5-dinitrotyrosine. J. Am. Chem. Soc., 37:,2194.
Jones, D. a, and Moeller, O. 1928. Some recent determinations of a~partic and glutamic acids in ,various proteins.
J. BioI. Chem., 79. 429.
Jonnard, R. 1945. Determination of total nitrogen in proteins and their hydrolyzates. Ind. Eng. Chem., 'Anal. Ed.,
17, 246.
Jukes, T. H., Ilnd Sanford, T. D. 1939. The vitamin D requirements of young turkeys. J. Nutrition, 18. 71.
Kapeller-Adler, R. 1932. Uber eine neue Reaktion zur qualitativen und quantitativen Bestimmung des Phenylalanins. Biochem. Z., 252, 185.
_ _ , 1934, tiber eine stufenphotometrische Bestimmung des Histidins. Biochem. Z .. 271, 206.
Kapfhammer, J., and Eck, R. 1927. I-Oxypro1in und I-Prolin. Ihre Darstellung aus Eiweiss mit Hilfe der
Reineckesaure. Z physiol. Chern., 170, 294.
KalTer, P., and Keller, R. !943. Die Natur der Waser'schen spezifischen Farbreaktion auf a-Aminosauren. Helv.
Chim. Acta, 26. SO.
Kassell, B., and Brand, E. 1938. The photometric determination of cystine, cysteine, ascorbic acid, and related
compounds with phosphotungstic acid. J. Bioi. Chern., 125. 115.
Kawai, K. 1916. p-Hydroxyphenylethylamine from tyrosine. Jap. Pat. 30,334; [Link], 11, 2027 (1917).
Kelling, A. H. 1944. Starch from com. U. S. Pats. 2,324,849, and 2,324,-850.
Kendrick, A. B., and Hanke, M. E. 1937. The use of iodine and other modifications in the Van Slyke manometric
amino nitrogen method. J. Bioi. Chem., 117, 161.
_ _ and _ _ 1940. The effect of iodine and mercury on amino nitrogen values with nitrous acid. J. BioI.
- 169-
_ _ and _ _ 1885. Untersuchungen fiber Milchzucker und Galactose. Ann 227. 221.
Kermack. W. O. 1946. Recent advances in science: biochemistry. Science Progress. 34. 778.
Kerr. R. W. and Trubell. O. R. 1943. Spectrophotometri'C analysis of starches. Paper Trade I . 117. 25.
Kiermayer. J. 1895. Uber ein Furfurolderivat aus Levulose. Chern. Ztg '19. 1003; Cent 1895. II, 214.
Kimbrough. W. D. 1942. Production of sweet potatoes for starch or other purposes. Louisiana Agr. Expt. Stll.
Bull. 348.
King. C. G. 1936. Vitamin C. ascorbic acid. Physiol. R~vs 16. 238.
_ _ 1941. Chemical methods for determination of vitamin C. Ind. Eng. Chern Anal. Ed;. 13. 225.
Ki~. E. J. 1932. The colorimetric determination of phosphorus. Biochern. I .. 26. 292.
Kingsbury. F. B . Clark. C. P Williams. G and Post. A_. L. 1926. The rapid determination of albumin in urine.
J. Lab. Clin. Mea.. 11. 981.
Kingsley. G. R and Schaffert. R. R. 1949. Determination of free and total cholesterol by direct chloroform extraction. J. Bioi. Chern 180. 315.
Kirk. E. '1948. The acid phosphatase concentration of the prostatic fluids in young. middle-aged and old individuals.
Gerentol 3. 98.
Kitson, R. E., and Mellon, M. G. 1944a. Colorimetric determination of phosphorus as molybdivanadophosphoric
acid. Ina. Eng. Chern., Anal. Ed., 16. 379.
_ _ and _ _ 1944b. Further studiesof the molybdenum blue reaction. Ina. Eng. Chern., Anal. Ea., 16,466.
Kjeldahl, J. 1883. Neue Methode fiir Bestimmung des Stickstoffs in organischen 'Korpern. Z. anal. Chern., 22, 366.
Klein, G., and Linser, H. 1932a. Colorimetrische Methode z_ur quantitativen Bestimmung von Glykokoll. Z. physiol.
z.,
11, 356.
_ _ and Blanco, J. G. 1925. Uber die Acetylierung von Aminosiiuren im Tierkorper. Z. physiol. Chern., 146, 267.
Koch, F. 1886. Experimentelle Priifung des Holzgummi und des sen Verbreitung im Pflanzenreiche. Pharm. Z.
Russland, 25, 619; Cent., 1886, 932.
Koch, F. c., and McMeekin, T. L. 1924. A new direct nesslerization micro-Kjeldahl method and a modification of
the Nessler-Folin reagent for ammonia. I. Arn. Chern. Soc., 46, 2066.
Koenigsfeld, H. 1912. Untersuchungen Uber die physikalisch-chemischen Grundlagen der Seliwanoffschen
Lavulosereaktion. Biochern.
38, ~10.
Koettstorfer, J. 1879. Neue Methode zur Untersuchung der Butter auf fremde Fette. Z. anal. Chern., 18, 199, 431.
Kolthoff, L M. 1935. Adsorption indicators. Chern. Revs 16. 87.
Konig, w. 1904. tiber eine neue. vom Pyridin derivierende Klasse von Farbstoffen. I. pTakt. Chern., 69, lOS.
Korenman, I. M. 1934. Uber die Empfindlichkeit der Jodstarkereaktion. Mikrochernie. 15, 25.
Koser. S. A., and Rettger, L. F. 1919. Studies on bacterial nutrition. The utilization of nitrogenous compounds
of definite chemical composition. J. Infectious Diseases, 24, 301.
Kossel, A. 1898. Uber die Constitution der einfachsten Eiweiss-stoffe. Z. physiol. Chem.. 25. 165.
Kraut, K. 1890. Uber die Bildung von Glycocoll aus Monochloressigsaure. Ber.. 23. 2577.
_ _ . 1891. Glycocoll und Derivate. Ann., 266. 292.
Kreger, D. 1946. X-Ray fibre pattern of part of a single starch grain: powder photographs of potato. wheat, and
arrowroot (maranta) starch. Nature, 158, 199.
Krehl, W. A.. Elvehjem, C. A., de la Huerga. J . and Hart. E. B. 1946. The distribution of niacinamide and
niacin in natural materials. I. Bioi. Chern., 166, 53.
_ _ , _ _ , and Strong, F. M. 1943. Determination of nicotinic acid. Ind. Eng. Chern., Anal. Ed., 15. 471.
Kroeker. Dr. 1846. Bestimmung des Starkmehlgehaltes in vegetabilischen Nahrungsmitteln. Ann.. 58, 212.
Kuhn. R., and Desnuelle, P. 1937. Uber die Arninosauren des gelben Fennents. BeT., 70, 1907.
Kulp, w. L . and'Rettger, L. F. 1924. Comparative study of Lactobacillus acidophilus and Lactobacillus
bulgaricus. I. Bact., 9, 357.
z.,
- 170-
LaForge, F. B., and Hudson, C. S. 1918. The preparation of several useful substances from com cobs. Inti. Eng.
.... Chern., 10, 925.
Lamb, F. W. 1943. Direct quantitative determination of'nicotinamide in vitamin mixtures. Ind. Eng. Chern., Anal.
Ed., 15, 352.
_ _ , Mueller, A., and Beach, G. ,W. 1946. Quantitative determination of ergosterol, cholesterol and
7-dehydrocholesterol. Ind. Eng. Chern., Anal. Ed., 18, 187.
Lamb, J., and Robson, W. 1931. The Erlenmeyer synthesis of amino-acids. Biochem. J., 25, 1231.
Lampitt, L. H., Fuller, C. H. F., and Goldenberg, N. 1948. The fractionation of potato starch. IL The reducing
powers, mean molecular weights, and solubilities of the fractions. J. Soc. Chern. Ind., 67, 38.
Lang, K. 1933. Eine Mikromethode zur Bestimmung des Prolins und des Oxyprolins. Z. 'physiol. Chern., 219, 148.
La~ley, W. D., and Evans, M. 1936. Th~ determination of creatinine with sodium 3,5-dinitrobenzoate. J. BioI.
Chern., 115, 333.
Lanyar, F. 1942. Degradation of the d- and I-forms of phenylalanine and the dI- and I-forms of tyrosine by
alkaptonurics. Z physiol. Chem., 275, 217.
Larson, H. W. 1932. A colorimetric method for the determination of allantoin. J. BioI. Chern., 94, 727.
Lawrow, D. 1899. Uber die Spaltungsprodukte des Histons von Leucocyten. Z. physiol. Chern., 28, 388.
Leiboff, S. 1.. 1942. The determination of total cholesterol and cholesterol esters in blood. J. Lab. CUn. Med.,
28, 219.
Levene, P. A. 1903. Darstellung und Analyse einiger Nucleinsiiuren (vierte Mitteilung). Z. physiol. Chern., 39, 4.
1923. Preparation of ct-mannose. J. BioI. Chern., 57, 329.
1924. Preparation of ct-mannose. Second paper. J. BioI. Chern., 59, 129.
1925. Phenylhydrazine derivatives of pyrimidines. J. Bioi. Chem., 63. 653.
__,__. 1935. Note on the preparation of crystalline d-mannose and of crystalline d-ribose. J. BioI. Chern., 108,419.
_ _ and Bass, L. W_ 1931. Nucleic Acids. Chemical Catalog Company, Inc., New York.
Levin, L. 1945. The effect of several varieties of stress on the cholesterol content of the adrenal glands and of
the serum of rats. Encrocrinology, 37. 34.
Levine, V. E. 1929-1930. A gen~ral test for carbohydrates. Proc. Soc. Exptl. BioI. Med., 27. 830.
Leviton, A. 1949. Methanol extraction of lactose and soluble proteins from skim milk powdl"r. Ind. Eng. Chem.,
41, 1351.
_ _ and Leighton, A. 1938. Separation of lactose and soluble proteins .of whey by alcohol extraction. Inti. Eng.
Chem.. 30, 1305.
Lewis, H. B. 1941. End products of nitrogen metabolism in animals. BioI. Symposia, 5, 20.
Lewis, R. C., and Benedict, S. R. 1915. A method for the estimation of sugar in small quantitiesof blood. J.
BioI. Chern., 20. 61.
Lewis, T., and Woodward, E. F. 1948. Papain. Drug & Cosmetic Ind.. 63, 734.
Lewk'Owitsch, J. 1901. The Laboratory Companion to Fats and Oils Industries. MacmiJIan & Co., Ltd., London.
_ _ 1921. Chemical Technology and Analysis of Oils, Fats and Waxes, Vol. I. Mactiullan & Co., Ltd., London.
_ _ . 1922. Chemical Technology and Analysis of Oils, Fats and Waxes, Vol II. Macmillan & Co., Ltd., London.
Lieben, F., and Muller, R. 1928. {fuer die Bromaufnahme von Casein, sowie von Casein- und Keratinhydrolysaten.
Biochern. Z, 197. 119.
Liebig, J. 1846. Uber die Abwesenheit der Kohlensauren Alkalien im Bl~e. Ann., 57. 127.
_ _ . 1847. tiber die Bestandtheile der Fliissigkeiten des Fleisches. Anno, 62, 257.
Linderstrpm-Lang, K. 1928. tiber die Einheitlichkeit des Kaseins. Vorliiufige Mitteilung. Z. physiol. Chern.,
176, 76.
_ _ 1929. Studies on casein. III. On the fraction~tion of casein. Compt. rend. lab. Carlsberg, 17, No.9.
_ _ and Kodama, S. 1925. Studies on casein. I. On the solubility of casein in hydrochloric acid. Compt. rend.
lab. Carlsberg, 16, No.1.
Lineweaver, H., and Schwimmer, S. 1941. Some properties of crystalline papain. Enzymologia, 10, 81.
Ling, A. R., and Nanji, D. R. 1923. The preparation of xylose from maize cobs. J. Chern. Soc., 123, 620.
Lintner, C. 1907. Determination of starch in barley by means of polarization. Z ges. Brauw., 30. 109; Chern.
Abstracts, I, 2740 (1907).
von Lippmann, E. O. 1921. Einige pflanzenchemische Beobachtungen. Ber., 54, 3111.
Loftfield, R. B. 1947. Preparation of C14-labeled hydrogen cyanide, alanine and glycine. Nucleonics, 1, 54.
Looney, J .. M. 1926. The col~rimetric estimation of tyrosine, tryptophane, and cystine in proteins., n. J. Biol.
Chern., 69, 519.
Lowe, J. 1870. Uber die Anwendung des Glycerin-Kupferoxyd-Natrons zur Nachweisung und Bestimmung des
Traubenzuckers. Z. anal. Chern., 9, 20.
-_.
- 171-
Lucas, C. c., and Beveridge, J. M. R. 1940. The analysis of hair keratin. I. A method for the quantitative removal of cystine from keratin hydrolysates. Biochem. I., 34, 1356.
Luecke, R. W., and Pearson, P. B. 1944a: The microbiological determination of free choline in plasma and uri~e.
I. BioI. Chern., 153, 259.
_ _and _ _ 1944b. The determination of free choline in animal tissues. I. BioI. Chern., 155, 507.
Lugg, J. W. H. 1932a. The application of phospho-18-tullgstic acid (Foli'3's reagent) to the colorimetric determination of cysteine, cystine and related substances. I. The reduction of phospho-l&tungstic acid by various
substances. Biochern. I., 26, 2144.
'
_ _ 1932b. Th~ application of phospho-l8-tungstic acid (Folin's reagent) to the color_imetric determination of
cysteine, cystil).e and related substances. It The deterIl,lination of sulphydryl compounds and disulphides
already existing in solution. Biochern. I., 26, 2160.
_ _ 1933. Sullivan's reaction for the [Link] determ,ination of cysteine and cystine. Biochem. I., 27, 668.
_ _ 1938. Investigations of sources of error in the. estimation of tyrosine and tryptophan in complex materials.
which are associated with hydrolysis. Bioche,rn. I., 31, 775,
Lund, H., ~d Lieck, H. 1936. A specific reaction for the qualitative and quantitative determination of ascorbic
acid in serum. Nature, 137, 784.
Lwoff, A., and Querido, A. 1938. Dosage de l'amide de l'acide nicotinique au moyen du test Proteus. Principe
de la methoqe. Co:npt. rend; soc. bioI., 129, 1039.
McArthur, C. S. 1946. Biological incorporation of a choline homologue into liver phospholipids. Science, 104, 222.
McCalla, A. G., and Corns, W. G. 1943. Effects of variety and environment on the starch c~ntent of wheat ap.d
parley. Can. I. Research, ~1 C, 307.
McCarthy, T. E., and Sullivan, M. X. 1941. A new and highly specific colorimetric test for methionine. I. BioI.
Chern., 141, 871.
McCollum, E. V., Simmond~, N., and' Becker, J. E. 1922. Studies on experimental rickets. XXI. An experimental
demonstration of the existenoe of a vitamin which promotes calcium deposition. I. BioI. Chern., 53, 293.
_ _ , _ _ , Shipley, P. G., and Park, E. A. 1922. Studies on experimental rickets. XVI. A delicate biological
test for calcium-depositing substances. I. BioI. Chern., 51, 41.
McCready. R. M., and Hassid, W. Z. 1943: The separation and quantitative estimation of amylose and amylopectin
in potato starch. I. Arn. Chern. Soc., 65, 1154.
McFarlane, W. D., and Guest, G. H. 1939. A new colorimetric method for the determination of hydroxyproline and
its application to gelatin hydrolysates. Can. I. Research, 17B, 139.
McGuigan, H. 1907. The oxidation of various sugars and the oxidizing power of <I+fferent tissues. Am. I. Physiol.
19, 175.
McIntire, J. M., Schweigert, B. S., and E1vehjem, C. A. 1944. The choline and pyridoxine content of meats. I.
Nutrition, 28, 219.
McKittrick, D. S., and Schmidt, C. L. A. 1945. Elimination of error in the Volhard analysis for chloride by
titration in acetic acid. Arch. Biochern., 6, 273.
McLean, F. C., and Van Slyke, D. D. 1915a. A method for the determination of chlorides in small amounts of
body fluids. I. BioI. Chern., 21, 361.
_ _ and _ _ 1915b. A method for the titration of small amounts of halides. I. Arn. Chern Soc., 37, 1128.
McRary, W. L.,and Slattery, M. C. 1945. The co19rimetric determination of pentoses and pentosans. Arch.
Biochem., 6, 151.
\ ,
MacFadyen, D. A. 1942. Determination of'amino acids in plasma by the ninhydrin-carbon dioxide reaction without
removal of proteins. I. BioI. Chern 145, 387.
_ _ . 1944. Determination of ammonia evolved fro_m a-amino acids by ninhydrin. I. Bioi. Chern., 153, 507.
_ _ 1945. Estimation of formaldehyde in biological mixtures. I. Bioi. Cliern 158, 105.
MacMasters, M.' M., and Hilbert, G. E. 1944. Glutinous corn and sorghum starches. Ind. Eng. Chem.,_36, 958.
Macpherson, H. T. 1942. Modified procedures for the colorimetric estimation of arginine and histidine. Biochem.
I., .36, 59.
Maher, F. T., and Wirth, E. H. 1946. The assay of papain with suggestion for the modification of the tentatively
accepted assay. Bull. Natl. Formulary Comm., 14, 56.
Maillard, L. C. 1914a. Synthese de polypeptides par action de la glycerine sur, Ie glycocolle. Ann. chim. (9), 1, 519
_ _ 1914b. Synthese de polypeptides par action de la glycerine sur Ie glycocolle: ~tude dynamique. Ann.
c~irn. (9),.2. 210.
Malaprade, M. L. 1934. Etude de l'action des polyalcools sur l'acide periodique et les p~riodates alcalins.
Bull. soc. chirn. (5), 1, 833.
Malpress, F. H., and Morrison, A. B. 1949. The semi-micro estimation of lactose alone and in the presence of
other sugars. Biochem. I., 45, 455.
-172
Mannich, C., and Lenz, K. 1920. Uber eine Methode zur polarimetrischen Bestimmung der Starke'in Calciumchloridlosung. Z. Untersufh. Nahr. U. Genussrn., 40, 1.
Maquenne, L., and Roux, E. 1903. Sur la r~trogradation de l'empois d'amidon. Compt. rend., 137, 88.
,,,,.Markham, R.,' and Smith, J. D. 1949. Chromatographic studies of nucleic acids. Biochem. I., 45, 294.
Marquardt, P. 1947. Uber eine Arbeitstheorie zu den Beziehungen zwischen dem Stadium elner Infektionskranlrheit
und der Hohe des Cholesterinspiegels im Blut. Experientia, 3, 203.
Marshall, E. K., Jr. 1926. The secretion of urine. Physiol. Revs., 6, 440.
Mathews, A. P. 1930. Physiological Chemistry, 5th ed. William Wood & Company, New York.
_ _ and McGuigan, H. 1907.
A study of the oxidizing power of cupric acetate solutions. Am. I. Physiol., 19, 199.
Mellanby, E. 1919. A further demonstration of the part played by accessory food factors in the aetiology of rickets.
I. Physiol., 52, Lill.
Melnick, D., and Oser, B. L. 1943. Chemical differentiation between niacinamide and niacin in pharmaceutical
products. Ind. Eng. Chern., Anal. Ed., 15, 355.
Merck and Co. 1939. Nicotinic Acid. Rahway, N. J.
Meredith, W: O. S. 1946. Analysis of'data for A.A.C.C. check sample service. II. Protein and thiamine results,
1944-1945. Cereal Chem., 23, 585.
Meyer, .K. H. 1942. Recent developments in starch chemistry. In Kroemer, E. 0., 'Advances in Colloid Science,
New York, I, 143.
_ _ , 'Fischer, E. H., and Bernfeld, P.- '1947. Physical and chemical properties of crystalline ct-amylase of hog
pancreas. Arch. Biochem.~ 14, 149.
_ _ , Wertheim, M., and Bemfeld, P. 1941. Recherches sur l'amidon XUI. Contribution ~ l'~tude de 1 'amidon de
pommes de terre. Helv. Chim. Acta, 24, 378.
Militzer, W. E. 1941. The qualitative chemical identification of the natural sugars. I. Chem. Education, 18, 25.
_ _ 1946. Note on the orcinol reagent. Arch. Biochem., 9, 85.
_ _ , Ikeda, C., and Kneen, E. 1946a. The preparation and properties of an amylase inhibitor of wheat. Arch.
Biochern., 9, 309.
_ _ , _ _ , and _ _ . 1946b. The mode of action or'an amylase inhibitor from wheat. Arch. Biochem.. 9, 321.
Miller, B. F., and Muntz, J. A. 1938. A method for the estimation of ultramicroquantities of lactic acid. I. BioI.
Chem., 126, 413.
Miller, G. L., and Miller, E. E. 1948. Determination of nitrogen in biological materials. Improved KjeldahlNessler method. Anal. Chem., 20, 481.
Milfer, L., and Houghton, J. A. 1945. Themicro-Kjeldahl determination of the nitrogen content of amino acids
and proteins. J. BioI. Chem., 159, 373.
Millon, E. 1849. Sur un r~actif propre aux compos~s prot~iques. Compt. rend., 28, 40.
Mills, M. B., and Roe, J. H. 1947. A critical study of proposed 'modifications of the Roe and Kuether method for
the determination of ascorbic acid, with further contributions to the chemistry of this procedure. I. Biol. Chern.,
170, 159.
Mitchell, H. K., ~nd Niemann, C. 1947. The competitive inhibition of the metabolism of:, <X-amino acids by
their halogenated analogs. I. Am. Chern. Soc .. 69, 1232.
_ _ and Nyc, J. F. 1948. Hydroxyanthranilic acid as a precursor of nicotinic acid in Neurospora. Proc. Natl.
- 174-
-175 -
Riggs, L. K., and,Beaty, A 1947. Some unique properties of lactose as a dietary caroohydrate. }. Dairy Sq.,
30. 939.
'
Rittenberg. D and Schoenheimer. R. 1939. Studies in protein metabolism. VI. J-lippuric acid formation studies
with the aid of the nitrogen isotope. J. Bioi. Chem 127. 329.
Ritthausen, H. 1869a. Asparaginsaure und Glutaminsaure. Zersetzungsprodukte des Legumins beim Kochen mit
Schwefelsiiure. J. prakt. Cbem.. 106. 445!
_ _ 1869b. Asparaginsiiure und Glutaminsiiure. Zersetzungsprodukte des Legumins und Conglutins beiln Kochen
mit Schwefelsaure. J. prakt. Chern 107. 218.
Roberts. E. C . and Snell. E. E. 1946. An improved medium for microbiological assays with Lactobacillus casei.
J. Bioi. Cbern 163. 499.
Robertson, E. B. 1934. Spectrophotometric estimation of ascOlilic acid. J. SOC. Cbem. Ind.. -53. 277.
Robertson, G. R. 1927. The reaction of chloro-acetic a'iid with ammonia and the preparation of glycine. J. Arn.
Chern. Soc 49. 2889.
_ - . 1943. LaboratorY Practices of Organic Chemistry, 2d ed. The Macmillan Company, New York.
Robiquet., 1805. Memoire sur les acides aspartique et malique. Quoted by Pasteur, L., Ann. cbim. et phys. (3),
34. 30 (1852).
Roche. J. 1946. Sur la biochimie general des phosphatases. Helv. Chim. Acta, 29, 1253.
Roman, W. 1930. Eine chemische Methode zur [Link] Bestimmung des Cholins und einige physikalischechemisclle Daten des Ch_olins und seiner Salze. Biochern. Z., 219. 218.
Rose. W. C. 1938. The nutritive significance of the amino acids. Physiol. Rev., 18. 109.
Rosenberg. H. R. 1942. Chemistry and Physiology of the Vitamins. Interscience Publishers, New York.
Rosenheim, O. 1906a. On the preparation of cholesterin from brain. J; Physiol 34. 104.
_ _ 1906b. A colour reaction of formaldehyde with proteids and its relation to the Adamkiewicz reaction.
Biochem. J 1. 233.
Rosenthaler. L. 1904. Eine titrimetrische Zuckerbestimmung. Z. anal. Chern 43. 282.
Rosin. H. 1903. Eine Verschiirfung der Seliwanoffschen Reaktion.
Z. physio/. Chem 38. 555.
Rothmund, V and Burgstaller. A. 1909. Uber die Genauigkeit der Chloroestimmung nach Volhard. Z. anorg.
Chern 63, 330.
Rudisch. J . and Celler, H. L. 1907. The quantitative determination of glucose in the urine by a new modification
oC Fehling's solution. }. Am. Med. Assoc 48. 324.
Ruhemann, S. 1910. Triketohydrindene hydrate. J. Chem. Soc., 97, 2025.
_ _ 1911. Triketohydrindene hydrate. Part V. The analogues of uramil and purpmi'c acid. J. Chern. Soc., 99,
1486.
Rundle. R. E.. Foster. J. F., and Baldwin, R. R. 1944. On the nature of the starch-iodine complex. J. Am. C~em.
Soc., ~6, 2116.
Rydon, H. N. 1948. The synthesis of the nuclear-C-methylated tryptophans. A hate on the aldehyde reactions for
. tryptophan. J. Chern. Soc., 705.
Saifer, A.. and Kammerer, O. F. 1946: Ph~tometric determination of total cholesterol in plasma or serum by a
modified Liebermann-Burchard reaction. J. Bioi. Chem 164. 657.
Sakaguchi. S. 1925a. A new color reac~ion of prOtein and arginine. J. Biochem. (J apan). 5. 25; Chern. Abstracts,
19. 3506 (1925).
_ _ 1925b. The mode of combination and determination of arginine in the protein molecule. J. Biochem. (J apan).
5. 133; Chern. Abstracts, 20, 925 (1926).
S~ami, W. 1949. The conversion of glycine into serine in the intact rat. }. BioL Chem., 178, 519.
_ _ , Evans. W. E and Gurin, S. 1947. The synthesis of organic compounds labelled with isotopic camon.
J. Am. Chern. Soc 69, 1110.
Salkowski, E. 1888. Kleinere Mittheilungen. Uber die Farbenreaktionen des Eiweiss. Z. physio/. Chem., 12, 215.
,Sannie. C. 1942. A simple synthesis of 2.5-diketopiperazines (Cl-amino acid anhydrides). Bull. soc. chim. (5),
9. 487; Chern. Abstracts, 37, 5065 (1943).
Sarett, H. P., Bennett, M. J., Riggs, T. R., and Cheldelin, V. H. 1946: Thiamine, riboflavin, nicotinic acid,
pantothenic acid and ascoro~c acid content of restaurant foods. J. Nutrition. 31, 755.
Sasaki, T. 1914. Uber die biochemische Umwandlung primarer Eiweiss-spaltprodukte durch Bakterien. Biochem.
Z., 59,429.
Satherfie1d, G. H. 1947. Chemicai methods for the estimation of ascoroic acid. Bioi. Symposia, 12. 397.
Scheele, C. W. 1780. Kong. Vetenskape Akademiens nya handlinger. I, 116. Quoted by Sutermeister and Browne
(1939).
Schenck, F. 1937. Uber die kristallisierte Vitamine D3 Naturwissenschaften, 25, 159.
-176 -
Schmidt, C. L. A.
J.
1929. The reaction between nitrous acid and certain amino acids and related compounds at 45~
_ _ and Allen, F. W. 1938. Fundamentals of Biochemistry with Laboratory Experiments, 1st ed. McGraw-Hill
. Book Company, Inc~, New York, 284.
:lchOberl, A., and Rambacher, P. 1937-1938. Uber die kolorimetrische Bestimmung von Cystein and Cystin mit
Phosphorwolframsiiure. Biochem. Z., 295, 377.
Schoch, T. J. 1942. Non-carbohydrate SUbstances in the cereal starches. J. Am. Chfm. Soc., 64, 2954.
Schoenheimer, R. 1932. Die Spezifitat der Cholesterinres<!rption und ihre biologische Bedeutung. Klin. Wochsch,.,
II, 1793.
Schopmeyer, H. H. 1945. Amioca - the starch from waxy corn. Food Jnds.~ 17, 1476.
Schott, H. F., Larkin, J. B., Rockland, L. B., and Dunn, M. S., 1947. Synthe~is of 1(~)leucy1glycy1glycine. J. Org.
Chem., 12, 490.
_ _, Rockland, L. B., and Dunn, M. S. 1944. Investigations of amino acids, peptides and proteins. XvI. A source of
errorin the manometric ninhydrin method for the analysis of amino acids and its suppression by the use of hydrazine.
}. BioI. Chern., 154, 397.
Schotten, C. 1888. Die Umwandlung des Piperidins in [Link] und in Oxypiperidin, Ber., 21, 2235.
Schulze, E., and Winterstein, E. 1~10. Studien iiber die Proteinbildung in reifenden Pflanzensamen. Z. physiol.
Chem., 65, 431.
Scott, E. M., and Sandstrom, W. M. 1942. The activation of papain. Arch. Biochern., I, 103.
Scotti, G. 1938. A new method for determining the iodine number of oils and fats. Olii minerali, grassi e saponi,
colon e vernici, 18, 96; Chem. Abstracts, 34, 4291 (1940).
Sebrell, W. H., Jr. 1934. Table showing the pellagra-preventive value of various foods. Public Health Reports,
U.S.P.H.S., 49, 754.
Seliwanoff, T. 1887. Notiz liber eine Fruchtzuckerreaktion. Ber., 20, 181.
Sen, R. N., and Sinha, N. N. 1923. Condensations of aldehydes with resorcinol and some other aromatic hydroxy
compounds. }. Arn. Chem. Soc.. 45, 2984.
Sendroy, J., Jr. 1937. Microdetermination of chloride in biological fluids, with solid silver iodate. I. Gasometric
analysis. }. Bioi. Chern., 120, 335.
Seshagirirao, P., and Giri, K. V. 1942. The mechanism of !3-amylase inhibition by vitamin C. Proc. Indian Acad.
- 177-
Snell, F. D., and Snell, C. T. 1948. Colorimetric Methods of Analysis. D. Van Nostrand Company, Inc., New York.
Sobel, A: E., Hirschman, A., and Besman, L. 1945. A convenient microtitration method for the estimation 6f amino
acids. J. BioI. Chem., 161, 99.
\
_ _ and Mayer, A. M. 1945. Improvements in the Schoenheim~r-Sperry method for the determination of free
cholesterol. J. BioI. Chem., 157, 255.'
,
I
.
Sobotka, H. 1938. Chemistry of the Sterids. Thl' Williams & Wilkins Company, Baltimore.
Society for Experimental Biology. 1947. Nu<;,l"tit,,~AcidS. Cambridge University Press, London.
Somogyi, M. 1938. Micromethods for the estillldHon of diastase. J. BioI. Chem., 125, 399.
__ 1945. A new reagent for the determination of sug!"rs. J. BioI. Chem., 160, 61.
Soodak, M., Pircio, A., and Cerecedo, L. R. 1949. A colorimetric method for the estimation of uracil and cytosine.
J. Bioi. Chern., 181, 713.
SlIIrenson, M., and Haugaard, G. 1933. Uber die Andwendbarkeit der Orcinreaktion zur Bestimmung der Art und
Menge von Kohlenhydratgruppen in Eiweiss-stoffen.1 Biochem. Z., 260, 247.
Speich, H. 1942. Optical properties of potato starch grains. Ber. schweiz. botan. Ges., 52. 175; Chem. Abstracts,
.38, 2233 (1944).
Stark, I. E. 1942. Equilibria in diastatic reactions. J. Bioi. Chtim., 142, 569.
Stein, R. S., and Rundle, R. E. 1948. On the nature of the interaction between starch and iodine. J. Chem. Phys.,
16, 195.
Stein, W. H., and Moore, S. 1949. Amino acid composition of e~lactog1obulin and bovine serum albumin. J. BioI.
Chem., 178, 79.
Steiner, A. 1948. Significance of cholesterol in coronary arteriosclerosis. N. Y. State J. Med., 48, 1814.
Steko1, J. A. 1941. Detorlcation mechanisms. Ann. Re~ Biochern., 10, 265.
Stephenson, C. H., and Parker, C. E. 1921. Some Microchemical Tests for Alkaloids. J. B. Lippincott Company,
Philadelphia, Pa.
Stern, A., Beach, E. F., and Macy, I. G. 1939. Polarographic microdetermination of cystine in protein hydrolysates.
J. BioI. Chern., 130, 733.
Stetten, D., Jr. 1942. Biological synthesis of choline by rats on diets with and without adequate lipotropic methyl.
J. Bioi. Chern., 142. 629.
Stewart, A. P., and Sharp, P. F. 1945. Determination of vitamin C in the presence of interfering reducing substances. Ind. Eng. Chern., Anal. Ea., 17, 373.
Stone, W. E., and Lotz, D. 1891. A new source for xylose. Am. Chem. J., 1.3, 348.
Strafford, N. 1936. Quantitative colorimetric analysis. Ann. Repts. Soc. Chem., on Progress Applied Chern., 456.
Strain, H. H. 1942. Chromatographic Adsoll>tion Analysis. Interscience Publishers, New Yolk.
Strain, W. H. 1943. The steroids.' In Gilman's Organic Chemistry, Vol. II, 2d ed. John Wiley & Sons, Inc., New
Yolk.
Strecker, A. 1849. Beobachtungen iiber die Galle verschiedener Thiere. Ann."70, 149.
Street,' H. E., Kenyon, A. E., and Watson, G. M. 1946. Estimation of free choline in plants. Biochem. J., 40, 869.
~tuart, N. W., and Brimhall, B. 1943. Starch from easter lily bulbs. Cereal Chern., 20, 734.
Stumpf, P. K., and Green, D. E. 1944. I-Amino acid oxidase of Proteus vulgaris. J. BioI. Chem., 15.3, 387.
Sullivan, J. T. 1935a. The estimation of starch-in plants with special reference to [Link] plants. 'J. Assoc.
O//ic. Agr: Chemists, 18, 621.
_ _ 1935b. The estimation of starch. Ind. Eng. Chern., Anal. Ea., 7, 311.
Sullivan, M. X. 1929a. Studies on the biochemistry of sulphur. IV. The colorimetric estimation of cystine in
casein by means of the beta-naphthoquinone reaction. Public Health Reports, U.S.P.H.S., Sup pl. No. 78.
_ _ 1929b. Studies on the biochemistry of sulphur. II. Further studies on the distinctive reaction for cysteine
and cystine. Public Health Reports, U.S.P.H.S., 44, 1421.
_ _ and Hess, W. C. 1929. Studies on the biochemistry of sulphur. nI. Chemical groups involved in the
naphthoquinone reaction for cysteine and cystine.
Public Health Reports, U.S.P.H.S., 44, 1599.
Summerson, W. H. 1939. Simplified test-tube photoelectric colorimeter, and the use of the photoelectric colorimeter in colorimetric analysis. J. Bioi. Chern., 130, 149.
Sumner, J. B. 192L Dinitrosalicylic acid: a reagent for the estimation of sugar in normal and diabetic urine.
J. Bioi. Chern., 47, 5.
_ _ . 1925. A more specific reagent for the determination of sugar in urine. J. BioI. Chem., 65, 393.
_ _ and Somers, G. F. 1947. Chemistry and Methods of Enzymes, 2d ed. Academic Press, New York.
Sunderman, F. W. 1942. Phosphatase. Recent advances in the significance and interpretation of phosphate
measurement in disease. Am. J. Clin. Path., 12, 404.
Suoma1ainen, H., and Arhimo, E. 1947. Oxidative bromination in the determination of malic acid and aspartic
acid. Anal. Chem., 19, 207.
- 178-
Sutermeister. E.. [Link]. 'F. L. 1939. Casein and Its Industrial Applications. Reinhold Publishing Corporation, New York.
,Suzuki. U Shinamura. T and Odake. S. 1912. Uber Oryzanin. ein Bestandteil der Reiskleie und seine physio- logische Bedeutung. Biochem. Z .. 43. 89.
'
Szent-GyOrgyi. A. 1928. Observations on the function of peroxidase systems and the chemistry of the adcenal,
cortex. Description of a new carbohydrate derivative. Biochem. J., 22, 1387.
Tauber, H. 1937-1938. A color test for pentoses. Proc. Soc. Exptl. BioI. Med., 37. 600.
1949. The Chemistry and Technology of Enzymes. John Wiley & Sons. Inc New Yolk.
_ _ and Kleiner, I. S. 1932-1933. A method for the determination of monosaccharides in the presence of disaccharides and its appli:cation to blood analysis. J. Bioi. Chern 99. 249.
_ _ and _ _ 1935. An enzymic method for the estimation of true vitamin C. J. BioI. Chern., 110, 559.
Taylor. W. H . and Smith. G. F. 1942. Micro-Kjeldahl nitrogen determination without use of titration procedure.
Inri. Eng. Chern Anal. Bri., 14, 437.
Tebb. M. C. 1906. The cholesterin of the brain. J. Physibi., 34, 106.
Teeri, A. E. 1944. The determination of esterified cholesterol. J. Bioi. Chem., 156, 279.
van Thoai, N., and Bern~e-Silhol, J. 1946. Sur la purification de l'amylase pancre.a,tique. Cornpt. rend., 223, 761.
Thompson. R. L and Wilkins. M. L. 1948. Inhibition of growth of the vaccinia virus by !3-2-thienylalanine and
its reversal b~ phenylalanine. Proc. Soc. Exptl. BioI. Med 68, 434.
Tillmans. J., Hirsch, P and Reinshagen, E. 1928. The use of 2.6-dichlorophenol indophenol as a reduction
indicator in the examination of foodstuffs. Z Untersuch. Lebensm., 56. 272; Chern. Abstracts. 23. 3277 (1929).
Tinker. J. T . and Brown. G. B. 1948. The characterization of purines and pyrimidines by the method of countercurrent distribution. J. Bioi. Chern., 173, 585.
Todd, J. C., and Sanforo, A. H. 1948. Clinical Diagnosis by Laboratory Methods, 11th ed. W. B. Saunders Company.
Philadelphia.
Toennies. G. and Bennett. M. ~ 1935-1936. Some observations on the isolation of cystine from wool hydro1ysates.
J. Bioi. Chern 112, 39.
_ _ and Lavine, T. F. 1930. On the optical rotation of L-cystine. Determination of its value for the sodium and
mercury lines and of the temperature factor. J. BioI. Chern., 89, 153.
Tompsett. S. L. 1931. A note on the determination of cystine in proteins by the method of Folin and Marenzi.
Biochem. J.; 25, 2014.
Tristram. G. R. 1939. The basic amino acids of leaf proteins with a discussion of various methods of analysis.
Biochem. J., 33, 1271.
Tswett. M. 1906. Adsorptionsanalyse und chromatographische Methode. Anwendung auf die Chemie des
ClUorophylls. Ber. deut. botan. Ges., 24. 384; Cent. 1906, TI. 1286.
Turner. N. C. 1943. Starch blue. Science, 108. 302.
v. Udrllnsky. L. 1888. Uber Furfurolreaktionen. Z physiol. Chern., 12, 355.
Underkofler, L. A. Christenson, L. M.. and Fulmer. E. I. 1936. Butyl-acetonic fermentation of xylose and other
sugars. Ind. Eng. Chern., 28. 350.
_ _ and Hunter, J. E . Jr. 1938. Butyl-acetonic fermentation of arabinose and other sugars. Inri. Eng. Chern.,
30, 480.
Uschinsky. 1893. Uber eine eiweissfreie Niihrlosung fur pathogene Bakterien nebst einigen Bermerkungen uber
Tetanusgift. Centro Bakt., 14, 316.
Van Slyke. D. D. 1911. A method for quantitative determination of aliphatic amino groups. Applications to the
study of proteolysis and proteolytic products. }. Bioi. Chern., 9. 185.
1912. The quantitative determination of aliphatic amino groups. n. J. BioI. Chern., 12, 275.
1913-1914. The gasometric determination of aliphatic amino nitrogen in minute quantities. J. Bioi. Chern.,
16. 121.
1915a. Improvements ~ the method for analysis of proteins by determination of the chemical groups characteristic of the differen_t amino acids. J. BioI. Chern., 22, 281.
_ _ 1915b. Note on the micro-method for the gasometric determination of aliphatit: amino nitrogen. }. Bioi.
Chern.. 23, 407.
- _ . 1917. Studiesof acidosis. VII. The determination of S-hydrQxy-butyric acid, aceto-acetic acid, and
acetone in urine. J. Bioi. Chern., 32, 455.
- - . 1929a. The determination of acetone bodies in blood and urine. J. BioI. Chern., 83, 415.
- _ . 192%. Manometric determination of primary amino nitrogen and its application to blood analysis. J. BioI.
Chern., 83, 425.
- - and Cullen. G. E. 1914. A permanent,preparation of urease, and its use in the determination of urea. J.
Bioi. Chern., 19. 21.1.
- 179-
1938. Gasometric determination of carboxyl groupsin amino acids. Compt. Tend. trav.
lab. Carlsberg, 22, 480.
_ _ , _ _ , MacFadyen, D. A., and Hamilton, P. B. 1941. Gasometric determination of carboxyl groups in free
amino acids. J. Bioi. Chern., 141, 627.
- - and Folch, J. 1940. Manometric carbon determination. J. BioI. Chem., 136, 509.
_ _ , Hiller, A., and Dillon, R. T. 1942. Solubilities and compositions of the pho'spho-l2-tungstate!j of the dia'l:nino
acids and of proline, glycine, and tryptophane. J. Bioi. Chern., 146, 137.
_ _ , _ _ , and MacFadyen, D. A. 1941. The determination of hydroxylysine in proteins. J. Bioi. Chem., 141,681.
_ _ , _ _ , _ _ , Hastings, A. B., and Klemperer, F. W. 1940. On hydr~xylysine. J. BioI. Chem., 133, 287.
_ _ , MacFadyen, D. A., and Hamilton, P. B. 1942. Application of the gasometric ninhydrin.:c0 2 method to
determination of amino acids in blood. Federation Proc., 1. 139.
_ _ , _ _ , and _ _ 1943. The gasometric determination of amino acids in urine by th~ ninhydrin-carbon
dioxide method. J. BioI. Chem., 150, 251.
_ _ and Palmer, W. W. 1920. Studies of acidosis. XVI. Determination of organic acids in urine. J. Bioi. Chem.,
_ _ and Dillon, R. T.
41, 567.
Van Slyke, L.L. and Baker,J.C. 1918. The preparation qfpure casein. J. Bioi. Chem.. 35,127.
_ _ and Bosworth, A. W. 1913. Method of preparing ash-free casein and paracasein. J. Bioi. Chem., 14, 203.
_ _ and Carpenter, D. C. 1924. Quoted by Zoller, H. F. Ca~ein and the Dairy Industry. In Bouge, R. IL, The
Theory and Application of Colloidal Behavior. McGraw-Hill Book Company, Inc., New York, 2. 784.
Vauquelin and Robiquet. 1806. Decouverte d'un nouveau principe vegetal dans les asperges (Asparagus sativus.
Linn.) Ann. chirn. et. phys. (1) 57, 88.
Vickery, H. B. 1946. Position of Arnold ill relationship to the Kjeldahl method. J. Assoc. Of/ic. Agr. Chemists,
29, 358.
_ _ and Leavenworth, C. S. 1929. The separation of cystine from histidine: the basic amino acidsof human
hair. J. Bioi. Chern., ~3, 523.
_ _ and Pucher, G. W. 1943. Amide metabolism in etiolated seedlings. I. Asparagine and glutamine formation
in Lupi'nus angusti/olius. Vicia atropurpurea. and Cucurbita pepo. J. Bioi. Chern., 150, 197.
_ _ , _ _ , Wakeman, A. J., _and Leavenworth, C. S. 1937. Chemical investigations of the tobacco plant. VI.
Chemical cpanges that occur in leaves during culture in light and in darkness. Connecticut Agr. Exp. Sta.
Buli., 399.
_ _ and Schmidt, C. L. A. 1931. The history of the discovery of the amino acids. Chern. Revs., 9, 169.
_ _ and White, A. 1932. The u,se of cysteine cuprous mercaptide in the determination of cystine. J. BioI.
Chern., 99, 701.
du Vigneaud, V. 1941. Interrelationships between choline and other methylated compounds. Bioi. Syrnposia, 5,
234.
_ _ 1942-1943. Significance of labile methyl groups in the diet and their relation to transmethylation. Haroey
Lectures, 38, 39.
_ _ and Irish, O. J. 1938. The role of the acetyl derivative as an intermediary stage in the biological synthesis
of amino acids from keto acids. J. BioI: Chern., 122, 349.
_ _ and'Meyer, C. E. 1932. The racemization of amino acids in aqueous solution by acetic anhydride. J. Bioi.
Chern., 98, 295.
Virtanen, A. I., and Rautanen, N. 1947. Determination of certain amino-acids by ninhydrin oxidation to volatile
aldehydes. Biochem. I., 41, 101.
Virtue, R. W., and Lewis, H. B. 1934. Toe iodometric determination of cystine in the urine. I. BioI. Chern.,
104, 415.
Vischer, l}, and Chargaff, E. 1948a. ~he separation and quantitative estimation of purines and pyrimidines in
minute amounts. J. BioI. Chern., 176, 703.
_ _ and _ _ . 1948b. The composition of the pentose nucleic acids of yeast and pancreas. J. Bioi. Chern.,
176, 715.
_ _ , Magasanik, B., and Chargoff, E. 1949. Chromatographic separation of ribonuc1eotides and its application
to the study of ribonucleic acid structure. Federation Proc., 8, 263.
_ _ , Zamenhof, S., and Chargoff, E. 1949. Microbial nucleic acids: the desoxypentose nucleic acids of avian
1:ubercle bacilli and yeast. J. BioI. Chern., 177, 429.
Volhard, J. 1874. Uber eine neue Methode der maassanalytischen Bestimmung des Silbers. J. prakt. Chern., 9,217.
- - . 1878. Die Anwendung des Schwefelcyanammoniums in der Massanalyse. Ann., 190, 1.
Volz, G. W., and Caldwell, M. L. 1947. A study of the action of purified amylase from Aspergillus oryzae, takaamylase. J. Bioi. Chern., 171, 667.
Vredenburg, R. M. 1950. Guide to Photometric Analysis. Fisher Scientific Company, Pittsburgh, Pa.
- 180-
Waddell, J., and Kennedy, G. H. 1947a. The antirachitic potency of pure crystalline vitamin D3 in comparison
with the U.S.P. ~eference cod liver oil when assayed by the chick method. J. Assoc. Offic. Agr. Chemists,
30,190.
- - and _ _ 1947b. The assay of vitamin D by the chick method. Bioi. Symposia, 12,435.
Waelsch, H., and Prescott, B. A. 1945. :1\ chemical method for the determination of glutamic acid. Federation
Proc., 4, 108.
"''''Wagner, E. C" 1940. Titration of ammonia in presence ~f boric acid. Inti. Eng. Chern., Arial. Eti., 12, 771.
Waldschmidt-Leitz, E., and Akabori, S. 1934. Bestimmung von Oxyprolin. Zur Methode nach K. Lang. Z. physiol.
Chern., 224, 187.
Warner, R. C. 1944. Separation of ex- and f3-ca~ein. J. Am. Chern. Soc., 66, 1725.
Waser, E., and Brauchli, E. 1924. Untersuchungen in der Phenylalanin-Reihe. V. Hydrierung des Tyrosins.
Helv. Chim. Acta, 7, 740.
Webb, B. H., SJld Ramsdell, G. A. 1944. The manufacture of crude and technical lactose from cheese whey.
Natl. Butter Cheese J., 35, No.7, 18, No.8, 18.
Weber, C.
J.
th~
J. Bioi.
J.'
- 181 -
_ _ and _ _ 1945. Quantitative determination of d-xylose by selective fermentation. Ind. Eng. Chern., Anal.
Ed., 17, 182.
_ _ , Ratliff, E. K., and Browning, B. L. 1948. Determination of.m~ll1l:0sr' Mannans in hardwoods. Anal. Chem.,
20, 825.
'
Wittenberg, J., and Shemin, D. 1949. The utilization o{ glycine for the biosynthesis of both types of pyrroles in
protoporphyrin. J. .Bioi. Chern., 178,47.
Wohlgemuth, J. 1908. liber eine neue Methode zur Quantitativen Bestimmung des Diastatischen Ferments.
Biochem.
z.,
9, 1.
Wood, H. G., Geiger, C., and Werkman, C. H. 1939-1940. Nutritive requirements of the heterofermentative lactic
acid bacteria. Iowa State Coll. J. Sci., 14, 367.
Woodard, H. Q. 1934. Colorimetric determination of iodine by the starch-iodine reaction. Ind. Eng. Chern., Anal.
Ed., 6, 331.
Woodhouse, D. L. 1949. A method for the colorimetric mcro-estimation of thymine. Biochern. J., 44, 185.
Woodson, H. W., Hier, S. W., Solomon, J. D., and Bergheim, O. 1948. Urinary excretion of amino acids by human
subjects on normal diets. J. Bioi. Chern., 172, 613.
Woolley, D. W., and Peterson, W. H. 1937-1938. Some observations on the Kapel1er-Adler method for the determination of histidine. The histidine content of yeast. J. Bioi. Chem., 122, 207.
Wu, H. 1920. Contribution to the chemistry of phosphomolybdic acids, phosphotungstic acids and allied substances. J. Bioi. Chem., 43, 189.
Yamafuji, K., and Yoshida, T. 1939. Eine Micromethode zur Zuckerbestimmung mittels a-Naphthol. Biochem.
-182 -
INDEX
A
Acetaldehyde, determination of, 80
formation of, from amino acids, 44
from lactic acid, 114
from threonine, 38, 41,44
reaction' of, with bisulfite, 44
with p-hydroxydiphenyl, 38, 44
Acetaldehyde bisulfite, determination of, 114
formation of, 114
Acetic acid, detection of proteins with, 111, 113
Acetic anhydride, determination of cholesterol with, j08
Acetoacetic acid, excretion of, in diabetes, 113
_formation of acetone from, 113
in urine, 110, 113
Acetone, complex of, with mercuric sulfate, 41, 44
with salicylaldehyde, 41
detection of, with ferric chloride, 113
with iodine, 113
with salicylaldehyde, 111, 113
determination of, by gravimetric analysis, 113
by iodimetry; 113
as mercuric sulfate complex, 113
by photometry, 113
as salicylaldehyde complex, 113
in urine, 113
excretion of, in normal and pathological conditions,
113
formation of, from acetoacetic acid, 113
from /3-hydroxybutyric acid, 113
from leucine and valine, 41
formation of iodoform from, 113
separation of, from methyl ethyl ketone, 41, 44
in urine, 110, 111, 113
Acetylcholine, determination of choline as, 96
liberation of, from parasympathetic nerves, 95
Acetylphenylalanine, specific rotation of D-, 107
synthesis of, from phenylalanine, 106
Acetylphenylalanine methyl ester, formation of
L-phenylalanine- from, 106
hydrolysis of L-, with chymotrypsin, 106
Acetylphenylalanine p-toluidide, asymmetric
synthesis of, with papain, 106
Acetylphenylalanylglycine anilide, asymmetric
synthesis of, with papain, 106
Achrodextrin, formation of, from starch, 102, 103
Acid number, determination of, 49
Acidity, determination of, in urine, 114
of urine, variations in, 114
Acids (see specific names of acids)
Adenine, detection of, with potassium-bismuth
iodide, 86
with silver salt reagent, 85
formation of hypoxanthine from, 84
in nucleic acids, 84
reaction of, with diazobenzenesulfonic acid, 87
salt of, with HI and Bil 3, 87
Adrenal cortex, corticosterone in, 53
Adrenaline, physiological action of, compared to
tyramine. 24
,... 183-
B
Bacteria, aSparagine in media for growth of, 29
decarooxylation of tyro sine by, 24
inhibition of, by phenylalanine analogues, 10
techniques in culturing of, 92
tyrosine decarboxylase in, 24
(See also Microorganisms)
Barometer, determination of atmospheric pressure
with, 155
Bean sprouts, isolation of asparagine monohydtate
from, 29
Beef, determination of papain proteolytic activity
with, 106
Beer-Lambert equation, derivation of, 150
Benedict reagent, detection of reducing carbohydrates
with, 79, 111
Benzaldehyde, reaction of, with xylose, 60, 76, 83
Benzhydrazide, reaction of, with arabinose, 76, 83
Benzhydrazone, formation of, from pentoses, 83
Benzidine, reaction of, with hexoses, 75
with nucleic acids, 81
with pentoses, 81
with riboflavin, 81
Benzoic acid in urine, 110
4-Benzylidine-2-phenyl-S-oxazolone in phenylalanine
synth~sis, 10, 11
-184 -
c
Caffeine, detection of, with potassium-bismuth iodide,
86
in natural materials, S4
Calciferol (see Vitamin Dz)
Carbohydrates, detection of, with amines, 71
by crystal form, 71
difficulties in, 71
by fermentation, 71
wiUlheavy-metal ions, 71
with phenylhydrazine, 71
determination of, with acidic cupric acetate lleagent,
80
with 2,4-dinitrophenylhydrazine, 83
with Molisch reagent, 7S
with, thymol reagent, 78
and formation of benzhydrazones, S3
interference of, in detection, of disaccharides, 72
of pentoses, 75
of sucrose and. raffinose, 75
in solubility separations, 80
interfering substances in detection of, by Molisch
test, 7S
reaction of, with acidic cupric acetate, 73, 80
with alkali, 79
with basic cupric citrate, 72, SO
with diazouracil, 75
with iodine, 72, 74, 78
with methylamine, 72, 79
with CX-naphthol, 77, 7S
with !titric acid, 76
with !titrochromic acid, 72, 79
with phenols, 77
with phenylhydrazine, 76, 82
- 185-
186 -
D
Decalso (see Permutit)
Decarboxylase, L-amino acid, determination of tyrosine
with, 23
,
L-tyrosine, in natural materials, 24
Dehydroascorbic acid, formation of, froin ascorbic
acid, 89
reduction of, with Escherichia coli, 88
7-DehYdrocholesterol, formation of, from 7-hydroxycholesterol, 98
formation of vitamin Ds by irradiation of, 98
Desoxypentose nucleic acids in thymus, 84
Desoxysaccharides, detection of, 73
reaction of, with periodic acid, 73
Dextrin, detection of, 72, 74
formation of, from glycogen, 102
from starch, 102
precipitation of, by ammonium sulfate, 74
Dialuric acid, formation of, from pyrimidines, 87
Diazobenzenesulfonic acid, determination of tyrosine
with, 22
reaction with, of putines, 87
of pyrimidines, 87
Diazouracil, reaction of, with carbohydrates, 75
with raffinose, 81
with sucrose, 81
Dibenzylidenexylose dimethyl acetal, formation of,
from xylose, 77, 83
Dibromoacetaldehyde, reaction of, with 2,4-dinitrophenylhydrazine, 43
Dibromoxyhydrouracil, formation of isobarbituric
acid from, 87
formation of, from pyrimidines, 87
reaction of, with arsenophosphotungstic acid, 87
2,6-Dichlorophenol indophenol, determination of
ascorbic acid with, 88
oxidized form of, 89
reduced form of, 89
2,6-Dichloroquinone chloroimide, determination of
uric acid with, 115
Digitogenin in noncardiac glycosides, 53
Digitonin, precipitation of cholesterol by, 108
Digitoxigenin in cardiac glycosides, 53
Dihydrocho1estero1, .formation of, fronl cholesterol, 108
Dihydroxyaceto'ne, reaction of, with p-hydroxydiphenyl,
44
Dihydroxybenzalacetone, formation of, from salicylaldehyde, 41, 111
1,8-Dihydroxynaphthalene-3,6-disulfonic acid (see
Chromotropic acid)
2,5-Dihdroxyphenylacetic acid, excretion of, in
alkaptonuria, 10
formation of, from phenylalanine, 10
from tyrosine, 10
in urine, 110
Dihydroxyphenylalanine, interference of, in detection
of aspartic acid, 43
2,5-Diketopiperazine (see Glycine anhydride)
E
Egg yolk, cholesterol in, '53
choline in, determination of, 95
Electrodialysis, purification of casein by, 45
separation of arginine by, 43
Electrophoresis, analysis of casein by, 45
End-group method, determination of starch constitution
by, 55
Enzymes, liberation of tyrosine from proteins by, 20
in urine, 110
Ergosterol, antirachitic activity of irradiated, 98
formation of products by irradiation of, 98
International units of vitamin D in, 99
in yeast, 53
Ergothioneine in urine, 110
Erythrocytes, acid phosphatase in, 104
cholesterol in, 108
Erythrodextrin, formation of, from starch, 103
Escherichia coli, decarboxylation of tyrosine by, 24
reduction of dehydroascorbic acid by, 88
Estrone in urine, 53
Ethanolamine, formation of choline from, 95
Ethereal sulfates in urine, 110, 114
Ethyl hydantoate, preparation of, from glycine ester, 4
Ethylene glycol'in preparation of glycylglycine, 6
F
Fat, absorption of, function of cholesterol in, 108
animal, calcification of bones by, 98
blood, in choline deficiency, '95
deposition of, in liver, 95
oxidation of, inhibition of, 49
in urine, 110
Fatty acids in starch, 56
Feathers, cystine in, 25
pigmentation of, in vitamin D deficiency of chick, 99
Feces, occurrence of coprosterol in, 53
Fehling reagent, detection of carbohydrates with, 72
modifications of, 70
Fermentation, determination by, of galactose, 82
of reducing substances, 113
inhibition of anerobic, by ~odium azide, 82
of lacto se, 75
of monosaccharides by zymase, 81
products formed from monosaccharides by, 82
of yeast-fermentable saccharide s, 76
Ferric chloride, detection of acetone with, 113
Ferric thiocyanate, formation in chloride-ion
determination, 19
- 187-
188 -
H
Hair, animal and human, cystine in, 25
isolation of tyrosine from, 25
Hanus method of determination of iodine number, 51
Hemoglobin, determination of papain proteolytic
activity with, 106
Hexoses, determination of, with orcinol, 80
interference of, in detection, of disaccharides, 79
of pentose s, 75
Hippuric acid, determination of liver function with, 114
determination of, in urine, 114
excretion of, after ingestion of foods, 114
formation of, from glycine, 2
in kidney, 9
in liver, 9
synthesis of, from glycine, 9
synthesIs of phenylalaninc< from, 10
in urine, 9, 110, 114
Histamine, reaction of, with bromine-acetic acid, 43
with sodium nitroprusside, 42
Histidine, detection' of, 34, 37
detection of interference of amino acids in, 43
determination of nitrogen in, 13
excretion of, in urine, 114
interference of, in detection, of lysine, 37
of phenylalanine, 42
reaction of, with bromine-acetic acid, 43
with nitrous acid, 15
with sodium nitroprusside, 4~
separation of, with phosphotungstic acid, '34, 35
Histidine phosphotungstate, solubility of, in
hYcb;ochloric acid, 40
Hog hair, isolation of cystine from, 25
Homocystine, formation of methionine from, 96
Homogentisic acid (see 2,5-Dihydroxyphenylacetic
acid)
Hopkins apparatus for condensing vapors, 148
Hormones, adrenal, sterol types of, 53
sex, ralation of cholesterol to, 108
sterol types of, 53
in urine, 110
Iminodiac~lc
189 -
K
Kaufman bromine method of determination of iodine
number, 51
Keratins, cystine in, 25
7-Ketocholesterol, formation of, from cholesterol, 98
formation of 7-hydroxycholesterol from, 98
Ketohexoses and ketopentoses, reaction of, with
resorcinol, 80
Kidney, hemorrhages of, in choline deficiency, 95
Kjeldahl method of determinations, of nitrogen, 13
of protein, 114
of uric acid, 114
Koettstorfer number (see Sapon,ification number)
-190 -
N
a-Naphthol, compounds reacting with, 43
reaction of, with arginine, 42
with carbohydrates, 77
,B-Naphthot,.reaction of, with carbohydrates 77
with 5-(t)ohydroxymethyl) furfural, 78
'
![Link] sulfonic acid (see 1,2,
Naphthoguinone-4-sulfonic acid)
1,2,Naphthoquinone-4-sulfonic acid, determination with,
of amino acids, 114
of cysteine, 28, 36, 41
reaction of, with amino acids, 40
Nerves, choline in, 95
liberation of acetylcholine from, 95
Nessler reagent, detection of ammonia with, 35
determination with, of ammonia, 112
of total nitrogen, 112
interference of sodium chloride with, 115
reaction of, with ammonia, 115
Neurine, formation of, from choline, 95
Neurospora crassa, inhibition of, by phenylalanine
analogues, 10
Nicotinamide, determination of; by microbiological
assay, 91
Nicotine, preparation of nicotinic acid from 91
Nicotinic acid, in coenzymes I and II, 91 '
daily allowances of, for humans, 91
deficiency symptoms of, 91
[Link]?n of, with cyanogen bromide-pyridine, 91
by Ill1croblOlogical assay, 91
by photometry, 91
in foods, 91
formation from, of N'-methylnicotinamide 96
of nicotinuric acid, 96
'
of trigonelline, 96
'.
formation of, from hydroxyanthranilic acid 91
function of, in glycolysiS, 91
'
in respiration. 91
history of, 92
interference of, in utilization of choline 95
isolation of, from rice, 91
'
methylbetaine of, 96
in natural materials, 91
preparation of, from nicotine, 91
in purified amylase, 102
treatment of natural products for assay of, 94
Nicotinuric acid, determination of, by microbiological
assay, 91
formation of, from nicotinic acid, 96
Ninhydrine, determination of amino acids with 114
reaction of, with alanine, 44
'
with amino acids, 33, 39.
with glutamic acid, 43
with glycine, 44
Nitric acid, detection with, of aromatic amino acids,
33,40
of proteins, 40, Ill, 113
p-Nitrobenzoyl chloride, reaction of, with amino acids 40
Nitrochromic acid, compounds reacting with 79
'
~eaction of, with carbohydrates, 72, 79 '
NItrogen, detection of, in starch preparations 56
determination of, with Nessler reagent, 112'
-191-
192 -
Q
Quinhydrone electrode, determination of pH with, 150
193 -
R
Rabbit. utilization of mannose by. 64
Rabbit hair. cystine in, 25
Rabbit intestine, determination of acetylcholine
with, 96
Raffinose, detection of, 72. 74-76
reaction of, with diazouracil, 81
solubility of. 73. 80
Rat. bioassay of vitamin D by line test with, 99
cholesterol in blood of, 108
dispensability of glYCine for growth of, 2
exsanguination of, 104
inhibition of. by phenylalanine analogues, 10
lethal dose Of choline for, 95
phosphatase in blood of, 104
potency of vitamins D2 and D3 for, 99
utilization by, of lactose, 66
of xylose ...60
vitamin D unit for, 99
Rat hair. cystine in, 25
Reducing substances. detection of, 111. 113
with bismuth reagent, 113
with copper reagent, 113
by optical rotation, 113
with phenylhydrazine, 113
determination of. with dinitrosalicylic acid, 114
by fermentation with yeast. 113
with ferricyanide, 114
by iodimetry, 114
with phosphomolybdic acid, 114
by photometry, 114
with picric. acid, 114
in urine. 114
formation of cuprous thiocyanate in oxidation of, 114
in potatoes, 56
in urine in pathological conditions, 114
Reinecke salt. determination of choline with, 96
Resorcinol, reaction of, with carbohydrates, 74, 77
with furfural, 81
with ketohexoses, 80
with ketopentoses, 80
Rfvalues of purines and pyrimidines, 85, 86
Rhamnose, detection of, 72, 75
formation of furfural from, 77
solubility of, 73, 80
Rhamnose benzhydrazone, formation of. 83
Riboflavin, determination of, by microbiological
assay, 92
reaction of, with benzidine, 81
Ribose in pentose nucleic acids, 84
Rice, isolation of nicotinic acid from. 91
Rotation (see Specific rotation)
s
Saccharic acid, formation of, from glucose, 82
monopotassium salt of, 82
silver salt of, 82
Saccharin, taste of, 2
Saccharogenic activity, variations in, 102
Saccharogenic time, definition of, 102
Salicylaldehyde, detection of acetone with, 111
formation>of dihydroxybenzalacetone from, 41, 111
reaction of, with acetone, 36, 38,41,44, 111
with methylethylketone, 36, 38, 41. 44
Saliva, isolation of amylases from, 102
Sapogenins, sterol types of, 53
Saponification number, determination of, 50
Sarcosine, formation of choline from. 95
194 -
T
Tachysterol, formation of, from ergosterol, 98
Takadiastase, solubilization of starch by, 58
Taurine in urine, 110
Tea, methylpurines in, 84
Testosterone in urine, 53
2,3,4,6,-Tetramethylglucose in methylated starch, 55
Theelin (see Estrone)
Theobromine, detection of, with potassium-bismuth
iodide, 86
in natural materials, 84 .
Theophylline, detection of, with potassium-bismuth
iodide, 86
in natural materials, 84
reaction of, with diazobenzenesulfonic acid, 87
Thiocyanates in urine, 110
L-Threo-3-ketohexuronic acid lactone (see Ascorbic
acid)
Threonine, detection of, 35, 38
detection of interfering substances in, 44
determination of, with periodic acid, 41
195 -
Tyrosine, preparation of tYramine hydrochloride from, 24 Urine, determination in, of acetone, 113
of acidity, 114
purification of, by crystallization from water, 20
of amino acid 50 39, 114
as ethyl ester hydrochloride, 20
with ninhydrin, 39
as mercuric chloride complex, 20
of ammonia, 112
as phosphotungstate, 20
of creatinine, 113
reaction of, with :Mi11on reagent, 36, 42
of hippuric acid, 114
with nitric acid, 33, 40
of nicotinamide, 91
with phosphotungstic acid, 28
of nicotinic acid, 91
in silk fibroin, 20
of nicotinur~c acid, 91
solubility of, in water, 40
of organic acids, 114
L-Tyrosine, separation of, from L-cystine, 25
of protein, 114
Tyrosine decamoxylase, in natural materials, 24
of purines, 114
Tyrosine ethyl ester, extraction of, from crude cystine
oJ reducing substances, 114
ethyl ester, 25
of total nitrogen, 112
Tyrosinuria, excretion of p-hydroxyphenylpyruvic acid
tlf ure a, 112
in, 10
of uric _acid, 113, 114
excretion in, of amino acids, 114
of ammonia, 115
of aspartic acid, 114
Ultraviolet light, formation of antirachitic substances
of cl}oline, 95
in skin by, 99'
of glutamic acid, 114
Uracil, detection of, with iron-bromine-ammonia, 85
of histilline, 114
determination of, with arsenophosphotungstic acid, 87
of methionine, 114
formation of, from cytosine, 86
of phosphatase, 104
formation of dialuric acid from, 86
formation of, by filtration of plasma, 110
in pentose nucleic acids, 84
glucose in, 110-112
reactions of, with bromine water, 86
hippuric acid in, 9, 110, 114
With diazobenzenesulfonic acid, 87
inorganic ions in, 110, 114
Uranium phosphate, precipitation of phosphate by, 47
nitrogen of, on high- and low-protein diets,
Uranyl salts, identification of nucleic acid derivatives
total, 111
as, 84, 87
undetermined, 111
Urea, decomposition of, with urease, 112
uric acid, 111
excretion of, in normal and pathological conditions, 115 nitrogen comp ounds in, 110, 115
formation of ammonia from, 112
organic acids in, 110, 114
reaction of, with nitrous acia, 15
pH of, 110
in urine, 110-112, 115
properties of normal and pathological, 110
determination of, 112
purines in, 110, 114
Urea nitrogen, excretion of, 114
reabsorption of constituents in, 110
in urine, on nigh- and low-protein diets, III
specific gravity of, 110
Urease, decomposition of urea with, 112
on high- and low-protein diets, 111
Uric acid, copper salt of, 114
te stosterone in, 53
I
determination of, as ammonium urate, 115
tyrosine decarboxylase in, 24
with arsenophosphotungstic acid, 115
volume of, excreted under ,normal and pathological
with 2,6-dichloroquinone chloroimide, 115
condition$, 110
with phosphotungstic acid, 115
on high- and low-protein diets, 111
by photometry, 113, 115
influence of drugs on, 110
with potaSSium permanganate, 115'
Urocanic acid in urine, 110
in urine, 113, 114
formation of, from 8-aminoxanthine, 87
from glycine, 2
hydrochloride of, 114
Vaccinia virus, inhibition of"by phenylalanine
in urine, 110, 111, 113, 115
analogues, 10
Uric acid nitrogen, determination of, by Kjeldahl
Valine, detection of, 35, 38
method, 114
determination of, by photometry, 41
in urine, on high- and low-protein diets, 111
Van Slyke factors, conversion of nitrogen volume to
Uridine, determinahpn of, with arsenophosphotungstic
weight by, 17, 18
acid, 87
Van Slyke method of determination of amino nitrogen, 15
Uridylic acid, destruction of, during hydrolysis, 86
Vapors, apparatus (Hopkins) for condensing, 148
Urine, acetone bodies in, 110, 111, 113
Vetch sprouts, isolation of asparagine monohydrate
acid content of, in normal and pathological
from, 29
conditions, 110
Vitamin C (see Ascorbic acid)
alkalinity, causes of, 110
Vitamin D, allowances of, for humans, 99
allantoin in, 110, 113
bioassay of, by chick bone-ash method, 99
ammonia nitrogen of, on high- and low-protein
by ratline test, 99
diets, 111
determination of,_by chick bone-ash method, 98
bacterial decomposition of, 110
comparative studies on, 99
carbohydrates in, 110, 111, 113
with turkeys, 99
creatinine nitrogen of, on high- and low-protein
discovery of, 98
diets, 111
disease symptoms of chicks in deficiency of, 99
detection of pentoses in, 81
196 -
x
Xanthine, detection of, with diazobenzenesulfonic
acid, 65
with silver salt reagent, 85
formation of, from guanine, 84
z
Zymase, fermentation of monosac"h"ri ""'''
- 197-
hv
oR 1
(Date)
(Desk No.)
(Student's name)
Cname)B
(formula)
(Date)
(Student's name)
(Desk No.)
Unknown ........................................................................... .
No. _ __
No., _ __
No. _ __
(name)Q
Per cent purity of the unknown b
(formula)
................................... .
ber percentage
(Desk No.)
(Date)
Trial ....................................................................................
(Student's name)
. .....................
e ..............................................................
(name)b
(formula)
DATA
(Date)
S~EET
FOR EXPERIMENT 11
(Desk No.)
Sample ..............................................................................
G. of hydrochloride unknown ......................................... .
G. of sample in 2S-ml. aliquot ...................................... ..
Ml. of _ _ _ N AgNO s ................................,.................. ..
Ml. of _ _ _N KSCN
First trial .................................................... ;................ .
Second trial ................................................................. .
Third trial ......... :......................................................... ".
Average ......................................................................... .
Equivalents of AgNO s .................................................... ..
Average equivalents of KSCN ....................................... .
Equivalents of AgNO s used by chloride ....................... .
Per cent of chloride in unknown hydrochloride ........... .
Theoretical percentage of chloride in unknown ........ ..
(name)B
(formula)
(Student's name)
Unknown
No. _ _
Unknown
No. _ _
Unknown
No._ _
(Date)
(Desk No.)
Sample ................................................................................
(Student's name)
Unknown
No. _ _
Unknown
No.
Unknown
No. - -
20.0
20.0
20.0
--
(Date)
(Student's name)
(Desk No.)
Sample ............................................................................... .
Casein
G. of casein ..................................................................... .
G. of casein in 10-ml. aliquot ....................................... .
20.0
................................................... .
20.0
(Date)
(Desk No.)
Sample ...............................................................................
(Student's name)
Unknown
No. - -
Un~nown
No. _ _
Unknown
No. - -
(Date)
(Student's name)
(Desk No.)
Sample ................................................................................
Unknown
No. _ _
Unknown
No.
Unknown
No.- -
20.0
20.0
20.0
--
(Date)
(Desk No.)
(Student's name)
STANDARD~CURVE DATA
MI. of
pH
10
............................................
Sample ......................................................................
Unknown asparagine
monohydrate No.
Unknown + added
asparagine monohydrate
5~mL
ammonia~borate
solution ..............................
"
"To be obtained from the instructor after grams of asparagine monohydrate recovered has been determined
and submitted.
(Date)
(Desk No.)
Sample ................................................................................. .
(Student's name)
20.0
20.0
G. of casein ....................................................................... .
G. of casein in 10-ml. aliquot ......................................... .
Mm. thickness of colored standard
phosphate solution......................................................... .
Depth in mm. of colored unknown
soiution ........................................................................... .
Mg. of phosphorus per 10.0 ml. of standard
phosphate solution ......................................................... .
Mg. of phosphorus per 10.0 ml. of unknown
solution analyzed. .......................................................... .
Per cent phosphorus in casein.......................................... .
G. of phosphorus recovered (from
10-mi. aliquo't) ............................................................... .
G. of phosphorus added B ................................................... .
Per cent recovery of phosphorus .................................... ..
[Link] be obtained from the instructor after grams of phosphorus recovered has been determined and submitted.
DATA SHEET
(Date)
FOR EXPERIMENT 24
(Desk No.)
Trial ................................................................................. .
G. of unknown lipid, No. _ _ _................................... .
Ml. of___N KOH to titrate sample ......................... .
MI. of- - - N KOH to titrate blank
First trial ..................................................................... .
Second trial ................................................................. .
Average ......................................................................... .
Ml. of
N KOH to titrate sample,
corrected for blank value .......................................... ..
Equivalents of fatty acid in sample ............................ ..
Acid number of lipid ...................................................... .
Average acid number of lipid ....................................... ..
(Student's name)
(Date)
Trial .........................................................................................
(Student IS name)
(Desk No.)
, 2
. ...................................
(Date)
(Desk No.)
Trial ................................................................................. .
G. of unknown lipid, No.
' - - - ............................... .
Ml. of
N Na~203 to titrate blank
First trial ................................................................. :... .
MI. of
N Na 2SPl equivalent to IBr
absorbed by sample ...................................................... .
Equfvalents of IBr absorbed by sample ...................... ..
Iodine number of lipid ..................................................... .
Average iodine number of lipid .................................... ..
(Student's name)
EXPERIMENT 27
(Desk No.)
(Date)
(Student's name)
(Date)
(Student IS name)
(Desk No.)
1.
g. of purified, dry starch was prepared from 500 g. of macerated slices of peeled
I
potatoes of the _ _ _ _ _ _ _ _variety.
2. In the color tests the following color changes were noted: sliced potatoes standing in air,
________ to
; sliced potatoes suspended in water and heated for 15
min., _ _ _ _ _ __ to
; and the supernatant liquids obtained in the preparation of purified starch, ________ to _ _ _ _ _---,_3.
g. of dry colloid was obtained in the last section from 100 mI. of aqueous extract.
The total colloid in 500 g. of potato slices is calculated to be
g.
4. The test for nitrogen in dry starch was ________ and in dry colloid _ _ _ _ __
(Indicate each answer as positive or negative.)
5. Observations on the reducing substances, presumably principally glucose, present in aqueous
extracts of potato pulp and from starch purification:
Aqueous extract of
potato pulp
Results from
Benedict test
"
Results from
Benedict test
(Date)
(Student's name)
(Desk No.)
a,
Sample No ................
Log
G. of sample ............
a,
Temperature of
solution ................
Specific rotation
Time, min.
Observed rotation
(corrected)
(Water added)
(Most sample dissolved)
.
_-
[a]D ..........................................
(Desk No.)
(Date)
Sample ................................................................................. .
(Student's name)
Unhydrolyzed
Corn Corn sirup
sirup
+ glucose
Hydrolyzed
Corn Corn sirup
sirup + glucose
20.0
20.0
2-m~.
aliquot .................................. ..
20.0
20.0
(Date)
(Student's name)
(Desk No.)
0.070
Sample ........................................................................
Unhydrolyzed
Corn sirup
Corn
sirup
+ glucose
. .....................
(brand)
0.120
0.300
0.200
.
--
Hydrolyzed
Corn sirup
Corn
sirup
+ glucose
--
(Date)
(Desk No.)
Sample ................................................................
Trial .................. ,:;, ..............................................
,
Ml. of
(Student's name)
Grapefruit juice
.........
MI. of
N KIO l to titrate aliquot
after dye titration..........................................
Ml. of _ _ N KIO, equivalent to
ascorbic acid in aliquot ..............................
Average equivalents of
N KI0 3
equivalent to ascorbic acid
in aliquot.............................................. :.........
Equivalents of KI0 3 in average
titration volume ............................................
Mg. of ascorbic acid in aliquot ......................
"
(Date)
(Desk No.)
(Student's name)
Ml. of
per tube
1 ....................................................................
2 ....................................................................
Average........................................................
, Ml. of vi tamin
solution
N NaOH
Ml. of
to titrate sample
<
Average
Nicotinic acid
p.g per
tube
0.5
1.0
1.5
2.0
2.5
Mean ......................................................................................
p.g per
ml.
Deviation from
the mean p.g
per mI.
Ml. of
NNaOH
to titrate sample
Average
Nicotinic aciq
p.g per
tube
p.g perl
ml.
Deviation
ftom the
mean p.g per ml.
Per cent
mean
deviation
0.5
1.0
1.5
2.0
2.5
Mean ...........................................................................................
Furnished by instructor after milligrams of nicotinic acid recovered has been petermined and submitted.
(Date)
(Student's name)
(Desk No.)
0.5
Choline
,
Ml. of sample solution
Mg. of
mycelium
1.0
2.0
3.0
4.0
5.0
Mean...........................................................................
Deviation from
the mean ILg per
ml.
Choline
Ml. of samplerecovery solution
Mg. of
mycelium
Deviation from
the mean fLg per
m1.
1.0
2.0
3.0
4.0
5.0
Mean............................................................................
................................................................................ ..
(Date)
(Student's name)
{Desk No.}
Tibiae
Left
Right
. .................
Middle toes
Left
Right
(Date)
(Student's name)
(Desk No.)
Time
Starch solution, per cent
Start
1.5 .............................. ..
2.0 .............................. ..
2.5 .............................. ..
Unknown No. _ _ __
Unknown No. _ _ __
Per cent starch in
Unknown No. _ _ __
Unknown No._ _ __
Ml. to which _ _ __;ml. saliva diluted .............................................. ..
Units of amylase activity ......................................................................... .
Finish
Elapsed
(Date)
(Desk No.)
Sample ..................................................................
(Student's name)
TotalS
phosphorus
Totals plus
added phos- Preformed
phosphorus
phorus
Blank
phosphorus
20.0
20.0
20.0
20.0
--
--
'.
--
(Desk No.)
(Student's name)
,
I
Optical density
Sample ..................................................................
Total a plus
Total a
added phos- Preformed
phosphorus
phosphorus
phorus
Blank
phosphorus
---
---
(Date)
(Desk No.)
(Student's name)
Ml. of standard (
Optical density......................................................................................
Sample ..............................................................
Serum
'.
s~bmitted.
(Date)
(Desk No.)
(Student's name)
Diet ................................................................................................. .
Volume of urine excreted in 24 hr..............................................
Final volume of urine after dilution ........................................... .
Nitrogen found
Constituent
Mg.
Per cent
of total
Nitrogen recovery
Mg. N/50 mI.
Per cent
recovery
Found
Added a
Total nitrogen......................
Urea ......................................
Ammonia ..............................
Creatinine ............................
Uric acid ..............................
Undetermined nitrogen
(by difference) ................
OrigimH urine
Albumin b ............................. .
Glucose c
Acetone b
............................. .
9Furnished by the instructor after milligrams of nitrogen recovered has been calculated and submitted.
b