Full text of "Food and drugs"

See other formats

FOOD ANP DRUGS

VOLUME !
THE ANALYSIS OF FOOD AND DRUGS

CCHEMICAL AND MICROSCOPICAL)

.»,Xji^,iUi,.

„ha,i

FOOD AND DRUGS

UNIFORM WITH THIS VOLUME

FOOD AND DRUGS

VOL. II

THE SALE OF FOOD AND DRUGS ACTS, 1875-1907

Royal Svo. 188 pages

CONTENTS

The Sale of Food and Drugs Act, 1875.

The Sale of Food and Drugs Act Amendment Act, 1879.

The Sale of Food and Drugs Act, 1899.

The Margarine Act, 1887.

The Butter and Margarine Act, 1907.

Index of Cases.

Price 75. 6d. net {post free, js. lod. home, Ss. abroad)

FOOD AND DRUGS

ERNEST Jf PARRY, B.Sc. (Lond), F.I.C, F.C.S.,

MEMBER OF THE SOCIETY OF PUBLIC ANALYSTS, ETC.,
BARRISTER«AT-LAW, OF CRAY'S INN

VOLUME I

THE ANALYSIS OF FOOD AND DRUGS

CHEMICAL AND MICROSCOPICAL)

WITH FIFTY-NINE ILLUSTRATIONS

LONDON
SCOTT, GREENWOOD & SON

THE OIL AND COLOUR TRADES JOURNAL" OFFICES
8 BROADWAY, LUDGATE, E.G.
1911
{All rights reserved)

(

PEEFACE.

In the preparation of this work the aim of the Author has been
to deal with the question of food and drugs from both the
chemical and the legal points of view, in a manner that will be
of assistance to those entrusted with the administration of the
Sale of Food and Drugs Act, primarily, and also to those
engaged in the examination of food and drugs for other
purposes.

It is hoped that the application of chemical knowledge to the
legal aspect of the question, as set out in the second volume
may prove of service to counsel and solicitors engaged in cases
under the Acts.

I have to acknowledge the kindness of Professor Greenish
in writing the chapter on microscopic analysis ; and to thank
Messrs. Cecil Cribb, Peter MacEwan, E. A. Pinchin, A. Searl,
and J. C. Umney, for much assistance in the reading of
proofs ; and Mr. H. Droop Pichmond for several valued sug-
gestions. My acknowledgments are also due to the Editor of
the " Pharmaceutical Journal " for permission to use numerous
illustrations for which he holds the copyright.

I am also much indebted to Mr. R. J. Preston, LL.D.
(Lond.), for reading the proofs of Volume II.

ERNEST J. PARRY.

Thanet House,

56a. Great Dover Street,

London, S.E., J^lhJ, 1911.

CONTENTS.

PAET I.
CHAPTER I.

PAOB

Tea, Cocoa and Chocolate, Cocoa Butter, Coffee 1

CHAPTER II.
Milk, Cheese, Butter, Lard, Suet, Olive Oil 41

CHAPTER III.

The Carbohydrate Foods : Cane Sugar, Molasses, Glucose, Honey,
Sugar of Milk, Maltose and Malt Extract, The Starches and
Starchy Foods, Wheat Flour and Bread IIT

CHAPTER IV.

Spices, Flavouring Essences, etc. : Ginger, Pepper, Mustard, Cloves,
Allspice, Cinnamon, Nutmegs, Mace, Cochineal, Saffron, Tub-
MERic, Annatto, Vinegar, Flavouring Essences .... 191

CHAPTER V.

Alcoholic Beverages : Brandy, Whisky, Rum, Gin, Wine, Malt

Liquors, Cider 275

CHAPTER VI.
Flesh Foods, Extract of Meat, Gelatine 368

CHAPTER VIL

Microscopical Analysis 416

vii

viii CONTENTS.

PAET II— DEUGS.
CHAPTER VIII.

PAOB

Crude Drugs : Acacia, Tragacanth, Ammoniacum, Araroba, Asafcetida,
Balsam op Peru, Balsam of Tolu, Benzoin, Cannabis Indica,
Catechu, Cardamoms, Copaiba, Creosote, Cubebs, Galbanum, Guai-
ACUM, Gamboge, Gentian, Kino, Liquorice Root, Male Fern,
Musk, Myrrh, Pepsine, Canada Turpentine, Burgundy Pitch,
Liquid Tar, Resin, Thus, Scammony, Senna, Aromatic Spirit of
Ammonia, Spirit of Nitrous Ether, Squills, Storax, Terebene,
Standards for Tinctures, Liquid Extracts, and Concentrated
Liquors 427

CHAPTER IX.

Drugs containing Alkaloids, etc., capable of Approximate Deter-
mination : Aconite, Aloes, Belladonna, Cantharides, Cinchona,
Coca, Colchicum, Colocynth, Conium, Digitalis, Elaterium, Ergot,
Gelsemium, Hydrastis, Hyoscyamus, Ipecacuanha and its Pre-
parations, Jaborandi, Lobelia, Jalap, Nux Vomica, Opium and its
Preparations, Podophyllum, Rhubarb, Stramonium, Strophanthus 499

CHAPTER X.
The Essential Oils of the British Pharmacopceia .... 606

CHAPTER XI.
The Fixed Oils, Pats, and Waxes of the British Pharmacopceia . 626

CHAPTER Xll.
The Chemicals op the Phabmacopgeia 648

Table op Chemicals 539

Index y-^y

PART I.
CHAPTER I.

TEA, COCOA AND COFFEE.

TEA.

Tea, as used as a beverage, is the dried and prepared leaf of various
species of Thea, a shrub belonging to the genus Camellia. Thea
sinensis is the principal species from which the tea leaf is derived, but
several other species are also employed, of which the most well
recognized are Thea viridis, T. Bohea, and T. Assamica. The differences
between the many kinds of tea known in commerce do not, however,
depend on any important botanical distinctions, but on methods of
preparation, age of the plant, time of gathering of the leaf, etc. The
finest teas are derived from the young leaves of young shrubs, the old
leaves and leaves of old plants being Of inferior quality. Black tea has
undergone a certain amount of fermentation before it is dried, whilst
green tea is the result of rapid drying of the leaves before fermentation
has set in.

The only questions in reference to tea that the analyst is called
upon to decide are in regard to its purity or otherwise, and o3casionally
in reference to the amount of caffeine (theine) and tannic acid present.
The quality of tea has practically no relation to its analysis, and the
expert tea taster is the recognized authority in reference to the quality
and value of tea.

The adulteration of tea, especially when it was a very expensive
product, was most gross and very common, and the importance of the
matter was recognized as far back as 1724 when the Adulteration of
Tea and Coffee Act was passed (11 Geo. 1, c. 30). Sect. 5 of this Act
imposed a penalty of £100, with forfeiture of the tea, for any adulteration
whatsoever. In 1730 an Adulteration of Tea Act was passed (4 Geo.
2, c. 14), which goes further and inflicts a penalty of £10 per pound
weight of the tea adulterated, and makes penal the use of exhausted
tea leaves. In 3 776 a further Act was passed (17 Geo. 3, c. 29). The
former Acts only touched dealers in tea, whilst this Act imposed a
penalty on any person who adulterated the tea, or had it in possession.
In 1875, the ISale.of Food and Drugs Act of that year, sect. 30, pro-
vided for the examination of all tea imported into this country by
the Customs authorities, since which date adulteration has been far
less common.

Sect. 31 of the same Act defines exhausted tea, as " any tea which
VOL. I. 1

2 FOOD AND DRUGS.

has been deprived of its proper quality, strength or virtue by steeping,
infusion, decoction, or other means ".

Apart from the analysis of tea as an article of food, tea is frequently
tested for its percentage of caffeine. Caffeine of commerce (which
substance is identical with theine) is manufactured from tea, and tea of
inferior quality is used for this purpose. It is allowed into this country
duty free if denatured by the addition of such substances as render it
impossible for use as a beverage. In these cases the value of the tea
is obviously in direct proportion to its caffeine content.

The Comiiositioji of Tea. — According to Eder (" Dingl. Poly. Jour-
nal," 131, 445, 526) the average composition of tea is as follows: —

Soluble in water.

Insoluble

in water.

Moisture

10 per cent

Chlorophyll

1 1-8-2 -2

per cent

Tannin

Wax

0-2

Gallic and oxalic acids |
Quercetin /

0-2

Resin

3-0

Colouring matter

1-8

Boheic acid

Extractive matter

16-0

Caffeine

Cellulose

Essential oil

0-6

Albuminous matter

12-7

Albuminous matter

Mineral matter

Carbohydrates

3 to 4 „

Mineral matter

1-7

Of these, caffeine is the principal constituent of tea. It is an alka-
loidal compound of the uric acid series (a trimethyl-xanthine) and of the
formula CgH^^N^O.,. It was originally described, when separated from
tea, under the name of theine, but was later shown to be identical with
caffeine, the alkaloid of coffee, under which name it is now generally
known. It crystallizes with one molecule of water. Its melting-point
is about 233°. In large doses caffeine exerts a poisonous effect, but
in moderate doses finds useful employment in medicine.

The essential oil which has been stated by Eder to exist to the
extent of 0"6 per cent was possibly obtained from a scented tea.
At all events, Van Eombugh only obtained 0'006 per cent from a
genuine tea. Schimmel & Co. have examined two samples which they
obtained from fermented leaves and suggest that the oil may be a re-
sultant of the fermentation process. These two oils had specific
gravities 0*866 and 0*856 respectively, and were only faintly optically
active. The oil contains methyl salicylate and an alcoholic body of the
formula C,.Hp^O. Acetone and methyl alcohol were found in the distilla-
tion water.

Quercetin is a compound, possibly of the formula C^^Hj^O-, which
is found to a minute extent in tea. It is of no importance from the
analytical point of view.

A glucoside quercitrin, C.,iH.j.,Op_,, is also said to be present in minute
amount. On hydrolysis this splits up into rhamnose and the above-
mentioned quercetin.

TEA. 3

The remaining constituents of tea do not require discussion.

The following figures represent ten samples of tea — five of ordinary
Ceylon tea sold in shops at from Is. 6d. to 2s. 6d. per lb., and five of
ordinary China tea of the values 2s. 6d. to 3s. 6d. per lb., which have
been analysed by the author : —

Ceylon Tea.

China Tea.

Per

Per Per Per

Per

Per Per

Per

cent

cent cent cent

cent

cent cent

cent

Moisture

7-9

8-5 10 11-1

7-7

8-2

9-8

7-6 8-8

Caffeine-

3-3

2-9 3-7 2-6

2-9

2-5

2-9

3-6 2-9

3-8

Tannin

12-5

14-1 16 13-8

17-5

9-4

10-2 9

Ash

4-9

5-2 5-5 4-9

5-8

5-4

5-7

51 5-2

Aqueous extract

32-4

42 32 40-3

40-9

38 36

Total nitrogen

5-5

5-3 6-2 5-9

5-9

5-3

5-6

5-9 5-2

Ash soluble in water

3-2

2-9 31 2-7

3-2

3-3 2-9

Ash insoluble in acid

0-2

0-46 0-29 0-33

0-2

0-3

0-4

0-38 0-5

0-2

The following table is quoted from " Food Adulteration " by J. P.
Battershall p. 28, as embodying the results of the analyses of samples
representing 2414 packages of high quality Indian tea : —

Minimum.

Maximum.

Average.

Per cent

Moisture

5-83

6-32

5-94

Insoluble leaf

4712

55-87

51-91

Extractive

37-80

40-35

38-84

Tannin

13-04

18-87

15-32

Caffeine

1-88

3-24

2-74

Ash — total

505

6-02

5-61

Soluble in H.^O

3-12

4-28

3-52

Insoluble in acid

0-12

0-30

0-18

The infusion of tea, as used as a beverage, does not contain the
whole of the soluble constituents of the leaf, since the conditions of
the extraction are not such as to entirely exhaust the tea.

The following table by Gfeisler (" Analyst," ix. 221) shows the char-
acters of the infusion made by allowing 100 parts of water to stand for
ten minutes on one part of tea. The water was distilled, and heated
to boiling-point. The figure " ratio to total " indicates the percentage
found in the infusion of that present in the tea : —

FOOD AND DKUGS.

Variety of Tea.

Extract.

Tannin.

Caffeine.

Ash.

Infusion.

Ratio to
Total.

Infusion.

Ratio to
Total.

Infusion.

Ratio to

Total.

Per

cent

Ceylon Pekoe Tips

33-25

76-6

17-19

75-3

2-44

3-44

91-0

Assam

29-15

78-6

11-48

60-8

3-30

3-80

70-0

28-57

9-5

58-4

2-75

4-40

79-5

Moyone

37-32

73-2

16-79

87-8

2-95

4-60

65-8

28-07

79-4

9-27

77-7

1-67

4-02

66-1

Japanese

31-76

75-6

11-23

74-5

2-17

4-27

80-8

34-37

79-6

13-41

94-4

2-07

3-67

63-6

Fprmosa

33-62

75-9

12-91

75-6

2-50

4-00

71-3

33-30

73-7

13-75

68-5

2-42

3-97

66-5

29-00

68-6

9-6

59-6

'^-12

3-66

62-3

Amoy

27-40

60-9

10-12

1-92

3-72

68-5

24-50

60-6

7-53

55-6

1-70

3-25

58-9

Moning

24-25

70-6

5-40

41-7

2-87

4-13

73-7

21-55

57-8

4-44

32-0

2-77

3-70

63-5

••

21-02

68-6

5-55

45-2

2-33

8-22

58-3

Kaisow

23-26

64-1

4-05

38-5

2-35

3-30

59-9

Moning

19-60

72-2

4-50

52-9

1-96

2-88

46-8

The following figures were obtained at the laboratories of the
Imperial Institute for teas grown in the Nyasaland Protectorate and
in Natal. In all cases the soluble extract is not the true extractive,
but the soluble matter extracted by infusing the tea in 100 times its
weight of boiling water for ten minutes : —

Nyasaland

Tea.

Tannin

No.

Description.

Moisture.

Ash.

Caffeine.

determined
by Eder's
methi^.

Soluble
Extract.

Per cent

Percent

Per cent

\" Orange
/Pekoe "

8-26

503

3-68

10-5

26-5

7-84

5-28

3-54

10-4

25-4

\" Broken
/Mixed"

8-32

509

3-35

9-5

23-7

7-77

5-60

3-22

9-8

23 0

V' Dust
jFannings "'

, 8-68

5-20

308

10-3

29-6

8-43

5-17

3-19

10-6

28-8

TEA.

Natal Tea.

Percentages calculated on material

Estate.

Description.

•i

dried at 100° C.

Ash.

Extract.2

Caffeine.

Tannin.'

Kearnsey

Grade 1 ^

9-1

5-8

26-1

8-9

7-8

„

Grade 2 i

7-6

5-6

28-8

3-6

6-8

Grade 3 1

7-4

5-2

27-4

6-7

Grade 41

8-7

5-9

25-0

3-4

6-8

„

Flowery Pekoe

7-6

5-1

not determined

7-0

» •

Broken Pekoe

6-9

5-8

not determined

7-8

Barnsdale

Pekoe

5-9G

5-8

20-2 ' 4-8

10-5

CUfton

j Pekoe

6-2

4-8

31-4

not deter-
mined

18 0

Barnsdale

Golden Pekoe

5-5

28-0

4.4

11-5

Flowery Pekoe

6-1

5-3

27-0

4-2

11-6

Aroma

Pekoe Souchong

7-1

5-5

24-3

10-4

Fine Natal Souchong

8-0

5-0

20-9

4-1

10-1

Barrow Green

Choicest Golden Pekoe

7-7

5-2

83-0

4.4

10-8

Average

7-1

5-4

27-1

4-0

9-2

Adulterants of Tea. — The adulterants that have been met with
from time to time in tea are either mineral matter added for the pur-
pose (1) of increasing its weight, (2) of causing a more complete ex-
traction of the colouring matter or the tannin, (3) of improving the
appearance, by the process of facing ; or organic matter added for the
same purposes.

Amongst mineral matters, the following have been met with :
sand, magnetic iron ore and brass filings (!), sodium carbonate, borax,
steatite, and prussian blue. The only organic matters that come into
serious consideration are exhausted tea leaves, the leaves of other plants
than the tea shrub, prussian blue or indigo for facing, and, rarely,
astringent matter such as gambier and catechu to increase the astring-
ency.

The Analysis of Tea.

Moisture. — The moisture in the tea should average about 6-8 per
cent, rarely up to 10 per cent, and never exceeding 12 per cent.

Mineral Matter. — Five grms. of the sample should be ignited, and

1 These four samples were taken from specimens in the Natal Court of the
Imperial Institute ; the remainder were from the South African Products Exhibi-
tion.

2 " Extractive matter," or " Extract," is the percentage dissolved by treating
a given quantity of the tea with one hundred times its weight of boiling water, and
allowing it to infuse for ten minutes.

=' Determined by Procter's modification of Lowenthal's process.

FOOD AND DEUGS.

the ash examined. The total ash (which is often green owing to the
presence of manganese or copper) should be weighed ; then the water-
soluble ash determined, and lastly the amount of siliceous matter.
The alkalinity of the ash should also be determined by titration with
ilecinormal sulphuric acid, using methyl-orange as indicator. The total
ash varies from about 5 per cent to 7 per cent, rarely reaching 8 per
cent ; of this about half is soluble in water. The Society of Public
Analysts suggested (in 1874) 8 per cent as the maximum ash of tea
which had been dried at 100° of which at least 3 per cent should be
soluble in water. The following figures represent the results obtained
on 9 samples of tea in the author's laboratory : —

Moisture.

Total Ash.

Soluble in Water.

SiUceous Matter.

Alkalinity as KaO.

Per

cent

7-72

4-0

0-39

2-21

7-5

7-61

3-7

0-41

2-31

9-0

6-04

3-2

0-67

1-87

6-4

6-33

4-15

0-62

1-71

5-8

7-81

4-01

0-24

2-22

8-2

5-42

2-8

0-19

1-28

6-6

6-11

2-79

0-33

1-41

7-1

5-82

2-72

0-71

1-49

8-6

5-91

3-18

0-75

1-81

Sheridan, in a private communication to the late A. H. Allen, gave
the following results of the examination of commercial black teas in
the Customs Laboratory : —

Tea.-

Total
Ash.

Ash Soluble in
Water.

Siliceous Matter.

Extract.

Per

cent

Indian

5-40

3-20

0-45

40-49

))

610

3-30

0-90

29-32

5-70

315

0-60

39-44

5-75

3-25

0-70

38-78

Ceylon

5-50

3-20

0-20

42-90

5-40

3-35

0-25

38-24

5-60

3-40

0-30

37-98

Japan

6-12

3-15

0-95

29-80

Java

5-60

3-05

0-50

34-60

7-65

3-75

105

30-72

China

5-70

3-25

0-50

32-95

5-85

2-95

1-00

31-71

5-60

305

0-65

33-57

5-65

3-20

0-70

34-10

5-45

3 05

0-55

35-70

Port Natal

5-65

3-10

0-45

34-80

TEA.

The ash of three samples of tea examined by the author contained
the following (average) : —

Calcium (as oxide)

Ma<,'nesium (as oxide)

Fe,0,

Manganese (as MnoO-)

P2O, .

SO,'

Chlorine

Alkalies (as K.O)

Silica

9-3 per cent

5-60

5-65

105
12-25

6-41

1-86
38-5

<5-9

Wanklyn has given the following figures for the ash of certain
other leaves said to be used for the purpose of adulterating tea :—

Total Ash.

Soluble in Water.

Per cent

Beech

4-52

2.0

Bramble •

4-53

1-84

Easpberry

7-84 ,

1-72

1 Hawthorn

8-05

3-78

Willow

9-34

416

Plum

9-90

5-66

Elder

10-67

3-19

Gooseberry

13-50

7-83

The following table compiled from analyses by Zoller, Hodges,
Wigner, and the author shows the difference between the ash of genu-
ine tea and of exhausted tea leaves : —

Pure Teas.

Exhausted Leaves.

Per cent

Potash

28-42 to .39-22

6-49 to 7-45

Soda

0-65 „ 14-43

0-59

Magnesia

4-40 „ 6-47

9-5 „ 11-45

Lime

4-24 „ 9-3

8-9 „ 10-76

Fe^O,

2-49 „ 5-65

9-2 „ 9-8

MnA

0-80 „ 1-05

1-97 „ 2-2

^.0,

9-18 „ 18-03

20-8 „ 25-41

SO.

trace to 7-41

trace to 1-8

0-81 „ 3-51

traces

Silica

0-5 „ 6-9

7-57 „ 9-2

According to Allen the adulteration of tea with magnetic iron filings
used to be very common. It is now, however, quite obsolete.

To detect it, 10 grms. of the powdered tea are spread out, and the
magnetic particles easily picked up by a strong magnet, washed, dried,
and weighed.

Aqueous Extract. — By this is understood the total solid matter

FOOD AND DRUGS.

which can be obtained by complete exhaustion of the tea by boiling
water. Tea takes a very long time to completely exhaust, and a
relatively large amount of water is necessary. To show the necessity
of this, the following figures were obtained on the same sample of tea,
which was first dried at 100° and then powdered. The tea was
boiled under an upright condenser for the time specified : —

Amount of Tea.

Amount of Water.

Time of foiling.

Extract.
Per cent

5 grams.

200 c.c.

Ihour

25-5

M M

300 „

M M

27-2

»> >J

500 „

28-4

2 hours

30-0

" "

3 „

31-0

Five grms. should be boiled with 500 c.c. of water for at least two hours
under a reflux condenser, and on cooling the liquid should be made
up to 500 c.c. again, and 100 c.c. of the clear solution evaporated, and
the extract dried in a water oven and weighed. The aqueous extract
includes the caffeine, tannin, most of the colouring matter, and various
other constituents. The infusion of tea, as made for drinking purposes,
does not contain nearly all the soluble matters, and the properties of
the ordinary infusion are only of importance in gauging the character
of a tea as used in practice, a complete extraction being necessary in
dealing with questions of adulteration. The table of Geisler's results
(page 4) gives the average values of the total extract and of that of an
infusion prepared by pouring on the leaves 100 times their weight of
boiling water, and allowing them to infuse for ten minutes.

The extract obtained by complete exhaustion varies from about
32 per cent up to as much as 40 per cent in certain classes of tea. It
is obvious, therefore, that the presence of exhausted leaves will not
necessarily be indicated if the original leaves present contain a very
high amount of extract. Green tea yields a rather higher extract than
black tea, the lowest permissible limits being 30 per cent for black tea
and 38 per cent for green tea.

Tatlock and Thomson ("Analyst," xxxv. 103) prefer to boil 1 grm.
of tea with 400 c.c. of water for one hour under a reflux condenser,
collect the insoluble matter, wash with 80 c.c. of hot water and dry
and weigh. The weight of the insoluble matter, plus the moisture,
deducted from 100 gives the percentage of water-soluble ingredients.

They give the following limits for a number of samples : —

Indian teas
Ceylon teas
China teas

Per cent
43-47 to 49-75
41-32 „ 48-25
38-43 ,, 46-94

Ordinary exhausted tea leaves, that is the residues from restaurants,
etc., contain about 25 to 30 per cent of their original extractive matter.

TEA.

The following are results obtained by various observers with
specified characters : —

Kenrick.i

Wigner.

Peligot.

Parry.

Per cent

Congou teas

23-37

26 to 332

36-8

31 to 38

Assam „

38-53

33-3

41-7

32 „ 42

Hyson „

34-22

36-8

43-8

Ceylon

27-45

—

29 to 37

Gunpowder ,.

28-55

33 to 40

48-5

—

Japan „

30-07

—

Pekoe

—

34-2

38-0

China „

—

29 to 41-5

Caffeijie. — This may be determined by the methods described under
cotfee (page 38), but it is preferable to use 5 to 6 grms. of tea for the
determination. The percentage of caffeine varies between 2*5 per cent
and 4 per cent, rarely falling below 2-9 per ,cent. Many of the older
figures given by various observers are due to the fact that the processes
adopted (as is now recognized) gave too low results.

Tatlockand Thomson {supra, loc. cit.) prefer to determine caffeine by
boiling 2 grms. of the tea under a reflux condenser with 800 c.c. of
water for an hour. The filtered liquid is evaporated to 40 c.c, cooled,
mixed with 10 c.c. of normal alkali, and extracted with three successive
quantities of 40, 30 and 10 c.c. of chloroform. The mixed chloroformic
liquids are washed with 10 c.c. of normal alkali, then with water and
finally the chloroform evaporated and the caffeine weighed.

Burmann ("Bull. Soc. Chim." 1910, [iv.], 239-44) gives the follow-
ing method for the determination of caffeine, which he found to give
more accurate results than any method hitherto adopted, including that
of Keller, which was found inapplicable to roasted coffee. Five grms.
of the finely powdered sample are dried and freed from fat by extrac-
tion with petroleum spirit, and the residue shaken for some minutes in
a stoppered flask with 150 grms. of chloroform after which 5 grms. of
a 10 per cent solution of ammonia are added and the shaking continued
for 30 minutes. The liquid is then filtered quantitatively into a tared
Erlenmeyer flask, the solvent expelled and the residue of crude caffeine
heated in the water oven until constant in weight. It is next dissolved
in a little chloroform and introduced into a test tube (150 to 280 mm.
in length and 15 to 18 mm. in diameter) which has a constriction near
the bottom and another near the top. The solvent is evaporated from
the tube, which is then dried at lOO'' C. or " in vacuo ". The lower
constriction is now closely packed with a plug of asbestos wool, while
the portion above the upper constriction is filled with cotton wool. The
bottom of the tube containing the crude caffeine is then immersed in
melted paraffin, the temperature of which is maintained at 210° to 240°
C. and, after three hours heating, the whole of the caffeine will have

By ten minutes infusion.

2 On the dried tea.

10 FOOD AND DRUGS.

sublimed and condensed in the portion of the tube between the con-
strictions. This portion may then be cut off with a file and the caffeine
dissolved from it with a little chloroform, w^hich is subsequently evapor-
ated. The weight of the dried residue gives the amount of pure caffeine.
In the case of tea, it is not necessary to exhaust with petroleum spirit.

Tannin. — The determination of the tannin in tea is of the greatest
value, inasmuch as an excess indicates the presence of added astringent
matter, whilst a low tannin value indicates the presence of either ex-
hausted tea leaves or leaves of some other plant which contain but little
tannin. The tannin present in genuine tea varies from 9*0 per cent to
about 18 per cent, the average being almost 12 per cent. Any quantity
less or more than these limits must be regarded with suspicion.

Janke (using the copper acetate process described below^) found,
with eighteen samples, 6-9 per cent as the lowest, and 9"2 per cent as the
highest limit in black tea, whilst green tea gave from 8-5 per cent to 9*9
per cent (but see below). Exhausted tea leaves — that is tea leaves in-
fused for the beverage — contain from 1-5 to 4 per cent of tannin.

Very many processes have been described for the determination of
tannin, none of which can be said to yield absolutely accurate results,
and in reporting on the amount of tannin present, it is necessary, in
order to compare results by different analysts, to state the process
used.

The most reliable only of the many processes described will be
here discussed.

(1) The copper acetate process. About 2 grms. of finely powdered
tea are extracted by boiling for an hour with two successive quantities
of 100 c.c. of water. The filtered extracts are heated to boiling and
then 30 c.c. of a 5 per cent solution of cupric acetate are added. The
precipitate, consisting of a tannate of copper, is collected on a filter,
washed, and ignited, the copper ash being fully oxidized by the addition
of a few drops of nitric acid. One part of CuO may be regarded as equi-
valent to 1-305 part of tannin. This process is due to Eder ("Ding.
Poly. Journ." 129, 81). By this process Eder found an average of 10
per cent of tannin in black tea, and 12*3 per cent in green tea.

(2) Lead acetate process. Fletcher and Allen ("Chemical News,"
29, 169 and 189) proposed the use of acetate of lead for the determina-
tion of tannin. This process, which includes the small amount of
gallic acid present, is carried out as follows : An infusion is made by
completely extracting 2 grms. of tea, in the same manner as for the
determination of the total extract, the liquid being made up to 250 c.c.
Three quantities of 10 c.c. of a 0*5 per cent solution of neutral acetate
of lead are placed in beakers, each being diluted to about 100 c.c. with
boiling water. The process is a titration one, the indicator being a
solution of -1 grm. of potassium ferricyanide in 100 c.c. of water
and 100 c.c. of strong ammonia of specific gravity 0*880. This indi-
cator gives a deep red colour with tannic or gallic acids. The standard
tea infusion (2 grms. in 250 c.c.) is run in from a burette into the three
trial quantities of lead acetate solution. The first beaker should receive
12, the second 15 and the third 18 c.c. of the tea infusion (if green tea
is being examined 8, 10 and 12 c.c. will be sufficient). After well

TEA. 11

stirring, a few drops of the liquid are filtered and allowed to fall on to
a few drops of the indicator on a porcelain slab. In the presence of
tannin a pink colour will be observed. The approximate quantity of
the infusion will be easily determined by this preliminary experiment,
and a fresh titration is now carried out, and the approximate amount
of tea infusion at once run in. If tannin is still indicated another
small addition of the infusion is made and a few drops again filtered
and tested. According to Allen 10 c.c. of a 0*5 per cent acetate of lead
solution will precipitate 0*010 grm. of pure tannic acid. So that if
the above -described quantities be adhered to, the number of c.c. of tea
infusion divided into 125 will give the percentage of tannin in the
sample. Fletcher and Allen thus found 8-5 to 11*6 per cent of tannin
in black tea. Catechu tested by this method gives a result of 105 to
125 - per cent of tannin. It is therefore obvious that the method is
only comparative, and not absolute.

(3) Lowenthal's process. This process depends on the oxidation
of the tannin by means of potassium permanganate. In principle the
process is as follows : Tannin is much more easily oxidized by
potassium permanganate than indigo. It is, however, impossible to
determine the end of the reaction when a coloured solution of tannin is
titrated with the permanganate. With indigo, however, the end re-
action is comparatively sharp. A known quantity of indigo solution is,
therefore, added to the solution of tannin from a given quantity of tea,
and the amount of permanganate required for oxidation of the tannin and"
the indigo is noted. As the tannin is oxidized first, the end reaction is
that of the indigo and is fairly sharp. The same quantity of the indigo
solution is now oxidized alone, and the quantity of permanganate
used is deducted from the former figure, by which the amount necessary
to oxidize the tannin is ascertained. But besides the tannic acid (in
whatever form it is present), the gallic acid and certain other substances
are also oxidized. A known volume of the extract of the tea is there-
fore heated with gelatine solution to precipitate the tannic acid, and
the filtrate, together with a known quantity of indigo solution, is
again titrated with permanganate solution. The quantity used for the
oxidation of the gallic acid, etc., is deducted from the amount required
for the " total astringent matter " and the amount used for the tannin is
thus ascertained. The tannin is now determined in terms of potassium
permanganate. But as it is uncertain in what form all the tannic acid
of tea is present, the conversion of the permanganate figure into tannic
acid must be an empirical matter.

In practice the determination may be carried out as follows
(Procter's modification). The solutions necessary are : —

(i) A solution of potassium permanganate of about 1 grm. per litre.

(ii) A solution of indigo, containing 5 grms. of pure indigo-carmine
(sodium sulphindigotate) and 50 c.c. of strong H._,SO^ per litre.

(iii) A solution of gelatine (2 per cent strength).

The extract of the tea used for determining the amount of water-
soluble extract may be used for the determination (1 in 100).

Lowenthal gives the following concrete example, as illustrating the
calculation necessary : —

12 FOOD AND DRUGS.

A 1 per cent solution of the extractive matter of sumach was used.
Ten c.c. were diluted with 75 c.c. of water and 25 c.c. of the indigo solu-
tion and 10 c.c. of dilute sulphuric acid were added. The potassium per-
manganate is slowly run in until the blue colour changes to yellow, when
reaction is considered to be complete. The same amount of indigo and
acid are now titrated and the result noted. For the most concord-
ant results, the indigo should require about twice as much permanganate
as the tannin. 100 c.c. of the watery extract are then mixed with
50 c.c. of the gelatine solution, and the mixture is well stirred ; 100 c.c.
of the salt solution are then added and the whole allowed to stand for
twelve hours. A portion of the liquid is now filtered and the same pro-
cess is repeated in an aliquot portion of the filtrate, which is, of course,
now deprived of tannic acid. For example : —

10 -cc. of the extract 1 -11/./. a x 1 x-

25 c.c. of indigo solution | '"^"^"^"'^ ^^''^' ^•'- °^ Permanganate solution.

25 c.c. of indigo solution „ 6-6 c.c.

.•. required for the "tannin" only 10*0 c.c.

25 of the filtrate from the gelatine^

( = 10 c.c. original extract) r „ 11-2 c.c.

and 25 c.c. indigo solution J

but 25 c c. indigo solution alone „ 6*6 c.c.

Therefore the gallic acid, etc., required 4-6 c.c But as the total
astringent matters, or " tannin," required 10*0 c.c, it follows that 5'4 c.c.
of the permanganate solution were used by the tannic acid and 4 '6 c.c.
for the gallic acid, etc.

The best quantities of the extract of tea leaves (1 per cent) to use,
are 4 c.c. for the titration in the first instance, and 8 c.c. of the extract
after treatment with gelatine.

The titration should be carried out as follows : —

The extract (corresponding to 0*04 grm. of tea) is diluted to about
500 c.c. with water and 20 c.c. of the indigo solution added. The
solution of potassium permanganate is then run in slowly with vigorous
stirring until the liquid is transparent, when the permanganate is run
in very cautiously and slowly until the yellow solution appears of a
faint pink colour in the margin. A second titration in an equal quantity
is made and the two results, representing 0"08 grm. of tea, are added
together (a).

The same titration is then made on 40 c.c. of the indigo solution
without the tea extract. The amount of permanganate used {b) is de-
ducted from a. Now {a - b) represents the amount of permanganate
used to oxidize the tannin and other similar matter in the tea. If {a - b)
is more than two- thirds of b, correct results will not be yielded and the
amount of indigo must be adjusted accordingly.

The extract of the tea corresponding to 0-080 gi'm. is then mixed with
about 25 c.c. of the gelatine solution and the mixture is saturated with
ordinary salt. 10 c.c. of 10 per cent sulphuric acid are added and the
whole diluted with water. The liquid is now well shaken with a little
kaolin and filtered ; the precipitate can either be well washed with
water, or an aliquot portion of the filtrate, after the liquid has been
made up to a definite volume, can be used and the necessary calculation

TEA. 13

made. The filtrate is now mixed with 40 c.c. of the indigo solution
and titrated as before. The permanganate solution used is that required
to oxidize the indigo and the oxidizable matter of the tea other than
tannin (c). Hence a - c represents the permanganate used to oxidize
the tannin.

It must be remembered that no estimation of tannin is absolute,
but is comparative as between determinations on the same material.
For example, the tannin of tea and that of, say, sumach, require
different quantities of permanganate for oxidation. Hence it is very
common to report the tannin value of tanning materials in terms of
permanganate, or rather of oxalic acid, which is quantitatively oxidized
at once by permanganate.

If 10 c.c. of a decinormal solution of oxalic acid be titrated with
the permanganate solution in the presence of a little dilute sulphuric
acid', the volume of permanganate required is that which will oxidize
63 milligrams of oxalic acid.

The above results, therefore, enable the tannin to be expressed in
terms of crystallized oxalic acid.

Any attempt to return the actual amount of tannin, obviously re-
quires the knowledge of the reducing power of the tannin as compared
with that of oxalic acid. The term " reduction equivalent " is used to
indicate the weight of tannin that will reduce the same amount of
potassium permanganate as 63 grms. of oxalic acid (a normal solution
contains 63 grms. per litre). The actual reduction equivalent of tea
tannin is a matter not yet settled, but it is generally believed to be
practically identical with that of oak bark tannin, which is 62*3. So
that excellent comparative results may be obtained by calculating as
though tea tannin possessed the same reducing power as oxalic acid.

(4) Vignon (" Comptes Eendus," cxxvii. 369) has suggested the fol-
lowing simple method for an approximate determination of tannin. An
extract is made which contains about 0*1 per cent of tannin — say 1
grm. of tea to 150 c.c. of water. The total solid matter is determined
in a portion of this and then 5 grms. of pure white silk free from any
dressing is placed in the liquid, which is kept at 50° C. and occasionally
shaken for two hours. The silk absorbs the tannin, and the difference
between the total solids before and after the treatment with silk is
returned as tannin.

(5) Gelatine process. An approximate estimation of the tannin
may be made by titrating a definite volume of the extract with a 2 per
cent solution of gelatine to which a trace of alum has been added.
The gelatine solution is standardized by a solution of tannin of known
strength. The tannin solution should be titrated with the gelatine
solution until no further precipitation occurs. The exact ending of the
titration may be observed by allowing the precipitate to settle after
each addition of gelatine solution, and then placing a few drops of the
clear liquid in a watch glass and testing with gelatine solution to as-
certain if precipitation be complete. The tannin value of the gelatine
solution being known, the tea extract is titrated in the same manner.

(6) Hide powder. The tannin is absorbed by hide powder, and
the difference in the solid matter of' the extract before and after

U FOOD AND DRUGS.

absorption gives, approximately, the amount of tannin. An extract
of tea is made so as to contain about 1 per cent of tannin (preferably
about 8 grras. of tea thoroughly exhausted and the extract concen-
trated to 100 c.c). Procter operates as follows : —

A piece of glass tubing about 4 inches long and 1 inch in dia-
meter is packed tightly, but without ramming down, with finely
powdered pure hide powder. A siphon tube is inserted into the cork
which closes the upper end of this tube, the opening of which is lightly
plugged with a piece of cotton wool and covered with muslin. The
lower end is covered with fine muslin held in position by an india-
rubber band. The hide powder tube is now immersed in 100 c.c. of
the extract, of which the total solid matter has been determined, and
after the hide powder is thoroughly well soaked, the siphon is started
by suction and 50 c.c. of the liquid collected ; the lower end of the
siphon should be connected with an india-rubber tube and burette
cUp, so as to regulate the flow. This should be slow, so as to ensure
the absorption of the tannin. The difference between the solid matter
in the extract that has passed through the hide powder is returned
as tannin. A blank experiment should be made w^ith hide powder and
distilled water (as there is always a little soluble matter in the hide
powder) and the necessary correction made.

Tatlock and Thomson {vide supra; loc. cit.) prefer to determine
tannin by precipitating the aqueous extract by a slightly acidified solu-
tion of quinine sulphate, and to wash and dry the quinine tannate,
which is stated to contain 75 per cent of its weight of tannin. The
least variation in the conditions of the experiment will, however, cause
the composition of the precipitate to vary.

Allen recommends the following for the detection of catechu in tea,
but states that the suspected sample should be tested side by side with
a sample of genuine tea. One grm. of pure tea, and an equal quantity
of the suspected sample, are infused in 100 c.c. each of boiling water,
and the strained liquids precipitated while boiling with a slight excess
of neutral lead acetate ; 20 c.c. of the filtrate from pure tea, which
should be colourless, when treated with a few drops of silver nitrate
solution gives only a greyish colour, and if cautiously heated gives only
a very slight greyish cloud or precipitate of reduced silver. But in the
presence of 2 per cent of catechu, there will be a copious precipitate
of a brownish colour, the liquid becoming decidedly yellow.

It has been said that foreign leaves are legitimately present in tea
to the extent of 1 or 2 per cent, being added in order to impart a
special bouquet to the tea. It is more probable, however, that any
foreign leaves added to tea for this purpose are removed after imparting
such bouquet to the tea. According to Wynter Blyth, a crystalline
sublimate, which he considers to be caffeine, is obtainable from every
single leaf of tea. A leaf is boiled for a minute in a watch glass with
a very little water, a little magnesium oxide added and the liquid
evaporated down to a single drop. This is transferred to a microscopic
cover glass and evaporated almost to dryness on a hot plate. It is
surrounded by a glass ring and on evaporation of the last drop of
water, a second cover glass is placed on the ring. In the presence

TEA. 15

of caffeine, further heating causes a crystalline sublimate to collect
on the upper glass which is clearly visible under the microscope.
Certain other leaves yield a sublimate, but according to Blyth, if no
sublimate is found, the leaf is not that of tea. The same chemist
has also claimed that the presence of manganese in tea leaves is a
certain method of recognizing them. The ash of a leaf is taken up on
a bead of sodium carbonate on a platinum looped wire, and on fusion
with a trace of KNO3 ^^^ green colour of potassium manganate is
found. Allen has found manganese in the leaves of Camellia Japonica,
Camellia sasanqua, Cotiea Arabica, the beech, blackberry and sycamore.
Other leaves examined showed either no manganese, or only faint
traces.

Structural and Microscopic Examination. — Some of the leaves
should be soaked in hot water. The facing, if any be present, as used
to regularly be the case with green teas, will become detached, and a
little may be examined under the microscope if necessary, and will
be found to be structureless. The bulk of the facing will sink to the
bottom of the water, or may be collected and examined. Indigo may
be recognized by giving a yellow colour wdth a drop of nitric acid ;
Prussian blue is detected by heating with caustic alkali, acidulating
with hydrochloric acid, and testing for ferrocyanide by ferric chloride.
Any matter left insoluble after warming with caustic alkali, should be
then treated with hydrochloric acid ; any insoluble matter will usually
be of a siliceous nature, such as steatite.

The venation and serration of the leaf can to some extent be ob-
served on the leaf which has been soaked in hot water and dried
between blotting paper, by means of a lens. Blyth proposes cleaning
the leaf by warming with a strongly alkaline solution of potassium
permanganate; sodium hypobromite is also useful. The leaf should
be kept between two cover glasses, the upper one being kept in position
by a small weight. The leaf, previously soaked in the alkaline perman-
ganate, is then washed in water, treated with HCl, again washed and
then examined. The details of its structure can then be examined.
A better method is to infuse the leaves in boiling water twice. Eemove
superfluous moisture carefully by means of blotting paper. The leaves
should then be immersed for ten days in a 70 per cent solution of
chloral hydrate. They can then be conveniently examined.

A leaf will be found to be bi- facial ; the epidermis, both upper and
lower, can be examined in a transverse section and will be found to con-
sist of small cells, and if the sections are suitable, characteristic long
hairs will be found. One or two rows of palisade parenchyma will be
found, and the spongy parenchyma with large air spaces. In the
centre of the leaf numerous cells are found containing crystals of calcium
oxalate in various form. The principal diagnostic characters are as
follows : —

(1) The long hairs with a radiate arrangement of cells at their bases.

(2) The rosette crystals of calcium oxalate.

(3) Peculiar sclerenchymatous cells, known as idioblasts, found in
the mesophyllic tissues. These are most common in the midrib and
petiole.

16 FOOD AND DKUGS.

The walls of these idioblasts are highly lignified and are well stained
by phloroglucinol and hydrochloric acid.

An examination of the surface preparation is also very useful. The
venation consists of well-defined loops, not found in leaves likely to be
met with as adulterants. Each of the serrations is a more or less hook-
shaped tooth consisting of a conical mass of parenchymatous cells.
They often fall off on old leaves, leaving a characteristic scar. The
under surface has numerous oval stomata with characteristic spaces
between the guard cells.

The hairs are very long — although often absent on old leaves. They
are unicellular, and very thick-walled, and are generally beni nearly
at right angles near the base, so as to lie almost flat on the leafy sur-
face.

In examining tea leaves in this manner, the surest guide is to con-
sider the above-described characters side by side with leaves of known
authenticity. The comparative rarity of cases in which one finds
other leaves present does not justify the space that would be necessary
for a description of the botanical or microscopical characters of leaves
said to be used as adulterants of tea. No such leaf shows the characters
described above and a comparison of suspected samples with genuine
tea leaves will at once demonstrate whether the suspected sample is
tea or not. Identification of foreign leaves can only be definitely made
by comparison with similar leaves of known origin, and is very rarely N^ .
necessary. ^^

COCOA AND CHOCOLATE.

Cocoa of commerce consists of the slightly roasted seeds of several
species of T/ieo6roma— principally T. cacao. The seeds consist of the
shells or husks, and the cotyledons or kernels, the latter being know^n as
cocoa nibs, and these when ground, after being freed from the husks,
constitute what is understood as cocoa. Chocolate is understood to
mean cocoa both sweetened and flavoured (principally with vanilla),
but no legal standard appears to exist for what chocolate should be,
hence its composition is very variable. This will be dealt with later.

The average compositions of (1) natural cocoa, (2) cocoa nibs, and
(3) cocoa husks, are given in the following tables which are the results
of numerous analysis by Konig, Heisch and others, together with
samples examined by the author : —

(1) Cocoa with the Husks (roasted).

Moisture 5 to 8 per cent

Fat 40 „ 50

Carbohydrates 10 „ 14 „

Ash (total) 3 „ 4-6 „

Ash (water-soluble) . . . . 1*5 „ 2*4 „

Nitrogen 1-7 „ 2 -4 „

Theobromine 1 ,, '^

COCOA AND CHOCOLATE.

(2) Roasted Cocoa Nibs.

Moisture

Fat .

Carbohydrates .

Ash (total)

Ash (water-soluble)

Nitrogen

Theobromine

Cellulose

3-5 to 4-5 per cent

45
10
2-3

0-9
1-6
1-5
3

14
4-0
1-8
2-15
2-5
4-5

(3 Cocoa Husks (roasted).

Moisture 3

Fat 4

Carbohydrates 7

Ash (total) 6

Nitrogen 1*5

Cellulose 13

8
5

10
20
2-4
18

per cent

The following tables are given by Booth, Cribb and Richards,
("Analyst," xxxiv. 137).

Granada Bean

Trinidad Bean

(with husk).

(without husk).

Raw.

Roasted.

Raw.

Roasted.

Per

cent

H20

632

3-10

6-67

4-45

Fat

46-50

49-96

54-60

55-70

Nitrogen

1-96

1-86

2-28

2-32

Fibre

3-60

3-90

2-45

2-48

Ash

2-86

312

2-87

2-73

Siliceous matter

0-10

0-12

0-03

0-08

Soluble ash

1-26

1-44

0-94

0-95

Alkalinity as K3O

0-68

0-75

0-42

0-43

Cold water extract

13-50

12-90

12-73

12-00

Analysis of

Nibs op Known Origin.

Ash.

Soluble
Ash.

Siliceous
Matter

Alkahnitv of
Ash as KgO.

Cold Water
Extract.

Nitrogen.

Fat.

Fibre.

Per

cent

African

2-52

0-98

0-05

0-38

11-58

1-84

50-2

Granada

2-60

1-04

0-03

0-55

9-8

2-26

50-8

2-97

Guayaquil

3-16

1-32

0-04

0-53

11-4

—

Trinidad

2-73

0-95

0-08

0-44

12-0

2-32

55-7

2-48

Caracas

3-24

1-58

0-08

0-74

—

Bahia

2-68

1-22

0-05

0-51

9-5

1-98

44-4

—

Accra

3-22

1-36

0-04

0-41

11-4

2-46

50-6

2-87

Ceylon

3-81

1-66

0-03

0-67

11-9

2-44

50-2

2-36

Para

3-22

1-14

0-06

0-45

12-1

—

Puerto Cabello

3-88

1-50

0-13

0-64

12-6

2-35

51-3

3-02

VOL. I.

FOOD AND DRUGS.

The ash of cocoa husks contains a variable amount of matter in-
soluble in acid, which is not the case with the ash of the cocoa nibs.

Analysis of Husks from Known Sources.

Ash.

Soluble
Ash.

Siliceous
Matter.

Alkalinity of
Ash as KgO.

Cold Water
Extract.

Nitrogen.

Fat.

Fibre.

Per

cent

Ceylon

6-61

4-78

1-0

2-54

20-7

2-4

3-1

12-8

African

5-63

3-53

1-79

2-63

20-4

2-91

3-5

15-7

Para

6-78

4-39

0-72

2-80

18-7

Guayaquil

8-19

5-25

1-45

3-36

24-6

2-13

5-9

12-85

Puerto Cabello

20-82

5-24

8-33

1-13

23-5

5-68

13-83

8-48

5-78

0-82

2-51

19-5

2-76

3-31

15-8

11-68

4-08

3-34

2-24

20-3

16-28

—

18-9

2-29

4-62

14-80

D (raw)

7-82

4-62

0-86

2-12

24-4

—

The following table is due to Ridenour (** Amer. Jour. Pharm." 1895,
207) :—

-©8BjaAV

oqifBOBaBiv^

O iH rH -^ us -^

•an«iio«w

•SVOVJLVQ

pa^sBoa

•oosBqox

OS fh e«5

•pBpiUUJ,

pa^SBoa

•BpBnsjf)

OS»C(MQOOOOC*"* rttSOO
Op OS O "^ (N t- CO «p op CO t>

^ ' 6^ 1^ ' ui ih a> iciNOT

•SBOBaBQ

•Bqixy^

•pBpiUUJ,

-BABf

•ui'BaiJns

-wq«a

(M t- 00 CQ t>

?0 iH ->* "*

«0»OOQOC100lOr-( <-t ^ <Z>

oopqicpcpoioo cccco

OCO W5 « iH t- O W O <N iH
■^ iH (N (?1 »p ,H rH Op ^'T' «

WSrHOSiH* ussbcb a>«5«3

cS ^ S "2 =2

^^ 2 s 5-^ o

o o'^

COCOA AND CHOCOLATE.

The following figures represent average values for the nibs and
husks : —

Nibs.

Husks.

Dry and Fat-
free Nibs.

Dry and Fat-
free Husks.

Per

•

cent

HaO

3-0

4-5

Ash

3-07

7-3

6-14

7-84

Siliceous matter

0-05

1-11

0-1

1-2

Fat

50 0

4-4

Fibre

2-8

14-0

5-9

16-5

Nitrogen

2-5

5-05

2-64

Cold water extract

11-6

22-0

24-2

23-6

The following figures are those of certain flours sometimes found
adulterants : —

Cold Water
Extract.

Nitrogen.

Ash.

Per

cent

Wheat

7-7

1-97

0-5

Barley

5-1

1-2

0-9

Maize

0-8

0-14

0-4

Rice

0-9

1-23

0-5

Sago

1-98

0-03

0-14

Arrowroot

0-4

0-13

0-3

Banana

1-5

0-8

The best cocoas should be made from the nibs only, but the husks
are frequently ground with the nibs in the preparation of inferior
qualities. So long as the husk is not in excess of that natural to the
bean, it may be properly sold as cocoa. In commerce there are two
varieties of (prepared or soluble) "cocoa" regularly to be met with.
(1) This consists of the ground nibs, with frequently some of the fat
removed ; (2) preparations to which sugar and frequently starch have
been added.

The fat is frequently partly removed, as excess of this constituent
renders the manipulation of the cocoa difficult, and its removal renders the
product more easy to digest. A small amount of alkali is often added
by certain makers, in order to soften the fibre, and to emulsify the fat,
so as to render the product more easily miscible.

Adulterants — or considering the legal aspect of the matter — dilu-
tents may be a better term, of cocoa are usually starch and sugar, al-
though other substances are sometimes met with. Brick dust, oxide of
iron, iron-earth, chalk and similar substances are given as adulterants
in many books — but they are rarely met with — although a little iron
preparation may occasionally be added for colouring purposes. Aniline

20 FOOD AND DEUGS.

dyes are also sometimes added. Any organic powder will usually be
detected microscopically.

The Analysis of Cocoa.

Ash. — After determining the moisture (if considered necessary) the
ash should be determined. This should be very little in excess of 4
per cent, usually less, unless a notable amount of husk is present, or
some inorganic adulterant has been added. The ash should be wholly
soluble in hydrochloric acid, but if much cocoa husk be present, a small
proportion will be insoluble. The ash of pure cocoa contains from 30 to
50 per cent of phosphoric acid. The alkalinity of the ash varies enor-
mously on account of the addition of a trace of alkali in the preparation
of the cocoa, so that it is practically only a measure of the amount of
alkali so added, and gives no further information. The presence of
sugar and starch naturally diminishes the amount of ash.

Fat. — This should be determined on 10 grms. of the sample, after
drying to remove moisture. It is best mixed with sand and ex-
tracted in a Soxhlet tube with anhydrous ether. In cases of cocoa to
which alkali has been added a little of the fat remains undissolved,
being fixed in the form of soap. A trace of acid is necessary to decom-
pose this. The fat should have the character of pure cocoa butter, as
described on p. 26. This is important in the case of chocolate and
chocolate creams, as many other fats are used to add to cocoa after the
more valuable cocoa butter has been extracted for sale as such. The
fat in ordinary powdered cocoa varies considerably — if the cocoa is in
the natural state usually between 45 per cent and 55 per cent, but
usually 20 to 28 per cent is found in cocoa which has been more or less
defatted before put on the market.

Sugar. — The dried residue, after the fat has been exhausted, is ex-
hausted with hot alcohol of specific gravity about 0*850. The hot
alcoholic solution is heated with a strong solution of lead acetate, which
precipitates tannates, tartrates, etc. The alcohol in the filtrate is
driven off, and the excess of lead removed by the addition of a strong
solution of sodium phosphate. The liquid is now ready for the sugar
determination. This is effected either by a polarimetric reading or by
inversion and reduction of Fehling's solution, as described under
" Sugar " (p. 122). The alcoholic extract of the cocoa has practically
no reducing power on the copper solution, so that the whole of the
sugar found may be approximately credited to added cane sugar.

Albumenoid Nitrogen. — The residue, after the extraction with ether
and alcohol contains the starch and albumenoids, together with cellu-
lose, fibre and gum, etc. After weighing this residue, an aliquot por-
tion may > be used for determination of the albumenoid nitrogen by
Kjeldahl's method, and this may be multiplied by 6*25 and returned as
albumenoids.

This residue is in a suitable condition for microscopic examination,
and if any foreign starch is detected it may then be estimated.

Starch. — A weighed portion of the residue which has been extracted
with ether and alcohol is heated for an hour with 50 c.c. of a 2 per cent
solution of hydrochloric acid at a pressure of two atmospheres (this is

COCOA AND CHOCOLATE. 21

conveniently done by effecting the conversion in a soda water bottle
with an india-rubber cork tightly wired in ; the bottle is heated to 120''
C. in an oil bath which will correspond with the necessary pressure).
The starch is now completely converted into dextrose and this is deter-
mined by the reduction of Fehling's solution, ten parts of dextrose
representing nine parts of starch.

An alternative method of determining the starch present consists in
extracting the fat-free sample with cold water, and washing the residue
with a -04 per cent solution of caustic soda to remove the albumenoids.
The residue is rinsed off the filter with warm water, the starch gelatinized
by heating, and the liquid heated with a measured quantity of a freshly
prepared cold infusion of malt, whose specific gravity has been as-
certained. The liquid is kept at 60° to 63° C, with occasional stirring,
until the conversion of the starch is finished, as shown by a drop of the
liquid giving no blue or brown colour with iodine solution on a white
tile. The solution is filtered, made up to a definite volume and the
specific gravity taken. From the excess of the specific gravity over
1*000, is subtracted the density due to the solids in the malt infusion
— allowing of course for the increase in volume — and the remainder
represents the increase in specific gravity due to the starch dis-
solved. This figure divided by 4-096 gives the number of gi'ms. of
starch in 100 c.c. of the solution being examined. For example (as
calculated by Allen) if 20 grms. of cocoa be taken, and the solution of
gelatinized starch be made up to 50 c.c, and 5 c.c. of infusion of malt of
specific gravity 1-060 be added : the liquid is made up eventually to
100 c.c. and is found to have a specific gravity 1-023. The correction for

the malt infusion will be (1060-0000) ^ 5 ^ 3 ^ ^^e density of

100 ^

the solution 1023 - 1000 = 23, and this figure - 3 (the malt density
value) = 20. This, divided by 4*096 gives 4-9 grms. per 100 c.c. or
24*5 per cent of starch on the sample.

Cellulose and Fibre. — The mixed cellulose, fibre, and siliceous matter
left after the above treatment is washed with 2 per cent caustic alkali
solution, then with dilute HCl, alcohol, and finally ether, and dried and
weighed. A direct determination of the crude fibre, which is of value
as indictating the presence of cocoa husk, can be made by removing the
fat from 2 grms. of the sample, and then boiling the residue for half an
hour under a reflux condenser with 200 c.c. of water and 2-5 c.c. of
strong sulphuric acid. The liquid is filtered through fine linen and
the undissolved matter washed with hot water several times and then
boiled with 200 c.c. of water and 2-5 grms. of caustic soda. The
residue is washed again with hot water, and then with alcohol and
finally with ether, dried at 110° and weighed. This — after deducting
the ash left on ignition — is to be returned as crude fibre. In cocoa free
from husk it will generally vary between 3*5 and 5 per cent, but will be
higher than this if husk is present. (Allen, " Commercial Organic
Analysis," 2nd edition. Vol. Ill, part ii, p. 567). This process gives
good results, and with the omission of the alcohol and ether washing,
is officially used in the United States.

Booth, Cribb, and Kichards ("Analyst," xxxiv. 141) remark that in

22 FOOD AND DRUGS.

the course of this process as applied to the analysis of chocolate it is
worth while to obsei-ve the tint of the solution obtained by the acid
treatment, as, if it be only a faint red, there is probably only a very
little cocoa present.

Estimation of Husk in Cocoa Poivders. — A. Goske (" Zeit. Unter-
such. Nahr. Genussm." 1910, 19, 154-8). To carry out the following
process the husk is first separated from the cocoa powder by means of
calcium chloride solution, advantage being taken of the greater sp. gr. of
the husk. Extract 5 grms. of the cocoa powder with ether for sixteen
hours, and weigh the amount of fat extracted. Dry the fat-free powder,
and mix one gramme of it in a tube with 20 c.c. of calcium chloride
solution, prepared by dissolving 107*1 grms. of calcium chloride in
100 c.c. of water, this solution having a sp. gr. of 1*535 at 30° C.
Warm the calcium chloride solution to a temperature of about 50° C.
before adding the cocoa, thoroughly mix together, and heat to boiling
for two minutes. Submit the tube and its contents to centrifugal
action for six minutes while still hot. Use a glass rod to break down
the froth, then decant the liquid portion from the almost solid sediment.
Collect the latter on a weighed filter, wash until free from chloride, dry
at 100° C, and weigh. Several samples of commercial husk when
submitted to this process yielded from 15 to 38*7 per cent of sediment,
the average being 24*5 per cent, calculated on the dry, fat-free substance.
The author suggests the use of the highest figure, 38*7 as a standard
in ascertaining the amount of husk present in a sample of cocoa. One
gramme of dry, fat-free cocoa, for example, yielded 0*0618 grm. of
sediment, corresponding with 0*16 grm. of husk, or 13 per cent on the
original cocoa, which contained 18*4 per cent of fat. When 6 per
cent was subtracted as the amount yielded by ordinary cocoa, the
sample contained 7 per cent of added husk.

Total Nitrogen. — Two or three grms. may be used for a determina-
tion by Kjeldahl's method. It must be remembered that theobromine
contains 31*1 per cent of nitrogen, so that if the albumenoid nitrogen is
required, the theobromine must be removed by exhaustion with ether,
alcohol, and chloroform, and the nitrogen determined on the residue.

Theobromine. — The principal alkaloid of cocoa is theobromine,
G^HgN^Og, but a little caffeine is also present. The amount of alkaloid
present in cocoa averages 1*4 to 1*8 per cent, the husks containing a
very small quantity. It is not usually necessary to determine the
amount of alkaloid present, as, if a cocoa is pure, the alkaloidal value
will only be an indication of its quality — and that probably only as a
stimulant. Numerous methods have been proposed to determine this
value, and in the author's experience that of Beckurts is, on the whole,
the most satisfactory. A mixture of 10 grms. of the powdered cocoa
and 10 grms. of fine sand is heated on the water bath with 150 c.c.
of water and 0*1 c.c. of strong hydrochloric acid, with repeated agita-
tion. After this has been done for about an hour, the fat is allowed to
solidify and the aqueous solution of alkaloids is filtered off. Excess of
magnesia is added, and the liquid evaporated to dryness, and the dry
residue is extracted with chloroform : the chloroform is evaporated and
the theobromine (and caffeine) weighed.

COCOA AND CHOCOLATE. 23

A full examination of numerous methods suggested for this deter-
mination has been made by Kunze (" Zeit. J. Analyt. Chem." 1894, 1).
Kunze prefers the following process, which, however, is more tedious
than that first described and but little more accurate. Ten gi-ms. of
cocoa are boiled with 150 c.c. of 5 per cent sulphuric acid for twenty
minutes and the liquid filtered and the residue washed with boiling water.

Phosphomolybdic acid is then added to the liquid, and after stand-
ing for twenty-four hours, the precipitate is filtered and washed with 5
per cent sulphuric acid. While still wet, the precipitate is transferred to
a flask and decomposed with baryta water. COg is then passed through
the liquid and the whole is evaporated to dryness. The dry mass is
exhausted with hot chlorofoim, the chloroform evaporated and the dry
residue, consisting of theobromine and caffeine, is weighed. Either of
these processes gives quite satisfactory results.

Cold Water Extract. — Two grms. are placed in 100 c.c. flask and
about 60 c.c. of cold water added. The whole is shaken well at inter-
vals for several hours and then made up to 100 c.c, well shaken and
allowed to stand all night. After again well shaking, 25 c.c. are
filtered off, evaporated, and the residue weighed. Good commercial
cocoa containing its full amount of fat contains on an average 12 to 13
per cent of cold water soluble extractive. In defatted cocoas this will
be proportionally higher.

Determination of Pentosans. — Bodies known as pentosans are anhy-
drides of five carbon glucoses, which yield bodies of the type of xylose
and arabinose on hydrolysis. According to Tollens these bodies ap-
pear to occur in greater quantity as lignification of plant substance
progresses. Cross and Bevan consider that bodies of the oxycellulose
type behave in this determination in the same way as pentosans. At
all events, on distillation with hydrochloric acid such bodies yield fur-
fural, from which the " pentosans " can be calculated. The method
of carrying out this determination is as follows : —

Three to four grms. of the substance are mixed with 100 c.c. of hydro-
chloric acid of specific gravity 1-06 (= 12 per cent HCl) in a Wurtz
flask and the contents of the flask distilled from a sand bath. When
30 c.c. has collected a further 30 c.c. of acid is added to the flask
through a tap funnel and this is repeated till 400 c.c. is distilled over.
During the distillation, a drop of the distillate is tested for furfural
by touching it on paper impregnated with a dilute solution of aniline
acetate with some sodium acetate. If no pink coloration appears,
no more furfural is coming over and the distillation may be stopped.
Usually between 300 to 400 c.c. will be distilled. A solution of phloro-
glucinol in dilute hydrochloric acid is then added to the distillate and
the mixture is allowed to stand over night. The black precipitate
formed is filtered through a weighed paper, washed with 150 c.c. of
cold water, dried in a water oven and weighed. The weight of the
phloroglucide, divided by 1*82, gives the amount of furfural. This
latter may be calculated into pentosans by subtracting 0'0104 grm. and
multiplying by 1-88. The following results are given by Hehner and
Skertchly (" Analyst," xxiv. 181), and refer to partially defatted
cocoas.

FOOD AND DKUGS.
Cocoas.

Cold Water
Extract.

Nitrogen.

Fat.

Ash.

Alkalinity as
K2CO3.

Pentosans.

Per

cent

18-60

3-23

28-82

8-18

4-15

2-18

18-08

315

29-74

8-12

4-08

2-31

18-56

3-06

28-57

8-38

3-79

2-35

18-08

3-29

28-24

9-03

3-79

1-69

18-08

3-20

28-21

8-84

3-75

1-84

18-48

3-24

27-51

9-30

4-08

1-89

19-00

3-32

28-19

8-61

4-08

2-08

17-44

3-07

26-82

7-18

2-77

2-81

None of these samples show indication of the addition of husk.
Chocolates.

Cold Water
Extract.

Per
cent
67-20
66-88
65-04
42-96
37-52

Nitrogen.

Fat.

Ash.

Pentosans.

Fibre.

Per

cent

0.76

23-76

217

1-27

—

0-80

23-12

1-98

0-83

1-18

0-70

23-59

2-16

0-81

1-33

0-60

4-20

1-88

2-95

0-57

4-21

—

1-81

2-71

The last two samples indicate the presence of about 25 per cent of
*' cocoa " which was almost entirely husks, as shown by the high fibre
and pentosans. The determination of pentosans, however, gives little
information that the estimation of fibre does not, and is far more
complicated.

Added Alkali. — If a cocoa has an alkaline reaction, and yields an ash
from 10 grms. which requires more than 1'5 c.c. of normal acid for neu-
tralization, and the amount of ash insoluble in water is less than 50
per cent of the total ash, there is no doubt that a soluble alkali has been
added to the cocoa. If the alkalinity is as high as above stated, and the
insoluble ash is more than 60 per cent of the total ash, then magnesium
carbonate has probably been added to the cocoa.

Microscojnc Characters. — To examine powdered cocoa, two prepara-
tions should be made : (1) By thoroughly mixing a few grains with
water ; (2) By mixing with a little ether for a few hours, washing with
alcohol and mounting in water. The tissues are now clearer than
when mounted with water before extraction. The principal diagnostic
characters of genuine powdered cocoa are as follow : —

(1) Thin-walled parenchyma of the cotyledons.

(2) Minute starch grains.

COCOA AND CHOCOLATE.

(3) Polygonal epidermis of the cotyledons with red-brown con-
tents.

(4) Abundance of oil globules.

(5) Cells containing cocoa-red.

(6) Note the absence of large starch grains.

The principal portion of the cotyledons consists of polygonal thin-
walled parenchyma, many of the cells containing minute starch grains
and fat and albuminous matter. The starch grains are round and not

Fig. 1. — Powdered Cocoa, a, starch grains; ae, outer layer of endosperm;
ai, inner layer of endosperm ; al, aleurone grains ; co, cotyledon ; cp, pigment
cells containing cocoa-red ; cr, crystals of fat : ec, epidermis of cot>iedon,
surface view ; e'c', epidermis of cotyledon, profile ; end, inner epidermis of
pericarp ; gr, crystals of fat ; I, bast from fibro-vascular bundles; ox, calcium
oxalate crystals ; p, pluricellular hairs ; pa, pi, parenchyma of seed coat ;
ra, cells of radicle ; sc, sclerenchymatous layer of seed coat, surface view ;
s'c', sclerenchymatous layer of seed coat, in profile ; te, outer epidermis of
seed coat to which the inner epidermis of the pericarp {fnd) is adhering;
ti, inner epidermis of seed coat; tr, v, vessels, etc., from fibrovasciilar bundle.

B7J 2J0rmission of the Editor of tJie "Pharmaceutical Journal ".

angular. Attached to the epidermal cells, a few club-shaped hairs
may be found. Characteristics of the husk are a large-celled epi-
dermis and small thick- walled cells ; and also spiral vessels of a well-

26 FOOD AND DEUGS.

defined type, and masses of mucilage, which are tinged rose pink with
ruthenium red dissolved in a 10 per cent lead acetate solution.

The diagnostic characters of powdered cocoa shells are as follows
(Greenish) : —

(1) The two epidermal layers : long narrow cells crossing larger
polygonal ones, often diagonally.

(2) The small polygonal thick- walled cells of the sclerenchymatous
layer.

(3) The large rounded parenchymatous cells with arm-like projec-
tions.

(4) The mucilage in masses, often tinged with brown.

(5) Spiral vessels or their fragments.

COCOA BUTTER

It will now be convenient to briefly consider the principal char-
acteristics of cocoa butter — the fat of the cocoa seed. This fat is used
in medicine — principally for the preparation of suppositories — and is a
drug official in the British Pharmacopoeia — although it is principally
used as a food, in the preparation of chocolate creams, etc. The official
requirements of the Pharmacopoeia are as follows : It softens at 26-6°
C, and melts between 31-1° and 33-9°. If 1 grm. be dissolved in 3 c.c.
of ether in a test tube at 17° C, and the tube be placed in water at 0°,
the liquid should not become turbid, nor deposit in less than three
minutes, and if the mixture after congealing be exposed to a tempera-
ture of 15*5° it should gradually afford a clear solution. This test is
highly unsatisfactory. The following may be taken as the average
character of genuine cocoa butter : —

Specific gravity ^ at 15° C. =

0-955 to

0-995

, 100°
» at jg, _

0-855 „

0-858

Melting-point

Melting-points of fatty acids
Reichert-Meissl value (5 grms.)

28^ „

48" „
0-2 „

33°

51°
0-6

(but variable if the fat has
been heated much).

Saponification value
Iodine value

192 „
32 „

196
38

Refractive index at 40° =

1-4565 „

1-4575

60° =

: 1-4496 „

1-4504

Butyro-refraotometer index at 40°=

46 „

„ at 35° =

= 48-5,,

49-5

Cocoa butter consists chiefly of the glycerides of stearic, lauric,
palmitic, arachidic and oleic acids (with an acid of the empirical formula
Cg^Hp^gOg termed by Kingzett theobromic acid). A little cholesterin
and some glycerides of the volatile fatty acids are also present. In de-
termining the melting-point of this fat the capillary tube should be al-
lowed to stand for forty-eight hours after the fat is first melted, before

^ The specific gravity of cocoa butter gradually rises from about 0-950 after

being freshly melted to a maximum of 0-995 after a few days. This figure should

100°
therefore be determined at — -, at which temperature this disturbing influence is

eliminated.

CHOCOLATE. 27

the determination is made. Or it may be stood in ice for six hours or
so.

Adulteration. — It is stated in textbooks to be frequently adulterated
with tallow, various fatty oils, paraffin wax and beeswax. This is quite
untrue. Cocoa butter is rarely adulterated. The author has examined
a very large number of samples and adulteration is rare — and when
practised, not with, for example, beeswax, which is usually worth at least
25 per cent more than cocoa butter. Substitution — usually quite openly
— of another fat is, however, common, and several so-called " cocoa
butter substitutes " are on the market. The basis of these is either the
stearin of cocoa-nut oil (sometimes from palm-nut oil) — which melts at
a very low temperature and is a poor substitute for cocoa butter — or one
of the harder and less known fats such as Shea butter.

The author has examined a number of these samples and found the
following average figures : —

Substitutes with Cocoa-nut Stearin as the Principal Ingredient.

100°
Specific gravity at y^ 0*8736

Reichert-Meissl value (5 grms.) .... 4-5 to 6-0

Iodine value 4 „ 8

Saponification value 250 „ 270

Melting-point 26° ,,28°

Refractive index at 60° 1-4400 „ 1-4420

Butyro-refractometer No. at 60° . . . . 33 „ 37

Substitutes op the Shea Butter Type.

100°
Specific gravity at ^5 0-855 to 0-865

Reichert-Meissl value (5 grms.) .... below 1

Iodine value 30 to 45

Saponification value 180 „ 195

Melting-point 28° „ 34°

Substitutes of the Palm-nut Stearin Type.

100°
Specific gravity at ^^^ 0-873 to 0-875

Melting-point 25°,, 30°

Iodine value ......... 10 „ 15

Saponification value 240 ,, 255

Reichert-Meissl value (5 grms.) 5 ,, 7

Refractive index at 60° 1-4430 „ 1-4450

Butyro-refractometer No. at 40° 36 „ 37

CHOCOLATE.

Chocolate is universally understood to be sweetened and flavoured
cocoa. There is no legal standard for chocolate, and if one should ever
arise, it would obviously be impossible to fix the relative proportions
of the sugar and cocoa except within very wide limits. Many analysts
are strongly in favour of fixing legal standards for this article, but in
doing so grave difiiculties would arise. The bulk of the chocolate
manufactured is sold as a sweetmeat, and not for ordinary nutritive

28 FOOD AND DRUGS.

pui-poses. It may be bought at very high prices, or by the poorer
classes at so Httle as three ounces for a penny in the form of chocolate
cream. The whole question becomes one of palatability, except in the
case of chocolate sold as a beverage, and an obvious hardship would
ensue if the sale of the lower grade, but quite wholesome, chocolate and
chocolate creams were to be interfered with, because they contain, for
example, the outer husk of the cocoa bean, or some cocoa-nut fat as a
filling. It appears to stand on quite a different footing to a purely
natural product where an obvious standard of quality exists. To re-
strict the sale of this product would logically necessitate carrying the
principle to many other articles, and would largely interfere wdth the
sale of inferior but still wholesome products. In the case of chocolate
for use as a beverage it may be that standards would be advisable.
Where nothing but cocoa husk is used, the term chocolate certainly
ought to be qualified in some way.

The quantity of sugar in the best chocolate averages about 50 per
cent, the remainder being pure cocoa either with or without the addition
of some extra quantity of cocoa butter, with a minute quantity of
flavouring. The principal flavour used to be derived from vanilla beans,
and much of it is still obtained from that source, but the bulk of the
vanilla flavour is now derived from synthetic vanillin obtained from
oil of cloves. A little cinnamon, benzoin, Tolu and Peru balsams, and
nutmegs are sometimes used, but vanilla is the favourite flavouring
employed all over the world.

The fact that chocolate is only sweetened cocoa (for in the analysis
the determination of the amount of flavouring is an impossibility in
nearly all cases) renders a long description of its analysis un-
necessary.

The points to which attention should be paid are as follows (1)
Presence of husk, (2) The quantity of sugar, (3) The presence of
foreign starch, (4) The addition of fats other than cocoa butter, especi-
ally common in the white portion of chocolate creams, and also in
cheaper bar chocolate, and in the coverings of chocolate creams. The
determination should include the following : —

Moisture — Mineral Matter. — If nothing but sugar and flavouring
have been added, the ash will be a direct indication of the amount of
cocoa present, especially if the microscopic examination shows that
husk is absent. Rarely, a little iron earth is added to improve the
colour of poor chocolate. This will give a higher ash in which iron
can be detected. The water-soluble ash is, in genuine cocoa, about 50
per cent of the total ; if it be higher than this, the presence of husk (or
alkali) is probable. The siliceous matter of the ash as determined by
the evaporation of the ash with HCl, and weighing the then insoluble
residue is important, as it rarely exceeds 0*05 per cent in the cocoa
nib, but reaches 1*2 per cent in the husk. The fat will be examined
in the same way as with cocoa, the fibre also determined as described
above, as well as the nitrogen (if required), the sugar and starch. The
cold water extract of cocoa being tolerably constant (on the fat-free
cocoa averaging 24 to 25 per cent), this figure which, in the case of
chocolate will represent the cold water extract of the cocoa present, to-

CHOCOLATE.

gether with the sugar added, allows the amount of cocoa to be calcu-
lated. In estimating the value of the determinations on a sample of
chocolate, corrections are to be made for the amount of added sugar
found, when- the results can be directly compared with cocoa standards.
For the composition of a number of samples of chocolate reference
should be made to the "Analyst" (xxxiv. 134).

Milk chocolate is a mixture of cocoa, sugar, milk-powder and
various flavourings. According to Dubois (" Jour. Amer. Chem. Soc."
1907, 556) the following analyses represent typical milk chocolate of
well-known brands :—

2
3

Polarization.

Sucrose.

Lactose.

Per cent of milk fat
in total fat.

Direct.

After inversion
at 24°

+ 21°
+ 23-22°
+ 23-88°

-2°

-2-22°

-2-20°

Per cent
40-9
45-7
46-8

Per cent
8-24
9-12
8-24

22-1
22-9
24-2

When both sucrose and lactose have to be determined, as in the
case of a milk chocolate Dubois' method may be used. Thirteen grms.
of the sample are freed from fat, and to the residue 100 c.c. of water is
added and the whole well shaken for ten minutes, 5 c.c. of basic lead
acetate solution are then added, and the solution filtered, and excess of
lead removed by HgS ; 25 c.c. of the filtrate are allowed to stand over-
night in order to attain its stable rotation and the polarimetric value
determined. Multiply this reading by 2. 50 c.c. of the filtrate are
inverted by acid, neutralized and made up to 100 c.c. Take the polari-
metric reading at the same temperature as the direct reading and also
take the reading at 86°. Multiply the readings by 2. The weights of
the two sugars may be calculated from the following formulae : —

Sucrose (in grms.) = (^ " ^) ^'Q^ ^ ;,^3

Lactose

142-66 -

19-152 c.

100

When a is the direct reading for normal weight.
When b is the invert reading for normal weight.
When c is the invert reading at 86°.

Baier and Neumann (" Analyst," xxxiv. 439) give the following de-
tails for the examination of milk or cream chocolate. The quantity of
milk- fat and casein present should be estimated ; the percentage of
milk-fat can be obtained from the Reichert-Meissl value of the fat
separated from the chocolate, whilst the amount of casein can be as-
certained by a modification of Hammarsten's method, which relies on

30 FOOD AND DRUGS.

the solubility of casein in ammonium oxalate solution and on the insolu-
bility of other proteins in this solution. The authors have found that
casein is completely soluble in sodium oxalate, but not in ammonium
oxalate. To estimate casein proceed as follows : Extract 20 grms. of
the powdered chocolate with ether in a Soxhlet apparatus for sixteen
hours ; allow the extracted residue to dry spontaneously, and rub down
10 grms. of it in a mortar with a small quantity of 1 per cent sodium
oxalate solution. Wash the paste into a 250 c.c. flask with about
200 c.c. of the oxalate solution, heat the mixture to boiling and add hot
sodium oxalate solution until the flask is nearly filled up to the mark.
Allow it to stand for about eighteen hours, occasionally shaking, then
dilute with cold sodium oxalate solution to a volume of exactly 250 c.c. ;
mix and filter. Add 5 c.c. of 5 per cent uranium acetate solution to
100 c.c. of the filtrate, then 30 per cent acetic acid drop by drop, con-
tinually stirring until the casein commences to precipitate. The
number of drops necessary vary from 30 to 120 according to the
quantity of casein present. Add 5 drops of the acetic acid in excess,
separate the precipitate by centrifugal action, and wash with a solution
containing 5 grms. of uranium acetate and 3 c.c. of 30 per cent acetic
acid per 100 c.c. As soon as the washings give no reaction for oxalates,
transfer the precipitate to a flask and determine the quantity of nitro-
gen present by Kjeldahl's process. Multiply this quantity of nitrogen
by 6' 37, thus obtaining the amount of casein, calculating it into a per-
centage quantity on the original chocolate. When calculating the
percentage of milk-fat it is assumed that cocoa butter has a Reichert-
Meissl value of 1*0 and milk-fat a value of 27*0. The total milk solids
present in the chocolate can be estimated from the quantities of casein
and fats therein. Fat, proteins, lactose and mineral matter are the
component parts of th© milk solids. To calculate the proteins multiply
the casein by 1*111 ; the lactose, multiply the protein by 1*3, and the
mineral matter, multiply the protein by 0*21 . The fat content of the
milk or cream used in the preparation of the chocolate can be estimated
by a simple calculation. The authors state that they consider that the
milk used should contain about 3*5 per cent of fat, and the cream
about 10 per cent. Milk chocolate should in their opinion contain at
least 15 per cent of dry milk solids, while cream chocolate should con-
tain at least 20 per cent of dry cream solids.

COFFEE.

Coffee berries are the seeds of Goffea Arabica and probably of
allied species of the natural order Cinchonacece. The commercial
product consists of the endosperm of the seed, the seed coats having
been removed during the preparation of the coffee. Small fragments
of the seed coats, however, may be found in the groove running along
the flat side of the berry, and naturally, also in ground coffee. The
coffee tree is cultivated in many tropical countries, into which it has
been introduced from Abyssinia and Ethiopia. India, Java, Ceylon and
Arabia furnish some of the best coffee, but at least half of the world's
supply comes from Brazil. The quality of coffee is a matter for the

COFFEE.

palate, and not for chemical analysis, the function of which is merely
to decide on its purity.

The raw cotfee berries are roasted to a greater or less degree before
use, and the greater part of the coffee sold in commerce consists either
of the roasted beans, as the berries are usually termed, or of the same
ground to a coarse powder. It is during the roasting of the berry that
the peculiar aroma of coffee is developed, and the original toughness
of the berry is destroyed, so that it can then be easily ground. The
changes brought about by roasting are as follows : A large quantity —
up to 20 per cent — of water and organic matter is driven off and the
sugar present is largely caramelized : a small amount of caffeine is
probably volatilized, but not very much. The fat and albumen are
partially decomposed, carbonic acid gas is given off and the berries
naturally swell. Traces of quinone, acetone, methylamine and similar
bodies are formed, and also a certain amount of caffeol, CgH^^O.^, which
is a heavy oil which appears to be responsible for the aroma of roasted
coffee. Samples of raw coffee, roasted in the author's laboratories to
a full rich brown colour, gave the following results, indicating the
changes which take place during roasting : —

II.

Raw.

Roasted.

Raw.

Roasted.

Per

cent

Moisture

12-45

4-1

11-9

3-7

Ash

3-72

3-95

3-66

3-82

Cellulose

26-82

25-00

28-5

26-8

Caffeine

1-2

1-36

1-40

Sugar

1-5

3-2

1-1

The Constituents of Coffee. — The principal constituents of coffee are
(1) Caffeine (identical with theine), (2) Caffetannic acid, Ci^H^gOg, in com-
bination with caffeine and with magnesia or lime, (3) Fat, (4) Albumen-
oids, (5) Carbohydrates, (6) Essential oil and aromatic substances, (7)
possibly other alkaloidal substances than caffeine.

Adulterants of Coffee. — Coffee is subject to a considerable amount
of adulteration ; this is generally only to be found in the ground variety,
although numerous cases of factitious coffee beans have been noticed,
but these latter are usually easily detected by the eye by any skilled
observer. The principal adulterant met with to-day is roasted chicory,
the root of Cichormm Intyhus.

Eoasted rye, wheat, dates, acorns and other similar vegetable
matter have occasionally been found, but to-day these are rarely used.
Sometimes, according to Konig, the berries are roasted with glucose,
which provides much caramel and makes the resulting infusion appear
stronger. Exhausted coffee, which has been used for the manu-
facture of extract of coffee, is also a recorded, but rare, adulterant. The

FOOD AND DKUGS.

factitious beans are moulded from one or more of the following ingredi-
ents; chicory and other roasted vegetable matter, china clay, wheat
flour, bran, sawdust, caramelized sugar, and lupin seeds. But, as

Parry

Krauth

Rupp

Parry

Krauth

Konig

Krauth

Per cent
21 to 26
22-5 „ 25-2

62 to 67
65-4

52 to 65

Water.

Per cent

1 to 4
1-5 „ 4-5
1-29

2 „ 4
4-3

12-85

•

■ B

1
1

Per cent

22 to 25

243

22-25
2214

"3 ' a ^ «,

O ! Ph

S i-l i-l

S o ,

(L, op op 'H«pt-
rH iH 'Jf* iH fH rtt (?q
i-l fH iH

Ash.

M5 Ttt

PL, (N f Op i-H Op
""^ "^ « OS O (N fH

Substance.

Coffee

Chicory (roasted)

>>
Acorns (roasted)
Wheat (roasted)

mentioned above, the adulterant found in the vast majority of cases is
roasted chicory root.

Eiley gives the following list of imitation "coffees" which have
been found in the course of examinations made in the laboratory of the

COFFEE.

United States Department of Agriculture. The mixtures had been
moulded and pressed into berries : —

Coffee, bran and molasses.

Wheat-flour, coffee and chicory.

Wheat-flour, bran and rye.

Chicory, peas and barley.

Wheat, oats and buckwheat.

Wheat-flour and sawdust.

Husks of leguminous seeds roasted, and molasses.

Pea hulls and bran.

Factitious coffee berries, however, are rarely met with to-day.

The analyses on page 32 represent the principal constituents of
coffee and several of the adulterants noted above.

Dyer (" Analyst," xxiii. 226) gives the following table of figures
for dried chicory (the moisture varied from 1 to 4 per cent) : —

Insoluble in
water

Ether
extract.

Nitrogen.

Ash.

Soluble Ash.

San I

Per

cent

ceilt

cent

Chicory nibs, medium roast

22-40

2-57

1-53

4-63

2-50

0-70

„ „ dark roast

50-30

1-43

1-67

4-70

2-99

.0-30

Ground chicory

22-27

2-17

1-33

5-53

2-43

il-43

21-50

1-90

1-34

5-23

2-07

1-43

35-50

3-43

1-50

513

2-57

0-77

37-80

3-87

1-52

8-23

1-60

3-97

22-77

3-17

1-25

5-13

3-30

1-66

22-50

3-67

1-23

573

3-23

1-63

23-50

2-60

1-29

5-63

2-97

1-47

22-50

2-60

1-29

5-33

3-22

1-47

22-63

2-57

1-29

5-70

2-80

1-47

The following figures are given by Allen as representing certain
factitious " coffees " : —

" Acorn

"Rye

"Barley

Coffee."

Per cent

Water

12-85

2-22

3-45

6-41

Nitrogenous matter

6-13

11-87

9-38

10-56

Fat

4-01

3-91

3-25

104

Sugar

8-01

- ^

Starch 1
Dextrin |-

62-00

8-34 \
49-51 {

70-13

68-38

Other non-nitrogenous matter j

9-83 J

Cellulose

4-98

9-78

4-25

10-50

Ash

2-02

4-54

3-36

3-04

! Soluble in H^O

61-33

31-20

34-37

Glucose (by inversion)
1

—

69-28

67-19

VOL. I.

FOOD AND DKUGS.

Tatlock and Thomson (" Jour. Soc. Chem. Ind." 1910, 29, 138)
give the following results of the analyses of a number of coffees, of a
coffee " free from caffeine," and a sample of chicory : —

CaflFeine.

Water
Extract.

Ash Sohible
in H2O.

Ash Insoluble
in H2O, less SiOg.

Silica.

Specific gravity

of 10 per cent

Infusion.

Per cent

CJosta Bica

1-22

30-80

3-06

0-77

trace

1-20

30-26

2-96

0-88

10102

1-38

30-77

3-21

0 98

1-0099

Mysore

1-18

31-02

301

0-93

—

1-25

2906

3-15

0-96

10102

E. India

1-46

29-10

3-32

0-97

1-0101

Mocha

1-19

30-76

3-14

0-85

1-0102

Caffeine-free

0-08

27-42

3-30

101

10101

Chicory

none

75-84

1-95

201

4-77

1-0274

These chemists consider that 30 per cent of matter extracted by
water for coffee and 75 per cent for chicory (calculated on the dried
substance) affords a fair basis for calculation of mixtures of the two.

The Analysis of Coffee.

Specific Gravity. — Much stress has been laid on the specific gravity
of coffee by the chemists of the Municipal Laboratory of Paris. By
the use of elaborate apparatus they have determined this figure for a
number of samples, and give 1-041 to 1*368 for unroasted, and 0*500
to 0635 for roasted coffee. The author has found that for ordinary
cases this figure yields absolutely no information. It is of use only in
a few cases where actual factitious whole beans are present, for these
are almost invariably heavier than water, whilst genuine roasted beans
are lighter — unless, according to Allen, they have been much over-
roasted. Allen takes the average specific gravity, if the beans do not
float in water, and are therefore presumably factitious, by immersing
twenty beans in brine and gradually adding water until ten sink and
ten float. The specific gravity of the liquid is taken as the mean
specific gravity. This determination is certainly of but little value and
does not justify the amount of trouble that has been taken over it.

The following determinations should be made : —

Ash. — The ash of pure coffee is usually between 3*5 and 4*5 per cent,
rarely exceeding 4*8 per cent. The ash of chicory is distinctly higher,
and this figure may afford useful information on analysis. The com-
position of the ash of coffee, however, differs in a marked degree
from that of chicory, and by determining the amount of soluble ash
and silica present an approximation as to the amount of chicory
present may be made. Not less than 60 per cent — usually 70 to
80 per cent of the ash of pure coffee — is soluble in water, whereas
only 25 to 35 per cent of the ash of chicory is soluble in water. The
ash of coffee is free from silica, whereas that of chicory contains a not-

COFFEE.

able quantity. The following figures may be taken as covering most
samples : —

CoflFee.

Chicory.

Per cent

Silica and Sand

CO,

Fe,03

P2O5

Soluble Ash

14-92

0-44 to 0-98
0-26 „ 1-11

10 „ 11

75 „ 85

10-69 to 55
1-78 „ 319
3-13 „ 5-32
3-28 „ 4-93
5 „ 6

21 „ 35

It is evident, therefore, that an analysis of the ash renders very
useful information.

A comparison, however, of the soluble ash in the two substances is
often vitiated by the high amount of sand present in the chicory root.
As Allen points out, by comparing the soluble ash with the total ash
minus the sand and silica, more reliable results are obtained, and a rough
indication of the percentage of chicory present may be deduced. With
pure coffee, the amount of soluble ash is, as given above, usually 75 to
85 per cent, whereas in chicory (after deducting the sand and silica) it
is from 38 to 45 per cent or thereabouts.

The ash of dandelion root also contains a high amount of silica —
varying from 10 to 14 per cent of the total ash.

Determinatio7i of Fat. — The fat of pure coffee is fairly constant in
amount. It may be determined in the dry powdered coffee by extraction
with ether and petroleum ether and varies between 10 and 14 per cent,
or in rare cases 16 per cent. Chicory yields about 1 to 2 per cent, so
that any undue proportion of this substance will be indicated by the
low fat yield. Most other adulterants yield a very low fat value.

Aqueous Extractive. — The watery extract of coffee is not only very
constant in amount, but is sensibly less than that of chicory. Instead of
actually weighing the extracted matter, Graham, Stenhouse and Camp-
bell (" Journal Chem. Soc." ix. 38) preferred to take the specific gravity
of the aqueous infusion. Their method was to take the powdered sub-
stance with ten times its weight of cold water and raise the liquid to
the boiling-point, and on cooling to 15*5° to take the specific gravity.
The following were the results obtained : —

Substance.

Specific gravity.

Coffee

Chicory

Leguminous seeds

Acorns

Dandelion root

Cereals

Per cent
1-008 to 1-009
10191 „ 1-0233
1-0057 „ 1-0084
1-0073
1-0219
1-0109 „ 1-0263

Per cent
Average 1-0087
1-02105

36 FOOD AND DKUGS.

Allen, by well boiling the cotfee and filtering and washing till the
filtrate measured 10 c.c. for each gi'amme of the coffee employed, found
1-0079 as the average specific gravity. He prefers to dry the sample
first, and then adopts the values 1-009 for coffee and 1*024 for chicory.
On this basis which yields fairly approximate results, the percentage of
coffee in a mixture of chicory and coffee may be approximately deduced
from the equation,

(1024 - d) 100
^" 15

where P is the percentage of coffee, and d is the specific gravity of the
10 per cent infusion.

iMcGill prefers to boil the finely powdered sample for an hour
under a reflux condenser and filter and make up to the requisite
volume. He then adopts the values 1-00986 for coffee and 1-02821
for chicory.

Macfarlane, after extracting the dried sample with petroleum ether,
and again drying the sample, then extracts with water. He gives the
following table for the dried watery extract obtained : —

Per cent

Soutas coffee 22-44

Mocha „ 21-92

Java „ 20-42

and 10 per cent chicory 25-90

„ 20 „ „ „ 30-75

„ 30 „ „ „ 37-40

„ 40 „ „ „ 43-36

„ 50 „ „ „ 49-84

„ 60 „ „ „ 53-82

„ 70 „ „ „ 60-34

„ 80 „ „ „ 65-93

„ 90 „ „ „ 71-41

Chicory 77-73

Calculated on the dried substance 24 per cent may be taken as the
value for the aqueous extract of normal coffee and 75 per cent for that
of chicory. These are fair average values.

As the results of the examination of over 100 samples the author
prefers to adopt Allen's suggestions — namely, of boiling the sample for
twenty minutes, with about six times its weight of water, filtering and
washing with warm water and making up to ten times the weight of the
coffee used at 60'. The values 1*009 for coffee and 1-025 for chicory

will then give very close results. The equation P = r^ —

will then give the amount of coffee in a mixture, where P is the pei--
centage and d is the specific gravity.

A. E. Johnson prefers to weigh the aqueous extract, which he gives
as 24 per cent for dried roasted coffee, and about 70 per cent for dried
chicory. (Hehner has found as low as 54-1 per cent in a pure chicory.)
To determine the extractive matter Johnson boils 5 grms. with 200 c.c.
of water for 15 minutes, strains, and again boils the residue with 50 c.c.
of water for 5 minutes. The liquids are mixed, made up to 250 c.c.

COFFEE. 37

when cold, and filtered ; 50 c.c, equivalent to 1 grm. of the sample, are
dried over a water bath and finally in a water oven, and weighed. The
amount of coffee in the sample is calculated on the average values 24
and 70 as mentioned above. This method is simple and gives excellent
results.

The following are average values for the water-soluble extractives
in certain substances used as adulterants of coffee : —

Roasted rye . . . . 30 to 36v5 per cent

Boasted wheat . . . . 47 ,, 55 „

Roasted figs .... 60 „ 67-5

The refractive index of the aqueous infusion (10 per cent) has been
used as a means of discriminating between genuine coffee and some of
its substitutes.

Lythgoe has examined a number of samples and finds the refractive
index of the 10 per cent extract at 20° to lie between 1-3374 and
1-33804.

The same extract of roasted chicory had a refractive index 1-34463.

Graham, Stenhouse and Campbell ("Journ. Chem. Soc." ix. 36)
suggested a comparison of the tinctorial value of coffee infusions as a
method of detecting and determining the amount of adulterants. They
found that infusions of pure coffee, compared with similar infusions of
chicory in the usual manner in Nessler glasses, showed only one-third
of the colour of the latter. Allen considers that 2-8 to 3-2 was a fair
value for the colour of the chicory infusion as compared with that of
coffee and recommends a standard colour being kept which is made up of
ferric, cobalt, and copper sulphate, of exactly the depth of tint as a
standard 1 in 200 infusion of 50 per cent coffee and 50 per cent chicory.
Such a standard is unalterable and can be kept. An infusion of the
sample to be compared is made (1 in 100) and if it be pure coffee, the
colour will be equal to that of an equal volume of the standard colour.
If chicory be present, the colour will be darker, and water must be
added until the two tints are identical. This is done in graduated
Nessler tubes and the amount of chicory present is calculated on the
basis of the figures 3 for chicory and 1 for coffee.

The author has examined this tinctorial comparison with samples
of coffee and chicory which have been roasted for different lengths of
time under different conditions, and consider that very variable results
may be obtained from the same sample thus differently treated.

It is considered a waste of time to make the comparison when so
much more reliable results can be obtained by so simple a method as
the determination of the specific gravity of the aqueous extract.

Determination of Gajfetannic Acid. — Digest 2 grms. of the powdered
sample for thirty-six hours with 10 c.c. of water : then add 25 c.c. of 90
per cent alcohol and continue the digestion for twenty-four hours. The
liquid is then filtered and the residue washed with 90 per cent alcohol.
The filtrate is heated and a boiling concentrated solution of lead acetate
is added, which causes the precipitation of caffetannate of lead, which
contains 49 per cent of lead. When this has become flocculent, it is
filtered off, and washed with 90 per cent alcohol, until the washings
are free from lead, and then with ether until free from fat, dried at

FOOD AND DRUGS.

100° and weighed. The precipitate may be taken as containing 50
per cent of caffetannic acid.

Deter7nination of Pentosa7is. — (See under cocoa, p. 23.) Hehner
and Skertchly give the following results for some coffees and chicories : —

Moisture.

Pentosans,

Crude Fibre.

Per cent

Raw coffee

—

2-86

Roasted „

—

2-50

7-36

Coffee with 32 per cent chicory

2-71

High dried Belgian chicory

5-51

514

5-47

„ „ roasted 5 minutes

6-17

5-55

6-57

„ „ „ 10 „

3-95

5-16

6-87

„ „ „ 15 »

3-73

4-80

867

„ „ „ 23 „

3-28

5-56

11-50

(Carbonized matter is included with the fibre.)
The Determination of Caffeine. — This determination is not often
required in samples of coffee, as the quality of coffee does not bear a
direct ratio to its alkaloidal value.

Paul and Cownley (" Ph. Jour.," [3] xvii. 565, 648, 821, 921) recom-
mend the determination as of value, as the amount of alkaloid is, they
claim, fairly constant, and a determination may give an indication as to
the amount of coffee in a mixture. They showed that most published
processes gave results below the truth. They find in dried coffee 1*20
to 1*39 per cent of caffeine, and adopt 1-3 per cent as a mean value.
Allen prefers the figure 1*2 per cent as a safer average. Paul and
Cownley operate as follows : —

Five grms. of the sample, finely powdered, are well mixed in a
mortar with 2 grms. of calcined magnesia, the whole moistened with
hot water, well triturated and then dried at 100°. The mixture is ex-
tracted with boiling alcohol, and the resulting liquid evaporated nearly
to dryness. It is then boiled with 50 c.c. of water and heated with
a few drops of dilute H2SO4. When cold, the liquid is repeatedly ex-
tracted with chloroform until exhausted. The mixed chloroform solu-
tion is then treated with a 1 per cent solution of caustic alkali to remove
colouring matter, and the separated chloroform solution evaporated
and the practically pure caffeine weighed.

According to Allen, a trace of caffeine is left in the sample by this
treatment, and can be extracted by water after the alcoholic extraction.
He also states that six or seven successive extractions with 30 to 40 c.c.
of chloroform are necessary.

In case the resulting caffeine — as is often the case w^ith caffeine ex-
tracted from coffee, but not so with caffeine from tea — should be some-
what brown (due to the presence of a waxy or resinous impurity), it
should be purified by re-solution in boiling water and subsequent filtra-
tion, and evaporation of the water.

There are numerous other methods of determining the caffeine, but
as they are usually slight variations one of the other, and give results

COFFEE. 39

of, generally, less accuracy than the above, it is not necessary to more
than refer the reader to the papers of Paul given above, that of Allen's
pupils (" Pharm. Journ." [3], xxiii. 215), and to briefly draw attention to
the process preferred by Allen and those recommended by Juckenack
and Hilger, and by Lendrich and Nottbohm.

Allen exhausts 12 grms. of the finely powdered coffee by boiling
under a reflux condenser with 500 c.c. of water. After six to eight
hours boiling, the liquid is filtered, and the residue washed on a filter,
making the filtrate up to 600 c.c. This is then heated to about 95°
C, and about 4 grms. of powdered lead acetate added, and the whole
l)oiled under a reflux condenser for ten minutes. If the precipitate
does not curdle and settle readily, leaving the liquid of at most a pale
colour, a further addition of lead acetate, and further boiling are neces-
sary. The liquid is now filtered, after it has been made up to 600 c.c.
when cold, and 500 c.c. of the filtrate (equivalent to 10 grms. of
coffee) are evaporated to about 50 c.c. and a little sodium phosphate is
added to precipitate the remaining lead. The liquid is filtered, the
precipitate washed, and the filtrate is further concentrated to about 40
c.c, when the caffeine is extracted in a separator by four to five succes-
sive treatments with chloroform. The chloroform solutions are mixed,
the chloroform distilled off in a tared flask, over a water bath. The
last traces of chloroform are removed by a current of air, and the alka-
loid weighed.

In the presence of chicory the caffeine is liable to be strongly
coloured, and it should then be redissolved in water, a few drops of
solution of caustic alkali added, and the liquid again extracted as be-
fore with chloroform.

Juckenack and Hilger exhaust 20 grms. of the sample with 900 c.c.
of water by boiling for about three hours, and whilst the liquid is cooling,
add 75 c.c. of a 10 per cent solution of basic aluminium acetate when
the thermometer marks about 70°, and then 2 grms. of sodium bicar-
bonate. The liquid is again boiled for five minutes, and on cooling
made up to 1020 c.c, 750 c.c. of which is filtered. This is equivalent
to 15 grms. of coffee. This liquid is evaporated with 10 grms. of
powdered aluminium hydroxide and the residue is extracted with carbon
tetrachloride in a Soxhlet tube. The solvent is evaporated and the
alkaloid weighed. The residue should be slightly moistened with water
before extraction — or all the caffeine will not be dissolved. [It is noted
that in treating an aliquot part of the liquid this process allows for the
20 grms. of coffee added, whilst Allen makes no such allowance.]

Lendrich and Nottbohm ("Zeit. Untersuch. Nahr. Genuss." 1909,
17, 241) give the following as the most accurate method of obtaining
the caffeine in a state of great purity. Twenty grms. of the coffee
ground to a fine powder, are moistened with 10 c.c of water, the mass
being stirred from time to time for a period of two hours. The moist
mass is then transferred to an extraction thimble and extracted with
carbon tetrachloride for three hours. To the extract 1 grm. of paraffin
wax is added, and the carbon tetrachloride is evaporated. The residue
is extracted with four successive quantities of boiling water, using 50
c.c for the first extraction and 25 c.c. for each of the subsequent ones.

40 FOOD AND DKUGS.

The united extracts are passed through a moistened filter paper which
is washed with hot water. The filtrate is treated with 10 to 30 c.c. of
a 1 per cent solution of permanganate of potash for 15 minutes at ordin-
ary temperature, and the excess of permanganate destroyed by the ad-
dition of a 3 per cent solution of hydrogen peroxide containing 1 per
cent of acetic acid. The whole is then heated on a water bath and
filtered, the residue being washed with hot water. The filtrate is now
evaporated to dryness, the residue dried in a steam oven and at once
extracted with warm chloroform. The chloroform is evaporated, and
the resulting caffeine dried at 100° for thirty minutes and weighed.

Microscopic Examination. — An examination of powdered coffee
under the microscope affords the principal means of identifying
adulterants in coffee. In this examination it is useful to compare the
tissues with those found in sections of the raw coffee beans, since, with
the exception of some alteration in the cell contents, roasting has no
material effect on the appearance of the tissues. To fully examine
coffee, the fine and coarse particles may be separated by a sieve
and sections of the coarser particles made. The tissues are too dark
for useful observation, and may be decolorized by sodium hypochlorite
and mounted in diluted glycerine. The principal portion of coffee is
composed of fragments of the endosperm — thick- walled angular cells,
very tough and hard, and adhering firmly to each other. Globules of oil
are enclosed in the cells but no starch is present in them. The epi-
dermal and immediately neighbouring cells have evenly thickened walls,
bnt the remainder of the endosperm tissue consists of parenchymatous
cells with thickened walls and very large pits — often as long as the cell is
wide. A small portion of the seed coat (the bulk of which has been
removed during the preparation of coffee for use) is always present, and
is revealed by its very long sclerenchymatous cells, with numerous
pits. A few spiral vessels are to be found — derived from the raphe or
groove of the berry, but these are not numerous. The minute number
of embryonic cells are rarely identifiable in ground coffee. These char-
acters are sufficient to distinguish coffee from all its adulterants. The
presence of the various starches is easily detected, and such adulterants
as ground dates, roasted figs, etc., are widely different in their microscopic
characters from coffee beans. The principal adulterant, chicory, re-
veals numerous loose, thin-walled parenchymatous cells, laticiferous
vessels, sieve-tubes with transverse plates and large vessels with well-
defined large pits, which cannot possibly be mistaken for ground
coffee.

Boasted date stones, which have been used as a substitute, and
probably also as an adulterant of coffee, have a very characteristic
appearance under the microscope. The epidermal cells are almost
oblong, whilst the parenchymatous tissue consists of very irregular-
shaped cells containing much tannin.

CHAPTER II.

MILK, BUTTEE, CHEESE, AND EDIBLE OILS.

MILK.

Milk is the fluid secreted by the mammary glands of female mammals
for the nourishment of their young. It contains all the essentials of a
complete food, namely, fat, sugar, proteids and mineral matter. The
only milk of importance from the point of view of the present work is
cow's milk.

Although milk is a natural product of variable quality, it is one of
the few foods in regard to which section 4 of the Sale of Food and
Drugs Act of 1899 empowers the Board of Agriculture to make legal
standards. In pursuance of that power, the Board of Agriculture
issued the following regulations on 5th August, 1901.

Milk. — (1) When a sample of milk (not being milk sold as skimmed,
or separated, or condensed milk) contains less than 3 per cent of milk-
fat, it shall be presumed for the purposes of the Sale of Food and
Drugs Acts, 1875 to 1899, until the contrary be proved, that the milk
is not genuine, by reason of the abstraction therefrom of milk-fat, or
the addition thereto of water.

(2) When a sample of milk (not being milk sold as skimmed, or
separated, or condensed milk) contains less
than 8*5 per cent of milk solids other than
milk-fat, it shall be presumed for the purposes
of the Sale of Food and Drugs Acts, 1875
to 1899, until the contrary is proved, that
the milk is not genuine, by reason of the ab-
straction therefrom of milk solids other than
milk-fat, or the addition thereto of water.

Skimmed or Separated Milk. — (3) When
a sample of skimmed or separated milk (not
being condensed milk) contains less than 9
per cent of milk solids, it shall be presumed ^^^'p faTanJeTZk. '^'
for the purposes of the Sale of Food and

Drugs Acts, 1875 to 1899, until the contrary be proved, that it is not
genuine, by reason of the abstraction therefrom of milk solids other
than milk-fat, or the addition thereto of water.

(4) These regulations shall extend to Great Britain.

There are several important legal decisions in reference to this
matter, which will be discussed in Volume II, but it may be well to
here mention that milk below the standard is not necessarily to be re-
garded as adulterated, nor is milk above the standard necessarily to be

(41)

42 FOOD AND DRUGS.

regarded as pure. The effect of these regulations is to fix the burden of
proof on the defendant where the milk was presumably adulterated ; in
other words, the failure to satisfy the standards raises a presumption of
adulteration, but one which is rebuttable by evidence.

Milk is essentially an emulsion of fine globules of fat in a solution
containing milk sugar, casein and mineral salts. Under the microscope,
the cream has the appearance shown on p. 41. The fat globules vary
much in size, varying from about 0"0015 mm. to 0*009 mm. in diameter.
Minute particles of separated proteids may sometimes be seen. Colos-
trum is the term applied to the milk yielded by the females for
a few days after the birth of their young and differs materially in
composition from normal milk, from which it should be absent. It
may be recognized under the microscope by its containing numerous
circular cells containing fat globules which have not yet been liberated
by the disintegration of the cell wall.

Fresh cow's milk is usually amphoteric in reaction, that is, it yields
both acid and alkaline reactions. To phenol-phthalein, however, it is
always slightly acid. It soon becomes distinctly acid owing to a
gradual conversion of lactose into lactic acid.

The Composition of Milk. — The qualitative composition of milk is
fairly constant, but the quantities of the various constituents are liable
to considerable variation. The following gives the average composition
of normal cow's milk : —

Water 87 -3 per cent.

Fat 3.6

Proteids 3-8

Lactose 4-5

Mineral matter 0-7

Citric acid 01

Colouring matter traces

The fat of milk, which will be dealt with under butter, is a mixture
of glycerides of the non-volatile fatty acids (olein, palmitin, stearin,
and myristicin, about 90 per cent) and of the glycerides of the soluble
volatile fatty acids (butyrin, caprylin and caprinin, 10 per cent).

The proteid matters of milk are composed of the following substances :
Casein constitutes about 80 per cent of the total proteids of milk and
is probably only partly dissolved in the milk and partly held in a some-
what colloidal state in the liquid. It is a white, colourless, almost
tasteless soUd, soluble in alkalies, but precipitated by dilute acids. It
is loBvo- rotatory, alkaline solutions having a specific rotation of about
- 90°. Lactalbumin is the soluble albumin of milk and is present to the
extent of about 15 per cent of the proteids. It resembles egg albumin,
and is coagulated at 70 to 72°. It is readily soluble in water and has
a specific rotation of about - 68°. There are traces of a proteid which
has been termed lactoglobulin, and numerous others have been described
from time to time. They are, however, of little importance from an
analytical point of view, and their literature may be found w^ell sum-
marized by Droop Richmond (" American Chem. Jour." 1893, October).

Lactose will be found fully described under " Carbohydrates ".

MILK.

The mineral matter of milk has the following average com-
position : —

Potassium oxide
Sodium ,,

Calcium „
Magnesium „
Iron „

SO,

PoO,
Chlorine

Per cent.

25-02

10-01

20-01

2-42

0-13

3-84

24-29

14-28

As the result of the examination of over 170,000 samples Vieth and
Richmond give the following average values : —

Specific gravity
at 15° C.

Total SoUds.

Fat.

Solids not Fat.

Per

cent

1-03215

Per
cent
12-86

Per
cent
4-02

Per
cent
8-84

These analyses extended over fifteen years, the minimum and maxi-
mum average for any year being as follows : —

Minimum
Maximum

Specific gravity
at 15° C.

Total SoUds.

Fat.

Solids not Fat.

Per

cent

1-0315

1-0323

Per
cent
12-66
13-06

Per
cent
3-84
422

Per

cent
8-68
8-88

For the past ten years, the average fat value has been 3*75 per
cent, according to Eichmond.

Naturally there are variations outside these limits, for even milk of
normal quality, and, naturally, still greater variations for milk obtained
under abnormal conditions.

Milk cannot be considered normal unless it consists of the well-
mixed total quantity obtained in a milking. The first portion of the
milk leaving the udder is known as fore milk and contains less fat
than the last portion, known as the strippings, which may contain up
to 8 or 10 per cent of fat. The evening milk is nearly always richer in
fat than the morning milk. The following represent average differences
between fore milk and strippings on the one hand, and morning and
evening milk on the other : —

FOOD AND DEUGS.

Water.

Total Solids.

Fat.

Per

cent

(a) Fore milk

88-0

12-0

1-4

Strippings

82-0

18-0

8-8

(6) Fore milk

88-5

11-5

1-45

Strippings

81-0

19-0

9-6

87-8

12-2

1-25

Strippings

80-8

19-2

9-7

Morning Milk.
Sp. gr. Solids.
1-0322 12-53

Evening Milk.
Fat. Sp. gr. Solids. Fat.

3-68 1-0318 12-94 4-04

The variations in the composition of morning and evening milks
throughout twelve months are fully recorded by H. D. Eichmond
("Analyst." xxiv. 197).

The following analyses of colostrum are due to Engling : —

I'ime after Calving.

Specific gravity.

Fat.

Casein.

Albumin.

Sugar.

Ash.

Total Solids.

Per

cent

Immediately

1-068

3-54

2-65

15-56

3-0

1-18

26-93

10 hours

1-046

4-66

4-28

9-32

1-42

1-55

21-23

24 „

1-048

4-75

4-50

6-25

2-85

1-02

19-37

48 „

1-042

4-21

3-25

2-31

3-46

0-96

1419

72 „

1-085

4-08

3-33

1-03

4-10

0-82

13-36

Milk from underfed cows comes within the category of abnormal
milk. It has been carefully examined by Carter-Bell ("Analyst,"
VI. 63) ; he gives the following figures : —

Specific gravity.

Total SoHds.

Fat.

Ash.

Per cent
1-028 to 1-031

Per cent
9-10 to 13-7

Per cent
1-06 to 4-34

Per cent
0-64 to 0-75

Milk that has been frozen is abnormal, since it is principally the
water that freezes out. Droop Eichmond (" Analyst." xviii. 53) gives
the following figures : —

MILK.

Frozen Portion.

Unfrozen Portion.

Per cent

Water

96-23

85-62

Fat

1-23

4-73

Sugar

1-42

4-95

Proteids

0-91

3-90

Ash

0-21

0-80

Specific gravity

10090

10345

Infected milk may result from disease in the cow, or by after-infec-
tion by contact with diseased persons, the use of dirty vessels, etc.
Tuberculosis, diphtheria, scarlet and typhoid fevers are all milk-borne
diseases, and as milk forms a most favourable medium for rapid de-
velopment, the organisms when once present increase with alarming
rapidity, except in a few cases, such as the bacillus of tuberculosis,
which does not increase in milk.

Milk may also be contaminated by non-pathogenic organisms which
cause marked changes in the physical character of the milk. Chromo-
genetic bacteria are sometimes present in the milk causing the condi-
tions known as blue milk (due to bacillus cyanogenus), red milk (due to
bacillus erythrogenus), and yellow milk (due to bacillus synxanthus).
The fermentation in blue milk is marked by the production of a blue
colour, which is changed to cherry red by alkalies, but the blue colour
is restored by acids. The bacillus may be identified by cultures from
the small patches formed in the milk ; on gelatine plates it forms
rounded, dirty- white, finely granular colonies with smooth outlines;
the surrounding gelatine takes on a light-green or greenish-brown
colour. In a primitive culture on gelatine, the remaining gelatine is
greenish-blue, sometimes nearly black. On potatoes it forms a
yellowish layer near the point of inoculation, the surrounding potato
being stained blue. It is an aerobic and exceedingly mobile bacillus of
1 to 4 /A in length and 0"3 to 0*5 /x in breadth, with numerous flagellae,
and, when spores are present, club-shaped ends. Spores are frequently
present in the middle as well as the ends of the rods. Grown in milk
soured by lactic acid, it causes an intense blue coloration.

Ropy milk is due to the presence of organisms, which cause the milk
to become very viscid and stringy, so that it may be drawn up in
threads by a spoon.

Diseased milk is a subject more for the veterinary surgeon and the
pathologist than for the analyst, but the detection of the bacillus of
tuberculosis is frequently asked of the analyst, so that a few words on
the subject may not be out of place.

To show the presence of tubercle bacilli in tuberculous milk, obtain
the sediment for examination, by passing through a centrifugal appa-
ratus. The sediment will be found to contain almost the whole of the
bacilli with the mucus and solid particles. When the apparatus is un-
obtainable the best plan is to allow the milk to stand in a funnel-
shaped separator, for about twenty-four hours. The sediment collected

46 FOOD AND DEUGS.

at the bottom of the separator can be drawn off by means of the tap
and a drop dried on a glass sHde.

The preparations are stained in a solution made as follows : —
Take 1 part fuchsine and dissolve it in 10 parts of absolute
alcohol, then add 100 parts of a 5 per cent solution of carbolic acid
and heat the mixture until it steams freely. It takes three or four
minutes, or even less, to stain cover glass preparations, whilst seven or
eight minutes are necessary for the staining of sections. After getting
rid of the superfluous fluid place the preparations in 90 per cent alcohol
for a second or two, then plunge into a 25 per cent solution of sulphuric
acid, when it will be noticed that the pinkish tinge has become a
yellowish-brown. Wash the preparations in alcohol, and if they have
sufficiently changed colour, place in water holding lithium carbonate
in suspension. This process being completed they may be stained with
a dilute solution of methylene blue.

The Analysis op Milk.

The usual determinations in milk analysis involve the following : —

Specific gi'avity.

Total solid matter.

Fat.

Mineral matter.

Sugar.

Proteins.

Unless required for special purposes, the last three determinations
are not usually made in the analysis of samples under the Food and
Drugs Act, as the first three are usually sufficient to decide as to the
purity of a sample.

Specific Gravity. — Pure normal milk rarely has a specific gravity
below 1*031, sometimes rising to 1-035.

Total Solid Matter. — The legal minimum (see p. 41) for the total
solid matter in milk is 11*5 per cent, but it is rarely that a genuine
milk falls so low, 12*5 to 13*5 per cent covering the majority of pure
samples, although occasionally 15 per cent will be found.

Determination of the Total Solids of Milk. — About 5 grms. of the
milk are dried on a water bath in a small platinum capsule to constant
weight and the residue weighed. The milk should preferably be weighed,
although the error is very small if 5 c c. be delivered from an accurate
pipette and the result obtained divided by the specific gravity of the
milk. The dishes should be flat-bottomed and the time of heating re-
quired will be, usually, about 5 hours for 5 grms. A skin forms over
the surface of the milk as it dries making it somewhat difficult for the
water to escape. This may be broken by a fine needle from time to
time. Allen and Stokes prefer the use of porcelain dishes for the deter-
mination of the milk solids. Numerous devices to accelerate the drying
of milk for the determination of the solid matter have been proposed,
but they are of no practical advantage. For two of these reference
may be made to the "Analyst" (xvn, 227, xvii. 79).

The use of a little recently ignited sand, or asbestos, accelerates the
drying of the milk, however.

MILK.

Determination of the Mi^ieral Matter of Milk. — The platinum dish
containing the above sohd residue is heated to a low red heat, cooled
in a desiccator when all the organic matter is consumed, and weighed.
Prolonged or excessive heat causes a slightly low result to be obtained
owing to the volatilization of chlorides. The ash of normal milk
amounts to about 8 per cent of the non-fatty solids, or say, from
0*68 per cent to 0*78 per cent of the milk.

Determination of the Fat. — There are numerous methods of deter-
mining the fat in milk, but only
a few of these will be described,
many others being fully described
in textbooks devoted to milk only.

{a) The Adams method. This
method has the advantage of pro-
bably being the most accurate
process known. A strip of fat-
free absorbent paper about 2
inches wide and 24 inches long
is rolled loosely into a coil and
held by a wire, so that it can be
conveniently suspended. Either
5 c.c. of milk are delivered by a
pipette, slowly on to the coil, so
that every drop is absorbed, or a
beaker containing about 5 c.c. of
milk is accurately weighed and
the coil inserted, and when as
much as possible is absorbed, the
beaker is re- weighed, so that the
amount of milk used is known.
The coil is now hung up and air-
dried and then transferred for a
short time to the water oven.
When completely dry it is trans-
ferred to the Soxhlet extraction
apparatus (Fig. 3).

The coil is extracted for two
to three hours, the tared flask
containing the fat being heated
at first on the water bath, and
finally in the air oven, until of
constant weight.

Soxhlet's serometric process is as follow^s : 200 c.c. of milk are run
into the flask H, and 10 c.c. of normal potash solution added, and 60
c.c. of ether, which has been saturated with water. The mixture is
well shaken in the closed flask for fifteen minutes, and the liquids
allowed to separate. By working the rubber bellows, sufiicient of the
ether solution may be transferred to the tube B, which is water-jacketed,
to float the special form of hydrometer called the aerometer. The
tube connecting the flask H with B is now closed by the clip. The

Fig. 3. — Soxhlet apparatus for milk fat.

FOOD AND DRUGS.

amount of fat is determined by taking the reading of the aerometer,
when the water jacket is exactly at 17'5° C. The reading then corre-
sponds with the quantities of fat found in the annexed tables.

Fig. 4. — Soxhlet's ferometric apparatus.

This process is not used in this country, but is still employed to n.
small extent on the Continent.

MILK.

Soxhlet's Fat Table.

(In the specific gravities given here the figure 7 is omitted all through, as the
small aerometer is thus graduated. Thus 21-1 implies a gravity of 0-7211.)

"rt

fee-

^?;

feo-

s
o

fee-

*2 ^

. ^

■*^ ^-

-»-» I-

.»«

-*-s 1-

»o

oT'r^

cg-ji:

211

0-00

25-5

0-41

29-9

0-82

34 3

1-22

38-7

1-64

21-2

0-01

25-6

0-42

0-83

34-4

1-23

38-8

1-65

21-3

0-02

25-7

0-43

301

0-84

34-5

1-24

38-9

1-66

21-4

0-03

25-8

0-44

30-2

0-85

34-6

1-24

1-67

21-5

0-04

25-9

0-45

30-3

0-86

34-7

1-25

.391

1-68

21-6

0-05

0-41)

30-4

0-87

34-8

1-26

39 2

1-69

21-7

00.)

261

0-47

30-5

0-88

34-9

1-27

39-3

1-70

218

0-07

20 2

0-48

30-6

0-88

1-28

39-4

1-71

21-9

0-08

26-3

0-49

30-7

0-89

35-1

1-29

39-5

1-72

0 09

26-4

0-50

MO-8

0-90

35-2

1-30

39-6

1-73

22-1

0-10

26-5

0-50

30-9

0-91

35-3

1-31

39-7

1-74

22-2

0-11

26-<)

0.51

092

35-4

1-32

39-8

1-75

22-3

0-12

26-7

0-52

31-1

0-93

35-5

1-33

39-9

1-76

22-4

013

26 8

0-53

31-2

0-94

35-6

1-33

1-77

22-5

0-14

26-9

0-54

31-3

0-95

35-7

1-34

401

1-78

22 0

015

0-.55

31-4

0-95

35-8

1-35

40-2

1-79

22-7

0-l()

27-1

0-56

31-5

0-96

35-9

1-36

40-3

1-80

22-8

0-17

27-2

0-57

31-6

0-97

1-37

40-4

1-81

22-9

018

27-3

0-58

31-7

0-98

36-1

1-38

40-5

1-82

0-19

27-4

0-59

31-8

0-99

.36-2

1-39

40-6

1-83

23-1

0-20

27-5

0-150

31-9

100

36-3

1-40

40-7

1-84

23-2

0-21

27-6

0-60

101

36-4

1-41

40-8

1-85

23;;

0-J2

27 7

0-61

:^2-l

1-02

36-5

1-42

40 9

1-86

23-4

0-23

27-8

0-62

322

1-03

36-6

1-43

1-87

23-i5

0 24

27-9

063

32-3

1-04

36-7

1-44

411

1-88

23-6

0'25

0-64

32-4

1-05

36-8

1-45

41-2

1-89

23-7

0-25

281

0-65

32-5

105

36-9

1-46

41-3

1-90

23-8

0-26

2S-2

0-66

32-6

1-06

1-47

41-4

1-91

2:{-9

0-27

28 3

0-67

32-7

1-07

37-1

l-4«

41-5

1-92

0-28

28-4

0-68

32-8

1-08

37-2

1-49

41-6

1-93

24-1

0-29

28-5

0-69

32 9

1-09

37-3

1-.50

41-7

1-94

24-2

0-30

28-6

070

1-10

37-4

1-51

41-8

1-95

24-3

0-30

28-7

0-71

381

1-11

37-5

1-52

41-9

1-96

24-4

0-31

28-8

0-72

33-2

1-12

37-6

1-.53

1-97

24-5

0-32

28-9

0-73

33-3

1-13

37-7

1-54

421

1-98

24-6

0-33

0-74

33-4

114

37-8

1-.55

42-2

1-99

24-7

034

29-1

0-75

33-5

1-15

37-9

1-.56

42-3

200

24-8

0-35

29-2

076

33-6

115

1.57

42-4

•201

24-9

o-3r,

29-3

0-77-

33-7

116

38-1

1-58

42-5

202

0-37

29-4

0-78

33-8

117

.38 2

1-59

42-6

203

25-1

0-38

29-5

0-79

33-9

1-18

38-3

1-60

42-7

2-04

25-2

0-39

29-6

0-80

34 ,119|

38-4

1-61

42-8

205

25-3

0-40

29-7

0-80

341

1-20

38-5

1-62

42-9

2 06

25-4

0-40

29-8

0-81 34-2 1

1-21 1 38-6 1

1-63

207

VOL. I.

FOOD AND DEUGS.

■»J

_jj

e9 ^

.fji

foo-

&o-

&o- 1 1

febo-

fee-

die

^ fe

d,;e

l«

die ^^

<^ aj

CQr^

^ ^

02 rH

^ a,

CCr-l fe Ph

MrH

^ p<

coJ-I

fe A

2-07

47-7

2-61

52-3

3-16

56-9

3-74

61-5

4-39

43-1

208

47-8

2-62

52-4

3-17

3-75

61-6

4-40

43-2

2-09

47-9

2-63

52-5

3-18

57 1

3-76

61-7

4-42

43-3

2-10

2-64

52-6

3-20

57-2

3-78

61-8

4-44

43-4

2-11

48-1

2-66

52-7

3-21

573

3-80

61-9

4-46

43-5

212

48-2

2-67

52-8

3-22

57-4

3-81

4-47

43-6

213

48-3

2-68

52-9

3-23

57-5

3-82

62-1

4-48

43-7

2-14

48-4

2-70

3-25

57-6

3-84

62-2

4-50

43-8

2-16

48-5

2-71

53-1

3-26

57-7

3-85

62-3

4-52

43-9

217

48-6

2-72

53-2

3-27

57-8

3-87

62-4

4-53

2-18

48-7

2-73

53-3

3-28

57-9

3-88

62-5

4-55

44-1

2-19

48-8

2-74

53-4

3-29

3-90

62-6

4-56

44-2

2-20

48-9

2-75

•53-5

3-30

58-1

3-91

62-7

4-58

44-3

2-22

2-76

53-6

3-31

58-2

3-92

62-8

4-59

44-4

2-23

491

2-77

53-7

3-33

58-3

3-93

62-9

4-61

44-5

2-24

49-2

2-78

53-8

3-34

58-4

3-95

4-63

44-6

2-25

49-3

2-79

53-9

3-35

58-5

3-96

63-1

4-64

44-7

2-26

49-4

2-80

3-37

58-6

3-98

63-2

4-66

44-8

2-27

49-5

2-81

54-1

3-38

58-7

3-99

63-3

4-67

44-9

2-28

49-6

2-83

54-2

3-39

58-8

4-01

63-4

4-69

2-30

49-7

2-84

54-3

3-40

58-9

4-02

63-5

4-70

451

2-31

49-8

2-86

54-4

3-41

4-03

63-6

4-71

45-2

2-32

49-9

2-87

54-5

3-43

59-1

4-04

63-7

4-73

45-3

2-33

2-88

54-6

3-45

59-2

4-06

63-8

4-75

45-4

2-34

50-1

2-90

54-7

3-46

59-3

4-07

63-9

4-77

45-5

2-35

50-2

2-91

54-8

3-47

59-4

4-09

4-79

45-6

2-36

50-3

2-92

54-9

3-48

59-5

4-11

64-1

4-80

45-7

2-37

50-4

2-93

3-49

59-6

4-12

64-2

4-82

45-8

2-38

50-5

2-94

55-1

3-51

59-7

4-14

64-3

4-84

45-9

2-39

506

2-v»6

55-2

3-52

59-8

4-15

64-4

4-85

2-40

50-7

2-97

55-3

3-53

59-9

4-16

64-5

4-87

46-1

2-42

50-8

2-98

55-4

3-55

4-18

64-6

4-88

46-2

2-43

50-9

2-99

55-5

3-56

60-1

4-19

64-7

4-90

4r3-3

2-44

3-00

55-6

3-57

60-2

4-20

64-8

4-92

46-4

2-45

511

3-01

55-7

3-59

60-3

4-21

64-9

4-93

46-5

2-46

51-2

3-03

55-8

3-60

60-4

4-23

4-95

46-6

2-47

51-3

3-04

55-9

3-61

60-5

4-24

65-1

4-97

46-7

2-49

51-4

3-05

3-63

60-6

4-26

65-2

4-98

46-8

2-50

51-5

3-06

56-1

3-64

60-7

4-27

65-3

5-00

46-9

2-51

51-6

3-08

56-2

3-65

60-8

4-29

65-4

5-02

2-52

51-7

3-09

5(3-3

367

60-9

4-30

65-5

5-04

47-1

2-54

51-8

3-10

56-4

3-68

4-32

65-6

5-05

47-2

2-55

51-9

3-11

56.5

3-69

61-1

4-33

65-7

5-07

47-3

2-56

3-12

56-6

3-71

61-2

4-35

65-8

5-09

47-4

2-57

52-1

314

56-7

3-72

61-3

4-36

65-9

511

47-5

2-58

52-2

3-15

56-8

3-73

61-4

4-37

5-12

47-6

2-60

The Werner- Schmidt Process. — Ten grms. of milk are weighed into
a 50 c.c. stout glass tube, about 8 inches long, and 10 c.c. of strong
hydrochloric acid added. The tube is then placed in boiling water
until the contents become dark brown, which usually takes place in
about ten minutes. The tube is then cooled by immersing it in cold water,

MILK.

and 30 c.c. of ether added. The whole is well shaken and the ether
transferred to a tared flask by means of a pipette (or more easily by
closing the tube with a cork pierced by tubes similar to those used in a
wash-bottle, of suitable length) ; the shaking with ether is repeated three
times, the ether evaporated and the fat weighed. This process is par-
ticularly applicable when the milk has become sour.

Gottlieb has modified a method devised by Eose for the determina-
tion of fat by extraction with ether and petroleum ether from an alkaline
solution of milk ("Chem. Zeit." xxii. 632).

Ten c.c. of milk, 1 c.c. of 20 per cent ammonia, and 10 c.c. of 95
per cent alcohol are shaken in a glass tube about 40 cm. long. Twenty-
five c.c. of ether are added, and the tube is inverted several times.
Then 25 c.c. of petroleum ether are added, and the whole is well
agitated. After complete separation the ethereal layer is removed,
and the milk again extracted with ether and petroleum ether, and the
second extract mixed with the first. This process is again twice re-
peated. The solvent is evaporated and the fat dried at 100° and
weighed. The fat should be dissolved in a little petroleum ether, and
the small residue of non-fatty solids subtracted from the weight.

The essential point of this method is the complete mixing of each
solvent before the addition of the next.

Bell's Process. — This depends on the evaporation of 10 grms. of
the milk with constant stir-
ring, until a not too dry, pul-
verulent, residue is obtained.
This is repeatedly treated
with warm ether which is
filtered into a tared beaker,
the last traces of fat being
washed through the paper
with more ether, the ether
evaporated and the residue
weighed.

Centrifugal Separation
of the Fat. — In all processes
of this type, the milk is
whirled in a centrifugal ap-
paratus in closed tubes, after
being treated by a suitable
reagent. The two most use-
ful instruments are those of
Gerber and Letfmann-Beam,
which are modifications of a
process devised by Babcock.

Leffmann-Beam Process.

Fig 5.

Centrifugal apparatus.

H. Leffmann and W. Beam were the
originators of this method of fat-separation. The centrifugal apparatus
used was made by Beimling. This apparatus is made to hold four,
eight or twelve bottles according to the number required. In using it
take 15 c.c. of the sample of milk and pour it into the specially con-
structed 30 c.c. bottle which has a long graduated neck. Each division

FOOD AND DEUGS.

represents 0-10 per cent of milk-fat. Then add 3 c.c. of a mixture of
equal parts of amylic alcohol and fuming hydrochloric acid (sp. gr. 1'16).
Shake well and slowly add sulphuric acid (sp. gr. 1-835) shaking all
the time. The contents are now hot, and the casein which was pre-
viously separated, now completely dissolves, and the liquid becomes a
dark reddish-brown colour. Mix two measures of water with one of
strong sulphuric acid and add until the liquid reaches the zero mark
on the bottle neck. Rotate the bottle for two minutes in the centri-
fugal machine. If the sample of milk used is poor in fat, or is skimmed
milk, the rotation should be continued a minute or two longer. It will
be noticed on stopping the machine that there is a layer of fat in the
neck of the bottle, the amount of which can be ascertained from the
graduations. To do this accurately take a pair of dividers, the legs of
which should be placed at the upper and lower limits of the layer of
fat, allowance being made for the meniscus. Then move the dividers
until one point is on the zero mark of the scale. The percentage of
fat will be shown from the position of the other leg.

The amount of fat in cream can also be estimated by the Leffmann-
Beam process. Take about 2 c.c. of the sample, put in the bottle and
add 15 c.c. of water. Multiply the reading by 15*25 and divide by the
weight (in grammes) of the sample taken. If only one test is being
made, the arms of the machine should be balanced by either a duplicate
test, or a bottle filled with diluted sulphuric acid, placed in the carrier
opposite the one containing the sample.

Gerber Process. — Another form of the centrifugal apparatus is that
devised by N. Gerber. He directs in the instructions issued with the
instrument, that 10 c.c. of sulphuiic acid, specific gravity not less than
1"820 or more than 1*825, should be poured into one of the bottles.
Eleven c.c. of the milk to be tested is then added, and finally 1 c.c. of
amylic alcohol, care being taken to pour gently down the side of the
bottle. Insert a tightly fitting india-rubber stopper in the bottle neck

Fig. 0. — Bottles for fat separation in milk.

Fig. 7 — Babcock bottles.

and shake well until the contents are thoroughly mixed. Place the
bottle, while still hot, in the rotator, screw the top on, and, by pulling
the cat-gut string make the whole revolve as quickly as possible. It
will take about two or three minutes to separate the fat, after which

MILK.

I
I

the bottle should be plunged into water at 60" to 70° C. and the volume
of fat ascertained from the graduations. When the sample is skimmed
milk the rotation and immersion in warm water should be repeated
several times before reading off the amount of fat. The same treatment
would be used for condensed milk, which should be first diluted with

nine times its weight of water.

The following table gives the necessary amounts of milk and reagents
required in the centrifugal methods of fat separation : —

Babcock.

Lettniaim -
Beam.

Gerber.

Stokes.

15 cc
13-5 „
1-820 to
1-830
none
1-5 cc

Milk

Sulphuric acid (volume) .
Sulphuric acid (specific graviiy) .

Hydrochloric acid . .
Amylic alcohol ....

17-5 c.c.

17-5 „

1-831 to

1-834

none

15 c.c.
9 „
1-835

1-5 c.c*
15 „

11 c.c
10 „
1-820 to
1-825
none
1-0 cc

Wollny (" Milch. Zeit." 1900, 50) recommends the determination of
the refractive index of an ethereal solution of the fat obtained under
definite 'conditions, given refractive indices (or arbitrary readings on the
refractometer) corresponding to given fat values. The grave objection
to this process is that during the transference of the few drops of ether
solution to the refractometer, loss of ether may easily occur by evapora-
tion, and thus too high results be obtained.

There is a fairly constant relation between the specific gravity, total
solids, and fat in a sample of milk and Droop Eichmond and Hehner
(" Analyst," xiii. 32) have adopted the following formula from which
the amount of fat can be approximately calculated when the specific
gravity and solids are known : —

T = 1-164 F + 0-254 G,

when T is the total solids, . F the fat, and G the specific gravity.
Several modified formulae have been proposed but none remove the
calculation outside the limit of experimental error, so that they are
but slight improvements, if any, on the above. According to Allen,
Richmond now prefers : —

P = 1-2 F + 0-14 + 0-25 G.

The table on page 54 is based on Richmond and Hehner's formula.
The figures in the body of the table are the total solids.

Determination of the Milk Sugai' in Milk. — The only practical, and,
at the same time, accurate, methods of determining the amount of sugar
in milk are based on the optical rotatory power of that substance, or on
its power of reducing cupric oxide. Heat has a variable effect, according
to circumstances, on the optical rotation of milk sugar or lactose, but
does not affect its reducing power. Hence sugar cannot be very accu-
rately estimated by the polarimeter if the milk has been boiled.

(A) Polari7?ietric Determination. — It is first necessary to obtain a

FOOD AND DKUGS.

!M oi c<j (M (M CO CO M rc rc CO CC •<!»< -«i<

C0<M<MC<J(M<M(MCOCOCOCCCOCOeOCO-*

(M(M(NC<J(M(N(MCCCCC0CO00COCOeO

COrHCOeOOii-l'^C^OIMOOOi-l'rJtt-OSCQlCaCOCOCOQO

iM(M(M(M(M<M<MClCOeOeOCCCC0(5

O O r-l rH r-l

iHi-(i-H(M<M0<>(M(M(?q<M(NCOCOCOCOCO

eC«DODrH-^COOSCQ»OC»OCOv^C5<M-^t>OrQ»OOOtHCO

«o^^-c»o^^<^^«p»o«pt>CiOr-^(N-^^o?OQOC50^Hcc■*

000'HrHTHTHrHrHrHrH(J5(?Jlfl(jqCqClCq01COCOcbcO

rHiHi-lr-li-l(M(M(MIM<MC0(?5(MCOCCeO

OOOOOrHrHr-li-li-l

C<><M<M(N(N(M<M(MeO0O

o o o o o o

<N<MiM(M(M(MC<llMeO

o o o o o o o

<M<?Q(M'N(M(»'N(M

oooooooo

CT (M O^ <M (M (M ©»

TlH|>-w5(M«CO'OCQCCiC0!— (Tt<t>.OcO»000'- -^SD Os'cvJ ■»)'
0>OTHCp-5t(lpt^(»CSOCTCp-^«pi^000SiHC<):0M<«'

OSOSOOOOOOOOi— li—ii—li— ii— I?— (iHi— ((MO-ICToaer^

OSOiOOOOOOOOOi— li— li— IrHi— IrHi— fC>50105IC^'ri

»pc^(»qirH(Nep•^^pl^-cOw^^-l(^qcQlc«ot>xO'H(^^co
OSOCiCSOOOOOOOOi— li— li— li— li— ii— (1— t(>l(jci(jqcji

C50C»CiC50000000

»Q00Oct5CDG0r-lTt<l>.C5<M»OaD'-i'^;0C5(M»CiC~Oe0»O

w-^P«p^-ccOfH(^1co"rf<l:ol^-ccoTH(^qoplO?ot*C50l-l

CSC5050S050iOOOOOOOi— trHf— IrH-Hr— li-HiHC<l C<1

-*C-05(N»OC~000«DaOi-lTj1t>OCOlCODt-i-*cOOSC<jTtl
(NC«THspt-(»OrH(NM»C«p^-CiOrH(7l-^100I>050
0S050S0i05C5OOOOOOOOi— (i— ii— li— li— IrHiHi— liil

OCSCSSSOSOSOOOOOOOOOrH

OTH(N«p>po^-cpcpr-^?0'-Hlo<:oapc50lH?'- '^loccco

C»CSC50»OSC5050SOOOOOOOOi— 1»— ItH-HfHrHrH

COCSC5CiC5C50SO>CSOOOOOOOO>— <

t-o^'oooocotoosi-iTficci-iTjtt^cicococirH'^jtL'-a;

t^<pOrH0»-*»0«pt^qsO'-(CC'^»0?pQ0CiO'>1CC-*»C
ODODCSCSCSOSOSCSCSOSOOOOOOOOrli ,h,h^^

:OC»tH-^C^O»'M»CQOOCOeOOCOCOOOrH-fi|^Ci-MlOt~
«0t~-0SO'H<>>-^'P«0C0OOCTc0-^>0t^G0C5OC<lCC-*

aD<i>a)C5Csc5C50>C5C»C5ooooooootHTH>o^

-^fooocoeDQOrHTft^cs'Miocsi— l■^l:^0'M^.'^QO^^-^t-
lra«o(»csOr-(cc•^»f5«occc50'^l«I-^eo^>•cccsrH'^Ico
aodoaDoocscscscicscsoscsoooooooorHfHrH

OOODOOQOC30C5C50C50CiCiCiOOOOOOO'— I— I

"^ o lo o ip cp>coLooioo«poipp«:>o»ooipo«50

■^- X ClfM-MfMIMS^lCOCOfOeCCOeO

oyioadg oooooooooooo

MILK.

clear solution of the sugar. Wiley (Bulletin 13, United States Dept. of
Agriculture) has examined this subject exhaustively, and recommended
the use of an acid solution of mercuric nitrate to clarify the milk. This
has the advantage over acetic acid and lead acetate as recommended by
Schmoeger, since heat is required in the latter process. Various other
precipitants may be used, but as mercuric nitrate is the most satisfactory,
the others need not be discussed. Eichmond, however, has shown
(" Analyst," xxxv. 576) that traces of proteids are left in solution, and
he adds 5 per cent of phosphotungstic acid and 5 per cent of 50 per
cent H.^SO^. This ensures complete precipitation, and by allowing for
the increase in volume, the polarimetric reading gives the amount of
lactose at once.

Lactose is estimated by determining the optical activity of its solu-
tion. When applying this method to milk, first prepare a clear whey
free from other optically active substances. Wiley ("Amer. Chem.
J." 1884, 6, No. 5.) proves that precipitation by basic lead acetate does
not completely remove the laevorotatory protein matters ; he suggests
two alternative mercurial reagents. His method, which has been offi-
cially adopted in America, is as follows : —

{a) Acid Mercuric Nitrate. — Dissolve mercury in twice its weight
of nitric acid, specific gravity 142, and dilute with an equal volume
of water. 1 c.c. of this reagent is sufficient for the quantities of milk
mentioned below. Larger quantities may be used without affecting the
results of polarization.

{b) Mercuric Iodide with Acetic Acid. — Mix 33 -2 grms. of potas-
sium iodide with 13*5 grms. of mercuric chloride, 20 c.c. of glacial
acetic acid and 640 c.c. of water.

Estimation. — The milk should be kept at one temperature and its
specific gravity determined.

The quantities of milk measured for polarization vary with the
specific gravity of the milk and also with the polariscope used. The
quantity to be measured will be found in any case in the accompanying
table : —

Specific gravity.

Volume of Milk to be used.

For Polariscopes of

which the Sucrose

Normal Weight

is 1619 grammes.

is 26-048 grammes.

c.c.

1-024

60-0

64-4

1-026

59-9

64-3

1-028

59-8

64-15

1-030

59-7

64-0

1-032

59-6

63-9

1-034

59-5 68-8

1-085

59-35 63-7

56 FOOD AND DEUGS.

Take a flask graduated at 102-4 g.c. for a Laurent or 102-6 c.c. for
a Ventzke polariscope and place in it the quantity of milk given in the
table. Add 1 c.c. of mercuric nitrate solution or 30 c.c. of mercuric
iodide solution (an excess of those reagents does no harm), fill to the
mark, shake well, filter through a dry filter and polarize. There is no
necessity to heat before polarizing. If a 2C0 mm. tube is used, divide
the polariscope reading by 3 when the sucrose normal weight for the
instrument is 16-19 grms. or by 2 when the normal weight for the in-
strument is 26-048. The lactose normal weight to read 100° on the
sugar scale for Laurent instruments is 20-496 grms. and for Ventzke
instruments 32'975 grms.

Wiley and Ewell in a later paper prefer to use the method of double
dilution, which does away with the necessity for allowing for the
volume of precipitated matter. They add 4 c.c. of the acid mercuric
nitrate solution to two portions of the milk and dilute one of these to
100 c.c. and the other to 200 c.c. The liquids are filtered and the optical
rotation taken in a 400 mm. tube. The reading of the dilute solution is
multiplied by two, and the product subtracted from the reading of the
stronger solution, the difference being called a. Then, the reading of
the stronger solution - 2 a, divided by four (assuming a Soleil- Ventzke
instrument is used) gives the percentage of lactose. A little considera-
tion will show that this is a mathematically correct proportion.

Rupp recommends treating 50 c.c. of milk with 25 c.c. of lead acetate
solution (10 per cent), heating, filtering, and when cool making up to
100 c.c. Each angular degree of rotation in 100 mm. tube corresponds
to 0*205 grm. of milk sugar. In the author's experience 0-195 is the
more correct value per 1°. A simpler process is described by Vieth
("Analyst," xiii. 63) but in the author's opinion is not so accurate.

(B) Volumetric determination.

Milk sugar may be determined by reduction of Fehling's solution.
In order to prepare the milk for the action of Fehling's solution, 25
grms. are mixed with 0*5 c.c. of 30 per cent acetic acid solution and
the mixture well shaken and 100 c.c. of boiling water is added ; 25 c.c.
of recently made cream of alumina are then added (cream of alumina
is prepared by adding a slight excess of NH3 to a saturated alum solu-
tion, and then adding more alum till faintly acid), the whole well
shaken and allowed to stand for fifteen minutes. The liquid is then
filtered through a moistened filter paper and the filtrate and w^ashings
made up to 250 c.c. The liquid must be quite clear. The liquid is
now titrated in the usual manner (see p. 123) against 10 c.c. of Feh-
ling's solution, or the precipitated copper is weighed as cupric oxide.
If the gravimetric process be used, 25 c.c. of the milk solution should
be boiled with 30 c.c. of Fehling's solution and 50 c.c. of water.
For the purposes of calculation, it may be taken that 0-067 grm. of lactose
reduces 10 c.c. of Fehling's solution, or that the weight of cupric oxide
X 0*6024 will represent anhydrous lactose. If more accurate results
be required the following table gives the amount of lactose correspond-
ing to given weights of metallic copper (which can be calculated from
the cupric oxide) : —

MILK. 57

Copper. Lfu-tose. Copjjer. Lactose. Copper. Laetose

120 = 8G-4 220 = 1G1-1> 310 = 232-2'

130 = 93-8 230 ^ 169-4 320 = 2400

140 = 101-3 240 = 176-9 330 = 247-7

150 = 108-8 250 = 184-8 340 = 255-1

160 = 116-4 260 = 192-5 350 = 263-9

170 = 123-9 270 = 200-3 360 = 272-1

180 = 131-6 280 = 208-3 370 = 280-5

190 = 139-3 290 = 216-3 380 = 289-1

200 = 146-9 300 = 228-3 390 = 297-7

210 = 154-5 400 = 306-3

Intermediate results can be interjDolated.

Determination of the Proteids of Milk. — A determination of the
total nitrogen by Kjeldahl's process (N x 6-33) gives a sufficiently ac-
curate result for the proteids in milk, as practically the whole of the
nitrogen in milk exists in the form of proteids. The determination
should be made in the following manner : —

Three c.c. of the milk are heated in a hard Jena glass round-
bottomed flask with 20 c.c. of strong sulphuric acid, with the addition
of a small globule of mercury, as recommended by Dyer. The liquid
is allowed to boil briskly and in ten minutes 10 grms. of potassium
sulphate are added, and the boiling is continued until the contents of
the flask are clear and nearly colourless. The contents of the flask
are allowed to cool and transferred to a large distilling flask (round-
bottomed Jena glass flasks are preferable). The flask having a cork
with two holes in it, is connected with any convenient condenser and
has a tap-funnel attached through the second hole. Though this
caustic soda solution is cautiously added in slight excess (as indicated
by means of the addition of a drop of phenol-phthalein solution to the
soda solution, which is run in until a pink colour is permanent, when
a further few c.c. should be added). A little pumice stone is added to
prevent bumping. If mercury has been used, a little sodium sulphide
should be added to the distilling flask in order to decompose any com-
pound of mercury and nitrogen, but the mercury is not essential and
may be omitted. The distillation is carefully conducted, the distillate
being received in the usual manner into an excess of decinormal sul-
phuric acid (80 c.c). Each c.c. of acid neutralized, as determined by
titration, corresponds to 0'0014 grm. of nitrogen or 0' 00886 grm. of
milk proteids. A blank experiment is necessary to allow for the traces
of nitrogen present in the reagents. A tin condensing tube is better
than glass, as there may be some action of the steam on the glass.

(For further details of the Kjeldahl process see " Trans. Chem. Soc."
1895, 118).

Kichmond and Bosely ("Analyst" xviii. 172) recommend the fol-
lowing method : 10 grms. of milk are diluted with about 200 c.c. of
water, and rendered faintly alkaline (to phenol-phthalein) with dilute
caustic soda solution. From 2 to 2 '5 c.c. of a 6 per cent solution of
copper sulphate are then added, the mixture shaken, and then allowed
to settle. The precipitate is washed five times by decantation, the
washings being poured through a tared filter paper. It is then trans-
ferred to the filter, washed once or twice more with water, dried for a
short time in a water oven, extracted with ether to remove traces of fat

58 FOOD AND DEUGS.

and dried at 130°, and weighed. It is afterwards burnt and the ash
deducted from the original weight. The result is the sum of the casein
and albumen. In practice, the author prefers Muter's suggestion of
not drying the precipitate before washing with ether, but of washing it
with absolute alcohol and then directly afterwards w^ith ether. Eupp
recommends using 10 grms. of milk and 100 to 150 c.c. of water, to
which is added 15 c.c. of a 6 per cent solution of copper sulphate ;
7 c.c. of a 1*5 c.c. of caustic soda are then added and the precipitate
of proteids, copper hydroxide and a little fat is transferred to a tared
filter paper, and washed successively with water (several times) alcohol
and ether. It is then dried and weighed and the ash deducted from
the weight. The disadvantage of this method is that a relatively large
amount of copper is precipitated, and the ash is very high as compared
with the amount of precipitate.

Sebelien separates the casein and albumin in the following manner :
About 10 grms. of milk are mixed with double its volume of a saturated
solution of magnesium sulphate, and then crystals of the salt added so
long as they dissolve on shaking. The liquid is allowed to stand for
three to four hours and the precipitate is washed with a saturated solu-
tion of magnesium sulphate. The precipitate with the filter paper is
treated with 30 c.c. of strong sulphuric acid and the nitrogen deter-
mined by Kjeldahl's process. The nitrogen x 6*33 may be taken as
representing the casein. The albumin, which is in the filtrate is pre-
cipitated by a solution of phosphotungstic acid, and the nitrogen in the
precipitate determined. A simple and comparatively accurate process
for the estimation of the casein and albumin is to dilute 10 to 20 c.c.
of milk with ten times its volume of water and acidify with a drop or
two of acetic acid. The mixture is now warmed to 40° C. and the re-
sulting precipitate of casein transferred to a tared filter, washed with
water, alcohol, and ether, dried and weighed. The filtrate is further
acidified with a little acetic acid and heated to boiling-point. The pre-
cipitate of albumin formed is collected on a tared filter, washed with
water, dried and weighed.' Greater accuracy is obtained by deter-
mining the nitrogen in the precipitates, using the factor N x 6*37 =
casein, and N x 6*73 = albumin.

Calculation of Adulteration.

The calculation of the amount of adulteration of milk is necessarily
based on an arbitrary standard for the original milk.

For legal purposes the solids other than fat must be present to the
extent of at least 8'5 per cent (assuming the fat to be only 3 per cent).
On the assumption of this figure as a standard, the formula

_ S X 100
W = 100-^^

will give the percentage of added water (i.e. calculated on the sample)
when W is that percentage and S is the amount of non-fatty solids in
the sample.

In cases where the adulteration is an abstraction of fat, the ap-

MILK. 59

proximate percentage abstracted, on the same assumption as above, is
calculated from the formula

100 (F, - F,)

when P is the percentage (calculated on the total fat normally present,
i.e. the legal minimum of 3 per cent) of fat abstracted, Fj is the normal
amount (i.e. 3 per cent) and F^ is the amount in the sample.

Poisonous Milk.

V. C. Vaughan named the poisonous ptomaine which he discovered
in stale milk, ice-creams and cheese, tyrotoxicon. It crystallizes in
needles which gradually decompose when exposed to moist air. Its
odour is similar to that of stale cheese, and it has a "dry " taste. It
is soluble in water, alcohol, and chloroform, but insoluble in ether when
pure, though it dissolves if other impurities are present.

Vaughan found that tyrotoxicon was exceedingly poisonous both to
man and the lower animals. The smallest portion placed on a child's
tongue caused symptoms identical with those of cholera infantum, viz.
sickness and diarrhoea. When ten drops of a solution of tyrotoxicon
taken from milk three months old, were given to a young dog, it caused
frothing at the mouth, vomiting, diarrhoea and muscular spasms.
Cats exhibited the same symptonjs. The mucous membrane of the
stomach showed no inflammation after death, being white and soft.

When a strong solution of tyrotoxicon is evaporated with some
platinic chloride on a water bath, a violent explosion takes place as
soon as the whole of the alcohol has evaporated.

Tyrotoxicon does not precipitate with most of the general reagents
for alkaloids.

Tyrotoxicon forms a potassium derivative, which crystallizes in six-
sided plates, soluble in alcohol from which it is precipitated by ether.

Vaughan considers this compound is diazobenzene potassoxide
CgHgNgOK, and believes that tyrotoxicon if not identical with diazo-
benzene butyrate is closely related to it.

He suggests the following test for detecting tyrotoxicon in milk
and cheese : —

Neutralize the filtrate from the curdled milk or the filtered cold
water extract of cheese, with sodium carbonate, place in a separator
and agitate with its own volume of ether. Allow the mixture to stand
for twenty-four hours, or until separation has taken place, then
leave the ethereal layer to .evaporate spontaneously in an open dish.
Dissolve the residue in water, again agitate the liquid with ether, separ-
ate the ethereal layer, and allow to evaporate as before. There should
not be repeated extractions with ether, as the tyrotoxicon becomes less
easy to dissolve the purer it is. Dissolve the residue in a few drops
of distilled water and examine the solution as follows : —

(a) Place a drop of the liquid on porcelain with a few drops of a
freshly prepared mixture of phenol and concentrated sulphuric acid,
free from nitrous compounds. If tyrotoxicon is present there will be a

60 FOOD AND DKUGS.

coloration vaiying from yellow to orange-red and finally becoming
violet.

(b) Add a concentrated solution of caustic potash to the remainder
of the solution and evaporate to dryness on the water bath. If tyro-
toxicon is present, diazobenzene-potassoxide will be formed, and can
be recognized by its crystalline form and green colour produced on
the addition of a mixture of phenol and strong sulphuric acid.

An acid solution of tyrotoxicon prepared from milk gives with auric
chloride a golden-yellow precipitate, but the gold salt forms very slowly
from some milks, probably on account of the presence of other organic
matter.

Preservatives in Milk,

It is very frequently nece"^sary to examine samples of milk for the
presence of preservatives. Of these the principal are boric acid and
formic aldehyde. Other preservatives are used from time to time,
but the efficiency of the two above-mentioned have caused them to re-
place other preservatives almost entirely." M. Blyth (" Analyst," xxvi.
149) claims that the presence of preservatives in milk is accurately
indicated by the following method.

Measure 10 c.c. of each milk into clean wide test tubes. Measure
10 c.c. of a sterile milk known to be free from preservatives into a test
tube (these control tubes can be kept ready for use). Add to each milk
and to the control 2 c.c. of a very "strong, slightly alkaline solution of
litmus. Now examine all the tubes, and if any of them are not the
same shade of blue as the control tube, drop in, drop by drop, a half
normal solution of sodium hydrate until the correct shade of blue is
obtained. This will be found unnecessary in the case of most milks,
and will only be requisite when the milks are two or three days old ;
this process must then be done very carefully. Plug all the tubes with
cotton wool, and heat them in a water bath, kept at a temperature of
80^ C. for ten minutes. Allow the tubes to cool, and inoculate each,
including the control, w4th half a c.c. of sour milk in water (half c.c.
milk to 100 c.c. water). Shake the tubes well. Now let the tubes
stand for twenty-four hours at any temperature between 15° C. and 22°
C, and then examine. If the control tube be not white, or nearly so,
they must be allowed to stand for a longer period. Those tubes which
contain preservatives will remain blue or pink, while the tubes which
contain no preservatives will behave in the same w^ay as the control
tubes, becoming quite white. The length of time the blue or pink
colour takes to become white depends upon the quantity of preservative
present in the sample. The quantities of the more common preserva-
tives which can be detected with certainty by this method, are 0"005
per cent of borax, boracic acid, or mixtures of these substances, 0*05
per cent of salicylic acid, and O'OOOS of formic aldehyde, quantities
very much smaller than are ever found, or which w'ould be of any value
in commercial milks. Having selected by this method those samples
which contain preservatives, the nature of these must be determined by
the ordinary methods.

MILK. 61

Boric acid, borates, and a mixture of boric acid and borax are ex-
tensively used for the presei-vation of milk. Boron compounds can be
detected in the following manner: Take not less than 10 grms. of the
sample of milk and evaporate to dryness. Ignite the solid residue and
add to the ash sufficient hydrochloric acid to render the whole slightlv,
yet distinctly, acid to litmus. Place a small slip of turmeric paper
in the capsule in such a way that only part of it can be wetted,
then evaporate to dryness at a temperature of 100° C. If that part of
the turmeric paper placed in the liquid has become a definite brownish-
red colour, owing to the formation of rosocyanin, boron compounds are
present. A drop of caustic soda on the paper will produce a variety of
colours, particularly green and purple, whereas hydrochloric acid will
bring back the original red colour, which will change to gi'een and blue
on the addition of an excess of alkali.

It is not an easy matter to accurately determine the amount of
boron compounds in milk, especially in a small quantity. R. T.
Thompson (" Analyst," xviii. 184) observes that free boric acid may be
titrated somewhat accurately by caustic alkali and phenol-phthalein, that
is, if. the liquid contains 30 per cent of glycerine. The neutral point
then corresponds to the formation of NaBO^,. Thus each c.c. of deci-
normal alkali required represents 0'003o grm. of boric anhydride,
B0O3 ; 0-0062 of crystallized boric acid, H3BO3 ; 0-00505 of anhydrous
borax, Na.,B^O-; or 0-00955 grm. of crystallized borax Na^B^O^H-
IOH.,0. It is better to titrate the solution against a known weight of
crystallized boric acid, than to assume the neutralizing power of the
standard alkali to be correct : 0-310 grm. should neutralize 50 c.c. of
decinormal alkali. Whsn the foregoing method is applied to the de-
termination of boric acid in milk, L. de Koningh recommends the
addition of 1 c.c. of a strong solution of caustic soda to 10 ginns. of
the milk, evaporation of the liquid and ignition of the residue.
The ash should then be boiled w^ith water, the residue again ignited
and again boiled with water. The two solutions will contain all the
borates present. They should be mixed together, a drop or two of
methyl-orange added, then decinormal sulphuric acid carefully dropped
in until the liquid becomes slightly pink after stirring. After boiling
the liquid for a minute or two to expel carbon dioxide, it is cooled and
half its amount of glycerine added. A few drops of phenol-phthalein
solution is next added and the liquid titrated with decinormal caustic
soda until it becomes pink in colour. This method is more reliable
for large amounts of boron compounds than for small quantities.

Another method for determining boron compounds in milk is to con-
vert them into volatile methyl borate, then to decompose this compound
after distilling with lime or "other base. This method, apparently de-
vised by Eosenbladt and Gooch, has been modified by Penfield and
Sperry, Gilbert, Cassal, and Hehner, the following being the preferable
method (C. E. Cassal, "Analyst," xv. 230; and O. Hehner, "Analyst,"
XVI. 141) :—

Add to 50 grms. of cream or 100 grms. of milk caustic soda to
render alkaline, evaporate to dryness and ignite the residue. Reduce
the ash, not necessarily white, to a powder and transfer by means of a

62 POOD AND DRUGS.

little methyl alcohol and a few drops of water, to a conical flask of 200
to 300 c.c. capacity. Insert a caoutchouc stopper provided with a
tapped funnel and delivery tube. Add acetic acid to make the contents
of the flask acid, then add 5 c.c. of methyl alcohol. The flask should
be connected with a condenser by means of a flexible joint (to allow
the contents to be occasionally shaken). Place the liquid on an oil-
bath, and distil almost to dryness ; 5 c.c. of methyl alcohol is again
added and the distillation again continued. Ten such treatments, with
distillation, are quite sufficient, and in some cases even less suffice.
Test the residue in the flask with turmeric paper to be certain that the
boric acid has completely volatilized. Cassal recommends the addition
of a few 'drops of water before distillation, as it helps greatly in the
operation. He also suggests several distillations of small quantities in
preference to once adding a large volume of methyl alcohol, better re-
sults being obtained. Mix the distillates which contain methyl
borate, in a vessel containing a known weight of freshly burnt lime.
The amount of boric acid can be calculated after evaporating to dry-
ness and burning the residue, by observing the increase in weight.
This is not the most satisfactory of methods as there are many diffi-
culties and sources of error to contend with. Hehner prefers to use a
solution of crystallized sodium phosphate instead of caustic lime.
The methyl borate becomes decomposed after evaporating with this
reagent. If boric acid is absent, the residue left after igniting will be
solely sodium pyrophosphate Na4P207, but if boric acid or methyl
borate are present, the reaction brought about will form sodium meta-
phosphate and biborate Na4P.20- + 2B.p^ = 2NaP03 + Na.^B^O^.

Thus 0-133 grm. of sodium pyrophosphate produced by the ignition
of 0'332 grm. of crystallized sodium phosphate Na.^HPO^ + IOH.,0
will react with and fix 0*070 grm. of B^Og, which represents 0'124 grm.
of crystallized boric acid, or 0*191 grm. of crystallized borax. When
performing the test it is preferable to use a solution of about 2 per cent
strength of sodium phosphate instead of a solution of exactly known
strength. Take e.g. 20 c.c. of this solution and add to the distillate
containing methyl borate. Take an equal quantity of the same sol-
ution, measured with the same pipette, evaporate separately and ignite
both residues. Weigh the two ignited residues and the difference in
weight represents the B^Og of the sample. Great caution must be
exercised in heating the residue after evaporation, to prevent any loss,
and ignition should be carried on in a covered platinum dish, the
temperature being gradually raised until the residue fuses. O. Hehner
prefers to collect the distillate containing methyl borate in a receiver
containing caustic soda, evaporates, adds dilute mineral acid until abso-
lutely neutral to methyl orange, then glycerine and phenol-phthalein,
and lastly titrates with standard caustic soda to determine the boric
acid. A. E. Tankard has proved by experiments that the results are
the same as those obtained by evaporation of the distillate with sodium
phosphate and weighing the ignited residue.

Cassal and Gerrans ("British Food Journal," 4, 210) describe the
following process for the determination of boric acid : —

From 15 to 20 grms. of the sample, such as milk, is made distinctly

MILK. 63

alkaline with a saturated solution of Ba(OH)._, in a platinum dish, evapor-
ated to dryness, well charred, broken up, made just acid with HCl and
extracted with successive quantities of hot water, the filtrates being
mixed in a 100 c.c. tlask. The filter and contents are transferred to
the platinum dish, again made distinctly alkaline with Ba(OH).„ and
carefully ignited. The ash is dissolved in a little HCl (1 : 3), the" solu-
tion and washings added to the first solutions in the flask, and the
whole made up to exactly 100 c.c. Ten c.c. of this ash solution is then
pipetted on to 10 to 15 grms. of pure sand, the mixture is made alkaline
with Ba(OH)._, solution and evaporated to dryness with occasional stir-
ring. When dry it is made just acid with HCl (1 : 3), when 2 c.c. of
saturated solution of oxalic acid and 2 c.c. of an alcoholic solution of
curcumin (1 grm. per litre) are mixed in. The dish is then placed on
a water bath, covered with a funnel, the stem of which is connected
with a set of potash bulbs charged with Ba(OH)., solution. Air is
then gently aspirated through the apparatus until the mass in the
dish is dry. An additional 1 c.c. of curcumin solution is then added,
well mixed in, and the mass again dried. The dry mass is then ex-
tracted with small successive quantities of alcohol or methylated spirit,
the solutions obtained being filtered into a flask. "When the sand
mixture is freed from colour the liquid in the potash bulbs is poured
upon it and dried, care being taken that it is alkaline with Ba(OH).^.
The mixture is treated as before with HCl, oxalic acid, and curcumin
solutions, and the processes of evaporation and alcoholic extraction are
repeated. The further yield of coloured alcohol is added to that it first
obtained.

A standard colour is prepared by using 10 c.c. of boric acid solu-
tion (1 c.c. = 0*1 Mgm. B2O3) in precisely the same manner, the
coloured alcoholic extracts being made up to 200 c.c. By comparing
the depth of colour given by the ash solution extracts with this standard,
the amount of boric acid in the quantity of solution operated on may
be determined, and from this, the amount in the original sample cal-
culated.

Formaldehyde is probably by far the most efficient of milk preser-
vatives. It is usually employed in the form of a solution containing
40 per cent of formic aldehyde, H . CHO. The following are useful
methods for the detection of formic aldehyde in milk : —

S. Eideal ("Analyst," xx. 158) considers that Schiff's reagent is a
delicate test for formaldehyde in milk, if the solution used is slightly
acid. To prepare Schiff's reagent mix 40 c.c. of a 0-5 per cent solution
of fuchsine with 250 c.c. of distilled water, add 10 c.c. of sodium bi-
sulphite solution of 1-375 specific gravity, then 10 c.c. of pure strong
sulphuric acid. Allow the mixture to stand until it becomes colourless.
Another method is to add sufficient of a solution of sulphurous acid to
take away the colour of the fuchsine solution. If too large a quantity
of sulphurous acid is added it will be impossible to find traces of
formaldehyde. This test is useful as a confirmatory reaction and
can be applied to milk as described by Richmond and Boseley
("Analyst," xx. 155). Sulphuric acid is added in small quantities to
precipitate the casein, the liquid filtered and a little of Schiff's reagent

64 FOOD AND DEUGS.

is added to the filtrate. The amount of formaldehyde can be some-
what roughly estimated by the intensity of the red colour. No distilla-
tion is required in this test, which is an advantage. Richmond and
Boseley found that aqueous solutions of milk sugar on the addition of
Schitf s reagent give no coloration even when boiled with dilute sul-
phuric acid, though it has been stated that under some conditions, not
defined, a red colour appears.

O. Hehner (" Analyst," xxi. 94) makes use of a test for detecting
formalin in milk, the chief feature of which is the addition of strong
sulphuric acid to the milk. If mere traces of formaldehyde are
present, the liquid becomes a violet-blue colour. Richmond and
Boseley apply the test by adding an equal measure of water to the milk
and using sulphuric acid of 90 to 94 per cent strength. By using
the test in this manner, its delicacy is considerably increased. If in
200,000 parts of milk there is one of formalin, a violet-blue colour will
be produced at the junction of the two layers, which will remain per-
manent for some days. If formaldehyde is absent, a slight greenish
colour may be observed, and lower down in the acid after some hours
a brownish-red colour becomes noticeable. This, however, cannot be
mistaken for the blue colour denoting the presence of formaldehyde.
Hehner's reaction is simple and delicate. It is not produced by acet-
aldehyde. Richmond and Boseley state that it is not practicable with
large quantities of formaldehyde — 0 5 per cent would, for example,
give no blue coloration. Richmond and Boseley attribute Hehner's
reactions to the proteids of milk, as they find that egg-albumin and
peptone give the reaction, whereas gelatine does not. Hehner, however,
could not obtain a reaction with a solution of peptone, and only the
very slightest response from egg-albumin, which he thought was due
to some impurity rather than to the albumin itself. He was of the
same opinion concerning the gelatine. N. Leonard (" Analyst," xxi.
157) states that Hehner's reaction can easily be obtained if commercial
sulphuric acid is employed, but it fails altogether when pure redistilled
acid is used. When ferric chloride or platinic chloride was mixed
with pure sulphuric acid, it was found that the milk containing for-
maldehyde became violets blue in colour. The pure acid contained no
trace of iron whereas it was found that the commercial acid did.
Leonard deduces from this, that a feeble oxidizing agent must be pre-
sent for the production of Hehner's reaction. The addition of ferric
chloride in considerable quantities does not improve the test. A trace
of ferric chloride, however, renders the reaction more distinct. Hehner
confirms this statement of Leonard.

O. Hehner ( "Analyst," xxi. 94) describes another test which is only
useful for testing the presence of a small amount of formaldehyde.
This method is as follows : Add one drop of a dilute aqueous solu-
tion of phenol to the distillate from a sample of milk, or other substance.
Pour the mixture upon some strong sulphuric acid contained in a test
tube and it will be noticed that where they meet, a bright crimson
colour appears if one part of formaldehyde in 200,000 be present.
If there is a larger proportion present a milky-white zone above tho
crimson tinge appears. An orange-yellow colour denotes acataldehvd .

MILK. 66

This reaction will not be successful unless carried out exactly as de-
scribed. It is important that only a trace of phenol should be used.

Trillat (" Compt. Rend." cxvi. 891) suggests the following test for
formaldehyde. Mix the solution containing the formaldehyde with
a solution of dimethylaniline in slight excess of sulphuric acid ; shake
well together. Heat the liquid for half an hour on the water bath,
make alkaline and boil until there is no odour of dimethylaniline.
Filter, and moisten the filter paper with acetic acid. A sprink-
ling of some powdered lead oxide will produce a blue colour if
formaldehyde is present. The blue colour, which is not stable, is due
to the formation of tetramethyl-diamido-diphenylmethane. Another
test depends on the fact that a white precipitate is formed from a
mixture of a solution of formaldehyde and 0-3 per cent solution of
aniline.. This precipitate is anhydro-formaldehyde-aniline. It can be
weighed and the amount of formaldehyde present ascertained. Acet-
aldehyde also gives a precipitate. As the precipitate dissolves in hot
water, and reappears on cooling, the test must be carried out in the
cold. The precipitate given by acetaldehyde is more soluble than that
given by formaldehyde. Trillat states that formaldehyde cannot always
be detected in preserved foods after some lapse of time as condensation-
products are formed. Richmond and Boseley are of the same opinion
as Trillat, but they state that formaldehyde can always be detected in
milk by this test unless the milk is curdled. If it be desired to weigh
the precipitate of anhydro-formaldehyde-aniline it should be allowed to
stand for forty eight hours, filtered off, dried at 40" and weighed. Its
formula is Cj-HgNCH^, and 100 parts are equivalent to 28*5 parts of
formaldehyde.

Hydrochloric Acid Test. — Commercial hydrochloric acid (sp. gr.
1-2) containing 0*2 per cent of ferric chloride per litre is the reagent
used. Add to 10 c.c. of milk, contained in a porcelain vessel, an
equal quantity of the acid reagent. Slowly heat over a naked flame al-
most to boiling, shaking the vessel in order to break up the curd.
A violet coloration denotes formaldehyde, varying in depth according
to the amount present. If formaldehyde is absent the solution gradu-
ally turns brown. One part of formaldehyde in 250,000 parts of milk
can easily be detected, before the milk turns sour, by this test. When
this occurs the limit of delicacy is 1 part in 50,000. Various alde-
hydes in milk give colour reactions when subjected to the above
treatment, but formaldehyde alone gives the violet coloration, which
is perfectly easy to distinguish.

Confirmatory Tests with Distilled Milk. — To confirm the above
tests distil 100 to 200 c.c. of the milk and use the first 20 c.c. of the
distillate for testing purposes.

1. Add a drop of SchilFs reagent to a few drops of distillate in a
test tube. If aldehyde is present a pink coloration will soon appear,
deepening on standing.

2. Add a few drops of a 1 per cent aqueous solution of resorcin or
phenol to 5 c.c. of the milk distillate. A crimson colour denotes
formaldehyde and not other aldehydes.

3. To a small amount of milk distillate (slightly acidified with
VOL. I. 5

66 FOOD AND DRUGS.

sulphuric acid to fix any free ammonia before distillation) add a few
drops of Nessler's reagent. Traces of formaldehyde give a yellow
colour, whilst larger amounts produce a darker colour on standing, and
forms a grey precipitate.

Determination of Formaldehyde in Milk. — The determination of
formaldehyde is a matter of considerable difficulty when it is present
in such minute amounts. Add 1 c.c. of 1 : 3 sulphuric acid to 100
c.c. of milk and distil in a 500 c.c. Kjeldahl flask. Use a low
circular evaporating burner to avoid frothing. Smith states that the
first 20 c.c. of the distillate or one-fifth of the original quantity contain
almost one-third of the total formaldehyde. Collect 20 c.c. of the
distillate and use the following potassium cyanide method for the
determination of formaldehyde. To 10 c.c. of tenth-normal silver
nitrate add 6 drops of 50 per cent nitric acid, using a 50 c.c. flask.
Then add 10 c.c. of a solution of potassium cyanide containing 3'1
grms. of KCN in 500 c.c. of water making up to the 50 c.c. mark. Shake
well, filter and titrate 25 c.c. of the filtrate with tenth-normal am-
monium sulphocyanate using ferric chloride as an indicator. Take
another portion of 10 c.c. of tenth-normal silver nitrate and acidify
with nitric acid, add 10 c.c. of the potassium cyanide solution to which
the above 20 c.c. of formaldehyde distillate has already been added.
Make up the whole to 50 c.c, filter, and titrate 25 c.c. of the filtrate
with decinormal ammonium sulphocyanate for the excess of silver as
before.

To calculate the amount of potassium cyanide used up by the
formaldehyde, in terms of decinormal ammonium sulphocyanate,
multiply the difference between the two results by two, and the total
amount of formaldehyde can be calculated by multiplying the amount
found to be present in the 20 c.c. of distillate by three.

The formula GH.O + KCN = KOCH,CN shows the reaction
which takes place between the formaldehyde and the potassium cyanide,
from which the formaldehyde is easily calculated.

When milk is preserved with formic aldehyde it usually contains
•0002 to -006 per cent. The Government Laboratory consider -001
per cent as the maximum allowable, but owing to the fact that the
Government chemists examine the milk late, much of the aldehyde
may have disappeared. It must be estimated as quickly as possible,
as micro-organisms destroy the aldehyde.

The following is a useful method : —

Blow a bulb of about 50 c.c. capacity on a piece of soft glass tubing
with a quarter-inch bore. Draw out one end of the glass tube close to
the bulb, into a capillary tube, and turn at right angles to the bulb.
Turn the tube at the other side of the bulb at right angles to the
bulb. Pour 10 c.c. of the milk sample into the bulb and make slightly
acid, if the milk is not already sour. Seal up the end of the capillary
tube and place the bulb in a paraflSn bath taking care that it is com-
pletely immersed. Connect the open tube with a series of bulbs
each containing 5 c.c. of water, a very short rubber connexion being
used. Place the bulbs in a bath containing cold running water. Heat
the paraffin to 120° C. and distil the milk almost to dryness. To pre-

MILK. 67

vent frothing up the tube, warm it with a burner. As soon as the
milk is nearly dry, break the capillary tube at the end, and pass
a slow steady current of air through the apparatus by means of a
water pump. Raise the temperature of the parafi&n until it reaches
200° C. then keep it at that temperature for at least fifteen minutes.
Disconnect the bulbs and test the second bulb for formic aldehyde.
Unless the original solution contained a considerable amount, none will
be found. If there is any present, determine the amount, then test the
third bulb. Wash out the first bulb with distilled water, see that the
liquid does not exceed 20 c.c, add two drops of litmus and a few
drops of decinormal sodium carbonate solution to neutralize the free
acid present. The formaldehyde can then be determined by the
cyanide method above described.

In 1909 a report was made to the Local Government Board by Dr.
J. M. Hamill on the " Use of Preservatives in Cream ".

The use of preservatives is of importance in the cream -trade, and
many convictions under the Sale of Food and Drugs Acts have been
recorded when the amount of boron preservative used has been above
that recommended by the 1901 Departmental Committee (0-25 per
cent, expressed as boric acid), which added that boron presei-vative
only should be employed. Dr. Hamill in his report first deals with
the effect of boric acid on health, the general conclusion being that
there is a preponderance of opinion that boron compounds cannot
safely be regarded as incapable of exerting a deleterious action upon
health. The use of cream " thickeners," such as gelatin, starch paste,
and " sucrate of lime," is commended to the analyst's attention ; the
last-named compound being openly sold by wholesalers for admixture
with cream. In the section on the use of preservatives, Dr. Hamill
remarks that " apart from preservative power the qualities especially
requisite in such a preparation are ready solubility and freedom from
any objectionable taste ". Mixtures of boric acid and borax in such
proportions as to produce a compound as nearly neutral as possible are
the chief commercial preservatives. Most cream-preservatives contain,
in addition, a small quantity of saccharine to mask incipient sourness.
The recommendation of the 1901 Committee is sometimes infringed
by the addition of sodium salicylate or sodium benzoate to boron
preservatives, the makers relying upon the presence of these organic
preservatives being overlooked when the boric-acid content is found
to be within safe limits. Formalin has proved to be unsuitable as a
cream-preservative. Dr. Hamill has been informed that sodium
fluoride is used in some cases, and he refers to it as a "dangerous
substance ". The composition of cream-preservatives examined are
represented in the table on page 68.

The directions for use are generally such as will ensure approxi-
mately the presence of 025 per cent of boric acid in the cream.
Cream sold in jars or in small bulk generally contains preservatives,
more being used in summer than in the cooler months. Imported
cream practically always contains preservatives. Hydrogen peroxide
differs from ordinary preservatives in that when properly used little or
none of it remains in the cream. One firm who use it add 100 c.c. of

FOOD AND DRUGS.

a 3 per cent solution to each gallon of freshly separated cream at 120°
F., and maintain at that temperature for ninety minutes in a closed
vessel. There " appears to be no doubt that it is impracticable to carry
on a jug-cream trade . . . without the use of preservatives," and this
trade. Dr. Hamill adds, is undoubtedly a convenience to the public.
Experimental inquiry has substantiated the statement that 025 per

No.

Boric Acid

Anhydrous

Borax
(NajBA).

Soluble

Other
Sweetening or

Moisture,

etc , by

Difference.

(H3BO3).

Saccharin.

Preservative
Substances.

Per cent

84-32

15-15

Nil.

0-53

54-96

2085

0-85

19-39 fa)

3-95

76-76

2212

1-18

Nil.

77-98

21-53

Nil.

0-49

77-13

22-22

Nil.

0-65

73-02

24-24

Nil.

2-74

75-66

22-61

0-93

Nil.

0-8

75-39

21-01

Nil.

0-59 (6)

301

59-56

32-09

Nil.

3-78 (c)

4-57

(a) Cane sugar, 17-45 ; salicylic acid, 1-94. (6) Cane sugar, (c) Sodium
benzoate.

cent of boric acid is insufficient to preserve cream for more than three
or four days at 71° F., but 04 per cent keeps it for four to seven days
at this temperature. Dr. Arthur Harden's results are summarized in
the following statements : —

Boric acid in the presence of an alkali (7 grms. of Na.^O per 100
grms. of boric acid) is a more efficient preservative than boric acid
alone.

With this proportion of alkali 0*4 per cent of the acid is practically
as effective as 0*5 per cent at 65° F., and slightly less at 71° F.
Cream may be preserved by either of these proportions for about four
to seven days at temperatures up to 71° F.

0"5 per cent of boric acid does not prevent development of moulds
in cream after four to seven days.

The recommendations which Dr. Hamill submits to the Local
Government Board are briefly as follows : —

1. Boric acid or mixtures with borax should be the only preserva-
tive allowed to be used in cream.

2. An exception might be made in respect of the use of hydrogen
peroxide, providing that only traces are allowed to remain in the cream.

3. (a) Declaration of the presence of preservative to the purchaser,
whether wholesale or retail, should in all cases be adequate, and pre-
served cream should be differentiated from cream containing no added
preservative, (b) The maximum amount of boron preservative allowed,
calculated as boric acid (H3BO3) should be 0*4 per cent from May to
October inclusive, and C'25 per cent during the remainder of the year.

MILK. 69

4. The presence of sweetening agents, such as saccharine, should
be notified to purchaser.

In an addendum to the report, Mr. G. W. Monier- Williams, Ph.D.,
deals with the detection of small quantities of benzoic acid, salicylic
acid, and saccharine in cream. His process is as follows : —

Concentrated phosphoric acid (1 c.c.) is added to 100 grms. of
cream, and the mixture heated with constant stirring on an asbestos
gauze over a Bunsen burner until all the water has been driven off
(mere traces only of benzoic and salicylic acids volatilize owing to
their great solubility in butter-fat). The temperature should not rise
above 120° C. The clear fat is filtered through a dry filter, cooled to
60° to 70° C. and shaken out with 50 c.c. of sodium-bicarbonate solu-
tion (0'5 per cent) heated to the same temperature. The separated
alkaline liquid is filtered, acidified with strong hydrochloric acid (1 c.c),
cooled, and extracted with three successive quantities (15 to 20 c.c.) of
ether. The combined ethereal solutions are dried .with calcium
chloride, and the ether distilled off.

If saccharine is present the residue will taste distinctly sweet. To
detect the presence of salicylic or benzoic acid, strong ammonia (1 c.c.)
is added to the residue, which is then evaporated to dryness and taken
up with four drops of water, and a minute drop of a 10 per cent solu-
tion of iron alum added. The characteristic purple coloration or buff
precipitate will indicate the presence of salicylic acid or benzoic acid
respectively. The limits of the test are benzoic acid 0'0075 per cent,
saccharine 0*001 per cent, salicylic acid 0*0002 per cent.

Cream. — Cream is the thick fatty layer which rises to the surface
of milk when it is allowed to stand, or is otherwise induced to separate,
leaving the skimmed milk (or separated milk when a centrifugal process
is used). There is no legal standard for cream, other than the necessity,
of course, that it shall not be adulterated ; but the best cream contains 50
per cent of fat and no good cream should contain less than 30 per cent.
The table on page 70 shows the composition of a number of samples
separated by the Aylesbury Dairy Company, the analyses being those
of Vieth and Droop Eichmond.

The principal adulterants of cream; are, as stated above, gelatine,
starch, and sucrate of lime.

Starch may be detected by means of iodine in the usual manner.

Gelatine can be detected by the method proposed by Stokes.

An acid solution of mercuric nitrate is prepared by dissolving
mercury in twice its weight of nitric acid of 1-42 specific gravity and
diluting this to twenty-five times its volume, with water; 10 c.c. of
the milk or cream is mixed with an equal volume of this solution, the
mixture shaken and 20 c.c. of water added. The liquid is again shaken
and allowed to stand for five minutes and then filtered. If much
gelatine be present the filtrate will be opalescent and cannot be obtained
clear. To a portion of the filtrate add an equal volume of saturated
solution of picric acid. In the presence of gelatine a turbidity or yellow
precipitate, according to the amount present, will be formed.

FOOD AND DKUGS.

Fat.

Solids not Fat.

Average.

Highest.

Lowest.

Average.

Highest.

Lowest.

Per ceut

Per cent

Percent

Per cent

1883

35-5

41-1

31-8

6-8

7-1

6-3

1884

35-3

390

32-6

6-8

7-0

6-4

1885

42-5

51-1

35-9

1886

44-25

46-0

41-5

1887

43-2

46-1

40-6

1888

45-55

48-0

43-5

1889

47-35

49-9

45-5

1890

48-35

50-5

45-3

1891

49-05

51-9

45-7

1892

46-85

49-5

43-9

1893

47-7

50-9

45-0

1894

49-0

51-2

46-5

1895

49-1

50-6

47-4

To detect sucrose or calcium sucrate, the process of Eothenfusser
(" Zeit. Untersuch. Nahr. Genuss.," 1910, xix. 465) may be used. A
portion of the sample is heated to 90° C, and then treated with an
equal volume of a mixture of 2 volumes of a solution of lead acetate
(500 grms. in 1200 c.c. of water) and 1 volume of ammonia solution
of specific gravity 0-944. The whole is well shaken for thirty minutes
and after a minute is filtered ; 3 c.c. of the filtrate is mixed with 3 c.c.
of a solution of diphenylamine (2 grms. of diphenylamine in 10 c.c. of
alcohol, 25 c.c. of glacial acetic acid and 65 c.c. of hydrochloric acid),
and heated in the water bath for five to ten minutes. Another portion
is tested with Fehling's solution to see if all the lactose has been removed.
If no reduction takes place in the latter test, a blue coloration in the
former is proof of the presence of sucrose.

In the detection of calcium sucrate in cream, the mixture of cream
and ammoniacal lead acetate should be heated to 65° and should pre-
ferably be allowed to stand some minutes before filtering.

Salicylic may be extracted from the milk which has first been
treated with acid nitrate of mercury solution (see under determination
of lactose), by ether and the ether extract tested with ferric chloride
solution, when the characteristic violet colour will be obtained. Car-
bonate of soda will be shown by the effervescence of the ash when
treated with hydrochloric acid.

Added Colouring Matter in Milk. — Annatto has practically been
the only colouring matter used until recently. Milk dealers have con-
sidered caramel unsuitable, as it has too much brown and too little
yellow in its composition, and therefore it is diffipult to imitate the
natural colour of milk. Annatto, on the contrary, gives a rich, creamy
appearance to the milk, even if watered, when carefully used with the
right dilution. This accounts for its popularity with milk dealers.
One or more of the azo-dyes have been much used of late, as they give
just as good a cream colour as annatto.

MILK. 71

Appearance of Artificially Coloured Milk. — The natural cobiir of
milk is to be found chiefly in the cream. Artificial colouring, on the
other hand, spreads through the whole of the milk. When the cream
has risen to the surface, the underlying milk, instead of having the bluish
colour characteristic of skimmed milk, is the same colour as the cream,
especially if much colouring matter has been used. An analyst in ex-
amining a number of samples can often judge artificially coloured milks
from their appearance alone.

Nature of Annatto. — Annatto, arnatto, or annotto is a reddish-
yellow colouring matter, which is derived from the pulp enclosing the
seeds of Bixa Orellana, a shrub grown in South America and the
West Indies. The form used in the coloration of milk is a solution of
the colouring matter in weak alkali (see p. 247).

Nature of " Anilin Orange". — The azo-dyes are the best of the
coal-tar colours for colouring milk, and they are, therefore, most used.
The Department of Food and Drug Inspection of the Massachusetts
Board of Health have had some samples of these commercial "milk
improvers" analysed, and they have found them to be mixtures of
two or more members of the diazo- compounds of anilin. A mixture
of what is known to milk dealers as " Orange G " and " Fast Yellow "
gives exactly the same colour as one of these preparations which was
obtained from a milk dealer who had previously used it. A generic
name, such as " a coal-tar dye " or " anilin orange " is more convenient
for purposes of prosecution or otherwise, than a particular description,
considering our present knowledge of the subject.

Systematic Examination of Milk }or Colour. — Leach employs the
following general method for examining suspected milk samples —
Curdle about 150 c.c. of the milk by means of heat and acetic acid,
in a porcelain basin over a Bunsen flame. Gather the curd into one
mass by aid of a stirring rod, when it is easy to pour off the whey. If,
however, the curd is too finely divided in the whey, strain it through
a sieve or colander. All of the annatto, coal-tar dye, and part of the
caramel present in the milk will be found in the curd. Place the curd,
which should be free from all whey, in a corked flask with ether and
allow it to digest with it for several hours until the fat has been
extracted and with it the annatto. Then pour o£f the ether, and if the
curd is perfectly white, either the milk is not coloured, or annatto has
been used. If, on the contrary, the curd is coloured more or less
deeply, anilin orange or caramel has been used, the amount being
roughly estimated by the depth of the colour. Hence it is obvious
that of the three colours annatto, caramel, and anilin orange, only
annatto can be extracted by ether. The curd will have a brown
colour if caramel had been present and a brightish orange if anilin
orange has been used. The following tests should then be applied : —

(a) Tests for Annatto. — Evaporate the ether extract containing the
fat and annatto, if present, on a water-bath. Make the residue alkaline
with sodium hydroxide, pour upon a small wet filter, which will keep
back the fat, and allow the annatto, if present, to permeate the pores
of the filter as the filtrate passes through. After washing off the fat
carefully under the water tap, it will be found that all the annatto of

72 FOOD AND DEUGS.

the nfilk used for the test has collected on the filter, giving it an
orange colour, which is fairly permanent and varying in depth accord-
ing to the amount of annatto present. To confirm this test add
stannous chloride to the coloured filter, when the characteristic pink
colour is produced.

(6) Tests for Caramel. — Take the curd after the ether has been
poured off, and after it is free from fat, place in a test-tube and shake
well with hydrochloric acid. If caramel is present, this acid solution
gradually turns a deep blue on shaking. The white fat-free curd of
uncoloured milk would show the same colour if all the fat has been
thoroughly extracted from the curd. It is most necessary for the
quick formation of the colour that the curd should be entirely free
from fat. The reaction will be quicker if gentle heat is applied.
Caramel is only indicated when the blue coloration of the acid ap-
pears in conjunction with a coloured curd. The coloration is of a
brownish- blue when much caramel is present. Even if there is a blue
coloration it is a good plan to confirm its presence by testing a
separate portion of milk as follows : —

Curdle about a gill of milk by adding strong alcohol. Filter off
the whey, and add a small quantity of subacetate of lead. Collect the
precipitate thus formed upon a small filter and dry in a room free from
hydrogen sulphide. Pure milk treated in this manner gives a residue
wholly white or at the most a very pale straw colour, but, if caramel
is present, the residue is a more or less dark- brown colour varying
according to the amount present.

(c) Tests for Coal-tar Bye. — When azo-dye has been used to colour
the milk, apply strong hydrochloric acid to the coloured curd in a
test tube and the liquid will immediately turn pink. If much anilin
dye has been added to the milk, the curd will sometimes have a pink
coloration when hydrochloric acid is applied directly to it, before
ether is added. The colour reaction with fat-free curd is unmistak-
able, and very delicate.

Lythgoe states that the presence of anilin orange in milk
can be determined by directly adding 10 c.c. of strong hydro-
chloric acid to an equal quantity of the sample, mix well together and
if the dye is present in more than minute traces, it will produce a
pink coloration.

Condensed Milk.

The evaporation of water from milk, especially in Switzerland,
and packing the condensed product and exporting it to other countries
is now a very common and lucrative practice. The term condensed
milk is usually applied to the thick syrupy product containing 40 to
50 per cent of water, but recently powdered milk has become a regular
commercial article. Of powdered milk there is little to be said, since
it is obvious that its composition should correspond with that of
milk minus its water. The composition, to come within the regulation
of the Board of Agriculture for milk, would have to be (on the basis
of 11*5 per cent solids and 3 per cent fat)

Fat = 26 per cent
Non-fatty solids = 72 per cent.

MILK.

The author has examined many samples and found the fat varies
in the best brands from 26 to 30 per cent. The further composition of
this preparation is indicated by the figures given under milk. The
usual liquid condensed milk is sold as either unsweetened or sweetened,
the latter containing added cane sugar and having much better keeping
properties when exposed to the air, than the unsweetened variety.

The following represent the composition of the better- class con-
densed milks, and it is to be noted that there exists no legal standard
for condensed milk so far as the amount of condensation is concerned,
but of course the proper proportion of the various constituents of the
milk, other than the water, must be preserved ; skimmed milk when
condensed may not be sold as condensed milk, without disclosing the
fact that it is made from skimmed milk.

Unsweetened Condensed Milk.

Total Solids.

Fat.

Proteids.

Lactose.

Ash.

Observers.

51-61

15-67

17-81

15-40

2-53

Percy

36-10

11-06

12-75

—

Allen

43-00

9-8

11-3

18-5

2-5

Pearmain and Moor

23-10

8-10

8-66

—

1-55

Leach

39-80

12-21

13-10

—

2-4

Parry

Sweetened Condensed Milk.

Solids.

Fat.

Proteids.

Lactose.

Cane Sugar.

Ash.

Observer.

68-10
/4-40
74-29
73-11

11-05
10-8
10-65
11-61

10-95

8-8

8-46

9-1

16-0

11-97

13-0

37-1

41-92

38-00

1-7

1-29

1-4

Allen

Pearmain and Moor

Leach

Parry

Condensed unsweetened milks made from skim milk will show a
low fat value, whilst the proteids and sugar will be high. The same is
true, of course, for sweetened varieties.

It may be noted that in the United States an official standard
exists for condensed milk. Standard condensed milk must contain at
least 28 per cent of milk solids, 27-5 per cent of which is milk fat.

Anahjsis of Condensed Milk.— 10 grms. of the well-mixed sample
should be diluted to 100 c.c. with water. 10 c.c. of this mixture will
then represent 1 grm. of the sample.

Total Solids.— 10 c.c. of this (1 grm. of sample) are heated in a
flat-bottomed capsule on the water bath and then transferred to the
water oven. Five to six hours are required for complete desiccation.
After weighing the total solids, these are ignited at a dull red heat and
the ash weighed. Where much sugar is present, it is necessary to dry

74 FOOD AND DRUGS.

the liquid on a little recently ignited asbestos, which is weighed with
the dish itself.

Fat may be approximately determined by extraction with ether of
paper saturated with 10 c.c. of the above solution (Adams' process)
as described above ; the Werner- Schmidt and centrifugal processes are
not very suitable unless modified, especially when sweetened milks
are in question. Leach (" Journ. Amer. Chem. Soc." 1900, 589) prefers
the following method : 15 c.c. of a solution of 40 grms. of the sample
made up to 100 c.c. (i.e. 6 grms. of the sample) are measured into a
Babcock bottle, and 4 c.c. of a solution of copper sulphate (of the
same strength as Fehling's solution) are added. The mixture is well
shaken and the proteid precipitate and fat are rapidly separated by
whirling in the centrifugal machine. The supernatant liquid contain-
ing the sugar is drawn off with a pipette, and the precipitate is twice
washed with water, and the washings drawn off and added to the
clear liquid. Water is now added to the precipitate up to 17-6 c.c.
and 17*5 c.c. of sulphuric acid are added, and the liquid is then placed
in the centrifugal machine and the reading of fat multiplied by three
to give the percentage of fat (the bottles being graduated in reference
to 18 grms. of the sample).

Proteins may be determined on 5 c.c. of the solution of 40 grms. in
100 c.c. This is diluted to 40 c.c, and copper sulphate solution added
carefully until no further precipitation takes place, avoiding much
excess of copper. Add a little very dilute caustic soda solution, leav-
ing the solution faintly acid. Filter through a tared filter paper,
wash, dry at 100 c.c. and weigh. Burn the paper, and take the differ-
ence between the weights of the precipitate and the ash as the pro-
teids and fat. This, minus the amount of fat, gives the proteins.

Lactose may be determined in the filtrate and washings from the
above process. These are made up to 100 c.c. with water and 10
c.c. of Fehling's solution are reduced by this liquid in the usual
manner. The lactose may be calculated from the following formula : —

100 X 0-067
^ ~ S X 0-02

where L is the percentage of lactose in the sample and S is the number
of c.c. of milk solution of the above strength required to reduce the
10 c.c. of Fehling's solution.

Cane sugar, in sweetened condensed milk, may be taken as the
difference figure after milk sugar, proteids, fat and ash have been de-
termined.

Polarimetric determinations are somewhat unreliable with con-
densed milks, since the heat to which the milk has been exposed dur-
ing evaporation causes changes which cannot be allowed for. The
following methods are approximately accurate : —

Stokes and Bodmer (" Analyst," x. 10) add 1 per cent of citric
acid to coagulate the milk without heating, dilute, filter and determine
the reducing power of the clear filtrate. One per cent of citric acid is
added to another portion of the filtrate, and the authors give ten
minutes as the length of time for boiling the solution, though, accord-

MILK. 75

ing to Watts and Tempany ("Analyst," xxx. 119) it is better to
boil for at least thirty minutes. The solution is allowed to cool, is
neutralized, and the reducing power again deternained. The invert
sugar formed from the sucrose is measured by the difference in the
two reductions.

Leffmann and Beam use invertase for inversion. They precipitate
the proteins with mercuric nitrate and polarize the clear whey. The
acid is carefully neutralized in a portion of the filtrate, one drop of
acetic acid is added, also a small quantity of invertase and a few
drops of an antiseptic. The whole is incubated for twenty-four hours
at 35° to 40°, and the liquid made up to known quantity and polarized
again. The difference between the two readings is due to inverted
sucrose.

Bigelow and McElroy ("Jour. Amer. Chem. Soc," 1893, 15)
suggest the following method for determining sugars, in condensed
milk. The reagents used are acid mercuric iodide and alumina
cream.

Place the entire contents of a can in a porcelain dish and mix
thoroughly. Weigh a number of portions of 26-048 grms. into 100
c.c. flasks. Add water to two of the portions and boil the solutions.
Boiling is necessary in order to ensure normal rotations. To one
portion add a few c.c. of a solution of 53 grms. of potassium iodide,
22 grms. of mercuric chloride and 30 c.c. of glacial acetic acid in 1000
c.c. of water ; and also a little alumina cream. Make up to 100 c.c.
and filter. The polarimetric reading of the filtrate is determined. Heat
the other weighed portion in the water-bath to 55°, add one-half of a
cake of compressed yeast, keeping the temperature at 55° for five hours.
Clarify the solutions as before, cool to room temperature, make up to
100 c.c, mix, filter and take the polarimetric reading. The amount of

cane sugar can be determined by the formula — C = f /o"cc — 77o' ^^Q*"®

C is the percentage of sucrose, D is the difference between the direct
and inverted readings, and t is the temperature.

Determine the total reducing sugar by one of the reducing processes
on a weighed portion of the original material ; if the sum of it
and the amount of cane sugar determined by the inversion method
is equal to that obtained by the direct reading of both sugars before
inversion, no invert sugar is present. If the amount of reducing sugar
seems excessive, milk sugar may be separated as follows : —

Dissolve 25 grms. in water, boil the solution, cool to 80°, add a
solution of about 4 grms. of glacial phosphoric acid, keep the mixture
at 80° for some minutes, then cool to room temperature, make up to a
known volume, mix and filter. Next, add potassium iodide in such
quantity as not to quite neutralize the acid, and water sufficient to
make up for the solids precipitated by the acid. Filter the mixture,
and measure the filtrate in portions of 100 c.c. into 200 c.c. flasks.
Add a solution containing 20 mgs. of potassium fluoride and half a cake
of compressed yeast to each flask, then allow the mixture to stand for
ten days at a temperature of from 25° to 30°. Fermentation will re-
move the invert sugar and cane sugar while the milk sugar is still

76 FOOD AND DRUGS.

unaffected. Fill the flasks to the mark, shake, and determine the milk
sugar by both reduction and the polariscope.

The amount of copper reduced by the milk sugar and invert
sugar in the original sample less the milk sugar remaining after fer-
mentation is due to invert sugar. C. B. Cochran ("Jour. Amer.
Chem. Soc." 1907, 29, 545-56) advises Wiley's acid mercuric-nitrate
solution to invert sucrose in analysing sweetened condensed milk.
He has found that this inverts sucrose only very slowly at temperatures
below 15° ; 50 c.c. of the solution to be inverted (containing 3 c.c. of
mercuric solution per 100 c.c.) are polarized immediately the solution
has been mixed at 15°, then heated in boiling water for 7 minutes then
polarized again. The following formula gives the sucrose content in
the case of normal solutions : —

Sucrose = — — — where D represents the difference in polariza-

142-66 -0-5^ ^ ^

tion before and after inversion and t the temperature. Leff-
mann employs the sesame oil test for detecting sucrose in condensed
milk or milk sugar, i.e. 1 c.c. of sesame oil, 1 c.c. of concentrated hydro-
chloric acid and 0*5 grms. of the sample are well mixed together by
shaking. The characteristic crimson coloration will be apparent within
half an hour. This test is perfectly reliable and better than that to be
found in the United States Pharmacopoeia which depends on carbon-
ization of the sucrose by strong sulphuric acid. A quick test for de-
termining sucrose in milk and cream is to boil a mixture of 15 c.c, of
milk, O'l grm. of resorcinol, and 1 c.c. of concentrated hydrochloric
acid. Pure milk remains unchanged while sucrose gives a fine red
coloration.

Baker and Hutton ("Analyst," xxxv. 512) have shown that the
most accurate results, when a biological process is used, are obtained by
using 0*5 grm. of washed brewer's yeast per 100 c.c. of a 2 to 3 per
cent solution, and fermenting for 60 to 70 hours at 27°. Results
varying from 90 to 100 per cent of the theoretical are obtained with
lactose in the presence of dextrose, sucrose, maltose or invert sugar, the
determination being made by a direct titration with Fehling's solution.

The Inteiyretation of Results. — In order to decide whether the
milk from which the sample was prepared was genuine or not, an
arbitrary standard must be agreed upon. Considering the Board of
Agriculture regulations, it is necessary to adopt their minimum limits
for this purpose. The fat value will be 3*0 per cent and the non-fatty
solids 8-5 per cent. In the case of unsweetened condensed milks the
fat of the original milk may be calculated from the formula : —

Fat of original milk 8*5

Fat of sample ' Non-fatty solids of sample*

Similar equations will give the ash of the original milk, and the
total solids will be approximately obtained by adding 8*5 to the fat
value.

In sweetened condensed milk, the fat and total solids in the sample
should first be calculated to the basis of the condensed milk less the cane

MILK. 77

sugar, that is by multiplying by ^qq _ pWhen P is the amount of cane^

sugar.

Now subtract the fat from the total solids (corrected as above), the
result being the non-fatty solids calculated to the sugar-free condensed
milk. This value divided by 8-5 (adopting the official minimum) will
give the number of times the milk has been condensed.

The percentage of fat in the cane-sugar-free sample (corrected as
above) divided by the number of times condensed gives the amount of
fat in the original milk.

Approximately accurate results are also obtained by dividing the
ash of the sample (corrected as above, in sweetened milks) by O"?, the
average ash value for normal milks. This gives the number of times
the milk has been condensed, and the fat value of the sample (corrected)
divided by this gives the fat value of the original milk.

The Analysis of Altered Milk. — If only a slight amount of acidity
has developed in milk, no appreciable differences are noted between
the analyses of such a sample, and of the unaltered milk. But if de-
composition has gone so far that it is not possible to obtain a uniform
emulsion, no analysis will give satisfactory results. If the sample can
be well emulsified by the use of an egg beater, the analysis presents no
difficulty. The following process is employed officially in the govern-
ment laboratories when milk is very sour : —

From 10 to 12 grms. are weighed into flat platinum capsules and
neutralized with decinormal strontia solution using phenol-phthalein
as indicator. The liquid is then evaporated in a water bath until it
has a nearly solid consistency, and whilst hot enough to keep the fat.
melted, 20 c.c. of dry ether (0*720) are poured on to the solids which
are well stirred with a glass rod. The ether is filtered through a dry
filter (10 cms. diameter) into a wide-mouthed weighing bottle. The solids
are similarly treated with eight successive portions of 10 c.c. each, of
ether. The filter paper is well washed with boiling ether, and any in-
crease in its weight when dried is added to the weight of the non-
fatty solids. These are dried in a water oven to "constant weight, and
the fat is weighed after evaporation of the ether. A deduction of
0*004:2 specific gravity each c.c. of the strontia solution used, is
necessary..

Allen recommends the examination of the whey when milk has
curdled but not undergone any further change. He finds that the
specific gravity of the whey varies in pure milks from 1*029 to 1*031
very rarely falling a little outside these limits. The solid matter in the
whey of pure milk varies from 6*7 to 7*1 grms. per 100 c.c. So that
a lower specific gravity or solid residue indicates watering.

, Fat may be conveniently estimated in sour milk by the Werner-
Schmidt process (p. 50).

Allen points out that the alteration in certain of the solids of milk
on keeping makes it inadvisable to base an opinion alone or chiefly
as to the genuineness of the milk on the amount of non-fatty solids.
He prefers to determine the nitrogen and the ash, since any nitro-
genous matter present will not, on decomposition, evolve nitrogen, so

78 FOOD AND DKUGS.

that the determination of the nitrogen value will not be altered by de-
composition within certain limits.

The value 0"5 per cent of nitrogen may be safely taken as the
lowest permissible limit for nitrogen in genuine milk, as determined
by the Kjeldahl process. In the same way adulteration may be pre-
sumed if the ash falls below 0*7 per cent, and not more than 30 to 33 per
cent of this should be soluble in water.

When it appears necessary to determine the nature and amount of
the loss in the non-fatty solids of a milk as a result of decomposition
by keeping, it is necessary to estimate the amount of alcohol and volatile
acids formed. The slight decomposition of nitrogenous matter with
the evolution of a trace of ammonia is of very little importance.

The alcohol is estimated by distillation in the usual manner, the
specific gravity of the distillate made up to the original bulk, giving
the amount of alcohol. In distilling the milk, the free acidity should
be first determined and the portion used for distillation first treated
with half the amount of alkali necessary to neutralize it. If more
completely neutralized, there is a risk of ammonia distilling over. The
first 20 per cent of the distillate usually contains a little free acid, so
that it should be kept separate, and then redistilled after complete
neutralization.

The amount of proof spirit (see table) by volume, multiplied by
0842, gives the weight of lactose which has undergone fermentation.

The amount of lactose lost by conversion into acetic acid is cal-
culated by determining the total acidity and the acidity of the fixed
residue of the milk. If 10 grms. be used, the number of c.c. of deci-
normal NaOH required by the volatile acids, is multiplied by 0*006
(acetic acid has a molecular weight of 60) and by 10. This gives the
percentage of acetic acid, which, when multiplied by 0*425 gives the
amount of lactose converted into acetic acid.

During the past few years, the use of soured milk as an article of
diet has come much into vogue, and although it differs but little in
chemical composition from ordinary milk apart from the presence of a
larger quantity of lactic acid, its examination from a bacteriological
point of view is sometimes required. The following interesting account
of the souring of milk is due to F. W. Gamble (" Pharm. Journ."
1909, 1, 253).

In natural milk caseinogen and fat are very closely associated ; a
large proportion of the calcium and of the phosphoric acid present
is also in more or less intimate association with the caseinogen.
Caseinogen is coagulated by rennet and some bacterial ferments with
conversion into casein. When milk is coagulated by rennet, the
casein and fat together form a curd, separating from the whey, which
contains in solution lactalbumen, lactose, salts, and a small quantity
of whey-proteid formed as a decomposition-product of caseinogen.
Kennet coagulates milk only in the presence of calcium salts ; if these
be completely removed by potassium oxalate before adding rennet
coagulation does not occur. Milk that has been boiled is not coagu-
lated by rennet, probably because the calcium salts are rendered in-
soluble ; if they be partially removed from solution by the addition of

MILK. ^^^ ^g

sodium citrate, the milk — known as " citrated " milk — is rendered less
amenable to the coagulating action of rennet, and advantage is fre-
quently taken of this means of preventing the formation of an indi-
gestible solid curd by the action of the rennet of the gastric juice in
the stomach of infants and invalids. Acids — such as lactic or acetic
acid — precipitate caseinogen from boiled or unboiled milk ; the calcium
salts are removed from their natural combination and pass into
solution.

Lactose, the principal carbohydrate of milk, is a disaccharide ; as
such it is not assimilable, but must undergo inversion before or during
absorption. It is not readily inverted by ordmary yeasts, and is, there-
fore, not very susceptible to alcoholic fermentation. It is, however,
very readily decomposed by a group of micro-organisms which by a
hydra-ting process convert a molecule of lactose into four molecules of
lactic acid, and are hence classed as " lactic acid bacilli ".

Milk is an almost ideal culture medium for both saprophytic and
pathogenic bacteria, since it presents in an alkaline or neutral liquid
proteid matter, carbohydrate, and salts, which, together, constitute a
complete bacterial diet. Unless drawn from the cow under the most
aseptic conditions, hiilk is immediately infected by a host of micro-
organisms, derived from the teats of the cow, from the byre, the milker,
and other sources. In so suitable a nidus very rapid multiplication
of these bacteria ensues, so that milk drawn in what may be called
strictly sanitary conditions, cooled to 45° F., and kept at that tempera-
ture, may contain an average of 4000 to 6000 bacteria per 1 c.c. after
five hours, whilst London milk, as ofifered for sale, may contain from
1,000,000 to 4,000,000 bacteria per 1 c.c. The actual number of
bacteria present in milk per 1 c.c. is, therefore, seen to vary very
greatly, the determining factors being chiefly the conditions under
which the milk is drawn, the temperature at which it is stored, and
the length of time that elapses before examination, A characteristic
rise and fall in the numbers has, however, been shown to take place.
At about fours after milking the number of saprophytic or putrefactive
organisms has reached an initial maximum, and a fall in the total
number of bacteria present is then noticed. This is due to the gradual
multiplication of lactic acid producing organisms, which, by rendering
the medium acid in reaction, inhibit the development of ordinary
putrefactive bacteria, and ultimately procure their extinction. The
total number of bacteria present then rises again to a second maximum
many times greater than the first, the organisms now consisting almost
wholly of lactic acid producing species, or those whose vitality has
withstood the action of the acid produced. At this stage, either owing
to exhaustion of pabulum or to the degree of acidity reached, growth
is again checked, the lactic bacteria rapidly die out, and only moulds
flourish.

Pasteur was the first to describe an organism characteristic of
lactic fermentation and to demonstrate the distinction between this
and alcoholic fermentation. Lister subsequently obtained from sour
milk a bacterium which he grew in pure culture, and termed Bacterium
lactis. In 1884 Hiippe grew the same organism on the then newly

80 FOOD AND DRUGS.

introduced solid media, and termed it Bacillus acidi lactici, a name
now applied to the whole family of micro-organisms possessing
similar bacteriological properties and capable of decomposing lactose
with formation of lactic acid. Many different species of lactic acid
producing organisms have since been described and the same species
have been described under different names, so that considerable
confusion exists in the terminology of the subject. The dif-
ferent species exhibit small cultural variations, some growing well
in the presence of oxygen, others better in deep vessels. They also
vary in size and shape and in the type of lactic acid produced.
Three organisms only need be described in any detail.

Ordinary lactic fermentation in this country is due chiefly to the
bacillus of Hiippe, which is a non-motile oval rod 0*6 to 2 microns
long. It is non-spore-bearing, and grows well at room temperature.
It forms acetic and optically inactive lactic acids, and produces a solid
curd separating from a clear fluid. Accompanying the Hiippe
bacillus in sour milk, there is frequently found another non-motile
organism in the form of short, thick rods = 1 micron long, called
Giinther's bacillus, or Bacillus acidi paralactici. It coagulates milk,
producing as a result of the decomposition of lactose, dextro-rotatory
or paralactic acid. These two organisms possess comparatively low
vitality, are destroyed as the proportion of lactic acid increases, and
are considered useless as therapeutic agents. In some parts of
Europe a native lactic organism is found which differs considerably
both from the bacillus of Hiippe and from that of Giinther. This,
known as Bacterium Gaucasiuni, the Bulgarian bacillus, Massol's, or
Boucard's bacillus, is very much larger than other lactic bacilli, is
slightly motile, and produces lactic acid in abundance. It grows very
slowly at room temperature, but freely at its optimum temperature of
100" to 105° F. It is possessed of great vitality, and withstands the
action of its autogenous lactic acid to a higher degree than any other
lactic acid producing organism yet discovered.

Milk that has been allowed to become sour is preferred by many
people to milk in its natural sweet condition. Spontaneously soured
milk contains fewer saprophytic bacteria than sweet milk a few hours
old. The casein undergoes a slight degree of peptonization as a result
of the fermentative process, and its partial precipitation by the lactie
acid prevents the formation of a solid curd in the stomach. Besides
being less cloying to the palate, sour milk is, therefore, also more di-
gestible ; amongst many country folk in Great Britain it is a regular
article of diet, and especially so in Scotland. Buttermilk also is largely
consumed ; it contains but a small proportion of fat, is distinctly acid
from the presence of lactic acid, and its proteids are present in a finely
flocculent form. In many countries a doubly fermented milk is pre-
ferred to that which has undergone simple lactic fermentation. Milk
does not readily undergo simple alcoholic fermentation ; certain forms
of yeast are, however, found to be symbiotic with the Caucasian
bacillus, and these set up jointly a mixed alcoholic and lactic fermenta-
tion. The best-know^n examples of this double fermentation are seen
in the preparations koumiss, kephir, yoghourt, matzoon, and leben.

MILK. 81

Koumiss, as prepared by the nomadic Tartars of Russia, is made from
mare's milk, the cultures being carried on by adding a small propor-
tion of old fermented milk to the newly drawn milk. Koumiss con-
tains about 1-7 per cent of alcohol and under 1 per cent of lactic acid.
Kephir is prepared in the Caucasus from the milk of goats, sheep, or
cows. Some old kephir is used to carry on the cultivation, or a few
kephir grains are soaked in warm water, and when swollen and soft
are added to new milk ; fermentation takes from one to three days,
according to the temperature, and the product remains good for a
considerable time. It contains about 2 per cent of alcohol. Kephir
grains are the dried scrapings from old vessels in which repeated fer-
mentations have taken place : they are yellowish-brown in colour, ir-
regular in size and shape, and have a characteristic odour resembling
that of peptone, of which they contain a considerable proportion.
Stored in a dry place, kephir grains retain their activity for many
years ; they contain, in- addition to the yeasts and lactic acid produc-
ing bacteria, one or two forms of streptococci. Yoghourt prepared from
maya ferment possesses properties resembling those of kephir, and is
a staple food of the Bulgarians and other Balkan races. The peoples
consuming these fermented milks as daily articles of diet are amongst
the most healthy and long-lived races of mankind.

Many other applications of lactic fermentation to the preservation
of foods might be quoted to show that the process is a general protec-
tion against putrefaction. The ubiquity of the bacilli of lactic fer-
mentation ensures their ingestion with food-stuffs of all kinds, and
their effect in the intestinal tube is, so long as they survive the
changed conditions, exactly comparable with what is observed outside
the body.

Lactic acid bacilli are a prominent feature of the normal flora of
the small intestine ; in this portion of the alimentary tract the alka-
line intestinal secretions are rendered acid in reaction by the gastric
juice and by the lactic, acetic, and other acids which are produced by
the fermentative processes carried on in the abundance of material
poured out from the stomach. In consequence of this acidity, and
w^hilst it is maintained, anaerobic putrefactive bacteria gain no position
in the small intestine. These proteolytic bacteria can exist only in a
distinctly alkaline medium, and so long as a sufficient quantity of acid
is produced, putrefaction does not occur. In the normal state the
same condition of affairs should exist in the large intestine. In the
normal colon and appendix aerobic bacteria and facultative anaerobes
(such as the Bacillus coll covimunis) are found, but no such strictly
anaerobic organisms as are present in pathological conditions. In the
normal state, such anaerobic organisms as are ingested with the food
are destroyed by the aerobic bacilli present. On the other hand, in
enteritis, appendicitis, and the conditions associated with auto-in-
toxication the flora of the large intestine is characterized by a diminu-
tion in the number of aerobic organisms, the bacteria inhibiting
putrefaction, and by an increase in the number of anaerobic micro-
organisms which are the bacteria giving rise to putrefaction. These
latter, by their growth in the albuminous intestinal contents, produce

VOL. I. 6

82 FOOD AND DEUGS.

soluble poisons which are absorbed into the circulation. Escaping
destruction by the liver or other defensive mechanisms of the body,
these toxins are considered to be active agents in the production of
many forms of ill-health and chronic disease.

To Metchnikoff must be attributed the idea of changing the
balance of power between these opposing forces of bacteria by the ad-
ministration, under proper conditions, of a culture of living and highly
active lactic acid producing bacteria. It should not, however, be
supposed that nothing more than the administration of suitable
organisms is necessary ; they must be assisted by the concurrent use
of a proper diet, calculated not only to favour the multiplication of
lactic organisms but also to inhibit the growth of putrefactive bacteria.
This is attained by the use of a farinaceous, milk, and vegetable diet,
as opposed to a diet rich in albuminous materials.

The desiderata in the lactic bacillus employed are high vitality and
a good degree of lactic acid producing power. These are best seen in
the Caucasian or Bulgarian bacillus, isolated originally from Bulgarian
maya, and described by Metchnikoff, Massol, and Boucard. When
ingested by man, this bacillus is not destroyed in the intestine, but
reaches the end of the colon in a living state, and is found in the
stools. Though not normally an inhabitant of the human intestine,
it readily becomes implanted there, and then acts efficiently against
anaerobic bacteria. This organism may be used in conjunction with
the organisms of Hiippe and Giinther, and these are usually added
for the improvement in flavour they impart to the soured milk.
Therapeutically, however, only the bacillus of Massol is an efficient
anti-putrefactive agent. Metchnikoff condemns the use of kephir and
other doubly fermented milks containing yeasts and cocci on the
grounds of their alcohol content and their irritating effect in some
forms of enteritis.

Soured milk for medicinal use must be prepared from milk that is
free from preservative and that has been rendered practically sterile
by boiling; milk so treated and subsequently inoculated with the
bacterial culture, is incubated at a temperature of 100° to 105° F.
until the desired degree of acidity is reached. Curdling takes place
when about 1 per cent of lactic acid is present in the milk, but this
degree of acidity will be nearly doubled if sufficient time be allowed to
elapse, though the flavour of the soured milk will be detrimentally
affected. The necessity for initial sterilization of the milk will be
realized when it is considered that the incubating process would at
first encourage the multiplication of any extraneous organism present,
and the ultimate destruction of such resistant pathogenic germs as
those of tubercle and typhoid fever could not otherwise be ensured.
The cultures are most conveniently employed in tablet form, which
retain their activity for a long period, and possess every advantage
over liquid cultures, except that they require a rather longer incubation
period to produce the full degree of acidity. Patients who are unable
to tolerate milk even in this readily digestible form are treated, in the
opinion of some physicians, with equal success, by the administration
of the tablets themselves, suitable saccharine matters being given at

CHEESE. 83

the same time. When taken regularly, the bacilli commence to appear
in the stools in about three days, and become established in the in-
testine in about eight days. The course of treatment usually lasts
about twelve weeks without intermission.

CHEESE.

Cheese consists essentially of the curd and fat of the milk of any
animal removed from the milk which has been curdled either by
natural souring of the milk or by the action of rennet. The mass of
curd and fat, after compression, is allowed to undergo certain fermen-
tative changes, due to the action of micro-organisms or enzymes. In this
country cheese is made entirely from cow's milk, and certain additions,
such as colouring matter and salt, are regarded as quite legitimate. On
the continent the milk of other animals, such as sheep and goats, is
used for the manufacture of cheese. Cream is also used as the source
of the cheeses known as cream cheeses. The nature of the decomposi-
tion which takes place in the ripening of cheese is but little known, but
there is no doubt that amongst the principal of these is the degradation
of proteid matter to compounds of much lower molecular weight, and
containing much nitrogen. In such "ripe" cheeses mould is very
common, and many cheese=? are never eaten until they are very mouldy.
The principal moulds existing in cheese are aspergillus glaucus, and
sporodonema casei. When decomposition has proceeded very far
large living organisms, such as the acarus domesticus, the ordinary
cheese mite, are to be found, and are much appreciated by many
cheese eaters. The following are the principal types of cheese met
with in commerce : —

(1) English cheese ; such as Cheddar, Cheshire, Stilton, Wensley-
dale, all being made from full cream milk.

(2) American cheese ; usually made on the type of English Cheddar
cheese, and made from full cream milk.

(3) Dutch cheese ; usually made from partially skimmed milk.

(4) Gruy^re ; a Swiss cheese, made from goats' milk.

(5) Soft French cheeses ; made from milk with cream added, such
as Brie and Neufchatel.

(6) Ewe's milk cheese, of which Koquefort is a type.

No standard can at present be laid down for cheese, other than
that it shall be the product of milk solely, or at all events with the
small allowable additions of salt and colouring matter. The important
point is that it shall be free from foreign fat, otherwise it must be
sold as margarine cheese. Section 25 of the Sale of Food and Drugs
Act, 1899, provides that—

,-^"The expression 'margarine cheese' means any substance,
whether compounded or otherwise, which is prepared in imitation of
cheese, and which contains fat not derived from milk."

" The expression ' cheese ' means the substance usually known as
cheese containing no fat derived otherwise than from milk."

The following analyses of cheese are due to WoU (" Dairy Calendar,"
223) :—

FOOD AND DKUGS.

Cheddar .

Cheshire .
1 Stilton .
I Brie

Neufchatel
I Roquefort
I Edam

; Swiss

; Cream

Water.

Per cent
S4-38
32-59
30-35
c0'3o
44-47
31-20
36-28
35-80
38-60

Casein.

Fat

Sugar.

Per cent

26-3 >

32-71

2-95

32-51

26-0o

4-58

28-85

35-39

1-50

17-18

25-12

1-94

14-60

33-70

4-24

27-63

33-16

2-00

24-06

30-26

4-60

24-44

37-40

25-35

30-25

2-03

Ash.

Per

cent I
58
31
83

'41 I
■99 1
■01 I
•90
-36
-07 ;

Muter ("Analyst," x. 3) has published the following series of
fuller analyses than the above : —

Ash.

Fat.

Water.

Fat.

Lactic
Acid.

Lac-
tose.

NaCl.

Sap.
Value.

Insol.

Sol.

Insol.
Acids.

Sol.
Acids.

American Ch

eddar 29-7

30-7

0-9

trace

2-16

1-54

1-2

89-98

3-3

220

Bondon (area

m) . 55-2

20-8

0-9

0-74

0-62

6-96

3-16

87-34

5-95

228

Camenbert

. 48-78

21-35

0-36

trace

0-16

8-64

3-46

87-15

6 09

229

Cheddar

. 33-40

26-6

1-53

2-3

2-0

1-52

87-66

5-60

227

Gloucester

. 37-2

22-8

1-8

2-56

2-0

1-64

87-00

6-28

229

Dutch .

. 42-7

16-3

1-35

2-26

9-1

4-02

87-2

6-09

229

Gruyere

. 33-2

27-3

1-35

3-12

1-58

1-05 I 87-3

5-98

228 •

Roquefort

21-56

35-96

0-72

—

1-70

8-64

3-42 87-0

6-27

229

Stilton .

. 28-60

30-70

1-08

—

1-80

2-22

0-75 86-2

7-02

231

A number of useful analyses are also published by Chattaway,
PeaMnain and Moore ("Analyst," xix. 145).

Reichert-

Valeuta

Water.

Fat.

Ash. '

Proteids.

Meissl
Value.

Test of
Fat.

31° C.

Cheddar .

33-8

30-5

4-1

4-2

26-7

24-4

Cheddar (Canadian) .

33-3

30-6

3-6

4-34

27-6

24-0

41-5^

American .

29-8

33-9

3-7

4-76

30-3

26-2

47-5°

Gorgonzola

40-3

26-1

5-3

4-36

27-7

22-1

26-5°

Dutch

41-8

10-6

6-3

5-11

32-5

27-0

40°

Gruyere .

35-7

31-8

3-7

4-49

28-7

31-1

41°

Stilton

21-2

45-8

2-9

4-14

26-3

45-5°

Cheshire .

37-8

31-3

4-2

4-03

25-7

31-6

43°

Gloucester

37-4

28-1

4-6

4-45

28-3

32-3

41°

Camenbert

43-4

22-6

3-8

3-83

24-4

33°

Parmesan .

32-5

17-1

6-2

6-86

43-6

28°

Roquefort

29-6

30-3

6-7

4-45

28-3

36-8

19°

Double cream .

57-6

39-3

3-4

3-14

19-0

31-2

40°

CHEESE. 85

The Adulteration of Cheese. — Cheese is adulterated by the addi-
tion of foreign fats, often in total substitution of the milk fat. If by
cheese one understands a cheese made from full cream milk, the
cheese made from skim milk would be regarded as adulterated, but
until the Board of Agriculture, make regulations, as they are em-
powered to do by the Food and Drugs Act, as to standards for cheese,
it appears to be legal to sell skim milk cheese as " cheese ".

The use of foreign fats — such as lard and compositions of the
margarine type, in the manufacture of " cheese "^forms the basis of a
very large industry, especially in America, where this margarine- cheese
is generally known as " filled " cheese.

The adulteration of cheese (gorgonzola) by the use of abnormally
thick artificial rinds, composed of tallow, iron oxide and barium
sulphate has recently formed the subject of successful prosecutions
(see Vol. II, p. 29).

The Analysis of Cheese.

Moisture. — Two to thtee grms. are heated in a water oven for
several hours until of constant weight. The loss is reckoned as
water.

Mineral Matter. — The residue from the moisture determination is
ignited at a low red heat and weight. To determine the salt, a separ-
ate portion should be charred at a low heat and the salt extracted by
repeated boiling with distilled water and titrated with standard silver
nitrate solution.

Fat. — Five grms. should be dried and rubbed down in a mortar
with 20 grms. of ignited sand till a powdery mixture is obtained.
This is then extracted in a Soxhlet with petroleum ether in the usual
manner.

This solvent is preferable to ether, as ether dissolves appreciable
quantities of lactic acid. Allen prefers to boil the powdered cheese
with several portions of the solvent and decant each time. He finds
that four boilings are sufiicient to exhaust the cheese.

The fat may also be determined by the Werner- Schmidt method
(see p. 50). Three grms. of the cheese should be boiled with 5 c.c.
of water and 10 c.c. of concentrated HCl, till, with constant shaking, all
but the fat is dissolved.

The Lythgoe-Babcock method is as follows : Take 5 to 6 grms. of
the cheese in a tared beaker, add 10 c.c. of boiling water and stir with
a rod until the cheese softens and an even emulsion is formed, adding
a few drops of ammonia to aid the process. The beaker may be kept
in hot water until the emulsion is complete and free from lumps.
Then add about half of the 17-6 c.c. of the sulphuric acid regularly
employed in the Babcock milk test (see p. 51), stir well, and pour into
the Babcock bottle. Wash the beaker out with the remainder of the
acid. Then proceed in the usual manner as in the centrifugal milk
test, reading the amount of fat in the neck as usual.

Lactic Acid. — Ten grms. of the cheese are shredded and made up
with water to 105 c.c. The mixture is heated to 50"" and well shaken

86 FOOD AND DEUGS.

for some time. The liquid is cooled, and filtered. Twenty-five c.c. of
the filtrate is practically equivalent to 2-5 grms. of cheese. This
quantity is titrated with decinormal alkali, using phenol-phthalein as
indicator. Each c.c. of alkali required may be regarded as being
equivalent to 0*009 grms. of lactic acid.

Milk Sugar. — Twenty-five grms. of cheese are divided as finely as
possible and extracted by boiling with three successive quantities of
100 c.c. of distilled water. The mixed filtrates are, when cold, diluted
to 250 c.c. and the milk sugar is determined in the ordinary way by
titration against Fehling's solution.

Examination of the Fat. — All that is stated under butter fat
applies to the examination of the fat extracted from the cheese, and a
judgment as to the presence of foreign fatty matter must be based on
the examination of the fat, especially by the Keichert process and the
refractometer.

Detection of Skimmed Mitk Cheese. — In a whole milk cheese, fat is
almost invariably in excess of the proteids ; when this is not the case,
it is only slightly below the nitrogenous constituents. If the fat is
materially below the proteids, the cheese has certainly been made
from skim milk. In such cheeses, the fat will often fall as low as 5
to 15 per cent.

Determination of Nitrogenous Matter. — The total nitrogen may be
determined in 2 grms. by Kjeldahl's process, and this, multiplied by
6-33, may be taken as the total "proteid matter".

Where a full examination of the nitrogenous matter is desired,
the process of Van Slyke may be adopted. This is as follows : —

Place 25 grms. of the sample in a porcelain mortar and mix with
the same amount of clear quartz sand. Transfer the mixture to a
450 c.c. Erlenmeyer flask and add about 100 c.c. of water at 50° C,
keeping the temperature at 50° to 55° C. for half an hour, and fre-
quently shaking. Transfer the liquid through an absorbent-cotton
filter to a 500 c.c. graduated flask. Heat, shake, and decant from
the residue repeatedly portions of water of 100 c.c, until the fil-
trate or water extract amounts to just 500 c.c. at room temperature,
without taking into consideration the fat floating on the top ; use
aliquot parts of this water extract for the various determinations.

Water-soluble Nitrogen. — To determine the nitrogen use Gunning's
method on 50 c.c. of the foregoing water extract corresponding to 2-5
grms. of cheese.

Nitrogen as Paranuclein. — To 100 c.c. of the above water extract
(corresponding to 5 grms. of cheese) add 5 c.c. of a 1 per cent solu-
tion of hydrochloric acid. Keep the temperature at 50° to 55° until a
clear liquid floats on the surface, showing that separation is complete.
Filter, wash the precipitate with water, and employ the Gunning
method to determine the nitrogen.

Nitrogen as Coagulable Protein. — Take the filtrate of the preceding
determination and neutralize with dilute potassium hydroxide. Heat
to the temperature of boiling water until any coagulum that there
may be present completely settles. Filter, wash the precipitate and
determine the nitrogen contained.

CHEESE. 87

Nitrogen as Caseoses. — To the preceding filtrate add 1 c.c. of 50
per cent sulphuric acid saturated with zinc sulphate and warm to a
temperature of about 70° C. till the caseoses settle out completely.
Allow to cool, filter, and wash with a saturated solution of zinc
sulphate made acid with sulphuric acid. Determine the nitrogen in
the precipitate.

Nitrogen as Amides and Pepto7ies. — Into a 250 c.c. graduated
flask pass 100 c.c. of the aqueous extract of cheese. Add 1 grm. of
sodium chloride and a solution containing 12 per cent of tannin until
the clear liquid floating on the surface does not precipitate further.
Dilute to the 250 c.c. mark, shake, pour upon a dry filter, and de-
termine the nitrogen in 50 c.c. of the filtrate, which indicates the
amount of nitrogen in the amido-acid and ammonia compounds. If
the amount of nitrogen as ammonia separately determined, is deducted
the difference is the amido- nitrogen.

Nitrogen as peptones can be obtained by deducting the total sum
of the amounts of nitrogen, as paranuclein, coagulable proteins,
caseoses, amido-bodies and ammonia from the whole amount of
nitrogen in the aqueous extract.

Nitrogen as Ammonia. — Take 100 c.c. of the filtrate from the fore-
going tannin-salt precipitation and distil into standardized acid, then
titrate in the usual way.

Nitrogen as Paracasein Lactate. — To the residue, which is found
insoluble in water when obtaining the aqueous extract, add several
portions of a 5 per cent solution of sodium chloride. This forms a
500 c.c. salt extract of the same, in a similar way to that employed in
preparing the water extract.

Take an aliquot part of this salt extract to determine the nitrogen.

Van Ketel and Antusch (" Nedeerl. Tydschr. Pharm." 1897, 82)
affirm from the analysis of a number of cheeses that only about 80 per
cent of the nitrogen is present in the form of proteids, the remaining
20 per cent existing as ammonia and amido-bodies. To determine
the nitrogen present as ammonia, they distil the sample, which should
have previously been powdered with the addition of sand, with water
containing barium carbonate in suspension. Transfer the distillate
into a measured quantity of standard sulphuric acid, boil, then neutra-
lize the excess of acid with standard soda, using rosolic acid as an
indicator. To determine the nitrogen present as amido-compounds,
steep the powdered cheese with water for fifteen hours at ordinary
temperatures. Add a little dilute sulphuric acid (1 : 4), then precipitate
the peptones and proteids by phospho-tungstic acid. Filter off the
precipitate, and wash with water containing a little sulphuric acid,
Make up the filtrate to a definite amount, and determine the nitrogen
in an aliquot part of the liquid by Kjeldahl's process, making allow-
ance for the nitrogen existing as ammonia. To determine the pep-
tones and albumoses together, boil the powdered cheese (mixed with
sand as already described) with water ; filter, leaving the undissolved
casein and albumin. Add dilute sulphuric acid and phospho-tungstic
acid to precipitate the peptones and albumoses in an aliquot part of
the filtrate. Wash with acidulated water, then treat the precipitate

FOOD AND DRUGS.

by Kjeldahl's process. The total amount of nitrogen in the cheese
can be estimated by Kjeldahl's process and, after allowing for the
nitrogen present in other forms, the balance is calculated to casein.
A very small amount of indigestible casein is present.

Another and more elaborate method of distinguishing the various
classes of nitrogenized compounds to be found in matured cheese is
described by A. Stiitzer (" Zeit. Anal. Chem." 1896, xxxv. 493;
" Analyst," xxii. 14). The following table of figures shows the results
obtained by Stiitzer in three cases : —

Cameiibert.

Swiss.

Gervais.

Per cent.

Per cent

Per cent.

Water

50-90

33-01

44-84

Fat

27-30

30-28

36-73

Fat-free organic matter ....

18-66

31-41

15-48

Ash

The ash contained —

3-14

5-30

2-95

Calcium

0-03

1-56

0-14

Phosphoric acid

0-76

0-82

0-23

Solium chloride

Total nitrogen . . . .

2-21

1-56

0-76

2-900

5-072

1-923

Nitrogen as ammonia ....

0-386

0-188

0-031

„ „ amides

1-117

0-459

0-099

„ ,, albumoses and peptones

0-885

0-435

0-298

,, ,, indigestible matter

0-115

0-119

0-166

,, ,, casein and albumin
In 100 p-irts of nitrogen there existed —

0-397

3-871

1-139

As ammonia ......

13-0

3-7

1-11

„ amides ......

38-5

9-0

5-2

„ albumoses and peptones

30-5

8-6

15-5

„ indigestible matter ....

4-0

2-4

8-6

,, casein and albumin ....

14-0

76-3

69-1

Percentage of casein and albumin dissolved

in pepsin solution —

In 30 minutes

100

„ 60 „

100

BUTTER.

Butter is the product obtained by churning milk, so that the fat
globules adhere in a compact mass, together with a certain amount of
water and non-fatty solids, the greater portion of the milk serum being
removed by washing and mechanical means. More or less common
salt is added, according to taste, the product being sold as salt butter
or fresh butter according to the amount of salt it contains. Butter is
one of the few articles of food for which a legal standard exists. The
Board of Agriculture, acting under the powers conferred on them by
section four of the Food and Drugs Act of 1899, framed regulations for
the sale of butter in 1902. If any butter be sold containing more

BUTTER.

than 16 per cent of water it shall be presumed to contain added water
and therefore not to be genuine butter, until the contrary be proved.
This point is dealt with fully in Vol. II. The average composition
of normal butter, made from cow's milk is as follows : —

Per lent

Water 1200

Butter fat 86-80

Casein 0-50

Lactose ........... 0-45

Mineral matter ......... 0-25

Naturally, abnormal samples are to be met with, but these are
usually due to the use of methods of manufacture not generally em-
ployed. ^

The determination of the proportions of the proximate constituents
of butter presents no difficulties. The whole problem of butter analysis
lies in the examination of the fat.

The Composition of Butter Fat. — Pure butter fat consists almost
entirely of triglycerides of the fatty acids. Traces of cholesterol and
colouring matter are also present, but the amount is rarely more than
0*5 per cent. In addition to the glycerides of oleic, palmitic and
stearic acid, there are small quantities of the glycerides of arachidic,
myristic and lauric acid. But the characteristic feature of butter fat
is the comparatively large amount of glycerides of volatile fatty acids,
amongst which are those of butyric, caproic, caprylic and capric acids
(and traces of acetic acid). It is the decomposition of these latter, with
the liberation of the volatile fatty acids that causes " rancidity " in
butter.

According to Violette ("Journ. Soc. Chem. Ind." 1890, 1157) the
following represent the percentage composition of butter fat : —

Per

cent

Butyrin

6-94

6-09

6-28

5-76

5-28

5-49

5-45

5-00

Caproin

4-06

3-58

3-70

3-39

3-09

3-23

3-10

2-94

Glyceride

3 of volatile solid acids

3-06

3-22

2-96

3-16

3-06

2-53

3-16

3-15

of non-volatile acids

85-98

86-62

86-60

86-93

88-10

87-60

88-42

Other observers give the following values :-

Butyrin

Caproin i
Caprylin and Caprin 1
Olein I
Palmitic, stearin, etc. l"

Bell.

Blyth.

Spallanzani.

Per
cent
7-01

2-28

37-73
52-98

Per
cent
7-7
0-1

42-2
50-0

Per

cent

5-08

1-02

0-31

93-59

90 FOOD AND DRUGS.

The Analysis op Butter.

Determination of Water. — This is determined by heating 3 to 5
grms. in a flat capsule. A small glass rod should be weighed with
the capsule, and the fat stirred repeatedly. The sample should be
heated for from five to six hours. A fairly accurate and rapid method
is to shake 10 grms. of butter with 30 c.c. of ether previously satu-
rated with water. The separated aqueous liquid is run off into a
graduated tube containing 5 c.c. of brine containing a drop or two of
acetic acid. The increase of volume of the aqueous liquid represents
nearly accurately the number of grms. of water in the 10 grms.
of butter.

As mentioned above, iDutter should not contain more than 16 per
cent of water. Where a higher percentage is present, it is nearly al-
ways the case that special methods have been used to incorporate a
higher percentage of water with the fat. No properly made butter
need contain more than 16 per cent. A practice started some few
years ago of blending milk with butter. This, of course, resulted in
getting considerably more than 16 per cent of water into the butter,
and is practically tantamount to the addition of water. Legal de-
cisions have now caused that it shall be sold under the qualified name
of " milk-blended butter ".

Martinez ("Land. Jahb." 1898, 773) reported on over 20,000
samples of normally made butter from various European countries,
and finds no figures outside the limits 11 -18 per cent and 13*99 per
cent of water. Most Irish butter, which is made at higher tempera-
tures than normal, contains up to 23 or 24 per cent of water, but it
is necessary to disclose this fact when selling it.

Determination of Non-fatty Solids. — The portions of the sample
used for the determination of water may be used for this determina-
tion. The dried residue should be repeatedly exhausted with warm
petroleum ether, the liquid poured off each time, and when no further
fat is extracted the non-fatty solids, consisting of lactose, casein, and
mineral matter, are dried and weighed. If it be necessary to determine
the casein separately, it may be done by washing the fat-free residue
several times with water acidulated with acetic acid. Nearly pure
casein is left behind. Accurate results are obtained by a determina-
tion of the nitrogen by the Kjeldahl process, and multiplying by 6'37.

Determination of Mineral Matter. — This, owing to its small amount,
should be determined on the ether-insoluble residue of 10 grms. of
the butter.

Great care must be taken not to ignite at too high a temperature
lest sodium chloride should volatilize.

To determine the common salt present, the half- charred ash should
be repeatedly extracted with hot distilled water and the sodium
chloride determined either by titration or gravimetrically as silver
chloride. Greater accuracy is attained by melting 10 grms. of the
butter with 10 grms. of paraffin wax and well shaking the mixture
with 50 c.c. of hot water acidulated with 1 c.c. of citric acid. The
cake is well washed on cooling and the sodium chloride determined in

BUTTER. 91

the liquid. No standard can be laid down for the amount of salt. It
is a matter of taste. Bell found in 113 samples amounts varying
from 0-4 to 9-20 per cent. If the non-fatty solids are, apart from the
salt, not more than 1 to 1*5 per cent no further examination of these
solids is necessary. An excess of non-fatty solids will at once call for
special examination when starchy matters, etc., may be looked for.

Determination of the Fat. — Generally the difference figure will
give the amount of fat with sufificient accuracy ; but if a direct deter-
mination be required the ether extract is evaporated and the residue
weighed. The examination of the butter-fat is described later (p. 95).

Added Colouring Matter. — Martin ("Analyst," xii. p. 70) gives
the following details for the detection of added colouring matter to
butter. He uses a mixture of 2 volumes of CS.^ and 15 volumes of
alcohol. Twenty-five c.c. of this are shaken with 5 gi-ms. of the
butter, and after separation into the layers, the lower one will consist
of the carbon bisulphide with the fat in solution, and the upper one
will consist of alcohol containing artificial colouring matter if present.
The following colours will react as indicated : —

Saffron. — If saffron be present, the alcoholic liquid will be coloured
green by nitric acid and red by hydrochloric acid and sugar.

Turmeric. — Ammonia will turn the alcohol brown. Turmeric will
also be detected by evaporating the alcohol, and boiling the residue
with a few c.c. of dilute boric acid solution, and soaking a strip of
filter paper in the liquid. On drying this will assume the usual red
coloration, turning olive green on treatment with potash.

Goal- tar Dyes. — These may be detected by boiling wool fibres in
the alcoholic extract diluted with water and acidulated with a few
drops of HCl.

The following special tests should be used for certain colouring
matters : —

Carrotin. — As this is more soluble in CSg than in alcohol, the
following process should be used. Fifty grms. of the sample are
melted, and 5 to 10 grms. of powdered fuller's earth stirred in. After
well stirring the earth is allowed to settle and the warm fat poured
off, 20 c.c. of benzol added, and after stirring, the liquid is decanted
through a filter. This process is repeated until all the fat is removed
and the precipitate washed on the filter with benzene. If the benzene
be evaporated, and the residue shaken with alcohol containing a drop
of dilute ferric chloride solution, and CS.,, the alcohol will be coloured
yellow, if carrotin be present.

Amiatto. — Two to three grms. of the fat, freed from water and non-
fatty solids, are warmed with a 2 per cent solution of sodium hydroxide.
After well stirring, pour the mixture on to a wet filter in a hot funnel.
In the presence of annatto the filter paper will absorb much of the
colour, and become dyed a straw colour. The filtrate should be re-
turned several times, if necessary, to the funnel, in order to thoroughly
extract the melted fat. If the paper, after drying, turns pink when
treated with a drop of stannous chloride solution, annatto is certainly
present.

A confirmatory test for annatto is to soak a filter paper for twenty-

92 FOOD AND DKUGS.

four hours in the solution rendered alkaline with a little Na^COg. The
paper is stained brown in the presence of annatto, the colour changing
to pink by the action of HCl.

CorneUson recommends the following process for the detection of
artificial colouring matter (" Journ. Amer. Chem. Soc." 1908, 1478).
Ten grms. of the dry filtered fat are shaken well in a separator wdth
10 to 20 c.c. of glacial acetic acid. At about 35", the fat will separate
almost completely, and the clear acid is drawn off. Natural butter
gives a colourless liquid, which is unaltered by the addition of nitric
or sulphuric acid. The acid extracts of butters containing annatto,
turmeric or carrotin are yellow in colour, changing to pink — especially
with turmeric — by the addition of sulphuric acid. If methyl-orange
be present it will respond to this reaction.

Leeds ("Analyst," xii. 150) dissolves 100 grms. of butter in 300

c.c. of petroleum ether (specific gravity = 0*638) in a separator, drains

off the curd and water, and washes several times with water. The fat

solution is kept at 0° for twelve to fifteen hours, so that the greater

part of the solid glycerides crystallize out. The liquid is poured off

N
and shaken with 50 c.c. of ^ alkali to remove the colouring matters.

N
The aqueous layer is drawn off and exactly neutralized by — hydro-
chloric acid, until a drop is just acid to litmus. The colouring
matters are precipitated, contaminated with a trace of fat. The pre-
cipitate is dissolved in alcohol and a few drops tested with the reagents
when the reactions given in the table on opposite page will be observed.
The usual preservative, when any is added, used for butter is boric
acid or borax. A Departmental Committee on Food Preservatives in
1901 recommended that the only preservatives permitted to be used
in butter or margarine should be boric acid or mixtures of boric acid
and borax, and that no more than 0*5 per cent expressed as boric acid
should be used. In general the methods described under milk may
be adapted to the detection of preservatives in butter. In practice,
however, boric acid, and, according to Hehner, sodium fl.uoride, are
the only preservatives commonly to be found. For the determination
of boric acid, 25 grms. of butter are mixed with 25 c.c. of a solution
containing 6 grms. of lactose and 4 c.c. of normal sulphuric acid per
100 c.c. The mixture is placed in a water oven till the fat is just
melted, and is then well stirred. The aqueous liquid is allowed to
settle, and 20 c.c. are drawn off, a few drops of phenol-phthalein added,

and the liquid is titrated with ^ sodium hydroxide till a faint pink

A
colour appears: add 12 c.c. of glycerine and again titrate till a pink
colour appears. The difference in c.c. between the two titrations, less
the amount of alkali required (as shown by a blank experiment) by
the 12 c.c. of glycerine, is multiplied by 0-031. This gives the
amount of boric acid in 20 c.c. of the liquid. So that this value
multiplied by IgO + pei^^cent of water in the butter ^.jj ^.^^^ ^^^

actual percentage present.

BUTTER.
Reactions of Colouring Matters.

Colouring Matters.

Concentrated
H2SO4.

Conc^entrated
HNO,.

HaSO^ - HNO3.

Concentrated
HCl.

Annatto"

Indigo blue,
changing to violet

Blue becoming

colourless on

standing

Same '

- No change, or
only slight dirty
yellow and brown

Annatto +

decolorized

butter

Blue, becoming
green, and slowly
changing to violet

Blue, then
green and
bleached

Decolorized

No change, or

only slight dirty

yellow

Turmeric

Pure violet

Violet

Violet, changing

to original colour

on evitporMtion of

HCl

Turmeric +

decolorized

butter

Violet to purple

Violet to reddish-
violet

Same

Very tine violet

Saffron

Violet to cobalt

blue, changing to

reddish-brown

Light blue

changing to light

reddish-brown

Same

Yellow, changing
to dirty yellow

Saffron +

decolorized

butter

Dark blue
changing quickly
to reddish-brown

Blue, through
green to brown

Blue, quickly
changing to purple

Yellow becoming
dirty yellow

Carrot

Umber brown

Decolorized

Do. with NO2

fumes and odour

of burnt sugar

No change

Carro r +

decolorized

butter

Reddish -brown to

purple similar to

turmeric

Yellow, and
decolorized

Sime

Slightly brown

Marigold

Dark olive green,
permanent

Blue, changing

instnntly to dirty

yellow green

Green

Green to yellowish -
green

Safflower

Light brown

Partially
decoloriz&d

Decolorized

No change

Sudan orange

• Pink

Pink

Martins yellow

Pale yellow

Yellow, reddish

precipitate.

Magenta at margin

Yellow

Yellow, precipitate

treated with NH,

and ignited ;

deflagrates

Victoria yellow

Partially decolor-
ized

Same

Same, colour
returns on neu-
tralizing with
NH,

Richmond and Harrison (" Analyst," xxvii. 179) recommend using
25 grms. of butter and enough water to make, with the water ah^eady
present in the butter, 25 c.c. Ten to 15 c.c. of chloroform are then

then titrated with standard ( — ) alkali, after the addition of glycerol,

94 FOOD AND DRUGS.

added and the contents of the cylinder warmed, well shaken and
allowed to separate. A measured portion is drawn off (each c.c. con-
tains the boric acid of 1 grm. of the butter), made alkaline, evaporated,
ignited, and the ash thoroughly extracted with hot water. The solu-
tion is rendered neutral to methyl orange, boiled to expel CO., and

•NX

N
with phenol-phthalein as indicator. One c.c. of - NaOH = 0-0124

grm. of boric acid.

Fluorides are detected, as recommended by Hehner, by separating
the aqueous liquid from 50 grms. of butter, adding a little calcium
chloride, boiling the liquid, and adding excess of Na^COg to precipitate
calcium compounds. The precipitate is collected, washed, ignited and
treated with hot dilute acetic acid. The insoluble residue is collected,
ignited, and treated with strong H^SO^ in a platinum crucible, covered
with a waxed glass on which a mark has been scratched. The crucible
is stood on a sand bath for two hours, and in the pressure of fluoride,
the glass will be distinctly etched.

It will be convenient to briefly discuss margarine, before passing
on to the examination of butter-fat.

Various names have been used for butter substitutes, but to-day
they are all covered by the word margarine, which is defined by sec-
tion 3 of the Margarine Act, 1887 (50 and 51 Vict. c. 29) as fol-
lows : —

"The word margarine shall mean all substances, whether com-
pounds or otherwise, prepared in imitation of butter, and whether
mixed with butter or not, and no such substance shall be lawfully sold,
except under the name of margarine, and under the conditions set
forth in this Act." By section 8 of the Sale of Food and Drugs Act,
1889, no margarine may be sold the fat of which contains more than
10 per cent of butter fat. This wholesome restriction is in the in-
terests of the poorer classes, who may know that, whatever assurances
the vendor might give, they can never get legitimately more than 10
per cent of butter fat in margarine.

The manufacture of margarine is to-day an enormous industry,
and as the fats usually employed are now of a perfectly wholesome
nature, there is no doubt that the industry has benefitted the poorer
classes to a considerable extent. In this country a very large propor-
tion of the margarine consumed is manufactured from the fat of cocoa-
nuts. In the United States animal fats are more largely employed,
together with such oils as cotton seed, arachis or' sesame. It is not
necessary to discuss the manufacture of margarine : the fat after suit-
able treatment is churned up with water, a little colouring matter,
sometimes a little milk or butter, salt, etc., etc., until the proper con-
sistency and the desired flavour are obtained. The actual composi-
tion of margarine is, of course, very varied. Many samples are made
up, merely with regard to the flavour and appearance of the finished
product. But the fact that cocoanut fat contains a large proportion of
the glycerides of volatile fatty acids has caused a certain class of

BUTTER. 95

manufacturer to adjust ,his formulie so that the composition of his
margarine may be such as to yield analytical results similar to those
of butter. It cannot be denied that the use of cocoanut fat has
caused a large increase in the adulteration of commercial butter.

The analyst will have frequently to decide (1) whether butter con-
tains any margarine, (2) whether margarine contains more than 10
per cent of butter fat.

In regard to the analysis of margarine, there is nothing to be added
to the details given for pure butter, so far. It is only in regard to the
characters of the fat, that the analyst will be able to base an opinion
as to the presence or absence of margarine in samples of butter.

The Analysis of Butter Fat.

The' fat should be separated from the water, and non-fatty solids
by allowing the butter to remain melted for a short time, when the
fat is poured off through a dry filter in a hot water funnel. The
following determinations are then necessary.

Specific Gravity. — It is usual to determine this value at 100° F.
(37*8° C), taking the specific gravity of water at that temperature as
unity. At this temperature the specific gravity of genuine butter,

qrr.QO p

D ' varies between the limits 0*910 to 0-9135. Some observers

o7"o

prefer to take the specific gravity at 100° C, and adopt water at

100° c
15-5° C. as unity. The specific gravity, D — ^-^ varies from

15*o

0*8668 to 0*8705. Most adulterants lower the specific gravity, but
cocoanut oil — which is the basis of much of the commercial margarine
of to-day — has a rather higher specific gravity (up to 0*9175) at
017.0° n

5-^. Judicious mixtures, however, can be prepared with the same

37*8

specific gravity as butter fat, so that this, as all other tests for butter,
will only give an indication to be judged in combination with other
results.

Melting-'point. — The melting-point of butter fat, determined in
very thin tubes, varies from 29° to 33°, rarely up to 34°. The insoluble
fatty acids separated in the usual manner melt at 39° to 44° or rarely
up to 45°.

Iodine Value. — The iodine value of butter fat (see p. 628) as de-
termined by Hiibl's solution usually varies from 28 to 34, but 26 to 28
are figures recorded by trustworthy observers. Cocoanut fat has a
much lower iodine value, but such fats as cotton seed, with very
high iodine values, can be used in small quantities to adjust this figure.

Saponification Value. — This figure varies, for pure butter fat, be-
tween the limits 220 to 234. Here again mixtures of other fats are
easily made which have the same saponification value as that of pure
butter fat.

Avi-Lallemant (" Zeit. Unter. Nahr. Genuss." 1907, 14, 317) re-
commends the following method in the examination of the fatty acids
of butter. Two grms. are saponified with alcoholic KOH, the liquid

FOOD AND DRUGS.

neutralized with - HCl, alcohol expelled, the soap dissolved in boiling

A
water, and the boiling solution (150 to 180 c.c.) treated with 50 c.c. of
a solution of 2*5 BaCU per litre. The flask is left on the water bath
for fifteen minutes, the contents cooled, made up to 250 c.c. and
filtered. The baiium remaining in solution is precipitated from 200
c.c. of the filtrate, which is acidified with HCl, by H._,SO^ and weighed
as BaSO^. The amount of barium so found calculated as BaO is
multiplied by 1'25, since only 200 c.c. of the filtrate were used, and
this subtracted from the amount of BaO originally used, calculated to
1 grm. of fat, gives the amount of BaO which has combined with the
fatty acids to give insoluble barium salts, i.e. = the " insoluble baryta
value ". If the saponification value of the fat be calculated into mgr.
of BaO, this minus the insoluble baryta value — is the " soluble baryta
value ". The following values are obtained from pure butter and cer-
tain other fats : —

Insoluble

Soluble

BaO Value.

Butter

247-4 0 254-8

50-8 to 76-7

Sesame oil

2.51-9

3-3

Cotton-seed oil

256-9

6-6

Cocoanut oil

296-5 to 299-2

54-1 to 57-6

Lard

257-4

7-6

Beef tallow

264-1

6-2

Befractive Values. — The absolute refractive index of a given sub-
stance is a far more more scientific figure than any empirical values,
such as the scale readings of instruments known as butyro-refracto-
meters, etc. At the same time these instruments have come into use
to a very large extent, and the values recorded by them are of con-
siderable value, and being well established call for general recog-
nition.

The following are the refractive indices of pure butter fat and cer-
tain other oils which may enter into the composition of butter substi-
tutes, as determined by the author : —

Butter fat at 2.5° C. .

. 1-4587 to 1-4615 {20 samp

OS)

„ 38=

. 1-4540 „ 1-4569 (20

CoeoaDut ,,38°

. 1-4.500 „ 1-4515 (10

Lard ,,38°

. 1-4490 „ 1-4505 ilO

Cotton seed ,,38°

. 1-4660 „ 1-4680 (10

Huet „ 38°

. 1-4605 „ 1-4620 ( 5

The butyro-refractometer is an instrument manufactured by Zeiss,
similar to the ordinary refractometer, but graduated in arbitrary scale
divisions. It has been said that the differences in the dispersive
values of various fats is such that the critical line, seen in the refracto-
meter of butter, is colourless, whilst that of other fats is blue. In the
author's opinion this point is of no value at all, and only the quanti-

BUTTER. ^^^* 97

tative values are of importance. If a reading on this instrument be
taken at one temperature, and it is desired to correct this from an-
other temperature, 0-55^ should be added for every 1" C. by which
the temperature of observation is reduced or added for every degree
by which it is increased. According to Wollny pure butter gives a
scale reading of 49-5 to 54 degrees at 25° C. At 38° the readings
vary from 42 to 46°. The value at 25°, however, rarely exceeds 52-5",
and at 38°, 44-5°. Cocoanut fat has a value 34° to 37° at 40° C.
But, of course, mixtures can easily be made having the same refractive
index as that of butter fat. A butter showing values outside the
limits 49 to 54 at 25° may be condemned as adulterated. The table on
pages 98, 99 gives the correspondence betw^een the refractive index and
the butyro-refractometer scale of reading. The figures given across
the table are those in the first place of decimals corresponding to
the figures in the fourth place of decimals of the refractive indices.
An asterisk in these figures indicates that the figure has to be added to
the scale reading of the next lower line.

Another instrument used in this respect is the oleo-refractometer,
devised by Amagat and Jean. It is, like the above, an instrument
graduated on a purely empirical scale. In this instrument a zero
point exists, and readings to the right of this are designated +
whilst those to the left are called - , Pure butter fats give a reading
of - 26° to - 34° with an average of about - 30° ; most vegetable fats
give + readings, whilst cocoanut oil behaves like animal fats and
gives a reading up to - 58°.

The refractive index of the fatty acids (insoluble) of butter varies
from 1-4370 to 1-4390 at 60° C, whereas that of the insoluble fatty
acids of cocoanut oil never rises above 1-4301 at the same tempera-
ture.

The Volatile Fatty Acids. — The most valuable determination in
connexion with the examination of butter fat is that of the volatile
fatty acids, although even the value of this must not be over-estimated,
on account of the fact that cocoanut fat contains so high an amount,
of volatile acids. Various modifications of the Reichert process,
exist, but as the following details of working have been agreed
upon by. official and unofficial analysts, they may be taken as practi-
cally official for the purposes of the Food and Drugs Acts. It is to
be remembered that the Reichert- Wollny (or Reichert-Meissl) values
refer to 5 grms. of the fat, whilst the Reichert values refer to 2-5
grms. only. To connect the lower with the higher value a slight
correction is necessary, and it is usual to multiply it by 2-2. The
following process has taken into account the errors in the original
process as formulated by Wollny (" Journ. Soc. Chem. Ind." 1887,
831).

The details of this semi-official process are as follows : —

Five grms. of the dry fat are placed in a 300 c.c. flask (see Fig.
100) and 2 c.c. of an aqueous solution of NaOH (1 grm. in 1 c.c), free
from carbonate, are added, with 10 c.c. of 92 per cent alcohol (by
volume). The mixture is heated under a reflux condenser on a water
bath for fifteen minutes. The alcohol is then removed by heating on
VOL. I. 7

98 FOOD AND DKUGS.

Table of Kefractive Indices and Kefractometer Numbers.

4th Decimal ot Refractive Index.

Ref.
Index.

Scale
No.

7 .

Scale
No.

1-422

1-423

•0

1-423

1-424

1-425

1-426

1-427

»4

1-427

1-428

•2

»6

1-428

1-429
1-430

8
10

*0
3

♦4

•5

1-429

*0
*3

10
11

1-430
1-431

1-431

1-432

1-433

•4

*'l

**o

1-433

1-434

1-435

♦1

•2

1-435

1-436

1-437

♦1

•3

1-437

1-438

1-439

♦4

1-489

1-440
1-441

23
24

0
3

2
5

8
6

1-440

♦1

1-441

1-442

♦5

1-442

1-443

1-448

1-444

1-445

1-446

1-447

•0

1-447

1-448

♦•0

1-448

1-449

•0

1*5

1-449

1-450

8
2

•0
3

♦1

*2
6

*6
9

*7
*0

•8
*2

•9
*3

1-450

1-451

"38

1-451

1-452

♦4

1-452

1-453

**o

**1

1-453

1-454

1-455

♦2

**o

1-455

1-456

1-457

1-458

1-459
1-460

♦2

1-459

♦2

1-460

BUTTEK.

Table of Eefractive Indices and Eefractive Numbers —
Co7iti7iued.

4th Decimal of Refractive Index.

Ref.
ludex.

Ref.
Index.

Scale

No.

1-461

1-462

1-463

1-464

♦2

»3

1-464

1-465

♦2

1-465

1-466

1-467

1-468

1-469

♦1

**1

**2

1-469

1-470

1-471

68,

1-471

1-472

* 0

1-472

1-473

1-474

♦*o

**1

1-474

1-475

1-476

1-477

**1

*»2

1-477

1-478

**o

1-478

1-479
1-480

81
82

1-479

**1

**3

**5

1-480

1-481

•3

**0

**2

1-481

1-482

**0

1-482

1-483

1-484

1-485

*»o

**2

**4

1-485

1-486

**o

**2

1-486

1-487

**o

1-487

1-488

1-489

1-490

100

♦1

**0

**2

**4

**6

100

1-490

1-491

102

**0

**2

**4

102

1-491

1-492

104

105

—

— .

—

the water bath, and then 100 c.c. of hot water (which has been well
boiled for ten minutes to remove GO,,) are added and the flask warmed
until the soap is dissolved. Forty c.c. of normal HoSO^ are then added
and a few fragments of pumice stone, and the flask immediately
connected with the condenser. The flask is supported on a circular
piece of asbestos 12 cms. in diameter, having a hole in the centre 5
cms. in diameter, and is just heated so that the insoluble fatty

100

FOOD AND DHUGR.

ilcidH nu'lt, and tluMi hcate-l uutil l.li(> li((iii(l boils ami 1 lOc.c. of liquid
aro coli(K5U'd, th(^ distillation lastinj^ about thirty miuutt's. Th(^ distillate
\H shaktMj and liltorod, and 100 c.c. of tlu^ liltratc. titiatod with doci-
uonnal alkali, pluMiol-phthaltMn Iumu^' uscsd as indicator. A blank ox-
jxM'imtMit should !)(> conducted on tlio material employed, but the
amount of decinormal alkali iiMiuirtMl should not (exceed {)'A c.c. This
Hhould be deiluctetl from the r(»sult obtained, and the nund)er of c.c.
thus found, multiplii'd by M, is the U(Mchert-Wollny numbei-, which
Ik a measure of the soluble volatile fatty acids.

Tlie following is a diagram of the apparatus employed.

It is admitted on all
hands that this figure may
vary within wide limits,
and that uidess a very low
standard be adopted, occa-
sionally a genuine butter
may be condemned. The
usual limit adopted as a
minimum is from 23 to 25 ;
in this country the usual
figure is 24. Since genuine
butt(irs give figures up to
I];3 and higher, it is obvious
that this limit errs on the
side of giving the benefit
of the doubt to many
samples of butter. Cocoanut oil gives a Reicluu't-Wollny value up
to 8, hence mixtures of genuine butter with values of 28 to 30 with a
considerable amount of cocoanut fat will satisfy this accepted limit
of 24.

The majority of the modifications of tlie Heichert process are not
of much importance, but a few contributions to the question have re-
cently been made which are of considerable interest.

Blichfeldt has recently proposed (".lourn. Soc. Chom. Ind." 29,
792) the following method, depending on the dilTeronces in the amount
of soluble and insoluble silver salts of the volatile fatty acids in butter
and cocoanut fats : —

\^y this method the fat is saponified by a mixture of a()ueous potash
and glycerol, and the fatty acids are liberated by acidilication with sul-
phuric acid. The resulting mixture is distilled in a specially designed
apparatus, in which the connecting tube, condenser tube, and receiver,
are all in one place. The distillate is treated with an excess of deci-
normal soda solution and transferred to a 200 c.c. measuring flask.
The total vtilatile acids are determined by titrating back witli deci-
normal sulphuric acids. The neutral solution is now treated with ex-
cess of decinormal silver nitrate solution, and 10 per cent of solid
sodium nitrate is dissolved in the liquid in oixler to salt out all the
sparingly soluble silver salts. After making up to 200 c.c, the pre-
cipitated silver salts are filtered otT, and the excess of silver nitrate is
determined volumetrically in the filtrate in the following manner. A

Pio. 8.— Ueichort-Wollny apparatus.

J3 UTTER.

101

slight, known excess of decinorrnal sodium cliloi'ide solution is added,
and the excess titrated hack with decinornnal silver nitrate solution.
This will allow the original excess of silver nitrate to he calculated.
The volatile fatty acids thus determined as soluhle and insoluhle silver
salts are shown in the following table : —

SiihMtance,

Volatilft AfiidH iti tenn^ of
Dcjcinoniial NaOH.

Silver PriBi;li)itation in tfirinN
of Dcsclnormal Solution.

Total.

8olul)le.

Insoluble.

Total. Soluble.

luHoluble.

Butter
Cooo^^nut oil
Palm kernel oil

82
20
15

28
6
6

4
14
10

82 29
20 , 4
U 8

8
1«
12

The ratios of soluhle to insoluhle silver salts derived from butter
and cocoanut fats respectively dii'i'er considerably from one another,
and afford a ready means of determining the substances in presence
of one anothc)-. The method, however, makes no distinction between
cocoanut oil and })alm kernel oil.

Kirschner (" Zeit, Nahr. Genuss." 1905,9, 07) proposes the follow-
ing method which is itself a modification of one previously suggested
by Gensen. The distillate obtained in the Keichert-Wollny process,
is filtered as usual and 100 c.c. titrated with decinormal baryta. To
the neutral solution, 0*5 grm. of silver sulphate is added, and the
mixture shaken from time to time during one hour and then filtered.
Th(i filtrate measures just over 100 c.c. Exactly 100 c.c. of this are
placed in a Hask, 3.0 c.c. of water and 10 c.c. of dilute H.^SO^ added,
and the whole distilled until 110 c.c. have been collected. This is
again filtered, 100 c.c. of the filtrate are titrated with decinormal baryta
and the result calculated for /j grms. of fat. It is claimed by the
author that the elimination of the acids precipitable by silver accen-
tuat(is the difference between pure butter and a mixture with cocoanut
fat.

Monhaupt ("Chem. Zeit." 1909, 33, 305) claims that small
<juantities of Iwitter fat can thus be detected in cocoanut fat prepara-
tion. He gives the following figures for margarines containing from
15 to 25 per cent of cocoanut fat and 1 to 2 per cent of butter fat : —

Per cent Cocoanut Fat.

Per cent Butter Fat.

Reichert-WoUny No.

Kirschner Value.

0-50

0-24

072

0-89

0-99

0-62

0-HH

0-30

0-99

0-55

1-21

0-69

110

0-8(>

1-82

0-68

1-50

0-81

102 FOOD AND DRUGS.

He states that if 20 grms. of fat be used, more marked differences
still will be obtained. For further details, the original paper should
be consulted.

Paal and Amberger {" Zeit. Untersuch. Nahr. Genuss." 1909,
17, 1) have recommended the quantitative determination of the vola-
tile fatty acids which are precipitable by cadmium sulphate. As
cadmium butyrate and caproate are soluble in water, this method
deals mainly with the insoluble volatile fatty acids.

They use 2'5 grms. only of the fat, and after distillation of the
fatty acids from the aqueous mixture, a little alcohol is distilled through
the condenser, in order to wash away any deposited fatty acids.
After titration, the exactly neutralized fatty acids are precipitated by the
addition of 2 to 4 c.c. of a 20 per cent solution of cadmium sulphate.
The precipitate is collected, washed with 50 c.c. of water, dried at
102° for one hour and weighed. The weight in milligi-ams is called
the cadmium value of the fat. In pure butter this varies between 70
to 90, or in rare cases, due to abnormal feeding, it may rise to 100 or
just over. Cocoanut fat has a " cadmium value " of 440 to 470.

The processes of Blichfeldt, Kirschner, Paal and Amberger are
based on the observations of Polenske, who showed that the greater
part of the volatile acids of butter are soluble, whereas the reverse is
the case with cocoanut oil. Polenske's observations showed that in
thirty-one samples of butter fat, the Reichert-Meissl values of which
varied from 23-3 to 30*1, the amount of decinormal alkali required for
neutralization of the insoluble acids was only from 1*5 to 3*0, whereas
in samples of cocoanut oil, the Reichert-Meissl values of which varied
from 6-8 to 7"7, the insoluble volatile acids required from 16'8 to
17"8 c.c. of decinormal alkali.

Polenske thereupon termed the number of c.c. thus required to
neutralize the insoluble volatile acids of 5 grms. of the fat, the " new
butter value ".

The details of the process adopted by him are as follows : —

Five grms. of the butter fat are saponified by the Leffman-Beam
process, using 20 grms. of glycerine and 2 c.c. of 50 per cent NaOH
solution, in a 300 c.c. flask over a naked flame. The solution is
allowed to cool below 100°, 90 c.c. of water are added, and the mass
dissolved by warming on a water bath to 50°. The solution must be
nearly colourless, otherwise the experiment must be repeated. To the
warm solution 50 c.c. of sulphuric acid solution (2-5 per cent) are
added, and some powdered pumice stone. The flask is at once con-
nected with the condenser (in order to obtain absolutely concordant
results, the apparatus should be of the exact dimensions given in the
original paper — "Arbeit, aus dem Kaiser. Gesundheitsame," 1904,
545 — but the apparatus illustrated on p. 100 is so near to it that it
may be used in this process with confidence). The liquid is distilled
until 110 c.c. are collected in nineteen to twenty minutes. The receiv-
ing flask is now removed and a 25 c.c. cylinder substituted. The re-
ceiving flask is now stood in water at 15°, and after fifteen minutes
the consistency of the insoluble acids which are floating on the surface
noted. Pure butter acids are solid, whereas in the presence of 10 per

BUTTER. 103

cent of cocoanut oil, the acids no longer solidify. The contents of the
flask are mixed by merely turning the flask upside down several times
and then 100 c.c. are filtered off, and the Reichert-Meissl value deter-
mined by titration. The insoluble acids collected on the filter are three
times washed with 15 c.c. of water, which have been passed succes-
sively through the tube of the condenser, the 25 c.c. measuring cylinder
and receiving flask.

Finally three portions of 15 c.c. each of 90 per cent alcohol are
passed through the condenser, the measuring cylinder and the receiv-
ing flask, and each portion passed through the filter, which is allowed
to drain before the next portion is added. The alcoholic filtrate is
then titrated with decinormal alkali.

In thirty-four samples of pure butter Polenske found 1'35 to 3"0
as the insoluble volatile fatty acid figure, whereas, as mentioned above,
the value reaches to 17'8 for cocoanut oil.

Polenske claims to be able to calculate the approximate amount of
cocoanut oil present in a mixture, on the basis of an experimental
table, and assumes that each 1 per cent of added cocoanut fat in-
creases the insoluble acid value by 0-1. But this is based on two fal-
lacies, firstly that the amount of soluble fatty acids and insoluble fatty
acids in genuine butter fat are in definite ratio, and secondly, that one is
dealing with a mixture of butter fat and cocoanut oil only. Still the
process is of distinct value, but the results require careful interpreta-
tion.

Shrewsbury and Knapp (" Analyst," xxxv. 385) have proposed a
process for the determination of cocoanut fat, which certainly gives very
promising results. The fatty acids of this fat contain much lauric and
myristic acid, which are not present in most other fats, and which are
practically insoluble in water, but soluble in dilute alcohol. Their
process, which is similar to the less useful one of Vandam (" Analyst,"
XXVI. 320), is as follows : 5 grs. of the filtered fat are saponified with
20 c.c. of glycerol soda as in the Reichert process, and the soap diluted
with boiling water and at once transferred to a separator, using 200
c.c. of water in all ; 5 c.c. of H^SO^ (1 in 4 of water) are added and the
mixture well shaken for 60 seconds. The liquid is allowed to stand
for 5 minutes, and the water run off from the insoluble acids. These
are dissolved in the separator in 50 c.c. of methylated spirit, the liquid
run into a flask and heated, adding a fragment of pumice to prevent
bumping, 36 c.c. of cold water are then poured into the separator, and
the boiling alcoholic solution poured into it. The mixture is poured
back into the flask to wash it, again transferred to the separator,
shaken for 30 seconds, left for 3 minutes, and 70 c.c. of the clear
solution run off and titrated with decinormal soda, using phenol-
phthalein as indicator.

The figures thus found for the acid value of the 70 c.c. so obtained

N
vary from 23*6 to 31*2, with an average of 27"7 c.c. of — alkali for

pure butters. Cocoanut oil, on the other hand, gives a value of about
163 c.c. of ^ alkali. By adopting the values of 32 for pure butter

104

FOOD AND DRUGS.

and 16-3 for cocoanut oil, a fair estimation of the amount present is
obtained.

Valentas Test. — After melting the butter fat, filter at as low a
temperature as possible, then dry still further by again filtering through
a dried filter-paper. Weigh 2-75 grms. of the fat and place in a
stoppered test tube, and add 3 c.c. of 99*5 per cent acetic acid.
Place the tube in a beaker of water, gradually heating the water until
the solution becomes clear on shaking the tube. Notice carefully the
temperature. The following figures represent the temperatures for
butter fat and margarine respectively : —

Margarine
Butter fat

Maximum.

Minimum.

Average.

39°
97°

29°
94°

36°
95°

In order to avoid any mistake always test the acetic acid first on a
sample or samples of genuine butter fat. Jean prefers to determine the
amount of acetic acid dissolved. He weighs about 8 c.c. of the fat
into a graduated test tube 1 cm. in diameter, which is placed in water
at 50° C. He then removes the excess of fat by means of a pipette
until the fat measures 3 c.c. at 50° C, and adds 3 c.c. of glacial acetic
acid (specific gravity 1'0565) which has been measured at 22°.
The contents are then warmed for a few minutes, and, after inserting
a cork in the test tube, well shaken. The tube is then immersed in
the water at 50° C, and the amount of undissolved acetic acid deter-
mined. Nine samples of butter averaged 63-33 per cent of acetic acid
dissolved ; margarines vary from 27 to 32 per cent. The turbidity
test and the amount of acetic acid dissolved can obviously be done on
the same sample at the same time.

M. Crismer suggests a turbidity test bearing the name of " the
critical temperature of dissolution ". This differs from the Valenta
test, but is doubtless of the same value. He proceeds as follows; 0*5
c.c. of filtered fat is weighed into a tube of small diameter ; 0*75 c.c.
of alcohol, specific gravity 0*7967 (containing 0*9 per cent of water) is
added. The tube is hermetically sealed and fastened to the bulb of a
thermometer by means of a platinum wire. The bulb and wire are
then placed in a small sulphuric acid bath, and the temperature is
raised gradually until the meniscus separating the two layers becomes
a horizontal plane. The thermometer and tube must now be taken
out of the bath and the tube well shaken to mix the two liquids to-
gether. They are again immersed in the bath and shaken all the
time whilst the temperature is allowed to fall. Immediately a marked
turbidity is to be seen, the temperature is noted. Genuine butters
gave from 51° to 57° C., margarines varied from 69° to 78° ; with
alcohol of specific gravity 0-8195 containing 8-85 per cent of water
butters give 98° to 105-5° C. and margarines 109° to 124° C.

Lewkowitsch considers that if an absolute test for the presence of a
vegetable fat in butter be required, when other tests fail to decide the

BUTTER. ^^^r 105

question, the phytosteryl acetate test may be successfully employed and
gives decisive results (p. 630). A very small amount of paraffin wax
will, however, interfere with this test.

The amount of insoluble fatty acids in pure butter fat, which is
■determined by saponification and separating the fatty acids in the
usual manner, should not exceed 89'5 per cent and is rarely more than
88 per cent. Cocoanut oil contains a low^ percentage of insoluble
fatty acids, but nearly all other oils contain considerably more than
90 per cent, cotton-seed oil containing as much as 95 to 96 per cent.
A butter yielding over 90 per cent is undoubtedly adulterated.

In America numerous empirical methods of examining butter are
practised, such as the foaming or failure to foam of the sample when
treated over a flame ; the ease with which the fat mixes with milk or
not, etc., but these are not of any serious value, and need not be dis-
cussed as methods of analysis.

Microscojncal Examination of Butter. — With an ordinary light,
and a low power of from about 120 to 150 diameters much informa-
tion can be gained from a sample. A small portion on the edge of a
knife blade is placed on the glass slide and gently pressed into a thin
film by means of the cover glass. The difference betw^een genuine and
renovated butter is easily seen. The fat film of fresh butter is much
more transparent than that of renovated, and the curd of the genuine
butter fat is much more finely divided throughout the mass and
the field is much more even than that of the renovated, the latter
often showing large and opaque patches of curd throughout the field.

When a renovated butter sample, i.e. a " process butter" — rancid
butter, melted and made palatable by blowing steam through it, etc.,
mounted as above — is examined by reflected light, obtained by turning
the microscope mirror in such a way as not to transmit light through
the instrument, a very dark and scarcely perceptible field is to be seen,
whilst the above-mentioned opaque patches of curd are distinctly seen
as white masses against a dark background.

With Polarized Light. — The crystalline structure of fat once
melted and afterwards cooled is easily seen by the microscope which,
as has already been stated, is useful in determining whether or not a fat
has been melted, especially when examined by polarized light. This
fact has been made use of for a long time in the identification of butter
and oleomargarine by the microscrope.

Pure butter, not previously melted, should show no crystalline
structure when seen by polarized light between crossed Nicols under a
low magnification, and it should be uniformly bright throughout. If
the selenite plate is used, there should be an evenly coloured field
with fat crystals entirely absent. With process butter or oleomar-
garine, previously melted then cooled, the crystalline structure presents
a marked appearance, which is more or less mottled when viewed by
polarized light, and if the selenite plate is used there is quite a play
of colours.

There are various circumstances which may affect the reliability
of the polarized light test. These distinctive features, already de-
scribed, are particularly obvious in cold weather. The appearance of

106 FOOD AND DRUGS.

pure butter is quite blank, whilst oleomargarine has a much more mottled
appearance than renovated butter. An expert cannot always detect
these well-defined points of variation in practice. Sometimes pure
butter will show a somewhat mottled field owing to a slight crystalliza-
tion at some previous time. In the summer, for example, these dis-
tinctions between pure and adulterated butter as indicated by polarized
light, are not as reliable when the butter easily melts at ordinary tem-
perature, as they are in winter. It is necessary, therefore, that both
the collector of samples and the analyst should keep the samples to be
examined from melting under ordinary conditions.

• (1) (2)

Fig. 9. — Unmelted and renovated butter under polarized light.

The above sketches illustrate the appearance of (1) unmelted and,
therefore, presumably pure, butter, examined under polarized light,
and (2) melted and solidified fat, in light and dark fields, under
polarized light. These are presumably either renovated or adulterated
butter.

LAED.

At one time, lard was regarded as the fat rendered from specified
parts of the hog, but as the fat is now rendered in enormous quantities
from other parts than the kidneys and the bowels— which used to
furnish the fat known as lard — the term lard has now a rather wider
signification.

According to Lewkowitsch, the following grades of lard are recog-
nized in the American packing trade.

(1) Neutral lard No. 1. The fat rendered between 40° to 50° in a
perfectly pure state from the leaf (kidneys and bowels). It is practi-
cally neutral.

(2) Neutral lard No. 2. This is the fat of the back rendered under
the same conditions.

(3) Leaf lard. On subjecting the residue not rendered for neutral
lard to steam heat under pressure, the leaf lard of commerce is
obtained.

(4) Choice lard. This is defined by the Chicago Board of Trade
as made from the leaf and trimmings only, either steam or kettle
rendered (i.e. in open steam-jacketed vessels). Neutral lard may
have first been rendered.

(5) Prime steam lard. This is made from any part of the hog and
is rendered in tanks by the direct application of steam.

LARD.

107

(6) A still lower quality is rendered from the whole of the ab-
dominal viscera.

The usual American standards for " standard lard" and " standard
leaf lard," are lard, and leaf lard respectively, containing at least 99
per cent of fat (including fatty acids), and having an iodine value not
above 60.

Lard consists of the glycerides of lauric, myristic, palmitic, stearic,
oleic and linoleic acids. According to Twitchell the composition of the
mixed fatty acids of lard is as follows : —

Per cent

Linoleic acid 10-06

Oleic acid 49-39

Solid fatty acids .......... 40-55

Lewkowitsch has summarized the following comparisons between
European lards and American lards made from various portions of the
hog (" Oils, Fats, and Waxes," Vol. II, p. 781, 3rd edition).

European Lards.

Iodine No. of

Fat from

Sp.gr.

M. Pt.
of Fat.

M. Pt. of

Fatty Acids.

Fat.

Fatty Acids.

Free Acids
as Oleic.

Back

Kidney

Leaf

0-8607
0-8590
0-8588

33-8°
43-2°
44-5°

40°

43-2°

42-9°

Per
cent
60-6
52-6
53-1

Per
cent
61-9
54-2
54-4

Per

cent

0-152

0-163

0-360

North American Lards.

Specilic gravity

Iodine No.

Melting-point.

Butyro-refractoraeter No.
at 40' C.

Per cent

Head

0-8637

66-2

44-8°

52-6

0-8629

66-6

44-8°

52-5

0-8631

65-0

45°

Back

0-8611

61-5 .

48-5°

52-4

0-8621

48-5°

51-8

0-8616

65-1

46°

51-9

Leaf

0-8637

62-2

45°

51-4

0-8615

44°

50-2

0-8700

44-5°

Foot

0-8589

68-8

40°

44-8

0-8641

68-4

45°

51-9

Ham

0-8615 .

66-6

44°

51-9

0-8628

68-3

44-5°

108

FOOD AND DKUGS.

The following are the average constants for a large number of genu-
ine lards. The determinations were made in the author's laboratory
except when otherwise mentioned : —

Lard.

Sp. gr. at

Melting-
point.

Saponification
Value.

Iodine
No.

Refractive
Index at 60= C

Butyro-refractometer
No. at 40°.

Observers.

0-859 to 0-862

— 0-861

0-859 „ 0-864

40 to 47°

194 to 197

195 „ 196

54 to 68
53 „ 77
59 „ 68

1-4527 to 1-4541

45 to 53

Parry

Lewkowitsch

Dennstedt

Fatty Acids from Lard.

'f'^V

M. Ft.

Mean Molecular
Weight.

Iodine
No.

59 to 66

Refractive
Index at 60° C.

0-837 to 0-840
0-836 „ 0-841

44°

42 to 45°

278
275

1-439 to 1-441

Allen
Parry

Genuine lard should contain under 0*5 per cent of unsaponifiable
matter, which is principally cholesterol.

Lard is ofi&cial in the British Pharmacopoeia, the melting-point
being given as about 3 7 "8° : —

Lard is adulterated to a considerable extent. Beef fat, cotton-seed
stearin and maize oil are the principal adulterants. The so-called
compound lard is sometimes a mixture of lard stearin from which the
more liquid portion of the oil has been removed and sold as lard oil,
with beef stearin and cotton-seed or maize oil. Frequently no lard at
all is present.

Interpretation of Analytical Results. — Any lard having a specific
gravity at ^""°/i50 above 0*864: must be regarded with great suspicion,
and is probably adulterated. The following table shows the effect of
individual adulterants on the specific gravity of lard, but it must be
remembered that mixtures are easy to make up which have the same
specific gravity as pure lard.

Fat.

Specific gravity at ^"oo/j^o.

Observers.

Pure lard
Lard stearin
Cotton-seed oil .
Cotton-i^eed stearin
Beef stearin
Arachis oil
Cocoanut oil

0-859 to 0-862

0-857 „ 0-858

0-868 „ 0-8725

0-8658 „ 0-8662

0-857

0-8673

0-8736

Parry
Parry

Allen — Pattinson

Parry

Pattinson

Allen

LARD.

10^

The iodine value must be regarded with caution, as reliable ob-
servers have shown that occasionally values well above the generally
accepted limits may be found. Lewkowitsch considers that 50 to 66
is a fair range of values, and that outside these figures, a sample
should be regarded with suspicion, or at least as inferior lard. The
average iodine absorptions for various adulterants are as follows : —

Beef stearin .
Beef tat .
Mutton fat
Cotton-seed oil
Cotton-seed stearin

20
38
40
110
90

Cocoanut stearin 4 to 6

The refractometric examination, like all the other results, must be
regarded in conjunction with all the other figm^es.

According to Mansfeld, Bomer, and Dennstedt and Voigtlander^
the following values cover a large number of authentic samples : —

Fat.

But)

^ro-refractometer No. at 40°

Lard from back 50-2 to 52-4

„ ,, head .

52-2 „ 52-6

„ „ leaf .

50-2 „ 52

„ outer part of leaf

50-7

,, from belly .

50-4

„ „ intestines

„ foot .

44-8

., „ ham .

49-1 „ 53

Beef tallow

Horse fat .

53-7

Cocoanut fat

•

35-5

Cotton-seed oil .

Generally speaking a low refractometer number indicates the pre-
sence of beef stearin or cocoanut fat, whilst a high value indicates
cotton-seed stearin.

The deviations in Amagat and Jean's instrument show wider
differences still. The following figures are due to Dupont : —

Pure lard
Lard stearin .
Beef stearin .
Cotton-seed stearin
Arachis oil
Cocoanut oil .

-12-5°
-10 to
-34°

+ 25°
+ 5°
-54°

Vegetable oils may, of course, be indicated by the above tests, ex-
amined collectively, but the only certain proof of the presence of a
vegetable fat, when the figures have been adjusted by judicious mix-
tures, is a positive response to the phytosteryl acetate test (p. 630).

If a vegetable oil has been indicated, however, by a high iodine
value, the following oils should be looked for : —

Arachis Oil. — This should be searched for by any modification of
Renard's test (see olive oil, p. 113). The amount of arachis oil may
be calculated from the amount of arachidic acid found.

Sesame Oil. — This is detected by the hydrochloric acid and furfural
test (p. 116).

110 FOOD AND DKUGS.

Maize Oil. — There is no direct test for this, but in the absence of
arachis, se3ame and cotton- seed oils, a high iodine value and a low
melting-point of the fatty acids would indicate maize oil.

Cotton-seed Oil or Stearin. — The Becchi and Halphen tests (see
p. 115) may be applied, but neither does a negative reaction prove
the absence of cotton-seed oil, nor does a positive reaction prove its
presence. It has been abundantly proved that lard obtained from hogs
fed on cotton-seed cake yields a cotton-seed oil reaction, and as it
cannot be suggested that such feeding is objectionable, no great reli-
ance must be placed on these reactions.

If a positive reaction, combined with a high iodine value, be obtained,
cotton oil may almost safely be presumed to be present. The final test
to decide with certainty is the phytosteryl acetate test referred to above.

A microscopical examination of lard may yield useful results. Five
grms. of the fat, free from moisture, should be dissolved in 20 c.c. of
a mixture of 90 per cent ether and 10 per cent alcohol and the liquid
allowed to stand at about 20° over night. In the event of no crystals
forming by the morning, the stopper should be removed from the tube
and a plug of cotton wool inserted, so as to allow a slow evaporation
of the solvent. If the crystals form rapidly, they must be redissolved
and redeposited. The mother liquor is decanted and the crystals ex-
amined under the microscope. Such crystals from pure lard usually
form oblong plates, either alone or in bunches, cut off" obliquely at one
end. Beef stearin, on the other hand, forms curved tufts of thin
needles, often resembling the letter/ in shape. The ends are sharp,
and the needles are often arranged in fan -shaped clusters. It is
necessary to use a high power in order to show the plate-like struc-
tures of lard crystals, as under low powers they may appear as
needles. Under low powers also, the lard crystals may appear curved,
but under a high power, this will be found due to the fact thai; several
crystals are joined at various angles. Stock (" Analyst," xix. 2)
compares the crystals deposited from ether with those from two
standard sets of mixtures, the first consisting of pure lard melting at
34° to 35°, containing 5, 10, 15 and 20 per cent of beef stearin melting
at 56°; the second consisting of pure lard melting at 39° to 40°, with
5, 10, 15 and 20 per cent of beef stearin melting at 50°. Three c.c. of
the melted fat are mixed with 21 c.c. of ether in a 25 c.c. stoppered
cylinder, and the mixture warmed to 20° to 25°. If the sample melts
at about 35°, 3 c.c. of each of the series of the first set of standards
are treated in exactly the same manner. If the sample melts at
about 40°, the second series is employed instead. The cylinders are
now cooled down to 13° and kept for twenty-four hours. The ether
is decanted and 10 c.c. of pure ether at 13° C. is added in each case.
After standing at 13° for twenty-four hourS; the contents are poured
off into shallow beakers, the ether poured off and the crystals
allowed to dry for fifteen minutes at 10° C, and weighed. The
approximate amount of adulterant is arrived at by comparing the
weight obtained with that of the type sample nearest to it. The
crystals should then be examined microscopically in comparison with
the type samples.

SUET— OLIVE OIL.

Ill

SUET.

The unrendered fat of various animals, principally oxen and sheep,
is known as suet. The corresponding rendered fat, freed from cellular
tissue, etc. is know as tallow.

Suet is usually sold as beef suet or mutton suet, and being handled
in its natural state gives little opportunity for adulteration.

The principal difference between beef and mutton suet is thajt
mutton suet is richer in stearin than beef suet. The following are
the principal characters of beef and mutton suet : —

Beef Suet.

Mutton Suet.

Specific gravity at 15° ...

0-942 to 0-958

0-937 to 0-963

„ ,,^""7X5° . . •

0-860 „ 0-864

0-858 „ 0-864

Melting-point

Saponification value ....
Iodine value

38 „ 46°

192 „ 200

35 „ 46

45° „ 50°

192 „ 198

34 „ 48

Rei chert value

0-25

Hehner value .....

95 to 96

Refractive index at 60°

1-4510

1-4500

Butyro-refractometer No. at 40°

48 to 49

Specific gravity of fatty acids, ^^o^ooo
Melting-point of fatty acids
Mean molecular weight of fatty acids .
Refractive index of fatty acids at 60° .

0-870

43 to 47°

270 „ 285

1-4375

0-869 „ 0-872

46° „ 54°

266 „ 275

1-4374

OLIVE OIL.

At one time the names olive oil and salad oil werfe considered
synonyms in the oil trade, but to day, owing to the fact that several
other vegetable oils are quite suited for edible purposes, oils sold
under the name of salad oil are not necessarily olive oil.

Olive oil is pressed from the fruit of the olive tree, Olea europaea.
It is true that some of the oil is also extracted by means of a suitable
solvent, but the resulting oil is only fit for use for industrial purposes,
so that from the present point of view, we need only consider olive
oil of the edible type.

There are very numerous species of the olive tree, and this fact,
together with the effect of climate, soil and method of cultivation,
accounts for the numerous types of genuine olive oil which may be
found on the market. The finest oil is obtained from hand-picked
fruits, and the olive stones -are not crushed in the pressing — the oil
from olive kernels is not quite identical with that from the pulp of the
fruits. There are various grades, such as the first pressed oil, which
is always the best, and the second pressings, which are usually ob-
tained by moistening the marc from the first pressings, and subjecting
it to a further pressing. Apart from the question of purity, the
essential feature of edible olive oil is that it should contain only a
very small amount of free fatty acids : as a rule from 0'2 per cent to
0"5 per cent will be found to cover the best samples. Samples con-

112

FOOD AND DEUGS.

taining more than this cannot be condemned as impure, but must he
judged as inferior oils. Commercial oils used for pharmaceutical
purposes, such as for the preparation of liniment of camphor, will
usually be judged as of good quality when they contain less than 4
per cent of free fatty acids, when calculated as oleic acid.

Olive oil consists of a large quantity of olein, mixed with some
stearin, palmatin, and other glycerides. It is a limpid liquid of pale
.yellow colour, or in the lower grades, sometimes green, but never
so in the case of the best oils. Its taste is sweet and bland, re-
calling that of the fruit. This last feature varies very greatly with
the district in which the olives are grown — Tunisian oil, for example,
has a rather harsh taste, and for many years there is no doubt that
Tunisian oils were improperly rejected by the French official analysts,
perhaps partly on this account, but also because they were found to
yield a slight reaction then believed to be absolutely characteristic of
sesame oil. It is now well established that injustice was done to this
oil, and a very large trade in it now exists. It is generally necessary
to blend the oil from such districts with sweeter oil from other dis-
tricts. In choosing an oil for such purposes as tinning sardines, the
finest oil alone should be used, as a poor-quality oil will spoil the
finest sardines.

In judging the purity of olive oil, the following are the principal
features to determine : —

Specific gravity : iodine value : solidifying point : saponification
value : the characters of the fatty acids : the refractive index, and
special tests for certain oils, which will be mentioned later. The
following table is compiled from the examination of about 100 samples-
of edible olive oils in the author's laboratories, but for the sake of com-
pleteness, the figures for ordinary oils, which are rather wider in their
limits, are added as well.

Characters op Olive Oil.

Edible Oils.

Olive Oil in general.
0-914 to 0-919

Specific gravity at 15° C

0-915 to 0-9175

Iodine value .....

82 „ 85

80 „ 88

Solidifying point ....

+ 1° „ + 3°

+ 1° „ + 4°

Butyro-refractometer No. at 20°

66 „ 67

Refractive index at 20° .

1-4690 „ 1-4671

1-4669 „ 1-4671

Saponiticatiop value (% KOH) .

18-9 „ 19-1

18-5 „ 19-5

Reichert value

0-2 „ 0-3

Maumene test (with HgSO^)

41° „ 46°

Melting-point of insoluble fatty acids

24° „ 25-5°

24° „ 27°

Mean molecular weight of ,,

280 „ 286

Free fatty acids ....

0-1 „ 1-5 %

—

Olive oil is considerably adulterated. At one time the favourite-
adulterant was poppy-seed oil, but this has long given way to other
oils of which the favourites are arachis, sesame and cotton-seed oils.

OLIVE OIL. 113

It will not be necessary to deal here with the adulteration with such
oils as castor and rape, as these are only found in industrial oils,
their taste preventing them from being employed for admixture with
edible oils.

The effects on the analytical results, of the principal adulterants
are as follows.

(1) Arachis Oil. — The specific gravity of this oil is slightly higher
than that of olive oil, but 30 to 40 per cent may be added without
raising this figure beyond that of normal olive oil. The iodine value
is about 95 to 98, so that a substantial addition of arachis oil may be
indicated by a high iodine value. As the fatty acids melt at 31° to
32°, a slightly higher melting-point may be expected if much arachis
oil is present. The action of nitrous acid on arachis oil differs from
that on olive oil, and may roughly indicate the presence of this or
another adulterant. Olive oil yields the hardest elaidin of all oils,
and much arachis oil will prevent the formation of more than a
buttery mass except after a long time. The elaidin test is best carried
out as follows. Ten c.c. of the oil are mixed with 5 c.c. of nitric acid
(specific gravity 1-4) and 1 grm. of mercury, and shaken till the
mercury dissolves. It is then allowed to stand for twenty minutes,
and then again shaken for one minute. Olive oil yields the hardest
mass as a result, but arachis oil yields a hard elaidin, only in a rather
longer time, so that nothing more than general indications are yielded
by this test, unless a very soft elaidin, such as would be due to the
presence of a large quantity of sesame oil, results, when a very strong
inference of adulteration may be drawn. The decisive test for arachis
oil, however, is the determination of the arachidic acid. The very
small quantity of arachidin present in olive oil does not interfere with
the approximate accuracy of the process. But it must be remembered
that the process is only approximate, since arachis oil contains a
variable amount of arachidic acid, so that an average factor must be
used to convert the arachidic acid into arachis oil. Keliable analysts
have isolated from 4"5 per cent to 5 per cent (Eenard) ; 4 '37 per cent
to 4-8 per cent (De Negri and Fabris) ; 5*5 per cent (Allen) ; 4*31 per
cent to 5*4 per cent (Tortelli and Kuggeri). The usually adopted
factor is based on an average value of 5 per cent, and this will give
results as near as are usually necessary. If the arachidic acid be
purified so as to melt at 74° to 75°, 4*8 is a more accurate factor The
process is carried out as follows (this was originally described by
Renard, " Comptes Rendus," 73, 1330) : 10 grms. of oil are saponified
with alcoholic potash, and the fatty acids separated in the usual
manner, by driving off the alcohjol, and acidifying the aqueous solution
with hydrochloric acid. The fatty acids are washed, and dissolved in
90 per cent alcohol. Excess of a solution of lead acetate is now
added. The precipitated lead salts are filtered off", washed with
alcohol, and dried. They are then extracted in a Soxhlet tube with
ether so as to separate the lead salts of the unsaturated and saturated
fatty acids. The lead salts of the saturated fatty acids remain un-
dissolved, and are decomposed by the addition of hydrochloric acid
under ether : the latter dissolves the fatty acids, and is separated, the
VOL. I. 8

114

FOOD AND DKUGS.

residue washed with ether and the whole of the solvent evaporated,
leaving the residue of fatty acids. This is dissolved in 50 c.c. of hot
90 per cent spirit, transferred to a covered vessel, and allowed to cool.
If any quantity of arachis oil be present a crop of crystals of arachidic
acid will be formed when the alcohol has cooled. If this be the case,
filter off the crystals, wash the filter with 90 per cent and then with
70 per cent alcohol, noting the quantity of 90 per cent alcohol used
for washing. Now dissolve the crystals by pouring boiling absolute
alcohol through the filter, evaporating the solvent in a small capsule
and weighing the residue. To the weight so found, add 0*0022 grm.
for each 10 c.c. of 90 per cent alcohol used for washing at 15° C, or
0-0045 grm. if at 20° C. This crude arachidic acid, multiplied by 20,
gives approximately the amount of arachis oil present in the sample.
Its melting-point should be 71° to 72°.

There are various modifications of this process. Lewkowitsch
prefers to neutralize the saponification liquids with acetic acid, using
phenol-phthalein as an indicator, and then precipitate with lead
acetate, in order to save the trouble of separating the fatty acids.
Tortelli and Ruggeri prefer to use 20 grms. of oil and to dissolve the
fatty acids liberated from the lead salts not dissolved by the ether
extraction, in 100 c.c. of 90 per cent alcohol, on the water bath at
about 60° C. A drop of hydrochloric acid may be added if the liquid
is turbid. The liquid is allowed to stand for three hours at 15° or
20°. The separated fatty acids are transferred to a filter (the filtered
liquid may be used to assist in this operation). They are then washed
thrice with 10 c.c. of 90 per cent alcohol, and then with 70 per cent
alcohol. The crystals are now dissolved in boiling absolute alcohol.
The solvent is driven off, and the residue again dissolved in 100 c.c.
of 90 per cent alcohol, and the separation and filtration and washing
carried out as before. The residue is now dissolved in absolute alco-
hol again, the solvent driven off, and the residue weighed. The
crystals should now melt at 74 to 75-5°. The following table (from
Lewkowitsch) gives the amount to be added to the weight found,
due to the solubility in 90 per cent alcohol according to Tortelli : —

100 c.c. OP 90 Per Cent Alcohol Dissolve Arachidic Acid Melting

AT 74-75°.

Amount of Acid found.

At 15°.

At 17-5°.

At 20°.

2-7 gr. down to 0-5 gr.
0-5 „ „ 0-17 „
0-17 „ „ 0-05 „

0-070 gr.

0-050

0-033

0-080
0-060
0-040

0-090
0-070
0-045

The following qualitative method will detect 10 per cent of arachis
oil. Saponify 1 c.c. of the oil with 5 c.c. of an 8*5 per cent solution
of ordinary caustic potash in absolute alcohol. This will only take a'
few minutes. Exactly neutralise with 90 per cent acetic acid, and then
add 50 c.c. of 70 per cent alcohol containing 1 per cent of ordinary

OLIVE OIL. 115

laboratory hydrochloric acid. Cool to 18° to 20°. In the presence of
10 per cent, often as low as 5 per cent, of nut oil — crystals of
arachidic acid will separate. In general pure olive oil will remain
quite clear, but occasionally a sample will give a slight flocculent
deposit easily distinguished from arachidic acid.

The French Codex prescribes the following test : —

One c.c. of olive oil and 15 c.c. of alcoholic potassium hydroxide
solution (5 per cent) are boiled for twenty minutes in a small flask
under a reflux condenser. The liquid is then to be kept in a cool
place for twelve hours, at the end of which it should still be limpid.

(2) Cotton-seed Oil. — The specific gravity of cotton-seed oil is about
0*922 so that appreciable " quantities would slightly raise this figure,
although moderate quantities might not affect it appreciably. The
iodine value is about 110, so that a high iodine value for any sample
of olive oil would suggest the presence of cotton oil. The most valu-
able quantitative process, however, is the determination of the iodine
value of the liquid fatty acids, which have — in the case of cotton and
olive oils — a far greater difference than the oils themselves. The liquid
fatty acids are prepared by converting the fatty acids into lead salts,
extracting the lead salts of the liquid fatty acids by ether (as de-
scribed in the previous paragraph) and decomposing the lead salts in
the usual manner. The iodine value of the liquid fatty' acids of pure
olive oil varies from 94 to 96-5. That of the liquid fatty acids of
cotton-seed oil is from 145 to 150, so that the presence of 10 per cent
of cotton-seed oil is clearly indicated.

Several colour reactions have been recommended, and are to some
extent trustworthy, but not altogether so. The principal of these tests
are those known as the Becchi and the Halphen tests.

Becchi's test as originally introduced involves the use of an alco-
holic solution of silver nitrate, and a solvent consisting of amyl alcohol
and colza oil, the use of which is far from clear. The test has under-
gone so many modifications, and been condemned by so many chemists,
yet approved by so many others, that a strong difference of opinion
exists in regard to it. The simplest and probably most useful form is
that of the British Pharmacopoeia. It is as follows : If 10 c.c. of the
oil be shaken with 2 c.c. of a reagent prepared by dissolving 1 grm. of
silver nitrate in 100 c.c. of absolute alcohol, with the addition of 20
c.c. of ether and one drop of nitric acid, no blackening should take
place when the mixture is heated on a water bath for ten minutes.
There is no doubt that many pure olive oils give a brown or even slight
black colour, so that the test is deceptive, but Milliau claims that if
the test be applied to the fatty acids instead of to the oil itself, no pure
olive oil gives the reaction. Hehner & Lewkowitsch see no advantage
in Milliau's suggestion, but the author finds that the results are more
trustworthy than on the oils themselves. The Halphen test is the
better of the two. It consists in heating in a water bath about 2 c.c.
of the oil, with an equal volume of amyl alcohol, and an equal volume
of a 1 per cent solution of sulphur in carbon bisulphide. In the pre-
sence of cotton-seed oil the mixture becomes of a red colour, its in-
tensity depending on the amount of cotton oil present, in from 5 to
30 minutes. Definite reactions can be obtained with 2 per cent of

116 FOOD AND DKUGS.

cotton oil, and according to Lewkowitsch, even with 1 per cent. It
must be remembered, however, that there are certain other oils which
give a similar reaction, and also that if cotton-seed oil be heated suffi-
ciently, the substance which is responsible for the Halphen and the
Becchi reactions is destroyed, and no reactions will beiobtained. These
colour tests, therefore, must only be considered as of an indicative or
confirmatory nature, and must not be too greatly relied on, unless con-
firmed by the iodine determination above mentioned.

(3) Sesame Oil. — The specific gravity of this oil is about 0*923 to
0*924, and the iodine value about 105. Only faint differences in these
figures would therefore be produced by the addition of substantial
quantities of sesame oil, although large quantities might be distinctly
indicated.

The iodine value of the liquid fatty acids (see p. 115) is from
130 to 140, so that better indications are given by this than by the
iodine value of the oils themselves.

There is, however, a colour test which appears to be absolutely re-
liable for this oil. It is known as Badouin's test (modified in various
methods). It consists in dissolving 0*1 grm. of sugar in 10 c.c. of
hydrochloric acid of specific gravity 1*19, andadding 20 c.c. of the
oil to be tested, and shaking for at least a minute in a stoppered test
tube, and allowing the mixture to stand and separate. In the pres-
ence of from 1 per cent to 2 per cent of sesame oil a distinct crimson
colour will result in the aqueous layer. This reaction appears to be
due to the formation of furfural by the action of the acid on the sugar,
and Villavecchia recommends the following as the best modification
of the test, which it certainly is. Use 0*1 c.c. of a 2 per cent solution
of furfural in alcohol, 10 c.c. of the oil, and 10 c.c. of hydrochloric
acid. This will certainly reveal the presence of 1 per cent to 2 per
cent of sesame oil. It must be noted that some olive oils, especially
Tunisian oils, give a slight reaction, but according to Milliau, this is
never the case with the fatty acids, so that in doubtful cases the fatty
acids should be tested instead of the oil. There is also another point
to be remembered. There are many factories where arachis and
sesame oils are prepared together. The press cloths are not changed,
and after a pressing of sesame oil, the first pressing of arachis oil con-
tains a trace of sesame oil, and the author has examined many
samples having every characteristic of pure arachis oil, which could
not have contained more than a trace of sesame oil, but which yielded
this reaction. Hence a faint trace of sesame oil may be accompanied
by a large amount of arachis oil, where such an oil has been used as
the adulterant.

(4) Poppy-seed Oil. — This is not now a common adulterant for edible
oils, but it will be indicated by a rise in specific gravity, and iodine
value (which are about -926 and 134 respectively for poppy-seed oil).

The requirements of the British Pharmacopoeia for olive oil are as
follows : It is a pale yellow or greenish-yellow oil with a faint odour
and a bland taste. Specific gravity at 15*5° from 0*914 to 0*919. At
10° it is liable to become of a pasty consistence, and at 0° to form a
nearly solid granular mass. It must not yield a black colour when sub-
jected to the silver nitrate test described above.

:aptek III.

THE CAEBOHYDKATE FOODS.

The carbohydrate foods fall naturally into two groups, the sugars and
the starches. Of these the sugars are far better understood from a
chemical point of view, the starchy substances being closely related
to them, but of a less definite nature. The sugars will therefore be
first considered.

Cane Sugar. — The natural sugars which are, or enter into the
composition of, food stuffs, are for the most part substitutive deriva-
tives of the hydrocarbon hexane C^-tii^. The principal member of
the series is cane sugar — known also as sucrose, and giving the
generic name of saccharoses to that group of carbohydrates which in-
cludes itself, maltose and lactose. Cane sugar forms hard, transparent
crystals melting at 160°. After melting it remains for a long time
in a transparent amorphous condition (barley sugar), and at higher
temperatures it becomes converted into a dark brown amorphous
substance known as caramel. Caramel is a mixture of various de-
composition products of the sugar.

The formula of sugar is 0^2-^22^11' ^^^ ^^ ^^ known — as are lactose
and maltose — as a twelve-carbon sugar or hexabiose (or saccharose).
On hydrolysis cane sugar yields dextrose and levulose (see below) in
equal amounts. It can be prepared artificially by the action of acetyl-
chlorohydrose on potassium-levulose in alcoholic solution, and un-
doubtedly is an oxygen ether of the anhydrides of dextrose and
levulose, of the constitution

Ca,OH.CHOH.CH(CHOH)2.CH.O.C(CH20H)(CHOH)2CH.CH20H.

I o \ I 0 ^1

This constitution explains why cane sugar neither reduces alkaline
copper solution, nor forms an osazone — both of which positive char-
acters are typical of, and extremely useful in identifying, the sugars
which have an aldehydic constitution.

Cane sugar is extracted from the juice of the sugar cane, the sugar
beet, the sugar maple, and the sorghum plant. It occurs in numerous
other fruits, often associated with other forms of "sugar," but the
sugar of commerce is obtained from the three sources above mentioned,
cane sugar and beet sugar being used in Europe, whilst these and
some maple sugar are employed in America, where the sugar maple
flourishes. Loaf sugar is a product which has been rapidly cry-

(117)

118 FOOD AND DRUGS.

stallized from hot solutions of the cruder sugar which have been de-
colorized by animal charcoal. "Without this treatment, the crystals
will be the cruder brow^n sugar. Sugar candy is the result of slow
crystallization from cold syrup, with a nucleus of string on which the
crystals are deposited. Cane sugar is strongly dextrorotatory, the specific
rotatory power for the sodium line being + 66° (the optical properties of
sugars will be referred to later). It is soluble in about half its weight
of cold water — forming the viscous liquid known as " syrup ". It is
nearly insoluble in absolute alcohol, but is readily soluble in dilute
alcohol, even in 90 per cent " spirits of wine " ; but insoluble in ether
and similar liquids. It is oxidized by nitric acid, in the cold slowly,
and if the temperature be kept below 50° C, the product is entirely
saccharic acid C^Hj^Og, but at 100° oxalic acid is the principal pro-
duct. Cane sugar is distinguished from the glucoses by the fact that
solutions of caustic alkalies have no immediate perceptible action on
it, whilst this is not so with the glucoses.

By the action of yeast cane sugar is first transformed into a mixture
of dextrose and levulose (invert sugar) which are ultimately further
changed by fermentation into alcohol, carbon dioxide and traces of
other compounds. Maltose and lactose may now conveniently be
shortly described.

Maltose CjoHggOji + HgO exists in malt, and of course malt extract,
and results, together with dextrin, by the limited action of dilute acids
or infusion of malt, which contains the ferment diastase, on starch.
It is strongly dextrorotatory, [a]d = -t 138°. Although isomeric with
sucrose, its constitution is very different, as it is of an aldehydic nature,
reducing alkaline copper solution and forming an osazone. Chemically
it is glucose-glucoside, of the constitution

CH.,.OH.CHOH.CH(CHOH)2.CH.O.CH2(CHOH),CHO.

L_o— I

and is stereo- isomeric with lactose. It yields dextrose only on
hydrolysis, and is directly fermentable by yeast, if its action be
long continued, since yeast usually contains a hydrolysing ferment
which causes dextrose to be formed, which is then directly fermented.

It forms fine crystalline needles, which contain one molecule of
water of crystallization, which is lost at 100°. It is less soluble in
alcohol than sucrose. Maltose exhibits the phenomenon known as
bi-rotation, that is, the rotatory power of a freshly made solution is
less than that of one which has been kept for some time or has been
heated. A cold solution of maltose takes several hours before it at-
tains its full optical activity — a fact of great importance in practice,
as will be seen in the sequel. Lactose and dextrose also exhibit bi-
rotation, but in these cases the rotatory power diminishes on keeping.

Maltose is hydrolysed by heating with dilute mineral acids, the
resulting product being dextrose, the solution increasing in its power
of reducing copper solutions, and decreasing in optical activity. Three
to four hours boiling with dilute acid is necessary for complete inver-
sion. Maltose is not inverted by the ferments diastase or invertase.

THE CARBOHYDEATE FOODS. 119

It resembles the glucoses, as mentioned above, in its power of re-
ducing copper salts (as Fehling's solution, vide infra). But its reduc-
ing power is considerally lower than that of the glucoses, which is
easily understood by an inspection of the constitutional formulae of the
compounds — the reducing power depending on the aldehydic com-
plex in the molecule.

Maltose in a more or less pure condition is manufactured by the
action of malt infusion on starch. Dextrin in some variety or other
is always found at the same time, the proportion varying with the
conditions of the reaction. The normal reaction may be represented
as follows : —

3(Ci,H,,0,,) + 2H,0 = 2(C,H,,0,) + 2(Ci,H,,0„)
Starch Dextrin Maltose

Lactose or milk sugar, Cy^,22^-^^, is the sugar found in mammalian
milk, in which it is present to the extent of about 5 per cent. It
forms hard white rhombic crystals containing one molecule of water,
but usually occurs in commerce as a powder, in which form it is
largely employed as a constituent of infant foods. It is far less
soluble and less sweet than sucrose. It melts, when anhydrous, at
205° and has a specific rotation of + 52-7°. It reduces metallic solu-
tions, sometimes even in the cold, and forms an osazone. It is not
capable of direct fermentation by yeast, but is converted by the lactic
ferment into lactic acid. It is hydrolysed by dilute mineral acids
yielding equal quantities of glucose and its isomer galactose. It is
a stereo-isomer of maltose, and may be chemically described as
galactose-glucoside.

A freshly prepared saturated solution contains 14-55 per cent of
C12H22O11 + H2O, but after standing for some time 21-64 per cent is
dissolved. This phenomenon has some connexion with that of bi-
rotation, for the specific rotatory powers of the two modifications which
may be assumed to exist here, are in inverse ratio to the solubility.
The least insoluble modification has a specific rotatory power-}- 80°
whilst that of the more stable modification is + 52-7°. In dealing with
solutions of lactose, therefore, this fact should be borne in mind.

There are other hexabioses, but they are not of importance from
the present point of view.

The six-carbon sugars, or hexoses, to which consideration must
now be given are dextrose (which is generally known as glucose), levu-
lose and, incidentally, galactose.

Dextrose, glucose or " starch sugar," CgHj.,0,,, is a colourless
crystalline substance, usually crystallizing with one molecule of water,
but can be obtained in an anhydrous condition by crystallization from
hot methyl alcohol. It loses its water by crystallization below 100° C,
and when anhydrous melts at 146°. It is soluble in a little more than
its own volume of water, forming a syrup which is much less sweet
than cane sugar syrup. It exists — with levulose — in honey, and in many
fruits, such as the grape, which often contains 15 per cent. It results
from the decomposition of many of the glucosides, and is artificially
prepared by the hydrolysis of starch or cane sugar. The glucose of

120 FOOD AND DRUGS.

commerce is nearly always the product of the hydrolysis of starch.
Some discrepancies exist in the figures usually quoted for the specific
rotatory power of dextrose but the probable value is + 53°. When
partly dissolved its rotatory power is much greater, owing to the initial
formation of a labile isomeride. The rotation decreases slowly if
the solution be left, or rapidly by heating or the addition of alkali,
until its permanent value is attained.

Dextrose is soluble in alcohol, is not charred by cold sulphuric
acid, and is coloured brown when warmed with caustic soda solution.
It has a strong reducing power on metallic solutions such as Fehling's
solution, and yields a characteristic osazone. It is directly ferment-
able by yeast.

The constitution of dextrose is important, as indicating the power
of reducing metallic solutions, and of forming an osazone, characters
which are always concomitant. It is at once a pentatomic alcohol,
containing an aldehyde group, known as an aldose or aldohexose of
the constitution CHpH . (CriOfl)^ . CHO. There is now do doubt
that dextrose is in reality a mixture of two bodies of this constitution,
differing only in stereochemical relationships. In general, dextrose
which has been prepared from cold solutions contains excess of
a-dextrose, of specific rotation + 105°, whereas that separated from
solutions that have been heated contains most y8-dextrose of specific
rotation + 22°. All forms of dextrose when in solution in water gradu-
ally attain a state of equilibrium in which there is nearly half of each
form present, hence the fact that ultimately a specific rotation of + 53°
is attained. It may be mentioned incidentally that dextrose exists,
on account of the atomic groupings it contains, in the laevorotatory
and optically inactive forms.

Levulose or fructose (fruit sugar) is a laevorotatory six-carbon sugar,
containing a ketonic grouping, and is classified as a keto-hexose,

It occurs in honey and in various fruits, and is formed in equal
amount with dextrose in the hydrolysis of cane sugar. It is also
easily obtained by the hydrolysis of inulin, a starchy matter found in
dahlia tubers, which yields levulose in the same manner as ordinary
starch yields dextrose. It is a colourless crystalline substance, melt-
ing at 95°. Its specific rotation is - 98-8° at 15° decreasing by
0'6385° for each degree until at 87° C, it is - 53°, which is identical
with that of dextrose, but opposite in sign. Being a keto-alcohol, it
forms an osazone and easily reduces metallic solutions.

The product of hydrolysis of cane sugar is a mixture in 'equal
quantity of dextrose and levulose, which is known as invert sugar.
But as the rotation of levulose is higher than that of dextrose, cane
sugar is converted by hydrolysis from a dextrorotatory to a laevo-
rotatory substance.

Levulose has the constitutional formula

CH2OH . CO,. (CH0H)3 . CH,OH

which indicates its power of forming an osazone and of reducing
metallic solutions.

THE CAEBOHYDRATE FOODS. 121

It is of great interest to note that levulose or fructose can be ob-
tained from glucose or dextrose by means of its osazone ; and in
spite of its laevorotation, the name fZ-fructose is retained for levulose in
the scientific nomenclature of the sugars, as indicating its genetic
relationships. It may also be here mentioned that numerous space
isomerides of most of the sugars may, and do, exist, but these the
analyst has never to deal with.

Levulose is separated from dextrose in the reaction products of
the hydrolysis of cane sugar by mixing the liquid with powdered
slaked lime, in a vessel surrounded by ice. The levulose forms a
nearly insoluble compound with calcium, whilst the dextrose compound
is soluble and can be filtered off. The levulose compound is decom-
posed by shaking with oxalic acid, or better, by a current of CO^ when
the filtered liquid yields anhydrous levulose by evaporation in vacuo
over sulphuric acid.

Invert Sugar. — This name is given to the mixture of dextrose and
levulose, either occurring naturally, when it may may have resulted
from the hydrolysis of cane sugar (natural invert sugar has not a con-
stant composition, as the conditions of its formation are not constant) ;
or prepared artificially, especially for brewers' use, and sold under
the names of " invert sugar," ** saccharum," or " saccharine ".

Other compounds of this group will be referred to as found neces-
sary under special paragraphs.

THE CHARACTERISTICS OF THE SUGARS.

The Phenyl-hydrazine Compounds. — The hexoses, of which dex-
trose and levulose are typical (as also galactose the product, together
with dextrose, of the hydrolysis of lactose) show the reactions of alco-
hols, and those of aldehydes or ketones. Of the hexabioses under
consideration, cane sugar contains no aldehydic or ketonic groupings,
whilst maltose and lactose both possess aldehydic functions. Those
sugars, then, which contain aldehydic or ketonic groupings, are
capable of reacting with phenyl-hydrazine, and in some cases of form-
ing compounds which are well suited to characterize the several
sugars. In general, the sugars, in the presence of excess of phenyl-
hydrazine C^H- . NH . NHg, react, with the formation of osazones, in
accordance with the equation (for the hexoses),

COH CH(N . NH . CgHj)

HOH + 2(H2N . NH . C.H^) = C(N. NH. C.-HJ + 2H20-hH2

(CH0H)3 (CH0H)3

I I

CH^OH CH2OH

The reaction is carried out by adding two parts of phenyl-hydra-
zine and two parts of 50 per cent acetic acid, to about I part of glu-
cose or other sugar in 20 parts of water. The mixture is digested for
about an hour on the water bath, when the osazone will separate
on cooling.

122 FOOD AND DEUGS.

Dextrose and levulose yield the same glucosazone under these
circumstances, which clearly indicates the identity of the CH.,OH
(CHOH)^ grouping in both compounds as will be seen from an in-
spection of the above equation. The crystalline osazone of dextrose
or levulose is collected on a filter, washed with a little water and.
dried. To ensure absolute purity, it may be recrystallized from hot
alcohol. It is then found to be a golden yellow crystalline powder
melting at 204°. The formation of this compound is absolutely indi-
cative of the presence of sugar in diabetic urine, where substances
may occur which simulate the other reactions of glucose. The osazone
of lactose, produced under similar conditions, melts at 200°, and that
of maltose at 206°. If the osazones are sufficiently well purified,
these melting-points are sharp, and quite characteristic (especially
when taken in conjunction with the other properties of the sugars) of
the several individuals.

THE SUGARS AS REDUCING AGENTS.

The majority of sugars are either aldehydic or ketonic in char-
acter, and as such, possess greater or less power of acting as reducing
agents. Those sugars which are not aldehydic or ketonic in character,
and do not form phenyl-hydrazine compounds, do not reduce metallic
solutions — (or if so only very slightly and with difiiculty).

In alkaline solutions, the aldehyde and ketone sugars reduce picric
acid to picramic acid ; indigotin to indigo white ; and ferricyanides to
ferrocyanides. Bismuth, mercury, silver, gold and platinum salts are
reduced to the metallic state ; and ferric and cupric salts to ferrous
and cuprous compounds.

In practice, the reduction of cupric salts to cuprous oxide is almost
universally used as a quantitative method for the determination of the
sugars. The use of mercuric salts is not uncommon, but other re-
ductive processes are rarely used. These, then, are the only reduction
processes, which need be described in any detail : —

(1) Copper Salts. — Many organic compounds prevent the precipi-
tation of cupric sulphate by caustic soda or potash. The best com-
pound for preventing such precipitation is tartaric acid or a soluble
tartrate. If a solution of cupric sulphate, caustic soda and sodium
tartrate be made, it can be boiled without any precipitation occurring,
but in the presence of a reducing sugar a yellowish-red precipitate of
cuprous oxide occurs, and if it be present in sufficient quantity the
blue colour of the solution entirely disappears, the whole of the copper
being precipitated in the form of cuprous oxide. This reaction has
been utilized in numerous ways as the basis of a quantitative deter-
mination of various sugars.

The most generally employed method is the volumetric process with
Fehling's solution. To prepare this, 34-64 grms. of pure crystalline
cupric sulphate is dissolved in distilled water to form 500 c.c. This
is labelled No. 1 solution. No. 2 solution is prepared by dissolving
about 70 grms. of caustic soda (as nearly pure as possible) and 180
grms. of potassium sodium tartrate, in water to form 500 c.c. These

THE CARBOHYDRATE FOODS. 123

solutions should be kept separate, and mixed in exactly equal volumes
as necessary, the mixture then forming Fehling's solution. This,
when boiled, should remain quite clear. If the two solutions are kept
mixed for any length of time no reliance should be placed on the mix-
ture for quantitative results.

To detect a reducing sugar by means of Fehling's solution, the
liquid to be tested must be clear and nearly colourless. If the solu-
tion is dark coloured it should "be clarified by the addition of lead
acetate solution, the excess of lead removed by a strong solution of
sulphurous acid, and a little washed moist alumina added. The liquid
is made up to a definite volume and filtered. With dark coloured liquids
qualitative reactions are difficult, and quantitative reactions impossible.

If a solution containing a reducing sugar be boiled with Fehling's
solution, a yellow, or orange red, precipitate of cuprous oxide is
formed. The hexoses, maltose and lactose reduce Fehling's solution,
but sucrose does not do so until after inversion. There are other
substances which reduce Fehling's solution, but, except in the case of
urine analysis, these do not as a rule give any difficulty in practice.
In doubtful cases, recourse must be made to the phenyl-hydrazine test.

As a quantitative reaction, the reduction of copper salts may be
used gravimetrically or volumetrically. The latter is the more generally
used method.

In this process, in which Fehling's solution is used, the saccharine
solution containing, as nearly as can be judged, about 1 per cent of
sugar, is placed in a burette, and 10 c.c. of Fehling's solution is placed
in a white porcelain basin with 30 c.c. of water, over a Bunsen burner.
When the solution is boiling the sugar solution is run in 2 c.c. at a
time at first, with boiling after every addition. As the cuprous oxide
is precipitated, the blue gradually lessens and as it is nearly gone, the
sugar solution is added more cautiously, but as rapidly as possible.
The end of the reaction is noted by allowing the precipitate to settle,
and noting that the blue colour has entirely disappeared, leaving
the supernatant liquid colourless or faintly yellow. Or a drop or
two of the liquid may be rapidly filtered through a little glass wool
and spotted with a mixture of acetic acid and potassium ferrocyanide
on a white tile, when excess of copper will produce a brown coloration.

As determined in the foregoing manner, 10 c.c. of Fehling's solu-
tion correspond to the following weights of sugar (these figures are the
average of numerous determinations by various observers : —

Dextrose or levulose (and invert sugar) .... 0*0505 grm.

Lactose 0-0685 „

Maltose . . . . • 0-0810 „

Sucrose, after inversion ....... 0-0475 „

In the determination of sucrose, it should be inverted by heating
50 c.c. of the saccharine liquid with 5 c.c. of strong hydrochloric acid
slowly to 68° C. and then allowing it to cool. If coloured, the liquid
may be treated with a little animal charcoal and filtered, or, if neces-
sary, clarified by treatment with lead as described above. Excess of
acid should be neutralized by sodium carbonate before the addition of
Fehling's solution.

124

FOOD AND DRUGS.

Fehling's solution should be, for exact work, standardized against
pure cane sugar : but if very exact results are required, the gravimetric
process must be employed.

The following are the details of the gravimetric process, using
Fehling's solution : —

About 50 c.c. of Fehling's solution is diluted with an equal volume
of boiling water that has been well boiled in order to expel dissolved
oxygen. The liquid is kept in a small beaker immersed in a second
beaker in which the water is kept boiling. When the temperature of
the diluted Fehling's solution is nearly 100° C, a known volume of
the sugar-containing liquid is added (previously neutralized, if neces-
sary) and the mixture kept in the boiling water for twelve to fifteen
minutes. If the amount of sugar present is large, the blue colour will
soon disappear, and a further quantity of Fehling's solution should at
once be added. The precipitated copper oxide is rapidly collected on
a double filter paper, washed with boiling, well boiled, water, dried,
ignited for five or six minutes in an open crucible and weighed as
CuO. As cupric oxide is hygroscopic, it must be kept in a desiccator
and rapidly weighed. A. H. Allen gives the following figures as the
equivalent amounts of sugars corresponding to 1 gram of cupric
oxide : —

Glucose.

Cane Sugar (iuverted).

Milk Sugar.

Maltose.

0-4585 gr.

0-4308 gr.

0-6153 gr.

0-7314 gr.

O'Sullivan introduced the symbols K and R as indicating the rela-
tive reducing values of carbohydrate mixtures, referred to dextrose and
maltose respectively. If the K value of dextrose be taken as 100, a
substance of half the reducing power of dextrose will have K = 50.
In the same way R is taken as 100 for the reducing value of maltose.

The following factors may be employed for the approximate calcula-
tion of the principal sugars from the weight of copper or copper
oxide obtained : —

Glucose.

Cane Sugar

(after
inversion).

Milk Sugar.

Maltose.

Copper

Cuprous oxide
Cupric oxide

0-5634
0-5042
0-4535

0-5395
0-4790
0-4308

0-7707
0-6843
0-6153

0-9039
0-8132
0-7314

Thus if 0-2 grm. of copper has been obtained 0-2 x 0*5395 will give
the equivalent of cane sugar.

These factors are not absolutely correct, especially for certain
values, and various tables from which the amount of dextrose can be
shown at once, have been constructed. The following is probably the
most accurate of such tables : —

THE CAEBOHYDEATE FOODS.

125

Cl,

Mg.

5-7

31-8

114

57-3

166

83-7

218

111-1

6-2

32-3

115

57-8

167

84-2

219

111-6

6-8

32-8

116

58-3

168

84-7

220

112-2

7-2

33-3

117

58-8

169

85-2

221

112-7

7-8

33-8

118

59-3

170

85-7

222

113-2

8-6

34-3

119

59-8

171

86-3

223

113-7

9-0

34-9

120

60-2

172

86-8

224

114-3

9-5

35-4

121

60-7

173

87-3

225

114-8

10-0

35-9

122

61-2

174

87-8

226

115-4

10-5

36-4

]23

61-7

175

88-3

227

115-9

11-0

36-8

124

62-2

176

88-9

228

116 4

11-6

37-3

125

62-8

177

89-4

229

117-0

12-0

37-8

126

63-3

178

89-9

230

117-5

12-5

38-3

127

63-8

179

90-4

231

118-1

130

38-6

128

64-3

180

91-0

232

118-6

13-5

39-0

129

64-8

181

91-5

233

119-2

14-0

39-4

130

65-3

182

92-0

234

119-7

14-5

40-0

131

65-8

183

92-5

235

120-3

15-0

40-5

132

66-3

184

93-1

236

120-8

15-5

41-0

133

66-8

185

93-6

237

121-3

16-0

41-5

134

67-3

186

94-1

238

121-8

16-5

42-0

135

67-8

187

94-6

239

122-4

17-0

42-5

136

68-3

188

95-1

240

122-9

17-6

42-9

137

68-8

189

95-7

241

123-5

18-0

43-4

138

69-4

190

96-2

242

124-0

18-5

43-9

139

69-9

191

96-7

243

124-6

19-0

44-4

140

70-4

192

97-2

244

125-1

15-5

44-9

141

70-9

193

97-7

245

125-7

20-0

45-4

142

71-4

194

98-3

246

126-2

20-4

45-9

143

71-9

195

98-8

247

126-8

20-9

46-4

144

72-4

196

99-3

248

127-3

21-4

46-8

145

72-9

197

99-8

249

127-9

21-9

47-3

146

73-4

198

100-4

250

128-4

22-4

47-8

147

73-9

199

100-9

251

128-9

22-9

48-3

148

74-5

200

101-4

252

129-4

23-4

48-8

149

75-0

201

101-9

253

130-0

23-9

49-3

150

75-5

202

102-5

254

130-6

*24-4

49-8

151

70-0

203

103-1

255

131-1

24-9

100

50-3

152

76-5

204

103-6

256

131-7

25-4

101

50-8

153

77-0

205

104-1

257

132-2

25-9

102

51-3

154

77-5

206

104-6

258

132-8

26-4

103

51-8

155

78-0

207

105-2

259

13B-3

26-9

104

52-3

156

78-5

208

105-7

260

133-9

27-4

105

52-8

-157

79-0

209

106-2

10()

53-3

158

79-6

210

106-7

28-4

107

53-8

159

80-1

211

107-3

28-9

108

54-3

160

80-6

212

107-8

29-3

109

54-8

161

81-1

213

108-4

29-8

110

55-3

162

81-6

214

108-9

30-3

111

55-8

163

82-1

215

109-4

30-8

112

56-3

164 1

82-6

216

109-9

31-3

113

56-8

165

83-2

217

110-5

126 FOOD AND DRUGS.

Pavy's Method. — Dr. Pavy has introduced a useful modification of
Fehling's process for determining sugar. It has the advantage of
yielding a sharp end reaction, and depends on the fact that in the
presence of excess of ammonia, the cuprous oxide is not precipitated
but forms a colourless solution. This solution is extremely easy to
oxidize, therefore contact with the air must be avoided. The am-
moniacal solution is prepared by adding 300 c.c. of strong ammonia
solution (0"880 specific gravity) and 400 c.c. of a 12 per cent solution
of caustic soda, to 120 c.c. of ordinary Fehling's solution, and making
up to one litre with distilled water. One hundred c.c. of this solution
has the same oxidizing power on dextrose as 10 c.c. of the ordinary
Fehling's solution, i.e. it corresponds to 0-050 grm. of dextrose. The
determination is carried out by introducing 100 c.c. of the copper
solution into a wide-mouthed flask having an india-rubber cork with
two perforations. The nose of the burette containing the sugar solution
is passed through one of these, and a bent tube to carry over steam
and ammonia vapour is passed through the other. A few fragments
of well-burnt pumice are added and the liquid boiled ; the sugar solu-
tion is then run in, boiling well after each addition, when the blue
colour fades and finally disappears. Hehner has shown (" Analyst,"
VI. 218) that the presence of varying amounts of salts, such as alkaline
tartrate or carbonate affects the accuracy of the process considerably.
The oxidizing power of this solution is only | of that of the ordinary
Fehling's solution on dextrose, levulose or invert sugar. Hence the
fact that 120 c.c. are diluted to a litre instead of 100 c.c. The reduc-
ing action of lactose and maltose on Pavy's solution is not identical
with that on Fehling's solution, and reliable figures for these sugars
are wanting. It is to be noted that the process of reduction is slower
with this solution than with Fehling's, hence longer boiling is neces-
sary.

Gerhard's Process.

The formation of a colourless double cyanide of potassium and
copper is the basis of a method devised by A. W. Gerrard. He pre-
pares the following three solutions : —

Solution No. 1.

CJopper sulphate recrystallized 69*30 grms.

Distilled water to 500 c.c.

Solution No. 2.

Tartrated soda crystallized 17o-00 grms.

Caustic soda (pure) 76-56 ,,

Distilled water to 500 c.c.

Solutio7i No. 3.

Cyanide of potassium (98 per cent) 83*00 grms.

Distilled water to 500 c.c.

THE CARBOHYDRATE FOODS. 127

For the purpose of testing the solutions, 5 c.c. of each are mixed
with 50 c.c. of distilled water, then boiled. Whilst boiling add a solu-
tion of grape sugar until the blue colour is discharged. If any pre-
cipitate is formed, more cyanide must be added to No. 3 until again
on boiling equal volumes of the mixed solutions with grape sugar, they
cease to precipitate.

As compared with Fehling's solution, the advantage is that the
end reaction is very sharp, filtration is avoided, time is saved, and
experimental error reduced.

The following are the details of a grape sugar estimation, when
using what may be termed the cyano-cupric test. Measure 5 c.c. each
of solutions No. 1,2, and 3 in the order given ; add 50 c.c. water, and
boil in a porcelain capsule. Run the sugar solution into the boiling
test solution until the blue colour has gone. This should be added
slowly towards the end of the reaction. For accurate determination
it is usual to make a second and more rapid estimation, so as to check
error that may arise from too long boiling.

It is to be noted that the copper solution here recommended is
twice the strength of Fehling's solution, but it is best to standardize
the solution against a known weight of inverted cane sugar.

The Reduction of Mercury Salts. — Knapp recommends the use of
an alkaline solution of potassio-mercuric cyanide. He prepares a
standard solution by dissolving 10 grms. of pure mercuric cyanide in
water, adding 100 c.c. of solution of sodium hydroxide of specific
gravity 1-145, and making the solution up to 1000 c.c. ; of this solution
10 c.c. are reduced by 25 milligrams of dextrose. The process of re-
duction is carried out as follows : 10 c.c. of the mercury solution
and 25 c.c. of water are heated to boiling, and the sugar solution (of
about I per cent strength) is run in from a burette until the whole
of the mercury is precipitated. To determine the end of the reaction,
the precipitate is allowed to subside, and a drop of the supernatant
liquid is spotted on to a piece of thin white filter paper. This paper
is held for a few seconds over fuming hydrochloric acid and then ex-
posed to sulphuretted hydrogen. If any mercury be left in solution
a yellow or brown stain is at once produced on the spot.

Sachsse proposes the use of a solution containing 18 grms. of pure
mercuric iodide, with 25 grms. of potassium iodide and 80 grms. of
caustic potash in 1000 c.c. The titration is conducted with the sugar
solution into the boiling mercury solution, the end of the reaction
being determined when a drop of the supernatant liquid ceases to give
a brown coloration with a drop of a strongly alkaline solution of
stannous chloride. Ten c.c. of Sachsse's solution are reduced by 33
mgs. of dextrose or 27 mgs. of invert sugar.

According to various experimenters, the conditions under which
the reduction of either copper or mercury salts is carried out, afifect
the results to a considerable extent, so that if accurate results are to
be expected the titrations should be carried out against experiments
with known quantities of sugar.

The relative reducing powers of the following sugars, taking that
of dextrose as 100 for each solution, are as follows : —

128

FOOD AND DEUGS.

Fehling's.

Kuapp's.

Sachsse's.

Dextrose

100

300

Invert sugar

100

120 (?)

Levulose

100

148 (?)

Lactose

Maltose

For numerous modifications of Fehling's process, none of which
appear to possess any particular advantage over the original method,
references maybe made to (" Journ. Amer. Chem. Soc." 1896, 749),
("Zeit. Anal. Chemie." 12, 27), ("Journ. Amer. Chem. Soc." 1907,
1744) and (" Zeit. Ver. Deut. Zuckerind." 1906, 1012).

The Polarimetric Determination of the Sugars. — Sugars, in general,
possess the power of rotating the plane of plane-polarized light. The
observation of this power of rotation is not conveniently effected on
the solid sugar, but is determined on a solution of the solid substance.
The rotation effected is approximately proportional to the concentra-
tion of the solution, but not strictly so. In determining this value, it
must be remembered that certain sugars possess the power of bi-rota-
tion, and their solutions should be allowed to stand for several hours
before a reading is taken.

The bi -rotation of sugars may be destroyed by adding a few drops
of strong ammonia to the solution before making it up to its normal
volume, or by boiling the solution for a few minutes. The state of
optical equilibrium is thus at once produced.

The specific rotatory power of an optically active substance is the
angular rotation effected on the plane of polarization by causing it to
traverse a thickness of 1 decimetre of the substance.^ This power is
different for different parts of the spectrum ; hence it is usual to indi-
cate the particular light which has been polarized. The symbol used
for specific rotatory power is [a] : that for 'the sodium light, or D line
of the spectrum, which is the usual light used, being indicated by [a]fi.

In practice it is usual to determine the angle of rotation for a
solution of known concentration, and from this to calculate the specific
rotatory power. Since, as has been mentioned above, the concentra-
tion is not always accurately in proportion to the observed angle, exact
results are only obtaining by alw^ays using solutions of approximately
constant concentration. The specific rotation of a substance in solu-
tion is calculated from the following formula : —

[a]

100 a

1 A more scientitic, and stricter, cletinitiou of specific rotatory power is one
which takes into account the density of the liquid.

The specific rotatory power of a liquid is the angle through which the plane of
polarized light is turned, when the light traverses a liyer whose thickness is in-
versely proportional to the specific gravity of the liquid. The decinietie is usually
adopted as the unit of length.

THE CARBOHYDRATE FOODS.

129

Where [a] is the specific rotation, a the observed angle of rotation
of the solution, I the length of the tube in decinaetres, and c the
number of grms. of substance in 100 c.c. of the solution. So long as
one is dealing, then, with a solution of only one sugar, it is obvious
that the percentage present in a solution can be calculated so long as
the specific rotation of the sugar be known. The following are the
mean values of numerous determinations of the apparent specific
rotations of the more common sugars, for solutions containing about
10 per cent and for solutions containing about 16 per cent of the
sugar. The values are for sodium light [a]a, and for the transition
tint as used on some instruments, [a]^-, at 15° C.

[al-

[«]y.

Sugar.

io per cent

16 per cent

10 per cent

16 per cent

Cane sugar

+ 66-6°

+ 66-5°

+ 73-8

+ 73-6°

Maltose (anhydrous)

+ 138°

-f 138°

+ 154-5°

+ 154-3°

(hydrated)

+ 132-2°

+ 139-4°

—

Lactose + IHjO

+ 52°

+ 68-5°

—

Dextrose

+ 63°

—

+ 58-6°

+ 58-3°

Levulose

- 98-8°

- 98-4°

- 109-7°

- 109-6°

Invert sugar

- 23-7°

- 23-6°

- 26-6°

—

(The rotation of levulose, and, ooneequently, of invert sugar, is affected greatly
by temperature — see below.)

Brown and Millar (" Trans. Chem. See." 1897, 71, 73) give the
following table for converting [a]a, into [a]j : —

Sugar.

Per cent solution.

[a]j=[a]a x by.

Cane sugar

1-107

Maltose

1-113

Mil

Dextrose

1-115

1-111

Starch sugar

1-111

1111

In practice, certain polarimeters are graduated on a scale which
indicates the percentage of sugar present in a given solution, under
definite conditions, whilst others are graduated in angular degrees
as well. So far as angular degrees are concerned, the above formula

[a] = — - — - will always apply when only one active sugar is present.

The graduations to read off the percentage of sugar present in the

solution are based on the use of an amount of sucrose in 100 c.c.

which will in a 2 decimetre tube cause the same rotation as a plate

of quartz 1 millimetre in thickness. This, for any given instrument,

VOL. L 9

130 FOOD AND DEUGS.

is known as the "normal weight". For other sugars than sucrose,
angular rotations in degrees should be observed.

In the Soleil-Dubosq instrument 16'350 grms. of sucrose are taken
as the normal weight in 100 c.c. of liquid. For other instruments of
this type, quantities varying from 16*19 to 16'35 have been adopted
— so that the value may be agreed as 16-35 grms. For polarimeters
of the Ventzke type the standard weight, 26'04:8 grms., should be
used (or 26 grms. if the strict metric c.c. be employed). If angular
determinations are to be made solutions of 16 per cent to 18 per cent
strength should be employed.

The amount of sugar used for graduating each particular type of
instrument is indicated by the maker, and should be adhered to in
making determinations. The standard instruments are gi-aduated so
that, for the transition tint 24 angular degrees (the rotation produced
by 1 mg. of quartz or the standard weight of sugar in 100 c.c, in a 2
decimetre tube), or for the sodium light about 21-8° (the similar value),
are divided into 100 sugar degrees. In making an observation, the
standard weight of the sample is dissolved in water to 100 c.c. of
solution and the reading taken, when the percentage of pure sucrose
will be directly indicated on the scale. If calculations are to be made
on instruments graduated in angular degrees, the percentage is de-
termined by comparing the rotation of a solution with that of a solu-
tion of pure sucrose of the same concentration. Thus if a solution of
20 grms. of the sample in 100 c.c. give a rotation of + 25° in a 2
decimetre tube, whilst an equally concentrated solution of pure sucrose
gives a rotation of +26*6°, the percentage of true sucrose in the

1 . 25 X 100 n^ o
sample is ______ = 94 -3 per cent.

2d'6

As 18*8 grms. of sucrose in 100 c.c. in a 2 decimetre tube efifect

an angular rotation of exactly 25°, it follows that if exactly this weight

of the sample be used, each angle of rotation equals 4 per cent of

sugar in the sample, thus facilitating calculation In determining the

amount of sucrose in liquids of unknown strength, it is obvious that

where an instrument graduated in angular degrees is used, the formula

r 1 100 a

applies, [a] being either 66*5'' for sodium light, or 73*6° for white light
with the neutral tint.

If the polarimeter be graduated in percentages of sugar, the con-
centration of the liquid is given by multiplying the standard weight of
sugar for which the instrument is designed by the observed number
of sugar degrees and dividing by 100.

The table at top of opposite page comparing the various instru-
ments will be found useful.

Very frequently solutions of sugar are not in a fit state for the
polarimeter, as no reading can be obtained unless the solution be both
clear and very pale in colour. If the solution be dark coloured it
should be agitated with about 20 per cent of its weight of fresh, dry
bone black, and agitated from time to time and then filtered, or, as

THE CARBOHYDRATE FOODS.

131

German Instruments, such as

Normal Weight
of Sugar.

1 Sugar

Division =

Angular Degrees

(for Sodium

Light).

1 Angular Degree
= Sugar Divi-
sions.

Sohmidt, and Haensch, Ven-

tzke, Scheibler, etc.
Soleil-Dubosq
Laurent

26-048

16-35
16 27

0-3468

0-2175
0-2167

2-8835

4-597
4-6154

preferred by A. H. Allen, the following method may be adopted. The
normal quantity of sugar sample is weighed out and dissolved in
about 50 c.c. of water in 100 c.c. flask. The solution may be (1)
colourless, but cloudy, (2) yellow, (3) brown, (4) black. In the first
case add about 3 c.c. of a cream of hydrated alumina (prepared by
precipitating a solution of alum by a hot solution of sodium carbonate
and washing the precipitate with hot water and then mixing it with
enough water to form a thin cream) and one drop of a 40 per cent
solution of basic acetate of lead. In the second case, the same amount
of alumina should be added, but 3 to 5 drops of the lead solution. In
the third and fourth cases, about 2 c.c. of a 10 per cent solution of
sodium sulphite should be added, and then the lead solution gradually
until no further precipitation takes place. The liquid is well agitated,
the precipitate allowed to settle, and then made up to the 100 c.c.
mark with water (the froth may be destroyed by the cautious addition
of a drop or two of methylated spirit or ether). The liquid is now
filtered and the rotation observed.

Bryan ("International Sugar Journal," 1908, 602) has shown,
however, that basic lead acetate causes a precipitation of both dextrose
and levulose, whereas normal lead acetate causes practically no such
precipitation. The American Association of Ofi&cial Agricultural
Chemists now use only the normal acetate for the purpose of clarifica-
tion. Eynon has shown that so long as only sufficient lead acetate is
used to leave only a very slight excess in solution, no serious error
results by the use of the basic salt, but considerable excess of lead
causes (in Clerget's process) an increase in the direct polarization,
and a decrease in the inversion polarization. A given sample for
example gave the following results : —

c.c. of Basic Lead
Acetate Solution
(21 per cent. Pt.).

Direct
Polarization.

Sucrose

per cent

(apparent).

Reducing Sugar

per cent

(apparent).

73-2°
73-4°
74°
75-1°

76
76-1
76-3
76-5

9-7
9-6
9-1
8-5

132 FOOD AND DKUGS.

Pellet (" Bull. Soc. Chem. Sucr. et Dist." 1906, 23, 1466) states
that the error due to clarifying sugar solutions with basic acetate of
lead, due to the volume occupied by the precipitate, is compensated
by the small amount of sugar mechanically precipitated with' the lead
compounds, and that no correction is necessary.

In dealing with sugars other than sucrose (so long as the standard
weight of the sample is used) the observed percentage recorded on the
instrument which is graduated in sugar units may be corrected with
actual percentages of the sugar in question by multiplying by the factor

t^ where [a]^ is the specific rotation of cane sugar, and [a]^ is that of
[a]2

the sugar in question. If angular degrees be employed the formula

ral= ^stands good, where [a] is the specific rotation, a the ob-

served angle, I the length of the tube in decimetres and c the number

of grammes per 100 c.c. of the actual sugar.

As a rule polarimetric observations are valueless when more than
one optically active substance is present. In certain cases, however,,
the presence of two such bodies does not prevent a fairly accurate
determination being made. This is the case where one of the bodiea
in question does not alter its optical properties by certain treatment,
whilst the other one does, and this alteration is capable of determina-
tion. For example, Clerget has proposed the hydrolysis of cane sugar
in the presence of glucose (dextrose) and the determination of the
optical properties of the inverted sugars in comparison with the same
values before inversion, as a means of determining the amount of
sucrose present when mixed with dextrose.

Dextrose is not affected by heating with dilute acids, whilst
sucrose is converted into " invert " sugar — a mixture in equal quan-
tities of dextrose and levulose, 100 parts of these being yielded by 95
parts of sugar. The specific rotations of both sucrose and dextrose
are not materially affected by change in temperature, whereas that of
levulose is markedly affected. From a specific rotation of about - 95°
at 20°, this value falls to - 53° at 87-2°.

As a matter of experiment it has been found that a solution of
sucrose which causes a rotation of 100 sugar degrees to the right in a
2 decimetre tube, will have a rotation of 39 degrees to the left after
inversion, the reading being taken at 10° C, and has therefore under-
gone a change of 139 divisions. The difference is less, the higher the
temperature, a reduction of 1 degree for each 2" of temperature taking
place. Hence at 0°, the difference in 144 sugar degrees, and, of

course, for any other temperature is given by the equation

*
D = 144 - ^, where t is the temperature centigrade.

Care should be taken that either the bulk of the solution is identi-
cal before and after inversion, or if it be increased after inversion it
should be by 10 per cent, and the readings taken before inversion in
a 200 mm. tube, and after inversion in a 220 mm. tube, when no cor-
rection will be necessary.

THE CAKBOHYDRATE FOODS. 133

The readings before and after inversion must be taken at the same
temperature, 15° being most suitable. At this temperature the change
by inversion is 136 "5°, so that the observed change in rotation, multi-

pHed by ^^ ^ (or 0"7326) represents the rotation due to the original

sucrose in the solution, from which the amount of cane sugar may be
deduced. This factor stands good for 15° C. however the change in
rotatory power be expressed, whether in sugar degrees, or in angular
degrees.

Thus if a solution gives a rotatory power of + 20° before inversion,
and after inversion a rotation of - 5°, then the actual change in
rotatory power is 25°, which, multiplied by 0-7326, is 18-31°. There-
fore the rotation of 18*31° is due to cane sugar originally present in
the solution, and that of +1-69° to dextrose (or some other dextro-
rotatory substance).

Glerget's original formula, then, is

S IQQK
lU - 0-5 f

where S equals the rotation in the original solution due to sucrose, K
equals the observed difference in rotation before and after inversion,
and f = the temperature centigrade.

The polarimeter employed by Clerget was a Soleil instrument
using 16-471 grms. of sugar as its standard. The usual polarimeter
employed by sugar analysts in this country is a Soleil- Ventzke-
Scheibler, or some modification of it, using 26 grms., and with this
instrument the figures of Clerget are not strictly accurate, although
very nearly so.

Herzfeld has investigated the inversion values more recently and
his figures are now accepted universally as accurate. Using the last-
named standard amount of sugar — viz. 13 grms. per 100 c.c, he finds
132-66° as the difference figure before and after inversion, if the read-
ings be taken at 20°, which is equal to 14266° at 0°. Hence the
formula of Clerget, corrected for present instruments becomes

S lOOK

142-66 - 0-5^

If any other concentration be employed, the inversion constant varies
slightly : the following are the values for given concentration,
at 0° C. :—

Per cent

1 = 141-85

5 = 14212
10 = 142-46
15 = 142-79
20 = 143-13

Inversion is usually best carried out by heating the solution with
10 per cent of its volume of strong hydrochloric acid at 68° to 70° (the
time taken to attain this temperature should be about ten minutes)
and then cooling down by immersion in cold water. If 50 c.c. be thus

134 FOOD AND DEUGS.

made up to 55 c.c, the reading in a 220 mm. tube will be comparable
with the reading of the original solution in a 200 mm. tube.

It is to be noted that heating for ten minutes at about 70"" with 10
per cent of hydrochloric acid completely inverts sucrose, but has
little action on maltose. To invert maltose, heating with 3 per cent
of strong sulphuric acid at 100° for three to four hours is advisable.

Hence an approximation to the amounts of sucrose and maltose
in a mixture can be obtained by using these two methods of inversion.

Pierraerts (" Bull. Assoc. Chim. Sucr. et Dist." 1909, 650) gives
the following formula for the determination of mixtures of sucrose
and maltose, when examined by the polarimeter before and after in-
version. Taking 66*5 as the specific rotation of sucrose and 130 as
that of hydrated maltose, and denoting the quantity of sucrose per
100 c.c. of the hydrolysed solution by x, and the corresponding quantity
of maltose by y, then

X = 0-57246 {a - a')

y = 0-3846154a - 0-2928363 {a - a')

where a is the polarimetric reading in sugar degrees before inversion
and a' is the reading after inversion.

Mixtures of Sucrose, Invert Sugar and Glucose. — Boseley (" Ana-
lyst," xxTii. 123) has published a number of observations on the
analysis of marmalade, which may be taken also to apply generally to
the examination of jams.

No difficulties are presented in the determination of the water or
free acids, the only point of real importance being an examination of
the sugars present. Supposing only cane sugar and invert sugar to
be present (due to the action of the acids of the fruit on the sucrose
used), the following method is the best to employ : —

65-12 grms. of the well-mixed sample are weighed out, and mixed
with about 50 c.c. of cold water : this is decanted into a 250 c.c. flask
and successive quantities of about 50 c.c. of water are used until the
whole is transferred to the flask. Add solution of basic acetate of lead,
make up to 250 c.c. and shake well. Excess of lead acetate should
be avoided, by seeing that the solution remains slightly acid. The
liquid is filtered and the polarimetric value taken. Fifty c.c. of the
filtrate are then treated with 5 c.c. of strong HCl, and inverted in the
usual manner. The cane sugar is calculated from the difference in
the polarization by Clerget's formula (see p. 132) and the invert
sugar from the formula

(cane sugar - direct reading) 100
Invert sugar = — -. •

If glucose be present it will be indicated by the reading after in-
version being positive instead of negative, or at all events much
smaller than usual if it be negative. If this be the case it will be
necessary to determine the cupric reducing power.

This is best done, on the assumption that the reducing power of
marmalade is in the neighbourhood of 25 per cent of sugar, by pre-

CANE SUGAR. 136

paring a solution of 13-024 grms. of pure cane sugar in 100 c.c. and
inverting by acid in the usual manner. Make up to 110 c.c. and take
11 c.c. and dilute to 100 c.c. Call this solution A.

Now take 20 c.c, of the filtrate of the marmalade solution (65'12
grms. in 250 c.c.) and dilute to 100 c.c. Call this solution B. It
contains four times as much of the marmalade, as solution A does
sugar. Boil in the usual manner with alkaline copper tartrate solu-
tion, using 10 c.c. of each solution in two small beakers, with the
usual precautions, when the cupric reducing power of the sample,
calculated into percentage of invert sugar, will be given by the fol-
lowing : —

Cu obtained from B .. ^^
four times Cu from A

The approximate amounts of cane sugar, invert sugar and glucose
in a marmalade can be calculated from the following : —

^ 100 X (direct - inverted reading).
Cane sugar = ^^ —^

144-1
2

J , _ Cane sugar - direct reading + (4 x (cupric reducing power))

nver sugar -

44-2

4+ __
100 •

Glucose = 2(cupric reducing power - invert sugar)
Notes. —

The cupric reducing power is in terms of invert sugar.
Glucose is assumed to contain 81-9 per cent of solids.

The factor 144 in Clerget's formula should be 142-66 ; and consequently 44
in the same formula should be 42-66.

COMMEECIAL CANE SUGAE AND ITS PRODUCTS.

Commercial sugar is manufactured either from the sugar cane, or
from the sugar beetroot, the latter forming the source of supply of the
bulk of the sugar manufactured on the European continent. Beet-
root sugar is the variety usually speculated in on the London market,
a polarization test being the accepted basis of sugar contracts. The
amount of cane sugar obtained from other sources is insignificant.

The principal types of sugar one meets with are as follows ; —

(1) Pure sugar, in the form of cones (loaves), chibes, small crystals
or large crystals (sugar candy), and powder.

(2) Brown crystals, often containing 97 per cent of sucrose.

(3) Raw sugars, containing about 88 per cent of true sucrose.

(4) Molasses or treacle. This is the syrup which is left after the
crystallization of the sucrose, and contains a considerable amount of
sucrose with more or less glucose, etc.

136

FOOD AND DRUGS.

The following analyses illustrate the average composition of vari-
ous types of sucrose : —

Organic

Sugar.

Sucrose.

Glucose.

Ash.

Water.

Matter
not Sugar.

Authority.

Raw cane, W.I.

94-4

2-2

0-2

2-8

0-3

Wallace.

88-0

514

0-96

4-23

1-67

Wigner and Harland

90-4

3-47

0-36

4-22

1-55

»i »»

Raw beet

87 to

0 to 0-2

l-4to2-l

2 to 5-1

1-2 to 3-0

Parry

(25 samples)

93-5

Pure cane

99-6 to

—

traces

—

„

(in various forms)

99-9

Pure beet loaves

99-1

trace

0-15

0-25

—

Wigner and Harland

A. H. Bryan (" U. S. Dept. Agriculture Bur. of Chem., Circular"
No. 40) gives the following analyses of pure maple products : —

Sugar.

Juice.

Water .

Sucrose .

Invert sugar .

Lead No .

Ash ...

Soluble ash .

3-05 to 11 per cent
72-6 „ 87-4 „
1-16 „ 8-37 „
1-83 „ 2-48
0-64,, 1-32 percent
0-33 „ 0-67

up to 32 per cent.

51 „ 62-2 „
0-34,, 9-17 „
1-19 „ 2-03
0-46 ,, 1-01 per cent
0-21 „ 0-63

The only determinations necessary as a rule in examining com-
mercial sugar, are the water, mineral matter, sucrose, reducing sugars —
and the difference figure of these, which is usually returned as organic
matter other than sugar.

The water is estimated by heating 5 grms. in a platinum capsule
at 100° to 110° until the weight is constant. If, however, much
glucose is present, the temperature should be kept below 65° C. and
the time of drying consequently lengthened, since glucose increases
in weight by prolonged exposure to heat.

Ash Determination. — The actual amount of ash of commercial
sugar is diflBcult and tedious to ascertain exactly, since it is difficult
to completely incinerate the sugar, and the ash is both very light and
liable to be blown away, and hygroscopic and difficult to weigh. Its
minute proportion, however, makes it a matter of almost indifference,
if it be sulphated and weighed as sulphates. Some prefer to deduct
10 per cent of the weight of the sulphated ash to convert into true
ash, but this refinement is scarcely necessary, considering its minute
amount. It is usually returned in this method however. Three grms.
of the sample are slightly moistened with water and then with a little
pure, strong, sulphuric acid, and the whole gently heated to a cinder,
when it is burned at a low red heat in a muffle, being moistened again
with sulphuric acid when it is nearly free from carbon. The presence

CANE SUGAR.

137

of sand or clay will be indicated by a high ash value, and by a high
proportion of matter insoluble in acid. The average value of the ash
in pure refined sugar is from a mere trace to O'l per cent, whilst in
raw sugar it may reach 2 or even 3 per cent.

Monier gives the following as the average composition of the ash
of cane and beet sugars : —

Cane Sugar.

Beet Sugar.

Alkaline carbonates
Calcium carbonates
Alkaline sulphates
Sodium chloride
SiOjandAlaOj

16-5

49-0

16-0

9-0

9-5

82-2
6-7

}xw

none

Actual Sugar. — Sucrose may be estimated by a direct polarimetric
reading on the principles given above. If no glucose, or practically
none is present, a direct reading is sufficiently accurate, but in the
presence of reducing sugars, Clerget's inversion process, as described
on page 132, should be used. If necessary, the sugar may be esti-
mated by inversion and the reduction of Fehling's solution, but the
results are not so accurate as the more simple polarimetric method.

Invert Sugar may be estimated by any of the copper or mercury
reduction processes described above. If the solutions are dark
coloured, the polarimetric process is preferable — and probably more
accurate. Any dextrose which may have been added intentionally
will be determined by either process.

It is usually sufficient to return the difference figures as "organic
matter other than sugar ". But for the purposes of the sugar refiner,
it is sometimes necessary to decide whether much gummy or albu-
minous matter be present, as such bodies have a deleterious effect on
crystallization.

The Refractive Index of Sugar Solutions. — If, as is often the case,
the estimation of cane sugar in a solution containing nothing else than
that sugar, be required, the determination of the refractive index will
yield the required information.

The following table, showing the amount of water in syrups as in-
dicated by their refractive indices, is due to Main : —

138

FOOD AND DRUGS.

1 -•

1^
lis

111

1-3330

100

1-3406

94-6

1-3488

89-2

1-3574

83-8

1-3331

99-9

1-3408

94-5

1-3489

89-1

1-3576

83-7

1-3333

99-8

1-3409

94-4

1-3491

1-3577

83-6

1-3334

99-7

1-3411

94-3

1-3492

88-9

1-3579

83-5

1-3336

99-6

1-3412

94-2

1-3494

88-8

1-3581

83-4

1-3337

99-5

1-3414

94-1

1-3496

88-7

1-3582

83-3

1-3338

99-4

1-3415

1-3497

88-6

1-3584

83-2

1-3340

99-3

1-3417

93-9

1-3499

88-5

1-3586

83-1

1-3341

99-2

1-3418

93-8

1-3500

88-4

1-3587

1-3343

99-1

1-3420

93-7

13502

88-3

1-3589

82-9

1-3344

1-3421

93-6

1-3503

38-2

1-3591

82-8

1-3345

98-9

1-3423

93-5

1-3505

88-1

1-3592

82-7

1-3347

98-8

1-3424

93-4

1-3507

1-3594

82-6

1-3348

98-7

1-3426

93-3

1-3508

87-9

1-3596

82-6

1-3350

98-6

1-3427

93-2

1-3510

87-8

1-3597

82-4

1-3351

98-5

1-3429

93-1

1-3511

87-7

1-3599

82-3

1-3352

98-4

1-3430

1-3513

87-6

1-3600

82-2

1-3354

98-3

1-3432

92-9

1-3515

87-5

1-3602

82-1

1-3355

98-2

1-3433

92-8

1-3516

87-4

1-3604

1-3357

98-1

1-3435

92-7

1-3518

87-3

1-3605

81-9

1-3358

1-3436

92-6

1-3519

87-2

1-3607

81-8

1-3359

97-9

1-3438

92-5

1-3521

87-1

1-3609

81-7

1-3361

97-8

1-3439

92-4

1-3522

1-3610

81-6

1-3362

97-7

1-3441

92-3

1-3524

86-9

1-3612

81-5

1-3364

97-6

1-3442

92-2

1-3526

86-8

1-3614

81-4

1-3365

97-5

1-3444

92-1

1-3527

86-7

1-3615

81-3

1-3366

97-4

1-3445

1-3529

86-6

1-3617

81-2

1-3368

97-3

1-3447

91-9

1-3530

86-5

1-3619

81-1

1-3369

97-2

1-3448

91-8

1-3532

86-4

1-3620

1-3371

97-1

1-3450

91-7

1-3533

86-3

1-3622

80-9

1-3372

1-3451

91-6

1-3535

86-2

1-3624

80-8

1-3373

96-9

1-3453

91-5

'1-3537

86-1

1-3625

80-7

1-3375

96-8

1-3454

91-4

1-3538

1-3627

80-6

1-3376

96-7

1-3456

91-3

1-3540

85-9

1-3629

80-5

1-3378

96-6

1-3457

91-2

1-3541

85-8

1-3630

80-4

1-3379

96-5

1-3459

91-1

1-3543

85-7

1-3632

80-3

1-3380

96-4

1-3460

1-3545

85-6

1-3634

80-2

1-3382

96-3

1-3462

90-9

1-3546

85-5

1-3635

80-1

1-3383

96-2

1-3463

90-8

1-3548

85-4

1-3637

1-3386

96-1

1-3465

90-7

1-3549

85-3

1-3639

79-9

1-8886

1-3466

90-6

1-3551

85-2

1-3640

79-8

1-3387

95-9

1-3468

90-5

1-3562

85-1

1-3642

79-7

1-3389

95-8

1-3469

90-4

1-3564

1-3644

79-6

1-3390

95-7

1-3471

90-3

1-3556

84-9

1-3645

79-5

1-3392

95-6

1-3472

90-2

1-3557

84-8

1-3647

79-4

1-3393

95-5

1-3474

90-1

1-3559

84-7

1-3649

79-3

1-3394

95-4

1-3475

1-3561

84-6

1-3650

79-2

1-3396

95-3

1-3477

89-9

1-3562

84-5

1-3652

79-1

1-3397

95-2

1-3478

89-8

1-3564

84-4

1-3654

1-3399

95-1

1-3480

89-7

1-3566

84-3

1-3655

78-9

1-3400

1-3481

89-6

1-3567

84-2

1-3657

78-8

1-3402

94-9

1-3483

89-5

1-3569

84-1

1-3659

78-7

1-3403

94-8

13484

89-4

1-3571

1-3661

78-6

1-3405

94-7

1-3486

89-3

1-3572

83-9

1-3662

78-5

CANE SUGAR.

139

> *>
'J3 eS

■B^

1.^-

SS6

•55 t3 o

-2 .^i

Is?

1.-

ll?

tf^Sg

^^1

«^§

^^i

C«£S

^^§

^^k

1-8664

78-4

1-3757

'13

1-3854

67-6

1-3955

62-2

1-3666

78-3

2-3758

72-9

1-3856

67-5

1-3957

62-1

1-3667

78-2

1.3760

72-8

1-3858

67-4

1-3959

1-3669

78-1

1-3762

72-7

1-3860

67-3

1-3961

61-9

1-3671

1-3764

72-6

1-3862

67-2

1-3963

61-8

1-3672

77-9

1-3766

72-5

1-3863

671

1-3965

61-7

1-3674

77-8

1-3767

72-4

1-3865

1-3967

61-6

1-3676

77-7

1-3769

72-3

1-3867

66-9

1-3969

61-5

1-3677

77-6

1-3771

72-2

1-3869

66-8

1-3970

61-4

1-3679

77-5

1-3773

721

1-3871

66-7

1-3972

61-3

1-3681

77-4

1-3774

1-3873

66-6

1-3974

61-2

1-3682

77-3

1.3776

71-9

1-3874

66-5

1-3976

1-3684

77-2

1-3778

71-8

1-3876

66-4

1-3978

1-3686

77-1

1-3780

71-7

1-3878

66-3

1-3980

60-9

1-3687

1-3782

71-6

1-3880

66-2

1-3982

60-8

1-3689

76-9

1-3783

71-5

1-3882

66-1

1-3984

60-7

1-3691

76-8

1-3785

71-4

1-3884

1-3986

60-6

1-3692

76-7

1-3787

71-3

1-3885

65-9

1-3988

60-5

1-3694

76-6

1-3789

71-2

1-3887

65-8

1-3989

60-4

1-3696

76-5

1-3790

71-1

1-3889

65-7

1-3991

60-3

1-3697

76-4

1-3792

1-3891

65-6

1-3993

60-2

1-3699

76-3

1-3794

70-9

1-3893

65-5

1-3995

601

1-3701

76-2

1-3796

70-8

1-3895

65-4

1-3997

1-3703

76-1

1-3798

70-7

1-3896

65-3

1-3999

59-9

1-3704

1-3799

70-6

1-3898

65-2

1-4001

59-8

1-3706

75-9

1-3801

70-5

1-3900

65-1

1-4003

69-7

1-3708

75-8

1-3803

70-4

1-3902

1-4005

59-6

1-3709

75-7

1-3805

70-3

1-3904

64-9

1-4007

59-5

1-3711

75-6

1-3806

70-2

1-3906

64-8

1-4009

69-4

1-3713

75-5

1-3808

70-1

1-3908

64-7

1-4011

59-3

1-3714

75-4

1-3810

1-3910

64-6

1-4013

59-2

1-3716

75-3

1-3812

69-9

1-3912

64-5

1-4015

59-1

1-3718

75-2

1-3814

69-8

1-3913

64-4

1-4017

1-3719

75-1

1-3816

69-7

1-3915

64-3

1-4019

58-9

1-3721

1-3817

69-6

1-3917

64-2

1-4021

58-8

1-3723

74-9

1-3819

69-5

1-3919

64-1

1-4022

58-7

1-3725

74-8

1-3821

69-4

1-3921

1-4024

58-6

1-3726

74-7

1-3823

69-3

1-3923

63-9

1-4026

58-5

1-3728

74-6

1-3825

69-2

1-3925

63-8

1-4028

58-4

1-3730

74-5

1-3827

691

1-3927

63-7

1-4030

58-3

1-3732

74-4

1-3828

1-3929

63-6

1-4032

58-2

1-3733

74-3

1-3830

68-9

1-3931

63-5

1-4034

58-1

1-3735

74-2

1-3832

68-8

1-3932

63-4

1-4036

1-3737

74-1

1-3834

68-7

1-3934

63-3

1-4038

57-9

1-3739

1-3836

68-6

1-3936

63-2

1-4040

57-8

1-3541

73-9

1-3838

68-5

1-3938

63-1

1-4042

57-7

1-3742

73-8

1-3839

68-4

1-3940

1-4044

57-6

1-3744

73-7

1-3841

68-3

1-3942

62-9

1-4046

57-5

1-3746

73-6

1-3843

68-2

1-3944

62-8

1-4048

57-4

1-3748

73-5

1-3845

68-1

1-3946

62-7

1-4050

57-3

1-3749

73-4

1-3847

1-3948

62-6

1-4052

57-2

1-3751

73-3

1-3849

67-9

1-3950

62-5

1-4054

57-1

1-3753

73-2

1-3850

67-8

1-3951

62-4

1-4056

1-3755

73-1

1-3852

67-7

1-3953

62-8

1-4058

56-9

140

FOOD AND DRUGS.

g So

12^

1-4060

56-8

1-4171

51-4

1-4283

1-4405

40-6

1-4062

56-7

1-4173

51-3

1-4285

45-9

1-4408

40-5

1-4064

56-6

1-4176

51-2

1-4288

45-8

1-4410

40-4

1-4066

56-5

1-4178

51-1

1-4390

45-7

1-4412

40-3

1-4068

56-4

1-4180

1-4292

45-6

1-4414

40-2

1-4070

56-3

1-4182

50-9

1-4294

45-5

1-4417

40-1

1-4071

56-2

1-4184

50-8

1-4296

45-4

1-4419

1-4073

561

1-4186

50-7

1-4298

453

1-4421

39-9

1-4075

1-4188

50-6

1-4300

45-2

1.4424

39-8

1-4077

55-9

1-4190

50-5

1-4302

45-1

1-4426

39-7

1-4079

55-8

1-4193

50-4

1-4304

1-4428

39-6

1-4081

55-7

1-4195

50-3

1-4306

44-9

1-4431

39-5

1-4083

55-6

1-4197

50-2

1-4309

44-8

1-4433

39-4

1-4085

55-5

1-4199

50-1

1-4311

44-7

1-4435

39-3

1-4087

55-4

1-4201

1-4313

44-6

1-4438

39-2

1-4089

55-3

1-4203

49-9

1-4316

44-5

1-4440

39-1

1-4091

55-2

1-4205

49-8

1-4318

44-4

1-4442

1-4093

55-1

1-4207

49-7

1-4320

44-3

1-4445

38-9

1-4095

1-4209

49-6

1-4322

44-2

1-4447

38-8

1-4097

54-9

1-4211

49-5

1-4325

44-1

1-4449

38-7

1-4099

54-8

1-4213

49-4

1-4327

1-4451

38-6

1-4101

54-7

1-4215

49-3

1-4329

43-9

1-4454

38-5

1-4103

54-6

1-4217

49-2

1-4332

43-8

1-4456

38-4

1-4106

54-5

1-4220

49-1

1-4334

43-7

1-4458

38-3

1-4108

54-4

1-4222

1-4336

436

1-4461

38-2

1-4110

54-3

1-4224

48-9

1-4339

43-5

1-4463

38-1

1-4112

54-2

1-4226

48-8

1-4341

43-4

1-4465

38 V

1-4114

541

1-4228

48-7

1-4343

43-3

1-4468

37-9

1-4116

1-4230

48-6

1-4345

43-2

1-4470

37-8

1-4118

53-9

1-4232

48-5

1-4348

43-1

1-4472

37-7

1-4120

53-8

1-4234

48-4

1-4350

1-4475

37-6

1-4123

53-7

1-4236

48-3

1-.4352

42-9

1-4477

37-6

1-4125

53-6

1-4238

48-2

1-4355

42-8

1-4479

37-4

1-4127

53-5

1-4240

48-1

1-4357

42-7

1-4482

37-3

1-4129

53-4

1-4242

1-4359

42-6

1-4484

37-2

1-4131

53-3

1-4244

47-9

1-4362

42-5

1-4486

37-1

1-4133

53-2

1-4246

47-8

1-4364

42-4

1-4489

1-4135

53-1

1-4248

47-7

1-4366

42-3

1-4491

36-9

1-4137

1-4250

47-6

1-4368

42-2

1-4493

36-8

1-4140

52-9

1-4253

47-5

1-4371

42-1

1-4496

36-7

1-4142

52-8

1-4255

47-4

1-4373

1-4498

36-6

1-4144

52-7

1-4257

47-3

1-4375

41-9

1-4500

36-5

1-4146

52-6

1-4259

47-2

1-4378

41-8

1-4503

36-4

1-4148

52-5

1-4261

47-1

1-4380

41-7

1-4505

36-3

1-4150

52-4

1-4263

1-4382

41-6

1-4507

36-2

1-4152

52-3

1-4265

46-9

1-3385

41-5

1-4509

361

1-4154

52-2

1-4267

46-8

1-4387

41-4

1-4512

1-4156

52-1

1-4269

46-7

1-4389

41-3

1-4514

35-9

1-4159

1-4271

46-6

1-4391

41-2

1-4516

35-8

1-4161

51-9

1-4273

46-5

1-4394

41-1

1-4519

35-7

1-4163

51-8

1-4275

46-4

1-4396

1-4521

35-6

1-4165

51-7

1-4277

46-3

1-4398

40-9

1-4523

35-5

1-4167

51-6

1-4279

46-2

1-4401

40-8

1-4526

35-4

1-4169

51-5

1-4281

46-1

1-4403

40-7

1-4528

35-3

CANE SUGAR.

141

S2§

e« «- c

Ilk

Refractive
Index at
20° C.

1-4530

35-2

1-4649

30-1

1-4774

1-4904

19-9

1-4533

35-1

1-4651

1-4777

24-9

1-4906

19-8

1-4535

1-4653

29-9

1-4779 !

24-8

1-4909

19-7

1-4537

34-9

1-4656

29-8

1-4782

24-7

1-4912

19-6

1-4540

34-8

1-4658

29-7

1-4784

24-6

1-4914

19-5

1-4642

34-7

1-4661

'29-6

1-4787

24-5

1-4917 j

19-4 ,

1-4544

34-6

1-4663

29-5

l-47b9 i

24-4

1-4919 I

19-8

1-4547

34-5

1-4666

29-4

1-4792

24-3

1-4922 1

19-2

1-4549

34-4

1-4668

29-3

1-4794

24-2

1-4925

19-1

1-4551

34-3

1-4671

29-2

1-4797

24-1

1-4927

1-4554

34-2

1-4673

29-1

1-4799 1

1-4930

18-9 i

1-4556

34-1

1-4676

1-4802

23-9

1-4933

18-8

1-4558

1-4678

28-9

1-4804

23-8

1-4935

18-7

1-4561

33-9

1-4681

28-8

1-4807

23-7

1-4938

18-6

1-4563

33-8

1-4683

28-7

1-4810

23-6

1-49U

18-5

1-4565

33-7

1-4685

28-6

1-4812

23-5

1-4943

18-4 i

1-4567

33-6

1-4688

28-5

1-4815

23-4

1-4946

18-3 i

1-4570

33-5

1-4690

28-4

1-4817

23-3

1-9449

18-2 1

1-4572

33-4

1-4693

28-3

1-4820

23-2

1-4951

18-1 '

1-4574

33-8

1-4695

28-2

1-4822

23-1

1-4954

1-4577

33-2

1-4698

28-1

1-4825

1-4956

17-9

1-4579

33-1

1-4700

1-4827

22-9

1-4959

17-8

1-4581

1-4703

27-9

1-4830

22-8

1-4962

17-7

1-4584

32-9

1-4805

27-8

1-4832

22-7

1-4964

17-6

1-4586

32-8

1-470^

27-7

1-4835

22-6

1-4867

17-5

1-4588

32-7

1-4710

27-6

1-4838

22-5

1-4970

17-4

1-4591

32-6

1-4713

27-5

1-5840

22-4

1-4972

17-3

1-4593

32-5

1-4715

27-4

1-4843

22-3

1-4975

17-2

1-4595

32-4

1-4717

27-3

1-4845

22-2

1-4978

17-1

1-4598

32-3

1-4720

27-2

1-4848

22-1

1-4980

1-4600

32-2

1-4722

27-1

1-4850

1-4983

16-9

1-4602

32-1

1-4725

1-4853

21-9

1-4985

16-8

1-4605

1-4727

26-9

1-4855

21-8

1-4988

16-7

1-4607

31-9

1-4730

26-8

1-4858

21-7

1-4991

16-6

1-4609

31-8

1-4732

26-7

1-4860

21-6

1-4993

16-5

1-4612

31-7

1-4735

26-6

1-4863

21-5

1-4996

16-4

1-4614

31-6

1-4737

26-5

1-4865

21-4

1-4999

16-3

1-4616

31-5

1-4740

26-4

1-4868

21-3

1-5001

16-2

1-4619

31-4

1-4742

26-3

1-4871

21-2

1-5004

16-1

1-4621

31-3

1-4744

26-2

1-4873

21-1

1-5007

1-4623

31-2

1-4747

261

1-4876

1-5009

15-9

1-4625

31-1

1-4749

1-4878

20-9

1-5012

15-8

1-4628

1-4752

25-9

1-4881

20-8

1-5015

15-7

1-4630

30-9

1-4754

25-8

1-4883

20-7

1-5017

15-6

1-4632

30-8

1-4757

•25-7

1-4886

20-6

1-5020

15-5

1-4635

30-7

1-4759

25-6

1-4888

20-5

1-5022

15-4

1-4637

30-6

1-4762

25-5

1-4891

20-4

1-5025

15-3

1-4639

30-5

1-4764

25-4

1-4893

20-3

1-5028

i 15-2

1-4642

30-4

1-4767

25-3

1-4896

20-2

1-5030

15-1

1-4644

30-3

1-4769

25-2

1-4898

20-1

1-5033

1-4646

30-2

1-4772

25-1

1-4£01

142

FOOD AND DRUGS.

An approximate valuation of crude sugar may be made by making
a saturated solution and taking the specific gravity of the solution at
17 "5°. The percentage of sucrose and its impurities is indicated by
the specific gravity and the following table gives the approximate
values : —

Specific Gravity

Percentage Composition of the Solution.

of Saturated
Solution at

17-5°.

Sucrose.

Impurities.

Water.

1-330

66-66

33-34

1-3322

64-85

2-66

32-49

1-3384

63-70

5-29

31-01

1-3446

62-65

7-76

29-68

1.3509

61-42

10-13

28^45

1-3572

60-28

12-48

27-24

1-3636

59-14

14-67

2619

1-3700

58-00

16-82

25-18

. 1-3764

57-85

18-87

24-28

1-3829

55-70

20-77

23-53

1-3894

54-56

22-59

22-85

1-3959

53-42

24-36

22-22

1-4025

52-28

25-98

21-74

1-4092

51-14

27-56

21-30

1-4159

50-00

29-00

21-00

The table on opposite page gives the specific gravity of pure
sugar solutions (from 1 per cent to 74 per cent) at 17*5° C.

Adulteration of Sugar. — Coarse adulteration of sugar is not now
common, and the old additions of sago and potato flour are now never
met with. The improvement in this respect may be seen when one
finds that prior to 1845 when the duty on sugar varied from 25s. to
63s. per cwt. according to the origin of the sugar, it has been estimated
by competent authorities that from 10,000 to 12,000 tons of intention-
ally added matter was annually sold fraudulently as sugar !

To-day it is accidental impurities such as wood fibres and a small
amount of clay or sandy matter that one meets with — and this princi-
pally in raw sugar.

Occasionally starch glucose in an anhydrous condition is met with.
This is rarely added to refined sugar, but to moist sugar, sold as cofifee
sugar, which may contain, naturally, a little glucose.

But genuine sugar of this type will never give a copper reduction
figure, corresponding to more than 4 per cent to 6 per cent of dextrose ;
hence any greater value obtained by Clerget's process will indicate the
addition of glucose. If the starch glucose contain much dextrose and
maltose, the initial rotation of the sugar may be such as to indicate
more than 100 per cent of sucrose. Such a figure is definite evidence
of adulteration.

Crystals of beet sugar are sometimes dyed in order to resemble the
coloured Demerara crystals which are, of course, more valuable. The

CANE SUGAR.

143

Table showing the Strength of Sugar Solutions by Specific
Gravity at 17-5° C.

Specific Gravity accord- 1

Specific Grayity accord-

Sugar Per

ing

Sugar Per

ing

cent.

Balling.

Niemann.

Balling.

Niemann.

1-0040

1-0035

1-1692

1-1681

1-0080

1-0070

1-1743

1-1731

1-0120

1-0106

1-1794

1-1781

1-0160

1-0143

1-1846

11832

1-0200

1-0179

1-1898

1-1883

1-0240

1-0215

1-1951

11935

7 .

1-0281

1-0254

1-2004

1-1989

1-0322

1-0291

1-2057

1-2043

1-0363

1-0328

1-2111

1-2098

1-0404

1-0367

1-2165

1-2153

1-0446

1-0410

1-2219

1-2209

1-0488

1-0456

1-2274

1-2265

1-0530

1-0504

1-2329

1-2322

1-0572

1-0552

1-2385

1-2378

1-0614

1-0600

1-2441

1-2434

1-0657

1-0647

1-2479

1-2490

1-0700

1-0693

1-2553

1-2546

1-0744

1-0738

1-2610

1-2602

1-0788

1-0784

1-2667

1-2658

1-0832

1-0830

1-2725

1-2714

1-0877

1-0875

1-2783

1-2770

1-0922

1-0920

1-2841

1-2826

1-0967

1-0965

1-2900

1-2882

1-1013

I-IOIO

1-2959

1-2938

1-1059

1-1056

1-3019

1-2994

1-1106

1-1103

1-3079

1-3050

1-1153

1-1150

1-3139

1-3105

1-1200

1-1197

1-3190

1-3160

1-1247

1-1245

1-3260

1-3215

1-1295

1-1293

1-3321

1-3270

1-1343

1-1340

1-3383

1-3324

1-1391

1-1388

1-3445

1-3377

1-1440

1-1436

1-3507

1-3430

1-1490

1-1484

1-3570

1-3483

1-1540

1-1533

1-3633

1-3535

1-1590

1-1582

1-3696

1-3587

1-1641

1-1631

1-3760

1-3658

dye may be detected by suspending silk fibres in a neutral solution of
the sugar, or one slightly acidified with hydrochloric acid.

Many samples of true Demerara sugar contain a trace of tin, due
to the use of chloride of tin to fix the natural colouring matter.

Traces of ultramarine are sometimes to be found. This substance
is added as a corrective to the yellowish colour of poorly refined sugar.
It is detected by dissolving the sugar in water, when the ultramarine
will sink to the bottom in fine particles.

144

FOOD AND DRUGS.

MOLASSES OE TREACLE.

Molasses, treacle, and golden syrup are to be understood as
practically synonymous names for the uncrystallizable syrup, in a
greater or less state of refinement, usually obtained as a secondary
product in the manufacture of sucrose. Molasses contains a con-
siderable amount of sucrose, together with (in molasses from cane
sugar) a good deal of invert sugar, and (in molasses from beet sugar)
impurities such as raffinose and other organic substances. It should
consist of partially hydrolysed sucrose, and should therefore contain
practically no sugars, other than sucrose and invert sugar.

Four samples of each variety, analysed by the author, had the
following composition : —

Cane Sugar Molasses.

Beet Sugar Molasses.

Sucrose.

Glucose.

Water.

Ash.

Sucrose.

Glucose.

Water.

Ash.

Green syrup
Treacle
Golden syrup

Per cent
56-8
38-1
42-1
39-5

Per cent

915

17-2

20-8

23-5

Per cent
241
25-1
23-9
25-2

Per cent
2-2
2-8
2-1
1-4

Per cent
55-4
521
49-9
561

Per cent
1-4
0-8
0-9
1-6

Per cent
22-1
20-9
19-6
23-0

Per cent

101

9-9

13-4

11-8

According to Winter Blyth the ash of ordinary treacle is 8 '21 per
cent and the reducing sugars 12'92 per cent. It is to be noted, how-
ever, that in the more highly refined golden syrup, the percentage of
sucrose is lower than in the green syrups and the proportion of re-
ducing sugars correspondingly higher. Ling and Maclaren give the
following as the sugar values of five samples of cane molasses : —

Sucrose
(Copper Process).

Sucrose
(Clerget).

Invert Sugar.

Per cent

40-2

40-6

17-9

34-8

17-1

28-7

27-8

11-7

35-3

34-9

15-3

65-6

66-0

7-2

(doubtful sample)

Bodmer, Leonard and Smith give the following as representing the
composition of pure golden syrups, or treacles : —

Per cent

Water

160

13-8

16-2

16-3

20-1

16-6

16-9

Ash

2-0

8-2

4-1

2-1

7-9

2-0

1-8

Sucrose

33-3

32-4

25-8

34-1

40-8

28-4

34-9

Invert sugar

45-7

32-5

45-6

45-3

22-0

440

40-8

Other organic matter

3 0

8-3

2-2

9-2

9-1

5-6

Specific rotation

15°

17-5°

11-5°

13°

25°

12-5^

16°

MOLASSES.

145

The analysis of raolassses is conducted on the same principles as
that of ordinary sugars, the principal adulterant to be looked for being
starch-glucose syrup, which is now often sold either slightly flavoured,
or mixed with a little genuine molasses under the name "amber
syrup ".

Water. — This averages from 20 to 30 per cent, and is best determined
by dissolving 10 grms. in water to make 100 c.c. and then drying 10
c.c. ( = 1 grm. of the sample) in a platinum capsule containing recently
ignited sand. This prevents a pellicle forming, which prevents the
water from escaping.

Mineral Matter. — This is determined as described under cane
sugar and averages from 2 to 6 per cent in cane molasses and from 9
to 13 per cent in beet molasses.

Sugars. — The reducing sugars may be determined by Fehling's
solution in the usual manner. On inversion of a genuine molassses,
the rotation will change from a dextro- to a laevo-rotation, whilst
if much added glucose be present the dextrorotation will remain
after inversion, as dextrose is not affected by the inversion.

In determining the copper reducing power, when necessary, the
gravimetric process is preferable, as the solutions are somewhat dark.
Ten c.c. of 10 per cent solution should be used with 20 c c. of
Fehling's solution. By heating with dilute acid to 68° for ten minutes,
little but the sucrose is inverted, so that the difference between the
two reducing powers will enable the sucrose to be approximately cal-
culated. A third determination, after inversion of 10 c.c. of the solu-
tion with 40 c.c. of water containing 1-5 c.c. of strong sulphuric acid
for three hours on a water bath, by which the dextrin and maltose are
inverted, will enable the amount of these substances to be determined.

The presence of dextrin and maltose is strong evidence of adultera-
tion with glucose syrup. Dextrin is indicated by a white precipitate
being formed when 20 c.c, of alcohol is added to 2 c.c. of the 10 per
cent solution of the syrup. Care must be taken, however, with this
reaction as a turbidity is sometimes produced by pure samples.

The following figures will show the difference in the character of
pure golden syrups and of those containing much glucose syrup : —

Specific

Reducing Sugars (as Glucose).

Rotation.

Original

Syrup.

Inversion

at 68°

Inversion
at 100°

for 10 minute.s.

for 3 minutes.

Per cent

Pure syrup

+ 16° 30'

+ 14°

.39-5

76-8

+ 17°

69-5

71-8

Adulterated syrup

+ 70°.

»» M

+ 63°

+ 92°

33-8

71-5

VOL. I.

146 FOOD AND DEUGS.

Bernard Dyer (" Analyst," xxv. 95) gives 42 as the K value which
may be safely used for glucose syrup (i.e. the percentage of reducing
sygars in terms of pure glucose taken as 100), and +113° as the
specific rotatory power. He uses the following formulae for calculating
the percentage of glucose syrup present. He estimates the amount
of sucrose present by a determination of the specific rotation before
and after inversion (see Clerget's process, p. 132) and determines
the cupric reducing power K. If P be the percentage of glucose syrup,
E the specific rotation, [a]^, of the original sample, and S the rota-
tion due to the sucrose, then

p _ 0-206K + (E - S)
1-217

if angular degrees be used, and

p^ 0-31K + (E - S)
1-83

if percentage degrees are used, E here being the percentage reading
before inversion and S the actual percentage of sucrose.

According to Leach the following formula gives approximate re-
sults (allowance being made for the average variable constituents
present) when commercial glucose is present in molasses etc. : —

c = (^ - ^)1QQ

175

where G is the percentage of commercial glucose ; a the direct polar-
ization on the sugar scale for normal weight, and S the percentage of
sucrose as determined by Clerget's process.

The variable composition of both natural molasses or treacle, and
of glucose syrup, prevents 'any absolutely accurate determination of
the amount of each of these substances in a mixture, but approximate
valuations may be made by a determination of the optical values of
the syrup. Glucose syrup consists principally of dextrose, with some
dextrin and maltose. These compounds being dextrorotatory, the
-optical rotation of the syrup at once affords a clue to the presence of
adulteration with glucose syrup. The average specific rotation for
sodium light of genuine golden syrup is about + 16°, and that of
glucose syrup + 110° (Bodmer, Leonard and Smith, " Analyst," xxiv.
252). So that if [a],; be the specific rotation of the sample, the

'.approximate amount of glucose present is — ^}!t — tt^. Bodmer,

100 - lb

Leonard and Smith (loc. cit.) recommend the following process to be

used in examining this syrup : —

Water. — This may be determined with sufficient accuracy by

making a 10 per cent solution of the syrup, and taking the specific

gravity of water as 1000, and d d acting this from the observed specific

gravity of the solution, and dividing the result by 3*86, the amount of

solid matter in the 10 per cent solution is found. Thus if the specific

GLUCOSE. 147

gravity be 1-032 (and it should not be much below this, the solid

32
matter in the original syrup is k-tttt ^ 1^ ^^ S2'9 per cent, the water

3'oo

being 17'1 per cent. Jones (" Analyst," xxv. 87) prefers the figure
4 as the divisor, which would give the water as 20 per cent. Bodmer,
Leonard and Smith then clarify, if necessary, the ten per cent solu-
tion and take its rotatory power, from which the specific rotation is

calculated from the usual formula [a],^ = — - — , where I is the length

% Ic

of the tube in decimetres and c the number of grms. per 100 c.c. If
a 200 mm. tube be used the observed angle has merely to be multi-
plied by 5.

The average specific rotation of pure golden syrup after inversion
is - 12° for sodium light ( - 14" for the transition tint). The approxi-
mate amount of glucose syrup may therefore be calculated, if the
syrup is dextrorotatory after inversion, from the formula
p _ (Wrf after inversion + 12) 100
110 + 12

Matthews and Parker (" Analyst," xxv. 89) calculate the sucrose
in genuine golden syrup by taking the rotation of a 10 per cent solu-
tion before and after inversion with yeast (1 grm. to 50 c.c. at 52° for
five hours — afterwards boiling to destroy bi-rotation). The sum of the
readings in a 200 mm. tube (neglecting their signs) is multiplied by
10 and divided by 5-02 (the divisor for a 1 per cent sucrose solution
in a 200 mm. tube when inverted). This gives the percentage of
sucrose. For the calculation of the reducing sugars by fermentation
processes and for the estimation of the dextrin and maltose the
original paper should be consulted.

It is to be remembered that there are to be found in some samples
of molasses — especially in beet molasses — optically active substances
other than sugars. It is probable that little effect is caused by these
on the ultimate sugar value determined by optical methods as the other
active bodies fairly neutralize each other. Most of such optically
active bodies are best precipitated by a little lead acetate and alcohol.

COMMEKCIAL GLUCOSE.

Under the names of glucose or glucose syrup is usually sold a
syrup, containing a large amount of dextrose, made by the hydrolysis
of starch by acid. The principal variety is that imported from
America, made from maize, starch. Solid glucose (saccharum or
saccharine) is also a regular article of commerce. Apart from the use
of such products of the conversion of starch in the brewing and other
industries, glucose syrup is frequently used as an adulterant of golden
syrup, and of honey. Since the attention of analysts has been drawn
to this, glucose syrup, coloured and slightly flavoured, is often sold
under the name "amber syrup". Glucose has also a legitimate use
in medicine, the British Pharmacopoeia prescribing a syrup of glucose.
The constituents of "glucose" of commerce include true dextrose.

148

FOOD AND DRUGS.

dextrin, maltose and, often, a notable proportion of unfermentable
carbohydrates to which the name gallisin has been given. Gallisin is.
probably a mixture of bodies (gluco-amylins) of the average specific
rotation [a](i = about +82°. According to Valentin (" Journ. b'oc.
Arts." XX. 14, 404), the following represent the compositions of five
samples of good quality commercial glucoses.

Dextrose ....

Maltose

Dextrin . . •.
Unfermentable carbohydrates
Mineral matter
Water

Per
cent
80-00
none
none
8-20
1-30
10-50

Per

cent

58-85

14-11

1-70

9-38

1-40

14-56

Per
cent
67-44
10-96
none
4-30
1-60
15-70

Per
cent
63-42
23-50
none
8-40
1-50
13-18

Per
cent
61-46
13-20
none
8-60
1-60
15-20

These were solid " glucose " of English, French or German make.
The small proportion of dextrin present is inexplicable, especially
when contrasted with the following analyses of Steiner : —

Per

cent

Dextrose

45-40

26-50

76-00

Maltose

28-00

40-30

5-00

42-60

Dextrin

9-30

15-90

—

39-80

The characters of numerous samples of glucose syrup examined
in the author's laboratory are as follows ; —

Spacific gravity .

K value (in terms of dextrose)

1-400 to 1-4370

40-5 „ 66-8

+ 89° „ +108°

For calculating the approximate amount of glucose syrup in golden
syrup, Dyer, as stated above, uses the K value 42, and specific rotation
+ 113''. These values are safe to use, in so far as they give the
mixer any benefit of the doubt. Dextrin is determined by precipitating
1 volume of a 25 per cent solution of the sample, by 10 volumes of
90 per cent alcohol, collecting the precipitated dextrin, washing it with
alcohol and weighing on a tared filter.

According to experiments by Wiley, inversion by dilute acid re-
duces the specific rotation to about + 54° showing that dextrose is the
principal ingredient present after inversion. The K value becomes
from 80 to 90.

If it be necessary to separate the glucose and maltose, the follow-
ing scheme of analysis may be adopted : —

Water and Mineral Matter. — Dry about 1 grm. of the sample to

HONEY.

149

This will give the

constant weight, and then ignite the residue,
water, mineral matter and organic matter.

Beducing Sugars. — Ascertain the K value by reduction of Fehling's
solution.

Specific Rotation. — Determine the specific rotation by using a 20
per cent solution.

Taking the specific rotations as follows : glucose + 53° ; maltose
+ 139-2°; dextrin + 198°, the following formulas can be deduced for
ascertaining the percentage of maltose, glucose, and dextrin. Let M =
percentage of maltose ; G that of glucose and D that of dextrin. Then

(2) G = K-0-62M

(3) D = 0-G-M

where 0 is]the percentage of organic matter, and K the usual cupric
reducing power, and [aj^ the specific rotation of the sample.

According to the Report of the Committee of the American Academy
of Sciences, the following are the average compositions of solid and
liquid " glucoses " : —

(1) M = ([a]

Water
Dextrose

Maltose
Dextrin
Ash

SoUds.

Liquids.

Per cent
1-4 to 17-6
72 „ 73-4
0 „ 3-6
4-2 „ 91
0-33 „ 0-28

Per cent
14-2 to 22-6
34-3 „ 42-8

0 „ 19-3
29-8 „ 45-3
0-32 „ 1-06

HONEY.

Although a saccharine substance is secreted by several species of
the Hymenopterae, commercial honey is, in at all events nearly every
case, the production of the bee. Apis mellifica.

Honey consists chiefly of a concentrated solution in water of
various sugars, dextrose and levulose being the principal. In certain
honeys cane sugar (sucrose) is present. According to Bell, there is
present from 5 per cent to 10 per cent of a substance which is with
difficulty hydrolysed to glucose, and which may be maltose. Mannite
is also present, and up to 3 per cent or thereabouts of dextrin, princi-
pally in the form of achroo-dextrin. A small amount of wax is
normally present, and flavouring matters, a small amount of mineral
matter, and some pollen grains.

The average composition of a normal honey is as follows : —

Per cent

Dextrose

. 25 to 40

Levulose

. 30 „ 45

Mannite

about 1 „ 3

Mineral matter

. 0-1 „ 0-3

Water .

. 15 „ 25

Sucrose .

. 0 „ 8(]

8 (11 per cent is found occasionally)

150 FOOD AND DEUGS.

It is often stated that in normal honey the dextrose and levulose
are present in about equal parts, but that when it has crystallized in
the comb there will be an excess of levulose, and the honey is laevo-
rotatory. It is said also that all honey will crystallize, and those
remaining syrupy are adulterated. This however is not the fact, as
many syrupy honeys are certainly genuine.

In the analysis of honey, a microscopic examination will afford
useful information. Pollen grains, fungus spores, tiny fragments of
wings, etc., are to be observed. This is sometimes of importance, as
there are honeys, so called, to be found, which are entirely factitious,
and the total absence of pollen grains would be a strong indication
of adulteration. Starch is not present. The usual adulterants are
other carbohydrates. The chief of these is ordinary glucose syrup,
obtained by the hydrolysis of starch, or invert sugar. Cane sugar is
sometimes used, and possibly molasses, but the strong taste of the last
named renders it an improbable adulterant to-day.

Moisture. — The water of genuine honey varies from 15 to 25 per
cent. It is best determined in the following manner : About 3 grms.
of the honey are weighed in a shallow platinum dish, and 3 c.c. of
absolute alcohol added, and the whole mixed to a thin solution, a
weighed quantity of freshly burned sand (about 3 grms.) is then added,
and the dish left for an hour or so on a water bath. A few c.c. of
alcohol are again added, and the dish again dried to a constant weight.

Ash. — The ash of genuine honey should not exceed 0*3 per cent
or at most 0'4 per cent. The afeh of artificial or adulterated honey is
often higher than this on account of the fact that starch glucose very
frequently contains calcium sulphate. If the ash be higher than 0*3
per cent, it should be digested with water, filtered and tested with
barium chloride. The presence of more than the faintest trace of
sulphates is strong evidence of the adulteration with starch glucose.
The ash of honey mixed with invert sugar is, however, 'usually very
low.

The British Pharmacopoeia allows 0'25 per cent of ash as a maxi-
mum for purified honey.

Carbohydrates. — The hydrolysis of starch to glucose is usually
carried to the point at which iodine gives a red reaction with the
product, so that erythrodextrin and amylodextrin are usually present.
This affords a useful means of deciding whether starch glucose is
present. For even if the starch glucose contain none of the two
dextrins mentioned above, it will still contain bodies of a dextrinoid
nature, which although not precipitated by alcohol, yield barium com-
pounds insoluble in methyl alcohol. Natural honey gives no such
precipitate in most cases, and even in the most unfavourable case,
that of coniferous honey, not more than 2*5 per cent. For a quali-
tative determination, 5 c.c. of a solution containing 20 grms. of honey
in 100 c.c, are shaken with 2 c.c. of a 2 per cent solution of barium
hydroxide and 17 c.c. of methyl alcohol. The precipitate should be
collected on a Gooch filter, washed with 10 c c. of methyl alcohol,
then with 10 c.c. of ether, and dried at 60° C. More than a trace of
precipitate is suspicious, and if the honey is not a coniferous one.

HONEY.

151

anything over 25 per cent will indicate the addition of starch glucose.
More than 2*5 per cent will certainly indicate this adulteration in all
cases. Molasses, if present, is best detected by examining the honey
for raffinose. Five c.c. of a 25 per cent solution of the honey is
treated with 2-5 c.c. of solution of basic acetate of lead (30 per cent),
and 22-5 c.c. of methyl alcohol. Pure honey gives about 0-5 per cent
to 1-5 per cent of precipitate, whilst molasses gives 50 per cent to
70 per cent. The specific gravity of pure honey is from 1-4150 per
cent to 1*430 per cent, rarely up to 1-440 per cent.

The most important method of examination, however, is the de-
termination of the optical rotation of the honey, both before and after
inversion. It is to be noted, however, that in any such determina-
tion, the solutions must be allowed to stand for several hours, before
the reading is taken, as they exhibit some amount of bi-rotation. The
specific rotatory power S, is calculated from the formula

S =

100a

where a is the observed angle of rotation in a 100 mm. tube, I is the
length of the tube in decimetres, and c is the number of grammes of
substance in 100 c.c. A 200 mm. tube is most convenient for the
reading, when, if the observed angle is used in the above formula, the
value of I becomes two. The usual limits given for honey are not
strictly correct, as since it has become known that cane sugar is a
normal constituent of some honeys, a wider limit is necessary. Taking
20 per cent of water as an average for honey, the following adulter-
ants give the specific rotatory powers before and after inversion.
The figures for honey are also added : —

Original substance
After inversion ^

Cane Sugar and

20 Per cent of

Water.

Invert Sugar and

20 Per cent of

Water.

Glucose
Syrup.

Genuine
Honey.

+ 53-2°
- 19-5°

- 18-4°
- 18-4°

+ 90° to 102°
+ 45° „ 50°

+ 5° to -8°
hardly altered

It is to be remembered that most honey has a specific rotation
within the limits + 3° and - 3°, but that where cane sugar is normally
present, as is undoubtedly the fact with certain honey ,♦ the rotation
may be -l- 5° or a trifle over for the honey, and naturally, inversion
will alter this figure to even - 1° or thereabouts. It is also true that
where a honey has crystallized in the comb, it may be more highly
laevorotary than even - 8°, although this is rare. So that unless a
wide difference from the above limits is noted, care must be exercised
in forming an opinion. For example, a strong dextrorotation, reduced
on inversion to a laevorotation, would indicate the , presence of cane
sugar. Glucose is indicated by a very high dextrorotation, which is

^ Inversion by heating a 20 per cent solution with 10 per cent of hydrochloric
acid for ten to twelve minutes to 70° C. The specific rotation is for sodium hght.

152

FOOD AND DRUGS.

reduced to about one half by inversion, but the heating should be for
twenty minutes to half an hour in this case. Mixtures of cane and
invert sugar or invert sugar and glucose may have correct initial
rotations but a mixture of invert sugar and glucose of specific rotatory
power about that of honey will undergo a greater change on inversion
than most honey does.

Jt is generally understood that honey derived from flowers is
always laevorotatory, and that from conifers dextrorotatory. This is
not correct, however, as it is easy to produce a dextrorotatory honey
by artificial feeding of the bees.

The optical rotation of the honey after fermentation by yeast is a
figure sometimes determined, but as it takes considerable time, and
possesses no advantage over inversion by means of acids, it is very
rarely, if ever, resorted to.

Haenle ("Analyst," xvi. 79) states that if a solution of one part of
honey in two parts of water (in the case of a lasvorotatory honey) be
dialysed for twenty-four hours, and the solution remaining in the
dialyser be then examined, it will never have become dextrorotatory,
unless glucose be present, when it will be found to be dextrorotatory.

Clerget's process (see under sugar) will give a pretty accurate de-
termination of the amount of cane sugar present.

The reducing power of honey when boiled with copper oxide {v,
p. 124) affords considerable information as to the purity of the sample.
The following are the cupric oxide reducing values (K) of the usual
adulterants, that of dextrose determined gravimetrically being taken
as 100 :—

Original value

Value after inversion with acid

Cane Sugar
and 18 Per
cent Water.

Invert Sugar
and 18 Per
cent Water.

Glucose
Syrup.

Genuine
Honey.

86-3

82
82

50 to 55

80 „ 85

60 to 80
little altered

Determined by Fehling's solution pure honey should give a result
equivalent to about 60 to 80 per cent of glucose (see under sugar).

Sieben (*' Analyst," x. 34) recommends the determination of the
optical rotation of the fermented honey, which if pure is nearly 0°,
whereas with starch-glucose it is highly dextrorotatory. The author
has found that no information is yielded by this, that is not given by
inversion with acid. The process recommended by the same author
depending on the determination of carbohydrates which do not reduce
sugar involves several unwarrantable assumptions and is not reli-
able.

According to Ditte (" Zeit. Untersuch. Nahr. Genussm." 1909, 18,
625 - 649) the genuineness or otherwise of a sample of honey is best
ascertained by estimating the total nitrogenous matter precipitated
by tannin, and the amount of cane sugar present. The last mentioned
may be calculated from the polarization of the honey before and after
inversion. The results obtained on the estimation of the amounts o f

HONEY.

153

total solids, invert sugar, acidity, and ash present in the sample affords
little evidence of the presence or absence of artificial honey. The
following figures were obtained on the analyses of fifty-two samples of
honey, and from the results these are classified as pure honey (twenty-
four samples), artificial honey (six samples), honey of suspicious quality
•(four samples), and adulterated honey (eighteen samples) : —

■ 1
<

•:^u9o laj
a3Bj9Av

-* iH ic cq o 00
eo 00 O -* iH fH

«b « 6 oDO 6

r-i C-

-6-20
-7-67

0-28

ranuiiXBpi

00 O "•* OS
O O CO OS (M 00
00 ■* O O CO w

os?b 6 oD 6 6

rH C^

-7-50
-8-35

0-45

•:;U90 J9J

ranuiiuij^

W •* 00 o »o »c

«0 00 O CO -^ CO

(N<5s6?'9 r"

r-( «0 i-H O O

-4-36
-6-80

la90 J9d
.93BJ9AV

C- -* 00 O O "*

»o 00 c^ w:> CO 'Jt*

rH t^ O CI O O

-5-77
-6-93

•;U90 19 J

uinuiix'Bj^

11 I> O 00 o o

-6-39
-7-43

100

Tunraiaij\[

CO -^ ^ O CO »H
« O O »0 O <M

CO 05 o o o o

fH t-

-6-31
-6-61

0-70

•%n90 J9<I
9S«I9AV

CQ O !>• t^

o 00 »o cq CO -^

OS 00 O CO tH rH

i) CO o do 6 6

i-l CO

M5 CO

+ 1

•!»U90 J9J

ranuiixBi^

C<J '"i^ OS 00 »c c2
■Tl< C- O t- Cq rH

-3-71

-8-18

•^U90 J9 J

uinraiuii\[

X O CO Tj<
O 00 fH »C !>• r-t
CO CO O Ol O tH

M5 CO O CO 6 6
iH lO

+ 7-73

•^U90 19 J

93«a9Ay

00 CO 00 OS OS OS
fH CO O CO r-i CO

t-- (N O (N O O
iH t^

-5-50
-6-57

1-49

•!jn90 I9d

rantuiXBj^

CO O Mt CO

-9-54
-10-32

3-95

•;U90 19 J

ranramii\[

CO O CO Tjt

CO i) o o o o

+ 1-25
-1-19

0-90

Water ....
Invert sugar .
Acidity (as formic acid)
Cane sugar .
Ash ... .
Total protein
(Nx6-25)

Polarization of 25 Per cent

Honey Solution in

'200 Mm. Tube.

Before inversion .
After

Vol. of tannin precipitate
in CO.

154 FOOD AND DRUGS.

The tannin precipitate is obtained by placing 10 c.c. of filtered 20
per cent honey solution in a graduated tube, adding 35 c.c. of water,
and then 5 c.c. of 0'5 per cent tannin solution. The contents of the
tube are well mixed, allowed to stand for twenty-four hours, and the
volume of the precipitate then read off ; in the case of pure honey the
volume of the precipitate is never less than 0*9 c.c.

Sugar of Milk.

Lactose or milk sugar, is a sugar prepared from the whey of
curdled milk, and is employed, inter alia, to a very large extent in the
manufacture of infant foods. It is official in the British Pharma-
copoeia, that authority prescribing that it should have the following,
character : —

It contains 1 molecule of water of crystallization, having the for-
mula G12H22O11H2O. It is soluble in 7 parts of cold water and in
about 1 part of boiling water. It should not yield more than 0*25
per cent of ash. One grm. dissolved in 10 c.c. of water, to whichi
three drops of decinormal solution of sodium hydroxide have been
added, give a red colour with phenol -phthalein. The water of crystal-
lization is driven off by heating to 130" to 135° for some time.

In general, these characters, together with the optical values of
the sample are sufficient to determine the purity of the sample.

Lactose exhibits the phenomenon of bi-rotation (see p. 128). In
its stable condition, after standing for a time in solution, the specific
rotation is [a]rf = + 52° for the crystals or + 55'8° for the an-
hydrous sugar.

On hydrolysis by boiling with 20 per cent acid for several hours,
it is converted into galactose and probably another sugar, which have
an average specific rotatory power of [aja = + 67'5°. (It is to be re-
membered that 1 molecule of lactose combines with 1 of water in.
yielding 2 molecules of six-carbon sugars.)

Lactose readily reduces Fehling's solution, 10 c.c. of the latter
requiring 0*0685 grm. of lactose ; or, if a gravimetric process be
employed, 1 grm. of cupric oxide =0*6153 grms. of anhydrous,
lactose.

In estimating lactose by means of Fehling's solution, the greatest
accuracy is obtained, according to Muter, by first ascertaining the
approximate amount required, and very rapidly bringing the Fehling's
solution and the sugar together when both are boiling.

The determination of lactose in milk is described on p. 55 under
milk.

It may here be convenient to give the analyses of a few typicall
infant foods.

MALTOSE AND MALT EXTRACT.

155.

Carbohydrates

HaO.

N X 6-25.

Fat.

mostly Lactose
with some Starch.

Ash.

P2O5. ^

Per

cent

10-7

18-6

66-6

3-95

11-29

10-43

1-10

75-62

0-96

0-29 ,

5-08

9-67

0-34

86-34

2-02

006

4-27

13-2

1-70

79-85

1-09

0-12

7-06

8-70

1-38

81-54

0-64

006

6-81

10-79

1-06

78-61

0-91

0-86

MALTOSE AND MALT EXTRACT.

Maltose is an important sugar which, although it does not come-
before the analyst as such very often, is of interest and importance,,
and may here be considered, together with extract of malt. The action
of diastase on starch matters, such as rice, malt and various grains,,
results in a mixture of maltose and dextrin, both of which are con-
verted into dextrose by hydrolysis with acids. To completely hydro-
lyse maltose by means of dilute acid, it is necessary to allow the
reaction to proceed .for three to four hours at 100° C. Neither
diastase nor invertase hydrolyse maltose, hence a determination of
sucrose in the presence of maltose is possible.

Maltose is bi-rotatory, and all solutions of this sugar should be
kept for several hours before being examined polarimetrically. The
specific rotatory power of anhydrous maltose varies slightly with the
concentration, but for all practical purposes [a]d may be taken
as +138°.

Maltose reduces Fehling's solution, the K value, however, being
only 62. Ten c.c. of Fehling's solution oxidize 0*081 grm. of maltose^
If a gravimetric estimation be used, the CuO x 0"7314 gives the amount
of maltose.

Extract of Malt. — A thick viscous extract of malt is now a com-
mercial article prepared on a very large scale by the evaporation of
an infusion of malt at a very low pressure and at a sufficiently low
temperature not to destroy the properties of the diastase. Extract of
malt is used largely in medicine, either alone, or more often in com-
bination with oils, such as cod liver or olive oil.

In examining an extract of malt, the most important determina-
tion is its diastatic power, as it owes its value principally to its power
of converting starch into easily assimilable carbohydrates. Diastase
is an unorganized ferment formed during the germination of barley
and other grains. In converting the starch, diastase changes | of it
into maltose and ^ into dextrin, which is slowly further converted into
dextrose. According to Ling and Baker, the starch is gradually con-
verted into maltose and a series of malto-dextrins of gradually de-
creasing molecular weight and optical activity, but of higher reducing
powers.

156

FOOD AND DRUGS.

The composition of pale malt is shown by the following analyses
by O'Sullivan, and this of course is in close relationship to the com-
position of the solid matters of extract of malt : —

Starch

Other carbohydrates (of which 60 to

70 per cent are fermentable sugars)

Cellular matter . . . .

Fat

Albumenoids

Ash

Water

11.

Per

cent

4415

45-13

21-23

19-39

11-57

10-09

1-65

1-96

13-09

13-80

2-60

1-92

5-88

7-47

The complete examination of extract of malt should include the
following determinations : — water, total solid matter, mineral matter,
sugars, diastatic value (and, of less importance, dextrin, proteids and
phosphoric acid).

V/ater and Total Solid Matter. — These may be determined by dry-
ing 10 c.c. of a 10 per cent solution in a platinum dish on recently
ignited sand. Approximate accuracy may be obtained by calculating
from the specific gravity of a 20 per cent, i.e. 20 grms. in 100 c.c,
solution of the extract in water. The following formula will give the
percentage of solid matter : —

(D - 1000) X 1-3

Where D is the specific gravity (water = 1000) at 15*5°.

A good malt extract should not contain less than 70 per cent of
solid, matter, as diastase does' not keep well in weaker solutions unless
a preservative be added which usually destroys the activity of the
■diastase. A thin fluid extract of malt exists which often contains
alcohol.

Mineral Matter. — This is best determined on 5 grms. of the
sample — which is nearly dried in an air oven at 110° and then carefully
ignited. The carbonized mass requires breaking to get a pale ash
within any reasonable time.

Sugars. — The reducing sugars are determined by using 2 c.c. of a
10 per cent solution and 30 c.c. of Fehling's solution, preferably using
the gravimetric process. Cane sugar is determined by making 20 c.c. of
■a 10 per cent solution up to 100 c.c, warming to 55° C, and adding
about 0*2 grm. of yeast (dried between blotting paper). O'l per cent
of thymol should be added to prevent fermentation. The whole is
allowed to digest at 55° for four hours (the loss by evaporation being
made up) and the reducing .sugars are then determined in 10 c.c. of
the filtered liquid. The difference in this and the preceding deter-
mination is due to glucose formed by inversion of the cane sugar,
which is thus easily calculated.

MALTOSE AND MALT EXTEACT. 15T

In most cases it is sufficient to assume that the whole of the original
reducing sugar consists of maltose, but this is not strictly accurate,
as about 12 per cent to 15 per cent consists of glucose. If it be
necessary to determine the amount of glucose present — and more than
about 16 per cent should be viewed with suspicion as indicating added
glucose — the only practicable manner is to determine all the constitu-
ents other than reducing sugars (i.e. the cane sugar — dextrin, proteids
water and ash) and call the difference of these and 100 the total
reducing sugars = S. Let W = the weight of cupric oxide precipi-
tated by 100 grms. of extract.

But 1 grm. of maltose precipitates 1"37 grms. of CuO, and 1 grm.
of glucose precipitates 2-21 grms. of CuO.

Let the percentage of maltose = M, and of glucose = G.

Then :—

M + G=S

and

1-37 M+ 2-21 G=W.

From these two equations, the values of M and G may be cal-
culated. The average amount of maltose in a well-prepared extract of
malt is over 50 per cent, and of glucose, according to Ling, from 12
per cent to 20 per cent.

Diastatic Value. — The usually adopted method of determining the
diastatic value of malt extract is that of Lintner, and the results are
expressed in " degrees Lintner ". On Lintner's scale, the diastatic
capacity of a malt or its extract is regarded as 100, when 0*1 c.c. of a
5 per cent solution converts enough starch in an hour at 70° F. to
reduce 5 c.c. of Fehling's solution.

The valuation is carried out as follows : Soluble starch is pre-
pared by treating ordinary starch (e.g. potato starch) with 7'5 per
cent hydrochloric acid for seven days at ordinary temperatures It is
washed with cold water until every trace of free acid is removed, when
it is dried. It is then soluble to a clear solution in hot water.

A 2 per cent solution of soluble starch is prepared, and cooled.
To ten test tubes are then added 10 c.c. of the solution and the tubes
numbered 1 to 10. A 5 per cent solution of the malt extract is then
made and quantities of 0*1 c.c, C"2 c.c. and so on up to 1 c.c. are
added to the tubes 1 to 10. They are now allowed to stand for one
hour in a water bath at 70° F. and then 5 c.c. of Fehling's solution
are added to each tube. The tubes are shaken up, and stood in
boiling water for ten minutes and the copper oxide allowed to subside.
The colours of the tubes are then noted. Some will be found to be
yellow, all the copper being reduced, and others will still be blue. If
tube 5 be completely reduced and tube 6 be still slightly blue, the
amount of the 5 per cent solution necessary to convert the starch re-
quired to reduce the Fehling's solution is obviously between 0-5 c.c.
and 0"6 c.c. The amount of blue colour left in the tube can be fairly
well judged by a little practice — so that the second place of decimals
can be gauged with fair accuracy. If the test tube gives 0*55 as the
amount necessary, then the apparent diastatic power will be

158 FOOD AND DEUGS.

or IS'S", subject, however, to a correction for the amount

0-55
•of reducing sugar, actually present in the small amount of malt
extract used in the experiment. This is determined by a previous
estimation, and a deduction made. The deduction usually given
is that deduced from a determination of the " Lintner value" of the
reducing sugars of the extract, that is, for example, if 1*5 c.c. of the 5
per cent solution reduced 5 c.c. of Fehling's solution, the Lintner

value would be ■ — — or 6"6°. But as a different amount of

the malt solution is used in each tube, this correction is obviously
inaccurate. The proper correction is to deduct from the 5 c.c. of
Fehling's solution reduced, the number of c.c. that would be' reduced
by the amount of sugars in the quantity of malt extract in the com-
pletely reduced tube. If, for example, 0"5 c.c. of malt extract solution
were judged to be the proper quantity, and a previous experiment
showed that this amount contained sugars sufficient to reduce 1-5 c.c.
of Fehling's solution, then the true amount of Fehling's solution re-
(duced through the diastatic action is only 3*5 c.c, and the observed

3*5

.apparent diastatic value must be multiplied by -— . Very high grade

malt extracts will show a Lintner value of well over 50 but these

;are rare.

If the diastatic value is very high it may be necessary to use a
•weaker solution of the extract of malt and make the corresponding
calculations.

Gadd has found that good commercial extracts of malt will convert
2| times their weight of potato starch (not soluble) at 100° F., in from
four and a half minutes to fifteen minutes.

Takamine suggests the use of standard solutions of taka-diastase
as a means of determining the value of malt preparations. Taka-
diastase is prepared from a fungus cultivated on wheat bran and its
•diastatic value is determined by Lintner's method and a standard is
■so fixed. The details of the process are as follows : —

Standard Solution of Taka-diastase. — A 10 per cent solution of the
^standardized ferment in distilled water is made.

Starch Solution. — Fifty grms. of neutral potato starch are made
in the usual manner into a 5 per cent solution, by rubbing the starch
•down with cold water and then boiling for two minutes.

Iodine Solution. — A 1 per cent solution of iodine in water (1 grm.
•of iodine and 2 grms. of KI are dissolved in a little water and made
up to 100 c.c).

Eight test vessels are used, to each of which is added 100 c.c. of
•the starch solution, the whole being kept in warm water at 104° F.
Into the first test glass 1 c.c. of the liquid to be tested is added ; into
the second, 1 c.c of standardized taka-diastase solution, into the third
2 c.c. of the solution, and so on. All the liquids are well stirred and
when the starch is limpid — the conversion being very rapid with taka-
diastase — in about ten minutes, a drop of each is mixed on a tile with

MALTOSE AND MALT EXTRACT.

159

a, drop of iodine solution and the colour of the sample to be tested is
matched with the nearest of the other seven — which range from blue
to purple brown. This gives a comparison of the diastatic value of
the sample in terms of the standard taka-diastase solution, which has
itself been standardized in terms of starch converted.

The author does not find that nearly so concordant results are
•obtained by this method as by Lintner's.

Ling (" Analyst," xxix. 244) gives the following figures for a
number of commercial extracts : —

Per

cent

Cent

cent

Specific gravity at 15-5^

1-8957

1-3951

—

1-4084

1-8778

Maltose ....

31-1

30-9

24-8

27-4

84-2

25-2

Dextrose ....

17-2

18-2

19-1

12-5

Dextrin ....

9-8

8-6

9-8

9-9

6-7

Ash

1-45

1-49

1-58

1-64

1-84

1-64

Water ....

24-30

24-67

27-36

24-84

24-38

29-52

Diastatic value .

30-8°

27-2°

32-3°

25-6°

39-2°

46-5°

Specitie rotatory power

91-8°

90-5°

84-2°

86-8°

94-5°

81-1°

Harrison and Gair (" Chemist and Druggist," 1906, ii. 180) give
the following analyses, which are of interest as many of them repre-
sent well-known brands of malt extract, and it is to be noted that the
diastasic value is not expressed in the conventional manner. The
methods of analysis adopted for those results are given below, includ-
ing the determination of diastasic value : —

Sample.

Total Solids

Maltose

Proteids

Diastasic

Remarks.

Per cent.

Value.

78-2

65-4

7-0

468

79-8

64-4

5-0

346

III

69-8

58-5

4-1

856

. 77-0

54-0

3-6

72-3

52-1

3-8

95-9

82-1

5-7

Solid extract

VII

76-8

66-0

5-4

VIII

74-3

62-5

5-2

Considerable salicylate
present

73-0

47-1

3-8

9-5 per cent of cane sugar
present

66-2

49-7

3-9

—

78-7

74-2

5-5

268

High maltose figure pro-
bably due to glucose

XII

64-9

58-8

3-9

—

XIII

73-9

63-6

6-6

187

—

Total Solids. — Twenty grms. of extract were dissolved in water
and made np to 100 c.c, and the specific gravity of the solution de-

160 FOOD AND DRUGS.

termined. The percentage of total solids in the extract was found by
the fornaula : —

^ o _ specific gravity - 1000 .

i. b. 3.92 ■^^-

Maltose. — Five c.c. of the solution as used for specific gravity were
diluted to 100 c.c. ; then 10 c.c. of Fehling's solution were diluted with
40 c.c. of water and boiled in a porcelain beaker, and the malt solution
run in from a burette until exactly all the copper was reduced.
Since 10 c.c. of Fehling's is reduced by 0-0805 of maltose, the per-
centage of maltose in the extract is given by the expression

where m stands for the number of c.c. used.

Proteids. — Total nitrogen was determined by the Kjeldahl-Gunning
method and the result multiplied by 6*3 was taken as proteid.

Diastase. — Dr. H. A. D. Jowett's excess of starch plan was fol-
lowed, potato starch being used. An amount of starch is taken con-
taining 1 grm. of the anhydrous substance, mixed in a mortar, with
a few c.c. of cold water and poured into 65 c.c. of boiling water.
The mortar is rinsed with a little more water to make 15 c.c. in all,
or a total of 80 c.c. of mucilage, which is boiled for about a minute
to ensure complete gelatinization. The mucilage is then cooled to
46° C, and to it is added 20 c.c. of the same solution of malt extract
as was used for the titration of maltose. This solution contains I'O
of extract in 100 c.c, so that the quantity of extract taken to digest
the starch is 0 2 grm. The mixture is then kept at 40° C. for exactly
half an hour, then boiled to stop the action going further. The liquid
is then cooled and adjusted to measure 100 c.c, and 100 cc. of
Fehling's solution is titrated with this as described under maltose.
From the maltose found it is necessary to deduct the maltose intro-
duced into the extract. The calculations may be combined by the
use of the following formula : —

Weight of anhydrous starch _ ^.^g. /8-05 _ 1-61\
completely converted ~ \ n m )

where n is the number of c.c. used in the last titration, m (as above)
is the c.c. used in the former maltose titration, and 1*104 is the factor

— — - for calculating maltose into starch.
o4'4

The diastatic power may be conveniently expressed numerically
by the weight of starch converted by 1 part of the extract, or, to avoid
fractions, by 100 parts. The figures given in the table for diastase re-
present accordingly the percentage of starch which the extract is
capable of completely converting in half an hour at 40° C. Since 0*2
grm. is the weight of extract taken for the test, the above result must
be multiplied by 500, or

Diastatic value = 592 ( ).

\ n m /

Ling (" Chemist and Druggist," 1910, 11. 52) has criticised Harrison's

MALTOSE AND MALT EXTRACT.

results in a very able paper. He shows that various considerations
necessitate a modification in the formula advocated by Harrison and
Gair, and the importance of his work justifies the reproduction of this
paper in extenso.

He points out that it is well known that when the reactions which
most etizymes bring about are graphically expressed on a co-ordinate
system, they appear, in their initial stage, as linear functions of the
time or mass ; also that after a certain point has been reached, the
curves representing them are logarithmic ; or, in other words, the
reactions then follow the law of mass action. It was the recognition
of these facts by J. Kjeldahl (" Comptes Rend.," Carlsberg, 1879, 1, 109)
with regard to the reaction between malt diastase and starch paste
that rendered diastasimetry — or, to speak more correctly, the measure-
ment of diastatic power — possible.

Kjeldahl showed (loc. cit.) that when time and temperature are
constant, the starch-hydrolysing power of a malt or malt extract is
directly proportional to its mass ; provided that not more than some
40 per cent of maltose, estimated by the cupric-reduction method,
and calculated on the original starch, is produced. It is after this
point that the reaction becomes logarithmic. Kjeldahl proposed a
method of measuring the diastatic power of malt based on these
observations, but it was not altogether satisfactory, and a modifica-
tion was proposed by C. J. Lintner in 1887 (''J- P^- Chem.,"
1887, [2], 34, 375). In this method soluble starch prepared by the
action of 7*5 per cent hydrochloric acid on potato starch was used in-
stead of ordinary potato starch as substrat. Still another modification
of the method was devised by Ling in 1900 ("J. Fed. Inst. Brewing,"
1900, 6, 355), and this is now almost exclusively employed by brewers
in the United Kingdom, it having been adopted by the Malt Analysis
Committee of the Institute of Brewing (see "J. Inst. Brewing," 1906,
12, 6). In consists in extracting 25 grms. of finely ground malt with
500 c.c. of distilled water at 21° C. for an hour, and allowing a portion
of the filtrate to act at 21° C. for an hour on 100 c.c. of a 2 per cent
solution of soluble starch prepared by Lintner's method. After this
10 c.c. of N/10 caustic potash is added to stop further diastatic action,,
the liquid is made up with water to 200 c.c, and titrated against 5 c.c.
of Fehling's solution. The diastatic power is calculated on Lintner's,
standard, the value of 100 being assigned to a malt O'l c.c. of the
extract of which, after acting on starch solution under the above
conditions, exactly reduces 5 c.c. of Fehling's solution. The results
are calculated by the formula

in which D is the diastatic powder of the malt, x is the number of cubic
centimetres of malt extract contained in 100 c.c. of the fully-diluted
starch-conversion liquid, and y equals the number of cubic centimetres
of the same liquid required for the reduction of 5 c.c. of Fehling's
solution. It is important that the volume of malt extract employed
for the conversion shall be less than would produce a reducing power
VOL. I. 11

162 FOOD AND DRUGS. > ^ 3

exceeding the limit of Kjeldahl's law {vide ante). In the case of
concentrated malt extracts an aliquot portion of a 5 per cent solution
would be employed.

Harrison and Gair's method, which has since been adopted by the
" Pharmaceutical Codex," aims at determining diastatic power by
measuring the amount of starch dissolved in half an hour at 40° C. by
a standard solution of malt extract. A weight of potato starch cor-
responding with 1 grm. of the anhydrous substance is made into a
paste with a convenient quantity of water. The paste is rinsed into
100 c.c. measuring flask with water, so that its total volume does
not exceed 80 c.c. After cooling to 40° C, a solution, at the same
temperature, of 0*2 grms. of the malt extract in somewhat less than
20 c.c. of water is added, and the mixture is then kept at 40° C. for
half an hour, when it is boiled to arrest further diastatic action, cooled,
made up to 100 c.c. and titrated against 10 c.c. of Fehling's solution.
From the maltose found, that present in the amount of malt extract
used is deducted, this being estimated by a separate experiment ; the
diastatic power is expressed by the weight of starch converted by 100
grms. of the extract.

In order to understand the formula by which the results may be
calculated on Harrison and Gair's assumptions, it will be necessary
to consider a few addditional points in the literature. It was shown
by H.»T. Brown and J. Heron ("C. S. Trans." 1879, 35, 634) that at
temperatures below 60° C, starch paste is rapidly hydrolysed by malt
diastase until the products show the following constants, calculated
on the total solid matter in the solution : — ^

[a]., 150°

R (percentage of apparent maltose) 80

Beyond this point further change takes place comparatively
slowly (see also Brown and Millar, " C. S. Trans." 1899, 75, 315).

Since every grm. of starch hydrolysed yields 1-05 grms. of ap-
parent maltose, 80 per cent of maltose will correspond with an in-
crease in solid matter on the original starch of 84 per cent and the

factor necessary for calculating maltose into starch will be -^ = 1'19.'-

The simplest formula by which the results can be calculated on these
assumptions is therefore

1-19M
^ ^E~'

in which M is the maltose formed per 100 grms. of starch and E is
ihe weight (in grms.) of malt or malt extract used. The formula of

* These values are calculated from the constants given by Brown and Heron
(loc. cit.), which are [oj/j-ag 162-6° ^3-36 (percentage of apparent glucose expressed on
100 grms. of the total solids of the solution, estimated by the 3-86 divisor) 49'3.

2 The factor used by Harrison and Gair is

i55=l-184.
84-4

MALTOSE AND MALT EXTRACT. 163

Harrison and Gair is open to the objection that it assumes a constant
maltose titre for FehHng's solution.

Since the amount of starch hydrolysed is measured by the amount
of maltose produced, it is obvious that the results cannot be accurate
unless this amount of maltose does not exceed 40 per cent of the
starch, in accordance with Kjeldahl's law. Harrison and Gair, how-
ever, ignore this law entirely.

More recently Harrison (" Pharm. J.," 1909, 82, 388) has recog-
nized that the method is faulty with highly diastatic malt extracts,
and in reference to these he says : —

" Clearly no higher results could be obtained than the conversion
by the extract of five times its own weight of starch (or a diastatic power
of 500), since no more is present. If it can convert more than this, the
best plan is to reduce the quantity of extract, keeping all the other
quantities constant. But if by proceeding in this way two extracts
were found to give values of, say, 450 (using 0*2 grm. of extract)
and 550 (using O'l grm. of extract), these two figures would not re-
present their respective powers with as near an approach to quanti-
tative truth as if they were, say, 100 and 150, since in the first case
nine-tenths of the starch present would have been used up, and in
the second only eleven-twentieths, and the rate of conversion becomes
less as the excess of starch becomes less. I find it best, therefore, if
an extract gives a diastatic value of over 250, to repeat the test with
only half the quantity of extract ; if a diastatic power of over 750 is
found, to reduce the extract to one-fourth the original quantity — i.e.
to 0*05 grm. This allows for values up to 2000 being recorded, and, if
necessary, the quantity can of course be further reduced : in each case
water is added to make the total liquid in which conversion occurs
measure 100 c.c."

In carrying out the experiments now to be described Ling found
it more convenient, instead of weighing out separate quantities of
starch and malt extract, to take definite volumes of starch paste and
of malt extract solution. In the following experiment 50 c.c. of a
starch paste containing 2 grms. of anhydrous starch per 100 c.c. was
treated with increasing quantities of a 1 per cent solution of malt
extract. The mixture was kept at a constant temperature of 40° C.
for half an hour, at the end of which time 0*5 c.c. of normal caustic
potash was. added to arrest further diastatic action,^ the volume made
up to 100 c.c, and the liquid titrated against 10 c.c. of Fehling's solu-
tion by the method devised by Ling and Rendle (" Analyst," xxx.
182).

The values given under the various columns in the tables are as
follows : —

A is the number of c.c. of 1 per cent malt extract solution taken.

M is the maltose produced, calculated as a percentage on the
original starch.

E is the weight (in grms.) of malt extract taken.

' This method of arresting diastatic action is far preferable to that of boiling
the solution, since the latter is not sufficiently rapid and may cause an error.

164 FOOD AND DKUGS.

A is the diastatic power of the malt on Harrison and Gair's stand-
ard.

D is the Lintner value for half an hour's action at 40° C, calcu-
lated by Ling's formula.

Experiment I.

No.

1-19M.

^ 1-19M

13-88

16-52

1652

168

29-61

35-23

1762

177

42-57

50-66

1689

170

59-92

71-30

1426

68-19

81-14

811

67-63

80-48

402

—

Experiment II.

This was carried out in precisely the same manner as Experiment
I., but a solution of soluble starch prepared by Lintner's method was
used instead of starch paste.

Ko.

•19M.

^_1-19M
E

j3_1000

1
2
3
4
5
6

1
2
3
5
10
20

15-57
31-26
46-33
58-48
61-21
62-99

18-51
37-19
55-13
69-58
72-84
74-94

1851
1859
1837
1391
728
374

185
185
183

Experiment III.

Twenty-five grms. of a low-dried distiller's malt was extracted
with 500 c.c. of water at 21° C. for an hour, as in the Lintner
method. Increasing volumes of the filtrate were then allowed to act
on a solution of soluble starch for half an hour at 40° C. as in Experi-
ments I. and II.

No.

1-19M.

.=-T

D_1000
xy

0-25

14-43

17-17

1373

133

0-50

29-15

34-68

1387

135

0-75

41-94

50-00

1333

131

1-0

49-74

59-18

1184

116

2-0

60-77

72-30

723

3-0 63-31

75-34

502

50 63-91

76-05

304

10-0

66-99

79-72

159

25-0

69-24

82-39

MALTOSE AND MALT EXTRACT. 165

It will be seen from these resulta that Harrison's latest sugges-
tions, whilst they tend to increase the accuracy of the method, do not
eliminate entirely the error inherent in the formula of Harrison and
Gair. Thus, as Ling's results show, with malt extracts having a dia-
static power of 1700 to 1800 on Harrison and Gair's scale, if a weight
of extract of 0"05 grm. were taken (as Harrison suggests for extracts
having a value up to 2000), the results would not be accurate,
because under these circumstances more than 40 per cent of maltose,
calculated on the original starch, is produced. But without a kQOW-
ledge of Kjeldahl's empirical law, it is possible to fix the limit of ac-
curacy of any method of diastasimetry by diminishing the weights of
diastatic substance taken until two experiments with different weights
give uniform results. With a malt extract of diastatic power 2000 on
Harrison and Gair's scale, it will be necessary, in order to attain ac-
curacy, to take a smaller weight than 0*05 grm. In the case of a
highly diastatic malt extract, three experiments should be made with
0*05, 0'03, and 0*01 grm. of the sample respectively ; those coming
within the limit of Kjeldahl's law would give identical values.

From the last column in the tables, it will be seen that the Lint-
ner value, calculated by Ling's formula, under the conditions of Har-
rison and Gair's method — that is to say, action of the diastase on
starch paste or on soluble starch for half an hour at 40° C. — is ap-
proximately one-tenth that of the value on Harrison and Gair's scale.
It must be remembered, however, that iu the Lintner method action
is allowed to take place for an hour — or, in other words, for twice the
period prescribed by Harrison and Gair ; consequently the Lintner
value at 40° C. will be approximately one-fifth the Harrison value.
This is a pure coincidence, since the basis of each of the methods is
quite distinct.

Ling points out that Harrison and Gair's method is based on the
assumption that the so-called stable equation of Brown and his co-
workers is an absolute constant, which it certainly is not, the value of
E {vide ante) varying according to the diastatic power of the sample
of malt or malt extract used.

In conclusion, Ling urges that the scale employed by Harrison
and Gair should for all purposes be replaced by that of Lintner, since
the former records values to four places, thus beyond the limit of ac-
curacy of the Fehling method, which, as he has shown elsewhere
("Analyst," xxxiii. 163), is, even in its most accurate form, only 1
in 300, Values of 1000 and upwards can therefore only tend to
mislead manufacturers.

Squire advocates the following method for the determination of
diastatic value. Two c.c. of a solution of iodine (about 0*2 grm. per
litre) are run into each of twelve test tubes. A 5 per cent solution of
the extract, and a 1 per cent solution of starch, well boiled, are made.
These solutions are warmed to 100° F., and 50 c.c. of the starch are
placed in a beaker kept in water at 100° F. To this 10 c.c. of the
solution of the extract are added. At the end of exactly one minute
draw off 2 c.c. of the solution and add it to the iodine in one of the
tubes, and repeat this each minute. If the tubes are kept in the order

166

POOD AND DRUGS.

in which the additions are made, the colours will depend on the amount
of action that has taken place. He states that if the malt extract be
of the best quality, the first tube will be of a blue colour, the second
red and the third or fourth yellow. This is a somewhat empirical
method, and can give no quantitative valuation.

Estimation of Dextrin. — To 20 c.c. of a 5 per cent solution of the
extract, add 250 c.c. of methylated spirit. If the precipitate be very
small, a fresh experiment should be commenced with a 10 per cent
solution. Determine the nitrogen in the washed and dried precipitate,
and deduct the weight of the calculated proteids (N x 6-25), return-
ing the remainder as dextrin. It is more accurate to redissolve in
water and precipitate a second time with alcohol.

Estimation of Proteids. — These are determined by a direct deter-
mination of nitrogen on 1 grm. of the extract by Kjeldahl's process.
The total nitrogen varies from 0-5 to 2*2 per cent.

Botatory Power. — The specific rotatory power of genuine malt ex-
tracts usually varies between + 80° and 90°.

The following figures are those of a number of commercial malt ex-
tracts. No. 1 being one of the best-known brands and of the best
quality.

Solids.

Sp. gravity.

Sp. Rotation.

Reducing Sugars as Maltose.

lintner Value.

Per cent

1-400

+ 76^

1-390

+ 80°

72-5

1-375

+ 80^

1-380

+ 82°

1-378

+ 79°

1-378

+ 78°

1-400

+ 84°

1-420

+ 88°

1-415

+ 88^

—

Any sample with a rotation of over +90° is suspicious, and
may be suspected to contain glucose syrup — the dextrin present in
this causing it to have an average specific rotation of over + 100°.

THE STARCHES AND STARCHY FOODS.

The second group of the carbohydrates which comes within the
scope of this work, is that embracing the starches and starchy foods.
The microscopic characters of starches are the only means of identi-
fying them with certainty, and their examination is principally
■microscopic rather than chemical.

Starch has the empirical formula (CgHjQOg)^, its structural formula
being unknown. It is a white tasteless powder not soluble in any
solvent without some alteration. It is a cellular substance consisting
of small masses of starch "granulose," encased in an outer layer of
starch cellulose. When heated with water the cellulose layer is rup-
tured and the granulose and similar matters are dissolved in the

THE STAKCHES AND STARCHY FOODS.

167

water. The characteristic blue colour which is given by starch when
treated with iodine is due to the granulose. Starch solutions are
highly dextrorotatory [a]^l = about + 200^ Starch is readily converted
into dextrin and maltose by boiling with dilute acids, dextrose eventu-
ally resulting if the treatment be prolonged. A similar change is
produced (on soluble starch) by malt extract, on account of the
diastase present.

The Microscopic Examination of Starches. — Numerous investiga-
tions have shown that the starch grains produced by any particular
plant are remarkably constant in size, shape and other characteristics.
A small quantity of the starch is mixed on a slide with a few drops of
water. In many cases a micrometer scale on the eyepiece is useful,
as the size of the grains is sometimes an important characteristic to
consider. Glycerine is not a good medium to mount starch in, since
its high refractive power renders the striations in the starch less visible.
When it is necessary to locate starch grains in a section, or in a
mixed powder, the specimen should be irrigated with iodine solution,
when the starch grains are stained from a violet to an almost black
colour. When the starch grains are very small, as in cayenne pepper,
such staining becomes necessary.

The following points are to be noted : —

(1) The shape, whether oval, ovate, ellipsoidal, oyster-shaped,
nearly round, or even angular. To get a proper idea of the shape of
a starch grain, a drop or two of alcohol should be brought to the edge
of the cover glass, by which means a current is set up and the starch
grains move and can be examined in motion. Some grains may be
found adherent to each other, forming compound grains.

(2) The size of starch grains is usually expressed in microns (a
micron, /x, is y^jVo P^^* o^ * millimetre). The size varies from 5
microns to 70 microns in their longest measurement.

(3) The hilum. Near to one extremity, generally the narrower
end, there is a point round which concentric striations are arranged ;
with a high focus it appears as a dark spot, and with a low focus as a
light spot. In some grains there is a V-shaped fissure through the
hilum, as this spot is called. The hilum is sometimes in the centre
of the grain, sometimes eccentric. The ratio of the distance of the
hilum from the nearer margin to that from the further margin is the
measure of its eccentricity.

(4) Striations. Round the hilum there are a number of striations
or strise, concentric lines probably due to varying proportions of water
in the different parts of the grain. With a high focus the strias are
dark, and light at low focus. Some starches exhibit no striations at
all, whilst in others they are well marked.

It will here be convenient to describe the microscopic characters of
the principal starches, many of which are illustrated on pp. 168-172.

Wheat Starch. — This starch is obtained from various species of
Triticum. It consists of large nearly round and oval grains ; with nu-
merous very small ones, but with few of intermediate size. There is no
evident hilum, nor striations in most of the grains, but by careful altera-
tion of the focus, a very few will be found to show a distinct hilum in

168

FOOD AND DEUGS.

the shape of a minute spot, a cleft or a small cavity. Rarely a faint
concentric striation will be observed. On observing the moving grains,
they will be found to be lenticular, and not spherical, and on side view
a longitudinal line may often be observed. The size of the larger grains
when lying flat varies from 20 to 35 /x.

"0% 0 0
O o

C C

■Q

Fig. 10.— Wheat starch x 240. Fig. 11.— Barley starch x 240.

(The illustrations of starches by Greenish & Collin are repro.^uced by permission

of the Editor of the Pharmaceutical Journal).

Barley Starch.—This is obtained from Hordeum distichon. This
starch is very similar to that of wheat, consisting of large and small
grains with but few of intermediate size. The large grains are rather
smaller than those of wheat starch, measuring from 15 to 25 /x, rarely
up to 30 IX. They are also less regular in shape, the rounded' grains
often being of a somewhat kidney shape. On moving they are seen
to be lemon shaped rather than lenticular. Hilum and striations aro
very rarely found.

Rije Starch.—This is obtained from Secale cereale. The grains
closely resemble those of wheat starch, but the larger ones measure
40 to 50 /x, and are not so regular in shape. The grains frequently

^-^?(S5?

Fig. 12.— Rye starch x 240.
(Greenish & Collin.)

Fig. 13. — Maize starch x 240.
(Greenish & Collin.)

show a fissure with 3 to 5 rays extending from an invisible hilum
nearly to the circumference. No striations are visible. ''

Maize Starch. — This starch is obtained from Zea inays.lO The

THE STARCHES AND STARCHY FOODS.

169

grains are circular or polyhedral, usually with more or less rounded
angles, measuring about 7 to 18 /x. They are nearly uniform in size,
generally about 12 to 15 /a, and exhibit a well-marked hilum, which
is sometimes a point, but usually a well-defined 3- or 4- rayed cleft.
No striations are visible. Compound grains do not occur.

Oat Starch, — This is contained in the fruits of Avena sativa. The
grains are nearly uniform in size, about 10 fx, but large grains up to
45 fji are plentiful, but are seen to be merely compound grains
which are easily separated. The grains are mostly angular, sometimes
round or lemon shaped and resemble rice starch in appearance. No
hilum nor striations are to be found.

Bice Starch. — This is ob-

tained from Oryza sativa. The
grains measure from 6 to 10 /x,
and many larger compound grains
are present. Many are polygonal,
usually 5-or 6-sided, rarely trian-
gular, and no lemon -shaped nor
round grains are present. No
hilum is seen (except rarely when
a small light central spot may be
seen) and there are no striae to
be found.

Potato Starch. — This starch
is obtained from the tubers of
Sola7iU7n tuherosum. The size of
the grains is very variable, the
larger ones measuring up to 100 /x or even 120 /x in length, whilst the
small and medium-sized grains measure from 15 to 65 /x. The grains

Fig. 14.— Rice starch x 240.
(Greenish & Collin.)

Fig. 15.— Potato starch x 240. (Greenish & Collin.)

are flattened, and are oval, ellipsoidal or, especially in the case of the
larger grains, oyster-shaped. The hilum is a spot nearly always close

■

170

FOOD AND DRUGS.

Fig. 16. — Arrowroot starch x 240.
(Greenish & Collin.)

to the smaller end of the grain, and the striations are concentric and
exceedingly well marked. A few compound grains are to be found.
Arroivroot Starch. — This is obtained from various species of

Maranta ("West Indies,
Natal, Bermuda, or St.
Vincent). The grains
are fairly large, measur-
ing from 10 to 70 yu. in
length. They somewhat
resemble potato starch,
but are smaller and less
regular in shape; they
are ovoid, and often
shaped something like
a mussel shell or a pear,
and sometimes tend to-
wards a triangular shape.
The small grains are
nearly spherical. The
hilum is well marked,
and is usually near the narrow end of the grain. It is either circular
or linear and often cracked, so that • the cleft appears like the open

wings of a bird. In St. Vin-
cent arrowroot, the linear or
stellate hilum predominates,
whilst in the Natal variety, the
rounded hilum is more usual.
Striations are numerous and
well defined, but not very
strongly marked. Tons les
mots arrowroot is obtained
from Canna species — the
grains are similar to normal
arrowroot starch, but are lar-
ger (50 to 150 /a) with a
rounded hilum and well
marked striations. Curcuma
or East Indian arrowroot is
also similar; the grains are
30 to 60 /A in length. The
grains taper to a small obtuse
projection in which the hilum
is situated. The concentric striations are well defined, but not very
strongly marked.

Sago Starch. — This is prepared from the sago palm, Metroxylon
Sagu. The grains vary in length from 25 to 65 fx. Pearl sago is pre-
pared from sago starch with the aid of heat, so that in commercial
sago that has thus been agglomerated, the granules are mostly broken.
The intact starch grains are ovoid, often rounded at the larger end and
truncated at the narrower end. Many are very irregular in shape,

Fig. 17. — Tous les mois starch x 240.
(Greenish & Collin.)

THE STAKCHES AND STARCHY FOODS.

171

some are simple, but many are compound, having one or more small
granules attached to short pro-
tuberances on a large grain.
The hilum is eccentric, and is
a circular spot or crack at the
broader end of the grain.
Striations are concentric,
often plain, but frequently
only faintly indicated.

Tapioca Starch. — Tapioca
is prepared in a similar man-
ner to prepared sago, but from
the starch obtained from the
tubers of manihot utilissima.
The grains of the manihot
starch are from 15 to 35 /x in
length, and are therefore
small. Many of them exhibit
a flat surface in places, having
probably been components of
a compound grain. They are
circular or kettle-drum in shape, many having the flat surface forming a
sharp angle when they meet. The hilum is a point or short cleft, and is
nearly central, but this depends on the point of observation. If the grain
be lying flat the hilum appears eccentric — if standing on a flat surface,
it will appear central. In commercial tapioca,
most of the grains exhibit the effects of heat.

Bean Starch. — This is prepared from Phase-
olus vulgaris, the haricot bean. The grains vary
in size from 25 to 60 yu., most of them measuring
about 35 yu, in diameter. They are oval, or reni-
form in shape, some being nearly round. The
hilum, owing to large fissures, appears either
stellate, or as a long and often branching cleft
running nearly the whole length of the grain,
and appearing very conspicuously under the
microscope. The striae are well marked.

Banana Starch. — This starch is prepared from the unripe fruits

Fig. 18.— Sago starch x
(Greenish & Gollin.

240.

Fig. 19.— Bean starch.

Fig. 20. — Banana starch.

172 FOOD AND DKUGS.

of Musa sapientum ; the grains are oval, ellipsoidal, or elongated.
The hilum is rounded near to the extremity and surrounded by con-
centric striations. Size from 7 to 65 /x.

Pea Starch. — This occurs in the seeds of Pisimi sativiLm. The
grains are very similar to those of bean
starch, but are rather smaller, measuring
about 20 to 40 /x. So similar is this to
bean starch that in a mixture of the two,
it could scarcely be detected.

Lentil Starch. — This also is so similar
to bean and pea flour that it is not easy to
distinguish between them. The characters
of pea, bean, and lentil starch, although
very similar, are distinct enough to enable
Fig. 21.— Pea starch. ^Yie microscopist to diagnose them as the

starch of a leguminous plant.
With polarized light, starch grains show dark crosses, the point of
intersection of the arms being at the hilum. Some starches show
colours with crossed prisms and a selenite plate, but in the author's
experience, the value of the polarization phenomena with starch has
been much exaggerated, and no information is yielded that is not
easily obtainable by the use of ordinary light.

Passing on from the separated starches to the flours of the prin-
cipal starchy foods, the following are the principal diagnostic characters.
Wheat Flour. — Microscopic characters. The characteristic fea-
tures are : —

1. The large starch grains.

2. The hairs with enlarged lumen at base.

3. The thick-walled pitted cells of the hypoderma.

4. The tabular cells of the outer epidermis with pitted transverse
walls.

Bye Flour. — The characteristic features are : —

1. The starch grains which are rather larger than those of wheat
and often show a stellate hilum.

2. The hairs with thinner walls than those of wheat, and less en-
larged lumen.

3. The lignified cells of the hypoderma which are usually longer
than the transverse cells, whereas in wheat they are shorter.

Barley Flour. — The characteristic features are : —

1. The typical starch grains.

2. Epidermal cells of the paleae with thickened sinuous walls.

3. Hairs on the inner epidermis of the paleae.

4. The thin-walled, not pitted, epidermal cells of the pericarp.

5. The aleurone grains of two or three rows of cells.
Oat Flour is distinguished by : —

1. The elongated hairs which are often found in pairs.

2. The cells of the outer epidermis of the pericarp which have
thin walls and numerous pits.

3. The polygonal cell of the hypoderma.

4. The cells of the seed coat which are polygonal, smooth, and
seldom pitted.

WHEAT FLOUR AND BREAD.

173

0. The starch grains, mostly compound, and composed of small
angular grains.

Rice Flour contains only a small proportion of the seed coats
consisting chiefly of the characteristic starch grains, compound and
small simple, rounded or polyhedral grains.

Maize Flour. — The characteristic features are : —

1. The typical starch grains.

2. The numerous small tabular grains of the pericarp.

3. The hypodermal cells with slightly pitted walls.
Buckivheat Flour. — The characteristic features are : —

1. The typical starch grains which resemble rice starch but are
rather more rounded, with a small central hilum, and usually agglo-
merated.

2. The epidermal cells of the seed coat which have sinuous walls»

3. The middle layer of the seed coat which consists of cells with
lacunae.

Haricot Bean Flour. — The characteristic features are : —
i. The typical starch grains which are ovoid with an elongated or
fissured hilum.

2. The palisade cells of the epidermis.

3. The rectangular cells containing prismatic crystal of calcium
oxalate.

4. The cells of the cotyledons which are polygonal and thickened
at the angles.

Pea Flour. — The characteristic features are : —

1. The palisade cell with square ends.

2. The starch grains which bear rounded protuberances, and have
a central hilum surrounded by concentric rings.

Lentil Flour. — The characteristic features are : —

1. Palisade cells with conical ends.

2. Hour-glass cells without calcium oxalate.

3. Starch grains intermediate in character between pea starch and
bean starch, the hilum being generally fissured with distinct striae.

Wheat Flour and Bread. — Wheat flour is understood to be the
ground fruit of Triticum sativum and allied species (freed from the
bran or episperm).

The average composition of wheat flour, according to various
chemists, is as follows : —

Graham.

Chiirch.

Konig.

Bell.

Per cent

Water

111

13 tol4'5

13-56

12-08 to 14-08

Starch

62-3

69 „ 74

64-07

63-71 „ 65-86

Fat

1-2

0-8 „ 1-2

1-70

1-48 „ 1-56

Cellulose

8-3

0-7 „ 2-6

2-62

2-93 „ 303

Sugar and gum

3-8

—

3-82

2-24 „ 2-57

Albumenoids

10-9

10-5 „ 11-0

12-42

11-59 „ 15-53

Mineral matter

1-6

0-7 „ 1-7

1-79

1-60 „ 1-74

174

FOOD AND DRUGS.

If the whole of the grain be ground, in order to prepare whole-
meal flour, the percentage of mineral matter is considerably raised,
bran containing as much as 7 per cent of ash.

The fat of wheat consists of olein and palmitin with a small amount
of similar glycerides, and about 6 per cent of free fatty acids.

Small quantities of dextrin are present, but it is never necessary
in practice to determine this.

The proteids of wheat are several in number and are generally
mentioned here as " albumenoids ". The amount of nitrogen present
in the greater part of the proteids and allied bodies of the cereals
contains an amount of nitrogen not differing much from 15'8. This
value is taken as a fair average one, so that the total nitrogen multi-
plied by 6'33 is usually returned as total albumenoids (5*7 gives more
accurate results — in any case the multiplier should be stated). A
more complete examination of the nitrogenous bodies of wheat flour
is, however, often necessary. According to Osborne and Voorhees
(** American Chem. Journal," xv. 392 ; xvi. 524), wheat contains
five different proteids which have the following composition : —

Properties.

Composition.

Per

S.
Per

Per

cent

Globulin

0-6 to 0-7

51-03

6-85

18-39

23-04

0-69

Albumin

0-3 „ 0-4

53-02

6-84

16-80

22-06

1-28

Proteose

0-3

51-86

6-82

17-32

24-

Gliadin

4-25

52-72

6-86

17-66

21-62

1-14

Glutenin

4 to 4-5

52-34

6-83

17-49

22-26

1-08

The two last named are the principal constituents of the gluten,
the first three being soluble in water and only present in small
amount. They are separated in the following manner: —

Four thousand grms. of flour are treated with 8000 c.c. of 10 per
cent brine, allowed to stand all night, and the supernatant liquid
filtered off. Another 2000 c.c. of brine are added to the residue, the
whole stirred up, allowed to settle, and again filtered. The filtrate is
at once saturated with ammonium sulphate, and the precipitated
globulin filtered off, redissolved in 10 per cent brine, and dialysed until
all the salts are removed when practically pure globulin is precipitated.
The albumin is coagulated by heating the globulin-free solution and
proteose is precipitated by saturating the filtrate with sodium chloride.

Glutenin is prepared by boiling crude gluten (made by kneading
flour in a stream of water, so that the starch and soluble proteids are
washed away) with alcohol of specific gravity 0-890. The gliadin
dissolves, leaving a residue of glutenin with a small quantity of im-
purities. It can be purified by dissolving in very dilute potash and

WHEAT FLOUR AND BREAD. 175

neutralizing with acetic acid, when it is precipitated as a whitish-
grey mass. It is practically insoluble in cold water and cold alcohol.

Gliadin is readily dissolved from the flour or the gluten by hot
dilute alcohol. It is easily soluble in dilute alcohol and in water
free from salts, and forms, in the hydrated condition, a sticky mass,
which, however, when dehydrated by treatment with absolute alcohol,
and then treated with ether, forms a white friable mass easily
powdered.

Wheat-gluten, then, consists essentially of a mixture of gliadin and
glutenin. Gluten cannot properly be made with distilled water, since
gliadin is easily soluble unless salts be present. It has been suggested
that the gluten is the result of a ferment action on the flour in the
presence of water. Osborne and Voorhees, however, state that " no
ferment action occurs in the formation of gluten ". No cereal but
wheat yields a gluten, which has considerable importance as a food
stuff. Bread made from gluten has long been recommended as a
suitable food for diabetic patients. Many preparations, however, con-
tain so much starch left in the gluten as to be totally unfit for this
purpose, and in addition such bread is very unpalatable. Gluten
bread should be examined for starch, and if present in more than
traces, this should be determined. It has now, however, been nearly
superseded by a bread made from the casein extracted from milk,
which is obtainable practically free from carbohydrates.

In the examination of flour (from the point of view of the baker,
technical and empirical valuations of the flour are often required which
are not within the scope of this work — see " Bread-making " — by W.
Jago), the following determinations are more or less necessary : —

(1) Water, (2) mineral matter, (3) fat, (4) cold water extract,
(5) total fnitrogen, etc., (6) starch, (7) percentage of gluten obtain-
able, (8) amount of free acidity, (9) fibre.

A microscopic examination, bearing in mind the diagnostic fea-
tures above given, especially when a comparison is made with a
standard sample, is, of course, essential.

Determination of Water and Mineral Matter. — Water should be
determined on about 2 grms. of the flour, at 105°, and the dried flour
may then be ignited for the ash determination. The water averages
about 12 to 14 per cent and the mineral matter from 0*5 to 0*7 per
cent. The ash consists principally (80 per cent) of potassium
phosphate.

Determination of Fat. — The flour must be dried before the fat is
determined, due allowance being made for the loss in weight. From
10 to 15 grms. should be exhausted with dry ether in a Soxhlet tube
in the usual manner. The fat will be found to be from 1 per cent to
1-6 per cent.

Cold Water Extract. — The estimation and examination of the cold
water extract afford much useful information. Ten grms. of the flour
are rubbed to a cream with water and the volume made up to 500 c.c.
and the whole well shaken at intervals from two to three hours.
The liquid is filtered, and aliquot parts used for the following deter-
mination. A portion — about 200 c.c. — is evaporated and the residue

176 FOOD AND DRUGS.

weighed. This, the cold water extract, should vary from 3 per
cent to 7 per cent, usually from 5 per cent to 6 per cent. To some
extent the cold water extract may be regarded as a measure of the
amount of degradation of the starch of the wheat, an excess of soluble
matter indicating unsoundness in the fiour. If desired the soluble
proteids may be estimated by evaporating down 200 c.c. of the extract
to a syrupy consistence and determining the nitrogen present by
Kjeldahl's process. The soluble proteids (using the nitrogen factor
6-33) will vary from 0*7 per cent to 1*5 per cent, occasionally being
a little lower or a little higher than these limits. If the sugar is
required, it is estimated with sufficient accuracy by inverting a portion
of the aqueous extract and determining its reducing powder. The
amount calculated as sucrose is from 2-5 per cent to 3 per cent. It
is probable that in perfectly fresh sound wheat sucrose is the only
sugar present, but that if any diastatic action has set in, a little maltose
may be present. This is best decided by extracting the flour with 70
per cent alcohol, and determining the sugar before and after inversion
in the alcoholic extract. The presence of much maltose would indi-
cate that the flour was not a fresh, sound one.

Determination of Total Proteids. — This is effected by a nitrogen
determination on 1 grm. of the flour, rubbed into a cream with water
so as to prevent the formation of lumps. As pointed out above, the
nitrogen factor 6*25 is usually used in calculating the proteids, but
6'7 is more correct. The multiplication used should, therefore, be
stated.

Determination of Starch. — O'Sullivan's process ("Journal Chem.
Soc." XLV. 1) is the most accurate, and is easily carried out. Five
grms. of the flour are placed in a flask holding about 120 c.c, and
which has been well wetted with alcohol ; 25 c.c. of ether are then
added. The closed flask is well shaken from time to time, and after a
few hours the ether is filtered off' and the residue washed several times
with more ether. To the residue 80 c.c. of alcohol of specific gravity
0*9 are added and the mixture kept at 35° C. and shaken from time ta
time, for several hours. The alcoholic solution is then filtered through
the same filter that was used to filter off' the ether, and the residue
well washed with more alcohol at the same temperature and of the
same strength. The residue in the flask and on the filter is then
treated with about 500 c.c. of cold water. After twenty-four hours
the clear supernatant liquid is carefully decanted through a filter and
the residue repeatedly washed with water at 88" (J. The residue is
then transferred to a 100 c.c. beaker, the solid matter and the water
.used to aid its transference to the beaker, measuring about 40 to 50
c.c. This is heated to 100° for a few minutes with constant stirring
to effect the gelatinization of the starch. When the contents of the
beaker are cooled down to 60°, about 0'03 grm. of diastase, dissolved
in a few c.c. of water, is added. After a short time, conversion into
maltose and dextrin is complete. The reaction is complete when a
drop of the liquid does not give a blue colour with starch. But by
keeping the mixture at about 62° for another hour, it filters far more
readily. It is now heated to 100° for ten minutes and filtered, tha

WHEAT FLOUR AND BREAD. 177

residue on the filter being well washed with hot water. On cooling
the filtrate is made up to 100 c.c. and the specific gravity is taken.
The maltose is estimated by reduction of Fehling's solution ; the
dextrin can be deduced by a polarimetric reading. The amount of
maltose multiplied by 0*95 gives the corresponding amount of starch,
which, added to the dextrin found, should amount to the solids deduced
from the specific gravity of the solution (i.e. by dividing the excess of
specific gravity over 1000 by 3-86). An example given by O'Sullivan
will make the calculations clear.

In the analysis 4-94 grms. of the ground wheat were taken. The
100 c.c. solution had a rotation equivalent =8*52° in a 200 mm. tube.
By Fehling's method it was found to contain 2-196 grms. of maltose.

But 2-196° X 2-78'' = 6-10°, the rotation due to the maltose {[a]d = + 139°),

0.40
therefore 2-42° is due to the dextrin ([a]^= +200-4°), so that— -— =

0-605 grms. of dextrin per 100 c.c. Hence maltose, 2-196x0-95 =
2-086. This, added to 0-605 = 2-691, the total amount of starch present
in the 4-94 grms. of wheat used.

If the acid conversion method be used, the gelatinized starch should
be heated to 95° to 100° with 200 c.c. of water and 20 c.c. of strong
hydrochloric acid for 2^ hours under a reflux condenser. When cool
the liquid is clarified, if necessary, with alumina cream, neutralized with
alkali solution and made up to 500 c.c. after filtration. The dextrose
is determined by reduction of Fehling's solution in the usual manner.
Dextrose x 0-9 may be taken as the equivalent of starch.

If diastase be not at hand, a fairly accurate result is obtained by
replacing it by sulphuric acid and inverting in the usual manner.
Wheat flour will usually contain over 60 per cent of starch.

Perce7itage of Gluten obtainable. — This determination is somewhat
empirical, but is necessary in certain cases : 10 or 20 grams are rubbed
in a mortar with 15 to 20 c.c. of water. The dough is then tied up
in a fine muslin cloth, rather tightly, and gently kneaded with the.
fingers under a gentle stream of water until no more starch is washed
out, as shown by the clearness of the water running away. The gluten
is then taken out of the cloth and flattened in a dish and slowly dried
at 110° to 115° C. It generally amounts to about 35 per cent.

Acidity of the Flour. — By rubbing the flour into a cream and then
adding water, and titrating the whole, a higher acid value will be
obtained than by titrating an aqueous extract. If the flour be so
titrated it should not show a higher acid value than about 0*1 per
cent calculated as sulphuric acid, H^SOg. By extracting the flour
with water and titrating directly the liquid is filtered 0-025 per cent
is about the highest value a good flour will give.

Crude Fibre. — The amount of fibre may be determined as de-
scribed on p. 21.

Adulterants of Flour. — Adulteration of flour with the flour or
meal of other cereals can, practically, only be determined by a care-
ful miscroscopic examination, and any suspected additions should be
compared with a standard sample of the suspected adulterant. Maize
flour, rye flour, rice meal, potato starch and leguminous starches have
VOL. I. 12

178 FOOD AND DKUGS.

been found, but the adulteration of wheat flour is not very pre-
valent.

Seeds, sometimes of an objectionable nature, occasionally get mixed
with the wheat, and some of these can be detected by the microscope.
The bearded darnel (Lolium temulentum) is one of these. The starch
grains of this seed are somewhat similar to those of rice, but the
principal cellular tissue of the husk in the darnel consists of cells
only twice or thrice as long as they are broad, whereas in the rice
they are long and narrow fibres. In the presence of much darnel an
extract of flour with 90 per cent alcohol will be of a sHght greenish
tint, gradually darkening, and on evaporation of the alcohol an acrid
greenish resin is left.

The seeds of the corn cockle {Agrostemma githago) are sometimes
also mixed with the wheat. The surface of the seed coat is quite
characteristic and shows very large thick-walled cells of up to half a
millemetre in diameter, with branching protuberances on the surface.
The starch grains, situated in the parenchymatous cells, are very minute,
and characteristic egg-shaped grains consisting of starch, saponin, and
other bodies are to be found.

Ergot in flour (which, although a natural impurity) is of sufficient
importance to be looked for carefully in every case of obviously low
grade, stale flour.

Vogel (" Chem. Central." 3, x. 559) recommends staining the flour
with aniline violet. Starch granules which have been attacked by
ergot or any other fungus, are stained an intense violet, the other
grains being left nearly colourless. Petri (" Zeit. f. Anal. Chem."
1879, 211) recommends exhausting 20 grms. of the flour with hot
alcohol. If ergot be present the extract will be more or less reddish
in colour. If the alcoholic extract be diluted with twice its volume
of water and shaken with ether, amyl alcohol, benzene or chloroform,
each of these solvents will be coloured reddish to full red. Or if 20
drops of dilute sulphuric acid be added to the alcoholic solution, and the
liquid examined spectroscopically, two well-defined absorption bands
will be seen if the flour be ergotized, one in the green near E, and a
wider band in the blue between F and G. According to Hoffmann and
Hilger (" Archiv. der Pharm." 1885, 827) O'Ol per cent of ergot in
flour can be detected by treating 10 grms. of flour with 10 drops of a
20 per cent solution of potash, and allowing it to stand for ten minutes.
Thirty c.c. of ether and 20 drops of dilute sulphuric acid are then added,
and the whole allowed to stand for six hours with occasional shaking,
and then filtered. The residue is washed with ether so that 20 c.c. of
filtrate are collected, 10 to 15 drops of a saturated solution of bicarbonate
of sodium solution are then added and the whole well shaken. In
the presence of ergot a well-marked violet colour is developed. This
reaction is very reliable.

Vogel (" Verfalschungen des Mehls') gives the following chemical
test foe several adulterants : Two grms. of the flour are warmed and
;shaken with 10 c.c. of 70 per cent alcohol to which 5 per cent of
hydrochloric acid has been added. The liquid is filtered. Wheat and
j-ye flour give a practically colourless extract ; if 5 per cent of^corn-

WHEAT FLOUR AND BREAD. 179

cockle be present, the extract will be of an orange colour ; barley or
oats give a yellow colour ; leguminous flours give a reddish colour if
from 5 per cent to 10 per cent be present, or a violet colour if more
be present ; in the presence of 5 per cent of ergotized wheat, the ex-
tract will be of a blood-red colour. Numerous " methods " for em-
pirically detecting more or less apocryphal adulterants have been
published from time to time, but as most of them are quite useless,
they are not referred to here.

Embrey points out ("Analyst," xxv. 315) that foreign so-called
wheaten flour, and certain flours sold under fancy names such as
" self-raising flour," etc., often contain from 10 per cent to 20 per
cent of maize flour.

He considers that an examination of the ash is of some value,
stating that the ash of maize starch contains 38*65 per cent of
phosphates calculated as tricalcium phosphate, whilst that of wheat
flour contains only 22-42 per cent. These figures, however, are quite
at variance with those of other observers, and are not, in the author's
opinion, of the slightest value.

Bauman states that as little as 2 per cent of maize in wheat flour
can be detected in the following manner: About 0*1 grm. of the flour
is mixed with lOc.c. ofal'8 per cent solution of potassium hydroxide
and the test tube shaken at intervals during two minutes. Four or
five drops of 25 per cent solution HCl are then added and the tube is
again shaken. The liquid must remain alkaline. If a drop of the
liquid be now examined miscroscopically the wheat starch will be com-
pletely ruptured, whilst the maize starch will be unaltered. It is sug-
gested that an approximate determination can be made by the
comparison of the sample with standard mixtures under identical
conditions.

A. C. Wilson states that maize starch may be detected by examining
the flour in a clove oil preparation. The hilum of maize starch appears
as a black dot or star, whilst that of wheat flour is invisible. This is
confirmed by E. J. Bevan {" Analyst," xxv. 316).

Mineral Adulterations. — The principal mineral adulterant to guard
against is alum, which is added in very small quantity in order to
whiten the flour. Any other mineral matter, such as plaster of paris,
is very rarely found. If such be suspected it will usually show in the
ash, as any figure over 0-75 per cent should be suspected. Mineral
matter may be detected by the chloroform test — it being remembered
that as a rule there is only a small amount added, if any, not to increase
the weight, but to conceal the .bad colour of an inferior flour. One
hundred grms. of the flour are shaken in a separator with 250 c.c. of
chloroform and allowed to settle ; the flour will be on the surface, and
the mineral matter will have settled at the bottom of the separator.
This is run off and diluted with a little more chloroform, and again
run off and the chloroform evaporated in a current of air. The de-
posit may be examined microscopically to detect crystals of alum, and
then extracted with water, in which the alum if present may be de-
termined as AlgOg : if the insoluble residue does not weigh more than
0*1 per cent of the weight of the flour it need not be further examined.

180 FOOD AND DRUGS.

Alum may be detected in flour by the following methods. Ten
grms. of the flour are rubbed down with 10 c.c. of water and 1 c.c. of
a tincture of logwood (5 per cent logwood in alcohol) and 1 c.c. of a
saturated solution of ammonium carbonate are then added and the
whole well mixed. If the flour be pure, a pink colour gradually chang-
ing to a dirty brown results : if alum be present, the colour changes
to a lavender or blue colour ; or strips of gelatine may be soaked for
twelve hours in the mixture of water and flour and then immersed in
the alkahne solution of logwood, when a decided blue colour is obtained
on the gelatine.

If bread has to be dealt with instead of flour, the following points
are to be noted. The following represent the average composition of
moderately fresh bread prepared from wheaten flour: —

Per cent

Water 28 to 45

Albumenoids 5 „ 7

Fat 0-4 „ 1

Sugar 0-8 „ 4-5

Starch (and dextrin, etc.) 38 ,, 58

Cellulose 0-2 „ 0-8

Mineral matter 0-75 „ 1-4

From the analyst's point of view, the examination of bread in
practice is usually restricted to the detection and determination of
alum. Cases of adulteration with other starchy matter are rare, and
as the starch grains are much altered by the action of heat, a micro-
scopic examination will be of little service, unless comparisons are
made with bakings from the flours suspected to have been used. A
little salt, or the reaction products of baking powders, are to be found
frequently, but no exception is to be taken to these. It is said that
sulphate of barium and plaster of paris are sometimes added, but if
so, this is very rare, and such additions will be found in the ash of the
bread, which will, of course, be correspondingly high.

The principal adulterant of bread, however, is alum, which is added
to cover the use of inferior flours.

Alum may be detected in bread by diluting 5 c.c. of the tincture of
logwood mentioned under flour, with 90 c.c. of water and 5 c.c. of
saturated solution of ammonium carbonate. The liquid is then poured
on about 10 grms. of the bread on a glass dish. After about five
minutes the liquid is drained away and the bread washed gently with
a little water and dried. If alum be present, the bread will assume a
lavender or dark blue colour when dry. According to Allen, 7 grains
of alum in a 4 lb. loaf can thus be detected. Young, however, considers
that this test is not absolutely reliable, as some breads free from alum
give the reaction.

For the quantitative determination of alum Dupre's process, slightly
modified by Young (" Analyst," xv. 61, 83) gives the most accurate re-
sults. One hundred grms. of bread are incinerated in a muffle, until the
ash does not weigh more than 2 grms. This is then moistened with
3 c.c. of hydrochloric acid and 25 c.c. of distilled water. The whole is
boiled, filtered, and the undissolved matter is washed, dried, ignited,

WHEAT FLOUR AND BREAD. 181

and weighed. This consists of siHca. Ammonia solution (dilute) is
then added until the precipitate formed barely dissolves. The liquid is
then raised to boiling-point and a faintly acid solution of ammonium
acetate added, and the boiling continued for a few minutes. The
precipitate of iron and aluminium .phosphates should be filtered off at
once. (If this is done in the cold or after standing for long the results
are much below the truth.) The precipitate is then washed, and re-
dissolved in a very small quantity of dilute hydrochloric acid. The
resulting solution is poured into an excess of a solution of jjiire caustic
soda, and after heating for a short time, the liquid is diluted and
filtered. The filtrate is acidified with slight excess of HCl and
ammonium acetate and a few drops of solution of sodium phosphate
are added, and finally a slight excess of ammonia. The liquid is
heated till all the ammonia has been driven off", and the precipitated
aluminium phosphate is filtered off, washed, dried, ignited, and weighed.
The weight of the precipitate multiplied by 3-87 gives the equivalent
of crystallized alum, but this requires a correction, as bread naturally
contains traces of aluminous silicates which have been derived from
the mill-stones, or from the soil, and the alumina thus naturally
present must be deducted.

From a series of forty analyses by Carter Bell, it is clear that the
silica and alumina are combined in such proportions that the correc-
tion should be to allow for any part of silica found an equal amount
of crystallized alum, which is deducted from the result above found.

Bleached Flours. — Nitrous fumes are sometimes used for bleaching
flours, in order to apparently improve their quality. It has been
shown that this seriously affects the ease with which certain constitu-
ents of the flour are digested, and also causes the bread baked from such
flour to have a greater tendency to go mouldy. Such bleached flours
may be recognized by determining the presence of nitrites, and also by
determining the iodine value of the fat extracted from the flour.
Normal flours yield a fat with an iodine absorption of 100 or over,
whereas bleached flours give a fat with an iodine value of from 80 to
90, owing to oxidation of the fat.

Nitrites may be detected by the Griess-Ilosvay's reaction. If a
bleached flour be treated with a solution of naphthylamine acetate and
sulphanilic acid, an amino-azo dye of a red colour is at once produced.
An unbleached flour will show no coloration for at least half an hour.
Occasionally an unbleached flour appears to answer this reaction how-
ever. Weil (" Chem. Zeit." 1909, 33, 29) states that it is well known
that bleached flours revert in colom- fairly rapidly if stored. This may
be used to detect the bleaching, if the reversion be accelerated. This
is best done by passing a current of sulphuretted hydrogen through the
flour in a closed vessel for an hour. An unbleached flour shows hardly
any change whereas the colour of a bleached flour is much darker than
it was before such treatment.

Maize Flour. — The flour of maize or Indian corn, used largely
under the name of corn flour.

The microscopic examination of this flour will yield most of the in-
formation that is necessary. Rice or potato flours are the most pro-

182 FOOD AND DEUGS.

bable adulterants and those should be specially looked for under the
microscope.

The average composition of maize flours is : —

Per cent

Fat 3-58

Starch 64-66

Cellulose . 1-86

Sugar 1-94

Albumenoids .......... 9'67

Other nitrogenous matter 4-60

Mineral matter 1"35

Moisture 12-34

The oil extracted from the flour may be examined. It should have
the following character : —

Specific gravity 0-921 to 0-925

Iodine value 116 „ 123

Kefractive index at 15° t 1-4768 „ 1-4775

Saponification value 188 „ 189

Oatmeal. — This meal has the average composition : —

Per cent

Water 10-07

Albumenoids 14-66

Fat 5-91

Sugar 2-26

Dextrin or gum 3-08

Starch 5939

Mineral matter 2-24

Cellulose 2-39

Its characteristics are its high nitrogen value, and its high amount
of cellulose tissues. Its fat content is also higher than most cereal
grains and should be determined.

The chief adulterant of oatmeal is barley meal. This is detected by
a microscopical examination. To determine the approximate amount
of such adulteration, standard mixtures must be made of the two
meals and a number of microscopic preparations made. The relative
number of starch grains of the two meals is counted over a number of
fields and the average taken. This is compared with standard prepara-
tions say of 20 per cent adulteration, 30 per cent, 40 per cent, etc.,
and so an approximate determination is arrived at.

Bice. — The only practical question arising for the analyst in regard
to rice, is that of " facing ". Some rice millers use various substances
to improve the appearance of rice grains, especially on the Continent.
The Local Government Board published a report by J. A. Hanwill on
the whole question (" Reports of Inspectors of Foods,"1909, No. 8, 1-
21). It is pointed out that the operation of milling rice includes the
polishing of the grains, which is carried out in revolving cylinders lined
with sheepskin. Some millers, however, add talc to improve the
polish of the grain. Traces of ultramarine or aniline blues are also
added to improve the colour, and a trace of oil to add translucency.

A glazing mixture — talc, glucose, glycerine and starch — is also some-
times used after the polishing process. From the results of analysis

MACARONI AND PASTES. 183

published in the above report, it may be seen that the amount of
such added substance is infinitesimal so far as the polishing is con-
cerned, but that up to 0'2 per cent of mineral matter may be added
by the glazing process. Fourteen samples of rice milled in Holland
contained from 0*16 to 1'25 per cent of added mineral matter (a nor-
mal rice contains from 0*2 to 0'3 per cent of mineral matter). The
conclusion arrived at by the reporter is that on the whole it seems re-
grettable that the practice of polishing rice with mineral matter has
been allowed to reach its present proportions, but as the matter has be-
come a trade custom, it is suggested that an outside of limit of 0"5 per
cent of mineral matter might be fixed for rice.

Borgherio (" Giorn. Farm. Chim." 1909, 58, 533) states that he has
found samples of rice dyed a faint yellow colour with an oil-soluble
aniline yellow. This is done to improve the appearance of poor
samples, and may be detected by heating the ethereal extract of the rice
with hydrochloric acid. The acid layer becomes a faint carmine pink
if this dye is present, changing to yellow on the addition of ammonia.

Richardson (" Analyst," xxxv. 293) determines the extraneous
mineral matter in rice in the following manner. Five grms. of the
rice grains are treated in a platinum dish with 0*5 grm. of ammonium
fluoride, 2 c.c. of water and 2 c.c. of strong hydrochloric acid, and
stirred occasionally during ten minutes with a platinum wire. The
rice is then well washed with water, which is decanted off, and the
so cleaned rice is incinerated.

The difference between the ash so found and the total ash (both
recarbonated) gives the amount of facing mineral matter. Richardson
finds the average natural ash of rice to be 0*2 per cent so that any
excess over this may be regarded as added mineral matter.

Semolina and Macaroni. — Semolina is the coarse meal ground
from certain varieties of hard or "durum" wheats now grown in
France and .certain parts of the United States and Canada, though
originally produced in Italy, Sicily, and Russia. The hard wheats
contain a considerable amount of gluten, and are therefore particularly
suitable for the preparation of macaroni and the various pastes. In
preparing the wheat, the husk is removed by wetting, heating, grinding,
and sifting. « The meal thus obtained — viz. semolina — is in small,
round glazed granules.

Italian Pastes. — Semolina is the chief constituent of the Italian
edible pastes. It is mixed wi' J warm water, kneaded and moulded
into different shapes by pressure through holes in an iron plate or by
other similar methods, then dried. It has been said that the juices
of carrots, onions, and other vegetables are mixed with the paste in some
parts of Italy, but this is only used locally. Saffron is sometimes
added to pastes for flavouring purposes, though sometimes in such
small quantities that it would appear as if it were solely used to give
a colour resembling egg-paste.

Macaroni is the larger slender tube or pipe-shaped variety ; vermi-
celli is the worm-shaped product obtained when the holes in the plate
are very small ; spaghetti is a cord-like variety of a size between the
two already mentioned.

184

FOOD AND DEUGS.

Other kinds of Italiaa pastes or pates are prepared by rolling the
kneaded semolina into thin sheets then cutting it into various shapes —
as animals, letters of the alphabet, etc.

The following table shows the composition of some of these
products : —

No. of
Samples.

Water.

Protein.

Fat.

Total
Carbo-hydrates.

Crude
Fibre.

Ash.

Per

6ent

cent

Semolina

10-50

11-96

0-60

75-79

0-50

0-65

Macaroni

10-3

13-4

0-9

74-1

—

1-3

Noodles

10-7

11-7

1-0

75-6

0-4

Spaghetti

10-6

12-1

0-4

76-3

0-4

0-6

Vermicelli

11-0

10-9

2-0

72-0

—

4-1

Adulteration of Pastes. — The cheaper forms of semolina have been
adulterated with rice, corn and potato flours but not often in this
country.

Analysis of Pastes. — Determination of lecithin-phosphoric acid —
Juckenack's method. This determination will indicate the presence
or absence of eggs in the pastes. Egg-free pastes will give a result of
about 0 02 per cent, and each egg per lb. of paste will give an average
increase of 0 028 per cent. Extract 30 grms. of the finely ground
sample with absolute alcohol for ten hours in a Soxhlet tube at a tem-
perature inside the Soxhlet not below 55° to 60° C. There should be
a small quantity of pumice stone in the extraction flask to prevent
bumping during boiling, and the Soxhlet should be enveloped in
asbestos if it is difficult to keep up the required temperature. After
the extraction add 5 c.c. of alcoholic solution of potash (made by dis-
solving 40 grms. of phosphorus-free caustic potash in 1000 c.c. al-
cohol), then distil off all the alcohol. Transfer the residue to a platinum
dish by means of hot water, evaporate to dryness on a water bath, and
char over asbestos. Add dilute nitric acid to the charred mass, filter
and wash with water. Again transfer the residue with the paper to the
platinum dish, and burn to a white ash ; then treat with nitric acid, filter
and wash as before, uniting the filtrates. Determine phosphoric acid by
the usual method.

Detection of Artificial Colours in Paste. — Turmeric, saffron, annatto,
naphthol yellow (Martins yellow), naphthol yellow S, picric acid,
aurantia, victoria yellow, tartrazine, metanil yellow, azo yellow, gold
yellow and quinoline yellow are the colours that have been used in
several of these pastes. Naphthol yellow, picric acid, metanil yellow
and victoria yellow are injurious to health, and it is, therefore, im-
proper to use them in European countries and the United States.
They are not often found in the products now on the market.

The natural colouring matter of the flour and the lutein of eggs
make it difficult to detect artificial colours. Ether will extract the

ITALIAN PASTES. 185

former, though it does not remove the artificial colours, which, how-
ever, mostly dissolve freely in the solvent if unmixed.

Juckenack's Method.— Tsbke two portions of the sample each of
about 10 grms. in test tubes with 15 c.c. of ether, and 15 c.c. of 70
per cent alcohol respectively ; allow to stand for twelve hours.

(a) If the ether remains colourless, or only slightly tinted, and the
substance below remains yellow while the alcohol is distinctly coloured
and the substance below is decolorized, then a foreign dye is pre-
sent.

(b) If both ether and alcohol are coloured either (1) lutein (egg
-colour) alone, or (2) this with a foreign dye is indicated.

(1) Add dilute nitrous acid to a portion of ether solution. If

the ether is not completely decolorized a foreign dye is
present.

(2) If the substance below the alcohol is decolorized, while

that below the ether is coloured, the following tests for
. foreign dyes should be applied. Shake the portion
previously treated with ether with three or more fresh
portions of the same solvent until no more colour can be
extracted, then shake the residue with 70 per cent alco-
hol and allow it to stand for twelve hours. Filter, then
concentrate the solution, slightly acidify with hydro-
chloric acid, boil with wool which will be coloured if
coal-tar dyes be present.
SchlegeVs Method. — Extract 100 grms. of the finely powdered
sample with ether in a continuous extraction apparatus. Shake the
residue with a mixture of 140 c.c. of alcohol, 5 c.c. of ammonia and
105 c.c. of water at frequent intervals for half a day. Filter, evaporate
to remove alcohol and ammonia, slightly acidify with hydrochloric acid,
then filter again. Boil the filtrate with fat-free wool and identify
the colour on the dyed filter by the usual methods.

Fresenius Method. — Extract 20 to 40 grms. of the powdered
sample with ether in a continuous extraction apparatus. Dry the
residue to remove the ether, shake for fifteen minutes with 120 c.c. of
60 per cent acetone, then allow to stand from twelve to twenty-four
hours. Filter, evaporate until free from the acetone, and divide into
two portions, a larger and a smaller. Add sufficient acetic acid to
dissolve any deposit to the large portion and boil with wool. Boil
the wool with dilute acetic acid to remove all natural colouring matter.
If the wool is dyed after this treatment, foreign colour is present
which can be identified by the usual tests.

To the smaller portion of the aqueous solution, obtained as just
described, add an equal quantity of alcohol, heat to dissolve the deposit,
divide into four portions, and apply special tests to three of these,
keeping the fourth for comparison.

Hydrochloric acid will decolorize the liquid if no artificial colour

be present; ammonia will intensify it; and stannous chloride will

not affect it. Saffron reacts in a similar way but is only slightly

bleached by the acid and is not affected by the other two reagents.

Piutti and Benitvoglio Method. — This method is especially in-

186 FOOD AND DRUGS.

tended to detect the four colours forbidden by Italian law and to dis-
tinguish them from naphthol yellow S. Boil 50 grms. of the paste in
500 c.c. of water alkalized with 2 c.c. of concentrated ammonia
water, add 60 c.c. to 70 c.c. of alcohol and continue to boil for forty
minutes. Filter, acidify the liquid with 2 c.c. to 3 c.c. of dilute hydro-
chloric acid and boil with 5 or 6 strands of fat-free wool, each strand
weighing about 0*5 grm. Wash the wool, dissolve the colour in
dilute ammonia and dye again. Dissolve a second time in ammonia,
then evaporate the solution of the dye to dryness, taking care to avoid
the formation of a skin, and take up the residue in water.

If, however, a skin has formed filter and test the insoluble matter for
metanil yellow, with dilute hydrochloric acid, and with ammonium
sulphide for picric acid.

Add stannous chloride solution and a little sodium hydrate or
preferably sodium ethylate to the filtrate. If there is no red color-
ation, nitro-colours are absent; if in another portion dilute hydro-
chloric produces no violet colour, metanil yellow is absent and no
other test is necessary. When these colours are present acidify the
remaining solution with acetic acid, and shake well with carbon
tetrachloride, and identify the colour from the following scheme : —

A. Carbon tetrachloride dissolves colour to a colourless solution.
Extract with exceedingly dilute ammonia, concentrate and divide
into two parts.

(1) Acidify with hydrochloric acid, then add the 2 drops of

stannous chloride and ammonia in excess. A ro e-
coloured solution and precipitate form.

Naphthol yellow.

(2) Acidify slightly with hydrochloric acid, add a little zinc

dust, then stir. Solution becomes rose-violet.

Victoria yellow.

B. Colour is insoluble in carbon tetrachloride. Evaporate to dry-
ness on a water bath, take up in water and divide into three parts.

(1) Hydrochloric acid gives a violet coloration.

Metanil yellow.

(2) Ammonium sulphide gives a red- brown coloration.

Picric acid.

(3) Stir on a water bath with zinc dust and ammonia, filter,

treat with zinc dust and hydrochloric acid, and again
filter.

{a) Potassium hydroxide produces a yellow colora-
tion.
(6) Ferric chloride gives a yellow coloration.

Naphthol yellow S.

Schmitz-Dumont Test for Tropeolins. — Add a few drops of dilute
hydrochloric acid to a portion of the sample. If there is a resulting
reddish or bluish colour, an azo colour or some other coal-tar colour
is present.

BAKING POWDEES.

187

Test for Turmeric. — Extract the colour from the ground sample by
alcohol and identify by the boric acid test.

Other Cereal Flours. — A microscopic examination, preferably side
by side with standard preparations, affords the only practicable means
of detecting adulteration of most of the cereal flours. The following
figures, however, are given as showing the average composition of the
chief of these flours, as occasionally information of a useful or confir-
matory character may be obtained by a detailed analysis. These
figures are those of A. H. Church : —

Water.

Albumenoids.

Starch.

Fat.

Cellulose.

Mineral Matter.

Per

cent

Wheat

130

10-5

74-3

0-8

0-7

Wheat bran

14-0

15-0

44-0

4-0

170

Oatmeal

5-0

16-1

63-0

10-1

3-7

2-1

Barley

14-6

6-2

76-0

1-3

0-8

1-1

Kice

14-6

7-5

78-0

0-5

0-9

0-5

Rye

130

10-5

71-0

1-6

Maize

14-5

9-0

64-5

5-0

2-0

Buckwheat

13-4

15-2

63-6

3-4

2-1

2-3

Peas

14-3

22-4

51-3

2-5

6-5

3-0

Beans

14-0

23-0

52-3

2-3

5-5

2-9

Lentils

14-5

240

49-0

2-6

6-9

3-0

Earth nuts

7-5

24-5

11-7

50-0

4-5

1-8

Baking Poivders. — It will be convenient to here briefly discuss the
composition and analysis of baking powders — not on account of any
chemical relationships with the starches, but because they are generally
used in conjunction with flour.

Baking powder is an article of considerable interest to analysts
working under the Food and Drugs Act, as it was largely due to the
decision in James v. Jones (58 J. P. 230), following on an earlier case,
Warren v. Phillips (44 J.P. 61), in which it was held that baking
powder was not a food within the meaning of the 1875 Act, that the
more extended definition of foods was embodied in the 1899 Act.

Baking powders are mixtures which, when added to flour, etc., give
off CO2 under the influence of moisture, and so enable bread and cakes
to be baked without the aid of yeast.

They may be roughly divided into three groups : —

(a) Tartaric acid powder, in which the acid constituent of the
powder is either tartaric acid .or cream of tartar or a mixture of both.

(b) Phosphoric acid powders, in which the acid constituent is an
acid phosphate.

Bicarbonate of sodium is invariably the alkaline constituent, and
nearly all baking powders, contain a good deal of starchy matter as a
diluent, and in order to absorb traces of moisture.

The properties of baking powder, from the point of view of the Food

188

FOOD AND DEUGS.

and Drugs Acts, are not merely a question of analysis. If proceedings
have to be taken ^ in regard to this substance, it is advisable that they
should be under s. 3 of the 1875 Act, when it will be necessary to
prove that such constituents as alum, etc., are harmful ingredients.

All the better-class baking powders are tartaric acid compounds.
Tartaric acid is more easily soluble than cream of tartar, and powders
made of the former evolve their CO2 more rapidly than those made
with the latter ; hence in many of the best powders, the two are mixed
so as to give an intermediate result.

The following are typical samples of well-known brands of baking
powder : —

Per cent

Tartaric acid

Cream of tartar

9-5

Bicarbonate of soda

23-5

Starchy matter

The theoretical quantities of the various acid materials necessary
for neutralization of the sodium bicarbonate are shown by the follow-
ing reactions : —

168
+ 2NaHC0,

Sodium bicarbonate
84
+ NaHCO-

Sodium bicarbonate
168

Na^C^HASH.O +200.,

KNaC4H408 + CO2 + H^O

150

HAH4O,

Tartaric acid

188

KHC4H40fi

Cream of tartar

234

CaH4(P04)2 + 2NaHC0, = CaHP04 + NaoHP04 + 2CO2 + 2HjO

Calcium hydrogen phosphate Sodium bicarbonate
516 504

KjAl 2(804)4 + ONaHCO, = Al2(OH)6 + 3Na2S04 + K2S04 4-6CO.,

Dry potash alum Sodium bicarbonate

In the analysis of baking powder the following determinations
should be made : —

Carbonic Acid. — This is the measure of the strength of the
powder, as its value depends on the quantity of gas liberated. Usually
the total and available carbon dioxide are both measured, as, through
deficiency in acid ingredients, the whole of the carbonates are not
always decomposed when the powder is employed for baking purposes.
The total carbon dioxide is obtained by treatment with excess of acid ;
the available by the use of water.

Any of the usual forms of apparatus for the measurement of
carbon dioxide may be used for this purpose. Thus, the well-known
Schrcedter flask may be used, in which the liberating acid and drying
tubes, etc., are all self-contained together with the powder. The loss
of weight after reaction is the amount of carbon dioxide evolved. In
using an apparatus of this form from 1 to 2 grms. of the powder are

BAKING POWDERS. 189

weighed out and transferred to the flask previously charged with dilute
sulphuric acid, and concentrated acid for drying the escaping gas ; the
whole apparatus is weighed and the acid allowed to enter very slowly.
Towards the end of the reaction, the flask should be carefully
warmed. Finally draw air through in the usual manner, and weigh
the flask again. Water must not be added to the powder before the
reaction is started. To estimate available carbon dioxide proceed in
the same manner, except that distilled water must be used for liber-
ating purposes, instead of dilute acid.

(2) Tartaric Acid. — Weigh out 5 grms. of the powder, transfer to
a 500 c.c. flask, and add 100 c.c. of water and 15 c.c. strong hydro-
chloric acid. When all action has ceased, make up with water to 500
c.c, and allow starch to subside. Filter and take 50 c.c. of the filtrate
and add 10 c.c. of 30 per cent potassium carbonate solution. Boil for
half an hour and filter into a porcelain dish, concentrate to 10 c.c, add
gradually and with stirring 4 c.c of glacial acetic acid, and then 100
c.c of 95 per cent alcohol, stirring the liquid until the precipitate float-
ing in it assumes a crystalline appearance. After standing some
hours, filter and wash with alcohol until free from acetic acid. Trans-
fer precipitate to a beaker, add water and boil. Titrate the resulting
solution with decinormal alkali — 1 cc of alkali corresponds to 0'0188
grm. of potassium bitartrate (cream of tartar), or 0"0150 grm. of
tartaric acid.

(3) Sulphuric Acid. — This may be estimated without previous
ignition of the powder. Weigh out 0*5 grm. and digest in a beaker
with strong hydrochloric acid until the whole of the powder including
the starch is dissolved, dilute with water, boil, and add barium
chloride in slight excess, allow to stand twelve hours, filter and weigh
the BaSO^.

(4) Alumina. — In the absence of phosphoric acid, from 0*5 grm.
to I'O grm. may be igaited, extracted with HCl, evaporated to dryness
to separate silica, treated with strong hydrochloric acid, again eva-
porated, filtered, and diluted with water, and alumina precipitated with
ammonia, washed, dried, ignited, and weighed. In the presence of
phosphoric acid, the following method may be used : Weigh out 5
grms. of the powder, heat until thoroughly carbonized, digest with
strong nitric acid, dilute, and filter into a 500 c.c. flask. Wash the
residue slightly, transfer it to a platinum dish, dry, burn, add mixed
potassium and sodium carbonates, and fuse. Dissolve in nitric acid,
evaporate to complete dryness, again dissolve in nitric acid, dilute,
and filter into a 500 c.c. flask. The flask will now contain both
series of filtrates ; make up to 500 c.c. with water. Take 100 cc and
precipitate with ammonium molybdate and nitric acid, digest and
filter. In the filtrate determine alumina by precipitation with
ammonia.

(5) Starch. — Starch may be determined by treatment with dilute
acid so as to convert into glucose, and then estimating this by Fehl-
ing's solution.

Phosphoric Acid. — This may be determined by igniting 0-5 grm. of
the sample, dissolving in nitric acid, diluting and filtering. The phos-

190 FOOD AND DRUGS.

phoric acid is precipitated by ammonium molybdate solution, the pre-
cipitate collected, washed with ammonium nitrate solution, then dis-
solved in ammonia, precipitated with magnesia mixture, the precipitate
washed with dilute ammonia, dried, ignited and weighed as pyrophos-
phate in ^e usual manner.

CHAPTER IV.

SPICES, FLAVOURING ESSENCES. ETC.

Under the above heading, the ordinary spices and condiments, and allied
bodies snch as cochineal and turmeric, etc. (used largely for colouring
foods), flavouring essences, and vinegar will be considered.

There are a number of useful determinations common to a number
of spices, which will be described before the individual substances are
dealt with.

Since most of the spices owe their characteristics to essential oils and
resins, which are soluble in ether, a determination of the ether extract
becomes of importance. But care must be taken to make it clear as
to what is meant by "ether extract," since the essential oils are
volatile at comparatively low temperatures. In the present chapter
the total ether extract is the residue left by the spontaneous evapora-
tion of the ether after a Soxhlet extraction, and then left in a sulphuric
acid desiccator for twelve hours. The extract is then dried at 100° for
several hours, the temperature being slowly raised to avoid oxidation
of the oil. The temperature is then increased to 110° till the weight
is constant. This gives the fixed ether extract, the difference between
the two being the volatile ether extract.

Starch Determination. — In substances containing much starch, it
is generally safe to convert by means of acid and estimate by titration
against Fehling's solution (p. 123). But when very small amounts
only are present, such as cayenne pepper, the diastatic conversion is
safer. Leach recommends that 4 grms. of the powdered sample be
extracted with 5 successive portions of 10 c.c. of ether and then with
150 c.c. of 10 per cent alcohol, on a filter paper. The insoluble matter
is washed into a 500 c.c. flask (if hydrochloric acid be used to convert
the starch) and 200 c.c. of water and 20 c.c. of HCl added. The
process is then carried out as described on p. 122.

If the starch is to be determined by the diastase method, the
residue is washed into a beaker with 100 c.c. of water and the pro-
cess carried out as described on p. 176.

Fibre. — This determination is best carried out on the residue left
from the ether extract. This residue is boiled for thirty minutes with
about 200 c.c. of sulphuric acid (containing 1-5 per cent H^SOJ. The
flask should be well shaken during the boiling, and after thirty minutes
the contents are poured on to a filter and the insoluble matter washed
with boiling water. The residue is • washed back into the flask and
boiled with a like quantity of 1*5 per cent solution of NaOH. After

(191)

192 FOOD AND DKUGS.

thirty minutes boiling the hquid is filtered, the residue washed with
boiling water until the washings are neutral, and then dried. The
weight of this residue, less the amount of ash it yields on incineration,
is the crude fibre.

Determination of Volatile Oil. — This is described at some length
under cloves (p. 224).

The Tannin Value. — Eichardson ("U. S. Dept. of Agriculture,.
Div. of Chem. Bull." 13, 167) determines the tannin value, either in
terms of oxygen absorbed, or of quercitannic acid, as follows : —

A standard indigo solution is prepared by dissolving 6 grms. of
pure potassium sulphindigotate in 500 c.c. of hot water, cooling, add-
ing 50 c.c. of concentrated sulphuric acid and making up to 1000 c.c.

A standard solution of potassium permanganate is made by dis-
solving 1*333 grms. of pure potassium permanganate in water to make^
1000 c.c.

Two grms. of the substance are extracted for twenty hours with
ether. The residue is boiled for two hours with 300 c.c. of water,
cooled, the liquid made up to 500 c.c. and filtered. Twenty-five c.c. of
this filtrate are run into a 1200 c.c. flask, 750 c.c. of distilled water
added and 20 c.c. of the standard indigo solution. Standard per-
manganate is then run in from a burette, until the colour changes to
golden yellow, which indicates the end of the reaction. The number
of c.c. of permanganate used is noted [a). The titration is repeated on
20 c.c. of indigo solution only. The number of c.c. used is noted {h).
a-h represents the number of c.c. of permanganate used to oxidize
the tannin present.

The permanganate solution is standardized against decinormal
oxalic acid, so that the amount of permanganate used is easily con-
verted into terms of decinormal oxaHc acid. Each c.c. of the latter is
equivalent to 0*008 grm. of oxygen absorbed, or to 0*062355 grm. of
quercitannic acid.

Wintons Lead Number. — Winton has shown that the amount of
precipitate obtained from a solution of certain substances by means of
subacetate of lead is fairly constant. The lead number represents the
amount of metallic lead precipitated by 100 grms. of the substance-
(essence of vanilla, for example). Twenty-five grms. of the liquid (or
an extract prepared by alcohol of expressed strength, from 25 grms.
of a solid), are mixed with 25 c.c. of a standard solution of lead sub-
acetate (made by diluting the ordinary liquor Plumhi subacetatis.
of pharmacy with four times its volume of water) and the whole made
up to 100 c.c. and allowed to stand, after well shaking for three hours.
It is then filtered, and to 10 c c. of the filtrate 40 c.c. of water are
added, excess of H.^SO^ and 100 c.c. of methylated spirit. After
standing for twelve hours the lead sulphate is filtered off, dried, and
weighed. The precipitate x 0*6829 gives the amount of metallic lead
in the sulphate (a).

Now determine the amount of lead in 2*5 c.c. of the standard lead
solution in the same manner {b). Then b-a represents the lead in
the lead precipitate from 2*5 grms. of the substance examined. So that.
(b-a) 40 represents the lead number of the substance in question.

GINGEK.

193

If a solution of normal acetate be uaed instead of the basic acetate,
different results are obtained, so that it is necessary to specify which
solution has been used.

GINGEK.

Ginger is the rhizome of Zingiber officinale, and although employed
principally as a spice, is a drug official in the British Pharmacopoeia.
It is usually sold in the scraped condition, and is so directed to be
used as a drug in the Pharmacopoeia.

It is not usually necessary to analyse whole ginger, except some-
times in order to determine whether there be any pieces of exhausted
ginger present. It is necessary, however, to examine the powdered
ginger, which is sometimes adulterated, especially with powdered spent
ginger : and preparations sold as prepared from ginger will often be
found which have been prepared from capsicum. This is true of
many samples of cheap ginger beer, which are frequently rendered hot
with capsicum and contain little or no ginger.

Ginger contains an oleo-resin, the resin being of a phenolic nature
and representing the pungency of the ginger, whilst the essential oil is
responsible for the aroma of the spice.

The presence of exhausted ginger is shown by a low cold water ex-
tract, a low soluble ash, and a reduced resin value. Other adulterants
of an organic nature are detected by the microscope.

Allen (" Analyst," xix. 124) gives the following tables of analyses of
ginger :—

VOL. I.

194

FOOD AND DKUGS.

Per
cent
4-50
1-37

7-39

Per
cent
3-29
0-97

8-08

^<Mr-<IOeOaD'*00<M <-H "H
^•S«trHC^aOS«Cqr-|Tt< t^- CO

*-i

0Ho(?«OO«b>HrH0bt?- rH l>-

t-H

0Hg«0^ol«»(N0S00 O 0»

»H^[^05 0 0CO«5 00 «0
•§J^«0(M'710 0CpCpcp to (N

CLigcbiHOiA- CO <?idb do A OD

" rH r-l

Per
cent
5-39

4-65

5-85

8-14

Per
cent
7-69
2-36
0-20

21-60

14-60
7-49

1-31

23-4

^+aC0l>«5»0 «3 O
fec»OaOr-lrf.«3 |lO 1 1 1

(^g«s<Modot>- I"* 1 ' 1

SprH»O.H(?q-* ,lO 1 1 1

Oh ^ »C(n6«5 6 1^1 1 1
rH rH

fee ^ CT ^ CO ® 1 ^ 1 1 1

PnglOli^oC^Oobwcb OS tH Tt<

.^tJHCOCOCOtJIIO C*C« rH »0
5J-5ipcOqiCp«p(N|»Ot> rH -rtl

0H®C«l?qOt>-rH6l-*05 rH »0
" (N Cq rH (M

Total ash

Ash Foluble in hot water .
Alkalinity of soluble ash as KgO .
Exti-aeted by rectified spirit

Exti acted by proof spirit . |

Containing ash ....

Extracted by cold water

Extracted by subf equent treat-
meat with proof spirit

Extracted by subsequent treat-
ment with rectified spirit

Total extract by the three solvents
used consecutively .

•93BI9AV

ssssss

CU g CO CO (M (N

•UBOLgV

CL, g CO CO cq o

•aBOUjY

P^ ?? W5 CO (^1 6
-^ rH rH

•aiqooo

^.-g (N00«0
Oh g CO CO <M rH

&H*

•aiqooQ

eu g CO eo (M 00

•mqooo

Per
cent
10-64

1-71
13-00

•BoremBp

CL, g CO 0(5 ,!( Al

•BoreuiBf

^ «0-0i U5 U3
Ph g (N CO rH (N

"BOIBUIBf

Om g CO CO CO -*

•BoreuiBf

Per

cent
10-98

1-41
13-25

:zi

•BOlBttlBf

Per
cent

11-26

1-70
15-65

Moisture
Total ash
Soluble ash
Cold water extract

GINGER

195

Richardson has pubHshed the following analyses of five samples of
ginger : —

Sample,

H20.

Ash.

Volatile
Oil.

Fixed Oil
and Resin.

Starch.

Fibre.

Albumenoids.

Nitrogen.

Per

cent

Calcutta

9-60

7 02

2-27

4-58

49-34

7-45

6-30

101

Cochin

9-4]

3-39

1-84

4-07

53-33

2-05

7-00

1-12

Unbleached Jamaica

10-49

3-44

2-03

2-29

50-58

4-74

10-85

1-74

KBleached

11-00

4-54

1-89

3-04

49-34

1-70

9-28

1-48

10-11

5-58

2-54

2-69

50-67

7-65

9-10

1-46

Moisture. — Normal samples of ginger contain from 9 to 12 per
cent of moisture. Any great excess of this quantity should not be
present, as the spice will have a tendency to become mouldy.

Mineral Matter. — The ash of genuine ginger is about 4 to 5 per
cent, sometimes falling as low as 3 per cent or rising to 6*5 per cent.
Any slight excess of this amount may be due to the presence of a little
dirt, but as ginger of an inferior quality is sometimes limed, any large
quantity of lime should be condemned. The following figures represent
about twenty-five samples examined by the author: —

I. Normal Pure Ginger.

Total ash
Soluble in H2O
Insoluble in acid

Jamaica.

Africa.

Cochin.

Per cent
3-4 to 5-4
1-8 „ 3
0-3 „ 0-8

Per cent
4 to 5-8
1-9 „ 2-9
0-4 „ 1

Per cent
3-8 to 5-6
1-7 „ 3-1
0-5 „ 1

II. Limed Gingers.

Total ash
Soluble in H2O
Insoluble in acid

Per cent
5-8 to 9-2 (CaO up to 3-1)
2-4 „ 3-9
0-6 „ 1-1

The amount of ash soluble in water should always be well over 50
per cent of the total. A lower amount indicates the presence of ex-
hausted ginger.

Cold Water Extract. — The sample should be well shaken at
intervals with twenty times its weight of water and an aliquot part of
the filtered Hquid dried at a temperature of 100°, until the loss between
the subsequent weighings at intervals of five minutes does not exceed
5 milligrammes. The cold water extract should not be less than 10 per
cent, and is usually from 14 to 16 per cent. Exhausted ginger materi-
ally reduces this figure.

196

FOOD AND DKUGS.

Carbohydrates. — The carbohydrates, determined by direct inversion
of the powder, and estimation by means of Fehling's solution, should
be not materially less than 50 per cent nor more than 55 per cent cal-
culated as starch.

Ether Extract. — The oleo-resin, extracted by ether, and dried at
about 65° C, until the weight is practically constant, varies from 3 per
cent to 6 per cent, rarely rising to 8 per cent- (in East Indian ginger).

Good quality Jamaica ginger rarely contains less than 5 per cent of
oleo-resin thus extracted. The extraction may be mside with alcohol,
when slightly higher results will be obtained, but it is less easy to drive
off all the solvent than when ether is used. According to Garnett and
Grier (" Pharm. Journal," 1909, ii. 159), the pure resin (gingerol) is
best determined by exhaustion with ether, recovering the solvent,
boiling the residue with several successive portions of petroleum ether
and then extracting the petroleum ether (which contains gingerol,
volatile and fatty oils and colouring matter) with three successive por-
tions of 60 per cent alcohol. The alcoholic liquid now contains all
the gingerol and some impurities. It is washed once with petroleum
ether to get rid of traces Of fat, the alcohol recovered, and the watery
liquid is extracted with three successive portions of ether, which is.
driven off and the residue weighed. If pure it should be quite soluble
in 1 per cent solution of KOH. The amount of gingerol present in
genuine ginger usually varies from 1*2 to 2 per cent.

The following may be taken as figures covering most pure samples
of ginger : —

Per cent

Moisture 8*5 to 14-0

Volatile ether extract 1-5 „ 3

Non-volatile „ 3-0 „ 6

Alcohol extract 3-6 „ 6-8

Cold water extract 10-5 „ 18

Starch ,50-0 „ 55

fibre 3 „ 7-5

Albumenoids (N x 6-25) 6-5 „ 11

Mineral matter 3 „ 6

Ash insoluble in HCl 0-02 ., 2-3

The following values indicate the composition of ginger after it has-
been partially exhausted, for essence or ginger ale manufacture : —

Essence Residue

Gine^er Ale Residue

(i.e. Exhausted by Spirit).

(i.e. Water-Exhausted).

Per cent

HjO

8-02

10-61

Volatile ether extract

0-13

1-61

Non-volatile „

0-54

3-86

Alcohol „

1-52

4-88

Cold water ,,

16-42

615

Starch ....

54-57

Fibre ....

5-17

Albumenoids

6-94

Ash

506

1-1 to 2-12

„ soluble in HjO .

3-55

0-2 „ 0-59

„ insoluble in HCl

1-50

0-18

GINGER

197

Dyer and Gilbard ("Analyst," xviii. 197) first called attention to
the water-soluble ash as a reliable means of indicating water-exhausted
ginger. Six samples of pure and exhausted ginger gave the following

results : —

Total Ash

Water

Alcohol Ext.

Sol. Ash.

After Ether Ext.
Per cent

Per cent

r Highest

4-1

8-0

3-8

Pure ginger

< Lowest

1-9

2-1

I Average

3-8

2-7

2-8

/-Highest

2-3

0-5

1-5

Exhausted ginger

< Lowest

1-1

0-2

0-8

I Average

1-8

0 35

1-2

Microscopic Examination. — The starch grains are more or less
sack-shaped, with somewhat rounded ends. The smaller grains are

Fig. 22. — Powdered ginger.

nearly circular. The hilum and striaiions are almost invisible, the
former being close to the pointed extremity of the grains. The grains

198 FOOD AND DKUGS.

measure from 20 to 30 /x, although a few will be found measuring so
little as 15 {x or so much as 50 /x.

An examination should be made, after the starch has been removed,
by boiling 5 grms. with 50 c.c. of 5 per cent hydrochloric acid. The
tissues should be washed with water and then examined in a 50 per
cent solution of chloral hydrate. Characteristic fibres will be found,
with spirally arranged pits.

The Detection of Capsicum in Ginger Preparations. — Garnett,
Grier and La Wall (" Analyst," xxxiv. 321) recommend the following
process. The ginger ale, etc., is warmed to expel CO^, and if alcohol
be present this is driven off also. The aqueous residue is acidified with
dilute sulphuric acid and shaken with 50 c.c. of ether for a minute.
If the residue from the ether, which is allowed to evaporate spontane-
ously, weighs less than 10 milligrammes it is treated with 2 c.c. of

•^ alcoholic caustic potash solution. An additional 1 c.c. of alkali is

added for each further 10 milligrammes. The mixture is transferred
to a test tube fitted to a reflux tube and gently boiled for thirty minutes
in a Water bath. The alcohol is drained off, and the test tube is half
filled with water, and the liquid well shaken with half its volume of
ether. The ether is separated and evaporated. If the residue has a
hot, pungent taste, capsicum is present — the phenolic constituents of the
ginger being retained by the potash. One part of capsicum in 10,000
of water can thus be detected. Samples of gingerine are similarly
examined, using 50 milligrammes of the sample.

Nelson ("J. Ind. and Eng. Chem." 1910, 2, 419) prefers to take
the ether extract from 100 c.c. of a beverage, which is first heated to
drive off alcohol, and evaporate it with 10 c.c. of twice normal alcoholic
potash. About 7 mgs. of manganese dioxide and 5 c.c. of water are
added, and the whole heated till volatile oils are driven off. The cold
liquid is acidified with dilute H2SO4, and at once extracted with
petroleum ether. The solvent is evaporated and the residue touched
with the tip of the tongue, when the burning taste of capsicum, if pre-
sent, cannot be mistaken.

PEPPER

Pepper consists of the not quite ripe fruit of Piper nigrum, culti-
vated in India and the islands of the Malay Archipelago. Black
pepper consists of the entire fruit, whilst white pepper consists of the
berries deprived of the outer portion of the pericarp.

Constituents. — Pepper consists of the usual plant tissues, together
with a small amount of an essential oil (from 1 to 2 per cent), a small
amount of a bitter, hot, pungent resin, and from 3 to 7 per cent of the
alkaloid piperine, with possibly a small quantity of piperidine. It
contains an appreciable amount of pepper starch.

The quality of pepper depends almost entirely on the amount of
resin and alkaloid, although the flavour is influenced by the amount
of essential oil.

PEPPER

199

A number of samples of various origins have been examined by the
author, and the following results obtained : —

Penang pepper (10 samples)
Suma.tra pepper (8 samples)
Malabar pepper (10 samples)
White pepper (origin unknown)

Moisture.

Ash.

Watery
Extract.

Alkaloid.

Resin.

Per cent

8-35 to 9-96

8-5 „ 110

9 „ 10-8

10 „ 12

Per cent
4 to 6
4-5 „ 61
4-8 „ 5-7
1-2 „ 1-6

Per cent
17 to 19-6

17 „ 20-3

18 „ 20
20 „ 22

Per cent
5 to 6-2
4-3 „ 5
4-0 „ 5-8
4-3 „ 5-3

Per cent
2 to 2-2
2 „ 2-3
1-9 „ 2
2 „ 2-8

The following table is abridged from one by Heisch (" Analyst,"
XI. 188).

Jlack peppers
White peppers
Long peppers
Black pepper

husks
^ Sittings before
■ grinding

Ash in Moisture Free Pepper

Total.

Per cent
4-35 to 8-99
1-28 „ 3-78
12 „ 13-5

11-9

51-4

Sol. in HaO.

Per cent
1-54 to 3-34
016 „ 0-61
2-3 „ 2-4

212

1-02

Insol.inHCl.

Per cent
0-04 to 4-38
0 „ 0-69
3-7 „ 5-7

3-41

43-90

Per cent on the Moisture and Ash
Free Pepper,

Starch.

Per cent
48 to 57
76 „ 85
46 „ 59

41-7

30-66

Alcohol
Extract.

Per cent

11-6 to 16-2

9-2 „ 10-6

8-3 „ 8-5

13-8
7-5

Piperine.

Per cent
•05 to 9-38
50 „ 614
70 „ 1-71

4-84 .

1-15

Richardson gives the following values

H2O.

Ash.

Volatile
Oil.

Piperine
and Resin.

Alcohol
Extract.

Per

1
Starch.

Per

Fibre.
Per

Per

cent

ceut

cent

Black

8-91

4 04

0-70

7-29

36-52

10-23

1-57

8-29

4-70

1-69

• 7-72

606

37-50

1002

202

9-83

3-70

1-60

7-15

5-74

37-30

10-02

1-93

White

9-85

1-41

0-57

7-24

40-61

7-73

1-83

10-60

1-34

1-26

7-76

2-57

43-10

4-20

1-90

Gladhill ("Amer. Jour. Pharm.'
of genuine samples of pepper : —

76, 71) gives the following analyses

This cannot be considered as pepper at all.

200

FOOD AND DRUGS.

•^

Per

cent
2-27
0-80

2-47
2-30

2-10

1-06

Per
cent
2-94
0-70

2-85
1-76

2-13

o o -rs c<>

1 rt r^ 6 6

t>l

« g Oa)C;-(?il<NO»{3r-tU5
^ CJ rHCqCT«5«S(?q(rqCTO

W tH o o

co'

-•

^+f aoioiocqocoQOO'*

^g CO'*OScp(?5tHCT»OTt<
^t) (NC5»HCOOqrH(N(NO

rH fH tH rH

-**

Per
cent
7-33
6-82

6-31

7-28

7-10

6-84

•>j5

Per
cent
6-58
6-56

5-98
7-00

7-67

CO O CO IM

fH (?q qp o
i> t- »b t>

S--<± 00<N>OrH00O00CO'*

« g coot>cpoicpcpq5qs
f^ o t>-cbcbvccbdbcbt>-05

O 00 CO o
O t~ I> CO

OS CD CO CO

^ COr-IOIMOCOOeOIN
gj- r-(qit>fHW5t;-0»00

Q-i*t>J5t>ibcbi>-.*-c^ b

iH CO -^ >C

qp (N i> «

CO l>- IC CO

OS
CO
CO

■^

Per
ceut
9-60
7-&2

8-78
9-58

9-20

fill

Per
ceut
9-52
7-26

8-83
8-76

9-80

do do CO b-

t>l

Pm8 dbaodsooosooscbo

rH rH iH

»H 00 t>- CO

-■

, jj e0'«!t<U5Tt<O(M00C0C0

Oig oioDCicbaDobosoo

tH t-H

t- 00 Tj( -^t*
OJ C^ O CO

do do t- i>

\e>

■^*

Per
cent
4-5
4-5

4-0
5-2

4-7

1 1 1 1

■<*

Per
cent
4-2
3-8

3-6
5-4

4-0

O IN CO (N

tH rH <?q rH

co'

i"c •>"<Jpt;-cpo>OOiC(M

Op O qp (X>
O fH O) 6

-■

§3 a ipt>c;-OiGpococou5

O tH rH OS
rH tH Cil tH

Singapore

Tellicheiiy

Aleppo

Trang

Lienburg

Lam pong

W. C. Sumatra

Acheen A

Coriander
Singapore
Penang
Decorticated

o
a

PEPPER

201

The following are numerous analyses of various American chemists
(Brooks, " Federal Spice Standards, 1909 "), of black, white, and long
peppers.

Black Peppers.

Lsifc

2 S

irts Nitrogen
100 Parts No
volatile Eth
Extract.

V^ariety, Etc.

•28

1 ^

Per

cent

Singapore (14 samples)

Minimum .

8-20

0-99

6-57

33-75

10-02

3-91

3-09

0-07

3-90

Maximum .

12-43

1-94

7-92

39-66

13-82

4-22

4-95

0-56

5-46

Tellicherry (7 samples)

Minimum .

8-42

0-65

6-72

36-03

11-98

3-88

4-06

0-00

4-13

Maximum .

11-86

1-55

7-02

41-75

13-21

4-14

4-69

0-10

5-11

Aleppo (5 samples)

Minimum .

8-46

1-12

7-48

34-65

11-36

3-72

4-74

0-07

4-29

Maximum .

10-01

1-90

8-87

41-60

13-01

3-98

5-02

0-30

5-41

Malabar (2 samples)

Minimum .

9-47

1-04

6-10

36-84

9-68

3-86

3-45

0-09

4-28

Maximum .

10-53

1-51

7-71

44-83

12-78

4-00

4-40

0-20

5-74

Lampong (10 samples)

Minimum .

8-09

1-11

6-81

33-41

10-25

3-82

4-86

0-48

3-32

Maximum.

12-17

2-10

9-05

39-46

13-50

4-27

6-52

1-80

3-79

Trang (5 samples)

Minimum .

8-09

1-22

6-60

35-73

10-58

3-79

3-43

0-33

8-82

Maximum .

11-57

1-60

6-97

41-00

13-11

4-10

416

0-41

418

Acheen A (3 samples)

Minimum .

8-73

1-09

9-17

28-00

13-07

4-02

5-04

0-48

3-20

Maximum .

12-09

1-71

10-44

33-72

16-97

4-21

6-49

0-96

3-69 1

Acheen B (3 samples)

Minimum . .

8-89

1-15

9-03

25-09

14-09

4-06

5-80

115

2-52

Maximum .

12-95

2-07

9-16

33-08

18-84

4-13

6-62

1-36

2-79

Acheen C (4 samples)

Minimum. ' .

9-62

1-28

7-99

22-05

16-40

3-94

6-10

1-00

2*12

Maximum ,

12-33

2-05

9-64

33-38

18-25

4-18

8-04

2-59

2-82

Acheen D (2 samples)

Minimum .

10-03

1-66

8-24

28-00

17-98

4-05

6-75

1-52

—

Maximum .

10-06

1-98

8-81

28-40

18-89

4-15

7-00

1-62

2-46

Mangalore (3 samples)

Minimum .

8-53

1-50

6-81

34-62

10-00

3-46

4-03

0-05

8-57

Maximum .

11-61

1-87

9-08

36-95

10-42

4-06

4-74

0-19

9-72

Shot pepper (3 samples)

Minimum .

8-40

1-16

6-66

33-19

10-58

3-29

3-66

0-20

4-84

Maximum .

11-50

1-41

7-49

38-60

13-04

4-07

4-15

0-28

6-00

202

FOOD AND DRUGS.

White Peppers.

§3

C fl fe

«- «

Variety, Etc.

|§3

arts Nitrogen
100 Parts No
volatile Eth
Extract.

'S = <5

:z;

Per

cent

Singapore (9 samples;

Minimum .

8-15

0-90

5-68

53-11

3-39

4-22

0-94

0-06

4-35

Maximum .

13-82

1-66

7-94

59-34

6-10

4-35

1-61

0-20

5-20

Penang (10 samples)

Minimum .

8-04

0-62

5-65

48-88

3-70

4-02

2-15

005

4-79

Maximum .

14-19

1-64

6-50

54-74

7-65

4-37

4-28

0-28

5-62

Siam (5 samples)

Minimum .

8-66

0-58

5-71

55-01

3-49

4-20

1-26

0-04

4-49

Maximum .

14-47

1-37

6-81

56-33

3-91

4-48

1-77

0-22

5-48

Tellicherry (1 sample)

10-49

115

6-09

57-09

3-39

4-31

0-86

0-07

5 40

Decorticated (8 samples)

Minimum .

8-14

0-49

5-96

57-38

0-10

4-23

1-00

000

2-56

Maximum .

13-34

1-50

7-26

63-60

2-07

4-53

2-24

0-15

3-47

Coriander (1 sample)

10-22

0-85

6-48

56-60

4-14

4-15

103

0-05

4-22

Long Peppers.

Minimum.

Maximum.

Per cent

Moisture . .

8-43

10-13

Volatile ether extract

0-79

1-55

Non-volatile ether extract ... . .

5-71

7-53

Stirch (diistase method)

28-43

45-87

Crude fibre

5-76

1001

Total ash

5-93

1439

Ash insoluble in acid .......

0-22

5-92

Parts nitrogen in 100 parts no i-volatile ether extract

3-12

3-56

The following determinations should be made : —
Moisture. — About 5 grms. of pepper coarsely bruised if the whole
berries are being examined, or of the powdered pepper, are dried in a
water oven to constant weight. Any excess over 13 per cent may be
regarded as moisture purposely added. The average value is from 8
to 11 per cent.

Mineral Matter.^-The portion used to determine the moisture is
ignited in the usual manner, and the ash weighed. It should never
exceed 8 per cent, and this ficrure may be regarded as rather excessive,
5 to 6 per cent being the average. Long pepper, however, yields an ash
of up to 14 per cent. Since the husks contain up to 14 per cent of
mineral matter, the white or decorticated pepper always has a lower

PEPPER. 203

ash value than black pepper. Pepper is said in most text books to
be adulterated with chalk, barium sulphate, and such mineral additions.
In ground pepper these adulterants are very rarely met with, but in
whole pepper, the author has on several occasions met with samples
which are not decorticated, but consist of black pepper carefully
coated with kaolin and sold as white pepper. The kaolin is rendered
adherent with a little gum tragacanth or similar substance, and the
black pepper can be treated so well in this manner as to deceive the
casual purchaser. Naturally a very high ash value is found in such
samples.

The Ash of Pepper. — An average black pepper ash contains the
following : —

Per cent

Potassium (as 'K.^O) ......... 24

Sodium (as Na-^O) 3-5

Magnesia 12-0

Lime 12-0

Iron 0-25

P2O5 8

SO3 9

Chlorine 8

Silica 6

The following values are representative of normal peppers.
(Leach) : —

Total Ash.

Soluble in Water.

Lisoluble in HCl.

Per cent

Black

3-49

2-10

012

4-21

275

001

6 05

2 37

106

504

2-78

0-48

White

106

0-47

001

„

1-33

0-33

0-09

1-47

0-38

0-10

2-84

0-65

015

Long pepper

5-93

4-20 *

0-22

Besin and Alkaloid. — The pepper is ground to a fine powder, and
well exhausted with strong methylated alcohol. The alcohol is eva-
porated at about 60°, and the extract left in a desiccator for 24 hours.
The extractive matter is almost entirely alkaloid and resin. A separa-
tion, not very exact, but sufficiently accurate for practical purposes,
may be affected by treating the extract with a 5 per cent solution of
caustic potash, pouring off the aqueous liquid, washing wdth water and
redissolving the alkaloid in 95 per cent alcohol. The solution is
filtered, the alcohol evaporated at 60°, and the residue allow^ed to stand
in a desiccator for twenty-four hours, and weighed. According to
Stevenson (" Analyst," xii. 144) black pepper contains up to 7-14 percent
of piperine, and about 1'5 per cent of resin, w^hilst w-hite pepper con-
tains up to 6-47 per cent of piperine and 0-69 per cent of resin.

Stoddart ("Analyst," xiv. 37) has recorded a sample of pepper

204 FOOD AND DRUGS.

containing a mixture of starch, barium sulphate, chalk and lead
chromate. The latter is said to have been added to improve the
oolour. Such an adulteration is not likely ever to be met with again.

Watery Extract. — Five grms. of the pepper should be powdered and
shaken at frequent intervals for twenty-four hours, with 100 c.c. of
water, the whole filtered, the residue washed with-two or three portions
of about 20 c.c. of water, the washings added to the filtrate, and the
liquid evaporated on a water bath. The extract should amount to 17
to 22 per cent.

Determination of Starch. — Lenz regards the determination of the
starchy matter in pepper as of the greatest importance. He carries out
the estimation in the following manner : —

About 3 grms. to 4 grms. of the powdered sample are heated
with 250 c.c. of cold water for a few hours with constant agitation.
The solution is filtered and the residue washed with cold water, and
the insoluble matter washed into a flask and diluted to 200 c.c. with
water. Twenty-five c.c. of a 25 per cent solution of hydrochloric
acid are added and the flask, attached to a tube or condenser, is heated
with occasional agitation, in a water bath for three hours. On cooling,
the liquid is made up to 500 c.c. after neutralization with caustic soda
solution. The copper-reducing sugars are then determined by means
of Fehling's solution.

Genuine pepper gives a result equivalent to at least 50 per cent of
starch calculated on the ash and moisture free pepper, whereas most
adulterants except starch itself give much lower results. Rottger gives
57 per cent to 60 per cenf as the average value for black pepper and
59 per cent to 74 per cent for white pepper.

This method, of course, involves the determination of all bodies
which yield sugars or reducing substances on hydrolysis by acids.
When the diastase conversion method is used the results will be
from 5 per cent to 10 per cent lower, averaging from 40 per cent to
48 per cent on the moisture and ash free pepper.

Determination of Nitrogen. — The most satisfactory method for the
determination of the nitrogen in pepper, which varies from 1*95 per
cent to 2*55 percent is ths Gunning- Arnold modification of Kjeldahl's
process : 1 grm. of the sample in powder is mixed with 1 grm. each of
copper sulphate and red oxide of mercury and about 16 grms. of K2SO4 ;
25 c.c. of H2SO4 are added and the digestion and distillation carried
out in a 600 c.c. flask. After about four hours digestion at boiling
temperature, the liquid is cooled and 300 c.c. of water and 50 c.c.
of a 4 per cent solution of potassium sulphate are added, and finally
enough NaOH solution to render the liquid alkaline. The ammonia
produced is then distilled in the usual manner for Kjeldahl's deter-
mination.

Adulteration of Pepper. — Such adulterations as chalk, sand, clay,
barium sulphate, and the like are readily indicated by the high ash
value of the sample. Further, as the mineral matter of pepper is in
combination with organic matter, it is not nearly so heavy as added
mineral matter. By shaking — say 5 grms. — of the powder in a
separating funnel with chloroform, the added mineral matter, with a

PEPPER.

205

small quantity of the natural husk material, rapidly sinks to the
bottom and may be drawn off and examined. The presence of added
mineral matter is not uncommon even in the whole peppercorns. In-
ferior black pepper is sometimes coated with lime and mixed with
white peppercorns and sold as white pepper.

Commercial adulterated pepper frequently contains ground rice
or other starchy matter. These starchy adulterants are recognized
by a microscopical examination of the sample, together with a low
ash value, a high starch content, and a low result for resin and alkaloid.

Exhausted ginger has been found in pepper. This may be recog-
nized by the appearance of the ginger starch under the microscope.

Perhaps one of the most common adulterants found in pepper of
recent years is ground olive stones, originally known under the un-
suitable name ot poivrette.

According to Campbell Brown the composition of ground olive
stones is as follows, a comparison also being made with ground almond
shells : —

Ash

Starch

Soluble in dilute HCl .
Insoluble in acid and alkali .

White
Poivrette.

Black
Poivrette.

Almond Shells.

Olive Stones.

Per cent
1-33

none
38-3

48-5

Per cent
2-47
none
34-5

Per cent
205
none
23-5
51-7

Per cent
1-61
none
39-1
45-4

No starch is present in this adulterant, but substances capable of
inversion are present and an apparent starch content of 10 per cent
will be found.

The following figures are of interest as showing the general char-
acter of some of the possible adulterants of pepper : —

Cocoanut Shells.

Almond Shells.

Date Stones.

Walnut Shells.

Per cent

Ash (total)

0-54

2-90

1-25

1-40

• „ H2O soluble

050

2-40

0-76

0-8

„ insoluble in acid

005

004

Ether extract

0-25

0-75

8-5

0-65

Alcoholic extract

1-2

5-20

16-7

190

Nitrogen

015 to 0-2

0-2 to 0-3

0-85

0-3

The shells of black pepper, which have been removed in the pre-
paration of white pepper, are sometimes ground in with the pepper of
commerce. It would be difficult to say this is an adulteration in a
legal sense, but to the analyst it would be indicated by an excess of
the characteristeristic stone cells seen under the microscope, and by a.
high ash value.

The estimation of pentosans (see under cocoa, p. 23) is of value

206 FOOD AND DEUGS.

in indicating the presence of both peppei- and other shells in ground
pepper. According to Hehner ('Analyst," xxiv. 181) genuine white
pepper yields 1-68 per cent of pentosans (ool per cent of crude fibre) ;
genuine black pepper, 4'58 per cent of pentosans (9"91 per cent of
crude fibre) ; and pepper husks 10*24 per cent of pentosans (14-63 per
cent of crude fibre).

Numerous colour reactions have been suggested to detect the pre-
sence of ground olive stones in pepper. The sample may be mixed
into a paste with dilute caustic soda solution, the paste diluted with
water and the residue washed by repeated decantation. Particles of
ground olive stones are coloured bright yellow, whilst the darker
particles are black pepper husk. Bleached pepper husk, however, takes
on a very similar colour.

Chevreau utilizes the fact that an acid solution of aniline colours
the sclerenchymatous tissues yellow, but leaves the other tissues un-
affected. The sample is moistened with a solution of aniline in three
times its volume of strong acetic acid. To the naked eye no change is
apparent, but under the microscope only a few isolated yellow cells can
be found, if the pepper is pure, whilst in the presence of olive stones or
almond shells, the sample becomes of a yellow colour and numerous
yellow cells can be observed. Martelli prefers to digest 1 grm. of
phloroglucinol in 50 c.c. of hydrochloric acid for a few days and decant
the clear liquid. The sample is covered with a reagent and heated for
a few minutes. Olive stones and similar tissues give a reddish-violet
colour whilst pepper is only tinged yellow or famtly brown. If water
be added and decanted the stained woody tissue is left as a sediment.

For other colour reactions, reference may be made to a note by
Pabst ("Journal Soc. Chem. Industry," 9, 770) who recommends a
solution of dimethyl-jLJ-phenylenediamine, which colours the ground
olive stones a carmine red colour, leaving the pepper almost unchanged.
Thalline sulphate (a 1 per cent solution in water) stains ground olive
stones a fine orange colour, leaving the pepper but little affected.

To form an approximate judgment as to the amount of ground
olive stones in a sample of pepper, the amounts of ash and of starch,
both of which will be reduced, are taken into account, together with
the amount of woody fibre. The last named may be determined in
the following manner : —

Two grms. of the powder previously exhausted by ether are boiled
with 200 c.c. of 15 per cent H2SO4 for 30 minutes, and the residue
filtered off. This is now boiled for 30 minutes with 1-5 per cent solu-
tion of caustic soda, and the fibre filtered off, washed with water, dried
and weighed. This should be done under a reflux condenser, and the
flask requires very firm attachment owing to the frequent bumping
that takes place. The residue is filtered through a dried tared filter,
washed with hot water, dried and weighed. The following are the
average amounts of woody fibre obtained : —

Per cent

White pepper 3 to 9*5

Olive stones . . 60 ,, 75

Black pepper 10 ,, 18

Almond shells . 65 „ 80

PEPPER.

207

It will be convenient to here mention long pepper, the fruit of Pipp.r
officinarum (Malay) and of P. longum (Bengal and Philippine Islands).
This is a spice used in pickles, but rarely found in retail commerce.
It is seldom, if ever, found in the powdered form, so that adulteration
is most exceptional. J. Campbell Brown gives the following analyses
of three samples of long pepper : —

Total ash

Ash insoluble in HCl ....

Starch (and other sugar-producing bodies)
Albuminous matter soluble in KOH
Cellulose ......

Alcohol extract . . . . .

Ether extract .....

Nitrogen

Per cent

8-91

1-2
4404
15-47
15-70

7-7

5-5

2-1

Per cent

8-98

1-1
49-34
17-42
10-50

7-6

4-9

2-0

Per cent

9-61

1-5
44-61
15-51
10-73
10-5

8-6

2-3

The Microscopic Examination of Pepper. — A small quantity of the

Sfei^

Fig. 23. — Powdered black pepper.
powder should be examined in dilute glycerine (1 in 4 of water). A

208 FOOD AND DEUGS.

large portion of the powder consists of fragments of connected polygonal
cells (of the perisperm) packed with minute starch granules : many of
the cells are broken, and tiny starch grains, isolated, or in connected
groups, can be identified. Irrigation with iodine solution will be useful
to identify the starch grains. A second preparation should be made
by boiling for ten minutes a little of the powder with a dilute solution
of hydrochloric acid (about 0'5 per cent). The powder should then
be washed with dilute solution of potash (1 per cent) and then with
water. It is best mounted in chloral hydrate solution (70 per cent).
The starch is destroyed and the cell structures are observed to
better advantage. Thick -walled sclerenchymatous cells, fibrovascular
bundles, with spiral vessels and empty parenchymatous cells are all to
be found. The general appearance of ground pepper under the micro-
scope is as shown below.

Hanausek (" Zeit. F. Untersuch. der Nahr. und Genussmittel,"
1898, 490,) records a case of adulteration with ground coriander seed.
The characteristics of this spice under the microscope are (1) bundles
of corrugated bent fibrous cells, (2) coarse parenchyma overlaid with
narrow cells of a yellow colour, with parallel walls, (3) colourless
cellular parenchyma enclosing crystals in rosettes.

Cayenne Pepper.

Cayenne pepper consisis of the powdered fruits of various species
of capsicum, of which Capsicum putescens is the principal. C. fasti-
gatum, C. annum and C. minimum are other species used. The so-
called " tasteless cayenne " is derived from a species of Pimento grown
chiefly in Spain and is without any pungency. It is used principally
as a colouring material for certain types of sweet pickles, and for
stuffing olives, or for imparting an orange colour to canaries, who
eat it mixed with their ordinary food.

The capsicum pods, in their entire condition, are known as
chillies.

As a drug, Capsicum minimum is official in the British Pharma-
copoeia. The only standard there laid down for the whole fruit, is that
it should not yield more than 6 per cent of ash.

Cayenne pepper is not very frequently adulterated, and certainly
never with the absurd adulterations usually enumerated in text books,
such as cinnabar.

It owes its virtue to the presence of an oleo-resin of a somewhat
complex character, and to a definite crystalline body isolated by Thresh „
and named by him capsaicin. Its formula is represented empirically,
at all events, by CjgH^.gNOg and it melts at 64'5''. It is present to-
the extent of about 005 per cent to 0*14 per cent.

On heating the minutest trace of cayenne pepper, these acrid prin-
ciples volatilize and produce such an intensely irritating vapour which
so affects the throat that the presence of even a minute quantity of
cayenne could scarcely be overlooked.

According to Eichardson, the following represents the average com-
position of cayenne pepper : —

PEPPEE.

209

Seed.

Pericarp.

Whole Fruit.

Per cent

Water .

8-12

14-75

11-94

Alburaenoids

18-31

10-95

13-88

Ether extract

28-54

• 5-48

15-26

Fibre .

17-50

23-73

21-09

Ash .

3-20

6-62

5-20

Nitrogen

2-93

1-71

2-22

Wynter Blyth gives the following figures as the means of several
samples : —

Per cent

Aqueous extract of dried pepper . . . . 32- 1

Alcoholic extract of „ „ .... 25-79

Benzol extract 20-00

Ether extract 10-43

Ash 5-69 (soluble 3-32)

Nitrogen 2-04

The ether extract is here far too low, and this figure must not be
taken as at all representative.

Miiieral Matter. — The ash of cayenne pepper should vary between
4 per cent and 7 per cent, with a maximum of 1*2 per cent of siliceous
matter insoluble in hydrochloric acid. Barely, a sample may contain
7'5 per cent of mineral matter. Over 50 per cent of the ash is soluble
in water.

Barium salts are sometimes found in adulterated cayenne, which
has been artificially coloured with a coal-tar lake on a barium base in
order to improve the colour.

Oleo-resm. — The amount of oleo-resinous matter obtained by
various solvents gives a good indication of the character of the sample,
and its determination will guard against the presence of exhausted
cayenne.

It must be- remembered, however, that capsicum fruit contains much
fat, so that all extracts with organic solvents are merely mixtures of so
much oko-resin, fat, and other extractive matters. The most reliable
figures from the extractive matter of genuine capsicum are those of A.
W. Gerrard. He has carried out a number of experiments with the
following results : —

Ten grms. of capsicum in No. 60 powder were packed in each of
six percolators, composed of glass syringe tubes, and slowly percolated
respectively with ether, 90 per cent alcohol, benzene, petroleum ether,
bisulphide of carbon, and chloroform until 100 c.c. of percolate had
collected from each. The percolates were evaporated over a steam
bath until the solvent was quite removed. The residues obtained were
weighed, and are given here as percentages : —

VOL. I. 14

210 FOOD AND DKUGS.

Percentage Yield
Solvents Used. of Extract.

Ether 18-2

Alcohol, 90 per cent 26-4

Benzene 18 6

Petroleum ether 16-4

Bisulphide of carbon 16-7

Chloroform 17-5

The physical characters of each extract differed somewhat ; all of
them on standing twenty-four hours deposited a soft granular fat, and
separated a fluid dark red resin. The fat yielded from the carbon
bisulphide was somewhat crystalline, and in the resinous portion of the
extract numerous small crystals were seen floating. The palest coloured
extract was obtained by petroleum ether, which solvent does not readily
remove the colour from capsicum. To make certain that the powdered
capsicum in each case had been properly exhausted, small portions of
each marc were tested, and except in the case of the alcohol-treated
marc, all were found to possess much pungency. The alcoholic marc,
though slightly warm to the taste, might certainly be considered as
practically exhausted. It is thus evident that a much larger yield of
extract is obtained by alcohol than by the other solvents, the alcohol
giving 26-4 per cent against an average of 17 5 per cent from the
others, a difference in favour of the alcohol of about 35 per cent. The
dried marcs or residues of the previous extractions, except that with
alcohol, were again packed in the percolators and treated with 90
per cent alcohol until 100 c.c. had been collected ; on evaporation of
the alcohol there was obtained in each case a brown, resinous, strongly
pungent residue, in the following proportions : —

Per cent
Prom the ether marc 7-9

benzene

petroleum ether
bisulphide of carbon
chloroform

7-5

9-0
7-2

7-4

By adding these figures to the previously obtained figures, under
iiheir proper solvents, we get approximately the same amount of extract
as when alcohol alone is employed. It is thus clear that alcohol is
the most powerful and perfect solvent of capsicum.

From numerous analyses by the author, these figures have been
'iully confirmed, and the following may be taken as the usual limits for
,the extractive matter of genuine cayenne pepper (Gerrard's figures
■.appear to have been obtained from Capsicum minimum) :

Per cent

Alcoholic extract 23-5 to 27-5

Ether extract 15-5 „ 19

Benzene extract 160 „ 18-6

CS2 extract 15-0 „ 19

Chloroform extract . . 16 0 „ 185

The following may be taken as the average values for the usual
Analytical figures for cayenne pepper : —

PEPPER.

211

3-5

to 7-5

M 7-2

0-05

„ 0-3

3-3

„ 0-3

15-5

„19

0-7

M 2-8

0-8

„ 1-5

„25

„ 15

„ 34

„30

Per ceut
Moisture .
Ash .

„ insoluble in HCl

„ soluble in HgO .
Non-volatile ether extract
Volatile ,, „

Starch
Fibre

Albumenoids
Aqueous extract
Alcoholic „ .

Microscopical Examination. — The best method of examining
cayenne pepper is to defat some of the powder with ether-alcohol, and
then mount in chloral hydrate. This should be done, after examining
some of the original powder in water, when the numerous red globules
of oleo-resin are observed.

In the defatted preparation, several characteristic elements are
easily observed.

It is remarkable that text books — even those published during the
last year — ^consistently draw attention to the absence of starch.
This statement appears to be due to an old statement of Dr. Hassel's
and has been faithfully reproduced ever since.

It is not, however, a fact. Very small starch grains are to be found
in all cayenne pepper. The author has powdered numerous species
of pods, and defatted the powder. The specimens on staining with
weak iodine show fairly numerous very small starch grains embedded
in the cells. A comparison of such a preparation of a commercial
sample, with a standard sample, will at once indicate whether any
added starchy matter is present.

Wallis (" Pharm. Journ." [4], 15, 3) thus summarizes the micro-
scopic characters of the three species of capsicum met with in com-
merce : —

C. Minimum.

C. Annum.

Japanese Chillies.

Thick and straight-
walled rectangular cells
with few pits ; often ar-
ranged in groups of 5
to 7 in a row and with
a uniformly striated
cuticle. Size of cells,
25 jti to 60 /x in either
direction.

Irregular polygonal
cells with evenly
thickened walls, tra-
versed by numerous
well-marked simple pits.
The cuticle shows stri-
ated ridges. Size of
cells, 50 fi to 100 fji long,
and 25 /j. wide.

Cells with strongly
thickened walls and a
radiated lumen. The
pits only rarely pene-
trate the whole thick-
ness of the wall. No
visible striation. Size
of cells, 30 /x to 80 jii long,
and 15 ;u to 45 /x wide.

Delicate thin-walJed
cellulose cells.

Several layers of
cuticularized collen-
chymatous cells, having
a rounded outline and
very few pits.

A singular layer of
regular ■ polygonal cells
with cuticularized fairly
thick walls, traversed
by numerous pits, which
gave them a beaded
appearance.

212

FOOD AND DEUGS.

Note. — For the detection of cayenne in ginger preparations, see
under ginger, p. 198. It is true that capsaicin (Thresh) is a phenolic
compound, but the resinous pungent matter is not of this nature, hence
the raison d'etre of the test there given.

eyfuL

Fig. 24. — Powdered capsicum, e^id, endosperm; ep, epidermis of same; ep. ca
upper epidermis of calyx ; ep. p, outer epidermis of pericarp ; ep^s, epidermis-
from fiat surface of seed ; ep.^s, epidermis from edge of seed ; ep>;iS, epidermis
of seed, side view ; cp^s, isolated epidermal cell of seed coat ; /, sclerenchy-
matous fibres; f.v, fibro- vascular bundle; ep.-st, epidermis of stalk; o, oil;.
par, parenchyma of inner epidermis of pericarp ; pi- P^'''^ parenchyma of
pericarp ; scl^, sclerenchyma of inner epidermis of pericarp, seen from above ;
scl^, the same, side view x 108. (Wallis.)

(By permission of the Editor of the "Pharmaceutical Journal".)

Closely allied to cayenne pepper is the so-called Paprika and also*
the spice known as Pimento.

MUSTARD.

213

Paprika is the ground fruit of Capsicum a7inum grown chiefly in
Hungary, and has a mildly pungent taste, whilst Pimento, which is
probably the product of an entirely different plant, is a nearly "taste-
less " red pepper, produced in Spain and used for garnishing — such
as for stuffing olives or giving a colour to various preserves, and also for
feeding canaries on, in order to impart a characteristic orange colour
to their plumage.

The principal adulterant of these " peppers " is olive or nut oil
added principally to brighten their colours.

Analyses op Papeika (Brooks).

-S^

§ X

0
'o

Per

cent

Hungarian (6 samples)

Minimum (whole pods) .

7-09

017

7-42

19-56

15-10

1406

5-63

4-67

Maximum „

8-14

1-25

11-99

21-40

19-83

17-44

7-06

5-68

0-22

Spanish (2 samples)

Minimum (whole pods) .

7-10

118

10-39

16-52

15-37

14-62

5-24

4-59

005

Maximum „

—

16-87

6-79

5-79

0-06

Hungarian (7 samples)

Minimum (shell alone) .

901

0-44

4-01

21-16

16-66

12-50

5-50

4-85

0-03

Maximum ,, . .

9-76

110

6-69

24-52

23-61

1587

6-90

6-10

008

Spanish (9 samples)

Minimum (shell alone) .

5-75

0-51

4-48

19-96

10-15

11-64

6-20

4-45

0-00

Maximum ,, . .

8-95

210

8-95

15-75

14-06

7-68

6-68

0-35

Hungarian (7 samples)

Minimum (seeds and placentae)

4-05

0-95

17-66

17-36

17-29

16-56

3-06

1-72

0-05

Maximum ,,

4-56

1-90

22-34

18-16

20-11

21-19

4-93

3-72

0-09

Spanish (4 samples)

Minimum (seeds and placentae)

3-63

1-56

18-99

16-12

19-48

15-50

3-41

2-23

004

Maximum „

4-33

2-25

19-80

24-01

16-25

5-20

435

0-11

Hungarian (7 samples)

Minimum (stems) .

2-83

0-27

1-38

19-86

14-37

10-03

6-93

0-30

Maximum „ i

8-80

0-78

2-39

29-94

18-00

12-25

9-28

0-61

Spanish (2 samples)

Minimum (stems)

3-15

0-29

0-98

29-99

11-56

15-50

1309

0-26

Maximum „ . . .

—

Added oil will be indicated by a high non-volatile ether extract.

MUSTARD.

The condiment mustard is the ground seed of Sinajns {Brassica)
alba, or Sinajns nigra or preferably, a mixture of the two. On the
continent, and in Asia, the seed of Sinajns juncea, the brown mustard,
is also used. In pharmacy, both white and black mustard seeds are

214 FOOD AND DRUGS.

ofiicial in the British Pharmacopoeia, a mixture of the powdered seeds
of the two plants being the official " Sinapis," used for the preparation
of plasters.

The composition of the white and black mustard seeds is very-
similar, but important differences exist between the two.

The pungency of mustard does not depend on the existence of any
ready formed compound in the seed, but to the decomposition of gluco-
sides in the presence of moisture by a ferment termed myrosin. This
decompositioQ results in the formation of glucose, and a very pungent
essential oil. Both seeds contain a fixed oil known as mustard oil,
together with a considerable amount of albuminous matter and
mucilage, but no starch. They each also contain sinapin thiocyanate
CigHg^NOgCNS, the white mustard containing about twice as much as
the black mustard. Each contains the ferment myrosin, a soluble
enzyme present in many of the cruciferous seeds. The most favourable
temperature for its action on the glucosides is 45° to 50° C. ; at 70° to
75° it is rendered inactive. The action is also inhibited by the presence
of dilute hydrochloric acid.

The Glucosides of Mustard. — The glucoside contained in black or
brown mustard seeds is sinigrin CiQHigNS.2K0g + HgO, or the potassium
salt of myronic acid. That of white mustard is known as sinalbin
CjjQH^gNgSgOi^ + HgO. Sinapis jimcea contains sinigrin or potassium
myronate. Under the influence of the ferment myrosin, the glucosides
split up in the following manner : —

Black Mustaed.

CjoHigKNOioSa = CeHiA + KHSO, + C3H5N . CS

Sinigrin Glucose Mustard oil (essential)

White Mustard.

CjoH^NaSaOig = G^U^^O^ + C.^H^-iiO^ . H2SO4 + C^H^O . SNC

, Sinalbin Glucose Sinapin sulphate Acrinyl thiocyanate

(" white mustard oil ")

The amount of myrosin present in the black seeds is not usually-
sufficient to convert all the glucoside present, whereas the white
seeds contain more myrosin than is necessary to act on the whole of
the glucoside present. Hence, although white mustard seed yields
but a minute quantity of essential oil, black mustard will yield a
higher amount of essential oil when mixed with white mustard. This
explains the fact that the British Pharmacopoeia describes white
mustard as almost inodorous when treated with water, and black
mustard as having a strong, pungent odour under the same circum-
stances, and directs a mixture of the two to be used as an official drug.
The ground mustard of commerce is the farina separated from the
husk by suitable sifting.

The following analyses by Piesse and Stansell ("Analyst," v. 161)
represent the composition of white and black mustards. The essential
oil, it is to be remembered, does not exist already formed in the seed»

MUSTAKD.

215

so that the following figures represent the amount of essential oil
formed in the mustard by hydrolysis : —

White Mustard.

Black Mustard.

Whole Seeds.

Farina.

Whole
Seeds.

Farina.

rer

Per

cent

Moisture .

9-32

8-0

5-78

6-06

8-52

4-35

4-52

5-63

Fatty oil .

25-56

27-51

37-18

35-74

32-55

25-54

36-96

38-02

36-19

Cellulose .

10-52

8-87

3-90

415

9-34

9-01

3-09

2-06

3-26

Sulphur- .

0-99

0-93

1-33

1-22

1-26

1-28

1-50

1-48

1-30

Nitrogen .

4-54

4-49

5-05

4-89

4-25

4-38

4-94

5-01

4-31

Total proteids .

28-37

28-06

31-56

30-56

26-56

26-50

29-81

30-25

26-06

Soluble albumen

and myrosin .

5-24

4-58

7-32

6-67

6-11

5-24

6-46

6-78

614

Volatile oil

0-06

0-08

0-03

0-04

0-03

0-47

1-44

1-50

1-38

Ash .

4-57

4-70

4-22

4-31

4-30

4-98

504

4-84

4-91

HgO soluble ash

0-55

0-75

0-44

0-55

0-33

1-11

1-01

0-98

0-77

Aqueous extract

27-38

26-29

36-31

36-60

33-90

24-22

31-64

32-78

31-41

The following may be taken as the limit values for genuine ground
mustards ; —

Per cent

Moisture 5-0 „ 9-5

Fatty oil (frequently part is extracted, which is regarded as a legitimate

practice, when the amount will fall to 15 to 25 per cent) . 32 to 39

Fibre (this may reach 10 per cent if much of the husk is ground with

the farina) 1'8 ,, 5-5

Nitrogen 6-0 „ 7-5

Ash .... 4-0 „ e-O

Ash insoluble in acid 0-08 „ 0-5

Alcohol extract 19-0 „ 25

"Starch" (diastase method) 0 ,, 2

" Starch " (by HCl Conversion — not true starch) .... 5 „ 12

The ash of mustard seeds has the following composition according
to the same chemists (the analyses were made on ash not free from
organic matter) : —

White Seeds.

Black Seeds.

Per cent

Potash, KgO .

18-88 to 21-29

21-41

Soda, NagO

0-18 „ 0-21

0-35

Lime, CaO

9-34 „ 13-46

13-57

Magnesia MgO

8 17 „ 10-49

10-04

FeA .

1-03 „ 1-18

1-06

SO3 . . .

706 „ 7-16

5-56

CI . . .

Oil „ 012

0-15

P2O5

32-74 „ 3500

37-20

Siliceous matter

2-92 ., 3-07

2-79

Carbon .

12-82 „ 15-14

7-57

216 FOOD AND DEUGS.

The Analysis of Mustard. — The following are the determinations
necessary : (1) moisture, (2) ash, (3) fixed oil, (4) total sulphur.

A microscopic examination is necessary, and tests for special adul-
terants, the determination of the essential oil is often useful, and occa-
sionally other determinations.

Moisture and Ash. — The moisture in a good quality mustard should
not exceed 6 to 7 per cent; nor the ash from 4 to 6 per cent, of which
from 0*3 to 1-2 is soluble in water. An ash of less than 4 per cent is
an almost certain indication of the presence of an organic adulterant
such as starch.

Fixed Oil — Ten grms. of the mustard should be dried at 100° and
extracted in a Soxhlet tube with ether. From 32 to 39 per cent should
be obtained, a lower amount indicating the presence of an adulterant
containing little or no oil.

Total Suljohur. — Genuine mustard contains from I'i per cent to
1-6 per cent of sulphur (usually about 1*3 per cent), as determined by
heating the mustard with five times its weight of fuming nitric acid
until completely oxidized and then precipitating the sulphates formed
with barium chloride. Or the oxidation may be carried out by means
of boiling with very strong alkaline permanganate of potassium.

The Detection of Adulterants. — Wheat flour or starch is added to
mustard, sometimes with the idea of making it keep better. Such
admixtures, however, must be disclosed or the sale of the mixed article
constitutes an otfence under the Food and Drugs Act. The detection
of starch which would first be found by a microscopic examination, is
simple, sines mustard contains practically no starch. A gram of the
sample should be boiled with water, and on cooling, a solution of
iodine is added gradually, but not in too great excess. The production
of a blue or blue-green colour is proof of the presence of added starch.
The nature of the starch can only be decided by the results of the
microscopic examination. If no adulterant but starch be present, its
approximate amount may be deduced by the shortage in fixed oil, which
averages 35 per cent in pure mustard. But as some of the mustard
oil is often expressed before the mustard is prepared, this will not
give necessarily reliable results. The most approximate method for
the quantitative determination of the starch is to exhaust the sample
first with ether and then with 60 per cent alcohol. The starch is now
converted in the usual way by dilute acid, into sugar and estimated
by reduction of Fehling's solution. Wheat flour may be taken as con-
taining 72 per cent of starch.

Mineral adulterants are now rare, and are at once revealed by the
ash determination. Turmeric is added, especially to mustards which
have already been reduced with starch and so rendered too pale in
<jolour to be attractive. The characteristic odour of turmeric is sufficient
to prevent much being used, and in France it is regarded as a legiti-
mate addition to mustard.

Turmeric is detected microscopically, and as it contains starch an
iodine reaction is obtained, which can be well observed under the
microscope. To verify the presence of this adulterant, 5 grms. should
be extracted with methylated spirit and the extract concentrated to

MUSTARD. ^^^r 217

about 1 c.c, and a piece of iQUer paper moistened with it and dried.
The paper so prepared is treated with a few drops of concentrated
boric acid solution, and dried. The characteristic reddish colour will
result, which turns green to purple on moistening the spot with
{blkali.

It is said that the fluorescence of turmeric colouring matter enables
one to detect a very small quantity. If 1 gi-m. be shaken for some
time with warm castor oil and filtered, the oil will have a distinct
green fluorescence if turmeric be present.

Martius yellow (dinitro-a-naphtholate of calcium) has been found
by Waller and Martin (" Analyst," ix. 166). It is detected by shaking
the mustard for several minutes with cold 95 per cent alcohol and
filtering. The filtrate will be of a light yellow colour, and if it is
evaporated to dryness and the residue is taken up with water, the
aqueous solution will dye wool a light yellow colour. This, of course,
is true of other yellow coal-tar dyes. The yellow solution is de-
colorized by hydrochloric acid, a yellow precipitate being formed.

Cayenne pepper has been found in mustard, especially in that which
is adulterated with starch, as it is added to impart pungency to the
•diluted mustard. It can be detected by boiling 1 grm, of the mustard
with alcohol and drying the extracted matter. The taste of the extract
at once reveals the presence of capsicum, and if the residue be burned
the acid fumes of capsicum cannot be mistaken. In foreign mustard,
sold ready mixed as a condiment, sugar, tartaric acid, citric acid,
turmeric, vinegar, wine, apple juice and a trace of sodium bisulphite
are to be found.

Microscopic Examination. — Only a few fragments of the seed
coats are to be found in ground mustard. Such as may be present
will — in the case of white mustard — show some large hexagonal cells
filled with stratified mucilage, the centre of each cell appearing as
perforated by funnel-shaped tubes, which seem to terminate on the
surface of the cell. When immersed in water these cells swell up enor-
mously and then rupture. Numerous roundish or polygonal cells
with thickened angles, or long palisade cells, are also characteristic of
the seed coat. The mucilage cells in the seed coat of black mustard is
scarcely stratified at all and is therefore less conspicuous. Nor do
they svvell so much in water. The sample should be defatted, and the
main portion will be found to consist of the fragments of the seeds
proper. Numerous cells will be found containing aleurone grains
which stain yellow with picric acid ; no starch grains are to be found.
The bulk of the powder will be found to consist of small masses of
delicate parenchymatous cells..

Essential Oil of Mustard. — The essential oil of mustard of com-
merce is obtained by distilling the seeds of the black mustard.

It is official in the British Pharmacopoeia, which authority requires
it to have a specific gravity 1-018 to 1-030. It should distil between
147° and 152°, and the first and last distillates should have the same
specific gravity as the original oil. In the formation of the essential
oil by the hydrolysis of the glucoside, the chief product of the reaction
is allyl iso-thiocyanate (allyl thiocarbimide), a pungent and disagree-

218

FOOD AND DRUGS.

able liquid. A small quantity of the normal allyl thiocyanate is also
formed, together with traces of cyanallyl and carbon disulphide. The
oil is a pungent and unpleasantly smelling liquid of specific gravity
1-015 to 1*030, and optically inactive. It boils almost completely be-
tween 148° and 155° and has a refractive index from 1-526 to 1-530.
As mustard oil consists almost entirely of allyl iso-thiocyanate, and

Fig. 25. — Powdered mustard.

the latter body is easily prepared artificially, there is an artificial oil
on the market. This is made by distilling allyl iodide or bromide
with alcoholic solution of potassium thiocyanate — a molecular re-
arrangement to the iso-thiocyanic radicle taking place. Thus —

ON . SK + C3H5I = CS . N . C3H5 + KI.

Pure allyl iso-thiocyanate is a liquid of specific gravity 1 017 at 10°,
boiling at 151'.

The artificial oil, however, is not far different in price from the
natural oil, and is not official in any pharmacopoeia. The amount of
allyl iso-thiocyanaie present in the oil can be approximately estimated
by heating a known quantity with an alcoholic solution of ammonia,
when allyl-thio-urea is formed.

For the determination of the amount of allyl iso-thiocyanate pre-
sent, which should not be less than 9'2-5 per cent in any oil; or of the

MUSTARD. 219

amount of mustard oil in spirituous preparations, or in the mustard
itself one of the following processes should be used : —

Three grms. of the oil and 3 grms. of alcohol are shaken in a flask
with 6 grms. of a 10 per cent solution of ammonia. It should become
clear after standing for a few hours, or rapidly if warmed to 50° C,
and deposit crystals of allyl-thio-urea (thiosinamine).

^^ NH.C3H,

To determine the quantity, decant the mother liquor and evaporate
it slowly on the water bath m a tared capsule, adding fresh portions,
slowly as the smell of ammonia disappears. Then add the crystals
from the flask to those in the capsule, rinsing the flask with a little
alcohol,- and heat the capsule on the water bath to a constant weight.
Three grms. of oil should yield between 3-25 and 3-5 grms. of thiosin-
amine, which should melt at 70° to 74°. One hundred and si-xteen parts
of thiosinamine, correspond to 99 parts of allyl iso-thiocyanate. Gada-
mer (" i\rch. Pharm." 1899, pp. 110, 237) recommends the following
process. The mustard oil is dissolved in alcohol to form an exactly 2
per cent solution. Five c.c. (4*2 grms.) of this solution are allowed to
remain with 25 c.c. of decinormal solution of silver nitrate and 5 c.c. of
ammonia for twenty-four hours in a well-stoppered 50 c.c. flask. It is
then made up to 50 c.c. with water and filtered from the precipitated
silver sulphide ; 25 c.c. of the filtrate are mixed with 4 c.c. of nitric acid
and a few drops of ferric sulphate solution, and titrated with deci-
normal ammonium thiocyanate solution, until the characteristic red
colour of the ferric thiocyanate appears. From 4*1 to 4*5 c.c. of the
solution (corresponding to 1-85 to 2'0 per cent of allyl thiocyanate in
the alcoholic solution) should be required.

Grutzner converts the thiocyanate into thiosinamine, which he oxi-
dizes with peroxide of sodium, and weighs the resultant sulphuric
acid as barium- sulphate. From the figures obtained in his analyses,
Grutzner concludes that a mustard oil containing 28*60 per cent of
sulphur (equivalent to 88*48 per cent of iso-thiocyanate) may be re-
garded as pure. P. Roeser proposes to modify Gadamer's method
for the determination of the sulphur content of oil of mustard so as to
determine the excess of silver nitrate in an ammoniacal solution, in-
stead of an acid solution according to Volhard's method, as is usually
done. Koeser operates in the following manner : when the conver-
sion of the thiosinamine with silver nitrate, after twenty-four hours
standing, has taken place, an excess of one-tenth normal solution of
potassium cyanide is added to 50 c.c. of the clear filtrate, and the
excess of potassium cyanide titrated back with one-tenth normal solu-
tion of silver nitrate, in the presence of a few drops of a weak am-
moniacal solution (5 per cent) of potassium iodide. Schimmel & Co.
prefer the following method, About 5 grms. of a solution of 1 grm.
mustard oil in 49 grms. alcohol are mixed in a measuring flask of 100
c.c. capacity with 50 c.c. decinormal solution of silver nitrate and 10
c.c. of ammonia (dj,^° 0-960) ; the flask is then closed, and with
frequent agitation left standing for twenty-four hours with the light

220 FOOD AND DEUGS.

excluded. The flask is then placed for half an hour in water at 80°,
•during which time it is again repeatedly shaken, then cooled down
to the temperature of the room, filled up with water to the mark,
shaken up, and filtered. Fifty c.c. of the filtrate are titrated with one-
tenth normal solution of ammonium thiocyanate, after adding 6 c.c.
nitric acid (dj5° 1-153) and a small quantity of solution of iron alum,
until a change of colour from white to red takes place. In order to
ascertain the whole quantity of silver solution which has entered into
reaction, the number of c.c. of ammonium thiocyanate solution used
is doubled, and the product subtracted from 50. The percentage of
allyl iso-thiocyanate in the mustard oil is obtained by means of the
following formula : —

CSNC3H, = ?i!§75j^

A = number of c.c. of decinormal solution of silver nitrate used,
6 = spirit of mustard used, in grammes. Mustard oil determinations
carried out by them in the manner described, showed in the case of
natural oil a content of about 94 per cent allyl iso-thiocyanate, whilst
in artificial oil about 98 per cent was found.

In the case of the mustard itself Forster recommends the following
process (" Journ. Chem. Soc." 54, 1350) : —

Twenty-five grms, of the powder is made into a thin paste with
water and allowed to stand for an hour, and then the essential oil dis-
tilled by steam through a condenser into a 250 c.c. flask, containing
50 c.c. of alcohol saturated with ammonia, the end of the condenser
•dipping under the surface of the fluid.

When about 150 c.c. has distilled, the flask is allowed to stand for
twelve hours, and then heated to boiling-point, and freshly prepared
mercuric oxide (made by decomposing a 5 per cent solution of mercuric
chloride with caustic potash and boiling the mixture) is added, to com-
bine with all the sulphur present. The mixture is again boiled and a
little potassium cyanide added before it is quite cold. The mercuric
sulphide is collected on a tared filter, dried and weighed. The amount
of HgS multiplied by 0'4266 gives the amount of mustard oil present.
Schlicht's process is very satisfactory also. He proposes (" Zeit. Anal.
Ghem." xxx. 661) to distil the mustard oil from the mustard in a
current of steam and then proceeds as follows : —

To the aqueous distillate containing the mustard oil are added 20
parts of potassium permanganate and 5 parts of caustic potash or
caustic soda (which reagents must be free from sulphate) for each part
of mustard oil supposed to be present. The mixture is shaken for
some time in a closed flask, and finally heated nearly to boiling. The
whole of the sulphur is thus oxidized to sulphuric acid. After cooling
the solution somewhat, 5 c.c. of alcohol should be added for every grm.
of permanganate previously used. The whole of the manganese is
thus precipitated. The mixture is then completely cooled, largely
diluted, made up to a known volume, and filtered. A measured
portion of the filtrate is slightly acidified with hydrochloric acid, and
treated with a solution of iodine in potassium iodide until a feeble

MUSTAED. 221

yellow colour remains even after warming. This reoxidizes any
sulphurous acid which may have been produced by reduction by
means of aldehyde, and also removes the aldehyde itself.^ The sul-
phuric acid is now determined by precipitation with barium chloride,
and the weight of barium sulphate multiplied by 0-42492. The product,
gives the amount of mustard oil.

Piesse and Stansell determine the essential oil formed in mustard
from the glucosides present, in the following manner, which according
to Sutton forms an approximate method for the estimation of the pro-
portion of brown mustard in a mixture of the two kinds.

Twenty-five grms. of the crushed brown seeds are mixed with
about 6 grms. of crushed white seeds, and 300 c.c. of cold water added.
The mixture is allowed to stand in a 700 c.c. flask for five to six hours
at ordinary temperature. The contents of the flask are then distilled
and the distillate collected in a flask containing 30 c.c. of strong
ammonia. Usually about 50 c.c. of distillate are collected, but distil-
lation should proceed till no more oil drops are carried over. When
combination is complete, ihe distillate is evaporated and the thiosin-
amine dried at 100° and weighed. The weight x 0-853 gives the^
amount of mustard oil — or if multiplied by 3-578 the amount of potas-
sium myronate from which it was derived. As the average amount of
potassium myronate is fairly constant in brown mustard seeds — about
5'15 per cent, yielding 1-33 per cent of thiosinamine so that the weight
of the latter multiplied by 75 will give the approximate amount of
brown mustard present in the sample.

Jorgensen recommends (" Analyst," xxxiv. 489) the estimation of the
amount of nitrogen in the "thiosinamine" obtained by distilling the
essential oil into a solution of ammonia and evaporating the solution^
He states that the following percentages of nitrogen are obtained from
the oils of various species of brassica : —

Per cent

Brassica (sinapis) nigra 2414

„ dichotoma ......... 20*52

,, glauca . . 20-34

„ ramosa 18-36

,, napus 21-21

„ rapa . 20-83

He recommends adding a little powdered white mustard to provide
the ferment and then distilling the essential oil.

White mustard seeds, from Sinajns alba, contain the glucoside
sinalbin, Caj^H^^N.^S.^O^g, which on decomposition in the same manner
as the glucoside of black mustard, yields glucose, sinapine sulphate,
and the evil-smelling oil, acrinyl-isothiocyanate (2?-hydroxy-benzyl-iso-
thiocyanate).

Acrinyl iso-thiocyanate, or "white mustard oil," is a yellowish oily
liquid, of pungent odour and unpleasant hot taste. It is prepared
synthetically by treating j;-hydroxy-benzylamine with carbon disul-
phide, and the resulting compound with mercuric chloride.

^ The reduction of sulphuric acid in dilute alkaline solution by aldehyde is-
highly improbable. Addition of bromine-water would do instantaneously and cer-
tainly what Schlicht effects by iodized potassium iodide.

222

FOOD AND DKUGS.

Fixed Oil of Mustard. — If it is coasidered necessary to examine the
fixed oil of mustard, it should have the following characters : —

Sinapls Nigra. '

Sinapis Alba.

Sinapis Juncea.

Specific gravity at 15°
Saponification value
Iodine value
Refractive index at 15°
Melting-point of fatty acids
Molecular weight of „ „

0-915 to 0-920
173 „ 176
96 „ 106
1-4672
15° to 17°
300

0-9125 to 0-916
170 „ 175
92 „ 98

1-4735
15° to 16°
302

0-915 to 0-922
172 „ 182
102 „ 110
1-4699
15° to 17°
296

CLOVES.

Cloves are the dried flower buds of Eugenia caryophyllata, and are
almost invariably sold whole, and are largely used as a spice. They
•are official under the name " Caryophyllum " in the British Pharma-
copoeia, which describes them as follows : —

" About fchs of an inch long, each consisting of a dark brown,
•wrinkled, suocylindrical, somewhat angular calyx tube, which tapers
'below and is surmounted by four thick, rigid, patent teeth, between
which are four paler imbricated petals enclosing numerous stamens and
a single style. Odour strong, fragrant, and spicy ; taste very pungent
and aromatic. Cloves should emit oil when indented with the finger-
nail. Incinerated they should not yield more than 7 per cent of
^sh."

As cloves are particularly characteristic in appearance, and are
.almost invariably sold whole, adulteration is not common. Admixture
with the fruits of the clove ("mother cloves") has been observed, but
this is not now practised. The fruit has the shape of a very small
•olive, and is crowned with the four teeth of the calyx and the remains
of the style. Clove stalks are also mixed with the clove buds at times.
These are distinguished by the absence of the stamens, style, etc., which
renders their detection quite simple by the naked. eye. Apart from such
^admixture, the only other cases noticed by the author are the addition of
partially spent cloves, from which a large amount of the essential oil
is distilled, and the. admixture with water-damaged cloves. In the
former case, it is a common thing for some distillers to dry the cloves
from which they have distilled the greater part of the essential oil, and
sell them to second-rate spice dealers. This fraud is practised to a
much larger extent than would be expected. The author is personally
•acquainted with a distillery which puts about 40 tons of such partially
spent cloves on the market por annum. The greater part of these
find their way, however, to the continent. In the case of water-
damaged cloves, but little exception is to be taken as it is little more
than a matter of appearance. It is, however, often a matter of import-
ance to decide whether such damaged cloves have been exposed to
river or sea water, since the conditions of insurance policies differ-
entiate between the two. The sea-damaged cloves, on soaking for a

GLOVES.

223

short time in distilled water, give a copious precipitate with silver
nitrate.

Cloves contain from 12 to 20 per cent of essential oil, rarely below
15 per cent. Clove stems, however, contain only about 6 per cent.

The following represent the average compositions of cloves and of
clove stems : —

Observers.

Parry

Eichardson

MoGill

Water.

Per cent

8-6

2-9 to 10-7

6 „ 11-8

Ash.

Per cent
12-4
5-5 to 13
5 „ 7-0

Essential
Oil.

Per cent

16-5

10-2 to 18-9

9-2 „ 19-6

Fixed Ether
Extract.

Per cent

8-8

7-1 to 10-2

0-9 „ 10-2

Fibre.

Per cent

10-1
6-2 to 9-7

Nitrogen.

Per cent

0-9

0-76 to 1-12

Tannin.

Per cent

17-5

11-7 to 22-1

Winton, Ogden and Mitchell (" Conn. Exp. Sta. Eep." 1898, 206),
give the following fuller analyses of eight samples of genuine cloves : —

Maximum
Minimum
Mean
Stems

H3O

Ash.

Ether Extract. ^^

Total.

HCl. Sol.

Insol. in HCl.

Volatile.

Non-volatile.

Per

cent
8-26
7-03

7-81
8-74

Per
cent
6-22
5-28
5-92
7-99

Per
cent
3-75
3-25
3-58
4-26

Per
cent
0-13
0-00
0-06
0-60

Per
cent
20-53
17-82
19-18
5-00

Per Per
cent cent
6-67 15-58
6-24 33-99
6-49 14-87
3-83 6-79

Maximum
Minimum
Mean
Stems

"Starch" by
HCl Conversion.

Starch by
Diastase Method.

Fibre.

Nitrogen.

Tann'c Acid.

Per cent

9-63

8-19

8-99

1413

Per cent
3-15
2-08
2-74
2-17

Per cent

9-02

7-06

8-10

18-71

'Per cent
1-13
0-94
0-99
0-94

Per cent
20-54
16-25
18-19
18-79

Mineral Matter. — Any excess over 7 per cent of ash is due either
to adulteration or to the presence of too much dirt. The average
figures are from 5'5 per cent to 6-5 per cent, of which 60 per cent is
soluble in water ; the ash insoluble in HCl should not exceed 0*1 per
cent. The ash of clove stems is almost 8 per cent so that a consider-
able proportion of stems might be present without the ash limit being
exceeded.

Extractives. — The ether extract, after driving off the essential oil

224 FOOD AND DRUGS.

will always fall between the values 5*5 per cent and 7 per cent in
normal cloves. The alcoholic extract, similarly dried varies from 14
to 16 per cent.

Tannin. — Cloves contain much tannic acid. If this be determined
as described on page 192, it should not be materially less than ]6
per cent (this is probably a modification of ordinary tannic acid).

The Essential Oil. — This can only be determined properly on a
large sample, with a proper experimental still. At least 1000 grms.
should be exhausted by steam distillation and the essential oil collected
and measured. Its specific gravity may be taken as 1*050. Any re-
sult lower than 15 per cent is strongly indicative of the presence of
partially exhausted cloves.

Cripps and Brown ("Analyst," xxxiv. 518) recommend the deter-
mination of the essential oil in this and other spices, by first estimating
the total amount of volatile matter, and then determining the moisture
by the amount of acetylene liberated from calcium carbide, returning
the difference as essential oil.

They use 0 5 grm. of the spice in fine powder, in a stout tube 5
in. long and | in. in diameter. Dried sand is added to the depth of
about J in. and then calcium carbide to within 1^ in. of the mouth of
the tube. This is connected with a calcium chloride tube to prevent
moisture from outside reaching the carbide, and then with a nitrometer
tube in strong brine. The tube with the spice, etc., is immersed in a
brine bath and heat is applied until no increase in the volume of the
gas in the nitrometer takes place in five minutes. The gas is then
measured after adjusting the temperature and pressure, and the number
of c.c. multiplied by 0-001725 gives the weight of water in grammes
in the amount of the sample used. In this way the above-named
chemists found the following amounts of essential oil in samples
known to be pure : —

Cloves 12-75 to 17-90

Allspice 1-64 „ 3-67

Caraway 2-49 „ 5-24

Whole mace 6-25 „ 10-80

Ground mace 2-86(?) „ 7-15

Ginger 2-24 „ 8-48

Fennel 1-97 „ 4-00

The presence of exhausted cloves, then, is indicated by a low es-
sential oil yield, a low soluble ash, and a low tannin content.

Microscopic Examination. — Under the microscope cloves in coarse
powder will show numerous large oil cavities — often broken in the
powdering, fibrovascular bundles embedded in parenchymatous tissue,
with spiral and other vessels. Crystals of calcium oxalate are plentiful
but no, or practically no, starch can be found. The presence of ex-
hausted cloves will be indicated by the disrupted nature of the tissues.
Clove stems are easily detected by the presence of well-marked " stone
cells," very thick- walled large cells, penetrated by radial pores (see
illustration). Starch is also to be found in small quantity. " Mother
cloves " or fruit, show similar " stone " cells, except that they are
usually much larger — often ten times as long as they are wide.

CLOVES.

225

Oil of Cloves. — Essential oil of cloves is largely used in flavouring
and is usually sold, diluted with alcohol, under the name of " essence
of cloves ". For its examination the alcohol should be removed in a
current of warm air, and the oil then examined. Oil of cloves is
official in the British Pharmacopoeia, that authority requiring it to
have a specific gravity not below 1-050.

The principal constituent of clove oil is eugenol, besides which there

Fig. 26. — Section of clove buds, o, oil receptacles ; a, parenchyma ; 6, spiral vessels.

Pig. 27a. — Clove stems. Pig. 27b. — Mother cloves.

G, scalariform vessels; B, thickened fibrei ; St, stone cells; Ep, epidermis.

are present, the sesquiterpene caryophyllene, esters of eugenol, methyl
alcohol, furfurol, amyl-methyl-ketoue and traces of other bodies.
Pure clove oil should have the following characters : —

Specific gravity
Refractive index at 20° .
Optical rotation
Eugenol (by absorption .
,, (Thorn's method)

Per cent
1-048 to 1-066
1-5280 „ 1-5320
-0°20'„ -1°35'
80 at least, usually 85 to 93
76 to 88

The method for the determination of the eugenol suggested by
VOL. I. 15

226 FOOD AND DRUGS.

Urnney gives useful approximate results, but is subject to a not in-
considerable error. This consists in shaking a known weight of the
oil with a 10 per cent aqueous solution of potassium hydroxide in a
Hirschsohn flask, and allowing the unabsorbed portion to rise into the
graduated neck. This is measured and its volume corrected by multi-
plying it by -908 — the specific gravity of the sesquiterpene — and the
unabsorbed portion returned as caryophyllene, the difference being
reckoned as eugenol. The globules of iincombined hydrocarbons have
a great tendency to stick round the top of the flask and require some
" coaxing " to rise and agglomerate in the neck of the flask. Heat will
accelerate and assist this however. But the source of error lies in
the fact that the aqueous solution of potash and potassium eugenate
dissolves some of the sesquiterpene, which is thus reckoned as eugenol,
and a too high result is obtained. The process proposed by Thorn, al-
though more tedious, gives more exact results. This depends on the
conversion of the eugenol into benzoyl-eugenol. The following are
the details, which should be carefully observed in order to secure ac-
curate results : —

Five grms. of the oil are heated on a water bath with 20 c.c. of a
15 per cent solution of caustic soda for thirty minutes.

After allowing the hydrocarbons to separate, the eugenol soda
solution is run off", and the hydrocarbons washed with dilute soda
solution twice, the washings being added to the original soda solution.
The reaction is now effected at water- bath temperature with 6 grms.
of benzoyl chloride. The whole is allowed to cool, and the crystalline
mass is transferred to a beaker with 55 c.c. of water. It is heated in
order to melt the crystals, and well agitated with the water to wash
the benzoyl eugenol. This washing is repeated twice. The crystal-
line mass is then transferred to a beaker with 25 c.c. of 90 per cent
alcohol, and warmed till complete solution takes place. The solution
is allowed to stand till the bulk of the crystals have separated out, and
is cooled to 17° and filtered through a paper 9 centimetres in diameter,
previously dried and tared. The filtrate measures about 20 c.c. and
the crystals are washed with more alcohol until it measures 25 c.c.
The paper and crystals are then dried in a weighing glass and weighed,
the temperature of drying being not more than 101° C. The solubility
allowance for 25 c.c. of alcohol is 0 55 grm. The total eugenol is cal-
culated from the formula.

^^ a -f- 0-55

where P is the percentage, a the weight of benzoyl-eugenol obtained,
and b is the weight of oil of cloves used.

Verley and Bolsing propose the following method : It depends on
the fact that acetic and other anhydrides react with phenols in excess
of pyridine. Eugenol reacts readily forming eugenyl acetate and
acetic acid, the latter combining w^ith pyridine to form pyridine acet-
ate. This compound reacts towards indicators such as phenol-
phthalein in the same way as acetic acid, and therefore a titration is
possible.

ALLSPICE.

227

Verley and Bolsing use from 1 to 2 grms. of the oil, which is placed
in a 200 c.c. flask, and 25 c.c. of a mixture of acetic anhydride (15
parts) and pyridine (100 parts). The mixture is heated for thirty
minutes on a water bath, the liquid cooled, and 25 c.c. of water added.
The mixture is well-shaken and titrated with normal potash, using
phenolphthalein as indicator. A blank experiment is carried out with-
out the eugeiiol, and the difference between the titration figures in c.c.
of normal alkali, multiplied by 0*582, gives the amount of eugenol in
the sample taken.

ALLSPICE.

Allspice or Pimento is the dried, fully grown unripe fruit of Pimenta
officinalis, and as such is official in the Jiritish Pharmacopoeia.

The fruits consist of dark reddish-brown, nearly globular two-celled
fruits, about 5 to 8 millimetres in diameter. The pericarp is rough
and brittle and covered by the remains of a four-toothed calyx in the
form of a raised ring, surrounding the remains of the style. Each
cell contains a single brownish-black reniform seed. Allspice owes its
characteristic flavour to from 3 to 6'5 per cent of an essential oil. The
spice is generally sold whole, and the author has never met with an
adulterated sample. The genuine spice has the following char-
acters : —

Per cent

Total ash

Ash soluble in water

Ash insoluble in HCl

Alcoholic extract .

Fixed ether extract

Essential oil

Fibre .

Nitrogen

Tannin

4-0 to 5-5

at least 50 of that

up to 0-2

10 to 13

4-0

, 6-2

3-0

, 6-5

, 19

0-65

, 0-90

, 13

Microscopic Examiiiation. — Under the microscope powdered allspice
is characterised by stone cells, similar to those in cinnamon, etc., a num-
ber of large oval cells of a port wine colour, parenchymatous cells con-
taining may small starch cells ; and cells containing essential oil.
Foreign starch grains are easily detectable.

Essential Oil of Pimento. — The principal constituent of the essential
oil is eugenol, the remainder consisting chiefly of a sesquiterpene.

The specific gravity of the oil is, to an extent, an indication of the
amount of eugeno! present, and should vary between 1*040 and 1*055.
The British Pharmacopoeia, in which this oil is official, states that the
specific gravity should not fall below 1-040. The oil is laevorotatory,
but never exceeds - 4^ usually about - 2°. It is easily soluble in 90
per cent alcohol and in twice its volume of 70 per cent alcohol.
Eugenol boils at 247'', consequently the fraction 245° to 250° should
be considerable— in genuine oils not below 60 per cent, usually 70
per cent or over. The amount of eugenol, as estimated by Them's

228

FOOD AND DRUGS.

process (see Oil of Cloves), should not be less than 65 per cent ; and
the residue not absorbed by caustic potash solution should not exceed
25 per cent.

Fig.

28. — Powdered allspice.
K, crystals ;

Ep, epidermal cells ; P. brown parenchyma ;
Po, oil cells ; St, stone cells.

CINNAMON.

This spice is the dried inner bark of shoots from the truncated stocks
of Cinnamomum zeylanicum, and is official in the British Pharmacopoeia.
It is usually sold in quills (the dried rolled bark of the shoots) and is
only adulterated in the form of powder. Cinnamon bark yields from
0-5 per cent to 1 per cent of an essential oil which is largely used for
flavouring purposes.

The following are the characters of genuine cinnamon : —

Per cent

Total ash 3-5 to 5-5 rarely up to 6

Ash soluble in water . . . . 1-6 „ 2-4

„ insoluble in HCl . , . usually under 0-5

Fibre 25 to 33

Alcoholic extract 10 „ 15

Nitrogen 0-5 ,, 0-65

Fixed ether extract 1'4 ,, 1*7

Essential oil 0-8 „ 1-30

The following analyses are those of Winton,
chell :—

Ogden and Mit-

CINNAMON.

229

Moisture

Total ash

Ash soluble in H^O

„ insoluble in HCl
Volatile ether extract
Non-volatile „ „

Alcohol extract
" Starch " by acid conversion

Fibre

Nitrogen ....

The powdered bark of Ginnamomum cassia, the Chinese cassia tree,
is' sometimes used as an adulterant, but can only be detected by its
odour if present in large quantity, or microscopically. The powder
should be fine enough to pass through a very fine sieve, and should be
bleached by immersion in a solution of chlorinated soda. If a specimen
so bleached be examined in glycerine, the characteristic thick -walled

Per cent

7-79 to 10-48

4-16 „

5-99

1-40 „

2-71

0-02 „

0-58

0-72 „

1-62

1-3.5 „

1-68

9-97 „

13-60

16-65 „

220

34-38 „

38-48

0-52 „

0-65

Fig. 29. — Powdered cinnamon bark.

sclerenchymatous cells with radial markings, ordinary parenchymatous
cells, many containing crystals of calcium oxalate (which appear well
marked under a polarizer) and bark fibres will be seen. A preparation
stained with a hot solution of soudan red shows up the characteristic
"secretion cells," elongated cells with suberized walls, which take the

230

FOOD AND DKUGS.

stain very deeply. The starch should be examined, and compared with
that of a standard preparation, and also with that of cassia bark. The
characteristic differences between cinnamon and cassia are that in
the latter the starch cells are rather larger, and the bark fibres are
stouter. But only a comparison with type samples will render these
differences useful. It has been pointed out by Greenish that the differ-
ences between the two barks become slighter as one compares the
lower-grade cinnamons with the best-grade cassias.

-^i s in

Pig. 30.— Cinnamon bark, transverse section. 6, bast fibres; A;, crystals of cal-
cium oxalate ; 7n, medullary rays ; _p&, primary bast fibres (pericyclic fibres) ;
pr, cortical parenchyma ; s, sieve tubes ; sch, secretion cells ; si, sclerenchy-
matous cells, forming an uninterrupted ring x 160. (Moeller.\

Essential Oil of Cinnamon. — This is understood to be the oil dis-
tilled from the bark, although nearly every part of the plant yields an
essential oil. It is distilled in Europe, but a good deal is prepared in
Ceylon and exported westwards.

The majority of that exported, however, is not genuine. Either
the leaves are added to the bark when distilled, or cinnamon leaf
oil is added to the oil after distillation. The important difference be-
tween the two oils is that the bark oil owes its characteristic odour to
the cinnamic aldehyde it contains, whilst the leaf oil contains only
traces of that body ; the chief constituent of the latter oil is eugenoi,
the characteristic phenol of the oils of cloves and pimento.

The pure bark oil has a specific gravity of 0*998 to 1'038.

CINNAMON.

231

Adulteration with the leaf oil or with clove oil increases this figure
The oil is optically inactive, or at most laevorotatory to the extent of
- 1°. The ascertained constituents are the terpene phellandrene,
cinnamic aldehyde, and eugenol. The British Pharmacopoeia, in
which this oil is official, gives the following limits: specific giavity,
1-025 to 1-035: cinnamic aldehyde at least 50 per cent; should not
yield a decided blue coloration with ferric chloride solution. ^ It is

Fig. 81. — Cassia bark, transverse section, b, bast fibres; K, sclerenchymatous
cork cells: m, medullary rays: pb, primary bast fibres (pericyclic fibres);
pr, cortical, parenchyma, with sclerenchymatous cells ; s, sieve tubes ; sch,
secretion cell ; st, sclerenchymatous cells, forming an interrupted ring x'160.
(Moller.)

very rare, however, to find an oil with only 50 per cent of aldehyde
present. Adulteration with much leaf oil causes the characteristic
blue colour given by eugenol to .be developed when a few drops of a
solution of ferric chloride are added to a solution of the oil in alcohol.
The amount of eugenol estimated as described under oil of cloves, should
not exceed 8 per cent. More than this indicates the presence of leaf
oil.

The most important method of examination is the determination of
the percentage of cinnamic aldehyde. In this process the following
details should be observed. Ten c.c. of the oil are run into a Hirsch-
sohn flask (capacity about 100 to 150 c.c, with a neck about 5 inches

232 FOOD AND DRUGS.

long and J inch in dianieter, graduated in y^gth c.c). The flask is
then filled about | full with a 30 per cent solution of sodium bi-
sulphite, and the whole well shaken. The flask is then placed
on the water bath for several hours with occasional shaking, until
the precipitated compound of the aldehyde and bisulphite is com-
pletely dissolved, and only a clear oil floats on the surface. Bisulphite
solution is then carefully poured in until the oil is driven up into the
neck, and when it has attained the temperature at which the oil was
measured, the amount is read off. This gives the percentage of non-
aldehydic constituents, the difference being returned as cinnamic alde-
hyde. Pure oils should not give less than 60 per cent of aldehyde,
the best oils yielding 65 to 70 per cent, or occasionally even higher.
Oils with much higher cinnamic aldehyde vajue usually contain the
synthetic aldehyde. Strictly speaking, these percentages are by volume,
but the errors of reading the result, and those due to solubility of the
non-aldehydes in the aqueous liquid render any correction for the
specific gravity of the constituents unnecessary in practice. Care must
be taken that every particle of the aldehyde compound is dissolved, as
otherwise the reading of the oily layer will be obscured, and a serious
error may be introduced.

Hanus has recently published a new method for the determination
of cinnamic aldehyde in cassia and cinnamon oils (" Pharm. Central."
1904, 37) depending on the combination of the aldehyde with semi-
oxamazide. Ten grms. of finely powdered hydrazine sulphate are
dissolved in a solution of 9 grms. of caustic soda in 100 c.c. of water
and the alkaline sulphate produced is precipitated by the addition of
100 c.c. of alcohol. After filtration the solution is warmed, 9 grms. of
oxamethane are added in small portions, the whole warmed for half
an hour and allowed to cool. The azide separates in crystalline tables
and these are separated and recrystallized. To estimate the aldehyde
by means of this reagent, a small quantity, not more than 0*2 grm. of
the oil is well shaken in 85 c.c. of water, and about 0*35 grm of semi-
oxamazide in 15 c.c. of hot water is added and the whole well shaken.
After five or ten minutes the compound begins to be precipitated, and
after standing twenty-four hours can be collected on a Gooch filter,
washed with cold water, and dried for a few minutes at 105°. The
amount of the precipitate is multiplied by 0'6083 to obtain the amount
of aldehyde. The constitution of the semi-oxamazone of cinnamic
aldehyde is NH2 . CO . GO . NH . N : CH . CH : CH. C.H^.

NUTMEGS.

As, in the case of mace, the nutmeg of commerce is derived from
Myristica fragrans, of which it is the dried seed, divested of its testa.
As a rule nutmegs are sold in the whole condition, powdered nutmegs
being rarely seen in retail shops. They are rarely adulterated except
(1) by admixture with other species of nutmegs, (2) when worm-eaten
nuts accumulate, the holes are sometimes filled with extraneous matter
and the nut is coated with lime, (3) by the addition of partially ex-
hausted nuts.

NUTMEGS. ^BRr 233

(1) The genuine nutmeg can be distinguished from the usual ad-
mixture the so-called " long " nutmeg, or Macassar nutmeg (M. argentea)
by its appearance. The true nutmeg resembles an olive in shape,
^whereas the Macassar nutmeg more closely resembles an enlarged date
stone. The latter are far less fragrant than the former.

(2) As some organic powder is used for this purpose, and that
only to a minute extent — since the filling up of the tiny holes is done
with a view of passing otf interior nuts as of better quality, this* can
hardly be detected except by carefully cutting and probing the nutmeg.
The custom of liming nutmegs originated for the purpose of perserva-
tion from the attacks of insects ; to-day it is often practised solely for
the purpose of concealing the inferior quality of low-grade nutmegs.

(3) Occasionally small holes are drilled in nutmegs and some of
the fat extracted by soaking in hot water, or some of the essential oil
driven off The holes are then filled and the nutmegs limed as before.
This can only be detected by a physical examination, and a determina-
tion of the amount of fat present.

According to Vanderplanten, damaged and exhausted powdered
nutmegs are sometimes made up with some medium to cause the
particles to adhere, into factitious nutmegs. These, however, are
easily crushed to powder, especially when heated in water, and on
cutting show no vegetable structure. No difficulty is experienced
in detecting these when compared with a genuine nutmeg. An analy-
sis of such spurious nutmegs gave the following figures : —

Per cent

Moisture 11-09

Asfi 11-34

„ insoluble in HCl 3-90

Ether extract 15-42

Essential oil . . . 1-76

Cellulose 8-44

Genuine nutmegs will show the following characters on analysis,
the estimation of mineral matter and fat being the principal useful
chemical determinations, except when a large quantity is available,
when the essential oil may be estimated : —

Per cent

Moisture 4 to 8

Total ash . 2 „ 4

Ash soluble in HaO 0-8 „ 1-4

„ insoluble in HCl 0 „ 0-15

Fat 32 „ 36

Essential oil 5 „ 15

Alcohol extract . . . " 11 „ 17

Total nitrogen 1 „ 1-5

" Starch " (acid conversion) 15 „ 25

Any considerable reduction in the amount of fat indicates the
presence of exhausted nutmegs.

Microscopic examination. — The uncoloured portion of powdered nut-
meg under the microscope consists of minute angular cells, containing oil,
and sometimes crystals of myristic acid ; aleurone grains are to be found

234

FOOD AND DRUGS.

and numerous small but distinct starch granules, which show a well-
marked central depression round the hilum. In the more> coloured
portions of the powder much brown pigmentary matter is to be found
but no starch and but little oil.

Fig. 32. — Powdered nutmeg.

The following illustrations represent the tissues present in nutmegs
(after Moller).

K = endosperm cells with starch grains ; F = crystals of fatty acids ;
E = cells with aleurone grains and crystals.

Nutmeg butter, which is a commercial article, is a mixture of
about 5 per cent of the essential oil of nutmeg, with 95 per cent of
fixed fat. Nearly all of the published figures for this fat are to be re-
garded with suspicion, as they have usually been obtained on the
butter as found in commerce, and as this contains a variable amount of
essential oil with a very high iodine value, any suggested methods of
differentiating between the fat of Bombay and Banda mace or nutmegs
depending on the iodine value, are useless.

The following figures are given by Spaeth : —

Origin.

M. Ft.

Sap. Value.

Iodine Value.

Refractometer No. at 40°.

Banda

Bombay

Menado

Penang

Macassar

Zanzibar

25° to 26°
31° „ 31-5°

25-5°

26°
25° to 25-5°
25-5°,, 26°

170 to 173
189 „ 191

169
171-8 to 172
171-8 „ 172
169-9 „ 170-5

77-8 to 80-8
50-4 „ 53-5
76-9 „ 77-3
75-6 „ 761
75-6 „ 76-1
76-2 „ 77

76 to 82
48 „ 49
74 „ 74-5

84-5 „ 85
78-5
77-5

An examination of these figures show that it is extremely probable
that with the exception of the Bombay sample, they all contained notable
proportions of essential oil. The author has examined three samples
of pure commercial nutmeg butter, firstly in their natural state, and
again after removing all traces of essential oil. The results are as
follows : —

NUTMEGS.

235

" Natural" Butter.

Freed from Essential Oil.

1
2
3

M. Pt.

Sap. Value.

Iodine Value.

M. Pt.

Sap. Value.

Iodine Value.

25°
26°
25°

174
172
175

76-5

74-9
78

29°

29-5°

30°

185
186
184

69-5

57-5

It does not appear that any great differences exist between the
solid fats of the various species of mace and nutmeg.

Drs. Power and Salway, who have carefully examined the fixed oil
of nutmeg, found that when the warmed Ceylon nutmegs were expressed,
they yielded 26-6 per cent of fat, but that to ether they yielded 42-9
per cent. They found the nutmeg butter to contain a new unsaponifi-
able body of a viscid consistence, but without physiological action,
to the extent of about 5 per cent, having the formula Cj^Hp^O^. The
other coQstituents of the expressed oil were trimyristin 73 per cent,
essential oil 12-5 per cent, oleic acid as glyceride 3 per cent, linoleic
acid as glyceride 05 per cent; formic, acetic, and cerotic acids, very
small amounts, and unsaponifiable constituents 8 5 per cent; resinous
material 2 per cent, and a little myristicin (" Journ. Chem. Soc." xciii.
p. 1659).

The essential oil of nutmeg is official in the British Pharmacopoeia,
being used in the preparation of aromatic spirit of ammonia. It is
there described as having a specific gravity of 0-870 to 0*910, and being
soluble in an equal volume of 95 per cent alcohol. It should be free
from the solid fat, as shown by leaving no crystalline residue when
evaporated on a water bath. In the author's opinion these limits
should be rather wider — from 0-868 to 0915. The presence of traces
of the fatty oil is said to be objectionable when the oil is used for the
flavouring of sal 'volatile (aromatic spirits of ammonia). It may be
detected by evaporating the oil, and purifying the residual crystals by
washing them several times with cold alcohol and recrystallizing from
boiling alcohol. The resulting myristic acid melts at 54° to 55°. They
should be dextrorotatory from +14° to + 40°, and should be soluble
in an equal volume of 95 per cent alcohol. The chemistry of this
oil requires elucidation.

The most reliable investigation of oil of nutmeg is that of Power
and Salway (" Journ. Chem. Soc." 1907, 2037). They showed that it
contained : —

Per cent

Eugenol and iso-eugenol 0-2

Dextropinene and dextrocamphcne ...... 80

Dipentene 8

d-linalool '^

d-borneol j- about 6

i-terpineoU

Safrol 0-6

Myristicin ........... 4

and traces of alcohols, aldehydes, esters, and free acids.

536

FOOD AND DRUGS.

MACE.

By mace is generally understood the arillus of the fruit Myristica
fragrans, the nutmeg of commerce, which is official in the British
Pharmacopoeia. This is known as Banda mace and is usually re-
garded as the only genuine one. There are other maces, each being
the arillus of another species of nutmeg, but the only ones which are
seen in commerce are Bombay mace, derived from M. Malabarica, and
Macassar or wild mace, derived from M. argentea. The so-called
false mace, from M. fatua, is said to be sometimes met with, but
this is improbable. Mace is usually sold in retail shops in the form
of powder. Banda, or genuine mace, owes its characteristic fragrancy
and its value as a spice largely to an essential oil, which is absent or
only present to a small extent in the other varieties of mace. It also
contains a solid fat, which has the same characteristics as that obtained
from the nutmeg and which is described on p. 234. A comparison of
the figures there given shows that it is not possible to detect the pres-
ence of Bombay mace by the iodine value, but the low refractive value
of the Bombay fat may give a useful indication. According to Leach,
the following are the absolute refractive indices of the fats : Genuine
mace, 14747 to 14975; Bombay mace, 1-4615 to 14633, at 35° C.
Mace is sometimes adulterated with starchy matter, but the usual
admixture is ground Bombay mace or wild mace. The following
characters are the average of the results of a number of samples which
were obtained as whole mace and ground in the laboratory : —

True Mace.

Bombay Mace.

Macassar Mace.

Per cent

Total ash .

1-87 to 2-36

1-9 to 2-1

1-7 to 2-08

Soluble in H2O .

1-08 „ 1-27

1-0 „ 1-2

M „ 1-25

Insoluble in HCl .

0-07 „ 0-2

0-07 „ 0-08

0-6 „ 0-075

Fixed ether extract

25 „ 32-5

58 „ 63

49-5 „ 52

Alcoholic extract .

22 „ 25

45-8

Volatile oil .

4 „ 8

4 to 8

4 to 7

Fibre ....

5 „ 9

3 ., 8

4 „ 8

Nitrogen

0-7 „ 1-2

0-8 „ 0-9

1 „ 1-2

The ether extract above given is that obtained by drying the mace,
extracting with ether and drying at 110° until the volatile oil has been
driven off.

In the examination of mace the following determinations should be
made : —

Mineral Matter. — If the ash be substantially higher than 2 per
cent the addition ot mineral matter is probable. Such apocryphal
adulterations as sawdust, which is stated to sometimes be found,
would cause an increase in the ash. Of the ash at least 50 per cent
should be soluble in water, and only a trace left insoluble in acid. If
starch be added, the ash will be reduced.

Extracts. — Three quantitative extractions of the dried mace should

MACE.

237

be made : (1) ether, (2) alcohol, (3) ether after exhaustion with
petroleum ether.

In genuine mace, the ether extract, dried so as to drive off volatile
matter, averages 22 per cent to 33 per cent, whereas in both Bombay
and Macassar mace it is considerably higher — up to 63 per cent in the
former and 52 per cent in the latter. The alcoholic extract, similarly
dried, is seldom over 23 per cent in genuine mace, whereas it reaches
45 per cent in Bombay mace and 38 per cent in Macassar mace. A
low alcoholic extract indicates the presence of exhausted mace. The
extraction with ether, after exhaustion with petroleum ether, is very
important. Genuine mace, extracted with petroleum ether, will only
yield from 2 to 3*5 per cent of extractive to ether. Bombay mace,
treated similarly, yields an extract up to 33 per cent, so that its presence
in comparatively small amount is thus easily recognized. Macassar
mace, on the other hand, behaves like Banda mace in this respect.

Umney gives the following figures/ for typical samples of known
origin : —

Penang ....
Pale West Indian .
Red „ „ . .
Bombay ....

Petrol. Ether Extract.

Ether Extract after Petrol. Extract.

Per cent
17-55
22-71
28-37
26-11

Per cent

2-68

2-04

3-90

29-11

Griebel (" Zeit. Unterreich. Nahr. Genuss." 1909, 18, 202) gives the
following method for the detection of Macassar mace in genuine mace :
0-1 grm. of the sample, and the same quantity of genuine mace are
placed in test tubes and well shaken for one minute with 10 c.c. of
petroleum spirit. The solutions are then filtered and 2 c.c. of each
filtrate are mixed with 2 c.c. of glacial acetic acid in separate test tubes.
Concentrated sulphuric acid is carefully added to both tubes so as to
form a layer under the acetic acid solution. If the sample contains
Macassar mace, a red ring forms at the junction of the two liquids,
whilst with pure mace only a yellow colour develops. Two minutes
should be allowed for the colour to develop. After this time even a
pure mace may become red — hence the necessity of a check experi-
ment on a pure sample. This • test is useful, but should only be relied
on as confirming more precise results obtained by quantitative deter-
minations.

Special Reactions. — Schindler's reaction is useful in the detection of
Bombay mace (" Chem. Central." 1902, (2), 849) : 5 grms. of powdered
mace are packed, after being moistened with 8 c.c. of 98 per cent alcohol,
into a percolating tube and placed over a suitable receiver ; 8 c.c»
more 98 per cent alcohol are added and the percolate collected. The
receiver is changed and another similar quantity of alcohol added.
This process' is repeated several times. A drop of lead acetate solution
is then added to the various receivers. With genuine mace, the first

238 FOOD AND DRUGS.

tube shows a deep yellow-red precipitate ; the second tube, less precipi-
tate ^nd of a paler colour ; the third tube, none or only a slight whitish
precipitate ; and the fourth tube will be utiatfected. With Bombay mace,
a coloured precipitate results even from the 25th extraction. Hefelmann
{" Pharm. Zeit." 1891, 122) recommends boiling the sample with alcohol
and filtering through paper. The paper is stained red at the edge in
the presence of Bombay mace ; if only a small quantity is present, the
stain may not appear until the paper is dried. If a slip of filter paper
be moistened with an alcoholic extract of the mace, and a drop of
weak caustic soda solution (decinormal) be added, a buff-pink colour
results. In the presence of Bombay mace, this will be of a more or
less deep orange.

If turmeric be suspected — it has occasionally been found — it can be
■detected by soaking filter paper in the alcoholic extract, drying and
testing in the usual mmner with boric acid.

Microscopical Examination. — Epidermal cells will be found to be
very elongated and very thick-walled. The principal part will be
found to consist of parenchymatous cells, mostly containing much
fixed oil and a considerable amount of small granules of amylo- dex-
trin which stain i ed with iodine. Large receptacles or cells will be
found which contain the essential oil. A few spiral vessels are pre-
sent. No starch is present in pure mace.

If Bombay mace be present, the oil glands situated in the outer
layers of this variety are very deeply coloured, so that deep red masses
of resinous matter will be foand, which are absent from ordinary
mace.

Note. — Essential oil of mace is a pale or colourless oil of specific
gravity 0-890 to 0*930 and having an optical rotation of about + 10°
to 4- 20°. It is soluble in 3 volumes of 90 per cent alcohol. It con
sists chiefly of terpenes, and some terpene alcohols, with a phenol
and some myristicin (a complex benzene derivative).

COCHINEAL.

Cochineal is the dried impregnated female insect. Coccus cacti,
which fixes itself firmly on to certain plants of the cactus family,
espeaially the nopel, or Nojjalea coccinellifera, an opuntia growing
chiefly in Mexico. It is employed largely as a colouring matter both
in foods and drugs.

The principal types of cochineal known in the market are the silver
grain, the black grain, and granilla, the last named being, in all pro-
bability, the unimpregnated females.

The British Pharmacopoeia, in which cochineal is official under the
name Coccus, describes the insect as " about 1th of an inch long,
somewhat oval in outline, flat or concave beneath, convex above, trans-
versely wrinkled, purplish-black or purplish-grey, easily reduced to
powder, which is dark red or puce-coloured. When cochineal is
macerated in water no insoluble powder is separated. Incinerated
with free access of air, it should yield not more than 6 per cent of
ash."

COCHINEAL. 239

Cochineal owe^ its colour to a complex acid of the probable
formula C^^H^fi^., named carminic acid, which is easily soluble in
alkaline solutions. This, with certain other subsidiary bodieSj is
precipitated by such salts as alum or stannic chloride and then forms
the carmine lake of commerce.

Under the name "liquid cochineal" is sold a fluid for colouring
food preparations, and which is, in substance, an alkaline decoction of
cochineal, preserved with more or less alcohol. The whole insect is
frequently adulterated, either by the addition of exhausted insects or by
dressing the natural insect with mineral matter rendered adherent in
the wrinkles by means of a little gum. The Britisn Pharmacopoeia
requires an ash value not exceeding 6 per cent, but Umney prefers 8
per cent as the maximum limit. It is probable than absolutely pure
cochineal rarely has a higher ash value than 3 per cent. At all
events plenty of cochineal is available with no higher ash value than
that.

Cochineal — both partially exhausted and natural insects — is largely
adulterated by facing with sulphate of barium, gypsum, mica, chma-
clay, and sometimes with bone-black and similar substances. All
these adulterants raise the ash value considerably, and it is not un-
common to find samples with 12 per cent to 20 per cent ot ash. The
"silver grain" variety is usually faced with some finely powdered
siliceous matter.

Exhausted cochineal always appears deeply wrinkled and is
generally slow to absorb water — so that on throwing the exhausted
insects into water, many of them will float for a time.

Apart from a determination of the ash value, the valuation of the
insect from a colour point of view is the only method of forming an
opinion on the quality of cochineal. This may be done by boiling 1
grm. of the powdered sample with 1 litre of water and 0"5 grm. of
alum for an hour. On cooling, the solution is made up to 1 litre and the
colour compared in Nessler glasses with standard specimens.

Merson ("Chemist and Druggist," 56, 517) recommends valuing
cochineal from the colour point of view, by noting the amount of solu-
tion of chlorinated soda (containing 1 per cent of available chlorine)
necessary to decolorize the colouring matter of 1 grm. of the sample.
He finds that the best samples require over 20 c.c. whereas the poorest
samples only require 9 to 10 c.c. He proposes the following method
of carrying out the determination : —

Weigh 0*5 grm. of finely powdered cochineal ; place in a 100 c.c.
flask with 30 c.c. of distilled water, and 5 drops of strong ammonia ;
heat to boiling-point, strain through cotton-wool into a 100 c.c. flask,
and wash with sufficient water to produce 100 c.c. The marc on the
wool should now be quite colourless. Put 25 c.c. of the liquid into a
100 c.c. stoppered test-mixer, add 5 c.c. of strong hydrochloric acid,
and sufficient distilled water to produce 100 c.c. Eun in 0*5 c.c. at
a time of solution of chlorinated lime (or soda), containing 1 per cent
of available chlorine, till the cherry-red colour changes to dull orange,
shaking briskly after each addition. Continue adding chlorinated
solution in 0"1 c.c. portions as long as the colour is being bleached.

240 FOOD AND DEUGS.

When almost completed, note the burette reading, and after adding a.
further O'l c.c. of solution, shake the liquid slightly and see if the top
layer is lighter than the lower. If there is no difference, the reaction
is finished ; if the lower stratum is darker, continue to add chlorinated
solution drop by drop till the action is quite complete.

An approximate determination of the amount of colouring matter
may be made by exhausting the cochineal with boiling water, precipi-
tating the colouring matter with a slightly acid solution of acetate of
lead, and washing and drying the lead precipitate. The lead precipi-
tate is ignited in a porcelain dish and the loss on weight returned as
colouring matter.

Lagorce recommends the following method for detecting cochineal
in alimentary substances.

The substance should be dissolved in water or weak alcohol faintly
acidified with acetic acid. The liquid is shaken with amyl alcohol,
which is separated and evaporated in the presence of water. The
water solution so obtained is treated with a few drops of a 3 per cent
solution of uranium acetate, when a beautiful bluish-green colour or
precipitate will be produced if cochineal be present. Acids destroy
this colour, with production of the orange tint of the carminic acid.
In the case of wine, a mixture of amyl alcohol and toluene should be

Logwood is distinguished by the black colour produced with ferrous
sulphate, and brazil wood by adding excess of lime water to a little of
the solution. This completely precipitates the colouring matter of
cochineal, but if brazil wood be present, the filtered liquid will have a
purple or violet colour.

SAFFKON.

Saffron consists of the dried stigmata, together with the tops of the
styles, of Crocus sativus. It is used, especially in certain parts of the
country, as^a colouring matter for cakes, and is also employed in medi-
cine. Under the name " Crocus " it is oJBQcial in the British Pharma-
copoeia.

That authority requires it to have the following characters : water,
not to exceed 12*5 per cent ; ash, not more than 7 per cent. On in-
cineration it does not deflagrate (absence of nitrates). When pressed
between blotting paper it does not leave an oily stain. When a small
portion is placed in a glass of warm water it colours the water orange
yellow, becomes paler itself in colour and does not deposit any white
or coloured powder.

Saffron contains about 0-6 per cent of an aromatic essential oil,
and a colouring matter, known as crocin, of the empirical formula
C^^'H.jqB..,^. It probably contains a second colouring matter, known as
picrocrocin CggHfis^n.

Saffron is adulterated to a very large extent, the greater part of
that known as Alicante saflron being a mixture of genuine saffron and
other fibres dyed with a coal-tar colour. Maisch (" Analyst," x. 200)
has given an excellent account of the adulterations of saffron. These

SAFFRON. 241

consist, as a rule, of either dyed or naturally coloured fibres, or of
mineral matter.

Moisture and Ash. — The moisture in commercial saffron averages
from 9 to 12 per cent or shouldnot exceed 12*5 per cent. Occasionally
a trace of glycerine is added in order to induce the saffron to absorb
moisture from the atmosphere. The mineral matter varies from 4*5
to 7 per cent — rarely reaching 8 per cent, of which not more than 0*5
per cent is siliceous. No deflagration should take place during incin-
eration, or nitrates are indicated, and should be tested for in an aqueous
extract. Nitrate of potash is often added, in a strong aqueous solu-
tion, to increase the weight of the saffron. In addition to this, chalk,
sulphates of lime or barium, sulphate of soda, and other salts are
sometimes found, being rendered adherent with a trace of glycerine or
glucose (a conviction was obtained in London in 1909 for the adultera-
tion of saffron with barium sulphate).

If the sample is suspected of being weighted with a mineral, it is
recommended to be placed on the surface of water and gently stirred,
when the water immediately becomes turbid and gradually the powder
subsides, if allowed to stand. In all samples a small quantity of pollen
thus deposits, but its nature can be detected under the microscope.
The nature of any soluble mineral matter present may be ascertained
by testing the aqueous infusion for ammonium salts, nitrates, etc., in
the usual way. The insoluble salts in the deposit have to be rendered
soluble by fusion with alkaline carbonates and then examined accord-
ing to the ordinary rules of mineral analysis.

Other Fibres. — The principal fibres which have been recorded as
adulterants of saffron -are as follows : the corolla tubes and stamens
of the Crocus, dyed with either brazil or santal wood dye ; or with a
coal-tar yellow (usually dinitrocresylate of sodium) ; Calendula, Cartha-
mus, Cyanara, red poppy, threads of algae, and various other plant
fibres.

For the general detection of such adulterations, the sample should
be scattered on the surface of warm water. The genuine saffron fibres
at once expand to a characteristic form, which are readily distinguished
from Crocus stamens, and such fibres as Carthamus florets, or Calendula
florets. A comparison with a standard sample will enable most adul-
terations to be thus readily detected.

Kraemer recommends adding the sample to dilute sulphuric acid.

With crocus only the stigmas become blue immediately, and in
half a minute the solution becomes blue, gradually changing first to a
violet, then to a deep wine-red colour. The flowers of Carthamus turn
yellow ; the solution remains colourless for a few minutes, then be-
comes yellow, and after a much longer time assumes a deep wine-red
colour. Calendula flowers turn brown, or blackish- brown, as if
charred, but the solution behaves much the same as with Carthamus.

The colouring matter from santal wood is characterized by being
soluble in alcohol with a red colour, and in ammonia with a purple-red
colour. Brazil and logwood dyes will tinge the water a red colour,
deepening by addition of ammonia and becoming paler by addition of
acid. Coal-tar dyes may be usually detected as follows : —
VOL. I. 16

242 FOOD AND DKUGS.

Nitrocresylate of sodium, which is the dye most usually employed,
may be detected by soaking in petroleum spirit, when the spirit ac-
quires a lemon- yellow colour, the colouring matter of saffron not being
soluble in that liquid.

According to Wanters (" Bull. Assoc. Beige. Chem." xii. 103) a
good test consists in trying the tinctorial power of a sample on
wool, silk, and cotton. These materials strike a citron -yellow colour
with a solution of the true drug, containing tartaric acid, which is not
altered by subsequently treating the materials with potassium hydrate.
Under treatment with a solution of the spurious article, the wool takes
a deep brownish-red, the silk a deep orange-yellow, and the cotton
a lighter yellow tint ; in each case the addition of potassium hydrate
causes a deepening of colour.

Pfyl and Scheitz (" Zeit. Nahr. Genuss." 1908, 16, 347) estimates
the value of saffron by determining the amount of sugar obtained by
hydrolysing the chloroform -soluble glucosides present. The sample
is dried, and 5 grms. of the dry, powdered saffron are extracted in a
Soxhlet with petroleum ether, and after drying with chloroform, the
solvent is evaporated from the chloroformic solution and the residue
taken up as far as possible in acetone. The latter is evaporated and
the glucoside in the residue hydrolysed by adding 5 c.c. of normal
HCl, and heating for fifteen minutes, water being added as required,
to bring the amount of the Hquid up to 25 c.c. The liquid is, if neces-
sary, filtered, and neutralized with normal alkali, and the sugar es-
timated by reduction of Fehling's solution, the copper being weighed
as CuO, and calculated to Cu.

Pure saffron, consisting only of stigmata of the crocus, yields from
0-199 grm. to 0-209 grm. of Cu, when treated in this manner.

The styles of the Crocus, logwood, poppy petals, peony petals,
marigold, safiflower. Cape saffron, and Spanish thistle flowers give
practically no copper, and turmeric and red sanders wood very much
less than saffron.

Microscopic Examination. — The principal tissues consist of long-
celled parenchyma, with a number of vessels, often spiral ; and some
of the parenchymatous cells have large papillae attached, and nearly
all the cells are red in colour. A comparison with genuine saffron
will enable most adulterants to be detected under the microscope.

Valuation of Saffron. — Proctor has suggested a colorimetric
method of valuing saffron from a colour point of view. Dowzard has
also worked in this direction and gives the following method ("Ph.
Jour." 4, VII. 443) which, in the author's opinion, is fairly accurate : —

The method is not designed to furnish evidence of adulteration, but
merely to test the value of a sample of saffron as a colouring agent.

A standard solution of chromic acid is prepared containing 78-7
grms. of chromic acid per litre. One hundred c.c of the above solution
are equal in tinctorial power to 0-15 grm. of crude crocin dissolved in
100 c.c. of water (crude crocin is obtained by extracting saffron with
ether, drying, a -id exhausting the residue with 50 per cent alcohol ; the
alcoholic solution is evaporated to dryness, and the residue taken as
crude ciociu).

TURMERIC. 243

The sample is reduced to a coarse powder by pestle and mortar,
0*2 grm. of the powder is transferred to a stoppered cylinder having a
capacity of about 35 c.c. ; 20 c.c. of 50 per cent alcohol are then intro-
duced into the cylinder, which is tightly stoppered, and placed in
water at 50" C. for 2^ hours. The solution is cooled and filtered, 10
c.c. of the filtrate ( = 0*1 grm. of saffron) are diluted with water to 50
c.c, and the depth of colour compared with 50 c.c. of the standard
chromic acid solution (for comparing the colours it is suf&cient to
use two Nessler glasses of equal bore). If the chromic solution is
deeper in tint than the solution under comparison, small quantities
are removed until equality is produced, or vice versa ; the solutions
tions are then measured, and the amount of crude crocin calculated.

Example. — Ten c.c. of saffron solution ( = 0'1 grm. of saffron) di-
luted to 50 c.c. had a depth of colour equal to 40 c.c. of the standard
chromic acid solution,

100 : 40 : : 0-15 : x

= 006.

.*. 50 c.c. contain 0*06 grm. of crude crocin.

0-1 : 100 : : 0-06 : x

= 600.

The above example therefore contains 60 per cent of crude crocin.

The finest samples on the market contain upwards of 75 per cent
of crude crocin ; good samples of saffron should not contain less than
50 per cent.

TURMERIC.

Turmeric is the rhizome of Curcuma longa, or so-called Indian
saffron (the name is derived from the word Kurkum, the Persian word
for saffron), and Curcuma rotunda. Powdered turmeric is used to a
very large extent in the preparation of curries, pickles, etc., the char-
acteristic colour and flavour of "picallili" being produced by turmeric.
It is usually sold in shops in the powdered form. Turmeric contains
about 10 per cent of a resin, and about 5 per cent of an essential oil
having the characteristic odour of the rhizome. It owes its colour to
a substance named curcumin, probably bearing the formula C^gHj^O^
(OCH3)2. For the chemistry of this body see Ciamician and Silber
(" Berichte," xxx. 192).

In the table on p. 244 A. E. Leach (" Journ. Amer. Chem." Soc.
26, 1210), gives the analyses of the three varieties of turmeric most
frequently met with in commerce.

To curcumin the characteristic boric acid reaction is due. The
reaction is best applied to blotting-paper soaked in an alcoholic extract
of the turmeric and dried. A solution containing boric acid, or borax
to which sufficient HCl has been added to show an acid reaction to
litmus, is spotted on to the paper and the latter again dried. The spot
will be of a red colour, changed by a drop of alcohol through a very
varying series of colours, in which green and purple predominate.

According to Bell (" Pharm. Journ." 4, 15, 551) turmeric may be

244

FOOD AND DEUGS.

4^'

^•

^°

.s
i

s
o

3
1

l-H

Per

52i

5
fa

^6.

So J

Per

cent

China

9-03

6-72

5-20

0-11

1-73

10-81

10-86

2-01

8-84

9-22

4-45

48-69

40-05

Pubna

9-08

8-52

6-14

0-97

6-06

12-01

4-42

7-60

7-28

5-84

50-08

29-56

Aleppo

8-07

5-99

4-74

1-56

9-75

10-66

3-16

7-51

4-37

5-83

50-44

33-03

Average

8-73

7-07

5-36

—

1-42

8-88

11-17

3-19

7-98

6-96

5-37

49-73

34-21

identified in mixtures by its reaction with diphenylamine. It gives
with diphenylamine, in acid alcohoHc solution, a fine purple coloration.
No other vegetable colouring matter has been found to give a similar
reaction, so that the test is available for turmeric mixed with other
substances. One part of turmeric in 200 parts of rhubarb, or in 1000
parts of mustard, is readily detected.

The following is the best method of applying the test : A drop of
the reagent is placed on a clean microscopic slide by means of a glass
rod, a small quantity of the powder under examination is spread evenly
over the entire surface of a cover glass, and carefully dropped over the
reagent on the slide.

The slide is then examined microscopically with an inch objective,
when, if turmeric be present, spots of a fine purple colour will be
observed scattered throughout the field. The number of these purple
spots can be employed in estimating approximately the amount of the
drug present by comparison with standard specimen slides containing
a known percentage of turmeric.

The reagent consists of pure diphenylamine, alcohol 90 per cent,
and pure sulphuric acid : —

Diphenylamine 1 gm.

Alcoliol 90 per cent 20 c.c.

Pure sulphuric acid 25 c.c.

The diphenylamine is dissolved in the 90 per cent alcohol, and the
sulphuric acid is then added. When cold the reagent is ready for
use.

In examining turmeric,' the following determinations will afford
most of the necessary information : —

Moisture. — The average amount of moisture is 8 per cent to 9 per
cent. Excess will cause the sample to cake and it will then rapidly
deteriorate.

Ash. — The total ash of turmeric varies from 6 per cent to 8-5 per
cent of which 70 per cent is soluble in water, and not more than 0-2 per
cent insoluble in hydrochloric acid.

TURMERIC.

245

Extractives. — The ether extract, free from essential oil, varies from
7*5 per cent to 10 per cent, the alcoholic extract from 5*5 per cent to
10 per cent.

Starchy Matter. — Turmeric contains a considerable amount of
starch, and if this is converted in the usual manner by acid, the
amount of glucose obtained by reduction of Fehling's solution should
vary from 47 per cent to 52 per cent. Excess of this indicates adultera-
tion with starchy matter. '

Microscojiic Examination, — The mass of the tissue consists of large
yellow parenchymatous cells filled with colouring matter ; many starch
granules are present, some stained yellow by curcumin, and swollen

Fig. 33. — Turmeric rhizome, transverse "section, g, vessel ; k, cork ; oe, oleo-resin
2J, parenchyma of wood, containing masses of gelatinized starch. (Moller.)

by^the process of scalding to which the rhizome is usually subjected
before being put on the market;. The starch granules are oval, oblong
or oyster-shaped and show w^ell-defined rings and an eccentric hilum,
but, owing to the method of preparation, the starch grains are usually
more or less disintegrated, and well-defined starch grains may indicate
the presence of an adulterant. The microscopic appearance is very

246

FOOD AND DKUGS.

similar to that of ginger, but there are no bast fibres. Foreign
starches should be searched for.

Oil of Turmeric. — This is a thick liquid of specific gravity about
0*940 and contains an alcohol termed turmerol, which is probably the
aromatic ingredient of the oil. It is of little commercial import-
ance.

ANNATTO.

Annatto is a colouring matter employed in the preparation of various
foods, to which a pale yellow tint is imparted. It consists essentially
of the soft tissue surrounding the fruits of Bixa orellana, a plant
growing in the East and West Indies and in South America. Such
American annatto usually arrives in this country in the form of hard
rolls or cakes, containing from 15 to 30 per cent moisture, whereas
Cayenne annatto is generally imported in the form of a soft paste. It
owes its colouring power to at least two bodies, bixin CggHg^Og,
orellin, of unknown composition. Annatto consists of about 25
cent of colouring matter, with cellular tissue, glutinous matter,
mineral matter.

It is soluble to a variable extent in water, and to a greater extent
in alcohol. Caustic alkalies, alkaline carbonates and borax in solution
dissolve it to a large extent, orange-red colouring matter being precipi-
tated on the addition of acids.

Annatto is adulterated to a considerable extent, the principal adul-
terants being starchy matter, oxide of iron, salt and aniline dyes. The
last named are added, together with a little potassium carbonate, to
improve the colour of samples which have been reduced with starchy
matter or similar adulterants.

Pure roll annatto has the following composition : —

and
per
and

Water .
Kesinous matter
Mineral matter
Extractive matter

Per cent
15 to 20
23 „ 30
18 „ 22
20 „ 28

Paste annatto has a proportional composition, with water up to
70 or 75 per cent.

Lawson (" Pharm. Journ." [3], xvi. 645) gives the following as his
results of the analysis of a number of commercial samples : —

Moisture.

Resin.

Extractive.

Total Ash.

Ash Soluble in HgO.

Per cent

21-75

3-00

57-29

17-9G

13-20

roll

21-60

2-90

59-33

16-17

12-57

20-39

1-00

65-00

13-61

7-50

69-73

8-80

19-47

2-00

paste roll

18-00

3-00

58-40

20-60

10-00

18-28

1-80

65-67

14-25

11-75

15-71

5-40

26-89

52-00

18-50

38-18

1-20

20-82

29-00

20-00

19-33

5-99

23-77

51-00

15-00

29-50

9-20

28-50

39-80

13-80

ANNATTO. 247

Anaiysis of Ajinatto. — The water should first be determined, and
the dry residue should be exhausted with boiling methylated spirit.
After driving otf the alcohol, the residue is re-dissolved in a solution of
sodium carbonate and a slight excess of dilute sulphuric acid added to
the liquid. The resin is precipitated, and is filtered off, washed with
water, dried, and weighed.

The ash is determined on a fresh portion of the sample.

The extractive matter (which however includes a little insoluble
matter) is taken as the difference.

Annatto should be examined microscopically. Owing to its
method of treatment, but little structure will be observed, only a few
starch granules being unbroken. The presence of added starchy matter
or turmeric is therefore fairly easily identified.

In solution of colouring matter in which annatto is suspected, the
following reaction may be applied : —

K dilute solution of the colouring matter is floated in a test tube,
on an equal volume of dilute HNO3, so that the two solutions do not
mix. In the presence of annatto the zone of contact at once shows a
deep blue colour, the colour spreads into the HNO3, which soon be-
comes green, and the upper aqueous layer shows a reddish turbidity.

Milk is often coloured with a trace of annatto to give it a fictitious
appearance of richness. To detect this addition. Leys employs the
following test : 50 c c. of the sample are shaken out with twice its
volume of ether-alcohol mixture composed of 240 parts of alcohol (93
per cent), 320 parts of ether, 20 parts of water, and 8 parts of solution
of ammonia of -920 specific gravity. After separation, the ethereal
layer is rejected, the colouring matter being retained in the aqueous
portion. This is transferred to another vessel and half its volume of
a 10 per cent solution of sodium sulphate gradually added, which
causes slow separation of the casein. The clear aqueous portion
is decanted and shaken out with amyhc alcohol, the washing being
conducted in test-tubes to facilitate the separation of the solvent.
After shaking, these' tubes are plunged into a cold water bath, the
temperature of which is gradually raised to 80° C, when separation
will be complete. The amylic alcohol solution is collected and eva-
porated. The deep yellow residue is re-dissolved in warm water
containing a little ammonia and alcohol, a strip of bleached cotton is
immersed in the solution, and the whole evaporated to dryness. The
cotton, which is now of a yellow tint, is washed and plunged into
a solution of citric acid. If the colouring be annatto, the thread will
at once assume a marked rose tint. Uncoloured normal milk imparts
a very slight yellow tint to cotton by this method, but does not give
the change of tint with citric acid, which is characteristic of annatto.

VINEGAR.

The definition of vinegar is by no means an easy matter. Origin-
ally, as is obvious from its name ("sour wine "), vinegar w^as the pro-
duct of the acetous fermentation of wine. To-day products which
are essentially of the same character are obtained by similar fermenta-

248

FOOD AND DRUGS.

tions of other alcoholic liquids, and to these the term vinegar is natur-
ally applied. The characteristic constituent of vinegar is acetic acid,
and the name has therefore been extended — rightly or wrongly — to
dilute acetic acid obtained by the distillation of wood. The propriety
of thus extending the name vinegar will be discussed in the second
volume of this work, dealing with the law on the subject. In the pre-
sent chapter, wood vinegar will be understood to be excluded from the
term vinegar, except when qualijfied by the word " wood " as denoting
its origin.

With this qualification vinegar may be defined as the product,
consisting of dilute acetic acid with small quantities of subsidiary con-
stituents, obtained by the oxidation of alcoholic solutions obtained from
vegetable source-;.

In practice this oxidation takes place by means of fermentative
changes brought about by an organism known as mycoderma aceti.
The principal varieties of vinegar are as follows : —

(1) Malt Vinegar. — This is the product of the fermentation of a
wort made from malt and barley (but see below).

(2) Wi7ie Vinegar. — This vinegar is made from grape juice and
low-grade red or white wines.

(3) Sugar Vinegar. — This is usually made by the hydrolysis of
starchy matter by means of dilute acids, followed by fermentation of
the starch, and subsequent acetification of the alcohol formed.

(4) Cider Vinegar, made from cider (or perry).

(5) Date vinegar, made from a fermented extract of dates.

The general characters of vinegars made from various sources differ
somewhat and are shown in the table on opposite page.

Fairley ("Analyst," xxxiv. 515) has shown that malt vinegars,
brewed from mixtures of malted and unmalted grain (maize in this
instance) will often contain less phosphoric acid than some whole malt
vinegars. He gives the following figures for a number of samples : —

Average.

Maximum.

Minimum.

Per cent

Acetic acid

4-50

5-30

3-65

Total solids

2-51

4-01

1-52

Ash .

0-45

0-96

0-14

Proteins .

0-40

0-79

0-22

PA. . . .

0-058

0083

0-040 ,

Specific gravity .

1-0169

1-0210

1-0120

How far such vinegars are properly described as " malt vinegar "
is a matter which has never been decided in a Law Court. The fol-
lowing remarks, due to Dr. Bernard Dyer, probably express the current
opinion on the subject (" Analyst," xxxiv. 518) : —

"He imagined that there were two accepted meanings for the
term "malt vinegar". The view of the extreme purist would be that
malt vinegar was vinegar in which everything but the water was de-

VINEGAR.

249

rived from malt. The more liberal interpretation, and one which for
a number of years had been largely accepted, was that malt vinegar

S •"'

. 6

o o

„

u ■"

^ s

Ift!

(>)

^ o

i»

g '^

o
6

(?5

«5

r-l

^ o

Tt<

i^'

(jl

■«*

<?q

1—1

" -2

;->

"!**

TJ4

C<J

«o

= -^

" o

p^ CO

-f^ '=5

«o

g '^

O o

.£3

»H

•^

i-i

-k3

-2

■"1
S

-a

"s
-^

-G

;zj

was vinegar in which the acid was derived from starch which had
been rendered soluble by the diastatic action of malt — i.e. vinegar
made by means of malt, but not necessarily wholly from malt. Malt

250 FOOD AND DRUGS.

vinegar, even according to this liberal interpretation, would differ
from vinegar made from sugar or from a mixture of malt and
sugar."

In addition to acetic acid, vinegar contains such bodies as ethyl
acetate, traces of alcohpl, sugar, gummy matter, acetates, tartrates and
phosphates, colouring matter, albumenoids and traces of some bodies
which impart the characteristic odours to the different vinegars.

Vinegar is known to English manufacturers by numbers — 18, 20,
22 and 24 being the usual strengths. These numbers refer to the
amount of dry sodium carbonate (in grains) required to neutralize one
ounce of the vinegar. Since 60 grains of acetic acid are neutralized
by 53 of sodium carbonate, the number of grains of acetic acid per
ounce corresponds with the number of the vinegar multiplied by 1"132.
The " number " multiplied by 0"259 gives the percentage of acetic
acid (i.e. lueight of acetic acid in 100 volumes of the vinegar).

Before discussing the analysis of vinegar, it must be recognized
that an expert nose and palate will often be of more value than a chemical
analysis, since factitious vmegars made from dilute acetic acid, coloured
with caramel, with a trace of phosphates and nitrogenous matter are
made which give results practically iodistinguishable from those of
genuine vinegars.

The analysis of vinegar should include the following determina-
tions : —

Acetic Acid. — Acetic acid may be determined with sufficient ac-
curacy by titrating a measured quantity with standard potash solution,
using phenol-phthalein as an indicator. The sample should be
diluted — say5. c.c. to 50 c.c. — with distilled water and titrated with
semi-normal alkaU.

Any refinements, such as adding excess of sodium carbonate and
titrating back with standard acid with the use of cochineal as indicator
are quite unnecessary.

The only error likely to occur is due to the presence of mineral
acids in the sample. Very weak vinegar is liable to putrefy, and on
this account an old excise regulation (under the Vinegar Act of 1818
since repealed) allows the addition of 1 gallon of sulphuric acid per
1000 gallons of vinegar. This addition is not necessary in good vinegars
and is not often practised, although it is not yet an obsolete custom.
The deliberate adulteration of vinegar with sulphuric or hydrochloric
acid is not a very common occurrence.

Free hydrochloric acid may be determined by distilling 100 c.c. of
the sample till 90 c.c. have been collected, adding 100 c.c. of distilled
water to the residue in the distillation flask and distilling a further 100
c.c. Practically the whole of the hydrochloric acid will be found in
the distillate and may be precipitated and weighed as silver chloride.

Hehner has examined this subject and has suggested a method for
determining the free mineral acids. Vinegar always contains potassium
and sodium salts of acetic acid (and tartaric acid, if present). Mineral
acids added in small quantity will therefore merely decompose such
salts and become converted into sulphates or chlorides as the case may
be, disappearing as free mineral acids entirely. Any excess, however,

VINEGAR.

over the quantity necessary for such reaction remains as free mineral
acid.

It is therefore clear that if any undecomposed acetate or tartrate
remains in the vinegar, it is impossible for any free mineral acid to be
present. To decide this point, the vinegar should be evaporated and
the ash tested. Acetates and tartrates yield carbonates on ignition, and
therefore if the ash is alkaline, no free mineral acid could have been
present. If the ash is neutral, free mineral acid is probably present.
To use this as a quantitative reaction, a known volume of the vinegar
should be carefully neutralized with standard solution of soda, and the
sample now evaporated to dryness. The residue is ignited and the
aqueous solution of the ash is titrated with standard acid. If the
original acid of the vinegar were entirely organic, the ash will require
for neutralization acid exactly equivalent to the alkali required for the
neutralization of the vinegar. Any deficiency in the amount of the stan-
dard acid required for neutralization of the ash is due to free mineral
acid originally in the vinegar. Fifty c.c. may be used for this deter-
mination. Free sulphuric acid may also be determined by evaporating
100 c.c. of the vinegar to a very small bulk and adding to the liquid
when cold four or five times its volume of alcohol. Sulphates are pre-
cipitated, free sulphuric acid remaining in solution. The filtered liquid
is diluted with water, the alcohol driven off, and the free sulphuric
acid precipitated as barium sulphate. If much chloride was present
in the water used for manufacturing the vinegar, some of the added
sulphuric acid will have decomposed the chlorides, and free hydro-
chloric acid will also be present, which must be estimated by distilla-
tion.

In regard to qualitative reactions, it must be remembered that
many waters used for brewing vinegar contain large amounts of
chlorides and sulphates, so that reactions for these salts are meaning-
less.

To detect free mineral acids Ashby ("Analyst," ix. 96) prepares a
2 per cent decoction of logwood. Drops of this liquid are spotted on
a flat porcelain plate, and evaporated to dryness over boiling water.
A drop of the suspected sample (concentrated if necessary) is added to
a spot of the logwood decoction and evaporated. In the presence of
free mineral acids, the residue is bright red, whereas in the presence
of pure vinegar it will be bright yellow.

Hilger (" Archiv. des Pharmazie," 1876, 193) adds a few drops of
a 0-1 per cent solution of methyl-violet to 25 c.c. of the vinegar. If this
be pure no change in colour results, but in the presence of as little as
0-2 per cent of free mineral acid, the colour becomes blue ; with 0*5
per cent it is blue-green, and with 1 per cent it is green.

Total Solid Matter. — Ten c.c. of vinegar should be evaporated to
dryness on a water bath, and as the residue obstinately retains traces
of acetic acid, it is best to add a little alcohol to this and evaporate
again to constant weight. The extract varies according to the nature
of the vinegar, as will be seen from the table on p. 249.

Mineral Matter. — The residue obtained in determining the total
solid matter is ignited at as low a temperature as possible, and the ash

252 FOOD AND DEUGS.

weighed. The average amounts for the various types of vinegar will
be found in the table on p. 249. Sugar vinegar, prepared by the in-
version of starches, yields an ash containing much sulphate. Cane-
sugar vinegar yields a readily fusible ash, whilst that of a malt or
glucose vinegar is not readily fusible. The ash of cane-sugar vine-
gar is composed chiefly of potassium salts.

Nitrogen. — Vinegar made from sugars contains hardly any proteid
matter. Grain vinegars, on the other hand, contain as much as 0-75
per cent of proteids. To determine the nitrogen 25 c.c. may be
evaporated to a syrupy consistency and the nitrogen determined by
any modification of the Kjeldahl method. A genuine malt vinegar
will contain from 0'088 to 0'125 per cent of nitrogen, whereas sugar
vinegars often contain less than O'Ol per cent, and factitious vinegars
made from acetic acid practically none at all.

PhospJioric Acid. — The phosphoric acid should be determined by
treating the ash with a little hydrochloric acid and evaporating to
dryness. The residue is then dissolved in about 10 drops of dilute
acetic acid ; 50 c.c. of boiling water are added, and about 1 grm. of
sodium acetate. This solution is then titrated with uranium acetate
solution in the usual manner.

Metallic Contamination. — Occasionally traces of copper, lead,
zinc, or tin are found in vinegar. To detect these, 10 c.c. of the
sample is boiled and 1 c.c. of strong hydrochloric acid is added. A
little potassium chlorate is added in very small quantities at a time,
whilst the liquid is still boiling, until the colour becomes pale yellow :
the boiling is continued for a minute, and the liquid is then treated
with sodium acetate, to remove the hydrochloric acid, and treated with
H^S. Very faint traces of lead, copper, or tin can be thus de-
tected.

Colouring Matter. — Factitious vinegar is usually coloured with
caramel. Amthor (" Zeit. f. Anal. Chemie," xxiv. 30) recommends
the following process for its detection : 10 c.c. of the sample and 30 c.c.
of paraldehyde are mixed with sufficient absolute alcohol to obtain
complete solution. The mixture is allowed to stand in a closed vessel
for twenty-four hours, when caramel, if present, will be precipitated.
The liquid is decanted, the precipitate washed with a little absolute
alcohol, dissolved in water, and the aqueous solution evaporated to 1 c.c.
To this a small quantity of a solution of phenylhydrazine in acetic acid
is added and the liquid heated to 100" for half an hour. x\n amorphous
compound is formed, which is probably a mixture of phenlhydrazones
and osazones, if the precipitate were caramel.

No accurate method is known for the estimation of caramel, nor,
as a rule, is one required. It has been suggested that colorimetric
tests, using caramel as a standard, give fairly accurate results, but this
is not so, as the colour of caramel varies so much, according to its
method of preparation, that one sample may be of a much greater
depth of colour than another — so that quantitative comparisons are
impossible.

Preservatives. — Preservatives are seldom added to vinegar (see
above as to the addition of sulphuric acid). Salicylic acid is, how-

VINEGAR. 263

ever, occasionally found to be present. It may be detected by adding
a few drops of sulphuric acid to the sample, and extracting with ether
or petroleum ether, and estimated colorimetrically by means of iron-
ammonia alum (see page 680).

Benzoic acid can be detected by extracting in the same manner,
and, after washing the ethereal extract with water, evaporating the
ether on a watch-glass and covering the extract with a piece of filter
paper and then with a clock-glass. If the watch-glass be heated over
a small flame, benzoic acid will sublime through the paper and
collects on the upper glass. The crystals are characteristic in form
and may also be recognized by their reaction towards ferric chloride.

The Detection of Methyl- acetol. — Pastureau has observed that
specimens of commercial vinegar give precipitates with 95 per cent
alcohol and possess a powerful reducing action in the cold on alkaline
cupric tartrate. He finds this to be due to the presence of methyl-
acetol, CHg.CO.CHOH.CH,^, in one case to the extent of "325 per cent.
This substance should not be present in naturally brewed vinegars, and
may possibly indicate the presence of wood distillation products. To
determine this substance 50 c.c. of the vinegar are neutralized with
Na.2C03 and distilled carefully to dryness. The distillate is collected
in a graduated 100 c.c. flask and is rendered alkaline by a few drops
of caustic soda solution and ammonia and then treated with 10 c.c.
of decinormal solution of silver nitrate. After being allowed to stand
for twenty-four hours, the liquid is made up to 100 c.c. with water,
filtered, and the excess of silver determined by the potassium cyanide
method. The amount of methyl-acetol present is calculated from
the amount of silver reduced, according to the equation : —
3(CH3CO.GHOHCH3) + AgN03= 3(CH3CO.CO.CH3) + SH.O -{- N -f Ag.

Eemarks on Special Types of Vinegar.

Wi7ie Vinegar. — This vinegar varies in colour, according to the
type of wine from which it has been made. Further, the darker
varieties are often distilled, and the product sold as " distilled wine
vinegar ". Naturally this type of vinegar is practically free from solid
matter, and can only be distinguished from a factitious vinegar by
its fine characteristic aroma. Distilled wine vinegar is often described
as white wine vinegar, but according to some, the latter term should
be restricted to normal vinegar made from lohite luine. This is usu-
ally of a pale straw colour.

Normal wine vinegar contains from 5 to 10 per cent of acetic
acid — usually from 6 to 8 per cent. Its specific gravity is usually
fairly low, averaging about 1-017. If the solid residue (averaging
about 2 per cent) be treated with absolute alcohol, nearly the whole
is dissolved, except a small granular residue of acid tartrate of potas-
sium, the presence of which is characteristic of wine vinegar. Vinegars
made from malt or sugar leave a more or less glutinous residue very
insoluble in alcohol. It is not uncommon for the grape juice to be
fortified with sugar during fermentation or acetification. Vinegars
resulting from this process will show a more or less glutinous insol-

254

FOOD AND DRUGS.

uble residue when treated as above described. To definitely identify
the cream of tartar, the crystalline residue should be separated by
pouring off the alcohol and dissolving it in the minutest quantity of
hot water. The liquid is cooled and the watch-glass on which it is
placed is rubbed with a glass rod when the cream of tartar will be
deposited in streaks along the track of the rod. Addition of a drop
or two of alcohol renders the test more delicate.

The following figures are those of the Municipal Laboratory of
Paris for white wine vinegars : —

Sp. Gr.

Solids.

Sugar.

Potassium Bitartrate.

Ash.

Acetic Acid.

Maximum
Minimum
Mean

1-0213
1-0129
1-0175

3-19
1-38
1-93

0-46
0-56
0-22

0-36
0-07
0-17

0-69
0-16
0-32

7-38
4-44
6-55

It may be necessary to determine the amount of potassium bitar-
trate in a wine vinegar. This may be done by evaporating 25 c.c. of
the vinegar to a syrupy consistency, and dissolving the residue in
water to its original volume. One hundred c.c. of a mixture of ether
and alcohol in equal parts are then added, and the mixture kept in a
cold place for forty-eight hours. The precipitated tartrate is collected
on a filter, washed with ether-alcohol, and finally dissolved in hot

Each c.c. is equivalent to 0'0188

water and titrated with —alkali.

gr. of potassium bitartrate.

Fleury (" J. Pharm. Chim." 1910, 2, 264) considers that the presence
of inositol is characteristic of wine vinegar (except distilled wine
vinegar). He evaporates 100 c.c. nearly to dryness, takes up the
residue with 50 c.c. of water, and adds 3 gr. of J3a(0H).,. The pre-
cipitate is separated centrifugally and washed with 30 c.c, of baryta
water. The filtrate and washings are freed from excess of barium by
a current of CO.,. Ten c.c, of a 30 per cent solution of acetate of lead
are added and the precipitate separated as before. The filtrate is
made up to 100 c.c, and treated with 10 c.c. of basic lead acetate
solution and 2 gr. of cadmium nitrate dissolved in water. The precipi-
tate now formed contains the inositol. It is separated, decomposed
by HgS and the filtrate evaporated to a syrup, and taken up wath 20
c.c. of absolute alcohol and 5 c.c. of ether. After standing for forty-
eight hours, inositol, if present, separates in crystals, which are sweet
and melt at about 250°.

Cider Vinegar. — This vinegar is of a pale yellow colour and has
a distinct odour of the fruit. It is usually the weakest of all brewed
vinegars, often containing as little as 3*5 per cent of acetic acid. The
solid residue, which averages about 2 per cent, is somewhat mucilagin-
ous, and during evaporation the odour of baked apples may be ob-
served. The residue contains malic acid, but no tartaric acid.

The sugar, as indicated by reduction of copper solutions, should
be the same before and after inversion (inversion should be accom-

VINEGAK. 255

plished by hydrochloric acid at 70° C). Cider vinegar is optically
active. If 25 c.c. of the vinegar be clarified by the addition of 2*5 c.c.
of 10 per cent lead acetate solution, the filtered liquid should show an
optical rotation between - 0°6' and - 4° in a 200 mm. tube. This
vinegar is characterized by the presence of malic acid. If it does not
give a precipitate with lead acetate which settles in a few minutes, it
is not cider vinegar. To confirm the presence of malic acid, 5 c.c.
of the vinegar is treated with 1 c.c. of a 10 per cent solution of calcium
chloride. The liquid is filtered and to the filtrate about three times its
volume of 95 per cent alcohol is added. In the presence of malic acid
a flocculent precipitate will be formed. Dextrin will give a precipitate
under these circumstances, but its presence will be indicated by a
dextro-rotation in the polarimetric test. Sulphate will also yield a pre-
cipitate, although not of the same character. If the precipitate be
collected, dried, dissolved in HNO3 and evaporated to dryness on the
water bath, malates will be converted into oxalates. The residue is
treated with a little hot sodium carbonate solution, acidified with
acetic acid, and tested with a solution of calcium sulphate. A pre-
cipitate of calcium oxalate is a decisive indication of malic acid.

Cider Vinegar, Analysis and Suggested Standards for. — A. E. Leach
andH. C. Lythgoe ("Journ. Amer. Chem. Soc." xxvi. 375) recom-
mend the following scheme of analysis of, and standards for, cider
vinegar : —

Acetic Acid. — Three c.c. of vinegar are diluted with about 300 c.c.
of water and titrated with N/lONaOH, using phenol-phthalein as in-
dicator. The number of c.c. of alkali used, multiplied by 0*2 gives
the percentage of acetic acid, which should not be less than 4*5
per cent. (This is incorrect : ic may be under 3 per cent.)

Solids. — ^Five grms. of vinegar are weighed into a tared, flat-
bottomed platinum dish, subjected for an hour to direct contact with
the live steam of a boiling water bath, and the residue weighed. It
should be approximately 2 per cent.

Ash. — The residue from the solids is carefully ignited in a muffle
and the resulting ash weighed. It should be about 6 per cent of the
solids. (It is often higher than this.)

Alkalinity of the Ash. — One hundred grms'. of vinegar are evapor-
ated to dryness in a platinum dish and the residue reduced to an ash
in a muffle. The resulting ash is boiled with water, the solution
filtered and the residue washed with boiling water till free from alkali.
The filtrate is then treated with an excess of' N/lOHCl, the solution
boiled to expel CO.,, and the excess of acid titrated with N/lONaOH,
using phenol-phthalein as indicator. The number of c.c. of N/lOHCl
required for neutralization should be equivalent to at least 65 c.c. for
each grm. of ash.

Soluble Phosphoric Acid. — The solution of the ash, after titration,
is made acid with HCl, and evaporated to dryness, after which 50 c.c.
of boiling H.^O are added, and the P2O5 determined by titration with
uranium acetate in the usual way. At least 50 per cent of the total
phosphates should be soluble in water.

Insoluble Phosphoric Acid. — The residue from the ash soluble in

256 FOOD AND DEUGS.

water is dissolved in HCl, and the acid solution evaporated to dryness.
The residue is then dissolved in about 10 drops of dilute HCl, 50 c.c.
of boiling H^O are added, then about 1 grm. of sodium acetate and the
solution titrated with uranium acetate, as in the case of the soluble
phosphoric acid.

Reducing Sugars. — Two portions of 25 c.c. each are rneasured into
100 c.c. flasks. One portion is diluted with 20 c.c. of water, 5 c.c. of
concentrated hydrochloric acid are added and the solution subjected
to inversion by heating to 70'^ C. for ten minutes and cooling. Both
portions are neutralized with sodium hydroxide and made up to a
known volume. The reducing sugars determined by the ordinary
methods should be the same before and after inversion. Any increase
denotes the presence of cane sugar or glucose.

Polarization. — Twenty-five c.c. of the vinegar is precipitated with
2*5 c.c. of 10 per cent lead acetate solution, and filtered bright. It
should show a rotation of between - 0*1° to - 4*0° in a 200 mm. tube.

Malic acid should be present, as shown both by the lead acetate
and the CaCl, tests. If the vinegar under examination does not give
a precipitate with lead acetate, settling in a few minutes, it is not
cider vinegar. To confirm the presence of malic acid 5 c.c. of the
vinegar is treated with 1 c.c. of 10 per cent CaCl.^ solution ; filter and
add to the filtrate about 3 volumes of alcohol 95 per cent. In the
presence of malic acid a flocculent precipitate will occur. Dextrin is
also thus precipitated by alcohol, but its presence will be indicated by
a dextro-rotation on the polarimetric test. Sulphate also will give a
precipitate of CaSO^. If the alcohol precipitate be collected, dried,
dissolved in HNOg, evaporated to dryness on the water bath, the cal-
cium malate is converted into oxalate, which may be decomposed
with Na.^COg by boiling, acidified with HC.^HgO^,, and precipitated with
Ca SO^. The last reagent is employed to prevent precipitation of any
sulphates present as CaSO^.

The value known as Winton's lead number (see under Vanilla) is
sometimes determined on cider vinegars, as it often shows a consider-
able variation from the value obtained with malt vinegars. The
average values are, according to Bailey, as follows : —

Cider vinegar 0-075 to 0-290

Malt vinegar 0-158 „ 0-548

Malt Vinegars. — The characteristics of malt vinegar are its com-
paratively high specific gravity, which averages above 1-020, its high
extractives, containing much nitrogen and phosphoric acid. The
water used for brewing malt vinegar often contains a high amount of
sulphates and chlorides, so that no attention need be paid to reac-
tions with barium chloride or silver nitrate.

Sugar Vinegar .—This vinegar is usually prepared from inverted
starchy matter, glucose being the starting-point of the fermentation.
This glucose vinegar generally contains dextrose, dextrin, and very
frequently, calcium sulphate, in notable quantity. It therefore usually
reduces Fehling's solution, and gives abundant precipitates with
ammonium oxalate and barium chloride. To detect dextrin, the vine-

FLAVOUBING ESSENCES. 257

gar should be evaporated to about one-fifth of its bulk and then three to
four times its volume of absolute alcohol added. Dextrin is precipi-
tated in this manner. Dextrose remains in the filtrate, which is
decolorized by boiling with animal charcoal, and the d*extrose esti-
mated by reduction of Fehling's solution, after the alcohol is boiled
otf. The optical rotation may be observed by concentrating the vine-
gar to half its bulk, clearing with lead acetate (see under cider vine-
gar) and filtering. It is usually, however, too small to give any
decided results.

The above details will enable the analyst to differentiate, in.
general, between genuine brewed vinegars and diluted acetic acid
coloured with caramel.

FLAVOURING ESSENCES.

Essence of Almonds. — Essence of almonds as sold to the public is
a dilute alcoholic solution of essential oil of almonds, usually contain-
ing 1 to 2 per cent of the essential oil. The oil itself is used in phar-
macy, but is not official in the British Pharmacopoeia. Two varieties
are recognized, one, the natural oil, the other deprived of its prussic
acid, when it is then known as " 01. Amygdalae essent. S. A. P." {sine
acid prussic).

Oil of Bitter Almonds. — The true bitter almond oil is obtained by
distilling the seeds (almonds) of Amygdalus communis var. amara
with water after the fixed oil has been extracted by expression. The
kernels of the apricot and peach yield essential oils practically
identical with that from the almond, and the greater part of the bitter
almond oil of commerce, especially the foreign oil, is obtained from
the former {Prunus Armeniaca). The following remarks may be
taken as generally applicable to all three oils : —

Bitter almond oil does not exist as such ready formed in the seeds
(almonds, kernels). It results from the action of water on the glu-
coside amygdalin, under the influence of the natural ferment emulsin
present in the seeds. The reaction taking place is expressed by the.
following equation : —

C,oH,,NOii + 2H2O = C^H.O + 2G,B.,,0^ + HCN.

Amygdalin. Benzaldehyde. Dextrose.

Amygdalin, taking up two molecules of water, yields benzaldehyde,.
dextrose, and hydrocyanic acid. Amygdalin is a crystalline body,
without any smell of the bitter almond, and does not yield the oil ex-
cept under the influence of a hydrolytic agent, such as the natural
ferment emulsin, or by boiling with dilute acids. The action of the
ferment is destroyed by heat or by warm alcohol. Hence if dried
and powdered bitter almonds are shaken with boiling water and
distilled, no oil is obtained. After the fixed oil has been expressed
the press-cakes are ground up and soaked for about twenty-four hours-
in twice their weight of water, to which a quantity of salt is usually
added. The whole is then subjected to distillation. Some trouble,
however, is experienced during the process, as the large quantity of
VOL. I. 17

258 FOOD AND DEUGS.

albuminoids present causes excessive frothing. To remedy this, the
press-cakes are coarsely powdered and at once immersed in boiling
water to coagulate the albuminoids and dissolve the amygdalin. The
emulsin is, or course, rendered inactive, so that on cooling a quantity
of emulsion of the fresh cake in cold water is added to the previously
treated mass. This is allowed to stand, until the emulsin will have
converted the whole of the amygdalin into essential oil. The mixture
is now distilled. As hydrocyanic acid is a very deadly substance, it
is necessary to use great care that none of the vapour is allowed to
escape into the air. The oil of almonds so obtained contains a con-
siderable amount of hydrocyanic acid, the remainder being principally
benzaldehyde, C^-Hr, . GOH. The absolutely natural oil is a regular
commercial article, but much is deprived of its hydrocyanic acid before
being sold, and is then listed as " Oil of Almonds (S.A.P.)".

Oil of almonds also contains the nitrile of mandelic acid, and a
trace of benzoin. Hydrocyanic acid is detected by adding to water
which has been well shaken with the oil a little ferric and ferrous chloride
or sulphate and solution of caustic potash. On acidifying the liquid
a blue-green colour or precipitate is formed in the presence of hydro-
cyanic acid (ferro- and ferri-cyanides). It may be estimated by dis-
solving 1 grm. of the oil in 5 c.c. of alcohol and adding 50 c.c. of
water. Ammonia silver nitrate solution is then added and the whole
well shaken. The solution is then acidified with nitric acid and the
silver cyanide washed, dried, and ignited.

The ignited preciiDitate corresponds to 25 per cent of its weight
of HON.

Natural almond oil is, in the crude state, a yellowish liquid, but
is white when rectified, of specific gravity 1-045 to 1*070, but usually
from 1'045 to 1-055. Its refractive index is about 1*5450, or if free
from HON, 1*550. It is optically inactive. It is often adulterated
with artificial benzaldehyde, and if the purest variety be used it is im-
possible, within certain limits, to detect it, except, possibly by the
nose. If the cheaper variety has been used, chlorine compounds will
be present and may be detected as follows : A piece of filter paper
is saturated with the oil and placed on a small porcelain dish stand-
ing in a larger one. A large beaker whose sides are moistened with
•distilled water is stood over the smaller dish, the paper having been
.set alight. The gases generated by the combustion are, to a certain
extent, absorbed by the water on the sides of the beaker, which is
rinsed out with a little more distilled water. The liquid is filtered,
;and one drop of nitric acid and a few drops of solution of silver nitrate
.are added to the filtrate ; the formation of insoluble silver chloride is
strong evidence that artificial benzaldehyde is present.

Another adulterant, of a much grosser character, is oil of mirbane.
This is the cheap almond oil substitute so largely used for perfuming
cheap toilet soaps. Chemically it is nitrobenzene, CgH.NOg, more or
less mixed with impurities, of which the most common is nitro-
toluene, which sometimes itself forms the greater part of cheap nitro-
benzene. Indeed, nitrotoluene in any great quantity may be regarded
as an adulterant of nitrobenzene. The latter, when pure, is a yellowish

FLAVOURING ESSENCES. 259

liquid of specific gravity at 0 °of 1-200, boiling at about 206°, and solidify-
ing at + 2"" to + 3°. It has a coarse almond-like odour, and is poisonous
when taken internally, and irritating to the skin when used externally.
The cheapness of benzaldehyde should discourage the use of mirbane
in even the cheapest toilet soaps, Nitrotoluene, CgH4(CH3)N02, ex-
ists in three isomeric modifications, and nitroxylene, (Cj.H3)(CH3).2N02,
in more still. It is these bodies which are found to a considerable
extent in the cheaper qualities of nitrobenzene. Consequently it is
important that commercial samples should have physical characters
in approximate agreement with those above quoted.

To detect the presence of this objectionable substituent in oil of
almonds a little of the oil is warmed with iron filings and acetic acid.
The nitrobenzene is reduced to aniline, C^Hj^NHg, which is distilled off
and collected. To the distillate a few drops of solution of ordinary
chloride of lime is added. If aniline be present the liquid yields the
characteristic violet colour. Pure benzaldehyde combines with sodium
bisulphite to form a crystalline compound without the characteristic
almond odour. Samples adulterated with nitrobenzene, when shaken
with excess of bisulphite of sodium solution, so that the benzaldehyde
is entirely combined, then have the characteristic coarse nitrobenzene
odour.

To determine the benzaldehyde in diluted solutions of the oil such
as ordinary essence of almonds, the following process (due to Denher,
and elaborated by Denis and Dunbar) may be used : —

The reagents used are : —

Beagent (1) : Phenyl-hydrazine hydrochloride ... 2 grnis.
Crystals of sodium acetate . . . . 3 ,,
Water 20 c.c.

Dissolve the sodium acetate in the water, add the phenyl-hydrazine
hydrochloride, shake for five or six minutes, and filter, or

Reagent (2) : Phenyl-hydrazine 1 c.c.

Glacial acetic acid 1-5 c.c.

Water 20 c.c.

Mix the acetic acid and water, then pour in the phenyl-hydrazine.

Reagent (2) is much more convenient on account of the rapidity
with which it may be prepared. Whether reagents (1) or (2) be used
the solution should be made up immediately before use, and solutions
more than an hour old should be discarded. The method of precipi-
tation is as follows : —

Two 10 c.c. portions of almond essence are measured into 300 c.c.
Erlenmeyer flasks ; to one portion is added 10 c.c, t'o the other 15 c.c.
of either reagent ; shake, stopper tightly, and allow to stand over
night in a dark place. The next day add 200 c.c. of cold water to
each flask, and filter on tared Gooch crucibles provided with thin
mats of asbestos. Wash with cold water and finally with a 10 c.c.
of 10 per cent alcohol. Dry for three hours in a vacuum oven at 70°
to 80° C. If a vacuum oven is not available, the drying may be ac-
complished in a vacuum desiccator over sulphuric acid, but will of

260 FOOD AND DEUGS.

course take much longer than when a higher temperature is employed.
The weight of the precipitate multiphed by 5-408 gives the number of
grammes of benzaldehyde in 100 c.c. of the solution.

The reason for using 10 and 15 c.c. of reagent on different por-
tions of the same extract is based on the fact that while the large
majority contain in the neighbourhood of 1 per cent by volume of oil of
bitter almonds, occasionally extracts are met with containing as much
as 6 per cent of benzaldehyde ; in such preparations it is obvious that
while the use of 10 c.c. of the reagent may give good duplicates, the
results would be far below the truth.

Several other processes have been suggested, especially a colori-
metric process (" Journ. Amer. Chem. Soc." 1908, 1607), which
the author finds gives fairly accurate results. It is carried out as
follows : —

A solution of fuchsin-sulphurous acid is prepared fresh each time
in the following manner : 0*5 grm. fuchsin is dissolved in water, and
sulphurous acid introduced into the solution until the weight has in-
creased by 20 grms. when it is further diluted to make 1 litre. In the
next stage alcohol free from aldehyde is employed ; this is obtained
by taking spirit which has undergone a preliminary treatment with
oxide of silver, and diluting it with 25 grms. phenylene- diamine hydro-
chloride per litre of alcohol, then passing a strong current of air
through it for three hours, and distilling off, rejecting the first 100
c.c. Afresh standard solution of recently distilled benzaldehyde in
aldehyde-free alcohol (1 mg. in 1 c.c.) is prepared. Ten grms. of the
almond essence are then diluted to 50 c.c. with aldehyde-free alcohol ;
2 c.c. of this solution is placed in a colorimeter tube and diluted to 20
c.c. Three control-solutions containing respectively 2, 4, and 6 mgs.
benzaldehyde are then poured into tubes of equal size, all the tubes
are brought to a temperature of 15°, the contents quickly diluted with
20 c.c of the fuchsin-sulphurous acid solution, shaken up, and allowed
to stand for ten minutes. As much of the sample-solution is now
run off as will make its colour resemble that of one of the control-
solutions. As the colour is in proportion to the degree of concentra-
tion of the benzaldehyde, the proportion of that body which is present
may be calculated. The method gives accurate results, and is applic-
able also to bitter almond oil.

Essence of Lemon.

Essence of Lemon as sold to the public is usually an alcoholic solu-
tion of the essential oil of lemon, varying in strength from 5 to 30 per
cent. The essential oil itself is frequently known under the same
name.

Two varieties of essence of lemon are commonly met with ; firstly
that prepared by dissolving the essential oil in alcohol, secondly, that
made by using terpeneless lemon oil — the essential oil freed from
terpenes.

The examination of the essence of the shops is restricted to the
separation and examination of the essential oil.

FLAVOURING ESSENCES. 261

The oil is best separated by adding about 30 volumes of water
to one of the essence, and allowing the liquids to completely separate
in a separatory I funnel, and, if necessary, dry the oil over a little fused
acid potassium sulphate. It should then be examined as described
under oil of lemon. The optical activity of the essence will indicate
whether it has been made from ordinary oil of lemon or from the
terpeneless oil, as the former has an optical rotation of about + 60°,
and the latter , - 7°, or thereabouts.

Essential Oil of Lemon. — This oil is official in the British Phar-
macopoeia which describes it as the oil obtained from fresh lemon peel.
That authority requires it to have the following characters : —

Specific gravity, 0"857 to 0'860. Optical rotation not less than
+ 59°. The optical rotation of the first 10 per cent distilled should
have a rotation not differing to more than 2° from that of the original oil.

These characters, especially the last, are seriously incorrect. Pure
oils may have characters well outside these limits, especially in re-
gard to the optical rotation of the first 10 per cent distilled. This
depends so entirely on the form of the distillation flask and the rate
of distillation, that very varying results can be obtained from the
same oil. This is referred to later.

The well-defined hydrocarbons of lemon oil are the terpenes, limo-
nene, leevo-pinene, camphene, and the sesquiterpene limene ; limonene
forming about 90 per cent of the oil.

The oxygenated bodies forming the other 10 per cent are citral,
nonyl, and decyl aldehydes, geraniol and linalool and their acetic esters
(the latter stated to be only present in Palermo oils, and thus probably
accounting for the slight difference in odour between this and Mes-
sina oils), citraptene, G^Hj^O^, and a stearoptene of unknown con-
stitution.

Adulteration, which was until recently very frequent, is still
common. Turpentine was the regular adulterant, with, at times, the
poorer-quality distilled oil of lemons. But adulteration with turpen-
tine is now so easily detected that the sophistication is frequently
carried out in a more scientific manner.

Mixtures with the proper specific gravity and optical rotation can
easily be made up from turpentine and orange oil — the poorer qualities
of the latter of course being used — and such mixtures are often
used to adulterate the oil.

But the most formidable adulterant from the analyst's point of
view is one that has only come into vogue during the last few years,
viz. the terpenes obtained in manufacturing the " terpeneless " or con-
centrated oil of lemon and oil of orange, the latter being sometimes
added to turpentine to raise the optical rotation.

The terpenes are sometimes used alone, sometimes together with
a little citral obtained from lemon-grass oil. Before discussing the
analysis of lemon oil, a few words on the citral content will not be
out of place. It is common custom to export oil of lemons with a
guaranteed citral content, and to sell it upon that basis. In judging of
the value of such a basis for market value of the oil, the following
points should be carefully noted : —

262 FOOD AND DRUGS.

1. The value of the oil depends on its percentage of oxygenated
constituents, which are soluble in weak alcohol.

2. The terpenes are practically odourless and insoluble, therefore
valueless for the general purposes for which lemon oil is employed.

3. The percentage of terpeneless oil obtained by careful fraction-
ation is an indication of the value of the oil.

It would be more rational in valuing this oil not to give the per-
centage of citral ^which might be, in fact often is, added as lemon-grass
citral) but to return the amount of terpeneless oil actually obtainable
by careful fractionation. It has been found in practice that genuine
oils only yield from 5 to 6 per cent of terpeneless oil, which in the
strict sense of the term are not terpeneless but contain a fair propor-
tion of sesquiterpenes ; however, such oils will not contain more than
50 per cent of total aldehydes, and therefore the amount of citral cal-
culated on the original oil would be 4 per cent. No lemon oil ever
contains anything like 7 per cent of citral, which is a figure given by
Messina analysts.

The specific gravity of a pure lemon oil should be between 857
and 862. These limits are rarely exceeded. The optical rotation
taken at 20° should not be less than 4- 57°, and is usually between
this figure and + 64°. Increase in temperature causes a slight diminu-
tion in rotatory power, but this does not amount to more than about
- 8' or - 9' per degree rise in the temperature of the oil. The refrac-
tive index for pure oils lies within the comparatively narrow limits
of 1-4748 to 1-4754. Pure lemon oil commences to boil at 170° to
172° and from 20 to 30 per cent will be obtained from 172° to 174°.
However, the percentage of the fractions at temperatures near to one
another are so dependent on the exact form of the distillation flask
that the results are not constant enough to be of much value unless
any oil distils below 170°.

To determine the purity or otherwise of lemon oil a somewhat
prolonged analysis is necessary, no one constant being of much value
by itself.

The following scheme will, however, detect any adulteration : —

1. Determine the specific gravity at 15° C.

2. Optical rotation (100 mm. tube).

3. Refractive index, at 20° C.

4. Fractionally distil as follows : —

Introduce into a distilling flask having three bulbs blown in the
neck 100 c.c. of the oil to be tested. The receiver is an ordinary
Bruhl apparatus with two vessels graduated at 10 and 80 c.c. respec-
tively. The whole apparatus is then exhausted by means of a water
pump (or other suitable means), and when a pressure of not more
than 20 mm. is obtained, as shown on a gauge, the distillation is
commenced by gently heating the oil-bath containing the flask.

The first fraction is collected in the 10 c.c. tube and the second
in the 80 c.c. flask. Directly the second fraction is collected the
pressure is released, and the distillation continued by passing a
current of steam into the distillation flask and collecting the oil and
water into a suitable vessel. When about 200 c.c. of water have been

FLAVOURING ESSENCES. 263

collectad the distillation is stopped ; the oil is then separated from the
water and carefully measured.

The rotation and refractive index of the three fractions are then
carefully determined, and- further with No. 3 the aldehydes are estimated
by absorption with acid sulphite of soda.

Interpretation of results : —

Fraction 1 will indicate the addition of turpentine ; there should
not be a greater difference than 8° between the rotation of this and
of the original oil, 5° to 6° being the average for normal oils.

Note. — Pineneis a natural constituent of lemon oil, but only in traces.

The refractive index will be about 2 points in the third place of
decimals lower than that of the original oil. This fraction will also
indicate substances of low boiling-point.

Fraction 2 will indicate chiefly the addition of added terpenes,
normal oils showing an increase of 6° to 7° rotation from the original
oil, whereas with added terpenes the increase will be considerably
higher and the refractive index lower.

Fraction 3 is in many respects the most important, inasmuch that
it indicates the true proportion of oxygenated constituents of the oil
and therefore the strength of the oil. The rotation will depend partly
on the amount of oil obtained by the steam distillation which is usu-
ally between 6-5 to 7*5 c.c, the rotation being about + 2° to + 14°, but
in some very rich oils the sign may be minus.

The refractive index will also have increased to nearly 1-4800 in
this fraction. The aldehydic content should be in direct ratio to the
amount of the fraction from 36 to 46 per cent.

It must be remembered that this method does not give the actual
amount of citral present in the oil, but only that obtainable in the
only practicable method of manufacturing terpeneless oil of lemon.
There are many methods of determining citral suggested, none of
which are very satisfactory. Of those, the following gives the most
exact results : —

The analysis is conducted as follows : —

Twenty c.c. of lemon oil are mixed with 20 c.c. of ^ solution of

hydroxylamine hydrochloride in 80 per cent alcohol, and to the mix-

N
ture is added about 8 c.c. of y alcoholic potash and 20 c.c. of strong

alcohol (which is sufficient to procure complete solution when hot).

The mixture is boiled gently under a reflux condenser for half an

hour, and then allowed to cool. The condenser is washed down,

and ihe contents of the flask diluted with about 250 c.c. of water, and

N
neutralized to phenol-phthalein. The liquid is then titrated with — •

sulphuric acid, using methyl orange as indicator. The number of
c.c. of acid required, subtracted from the number used in a blank ex-
periment, in which no lemon oil is present, gives the amount of
hydroxylamine which has entered into reaction with the citral, and
multiplied by 0*076 gives the weight of citral.

264 ' FOOD AND DKUGS.

When the titration of methyl orange is performed in the usual
way with addition of a drop of the indicator to the solution, the end
point is often not very satisfactory. Much sharper results are
obtained by making use of drops of a very dilute aqueous solution of
methyl orange scattered on a white plate. When drops of the solu-
tion which is being titrated are brought into contact with these the
change of colour when neutralization is complete is well marked.

Hiltner proposes (" J. Ind. Eng. Chem." 1909, 1, 798-800, through
"J. Soc. Chem. Ind." 1910, 29, 172), a method of citral determina-
tion based on the fact that a 1 per cent solution of m-phenylenedia-
mine hydrochloride in 50 per cent alcohol gives a clear yellow
coloration with citral at the ordinary temperature. The determina-
tion is performed in the usual manner, the solution being made up to
volume with 90 to 95 per cent alcohol in the case of lemon extracts
and with 50 to 60 per cent alcohol in the case of terpeneless ex-
tracts. He claims that this method is more accurate than the
magenta sulphurous acid process since under the experimental con-
ditions the reagent gives no coloration with acetaldehyde or citron-
ellal. The method, however, fails in the case of lemon oil which
has become altered by oxidation, the coloration produced varying from
yellowish-green to greenish-blue, according to the degree of oxidation
of the oil.

Chace (" U. S. Dept. of Agriculture, Bureau of Chemistry," Circular
No. 46, 1909) has published the following method, which he claims
will detect trace3 of pinene in lemon oil, and thus indicate adulteration
with turpentine.

Fifty c.c. of the oil is distilled and is then tested for pinene as
follows : —

The distillate is mixed with an equal volume of glacial acetic acid
in a 2-ounce Erlenmeyer flask, and immersed in a freezing mixture.
Ten cubic centimetres of ethyl nitrite are next added and finally,
slowly with constant stirring, 2 c.c. of a mixture of two parts of con-
centrated hydrochloric acid and 1 part of water, all previously cooled.
The whole is allowed to remain fifteen minutes in the freezing bath,
then rapidly filtered on a Gooch crucible provided with a filter paper
disc, using a vacuum. The resulting crystals of nitroso-cbloride of
limonene are dis-olved in the smallest possible amount of chloroform
and reprecipitated with methyl alcohol. After filtering off these
crystals, they are mounted with olive oil and examined under the
microscope, using a magnification of 100. (See Fig. 34.) If pre-
sent, pinene nitroso-cbloride is easily detected by its comparatively
broad crystals having irregular pyramidal ends, limonene nitroso-
cbloride crystallizing in needle shapes or columns.

This method has been very severely criticized by the author, and
the consensus of opinion is that a positive reaction cannot be taken
as any proof of adulteration of lemon oil, on account of the small,
varying amount of pinene naturally present in lemon oil.

Further, the published results of such authorities as Wallach, in
reference to the crystallographic characters of the nitroso-chlorides, do
not support the contentions raised by Chace.

FLAVOURING ESSENCES.

265

•T" For fuller details of this subject " The Chemistry of Essential Oils "
(by the author) should be consulted.

Fig. 34. — Photomicrograph of crystals from lemon oil (x 100). a, 6, Limonene
nitroso-chloride crystals from lemon oil ; c, limonene and pinene nitroso-
chloride crystals from a lemon oil mixed with 5 per cent of turpentine ; d,
pinene nitroso-chloride crystals from turpentme.

Vanilla.

Vanilla is the fruit of an orchid, Vanilla planifolia, and other
•closely allied species, grown chiefly in Mexico, Bourbon, Tahiti, and
the Seychelles. The fruits are long pods varying in length from about
3 to 8 or 9 inches, and differing materially in odour according to the
country in which they are grown.

Vanilla owes its value chiefly to the presence of the odorous body
vanillin — small quantities of other odorous bodies, of course, being
present.

The fruits, or beans, as they are called, are judged by their appear-
ance and odour, and do not often come, as such, before the analyst.
The only forms of adulteration possible are the addition of exhausted
beans, or the apparent improvement in appearance of inferior beans
by coating them with crystals — usually of benzoic acid. The most
valued kinds of vanilla are incrusted with fine crystals of vanillin —
which gradually find their way to the surface of the bean from the
interior.

The following analyses of vanilla beans are due to Konig: —

266

FOOD AND DRUGS.

Water

Nitrogenous matter ....
Fat and wax .....

Reducing sugars

Non- nitrogenous extractives

Cellulose

Ash

Per cent

25-85

4-87

6-74

7-07

30-50

19-60

4-73

Per cent

30-94

2-56

4-68

9-12

32-90

15-27

4-53

Vanillin is a methyl-protocatechuic aldehyde, GgHg(C0H)(0CH3)-
(OH), forming small white crystals melting at 81° to 82°, and possess-
ing an intense vanilla odour. It is now produced synthetically in
enormous quantities, and has largely replaced the natural vanilla bean
as a flavour. A very large amount of the chocolate of commerce is
flavoured with artificial vanillin. It is prepared in many different
ways, the principal of which is by the oxidation of eugenol, the prin-
cipal constituent of oil of cloves.

The eugenol is first acetylated by means of acetic anhydride, and
the resulting acet-eugenol is dissolved in acetic acid and oxidized with
permanganate of potassium. The liquid is then filtered, and rendered
alkaline, and the whole is then evaporated, and the residue treated
with moderately dilute acid, and extracted with ether. The ethereal
solution is extracted with a solution of sodium bisulphite, which com-
bines with the vanillin. The double sulphite compound is decomposed
with dilute sulphuric acid, and the vanillin is extracted with ether,
from which solvent it is obtained in fine white crystals.

The best yield, however, is obtained by first converting the eugenol
into iso-eugenol OH . OCH3 . CyHg . CH : CH . CHg by treating it with
solution of potassium hydrate. The acetylation product is oxidized,
by which acetyl- vanillin is chiefly formed, which yields vanillin by
splitting off the acetyl group.

Some of the cheaper commercial samples are heavily adulterated
with acetanilide. The effect of this body is to lower the melting-
point even if present in large quantity, but it is very easily detected,
as by boiling with solution of potash, aniline is formed, which is
easily detected by any of the usual reactions. A quantitative separa-
tion may be effected as follows : the substance is dissolved in ether
and the liquid repeatedly shaken with concentrated solution of sodium
bisulphite. The vanillin is thus extracted, and the ether, alter being
washed twice with water, is allowed to evaporate, when the acetani-
lide remains. This will then be found to have a melting-point close
to 113°. Benzoic acid and coumarin are also occasional adulterants
of vanillin. A little isovanillin C^H3(CHO)i(OH)3(OCH3)-^ is occasion-
ally present, but this is due to the fact that it is generally formed in
small quantity with vanillin, in many reactions.

Sugar is occasionally found as an adulterant, but is easily detected
by its insolubility in ether : after extraction with the vanillin it can
be dissolved in water and recognized by any of the usual tests.

FLAVOUEING ESSENCES. 267

Acet-iso-eugenol, one of the intermediate bodies in the manu-
facture of vanillin is sometimes found ; it lowers the melting-point of
the sample, yields acetic acid in hydrolysis, and gives a fine red
colour with strong sulphuric acid, whereas pure vanillin only gives a
lemon-yellow colour. Benzoic acid is also found as an adulterant.
This is easily detected by the high acid value of the substance (vanillin
is neutral), and by dissolving the sample in ether, extracting the
vanillin by means of sodium bisulphite solution, and neutralizing the
residue from the ethereal solution with potash, dissolving it in water,
and testing it with a neutral solution of ferric chloride, when red
ferric benzoate is precipitated.

In examining vanilla beans the determination of the vanillin is a
matter of importance. Busse recommends the following process for
the determination : 20 grms. of the pods, crushed with sand, are ex-
hausted, with ether in a Soxhlet tube, and the ethereal extract is-
shaken out with 20 per cent sodium bisulphite solution. From the
latter, vanillin is removed by treatment with dilute H.^SO^, the SOg
generated removed by a current of COo, and the vanillin extracted by
shaking out with ether, evaporating the solvent and weighing the
residue. In East African vanilla the author found 2*16 per cent of
vanillin, in that from Ceylon 1'48 per cent, and in Tahiti vanilla
from 1-55 to 2*02 per cent. Tiemann and Haarman found in the
best Bourbon vanilla 1'94 to 2-90 per cent, in the best Java vanilla.
2*75 per cent and in Mexican vanilla from 1'7 to 1*9 per cent.
Tahiti vanilla sometimes contains less than 1 per cent of vanilla.

In suspected cases the crystals on the beans should be carefully
separated and examined for benzoic acid as above described.

Hanus (" Analyst," xxv. 318) recommends that yS-naphthyl hydra-
zine hydrochloride should be added to the solution of vanillin in such
proportion that from two to three parts are present for each part of
vanillin. After standing for five hours the precipitate is transferred
to a tared filter, washed with hot water until the washings no longer
precipitate silver nitrate, dried at 90° and weighed. The weight of
the hydrazine formed, divided by 1*92 gives that of the vanillin pre-
sent. This method is available in all cases where an aqueous solution
of the vanillin can be prepared.

Hanus has more recently recommended the following method for
the determination of vanillin in vanilla beans and in preparations
thereof (" Pharm. Zeit." 50, 1022, 157). Three grms. of the crushed
pods are extracted for three hours in a Soxblet tube with ether, the
solvent distilled off cautiously, and the residue dissolved in a little
ether, the solution filtered and the filtrate evaporated cautiously. The
residue is treated with 50 c.c. of water at 60° on a water bath : 0-25-
grm. of me^a-nitrobenzhydrazide is then added to the aqueous solution
in a stoppered flask, which is kept for two to three hours at 60°, and
then set aside with occasional shaking for twenty-four hours. The
vanillin is precipitated quantitatively as vanillin-me^a-nitrobenzhydra-
zone, NO.^ . C.H^ . NH^ . N : CH . C,;H3(OCH3) . OH. The precipitate
is washed with three successive quantities of petroleum ether to re-
move fat, then washed with water, and then again with petroleum

268 FOOD AND DRUGS.

ether, and then dried at 100° for two hours. The weight, multipUed
by 0"4:829 gives the amount of vanillin present. Preparations of
vanillin are treated similarly, alcohol being removed by evaporation.
The presence of other aldehydes, such as heliotropin, of course, will
vitiate the results.

Essence of Vanilla. — The substance sold under this name is, pro-
perly, a spirituous extract of the vanilla bean. Many samples, how-
ever, are little more than alcoholic solutions of artificial vanillin,
coloured with caramel. Some samples, which cannot be described as
adulterated, contain a little coumarin or other odorous substance,
added to vary the characteristic vanillin odour and flavour somewhat.

A genuine extract can be recognized by the fact that it contains
some dark red or red-brown resin, soluble in 50 per cent alcohol, but
precipitated on further dilution.

Coumarin, or extract of Tonka beans, which contain coumarin,
may be detected as follows : a small quantity of the essence is eva-
porated to dryness, the residue fused with caustic potash, saturated
with hydrochloric acid and treated with a drop of ferric chloride solu-
tion. If coumarin be present, a violet colour, due to the formation of
salicylic acid, will be produced.

Winton and Silverman (" Jour. Amer. Chem. Soc." 2% 1128) re-
commend the following methods for examining essence of vanilla : —

De-alcoholize 25 grms. of the extract in an evaporating dish upon a
water bath, at a temperature of about 80° C, adding water from time
to time to retain the original volume. After removal of the alcohol,
add normal lead acetate solution, drop by drop, until no more pre-
cipitate forms. Stir to facilitate flocculation of the precipitate, filter
through a moistened filter, and wash three times with a few c.c. of
hot water. Cool the filtrate and extract with ether by shaking out in
a separator. Use 15 c.c. to 20 c.c. of ether at each separation, re-
peating the process three or four times, or until a few drops of the
ether, evaporated upon a watch glass, leaves no residue. Place the
combined ether extracts containing all of the vanillin and coumarin
in a clean separator, and shake out four or five times with 2 per cent
ammonia, using 10 c.c. for the first, and 5 c.c. for each subsequent
shaking.

Set aside the combined ammoniacal solutions for the determination
of vanillin.

Wash the ether solution into a weighed^dish, and allow it to evapor-
ate at the room temperature. Dry in a desiccator and weigh. Usually
the dried residue is pure coumarin. Treat the residue with 5 c.c. to
10 c.c. of cold petroleum ether, boiling between 30° C. and 40° C, and
■decant off the cl^ar liquid into a beaker. Repeat the extraction with
petroleum ether until a drop, evaporated on a watch glass, leaves no
residue. Dry the dish for a few moments in a water oven, cool and
weigh. Subtract the weight of the dish and the residue (if any) from
the weight previously obtained after evaporation with ether, thus
obtaining the weight of pure coumarin. Allow the petroleum ether to
evaporate at the room temperature, and dry, if necessary, in a desic-
cator. The residue should be crystalline and have a melting-point of

FLAVOUKING ESSENCES. 269

67° C. This, with the characteristic odour of coumarin, is sufficient
for its identification.

Slightly acidulate the reserved ammoniacal solution of vanillin
with 10 per cent hydrochloric acid. Cool and shake out in a
separatory funnel with four portions of ether of about 15 c c. to 20
c.c. each. Evaporate the ether at room temperature in a weighed
platinum dish, dry over sulphuric acid, and weigh. Treat the residue
with boiling petroleum ether (boiling-point 80°), decanting into a dry
beaker. Kepeat the treatment until all vanillin is removed. Dry the
dish and residue (if any) for a few moments at 100"" C. and weigh ;
deduct the weight from the weight of the ether residue. The differ-
ence is the weight of the vanillin. Evaporate the petroleum ether at
ordinary temperatures, and dry in a desiccator. The residue should
be crystalline, and melt at 80° C. to 81° C.

Tests for Caramel. — Valuable indications of the nature of an
extract are obtained in the process of determination of vanillin and
coumarin. Pure extracts of vanilla beans give, with lead acetate, a
bulky, more or less glutinous, brown-grey precipitate, and a yellow
or straw-coloured filtrate, whereas purely artificial extracts coloured
with caramel give a slight dark brown precipitate and a dark brown
filtrate. If both vanilla bean extract and caramel are present the
precipitate is more or less bulky and dark-coloured, and the filtrate
is more or less brown. The solution remaining after extraction of
the vanillin and coumarin with ether, if dark- coloured, should be
tested for caramel.

The most satisfactory test for caramel is to shake with fuller's-
earth, as recommended by Crampton and Simons. If the colour is
due to caramel and a grade of fuller's earth is used, which experience
has proved suitable, the solution, after filtering, is yellow or colourless.
This test does not positively identify the colour, as some other brown
substances may give similar reactions, but no such substance is liable
to be present in vanilla extract.

Winton and Bailey determine vanillin, coumarin and acetanilide
(which is sometimes found as an adulterant of artificial vanillin, and
therefore indicates its presence), in the following manner, which is a
modification of the method devised by Hess and Prescott (" Jour.
Amer. Chem. Soc." 1899, 256) :—

Twenty- five grms. of the essence are weighed into a 200 c.c.
beaker, marked to indicate volumes of 25 c.c. and 50 c.c. The es-
sence is diluted with water to 50 c.c. and evaporated on a water bath
to 25 c.c. at a temperature not exceeding 70°. It is now again
diluted to 50 c.c, and evaporated to 25 c.c. Solution of acetate of
lead is then added until no further precipitation takes place. The
liquid is then, after being well stirred, filtered through a moistened
filter paper, and washed three times with hot water, so that the total
filtrate does not exceed 50 c.c. The filtrate, when cold, is shaken
with 20 c.c. of ether in a separator. The ether is separated, and thfe
liquid extracted with three further portions of 15 c.c. of ether. The
combined ether extracts are then shaken with 10 c.c. of 2 per cent
ammonia solution and with three subsequent portions of 5 c.c. The-

^70 FOOD AND DEUGS.

ethereal solution is reserved (B) andithe combined amraoniacal solutions
are rendered slightly acid with 10 per cent hydrochloric acid. The
liquid is then extracted four times with ether, and the ether evapor-
ated and the residue dried at room temperature, and finally in a desic-
cator and weighed (A). If acetanilide is absent, this may be taken
as pure vanillin, which should melt at 79" to 81°. If acetanilide has
been detected {vide infra), the residue should be dissolved in 15 c.c. of
10 per cent ammonia, and the liquid shaken twice with ether. The
ether, on evaporation, will leave a residue of acetanilide, which is dried
at room temperature and then in a desiccator and the weight de-
ducted from the " vanillin " (A) previously weighed. The total amount
of acetanilide is the amount thus obtained, together with that present
in the ethereal solution (B) reserved above. The latter is transferred
to a tared dish and the ether allowed to evaporate at room temperature.
The residue is dried in a desiccator and weighed. It is then extracted
several times by stirriog well with petroleum ether, which is decanted
each time. If the residue is thus completely dissolved, it may be
taken to be entirely coumarin. Any undissolved residue is probably
acetanilide (melting-point 112° to 113°) and its weight deducted from
the total residue gives the coumarin.

The acetanilide here found is added to the amount extracted with
the vanillin to give the total amount present.

The presence of acetanilide in these residues may be confirmed by
boiling the residue for two to three minutes with HCl, and when cool,
adding a few drops of 0*5 per cent of chlorinated lime solution, in
such a manner that the liquids do not mix. A fine blue colour re-
sults if acetanilide be present.

Commercial essence of vanilla is usually made with about 5 per cent
of vanillas, the menstruum varying in strength from 40 to 50 per cent
alcohol in the best varieties. Sugar is sometimes added, but not
always. The average vanillin content is 0*1 to 0-2. Much higher
values than these indicate the presence of synthetic vanillin.

Jackson and McGeorge have determined the lead number (see
p. 272) of vanillas (calculated to 100 grms. of the beans) and consider
that for a given bean, this value is constant irrespective of the strength
of the alcohol used to extract the vanillas, so long as not more than
10 per cent of bean be used for the essence.

The table at top of opposite page represents their analytical
results.

It is thus noticed that the "Lead Number" is practically a con-
stant for any 'particular bean regardless of the strength of alcohol
used in the percolation, which would indicate that practically all of
the bodies precipitated by basic lead actetate are removed by any
strength of alcohol likely to be used, and that the "Lead Numbers"
oould be used as a measure of the quantity of vanilla beans used in
the manufacture of an extract.

FLAVOUEING ESSENCES.

271

Bean.

Prime :

Mexican
Mexican :

Cuts .

Pri-ne :
Mexican

Ordinary :
Mexican

Mexican

Cuts

Bourbon

South American

Vanillons

Per cent Alcohol.

50
50
50

65
20
30
40
65

20
30
40
65

20
30
40
20
30
40
65
20
30
40
65
20

Lead Number.

Duplicate.

1-62

1-82

1-91

—

1-90

—

1-60

1-64

1-63

1-68

1-66

1-65

1-80

1-82

1-80

1-81

1-83

1-80

2-00

203

2-01

2-10

2-09

1-56

1-55

1-59

1-68

1-60

1-62

1-62^

1-42

1-48

1-47 f

1-51

149

1-02

1-02'

To test this point three extracts were made up as before, but with
5, 10 and 15 grms. of beans to each 100 c.c. The results follow: —

Le\d Numbers.

Bean.

Experiment.

Per cent Alcohol.

Lead Number.

^Duplicate.

Grms. in 100 c.c.

Bourbon

>»

A
B
C

65
65
65

0-84
1-68
1-79

0-88
1-68
1-80

5
10
15

Experiments A, B and G seem to indicate that the " Lead Num-
ber " is a measure of the quantity of beans used when this does not
exceed 10 grms. in 100 c.c.

Although, as indicated above, there is no standard in this country
for essence of vanilla, it may be regarded as certain that, if an essence
gives no precipitate with a solution of lead acetate, it is made entirely
from artificial vanillin, and contains no natural vanilla at all.

Winton and LotD {" U.S. Dept. Agricul. Bull.," 132, 1910) state that
if normal lead acetate solution be used instead of the basic acetate the

272

FOOD AND DRUGS.

lead numbers are lower. For artificial essences, they find values be-
tween 0'03 and 017 ; for natural essences, between 0-29 and 0*34,
but the last-named figures must depend entirely on the strength of the
essence. To determine the lead number they dilute 50 gr. of essence
to 80 c c. with water, evaporate to 50 c.c, add 30 c.c. of water, again
evaporate to 50 c.c. and add 25 c.c. of an 8 per cent normal lead ace-
tate solution. The whole is diluted to 100 c.c , allowed to stand
several hours and filtered. To 10 c.c. of the filtrate, 25 c.c. of water,
excess of sulphuric acid and 100 c.c. of 95 per cent alcohol are added.
After sixteen hours, the PbS04 is collected and weighed. The lead
number is given by the formula

p ^ 100 X 0-6831(S - W) ^ ^3.gg2(g _ ^
5

where P is the lead number, S = gr. of PbSO^ obtained from 2*5 c.c.
of the lead acetate solution, and W the weight of PbSO^ obtained from
10 c.c. of the filtrate.

CHAPTER V.

ALCOHOLIC BEVEEAGES.

Befoee dealing with spirits, wines, and beer, it is necessary to deal
shortly with pure alcohol, which, of course, is never sold in the
ordinary way as a beverage, but is ofi&cial in the British Pharma-
copoeia. - All alcoholic beverages may be classed as products of fer-
mentation of saccharine matter. In the case of fruits, sugars exist
ready formed in their ^ juices, and any expressed fruit juice will
' rapidly commence to ferment when left to itself, since the microscopic
organisms responsible for the process are universally diffused. In
the cases of beverages made from cereals the first process is the con-
version of the starchy matter of the grain into sugars, by the action
of an unorganized ferment such as diastase, the active ferment of
malt. In the case of beers, sugars are often added to a mixture of
malted and unmalted grain, and in this country no objection is taken
to such additions. The manufacture of beers is under much greater
control than is the case with wines, so far as the fermentation pro-
cess is concerned, since in the former case there is a good deal of
selected yeast used, whilst in the latter, the yeasts mostly find their
way into the fruit juice, either from the air or from the skins of the
grapes. The greater part of the sugar present in fruits, especially in
the grape, is invert sugar, which is, of course, capable of direct fer-
mentation. The theoretical reaction usually expressed as represent-
ing ordinary fermentation processes is as follows : —

CgHigO^ = 2C2H6O + 2G0,

Six-carbon sugars = alcohol carbonic acid.

As a matter of fact, however, one obtains no more than 48*5 per
cent of alcohol and 46'5 per cent of carbonic acid, the remaining 5
per cent being glycerin, succinic acid and traces of higher alcohols
and esters, which may be termed the secondary products of alcoholic
fermentation.

Alcohol. — Alcohol, or ethyl alcohol CgHgOH, is, chemically, methyl-
carbinol. It is a colourless liquid with a pleasant odour, boiling at
78*3°, and of specific gravity 0*7939 at 15'5°. Alcohol and water are
miscible in all proportions with evolution of heat and a contraction
in volume. Alcohol obtained in as strong a form as can be made by
ordinary distillation is known as rectified spirits of wine, or "rectified
spirit," the latter being the official name of the British Pharmacopoeia,
which requires it to have a specific gravity 0"8340 and to contain
85'65 per cent of alcohol by weight ( = 90 per cent by volume).
VOL. I.— 18 (273)

274 FOOD AND DRUGS.

Proof spirit is an excise term, having its origin in the fact that in
olden days, the excise tested alcohol by pouring it on to a weighed
quantity of gunpowder. If it was above a certain strength, the gun-
powder ultimately exploded when a light was applied, but if it was
below that strength, the powder was too saturated with moisture to
ignite. Hence the terms "over proof" and "under proof". Proof
spirit is now defined by statute to mean of such density that at 51° F.
13 volumes shall weigh the same as 12 volumes of water. This spirit
has a specific gravity at lo"5 C. of 0'9198 and contains 49*24 per cent
of alcohol by weight' or 57"06 per cent by volume. The expression
" degrees under proof " and " degrees over proof " are very confusing
and rarely understood out of this country. If a liquid is 20° under
proof it is meant that it contains 80 volumes of proof spirit and 20
volumes oi water. If a liquid is spoken of as being 20° over proof it
contains so much alcohol that if a 100 volumes be diluted with water
to 120 volumes it will be of proof strength. Absolute alcohol is 75*25°
over proof. As nearly every determination of alcohol is made by
taking the specific gravity of the liquid containing (practically) nothing
but alcohol and water, it is necessary to have a table of reference from
which the values can at once be taken. A very elaborate table is in the
course of preparation by the excise authorities at the Government
laboratories, but is not available at the time of going to press. The
table on following pages is of a high degree of accuracy and gives the
percentage of alcohol by weight and by volume, and also the degrees
under and over proof of all mixtures of alcohol and water, of specific
gravities between 0*7938 and 1*000, with differences of 0*0005
throughout : —

Methylated spirit is alcohol suitably denatured so as to be unfit
for human consumption. In this country two varieties of methylated
spirit are known, both being allowed to be sold without an excise
duty. Of these, that which is used solely for manufacturing purposes
on the premises to which it has been delivered is of much less objec-
tionable odour than that which is resold from such premises. Wood
naphtha is the principal denaturant so that it is a matter of import-
;ance to examine certain alcoholic preparations for methyl alcohol,
ivhich would indicate the illegitimate use of duty free spirit.

The detection of alcohol is not a difiicult matter, and is a question
t)f some importance in the examination of temperance beverages The
following are the principal reliable tests for the detection of alcohol : —

(1) The iodoform test. This test will detect from 0*1 per cent to
0*2 per cent of alcohol. The liquid is warmed with a few drops of
concentrated solution of iodine in potassium iodide, and then enough
solution of caustic sodi to nearly decolorize. The characteristic odour
of iodoform will be developed, and on standing a slight yellow precipi-
tate of iodoform appears *. if the amount exceeds 0*3 per cent to 0*4 per
cent the precipitate is formed quickly. The iodoform crystals, when
examined with a powerful hand lens, will often show their crystalline
structure, and under the microscope they may be seen as star-shaped
groups or six-sided tablets. It must be remembered that several other
substances such as acetone and lactic acid yield this reaction.

ALCOHOLIC BEVEEAGES.

Alcohol Table.

Under Proof.

276

Sp. Gr.

Per

cent of

Per
cent of

Per

cent
under
Proof.

Sp. Gr.

Per

cent of

Per
cent of

Per

cent
under
Proof.

at
60° F.

Alcohol

by
Weight.

Alcohol

by
Volume.

at
60° F.

Alcohol

Weight.

Alcohol

Volume.

1-000

0-00

000

0-00

•9775

15-25

18-78

67-10

•9995

0-26

0-31

99^43

•9770

15-67

19-28

66-20

•9990

0^55

0-65

98^80

-9765

16-08

19-78

65-34

•9985

0^80

1-00

98^24

•9760

16-46

20-24

64-53

•9980

105

1-30

97^62

•9755

16-85

20-71

63-72

•9975

1-34

1^65

97-05

•9750

17-25

21-19

62 87

•9970

1-60

2^00

96 46

•9745

17-67

21-69

62-00

•9965

1-89

235

95-84

•9740

18-08

22-18

61-13

•9960

215

2-70

95-23

•9735

18-46

22-64

60-32

•9955

245

3-10

94 62

•9730

18-8^

23-10

59-52

•9950

2-75

3-50

93-98

-9725

19-25

23 58

58-67

•9945

302

3-82

9336

•9720

19-67

24 08

57-80

•9940

330

415

92-70

•9715

20-08

24-58

56-93

•9935

361

452

92-08

•9710

20-50

25-07

56-06

•9930

3-90

4-90

91-46

•9705

20 92

25-57

55-20

•9925

4-20

5-27

90-80

•9700

21-31

2604

54-37

•9920

4-50

5^65

90-12

-9695

21-69

26-49

53-57

•9915

4-82

6^02

89-45

•9690

22-08

2695

52-77

•9910

615

6-40

88-78

•9685

22-46

27-40

51-98

•9905

5^45

6-78

88-08

•9680

22-85

27-86

51-18

•9900

5-75

7-15

87-40

•9675

23-23

28-31

50-38

•9895

609

7-42

86-70

•9670

2362

28-77

49^60

•9890

6^40

800

85-98

-9665

2400

29-22

48-80

•9885

6^75

8^40

85-27

-9660

24-38

29^67

48-00

•9880

710

8^80

84-53

•9655

24 77

30-13

47-20

•9875

7-43

9^22

83-80

•9650

25-14

30-57

46-44

•9870

7^80

9^65

83-05

•9645

25-50

30-98

45-70

•9865

8^13

1010

82-30

-9640

25-86

31-40

44-97

•9860

8^50

10^55

81-54

•9635

26-20

31-80

44-27

•9855

8^84

10^97

80-77

•9630

26-53

32-19

43-60

•9850

9-20

11^40

79-99

•9625

26-87

32-58

42-90

•9845

9-56

11^87

79-22

•9620

27-21

32-98

42-20

•9840

■9-90

1235

78-45

-9615

27-57

33-39

41-47

•9835

1035

12-77

77-53

-9610

27-93

33-81

40-74

•9830

10^81

13-20

76-53

-9605

28-25

34-18

40-10

•9825

11^23

13^65

75-64

•9600

28-56

34-54

39-47

•9820

11-45

14^10

74-83

•9595

28-87

34-90

38-84

•9815

1200

14-60

7400

•9590

29-20

35-28

38-18

•9810

12-25

15-10

73-18

•9585

29-53

35-66

37-50

•9805

12-77

15-65

72-36

-9580

29-87

36-04

36-83

•9800

13-00

16-00

71-54

-9575

3017

36-39

36-23

•9795

13-54

16-70

70-73

-9570

30-44

36-70

35-68

•9790

13-92

17-17

69-90

•9565

30-72

37^02

35-13

•9785

14-36

17-70

68-97

•9560

31-00

37-34

34-57

•9780

14-82

18-25

68-00

•9555

31-31

37-69

33-95

276

FOOD AND DRUGS.

Alcohol Table.

Under Proof.

Sp. Gr.

Per
cent of

Per

cent of

Per

cent
under
Proof.

Sp. Gr.

Per
cent of

Per

cent of

Per
cent
under
Proof.

at
60° F.

Alcohol

by
Weight.

Alcohol
by

Volume.

at
60° C.

Alcohol

by
Weight.

Alcohol

by
Volume.

•9550

3162

38-04

33-32

•9370

4130

48-75

14^57

•9545

Sl-94:

38^40

32-70

-9365

41^55

49-02

1410

•9540

3225

38^75

32-08

-9360

41^80

49-29

13-63

•9535

32^56

39^11

31-46

•9355

42^05

49-55

13-16

•9530

32-87

39-47

30-84

•9350

42^29

49-81

12-70

•9525

3318

39-81

30-24

-9345

42^52

50-06

12-27

•9520

33-47

40-14

29-66

•9340

42-76

50-31

11-82

•9515

33^76

40-47

29-08

•9335

43-00

50-57

11-38

•9510

34^05

40-79

28-52

•9330

43 24

50-82

10-94

•9505

34^29

41-05

28-06

•9325

43-48

51-07

10-50

•9500

3452

41-32

27-60

•9320

43^71

51-32

1005

•9495

34^76

41^58

27-13

•9315

43^95

51^58

9-60

•9490

35^00

4r84

26-67

•9310

44^18

51^82

9-20

•9485

3525

42^12

26-20

9305

44-41

52^06

8-77

•9480

35-50

42^40

25-70

-9300

44-64

52^29

8-36

•9475

35-75

42^67

25-22

•9295

44-86

52^53

7-94

•9470

3600

42-95

24-74

•9290

45-09

52^77

7-52

•9465

36^28

43-26

24-20

•9285

45-32

5301

7-10

•9460

36-56

43-56

23-66

-9280

45-55

5324

6-70

•9455

36-83

43-87

23-12

-9275

45-77

53^48

6-27

•9450

37-11

44-18

22-58

/•9270

46^00

5372

5-86

•9445

37-39

44^49

2204

■9265

46^23

63^95

5-45

•9440

37-67

44-79

21-50

•9260

46-46

54^19

5-03

•9435

37-94

4510

20-96

•9255

46-68

54^43

4-62

•9430

38-22

45^41

20-43

•9250

46-91

54^66

4-20

•9425

38-50

45-71

19-89

•9245

47-14

64^90

3-80

•9420

38-78

46-02

19-36

•9240

47-36

55^13

3-38

•9415

39-05

46-32

18-83

-9235

47-59

5537

2 97

•9410

39^30

46-59

18-36

•9230

47^82

55^60

2-56

•9405

39^55

46-86

17-88

•9225

48^05

5583

2-15

•9400

39-80

47-13

17-40

•9220

48^27

56^07

1-74

•9395

4005

47-40

16-93

•9215

48-50

56^30

1-33

•9390

40-30

47-67

16-46

•9210

48-73

56-54

0^92

•9385

40-55

47-94

15-98

•9205

48-96

56-77

0-50

•9380

40-80

48-21

15-50

•9200

49-16

56-98

014

•9375

4105

48-48

1504

•9198

49-24

57^06

Proof

ALCOHOLIC BEVERAGES.

Alcohol Table.

Over Proof.

277

Per

cent

Per

cent

Sp. Gr.

cent of

Sp. Gr.

cent of

Alcohol

60° F.

Weight.

by
Volume.

over
Proof.

60° F.

by
Weight.

by
Volume.

over
Proof.

•9195

49^39

57.20

0-25

•8970

59-^9

67-11

17-61

•9190

49-64

57-45

0-68

8965

59-61

67^32

17-98

•9185

49-86

57-69

1-10

-8960

59-83

67-53

18-34

•9180

5009

57-92

1-51

-8955

60-04

67-73

18-70

•9175

50-30

58-14

1-89

-8950

60-26

67-93

1905

•9170

50-52

58-36

2-28

•8945

60-46

68-13

1939

•9165

50-74

58-58

2-66

•8940

60-67

68-33

19-74

•9160

50^96

58-80

3-05

-8935

60-88

68-52

20-08

•9155

51^17

59-01

3-41

•8930

61-08

68-72

20-42

•9150

51^38

59-22

3-78

•8925

61-29

68-91

20-77

•9145

51-58

59-43

4-14

•8920

61-50

69-11

21-11

•9140

51-79

59-63

4-50

•8915

61-71

69-30

21-45

•9135

52^00

59-84

4-87

-8910

61-92

6950

21-79

•9130

52-23

60-07

5-27

•8905

62^14

69^71

22-16

•9125

52-45

60-30

5-67

•8900

62^36

69^92

22-53

•9120

52-68

60-52

6-07

•8895

62-59

70-14

22-91

•9115

52-91

60-74

6^47

•8890

62-82

70-35

23-29

•9110

5313

60-97

6-86

•8885

6304

70^57

23-66

•9105

53^35

61-19

723

•8880

63-26

70-77

24-02

•9100

5357

61-40

7-61

•8875

63-48

70-97

24-37

•9095

53^78

61-62

7^99

•8870

63-70

71-17

24-73

•9090

54^00

61-84

8^36

•8865

63-91

71-38

2509

•9085

54-24

6207

8^78

•8860

64-13

71-58

25-44

•9080

54-48

62-31

9^20

•8855

64-35

71-78

25-79

•9075

54-71

62-55

9-62

•8850

64-57

71-98

26-15

•9070

5495

62-79

1003

•8845

64-78

72-18

26-50

•9065

5518

63-02

10^44

•8840

65-00

72-38

26-85

•9060

55^41

63-24

10-84

•8835

65-21

72-58

27-19

•9055

5564

63-46

11-24

•8830

65^42

72-77

27-52

•9050

5586

63-69

11-64

•8825

65^63

72-96

27-85

•9045

5609

63-91

12-03

•8820

65-83

7315

28-19

•9040

. 5632

64 14

12-41

•8815

66-04

73-34

28-52

•9085

5655

64-36

12-80

•8810

66-26

73-54

28-87

•9030

56-77

64-58

13-18

•8805

66-48

73-73

29-22

•9025

5700

64-80

13-57

•8800

66-70

73-93

29-57

•9020

57^22

65^01

13-92

•8795

67-91

7413

29-92

•9015

57^42

65-21

14-27

•8790

67-13

7433

30-26

•9010

57^63

65-41

14-62

•8785

67-33

74^52

30-59

•9005

57-83

65-61

14-97

-8780

67-54

74^70

3092

•9000

58-05

65-81

15-33

-8775

67-75

74^89

31-25

•8995

58-27

66-03

15-72

•8770

67-96

75^08

31-58

•8990

58^50

66-25

16-11

•8765

68-17

7527

31-90

•8985

58-73

66-47

16-49

•8760

68-38

75-45

32-23

•8980

58-95

66-69

16-88

•8755

68-58

75-64

3>-56

•8976

59-17

66-90

17-25

•8750

68-79

75-83

32-89

278

FOOD AND DKUGS.

Alcohol Table.

Over Proof.

Sp. Gr.

Per
cent of

Per

cent

Sp. Gr.

Per

cent of

Per

cent of

Per

cent

Alcohol

60° F.

over
Proof.

60° F.

by.

.r}^

over
Proof.

Weight.

Volume.

-8520

Weight.

Volume.

•8745

6900

76^01

33-21

78^52

84-27

47-70

•8740

6921

76-20

33-54

•8515

78^72

84^44

47-98

•8735

69^42

76^39

33-86

•8510

78^92

84 60

48-27

•8730

69^63

76-57

34-19

-8505

7912

84^77

48-56

•8725

69^83

76-76

34-51

-8500

79-32

84-93

48-84

•8720

7004

76-94

34-84

-8495

79-52

85-10

49-13

•8715

70^24

77^12

35-14

•8490

79-72

85-26

49-38

•8710

70-44

77-29

35-45

-8485

79-92

85-42

49-67

•8705

70-64

77-46

3576

•8480

80-13

85-59

50-00

•8700

70-84

77-64

3607

•8475

80-38

85-77

50-31

•8695

71^04

77-82

36-37

•8470

80-54

85-94

50-61

'8690

71-25

78-00

36-69

•8465

80-75

86-11

5091

•8685

71-46

78-18

37-01

■8460

80-96

86-28

51-21

•8680

71^67

78-36

37-33

-8455

81-16

86-45

51-50

•8675

71^88

78-55

37-65

•8450

81-36

86-61

51-78

•8670

72^09

78-73

37-98

•8445

81-56

86-77

52-06

•8665

72-30

78^93

38-32

•8440

81^76

86-93

52-34

•8660

72-52

79^12

38-65

-8435

81^96

87-09

5262

•8655

72^74

7931

38-99

-8430

82-15

87-24

52-90

•8650

72-96

79-50

39-32

•8425

82-35

87-40

53-16

•8645

73-17

79^68

3964

•8420

82-54

87-55

53-43

•8640

73-38

79^86

39-96

•8415

82-73

87-70

53-70

•8635

73-58

80^04

40-27

•8410

82-92

87-85

53-96

•8630

73-79

80-22

40-60

•8405

83-12

8800

54-23

•8625

7400

80-40

40-91

-8400

83-31

88-16

54-50

•8620

74-23

80-60

41-26

•8395

83-50

88-31

54-75

•8615

74^45

80-80

41-61

•8390

83-69

88-46

55-07

•8610

74^68

81-00

41-96

-8385

83-88

88-61

55-28

•8605

74-91

81-20

42-31

-8380

84-08

88-76

55-55

•8600

75^14

81-40

42-66

-8375

84-28

88-92

55-85

•8595

75^36

8160

43-00

-8370

84-48

89-08

56-10

•8590

7559

81-80

43^35

-8365

84-68

89-24

56-38

•8585

75-82

82-00

43-70

-8360

84-88

89-39

56 66

•8580

76-04

82-19

44-04

-8355

85-08

89-55

56-93

•8575

76 25

82-37

44-35

-8350

85-27

8970

57 20

•8570

76-46

82-54

44-66

•8345

85-46

89-84

57-45

•8565

76-67

82-72

44 97

•8340

85-65

89-99

57-71

•8560

76-88

82-90

45-28

-8335

85-85

9014

57-97

•8555

77-08

83-07

45-60

•8330

86 04

90-29

58-23

•8550

77-29

83-25

45-90

•8325

86-23

90-43

58-48

•8545

77-50

83-43

46-20

-8320

86-42

90-58

58-74

•8540

77^71

83-60

46-51

•8315

86-62

90-73

59-00

•8535

77-92

83-78

46-82

-8310

86-81

90-88

59-26

•8530

78^12

83-94

47-11

-8305

87-00

91-02

59-51

•8525

78^32

84-11

47-40

•8300

87-19

91-17

59-77

ALCOHOLIC BEVEEAGBS.

Alcohol Table.

Over Proof.

279

Per

Sp. Gr.

cent of

Sp. Gr.

cent of

Alcohol

cent

Alcohol

cent

60° F.

Weight.

Volume.

over
Proof.

60° F.

Weight.

Voliune.

over
Proof.

•8295

87^38

91-31

60-02

-8110

94^28

96-32

68-80

•8290

87^58

91-46

60-28

-8105

9445

96-43

69-00

•8285

87^77

91-60

60-53

-8100

9462

96-55

69-20

•8280

87-96

91-75

60-79

-8095

94-80

96-67

69-40

•8275

88-16

91-90

61-05

-8090

94-97

96-78

69-61

•8270

88-36

9205

61-32

-8085

95-14

96-90

69-82

•8265

88^56

9221

61-60

-8080

95-32

97-02

70-03

•8260

88^7B

92-36

61-86

-8075

95-50

97-15

70-25

•8255

88^96

92-51

62-12

-8070

95-68

97-27

70-46

•8250

8916

92-66

62-38

-8065

95-86

97-39

70-67

•8245

89^35

92-80

62-63

•8060

96-03

97-51

70-88

•8240

89^54

92-94

62-88

•8055

96-20

97-62

71-07

•8235

89-73

93-09

6313

•8050

96-37

9773

71'26

•8230

89-92

93-23

63-38

•8045

96-53

97-83

71-45

•8225

90^11

93-36

63-62

•8040

96-70

97-94

71-64

•8220

90^29

93-49

63-84

•8035

96-87

98-05

71-83

•8215

90^46

93-62

64-06

•8030

97-03

98-16

72-02

•8210

90^64

93-75

64-30

•8025

97-20

98-27

72-20

•8205

90^82

93-87

64-51

•8020

97-37

98-37

72-40

•8200

91^00

94-00

64-74

•8015

97^53

98-48

72-58

•8195

91^18

94-13

64-96

•8010

97^70

98-59

72-77

•8190

9136

94-26

65-18

•8005

97^87

98-69

72-95

•8185

91^54

94-38

65-40

•8000

98^03

98-80

73 14

•8180

91^71

94-51

65-62

•7995

98-19

98-89

73-30

•8175

91^89

94-64

65-85

•7990

98 34

98-98

73-47

•8170

92^07

94-76

66-07

•7985

98^50

99-07

73-64

•8165

9226

94-90

66-30

-7980

98-66

9916

73-81

•8160

9244

9503

66^53

•7975

98-81

99-26

73-97

•8155

92.63

9516

66-76

•7970

98-97

99-35

7414

•8150

92^81

35-29

67-00

•7965

99-13

99-45

74-31

•8145

93^00

95-42

67-23

•7960

99-29

99-55

74-50

•8140

93^18

95-55

67-46

•7955

99-45

99-65

74-66

•8135

93-37

95-69

67-70

•7950

99-61

99-75

74-83

•8130

93^55

95-82

67-92

•7945

99-78

99-86

75-01

•8125

93^74

95-95

68-15

-7940

99-94

99-96

7518

•8120

93^92

96-08

68-38

-7938

Absolute

Alcohol

75-25

•8115

9410

96-20

68-60

(2) Berthelot's benzoyl- chloride test. The liquid is shaken with a,
few drops of benzoyl-chloride. Alcohol reacts with this, forming
ethyl benzoate which, if present, sinks to the bottom with excess of
benzoyl-chloride The heavy layer is drawn off and warmed with a
little caustic potash solution. The characteristic odour of ethyl
benzoate is at once perceived.

280 FOOD AND DEUGS.

(3) Hardy's test. This test is carried out as follows : two ordin-
ary Nessler cylinders are stood side by side on a white tile. Into one
is placed the sample to be tested, after being well shaken up with a
fragment of guaiacum resin which has been freshly detached from a
lump, and then filtered. A few drops of dilute hydrocyanic acid and
a drop or two of dilute copper sulphate solution are then added. The
same procedure is adopted with the second tube, except that distilled
water is used in place of the sample. If alcohol be present, the
sample will have a much deeper blue colour than the blank cylinder.

Water in Alcohol. — In the examination of absolute alcohol, the
presence of water is important ; 0*4: per cent of water may be detected
by shaking the liquid with a crystal of potassium permanganate. In
the presence of this amount of water, the liquid will acquire a pink
tinge, whereas potassium permanganate is totally insoluble in alcohol.
If alcohol be shaken with anhydrous cupric sulphate it will give the
salt a blue colour in the presence of about 0*8 per cent of water,
or calcium carbide may be shaken with the alcohol. In the absence
of water no reaction takes place, whilst in the presence of water,
bubbles of acetylene are liberated and the liquid becomes turbid owing
to the formation of traces of calcium hydroxide.

Ihe Detection of Methyl Alcohol. — The following methods may
be used : —

(1) MuUiken and Scudder's test depends on the oxidation of
methyl alcohol to formic aldehyde by means of a hot copper wire. It
has been modified and adopted by the United States Pharmacopoeia
and will, in this form, detect as little as 2 per cent of methyl alcohol
in ethyl alcohol. It is carried out as follows : —

Into a test tube 1 c.c. of the preparation to be tested is introduced,
and, unless dilute, made up to 10 c.c. The proportion of alcohol pre-
sent should not exceed 10 per cent by volume. A copper wire spiral
is made by winding 1 metre of No. 18 copper wire closely round a
glass rod 7 mm. thick, making a coil 3 cm. long, the rest of the wire
being used for a handle. The coil is heated to redness in a smokeless
flame, then immersed steadily quite to the bottom of the alcoholic
fluid. This treatment is repeated five or six times, immersing the
tube meanwhile in cold water to keep down the temperature of the
liquid. The spirit is now filtered into a wide test tube and boiled
very gently. If any odour ol acetaldehyde is perceptible, boiling is con-
tinued until this has been dissipated. The liquid is then cooled and a
drop of 1 : 200 solution of resorcinol added to it. A portion of this
mixture is then floated on H2SO4 in another tube and allowed to stand
for three minutes, then slowly rotated. No rose-red ring should be
evident at the zone of contact of the two liquids, indicating the ab-
sence of more than 2 per cent of methyl alcohol.

(2) Kiche and Bardy's method. Ten c.c. of the sample, rectified
over potassium carbonate if necessary, are placed in a flask with 15
grms. of iodine and 2 grms. of red phosphorus. The flask is kept in
ice-cold water until the reaction is over. The ethyl and methyl
iodides are then distilled into about 30 c.c. of water. The free iodine
may be removed by washing with dilute alkali. The heavy oil is now

ALCOHOLIC BEVERAGES. 281

separated and transferred to a flask containing 5 c.c. of aniline. The
flask should be kept in cold water in case the reaction is violent, or it
may be warmed if the reaction appears to require stimulating. After
an hour the reaction product is boiled with water, and about 20 c.c. of a
15 per cent solution of caustic soda added. The bases formed now rise
to the surface of the fluid and are separated. To 1 c.c. of the oil add 10
grms. of a mixture of 100 parts of sand, 2 of common salt, and 3 of cupric
nitrate. After thorough mixing, introduce into a glass tube and heat
to 90° C. for eight to ten hours. The product is then exhausted with
warm alcohol, the alcohol filtered and made up to 100 c.c. If the
sample were free from methyl alcohol, the liquid is of a red tint, but
in the presence of 1 per cent of methyl alcohol, it has a distinct
violet shade. With 2*5 per cent the violet is very marked. Even
smaller quantities are indicated by diluting 5 c.c. of the coloured alco-
holic extract to 100 c.c. with water, and again diluting 5 c.c. of this to
400 c.c. If this be now heated and a fragment of white wool, abso-
lutely free from sulphur, be inserted in it for an hour, it will be dyed
a faint violet, if traces of methyl alcohol are present.

(3) Trillat's test. This test is a very delicate but very tedious
one. To 50 c.c. add 8 grms. of lime, and 50 c.c. of water, and then
•distil with the aid of a pear head, or Glinsky bulb tubes. Dilute the
first 15 c.c. distilled to 150 c.c. with water, and add 15 grms. of potas-
sium bichromate and 70 c.c. of 20 per cent sulphuric acid. Allow
the whole to stand for an hour with occasional shaking. Eedistil,
rejecting the first 25 c.c. and reserving 100 c.c ; 50 c.c. of this are
mixed with 1 c.c. of pure dimethyl-aniline and transferred to a well-
stoppered, stout glass flask. The mixture is rendered distinctly
■alkaline with caustic soda, and the excess of dimethyl-aniline distilled
■off, the distillation being stopped when 25 c.c. have come over. The
residue in the flask is acidified with acetic acid, well shaken and
5 c.c. of water containing 1 per cent of lead dioxide in suspension
added. If methyl alcohol be present, a blue coloration is developed
which is increased by boiling (ethyl alcohol yields a blue colour,
changing at once to green and then to yellow, becoming colourless
when boiled).

(4) Hinkel's method. This is" not particularly delicate, only de-
tecting about 4 to 5 per cent of methyl alcohol in ethyl alcohol. But
since in preparations such as tinctures, etc., when it is likely methy-
lated spirit might be used, the amount of methyl alcohol is usually
appreciable, if any be present at all, the method is frequently of
service.

To 1 c.c. of the alcoholic distillate 0'8 grm. of ammonium per-
sulphate and 3 c.c. of 20 per- cent sulphuric acid are added. The
whole is diluted with water to 20 c.c. and distilled. The distillate
is collected in fractions of 2 c.c. of which the first five are reserved.
The first two, containing acetaldehyde, are rejected, the remaining
three being tested as follows : —

A few drops of a 0*5 per cent solution of morphine hydrochlorida
are added to each, and then strong H.^SO^ is added to form a layer at the
•bottom of the liquid. In the presence of formic aldehyde, which

282 FOOD AND DRUGS.

indicates the presence of methyl alcohol, a violet ring is formed at
the junction of the liquid. As this reaction will detect minute
quantities of formic aldehyde, and as pure alcohol yields traces of
formic aldehyde, a control experiment should be conducted on pure
ethyl alcohol, so that the colours may be compared.

Where very small quantities of methyl alcohol are in question
the alcohol should be fractionally distilled under a pear head or rod
and disc apparatus, and the earlier fraction subjected to one or more
of the above tests.

Sangl6-Ferri6re and Cuniasse detect methyl alcohol as follows
(" Annales de Chim. Analyt." 8, 82) : 50 c.c. of the liquid is dis-
tilled off, the distillate acidulated with 1 c.c. of pure H^SO^, and
treated with 5 c.c. of saturated solution of KgMn^Og. After allowing
to stand for a few minutes the colour should be distinctly brown,
without any reddish tinge due to excess of K^MngOg. If this excess
should occur it must be removed by the addition of a drop or two of
solution of tannin. The liquid is then made faintly alkaline with
Na^COg, filtered, and treated with 2 c.c. of a 1 per mille solution of
phloroglucin and 1 c.c. of strong solution of KOH. In the presence
of added methyl alcohol a marked red colour reaction will be obtained..
A slight yellowish-red or violet tint may be disregarded, since a trace
of methyl alcohol may occur in pure wine alcohol : the reaction, due
to added methyl alcohol, being bright red, is unmistakable. A con-
firmatory test may be obtained with gallic acid. The alkaline filtrate
is acidified with a dilute H.^SO^; a few grains of gallic acid are dis-
solved in the liquid, when a few drops of strong H.2SO4 are carefully
run down to the bottom of the vessel. In the presence of methyl
alcohol a blue colour will form at the zone of contact of the two-
liquids. It will be seen that these reactions depend on the formation
of formaldehyde by the oxidation of the methyl alcohol.

The Determination of Methyl Alcohol. — The method of Thorpe
and Holmes (" J. Chem. Soc." 1904, 85, 1) is accurate and compar-
atively simple. The sample is mixed with water so that 50 c.c. shall
contain not more than 1 grm. of methyl alcohol, or 4 grms. of mixed
methyl and ethyl alcohols. Fifty c.c. of this mixture are placed in a
300 c.c. flask, which can be closed by a ground-in stopper, and
which is fitted with a funnel and side tube ; 20 grms. of potassium
bichromate are added and 80 c.c. of 25 per cent H.2SO4. The mixture
is allowed to stand for eighteen hours. A further quantity of 10
grms. of potassium bichromate and 100 c.c. of 50 per cent H.^SO^ (by
volume) are now added, and the liquid raised to the boiling-point for
ten minutes, the CO2 evolved being swept out of the flask by a current
of air, and after passing through drying tubes of CaCl2 and H^SO^,
collected in a soda lime tube and weighed. Thirty-two parts of
methyl alcohol yield forty-four parts of CO.2. A correction must be
made for the ethyl alcohol present by deducting 001 grm. of CO^ for
each grm. of ethyl alcohol. As there is always at least ten times as
much ethyl alcohol present as methyl alcohol, the amount of ethyl
alcohol may be taken as that indicated by the specific gravity of the
liquid or distillate. Certain other constituents of wood naphtha are

ALCOHOLIC BEVERAGES. 283

completely oxidized to CO^ in this manner, so that the above results
are comparative rather than absolute, being from 5 to 10 per cent
above the actual truth.

Thorpe and Holmes recommend, in the case of tinctures, that the
alcohol from 25 to 50 c.c. of the sample be distilled, and any essential
oils present removed by shaking with petrolum ether, and then distilled
and diluted to 250 c.c. Fifty c.c. of this mixture are then treated as
above described. If the weight of CO.^ does not exceed O'Ol grm.
per grm. of alcohol present, it may be concluded that methyl alcohol
is absent.

An alternative method is that devised by Leach and Lythgoe,
which depends on the fact that the refractive indices of methyl and
ethyl alcohols are very different. These chemists use the Zeiss im-
mersion refractometer, but any form of refractometer will be found
suitable so long as the absolute index of refraction be determined.
As most chemists use a Zeiss Abbe refractometer, the following values
refer to this treatment. It is only necessary to distil say 75 per
cent of the liquid and make up to original volume and determine the
specific gravity. The refractive index at 20° is then taken. If the
amount of alcohol as indicated by the specific gravity agrees with
that indicated by the refractive index, no methyl alcohol is present.
If a difference is indicated, the amounts of methyl alcohol present can
be deduced by interpolation and calculation. The table on page 284
gives the amounts of the two alcohols indicated by various refractive
indices : —

For example, if the distillate showed a specific gravity which cor-
responded, as found by the tables, to 18 per cent of ethyl alcohol, and
has a refractive index 1-34116.

The correct readings for ethyl and methyl alcohols of 18 per cent
strength are 1-34518 and 1-33712 respectively, the difference being
0-00806. The difference between the refractive index of pure ethyl
alcohol of this strength and of the sample is 1-34518 - 1*34116 =
0-00402.

So that 0-00402 : 0-00806 \ \ x \ IOC, where x is the amount of
methyl alcohol in the total alcohol present, that is, in the above case,
49-9 per cent.

The Determination of Alcohol. — Where only alcohol and water are
present, it is sufficient to determine the specific gravity, from which
the amount of alcohol is at once found from the table on p. 275. In
the presence of fixed matter (volatile acids may be fixed by neutrali-
zation with alkali) 100 c.c. (if possible) should be distilled until
80 c.c. has been collected, and this should then be made up to the
original volume with distilled water. In the case of many wines the
addition of a little tannin will be found to assist a quiet distillation.
When the result has to be expressed-by weight, the amount, by weighty
in the distillate so made up to original volume is taken from the

table and then corrected by multiplying by the factor ^' ° ' , t- ,

•^ tf J ^ J gp gj, Qf sample

which then gives the amount by weight in the original sample.

An alternative method where fixed matter is present is to evapor-

"284

FOOD AND DRUGS.

ate a measured portion of the sample to about 25 per cent of its volume
and when cold make up to its original volume with distilled water

Refractive Indices.

Per cent by Weight of Alcohol.

Methyl Alcohol.

Ethyl Alcohol.

Per cent

1-33312

1-33358

1-33335

1-33420

1-33358

1-33478

1-33381

1-33540

1-33405

1-33601

1-33427

1-33671

1-33451

133739

1-33474

1-33812

1-33497

1-33881

1-33521

1-33949

1-33543

1-34018

1-33567

1-34086

i-33590

1-34158

1 33613

1-34226

1-33636

1-34294

1-33663

1-34369

1-33686

1-34444

133712

1-34518

1-33735

1 34593

1-33762

1-34668

1-33812

1-34809

1-33862

1-34954

1-33907

1-35091

1-33957

1-35223

1-34002

1-35352

1-34052

1-35450

1-34094

1-35547

1-34135

1-35638

1-34173

1-35718

1-34203

1-35797

1-34229

1-35869

1-34248

1-35937

1-34256

1-36002

1-34264

1-36063

1-34267

1-36120

1-34260

1-36149

1-34256

1-36217

1-34245

1-36255

1-34222

1-36294

1-34195

1-36329

1-34162

1-36362

1-34124

1-36394

1-34086

1-36419

1-34048

1-36443

1-34010

1-36464

and take the specific gravity of this de-alcoholized liquid. Add 1*000
to the original specific gravity and subtract the second gravity. The
difference is the specific gravity corresponding to the alcohol present,

ALCOHOLIC BEVEKAGES. 285

from which the amount of alcohol is determined by reference to the
table. Thus, if the specific gravity of the original sample be 0'9850,
and that of the de-alcoholized sample is 1-0040. Then 1-9850 - 1-0040
= 0-981, and this corresponds to 15*1 per cent by volume of alcohol.
To convert this into the percentage by weight it should be multiplied
by 0-7938 and divided by the specific gravity of the original liquid.

The determination of alcohol by the vaporimeter is described under
wines (p. 315). Wiley (" Journ. Amer. Chem. Soc." 1896, 18, 1063)
has described a method for the determination of alcohol based on the
boiling-point of the liquid. For the details of this method the ori-
ginal paper should be consulted, but the following remarks may be
made upon it. Where the alcoholic liquid contains solid matter in
solution, the actual boiling-points of liquids are considerably altered, and
a diluted whisky and a port wine, each containing the same quantity of
alcohol, will give different boiling-point results. Where there is no
solid matter present, the specific gravity gives accurate results, and
so cumbersome a method as this is unnecessary.

Where the volatile substances are present, such as essential oils
and the like, Thorpe and Holmes (" Journ. Chem. Soc." 1903, 83,
314) use the following process : 25 c.c. of the sample, at 60° F.,
are mixed with water in a separator to a volume of 100 to 150 c.c.
and sodium chloride added in sufiicient quantity to saturate the
liquid. The mixture is now shaken well with petroleum ether (50
to 80 c.c), and after standing for half an hour the lower layer is
drawn off and extracted again with petroleum ether and then drawn
off again, and the petroleum ether liquid washed twice with salt,
solution, and the washings added to the main bulk of the liquid
and the whole distilled, and the distillate made up to 100 c.c.
(four times the original bulk). From the specific gravity of the distil-
late the amount of alcohol is at once found. If ammonia be present,
the liquid must be rendered slightly acid. If camphor be present,
dilute sulphuric acid is better to use than salt.

Deter vvuiation of Higher Alcohols in Spirits of Wine. — C. Bardy
(" Comptes Kendus," cxix. 1201-1204) recommends the following
process for the determination of higher alcohols in spirits of wine :
A preliminary examination is made by agitating 10 c.c. of the alcohol
to be tested with 50 c.c. of saturated solution of sodium chloride.
Two results may thus be produced : —

1. The salt solution forms a clear mixture with the alcohol, thus
indicating that the amount of impurity is small. In this case 500 c.c.
of the alcohol a^e mixed in a capacious separator with 450 c.c. of
solution of sodium chloride, and subsequently with sufficient water
to re-dissolve the salt separated ; 60 c.c. to 70 c.c. of carbon
bisulphide are then added, the whole is well shaken, and after some
minutes' rest the bisulphide is separated. This operation is repeated
three times. The bisulphide will then contain the whole of the butyl
and amyl alcohols, and to extract these it is shaken with 2 c.c. of
strong sulphuric acid, and the acid removed, after settling, into a flask
of 125 c.c. capacity. This operation is also repeated several times,
and the united acid liquor is freed from bisulphide by warming. An

286 FOOD AND DRUGS.

equal volume of glacial acetic acid is now added, the neck of the flask
closed with a reflux condenser, and the mixture heated to 100° C. for
four hours to promote formation of acetic ethers. The contents of
the flask are then mixed with 100 c.c. of salt solution ; if higher
alcohols were present, the ethers will separate as an oily layer on the
surface. This oily liquid is separated and measured at 15° C. ; the
volume expressed in c.c. and multiplied by 0'8 gives the percentage
of butyl and amyl alcohols.

2. An oily layer separates at the surface of the salt solution in
the preliminary experiment. In that case larger amounts of the
higher alcohols are present, and the operations above described are
now carried out with a smaller quantity (25 c.c.) of the alcohol, 100
c.c. of saturated salt solution, and 8 to 10 c.c. of water. The quantity
of bisulphide of carbon should not be reduced. Since the latter dis-
solves only the butyl and amyl alcohols, the liq:uid from which the
bisulphide has been separated must be examined for propyl and
isopropyl alcohols. For this purpose it is filtered through moist
paper and distilled, the distillate being collected in a tube containing
an alcoholometer until this instrument indicates 50°. At that point
the whole of the propyl alcohol will have passed over, and may be
•determined in the distillate by titration with permanganate.

BRANDY.

Brandy is a spirit resulting from the distillation of fermented
^rape juice or wine. Apart from the question of alcoholic strength,
for which a prima facie standard exists, the only question with
which the analyst is usually concerned is the admixture with true
brandy of alcohol derived from other sources, a form of adulteration
which is referred to in the oldest books available. Brandy is a term
sometimes applied to a spirit derived from other sources than the
^rape, but if the use of the word is at all justifiable in this sense it
should certainly be qualified in such a manner that the source is indi-
cated, e.g. plum brandy. The following definition of brandy has been
agreed to by the recent Royal Commission on whisky and other pot-
able spirits, 1909. " The term ' brandy ' is applicable to a potable
spirit manufactured from fermented grape juice, and from no other
materials."

They are, however, of opinion that the compounded spirit long
recognized by the name of British brandy is entitled still to be so
named and sold as "British brandy".

The bulk of the brandy of commerce is prepared in France, but,
of course, pure brandy is made in other countries. The most esteemed
type of spirit is that distilled in the Charente district, and it is reason-
able that the term Cognac should, in this country, carry the same
meaning as that which it does officially in the home of the industry.
By a decree dated 1 May, 1909, of the French Republic, no brandy
shall be entitled to the name " Cognac," " Eau-de-vie de Cognac " or
"eau de vie des Charente," except it be distilled on the spot from
vines grown in the following districts : —

BRANDY. 287

1. Dej)artme7it of Charente-hiferieure. — The arrondissements of
Eochefort, Marennes, Saintes, St. Jean d'Angely, Jonzac, parts of
La Rochelle.

2. Department of Charente. — The arrondissements of Cognac,
Barbejiieux, parts of Angoul^me, parts of Ruffec.

3. Department of Dordogne. — Parts of the arrondissement of
Riberac.

4. Department of Deux Sevres. — Parts of the arrondissements of
Niort and Melle.

The French Government have further restricted the use of the
names Armagnac and T6narfeze to brandies distilled from wine grown
and made within suitable geographical limits (Decree of 25 May, 1909).

The functions of the analyst are principally of importance, how-
ever, in deciding whether a brandy is pure, and no chemical means
are available for discriminating between pure brandies. The trade
expert would, however, be able to decide the place of origin of a
brandy with a fair degree of accuracy.

There can be no reasonable doubt that the medicinal value of
brandy is not entirely due to the alcohol it contains, but also to the
secondary constituents, which are either the result of the original fer-
mentation and distillation, or are formed during the process of maturing.
The characteristic flavour of brandy is, in the same manner, due to
such secondary constituents. Vasey (" Potable Spirits ") has remarked
that " the patent or fractionating still is practically the key to the
situation as regards the analysis of potable spirits ". This is, of
course, true of brandy, whisky, and rum at all events, and requires
comment at the present stage.

It is not proposed to enter into controversial matter as to the
actual merits of the pot or simple still, as against the patent or frac-
tionating still. Ample details of this — an essentially trade matter —
will be found in the report of the recent Royal Commission on whisky
and other potable spirits. The facts, however, amount to the following,
which apply equally to all such distilled spirits, but principally to
whisky : —

1. The distillate from the pot still contains considerably more
secondary constituents, as would be expected, than does the distillate
from a patent still.

2. The presence of a large amount of these secondary constituents
renders it necessary to mature a pot-still product for a considerable
time, in order to allow sufiicient change in the character of the
secondary constituents to take place for the spirit to be palatable.
As to the physiological effect of new and matured pot-still spirits,
great differences of opinion exist.

3. Patent-still spirits contain very little secondary constituents
and thus require but little maturing, but are are usually correspond-
ingly flavourless. They approximate in character to a pure diluted
alcohol obtained from any source, more or less, according to the
nature of the still.

In regard to whisky, as will be seen later, it appears that no
attempts will be made to restrict the patent stills which have been

288

FOOD AND DEUGS.

in use for very many years, as the beverage sold as whisky, made
from grain spirit, has for many years been either a pot-still product
or a blend of the two distillates. It is therefore a matter of the
public taste — " de gustibus non est disputandum ".

The brandy question is, however, not quite on the same footing, and
from the point of view of public expediency it is desirable that stills
of the pot or simple type should be used for the manufacture of brandy.

The addition of spirit from other sources than the grape to brandy
is, of course, deliberate adulteration.

Incidentally it may be mentioned that the British Pharmacopoeia
defines brandy as a spirituous liquid distilled from wine and matured
by age, and containing not less than 36'5 per cent by weight or 43'5
per cent by volume of alcohol.

Freshly distilled brandy is colourless, the colour of commercial
brandy being due to colouring matter derived from the casks in which
it is stored.

Brandy may be sold as such, diluted with water so that the
strength is not below 25° under proof (=35-93 per cent by weight, or
42-8 per cent by volume). The secondary constituents of brandy
have been examined with minute care by Ordonneau (" Comptes
Eendus," cii. 217). The typical sample upon which his results were
based was a Cognac brandy twenty-five years old.

He found in each 100 litres of absolute alcohol present : —

Grms.

Acetic aldehyde 3

Aeetal ........... 36

Butyl alcohol 218-(5

Hexyl „ 1-5

Proprionic, butyric, and caproic esters 3

Amine bases 4

Ethyl acetate 35

Propyl alcohol , 40

Amyl „ 83-6

Oenanthic ether 4

It appears to be agreed that the principal flavouring constituent
of brandy is oenanthic ether, and as this is easily made artificially from
the products of distillation of castor oil, it forms the basis of so-called
"artificial cognac oil," which is sold in order to mix with "silent"
spirit in order to prepare a factitious brandy.

The higher alcohols in brandy, separated by Ordonneau and by
Claudon and Morin under the name " fusel oil," have been carefully
examined, and the results of both investigations are very concordant.
According to these authorities the composition is as follows : —

Ordonneau.

Claudon and Morin.

Propyl alcohol .
Normal butyl alcohol
Isobutyl alcohol
Amyl alcohol .

Per cent

11-9

49-3

4-5

34-4

Per cent

11-7

63-8

24-5

BEANDY. 289

The higher'alcohols of potato spirit have approximately the follow-
ing composition : —

Per cent

Isopropjl alcohol 15

Isoarayl ,, 30

Butyl „ 8

Isobutyl „ 5

Amyl „ 18

Other bodies 26

It is impossible to estimate individual compounds in a spirit, but
allied substances can be determined in gi^oups, which materially
assist the analyst in coming to a conclusion.

But one cannot emphasize too strongly that any standards that have
been published for limits of esters, aldehydes, .etc., are liable to be found
untrue for occasional cases. The author is acquainted with some of
the finest champagne brandies which invariably show a deficiency in
esters according to the standards generally accepted, whereas by
"doctoring" with a trace of artificial cognac oil, they will at once
appear to pass the standards.

In this regard, attention should be paid to the remarks contained
in the final report of the Koyal Commission on whisky and other po-
table spirits, 1909. The Commissioners state that whilst certain
benefits have been obtained from the adoption of an " ether standard "
of 80 parts of ethers per 100,000 parts of absolute alcohol (a stan-
dard largely adopted by public analysts), the adoption of such a stan-
dard based on a minimum quantity of ethers alone is quite incapable
of affording general protection against fraud. For instance, genuine
brandies are frequently found to contain upwards of 100 parts of
ethers per 100,000 of absolute alcohol : admixtures of such with 20
per cent of neutral spirit would be passed as genuine brandies by the
standard. Again, the requisite proportion of ethers in any spirit can
be insured by the addition of suitable ethers. The standard can
therefore no longer be regarded as useful, and may even be mischiev-
ous. All the usual analytical data, and in addition, the flavour, and
any information obtainable from the excise records as to the origin of
the sample should be taken into account in order to form a reliable
opinion as to the genuineness or otherwise of a sample of reputed
brandy.

There is no doubt that this is true, but it must be remembered
that there is one case, and a not uncommon one, where the ether
value, as the Commissioners term it, is of definite service, that is,
where it falls inaterially below 80. For example, spirit with an
ether value of 25 to 50, as often- met with, would be condemned as
containing neutral spirit on all hands.

The quantitative determinations hereinafter described, are there-
fore to be judged in the light of the above remarks.

Genuine brandy contains a small amount of volatile acids of the
acetic acid series ; aldehydes of the aliphatic series, and traces of
furfural ; esters and higher alcohols.

The examination of brandy should include the determination of
VOL. I. 19

290

FOOD AND DKUGS.

Icoholic strength; solid residue; free volatile acids calculated as
acetic acid ; total aldehydes ; furfural ; esters calculated as ethyl
acetate, and higher alcohols.

Alcohol. — When the solid residue is under 0*5 per cent, as it fre-
quently is, the specific gravity of the sample is a sufficiently correct
indication of the amount of alcohol present. If a high residue be
present, 50 c.c. should be distilled from 60 c.c. and the distillate
made up to 60 c.c, and the specific gravity taken.

Solid Residue. — The solid residue of genuine brandy. averages under
1 per cent, usually from 0-3 per cent to 0*6 per cent, but sometimes,
especially in brandies which have been stored in new casks, up to
3 per cent of solid residue may be present. It is customary to re-
turn the secondary constituents of brandy in terms of parts per 100,000
of absolute alcohol present. Thus if a brandy contain 40 per cent of
alcohol by weight, and two parts of furfural per 100,000 be found,
this would be returned as 5 per 100,000. The following represent the
analyses of a large number of pure French brandies made by the
author and various observers, whose names are given : —

Parts per 100,000 of Absolute Alcohol.

No. of Samples.

Volatile Acids
as Acetic.

Total
Aldehydes.

Furfural.

Esters as
Ethyl Acetate.

Higher
Alcohols.

Observers.

Per cent

30 (average of) .

81-5

21-5

1-6

116

138

Parry

New brandy

740

14-5

2-6

108

195

„

Brandy 25 years old .

119-5

27-8

115

126

219

40 „ „ .

202

120

133

345

Girard

6 „ „ .

229

11-5

1-20

101

260

Various samples (average)

81-9

24-2

016

212

289

Konig

» » »

120

210

1'34

165

Mohler

The following are types of cpirit used in mixing with natural pot-
able spirits : —

Parts per 100,000 of Absolute Alcohol.

Samples.

Volatile Acids
as Acetic.

Aldehydes.

Furfural.

Esters.

Higher
Alcohols.

Observers.

Per cent

Grain spirit .

0-1

nil

3-6

2-9

Girard

8-4

4-9

0-35

23-8

traces

Vasey

61-2

11-22

0-23

32-65

nil

3-70

nil

47-6

67-7

Schidrowitz

Beetroot spirit

2-5

nil

3-6

2-5

Girard

Potato

(average of 8 samples)

1-8

4-9

0-1

5-0

19-9

Parry

BEANDY. 291

The use of pure alcohol will be frequent in the examination of
potable spirits, and it is essential that it shall contain none of the
impurities which are estimated. Pure 90 per cent alcohol can easily
be obtained free from esters and aldehydes, but should never be relied
on unless carefully checked. To ensure a pure alcohol, it should first
be boiled for an hour under a reflux condenser to destroy esters, and
then, after distillation, be again boiled for an hour with 0*4 per cent of
sodium phenyl-hydrazine parasulphonate or metaphenylamine diamine
hydrochloride.

The Determination of Alcohol. — When exact determination is neces-
sary 90 per cent of the liquid may be distilled and the distillate made
up to the original volume and the specific gravity taken.

Approximately accurate results may be obtained by taking the
specific gravity of the liquid (S), then evaporating off all the alcohol
and making up to the original volume with water, and taking the

specific gravity of this liquid (S'). Then — = specific gravity due to

the diluted alcohol, from which the amount of alcohol is deduced.

The Free Acids. — Twentv-five c.c. of the sample are titrated with

— baryta water using phenol-phthalein as indicator. This gives the

total acidity, which is expressed in terms of acetic acid. A second
25 c.c. are evaporated to dryness, the residue being redissolved in
water and dried again twice, and finally again dissolved in water and

N N

titrated with — baryta water. Each c.c. of — baryta water corre-
sponds to 0'0075 grm. of tartaric acid ; the difference in the number
of c.c. required for the fixed and total acidity x 0*006 gives the
amount in grams, of volatile acids calculated as acetic acid.

[The French official method, fixed by ministerial decree in 1907,

N
requires the use of 25 c.c. for the total acidity, using j-~ soda solution.

For the fixed acidity, the liquid is evaporated from 25 to 5 c.c, and
the drying completed in vacuo.]

Total Aldehydes. — The best process for the determination of alde-
hydes which will include the furfural present, is the official process
of the French Government, which has taken account of all the errors
probable in similar previously described processes. It depends upon
a comparison of the colours developed by the reaction of aldehyde
and fuchsine-sulphurous acid. The precautions which must be taken
are that the determinations should be carried on under such conditions
that the standard and the sample are of as nearly as possible the same
aldehydic strength, and the alcoholic strengths of the liquids must be
practically identical. If colorimeters are used where the same tint is
obtained on different thicknesses of the coloured solution, account must
be taken of the fact that the colour developed is not in direct proportion
to the amount of aldehyde present, and corrections must be made. In
the following process, although approximate results may be obtained
by au\ suitable colorimetric comparison, greater accuracy is ensured

292 FOOD AND DRUGS.

by using a standard colorimeter such as that of Dubosq or Mills, and
comparing the colour against 10 millimetres of the standard solution.
If the thickness of the sample under examination be materially differ-
ent from 10, the following table, due to Cuniasse, enables the necessary
corrections to be made : —

Indication of colorimeter matching 10 mm. of Aldehydes per 100,000 of

standard described below. absolute alcohol.

100 4

40 9

25 12

16-7 15

10 20

6-9 25

5-4 30

4-2 35

3-4 40

The process is carried out as follows : —

A standard solution of 0*1 grm. of aldehyde per litre (10 per
100,000 c.c.) is prepared by washing commercial aldehyde-ammonia
(which is stable) several times with ether, pouring off" the ether and
drying the crystals, which have been rubbed down in a mortar, over
HgSO^ in vacuo. 1-386 grms. (=1 grm. aldehyde) - are dissolved in
about 50 c.c. of 95 per cent alcohol. 22-7 c.c. of normal H^^SO^ are
added, when ammonia sulphate is precipitated. The liquid is made
up to 100 c.c. with 95 per cent alcohol, and then 0*8 c.c. of alcohol is
added to compensate for the volume of the sulphate of ammonia.
The liquid is well shaken, left for twelve hours, and filtered. It is
now a solution of 1 grm. of pure aldehyde in 100 c.c. of alcohol.
About 90 c.c. of water are then added and the whole made up to 1000
c.c. with 50 per cent alcohol. It is now a O'Ol per cent solution of
aldehyde in practically 50 per cent alcohol.

The fuchsine-sulphurous acid solution is made by mixing 30 c.c.
of 0-1 per cent solution of fuchsine in 95 per cent alcohol, 15 c.c. of a
solute of sodium bisulphite of specific gravity 1-308, and 30 c.c. of
water. The mixture is shaken, and allowed to stand for an hour,
and then 15 c.c. of 30 per cent sulphuric acid added. The liquid is
then made up to 250 c.c. with 50 per cent alcohol. After standing
for a short time this solution becomes quite colourless.

A portion of the 90 per cent distilled for the determination of the
alcohol is used for the determination, which, however, is diluted with
pure alcohol (so as to be of 50 per cent strength), or by water if above
that strength.

In the comparisons 10 c.c. of the standard aldehyde solution are
placed in one tube, and 10 c.c. of the diluted distillate in another.
For the original alcoholic strength, the amount of dilution is calculated
and the observed results corrected accordingly. For example, if 100
c.c. of the distillate have to be diluted to 130 c.c. for the determination,
the observed results used have to be multiplied by 1-3.

Four c.c. of the fuchsine-sulphurous acid solution is added to each

BRANDY. 293

tube and the contents well mixed and allowed to stand for twenty
minutes when the reading can be taken.

For example : If from 100 c.c. of brandy 90 c.c. be distilled and
made up to 100 c.c. and it is found that it is of 33 per cent strength,
it is necessary to add sufficient 95 per cent alcohol to 100 c.c. to bring
the volume up to 134-8 c.c. so as to bring the alcoholic strength to 50
per cent. Therefore the observed result must be multiplied by 1-348.
If 10 c.c. of the so diluted distillate exactly match 10 c.c. of the stan-
dard, then it will contain 0-1 grm. of aldehyde per litre, but being of
50 per strength, this is 0*2 grm. per litre of absolute alcohol or
20 per 100,000. This multiplied by 1-348 is 26-96. The brandy
therefore contains 26*96 parts of aldehyde per 100,000 parts of abso-
lute alcohol. The tables on page 294 give the amounts of 95 per cent
alcohol or water to be added to alcohols (distillates) in order to bring
them to exactly 50 per cent strength by volume.

An alternative method for the estimation of the aldehydes is as
follows, provided the amount of aldehydes is high, as is the case
sometimes : —

The following solutions are necessary : —

(1) Pure sulphite of sodium (anhydrous) 12-6 grms. dissolved in
400 c.c, 100 grms. of normal sulphuric acid added, and the whole made
up to 1000 c.c. with 95 per cent alcohol. If crystals of Na^SO^
separate they should be filtered off.

(2) Decinormal solution of iodine in iodide of potassium 1 c.c. =
0-0032 gr. SO.^ or 0-0022 of ethyl aldehyde.

Into a 100 c.c. flask, 10 c.c. of the solution to be tested (if the
aldehydes are present to the extent of 0-5 per cent to 1 per cent, or
correspondingly more, if the aldehyde value is lower) are placed, and
50 c.c. of the above sulphurous acid solution added. The volume is
made up to 100 c.c. with 50 per cent alcohol. The whole is well
shaken and the flask securely stoppered.

A blank experiment is conducted in the same manner, only omitting
the solution to be tested. The two flasks, securely stoppered are
placed in a water bath at 50° for four hours, and then cooled and
well shaken. Fifty c.c. are then titrated with the iodine solution.
The difference in the amount of free SOg indicates the amount that
has continued with the aldehydes, each c.c. of decinormal iodine
being equivalent to 0-0022 grm. of ethyl aldehyde.

Estimation of Furfural. — A standard solution of 10 milligrams of
furfural per 1000 c.c, in 50 per cent alcohol is used.

Ten c.c of the alcohol (distillate) brought to 50 c.c strength, and
10 c.c. of the standard solution are each treated with 0-5 c.c. of freshly
prepared aniline and 2 c.c. of glacial acetic acid. After twenty minutes
the solutions are compared colorimetrically. The same remarks as to
correction for dilution, and the irregular ratio of the colour produced
to the amount of furfural, apply here as to the fuchsine-sulphurous
acid method of determining aldehydes (p. 292). The following table
will enable the observer to make the necessary corrections, but the
greatest accuracy is obtained by repeating the experiment if the columns
are much different with quantities so adjusted to give as nearly

294

FOOD AND DRUGS.

Original Strength of Alcohol.

Volume of .95 per cent Alcohol
to add to 100 Vols.

Final Volume Obtained.

Per cent

42-2

140-2

401

138-2

380

1368

360

184-3

83-9

132-4

31-8

130-4

29-7

128-4

27-6

126-5

25-5

124-5

23-4

122-5

21-3

120-5

19-2

1185

17-1

116-4

14-9

114-4

12-8

112-4

10-7

110-8

8-6

108-2

6-4

106-2

4-3

104-1

102-0

1000

Strength of Alcohol.

Volume of Water to
add to 100 Vols.

Strength of Alcohol.

Volume of Water to
add to 100 Vols.

Per cent

100

107-4

50-3

105-6

48-1

102-7

46-0

100-4

43-9

98-1

41-8

96-9

39-7

98-6

37-6

91-4

35-4

89-2

33-3

87-0

31-2

84-8

29-1

82-6

27-0

80-4

25-0

78-2

22-9

760

20-8

78-8

18-7

71-7

16-6

69-5

14-5

67-4

124

65-2

10-4

63-1

8-3

60-9

6-2

58-8

4-1

56-7

2-1

54-5

0-0

52-4

BRANDY.

295

identical colours as possible. Assuming a 10 mm. layer of the
standard has been used in the colorimeter, then the following are
the amounts of furfural per 100,000 parts of absolute alcohol indicated
by the following depths of the solution being found equal to the
standard : —

ram. Parts per 100,000 of absolute alcohol.

200 0-1

100 0-2

66-7 0-3

50 0-4

33-3 0-6

25 0-8

20 10

13-3 1-5

10 20

8 2-5

5 40

In carrying out these colorimetric processes for the determination
of aldehydes or of furfural, it is well, as Vasey has pointed out, to
subject the standards to exactly the same treatment as the sample
under examination, by actually distilling the standard furfural solution
under conditions identical with those used in the distillation of the
sample.

J. T. Hewitt (" Jour. Soc. Chem. Ind.," Jan. 1902) prefers to distil
the sample (unless it be colourless) nearly to the last drop, a little
fresh pure spirit is poured into the distillation flask and the process
repeated several times and the mixed distillates made up to a certain
volume and then matched in glass cylinders by the standard solution
as in the well-known process of " Nesslerizing ". There is, however, a
risk of furfural being actually formed during this repeated distillation,
from the heated residue in the flask, and the results may be somewhat
too high.

Determination of Esters. — The fact that alkalies act upon alde-
hydes renders it necessary to remove these if a correct determination
of esters is required. This is best effected by boiling the alcohol
under a reflux condenser for an hour with 3 per cent of meta-phenyl-
ene diamine hydrochloride. The liquid is then distilled, 90 per
cent being collected and being made up to the original volume. To
100 c.c. of this liquid, a few drops of phenol-phthalein are added and
decinormal' baryta solution added to exact neutralization. Twenty-
five c.c. of decinormal alcoholic solution of soda are then added and
the whole boiled under a condenser for an hour. The excess of soda
is then determined by titration with decinormal hydrochloric acid,
each c.c. of alkali used being equivalent to 0-0088 grm. of ethyl
acetate, in which form the esters are returned. From the alcoholic
strength of the liquid the amount per 100,000 of absolute alcohol
is calculated.

[The French official method takes no notice of the presence of
aldehydes, but recommends that where the proportion is appreciable,
the saponification should be by means of a standard solution of lime
in sugar solution, which does not act appreciably on aldehydes.]

296 FOOD AND DKUGS.

Determination of Higher Alcohols. — Whatever the exact nature of
the alcohols present in what is usually termed the " fusel oil " of
distilled spirits, or more correctly, the "higher alcohols," it is certain
that isomeric amyl alcohols (principally 3 methyl-butanol 1, and 2
methyl-butanol 1) are the most important, and after these, normal
iso-primary butyl alcohols. Any process of determining these alco-
hols is necessarily more or less empirical, and most processes described
are now quite discredited. Schidrowitz and Kaye have exhaustively
examined the more promising process and have shown that the Ger-
man official process (Eose-Stutzer-Windisch) is useless in most cases ;
that Beckmanns process (" Zeit. Unter. Nahr. v. Genuss.' ii. 709;
IV. 1057) is quite misleading; and that the French official process
(Girard and Cuniasse), whilst giving fair results with brandy, is mis-
leading in regard to whisky.

The process upon which most reliance is to be placed is that of
Marquardt, as modified by A. H. Allen and slightly modified by
Schidrowitz (" Jour. Soc. Chem. Ind.," 1902, 815). The details of the
process are as follows : —

Two hundred c.c. of the spirit are boiled under a reflux condenser
for an hour with 0*2 grm. of KOH, by which means acids are com-
bined, and esters and furfural are decomposed. The liquid is now
distilled until 180 c.c. have passed over, and steam passed through
the residue till 300 c.c. are collected. The alcoholic strength of
the liquid should be as near 50 per cent by volume as possible. If
too low, it should be raised by the addition of pure alcohol. The
exact volume of the liquid is now noted and 100 c.c. taken for the
estimation. This is mixed with saturated brine until the mixture has
a soecific gravity 1*100. The mixture is then extracted with three
successive quantities of 40 c.c. of carbon tetrachloride (which has
been purified by washing with water, boiling with chromic acid
mixture, washing with sodium bicarbonate solution and finally again
with water until neutral. It is then distilled and ready for use). The
carbon tetrachloride solution is shaken with 50 c.c. of a saturated
solution of potassium sulphate, the dry carbon tetrachloride separated
and filtered, and then oxidized with a solution of 5 grms. of potassium
bichromate, 2 grms. of HgSO^ and 10 c.c. of water, for at least eight
hours in a water bath. Any loss during the heating for eight hours is
prevented by having the flask ground to fit the reflux condenser, the
tube of which is fitted with the rod and disc condensing device. The
liquid is then distilled, first over a Bunsen burner, and then with
steam, until about 300 c.c. have passed over. The carbon tetrachloride
may be separated and washed once with water, the washings being
added to the main bulk of the aqueous liquid. The aqueous distillate
is now titrated with decinormal baryta solution. Methyl orange is,
according to Allen, first used as an indicator, the end reaction in-
dicating the neutralization of the traces of free mineral acid (HCl)
that may have been found. It is continued with phenol-phthalein,
this result giving the amount of organic acids which are calculated to
valeric acid, and thence to amyl alcohol (but see belowj. Each c.c.
of decinormal bartya solution used for the neutralization of the or-

BRANDY. 297

ganic acids, corresponds to 0*0088 grm. of amyl alcohol (or to 0-0074
grm. of butyl alcohol).

Schidrowitz and Kaye (" Analyst," xxxi. 181) have shown that the
organic acids have some effect on methyl-orange, and that the apparent
mineral acid value is not, at all events principally, due to mineral
acid in reality. So long as the apparent mineral acid value does not
exceed ^^th of the total acid value it may be neglected. In doubtful
cases, the chlorine should be determined gravimetrically, and the
mineral acid as HCl deducted from the total acidity.

The same authorities have very exhaustively examined the AUen-
Marquardt process, and have, by starting from weighed quantities of
amyl and butyl alcohols, shown that so long as the higher alcohols are
not present to the extent of more than 0"15 per cent — which is usually
true for commercial spirits, the process yields exceedingly accurate
results. When the amount is over 0*15 per cent, the oxidation should
go on for ten hours ; when over 0'3 per cent, the determination is not
reliable,

Marquardt's original process possesses one useful feature. When
the organic acids were distilled he warmed their aqueous solution with
excess of barium carbonate for some time, filtered the solution, and
evaporated the water and weighed the barium salts. If an indication
of the nature of the acids — and therefore of the alcohols — be required,
a determination of the amount of barium in the barium salts, by con-
version into barium sulphate, will give the mean combining weight of
the acids and thus indicate the nature of the alcohols. If mineral
acids be actually present in the distillate, the necessary allowance for
the barium chloride found must, of course, be made.

Bell some time ago suggested replacing the bichromate used in
the Marquardt process by permanganate of potassium, but this was
generally regarded as a retrograde step. Mitchell and Smith ("U.S.
Dept. of Agriculture Bull." 122, 1909, 199) have again suggested this,
and as the bulletins of the Department carry official weight in
America, it is advisable to describe their process. The carbon tetra-
chloride solution of higher alcohols is placed in a separator with 10
c.c. ot a 50 per cent solution of KOH and the mixture cooled to 0°.
One hundred c.c. of a 2 per cent solution of potassium permanganate
are placed in a flask, cooled to 0° and then added to the separator.
The mixture is well shaken for five minutes and then set aside for
thirty Jiiinutes at the laboratory temperature. One hundred c.c. of
H.,0.^ solution which is rather stronger, relatively, than the perman-
ganate solution are now placed in a 1000 c.c. flask, 100 c.c. of 25 per
cent H._,S04 ^-dded, and the contents of the separator added with con-
tinual shaking. The separator is rinsed with water, which is added to
the HgO^ solution. The excess of Hy02 is titrated with a standard
permanganate solution (about 1 per cent). A blank experiment
is carried out at the same time, and the amount of permanganate
used for oxidation is noted. It is found that 1 grm. of permanganate
oxidizes 0*475 grm. of propyl alcohol, 0*585 grm. of isobutyl alcohol,
and 0*696 grm. of amyl alcohol.

298 FOOD AND DRUGS.

In the same bulletin, processes due to Tolman and Hillyer are
described as follows : —

(1) Estimation of Colouring Matter. — Fifty c.c. of the spirit are
evaporated to dryness, the residue dissolved in il6-3 c.c. of 95 per
cent alcohol and this solution diluted to 50 c.c. with water. Twenty-
five c.c. of this solution are treated in a separator with 20 c.c. of a
solution consisting of 100 c.c. amyl alcohol, 3 c.c. of syrupy phos-
phoric acid and 3 c.c. of water. The whole is well agitated and al-
lowed to separate, three times. The aqueous layer is drawn off and
diluted to 25 c.c. with 50 per cent alcohol. The colour of this solu-
tion is now compared with the colour of the other 25 c.c. of the ori-
ginal solution of the dried residue, which has not been treated with
the amyl alcohol mixture. The percentage of the colour which has
not been dissolved by atnyl alcohol is thus obtained. This is stated
by the authors to be due to added caramel. In the determination
of the higher alcohols, these chemists prefer to determine the excess,
of bichromate left after oxidizing the carbon tetrachloride solution,
by the liberation of iodine from potassium iodide and titrating this
with standard solution of sodium thiosulphate. A blank experiment
is carried out, and each c.c. of decinormal thiosulphate required by
the blank in excess of the sample is equivalent to 0*001773 grm. of
amyl alcohol.

The French official method for the determination of higher alcohols
appears to give very fair results with brandy, although its indications
with whisky are erratic. In spite of the fact that amyl alcohols are
the most important of the higher alcohols naturally found in spirits,
this process adopts iso-butyl alcohol for its standard solution. This
solution contains 0'667 grm. of iso-butylic alcohol in 1 litre, the solvent
being alcohol of 66"7 per cent strength.

One hundred c.c. of the brandy, etc., are distilled and the distillate
adjusted to exactly 50 per cent alcoholic strength. This is placed in a.
250 c.c. flask, 1 c.c. of pure aniline and 1 c.c. of s5Tupy phosphoric acid
added, together with a few pieces of pumice stone. The liquid is
gently boiled for an hour under a reflux condenser. It is then allowed
to cool, and then distilled until 75 c.c. have passed over. This will,
of course, be of 66'7 per cent alcoholic strength. To 10 c.c. of this
10 c.c. of pure colourless monohydrated H.^SO^ (sp. gr. = 1*799) are
added. The acid and alcohol are well mixed and heated to 120° for
one hour in a chloride of calcium bath. Ten c.c. of the standard
solution are treated in the same manner with sulphuric acid and
the colours of the two liquids compared in a colorimeter. The
standard solution corresponds to 0667 grm. of iso-butyl alcohol per
litre of 66*7 per cent alcohol, so that if the colour of the two liquids
is identical, the amount of higher alcohols (calculated as iso-butyl
alcohol) per 100,000 of absolute alcohol, would be 100. Jn the case
of any material divergence in colour, the only correct method is to
repeat the experiment with such a quantity of the alcohol as to give
practically identical colours. Approximate results may, however, be
obtained by constructing a curve from the following values, and by
intercalating, the true result may be found. As the colours are not.

BRANDY.

299

directly proportional to the amounts of alcohols present, assuming 10
mm. of the sample matches 10 mm. of the standard solution, the
value will be 100 parts per 100,000 of absolute alcohol ; but

260 mm. matching 10 mm. of the standard =

10 per 100,000
20

33
19-5
13-2
4-4
3-7
31
2-3

= 40
= 60
= 70
= 200
= 250
= 300
= 400

The empirical nature of this process is very apparent, when one
remembers that no two alcohols give exactly the same colour with
sulphuric acid, and consequently no two "fusel oils " can be ex-
pected to give similar results. Schidrowitz has stated that this process
gives fair results with brandy, but not with whisky. Certainly the
process based on the Allen-Marquardt process gives the best result of
any so far devised. The French official process, giving results ex-
pressed in terms of butyl alcohol, is always below the truth — since
amyl alcohol is the predominant alcohol present. The cardinal de-
fects in this process are discussed by Schidrowitz and Kaye in the
" Analyst," xxxi. 185.

Vasey recommends, as an approximate method, enabling a fair
distinction to be drawn between genuine distillates of the grape or
malt and mixtures of these with silent spirit, the following process
(which in the author's experience, although giving careful results,
does not enable the discrimination claimed for it to be made).

Ten c.c. of a distillate from spirit whose alcoholic strength has
been adjusted to exactly 50 p.c. are taken and to it are added 10 c.c.
of monohydrated H^SO^ specific gravity 1*794, in a test tube 6 inches
long by 1 inch. The contents of the tube are mixed by shaking, and
a small piece of glass tubing is dropped into the mixture. It is then
heated over a flame, and directly bubbles of steam arise from the frag-
ments of glass tubing, the test tube is withdrawn from the flame for
twenty seconds and then returned, and so on, until five minutes have
elapsed. The tube is then cooled, and the volume made up to 20 c.c.
with 50 per cent alcohol. Ten c.c of a standard solution of isobutyl
alcohol (0-2 per cent in 50 per cent alcohol) are heated in the same
manner, and the colours compared, and the approximate amount of
higher alcohoh thus calculated as isobutyl alcohol.

Bedford and Jenks have proposed a process depending on the for-
mation of nitrous acid by nitration of the alcohols, and a determina-
tion of the iodine liberated by the acid from potassium iodide, but as
neither this process nor that of Beckmann give nearly so concordant
or accurate results as the Allen-Marquardt, they are not described.
The same is true of Rose's process, which is recognized ofiicially in
Germany, and which depends on the increase in volume of a mea-
sured volume of chloroform when shaken with a measured quantity
of the alcohol at 30 per cent strength and a small quantity of sul-
phuric acid, under rigidly defined conditions.

300 FOOD AND DRUGS.

In addition to the above determination, it may be necessary to
'examine the spirit for methyl alcohol. This can be done in any of
the usual methods (see p. 280), but the following may be quoted as
being the official method in France.

Fifteen grms. of potassium bichromate are dissolved in 130 c.c.
of water, and 70 c.c. of H^SO^ (1 to 5 of water). Ten c.c. of alcohol
of about 90 to 95 per cent or an equivalent amount of a diluted spirit
to be tested, are added. The mixture is allowed to stand for twenty
minutes. It is then distilled, the first 25 c.c. being rejected and the
next 100 c.c. collected. To 50 c.c. of this 1 c.c. of dimethyl-aniline
is added, and the mixture kept at 70 to 80° for three hours with
continual shaking, in a well-stoppered bottle. It is then rendered
■distinctly alkaline with caustic soda solution (about 5 c c. of a 16 per
cent solution) and 30 c.c. are distilled off to drive off dimethyl-
aniline. To the residue in the flask, 25 c.c. of water, 1 c.c. of acetic
acid, and 4 or 5 drops of water containing lead dioxide in suspension.
The solution must be acid. In the presence of methyl alcohol, the
liquid becomes blue, the colour being intensified by boiling. Ethyl
alcohol becomes blue, changing at once to green, then to yellow, and
'becoming colourless on boiling.

It must be obvious to the meanest understanding that so long as
the determination of the groups of bodies as outlined above, be con-
sidered as a standard of purity, the addition of such bodies to what
may fairly be called " neutral spirit" can be practised, and so succeed
in deceiving the analyst.

So far, all researches on the analysis of genuine potable spirits
come to this, and to no more. Natural genuine potable spirits con-
tain certain quantities of acids, esters, aldehydes and higher alcohols.
With the determination of these, the resources of the analyst finish.
■*' Silent " spirit — such as potato alcohols, contain very little of such
bodies. All these bodies are commercial products easily obtainable.
If such bodies are added to silent spirit in due proportions — there
are no chemical means available to decide whether they are naturally
present or have been added. It is true that the absence of due pro-
portions of such bodies may prove adulteration, but presence of them
is no more than presumptive evidence of purity. The honest analyst
must stop at that until further developments may arise to assist
him.

Hence the absolute necessity of the opinion of .the trade expert in
conjunction with that of the analyst.

The analyst, however, can obtain much useful information by cul-
tivating the sense of taste, and by making up numerous mixtures of
such secondary constituents, and thus enabling himself to reject arti-
ficial spirits, not only by the results of his analysis, but also by the
use of the palate. The great danger to be avoided is in too implicit
reliance on arbitrary chemical standards.

Assuming that a spirit is genuine and not artificially prepared
with silent spirit and added secondary constituents, the following
remarks as to the interpretation of results obtained as above may be
useful : A genuine brandy will rarely contain less than 280 parts

WHISKY. 301

per 100,000 of absolute alcohol of all secondary constituents calculated
in the manner described above. Very old spirits may show a very
much higher amount than this. The proportion of higher alcohols
to esters in brandy rarely if ever varies outside the limits of from
one to two of alcohols to one of esters. In general, secondary pro-
ducts increase with age, but the furfural diminishes. Assuming no
artificial products have been added, it is safe to say that " silent "
spirit contains very little esters, and a low ester value is indicative of
adulteration.

WHISKY.

The question as to what whisky is has aroused a good deal of dis-
cussion during the past three or four years. On the one hand it was
contended that only the product of cereals distilled from a pot still
could properly be described as whisky, since the patent still elimi-
nated the greater portion of the secondary constituents of the fer-
mented liquid. On the other hand, it was contended that the term
whisky was equally applicable to pot and patent still products. Some
went so far as to allow the spirit distilled from potatoes to be termed
whisky. Indeed, a statement to this effect appears so recently as in
the 1901 edition of Vol. I of Allen's Commercial Organic Analysis (p
143). The question has now been settled, for this country at all
events, by the publication of the Eoyal Commission on whisky and
other potable spirits, 1909. The following remarks cover the whole
of the question from the point of view of the legal standard. The
Commissioners state : —

"The evidence which we received, shows that such spirits have
been frequently described as ' whisky ' by distillers and traders since
the patent still came into use ; and that for many years a section of
the pubhc, particularly in parts^ of Scotland and Ireland, has recog-
nized patent-still spirit without admixture under the name of whisky,
and has purchased it as whisky, no attempt being made by distillers
or vendors to conceal the method of distillation. Moreover, spirit
produced in the patent still, as we have shown, has long been em-
ployed for blendincr with or diluting whiskies of different character
and distilled in different forms of still. This has been by far its
largest use, and most of the whisky now sold in the United Kingdom
contains in greater or less degree spirit which has been obtained by
patent-still distillation.

" Again, apart from the fact that pot stills differ so much that a
comprehensive legal definition would be difficult to frame without
either excluding certain types of still which are now commonly re-
cognized as pot stills, or including other types which are not now
looked upon as legitimate variations of the pot still, there are strong
objections to hampering the development of an industry by stereo-
typing particular forms of apparatus.

" Finally, we have received no evidence to show that the form of
still has any necessary relation to the wholesomeness of the spirit
produced.

" For these reasons we are unable to recommend that the use of

302

FOOD AND DRUGS.

the word ' whisky ' should be restricted to spirit manufactured by
the pot-still process.

" The taste of the consumer creates the demand which ultimately
controls the trade. The public purchases the whisky that meets its
taste, and the blender must satisfy that taste or lose his trade. It is
not for the State to say what that taste ought to be.

"In our opinion, the use of the term 'Scotch' and 'Irish' as
applied to whisky cannot be denied to any whisky distilled in Scot-
land and Ireland respectively.

" Our general conclusion, therefore, on this part of our inquiry is
that ' whisky ' is a spirit obtained by distillation from a mash of
cereal grains saccharified by the diastase of malt : that ' Scotch
whisky ' is whisky, as above defined, distilled in Scotland, and that
' Irish whisky ' is whisky, as above defined, distilled in Ireland."

The general remarks with reference to the question of the second-
ary constituents of brandy, and the methods of analysis, apply in
general also to whisky. Whisky may be diluted with water to a
minimum strength of 25° under proof and sold as whisky without an
offence being committed under the Sale of Food and Drugs Acts.

The solid residue in whisky averages about O'Ol per cent, or from
0*004 per cent to 0038 per cent. The mineral matter varies from 0
to 0'02 per cent. This residue is, of course, derived from the casks in
which the spirit is stored.

For a series of analyses of whiskies of all types, reference should
be made to papers by Schidrowitz and Kaye (" Jour. Soc. Chem. Indus."
1902 and 1905).

Parts per 100,000 of Absolute Alcohol.

Volatile
Acids.

Aldehydes.

Furfural.

Esters.

Higher
Alcohols.

Observers

New whisky, Scotch .

25-4

11-4

61-9

199-4

Schidrowitz

Whisky 5 years old Scotch

'20.1

213

3.7

109-4

148-3

»»

„ 9 „

65-4

280

3 9

75-6

239-7

Girard

Irish whisky, new .

20-88

6-52

0-43

7-65

174

Vasey

„ 10 years old

51-9

14-41

346

30-44

259-5

Cuniasse

Maize whisky .

14-3

3-0

263-3

Schidrowitz

Rye „ . . .

13-5

12-8

69-4

76-2

Malt „ . . .

28-1

13-2

1-8

112-7

182

„

All malt pot still

19-5

12-6

2-2

280

Parry

All malt patent still .

6-5

0-4

120

In deahng with the question of the estimation of the higher alco-
hols in whisky by the French official process, Schidrowitz states that
it is useless in the case of this spirit, since amongst other reasons,
the higher alcohols of whisky consist of more than one individual
(probably much more variable than in the case of brandy) and as each

WHISKY.

303

:alcohol gives a very different colour with sulphuric acid, no quanti-
tative result can be of value.

The analyses of whisky given on page 302 will show the average
amounts of secondary constituents, of pure whisky. No differentia-
tion is here made, except in one case, between pot and patent still
whiskies ; analyses of " silent spirits " will be found on p. 290.

American whisky has been exhaustively examined by Crampton
and Tolman ("Jour. Amer. Chem. Soc." 1908 30, 98).

The results of their analyses of numerous samples of rye and
Bourbon whisky are as follows : —

(1)

Eye

Whisk

Grams, per 100 Litres of Proof Spirit.

Maximum
Minimum

Proof Spirit
Value.

Extract.

Acids.

Esters.

Aldehydes.

Furfural.

Fusel Oil.

132
100

339

5 (new)

112
12

126-6
4-3

26-5
0-7

9-2
trace

2803
43-7

The samples varied from quite new to eight years old.
(2) BouKBON Whisky.

Maximum

124

326

91-4

98-6

28-8

100

241-8

Minimum

100

4-0

7-2

10-4

1-0

trace

420

In interpreting the results of the analysis of potable spirits, it must
be remembered that it is only in certain cases that positive deductions
•can be drawn. Hehner has gone so far as to say the chemical analysis
cannot decide, but that the expert taster must be called in. There is
no doubt that in the large majority of cases this is true, but where a
brandy, for example, shows secondary constituents appreciably below
250 parts per 100,000 of absolute alcohol or esters much below 80 (a
low figure in itself) it is quite fair to pronounce it as mixed with
silent spirit — unless it can be shown that it was in fact made in some
parts of the world where patent stills were used.

Equally, a genuine whisky will almost invariably show a consider-
able excess of higher alcohols over esters, whereas in Jamaica rum
the esters are always far in excess of the higher alcohols.

The author cannot agree with the general deduction of Vasey as to
the utility of the analysis of these spirits (Vasey, " Analysis of
Potable Spirits," passiin). The analyst cannot be too careful in his
deduction, and except where the figures are obviously those outside

304 FOOD AND DRUGS.

the limits of a pure spirit, should hesitate to condemn a sample with-
out joining his opinion with that of an expert spirit taster.

RUM.

By rum, the spirit distilled from fermentation products of the juice
of the sugar cane was at one time invariably understood. In old
works of reference it is always so described, and in an old volume of
the eighteenth century (Shaw's " Essay on Distilling") the following
remarks occur : " Rum is usually very much adulterated in England ;
some are so barefaced as to do it with malt spirit ; but when it is done
with molasses-spirit, the tastes of both are so nearly allied, that it is
not easily discovered." Up till recently much so called " rum " was
to be found in commerce which was either made from beet sugar
molasses, or from neutral spirit which is flavoured with artificial rum
essences.

So far as this country is concerned the word rum may.be taken as
indicating the product as defined in the report of the Royal Com-
mission on Whisky and other potable spirits (1909). In the course of
their report, the Commissioners state that it was suggested during the
inquiry that the principal cause for the difference in flavour between
rums produced in various places lies in the methods of fermentation
used rather than the process of distillation. According to the evidence,
there are two distinct types of rum, Jamaica rum 'being representative
of the first, and Demerara rum of the second.

The Commissioners see no reason, however, to deny the nam'? of
rum to either of these types. They consider that the definition of
rum as " a spirit distilled direct from sugar-cane products in sugar-
cane growing countries," submitted by Mr. Aspinall, on behalf of the
West India Committee, fairly represents the nature of the spirit
which a purchaser would expect to obtain when he asks for "rum ".
The Customs already recognize the distinction between " rum," " rum
from Jamaica," and "imitation rum," and they consider that this
differentiation should be continued.

The characteristic flavour of rum is due to a mixture of esters in
which butyric and acetic esters of ethyl alcohol predominate. Arti-
ficial mixtures of esters of this type are regular commercial articles,
and are used largely in the preparation of factitious rums. The re-
marks made under brandy as to the limits to which chemical analysis
can go to apply equally to rum, as, of course, do the various processes
there described.

Rum may be sold as such, when diluted with water, provided the
strength be not below 25" under proof.

The characteristic feature of the secondary constituents of genuine
rum is the large excess of esters over higher alcohols. The following
analyses of genuine rums are due to Collingwood Williams. They are
all genuine rums, the flavoured samples not being those to which a
small quantity of fruity flavouring matter has been added — such as,
possibly, a trace of pineapple, etc. — as is sometimes stated, but are
samples whose flavour is developed by a special method of fermentation.

KUM.

305

The Jamaica rums are placed in the order of their quality as judged
by the smell of a diluted sample.

Common Clear Jamaica Rums.

Parts

per 100,000 of Absolute Alcohol.

No.

Volatile Acid.

Esters.

Higher Alcohols.

Furfural.

Aldehyde.

557

1-8

7-6

74-5

565

—

3-5

7-5

832

—

20-0

146

297

3-7

10-0

310

3-3

9-0

216

6-3

160

355

2-3

12-5

164

351

—

4-0

17-5

480

5-0

303

120

2-7

250

308

—

200

372

—

4-5

18 0

516

—

150 1

321

4-5

6-0

388

4-5

17-5

266

11-5

200

181

4-6

300

211

3-2

250

Average

333

—

4-6

75-5

Highest

146

565

—

11-6

300

Lowest

— 10

600

Flavoured Jamaica Rums.

No.

Volatile Acid.

Esters.

Higher Alcohols.

Furfural.

Aldehyde.

137

981

122

1204

—

2-9

116

662 .

—

120

787

—

2-7

17-5

599

—

4-5

37-5

1058

—

3-6

12-5

109

866

—

4-1

391

—

Average

805

VOL. I.

306

FOOD AND DRUGS.
Demerara Rums.

No.

Volatile Acid.

Esters.

Furfural.

Vat Still.

Volatile Acid.

Esters.

2-7

Average

33-1

69-9

1-6

Continuous

18-4

44-4

0-6

2-6

The following are confirmatory analyses by various observers

Samples.

Volatile
Acids.

Esters.

Higher
Alcohols.

Furfural.

Aldehydes.

Observers.

Jamaica rum .
(Average of 10 samples)

28
176

} 48

399
443

338

90-6
93-9

84-0

2-8
2-9

3-2

8-4
22-1

11-9

Vasey
Girard

Parry

Bonio (" Annales Falsific," 1909, 12, 521) states that the better-
class rums (Martinique rums) usually contain more secondary con-
stituents than lower-grade samples, but that the ratio of the esters to
the free acids and the higher alcohols is more important than the
total amount.

He gives the following analyses : —

Per 100,000 of Absolute Alcohol,

Molasses
Rum.

High grade

Average
Low grade

Sugar cane
juice rums

No.

.1
2
3
4
5

.9
10
11
12
13
14

Vola tile
Acids.

201-3

201-0

1740

165-3

173-2

145 2

196-6

158-5

53-5

80-7

80-8

83-4

42-4

Aldehydes.

34-5

16-3

14-6

10 4

10-0

19-0

18-6

17-3

Esters.

443-5
91-5
93-2
61-6
82-7

117-9
95
89-7
61
77
63-3
74-0
68-6
61-6

aSoI Furfural.

67-5
385-0
425
339
244
167

97
143
280
300
256
243
214
283

8-8
5-3
11-0
0-9
05
6-3
3-8
01
0-7
1-4
1-5
0-8
1-8
1-2

Total.

Fixed
Acids.

813

2-2

742

0-46

735

0-54

601

0-37

520

0-48

459

0-81

409

0-48

406

0-57

396

0-95

448

1-28

422

0-79

418

0-91

390

0-82

406

1-45

Ratios of Esters
to Higher Alcoholiu

6-6

0-24

0-22

0-18

0-34

0-71

0-98

063

018

0-26

0-25

0-30

0-31

0-22

GIN. 307

These figures apply to Martinique rum, but it is not easy to un-
derstand them, as Jamaica rums may be said to practically invariably
contain considerably more esters than higher alcohols. It is probable
that if these determinations had been made by the Allen-Marquardt
process, the results might have been very different.

Simon (" Annales Falsific," 1909, 12, 494) considers that the
quality of rum is mainly dependent on the esters, but that the free
acids are important as regards its flavour. In Martinique rum he also
finds the alcohols are frequently higher than the esters.

Micko (" Zeit. Untersuch. Nahr. Genuss." 1908, 16, 433) states
that he has found a peculiar, typical aromatic substance in Jamaica
rum, which is absent from artificial rums and also from rum made
in Europe from molasses. He states that by distilling a mixture of
200 G.c. of rum and 30 c.c. of water, and collecting the distillate in
fractions of 25 c.c. each, the aromatic substance in question is found
in the fifth or sixth fraction. But as he has not characterized the
body other than by describing it as " aromatic " it is of little assistance
to the analyst at present.

The solid residue of rum averages 0*3 to 05 per cent.

The author is entirely at variance, as are all other observers in
Europe, with the standards adopted by the Joint Committee of the
American Association of Oflicial Agricultural Chemists and of State
and National Food and Dairy Departments, which require that the
principal part of the secondary constituents of rum should be higher
alcohols calculated as amyl alcohol — whereas they are in fact princi-
pally esters.

GIN.

Gin is a more or less neutral alcoholic liquid, flavoured with
juniper and sometimes with other substances, and frequently
sweetened by the addition of sugar. The definition accepted by the
Koyal Commission above referred to is as follows : " Gin may be
defined as a spirit distilled from grain doubly rectified, and then
flavoured by distillation with juniper berriesand other herbs. Geneva,
also called Hollands, is a foreign spirit imported into this country :
it resembles gin, inasmuch as in both cases the genuine article is made
from grain only, and flavoured with juniper."

It may not be out of place to reproduce the opinions held in regard
to what gin should be in the middle of the eighteenth century as
stated in Shaw's " Essay on Distilling ". Dr. Chambers in his " Ency-
clopedia " (1783) states that " Geneva or gin, is a popular name for a
compound water which is, or ought to be, procured from the berries
of the juniper tree, distilled with brandy or malt spirits. The word
is from Genevre, the French name of the Juniper berry." He then
quotes from Shaw as follows : —

" The best Geneva we now have is made from an ordinary spirit
distilled a second time with an addition of some juniper berries : but
the original liquor of this kind was prepared in a very different
manner" Shaw then describes how the berries were added to the

308

FOOD AND DRUGS.

malt in the grinding and observes that " the spirit thus obtained was
flavoured ah origine with the berries and exceeded all that could be
made by any other method. Our common distillers leave out the
juniper berries entirely from the liquor they now make and sell under
that name. Our chemists have let them into the secret that the oil
of juniper berries and that of turpentine are very much alike in flavour,
though not in price : and the common method of making what is
known in London as Geneva is with a common malt spirit and a
proper quantity of oil of turpentine distilled together."

Although the Royal Commission's definition speaks of this spirit
being distilled with juniper berries, it is difficult to conceive that a
spirit which had been made with a grain spirit and essential oil of
juniper would ever be condemned as not being gin. Gin may be
diluted with water to a minimum strength of 35° under proof, and
sold as gin without an oft'ence being committed under the Sale of
Food and Drugs Acts. Various aromatic flavourings are used in
certain varieties of gin, including cardamon, coriander, angelica,
acorus calamus, grains of paradise, etc.

It must be definitely stated that the standards for secondary con-
stituents as indicated for l)randy, rum or whisky have no meaning
ivhatever as applied to gin. Gin stands quite alone in this respect,
and the analyses quoted by Vasey ("The Analysis of Potable Spirits,"
p. 25) are quite useless. They are as follows : —

Parts per 100,000 of Absolute Alcohol.

•2

Volatile Acids

1 Aldehydes.

Fnrlural.

Esters.

Higher Alcohols.

1
Observers.

nil
40-4

1-78
9-90

nil
0-3

37-28
18-50

83 66
97-00

Vasey
Girard j

Gin is essentially a neutral spirit, flavoured with the volatile con-
stituents of juniper berries — and to a lesser extent with those of
other aromatics. It is open to any maker to flavour heavily or lightly,
and the addition of many times as much flavouring in one case as in
the other, in no way alters the legal character of the gin. But, as the
essential nature of gin is its juniper flavour, and as this is due to
essential oil of juniper, it is obvious that none of the groups of com-
pounds enumerated for other spirits is the essential secondary con-
stituent of gin. Juniper oil consists almost entirely of terpenes and
sesquiterpenes, which would possibly be returned as " higher alcohols "
although not in the least related to them.

It should not be forgotten that the " secondary constituents " of
the other potable spirits are the results of natural processes in the
formation of the alcohol, and, subject to the necessary limitations
above given under "Brandy" can be dealt with as "standards,"
whereas in the case of gin, the secondary constituents are deliberately,

WINE. 309

but legitimately, added, and to what extent is merely a matter of
taste.

Apart from the alcoholic strength, gin should be tested for methyl
alcohol and if required, the sugar, when present, determined in the
usual manner in an aqueous solution of the solid residue left on evapora-
tion. If an extract be made in the same manner as by the AUen-
Marquardt process, but with the lightest petroleum ether obtainable,
and the solvent allowed to evaporate, the tasteof the residue will afford
considerable information as to the nature of the essential oil present.
If 500 c.c. be so treated, usually at least 0'5 grm. can be obtained,
which is sufficient for the determination of the refractive index.
The author has examined ten authentic samples in this manner and
found the refractive index to never fall below 1"4750, usually about
1-4770 at 20". If turpentine be used, the refractive index will fall to
1-4725 or lowar, but the author has never met a case where it has
fallen so low.

WINE.

Wine, without further qualification, is understood to be the pro-
duct of fermentation of the juice of the grape, with at most such
additions as are essential to its preservation. Public taste in various
countries has to a great extent altered the primitive meaning of the
word, in the sense that a wine containing more alcohol than a natural
fermentation of grape juice will yield, is demanded. Hence many of
the wines of to-day are fortified or increased in alcoholic strength by the
addition of alcohol. Where such fortification is effected by the addition
of brandy-^a wine product — the wine may be correctly described as
such, but where alcohol derived from another source is used, the
finished product is not in the proper sense a pure wine.

The.manufacture of wine is, of course, a subject which would re-
quire a special volume to itself, but from a broad point of view it con-
sists in the conversion of the saccharine matter dn the expressed juice
of the grape by the action of a yeast, Saccharomyces elUpsoideus, and
allied species, into other constituents, of which ordinary alcohol is the
principal. At the same time, the nitrogenous constituents of the grape
juice assist in the feeding of the organisms, and are thus changed in
their nature, and other subordinate changes in other constituents are
effected at the same time. Details of the manufacture of wine, how-
ever, do not come within the scope of the present work.

Classificatio7i of Wines. — Wines may be divided into numerous
classes, such as red and luhite, depending on the colour of the grape
used, or upon the use or rejection of the skin of the black grape ;
dry or sweet, depending on the- absence or presence in large amount of
sugar in the finished product ; still or sparkling, dependent on the
absence or presence of carbonic acid gas ; and — of course — natural
or fortified, dependent on the absence or presence of added alcohol.

The distinction of wines, however, into geographical groups, such
as Burgundy, port, sherry and so on, is a matter which has more
importance in connexion with the administration of the Merchandise
Marks Act than with that of the Food and Drugs Acts. Such wines

310 FOOD AND DEUGS.

can be instantly distinguished by the palate ; so that the sale of port
for Burgundy would be an absurdity ; but as port is recognized as the
product of Portugal shipped from the neighbourhood of Oporto, the
sale of a Spanish wine under the name of port (a not uncommon pro-
ceeding) is an offence under the Merchandise Marks Act, since it
should be described with the qualification Spanish port or Tarragona
port. The wine produced in the Burgundy district has, equally, earned
the right to the sole use of its geographical description, and wines of
the Burgundy type produced in California or Australia should be de-
scribed as such. Whether the name claret is restricted legally to a
wine produced in the Bordeaux district, is less certain.

The wine country of the world is undoubtedly France. Germany
produces excellent wines ; Portugal and Spain produce port and sherry
respectively. Austria and Italy produce much excellent wine ; and
during the past twenty years Australia and California have given to the
world wines of very high grade indeed. All these are the true product
of the grape. But so long as properly qualified, there is no reason
why the word wine should not be used for the product of fermentation
of other fruits, such as, for example, the well-recognized " British
Wines ". These are produced by the fermentation of various fruits
and are never suggested as being the product of the fermentation of
grape juice.

At the outset it is necessary to inquire what are the legitimate
additions to the juice of the grape in the manufacture of wine. In
this country there are no statutory definitions, so that we naturally
turn to the home of the industry to see what legal restrictions exist in
the matter.

In this connexion it is easy to see that the addition of a saccharine
solution to the juice of the grape will result in the formation of a
correspondingly increased amount of fermented alcoholic liquid, but
which will be proportionately deficient in the minute quantities of
secondary constituents which discriminate wine from a mere solution
of alcohol.

Every precaution is taken in France to preserve the good character
of its wine industry. By the law of 29 June 1907, amending previous
laws on the subject, every vine grower must declare his acreage and
his average produce, and the quantity of sugar which he may have
delivered to his home may not exceed 20 kilograms per head of his
establishment for personal use, subject to a heavy penalty. This law
is intended to entirely prevent the addition of sugar to the must to
increase the amount of "wine" obtained.

A decree dated September, 1907, issued under statutory authority,
defines clearly what may be sold as wine in the republic of France.

All beverages sold as wine must be derived exclusively from the
fermentation of fresh grapes, or the juice of fresh grapes. The
following " manipulations " are not to be considered as illegal in any
way in the manufacture of wine.

(1) Blending of wines.

(2) Freezing of wines to increase their alcoholic strength.

(3) Pasteurization.

WINE. 311

(4) Clarification by the use of well-known agents such us albumen,
fresh blood, casein, gelatine.

(5) The use of tannic acid in the amount necessary to effect clari-
fication by means of albumen or gelatine.

(6) Decoloration of white wines by means of charcoal.

(6) The use of SOg resulting from the combustion of sulphur, and
of alkaline bisulphites. The amounts which may be employed are
such that the wine shall not retain more than 350 mg. per litre of
SO2 free and combined. In no case shall alkaline bisulphite be em-
ployed to a greater extent initially than 20 grms. per hectolitre.

In regard to the treatment of the grape.-juice or must, a little tar-
taric acid may be added to musts not sufiBciently acid, as well as
selected yeasts where necessary.

In regard to sparkling wines, a further rule is in force, that, whilst
the artificial aeration of sparkling wines by means of COg is allowed,
such practice must be indicated on the label, by the use of the word
" fantasie,"so that the resulting wine must not be simply described as
" sparkling".

The Adulteration of Wine. — Wine is adulterated in the following
manners : —

(1) By the addition of saccharine matter to the must in order to
increase the amount of alcohol produced. Water is generally added
at the same time, so as to increase the volume of the wine also.

(2) By the addition of fermented liquors from other fruits than
the grape, and as these are usually prone to undergo acetous fermen-
tation, antiseptics are often added as well.

(3) By the addition of ordinary alcohol.

(4) By dilution with watei.

(5) By the addition of extraneous colouring matter, especially where
a pale-coloured liquid has been used as the adulterant.

(6) By plastering, that is by adding more than a small quantity
of plaster of paris, in order to fine the wine and remove tartrates.

Before passing on to the question of the analysis of wine, a few
words on natural and fortified wines may not be out of place. A
natural wine is the product of the fermentation of the pure juice with-
out the addition of sugar or alcohol. The fermentation has gone on
until either the whole of the sugar has been used up, or till the
nitrogenous food for the yeast has been exhausted, or until the
alcoholic strength is such as to check further growth of the yeast.
With an alcoholic strength of 14 per cent to 14*5 per cent by weight,
no further fermentation due to yeast can take place, so that this
figure may be taken as the highest limit for a natural wine. Forti-
fied wines have frequently received the addition of alcohol before
fermentation has finished, so that a wine of this type would be sweet
on account of a large amount of grape sugar left in the wine. Spark-
ling wines may be perfectly natural, havmg been bottled before fer-
mentation has finished, or they may be fermented to their full extent
and then bottled with a little sugar to induce a secondary fermenta-
tion in the bottle. Or they may be changed artificially with carbonic
acid gas. A dry wine has been allowed to ferment to the fullest ex-

312 FOOD AND DRUGS.

tent, even perhaps with the addition of a little nitrogenous matter
such as gelatine or albumen. It contains practically no sugar.

The analysis of wine is limited in value and rarely affords more
information than that which may establish the purity or otherwise of
a given sample. Speaking broadly, the analyst obtains no results
which enable him, qua analyst, to pass an opinion as to the quality
of a pure wine. This is essentially a matter of flavour and bouquet,
and requires the experience of a trained wine taster. It is true that
one may discriminate between a highly acid hock and one much less
acid, or a claret containing enough tannin to feel rough to the palate,
and one which is practicajly free from tannin, but excluding such
simple cases, there is no guide to the analyst which will allow a dis-
crimination between a claret worth 2s. and one worth 20s. a bottle,
or a port 5 years old and one 50 years old. It is necessary, however,
to understand the general character of the principal groups of wine,
in order to be able to correctly interpret the results of analysis. The
following are the chief types of wine in general use in this country.
A number of typical analyses are appended.

(Ij Claret. — As generally understood, claret is a red wdne of deep
colour and low alcoholic strength — from 7 per cent to 11 per cent by
volume being the usual amount. It is made in various districts in the
South of France, the Bordeaux district being the principal. Sauternes
may be taken as typical white clarets, the grapes being usually grown in
the Gironde district in the neighbourhood of Bordeaux. Eed clarets
contain very little sugar: white clarets are frequently sweet.

(2) Burgundy. — This is a red wine grown in the Burgundy district
(Cote d'or, Saone et Loire, and the Yonne). It resemble clarets in its
general character, but is of a different bouquet, rather fuller-bodied
and usually of slightly higher alcohol content. Chablis is the type of
a white Burgundy.

(3) Port is a Portuguese red wine, practically always fortified,
containing from 15 per cent to 22 per cent of alcohol by volume and
from 3 per cent to 7 per cent of sugar. It is thus a typical sweet wine.
Tarragona port is a wine of similar type made in Spain. Port de-
rives its name from Oporto, whence it is shipped.

(4) Sherry. — The " wine of Xeres " is a Spanish wine, varying in
colour from very pale yellow, to a deep brown. It is a wine which
may be either dry or sweet, the sugar varying from almost nil to a
quite considerable amount. Its characteristic bouquet is dependent
on its ethers, amongst which nitrous ethers are to be found.

(o) Hock a7id Moselle are German wines, produced in the Ehine
and Moselle districts respectively. They are characterized by the very
small amount of sugar they contain, so that they are often thought to
be considerably more acid than French wines. This is not usually
the case, however. The alcohol content varies from 8 per cent to 12
percent by volume, l^ed German wines are not drunk in this country
to any large extern.

(6) Champagne. — The wines produced in the old district of cham-
pagne— now covering the departments of the Ardennes, the Marije,
the Aube and the Haut Marne, are, of course, very varied in character,

WINE.

313

and are generally considered amongst the best class of their type. It
is the sjMrkling wine of this district, however, that is meant when
champagne is referred to in this country. This is a white wine which
is allowed to undergo some fermentation in bottle, frequently assisted
by the addition of a little sugar. It is a wine containing a fair amount
of sugar — sometimes, in very dry wines, only small quantities are
present — and from 9 per cent to 13 per cent of alcohol by volume.

(8) Madeira, made in the neighbourhood of Madeira, and Mar-
sala made in Sicily, are very similar wines of the sherry type, but
containing ethers which give to the wine a characteristic flavour.
Both are almost invariably fortified and contain 18 per cent to 22 per
cent of alcohol by vol'ime.

Italy and Hungary produce excellent wines, more or less assimi-
lating to the above types and similar wines are produced in Australia
and California. Where custom has established a name for the wine
of a given country, there is now no doubt that it is an offence under
the Merchandise Marks Acts to apply that name to the wine of an-
other country without a geographical qualification.

The following are typical characters of the principal varieties of

Sp. Gravity.

Alcohol
by Volume.

Solid Residue.

Total Acids
as Tartaric.

Sugar.

Ash.

Per cent

Claret

0-990 to 1-025

7 to 11

2-2 to 3-0

0-4 to 0-8

0-1 to 0-8

0-18 to 0-3

Burgundy

0-990 „ 1035

8 „ 12

2-2 „ 3-6

0-4 „ 0-8

0-1 „ 0-9

0-18 „ 0-3

Port

0-990 „ 1-050

15 „ 22

5 „ 14

0-3 „ 0-7

3 „ 9

0-2 „ 0-4

Sherry

0'980 „ 1-020

15 „ 20

2-0 „ 5-5

0-35 „ 0-7

2 „ 5

0-2 „ 0-5

Hock

0-988 „ 1-010

8 „ 12

1-8 „ 3-5

0-4 „ 0-9

0 „ 0-2

0-18 „ 0-4

Moselle

0-985 „ 1-015

8 „ 12

1-8 „ 4

0-4 „ 0-9

0 „ 0-4

0-18 „ 0-4

Champagne

1-040 „ 1-060

9 „ 13

10 „ 19

0-5 „ 0-8

8 „ 17

0-1 „ 0-2

Madeira

0-995 „ 1-010

18 „ 22

4-5 „ 7

0-4 „ 0-6

3 „ 5

0-3 „ 0-5

The Analysis op Wine.

In examining wine, the following determinations are made if the
fullest information is desired : —

Specific gravity.

Alcohol.

Fixed residue.

Mineral matter.

Sugar and polarization value.

Total acidity — free and volatile.

Glycerine.

Sulphates.

Sulphurous acid.

Added colouring matter.

Tartaric acid.

Tannic acid.

Salicylic acid.

Succinic acid.

Saccharin.

314

FOOD AND DRUGS.

The following represent the composition of a number of samples of
wines of various origins.

In Grms. per 100 o.o.

i ■
11

<j ft

b^<

Per

cent

French red wine *

Minimum ....

0-9890

6-4

1-96

0-190

0015

0-38

0-11

0-57

0-11

Maximum ....

10401

9-3

14-20

0-300

0062

0-78

0-84

1-04

0-30

Average ....

0-9982

7-8

2-56

0-248

0-030

0-57

0-30

073

0-18

French white wine

Average ....

0-9963

10-3

303

0-250

0-032

0-66

—

0-97

—

Swiss red wine

Average ....

09963

8-0

2-31

0-220

0-030

0-79

0-17

0-61

0-20

Swiss white wine

Average ....

0-9904

7-6

1-860

0-244

0-030

0-43

0-07

0-64

—

Tyrol red wines

Minimum ....

0-9905

6-7

1-50

0-182

0-017

0-48

0-41

0-08

Maximum ....

1-0140

11-0

6-55

0-269

0-055

0-85

114

0-27

Average ....

0-9940

9-0

2-34

0-222

0027

0-62

—

0-65

0-17

Tyrol white wine

Average ....

0-9927

8-8

1-87

0175

0-022

0-59

—

0-65

—

Austrian white wine

Minimum ....

0-9918

5-8

1-43

0'144

0-024

0-45

—

0-44

—

Maximum ....

0-9986

11-4

3-91

0-311

0048

1-04

—

1-01

—

Average ....

0-9949

7-9

2-13

0-189

0034

0-67

—

0-68

—

Austrian red wine

Average ....

0-9958

8-4

2-54

0-241

0'037

0-62

—

0-81

0-11

Hungarian red wine

Minimum ....

0-9916

6-3

1-40

0158

0-019

0-53

0-33

0-06

Maximum ....

0-0974

11-1

3-43

0-272

0-051

1-05

1-41

0-28

Average ....

0-9952

9-0

254

0-215

0038

0-67

—

0-79

0-15

Hungarian white wine

Mmimum ....

6-9907

5-4

1-45

0126

0-014

0-45

—

0-41

—

Maximum ....

0 9993

10-0

3-50

0-504

0-068

1-01

0-78

1-22

—

Average ....

09955

8-0

2-33

0204

0034

0-69

—

0-77

—

Italian red wine

Average ....

0-9940

10-5

3-44

0-290

0 032

0-52

0-44

1-45

—

Barletta ....

11-7

3-99

0-340

0033

0-36

065

1-40

—

„

0-9955

103

3-10

0-290

0031

0-60

0-50

0-85

—

0 9960

8-8

3-46

0-326

0030

0-63

0-30

0-70

—

Brindisi ....

11-2

3-83

0-280

0-034

0-55

037

0-90

—

Chianti ....

0-9960

8-2

2-36

0-234

— .

0 70

0-18

—

0-32

Sicilian wine

Average ....

1-0094

12-7

7-55

0-380

—

0-63

8-41

—

0-16

Spanish red wine

Average ....

12-1

3-53

0'610

0-027

0-49

0-38

1-09

0-22

Spanish (AUcante) sweet wine

Average ....

1-0233

12 7

9-69

0-740

0 039

0-59

6-55

063

0-20

WINE.

315

The Determination of Alcohol. — The amount ot alcohol (by volume)
may be determined by distilling 200 c.c, previously neutralized with
a few drops of alkaline solution, and to which a few pieces of pumice
stone have been added, and after collecting 150 c.c, making the dis-
tillate up to 200 c.c. and calculating the alcohol from the specific gravity
of this (see table, p. 276). If the percentage by weight be required it
is calculated by weighing the sample, and the distillate, when the
amount of alcohol may be calculated as follows : —

_ ^ , • 1 ^ • XI, 1 weight of distillate

Per cent by weight in the sample = 2_^ — — x per cent

-' ^ ^ weight 01 sample

of alcohol in distillate.

It must be remembered that all alcohol determinations based on
specific gravities are slightly erroneous in so far as esters or other
secondary constituents are present.

Tabarie's method, which gives results close to the truth, is to eva-
porate the wine until all the alcohol is driven off and make up with
water to the original volume. Then

Specific gravity of original sample .„ . „ , , ,

^ TTT-^ ^ „ ,, "^^^ ^ TTT = specific gravity of the alcohol

Specific gravity of the " extract ^ o j

present ; that is, the specific gravity of the sample minus its solid ex-
tract, from which the alcohol is at once calculated from the tables.

Alcohol in wines may also be deter-
mined with approximate accuracy by the
use of the vaporimeter. This determin-
ation depends on the fact that the vapour
tension at given temperatures of mixtures
of alcohol and water can be measured by
reference to the height of a column of
mercury which is supported by the vapour
pressure. In Geissler's vaporimeter,
which is illustrated, the bulb A, quite dry,
is filled up to the mark with mercury and
then filled completely by the addition of
the sample. The limb carrying the glass
tube B and the scale, which has been ex-
perimentally determined, is fitted into the
bulb, which the tube is ground to fit, and
the bulb with the attached limb turned
upside down, the wine thus rising to the
closed end of the bulb ; the water jacket
is now put into position and the water
bath heated. As the temperature rises .

the alcohol becomes partially vaporized ^^' '~ ^.ponme er.

and the mercury rises in the tube B. When the thermometer shows
a constant temperature of 99-5° to 100°, and the mercury becomes
steady, the amount is read off on the scale. Most vaporimeters are
graduated for both volume and weight percentage. It is obvious that
the weight percentage depends on the specific gravity of the wine —
so that if this is materially different from unity, the observed result for
alcohol by weight should be divided by the specific gravity.

316 FOOD AND DBUGS.

Fixed Besidue. — The determination of the fixed residue of wine
presents several difficulties, owing to dehydration of sugars at elevated
teroperatures, and the fact that glycerine is present in small quan-
tities. The following details should be observed, when the most
accurate results possible are obtained. If the amount of extractive is
less than 4 per cent a direct determination should be made. So
much of the sample as will not leave more than 1*5 grms. of residue
is evaporated for six hours on the water bath, and then transferred
to a water oven for two hours, cooled in a desiccator and rapidly
weighed. The evaporation should be done in a wide flat-bottomed
dish. If the amount of residue be more than 4 per cent, a consider-
able proportion is generally sugar. The results of a direct determina-
tion are then inaccurate, and the amount of fixed residue may be
calculated by determining the specific gravity of the liquid obtained
by evaporating the alcohol from the sample and making up to the
original bulk with water. On the assumption that a 10 per cent
aqueous solution of wine solids has a specific gravity 1-0386 (which
is not strictly correct, but is approximately so), the amount of fixed
residue may be calculated from the specific gravity of the de-alcohol-
ized sample by the table on following pages which is due to
Windisch.

Mineral Matter. — The dry or nearly dry residue from the evapora-
tion of 5 c.c. to 50 c.c. of the wine is ignited to whiteness and weighed.

Sugar and Polarizatio7i. — The polarization values of wine n>ay
often yield very useful results. The best methods are those laid down
by the German official processes for wine analysis (" Centralblatt f. d.
Deutsche Keich," 1896, No. 27) which are substantially as follows : An
instrument of the Schmidt and Haensch type should be used and the
results expressed in degrees of that instrument (or x 0*3468 as
angular rotation) on 200 mm. of the wine classified as described under
sugar solutions. The following inferences are to be drawn : —

1. The wine is optically inactive. Either no sugar or other
rotatory body is present in the wine, or there is a mixture of dextro-
and Isevo-rotatory sugars present. If after inversion by acid (see
under sugars) the wine becomes laevo-rotatory cane sugar is almost
certainly present. If the alcohol is driven off" and the wine made up
to its original volume, and the liquid is allowed to ferment by the
addition of 2 grms. of air-dried yeast, and if it should then be dextro-
rotatory, the indication is that laBVO-rotatory sugar and commercial
glucose, containing dextro-rotatory unfermentable substances were pre-
sent. If no change is produced by inversion or fermentation, cane
sugar, levulose and commercial glucose are absent.

2. The wine is dextro-rotatory. Cane sugar and/or commercial glu-
cose are preseni. If after inversion it becomes laevo-rotatory, cane
sugar was present. Jf after inversion it is dextro-rotatory to the
extent of over 2-25° it is practically certain that this is due lo the
presence of unfermentable constituents of commercial glucose. If,
after inversion, it is dextro-rotatory to the extent of from -f 1° to + 2*25°
it is treated as follows : 210 c.c. of the wine are evaporated to 70 c.c,
made up with water to the original volume and fermented with 2

WINE.

317

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Per cent

Per ceut

Per cent

1-0000

0-00

1-0050

1-29

1-0100

2-58

1-0150

3-87

003

1-32

2-61

3-90

005

1-34

2-63

3-93

008

1-37

2-66

3-95

0-10

1-39

2-69

3-98

0-13

1-42

2-71

4-00

015

1-45

2-74

403

0-18

1-47

2-76

4-06

0-20

1-50

2-79

408

0-23

1-52

2-82

411

10010

0-26

1-0060

1-55

1-0110

2-84

10160

413

0-28

1-57

2-87

4-16

0-31

1-60

2-89

4-19

0-34

1-63

2-92

4-21

0-36

1-65

2-94

4-24

0-39

1-68

2-97

4-26

0-41

1-70

3-00

4-29

0-44

1-73

302

4-31

0-46

1-76

305

4-34

0-49

1-78

3-07

4-37

1-0020

0-52

1-0070

1-81

10120

3-10

1-0170

4-39

0-54

1-83

312

4-42

0-57

1-86

315

4-44

0-59

1-88

3-18

4-47

0-62

1-91

3-20

4-50

0-64

1-94

3-23

4-52

0-67

1-96

3-26

4-55

0-69

1-99

3-28

4-57

0-72

201

3-31

4-60

0-75

2-04

3-33

4-63

10030

0-77

10080

207

1-0130

3-36

1-0180

4-65

0-80

2-09

3-38

4-68

0-82

212

3-41

4-70

0-85

214

3-43

4-73

0-87

2-17

3-46

4-75

0-90

2-19

3-49

4-78

0-93

2-22

3-51

4-81

0-95

2-25

3-54

4-83

0-98

2-27

3-56

4-86

100

2-30

3-59

4-88

1-0040

1-03

1-0090

2-32

10140

3-62

1-0190

4-91

1-05

2-35

3-64

4-94

1-08

2-38

3-67

4-96

1-11

2-40

3-69

4-99

1-13

2-43

3-72

5-01

116

2-45

3-75

504

1-18

2-48

3-77

5 06

1-21

2-50

7 1

3-80

409

1-24

2-53

8 I

3-82

5-11

1-26

2-56

3-85

614

318

FOOD AND DRUGS.

Sp. Gr.

Resi.lue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Per cent

10200

517

10250

6-46

1-0300

7-76

1-0350

9-05

519

6-49

7-78

9-08

5 22

6-51

7-81

910

5-25

6-54

7-83

9-13

5-27

6-56

7-86

9-16

5-30

6-59

7-89

9-18

5-32

6-62

7-91

9-21

5-35

6 64

7-94

9-23

5-38

6-67

7-97

9-26

5-40

6-70

7-99

9-29

10210

5-43

10260

6-72

10310

8-02

1-0360

9-31

5-45

6-75

804

9-34

5-48

6-77

8-07

9-36

5-51

6-80

8-09

9-39

5-53

6-82

8-12

9-42

5-56

6-85

8-14

9-44

5-58

6-88

8-17

9-47

5-61

6-90

8-20

9-49

5-64

6-93

8-22

9-52

5-66

6-95

8-25

9-55

1-0220

5-69

10270

6-98

10320

8-27

1-0370

9-57

5-71

701

8-30

9-60

5-74

7-03

8-33

9-62

5-77

7 06

8-35

9-65

5-79

7-08

8-38

9-68

5-82

711

8-40

9-70

5-84

7-13

8-43

9-73

5-87

7-16

8-46

9-75

5-89

719

8-48

9-78

5-92

721

8-51

9-80

1-0280

594

1-0280

7-24

1-0330

8-58

1-0380

9-83

5-97

7-26

8-56

9-86

600

7-29

8-59

9-88

6-02

7-32

8-61

9-91

605

7-34

8-64

9-93

6-07

7-37

8-66

9-96

610

7-39

8-69

9-99

612

7-42

8-72

10-01

615

7-45

8-74

10-04

6-18

7'47

8-77

10-06

1-0240

6-20

10290

7-50

10840

8-79

1-0390

10-09

6-23

7-52

8-82

10-11

6-25

7-55

8-85

10-14

6-28

7-58

8-87

10-17

6-31

7-60

8-90

aoi9

6-33

7-63

8-92

10-22

636

7-65

8-95

10-25

6-38

7-68

8-97

10-27

6-41

7-70

9-00

10-30

6-44

7-73

9-03

10-32

WINE.

319

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

iOr.

Residue.

Per cent

10400

10-35

1-0450

11-65

1-0500

12-95

1-0550

14-25

10-37

11-68

12-97

14-28

10-40

11-70

13-00

14-30

10-43

11-73

13-03

14-33

10-45

11-75

13-05

14-35

10-48

11-78

13-08

14-38

10-51

11-81

13-10

14-41

10-53

11-83

13-13

14-43

10-56

11-86

13-16

14-46

10-58

11-88

13-18

14-48

1-0410

10-61

1-0460

11-91

1-0510

13-21

1-0560

14-51

10-63

11-94

13-23

14-54

10-66

1196

13-26

14-56

10-69

11-99

13-29

14-59

10-71

12-01

13-31

14-61

10-74

12-04

13-34

14-64

10-76

1206

13-36

14-67

10-79

1209

13-39

14-69

10-82

12-12

12-42

14-72

10-84

1214

13-44

14-74

1-0420

10-87

1-0470

12-17

1-0520

13-47

1-0570

14-77

10-90

12-19

13-49

14-80

10-92

12-22

13-52

14-82

10-95

12-25

13-55

14-85

10-97

12-27

13-57

14-87

11-00

12-30

13-60

14-90

11-03

12-32

13-62

14-93

11-05

12-35

13-65

14-95

11-08

12-38

13-68

14-98

11-10

12-40

13-70

15-00

1-0430

11-13

1-0480

12-43

1-0530

13-73

1-0580

15-03

11-15

12-45

13-75

15-06

11-18

12-48

13-78

15-08

11-21

12-51

13-81

15-11

11-23

12-53

13-83

15-14

11-26

12-56

13-86

15-16

11-28

12-58

13-89

15-19

11-31

12-61

13-91

15-22

11-34

12-64

13-94

15-24

11-36

12-66

13-96

15-27

1-0440

11-39

1-0490

12-69

1-0540

13-99

1-0590

1530

11-42

12-71

14-01

15-32

11-44

12-74

14-04

15-35

11-47

12-77

14-07

15-37

11-49

12-79

14-09

15-40

11-52

12-82

14-12

15-42

11-55

12-84

14-14

15-45

11-57

12-87

14-17

15-48

11-60

12-90

14-20

15-50

11-62

12-92

14-22

15-53

320

FOOD AND DEUGS.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Per cent

1-0600

15-55

1-0650

16-86

1-0700

18-16

1-0750

19-47

15-58

16-88

18-19

19-50

15-61

16-91

18-22

19-52

15-63

16-94

18-24

19-55

15-66

16-96

18-27

19-58

15-68

16-99

18-30

19-60

15-71

17-01

18-32

19-63

15-74

17-04

18-35

19-65

15-76

17-07

18-37

19-68

15-79

17-09

18-40

19-71

1-0610

15-81

1-0660

17-12

1-0710

18-43

1-0760

19-73

15-84

17-14

18-45

19-76

15-87

17-17

18-48

19-79

15-89

17-20

18-50

19-81

15-92

17-22

18-53

19-84

15-94

17-25

18-56

19-86

15-97

17-27

18-58

19-89

16-00

17-30

18-61

19-92

16 02

17-33

18-63

19-94

1605

17-35

18-66

19-97

1-0620

16-07

1-0670

17-38

1-0720

18-69

1-0770

20-00

16-10

17-41

18-71

20-02

16-13

17-43

18-74

20-05

16-15

17-46

18-76

20-07

16-18

17-48

18-79

20-10

16-21

17-51

18-82

20-12

16-23

17-54

18-84

20-15

16-26

17-56

18-87

20-18

16-28

17-59

18-90

20-20

16-31

17-62

18-92

20-23

1-0630

16-33

1-0680

17-64

1-0730

18-95

1-0780

20-26

16-36

17-67

18-97

20-28

16-39

17-69

19-00

20-31

16-41

17-72

19-03

20-34

16-44

17-75

19-05

20-36

16-47

17-77

19-08

20-39

li)-49

17-80

19-10

20-41

16-52

17-83

19-13

20-44

16-54

17-85

19-16

20-47

.16-57

17-88

19-18

20-49

1-0640

16-60

1-0690

17-90

1-0740

19-21

1-0790

20-52

16-62

17-93

19-23

20-55

16-65

17-95

19-26

, 20-57

16-68

17-98

19-29

20 60

16-70

18-01

19-31

20-62

16-73

18-03

19-34

20-65

16-75

18-06

19-37

20-68

16-78

18-08

19-39

20-70

16-80

18-11

19-42

20-73

16-83

18-14

19-44

20-75

WINE.

321

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Sp. Gr.

Residue.

Per cent

1-0800

20-78

1-0840

21-83

10880

22-88

1-0920

23-93

20-81

21-86

22-91

23-96

20-83

21-88

22-93

23-99

20-86

21-91

22-96

24-01

20-89

21-94

22-99

24-04

20-91

21-96

23-01

2407

20 94

21-99

23-04

24-09

20-96

22-02

'23-07

24-12

20-99

22-04

23-09

24-14

2102

22-07

23-12

24-17

1-0810

2104

1-0850

22-09

1-0890

23-14

1-0930

24-20

21-07

22-12

23-17

24-22

2110

22-15

23-20

24-25

21-12

22-17

23-22

24-27

21-15

22-20

23-25

24-30 i

21-17

22-22

23-28

24-33 !

21-20

22-25

23-30

24-35

21-23

22-28

23-33

24-38

21-25

22 30

23-35

24-41

21-28

22-33

23-38

24-43

1-0820

21-31

1-0860

22-36

1-0900

23-41

1-0940

24-46

21-33

22-38

23-43

24-49

21-36

22-41

23-46

24-51

21-38

22-43

23-49

24-54

21-41

22-46

23-51

24-57

21-44

22-49

23-54

24-59

21-46

22-51

23-57

24-62

21-49

22-54

23-59

24-64

2152

22-57

23-62

24-67

21-54

22-59

23-65

24-70

1-0830

21-57

1-0870

22-62

1-0910

23-67

1-0950

24-72

21-59

22-65

23-70

24-75

21-62

22-67

23-72

24-78

21-65

22-70

2375

24-80

21-67

22-72

23-77

24-83

21-70

22-75

23-80

24-85

21-73

22-78

23-83

24-88

2175

22-80

23-85

24-91

21-78

22-83

23-88

24-93

21-80

22-86

23-91

24-96

grms. of air- dried yeast. The liquid is then evaporated with a little
sand and a few drops of 20 per cent solution of potassium acetate, to
a thin syrup ; 200 c.c. of 90 per cent alcohol are then added, with
constant stirring. The liquid is filtered and the whole evaporated to
about 5 c.c. The residue is mixed with bone black, filtered, and the
filter washed with water until the filtrate measures 30 c.c. If the fil-
trate has a dextro-rotation of more than 1-5° the unfermentable
constituents of commercial glucose were present.

3. The wine is laevo-rotatory. It must contain unfermented laevo-
VOL. I. 21

322 FOOD AND DEUGS.

rotatory sugar which may be natural and/or inverted cane sugar. Some
dextro-rotatory sugar may, of course, also be present. If after fer-
mentation as above described the laevo-rotation is at least - 3°, only
laevo-rotatory sugar was present. If it now rotates to the right,
laevo-rotatory sugar and commercial glucose were present. If the
lasvo-rotation is increased by inversion, both laevo-rotatory sugar and
unchanged cane sugar are present.

The German official processes are substantially as follows : —

WTiite Wines. — Sixty c.c. are neutralized with alkali, evaporated
to one-third their volume and made up again with water. Three c.c.
of basic acetate of lead solution (10 per cent solution) are added and
the liquid filtered; 31-5 c.c. of the filtrate are treated with l-o c.c. of
a saturated solution of Na^COg and filtered. The liquid is now di-
luted in the proportion of 10 to 11 so that for a 200 mm. reading a
220 bube must be used.

Bed Wines. — To the 60 c.c. of de-alcoholized wine 6 c.c. of the
lead subacetate solution are added, and to 33 c.c. of the filtrate 3 c.c.
of a saturated solution of Na2C03 are added. The filtered liquid now
represents the wine diluted from 5 to 6, so that if the reading be taken
in a 200 mm. tube it must be multiplied by 1-2.

A little animal charcoal may be used, if decolorization is not com-
plete by the use of lead subacetate.

For the inversion of the wine to correspond with the values given
above, the following process must be used : —

One hundred c.c. are neutralized, evaporated to one-third, made
up to original volume and decolorized with 2 c.c. of lead subacetate
solution, and 8 c.c. of water added. To 55 c.c. of the filtered solution
0*5 c.c. of saturated Na^COg solution is added, and 4*5 c.c. of water,
and the whole filtered. The dilution is now 5 to 6, so that the 200
mm. direct reading is multiplied by 1-2. Thirty-three c.c. of the
filtrate from the lead subacetate is now inverted by adding 3 c.c. of
istrong HCl, and heating in ten minutes to 70° C. It is then quickly
.cooled, filtered, and the rotation is a 200 mm. multiplied by 1*2. This
gives the true value after inversion.

For readings after fermentation, except in the special case above
described, 50 c.c. are de-alcoholized, made up to original volume and
kept at 30° for sixty hours with washed yeast. A few drops of a
solution of acid mercuric nitrate followed by a few drops of solution
of subacetate of lead, are then added and finally a little sodium car-
bonate solution. The whole is then filtered, made up to 100 c.c. and
the reading taken in a 200 mm. tube. The dilution is 1 to 2, so that
the reading is multiplied by 2.

For the determination of the reducing sugars, the French official
method is the volumetric process. One hundred c.c. of wine are
neutralized by sodium bicarbonate, and a few c.c. of 10 per cent
subacetate of lead solution added, excess being avoided. The volume
is made up to 110 c.c, the whole well shaken and filtered. If the
liquid is still coloured, it is shaken with some animal charcoal, and
again filtered. Five c.c. of Fehling's solution ( = 0*025 grm. of glucose)
are used for the titration. If the amount of wine necessary to de-

WINE. 323

colorize the Fehling's solution is less than 5 c.c, it is diluted suflBci-
ently for the amount finally used to lie between 5 c.c. and 10 c.c.
In calculating, it is to be remembered that 11 c.c. of the liquid are
equivalent to 10 c.c. of the wine.

In Germany the official method is the gravimetric process. If 25
c.c. of the above filtrate be used, and the results multiplied by 1-1, the
resulting precipitate can be weighed as metallic copper after reduction
by a stream of hydrogen. This is the official German process, and
the results are calculated from Weiss' tables, which are given on pp.
324-5, showing the amount of invert sugar present.

If the precipitate be weighed as CuO, then the amounts of sugar
in the above table must be multiplied by the factor 0"8.

For the determination of cane sugar 50 c.c. of the filtrate used for
the determination of reducing sugars are treated with 5 c.c. of 5 per cent
HCl and the liquid heated on a water bath for twenty minutes. The
liquid is exactly neutralized with Na^COg, evaporated slightly, rendered
faintly alkaline with Na^COg, and filtered, any residue being washed
with water until the filtrate measures 50 c.c. The amount of reducing
sugar is now determined by means of Fehling's solution as before, 11
c.c. of the solution being equivalent to 10 c.c. of the original wine.
Further, as 95 parts of cane sugar yield 100 parts of invert sugar, the
real amount of cane sugar present is given by the formula x = 0*95
(6 - a) where b is the amount of invert sugar formed after, and a
the amount before, inversion. The gravimetrc process is pre-
ferable.

Acidity. — Twenty-five c.c. of the wine are heated until boiling

N
just commences, and titrated with — potassium hydroxide. Litmus

paper should be used as an indicator, and in order to ensure

accurate results the alkali should be standardized against a solution

of tartaric acid, using the same indicator. If standardized exactly,

N
each c.c. of the — alkali is equivalent to 18*75 mg. of tartaric acid

(assuming that the whole of the free acids are tartaric, which is not

strictly true). The official French standards are calculated to grams of

N
sulphuric acid per litre, and — sodium hydroxide solution is used for

the titration, using phenol-phthalein as indicator. This is somewhat
difficult in the case of red wines, unless used in the form of spots on
a white tile.

To separate the fixed and volatile free acids, the following apparatus,
which is used officially in Germany will be found the most useful.
Fifty c.c. of the wine are placed in the flask B, which holds 200 c.c,
and a little tannic acid added in order to prevent foaming.

At first the connexion between the distilling flask and the steam-
generating flask A, is interrupted by a clip on the india-rubber por-
tion of the connexion. The wine is distilled until reduced to half its
volume, the distillate being collected in the flask C. Steam is then
turned on, the flame below B being lowered, and 200 c.c. is collected.

324

FOOD AND DRUGS.

Cu.

Sugar.

Cu.

Sugar.

Cu.

Sugar.

Per cent

Per ceut

Per cent

0-010

0-0061

0-063

0-0323

0-116

0-0607

0-169

0-0892

O'Oll

0-0066

0-064

0-0328

0-117 !

0-0612

0-170

0-0897

0-012

0-0071

0-065

0-03.33

0-118 i

0-0617

0-171

0-0903

0-013

0-0076

0-066

0-0338

0-119 !

0-0623

0-172

0-0908

0-014

0-0081

0-067

0-0343

0-120 1

0-0628

0-173

0-0914

0-015

0-0086

0-068

0-0348

0-121 1

0-0633

0-174

0-0919

0-016

0-0090

0-069

0-0353

0-122 j

0-0639

0-175

0-0924

0-017

0-0095

0-070

0-0358

0-123 1

0-0644

0-176

0-0930

0-018

0-0100

0-071

0-0363

0-124

0-0649

0-177

0-0935

0-019

0-0105

0-072

0-0368

0-125

0-0655

0-178

0-0941

0-020

0-0110

0-073

0-0373

0-126

0-0660

0-179

0-0946

0-021

0-0115

0-074

0-0378

0-127

0-0665

0-180

0-0952

0022

0-0120

0-075

0-0383

0-128

0-0671

0-181

0-0957

0-023

0-0125

0-076

0-0388

0-129

0-0676

0-182

0^0962

0-024

0-0130

0-077

0-0393

0-130

0-0681

0-183

0-0968

0-025

0-0135

0-078

0-0398

0-131

0-0687

0 184

0-0973

0-026

0-0140

0-079

0-0403

0-132

0-0692

0-185

0-0978

0-027

0-0145

0-080

0-0408

0-133

0-0697

0-186

0-0984

0-028

0-0150

0-081

0-0413

0-134

0-0703

0-187

0-0990

0-029

0-0155

0-082

0-0418

0-135

0-0708

0-188

O-O905

0-030

0-0160

0-083

0-0423

0-136

0-0713

0-189

0-1001

0-031

0-0165

0-084

0-0428

0-137

0-0719

0-190

0-1006

0-032

0-0170

0-085

0-0434

0-138

0-0724

0-191

0-1012

0-033

0-0175

0-086

0-0439

0-139

0-0729

0-192

0-1017

0-034

0-0180

0-087

0-0444

0-140

0-0735

0-193

0-1023

0-035

0-0185

0-088

0-0449

0-141

0-0740

0-194

0-1029

0-036

0-0189

0-089

0-0454

0-142

0-0745

0-195

0-1034

0-037

0-0194

0-090

0-0469

0143

0-0751

0-196

0-1040

0-038

0-0199

0-091

0-0474

0-144

0-0756

0-197

0-1046

0-039

0-0204

0-092

0-0479

0-145

0-0761

0-198

0-1051

0-040

0-0209

0-093

0-0484

0-146

0-0767

0-199

0-1057

0-041

0-0214

0-094

0-0489

0-147

0-0772

0-200

0-1063

0-042

0-0219

0-095

0-0495

0-148

0-0778

0-201

0-1068

0-043

0-0224

0-096

0-0500

0-149

0-0783

0-202

0-1074

0-044

0-0229

0-097

0-0505

0-150

0-0789

0-203

0-1079

0-045

0-0234

0-098

0-0511

0-151

0-0794

0-204

0-1085

0-046

0-0239

0-099

0-0516

0-152

0-0800

0-205

0-1091

0-047

0-0244

0-100

0-0521

0-153

0-0805

0-206

0-1096

0-048

0-0249

0-101

0-0527

0-154

0-0810

0-207

0-1102

0-049

0-0254

0-102

0-0532

0-155

0-0816

0-208

0-1108

0-050

0-0259

0-103

0-0537

0-156

0-0821

0-209

0^1113

0-051

0-0264

0-104

0-0543

0-157

0-0827

0-210

0-1119

0-052

0-0269

0-105

0-0548

0-158

0-0832

0-211

0-1125

0-053

0-0274

0-106

0-0553

0-159

0-0838

0-212

0-1130

0-054

0-0279

0-107

0-0559

0-160

0-0843

0-213

0-1136

0-055

0-0284

0-108

0-0565

0-161

0-0848

0-214

0-1142

0-056

0-0288

0-109

0-0569

0-162

0.08.54

0-215

0-1147

0-057

0-0293

0-110

0-0575

0-163

0-0859

0-216

0-1153

0-058

0-0298

0-111

0-0580

0-164

0-0865

0-217

0-1158

0-059

0-0303

0-112

0-0585

0-165

0-0870

0-218

0-1164

0-060

0-0308

0-113

0-0591

0-166

0-0876

0-219

0-1170

0-061

0-0313

0-114

0-0596

0-167

0-0881

0-220

0-1175

0-062

0-0318

0-115

0-0601

0-168

0-0886

0-221

0-1181

WINE.

325

Cu.

Sugar.

Cu.

Sugar.

Cu.

Sugar.

Cu.

Sugar.

Per ceut

Per cent

0-222

0-1187

0-278

0-1507

0-334

0-1841

0-390

0-2187

0-223

01192

0-279

0-1513

0-335

0-1847

0-391

0-2193

0-224

0-1198

0-280

0-1519

0-336

0-1854

0-392

0-2199

0-225

0-1204

0-281

0-1525

0-337

0-1860

0-393

0-2205

0-226

0-1209

0-282

0-1531

0-338

0-1866

0-394

0-2212

0-227

0-1215

0-283

0-1537

0-339

0-1872

0-395

0-2218

0-228

0-1221

0-284

0-1543

0-340

0-1878

0-396

0-2224

0-229

0-1226

0-285

0-1549

0-341

0-1884

0-397

0-2231

0-230

0-1232

0-286

0-1555

0-342

0-1890

0-398

0-2237

0-231

0-1238

0-287

0-1561

0-343

0-1896

0-399

0-2243

0-232

0-1243

0-288

0-1567

0-344

0-1902

0-400

0-2249

0-233

0-1249

0-289

0-1572

0-345

0-1908

0-401

0-2257

0-234

0-1255

0-290

0-1578

0-346

0-1914

0-402

0-2264

0-235

0-1260

0-291

0-1584

0-347

0-1920

0-403

0-2271

0-236

0-1266

0-292

0-1590

0-348

0-1926

0-404

0-2278

0-237

0-1272

0-293

0-1596

0-349

0-1932

0-405

0-2286

0-238

0-1278

0-294

0-1602

0-350

0-1938

0-406

0-1^293

0-239

0-1283

0-295

0-1608

0-351

0-1944

0-407

0-2300

0-240

0-1289

0-296

0-1614

0-352

0-1950

0-408

0-2307

0-241

0-1295

0-297

0-1620

0-353

0-1956

0-409

0-2314

0-242

0-1300

0-298

0-1626

0-354

0-1962

0-410

0-2321

0-243

0-1306

0-299

0-1632

0-355

0-1968

0-411

0-2328

0-244

0-1312

0-300

0-1638

0-356

0-1974

0-412

0-2335

0-245

0-1318

0-301

0-1644

0-357

0-1980

0-413

0-2343

0-246

0-1323

0-302

0-1650

0-358

0-1986

0-414

0-2350

0-247

0-1329

0-303

0-1656

0-359

0-1992

0-415

0-2357

0-248

0-1335

0-304

0-1662

0.360

0-1998

0-416

0-2364

0-249

0-1341

0-305

0-1668

0-361

0-2004

0-417

0-2371

0-250

0-1346

0-306

0-1673

0-362

0-2011

0-418

0-2378

0-251

0-1352

0-307

0-1679

0-363

0-2017

0-419

0-2385

0-252

0-1358

0-308

0-1685

0-364

0-2023

0-420

0-2392

0-253

0-1363

0-309

0-1691

0-365

0-2030

0-421

0-2399

0-254

0-1369

0-310

0-1697

0-366

0-2036

0-422

0-2406

0-255

0-1375

0-311

0-1703

0-367

0-2042

0-423

0-2413

0-256

0-1381

0-312

0-1709

0-368

0-2048

0-424

0-2420

0-257

0-1386

0-313

0-1715

0-369

0-2055

0-425

0-2427

0-258

0-1392

0-314

0-1721

0-370

0-2061

0-426

0-2434

0-259

0-1398

0-315

0-1727

0-371

0-2067

0-427

0-2441

0-260

0-1404

0-316

0-1733

0-372

0-2073

0-428

0-2449

0-261

0-1409

0-317

0-1739

0-373

0-2080

0-429

0-2456

0-262

0-1415

0-318

0-1745

0-374

0-2086

0-430

0-2463

0-263

0-1421

0-319

0-1751

0-375

0-2092

0-264

0-1427

0-320

0-1756

0-376'

0-2099

0-265

0-1432

0-321

0-1762

0-377

0-2105

0-266

0-1438

0-322

0-1768

0-378

0-2111

0-267

0-1444

0-323

0-1774

0-379

0-2117

0-268

0-1449

0-324

0-1780

0-380

0-2124

0-269

0-1455

0-325

0-1786

0-381

0-2130

0-270

0-1461

0-326

0-1792

0-382

0-2136

0-271

0-1467

0-327

0-1798

0-383

0-2143

0-272

0-1472

0-328

0-1804

0-384

0-2149

0-273

0-1478

0-329

0-1810

0-385

0-2155

0-274

0-1484

0-330

0-1816

0-386

0-2161

0-275

0-1490

0-331

0-1822

0-387

0-2168

0-276

0-1495

0-332

0-1828

0-388

0-2174

0-277

0-1501

0-333

0-1835

0-389

0-2180

326

FOOD AND DRUGS.

The distillate is titrated with standard alkali, and the results calculated
to acetic acid. By deducting the amount of alkali used for the
neutralization of the volatile acids for 100 c.c. of wine, from that used
for the total acids, the remainder is calculated into tartaric acid and
returned as fixed acids.

Glycerine. — Approximate re-
sults may be obtained by the use
of the process devised by Trillat
(" Comptes Eendus," 135, 903),
which is as follows : —

Fifty c.c. of wine is evaporated
in a small silver dish on the water
bath at 70° C. to one third of its
volume. Five gi-ms. of animal
charcoal are then added, inti-
mately mixed with the residue,
and evaporation continued to
complete dryness. After cooling,
this residue is mixed with o grms.
of quicklime. The powder thus
obtained is transferred to a flask
and agitated for five minutes with
30 c.c. of pure dry acetic ether.
The liquid is decanted and filtered,
and the powder extracted twice
more with the same quantity of
solvent. The acetic ether is then
evaporated, in small quantities at
a time, in a tared capsule on the
water bath, then dried to constant
weight at 60° C. and weighed. It
may then be ignited and the ash
weighed, this weight being de-
ducted from that of the glycerin ;
but, as a rule, the amount of ash
is so small that it may be disre-
garded.

The German official method is tedious but gives fairly exact results
except in the case of plastered wines when they are too high. The
following are the details of this method : —

(1) Wines containing less than 2 per cent of sugar : 100 c.c. are
evaporated down to 15 c.c. on a water bath, and 1 grm. of fine sand
added. Two c.c. of 40 per cent emulsion of lime are added for each
grm. of fixed residue present, and evaporation continued. When the
water is nearly driven off, 5 c.c. of 96 per cent alcohol are added. The
particles adhering to the sides of the dish are loosened with a glass
rod and rubbed into a cream with a little more alcohol. The mixture
is heated on the water bath with constant stirring until it begins to
boil, when the liquid is decanted into a 100 c.c. flask. The residue in
the dish is extracted with five or six portions of 10 c.c. of 96 per cent

Fig. 36. — Apparatus for determining
volatile acids in wine.

WINE. 327

alcohol, each portion being decanted into the flask, which is then made
up to 100 c.c. with alcohol. After filtration, 90 c.c. of the filtrate are
evaporated in a porcelain dish on the water bath, which is only al-
lowed to boil very gently. "When the alcohol is driven off the residue in
dish is washed out with three successive portions of 5 c.c. each of abso-
lute alcohol, which are transferred to a graduated cylinder, and made up
to exactly 15 c.c. with absolute alcohol. Three successive portions of
7*5 c.c. of absolute ether are added to the contents of the cylinder
which are well shaken after each addition. When the solution is
quite clear it is transferred to a tared glass dish, and the cylinder
washed out with a mixture of 2 c.c. of alcohol and 3 of ether. The
alcohol-ether is evaporated in a warm water bath — care being taken
that the solvent does not actually boil — the residue is dried in a water
oven for one hour, cooled in a desiccator and rapidly weighed.

Wines containing more than 2 per cent of sugar : 50 c.c. are warmed
in a large flask on the water bath, and 1 grm. of fine sand, and milk
of lime until the colour is quite pale, are added. On cooling 100 c.c.
of 96 per cent alcohol are added > the precipitate is allowed to subside,
and the liquid filtered, the precipitate and filter being washed with
strong alcohol. The filtrate is then treated as in the former case.

Stierlin's method is to evaporate the liquid, without addmg any-
thing, to one-fifth or one-sixth of its volume. He then extracts with
hot absolute 'alcohol, and estimates sugar, non-volatile acids, alka-
loids, bitter matters and glycerin in this alcoholic extract. Glycerin
is estimated by freeing a given quantity from alcohol by evaporation,
then again evaporating to dryness with a slight excess of caustic Hme.
It is then extracted with alcohol and ether (2 : 3), or alcohol and
chloroform may be used.

Kaynaud has stated that the processes used for the estimation of
glycerin cannot always be depended upon, especially with plastered
wines, when the results obtained are too high, since lime de-
composes a large amount of sulphate of potash and hydrate of potash
is formed, which is dissolved by glycerin when alcohol is present,
and is of course weighed with it. He suggests the following process.
Evaporate the liquid to one-fifth of its volume, and precipitate the potash
by hydrofluosilicic acid. Then filter the liquid. Add baryta water to
make slightly alkaline, also a small amount of sand, and evaporate to
dryness in a vacuum : to extract the dry residue add a very large
quantity of absolute alcohol and ether ; as much as 300 c.c. for 250 c.c.
of wine, can be used. This, however, is unnecessary with proper
extracting apparatus, and 50 c.c. to 100 c.c. in a Soxhlet's apparatus
will have just the same effect. When the alcohol and ether have
evaporated the glycerin should stand for twenty-four hours in a
vacuum over phosphoric anhydride ; it is then put into a tube, a
perfect vacuum formed, and at a temperature of 180° it will distil
into the cool part of the tube.

A useful method of estimating glycerine is that of Parthiel, by
distillation in a vacuum to separate the more volatile substances and
oxidation of the glycerin to oxalic acid. He takes 50 c.c. of the
liquid and adds a little calcium carbonate to neutralize it. It is

328 FOOD AND DRUGS.

evaporated down to 15 c.c. and placed in a small retort which is en-
closed in an air bath, the bottom of the bath being made of sheet iron,
while the sides and top are made of asbestos card. A globular re-
ceiver is connected with the neck. The second opening of the receiver
is connected with an inverted condenser, and then to a pump. The
receiver is kept cool. The liquid is first distilled almost to dryness at
ordinary pressure, the temperature being 120° C. Then it is cooled to
60'' C. The pressure is reduced by means of a pump, the temperature
being 80" C. and it is now distilled for one and a half hours : then the
vacuum is broken, the retort cooled, 10 c.c. of water are added and
distillation is continued at the ordinary pressure, the temperature
in the bath being 120° C. The distillate is then diluted to 200 c.c,
8 grms. to 10 grms. of caustic soda are dissolved in it, and 5 per cent
of potassium permanganate are added until there is an unmistakable
blue-black colour. The whole should then be heated for an hour,
SO^ added to decolorize, 20 c.c. of acetic acid added, the whole heated
to get rid of SO.2 and the oxalic acid precipitated by calcium chloride.

The iodide method proposed by Zeisel and Fanto for the estimation
of glycerin in wine is quite trustworthy. In this method the wine is
prepared for analysis by treating 100 c.c. of it with tannin and barium
acetate, distilling off about 70 c.c. and diluting the residue to 100 c.c.
Five c.c. of this solution are then distilled in a current of carbon di-
oxide after the addition of hydriodic acid. The isopropyl iodide
formed by the action of the hydriodic acid on the free and com-
bined glycerin distils over, and after being passed through a small
wash bottle containing amorphous phosphorus suspended in water,
is collected in an alcoholic solution of silver nitrate. The quantity of
silver iodide produced corresponds with the amount of glycerin
present.

The lime method, in the case of wines containing not more than 5
per cent of sugar, yields results which are somewhat lower than those
obtained by the iodide method.

With sweet wines much lower results are obtained by the lime
method than by the iodide method.

There are numerous other methods for the estimation of glycerin,
but all of a more or less complicated character and none yielding
«triptly accurate results.

For these reference may be made to the original publications, as
follows : —

Bordas and de Raczkowski (oxidation by chromic acid, " Comptes
Rendus," 1896, 1021).

Bottinger (conversion into triacetin," Comptes Rendus," 1897, 240).

Sulphates. — The significance of any excessive quantity of sulphates
in a wine has reference to the practice known as plastering. A good deal
has been said against the practice of adding a small amount of calcium
sulphate to wines, but so long as but little is used, the facility with
which the wine is clarified entirely outweighs any sentimental disad-
vantages attached to the process. The greater part of the cream of
tartar present in the wine is converted into insoluble calcium tartrate,
which mechanically carries down various impurities in the wine which

WINE. 329

would otherwise require an exceedingly long time to settle down.
The practice of plastering is general in the sherry district — indeed the
author is informed by leading wine experts that it is a commercial
necessity with this wine, but it is also resorted to to a lesser extent in
other districts. There is always a small amount of potassium sulphate
present in grape juice, and a small quantity results from the practice
of sulphuring the casks, the SO^, generated becoming oxidized to sulp-
huric acid. An unplastered wine will contain sulphates to the ex-
tent of O'l per cent calculated as potassium sulphate. Any excess
over 0"2 is usually accepted as evidence of plastering ; indeed, any excess
over O'l per cent is nearly always due to plastering. Native wines
in Germany and most wines in France or Switzerland are not allowed
to be sold with over 0'2 per cent of sulphates, calculated as potassium
sulphate. The sulphates are determined by evaporating lOO'c.c. to
about. 30 c.c, and adding excess of hot solution of BaCl^ in the
usual manner, after acidification with hydrochloric acid, and weighing
as BaSO^.

Sulphitrous Acid. — The presence of sulphurous acid in wine may
be due to traces being absorbed from sulphured casks, or it may be
due to the addition of sulphurous acid or sulphites for the purpose of
preserving the wine, a quite necessary precaution for certain types of
wine. Sulphurous acid, when added to wine, appears to enter, to a
very considerable extent, into combination with normal constituents
of the wine, leaving a relatively small amount in the free state. In the
determination of sulphurous acid it is customary to return the SO2 as
" free " and " total," the total including this combined or "aldehyde "
sulphurous acid. The combined sulphurous acid is regarded as almost
innocuous, whilst objection is taken to more than traces of free acid.
The usual official limits are 200 mg. of SOg per litre for the total, or
20 mg. to 30 mg. per litre for free SOg.

A recent investigation of the white wines of France has shown that
these limits are a serious hindrance to commerce, and that it is not a
fact that it is the sweeter wines which require the most SO^ for pre-
servation. . Wines containing but little sugar frequently contain so
much of constituents which combine readily with SO.,, that if limit
quantities are added, the whole is almost at once combined. The
leading experts in the Bordeaux white wine trade consider that the 350
mg. per litre allowed for these wines should be raised to 400 mg., with a
10 per cent allowance to meet special cases, without distinction of free
or combined SO.,, or a limit of 100 mg. of free SO3 per litre without
reference to the amount of combined SO^.

The total SO.^ may be determined by passing a current of CO^
through a 400 c.c. flask, the CO^ entering through one tube through
an india-rubber cork, the exit tube leading to a tube containing ab-
sorption bulbs. The absorption tube contains 50 c.c. of a 5 per cent
solution of iodine (in 7-5 grm. KI per litre). When the air is dis-
placed, 100 c.c. of the sample and 5 c.c. of syrupy phosphoric acid are
poured into the flask and the cork at once replaced. The current of
CO.2 is allowed to proceed, and after fifteen minutes, the contents of
the flask are carefully heated, so that half of the contents distil over

330 FOOD AND DEUGS.

into the absorption tube in the current of CO^. The tube should be-
kept cold by immersion in cold water. The contents of the absorption
tube are then poured into a beaker and the sulphuric acid formed by-
oxidation of the SOo precipitated and weighed as barium sulphate.
The amount of BaSO^ x 2-7468 gives the amount of SO.^ per litre.

The free SO^ is determined by diluting the wine if red — but not
if white — and adding a little sodium carbonate. Excess of dilute
sulphuric acid is then added and, with the flask filled with CO.^, the
SO.j can be titrated with one-twentieth normal iodine, using a little
starch water as indicator. The results are approximately accurate.
Each c.c. of one-twentieth normal iodine is equivalent to 1-6 mg. of SO.^.

Colouring Matter. — The examination of the colouring matter of
wine is not an easy matter. The following three tests are the French
official methods : —

1. Fifty c.c. of the wine rendered alkaline by ammonia are
shaken with 15 c.c. of pure amyl alcohol. The amyl alcohol should
remain colourless and after separation should remain colourless when
acidified with acetic acid.

2. The wine is treated with a 10 per cent solution of acetate of
mercury until the precipitate formed does not change colour, when a
slight excess of magnesia is added, to render the liquid alkaline. The
mixture is then boiled and filtered. The liquid, acidified with dilute
sulphuric acid, should remain colourless.

(3) Fifty c.c. of wine are placed in a porcelain dish and 2 drops
of 10 per cent sulphuric acid added, and a small piece of white wool
plunged into the liquid. The liquid is boiled for 5 minutes, water
being added to replace the loss due to the boiling. The wool is then
washed in a current of water. It should then have only the slightest
rose-coloured tint, and when dipped into ammonia solution, it should
only yield a very pale green colour.

Other useful tests are as follows : —

Five c.c. of a 10 per cent solution of subacetate of lead are added
to 20 c.c. of the wine. If the precipitate is of a red-violet colour, it is
nearly certain that a vegetable colouring matter has been added —
probably that extracted from the berries of Phytolacca decandra. If
the liquid be heated and then filtered, and amyl alcohol extracts any
red colour from the filtrate, the addition of foreign colouring matter
is certain.

Test No. 3 alone may be modified by first mordanting the wool by
dipping it into a solution of alum and sodium acetate, and the test is
then carried out as above. If the colour of the wool is a deep red,
aniline colours are present,- and this may be confirmed by the behavi-
our of the wool when treated with ammonia. The pale reddish
colour which may result with natural wines is changed to a dirty,
pale green. The red due to aniline colours is either unchanged or
turned to a yellowish tint, the red colour being restored by washing
out the ammonia.

Dupre's gelatine test is a useful one in the hands of one having
had experience of the test. Small gelatine cubes are prepared, by soak-
ing 5 grams of gelatine in water and then adding hot water to 100 c.c.

WINE. 331

When the jelly has set, it is cut into cubes of about | in. on each sur-
face. If one of these cubes be inserted in the sample for forty-eight
hours, the colouring matter of pure wine will be found, on cutting
sections of the cubes, to have only penetrated just below the surface,
most other colouring matters, including aniline reds, cochineal, log-
wood, beetroot, litmus, etc., will permeate the jelly and colour it
nearly, if not quite, throughout.

There are certain aniline colours which escape the lead acetate test
given above. These however are identified by Cazeneuve's oxide of
mercury test.. Ten^ c.c. of the sample are shaken with 0*2 grm. of
yellow mercuric oxide for at least a minute, and the liquid filtered,
preferably after boiling. A clear but coloured filtrate indicates the
presence of coal-tar colours — but a colourless filtrate does not prove
their absence.

There are other coal-tar colours which may be detected by tbe
following method : Two portions of 100 c.c. each of the wine are
extracted with ether, one of them being rendered alkaline by 5 c.c. of
ammonia.

The ether from both is evaporated in a porcelain dish with a thread
of wool. If the wool in the experiment in which ammonia was used is
dyed red, a coal-tar colour was present.

The experiment in which no ammonia was used may give a
brownish tint with pure wines, but no pronounced red colour.

The French official test with amyl alcohol should be performed not
only on the wine rendered alkaline, but also on the natural and the
acidified wine. The following are the inferences to be drawn from the
results of these tests.

(1) On the natural wine. A small amount of red colouring matter
may be extracted from pure wines by amyl alcohol. But it will be
changed to a green or blue-green by the addition of ammonia. Any
red colour, not so changed, is a powerful indication of added coal-tar
colouring matter.

(2) On the alkaline wine. A red extract with amyl alcohol is a
fairly certain indication of a coal-tar colour. The ammonia should not
be present to a greater extent than 3 c.c. of ordinary strong ammonia
(specific gravity = 0-880) per 100 c.c.

(3). On the acidified liquid. Eed colouring matter is extracted
from pure wines. The amy! alcohol should be shaken with water, and
the aqueous solution tested by ammonia, which, in the presence of
coal-tar colours, leaves the solution red and not green, or it may be
tested by dyeing wool as described above.

The above reactions are sufficient for any ordinary case. Lengthy
researches — usually yielding inconclusive results — as to the exact
nature of the added colouring matter when present, may be under-
taken, but it is rarely necessary to go further than deciding if foreign
colouring matter be present or not.

A systematic examination of the colouring matter was published in
1876 by Gautier, and may be consulted in the "Analyst " (xxi. 1).

Tartaric Acid. — The total amount of tartaric acid present (either
free or as potassium bitartrate) may be estimated by mixing 20 c.c. of

332 FOOD AND DRUGS.

the wine v?ith 1 c.c. of 10 per cent solution of potassium bromide
and 40 c.c. of a mixture of equal volumes of ether and alcohol. The
whole is well shaken and left in a closed flask for three days. The
liquid is then decanted on to a small filter, which is washed with a
little of the alcohol ether mixture. Forty c.c. of warm water is then
passed through the filter into the original flask, the contents of which
are warmed until the acid tartrate of potassium is dissolved. This is
now titrated with one-twentieth normal alkali, using phenol-phthalein
as indicator. From the amount of alkali used, the percentage of tar-
taric acid can be at once calculated, from the following formula,
which allows for the solubility correction : —

(?ix 0-47) + 0-2

gives the number of grms. of tartaric acid per litre, where n = the
number of c.c. used.

The genuine character of certain French wines has for many years
past been called in question by German chemists on account of their
abnormal values for total extract and free tartaric acid. Apart from
the fact that many of the wines contain less than the lowest propor-
tion of extract regarded as genuine in Germany, most of them also
contain a considerable amount of tree tartaric acid. In accordance
with German regulations, the free tartaric acid in wines containing
not more than 0'8 grm. of total acids in 100 c.c. should not, as a rule,
exceed one-fifth to one-sixth of the total non-volatile acids ; in wines
containing more than 0*8 grm. of total acids per 100 c.c. the propor-
tion of free tartaric may be higher. French chemists, on the other
hand, have repeatedly asserted that these wines from the South of
France are genuine, and that the free tartaric acid may amount to
about 2-0 grms. per litre. In order to obtain complete certainty on
the points in dispute, the wines from grapes grown near Barbonne
have been examined, and the results show that the frequently high
proportion of free tartaric acid must be attributed to the conditions of
the soil and not to any additions of that acid. The results on opposite
page are typical of those obtained, the amounts of total and free
tartaric acid having been estimated by the official German method of
Halenske and Moslinger (" Zeit. Anal. Chem." 1905, 34, 279).

Tan?iic Acid. — Any one of the modified Lowenthal's processes
(see p. 11) may be used for this determination. Ten to twenty c.c.
of the sample should be employed, according to the astringency of the
wine.

Salicylic Acid. — For the detection of salicylic acid in wine, see
pp. 679 et seq.

Disinfection of wine barrels is sometimes carried out by means of
formaldehyde (paraformaldehyde being volatilized in the barrels for
this purpose), and when this is done the formaldehyde is removed by
treating the barrels with sodium carbonate solution, water and steam.
After this process has been carried out distinct traces of formaldehyde
were found in a wine which had been stored in a barrel disinfected
with formaldehyde fourteen days previously. For the detection of
formaldehyde the reaction described by Arnold and Mentzel (" Analyst,"

WINE.

333

"3 •

t»

5*1

^Mp

1hPh

■3 •

|.2 §

-f

"'l^

1^1

,— t

r-t

r-l

r— 1

'■J3 -S

OS w 0

o3 8

•«*

tr-

Ifi

"?*

EH«< g

t<-l

0 .

>>^-;

^z ^

»c

.2.S 0

»p

l>-

-s^ w

g^s^h

-<*<

■rt<

»o

■s-^^

T)<

^ 5r!

«(S

— -4^

Co a

XTb

'"'

2 >.

■*

Tjt

S *^

r-l

C<1

•73

• ^

^ s

§ g

E3 OS
0 >-*

■ PI

1 ^
<1

CS
i-H

OS
rH

334 FOOD AND DEUGS.

XXVII. 227) is recommended, but, as wine frequently contains sulphur-
ous acid, some preliminary treatment is necessary to decompose the
formaldehyde-sulphurous acid compound before the test is applied
to the distillate of the wine. The wine is distilled after the ad-
dition of phosphoric acid, and the distillate is rendered alkaline with
potassium hydroxide. After the lapse of fifteen minutes the solution
is neutralized with sulphuric acid, and any formaldehyde test is then
:applied. For the estimation of formaldehyde the method proposed
by Legler may be used ; it is based on the combination of the for-
maldehyde with ammonia according to the equation : —

6CH,0 + 4NH3 = N,(CH2), + 6H,0

The distillate from a definite volume of the wine is treated with alkali

and then neutralized with sulphuric acid, using rosolic acid or litmus

as indicator. An excess of standardized ammonia is next added, the

mixture is set aside for three hours, and the excess is then titrated

N
with — - sulphuric acid. The quantity of formaldehyde present is then

•calculated from the ammonia used to combine with the aldehyde.

Succinic Acid. — The best method for the determination of succinic
acid in wine is that of Raw (" Zeit. fiir Anal. Chem." xxxii. 482).
■One hundred c.c. of the sample are evaporated to a syrupy con-
sistency, repeatedly extracted with boiling alcohol, and the cooled
alcoholic solutions filtered, mixed, and distilled. The residue is dis-
solved in a little hot water, and the cooled solution filtered, if turbid ;
it is then treated with barium nitrate, 3 to 4 vols, of 90 per cent
alcohol are added, and the mixture is well stirred. The precipitate,
•containing tartaric, malic, and succinic acids, is collected, washed well
with 70 per cent alcohol, warmed with sodium carbonate solution, and
filtered ; the filtrate is neutralized with nitric acid, evaporated to a
small bulk, and after neutralization with ammonia, is precipitated with
a magnesia mixture, made with magnesium nitrate, ammonium nitrate,
and ammonia. The precipitate, which contains the tartaric acid, is
filtered off after three or four hours' repose ; the filtrate is heated with
potash until all the ammonia is expelled, then filtered from magnesia,
neutralized exactly with nitric acid, diluted to 100 c.c. to 150 c.c, and
precipitated with silver nitrate (1 : 20). Silver nitrate precipitates
succinic acid completely, but produces precipitates in malic acid
solutions only when they are stronger than 1 : 800 The precipitate
is collected on a tared filter, washed well dried, and weighed. As a
control, it may be ignited and the silver weighed. Should the solution
to which silver nitrate is to be added contain chlorides, which may
happen if too much alcohol has been added after the barium nitrate,
or too long an interval has been allowed before filtration, a portion of
it must be evaporated, incinerated, the chlorine determined, and a
corresponding quantity of silver chloride subtracted from the weight
of the silver succinate.

The method described by Kunz (" Analyst," xxviii. 314) has been
subjected to a critical examination, and it is found that while the
process as a whole is the best of many methods which have been

WINE. 335

proposed for the estimation of succinic acid, certain modifications in
the method of procedure are necessary in order to obtain accurate
results. The process, as modified by Heide and Steiner, is as
follows : —

Fifty c.c. of the wine are evaporated in a basin of about 200 c.c.
capacity until the alcohol has been removed ; after the addition of 1
c.c. of 10 per cent barium chloride solution and a little phenol-
phthalein, the residual solution is treated with powdered barium
hydroxide until all the acidity has been neutralized. Excess of barium
hydroxide is removed by treating the mixture with carbon dioxide, and
85 c.c. of 96 per cent alcohol are then added to the mixture with con-
stant stirring. After the lapse of at least two hours the precipitate,
consisting of barium succinate, tartrate, and malate together with other
barium salts, is collected on a filter, washed with a small quantity of
80 per cent alcohol, and then washed back again into the basin by the
aid of a jet of boiling water. The contents of the basin are now heated
until all alcohol has been removed, and 5 per cent potassium perman-
ganate solution is then added in quantities of about 3 c.c. at a time
until the red coloration does not disappear after the mixture has stood
for five minutes. A further 5 c.c. of potassium permanganate are
then added, and the mixture is heated on the water bath for fifteen
minutes. The excess of permanganate is destroyed by the addition
of sulphurous acid, and after the mixture has been acidified with sul-
phuric acid, more sulphurous acid is added until all the manganese
dioxide has been redissolved. The mixture is then evaporated to a
volume of about 30 c.c. and extracted with ether for twelve hours in
a percolating apparatus after the addition of so much 40 per cent sul-
phuric acid that the solution contains about 10 per cent of free sul-
phuric acid. The ethereal extract is diluted with water, the ether is
evaporated, and the residual solution, after neutralization, is trans-

N
ferred to a 100 c.c. flask, 20 c.c. of — silver nitrate solution are added

and the whole is diluted to the mark and filtered. The excess of sil-
ver is then titrated in 50 c.c. of the filtrate, Volhard's method being
used for the purpose.

Saccharin may be detected by the method described on page 674.

Identification of Inosite m Natural Wines. — The fact that all
natural wines contain inosite affords, according to Perrin (" Ann. de
Chem. Anal. Appl." 1909, 14, 182) a simple means of distinguishing
them from artificial products. For the identification of inosite 200
c.c. of the wine are treated with 20 c.c. of basic lead acetate solution
and a few drops of alcoholic solution of tannin, and filtered. The fil-
trate is freed from lead by means of hydrogen sulphide, and the
filtrate from the lead sulphide decolorized with animal charcoal and
concentrated to about 10 to 20 c.c. on the water bath. The following
tests are then applied to this liquid : —

(1) Two drops are heated on platinum foil with one drop of a 10
per cent solution of silver nitrate, and the residue carefully ignited.
In the presence of an inosite a violet-rose coloration is obtained. This
disappears on cooling, but reappears on again heating the foil.

336 FOOD AND DRUGS.

(2) Two drops of solution are heated on platinum foil with one
drop of nitric acid and the carbon incinerated as before. The residue
is then treated with a drop of ammonia solution and the liquid again
evaporated. A rose coloration which is less pronounced than that
obtained in the first test indicates that the wine contained inosite.

Hexamethylene Tetramine m Wines. — The French Minister of
Agriculture has recently drawn attention to the fact that hexamethylene
tetramine is being used for the purpose of removing sulphites from
wine. This is considered to be a fraudulent proceeding and in a cir-
cular issued by the Minister the following methods of detecting it are
recommended

(1) Strongly acidify a few c.c. of the wine with sulphuric acid,
and thea add an equal volume of solution of rosanilin bisulphite. An
intense violet colour results if this body be present.

(2) Distil 20 c.c. of the wine after acidifying with a few drops of
sulphuric acid. Collect the first 5 c.c. add to it 1 c.c. of dilute sul-
phuric acid and then 5 c.c. of solution of rosanilin bisulphite. An
intense violet colour results if hexamethylene tetramine be present.

Numerous other reactions are available for which see Blarez (" Bull,
de la Soc. de Pharm. de Bordeaux," 1910, February) and Voisenet
("Ann. de Chim. Anal." 1910, 266).

The Significance of Certain Results in Wine Analysis. — In the
author's opinion, too much stress is frequently laid on certain
analytical determination on samples of wine. The following are the
most important deductions that can be drawn : —

(1) Alcohol. Any excess of alcohol over 14'5 per cent by weight
may be regarded as definite evidence of added alcohol. It must be
remembered that many wines, especially the sweet wines of Spain
and Portugal, are regularly fortified, and must therefore be judged as
such, and certain German wines also regularly receive a small addition
of alcohol.

(2) Glycerine. Genuine wines usually contain 0'4 per cent to 1
per cent of glycerine but these limits are sometimes exceeded.

(3) Alcohol-glycerine ratio. The researches of Pasteur tended to
show that the ratio of alcohol to glycerine in wines undergoing normal
fermentation was nearly constant, but later researches have thrown a
somewhat different light on the subject. Semichon has shown that
where traces of sulphurous acid or fluorides, for example, have been
added to the must, less glycerine is developed during the fermentation,
Mathieu has shown that the more acid the must naturally is, the
greater the amount of glycerine developed. Roos has made exhaustive

investigations which appear to prove that the ratio — -. — shows

glycerme

great variations, whilst the ratio -. — ; r^ is far more constant.

succmic acid

It is generally to be found that the alcohol-glycerine ratio of a genuine,

unfortified wine, lies between 100 : 5 and 100 : 15 but even these limits

are sometimes exceeded.

(4) There is a fair amount of constancy to be observed in the
amount of solid extract in wines, so long as they have not been kept

MALT LIQUORS. 337

long, if allowance be made for sugar, and, when the wine is plastered,
potassium sulphate. Official French standards include a so-called
" reduced extract," which is defined as ic - (S - O'l) - (K - 0*1)
where x = the percentage of extract, S that of sugar, and K that of
potassium sulphate. The " reduced extract " of red wines falls
usually between the values 1*8 to 2-6, and of white wines I'O to 2-6.
Even in old wines the reduced extract rarely falls below 1*5 per cent.

(5) The ratio ^j — r— ; is used by French chemists to a considerable

extent as an indication of watering of wine. Gautier has advocated
the sum of the total acids and the alcohol as an indication of water-
ing, as he considers that the free acids are higher as the alcohols are
lower, and that the sum of the two is fairly constant ; but Hapter con-
acid
siders the ratio -i — ^-^ to be of greater value. Gautier's value is that

usually accepted in France. The figures are expressed by the sum
of the alcohol (in per cent by volume) and the acid (as grm. of HgSO^
per litre) ; this figure is not less than 12-5, and for most wines any
lower figure is strong evidence of adulteration with water.

(6) The ratio ^ — — '— is regarded with much im-

reduced extract

portance in France.

In genuine, unfortified, red wines this figure never exceeds 4-5 to
4-6, and for white wines 4-8 to 6-5. Any higher ratio indicates the ad-
dition of alcohol.

(7) If the ratio indicated in 6 shows that alcohol has been added,

then the ratio —^ — r— f as shown in (5), must be adjusted to the

amount of alcohol naturally present, if the question of added water is
to be settled. Thus, if a wine (red) contains 12 per cent alcohol and
1"5 per cent reduced extract, it is obvious that alcohol has been added.
If the extract be multiplied by 4*5 the approximate amount of natural
alcohol is obtained (by weight). This is then divided by 0*8 to give
the percentages by volume. To this value — 8'5 — the acidity in grm.
of £[280^ per litre is added, and if the total be less than 12*5 it may
be inferred that water has also been added.

MALT LIQUORS.

Beer, including ale, stout, etc., in this country is not restricted, as
it is in Bavaria, for example, to the product of fermentation of a
mixture of barley malt, hops, and water, by the aid of yeast. It is
more properly described as the product of the fermentation of a sac-
charine infusion, suitably bittered by a harmless bitter substance.

Many of the best brands of beer, however, are brewed from nothing
but malted barley and hops, so that some account of malt is neces-
sary.

In the preparation of malt, the grain — barley or other grain — is well
sieeped in water and after fermentation, it is dried and heated, or
cured, in a kiln. The principal changes taking place during this pro-
VOL. I. 22

338

FOOD AND DRUGS.

cess are an increase in the amount of soluble carbohydrates at the
expense of the starch, and a conversion of the insoluble nitrogenous
matter into a more soluble form. The following figures illustrate the
composition of barley malt. They are due to 0' Sullivan and are both
pale malts : —

Starch

Other carbohydrates (of which 60 to 70 per cent is ferment-
able sugar

Cellular tissue

Fat

Albumenoids —

(a) Soluble in alcohol of sp. gr. 0*820, and in cold water .
(6) Soluble in cold water and at 68° . . . .
(c) Insoluble in cold water but soluble at 68° to 70°
{d) Insoluble at 68° to 70° but soluble in cold water
(e) Insoluble in cold water and at 70° .

Ash

Water

Per cent

44-15

45-13

21-23

19-39

11-57

10-09

1-65

1-96

0-63

0-46

3-23

3-12

2-37

1-36

0-48

0-37

6-38

8-49

2-60

1-92

5-83

7-47

Well-malted barley is of a yellowish colour, unless it has been
heated in order to partially caramelize the sugar, in order to obtain a
high-coloured malt such as is used for stouts and porters.

Good malt should float on cold water. It should not be too
hard, being easily crushed between the fingers, but at the same time
should be crisp. The acrospire should be from two-thirds to three-
fourths of the length of the grain, but should in no case protrude, as
too much albuminoid matter would be extracted in the washing.
Malt dried at about 32° to 36° is pale in colour and is used for the
palest grades of beer. Malt dried at from 38° to 50° is used for the
various grades of beer up to very dark brown beers, whilst much
higher temperatures are employed when the malt is used for black
beers.

The principal object in malting is to produce a relatively large
amount of diastase, a ferment capable of converting starch into the
soluble carbohydrates, maltose, and dextrin. Malt, however, contains
far more diastase than is necessary to convert the starch contained
therein into maltose, so that for many beers, considerable amounts of
X)ther grain, such as rice, are added to a small quantity of malt.

The Valuation of Malt. — The quantity and nature of the water-
soluble constituents of malt, together with its diastatic value, are the
principal chemical criteria of its quality.

The following may be regarded as standard methods for this
country (being recommended by a special committee appointed by
the Council of the Institute of Brewing {vide " Jour. Inst, of Brew-
ing," 1906, 12).

In the report of this committee it is recommended that samples
should be from 10 per cent of the number of sacks, and the samples

MALT LIQUOKS. 339

should be taken from a depth of at least six inches below the surface.
The samples should then be ground in a Seek mill set at 25°. Only-
enough for each determination is ground at a time, and the determina-
tion at once proceeded with.

Extractive Matter. — Three hundred and sixty c.c. of distilled
water heated to 155° F. are mixed with 50 grms. of the ground malt,
in a beaker of about 500 c.c. capacity. The beaker is kept covered
with a clock glass and kept at 150° F. for fifty-five minutes in a warm
water bath. The temperature is then raised to 158° in five minutes,
and the mixture washed into a flask graduated to 515 c.c. (the volume
of the insoluble part if 50 grms. of malt are assumed to occupy
a volume of 15 c.c), the whole cooled to 60° F., made up to the
mark with water, well shaken and filtered through well-ribbed filter
paper.

The amount of extract in the clear wort is now deduced by taking
the specific gravity, deducting 1000 (water = 1000) and dividing the
remainder figure by 4. This gives the grms. of extract in 100 c.c. of
the wort. [The figure 4 is not universally accepted, figures varying
from 3*8 to 3 95 being used by some workers.]

In the brewing trade, the value of worts is usually expressed in
pounds per barrel, this being the number of lb. in excess of 360 con-
tained in a barrel of 36 gallons. An instrument used largely in this
connexion is the hydrometer known as Bates' saccharometer, which is
graduated to read lb. per barrel directly. These readings are con-
vertible into specific gravities by dividing the value by 0'36 (or
multiplying by 2 '778) and then adding 1000. Thus a barrel of
wort weighing 380 lb., is said to have a saccharometric value of 20
lb. per barrel. The specific gravity of this would be 1055*5 (water =

360 1000

1000) since oqq == iqkk.k 5 and from the above-given figures it would

contain 13*8 grms. of residue per 100 c.c. or 50*1 lb. per barrel.
And a wort of specific gravity 1*055 (the standard strength at which
the duty per barrel is levied) has a saccharometric value 19'80 (1055 -
1000) X 0-36. The above considerations, of course, apply to unfer-
mented worts.

The solid matters of worts consist largely of maltose, with dextrins,
albuminoids, ash, soluble starch, etc.

The Colour of the Wort. — The colour of the wort is a matter of im-
portance to the brewer's chemist, but has no particular interest in
connexion with the analysis of the finished product. For particulars
of this reference should be made to the " Journal of the Institute of
Brewing " (1906, 12, 302 and 1907, 13, 26).

Moisture. — About 5 grms. are dried at 99° to 100° for five hours in
a shallow dish.

Diastatic Vahie. — Ling's method of carrying out the determination
of the diastatic value (Lintner value) yields most concordant results.
Twenty-five grms. of ground malt are exhausted with 500 c.c. of
absolutely pure distilled water at 70° F., and filtered. When the
filtrate is perfectly clear, 3 c.c. are allowed to react with 100 c.c. of a
2 per cent solution of soluble starch at 70° F. for one hour, in a 200

340 FOOD AND DRUGS.

c.c. flask. The starch solution is prepared by digesting pure potato
starch with HCl (sp. gr. 1'037) for a week at ordinary temperatures,
stirring well each day. Two grms. of acid should be used for each
grm. of starch. The powder should then be repeatedly washed with
water until free from acid, freed from water as far as possible by
means of an exhaust filter, and allowed to dry on a clean unglazed
tile. It is then dried at about 110'' F. for a short time. Two grms.
of this are then dissolved in 100 c.c. of boiling water.

At the end of the hour's reaction of the malt extract and soluble
starch, 1 c.c. of normal potash solution is added to stop further
action, the liquid cooled to 60° F. and made up to 200 c.c. It is
then titrated in the following manner : Five c.c. of Fehling's solution
are measured into a 150 c.c. flask and heated to boiling. The solu-
tion of starch (converted) is then run in 5 c.c. at a time, the liquid
being well shaken and kept at boiling temperature. After each addi-
tion, the liquid is well boiled, a drop is withdrawn by a glass rod,
and brought into contact with a drop of ferrous thiocyanate, spotted
on a white tile. So long as any cupric salt remains the red colour of
ferric thiocyanate is at once developed, so that the end reaction is
well marked. The results are to be calculated by the formula A =

, where A = the diastatic activity in the empirical Lintner de-

grees, x = the number of c.c. of malt extract prepared as above de-
scribed in 100 c.c. of the fully diluted (i.e. to 200 c.c.) converted
starch liquid, and y = the number of c.c. of the solution necessary to
reduce 5 c.c. of Fehling's solution.

For malts showing a higher value than 50, 2 c.c. only of the malt
extract — or if the value be over 80, only 1 c.c. — should be used. An
alternative method of determining the Lintner value is described
under Extract of Malt, which, as a matter of convenience, is described
with the carbohydrate food stuffs.

Cold Water Extract. — There is considerable difference of opinion
as to the value of this figure, some chemists holding that if a malt
has been forced in its rate of sprouting, too much starch is converted
into soluble sugar ; other che;iiists see no objection to the use of a
forced malt. Twenty-five grms. of ground malt are digested with
250 c.c. of distilled water containing 2 c.c. of normal solution of
ammonia, for three hours at 70° F., with occasional stirring. The
specific gravity of the filtrate is then taken and the excess of this
figure over 1000, multiplied by 10 and divided by 3*86, gives the per-
centage of extract, which averages 16 to 20 per cent. The correction
in specific gravity due to the trace of ammonia is practically negli-
gible.

Nitrogenous Matter. — If this value is required, it should be deter-
mined in the usual manner by Kjeldahl's method (p. 403). The
average, based on the multiplying factor 6*25, is 10 to 11*5 per cent.

Arsenic. — The determination of arsenic may be carried out as de-
scribed under drugs (pp. 662 et seq.).

The wort, either from the malted barley or a mixture of this with
other grain, is boiled in order to concentrate and utilize it, when hops

MALT LIQUORS.

341

or other bitter material are added to it, and the boihng continued.
After cooling the clear liquid is run into fermenting vats where
properly selected yeast, usually Saccharomyces cerevisiae, but some-
times other species, is added, and alcoholic fermentation is allowed
to proceed. If the fermentation be conducted at low temperatures,
from 4° to 10^ C, it is a slow process and yeast settles at the
bottom of the liquid. If the process be conducted at 15° to 22°, quick
fermentation goes on and most of the yeast rises to the surface and
is skimmed off. Considerable differences in the finished beer result
according to the condition of fermentation. The following are the
appearances of the top and bottom yeasts which grow in the two pro-
cesses of fermentation.

Fig. 37.— Top yeast.

Fig. 38. — Bottom yeast.

In England the following are the principal varieties of beer
brewed : —

Beer is generally used to indicate the pale to amber-coloured ale,
as distinguished from stout.

Sto2U is the black variety of beer, prepared by the use of a
caramelized malt.

Porter is a weak variety of stout.

There are several varieties of " lager " beer brewed in this country,
in imitation of German beers. But in the author's opinion (based on
experience as a juror in the beer section of the Paris Exposition, 1900),
no English-brewed " lager " beer is identical in character with the
high-grade German lager beers. The distinction in the varieties of
German beer are not to-day of so much importance as they were
formerly, owing to the improvements in methods of storage, refrigera-
tion, etc.

" Schenck " or winter beer is a quickly fermented beer made for
immediate use, containing a low proportion of alcohol, so that it will
become acid by keeping.

" Lager "or " summer " beer, so named because it is stored —
lager = a store-house — is higher in alcohol than schenck beer, and will
keep for a much longer time ; formerly it was brewed in the winter
and kept for summer use.

Bock beer occupied an intermediate position and v/as a fairly strong
beer which would keep longer than schenck beer.

Export beer is the strongest variety of lager beer made, but is

342

FOOD AND DRUGS.

usually pasteurized before it is sent abroad, where it is consumed
fairly rapidly.

The following represent the compositions of a number of typical
beers examined by Konig : —

-.1

c
■c

0-120

.2 .

Sohenk beer

1-0144

91-11

0-197

3-36

5-34

0-74

0-95

3-11

0-156

0-204

0-055

Lager „

1-0162

90-08

0-196

3-93

5-79

0-71

0-88

3-73

0-165

0-153

0-228

0-077

Export „

1-0176

89-01

0-209

4-40

6-38

0-74

1-20

2-47

0-154

0-161

0-247

0-074

Bock

1-0213

87-87

0-234

4-69

7-21

0-73

1-81

3-97

0-176

0-165

0-263

0-089

Ale . .

1-0140

88-00

0-200

5-00

6-40

0-54

0-95

1-70

0-250

0-260

0-300

0-160

Porter.

1-0200

88-10

0-190

4-90

9-60

0-60

2-40

2-80

0-240

0-250

0-340

0-085

Some strong Burton and Scotch ales will contain as much as 12
per cent to 14 per cent of extract and from 8 per cent to 10 per cent
of alcohol. Porter rarely contains more than 6 per cent to 7 per cent
of alcohol, the sample examined by Konig being exceptionally low.

The ash of beer has the following average composition : —

Per cent

Soda 8-94

Potash 33-67

Lime 2-78

Magnesia • • • 6-24

FeA 0-48

PA . 31-35

Chlorine • . . 2-93

SO3 3-47

SiOa 9-29 .

The physical differences as indicated by taste are of more import-
ance than the chemical characters in discriminating the various
qualities of beers. The typical differences between English beers and
German beers are due to the fact that English beers are generally pre-
pared by a top fermentation at a more elevated temperature than that
employed in the fermentation of German beers, which are produced
by bottom fermentation. The yeasts used, too, are different varieties.
The German beers contain less alcohol, but more dextrin, sugar and
nitrogenous substances than English beers.

The Analysis 0/ Beer. — The duty on beer is calculated from the
strength of the unfermented wort as indicated by its specific gravity.
So that where beer is exported and a rebate of duty claimed, it be-
comes a matter of importance for the analyst to be able to determine
from the finished beer what this value was. Since the amount of
alcohol is about 50 per cent of the sugars fermented, it is clear that a
determination of the alcohol will give the means of determining the
original specific gravity of the wort. The reduction of the specific

MALT LIQUOES.

343

gravity by fermentation is known technically as the " attenuation " of
the wort.

To obtain this figure the specific gravity of the beer freed from
alcohol by evaporation and made up to its original volume, and the
specific gravity of the alcohol distilled from the beer, made up to its
original volume, are taken.

The beer is first freed as much as possible from COg by pouring
from one vessel to another and filtering through either cotton wool or
paper and then 100 c.c. are diluted with 40 c.c. of water, and about
80 c.c. distilled. Both the distillate and the residue are made up to
100 c.c. and the specific gravities taken.

The distillate now represents the fermented matter as a mixture of
alcohol and water, whilst the residue indicates the unfermented
matter left from the original wort. The specific gravity of the distil-
late is subtracted from 1000 (specific gravity of water) and the differ-
ence is called the degree of spirit indication.

From the table compiled by Graham, Bed wood and Hobhouse
which is legalised for use by the excise in this country, the degree of
specific gravity lost by the fermentation is found, and this figure when
added to the " extract gravity " — i.e. the gravity of the de-alcoholized
beer made up to its original volume — gives the specific gravity of the
original wort. The table is as follows : —

Fractions of a Degree of Same.

Degree of Spirit
Indication.

0-1

0-2

0-3

0-4

0-5

0-6

0-7

0-8

0-9

0-3

0-6

0-9

1-2

1-5

1-8

2-4

2-7

3-0

3-3

3-7

4-1

4-4

4-8

5-5

5-9

6-2

6-6

7-0

7-4

7-8

8-2

8-6

9-0

9-4

9-8

10-2

10-7

11-1

11-5

120

12-4

12-9

13-3

13-8

14-2

14-7

15-1

15-5

16-0

16-4

16-8

17-3

17-7

18-2

18-6

19-1

19-5

19-9

20-4

20-9

21-3

21-8

22-2

22-7

23-1

23-6

24-1

24-6

25-0

25-5

26-0

26-4

26-9

27-4

27-8

28-3

28-8

29-2

29-7

30-2

HQ-7

31-2

31-7

32-2

32-7

33-2

33-7

34-3

34-8

35-4

35-9

36-5

37-0

37-5

38-0

38-6

39-1

39-7

40-2

40-7

41-2

41-7

42-2

42-7

43-2

43-7

44-2

44-7

45-1

45-6

46-0

46-5

47-0

47-5

48-0

48-5

49-0

49-6

50-1

50-6

51-2

51-7

52-2

62-7

53-3

53-8

54-3

54-9

55-4

55-9

56-4

56-9

57-4

57-9

58-4

58-9

59-4

60-0

60-5

61-1

61-6

62-2

62-7

63-3

63-8

64-3

64-8

65-4

65-9

66-5

67-1

67-6

68-2

68-7

69-3

69-9

70-5

711

71-7

72-3

72-9

73-5

74-1

74-7

75-3

75-9

For example, if the " extract gravity " be 1*0420, and the specific
gravity of the alcohol distillate 0*9905 ; then the degree of spirit in-
dication is 1000 - 9905 = 9-5. From the table the corresponding
degree of gravity lost is 41'7. So that 1-0420 + 0-0417 = 1-0837 is
the original gravity of the wort.

These values are based on the presence of about O'l per cent of
free acid calculated as acetic acid in the beer. In the case of sour

344

FOOD AND DRUGS.

beers or beers containing much free acid, a correction must be added
to the apparent spirit indication. This may be obtained from the
following table, which allows for the conversion of the small quantities
of alcohol into acetic acid.

^-^ of a Per cent.

Excess of Acetic Acid
over 0-1 Per cent.

•00

•01

•02

•03

•04

•05

•06

•07

•08

•09

•02

•04

•06

•07

•08

•09

•11

•12

•13

•1

•14

•15

•17

•18

•19

•21

•22

•23

•24

•26

•2

•27

•28

•29

•31

•32

•33

•34

•35

•37

•38

•3

•39

•40

•42

•43

•44

•46

•47

•48

•49

•51

•4

•52

•53

•55

•56

•57

•59

•60

•61

•62

•64

•5

•65

•66

•67

•69

•70

•71

•72

•73

•75

•76

•6

•77

•78

•80

•81

•82

•84

•85

•86

•87

•89

•7

•90

•91

•93

•94

•95

•97

•98

•99

1-00

1^02

•8

1-03

1-04

1^05

1^07

1-08

1^09

110

111

113

114

•9

115

1^16

1-18

1-19

1^21

122

1^23

1-25

1-26

1^28

1^29

131

133

1^35

1-36

1^37

1-38

1^40

1^41

142

The acetic acid for the purposes of this correction should be deter-
mined by deducting the amount of free fixed acids as determined by
titration of the dried residue of the beer redissolved in water from the
total free acids, both being calculated as acetic acid. Decinormal
solution of ammonia, using red litmus as indicator, should be used
for the titrations.

For example, if in the above illustration the free acidity, of the
beer (volatile acidity) equal 0-33 per cent (i.e. 0-23 in excess of O'l
per cent) then the above table gives the correction as *31. This is added
to the apparent spirit indicator value, namely 9-0, which is now 9'81»
From the former table this figure gives a " specific gravity lost " of
43-2, which, added to the extract gives 1-0852 as the original gravity
of the wort.

An alternative method for determining the original specific gravity
of the wort consists in determining the " spirit indication " by merely
deducting the specific gravity of the beer itself from that of the beer
from which the alcohol is expelled, made up to original bulk with
water. The alcohol need not be collected if this determination is
made.

The calculations and corrections for acidity are identical with the
former method, except that when the " gravity lost " is found from
the table one-fortieth of the figure found must be added as a correction
For example : —

Specific gravity of the de-alcoholized beer made up to \olume. 1-046

Specific gravity of the beer ....... 1*035

Difference = spirit indication 11

Specific gravity lost (from table) 49

Add one-fortieth . . • 1^22

Corrected " gravity lost " 50*22

Add " extract gravity " 1^046

Original gravity of wort 1-0692

MALT LIQUOES. 345

The following determinations on the finished beer are at times
necessary. Alcohol, extractive, residual fermentable matter, acids,
glycerine, reducing sugars and dextrin, proteids, phosphoric acid,
chlorine, carbonic acid, detection of bitter principles, arsenic, and
preservatives.

Alcohol and Extract. — The alcohol may be determined by distil-
ling 80 per cent of the beer freed as much as possible from CO2 and
making the distillate up to original volume, whence the alcohol is de-
duced from the specific gravity by the table on p. 275. The extract
is not accurately determined by evaporation as maltose is dehydrated
at temperatures over 75" C. Unless very accurate results are required,
however, drying at 100'' gives fairly approximate results. A fairly ac-
curate method is to de-alcoholize the beer, make up to original volume,
and take the specific gravity of the liquid. The excess of 1000 of this
value', divided by 4, gives the amount in grms. of dry residue per 100
c.c. If is probable that the divisor should be a little lower than this,
but 4 is usually accepted, and gives fairly accurate results.

An accurate method for the determination of the alcohol and the
•extract in beer depends on the determination of the refractive index
•of the beer itself (E) and of the distillate from the beer, made up to
•original volume (E'). For full details of this the following papers
may be consulted : Ackermann and Steinmann (" Zeit Gesamt.
Brau.," 1905, 28, 259 and 1906, k9, 146); Ling and Pope ("J. Fed.
Inst, of Brewing," 1901, 7, 170) and Eace (" J. Soc. Chem. Ind."
1902, 27, 544).

Residual Fermentable Matter. — One hundred c.c. of the beer are
heated to drive off the alcohol, made up to original volume with
water and fermented with 1 to 2 grms. of pressed yeast for forty-
eight hours. The refermented liquid is boiled to drive off alcohol, made
up to original volume with water and a little cream of alumina and
filtered. The " maltose " is now determined by reduction of Fehling's
solution (p. 156) in both this liquid and in the original beer itself, first
deprived of its alcohol. The difference between the two values gives
the amount of residual fermentable matter in the original beer, in
terms of maltose (but in reality including degradation products of
maltose).

Free Acids. — The beer is freed from carbonic acid by pouring
from vessel to vessel and filtering through paper or wool, and then

N
titrated with — solution of soda or ammonia using litmus as indi-
cator. The total acidity is usually expressed as lactic acid, each c.c.
of -— alkali being equal 0-009 grm. of lactic acid. The fixed acidity

is determined by evaporating to dryness and titrating the residue

after re-solution in water. This is expressed as lactic acid. The

difference between the two values is calculated to acetic acid (1 c.c.

—, alkali = 0'006 grm. acetic acid) and expressed as such.

Glycerin. — This is determined, when necessary, as directed under
wine (p. 326), except that the milk of lime is added after the CO., is

34.6 FOOD AND DEUGS.

expelled ; and it is advisable to use a further portion of alcohol and
ether in the extraction.

Beducing Sugars and Dextrin. — Fifty c.c. of the beer are diluted
to 200 c.c, heated on a water bath for two and a half hours with 20
c.c. of HCl (specific gravity 1"13), almost neutralized by soda solution,
made up to 300 c.c. when cold, filtered and the dextrose determined by
reduction of Fehling's solution.

Determine the amount of reducing sugars in the beer by a direct
reduction of Fehling's solution, and calculate as maltose. Multiply
this value by 0*95 (to convert to dextrose equivalent) and subtract
the product from the amount of dextrose after the above inversion,
calling the result the dextrose derived from dextrin. Multiply this
figure by 0*9 which gives the dextrin in the beer.

Proteids. — Fifty c.c. are treated with 5 c.c. of dilute HgSO^, and
then concentrated to a syrup. The' nitrogen is then determined by
Kjeldahl's process, and multiplied by 6-25 to give the proteid value.

Phosphoric acid and other mineral constituents. The principal
value in the determination of the mineral constituents of a beer lies
in the fact that owing to the characters of a given water supply it is
often possible to decide whether a given beer is the product of a
given brewery. Further by English excise regulations no excess over
50 grains of sodium chloride per gallon is allowed in beer. The phos-
phoric acid may, except in very dark beers, be determined by titration
with uranium acetate solution in the usual manner. The uranium
solution should be standardized so that 20 c.c. corresponds to O'l
grm. PgOg, using potassium ferrocyanide in spots on a white tile as
indicator. Fifty c.c. of the beer are titrated, each c.c. of uranium
solution being equivalent to 0*01 per cent of PgO^.

In dark beers, the ash must be moistened with HCl, dried and
boiled with 50 c.c. of water, which is then titrated as above.

All malt beers contain the highest amount of PgOg.

The chlorides are determined by evaporating 50 c.c. of beer, with

a little NagCOg, and incinerating at as low a temperature as possible,

until a black ash is obtained, which is extracted with water in the

N
usual manner, and the chlorine titrated with —silver nitrate.

The determination of sulphates is often of importance from the
point of view indicated above. It is important, in making this de-
termination, that the beer should be evaporated in the presence of a
little sodium hydroxide, or there will be a loss of SO^.

Carbonic Acid. — The determination of CO^ is not often required,
but if it is, it is obvious that the case of beers bottled with the
common screw stopper cannot be dealt with, as the act of opening the
bottle causes an immediate loss of the gas. In other cases a metal
champagne tap may be inserted through the cork, and this is con-
nected by rubber tubing with (1) an empty safety flask whose exit
tube passes on to a series of four absorption tubes, of which the first
three contain (1) calcium chloride (2) sulphuric acid, and (3) sulphuric
acid, in order to absorb moisture, and (4) concentrated solution of
caustic potash in which the CO., is absorbed. The tap is turned on

MALT LIQUOES. 347

slightly so that the gas bubbles through the first three absorption
tubes slowly, and when no more gas is given off the water bath in
which the bottle is standing is heated gradually, so that all the COg
is driven off. The difference between the weights of the COg bulb
after and before the process gives the amount of CO.^.

The Bitter Substances in Beer. — Most of the bitter substances al-
leged to have been found in beer are, to-day at all events, apocryphal.
Alkaloids — apart from cases of poisoning — are rarely if ever to be
found in beer. The absence of a precipitate with any of the usual
alkaloidal reagents (see p. 502) is sufficient guarantee of the absence
of added alkaloids.

Quassia, chiretta, gentian, and aloes may be present, the former
two especially. Since the bitter principle of hops is readily soluble
in ether, it follows that if a sample of beer be evaporated to a syrupy
consistence and extracted with ether, and the ether separated and
evaporated, the absence of a bitter taste in the ether extract is proof
of the absence of hops ; whilst the presence of a bitter taste is not
proof of the presence of hops.

The following methods will usually be found sufficient, in addition
to the above, for the examination of the bitter principles in beer. The
bitter principle of hops is completely precipitated by solution of lead
acetate or subacetate, so that the filtrate after such treatment, when
concentrated, has no bitter taste unless some other bitter be present.
The filtrate should be treated with HgS to remove excess of lead, and
then concentrated to a syrup. If ether extracts anything with a
bitter taste from this concentrated filtrate, it is certain that a foreign
bitter has been used.

Chapman (" Analyst," xxv. 35) proposes to differentiate
between hops and quassia by the following process : 500 c.c.
of the beer are evaporated to dryness with a little sand, the whole
being constantly stirred, and the residue is dried in an air oven, and
powdered. It is then extracted with ether and after the ether is eva-
porated, the residual extract is oxidized by the careful addition of an
alkaline solution of potassium permanganate (40 grms. of permangan-
ate and 10 grms. of KOH per litre). This should be added gradually
with shaking and warming. When the permanganate is but slowly
reduced, a few drops of hot solution of oxalic acid are added, which
decolorizes the slight excess of permanganate,, and the colourless liquid
is filtered and evaporated to dryness. The dry residue is then treated
with dilute sulphuric acid, when, if hops be present, the distinctive
odour of valerianic acid is evolved. Neither chiretta nor quassia give
this result, whereas chamomiles give a distinct valerianic acid reaction.
If quassia be present, the chloroform extract from the beer (rendered
slightly acid with H^SO^) when dried, gives, with a weak alcoholic
solution of ferric chloride, a distinct mahogany brown colour ; or, if
the residue be treated with bromine and ammonia, a bright yellow
colour is given.

Chiretta is indicated by the ether residue giving a straw colour,
changing to a dull purple-brown, with bromine and ammonia. The
chloroform residue does not yield this reaction.

348

FOOD AND DRUGS.

Gentian is indicated by the chloroform residue (from acidified
beer) yielding a carmine-red colour when treated with warm concen-
trated H^SO^. A trace of ferric chloride converts this into a green-
brown colour.

The bitter principle of aloes is indicated by treating the dried
residue of 200 c.c. of beer with warm ammonia, and filtering and
cooling the resulting liquid and then adding HCl. Aloe resin is pre-
cipitated and is collected. It is insoluble in ether, chloroform or
petroleum ether, but soluble in alcohol. Its taste and odour render
it easy for identification when so collected.

The following outline of processes for the examination of the
bitter substance in beer is due to Allen (Vol. I, 4th edition, p. 162,
Baker).

One thousand c.c. of beer are evaporated to 500 c.c. and neutral
lead acetate solution added : the liquid is boiled for fifteen minutes
and filtered hot. If any precipitate separates on cooling, the liquid
is again filtered.

Precipitate con-
tains hop bitter
and chiretta
bitter.

Filtrate. Remove Pb by H^S ; filter : concentrate to 150 c.c. and taste. It any bitter
taste remains, the liquid is acidified with dilute HoSOj and repeatedly shaken
with CHCI3.

Chlorofmyi layer, on evaporation, leaves a bitter ex-
tract in the case of gentian, calumba, quassia
(and old hops) and only slight bitter taste with
chiretta. Residue is dissolved in a little alcohol,
hot water added and the hot solution treated
with NH;< and basic lead acetate and filtered.

Precipitate contains the
bitters of old hops,
gentian or caramel. It
IS suspended in water,
decomposed by H2S,
and the solution
shaken with CHCl-j.

Chloroform
lajrer con-
tains gen-
tian, or
old hop
bitter.

Aqueous
layer con-
tains
traces of
caramel
bitter.

Filtrate boiled to remove
NH;{, slight excess of
H.2SO4 added, the
liquid filtered and
tasted. If bitter it is
shaken with CHCl;; and
the residue tasted. If
bitter it indicates cal-
umba or quassia.

Aqueous liquid is now well shaken with ether.

Ether layer leaves bitter residue in
the case of chiretta, gentian or
calumba. It is dissolved in a
little alcohol, hot water added,
and the NH3 and basic lead ace-
tate solution. It is then filtered.

Precipitate sus-
pended in wa-
ter and decom-
fosed by H-^S.
f filtrate is bit-
ter, gentian i-
indicated.

Filtrate is treated
with slight ex-
cess of HoSt)4

and tasted.
Bitter taste in-
dicates calumba
or chiretta.

Aqueous liquid
is, if still Dit-
ter, rendered
alkaline and
extracted
with ether-
chloroform.
A bitter ex-
tract indi-
cates cal-
umba or
strychnine.

Saccharin. — This may be searched for as described under saccharin
(p. 674).

Preservatives. — Salicylic acid may be searched for as described
under that acid (p. 679).

Sulphites are sometimes present in beer, and may be searched for
by adding 5 c.c. of phosphoric acid to 500 c.c. of beer, and distilling 250
c.c. of the liquid. The SO^, present is estimated by titrating with
one-hundreth normal iodine solution. One c.c. = 0-00032 grm. SOg.
The presence of SOg may be confirmed by precipitating the oxidized
SOg in this distillate by BaCl, in the presence of HCl.

CIDER.

349

Fluorides are occasionally to be found in beer. To detect this
200 c.c. of the sample are rendered alkaline with ammonium carbon-
ate, boiled with 2 c.c. of a 10 per cent calcium chloride solution, for
ten minutes, and the precipitate collected, washed, and dried. The
dried precipitate is ignited (without the paper) in a platinum crucible,
powdered, and moistened with 2 or 3 drops of water and 1 c.c. of
strong H.,SO^. The crucible is covered with a watch-glass which is
coated with wax and the wax cut through to the glass with a style.
The crucible is now warmed on a water bath, and if fluorides are
present the glass will be etched. The wax is kept from melting by
coating the convex surface, and placing pieces of ice on the upper con-
cave surface.

Boric acid may be found as described under milk (p. 60).

CIDER.

Cider is understood to be the product of alcoholic fermentation of
apple juice.

Much cider is made from bruised fruit, and, especially in America,
with but little care as to the cleanliness employed during the fermen-
tation process. In such cases it is necessary to use a preservative,
or the cider will rapidly turn to vinegar. France has paid consider-
able attention to the manufacture of cider, and most experts consider
that French ciders are the best obtainable.

There is no doubt that the absence of a legal definition of cider in
this country has materially prejudiced the development of the industry
in a healthy manner. In France the legal position is as follows (Dur-
ham, " Journal of the Royal Institute of Public Health, May 1908 ") :
" No drink may be sold as cider unless it is the product of fermenta-
tion of the juice of fresh apples, unless it contains a certain propor-
tion of certain chemical constituents, and unless not less than a given
proportion of apples has been used to produce a given volume of fluid.
Moreover it must not contain any artificial sweetening chemicals such
as saccharin ; it must not contain any chemical flavouring additions ;
the employment of certain antiseptics (as sulphurous acid) may be
permissible within defined limits, or entirely prohibited (as borates).
Artificial colouring agents are likewise prohibited.

Much of the modern systematic work on the analyses of cider is.
due to A. H. Allen ("Analyst," xxvii. 183). The following represent
the compositions of several varieties of apples examined by Allen : —

Water.

Free xVcid
(as malic).

Glucose.

Sucrose.

Ash.

Per

cent

Table apples

81-62

0-88

9-28

6-28

0-44

Cooking apples

84-74

0-56

8-75

2-29

0-33

Cider apples

80-29

0-097

9-43

2-95

0-54

84-14

0-36

7-21

2-84

0-44

350

FOOD AND DRUGS.

The following represent the composition of apple (and incident-
ally of pear) juice, as recorded by Truelle : —

Sp. gravity
Total solids
Acidity (as H2SO4)
Sucrose
Glucose
Tannin

Apple Juice.

Pear Juice.

Per cent

1-057 to 1-1110
14-94 „ 28-57

0-07 „ 0-74

0-56 „ 7-17
10-84 „ 18-18

0-026,, 0-81

Per cent

1-067 to 1-098
17-53 „ 25-32

0-08 „ 0-24

1-67 „ 6-14
10-81 „ 20-99

0-10 „ 0-32

Allen gives the following figures for apple juice prepared in Eng-
land. The analyses were made about thirty-six hours after the apples
were pressed, so that a little alcohol was found : —

•

Per cent

Sp. gravity

1-0550

1-0530

1-0470

Alcohol by weight

0-10

—

1-04

113

Solids

14-63

12-74

11-91

11-95

Glucose

13-51

10-48

9-13

8-82

Sucrose

1-34

0-69

0-66

0-38

Fixed acid, as. malic

0-28

0-42

0-45

0-50

Ash

0-35

0-30

0-22

0-26

Tannin

—

0-22

—

The following tables show the average composition of the apple
and its juice, cider, and cider vinegar, as recorded by C. A. Brown
(" Jour. Amer. Chem. Soc." 1901, xxiii. 809) :—

Unripe apples
Summer apples
Winter apples

Water.

Reducing
Sugars.

Sucrose.

Starch.

Ash.

Fixed Acid
(as Malic).

Per cent
80-67
85-00
82-16

Per cent
6-43
7-10
8-16

Per cent
2-84
3-36
4-16

Per cent
3-92
1-04

Per cent
0-27
0-28
0-26

Per cent
1-14
0-68
0-59

Summer apple juice

Winter apple juice

Cider

Cider vinegar

Sp.gr.

Per cent
1-0502
1-0569
1-006
1-0184

Solids.

Per cent

12-29

13-96

2-34

2-00

Reducing
Sugars.

Per cent
6-76
8-57
0-32
0-52

Sucrose.

Per cent
3-23
3-40

Acid as
Malic.

Per cent
0-72
0-43
0-25
0-14

Ash.

Per cent
0-12
012
0-04
0-01

CIDER.

351

Brown also gives the following analyses of five typical samples of
cider :-^

Sp.gr.

SoUds.

Reducing
Sugars.

Malic Acid.

Acetic Acid.

Alcohol.

Ash.

Per cent

0-9981

1-94

0-19

0-21

0-24

6-85

0-25

1-0012

2-71

0-19

0-24

0-42

513

0-32

1-0052

3-26

0-89

0-30

0-48

4-67

0-29

1-0007

1-93

0-34

0-27

0-21

4-95

0-23

1-0051

2-71

0-24

0-29

1-96

4-26

0-36

Allen publishes {loc. cit.) numerous analyses of English ciders of
which the following are the average figures : —

Norfolk Bottled.

Devonshire Bottled.

Draught Ciders.

Per cent

Sp. gravity

1-002 to 1-012

1-003 to 1-032

1-006 to 1-028

Alcohol by weight .

5-3 „ 7-69

2-57 „ 5-39

2-49 „ 5-86

o /Extractive

2-07 „ 5-47

2-12 „ 7-93

2-59 „ 7-63

1 « 1 Glucose .

§ o Fixed acid as malic .

0-77 „ 4-55

0-94 „ 7-24

0 „ 4-17

0-31 „ 0-42

0-12 „ 0-35

r§ 7| Volatile acid as acetic

0-07 „ 0-21

0-19 „ 0-37

0-2 „ 0-43

^ I .Ash ....

0-26 „ 0-33

0-23 „ 0-36

0-16 „ 0-23

There is no evidence as to whether these ciders were genuine or
not. Roques {" Le Cidre," p. 128) gives the following as the results
of numerous analyses of French ciders : —

Maximum.

Minimum.

Mean.

Per cent

Sp. gravity

1-041

1-0012

1-0159

Alcohol per cent ....

6-5

3-5

5-2

Sugar free extract per cent .

6-46

2-262

3-39

Sugar per cent ....

6-08

traces

2-162

Ash per cent . . " .

0-432

0-248

0-326

Alkalinity of ash (K2CO3) per cent

0-368

0-204

0-256

Kulisch (" Land. Jahrb.'
for German ciders : —

19, 83) gives the following as the figures

352

FOOD AND DKUGS.

Specific gravity

AJcohoi per cent

Total extract .

Sugar ....

Ash

Minimiim.

Maximum.

Per cent
0-9977
5-4
1-923
0-1
0-225

Per cent
1-050
7-3
3-023
0-3
0-336

Barker and Kussell (" Analyst," xxxiv. 125) find the amount of
P2O5 present to vary between 0'013 per cent to 0*023 per cent.

Grignon gives the following analyses of French ciders (" Le
Cidre," Paris, 1887) :—

Alcohol (by volume)
Extract .

Ash . . .
Acidity as H2SO4 .
Sugar .

X'^l'

Sweet.

Dry.

Per cent.

Per
cent

Per

cent

3-8

4-1

5-4

6-41

6-40

3-03

2-95

0-29

0-28

0-27

0-26

0-36

0-39

0-52

0-58

3-47

3-75

0-65

0-58

Old Ciders, annually treated
with fresh must.

Per cent

2-22

0-24

0-54

0-27

The Paris Municipal Laboratory authorities hold that pure cider
should contain a minimum of 3 per cent of alcohol by volume, 1*8
per cent of extract, and 0'17 per cent of ash, but these low limits are
rarely found in practice. Other French authorities are content with
0'9 per cent of extract and 0*12 per cent of mineral matter.

By a decree of 20 July, 1908, cider and perry are legally defined
in France as follows : —

" No drink is to be sold (1) under the name of cider unless it is
derived exclusively from the fermentation of the juice of fresh apples,
or a mixture of fresh apples and pears extracted with or without the
addition of water or (2) under the name of perry unless it is derived
exclusively from fresh pears with or without the addition of water.
The term cidre pur jus or poire pur jus is reserved for cider
or perry obtained without the addition of water. The term cider
or perry is reserved for cider or. perry containing at least 3*5 per
cent of alcohol '12 grm. per litre of extract (sugar being deducted),
and 1'2 grm. of mineral matter per litre. Cider or perry falling be-
low these limits is to be called petit cidre or petit poire.

The presence of a trace of boric acid in genuine cider (see below)
may be used as evidence of its purity, since most artificial ciders are
free from boric acid.

Artificial ciders are also usually free from tannin, and the pre-
sence of tannin is useful evidence of the authenticity of a given
sample.

CIDEE. ^^BT 353

According to Barker and Eussell (" Analyst," xxxiv. 125), if
genuine cider be shaken with an equal volume of ethyl acetate for five
minutes, and the ethyl acetate separated and poured on to lime water,
a band of yellow colour — not persisting long — is developed at the
junction of the liquids if pure apple juice be present.

The ash of genuine cider has the following composition : —

Per cent

Silica . 0-94

Phosphoric acid . . . 12-68

Lime 2-77

Maj^nesia ........... 0-94

Oxide of iron and manganese 0*94

Potash . . . . , 53-74

Soda 1-10

Carbonic acid 25-78

Pure cider is always laevorotatory. In the presence of added
cane-sugar, the rotation will frequently be to the right. If the cider
be dextrorotatory, and after inversion is still dextrorotatory, it is
certain that commercial glucose is present.

Allen gives the following details for the determination of boric
acid in cider.

The detection of boric acid in cider and fruits can be readily effected
by evaporating 20 c.c. of cider or apple-juice to dryness and igniting
the residue, or by directly igniting 25 grms. of apple or other fruit.
The ash is rendered distinctly acid to litmus with dilute hydrochloric
acid, a piece ot turmeric paper partially immersed in the liquid,
and the whole evaporated to dryness on the water bath in a flat
porcelain dish. The residue is further dried in the water oven for a
short time. In the presence of boric acid the turmeric paper will
acquire a brownish-red colour ,which, on being moistened with a drop
of caustic soda, is changed into a variety of colours, chiefly green and
purple.

The quantitative determi7iation of boric acid in cider and fruits is
very troublesome, and this has been the subject of numerous experi-
ments. The difficulty of the analysis is enhanced owing to the
minute quantity of boric acid present, and the determination is further
complicated by the presence of phosphates. These salts render in-
applicable the direct employment of E. T. Thomson's well-known pro-
cess (" J.S. C. I.," 1893, p. 433), in which the solution is first made
neutral to methyl-orange and then titrated with caustic soda and
phenol-phthalein in presence of glycerin, the end-point of the titration
corresponding to the formation of NaBO^. The unsuitability of Thom-
son's method without modification in the presence of phosphates is
due to the fact that while phosphates of the formula MH^PO^ are
neutral to methyl-orange, they are acid to phenol-phthalein. A
number of experiments were made with a view of overcoming the
difficulty caused by the presence of phosphates in quantity, but with-
out success. It does not seem possible to make an allowance for
the disturbing action of the phosphates, nor does the addition of
glycerin after the aqueous liquid has been rendered neutral to phenol-
VOL. I. 23

354 FOOD AND DRUGS.

phthalein overcome the difficulty, owing to the fact that boric acid
is distinct!)', but indefinitely, acid to phenol-phthalein, even in the
absence of glycerin.

After a large number of experiments, the following method for the
determination of boric acid in cider, etc., based on the moderate solu-
bility of calcium borate in water, was devised : About 100 c.c. of cider
or other liquid is evaporated to dryness with a few cubic centimetres
of a 10 per cent solution of calcium chloride ; or, in the case of fruits,
about 50 grms. weight is cut up into small pieces and the solution of
calcium chloride poured over the mass, which is then evaporated to
dryness. The dry residue is well charred, boiled with about 150 c.c.
of distilled water, and the liquid filtered. The carbonaceous residue
is thoroughly incinerated at a moderate temperature, and when cold
boiled with a further quantity of 150 c.c. of water, and allowed to
stand in the cold for some hours, or preferably overnight. The liquid
is then filtered cold, and the filtrate added to the first extract. i The
mixed aqueous extracts are next evaporated to a volume of 25 or 30
c.c, and after cooling neutralized' by decinormal acid, using methyl-
orange as indicator. "-^ An equal volume of glycerin is next added,
and the liquid titrated with phenol-phthalein and one-twentieth nor-
mal caustic soda solution (free from carbonate). About 10 c.c. more
glycerin should now be added, when, if the titration is complete,
the red coloration ' will remain. Each cubic centimetre of the one-
twentieth normal solution of caustic soda required represents 0'00175
grm. of boric anhydride, BgOg) 0*0031 grm. of crystallized boric
acid, H3BO3; or 0-004775 grm. of crystallized borax, Na^B^O^ -i-
lOH^O. The above process gives good results when the amount of
boric acid present in the sample taken is not less than 0*005 grm.

Allen has also examined the well-known method for the deter-
mination of boric acid based on the volatility of methyl borate, and
find the following to be the best method of operating : A suitable
quantity of the substance under examination is treated with calcium
chloride solution as already described, and well charred, and the main
portion of the salts extracted with about 50 c.c. of water. This
aqueous extract is transferred to a distillation-flask of about 100 c.c.
capacity, and cautiously evaporated nearly to dryness over a naked
flame. Meanwhile the charred residue is incinerated, the ash (nearly
-white) moistened with 2 c.c. of strong sulphuric acid, and the mix-
ture warmed. When the evolution of hydrochloric acid gas is nearly at
:an end, the acidified residue is transferred to the distilling-flask con-
taining the evaporated aqueous extracts. The last portions are
washed in with 10 c.c. of methyl alcohol,^ the flask immersed in a
boiling water bath, and the liquid distilled almost to dryness. A

1 It is desirable to extract the residue for a third time with hot water, allowing
the liquid when cold to stand for some time before filtration. This third extract when
titrated separately will generally be found to be free from boric acid. If not, the
amount found must be added to that already extracted.

- Care should be taken that all the borate is in solution before the titration is
begun.

^ Ordinary wood-spirit of good quality, purified by redistillation over caustic
potash, is suitable for this purpose. -

CIDEK.

355

further addition of 10 c.c. of methyl alcohol is then made, and the
distillation repeated. As many as six such treatments are usually
required. Between each distillation the residue in the flask should
be allowed to cool before the next addition of methyl alcohol is made.
The residue finally contained in the distilling-flask should be tested
by the flame-reaction with alcohol to ensure that the whole of the
boric acid has been volatilized. If this is not found to be the case,
the distillation should be repeated once or twice more.

The alcoholic vapours are passed into 25 c.c. of water contained
in a flask, the end of the condenser-tube dipping into the liquid.
When the process is completed, the distillate is evaporated over a
water bath until free from alcohol. By this treatment the methyl
borate is hydrolysed, and the boric acid left in a free state. The re-
sidual liquid is diluted with a little water and rendered exactly
neutral to methyl orange. An equal volume of glycerin is then
added, and the liquid titrated with one-twentieth normal caustic soda
and phenol-phthalein as already described.

The glycerin used in these processes should be rendered neutral
to phenol-phthalein just before use, as it is generally slightly acid in
reaction.

In many of the processes already in use for the separation of
boric acid by distillation, the methyl borate is distilled into a solution of
caustic soda, and after evaporation of the alcohol the aqueous liquid
is titrated in the usual way. In Allen's experience, however, when an
alkali was used, the results were always above the truth, even when
specially purified methyl alcohol was employed. For this reason the
use of caustic soda is not to be recommended, and, as previous experi-
ments have shown, is quite unnecessary.

The following results were obtained in a series of experiments
made to test the accuracy of the processes here described. A known
weight of crystallized borax was added either to a mixture of calcium
chloride, magnesium sulphate and sodium phosphate, or to a known
weight of apple. In the latter case an exactly similar portion and
weight of the same apple was treated with calcium chloride and the
boric acid determined, and deducted from that found in the other
portion to which borax had been added : —

No. of

Experiment.

(D)

(2)f

{3)\

(4)j

(5)

(6)

(7)

(8)

Substances Added to the
Borax.

Calcium chloride, mag-
uesium sulphate, and
sodium phosphate

50 grms. of apple

None

None
Sodium phosphate
Sodium phosphate

Borax

Taken.

Found.

Gm.

^ 0-200

0-198

U-200

0-204

\0-020
/0-020

0-019

0-020

0-200

0-197

0-020

0-022

0-200

0-201

0-020

0-023

356

FOOD AND DKUGS.

Kichmond and Harrison's method (" Analyst," xxvii. 179) for
the determination of boric acid in butter is rapid and accurate for its-
intended purpose, but the presence of phosphates in fruits and fruit-
products renders the process unsuitable for the determination of boric
acid in these substances.

A colorimetric method for the determination of boric acid in milk
and other foods has been devised by Cassal and Gerrans (" Brit.
Food Journal," October, 1902). The process is based upon the fact
that in the presence of oxalic acid the colouring matter of turmeric
forms with boric acid an intense magenta-red colour more delicate
than the ordinary turmeric reaction (that is, when obtained in the
absence of oxalic acid), and permanent for many hours. The alco-
holic solution of the colour formed in the reaction is compared with
that from a known weight of boric acid. The method is said to be
reliable and accurate, but appears to be rather lengthy and tedious.

The following table shows the proportion of boric acid contained
in various fruits and ciders, etc., examined : —

Fruits, etc.

Boric Acid, H3BO3.

(1) Apple (Norfolk)

0-009 per

cent.

(2) Apple (fox whelp)

0-013

(3) Apple (old fox whelp)

0-011

(4) Pear, no. 1

0-007

(5) Pear, no. 2

0-016

(6) Quince

0-016

'7) Pomegranate

0-005

(8) Grapes

0-004

(9) Norfolk cider

0-009 grm

. per 100 c.c.

(10) Hereford cider

0-017

(11) Devonshire cider

0-004

)) >i

(12) Apple juice (Devon)

0-004

» »

Perry, which is the fermented juice of the pear, differs from cider
principally in containing less malic acid, and therefore appearing ta
be more sweet.

Allen gives the following analysis of sparkling perry : —

Worcestershire.

Devonshire.

Gloucestershire.

Per cent

Specific gravity

1-020

1-021

1-070

Alcohol by weight

4-61

4-81

3-64

Solids

6-51

6-49

4-50

Volatile acid as acetic

0-41

0-35

0-22

Fixed acid as malic .

0-25

0-20

0-24

Glucose

2-71

3-60

0-36

Sucrose .

none

0-31

none

Ash .

0-40

0-28

0-30

CIDER.

357

Truelle gives the following figures for pear juice before fermenta-
tion in parts per 1000 : —

Specific gravity ....

Mean.

Maximum.

Miuimum.

1-0845

1-0980

1-0675

Invert sugar ....

145-64

200-

108-1

Sucrose

36-74

61-41

16-69

Total fermentable sugar as dextrose

184-14

220-

143-78

Tannin

1-78

3-2

1-01

Pectin

13-08

18-

Acidity as H0SO4 . . . .

1-47

2-40

0-76

The only practical adulterant of cider or perry is the dilution of
either the must or the fermented liquor with water. Sometimes pre-
servatives are added, which may be detected as in wine, and occasion-
ally citric or tartaric acid is added to modify the acid flavour.

The preservatives used are salicylic acid, boric acid and sulphites.
But as boric acid is a constituent of apples it is probable that in many
reported cases it was merely a natural constituent. A boric acid re-
action with turmeric can be obtained from 20 grms of pure cider.

The best method of deciding whether water has been added is to
calculate the amount of solids present in the original must, as
follows : —

Alcohol per cent by weight x 2-07 = original sugar fermented.
Acetic acid x 1*5 = ,, „ „

Extractive matter in sample.

The sum of these is equal to the original solids of the juice, and
rarely falls below 12 per cent. An excessive amount of solids indicates
added saccharine matter.

CHAPTER VI.

FLESH FOODS.

The inspection of fresh flesh foods does not come within the scope
of the present work. Chemical methods are rarely applicable to the
examination of fresh flesh from an analytical point of view, and such
flesh inspection comes rather within the purview of the meat inspector
and the veterinary surgeon than the analyst. The present section is
intended to deal more with preserved foods in the sense of flesh foods
preserved in tins (or glasses) and in the form of sausages. Further,
chemical methods for discriminating between various meats are en-
tirely lacking, and no attempt will be made in this chapter, to so
discriminate except so far as the detection of horse-flesh in preserved
meat is concerned. The subject will be dealt with from the following
points of view.

(1) The principles underlying the decomposition of flesh food.

(2) .The examination of preserved food from the point of view of
(a) decomposition products, (b) metallic contamination, (c) the pres-
ence of preservatives.

(3) Sausages.

(4) Meat extracts.

The Decomposition of Flesh.

It has long been known that the products of putrefaction of flesh
are ultimately of a very dangerous character. This is especially the
case in even the early stages of the decomposition of fish ; it is usually
only at a later stage that the decomposition products of ordinary meat
become dangerous ; and at a still later stage, those of the foods
embraced by the word " game " — which may be eaten safely when at a
decidedly later stage of decomposition than either ordinary meat or fish.
Early observers such as Barrows, Kerner, and Panum appear to have
recognized in the products of flesh decomposition certain nitrogenous
substances, which they believed to be in some way similar to the vege-
table alkaloids. But it was Zuelzer and Sonnenschein who first defin-
itely described a flesh-decomposition product as an alkaloid. Later
researches have indicated that the principal poisonous decomposition
products of flesh are those nitrogenous principles known as animal
alkaloids or ptomaines.

The leucomaines are closely allied to the ptomaines and are also
known by the name physiological alkaloids. These are formed by

(358)

369

the breaking down of the nitrogenous matter in the living cell and
are usually non-toxic. The classification, however, overlaps, as many
of these bodies are elaborated by the livmg cell, as well as formed by
the decomposition of dead flesh through the agency of bacteria.

The modern advances in our knowledge of this branch of a most
difficult subject are due to Selmi, Nencki, Gautier and Brieger. The
first ptomaine to be separated in a state of purity was that i:-olated
by Nencki, and later many were described in detail by Gautier
and Brieger. The following summary of the principal ptomaines
is due to Gautier and has been adopted by Mitchell (" Flesh Foods ").

Monamines of the Fatty Acid Series.

Trimethylamine (CH3)3N. Herring pickle.
Diethylamine (C.2H5)2NH. Putrid meat extract.
- Triethylamine (C2H5)3N. Decomposed cod-fish.
Propylamine (C3H_)NH2. Decomposing cod-liver.
Butylamine (C4Hg)lS[H._2. Decomposing cod-liver.
Amylamine (C5Hij)NH2. Cod-liver oil.

Diamines of the Fatty Acid Series.

Putrescine or Tetramethylene-diamine C4Hj2^2- Putrid horse-
flesh.

Cadaverine, or Pentamethylene-diamine C^H^^Ng. Putrid fish
and blood.

Neuridine CgH^^N.^. Putrid meat, albumin, gelatin.

Saprine C^Hj^Ng. Decomposed flesh.

Gitanidines.
Methylguanidine C2H7N3. Putrid horse-flesh and beef.

Aromatic Ptomaines, free from Oxygen.

Collidine CgH^^N. Putrid fish and putrid gelatin.
Parvoline C9Hj3N. Putrid horse-flesh after several months.
Corindine CjoHjgN. Putrid cuttle-fish.
Dihydrocollidine CgH^gN. Putrid fish and horse-flesh.

Oxygenated Ptomaines.

Neurine C5H13NO. Putrid meat on fifth or sixth day.
Choline C5HJ5NO2. Accompanies neurine.
Muscarine C5Hi_^N03. Putrid fish.
Betaine C^HjjN02. In mussels (leucomaine).
Homopiperidinic Acid C5HJJNO2. Decomposition of meat fibrin.
Mytilotoxine C,,i:lj5N02. In poisonous mussels (? leucomaine).
Mydatoxine CyHj3N02. Putrid horse-flesh after nine to fifteen
months.

Gadinene C-Hi-NO.. ) r> ^- -j ^ u • n j

Methylgadinene C, a,„NO, } ^""-"i ^'■'^' «^P««'^"y "o^.
Unnamed base of Brieger CyHj^N02. Accompanies mydatoxine

360 FOOD AND DRUGS.

Aromatic Oxyge^iated Bases.

Tyrosamines C^HgNO ; CgH^iNO ; CyHjgNO. Decomposing cod-
liver.

Mydine C^H^jNO. Decomposing human flesh.

It may be mentioned that Brieger considers that these poisonous
base? whicn Gautier claims to be incUided in the group of the leuco-
maines. are not in reality the products of the cell metabolism, but are
in fact absorbed into the cell from the intestines.

The question of ptomaines in regard to toxicological analysis will
not be discussed, but it may not be out of place to mention that there
are many ptomaines, closely rese-nbling well-known vegetable alkaloids
in their reactions, especially their mydriatic effects. Zuelzer and
Sonnenschein some years ago isolated a septic alkaloid which re-
sembled atropine and hyosciamine in a very remarkable manner.

A brief reference to the symptoms of ptomaine poisoning may now
be made.

Symptoms of Ptomaine Poisoning. — The usual symptoms of
ptomaine poisoning are as follows : A dilated then contracted pupil
of the eye, feeble respiration, weak pulse, temperature sub-normal,
skin moist, loss of the power of contracting the muscles, stupor,
convulsions, and death. The loss of muscular contractibility takes
place even when under the influence of electricity, and is one of the
determining features of poisoning by muscarine, a ptomaine found in
putrefying fish and in poisonous mushrooms.

The action of ptomaines on the body varies considerably ; some
have little effect whilst others are fatal in even small quantities. It
is not unlikely that the symptoms of flesh poisoning vary in nature
and extent according to the kind and amount of the bases present,
some of which probably modify in a greater or less degree the action
of the others.

The methylamines and ethylamines formed during the putrefaction
of flesh are the only monamines not very poisonous ; large quantities
of butylamine produces convulsions and muscular paralysis ; and
amylamine which is extremely poisonous causes the pupils of the eye
to dilate and, finally, convulsions. The diamines (putrescine, cada-
verine, neuridine and saprine) have very little or no effect on the
body, and are only considered slightly poisonous. Cadaverine may
produce inflammation of the mucous membrane.

Methylguanidine, which may be considered the representative
guanidine ptomaine, is extremely poisonous. When it is injected into
a small animal it causes dilation of the pupils, convulsions, and
death within twenty minutes.

Of the aromatic non-oxygenated ptomaines, collidine, parvoline,
corindine and dihydrocollidine are all exceedingly poisonous. Corin-
dine, like curare, produces paralysis. Dihydrocollidine produces
torpor, muscular paralysis, and convulsions.

Of the better- known oxygenated ptomaines neurine causes an ex-
cessive flow of saliva, contraction of the pupils, sudden convulsions,
and death.

FLESH FOODS. 361

Choline acts physiologically much in the same way as neurine, but
not so violently.

Muscarine is very poisonous, and small doses will produce saliva-
tion, contraction of the pupils, diarrhoea, convulsions, and death.
Atropine is used as an antidote, as its action is opposite to that of the
three foregoing ptomaines. Betaine is non-poisonous. Mydatoxine
is somewhat poisonous. Large doses cause diarrhoea, redness of the
eyes, convulsions, and death. Gadinene is not very poisonous, though
methylgadinene in large doses produces symptoms of paralysis. An
unnamed base of Brieger (C7HJ-.NO2), found with mydatoxine in
putrid horse-flesh, resembles curare in its poisonous properties.

Botulism or Sausage Poisoning. — Cases of botulism, like the attacks
of trichinosis, have been most prevalent in those parts of Germany,
Saxony, for example, where raw ham and raw sausage are largely con-
sumed. There have been wholesale c ises of poisoning as at Chemnitz
in 1879, when 241 persons were poisoned by Mettwurst, and 160
met the same fate seven years later. Ostertag mentions similar but
smaller outbreaks since 1886, as for example in Dresden (11), in
Gerbstadt (over 50), and in Gera (30).

The distinguishing symptoms of pure botulism can be detected after
a period of incubation of from eighteen to forty-eight hours. They are
an uneasy and heavy feeling in the stomach accompanied by vomiting
and sometimes diarrhoea, faintness, blurred vision, flaccidity of the
muscles and collapse. If the case is fatal, death ensues in from four
to eight days. If the toxine of B. bokdinus is the sole cause of the
illness, neither fever nor mental disturbances occur as symptoms.
According to Senkpiehl out of 412 cases recorded between 1789 and
1886, 165 proved fatal, thus the mortality is very high.

Eber considered both sausage poisons and ptomaines as toxigenic
substances, and not as toxines. He grouped together under the term
" toxigenes " those chemical products which, when injected into an
animal, are not poisonous until modified by the vital activity of the
cells. He compared them with sodium iodide and similar inoruanic
substances which, when injected into an animal, produce no ill effects
for at least six or eight hours. For years the origin of the poison
could not be determined, though it was well known to be distinct from
that produced by ordinary putrefaction. Hilger was the first to ob-
tain from the intestines of six persons who had died from sausage
poisoning, a semi-fluid substance which closely resembled curare.
Tamba also found a similar substance in liver sausage which had been
exposed to the air. Haupt concluded that the decomposition products
formed by B. proteus mirahilis caused the disease. Ostertag, however,
showed that the symptoms of botulism were not identical with those
produced by the inoculation of cultivations of that micro-organism.
In 1895 van Ermengem extracted an anaerobic bacillus from the body
of a person who had died from sausage poisoning, and the cultivations
of this produced the same symptoms. •

Brieger and Kempner have recently extracted a toxine from a pure
cultivation of B. botiUinus, which they consider closely related to the
toxines of diphtheria and tetanus in chemical composition.

362 FOOD AND DEUGS.

The isolation of ptomaines in preserved meat would, in general,
be very powerful evidence of the decomposition of the food which
would probably have caused more or less severe illness.

The most important of the earlier methods devised for the separa-
tion of the ptomaines is that used by Brieger, who is perhaps the
greatest authority on this subject. He uses the salts of the heavy
metals, and picric acid. For the purpose of separating an alkaloid
from a putrefying mass, this mass is first boiled with acidified water,
and then filtered ; the filtrate is treated with subacetate of lead ; from
this, excess of lead is precipitated by sulphuretted hydrogen which is
passed through the filtrate, and the fluid is again filtered to keep back
the lead sulphide. This second filtrate is evaporated to about one-
third of its original bulk, and is washed with amyl alcohol to remove
fat, etc., and again reduced in bulk by evaporation, and sulphuric acid
and ether added; the ether is removed, after which the remaining
liquid is concentrated by careful evaporation to one-fourth of its
bulk ; the evaporation drives off most of the volatile fatty acids present,
after which the fluid, neutralized by the addition of baryta, is again
filtered, carbonic acid gas is passed through it, by which barium car-
bonate is thrown down, which is separated by filtration. After careful
heating over a water bath, the fluid is cooled, and bichloride of mercury
is added, when a somewhat dense precipitate is formed. This pre-
cipitate is carefully washed and decomposed by sulphuretted hydrogen,
when sulphide of mercury is thrown down ; the fluid is again filtered
and the filtrate is evaporated to obtain as great concentration as pos-
sible. From the liquid so obtained all inorganic substances crystallize
out first; these are removed, and then in the fluid that remains "or-
ganic " acicular crystals are thrown down. These may be dissolved
in water, but they are insoluble in absolute alcohol, ether, benzine, or
chloroform. It is found that these substances, the ptomaines, may be
precipitated by the salts (especially the chlorides) of the heavy metals.
These precipitates or crystals differ, however, very considerably as to
their solubility; hydrochloride of putrescine obtained by the above
method separates out in acicular crystals, and on the addition of
chloride of gold gives very insoluble crystals of an octahedral form,
whilst on the addition of chloride of platinum, octahedral crystals,,
which are much more soluble, are also formed. Phospho-molybdic
and phospho-tungstic acid added to this substance give respectively
a yellow and a white crystalline precipitate. Iodide of mercury dis-
solved in iodide of potassium also gives rise to the formation of
prisms ; with ferrocyanide of potassium there is a yellowish amorphous
precipitate; with picric acid a yellow precipitate composed of delicate
needle-shaped crystals ; and with an aqueous solution of bichloride of
mercury an exceedingly insoluble acicular crystalline precipitate is
thrown down. This substance and the reactions obtained with it
may be taken as typical of the whole group, although there are certain
diff^ences ; for instance, cadaverine treated with chloride of gold gives
a very soluble substance, whilst with chloride of platinum there are
thrown down well-formed very insoluble crystals. Mydaleine is ex-
ceedingly soluble in most of its combinations, and it is at present.

FLESH FOODS. 363

almost impossible to separate it from the mother liquid ; in fact, its-
salts have not yet been separated, and in consequence it has been
found impossible to determine its exact chemical nature. These,
along with saprine, were obtained by Brieger from flesh that was
being decomposed by the action of putrefactive micro-organisms.

Brieger 's later method consists in extracting the finely divided sub-
stances with very dilute hydrochloric acid, evaporatiog the extract on
a water bath, filtering and finally concentrating to the consistency of
a syrup. This is dissolved in 90 per cent alcohol and the Uquid
filtered and excess of an alcohoUc solution of HgCl^ added. The
whole is allowed to stand for twenty-four hours when the precipitate
is collected and washed with water, then suspended in water and de-
composed by a current of H.^S. The precipitated sulphide of mercury
is removed by filtration and the ptomaines are now present as hydro-
chlorides in the filtrate, after which they are examined as in the
former method.

The Stas-Gautier method is as follows : —

To the finely divided substance add water containing 0*5 per cent of
tartaric acid and allow to digest for twenty-four hours. Filter the liquid
and separate the last portions by pressure. If the substance submitted
for examination is liquid or almost liquid, sHghtly acidify with tartaric
acid; if it is oily, shake in a flask containing carbon dioxide, with an
aqueous 0*25 per cent solution of oxalic acid.

Heat the slightly acid extract, or acidified original liquid, for a.
moment at 100° C. to coagulate albuminous substances, then cool and
filter. Evaporate the filtrate in a vacuum at 40° C. to a syrup, collect-
ing the distillate. This is extract A.

The distillate generally contains substances carried over with the
water, as phenols, indol, volatile fatty acids, ammonia, substituted
ammonias, etc., with traces of volatile ptomaines. To recover the
latter, acidify with a slight excess of sulphuric acid, and, to free
the bases and get rid of the larger proportion of the ammonia they
contain, treat the dried sulphates with lime. Shake the mixture of
calcium sulphate and free bases with ether, then with alcohol. Any
calcium oxide which dissolves can be precipitated with a very small
quantity of sulphuric acid, leaving the bases in solution.

To remove fatty substances, lactic acids, excess of acid added, etc.,.
from extract A, extract with ether then add boiling alcohol, which
gives solution B and leaves residue C.

Take up with water and dialyse residue C, which contains salts,,
extractives, xaathic bodies, acid amides, etc. Concentrate the part
passing through, by evaporation, precipitate the bases present with
lead acetate remove the lead bv adding hydrogen sulphide, concentrate
the filtrate and add alcohol. The substances gradually accumulating
consist of oxygenated bases, such as leucine.

Evaporate solution B, which contains the most important bases,
with peptones, etc., to a syrup. Make alkaline with potassium bi-
carbonate, mix with powdered glass, and extract first with ether, then
chloroform, and lastly amyl alcohol.

The first two extracts after being evaporated leave a residue of any

364 FOOD AND DKUGS.

alkaloidal substances extracted. Shake the amyl alcohol with water
to which has been added a Httle sulphuric acid, thus extracting the
bases in solution. Boil the liquid and add a hot solution of barium
hydroxide so long as a precipitate forms. Separate the bases, which
are left in solution, into fixed and volatile bases, by distillation, and
allow the distillate to pass into acidulated water.
Dragendorff has devised the following process : —
Dragendorff's Method. — Mix the finely divided substance with
water acidified with a little sulphuric acid, digest for several hours at
50" C, then wash with water. Evaporate the liquid to a syrup,
then digest for twenty- four hours with three or four times its volume
of 95 per cent alcohol. Filter off the separated substances, evaporate
the alcohol from the filtrate, shake the aqueous residue with benzene
to remove certain impurities. Make the residue alkaline with
ammonia and again extract with benzene, which this time removes
some free bases. Acidify the liquid and extract with chloroform,
again make alkaline with ammonia or sodium carbonate, and again
extract with the same solvent. Extractions are made in a similar
manner with amyl alcohol, first from acid and then from alkaline
solution. Finally recover and examine the bases from each of the
various extracts.

The following summary of the properties of the principal of the
ptomaines is based on the work of Brieger and Gautier, and is due to
C. A. Mitchell :—

(A.) AMINES OF THE FATTY SERIES.

(1) MONAMINES.

Trimethylmnine (CH3)N. Met with in herring pickle, ergot of
rye, and in putrefaction products of meat, cheese, etc. Is a gas with
fish-like odour. Boils at 9'3°, solidifies at about - 75°. Very soluble
in water, forms well-marked salts ; the aurochloride forms yellow
monoclinic prisms and the platinochloride orange prisms.

Ethylamine {C.2^r^^IL.^. Met with in putrefying flour. Strongly
alkaline liquid, boiling at 18"7° with an ammoniacal odour.

Diethylamine {G2'H.^).2NB.. In decomposing fish, meat extracts, or
sausages. A volatile, inflammable liquid, boiling at 57'5°. Very
soluble in water. It can be separated from ethylamine by treating the
mixed mercuro-chlorides with acetic acid, in which the diethylamine
salt is insoluble.

Triethylamine {0.2^.^.^. Accompanies the two last described
and other bases in decomposing fish or peptones. Strongly alkaline,
inflammable, liquid boiling at 89". Is slightly soluble in water, and is
precipitated from its solution by salts of mercury, copper, lead, or iron.
The aurochloride soon darkens by reduction to amino chloride,

Projjylamine (C3H-)NH^. — Is found in decomposing gelatine and
cod-liver. Is an alkaline liquid, boiling at 78° to 82°, soluble in water.
Its platinochloride forms monoclinic prisms.

Iso2)ropylamme{C.^B.j)'^B..2. — Is an ammoniacal liquid, boiling at

FLESH FOODS.

365-

32°. It is soluble in water. Its platinochloride forms orange plates.

Butylamine (C4H,,)NH2. In decomposing cod-liver. Is an alkaline
liquid boiling at 76". Its solution reduces copper and silver salts on
warmin.^. Its platinochloride forms yellow plates fairly soluble in water.

Isoamylajuine (C,-H^j)NH.^. — In decomposing cod-liver. A colour-
less liquid of disagreeable odour, of specific gravity 0*797.

Hexylamina (C,jHj3)NH.^. — In decomposing cod-liver and putrefy-
ing yeast. A liquid boiling at 129°. It forms a hydrochloride in
crystalline lamellae, and a platinochloride in orange scales.

(2) Diamines.

Ethylidene diamine {G.^^i^H.^.^. — Properties doubtful. To be
found in putrefying fish. Is probably identical with ethylene diamine.
An alkaline liquid boiling at 116°.

Tetramethylene Diamine (CH^. CH.^NH^)^ or Pz^^rescme. Is found
in the putrefactive products of flesh. Neuridine appears to be formed
first (q.v.), to be replaced by cadaverine and putrescine. Putrescine
is a clear mobile liquid with a strong characteristic odour. It rapidly
absorbs CO^ from the air, forming a crystalline carbonate, boils at
158° when quite pure, and melts at 24°. It forms a crystalline
hydrochloride in long transparent needles, crystallizable from hot
dilute alcohol. Brieger gives the following summary of the reactions
of the free base : —

Phosphotungstic acid = white ppt., soluble in excess.

Phosphomolybdic acid = yellow ppt.

Potassio -mercuric iodide =oily ppt., afterwards becoming crystalline.

Potasslo-bismuth iodide = „ „ ,,

Potassio-cadmium iodide — ,, ,, „

Picric acid = yellow needles.

Tannic acid = dirty white ppt.

The following method of separation from neuridine and cadaverine
may be adopted. The solution is precipitated with platinum chloride,
and the separating platinochlorides are treated with excess of cold
water. The putrescine salt is very insoluble, and on filtration is left
with some cadaverine salt. On heating the precipitate diluted with
more water, the putrescine and cadaverine salts dissolve, and the
putrescine salt separates out first on cooling.

Pentamethylamine Diamine (CH.,)^(NH2)2. Cadaverine. It ap-
pears after about the third day of putrefaction of flesh. Is often
associated with neuridine and putrescine. It is a viscid liquid, boiling
at 178° and rapidly absorbing 00^ from the air. It has a penetrating
odour. It yields the following reactions when in solution in water : —

Phosphotungstic acid
Phosphomolybdic acid •
Potassio-mercury iodide
Potassio-cadmium iodide

Iodine in potassium iodide
Potassio-bismuth iodide -

Picric acid
Tannic acid

Potassium ferrieyanide and ferric
chloride =

= white ppt., soluble in excess.

resinous ppt.

brown ppt.

gradually becoming
granular.

yellow needles,
white amorphous ppt.

blue coloration.

366 FOOD AND DEUGS.

It can be separated from putrescine and neuridine by precipitating
the hydrochlorides with a solution of platinum chloride. By fractional
crystallization the platinochlorides of cadaverine and putrescine
separate first, the more soluble salt of neuridine being left in the
mother liquor. The crystals separating are suspended in water and
decomposed by a current of H^S. By filtering off the platinum
sulphide, the hydrochlorides are left, which are obtained by evaporat-
ing the solution, and by treatment of the residues with 96 per cent
alcohol at a temperature of 60° to 70° ; the cadaverine hydrochloride
is dissolved and the putrescine salt left insoluble.

Neuridine C^Hj^Ng. Is formed in the putrefactive decomposition
of meat, fish, albumin, or gelatine, and reaches its maximum on the
eleventh or twelfth day. Brieger separates it in the following manner.
The finely divided mass is extracted with hot water slightly acidified
with HCl, the extract filtered, the filtrate concentrated to a syrupy
liquid on the water-bath, and this repeatedly extracted with alcohol.
The alcoholic filtrate is treated with mercuric chloride solution, the
precipitate collected and washed and then decomposed in suspension
by a current of H.^S. The liquid is filtered, and concentrated on the
water bath, and on cooling long needle-shaped crystals of neuridine
hydrochloride separate out, which can be purified by recrystallization
from hot dilute alcohol. Free neuridine is insoluble m alcohol or ether,
but is soluble in water. It gives the following reactions : —

With phosphotungstic acid = white amorphous ppt., soluble in excess.
,, phosphomolybdic acid = „ crystalline ppt.
„ picric acid = ppt. appears slowly, and becomes yellow

needles.
,, potassio-bismuth iodide = red amorphorus ppt.
„ gold chloride = crystalline ppt.

Saprine CgH^^N.^. Occurs in decomposing flesh. Not very
poisonous.

(3) GuANiDiNE Derivatives.

Methyl-guanidine, NH : C(NH^) (NH.CHg). Occurs in decom-
posed flesii products. The free base is crystalline and deliquescent. It
forms a crystalline hydrochloride, insoluble in alcohol. The platino-
chloride forms rhombic crystals, readily soluble in ether. Is very
poisonous.

(4) Aromatic Amines.

Pyridine CgHj^N. Found in the decomposition products of pro-
teids. Liquid of penetrating odour, miscible with water. Boils at
114°.

Collidine CgHjiN. Found in putrid fish. A yellow liquid of acrid
odour. Slightly soluble in water. Specific gravity 0*986 ; boiling
point 168". Is very poisonous.

Parvoiine CcjHjgN. From putrid horse-flesh. Amber-coloured
oil, boiling above 200°. Slightly soluble in water.

FLESH FOODS. ^^^ 867

Corindine CioHjr.N, From putrid fish. A yellow viscous liquid
boiling at about 230°. Is poisonous.

DihydrocoUidifie CgH^gN. In putrid meat and fish. Boils at
about 208°. Specific gravity at 0° = 1-0296. Forms a crystalline
hydrochloride. Is very poisonous.

(5) Oxygenated Bases.

Neurine C^H^aNO. In putrid flesh, appearing about the fifth or
sixth day. Is a syrupy liquid, of strong alkaline reaction. Soluble in
water. Forms a crystalline hydrochloride. . Is very poisonous. It
gives the following reactions : —

With phosphomolybdic acid = white crystalline ppt., soluble in excess.
,, phosphotungstic acid = nil.

,, potassio-mercury iodide = voluminous yellowish-white ppt.
„ potassio-bismuth iodide = amorphous red ppt.

Muscarine C^Hj^NOg. In putrid fish. Is one of the most
poisonous ptomaines known. It is separated by Brieger as follows : —

The alcoholic extract of the putrid mass is treated with mercuric
chloride to separate choline and neurine. The filtrate is treated with
H.^S to remove mercury, and the filtrate concentrated after being
neutralized with sodium hydroxide. The syrupy liquid is taken up
in alcohol and excess of platinum chloride added. The platinochloride
of neuridine crystallizes out first and is filtered off. The filtrate is
concentrated, and a fresh crop of crystals (ethylidenediamine platino-
chloride?) is filtered off. On further concentration, the platino-
chloride of muscarine separates, and this, treated with H^S, yields the
hydrochloride, which is converted into the sulphate by treatment
with silver sulphate, and this into the free base by treatment with
barium hydroxide. Muscarine forms colourless, deliquescent
crystals, which are alkaline, and rapidly absorbs COg from the air.

Mytilotoxine C^-H^^NO^. From poisonous mussels. Its auro-
chloride melts at 182°.

Gadinene C^B.-^^1^0^. From putrid fish. Its platinochloride melts
at 214°.

Mydaleine. — Found in decomposing human flesh. Composition
uncertain. Is poisonous.

Mydine CgHj^NO. In putrid flesh. Picrate melts at 195°. Is
not very poisonous.

Tyrotoxine C^-H^N^OH. From decomposing cheese, and in ice
cream. Fine needles melting with decomposition at 90°, in the pres-
ence of moisture. Is poisonous.

Apart from the exceedingly difficult question of the isolation of
ptomaines, there is another problem connected with the question of
preserved meats which, presents difficulties which are generally in-
superable. This is the question of deciding what meats are actually
present in a given sample. It is comparatively rare that this ex-
amination is necessary, since high-grade makers keep their prepara-
tions true to description, using the necessary palatable admixtures only.
But in low-grade preparations, the main constitutent is frequently

368 FOOD AND DKUGS.

not true to name, but as no legal standards for this type of prepara-
tion exist, it is very difficult to bring cases dealing with the composi-
tion of preserved meats, etc., into court, except on the grounds of
containing preservatives or being unfit for human food.

Warden and Bose have published (" Chem. News," 1890, Lxi. 304)
some particularly complete analyses of typical samples of canned
beef and mutton. They estimated the moisture to vary from 49 per
cent to 57 per cent, the fat from 10 to 22 ; the porteids (i.e. N x 6-25)
from 24-5 to 29 ; the ash from 0-62 to 4-36 ; the chlorine from O'll
to 2.65 ; the phosphoric acid from 0'31 to 0*40; the hot water extract
from 5-35 per cent to 10-1*4 per cent, with a content of nitrogen varying
from 0-88 per cent to 1*10 per cent.

The following methods of analysis are satisfactory : —

Thoroughly pulp the entire contents of a can in a large mortar,
taking care to scrape out any fat and jelly that may be left in the
can. Warden considers it a mistake to regard a slice of the contents
as a fair sample.

To determine the moisture, place from 5 grms. to 6 grms. of the
sample with forceps in a fiat platinum dish, and dry first at 100°,
then at 120°. Moisten the samples with alcohol then dry again.
The whole time of heating takes from eight to nine hours. In another
large platinum dish heat from 30 grms. to 40 grms. of the pulp in the
manner just described ; reduce to a fine powder, and again heat. This
dried pulp, preserved in a closely stoppered bottle, can be used for the
determining fat, nitrogen, and aqueous extract.

To determine the ash, char that portion of the pulp used for as-
certaining the moisture, at a temperature below redness, crush with a
glass rod, exhaust with boiling water and again ignite. Treat the
residue with boiling water again and ignite and weigh the insoluble ash.
Evaporate the aqueous extract to dryness, heat the residue almost to
redness, and weigh the soluble ash. The total ash is estimated as the
sum of the soluble and insoluble ashes determined as just described,
and it will be found that the figures thus obtained coincide with
determinations of the total ash by direct ignition, avoiding at the
same time the difficulty experienced in the latter case of brmging
about complete combustion of the carbon without losing any ot the
alkali-metal salts by volatilization.

The soluble ash is used for determining potassium and sodium by
dissolving it in water, then adding to the warm solution barium chloride
ferric chloride, and ammonia successively. The last reagent is used
in such quantity as to make the liquid just alkaline. Filter off" the
precipitate which consists of BaS04, FeP04, and FeHgOg ; treat the
filtrate with ammonium carbonate and ammonium oxalate and warm
on the water bath for some time. Kemove the precipitate, consisting
of BaCOa and CaC.,04 by filtration ; evaporate the filtrate in platinum
to dryness and gently ignite the residue ; re-dissolve the residue in
water, filter the solution from a little barium carbonate, add a drop of
hydrochloric acid to the filtered liquid and evaporate with platinic
chloride to separate the potassium and sodium.

Warden and Bose, to determine the chlorine and phosphoric acid,

FLESH FOODS. 369

employ the following method : Mix 20 grms. of the freshly pulped
meat with about two grms. of pure sodium carbonate dissolved in
sufficient water to cover the pulp. Evaporate the resulting magma
to dryness, carbonize, extract first with water, then with nitric acid,
again ignite the residue, dissolve in nitric acid, and determine the
chlorine and phosphates in the mixed solutions by the usual methods.

The total nitrogen in the dried pulp is determined by Kjeldahl's
process and multiplied by the factor 6" 25 to find the proteids.

The extractive matter is determined by boiling 1 grm. of the dry
pulp with distilled water in a 100 c.c. flask and when cold diluting to
100 c.c. Pass the liquid through a dry filter and evaporate to dryness
an aliquot portion of the very faintly opalescent filtrate in a platinum
dish ; weigh the residue. The greater part of the filtrate is used for
determining extractive nitrogen by Kjeldahl's method.

To determine fat, take 0*5 grm. of the dried pulp in a small well-
stoppered weighing bottle, add a measured volume of fight petroleum
ether from a burette. Allow the mixture to stand for two days,
occasionally shaking it, then draw off by a small pipette a portion of
the perfectly clear liquid floating on the top, carefully measure a small
volume of it and pass into a small beaker. Distil off the petroleum
ether, dry the residual fat at 100° and weigh. From this, calculate the
fat in the total amount of petroleum ether used. This method of Dragen-
dorff's was found by Warden and Bose to have similar results to those
produced by exhausting the substance with a solvent of fat in the
customary manner. Warden and Bose have compared their analyses
of canned meats examined by the foregoing methods with the figures
arrived a: by Konig in the analysis of fresh beef and mutton. They
find that whereas the percentage of moisture in canned meat is less
than in fresh meat, the fat in canned meat usually excaeds that of
fresh meat. They obtain the following amounts of albuminous matters
in the anhydrous and fat-free samples examined by multiplying the
total nitrogen by 6-25.

Average of canned beef samples
Average of canned mutton samples
Average of all fresh cow and ox flesh
Averags of all fresh mutton
Average of all canned meat samples
Average of all fresh meat samples

Albuminous Matters in
Anhydrous Fat-free Meat,

87-06 per cent

87-19 „ „

93-94 „ „

93-81 „ „

. 87-12 „ „

93-87 „ „

Konig's analyses of seven specimens of canned meats showed
them to have the following average composition : —

Proteids, etc., 28-97 ; fat," 12-63 ; ash, 3-71 ; and water, 54-69 per cent.
These figures correspond to 10-33 per cent of nitrogen and 27-27 per
cent of fat in the anhydrous samples, and to 88-63 per cent of al-
buminous matters in the anhydrous and fat-free samples.

The following table of analyses' of preserved foods is taken from
Bulletin 13, part 10, of the United States Department of Agriculture
Bureau of Chemistry : —

VOL. I. 24

370

FOOD AND DEUGS.

Sodiam
Chloride.

0-08
0-24

trace

en ^ -^

lO «5 o

6 c<i 6

t- 00 ?c>
CO «5 O
CO -^ (?i)

9-67

11-33

7-15

001

0-09

trace

6A6

00 iH »0
<M U5 «0

.H «6

OS 00 c*

«p CO «5
"* t?- CO

rH O 00
»C CO lO

o« t>- 6

rH f^

rH t» CO
rH rH O

OS »« (M

CQ cp 9

tH »H fH

00 rH eq
kO c^ «o
-H cb6

rH t- (M

rH Ht «P
fH fH 6

CO (M OS

CO 9 ip

Ol '^ fH

O rH !>
ti 00 -«><
rH C06

Gelatinoids

and Proteins

Ppt'd by

Bromine.

iH O t-
Ofl »(5 GO

.H At 6

OS »o o
»p o« ip

CO t- fH

<M rH lO

O CO Cfl
(Tl M rH

IM rH 00

05 CO CO

6 (fi 6

fH (fl 6

ill
III

»0 U3 "*

iH fH iH

6 1> «

iH C<l iH

C* CO rH

CO o 00

H< do 00

r-{ rH

i3> Tj* Oi

00 OS t-
tH rH fH

»0 -^ OS

qs^cp
»o -«*t o

(N CO «

CO «0 Wi

<x> «5 (?q
6 6 CO

Cq CO (M

??9
<N 6 >o

CO -^ (M

Tl< >p CO
6 « CO
rH W rH

£1

CO CO CO
-^ lO Cfi

6 66

05 CO OS

"* o o

6 -H 6

66 6

1/5 OS OS
00 (M W5
6 rH 6

»0 IM »0

U5 (M rH

6 rH6

OS ""l* ■*
iH (M iH

66 6

t- so "*

»0 rH C<I

6 fH 6

66 6

W) c^ <o

rH CO O

666

6 66

ill

rH C<) <M
-* »p CO

<N <?q o»

9??

CO ■* tH

»p CO o

CO -^ CO

(N rH O

cq OS CO

■^ "^ CO

t^ C~ CO

CO OS ^
CN eq rH

i '^»

9??
CO CO cq

MS iH rH
tH »C CO
Tt< O CO

«0 OS <M
«Qpt>
■^ "* CO

rH OS O

(M "^t* O

»b 6 "^

rH O 00

'r'9'r'

eoeo «

O CO OS
«0 CO CO
<N «5 6
i-l iH

OS 00 OS
OS t- op

CO fH U5
rH CO

00 00 CO
'^ SO CO

so t~ OS
'•it CO »o
t- 01 -^

rH «0 -^
«p CO «
6 CO rH

2^o3

(M 00 cq

-^ W5 (>l

"* >0 CO

rH rH CO
OpOSrH
OS SO CI

so t- ws

u, , 1

S 1 1

:^ 1 1

;^ 1 1

:s 1 1

Fresh beef

Average

Maximum .

Minimum .
Canned beef, roast
and boiled

Average

Maximum .

Minimum .
Canned beef, corned

Average

Maximum .

Minimum .
Canned beef, smoked
and dried

Average

Maximum .

Minimum .

Fresh horse meat
Average
Maximum .
Minimum .

FLESH FOODS.

371

S-rt*

-^ ^ ©

O (M t>

U3 CO 0)

<X> CO c^

on 00 -^

S3fl

CO 00 «

CS Tt* Op

6 6 J

rH (N CO

oceo

Oh g

^■i ^ 6

rH ^ 6

rH -t O

i-t -»a

"TJ •

00 O 00

^ <M 00

OS tH 00

rH (N O

CO t^ O

S-^

§3 a

op Oi t^-

t> « !>.

Ocpo

CO 00 O

00 CO CO

^-j_

A.S

« OS o

i-H

CO 6 6

rH (M O

« CO rH

C?l|6 rH

C-rH«0

CO «5 (M

rH •«1< ,

OS CO «o

t- rH (N

«^ s

'n «5 rH

•T" T" ?"

00 -^

CO O >f)

CO 00 rH

Sfi

&H g

tH -* 6

rH ^ 6

6r:H '

rH (N 6

6 -H 6

latinoids
Proteins
pt'd by
romine.

C<l ■M OC

CO O IN

CS O

«o ■* i«

rH 00 OS

« CO JO
rH -^ 6

CO O rH
r^ CO 6

CO l>. 1

«o -* t>

r^ IN 6

IN 00 rH
rH CO 6

»'gCu,PQ

a:-5s

«OCT>0>

"* O O

«0 rH CO

OS OS O

t> U5 00

CO «p «p
■4»t OS »0

ooo

U5 O r-(

oopo

OS t- rH

CO rH IC

db t- c^

00 l> CO
rH 6 IN

iH (M

rH <>> rH

<-( 7^ <-i

rH <N

rH<N

.2^'

CO >0 OS

CO 05 «C

t- (?q t-

rH t^ O

(N IN »0

-^ t>. «c>

OS rH «b

fP 9

OS -^ CO

OrHCp
-M C<« CO

«p qp o

<N rH 6

OS rH (N
CO -^ CO

1-1 'tjt

rH 3<I ^

Cfl CO rH

•N CO rH

rH <N

Tj( 00 lO

OS CQ t-

i£> iO<X>

■<# IC 00

rH (N -^

»0 CO o

co t- o

(M M* O

rf.«C,H

(N ^ O

6 rlf 6

6 6 6

66 6

o o o

6 6 6

l-i

-2 s

.JJ

o ■* «o

■H 00 (M

IN «0 rH

«0 OS <N

OS IN CO

•a.o

1-^ u

(M t~ O

(?« -rl< O

IN C- rH

<N CO rH

'7^99

666

6 6 6

6 66

o o o

iJj

u -^

00 »o ^

oo<o

»C IC t~

CO >c o

IN CO 00

CO b- OS

« "* 6

>C iN t--

^ CO rH

*? "* T

CO -^ rH

00 CO IN
IN "* rH

op op CO
rH CO 6

-^i

t-i "*^

rH 00 t-

rH ICOS

00 -* ■*

(N O O

CO CO IN

S s
oig

rH «p O

rH OS O

lO rH rH

CO rHCp

<N qp W5

«5 6 rH

CO CO (?q

CO »0 (N

CO W5 rH

(N CO 6

rH C- -^

CO OS -*

O rH (N

Cq t- <N

IN "* CO

§3 c

op O (X

OJ o qp

CO CO qp

C-- OS O

00 op »p

f^S

b- -^ 6

6 db 6

6 6 rH

IN 6 <N

rH rH CO

CO ?D rH

(M CO

rH CO

OH -*

5ri

t- o -^

C-O 00

iNSOSO

■<*• O U5

rH rH Ol

^ s

C~ CO CO

rH 00 »0

-* O "*

•^ CO OS

OcpO

■ 05

ft^g

6 so irt*

O rH 6

605t-

■N 6 '^

00 00 -*

CO « rH

«5 C- CO

«o to -*

CO C- Ti*

IC 00 -*

»- -*j

^ t-.

«® 2

CO 1 1

"^ I 1

<=• 1 1

»c 1 1

»C 1 1

S-rt

o.g

I-H 1 1

r-4 1 1

rH 1 1

i-t \ \

IN 1 1

s c

;2;<<

rr! . . .

• • ' n

. . .

00
ID

a * ' *

0) . . •

■5 'sal

^ . . .

fl ® S r3

c M

5 S

^ is

O ® S H

O OJ H S "^

fl> ID a H

d O 3 H

nedf
verag
axim
inimi
ned

gara(
verag
axim
inimi

ned 8
verag
axim
inimi

g -^^^^

o o

-.JSS

g-:sa

372 FOOD AND DEUGS.

Properly prepared preserved foods in cans or glasses should be
hermetically sealed and sterilized by heat. They will then keep for
an indefinite time.

If not properly sterilized, gradual decomposition will take place —
unless preservatives have been added — and gases due to fermentation
will be given off. Any tin that is bulged by internal pressure should
be rejected at once. In the trade such tins are known as " blown
tins ". If it be necessary to examine the gases given off from such a
tin, the process of collecting them described by Doremus (" Jour. Amer.
Chem. Soc." 1897, 19, 730) may be used. The tin, held in place by
a clamp, is pierced by a hollow steel needle passing through a closely
fitting hole in a rubber cork, which, being pressed hard into the tin
forms a tight joint ; the needle is connected by a fine glass tube — al-
most capillary — to an eudiometer or nitrometer tube filled with mer-
cury, and the gas, which may amount to 50 to 80 c.c. in large tins that
have decomposed is collected in the ordinary manner and examined.
A drop of lead acetate solution will indicate the presence of sulphur-
etted hydrogen, but the bulk of the gas will usually be found to be COg.
Sometimes, however, hydrogen predominates. It must not be inferred
that because a slight discoloration is found on the metal of the can
that putrefaction has taken place. In a case recently exhaustively
investigated by the writer, in which black patches were found, it was
proved that the gelatine used to "set" tongues in contained an
appreciable amount of sulphurous acid (used to bleach it).

The solder and the surface of the tin were of different metallic com-
position and appear to have set up electrical action, and the sulphur-
ous acid was partially reduced to H^S.

No bad odour should be detected on opening the tin; but it must
be remembered that, especially with smoked products, there may be a
trace of volatile matter derived from the smoking process, which is
smelt directly the tin is opened, but which in a few minutes has disap-
peared. If a piece of red litmus paper held close over the tin which
is stood in hot water, turns blue, the contents should be rejected. At
the same time a slight alkaline reaction of the actual contents is not
evidence of decomposition, as, for example, normal tinned lobsters
are often alkahne, but an alkaline reaction should be regarded with sus-
picion. A microscopic examination is necessary, as, if the meat is in
good condition, the muscular fibres will show their characteristic cross
striations, whereas if numerous bacteria are found, they often give
little coloured patches which destroy these striations in places. Such
a sample should be rejected.

The question of metallic contamination is one of considerable im-
portance. In general, one has to be prepared for the presence of tin
in tinned goods ; lead in such goods when a lead soJder has been used ;
and copper under the following circumstances : (1) where copper
utensils have been used for cooking ; (2) where much gelatine has been
used, and the gelatine is contaminated with copper in the course of its
manufacture ; (3) where it has been added as copper sulphate to pre-
serve the colour of green vegetables ; this, however, being rare in
the case of mixed meat and vegetables, but common in the case of

FLESH FOODS. 373

preserved peas and spinach, which should always be examined for
copper.

Tin. — In a recent report issued by the Local Government Board,
Schryver details the following method for the determination of tin in
tinned meats : —

He destroys the organic matter as far as possible by heating with
sulphuric acid and potassium sulphate in the manner used in the
Kjeldahl process. The clear liquid remaining in the flask is diluted
to 100 c.c, and the tin precipitated as sulphide by sulphuretted
hydrogen, and filtered after standing overnight. If the quantity
of tin present is comparatively large (over 2 grains per lb.), 50 grms.
of the sample are treated as above (in two flasks), and the tin con-
verted into oxide and weighed. For smaller amounts of tin, Dr.
Schryver devised the following colorimetric process.

The filter paper with the precipitate of tin sulphide, sulphur, etc.,
from 10 grms. of foodstuff is transferred to a test tube and boiled with
5 c.c. of concentrated hydrochloric acid. The liquid is filtered on a
suction filter, the filter being washed with another 2-^ c.c. of acid.
The air is replaced above the filtrate by a current of carbon dioxide,
and a standard strip of zinc foil (2 x -^ in.) added to the hot liquid.
Two c.c. of the special reagent (0-2 grm. of dinitro-diphenylamine-

sulphoxide, NH (' p^h^NO^ / ^^' ^^ ^^^ ^'^' ^^ decinormal sodium
hydrate solution) are then added and the solution boiled for one or two
minutes, then diluted to 100 c.c. with water, and filtered by suction.
If tin is present the solution turns violet during filtration, the full
depth of colour being attained by the addition of a drop of ferric
chloride solution. The process, after obtaining the sulphide, only
takes a few minutes. The amount of tin present is found by com-
parison with a standard solution of tin chloride (containing 11'28
mgm. of tin per 100 grms. or the equivalent of 1 grain per lb.). The
colour, due to the formation of Lauth's violet, is not strictly propor-
tional to the amount of tin present. One-tenth of a grain of tin per
lb. gives an appreciable colour, while with over one grain it is advis-
able to take less than 10 grains of foodstuff or estimate the tin gravi-
metrically.

The table on page 374 gives an abstract of the results of Dr.
Schyrver's examination of various canned foods.

Commenting upon the results, Dr. Buchanan states that meat
extracts and essences, owing to their natural acidity, take up tin to a
greater extent than other meat products. From the same cause
canned fruits and vegetables are also specially liable to take up tin,
the metal may penetrate into the substance of the solid food, which
may come to contain relatively larger proportions of tin than the
liquid. If solder gains access to the interior of the can, a very con-
spicuous solution of tin may take place. The results obtained by
Dr. Schryver, from experiments on himself and animals, are briefly
as follows : —

That there is no evidence of a cumulative action of tin until the
daily dose exceeds 2 grains.

374

FOOD AND DEUGS.

Foodstuffs.

Origin.

Grains of tin per lb.

Bacon, sliced
Beef essence
Beef extract
Curried rabbit
Fruits

Jams

Lobsters

Plum-pudding

Pork-pie

Roast fowl

Salmon

Tomato soup

Vegetables

U.S.A.

England (3 makers)
S. America (2 makers)
Australia
London importer
j England

(U.S.A. (tin pierced)
U.S.A.
England
England
England

British Columbia
U.S.A.
Australia

0-61 1
1-58 to 1-92
0-40 „ 5-33 1
0-19

0-33 „ 1-03
1-42 „ 2-81
513
2-39
trace
2-92

0-58 ., 1-44
0-4 „ 0-6
3-5
1-51 „ 2-19

That there is relatively small amount of absorption of tin from
the alimentary tract.

The experiments support in the main Lehmann's conclusions,
that there is not much probability of serious risk of chronic poison-
ing from a diet consisting largely of canned foods and continued over
considerable periods of time.

Dr. Buchanan states, however, that the presence of tin in a sample
can in quantities approaching 2 grains to the pound may be taken
to signify that the food has become potentially deleterious to health,
and calls for the examination of further samples.

F. Wirthle (" Chem. Zeit." xxiv. 263), reports on the examination
of samples of preserved meats of various ages up to five years. The
metal of the tins contained only 0*21 per cent of lead, and there was
no soldered joint. The amount of tin present was found to increase
with the time of preservation, and the meat to contain three times as
much tin as the juice. The interior of the tins was found to be cor-
roded almost solely where there was an accumulation of fat. They
were not acted on where they came in contact with gelatin. In five-
year-old tins a white crust was formed which consisted of basic tin
chloride, due to the action of the sodium chloride present on the tin.
In four-year-old tins a black layer of sulphide was present. The tin
was determined in the meat and juice, by the following modification
of Orfila's method : About 120 grms of meat (or juices, separated
from the meat) were placed in a large porcelain dish of nearly 1 litre
capacity, moistened with 5 c.c. concentrated sulphuric acid, and heated
carefully on a sheet of asbestos. It was frequently stirred, and from
time to time small quantities of sulphuric acid were added, altogether
about 15 c.c. to 20 c.c. ; the mass was occasionally removed from the
sides of the dish, to which it adhered, by means of a porcelain spatula.
After four or five hours, a porous carbonaceous mass was thus ob-
tained, which was pulverized and incinerated in a porcelain crucible.
The particles adhering to the porcelain dish were transferred to the
crucibles with the assistance of powdered anhydrous carbonate of soda;

FLESH FOODS.

376

a further proportion of carbonate of soda was added, together with a
sufficient quantity of nitrate of soda, the whole thoroughly mixed, and
heated to gentle fusion. After cooling the melted mass was taken up
with water, and when the cloudy solution had become clear (which
generally takes places after about twelve hours), the precipiiate was
collected on a filter, well washed, dried, and incinerated. The ash
was treated with a sufficient quantity of potasssum cyanide, and the
mixture heated to dull redness, the crucible being closed. The melted
mass waa taken up with warm water, and the metallic tin and the
iron collected on a filter, washed, and dissolved in a little warm hydro-
chloric acid. In the solution, which should not be very acid, the tin
was precipitated with sulphuretted hydrogen, the precipitated sulp-
hide was washed with water, saturated with sulphuretted hydrogen,
containing a small quantity of nitrate of ammonium, and then
dried, incinerated, and calcined until the weight was constant. The
weighed stannic oxide was reduced once more by means of potassium
cyanide ; the tin thus obtained was dissolved in hydrochloric acid,
precipitated in the form of sulphide, and weighed as stannic oxide.
The minimum amount of tin found in the meat was 0*0029 per cent,
and in the juice 0-0011 per cent. The maximum was 0*016 per cent
in meat, and 0*0036 in the juice.

Allen detects heavy metals according to the following scheme : —
The substance to be examined is heated on a water bath, and
finally at a rather higher temperature with sufficient strong sulphuric
acid to well moisten the whole of the substance, with which it is incor-
porated ; 1 c.c. of HNO3 is then added, and the heating continued
until red fumes are given off. Ignited magnesia (0*5 grm. for each
grm. of acid used) is now well mixed with the substance, and the
whole ignited at a dull red heat. After cooling, the ash is moistened
with nitric acid and reignited. This treatment is repeated until the
ash is white or grey. Ten drops of H^SO^ are added, the whole heated
until fumes are evolved, cooled, boiled with water, diluted to about
100 c.c. and without filtering, treated with HgS to saturation. The
solution is now filtered and the following scheme of analysis
followed : —

Aqueous solution may contain zinc and iron, j Ppt. may contain Pb. Sn. or Cu. Fuse in porce-
Add Br. water to destroy H2S. Add excess lain with 2 grs. of sodium-potassium carbo-

of Nfla, boil, and filter again.

Ppt. may con-
tain iron.

Filtrate, if blue, contains
nickel. Divide into two
parts^

1. Heat to boil-
ing and add
potassium
ferrocyanide
white ppt.=

2.- If zincf ■ be
found in 1,
determine it
by adding
acetic acid
and precipi-
tate with
H2S. Nickel
if present
will be in-
included.

nate and 1 gr. of sulphur. Cool, boil with
water and filter.

Residue. Boil with strong
HCl, add Br. water.
Filter. Add NHn— a blue
colour indicates copper.
Acidulate with acetic
acid and divide into two
parts. To one add potas-
sium bichromate, a
yellow ppt. indicates Pb.
To the other add
potassium ferrocyanide,
a >)rown ppt. indicates
copper.

Filtrate. Add
excess of acetic
acid, a yellow
ppt. indicates
Sn.

376 FOOD AND DRUGS.

Exceedingly minute amounts of copper may be detected by insert-
ing a bright steel needle into a slightly acidulated concentrated ex-
tract of the ash, removing it after some hours, cautiously rinsing with
water and then immersing it in ammonia with free contact of air.
Copper will be detected by acidulating the ammonia solution with
acetic acid and adding potassium ferrocyanide when a brown colour
or precipitate w'ill be formed.

Preservatives in Tinned (or glass-contained) Meats. — The most
important publication dealing with this matter that has recently ap-
peared is the report of Dr. A. W. J. MacFadden to the Local Govern-
ment Board of '26 May, 1908, with an analytical addendum by P. A.
Richards. The following are the most important portions of this
report : —

During the summer and autumn of 1906 a large number of public
analysts throiighout the country received samples of canned and glass-
packed meats, and in response to a request by Dr. Buchanan furnished
special reports on the result of their examination of these foods. From
these reports it was noticed that the percentage of samples in which
preservatives were found was much higher than might have been ex-
pected having regard to the fact that the cans or glasses in which
these articles were packed had been hermetically sealed and, presumably,
sterilized by heat in the usual way. Out of a total of 1733 hermeti-
cally sealed tins or glasses dealt with in the above-mentioned returns,
no fewer than 333, or over 19 per cent, were reported to contain
chemical preservatives other than salt and saltpetre. Of these pre-
servatives, 243 were boron compounds, forming 14 per cent of the
whole, and in the remaining 90 samples the presence of sulphite pre-
servatives was reported.

In the case of meat foods of this kind it is generally understood
that, with certain possible exceptions, there should be no need for the
addition of chemical antiseptics at the time of preparation of the meat
for canning, and that the meat which is canned should ordinarily be
fresh meat, or cured meat — not meat which has been subjected to
treatment involving the introduction of preservatives such as boric
acid or sulphites. It is to be expected that meat foods of this kind
which have been submitted to a process of sterilization in hermetically
sealed containers should be sufficiently protected by these means from
processes of decomposition so long as they remain unopened ; chemi-
cal preservatives are not required to further this object.

In these circumstances the use of preservative materials of the
kind referred to appeared liable to objection, apaii from any risk to
health arising from consumption of the preservatives themselves, in
that their presence pointed to the probability that by this means it had
been sought to overcome undesirable conditions either in the meat
itself or in the processes of its manufacture.

The following general conclusions are drawn : —

(1) Preservatives in Imported Canned Meat Foods.

The finding of preservatives in a considerable proportion of Ameri-
can and other imported canned meats, examined by public analysts in

FLESH FOODS. 377

this country, is a matter of importance from a public health aspect.
The question is not merely one of the ill-effects likely to be produced
in persons who, in this way, consume what may possibly be large doses
of these substances or who, by taking these foods, add unnecessarily
to the total quantity of antiseptic substances in their diet. The chief
objection which may be raised to preservatives occurring in meat foods
which have been subjected to the process of sterilization by heat in
hermetically sealed vessels, is that the presence of preservatives must
be regarded as indicating that conditions as to care and cleanliness
which are essential in the preparation of wholesome food materials
may not have been observed.

Dr. Eugene H. Porter, Commissioner of Health of the State of
New York, has referred to this matter in the following terms : —

« " The use of any preservative in a food to be enclosed in a can which
can be satisfactorily sterilized by the use of heat and sealed hermetic-
ally,' indicates that the materials to be placed in the can were in such
a state, or were kept under such conditions, as to lead the canner to
believe that they required the use of a preservative for the prevention
of decomposition until they could be safely canned.

" The finding of a boron preservative in a sample of "potted ham,"
in which were found numbers of the Trichina sjnralis, clearly indicates
that in the minds of those who prepared this meat, it required a pre-
servative in order to prevent its decomposition before it could be
sterilized in the can.

" It is not easy or always possible to ascertain from an examination
of a sample of canned meat containing a preservative whether decom-
position had set in at the time the preservative was added or not, but
the possibility or the probability of the development of such undesirable
changes must have been present or the preservative would not be
added."

The views expressed above undoubtedly offer the correct explana-
tion of the circumstances which lead to the presence of boron com-
pounds in foods of this kind in the great majority of cases, and it is
-difficult to accept the explanations in this matter which are commonly
^iven by representatives of canning firms.

Similar considerations may be said to apply to sulphite preserva-
tives. These seem to have been reported more frequently in American
canned meats than in those of British manufacture, and though there
are chemical and analytical conditions which impair the value of
quantitative analytical results obtained in regard to them, the fact
that sulphites are frequently found in canned meats cannot seriously
be questioned, and their presence can only suggest the same inter-
pretations as that just given.

Extension of Protection Afforded by Foreign and Colonial Laws. —
The United States Meat Inspection Act of 1906, which was the out-
come of inquiries made into this, together with other aspects of the
American meat-packing trade, prohibits the use of artificial preserva-
tives in meat and meat food products, and it may be hoped that
the very comprehensive and stringent provisions contained in that
measure are already having, or will have, the effect of abolishing those

378 FOOD AND DRUGS.

practices which lead to the use of chemical preservatives in canned
meats.

Legislation as to supervision of the preparation of meat foods in
various States of Australia, in New Zealand, and more recently in
Canada may also be alluded to as probably affording a substantial
check on the use of preservatives in canned meat foods which are
subject to official supervision in those Dominions.

It must, however, be remembered that the extent to which foreign
and colonial laws and regulations regarding canned foods are likely
to be and to remain operative in practice in the case of the exports
to the United Kingdom, depends very largely on the efficiency of
precautionary measures taken in this country. At a time when
systems of official inspection, regulation and certification in connexion
with the manufacture and preparation of meat foods are being rapidly
developed in foreign countries and British dominions, there would
appear to be special advantage in specifying British requirements in
such a way that foreign and colonial manufacturers and officials are
in no doubt as to what the public health authorities in Great Britain
expect them to avoid or to do in regard to the use of preservatives
in canned meats exported to this country, and in arranging some
systematic method of analytical control for the purpose.

(2) Preservatives in Meat Foods Canned or Packed in
Glass in the United Kingdom.

Preservatives in the Baw Materials. — Boric acid in large amount
has been found in canned meats of British manufacture, as the result
of employing meats imported from abroad in this preservative. Con-
victions have been obtained by local authorities under the Sale of
Food and Drugs Act in such cases. But the number of instances
coming to the notice of these authorities in which amounts of boric
acid are sufficiently large to ensure successful prosecution is neces-
sarily small and forms but a slight check on the practice of employing
such meats for canning purposes.

The second and larger class of meats which come under this head
are bacon and hams. These are meats which in normal circumstances
are cured by means of salt and saltpetre, and occasionally sugar. The
process of curing renders the meat more or less immune to the in-
fluences of putrefactive organisms according to the length of time during
which the meat has been kept in " pickle ". It was the custom formerly
to continue the curing process for a sufficient length of time to ensure
that when removed from the pickle and properly dried, hams and
bacon would remain sound for long periods. They were, however,
liable to be affected in various ways where conditions as to curing or
storage were faulty. Thu3 putrefactive processes like " taint," and con-
taminations such as flyblows were accidents which required constant
guarding against by bacon curers.

When borax came into use as a preservative it was found that these
conditions were prevented by sprinkling the surfaces of the meat with
powdered borax. Later this substance came to be used as a con-

(

FLESH FOODS. 379

stituent of the pickle, and it was found that by employing the pre-
servative in these ways the curing process could be shortened without
to a corresponding extent endangering the keeping qualities of the
meat. The lessened salty flavour of hams and bacon which followed
the introduction of these methods appears to have met with favour
among consumers generally, and it has since been the aim of most
bacon and ham curers to produce materials as mildly cured as pos-
sible. In these circumstances the use of boric acid has come to be re-
garded by many bacon manufacturers as a necessity, more especially
where their products have to be kept for long periods or are intended
for distant markets. On the other hand, some well-known British
manufacturers declare that no such necessity exists, and that they are
able to prepare, store and export their products to all parts of the
world without having resort to boron preservatives, or other anti-
septics, at any stage.

Before the use of borax was introduced for the purpose mentioned
above, hams and bacon were sent from the United States to this
country packed in salt. This method of transport amounted prac-
tically to a continuation of the curing process for a number of weeks
longer than was the custom in the case of home-cured materials and,
owing to their consequent increased saltness, American hams and
bacon suffered greatly by comparison with British produce in the
home market. The use of borax in the manner above mentioned has
enabled American and Canadian curers to send their produce to Great
Britain in a condition which, as regards " mildness " of flavour, com-
plies with the taste of consumers in this country.

Borax-packed Hams ajid Bacons as a Source of Preservative in
Canned Meats. — The amount of borax acid found in bacon and hams
imported under the above conditions will depend on whether the pre-
servative has been employed only as a packing material, or has in
addition been used in the curing of the meat. Examination of samples
of hams said to have been only packed in borax shows that amounts
varying from 2-6 grains to IS'O grains per pound of the minced and
mixed ham may result from mere contact with the preservative in the
packing cases. Larger quantities than this, up to 24 grains per lb.
of boric acid, were mentioned to the Departmental Committee on Pre-
servatives and Colouring Matters as having been found in hams, and
the Committee refer to this as an indication that boric acid had been
used in the curing of these products.

Since the passing of the United States Meat Inspection Act the
use of borax in hams and bacon sent to this country is said to have
been confined to external application as a packing material. Generally
it may be said that this use of borax is free from the objection that
the preservative has been used to mask the effect of objectionable
conditions of preparation, etc. It may be sufficient in this connexion
to observe that if the use of boron preservative in food were to be re-
stricted with the object primarily of limiting the number of foods
which at present may contain this preservative, and were to be con-
fined to such foods as in present circumstances appear to demand
borax as an essential condition of trade, a good case would be made

380 FOOD AND DRUGS.

for permitting the continuance of borax packing in the case of
American and Canadian hams and bacon, provided the amount of
preservative absorbed by the meat was not excessive. On the other
hand no trade necessity can be alleged for using borax in the curing
process itself, a view which is borne out by the fact that for some
time past American manufacturers have apparently been able to dis-
pense with this use of the preservative.

The amount of preservative which may be found in certain varie-
ties of canned meats, due to their having been manufactured from
hams and bacon in which boric acid had been employed solely as a
packing material, would in most cases be small and of little import-
ance as a public health consideration. But it would be necessary to
take account of such small quantities of preservative in any scheme
which might be arranged for controlling the use of preservatives in
canned meats. If potted meats made from imported hams were
exempted from any restriction as regards preservatives that might
be applied to canned meats in general, such exemption would be
certain to afford a convenient plea for manufacturers who desired
to continue the improper use of boric acid in potted meats, and the
•difficulty of administration would be greatly increased in such circum-
stances. It should be remembered in this connexion that hams and
bacon which are free from preservative can be obtained in British
markets, and that some well-known manufacturers, in order to en-
sure the freedom of their products from boric acid, have already
taken the precaution of using only those American hams and bacon
which'.are sent to this country in salt packing and guaranteed to be
free from chemical antiseptics. Apparently an unlimited supply of
these salt-packed hams could be had if demanded, and for a price not
greater than that charged for borax-packed materials. Danish bacon
free from borax is also available, and it could no doubt be arranged
by potted-meat makers who desired it, to obtain bacon and hams of
British origin which did not contain any preservative of this kind.

Addition of Preservatives by the Canner. — As already stated the
presence of boric acid in canned goods of British manufacture may to
a certain extent be attributed to sources referred to in the foregoing.
In a great number of instances, however, the presence of preservatives
in these materials cannot, for various reasons, be explained in this
way. Such is the case with, for example, many of the " potted meats "
which have been shown by analysis to contain exceptionally large
quantities of boron preservative. Notwithstanding the fact that ^hese
materials are in most cases mixtures of two or more meats and often
contain in addition a large percentage of farinaceous matter, the
amounts of boric acid found in them have frequently been much
greater then is ever likely to be found in raw material of the kind
above-mentioned from which they might have been made. The only
possible explanation of the presence of preservatives in such circum-
stances is that they have been deliberately added to the material in
process of manufacture.

It has been indicated in section 3 of the report that the use of
meat which has become, or is on the point of becoming, tainted may

FLESH FOODS. 381

lead to the use of chemical antiseptics (as by immersing in fluids,
spraying or powdering) for the purpose of preserving it. It has also
been shown that even where sound materials are used to start with,
faulty, dilatory and uncleanly methods of manufacture may so far
affect the condition of the meat previous to packing that the use of
antiseptics has to be resorted to.

It will be gathered from descriptions given in preceding pages that
the various steps in the manufacturing process of canned and glass-
packed goods are operations which require considerable skill and care.
In handling the perishable material dealt with, great cleanliness is
needed and every convenience for securing this and for enabling the
various steps to be carried out with despatch, is essential if the'material
is to reach the consumer in a wholesome condition. In the majority
of the factories visited, care seemed to be taken to ensure that these
essential requirements were observed. Reference has, however, been
mad6 to instances in which one or other important point was not at-
tended to. A few of the factories were, indeed, quite unfitted for the
use to which they were put. Places hidden away among the dila-
pidated buildings of dingy streets or beneath remote railway arches —
sometimes even within the same curtilage as premises in which of-
fensive trades are conducted — cannot be considered fit for the prepara-
tion of food material intended for human consumption. The structural
defects of such places are sufficient in themselves to render the cleanly
preparation of canned meats in them impossible, but when, in addi-
tion, as is frequently the case, the appliances used at these factories,
and the workpeople who use them, are of no higher standard of clean-
liness than the buildings themselves, it is not difficult to understand
the need which arises for means which will counteract in some way
influences so adverse to the preservation of the meat used.

I found many grades of greater or less efficiency between such en-
tirely undesirable food factories and others which could be pronounced
without hesitation to be satisfactory in all essential respects. But
neglect of some obvious precaution, such as the necessity for carrying
out certain stages in the manufacture speedily, and under suitable
conditions of temperature, etc., may render unavailing the care taken
in other steps of the process, and may lead to the addition of preser-
vatives at this stage.

There is a tendency on the part of some manufacturers to leave
the technical details of their business too much in the hands of the
workmen who carry them out. Many of these workmen look upon
the details of the processes which they are conducting as technical
secrets of their own, and resent any interference with the preconceived
notions which they may have regarding them. A manufacturer who
has not a well-grounded knowledge on these matters, or who, while
possessing this, fails to insist on his ideas being carried out, is at the
mercy of his foremen. The latter are in some cases content to pro-
ceed with the methods in which they were brought up, and are natur-
ally inclined to continue practices which protect them from the
consequences of miscalculations or mistakes. The use of antiseptics
from their point of view, possesses this advantage, and it is not to be

382 FOOD AND DKUGS.

wondered at that when given a free hand, they should avail themselves
of it.

Greater Use of Preservatives in Glass-packed Meats. — The use of
chemical antiseptics is especially undesirable in the case of meat foods
packed in glass, which have lately come into much favour with the
public. . Many complaints were made during my inquiry as to the
difficulties which had to be overcome in connexion with the manu-
facture of meats so packed. The chief of these was the difficulty ex-
perienced in. the sterilizing process, of steering a middle course between
the risk of breakage on one hand, and that of insufficient sterilization
on the other. This is doubtless an operation requiring care and ex-
perience. Most manufacturers, however, have been able to carry it
out with success, so far as the conferring of keeping qualities on their
meat is concerned, and some of them have even stated that they ex-
perience no difficulty whatever in the matter. The question seems to
be entirely one of efficient technique, and to avoid the danger by add-
ing sufficient antiseptic to preserve the packed material against the
consequences of faulty sterilization, amounts to a pretence and cannot
be justified, k still more unjustifiable practice is that which has been
referred to of adding boric acid to potted meats hermetically sealed in
glass, in order to avoid sterilizing them at all. To sell meats pre-
pared in this way in hermetically sealed containers, which purport to
be preserved by means of heat, or are of a pattern which the purchaser
generally associates with goods preserved in this way, is a form of de-
ception which may be prejudicial to the health of persons consuming
them.

The alleged inefficiency of certain forms of metal caps for pro-
ducing a hermetic seal with the glass containers for which they are
designed is, I am convinced, unreal. Complaints as to this were
made only by those who, either from fear of breakage, or in order to
preserve the homogeneous appearance of their potted meats, sterilized
these products imperfectly or omitted the process altogether. Many
of the better manufacturers had no difficulty with these covers, and
were able to rely with considerable certainty on their process of
sterilization alone to preserve the meats sealed by means of them.

It is undoubtedly more difficult to produce meats properly pre-
served by means of heat in glass containers than in cans. The high
temperature to which canned materials may be subjected for long
periods without risk renders the sterilization of such meats a much
more certain process than is the case with glass-packed goods. None
of the manufacturers whom I saw had experienced any real difficulty
in preserving their canned products by means of heat, and all denied
using preservatives, or the necessity for using them to supplement
this process in canned goods. To this extent canned meats in general
might be said to possess advantage, from the point of view of whole-
someness, over meats which are packed in glass. On the other hand,
the difficulties met with in sterilizing glass-packed meats necessitate
the observance of greater cleanliness and care in their preparation {so
long as preservatives are not added) if satisfactory results are to be ob-
tained. Added to this is the fact that their contents are more or less

_ FLESH FOODS. 383

capable of inspection by the purchaser, and for this reason very un-
likely to contain any of the grosser contaminations which have been
reported in some American canned foods, for example.

The freedom of glass-packed meats from liability to metallic con-
tamination is another point which may be noted in their favour, though
the practice among some British manufacturers of re-packing in glass
imported meats which have just been turned out of their cans for the
purpose, is calculated to shake the confidence of those who rely for
protection in this respect on buying only meat preserved in glass
containers.

(3) Administrative Considerations in Regard to Preservatives
IN Canned And Glass-packed Meats {Imported or Home-
prepared).

On review of the above, it seems desirable that steps should be
taken to secure that specified chemical preservatives should not be
used in the preparation of canned meats intended for consumption in
this country. In any schedule of prohibited preservatives, boron
compounds, sulphites and preparations of sulphurous acid, benzoic
acid and formalin should, I think, he included.

The action which for this purpose may be recommended for the
Board's considerations is such as could apparently be made avail-
able by the issue of suitable regulations under the Public Health
(Regulations as to Food) Act, 1907, made applicable (1) to imported
canned foods at the ports of entry; (2) to manufacturers in this
country of meat foods which are packed in cans or glass. Should
regulations be prepared with this object (either as a separate series
or as part of a larger series dealing with preservatives in a variety of
food-stuffs), it would seem desirable to make due allowance for trade
requirements in the matter of existing contracts, stocks on hand, and
so forth.

(4) Additional Observations.

In concluding this section attention may be directed to certain
matters which have incidentally come into prominence as a result of
my inquiries : —

[a) Proprietary Antiseptic Preparations. — It seems desirable that
manufacturers of meat foods should refuse to purchase any prepara-
tions offered to them for the purpose of treatincr meat in storage or
of adding to meat preparations or brines unless they are fully assured
as to the composition of any preparations so offered.

{h) Lahelling. — So far as my inquiry has been concerned with the
question of labelling and .marking of canned foods, it has brought
into prominence the fact that the custom which at present prevails in
this matter in many cases does not permit the origin of particular tins
of canned meats to be traced further than to the middleman or im-
porter. In some cases (including to my knowledge at least one in
which the wholesomeness of the foods was open to very serious
question) neither the can nor the label bears any name or other mark

384 FOOD AND DRUGS.

by which it can be referred to retailer, middleman, manufacturer, or
anyone at all. Reform in the trade custom in this respect seems on
many grounds desirable,

(c) Admixture of Starch with Potted Meats. — Starch, usually in
the form of rice flour, is sometimes present in such amounts in some
kinds of potted meats that these may be thought to constitute a fraud
on the consumer notwithstanding their cheap price. The matter
seems worth the attention of better-class manufacturers with a view
to arriving at a reasonable maximum limit of permissible starchy
matter in specified canned foods of this class.

(d) Preservatives in Sausages, etc. — Some of the objections above
set out to the use of chemical antiseptics in canned meats apply also
to such articles as sausages, pork pies, mincemeats, brawns, and the
like, which are not preserved by canning or cold storage, but in the
ordinary course of trade may be kept for several days on the shop
counter and similar places. If on the ground of public convenience
and trade requirements the use of chemical antiseptics is permitted
in these articles, it appears very desirable that their employment
should be restricted within narrow limits. Quantities of boric acid
are not seldom reported by public analysts in some of these goods
which cannot be otherwise than prejudicial in themselves, besides
being wholly unnecessary. If boron preparations are used for this
purpose, a limit of \ per cent of boric acid would probably be ample
to meet legitimate trade requirements, and even in this case it appears
desirable to consider whether notification of the presence of the pre-
servatives should not be given to the purchaser. The practice of using
solutions of sulphurous acid or sulphites as a spray or wash, or for
mixing with the meat, appears to be open to many abuses and to be
generally undesirable.

{e) Places where Food is Prepared. — The Board have recently re-
ceived representations from several sources to the effect that stricter
sanitary supervision and control should be exercised by local author-
ities in this country over premises on which meat foods are manufac-
tured and over the processes of preparation employed on such premises.
It has likewise been urged that all practicable steps should be taken
to require evidence from official sources as to the inspection or super-
vision in foreign or colonial establishments whence meat foods are
imported into the United Kingdom. The inquiries above recorded
have brought out many circumstances which appear to support these
views.

The United States Department of Agriculture in a bulletin pub-
lished in 1907 (No. 76 — Foods and Drugs Inspection Section) decided
that no drug, chemical, or harmful or deleterious dye or preservative
may be added to foods, or used in preparing them for the market,
except common salt, sugar, wood-smoke, potable distilled liquors,
vinegar, condiments, and, until further investigation, saltpetre. Sul-
phur dioxide is permitted within limits for wines and food-products
provided the amount does not exceed 850 mgrms. per litre in
wines, or per kilogram of food-products, but not more than 70
mgrms. should be in the free state. Sodium benzoate not ex-

FLE'SH FOODS. 385

ceeding 1 per mille or an equivalent amount of benzoic acid may be
used as a food preservative, and in this case, as well as sulphur
dioxide, the fact must be stated on the labels. The effects of coal-tar
dyes in foodstuffs are being investigated, and until the investigation
is complete they propose to permit the use of the following as noted
in Professor A. G. Green's edition of the " Schultz-Julius Systematic
Survey of the Organic Colouring-matters " published in 1904 : —

Bed Shades : 107. Amaranth. 56. Ponceau 3 K. 517. Ery-
throsin.

Orange Shade : 85. Orange I.

Yellow Shade : 4. Naphthol yellow S.

Green Shade : 435. Light Green S. F. yellowish.

Blue Shade : 692. Indigo disulphoacid.

Each of these colours shall be free from any colouring matter
other than the one specified, and shall not contain any contamination
due to imperfect or incomplete manufacture.

The various preservatives present in preserved foods may be
searched for by the methods described under milk (p. 160), wine
(p. 332), and meat extract (p. 413).

The use of benzoic acid has been common of recent years as a
preservative, especially in certain American preparations. The pre-
paration— if solid or semi-solid — should be well extracted by macer-
ating it with dilute alkali, straining through fine muslin and then
acidifying and extracting with an immiscible solvent as in the case of
salicylic acid. The ether or chloroform containing the free acid may
be, in turn, extracted with dilute ammonia, and the liquid evaporated
nearly to dryness, and the concentrated liquid tested. To a few drops,
a drop of neutral ferric chloride is added, when a characteristic flesh-
coloured precipitate is thrown down. In case the food contains
an artificial colour, which might mark the reaction, Mohler's
test may be applied. The ether extract is dried and the residue
heated with 2 c.c. of strong sulphuric acid, which converts it into
sulphobenzoic acid. A few crystals of KNO3 are then added, which
causes the formation of meta-dinitro-benzoic acid. When cold, the
acid is diluted with water, and ammonia added in excess, and then a
few drops of colourless ammonium sulphide solution. A red colour
is at once developed owing to the reduction of the acid to meta-
diamido-benzoic acid, whose ammonium salt is red.

A useful confirmation (in the absence of salicylic acid and saccharin)
is obtained by dissolving about O'l grm. of the ether residue sus-
pected, in 5 to 6 c.c. of H2SO4. A small quantity of barium peroxide
is then added, the tube being immersed in cold water, as fragments
are successively added, in all about 0'75 grm. After standing for
half an hour, the liquid is diluted with water and extracted with ether,
the residue being tested for salicylic acid, into which the benzoic acid
has been converted, in the usual manner.

Benzoic acid may be determined by the method of La Wall and

Bradshaw (" Amer. Jour. Pharm." 80, 1908, 171). Twenty grms.

of the material are well mixed with 2 grms. of sodium chloride, 5 c.c.

of HCl. and 25 c.c. of brine. The whole is well mixed and shaken for

VOL. I. 25

386

FOOD AND DRUGS.

ten minutes. Transfer to a moistened filter and after the liquid is
drained, the residue should be treated with three more quantities of
25 c.c. of brine, and drained into the filter each time, being washed
with more brine till the filtrate measures 100 c.c. Shake out the fil-
trate with three portions of chloroform (25, 15 and 10 c.c). Allow
the chloroform to evaporate at ordinary temperature, and dry to con-
stant weight in a desiccator. The result should be confirmed by
titrating the residue dissolved. in a little alcohol, with one-twentieth
normal KOH.

Sausages. — Sausages, although made to a considerable extent in
this country, are essentially a German delicacy, and the German
sausages imported into this country are many in number.

In this country sausages are principally made from pork and from
beef with certain spices and condiments, usually some colouring
matter, and frequently bread or other starchy material. The so-called
" German " sausages made in this country resemble no true German
sausage that the author has ever examined.

The following are the principal types of sausage manufactured in
Germany (most of which find their way to this country), after the
classification of Konig and by Merges.

Bothwurst (or Buntwurst) resembles the English *' black pudding ".
It is made from pork, bacon, often with the addition of heart or kidney,
various spices and frequently amylaceous material.

Mettwurst is made from pork, with a large addition of lard,
frequently beef and horse-flesh. It is frequently coloured with coal-
tar dyes.

Cervelatwurst. — This is generally made from the brains of pigs and
horses, with the addition of pork and lard, and usually a little colour-
ing matter.

Leberwurst. — This is made from the livers of pigs and calves, with
the addition of pork and lard. Frequently the liver and lungs enter
into the composition of this sausage, as well as some starchy matter.

Magenwurst is made from the stomach, skin and other parts of
the pig with blood and unsalted bacon.

Bratwurst is made from raw pork, and bacon with lemon and
cumin as flavourings.

Erbsiourst is made from suet, bacon, pea-flour, onions and various
other seasonings.

Frankfort sausages are small sausages made of raw pork and
seasonings.

The following are typical analysis of several types of German
sausages purchased and examined by the author : —

Leberwurst
Mettwurst
Cervelatwurst
Frankfort sausages

Water.

Fat.

Carbohydrates.

Nitrogen.

Ash.

Per cent
43-5
27-4
33-8
40-8

Per ceut
24-5
36-9
41-0
31-9

Per cent

10-6

7-3

2-8

6-5

Per cent
2-3
3-26
2-74
2-45

Per cent
5-4
6-0
4-95
4-16

FLESH FOODS.

387

Konig has published a number of analyses of these and other
sausages, but they all have similar compositions to the above.

Allen gives the following as the composition of average quality
English- made sausages: —

Water.

Fat.

Proteids.

Gristle, etc.

Starch.

Ash.

Per ceut

Per cent

Pork

54-99

21-04

12-88

0-67

1-05

3-52

Mutton

55-58

30-57

1-89

3-11

3-90

2-50

" German "

49-54

17-87

16-38

1-lH

15-00

4-47

Poloney

45-57

32-66

17-26

0-54

2-30

2-80

As a rule the determinations required in the examination of saus-
ages are those of preservatives and artificial colouring matters, with a
further examination for the presence of parasites, and, occasionally,
for the presence of horse-flesh. A fuller investigation is sometimes
required, however, when the following methods will be found suffi-
cient : —

Moisture and Ash. — This is determined by drying 10 grains at 105"
to 110° to constant weight. The ash is determined by drying and in-
cinerating 2 to 3 grms.

Fat. — The fat may be determined by extraction of the above dried
residue (preferably rubbed down with sand) with ether in a Soxhlet
tube, if an examination of the fat be necessary, a larger quantity
must be extracted. The following are the typical characters of the
commoner fats met with in sausages : —

., 100°
Sp. gravity -j^

Ox.

Sheep.

Pig.

Horse.

Per cent
0-860

Per cent
0-861

Saponification value
Iodine value
Rei'ihert value
Refractive index at
60°

192 to 198

36 „ 44

0-25

1-4510

192 to 196

34 „ 44

0-3

1-4500

194 to 196
50 „ 68
0-4 „ 0-6

1-4540

195 to 198
75 „ 85
0-8 „ 1-1

Butyro-refractometer
No. at 40°

48 to 49

50 to 51

53 to 54

Starchy Matter. — A small portion of the sausage may be teased on
a microscopic slide and a drop of iodine solution added. The pres-
ence of starch is easily seen by observing the black stained starch
granules under the microscope.

Mahrhofer (" Analyst," xxii. 2) determines starch in the follow-
ing manner : From 60 grms. to 80 grms. of the sample are heated
on the water bath with alcoholic KOH (10 per cent). Nearly every-
thing is dissolved. The solution is diluted with warm alcohol in order

388 FOOD AND DEUGS.

to prevent gelatinization, and then filtered. The insoluble residue con-
tains the starch, if any be present, and is washed with alcohol until
free from alkali, and then treated with aqueous potash, which dis-
solves the starch, and made up to a definite volume. An aliquot por-
tion is precipitated with alcohol, when the starch can be collected on
a dry tared filter, washed with alcohol and ether, dried and weighed.

Medicus and Schwab ("Berichte," xii. 1285) recommend the diges-
tion of a weighed quantity of the sample for ten hours with a definite
volume of an infusion of malt, at 30° to 40°. The mixture is then
allowed to stand for eighteen hours at ordinary temperature, and then
filtered, the insoluble matter washed, the filtrate boiled, and the pre-
cipitated albumen filtered off. The filtrate is then boiled with HCl to
convert the dextrin and maltose into dextrose, which is determined in
the usual manner by Fehling's solution. The amount in the volume
of malt infusion used, is determined, and deducted from the result.
Ten parts of dextrose may be taken as representing 9 parts of dry
starch originally present. It must be remembered that a very small
amount of starch will usually be present on account of the pepper
added to the sausage meat.

Determination of Total Nitrogeii. — Two grms. of the sample are
submitted to the Gunning or Kjeldahl method of treatment. In the
case of meat the customary method of taking N x 6*25 as representing
the total protein or nitrogenous substances, cannot always be relied
upon, on account of the varying amount of nitrogenous matter con-
tained in various compounds present, though a fairly close approxi-
mation of the nitrogenous substance present can be calculated by
making use of this factor, as proteins are by far the largest group
contained.

Separation and Examination of Nitrogenous Bodies. — It entirely
depends on the nature of the sample in hand how far an analyst
should subdivide the various nitrogenous bodies present in meat. The
above-mentioned simple determination of total nitrogen is frequently
sufficient. As a rule there is no necessity to do more than divide the
nitrogenous bodies into several main groups according to their solu-
bility in water or other solvents, and their attitude towards certain
reagents. Nitrogen can be determined separately in each of these
classes, and the corresponding nitrogen substance or class of sub-
stances can be obtained by the appropriate factor.

To more completely separate the various classes of nitrogenous-
bodies found in meat, agitate a portion of the fat-free sample with
cold water to remove the soluble proteins (soluble globulins, proteoses.
and peptones) and meat bases, leaving behind the insoluble globulins,
the sarcolemma, the albuminoids of the connective tissue and the
collagen. Then treat w^ith boiling water, thus removing collagen in
the form of soluble gelatin. The soluble proteins, including the
peptones and gelatin, can be precipitated from the meat bases by add-
ing zinc sulphate, sodium chloride, and tannic acid to the combined
aqueous extract.

Determinatio7i of Nitrogenous Substances Insoluble in Water. —
Thoroughly wash the sample with cold water, transfer the filter and

FLESH FOODS. ^^^^^ 389

insoluble material to a flask, then determine the nitrogen by the
Gunning or Kjeldahl method. Multiply the insoluble nitrogen thus
obtained by 6-25 to obtain insoluble proteins. The insoluble nitrogen
can also obviously be obtained by deducting the soluble from the
total nitrogen. Dilute the cold water extract to definite volume,
determine the nitrogen in an aliquot portion, and calculate to per-
centage of soluble nitrogen in the weight of total extract. Having
obtained the percentage, deduct it from the percentage of total
nitrogen, and the result is the percentage of insoluble nitrogen.

Trowbridge and Grindley take a sample previously ground in a
meat chopper, and immerse it for one hour in ice w^ater, in the pro-
portion of 1000 grms. of meat to 1500 c.c. of water. This solution is
then filtered through a cheese cloth, at the same time assisting the
process by squeezing the cloth with the hand. The residue thus ob-
tained is divided into smaller portions, transferred to beakers washed
in series, fresh water being used w^ith No. 1 only, filtering through
cheese cloth from one beaker to another until the last filtrate is
colourless, neutral to phenol-phthalein, and gives no reaction for pro-
teins by the biuret test. The mixed filtrates and washings easily
filter through paper giving a clear red filtrate, in which soluble
nitrogen can be determined.

Pennington employs the following process with the meat of
chickens : Place a portion of the finely divided red or white meat,
60 grms. in weight, into a tall slender bottle of 500 c.c. capacity, made
to fit a centrifuge which can hold 1 litre of material ; add 300 c.c.
of water, and shake the flask gently for fifteen minutes. This move-
ment is only sufficient to keep the particles of meat in motion and
the composition of the extract homogeneous. An emulsion is formed
when the shaking is violent and when the tissue is very finely ground.
Having shaken for the specified length of time, rotate the flask in a
centrifuge for twenty minutes, thus causing the heavier particles
to settle in a compact mass, and allowing the decantation of the
liquid floating on the top, which should be then filtered through
paper. Kepeat the extraction as outlined, with portions of 800 c.c.
of water until the filtrate is practically free from protein as indicated
by the biuret reaction. A volume of 1500 to 2500 c.c. is generally
necessary to obtain this result. Add thymol to both the flesh and
the extract to prevent bacterial decomposition, and keep cold, using
ice if necessary to keep the meat immune from the naturally occur-
ring enzymes.

The extraction of the white meat is a much simpler process than
the extraction of the dark meat. The latter does not settle so com-
pactly after rotating in the centrifuge, it is slower in filtering and
continues to show a distinct biuret reaction for a long time after the
white meat is freed from water-soluble proteins. Certain fowls in
fact, especially those kept in cold storage for a considerable time,
never show a dark meat completely free from water-soluble nitrogen.
In these cases, the question of the error owing to long handling and
enzyme action, causing an increase in the actual quantity, has to be
taken into consideration. It has been noticed after experiment that

390 FOOD AND DRUGS.

when there has been a long extraction of such tissue, a point is reached
when a very faint biuret reaction appears indefinitely and does not
seem to diminish. These extractions are continued for about twenty-
six hours, as it is probable that a greater error would arise in the gain
of >'.hat has been originally insoluble material, than the loss of the
originally-formed water-soluble nitrogen. The total extract of the
muscle is made up to a definite volume and made neutral to litmus
paper with tenth normal sodium hydroxide.

Cook's method is to weigh 200 grms. in a 450 Erlenmeyer flask,
to add 250 c.c. of water and agitate for three hours in a shaking
machine. The material is then filtered through linen bags, vigorously
and repeatedly immersed with the hands in successive portions of
water, pressing out after each extraction until negative biuret reaction
results ; 2200 to 2500 c.c. of water are generally necessary for this
operation, and a small quantity of phenol or thymol should be added.

Weber employs Cook's method at room temperature and with ice
water when examining samples of fresh and storage meat, also samples
which he had kept for varying lengths of time in his laboratory. A
larger amount of soluble proteins resulted when he worked at room
temperature. It has not been stated whether this was due to the
greater extracting power of water at room temperature, or to greater
enzymic action whilst the extracting process was being carried out.

Determination of Collagen. — Place the insoluble proteins, obtained
by the above-mentioned directions, in a beaker, add water and heat to
boiling for some minutes.

Separate by filtration, wash with boiling w^ater. Deduct the nitro-
gen of the residue insoluble in boiling water from the nitrogen insol-
uble in cold water, and multiply by 5*55 for the percentage of collagen.
There are drawbacks to this method on account of the difficulty ex-
perienced in rendering collagen soluble and the tendency towards
decomposition of the protein.

Determination of Coagulahle Proteins. — Heat the entire filtrate (or
an aliquot portion from the determination of nitrogenous bodies insol-
uble in water) sufficiently to coagulate the coagulahle proteins, filter,
wash the insoluble material with hot water, and transfer the filter and
contents to a Kjeldahl flask, and determine the nitrogen by Gunning's
method, multiply the percentage of nitrogen by 6*25, which gives the
percentage of coagulahle proteins.

The amount of heating required to obtain maximum coagulation
varies with different materials. The Association of Official Agri-
cultural Chemists of the United States directs that the solution should
be almost neutralized, but left still slightly acid, and boiled until the
globulins are coagulated.

Pennington, experimenting with chickens, evaporates 350 c.c. to a
volume of about 100 c.c. before filtering. Grindley and Emmett use
200 c.c. of the solution, add alkali till neutral to litmus paper, and
evaporate to 50 c.c. Trowbridge and Grindley, in a later paper, re-
port maximum results from the cold water extract of fresh beef by
neutralizing one-fourth of the acidity to phenol-phthalein before co-
agulation.

FLESH FOODS. 391

Determination of Proteoses, Pejjtones and Meat Bases. — Dilute the
filtrate from coagulated proteins with wash water, concentrate by-
evaporation, and make up to 100 c.c. The proteoses can be deter-
mined by saturating an aliquot portion of the filtrate with zinc
sulphate which precipitates the proteoses. The nitrogen found in the
precipitate should be multiplied by 6-25, and the meat bases determined
by Sjerning's method, as modified by Bigelow and Cook. (See under
Meat Extract). To determine peptones deduct from the total nitrogen,
the sum of the nitrogen occurring in insoluble nitrogenous bodies,
coagulable proteins, meat bases and proteoses.

Determinatioti of Gelatin (modified Stutzer's method). — Thoroughly
extract say 10 grms. of the sample with boiling water, then place the
extract in a porcelain dish containing about 20 grms. of previously
ignited sand and evaporate to dryness. Stir the residue with four
successive portions of absolute alcohol using about 50 c.c. each time,
and pouring it off through a filter made up of a layer of asbestos fibre
on a perforated porcelain plate inside a funnel. Pack the funnel
round with chopped ice and arrange it so that gentle suction may be
used to help on the filtration. Eepeatedly stir the residue with suc-
cessive portions of about 100 c.c. each of a mixture containing 100
c.c. of 95 per cent alcohol, 300 grms. of ice and 600 grms. of cold
water, passing each portion through the asbestos filter. Continue
the washing until the solution issuing from the filter is colourless,
always keeping the temperature below 5°. Transfer the asbestos with
the washed residue to a beaker and thoroughly extract the whole
with boiling water. Evaporate the hot-water extract to a small
volume, wash into a Kjeldahl flask, in which evaporate to dryness
and determine the nitrogen by the Gunning method; Nx5-55 =
gelatin.

Detection of Parasites.— The two principal parasites which are
found in sausages are Trichina spiralis (fig. 39b) and Cysticercus
cellulosce (fig. 39a), the latter being the cause of " measles " in pork.
The importance of the absence of these parasites is obvious, since the
former is responsible for the disease known as trichinosis, whilst the
latter is the larva of Taenia colium, a common tape-worm, whose
principal host is man. Other forms of Cysticercus are found, which
are the larvae of other tape-worms. For their detection, the fat should
be removed by a mixture of two parts ether and 1 part alcohol.
Schmidt treats the residue (from which pieces obviously not meat may
be removed by a needle) with ten times its weight of water containing
0-5 per cent of HCl, and a little pepsin. The mixture is allowed to
stand for six hours at 40°. The flesh is thus dissolved, the fat floats
on the surface, and the parasites sink to the bottom of the liquid. If
the digestion be performed in a separator, the deposited parasites can
be run off' in a few drops of liquid and examined under the microscope.
The Trichina are easily recognized as thread-like worms coiled in flat
spirals, whilst the Cysticerci have tape-worm heads and bladder-like
tails. The parasites will survive nearly any treatment, except exposure
to boiling water temperature.

Detection of Horse-flesh. — Horse-flesh is a common constituent of

392

FOOD AND DEUGS.

continental sausages, some of which find their way to this country.
In England a heavy penalty attaches to the sale of horse-flesh without
declaring it, so that it is very rarely to be found in English sausages.
The detection of horse-flesh, especially when in the minced state, in ad-
mixture with other meats is a matter of considerable difiiculty, and is
often impossible. Much stress has been laid on the presence of a
considerable amount of glycogen in horse-flesh, but the methods of de-

PiG. 39a. — Cysbicercus cellulosae.
Free (A) with head withdrawn x 10 ;
(B) with head protruding x 10 ; and
imbedded in muscular tissue x 5.

Fig. 39b. — Trichina spiralis.
Free x 100 : imbedded x 50.

tecting and determining this body are of sufficient uncertamty to
render them unreliable except in certain well-defined cases.

Glycogen (C^Hi^jO^),, was discovered by Bernard in 1857 and has
been termed " animal starch ". It is found in the livers of many animals
in which it is probably stored as a reserve material in times of fasting.
It is hydrolysed by ferments into maltose, and by dilute acids direct
into glucose. It is coloured red with iodine, a reaction differentiating
it from the ordinary starches.

Many processes have beeti published for the determination of
glycogen, but before describing any of these, the following figures due
to Bujard (" Forsch. Bericht," 1897, iv. 47) should be examined in

FLESH FOODS.

393

order to indicate the danger of drawing any strong inferences from
the results of glycogen determinations : —

Water.

Per cent Glycogen.

Per cent Glycogen on dried substance.

Per cent

Horse-flesh

74-44

0-440

1-72

74-87

0-600

2-39

76-17

1-827

7-69

7600

0-592

2-475

61-83

0-346

2-24

72-90

0-174

0-64

70-47

1-366

4-62

71-84

0-59

2-09

„ (smoked)

43-00

0-108

0-19

Beef

73-62

0-206

0-74

75-55

0-018

0-073

Veal

76-12

0-346

1-44

74-47

0-066

0-25

Pork

54-05

trace

61-29

Pork sausage

67-25

.0-240

0-73

Horse sausage

70-04

0-504

1-68

„ liver sausage

67-00

1-762

5-34

From these results it appears that the presence of glycogen can-
not be considered as definitely indicating the presence of horse-flesh,
but that the presence of quantities much over 1 per cent (on the
dried substance) would indicate its presence if corroborated by other
results, unless much liver were present. If more than 2 per cent (on
the dried substance) be present, the presence of horse-flesh is very pro-
bable.

Braiitigam and Edelmann (" Chem. Central," 1894, 1, 485) give
the following quantitive test for the detection of glycogen : 50 grms.
of the finely divided flesh are boiled for an hour with four times its
volume of water, and dilute nitric acid added to the strained liquid
after cooling. Proteids are precipitated and the liquid is partially de-
colorized. The filtrate is then tested by gently pouring a saturated
aqueous solution of iodine on to its surface. In the presence of gly-
cogen a wine-red ring is formed at the point of contact. If the colour
is not decided, the flesh may be heated on the water bath with 3 per
cent of its weight of KOH dissolved in water, until the muscular
tissue is dissolved. The strained liquid is evaporated to half its
volume, the proteids precipitated by HNO3 and the iodine solution
added as described above.

Piettre (" Anal. Chem. Anal." 1909, ii, 206) estimates glycogen
by boiling 25 grms. of the sausage under a reflux condenser with 80
c.c. to 90 c.c. of an alcoholic solution of KOH (aqueous solution of
specific gravity 1*3 diluted with four times its volume of absolute
alcohol). The insoluble residue is collected on a filter, washed with
hot 80 per cent alcohol, and then with cold alcohol rendered slightly
-acid with HCl, until all the alkali is removed. The residue is then

394 FOOD AND DKUGS.

heated with sHghtly alkaline water which dissolves both starch and
glycogen. An equal volume of water is then added, thus precipitating
the starch. This is filtered off and washed with 50 per cent alcohol ;
the filtrate is concentrated to a small volume and absolute alcohol
added to precipitate the glycogen, which is collected, washed with
alcohol, dried and weighed.

Considering the uncertainty attaching to this reaction, further de-
tails of other but similar processes are unnecessary.

Niehl gives the following method for the quantitative determina-
tion of glycogen : —

The flesh is heated on the water bath for six hours to eight hours
with 3 per cent to 4 per cent of its weight of KOH, and four times
its volume of water. The liquid thus obtained is evaporated to half
its bulk, and HCl, and a solution of mercuric-potassium iodine, added
to the liquid when cold, in order to precipitate nitrogenous matter.

The clear filtrate is mixed with 2-5 times its volume of 90 per cent
alcohol, and the precipitated glycogen collected on a filter, washed
successively with 60 per cent, 90 per cent, and absolute alcohol, and
then with ether, and finally with absolute alcohol, dried at 110° and
weighed.

Mayrhofer's method consists of dissolving the flesh in aqueous
solution of KOH, precipitating proteids by adding HCl and Nessler's
reagent, and then precipitating the glycogen with alcohol, and washing
it on a tared filter with alcohol and ether and then drying and
weighing.

The differences between the fat from the horse and that from other
animals have been discussed above (p. 387).

Perhaps the best indications are those given by the examination of
the intra-muscular fat and its liquid fatty acids as to the amount of
iodine they absorb. Bremer (" Forsch. Ber." 1897, iv. 1) recom-
mends the following process : All visible fat is mechanically removed,
and the remaining meat, finely divided, is heated for an hour on the
water bath with water. The fat rising to the surface is poured off
with the water, and the flesh after several washings with hot water is
dried at 100° for twelve hours and extracted with petroleum ether.
The fat so obtained is saponified, excess of alkali neutralized with
acetic acid and the alcohol evaporated on the water bath. The soap
is dissolved in hot water, and hot solution of zinc acetate added.
The precipitated zinc soap is washed with hot water and alcohol,
dried, and extracted with ether in a Soxhlet. The ether is shaken
with dilute sulphuric acid, to decompose the zinc salts of the liquid
fatty acids, and then washed three times with water. The ether is
then evaporated and the liquid fatty acids are dried at 100°. The
iodine value of the fat itself and of the so separated liquid fatty acids,
is then determined. On opposite page are tabulated Bremer's results.

When horse-flesh is present the petroleum ether extract has a
reddish-brown colour, and the fatty acids also have a slight reddish
colour. Bull's flesh gives similar colours, but if this reaction is
observed, and the glycogen exceeds 1-5 per cent on the dry substance
— or even 1 per cent — and the iodine value of the intra-muscular fat

FLESH FOODS.

395.

exceeds 65 and that of the Hquid fatty acids exceeds 95, there is
practically no doubt that horse-flesh is present.

Iodine Values of

Intra-muscular Fat.

Liquid Fatty Acids.

Horse-flesh sausage

„ „ with about 6 per cent

bacon

Horse-flesh (brain) and 22 per cent bacon

sausage ......

„ „ „ 25 per cent bacon
Pork (T-huringian) cervelat with 65 per cent

pig's fat . '

Per cent
75-8

74-0

53-7
741

64-3

Per cent
108-1

104-1

92-4
1021

95-8

For further details as to the presence of horse-flesh in sausages;
the following papers may be consulted : —

Pfluger and Nerking ("Arch. Ges. Physiol. " 1899, 76, 531).

Mayrhofer (Forsch. Ber." 1897, iv. 47).

Schiitze ("Deutsche Med. Wochs." 1902, 46, 804).

Colouring Matter in Sausages. — Sausages are very commonly
mixed with colouring matter, either with the intention of improving
the colour or of concealing a large addition of farinaceous matter^
The following details as to the detection of colouring matters apply
to preserved foods generally, as well as to sausages : —

The colouring matters usually added are (1) cochineal, (2) aniline'
colours, (3) iron oxide — often added as Armenian Bole, a form of iron
oxide diluted with chalk. There are certain cases where trade usage
certainly justifies a small addition of colouring matter. One of these
is the colouring of anchovy essences and pastes with a trifling amount
of oxide of iron. The preparation is of a colour not acceptable to the
public taste, and a little oxide of iron renders it fanmore inviting, and
has been used for many years, and is in every way unobjectionable.

The principal objection to the use of colouring matters in such
food stuffs, is that its purpose is to cover the employment of unsound
meat, which may be of bad colour. The following methods will reveal
the presence of added colouring matters.

The usual colouring matters added to sausages are either some-
form of oxide of iron or an aniline red. In most cases a small amount
of colouring matter is not objectionable, but it is usually necessary to>
examine the sample in order to decide whether any excess of colour-
ing matter has been added.

Any oxide of iron colour is at once revealed by the examination of
the ash, which should at most contain but a trace of iron — say up
to 2 per cent of the total ash. Sausages coloured, for example, with
Armenian Bole will have a high ash value, and the ash will contain,
much iron.

396 FOOD AND DEUGS.

Cochineal is sometimes added to sausages. It may be detected by
the method described by Klinger and Bujard (" Zeit. Angew. Chem."
1891, 515). The sample, in a fine state of division, is heated with
twice its volume of a mixture of equal parts of glycerin and water for
three hours on the water bath, the whole being slightly acidified.
The yellow solution is poured on to a wet filter, and the colouring
matter, if present, is precipitated as a lake by adding alum and
ammonia. The precipitate is filtered off and washed, and then dis-
solved in a small amount of tartaric acid, and the concentrated solution
is then examined by the spectroscope against a standard solution of
cochineal carmine, when the absorption bands, which should be identi-
cal, should be seen, as well-marked bands, lying between h and D.

Bremer ("Analyst," xxii. 216) has confirmed the utility of this
method.

Spaeth ('* Pharm. Central." 1897, 38, 884) finds that the artificial
colouring matters usually added to sausages can be extracted by
warming the finely divided matter with a 5 per cent solution of sodium
salicylate on a water bath for a short time. On adding ammonia to
the extract, red precipitates are often thrown down, which contain the
colouring matter.

Many colouring matters may be extracted by alcohol slightly
acidified with hydrochloric acid. A small fragment of white wool is
boiled in the liquid, and if it is distinctly dyed, a coal-tar colour
is certainly present.

Marpmann (" Zeit. Angew. Mikrosk." 1895, 12) considers that a
microscopic examination will reveal the presence of most colouring
matters. A section about 1 cm. thick of the sausage is made, and
thoroughly moistened with 50 per cent alcohol, and then examined
under the microscope. The cell tissue or contents are dyed by most
artificial colouring matters, and such dyed cells indicate the presence of
added colouring matter. When traces only of a colouring matter have
been added, the section may be treated first with xylene, then with
carbon tetrachloride, and finally immersed in cedar wood oil and
examined under the microscope.

EXTRACT OF MEAT.

Numerous meat preparations exist at the present time, which are
prepared in different manners, and which rarely justify the extravagant
claims made for them in regard to their nutritive value. One such,
claimed in advertisements to be the 7nost nutritious of all beef bever-
ages, was found by the author to contain over 70 per cent of mineral
matter, principally salt, and no true proteids.

The analyst is called upon frequently to judge the quality of such
preparations, but no legal standard can be said to exist for meat
extract, hence the rarity with which this class of preparation is dealt
with under the Food and Drugs Acts.

Direct adulteration of extract of meat is not common, but the
author has had several cases before him in which samples sold as
genuine extract of meat contained extract of yeast, an extract which

EXTRACT OF MEAT. 397

is now made to closely simulate extract of meat in general characters.

Numerous food products are also on the market under fancy
names such as would often lead a person of ordinary intelligence to
believe he was dealing with a pure meat preparation, which are in
fact little else than extract of yeast containing a small amount of meat
extract and various covering flavourings.

The extract of meat with which one is principally concerned
analytically is that known as Liebig's Extract of Meat. The name
Liebig is not a proprietary one and is open to any one's use, nor is
meat extract now made by Liebig's original process.

The best meat extracts to-day consist principally of the portions
of the meat, freed from bone and most of the fat, which are soluble in
water at a temperature not exceeding 75° C. When the water is used
at 100", the gelatine extracted will be higher than in the former case.
When ivarm water is used, the gelatine is low, but albumoses and
peptones and the meat bases are present to the full extent, and
albumin to a greater or less extent.

It is now generally recognized that extract of meat is rather a food
adjunct and a stimulant than a food in the proper sense of the word.

A large number of the analyses quoted below have been made by
the author, but for much of the information he is indebted to Allen
(" Commercial Organic Analysis," 3rd edition, Vol. IV, pp. 300 et
seq.).

Many of the extracts of beef of the present day contain added
gelatine, meat fibre and peptones. Few if any are made according
to the original or modified directions of Liebig. Without wishing
to deroo;ate from the admitted value of these preparations, the author
cannot help agreeing with the late A. H. Allen in his statement :
" It is claimed on behalf of these preparations that the various
additions and methods of treatment give them value as real foods,
but this is true in but a very limited sense, since the amount of
such preparations which would require to be taken to support life
is enormously beyond the quantity of any of the preparations which
could be consumed without upsetting the system, to say nothing of
the extravagant cost of all such preparations if used in quantity
necessary to sustain life. In judging of the amount of credence
to be attached to statements of the nutritive value and concentration
of meat extracts and similar preparations, it should be borne in
mind that fresh lean meat contains about 20 per cent of nutritive
matter and 75 per cent of water. Hence by the desiccation of 4 lb.,
of meat there will be obtained 1 lb. of dry substance of which 80 per
cent is nutritive proteid matter, the remaining 20 per cent consisting
of fat, meat bases, salts, etc. By no possible means can further
material concentration of the nutritive matter be effected."

At the same time there are several high class preparations of meat
extract to which certain additions have been made, which give to them
a true food value, so that such preparations are both stimulants and
foods. No standards, other than a requirement of purity, exist in
this country for meat extracts or essences, but the following require-
ments, which have been adopted by the American Association of

:398 FOOD AND DKUGS.

Ofificial Agricultural Chemists, are of considerable interest in showing
what is expected of a normal preparation in the United States : —

(1) Meat extract is the product obtained by extracting fresh meat
with boiling water and concentrating the liquid by evaporation, after
removal of the fat. It contains at least 75 per cent of total solid
matter, of which not more than 27 per cent is ash, and not over 12
per cent sodium chloride. The fat should not exceed 0*6 per cent and
the nitrogen be not less than 8 per cent. The nitrogenous compounds
contain not less than 40 per cent of meat bases and not less than 10
per cent of creatine and creatinine.

(2) Fluid extract of meat differs only from the above in containing
not less than 50 per cent of solid matter, and not more than 75 per
-cent. The proportionate amounts of the other ingredients, after allow-
ing for the extra water, are the same.

(3) Meat juice is the fluid portion of muscle fibre, obtained by
pressure or otherwise, and may be concentrated by evaporation
tit a temperature below the coagulating point of the soluble proteins.
The solids contain not more than 15 per cent of ash, and not more
than 2 '5 per cent of sodium chloride ; and between 2 and 4 per cent
of P2O5 ; and not less than 12 per cent of nitrogen. The nitrogenous
bodies contain not less than 35 per cent of coagulable proteins and
not more than 40 per cent of meat bases.

In judging the value of a meat extract the following are the chief
•considerations which should be taken into account : —

(1) The amount of water present, (2) the amount of mineral salts,
(3) the amount of meat bases, (4) the nature and amount of the pro-
teid matters and other nitrogenous bodies present.

The following tables, for which acknowledgment is made to Mr.
Otto Hehner, show the average composition of a number of meat ex-
tracts and similar preparations : —

EXTRACT OF MEAT.

399

•uagoj^i^ IB^OX

53 S O N 00 .-1

Ph S 6i 00 05 Ci

(M«OOaOC5<M-XiOrH

(T^co-icorHobibciO

5-02

•ppv
ouoqdsoqd:

Ph o <b <ib lb o

»0rH00t:~O":>'<*<lCJ0

oooiocp-^ocraep-^
cicocooo-^cbcorH

•apuomo
ranipog

Ph g »o OS « >o

kOCOOSrHC0C-(MC(?CO
COqSOrHCOO'^WTji

(?q«b0«)06srHlb<fl

•9DuaJ8i)ia

.9 .9 . ^

(2^ g (N tH GO ^ «p ^O ^t^-CpOopcfOCS-H « O -f, rg

•qsy

^ rH «0 O >0

^ ^ lO Cp 00 -^

p^ g so 05 cb o

(M 05 r-l C^

rH
rH

rHOOCOlOOC^OOO
Ot>-CS«pO«p05>pcp
Cd'^t^-^JrHt-OSlisb
rH rH rH r-t j-I i-t ^

•sas^a ;b9h

, ^ (M (M O 00

§3 -g «i ^ o »p

^1 «^ as -H do 00

" CO -* CO CO

o
(?q

oO'*iO)cqcooo<Mt'a>
■^TPC^oO'-^eprHOao

C505'*(NcbC5t-'^rH
1-t Cq rH rH CO CO

•S9UO|d9(J

^ ^ «0 CO «o "^
Ph g 00 »C O 00

t-soosc-t-aoio-^c^i
cpcpoco»c>rHoq'*cri
(fqArHoocoebsboo

•sasoranqxY

+f rH IC CS (M
S g O t- rH «p
P-l « (J? fH tH CO

(T^rHrHcbodbwOrH

•niranq
-IV p^:^BIn■SBo^

puB 9Jqi^-;B9H

Per
cent
2-12

1-81
1-30

O !>• C^ rH t- lO

1 1 O CO 1 CO CO CO CI

•nituuqiY

Per
cent

0-25
5-62

16-44
4-43

05
C5

•aai^BpQ

fe 1 ? ? ? ?

P-l " kC CO ■* W5

0(MC-»0(MrH«Ci?0«0
t-THCOrJiT-HGOOipwS
OrHrHO»bcb-H-^(?q

^ g CO (?q CO cfl
P^ « o o o o

000»0(MCO(N(Mt-rH
r-IOCClC0OO«pO»0)
OOOOO-HOrHO

rH
O

•a^:^BAV

^ "g CT O^ op (N

P-i g »b o t>- (fi

iH 1— I rH (M

corHosasoo-<*iO'*t-

»O«O.HrHCpC0t--C0-^

ibrHsbocsdb'^'^t-

»O)5C>C0t-Q0(M-<*(MrH

«o

§
s

•1

03
>

Extract of meat.
>> >>

<v
.2

•2.

"c8

Bouillon .
Meat juice .
Essence of beef .
Fluid beef .

A proprietary brand
„ (invalid) ,

.2.

"c3

•J9qmn^

iH cq CO -*

ITS

COt-OOCSOrHIMCOTt*

,-i ,-i T-1 ^^ ,-1

U5
rH

«o

»H

The following are the amounts of nitrogen existing in different
forms in the same samples : —

400

FOOD AND DKUGS.

Description.

Nitrogen existing as—

2
1

0-83

0-34

0-32

1-29

6-29

9-07

0-53

0-28

0-82

6-58

8-21

0-73

0-29

0-67

1-69

6-42

9-80

0-88

0-21

0-58

1-35

6-17

9-19

0-11

0-16

017

0-40

2-00

2-84

0-12

0-04

0-32

0-46

1-98

2-92

0-18

0-90

0-17

0-30

1-51

3-06

0-22

0-64

019

1-77

3-88

6-70

0-07

2-(53

0-06

0-01

0-06

0-43

3-28

0-82

0-03

0-09

0-65

1-49

0-61

0-86

134

2-11

3-10

8-02

0-17

117

0-38

1-00

2-74

5-46

0-73

0 94

0-89

1-03

5-61

9-20

0-41

071

2-44

0-17

1-41

5-07

10-21

0-04

0-35

0-15

0-58

0-16

1-81

3-09

0-27

0-98

—

0-28

0-77

2-72

502

Nos. 1 to 4 represent concentrated beef extracts ; 5 to 10 repre-
sent specimens of the meat juice type ; and 11 to 16 meat prepara-
tions containing added matter such as meat fibre, seasoning, etc.

Bigelow and Cook give the following analyses (" U. S. Dept. of
Agriculture, Bull. 114," p. 19) of meat jidces prepared by themselves,
from which comparison with meat extracts may be made : —

Per cent

Water

85-76

86-85

90-65

91-90

Ash .

1-53

1-86

1-36

1-29

NaCl .

0-12

0-20

0-15

0-19

PaOg .

0-37

0-31

0-36

0-29

Fat .

0-27

0-30

0-19

0-64

Nitrogen (total) .

2-08

1-74

1-16

1-09

„

(insoluble protein)

0-16

0-29

1 0-68

0-12

(coagulable „ )

1-37

0-98

0-41

(proteose) .

0-06

0-07

0-04

0-07

„

(peptone) .

0-16

0-11

0-01

0-21

(amido)

0-33

0-29

0-43

0-27

A few preliminary remarks on some of the nitrogenous matters
present in extract of meat are necessary.

The meat bases are amongst the most important of the consti-

EXTKACT OF MEAT.

401

tuents of meat extract, being largely responsible for its stimulating
value.

The following bases have been stated to be present in extract of
meat, but only a few of them can be said to have been definitely
identified : —

Nitrogen Factor.

Creatinine .

C4H7N3O

2-69

Creatine

C4H3NA

3-12

Neosine

CeHj^NO

8-43

Carnitine .

C,H,«NO,

11-57

Vitiatine .

C5H14NB

1-88

Histidine .

C,U,K,0,

3-69

Methyl guanidine

C.HvN,

1-74

Adenine

C,H,NO,

6-36

Xanthine .

CsH.N.O^

2-71

Xanthocreatinine

C5H10N4O

2-54

Hypoxanthine

C,H^Np

2-44

Carnine

C^H.N.O,

3-50

Leucine

CfiH,3N0,

9-n6

Tyrosine

CpH^NO,

12-93

•

No satisfactory method exists for their determination, and any re-
sults are useless unless accompanied by a statement of the method
adopted. Their amount is frequently deducted with some degree of ap-
proximation by attributing to them all the nitrogen existing over and
above that found to exist in other forms. But an examination of the
above figures shows the impossibility of fixing an average nitrogen
factor, by which to multiply the amount of nitrogen found in order to
convert it into meat bases. Hehner prefers the usual albumen factor
6-25. Stutzer adopts the factor 3-12 which is that for creatine. There is
no known means of ascertaining the proper factor, therefore it ap-
pears to be the easiest plan to return the " meat bases " in the form of
an equivalent of nitrogen. An approximate method for the deter-
mination of creatinine, however, exists, which will be discussed later.

The Analijfiis of Meat Extract. — The analysis of meat extract is
admittedly somewhat unsatisfactory, but the following scheme will
afford the most useful information available.

Water. — The water is not easily driven off from the extract by the
ordinary method. From I'o grms. to 2 grms. are weighed into a
flat-bottomed platinum dish and dissolved in a little distilled water, and
a weighed quantity of recently ignited sand added. The pasty mass
is dried in a water oven to constant weight, or more rapidly in an air
bath at 105°.

Mmeral Matter. — Since the addition of ordinary salt is common
in meat extracts, an ash burnt to whiteness or anything approaching
whiteness has necessarily lost some of its chlorides. The well-
burned ash, therefore, of meat extract is usually below the truth and
represents the mineral matter less an uncertain amount of chlorides.
The chlorides, if it be considered necessary to determine them, should
VOL. L 26

402 FOOD AND DRUGS.

be estimated by thoroughly extracting the half-burned ash with dis-
tilled water. It is customary to return the chlorides as sodium
chloride, but as a matter of fact, the natural chlorides of the meat ex-
tract are principally potassium chloride. Meat extract in its natural
condition contains about 0-06 per cent of chlorides calculated as
NaCl for every 1 per cent of dry solid matter it contains. Any excess
over this amount is to be regarded as added salt.

Total Nitrogen. — The nitrogen should be estimated by treating
about 1 grm. of extract (i.e. the ordinary Liebig's extract ; up to 5 grms.
of liquid preparations may be used) by Kjeldahl's process.

The Sejjaration of the Nitrogen. — It then becomes necessary to de-
termine the nitrogen existing in the different types of nitrogenous
compounds present in the extract.

It will here be convenient to describe the determination of nitrogen
by moist combustion. To determine nitrogen by this method, decom-
pose the organic matter by digesting with sulphuric acid and an oxi-
dizer, thus driving off the carbon and hydrogen as carbon dioxide and
water respectively, and converting the nitrogen into an ammonium
salt from which free ammonia NH3 is liberated later by making alka-
line. Distil the ammonia into an acid solution, the value of which is
known, and calculate by titrating the excess of acid. The decomposi-
tion, in the Kjeldahl process, is brought about by means of a mercury
compound, whilst in the Gunning method it is effected by potassium
sulphate which forms bisulphate with the acid.

If nitrates are present and neither method in its simplest form is
practicable, it is necessary to use a modification.

For the determination of nitrogen in pepper the Gunning-Arnold
method is used, as it is impossible to completely decompose the
piperin by the usual processes.

The Kjeldahl-Gunning Method. — Reagents : —

Standard alkali solution N/lONaOH.

Pulverized potassium sulphate.

Sulphuric acid, concentrated.

Sodium hydroxide, saturated solution.

Standard acid solution, N/IOH2SO4 or HCl.

An indicator, cochineal.

Granulated zinc.

Take a pear-shaped flask with flat or round bottom, and made of
fairly thick Jena glass, and digest and distil preferably in the same
flask. The following dimensions are suitable: length 29 cm., maxi-
mum diameter 10 cm., tapering gradually to a long neck which is 28
mm. in diameter with a flaring edge. Its capacity should be about
550 c.c.

If preferred, a smaller flask of about 250 c.c. and of the same,
shape as the one already described may be used for the digestion
and an ordinary round-bottomed flask of 500 c.c. capacity for the dis-
tillation. Transfer 0-5 grm. to 3-5 grms. of the sample to the digestion
flask, add 10 grms. of potassium sulphate and from 15 c.c. to 25 c.c. of
concentrated sulphuric acid.

Hold the flask over a flame, gently heating for a few minutes

EXTEACT OF MEAT. 403

"telow the boiling-point of the acid until the frothing ceases, then
gradually increase the heat until the acid boils ; continue the boiling
until the contents are either a pale straw colour or quite colour-
less. Place a wire gauze between the flask and flame, or better still
a triangle or some similar support.

Cool the contents of the flask, and if the digestion has been
brought about in the larger flask suitable also for distilling, as men-
tioned above, add cautiously 300 c.c. of water and sufficient strong
sodium hydroxide to make the contents strongly alkaline, using phenol-
phthalein as an indicator. "When, however, a separate flask is used
for distillation add the contents of the digestion flask to the water and
the alkali. Add also a few pieces of granulated zinc to prevent, by
the evolution of the gas, any bumping and sucking back of the distil-
late. Shake the flask well and connect with the condenser the bottom
of which is provided with an adapter dipping below the surface of the
standard hydrochloric or sulphuric acid, a measured quantity of which
should be contained in the receiving flask. Continue the distillation
until all the ammonia has passed over into the acid, which operation
should take from about forty-five minutes to an hour and a half.
Usually the first 250 c.c. of the distillate contains all the ammonia.

Titrate with standard alkali the excess of acid in the receiving
flask, and calculate the amount of nitrogen absorbed as ammonia.
Unless the reagents are known to be absolutely pure and free from
nitrates and ammonium salts they should be tested by means of a blank
experiment with sugar, thus reducing any nitrates present. Allow-
ance should be made for any nitrogen due to impurities.

It is necessary when purchasing sulphuric acid for the determina-
tion of nitrogen to obtain that which is " nitrogen-free," as often the
so-called chemically pure acid contains a large amount of nitrogen.

Modifiaction of Gunning s Method to Include Nitrates. — Sodium
thiosulphate and salicylic acid are used in addition to the reagents
employed in the simpler Gunning method. These should be mixed
in the proportion of 30 c.c. of concentrated sulphuric acid to 1 grm.
of salicylic acid. Add from 30 c.c. to 35 c.c. of the mixture to 0*5
grm. to 3'5 grms. of the substance in the digestion flask. Agitate the
flask well and allow it to stand for a few minutes, shaking occasion-
ally. Next add 5 grms. of sodium thiosulphate, then 10 grms. of
potassium sulphate. Heat very gently at first, gradually increasing
until the frothing has ceased. Continue to heat until the contents
have boiled and are colourless. Then proceed as in the Gunning
method.

The Kjeldahl Method. — Transfer 1 grm. of the air-dry substance or
a corresponding larger amount of a moist or liquid substance, and 0'7
grm. of mercuric oxide (or a" similar amount of metallic mercury) to
a 550 c.c. Jena flask. Add 20 c.c. of sulphuric acid. Incline the
flask over a Bunsen burner, and heat the mixture below boiling-point
for five minutes to fifteen minutes or until the frothing ceases, then
increase the heat until the mixture boils quickly. Continue the boiling
until the liquid has become almost colourless and for half an hour
afterwards. Turn the lamp out, place the flask in an upright position.

404 FOOD AND DKUGS.

slowly add potassium permanganate, shaking until the solution be-
comes a permanent green or purple colour. Cool, then add sufficient
saturated sodium hydroxide solution to render the solution alkaline,
and lastly a few grains of granulated zinc, shaking the flask well after
each addition. Immediately connect w4th the distillation apparatus
and proceed as in the Gunning method.

One of the most commonly employed processes, especially in
works using large quantities of extract of meat, is the following which,
however, is admittedly a somewhat empirical process. It consists in
making a rough differentiation between the greater part of the pro-
teid and gelatinous matter on the one hand, and the meat extractives
and salts on the other. Hehner (" Analyst," x. 221) recommends that
2 grms. of the sample should be dissolved in 25 c.c. of water and 50 c.c.
of alcohol be added. The precipitate, consisting principally of gela-
tinoid and proteid matter, is allowed to settle overnight and the clear
liquid decanted in the morning. The precipitate with its adherent
liquid is dried in a small basin and weighed. The alcohol precipitate
thus obtained is usually about 5 per cent to 6 per cent. Much higher
results would indicate added gelatine.

Allen recommends the precipitation from an aqueous solution of
the meat extract of the proteid and gelatinoid bodies by means of zinc
sulphate ; then precipitating peptones and similar bodies by means of
bromine, leaving the meat bases in solution. For the best available
separation of the various nitrogenous constituents he recommends the
following scheme of analysis : —

In the fullest possible analysis of a meat extract, an attempt will be
made to discriminate between and determine the amount of nitrogen
existing in the various forms of meat fibre and insoluble albumin, coagul-
able albumin, acid-albumin, albumoses, peptones, coagulable gelatin,
gelatin-peptones, meat bases, amido-compounds, and ammonia. Such
an analysis is necessarily tedious and rarely necessary, but some of the
more important of the above determinations can be affected with
reasonable ease and accuracy, and are not uncommonly required of the
analyst.

In consequence of the uncertainty attaching to the composition
of certain of the nitrogenized constituents of meat extracts, it is often
convenient to state simply the amounts of nitrogen found to exist in
the various forms, and in cases where it is preferred to state the
actual amounts of the nitrogenized bodies present, the corresponding
amounts of nitrogen should always be given in addition.

Ammoniacal Nitrogen should be determined by distilling the aqueous
solution of a known weight of the preparation with barium carbonate,
which is preferable to magnesia.

Unaltered Proteids and Meat Fibre. — Bovril and certain allied
high class preparations contain finely powdered meat-fibre. This may
be detected by treating the meat extract with cold water, and ex-
amining the insoluble portion under the microscope. If meat fibre be
found, 5 grms. of a dry preparation, 8 grms. to 10 grms of an extract
or 20 grms. to 25 grms. of a fluid preparation should be treated with
cold water, the insoluble matter collected on a filter, washed with

EXTKACT OF MEAT. 405

cold water, dried at 100° C, and weighed. The weight obtained re-
presents the meat fibre and insoluble matter of the preparation. An
alternative and in some respects preferable plan is to treat the moist
residue by Kjeldahl's process. The nitrogen found, multiplied by the
usual factor, will give the meat fibrin, as distinguished from the crude
meat fibre, etc., obtained by weighing the insoluble matter.

Coagulable Albumin can be determined in the filtrate from the in-
soluble matter, by rendering the liquid distinctly acid with acetic acid,
boiling for five minutes, filtering, and determining the nitrogen in the
coagulum. Only insignificant amounts of albumin are usually pre-
sent in meat extracts, but in certain preparations which have received
an addition of scale-albumin the amount may be considerable.

Syntonin. — An aliquot portion of the liquid filtered from the
coagulable albumin should be further acidulated with acetic acid and
tested with potassium ferrocyanide. If any precipitate be formed the
liquid should be heated, and if re-solution does not ensue the pre-
sence of acid-alhumin is certain. If found, the remainder of the
liquid should be rendered exactly neutral to litmus, the precipitate
filtered off and the contained nitrogen determined.

Albumoses and Peptones. — The filtrate from the precipitate of
syntonin, or, in the absence of syntonin, the liquid filtered from
the coagulable albumin, is satur^/ted with zinc sulphate. Fifty
c.c. of the solution, containing from 1 to 2 grms. of solid matter, is
freed from insoluble and coagulable matters, and treated with 1 c.c.
of dilute sulphuric acid (1:4) to prevent the precipitation of zinc
phosphate. It is then completely saturated with zinc sulphate at
the ordinary temperature, by adding the powdered salt as long as
it continues to dissolve on stirring. The precipitate, which will
contain any gelatin and all proteids other than peptones, is filtered
off' and washed with a cold saturated solution of zinc sulphate.
The filter and its contents are then transferred to a flask and treated
by Kjeldahl's process. The precipitate produced contains all the
albumose of the extract together with any gelatin which may be
present and any coagulable or insoluble proteids not previously re-
moved ; peptones, meat bases, amido-compounds, and ammoniacal
salts are not precipitated.

In an aliquot part of the filtrate, peptones may be determined by
precipitation with bromine.

A quantity of the solution containing about 1 grm. of the albumin-
oid matter is diluted with cold w^ater to a volume of about 100 c.c,
and treated in a conical beaker with sufficient hydrochloric acid to
render the liquid distinctly acid to litmus. Bromine water, is then
added in considerable excess, and the liquid stirred vigorously for
some time. The yellowish precipitate which separates is at first
flocculent, but becomes more viscous on stirring, and finally adheres
in great part to the sides of the beaker. When this occurs the
liquid is allowed to stand at rest for about half an hour, or until the
precipitate has settled. It is then decanted through an asbestos
filter.

The precipitate adhering to the sides > of the beaker is washed

406 FOOD AND DEUGS.

several times with cold distilled water, the washings being poured
through the filter. Occasionally, when the greater part of the bromine
has been washed out of the precipitate, the liquid does not filter clear.
It is therefore advisable to keep the washings separate from the filtrate,
and if necessary, to add bromine or sodium sulphate to the wash-water.

The contents of the filter-tube (including the asbestos, and, if
necessary, the glass-wool) are returned to the beaker used for the
precipitation, 20 c.c. of strong sulphuric acid added, and the beaker
covered with a watch glass and heated over wire gauze. The sub-
stance chars and bromine vapour is evolved. When frothing has
ceased, about 10 grms. of powdered potassium sulphate should be
added, and the liquid boiled vigorously until colourless. It is then
allowed to cool, diluted with water, an excess of caustic soda added,
the ammonia distilled off into a known volume of standard acid.
From the nitrogen found the amount of peptones present is deduced
by multiplying by 6-25.

Ammonia is estimated by distillation with barium carbonate ; and
total nitrogen by Kjeldahl's process. These two latter determinations
are, however, preferably made on a filtered aqueous solution of the
original sample. The difference between the total nitrogen and that
found in other forms is regarded as existing as meat bases, etc., the
actual weight of which is usually, calculated by multiplying the nitro-
gen by the factor 3 •12. (But see above).

If the preliminary precipitation with zinc sulphate be omitted the
bromine precipitate will include the gelatine, gelatine-peptone, albumen,
and similar bodies.

True peptones are present in very small quantity in extract of
meat. They may be tested for by the biuret reaction applied as
follows : To the aqueous solution, from which gelatine and albumoses
have been precipitated by the addition of excess of ammonium sul-
phate, a few drops of a very dilute solution of copper sulphate are
added, and then a large quantity of strong caustic soda solution. A
characteristic rose-red colour is produced in the presence of peptones.

Konig and Bomer hold the following views with respect to the
chemical examination of meat extracts and commercial peptones : —

1. Precipitation with 80 per cent alcohol is of no value in deter-
mining the form of combination in which nitrogen exists.

2. Albumoses should be determined by salting out with ammonium
sulphate or zinc sulphate.

3. The filtrate fiom the ammonium or zinc sulphate precipitate
should be decolorized with animal charcoal, and tested for peptones
by the biuret reaction.

4. A determination of the ammonia by distilling an aqueous
solution of the extract with ignited magnesia is valuable.

5. When peptone has been proved to be absent, the nitrogen in
the phospho-tungstate precipitate, after deducting the nitrogen de-
rived from gelatin, albumoses and ammonia, may be ascribed to the
flesh bases. The phospho-tungstate precipitate should stand at least
one day before filtration.

6. The difference between the total nitrogen and the sum of the

EXTEACT OF MEAT.

407

nitrogen in the forms of gelatin, albumoses, flesh bases, and ammonia
gives the amount of nitrogen present in compounds not precipitated
by phospho-tungstic acid. No evidence was obtained of the presence
of amido or acid amido-compounds.

By the application of these principles to the analysis of typical pre-
parations Konig and Bomer obtained the following results : —

Nitrogen in the form of —

1. Soluble albumin

2. Nitrogenous compounds insoluble in 60 to 64

per cent alcohol

3. Albumoses

4. Peptones .......

5. Flesh bases . . . . * .

6. Ammonia

7. Other nitrogenous compounds

Liebig's Extract.

Per cent of
Substance.

Per cent of
Total Nitrogen.

trace
0-21

0-96
0 to trace
6-81
0-47
0-83

trace
2-26

10-34

0 to trace

73-38

5-06

8-96

Total

9-28

—

These amounts of nitrogen represent the following percentages of
nitrogenous compounds : —

1. Soluble albumin

2. Gelatin and proteids insoluble in 60 to 64 per cent

alcohol

3. Albumoses

4. Peptones

5. Flesh bases

6. Ammonia

7. Other nitrogenous matters

Liebig's Extract.

Per cent
trace
1-14

6-05

0 to trace

21-25

0-57

5-23

Total

34-24

The following appears to be the most reliable scheme for separat-
ing the nitrogenous constituents of meat extract. It is largely based
on the work of Bigelow and Cook : —

Complete separation of nitrogen compounds involves a discrimina-
tion between meat fibre and insoluble protein, coagulable proteins, acid
albumin (syntonin), albumoses, peptones, meat bases, gelatin and
ammonia.

(1) Insoluble Proteins. — Agitate 5 grms. of the extract of the dry,

408 FOOD AND DRUGS.

or 20 grms. to 25 grms. of the fluid variety, with 200 c.c, to 250 c.c.
water at about 20° C, and collect the residue on a tared filter. It is
not easy to filter such an extract in the ordinary manner, so that a
centrifugal apparatus is of considerable use in getting the insoluble
matter to settle. After washing the residue, dry at 100", and weigh,
or determine the nitrogen by the Gunning method. Another method
is to place the solution in a graduated flask, immerse in plenty of cold
water for several hours, frequently shakin^;. Determine the nitrogen
in an aliquot part of the filtrate. Deduct this from total nitrogen and
the nitrogen of insoluble proteins is obtained. N x 6-25 = total insoluble
matter including both the meat fibre and insoluble proteins.

(2) Coagulahle Proteins. — Neutralize the filtrate from (1) exactly
to litmus, and add dilute acetic until acidity is just noticeable. Boil
for some minutes to make the coagulahle proteins insoluble ; collect
the latter upon a filter (using a centrifuge as recommended above).
Determine the nitrogen in the washed residue using the factor 6-25 for
coagulahle proteins.

(3) Albumoses or Proteoses. — Saturate an aliquot part of the fil-
trate from (2) with zinc sulphate addin;^ the powdered salt as long
as it continues to dissolve with stirring and shaking. This precipitates
any proteoses, traces of gelatin or insoluble proteins that have evaded
being removed but not the peptones or meat bases. Filter, wash, and
determine the nitrogen in the residue, using the factor 6-25 for the
proteoses, etc.

(4) Pei^tones. — Sjerning's tannin-salt method, modified by Bigelow
and Cook. Take an aliquot part of the filtrate from (2) concentrated
by evaporation to 20 c.c. or less, in case it is necessary to take more
than 20 c.c. and place in a 100 c.c. flask. Add 50 c.c. of a solution
containing 30 grms. of sodium chloride, and thoroughly shake the
flask to ensure the contents being well mixed with the solution of the
sample. Then cool the flask to about 10°. When the solution has
reached this temperature add 30 c.c. of a 24 per cent solution gf tannin,
which must be at the same temperature. The total volume is now at
100 c.c. Again thoroughly mix the contents of the flask, and place
it in a cool place and allow it to remain there overnight. In the
morning filter the solution at from 8° to 10° into a 50 c.c. graduated
flask. Determine the nitrogen in this filtrate, also in an aliquot portion
of the filtrate from a blank in which the reagents alone are employed.
Multiply the nitrogen found in the 50 c.c. portion by 2 (after correc-
tion for the nitrogen in the blank) which gives the total nitrogen in
the filtrate, and is calculated to per cent, of nitrogen on the sample
used. This includes the nitrogen present as ammonia and all the
nitrogen of the meat bases except that portion of the creatin precipi-
tated by the tannin-salt reagent. Add the figure thus obtained to the
per cent of nitrogen as determined in (1), (2) and (3). This sum
after deduction from the total nitrogen is generally given as the per
cent, of nitrogen existing as peptones and is multiplied by 6'25 for the
per cent of peptones.

Probably these substances, how^ever, are not true peptones, as the
filtrate from (3) usually gives no biuret reaction. It is not unlikely

EXTKACT OF MEAT. 409

that they consist chiefly of peptoids, formed by the action of the hot
solution Qn gelatin and polypeptides.

Bigelow and Cook state that the tannin-salt precipitate is not
contaminated with other meat bases than creatin. They consider that
about one-quarter of the creatin is found in this precipitate. Hence
they advise that the percentage of creatin should be determined before
and after precipitation with tannin-salt reagent, thus correcting the
results obtained. Street considers that this correction is impracti-
able. He thinks that it is exceedingly difficult if not impossible to
completely remove tannin from the filtrate, and that the least trace of
tannin prevents the colour reaction for creatin.

(5) Meat Bases. — Deduct the per cent of nitrogen found as ammonia
in (6) from the per cent of nitrogen found in the filtrate from the
tannin-salt precipitate in (4) and multiply the result by 3-12 to obtain
the per cent of meat bases.

(6) Ammonia. — Dissolve from 5 grms. to 10 grms. of the original
sample in a convenient volume of water, add powdered magnesia, then
distil. Titrate the distillate and estimate its alkalinity as per cent of
NH . Calculate the corresponding percentage of nitrogen also, as it is
necessary for determining meat bases in (5).

Determination of Creatin and Creatinin. — An aliquot portion of
the filtrate from the insoluble and coagulable protein determination
can be used for this determination. This portion, however, must
contain sufficient total creatinin after dehydration of the creatin to
creatinin to give a reading not far from Q" on the scale of the Dubosc
colorimeter, after applying the colorimetric method as outlined by
Eolin for the determination of creatinin in the urine. Add 5 c.c. of
semi-normal hydrochloric acid to this aliquot portion and heat for
three and a half hours on a water bath under a reflux condenser.
Add 5 c.c. of half-normal sodium hydroxide to neutralize the hydro-
chloric acid, then add 15 c.c. of a saturated picric acid solution, and
5 c.c. of 10 per cent sodium hydroxide. Agitate the solution and
allow it to stand for five minutes ; make up to 500 c.c. and compare
the colour with a half-normal solution of potassium bichromate in the
Dubosc colorimeter. The half-normal bichromate solution corresponds
to 10 mg. of creatinin, w^hen the scale is set at 8°, and the amount of
creatinin in the aliquot can therefore be estimated without difficulty.

Hehner does not consider this method suitable for application to
meat extracts. He concludes that more satisfactory results are ob-
tained from using 25 c.c. of a 1*01 per cent of picric acid with " a
quite small amount of alkali". He contends that the precipitate is
somewhat soluble in excess of alkali. Emmett and Grindley, who have
made an exhaustive study of the method as applied to meats, meat
extracts and wines, point out that 15 c.c. of 1-2 per cent picric acid
should be used for the original creatinin determination and 30 c.c. for
the dehydrated creatinin. They also suggest 5 c.c. of alkali for the
original creatinin and 10 c.c. for the dehydrated creatinin, though an
additional 5 c.c. does not produce lower results.

Determination of Xanthin Bases. — A true meat extract or meat
juice should contain in addition to creatin and creatinin, small quan-

410 FOOD AND DRUGS.

titles of xanthin bases including xanthin, hypo-xanthin, guanin, and
adenin. The nuclei of the cells produce these bodies, hence a certain
amount of the latter should be obtained in an extract that is prepared
from fresh unaltered beef, as well as salts and other extractive matter.
The determination of the xanthin bases is consequently valuable in
determining the origin of an alleged extract of meat. They are deter-
mined by the following method : —

Schittenhelm s Method Modified by Cook. — Take an amount of the
standard solution equivalent to 5 grms. of the original extract. Trans-
fer to a large evaporating dish then add 500 c.c. of 1 per cent sul-
phuric acid. Evaporate to 100 c.c. within four hours to five hours.
Cool, and add sodium hydroxide to neutralization, allow to stand
overnight, filter, and wash. Treat the precipitate held in suspension
in the water with sodium sulphide and warm on the water bath.
Add acetic acid to acidify, and filter hot. Add 10 c.c. of 10 per cent
hydrochloric acid to the filtrat^e and evaporate to a volume of about
10 c.c. Filter, make ammoniacal, and add ammoniacal silver nitrate of
3 per cent strength. Allow to stand for several hours, filter the solu-
tion, and wash the precipitate with distilled water until it is no longer
alkaline. The nitrogen in the precipitate is that of these xanthin bases.

Determination of Gelatin. — This is accomplished by the modified
Stutzer method given on page 391.

The recent adulteration of extract of meat with extract of yeast
first received attention at the hands of Searl. The following method
of detecting this adulteration was published by him, but has been
shown to be somewhat unreliable, unless large quantities of the
adulterant are present, when by a comparison with genuine extract
useful deductions can be draw^n.

Make a modified Fehling's solution by dissolving 200 grains of
sulphate of copper and 250 grains neutral sodium tartrate in 4 oz.
water ; add to this 250 grains caustic soda dissolved in 4 oz. of water.
Dissolve 10 grains of the sample to be examined in 1\ oz. water, and
add to it half volume of the above solution, and boil for a minute or
two.

With genuine meat extract no precipitation occurs, but with yeast
extract a bulky, curdled precipitate of a bluish-white colour is thrown
out, which is almost insoluble in water. When collected, washed,
dried, and weighed, several samples of yeast extract have been found
to give approximately 1 grain of this precipitate (it looks to the eye
more like 20 grains) from 10 grains of extract. It naturally varies a
little, according to the amount of moisture and ash contained in the
sample. Only one sample of yeast extract has yet been found which
did not respond to this test, and in that case it readily reduced the
copper. Continental extracts of yeast have given the best results with
this test. An English make does not respond to it.

Since yeast extract can be manufactured at a nominal cost from
brewers' and distillers' waste products, and its physical characters
closely resemble meat extract, it forms an excellent material for fraud-
ulent admixture, for which, until now, no simple chemical test has
been available.

EXTKACT OF MEAT.

lasHSeen mentioned above, a method is available for the deter-
mination of the creatin and creatinin present in meat extracts. The
amount of these bodies, expressed as creatinin, present in normal
extract of meat, varies from 4 '5 to 6 per cent, and any considerable
shortage below the lower limit will indicate either a badly prepared
or very gelatinous extract or the presence of yeast extract.

This reaction has been most fully studied by A. C. Chapman
("Analyst," xxxiv. 475). For a full account of the raison d'etre of
the reaction, reference should be made to the original paper. The
estimation is carried out in the following manner : —

A 10 per cent solution of the meat extract in distilled water is
prepared. Several 10 c.c. quantities of this solution are then trans-
ferred to small beakers, and to each 10 c.c. of normal hydrochloric
acid are added, after which the beakers are heated in an autoclave
for half an hour at 120" C. ; the whole of the creatin present is thus
converted into creatinin with the minimum amount of decomposition.
To the contents of one of these beakers, cooled to 20°, 30 c.c. of a
saturated solution of picric acid, and 15 c.c. of a 10 per cent solution
of sodium hydroxide are added. After standing for five minutes, the
coloured liquid is made up to 500 c.c. This solution is then matched
in a Dubosc colorimeter, against 8 mm. of a standard bichromate
solution containing 24*54 grms. per litre. The colour of the 8 mm.
column of this solution is practically identical with that oi 8*1 mm.
of a solution containing 10 milligrams of creatinin per 500 c.c. treated
with picric acid and alkali. From the reading obtained, it will be
easy to dilute a second 10 c.c. of the solution being tested to corre-
spond practically exactly with the standard bichromate, from which
the amount of creatinin can be calculated from the details above
given. Or a solution of pura creatinin of 20 milligrams per litre
may be prepared and treated side by side with the sample being tested,
and the colours matched in Nessler glasses, from which the amount
of creatinin is calculated.

Micko states (" Zeit. Untersuch. Nahr. Genussm." 1910 19, 426-
434) that although no substance other than creatinin, which yields
Jaffe's reaction, is likely to be present in meat extract, etc,, it is ad-
visable, especially in the case of extracts of unknown composition, to
have some means besides the reaction just mentioned of ascertaining
the actual presence or absence of creatinin in a sample and he there-
fore gives the following process for isolating the base : 10 grms. of
meat extract (larger quantities are taken in the case of mixtures of
meat and yeast extracts) are dissolved in water, lead acetate is added
until a precipitate is no longer formed, and the whole is diluted with
water to a volume of 1 litre. After the lapse of several hours the
liquid portion is passed through a filter, and the excess of lead is re-
moved by evaporating the filtrate after the addition of hydrochloric
acid, and adding alcohol to the residual solution. The dilute
alcohol solution, thus freed from lead, is evaporated to dryness, the
residue is dissolved in about 100 c.c. of water, the solution is neutra-
lized with sodium hydroxide and then treated with 10 c.c. of 20 per
cent sodium bisulphite solution and 10 c.c. of 13 per cent copper

412 FOOD AND DRUGS.

sulphate solution ; the mixture is boiled, allowed to cool, filtered, and
the precipitate is washed with cold, previously boiled water. The fil-
trate is acidified with hydrochloric acid and heated to expel sulphurous
acid, and the copper is removed as sulphide. The solution is now eva-
porated and the residue is extracted with alcohol in order to separate
the bases, etc., from the large quantity of alkali salts present. The
alcoholic solution is evaporated to a syrupy consistency, then acidi-
fied with about 50 c.c. of dilute sulphuric acid, and 30 per cent
phospho-tungstic acid solution is added in slight excess. After
standing for two days the precipitate is collected on a filter, washed
with acidified dilute phospho-tungstic acid solution, until free from
chlorides and then rinsed into a beaker with hot water. Hot satu-
rated barium hydroxide solution is added to the contents of the
beaker in sufficient quantity to render the mixture distinctly alkaline
in reaction, the precipitate is collected on a filter, washed with hot
water, and the filtrate, after neutralization with sulphuric acid, is
evaporated to a syrup. The latter is dissolved in dilute sulphuric
acid, again evaporated, and then dissolved in water and evaporated
once more. The syrup obtained is dissolved in the least possible
quantity of water, hot alcohol is added, and the mixture is placed aside
for about twenty-four hours. The alcoholic solution is then decanted
and evaporated, the residue is extracted with alcohol, and the solution
is separated from the insoluble portion. The first syrupy residue is
also extracted a second time with alcohol, and the united alcoholic
extracts are now evaporated : the residue is dissolved in 30 c.c. of
water, the solution is heated to boiling and rendered alkaline by the
addition of lead hydroxide, the whole being then diluted with several
times its volume of hot alcohol. x\fter standing for some hours the
mixture is filtered, the alcohol is evaporated from the filtrate, and the
lead is removed as sulphide. On evaporating the lead-free solution a
crystalline mass is obtained which is dissolved in 40 c.c. of 1*2 per
cent picric acid solution. Next day the creatinin picrate formed is
collected on a filter, the filtrate is evaporated under reduced pres-
sure, the residue is again dissolved in picric acid solution and allowed
to crystallize — these operations being repeated until crystals of creatinin
picrate are no longer formed. The creatinin picrate thus obtained
is heated with dilute hydrochloric acid, and the picric acid is extracted
with toluene, the aqueous solution of the creatinin hydrochloride is con-
centrated, treated with animal charcoal, and evaporated until crystals
begin to form. After cooling, the moist crystalline mass is treated
with a mixture of one part of acetone with two parts of absolute
alcohol, and the insoluble salt is then collected on a filter. A small
quantity of creatinin hydrochloride passes into the filtrate, and the
latter must be evaporated and the residue once more treated with the
acetone-alcohol mixture. The total quantity of crystals is then dried
at 100° C. and weighed. The sample of meat extract examined,
yielded 4-0 per cent of creatinin hydrochloride. Yeast extracts do
not yield any creatinin hydrochloride when examined by this process.
Cook (U. S. Dept. of Agriculture, Circular 62, 1910) states that
meat extracts contain from 6 to 8 per cent of ether-soluble material,

GELATINE. "^^^BT 415

whilst yeast extract only contains 1 per cent. On the dry and fat-tree
extracts, he finds the following differences : —

Yeast extract. Meat extract.

Per cent. Per cent.

Ash 27-3 to 30-4 18-2 to 24-9

Nitrogen 7-4 ,. 7-5 at least 11-5

Phosphoric acid in ash .... 33 „ 3-9 2-6 to 34

Boric Acid is sometimes added to meat extracts as a preserva-
tive. The presence of such a substance in an article intended for the
use of invalids and persons whose digestion is impaired is very un-
desirable. Boric acid may be detected and determined by the
methods employed for milk. A modified process recently proposed
by C. Fresenius and Popp (" Analyst," xxii. 282) and applied by
them to the examination of sausages, etc., may also be employed
for the determination of boric acid in meat extracts. An amount of
the extract corresponding to about 3 grms. of dry substance should be
concentrated to a syrup, if necessary, and mixed in a mortar with
from 40 to 80 grms. of recently ignited sodium sulphate. The mixture
is heated in the water-oven for about an hour, and as soon as the
mass is dry some more sodium sulphate is added, and the whole re-
duced to a fine powder. This is digested with 100 c.c. of cold methyl
alcohol for twelve hours, with frequent shaking, after which the
alcohol is distilled off. As a rule the boric acid passes over completely
in one distillation, but it is desirable to extract the residue a second
time, using 50 c.c. of methyl alcohol. The distillate is made up to
loO c.c, and 50 c.c. treated with 7*5 c.c. of water and 25 c.c. of pure

N
glycerine. The mixture is titrated with — solution of caustic soda.
^^ 20

(free from carbonate), using phenol-phthalein as an indicator. A pale
rose colour indicates the end of the titration. When it appears, some
more glycerine should be added, and if the colour is not permanent
the titration is continued till that point is attained. The volume of
alkali used (in c.c.) multiplied by 0-0031 gives the boric acid, H3BO3
(in grms.), in the volume of the distillate titrated. Borates will be
dissolved out of the organic matters by the methyl alcohol, but will
not pass over with the free boric acid. They may be determined in
the usual manner in the methyl alcoholic extract, after evaporation,
ignition, etc..

GELATINE.

It will be convenient to here briefly discuss gelatine as it affects
the food analyst.

Gelatine is employed to a considerable extent in the manufacture
of certain food products, and is sold retail to householders to a con-
siderable extent.

From an industrial point of view the examination of gelatine and
glue may be of an exhaustive nature, but for the purposes of an
analysis of gelatine used for edible purposes, the examination will be-
of a more restricted nature.

414

FOOD AND DRUGS.

Gelatine does not occur — at all events to any extent — ready formed
in nature, but is a protein-like body resulting from the decomposition
of other substances by the action of boiling water or dilute acids.

The composition of gelatine is approximately as follows : —

Per cent

Carbon 50

HyArogen 6-5 to 7

Nitrof^en 17 „ 19

Oxygen . 25

Traces of sulphur are usually present in gelatine — up to 0*5 per
cent — but its significance is unknown, and it does not, probably, enter
into the constitution of the gelatine molecular complex. Isinglass is
a closely related substance, which readily yields either gelatine or a
substance nearly indistinguishable from gelatine. It is obtained from
the swimming bladder of numerous species of fishes.

The following analyses of edible gelatine and isinglass are due to
Tankard : —

Moisture •...*...

Sulphur

Nitrogen in precipitate from aqueous solution

by ZnSOj

„ X 5-42 = "gelatine".

Total nitrogen

Ash

Isinglass.

Per cent

15-05

0-38

14-00

75-88

14-56

1-30

Gelatine.

Per cent

17-90

0-17

13-09

70-95

14-10

3-70

The British Pharmacopoeia describes gelatine as the air-dried
product of the action of boiling water on such animal tissues as skin,
tendons, ligaments, and bones. It is required to be free from chondrin,
since it is stated to give no precipitate with acids, alum, lead acetate
or 5 per cent solution of ferric chloride.

From the food point of view, the following are the essentials of a
good gelatine : —

(1) When soaked in cold water for four hours and then made into
a jelly by heating with water, it should yield no offensive odour. If
it has an objectionable smell, it should at once be rejected as unfit
for food.

(2) If in aqueous solution, it yields an appreciable precipitate
with the above-named reagents it is to be regarded as of inferior
quality as containing excess of chondrin, resulting from the decomposi-
tion of hyaline cartilage.

(3) Its ash should contain no heavy metals, such as copper, which
is sometimes present to the extent of 1 grain per lb., and not more
than 10 parts of iron per 100,000.

(4) It should not contain more than 5 parts of SO.j per 100,000.
This is determined by distilling a 5 per cent solution of the gelatine,

GELATINE.

415

oxidizing the distillate with bromine water and precipitating the SO^
formed, by BaCl^. Excess of SO.2, due to the use of SO., as a bleaching
agent may cause action to be set up with the metal of the tins in
which the product is frequently packed, and consequent discoloration
of the product, due to the formation of metallic sulphides.

CHAPTER VII.

MICROSCOPICAL ANALYSIS.^

While chemical analysis furnishes the means of determining the
chemical composition of foods and drugs, and thus ascertaining their
freedom from adulteration, it is often by microscopical analysis
alone that the identity and purity of such as are powdered can de-
finitely be determined. The microscopical examination of powdered
foods and drugs should, therefore, never be omitted ; even with
many substances other than powders valuable results may be ob-
tained. It often affords, in a minimum of time and w^th a minimum
of material, information that cannot be obtained by any other known
means.

Apparatus Required.

For microscopical analysis the following apparatus will be re-
quired : —

1. Microscope. — This should be capable of magnifying from 50
to 500 diameters, and should possess a revolving nosepiece and a
substage condenser. It should be provided with an Abbe- Zeiss
camera lucida for . sketching, and a separate eyepiece in which an
ocular micrometer is permanently fixed ; the value of the divisions of
the micrometer, when that particular eyepiece is used in conjunction
with each objective, should be determined and kept' ready for im-
mediate reference. A polarizing apparatus is of service in detecting
crystals, and a mechanical stage is useful when preparations have to
be thoroughly searched, but neither of these is absolutely necessary.

The best light is that obtained from a north or east window.
Direct sunlight is to be avoided ; if that is not possible, it should be
modified by means of a white blind. As artificial light, a small, in-
verted, incandescent gas burner with a ground glass globe, or a car-
bon filament lamp answers well. In no case should the field be more
brightly illuminated than is necessary, and both eyes should be kept
open during the work.

2. Centrifuge. — A small centrifuge is very effective in separating
fine powders from liquids, after bleaching or staining, and saves much
time ; plain centrifuge tubes answer every purpose.

3. Dissecting Needles. — Two plain and two glover's needles,.

^ The author desires to acknowledge his indebtedness to Professor Greenish,,
who has kindly written this chapter.

(416)

MICKOSCOPICAL ANALYSIS. 417

liidiiniied in handles with screw caps, which hold needles of any size
firmly and allow of ready changing.

4. Glass Dishes. — Several small hollowed glass blocks, with
covers ; these are far preferable to watch-glasses.

5. Slides. — Glass slides of the usual size and thickness.

6. Ooverslips. — Three-quarter inch square No. 2 coverslips are the
"naost generally useful ; for the highest power No. 3 may be employed.

7-. Reagent Bottles. — Half-ounce or 6 drachm square bottles with
glass peg- stoppers ; they are conveniently kept in wooden trays pro-
vided with covers.

8. Razors. — One solid and one hollow-ground razor.

9. Elder Pith. — This may be obtained from opticians or clock-
•makers.

10. Glass Plate with black and white fields.

Preparation of thb Material.

The preparation of the material for examination is comparatively
simple.

Carefully bulk the powder and set aside from 5 to 10 grms. for
examination. Sift 1 or 2 grms. of this through a No. 60 sieve.
Should any fragments fail to pass through the sieve, examine thena
with a lens and pick out such as appear suitable for section cutting.
Make and examine sections. Then powder a fresh quantity until it
all passes through the sieve.

Transfer about O'l grm. of this sifted powder to a glass dish, add
a drop of water, and triturate with a glass rod until thoroughly mixed.
Then add water, drop by drop, until the mixture acquires the consis-
tency of a thin cream ; cover, label, and set aside. Make a similar
preparation with a mixture of equal volumes of water and glycerin,
and a third with solution of chloral hydrate (see list of reagents). Let
these preparations stand for about twelve hours. Then mix the
water preparation thoroughly with a glass rod, transfer to a slide a
small portion of the mixture, which should be so viscous as to ensure the
removal of a representative sample, add a small drop of water, mix, and
cover with a coverslip, which should be carefully lowered on to the
slide so as to avoid the introduction of air bubbles. The quantity of
water should be just sufficient to fill the space between the slide
and the coverslip ; it should not be so large as to allow the cover-
slip to float, as then the particles of powder will be in more or
less constant motion, nor should it be so small as to cause the
coverslip to press on the slide, as this pressure is liable to distort
delicate tissues. The quantity of powder should be small enough to
avoid any overlapping of the particles, but not so small as to make
the distance between them unnecessarily great. Practice will soon
indicate the correct quantities. It is better to make a fresh prepara-
tion than to proceed with the examination of one that is defective.

Make in the same way a preparation in dilute glycerin and one
in solution of chloral hydrate with the powders that have been stand-
ing in these liquids.

VOL I. 27

418 FOOD AND DRUGS.

If the powder must be examined without delay, the following more
rapid method may be adopted : —

Place a small portion of the powder, about the size of a mustard
seed, on a slide, add a small drop of alcohol, allow most of the
alcohol to evaporate, add a drop of water, mix and cover with a cover-
slip. Mount, similarly, a little of the powder in dilute glycerin and
in solution of chloral hydrate. The use of alcohol may be omitted if
the powder contains much resin, but in this case the particles of
tissue may contain numerous air bubbles which seriously interfere
with the examination ; these may be driven out by gently warming
the preparation until a few gas bubbles escape.

Examination of the Preparations.

1. Water Preparation. — Examine the preparation in water first,
using the low power. Observe the colour of the fragments ; this
often atfords valuable information. If the powder is a very fine one,
sclerenchymatous cells, bast fibres, and thick-walled hairs may be
found intact; they should be examined under the high power and
sketched. Fragments of cells and, of cell walls will probably be numer-
ous ; they should be similarly dealt with. Portions of the epidermis
of leaves, held together by the resistent cuticle, may be found. The
larger fragments of tissues, which will be particularly numerous in
coarse powders, will probably be too opaque, and may be better ex-
amined in dilute glycerin, or in solution of chloral hydrate. On the
other hand, cell contents, liberated from the cells, may be found in
abundance. Examine for the following cell contents and cells : —

(a) Starch. — Make a fresh preparation in water. Bring a drop of
dilute solution of iodopotassium iodide into contact with the edge of
the coverslip, and, under the high or low power, watch it as it pene-
trates ; if necessary, draw the solution under the coverslip by applying
a pointed piece of filter paper to the opposite side. The iodine will
stain the starch grains deep blue or nearly black. If starch is present,
determine, by means of the ocular micrometer, the length of the
largest, of the smallest, and of the most frequently occurring grain.

If the powder consists mainly of starch, it may be desirable to
remove this and examine the residue (for methods of doing this, see
below).

{h) Oil and Besin. — Dilute 1 c.c. of tincture of alkanna with an
equal volume of water ; mount a little of the preparation in the mix-
ture, allow it to stand for half an hour and^K amine. Oil and resin
will be stained red. If much oil is present, it may advantageously
be removed by treating the original powder with ether-alcohol or
ether, drying, and making fresh water and other preparations from
the defatted powder.

(c) Aleurone Grains. — As maceration with water disintegrates
aleurone grains, a fresh preparation should be made as follows : —

Moisten a little of the powder with alcohol, let it stand till the
alcohol has nearly evaporated, and add a drop of solution of picric acid.
After two or three minutes, remove the aqueous liquid with filter

MICROSCOPICAL ANALYSIS. 419

paper, add a drop of glycerin, cover and examine. The aleurone
grains will be stained yellow ; they may contain one or more crystal-
loids, globoids, or crystals of calcium oxalate ; frequently two of these
varieties of contents are present, but very rarely all three. Jirigate
with very dilute solution of potassium hydroxide ; the ground substance
and crystalloids will dissolve instantly, leaving the globoids and
calcium oxalate.

(d) Mucilage. — This, also, is best detected in a fresh preparation: —
Moisten a little of the powder with alcohol, and add a drop of

solution of ruthenium red .(see list of reagents). Many, but not all,
mucilages will be coloured bright pink. Treat another portion
similarly, with solution of corallin-soda ; some mucilages will stain
pink ; if sieve tubes are present the callus plates will acquire the
same colour. Treat another portion of the powder with Indian ink
diluted with water ; mucilage is often readily seen as colourless or
nearly colourless masses.

(e) Lignified Tissue. — Mix a small portion of the powder with
alcoholic solution of phloroglucin, and allow it to stand for a minute
or two, covering to avoid evaporation. Then remove any excess of
the solution, add a drop of strong hydrochloric acid, cover, and
examine. Lignified tissue will be stained bright red.

2. Glycerin Preparation. — Examine next the glycerin prepara-
tion, mounting as directed for the water preparation. The larger frag-
ments of tissue now appear clearer, since the refractive power of
glycerin more nearly approaches that of cellulose, but this advantage
is partly counterbal9,nced by the loss of delicate details, such as the
striations of starch grains, which, for the same reason, become invisible.
Hence a glycerin preparation is more suited for the examination
of the larger particles, groups of cells, etc., than it is for that of frag-
ments of cell walls, the finer details of w^hich have often to be deter-
mined. The longer the preparation is kept the clearer the particles
become, but this is not always an advantage.

3. Chloral Hydrate Preparation. — Solution of chloral hydrate
(see list of reagents) has a very powerful solvent action, dissolving
protoplasm, aleurone grains, colouring matter and, in the course of a
few hours, starch. This property, added to its high refractive index,
makes it a more powerful clearing agent than glycerin, but at the
same time more liable to obliterate delicate markings.

Examine, in this preparation, the larger fragments of tissue and
endeavour, by gradually focussing downwards, to determine the char-
acters of the successive layers of which they consist. Groups of bast
fibres and of sclerenchymatous cells, fragments of cork and of
epidermis are usually very clear, and crystals of calcium oxalate con-
spicuous. This medium is excellent for giving a rapid, clear survey
of the tissues present, the details of which are often better seen in a
dilute glycerin or in a water preparation. Hence it is often desirable
to return to one of these after having examined the chloral hydrate
preparation. It must also be remembered that solution of chloral
hydrate may induce a swelling of the cell wall, particularly if the pre-
paration has been warmed.

420 FOOD AND DEUGS.

In addition to the foregoing preparations, which should invariably
be made, the following are useful for special purposes : —

i. Potash Preparation. — Mount a little of the pow^der in an
aqueous 10 per cent or 20 per cent solution of potassium hydroxide ;
warm gently, cool, and examine. Starch will have been gelatinized,
the cell walls expanded, and much colouring matter removed, as
is the case with solution of chloral hydrate, but the digestion wnth
hot solution of potassium hydroxide loosens the cells, especially par-
enchymatous cells, to such an extent that they may frequently be
separated from one another. Press the coverslip dow^n moderately
firmly with the finger and then push it rather sharply along the slide.
By this means disintegration of the tissues can often be effected.

6. Oil Preparation. — Mount a little in almond oil or in a mixture
of castor oil and alcohol, and examine. Substances soluble in water,
or in the other mountants mentioned, may often be detected.

Further treatment of the powder is occasionally desirable.

6. Removal of Starch. — This is advantageous when dealing with
powders consisting largely of starch, as it permits of the concentration
and ready examination of the tissues present. Mix 5 grms. or less
of the powder with 50 c.c. of water, add 2 c.c. of hydrochloric acid
(specific gravity 1-16), boil gently for five minutes, and cool. Centri-
fuge the turbid liquid, pour oE the supernatant solution and wash
the deposit once with water, again separating by the centrifuge. Add
now a few c.c. of solution of chloral hydrate, stir well, centrifuge again,
and finally mount in water, dilute glycerin, or chloral hydrate.

7. Bleaching. — Some powders are so dark in colour that it is
necessary to bleach them before applying colour reactions. Mix
about 0-1 grm. of the powder with 5 c.c. of solution of chlorinated
soda (see list of reagents), and allow the mixture to stand, shaking
occasionally, until most of the colour has been removed ; dilute with
an equal volume of water, separate by centrifugation, and w^ash the
deposit several times with water. The bleached powder may be
mounted in water or glycerin, or it may be stained with phloroglucin,
chlorzinciodine, etc. It may also be double-stained as follows :—

8. Double-staining. — To the deposit, bleached as above described,
add 5 c.c. of Cordonnier's double-stain (see list of reagents), allow it
to stand for ten minutes, dilute with an equal volume of water, separ-
ate by centrifugation and wash several times with water. Then pass
it once through 60 per cent alcohol, once through 90 per cent alcohol,
twice through absolute alcohol, and twice through xylol, separating
each time by the centrifuge. Mix the mass uniformly wdth a small
glass rod, and mount a little in balsam. Oily and fatty powders
should be previously defatted.

9. Disintegration by Schulze's Maceration Mixture.— The
elements of which sclerenchymatous tissue consists may be separated
by maceration with potassium chlorate and nitric acid (Schulze's
maceration mixture), which also removes colour, proteid matter,
starch, calcium oxalate, and many other of the cell contents, but
does not readily attack lignified tissue, and still less readily suberized
tissue. Mix about 1 grm. of the powder in a small flask with 5 c.c.

MICKOSCOPICAL ANALYSIS. 421

of water, add 10 c.c. of nitric acid (specific gravity 1-42) and 1 grm. of
potassium chlorate in crystals. Warm gently until evolution of
chlorine commences. When completely bleached, dilute with an equal
volume of water, separate by centrifugation, wash with water, centri-
fuge again, and mount the deposit in water. The sclerenchymatous
cells and fibres may be separated from one another by gentle pressure
and then examined ; they are generally very clear.

10. Mechanical Separation. — Mechanical separation, either by
elutriation or by sifting the powder through sieves of varying degrees
of fineness, occasionally yields useful results, especially in the detec-
tion of certain tissues, cells or cell contents present in small quantity
only. Water is the most useful liquid for elutriation. Groups of
sclerenchymatous cells and fibres, and calcium oxalate crystals are
among the first fragments to subside ; these are followed by groups of
parenchymatous cells, while fragments of such cells, starch grains and
minute cell contents are among the last. Sifting requires a rather
large amount of material ; the most useful sieves are those with 120,
100, 80 and 60 meshes to the linear inch.

Determination of the Okgan from which the Powder is
Derived.

The following procedure may be adopted to determine the organ
from which a powder is derived : —

{a) Observe the colour of the particles, as seen in dilute glycerin.
A green colour indicates leaf, leaf-stalk, herbaceous stem, or, possibly,
calyx of a flower. Examine the chloral hydrate preparation. The
presence of an epidermis with stomata and polygonal or wavy epider-
mal cells, of branching veinlets and of palisade tissue or spongy
parenchyma indicates a leaf. Elongated, rectangular epidermal cells
are probably derived from the midrib, or from an herbaceous stem ;
fragments consisting of large, elongated, colourless parenchymatous
cells point to the former. If pollen grains are found a flower may be
suspected, and search should be made for portions of the petal, which
will probably be coloured and have a papillose epidermis, and for the
characteristic spirally or reticulately thickened cells from the endo-
thecium of the anthers.

(b) If chlorophyll is absent, examine the chloral hydrate pre-
paration-for vessels. In the absence of chlorophyll these wall indi-
cate the presence of wood, which may be derived from a trunk, a
root, or a rhizome. Abundant, irregular fragments consisting of
wood fibres wiih medullary rays crossing them at right angles, ac-
companied by comparatively little calcium oxalate, indicate a wood.
On the other hand, abundant parenchymatous tissue, with starch, oil
or other reserve material, indicates a root or rhizome ; in this case,
sclerenchymatous cells, or bast fibres, isolated, or in more or less
regular groups, may be present.

(c) If vessels are absent, stain a bleached preparation (7) with
corallin-soda (see list of reagents) and examine for laige sieve tubes.
These, in conjunction with fragments of cork, and possibly with iso-
lated or grouped bast fibres or sclerenchymatous cells, indicate a bark.

422 FOOD AND DKUGS.

(d) If the powder is free from chlorophyll, large vessels and sieve
tubes, it is probably derived from a seed or fruit. Examine for
parenchymatous tissue with reserve material, which may be present
in the form of starch, oil, cellulose, aleurone grains, etc. Such tissue
is commonly found in seeds and these may, of course, form part of a
fruit. The presence of an epidermis with more or less distorted
stomata, or of much empty parenchymatous tissue, indicates a fruit.

Identification of the Powder.

Having determined the organ from which an unknown powder is
derived the next step is its definite identification. This demands
considerable skill and experience. It is best effected by comparing
the sketches made of the tissues and elements of the powder with
illustrations published with the various works dealing with this
subject : —

Greenish and Collin, "Anatomical Atlas of Vegetable Powders ".

Greenish, " Microscopical Examination of Foods and Drugs ".

Winton, " Microscopy of Vegetable Foods ".

Schneider, "Powdered Vegetable Drugs".

Vogl, " Die wichtigsten vegetabilischen Nahrungs- und Genuss-
mittel ".

Solereder, " Systematic Anatomy of the Dicotyledons ".

The identification should invariably be confirmed by powdering
the substance indicated, and comparing the two powders under
similar conditions.

Determination of Purity.

The microscopical examination of a vegetable powder often has
for its object the determination of the purity or otherwise of a powder
of given origin. In such case also, comparison of the powder with
a specimen of about the same degree of fineness, known to be
genuine, and under exactly similar conditions, is absolutely necessary.
Hence the gradual accumulation of a set of authentic specimens of
the more commonly occurring foods, drugs, spices, etc., has much
to recommend it. Care must be taken to interpret correctly the re-
sults of the microscopical examination.

The methods by which powders are sophisticated may be classed
under the following heads : —

1. Total Substitution.

2. Intentional Addition.

3. Intentional Abstraction.

4. Intentional Abstraction and Addition.

1. Total Substitution.— -The total substitution of one powder for
another is, as a rule, readily detected, although in some cases as, for
instance, the substitution of cassia for cinnamon, considerable caution
has to be exercised.

2. Intentional Addition. — Here, also, care is necessary not to con-
sider as intentional addition isolated particles accidentally or un-
avoidably present. Drug mills are frequently cleansed with saw-

MICROSCOPICAL ANALYSIS 423

dust, and an occasional fibre of pine wood is often met with in com-
mercial powders. Only when the quantity is considerable can this
be regarded as serious. Many barks have fragments of wood adher-
ing to them ; woods contain portions of bark ; rhizomes have the
lower leaves, or leaf-bases, or portions of the stem attached ; leaves
are accompanied by stalks, flowers, fruits, etc. The quantity, there-
fore, in which foreign substances occur must be taken into con-
sideration.

3. Intentional Abstraction. — This may be either mechanical or
chemical. Mechanical abstraction may be the result of improper
sifting, whereby an undue quantity of those elements which resist
pulverization may have been separated from those that are easily
powdered ; such alteration in the composition of the powder may be
detected by carefully noting the proportion in which the various ele-
ments occur in a genuine powder with that in which they are present
in the sample under examination. -

Chemical abstraction may take the form of the removal of some of
the soluble constituents, or of admixture with a powder that has
already been exhausted. In some cases the absence of colouring
matter, o^ of various secretions, etc., or the presence of gelatinized
starch grains may indicate that such form of sophistication has been
practised, but, in general, this is a point that must be decided by
chemical analysis. In fact, in the analysis of powdered foods and
drugs, chemical and microscopical analysis should always go hand in
hand.

4. Intentional Abstraction and Addition. — It occasionally happens
that the attempt is made to cloak the removal of active constituents
from a drug by adding a foreign material to it. Such sophistication
is readily detected.

Detection op Insect Pests.

Powders are occasionally prepared from material that has been at-
tacked by insect pests. The most common of these is the larva of the
beetle, Sitodrepa panicea. As the mature beetle is present in very
minute quantity only, a special method of procedure has to be adopted ;
the following ^ will be found satisfactory : —

Defat 5 grms. of the powder with ether in a Soxhlet ; dry the de-
fatted powder and boil it with 100 c.c. of 5 per cent hydrochloric acid
for five minutes in a tared flask ; add about 150 c.c. of water, allow
the powder to settle, and wash once by decantation. For every 35
grms. of water and powder in the flask add 6 c.c. of concentrated sul-
phuric acid, cool, and then add, in small portions and cooling again if
there is any considerable rise in temperature, 10 c.c. of a 1 in 1 aque-
ous solution of chromic acid. Allow the mixture to stand with occa-
sional agitation for thirty-six hours or longer. Separate the solid
particles by centrifugation, wash them with water, alcohol, and ether
successively, dry, remove from the tube, and mount in solution of

' Greenish aad Braithwaite, " Pharmaceutical Journal," Vol. LXXXV, p. 580,

424 POOD AND DEUGS.

chloral hydrate, or, if permanent preparations are desired, in xylol
balsam.

If the powder contains but little that is soluble in ether the treat-
ment with this solvent may be omitted, and similarly that wath the
hydrochloric acid. As these pests are almost ubiquitous care must be
exercised in condemning a powder in which they have been found.

List of the Principal Keagents.

The following reagents will be found sufficient for the examination
of most powdered foods and drugs : —

Acetic Acid. — The Acidum aceticum of the British Pharmacopoeia,
containing 33 per cent of real acetic acid. It is used for distinguish-
ing between calcium oxalate and calcium carbonate.

Alcohol. — Absolute alcohol is to be preferred, but 90 per cent
alcohol answers most purposes ; methylated spirit made w^ith wood
naphtha may also be used. It is employed for removing air from sec-
tions, and for dissolving resin, volatile oil, tannin, etc. Fats, waxes,
and, with the exception of castor oil, fixed oils are only sparingly
soluble in it.

Alkanet, Tincture of. —

Alkan^t root 20 grms.

Alcohol, 90 per cent 100 c.c.

Macerate for a week and filter.

Tincture of alkanna is much used as a staining agent for fats and
fixed oils. For this purpose it should be diluted with an equal volume
of distilled water immediately before use, and the powder left in con-
tact with it for about half an hour ; suberized cell walls wall also be
stained (compare Soudan glycerine).

Bismarck Broivn. — A very dilute, aqueous solution is used to
stain elements after maceration with potassium chlorate and nitric
acid, by which they are rendered very transparent. A saturated
aqueous solution is sometimes used as a stain for mucilage.

BrcBtner's Reagent. —

Sodium tungstate 1 grm.

Sodium acetate 2 grms.

Water to make 10 c.c.

Dissolve. This is one of the best reagents for tannin, with which
it produces a yellowish-brown precipitate.

Chloral Hydrate, Solution of. —

Chloral hydrate ....... 50 grms.

Water ......... 20 c.c.

Dissolve. The solution dissolves many of the commoner cell
contents, and hence is a most valuable clearing agent.

Chloral Iodine. — Solution of chloral hydrate saturated with
iodine, a few crystals of which should be kept in it. Useful for the
detection of minute starch grains.

MICKOSCOPICAL ANALYSIS

425

Chlorinated Soda, Solution of. —

Chlorinated lime 200 gnus.

Distilled water 1750 c.c.

Triturate the chlorinated lime with the water, added gradually,
transfer to a stoppered bottle and add

Sodium carbonate 250 grms.

dissolved in

Distilled water 750 c.c.

Shake together for four days and filter. To the filtrate add a 10 per
cent solution of potassium oxalate as long as a precipitate occurs ;
stand and filter.

The solution, which is used for bleaching, should be kept protected
from light.

Chlorzinciodine,. S&hdion of, —

Liquor Zinci Chloridi, B. P 175 c.c.

Evaporate to 100 ex. and add

Potassium iodide . . ... . . . 20 grms. ,

Iodide 0-5 grm.

Dissolve with a gentle heat and add to the hot solution of zinc
chloride ; let the mixture stand till cool. This reagent colours
cellulose ceH walls, blue or violet, lignified and suberized walls yellow
or brown ; it also swells starch grains and colours them blue.

Corailim. Siodia^ Solution of. —

SQdlum carbonate ....... 30 grms.

Distilled water 70 c.c.

Dissolve.. To a little of this solution, add a small fragment of
ciasiralllin. The reagent should have a bright pink, not wine-red, colour
aiadl should be freshly prepared. It is used for the detection of mucilage
aja.d sieve tubes.

Double-Stain, Cordonniers. —

Iodine green (Griibler's) 1 grm.

Chloroform 10 grms.

Alum-oarmine 1000 c.c.

Put them into a flask capable of holding 1200 c.c, in the order
named.. Shake till dissolved and filter.

Make the alum-carmine for the above as follows : —

Mix 1 grm. of carmine with 5 grrcs. of powdered alum and a
little distilled water. Evaporate to dryness at a gentle heat. Allow
the residue to stand twenty-four hours, dissolve it in 100 c.c. o^
cold distilled water and filter.

Ferric Chloride, Solution of. — A 1 per cent aqueous solution <^ii
ferric chloride ; it is frequently used as a reagent fgr tannin.

Glycerin. — Pure glycerin of specific gravity r-260.

Glycerin, Dilute. — Glycerin diluted with, ^.n equal yqh^m^ q^
distilled water.

426 FOOD AND DRUGS.

Hydrochloric Acid. — Pure hydrochloric acid of specific gravity 1-16.

Iodine Water. — Distilled water saturated with iodine, a few crystals
of which should be kept in the solution. Diluted solution of iodo-
potassium iodide is often used in its place. It is used for the detec-
tion of starch and aleurone grains.

lodopotassium Iodide, Solution of. —

Iodine . . . . . . . . . 2 grms.

Potassium iodide ....... 1 grm.

Distilled water 200 c.c.

Dissolve.

Maceration Mixture, Schulze's. — Potassium chlorate and nitric
acid ; the strength of the latter may be varied to suit the require-
ments of the case ; an acid of specific gravity 1'3 is very generally
useful.

Phloroglucin, Solution of. —

Phloroglucin 1 grm.

Alcohol (90 per cent) 100 c.c.

Dissolve. The solution gradually darkens and loses its power ; it
should not be kept more than three months. It is used in conjunc-
tion with hydrochloric acid for staining lignified cell walls.

Picric Acid. — A saturated aqueous solution is used to stain aleurone
grains.

Potash, Solution of. — A 5 per cent aqueous solution of potassium
hydroxide. It dissolves starch, proteid matter, tannin, etc., and is
largely used as a clearing agent.

Potash, Strong Solution of. — A 20 or even 50 per cent solution is
used to induce swelling of collapsed cell walls.

Buthenium Bed, Solution of {in Solution of Lead Acetate). — To a
few c.c. of a 10 per cent, aqueous solution of lead acetate, add sufficient
ruthenium red to produce a wine-red colour ; the solution should be
freshly prepared, as it will not keep long. The reagent is extremely
useful for the detection of mucilage.

Soudan Glycerin. —

Soudan red iii ....... O'Ol grm.

Alcohol (90 per cent) 5-00 c.c.

Dissolve and add glycerine 5*00 c.c.

The reagent colours suberized cell walls red, especially when
gently warmed, and hence is used to detect secretion cells (the walls
of which are commonly suberized). It also colours fixed and volatile
oils.

Sulphuric Acid. — Pure sulphuric acid of specific gravity 1-843.

PART II— DRUGS.
CHAPTER VIII.

CRUDE DRUGS AND CERTAIN GALENICALS.

In many cases the identification of drugs is a matter of botanical
knowledge, and when recognized, the form in which they exist often
precludes the possibility of adulteration. It is, in these cases, in the
form of powdered drugs, where one meets with sophistication.

It is true, however, that many " drugs " exist which although pure
may be practically useless for the purpose to which such drugs are
usually put. For example, samples of belladonna root may be met
with that contain so little alkaloid, that they may be practically use-
less.

An estimation of the active principle present is of course necessary
in such cases.

In the present section, apart from tables showing the ash content
of the principal crude drugs used, only those where adulteration is
practicable and probable will be dealt with.

In general the microscopic examination of a powdered drug, to-
gether with a determination of its ash content, will afford sufficient
information to decide on its purity or otherwise, except in those
cases, of course, where a reliable method exists for determining the
amount of a given active principle present.

The following table gives the ash limits for the principal drugs :
the figures are compiled from a large number of analyses by Umney
Moor, and the author. In all cases they are for the crude drug in
its original form. It is to be remembered, as pointed out by Umney
(" Pharm. Journ." 4, 15, 492), that there is a loss in weight on grind-
ing, especially in the case of such drugs as gum -resins, where volatile
oils will be lost, so that a slight allowance must be made for powdered
drugs. In the cases of barks, seeds, etc., this rarely amounts to much
and the figures here given would stand for most of the powdered
drugs. In the case of those containing volatile constituents, due
allowance must be made.

It is also to be noted that roots and rhizomes of a fibrous char-
acter are apt to have a comparatively large amount of soil adherent
to them, and if this is not removed by careful washing, a high ash
value follows : —

(427)

428

FOOD AND DEUGS.
Ash Standards.

Drut;.

Acacioe gummi
Aconiti radix ,
Aloe barbadensis
„ socotrina .
Ammoniacum .
AmyJum .
Anethi fructus
Anisi fructus
Anthemidis flores
Araroba .
Arnicee rhizoma
Asafoetida
Aurantii cortex
Belladonn8B radix
Benzoinum
Buchu folia
Calumbse radix
Cambogia
Cannabis indicia
Cantharis
Capsici fructus
Cardamomi sem
Carui fructus .
Caryophyllum
Cascarse sagradae cort.
CascarillsB cortex
Catechu .
Chirata
Cimicifug8Ba rhizoma
Cinchonse rubrse cortex
Cinnamomi cortex
Cocse folia
Coccus

Colchici cormus
Colchici semina

Colocynthidig pulpa

Conii folia
Conii fructus .
Coriandri fructus
Crocus

Cubebse fructus
Cusparise cortex
Cusso

Digitalis folia .
Elaterium
Ergota

Eucalypti gummi
Euonymi cortex

Ash Standard of B. P.. 1898.

Proposed Ash
Standard.

Not exceeding

Not exceeding 4 per cent

4 per cent

Not stated

6 „

>» »

3 „

„ „

4 „

„ ,,

7-5 „

»» ».

0-5 „

j> »>

„ ,,

8-5 „

'»

6 „

»> M

10 „

Not exceeding 10 per cent

20 „

Not stated

<> .»

8 „

2 „

».

5 „

»>

6 „

Not exceeding 3 per cent

Not stated

»• M

Not exceeding 6 per cent

Not exceeding 4 per cent

6-5 „

Not exceeding 8 per cent

Not stated

7-0 „

M »»

Not exceeding 5 per cent

Not stated

6 „

» n

10 „

»» M

Not exceeding 6 per cent

Not stated

Not exceeding 6 per cent

Not stated

« >.

Not less than

Not less than 9 per cent

10 per cent

Not exceeding

Not stated

„ „

,, „

Not exceeding 7 per cent

Not stated

" 'I

» »>

>» >»

>> M

6 „

M »»

0-7 „

10 „

CBUDE DRUGS.

429

Drug.

Ash Standard of B.P., 1898.

Proposed Ash
Standard.

Not exceeding

Filix-mas

Not stated

5 per cent

Fceniculi fructus

»» ?>

Galbanum

u »»

Galla .

Gelsemii radix

GentianaB radix

Glycyrrhizse radix .

Granati cortex

Guaiaci lignum

•» J»

Guaiaci resina

Hoeraatoxyli lignum

2 „

Hamamelidis cortex

Hamamelidis folia .

Hemidesmi radix .

4 „

Hydrastis rhizoma ,

Hyoscyami folia

Ipecacuanhse radix .

Jaborandi folia

7 „ !

Jalapa .

6-5 „ :

Kino

2 ,1 '

Kramerise triandree

radix .

)»

2 „ i

KraraeriaB argenteee

radix .

,, V

Limonis cortex

))

Linum .

Not exceeding 5 per cent

5 ,, i

Lobelia .

Not stated

Lupulinum

Not exceeding 12 per cent

14 „ i

Lupulus .

Not stated

7 „ \

Mezerei cortex

„ ,.

4 „ ;

Moschus .

Not exceeding 8 per cent

8 „ i

Myristica

Not statec

Myrrha .

M .,

Nux vomica

,, ,,

Opium

,, ,,

5 „

Papaveris capsules .

5> )»

Pareirse radix .

>> >.

Physostigmatis sem.

»5 M

4 „

Pimenta .

Piper nigrum .

V >>

Pix burgundica

,, „

Podophylli rhizoma

„ „

Pruni virginianae

cortex .

Pterocarpi lignum .

„ ,,

Pyrethri radix

,, ,,

Quassiae lignum

M »>

Quillaiae cortex

•

Rhei radix

Rhoeados petala

Rosae gallieee petala

„

Saccharum lactis .

Not exceeding 0-25 per cent

0-25 „

Sambuci fiores

Not stated

430

FOOD AND DKUGS.

Drug.

Ash Standard of B.P., 1898.

Proposed Ash
Standard.

Not exceeding

Sarsae radix

Not sta

ted

Sassafras radix

Scammoniae radix .

ScilJa .

Scoparii eacumina .

Senegse radix .

Senna alexandrina .

Senna indica .

Serpentariee rhizoma

Sinapis .

Staphisagriae sem. .

Strophanthi sem. .

Stramonii folia

Stramonii sem.

Styrax praeparatus .

0-5 „

Sambul radix .

Taraxaci radix

Tragacantha .

Uvae ursi folia

Valerianae rhizoma .

Zingiber .

ACACTA.

Gum acacia is officially described as a gummy exudation from the
stem and branches of Acacia Senegal and other species of Acacia.

It is described as being insoluble in alcohol, entirely soluble in
water, yielding a feebly acid solution. When dissolved in an equal
weight of water, the solution should neither form a glairy mucilage,
nor, after admixture with more water, should it give a gummy deposit
on standing. An aqueous solution forms an opaque jelly with lead
subacetate solution, and with borax solution a more or less translucent
white jelly. Its solution gives no precipitate with lead acetate sol-
ution, and is not coloured blue or brown by a small quantity of iodine
solution (absence of starch and dextrin), nor bluish-black by ferric
chloride solution. It does not reduce Fehling's solution, nor yield
more than 4 per cent of ash.

Gum acacia, or gum arable, as it is also frequently termed, con-
sists principally of the calcium salt of arable acid, which is also
present in combination with traces of magnesium and potassium.
The formula for arable acid is uncertain, but O Sullivan considers it
to be Cg(,Hi^.20-4 and that of its calcium salt, 0^,^11142^74^^0.

Arabic acid is obtained by dialysing an acidified solution of the
gum, the colloidal solution remaining in the dialyser being laevorotatory.

Gum acacia should have the properties ascribed to it by the
Pharmacopoeia, as mentioned above. It should contain from 10 to
13 per cent of water. Its solution should not yield more than a slight
precipitate with solution of mercuric chloride.

ACACIA.

431

According to Palladino, dextrin may be detected as follows : —
If an alkaline solution of gum is boiled for a minute with
aniline sulphate, the liquid remains pale yellow with a greenish tinge
in the absence of dextrin, but becomes orange-yellow^ or brownish-red
if dextrin is present. Other results are given in the following table,
in which (1) represents the specific gravity at 15° of solutions con-
taining 13*024 grms. of the gum in 100 c.c. ; (2) is the viscosity of
the same solution as compared with water ; (3) is the acidity in terms
of arabic acid ; (4) is the specific rotation, [a]^, at 16°.

Kordofan

1-0450

1-4166

6-29

- 26-47

Galam

1-0448

1-3333

7-23

+ 2-11

Salabreda .

1-0448

1-4166

8-18

+ 14-57

Bas du Fleuve

1-0450

1-5000

6-92

- 28-47

Arabic (Kordofan)

1-0454

1 -.3.833

6-92

- 23-02

Zula .

1-0448

1-1666

7-23

+ 12-84

Gheziri

1-0446

1-3333

9-75

+ 45-01

Amrad

1-0425

1-.S333

5-03

+ 71-81

Australia

1-0438

1-1666

5-03

+ 62-21

Cape .

1-0395

1-5000

7-86

+ 33-09

Suakim . . ,

1-0450

1-3333

10-06

-21-17

Turique

1-0450

1-5833

9-12

+ 34-41

Geddah

•

1-0449

1-4166

5-34

- 24-87

Guichard has examined the rotatory powers of the various acacia
gums in the market, and finds that they form three series : those of
Galam, Mogador, and Australia have a rotatory power near +16°;
Arabic, Aden, and Amrad gums border upon + 32°, whilst gum Ghatti
has a rotatory power close upon + 64°. The differences may be ex-
plained by the view that the gums are mixtures of several dextro-
rotatory and laevorotatory substances.

Inferior gums often contain a trace of reducing sugar, but any
notable amount would probably be due to added dextrin which is
sometimes added to powdered gum acacia.

Dextrin may be separated by dissolving the sample, concentrating
the liquid to a syrup, and precipitating this by means of 10 times its
volume of 90 per cent alcohol. One grm. of the dried precipitate is
then dissolved in 10 c.c. of water, the solution mixed with 30 c.c. of
60 per cent alcohol, and 4 drops of 25 per cent ferric chloride solution.
About 0-2 grm. of powdered chalk is then added and the whole
well stirred. The precipitate is washed with 60 per cent alcohol, and
the filtered liquid containing the dextrin is mixed with methylated spirit
and the precipitated dextrin allowed to stand for twenty-four hours.
The liquid is then decanted, the dextrin dissolved in a little water,
and the liquid filtered, if necessary, and evaporated, and the residue
weighed. To determine the arabin, the ferric chloride and chalk
precipitate is dissolved in a slight excess of hydrochloric acid, the
arabin precipitated by strong alcohol, washed with alcohol, then dis-
solved in water, the water evaporated, and the residue weighed.

432

FOOD AND DRUGS.

The following figures were obtained in samples of "gum ai^^^ic "'
from acacia and allied plants by the Technical Department o^i the
Imperial Institute : —

Gum
from

Gums from Garfung

Gumj
from,

(fWVQf
Apacia
i cqfra

Borgu.

Geldam

j\ano.

Borujt,

Per

?^r.

Per

cent

Q^qti

Qent

Moisture ,

14-5

14-0

17-8

17-4

1,5-Q,

i.7;7

Ash .

2-26

2-9

2-6

3-2

3.1.

2-6

Dry matter (soluble in

water ,

82-2

86-0

82-0

79-0

78-0

1 m'%

81-2

Reducing sugars

0-9

—

traces

1-2

traces

1 i^ii

—

Acidity {milligrams

KHO for 1 gram of

gum) ..

—

1-6

2-0

0-8

1 tfftfies

—

Relative viscosity of

10 per cent solution

220

21-0

14-2

21-2

22-5,

2^1 -8:

Colour of solution

pale

almost

pale

pal?,

pale

almost

yellow

colour-
less

brown

bro\y^ix

yellow

colour-
less

Tihree samples of gum from the Gold Coast Qplpny were ex-
amined, these gums being obtained from Acacia S^i^barifina, Burkea
Africana, and Pseudocedrela Kotchyi.

■Botanical Origin.

Acacia
Sieberiana.

Burkea,
Africana.

Pseudocedrela
Kotchyi.

Moisture ....

14-9

15-0$,

13-7

Ash

1-8

2-3

2-6

Insoluble matter .

7-02

0-9

0-35

Reducing sugars .

traces

—

considerable
traces

.Relative viscosity of 10 per

cent solutions

27-0

18-2

10-8

Acidity (milligrams KHO

per 1 gram gum)

3-6

3-92

3-1

Colour of solution .

yellowish-brown

dark browi^

yellowish-brown

and turbid

TRAGACANTS,

Tragacanth is officially described as 9^ gummy exu<Jation obtained
itom Astragalus gummifer, and other spegiea, kijQWft in commerce as
Syrian tragacanth.

TEAGACANTH. 433

It is officially stated that it is sparingly soluble in water, but
swells into a gelatinous mass which may be tinged blue or violet with
iodine.

There are numerous grades and varieties of tragacanth found in
commerce, but it is only the Syrian tragacanth — and only the
flattened flakes of that — that is official. Small masses of nondescript
shape constitute the "hog" tragacanth of commerce, which may or
may not be pure — but is not official.

The composition of tragacanth is very complex, but according to
O' Sullivan the portion soluble in water consists of a mixture of gum
acids. They belong to a series of poly-arabinan-trigalactane-geddic
acids, the chief of them being represented by the formula

VftgOg . 3C12H20O10

This has a specific rotation an = 88".

On hydrolysis these acids yield arabiriC3e and possibly galactose.
The insoluble portion consists of an acid body which is termed
bassorin.

Bassorin has not been obtained perfectly pure ; it is of an acid
nature, having the rotation av = + 98°. When treated with excess
of alkali it yields two acids, a- and y8-tragacanthan-xylan-bassoric
acid. The former Co^'H^fi.^QH.fi, is soluble in water, and has a
rotation aD= + 138'6° ; the latter is insoluble in water, and has a
rotation an = + 163° to 164°. Both acids, when hydrolysed with sul-
phuric acid, yield the same products, Tragacanthose and xylanbassoric
acid. Tragacanthose is laevorotatory, ao = - 30°, and is a pentose.
Xylanbassoric acid is strongly dextrorotatory, aD= +200°. The
hydrolysis is represented by the equation

^2 AeOgi + HgO = C5H10O5 + CigHggOi^.

Tragacanthose Xylanbassoric acid.

Xylanbassoric acid is almost insoluble in cold water, but its alkaline
salts are soluble ; when further hydrolysed it yields bassoric acid and
xylose, according to the equation

^lAsOiT + HgO = C5H10O5 + C14H20O13.

The last acid is insoluble in water ; the optical rotation of its alkaline
salts is found to be aD= + 225°.

A genuine tragacanth contains about 10 to 15 per cent of moisture,
and yields from 2"5 to 4 per cent of ash.

If it be allowed to stand with water for four hours and then heated
until a thick solution results and then diluted with a large volume
of water, tragacanth forms a ropy liquid which may be passed through
a filter, leaving a small amount of starch and cellulose. This liquid
should not be precipitated nor form a jelly, with borax solution,
sodium silicate or ferric salts.

Powdered acacia is often added to powdered tragacanth as an
adulterant. The sample should be rubbed into a cream with water,
and then diluted and vigorously shaken for a time, and then the
VOL. I. 28

434

FOOD AND DRUGS.

liquid filtered. In the presence of acacia, a precipitate or jelly results
on the addition of borax or sodium silicate solutions.

According to Scoville (" Druggist's Circular," March, 1909) gum
tragacanth in the form of powder is liable to be frequently adulterated
with the product known as " Indian gum ". This gum is generally
the product of either Sterculia urens, or Cochlospermum gossypium.
Scoville gives the following table of reactions of genuine tragacanth
and of the so-called Indian gum. He states, however, that so far as
the detection of the latter in mixtures with the former is concerned,
the only characteristic reactions are the acidity of the Indian gum
and its behaviour towards borax solution.

Test.

Tragacanth Solution
2 per cent.

Indian Gum Solution
2 per cent.

Appearance of solution.

Reaction to litmus.

Iodine test.

Solution heated with 5

per cent KOH.
Ferric chloride solution.
Subacetate of lead.
Alcohol (equal volume).
Alcohol (two volumes).
Heated with 2 per cent

HCl.
Borax solution.

Opaque, slimy and semi-
fluid.
Neutral.
Blue colour.
Froths on shaking.

Gelatinizes.
Precipitates in mass.
Precipitates slowly.
Precipitates at once.
Slight darkening.

No change in consistency
in 3 days.

Transparent, non-adhesive

jelly.
Acid.

No reaction.
No frothing on shaking .

Slightly hardens.
Precipitates in clots.
Clear mixture.
Precipitates slowly.
Red-brown colour.

Becomes slimy and tacky
with marked stringing
when poured from the
vessel.

The acidity test is certainly not very reliable and should be ig-
nored unless the reaction be verj'' markedly acid. The borax test is
certainly indicative of adulteration, but the nature of the adulterant
is not decided by it. It is best carried out as follows : 2 grms. of
powdered gum are shaken with 100 c.c. of water until quite free
from lumps. If the powder is first moistened with 3 c.c. of alcohol,
and the water added quickly, the semi-solution is prepared more
rapidly. Two grms. of powdered borax are then added and the
mixture shaken until the borax is dissolved. The mixture is allowed
to stand overnight. Pure tragacanth will not have altered, except
by a slight darkening in colour. In the presence of the Indian gum,
the liquid will have lost its transparency and have become more or less
slimy and tacky according to the amount of adulterant present.

AMMONIACUM.

This gum resin is the product of Dorema ammoniacum and pro-
bably other species, and is official in the Pharmacopoeia. |o
The following tests are given : The freshly fractured surface is

AMMONIACUM.

435

coloured yellow by solution of potash, and dark red or orange by solu-
tion of sodium hypochlorite. If a small fragment be strongly heated
in a test tube, and the contents of the tube, after cooling, boiled with
water, the resulting solution when largely diluted with water and
rendered alkaline with ammonia does not exhibit a blue fluorescence
(absence of galbanum and asafoetida). The last described reaction is
due to the characteristic fluorescent nature of umbelliferone (see
under galbanum). Ammoniacum of commerce consists of about 60
to 70 per cent of true resinous matter, which is a mixture of esters of
ammoresinotannol C^gHgoOg (in which the salicylic ester is predomin-
ant), and of free resin acids. Traces of free salicylic acid and of es-
sential oil are present and from 10 to 25 per cent of gum, which is
similar in properties to gum acacia.

The ash of ammoniacum varies from 2*5 to 7*5 per cent, the latter
being the highest that should be allowed.

The following are the most reliabrle analyses of good commercial
ammoniacum : —

Constituents.

Plugge.

Buchholz.

Braconnet.

Moss.

Hirschsohn.

Per cent

Ethereal oil .

1-27

}.e

1-43 to 6-68

Moisture

5-10

[ 4-0

[ 7-2

0-81,, 3-27

Ash . . .

2-00

2-3

2-02 „ 16-88

Kesin .

65-53

72-0

700

68-6

47-12 „ 69-22

Gum

2610

22-4

18-4

19-3

—

Bassorin

1-6

—

Gelatinous sub-

stances

4-4

5-4

Extractives .

1-6

Sugar, etc. .

—

1-61 to 4 -59

Per cent sol. in

water

—

11-85 „ 25-74

Residue

—

0-81 „ 3-09

A genuine ammoniacum should respond to the reactions of the
British Pharmacopoeia and should not contain more than 7*5 per cent
of mineral matter, nor more than 40 per cent matter insoluble in
90 per cent alcojiol. Good samples will often contain as little as 15
per cent to 25 per cent of matter insoluble in alcohol.

The resin extracted by alcohol from ammoniacum should have an
acid value between 70 and 100, and an ester value of 50 to 100.
These limits are occasionally exceeded.

The gum may be determined by dissolving 2 grms. of the sample
in 15 c.c. of a 60 per cent chloral hydrate solution, and filtering this
into 100 c.c. of alcohol, which precipitates the gum, which is collected
washed with alcohol, dried and weighed.

436 FOOD AND DEUGS.

ARAEOBA.

This drug, also known as Goa powder, is usually imported in its
crude form, not powdered. It usually contains a good deal of woody
fibre which is directed by the British Pharmacopoeia to be picked out
as far as possible. When powdered it forms a brown to umber-coloured
powder containing a large proportion of chrysarobin, but no chryso-
phanic acid.

Most samples of this drug are sold for the purpose of manufactur-
ing chrysarobin, so that the only point of importance to determine is
the amount of that acid present. This is often as high as 65 per
cent in the crude drug.

No limits for foreign matter are given in the Pharmacopoeia, and
if a sample contains 50 per cent of chrysarobin it will correspond
with the ofi&cial requirements.

Commercial samples vary in their content of mineral matter very
greatly. On an average of thirty samples of parcels as imported into
London, the author has found the mineral matter to vary from 7 per
cent to 28 per cent (values of 80 per cent given by Pearmain are
obviously for grossly adulterated samples), and the chrysophanic acid
from 42 per cent to 69 per cent.

In order to determine the ash and chrysarobin, a large amount of
the sample should be finely ground, as the size of the fragments of
woody tissue are so variable as to make it difficult to get a repre-
sentative sample. Two grms. should be ignited and the residue
weighed.

For the determination of the chrysarobin, 2 grms. to 3 grms, may
be extracted with chloroform in a Soxhlet tube. The resulting
chrysarobin, obtained by evaporation of the chloroform, should answer
to the following characters : it should leave not more than 1 per cent
of ash when incinerated, and should be almost entirely soluble in 90
per cent alcohol.

Jowett and Potter find " chrysarobin " to contain the following
compounds (" Jour. Chem. Soc." 81, 1575) : —

Chrysarobin CjjHjgOg is the anthranole of chrysophanic acid and
identical with chrysophanohydroanthrone obtained by the reduction
of chrysophanic acid. It melts at 204° C. When acetylated with
acetic anhydride alone, a mixture of diacetylchrysarobin (m.p. 193°)
and triacetylchrysarobin (m.p. 238° C.) is obtained, but if sodium
acetate and acetic anhydride are used, the triacetyl compound is alone
produced.

Methyl ether of dichrysarobin CgiHgeO^j melts at 160° C. It
yields a soluble pentacetyl compound (m.p. 135° C), identical with
Hesse's hexacetyldichrysarobin.

Dichrysarobin C^oH^fii does not melt below 250° C, but blackens
and chars gradually. On acetylation, hexacetyldichrysarobin (m.p.
179° to 181° C.) is obtained.

A substance C^jB.^fi^ (m.p. 181° C), which yields an acetyl com-
pound (m.p. 215° to 216° C).

ASAFCETIDA. 437

AS/VFCETIDA.

Asafoetida is a gum resin obtained by the incision of the roots of
Ferula fcetida, and other species.

It contains the following substances : asaresinotannol C24H33O4OH,
and its esters ; gum ; essential oil of a foul-smelling nature ; traces
of vanillin, and mineral matter and woody fibre.

It is an evil-smelling substance, occurring in masses or in tears —
the latter being the purer variety. The British Pharmacopoeia re-
quires it to contain not less than 65 per cent of matter soluble in 90
per cent alcohol, and not more than 10 per cent of ash. According
to most observers the ash value should be 20 per cent, as the great ma-
jority of commercial samples fail to satisfy the Pharmacopoeial re-
quirements. However, some samples occurring in tears are practically
pure glim resin, and will satisfy the above requirements.

Of twenty samples examined by the author, the ash varied from
16 per cent to 55 per cent, and the amount soluble in 90 per cent
alcohol from 31 per cent to 68 per cent. The examination of asafoetida
should include the determination of the mineral matter and of the
resinous matter extracted by 90 per cent alcohol. The petroleum
ether extract should not exceed 7 per cent. The presence of colo-
phony would be indicated by a high solubility value in petroleum
ether.

The pure resinous matter extracted with alcohol should have the
following characters : —

Acid value ..... Rarely exceeds 65
Ester value .... Rarely below 150

The figures are rather variable, but a high acid and low ester value
indicates the presence of colophony ; this, however, should be con-
firmed by the solubility in petroleum ether and by the transient violet
coloration produced by carefully pouring a few drops of 50 per cent
sulphuric acid on to a solution of the resin in acetic anhydride.

Puckner has published ("Proc. American Pharm. Assn." 1890) the
analyses on page 438 which give the characters of the ash of
asafoetida. Sample No. 5 was probably grossly adulterated with
siliceous matter.

Tincture of Asafoetida is an extract of 4 ounces of the drug with 20
fluid ounces of 70 per cent alcohol. There is obviously no official
standard of extractive matter in this tincture, since the 65 per cent
of resinous matter, etc., present in the official drug is not necessarily
extracted by 70 per cent alcohol. Tinctures made from the drug
having the official characters gave the following results on analysis : —

Specific gravity . . • . . 0-910 to 0-918

Solid residue 9 ,,10 per cent

Alcohol (by volume) . . . . 60 „ 63 „

Commercial samples, being made with low-grade asafoetida, are
very frequently below the proper standard of solid matter. It is not
clear why this tincture is not made with 90 per cent alcohol.

438

FOOD AND DKUGS.

-d ^

-c

8
<

,fl

-a

«'2

.^ 0

•2

"§, >>

?•

a °8

6 ,

Per cent

Per cent Per cent

Per cent

Per cent Per cent

No. 1 in mass .

59-49

19-45

2-32

1-16

7-12

1-14

8-57

1-09

No. 2

27-39

56-03

1-37

-42

25 07

2-03

20-49

10-78

No. 3 powdered.

44-48

3859

2-36

•60

16-57

1-39

16-18

5-32

No. 4

33-47

47-86

1-95

-60

18-85

•42

23-51

2-62

No. 5 „ .

31-35

55-38

21-96

2-57

9-91

6-49

8-96

4-25

Essential Oil of Asafceiida. — This is a foul-smelling oil, obtained
to the extent of 4 per cent to 10 per cent from fair-grade samples of
asafoetida. It has a specific gravity -975 to -990 and optical rotation
about - 10°. Semmler has investigated this oil, having separated by
fractional distillation under reduced pressure two terpenes, one of
which was probably pinene, and a sesquiterpene, which had a
lavender-like odour. The remainder of the oil consists chiefly of
compounds containing sulphur. According to Brannt, the oil con-
tains allyl sulphide and allyl disulphide, but Semmler denies this.
Sulphur compounds of the formulae C7HJ4S2, CjoHgoSg, CgH^gSg, and
CioHjgSg were found, together with an oxygenated body of the formula
CjoHigO, or a multiple of this.

A very minute trace of the oil, or of an alcoholic extract of the
gum resin, is present in several of the best-known sauces made in
England and America.

BALSAM OF PEEU.

This balsam is exuded from the trunk of Myroxylon PereircB.
The British Pharmacopoeia gives the following tests for this drug.
It is soluble in chloroform ; 1 volume is soluble in 1 volume of 90
per cent alcohol, but on the further addition of 2 or more volumes, the
liquid becomes turbid. Specific gravity 1-137 to 1-150, 10 drops
rubbed with 0"4 grm. of lime produces a soft mixture (absence of
copaiba and " resins " (?) ). It should not diminish in volume when
shaken with an equal volume of water (absence of alcohol). About 40
per cent of resin should separate on the addition of three times its
volume of CSg, and the clear supernatant liquid should not have more
than a pale brown colour and slight fluorescence (absence of gurjun
balsam). Five grms. shaken with 5 c.c. of a solution of NaOH of
specific gravity 1-16, and then washed with three successive quantities

BALSAM OF TOLU. 439

of 15 c.c. of ether, and the ether cautiously evaporated, should give
a residue weighing from 2-85 grms. to 3 grms. This residue should
require from 11-9 c.c. to 12-8 c.c. of normal alkali for saponification
(presence of a due proportion of cinnamein).

The principal constituents of Peru balsam are the esters benzyl
cinnamate (cinnamein) and cinnamyl cinnamate (styracin). A little
free benzyl alcohol is also present, and traces of vanillin and free
cinnamic acid. The cinnamic and benzoic esters of an alcohol
peruresinotannol C^gHgoOj^ are also present.

Balsam of Peru is a black, viscous liquid and has an aromatic,
agreeable odour, and being an expensive product is frequently adulter-
ated. The principal adulterants are colophony, fatty oils, storax,
copaiba (rarely met with now, however) and the so-called synthetic
balsam of Peru.

The last named is a thick blaqk liquid, closely resembling the
natural balsam, made from synthetically prepared esters with various
oleo-resinous matters in order to give it the proper consistency.

In addition to the Pharmacopoeial tests, the balsam should be ex-
amined as follows : —

The acid and ester values should be determined. The acid value
is usually from 70 to 80, and the ester value 180 to 200. A high
acid value and low ester value indicates the presence of colophony or
copaiba.

Solubility in alcohol. The balsam is practically entirely soluble
in 90 per cent alcohol. Fatty oils will be detected by the insolubility
of the sample. The iodine value varies from 38 to 45.

The refractive index at 20° varies between 1*585 and 1*595.

The copper acetate and petroleum ether test described under
balsam of tolu is not altogether reliable. A faint green colour is often
produced by pure balsam, but a brilliant emerald green is strongly
indicative of the presence of colophony.

If the balsam be extracted with petroleum ether and the residue
obtained by evaporation of the solvent be treated with a drop of HNO3, a
marked green coloration is indicative of the presence of colophony.

BALSAM OF TOLU.

Balsam of Tolu is obtained by making incisions in the trunk of
Myroxylon toluifera. It is an aromatic balsam, oflBcial in the British
Pharmacopoeia, which authority describes it as a soft, tenacious solid
becoming harder on keeping and, in cold weather, brittle. Pressed be-
tween pieces of glass by the aid of heat, and examined with a lens,
it exhibits an abundance of crystals. It is soluble in alcohol, the
solution having an acid reaction. If 5 grms. are warmed with two
successive portions of 25 c.c. and 10 c.c. of CSg, the solutions should
yield, when evaporated, a distinctly crystalline residue which should
require not less than one-third of its weight of KOH for its saponi-
fication (presence of a due proportion of benzoates and cinnamates).
No other official tests are given.

440 FOOD AND DEUGS.

Balsam of Tolu contains benzyl benzoate, benzyl cinnamate,
cinnamic and benzoic acids, traces of vanillin and the benzoic and
cinnamic esters of an alcohol, toluresinotannol, Ci,5Hj^030CH30H.

The principal adulterant of balsam of Tolu is colophony, but other
resinous matter is sometimes added, and occasionally balsam already
exhausted by water. The examination of this drug should embrace,
in addition to the Pharmacopoeial tests, the determination of its solu-
bility in vadous solvents, and the acid and ester values. The fol-
lowing are the minimum solubilities of a genuine balsam : —

Per cent
In 90 per cent alcohol .... 90

,, chloroform 95

„ petroleum ether 2 to 8

The acid value of the balsam varies from about 105 to 140 rarely
up to 150 ; and the ester value from 35 to 70. In the presence of
colophony or exhausted tolu balsam, the ester value is lowered.

The presence of colophony is confirmed by the following test : 5
grms. of the sample are exhausted with petroleum ether (preferably
by first rubbing down the balsam with a little CSg to make it viscous),
and the filtered petroleum solution is shaken with an equal volume
of O'l per cent aqueous solution of copper acetate. Copper abietate
is soluble in petroleum ether, and therefore if colophony be present,
the petroleum solution will show a marked emerald green colour.
In the author's experience, pure samples will never give more than
the faintest green tint under these conditions, 2 per cent of colophony
giving a marked green colour.

Tincture of Tolu. — This drug is official, being a solution of two
ounces of balsam in sufficient 90 per cent alcohol to produce 20 fluid
ounces. It should have the following characters : —

Specific gravity .... 0-862 to 0-870

Solid residue .... 8-5 „ 9 grms. per 100 c.c.

Alcohol by volume ... 81 „ 84 „

BENZOIN.

Although there are many varieties of benzoin or gum Benjamin,
as it is often termed, found in commerce, the official drug in the
British Pharmacopoeia is restricted to the products known as Siam
and Sumatra benzoin. This is probably not the intention, however,
of the compilers, since the words " and probably from other species
of styrax " are used, which would appear to allow the use of Penang
and other benzoins. The only tests given in the British Pharma-
copoeia are that the drug is to be almost entirely soluble in 90 per
cent alcohol and in solution of potassium hydroxide.

The British Pharmaceutical Codex states that the ash should not
exceed 2 per cent, nor the matter insoluble in 90 per cent alcohol
more than 10 per cent.

Benzoin occurs in commerce in tears, lumps or blocks, according
to quality and always contain some bark and mechanical impurities.

BENZOIN.

441

As this drug is used for the manufacture of " natural " benzoic acid,
the proportion of that acid present is a matter of importance. For
the manufacture of the tinctures (compound and simple) the percent-
age of matter soluble in 90 per cent alcohol is the principal considera-
tion.

Ash. — The ash will naturally rise with the amount of woody fibre,
bark, etc., present in the resin. Samples containing more than 2 or at
most 2-5 per cent of mineral matter must be regarded as. of inferior
quality, and although no standard for this exists officially, this figure
must be taken into account in judging the quality of samples.

Solubility in Alcohol. — Not less than 90 per cent should be soluble
in 90 per cent alcohol. Solubility tests with other solvents afford no
useful information.

Acid and Ester Values. — These figures should be determined, as the
presence, of other resinous matter is indicated by any wide variations
from the following limits, which represent the analyses of 12 samples
of benzoin of each type named, obtained from reliable sources, and
having the ash and solubility values above mentioned : —

Siani Benzoins.

Sumatra Benzoins.

Other Varieties.

Ash .

Soluble in 90 per

cent alcohol
Acid value
Ester value

0-24 to 1-77 p.c.

90 „ 96 „
130 „ 158
42 „ 68

0-4 to 1-82 p.c.

91 „ 93-5 „
98 „ 133-

58 „ 98-

0-4 to 2-23 p.c.

90-5 „ 93-9 „
106- „ 135-
60- ., 89-

Siam benzoin (made into a tincture and diluted with water) has a
characteristic odour of vanilla, Sumatra benzoin rather recalls that of
a mixture of styrax and vanilla, whilst Penang has an odour allied to
that of styrax.

Estimation of Benzoic Acid. — Benzoin contains from 12 per cent to
20 per cent of benzoic acid, sometimes as much as 22 per cent. It
may be approximately determined by powdering the sample, mixing
it with twice its weight of sand, and heating it in a beaker covered
with a perforated filter paper. The benzoic acid sublimes and
may be condensed in a porcelain or other cone kept well over the
top of the beaker, and kept as cool as possible by any suitable
means.

Tincture of Benzoin (conijwund). — This tincture, known also as
friar's balsam, is prepared with 90 per cent alcohol, by extracting
benzoin, storax, balsam of tolu and socotrine aloes.

It is obvious that such a complex mixture allows of no separa-
tion of its ingredients that will give any approximate quantitative
results.

The specific gravity of properly prepared compound tincture of
benzoin varies between 0-890 and 0-904. It should contain 75 per cent
of alcohol, and not less than 18 grms. of solid resinous matter per

442 FOOD AND DRUGS.

100 c.c. This is not a standard given by the British Pharmacopoeia,
but is based on the fair average values of the soUible matter in the
drugs employed for the manufacture of the tincture. Samples con-
taining 17 per cent of residue would probably not be taken excep-
tion to, but a tincture yielding anything below this must be regarded
as having been carelessly prepared, or made from drugs containing
too little soluble matter.

From the description of benzoin in the Pharmacopoeia, it should
certainly not -contain under 90 per cent of soluble matter. Purified
styrax never contains less than 90 per cent of soluble matter ; the same
is true for balsam of tolu, and socotrine aloes contains from 80 per
cent to 86 per cent of matter soluble in 90 per cent alcohol. The
use of low-grade benzoin containing 60 per cent to 75 per cent of
soluble matter, and much mechanical impurities, or of crude storax,
containing less than 60 per cent of soluble matter is often the cause
of the production of an inferior tincture. It is to be remembered
that a certain amount of volatile solid matter is lost when drying the
tincture, but this is allowed for by the above comparative figures
which were all obtained by heating the extract of the drug to con-
stant weight. Dowzard (" Chemist and Druggist," 20 Feb. 1904)
recommends drying with about 10 per cent of the weight of the
tincture of freshly ignited magnesium oxide. This fixes some of the
volatile matter, and gives a result about 2 per cent higher than when
dried without such addition.

CANNABIS INDICA.

Cannabis Indica, the dried flowering or fruiting tops of the female
plant of Cannabis sativa, is official in the Pharmacopoeia. It is used
for the preparation of a tincture and an extract, both of which are
official. No official standards are given.

This drug, which is known as guaza (Bombay variety) or ganjah
(Bengal variety) is closely related to " bhang " or " hashish " of the
native Indians. This latter is the dried leaf of the plant, whilst the
" charas " or " churrus " is the resin extracted by heating the plants
in a cloth.

The narcotic effect of this drug is produced by a resinous sub-
stance known as cannabinone. The principal constituent of this oleo-
resinous matter is cannabinol, a dark-coloured oil of the formula
C18H24O2, boiling at 265° and of specific gravity 1*0424. Traces of a
laBvorotatory terpene (probably an olefinic terpene) and a sesqui-
terpene (cannabinene) are also present, and a minute quantity of the
alkaloid choline.

Cannabis indica yields from 14 per cent to 17 per cent of mineral
matter.

The following table represents the (1) amount of extract with 90
per cent alcohol, (2) the same washed with water, (3) ether-alcohol-
soluble resins, (4) per cent of (1) soluble in water, (5) moisture, in a
number of type samples examined by David Hooper : —

CANNABIS INDICA.

448

Ether-

Per cent

Rectified
Spirit

Washed
Spirit

and
Spirit-

of Spirit
Extract

Moisture
in Ganjah.

Extract

Extract.

Soluble

Resins.

in Water.

Per cent

Bengal, Navagon . . .1

23-6

21-2

21-8

10-1

9-0

„ • . . 2

221

20-4

201

7-6

7-1

„ . . . 3

211

19-5

18-8

7-6

6-7

» . . . 4

19-8

18-1

18-4

8-5

9-2

Bombay, Sholapur (exported)

20 9

19-4

20-1

7-1

„ Khandesh

18-0

16-5

16-8

8-3

7-4

„ Satara .

17-8

16-6

7-0

8-9

Independent State, Hydera-

bad ....

17-7

16-8

16-5

7-6

N.W. Provinces, Basti .

17-2

* 15-8

108

Central Provinces, Nimar . 1

16-7

150

15-6

10-2

8-4

• 2

151

13-7

14-4

9-2

8-6

Bombay, Ahmednagar . . 1

16-7

15-4

7-7

12-4

• 2

16-2

15-2

14-9

6-1

11-6

• . 3

14-6

13-4

13-3

8-2

9-8

Nasik .

16-8

14-3

14-8

8-2

„ Sholapur

14-8

140

13-9

5-4

8-4

N.W. Provinces, Ghazipur .

17-1

137

13-8

19-8

100

Sind

16-3

13-9

14-7

8-4

Bombay, Surat

15-6

13-4

141

14-3

10-0

„ Bijapur .

14-5

13-4

7-5

9-2

Madras, Kistna Dist. .

81-0

24-0

23-4

22-5

7-6

„ Ootacamund .

28-1

20-8

20-1

25-9

9-8

„ Ganjam .

23-7

18-0

17-6

24-0

10-3

„ Bangalore

21-6

170

17-3

21-3

8-2

„ Tanjore .

241

15-9

16-1

34-0

9-7

„ Madras City .

19-4

130

13-2

32-9

7-9

In the above table the second column represents the amount in
the first column washed with hot water, and then dried and weighed.
The fourth column expresses the per cent of the amount in column
No. 1 dissolved by the water. The third column is the amount of
resin directly extracted from the air-dried drug with ether, and then
extracting the alcoholic extract with ether, leaving the resin soluble
in alcohol but not in ether ; this residue is added to the direct ether
extract and should correspond closely with the washed alcohol ex-
tract (column No. 2).

Extract of Cannabis Indica is the 90 per cent alcohol extract dried
to the consistence of a soft extract. Well-made extracts have the
following average characters : —

Water

Ash

Water-soluble extract

Soluble in 90 per cent alcohol

Per cent
4 to 8
1-5 „ 3-5
6 „ 15
practically complete

Tincture of Cannabis Indica should have a specific gravity of 0-845
to 0'850 ; a solid residue of 3-5 grms. to 4-2 grms. of solid matters per

444 FOOD AND DEUGS.

100 c.c. ; and should contain 85 per cent to 87 per cent of alcohol by
Tolume.

CATECHU.

The official variety of this drug is the light-coloured extract of
the leaves and grey shoots of Uncaria gambler. It occurs in com-
merce under the name of gambier, in small cubes about two-thirds of
an inch in measurement each V7ay. The official standards are that
when examined under the microscope it will be found to consist
chiefly of minute acicular crystals. It is almost entirely soluble
in boiling water. At least 70 per cent should be soluble in 90 per cent
-alcohol. No reaction should be given for starch, and it should not
yield more than 5 per cent of ash.

(Black catechu or cutch is an extract from the heart wood of Acacia
oatechu.)

Catechu consists of 10 to 30 per cent of a body called catechin,
which is probably a phloroglucide of tetrahydro-protocatechuic acid,
and 30 to 50 per cent of catechu-tannic acid.

Commercial catechu should contain from 8 to 10 per cent of
moisture and from 3 to 5 per cent of mineral matter. Genuine
samples should answer the requirements of the Pharmacopoeia as
given above, and should yield from 30 to 50 per cent of tannic acid
when determined by Lowenthal's permanganate process (see p. 11).
At least 45 per cent should be soluble in ether.

Catechu is sometimes adulterated with starch, which is detected
by the iodine reaction with an aqueous decoction of the sample, and
by a microscopic examination.

Chalk and calcium sulphate are common adulterants, up to 30
per cent being found in some samples. These are found in the usual
manner in the ash.

The difference between gambier and cutch is indicated by a
fluorescence test as suggested by Dieterich (" Pharm. Central. H."
1896, 855). Three grms. of catechu are dissolved in 25 c.c. of normal
caustic alkali and 100 c.c. of water. Fifty c.c. of petroleum ether are
added and the mixture well shaken. With pale catechu or gambier,
the petroleum shows a green fluorescence, but with acacia cutch no
fluorescence is shown.

Tincture of Catechu. — The characters of this tincture are given
in the table on page 495.

CAEDAMOMS.

The seeds of Elettaria cardamomum are official under the name of
cardamom seeds. The Pharmacopoeia describes the pericarps of the
fruits and states that the seeds shall be kept in them until they are re-
quired for use.

A description of the fruits is given, but the only standard is that
the seeds should not yield more than 4 per cent of ash.

CARDAMOMS,

445

According to the " Chemist and Druggist," (Diary, 1899, 500) the
principal varieties imported are the following : —

My sores. — Divided into rounds and longs. The former are what
the B.P. calls " ovoid " ; they vary in length from a quarter of an inch
to four-fifths of an inch (the latter 1 in 10), and have a smooth peri-
carp of a cream colour, due to the use of bleaching agents. Their
quality is judged by their weight. Sometimes the seeds are shrivelled
(unripe), so that the fruit is husky. This is not so frequent in the
longs, which are simply thinner than the rounds, and are not so
smooth on the surface, nor so pale, as a rule. The B.P. description,
" longitudinally striated," might exclude most of the rounds, as they
look smooth until closely examined.

Malabars, — These are smaller than Mysores, and there is a greater
proportion of seed to pericarp in them. They are fat pods, with a
pointed apex. Generally pale-brown 'or pink and longitudinally
striated. • Rarely more than half an inch long. They have a full
flavour.

Mangalores. — These are almost globular in shape and not unlike
Malabars. All three are washed or bleached before exportation.

In addition to these there are the so-called " Ceylon Wilds " which
are probably derived from another species.

It is clear that Mysore cardamoms are those usually employed,
but the Pharmacopoeial description would certainly allow the use of
Malabar cardamoms also.

The only methods of examination of cardamom fruits or seeds are
the determination of the ash value, a microscopic examination, and
the estimation of the essential oil.

The ash of cardamoms is usually, in the case of Mysore seeds, well
within the official requirements, but in the case of Malabar seeds, it is
often as high as 8 per cent to 9 per cent, so that in such cases Mala-
bar seeds would be excluded. According to Cowley and Catford,
the following figures are average ones for the three varieties men-
tioned : —

Variety.

Malabar.

Mysore.

Mangalore.

Number of fruits in 10 grms. .
Percentage proportion of pericarp .

Percentage proportion of seed .

Percentage of ash from dark seed .
Percentage of ash from light seed .
Percentage of ash from pericarp

„^ I dark, 57

'" i light, 13

5-0

8-5 to 9

13-0

55
25

3-3
4-5

7-1

45
20

2-9

7-6

Greenish (" Pharm. Journ." 4, xii. 264-393) is of opinion that
the minimum limit for ash for the seeds should be 5*5 per cent, and
states that the ash of the pericarp is so near this figure that the ash
limit will not discriminate between the powdered seeds and the
powdered whole fruits.

446

FOOD AND DRUGS.

In the author's experience this is hardly the case, and the limits
for ash for the seeds should be 6 per cent as a maximum, the peri-
carps yielding as much as 8*5 to 10-5 per cent.

Cardamom seeds should yield at least 3 per cent usually up to 4*3
per cent of essential oil when steam-distilled and the separated oil
measured.

The characters of the essential oils yielded by various cardamoms
are a matter of some uncertainty, but the author has examined the
subject to some extent, and the following is the outcome of the exam-
ination of samples of reliable origin. A good deal of genume oil
distilled from good cardamoms, however, has an optical rotation of
about +30".

Sp. gr. at 15-5°.

Optical Rotation

at 16°
(100 mm. tube.)

Oil of Malabar cardamoms
Oil of Mysore cardamoms

0-9418
0-9418

+ 40° 41'
+ 46° 39'

These figures are in fair agreement with those given for Malabar
oil, but in no way resemble those quoted by Schimmel for Ceylon oil.

The oils were soluble with a slight opacity in 40 to 45 volumes of
60 per cent alcohol.

There is little difference between the two oils. On distillation at
ordinary pressure, the oil, which is very rich in esters, in both cases
decomposes partially, and a considerable quantity of free acid distils
over. According to Weber ("Annalen," 238, 89), formic and acetic
acids are found in the distillate. Acetic acid is undoubtedly the chief
acid constituent of the esters, but the presence of formic acid could
not be confirmed. If it is present, it is only in faint traces. On dis-
tillation under reduced pressure the earlier fractions (the boiling-point
rises gradually until 50 per cent has distilled over) contain cineol, but
only to the extent of 5 to 10 per cent of the oil. This figure is the
result of an approximate estimation by means of phosphoric acid.
The earlier fractions also contain one or more terpenes, amongst
which is limonene. Weber states that terpinine is also present, but
this is doubtful, nor could Schimmel find it in Malabar oil ; and as it
easily forms a well-defined nitrite when present, it cannot exist in an
appreciable quantity. A small quantity of terpineol is present in both
oils, and is easily identified by its phenyl-urethane. The terpineol
comes over with the fraction obtained at 160° to 170° C. at 18 mm. The
nature of the alcoholic constituent of the greater part of the esters
requires further elucidation.

The following description is sufficient for the recognition of the
pericarp in the powdered drug, as is recommended by Greenish to be
included in the next edition of the Pharmacopoeia : —

" Powdered cardamoms, when examined under the microscope,

COPAIBA.

447

should exhibit masses of thin-walled parenchymatous cells packed with
minute starch grains ; long straight epidermal cells with moderately
thick walls, and small polygonal reddish-brown cells with very thick
walls. It should be free from sclerenchymatous fibres or elon-
gated cells, or small cells containing brown resin."

Compound Tincture of Cardamoms. — The characters of this offi-
cial tincture are given in the table on page 495.

Fig. 40. — Powdered Cardamoms.

It is also to be noted that this tincture is optically active, on ac-
count of the sugar present which is derived from the raisins used in
its preparation. A genuine sample should always be laevorotatory to
the extent of - 2° 10' to - 2° 40', when cleared as described under
sugars and examined in a 100 mm. tube. Dextrorotatory samples are
always prepared with cane sugar to save the use of raisins, and should
be condemned.

COPAIBA.

Copaiba, or balsam of copaiba (or capivi) is described in the British
Pharmacopoeia as an oleo-resin obtained from various species of
Copaifera.

It is a mixture of resins and an essential oil, the Pharmacopoeial
standards being as follows : Specific gravity 0*916 to 0993. It should

448 FOOD AND DEUGS.

yield at least 40 per cent of essential oil leaving a hard friable residue.
The essential oil is described as rotating the plane of polarization from
- 28° to - 34° (100 mm. tube). The essential oil should not boil
under 250° C. If 2 drops are dissolved in 20 parts of CS^ and a drop
of a cooled mixture of equal parts of nitric and sulphuric acid added,
a transient violet colour is not produced (absence of gurjun balsam).
If 4 drops be added to a mixture of half an ounce of glacial acetic
acid and 4 drops of nitric acid, a reddish or purple-red colour should
not result (absence of gurjun balsam).

This monograph is totally inadequate, and has been to some extent
corrected by that in the Pharmaceutical Codex. Before discussing
this important drug in detail it is necessary to call attention to the
fact that the optical rotation of the British Pharmacopoeia, besides
being erroneous in itself, is for a tube of 200 mm., which fact was in-
advertently omitted by the compiler of the monograph. The rotation
limits thus corrected are far too narrow, and should read from - 7°
to - 35°, or thereabouts. Further, the test with nitric and acetic
acids, will not reveal the presence of gurjun balsam unless it be
present to a fairly large extent : whereas if the test be applied to the
essential oil instead of the balsam, it will be found to be very delicate.
Oil of copaiba is also official, its specific gravity being given in the
Pharmacopoeia as 0*900 to 0'910.

The British Pharmaceutical Codex states that the acid value of the
thick balsams varies from 77 to 83 and the ester number does not
exceed 10. These figures are a little too stringent.

There are numerous varieties of copaiba found in commerce, the
principal of which are Maranham, Maracaibo, Cartagena, Bahia and
Para balsams. In addition to these the following are sometimes met
with : Surinam, Angostura, Maturin and British Guiana balsams.

Copaiba has for many years been subject to gross adulteration.
Fatty oils and turpentine were at one time met with, but in the course
of the last ten years the author has examined several hundred samples^
and has not met with any adulteration other than with either gurjun
balsam or the so-called African copaiba, or frequently, a mixture of
the two. For analytical purposes, copaiba is to be regarded as a
mixture of an essential oil and a resin, which must be separated and
the characters of the two examined. It is obvious that, since the
essential oil is practically a mixture of neutral sesquiterpenes, such
important characters as the acid and ester values will vary according
to the percentage of essential oil, whilst they may be fairly constant
for the resins present.

The thicker balsams, such as Maranham, Maracaibo, etc., are those
usually preferred for use in medicine, but the thinner ones, such as
the Para and Bahia varieties, are used for the distillation of the
essential oil.

In the examination of this drug, the following are the figures that
should be obtained and the methods adopted. As the various balsams
have somewhat different characters, the limit figures for each are
summarized in tables (p. 449).

(1) The specific gravity of the balsam, and of the essential oiL

COPAIBA.

449

CM
>^
EH
OS

I— I

lO (N

«D r-l lO O O
Oi U5 C- so (M

OS o

•

c<) ec

6 .H

o o w

!h O <M

o .... 1

o .

6th

o »o

OS iH

:!|'

§ i-i oc »c o 2

»o

ooo

O o o

OS o t> so « ^

<M (M

6-^ 1

O t- (M ,

O . . . r .

o -

S^ '

OS U5 "-^

or:^ 1

o o

iH i-H

2'^'S^

O 08 S

OS r-l lO 00 ?D

lOO

o o
O <N

CS W5 -^ 00 i-l

eo -«J<

W rH

6-:^ Sj

^S^ 1

• 1

o . . . .cc

O J

o o

«£ 1

OS lO

6 ^

>•

eo o ^ o (M

o >o

o . r . =11

00 <-# 00 t> o t? 1*

«5 eo

3§

6 A. ci.

t- r-l (M O «?

R
H

o
a

O U5 iH

OS ITS iO OS .-1
O rH

o = . r r 1

so 00 (N (M M5
O O ''J* 00

-* CO

o .

so iH

tils

OSIO

S »S 1

6 rH

»0 (M

OS ,-1 ICO 00

O "O

o o
-<»< O <M
O U5 iH

05 XC -* OS .-H

6 fH

-2 = = — 1

6^" '

irq -* 00 «ti >o
t- O Cft I>

^'^l'^

qs»p

» * 1

O <-!

o*.

OS i-i -* ec 00

U5 »0

O o o

OS W5 TtH 00 .-1

CO CO

OS o O

O rH

o .

6>^

2 rH o n

O 03 eg

o S

•1 • •

o ^1 • . .

■

o "^ . .

W5 X S

^ X 1

^H 0) g C

"" « S

--^•s . -1

£ o fl ^ ^le

llll

SC o3 « »- o

°*^ J- o rH « .i;

ts a>

"^ I- ;r; .^

. <« iH .2 -*a ■*»

.2 -t^

o, a> 0) o m Oi

o tc

|-§lt

a2«pM-iJWO

-tJW

VOL. I.

450

FOOD AND DRUGS.

Whilst the former varies within very wide limits, the latter will be
found to be much more constant.

(2) The refractive index of the balsam and the oil.

(3) The optical rotation of the essential oil.

(4) The amount of essential oil and resin. This may be deter-
mined by drying about 2 grms. in a flat platinum capsule at about
120° C. in an air oven to constant weight. The essential oil, however,
should be separated for examination by passing a brisk current of
steam through 30 grms. of the balsam. This will yield sufi&cient oil
for examination, and if the distilling flask be kept at 100" by immer-
sion in a water bath whilst the steam is passing briskly through, the
distillation will be complete in from three to four hours. If the oil
be distilled over a naked flame, it should be under reduced pressure,
but even then it is not possible to decide when all the oil has passed
over, and some decomposition is bound to occur, with the result that
the essential oil is contaminated with products of destructive distil-
lation of the resin.

(5) The following is the only reliable colour reaction that can be
used. It will detect very small additions of gurjun balsam. Five or
six drops of the essential oil are added to a mixture of 15 c.c. of
glacial acetic acid, and 5 drops of nitric acid. If no coloration takes
place in five minutes, gurjun oil may be regarded as absent. In the
presence of this adulterant a red or purple-red coloration is developed
in a minute or two, the time taken and depth of colour depending on
the amount of adulteration.

(6) The acid and ester values of the balsam and of the separated
resin should be determined. The residue left in the distilling flask
should be cooled, pressed between filter paper to remove as much
moisture as possible, and then dried in a water oven.

The figures on page 449 represent fair average values, but it must
not be forgotten that from tioae to time abnormal samples will be
found which have figures outside these limits, but are still pure.

The necessity of insisting on a high optical rotation for the essen-
tial oil is shown by the following table which includes the values de-
termined on a number of samples of direct importation by Messrs.
Evans, Sons, Lescher & Webb's chemists : —

Maranham.

Specific Gravity.

Optical Rotation,

Specific Gravity.

Optical Rotation.

0-902

- 13° 0'

0-899

- 16° 12'

0-900

- 20° 0'

0-898

- 17° 30'

0-900

- 16° 0'

0-898

- 10° 0'

0-898

- 17° 20'

0-905

- 9° 30'

0-900

- 21° 40'

0-900

- 13° 20'

0-901

- 14° 10'

0-902

- 16° 40'

0-902

- 13° 0'

0-903

- 16° 12'

0-900

- 13° 44'

0-0025

- 12° 30'

0-904

- 10° 30'

0-899

- 17° .

0-901

- 16° 0'

0-898

- 18°

COPAIBA.
Maraoaibo.

451

Specific Gravity.

Optical Rotation.

Specific Gravity.

Optical Rotation.

0-903

- 6° 30'

0-900

- 6°0'

0-900

- 7°0'

0-902

- 6°0'

0-901

- 6°0'

0-901

- 8°0'

0-898

- 21° 0'

0-895

- 19° 30'

0-890

- 30° 0'

0-895

- 20° 36'

0-894

- 21° 30'

0-887

- 26° 26'

0-888

- 28° 0'

0-896

- 18° 30'

0-896

- 20° 40'

0-893

- 28" 0'

0-894

- 26° 0'

0-886

- 32° 40'

0-891

- 28° 30'

0-891

- 28° 0'

0-893

- 23° 44'

0-892

- 25° 0'

. 0-891

- 24° 0'

0-886

- 31° 0'

0-889

- 26° 0'

0-889

- 26° 0'

0-886

- 31° 0'

Cartagena.

Specific Gravity.

Optical Rotation.

Specific Gravity.

Optical Rotation.

0-896

- 30° 0'

0-895

- 40° 0'

Bahia.

Specific Gravity.

Optical Rotation.

Specific Gravity.

Optical Rotation.

0-898
0-897
0-888

- 9°

- 10°

- 28°

0-898
0-909
0-901

- 8°

- 2° 42'

- 8°0'

Adulterants. — As stated above, the only adulterants met with
to any extent at the present time are African copaiba and gurjun
balsam. The colour reaction described above (p. 450) will detect as
little as 5 per cent of gurjun oil with certainty. Dextrorotatory oils,
or oils with a rotation below - 6° are very suspicious and African
balsam of copaiba is to be suspected. Under oil of copaiba the
British Pharmacopoeia states that the oil is soluble in its own volume
of absolute alcohol and gives this as a distinction from African copaiba
oil. This, however, is not so, as both are usually soluble in their own
volume of absolute alcohol.

The following characters of these two adulterants will assist the
analyst in forming an opinion on the character of the sample ex-
amined.

452

FOOD AND DEUGS.

Gurjun Balsam.

African Copaiba.

Specific gravity at 15° .

0-955 to 0-980

0-985 to 1-000

Specific gravity of essential oil

0-910 „ 0-930

0-916 „ 0-925

Rotation of essential oil . . .

up „ - 135°

+ 12° „ +45°

Refractive index of essential oil .

1-5050

1-5000 „ 1-5080

Acid value of balsam ....

10 „ 20

55 „ 60

Ester value of balsam ....

1 „ 12

Acid value of resin ....

40 „ 80

110 to 120

Ester value of resin ....

2 „ 25

In cases where African copaiba is suspected, the essential oil may
often be fractionated under reduced pressure, with advantage. African
copaiba oil yields fractions becoming steadily more dextrorotatory —
so that if an oil from a given sample shows a rotation of say - 4° and
on fractionation, the fractions become less laevorotatory and then
dextrorotatory, it is almost certain that African copaiba is present.
The following figures were obtained on three authentic samples of
African copaiba by the author and Bennett : —

Fraction.

II.

III.

Sp.
Gravity.

Ref.
Index.

Rota-
tion.

Sp.
Gravity.

Ref.
Index.

Rota-
tion.

Sp.
Gravity.

Ref,
Index.

Rota-
tion,

25 per cent

25 „

26 „
20 „

0-917
0-918
0-921
0-927

1-5030
1-5043
1-5061
1-5082

+ 17°30'
+ 28°30'
+ 46°

+ 55°

0-915
0-917
0-9-20
0-924

1-4960
1-4965
1-4980
1-5089

+ 16°
+ 19°

+ 24°
-f48°

0-914
0-917
0-919
0-923

1-4975
1-4980
1-4981
1-5090

+ 24°
+ 26°
+ 29°
+ 43°

These results have been confirmed by Cocking (" Chemist and
Druggist," 1910, ii., 51) who gives the following table showing the
optical value of pure and adulterated copaiba oils, and of their 10
fractions of 10 per cent each. He points out that if the sample be
pure the figures obtained will all be negative, and they will increase
from the first to the last fraction, although not regularly. If the
rotation of the first fraction be subtracted from that of the tenth, a
figure will be obtained which varies very little for genuine samples,
and is always a negative quantity. This figm*e (the "difference
value") varies from - 3*7° to - 7'6°.

When African copaiba is examined in this manner, the rotations
of all the fractions are, as would be expected, dextrogyrate, and the
rotations of the successive fractions increase, but to a much greater
extent than with the South American copaiba, in consequence of which
the difference value is much greater than with copaiba and is a posi-
tive figure. The figures also show a curious feature in that the tenth
fraction has a considerably lower rotation than the ninth. As would
be expected from the fact <^hat the range of boiling-points of the

CEEOSOTE. 453

constituents of the volatile oils from the two varieties are practically
identical, a mixture of the two will distil over containing proportional
parts in each fraction, and the presence of the African will be shown
at once by the difference value being positive.

In some cases, as will be seen from the tabulated results below,
where only a small percentage of the adulterant was present, all the
fractions were laevogyrate, but the difference value was positive. The
same process was applied to gurjun oil, which, like copaiba, gives
laevogyrate fractions, but, unlike it, they successively decrease instead
of increasing, and thus give a positive difference value similar to
African copaiba.

With the true copaibas the rotation of the first fraction is in every
case lower than that of the original oil, but in the adulterated samples
it is higher. It is important that the distillation of the oil should be
conducted in vacuo, since, if carried on under atmospheric pressure,
the higher temperature necessary causes some decomposition, which
entirely alters the optical rotation, as shown in the table on page 454
under *' u ".

The presence of such adulterants as fatty oils or turpentine gives
no difficulty to the analyst. In the case of fatty oils the sample is
saponified and the liquid neutralized, the acids precipitated with silver
nitrate and the mixture diluted with water. The salts of the resin
can be shaken out with ether, in which the fatty acid salts are in-
soluble. These can be decomposed by hydrochloric acid and the
liberated fatty acid examined. Their liquid or semi-liquid character
enables them to be at once distinguished from the resin acids.

Turpentine is at once detected by its odour on evaporation. Its
boiling-point — about 160° — enables it to be easily distilled off before
the essential oil of copaiba passes over, and its refractive index, about
1"4720, at once discriminates it from oil of copaiba.

CREOSOTE.

This drug is obtained by the distillation of wood tar. It is official
in the British Pharmacopoeia which describes it as a mixture of
guaiacol, creosol and other phenols. It requires it to have the
following characters : —

It is to be neutral or only slightly acid to litmus ; soluble in
150 volumes of water at ordinary temperatures; specific gravity
not below 1-079 ; it is soluble in alcohol, ether, chloroform, glycerin,
and glacial acetic acid. It distils between 200° and 220°. A 1 per
cent alcoholic solution gives a green coloration, rapidly changing to
reddish-brown, with a drop of ferric chloride solution. It is laevo-
rotatory. A drop on filter paper, heated to 100°, leaves no translucent
stain. It is miscible with an equaJ volume of collodion without
gelatinization ; when shaken with five times its volume of ammonia
(sp. gr. 0*959), its volume is not materially diminished (distinction from
phenol).

As a matter of fact, beechwood creosote is optically inactive, or
faintly dextrorotatory.

454

FOOD AND DKUGS.

t-H S

o -<

" I

m CO
g i
o

E-t
O
C4

-<
o

^•2

^^1

5 1

53 2

•si

"SI

•si

'-' I I I i I I I I I I I I

qpb-qppt-005ap<Ng5»prH

<N<N(M<N(N»H(M.H<N(MCM<M
I I I I I I I I I I I I

•Sd
I

.oop9>tHO«0»piHqpb-qpcp
•-' I I I I I

I I I

.t-t-OcpqDt-Qp»p<N'*005

/S<Nr-t(MiH<NrH(MrHCNCN(N(M
'-' I I I I

I I I

I I I I I I I I I I t I

..HO>ooO'*oocoo«Npqp<N

'^ I I I I I I I I I I I I

.'<}<pcNcpoppb-Qp9>aDqDO»

m^do cb6sQbTH>b»Ha205»H00O
'"' I I I I I I I I I I I I

O-*t-«p»Ht-»pCToq0Qpt-
'doiboodoTHTHTHOTosodbo

rH.Hr-lTHC<lT-l<»rH,H55rH(M
I I I I I I I I I I I I

tHCOTHTHQp-^rHpODipipCT

*' ' *' '" ~ ■ ' *■ '' p db 6

C5 rH <N
I I I I I I I I t I I I

. p ppcppTj<<»pt-'*>prJ(
'-' I I I I I I I I I I I I

I I

.CXXNp-^pt-pCNppoO^

^d> >oa>65<>i>b5^a365<NdbTH

^i-HTHrHr-l(NrHOJiHi-l<M»HCN

^ I I I I ; I I I I I I I

CM ©» -«}(

+ + + + + + + + I

>P'*pop(N05pC^

CO tH CO r-l

+ + + I I I I I I

ij-ci op p »p p

+ + + I I I I I I

QOCO Ot-OCDtO-*

+ + + I

a> q> r^ cp q> r-i cxp <c q> ^ (C ^

COCX)<MC5C^COCM05"5

;! + + + I I I I

OiOO->*Ot-r-lt0050

+ + +

•HTt4^t-pb-pp»C
CO CO O tH

+ + I I I I I I I

t-pC<ippTjlTjipTji

OlOOTrHt-cbiOOO
C^ 00 O 1— I

+ + I I

; rfS « f^ <s> H^ Cft,^ •«• •<>»,*» -^

prj^pppooTtfoocq

Q0«3TdlrJ(t-CDi-icDO

r-l iH

+ + 1

1 1 1 1 1 i

>p»0-*lOCO<MOOC75 0

aOgS-^tiOODijt-oo

+ + 1

1 1 1 1 1 1

"*9«P

iti^m^coco

■*o}-*iooo«>^t-ds|

+ + 1

1 1 1 1 1 r

t-t>C0C5C0»0(MOCi

^co^cot-^gt-c.

+ + 1

1 1 1 1 1 1

g g o

a,wv «o-. 3

•Si

o o

• otS CO
CH w 2

S .us

|g^
o "-S

08 C- «

e8 rt O

oS3

« O -*3

c6 »- O

o • P

•*3 fl >

CREOSOTE.

466

Creosote is usually distilled from beechwood — sometimes from oak
or pine. It consists essentially of a mixture of phenoioid compounds
in varying proportion, amongst which are phenol (boiling at 182") ;
paracresol (203°) ; guaiacol (200°) ; creosol (219°) ; dimethyl-guaiacol
(230°) ; and propyl-guaiacol (241°).

Guaiacol is present to the extent of about 15 per cent to 25 per
cent and is one of the most important constituents of creosote.

A good creosote should have a specific gravity of at least that re-
quired by the British Pharmacopoeia, preferably a little higher — up to
1-085. On fractionation three typical samples gave the following
results, with which pure samples will approximately correspond : —

Sp. Gr.

Guaiacol.

Under 200°.

200^-205°.

205°-210°.

210°-215°.

215°-220°.

1-0815

Per cent
21-5

Per cent
6

Per cent
39

Per cent
22

Per cent
25

Per cent
6

1-0820

19-8

7-5

1-0800

In order to differentiate between creosote and coal-tar phenols
(coal-tar creosote), one volume of the sample is shaken with a mixture
of 3 volumes of glycerine and 1 of water. The diminution in the
volume of the creosote indicates the amount of soluble impurities
derived from the coal-tar.

A pure wood creosote loses at least 10 per cent of its volume when
shaken with 5 volumes of 10 per cent ammonia.

Guaiacol is usually present to the extent of 10 per cent to 25 per cent,
and may be determined as follows (Kebler, " American Jour. Pharm."
39, 933). Five c.c. of creosote are mixed with 50 c.c. of a 20
per cent alcoholic solution of potassium hydrate. The crystalline mass,
which separates in 10 minutes to 30 minutes, consists of a compound
of guaiacol and creosol with potassium. The dried crystals are heated
for a moment with 5 c.c. of a 10 per cent solution of sulphuric acid,
the liquid is diluted, and the mixture of guaiacol and creosol, which
separates as a heavy oil, removed. By treating this oil with 4 c.c. of
a concentrated solution of ammonia, the guaiacol ammonium com-
pound is formed as a crystalline mass, which separates before the
less crystalline creosol compound. The latter is removed by means
of benzol, and the guaiacol ammonium compound decomposed by a
10 per cent solution of sulphuric acid. The liberated guaiacol is dis-
solved by shaking with benzol; and finally weighed after evaporating
the solvent.

If a fuller analysis of creosote is required, Behal and Choay
" Comptes Rendus " cxvi. 200) advise a separation based on the fol-
lowing facts : —

(1) Hydrobromic acid removes methyl from the methyl ethers of
the phenols, (2) that monophenols can be removed by steam, (3) that

456 FOOD AND DEUGS.

polyphenols cannot be thus removed, (4) that ether abstracts from
aqueous solutions pyrocatechin and homopyrocatechin as well as
monophenols, and (5) that pyrocatechin and homopyrocatechin are
separable by benzene. In carrying out the process a current of hydro-
bromic acid is passed into creosote mixed with some water, by which
proceeding the ethers of polyphenols are demethylated. By distil-
lation with steam the monophenols are carried over and can be
separated from the distillate by shaking with ether.

CUBEBS.

The dried full-grown unripe fruits of Pij^er cubeba are the official
cubebs of the British Pharmacopoeia.

The only official test is that the crushed fruit should impart a
crimson colour to sulphuric acid.

The principal constituent of this drug is from 12 per cent to 15 per
cent of essential oil (see p. 610). It also contains resinous matter, so
that by extraction with ether it yields from 20 per cent to 22 per cent
of oleo-resin. Cubebs yield from 6 per cent to 8 per cent of ash, usually
about 7 per cent. Traces of an acid, termed cubebic acid, are respon-
sible for the crimson colour with H^SO^. This drug is frequently found
in commerce mixed with similar fruits. The various piperaceous and
other fruits used for adulterating cubebs may be, for the most part,
distinguished from true cubebs simply by their external characteristics,
whilst other-i resemble the genuine drug so closely that a microscopic
examination of a section of the fruit is necessary. A few of the adulter-
ants, however, can only be distinguished by the fact that they do not
contain cubebic acid, and therefore do not give a purple-red coloration
with strong sulphuric acid. Cubebic acid does not occur in the peri-
sperm only, as has hitherto been supposed, but also in the pericarp,
and the same is true of the occurrence of piperine in black pepper.
Piperaceous plants which contain cubebic acid or an allied compound
do not, as a rule, contain any alkaloid, such as piperine.

The characters of genuine cubebs are those given above, and as
the percentage of essential oil is high, a determination of this should
be made : 125 grms. well bruised, should be steam-distilled until no
more oil is carried over. At least 12 per cent should be obtained from
good cubebs, and this should have the chuacters described under oil
of cubebs (p. 610). The crushed fruits should yield about 20 per cent
or more of oleo-resin to ether, the ether being driven otf at a tempera-
ture of 70° to 80°, and the residue weighed. The characters of tincture
of cubebs will be found in t^e table on p. 495.

GALBANUM.

This gum resin is officially described as the product, of Ferula
galbaniflua and probably of other species.

The only characteristic test given in the Pharmacopoeia is that if a
fragment is heated to redness in a dry test tube, the contents of the
tube, after cooling, yield with boiling water a solution which when

GUAIACUM.

457

largely diluted with water and rendered alkaline with ammonia ex-
hibits a blue fluorescence.

This test enables galbanum to be detected in certain other gum
resins, such as ammoniacum.

The official galbanum is that known as Levant galbanum, and for
medicinal purposes it must be in tears either separate or agglutinated.
The lumps containing tears embedded in a resinous mass are there-
fore not official, but are frequently met with in commerce.

Galbanum contains about 5 per cent to 10 per cent of essential oil ; in
very soft varieties, as much as 20 per cent being found. It contains from
20 per cent to 30 per cent of gum and mechanical impurities and from 60
per cent to 70 per cent of resin soluble in alcohol. It also contains traces
of free umbelliferone, which is the anhydride of umbellic acid C^HgO.
OOOH ; about 20 per cent of the resin consisting of umbelliferone (or
umbellic acid) combined as an ester with the alcohol galbaniresinotannol
CjgH^gO^ . OH. It is to the umbelliferone that galbanum owes the
characteristic fluorescent reaction described above. A petroleum ether
extract of galbanum should yield only a very slight green coloration
when shaken with aqueous copper acetate. Pure galbanum should
have the following characters : —

Mineral matter, 5 to 8 per cent

Resin soluble in 95 per cent alcohol not below 55 per cent
Acid value -^ 20 to 40

[. of extracted resin 60 ,, 100

Ester value
Saponification value J

80 „ 120

GUAIACUM.

Guaiacum resin, the product of Guaiacum officinale or of G.
sanctum, is official in the Pharmacopoeia.

The characteristic official test is that an alcoholic solution assumes
a blue colour on the addition of dilute ferric chloride solution.

Guaiacum usually occurs in large blocks, but sometimes in tears.
It breaks with a clean glassy fracture, showing a greenish or reddish-
brown colour.

The principal constituents of this resin are a-guaiaconic acid
^22^24^0 ' y8-guaiaconic acid C^^H.^^Pg ; guaiaretic acid, Cj^QS..j^,Jd.^(OT£)
(about 10 per cent) ; guaiacic acid G.^^-^^O^iOH).^ (10 to 12 per cent) ;
and small quantities of gum and other indefinite substances. This
resin is of great interest on account of the fact that Doebner has
succeeded in condensing tiglic aldehyde, guaiacol and cresol to a
resinous acid, C^^H.^^O^, isomeric with guaiaretic acid, a result which
throws some light on the formation ot natural resins, and indicates
that they owe their origin to the condensation of phenols and alde-
hydes rather than to the oxidation of the terpenes.

Good guaiacum resin in tears yields about 1 per cent of mineral
matter and 98 per cent soluble in 90 per cent alcohol. Block guaiacum
contains more mechanical impurities than the tears and usually gives
3 per cent of ash and about 90 per cent soluble in 90 per cent
alcohol.

Numerous oxidizing agents produce a blue colour when brought

458 FOOD AND DEUGS.

into contact with an alcoholic solution of guaiacum (this is due to the
oxidation of a-guaiaconic acid).

The well-known reaction for blood with tincture of guaiacum is
based on this fact.

Genuine guaiacum should have the following characters : —

Per cent.
Mineral matter . . . . . , . 1 to 4

Acid value

Soluble in 90 per cent alcohol

Acid value of acetylated resin

Ester value of acetylated resin

Methoxyl number

Soluble in petroleum ether

60 to 70

87 to 98
not above 50
125 to 150
70 to 85
not above 2

The ester value of the acetylated resin is valuable on account of the
large amount of hydroxy-bodies present in this resin ; and the high
methoxyl number is characteristic.

Colophony as an adulterant of powdered guaiacum may be detected
by the Storch-Morawski reaction (p. 478).

Starch is sometimes added and may be detected by testing the cold
aqueous decoction with iodine solution.

Guaiacum adulterated with colophony will yield a large proportion
of resin soluble in petroleum ether.

Ammoniated Tincture of Guaiacum. — This is ofi&cial in the Pharma-
copoeia, but no standards are given. It is a solution of the resin in
alcohol and ammonia, flavoured with essential oils of lemon and nut-
meg. A pure tincture should have the following characters : —

Specific gravity = 0-898

to 0-907

Solid residue =14

„ 17-5 per cent

Alcohol by volume = 69

„ 71

Ammonia NHg = 1-9

„ 2-2 „ (by weight)

The alcohol should be determined by rendering the tincture exactly
neutral with dilute H2SO4 and then distilling the alcohol and determin-
ing the specific gravity of the distillate made up to the proper volume.

Ammonia is determined by distilling 25 c.c. rendered alkaline with
KOH and diluted with 175 c.c. of water and collecting 100 c.c. of the
distillate through a well-cooled condenser, and then titrating with
decinormal H^SO^.

GAMBOGE.

Under the name Cimbogia, this is official in the Pharmacopoeia.
It is a gum-resin obtained from Garcinia Hanhurii.

The official standards are that it is completely dissolved by successive
treatment with 90 per cent alcohol and water ; that a cooled aqueous
solution should not become distinctly green with solution of iodine
(absence of more than a trace of starch) ; and that it should not yield
more than 3 per cent of ash.

This drug consists of about 70 to 80 per cent of resin acids, known
as " gambogic acid " ; 15 per cent of gum ; with small quantities of
mineral matter, vegetable debris, etc.

GENTIAN.

459

It usually occurs in commerce in the form of pipes, the gum
resin having been allowed to dry inside hollow bamboos. This type
of gamboge is produced in Siam, Cochin China and Cambodia. The
product of Garcinia Morella, a tree growing in India and Ceylon, is
known as Indian gamboge. It is found in pieces of irregular shape,
but should have the same characters as Siam gamboge. The usual
adulterants are starch, rosin, turmeric and mineral matter. Starch
is detected by its reaction with iodine, and also microscopically, mineral
matter is detected by a high ash value, and rosia by the Storch-
Marowski reaction (heating the powdered sample with acetic anhydride
and allowing H2SO4 of specific gravity 1-5 to run slowly on to the
surface of the cooled liquid when a violet colour is developed at the
surface of contact of the liquids if rosin be present).

Eosin is also indicated by the high acid and low ester value of the
sample. Ten samples of pure gamboge, examined by the author, gave
the following figures : —

Acid Value.

Ester Value.

Saponification
Value.

3
4
5
6
7
8
9
10

76
81
82
74
73
81
84
80
73
78

60
58
64
60
60
55
61
69
62
51

136
139
146
134
133
136
145
149
135
129

Eberhardt gives the following limits for the reaction for starch : —
One grm. of the powder to be tested is dissolved in 5 c.c. of potash,
followed by an addition of 45 c.c. of water, and finally an excess
of hydrochloric acid. The turbid liquid is then filtered through
cotton wool, and one or two drops of iodine are added to the clear
filtrate. In presence of over 2 per cent of starch there immediately
ensues a dark blue coloration, or a similarly coloured precipitate is
formed. The powdered commercial drug usually gives a yellow
coloration, which afterwards turns blue ; pure gamboge, with 1 per
cent of added starch, gives a dull blue, which deepens on standing, and
deposits a precipitate after several hours. Five per cent to 10 per-
cent of starch gives a blue precipitate immediately. Five per cent
and under of turmeric gives a decided starch reaction. Turmeric
may also be detected by the borax reaction (see under Turmeric).

GENTIAN.

The dried rhizome and roots of Gentiana lutea are official in the
Pharmacopceia, but no standards are given.

The fresh root contains at least three bitter principles, gentiopicrin,

460 FOOD AND DRUGS.

gentiin and gentiamarin, of which the last two exist in the dried drug
the first having been decomposed by changes taking place during
drying. Gentiopicrin is a glucoside of the formula C.,„H3yOj2 which
yields gentiogenin and dextrin on hydrolysis, according to the
equation

Amongst the other substances present are sugar, gentianose, which
yields on hydrolysis gentiobiose and levulose, the former finally splitting
up into dextrose.

The only methods of analysis available are the determination of
the ash and the amount of extractive; and a microscopical examin-
ation.

Genuine gentian root, which is largely sold in powder, should not
contain more than 5 per cent of mineral matter. The cold water ex-
tract of a good root varies from 30 per cent to 40 per cent (but in a
highly fermented root this may be much lower). The amount of ex-
tractive obtained by 60 per cent alcohol is usually from 34 per cent
to 44 per cent.

A good deal of powdered gentian is adulterated with either ground
olive stones, powdered almond shells, or even pine wood.

A genuine powdered gentian consists chiefly of parenchymatous
tissue, most of the cells containing minute crystals, and small oil
globules. Only a few starch grains are present. The vessels are
scattered and either isolated or in small groups. There are no
sclerenchymatous cells or fibres present ; this is the most character-
istic diagnostic feature of the drug, as most of the adulterants used
contain much sclerenchymatous tissue.

Collins (" Chemist and Druggist," 64, 403) has found almond shells
and ground pine wood as adulterants, but olive stones are probably
more common. These are all easily detected by the microscope. To
detect this type of adulterant, the sample is preferably shaken with
either water or 70 per cent alcohol and the heavier portion which
sinks to the bottom of the water examined. The illustrations on
opposite page represent pure gentian and powdered date stones.

Compound Tincture of Gentian. — The characters of this official
preparation are given in the table on p. 495.

KINO.

The drug is officially described as the evaporated juice of Ptero-
carpus marsupium. It is probable, however, that it is also obtained
from other plants.

Kino is described in the Pharmacopoeia as being almost entirely
soluble in 90 per cent alcohol, and practically insoluble in ether.
Not less than 80 per cent should dissolve in boiling water.

Kino occurs in small angular reddish-black fragments, and some-
times in cakes. The drug is of value solely on account of its astrin-
gent properties, and kino-tannic acid is its active constituent.

A genuine kino should not contain more than 15 per cent of
moisture and from 1 per cent to 3 per cent of mineral matter.

GENTIAN.

461

Fig. 42. — Powdered date stones

462

FOOD AND DEUGS.

When determined by the method described on page 11 the tannic
acid should vary between 70 per cent and 80 per cent, sometimes
even up to 83 per cent.

Tincture of Kino is official. Its characters are given in the table
on page 495.

LIQUORICE ROOT.

Liquorice root is official in the British Pharmacopoeia, being the
drug from which several galenical preparations are made. It is de-
scribed as the peeled root and subterranean stem of Glycyrrhiza glabra,
and other species. No standards are given in the Pharmacopoeia.

The only methods of analysis available are the determination of
the ash, and a microscopic examination.

The ash should vary between 3 per cent and 5 per cent.

Fig. 43. — Powdered liquorice root.

A microscopic examination of powdered liquorice should reveal
numerous parenchymatous cells containing very characteristic starch
grains, oval or kidney-shaped, and generally showing a central cavity
of the same shape as the grain ; numerous pitted vessels and bast
fibres are observable. Foreign starchy matter should be looked for
and also any very thick-walled cells as may be present through adul-

LIQUORICE ROOT.

463

teration with ground olive stones and similar very much-hardened
tissues.

The principal product of liquorice root is liquorice juice, or simply
•' liquorice ". This is the juice of the root, filtered and evaporated to a
nearly solid consistency. It is frequently adulterated, either with
starch, which is detected by the microscope and by the high amount of
substances yielding sugar on inversion. Liquorice juice contains a vari-
able amount of glycyrrhizin, the characteristic substance of the drug,
with variable amounts of sugar, gum, starch, and insoluble matter.
In examining it, the following determinations are necessary : —

Moisture. — This should not exceed 15 per cent.

Starch a7id Gums. — Weigh out 2-5 grms. of liquorice juice in a
small beaker. Add 15 c.c. hot water, cover with watch-glass, and stand
on a hot water bath until thoroughly dissolved, stirring as may be re-
quired. Cool. Add 25 c.c. of 80 per cent (by vol.) alcohol, stirring
meanwhile. Then add 50 c.c. 95 per cent alcohol with stirring and
allow to settle thoroughly, while covered, for about half an hour.
Filter through a dry weighed filter. Wash until colourless with 80
per cent alcohol. Dry in water oven to constant weight. This gives
the starch and gummy matter.

Glycyrrhizin. — Transfer the filtrate and washings to a flask and
distil off the greater part of the alcohol, or until there is only enough
liquid for conveniently transferring to a small porcelain evaporating
dish. Evaporate to a syrup or to the removal of alcohol, and trans-
fer to a stoppered flask graduated to hold 30 c.c. and make up to the
mark with water. Add 3 c.c. of dilute sulphuric acid (10 c.c. cone.
H2SO4 to 300 c.c. water) slowly and with constant stirring. Aljow
to stand all night at a temperature of about 60° F. Decant the
supernatant liquid, wash the precipitate three or four times with ice
water and dissolve in a little dilute alcohol with 2 or 3 drops of
ammonia to neutralize traces of sulphuric acid, and evaporate to
dryness in a flat-bottomed porcelain dish, till the weight is constant.

B. Hafner (" Zeitschr. des Oesterr. Apoth. Ver." xxxvi. 542) prefers
the following method. Ten grms. of the coarsely powdered extract
are warmed for several hours with 200 c.c. of 95 per cent alcohol and
25 c.c. of N sulphuric acid, and the insoluble matter is washed with
alcohol. The filtrate, made feebly alkaline with ammonia, and
diluted with an equal volume of water, is evaporated, made, up to 100
c.c. with water and a few drops of ammonia, filtered, and precipitated
with dilute sulphuric acid. The precipitated glycyrrhizin is washed
with 2 per cent to 3 per cent sulphuric acid, dried in the desiccator,
and then extracted with acetone on the water bath. After adding
water and barium carbonate, the acetone is expelled on the water
bath, the residue. digested with 200 c.c. of hot water, and the filtered
solution evaporated, dried at 100°, and weighed. The barium glycyr-
rhizate thus obtained should contain 18*76 per cent of barium, which
may be confirmed by evaporating with sulphuric acid, and igniting.

The following method is more rapid and not much less accurate : —

Weigh 2 5 grms. of well-ground juice into a small beaker, cover
with 15 c.c. of water and heat on a water bath until dissolved. Cool

464 FOOD AND DRUGS.

and add gradually with stirring 75 c.c. of methylated spirit. Set aside
to settle about thirty minutes, filter through a tared paper into an
evaporating dish, washing dish and paper with 50 c.c. methylated
spirit mixed with 5 c.c. of water. This leaves the insoluble starch
and gum on the paper, which is dried and weighed. Bulk the filtrates,
evaporate the alcohol off on a water bath. Transfer the syrupy liquid
to a cylinder with the aid of 30 c.c. of water, cool strongly in a melt-
ing ice bath, and add 3 c.c. H^SO^ (5 per cent) with constant agitation,
then freeze solid in an ice-salt jacket. If gradually melted the gly-
cyrrhizin forms a compact mass at the bottom of the cylinder. Wash
by decantation with about 50 c.c. of Hfi at 0°, drain as far as possible,
add -1 c.c. of ammonia, and transfer to a tared dish with absolute
alcohol ; evaporate and dry at 100° until constant.

Cederberg proposes the following as an accurate method of deter-
mining the glycyrrhizin : 10 grms. of the powdered juice are well
shaken for an hour in a flask with 200 c.c. of 95 per cent alcohol and
25 c.c. of normal HgSO^. The liquid is filtered and the filter washed
with 100 c.c. of hot alcohol. The filtrate is diluted with half its
volume of water and rendered alkaline with NHg. The liquid is now
evaporated to expel alcohol, and made up to 100 c.c. with water, and
100 c.c. of 20 per cent sulphuric acid added when it is cold. The
glycyrrhizin is precipitated and allowed to settle, and the supernatant
liquid poured off through a filter. The precipitate is then washed with
50 c.c. of 10 per cent H.2SO4, again allowed to settle, and the super-
natant liquid decanted through the same filter paper. One hundred
and fifty c.c. of 90 per cent alcohol is now added to the precipitate and
the whole warmed so long as anything will dissolve. This solution of
the glycyrrhizin is then filtered through the same paper, which is
washed with 50 c.c. of warm alcohol, but the filtrate is not mixed
with the previous acid filtrates from the original precipitate. The
filtrate is diluted with half its volume of water and rendered neutral
with potash solution. It is then made up to 500 c.c. One hundred
c.c. is evaporated to constant weight and dried at 110°. A second
100 c.c. is heated and treated with BaCl^ and the precipitate filtered
on to a tared filter paper, washed with hot water, dried at 110° and
weighed. The residue obtained by the evaporation of the 100 c.c.
represents the potassium glycyrrhizinate in 2 grms. of juice + K.^S04.
The amoijnt of K^S04 i^ calculated from the amount of barium sul-
phate found by the precipitation of the second 100 c.c. Thus : —

Residue found in 100 c.c. = 0-5005 grm.

BaS04 = 0-385 = K^SO^ 0 28b0 „

Potassium glycyrrhizinate 0*2125 „ (contains 11-58 per cent K)

= glycyrrhizin 0-1884 ,, = 9*42 per cent.

Sugars. — Ten grms. of the juice are dissolved with constant stirring
in about 100 c.c. of cold water and transferred to a 250 c.c. flask ;
colouring matter, etc., is precipitated by lead subacetate solution and
excess of lead removed by a strong solution of ammonium sulphate.
The liquid is filtered and an aliquot part titrated in the usual manner
with Fehling's solution. The result is calculated to invert sugar. For
the sugars after inversion, 50 c.c. of the filtrate, freed fiom gumny

LIQUORICE ROOT. 465

matters, etc., by means of alcohol, are inverted with 2 c.c. of strong
HCl at 70" for ten minutes, cooled, neutralized and then titrated as
usual with Fehling's solution. The difference between this and the
former result is calculated to cane sugar if necessary.

Eriksson (** Archiv der Pharm." 1911, 157) proposes the following
method, depending on the fact that glycyrrhizin is hydrolysed with
the formation of glycyrrhetinic acid and glucuronic acid, the latter of
which, as it contains an aldehydic residue, reduces Fehling's solution.

Ten grms. are powdered and dissolved in 100 c.c. of water, and
100 c.c. of 90 per cent alcohol added. The mixture is heated on the
water bath for half an hour, filtered, the filter washed with 50 c.c. of
hot alcohol, the filtrate evaporated until all the alcohol is' removed and
finally made up to 200 c.c. with water. Forty c.c. of this solution
(= 2 grms. of juice) is treated with 25 per cent H.2SO4 until no further
precipitation occurs. After a few hours, the precipitate is filtered
through a small filter, and the precipitate washed with 5 per cent
H2SO4. The filtrate is reserved for the determination of sugars. The
filter and precipitate are transferred to a porcelain capsule and heated
for fifteen minutes, with 50 c.c. of 90 per cent alcohol. The liquid is
filtered, the filter washed with a little alcohol, and 30 c.c. of water
added. The alcohol is evaporated off and 30 c.c. more water added
and the glycyrrhizin precipitated with 25 per cent H2SO4. After an
hour it is again filtered off. The filter and precipitate are then treated
in a porcelain capsule with cold 5 per cent alkali. The solution is
filtered into a flask and the filter washed with 100 c.c. of water, and 120
c.c. of Fehling's solution added, and the whole boiled under a reflux
condenser for fifteen minutes. The precipitated CugO is collected and
weighed in any of the usual methods, and calculated to glucose. The
amount of glucose indicated x 2*77 gives the amount of glycyrrhizin.

The filtrate reserved for the determination of sugars is neutralized
with 5 per cent alkali and the amount of reducing sugars estimated by
the amount of copper oxide precipitated in the cold after standing over
night. The saccharose is determined in the filtrate from this by boiling
for three minutes with excess of Fehling's solution. Or, alternatively,
an aliquot portion may be used for the determination of glucose by re-
ducing boiling Fehling's solution, and the saccharose determined by
inverting another portion of the filtrate and determining the total re-
ducing sugar now present.

By the .above process, Eriksson finds the following amounts of
glycyrrhizin and sugars in typical roots and in pure liquorice juices : —

Roots.

Glucose.

Saccharoses.

Glycyrrhizin

Italian (dried)

1-39-1-43

2-4-2-57

6-65-7-10

Spanish

1-28

3-20

6-49

Kussian

6-48

7-70

Russian

traces

6-50-

8-15

Russian

3-80

6-25

7-33

(fresh)

—

2-60

6-72

VOL. I.

466

FOOD AND DKUGS.

Juices.

Glucose.

Saccharoses.

Glycyrrhizin.

Glycyrrhizin
(Cederberg's method)

6-30

11-80

16-45

14-28

3-79

4-52

14-22

—

2-70

8-17

23-90

— ■

7-82

•.*-06

1210

1110

5-20

11-90

11-59

10-24

5-90

12-48

10-20

9-30

4-60

13-60

9-85

—

Liquorice root varies so enormously according to the country in
which it is grown, and even the locality in the same country, that it
would be very inadvisable to attempt to lay down any standard figures.
Samples must be judged individually and full account taken of their
place of origin. The principal variation is in the amount of glycyr-
rhizin contamed in the root, which may be twice as much in a root
grown in one district as in one grown elsewhere. This fact -divides
liquorice juice into three distinct species : —

Firstly, there are what may be described as the ordinary edible
juices. These are typified by a.glycyrrhizin content of about 10 per cent
to 13 per cent, and are sufficiently palatable to be used for the manu-
facture of stick liquorice. Of these the principal is the Calabrian
juice, which forms the basis of nearly the whole of the 'piire stick
liquorice of commerce.

Secondly, there are the juices which contain from 17 per cent to 25
per cent of glycyrrhizin, and which are too bitter to be palatable. Of
«uch juices the x^natolian is a type. Juices of this kind form the
principal source of supply for the pure block juice and for so many
purposes, such as confectionary and the tobacco trades. But they
are not made into stick liquorice except with the addition of sugar
of some kind or other, when they cannot, of course, be sold legiti-
mately as pure liquorice.

Thirdly, there is the sweet Spanish juice which frequently con-
tains 6 per cent or less of glycyrrhizin. This juice has too little
"' body " to be used much as an ordinary liquorice.

The following analyses are typical of liquorice juice of various
origins : —

Italian (Calabrian) Juices (Stick and Block).

Moisture ....
Ash

Soluble in water

Insoluble in water

Starchy and gummy matter

Glycyrrhizin

Sugars before inversion .

Sugars after inversion

Per cent
13-50

6-20
63-90
22-60
21-48

9-95
12-50
15-25

Per cent
12-80

5-98
69-25
17-95
20-80
10-18
13-50

4-95

Per cent
10-95
7-10
63-90
25-15
22-80
12-50
12-90
14-90

Per cent
14-65
6-69
64-80
20-55
24-50
11-42
13-00
15-50

Per cent
11-85
7-55
64-65
23-00
26-00
10-50
12-00
14-70

Per cent
13-6
5-9
65-9
20-5
25-2
10-5
11-9
14-5

LIQUOEICE ROOT.
Anatolian and Similar Juices (Block).

467

Moisture ....
Ash

Soluble in water

Insoluble in water

Starchy and gummy matter

Glycyrrhizin

Sugars before inversion

Sugars after inversion

Per cent
18-95

6-80
73-55

7-50
18-61
23-50
11-50
12-90

Per cent
20-50
6-90
72-45
7-05
19-00
18-75
12-00
13-90

Per cent
17-55

7-22
75-55

6-90
17-50
20-40
10-94
13-20

Per cent
16-95

5-80
74-55

8-50
19-65
21-55
10-88
13-00

Spanish Juices (Block).

Per cent

Moisture .....

9-40

10-50

8-55

Ash ... .

6-50

. 5-95

7-12

Soluble in water .

68-55

65-00

64-90

Insoluble in water .

22-05

24-50

26-55

Starch and gummy matter

20-48

21-00

23-50

Glycyrrhizin .

6-50

5-95

6-65

Sugars before inversion .

14-50

13-09

12-50

Sugars after inversion .

15-08

15-25

14-45

The following are analyses of adulterated samples

Pr cent

Moisture ....

13-50

12-95

12-50

12-90

13-50

14-00

13-2

Ash

3-9

4-7

5-0

4-2

4-6

16-1

4-8

Soluble in water .

80-50

78-00

80-56

74-50

77-0

74-60

76-90

Insoluble in water

6-0

9-05

6-94

12-60

9-5

11-4

9-85

Starchy and gummy matter .

17-41

16-50

18-00

17-50

16-90

17-05

16-05

Glycyrrhizin

6-40

7-00

7-25

14-25

16-50

8-12

16-00

Sugars before inversion

9-8

12-50

14-00

10-50

18-00

11-5

11-00

Sugars after inversion .

24-5

19-5

26-5

18-6

20-50

23-1

19-9

Liquid Extract of Liquorice. — This official galenical is made by
exhausting liquorice root with water and adding 25 per cent by volume
of 90 per cent alcohol, to the concentrated aqueous liquid when the
water has been evaporated until the liquid has a specific gravity of
1-200 at 15°. No official standards are given.

A pure liquid extract of liquorice should have a specific gravity of
1'130 to 1"150 : it should contain not less than 39 per cent of solid

468

FOOD AND DEUGS.

matter — often as much as 46 per cent to 48 per cent, and 17 per cent
to 18 per cent of alcohol by volume.

Comjoound LiqiLorice Poivder.—Tlhi^ powder is ofiBcial in the
Pharmacopoeia. It consists of : —

Powdered senna = 2 parts
„ liquoiice =2 „
„ fennel =2 „
„ .sulphur =1 „
„ sugar = 6 „

The only methods of analysis available are a determination of the
ash, which should vary from 4-5 to 5*3 per cent ; a microscopic com-
parison with powder of known authenticity ; a determination of the
sulphur ; the estimation of the matter extracted by 70 per cent
alcohol and the amount of sugar. The following analyses are by
Evans (" Pharm. Journ." (4) 20, 363) :—

-3

-S

—

•i

CO
<

4
<

2 §3

1*=^.

8S)

PL,

hSw

Per cent

1 . . .

3-86

5-12

2-26

2-86

60-52

50*0

10-62

8-92

2 . . .

4-20

6-54

3-98

2-56

60-16

47-7

12-46

8-49

3 . . .

3-98

4-84

2-36

2-48

60-08

49-2

10-88

8-91

4 . . .

3-68

4-88

2-00

2-88

63-44

50-0

13-44

8'90

5 . . .

3-84

4-66

1-68

2-98

63-10

49-6

13-50

8-78

The sugar is determined in the usual manner after inversion with
2 per cent of HCl at 70° for ten minutes, with Fehling's solution.

The sulphur is best determined by heating 1 grm. of the powder
with 25 c.c. of strong HNO3, 5 grms. of KNO3 and 25 c.c. of w^ater.
When oxidation is complete, 25 c.c. of hydrochloric acid are added
and the liquid evaporated to dryness in the fume chamber. The mass
is then extracted by boiling with 5 c.c. of HCl and 25 c.c. of water,
and the insoluble matter washed with water until free from sulphate.
The sulphate is then precipitated as BaSO^ and weighed. From the
weight the sulphur is calculated. Not less than 8 per cent is usually
obtained in this way, as besides the free sulphur, there are always
traces of sulphates present in the mineral matter of the drugs used.

MALE FEEN.

The rhizome of male fern, Aspidium filix-mas is official in the
Pharmacopoeia, being used for the preparation of an ethereal extract,.

MUSK. 469

which is known as the liquid extract of male fern. No standards are
given for either.

Male fern rhizome should not contain more than 5 per cent of
mineral matter.

The active constituent of male fern is filicic acid which either has
the formula C^^H^fii^ or C^^JI^qO^.^' ^^ exists both free in the rhizome
and also combined in the form of filmarone which slowly decomposes
in solution into filicic acid and aspidinol CjgHjgO^. Amongst the other
compounds isolated from the drug are flavaspidic acid C24H28O8 or
C24H.50O8 ; albaspidin Cg^HjgOg ; and filicinylbutanone G^.^H^Jj^.

Kraft (" Zeit. d. Oesterr. Apoth. Verein." xxxiv. 789.) regards an
amorphous acid which he describes as filicic acid as the only active
constituent of the drug, but according to other investigators, several
constituents have therapeutic activity as vermifuges.

Kraft recommends the following process for the assay of the ethereal
extract : —

Five grms. are shaken for a quarter of an hour with 60 grms. of
95 per cent alcohol and a solution of 2 grms. of potassium carbonate
in 40 grms. of water, after which 80 grms. of the mixture are quickly
filtered into a separating funnel, and agitated with 50 grms. of ether,
35 grms. of water, and 9 grms. of dilute hydrochloric acid. The
ethereal layer, after separation, is washed with another 35 grms. of
water, and then slowly evaporated in a 100 c.c. Erlenmeyer flask
until only about 2 grms. or less remain. The residue is dissolved in
1-5 grm. of hot amyl alcohol, the solution mixed with 5 grms. of
methyl alcohol, the mixture precipitated slowly by the very gradual
addition of another 25 grms. of methyl alcohol, and allowed to stand
in the stoppered flask overnight in a very cool place. The precipitate
is then collected on a tared filter, washed with 10 c.c. of methyl alco-
hol, and both the filter and flask dried at 60 to 70° C. until the
weight is constant. This weight represents the proportion of filicic
acid contained in 4 grms. of the extract. Kraft has found the filicic
acid in a number of extracts examined to vary between 0-4 and 10 per
cent ; but he considers that a good extract ought not to contain less
than 5 per cent.

MUSK.

Musk, the dried secretion of the preputial follicles of Moschus
moschiferus, is still ofiicial in the Pharmacopoeia, although the reason
for its inclusion as a drug is not easy to find.

The only standards given are that it should be free from earthy
impurities and should yield not more than 8 per cent of ash.

It appears that the grain musk — that is the contents of the sac
cut from the animal — is the official drug.

When pure — which is rarely the case — grain musk should yield
50 to 75 per cent to water, and 10 to 12 per cent to 90 per cent alcohol.
It should not contain more than 12 to 15 per cent of moisture, nor
more than 6 to 8 per cent of mineral matter.

The odoriferous principle of musk is principally a ketone, musk-

470 FOOD AND DEUGS.

one, but as musk is practically entirely used as a perfume material it
need not be further discussed.

MYBEH.

This gum-resin is an official drug, being described in the Pharma-
copceia as obtained from the stem of Balsamodendron myrrha and
probably other species. The only official test is that it should assume
a violet colour when moistened with nitric acid (distinction from
bdellium and false myrrh). The myrrh of commerce, when genuine,
is known as Herabol myrrh, but as imported is usually mixed with
more or less bdellium and occasionally a little Bisabol myrrh, so that
it often requires picking before the absolutely pure gum -resin can be
obtained. If a 10 per cent ethereal solution be prepared and a few
c.c. evaporated, the residue will at once become deep violet-black on ex-
posure to bromine vapour. The chemistry of myrrh is in a very chaotic
condition, and the formulae assigned to various constituents of it by
Tschirch can only be regarded as empirical, even if the constituents
themselves have been isolated in a pure condition. It is stated by
Tschirch to consist of 50 to 60 per cent of a gum of the formula
CgHjoOg ; two dibasic resin acids CiglligOg and C26H32O9 and a resene
CggHgi 02(011) 3. About 5 to 8 per cent of essential oil is also present.

0. von Friedrichs (" Archiv Pharm." 245, 427) gives the follow-
ing account of the chemistry of myrrh. The resinous portion of the
drug after distilling off the volatile oil was extracted with petroleum
ether ; the portion soluble in that solvent gives acetic acid on destruc-
tive distillation. The portion insoluble in petroleum ether, when
treated with ether, gave three soluble resin acids, a-, /3-, and y-com-
miphoric acids ; the fii'st two are isomeric, with the formula Ci4Hjg04 ;
y-commiphoric acid has the formula CijHggOg and is, therefore, iso-
meric with myrrholic acid. After saponifying the resin ester a mono-
basic acid, C28Hgg08, commiphorinic acid, was obtained. Two resin
phenols were isolated, both containing two hydroxyl groups ; they
were a-heerabomyrrhol. CigHg^Pg, and ^-heerabomyrrhol, C^^oH^^fif..
A monovalent volatile alcohol, G^^H.^2^,^, was also liberated by saponi-
fication. Heeraboresene was found to have the formula C42ll5g08,
and to contain a methoxyl group. The resin insoluble in ether con-
tains two acids, a- and y8-myrrhololic acids, the former has the formula
C15H22O7, the latter C25H320g. Both are monobasic. The gum,
which was dextrorotatory, {a] p + 23*78°, afiforded mucic acid on oxida-
tion with HNO3 and furfural on distillation with HCl. It probably
contains galactose and arabinose.

The drug when steam-distilled yielded 8*8 per cent of thick, light
yellow to greenish, very aromatic essential oil ; specific gravity 1-011 at
15° C. ; [a]'p - 73 "86°. It contains free formic and acetic acids, also
a non-volatile crystalline acid with the m.p. 159° C, which probably
exists in the drug as ester. After saponifying the esters another mono-
basic crystalline acid was isolated, myrrholic acid, CJ7H22O5, m.p.
236° C, separating from ether and benzol in small yellow crystals ; it is
soluble in most solvents, but not in benzol or in petroleum ether. This

MYRRH.

471

is isomeric with the y-commiphoric acid obtained from the ether-
soluble resin. It forms amorphous salts with silver, lead, and copper.
The oil contained metacresol, also cuminic and cinnamic aldehydes.
By fractionation over sodium under reduced pressure a new tricycHc
sesquiterpene, heerabolene, Cjr.H.^^, was isolated. No terpenes were
found in the oil distilled by the author, but pinene was found in a
commercial sample.

Lewinsohn ("Archiv. Pharm." 244,412) describes the essential
oil as bright yellow, neutral, and having the specific gravity 0*997 at
20° C, and I'OOl at 15° C. ; a^ - 70° 25' at 20° C. Three commer-
cial samples examined were reddish-brown in colour, and more or less
acid, the specific gravity was about 1*014, and the a^ ranged from
- 40° 3' to - 69° 5' at 18° C. The characters and constituents of
myrrh oil vary with age and method of distillation. Three of the
samples contained about 1 per cent of cuminic aldehyde. A fair
amount of eugenol and a little metacresol are also present ; also
penene, depentene and limonene ; and two sesquiterpenes having the
common formula C15H24. One has the specific gravity 0'926 at 20° ;
ajj + 22*75, and b.p. 163° C. under 12 mm. The other has the
specific gravity 0*911 at 21° C. ; ajj + 30° 4' ; b.p. 151 under 15 mm.
They have not been identified with any known sesquiterpenes, al-
though one resembles cadinene. When myrrh oil has been kept it
becomes acid and yields acetic and palmitic acids, due to the breaking
down of esters.

Myrrh should not contain more than 8 per cent of mineral matter —
usually from 5 to 6 per cent. The amount soluble in alcohol (90 per
cent), water, and petroleum ether may be determined, and also the acid
and ester values. These figures should be in accordance with the follow-
ing which were obtained by the author on six samples of myrrh freed
from all extraneous gum resins : —

1. Soluble in alcohol

33'8

41-9

37-5

2. Soluble in water .

29-5

31-2

40-5

38-5

3. Soluble in petroleum ether .

19-6

201

17-5

18-5

20-8

16-5

Acid value of (1) . . .

66-4

Este- value of (1) .

108

121

117

131

119

124

Acid value of the myrrh

20-5

Ester value ot the myrrh .

The nitric acid test for myrrh is, according to Greenish, best ap-
plied to the ethereal or petroleum-ether extract (" Pharm. Jour.
1901, II. 666). The extract is allowed to stand in an inverted dish
over the fumes of nitric acid when it gradually acquires a violet colora-
tion. Alcoholic solutions of myrrh such as the tincture are best diluted
with water and the dilute emulsion extracted with petroleum ether,
and the test applied to the residue left after evaporation of the petroleum
ether. Bisabol myrrh and bdellium do not give the re iction. Bromine
water or vapour gives a similar reaction, but not so well-marked.

Tucholka ("Year Book of Pharmacy," 1898, 180) gives the follow-

472 FOOD AND DRUGS.

ing test for Bisabol myrrh. A solution of 1 part of the sample in 15
of petroleum ether and 3 parts of glacial acetic acid is made, and
6 drops of this are cautiously mixed with 3 c.c. of str ;ng H^,S04. In
the presence of Bisabol myrrh a rose-red coloration appears at the
juncture of the liquids, and the whole of the acetic acid layer soon
acquires a red colour. With genuine myrrh, only a slight red colour
is acquired by the acid layer, whilst the line of contact is dull green.
Tincture of Myrrh is an extract of 4 ounces of myrrh by alcohol
(90 per cent) sufficient to produce 20 fluid ounces of the tincture. It
should have the following properties : —

Specific gravity . . . 0-848 to 0-858

Solid residue . . . 4 ,,6 grms. per 100 c.c.

Alcohol by volume . . 84 ,,86 per cent

It is to be noted that, as no official standard exists for the per-
centage of matter soluble in alcohol, in the gum resin, it is difficult
to condemn samples containing less than 4 per cent of solid residue.
Tinctures, however, prepared from a good myrrh will contain fully 5
per cent of solid residue.

PEPSINE.

The British Pharmacopoeia describes pepsine as an enzyme ob-
tained from the mucous lining of the fresh and healthy stomach of
the pig, sheep, or calf. It should dissolve 2500 times its weight of
hard-boiled white of egg when tested as follows : —

If 12*5 grms. of coagulated firm white of eggs, 125 c.c. of water
containing about 0*2 per cent of HCl, and 0-005 grm. of pepsine be
digested together at 105° F., for 6 hours, with frequent shaking, the
coagulated albumen dissolves leaving only a few small fiakes, in an
almost clear solution. The white of eggs should be prepared by
boiling quite fresh eggs for fifteen minutes, cooling, removing adhering
water with a towel, and at once rubbing the white through a sieve
having twelve meshes per centimetre, and at once using the product.
Pepsine is also required to be soluble in 100 parts of alcohol (90 per
cent).

Pepsine occurs as a powder or in scale form. Good specimens are
always pale in colour. It should not be very hygroscopic, otherwise
the presence of peptones is indicated. It should have no odour, and
should always be of faint acid reaction.

Many samples of commercial pepsine are mixed with sugar of milk
or powdered starch. Such samples will not satisfy the requirements
of the British Pharmacopoeia, and should not be sold as pepsine without
qualification. Many of them are honestly reduced to a standard
scrength, when the original pepsine is found to possess a very high dis-
solving power on albumen. The following analyses represent the com-
position of average samples of p6psine : —

PEPSINE.

473

Moisture
Pepsine (true)
Peptones
Mineral matters

A. H. Allen.

Parry.

Per cent

5-00

61-02

5-29

1-00

Per cent
4-00 to 6-5
62-706,, 69-55
2-54

0-9 „ 1-87

The examination of pepsine is almost confined to the determina-
tion of its value as a solvent of albumen, but when a complete analysis
is required, it is rarely necessary to do more than determine the
moisture, the ash, and the total nitrogen. If peptone is suspected, the
solution may be precipitated with zinc sulphate, and the filtrate from
this again precipitated with bromine. The nitrogen found in the brom-
ine precipitate multiplied by 6- 3 will give the approximate amount of
peptones. The moisture should not exceed about 5 to 6 per cent, and
the mineral matter should not be more than 1 to 1-75 per cent.

If not soluble in water and alcohol, the samples should be tested
for starch with iodine, and for sugar of milk in the usual manner.

In attempting any assay of pepsine for its proteolytic value, it is to be
remembered that the conditions of the experiment are very important, as
the formation of peptones eventually retards the action of the enzyme.
Hence different conditions of experiment will cause greatly different
results to be obtained.

Hercod and Maben, have, in a report presented to the 1910 Inter-
national Congress of Pharmacy at Brussels, made an exhaustive com-
parative study of the methods of pepsin assay of the principal
Pharmacopoeias. They have examined the methods official in the
following authorities ; the Belgian, British, German, Italian, Swiss
and United States Pharmacopoeias, and the French Codex, the follow-
ing being the quantitative requirements of each of these authorities in
reference to the assay process.

Acidity of
Digestive

Proportion

of Acid to

Pepsin per

cent.

Tempera-
ture of

Digestion
deg. C.

Duration
of Diges-
tion in
Hours,

Preparation of Al-
bumin.

Standard.

„.,_

Solution
per cent
HCl abs.

Egg-boiled
Minutes.

Sieve-
meshes
per cm.

1 Pepsin

digests

Albumin.

Belgian

0-25

250

100

British

0-2

5000

40-5

2500

German

0125

125

100

lltalian

009

38-40

1 to 2

100

iSwiss

0-2

200

1 to 2

100

jUnited States .

0-3

3600

3000

iFrench

0-25

150

—

25 (fibrin)

iStandard proposed

1 by authors .

0-25

250

2000

Hercod and Maben recommend the following method : —

474 FOOD AND DRUGS.

Take coagulated white of egg (obtained by boiling fresh eggs for
ten minutes), pass through a No. 40 sieve, and press between two-
sheets of filter-paper to remove surplus moisture ; weigh 10 grms., and
place it in a flask of 200 c.c. capacity, containing 100 c.c. of distilled
water previously heated to 52° C, 0*25 per cent absolute HCl, and
5 c.c. of a O'l per cent solution of pepsin. Place the flask in a water
bath at 52" C, and digest at that temperature for two hours, stirring
gently every fifteen minutes with a rotatory movement by means of a
glass rofl. At the expiration of two hours the albumin should be dis-
solved, the solution having an opalescent appearance.

To get a true idea of the value of pepsine, it is not sufficient to deter-
mine the amount of albumen dissolved, but also the amount of peptone
which has been produced. Weak samples of pepsine may dissolve a large
quantity of albumen but may only convert it into syntonin, whereas a
strong pepsine will carry the digestive process further, and convert it
all into peptones. Again, the colloidal nature of the substance causes
the action to take place at the surface where the pepsine meets the
albumen, no penetration taking place. So that the finer the particles
of albumen the greater the dissolving action. Still further, even when
albumen in an experiment appears not to be dissolved, it is usually in
an advanced stage of digestion so that it is difficult to estimate the
digestive action if any albumen remain undissolved. It is therefore best
to arrange experiments so that the end of the time reaction corresponds
with the solution of the whole of the albumen. Bartley ("American
Druggist and Pharmaceutical Record," Oct. 1893) has described a pro-
cess which in the author's experience gives exceedingly good results.
He eliminates the varying nature of egg albumen by using a solution
which contains the whites of several fresh eggs instead of the coagulated
albumen, and tests the liquid at regular intervals to see if conversion
is complete. His process is as follows : —

Solution No. 1. — Take the whites of several fresh eggs, mix them
thoroughly, and to 100 grms. of the mixed egg albumen add 900 c.c. of
distilled water, or in this proportion if smaller quantities are used..
Mix the solution well, and heat from three to five minutes. After
cooling, make up the mixture with water to the original volume.
The liquid may be strained, if necessary, through fine musHn ; but if
the eggs are fresh only a slight coagulum will form during the heating,
and will yield a slightly opalescent liquid, containing 10 per cent of
white of egg. As the latter contains, on an average, about 12-2 per cent
of dry albumin, 100 c.c. of this liquid will contain 10 grms. of egg-white,
or 1-22 grms. of dry albumen.

Solution No. 2. — Weigh out 1 grm. of the pepsin to be tested, add
25 c.c. of water, and then add 2 c.c. of diluted hydrochloric acid.
Now add water enough to make the solution up to 50 c.c, or if it be a
high-grade pepsin make up to 100 c.c. after adding 4 c.c. of diluted acid.

Procedure. — Measure out into a beaker or bottle 50 c.c. of the
albuminous liquid, and warm in a water bath to 35° to 40° C. (95° to-
104° P.). Now add to this solution 2 c.c. of diluted hydrochloric acid,
and from one-half to five c.c. of the pepsin solution. The more,
active the pepsin, the less the quantity to be taken. In the valuation

PEPSINE. 475

of high-grade pepsins it is best to use 100 c.c. of albumen solution,
containing 10 grms. of egg-white, and 1 c.c. of pepsin solution con-
taining 0*010 grm. of pepsin It may sometimes be necessary, with
an unknown pepsin, to perform a preliminary test to determine the
approximate time before spending too much time on an accurate
test. It is best to so regulate the quantity of pepsin and albumen
that the time shall be about two hours.

The time when the pepsin is added must be carefully noted, and
the temperature of the solution must be kept between 35° and 40° C.
(95° to 104° F.). At intervals of ten minutes, after the first hour, draw
out a few drops of the solution with a nipple pipette (dropper), and
float it upon a small quantity of pure nitric acid in a conical minim
glass. The digestion is incomplete as long as a white zone of coagu-
lated albumen appears at the line of contact of the two fluids. Note
the time when the nitric acid ceases to give this coagulation. This
end-reaction can generally be easily determined. In this manner three
elements in the calculation of the digestive power of the pepsin are
obtained, viz. : —

The weight of the egg-albumin, a,

The weight of the pepsin taken, p.

The time consumed, t.

As regards a standard time, the author fixes upon three hours as
the average time of stomach digestion. The relation between the

quantities of albumen and pepsin is expressed by the fraction p, i.e. it

is found by dividing the amount of albumen (5 grms. in the above
directions for weaker pepsins) by the amount of pepsin used when 1
c.c. of the solution above mentioned is taken for the test, viz. -02 grm.
This would give the amount of albumen digested by 1 part of pepsin
in the observed time of the experiment as 250 grms. But the time
is not the standard time. Assume that the time required for the
digestion was two hours. The relation of this to the standard time,
three hours, would be |. The above result must then be multiplied
by this ratio in order to give the amount of albumen capable of being
digested in the standard three hours. Expressed in the form of an

A 3

algebraic equation we have : D (digestive power) =— x =, and substitu-
ting the above values : —

D^.^'^-r X # = .i| = 375 grms., showing that 1 grm. of this pepsin is
capable of digesting 375 grms. of egg-albumen in three hours, or 750
grms. in six hours.

As egg-white contains about 12*2 per cent of dry albumen, 1 grm.
of this pepsin will digest 45-75 grms. of dry albumen in three hours,
or 91*5 grms. in six hours.

The advantages claimed for this process over other methods
are : —

1. The shorter time consumed.

2. Uniformity in results.

3. The avoidance of the necessity for shaking the solution during
digestion.

476 FOOD AND DRUGS.

4. A more exact statement of results.

5. The weaker solution of albumen used causes less interference
with the action of the pepsin by the peptone formed.

Stebbings ("Analyst," xiv. 197, i510, 229) recommends the pro-
cess suggested by Kremel. Egg albumen in scales is dried at 40° and
powdered. One grm. is treated in a 100 c.c. flask with 0*1 grm. of
the pepsine and 50 c.c. of 0-2 per cent HCl. The liquid is kept at 40°
for three hours. It is then neutralized with alkaline carbonate, heated
to 90° C, and cooled after coagulation is complete. The liquid is then
made up to 100 c.c. and 50 c.c. filtered off and evaporated to dryness.
This residue represents the albumoses and peptones formed and is
probably the truer measure of the digestive power of the pepsine, than
any experiment which determines the amount of albumen dissolved,
of which much is only converted into syntonin. From the weight of
the peptones, etc., thus determined, a deduction must be made for the
amount of mineral matter present, and also for the amount of pepsin
in solution, which may be determined by a blank experiment without
the albumen.

Allen ("Analyst," xxii. 258) prefers the following process. About
1 grm. of egg albumen in scales is powdered and treated with 20 c.c.
of water in a 100 c.c. flask. When it is dissolved the liquid is heated
in a water bath to coagulate the albumen and cooled to 40° C. 0*1 grm.
of the pepsine is then added and also 25 c.c. of decinormal hydrochloric
acid. The liquid is then warmed to 40° C. for three hours. The hquid
is neutralized by sodium carbonate solution, and is then heated to 90°
C. for ten rninutes. It is then cooled, made up to 100 c.c, and filtered.
The precipitate consists of syntonin and any unaltered albumen, while
the filtrate contains peptones and albumoses.

Fifty c.c. of this latter are saturated with zinc sulphate, allowed
to stand for half an hour, with occasional agitation, and then filtered.
The precipitate is washed with cold saturated solution of zinc sulphate,
and the filtrate made up to 250 c.c. with water slightly acidulated
with HCl, and the filtrate treated with excess of bromine water.

The albumoses may be calculated from the amount of nitrogen in
the zinc sulphate precipitate and the peptones from that in the
bromine precipitate (see under Extract of Meat, p. 405). An allow-
ance must be made for the amount of nitrogen present in the pepsine
used.

The proportion between the amount of albumen dissolved and the
amount of true peptones and albumoses formed is very small, and it
must be remembered that mere solution processes are rather a
measure of the amount of syntonin formed than of peptones.

CANADA TURPENTINE.

This oleo-resin, better known as Canada balsam, is official in the
Pharmacopceia. It is an oleo-resin obtained from Abies balsamea. It
is officially required to solidify when mixed with about one-sixth part
of its weight of magnesia moistened with a little water.

Canada balsam contains about 25 per cent of an essential oil con-

LIQUID TAR 477

sisting almost entirely of terpenes, principally lasvo-pinene. This oil
is laevorotatory and boils at 160° and almost completely distils be-
low 170°. The balsam also contains about 60 per cent of resin acids,
which have been described under the names canadinic acid Cj^Hg^Og ;
canadolic acid Cj^H^gOg ; and canandinolic acid CjgHg^Og. An in-
different resene, canadoresene C2iH^(jO, has also been isolated, and oc-
curs to the extent of about 6 per cent.

Canada balsam should have the following characters : —

Specific gravity at 15° 0-985 to 0-995

Kefractive index 1-5200 (about)

Essential oil 20 to 25 per cent

Optical activity . + 1° to + 5°

Optical activity of essential oil laevorotatory

Acid value 70 to 90

Ester value . . . . . . . . 4 „ 15

Acid value of the oil-free resin 100 „ 120

Colophony is a frequent adulterant, mixed with ordinary turpentine
oil. Such mixtures will generally give higher acid values, and the es-
sential oil may be dextrorotatory.

BUEGUNDY PITCH.

This resin is officially described as the resinous extract of the
stem of Picea excelsa.

It consists principally of pimaric anhydride, with, a small amount
of essential oil.

The analytical examination of this substance is not well under-
stood, but genuine samples examined by the author show that it
should have the following characters : —

Acid value 130 to 145

Ester value ...... under 20

Iodine value about 120 to 130

It should be soluble in twice its weight of glacial acetic acid.

Many samples of so-called Burgundy pitch consist merely of com-
mon rosin, pitch, and turpentine. Most of these are not completely
soluble in twice their weight of glacial acetic acid.

LIQUID TAB.

Under the name Pix liquida, the bituminous liquid obtained by
the destructive distillation of Pinus sylvestris and other species of
pine, is official in the British Pharmacopoeia. It is known com-
mercially as Stockholm tar.

The official requirements are that its speciJQc gravity should be
from 1-020 to 1*150. If it be shaken with water, the water acquires
an acid reaction and gives a red colour with dilute ferric chloride
solution. It should be completely soluble in 10 volumes of 90 per
cent alcohol.

Genuine Stockholm tar should be soluble in an equal volume of
absolute alcohol, ether, or chloroform, and almost entirely soluble in

478 FOOD AND DRUGS.

3 volumes of 5 per cent solution of potash. It is completely soluble
in 96 per cent acetic acid, which distinguishes it from other tars ex-
cept beechwood tar. If 1 volume be well shaken with 5 volumes of
petroleum ether and the petroleum separated and shaken with an
aqueous solution of cupric acetate (0*1 per cent), the petroleum ac-
quires a green colour, due to the formation of soluble copper salts of
the tar acids. Beechwood tar does not give this reaction.

RESIN.

Resin or colophony is an official drug. It is the residue left after
the distillation of the oil of turpentine (q.v.) from the crude oleo-
resin of various species of Pinus.

The official requirements are that it should be soluble in 90 per
cent alcohol, ether, benzol, and carbon disulphide, and that it should
leave no appreciable ash.

The bulk of the resin of commerce is obtained from the crude
American turpentine, but French turpentine yields a very high-grade
product also.

The chemistry of this resin is in an unsettled state, but it is clear
that it consists of several resin acids either isomeric or closely related.
It is generally agreed that abietic acid is the typical resin acid present.
The probable formula for this acid is C^oHg^O^i and it is possibly
identical with sylvic acid. Some investigations tend to support the
theory that the acids are present as anhydrides, but the ready solu-
bility in alkaline solution is against this hypothesis. At all events
from the analytical point of view, it is certain that resin consists
almost entirely of abietic or closely allied acids, with traces of esters
and up to 5 per cent of neutral resins or resenes. Pure resin should
have the following characters : —

Specific gravity at 15°
Acid value
Ester value
Iodine value .
Unsaponifiable matter
Specific rotation in alcohol

The very low price of resin makes it more interesting as an
adulterant than anything else — for, after the examination of many
hundreds of samples, the author has never found one adulterated.

The following details of the characteristics of resin, therefore, will
-be of interest': —

A valuable reaction for colophony, suggested by Liebermann, but
modified by Storch and Morawski, consists in treating the substance
with acetic anhydride, cooling the liquid, and separating the acetic
anhydride. Sulphuric acid of 1*5 specific gravity is then allowed to
flow gently into the tube containing the acetic anhydride, when a
reddish- violet colour will be immediately produced at the junction of
the two liquids if colophony be present. The colour soon changes to
reddish- brown.

It is frequently necessary to separate fatty acids and resin acids.

1-070 to 1-085

150

„ 182

„ 20

118

„ 130

about 6

„ 8 per cent

+ 58°

„ + 68°

EESIN. 479

This is best done by the following method which is due to Twitchell
and Gladding.

About 5 grms. of fatty and resin acids are boiled with excess of
alcoholic potash for half an hour under a reflux condenser. The
alcohol is then evaporated, the residue dissolved in water, and un-
saponifiable matter removed by agitation with ether. The aqueous
liquid is separated and acidified with hydrochloric acid. The
separated acids are removed by shaking with ether ; the aqueous acid
solution is neutralized, .evaporated to about 25 c.c, re-acidified, and
shaken out with ether. After distilling off the ether from the united
ethereal extracts, the residue of resin and fatty acids is dissolved in
50 c.c. of absolute ^alcohol, and the fatty acids converted into esters
by passing a moderately rapid current of dry hydrochloric acid gas
through the solution cooled by ice-water to a temperature not above
10° C. When the operation is complete (which is usually the case
in from- one to two hours), the liquid is allowed to stand for half an
hour at the ordinary temperature. It is then diluted with five times
its volume of water, and boiled under a reflux condenser for half an
hour. The cool solution is agitated with several successive quantities
of ether until the extracts are colourless. The aqueous liquid is neutral-
ized, evaporated to 50 c.c, acidified and repeatedly extracted with small
quantities of ether to recover the water-soluble constituents of colo-
phony. The mixed ethereal solutions are shaken out with about 50
c.c. of a solution containing 10 grms. of caustic potash, 10 grms. of
alcohol and 100 c.c. of water, when a brown layer usually separates
out between the ether and the alkaline solution and is drawn off with
the latter. This layer contains a considerable portion of the resin-
soap, which is only slightly soluble in the potash solution. The
ether is shaken with water to remove soluble resin-soaps ; then with two
successive quantities (10 c.c.) of the potash solution ; and finally with
water until the washings are colourless. The alkaline liquid is now
acidified and agitated with ether until completely extracted. The
acid solution is neutralized, evaporated to a small bulk, re-acidified,
and again shaken out with ether. The total ether extracts are
washed with 20 c.c. of water, and the ether distilled off. The residue
of resin acids so obtained — still contaminated with unchanged fatty
acids — is treated with several small successive additions of absolute
alcohol to remove the last traces of water, and weighed. The fatty
acids still remaining in the resin-acids are removed by Gladding's pro-
cess. From 04 grm. to 0-6 grm. of the resin-acids, obtained as above,
should be placed in a 100 c.c. stoppered and graduated cylinder, and
dissolved in 20 c.c. of 95 per cent alcohol. A drop of phenol-phthalein
solution is added to the alcoholic solution, and then concentrated
caustic soda solution (1 of NaOH to 2 of water) until the reaction is
just alkaline. The loosely-stoppered cylinder and its contents are
heated for a short time in the water bath, then cooled, and ether
added up to the 100 c.c. mark. One grm. of dry powdered silver
nitrate is added, and the contents of the cylinder are shaken for
fifteen minutes to conveit the fatty and resin acids into silver salts.

When the insoluble salts have completely settled (preferably after

480 FOOD AND DEUGS.

standing overnight), 70 c.c. of the solution should be pipetted into
a second 100 c.c. cylinder and shaken with 20 c.c. of dilute hydro-
chloric acid (1 : 2). The ethereal layer is drawn off, and the aqueous
liquid twice shaken with ether. The united ether extracts are washed
with water, filtered, and the ether distilled off. The residue, amounting
to about 10 c.c, is evaporated, dried for a short time at 100° to 115"
C, and weighed. The weight of the resin acids so found is calculated
back into the first weight (impure acids) obtained, and then on the
original substance taken. The percentage found is corrected by the
subtraction of 0*4 per cent, this allowance being made for a small
amount of unesterified fatty acid, which is always present. As colophony
contains an average of 8 per cent of unsaponifiable matter, a second
correction is necessary, the true percentage of colophony in the sub-
stance under examination being found by the following equation, in
which the corrections are combined : —

percentage of resin acids found - 0*4
100 02 = percentage of colophony.

A useful method for separating the resin acids (which represent
90 per cent of the colophony) from other resins which yield silver
salts insoluble in ether, such as shellac, etc., is that used by the author
for shellac analysis. The process is as follows : —

About 0'5 grm. of the sample is dissolved in the smallest possible
quantity of alcohol and the solution cautiously treated with alcoholic
potash till it is just neutral to phenol-phthalein. This solution, con-
taining the potash salts of the acids, together with the neutral con-
stituents of the sample, is poured into about 100 c.c. of water
contained in a separator, and about 0*5 grm. of silver nitrate, dis-
solved in a little water, added. The acids are precipitated as silver
salts, and on shaking the Hquid twice with ether, the silver salts of
the resin acids of colophony are completely dissolved, while the silver
compounds of the other acids remain insoluble. The ethereal
solution is filtered, repeatedly agitated with water to remove silver
nitrate, dilute hydrochoric acid added, and the liquid well shaken.
The silver salts are decomposed, silver chloiide being precipitated, and
the resin acids recovered by evaporating the washed ethereal solution
to dryness.

THUS.

Thus, or gum thus as it is known commercially, is the so-called
American frankincense. It is an official oleo-resin scraped from the
trunk of Pinus palustris.

No official standards exist.

As a matter of fact, thus is collected from other species of Pinus,
and so far as the author can ascertain is nothing other than crude
turpentine from which a portion of the essential oil of turpentine has
evaporated, leaving a crude concrete oleo-resin containing rather less
essential oil than the oleo-resin ous turpentine. Genuine samples ex-
amined by the author were treated to drive off the small quantity of
oil of turpentine, and the residues were then found to be indis-
tinguishable analytically from common resin, or colophony.

SCAMMONY. 481

SCAMMONY

Scammony (Scammonium of the British Pharmacopoeia) is a
natural gum resin obtained by incision of the Hving roots of Convol-
volus scammonia. The root itself is also official in the Pharmacopoeia,
and contains the following : —

Per cent

Resin 6 to 9

Extractive matter 12 ., 15

Starch 7 ,, 8

Mineral matter 9 „ 13

The pure resin of scammony prepared from the dried root is also
official as scammony resin. This should be entirely soluble in 90 per
cent alcohol and practically free from ash.

The principal constituent of scammony is scammonin Cg^H^gOig
which is closely related to or probably identical with the glucoside of
tampico jalap. It is a glucoside, melting at at 131° and having a
specific rotatory power - 23°.

Pure scammony, i.e. the crude gum resin obtained naturally, forms
masses of varies sizes of a brown, dark grey or nearly black colour.
It forms a grey powder when pulverised, which should yield only the
slightest reaction for starch, and should contain, according to the
Pharmacopoeia, at least 70 per cent of resin soluble in ether and not
more than 3 per cent of ash. An alcoholic solution should not afford
a blue colour with solution of ferric chloride. No other Pharma-
copoeial standards are given.

A genuine scammony will certainly rarely contain less than 70
per cent soluble in ether when powdered and extracted in a Soxhlet
tube, but pure samples will often contain as much as 6 per cent or
even 7 per cent of mineral matter. The greater part of the scam-
mony of commerce is grossly adulterated, especially the so-called
Aleppo scammony. Starchy matter is the principal adulterant, which
is detected under the microscope, or by testing the cooled aqueous,
decoction with iodine ; chalk is sometimes present, which will raise
the ash value, and will cause a little of the powdered drug to effervesce
with dilute hydrochloric acid. Lead sulphide is occasionally found.
The resin, extracted by means of ether (or preferably a mixture of
85 per cent of ether of specific gravity 0'735 and 15 per cent of 90
per cent alcohol) should be examined, and should have the following
characters : —

Acid value 14 to 21

Ester value 200 „ 226

Iodine value (Hiibl) 10 „ 15

Taylor (" Amer. Journ. Pharm." 1909, 81, 105) has examined a
number of genuine scammony resins and of the so-called " Mexican "
scammony resin which is obtained from the root of Ipomosa Oriza-
bensis. He gives the following figures : —
VOL. I. 31

482

FOOD AND DRUGS.

Resiu

Acid

Ester

Iodine

True scammony

per cent.

Value.

8-1

21-1

211-3

13-3

»» »

7-93

15-5

222-5

10-8

8-06

15-6

219-8

13-0

7-71

18-2

221-7

14-3

»» •>

8-52

18-8

218-1

14-6

Mexican scammony

16-75.

15-5

171-1

8-7

16-83

21-5

165-6

11-5

According to Guignes (" Bull Soc. Chem." 1908 [iv.] 3, 872) the
specific rotation of scammony resin affords a means of detecting certain
adulterants. He gives the following values for alcoholic solution : —

Scammony resin extracted from the gum resin, [0]^= up to- 24° 30'

„ „ „ „ roots " = - 18° 30' to - 23° 30'

Tampico jalap resin = - 34° 20'

Orizaba jalap resin = - 24° 45'

Jalap resin = - 30° 10' to - 36°

Colophony =+6° to+ 7°

Sandarac =+31° to + 34°

Mastic = + 29° 30'

Guaiacum resin = - 17°

The author has examined numerous samples of Mexican scam-
mony root and finds that it contains from 14 to 20 per cent of resin.
The ester value is a most valuable method of discriminating between
the genuine resin and that from Mexican scammony, and as many
tons of Mexican root are imported annually into this country, it is
necessary to carefully examine samples of the resin, many commercial
specimens of which are made entirely from Mexican root.

These characters will ensure the absence of colophony, and
guaiacum resin is detected by the blue colour imparted to an alcoholic
solution by ferric chloride solution.

If the presence of guaiacum resin be proved, its amount can be
determined approximately by an estimation of the methoxy value of
the resin. Pure scammony resin has a value of 0 to 2, whilst
guaiacum resin gives a figure of 72 to 85.

This value is determined in the following manner : About 0'3
grm. to 0*4 grm. is treated in a glycerine bath at 120° to 140°
with 10 c.c. of hydriodic acid (1*70 specific gravity) in a flask of 40
c.c. capacity, connected with three bulbs, the first being empty, the
second containing water, and the third water with red phosphorus
in suspension. After passing through the bulbs, which absorb
hydriodic acid and iodine, the alkyl iodide is absorbed in a flask con-
taining 5 c.c. of a 40 per cent aqueous solution of silver nitrate and
60 c.c. of alcohol ; a second flask with half the quantities of silver
nitrate and alcohol may be added as a precaution. The mixed silver
solutions are rendered acid with dilute nitric acid, and the silver
iodide filtered off and weighed. The weight of silver iodide multiplied
by 0-132 gives the amount of methoxyl CH3O, the methoxyl number

SENNA. 483

indicating the number of milligrams of CH3O in 1 grm. of the sub-
stance.

SENNA.

Two varieties of senna leaves are official in the Pharmacopoeia,
those of Cassia acutifolia, known as Alexandrian senna ; and those of
Cassia afigustifolia, known as East Indian or Tinnivelly senna. No
standards are given.

The constituents of senna are, in spite of very numerous investiga-
tions, but poorly understood. There appear to be present bodies either
isomeric or identical with emodin, iso-emodin and chrysop'hanic acid.
These are known as senna-emodin, etc., and are probably the result of
the decomposition of glucosides, which are usually present in small
quantity in the dried leaves, although the greater portion of these has
decomposed.

The only available means of examining senna are the determination
of the ash, and a microscopic examination.

The ash of senna leaves usually varies between 9 per cent and 14
per cent. If a higher ash be found in a powdered senna, it may still
be genuine,, but is probably of very inferior quality, containing much
" sif tings ". The ash should be almost entirely soluble in hydrochloric
acid.

Greenish describes the microscopic characters of the powder as
follows : —

The powder exhibits fragments of epidermal tissue consisting of
polygonal cells and bearing stomata and hairs or the scars of fallen
hairs. Each stoma is enclosed between or bordered by two cells, ar-
ranged parallel to it ; the hairs are one-celled, thick- walled and warty.
It also exhibits groups of sclerenchymatous fibres, which, however,
should not be present in excessive quantity.

In addition to these characters, characteristic rows of cells with
prismatic crystals are to be found. An examination of the sample
side by side with the powder of an authentic specimen will enable
most possible adulterants to be detected easily.

According to Sayre (" Amer. Jour. Pharm." 1896, 585) the characters
of the hairs enable one to distinguish between the two varieties of leaves.

Alexandrian senna is more hairy than Indian, a mixture of equal
parts of the two in No. 60 powder containing ten hairs of the former
to one of the other variety. The hairs of Alexandrian senna have a
sharp curve near the base, while those of Indian senna are straighter,
shorter, and stouter. Almost invariably the hairs of both kinds re-
main unbroken when the drug is powdered, and these distinctions
may therefore be of practical value. The epidermal cells also differ
in form, those of Indian senna being somewhat smaller and more
uniform in size, whilst the angles are more acute than in the Alex-
andrian variety. Careful measurement of the oells gives the following
average results : Indian senna, 35 micromillimetres in diameter ;
Alexandrian senna, 40 micromillimetres.

Compound Tincture of Senna is an official drug, the characters of
which will be found in the table on p. 496.

484 FOOD AND DRUGS.

Compound Mixture of Senna, or black draught, is also an official
preparation of senna. It is a mixture of liquid extract of liquorice,
compound tincture* of cardamons, aromatic spirit of ammonia, infusion
of senna, and sulphate of magnesium. The usual adulterant is an excess
of its cheapest ingredient, magnesium sulphate. A genuine mixture
should have the following characters : —

Solid residue 16 to 16-5 per cent

Alcohol by volume .... 9*0 „ 9-5 „

Magnesium sulphate (MgS047H20) . 25 grms. per 100 c.c.

AROMATIC SPIRIT OF AMMONIA.

This official drug is an aromatic alcoholic solution of ammonia
and ammonium carbonate.

The official requirements for the drug are as follows : —

Specific gravity 0"888 to 0*893. Twenty c.c. require 25*5 c.c. of
normal sulphuric acid for neutralization, corresponding to about 2*4
per cent of NH3, or 2-16 grms. in 100 c.c. Twenty c.c. after the
addition of 16 c.c. of a solution of barium chloride (10 grms. per 100
c.c), should yield a precipitate which becomes more copious on heating
to 71° C, and after filtering, the filtrate should yield a further precipi-
tate when more of the barium chloride solution is added, and the
liquid is again heated.

White (" Pharm. Journ." 4, x. 144-148) has shown that the state-
ments in the British Pharmacopoeia in reference to the strength of the
solution of ammonia directed to be used in this preparation, are incor-
rect. The Pharmacopoeia states that ammonia solution, of specific
gravity 0*891 contains 32*5 per cent of NHg. He has shown that
31*6 per cent or a figure very near to this is the true value. He
also shows that, whereas the Pharmacopoeia requires ammonium
carbonate to be of such strength that 1 grm. should require at least
18*7 c.c. of normal HgSO^ for neutralization, no samples are to
be found which require more than 18*2 c.c. From these facts it
follows that the calculated alkaline value given by the Pharmacopoeia,
viz. that 20 c.c. should require 25*5 c.c. of normal H2SO4 for
neutralization is incorrect, and samples must be made with rather
more ammonia than directed, in order to come up to the official re-
quirements.

The barium chloride test has been shown to be quite unreliable and
has been criticized severely by White and by F. C. J. Bird.

It has been shown that any quantity between 18 and 23 c.c. of
the spirit will behave as directed when 20 c.c. are used according to
official directions. By the addition of a little ammonium chloride,
the reaction is, as pointed out by Bird, rendered reliable. He re-
commends the test to be carried out as follows : —

To 20 c.c. of the aromatic spirit of ammonia add 5 grms. of
ammonium chloride, agitate vigorously, and add 16 c.c. of solution
of barium chloride. Warm to 160° F., cool to normal tempera-
ture, and filter. The filtrate, on the addition of more barium chlo-
ride and warming gives no further precipitate. Should a slight

SPIRIT OF NITROUS ETHER.

485

opalescence be produced by the barium chloride it should disappear
completely on heating, but any precipitate of barium carbonate would
remain permanent. Twenty-one c.c. of aromatic spirit of ammonia
tested under exactly similar conditions will be found to yield a pre-
cipitate on the further addition of barium chloride, which does not
disappear on warming. The reaction appears to be quite complete at
the time of filtering, as is evident from the following table : —

16 c.c. Barium Chloride Solution.

Spt. Am. Ar.
taken.

Filtrate on
Standing.

BaClg to Portion of Filtrate.

(NH.kCO^to
Portion of Fil-
trate.

20 c.c.

21 c.c.

22 c.c.

No ppt.
No ppt.
No ppt.

Faint opalescence which disappears

on warming.
Peimanent ppt. on warming and

allowing to stand for ten

minutes.
Copious ppt.

ppt.
No ppt.

No ppt.

In examining this drug, it must be remembered that a deficiency
in alkaline strength of from 1 per cent to 5 per cent of the total neces-
sary, may be due to the fact that, in the manufacture of the spirit,
the directions of the Pharmacopoeia have been minutely followed, the
deficiency being due to the errors in that authority. If the spirit be
of full strength, the directions of the Pharmacopoeia have probably
not been literally adhered to. i

SPIRIT OF NITROUS ETHER. ;

This drug is ofl&cially directed to be made by distilling a mixture;
of alcohol, sulphuric and nitric acids, and copper.

The official requirements for the drug are as follows : —

Specific gravity 0-838 to 0-842. If the spirit be poured on to a
layer of acid solution of ferrous sulphate (10 per cent) a deep olive-
brown colour is produced at the surface of contact of the liquids. Ten
c.c, mixed with 5 c.c. of normal soda solution, and 5 c.c. of water
should become yellow, but should not turn brown on standing for
twelve hours (limit of aldehyde). It should not effervesce, more than
very faintly, when shaken with sodium bicarbonate. One volume
agitated briskly at intervals during five minutes in a brine-charged
nitrometer, with 1 volume of 10 per cent solution of potassium iodide
and 1 volume of dilute sulphuric acid should yield at normal tempera-
ture and pressure, and when freshly prepared, from 6-25 to 7 volumes
of nitric oxide gas, corresponding to 2*5 per cent of ethyl nitrite. And
even after it has been kept some time and the vessel containing it
has been occasionally opened, it should yield not much less than five
times its volume of gas, corresponding to nearly 2 per cent of ethyl
nitrite, or a minimum of 1-75.

A more unscientific and unsatisfactory standard than the above

486

FOOD AND DRUGS.

would be very difficult to invent. It necessarily fixes the amount of
1-75 per cent of ethyl nitrite as the bottom limit, and so long as the
sample is of this strength no exception can be taken to it, in spite of
the standard for freshly prepared spirit which is 2-5 per cent.

Spirits of nitrous ether, or sweet spirit of nitre as it is now also
termed alternatively in the Pharmacopoeia, is a liquid of very complex
composition.

It contains ethyl nitrite, alcohol, aldehyde, ethyl acetate, ethyl
nitrate, free acids, and water. Various other compounds are also
present in traces.

The literature of this drug is voluminous, the greater part of it
being devoted to proving that it is very unstable and is very liable to
deterioration. This is well recognized to be true, and the majority of
successful prosecutions for selling this drug below its proper strength
are certainly due to the deterioration of an originally genuine
article.

MacEwan (" Pharm. Journ." 3, xiv. 817) has given the following
figures, which show the efifect of keeping on the drug : —

•

Ethyl Nitrite.

HNO2

Acetic Acid.

Aldehyde.

Per cent

(1) B. P. Spirit (old) .

0-87

0-47

1-20

0-80

(2) „ „ (1 week old)

3-54

0-22

0-21

0-85

„• ., (2 „ „)

—

0-26

0-25

0-95

„ „ (3 „ „)

314

0-27

0-35

—

(3) B. P. „ (2 days old)

2-01

—

0-80

„ (4 „ „)

—

0-24

0-22

1-14

„ „ (7 „ „)

1-24

0-32

0-25

2-00

(4) B. P. „ (1 month .

old) ....

1-93

0-24

0-41

1-67

(5) London Pharmacopoeia

(4 months old)

3-53

0-16

0-29

1-50

(6) London Pharmacopoeia

(4 months old)

1-64

0-35

0-49

1-43

(7) London Pharmacopoeia

(4 months old)

0-22

0-19

0-25

0-20

The principal factors which cause the amount of ethyl nitrite to
be reduced on keeping are the traces of water present, which cause
decomposition to be rapid ; exposure to air, light and excessive tem-
perature.

Although the standards laid down for this drug must be adhered
to for official purposes, it is a fact that very carefully prepared samples
may have a specific gravity up to 0*848 or even 0'850. No exception
would be taken to this, so long as the amount of ethyl nitrite is
maintained.

Free Acid. — The amount of free acid is determined, as recom-
mended by MacEwan, in the following manner : —

Ten c.c. of the sample are placed in a flask with a drop of phenol-
phthalein solution, and a few drops of solution of methyl-orange are

SPIKIT OF NITEOUS ETHER. 487

added. A porcelain slab is also spotted with drops of methyl-orange
solution. Semi-normal soda solution is run in until the pink colour
of the acid solution and methyl-orange begins to change, when a drop
is removed by a glass rod and brought into contact with a spot of
methyl-orange solution. If the spot assumes a pink tint the nitrous
acid is not quite neutralized, in which a little more alkali is lun in
until a spot of methyl-orange is rendered only faintly pink. The
amount of alkali used is noted, and the titration continued until
neutrality is indicated by the pink colour of the phenol-phthalein.
Each c.c. of semi-normal alkali used for producing neutrality to
methyl-orange = 0-0235 grm. of HNOgi and each c.c. of additional
alkali used is equivalent to 0-030 grm, of acetic acid. The results
are sufiQciently approximate for all practical purposes.

Aldehyde is best determined by Thresh 's colorimetric process. Ten
c.c. of the sample are diluted with 20 c.c. of water and 3 c.c. of a
saturated solution of caustic soda added. The mixture is heated to
boiling-point for a few seconds, then cooled and after two hours is
diluted with 20 c.c. of warm alcohol (free from aldehyde) and made
up to 60 c.c. with water. The liquid is quite clear and of a reddish-
yellow colour. As the colour soon alters, it is best to immediately
make a solution of potassium bichromate to match the colour. A
solution of aldehyde containing 1 per cent of aldehyde in pure
alcohol is then treated in the same manner and the colours of the
two matched by dilution of the one having the deeper colour as in
the process of Nesslerizing. The amount of aldehyde present is thus
calculated with approximate accuracy.

Estimation of Ethyl Nitrite. — Eykman's process is a most accurate
one for the determination of ethyl nitrite. The following, according
to A. H. Allen, is the most reliable method of carrying out this process : —

Take a small flask A, tubular in shape with a round bottom, and
insert a tight-fitting rubber-stopper, through which passes a narrow
glass tube B. This tube should extend nearly to the bottom of the
flask terminating in a turned-up point to prevent any gas from enter-
ing. The rest of the tube outside the flask should be bent over and
joined to a long, narrow vertical tube by an india-rubber joint. This
tube should also terminate in a point, and when placed in a conical
glass D should nearly reach the bottom. From one side of the flask
should branch a tube E which is connected with the stopper of a
Lunge's nitrometer G by means of a few inches of india-rubber F.
Take a solution of soda of about 1-10 specific gravity specially prepared
for the experiment by being previously shaken with a small quantity
of ferrous sulphate, thus ensuing freedom from dissolved oxygen, and
allowing the precipitated oxide of iron to subside. Use a solution of
ferrous sulphate containing lOO grms. of the powdered crystallized salt
in 500 c.c. of water with 0-5 c.c. of strong sulphuric acid.

Pour about 30 c.c. of the iron solution into the flask. Wet the
india-rubber cork well and insert firmly in the neck. Then connect
with the nitrometer, which contains a small quantity of soda solution
in the cup, and see that the tap of the nitrometer is closed. The
glass D should contain the solution of iron into which the tube C

488

FOOD AND DEUGS.

must be immersed. The screw-clip at H must be left open. Heat
the flask to expel the air through C, then remove the flame and allow
about 30 c.c. of iron solution to enter the flask, firmly closing the clip
at H. Heat the contents of the flask to boiling. When the india-
rubber at F begins to swell, open the tap at G, and allow the air from
the flask to bubble through the soda solution in the cup of the nitro-
meter. Close the tap G when all the air has been expelled, remove
the flame and allow the contents of the flask to cool. Put 5 c.c. to 10
c.c. of the sample (according to its strength) in the glass D with 10 c.c.
to 20 c.c. of water containing 1 or 2 grms. of common salt. Then

Fig. 44. — Eykman's apparatus for nitrous ether.

very carefully open the clip F and allow the liquid to flow into the
flask until the opening of tube C is covered. Pour a little iron solu-
tion into glass D, also 5 c.c. of the dilute sulphuric acid, and allow
this to pass into the flask. Continue to do this until the glass and
tube have lost their brown colour, at the same time being careful not
to allow any air to enter the flask. Heat the contents of the flask
to boiling, having previously closed the clip at H. Turn the tap G,
to open connexion between the graduated tube K of the nitrometer
and the flask, as soon as the india-rubber joint F shows signs of pres-
sure. This reaction produces nitric oxide gas. which passes into K
where it is collected. As soon, however, as the contents of the flask
are no longer brown, the tap G is closed and the clip at H opened
simultaneously, thus forcing back the liquid into A. The apparatus is
then ready for another experiment. When the liquid has had time to
assimilate the temperature of the air, i.e. in about half an hour, notice
the volume of gas in the nitrometer, being careful that the level of the
liquid in the tube L is identical with that in K.

The following formula expresses what has occurred : —

2C,H5NO., + 2FeS04 -i- H^SO^ = FeCSOJg + 2C.,H,0 -f 2N0.

The percentage of ethyl nitrite found in the volume of nitric oxide
obtained can be calculated as follows, when v represents the number

SPIRIT OF NITEOUS ETHER 489

of c.c. of gas obtained,^; the barometric pressure in mm., e the tension
of aqueous vapour at the temperature at which gas is measured,
d the density of the sample (water = 1) ; n the number of c.c. em-
ployed ; and t the temperature in centigrade degrees : —

aH.NO., = -^^ -^^ X 0-1207.
2 ^ ^ dxn 273 + «

When it is not necessary to have strictly accurate results, omit
the corrections for pressure, temperature, and tension of aqueous
vapour. The calculation will be much simpler. If the volume
of 0-030 grm. of nitric oxide (representing 0-075 grm. of C2H5NO2)
under the ordinary conditions of pressure and temperature be taken
at 23-55 c.c, then

volume of gas in c.c. x 0'3184 _ f h * hf

measure of sample in c.c. x density of sample & j 5

ofC^H^NO.,.

The official process with potassium iodide and dilute sulphuric
acid is, however, quite accurate enough for most purposes. The nitro-
meter should be charged with a saturated salt solution and the end
immersed in the same liquid. Five c.c. of the spirit should be placed
in the cup and carefully drawn in, avoiding the inclusion of any air
bubbles. Five c.c. of a 10 per cent solution of potassium iodide is
then drained in, and this is followed by 5 c.c. of dilute sulphuric acid.
It is advisable to place about 6 c.c. of the latter in the cup and leave
a little therein, as this will guard against the possibility of air bubbles
being drawn in, which would now be unnoticed on account of the
presence of nitric oxide in the top of the nitrometer. The reaction
is as follows : —

C.,H5(N0.,) + KI + H2SO4 = C2H,(0H) -t- KHSO^ + 1 + NO

from which the amount of ethyl nitrite can be calculated. The nitro-
meter should be read off at about 15" C, after adjusting the levels of
the liquids inside and outside by lowering or raising the nitrometer as
may be found necessary.

Dott (" Pharm. Journ." 3, xv. 492) has proposed titrating the
iodine liberated in the above reaction by a solution of sodium thio-
sulphate, but if this be done in the open air, the process is useless,
whilst if it be done in a confined space, the measurement of the gas
is at least as accurate as, and more rapid than, the titration process.

Muter (" Analyst," iv. 125) has published a process based on the
oxidation of the ethyl nitrite by means of permanganate of potassium,
but as this process includes other oxidizable bodies, such as aldehyde,
and is more tedious than processes giving more accurate results it
need not be described here.

If the sample contains free nitrous acid, this will, in Eykman's and
similar processes, yield nitric oxide gas. So that if absolutely accur-
ate results are required, the amount of free nitrous acid as indicated
by titration, as described above, must be multiplied by 1-59 and sub-
tracted fro'n the apparent amount of ethyl nitrite found. The Pharma-
copoeia does not differentiate between ethyl nitrite and other nitrous

490 FOOD AND DRUGS.

compounds, but guards against more than traces of nitrous acid by
the test with sodium bicarbonate.

Methyl alcohol compounds should be searched for, in case methyl-
ated spirit should have been used in the manufacture of the sample.
Fifty c.c. should be dehydrated with ignited potassium bisulphate or
carbonate and the dehydrated spirit poured off. Fifteen c.c. or 20 c.c.
are distilled with 10 grms. to 12 grms. of dry calcium chloride, from
a water bath, until practically nothing more comes over. Five c.c.
of water are then added to the flask and another 2 c.c. distilled. This
2 c.c. is then tested for methyl compounds in the following manner : —

Two grms. of potassium bichromate, and 2*5 c.c. of concentrated
H2SO4 are mixed in a small distilling flask with 20 c.c, of water and the
2 c.c. of distillate to be tested. After standing for fifteen minutes the
mixture is distilled, and when 20 c.c. have passed over, the acid dis-
tillate, which contains formic acid if methyl alcohol were present, is
treated with a slight excess of sodium carbonate, evaporated down to 10
c.c, and enough acetic acid added to give the liquid a distinct acid
reaction. O'l grm. of silver nitrate dissolved in 3 c.c. of water is then
added and the whole heated to 80° for a few minutes. In the presence of
methyl alcohol, a precipitate of brown or brownish-black metallic
silver is formed, and a thin film of silver is deposited on the tube. A
slight darkening may be neglected.

Liquor Ethyl Nitritis is also an official preparation of ethyl
nitrite, slightly stronger than the spirit, and much more stable. Its
official characters areas follows: Specific gravity 0-823 to 0*826.
It should not effervesce when shaken with sodium bicarbonate. Ten
c.c. when mixed with 5 c.c. of normal caustic soda solution and 5 c.c.
of water should not turn yellow (absence of aldehyde). It should
yield, when tested in a nitrometer, as in the case of the spirit, at
least 7*6 times its volume of nitric oxide (when freshly prepared) ; or
at least 6-33 times its volume after it has been kept for some time.

SQUILLS.

The dried bulbs of Urginea scilla are official under this name. No
standards are given.

This drug contains several glucosides of which scillitoxin is probably
the most active ; scillipicrin and scillin are also present. All of these
require investigation, but for such details of their chemistry as are
known, reference should be made to a paper by Merck (" Pharm.
Journ." 3, ix. 1038). Scillain is probably a non-toxic glucoside but is
better defined chemically than the others. Its formula is (CgHjo03)x.
On hydrolysis it yields dextrose, butyric acid, isopropyl alcohol, and
bodies not investigated. Squills should yield from 2*5 per cent to 4
per cent of ash on incineration.

Tincture of Squills. — The characters of this preparation are given
in the table on p. 496.

Vinegar of Squalls. — This drug is an extract of squills by acetic
acid containing 4*27 per cent of acetic acid. No standards are given.

A small quantity of acetic acid is lost during the process of making,

SQUILLS. 491

so that the resulting vinegar will not contain quite as much acid as
is used in its preparation. The amount of extractive from the squills
also slightly lessens the percentage of acid. It has been alleged that
this drug loses its acidity somewhat rapidly by keeping, but this is
not the case. The following are figures which cover all properly
prepared samples : —

Acetic acid

Freshly Prepared.

One Month Old.

Twelve Months Old.

Per cent
3-6 to 4-1

Per cent
3-5 to 4

Per cent
3-4 to 3-9

This drug formed the subject of the well-known appeal case of
Hudson V. Bridge, which is dealt with fully in Vol. II.

Oxymel of Squills. — This official drug is a mixture of an acetic
acid extract of the squills, and of clarified honey.

The only standard given is a specific gravity of 1*320.

Lucas (" Pharm. Journ." iv. 17, 778) has made a careful examina-
tion of a number of samples of oxymel of squills and gives the fol-
lowing details. As most commercial honey is laevorotatory, it
follows that oxymel of squills will also be laevorotatory. Even if a
genuine, slightly dextrorotatory honey be used, squills contain a laevo-
rotatory sugar, which will more than neutralize the dextrorotation of
the honey. If a dextrotatory sample be found it is almost certainly
made with an adulterated honey, containing glucose ; or with cane
sugar. The presence of glucose is, apart from the dextrorotation, re-
vealed by the following tests : —

One volume of the sample is mixed with 4 volumes of water and
filtered through animal charcoal, the filtrate being returned until
nearly colourless. It is divided into two portions. To one is added 5
volumes of absolute alcohol. Genuine honey gives only a slight opal-
escence, while an opaque precipitate forms at once if glucose be pre-
sent, in more than small amount. To the second portion of filtrate
is added 1 drop of 10 per cent iodine solution. Pure honey is .unaf-
fected ; if glucose be present the iodine is at once bleached, as the
former rarely contains less than 0*05 per cent, and frequently as
much as 0*1 per cent of sulphurous acid. If more iodine be added,
drop by drop, the slightest excess gives rise to a reddish-brown colour
due to the amylo- and erythro-dextrins present. Honey containing
glucose is strongly dextrorotatory.

Lucas gives the following figures for authentic samples, Nos. 1-8
and No. 10. Nos. 9 and 11 are adulterated with glucose.

The sugar values were determined on a Ventzke scale polarimeter,
using 65-12 grms. of oxymel of squills (i.e. 2*5 times the normal sugar
weight) diluted with water, cleared with lead subacetate, alumina
cream, and sodium sulphate and made up to 250 c.c. and then filtered
through animal charcoal.

492

FOOD AND DEUGS.

Density at
15-6° C.

Gxn. of Real
Acetic Acid
in 100 c.c.

Alcohol Test.

Iodine Test.

Cupric

Reducing

Power.

Direct
Reading.

After
Inversion.

(1) 1-310

1-92

Nil

49-1

-15-5

-20-4

(2) 1-323

1-87

Nil

53-2

-16-3

- 21-6

(3) 1-336

1-08

Nil

56-1

-15-7

-19-1

(4) 1-326

2-46

Nil

53-0

-11-0

-14-8

(5) 1-318

1-06

Nil

49-2

- 15-1

-16-7

(6) 1-322

0-36

Nil

55-4

-14-0

- 15-7

(7) 1-327

0-53

Nil

52-3

-12-1

-16-9

(8) 1-303

1-20

Nil

53-2

-13-9

-15-7

(9) 1-325

1-05

( heavy

I ppt.

Nil

/deep red

1 brown

Nil

45-6

+ 27-4

+ 23-9

(10) 1-350

0-26

57-5

- 12-1

-17-4

(11) 1-321

1-81

/ heavy
I ppt.

f deep red
"( brown

46-2

+ 26-4

+ 23-6

These, calculated to cane sugar, invert sugar and glucose, are
follows : —

Percentage reckoned as

Percentage

Sucrose.

Invert Sugar.

as Glucose.

3-6

48-5

Nil

3-9

53-4

2-5

66-6

2-8

51-8

1-18

48-9

1-25

64-3

3-5

61-5

1-3

52-3

2-57

361

19-1

3-9

56-4

2-06

86-8

18-8

STORAX.

Storax is a balsam obtained from the trunk of Liquidambar orientalis
apparently as a pathological secretion only arising after damage to the
bark or wood. [American storax is a different substance, known as
sweet gum, and is obtained from Liquidamhar stryraciflua. The so-
called Styrax calamitus is probably the powdered bark of one of the
species of Microstemon from which the bulk of the balsam has been
expressed, mixed with more or less sawdust.]

The crude drug is not official in the British Pharmacopoeia, which
only recognizes " prepared storax," which is made by dissolving the
crude drug in alcohol, filtering and evaporating the solvent. The
official requirements for this drug are that it should contain no moisture,
and when boiled with a solution of potassium bichromate and sulp-
huric acid it evolves an odour resembling essential oil of bitter
almonds.

TEREBENE. 493

Storax contains styrol C,;Hr,CH : CH^ ; cinnamic acid, cinnamyl
cinnamate, phenyl-propyl cinnamate, ethyl cinnamate, traces of
vanillin, and lesser known resin alcohols, esters, and hydrocarbons.

The principal adulterants of storax are colophony and fatty oils.
Storax after driving off water is soluble to the extent of 95 per cent,
in 90 -per cent alcohol, the remainder being principally mechanical
impurities. The examination of this drug should commence with the
determination of the amount soluble in 90 per cent alcohol. The-
residue should be examined and if it be of an oily nature, it should be
quantitatively saponified. A high saponification value of this residue
indicates the presence of a fatty oil, and the fatty acids can be
separated from the saponification liquor in the ordinary way.

Fatty oils are indicated by the behaviour of the dried storax with
petroleum ether. This solvent has little effect on pure dried storax,
but in the presence of the fatty matter usually employed as an adul-
terant it becomes quite milky. Another simple method for the de-
tection of fatty adulterants is as follows : Mix 4 grms. of the undried
storax with 9 mgs. of absolute alcohol, make up to 40 mgs. with 90
per cent alcohol (at about 20° C.) and Shake the liquid for a few
minutes. After filterinir, pour into a flask and place in cold water or,
better still, on ice for about six hours. In the presence of much fatty
matter a flocculent precipitate will make its appearance.

Crude storax should not contain more than 30 per cent of water,
and 1 per cent of ash.

The "purified storax " thus prepared by solution in alcohol should
then be examined, and should have the properties of purified storax
of the Pharmacopoeia.

Purified storax should be entirely soluble in 90 per cent alcohol.
It should have a specific gravity between 1*110 and 1-123. The acid
value of the purified balsam should be between 75 and SiO. The ester
value varies from 120 to 140, although these limits are sometimes
exceeded.

Colophony may be detected by extracting the storax with petroleum
ether. A genuine storax will rarely give more than 45 per cent of
extract and this will have an acid value not exceeding about 60. Storax
adulterated with colophony will give a high amount of extract, and
this will have an acid value up to over 100. The ester value of the
extract from pure storax will be over 140. This figure is reduced
by the addition of colophony.

TEREBENE.

Terebene is an official drug which is described as a mixture of
dipentene and other hydrocarbons, obtained by agitating turpentine oil
with sulphuric acid until it becomes optically inactive, and then distil-
ling the oil in a current of steam.

The official tests are as follows : specific gravity 0-862 to 0-866.
It is optically inactive : it distils between 156° and 180° and should
leave only a slight viscid residue. Not more than 15 per cent should
distil below 165°.

494

FOOD AND DKUGS.

Very few commercial samples comply strictly with these require-
ments, and it is probable that no sample can be made to exactly
satisfy the official standards. Very few samples are quite optically
inactive, and the unreasonableness of the requirement in this direction
is shown by the fact that however near to optical inactivity the pro-
duct is, on fractional distillation nearly every fraction will be found to
be optically active. Further, it is impossible to make a sample of
terebene which boils entirely below 180°. All samples will contain up
to 5 per cent or 7 per cent boiling about 180°. Oils made from
French turpentine more nearly meet the requirements of the Pharma-
copoeia.

A terebene may be considered satisfactory if it possesses the
following characters : Specific gravity 0*862 to 0'866 ; optical rotation
not exceeding 1° either way ; not more than 5 per cent boiling below
160° ; not more than 8 per cent boiling above 185° ; not more than 2
per cent of viscid residue (which generally results from the oxidization
during evaporation). A normal sample will give results approximating
to the following, on fractionation (Tyrer and Wertheimer, " Pharm.
Journ." 1900, ii. 101).

Fraction.

Sp. Gr.

Rotation (186 mm.).

Refractive Index.

150 .

0-8753'

+ 2-8

1-46207

150-164 .

0-8766

+ 1-4

623

164-169 .

0-8754

+ M

584

169 .

0-8733

-fO-8

603

169-169-5 .

0-8832

-fO-5

603

169-5-169-7

0-8807

-fO-4

652

169-7-169-9

0-8835

4-0-4

702

169-170 .

0-8968

771

170-170-5 .

0-8942

672

170-5-171 .

0-8892

-0-1

683

171 .

0-8855

-0-4

702

The following tables represent the principal characters of a
number of important galenical preparations of the British Pharma-
copoeia : —

TINCTURES. 495

Standards for Tinctures of the British Pharmacopceia.
In most Cases not Official.

Name of Tiiicture

Specific Gravity

Solid Residue

Alcohol by

Gr. per 100 c.c.

j.^aixu.c^ wi X J.ui.' t'Ui c*

at 15= '

Gr. per 100 c.c.

Volume.

Active Ingredient.

Per cent

Tinot. Aconiti

•890 to

•895

1-3 to 1^6

to 68

©•025 to 0-065

„ Aloes .

•970 „

•980

6^5 „ 7^6

« 42

„ Arnicse .

•893 „

•899

©•65

„ 0-8

„ 69

„ Asafoetidae .

•910 „

•918

, 10

» 63

„ Aurantii recentis .

•875 „

•885

1^6

, 1-9

M^re

„ Belladonnee .

•910 „

•915

, 0^65

M 58

0-048 to 0-052

„ Benzoin Co. .

•890 „

•904

„ 20

„ Buchu .

•925 „

•935

2^9

, 4^0

to 58

„ Calumbee

•915 „

•925

0-8

, 1-2

„ 58

„ Camphoree Co. , .

•913 „

•923

0^3

„ 0-37

» 59

0-43 to 0-49

„ Cannabis Indicae .

•845 „

•850

3-5

M 4-2

» 87

„ Cantharidis .

•835 „

•840

015

, 0^17

„ 90

„ Capsici

•890 „

•898

, 1-2

„ 69

„ Cardamomi Co. .

•945 „

•955

, 7-2

„ 54

„ Cascarillae

•895 „

•902

, 25

» 67

„ Catechu

•978 „

•984

, 16

„ 53

„ Chiratee

•920 „

•925

, 1^2

„ 58

„ Chloroform et
Morphinae Co.

I^OIO „

1^015

, 30

„ 52

—

„ Ciraieifugee .

•922 „

•928

1-2

, 15

„ 58

„ Cinchonae

•914 „

•924

6^2

, 6^9

0-95 to 1-05

•914 „

•924

4^6

. 5-2

0-45 „ 0-55

„ Cinnamomi

•900 „

•905

1^5

, 22

to 67

„ Cocci .

•950 „

•960

2 ,

, 2-5

» 44

•

„ Colchici Sem

•950 „

•960

1^9

, 2^4

„ 43

0-05 to 0-09

„ Conii .

•895 „

•902

1^3

, 1^45

„ 68

0-05 „ 0-1

„ Croci .

•925 „

•930

, 2^9

„ 58

„ Cubebae

•840 „

•845

, 1-5

„ 86

„ Digitalis

•930 „

•935

2-9

, 3^7

„ 56

0-4 to 0-75

„ Ergotae Ammon. .

•935 „

•942

2^8

, 4^0

» 52

„ Ferri Perchloridi .

1-085 „

1^088

„ Gelsemii

•920 „

•928

1^2 to 1^3

to 57-5

0-02 to 0^03

„ Gentianae Co.

•961 „

•970

4-0 „ 5^5

„ 43

„ Guaiaci Ammon. .

•898 „

•907

14 „ 17-5

M 71

,, Hamamelidis

•947 „

•954

1-4 „ 2^0

„ 45

„ Hydrastis

•923 „

•929

20 „ 2-5

„ 58

0-4 to 0-6

„ Hyoscyami .

•950 „

•960

23 „ 3^6

» 44

0-008 „ 0^015

„ lodi .

•875 „

•880

—

„ 86

„ Jaborandi

•956 „

•959

2^6 to 4^3

„ 43

0-08 to 0-15

„ Jalapae .

•910 „

•915

3^5 „ 4^7

„ 66

1-45 „ 1-55

„ Kino .

•995 „

1^000

20 „ 24

„ 53

„ Krameriae

•935 „

•940

4^5 „ 5^0

„ 56

„ Lavandulae Co.

•835 „

•840

045 „ 0-52

„ 89

„ Limonis

•875 „

•885

1^4 „ 1-5

„ 77

„ Lobelias ^Etherea .

•812 „

•817

0-9 „ 1^5

-02 to -04

„ Lupuli .

•935 „

•94a

3^5 „ 4-0

„ 54

„ Myrrhae

•848 „

•858

4 „ 6

M 86

„ Nueis VomicsB

•910 „

•915

!•? „ 1^8

„ 61

0^24 to 0^26

„ Opii .

•955 „

•962

3-4 „ 3^7

M 44

0^7 „ 0^8

„ ,, Ammon.

•894 „

•901

2^7 „ 2^9

» 64

0-1 „ 0-12

„ Podophylli .

•844 „

•848

3-4

3-3 „ 3-5

„ PruniVirg. .

•931 „

•938

2^3 to 2^8

to 55

„ Pyrethri

•900 „

•905

1^5 „ 1^9

» 69

„ Quassias

•945 „

•950

0^2 „ 0-5

„ 45

„ Quillaiae

•920 „

•927

1 „ 1^4

„ 58

„ Quininae

•885 „

•893

3-5 „ 3^9

1-634

496 FOOD AND DRUGS.

Standaeds for Tinctures qf the British Pharmacopceia.
In most Cases not Official.

Name of Tinctixre.

Specific Gravity

Solid Residue

Alcohol by

Gr. per 100 c.c.

at 15°

Gr. per 100 c.c.

Volume.

Active Ingredient.

Percent

Tinct. Quininae Ammon.

•925 to

•930

1^8

1-471

„ EheiCo.

•970 „

•975

12 to 13

49 to 50

„ Scillffi .

•962 „

•970

105 „ 12^5

51 „ 52

„ Senegee .

•935 „

•944

42 „ 5^0

54 „ 55

—

„ Sennse Co.

•985 „

•995

8 „ 9

38 „ 40

—

., Serpentariee

•895 .,

•900

1^4 „ 2-0

66 „ 68

—

„ Stramonii

•953 „

•962

32 „ 40

42 „ 43

0-02 to 0-03

„ Strophanthi

•894 „

•897

0^4 „ 0-7

68^5 „ 69

0-05 „ 0^08

., Sumbui

•900 „

•905

2^2 „ 2^8

65 „ 67

—

„ Tolutanae

•862 „

•870

8^5 „ 9-0

81 „ 84

—

„ ValerianBB Ammon.

•940 „

•945

2^7 „ 3^7

49 „ 51

„ Zingiberis .

•840 „

•845

OS „ 0-4

87 „ 89

—

Standards for Fluid Extracts of the British Pharmacopceia.
In most Cases not Official.

• Liquid Extract.

Specific
Gravity.

Extractive
(Gm. per
100 cc).

Average
Alcoholic
Strength
per cent
by Vol.)

Gr. of Ac-
tive Ingredi-
ent per 100
c.c.

Extract. Belladonnae liq. .

0-890 to 0-920

11 to 14

66 to 69

„ Cascarae liq.

1^054 „ 1^066

0-75

„ Cimicifugse liq. .

0-875 „ 0^890

—

,, Cinchonee liq.

1^115 „ 1-130

38 to 43

11 to 13

5-0

„ Cocse liq. .

0-995 „ 1-031

18 „ 20

49 „ 52

0-2 to 0-6

„ Ergotse liq.

1-005 „ 1-025

12 „ 15

30 „ 32

—

„ Glycyrrh. liq.

1-130 „ 1-150

17-5

—

„ Hamamelidis liq.

1-025 „ 1-040

—

„ Hydrastis liq.

1-025 „ 1-040

20 to 24

36 to 40

4 to 6 total
alkaloids)
2 to 2-25

„ Ipecac, liq.

0-885 „ 0-915

9 „ 12

78 „ 79

„ Jaborandi liq.

1-020 „ 1-050

21 „ 22

33 „ 35

0-2 „ 0-75

„ Nucis vomicae liq.

0-945 „ 0-965

11 „ 12-5

61 „ 63

1-5

„ Opii liq.

0-985 „ 0-995

3 „ 31

0-7 „ 0-8

„ Pareirae liq.

1-025 „ 1-050

—

„ Sarsae liq. .

1-055 „ 1-085

—

„ Taraxaci liq.

1-045 „ 1-060

—

The Concentrated Liquors of the British Pharmacopoeia.

A number of concentrated liquors are official in the Pharmacopoeia,
having been introduced as official representatives of a class of concen-
trated preparations known as concentrated infusions which although
commanding extensive employment, are not official. Thes-e concen-

CONCENTKATED LIQUOKS.

497

trated preparations do not, on dilution, exactly represent the freshly
prepared infusions, hence the choice of the name "concentrated
liquors" by the Pharmacopoeia authorities.

The following are the average characters of these galenicals : —

Liquores.

Specific Gravity.

Extractive (Gm.
per 100 c.c.)

Alcoholic Strength
(by Vol.)

Liquor. CalumbsB cone. .

0-987 to 0-997

3-5 to 6

18 to 20

,, Chiratae cone.

0-978 „ 1-000

3-8 „ -S-S

18 „ 19

„ Cuspariae cone. .

■ 1-008 „ 1-020

8 „ 10

18 „ 19

„ Kramerise cone. .

1-007 „ 1-015

8 „ 9

18 „ 19

,, QuassisB

0-976 „ 0-990

0-25 „ 0-5

18 „ 19

„ Khei .

0-998 „ 1-035

10 „ 13

17 „ 18

„ Sarsae Co. cone. .

1-020 „ 1-045

9 „ 10

18 „ 19

„ Senegas cone.

1-010 „ 1-032

10 „ 14

21 „ 22

„ Sennas cone.

1-000 „ 1-080

12 „ 16

17 „ 18

,, Serpentariae cone.

0-990 „ 1-005

5 „ 5-5

18 „ 19

It must be borne in mind, however, that in the case of liquor
calumbae concentratus, the official formula is so unsatisfactory that
the finished product may, unless prepared on a small scale, have char-
acters outside the above limits.

It has been stated that the characters of the finished product de-
pend on the power of the press used for expressing the liquid from
the macerated drug, and as pointed out by F. C. J. Bird, this state-
ment has been practically investigated by A. C. Abraham, who showed
that when using a hand-press the finished liquor had a specific gravity
of 1"029 and extractive 2*16 per cent, whilst with a powerful hydraulic
press these figures became respectively 1-032 and 3-66. An inspec-
tion of the formula even suggests this result, for in the first instance
(hand-press) the volume of the expressed liquid being smaller, and the
quantity of spirit added remaining the same, the proportion of alcohol
in the mixture is greater, and precipitation consequently is more
copious. Loss of extractive therefore follows both from the diminished
amount in the pressings and the larger quantity precipitated by the
spirit ; there is also much loss of alcohol owing to the greater alco-
holic strength of the liquid absorbed by the filter paper and the more
voluminous precipitate. There is no compensation for this in the
official process, as the deficiency of the volume is directed to be made
up by addition of water to the filtrate.

The formula for liq. calumbae cone, is one which furnishes a pre-
paration, varying greatly according to the conditions of manufacture,
both in extractive and percentage of alcohol, a low proportion of the
latter bringing in its train a continuous deposition of sediment from
the development of acidity in the liquid.

As the non-official " concentrated infusions " are still used to a
very large extent, being diluted with seven times their volume of
water when the fresh infusion is prescribed, it will be as well to give
the following figures which represent a large number of samples of
commercial products. These figures are those obtained in the author's
VOL. I. 32

498

FOOD AND DEUGS.

laboratory. It may be pointed out that even if the amount of extractive
matter be 8 times that of the official fresh infusion, their employment
is, strictly speaking, not absolutely justifiable when the fresh infusion
is ordered, since alcohol is introduced into the medicine. This, how-
ever, is almost an" academic point, and the actual quantity would be
very small.

All concentrated infusions should be tested for salicylic acid, a
preservative sometimes found, which is used to save the cost of
alcohol. The liquid should be freed from alcohol by evaporation,
acidified with H^SO^, shaken with a mixture of equal volumes of
ether and petroleum ether, the solvent separated and extracted with
dilute aqueous potash solution, and this solution, containing the
salicylic acid, neutralized and treated with a few drops of iron alum
solution, when a purple colour results if salicylic acid be present.

Infusion of

Specific Gravity.

. Extractive
Gr. per 100 c.c.

Alcohol
by vol.

Extraction of fresh
Infusion x 8.

Crflumba

Cascarilla

Gentian (compound)

Quassia .

Bhubarb

Senega .

0-985 to 0-995
0-980 „ 0-996
0-990 „ 1-010
0-975 „ 0-985
0-996 „ 1-010
1-000 „ 1-035

2-5 to 3-5
2-5 „ 4-6
6 „ 10
0-1 „ 0-4
9 „ 12-8
9 „ 12-5

15 to 20
15 „ 18-5
14 „ 19

14 „ 20
13 „ 18

15 „ 18

3-2 to 3-5
5-0 „ 6-0
8-5 „ 10
0-2 „ 0-3
12 „ 15
10 „ 14

The Spirits op the Pharmacopoeia.

On opposite page are the characters which the various spirits of
the Pharmacopoeia should have.

The following method is recommended by Thorpe and Holmes
(Proc. Chem. Soc. 19, 13) for the determination of ordinary alcohol
in essences and medicinal preparations containing essential oils and
volatile substances, such as ether, chloroform, benzaldehyde, camphor,
and compound ethers, in preparations for which "drawback" is
claimed from the Inland Ee venue, on exportation. It has been used
for some time past in the Government Laboratory, and has been found
to be both accurate and of very general applicability. Twenty-five c.c.
of the sample, measured at 15-5° C. are mixed with water in a separ-
ator to a bulk of from 100 c.c. to 150 c.c. and common salt is added in
sufficient quantity to saturate the liquid. The mixture is now shaken
vigorously for five minutes with from 50 c.c. to 80 c.c. of light petroleum
boiUng below 60° C. and after standing for about half an hour the
lower layer is drawn off into another separator, extracted if necessary
a; second time with petroleum, and then introduced into a distillation
flask. Meanwhile, the petroleum layers are washed successively with
25 c.c. of saturated brine, the washings added to the main bulk, which
is neutralized, if necessary, and then distilled, and the distillate made
up to 100 c.c, and its relative density determined at the standard
temperature in the usual manner. The results thus obtained require

SPIKITS.

499

a small correction from the circumstance that, as the alcohol present
is distilled into four times its initial volume, the errors of the spirit
tables are necessarily quadrupled. The mean error of the tables at

Specific Gravity.

Optical Rotation
in 100 mm. tube.

Other Characters.

Spirit of ether .

0-801

nil

About 33 should
distil below
45° C.

Compound spirit of ether .

0-808 to 0-812

nil

Spirit of nitrous ether

(see p. 485)

Aromatic spirit of ammonia

(see p. 484)

Fetid spirit of ammonia

0-838 „ 0-845

nil

To contain 2-88
grms. NH3 in
100 c.c.

Spirit of aniseed

0-848 „ 0-850

too small to observe

Alcohol 81 by
vol. Aniseed
oil 10 by vol.

Compound spirit of horse-

radish ....

0-895 „ 0-900

M It

Spirit of cajaput

0-842 „ 0-844

Alcohol 81 by
vol. Cajaput
oil 10 by vol.

Spirit of camphor

0-848 „ 0-851

+ 3-4° to +3-5°

Alcohol 81 by
vol.

Spirit of chloroform .

0-867 „ 0-868

Alcohol 85-5 by
vol.

Spirit of cinnamon

0-853 „ 0-855

too small to observe

Alcohol 81 by
vol. Oil 10
by vol.

Spirit of juniper

0-836 „ 0-887

0° to - 0° 50'

Alcohol 81 by
vol. Oil 10
by vol.

Spirit of lavender

0-839 „ 0-841

- 0° 20' to - 0° 55'

Alcohol 81 by
vol. Oil 10
by vol.

Spirit of peppermint .

0-840 „ 0-843

about - 2° 30'

Alcohol 81 by
vol. Oil 10
by vol.

Spirit of nutmeg

0-838 „ 0-842

+ 1° 20' to + 3°

Alcohol 81 by
vol. Oil 10
by vol.

Spiritus rectificatus .

(see p. 273)

Spirit of rosemary

0-8405 „ 0-8425

O°to + 1°

Alcohol 81 by
vol. Oil 10
by vol.

Spiritus vini gallici

(see brandy,
p. 286)

below 40 per cent proof (for example, 0-972 sp. gr.) may be set down
as +0-2 per cent of proof spirit, and hence the observed determina-
tions, expressed as percentage, of proof spirit, require a subtractive
correction of O'S per cent.

The essential oil, which should amount to 10 per cent by volume
in the case of all the spirits of essential oils in the above table, may

500 FOOD AND DRUGS.

be determined by evaporating the petroleum ether in a current of
warm air and weighing the oil ; or approximately by introducing 10
c.c. into a 200 c.c. flask with a neck graduated in ^jj c.c. and adding
brine, shaking gently, and adding sufficient brine to drive the oil into
the neck, where, after standing for twenty-four hours, it is measured.

>■ I

CHAPTER IX.

DEUGS CONTAINING ALKALOIDS, ETC., CAPABLE OF
APPEOXIMATE DETEEMINATION.

The present chapter is devoted to a number of drugs containing
physiologically active substances which are capable of determination
with more or less accuracy, and are frequently of an alkaloidal nature.
The principal active substances of this type often exist in very small
amount in their respective drugs — sometimes to the extent of less
than 1 per cent, but occasionally, as in opium, to a comparatively
large extent. The difficulty of exact determination is considerably in-
creased when more than one alkaloid exist together — and in some
cases a separation is practically impossible.

The Nature of the Alkaloids. — The greater number of alkaloids are
derivatives of cycloid bases such as pyridine and quinoline or of com-
plex phenanthrene compounds. The majority of the alkaloids contain
carbon, hydrogen, nitrogen, and oxygen, and are then generally crystal-
line solids. Those which are free from oxygen are usually liquid —
such as Conine. A small number are closely related to uric acid
such as caffeine and theobromine but there is a tendency to restrict
the name " alkaloid " to the derivatives of nitrogenous cyclic com-
pounds. Generally speaking the alkaloids are bases analogous to am-
monia, combining with free acids without the elimination of water.

Many of them are powerfully alkaline, neutralizing acids perfectly
and forming well-defined crystalline salts. In some cases, however,
their basic properties are very weak and even their salts with the
mineral acids are decomposed by solution in water. The majority of
the alkaloids are very sparingly soluble in water. Hence they are
usually precipitated from solutions of their salts by alkalies. Nearly
all the alkaloids are easily soluble in alcohol. Their salts are usually
soluble in water and fairly so in alcohol. Numerous double salts
exist which are practically insoluble in water, a fact which enables
most of them to be precipitated in a highly insoluble condition.

The solvents which dissolve most of the alkaloids are amyl alco-
hol and chloroform, but many of them are freely soluble in other
organic liquids. In most cases the salts are not soluble in these
solvents. These facts are taken advantage of in the separation of the
alkaloids from the crude drugs.

Numerous reagents have from time to time been proposed as pre-
cipitants of the alkaloids, of which the following are the most im-
portant.

(501)

502 FOOD AND DKUGS.

Sonnenschein s Reagent. — This reagent, which is exceedingly use-
ful for separating most of the alkaloids from'foreign matters, consists
of a solution of phosphomolybdic acid. It is best prepared from
ammonium molybdate, by dissolving sodium phosphate in hot water,
rendering acid with nitric acid, and adding an excess of a saturated
solution of ammonium molybdate. The yellow precipitate formed is
filtered off, washed, rendered acid with nitric acid and dissolved in
warm solution of Na2C03. The solution is evaporated to dryness
and ignited to drive off ammonium salts, and the cold residue again
moistened with nitric acid and ignited. The product is phospho-
molybdate of sodium. It is dissolved in ten times its weight of
water containing 10 per cent of nitric acid and is then ready
for use.

This reagent gives a yellow amorphous precipitate with nearly
every alkaloid, but other substances than alkaloids are sometimes pre-
cipitatated so that a precipitate does not always indicate the presence
of an alkaloid ; whilst a negative reaction is usually proof of the
absence of alkaloids. To recover the alkaloid from the precipitate
for further examination, the moist precipitate is treated w^ith a
solution of ammonia and the aqueous liquid, in which the alkaloid
will often be floating as a white precipitate, is extracted by chloroform,
amyl alcohol, etc. If a blue or green colour results it indicates re-
duction of the molybdic acid attended with decomposition of the
alkaloid. If this is the case, the moist precipitate should be made
into a paste with sodium carbonate and extracted with absolute
alcohol.

Mayer's Reagent is a solution of potassio-mercuric iodide. It is a
valuable precipitant of most alkaloids, and was originally recom-
mended as a solution for the volumetric determination of alkaloids.
The composition of the precipitate varies, however, with the slightest
change in the conditions of the experiment, so that it has fallen into
disrepute — so far as quantitative work is concerned. It is made by
dissolving 6-775 grms of HgClg in water, and 25 grms. of KI in
an equal volume of water. The two solutions are mixed and the
mixture made up to 1 litre. As a test for alkaloids it should be
applied to solutions rendered faintly, but distinctly, acid with HCl or
H2SO4, and as free from alcohol as possible. Alkaloidal precipitates
with Mayer's reagent are generally only faintly yellowish in colour and
are amorphous and flocculent. They are soluble to some extent in
alcohol, acetic acid and in much excess of the reagent. Other sub-
stances are precipitated besides alkaloids, so that an examination of
the precipitate is necessary, especially when the solution is known to
contain other organic matters. The quantitative determination of
alkaloids by Mayer's solution is dependent on the fact that the solu-
tion above given is approximately one-twentieth normal: but it is
necessary for anything like accuracy to check the value of the solution
against a known quantity of the alkaloid being determined, since the
conditions will vary with different alkaloids, and the composition of
the precipitates is uncertain and variable. A. B. Lyons has examined
the question very fully, and for further criticisms on it, the reader is

DKUGS CONTAINING ALKALOIDS, ETC. 503

referred to his " Manual of Pharmaceutical Assaying ". The following
is the method of carrying out the titration. The liquid containing
the alkaloid should be acidulated with hydrochloric acid, and
adjusted as nearly as possible to contain about 0*5 per cent of
alkaloid. Mayer's solution is run in carefully up to the point of
no further precipitate being produced. There is no possible indicator
for the end reaction, so that a few drops must be filtered through a
very small filter and dropped on to black glass or ordinary glass
backed with black paper. A drop of the solution from the burette
is added, and the slightest turbidity can then be noted. All the trial
portions must be returned to the titration flask before the final reading
is decided upon.

Phosphotungstic Acid (Schleibler's reagent) is prepared by dis-
solving 100 parts of sodium tungstate and 60 parts of sodium
phosphate in 500 parts of water, and adding sufficient nitric acid to
produce an acid reaction. The general characters of this reagent are
exactly the same as those of phosphomolybdic acid.

Wagner's Reagent is a 2 per cent solution of iodine in a 5 per cent
potassium iodide solution. It should be added to solutions of the
alkaloids rendered slightly acid with sulphuric acid, and only a small
quantity is to be used — not sufficient to colour the liquid yellow.
Under these circumstances a red or red-brown precipitate is produced
in very dilute alkaloidal solutions. A negative reaction is practically
proof of the absence of all common alkaloids, but a positive reaction
requires confirmation. Much alcohol should be avoided, since the
precipitation is often very slow in the presence of alcohol. The pre-
cipitates consist of complex iodides of the alkaloids, from which the
free bases may be recovered by treatment with sulphurous acid or
sodium thiosulphate, and then adding alkali and extracting with
chloroform.

Dragendorff's Reagent is a reagent of great delicacy. It is a
solution of potassio-iodide of bismuth, and is easily prepared by mix-
ing 15 c.c. of liquor bismuthi of the Pharmacopoeia, with 1 c.c. of
HCl and then adding 1'2 grms. of potassium iodide. With this
solution, solutions of the alkaloids strongly acidulated with HgSO^
give orange red precipitates which are quite insoluble in water.

Gold and Platinum Chlorides are useful precipitants of the
majority of the alkaloids. They are best added as aqueous solutions
to solutions, of the alkaloid containing pure HCl. The melting-points
of these double salts are generally characteristic and assist in the
identification of the base.

Colour Reactions. — The minute quantities of the alkaloids with
which one often has to deal, together with the fact that these bodies are
so similar in their behaviour to most reagents, have caused very
numerous colour reactions to be published. Colour reactions are,
of course, the most unsatisfactory reactions with which the analyst
has to deal, and it may be definitely stated that the great majority
are totally useless, whilst many are only sufficiently reliable to give
general indications. It is only a very few colour reactions that can
be regarded as characteristic, and giving definite information.

■504 FOOD AND DEUGS.

Absolute purity of the reagent used is necessary. For example,
tlie merest trace of nitric acid, a not uncommon impurity, in sulphuric
acid will act as an oxidizing agent and may materially alter the colour
of the reaction.

A careful examination of the mass of published work on colour
reactions of the alkaloids compels one to reject the bulk of them as
useless and misleading. Non-alkaloidal bodies will often give colours
identical with well-known alkaloids, and several alkaloids will often
give indistinguishable colours.

For instance, nitric acid is usually described as giving the follow-
ing colours when a drop is placed on a fragment of the alkaloids : —

Codeine

= orange-yellow

Papaverine

= orange

Sabadilline

= yellow

Morphine

= yellow to red

Berberine

= red to red-brown

Brucine

= blood-red

Pseudomorphine = orange-red

Again, Frohde's reagent (5 mgs. of sulphomolybdic acid in 1 c.c. of
H2SO4) is generally stated to give the following colours : —

Codeine = deep blue (gradual)

Morphine = violet-blue to dirty green and then to deep blue

Narcine = yellowish-brown to red, to blue

Berberine = brown -green

Quinine = pale green

Apomorphine = green to violet

And the same holds good for other reagents. Such few colour
reactions (such as that for strychnine with chromic acid and H2SO4 ;
or the thalleoquin reaction) as yield useful confirmatory information
are mentioned under the alkaloids treated in the present chapter.

The Estimation of Alkaloids. — The alkaloids are rarely found in a
free state in a plant, possibly never. They are combined with either
an organic acid such as malic or tannic acid, or with an inorganic
acid. The alkaloidal salt is usually, almost invariably indeed, soluble
in alcohol, but not so in ether.

It is necessary in the first place to extract the alkaloid from the
drug as free from foreign matter as possible, and then either purify
it thoroughly and weigh it, or obtain it in as nearly a pure condition
as possible and titrate it.

Allen ("Commercial Organic Analysis") gives the table on op-
posite page as showing the solubility of the leading plant constituents
in water, alcohol and ether, thus indicating the bodies likely to be
extracted from given plants.

As a rule the alkaloids can be best extracted from drugs by
means of alcohol (90 per cent), (sometimes after the addition of an
alkali) and then extracted from the alcoholic solution by the use of
ingi^miscible solvents. In some cases, however, it is better to render
the drug alkaline so as to liberate the alkaloid, and extract with
ether or chloroform at once, and then further purify the extracted
alkaloid. Speaking generally, salts of alkaloids are insoluble in

DRUGS CONTAINING ALKALOIDS, ETC.

505

solvents which are immiscible with water, hence an aqueous solution
of an alkaloid rendered slightly acid and shaken with chloroform,
ether, benzol, etc., does not lose its alkaloid. Glucosides, on the other
hand, are dissolved out by the immiscible solvent. A certain
number of alkaloids which possess very weak basic properties are,
however, extracted from acid solution by immiscible solvents.

In shaking out an alkaloid by means of an immiscible solvent, the
aqueous solution should be rendered alkaline, with ammonia or fixed

Alkaloidal salts .

Other salts of inorganic acids

Other salts of organic acids

Free organic acids

Tannin and colouring matter

Sugars

Gums and pectins
Albumenoids

Starch

Cellulose ....

Eesins

Fixed oils ....

Essential oils

Chlorophyll ....

Water.

Alcohol.

Ether.

Soluble
Mostly soluble
Soluble

„ (in hot water)
Insoluble

Soluble

Mostly insoluble

Soluble

Insoluble

Soluble

Sparingly soluble
Soluble
>>

Insoluble

Mostly insoluble

Variable
Insoluble

>»
»>

Variable
Soluble

alkalis or alkaline carbonates. It is usually necessary to shake the
aqueous liquid at least three times with fresh portions of the immis-
cible solvent to ensure complete extraction, especially when ether is
used, as ether is very soluble in water and after separation of the two
liquids, a considerable amount of ether, with alkaloid dissolved in it,
remains dissolved in the water after the first extraction. The methods
and solvents applicable to the various drugs described will be found
in detail under each drug.

After the alkaloid is obtained in a more or less pure state — which
may have required further extraction from the immiscible solvent by
acidulated water and then a final extraction from this water after
being rendered alkaline, with the immiscible solvent again — it may be
weighed after evaporation of the solvent, and if it is of such a nature
that it has been obtained in a state of purity, the weight may be taken
as being that of the alkaloid, or, if it be impure, it may be determined
by dissolving the residue, on evaporation of the solvent, in an excess
of standard acid, usually one-twentieth normal, and titrating the ex-
cess of acid by standard solution of an alkali, baryta being preferable,
although soda is generally used. Of indicators methyl-orange is the
most generally useful, although rosolic acid, iodeosine, cochineal, and
litmus frequently give good results.

When a complex or more or less unknown organic mixture is to
be examined for alkaloids, it is best to precipitate any alkaloids that
may be present from a solution containing but little alcohol, and a
slight excess of hydrochloric acid, by one of the alkaloidal reagents

506 FOOD AND DKUGS.

described above (p. 502). It is necessary to remove as much as pos-
sible of the inert organic matter present, which is best effected by a
solution of lead acetate, which should be added to the neutral solution
as long as any precipitate is formed. Much excess should be avoided.
After the lead precipitate containing much organic matter has been
filtered oti and the precipitate washed with water, a little solution of
basic acetate of lead is added which produces a further precipitate.
The liquid is then rendered faintly alkaline with ammonia, again
filtered, evaporated till the ammonia is driven off, and then the excess
of lead removed by a current of sulphuretted hydrogen. After filtering
and driving off excess of HoS, the alkaloids are precipitated by one of
the general precipitants such as phosphomolybdic acid, after the solu-
tion is rendered faintly acid with hydrochloric acid. The solution
should stand for twelve hours and then be filtered. The precipitate
is rendered alkaline with potassium carbonate and the free alkaloid is
extracted therefrom with strong alcohol.

ACONITE.

The root of Aconitum napeilus is ofl&cial in the British Pharma-
copoeia, but no standards are given.

The potency of this drug is due to the presence of the alkaloid
aconitine (see below), but there are also present small quantities of
aconine C25H4JNO9 and picraconitine. Good roots contain from 0*45
per cent to 1 per cent of alkaloids.

Aconite root contains from 3 per cent to 6 per cent of mineral
matter, any excess over this amount indicating the presence of earthy
matter.

The determination of alkaloids is carried out by exhausting
the roots in fine powder with 75 per cent alcohol rendered slightly
alkaline with ammonia. Twenty grms. of root are so treated and the
resulting alcoholic liquid is evaporated to a syrupy consistence. This
is then rendered alkaline with ammonia and extracted with three
successive portions of 20 c.c. of chloroform. The chloroform solu-
tions are extracted twice with dilute hydrochloric acid which is again
rendered alkaline and the purified alkaloids finally extracted with
chloroform three times, and the solvent evaporated. The weight is
recorded, and the residue is then titrated by adding excess of one-
twentieth normal acid and titrating back with one-twentieth normal
alkali, using methyl-orange as indicator. One c.c. of one-twentieth
normal acid is equivalent to 0-03225 grm. of aconitine (this deter-
mination is not quite exact, as aconine and picraconitine, with different-
molecular weights, are present).

Tincture of Aconite. — This official preparation is an extract of
1 ounce of the drug with sufficient 70 per cent alcohol to make 1
pint. No official standards are given, but a genuine tincture should
have the following characters : —

Specific gravity 0-890 to 0-895

Solid residue 1-3 ,, 1-6 gr. per 100 c.c.

Alcohol by volume .... 66 ,, 68 per cent

Alkaloids 0-025,, 0-065 per cent

ALOES. 507

The alkaloids are determined by evaporating 100 c.c. to 200 c.c. of
the tincture to a syrupy consistence, and then proceeding as above,
weighing the alkaloids (by titrating them rather lower results will be
obtained).

Aconitine is official in the Pharmacopoeia, where its formula is
wrongly given as CggH^jNOja. It is there described as melting at
189^ to 190", and when heated slightly above this temperature it yields
acetic acid. It is soluble in 90 per cent alcohol, and chloroform, but
less readily so in ether. It is nearly insoluble in water and petroleum
spirit. It is IsBvorotatory. A drop of a solution of less than one-tenth
per cent strength produces a persistent tingling sensation on the
tongue. The salts are crystalline, the hydrochloride melts at 149°
and the hydrobromide at 164°. A dilute solution — 1 in 4000 in water
— faintly acidulated with acetic acid yields a red crystalline precipitate
with a few drops of a dilute solution of permanganate of potassium.

The true formula for aconitine is Cg^H^^NO^j. It melts at 195° or
if heated slowly at 1^2", with decomposition. If 10 mgs. be evapor-
ated with 3 c.c. of fuming HNO3 and the residue on cooling be treated
with KOH solution no violet colour is produced — (diffentiation from

N
atropine and pseudoaconitine) ; 0*1 grm. should require 3'3 c.c. of 7^7.

hydrochloric acid for neutralization, as determined by dissolving it in

N
excess- of the acid and titrating back with — caustic soda solution

using methyl orange as indicator.

The hydrochloride has the formula Cg^H^-NOipHCl . 3H2O;
hydrobromide '2{C>^^^^^O^.U^y]6U^O ; and nitrate Cg^H^^NOuHNOg.

ALOES.

The drug known by this name is the inspissated juice of the leaves
of various species of aloe. There are numerous varieties of the drug,
of which those known as Barbadoes aloes, Socotrine aloes, Cape aloes,
and Natal aloes are the principal. The aloes official in the British
Pharmacopoeia are two in number. Aloe Barhadensis is described
as the product of Aloe vera, Aloe chinensis and probably other species.
It occui-s in yellowish, red-brown or almost black masses, with either
a dull waxy, or smooth glassy fracture. The following are the official
tests for this variety of aloes, which are known as either Barbadoes
aloes or Cura9ao aloes. The powdered drug imparts a crimson
colour to nitric acid, and when treated with sulphuric acid and the
vapour of nitric acid, should only yield a slight bluish-green colour, but
not a bright blue colour (absence of Natal aloes). It is almost en-
tirely soluble in a mixture of 2 volumes of 90 per cent alcohol with 1
volume of water. At least 70 per cent should dissolve in cold water.

Socotrine aloes is described as the product of Aloe Perryi and
probably other species. It has always a dull waxy fracture, and is
known in commerce as Socotrine or Zanzibar aloes. These aloes
impart a reddish or yellowish-brown colour to nitric acid. If the

508 FOOD AND DRUGS.

vapour of nitric acid be blown over the powder moistened with
sulphuric acid, no blue coloration is produced. The Pharmacopoeia
also requires about 50 per cent to be soluble in cold water, and states
that the drug is almost entirely soluble in a mixture of 2 volumes of
90 per cent alcohol and 1 volume of water.

Natal and Cape aloes are not official in the British Pharmacopceia.
The principal constituent of aloes is known as aloin. There are
several varieties of " aloin " of which the following are the principal : —

Barbaloin C^gH^gO^, SH^O (the Pharmacopoeial formula is Ci^fligO-),
is a yellow crystalline powder sparingly soluble in cold water, insol-
uble in ether and fairly soluble in 90 per cent alcohol. It occurs in
Barbados aloes, Cura9ao aloes, Cape aloes, etc., in fact in every variety
of aloes known, except Natal aloes, so that it might well be described
as aloin without any prefix. It melts at 147°. It is usually found
mixed with iso-barbaloin, although the best Barbadoes aloes contains
about 20 per cent of barbaloin and practically no iso-barbaloin.
Curasao aloes contains about equal quantities of the two varieties.
Leger considers barbaloin to have the formula CgjH^oOc,, but it is prob-
able that its formula is C^qH^qO^. Nataloin (CysHggOjo Leger) and
homonataloin (C22H24O10, L6ger) are found in Natal aloes.

Capaloin, another variety of aloin, is present in Cape aloes.

The neutral resinous matter of Barbadoes aloes is principally com-
posed of the cinnamic ester of an alcohol, aloeresinotannol C22H^gOg,
whilst that of Cape aloes is the paracumic acid ester of the same
alcohol.

All varieties of aloes which contain barbaloin also contain a small
amount (about 0*2 per cent) of emodin, a trioxy-methyl anthraquinone.
It is probable that this is really the cathartic principle of aloes, since
barbaloin is rapidly oxidized to emodin by the action of the air, if it
be dissolved in alkalis : —

C,gH,gO, + 03 = C,,H,oO, + CO2 + 3H2O
barbaloin emodin

The Examination of Aloes. — The ash of genuine aloes varies from
1 per cent to 4 per cent, rarely exceeding 3 per cent. The water varies
from 9 per cent to 14 per cent. The only other quantitative deter-
minations available are the amounts soluble in water and alcohol
(which should correspond with the above-given Pharmacopoeial re-
quirements) ; and an approxmate determination of the aloin. Tschirch
("Pharm. Post." [37], 2 33, 149, 265) consider the following an ac-
curate process : —

Since the active principles of all aloes, chiefly aloins, are soluble
in CHCI3, while the inert resins are insoluble, the determination
of the CHCI3 soluble constituents suffices for the assay. Five grms.
of aloes are macerated for twelve hours with 5 c.c. of methyl
alcohol, then warmed to 50° C. to 60° C. and treated with 30 c.c. of
CHCI3. After thorough agitation the mixture is set aside and the
chloroform separated and filtered into a tared flask. The residue
i* again treated with another portion of CHClg, the solution added to
that first obtained, the solvent distilled off, and the residue dried at

ALOES.

509

100° C, and weighed. Cape and Uganda aloes yield 80 percent to 85
per cent of CHCI3 extract, Socotrine aloes up to 55 per cent. The aloin
in the chloroform residue may, if desired, be determined colorimetri-
cally by Schouten's reaction, the production of a yellow colour and
strong green fluorescence with a saturated solution of borax. A
standard solution of 0*004 mgm. of aloin in borax solution is prepared ;
thi^ shows a just visible green fluorescence when observed through a
depth of 12 mm. in a vessel placed on black paper. A known weight
of the above CHCI3 residue is treated with a saturated aqueous solu-
tion of borax, and diluted until its degree of fluorescence is identical
with that of the standard. A simple calculation then gives the
amount of aloin presenu.

Tschirch and Hotfbauer (" Schenciz. Woch. fiir Chem. und
Pharm." 42, 12) give the following as the average composition of
commercial samples of aloes : —

Other Chrysaminic

Substances not yield-

Variety.

Aloin.

Acid yielding

bodies soluble in

CHpHO and CHCI3

ing Chrysaminic Acid
soluble in CH3HO and
CHCI3

Inert Resin.

Per cent

Cape aloes, soft

11-8

13-2

Cake aloes, hard

6-2

18-8

Uganda „

30-4

19-6

Barbadoes „

22-4

27-6

Curasao „

16-6

33-4

Socotrine „

3-6

63-4

The Detection of and Distinction between Varieties of Aloes. — Leger
gives the following reaction ("Jour. Pharm Chem." (6), 15, 335) for
detecting aloes and for distinguishing between Cape and Barbados
aloes.

Detection of Aloes. — 0-05 grm. of the sample is dissolved in 100 c.c.
of hot water. After rapid cooling in a current of cold water, the resin
which is thrown down is filtered with the aid of a little talc ; 20 c.c.
of this filtrate are heated on a water bath to 80° C, when a few
particles of sodium dioxide are added to the liquid. Simultaneously
with the evolution of oxygen, the liquid becomes at first brown, then
on adding more dioxide, a §ne cherry-red colour.

Distinction of Cape arid Barbadoes Aloes. — Twenty c.c. of the above
filtrate are treated with 1 drop of saturated solution of cupric sulphate ;
the yellow colour is somewhat darkened ; 1 grm. of NaCl is then added,
the flocculent precipitate thus formed being disregarded, since it is re-
dissolved in the 10 c.c. of 90 per cent alcohol, which is next added,
Cape and Socotrine aloes give a vinous red colour which gradually
fades to a permanent yellow tint. Barbadoes and Curagao aloes give a
bright cherry-red colour which persists for twelve hours. The first
reaction is sensitive to a 1 per mille dilution of the aloes. Since the
colour is then feeble, it may be rendered more evident by acidulating
the coloured solution with HCl and shaking out with ether. The

510 FOOD AND DRUGS.

ethereal solution, when shaken wi^h alkali, gives a marked cherry-red
colour.

Detection of Aloes in Mixtures. — Since aloes are frequently pre-
scribed associated with other drugs which contain oxy-methyl-anthra-
quinones, such as rhubarb, cascara, etc., these are iDest removed by
the addition of a few drops of basic lead acetate solution. Aloins, in
dilute solution, are only very slightly precipitated by this reagent,
while the oxy-methyl-anthraquinones and their glucosides are com-
pletely thrown down. The above reactions are then applicable as
described. Where only rhubarb is present with the aloes, alum and
ammonia may be used as the precipitants, since rhubarb extract thus
treated only gives the faintest peach tint with sodium dioxide. In-
cidentally it has been found that tinctures containing aloes which had
been stored for several years failed to give the reaction with sodium
dioxide, thus confirming the statement of Hirschsohn that these pre-
parations are not stable.

Fawsett (" Pharm. Journ." [4], 19, 401) gives the following method
for detecting the nature of aloes present in compound rhubarb pills.
The coating of the pill is removed and the pill mass powdered.
About 0-2 grm. of the powder is mixed thorougly with 0*035 grm. of
potassium ferricyanide. A small portion of the mixture is placed on
a microscopic slide and made into a thin paste with water, and after it
is spread out in a very thin layer, it is allowed to dry, and is then ex-
amined under the microscope, under a low power with lamplight.
The following appearances will be presented : —

Socotrine. — Rounded pieces of a yellow colour (sometimes brown
or green) ; often looking somewhat like potatoes.

Barbadoes. — Rounded pieces of a decidedly red colour, and similar
in shape to Socotrine. This kind of aloes stains the ferricyanide red
beyond the margin of the aloes itself.

Gape. — Irregularly shaped glassy pieces of a pale green colour.

If the ferricyanide at the edges of the spot is coloured even
slightly red, either Barbadoes or Cura9ao aloes is probably present.
All the other ingredients of Pil. Rhei Co. appear to be unaffected by
ferricyanide of potassium of the strength used.

The three kinds before named are probably those most likely to
be used in pill-making at the present moment, but the following
-colour reactions with ferricyanide may be observed with some varieties
•of aloes not in such general demand viz. : Cura9ao " Livery,"
greenish-brown ; Cura9ao, " Capey," greenish-brown, turning slowly
crimson ; Natal, pale greenish-brown ; Zanzibar, pale brown.

If the still moist spots of the above experiments have the vapour
of ammonia passed over them, " livery " Curasao, Natal, Zanzibar,
Socotrine and Cape all change to various shades of brown, but
Barbadoes and " Capey" Cura9ao turn purple.

Tschirch and Pedersen have examined the well-known test of Born-
triiger, which consists in extracting the sample with alcohol, filtering,
shaking the residua left after evaporation of the alcohol with benzene,
and shaking the benzene solution with ammonia, when a pink or
violet -red colour is obtained. The colour takes some time to develop

ALOES. 511

and the liquids should be left standing for twenty-four hours. The
substance to which this reaction is due is emodin. Most bodies which
are derivatives of oxyanthraquinone yield this reaction, so that such
drugs as araroba, rhubarb or cascara give the reaction. Practically
all aloes except Natal aloes give the reaction.

Tschirch and Hoffbauer give the following details for the recog-
nition of certain varieties of aloes : —

Becognition of Cape aloin. — A 0-1 per cent aqueous solution of the
aloes gives a green fluorescence on the addition of 5 per cent of
powdered borax.

Becognitio7i of Aloe-emodin. — Ten c.c. of an aqueous 0*1 per cent
solution of aloes is shaken for a minute with 10 c.c. of benzene. The
separated benzene is withdrawn, and shaken with 5 c.c. of strong
solution of ammonia. A rose colour is developed.

Distinction of Cape Aloes from Barbadoes Aloes. — Ten c.c. of a 0*1
per cent aqueous solution of aloes is treated with a drop of 5 per cent
CUSO4 solution. An intense yellow colour is developed by Cape
aloes, which after the addition of a trace of NaCl and a little alcohol,
does not change to red.

Distinction of Cape Aloes from Natal Aloes. — A spot of the yellow
solution obtained by the action of strong H2SO4 on the aloes, placed
in a porcelain capsule, should not develop a green colour with a trace
of fuming nitric acid.

Anthraquinone Beaction of Aloes. — One grm. of the aloes is treated
in a porcelain capsule with 20 c.c. of concentrated HNO3 and heated
on the water bath for two hours, the evaporated acid being made up
from time to time ; evaporation is then carried to dryness, and the
residue, treated with water, leaves an insoluble brown powder. This
dissolves in water containing ammonia, giving a violet-red colour.

Kremel (" Zeit. Analyt. Chem." xxxviii. 193) identifies aloes in
medicinal combination by the following method, which gives excellent
results : —

Solid substances are exhausted with alcohol, alcoholic or aqueous
solutions are evaporated on the water bath, and the residues dissolved
in alcohol or water respectively. The latter solution is again eva-
porated, and the residue taken up with water ; the aqueous solution
is then precipitated with excess of basic lead acetate, and the ex-
cess of lead removed from the filtrate by sodium sulphate. By these
operations, all substances which interfere with the reactions are re-
moved. The special reactions for aloin may then be applied. One of
the most characteristic is the conversion into chrysammic and picric
acids ; the solution is evaporated to dryness, and the residue digested
for some hours with 6 parts of concentrated nitric acid of specific
gravity 1*45, 3 parts of water "are added, and the solution heated on the
water bath. On the further addition of water and cooling, the chry-
sammic acid separates in deep yellow to orange crystals. Chrysammic
acid may be identified by the carmine-red colour of its alkali salts, the
violet colour of its ammonium salt, and the insolubility of its barium
salt. The picric acid is recognized by dyeing wool yellow.

Cripps and Dymond's method may be used to confirm the above

512 FOOD AND DEUGS.

reactions when necessary. About 0*1 grm. of the substance if a
solid, or of the solid residue if a liquid, is treated with 1 c.c. of strong
HgSO^ in a porcelain dish and triturated until dissolved, three or four
drops of strong nitric acid are then added and then 30 c.c. of water.
A deep orange to crimson colour results according to the variety of
aloes present. If ammonia be now added the colour is intensified,
usually to a deep claret red. Senna and rhubarb are the only likely
substances which interefere with this reaction.

Aloes in beer, in which it has occasionally been found as a hop
substitute, is detected by extracting the beer with benzene, and apply-
ing the above reactions to the dry residue.

Aloin. — The aloin of the Pharmacopoeia is barbaloin probably
mixed with isobarbaloin. No good indication is given as to its pre-
paration and no satisfactory tests are given. Pure barbaloin melts at
147°, but commercial samples are usually found to melt at about 140°
to 153° and to contain from 1 per cent to 4*5 per cent of ash. Some
samples contain resin, from which they should have been freed. If 1
grm. shaken up with 25 c.c. of warm distilled water yields a perfectly
clear solution, it may be considered free from resin. One per cent
or 2 per cent at most is the highest ash that should be obtained.

If a drop of CuSO^ solution be added to an aqueous solution of
aloin a bright yellow colour is produced, which is changed to red by
the addition of a few drops of concentrated solution of sodium chloride,
and to violet by the further addition of alcohol. If 1 grm. be shaken
with 10 c.c. of petroleum ether and filtered, the filtrate should only
give a faint pink colour when shaken with an equal volume of 5 per
cent ammonia (limit of emodin).

Tincture of Aloes. — This official tincture is made by dissolving the
solid extract of Barbadoes aloes and liquid extract of liquorice in 45 per
cent alcohol. Genuine tincture of aloes has the following characters : —

Specific gravity 0-970 to 0-980

Solid residue 6-5 „ 7-6 per cent

Alcohol by volume 38 ,, 42 ,,

Decoction of Aloes. — A comparatively weak decoction of aloes is
official in the Pharmacopoeia, and it is customary to use a concentrated
decoction, using, of course, a proportionately less quantity in dis-
pensing a prescription. Until recently, when proceedings under the
Pood and Drugs Act were taken, it was usual to use a concentrated
decoction which could not possibly contain the proper amount of
alcohol at so high a reputed degree of concentration. Most concen-
trated samples are now sold as three times the strength of the British
Pharmacopoeia, and the average decoction sold in retail shops is pre-
pared by diluting this preparation with twice its volume of water.
To correspond with the requirements of the British Pharmacopoeia
a genuine decoction should have the following characters : —

Specific gravity 1-004 to 1-006

Solid residue 5 „ 6-5 per cent

Alkalinity of ash as K2CO3 .... 0-5 „ 0-6 „

Alcohol by volume 17 „ 18 „

BELLADONNA BOOT.

513

Decoctions made by diluting concentrated decoctions whose reputed
concentration is too high, will show a deficiency in alcohol and in
solid residue ; the latter must never fall below 5 per cent.

BELLADONNA BOOT.

This drug is official in the British Pharmacopoeia, being the root
of Atropa Belladonna. No official standard exists for the drug. When
it is examined in the entire state, it is necessary to see that obviously
foreign roots are absent, but the principal determination is that of the
alkaloid present, as the drug is used to prepare the standardized pre-
parations official in the British Pharmacopoeia.

In the examination of the powder, the moisture should be deter-
mined, and the ash should not exceed 8 per cent.

A microscopic examination will also afford considerable informa-
tion as to the purity of belladonna root.

Fig. 45. — Powdered belladonna root.

If any doubtful pieces of the whole root are noticed in a sample,
transverse sections should be cut and compared with standard
specimens. In the powder practically no sclerenchymatous cells
are to be found, but plenty of parenchymatous cells, most of which
contain starch grains which measure from 15 to 20 />t, with a few, very
VOL. I. 33

514

FOOD AND DKUGS.

small ones, and a few measuring up to 30 /a. There are plenty of
vessels to be found and many elongated tracheids with narrow blunted
points and numerous large round or oval pits, often arranged in a left
ascending spiral.

Fig. 46. — Powdered belladonna leaves x 240. ccr, cells with sandy crystals of
calcium oxalate; co, collenchymatous cells from cortical tissue of midrib;
ei, epidermis of under surface ; en, epidermis over the veins, with striated
■cuticle ; es, epidermis of upper surface, with striated cuticle and occasional
:Stomata; Z, bast; me, branching cells of spongy parenchyma; nv, fragments
of small vein ; pa, p'a', palisade cells ; pgr, glandular hairs, long and short,
•with unicellular and pluricellular glands ; si, stomata surrounded by three
•or four cells, one of which is smaller than the others ; if, cortical tissue of
midrib ; tr, v, vessels, etc. (Greenish & Collin).

By permission of the Editor of the " Pharmaceutical Journal ".

The following analyses are those of E.
specimens of the root : —

M. Holmes, on air-dried

Woody Roots.

Soft Roots.

Moisture
Soluble ash .
Insoluble ash
Alcoholic extract .
Aqueous extract

Per cent

7-94

3-43

4-60

22-53

15-96

Per cent

10-28

2-20

3-68

29-87

10-50

BELLADONNA BOOT. 516

Determination of Alkaloids. — The official process for the deter-
mination of the alkaloids of belladonna root is described on page 517
under liquid extract of belladonna. The principal alkaloid present
is hyoscyamine. The statement that atropine is present appears to
lack confirmation. It is possibly formed during the extraction of the
root, being isomeric with hyoscyamine. It is probable that scopol-
amine and traces of other alkaloids are present. These are described
on pp. 520 and 521 (atropine, hyoscyamine and hyoscine). Dunstan
and Eansom have devised the following process for the assay of this
root, and in the author's opinion it gives satisfactory results : —

Twenty grms. of the dry powdered root first moistened with a
dilute solution of caustic soda are exhausted in a Soxhlet tube by a
mixture of equal volumes of chloroform and absolute alcohol. The
solution is removed and washed twice with 25 c.c. of water. If
separation is not rapid the solution should be warmed. The chloro-
form retains nearly all the colouring matter, the alcohol and alkaloids
as salts passing into the water. The watery liquid is washed with
chloroform, and then rendered alkaline with ammonia, and extracted
twice with chloroform. The alkaloids are dissolved out by the chloro-
form, which is then evaporated after being once washed with very
dilute ammonia water. The residue is dried and weighed, or titrated
in the manner described on page 506 with methyl-orange as indicator.

Keller's process (" Pharm. Post," 18, 67) as modified by Beckurts
gives good results. Twenty grms. are dried and extracted with a
mixture of 90 grms. ether and 30 grms. chloroform ; 10 c.c. of a 10
per cent solution of caustic soda are then added and the whole shaken
for three hours at intervals. Ten c.c. of water, or rather more if
necessary to make the powder agglomerate, are then added, and the
ether-chloroform separated. The aqueous liquid is washed twice
with more ether, and the mixed solvents are extracted with 25 c.c. of
centinormal hydrochloric acid. The acid is carefully separated, and
the solvent washed with water, the washings being added to the first
portion of acid liquid, and the acid titrated with centinormal alkali.
Prom the amount of acid left, the amount of alkaloids can be calcu-
lated, each c.c. of centinormal acid consumed being calculated as equal
to 0-00287 grm. of alkaloid.

P. C. J. Bird prefers the following method, which the author has
used for some years and found very satisfactory : —

Belladonna root in fine powder 10 grms.

Potassium carbonate 2 grms.

Water 6 c.c.

Dissolve the potassium carbonate in the water, and rub the whole
in a small mortar to a uniform moist granular powder.

Amyl alcohol 3 volumes ^

Chloroform 1 volume Ig.V

Ether 4 volumesj

Add the moistened powder to 20 c.c. of the above solvent, and
macerate for half an hour, with occasional shaking. Force out the
liquid by pressure and cover the powder with 10 c.c. more men-

516 FOOD AND DKUGS.

struum. Agitate vigorously, let stand fifteen minutes and again
force out the liquid. Kepeat this at intervals of a quarter of an hour
until six to ten quantities of menstruum have been used or the powder
is exhausted.

Agitate the mixed ethereal liquids in a][separator with —

Half saturated solution of chloride of sodium .... 10 c.c.

Kun this off and reject. Eotate with 1 c.c. water, separate and
shake the mixed ethereal extracts successively with —

Normal sulphuric acid 4 c.c. '^^

Water 6 c.c. )

Water 5 c.c.

To the mixed acid solutions add —

Solution of ammonia g.s.

to render alkaline. Shake out the alkaloid with successive quan-
tities of —

Chloroform . 10 c.c.

Chloroform 10 c.c.

Chloroform ....*. 5 c.c.

Run off the chloroform into a tared dish, evaporate, dry, weigh,
and titrate as directed by the Pharmacopoeia under liquid extract of
belladonna. The figures obtained by weight and titration should not
differ by more than 1 or 2 per cent.

Before passing on to the galenical preparations of the root, it will be
convenient to briefly notice belladonna leaves, which are also official
in the Pharmacopoeia. These are the leaves of the same plant collected
when the plant is in flower. No official tests exist.

Belladonna leaves should contain from 0 2 per cent to 0"6 per cent
of alkaloids of which the principal part is hyoscyamine, the remainder
being principally atropine.

The leaves may be assayed in the same manner as the root, but
Dunstan and Ransom prefer to use boiling absolute alcohol to extract
the powdered leaves with, diluting the alcoholic liquid with a large
volume of dilute hydrochloric acid, and then removing fat and chloro-
phyll by means ol" chloroform. From the liquid thus purified, the
alkaloids can easily be extracted by adding excess of ammonia and
extracting with chloroform.

Liquid Extract of Belladonna. — This official preparation is an ex-
tract of the root, made with alcohol containing about 78 per cent of
alcohol. It is to be of such strength that it contains 0 75 grms. of
alkaloids in 100 c.c. No other standards are given officially.

A genuine extract should have a specific gravity of 0'890 to 0-920, and
should contain from 11 per cent to 14 per cent of solid matter (but owing
to the veiy variable strength of belladonna root, this figure is liable to
vary outside these limits). The alcohol present in the finished extract
should be from 66 to 69 p?r cent by volume.

BELLADONNA ROOT. 617

The process by which this preparation should be assayed for offi-
cial purposes is as follows : —

Ten c.c. are mixed in a separator with 10 c.c. of chloroform, 50 c.c.
of water and a good excess of ammonia. After shaking and separating,
the extraction with chloroform is twice repeated The mixed chloro-
form liquids are washed with 15 c.c. of warm 5 per cent sulphuric acid
twice. The mixed acid liquids are washed with 3 c.c. of chloroform,
and then rendered alkaline with ammonia and three times washed
with 10 c.c. of chloroform. The mixed chloroform liquids are washed
with 5 c.c. of water containing 1 drop of ammonia, and the chloroform
separated, and evaporated in a tared' dish. The residue, dried below
100" C, is weighed and then dissolved in 10 c.c. of decinormal HCl.
The excess of acid is determined by titration with centinormal soda,
using cochineal as indicator. Each c.c. of centinormal alkali less than
the 100 that would have been required for neutralization of the 10 c.c.
of decinormal acid is equivalent to 0"00287 grm. of alkaloid.

As Bird has pointed out, the B.P. method consists of three opera-
tions : (1) The liberation of the alkaloid by ammonia and solution
of the crude alkaloid in chloroform, (2) partial purification by conver-
sion into sulphate, and (3) complete purification by again rendering
alkaline with ammonia and shaking out the alkaloid with chloro-
form. The emulsification in stage (1) which has proved almost a
complete bar to the use of this process as written, is well known, and
some twelve months ago a modification was proposed (see " Pharm.
Journ." [4], 8, 432) which consisted in first acidifying the diluted
liquid extract, and then removing fatty and resinous bodies by agita-
tion with chloroform, according to Dragendorff's plan, the acid chloro-
form being washed and the washings returned to the original liquid.
Although this adds two more operations to the three already existing,
infinitely less time is consumed in their performance ; also the figures
obtained are generally about 3 or 4 per cent higher, owing to there
being less loss of alkaloid. Since that time an extended experience
of this modified method has proved it to be an absolute preventative
of the troublesome emulsifications incidental to the strict adherence
to the process of the Pharmacopoeia.

Tincture of Belladonna. — This preparation is official and is made
by diluting 2 volumes of the liquid extract with 28 volumes of 60 per
cent alcohol. A properly made tincture should have a specific gravity
of 0-910 to 0-915 and should contain 0-6 grm. per. 100 c.c. of extract-
ive. Its alcohol strength is 57 to 58 per cent by volume. The only
official standard, however, is that 100 c.c. when assayed by the pro-
cess above described should contain 0-048 to 0-052 grm. of alkaloids
in 100 c.c.

These preparations can, of course, be assayed by the other pro-
cesses above described. If this is done, it is best to evaporate the
bulk of the alcohol and then to commence the extraction of the more
or less syrupy liquid.

Liniment of Belladonna. — This is also an official preparation of the
root. It is a mixture of 10 ounces (fluid) of the liquid extract, with 1
ounce of camphor, 2 ounces of water, and sufficient 90 per cent alcohol

518 FOOD AND DKUGS.

to produce 20 ounces. No official standards are given but the pure
preparation should contain 0*375 grm. of alkaloids when assayed by
the official process. In carrying out the process it is necessary first
to dilute 20 c.c. of the liniment with very dilute sulphuric acid, and
filter off the camphor precipitated, or the camphor may be removed
by evaporation. Any camphor remaining is removed by the chloro-
form. A properly prepared sample should have a specific gravity of
0*875 to 0-890, and should contain about 6 per cent of non-volatile
residue. Its alcoholic strength should be 70 per cent to 73 per cent.

There are two semi-solid extracts of belladonna official in the
Pharmacopoeia. The alcoholic extract of belladonna is made by
evaporating the liquid extract to a syrupy consistency, and diluting
with sugar of milk so that 20 fluid ounces of the liquid extract shall
yield 15 ounces of semi-solid extract. This extract contains 1 per
cent of alkaloids.

The green extract of belladonna is the juice of the fresh leaves and
young branches of the plant, evaporated, with the removal of certain
inert coagulable constituents. No standards are given.

In reference to these extracts, it has been pointed out by Farr and
Wright (" Pharm, Journ." 4, 20, 546) that by an oversight the char-
acters of the alcoholic extract are not given in the Pharmacopoeia,
and that the preparation is often sent out in the semi-solid form,
whereas a powdered extract was intended. The green extract of the
leaves contains a very variable amount of alkaloid, and as no standard
exists officially, this preparation must be looked upon as an unsatis-
factory one.

The following figures have been recorded : —

Per cent

Barclay 0-77 to 1-24

Farr and Wright 0-52 „ 1-33

Naylor and Bryant 0-55 „ 1-80

Umney . . • 0-94 „ 1-26

In determining the amount of alkaloids present in these solid or
semi-solid extracts, the alcoholic extract must be thoroughly well
mixed wfth water slightly acidulated with sulphuric acid, and the
assay carried out on the |liquid so obtained. For the assay of the
green extract the process devised by Naylor and Bryant gives the
best results. It is carried out as follows : —

From 2 grms. to 5 grms. of the extract are weighed into a wide-
mouthed flask (for convenience an Erlenmeyer flask is recommended),
25 c.c. of 90 per cent alcohol is added, and the flask with its contents
heated on a water bath under an inverted condenser or other arrange-
ment that prevents loss of alcohol and provides facilities for exhaustion.
This operation is twice repeated with two more quantities of 25 c.c.
of 90 per cent alcohol. After each operation the alcoholic solution
in the flask is allowed to become cold, and filtered, and the filtrates
are united.

To make sure that extraction of the alkaloidal content is complete,
the residue in the flask is warmed with a 5 per cent solution of
hydrochloric acid and filtered. The filtrate is then tested with solution

BELLADONNA EOOT. 519

of iodine in potassium iodide. Three extractions with alcohol are
usually sufficient for the purpose.

To the alcoholic solution of the alkaloid an equal volume (75 c.c.)
of a 5 per cent solution of the hydrochloric acid of the Pharmacopoeia
is added, and the mixture shaken up three times successively with 15
c.c. chloroform. After separation and rejection of the chloroformic
liquids the acid solution is rendered distinctly alkaline by the addition
of solution of ammonia and again shaken up three times successively
with 10 c.c. chloroform. The chloroformic solutions, after separation
are mixed and evaporated, and the residue dried over a water bath
until it ceases to lose weight. The dry alkaloidal residue is titrated
as the Pharmacopoeia directs in the final stage of the process for de-
termining the proportion of alkaloid as given under extractum bella-
donnas liquidum.

The chloroformic separations take place quicker and cleaner than
is the ease in the Pharmacopoeia process for liquid extract of belladonna.

It may be noted that the difference between the amount of alka-
loid obtained by weighing and that indicated by subsequent titration
is less than 0*01 grm.

Belladonna Ointment. — This official preparation should contain
0"6 per cent of alkaloids. No official method of assay is given. Bird
has given the following process, which works well : —

Belladonna ointment, B.P 10 grms.

Benzol 20 c.c.

Water 10 c.c.

Diluted sulphuric acid 7 c.c.

Melt the ointment in a small dish, pour into a separator, rinse the
dish with the water and acid, add the benzol and agitate vigorously.

Separate the benzol, and wash the aqueous layer twice with suc-
cessive quantities of

Benzol 10 and 10 c.c.

Warm the mixed benzol washings with

Diluted sulphuric acid 3 c.c.

Water 3 c.c.

agitate and separate. Shake a second and third time with

Water 10 and 10 c.c.

Eeject the benzol and return the acid liquids to the separator
Then make alkaline with

Solution of ammonia ........

Shake out with three successive quantities of

Chloroform 10, 10, and 10 c.c.

adding if necessary

Saturated solution of ammonium carbonate ..... g.s.

Wash the mixed chloroform solutions with

Water 3 c.c.

Solution of ammonia 10 drops.

Solution of ammonium carbonate 2 c.c.

520 FOOD AND DEUGS.

Evaporate in a tared dish, dry below 100° C, weigh and titrate as
directed in the Pharmacopoeia.

Belladonna Plaster. — An ofificiai belladonna plaster exists which
is a mixture of resin plaster (which contains colophony, soap and lead
plaster) with liquid extract of belladonna from which the bulk of the
liquid constituents have been removed by evaporation.

It should contain 0*5 per cent of alkaloids. According to Hender-
son the best means of getting this preparation into a suitable con-
dition for assay, is to disintegrate it with ether, and then shake the
emulsion with a mixture of acetic acid and water, the alkaloids and
lead being dissolved in the acid. The details of the method are as
follows : —

Weigh 5 grms. of the plaster, and introduce it into a stoppered
glass separator, with 25 c.c. of ether ; allow the plaster to disintegrate.
When the contents of the separator present the appearance of an
emulsion, add 5 c.c. of a mixture of glacial acetic acid and water (three
partsof the former to two parts of the latter), shake for thirty seconds
and set aside until the acid liquor has completely separated. Draw
off the lower layer into a small beaker, and again agitate the ether
solution with 5 c.c. of the dilute acetic acid of the B.P. and draw off
as before. To the united acid liquors in the beaker add dilute sul-
phuric acid in slight excess, stir well, and allow the sulphate of lead
to subside. Filter the solution through a small filter into a separator,
transferring the whole of the sulphate of lead on to the filter by
means of a glass rod tipped with rubber ; allow to drain. Eemove the
funnel from the separator, and wash the lead precipitate with distilled
water until a drop of the filtiate gives no precipitate with Mayer's
reagent. Concentrate the washings to a small bulk and add them to
the contents of the separator.

It wiil be found to be advantageous to use a filter pump in wash-
ing the lead precipitate, but it is not essential. The separator now
contains the extract of belladonna, freed from the other constituents
of the plaster. Add excess of solution of ammonia and 10 c.c. of
chloroform, shake well for thirty seconds, and draw off the chloroform
into another separator. Eepeat this treatment with two more suc-
cessive portions of chloroform of 5 c.c. each. Mix the chloroformic
solutions of the alkaloids, and shake out the alkaloids with three suc-
cessive portions of dilute hydrochloric acid, using 5 c.c. for each
shaking. To the mixed acid solutions, in a separator, add excess of
solution of ammonia and 10 c.c. of chloroform, shake well, and draw
off the chloroform into a weighed dish, repeat the shaking with two
successive portions of chloroform, using 5 c.c. for each, draw off as
before, and allow the chloroformic solutions to evaporate spontane-
ously. Dry the residue in the air oven at a temperature not exceed-
ing 93" C, until the weight is constant, and weigh.

Atrojnne. — Atropine C1-H23NO3, and its sulphate (Ci7H23N03)2-
HgSO^ are official in the Pharmacopoeia.

Atropine is described as being soluble in 300 parts of water (tem-
perature not stated), and readily soluble in alcohol, chloroform, and
ether. It has an alkaline reaction, and when applied to the eye has

BELLADONNA EOOT. 521

a powerfully dilating action on the pupil. An alcoholic solution, on
warming, precipitates a solution of mercuric chloride, the precipitate
being yellow but soon turning red. The aqueous solution when
treated with solution of auric chloride gives a citron yellow precipi-
tate which when recrystallized from boiling water acidulated with HCl
has a minutely crystalline appearance, and when dry a dull powdery
appearance (distinction from hyoscyamine). When moistened with
fuming nitric acid and evaporated to dryness on a water bath, the resi-
due gives with freshly prepared alcoholic solution of potash, a fugitive
reddish-violet colour. It leaves no ash.

Sulphate of atropine is a crystalline substance melting at 183'' C.

It is insoluble in ether and chloroform, and leaves no ash on
ignition.

No other official requirements are given.

As has been mentioned above, it is doubtful whether atropine
exists in belladonna, and it is possible that it is formed by isomeriza-
tion of the hyoscyamine present. It is frequently obtained from the
rhizome of Scopola Carniolica. It is soluble in 500 parts of cold
water, not 300 parts as stated in the Pharmacopoeia, and should melt
at 115° to 116°. Commercial atropine frequently contains a little
hyoscyamine, which low^ers its melting-point, but raises that of the
aurichloride, which with pure atropine melts at 137°, as against 160°
for the hyoscyamine compound. A solution of atropine is optically
inactive, whilst that of hyoscyamine is optically active.

According to the British Pharmaceutical Codex, pure atropine sul-
phate melts at 185° to 186°.

Hyoscyamine. — The pure alkaloid, C^^HggNOg, which is isomeric
with atropine is rarely employed. The only salt that is official is the
sulphate 2(Ci7H23N03)H2S04 2Hp. This is described in the Pharma-
copoeia as a crystalline powder, deliquescent, odourless and having a
bitter taste. It melts at 206°. It is soluble in 0*5 part of water and
in 2-5 parts of 90 per cent alcohol. A solution in water yields no
precipitate with platinum chloride, and with auric chloride it yields a
precipitate of a yellow colour, soluble in boiling water acidulated with
HCl, and deposited on cooling in the form of brilliant golden scales
(distinction from atropine). It leaves no ash. According to the
British Pharmaceutical Codex it is soluble in 4*5 volumes of 90 per
cent alcohol. Commercial samples melt below 206°, but they should
not be allowed to melt below 200°, or the limits of impurities will be
too great. The free base crystallizes in needles or prisms.

If 10 mgs. be added to 2-5 c.c. of nitric acid, evaporated to dry-
ness, and alcoholic solution of potash added, a violet colour should
result.

Hyoscine. — This alkaloid, also known as scopolamine, is official in
the form of its hydrobromide. It is described here for convenience,
although it probably does not occur in belladonna root, but is obtained
from other solanaceous plants. It is a mixture of stereo-isomeric
varieties of the base, having the formula C-i7H22N04HBr . 3HoO
(hydrobromide). The Pharmacopoeial formula is incorrect. The
Pharmacopoeia requires this salt to lose rather more than 12 per cent

522

FOOD AND DEUGS.

of its weight on heating to 100° C. and the resulting mass to melt at.
193° to 194°. It forms with auric chloride a compound melting at
190°. Its aqueous solution slightly reddens litmus. The statements
as to its melting-point are rather misleading. When heated in a capil-
lary tube the hydrated salt melts at about 100°. If dehydrated over
sulphuric acid, the salt as met with in commerce melts at 181°. The
purer optically inactive variety melts at 180°, and the laevorotatory
variety at 193°. The aurochloride formed without the addition of
free hydrochloric acid melts at 215°, but the product formed in the
presence of free HCl melts at 193°. The nitric acid colour reaction
yielded by hyoscyamine is also yielded by hyoscine. The free base is
a syrup, and does not crystallize.

CANTHAEIDES.

Cantharides, or Spanish flies, as the insects are often termed, are the
dried beetle, Cantharis vesicatoria. This is the only variety which is
official in the British Pharmacopoeia under the name Cantharis. The
" Chinese cantharides " or Chinese flies, are the dried beetle, Mylabris
dehor ii, and are quite similar in properties to the Spanish beetle, both
of them being used as vesicating agents.

The important constituent of these beetles is cantharidin, which
occurs both in the free and in the combined condition. The estima-
tion of this ingredient is the most important determination to be made
in their examination. The two varieties of cantharides have the fol-
lowing characters : —

Mineral matter
Free cantharidin
Combined cantharidin
Total cantharidin .
Moisture

Spanish flies,"

5
0-3
0-05
0-4
10

Per cent
to 6-5

0-57
0-3
0-85
13

Chinese flies.

Per cent
4-0 to 5-8
0-6 „
0-1 .,
0-7 „

10 „

1
0-8
1-9
13

The Chinese insects contain more cantharidin than the Spanish,
and are the better source for the preparation of cantharadin.

Colledge (" Pharm. Journ." 1910, 674) has examined six samples of
Cantharides of different species in the following manner : —

The powdered flies were exhausted with benzene and the solvent
evaporated. The residue was exhausted with water slightly acidified
with HCl, at boiling temperature. The acid solution was then ex-
hausted with chloroform and-the chloroform evaporated. This residue
was then extracted with petroleum ether to remove fat, and the

residue finally dried at 60° to 65°
tharidin were obtained : —

The following amounts of can-

CANTHARIDES. 523

Mylabris oculata 0-615 per cent.

,, holocericea 1-3 ,,

Decatoma lunata 1-0 „

Electica wahlbergia 0-32 „

Cantharis vellata 2*73 ,,

Lytta coelestina ...... 1'89 „

Chinese cantharides gave 1'2 per cent of cantharidin.

A useful method for the determination of cantharidin is that of
Greenish and Wilson ("Pharm, Journ," 4, vi. 255). Their method is
as follows : —

Determination of Total Cantharidin. — Twenty grms. of the flies
in No. 40 po\9der are mixed in a small mortar with 25 c.c. of a mixture
of:—

Glacial acetic acid . . . . . . 1 volume

Itectified spirit ....... 2 volumes

Chloroform 3 „

The moistened mass is covered over for about an hour, and then
allowed to dry spontaneously or at a slightly raised temperature.
This is easily accomplished without loss of cantharidin. The dried
mass is then packed in a Soxhlet extractor, and exhausted with
chloroform, the latter being first used to rinse out the mortar em-
ployed.

About one hour will usually suffice for complete extraction, but
complete exhaustion should always be ascertained by removing the
flask with the chloroformic solution, and continuing the extraction with
a little fresh chloroform.

The chloroformic solution thus obtained is placed in a separator
containing a little water, and the acetic acid, which passes into the
water, is almost, but not quite, neutralized with solution of potash,
and the whole well shaken.

The chloroformic layer is run off into a glass dish and evaporated,
cautiously towards the end. The residue consists of fat, in which can
be seen crystals of cantharidin. The fat is removed by washing with
petroleum spirit (the washings being set aside), leaving in the dish
crystals of cantharidin mixed with a green substance of a resinous
nature. This residue is allowed to dry, and is then washed with suc-
cessive small quantities of absolute alcohol until all green matter is
removed, and perfectly white cantharidin remains. The alcoholic wash-
ings are carefully evaporated.

The cantharidin, dissolved or mechanically removed whilst wash-
ing out the fat with petroleum spirit, is now recovered ; 20 c.c. of 10
per cent solution of caustic potash are added to the petroleum spirit
solution, and the mixture warmed until the fat is completely saponi-
fied; during the process most of the petroleum spirit is dissipated.
The soap solution thus produced is diluted with warm water and
transferred to a separator, sufficient petroleum spirit being added
to dissolve the fatty acids when liberated ; it is now acidified with
hydrochloric acid, when the fatty acids rapidly rise and dissolve in the
petroleum spirit. The aqueous layer is quickly run off from beneath
the petroleum spirit solution into another separator, the petroleum

524 FOOD AND DEUGS.

spirit solution washed with water and the washings added. The can-
tharidin is then removed by shaking with successive quantities of
chloroform as long as cantharidin is removed ; this must be ascer-
tained. In the chloroformic solution thus obtained the residue from
the alcoholic washings of the crystallized cantharidin is dissolved.

The chloroform now contains in solution chiefly cantharidin and
the green resinous matter. It is placed in a separator and shaken
with lime water, containing excess of calcium hydrate suspended in
it, and solution of common salt, the latter causing the chloroformic
layer to separate more readily.

In this way the cantharidin passes into aqueous solution, probably
as cantharidate of calcium, whilst the chloroformic layer containing
green resin and colouring matter is rejected.

The aqueous solution is filtered, acidified with hydrochloric acid,
and shaken out with chloroform as before. This chloroformic solution
is added to the cantharidin previously separated, evaporated cautiously,
dried in a desiccator, and weighed. In this way a crystalline residue
of cantharidin only very slightly coloured is obtained.

Determination of Free Cantharidin. — This is accomplished in the
same way as the determination of total cantharidin, with the exception
that the drug is not moistened with the acetic acid mixture before ex-
traction, and, no acetic acid being present, the washing of the chloro-
formic solution with water becomes unnecessary.

A useful summary of the proposed methods for the assay of can-
tharidin by Self and Greenish ( 'Pharm. Journ." [4], 24, 324) has been
published and the authors finally adopt the following process : —

Twenty grms. of cantharides in fine powder are moistened with 3
c.c. of strong HCl and extracted in a Soxhlet with 80 c.c. of benzene.
The benzene is driven off, the last traces being removed in a current of
air on a water bath. The benzene (recovered by distillation) is ex-
tracted three times with a 1 per cent solution of KOH to recover traces
of cantharidin. The alkaline liquid is acidified with HCl, made up
to 105 c.c. with water and added to the mixed fat and cantharidin in
the extraction flask. The mixture is boiled for ten minutes under a
reflux condenser, the fat allowed to separate and as much as possible
of the aqueous solution transferred to a large separator. The boiling
with water (50 c.c.) is repeated four times and the mixed aqueous ex-
tracts are rendered thoroughly acid with 3 c.c. HCl, and extracted
four times with chloroform. The residue from the chloroform extract
is washed three times with three portions of 5 c.c, 5 c.c, and 2 c.c.
of equal volumes of absolute alcohol and petroleum ether which has
previously been saturated with cantharidin. The washing fluid is
poured through a funnel containing a plug of cotton wool, and the
flask and wool finally washed with a few c.c of petroleum ether, until
nothing further is dissolved. A few c.c of chloroform are then poured
through the wool into the flask, in case any crystals of cantharidin
have been transferred to the wool and the cantharidin in the flask
dried to constant weight at 60° to 65°.

The following is the official process in the German Pharmacopoeia : —

The powder is extracted with chloroform and dilute hydrochloric

CINCHONA. 525

acid for twenty-four hours, and an aliquot part filtered out and
evaporated ; the residue is extracted with petrolem benzin for twelve
hours, the liquid filtered off, and the extraction repeated several times.
The undissolved portion is then treated with water containing a trace
of ammonium carbonate as long as this removes any colour, then dried
and weighed ; if it is then resinous or dai k in colour, the cantharidin
is extracted from it with hot acetone. Not less than 0*8 per cent of
crystalline cantharidin is required.

The method proposed by L6ger (" Journ. Pharm. Chem." 6, 17,
457) gives good results. He prefers to extract with benzene, after
acidifying with hydrochloric acid.

Tincture of cantharides is an extract of 1 part of cantharides with
80 parts of 90 per cent alcohol. It is so weak in cantharidin as to be
almost impossible to assay accurately. The following are the char-
acters that a genuine tincture should have : —

Specific gravity at 15° . . 0-835 to 0-840

Eesidue dried carefully at 90° to 100° 0-15 grm. to 0-17 grm. per 100 c.c.

Alcohol 89 per cent to 90 per cent by volume.

CINCHONA.

There are a number of species of Cinchona whose barks are more
or less rich in alkaloids of which quinine is the principal. Various
species of Bemijia also contain the same alkaloids.

Any of these plants may be used as the source of preparation of
the cinchona alkaloids, but there is only one that is official for other
purposes in the British Pharmacopoeia. This is the bark of Cinchona
succirubra, from which the galenical preparations of Cinchona should
be made.

The Pharmacopoeia requires that this bark, when used for any
purpose other than that of obtaining the alkaloids and their salts,
should contain from 5 per cent to 6 per cent of alkaloids of which
not less than half should be quinine and cinchonidine when estimated
by the following official method : —

Twenty grams of the bark in powder are mixed with 6 grms. of
calcium hydroxide and moistened with water (20 c.c.) ; allow the mix-
ture to stand for an hour or two, and then transfer the whole to a
flask connected with a reflux condenser. Add 130 c.c. of a mixture of
of 3 volumes of benzol and one of amyl-alcohol, and boil for half an
hour. Decant the liquid and repeat the boiling, etc., repeat a third
time and mix the liquids and wash the powder on a filter with more
of the liquid until exhausted. Place the mixed liquids whilst still
warm in a stoppered separator. Add 2 c.c. of dilute HCl mixed with
12 c.c. of water. Shake well, and separate the acid layer. Eepeat the
extraction with slightly acidified water until all the alkaloids are ex-
tracted. Neutralize w4th ammonia, and concentrate to 16 c.c. Add
about 1-5 grms. of sodium potassium tartrate dissolved in 3 c.c. of
water, and stir the mixture with a glass rod. Insoluble tartrates of
quinine and cinchonidine will separate in about an hour. These are
collected on a filter, dried and weighed. They will contain 0-8 of

526

FOOD AND DKUGS.

their weight of alkaloids, from which the percentage is calculated. To
the mother liquor add ammonia in slight excess. Collect, wash, and
dry the precipitate, which contains the other alkaloids. The sum of
the two may be taken as the total alkaloids.

The principal alkaloids occurring in cinchona bark are the following.
Of course, there are numerous other less important alkaloids, and
they do not all occur associated in each species of cinchona, but those
now described are the only ones having any practical importance
irom the analyst's point of view.

Fig. 47. — Powdered cinchona bark.

Quinine, quinidine, and quinicine are isomeric bases of the formula
C2oH,,N,0,.

Cinchonine and cinchonidine are isomers of the formula

Hydrocinchonine and hydrocinchonidine are isomeric, and have
the formula C19H24N2O.

• Quinamine has the formula C19H24N2O2.
5 Cupreine has the formula Ci()H22N202.

;-;*'J^Of these the only official alkaloid is quinine. This is described
on p. 532 and tests for cinchonine and cinchonidine, which have some
practical importance, will be found mentioned on the same page.

The examination of cinchona bark is confined to the determination

CINCHONA. 527

■of the mineral matter : a microscopic examination (if in powder) : and
a determination of the alkaloidal value. If the last named be re-
quired for official purposes, the process described above should be used ;
but from the point of view of the manufacturer of the alkaloids, a
fuller separation will be necessary.

The ash of cinchona bark should not exceed 4 per cent to 5 per
cent.

Under the microscope the powdered bark should show very large
bast fibre with characteristic pits, but no other cells should be found
of a sclerenchymatous character. The parenchymatous cells are
deep red or brown in colour, and after digestion in weak potash solu-
tion, crystals of precipitated alkaloids may be found.

As alternative methods for the assay of cinchona bark, the follow-
ing are those which give the best results : —

De Vrij (modified). — Twenty grms. of the bark are powdered and
mixed with 5 grms. of quicklime and 50 c.c .of water. The whole is dried
at 70", and transferred to a flask with a reflux condenser, and boiled
with 200 c.c. of alcohol of at least 93 per cent strength. After an
hour's boiling the liquid is cooled and filtered off. The residue is
again boiled with 100 c.c. of the alcohol, and this is also filtered ott".
The residue is washed twice with 50 c.c. of alcohol, and the mixed
alcoholic liquids are rendered acid with dilute sulphuric acid. Calcium
sulphate is filtered off, washed with a little alcohol which is added to
the main alcoholic liquid, and this is then concentrated to expel
alcohol and again filtered, the insoluble matter being well washed \^ith
water acidulated with sulphuric acid. The filtrate which contains
the alkaloids as acid sulphates is then transferred to a separator after
being concentrated to about 40 c.c. and rendered alkaline by soda
solution. The liquid is now extracted four times with 30 c.c. to 35
c.c. of chloroform, and the mixed chloroformic liquids, after being
once washed with water, are evaporated, and the residue dried and
weighed. The weight represents the total alkaloids in the 20 grms.
of bark.

Prollius Process {modified). — Ten grms. to 20 grms. of very finely
powdered bark are treated with twenty times its weight of a mixture
of 85 parts of ether, 10 of alcohol and 5 of ammonia of specific gravity
0-960 (all by weight).

Place the mixture in a well-fitting glass-stoppered bottle, weigh
and shake well at intervals for four hours. Maintain the original
weight by the addition, if necessary, of more solvent. Pour off as
much as possible perfectly clear, rapidly closing the bottle again.
Ascertain how much has been poured off by weighing the bottle again.
Distil off the solvent, evaporate the residual liquid, and weigh the
residue in a tared beaker, when dry. The weight of the residue re-
presents the alkaloids in the portion of the solvent evaporated, and
as the proportion of this to the whole amount used is known the
percentage can bo calculated.

Squibb's Method. — Squibb effects the exhaustion of the powdered
bark by means of acetic acid of 10 per cent strength in an apparatus
which is in effect a small percolator. The complete extraction of 10

528 FOOD AND DEUGS.

grms. of the powder is effected in thirty-six hours, the entire percolate-
measuring 180 c.c. to 200 c.c. This percolate is evaporated until the
residue, though still liquid whilst hot, is semi-solid on cooling. The
weight of this residue usually amounts to about 35 per cent to 38 per
cent of the bark used. The extract is dissolved in a mixture of
ammonia and alcohol, more ammonia is added to ensure the liberation
of the whole of the alkaloids, and the separation effected by shaking
out with chloroform. The alkaloids are then taken up with decinormal
sulphuric acid, precipitated with decinormal potassium hydrate, and
again taken up with ether. Finally, the varnish-like residue left on
evaporating the ethereal solution is weighed in order to get the ap-
proximate percentage of alkaloids, after which the alkaloids are titrated
with decinormal acid.

Where a separation of certain of the alkaloids is required, the fol-
lowing processes may be employed. It is to be remembered, however,
that nearly every process for the separation of cinchona alkaloids is
only approximate, and will often give erratic results in unskilled
hands.

As a rule the only practical question for solution is the amount of
quinine that can be obtained from the bark, the remainder of the
alkaloids being of but quite secondary importance.

For this purpose a modification of a process devised by De Vrij is
fairly accurate if carefully carried out.

The alkaloids from 50 grms. of the bark, extracted by one of the
above-described processes and in a fine state of division, are treated in
a closed vessel with ten times their weight of pure ether free from
alcohol. The mixture is well shaken and left for twelve hours, when
it is filtered and the residue washed with a small quantity of ether.
The ethereal solution is evaporated to dryness, and the residue weighed.
It consists of quinine, quinidine, and cinchonidine in heavy traces, and
amorphous alkaloids. It is dissolved in ten times its weight of 60 per
cent alcohol and rendered acid wiih decinormal sulphuric acid. An
alcoholic solution of iodine is then added until no further precipitation
takes place. Excess of iodine must be avoided. A black precipitate
of herepathite, an iodine compound of quinine of the formula
4C^oH24N202-3H^S042HI.I-f 3Hp, is formed. This is allowed to
stand for twelve hours, and the precipitate is then filtered off", washed
with strong alcohol, dried at 100°, and weighed. The weight multi-
plied by 0*55055 gives the weight of quinine in the mixed alkaloids
operated upon.

De Vrij has later recommended using a solution of the iodosulphate
of the amorphous mixture of alkaloids (known as quinoidine) as a pre-
cipitant instead of iodine. This prevents the possibility of the forma-
tion of periodized products.

David Howard, instead of converting the ethereal solution of quin-
ine and impurities into this iodine compound prefers to agitate the
ether with excess of dilute sulphuric acid, and, after heating the
aqueous liquid to boiling, to add ammonia until the liquid is neutral
to litmus, when the quinine crystallizes out almost entirely as sul-
phate (this salt is practically insoluble in a solution containing

CINCHONA. 529

ammonium sulphate). The crystals are filtered off, and washed with
a little cold water, pressed between filter paper, and dried at 100° ;
84*7 of the anhydrous salt are equivalent to 100 of the crystallized
sulphate.

The summary on pages 530, 531 of De Vrij's process for the
complete separation of the cinchona alkaloids is due to A. H. Allen.

Vigneron's Process. — This process ("Jour. Pharm. Chem." 21, 180)
gives excellent results if carefully carried out.

The total alkaloids of 25 grms. of bark are treated with twenty
times their weight of pure ether and shaken well with five or six
small pieces of pumice stone the size of a pea, previously moistened
with 98 per cent alcohol. The small amount of alcohol thus intro-
duced facilitates the separation of the quinine from the other alkaloids.
The mixture is allowed to macerate for six hours at about 15° C, with
occasional agitation, then filtered into a porcelain capsule, from which
the ether is allowed to evaporate spontaneously. The residual alka-
loids insoluble in ether are again macerated with a similar quantity of
ether for twelve hours ; the ethereal liquid is filtered into the same
capsule and gently evaporated at about 15° C. To the residue 5 c.c.
of alcohol and 100 grms. of a saturated aqueous solution of quinine
sulphate are added, followed by 10 drops of 1 per cent aqueous haemo-
toxylin solution. The capsule is then placed on the boiling water bath
to drive off the ether and alcohol. Meanwhile 2 or 3 c.c. of 10 per
cent sulphuric acid is added, then gradually a little 5 per cent acid
uutil the liquid assumes a lemon-yellow tint. If thie faint acidity re-
quisite be exceeded, a few drops of dilute ammonia are added until
only a faint yellow colour is visible. The solution is then set aside in
a cool place for twenty-four hours, and the crystals which have formed
are collected on a tared filter, washed first with saturated quinine sul-
phate solution, then with a few c.c. of distilled water used in portions.
The mixed sulphates of quinine and cinchonidine are then dried and
weighed ; 0*75 grm. of these sulphates is then weighed off, dissolved
by boiling in 85 c.c. of saturated solution of pure quinine chromate,
and treated with 0*20 grm. of pure KgCrO^ dissolved in a little water,
allowed to cool, and the precipitated quinine chromate collected on a
small tared filter, and washed with saturated solution of quinine chro-
mate to bring the volume of the filtrate to 100 c.c. This filtrate may
be tested for cinchonidine by the addition of NaOH solution. The
crystals are then slowly washed with another 100 c.c. of saturated
solution of quinine chromate, drained, dried at 100° C, and weighed
as (C^(,H24N202)2Cr04. Since 0*746 grm. of pure quinine sulphate
gives under these conditions 0*764 grm. of chromate, the equivalents
may be taken as practically 75 and 76. If the first filtrate from the
precipitated chromate gives no precipitate with NaOH, the amount
of free chromate in the liquid may be determined volumetrically by
means of KI and sodium thiosulphate. In this case, a solution of 0'2
grm. of K^CrO^ in 154 c.c. of water may be conveniently used as the
precipitant ; each c.c. of this will be equivalent to 0*005 grm. of an-
h\drous quinine sulphate. At the same time the amount of thio-
sulphite used up by the iodine liberated by 100 c.c. of saturated

VOL. I. 34

530

FOOD > AND DEUGS.

w
6

o
o

<
Ph

-^ £

2«

— » <a

lO g o

o'g,

o "« r-° -5 "o ^ *
"H 3 .2 ^ S 2 -n

i •§ ^ '3 ^ M ^ .2

I ^" Ml si

2 o o 2 o^»'g ft

ai rsi>,°ft

O O a? <*-

P,£ o
^ Sig

^|'S;S a « § ^'S

a) r' ^

g^'g.s grig's^
llilill^iir^

2^ ss 13^ ca.^"^

■e^-g-

^lg5&^5

§S

& 5 c

eS +s

g--^ a o 2 -a S ^ . -

bib-fl o

>-3,

- « « cJ

Jll.S||gl|7

.;2 i^J O o ^ 'C .S

.2.5 ^ S.2 4> c b c o
s S ^®'o ® 2 oT^^

EHS-§.si;Ss-3^5

fiillii

BSg.

•f-^-e

S SbhC'

» *" S -r C . o
■S -g o So ® « g

filtlil

=» 2:2 t ■SS3
"— ija s.s

x 3 ="■•- S *>

* 3^.5 ji« =« C

S ,2 03 d * -3-

« $-= - - "-

c * s s >•

®x tr5 2 2j=

8oa^-3;^tjsi

•5 « cj f= og -jr w
'3«2.t2o<g--2

X t? ft— » "

05 as ° JS C *

S3"ce«^^Sg
13 fc c ^ o g Q-3

JS«

eg c§ 8:5 go-

o t« W X « o£^
"" -►i . -is <= "C c

CINCHONA.

531

M
o

w
o

o
<J

CZ2

^•^ 2 ca ^

o o s .B

'.a

-^-rt o

« >.2 §«

« .2 ^ .& g

-iJ -H (rt t) ?

"S ^ ^ .2 ^
^3g p a
;S g g o^«

^■S'^J'F

•5^ « o o
> 'd . ^ 3

.2 S T3 p g

I nil

-I g 03

o S o

•3 3 ^ >.l .2

-3 05 • « «
a S " a;-

;g c: ^-^ |

o| li-
ft
£-6

-d" a ;i 2 .'&

"-go X! ®

p -*, rH 5
•^ ° ^ "-TS

rt .2 X o "^

° = o ^ a

p § S.^3

o « .S o p
o SU'-^ 05 o

.^ S fP ^ ^
OJ +3'' ,p j3

=« p 3j la

^^ 2-1.2 8

a?'S'33'^ ' 'tis

is^

-S-

^ -2 S « 3 o 2

5C t--

tP CC

-51 = 0^

•2 X 0 « ^

0,-73 la's 6x

^
1^^^

i.-S p >.
be a) ''j ^

11822

O,'^ 13 "rt -^

"^ .2 ■*^ Sb 3

-73 t3
m -t-3

S-.S-

■I I «3^
■»^t..pW)«<D.SP!»2sJ'gajpa!«?3
•2*-o^..-§»-S'5j^

g ^ 8 g § ^'> . =0 g^
^^«8^^o|P'^

.a
Eh

o .2 .2 'd 13

'o 'O J C ■

.p o o .-ti =* •

^ J-- ® o
oj .2 fl 5 S
^■S^'*= p
a p '^^ 03
g yii-^13
2'3x!5a'T3

sC O 05 O 03

5m FOOD AND DRUGS.

solution of quinine chromate is noted ; this number will be jS. Oper-
ating on the above quantities, the first 100 c.c. of chromate filtrate
collected will contain the equivalent of 4 c.c. of the titrated solution
of KgCrO^. If this filtrate requires x c.c. of thiosulphate to titrate the
iodine it liberates, the amount of quinine sulphate present in the

X — 3 + 4:

mixed sulphates may be found from the formula 75 - ^ = the

number of centigrams present in the 0*75 grm. of sulphates taken.

Quinine CgoHg^NgOg, SHgO is ofiicial in the British Pharma-
copoeia in the form of its hydrochloride C20H24N2O2, HCl, SH^O ;
acid hydrochloride G2oH24N2^2' 2HC1, SHgO ; and sulphate
[(C,„H,,N,0J,H,S0J,15H,0.

The pure alkaloid contains 14 per cent of water of crystallization
which is lost at 120° to 125°, the anhydrous alkaloid melting at about
175°. It should be practically free from cinchonine and cinchonidine,
which is assured by 1 grm. dissolving on warming in 6 c.c. of abso-
lute alcohol and 3 c.c. of ether, and remaining perfectly clear when
the solution is cooled.

The following typical reactions are characteristic of the base or its
salts. To 1 c.c. of a 1 per cent solution in water containing sufiicient
H2SO4 to dissolve the alkaloid, a small quantity of bromine water is
added, and then a little dilute ammonia. An emerald-green colour
results (thaleoquin reaction). If 0*05 grm. be dissolved in 5 c.c. of
alcohol with 5 c.c. of dilute sulphuric acid, and a little acetic acid,
and 5 c.c. of a saturated solution of iodine in alcohol be added to the
former solution heated to boiling-point, bronze or olive-green crystals
of quinine iodo-sulphate will separate on cooling (Herapath's re-
action).

Quinine hydrochloride contains 9 per cent of water which it loses
at 100° ; 92 per cent of hydrochloric acid, which is determined as
silver chloride in the usual manner ; and 81-7 per cent of anhydrous
quinine. This is determined by dissolving 1 grm. in slightly acidu-
lated water, rendering alkaline with KgCOg, extracting with ether and
weighing the residue.

The acid hydrochloride should contain 12 per cent of water of
crystallization ; 16'2 per cent of hydrochloric acid, and 71"86 per cent
anhydrous quinine.

Quinine sulphate is the " quinine " of commerce, and is by far the
most important form in which the alkaloid is found. By drying at
100° it should lose not more than 15-3 per cent of water of crystal-
lization ; and should contain 11-12 per cent of sulphuric acid, which is
determined by dissolving 1 grm. in slight excess of dilute hydrochloric
acid and precipitating in the usual manner with barium chloride. It
contains 73-55 per cent of anhydrous quinine, which is determined as
described under the hydrochloride. The Pharmacopceial standards of
quinine sulphate are somewhat stringent. It is directed not to afford
any appreciable reaction characteristic of cinchonine, quinidine,
cupreine or amorphous alkaloid, and should not yield more than 3 per
cent of (impure) cinchonidine when tested as follows : —

Test for Cinchonidine and Cinchonine. — Four grms. are dissolved

CINCHONA. 633

in 120 c.c. of boiling water. The solution is cooled to 50° C, with
constant stirring. The sulphate of quinine which crystallizes is
separated by filtration. The filtrate is reduced to 10 c.c. by evapora-
tion. When cold it is shaken with 10 c.c. of ether and 5 c.c. of
ammonia (specific gravity = 0-959). The whole is set aside in a
stoppered vessel for twenty-four hours. The crystals separating are
collected on a tared filter, washed with ether, dried at 100° and weighed.
They consist of cinchonidine, cinchonine and a little quinine, and
should not weigh more than 0*12 grm.

Test for Quinidine. — One grm. is dissolved in 30 c.c. of boiling
water, the solution is cooled and weighed. Solution of potassium
iodide and a little alcohol are added. Any quinidine hydriodide is
collected, washed with a little water, and weighed. Not more than
the slightest trace should be obtained.

Test for Cupreine. — The recrystallized sulphate of quinine, ob-
tained in testing for cinchonidine, is shaken with 25 c.c. of ether and
6 c.c. of ammonia (specific gravity 0'959). To the separated ethereal
liquid, add the ethereal liquid and washings obtained in testing for
cinchonidine, and add 6 c.c. of a 10 per cent solution of NaOH,
adding water if any solid matter should separate. Eemove the
ethereal liquid, treat the aqueous liquid with more ether and remove
the ethereal washings. Heat the aqueous liquid to boiling and
neutralize with dilute sulphuric acid when cold, collect any sulphate
of cupreine that may have separated, on a tared filter. Only the
slightest traces should be found.

Test for Cinchonine and Amorjjhous Alkaloid. — Dissolve 1 grm. of
quinine sulphate in 30 c.c. of boiling water, and add 1 grm. of sodium
potassium tartrate. Allow to cool, w.th frequent stirring ; filter.
The filtrate when evaporated to a small bulk should give little or no
precipitate with solution of ammonia.

The Detection and Determination of Quinine. — One of the most de-
finite indications of the presence of quinine is its fluorescence in a
dilute sulphuric acid solution. This property is impaired by the
presence of chlorides and other salts, but is under most conditions
still observable in the presence of excess of sulphuric acid. In the
presence of other organic matter, quinine is extracted by rendering the
mass alkaline with potassium carbonate and thoroughly extracting
with ether. The ethereal liquid is extracted with dilute sulphuric acid,
and in the presence of quinine, this will show a marked blue-violet
fluorescence and have a typically bitter taste. The presence of quinine
may be confirmed by the thalleoquin reaction (see quinine sulphate,
p. 532), but it must be remembered that other cinchona alkaloids
yield this reaction. It may also be confirmed by Herapath's reaction
(p. 532).

If the alkaloid be extracted by ether from an alkaline mass, the
ether washed with water, evaporated, and the residue weighed, an ap-
proximate determination results, or the residue may be dissolved in
excess of decinormal sulphuric acid and the excess of acid remaining
may be titrated with decinormal soda, each c.c. used being equivalent
to 0*0324 grm. of anhydrous quinine.

534

FOOD AND DRUGS.

Vigneron's process (p. 529) is applicable to the determination of
quinine, when the alkaloids present in the substance to be examined
have been extracted by ether from the mass previously rendered alkaline.

Iron and Quinine Citrate is an official preparation. It occurs as
brownish-green scales soluble in water. It should not contain more
than 1 1 per cent of water and should yield an ash which is not alkaline
to litmus; the official standard is that it should contain 15 per cent of
quinine, as determined by rendering an aqueous solution alkaline with
ammonia and extracting with ether, and drying the residue at 120°.
The ash value is from 18 to 20 per cent which should consist almost
entirely of Fe203.

Liquid Extract of Cinchona. — This official preparation is a liquid
extract made by percolating the powdered bark with a mixture of
water and glycerin with a little hydrochloric acid, and adding alcohol
to the concentrated percolate. The official standard is that it should
contain 5 grms. of alkaloids per 100 c.c, when assayed in the follow-
ing manner : —

Five c.c. and 25 c.c. of water are well shaken in a separator with 30
c.c. of a mixture of 3 volumes of benzol and 1 volume of amyl alcohol,
and 15 c.c. of a 10 per cent solution of potash. Run off the lower aqueous
layer, and well wash this with 30 c.c. of the same solvent, and mix
the two portions of solvent. Wash with water and then shake well
with 30 c.c. of 2 per cent HCl, separate the acid liquid and repeat the
extraction. Mix the acid liquids. Render alkaline with ammonia and
extract three times with 10 c.c. of chloroform. Evaporate the chloro-
form and dry the residue at 110° and weigh. A genuine liquid ex-
tract of cinchona should have the following characters : —

Specific gravity 1-115 to 1-180

Solid residue 38 „ 43 per cent

Alcohol by volume 11 „ 13 „

There are four tinctures of cinchona or quinine official in the
British Pharmacopoeia, which should have the following characters : —

Solid
Residue.

Alcohol

Quinine

Specific Gravity.

by
Volume.

Gr. per
100 c.c.

Per cent

Tincture of cinchona .

0-914 to 0-924

6-2 to 6-9

0-95 to 1-061

Compound tincture of

cinchona .

0-914 „ 0-924

4-6 „ 5-2

0-45 „ 0-551

Tincture of quinine

0-885 „ 0-893

3-5 „ 3-9

1-6342

Ammoniated tincture of

quinine

0-925 „ 0-930

1-8

1-4713

1 Official standards : to be determined as described under liquid extract of
cinchona.

^ In the form of hydrochloride, which should be present to the extent of 2 grms.
per 100 c.c.

3 As quinine sulphate (2 grms. per 100 c.c). NH3 should be present to the
extent of about 1 per cent to 1-03 per cent.

COCA. 535

Qiiinine Wine. — This galenical is official and is a solution of 20
grains of quinine hydrochloride in 20 fluid ounces of orange wine.
The official requirements are to be deduced from the directions given
for its preparation. It should have the following characters : —

Alcohol by volume . . . . 10 to 12 per cent
Quinine as alkaloid . . . -187 grm. per 100 c.c.

It should contain no salicylic acid, which is detected by acidifying
with sulphuric acid, distilling off the alcohol, and then extracting the
aqueous distillate with ether and testing the ether residue with ferric
chloride, when no violet colour should result.

COCA.

The dried leaves of Erythroxylon Coca (and its varieties) are official
in the Pharmacopoeia, and also a liquid extract, but no official standards
are given.

The principal constituent of the leaves is the alkaloid cocaine
Ci7HyiN04, which is described below ; this alkaloid is methyl-benzoyl-
ecgonine, and is associated in the leaves with the bases truxilline (also
known as cocamine) CggH^gNaOg which is iso-atropyl-cocaine ; methyl-
cinnamyl-ecgonJne CjciHggNO^, and tropococaine Ci-H^gNOg. Other
alkaloids are present in traces, nearly all of them being derivatives of
ecgonine CgH^^NOg. The leaves contain from 0*2 per cent to 1*1 per
cent of alkaloids, the Peruvian variety, known as Truxillo leaves,
usually containing more alkaloid than the Bolivian leaves, but only
about 50 per cent of the total is cocaine, whereas the alkaloids of
the Bolivian leaves contain about 80 per cent of cocaine.

Coca leaves contain from 6 per cent to 8 per cent of mineral
matter. The only determination necessary with this drug, which is
practically always met with in the whole condition, is the alkaloidal
value.

A microscopic examination, however, may be made of the powder
when necessary.

Prismatic crystals of calcium oxalate are present, and characteristic
papillose cells on the lower epidermis. Sclerenchymatous fibres, and
pitted and spiral vessels, are present in numbers. The sketch on page
536 represents powdered coca leaves.

The cocaine may be determined in coca leaves by one of the
following processes : —

Pfeiffer digests 100 grms. of the powdered leaves with 400 c.c. of

water, 50 c.c. of a 10 per cent solution of soda, and 250 c.c. of light

petroleum. The whole is kept at a temperature of about 40° C. for three

to four hours with occasional shaking, and then strained and the residues

pressed. There is no fear of the petroleum being retained by the leaves to

any extent, as it is sharply separated as an oily layer on the watery

solution. The aqueous layer is run off and the oily layer titrated with

— hydrochloric acid, of which 1 c.c. is equivalent to 0*0303 grm.

Methyl orange, or tincture of Cochineal, may be used as indicator.

536

FOOD AND DKUGS.

Lyons recommends that the finely powdered leaves should be
macerated for twenty-four hours with eight times their weight of a
mixture of 95 volumes of ether and 5 of ammonia. From an aliquot
part of this liquid the alkaloid is extracted by agitation with acidulated
water, the ether separated and the alkaloid liberated from the aqueous
liquid by means of ammonia and again extracted with ether, which is
then evaporated and the cocaine weighed. The other alkaloids are
soluble in water, but insoluble in ether, so do not interfere with the
determination of the cocaine. Gunn prefers to moisten 5 grms. of the

Fig. 48. —Powdered coca leaves x 240. cr, prismatic crystals of calcium oxalate ;
ei, lower epidermis, with surface view of papillose cells (pr) ; e'i', lower epi-
dermis in section ; /, sclerenchymatous fibres ; ffv. fragments of vessels from
midrib ; Z, bast ; me, spongy parenchyma ; ^a, i^'a' , palisade cells ; st, sto-
mata, with two subsidiary cells parallel to the ostiole ; tc, crystal cells ; tf,
cortical tissue of midrib ; tr, vessels, etc. (Greenish & Collin.)

By permission of the Editor of the *' Pharmaceutical Journal ".

powdered leaves with dilute ammonia and after allowing them to
stand for thirty minutes to proceed as follows : —

They are then placed in a narrow tubular percolator (10 inches
long and of J-inch bore) and percolated with ammoniated ether until
100 c.c. are collected. This is shaken out with three washings by a
2 per cent solution of hydrochloric acid, collecting about 50 c.c. of
the washings. This acid solution is now washed once with ether, then
made alkaline with ammonia, and the alkaloid shaken out with three
washings of ether. The collected portions of ether are transferred to
a weighed porcelain dish, the ether blown off, and the residue dried at
75° C.

When the bases of coca leaves have been extracted with alcohol
(as is the case with much crude cocaine) the cocaine may be determined

COCA. 537

by dissolving the mixed alkaloids in the minimum quantity of dilute
hydrochloric acid and then using the process of Garsed and Collie
(" Proc. Chem. Soc." xvii. 89). Advantage is taken of the fact that
cocaine forms a very stable insoluble di-iodohydriodide Cj-H.^jNO^HIIg,
so that by adding an excess of decinormal iodine to a solution con-
taining a salt of cocaine, and then titrating the excess of iodine in the
usual manner, the amount of cocaine may be determined, or the di-iodo-
compound may be collected and weighed. Ecgonine does not inter-
fere with the results since it forms soluble iodo-compounds. Benzoyl-
€cgonine, however, interferes, and should be removed by treating the
liberated bases with petroleum ether, or ether in which only cocaine
is soluble.

Liquid Extract of Coca is an extract of the drug by 60 per cent
alcohol, of such strength that 1 fluid ounce of the extract contains the
soluble matter of 1 ounce of the drug.

A properly prepared extract should have the following char-
acters : —

Specific gravity = 0-995 to 1-031
Solid residue = 18 „ 20 grms. per 100 o.c.

Alcohol by volume = 49 „ 52 per cent

Cocaine = 0-2 „ 0-6 „

This preparation should be of certain alkaloidal strength, and the
fact that the leaves contain so variable an amount of cocaine renders
it probable that the next edition of the British Pharmacopoeia will,
as in many other cases, require the extract to contain a definite pro-
portion of cocaine.

For the determination of the cocaine the following process may be
employed, which is due to Garsed : —

One hundred c.c. are evaporated to 50 c.c. in a shallow dish on a
wa f r bath, at a temperature never exceeding 80° C. with constant
stirring, to remove alcohol. When cold, the extract is made alkaline
by the addition of 5 c.c. of 10 per cent ammonia, and transferred to a
•separating funnel. The dish is washed out first with 45 c.c. of distilled
water, then with 100 c.c. of ether. The water and ether washings
are added to the rest in the separator, the whole well shaken and
allowed to stand until the ether separates, when the alkaline liquid is
■drawn off. The extraction with ether is three times repeated. Four
ether solutions are thus obtained. The first three are mixed together,
washed with a few c.c. of water containing a little ammonia, and
•shaken out first with 5 c.c. of 5 per cent sulphuric acid, then twice
with 5 c.c. of 1 per cent acid. This is sufficient to completely exhaust
the ether solution, the test being the addition of a few drops of
Mayer's reagent to the last few drops of the third quantity of acid,
w^hen, as a rule, no precipitate or opalescence is produced. The three
acid solutions are mixed together, made alkaline by the addition of
10 per cent ammonia, and three times shaken out with 10 c.c. of pet-
roleum ether, the bulked petroleum-ether extract evaporated to dry-
ness on a water bath in a tared dish, then placed in a desiccator for
some hours, and finally weighed.

538 FOOD AND DRUGS.

Cocaine. — Both cocaine and its hydrochloride are official in the
Pharmacopoeia. The following are the official requirements for the
alkaloid and its salt : —

Cocaine. — The alkaloid should melt at 96° to 98° C. It is almost
insoluble in water, insoluble in glycerine, soluble in 10 parts of 90 per
cent alcohol, in 4 parts of ether, in 0*5 part of chloroform, in 12
parts of olive oil and in 14 parts of turpentine. Its solution in dilute
nitric acid should give no reactions for sulphates or chlorides, and its
solution in dilute hydrochloric acid, when evaporated to dryness,
should give the reactions described under the hydrochloride.

Cocaine Hydrochloride. — The British Pharmacopoeia requires that
this should melt at 180° to 186° C. The U.S.P. gives 193° as
the melting-point of the pure substance. It is soluble in half its
weight of cold water, forming a clear colourless solution of neutral
reaction ; and in four times its weight of 90 per cent alcohol, or
glycerin. It is insoluble in olive oil and nearly so in ether. It
gives a yellow precipitate with auric chloride solution, and a white
precipitate with ammonium carbonate or borax solutions. It dissolves
without colour in cold HgSO^ or HNOg, but chars with hot sulphuric
acid, yielding a sublimate of benzoic acid. An aqueous solution yields
a white precipitate with potassium hydroxide solution, which is
soluble in alcohol and ether ; and a yellow precipitate with solutions
of picric acid ; with solutions of HgCl.2 it gives a white precipitate
in solutions slightly acidified with HCl, which is soluble in hot
water. If a fragment be moistened with HNOg, evaporated to dry-
ness and a drop of alcoholic potash solution added, a characteristic
odour recalling that of peppermint is evolved. A solution of not less
than 1 per cent strength gives, with excess of potassium perman-
ganate, a copious red precipitate which does not change colour within
an hour (absence of cinnamyl-cocaine {so called), and cocamine,
etc.) If 0*1 grm. be dissolved in 100 c.c. of water and 0*25 c.c. of a
solution of ammonia (10 per cent) be added, a clear solution should
result, from which a cirystalline deposit should gradually separate on
stirring. It should contain no sulphates. Dried at 100° it should
not lose more than 1 per cent of moisture, and it should contain no
mineral matter.

Pure cocaine melts at 98° and is laevorotatory, the specific rotation
in chloroform solution being about - 16°, whilst that of the hydro-
chloride in alcoholic solution is about - 70°.

. Synthetic cocaine, that is, cocaine made by hydrolysing allied
alkaloids of little or no therapeutic value, which yield Z-ecgonine, is
made by some manufacturers. This is benzoylated and methylated,
and the resulting product is identical with natural cocaine. A
synthetic, optically inactive, cocaine is made in a similar manner
from inactive synthetic ecgonine. Cocaine has the constitution of a
methyl-lasvo-benzoyltropine carboxylate. It gives a rose-coloured
precipitate with a solution of iodine in iodide of potassium, or if the
solution be strong, the precipitate is brown. If a drop of ferric
chloride be added to a solution of cocaine and the liquid boiled, an
intense red colour is developed. The usual alkaloidal precipitants

COLCHICUM.

539

yield precipitates with cocaine, phosphomolybdic acid being one of
the most delicate reagents with this alkaloid.

For the distinctions between Cocaine and similar local anaesthetics
see Hawkin ("Analyst," xxxvi. 2).

The Examination of Cocaine. — Cocaine can readily be obtained in
a state of great purity. In addition to the official test given above,
cocaine should comply with the following requirements, the alkaloid
being usually examined as the hydrochloride. The optical activity of
the hydrochloride should be taken in dilute alcoholic solution, and at
20° its specific rotation (see under sugars) should, in 2 per cent
aqueous solution, be - 71°, or, in 40 per cent alcohol - 69° (Antrich,
" Berichte," xx. 310).

The specific rotation - 52° which is frequently given in text-books,
is erroneous, and arises from a misinterpretation of Antrich's
equation.

If 0*1 grm. of the hydrochloride be dissolved in 5 c.c. of water,
and 3 drops of dilute H2SO4 be added, and then 1 drop of a 1 per
cent solution of potassium permanganate added, the liquid, kept in a
closed vessel, should only slightly decrease in colour in thirty minutes.

Maclagan proposed the following test : One grain of the salt is
dissolved in two ounces of water, two drops of strong ammonia solu-
tion ('880 specific gravity) are added and the walls of the vessel rubbed
from time to time with a glass rod : in fifteen minutes a good crop of
glistening crystals separate. If the cocaine be not very pure either no
crystals appear, or at most only a slight crop. If more than four per
cent of amorphous alkaloid be present, the liquid becomes milky.

B. H. Paul (" Pharm. Jour." 3, xviii. 783) has improved this test.
He adds ammonia gradually with constant stirring to a 2 per cent
solution of the salt, as long as a crystalline precipitate forms and the
liquid clears quickly. Directly clots begin to be precipitated the crys-
talline precipitate is filtered off and the amorphous precipitate pro-
duced by adding more ammonia is collected separately. Calculated
in the dry salt, the crystalline precipitate should weigh not less than
82 per cent to 84 per cent.

Cocaine and its hydrochloride should not contain more than 1 per
cent of moisture.

COLCHICUM.

The corms of Colchicum autumnale, as well as the seeds, are official.
An extract of the fresh corms, a wine prepared from the corms, and
a tincture of the seeds are all official, but no standards are given for
any of them.

The active constituent of both the corms and the seeds is the toxic
alkaloid colchicine C22H25NOg, which is present in the corms to the ex-
tent of 0*4 per cent to 0*65 per cent and in the seeds to the extent of
0-6 per cent to 0*8 per cent.

Colchicum seeds contain from 4 per cent to 5 per cent of mineral
matter, and the corms from 2 per cent to 3 per cent.

Assay of Colchicum Seeds. — Farr and Wright's process (" Pharm.
Jour." Vol Lxxxv. 1910, p. 579). Pack 5 grms. of the seeds in No.

540 FOOD AND DKUGS.

30 powder in a glass tube about 2 cm. diameter and exhaust by slow
percolation with 50 per cent alcohol. Transfer the percolate to a porce-
lain dish add 25 c.c. water and evaporate to about 20 c.c. over a water
bath. Transfer to a separator, rinsing the dish first with a little water
and then with 25 c.c. petroleum ether, add the rinsings to the separator
and shake vigorously. When the liquids have separated reject the
upper layer, return the residual liquid to the separator, and twice re-
peat the washing with 20 c.c. of petroleum ether. Saturate the aqueous
liquid with sodium chloride and shake vigorously with 20 c.c. of chloro-
form. Twice repeat the extraction with 10 c.c. of chloroform and mix
the chloroform solutions. Eecover the chloroform and treat the residue
first with a mixture of 19 c.c. water and 1 c.c. solution of ammonia,
used in four portions and then with a mixture of 16 c.c. water and 4
c.c. diluted sulphuric acid. Strain the solutions through cotton wool
into a flask, shake, add 20 c.c. of decinormal solution of iodine, set
aside for 5 minutes, and collect the precipitate on a small filter, washing
the flask and precipitate with 20 c.c. of distilled water containing 1 c.c.
decinormal iodine and 1 c.c. of dilute sulphuric acid. Drain the filter,
then place it in a small mortar with 2 c.c. of sodium carbonate test
solution and 20 c.c. of decinormal solution of sodium thiosulphate until
the filter has been reduced to a pulp. Filter the mixture through
cotton wool into a separator, rmse the mortar with several small
portions of distilled water until a few drops of filtrate acidulated with
dilute sulphuric acid cease to give a precipitate with a few drops of
iodine solution.

Shake the liquid in the separator vigorously for 1 minute with 20
c.c. of chloroform, and draw off the chloroform into a tared platinum
dish, repeat the process twice with 10 c.c. chloroform and evaporate
the chloroform solutions to dryness at a low temperature. Dissolve
the alkaloids in a little 90 per cent alcohol, evaporate over a water
bath and dry at 100° to constant weight.

The alkaloid obtained by this process is a very pale straw colour
perfectly soluble in water. Dissolved in chloroform and poured into
excess of petroleum ether, the alkaloid is precipitated quantitatively in
a nearly colourless condition.

For the determination of Colchicine, Lyons (" American Druggist
and Ph. Eecord," Feb. 1909) uses the following gravimetric method
which gives accurate results : —

Place in a small beaker 25 grms. of colchicum corm in moderately
fine powder. Add 15 c.c. of solution of lead subacetate, and 80 c.c.
of warm distilled water, and macerate vnth occasional stirring for six
hours at a temperature of about 50"^ C. Transfer to a small percolator
(a funnel answers the purpose well), having the tube so packed that
percolation will go on at the rate of about 2 c.c. per minute. When
the fluid has disappeared from the surface of the drug add warm
water, about 20 c.c at a time, and so continue the percolation until
250 c.c. of fluid has been collected. This should practically exhaust the
drug. To the percolate add 5 grms. of powdered sodium phosphate,
or enough to precipitate the whole of the lead present in the percolate.
Filter, returning the first portion of filtrate if it is not quite cjear.

COLOCYNTH. 541

Use for duplicate assays two portions, 100 c.c. each of the filtrate,
representing 10 grms. of drug. Shake out each portion with three
successive portions of chloroform, 25, 20, and 15 c c. or enough to
extract the whole of the colchicine. Evaporate off the chloroform
and treat the residual alkaloid repeatedly with 90 per cent alcoho to re-
move persistently adhering traces of chlorororm, dry at a temperature
below 100° C. to constant weight. The weight of the alkaloid in grms.
multiplied by ten gives the percentage of colchicine in the drug. This
may be confirmed by titration with Mayer's reagent. The alkaloid
will be found to be remarkably free from impurities.

The more recent method is that of Farr and Wright (p. 539).

Tincture of Colchicum is a 45 per cent alcohol extract of the seeds,
of which A ounces are used for 1 pint and should have the following
characters : —

Specific gravity = 0-950 to 0-960

Solid residue = 1-90 „ 2-4 grms. per 100 c.c.

Alcohol by volume = 41 „ 43 per cent

Akaloids = 0-05 „ 0-09

Vinitm Colchici. — This preparation is made by macerating 4
ounces of colchicum corms in powder in 1 pint of sherry. It should
contain from 14 per cent to 15 per cent of alcohol by volume, and
should be free from salicylic acid (see quinine wine, p. 535).

Colchicine. — The alkaloid colchicine C22H25NOg, and its salicylate
C.>2H25NO,j . CjHgOg are met with in medicine and are frequent consti-
tuents of gout remedies.

Colchicine is an amorphous powder, soluble in water, alcohol, and
chloroform. It melts at 145°. A minute quantity dissolved in
sulphuric acid, gives with nitric acid a rich greenish-blue colour,
which changes to pale blue and then to red and yellow. The yellow
solution is turned red by caustic soda solution. Nitric acid gives a
dirty violet colour, passing to greenish and then to yellow. An alco-
holic solution gives a garnet red colour with ferric chloride. In
organic mixtures suspected of containing colchicine it is easily preci-
pitated by phosphomolybdic acid, and the resulting precipitate may be
treated with ammonia and the free alkaloid extracted by chloroform,
when it will give the foregoing reactions.

COLOCYNTH.

The dried pulp of the fruit of Gitrullus colocynthis, free from the
seeds, is the official drug of the British Pharmacopoeia.

The ofi&cial standards are as follows : It should not yield a reaction
for starch, and should only yield traces of fixed oil to ether. It should
yield not less than 9 per cent of ash.

Numerous bodies have been described as active principles of this
drug, but the recent researches of Power and Moore (" Chemist and
Druggist," 1910, I. 150), have corrected many of the erroneous earlier
statements. These chemists have isolated a dihydric alcohol C22H3^02
(0H)2 which they have termed citrullol (apparently a homologue of
ipuranol). A very small quantity of an alkaloid was obtained, which

542

FOOD AND DKUGS.

neither crystallizes nor yields crystalline salts, but which has
a powerful physiological action. Traces of a glucoside are also
present and about 1 per cent of a-elaterin. The results of this in-
vestigation have established the fact that the so-called " colocynthin,"
'' colocynthitin," and other products heretofore obtained from colocynth
to which specific names have been attached, are not pure substances,
but very indefinite mixtures, and that the amount of glucosidic sub-
stance present is extremely small. On the other hand, it has been
shown that the activity of colocynth is due to at least two principles,
one of which is alkaloidal, although a very weak base, whilst the other
source of activity is represented by some non-basic principle or

Fig. 49. — Powdered colocynth.

principles contained in the ether and chloroform extracts of the resin,
but which did not permit of being more definitely characterized. The
colocynth contains, furthermore, a quantity of a-elaterin, but no evi-
dence could be obtained of the presence of y8-elaterin, which is the
physiologically active constituent of the fruit of Echallium elaterium.
According to the British Pharmaceutical Codex, the ash should
vary between 7 per cent and 13 per cent, but it is generally maintained
that 10 per cent to 13 per cent is the better standard. Not more than
1*5 per cent of fixed oil should be extracted by petroleum ether.

CONIUM. 543

The indefinite nature of the active principles of this drug render
any process of assay of doubtful value. The following process, however,
due to Brautigam (** Journ. Pharm. Chim," 6, 16, 130) has in the
author's hands given concordant results on a number of samples ex-
amined. Three grms. of the pulp are thoroughly exhausted with
water, and the water evaporated. The residue is then extracted with
two successive quantities, each of 30 c.c. of 90 per cent alcohol, for
one hour at 20" to 25° C, with frequent agitation. The residue is
washed with 20 c.c. of alcohol. The bulked alcoholic solution is
filtered and evaporated to dryness. The residue is triturated with
water, made up to about 120 c.c. and left in contact for twenty hours
at 25° C, with frequent and thorough agitation. The mixture is then
filtered, the filter washed with 20 c.c. of water, and then O'iiS grm.
of lead acetate is dissolved therein, and 3 grms. of 5 per cent solution
of basic lead acetate added. When precipitation is complete, the pre-
cipitate is filtered off and washed with two portions, each of 30 c.c.
of water. To the filtrate, aluminium sulphate 2 grms., and animal
charcoal 4 grms., are added, and the mixture is evaporated to dry-
ness. The residue is taken up with two successive 30 c.c. of ether,
and the ethereal extract evaporated. The residue is macerated
twice in succession, each time for one hour, with alcohol 40 c.c, and
finally washed with another 30 c.c. The bulked alcoholic solutions are
filtered and evaporated to dryness. The residue is taken up with a
little absolute alcohol, and filtered through a small filter, previously
moistened with alcohol, the filtration being repeated until the liquid
is quite bright. The filter is washed with a little absolute alcohol ;
the bulked liquids are evaporated in a small tared capsule. The
residue is dried to constant weight, and weighed. It should not be
less than 0-04 grm., and should be completely soluble in 2 c.c. of
absolute alcohol. On adding 2 drops of this solution to 4 c.c. of ether,
a flocculent white precipitate should be obtained ; and the same
quantity should give with 4 c.c. of water a cloudy solution which pre-
cipitates on standing. One or two drops of the " colocynthin " solution,
evaporated to dryness at a gentle heat, should give a fine red colour
when treated with HgSO^. A similar residue should give a cherry-red
colour with Frohde's reagent ; and with sulphuric acid containing
0*5 per cent of ammonium vanadate, a red colour gradually becoming
blue at the edge of the liquid, results.

CONIUM.

The fresh leaves and the dried fruits of Conium maculatum are
official drugs ; a juice prepared from the former, and a tincture from
the latter, being also official. No standards are given for either the
drugs or their preparations.

The active constituent of the leaves is the alkaloid coniine
CgHjyN with a certain amount of subsidiary compounds. The
amount of alkaloid present, however, rarely exceeds 0*25 per cent,
whilst the dried fruits in their best condition contain as much as 3 to
3'5 per cent of coniine. As found in commerce, however, the fruits

544

FOOD AND DEUGS.

rarely contain more than 1 per cent of alkaloid, owing to the fact
that they are collected after ripening has commenced.

This is shown in the following table, due to Farr and Wright : —

Hydrochloeates of Mixed Alkaloids per Cent.

1892.

1893.

Fresh.

Dried.

Fresh.

Dried.

Immature, ^ to i grown .

•896

3-00

frtol „ . . .

•975

^toj „

1-049

3-32

Nearly mature, f to full grown .

•935

—

Mature, J to full grown .

—

1-088

3-86

Mature, a few outer ones beginning

to turn slightly yellow .

—

1-049

Mature, yellowish-green to yellow .

•475

—

Mature, yellow ....

•434

1-44

—

Ripe, grey

—

1^32

—

The amount of moisture in the fresh fruit varies from about 6Q
per cent in the older stages to about 68 per cent in the younger, but
is not a constant proportion.

The amount of ash yielded by conium leaves varies from 12 per
cent to 15 per cent ; whilst that of the seeds lies between the limits
5 per cent and 7 per cent.

A good commercial sample of conium seeds will contain from 0-5
per cent to 1*1 per cent of alkaloid, whereas the leaves rarely contain
more than 0'2 per cent.

The best method for the determination of conium alkaloids is that
of Cripps.

Cripps (" Pharm. Journ." 3, xviii. 13, 511) exhausts 5 grms. of
the finely powdered fruit (20 grms. of the leaves should be used)
mixed with sand, by a mixture of absolute alcohol (25 c.c), chloroform
(15 c.c), and chloroform saturated with dry HCl gas (10 c.c). After
complete exhaustion the liquid is shaken with 25 c.c. of water twice,,
the mixed aqueous liquids, containing the alkaloids as hydrochlorides^
being then once washed with chloroform, rendered alkaline with soda,
and extracted three times with chloroform. The chloroform is washed
with a little slightly alkaline water and then run into an ethereal solu-
tion of dry HCl gas. The solvent is evaporated in a current of warm
air, and the residue dried at a temperature not exceeding 90° C. The
hydrochlorides of this alkaloid should be crystalline and practically
white. 163*5 parts of hydrochloride contain 127 of coniine (the small
amounts of other alkaloids do not materially interfere with this ratio).

This process is improved by titrating the free alkaloid which is in

the chloroform after its liberation by means of alkali (and after the-

chloroform has been washed) with decinormal HCl and methyl orange,.

until after shaking well the pink colour does not disappear. Each c.c oF

— HCl is equivalent to 0*0127 grm. of coniine.

DIGITALIS. 545

Tine Mire of Conium is a 70 per cent alcohol extract of the fruits,
four ounces of the drug producing one pint of tincture. A properly
prepared tincture should have the following characters : —

Specific gravity .... 0-895 to 0-902

Solid residue 1-3 „ 1-45 grms. per^lOO c.c.

Alcohol, by volume .... 66 „ 68 per cent

Alkaloids as coniine . . . 0-05 ,, 0-1 „

Farr and Wright (" Pharm. Jour." 3, xxi. 857^ assay the tincture by
evaporating 50 c.c. with 1 c.c. of normal sulphuric acid, down to a low
bulk, and shaking the liquid twice with chloroform. It is then
rendered alkaline, and the free alkaloids extracted with chloroform
three times. The chloroform is freed from traces of alkali by washing
with water, separated and run into a solution of dry HCl in chloro-
form. The solvent is then evaporated and the hydrochloride weighed
as in Cripps' process described above.

DIGITALIS.

The leaves of Digitalis purpurea are oflBcial in the British Pharma-
copoeia, as well as a tincture, but no standards are given for either.
The principal constituents of this drug are glucosides. Much contro-
versy has taken place in reference to the chemistry of these bodies, but
the following appears to be now settled.

Digitalin 035115^014, is the principal constituent of the commercial
" digitalin " which is generally very impure. It forms fine crystals,
but is generally obtained as an amorphous powder. On hydrolysis it
yields digitaligenin CoHg^^Og, glucose, and digitalose, a sugar of the for-
mula C7H14O5.

It yields a somewhat characteristic reaction w^hen dissolved in a
minute quantity of concentrated sulphuric acid and a drop of a solu-
tion of potassium hypobromite added : a fine rose or violet-red results.
Sulphuric acid containing a trace of ferric sulphate gives at first a
yellow colour, changing to red and then to violet-red, which is fairly
permanent. It melts at about 217°.

Digitoxin Cg^Hg^Ou, forms colourless crystals, generally very
small. It crystallizes from methyl alcohol and. chloroform in an an-
hydrous condition, but from diluted ethylalcohol, with 5 molecules
of water of crystallization. The latter form melts at 145°, the an-
hydrous variety commencing to melt somewhat indefinitely at 240°.
On hydrolysis it yields digitoxigenin CggHg^O^ and digitoxose
C^Hi204. If a few milligrams are dissolved in acetic acid and a drop
of dilute solution of ferric chloride be added, and then concentrated
H2SO4 be poured down the side of the tube, so as to form a layer under
the acetic acid a brownish-green band appears, altering quickly so that
the top layer of the sulphuric acid is coloured browmish-red and above
this is a broad bluish-green band, which soon becomes indigo blue.
After a long time green again appears, and finally fades to a brownish
colour.

Digitoxin is the principal glucoside present in the leaves, and is the
most reliable preparation to use.
VOL. I. 35

546

FOOD AND DKUGS.

Commercial "digitalin" is frequently a mixture of true digitalin
and digitoxin. According to Merck the following are the characters
of the principal commercial " digitalins : —

(1) German digitalin ; consists principally of digitalein with some
digitoxin and digitalin. It is freely soluble in alcohol, but insoluble
in ether and chloroform.

(2) Nativelle's crystallized digitalin. This consists almost entirely
of digitoxin. It forms fine white needles, insoluble in water, ether, or
petroleum ether. It is the type of " French digitalin ".

Fig. 50. — Powdered digitalis leaves x 240. co, coUenchymatous cells of the mid-
rib ; ei, lower epidermis, cells with sinuous walls ; en, neural epidermis ; es,
upper epidermis ; ip, scar of fallen hair ; Z, bast ; me, spongy parenchymatous ;
jpa, ^'a', palisade cMls ; 'pg, glandular hairs ; pi, simple hairs ; si, stomata ;
%j, cortical tissue of midrib ; tr, v, vessels, etc. (Greenish & Collin.)
By permission of the Editor of the " Pharmaceutical Journal ".

(3) Hommolle's amorphous digitalin. A white or yellowish-white
powder, slightly soluble in water and ether, freely soluble in 90 per
cent alcohol and in chloroform. It consists principally of digitalin
with some digitoxin and corresponds with " French amorphous digi-
talin ".

(4) Pure digitalin Merck. A yellowish-white powder correspond-
ing with No. 1.

A microscopic examination of the leaves in powder shows a lower
epidermal tissue with sinuous walls, many single hairs, spiral and
pitted vessels and numerous glandular hairs. The above illustra-
tion represents the powdered leaves.

DIGITALIS. 547

Digitalis should not contain more than 10 per cent or at most 11
per cent of ash. The usual amount is 8 per cent to 10 per cent.

The leaves may be assayed as described under the tincture, being
first extracted with alcohol. There are, however, advocates of a
physiological standardization, on account of the difiBculty of deciding
the relative proportions and activities of so many nearly related
bodies as are present in this drug.

Digitoxin is, as has been mentioned, the principal of these bodies,
and Keller (" Ber. Deutsch. Pharm. Ges." 1897, 7, 125) estimates this
in the following manner (slightly modified by Barger and Shaw) : —

Twenty grms. of leaves, or 146 grms. of tincture, are employed.
The British Pharmacopoeia directs that 125 grms. of leaves should be
percolated with sufficient 60 per cent alcohol to produce 1000 c.c. of
tincture. The density of this alcohol is 0*913, hence 1 grm. of leaves

is percolated with =-p^ = 7*3 grms. of alcohol. In order to

have quantities equivalent to Keller's 20 grms. of leaves, 146 grms.
of tincture may be used for an estimation. The 146 grms. of tincture
are evaporated on a water bath to 25 c.c. or less to remove the
alcohol, and made up with water to 222 grms. Here the first diffi-
culty presents itself, for by the evaporation of the alcohol a resin
separates out, which is mostly insoluble in water, and any digitoxin
which may be contained in it will escape estimation. The water is
best added in small quantities, and the dish containing the resin
warmed on the water bath, while its contents are stirred vigorously,
so as to have it suspended in as fine a condition as possible.

To the 222 grms. of turbid solution 25 grms. of a saturated basic
lead acetate solution are added, and the precipitate filtered off.

In some cases rather more lead solution is necessary.

A voluminous precipitate results, which is filtered off.

As the total bulk now weighs 247 grms. of which, as experiment-
ally shown, 7 grms. are precipitated, there remains 240 grms. of solu-
tion. Of this 132 grms. are easily obtained in the clear filtered condition
and represent 11 grms. of leaves. Five grms. of sodium sulphate
dissolved in 6 grms. of dilute sulphuric acid are now added. The lead
sulphate is filtered off and 130 grms. of the filtrate ( = 10 grms. of
leaves) are rendered alkaline with 2 c.c. of 10 per cent ammonia ; the
solution remains perfectly clear, and is shaken out four times with 30
c.c. of chloroform. The chloroform extract is filtered, evaporated to a
small bulk, and then washed into a small wide weighing bottle with
ground stopper, in which it is evaporated to dryness, first on the water
bath, then in the steam oven, till of constant weight. The residue is
" crude digitoxin ". Keller purifies this by dissolving it again in chloro-
form, adding ether and petroleum ether, and collecting the precipitate
on a small filter, from which it is dissolved again by hot absolute
alcohol, after it has been washed with petroleum ether. Barger
and Shaw prefer to place the chloroform solution in a tall 50 c.c. or
100 c.c. stoppered measuring cylinder, in which the digitoxin is pre-
cipitated and allowed to settle overnight. The following day the clear
liquid, containing impurities, is decanted, and the digitoxin washed

548 FOOD AND DRUGS.

by shaking it with a further quantity of petroleum ether. This is
decanted, and finally the digitoxin, mixed with some petroleum ether,
is dissolved in hot absolute alcohol. The solution is washed into a
weighing bottle and evaporated ; dry ether is added and evaporated
off, and then the substance, " pure digitoxin Keller," is weighed.
Almost two-thirds of Keller's " digitoxin " is really digitoxin.

The results to be obtained from well-prepared tinctures vary from
0*45 to 0-75 per cent of crude digitoxin, the leaves containing from
0'06 to 0*1 per cent as assayed by this method.

It is probable that this method does not give really correct, but
only comparative results.

Tincture of Digitalis, — This is an extract of the leaves with 60
per cent alcohol. A genuine tincture should have the following
characters : —

Specific gravity 0-930 to 0-935

Solid residue 2-9 „ 3-7 per cent

Alcohol by volume 54 „ 56 ,,

Digitoxin (as estimated by Keller's process) . 0-4 „ 0*75 „

ELATERIUM.

Elaterium, the sedimentary matter from the juice of the fruit of
Echallium elaterium, is official in the Pharmacopoeia, as well as its
active principle elaterin CgoHggOg.

The official requirements for elaterium are that it should not give
any reactions for carbonates and starch. It should yield 50 per cent
to boiling alcohol. When exhausted with chloroform and the solution
evaporated, and the residue washed with ether, and the process of
solution, evaporation, and washing repeated, at least 20 per cent of
elaterin should be so obtained.

The drug occurs as light, thin, pliable pieces, and should not yield
more than 14 per cent to 15 per cent of ash. The average composi-
tion of the drug is as follows : —

Per cent

Water 10 to 12

Mineral matter 12 ,, 15

Elaterin 22 „ 30

Glucosides ........ traces

Inert matter 40 to 50

Elaterin is described in the Pharmacopoeia as being almost insol-
uble in water, sparingly soluble in 90 per cent alcohol, but readily
soluble in chloroform. It is neutral to litmus ; with melted phenol
it yields a solution which gives a crimson colour, rapidly changing to
scarlet, on the addition of sulphuric acid. It is not precipitated by
tannic acid, mercuric chloride, or platinum chloride solutions.

Pure elaterin melts at about 225° to 230°, but the commercial pro-
duct which fulfils the requirements of the Pharmacopoeia, melts at
from 208° to 215°. It is highly laevorotary, having a specific rotation
of about - 42° in chloroform solution.

Power and Moore (" Pharm. Jour." 1909, 83, 501) state that an

ERGOT. 549

English-made elaterium gave 5-3 per cent of moisture, 6*7 per cent of
mineral matter, and an inert aqueous extract of 6 per cent. The dried
insoluble matter gave 57 per cent of matter soluble in chloroform and
alcohol used successively, the insoluble matter being quite inert.
The chloroform and alcohol extracts (continued) gave the following
extracts : —

(1) With petroleum ether, 15 per cent.

(2) With ether, 73 per cent consisting chiefly of elaterin (crystal-
line) melting at 217° to 220^

This latter result indicates 30 per cent of elaterin in the elaterium.
The crude drug contains in addition to this amount of elaterin, a con-
siderable proportion of an inert crystalline compound melting at about
230° with decomposition, and having a specific rotation of over - 50°.

ERGOT.

Ergot, the sclerotium of Claviceps purjmrea, a fungus whose
spores have developed in the ovary of Secale cereale, known usually as
" ergot of rye," is official in the Pharmacopoeia. No standard is given
for it, nor for its three official preparations, extract of ergot ; liquid
extract of ergot ; and ammoniated tincture of ergot.

Numerous constituents of ergot have been described, but it appears
that many of them are impure ; the well-defined bodies to which
the drug owes its activity are two alkaloids cornutine (ergotoxine)
<^35H4iN50,. and ergotinine C35H39N5O5.

An acid which has probably not yet been separated in a state of
purity exists which is termed ergotinic acid (sclerotic acid?). For
details of the indefinite and uncertain compounds reported upon, re-
ference may be made to the following papers : Tanret (" Jour.
Pharm. Chim." 6, 24, 397) ; Barger and Carr (" Pharm. Jour." 4, 23,
257), and Jacoby ('' Chem. Central." 1897, 483 and 1059).

iirgot should contain from 3 per cent to 6 per cent of mineral
matter. Various methods have been proposed for its assay, but
most authorities hold the opinion that it must be physiologically
standardized.

Keller (" Apoth. Zeit." 22, 183) gives the following process for
estimating the active principle which he terms ergotin : —

Twenty-five grms. of the powdered ergot is freed from fat by ex-
traction with petroleum ether, and then treated with 100 c.c. of
ether ; 20 c.c. of water is added after an hour, and 1 grm. of magnesia,
and then well shaken for an hour. After standing, 80 c.c. of the
ether is separated, corresponding to 20 grms. of ergot, and this is ex-
tracted with dilute hydrochloric acid. The acid extraction is repeated
several times, the mixed acid liquids are then rendered alkaline,
and extracted with ether. The ether extraction is repeated three
times, and the mixed ethereal liquids are evaporated in a tared basin,
and the residue, consisting of fairly pure ergotin, is weighed.

Liquid Extract of Ergot is an aqueous extract of the drug, pre-
served by means of alcohol. A pure extract should have the follow
ing characters : —

550 FOOD AND DRUGS.

Specific gravity . . . 1-005 to 1-025

Solid residue . . . 12 „ 15 grms. per 100 c.c.

Alcohol by volume . . 30 ,,32 per cent

Ammoniated Tincture of Ergot. — This is an extract of the drug
made by means of a mixture of ammonia and alcohol. A genuine
preparation should have the following characters : —

Specific gravity . . . 0-935 to 0-942

Solid residue . . . 2-8 „ 4 grms. per 100 c.c.

Alcohol by volume . . 50 ,,52 per cent

On adding solution of caustic soda and distilling the liquid through
a well-cooled condenser, about 1 per cent of NHg should be obtained.
The precautions necessary in determining the alcohol are the same as
in the case of ammoniated tincture of guaiacum (p. 458).

Wood ('* Amer. J. Pharm." 1909, 81, 215) claims that the thera-
peutic activity, as determined physiologically, of extract of ergot is in
almost direct ratio to the amount of resin precipitated by water. He
recommends the estimation of this resin by diluting the liquid extract
with twice its volume of water and repeatedly extracting with
benzene, evaporating the solvent and drying the resin on a water
bath. A small amount of another active principle is not extracted
by benzene, but this amounts to only a small proportion of the
whole of the active principles. The proportion of resin so extracted
should vary from 0*48 per cent to 0*68 per cent of the liquid extract.

It is to be noted, however, that Tanret has recently isolated a
sulphur-containing base, ergothionine, C9H15O.2N3S, which is soluble in
water, which is possibly therapeutically active ; and Carr and Barger
("Journ. Chem. Soc. Trans." 1907,91, 337) and Barger (" Journ.
Chem. Soc." 1909, 1123) have isolated soluble alkaloids (ergotoxin
and p-hydroxyphenylamine) which are certainly physiologically active
substances.

GELSEMIUM.

The root and rhizome of Gelsemium nitidum are official in the
British Pharmacopoeia. No standards are given for the drug.

It contains an alkaloid gelsemine, which probably has the formula
C20H22N2O2 (o^ according to Sayre C14HJ5NO). It is a crystalline
base possessing unknown therapeutic activity. Traces of gelseminine
are also present, a powerfully toxic alkaloid which occurs to the extent
of about 0*5 per cent in the drug. It is therefore obvious that a de-
termination of the alkaloidal value of the drug or its preparations gives
little information as to the real value of the substance. Scopoletin is
also present in small amount.

Gelsemium root yields from 2 per cent to 4 per cent of ash, rarely
up to 5 per cent.

It contains from 0*3 per cent to 0-9 per cent of alkaloids as deter-
mined by the process described under tincture of gelsemium.

Tincture of Gelsemium. — This is the extract from two ounces of
the drug with 60 per cent alcohol to make one pint of tincture. It

HYDRASTIS. 561

should have the following characters (none of which are mentioned in
the Pharmacopoeia) : —

Specific gravity . . . 0-920 to 0-928

Solid residue ■ . . . 1-20 „ 1-30 grms. per 100 c.c.

Alcohol (by volume) . . 56 ,, 57-5 percent

Alkaloids .... 0-02 „ 0-03

The alkaloids may be determined as follows : —

Two hundred c.c. of the tincture (if so much is available) are eva-
porated to a thick liquid, rendered alkaline and extracted with chloro-
form three times. The chloroform is then extracted with slightly
acidulated water, the aqueous liquid rendered alkaline and the
alkaloid agam dissolved out with chloroform. The chloroform is
washed with water until the water ceases to give a pink colour with
phenol-phthalein and then evaporated in a thin porcelain dish and the
alkaloid weighed. It is preferable, however, to titrate it, by adding a
calculated excess of one-twentieth normal hydrochloric acid and titrat-
ing back with one-twentieth normal solution of barium hydrate, using
methyl orange or iodeosin as indicator. Each c.c. of one-twentieth
normal acid is equivalent to 0-0161 grm. of gelsemine, taking
C.,qH22N202 as the formula.

Gelsemine must be distinguished from the resinoid substance to
which the same name has been applied and which is a powdered alco-
holic extract of the root. Gelsemine melts at 178° and crystallizes
from acetone in needles. If a minute fragment be allowed to stand
with a drop of nitric acid, which is allowed to evaporate spontane-
ously, a permanent blue-green colour is produced. If a fragment
of gelsemine be treated with sulphuric acid and an oxidizing agent,
it behaves much like strychnine, except that the colour produced
is of a reddish-purple soon changing to blue or red-blue. Gelsemine
forms crystalline salts of which the hydrochloride is the only one
found in commerce. It is a combination of 1 molecule of alkaloid
and 1 of acid.

HYDEASTIS,

The dried roots and rhizome of Hydrastis canadensis are official
in the Pharmacopoeia, but no standards are given. A liquid extract
and a tincture of the drug are also official.

The principal constituent of the drug is the alkaloid hydrastine
C2iH2iNOg (see below), together with some berberine C2oHj7N04 and
canadine. The first-named alkaloid is present to the extent of 1*5 per
cent to 4 per cent, berberine to the extent of about 3 per cent, and
canadine only in traces.

This drug yields from 4 per cent to 10 per cent of ash on incinera-
tion.

The drug should be assayed for its alkaloidal value, it being neces-
sary to separate the hydrastine from the berberine. The best method
for this determination is Maben's (" Year-book of Pharmacy," 1901,
408). Ten grms. of finely powdered hydrastis are extracted with hot 90
per cent alcohol in a Soxhlet tube. The liquid, containing the whole of
the extractive, is made up, when cool, to 100 c.c.

552 FOOD AND DRUGS.

Place 25 c.c. of the above extract in a wide-mouthed flask of about
8 ounce capacity ; add 1^ c.c. of hydrochloric acid (32 per cent), i c.c.
of concentrated sulphuric acid, and 125 c.c. of ether. Cool, shake
well and allow the mixture to stand twenty-four hours in a refriger-
ator, and the crystals of berberine hydrochloride will separate.
Filter through a tared paper and preserve the filtrate. Wash the
crystals with a mixture of equal volumes of alcohol and ether until
the washings cease to give an acid reaction. Add the washings to
the filtrate preserved as above directed. Dry the crystals at 105° C,
and weigh. The result multiplied by 0-9017 gives the berberine.
This multiplied by 4 is equivalent to the berberine in 10 grms. of the
drug.

Render the combined filtrate and washings from the berberine
neutral or only faintly acid. Evaporate nearly to dryness on the water
bath ; treat the residue with hot water in small quantities, filtering
same into a stoppered separating funnel until the washings from the
residue cease to give an alkaloidal reaction with the ordinary reagents.
(The extraction of the alkaloid from the resinous mass left after the
evaporation of the combined filtrate and washings may be somewhat
expedited at this point by the addition of a few drops of alcohol at
each extraction with water, evaporating off the alcohol each time be-
fore the aqueous washing is poured off.) Add to the aqueous extract
in the separating funnel ammonia water to render alkaline, and ex-
tract the hydrastine by agitation with ether. Continue the extraction
with ether until the hydrastine is entirely removed ; evaporate off the
excess of ether, and re-extract the hydrastine by means of several por-
tions of 5 per cent sulphuric acid, and from the combined acid wash-
ings extract the hydrastine again by shaking with several portions of
ether, after having rendered the solution alkaline with ammonia.
Finally evaporate off the ether, dissolve the hydrastine in an excess

N N

of — acid, titrating back the excess with —-—. alkali in the usual

manner, using cochineal as an indicator. Each c.c. of alkali is equal
to 0-00383 grm. of hydrastine and this multiplied by 4, gives the
hydrastine in 10 grms. of the drug.

Gordin and A. B. Prescott (" Amer. Journ. Pharm." 1899, 518-
22) recommend the following process of assay. Ten grms. of the
powdered root are stirred into a paste with a mixture of alcohol, con-
centrated ammonia, and ether (1:1:6 parts by volume), and allowed
to remain in a closed vessel for several hours. The mixture is then
dried, at first in a current of air, and then over sulphuric acid under
diminished pressure ; the residue is transferred to a Soxhlet apparatus,
being rinsed out with powdered barium nitrate, and the hydrastine is
extracted completely with absolute ether ; the ether is evaporated
from the extract, and the residue dissolved in acidified water, and the
solution diluted to 100 c.c. In a graduated 100 c.c. flask, 20 c.c. to
30 c.c. of a standard iodine solution (of about 1 per cent strength) are
placed, 20 c.c. of the filtered hydrastine solution run in, and the mix-
ture is diluted to the mark and shaken until the pentiodide has all
separated ; the mixture is then filtered, and the excess of iodine deter-

HYDRASTIS. 553

mined by titrating 50 c.c. of the filtrate with standard sodium thiosul-
phate solution. Every 1 part of iodine used corresponds with 0-607
part of hydrastine. Or the alkaloid may be estimated gravimetrically
by shaking 20 c.c. of the filtered hydrastine solution with petroleum
ether and ammonia, removing the alkaloid from the petroleum ether
solution by shaking with acidified water, and then from the acid
solution with ammonia and ether ; the ethereal solution is finally eva-
porated at the ordinary temperature, and the residue of hydrastine
weighed.

The residue in the Soxhlet apparatus contains the berberine, which
is not soluble in absolute ether ; it is dried by passing a current of
dry air through the apparatus, and is then extracted with alcohol.
The alcohol is removed from the extract by heating it with 200 c.c.
of water on the water bath ; the residual liquid is acidified with acetic
acid, cooled, and filtered into a conical flask ; in this it is shaken for
ten minutes to fifteen minutes with 6 c.c. to 8 c.c. of acetone, and
enough 10 per cent caustic soda solution to render it alkaline, and set
aside for two hours to three hours. The precipitated acetone compound
is washed, and warmed in the same flask with 200 c.c. to 300 c.c. of
very dilute sulphuric acid until it has all dissolved, the solution is
poured into a long-necked Kjeldahl flask and boiled for one and a half

N
hours to two hours; when cold, it is added to 100 c.c. of -- potassium

iodide solution contained in a graduated 1000 c.c. flask, diluted to the

mark, shaken, and left overnight. Then 500 c.c. are filtered from the

precipitate of berberine hydriodide into another 1000 c.c. flask, treated

N .
with 50 c.c. -— - silver nitrate and nitric acid, diluted to the mark,
20

and filtered ; the excess of silver is determined by titrating 500 c.c.

N
of the filtrate with —- ammonium thiocyanate. The number of c.c.

of the iodide solution used, multiplied by 0*167125, gives the percent-
age of berberine in the root.

Schreiber's process gives good results. It is as follows (" Pharm.
Post," 36, 321) : 10 grms. of the powdered root are dried on the
water bath, the moisture being thus determined. The dry residue is
moistened with a mixture of ammonia, 5 c.c, alcohol, 5 c.c, and
ether, 30 c.c, and dried. It is then extracted in a Soxhlet with ether ;
and the ether extract shaken with 15 grms. of 5 per cent hydrochloric
acid in a graduated cylinder. The ethereal layer is decanted, the
acid extract washed with more ether to remove resinous matter, and
the ether decanted. The volume of ether over the acid liquor is then
adjusted to exactly 50 c.c. . Ten c.c. of ammonia are added and the
whole well shaken until all the precipitated alkaloid is dissolved in
the ethereal layer. After separation, 40 c.c. of this is decanted (= f
of the whole), into a tared capsule, about half the ether evaporated
off at a gentle heat, the rest allowed to evaporate spontaneously.
In this manner almost colourless crystals of hydrastine are obtained
which are finally dried to constant weight on the water bath.

554 FOOD AND DRUGS.

Matthes and Rammstedt (" Archiv der Pharm." 245, 112) have
recommended picrolonic acid (dinitro-phenyl-methyl-pyrazolone) as a
precipitant of hydrastine — and also for the pilocarpine and the mixed
alkaloids of nux vomica. The process described is interesting, but it is
neither so accurate nor so simple as the above, and therefore for details
the original paper should be consulted.

Liquid Extract of Hydrastis. — This preparation is a 45 per cent
alcohol extract of the drug of such strength that 1 fluid ounce contains
the active principles of one ounce of the drug. No official standards
exist. When properly prepared it should have the following char-
acters : —

Specific gravity
Solid residue .
Alcohol by volume
Total alkaloids
Hydrastine

1-025 to 1-040
20 „ 24 grms. per 100 c.c.
36 ,,40 per cent

4 „ 6 gr. per 100 c.c.

1-5 „ 3

Tincture of Hydrastis is an extract with 60 per cent alcohol of
one-tenth the strength of the liquid extract. No standards are official.
It should have the following characters : —

Specific gravity
Solid residue .
Alcohol by volume
Total alkaloids
Hydrastine

0-923 to 0-929

2 „ 2-5 grms. per 100 c.c.
56 „ 58 per cent
0-4 „ 0-6 gr. per 100 c.c.
0-15 „ 0-3

. The alkaloids in the above two preparations should be assayed in

the same manner as in the root. In the liquid extract 10 c.c. may be

used, and for the tincture 100 c.c.

Hydrastine C^iH^jNOg forms white prisms soluble in 120 parts

of 90 per cent alcohol, and in two parts of chloroform. When pure

it melts at 132°. The approximate purity of the alkaloid should be

checked by titrating it with one-twentieth normal hydrochloric acid.

N
One c.c. of — HCl is equivalent to 0*01916 grm. of hydrastine,

using methyl orange or cochineal as indicator. A solution in dilute
sulphuric acid is rendered fluorescent (blue) by the addition of a
trace of potassium permanganate. The alkaloid should not be colored
red on the addition of chlorine water — berberine gives a strong red
colour. A solution of hydrastine (neutral) is precipitated by potas-
sium bichromate solution, and if the separated precipitate be touched
with H.2SO4 it instantly becomes bright red — the colour fading in a
few seconds.

Berberine C.20H17NO4 exists also in Berheris vulgaris and other
plants. The alkaloid crystallizes with 4 molecules to 6 molecules of
water of which the equivalent of 2*5 molecules remain after heating
to 100°. It forms yellow silky needles melting at 145° and decom-
posing at 150". It is only slightly soluble in cold water, almost in-
soluble in ether, but readily soluble in hot alcohol. An aqueous
solution is coloured blood-red with chlorine water. If a trace of ber-
berine be boiled with hydriodic acid, the liquid diluted with water

HYOSCYAMUS. 555

and rendered slightly alkaline with ammonia, an intense blackish-
violet colour is produced.

Berberine hydrochloride, C00HJ-NO4 . HCl, SH^O and berberine
phosphate C2oHj,N04 . 2H3PO4 . 2H.,0, are the commoner salts of the
alkaloid. They should give the reactions described under berberine
and yield the amount of alkaloid indicated by the above formula,
when dissolved in water, the liquid rendered alkaline with ammonia
and extracted with warm amyl alcohol.

HYOSCYAMUS.

Hyoscyamus leaves are official in the Pharmacopoeia, being de-
scribed as the fresh leaves and flowers with the branches to which they
are attached, of Hyoscyamus niger ; also the leaves and flowering tops
separated from the branches and carefully dried, collected from the
flowering biennial plants. No standards are given.

From the fresh leaves, etc., an official green extract is prepared,
whilst from the dried leaves and flowering tops an official tincture is
prepared.

The principal constituent of the leaves is the alkaloid hyoscyamine
(p. 521), together with snaaller quantities of atropine and hyoscine.
The alkaloidal value of the dried kaves varies from 0*05 per cent to
0*18 per cent, rarely up to 0*25 per cent.

The mineral matter varies from 9 per cent to 14 per cent.

A microscopic examination of powdered hyoscyamus leaves (hen-
bane leaves) shows a marked difference between these leaves and
those of stramonium and belladonna. In henbane leaves the
mesophyll is heterogeneous and asymmetrical, the cells of the spongy
parenchyma often containing prismatic crystals of calcium oxalate ;
whereas stramonium and belladonna leaves contain chiefly cluster
crystals and sandy crystals respectively (Greenish).

According to Greenish the diagnostic characters of this drug are : —

(1) Characteristic glandular hairs.

(2) Prismatic crystals of calcium oxalate.

(3) Epidermal cells with wavy walls.

(4) Stomata surrounded by three or four cells of which one is
larger than the other.

(5) The absence of pericyclic fibres.

The sketch on page 556 represents the powdered leaves.

The alkaloids in the leaves may be estimated by the process used
for the tincture.

Tincture of Hyoscyamus. — This is made by extracting 2 ounces of
leaves with sufficient 45 per cent alcohol to produce 1 pint of tincture.
No official standards exist, but a genuine tincture should have the
following characters : —

Specific gravity .... 0-950 to 0-960

Solid residue
Alcohol by volume
Alkaloids

2-3 „ 3-6 grms. per 100 c.c.
43 „ 44 per cent

0-008 „ 0-015 „

V The alkaloids may be determined in exactly the same manner as
that described on p. 603 for tincture of stramonium.

556

FOOD AND DRUGS.

Green Extract of Hyoscyamus.—'^o standard exists for this pre-
paration, but properly prepared samples will contain from 0*2 to 0'4

r r

Fig. 51. — Powdered henbane leaves x 240. ccr, crystal cells ; a', crystals of cal-
cium oxalate ; ei, lower epidermis ; es, upper epidermis ; ffv, portion of
fibrovascular bundle of midrib ; ip, scar of fallen hair ; m, spongy parenchyma ;
pa,p'a', palisade cells; pg, glandular hairs; pt, simple hairs; st, stomat©- ;
tf, cortical parenchyma of midrib ; tr, tracheids and vessels. (Greenish &
Collin.)

By permission of the Editor of the " Pharmaceutical Journal ".

per cent of alkaloids, with an average value of 0'3 per cent, when
determined in the manner described on p. 518 for green extract of
belladonna.

IPECACUANHA.

Ipecacuanha root is the dried root of Psychotria ipecacuajiha
{Cephcslis ipecacuanha) and is thus described in the British Pharma-
copoeia : —

" Ipecacuanha occurs in somewhat tortuous pieces not often ex-
ceeding six inches in length, and one quarter of an inch in thickness.
It varies in colour from dark brick-red to very dark brown and is
closely annulated externally, the annulations not taking the form of
narrow merging ridges (distinction from Carthagena ipecacuanha).
It breaks with a short fracture, the fractured surface exhibiting a thick
greyish cortex, which usually has a resinous but sometimes a starchy
appearance and a small dense central portion. When examined under
the microscope the cortex exhibits small compound starch grains and

IPECACUANHA. 557

raphides ; the wood contains no vessels. The odour is slight, the
taste bitter."

Ipecacuanha occurs on the markets in its natural state, but in the
retail shops is frequently purchased in the form of powder. In the
former case it is liable to be mixed with roots of similar appearance,
as well as with the stems of the genuine plant, whilst in the state of
powder, other adulterants may be present.

The following description of the commercial root is due to E. M.
Holmes : —

The ipecacuanhas of English commerce may be divided into two
sections : —

1. Those that are derived from the genus Cephcelis.

2. Those that are derived from other genera belonging to the same
or to different natural orders.

1. Official Ipecacuanha (Cephcelis ipecacuanha, Eich). — Of this
kind there are several commercial varieties or qualities.

A. Brazilian or Bio Ipecacuanha. — When of good quality the roots
are one or two lines in diameter, and externally of a reddish or
blackish- brown colour. Specimens without a powdery surface are to
be preferred, since the powdery appearance is often due to the re-
mains of moulds. A good sample should yield about 80 per cent of
bark.

B. Indian Ipecacuanha. — This is derived from the plant culti-
vated in Johore (Straits Settlements), and has only been introduced
during recent years. It is imported from Singapore. Commercially
it is distinguished from the Brazilian kind by the presence of the deli-
cate rootlets, which usually occur to a much smaller extent in the
South American drug. According to an analysis by Eansom, it con-
tains 1'7 of (the so-called) emetine as against an average of 1-66 per
cent in the Brazilian kind, and may therefore be supposed to be of
good quality.

C. Mouldy Ipecacuanha. — It is calculated that about three out of
every four serons of ipecacuanha root imported have been damaged
by sea- water during the voyage to Europe or during transit to the
coast from the place of collection ("Pharmacographia," 2nd ed., p.
375). It has been maintained by some that the mouldiness does not
affect the amount of alkaloid present. This statement needs confir-
mation.

D. Woody Ipecacuanha. — It is of the prevalence of this quality in
commerce that complaints have recently been made. It is character-
ized by the presence of an unusual amount of stem. A small piece of
the woody stem is often attached to the root in good samples, but in
woody ipecacuanha it may amount to 30 or 50 per cent of the whole^
The stem is easily recognized by its smooth not annulated surface,
remarkably thin bark, and by the presence, visible under a good lens,
of pith in the centre of the woody column. As the stem is not official
in the Pharmacopoeia, and is probably one-third weaker than the root,
it should not be used for pharmacopceial preparations.

E. Doctored Ipecacuanha. — This quality consists of inferior, woody
or mouldy ipecacuanha that has been washed and dried. It has a

558 FOOD AND DRUGS.

dark colour and clean epidermis, contains few large pieces, and the
bark has been much broken off the root in the process of washing.
By this latter character and its dark colour it is easily recognized.

2. Carthagena or Savanilla Ipecacuanha {CejihcBlis acuminata,
Karsten). — This kind of ipecacuanha has recently been imported in
increasing quantities. It is, however, by no means a new article in
commerce. It is probably identical with the grey annulated ipecacu-
anha of Pereira, which he describes as "occurring in pieces of larger
diameter than in ordinary ipecacuanha, with fewer, more irregular and
less prominent rings ". Professor Guibourt remarked that consider-
able quantities of it arrived unmixed with the ordinary sorts, and that
he thought it to be a distinct kind coming from a different part of
Brazil, and derived from another species of Cephcelis (Pereira, " Mat.
Med." Vol. II, pt. 2, p. 58). This description is exactly applicable to
the Carthagena ipecacuanha of the present day, which is characterized
by the less prominent and more distant rings and transverse fissures.

Under the microscope it presents, according to Karsten, a distinc-
tive feature in the fact that the cortical parenchyma forms two dis-
tinct layers, which is not the case in ordinary ipecacuanha. The
radiate structure of the central woody column is also more distinctly
visible than in the ordinary ipecacuanha.

Carthagena ipecacuanha has been analysed by Dr. Wimmel, Con-
roy, and others, and the results obtained indicate that it varies, like
the Brazilian drug, in percentage of alkaloids, but that on the whole
it is probably not inferior to it in the amount of alkaloid present. It
must be remembered, however, that it contains a different crystalline
alkaloid which is not chemically identical with that of the Brazilian
drug.

Spurious Ipecacuanhas.

Owing to the name " Poaya " being used in a generic sense in
South American countries for roots possessing emetic properties,
various drugs bearing this name are sent to this country by mer-
chants at intervals of a few years. None of them approach ipecacuanha
in therapeutic value. Hence a description of their appearance in the
crude state may prove useful.

The plants from which these Poayas are derived belong chiefly to
the natural orders Buhiacece and Violacece, and one to the Polygalacece.
Those which have been identified in English commerce are three in
number, viz. : (1) Psychotria emetica, (2) Richardsonia scabra, and
(3) lonidium ipecacuanha.

Several other spurious ipecacuanhas, more or less resembling these
three, have at intervals been imported into Europe, but probably
have not been distinguished from them, except in one or two cases
in which a microscopic examination has been made. Of these I only
propose to notice those that have been met with in commerce in this
country.

A. Black or Greater Striated Ipecacuanha {Psychotria emetica,
Mutis).— This is so called from its black epidermis. The root is
slightly larger than Rio ipecacuanha and strongly constricted at inter-

IPECACUANHA. 559

vals of about an inch, more or less, the intermediate portions being
cylindrical and striated longitudinally. Internally the cortical por-
tion is thick in proportion to the woody column, and presents a horny
appearance, and sometimes a purplish tint. A decoction of the root
gives evidence of the presence of a reducing sugar, but not of starch.
According to Ransom it contains traces ("016 per cent) of emetine, or of
an alkaloid giving the same reactions. The woody column is dense,
and not visibly porous.

B. Lesser Striated Ipecacuanha {Bichardsonia species). — This
drug externally has also a black colour and striated appearance, and
constrictions at intervals like the greater striated ipecacuanha, but it
presents marked differences internally. The cortical portion is often
of a dark violet tint, and is full of starch, which can readily be de-
tected in a cold decoction by iodine, and the woody column is seen to
be distinctly porous when viewed under an ordinary lens. Professor
Planchon refers it provisionally to the genus Bichardsonia. Ex-
amined by Ransom it was found to contain '027 per cent of emetine.

C. Undulated Ipecacuanha {Bichardsonia scahra). — Externally
the root is of a greyish-brown colour, and differs from ipecacuanha
in not having raised rings. It is. however, marked with deep con-
strictions, often on alternate sides, which gives the root a some-
what undulated or falsely annulated appearance. In transverse
section the root is seen to be white, and starchy, and sometimes has
a faint violet tint, and the woody column is yellow and porous. It
has been stated to contain emetine, but the statement needs confir-
mation.

D. (1) White Ipecacuanha {lonidium Ijjecacuanha) . — This- drug
differs from the foregoing in its pale yellowish-brown colour and much-
branched character. The woody column is large, yellow, and porous,
and the cortical portion is thin, so that the root is more woody in
character than Bichardsonia, but it has transverse fissures and con-
strictions like the latter. It does not contain starch.

•D. (2) A root, supposed to be that of lonidium ipecacuanha,
entered the London market in 1884, and was examined by W. Kirkby,
who pointed out that it differed from the root of that plant in having
large wedge-shaped groups of sclerenchymatous cells in the cortical
portion, and more or less broad medullary rays in the woody column
("Pharm. Journ." 3, xvi. 126).

E. False Indian Ipecacuanha. — Some years ago a quantity of a
small root said to be imported from Southern India, was offered in
the London market as ipecacuanha. It differs from true ipecacuanha
in colour, which is of a pale reddish -brown, but it presents a ringed
appearance.

The following key to the microscopical structure of the com-
mercial ipecacuanhas may perhaps prove useful. It is based upon a
paper on this subject by Tschirch and Ludtke in the "Archiv der
Pharmacie," 1883, p 441.

I. Woody column containing chiefly tracheids, but no vessels.

A. Boot-bark containing starch and raphides.

1. Parenchyma of bark unilorm = Bio Ipec acua^iha.

560 FOOD AND DRUGS.

2. Parenchyma of bark forming two lawyers = Car thagena Ipe-
cacuanha.

B. Boot-bark containing no starch, but sugar.

Woody centre not visibly porous = Greater Striated Ipecacuanha.

II. Woody cylinder containing vessels, wood- cells, and medullary
rays.

A. Boot-bark, containing starch.

1. Medullary rays composed of a single row of cells, woody centre
visibly porous = Lesser Striated Ipecacuanha.

2. Medullary rays forming two or three rows of cells = Undulated
Ipecacuanha.

B. Boot-bark containing inulin.

1. Medullary rays of a single row of cells, no starch, sphsBraphides
in the bark= White Ipecacuanha (a).

2. Bark contains stone cells.

3. Medullary rays broad = White Ipecacuanha (b).

III. Rhizome having a monocotyledonous structure, brown pig-
ment cells in parenchyma, acicular raphides and starch present = False
Indian Ipecacuanha.

It is obvious that the microscopic examination of this drug is a
matter of extreme importance. The commonest adulterant of the
powdered root is Carthagena ipecacuanha, which is not ofi&cial in
the Pharmacopoeia, as it contains the characteristic alkaloids, emetine
and cephaeline in quite different proportions to those found in the
Rio root.

Samples of powder should be compared microscopically with type
powders from the two roots. It will be found that the starch grains
from Rio root are about half as large as those in Carthagena root, but
sometimes the larger grains of Rio root are equal in size to the
smallest of Carthagena root.

The powder should show an absence of vessels, but there are to be
found perforated tracheids, and acicular raphides. The presence of
stem in the sample is revealed by the sclerenchymatous cells,
lignified pith cells and spiral vessels.

The principal constituents of ipecacuanha root are two alkaloids,
emetine and cephaeline, which also occur in Carthagena root in which
the cephaeline predominates over the emetine, whereas in Brazilian
root the reverse is the case.

The chemistry of these bodies has received much attention from
Paul and Cownley, who give an interesting account of the matter in
the " Pharmaceutical Journal (3, xxv. Ill, 373, 690).

According to these chemists emetine is a practically colourless
amorphous alkaloid of the formula C15H22NO2 (Kunz-Krause considers
the formula to be C3QH40N2O5 ; — it is probable that the formula of
Paul and Cownley requires doubling). It melts at about 68°, and is
readily soluble in ether, alcohol, and chloroform but insoluble in
alkalis. It forms salts, containing one equivalent of acid, and easily
soluble in water, but not easily crystallizable. Cephaeline has the
formula Cj^HgoNOg (or C28H4QN2O4), and is a crystallizable alkaloid
melting at about 102". It is less soluble in ether than emetine, but

IPECACUANHA. 561

it is freely soluble in alcohol and chloroform. It is much more
soluble in warm petroleum ether than emetine, and is easily soluble
in alkaline solutions. It forms well-defined neutral salts, which
crystallize from acid solutions.

A third alkaloid exists in ipecacuanha root, melting at 138° and
forming lemon-yellow prisms. It is present, however, in very small
amount, and is known as psychotrine. Apart from these alkaloids,
ipecacuanha contains numerous other substances, the following being
the composition of a root most exhaustively examined bv Cripps and
Whitby :—

Per cent

Moisture 10-85

Volatile oil trace

Free fatty acid 0-16

Neutral fat 0-11

Wax (?) 0-03

Acid resins soluble in ether 0-05

Indifferent resins 0-23

Substance allied to quercitrin 0-03

Tannin (total) 1-13

Phlobaphane 0-34

Saccharose 2-12

Dextrose (total) 4-06

Dextrine 2-08

Mucilage . . . . - 3.8I

Albumen precipitated by boiling 3-10

Albumen not precipitated by boiling ...... 0-23

Albumen, pectin, etc., insoluble in HgO 3-34

Albumen not precipitated by alcohol 3-12

Organic acids and allied bodies 1-48

Alkaloid removed by ether from alcoholic extract . . . 1-91

Alkaloid removed by chloroform ...... 0*24

Alkaloid, etc., removed by chloroform from acid solution . . 0-10

Alkaloid from aqueous extract ....... 0*17

Colouring matter and decomposition products .... 2-52

Besinous (?) matter not removed by agitat'on with ether, etc. . 0-07

Starch • 44-44

Cellulose, lignin, etc 11-30

Ash, soluble in H^ 0-53

Ash, soluble in HCl 1-69

Ash, insoluble in HCl ........ 0-21

99-65

Arndt ("Year-book of Pharmacy," 1889, 136) claims to have iso-
lated 0*3 .per cent to 0*5 per cent of- choline from the root, but this is
denied by Cripps.

The ash of ipecacuanha is rarely above 4 per cent, any excess be-
ing usually due to residual earthy matter left on the roots. On
page 562 are figures for pure samples of Eio and Carthagena roots
obtained by the author.

The Assay of Ipecacuanha Alkaloids. — The official process of the
British Pharmacopoeia for the determination of alkaloids in the liquid
extract, may be applied to the root, using 20 grms. for the determina-
tion. The root in coarse powder should be exhausted with 90 per
cent alcohol in a small percolator, and when exhausted, the mass
should be mixed with 2 grms. of pure lime, and allowed to stand for
VOL. I. 36

562

FOOD AND DKUGS.

Ash

Moisture.

Ash.

Soluble iu

Insoluble in

H2O.

HCl.

Eio root

10-42

2-^0

0-49

0-24

» ..

11-55

3-25

0-58

0-38

• > >>

11-90

3-90

0-52

0-69

M M

10-80

4-35

0-60

Carthagena root

10-25

4-65

0-62

0-75

„ ,,

11-80

3-98

0-59

0-89

1205

5-01

0-62

0-91

Fig. 52. — Powdered ipecacuanha.

twenty-four hours, and then again exhausted with 90 per cent
alcohol and the percolates mixed. The liquid should be concentrated
to about 20 C.C., and diluted with 20 c.c. of water. The alcohol is
then evaporated on the water bath, and to the warm solution a slight
excess of solution of subacetate of lead is added. The liquid is
filtered, and the precipitate well washed. Excess of lead is removed
from the filtrate by dilute sulphuric acid. The liquid is again filtered,
the precipitate washed and the liquid transferred to a separator.
Excess of ammonia is added, and the liquid well shaken with 25 c.c.
of chloroform. The chloroform is separated, and the extraction re-

IPECACUANHA. 563

peated twice. The mixed chloroform solutions are evaporated, and
the residue dried at 80° C, and weighed.

There is always a slight loss by this process, as alkaloids are pre-
cipitated with the lead, but it does not exceed 0*1 per cent.

Keller prefers the following process which gives very concordant
results : 10 grms. of the dried and finely powdered root are well
agitated in a bottle of 150 c.c. capacity with 40 grms. of chloroform
and 60 grms. of ether ; 10 grms. of solution of ammonia are then
added, and the agitation repeated at frequent intervals during one
hour, after which another 5 grms. of solution of ammonia are added,
and again well agitated with the mixture. After settling, 50
grms. of the decanted solution, representing 5 grms. of the dried root,
are carefully distilled in a weighed Erlenmeyer flask ; the varnish-
like residue is twice treated with 10 c.c. of ether, and evaporated by
forcing a current of air through the flask. After the last traces of
ether have been removed, the residue is dried in a water bath and
weighed. For the titration of the alkaloid it is dissolved in 10 c.c. of
absolute alcohol with the aid of heat, sufficient water added to pro-
duce a permanent turbidity, and the titration then carried out with
decinormal hydrochloric acid in the presence of a few drops of haema-
toxylin solution as an indicator. Each c.c. of the decinormal acid
represents 00254 grm. of -emetine. An improvement of this assay
consists in the removal of the fat from the ipecacuanha root by per-
colation with ether previous to the process described. This preliminary
treatment renders the subsequent titration more easy and distinct.

Kottmayer (" Phar. Post." 1892, 913, and 933) claims that the
following process gives the most accurate results, and according to
the author's experience this is true, but it is far too tedious for ordinary
use : —

The powdered root should be used without drying, since heating
renders the extraction of the alkaloid more difficult. Fifteen grms.
of the powdered root are placed in a bottle with 148 c.c. of 90 per
cent alcohol and 2 c.c. of hydrochloric acid of specific gravity 1*12,
and digested, with frequent agitation, at 40° C. for four days ; after
cooling to 15° C, 100 c.c. are removed, mixed in a capsule with 20
c.c. of a 10 per cent alcoholic lead acetate solution (50 per cent
alcohol), and, after the addition of 1*5 grms. of slaked lime, evaporated,
with occasional stirring, to a pasty consistency ; 5 grms. of powdered
glass are then incorporated, and the heating is continued on a water
bath with constant stirring until a dry powder results. This is ex-
tracted for ten hours with chloroform, the chloroform solution evapor-
ated in a weighed vessel, dried at 100° C, and weighed. The crude
alkaloid thus obtained is dissolved in 2 c.c. of normal hydrochloric
acid, the insoluble matter collected on a weighed filter, thoroughly
washed, dried, and weighed. The total residue minus the weight of
the insoluble resin leaves the weight of the pure alkaloid.

Cripps (" Pharm. Journ." 3, xxv. 1093) recommends the
following modification of Lyons' method : 2*5 grms. of the powdered
sample are exhausted by a mixture of 250 parts of ether, 10 of
ammonia and 20 of alcohol. The alkaloids are separated from the

564 FOOD AND DEUGS.

ethereal solution by repeated extraction with dilute hydrochloric
acid, the aqueous solution rendered alkaline with ammonia, and
the alkaloid finally dissolved out first by ether and then by
chloroform. The solvents are evaporated in a current of air, and
the residue weighed after drying at 50° to 60°. This gives the
approximate amount, after which it should be dissolved in 5 c.c. of
one-twentieth normal hydrochloric, and the excess of acid titrated
back with caustic soda. Each c.c. of decinormal acid represents
0-0244 grm. of alkaloids. This figure is based on the average pro-
portions of emetine and cephaeline present. Calculated as emetine
the value 0*0254 would be used.

F. C. J. Bird has devised a process for the alkaloidal assay of the
root which is both convenient and accurate, and well suited for general
work. The drug should be in a fine powder and 10 grms. should be
used for the determination. The following are the details of the pro-
cess. In the event of the chloroform not separating quickly a little
ether should be added : —

Rio ipecacuanha in fine powder .... 10 grms.

Sodium bicarbonate 2 ,,

Water 5 c.c.

Mix about half the soda with the powdered ipecacuanha, shake the
remainder with the water and rub the whole in a small mortar to a
uniform moist granular powder.

Amyl alcohol 1 volume »

Chloroform 1 „ U.S.

Ether 3 volumes'

Add the moistened powder to 20 c.c. of the above solvent, previ-
ously placed in D (plugged with cotton wool, as shown on page 565)
and macerate for half an hour, with occasional shaking. Force out
the liquid by compressing H, and cover the powder with 10 c.c. more
menstruum. Agitate vigorously, let stand fifteen minutes and again
force out the liquid. Kepeat this at intervals of a quarter of an hour
until ten or twelve quantities of menstruum have been used or the
powder is exhausted.

(If time is of no importance, percolation in the ordinary way may
be substituted for the above procedure.)

Agitate the mixed ethereal extracts successively with : —

Dilute sulphuric acid 4 c.c.

Water 6 „

Water 5 „

To the mixed acid solutions add carefully ammonium

bicarbonate 0-5 grm.

Shake out the alkaloid with chloroform containing about one-sixth
its volume of ether four times, adding 1 drop of ammonia to the chloro-
form. Mix the four portions of chloroform and either (1) evaporate, dry
below 80° C, and weigh, or (2) dilute with chloroform to 100 c.c. and
divide into two equal volumes. Evaporate, dry, and weigh the one.

IPECACUANHA.

565

as usual, but remove the chloroform from the other by a current
of air and titrate the residue, using the equivalent 0"0244. The
titration figure should come within 2 per cent or 3 per cent of the
weight.

For the official process of assay of the liquid extract Bird proceeds
as follows, the apparatus devised by him greatly expediting the pro-
cess : —

Liquid extract of ipecacuanha, 20 c.c. ; distilled water, 20 c.c. ;
acetic acid, q.s. to faint acid reaction. Evaporate off the alcohol and add
distilled water, 20 c.c. liq. plumbi subacet., 10 c.c. Keep the mixtureon .
the water bath for a few minutes until the magma which at first forms
changes to a thin liquid, and the precipitate assumes a finely granular
condition. Transfer to filter

B, and connect A with a
water pump. (In the absence
of the latter exhaustion of
the air by forcibly sucking
with the mouth at A, retain-
ing the vacuum by the clip

C, ensures a very fair rate of
filtration.) Wash the nearly
dry solid cake remaining on
the filter with distilled water,
30 c.c, added in small por-
tions. To the filtrate in C
add acid sulph. dil., 25 c.c. ;
change the filter paper on
B ; transfer B to another
filter flask, and pour upon it
the liquid in C, aiding filtra-
tion by a vacuum as before.
Wash the cake of lead sul-
phate with distilled water,
15 c.c. To the filtrate in the
same flask add chloroform
and 5 c.c. ammonia. Cork
the flask, agitate vigorously
and transfer contents to D.

Connect D with H by
the rubber cork G, and hav-
ing inflated the pressure-ball H, force the chloroform and a portion
of the aqueous liquid in D through the filtering medium shown,
into E. Draw off the chloroformic layer, which should be perfectly
clear and bright, into the "tared glass dish F. Return the aqueous
liquid in E and D to the filter flask, add chloroform, 25 c.c. ; and
proceed as before. Repeat a third time with chloroform 25 c.c.
Finally collect the three chloroformic layers in F. Evaporate and
dry the residue below 80° C.

Fig. 53. — Bird's apparatus for the assay of
ipecacuanha, belladonna and nux vomica
preparations. H, is an india-rubber bel-
lows ; B, is a Buchner's funnel, with a flat
filter paper.

566 FOOD AND DRUGS.

Preparations op Ipecacuanha.

Acetum fyecacuajihce. — Vinegar of ipecacuanha is prepared by
mixing 1 part of liquid extract of ipecacuanha, with 2 parts of 90 per
cent alcohol and 17 parts of dilute acetic acid. No standards are
given, but as the process is merely a mixing of these liquids, no loss
should occur, and the arguments as to loss in acetic acid adduced in
the case of Hudson and Bridge (see Vol. II) would hardly apply here.
This preparation should contain O'l to 0'1125 grm. of alkaloids per 100
c.c, when 50 c.c. (evaporated to 5 c.c.) are assayed by the process de-
scribed under Extract of Ipecacuanha ; 9 per cent of alcohol (absolute)
and from 3 '6 to 3-7 per cent of acetic acid. Its specific gravity should
lie between 0-982 and 0-987.

Liquid Extract of Ipecacuanha. — The official requirement for the
liquid extract in the Pharmacopoeia is that it should contain from 2 to
2-25 grms. of alkaloids wiien assayed by the official process, which is
described under the root ; 20 c.c. of the extract should be used for the
determination. It should be pointed out that the Pharmacopoeia re-
quires the alkaloids to be weighed and not titrated.

This official process is, however, by no means the best available,
and Farr and Wright have strongly recommended the following, which
certainly gives very accurate results : —

Five c.c. of the fluid extract is placed on a small porcelain dish,
10 drops of diluted sulphuric acid added, with 5 c.c. of water, and the
mixture evaporated over a water bath until the volume of liquid is re-
duced to about 3 c.c. This is run into a separator, the dish carefully
rinsed with 10 drops of water, and then with 15 c.c. of chloroform,
the whole being transferred to the separator. An excess of ammonia
is added, and the mixture well shaken, and allowed to stand until the
chloroform has separated. This is run off, and the agitation and
separation repeated with two successive quantities of 5 c.c. of chloro-
form. The chloroformic solutions are bulked, and the alkaloids ex-
tracted by shaking with three successive quantities of 10 c.c. 1 per
cent sulphuric acid. The acid alkaloidal solutions are drawn off in
turn and mixed. The alkaloids are finally recovered from this solution
by repeating the treatment with ammonia and chloroform. The solu-
tion of the alkaloids in chloroform is then evaporated in a tared dish
over a water bath until all chloroform has been removed. The weight

is taken, and the alkaloidal residue titrated with — — - and — — — as

10 20

previously described.

The following process, due to Naylor and Bryant, also gives very
accurate results : —

Place 10 c.c. of liquid extract in a basin over a warm water bath
until the alcohol is dissipated. The solution is transferred to a 50 c.c.
flask, and the basin is washed with small portions at a time of a mix-
ture of 2 c.c. of diluted sulphuric acid and 30 c.c. of water. The
solution is filtered, and water passed through the filter until the
volume measures 50 c.c. Of the filtrate 25 c.c, representing 5 c.c.
of liquid extract, are transferred to a separator, together with the

IPECACUANHA.

567

small portions of water used for washing the measure, and the solu-
tion is shaken up with 10 c.c. of chloroform. After removal of the
separated chloroform the solution is agitated with another 10 c.c. of
chloroform which after separation is also withdrawn. The solution
is then made alkaline with ammonia, and extracted successively with
3 X 10 c.c. of chloroform. The chloroform solutions are mixed, eva-

N
porated, and the residue weighed and titrated with — HCl.

Liquid extract of ipecacuanha has a specific gravity varying be-
tween 0-885 to 0-915, and yields from 9 per cent to 12 per cent of
extractive matter when dried at 100°. The average alcoholic content
is 78 per cent to 79 per cent.

Ipecacuanha Wine. — This is a mixture of one part by volume of
the liquid extract with 19 parts of sherry. It should be of the same
alkaloidal strength as vinegar of ipecacuanha : 50 or 100 c.c. should
be evaporated to 5 c.c. and the assay then carried out as described
under the liquid extract. It should contain 20 per cent to 21 per cent
of alcohol. This preparation should be tested for salicylic acid, which
is sometimes present in sherry. It should be acidified with dilute
sulphuric acid and extracted with ether and the ether evaporated.
The residue should show no violet coloration with a drop of ferric
chloride solution.

Pulvis lyecacuanhce Compositus. — Compound ipecacuanha powder
or Dover's powder is a mixture of 1 part of ipecacuanha 1 part of opium
and 8 parts of potassium sulphate. It should contain 80 per cent to
82 per cent of ash, consisting practically entirely of potassium sulph-
ate. The total alkaloids present when determined as described under
opium should be 1*2 per cent.

Colour Eeactions of the Isolated Alkaloids of Ipecacuanha.

Reagent.

Emetine.

Cephaeline.

Psychotrine.

Ferric chloride.

Indefinite.

Bluish-green.

Pale cherry-re-^.

Indefinite.

Froehde's reagent.

Dirty green.

Pink, changing to

Pale pink.

Bluish.

green.
Reddish-purple.

Dull purple.

Frohde's reagent

Grass-green.

Prussian blue.

Pale pink, chang-

and hydrochloric

ing to pale green

acid.

Starch and iodic

Negative.

Blue.

acid.

Ferric chloride and

Gradual blue

Almost immediate

Immediate blue.

potassium ferri-

coloration.

blue.

cyanide.

Immediate blue.

Allen and Scott Smith (" Analyst," xxvii. 346) have called attention
to the marked resemblance between some of the colour reactions of the
alkaloids of ipecacuanha and those of morphine, which might well lead
to confusion if qualitative tests were applied for the detection of one
or the other alkaloids.

568

FOOD AND DKUGS.

They give the following summary of the principal reactions

En
X

-3
O

<
<

:3
Q

-«5
Q
I— I
O

(in
O

cn
55

D
O

1-3
O

Cud
<

Greenish-blue
Purple

Purple, fading

Immediate
blue

1
1

1— 1

Blue,

changing to
green

Violet-blue,
changing to
dirty pink

Deep blue

Pink,

chanoting to
blue slowly

Immediate
blue

Blue,

changing to
green

Violet-blue,
changing to
dirty pink

Deep blue
Negative

Immediate
blue

•d

1— 1

1
1

Indefinite

Pink,

changing to
blue and
green

Deep blue
Negative

Immediate
blue

Blue,

changing to
green

Purple

Deep blue

Blue,

changing to
green

Immediate
blue

1— t

Blue,

changing to
green

Bluish-purple

Deep blue

Immediate
blue

Ferric chloride ....

Froehde's reagent ....
{5 rag. of molybdic acid in 1 c.c.

Froehde's reagent and hydrochloric
acid

Starch and iodic acid

Ferric chloride and potassium ferri-
oyanide

JABOEANDL 569

A most valuable means of detecting ipecacuanha alkaloids consists
in the production of psychotrine in a crystallized form. Paul and
Oownley describe the crystals as well-defined transparent prisms of a
pale lemon-yellow colour. Under the microscope, psychotrine forms
Yery minute crystals, which appear to belong to the regular system.
Many of them appear to be octahedral, and closely resemble microscopic
•crystals of arsenious oxide. Other crystals present a remarkable
resemblance to granules of rice-starch. Crystals of psychotrine
for microscopic observation are readily obtained by shaking out
an amylic alcohol or chloroform solution of the alkaloid with a little
'dilute acetic acid. The acid liquid is separated, concentrated if neces-
sary, and placed in a watch-glass, or, preferably, on a microscopic
«lide furnished with a cell. A watch-glass or small beaker is then
moistened internally with ammonia, and inverted over the alkaloidal
ticetate solution. After a time the vapours of ammonia are absorbed,
and liberate the alkaloid in characteristic crystals, which are observed
under the microscope. There is no occasion to employ pure psycho-
trine for the purpose, the crystals being readily obtainable from the
mixed alkaloids of ipecacuanha.

JABOEANDI.

The leaves of Pilocarpus Jaborandi are official in the Pharma-
•copceia, but no standards are given.

The principal constituents of this drug are the alkaloids, pilo-
carpine CjiHjgN.^O^, isopilocarpine CjjHjgNgO^ and pilocarpidine
'C10H14N2O2. Pilocarpine is by far the most important of these, and
the drug may be regarded as owing its therapeutic activity to this
alkaloid, which is described below. It occurs to the extent of about
from 0*2 per cent to 0*5 per cent in the leaves.

Jaborandi leaves should yield from 6 per cent to 8 per cent of ash,
rarely up to 9*5 per cent.

Tincture of Jaborandi is official. It is prepared by extracting four
•ounces of the leaves with 45 percent alcohol, to make 20 fluid ounces
of tincture. No standards are given, but a genuine tincture, prepared
ifrom leaves of good quality, will have the following characters : —

Specific gravity . . . 0-956 to 0-959

Solid residue . ' ^^ • 2-6 ,, 4-3 grms. per 100 c.c.

Alcohol (by volume) . . 42 ,, 43 per cent

Pilocarpine . . . 0-08 ,, 0-15 grm. per 100 c.c.

The following process is that of Farr and Wright (" Pharm.
Jour." 3, XXII. 1) for the determination of the pilocarpine. Fifty c.c.
•of the sample to be assayed are introduced into a porcelain dish and
evaporated over a water bath, water being added, if necessary, until
all spirit is driven off. The alkaloidal liquor is allowed to cool, 1
c.c. of semi-normal sulphuric acid added, and the solution filtered
through cotton wool, the dish being rinsed with acidulated water,
and the rinsings added to the filtered liquid. The latter is then
.rendered alkaline by the addition of 2 c.c. of B.P. liquor ammonias.

570 FOOD AND DRUGS.

and the liberated alkaloid taken out by agitation with two successive
quantities of 15 c.c. of chloroform.

To obtain the alkaloid in a pure condition, it is withdrawn from
solution in chloroform by shaking with acidulated water, 25 c.c. of
distilled water being acidified with 2 c.c. of semi-normal sulphuric
acid, and added in three successive portions. The mixed acid solutions
are again rendered alkahne with ammonia, and shaken with two suc-
cessive quantities of 15 c.c. of chloroform. The chloroformic alkaloidal
solution is then agitated with a little shghtly ammoniated water, and
after separation is drawn off and evaporated, and the residue heated
in a water oven at 100" till the weight is constant.

The amount of alkaloid may be checked by dissolving the residue
in a calculated excess of one-twentieth normal hydrochloric acid, and
titr-iting the excess of acid with one-twentieth normal soda or baryta
solution, using iodeosine or methyl-orange as indicator. Each c.c. of
one-twentieth normal acid is equivalent to 0-0104 grm. of pilocarpine.

The amount of mucilaginous matter present in the tincture is so great
as to produce emulsitication of the chloroform when that liquid is shaken
up with it, and it is therefore necessary to remove such matters
by adding strong alcohol, before proceeding with the estimation
of the tincture.

Fluid extract of Jaborandi is five times the strength of the tincture
and should have the following characters (which are not ofi&cial) : —

Specific gravity . . . 1-020 to 1-050

Solid residue . . . . 21 ,,22 grms. per 100 c.c.

Alcohol by volume . . . 33 ,,35 per cent
Pilocarpine .... 0'2 ,, 0-75 grm. per 100 c.c.

Pilocarpine CuHigNgOg has not been obtained in the crystalline
condition, but only as a thick syrup. It is official in the form of its-
nitrate, which, as in the case with the hydrochloride, forms well-
defined crystals. The Pharmacopoeia describes pilocarpine nitrate
C^jHj^gN.^O^ . HNOg as a white crystalline powder, soluble in 8 parts to
9 parts of cold water and freely soluble in hot alcohol. Strong sul-
phuric acid forms with it a yellowish solution, which on the addition
of potassium bichromate gradually acquires an emerald-green colour.
It leaves no ash when burned.

Pilocarpine hydrochloride CjjHj,;No02 . HCl is not official, but is a
salt sometimes met with in pharmacy.

Pilocarpine nitrate, when rendered alkaline, and the free alkaloid
extracted with chloroform, should yield 76-75 per cent of free pilo-
carpine. The nitrate should melt at 173° to 175° and in aqueous
solution should show a specific rotatory power of about -f- 88° ; a
concentrated aqueous solution does not yield a precipitate with am-
monia or caustic soda solution. Ten or 20 milligrams dissolved in 2
c.c. of water and 2 c.c. of slightly acidified hydrogen peroxide added,,
and 5 c.c. of benzol added, and finally 3 or 4 drops of a dilute solution
of potassium bichromate (1 in 300), and the mixture gently shaken,
the benzol layer will acquire a violet colour and the aqueous layer
will be yellow. If the pure alkaloid be separated as above mentioned,

LOBELIA— JALAP. 571

it should have a specific lotation in aqueous solution of +101°.
It possesses both acid and basic properties, and forms a crystalline
picrate melting at 147°.

LOBELIA.

The dried flowering herb of Lobelia inflata is official, but no
standards are given.

The principal constituent is the liquid alkaloid lobeline CjgHggNO
(?) which can be determined with comparative accuracy.

The drug yields from 10 per cent to 12 per cent of ash.

The alkaloid may be determined in the same manner as coniine

N
in hemlock fruits (p. 544). One c.c. of —- HCl is equivalent to

0-01425 grm. of alkaloid, calculated as lobeline.

Ethereal tincture of lobelia is an official preparation made by ex-
hausting 4 ounces of the drug by a mixture of 1 volume of ether and
2 of 90 per cent alcohol, the resulting product measuring 1 pint.
There are no standards given. A genuine tincture should have the
following characters : —

Specific gravity . . ' . 0*812 to 0-817

Solid residue .... 0-9 ,, 1-5 grnis. per 100 c.c.

Alkaloids . . . . 0-02 „ 0-04 per cent

By careful fractionation 30 per cent should be obtained boiling
below 50°, indicating the presence of a due proportion of ether.

The alkaloids are determined in the same manner as in the case
of tincture of conium.

JALAP.

Jalap consists of the dried root tubercles of IpomcEa purga. It is
official in the British Pharmacopoeia, and is required by that authority
to contain from 9 to 11 per cent of resin, when assayed by the follow-
ing process : —

Ten grams of the jalap in powder are digested with 20 c.c. of 90
per cent alcohol in a covered vessel, heated gently for twenty-four
hours. It is then transferred to a small percolating apparatus and
exhausted with alcohol. Five c.c. of water are added to the alcoholic
extract, and the alcohol removed by distillation. The residue, whilst
still hot, is transferred to an open dish, allowed to cool, and the separ-
ated resin washed several times with water, dried, and weighed.

This resin must not yield more than 10 per cent to ether, indicat-
ing absence of scammony and Tampico jalap resins ; and an alcoholic
solution should not yield a" blue-green colour with solution of ferric
chloride, indicating the absence of guaiacum resin.

The principal constituent of jalap is generally said to be the so-
called jalap resin or jalapin, which is essentially a glucoside of the
formula C54Hc,g02-.

This body, also known as jalapurgin ,or convolvulin, must not be
confused with the glucoside of Ipomoea simulans, the Tampico jalap,

572 FOOD AND DRUGS.

which is often termed jalapin, but is probably identical with scam-
monin. When pure, jalapin (Jalajmrgin, convolvulin) is a white amor-
phous powder, almost if not quite insoluble in ether, petroleum ether,
benzene or water ; slightly soluble in chloroform, and easily so in alcohol
and acetic acid. It melts between 150° and 155°. It reduces am-
moniacal silver nitrate solution on warming, and after boiling with dilute
acids, the reaction products reduce Fehling's solution. The products of
hydrolysis are glucose, methyl-ethyl acetic acid, purgic acid C^H^gO^g
and convolvulic acid C45Hgo0.28.

Jalapurgin dissolves in sulphuric acid with a fine red coloration.

Recent researches by Power and Rogerson (" Pharm. Jour." 1909
[iv.] 29, 7) indicate that jalap resin is of a much more complex com-
position, and that its physiologically active components are all indefi-
nite and amorphous, and that there is no justification for the formulae
usually assigned to them. From jalap resin, a small quantity of
ipurganol C.2iIl3o02(0H)2 was isolated, as well as ^-methyl-aesculetin.

The Examination of Jalap. — The examination of the whole tubers
is practically confined to an estimation of the amount of resin present,
and an examination of the resin itself. In the case of powdered jalap,
this should be supplemented by the estimation of the ash and a micro-
scopic examination.

The ash of jalap should not exceed 6"5 per cent.

The Besin Value of Jalap. — A large number of samples do not
contain as much resin as required by the British Pharmacopoeia. Such
samples, however, can be used for the manufacture of "jalapin," and
as the tincture of jalap of the Pharmacopoeia is a standardized pre-
paration, it does not appear to be of much importance whether a
weaker jalap is used in its preparation. It would certainly be advis-
able to reduce the ofiicial standard to a minimum of about 7*5 per cent.

Alcock ("Pharm. Jour." 3, xxii. 107) prefers the following process
for the determination of resin. In the author's experience it is a
better process than that of the Pharmacopoeia, in that less extraneous
matter is extracted by amyl alcohol than by ethyl alcohol, and it
obviates loss of resin which may become attached in films to the dish
in which it is washed with hot water. This process is as follows : —

Place 1 grm. of powdered jalap — free from agglutinated lumps —
in a suitable bottle, add 20 c.c. of amylic alcohol, and shake well from
time to time. After a few hours, strain the liquid off through a little
cotton wool into a glass separator, wash out the bottle with 5 c.c. of
amylic alcohol, and place the washings on the marc in the funnel ;
repeat with 5 c.c. more if necessary, so as to ensure the presence of
all the resin in the separator.

Now shake up the amylic solution of the resin with small quan-
tities of water at 50° C, set aside for the liquids lo separate, remove
the lower aqueous layer, and repeat the washing with water until
nothing more of a non-resinous nature is removed. Afterwards transfer
the solution of the resin to a weighed dish containing 10 c.c. of distilled
water, wash out the separator with a little amylic alcohol, placing the
washings in the dish, evaporate on a water bath in the usual way, and
when dry, weigh.

JALAP.

573

After the resin is extracted it should be powdered and a weighed
quantity exhausted with anhydrous ether. If more than about IQ
per cent be dissolved, admixture with foreign bodies, such as Tampico-
jalap, is to be suspected.

Four samples of genuine jalap were extracted by the author and
the resins examined. They were found to have the following char-
acters, after drying at 105" : —

Soluble in ether .
Soluble in alcohol
Acid value .
Ester value .

4*95 per cent
Complete
14-6
116

6'22^per cent
Complete
13-0
124

5*12 per cent
Complete
15-0
120

5-3 per cent

Complete

16-5

122

These results are in agreement with those of Kremel and Beckurts.
Commercial " jalapin " should have substantially the above characters.

Microscojnc Characters. — Powdered jalap should be examined both
in its natural condition, and after being bleached wiih sodium hypo-

Powdered jalap.

chlorite. Many rounded cells containing starch grains will be found,
and many dark, somewhat angular, resin cells. Parenchymatous

574 FOOD AND DEUGS.

cells, fibres and sclerotic cells with very thick walls are to be ob-
served. The starch granules are circular and flattened, or oyster-
shaped. The hilum is distinct and a few concentric rings can be
traced. Anything more than quite a small proportion of pitted vessels
and wood fibres should be regarded with suspicion.

Tincture of Jalaj). — The official tincture of jalap is an extract of
jalap with 70 per cent alcohol, of such strength that when assayed
by the process for determining the resin in jalap, it contains from
0*145 to 0-155 gram of resin in 10 c.c.

The. specific gravity of a properly prepared tincture lies between
0-910 to 0-915. The solid residue should not be less than '6-5 per
cent nor more than 4*7 grms. per 100 c.c. ; and the alcoholic strength
should not be less than 65 to 66 per cent.

NUX VOMICA.

This drug, which is official in the Pharmacopoeia, consists of the
dried ripe seeds of Strychnos nux vomica.

There are no standards in that authority for the drug, its pre-
parations being standardized to a given amount of strychnine.

As the seeds, which are from three-quarters of an inch to one
inch in diameter, are almost entirely sold whole, the analyst has
rarely to consider the question of adulteration, except in so far as from
time to time a false nux vomica is to be found mixed with the true
seeds.

The examination of the seeds is therefore usually confined to the
determination of the strychnine present, or sometimes the brucine
also.

Strychnine and brucine are the two characteristic alkaloids of this
drug, the strychnine being the more important.

Nux vomica should yield from 2 per cent to 2-5 per cent of ash on
incineration.

Strychnine C^iH^gNgOg is dealt with on page 578. It is present
in the seeds to the extent of 0*7 per cent to 1-60 per cent.

Brucine C23H^,5N204, is possibly a dimethyl-strychnine, and is a
bitter, white, odourless, crystalline compound, usually containing four
molecules of water. It melts at about 115°. It is less poisonous than
strychnine. The most satisfactory reaction for this alkaloid is the
following. On adding a drop or two of cold concentrated nitric acid
to an ether-chloroform residue or any other solid matter containing
brucine, a scarlet or blood-red colour is produced which on heating
changes to yellow. If the mixture be then cooled and heated with a
trace of stannous chloride or sodium thiosulphate, a purple colour
results, which is destroyed by excess of either HNO3 or stannous
chloride. From the analyst's point of view, brucine is only important
in reference to its separation from strychnine.

Assay of Nux Vomica. — The process described as official under
liquid extract of nux vomica is the result of work by Dunstan and
Short (" Pharm. Jour." 3, xiv. 290). It has, however, been shown
that brucine ferrocyanide is not completely soluble in acidulated water.

NUX VOMICA. 575

and that the ferrocyanide salts are very unstable. It is also probable
that some of the strychnine is carried into solution in the course of the
washing. Farr and Wright have shown by an exhaustive series of ex-
periments that the official process (see page 578) is sufficiently accurate
for all practical purposes, but that not more than 5 c.c. of the
liquid extract should be used, and not more than 30 c.c. of the
tincture. They improve the process, however, by using 200 c.c. of
wash water at 100° F., and making an allowance for the strychnine
dissolved. This may be taken as 0*002 grm. per 100 c.c.

Beckurts (" Pharm. Post," 18, 67, and " Apoth. Zeit." 1891, 537)
prefers the following method of assay : —

Ten grms. of powdered seeds are exhausted with about 45 per cent
alcohol and the percolate evaporated to a thin syrup. This is dis-
solved in a mixture of 10 c.c. of alcohol, 5 c.c. of water and 5 c.c.
of ammonia (10 per cent). The alkaloids are shaken out with chloro-
form. The alkaloids are then weighed, or titrated, but no separation
of the two bases is attempted.

F. C. J. Bird has found the following to give most accurate
results : —

Nux vomica in powder . . ' . . . . 5 grms.
Solution of potash, 10 per cent .... 2 c.c.

Triturate in a mortar until uniformly moistened

Arayl alcohol, 1 vol | «; • +

(Solvent) Chloroform, 3 vo's } ^ sutticient

' Ether, 4 vols / ^^^^^'^J-

Add the moistened powder to 20 c.c. of the above solvent, pre-
viously placed in a separator plugged with cotton wool, and macerate
for half an hour with occasional agitation. Adapt a pressure-ball to
the separator and force out the liquid as completely as possible by air
pressure. Add sufficient solvent to just cover the powder, insert the
stopper of the separator, agitate vigorously, let stand fifteen minutes
and again force out the liquid. Repeat this until no more alkaloid is
extracted, as shown by evaporating a few drops and testing with
diluted acid and Mayer's reagent. Usually five to six extractions will
be found sufficient.

Agitate the mixed ethereal extracts with : —

Diluted sulphuric acid, 6 c.c. . . ) -.i ,n j -.^

Water, 25 c.c. . . . . . ) ^1' ^O, and 10 c c.

in three successive quantities. Transfer the united acid liquids to a
200 c.c. separator half filled with water at 70° F. (21-1" C), and having
the neck above the stopcock plugged with a very small pledget of
cotton wool. Add a freshly prepared solution of

Potassium ferrocyanide ." 1*25 grms.

Water 25 c.c.

and completely fill the separator with water at 70° F. (21" C.) Re-
place the stopper by a cork carrying a thermometer; if necessary
raise the temperature of the contents to 70° F., by rotating the sepa-
rator in the steam of a water bath. Agitate occasionally during half

576 FOOD AND DRUGS.

an hour, then allow to remain at rest for an additional hour and a-
half, maintaining the temperature of the liquid at 70° F. by oc-
casional warming when necessary. (At 70° F. precipitation of strych-
nine ferrocyanide invariably commences well within a minute after
the addition of the potassium ferrocyanide solution.) Adapt an air-
pressure ball to the separator and force out the mother liquor.

Diluted sulphuric acid .... 5 c.c. ^ At 100° F.
Water 195 c.c, | (37-7° C.)

Add about 50 c.c. of the above wash water to the precipitate, ro-
tate, and apply air pressure as before, regulating the flow of liquid by
the stopcock to a quick succession of drops. Then add the re-
mainder of the wash water, agitate and repeat. Insert the stopper of
the separator, invert and displace the cotton wool plug by means of a
stiff wire passed through the open stopcock. Then add

Water 10 c.c.

Agitate to diffuse the precipitate, and add

Chloroform 7-5 c.n.

Strong solution of ammonia ..... 2 c.c.

Shake well and separate. Eepeat with

Chloroform 7'5 c.c.

and again separate. To the mixed chloroformic solutions in a tared
glass dish (preferably with a flat bottom) add

Amylic alcohol 2 c.c.

(This prevents decrepitation.)

Evaporate on a water bath and dry the residue to a constant weight.
Add 8 mg. to the weight of the strychnine thus obtained (to com-
pensate for strychnine ferrocyanide lost in the wash water) and
multiply the result by 20.

Extract of Nux Vomica. — Two grms. of the triturated extract are
agitated with 5 c.c. of ammonia, 5 c.c. 'of water, and 10 c.c. of
alcohol until solution is effected ; the solution is then shaken with
three portions of chloroform, 20 c.c, 10 c.c, and 10 c.c. The united
chloroform solutions are evaporated or the chloroform distilled off, the
residue warmed upon a water bath for several minutes with 15 c.c. of

r^r hydrochloric acid, then filtered, and the filter thoroughly washed.

The filtrate is titrated with alkali, using cochineal as the in-

100 ^

dicator ; if the number of c.c. of alkali be subtra.cted from 150 (corre-

N
spending to 15 c.c. of y^acid), and the remainder multiphed by 0"00364

(assuming that the alkaloids are present in equal amounts), the product
will represent the total alkaloid present in 2 grms. of extract ; multi-
plying this by fifty will give the percentage.

NUX VOMICA. 577

Separation of Strychnine from Brucine.

Lyons (" Pharm. Keview," 20, 253) separates the brucine from the
strychnine — in the total alkaloids obtained, for example, by the above
process, in the iollowing manner.

Advantage is taken of the fact that while strychnine sulphate is
practically insoluble in 10 per cent H.2SO4, brucine sulphate is very
soluble. Working with an experimental mixture of strychnine, 45,
and brucine, 55, portions varying in amount from 50 to 150 mg. of
total alkaloid were taken and treated with 10 per cent H2SO4 in the
proportion of 1 c.c. to every 10 mg. of alkaloids. After constant agita-
tion for ten minutes, and then at frequent intervals for two hours, the
solution is passed through a small filter, washed with a few drops of
H2SO4 10 per cent ; and the strjchnine sulphate left on the filter, de-
composed with ammonia and extracted with chloroform. In each
case the loss of strychnine was found to be about 1*75 mg. for each
c.c. of acid used.

To determine the proportion of strychnine in the total alkaloid ex-
tracted from nux vomica or its preparations, the above process is thus
conducted : For each 15 mg. of alkaloids 1 c.c. of 10 per cent HgSO^
is added in a capsule and frequently agitated for at least one hour. The
mixture is then filtered, the insoluble residue being entirely transferred
to the filter, and washed with 1 c.c. of acid. The filter and its con-
tents are then replaced in the capsule and treated with 10 c.c. of
CHCI3 and 3 c.c. of 10 per cent ammonia, agitated with a glass stirrer
until all the alkaloid is dissolved and the liquid transferred to a
separator. The filter is then washed with two successive washings,
each of 5 c.c. of CHCI3, which are added to the rest in the separator.
The chloroform solution is then received in a tared capsule, evaporated
to dryness, after adding 2 c.c. of alcohol, dried to constant weight and
weighed. To the weight obtained 1*75 mg. is added for each c.c. of
acid used, the result being the strychnine present in the total alkaloids.

The method adopted in the United States Pharmacopoeia depends
upon the destruction of the brucine by means of nitric acid. The total
alkaloids are extracted in the usual manner from 10 c.c. of fluid ex-
tract, and dissolved in 15 c.c. of 3 per cent sulphuric acid and cooled.
To this is added 3 c.c. of a mixture of equal parts of nitric acid and
distilled water, and the whole set aside for ten minutes. (Farr and
Wright suggest that the mixture be heated to 50° C, which ensures the
complete destruction of the brucine.) The mixture is then placed in
a separator, 25 c.c. of 10 per cent solution of soda added, and the
strychnine extracted with three quantities of chloroform (20 c.c, 10
c.c. and 10 c.c). The chloroform solutions are evaporated to dryness,
the residue is dissolved in 10 c.c of decinormal sulphuric acid and the
solution titrated with fiftieth normal sodium hydrate. Each c.c of deci-
normal sulphuric acid absorbed corresponds to 0*0332 grm. of strych-
nine. It is preferable to use twentieth normal solutions in each case.

This method gives results corresponding very closely with the
method official in the British Pharmacopoeia.

Liquid Extract of Nux Vomica. — This is the principal official pre-
voL. I. 37

678 FOOD AND DEUGS.

paration of nux vomica, and is a diluted alcoholic extract of the drug
which should contain ]-5 grms. of strychnine per 100 c.c. when
assayed by the official process.

A properly prepared extract should have a specific gravity of 0*945
to 0"965 and should contain 11 to 12-5 grms. of solid matter per 100
c.c. It should contain 61 to 63 per cent of alcohol (by volume).

The following is the official process of assay, except that half the
quantities are given, as being more accurate in the result. Evaporate
5 c.c. to a thick syrup on the water bath, dissolve the residue in 10
c.c. of water with gentle heat. Place the solution in a separator and
add 2-5 grms. of sodium carbonate dissolved in 12-5 c.c. of water, and
5 c.c. of chloroform, and then agitate well. Kun off the chloroform
when separated, and repeat the extraction with chloroform twice.
Extract the mixed chloroform solutions with 5 c.c, of about 3 per
cent sulphuric acid three times, and dilute the united acid liquids to
88 c.c. Transfer to a stoppered flask, and add 12-5 c.c. of a 5 per
cent solution of potassium ferrocyanide. Shake well and frequently
for a minute or so at intervals during half an hour. Allow to stand
for six hours, and then collect the precipitate on a small filter, rinsing
out the last portions from the flask with water containing one-fifth per
cent of sulphuric acid, and wash until the washings are free from
bitterness. Wash the precipitate into a separator, and add 2 '5 c.c.
of solution of ammonia (10 per cent). After well shaking add 5 c.c. of
chloroform and shake well then add a further 2-5 c.c. and after well
shaking separate the chloroform and allow the chloroform to evaporate
in a current of warm air in a tared dish, and then dry on the water
bath for an hour, taking care that the dish is covered as otherwise
loss will take place owing to decrepitation. The resulting strychnine
is then weighed.

Extract of Nux Vomica is a semi-solid preparation made by eva-
porating the liquid extract with sugar of milk so that the resulting
extract should contain 5 per cent of strychnine. It is assayed in the
same manner as the liquid extract, except that there is no preliminary
evaporation of alcohol, or in the manner described on p. 576.

Tincture of Nux Vomica is prepared by diluting 2 fluid parts of the
liquid extract with 3 of distilled water, and 7 of 90 per cent alcohol.

It should contain, when assayed by the official process above
described, from 0-24 grm. to 0*26 grm. of strychnine per 100 c.c.

It has a specific gravity about 0*910 to 0*915 and should contain 60
per cent of alcohol by volume, and 1*7 grms. of solid matter per 100 c.c.

Strychnine. — CgiH^gN^Og is one of the alkaloids official in the
British Pharmacopoeia. The official requirements for it are sufficient
to ensure it being of comparative purity. It is described as occurring
in trimetric prisms, soluble in 150 parts of cold water and in 6 parts
of chloroform and in 40 parts of boiling absolute alcohol. It is nearly
insoluble in ether. Sulphuric acid forms with it a colourless solution,
which, on the addition of a crystal of potassium bichromate, acquires
an intensely violet hue, speedily passing through red to yellow. With
sulphuric acid containing 2(Jotj P^^^ ^f potassium permanganate, a
minute particle of strychnine gives a violet coloration. It is not
coloured by nitric acid (absence of brucine) and leaves no ash.

NUX VOMICA. 579

Strychnine melts at 265° to 266°.

The only salt that is oificial in the Pharmacopoeia is the hydro-
chloride CgiHygN^O.^ • HCl, 2H2O. If dried at 100° C, it should con-
tain from 7*3 per cent to 8'8 per cent of water.

The detection of strychnine. The following qualitative tests are
suitable for the detection of strychnine, which is best extracted from
substances supposed to contain it by a mixture of equal parts of
chloroform and ether, after the addition of ammonia : —

(1) A solution of sodium phosphomolybdate in nitric acid pre-
cipitates strychnine as a yellowish amorphous mass from complex
organic liquids. The precipitate (which may consist of other alkaloids)
is separated, and treated with dilute ammonia, and the liquid extracted
with ether-chloroform and tested by the colour reactions given
below.

(2) A solution of strychnine as dilute as 1 in 105,000 gives a red-
brown precipitate with iodine in potassium iodide solution.

Mayer's reagent gives a precipitate in solutions as dilute as 1 in
150,000.

(3) Potassium ferrocyanide precipitates strychnine as a white or
faintly yellow crystalline powder. ■

(4) Strychnine in a very minute quantity, moistened with sulphuric
acid, gives a deep violet colour, (due to an oxidation reaction) when a
small crystal of potassium bichromate is brought into contact with it.
The colour is transient and is rapidly changed and masked by the
green of the chromium salt. A drop of very dilute solution of per-
manganate of potassium gives the reaction with more distinctness.
Lead and manganese dioxide are equally effective. Potassium ferricy-
anide also gives the reaction, but probably the best reaction is ob-
tained with a 1 per cent solution of ammonium vanadate in sulphuric
acid, withwhich the strychnine is moistened. Cerosoceric oxide Ce304
is also very effective. According to Allen the reaction is best ob-
tained as follows : —

The solution of the strychnine in ether-chloroform should be eva-
porated in a porcelain dish. If the quantity of strychnine is likely to
be very small the dish should be immersed in hot water and the
ether-chloroform solution dropped slowly into the dish from a burette
30 as to allow the solvent to evaporate rapidly so as to concentrate the
residue on a small spot. The cold residue should be treated with 2 or
3 drops of. pure concentrated sulphuric acid, which should be mixed
with it with a glass rod. The mixture should be allowed to stand for
five minutes in order to note whether any coloration be produced.
If any marked colour is produced the dish should be gently heated
(not to boiling-point of water) for half an hour, the contents diluted
with water, filtered, made alkaline with ammonia, and the strychnine
again recovered by ether-chloroform, and the solvent evaporated in
the same manner. This residue is again treated with a drop or two
of sulphuric acid. The oxidizing agent (which Allen prefers to be
manganese or lead, dioxide) is then moistened with the sulphuric acid
solution by means of a glass rod, and the mixture stirred. The blue-
violet colour will be at once developed, passing to purple and then to

580 FOOD AND DRUGS.

cherry red, the last tint being fairly persistent. This reaction will
detect .20I06 '^^ ^ grain of strychnine.

(5) If a trace of solid strychnine be dissolved in dilute nitric acid^
the liquid gently heated, and a minute particle of potassium chlorate
then added, an intense scarlet colour results.

R. H. Davies has proposed the following method for the estima-
tion of small quantities of strychnine. The results, however, are
only approximate.

He modifies the chromate test in such a way as to make it applic-
able also for approximate quantitative estimations of traces of this
alkaloid. A very weak solution of potassium bichromate in strong
sulphuric acid is placed in a test-tube, and the solution of the strych-
nine then added to it, when the reaction can be readily observed.
The colour thus produced soon disappears, giving place to a reddish-
orange, which is fairly persistent. This coloration is compared with
those obtained under the same conditions with exceedingly weak
strychnine solutions of various but known strengths. An approximate
idea of the amount of this alkaloid in the solution under examination
is thus arrived at by a colorimetric process similar to Nesslerizing.

Liquor Strychnines hydrochloridi. — The Pharmacopoeia recognizes
a solution of the hydrochloride of strychnine, which contains 17-5
grains of the salt in 4 fluid ounces of alcohol (22-5 per cent by volume).
It should have a specific gravity of about 0*970, and, after evaporat-
ing most of the alcohol, on the addition of ammonia and extraction
with ether-chloroform, should yield strychnine equivalent to 1 grm.
of the hydrochloride per 100 c.c.

Easton's Syrup and Syrup of Hypophosphites. — Under the name
of Easton's syrup, a preparation containing phosphates of iron, quin-
ine and strychnine is largely sold.

The Pharmacopoeia has not recognized the name "Easton," but
contains a preparation of the same nature, under the name " Syrup of
Phosphate of Iron, with Quinine and Strychnine ". This syrup should
contain 1 grain of dry ferrous phosphate, 0-8 grain of quinine calcu-
lated as sulphate and ^^ of a grain of strychnine, per fluid drachm.

The syrup known as " compound syrup of hypophosphites " is a
similar preparation, but contains hypophosphorous acid instead of
phosphoric acid. It is not official in the Pharmacopoeia.

In examining these preparations the iron is determined in the
usual way, on the dried and ignited residue of 5 grains of the syrup.
Free phosphoric acid is determined by titration with caustic soda using
methyl-orange as an indicator.

The alkaloids are best determined by the following process, due in
the main to Harrison and Gair.

The alkaloids are extracted from 150 c.c. of the syrup by diluting
with 250 c.c. of water, adding a little citric acid and ammonia and ex-
tracting with ether- chloroform. The alkaloids obtained on evaporating
the solvent are dissolved in 60 c.c. of water slightly acidulated with
sulphuric acid ; ammonia is added as long as the precipitate redis-
solves. Fifteen grms. of powdered sodium potassium tartrate are
then added gradually with stirring ; then more ammonia, until the

OPIUM AND ITS PEEPAEATIONS. 581

mixture is only just acid to litmus paper, and it is then warmed on
the water bath for about fifteen minutes, and allowed to stand till
quite cold (about two hours). The quinine tartrate is then filtered
off with the aid of a pump, and washed with a solution of 15 grms.
sodium potassium tartrate in 45 c.c. of water, made just acid with
sulphuric acid. The filtrate and washings are mixed, made strongly
alkaline with ammonia, and extracted three or four times with chloro-
form ; the chloroformic solution is washed with 10 c.c. of water, con-
taining a few drops of ammonia solution, evaporated to about 4 or 5
c.c, 10 c.c. of alcohol added, and the mixture evaporated to dry-
ness ; the residual alkaloid is washed three times with 1 c.c. each
time of washed ether, and the washings rejected ; the residue is
practically pure strychnine, and is dried and weighed. The alcohol is
added not only to prevent decrepitation, but also to avoid retention
of chloroform by the strychnine, which otherwise occurs. If the
amount of strychnine in the total alkaloid taken is much over O'l grpa.
it is necessary to increase the quantity of the first solution and of the
Hochelle salt, otherwise the same treatment is employed.

OPIUM AND ITS PEEPAEATIONS.

Opium is the inspissated juice of the unripe fruit capsules of
Fapaver somniferum. It is produced in various countries, notably
Persia, India, China, Asia Minor and Turkey.

Practically the whole of the supply of opium consumed in this
country is derived from Turkey and Asia Minor on the one hand, and
Persia on the other.

So-called Turkey opium is the variety principally employed for
preparations in pharmacy and is the usual " druggist's opium ". It
occurs in rounded or flattened cakes covered with the small triangular
fruits of a species of rumex. Persian opium is the variety princi-
pally used by manufacturers of morphia and other opium alkaloids.
It is imported in conical or more or less brick-shaped masses
wrapped in red paper, and, rarely, in sticks or flat cakes wrapped
in white paper. The characteristic difference in the appearance of
these two varieties is that Turkey opium is granular, whereas Persian
opium is a homogeneous mass.

The Composition of Opium. — The active constituents of opium
consist of a series of alkaloids, the remainder consisting of water, in-
soluble inert matter and extractives and colouring matter. The well-
defined alkaloids of opium are as follows : morphine, narcotine,
codeine, thebaine, narceine, papaverine, meconidine, codamine, laudi-
nine, laudanosine, lanthopine, protopine and several others. The
only ones which are employed to any great extent in medicine are
morphine and codeine, which will be discussed later. The indifferent
matter of opium consists of mucilage, sugar, wax, resins and mineral
matter.

The Adulteration of Opium. — The adulteration of opium may be
divided into two types. In the cakes or lumps, there may be found
from time to time, coarse adulterants such as shot, stones, gravel,

582 FOOD AND DKUGS.

pieces of metal and the like ; but the principal adulterants are
organic matters of various types which are beaten up with the juice
before it is dried. Writing of Indian opium fifty years ago Dr.
Eatwell said : "Flour is a very favourite article of adulteration, but is
readily detected; opium so adulterated speedily becomes sour, it
breaks with a peculiar, short, ragged fracture, the sharp edges of
which are dull and not pink and translucent as they should be ; and
on squeezing a mass of the drug after immersion in water, the starch
may be seen oozing from the surface." The application of the iodine
test, however, furnishes conclusive evidence of the presence of an
amylaceous compound. The farina of the boiled potato is not unfre-
quently made use of; impure treacle is also occasionally used. In
addition to the above, a variety of vegetable juices, extracts, pulps
and colouring matters, are occasionally fraudulently mixed with the
opium ; such are the inspissated juice of the common prickly pear,
and the extracts produced from various other narcotic plants. The
juicy exudations from various plants are frequently used, and of pulps,
the most frequently employed are those of the tamarind and of the
bael fruit.

There is no doubt that opium is still considerably adulterated, but
as its morphia content forms the basis of its value, and practically all
the opium used in this country is sold and used on that basis, ad-
mixture with inert foreign matter is not a very serious matter.

Genuine Turkey opium contains from 8 to 13 per cent — rarely up
to 16 per cent of morphine, calculated on the dry drug. Persian
opium contains in the natural state from 7*5 to 13 per cent of mor-
phine, rarely up to 16 per cent. It is probable that absolutely pure
Persian opium contains from 10 to 15 per cent of morphine as its
usual quantity. Codeine exists in the proportion of 0*3 to 2 per cent,
Persian opium containing considerably more than Turkey opium.
Narcotine exists to the extent of 3 to 8 or 9 per cent, the remaining
alkaloids only forming about 1 per cent of the opium.

The Analysis of Opium. — The analysis of opium is, in practice,
confined almost entirely to the determination of morphine. At times
a fuller analysis is required, however. The figures on opposite page
were obtained in six samples each of typical Turkey and Persian
opiums.

The insoluble residue of pure opium, when examined under the
microscope, should only contain quite small quantities of starch and
only a small amount of the cellular matter of the outer epidermis of
the poppy capsule. Persian opium usually contains more starch
than does Turkey opium.

The Estimation of Morphine. — Many processes have been published
for the determination of morphine, but it is proposed to here give de-
tails of a few which after exhaustive examination have proved to give
good results. The official process of the British Pharmacopoeia,
however, must be described, for, although open to much criticism, it
is the official process which the manufacturing druggist is bound to
employ. This process is as follows : —

Take the following quantities : —

OPIUM AND ITS PEEPARATIONS.

583.

Opium dried at 212° F. (100° C.))
and in No. 50 powder j
Calcium hydroxide, freshly prepared
Ammonium chloride
Alcohol (90 per cent)^
Ether V

Distilled water J

14 grms.

of each a sufficient quantity.

Triturate together the opium, calcium hyroxide, and 40 c.c. of
water in a mortar until a uniform mixture results ; add 100 c.c. of
water and stir occasionally during half an hour. Filter the mixture
through a plaited filter, about 10 cm. in diameter, into a wide-mouthed
bottle having a capacity of about 300 c.c, and marked at exactly 104
c.c, until the filtrate reaches this mark. To the filtered liquid (re-
presenting 10 grms. of opium) add 10 c.c. of alcohol (90 per cent) and
50 c.c. of ether ; shake the mixture ; add the ammonium chloride,
shake well and frequently during half an hour ; set aside for twelve
hours for the morphine to separate. Counterbalance two small filters ;
place one within the other in a small funnel in such a way that the
triple fold of the inner filter shall be superposed upon the single fold

Turkey Opium.

Mineral

Aqueous

Matter Insoluble

Anhydrous

Moisture.

Matter.

Extract.

in HgO.

Morphine.

Per cent

16-4

5-1

49-9

33-7

11-15

20-2

3-9

50-2

29-6

8-56

18-0

3-6

53-6

28-4

9-20

16-9

4-4

46-2

36-9

12-31

17-8

4-8

49-9

32-3

10-97

21-5

4-3

55-0

23-5

9-95

Persian Opium.

10-5

4-8

59-4

30-1

11-25

14-8

3-2

57-9

27-3

9-40

28-2

3-1

550

16-8

8-0

17-6

4-6

57-8

24-6

10-35

18-1

4-8

55-9

26-0

10-80

14-5

60-0

25-5

11-65

of the outer filter ; wet them with ether ; remove the ethereal layer
of the liquid in the bottle as completely as possible by means of a
small pipette, transferring the liquid to the filter; rinse the bottle
with 20 c.c. of ether, again transferring the ethereal layer, by means
of the pipette, to the filter-; wash the filter with a total of 10 c.c. of
ether, added slowly and in portions. Let the filter dry in the air, and
pour upon it the contents of the bottle in portions, in such a way as
to transfer the granular crystalline morphine from the bottle with
morphinated water, until the whole has been removed. Wash the
crystals with morphinated water until the washings are free from
colour ; allow the filter to drain, and dry it, first by pressing between

584 FOOD AND DRUGS.

sheets of bibulous paper, afterwards at a temperature between 131°
and 140° F. (55° C. and 60° C.) finally at 230° F. (110° C.) for two
hours. Weigh the crystals in the inner filter, counterbalancing by
the outer filter. Take 0*5 grm. of the crystals and titrate with deci-
normal volumetric solution of sulphuric acid until the liquid, after
boiling, slightly reddens blue litmus paper. One c.c. of this volumetric
solution represents 0-0285 grm. of pure anhydrous morphine. The
weight of pure anhydrous morphine indicated by the titration + 0-104
grm. (representing the average loss of morphine during the process),
should amount in total to 1 grm., that is to say, to a total of not less
than 0*95 grm. and not more than 1*05 grms., corresponding to about
10 per cent of anhydrous morphine in the dry powdered opium.

Opium is one of the standardized preparations of the British
Pharmacopoeia (see Vol. II, under "standardization"). It directs
that any suitable variety of opium may be used for the preparation
of the tincture and extract of opium, provided that when dry it con-
tains at least 7-5 per cent of anhydrous morphine. But when other-
wise used for officially recognized purposes, opium is to be of such a
strength that when dried and powdered it contains from 9*5 per cent
to 10-5 per cent of morphine. Opium stronger than this is to be
diluted either with a weaker opium of from 7*5 per cent to 10 per
cent strength, or with sugar of milk.

The following process is probably the most accurate for the de-
termination of morphine. It is based on the work of Prollius and
Fliickiger, with modifications by various other chemists. Ten grms.
of the opium are exhausted by digestion for half an hour, after disin-
tegration, with 100 c.c. of water at about 40° C. After the disinte-
grated matter has settled, the liquid is decanted as completely as
possible, and then the solid matter poured on to the filter. After this
has drained well, the solid matter is washed back into the beaker in
which it was digested, and the process repeated. The solid matter
on the filter is now washed with warm water to complete exhaustion,
and the mixed filtrates evaporated to a thin syrup on the water bath.
The liquid is transferred to a flask, the dish being rinsed out with
warm water, and 2-5 c.c. of 90 per cent alcohol are added, and 30 c.c.
of ether. The flask is corked, and the contents are gently shaken to
ensure as complete mixture as possible, when 2-5 c.c. of ammonia
(specific gravity 0-935) are added. The flask is well shaken so as to
cause the morphia to be precipitated in the most convenient form,
and allowed to stand for eighteen hours with occasional shaking.
The liquid is then filtered, preferably with the aid of a pump, and the
precipitate is dried and washed with benzene to remove narcotine
and other alkaloids. It is then dried and weighed, but will contain
up to 10 per cent of impurities, so that the weight is only taken as a
check. It is then titrated with decinormal sulphuric acid using
litmus as an indicator. One c.c. of decinormal sulphuric acid = 0*0303
grm. of hydrated morphine or 0-0285 of anhydrous morphine.

Dott (" Pharm. Journ." [3], 51, 746) prefers the following modi-
fication of this process : —

Ten grms. of the opium in powder are exhausted with spirit of

OPIUM AND ITS PREPARATIONS. 585

-920 specific gravity. One or two drops of solution of ammonium
oxalate are added, and then ammonia, until the liquid is only slightly
acid. The spirit is now evaporated to one-third of its original volume,
allowed to cool, and filtered. The filtrate is concentrated to about 5
C.G., transferred to a small flask, 4 c.c. of water and 3 c.c. of methylated
spirit being used to wash the capsule. 2"2 c.c. of solution of ammonia
(•960 specific gravity) are then introduced, 25 c.c. of ether being
introduced at the same time. The flask is now closed with a well-
fitting cork and shaken so as to mix the contents. After eighteen
hours the ether is decanted as completely as possible, the precipitate
collected on counterpoised filters, and washed with morphinated water.
It is then dried, washed with benzene, dried, and weighed, and finally

N
titrated with ^ sulphuric acid. One c.c. of the acid = -0303 grm. of

hydrated morphine. Although it is not essential, it is preferable to
weigh the morphine before titrating, as an idea is thereby given of
the amount of acid which will be required. This process is only
recommended where the morphine is to be titrated.

Schidrowitz (" Analyst," xxix. 144) has devised a process which
gives very constant results, and which is carried out as follows : —

Six grms. of opium roughly powdered are weighed into a small
porcelain dish, 6 c.c. of distilled water are added, and the whole
allowed to stand for about fifteen minutes. The contents of the dish
are then worked up to a homogeneous consistence with a pestle, and
then transferred (by means of successive small quantities of water)
to a 100 c.c. tared Erlenmeyer flask. The total weight of opium and
water is then made up to 54 grms. The flask, after corking, is shaken
vigorously for five. minutes, and is then allowed to stand for one hour,
with an occasional shaking. The contents are then filtered through
a filter, 10 cm. in diameter, into a second tared 100 c.c. Erlen-
meyer flask. If the filtrate does not run clear at first it must be
returned. When exactly 42 grms. of filtrate have been collected
filtration is stopped. Next, to the 42 grms. of filtrate, exactly 2 grms.
of a 50 per cent solution of salicylate of soda in water is added ; the
whole is then shaken for about half a minute, and immediately
filtered as before. Of the filtrate 36 grms. are collected, and to this is
added 15 c.c. of ether, and, after rotating the flask once or twice,
5'2 c.c. of a solution of ammonia, prepared by mixing 17 grms. of
ammonia (specific gravity 0*960) with 83 grms. of water. The whole
is then vigorously shaken for ten minutes, and the flask and contents
are subsequently kept for twenty-four hours at a temperature of 12°
C. After this, as much of the ether as is possible is poured off through
a filter of 8 cm. in diameter, 15 c.c. of fresh ether is run into the
flask, the latter rotated briskly (but so as to avoid forming an
emulsion), and the ether again poured off through the filter. The
whole of the liquid is then poured through the filter, the greater
part (roughly two-thirds) of the crystals, however, being retained in
the flask. The flask and filter are then washed with three lots each
of 5 c.c. of water saturated with ether, and delivered from a pipette.
Of each 5 c.c, 3 c.c. should be used to rinse the flask, and 2 c.c. run

586 FOOD AND DRUGS.

directly on to the filter. The filter with its contents is removed from

the funnel, folded, and gently but firmly pressed between sheets of

filter-paper. The filter is then opened, and the greater part of the

crystals returned to the flask. Filter and flask are then placed in an

air oven at 55° C. until dry. It is then perfectly easy to transfer the

small quantity of crystals still adhering to the filter to the flask.

N
Subsequently the crystals are dissolved in 25 c.c. ^p] H2SO4, and

N
the excess of acid titrated with -^ alkali, using methyl-orange as

an indicator. It is preferable, prior to this titration, to dilute the

liquid to roughly 50 c.c, and to fix the end-point by means of the

droplet method. The percentage of morphine in the sample is then

calculated as follows : —

N . 1

Let X = number of c.c. j^ acid employed, then x x 0*7575 -I- ^7^

{x X 0-7575) = per cent morphine.

Nagelvort (" American Journal of Pharmacy," November 1900) has
devised the following rapid process which gives fairly accurate
results. Ten grms. are dried at 100° for three hours and powdered.
The powder is transferred to a suitable filter and a mixture of 10 c.c.
of ether and 10 c.c. of chloroform poured over it. The filter is
covered and allowed to drain, and then 10 c.c. of chloroform poured
on to it. After draining, it is dried by exposure to warm air and the
powder is transferred to a flask holding about 120 c.c. To this, 100
c.c. of water are added. The flask is corked and well shaken at fre-
quent intervals during two hours. Fifty c.c. is now filtered off and
shaken with 10 c.c. of 95 per cent alcohol, 20 c.c. of ether and 1 c.c.
of 10 per cent ammonia water. It is then allowed to stand for six
hours. The precipitated morphine is collected on a tared filter, washed
with morphinated water,, pressed between filter paper, dried at 100°
and weighed. The weight multiplied by 20 gives the percentage of
morphine. It is to be noted, however, that no correction is made
for the increase in volume of the 100 c.c. of liquid, due to the soluble
portion of the opium.

Preparations op Opium.

Extract of Opium is a semi-solid extract made by exhausting
opium with water. Analysed by the official method described under
opium, it should yield 20 per cent of morphine ; 7 grms. of the extract
should be used for the assay.

Liquid Extract of Opium. — This preparation is made by dissolving
the extract in water and adding alcohol. It is a deep-coloured
liquid having a specific gi'avity 0*985 to 0*995 (official), and when
assayed for morphine, as described under tincture of opium, should
contain between 0*7 and 0*8 grm. of morphine in 100 c.c. Twenty
per cent by volume of 90 per cent alcohol is used in its preparation.
The final product should contain 18 per cent of alcohol by volume.
The solid residue averages 3 to 3*1 per cent.

OPIUM AND ITS PREPARATIONS. 587

Tincture of Opium. — This preparation is of the same alkaloidal
strength as the liquid extract, but contains more alcohol. It is made
with 45 per cent alcohol, but owing to the extractive matter, and al-
lowing for slight loss during manufacture, the finished tincture
should contain from 42 to 44 per cent of alcohol ; no standards, other
than the morphine content, are given in the Pharmacopoeia. The
extractive matter averages from 3-4 to 3'7 grms. per 100 c.c. The
specific gravity varies from 0-955 to 0*962. The morphine should be
between 0*7 and 0*8 grm. per 100 c.c, when assayed by the follow-
ing process.

Pour 80 c.c. of the liquid into a porcelain dish ; evaporate
on a water bath until the volume is reduced to 30 c.c. ; mix the
residual liquid in a mortar with 3 grms. of freshly-slaked lime ;
dilute the mixture with water to 85 c.c. ; set aside for half an hour,
stirring occasionally. Filter off 50 c.c. of the liquid (representing
50 c.c. of the tincture) through a plaited filter having a diameter
of about 1 decimetre, into a wide-mouthed stoppered bottle, having a
capacity of 200 c.c. ; add 5 c.c. of alcohol (90 per cent) and 30 c.c.
of ether ; shake the mixture ; add 2 grms. of ammonium chloride ;
shake well and frequently during half an hour ; set aside for twelve
hours for the morphine to separate. Counterbalance two small filters ;
place one within the other in a small funnel in such a way that the
triple fold of the inner filter shall be superposed upon the single fold of
the outer filter ; wet them with ether ; remove the ethereal layer of
the liquid in the bottle as completely as possible by means of a small
pipette, and transfer it to the filter ; pour into the bottle 15 c.c. of
ether ; rotate the contents and set the bottle aside ; transfer the separ-
ated ethereal layer carefully, by means of the pipette, to the filter ; wash
the filter with a total amount of 10 c.c. of ether added slowly, and in
portions ; let the filter dry in the air ; pour upon it the liquid in the
bottle, in portions, in such a way as to transfer the granular crystal-
line morphine as completely as possible to the filter. "When all the
liquid has passed through, wash the remainder of the morphine from
the bottle with morphinated water, until the whole has been removed.
"Wash the crystals with morphinated water until the washings are iree
from colour ; allow the filter to drain ; dry it, first by gentle pressure
between sheets of bibulous paper, afterwards at a temperature between
131° and 14.0' R (55° and 60° C), finally at 230° F. (110° C.) for two
hours. Weigh the crystals in the inner filter, counterbalancing by
the outer filter. Take 0*3 grm. of the crystals, and titrate with deci-
normal volumetric solution of sulphuric acid, as directed under opium.

Dowzard (" Pharm. Jour.," [4] 17, 908) has pointed out that in
the B.P. method for determining the morphine in the tincture a
serious mistake has been made. Eighty c.c. of tincture and 3
grms. of slaked lime are used, and the mixture made up to 85 c.c.
This is a very grave blunder, as the volume should only be made up
to 81*9 c.c. Three grms. of slaked lime displace 1*44 c.c. of water.

Linimentum Opii. — This preparation contains half the quantity
of morphine that the tincture does. It should contain about 55 per
cent of alcohol.

588 FOOD AND DRUGS.

Ammoniated Tincture of Ojnum. — This is a weak tincture of opium,
and should contain from 0*1 to 0-12 grm. of morphine in 100 c.c.
The specific gravity of properly prepared tinctures varies from 0-894
to 0'901. It should contain 2-06 grms. of benzoic acid per 100 c.c.
which may be determined as described under paregoric. The solid
residue varies from 2-7 to 2'9 grms. per 100 c.c.

Compound Tincture of Camphor, or Paregoric. — This preparation
is a mixture of tincture of opium, benzoic acid, camphor, aniseed oil,
and diluted alcohol. It should officially contain practically 0*46 mg.
of anhydrous morphine per c.c. but otherwise no official tests are
given. The average specific gravity of this tincture lies between
0*913 and 0*923, and the amount of extractive, dried at 105°, 0*3
to 037 grm. per 100 c.c. The alcohol content varies from 57 to
59 per cent by volume. The amount of morphine present may be
determined by using 250 c.c. of the tincture, evaporating to 6 c.c.
and then continuing the official process described under tincture of
opium using one-fifth of the quantities throughout. It is to be noted
that the statement that the tincture should contain 0*4G mg. of mor-
phine per c.c. which appears in the Pharmacopoeia is incorrect, since
the official tincture of opium from which it is prepared is allowed to
contain from 0*7 to 0*8 per cent, so that the proper limits for this
preparation are 0*43 to 0*49 mg. per c.c.

Bird recommends the following as the best process for the de-
tection of morphine in this tincture.

Compound tincture of camphor, 10 c.c. Evaporate to dryness on
a water bath, take up with dilute alcohol and a minute drop of acetic
acid, evaporate again to dryness, and dissolve the residue in 2 c.c.
distilled water. One drop of this solution tested with Mayer's solu-
tion should give a copious precipitate.

Filter the aqueous solution and wash filter with distilled water.
Transfer to a small separator and extract with hot amylic alcohol and
a few drops of a saturated solution of potassium carbonate. Separate
the amylic alcohol and wash the same with a half c.c. distilled water.
Eepeat the amylic extraction twice and evaporate the mixed amylic
extracts on a water bath to dryness.

The amylic residue from a genuine tincture is at this stage
brownish-yellow, but if no opium is present, nearly colourless.

Dissolve the amylic residue in 2 c.c. distilled water and four drops
of diluted hydrochloric acid. Filter the solution through a small filter,
with a little French chalk to remove colour, until perfectly bright,
and wash filter with distilled water. Extract the clear aqueous solu-
tion in a separator with 4 c.c. hot amylic alcohol and sufficient
powdered ammonium bicarbonate to make alkaline and repeat the
extraction twice with successive 2 c.c. quantities of hot amylic
alcohol. The mixed amylic extracts should be quite colourless and
measure 8 c.c. Evaporate 2 c.c. of the amylic extract to dryness in
a very small glass basin, concentrating the residue on to one spot,
place on a white surface and moisten the residue with a very dilute
solution of neutral ferric chloride. A perfectly distinct dirty blue
coloration characteristic of morphine should appear. Another 2

OPIUM AND ITS PKEPARATIONS. 58^

c.c. evaporated should afford an orange-yellow colour with nitric
acid.

The reactions may be compared with those obtained from 10 c.c.
of a known sample of Tr. Camph. Co. carried through the process at
the same time, when there should be no difficulty in coming to a con-
clusion as to the approximate correctness or otherwise of any sample
in question. The reactions are also given quite distinctly with the
residue from 2'5 c.c. tincture, but when that amount is taken one-
fourth only of the quantities of solvent, etc., mentioned in the pro-
cess must be used throughout.

It is very important that the amylic alcohol be specially redistilled ;
20 or 30 c.c. evaporated in a glass capsule on the water bath should
not leave the slightest residue.

The presence of the proper proportion of alcohol, which is from.
57 per cent to 59 per cent by volume, is practically safeguarded by
the specific gravity, which should lie between 0'913 and 0-923.

The benzoic acid may be determined by rendering 25 c.c. alka-
line with soda, evaporating to 10 c.c, and extracting the last traces
of camphor and aniseed oil with ether. The separated aqueous liquid
is acidified with hydrochloric acid and extracted twice with ether.
If the washed separated ethereal solution be allowed to evaporate
spontaneously the benzoic acid may be weighed ; or better, dissolved
in excess of decinormal alkali, and the excess titrated with standard
acid. Each 1 c.c. of decinormal alkali is equivalent to 12*2 mg. of
benzoic acid.

A qualitative test for the presence of tincture of opium in this
preparation is the reaction produced by the meconic acid always
present in opium, when treated with ferric chloride. The liquid ia
diluted with 60 per cent alcohol until it is of a pale yellow colour,
and a drop or two of ferric chloride solution added. A more or less
deep red colour, due to meconate of iron, is produced

Allen and Scott Smith (" Analyst," xxvii. 350) recommend the
following process for detecting opium in such preparations as paregoric
or cough mixtures. If 25 c.c. of the liquid be rendered alkaline
with caustic soda, and evaporated to about 10 c.c, the alcohol and
a portion of the camphor and oil of anise if present will be volatilized,
and the amount of alcohol can be deduced with sufficient accuracy
from the specific gravity of the distillate. On shaking the residue with
ether, the remaining camphor and oil of anise will be extracted.
If the ether be separated, and the aqueous liquid acidulated with
hydrochloric acid, benzoic acid will in some cases be separated ; but
whether it separates or remains in solution it can be dissolved out by
agitating the acidified liquid with ether. On allowing the separated
ethereal solution to evaporate spontaneously in a small beaker, the
benzoic acid is obtained in a state fit to weigh ; but a better and more
rapid plan is to repeatedly agitate the ethereal liquid with water
until the washings no longer redden litmus, add a little more water and
a few drops of phenol- phthalein solution, and titrate the liquid with

— caustic alkali (preferably baryta- water), which should be added

590 FOOD AND DEUGS.

until the aqueous layer acquires a pink colour, not destroyed by

N
agitation with the ether. Each 1 c.c. of ^^ alkali required represents

0*0061 grm. of benzoic acid. If 25 c.c. of the liquid has been em-
ployed, the number of mgs. of benzoic acid found, multiplied by 0*35,
gives the grains of benzoic acid per pint of the liquid.

The detection of meconic acid proves the presence of opium in the
tincture. When this information alone is sought the liquid may be
diluted in a test-tube with 60 per cent alcohol till it is of a light
yellow colour, and a drop or two of solution of ferric chloride then added.
If opium be present, a more or less deep red coloration will be pro-
duced, owing to the formation of iron meconate. By comparing the
depth of red colour with that given by a standard tincture a rough
indication of the amount of opium present may be obtained.

Unmistakable confirmatory evidence of the presence of morphine
in cough mixtures may be obtained by obtaining a microscopic pre-
paration of its typical crystals in the following manner : A portion of
the amylic alcohol alkaloidal extract is shaken out with a little dilute
acetic acid, a few drops of the aqueous acetate solution are put in a
watch glass or a celled microscope slide, covering it with another
watch glass moistened with strong ammonia, and allowing to stand
for half an hour. If morphine be present, the characteristic elongated
prisms of the crystalline base will be detected on examining the
liquid under the microscope.

The remaining galenical preparations of opium need not be dis-
cussed here, as they present no special individual features.

Morphine Cj-HjgNOg.HgO is the principal alkaloid of opium, as
mentioned above. It is only official in the British Pharmacopoeia in
the form of three of its salts, the acetate, hydrochloride and tartrate.

It crystallizes in fine colourless trimetric prisms, containing one
molecule of water, which is lost slowly at 90° to 100°. It dissolves in
about 33,000 parts of water at 0° ; 4500 parts at 10° and about 2500
parts (according to Dott) at 15°. In boiling water it is soluble to the
extent of about 1 in 400. It dissolves in 100 to 150 volumes of
90 per cent alcohol at 15°. It is very slightly soluble in chloroform,
ether, and benzene. The solutions are alkaline and slightly laevo-
rotatory. When pure it yields practically no colour with cold sulphuric
acid ; with nitric acid an orange-red colour changing to yellow is pro-
duced. Sulphuric acid containing 0*4 per cent of formaldehyde gives
a purple colour. A solution in very dilute acids gives a copious pre-
cipitate with potassio-mercuric iodide and other alkaloidal precipitants.
Traces of morphine mixed with a solution of starch and evaporated
give a blue coloration with a spot of iodic acid. If a solution of
morphine be heated with an aqueous solution of potassium ferro-
cyanide containing a drop of neutral ferric chloride solution, a deep
blue colour is developed, and on standing, a blue precipitate.

The following are the official salts of morphine : —

Morphine acetate Cj-HipNOg, C^H^Oo, 3^,0, contains about 71*7
per cent of alkaloid. It is required to have the following characters
by the British Pharmacopoeia : —

OPIUM AND ITS PREPARATIONS. 591

A white crystalline or amorphous powder, almost entirely soluble
in two and a half parts of water and in about 100 parts of alcohol
(90 per cent). It loses acetic acid when exposed to the air. It
affords the reaction for morphine mentioned under " Morphinae Hydro-
chloridum " and the reactions characteristic of acetates. Two grms.
of the salt form with 6 c.c. of warm morphinated water a slightly
turbid solution, which is rendered clear by the addition of O'l c.c.
of acetic acid, and this solution, when mixed with solution of
ammonia in slight excess, yields a precipitate which, after washing
and drying, as described under " Morphinae Hydrochloridum," weighs
1*42 grmsi If the salt yield a larger proportion of morphine than
this, it should be recrystallized from hot water acidulated with
acetic acid. Heated to redness with free access of air, it leaves no
residue (absence of mineral impurities).

Morphine hydrochloride Cj^^H^gNOg, HCl, SHgO, contains about
76 per cent of alkaloid and is officially required to have the following
characters : —

Acicular prisms of a silky lustre, or a white powder consisting of
minute cubical crystals, unchanged by exposure to the air. Soluble
in 24 parts of cold water, 1 part of boiling water, and in 50 parts of
alcohol. It should be without action on litmus. Solution of am-
monia causes a white precipitate in the aqueous solution with diffi-
culty soluble in excess ; solution of potassium hydroxide a similar
precipitate readily soluble in excess. This precipitate yields mere
traces to benzol (absence of other alkaloids). Moistened with nitric
acid the salt yields an orange-red coloration; with test solution of
ferric chloride a dull greenish-blue coloration. Heated on a water
bath for ten or fifteen minutes with a few drops of sulphuric acid,
cooled, and treated with a few drops of diluted nitric acid, it gives a
violet colour rapidly passing to blood-red. It dissolves without colora-
tion in strong sulphuric acid; the addition of a small quantity of
sodium arsenate to a portion of this solution causes a bluish-green
coloration, and a small quantity of bismuth oxynitrate added to an-
other portion gives a purplish-brown coloration. It affords the reac-
tions characteristic of hydrochlorides. Two grms. of morphine
hydrochloride dissolved in 250 c.c. of warm morphinated water, with
solution of ammonia added in the slightest possible excess, will give on
cooling a crystalline precipitate which, when washed with a little cold
morphinated water and dried, should weigh 1-51 grms. The drying
should be accomplished, first by pressing the precipitate between
sheets of bibulous paper, then by exposing it to a temperature between
131° and 140° F. (55° and 60° C), and finally to a temperature of
230° F. (110° G.) for twenty minutes. Heated to redness with free
access of air, it burns, leaving no residue (absence of mineral im-
purities).

Morphine tartrate (C^jR^gNO^)^, G^B.qO(^, SHoO, contains about 74
per cent of alkaloid, and is officially required to have the following
characters : —

A white powder consisting of fine nodular tufts of minute acicular
crystals. Efflorescent at 68° F. (20° C). Soluble in 11 parts of cold

592 FOOD AND DRUGS.

water, almost insoluble in alcohol (90 per cent). It affords the re-
actions characteristic of morphine and of tartrates. Two grms. dis-
solved in 20 cubic centimetres of warm morphinated water, with
solution of ammonia added in the slighest possible excess, will give,
on cooling, a crystalline precipitate, which, after washing and drying
as described under " Morphinae Hydrochloridum " should weigh 1-47
grms. Heated to redness with free access of air, it burns without
leaving any residue (absence of mineral impurities).

The Detection of Morphine. — In the solid state morphine is readily
identified by the following reactions : —

A minute fragment, moistened with a drop of neutral* solution of
iron alum, gives a characteristic greenish-blue colour, which is de-
stroyed by free acids or by heat. Contact with strong nitric acid gives-
an orange-red colour, changed to yellow on standing, and destroyed
by the addition of a few drops of solution of sodium thiosulphate. If
a fragment be dissolved in sulphuric acid and a few drops of a solu-
tion of sodium or ammonium molybdate be added, a fine violet colour
is produced, changing to blue-green and finally disappearing. This re-
action is best applied by dissolving 5 mgs. of molybdate of ammonium
in 1 c.c. of strong H2SO4 (Frohde's reagent) and adding a few drops
of this to the solid morphine. If a fragment be dissolved in HgSO^,
heated to 100° and a fragment of potassium perchlorate added, a deep
red-brown colour is produced which rapidly spreads through the
liquid. This is a very characteristic reaction. If a fragment of mor-
phine be mixed with a little cane sugar, a drop of concentrated sul-
phuric acid will produce a beautiful purple colour. If a solution of
morphine be saturated with sugar and the liquid poured into con-
centrated H2SO4 a purple or rose-red colour appears at the junction
of the liquids. The addition of a drop of bromine water after the
sulphuric acid increases the delicacy of this reaction. A fragment
moistened with sulphuric acid containing 0*4 per cent of formaldehyde
giv^s a fine purple colour.

Morphine in solution is precipitated by alkalis or alkaline car-
bonates but is redissolved by excess, except by alkaline bicarbonates.
In solutions free from interfering substance, morphine can be precipi-
tated by sodium bicarbonate, collected, washed with morphinated
water, and weighed, and thus quantitatively determined. In complex
solutions morphine is liberated from its salts by the use of an alkali,
and the solution Well shaken for a long time with hot amyl alcohol.
The solvent should first be added and then the alkaline bicarbonate,
as, if the morphine be allowed to crystallize, it is very difficult to dis-
solve in the solvent. Alter separation, if a determination be required,
the process should be repeated three times, and the amyl alcohol eva-
porated at 100°, and the morphine weighed or titrated with dilute
sulphuric acid using methyl-orange as indicator. Or, better, the amyT
alcohol is repeatedly shaken with very dilute hydrochloric acid, and the
acid solution precipitated by sodium bicarbonate. If only a quantitative
reaction is desired, the hydrochloric acid solution is concentrated and
a few drops of a mixture of ferric chloride and potassium ferricyanide
added, when prussian blue is formed. A solution of iodine in hy-

OPIUM AND ITS PKEPARATIONS. 695

driodic acid gives a crystalline precipitate in very dilute solutions of
morphine. A very delicate test, but one which is merely confirmatory,
since many other bodies give it — is as follows : If a dilute solution
of morphine be mixed with a few drops of starch solution and eva-
porated to dryness, the residue, moistened with iodic acid, will give a
blue colour if as little as airioo of a grain of morphine be present.

Codeine. This alkaloid, Ci7Hig(CH3)N03 . H.^O, is methyl-morphine
and may be obtained from opium, or by the methylation of morphine.
It is thus described in the British Pharmacopoeia : —

In colourless or nearly colourless trimetric crystals soluble in
80 parts of water or of solution of ammonia, readily soluble in
(90 per cent) chloroform, and in diluted acids. It is soluble in
30 parts of ether. The aqueous solution has a bitter taste and
an alkaline reaction. The alkaloid dissolves in an excess of sulphuric
acid, forming a colourless solution, a small quantity of which, when
gently warmed on a water bath with 2 drops of solution of ammonium
molybdate, or with a trace of ferric chloride or potassium ferricyanide,
develops a blue or bluish-black colour,- which on the addition of a
minute trace of diluted nitric acid, changes to a bright scarlet, be-
coming orange. Heated to redness in air it yields no ash. Moistened
with nitric acid the liquid becomes yellow but not red. A 2 per cent
solution of codeine in water acidulated with a few drops of hydro-
chloric acid, gives a whitish precipitate with solution of potassium
hydroxide, but not with solution of ammonia. A saturated solution
of codeine in water acidulated with hydrochloric acid, should give no
blue colour, but only gradually a dull green, on the addition of test
solution of ferric chloride and a very dilute solution of potassium
ferricyanide (absence of morphine and other impurities).

[It is to be noted that codeine frequently gives a green colour with
cold sulphuric acid. This, however, is due to the presence of a trace
of selenium as an impurity in the acid.]

Codeine is sharply differentiated from morphine by its ready
solubility in ether and chloroform, by which solvents it can be ex-
tracted from its solutions when rendered alkaline. When warmed with
sulphuric acid, codeine (and other bodies also) gives a blue colour, in
the presence of a trace of any oxidizing agent such as arsenic acid.
A fragment, treated with two drops of sodium hypochlorite solution
and four drops of sulphuric acid, gives a fine blue colour.

Claassen (" Jour. Chem. Soc." 58, 1198) proposes estimating
codeine when it exists in the free state in neutral solutions, by allow-
ing it to decompose morphine sulphate, to a solution of which it is
added. The precipitated morphine, multiplied by 0-9868 represents
the codeine.

Codeine phosphate (Ci-Hi8[CH3]N03 . HgPOJg, ^HgO, is thus de-
scribed in the British Pharmacopoeia : —

White crystals which have a slightly bitter taste. It is soluble
in 4 parts of water, much les soluble in alcohol (90 per cent). A 5
per cent aqueous solution has a slightly acid reaction, and yields a
whitish precipitate with solution of potassium hydroxide, but not with
solution of ammonia. It affords the reactions characteristic of codeine
VOL. I. 38

594 FOOD AND DRUGS.

and of phosphates. It loses its water of crystalHzation when dried at
212° F. (100° C), and at a higher temperature melts, forming a
yellowish-brown liquid. It should yield no characteristic reaction
with the tests for chlorides or sulphates. It should not be coloured
blue by test solution of ferric chloride (absence of morphine).

Ajjomorphine hydrochloride C^-Hj-NO^. HCl is an alkaloid obtained
by heating morphine or codeine hydrochloride in a sealed tube to
150° with excess of hydrochloric acid or with zinc chloride. Apomor-
phine hydrochloride is official in the British Pharmsecopoeia, which ,

describes it as follows : —

Small, greyish-white, shining, acicular crystals, turning green on
exposure to light and air. Inodorous. Soluble in 50 parts of water
and more soluble in alcohol (90 per cent), the solutions being decom-
posed with production of a green colour when they are boiled.
Neutral or very feebly acid to solution of litmus. From solutions,
solution of sodium bicarbonate throws down a precipitate which be-
comes green on standing and then forms a solution which is purple
with ether, violet with chloroform, and bluish-green with alcohol (90
per cent). With dilute test-solution of ferric chloride it gives a deep
red, and with nitric acid a blood-red coloration. If the salt imparts
an emerald-green colour to 100 parts of water, after shaking the mix-
ture, it should be rejected.

A solution containing 1 part in 100,000 will yield a green color-
ation when rendered faintly alkaline with potassium bicarbonate and
■exposed to the air.

Dr. Hasse has shown that much of the so-called " Apomorphinum ^

Hydrochloricum " on the market is not apomorphine hydrochloride, p

ibut a hydrochloride of trimorphine, and only contains traces of apo-
morphine. Frerichs has investigated this substance and found that
the " apomorphine hydrochloride " in question did not conform to the -

'German Pharmacopoeia requirements owing to the fact that it is not *

^crystalline, and instead of dissolving in 40 parts of water only its
own weight of water was required for solution. Dr. Frerichs shows
that as regards the other Pharmacopoeia tests the spurious preparation
does not differ noticeably from the authentic. In the case of pure
apomorphine hydrochloride the solution darkens more rapidly with
soda solution than in the case of the spurious hydrochloride, and the
deposit obtained with sodium bicarbonate from a solution of true
apomorphine hydrochloride quickly assumes a green shade, while
this is hardly perceptible in the case of the spurious. A solution of
the precipitated alkaloid in ether or chloroform is decidedly coloured
in the case of the authentic apomorphine, whereas in the case of
the spurious preparation it is scarcely tinged. Dr. Frerichs considers
that the new edition of the Pharmacopoeia should distinguish by tests
between the trimorphine and apomorphine salts, and he gives the
following test as suitable for the purpose : —

Place 10 eg. of the apomorphine hydrochloride on a small dry
filter and pour over it 5 c.c. of a mixture of hydrochloric acid 1 part
and water 4 parts. To the filtrate add potassio-mercuric iodide
solution. Pure apomorphine hydrochloride gives at the most an

PODOPHYLLUM.

595

opalescent turbidity, but if other alkaloids are present the hydro-
chloric acid filtrate gives a distinct precipitate with potassio-mercuric
iodide.

According to Harnack and Hildebrand, however, this impurity is
probably yS-chloromorphide, with, at most, traces of trimorphine.

As little as 10 per cent of trimorphine hydrochloride in the apo-
morphine salt is detectable by this test, as also are morphine and
cinchona alkaloids.

PODOPHYLLUM.

The official drug of the Pharmacopoeia is Podophyllum peltattim,
which is used as the source of manufacture of podophyllum resin.
It will be advisable, however, to discuss the allied drug. Podophyllum
emodi, which will probably be rendered official in the next edition of
the Pharmacopoeia. The drug consists of the dried rhizome and roots.
The resin is obtained by exhausting the powdered drug with 90 per
cent alcohol, recovering the bulk or the alcohol and then precipitating
the resin by pouring the remaining liquid into water rendered acid
with HCl. The precipitated resin is collected and dried at a tempera-
ture not exceeding 38° C. It is described officially as soluble, or nearly
so, in 90 per cent alcohol and in ammonia, but not in acid liquids. It
should not yield more than 1 per cent of ash. [The root yields an
average of 5 per cent of ash.] The value of the drug lies in the
resinous matter present. Dunstan and Henry {" Proc. Chem. Soc."
1898, 189) have examined the constituents of Podop)hyllum peltatum
(American podophyllum) and of Podophyllum emodi (Indian), and
found them to be identical. The principal constituent is podophyllo-
toxin, a neutral crystalline substance of the formula Cj^H^^O^, melting
at 117°., first isolated by Podwyssozki and Kiirsten. An uncrystalliz-
able resin, podophylloresin is also present. The product known
commercially as podophyllin is the mixed resinous matter of the drug.
American podophyllum, the official variety, contains 4 to 5 per cent,
whilst the Indian variety contains 9 to 12 per cent. According to
Dunstan and Henry the cpystalline podophyllotoxin is present to the
extent of about 1 per cent in the American, and from 2 to 5 per cent
in the Indian drug. According to Umney the following represents
the average characters ot the resins of the two drugs : —

P. Emodi.

P. Peltatum.

Kesin, by official process for podophyllin

resin . . . ...

Constituents of the resin—

Podophyllotoxin (crude) ....

Pure crystalline picropodophyllin .

Picropodophyllic acid ....

Podophyllic acid .

Podophylloquercetin ....

Fatty matter

Per cent
11-4

17-8

2-6

not determined.

30-8

1-3

2-3

Per cent

5-9

33-8
4-5
not determined.
6-9
2-4
5-7

596

FOOD AND DRUGS.

1 . 1 ;^

■ S-l •

Vl ■

^ «3

a n P oj
? o p p.

§^g

a §

$. S a

fl p'

O o

t: «

o S

-^Z^-^ 8

.P 8^3

"^ 1 f^

g^-^

1 .^ "^ 1

o §

21< 2

Ti W

^ tens

OCr^

]

ce 2 <s r^j

ce "^ oe

03 ' eS t3 OS 05

' oi 0) '

.S.S.S

2 «

« m P

^■*^^J

H^H

* H

Sh-

ittle brown

mass,
mixture of
nish masse
d a yellow
powder.

tnixture of
nish masse
d a yellow
powder.

-= 1

mixture of
nish masse
d a yellow

m o 1

••i9qm

uni9ioj:}^9j qq.iAV

»0 lO

«5

CO M3

»o

paqsBAV lo^-^xg

^ §

U5 -*

S '

co-^ 1

jaqi^a JO o'd

.jj

s .
•11.

= =

ill

^ p

.p °

> P

.1?

si '

^ rtJ

oj .S

^
M

1 a
1-^

•jaq^a oanj

CN «

"^ 1

ui aiqnpsui 'O" j;

^ g

00 »o

(M CO

;3 1

sg 1

"tS

"K os

>.4

J ll .

= P V ^ -

§^»

1 §

« s

■J9qm

uinaioa^jaj q^iAv

•«5

»o

«5

•CO .CO

»o

paqs^AV ^oiB-i;xa

SO CO

o t>

^ ^

■^

(NO CO
■* lO 00

CS
CO

uijojojoiqojo-o-j

•mjoioaoTuo

•CO

C5 -co

CO 00 -co

•CO

lO ■*

Tt* CO

■>* CO

C<]

CO iH rH

ui aiqniosai -o j

Tj< «5

CO CO

"* CO

■^ Tt* lO

»o

"Biuouiuiy

JO uoi:^nios
UI aiqniosui -o-j

1 5

iH CO

s 1

qoN 1
OO '

•o-d 06 loqooiv

G^l

1 "^

•CO

""^

-* 1

UI aiqniosui -o- j

O (N

1 (M

CO iH

rW CO 1

r-l

•qsyjo-o-j

Tt( -*

Tj* -5t(

Tl* ■<*

»o

•CO

Oi CO '^

O O

o o

O iH rH

Very dull
yellow.

Dull light
yellow.

Dull Ught

yellow.

Dull greyish-

green.

Deep bright

yellow.

Dark greenish-

yellow.
Dull greenish-
yellow.

Brown.

Greenish-

yellow.

Light bright

O ^ O

lit

JP
CO

•aidiuisg

iH (M

CO '^

U5 CO

!>■

^S ;^

r-l

00
rH

EHUBARB. 597

Apart from the official tests given above, a genuine podophyllum
resin should yield from 30 per cent to 40 per cent of podophyllotoxin,
when examined in the following manner : The resin should be ex-
tracted with cold chloroform, and the bulk of the chloroform evapor-
ated. The liquid remaining is poured into petroleum ether and
the precipitate washed with petroleum ether, dried and weighed
(Kremel, " Year-Book of Pharmacy," 1889, 180). The Indian resin
is stated by Verney to contain a smaller percentage of podophyllo-
toxin.

About 50 per cent of the resin dissolves in chloroform.

Taylor ("Pharm. Journ." 4, 16, 368) has made a number of analyses
of typical samples and considers that : (1) The ash limit of 1 per cent
is reasonable (high ash values indicate the use of alum in precipitating
the resin, in order to get it of good colour). (2) The solubility in
ammonia is not of much value. (3) That at least 95 per cent should
dissolve in 90 per cent alcohol. (4) That 50 per cent should be
s::)luble in chloroform. (5) That at least 40 per cent of crude podo-
phyllotoxin should be obtained by precipitation of the chloroformic
solution with petroleum ether. (6) That 60 per cent should be sol-
uble in pure ether, and that the residue should consist of a resinous
and sticky substance.

The table on opposite page embraces Taylor's examination of
thirteen typical samples.

Millard ("Pharm. Journ." 4, vi. 304) gives the following as a reli-
able means of discriminating between the American and Indian
resins : —

0-4 grm, of resin is mixed with 3 c.c. of alcohol of specific gravity
0*920 and 0*5 c.c. of liquor potassas. The test tube is shaken, and in
the case of the Indian resin, the mixture becomes semi-solid, so that
the tube can be safely inverted. If solidification does not take place
at once, it will do so on boiling and then cooling the liquid. The
American resin gives a dark fluid under the same conditions.

Tmckire of Podophyllum is a solution of 320 grains of the resin in
90 per cent alcohol to make 1 pint. It should have a specific gravity
from 0*844 to 0*848. If made with a well-prepared resin, it should
yield 3*5 grms. of solid residue per 100 c.c, but many samples do not
contain more than 3 grms. per 100 c.c The amount of alcohol should
not be below 86 per cent by volume.

EHUBAEB.

The rhubarb official in the British Pharmacopoeia is the rhizome
of Bheum palmakim, Bheum officinale, and probably other species ;
collected in China and Tibet.

No standards of purity are given in the Pharmacopoeia, other than
external appearance.

Ehubarbs grown in Europe are usually from other species and are
quite different in their constituents, and are therefore not official.

The constituents of rhubarb which are of importance have been

598 FOOD AND DEUGS.

described from time to time by different chemists, and their descrip-
tions are very discordant.

A careful examination of the most reliable researches shows that
these discrepancies are probably due to the fact that rhubarb contains
a very unstable substance which easily breaks down into simpler
compounds.

This substance is known as rheopurgin, which decomposes into
four glucosides.

These are bodies which yield a sugar on hydrolysis, and a deriva-
tive of anthraquinone in each case.

They are as follows : —

(1) Chrysophanein C21H20O9, which yields chrysophanic acid on
hydrolysis as follows : —

C21H20O9 + HP = C,,H,oO, + C,Hi,0,.

According to Gilson pure chrysophanic acid melts at 195" to 196°-

(2) Kheochrysin C.^^H^Piq, which yields rheochrysidin (formerly
known as isoemodin or rhaberone) as follows : —

C,2H,Pio + H,0 = Ci,H,A + CeHiA-

(3) A glucoside of unknown characters which readily yields
emodin C^gH^Pg melting at 256°.

(4) Another unknown glucoside which readily yields rhein CjjHgOj.
melting at 314°.

In addition to these bodies there are several astringent bodies of
which several are glucosides yielding gallic acid on hydrolysis.

As this work is going to press, an important paper on this subject
has just been read at the meeting of the Chemical Society held on
6th April, 1911, by Tutin and Clewer, to which reference should be
made in the Society's Journal.

The analysis of rhubarb has I een the subject of numerous experi-
ments, but no process for the determination of the active principles
can be said to be very satisfactory. The determination of the moisture
and mineral matter and the microscopic examination are the most
useful methods available, together with an approximate determination
of the emodin and chrysophanic acid of the root and special tests
for such adulterations as turmeric and added oil. A compound
tincture is official. Its characters are given in the table on p. 496.

Moisture and Ash. — The moisture in normal powdered rhubarb
does not exceed 8 per cent to 9 per cent, and the ash varies from 5
per cent to 12-5 per cent, the high ash being due to the presence of
a considerable amount of calcium oxalate in the drug.

Extractive. — Genuine rhubarb should yield not less than 33 per
cent of dry extractive matter to 45 per cent alcohol.

The Glucosides. — The separation of the glucosides, which is rarely
necessary in actual practice, may be effected in the following manner :
the rheopurgin is extracted by percolation with a mixture of 5
volumes of methyl alcohol and 95 volumes of ether. As extraction
proceeds, the amount of methyl alcohol in the solvent liquid is gradu-
ally increased up to 40 per cent. The percolate is concentrated

EHUBAEB. 599

when a yellow crystalline powder commences to be deposited. To
avoid decomposition, the concentration must now go on in vacuo.
The deposited yellow powder is washed with a mixture of methyl
alcohol (1 volume) and ether (3 volumes), and finally with pure ether,
and dried in vacuo at ordinary temperature. Twenty grms. of the
crude substance are macerated for 5 days with 1 litre of 2 per cent
sodium carbonate solution. It is then filtered and the insoluble
portion again treated with 750 c.c. of the same solution for half an
hour. After standing in a cool place for twenty-four hours it is
filtered. The filtrates contain the rhein and emodin, and are washed.
On acidifying with dilute H2SO4 and warming for a short time on the
water bath, rhein and emodin are precipitated. From the washed
and dried precipitate emodin is extracted by boiling chloroform, and
rhein is extracted from the residue by boiling pyridine and recrystal-
lized from methyl alcohol.

The portion insoluble in Na.jCOg solution is then digested at 70"
for forty-five minutes with the same solvent, and filtered hot. After
cooling the filtrate throws down a precipitate, which is collected,
washed and dried, and kept separate. The filtrate is then acidified and
the pracipitate collected. The melting-points of the products of
hydrolysis of the two precipitates are determined and those melting
from 184° to 186° are put together ; and those melting at 199° to 201°
are put together ; and each bulking is again treated with hot sodium
carbonate solution as before, and so on until no further separation
can be effected. The two bodies are recrystallized from 90 per cent
alcohol, when the chrysophanein melts at 248° to 249°, and yields
on hydrolysis chrysophanic acid melting at 193°. The rheochrysin
yields rheochrysidin melting at 204°.

The determination of emodin (which is approximate only) is best
determined by the following colorimetric method, the standard colour
being that given by 0*001 grm. of pure emodin obtained from aloes,
dissolved in 1 litre of water rendered slightly alkaline with KOH.

This has a pale rose colour. 0*5 grm. of rhubarb, in very fine
powder, is boiled for fifteen .minutes, under a reflux condenser, with
50 c.c. of 50 per cent H^SO^ ; the anthra-glucosides are thus hydro-
lysed and anthraquinone derivatives set free. When cold, the liquid,
without filtration, is shaken out with successive 50 c.c. of ether until
that solvent is no longer coloured and does not give a rose colour
when a portion is tested with KOH. The separated aqueous liquid
is again boiled for fifteen minutes, cooled, and again shaken out with
ether. The bulked ether extracts are then shaken out with succes-
sive washings of 5 per cent KOH solution until a rose tint is no
longer obtained. The bulked alkaline liquid is then made up to 500
c.c. One hundred c.c. of this solution is diluted to 1 litre ; the colour
is then matched against that of the standard emodin solution, on a
white surface, in the usual manner. The tint of the rhubarb solution
will generally be too dark ; it must therefore be diluted with a known
volume of water.

The method advocated by Tschirch and Edner ("Archiv der
Pharm." 245, 150) for the approximate determination of chrypophanic

600

FOOD AND DEUGS.

acid gives good results. From 0*5 grm. to 1 grm. of the rhubarb in
fine powder is exhausted by boiUng several times with 5 per cent
alcoholic potash solution. The alkaline liquids are then distilled to
remove nearly all the water, and the residue is slightly diluted with
water and rendered acid with HCl. The precipitate is washed with
slightly acidified water and dried, and then extracted with chloroform in
a Soxhlet tube. The oxymethyl-anthraquinone derivatives are thus
removed. The chloroform is recovered and the residue dissolved in
10 c.c. of 5 per cent solution of NaOH, and diluted with water to 50
c.c. A solution of p-diazonitraniline is prepared by shaking 5 grms.
of p-nitraniline with 25 c.c. of water and a little strong sulphuric
acid. Another 100 c.c. of water and 3 grms. of NaNO.j in 25 c.c. of
water are then added. The whole is then made up to 500 c.c.
Twenty c.c. of this reagent are then added to the alkaline rhubarb
extract and well shaken, and hydrochloric acid is added drop by drop
until the red colour is discharged and an acid reaction obtained. The
liquid is set aside for four hours, and the precipitate collected on a
rated filter, washed with water, dried at 70" and weighed. 4-47 parts
of the precipitate are representative of 2*54 parts of chrysophanic
acid (or say 4*5 = 2*5 parts). According to the most reliable results,
the following are the amounts of chrysophanic acid and emodin
present in various types of rhubarb : —

Chinese
English
French
Austrian

Chrysophanic Acid.

Per cent
2-5 to 4-3
1-5 „ 1-9
1-0 „ 1-5
0-9 „ 1-6

Emodin.

Per cent
1-8 to2-8
0-5 „l-5
0-4 „l-5
0-5 „l-5

Oil. — Not more than 0*3 per cent of fat should be present in pure
rhubarb ; if, on extracting with ether in a Soxhlet tube, more than this
be found, it is practically certain that a little oil has been added in
order to improve the colour of the powder.

The Detection of Turmeric. — One grm. of powdered rhubarb is
shaken for a few minutes with 10 c.c. of chloroform, and the mixture
is filtered. The filtrate is agitated with 15 times its volume of
petroleum ether and the mixture divided into 2 parts, one of which is
shaken once or twice with 2 c.c. to 3 c.c. of pure strong sulphuric
acid, while the other is shaken with 1 c.c. to 1-5 c.c. of saturated
solution of borax. If the sample be pure, the original chloroform
solution will show a pale, straw-yellow colour, which disappears on
mixing with the petroleum ether. The treatment with sulphuric acid
will impart a pale brown colour to the latter, while the supernatant
liquid remains colourless. The treatment of the second portion with
strong borax solution should produce no change in colour. If, how-
ever, the sample under examination was adulterated with turmeric,
the following reactions will be obtained : the chloroform solution

RHUBAKB.

601

will show a yellowish-brown colour and a well-marked greenish
fluorescence. The addition of petroleum ether will cause the forma-
tion of a yellow flocculent precipitate in the chloroform solution, while
the yellow colour of the liquid and the green fluorescence remain un-
changed. The mixture of chloroform solution and petroleum ether,
when shaken with sulphuric acid as stated, will change to violet,
while the acid itself will assume an intense red coloration, changing
rapidly to reddish-brown and then slowly to yellowish-brown. The
agitation of the second portion with borax solution will cause the
latter to turn violet, the upper layer remaining unchanged.

Microscopic Examination. — On examination under the micro-
scope, characteristic star-like aggregations of calcium oxalate are to
be seen, and small starch grains somewhat resembling those of the pea

Fig. 55. — Powdered rhubarb.

or bean, with a strongly marked hilum. Large reticulated vessels
and thin parenchymatous cells containing a few starch grains are
:also to be found. Added starchy matter will be detected by the
character of the starch grains. The presence of turmeric is indicated
by irregular masses of gelatinized starch, and the universally distri-
buted yellow colouring matter, changed to a deep red by sulphuric
acid diluted with an equal volume of alcohol, in which the red
)ur dissolves.

602

FOOD AND DKUGS.

STEAMONIUM.

Both the dried leaves and the dried ripe seeds of Datura stra-
moniuvi are official in the Pharmacopoeia, as well as a tincture of the
former and a semi-solid extract of the latter. No standards are given
for any of these.

The leaves contain about 0*25 (from 0-15 to 0-32 per cent) of alka-
loids, consisting principally of hyoscyamine, with some atropine and
hyoscine. The seeds contain about the same amount of alkaloids
and in about the same proportions. Hyoscyamine and atropine are
described on pp. 520, 521 ; hyoscine is described on p. 521.

The ash of stramonium leaves varies from 14 to 22 per cent ; that
of the seeds from 2 to 3 per cent.

The alkaloids may be estimated by exhausting the leaves or seeds
with alcohol of 60 per cent strength and then using the process de-
scribed under the tincture.

Microscopic Examination. — In powdered stramonium leaves,
numerous fragments will be found which show partial sections of the

Fig. 56. — Powdered stramonium leaves x 240. cr, crystals ; ccr. crystal cells ;
ei, lower epidermis ; en, neural epidermis ; es, upper epidermis ; ffv, debris
of fibro-vascular bundles ; I bast ; me, spongy parenchyma ; pa,p'a', palisade
tissue ; pg, glandular hairs; po, pollen grains; pt, simple hairs; tf, cortical
tissue of midrib ; tr,v, vessels, etc. (Greenish & Collin).

By permission of the Editor of the " Pharmaceutical Journal ".

leaf, in which numerous rosettes of calcium oxalate are to be found.
A few glandular hairs will be seen and a fair number of pitted and
other vessels.

STROPHANTHUS. 603

Tincture of Stramonium. — This is prepared by exhausting 4
ounces of the leaves with sufficient 45 per cent alcohol to give 20
fluid ounces of the tincture. It should have the following char-
acters : —

Specific gravity 0-953 to 0-9(;2

Solid residue 3-2 „ 4 grms. per 100 c.c.

Alcohol by volume . . . . 42 to 43 per cent

Alkaloids 0-02 to 0-03 per cent

The alkaloids are determined by the process devised by Farr and
Wright ("Pharm. Journ." 3, xxii. 569) which is as follows: —

Fifty c.c. of the tincture to be estimated are introduced into a
porcelain dish, and evaporated over a water bath to low bulk, water
being added, if necessary, until all the spirit is removed. The re-
sidual liquor is allowed to cool, and is acidified with 1 c.c. of semi-
normal sulphuric acid, and the liquid filtered through cotton-wool
into a separator. The dish and filter are rinsed first with a little
acidulated water, and then with 15 c.c. of chloroform, the rinsings
added to the contents of the funnel, and the whole well shaken. After
separation the chloroform is drawn off, and the process repeated with
10 c.c. of chloroform. The washings are mixed and freed from traces
of alkaloid by shaking with three successive small portions of acidu-
lated water, and these are separated and added to the original solution.
The latter is then made alkaline with ammonia, and the alkaloids ex-
tracted with three successive quantities of chloroform of 15 c.c. each.
To obtain the alkaloids in a pure condition, they are withdrawn from
solution in chloroform by agitation with three successive small por-
tions of acidulated water, the mixed acid solutions made alkaline with
ammonia, and the alkaloids taken out by agitation first with 10 c.c,
and then with two successive quantities of chloroform of 5 c.c. each.
In cases where the final acidified aqueous solution is not colourless,
the process of shaking out is repeated. The mixed chloroformic
alkaloidal solutions are afterwards shaken with ammoniated watei,
and after separation are drawn off and evaporated over a water
bath, and the alkaloidal residue heated at 100° until the weight is
constant.

If the alkaloids are titrated, which is perhaps, the more correct
method, as in the case of gelsemium, 1 c.c. of -}q^^ normal HCl is
equivalent to 0*01445 grm. of alkaloid.

STROPHANTHUS.

The dried ripe seeds of Strophanthus kombe are official in the Phar-
macopceia, as well as an extract and a tincture, but no standards are
given for them, other than the following colour test which is intended
to distinguish the kombe seeds from those of other species.

The thin endosperm surrounding the cotyledons of the seed is
coloured dark green by sulphuric acid (presence of strophanthin).

The principal constituent of the seeds is the glucoside strophan-
thin C4yIIg^.0ic), which is present to the extent of from 1-8 to 3*2 per
cent.

604 FOOD AND DEUGS.

Seeds from other varieties of strophanthus are frequently present
in commercial parcels of the drug and it is not easy to distinguish be-
tween them and the kombe seeds.

According to Gordon Sharp (" Pharm. Journ." 4, xxiii, 258) the
official test with sulphuric acid is not always reliable and should be
modified as follows : a seed should be cut into four pieces and
placed on a white dish in which are 20 drops of 13'6 per cent sul-
phuric acid. Allow to stand for one minute. The dish is then
rotated over a Bunsen flame and in half a minute, the dark green
colour will appear at the edge of the fluid if the seeds are genuine.
The green colour rapidly spreads and if heating be continued, a red,
and finally black colour appears.

E. M. Holmes recommends the use of cold 80 per cent sulphuric
acid.

The seeds should yield from 3'5 to 4*5 per cent of ash on incinera-
tion. The most reasonable method of valuing the seeds appears to
be that of Barclay ("Pharm. Journ." 4, iii. 463). Twenty grms. of
the seeds in coarse powder are extracted with carbon bisulphide in
order to remove the fat. The seeds are then exhausted with 70 per
cent alcohol, in a Soxhlet tube, and the alcoholic liquid diluted with
its own volume of water and the alcohol evaporated. The filtered
aqueous liquid is then digested for an hour on the water bath with 1
per cent of sulphuric acid. This results in the hydrolysis of the
strophanthin with the formation of strophanthidin CgjjHggO^. This
can be extracted by shaking with three successive quantities of warm
amyl alcohol, the solvent evaporated and the residue weighed. One
part of strophanthidin is equivalent to 1*84 parts of strophanthin.

Mann (" Year-Book of Pharmacy," 1906, 249) has recorded much
higher figures for strophanthin in strophanthus seeds, but they appear
to lack confirmation. Caesar and Loretz (Report, September, 1905)
give the following method for the assay of the seeds : —

Seven grms. of crushed strophanthus seeds are treated in a flask
with 70 grms. of absolute alcohol, and the gross weight noted. The
whole is then digested, under a reflux condenser, on the water bath
for one hour. When cold the original weight is made up by the
addition of more absolute alcohol, and 50-5 grms. are filtered off ( = 5
grms. of seeds). The solvent is then evaporated, and the alcohol-free
residue treated with petroleum ether, to remove the fat, the solution
being passed through a small filter. The insoluble residue on the
filter is then washed back into the rest, in the capsule, with 5 to 8
c.c. of boiling water. The whole is then heated to boiling and treated
with 5 drops of basic lead acetate solution. The precipitate is
collected on a filter, and washed with boiling water until the filtrate
is free from bitterness. This aqueous filtrate is boiled and freed from
excess of Pb by means of SH2, the PbS being filtered off. On evapor-
ating an aliquot part of this filtrate, the residue may be weighed, when
dry, as crude strophanthin. To determine the amount of pure strophan-
thin, the above aqueous filtrate is hydrolysed by boiling for two hours
with 5 drops of pure HCl. When the volume of liquid is reduced to 10
c.c, it is made up to 20 c.c. with water, and, when cold, shaken out

STKOPHANTHUS. 605

with successive washings of CHCI3, the CHCI3 extracts being bulked
in a small tared flask. The aqueous portion, after shaking out, is.
again boiled for thirty minutes, and again shaken out with CHCl,.
the process being repeated as long as any bitter taste is evident. The
bulked CHCI3 solutions are then distilled to dryness and the residue,,
when constant, weighed as strophanthidin. The product x 1-84
gives the equivalent of strophanthin.

As has been pointed out by Gilg, Thoms and Schedel (" Berichte
Pharm." 14, 90) the various species of strophanthus yield glucosides-
which are not identical, hence any attempts at standardization, must,
to be of value, have reference to the botanical origin of the seeds.

Tincture of Strophanthus. — This is prepared by exhausting half
an ounce of the seeds with sufficient 70 per cent alcohol to produce
1 pint of tincture. A genuine tincture should have the following
characters : —

Specific gravity 1 . . . 0-894 to 0-897
Solid residue .... 0-4 „ 0-7 grm. per 100 c.c.

Alcohol by volume . . . . 68-5 „ 69 per cent
Strophanthin .... 0-05 „ 0-08 „

The strophanthin may be determined by evaporating 100 c.c. of
the tincture, diluted with an equal volume of water, until the
alcohol is removed and then proceeding as with the assay of the
seed.

Extract of Strophanthus. — This is an official semi-solid extract con-
taining the active principles of half its weight of the seeds. Ten grms.
should be rubbed down with 70 per cent to a cream and then warmed
for an hour with about 50 c.c. of 70 per cent alcohol. On cooling it
is filtered, the filter washed well with more 70 per cent alcohol, and the
strophanthin estimated as in the case of the tincture. Three authentic
samples made from kombe seeds gave the following results : 0*9 per
cent ; 1-15 per cent ; 1*26 per cent. About 1 per cent to 1"3 per
cent should be found in well-made extracts.

CHAPTER X.

THE ESSENTIAL OILS OF THE BEITISH PHAEMA-

COPCEIA.

1. Oleum Anethi.

Dill oil is the product of the distillation of the fruit of Anethum
£raveolens.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from Dill fruit.

'^Characters arid Test. — Colour pale yellow, odour that of the
fruit, taste sweet and aromatic. Specific gravity 0*905 to 0-920. It
rotates the plane of a ray of polarized light not less than 70° to the
right, at 60° F. (15-5° C), in a tube 100 millimetres long."

The British Pharmaceutical Codex describes the oil as having a
specific gravity 0905 to 0-915 and an optical rotation of + 75° to + 80''.

The oil is also distilled from a plant grown in India which is pro-
bably Anethum soma. The oil from the European plant is a pale
yellow liquid, which sometimes has a specific gravity as low as 0-895
but which then is of too low a standard to be used in medicine. The
oil from the Indian plant usually has a specific gravity of 0*945 to 0*970
and an optical rotation from -f 40° to + 50°.

It is the oil from the European plants which is official in medi-
cine in this country. This oil has a refractive index of about 14900.
It consists almost entirely of carvone and limonene. The oil should
possess the characters above given, and on distillation not more than
15 per cent should distil below 185"" and not less than 40 per cent
above 220°. The carvone may be estimated by the process described
under oil of cinnamon.

2. Oleum x\nisi.

This oil is either distilled from the true aniseed, Pimpineila anisum
•or from the star aniseed, Illicium verum. The latter plant is cultivated
in Southern China and Tonkin and furnishes the greater portion of
the aniseed oil of commerce.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from Anise fruit ; or from the fruit of the star
anise, Illicium verum, Hook. fil. [" Bot. Mag." plate 7005].

" Characters and Tests. — Colourless or pale yellow ; with the odour
of the fruit, and a mildly aromatic taste. It congeals, when stirred,
at temperatures between 50° F. and 59° F. (10° C. to 15° C.) and
should not again become liquid below 59° F. (15° C). Specific
gravity— at 68° F. (20° C.)— 0*975 to 0*990. It rotates the plane of
a, ray of polarized light slightly to the left."

(606)

ESSENTIAL OILS OF THE BEITISH PHARMACOPCEIA. 607

It is a pale yellow oil of a syrupy consistence. The specific
gravity lies between 0-975 and 0*990 at 20°. The rotation varies
between + 0° 30' and - 2". It is soluble in three volumes of 90 per
cent alcohol. The refractive index varies from 1-5520 to 1-5600.
The usual adulterants of this oil are petroleum, fennel oil, and the
waste liquid portion of aniseed oil obtaine'd in the manufacture of
anethol. The value of the oil depends upon the quantity of anethol
it contains, and as this melts at 21° to 22° and boils at 232° the melt-
ing-point and behaviour on distillation furnish valuable information
as to the value of the oil. Not less than 80 per cent should distil
between 225° and 235°.

Aniseed has a great tendency to exist in a state of superfusion so
that the oil may often be cooled down below its normal solidifying
point, when it may be necessary to add a crystal of anethol to induce
the oil to solidify, the temperature at the same time rising to what
may be described as the correct solidifying point, which should not be
below 13°.

A good deal of oil, which was possibly aduherated with a camphor
oil fraction, is to be found from time to time on the Loudon market.
The author has examined a number of such samples recently.

On fractionating large samples of the oil in question, the first point
to be noticed was the comparatively small amount distilling between
225° and 235°. In one case only 69 per cent was obtained, and in no
case more than 75 per cent. The average for normal oils is 83 per
cent. It was also noted that the first 10 per cent distilled had char-
acters quite different from the corresponding fraction of pure oil. The
following values are those of pure and suspected samples : —

Pure Oil.

Suspected Oil.

Fraction.

Amount.

M.Pt.

Per cent

No. 1 .

9°

Not at 0°

No. 2 .

16°

12°

No. 3 .

19°

15°

No. 4 .

20°

17°

No. 5 .

20-5°

18°

Eesidue

14°

10°

The above are Umney's figures.

Fraction.

Amount.

Pure Oil.
M.Pt.

Suspected Oil.
M.Pt.

Per cent

No. 1 .

8° .

-3°

No. 2 .

18°

15°

No.3 .

20°

17-5°

No. 4 .

20°

18°

No. 5 (Eesidue) .

15°

11°

608

FOOD AND DEUGS.

The above are the author's figures.

It will be noticed that in every case the fraction of the suspected
oil had a lower melting-point than the corresponding fraction of pure
aniseed oil.

Erom the following figures it will be seen that the same fact is
noticeable in regard to the refractive index, which were determined at
20° to 21" :—

Fraction.

Amount.

Pure Oil.

Suspected Oil.

No. 1 .
No. 2 .
No. 3 .
No. 4 .
No. 5 .
Residue

Per cent
10
15
20
20
20
15

1-5308
1-5470
1-5550
1-5575
1-5581
1-5540

1-5110
1-5391
1-5463
1-5513
1-5538
1-5478

3. Oleum Anthemidis.

This oil is distilled from the flowers of Anthemis nohilis.

The optical rotation varies from + 1° to +3°. It consists princi-
pally of the esters of angelic and tiglic acids. Its use in medicine is
extremely limited.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from Chamomile flowers.

" Characters. — Pale blue or greenish-blue when freshly distilled,
but gradually becoming yellowish-brown. It should have the aromatic
taste and odour of the flowers.

Specific gravity 0-905 to 0-915."

Its refractive index is between l'44:40-l-4470

4. Oleum Cadinum.

This oil contains a large proportion of Cadinine, one of the best-
known sesquiterpenes. It is not an essential oil in the proper sense
of the word, being obtained as mentioned below, by a process of de-
structive distillation.

The British Pharmacopoeia describes this oil as follows : —

" An empyreumatic oily liquid obtained by the destructive distilla-
tion of woody portions of Ju7iiperus Oxycedrus, Linn. [Moggridge,
•'Flora of Mentone," tab. 65], and some other species.

" Characters and Tests. — A dark reddish-brown or nearly black,
more or less viscid, oily liquid, with a not unpleasant empyreumatic
odour and an aromatic, bitter and acrid taste. Specific gravity about
0-990. It is soluble in ether and chloroform ; partially soluble in cold,
almost wholly in hot alcohol (90 per cent). It is very slightly soluble
in icater. The filtered aqueous solution is almost colourless and
possesses an acid reaction."

ESSENTIAL OILS OF THE BRITISH PHARMACOPCEIA. 609

5. Oleum Cajaputi.

This oil is distilled from the leaves of various species of Melaleuca.
The Pharmacopoeia, however, restricts the oil to a given species.
The British Pharmacopoeia describes this oil as follows : —
" The oil distilled from the leaves of Melaleuca leucadendron, Linn.
{Melaleuca cajaputi, Roxb.) ['* Bentl. and Trim. Med. PI." Vol. II, plate
108].

" Characters and Tests. — Bluish-green, with an agreeable penetrat-
ing camphoraceous odour, and an aromatic bitterish camphoraceous
taste. Specific gravity from 0-922 to 0-930. It should become semi-
solid on being stirred, when cold, with a third or half its volume of
phosphoric acid of commerce of specific gravity 1-750 (presence of a
due proportion of cineol)."

The specific gravity of the British Pharmacopoeia is higher than
that found in many samples of pure oil. It is well recognised that
0-917 is a permissible limit for genuine cajaput oil.

The oil is nearly inactive optically, the rotation usually varying
from 0° to - 2°. The refractive index varies from 1-4650 to 1-4680.
Genuine oils contain from 55 per cent to 65 per cent of eucalyptol as
determined by the phosphoric acid process, which is described under
eucalyptus oil. The author has, in recent years, found no adulter-
ants present in this oil except eucalyptus oil, which is detected by its
odour, and alcohol, which can be estimated by shaking a known vol-
ume of the oil with ten times its weight of water. The oil is always
of a pale green colour, but can be obtained white by redistillation.
It owes its use in medicine entirely to the presence of eucalyptol.

6. Oleum Caeui.

This oil is obtained by the distillation of the fruit of Carum carui.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from Caraway fruit.

" Characters. — Colourless or pale yellow, with the characteristic
odour of the fruit, and a spicy taste. Specific gravity 0*910 to
0-920."

Caraway oil resembles Dill oil very closely in its composition, con-
sisting practically entirely of carvone and limonene. Normal oils-
sometimes have a specific gravity slightly below that given in the
British Pharmacopoeia, but it is as well to adhere to the higher limit,
as such low-gravity oils have usually been deprived of carvone. The
oil has an optical rotation of -1-70° to -I- 85°, and a refractive index
1-4870 to 1-4900. The usual adulteration of the oil consists either in
the abstraction of carvone, or in the addition of oil from which car-
vone has been abstracted. The estimation of the carvone is therefore
a matter of importance. This is best estimated by the process de-
scribed under oil of cinnamon. The following process also yields ex-
cellent results : —

When 5 c.c. phenylhydrazine are added to 5.c.c. of caraway oil the
mixture becomes warm owing to chemical combination taking place^
and if the action be accelerated by placing the test tube in boil-
VOL. I. 39

610 FOOD AND DKUGS.

ing water for a few minutes a copious crystallization of carvone
phenylhydrazone CioHi^ : N . NH , G^B.^ appears, and on cooling the
whole solidifies to a crystalline mass. After heating for one hour the
reaction is complete, the excess of phenylhydrazine is removed by add-
ing 5 c.c. glacial acetic acid whilst hot, shaking and diluting with 20 c.c.
water. The contents of the test tube are then cooled and filtered
through a paper disc by means of a pump, and the crystalline mass
washed with water until of a pale yellow colour. By this process not
only is the excess of phenylhydrazine removed in aqueous solution as
acetate, but nearly all the oily terpene adherent to the crystals is washed
away. On crystallizing from a definite volume of 95 per cent alcohol
the carvone phenylhydrazone is obtained in long silky pale yellow
needles, melting at 106° C, but so difficult to dry without decomposi-
tion as to render the determination only approximate.

7. Oleum Caryophylli.
This oil has already been described under the spice " Cloves ".

8. Oleum Cinnamoni.

This oil has been described under the spice " Cinnamon ".

9. Oleum Copaibae.

This oil will be found described under " Balsam of Copaiba ".

10. Oleum Coeiandri.

This oil is distilled from the fruit of Coriandrum sativum.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from Coriander fruit.

•' Characters and Tests. — Colourless or pale yellow, having the odour
and flavour of the fruit. Specific gravity 0-870 to 0-885. If 1 c.c. of
the oil be mixed with 3 c.c. of alcohol (70 per cent), a clear solution
results (absence of oil of turpentine and added terpenes)."

The optical rotation of this oil varies from + 7° to + 15°. On
fractional distillation from 45 per cent to 55 per cent should be ob-
tained between 190" and 200° indicating a due proportion of linalol,
Avhich is the principal odorous constituent of the oil. The refractive
index is about 1-4650 and the ester number varies from 4 to 23. The
oil should be soluble in three times its volume in 70 per cent alcohol.
The only adulterant now met with in this oil, is sweet orange oil,
which interferes very greatly with the solubility of the oil and raises
the optical rotation enormously.

11. Oleum Cubebae.

This oil is the product of the distillation of the fruit of Piper
cuheha. ,

The British Pharmacopoeia describes this oil as follows : —
" The oil distilled from cubebs.
Characters. — Colourless, pale-green, or greenish-yellow ; with the

ESSENTIAL OILS OF THE BRITISH PHARMACOPCEIA. 611

odour and camphoraceous taste of cubebs. Specific gravity 0*910 to
0-930."

The oil has an optical rotation of - 30° to - 40° and a refractive
index of about 1*4950. The solubility of the oil in 90 per cent alcohol is
very variable, some oils dissolving in one volume, others requiring 10
volumes to effect solution. The oil is a mixture of terpenes and
sesquiterpenes, cadinine being the principal of the latter with a small
amount of so-called cubeb-camphor. Samples are sometimes found
adulterated with turpentine. A genuine oil on distillation yields the
following fractions : —

Below 250° = 10 per cent ; 250° to 260° = 25 per cent ; 260° to
270° = 50 per cent ; 270° to 280° = 5 per cent.

12. Oleum Eucalypti.

This oil is distilled from the leaves of various species of Eticalyptus.
Its reputation in medicine has been built up on a description under
the name Eucalyptus globulus, although but little of the oil reaching
this country under that name is really distilled from the globulus
species.

In fact, to most patients the globulus oil <is irritating and objec-
tionable, and many oils distilled from other species are, in the author's
opinion, much to be preferred to the globulus oil.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from the fresh leaves of Eucalyptus globulus,
Labill [" Bentl. and Trim. Med. PI." Vol. II, plate 109] and other
species of eucalyptus.

" Characters and Tests. — Colourless or pale yellow, having an aro-
matic camphoraceous odour, and a pungent taste, leaving a sensation
of coldness in the mouth. Specific gravity 0*910 to 0*930. It should
not rotate the plane of a ray of polarized light more than 10° in either
direction in a tube 100 mm. long, and it should become semi-solid on
being stirred, when cold, with a third or half its volume of phosphoric
acid of commerce of specific gravity 1*750 (presence of a due propor-
tion of cineol). If to 1 c.c. o'f the oil be added 2 c.c. of glacial acetic
acid and 2 c.c. of a saturated aqueous solution of sodium nitrite, the
mixture, when gently stirred, should not form a crystalline mass (ex-
clusion of eucalyptus oils containing much phellandrene)."

Although there are many pure oils to be met with having figures
well outside the limits given by the British Pharmacopoeia there is
a plentiful supply of oil up to the standards of that authority and it is
wise that that standard should not be relaxed.

The British Pharmacopoeia gives only a qualitative test for euca-
lyptol. A quantitative determination is therefore of considerable im-
portance. No method, however, yields absolutely accurate results.
It is therefore necessary in stating the eucalyptol value of a given oil,
to describe the method by which the determination has been made.
A convenient and approximately accurate method is as follows : —

To a known weight of oil from one to one and a half times its
weight of phosphoric acid of specific gravity 1*75 should be added,

612 FOOD AND DEUGS.

drop by drop, the oil being kept cold and continually stirred. The
crystalline magma formed is pressed between filter paper, after as
rauch as possible has drained off; and when the adherent terpenes
and phosphoric acid have been removed as far as possible, the crystals
are decomposed by hot water in a graduated tube. On cooling, the
cineol is measured, and from its specific gravity (-930) the weight i&
easily calculated. The separated cineol should readily crystallize on
cooling to - 3°, otherwise it must be regarded as impure and the pro-
cess repeated. Oils rich in cineol yield a correspondingly high frac-
tion distilling between 170° and 190°.

The United States Pharmacopoeia directs that the oil is to be
diluted with petroleum ether before treatment with the phosphoric acid.

Messrs. Schimmel & Co. have recently recommended absorbing
the eucalyptol by a 50 per cent solution of resorcinol in water, and
reading the unabsorbed portion in the neck of an ordinary absorbing
flask. Although this method gives approximately accurate results in
some cases, so many other constituents of essential oils are absorbed
by this reagent that the process cannot be entirely relied upon.

13. Oleum Juniperi.

This oil is obtained by the distillation of the fully grown unripe
fruit of Juniperus communis.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from the full-grown unripe green fruit of Juniperus
communis, Linn, ["Bentl. and Trim. Med. PI." Vol. IV, plate 255].-

" Characters and Tests. — Colourless or pale-greenish yellow, with
the characteristic odour of the fruit, and a warm, aromatic, bitterish
taste. Specific gravity 0'865 to 0*890. The oil is soluble, with slight
turbidity, in four times its own volume of a mixture of equal parts of
absolute alcohol and alcohol (90 per cent)."

In regard to the above tests it should be noted that pure juniper
oil rapidly loses its solubility by keeping, so that pure samples will
frequently fail to comply with the solubility test of the Pharmacopoeia.

Eectification also naturally alters the specific gravity, which depends
chiefly on the relative proportions of terpene (specific gravity = 0'845),
and sesquiterpene (specific gravity =-920). The limits -865 and -890
are certainly those which should be accepted for genuine good oils.
The approximate proportions of pinene and cadinene may be judged by
a fractional distillation, as pinene boils at 156° and cadinene at 274°.
The results vary largely according to the fractionating apparatus used,
but with a series of bulbs, from 25 per cent to 35 per cent is obtained
between 155° and 160°, and at least 20 per cent ever 200°, having a
refractive index of over 1*4950 and a specific gravity over 0*904.

The oil is always laevorotatory, usually between - 4° and - 10° ex-
cept in the case of Hungarian oil, which may have a rotation up to
- 19°. The refractive index varies from 1*4740 to 1*4880. After
distilling off 90 per cent of the oil the 10 per cent residue should have
a refractive index of not less than 1*5000 indicating a sufficient pro-
portion of cadinine.

ESSENTIAL OILS OF THE BRITISH PHARMACOPCEIA. 613

The only constituents of oil of juniper which have been ascertained
with certainty are (1) the terpene, pinene CjoH^,, ; (2) the sesquiter-
pene, cadinene C^Jl.,^ ; (3) juniper camphor, a crystalline body pro-
bably belonging to the series of terpene alcohols ; (4) an ester boiling
at about 180°, probably the acetic ester of the above-mentioned alcohol.
According to Schimmel, the chief, if not only, constituent of the
stearoptene is a sesquiterpene alcohol melting at 165° to 166".

14. Oleum Lavandula.

This oil is distilled from the flowers of Lavandula vera.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from the flowers of Lavandula vera, D. C.
[Bentl. and Trim. Med. PI." Vol. HI, plate 199] .

** Characters a7id Test. — Pale yellow or nearly colourless, with the
fragrant odour of the flowers, and a pungent bitter taste. Specific
gravity not below 0-885. It should dissolve in 3 times its volume of
alcohol (70 per cent)."

There are two distinct varieties of genuine lavender oil ; one is that
distilled from the plants grown in certain districts in England, the
other distilled principally from plants grown in France, and to some ex-
tent in Spain.

The chief difference between English and French oils of lavender
lies in the fact that the former only contains about 7 to 10 per cent of
esters calculated as linalyl acetate, whereas the latter contains up to 40
per cent and over. Messrs. Schimmel have actively endeavoured to
establish this ester content as the basis of the valuation of the oil.
They maintain the superiority of fine French oil over English oil, and go
so far as to say that the latter cannot compete with the former. The
author, in common with most others, holds the opposite opinion, and
considers that no comparison can be made between the two oils on the
basis of their ester content. This is much accentuated, if such were
necessary, by the fact that linalyl acetate is not the odoriferous in-
gredient of oil of lavender. It is so much modified by the presence of
other bodies, as to be regarded as only one of the odoriferous com-
pounds in the oils. Pure linalyl acetate has a marked bergamot
odour, and may be regarded as the characteristic ingredient of that oil.
The fact that English oil fetches from 5 to 8 times as much as
French oil speaks for itself. For a comparison of oils grown in the
same locality, the ester comparison may, however, be of service.
The oils produced in the south of Europe are finer according as the
plants are growing at greater elevations, and according to Schimmel
& Co. the very finest oils are produced from the higher valleys of the
Savoy Alps, yielding 44 per cent of ester. The fine oils yielding 38
to 40 per cent of ester are usually obtained from the Alps Maritimes
and the Basses Alpes, close to the Italian frontier. Less fine, but
still excellent oils, with 28 to 32 per cent of ester, are obtained from
the French departments of the Gard Drome and Herault.

Genuine lavender oil is a pale yellow oil of specific gravity 0*885
to 0-900, with an optical rotation of - 3^ to - 10°. Rarely the specific

614 FOOD AND DRUGS.

gravity falls to 0-883. The refractive index varies from 1-4622 to
1-4675. Coarse adulteration with such bodies as turpentine would re-
veal itself by the decrease in the solubility of the sample. Oil of spike
lavender is used very commonly for the purpose of adulteration, and
causes a reduction in the ester value, a rise in the specific gravity and
usually a diminution in the optical rotation of the oil. Spike lavender
oil being generally dextrorotatory, the most formidable adulterant
with which the analyst has to cope to-day is a mixture of spike laven-
der oil and artificial esters, such as ethyl citrate, ethyl succinate, or
ethyl oxalate. As these esters require considerably more alkali for sap-
onification than does linalyl acetate, which is the principal constituent of
French oil of lavender, a small quantity of any one of them appears in
an ester determination, as indicating a considerably higher proportion
of the natural ester.

The following is the best method for detecting adulteration with
these artificial esters : ten grms. of the oil to be examined are
saponified on the water bath for one hour with alcoholic potash, the
contents of the flask then placed in a porcelain dish and the bulk of
the alcohol evaporated. After this the liquid is washed in a separating
funnel with about 100 c.c. water, the oil portions removed by extraction
with ether, but the aqueous solution returned to the porcelain dish
and the bulk evaporated on the water bath. When the alkaline
solution has cooled down it is acidified with sulphuric acid, and the
organic acids thus liberated absorbed by ether. In this manner a fine
crystal residue remains behind. On recrystallization, however, from
a small quantity of alcohol, white crystals are obtained, and the
melting-point can be determined. In the event of the organic acid
being insoluble in ether, it can be precipitated as a barium salt and
examined. On fractional distillation the artificial esters will be
found in the residues left after distilling off the more volatile portion,
and will be found to have a very high specific gravity and low refrac-
tive index. A comparison with similar fractions of a normal oil will
at once reveal the characteristic differences.

15. Oleum Limonis.
This oil has been described under " flavouring essences ".

16. Oleum Menthae Piperita.

This oil is distilled from the flowering herb of Mentha piperita.

The British Pharmacopoeia describes the oil as follows : —

" The oil distilled from fresh flowering peppermint, Mentha piperita,.
Sm. [" Bentl. and Trim. Med. PI." Vol. Ill, plate 203].

" Characters and Tests. — Colourless, pale yellow, or greenish-yellow
when recently distilled, but gradually becoming darker by age. It has
the odour of the herb, and a strong penetrating aromatic taste,
followed by a sensation of coldness in the mouth. Specfic gravity
0-900 to 0-920. It should dissolve in four times its volume of alcohol
(70 per cent). If a portion of the oil be cooled to 70° F. (-8-3° C.)

ESSENTIAL OILS OF THE BEITISH PHAKMACOPCEIA. 615

and a few crystals of menthol be added, a considerable separation of
menthol should take place."

So far as English commerce is concerned the only true oils of
peppermint which come under consideration are the English and
American distillates. It is true that a large business is done in
Japanese peppermint oil but this is distilled from a different species —
Mentha arvensis. The figures given here in regard to the oil refer to
these two species only. Few plants alter more largely in the character
of their essential oil according to the districts in which they are
cultivated than does peppermint, so that French, Italian, and Spanish
oils have characters quite different from those here discussed, but as
they do not enter into English commerce to any extent they need
not be further considered. The principal constituent of peppermint
oil is menthol, principally in the free condition and to a smaller
extent as esters, together with a certain amount of menthone. Numer-
ous other bodies exist in this oil, for an account of which the author's
work " The Chemistry of Essential Oils " should be consulted. True
peppermint oil has a specific gravity of 0-900 to 0*920, rarely up to
0'925, and an optical rotation of - 22° to - 33°, and refractive index
up to about 1-4650. It is soluble in from 3 volumes to 4 volumes of
70 per cent alcohol but in the case of certain American oils, the solu-
bility is not complete. American and English oils contain from 50
per cent to 65 per cent of menthol. The Japanese oil has a specific
gravity of 0-895 to 0-905 and an optical rotation of - 25" to - 43°.
The menthol of commerce is almost entirely derived from Japanese
peppermint oil. The normal Japanese oil contains over 70 per cent
of menthol, and after the abstraction of a portion of this, the residual
oil is sold on this market as Japanese dementholized peppermint oil
and contains about 40 per cent of menthol. The table on page 616,
due to Power and Kleber, shows the characters of a number of typical
peppermint oils : —

The most important determination apart from the physical char-
acters for this oil, is the estimation of the amount of menthol. The
following are the details of the necessary process : —

About 10 grms. (accurately weighed) of the oil together with 20
c.c. of an alcoholic normal solution of sodium hydroxide, are either
heated to boiling for half an hour in a flask provided with a reflux con-
denser, or the mixture, contained in a strong, securely-stoppered
glass bottle, is heated for an hour in a bath of boiling water, and sub-
sequently the uncombined alkali titrated with normal sulphuric acid
with the use of phenol -phthalein as an indicator. From this the com-
bined menthol is calculated as menthyl acetate.

The saponified oil is then repeatedly well washed with water and
finally boiled for 2 hours with an equal volume of acetic anhydride
and 2 grms. anhydrous sodium acetate in a flask provided with a
suitable condensing tube, ground at one end so as accurately to fit the
neck of the flask. The product, after cooling, is washed with water,
then with a dilute solution of sodium carbonate, dried in contact
with calcium chloride, and filtered. From 3 grms. to 4 grms. of the
resulting oil are then saponified as above, now using 25 c.c. of

616

FOOD AND DRUGS.

o «
o

•anpisaa

iH t- CO -^

«b cb CO «3

(>lrH0pCprH»OFHI>-':O»r5->*C^ »H

lo »o »o CO -^ <N 4< db CO CO c« «s (n

o58Z-08S

2^22 I IS M I I IS M I I

.08Z-SZS

>^zz-os&

•oOSS-SIS

W3 i?) -rl* OS
Tt< "* (f) !?1

I I

122 M I

qp (N CO «p

Oi OS !>• !>

iNcpqs'^cp ir»<(>iqpr-icp i rf
■^ (k <x> ^ oi IcoorHcqcb I (N

O ^ OS OS
«N rH (>l CO

i?lrt<(NOi-t(?1-^qS(?5010«p

<?i<N^rHocb6so5«bt:-(Ndb

•o5IS-0I2

•^ M5 O l>

•oOiz-eos

lt>CpO ap'*»00«P'^OSrHCT-*lN(N OS

iiosbws dbosio(fit^eo6s(X)rHcboi?5 o

•oSOS-OOS

iHiHfHfH CO»OiHC^I>05»r5i)00»HO I j^

•Oo002<4i

«3qst-.qsTt<rttt>coosi>-»p'*

iHrHfHrHCO-rf<rHCOOOO>H

I-

?ooo«oao«oosr-tc-!Moooo

Susc^iQprH Oi:pcpq5cpiyici»r505t>tHq5 <?!I
"»ccooo THcoodb6s-^'^«b'^o5>cco »h

S a Sh 3c

fc- -* iH OS OS

•^ Oi C^ W5 >— I <M

i>-o-^Tt<t>qs

assbt^-cb-^-^sboOfHcows-*

I § S 8
■+^ '-S o '2

f§

^ ^ «5 O
O O ""t «5

C5 o o CO

CO CO CO CO

'<JtC<ICOTtHCOCO«5CO'^Tt<C<)»0
oooooooooooo
(MOSOOOSOSCOCOCOCO'^i-lt-
eOCOCO i-l(MCQC<lrHCOCOC0

I I I I I I I I I I I I

OOQOO Ot>»0»CCOO»OW50000 O

Tj*CO»Or-t »-l«0C0C0G0«5OO»0OC0t- O

.HrHlH^ i-HOi-lrHOOr-liHCOr-lOO fH

oscsosos osososososososososososqs os

oooo oooooooooooo o

ESSENTIAL OILS OF THE BRITISH PHARMACOPCEIA. 617

alcoholic normal solution of sodium hydroxide, and the uncombined
alkali determined by titration.

As each c.c. of normal alkali required for saponification corre-
sponds to 0*156 grm. menthol, and as this yields 0-198 grm. menthyl
acetate, it is necessary, in order to calculate the found amount of
menthol with reference to its percentage in the non-acetylized oil (free
from ester), to subtract from the amount of oil used for saponification
0'042 grm. (the difference between 0-156 grm. and 0-198 grm.) for
each c.c. of normal alkali consumed. If, for example, s grms. of
acetylized oil had required for saponification a c.c. of normal alkali,
the total percentage P of menthol, free and in the state of ester, may
be calculated by the following formula : —

ax 15-6
■^~5- (ax 0-042)'

This, indeed, does not indicate with absolute exactness the per-
centage of menthol contained in the original oil, for it is assumed
in this calculation that all the menthol which is present as ester is
combined with acetic acid, whereas as a matter of fact it is partly in
combination with iso- valerianic acid, etc. But the error so introduced
is so small that it may be left out of account.

As menthone may readily be converted into menthol by reduction,
the above-described method may be also employed for the determina-
tion of the amount of menthone in the oil, in the following manner.
The oil is saponified, and in a portion of the product, previously de-
prived of alcohol, the percentage of menthol is determined. Another
portion is diluted with twice its volume of alcohol, and treated at
the boiling temperature of the liquid with metallic sodium. Of the
oil which separates by the subsequent addition of water, a weighed
quantity is used for another estimation of menthol. The increase
corresponds to the amount of menthone present.

The above formula, which gives the total percentage of menthol,
is not quite accurate, as it is referred to the ester-free (saponified) oil.
The correction necessary to be introduced is not of great importance,
as the quantity of menthyl esters is not nearly so great as that of free
menthol, but to be perfectly correct it must be remembered that to
calculate the ester-free oil to the original peppermint oil, the latter has
lost weight as compared with the former to the extent of -75 per cent
for each 1 per cent of KOH required for the preliminary saponification
of the esters, assuming that these are all present as menthyl acetate.
Thus if an oil gives an ester content of 10-6 per cent, equivalent to 3
per cent of KOH, or 8-4 per cent of menthol, and a total menthol
content as calculated from the above formula of 60 per cent, it is

necessary to multiply this by- the factor to obtain the total men-

thol content in the original oil, i.e. 58-7. Hence the free menthol will
be 50-3 per cent and the combined menthol 8*4 per cent.

Peppermint oil is frequently adulterated. The American oil is
sometimes enriched by the addition of menthol ; or it is adulterated
with camphor oil, petroleum oil, cedar wood oil, and African copaiba

618

FOOD AND DRUGS.

oil. In one case Bennet has observed the use of glyceryl triacetate.
Camphor oil, petroleum oil, cedar wood oil and copaiba oil impair the
solubility so seriously as to at once be indicated. Glyceryl triacetate
is usually soluble in spirit and is therefore not indicated by the solu-
bility test. The fractional distillation of the oil is absolutely essential
in considering adulterants of this. type. The pure oil will give figures
not differing materially from the following, which were obtained on a
normal sample : —

Quantity.

Specific Gravity.

Opt. Rotation.

Refractive Index.

Per cent

12i

0-898

1 10

1-4660

0-903

-14

1-4635

0-907

-16

1-4645

0-910

-20

1-4640

)

0-912

-23

1-4615

0-912

-23

1.4615

0-915

-24

1-4630

0-962

—

1-4790

If cedar wood or copaiba oils are present the higher boiling frac-
tions will be found to have refractive indices up to 1*490 or even
higher, and to consist of hydrocarbons quite insoluble in 70 per cent
or 80 per cent alcohol. With cedar wood oil the rotations of the
higher boiling fractions may reach - 45° and in the case of African
copaiba will be much lower than normal. In the case of glyceryl
triacetate, the residues after distilling off the more volatile portion of
the oil will be found to have a high specific gravity even up to 1-14,
and a low refractive index down to 1-445 ; such residues will also be
found to be freely soluble in 70 per cent or 80 per cent alcohol.

17. Oleum MENTHiE Viridis.

The true spearmint oil is obtained from the green herb Mentha
viridis, but a good deal of the oil of commerce is obtained from
Mentha crispa. The two oils however are practically identical.

The British Pharmacopoeia describes this oil as follows : —

"The oil distilled from fresh flowering spearmint, Mentha viridis,
Linn. ["Bentl. and Trim. Med. PI." Vol. Ill, plate 202].

" Characters and Tests. — Colourless, pale-yellow, or greenish-
yellow when recently distilled, but becoming darker by age. It has
the odour and taste of the herb. Specific gravity 0-920 to 0*940.
The oil forms a clear solution with its own volume of a mixture of
equal parts of absolute alcohol and alcohol (90 per cent)."

The specific gravity of the oil may reach considerably higher
limits than those of the B. P., a pure oil with a specific gravity of
over *970 having been observed by Schimmel & Co. The optical
rotation of the oil usually varies between - 40° and - 50°. The oil
should dissolve in 1 volume of 90 per cent alcohol.

ESSENTIAL OILS OF THE BEITISH PHARMACOPCEIA 619

When estimated as described under cinnamon oil this oil should
show a carvone content of between 35 per cent and 45 per cent,
and on fractional distillation about that quantity should be obtained
between 200° and 226°. The oil is not much applied in medicine.

18. Oleum Mybistic^.
This oil has already been described under the spice " Nutmeg".

19. Oleum PiMENTiB.
This oil has already been described under the spice " Allspice ".

20. Oleum Pini.

There are numerous varieties of essential oils distilled from pine
leaves, the British Pharmacopoeia only recognizing that distilled from
the leaves of Pinus pumilio.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from the fresh leaves of Pinus pumilio, Haenke
[Lambe. Gen. Pin. I. plate 2] .

" Characters and Tests. — Colourless or nearly so, with a pleasant
aromatic odour and pungent taste. Specific gravity 0-865 to 0-870.
It should rotate the plane of a ray of polarized light from 5° to 10° to
the left at 60' F. (15-5° C.) in a tube 100 mm. long. Not more than
10 per cent should distil below 329° F. (165° C)."

A genuine oil may have a specific gravity up to '875, but accord-
ing to Umney should give only a minute distillate below 165°, a
typical sample examined by him yielding only 2 per cent. About 60
per cent should distil between 165° and 180°.

Many samples of pine needle oil correspond with the description
and tests of the British Pharmacopoeia, but are in reality distilled
from other species of pine.

21. Oleum Ros^.

This oil is distilled from the flowers of Bosa damascena which is
the only official variety recognized by the British Pharmacopoeia
although French oil of rose is distilled from Bosa centifolia.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from the fresh flowers of Bosa damascena, Linn.
[Redout6, " Les roses," plate 109] .

" Characters and Tests. — A pale yellow crystalline semi -solid, with
the strong fragrant odour of rose and a sweet taste. Specific gravity
0-856 to 0-860 at 86° F. (30' C). The congealing and melting-points
vary according to the proportions of crystalline matter, but should lie
between 67° and 72° F. (19-4° and 22-2° C.)"

The Pharmacopoeial description of this oil practically restricts it to
the Bulgarian product. Few essential oils are so grossly adulterated as
is this, but as its use is exceedingly restricted in medicine,. it only being
used to perfume a few preparations, there is no justification for devot-
ing much space to it in a work devoted to food and drugs.

620 FOOD AND DRUGS.

When the otto has to be examined from a perfumer's point of view
reference should be made to the " Chemistry of Essential Oils," by the
author, second edition, pp. 396-409.

The following, however, are to be regarded as the limits for the
figures yielded on analysis by pure Bulgarian otto of rose : —

Specific gravity at 30° . . . . 0-850 to 0-861

Optical rotation -2 ,, -5°

Refractive index at 25° . . . . 1-4610 „ 1-4650

Melting-point 19° ,,23°

The total alcohols calculated as geraniol, when estimated in the
method similar to that described for menthol and oil of peppermint,
should not exceed 75 per cent or rarely 76 per cent. A small quantity
of alcohol is frequently added as an adulterant. This may be detected
by shaking the otto with warm water and testing the water separated
by the usual iodoform reaction. Otto so washed with vjater will if
alcohol be present show a rise in its refractive index. A pure otto in
these circumstances will not show an increase in its refractive index
of more than 0*002.

Further, if instead of using acetic anhydride to convert the alcohols
into esters as in the determination of the total alcohols, formic acid be
used, the citronellol, which is one of the constituents of the alcohol of
this oil, will alone be esterified, so that separation of the geraniol and
citronellol is thus practicable. In genuine otto of rose the approximate
amount of citronellol in the total alcohols will be about 35 per cent.
A much lower figure than this indicates the addition of geraniol pre-
pared from other sources.

22. Oleum Rosmarini.

This oil is distilled from the flowering tops of Rosmarinus
officinalis.

The British Pharmacopoeia describes this oil as follows : —

" The oil distihed from the flowering tops of Bosmarinus officinalis.
Linn. ['' Bentl. and Trim. Med. PL" Vol. Ill, plate 207].

" Characters and Tests. — Colourless or pale yellow, with the odour
of rosemary, and a warm camphoraceous taste. Specific gravity 0*900
to 0*915. It should dissolve in twice its volume of alcohol (90 per
cent), and should not rotate the plane of a polarized ray of light more
than 10° to the right in a tube 100 mm. long (absence of oil of tur-
pentine)."

In regard to the British Pharmacopoeia figures given above, it may
be remarked that many samples of pure rosemary oil, especially those
distilled in Spain, are laevorotatory up to - 9°.

Apart from the determination of the physical characters of this oil
the only estimation that is usually necessary is that of the boneol,
which is determined in a manner similar to menthol in oil of pepper-
mint. This will usually vary from about 12 per cent in low-grade
samples to 20 per cent in the best samples.

ESSENTIAL OILS OF THE BKITISH PHARMACOPCEIA. 621

23. Oleum Santali.

This oil is distilled from the wood of Santalum album.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled from the wood of Santalum album, Limi.
[- Bentl. and Trim. Med. PI." Vol. IV, plate 252].

" Characters and Tests. — Somewhat viscid in consistence, pale
yellow in colour, having a strongly aromatic odour and a pungent and
spicy taste. Specific gravity (-975 to 0-980. It forms a clear solu-
tion with six times its volume of alcohol (70 per cent) (absence of
cedar wood oil). It rotates the plane of a ray of polarized light to the
left, through an angle not less than 16° and not more than 20°, in a
tube 100 mm. long (absence of other varieties of sandal wood oil)."

This oil consists essentially of from 90 per cent to 95 per cent of
alcohols, which although a mixture of several bodies, are usually known
as santalol, to which the formula Cj^HgeO is usually assigned. The
chemistry of this oil is fully discussed in the work above quoted, by
the author, pp. 244 to 260. The figures of the Pharmacopoeia are not
sufficient to determine the purity or otherwise of this oil. A genuine
oil has a specific gravity varying between 0-973 and 0-985. The
optical rotation occasionally falls below - 16°, but as a rule when this
is the case it is due to defective distillation of the wood. The oil has a^
refractive index of 1-505 to 1-510. It contains a minute quantity of
free acid and requires 0-7 grm. to 1*5 grms. of potassium hydroxide to
saponify the esters present. The alcohols calculated to the above
given formula, when estimated by a process similar to that described
for menthol in oil of peppermint, should never fall below 90 per cent.
A pure oil rarely shows below 92 per cent. Most adulterants cause
the oil to be considerably less soluble than the pure oil should be,
and also reduce the amount of total alcohols, which is usually
known as the santalol value. Fractional distillation of the oil may
occasionally be necessary. When this is the case, the results should
agree with the following, which were obtained on a normal pure
sample of sandal wood oil : —

Fraction.

Specific Gravity.

Opt. Rotation.

Refract. Ind.

Per cent

0-970

- 19° 30'

1-5055

.970

- 17° 20'

1^5060

.927

-16°

1^5060

•974

-16°

1-5065

•977

- 15° 30'

1-5068

•978

-15°

1-5068

•980 •

- 16° 40'

1-5079

.980

-18°

1-5080

•984

-21°

1-5084

The acetylated oil should have a specific gravity 0-986 to 0*989,
an optical rotation of - 13° 30' to - 18° and a refractive index
1-4899 to 1-4920 at 20°.

622 FOOD AND DRUGS.

24. Oleum Sinapis Volatile.

This oil has been described under the condiment " Mustard ".

25. Oleum TEREBiNTHiNiE.

This oil is obtained by the distillation of the oleo-resin obtained
from Pinus sylvestris and other species of pine.

The British Pharmacopoeia describes this oil as follows : —

" The oil distilled, usually by the aid of steam, from the oleo-resin
(turpentine) obtained from Pinus sylvestris, Linn. [" Bent, and Trim.
Med. PI." Vol. IV, plate 275], and other species of Pinus; rectified if
necessary.

" Characters and Tests. — Limpid, colourless, with a strong peculiar
odour, which varies in the different kinds of oil, and a pungent and
somewhat bitter taste. It is soluble in its own volume of glacial acetic
acid. It commences to boil at about 320° F. (160° C), and almost
entirely distils below 356° F. (180° C), little or no residue remaining."

The following are the best-known varieties of oil of turpentine : —

American Turpentine. — This is chiefly obtained from Pinus
Australis, but also to a certain extent from Pinus tceda, the loblolly
pine. It is a colourless limpid liquid of specific gravity '855 to •870.
It is almost invariably dextrorotatory, to the extent of about + 3° to
+ 15°, but is rarely slightly laevorotatory. It commences to boil at 156°
to 157°, and in good samples 88 per cent to 99 per cent will distil be-
low 165°. Its chief constituent is pinene Ci^H^g and a little dipentene
is also present.

French Oil of Turpentine. — This variety is chiefly obtained from
the oleo-resin of Pinus pinaster. Here again the chief constituent is
the terpene pinene, and the great difference between this and American
turpentine lies in the fact that the former is lasvorotatory, about - 18°
to - 40°.

German Oil of Turpentine is chiefly the product of Pi^ius sylvestris,
but Pinus abies, Pinus vulgaris and Pifius picea also contribute to it.
Its specific gravity is -860 to -870, and it is dextrorotatory, about + 15
to + 20°. It contains pinene and sylvestrene.

Bussian and Siuedish Oil of Turpentine. — This variety is almost
entirely obtained from Pinus sylvestris and Pifius ledebourdii. In
general properties it resembles German oil, but it is rather more vari-
able in specific gravity, etc. According to Tilden, it contains as much
as 60 per cent to 70 per cent of sylvestrene. Its specific gravity is
usually about -870 to -875, and its boiling-point about 170°. It is
dextrorotatory to the extent of + 20°. Of all the commercial turpen-
tines it is of the least technical importance. It often possesses a dis-
agreeable empyreumatic odour, due to the presence of the products of
destructive distillation of the pine-wood.

Other less important turpentines are Hungarian (from Pinus
pumilio) ; Austrian (from Pimis laricio) ; Carpathian (from Pinus
cemhra), and Finnish (similar to German). In addition, Venetian
turpentine and Canada balsam yield oils. These latter, however, have
practically no commercial interest.

ESSENTIAL OILS OF THE BRITISH PHARMACOPCEIA. 623

The terebene of pharmacy consists of optically inactive terpenes, the
result of the action of sulphuric acid on turpentine, which causes a
certain amount of isomerization, and also changes the active terpenes
into their inactive variety. For the manufacture of this, it is preferable
to employ rectified oil of turpentine. Indeed, for pharmaceutical pur-
poses in general, it is usual to employ turpentine purified by redistilla-
tion.

Turpentine is sometimes adulterated with petroleum and with
rosin spirit, and, rarely, with volatile portions of shale oil and coal tar.
It is itself used very largely to adulterate other essential oils, both on
account of its price and because it so closely resembles many other
oils in chemical constitution.

The accompanying table (p. 624) is given by Allen {Commercial
Organic Analysis), embracing certain properties of these bodies.

The chief points of importance to be noted in the examination of
the oil are the specific gravity, boiling-point and temperature of distil-
lation, optical activity, and flashing-point.

Good commercial turpentine has a specific gravity of -858 to '870,
only occasionally passing these limits shghtly. Russian oil has a
higher gravity — often reaching -875. The optical activity, as stated
above, varies with the source, and this factor is only of value when
studied in conjunction with the other features of the oil. The boihng-
point is usually 155° to 156°, and a considerable portion distils at be-
low 160°. In the best class of oils at least 85 per cent distils below
165°, often several degrees below this temperature. Russian oil, on
the other hand, distils chiefly between 170° and 180°. When adulter-
ated, the temperature of distillation rises gradually, and no large frac-
tions are obtained at any definite temperature when the adulteration
is at all excessive. The presence of ordinary petroleum spirit lowers
the flash point of turpentine. When pure, it flashes at 92° to 95° F.
when tested in Abel's flash-point apparatus. With only 1 per cent
of ordinary petroleum spirit this temperature is reduced by 10°.

According to Armstrong, a good indication of the presence of the
usual adulterants is obtained by distillation with steam. A current of
steam is allowed to pass into a definite volume of the turpentine con-
tained in a flask attached to a condenser. Unless it has been allowed
free access to the air for some time, the genuine oil leaves only traces
of non-volatile matter, but old samples may leave up to 2 per cent.

Usually, however, the presence of more than -5 per cent after
steam distillation indicates the presence of unvolatilized petroleum oil.
This is easily recognised by its low specific gravity and its fluorescence
when dissolved in ether. If the residue consists of resin oil, it will
form a bulky soap when rubbed with slaked lime. The specific gravity
of the fractions coming over with the steam will largely assist in de-
termining the presence of volatile adulterants.

For the approximate estimation of the amount of petroleum
naphtha in adulterated turpentine, Armstrong recommends the fol-
lowing process : 500 c.c. of the sample are placed in a separator and
treated with about 150 c.c. of sulphuric acid (two volumes of acid to
one of water). The mixture is cautiously agitated, and if much rise

624

FOOD AND DEUGS.

-2
o

1
1

g
1

ii^

»i

M QQ

-a

-2

.2
"S

ill

4) O

gee
II

O
CO

-2
o

'2 "S

'53 e8

o ^
J3 "^

0)
O

ID
CO

eS o

.1
If

3-2

"o
eg

"to

^ rH

2 2

CO
O

d
0

to ><

-2

>;,

-£3

— •
O

r-l

_>■

C CS

C3 0

«'s5

•■C

03 «

■4S

-»2
CO

o ^

C<J

a a

•

1 •

(X) •

•

s
.2

Vis

i .

CO •

"^i

Jo
'So '-I

_>»

'% .

■-I QQ

■^

-3

.s •
g

> — <

■N

■ll

S 2
-§1

C3-I

r-5

eri

T)5

«D

ESSENTIAL OILS OF THE BRITISH PHARMACOPCEIA. 625

of temperature is observed, the separator must be placed in cold water
for a short time. The turpentine is gradually converted into a viscid
oil, and when this has taken place, and no more heat is developed on
repeated agitation, the acid is tapped -off. The oily layer is then
transferred to a flask and subjected to steam distillation. When all
that is volatile with steam has passed over, the oily portion of the
distillate is separated from the aqueous layer, and heated with half its
volume of sulphuric acid previously diluted with one-fourth of its
measure of water. The mixture is well agitated, the acid liquid sepa-
rated, and the oily layer again distilled with steam. When genuine
turpentine is operated upon, the volatile portion of this second treat-
ment consists merely of cymene and a small quantity of parafiinoid
hydrocarbons. It never exceeds 4 to 5 per cent of the volume of the
original sample, and with care is as low as 3 per cent. If the volume
notably exceeds 5 per cent, it is advisable as a precaution to repeat
the treatment with the (4 to 1) acid. When treated in this manner,
petroleum naphtha is not appreciably affected, hence the proportion
may be fairly estimated by making an allowance of 4 to 5 per cent
from the volume of volatile oil which has survived the repeated treat-
ment with sulphuric acid. A further purification may be effected by
violently agitating the surviving oil with several times its volume of
concentrated sulphuric acid heated to 50° or 60°. This treatment can
be repeated if necessary, after which the residual hydrocarbon is sepa-
rated, steam-distilled, and again measured, when the surviving oil
from pure turpentine oil will not exceed from one half to one per cent
of the original sample. Any excess over this will be the minimum
quantity of petroleum naptha present. Shale naphtha cannot be at all
estimated in this way. The behaviour of the oil on distillation is the
best indication of the presence of rosin spirit, as the temperature rises
gradually, and no considerable fraction is obtained at 158° to 160° if
much rosin spirit be present.

VOL. I.

CHAPTER XL

FATTY OILS, WAXES AND SOAPS OJ^ THE BEITISH
PHAEMACOPCEIA.

The examination of the fixed oils is well understood, and the only
difiiculty in the analysis of these bodies is the interpretation of results,
which is rendered all the more difficult on account of the fact that the
figures obtained for various oils frequently overlap to a considerable
extent. Apart from the determination of the usual physical consta-its
of an oil, the following are special processes which are usually neces-
sary to adopt in their examination : —

Saponification. — The saponification value, or Koettstorfer value, of
a fat or wax is the number of mgs. of KOH requisite for the
complete saponification of a given sample. As a rule a fatty oil con-
tains some free oleic or similar acid, and a certain amount of alkali is
necessary to neutralize this, but generally speaking one understands
the saponification value to include the amount of KOH necessary for
the neutralization of the free acids as well as that for the hydrolysis of
the esters. If these values are expressed separately they become the
:acid and ester values respectively, being the number of mgs. of
KOfI necessary to neutralize the acids, and decompose the esters
respectively.

For the determination of these values the following process should
fee used. About 2 grms. to 4 grms. of the oil or wax accurately
■weighed is warmed with about 10 c.c. of alcohol, and well agitated
with it. A few drops of phenol-phthaLein are added, and alcoholic
potash of about semi-normal strength run in until the pink colour
is permanent. The amount of alkali used is noted and a further 25 c.c.
of the alcoholic potash solution run in. The liquid is now boiled briskly
for thirty minutes under a reflux condenser. After cooling it is diluted
with 50 c.c. of water, and semi-normal hydrochloric acid run in until
the pink colour is discharged. A blank experiment is conducted at
the same time, using the same reagents but omitting the oil. This will
give the exact value of the potash solution. From these results, the
amounts of potash used for the neutralization of the free acids and for
the hydrolysis are given, and expressed in mgs. per gram of the
sample give the acid and ester — or added together — the saponification
value of the fat.

Characters of the Fatty Acids. — In order to determine the amount
and character of the fatty acids, 10 grms. — or 5 grms. if it is not
necessary to make a very full examination — of the sample is saponified

(626)

FATTY OILS, WAXES AND SOAPS. 627

with 125 c.c. of a semi-normal alcoholic potash solution. The bulk of
the alcohol is evaporated, water added, the whole warmed to ensure
complete solution, and the fatty acids liberated by HCl, and allowed
to rise to the surface. The lower layer is run off, the fatty acids
washed twice in a separator with hot water, and finally filtered, dried
and weighed. The iodine value, neutralization value and melting-
point can then be determined on these in the usual manner. The
neutralization value is the number of mgs. of KOH necessary to
neutralize 1 grm. of the fatty acids. This is determined by dissolving
about 1 grm. in 10 c.c. of alcohol, and titratinsj with semi-normal
alkali, using phenol-phthalein as indicator. The mean molecular
weight can be calculated from this value, as the molecular weights of
the free acids, and that of potassium hydroxide are in direct proportion
to the amounts of the two which neutralize each other. The melting-
p^int is determined in the usual manner in a capillary tube, the acids
being first cooled on ice and allowed to stand thereon for an hour be-
fore the determination is made. The iodine value is determined as in
the case of the oil itself, except that the fatty acids do not require the
addition of any chloroform, as they are soluble in the alcoholic solu-
tion of iodine.

Unsajwnijiable Matter. — The unsaponifiable matter may be deter-
mined on the quantity of oil used for the determination of the saponi-
fication value. The saponification liquor is evaporated on a water
bath, with the addition of water, until the alcohol is driven off. The
cold aqueous solution is then transferred to a separator and shaken
with its own volume of ether. This is allowed to separate, and then
the aqueous layer is run off. This is repeated twice, and the mixed
■ethereal solutions are washed in a separator with a little distilled water.
The water is run off, and the ether filtered if necessary. The ether is
evaporated and the unsaponifiable matter dried and weighed.

Determination of the Hehner Value. — This value is generally
understood as the percentage of insoluble fatty acids present in an
oil.

Three to 5 grms. of the oil are saponified with alcoholic potash in
the usual manner, and the alcohol removed by evaporation. The soap
is dissolved in water, and .decomposed by excess of sulphuric acid.
The liquid is then warmed until the free fatty acids float on the sur-
face of the liquid as an oily layer. A weighed quantity — about 2 grms.
to 3 grms..ot dry hydrocarbon wax — is now added, and when melted the
whole is well stirred and allowed to cool. The solid cake of fatty
acids and neutral wax is then freed from the aqueous liquid by piercing
the cake by a glass rod, and pouring off the liquid. Hot water is then
added, and the cake remelted, st rred with the water, and allowed to
set again. This washing is repeated until the wash water is free from
acid, when the cake is removed, adherent water removed by absorbent
paper, and the cake transferred to a porcelain dish, dried at 105°, and
weighed. The weight, minus the weight of wax added, gives the fatty
acids insoluble in water. Most fats contain about 95 per cent ol
such acids, but the following deviate to a considerable extent from
this figure : —

628 FOOD AND DKUGS.

Per cent.

Maize oil 88-5 to 95

Shark liver 88 „ 93

Many fish oils of uncertain origin . . . . 75 ,, 85

Cocoa-nut oil 88 ,, 90

Japan wax 89 ,, 92

Butter 86 „ 90

TJie lodme Value. — The iodine value is an expression of the amount
of iodine (as a percentage of the fat used) which will coed bine with the
fat under definite conditions. This value denotes the amount so com-
bined, when the conditions laid down by Hubl are observed. An al-
ternative process is that of Wijs, but as Hubl's process is still generally
employed, the iodine value without qualification is here intended to
mean the value as determined by Hubl's process. The Wijs values
are qualified by the use of the name of the chemist responsible for the
process.

Hubl's process is carried out as follows : —

Solutions Necessary. — The Iodine Solution is prepared by dissolving
25 grm3. of pure iodine in 500 c.c. of 95 per cent alcohol : and 30
grms. of mercuric chloride in a separate portion of 500 c.c. of alcohol
of the same strength. The two solutions are then mixed, and allowed
to stand for at least three days before use (twenty-four hours is gen-
erally said to be sufficient, but the author finds a gradual diminution
in iodine value of the mixed solution goes on for quite two days, and
it is safer to allow the solution to stand for three days). A small dim-
inution goes on for a long time after the three days, but this is not im-
portant, as standard blank experiments are always carried out when
an iodine determination is being made.

The Thiosulphate Solution. — About 25 grms. of pure sodium thio-
sulphate are dissolved in 1000 c.c. of water. This is standardized at
least once a week by weighing out about 0*25 grm. of pure re-sublimed
iodine, dissolving in 5 c.c. of water containing 3 grms. of potassium
iodide, and diluting to 25 c.c. The thiosulphate is run in from a bur-
ette, until the yellow colour is nearly discharged, when a little starch
solution is added. The thiosulphate is then carefully run in until the
blue colour is just discharged. The number of c.c. used divided into
the number of mgs. of iodine used will give the iodine value of the
thiosulphate solution.

From 0"1 to 0'2 of the sample — according to its probable iodine value
— is weighed into a stoppered bottle holding about 250 c.c, and 10 c.c.
of pure chloroform is added ; 25 c.c. of the iodine solution are then
added, and the whole mixed by rotating the bottle. The bottle is then
allowed to stand, concealed from the light, for about eighteen hours.
At the same time a blank experiment is conducted, the bottle contain-
ing the reagents in the same quantity, but none of the sample.

At the end of eighteen hours, the contents of the bottles are titrated.
The bottle containing the sample should be of a deep brown colour,
so that not more than half the iodine used has been consumed. If on
titration it should be found that substantially less than half the original
quantity of iodine remains, then the recorded iodine value is probably
too low, and the experiment should be repeated.

FATTY OILS, WAXES AND SOAPS. 629

To each of the bottles 10 c.c. of a 10 per cent of potassium iodide
solution in water is added, and after gentle agitation to mix the liquids,
50 c.c. to 100 c.c. of water are added. Thiosulphate solution is then
run in, until the yellow colour of the aqueous solution and the red
colour of the chloroform are nearly discharged. The liquid is then well
agitated in order to cause the remaining iodine to pass entirely into
the aqueous solution, and a little starch solution added. Thiosulphate
solution is then run in until the colour is discharged, and does not im-
mediately return on agitation of the contents of the bottle. The
number of c.c. required for the blank experiment, minus the number
required for the sample, gives the iodine in terms of thiosulphate that
has combined with the oil. Since the iodine value of the thiosulphate
solution is known, the actual iodine absorbed is easily calculated, and
from this the iodine value (per cent of iodine absorbed) is deduced.
In the case of fatty acids, the chloroform may be omitted.

The Wijs process depends on tlie use of a solution of iodine trichloride
in glacial acetic acid. The advantages of this solution are that it does
not alter materially in strength even after keeping for a long time,
and that the process can be completed in under an hour. The ab-
sorption should be allowed to go on for thirty minutes, or in the case
of drying oils with an iodine value of over 100, for one hour.

The iodine solution is prepared by dissolving 9-4 grms. of iodine
trichloride and 7*2 grms. of iodine in glacial acetic acid on the water
bath, each being dissolved in separate portions. The solutions are
then mixed and made up to 1000 c.c. with the acid. The chloroform
used in Hubl's process must be replaced by carbon tetrachloride, as
chloroform often contains traces of alcohol, which interfere with the
reaction.

Otherwise the test is carried out as described above, except that
the absorption should only go on for thirty to sixty minutes. The re-
sults are practically identical with those obtained by Hubl's process
(but this is not the case with resins, which appear to behave in an
erratic fashion with the Wijs solution).

The Bromine Thermal Value. — This test, due to Hehner and
Mitchell, depends on the rise in temperature of a given quantity of
the oil when mixed with a given quantity of bromine under definite
coaditioQS. This value is in close relation to the iodine value, and the
latter may be approximately calculated fi-om the bromine value. A
convenient method of applying this test is as follows : —

Five grms. of the oil are dissolved in 25 c.c. of chloroform, and 5
c.c. of this solution are placed in a small Dewar's vacuum tube, taking
care that the liquid does not flow down the side of the tube. The
temperature of the liquid is taken with a thermometer graduated in
one-fifth degrees. A solution of 1 volume of bromine in 4 volumes of
chloroform is prepared, and brought to the same temperature as the
oil solution, poured into the vacuum tube, the whole gently stirred with
the thermometer, and the rise in temperature noted.

Hehner and Mitchell in their original test use 1 c c. of pure bromine
and 1 grm. of oil in 10 c.c. of chloroform. Under these circumstances
the bromine thermal value x 5-5 gives a very close approximation to

630 FOOD AND DRUGS.

the iodine value. The following values were obtained by Hehner and
Mitchell :—

Lard 10-6°

Butter 6-6°

Olive oil 15°

Corn o 1 21-5°

Cottonseed oil 19- 1°

The Detection and Sejjaration of Cholesterol and Phytosterol. —
These similar, and probably isomeric, alcohols are found, the former in
numerous animal oils, the latter m most vegetable oils. The presence
of phytosterol is considered conclusive evidence of the presence of a
veg etable fat. The best method for the separation of these bodies is that
of Borner (" Zeit. Unter. Nahr. Genuss." 1898, 1, 31). He saponifies
100 grms. of the fat with 200 c.c. of 20 per cent alcoholic potash, dilutes
the liquid with 400 c.c. of water and shakes the whole, when cold, with
500 c.c. of ether. This is separated, and the soap solution extracted three
times more with 250 c.c. of ether. The ether is distilled off and any
traces of alcohol present removed by heating on the water bath. The
residue is again boiled with a little alcoholic potash in order to be
certain that all fat is saponified, and the liquid diluted with water and
again thoroughly extracted with ether. The ether is washed with
water and filtered and evaporated, leaving the crude cholesterol and
phytosterol. In the case of animal fats, the residue is chiefly choles-
terol, whilst with vegetable fats it is mostly phytosterol. It is dis-
solved in about 10 c.c. to 15 c.c. of absolute alcohol with the aid of heat,
and the liquid allowed to deposit crystals in a shallow dish. In the
case of cholesterol alone, crystallization commences at the margin and
gradually extends towards the centre of the liquid on the surface. This
crop, which soon separates, is separated and dried on blotting paper
With phytosterol no surface film is formed, but needles are separated
from the margin inwards, under the surface of the liquid. These
crystals are best separated by filtration. The crystals may be washed
with a very small amount of absolute alcohol and then examined
microscopically. The general appearance of (1) pure cholesterol, (2)
pure phytosterol, (3) mixtures of both, are shown by the following
diagrams on opposite page.

The crude alcohols may also be dissolved in the smallest possible
amount of absolute alcohol and allowed to crystallize, and after
microscopic examination, the crystals obtained, together with the
residue left after evaporating the alcohol, are boiled in a small dish,
covered with a watch-glass, with 2 c.c. to 3 c.c. of acetic anhydride.
The excess of acetic anhydride is driven off on the w^ater bath, and the
residue dissolved in a little hot absolute alcohol, so that crystallization
shall not take place directly the alcohol cools. Allow the crystals to
separate slowly, and when about half the alcohol has spontaneously eva-
porated, remove the crystals with a spatula, and wash them in a small
filter with 3 c.c. of 95 per cent alcohol. Redissolve in 5 c.c. to 10 c.c. of
absolute alcohol and again allow to crystallize. The crystallization
should be repeated five to six times, the melting-point being deter-

^^^^^^ mi

FATTY OILS, WAXES AND SOAPS.

631

mined after the third crystalHzation onwards. Cholesterol acetate
melts at lll'S" to 114-8°, whilst phytosteryl acetate melts at 125*6° to
137° according to the som'ce Irom which it is obtained. If the
crystals melt at 116°, vegetable oil is probably present; if at 117° or
over, the presence of a vegetable oil is certain.

The Beichert Value. — The Eeichert (or Eeichert-Meissl) value in-
dicates the number of c.c. of decinormal potash solution requisite for
the neutralization of that portion of the soluble volatile fatty acids ob-
tained from 2-5 grms. (or 5 grms.) of the fat when saponified and
distilled by Eeichert's method.

The Reichert value refers to 2-5 grms. of the fat, whereas the
Eeichert-Meissl or Eeichert-Wollny value refers to 5 grms. of the fat.

Five grms. of the fat are accurately weighed into a flask of about 200

(1)

■»oVl

Fig. 57.— (1) Cholesterol ; (2) Phytosterol ; (3) Mixture of both.

C.c. capacity. About 2 grms. of stick potash and 50 c.c. of 70 per cent
alcohol are then added, and the oil is saponified on a water bath, and the
alcohol driven off completely. The soap is dissolved in 100 c.c. of
water, and 40 c.c. of 10 per cent sulphuric acid added to the liquid (cold).
A few small pieces of pumice stone are added to prevent " bumping ".
Distil from the flask through a Liebig condenser, placing a safety
bulb between the flask and condenser so as to avoid spurting, and collect
110 c.c, which should take about 1 hour to come over. One hundred c.c.
of this are filtered and titrated with decinormal potash using phenol -
phthalein as indicator. The value so obtained is multiplied by 1*1 and
this, in c.c. of potash solution gives the Eeichert-Meissl value. This is
not exactly double the Eeichert value, but is usually about equal to the

632

FOOD AND DRUGS.

Eeichert value multiplied by 2-2. Filtration of the distillate is necessary,
since a certain amount ol volatile insoluble fatty acids are distilled over.

This process being an empirical one, requires careful attention to
the exact details which should not be allowed to vary at all. (For
further details see under butter.)

Befractive Values. — The determination of the refractive index is
often a matter of importance with fats and oils, but it is usual to use
an instrument which is not graduated in absolute indices, but in arbit-
rary degrees, when the value is returned as the " refractometer number ".
Thi§ is determined on the Zeiss-Abb6 instrument, which is that found
in most laboratories, and is known as the " butyro-refractometer". In
quoting figures, the instrument should be quoted, as the values for a
Zeiss- Abb6 instrument differ materially from those of the Jean-Amagat
refractometer.

For further, details see under butter (p. 96).

The following table gives the limits of the usual values determined
analytically for the commoner adulterants of some of the fatty oils. It
is to be remembered that many of these oils are themselves edible, and
there is no reason that they should not be sold, so long as they are
properly labelled or described : —

Specific
Gravity.

Refractive ludex.

Sap.
Value.

Iodine
Value.

Refract. No.

Poppyseed

0-923 to 0-927

1-458 at 60"

190 to 197

130 to 145

63 to 64 at 40°

Sunflower

0-927 „ 0-926

1-461 „ 60°

193 „ 196

120 „ 132

72 „ 73 „ 2e5°

Maize

0-921 „ 0-926

1-4762 „ 20°

188 „ 193

110 „ 130

69 „ 20°

Cottonseed

0-922 „ 0-926

1-4740 „ 20°

193 „ 196

108 „ 112

67 „ 69 „ 20°

Sesame

0-923 „ 0-925

1-475 to 1-477 at 15°

189 „ 194

103 „ 114

68 „ 25°

Arachis

0-917 „ 0-921

1-4550 at 60°

190 „ 196

84 „ 102

66 „ 68 „ 25°

Mustard

0-915 „ 0-920

1-4750 „ 15°

170 „ 174

95 „ 110

58 „ 60 „ 40°

Rapeseed

0-915 „ 0-917

1-4725 to 1-4758 Ht 15°

170 „ 180

95 ., 105

68 „ 25°

Cocoanut

0-911 „ 0-913

1-4400 „ 1-4420 ,. 60°

245 „ 268

8 „ 10

34 „ 40°

THE FIXED OILS,

FATS, AND WAXES OF THE BRITISH
PHARMACOPOEIA.

Adeps Lan^.

Adeps Lanae, or wool "fat" is described in the Pharmacopoeia as
the purified cholesterin fat of sheep's wool. It is stated to be a
yellowish, tenacious, unctuous substance, almost inodorous ; melting
at 40° to 44-4° C, readily soluble in ether or chloroform, sparingly so
in alcohol. One grm. should dissolve almost completely in 75 c.c. of
boiling 90 per cent alcohol, the greater part separating in flocks on cool-
ing ; it should not yield more than 0-3 per cent of ash, and this should
not be alkaline. It should not contain more than 028 per cent of free
acids calculated as oleic acid. A solution in chloroform poured carefully
into sulphuric acid acquires a purple-red colour. Heated wdth caustic
soda solution no odour of ammonia should be evolved.

FIXED OILS, FATS AND WAXES. 633

Anhydrous wool fat — or wool wax as it is more properly termed —
is the natural grease extracted from the sheep's wool, purified and
freed from fatty acids. Its extensive use depends on the tact that it
forms an emulsion with 75 per cent of its weight of water, which is
readily absorbed by the human skin, so that it forms a useful vehicle
for certain forms of medication.

Wool wax consists of a complex mixture of esters and free alcohols.
Amongst the alcohols, cholesterol and isocholesterol are the principal.
Pure wool wax, freed from free fatty acids, should have the following
characters : —

Specific gravity at 15° 0-940 to 0-950

„ W 0-899 „ 0-908

Melting-point 35° „ 45«'

Saponification value 98 „ 105°

Iodine value 20 „ 30

Fatty acids 50 „ 60 per cent

Alcohols (determined as unsaponifiable matter) 40 ,, 50 „

Almond Oil.

The oil expressed from the bitter or sweet almond is described
officially as a pale yellow nearly inodorous oil with a nutty taste. Its
specific gravity is given as 0*915 to 0-920, and it should not congeal
until cooled to nearly - 4° F. If 2 c.c. of the oil be shaken with 1
c.c. of fuming nitric acid and 1 c.c. of water, a whitish, not brownish-
red, mixture should be formed, which after standing for six hours at
50" F. should separate into a solid white mass and a nearly colourless
liquid. (Absence of peach kernel and other fixed oils.)

Almonds yield from 35 to 45 per cent of fixed oil, which consists
essentially of glycerides of oleic and other liquid unsaturated fatty
acids. No solid fatty acids — or not more than traces — are present in
the oil. The usual adulterants of this oil are the fixed oils of the apri-
cot and peach kernel, and of recent years a good deal of hazel nut oil
has been used for the purpose of sophistication. From time to time
other oils such as arachis, sesame, and olive oils have been used but
these are not commonly met with.

Pure almond oil should have the following characters : —

Specific gravity at 15°
Saponification value
Iodine value ....
Refractive index at 15°
Butyro-refractometer No. at 15°
Melting-point of fatty acids .
Solidifying point ,, ,, .
Neutralization value of ,, (per cent KOH)

Hazel nut oil may be readily recognized by the characteristic taste
of the hazel nut. The iodine value of this oil is about 84 to 88 so that
any considerable quantity will be indicated by a reduced iodine num-
ber. The fatty acids of hazel nut oil melt at from 19"" to 25°, so that
this fi'Jure will be raised if much of this oil is present.

Poppy seed oil has been found in a number of samples by the
author during the past few years. This will be indicated by a higher

0-914

to 0-920

189

„ 196

„ 104

1-4710

„ 1-4728

„ 71

13°

„ 14°

9-5

„ 11-5°

20-4

634 FOOD AND DKUGS.

specific gravity, a higher refractive index, and a higher iodine value,
that of poppy seed oil being from 135 to 140.

Practically no tests exist which will definitely prove the presence
of apricot or peach kernel oils, except, to some extent, colour reactions.
The official nitric acid test (see above) is a useful one, as both these
adulterants give a yellow to red-brown fatty mass, when shaken with
the acid.

Bieber's test is also fairly reliable. It consists in shaking 5 volumes
of oil with one volume of a mixture of equal parts by weight of con-
centrated sulphuric acid, fuming nitric acid and water. Pure almond
oil does not change colour, while apricot kernel oil gives a pink colour,
and peach kernel oil a faitit pink colour after standing for some time.

Figures have been from time to time published to show that the
usual quantitative determinations may be of assistance in discriminat-
ing between pure almond oil and mixtures with peach or apricot
kernel oil, but from the experience of a very large number of samples,
the author has no hesitation in saying that these figures overlap so-
much that they are perfectly useless for the purpose of detecting these
two oils.

Lewkowitsch ("Analyst," xxix, 105) gives the figures on opposite^
page, for almond, peach, and apricot kernel oils : —

Croton Oil.

This oil is expressed from the seeds of Croton tiglium. The specific
gravity is officially given as 0-940 to 0*960. The oil should be soluble
in absolute alcohol, ether, and chloroform. An alcoholic solution should
not redden litmus. If 2 c.c. be shaken with 1 c.c. of fuming nitric
acid, and 1 c.c. of water, the mixture should not solidify, but only
thicken slightly after standing for two days (absence of other non-
drying oils).

Croton seeds contain about 55 per cent of fixed oil.

The oil contains glycerides of various fatty acids, amongst which
are the lower fatty acids, such as formic, acetic, butyric, etc., so that a
high Eeichert value is always found. A small quantity of a resinous
matter, probably of a lactone nature, exists, and is probably the purga-
tive principle of the oil.

The statement which is contained in the Pharmacopoeia that croton
oil is soluble in absolute alcohol requires some quahfication. According
to Lewkowitsch, this is only true where the oil has been extracted from:
the seeds by alcohol. It is only true for expressed oils if less than an
equal volume of alcohol be used : more alcohol at once causes turbidity..

Pure croton oil has the following characters : —

Specific gravity at 15°

. 0-937 to 0-943

Refractive index at 15°

. About

1-4770

Saponification value .

. 197

to 215

Iodine value ....

. 101

„ 112

Butyro-refactometer No. at 40° .

. 67

M 69

Melting-point of fatty acids

. About

22°

Solidifying point of fatty acids .

. 18°

„ 19°

Neutralization value of fatty acids

. 20

„ 20-5 (per cent KOH)

Reichert-Meissl value

. 12

„ 14

Acetyl value ....

. 25

„ 36

FIXED OILS, FATS AND WAXES.

635

Colourless.
Colourless.
Colourless.
Colourless.
Colourless.
Colourless.
Colourless at first

then pink.
Pink coloration.

Slightly pink.

Very slightly pink.

«3 t> oi rH i>- !yi o p t- qp

t-rH^lCCOWIib c^ o S

is
0
li

00 Tt< op 00 cc th op p p op
i>.«bQb«bu5c--?b do tih t>-

O 05 CS Ci CS CS O C5 CS 05
(>lr-lr-iiHr-(i-Hi-l r-i iH iH

-H p t- .-1 CI «p O «5 op CNl

lo I?? o cb oi th sc (fa (fq tH

Butyro-

Refracto-
meter at
40° C.

O lO o >o o p »o <z> <^ <C>
b- t- t- «c i>- t>- t> do t> 00

lO IC W5 »0 lO >0 iO »0 »C »0

Saponifica-
tion
\alue.

«5 r^ CC :0 (M S •* -* (M 00

sss^i^s § g i

(Mi-llHrHrHrHr-t rH i-H ,H

^^ 11

is^glsgs § ?2 i

T-Hr-I-Hi— 1.— Ii-Hi-I (M 1— 1 (M
ppppppp p p p

ooooooo o o o

Almond oils, expressed from : —

1. Valencia sweets

2. Blanched Valencia sweets

3. Sicily sweets .

4. Mazagan bitters

5. Small Indian almonds

6. Mogador bitters

7. Peach kernel oil

8. Apricot kernel oil .

9. Apricot kernel oil from

Mogador kernels .
10. Californian apricot kernel
oil ... .

636 FOOD AND DRUGS.

This oil (which is not a noa-drying oil, as would be indicated by the
wording of the Pharmacopceial monograph, but a semi-drying oil) is
not often adulterated. Most other fixed oils are revealed by a lowered
specific gravity, and a reduced Reichert value. Castor oil is the
only adulterant met with by the author, and this would be detected
by a high acetyl value, a lower Reichert value, and a higher specific
gravity.

Linseed Oil.

Linseed oil is expressed from the seeds of Linum usitatissimum,
which yield about 35 per cent of the oil, the pressed cake retaining about
10 per cent. Linseed oil is one of the most typical of the drying oils,
and is used for technical purposes to a very large extent, its use in
medicine being very small. For technical purposes valuations of the
oil are required which are not necessary when dealing with the oil
from a pharmaceutical point of view.

The only standards given in the Pharmacopoeia are that the specific
gravity should be from 0*930 to 0*94:0 : the oil should be soluble in 10
parts of 90 per cent alcohol, and in turpentine. It gradually thickens
by exposure to the air. It does not congeal above - 20° G.

Linseed oil is, even when generally accepted as pure, a mixture of
the oil from linseed with that from a small quantity of hemp or rape
seed. This is due not to deliberate adulteration, but to the fact that
the seeds are more or less accidentally mixed, owing to the fact of the
proximity of the plants, which grow together in certain districts.

This oil consists of glycerides of palmitic and myristic acid (10 per
cent) but principally of the glycerides of liquid fatty acids of which the
principal are linolic, linolenic and isolinolenic acid, all being highly un-
saturated. Pure linseed oil should have the following characters : —

Specific gravity at 15° .
Refractive index at 15°
Saponification value .
Iodine value ....
Butyro-refractometer No. at 20° .
Melting-point of fatty acids
Solidifying point of fatty acids .
Neutralization value of fatty acids
Unsaponifiable matter

If a sample complies with the above figures it is practically certain
to be pure. The iodine value is characteristic, and if this falls below
170 the oil should be condemned. The only adulterations met with to
any extent are those with mineral and rosin oils, the latter especially.
Mineral oils will be indicated by the low specific gravity, the low
iodine, value, and the high unsaponifiable matter. Rosin oil will
raise the specific gravity, and lower the saponification value, at the
same time raising the amount of unsaponifiable matter. It is also
dextrorotatory, by which feature it can be detected if present in any
quantity. The amount of unsaponifiable matter however is the best
criterion of the presence of these adulterants. Rosin oil may also be

0-930

to 0-941

1-4830

„ 1-4845

190

„ 196

172

„ 192

„ 85-5

19°

„ 23°

13°

„ 16°

19-5

„ 20 (per cent of

KOH)

Under 1 per cent

FIXED OILS, FATS AND WAXES.

detected in the following manner : Warm 5 c.c. of the oil with 10 c.c.
of 90 per cent alcohol. When cold separate the alcoholic liquid,
and evaporate the alcohol. Dissolve the residue in a few c.c. of acetic
anhydride, and carefully pour on to the surface of the liquid a few
drops of cold 50 per cent sulphuric acid. A fine violet colour, which
is transient, results if rosin oil be present.

Most other oils if present will be revealed by the failure of the oil
to comply with the standards above given. But as linseed oil is a
cheap oil, most other oils are precluded from use as adulterants,

COD LIVER OIL.

Cod liver oil is required by the PhaTmacopoeia to have been freed
from solid fat by filtration at about - 5° C. The oil should be obtained
from the livers of the codfish only, and is required to have the follow-
ing characters : —

Specific gravity at 15° 0920 to 0-930 : no solid fat should separate
by exposing the oil to a temperature of 0° C. for two hours : a drop of
sulphuric acid added to a few drops of the oil on a porcelain slab de-
velops a violet coloration : when nitric acid is carefully poured inta
some of the oil contained in a test tube, a precipitate of coagulated
albumen should be formed at the surface of contact of the two
liquids.

Cod liver oil, to be fit for medicinal purposes, should be prepared by
steaming the livers within twenty-four hours after the fish are caught.
The pale cod liver oil of pharmacy results by this treatment. When
the fishing boats are unable to come ashore quickly, the fish are killed
on board, and the livers stored. These may be brought to shore after
several days or a week or more when they are often in a more or less
decomposing condition. The oil obtained from these livers may be a.
pure cod liver oil but it is dark brown, and has an objectionable odour,
and is only fit for veterinary purposes. At one time there was believed
to be a great difference between the Norwegian cod liver oil, and the
Newfoundland oil. This difference probably was due to the fact that
other livers were formerly used in the preparation of the Newfoundland
oil, and the livers were not pressed in a sufficiently fresh state. To-
day, however, the difference is somewhat sentimental, as Newfoundland
oil can be obtained of the highest grade, and perfectly pure. Oil of
high grade is also made now on the East Coast of Scotland. The
standards of the British Pharmacopoeia are totally inadequate to dis-
criminate between cod and many other liver oils. Indeed few fatty
oils give so much difficulty to the analyst as this one, and in some
cases it is impossible to decide whether an oil is pure or not.

The specific gravity given in the Pharmacopoeia is common to a
whole group of liver oils : the sulphuric acid colour reaction is equally
common to a number of liver oils ; and the nitric acid test for albumen
is quite unreliable. New standards for this oil are therefore obviously
required. A genuine cod liver oil should have the following char-
acters : —

638 FOOD AND DEUGS.

Specific gravity at 15° . . . 0-920 to 0-930 (rarely up to 0-932).

Eefractive index at 15° . . . 1-4800 to 1-4825.

Saponification value . . . 380 to 190 (rarely a little higher or lower).

Iodine value 158 to 168.

Butyro-refractometer No. at 15° . 81 to 86.

Neutralization value of fatty acids . 19-7 to 20-3.

Mean molecular weight of fatty acids About 290.

Melting-point of fatty acids . . 21-25.

Iodine value of fatty acids . . 165 to 172.

A properly-prepared cod liver oil will contain no glycerides of
volatile fatty acids, and will never give a Eeichert value of more than
1. Any higher value than this indicates the decomposition of the livers
used in preparation of the oil.

The amount of unsaponifiable matter present in this oil is a very
important determination. Genuine cod liver oil contains a Httle chol-
esterol, but in medicinal oils the total amount of unsaponifiable matter
rarely exceeds 1-0 per cent, often being less than 0*5 per cent. Cer-
tainly any higher amount than 1*5 per cent should be condemned, as
this is almost certainly due to the presence of other oils, of which the
most usual is shark liver oil which usually contains as much as 5 to 8
per cent of unsaponifiable matter.

The amount of free fatty acids is important, and oils for medicinal
use should not contain more than 0-6 per cent, or at most 0-8 per cent
of free acids calculated as oleic acid. Higher values point to crude
unrefined oils.

Apart from the adulteration with other fish oils which have very
-similar analytical values, cod liver oil is sometimes — although rarely —
adulterated with vegetable oils. This is, as a rule, at once indicated
by the lower iodine value. In cases of doubt the phytosteryl acetate
test may be applied (see p. 630).

Numerous colour tests have been recommended for this oil.
Most of these are absolutely useless, but the following are useful within
certain limits : —

If one part of the oil be dissolved in five of carbon disulphide, and
a few drops of concentrated sulphuric acid added, a fine blue colour
will result, which is more purple if the oil is rancid. This colour test
is a general one for liver oils, and not restricted to cod liver oil. It is
not yielded by oils from other parts of the fish, such as blubber oil.
The following is also a general test for liver oils : —

One gram of the oil is dissolved in 5 c.c. of chloroform in a test
tube, and shaken with 2 c.c. of freshly prepared solution of phospho-
molybdic acid. A blue ring is formed at the zone of contact of the
liquids, in the presence of a liver oil.

Except in rare instances, where perhaps the oil has been exposed
to the influence of air and light for some time, the following reaction
is yielded by pure cod liver oil (but other oils may be in admixture
with the cod liver oil without interfering \^ith the reaction). If 10
drops of nitric acid be stirred for a minute with 5 c.c. of cod liver oil
on a white tile, a pale rose colour results, which after standing becomes
pale yellow. In the presence of some liver oils, the colour will be deep
red, very soon changing to a dirty brown instead of a pale yellow.
This reaction is only of value if a positive result is obtained.

FIXED OILS, FATS AND WAXES.

639

The following values are those (Lewkowitsch) of some of the liver
oils which are used for adulterating cod Hver oil : —

Liver Oil from

Specific Gravity.

Saponification
Value.

Iodine Value.

Acid Value.

Unsapon.
Per cent.

Skate

0-9307
(at 15° C.)

185-4

157-3

—

0-97

Tunny

—

155-9

0-2 to 34

1-0 to 1-8

Haddock

0-9298
(at 15° C.)

188-8

154-2

—

1-1

f 0-925

177 to 181

123 to 137

1-26 to 1-68

—

Coal fish

\ (at 15° C.)

137 to 162

7-2; 21-6

I 0-9272

186-1

139-1

2-8

6-52

Ling

0-9200
(at 15° C.)

184-1

132-6

10-9

2-23

Shark (Arctic)

0-9163
(at 15° C.)

161-0

114-6

—

10-2

>> n

0-9105 to 0-9130

146-1 to 148-5

111-9 to 114-9

2-6 to 6-2

20-8 to 21-5

„ (Japan)

0-9156 „ 0-9177

163-4 „ 163-5

128-3 „ 136

0-88 „ 1-5

14-4 „ 21-5

)) )>

0-9158

157-2

Bay

0-9280
(at 15-5^ C.)

—

Hake

0-9270
(at 15-5" C.)

The interesting figures on page 640 are due to Barclay, and repre-
sent a number of samples of cod liver and other fish oils.

Olive Oil.
This oil has already been dealt with under foods (see page 111).

Castor Oil.

This oil is required by the British Pharmacopoeia to have a specific
gravity between 0*950 and 0-970 : to be soluble in one volume of
absolute alcohol, and in 5 volumes of 90 per cent alcohol. Equal
volumes of castor oil and 'petroleum spirit are stated not to yield a
clear mixture at 15*5°, but in the presence of other fixed oils the
mixture is clear (as a matter of fact, the amount of other fixed oils
present materially influences this test). Another ofiicial test is given
which requires that 3 c.c. of the oil dissolved in 3 c.c. of carbon disul-
phide should not become brown when shaken with 1 c.c. of sulphuric
acid. This test is quite incorrect, and the author has never met with
a sample which literally answers it.

Castor oil consists principally of the glycerides of two or more un-
saturated hydroxy acids, riciholeic and isoricinoleic acids C^gHg^Og.

The glycerides of stearic and dihydroxystearic acids are also
present to a small extent, as well as traces of other glycerides.

There are numerous grades of castor oil, the quality employed in
medicine being white or at most pale yellow in colour, and usually
being a cold pressed oil.

640

FOOD AND DKUGS.

•5c
§

•-3

O O r-l O rH O 00 '

S
o

•<

CO
CSt-OO'MrHCOO 1

050t-05«0(?q!NO 1

O t- CO O rH O 00 '

«o

OSC^COO"^CO'-l(jq 1

O O ^ O r- 1 O 00 '
00 O 1 rH

£5

OSOOSOCOrHOOlC 1
t-CO 1 rH "^

1
1

O Oi O O .-H O CO '

•s

C35

Cit>C-0-*rHCSCO 1
t- «0 1 iH "

t;-

OCOt-OrHOOO iH
t^ «0 1 iH

2 u

c»ocoS^ao-Hcooo l

C-«0 1 rH

CO
CO

hi
O

«
>

O CiO O rH O 00 1

^ooo^-^cpoo
o»c«OrH*=^ooo2'-'

«0 U5 , rH "^

CO o

Cil>-CO-<**^W5CO00?O

6 6 rin 6 ^ 6 o "^ 6

CO t> 1 iH

CO
U5

o? c- cp Ss g « t> g «^^
6a.oo..ooo -H

oloookoo5«ooocoo

6 -^ ^ 6 ^ 6 6i "^ r^H
00 t> 1 " >H

o o o o th o 00 '

t-t- 1 rH

ft 1

5seoo^C5koc<j«oo5

00
OSOoSsOOiHt-lO 1

00 t- 1 -H

Ci CO t- S? (M cq o »o o 1
00 t- 1 "^ 00

o
-^

J^ "*

2
"5

'53

0?I>OOCO>0'*<000
l>«0 "^ rH

do 1

11^ 1 ^

sS eg "3 , «« . o
•O « « *3 ^ ^

lllllSSli

•do • •:2 • • •

1^1 '^ • "^

oils -^ • -^

§,-S -S o '^ W ffl 7. sa
cg^N N P2 ►S W W !> t^

f^2

FIXED OILS, FATS AND WAXES. 641

A medicinal castor oil should not contain more than from 1 per
cent to 2 per cent of free fatty acids calculated as oleic acid. It
should have the following characters : —

Specific gravity at 15° 0-955 to 0-968

(the official limits are too wide)

Saponification value 176 „ 186

Iodine value 82 „ 88

Acetyl value 145 „ 150

Refractive index at 15° 1-4790 „ 1-4810

Butyro-refractometer No. at 25° .... 77 „ 79

Optical rotation (100 mm.) 3° „ 5°

Solidifying point of fatty acids ..... 3°

Mean molecular weight of fatty acids . . . 290 to 303

Refractive index of fatty acids at 15° ... 1-4540 to 1-4558

Castor oil is distinguished from other fixed oils by its very high
acetyl value, its optical activity, and its solubility in alcohol and
petroleum spirit. Nearly every genuine sample will dissolve in 3 to 3-5
volumes of 90 per cent alcohol. The official test for this oil in refer-
ence to its solubility in petroleum spirit is misleading. It is true that
1 per cent of castor oil is not soluble in petroleum spirit, but castor
oil dissolves its own volume of the spirit, so that mixtures in equal
parts give a clear solution, but if more than 1 volume of the spirit be
employed with 1 volume of the oil a turbid mixture results. Most
samples also give a clear solution with one and a half times their
volume of kerosene, but not with larger quantities. Small quantities
of other fatty oils, the amount depending on the nature of the oil,
cause the oil to lose the characteristic insolubility, and such adulterated
oils are soluble in all proportions in petroleum spirit.

An acetyl value below 140 is strong evidence of adulteration, as
no oil, except perhaps grape seed oil (which is not used as an adulter-
ant) has an acetyl value anywhere near that of castor oil.

The sulphuric acid test of the Pharmacopoeia must be ignored as
it is quite incorrect.

Oil of Theobroma.

This has been dealt with under cocoa (see page 26).

Beeswax.

Beeswax is official both as white and yellow wax (Cera alba and
Cera /lava). The official tests for both varieties are identical, and are
as follows': —

The wax should have a specific gravity at 15° of 0*960 to 0-970 :
should melt at 62-5° to 63-9° : it should not yield more than 3 per cent
to cold 90 per cent alcohol : it should not yield anything to water or
boiling solution of caustic soda, the liquids filtered after such treatment
not giving any precipitate when acidified by HCl (absence of fatty
acid, Japan wax and resin) : 5 grms. should require not less than 1-6
c.c. of normal alkali to neutralize the free fatty acids present, and be-
tween 6-2 and 6-8 c.c. of normal alkali to saponify the esters present.
After heating 5 grms. with 25 grms. of sulphuric acid to 160° C. for
fifteen minutes and diluting the mixture with water no solid wax
VOL. I. 41

642

FOOD AND DRUGS.

should separate (absence of paraffin wax) . It should yield no reaction
for starch.

Beeswax consists essentially of a mixture of free cerotic acid
C^^HjgOg and myricin (myricyl ,palmitate) CaoHg^O . CO . Cj^Hgp

It also contains small quantities of other free acids, traces of free
myricyl and ceryl alcohols, and other bodies which are not well under-
stood. A small amount of hydrocarbons also exists in beeswax.

Beeswax is very often adulterated, the common adulterants being
paraffin or ceresin wax, Japan wax, tallow, stearic acid, resin, and
insect wax.

Carnauba wax used to be a common adulterant, but its price is
now too high for it to be so employed.

The usual adulterants to-day, are mixtures of some of the above,
so blended as to give analytical results very similar to those of pure
IX. Pure beeswax should have the following characters : —

Specific gravity at 15° . . 0-962 to 0-970

Specific gravity „ 100° . . 0-818

Melting-point .... 60-5°

Acid value .... 18

Ester value .... 72

Saponification value (total) . . 90

Iodine value . . . . 8

0-824 (water 15° = 1)

64°

21-2 (rarely as low as 16-8)

99 (rarely 88)

In judging the above figures it must be remembered that a large
number of samples of Indian beeswax give figures which are far out-
side the above limits. Whether these samples are due to abnormal
conditions or whether they are regularly adulterated with some un-
known adulterant is not yet definitely known. At all events such bees-
wax cannot safely be employed in medicine. The specific gravity at
15° is best taken by carefully melting the sample and cutting small
portions with a sharp cork borer, and mixing methylated spirit and
water of various strengths, so that in one mixture (at 15°) the frag-
ments just float, whilst in another containing a trace more of the
alcohol they just sink : the specific gravities of the liquids are taken,
and the mean of the two is taken as that of the wax. Care must be
taken that no air bubbles are adherent to the wax.

The ratio of the ester value to the acid value, the " ratio no." as it
is called, is f-airly constant, and will.be found to vary between 3-5 and
4:'l usually about 3-7. Most adulterants will upset this ratio.

The following figures are those of some of the commoner adulter-
'ank of beeswax : — *

Acid Value.

Ester Value.

Ratio Number.

Japan wax .
Chinese wax .
Spermaceti .
Myrtle wax .
Tallow ....
Stearic acid .

Resin ....
Parafl&n and ceresin

19 to 22
Traces

2 to 5

3 „ 5
190 „ 200

140 „' 165

200 to 210
78 „ 82
130 „ 135
205 „ 210
190 „ 198
None

20 to 30

About 11
Very high

Very infinitesi-
mal.
^ to*

FIXED OILS, FATS AND WAXES. 643

When mixtures for adulteration have been prepared so as to give
correct acid ester and ratio numbers, hydrocarbon wax and stearic
acid are usually present together with a wax, such as Japan wax, with
a high ester value.

These bodies must therefore be searched for.

Japan wax will be indicated by the presence of glycerin. If this
has to be determined the following process is the best. Twenty grms.
are saponified in the usual way, and the alcohol evaporated, the re-
sulting mass boiled with water, and excess of sulphuric acid added.
The separated waxy matter is filtered off, and washed with boiling
water, and the glycerin in the filtrate determined by Lewkowitsch's
process, which is as follows : —

The filtrate is neutralized with an excess of barium carbonate and
boiled down on the water bath until most of the water is driven
off. The residue is exhausted with a mixture of ether and alcohol,
and the ether-alcohol driven off for the most part by gently heat-
ing on the water bath, and the residue dried in a desiccator and
weighed. It is not necessary to dry until constant weight is obtained,
since the glycerol is determined in the crude product by the acetin
method.

This process is based on the conversion of glycerol into triacetin
when concentrated glycerol is heated with acetic anhydride. If the
product of this reaction is then dissolved in water, and the free acetic
acid has been carefully neutralized with alkali, the dissolved triacetin
can be easily estimated by saponifying with a known volume of
standard alkali and titrating back the excess. The solutions re-
quired are : —

1. Half normal or normal hydrochloric acid.

2. Dilute caustic soda, containing about 20 grms. of NaOH in 1000
c.c. Its strength need not be known accurately.

3. A 10 per cent solution of caustic soda.

The estimation of the glycerol is carried out as follows : —
About 1*5 grms. of the crude glycerin weighed accurately are heated
with 7 c.c. to 8 c.c. of acetic anhydride and 3 grms. of anhydrous sodium
acetate for one and a half hours in a flask, of about 100 c.c. capacity,
connected with an inverted condenser. The mixture is then allowed
to cool a little, 50 c.c. of warm water are poured down through the
tube of the condenser, and the acetin made to dissolve by shaking the
flask ; if necessary, the contents of the flask may be slightly warmed,
but must hot be boiled. As triacetin is volatile with water vapours,
these operations must be carried out whilst the flask is still connected
with the condenser. The solution is next filtered from a flocculent
precipitate, containing most of the impurities of the crude glycerin,
into a wide-mouthed flask of about 500 c.c. to 600 c.c. capacity, and
the filtrate allowed to cool to the ordinary temperature. Phenol-
phthalein is then added, and the free acetic acid neutralized with the
dilute caustic soda solution. Whilst running in the soda the solution
must be agitated continually, so that the alkali may not be in excess
locally longer than is unavoidable. The point of neutrality is reached
when the slightly yellowish colour of the solution just changes into

644 FOOD AND DKUGS.

reddish -yellow. If the solution is allowed to become pink, the point
of neutrality has been exceeded, and a fresh test must be made ; the
excess of soda cannot be titrated back, as partial saponification of the
acetin takes place in presence of the slightest excess of alkali. The
change of colour is very characteristic, and is easily noticed after some
little practice.

Twenty-five c.c. of the strong soda solution are now run in and the
solution boiled for a quarter of an hour ; the excess of soda is then
titrated back with the standard acid. Side by side, operating in the
same manner, 25 c.c. of the strong caustic soda are boiled and titrated
with acid. The difference between the two titrations corresponds to
the amount of alkali required for the saponification of the triacetin.
From this the quantity of glycerol in the sample can be calculated, as
shown in the following example : Suppose 1'324 grms. of the sample
have been treated as described above. Let 25 c.c. of the strong alkali
require 60"5 c.c. of normal hydrochloric acid, and let the number
of c.c. required for titrating back the excess of soda in the sample be
21-5 c.c, then 60*5 - 21-5 = 39*0 c.c. have been used. One c.c. of
normal acid corresponds to ^^^^ = 0-03067 grm. of glycerol. Hence
the sample contained 0-03067 x 39= 1-1960 grms. or 90-3 per cent of
glycerol.

The percentage found is calculated to the crude glycerin obtained
on saponifying the original quantity of 20 grms.

Added stearic acid or resin are indicated by a high acid value, and
resin by a high iodine value. But if the acid value is adjusted by the
presence of hydrocarbon wax, stearic acid may be detected by the fol-
lowing method : 1 grm. is boiled with 10 c.c. of 80 per cent alcohol.
On cooling the filtered alcohol is poured into water. In the case of
pure beeswax the liquid will remain clear or at most slightly opales-
cent. If stearic acid be present flocks of the acid will be precipitated,
and rise to the surface.

An approximate determination of the stearic acid present may be
made by boiling the wax with 90 per cent alcohol and titrating the
filtered liquid with semi-normal alkali. This will include the free
acids of resin if present. Paraffin and ceresin wax, if present in large
quantity, will be revealed by the low acid and ester values, unless these
have been adjusted by other adulterants. If so, the charring by sul-
phuric acid may be resorted to, or the actual amount of unsaponifiable
matter determined. By the usual process the alcohols of beeswax will
be returned as unsaponifiable matter. This will vary for pure waxes
between 48 and 54 per cent, so that any considerable amount of
hydrocarbons present will be detected. But the most accurate process
is that of Buisine. About 5 to 10 grms. of the wax are heated with
potash-lime to 250° C, and the mass powdered and extracted in a
Soxhlet tube with petroleum ether. The extract is filtered if necessary,
the solvent evaporated and the residue dried and weighed. Genuine
beeswax yields from 12-5 to 16-5 per cent of hydrocarbons under
these circumstances, so that the presence of more than 5 per cent of
paraffin wax will be indicated.

FIXED OILS, FATS AND WAXES. 645

Spermaceti.

Purified spermaceti is described in the Pharmacopcfiia as a concrete
fatty substance obtained mixed with oil from the head of the sperm
whale, Physeter macrocephalus. As a matter of fact it is obtained
both from the head and the blubber of the sperm whale, and also from
the bottle-nose whale, Hyjjeroodon rostratus, and possibly from other
allied species.

The official requirements > for this substance are that it should melt
at 46° to 50° C, that it should be reducible to powder by the aid of a
little alcohol, and that it should be insoluble in water, nearly insoluble
in cold alcohol but soluble in ether, chloroform, boiling alcohol and in
fixed and volatile oils. The absence of stearic acid is provided for by
the following test. When boiled with 90 per cent alcohol, and the
liquid cooled and filtered, the filtrate should not give a flocculent pre-
cipitate when added to water. The following test is given to limit
the free acidity : 0-2 grm. is dissolved in 20 c.c. of hot alcohol (90 per
cent), and 2 drops of phenol-phthalein solution added. One drop of
decinormal soda solution should produce a permanent red colour.

Spermaceti consists chiefly of acetyl palmitate (acetin) C^yllggO.
CO . C,,H3,.

A small quantity of other esters is present, and also a small amount
of free acetyl alcohol.

Spermaceti is rarely adulterated, as its characteristic crystalline
appearance is destroyed by nearly every possible adulterant. Pure
spermaceti should have the following characters : —

Speeifie gravity at 15°
Specific gr-ivity at 100°
Melting-point .
Iodine value .
Saponificition value
Fatty acids
Alcohols

0-950 to 0-960

0-808 „ 0-816 (water at 15" = 1)

44 to 48°
3„ 4-5
125,, 135

51 ,, 54 per cent

49 „ 52

The free acids and the alcohols are determined on the portion of
the sample used to determine the saponification value. The alcohols
are extracted from the saponification liquor, after driving off the alcohol
by extraction with ether, as in the determination of unsaponifiable
matter. From the aqueous liquid the fatty acids are precipitated by
hydrochloric acid, collected, washed, dried, and weighed. The sum of
the acids and alcohols will be more than 100 per cent as in the de-
composition water is taken up.

The slightly variable figures for the iodine value and the melting-
point of a spermaceti are due to a varying amount of sperm oil which is
left in the purified spermaceti. Pure spermaceti is practically neutral,
and any excess of free fatty acids over O'S per cent will be due to
careless preparation, or, more probably, to free stearic acid, which may
be detected by the official test mentioned above. A small amount of
sperm oil will raise the iodine value considerably, as the iodine value
for the oil is over 80.

646 FOOD AND DRUGS.

Petroleums.

Three varieties of petroleum are ofiBcial, the liquid, soft, and hard
paraffins.

The official tests are sufficient to ensure their purity, but in case of
any doubt, 5 grms. of the sample should be boiled with 20 c.c. of alco-
holic potash (semi-normal) for half an hour. Not more than the
slightest trace of alkali should be used in the process, otherwise fatty
substances are present.

The following are the official tests : —

Liquid Paraffin. — A colourless, odourless, and tasteless hydrocar-
bon liquid, free from fluorescence. It should not boil below 360° C.
Specific gravity 0-885 to 0*890.

Three c.c. heated with an equal volume of sulphuric acid to 100°
C, for ten minutes with frequent agitation shpuld not colour the acid
more than a pale brown (many samples do not pass this somewhat too
stringent test).

Alcohol when boiled with the sample should not colour blue litmus
paper red. A mixture of 4 c.c. of the sample and 2 c.c. of absolute
alcohol, and 2 drops of a saturated solution of lead oxide in 20 per
cent solution of caustic soda, should remain colourless when kept at
70° C. for ten minutes (absence of sulphur compounds).

Soft Paraffin. — This may be either white or yellow. It should be
free from acidity and alkalinity, and free from unpleasant odour and
taste when warmed to 120° F. Its specific gravity at its melting-point
should be 0*840 to 0-870 (water at 15-5° presumably being 1). It
melts at 35-5" to 389° C, or even somewhat higher, and gives no acrid
vapour when volatilized, and leaves no ash. It is insoluble in water,
slightly soluble in absolute alcohol, and freely soluble in ether, chloro-
form, and benzol. On treatment with boiling 20 per cent caustic soda
solution (aqueous) the separated aqueous liquid should yield no precipitate
on the addition of excess of acid (absence of fixed oils, fats and resin).

Hard Paraffin. — This should be colourless, inodorous, and tasteless.
Its specific gravity is 0-820 to 0*940. It is insoluble in water, slightly
soluble in absolute alcohol, and almost entirely soluble in ether. An
alcoholic solution should not redden litmus. It melts at 54-5° to 57*2°,
and leaves no ash when burned.

Lard.

This fat is official in the Pharmacopoeia, and has been dealt with
under Foods (see page 106).

Suet.

This fat is official in the Pharmacopoeia, and has been dealt with
under Foods (see page 111).

Soaps.

Three varieties of soap are official, curd or animal soap ; hard
(olive oil) soap ; and soft (olive oil) soap.

Curd Soap or Sapo Animalis, as it is officially termed, is directed
to be a soda soap made with purified animal fat consisting principally

FIXED OILS, FATS AND WAXES. 647

of stearin. It is to contain about 30 per cent of water. The ofiBcial
standards for this soap are as follow : —

White or pale grey in colour ; becomes horny and pulverizable when
kept in warm dry air. It is soluble in 90 per cent alcohol, sparingly
so in cold, but easily in hot, water If 5 grms. of the dried and pow-
dered soap be digested with boiling 90 per cent alcohol, and filtered
while hot, and the filter washed with a Httle more alcohol, the filtrate
should not give a red or pink colour with phenol-phthalein. And if
the filter be then washed with hot water the washings shall not require
more than 3 c.c. of decinormal sulphuric acid to discharge the red
colour imparted to phenol-phthalein. It should not impart a greasy
stain to unglazed paper. The ash yielded on incineration does not
deliquesce. It should contain about 30 per cent oi moisture.

A well-made animal soap should contain about 60 to 62 per cent
of fatty anhydrides and 7*2 to 7*5 per cent of alkali calculated as NagO.

The fatty anhydrides may be determined by dissolving 5 grms. of
soap in hot water, decomposing with HCl, and adding 3 grms. or
thereabouts of paralSin wax accurately weighed, to the hot liquid. The
fatty acids and wax solidify and can be removed in a cake from the
liquid. They are melted with distilled water, well stirred in order to
wash them, and again separated. This cake is removed, adherent
moisture removed by filter paper, and the cake then dried at 105°.
The weight less the weight of wax added gives the fatty acids. From
this an average of 7 per cent of the weight must be deducted to convert
into fatty anhydrides, two molecules of the acids losing one of HgO in
becoming anhydrides.

Sapo durus is to be made from olive oil. The tests for free alkalis,
mineral matter, and moisture are identical with those for Sapo animalis.

No ofi&cial method of deciding whether the oil used for its manu-
facture is olive oil or not is given. The soap (about 20 grms.) should
be dissolved in hot water and the fatty acids liberated by the addition
of hydrochloric acid, the free fatty acids separated, washed twice to
render them free from HCl, and dried. They should then have the
characters given under olive oil (see page 111) for the fatty acids of
olive oil. The tests there described for arachis, sesame and cotton
oil may be applied to the free fatty acids. A properly made hard soap
should contain about 30 per cent of water, 60 to 62 per cent of fatty
anhydrides, and 7'2 to 7"5 per cent of alkali calculated as NagO.

Sapomollis is the potash soap made with olive oil. It is described
as containing not more than 3 per cent of matter insoluble in warm
90 per cent alcohol ; it must not contain more free alkali than that
allowed by the tests given under Sapo animalis (see above). It yields
a deliquescent ash, which should not afford any reaction for copper.

No limit for water is given, but a pure soft olive oil soap will usu-
ally contain about 48 per cent to 50 per cent of water, and 38 per cent
to 40 per cent of tatty anhydrides. The ash should consist almost
entirely of potassium carbonate, and should on titration with standard
acid, yield results equivalent to from 6-5 per cent to 7 per cent of alkali
calculated as K^O. The fatty acids should be examined in the same
manner as those of Sapo durus.

CHAPTER XII.

THE CHEMICALS OF THE PHARMACOPCEIA.

In the present chapter, a number of the purely chemical substances
included in the Pharmacopoeia are dealt with merely in tabular form,
the figures giving certain well-marked characters, and indications of
probable impurities, etc. Others, especially when their examination
involves something more than simple inorganic testing, are dealt with
at greater length as their importance appears to justify.

The presence of small quantities of lead or arsenic in chemicals has
of late years attracted considerable attention, and it is probable that in
the next edition of the Pharmacopoeia, limits of such impurities will be
fixed. The necessity of such limits has become obvious when it is re-
membered that the harmlessness of given quantities is often a matter
of conflicting evidence in the police courts, and that convictions have
taken place when, for example, cream of tartar has been contaminated
with |rd of a grain of lead per lb., whilst in another court, acquittal
followed when there was over 1 grain per lb. The attention which has
recently been paid to this matter justifies its full treatment in this
chapter. At the same time attention may be called to the fact that, al-
though no quantitative standards exist officially, in March, 1907, Dr.
MacFadden reported to the Local Government Board (Reports of In-
spector of Foods, No. 2, 14 March, 1907) on the question of lead and
arsenic in citric and tartaric acids, and cream of tartar. The limits set out
in that report, although not legal " standards " have sufficient weight to
large 'y influence magisterial decisions in the case of either these or
similar chemicals, and are therefore of much importance, at all events
pending the issue of a new edition of the Pharmacopoeia. The con-
clusion arrived at by the reporter was that less than 0*002 per cent
of lead and 0*00014 per cent of arsenic (Ar^Og) would not be sufficient
to justify the condemnation of such substances.

The Present Official Tests for Lead and Arsenic. — The tests for ar-
senic are not described in the monographs of the Pharmacopoeia, but
are grouped in Appendix III, pp. 418-9. They are as follows : —

Arsenium.

Hydrogen sulphide affords in solutions containing hydrochloric
acid a yellow precipitate, soluble in solution of potassium hydroxide,
potassium carbonate, ammonium hydrosulphide, and potassium hydro-
gen sulphite, and iti solution of the official ammonium carbonate, but

(648)

THE CHEMICALS OF THE PHARHACOPCEIA. 649

re-precipitated on addition of hydrochloric acid. The precipitate is
insoluble in the strongest hydrochloric acid.

Nascent hydrogen, generated by the interaction of zinc and diluted
sulphuric acid, converts arsenium compounds into hydrogen arsenide.
A cold porcelain tile held in the flame of this gas acquires a dark
metallic deposit, which is readily dissolved by solution of chlorinated
soda. The gas, when passed into excess of solution of silver nitrate,
causes a black precipitate of silver, and the cautious addition of solu-
tion of ammonia to the supernatant liquid causes a yellow precipitate.

Hydrogen, generated by the interaction of zinc and solution of
potassium hydroxide or sodium hydroxide, converts arsenium compounds
into hydrogen arsenide. This gas gives a black stain to filtering paper
soaked with solution of silver nitrate and placed as a cap over the tube
in which the test is being performed. Hydrogen antimonide is not
evolved from antimony compounds under similar circumstances. The
operation should be performed in an atmosphere which is free from
hydrogen sulphide.

Stan7ious chloride dissolved in a large excess of hydrochloric acid
gives on boiling with a solution containing arsenium a brownish-black
precipitate.

Bright copper foil precipitates arsenium from solutions acidulated
by hydrochloric acid, and the arsenium may be volatilized by heat in
an open tube, when it condenses, at some distance from the copper, as
a white sublimate of characteristic octahedral crystals.

Ar seniles. — Solutions of arsenites yield a yellow precipitate with
solution of silver ammonio-nitrate.

Arsenates, — Solutions of arsenates yield a reddish chocolate pre-
cipitate with solution of silver ammonio-nitrate. Solution of mag-
nesium ammonio-sulphate affords a white crystalline precipitate.

Lead.

With lead some confusion exists. In the preface to the Pharma-
copoeia pp. xiii. to xiv. it is stated as follows: "The qualitative tests
by which the basylous and acidulous radicals of ordinary salts are re-
cognized, and by which common impurities are detected, instead of
being many times repeated in the text, as in previous editions of the
Pharmacopoeia, are given once for all in an Appendix, the text simply
stating the names of the radicals or other matters which should be
present or absent respectively. Special tests or tests rarely employed,
are still given in the text."

This raises a very important point in reference to many of the
convictions obtained during the past few years for cream of tartar
alleged to contain lead, so that it failed to correspond with 'the Phar-
macopoeial requirements. Under acidum citricum, a very delicate
test for lead is specifically described in the monogi'aph. It provides
for an acid which shall not even darken when dissolved in ammonia
and treated with HgS. Under acidum tartaricum, the acid is directed
to comply with the test for lead given under acidum citricum. But
on referring to potassii tartras acidus, or purified cream of tai-tar, it

650 FOOD AND DRUGS.

states that " It should yield no characteristic reaction with the tests for
lead". These tests are, of course, those on pages 424 to 425 of the
third appendix, and are as follows :—

" Hydrochloric acid affords, except in very weak solutions, a white
precipitate, soluble in boiling water. The aqueous solution as it cools
deposits the lead chloride in the crystalline form.

" Hydrogen sulphide, in not very strongly acid solutions, yields a
black precipitate insoluble in dilute hydrochloric acid, solution of
potassium hydroxide, and solution of ammonium hydrosulphide. It
is decomposed by boiling with diluted nitric acid, being partly con-
verted into soluble lead nitrate and partly into white insoluble lead
sulphate and sulphur. Dilute sulphuric acid causes a white precipi-
tate almost insoluble in water, and still less soluble in dilute sulphuric
acid and in alcohol, but soluble in solution of ammonium acetate.

" Solution of potassium chromate produces a yellow precipitate
readily soluble in solution of potassium hydroxide, in strong hot nitric
acid, sparingly soluble in diluted nitric acid, insoluble in acetic acid.

" Solution of potassium hydroxide gives a white precipitate soluble
in excess of the reagent but insoluble in solution of ammonia."

Not one of these tests is satisfactory for detecting small quantities
of lead, except the hydrogen sulphide test, and this is vitiated in the
case of cream of tartar, since it is directed to be applied in solution
"not very strongly acid," and as cream of tartar is about 1 in 200,
heavy traces of lead — certainly heavier than ought to be present —
would fail to produce a black precipitate.

Non-official Considerations.

The report of Dunstan and Robinson to the Pharmacopoeia com-
mittee of the General Medical Council, with possible modifications in
cases where the requirements may be considered rather too stringent,
will probably largely influence the question in the next edition of that
work. The recommendations embodied in this report are as follows : —

Tests for Arsenium.

The tests described on pp. 418 and 419 of the British Pharmaco-
poeia, 1898, to be replaced by the following : —

Those drugs which are directed not to yield any characteristic re-
action with the tests for arsenium should be proved to contain less
than three parts of arsenium in one million parts of the drug (three
parts of arsenium are equivalent to four parts of arsenious anhydride) ,
except in the cases of acidum citricum and acidum tartaricum, which
should be proved to contain less than one part of arsenium in one
million parts of the drug ; and in the cases of acidum hydrochloricum,
acidum nitricum, and acidum sulphuricum, which should be proved
to contain less than three-tenths of one part of arsenium in one million
parts of the drug ; and in the case of liquor ammonias fortis, which
should be proved to contain less than one-tenth of one part of arsenium
in one million parts of the drug.

The freedom of the drug from these quantities of arsenium is to be

THE CHEMICALS OF THE PHARMACOPCEIA. 651

proved by comparing the stain it yields when submitted to that one of
the following tests suited to its nature with the stain yielded by liquor
arsenici hydrochloricus suitably diluted and submitted to the same
test.

Each reagent employed must contain less arsenium than the limit
prescribed for it ; allowance can be made, on the one hand, for an
increase in the stain due to any minute quantities of arsenium (below
these limits) contained in the reagents, and, on the other hand, for
any diminution in the stain due to the process, by employing the same
reagents in a similar manner when preparing the stain used as a
standard for comparison.

The analyst should satisfy himself, especially in using the tests
involving more than the simplest operations, that his method of pro-
cedure is capable of finding the arsenium, by first testing the drug
with the addition of 3 c.c. of the diluted liquor arsenici hydro-
chloricus or of a solution of sodium arsenate of the same strength
as regards arsenium, by the prescribed test. It will be found that in
such cases the stain is not quite so deep as that obtained in the case
of water and the easily soluble drugs.

Test A.

A solution of 4 grms. of the drug is to be prepared as described
below, and it is to be diluted with luater to a volume of 25 c.c. This
solution is to be placed in a test-tube of about three-quarters of an
inch (about 18 mm.) in diameter and 7 to 8 inches (18 to 20 cm.) in
length. Fragments of granulated zinc are to be put into the test-tube
until they reach to about two-thirds of the height of the liquid. Im-
mediately after adding the zinc a small plug of cotton-ivool is to be
placed in the test-tube above the liquid, and then a plug of plumbized
cotton-iuool, so as to leave a short space between the two plugs, and a
closely fitting cap formed of two mercurialized test-papers is to be
fastened on ; it must not be torn at all when fastened on the test-tube.
The test is to be allowed to continue for two hours at least, and the
test-paper cap is to be exapiined by daylight for a yellow stain. The
test should be conducted in a place protected from strong light.

Ten c.c. of the liquor arsenici hydrochloricus are to be diluted to
75 c.c. (1 c.c. of the product contains 1 mg. of arsenium). Four c.c.
of this solution are to be diluted, not more than a week or two before
the test is made, to 1 litre (dilute solutions of arsenium have some-
times been found to give weaker reactions after keeping than when
fresh). Each c.c. of this solution contains 4 one-thousandth parts
of a mg. of arsenium, and is equivalent for purposes of comparison
with 4 grms. of drug to 1 .part per million, so that the yellow stain
'from 4 grms. of drug should be less than the yellow stain from 3 c.c.
of this solution mixed with water and with 5 c.c. or other suitable
quantity of hydrochloric acid, and diluted to 25 c.c. and tested in a
similar manner and at the same time.

When the drug cannot be conveniently obtained dissolved in 25 c.c.
of liquid, or when the liquid froths excessively, the experiment can be

652 FOOD AND DKUGS.

conducted in a small flask, the stain being compared with a standard
stain obtained from an equal volume of liquid in a similar flask having
a mouth of the same diameter. The flask should be shaken occasion-
ally to mix the liquid and prevent the heavy zinc-chloride solution from
settling at the bottom.

The mercurialized test-paper cap is to be prepared by moistening
two pieces of smooth white filter-paper placed together with a few drops
of test-solution of mercuric chloride and drying them. Hydrogen
arsenide produces a yellow stain on this test-paper, and thus shows
the presence of arsenium in the drug. The stain may be examined the
day after performing the test if exposure to light is avoided. Damp-
ness of the paper diminishes the intensity of the stain produced. Light
acts on the yellow stain, causing it to fade or turn grey ; the action is
only noticeable after a few days if the light is dull, but if the light is at
all bright the action is rapid. A stain placed between glass plates and
exposed to bright daylight fades considerably in an hour or two ;
without the glass plates it turns grey. The stain lasts longer in the
dry air of a desiccator than in ordinary air. Access of ammonia must
be avoided, as it turns the stain grey. Hydrogen antimonide produces
an orange or grey stain ; hydrogen phosphide and hydrogen sulphide
also produce yellow stains, and any sulphur dioxide in the solution is
changed into hydrogen sulphide by the action of the zinc.

The plug of j^lumbized cotton-ivool is to be made of cotto7i-ivool pre-
viously soaked in solution of lead acetate squeezed and dried. It is
used in order to remove any traces of hydrogen sulphide, and the lower
plug of cotton-wool is to prevent the spray from washing down the
lead acetate into the liquid beneath. Sulphur compounds should be
oxidized to sulphates when preparing the liquid for testing, and the
2)himbized cotton-ioool plug should be relied on only to remove traces
of hydrogen sulphide. A yellow stain due to sulphur, when cut out
and treated with a few c.c. of hydrochloric acid disappears in less than
ten minutes, and can thus be distinguished from a stain due to arsenium,
which, when thus treated, changes to an orange colour and lasts for
one or two hours.

As the rate of evolution of the hydrogen varies with different
samples of zinc and with the temperature, the amount of hydrochloric
acid used should be varied if necessary from 5 c.c. so that the effer-
vescence may be brisk but not violent. If the effervescence is very
soon over a further addition of hydrochloric acid can be made, lo see
if the stain becomes deepened by further evolution of hydrogen. By
using large and long fragments of granulated zinc it can be made to
extend high up in the tube without employing a great weight of it.
Granulated zinc is liable to absorb sulphur compounds on its surface ;
it can be freed from these by washing with hydrochloric acid for a few
seconds, and then with water, shortly before use.

In certain cases the oxidation of sulphur compounds in the solution
to be tested can be effected by means of the treatment with bromine,
and then with hydroxylamine hydrochloride, as described in Test B.

The presence of iron in the zinc or in the liquid must be avoided,
as it diminishes the amount of hydrogen arsenide evolved, and nitrates

THE CHEMICALS OF THE PHARMACOPCEIA. 653

and other oxidizing agents must be absent, as they also diminish the
intensity of the stain or prevent its formation.

The test can be simpHfied by omitting the use of the plumbized
cotton-ivool plug and other precautions against hydrogen sulphide, and
inferring the absence of arsenium if no yellow stain is produced, and
repeating the test with the proper precautions if a yellow stain is
found.

Test B.

Four grms. of the drug are to be placed in a flask of about 60 c.c.
capacity, together with 2 grms. of potassium metasulphite and 22 c.c.
of a mixture of hydrochloric acid and luater, in such proportions that
after reacting there shall be hydrochloric acid solution approximately
of the constant boiling strength — that is 20 parts of free hydrochloric
acid to 80 parts of water. The 2 grms. of j^otassium metasulphite,
together with 4-1 c.c. of hydrochloric acid, produce such acid ; if the
drug contains no water of crystallization, and yields no water or
volatile acid or free organic acid by its reaction with hydrochloric acid,
then 11 c.c. of hydrochloric acid and 7 c.c. of ivater will produce 18
c.c. of such acid, thus making 22 c.c. in all. If hydrochloric acid is
decomposed and water or volatile acid or free organic acid produced,
then more hydrochloric acid and less ivater must be used.

The flask is to be immediately attached to a condenser in the
position suited for distilling, and having a receiver at the lower end.
The internal diameter of the condenser-tube should not exceed 8
mm. The liquid is then to be heated gently for about one hour
in order to reduce arsenic compounds to arsenious compounds ; it is
then to be distilled until about three-fourths of it have passed over»
The distillate is to be partially neutralized with strong solution of
ammonia, so as to lea^e about 4 c.c. or other suitable quantity of
hydrochloric-acid solution of the constant-boiling strength unneutra-
lized (1 c.c. of strong solution of ammonia neutralizes 2-8 c.c. of the
constant-boiling hydrochloric-acid solution). Some distillates, especi-
ally those from antimony and bismuth compounds, effervesce with the
zinc more violently than the solutions in other cases, so in these less
than 4 c.c. of acid should be left unneutralized. In order to oxidize
the sulphur dioxide in the distillate strong solution of bromine is to be
added, a few drops at a time, until the colour due to the bromine is
permanent even after warming for a minute or two, showing that
there is a slight excess. In order to remove this excess, solution of
hydroxylamine hydrochloride is to be added, a few drops at a time,
until the liquid is colourless. The liquid is then to be diluted to 25
c.c. with ivater, and tested as described in Test A. The plug of
plumbized cotton-iuool must be used, as the treatment with bromine
does not altogether prevent the evolution of hydrogen sulphide.
"When the effervescence has ceased a further addition of acid should
be made to ensure that all the arsenium is evolved. For the purpose
of obtaining a stain for comparison, 3 c.c. of the diluted liquor
arsenici hydrochloricus should be submitted to the same process.

654 FOOD AND DRUGS.

Methods fob Drugs Specified.

The following are the methods to be employed in the cases of
particular drugs : —

For acidum aceticum, acidum hydrobromicum dil., acidum lacticum,
acidum phosphoricum cone, alumen, ammonii bromidum, ammonii
chloridum, ammonii phosphas, calcii chloridum, glycerinum, liquor
zinci chloridi, lithii citras, magnesii sulphas, phenazonum, potassii
acetas, potassii bromidum, potassii citras, potassii tartras, soda tarta-
rata, sodii bromidum, sodii sulphas, zinci acetas, zinci chloridum, zinci
sulphas, and zinci sulphocarbolas.

Four grms. of the above-named drugs are to be dissolved in nearly
20 c.c. of water, and the solution is to be mixed with 5 c.c. or other
suitable quantity of hydrochloric acid, and diluted to 25 c.c. with
water, and tested as described in Test A.

For Potassii Sulphas a^id Sodii Phosphas. — Four grms. of these
drugs are to be dissolved in the smallest convenient quantity of ivater,
and the solution is to be mixed with 5 c.c. or other suitable quantity
of hydrochloric acid, and tested in a small flask as described in Test A.

For Potassii lodidum and Sodii lodidum. — Four grms. of these
drugs are to be dissolved in 5 c.c. of icater and are to be tested by
Test A, modified in the following manner : 5 c.c. or other suitable
quantity of hydrochloric acid are to be mixed with 14 c.c. of ivater in
the test-tube. The zinc is then to be added and the effervescence is
to be allowed to proceed for two minutes, then the above solution of
the iodide is to be poured in and the plugs and cap are at once to be
put into position. This procedure avoids the liberation of iodine in
the liquid ; a little iodine appears on the plug but does not materially
diminish the stain.

For Syrupus Glucosi. — Four grms. of this drug are to be dissolved
in 10 c.c. of tvater. In order to oxidize any sulphur dioxide that may
be present, 3 c.c. of strong solution of bromine are to be added and
then 5 c.c. of hydrochloric acid, and the mixture is to be warmed for
a few minutes, care being taken to stop whilst a distinct amount of
free bromine is still present. When cold the free bromine is to be
removed by adding a little solution of hydroxylamine hydrochloride ;
3 c.c. or other suitable quantity of hydrochloric acid are to be added,
and the liquid is to be diluted to 25 c.c. with icater and tested as de-
scribed in Test A. In presence of glucose the stain obtained from 3 c.c.
of the diluted liquor arsenici hydrochloricus is only about three-
fourths of its proper intensity, and for this diminution allowance must
be made by means of a comparative experiment made with the syrupus
glucosi and the arsenium solution. The effervescence should be pro-
longed by a second addition of hydrochloric acid.

For Acidum Boricum and Borax. — Four grms. of these drugs are
to be mixed with 8 grms. of citric acid and dissolved in 55 c.c. of ivater,
and the solution is to be mixed with 5 c.c, or other suitable quantity
of hydrochloric acid, and tested in a small flask as described in Test A.

For Acidum Citricum and Acidum Tartaricum. — Twelve grms.

THE CHEMICALS OF THE PHARMACOPOEIA. 655

of these drugs are to be dissolved in 40 c.c. of ivat$r, and the solution
is to be mixed with 15 c.c. or other suitable quantity of hydrochloric
acid, and tested in a small flask as described in Test A. The stain
should be less than that given by 3 c.c. of the diluted liquor arsenici
hydrochloricus similarly treated, thus proving that the drugs contain
less than one part of arsenium in one million parts of the drug.

For Acidum Hydrochloricum. — Forty grms., or 34-5 c.c, of this
drug are to be placed in a porcelain basin and mixed with 2 c.c. of
strong solution of bromine. The mixture is to be gently evaporated
on a sand-bath, adding small quantities of strong solution of bromine
from time to time, so that the liquid is always kept orange-colour and
smelling of bromine (about 4 c.c. or 5 c.c. of the strong solution of
bromine will be required in all). The presence of free bromine pre-
vents loss of arsenium during the evaporation. When the volume is
reduced to about 15 c.c. the acid is to be partially neutralized with
strong solution of ammonia (1 c.j. of this neutralizes 2*8 c.c. of hydro-
chloric acid solution of the constant-boiling strength), so as to leave
unneutralized 5 c.c. or other suitable quantity of hydrochloric acid
solution of the constant-boiling strength. The excess of bromine is to
be removed by adding a little solution of hydroxylamine hydrochloride ;
the liquid is then to be diluted to 25 c.c. with water, and tested as de-
scribed in Test A. The stain should be less than that given by 3 c.c. of
the diluted liquor arsenici hydrochloricus, thus proving that the drug
contains less than three-tenths of one part of arsenium in one million
parts of the drug.

For Acidum Nitricum. — Forty grms., or 28-2 c.c, of this drug are
to be mixed with 2 c.c. of sulphuric acid, and with 0*1 grm. of sodium
bicarbonate, and the liquid is to be evaporated in a porcelain basin on
a sand-bath until all the nitric acid is expelled and fumes of strong
sulphuric acid are given off. The residual liquid is to be allowed to
cool and is then to be mixed with about 15 c.c. of ivater, and then
with 3 c.c or other suitable quantity of hydrochloric acid. The mix-
ture is to be diluted to 25 c.c with ivater, and tested as described in
Test A. The stain should be less than that given by 3 c.c of the
.diluted liquor arsenici hydrochloricus, thus proving that the drug
contains less than three-tenths of one part of arsenium in one million
parts of the drug.

For Acidum Sulphur icum. — Forty grms., or 21'7 c.c, of this drug
are to be mixed with 5 c.c of 7iitric acid, and with 0*1 grm. of sodiimi
bicarbonate, and the liquid is to be evaporated in a porcelain basin on
a sand-bath until only about 2 c.c. remain. The residual liquid is to
be allowed to cool and is then to be mixed with about 15 c.c. of ivater,
and then with 3 c.c. or other suitable quantity of hydrochloric acid.
The mixture is to be diluted to 25 c.c with water, and tested as de-
scribed in Test A. The stain should be less than that given by 3 c.c
of the diluted liquor arsenici hydrochloricus, thus proving that the
drug contains less than three-tenths of one part of arsenium in one
million parts of the drug.

For Liquor Ammonice Fortis. — One hundred and twenty grms., or
135 c.c, of this drug are to be mixed with 0*1 grm. of sodium bicar-

56 FOOD AND DRUGS.

bonate, and the solution is to be evaporated to dryness, or nearly to
dryness, on a water-bath. The residue, when cold, is to be dissolved
with a mixture of 5 c.c. or other suitable quantity of hydrochloric acid,
and about 20 c.c. of ivater, avoiding heating except for a minute or
two. The solution is to be diluted to 25 c.c. with water, and tested
as described in Test A. The stain should be less than that given by

3 c.c. of the diluted liquor arsenici hydrochloricus, thus proving that
the drug contains less than one-tenth of one part of arsenium in one
million parts of the drug.

For aramonii carbonas, calcii carbonas praecipitatus, calcii
hydras, calcii phosphas, calx, liquor potassae, lithii carbonas, mag-
nesia levis, magnesia ponderosa, magnesii carbonas levis, magnesii
carbonas ponderosus, potassa caustica, potassii bicarbonas, potassii
carbonas, potassii tartras acidus, sodii bicarbonas, sodii carbonas,
zinci carbonas, zinci oxidum, and zinci valerianas.

Four grms. of these drugs are to be dissolved in hydrochloric acid
and water, using enough hydrochloric acid to acidify and dissolve the

4 grms. of drug taken, and to produce a suitable effervescence with the
zinc.

Care must be taken not to warm hydrochloric acid and drug to-
gether except for a minnte or two, and with only a small area of sur-
face, so as to avoid loss of arsenium. If necessary, loss of arsenium
can be avoided by mixing a little strong solution of bromine with the
hydrochloric acid and water, so as to oxidize the arsenium to arsenic
acid, which is not easily volatilized when heated with hydrochloric acid ;
when solution is effected the excess of bromine is to be removed by
the addition of a little solution of hydroxylamine hydrochloride.

The solution is to be diluted if necessary and tested in a test-tube
or flask as described in Test A.

If a drug contains any iron, it must be tested as described in Test
B.

For Cerii Oxalas. — Four grms. of this drug are to be added to a
small flask containing a hot mixture of 15 c.c. of hydrochloric acid, 10
c.c. of water, and 1 c.c. of strong sohttioii of bromine. The mixture is
to be heated for about a minute, when the cerium oxalate will dis-
solve, but a precipitate will very soon separate. As soon as solution
has occurred the flask is to be removed from the flame and the acid is
to be partially neutralized by the addition of about 7*25 c.c. of strong
solution of ammonia, and the free bromine is to be removed by the
addition of a little strong solution of hydroxylamine hydrochloride.
The mixture is then to be tested in the flask as described in Test A,
shaking it occasionally to promote the circulation of the liquid, which
is checked by the presence of the precipitate.

For lodum. — Four grms. of this drug are to be mixed with O'l
grm. sodium bicarbonate, and then with 3 c.c. of water and 4 c.c. of
sulphuric acid in a porcelain basin, and the mixture is to be heated
with stirring until all the iodine is driven off. The residue of sul-
phuric acid is to be diluted with about 15 c.c. of water, and then
mixed with 2 c.c. or other suitable quantity of hydrochloric acid, and
then diluted to 25 c.c. ^vith water, and tested as 'described in Test A.

THE CHEMICALS OF THE PHAKMACOPCEIA. 657

For Liquor Hydrogenii Peroxidi. — Four grms. of this drug are to
be EQixed with 4 c.c. of luater and with 2 c.c. of suljjhiiric acid. Potas-
sium permanganate is then to be added in small quantities at a time
until the hydrogen peroxide is all decomposed and a slight permanent
coloration is produced. The solution is to be mixed with 7 c.c. of
water, and the coloration is to be destroyed by the addition of a little
solution of hydroxy lamine hydrochloride. Three c.c. or other suitable
quantity of hydrochloric acid are to be added, and the solution is to be
diluted to 25 c.c. with ivater, and tested as described in Test A.

For Potassii Chloras. — Six c.c. of sulphuric acid are to be mixed
with 3 c.c. of water, and the mixture is to be heated. Four grms. of
this drug are to be added cautiously in small portions at a time to the
above liquid whilst hot. When effervescence has ceased the liquid
is to be evaporated in a porcelain basin until only about 2 c.c. of sul-
phuric acid are left. The residue is then to be dissolved in about 15
c.c. of water and mixed with 2 c.c. or other suitable quantity of
hydrochloric acid and diluted to 25 c.c. with water, and then tested
as described in Test A.

For Potassii Nitras. — Four grms. of this drug are to be added to

4 c.c. of sidjjhuric acid in a porcelain basin, and then heated until all
the nitric acid is driven off and fumes of sulphuric acid escape. The
residue is then to be dissolved in about 15 c.c. of ivater, 3 c.c. or other
suitable quantity of hydrochloric acid are to be added, and the solution
is to be diluted to 25 c.c. with water, and tested as described in Test
A.

For Potassii Permanganas. — Four grms. of this drug are to be
added, in small quantities at a time, to 30 c.c. of hydrochloric acid.
When it has all dissolved 2 c.c. of solution of hydroxylamine hydro-
chloride are to be added in order to decolorize the liquid, and then
about 4 c.c. of strong solution of ammonia in order partially to neutra-
lize the free hydrochloric acid. One c.c. of solution of hydroxylamine
hydrochloride is then to be added in order to remove the last traces of
free chlorine, and the liquid is to be tested in a flask as described in
Test A.

For Galcii Hypophosphis and Sodii Hypophosphis. — A mixture of
12 c.c. of nitric acid and 12 c.c. of water is to be warmed, and 4 grms.
of these drugs are to be added in small quantities at a time, so as to
prevent the action being too violent. When all is added the liquid is
to be evaporated to dryness on a sand-bath and the residue heated, but
not strongly, until the nitric acid has been driven off. The residue,
when cold, is to be dissolved in 5 c.c. or other suitable quantity of hy-
drochloric acid mixed with ivater, avoiding loss by warming, or using
the bromine and hydroxylamine-hydrochloride treatment. The solu-
tion is then to be diluted to 25 c.c. with water, and tested as described
in Test A.

For Phosjjhorus. — Six cgms. of this drug are to be dissolved by heat-
ing them cautiously in a flask of about 100 c.c. capacity, having a small
funnel placed in its mouth, with a mixture of 5 c.c. of nitric acid and

5 c.c. of ivater. The solution is then to be transferred to a porcelain
basin, and in order to oxidize any phosphorus acid 5 c.c. of nitric acid

VOL. I. 42

658 FOOD AND DRUGS.

are to be added, and the mixture is to be heated until it has concen-
trated to about half its volume. In order to remove nitric acid 0"1
grm. of sodium bicarbonate is then to be added, and 3 c.c. of sulj)huric
acid, and after mixing the liquid is to be evaporated down to about 3
c.c, and then, in order to decompose any nitrosulphonic acid, the re-
sidue is to be allowed to cool and mixed with 10 c.c. of ivater. It is
then to be evaporated until fumes of strong sulphuric acid escape ;
when cold the residue is to be diluted with about 10 c.c. of ivater and
mixed with 5 c.c. or other suitable quantity of hydrochloric acid, and
diluted to a volume of 25 c.c. with water, and tested as described in
Test A. The stain should be less than that given by 3 c.c. of the
diluted liquor arsenici hydrochloricus thus proving that the drug
contains less than 0'02 per cent of arsenium.

For Sulphur PrcBcipitatum and Sulphur Sublimatum. — Four grms.
of these drugs are to be dissolved by heating them in a large flask,
having a small funnel placed in its mouth, with 25 c.c. oi fuming nitric
acid, and adding more fuming nitric acid when necessary (about 60 or
70 c.c. will be required). When the sulphur has all dissolved 0*1 grm.
of sodium bicarbonate is to be added, and the liquid is to be evaporated
in a porcelain basin on a sand-bath until all nitric acid is expelled and
fumes of sulphuric acid are given off; the volume is to be reduced to
about 2 c.c, it is then to be diluted with about 15 c.c of water and
mixed with 2 c.c or other suitable quantity of hydrochloric acid, and
diluted to a volume of 25 c.c. with water, and tested as described in
Test A. The amount of arsenium in the fuming nitric acid used can
be determined by the method described for testing acidum nitricum
and allowed for. It should be less than one-tenth of one part of
arsenium in one million parts of the acid, and the acid should be free
from the impurities mentioned in the case of acidum 'nitricum, especi-
ally iron.

For Acidum Salicylicum, Adeps Lance, Glusidum, Phenacetinum,
Sapo Animalis, Sapo Durus, and Sulphonal. — Four grms. of these
drugs are to be mixed with 2 grms. of magnesia and 2 grms. of exsiccated
sodium carbonate, and the mixture is to be made into a thin paste by
warming it with a small quantity of tcater and stirring. The mixture
is then to be dried and ignited in a porcelain basin or in a porcelain
crucible until the volatile organic matter is driven off and the residue
is greyish-white. The temperature must not approach a white heat.
Fifteen c.c. of water are to be mixed with 21 c.c of hydrochloric acid and
3 c.c of strong solution of bromine. The bromine is used in order to
oxidize any sulphur compounds and to prevent loss of arsenium by the
heating of the liquid, which should be cooled by the use of an outer
vessel of water. The ignited residue is to be added to this mixture
in small portions at a time. When solution is effected (some carbon-
aceous particles will remain undissolved) the excess of bromine is to be
removed by adding a little solution of hydroxylamine hydrochloride,
and the liquid is to be tested in a flask as described in Test A. The
plug of plumbized cotton-wool must be used, as the treatment with
bromine does not altogether prevent the evolution of hydrogen sulphide.
For the purpose of obtaining a stain for comparison, 3 c.c. of the diluted

THE CHEMICALS OF THE PHAEMACOPCEIA. 659

liquor arsenici hydrochloricus should be submitted to the same
process.

For cupri sulphas, ferri phosphas, ferri sulphas, gelatinum, plumbi
acetas, quinincB hydrochloridum, quinmcB hydrochloridum acidum, and
quinincR sulphas.

Four grms. of the above drugs are to be tested as described in
Test B.

For Antimonii Oxidum and Antimonium Tartaratum. — Four
grms. of the above drugs ara to be tested as described in Test B, but as
the distillate will still contain a little antimony chloride, the condenser
is to be washed free from any traces of antimony chloride, and the
distillate is to be re-distilled until about three-fourths of it have collected
in the receiver, and this distillate is to be treated as directed in Test
B. Twenty-three c.c. of hydrochloric acid and no luater are to be used
with the antimonii oxidum, and a mixture of 20 c.c. of hydrochloric
acid and 3 c.c. of icater with the antimonium tartaratum.

For Antimonium Nigrum Purificatum and Antimoni^im Sulphur-
atum. — Four cgms. of these drugs are to be heated in a flask of
about 100 c.c. capacity, having a small funnel placed in its mouth,
with 10 c.c oi filming nitric acid, until all sulphur or black sulphide^
has been oxidized. A white precipitate will be formed in the liquid,
but the absence of free sulphur or of black sulphide can be easily seen..
The mixture is then to be transferred to a porcelain basin, and is to
be mixed with 0*1 grm. of sodium bicarbonate and with 3 c.c. of
sulphuric acid. All the nitric acid is to be removed by evaporating
the mixture down to about 3 c.c, and then, in order to decompose
nitrosulphonic acid, mixing the residue when cold with 10 c.c. of
water, and evaporating again until fumes of strong sulphuric acid
escape. When the residue is cold it is to be transferred to a flask of
about 60 c.c. capacity, by means of a mixture of 15 c.c. of hydrochloric
acid and 7 c.c. of ivater. Two grms. oi potassium metasulphite are to
be added, and the flask is to be immediately attached to a condenser
and treated as described in Test B, but as the distillate will still con-
tain a little antimony chloride, the condenser is to be washed free
from any traces of antimony chloride, and the distillate is to be re-
distilled until about three-fourths of it have collected in the receiver,
and this distillate is to be treated as directed in Test B. The stain
should be less than that given by 3 c.c. of the diluted liquor arsenici
hydrochlaricus, thus proving that the drug contains less than 0*03
per cent of arsenium.

For Bismuthi Oxidum. — Four grms. of this drug are to be tested as-
described in Test B, using 20 c.c. of hydrochloric acid and 2 c.c. of
water ; but if the drug contains any nitrate it must be tested in the
same manner as bismuthi carbonas.

For Bismuthi Carbonas and Bismuthi Subnitras. — Four grms. of
these drugs are to be mixed with 5 c.c. of nitric acid in order to
oxidize any arsenious compounds to arsenic acid, and then with 8 c.c-
of sulphuric acid, and the mixture is to be heated in a porcelain basin
on a sand-bath until all the nitric acid is expelled and a considerable
proportion of the sulphuric acid has been driven off in fumes. When

660 FOOD AND DEUGS.

evaporating off sulphuric acid, in order to avoid loss of arsenium, the
latter should be present as arsenic acid and not as arsenious com-
pounds. The residue is to be allowed to cool, and then 6 c.c. of tcater
are to be added. The mixture is again to be allowed to cool, and is
then to be transferred to a flask of about 60 c.c. capacity, together wdth
17 c.c. of hydrochloric acid; 4 grms. of ferrous suljjhate and 2 grms.
of 2^otassium metasuli^hite are to be added, and the rest of the test is
to be conducted as described in Test B.

For Bismuthi Salicylas and Liquor Bismuthi et Ammonii Cit-
ratis. — Four grms. of these drugs are to be mixed with 2 grms. of
magnesia and 2 grms. of exsiccated sodium carbonate, and the mixture
is to be made into a thin paste by warming it with a small quantity of
ivater and stirring. The mixture is then to be dried and ignited in a
porcelain basin or in a porcelain crucible until the volatile organic
matter is driven off and the residue is greyish. Fifteen c.c. of ivater
are to be mixed with 21 c.c. of hydrochloric acid, and 3 c.c. of strong
solution of bromine, in a flask of about 60 c.c. capacity. The bromine
is used in order to convert the arsenium into arsenic compounds, and
prevent its loss by the heating of the liquid, which should be cooled
by the use of an outer vessel of water. The ignited residue is to be
.^dded to this mixture in small portions at a time. When solution
is effected (some carbonaceous particles will remain undissolved), the
flask is to be attached to a condenser as described in Test B, and dis-
tilled until about half the volume of the liquid has passed over. This
•distillate will contain the free bromine and no arsenium ; but for
greater security it may be tested for arsenium. A fresh receiver is to
be placed in position, and 20 c.c. of hydrochloric acid are to be
added to the residue in the distilling flask, and then 2 grms. of potas-
sium metasulphite, and the mixture is to be heated gently for about
•one hour in order to reduce arsenic compounds to arsenious com-
pounds. It is then to be distilled until about three-fourths of it have
passed over, and the distillate is to be treated in the same manner as
the distillate described in Test B, but as the volume will exceed 25 c.c,
it must be tested in a small flask

For Ferrum. — Four grms. of this drug are to be dissolved in a
mixture of 3 c.c. of nitric acid and 3 c.c. of ivater, and the solution is
to be evaporated to dryness in a small porcelain basin, and the residue
is to be ignited until the ferric nitrate is converted into ferric oxide.
The residue is then to be transferred to a flask of about 60 c.c. capa-
city, together with 10 c.c. of hydrochloric acid and 6*5 c.c. of ivater,
•scraping out as much as possible, and treating the remainder with the
mixed acid and water, but not warming unless very slightly and only
for a minute or two. The flask is to be attached to a condenser, as
■described in Test B. The mixture is to be warmed until the ferric
oxide has all dissolved, and then 4 grms. of ferrous sul2)hate and 2
grms. of ])otassium metasuljjhite and 7 c.c. of hydrochloric acid are to
be added, and the rest of the operation is to be conducted as described
in Test B. At the end of the distillation the residue in the distilling
flask should be tested and some ferrous iron should be found to be
present. The stain should be less than that given by 3 c.c. of the

THE CHEMICALS OF THE PHAEMACOPCEIA. 661

diluted liquor arsenici hydrochloricus, thus proving that the drug
contains less than 0'03 per cent of arsenium.

For Ferrum Redactum. — Two decigrammes of this drug are to be
heated in a flask having a small funnel placed in its mouth with a
mixture of 10 c.c. of nitric acid and 10 c.c. of water. When the ac-
tion has ceased, if an insoluble residue is left, it is to be dissolved
by adding '6 c.c. of hydrochloric acid and continuing the warming.
The solution is then to be transferred to a small porcelain basin, and
5 c.c. of nitric acid are to be mixed with it, and the liquid is to be
evaporated to dryness and ignited until the ferric nitrate is converted
into ferric oxide. The ignited residue is then to be treated in the same
way as the ignited residue obtained in testing ferrum. The stain
should be less than that given by 3 c.c. of the diluted liquor arsenici
hydrochloricus, thus proving that the drug contains less than 60 parts
of arsenium in one million parts of the drug.

For Liquor Ferri Acetatis. — Four grms. of this drug are to be
put into a flask of about 60 c.c. capacity, together with 4 grms. of
ferrous sulphate. A mixture of 15 c.c. of hydrochloric acid and 2 c.c.
of luater is to be added, and then 2 grms. of potassium metas^iljjhite.
The flask is then to be attached to a condenser, and the mixture is to
be treated as described in Test B. The ferrous sulphate will effect the
decomposition of any traces of nitric acid. At the end of the distilla-
tion the residue in the distilling flask should be tested, and some ferrous
iron should be found to be present.

For Liquor Ferri Perchloridi Fortis. — Twenty-five cgms. of
this drug are to be put into a flask of about 60 c.c. capacity, together
with 4 grms. of. ferrous sulphate. A mixture of 15 c.c. of hydrochloric
acid and 6 c.c. of loater is to be added, and then 2 grms. of potassium
metasulp)hite. The flask is then to be attached to a condenser, and the
mixture is to be treated as described in Test B. The ferrous sulphate
will etfect the decomposition of any traces of nitric acid. At the end
of the distillation, the residue in the distilling flask should be tested
and some ferrous iron should be found to be present. The stain
should be less than that given by 3 c.c. of the diluted liquor arsenici
hydrochloricus, thus proving that the drug contains less than 48 parts
of arsenium in one million parts of the drug.

For Liqtior Ferri Pernitratis. — One grm. of this drug is to be
treated in the same manner as that described above for liquor ferri
perchloridi fortis. The stain should be less than that given by 3 c.c.
of the diluted liquor arsenici hydrochloricus, thus proving that the drug
contains less than 12 parts of arsenium in one million parts of the drug.

The reporters suggest certain alterations in the B.P. monographs
so as to provide certain definite limits of arsenium content. The
subjoined paragraphs embody their specific recommendations, no
attention being paid here to w^hat they say about the text of the
monographs.

Antimonium Nigrum Purificatum. — Should be proved to contain
less than 0*03 per cent of arsenium by the tests for arsenium.

Antmionium Sulphuraticm. — Should be proved to contain less
than 0-03 per cent of arsenium by the tests for arsenium.

662 FOOD AND DRUGS,

Ferrum. — Should be proved to contain less than 0'03 per cent of
arsenium by the tests for arsenium.

Ferrum Bedactum. — To be prepared by reducing ferric hydroxide
free from arsenium, heated to dull redness, by a stream of dry hydro-
gen which has been purified from arsenium compounds. It should
be proved to contain less than 60 parts of arsenium in one million
parts of the drug by the tests for arsenium.

Glycerinum. — Should yield no characteristic reaction with the
tests for arsenium.

Liquor Bismuthi et Ammonii Citratis. — Should yield no character-
istic reaction with the tests for arsenium.

Liquor Ferri Perchloridi Fortis. — Should be proved to contain less
than 48 parts of arsenium in one million parts of the drug by the
tests for arsenium.

Liquor Ferri Pernitratis. — Should be proved to contain less than
12 parts of arsenium in one million parts of the drug by the tests for
arsenium.

Phosphorus. — Should be proved to contain less than 0-02 per cent
of arsenium by the tests for arsenium.

Sulphur PrcBcipitatum. — Should yield no characteristic reaction
with the tests for arsenium.

Sulphur Sublimatum. — Should yield no characteristic reaction
with the tests for arsenium.

The Following Drugs should also yield no characteristic reaction
with the tests for arsenium : acidum boricum, acidum citricum,
acidum salicylicum, adeps lanae, alumen, ammonii bromidum, ammonii
carbonas, calcii carbonas praecip., calcii chloridum, calcii hydras, calx,
ferri sulphas, gelatinum, glusidum, iodum, liquor hydrogenii peroxidi,
magnesia levis, magnesia ponderosa, magnesii carbonas levis, magnesii
carbonas pond., magnesii sulphas, phenacetinum, phenazonum, potassii
carbonas, potassii chloras, potassii citras, potassii tartras, potassii
tartras acidus, quininae hydrochloridum, quinina3 hydrochlor. acid.,
quininaB sulphas, sapo animalis, sapo durus, soda tartarata, sodii bi-
carbonas, sodii carbonas, sodii hypophosphis, sodii phosphas, sodii
sulphas, sulphonal, syrupus glucosi.

There are so many methods, many merely slight modifications of
each other, for the detection and estimation of arsenic in traces, that
only a few which are fairly accurate and satisfactory will be described.
It is to be noted that Dunstan and Eobinson in their prepared stand-
ards, have given them in terms of the metal arsenium, which is con-
trary to the usual practice. In any references in this work, the
amount of arsenic, expressed as As^O^. is intended unless otherwise
indicated.

F. C. J. Bird (" Pharm. Journal," i, 19, 424) has criticized the
above detailed report. He points out that the method of fastening the
cap of mercurialized paper is left to the judgment of the operator.
He considers that it should be tied tightly over the mouth of the test-
tube or flask, so that the evolved gas is obliged to force its way
through the pores of the mercurialized paper.

With regard to the intensity of the stain, he points out that it is

THE CHEMICALS OF THE PHARMACOPCEIA. 663

diminished by moisture. This should be emphasized, for a given
stain, having naturally absorbed moisture by contact with the evolved
gas charged with aqueous vapour, will often nearly double its depth
of colour on exposure to a temperature 80" - 90° for a minute or
two. It would, therefore, appear to be a desirable addition to tha
directions that the stain from the material under examination and that
from the standard arsenical solution, for comparison, should be placed
in a water oven for a few minutes, in order to ensure equal conditions
and guard against one stain being damper, and, therefore, fainter than
the other. Drying in a water oven will also sometimes render evi-
dent a stain otherwise indistinguishable.

He further points out — a fact confirmed by the author's experience
— that it is necessary to pass the evolved gas through a solution of
lead acetate, as cotton-wool soaked in the solution may fail to arrest
all the possible H^S present in a rush of gas. He also recommends
placing the stained papers in a little HCl on watch glasses and heating
to the boiling-point of the acid, and drying the stains. Any stain due
to H^S would be destroyed, and the resulting brick-red colour is more
characteristic and more easily compared with the standard stains.
Bird has modified Gutzeit's test with considerable success as outlined
below, and in the author s experience, this modified process yields
exceedingly accurate results, even with substances so refractory as
reduced iron or oxide of iron.

This improved test, in which all the disadvantages and unreliability
of the Gutzeit reaction are overcome, consists in evolving hydrogen in
a flask from definite quantities of pure zinc and hydrochloric acid at
a boiling temperature in presence of a definite amount of the sub-
stance to be examined. Any arsenic present is converted into arsen-
iuretted hydrogen, and this, together with the excess of hydrogen and
water vapour, is passed through a vertical condenser when the aqueous
vapour is condensed and runs back into the flask. The gases are then
made to bubble through lead acetate solution (to remove SHg) and
then force themselves through the pores of a small disk of filter paper
impregnated with mercuric chloride. Here, if in small proportion,
the whole of the arsenic is arrested forming a characteristic lemon-
yellow stain, but to guard against any escaping the first disk, a second
disk is employed to absorb the last traces. The two disks are then
removed and heated in a watch glass with pure HCl, when, if the
yellow stain be due to arsenic, it at once assumes a characteristic brick-
red hue, differing in this respect from a sulphuretted stain which
disappears, a phosphorettsd stain which remains yellow, and an anti-
moniuretted stain which turns grey and nearly fades. The brick-red
paper disk is then treated with hydrochloric acid containing a little
bromine, when the stain dissolves and the solution (now containing
the arsenic) may be transferred to a small test-tube and a special
stannous chloride reagent added, when the brown arsenical coloration
is developed either immediately or on standing. In this manner yi^
mg. of As^O,; can be detected and identified quite easily.

Description of the Ajjparatus. — A, flask capacity of about 100 c.c,
D, bent thistle tube passing through I. R. cork to bottom of A. C,

664

FOOD AND DRUGS.

separator with stopcock, capacity about 50 c.c. B, vertical condenser
connected with F by bent glass tube E. r.r. rubber joints. F, gas
washing bulb containing a 1 in 10 sclution of lead acetate. G, special
glass nozzle carrying a disk of mercuric chloride paper. rr\ nib-

ber joints. H, second glass
nozzle similar to G. J, curved
glass open nozzle. I, Bunsen
burner.

Directions jor tise. — In the
flask A place 30 c.c. water, four
grms. of zinc(distilled, As.-free)
and a w^eighed or measured
quantity of the substance to be
tested. Introduce into C 15
c.c. of pure hydrochloric acid
(As.-free) and heat the contents
of the flask A to boiling-point.
Then open the stopcock to such
an extent that the hydrochloric
acid runs out in a steady suc-
cession of drops so that about
seven minutes are required for
the whole to pass into A. The
contents of A should be kept
in a gentle state of ebullition
throughout the experiment.
The evolved gas first passes

f1ER.CUK.lC
<-HLQR.iDB
P/\PeK DISC.

Fig. 58.— Bird's arsenic apparatus. Fig. 59.— Enlarged dia-

gram of exit nozzles in
Bird's arsenic apparatus.

through the lead acetate solution F (where SH, is removed) and then
through the pores of the mercuric chloride paper disks fixed on G
and H, which should absorb the whole of the arsenic, finally escaping
at J. At the expiration of about fifteen minutes the paper disk s

THE CHEMICALS OF THE PHAEMACOPCEIA. 66J

should be detached by touching the edges of the disks with a
moistened glass rod, placed in a watch glass, dried for a minute or
so in a water oven, and about 3 c.c. pure HCl {jree from CI) added,,
and the whole heated on a thin piece of asbestos millboard until the
acid just commences to boil. Any yellow stain due to arsenic now
becomes of a deep brick-red, a reaction entirely characteristic of an
arsenical stain, all interfering stains (S, P, Sb) behav.ng quite differently
under this treatment. (See F. C. J. Bird, "Analyst," xxvi. 181.)
The depth of stain should then be compared with that produced by
y^ of a mg. of As^O,. which amount forms a convenient standard for
comparison. With regard to the standard, with very pure zinc and
HCl y^o mg. is quite definite, but ^i^ mg. will often be found more-
convenient.

For ordinary testing the process may terminate here and the further
confirmation be omitted.

For further confirmation of the arsenical nature of the stain, the
hydrochloric acid is poured away, leaving the paper disks on the watch
glass and a second 3 c.c. of pure hydrochloric acid added, warmed
geatly, and again poured off. (The second washing with acid is for the
purpose of removing from the paper excess of mercury salt which
would otherwise affect the stannous chloride reaction.)

The disks are now warmed with ^ c.c. hydrochloric acid and the
arsenical compound dissolved by the addition of 1 or more drops of
bromine hydrochloric acid, avoiding any large excess. The pale yellow
liquid is poured off into a small test-tube 3 inches by \ inch, any
residual acid displaced with one or two more drops of hydrochloric
acid, an equal volume of 30 per cent stannous chloride solution in
HCl added, and the contents of the test tube warmed, when if y^ of
a mg. of arsenic be present, the characteristic pink brown coloration
of Bettendorff s reaction makes its appearance almost immediately.

Beagents. — Bromine hydrochloric acid. This is strong hydro-
chloric acid containing sufficient bromine to impart a deep yellow
colour.

The stannous chloride reagent is prepared by dissolving 30
grms. of crystaUized stannous chloride in 150 c.c. of pure hydro-
chloric acid and boiling down with a few fragments of metallic tin to
100 c.c. It should remain perfectly colourless when heated with an
equal volume of hydrochloric acid and allowed to stand for some time.

Mercuric chloride disks 5 mm. diameter. Filter paper is satur-
ated with a solution of mercuric chloride (1 in 20) and dried ; the
disks are cut out with a cork borer.

By taking an amount of the substance under examination corre-
sponding to any standard limit of arsenical impurity for that substance,
it is possible to see with the minimum expenditure of time and
trouble whether the limit has been exceeded or not In this way
with a limit of y^^ of a grain per gallon (e.g. beer) 70 c.c. would be
taken for the test = 1 part in seven millions. With 7 grms. of
material y^ of a mg. of arsenic equals yi^ of a grain per lb. With
3 "5 grms. ^l of a grain per lb., with 1 grm. 1 part in one hundred
thousand, etc.

€66 FOOD AND DEUGS.

Cowley and Catford recommend (" Pharm. Journ." 4, 19, 897) the
following process : —

A few inches of fine copper wire, coiled into a spiral are immersed
in 10 c.c. of the liquid to be tested, to which one-fifth of its volume of
pure HCl has been added. The liquid is contained in a test-tube, which
is supported upright in a brine bath. The coil of wire is arranged so
that it shall reach from the bottom of the arsenical liquid to above its
surface. The test-tube must be immersed in the brine bath so that
the liquid it contains shall be below the level of the liquid in the bath ;
the bath is kept simmering without, however, reaching the boiling-
point, for about an hour. The projecting end of the copper is now
pressed below the surface of the liquid, and if it (that is, the bright
end that has not been previously in the liquid) remains bright after
remaining for another fifteen minutes, the arsenic will be all re-
moved from the liquid, and the wire may be removed to a small dish,
rinsed without touching it with the fingers, and the deposit then dis-
solved off by 1 c.c. of bromine water containing a little hydrobromic
^cid. The clean wire is lifted out, rinsed with water, and if thought
necessary may be returned to the acid liquid to make sure that all the
arsenic has been dissolved from it. The bromine solution now con-
tains the arsenic as arsenious acid. To it 1 c.c. of solution of potash
is added, and the liquid is boiled until the light green copper com-
pounds are broken up. The arsenic in the neutralized filtrate is reduced

■completely to arsenite and titrated with , or other suitaby weak

solution of iodine. A solution of iodine of convenient strength is made

by diluting 10 c.c. of — solution to about 150 c.c, and comparing it

with a standard arsenical solution.

For a burette a pipette graduated in hundreths of a cubic centi-
metre is used. To control the flow, a piece of rubber tubing is slipped
on the upper end and compressed by a screw clamp. One-hundredth
part of a c.c. of the iodine solution gives a blue colour, with starch, in
a volume of liquid not exceeding 10 c.c.

For the detection of arsenic in such liquids as beer (which is con-
veniently described here) Hehner recommends the following process,
which is based on the Marsh-Berzelius method : —

The hydrogen is obtained from zinc and hydrochloric acid. It
is important that the zinc should be arsenic free, and sensitive to the
presence of arsenic under the conditions of the test. The fact pointed
out by Dyer, that certain forms of zinc do not yield a good mirror is
confirmed, Hehner stating that this is the case with the metal cast in
rods. Granulated zinc should be used. The hydrochloric acid should
be subjected to vigorous boiling to drive oft" every trace of arsenic.
About 10 grms. of zinc are then introduced into a 250 c.c. flask, fitted
with a two-hole rubber cork. Through one hole passes a tap funnel,
through the other, the exit tube, connected with a tube holding a roll
of dry lead acetate paper, then a plug of cotton-wool, then granu-
lated calcium chloride, about 3 inches in length, and then another
plug of wool. To this is attached the hard-glass reducing tube, quill-

THE CHEMICALS OF THE PHARMACOPCEIA. 667

size, drawn out in the middle to a thickness corresponding to a standard
wire gauge No. 13 (0*092 inch) at the place where the arsenical mirror
is to make its appearance. Five c.c. of water are run into the flask, fol-
lowed by 10 c.c. of the pure HCl. The issuing hydrogen is then ignited,
a Bunsen flame applied to the reducing tube, and the time noted. The
Bunsen is removed in fifteen minutes, when no arsenical mirror should
be apparent. Having thus assured the absence of arsenical impurity
in the reagents and apparatus, 10 c.c. of beer are slowly dropped in, care
being taken to introduce no air, and the Bunsen flame reapplied to the
reducing tube. If frothing occurs a little strong alcohol may be used,
but if possible this should be avoided. The test should be continued
for fifteen minutes. The mirror- bearing tube is now disconnected,
hydrogen removed by suction, and the narrow parts fused up on both
sides of the mirror. On gently drawing this closed tube through a
flame until the mirror disappears, the arsenic therein is oxidized by the
contained air, and, on cooling, glistening crystals of Asfi^. will be
obtained, which are evident to the naked eye with even so little as one
or two-thousandths of a mg. Selenium and tellurium do not, in Heh-
ner's opinion, interfere with the production of the mirror. For quan-
titative determination, a series of standard mirrors should be prepared
by this method, with arsenical solutions of known strength which
should not, at the maximum, contain more than 0*01 mg. These
standard mirrors are fused off, mounted on white card, and kept in the
dark, for comparison with those obtained in ordinary tests. With
sulphuric acid, a preliminary test with 10 grms. diluted with water
should be made. If the mirror be too strong, a fresh experiment with
a less quantity should be performed. With glucose and sugar, 10 grms.
is a convenient quantity to work with. In the case of malt, 10 grms.
should be washed with dilute HCl at first cold, then warmed, three or
four times and the test made on the extract.

Allen (" Chem. News," lxxxii. 305) prefers the following modifica-
tion of the Reinsch test. Pure hydrochloric acid from which the first
10 per cent has been removed by vigorous boiling, so as to ensure
freedom from arsenic, is employed. One hundred c.c. of beer are used
for the analysis, which is first boiled for a few minutes with a little
HCl and bromine water. To ensure the reduction of arseriic to
arsenious oxide, a little cuprous chloride in hydrochloric acid solution
is then added. About 1 c.c. of copper foil is now introduced and the
whole boiled for 30 minutes, water being added to keep the volume
approximately constant. If the copper has been stained, it is dried in
the water oven, cut into strips and heated in a long narrow tube,
when characteristic crystals of arsenious oxide acid are visible. In
doubtful cases, it is best to repeat the whole test several times, and sub-
mit the combined deposits to Marsh's test as described by Hehner above.

For the quantitative determination of arsenic Allen (" J. S. C. I."
XX. 197) modifies the above process.

One litre or 500 c.c. of the beer, according to the qualitative in-
dication of arsenic, is evaporated down to 200 c.c. in a basin, about 20
c.c. bromine water and 20 c.c. hydrochloric acid added, and the excess
of bromine boiled ofi", the volume of the liquid being kept at about 200

668 FOOD AND DKUGS.

c.c. A few drops of a freshly prepared solution of cuprous chloride

in hydrochloric acid, and three or four pieces of pure copper foil are

then added, and the boiling continued for half an hour. The pieces

of copper are removed and replaced by .fresh ones till no darkening

takes place. They are treated in a beaker with hydrochloric acid

and crystals of potassium chlorate, taking care to have excess of the

latter, till the arsenic is removed. The solution is warmed till free

from oxides of chlorine, and transferred to a distilling flask. An

alternative method, due to Clark and Jones, may be used. This is to

cover the copper with water in a beaker, add 10 c.c. of 5 per cent caustic

soda and 10 drops of solution of hydrogen peroxide, and allow to stand,

in the cold, till the arsenic is dissolved. A few drops of cuprous chloride

solution and about 15 c.c. hydrochloric acid are added and the liquid

distilled into water till the residue in the flask measures about 15 c.c.

The distillation is repeated with 20 c.c. fuming hydrochloric acid, the

combined distillates rendered alkaline with ammonia, and then slightly

acidified with hydrochloric acid, keeping cool by immersion in water.

It is then neutralized with sodium bicarbonate, a slight excess of

sodium bicarbonate added, and titrated with iodine solution

iiOO

(using starch as an indicator), 1 c.c. of which represents 0-0002475

grm. As^O^,.

A blank determination should be made on the reageants employed,
the amount of iodine solution required being deducted. The hydro-
chloric acid, copper, cuprous chloride, caustic soda, and hydrogen
peroxide are all liable to contain traces of arsenic.

Thorpe (" Journ. Chem. Soc." 83, 974) has recommended a pro-
cess which depends on the electrolytic formation of hydrogen, instead
of the use of zinc and acid. In the author's opinion it is too compli-
cated for general use and does not give any better results than Hehner's-
modification of the Marsh-Berzelius test.

LEAD.

The committee of reference in pharmacy have recommended a
quantitative test for the determination of lead, to the British Pharma-
copceia committee, which is based on the colorimetric test of Warington..

The following are the contents of this report : —

Quantitative Colorimetric Lead Test.

Apjmratus.

Note. — All glass apparatus used should be lead-free.

Nessler Glasses. — These should be thin and of lead-free glass.
They should be about 25 mm. in diameter, and about 100 mm. in
height to the 50 c.c. mark.

Solutions.

Strong Lead Solution. — Dissolve 0*16 grm. of pure re-crystallized
lead nitrate in water, adding 50 c.c. of strong nitric acid, and dilute

THE CHEMICALS OF THE PHAKMACOPCEIA. 669

with water to 100 c.c. This solution is of such strength that 1 c.c. =
O'OOl grm. Pb. and forms a permanent stock solution.

Dilute Lead Solution. — Dilute 1 c.c. of the strong lead solution,
measured from a burette, with water so that the resulting solution
measures 100 c.c. This solution is of such strength that 1 c.c. =
O'OOOOl grm. Pb. It is the solution actually used in the tests and
should be freshly prepared.

Potassium Cyanide Solution. — Dissolve 10 grms. of potassium
cyanide (98 per cent) in water, add 2 c.c. of solution of hydrogen per-
oxide and make up to 100 c.c.

Note. — This solution, after being allowed to stand, should be tested
under the conditions of the quantitative colorimetric test to see that it
gives no colour with the dilute lead solution.

Sodium Suljohide Solution. — Dissolve 10 grms. of crystallized
sodium sulphide in water and make up to 100 c.c.

Mode of Testing {Geiieral).

Two solutions in hot water of the substance under examination
are made : —

(1) The primary solution containing 12 grms. of the substance.

(2) The dilute solution containing 2 grms. of the substance.
Each solution is filtered (if necessary), made alkaline with ammonia,

and treated with 1 c.c. of the potassium cyanide solution.

If the colour of the two solutions differ much, this may be rectified
by the cautious addition of a highly diluted solution of burnt sugar.

Then, by the method of trial and error (well known in water
analysis as " Nesslerizing ") is determined the quantity of dilute lead
solution which must be added to the dilute solution, in order that
there may be equal colorations produced upon the addition of two
drops of sodium sulphide solution to both the primary and dilute
solutions, after dilution to the 50 c.c. mark. In these circumstances,
each c.c. of lead solution required represents 1 part per million of lead
in the substance examined.

The colorations may be viewed by light reflected from a white tile
through the Nessler glasses inclined at an angle to the observer.

Note. — In some cases, 7 or 4 grms. only are used in the primary
solution. In these cases each c.c. of dilute lead solution required will
represent 2 or 5 parts per million of lead respectively.

This report has been criticized somewhat severely by Harvey and
Wilkie ("Chemist and Druggist," 1909 ii. 92) with whose remarks the
author, in the main, agrees. They state that in most cases incon-
veniently large quantities of substance are used, the natural limitations
imposed by" solubiUty not being taken into consideration. This is ob-
jectionable, as one must either work with hot solutions or permit a
certain amount of crystallization to occur. Pure metallic lead forms
a better stanidard than the nitrate, and in preparing the strong lead
solution it is highly necessary to cool before finally adjusting to volume.
As alkaline hydrogen peroxide under certain conditions oxidizes lead
sulphide to the sulphate in the cold, the potassium-cyanide solution
should be tested to ensure that it exercises no apparent solvent action

670 FOOD AND DRUGS.

on lead sulphide. The committee has also failed to direct that solutions
should be cooled to laboratory temperature before adding the sodium
sulphide, failing which consistent results cannot be obtained. They
do not approve the viewing of the tests at an angle inclined to the ob-
server. It is much better to stand them on a good white surface and
view them from above.

The recommendation to filter the aqueous solution is obviously bad
as, in the case of cream of tartar for example, lead often occurs as
minute particles of metallic lead, which would thus be filtered off and
ignored. Since iron and copper frequently occur with lead as im-
purities, these have to be reckoned with and as potassium cyanide
causes a yellow colour to appear with iron salts, it renders the process
unworkable when more than the faintest traces of iron are present.
The use of a colouring matter such as burnt sugar is also to be de-
precated.

The addition of tartaric acid prevents the interference of iron with
the test. Teed (" Analyst," xvii. 142) pointed this out, and recom-
mends the following details for carrying out the test : —

A measured quantity of the liquid is mixed in a cylinder or white
basin with a few c.c. of ammonia, a few drops of solution of potassium
cyanide, and then with a drop of ammonium sulphide. A small quan-
tity of pure tartaric acid is added unless it be already present. The
presence of lead will be indicated by the dark coloration produced, and
its quantity can be estimated by imitating the coloration with known
quantities of lead precipitated under the same conditions. Iron does
not interfere with the test, as it is kept in solution by the tartaric acid,
and is then converted by the potassium cyanide into a ferro- or ferri-
cyanide, which is not affected by ammonium sulphide. Copper does
not interfere with the test, as copper sulphide is soluble in potassium
cyanide.

As a very delicate test for the detection of lead in sulphuric acid,
Teed proposes the addition to the strong acid of a drop of hydrochloric
acid or of a small crystal of common salt. Chloride of lead is thus
precipitated and recognized by a peculiar pearly opalescence of the
liquid.

The colorimetric process is due in the first instance to Warington.
Bennet (" Chemist and Druggist," 64, 633) gives the following details for
the determination of lead in nitric and tartaric acids and cream of
tartar. In other cases, when these details are appUcable, tartaric acid,
as recommended by Teed, should be added to prevent the interference
of iron salts.

Ten grms. are dissolved in 15 c.c. of distilled water, 25 c.c. of solution
of ammonia (10 per cent) added (for cream of tartar 10 c.c. is sufficient)
and made up to 50 c.c. One drop of solution of sodium sulphide (10
per cent) is added, and the coloration produced is matched in Nessler
glasses by adding from a burette a standard solution of lead acetate
(containing O'OOOl grm. of lead in 1 c.c.) to 50 c.c. of distilled water
containing a drop of sodium sulphide solution. Each tenth part of 1
c.c. will then represent 1 part of lead per million.

If iron be present, the addition of 1 c.c. of a 10 per cent solution

THE CHEMICALS OF THE PHAKMACOPCEIA. 671

of potassium cyanide is necessary, copper also being eliminated since
copper sulphide is soluble in potassium cyanide. A yellow coloration
is often caused by the addition of the cyanide, but this gradually dis-
appears on warming. If only slight, it may be matched before add-
ing the sodium sulphide, and the amount of standard lead solution so-
used deducted from the total quantity required. It is essential that
the solution should be distinctly alkaline, or the full colour is not de-
veloped.

Sodium sulphide is much preferable to either sulphuretted hydro-
gen or ammonium sulphide, as no turbidity or coloration is produced
in the absence of metals, while its comparative freedom from odour is
also a distinct advantage.

x\ccording to Carles, one of the leading French experts on the wine-
and tartar industries, lead exists in cream of tartar, not as tartrate but
as sulphate dissolved by the bitartrate. To detect and estimate it in
cream of tartar Carles uses 10 grms. of the sample. It should be-
finely powdered, carbonized in a porcelain capsule, extracted with hot
water, and filtered through a small folded paper. The insoluble sul-
phate and carbonate of lead remain with the carbon on the filter, which
is washed to get rid of potassium carbonate. It is then treated with
dilute nitric acid and filtered.

The filtrate is rendered alkaline with excess of ammonia, and the
solution containing any copper present is filtered off, the precipitate is-
dissolved in hot dilute HCl and the lead precipitated as sulphide and
weighed. With minute quantities of lead, however, this gravimetric
process does not give very useful results.

If iron be absent, the amount of copper present can be ap-
proximately determined by (1) matching the coloration of a dilute
acetic acid solution of the substance with a few drops of potassium
ferrocyanide solution, against standard amounts of a solution of pure
copper sulphate containing 0*1 mg. of Cu per c.c.

(2) By determining the lead by the sodium sulphide coloration pro-
cess, and at the same time preparing a solution under identical con-
ditions, but without the addition of potassium cyanide. A solution of
copper sulphate (equivalent to 0*1 mg. Cu per c.c.) is then added
to the tube containing the standard amount of lead, until the colour
is matched, which then gives approximately the amount of copper
also.

A satisfactory method of determining the copper in the presence
of lead is an electrolytic process, but if only slight traces are present,
the amounts obtained are too small to weigh accurately. The details
of the experiment must be carefully observed, since whilst lead
separates normally as an oxide at the anode, some of it sometimes
separates simultaneously in the metallic form at the kathode. Sul-
phuric acid, as recommended by Hill ("Chemist and Druggist," 23
May, 1908), is not so suitable as nitric acid. It is necessary to use at
least 100 grms. of most chemicals, dissolved in about 200 c.c. to 250 c.c.
of a mixture of 9 parts of water and 4 of concentrated nitric acid.

A current of 1*0 to 1'5 amperes should be maintained, with an
electromotive force of about 1-4 volts. Under these conditions, all

672 FOOD AND DEUGS.

the lead will be separated as peroxide of lead at the anode ; this
electrode is removed, washed with distilled water, dried at 180° to
^00° and w^eighed. The increase in weight may be calculated as
Pb02, and if multiplied by 0-866 gives the amount of metallic lead.
A fresh electrode is put in, and the current allowed to continue for
about 4 hours when the copper is entirely deposited, and is weighed on
the kathode. It is convenient to use a " cone "or " jacket " electrode as
the anode, the kathode being smaller. After replacing the cone by a
fresh one, the current should ba reversed and the copper originally
deposited on the smaller surface, is dissolved again by the electrolyte
and it is all deposited on the cone which is now acting as kathode.

Numerous experiments have convinced the author that the follow-
ing method is the most accurate available. It is the combination of
the principle, first suggested (so far as the author can trace) by C.
Hill, that while the trustworthiness of these colorimetric tests is de-
pendent upon the comparison being made between two solutions of the
same substance, it is independent of the concentration of ihose solu-
tions within wide limits.

With the necessary corrections for the presence of iron suggested
by either Teed or Harvey and Wilkie, it is substantially that of the
report of the Reference Committee detailed above, with the slight
modifications necessary when iron is present, and is that given by
Harvey and Wilkie in the criticism already mentioned.

Mode of Testing {General). — Two solutions of the substance under
examination are made in water — (1) The primary solution, containing
5 grms. of the substance ; (2) the dilute solution, containing 2-5 grms.
of the substance and 5 c.c. of the dilute lead solution. The volume
of each solution should be about 40 c.c. Four drops of acetic acid
(33 per cent) are added and then 1 c.c. of the potassium-cyanide
solution. After well mixing, excess of '880 ammonia is added. The
volume of each is adjusted to 50 c.c. at laboratory temperature, and
two to three drops of sodium-sulphide solution are added. Under these
-conditions the coloration developed in the primary solution should not
be darker than that of the comparison solution, thus ensuring that the
lead present does not exceed twenty parts per million.

Notes. — In a few cases 7-5 grms. are used in the primary solution, the compari-
son solution as usual containing 2 5 grms. and 5 c.c. of the dilute lead solution. If
the colour in the primary solution be darker it must be matched and the amount of
lead calculated.

Should any insoluble matter be present, the solutions should be boiled if neces-
sary with more acetic acid, correspondingly more ammonia being subsequently
used.

The primary and secondary solutions prior to the addition of sodium sulphide
must be colourless ; if this is not the case special treatment must be given (cf.
infra) ; in addition they must in all cases be cooled to laboratory temperature.

In all cases where the acid solutions, treated with potassium cyanide,
and then by ammonia, give a colourless solution, or one so faintly
yellow as to be almost inappreciable, the interference of iron need not
be feared. But if this be not the case then, if the colour be due to
ferric iron, it will disappear on the addition of a little more ammonia and

SACCHAEIN. 673

gently boiling. If this is not the case, a little tartaric acid may be
added to a fresh preparation, and if the solution still remains coloured
after treatment with potassium cyanide and ammonia, then Harvey
and Wilkie's more drastic treatment should be employed. This is as
follows : —

After solution in water, four drops of hydrochloric acid, specific
gravity 1*16, is added, then 1 c.c. of a saturated aqueous solution of
sodium sulphite (Na^SOg . 7H^,0), and the solution heated gently until
the colour due to ferric iron suddenly bleaches. To the colourless
solution is added a mixture of 1 c.c. 10 per cent potassium-cyanide
solution and 2 c.c. '880 ammonia. The whole is again heated and
gently boiled, if necessary, until quite colourless ; then it is cooled,
adjusted to 50 c.c. and treated with sulphide in the usual manner.

Saccharin.

Sacchariny ortho-benzoic-sulphinide C^H^ (^ <:,^ yNH, is official

in the British Pharmacopoeia under the name Glusidum.

The official requirements for this substance are as follows : It is
soluble in 400 parts of cold water, in 24 parts of boiling water and in
25 parts of 90 per cent alcohol : only slightly soluble in ether and
chloroform. Eeadily soluble in alkalis. When a warm solution of
NaoCOg is neutralized with saccharin and the liquid evaporated to
dryness, 100 parts yield about 113 parts of "soluble saccharin".
Neither saccharin nor soluble saccharin is blackened by HgSO^ even
when gently warmed. On evaporating with solution of KOH and
gently fusing for a few minutes, the residue when cooled and dissolved
in water and slightly acidulated with HCl, gives a purplish colour
with ferric chloride solution. 0*5 grm. in 80 c.c. of warm water set
aside for twelve hours deposits tabular crystals which melt at 218*8°
to 220°, and should not, even when briskly shaken, deposit crystals
melting at a higher temperature (absence of sulphamido-benzoic acid; .

Pure saccharin is a white crystalline powder melting at 220°
(according to Allen at 224°, but this is not correct). It is soluble in 50
parts of amyl acetate and in 20 parts of ethyl acetate.

Commercial Saccharin. — In commerce, saccharin occurs in twa
forms, (1) "550" which is intended to be 550 times sweeter than
sugar and which corresponds to the B. P. glusidum, and (2) " 330 " an
impure form, stated to be 330 times as sweet as sugar. Each has its-
corresponding soluble variety.

Mineral adulterants, which may be present, are at once indicated
by the ash, which should be infinitesimal in the case of saccharin.

Sugars can be detected by exhausting the solid substance repeatedly
with a mixture of equal volumes of ether and petroleum ether. 0*5
grm. should be extracted three or four times by well shaking with
successive quantities of 50 c.c. of the solvent. If the undissolved
residue contains sugar it will have a sweet taste, and this may be de-
termined by inversion and titration with Fehling's solution, which is.
not reduced by saccharin.

VOL. I. 43

674 FOOD AND DKUGS.

When unmixed with other substances saccharin may be approxi-
mately determined by titration with decinormal alkah, 1 c.c. = 0-0183
grm. of saccharin.

Sulphamido-benzoic acid is an impurity specially recognized in
the Pharmacopoeia. This acid melts at 281° to 282° and may be
detected by shaking 1 grm. with 70 c.c. of ether for five minutes, and
filtering and collecting the undissolved residue. As saccharin is far
more soluble in ether than the sulphamido-benzoic acid, the latter, if
present, accumulates in the undissolved portion, and its melting-point
will be above that of saccharin. If the residue melts at above 224°,
the presence of this impurity is almost certain.

The Determination of Saccharin in Commercial Samjjles. — The pro-
cess of Kemsen and Burton gives accurate results (" Amer. Chem. Journ."
1889, 403). Two grms. are boiled under a condenser for an hour with
100 c.c. of dilute HCl. The liquid is evaporated to 15 c.c. when sulph-
amido-benzoic acid will separate if present. After four hours standing
they are collected on a tared filter, washed with a little cold water and
weighed. The solution and washings now contain acid ammonium
sulphobenzoate, which results from the decomposition of the saccharin
in the sample, and any acid potassium ortho-sulphobenzoate which
may have been present in the sample. . The liquid is evaporated to
dryness and the residue weighed. The potassium is then converted
into sulphate and weighed. From the weight of the potassium sul-
phate, the amount of acid potassium sulphobenzoate present can be
calculated (its formula is COOH . C,5H4(S03K)). The dry residue minus
this gives the weight of acid ammonium sulphobenzoate from which
the amount of true saccharin is calculated, by multiplying by 0"835.

Reid (" Amer. Chem. Journ." xxi. [6], 461) has devised a useful and
fairly accurate process of assay for commercial saccharin, based on the
fact that it is converted into the acid ammonium salt of o-sulphobenzoic
acid by boiling with hydrochloric acid, whilst p-sulphamido-benzoic
acid is unaffected.

The process is conducted as follows: 0-650 grm. of "saccharin"
is weighed out into a 100 c.c. flask, and 50 c.c. of dilute hydrochloric
acid are added (120 c.c. pure concentrated HCl in 1 litre). The flask
is fitted with a cork, through which a glass tube passes, 8 mm. wide
and 45 cm. long. After two hours' gentle boiling on a sand-bath, the
stopper is removed, and the solution allowed to evaporate to about 10
c.c. After diluting, the contents are washed out into an ordinary dis-
tilling flask. Twenty c.c. of a caustic soda solution (equal to 10 grms.
'Of NaOH) are added, the ammonia is distilled off into standard acid, and
the excess titrated back with KOH, cochineal being the indicator. To
cause the caustic soda solutions to boil evenly, steam is passed into the
distilling flask, from a flask containing water acidulated with H^SO^.

For alternative processes, Hefelmann (" Pharm. Central " 36, 228)
and Proctor ("Journ. Chem. Soc." 1905, 242) may be consulted.

The Detection of Saccharin in Beverages, etc. — Saccharin may
be extracted from foods and estimated by the following method.
If a liquid, 50 c.c. are taken ; if a solid, a suitable aqueous ex-
tract is made by warm alcohol and then diluted with an equal

SACCHARIN. 676

volume of water. The liquid is concentrated to about 25 c.c. and
the alcohol driven off, a little hydrochloric acid added, and the
liquid is extracted twice with 25 c.c. of amyl acetate. This solvent
dissolves but little colouring matter. The amyl acetate solution is eva-
porated to dryness, and the residue dissolved in a little bicarbonate of
potassium solution. A few drops of lead acetate solution are added,
the liquid filtered and the excess of lead removed by H.,S, and the
liquid again filtered. It is then extracted twice with ethyl acetate,
after being rendered acid with a trace of hydrochloric acid, and the
residue dried and tested by taste and other means. The best method
for a quantitative determination is to remove as much extraneous matter
as possible from the solution containing the saccharin, by means of lead
acetate, and then extracting twice with 50 c.c. of a mixture of equal
volumes of ether and petroleum ether. The residue obtained on eva-

N
poration is titrated with iqO ^''^''''^ hydroxide solution, and almost

theoretical results are obtained.

Bianchi and Di Nola ("Boll. Chim. Farm." 1908, 47, 599) give
the following method for detecting saccharin in beverages and foods.
The liquid, or in case of a solid a suitable alcoholic extract, diluted
with water, is evaporated to free it from alcohol, heated to boiling,
and acidified with about 20 drops of acetic acid per 100 c.c. The
liquid is cooled and then mixed with 10 c.c. of a 20 per cent lead
acetate solution. After half an hour, the excess of lead is removed by
a solution containing 10 per cen,t each of sodium sulphate and phos-
phate. The filtered liquid is concentrated to 70 c.c, acidified with 6
c.c. to 8 c.c. of 25 per cent sulphuric acid and extracted with its own
volume of a mixture of equal quantities of ether and petroleum ether.
The residue from this extraction is tested. (1) By tasting, (2) by fusing
with potash at 270°, adding a few drops of HCl, then ferric chloride,
when a violet colour results, (3) also for salicylic acid which would
be present in this residue if present in the beverage, by direct testing
with ferric chloride when a violet colour results. If salicylic acid be
found the potash fusion test will not, of course, be relied on, until the
salicylic acid is removed. -In this case the residue should be mixed
with dilute HCl, and bromine water added. The precipitated bro-
mine derivative of salicylic acid is filtered off, and the filtrate is
rendered alkaline with KOH, and again dried, and the residue, now
free from salicylic acid, tested as above.

Allen's method is especially adapted to the detection of saccharin in
malt liquors, in the analysis of which bitter substances may interfere
with the characteristic taste of the saccharin separated. It is claimed
that minute quantities can be detected by this method. The liquid
to be tested is evaporated to a small bulk, acidified with a little phos-
phoric acid, and extracted with ether. The residue from the ether is
rendered alkaline with sodium hydroxide, ignited, and the ash tested
for sulphate. The presence of sulphate is regarded by Allen as con-
clusive of the presence of saccharin. The process is facilitated if the
liquid is first treated with a little lead acetate solution and filtered.
Excess of lead — if not too great — is immaterial.

676 FOOD AND DEUGS.

Kastle ("Jour. Chem. Soc." 1905, 503) gives the following delicate
reaction which will detect as little as one quarter of a mg. of saccharin.
A small quantity of the solid residue, e.g. the residue from an ex-
traction, is mixed with a minute amount of a mixture of 5 c.c. of
phenol and 3 c.c. of sulphuric acid, and heated to 160° to 170° for
five minutes. The mass is dissolved in cold water and a little NaOH
added. A purple or rose-red will result if saccharin be present.

Truchon gives the following method, which is used in the muni-
cipal laboratory of Paris. At least 200 c.c. of liquid, after acidifying
with phosphoric acid, are extracted three times with 35 c.c. to 40 c.c. of
a mixture of equal parts of ether and petroleum spirit. The extract
is washed with water, evaporated in a platinum dish, 5 to 6 drops
of a solution of pure caustic soda are added, and the mass is carefully
brought to quiet fusion over a small Bunsen flame. The end of the
reaction is indicated by the disappearance of the small gas bubbles.
The mass is extracted with distilled water, the solution acidified with
sulphuric acid, extracted twice in succession with 30 c.c. of petroleum
spirit, the petroleum separated, evaporated in a porcelain dish, and a
drop of a very dilute (1 : 10,000) solution of iron chloride added to
the residue. If saccharin was originally present, the well-known
violet coloration is produced by the salicylic acid formed from the
saccharin.

Tertoni (" Zeit. Untersuch. Nahr. Genuss." 1909, 12, 577) has
devised processes for the identification of saccharin in the presence of
benzoic, salicylic, and citric acids, etc., which separate with the saccharin.
The substance is extracted as described in the above processes, by a
mixture of ether and petroleum ether, and the residue obtained by
evaporation, tested as follows : —

In the Presence of Benzoic Acid. — Weigh the residue so obtained,
in a tared porcelain capsule, and heat in an oven the temperature of
which is 110° to 115° C., until the weight becomes stable. The
saccharin will remain unchanged whilst the benzoic acid present will
be partly volatilized. To get rid of the benzoic acid entirely dissoh e
the residue in alcohol. The saccharin may be precipitated by adding
silver nitrate solution in small quantities. Collect the precipitate on
a filter, wash with alcohol, dry at 100° C. then weigh. Its formula is
C^H^SOgNAg, and 37'24 per cent of it is silver. Benzoic acid cannot
be precipitated in this way.

In the Presence of Salicylic Acid. — Precipitate the salicylic acid with
bromine. Eemove the tetra-bromophenol by filtration and then extract
the saccharin by petroleum ether and ether.

In the Presence of Fats, Fruit Essences, etc. — If there are fatty
substances as well as the saccharin in the residue, the latter can be
estimated by using a mixture of sodium carbonate and potassium
nitrate to fuse with the residue, then precipitating the resulting
sulphuric acid in the usual manner. Heat the fatty residue with
hydrochloric acid (sp. gr. I'l) to a temperature of 120° to 130° C. in
an autoclave. The nitrogen of the saccharin is converted into am-
monia, and the saccharin can be estimated by rendering the liquid
alkaline and distilling the ammonia into normal acid. There must, of

SALICYLIC ACID. 677

course, be no sulphur and nitrogen compounds except the saccharin
in the ethereal extract.

Jorgensen's process is a most useful one for the detection of
saccharin where easily-oxidizable interfering substances are present.

Evaporate 500 c.c. of beer on a water bath until it is the consistency
of a syrup, then mix the residue with a quantity of 96 per cent alcohol
and stir well. Decant the alcoholic solution, evaporate the residue to
dryness, then moisten the solid residue with water and again extract
with alcohol. The united extracts should be evaporated until all the
alcohol has been expelled. Add a few drops of sulphuric acid to the
aqueous syrup, filter, and shake out the filtrate with several successive
portions of ether. Evaporate the ethereal solution, take up the residue
with a little water, add a small quantity of dilute sulphuric acid and
saturated potassium permanganate solution, drop by drop, until there
is a permanent pink coloration ; to remove excess of permanganate add
saturated oxalic acid solution, avoiding an excess of oxalic acid. After
filtering, extract the colourless solution with a mixture of ether and
petroleum spirit.

If the beer contains saccharin, the crystalline residue obtained by
evaporating the ethereal extract, will have an extremely sweet taste.
The treatment with permanganate will have removed any salicylic acid
present in the beer, and the saccharin obtained may be further identi-
fied by converting it into salicylic acid and applying the usual tests for
the latter substance.

Salicylic Acid.

Salicylic acid CgH^ . OH . COOH (ortho-hydroxy-benzoic acid) is an
acid which occurs in nature, generally as its methyl ester, in numerous
plants ; but the greater part of the commercial article is prepared
artificially by heating potassium phenate with CO2, or by a similar
process.

It is official in the Pharmacopoeia, which requires it to have the
following characters. Soluble in about 500 parts of water, in 3 of
alcohol, in 2 of ether and in 200 of glycerin. A 1 per cent solution
in alkalis or in certain saline solutions yields a yellowish precipitate
with uranium nitrate solution. Melting-point 156° to 157°. Volatile
below 200° without decomposition. Solution of ferric chloride gives
with an aqueous solution a violet colour. If shaken with water, the
water filtered and evaporated, the residue is white without a buff-
coloured fringe. It dissolves in cold H2SO4 imparting no colour to it
in fifteen minutes. If 1 grm. be dissolved in solution of Na^COj and
the liquid shaken with ether, and the ether allowed to evaporate
spontaneously, the residue, if any, should not smell of phenol.

Salicylic acid occurs in commerce in four varieties (1) natural, that
is made from a naturally occurring methyl salicylate, such as oil of
sweet birch, (2) physiologically pure, that is, freed from traces of im-
purities which may have a bad effect therapeutically, (3) crystal, and
(4) powder, which may or may not contain traces of such impurities.
Apart from its use in medicine, it acts as a very powerful preservative.

678 FOOD AND DRUGS.

and as such is often added to foods and wines, so that its detection is
a matter of considerable importance.

Pure salicylic acid melts at 156-.7° and has a specific gravity 1-4:83
at 4°. Traces of impurities such as para-hydroxy-benzoic acid and
hydroxy-phthalic acid lower the melting-point appreciably.

This drug is rarely adulterated, but traces of impurities must be
guarded against, although the ordinary grades are now usually suf-
ficiently pure for use in medicine.

xlccording to Kolbe the absolute whiteness of the residue obtained
on evaporating a solution of 0*5 grm. in 5 c.c. of alcohol, is important.
If a coloured deposit is obtained the sample should be rejected.

Phenol should be absent, and the following test (due to Muter)
may he applied in addition to that given in the Pharmacopoeia : —

Boil 1 grm. in 15 c.c. of water, cool, pour off the solution and add
to it 1 drop of a saturated solution of KHCO3, 1 drop of aniline and
5 drops of a solution of calcium hypochlorite. In the presence of
phenol a deep blue colour will be produced.

The only natural impurity found to-day in salicylic acid made from
phenol is para-cresotic acid, which may occur to the extent of 3 per
cent in badly made specimens ; whilst twenty years ago as much as 15
to 20 per cent of this and similar impurities was common. Fischer
has devised the following method for the detection of para-cresotic acid.
CaCOg (1 to 2 grms.) is boiled with 15 c.c. of water and 3 grms. of the
acid. The flask is kept well shaken over a flame until the volume is
reduced to 5 c.c. The liquid is cooled and the crystals separated by
filtration, and the mother liquor poured into a test-tube and evaporated
to 1 c.c. By rubbing the liquid against the side of the tube, crystalliza-
tion sets in. One c.c. of water is then added and the liquid filtered
through a plug of cotton-wool. A few drops of HCl are then added.
If over 1 per cent of cresotic acid be present, it separates and melts
when the water is boiled, sinking to the bottom of the tube as oily drops.

Cresotic acid may be determined, if previously found to be present,
by titrating the sample with baryta water. But unless a comparatively
large amount be present, the results are not reliable, for the following
reason. One grm. of salicylic acid requires 726-3 c.c. of centinormal
baryta water for neutralization, whereas 1 grm. of cresotic acid only re-
quires 659-4 c.c. Each 1 per cent of cresotic acid as an impurity
therefore only lowers the quantity required by 0*67 c.c. It is impos-
sible to work accurately on so enormous a volume, so that by using
0-2 grm. of the sample, this difference is reduced to 0-134 c.c, which
means that an error in titration of jq of a c.c, where over 100 c.c of
titration liquid are used, entirely vitiates the result. If it be necessary
to use this process 0-2 grm. of the acid (previously purified by solution
in ether, filtration, evaporation, and drying at 60°) are placed in a fiask,
a few drops of alcoholic solution of phenol-phthalein added, and the
centinormal solution run in to near the neutral point. The flask is
then shaken over a flame, and directly the remainder of the acid is
dissolved, the titration is completed.

Ewell and Prescott {v. " Analyst," xiii. 237) distil with lime and so
convert the cresotic acid present into cresol, 15 grm. of the acid and 15

SALICYLIC ACID.

679

grm. of CaO are dried and well mixed together with 15 grm. lion filings.
The mass is distilled in a retort and the distillate collected. It is treated
with just enough water to liquefy it and the liquid mixed with an
equal volume of 9 per cent aqueous NaOH. Water is then added until
precipitation commences. Erom the volume of water necessary to cause
precipitation the proportion of cresotic acid may be approximately
calculated by the following table, but amounts under 5 per cent are
not to be detected with certainty : —

Volume of H2O
necessary for
Precipitation.

Per cent of

Cresol
in Distillate.

Per cent of

Cresotic Acid

in Sample.

6-7
6-0
5-25
4-5

10
15
20

4-9

9-8
14-8
19-8

The Detection of Salicylic Acid. — An acidified alcoholic solution o._>
salicylic acid is slowly reduced by sodium if warmed. The reduction
product is salicylic aldehyde, easily recognized by its characteristic odour
of meadow-sweet. On heating salicylic acid with H.)S04 and methyl
alcohol, methyl salicylate is formed which has the characteristic winter-
green odour. Curtman (" Jour. Chem. Soc. " Lii. 188) recommends
the latter test being carried out if salicylic acid be suspected in wine,
etc., in the following manner. Four c.c. of the wine or beer should
be mixed with methyl alcohol (CH3OH) and then 2 c.c. of pure HSO^
added cautiously. The liquid is agitated, heated for 2 minutes, allowed
to cool and then heated to boil.ng. If salicylic acid be present,
the odour of wiuter-green is perceptible. If only minute traces be
present, it may be necessary to heat a third time.

Jorissen gives the following reaction, which is not yielded by
benzoic or cinnamic acids. A solution of salicylic acid or a salicylate
is treated with sodium nitrate and acetic acid and then a drop or two
of copper sulphate solution, and the liquid boiled, when, if salicylic
acid be present, a blood-red colour results.

The test, however, on which practically every analyst relies is the
intense violet colour produced by solution of salicylic acid with a drop
of neutral solution of ferric chloride. The test answers wdth solutions
containing 1 in 110,000. The colour is destroyed by alkalis and
most acids. In examining foods, wines, etc., suspected of containing
salicylic acid it is necessary to remove various interfering substances,
so that the liquid (or the extract by dilute KOH solution from the
solid matter) is acidified and extracted with ether, petroleum ether or
chloroform. The best solvent is a mixture of equal volumes of ether
and petroleum ether. The solvent is extracted with w^ater containing
a trace of free alkali and the aqueous liquid exactly neutralized with
HCl, and a few drops of ferric chloride solution (or better, solution
of iron-alum) added, when a violet colour results if salicylic acid be

680 FOOD AND DRUGS.

present. If minute quantities only are present, the ether may be
evaporated and the residue tested with the iron solution.

Or the residue from the evaporation of the solvent may be treated
with 2 drops of HNO3 and the ammonia added. The acid is converted
mto picric acid, and a thread of white wool will be dyed a deep
yellow if treated with the few drops of liquid.

Vitali ("Repertoire de Pharmacie," [3], 19, 39) prefers toluene as
the solvent as it does not dissolve out any interfering substances at all.
He also confirms the presence of salicylic acid by adding a drop of
dilute HgSO^ and 1 drop of a very dilute and almost colourless solution
of copper sulphate to the residue left by evaporation of the solvent.
In the presence of a minute trace of salicylic acid, a green spot will re-
main on evaporation.

S. Harvey ("Analyst," xxviii, 2) describes the quantitative applica-
tion of this reaction to salicylic acid in wine, beer, etc., his method
being a slight improvement on previously described similar processes.
Jn the author's laboratoiy this has been found to give very accurate
results, using a solvent consisting of equal volumes of ether and
petroleum ether.

An aqueous 1 per cent solution of iron-alum, to w^hich a few drops

of H2SO4 have been added as a preservative, is recommended for the

colorimetric determination of salicylic acid. The tint given by this

reagent is more definite and persistent than that obtained with Fe.,Cl,;.

The acid is extracted from a known volume of the previously acidified

solution by two successive shakings out with ether. The bulked -ether

N N .

extracts are then shaken out with _. or — - alkali, the alkaline solu-

2 10

tion exactly neutralized wdth acid and diluted to a definite volume of

250 c.c. or 500 c.c. One hundred c.c. of this solution are treated, in a

Nessler glass, with 2 c.c. of the iron-alum reagent, and the colour

matched with a known volume of freshly prepared standard solution

of salicylic acid containing O'OOl grm. or 0-0001 grm. of salicylic in

each c.c.

No larger quantities than 2 mgs. of salicylic acid should be used
for the actual determination.

Messinger and Vortmann (" Berichte," xxii. 2312, and xxiii. 2753)
have proposed an iodometric process based on the addition of an excess
of standard iodine to the acid in a dilute alkaline solution. The acid
is precipitated as a substituted iodo compound and the excess of iodine
is determined by titration with thiosulphate solution. Six atoms of
iodine enter into the reaction with one molecule of salicylic acid : —

^«^Kca^a "^ ^^^^^ "^ ^^ ^ ^'ft^Kcak + ^^^^ + ^^^^

One c.c. of decinormal iodine corresponds to 0-0023 grm. of salicylic
acid. In order to get the best results, about 0'15 grm. of salicylic
acid is the best quantity to work upon ; the solution should not con-
tain more than 0*5 c.c. of a 10 per cent solution of NaOH : excess of
decinormal iodine is added at about 50° C, and the liquid, in a closed

SALICYLIC ACID. 681

flask, allowed to stand for a few minutes. Excess of dilute sulphuric
acid is then added, the liquid filtered, the precipitate washed with
water and the filtrate titrated with standard thiosulphate solution.

Fresenius and Grunhiit (" Zeit. Anal. Chem." xxxviii. 292)
have criticized this method very adversely, but as long as carried out
carefully, avoiding excess of alkali, fairly good results may be obtained.

Freyer's process (" Chem. Zeit." xx. 820) is one of the best methods
for the determination of salicylic acid, where the quantity is not too
small (when of course the colorimetric process must be adopted).
This process is based on the fact that bromine water reacts with
salicylic acid as follows : —

OH
C,H / + 8Br = C^HBrg . OBr + 4HBr + CO,
\CO2H

Excess of bromine water is added to the salicylic solution, whereby
the bromine compound is precipitated and the excess of bromine
liberates iodine from potassium iodide ; the tribromphenol bromide also
enters into the reaction.

CgHBrgOBr + 2KI = C^HBraOK + 21 + KBr.

So that in calculating the results, only 6 atoms of bromine cor-
respond to 1 molecule of salicylic acid. This process should be carried
out as follows. A known weight of the sample is dissolved in water
with the aid of a little caustic soda, and a measure corresponding to
about O'l grm. of salicylic acid is diluted to about 100 c.c. with water, in
a stoppered bottle. Ten c.c. of HCl are added, and then a known
volume (50 to 60 c.c.) of a solution of sodium bromide and bromate
(about 10 grm. of NaBr and 3 grm. of NaBrOg per litre) added.
This solution is prepared by adding bromine to caustic soda solution,
or by dissolving the salts themselves. The bottle is closely stoppered,
well shaken and stood in the dark for an hour. A blank experiment
is conducted at the same time omitting the sample only. Excess of 10
per cent solution of potassium iodide is then added and the liberated
iodine titrated with decinormal thiosulphate solution. Each c.c. deci-
normal thiosulphate is equivalent to 0*0023 grm. of salicylic acid, so
that the excess of thiosulphate required for the blank experiment
allows the amount of salicylic acid to be calculated. Owing to the
fact that starch must not be added till quite close to the end re-
action, substances containing starch should be extracted with 90 per
-cent alcohol and the determination carried out on the extract. In sub-
stances containing sulphites or sulphurous acid, the liquid, such as
wine, etc., must be rendered alkaline, concentrated, rendered acid
and extracted with ether and petroleum ether. The solvent is then
extracted with dilute alkali and this aqueous solution used for the
•determination. Salicylic acid may be estimated in this manner in the
presence of benzoic acid, which does not react with the bromine solution.

For the examination of milk, fat and proteids should be removed

682 FOOD AND DEUGS.

by adding a mercuric nitrate in acid solution, and filtering the liquid.
The above methods can then be applied to the clear whey.

Metallic salicylates can be assayed direct if soluble in water, or
by dissolving in acid and precipitating the metal with KOH if not,
and examining the alkaline filtrate.

Seidell (" J. Amer. Chem. Soc." 1909, 31, 1168) considers that both
Freyer's and Messinger and Wortmann's process are of uncertain value
and strongly recommends the precipitation of dibrom-salicylic acid as
more reliable. This process, which is carried out as described below,
has given excellent results in the author's hands.

About 0-25 grm. is placed in a stoppered flask and a few c.c. of

N
water and about 50 c.c. of strong HCl added. — bromine solution (in

HCl) is added, until after well shaking and warming to about 90°, the
yellow colour of the last two drops of bromine solution is persistent for
five minutes.

The reading may then be taken as final, four atoms of bromine repre-
senting one molecule of salicylic acid. The results are, in the author's-
experience, within 2*5 per cent of the truth.

Camphor.

Camphor is official in the Pharmacopoeia, being described as a
white crystalline substance obtained from Cinnamomum camphor a.

Its specific gravity is given as about 0*995. It forms a liquid when
triturated with menthol, phenol, or thymol. Its solubility in water
is officially given as 1 in 700 ; in alcohol (90 per cent) 1 in 1 ; in
chloroform 4 in 1, and in olive oil as 1 in 4.

Camphor C^^HiyO, is the stearoptene of the essential oil of Laurus
camphor a {Cinnamomum camphor a). The commercial article is
dextrorotatory, the much rarer laevorotatory variety occurring in the
oil of Matricaria parthenium. Camphor has also been prepared
synthetically, but in the optically inactive variety. Camphor forms a
translucent, colourless mass (or powdery "flowers") melting at 175°
and boiling at 204°. It is dextrorotatory, the specific rotation varying
with the solvent. Landolt's formula gives the specific rotation for
different degrees of concentration : —

[a], = +55-4°-(ax^)

where q is the number of grms. of solvent in 100 grms. of solution,,
and a is a constant for each solvent. For alcohol « = 0*1372 and for
benzol 0"1632. Thus the apparent specific rotation of camphor, when
10 grms. are dissolved in 90 grms. of alcohol is

+ 55-4° -(90x0-1372°)= +43-052°.

The most exhaustive determinations are those of Partheil and Van
Haaren (" Journ. Soc. Chem. Ind." 1900, 684). They show that the
more dilute the alcohol, the lower the apparent specific rotation. Also,
as the percentage of camphor increases, the apparent specific rotation
diminishes. They give the formula P= 1*5152 a, where P is the^

CAMPHOR

683

percentage of camphor by weight, and a is the observed rotation for
200 mm.

To ascertain the volume percentage, the specific gravity of the
alcohol must be arrived at. This is given by the formula :■ —

100 -jp
^ ^ 100 where S is the specific gravity of the alcohol, p is the

1-05 p

percentage by weight of camphor and a is the specific gravity of the
alcoholic solution of camphor. They give the following table : —

No.

Specific Gravity
of Alcohol.

Camphor.

Rotation for
200 mm.

Specific
Rotation.

1
2
3
4
5
6
7
8
9
10
11

0-7896
0-8212
0-8505
0-8637
0-8781
0-8909
0-9007
0-7895
0-7895
0-9007
0-9007

Per cent

10
8-37
6-81
8-35
6-82

+ 6-98°
+ 6-78°
+ 6-69°
+ 6-65°
+ 6-60°
+ 6-59°
+ 6-59°
+ 5-79°
+ 4-69°
+ 5-48°
+ 4-40°

43-4362
40-6666
39-0439
38-1439
37-2755
36-7622
36-4008
43-2142
43-1411
36-2929
35-6951

The above formula enables the camphor to be accurately determined
in spirit of camphor.

The Spirit of Camphor of the British Pharmacopoeia is a solution
of 1 ounce of camphor in 9 fluid ounces of 90 per cent alcohol (the
volume of the camphor when in solution being almost identical with
that when in the solid state).

It should have a specific gravity about 0*850, and should have an
optical rotation in a 200 mm. tube of + 7*7° to + 7*8°, equivalent to
about 11'7 per cent of camphor by weight. SchmatoUa (" Apoth.
Zeit." XVI. 290) proposes "the following method for determining the
camphor in spirit of camphor. Ten grms. are placed in a 50 c.c. burette,
and shaken with 30 c.c. of saturated solution of salt. When the cam-
phor has risen to the surface, exactly 1 c.c. of petroleum ether is poured
in and the camphor dissolved by careful agitation. The volume of the
petroleum layer is then read off. After subtracting the original 1 c.c.
each c.c. increase in the volume corresponds to 0-99 grm. of camphor.

The specific rotation of camphor in olive oil is a matter of import-
ance as it enables its rapid determination in Linimentum camphorcB
(camphorated oil) to be made.

Linimentum camphorce (camphorated oil) is a solution of 1 ounce
of camphor in 4 fluid ounces of olive oil. It therefore contains 21-45
per cent by weight of camphor. The apparent specific rotation of
camphor in olive oil varies from -f- 52° to + 55° according to concentra-
tion. Leonard and Smith (" Analyst," xxv. 202) give the following

684

FOOD AND DKUGS.

figures, after correcting for the slight optical activity of olive oil, which
varies from + 0-1° to + 0-2° for a 200 mm. tube :—

Percentage of camphor by

weight ....

5-32

11-26

20-66

26-78

Specific gravity at 15-5°

0-91903

0-92173

0-92911

Observed rotation (200 mm.)

+ 5-26°

+ 11-35°

+ 20-74°

-f 26-79

Angular rotation for 1 per

cent camphor .

0-964°

0-998°

0-996

Apparent specific rotation of

camphor ....

+ 52 •4°

+ 54°

+ 53-9°

+ 53-6°

It is clear, therefore, that each 1 per cent of camphor, when the
amount is over 10 per cent, produces practically 1" of rotation. Hence
a calculation is easy and the camphor can rapidly be determined.

Liversege (" Chemist and Druggist," lviii. 167) gives the following
formula for determining the amount of camphor per 100 grms. of
sample : —

P =

100 (L-o)

(C-o) (S + kF)

Where P is the percentage of camphor ; L the rotation in a 200
mm. tube ; C = 104 (that is twice the specific rotation of camphor in
olive oil) ; o is the rotation of the oil itself, if any, in a 200 mm. tube
(this may be taken as about + 0*2°) ; S is the specific gravity of olive
oil, say 0-915 ; and k is the increase in specific gravity produced by 1
per cent of camphor, which is about 0*0004.

The camphor may also be determined gravimetrically by evapor-
ation. Three to 5 grnis. are heated in a flat-bottomed capsule at
120° for two or three hours. The loss, after adding 0*15 per cent to
compensate for gain in weight of the oil, due to oxidation, may be
taken to be camphor. The oil itself can be examined for mineral or
nut oil, etc.

Chloroform.

Chloroform CHClg, is, when pure, a liquid of specific gravity
1-5020 at 15°, and boils at 60-8°. The British Pharmacopoeia, how-
ever, requires it to contain a little absolute alcohol in order to hinder
decompositioQ.

It is officially described as follows : —

Specific gravity 1-490 to 1-495. Boils between 60° and 62°. If
20 c.c. be allowed to evaporate from filter paper on a warm plate, no
foreign odour is perceptible at any stage of the evaporation. Water
shaken for five minutes with half its volume of chloroform is neutral
to litmus, and gives no colour with 1 c.c. of a 5 per cent solution of
cadmium iodide and a few drops of starch solution (absence of acid
and of free chlorine), and should not yield more than a very slight

CHLOROFORM. 685

opalescence with four drops of silver nitrate solution (absence of chlo-
rides). After shaking H.,S04 with 10 volumes of chloroform for twenty
minutes, and setting aside for fifteen minutes, both the acid and the
chloroform should be transparent and nearly colourless. Two c.c. of
the H2SO4 layer, diluted with 5 c.c. of water, should remain trans-
parent and almost colourless and should have a pleasant odour. On
further dilution with 10 c.c. of water, and the whole stirred with a
glass rod and four drops of silver nitrate solution added, the trans-
parency should only be slightly diminished. Water which has been
shaken with half its volume of chloroform, first treated as above with
H2SO4, should show a transparency which is only slightly diminished
with silver nitrate solution. It should evaporate without residue.

The purity of chloroform is a matter of the highest importance on
account of its use for anaesthetic purposes. Impurities may be present
as a result of faulty manufacture, or as the result of decomposition.

The following are products of decomposition : chloro-carbonic ether
C^HgCOjjCl ; carbon oxychloride COCl.^ ; possibly allylene dichloride
C3H4CI2 ; hydrochloric, hypochlorous, and, possibly, formic acids.

Ethylene dichloride CgH^Clg and ethyl chloride CgH^Cl may be
present as impurities.

The deliberate adulteration of chloroform is not common.

The proportion of alcohol in chloroform may be approximately
deduced from the specific gravity. According to Schacht (" Pharm.
Journ." [3], xxiii. 1005), the following figures are accurate: —

Pure ohloroform

1-5020 at 15°

„ with 0-25 per cent alcohol .

1-4977

.» » I) 0"5 ,, „ „

1-4939

„ 1-e „ „

1-4839

)> >> >) ^'0 ,, ,, J)

1-4705

Oudemanns (" Zeit. Anal. Chem." xi. 409) determines the alcohol,
by estimating the solubility of pure cinchonine in the chloroform. If
10 c.c. be well shaken for an hour at 17° with excess of dry cinchonine,
filtered, and 5 c.c. evaporated and the residue weighed, the following
amounts will be obtained : —

Pure chloroform

•021 grms

„ ,, with 1 per

cent alcohol

•067 „

„ 2 „

•111 „

M • „ ,,3 „

•152 „

„ 4 „

•190 „

„ 5 „

•226 „

„ 6 „

•260 „

„ 10 „

•346 „

Organic Impurities. — Brown prefers to detect impurities which
have an offensive odour by a more extended test than the simple eva-
poration of the chloroform, as required by the Pharmacopoeia. He
states (" Pharm. Journ." [3], xxii. 769) that by careful fractional dis-
tillation, and dividing the sample under examination into two frac-
tions, one of 10 per cent, the other of 75 per cent, and a residue of
15 per cent, both the more and less volatile impurities are obtained in

686 FOOD AND DRUGS.

a concentrated form, and thus betray themselves by their odour. The
non-volatile impurity may be determined by slow evaporation, with
precautions to exclude dust, at a temperature of about 90° F. This
process, which may be considered as an extension of the present test
of the British Pharmacopoeia, requires about 130 c.c. of the sample,
and several days' time for each experiment. In order to show this,
the following test, based on observations by Ramsay, who considered
that traces of carbonyl chloride in a sample of chloroform examined,
were responsible for the death of a patient, has been recommended as
being the best for detecting incipient decomposition in chloroform.
To 5 c.c. of the chloroform in a test-tube, 4 c.c. of perfectly clear, satu-
rated solution of barium hydroxide are added without agitation, and
the tube securely closed. It is then set aside in a dark place for six
hours, when no film should be found at the contact of the two liquids.

Brown (who originally suggested the cadmium iodide test in the
form of zinc iodide, however) at one time considered the baryta
water test the more delicate, but as the result of exhaustive investiga-
tions(" Pharm. Journ." [3], xxiii. 792) he considers the iodide test the
most reliable. He states that during the first stages of decomposition
a distinct reaction is obtained with zinc iodide and starch, but none
with baryta water, a separation of water being also observed. After
iurther decomposition, zinc iodide and starch give a more marked re-
liction than at first, and baryta water also reacts, but faintly. Still
following the decomposition, it is found that both reagents continue to
give marked reactions until a point is reached, when that produced by
zinc iodide and starch is observed to become less marked, and finally
to disappear altogether, while the reaction with baryta water may still
be obtained. A small quantity of deep straw-coloured liquid is also
observed at this stage floating on the surface of the chloroform. At
this point there remains a considerable quantity of undecomposed
chloroform, which may, either before or after separating the decom-
position products, be again put into an active state of decomposition
by simply removing the stopper from the bottle for a few seconds, re-
placing it, and again exposing it to sunlight, when reactions similar to
those already described with zinc iodide and starch are obtained. The
author thinks that results such as those described could not have been
obtained, if Professor Ramsay were correct in stating that carbonyl
chloride and hydrochloric acid are the only products obtained from
chloroform decomposing in the presence of air.

The following equations are given as a probable explanation of the
changes observed : —

4CHCI3 + 3O2 = 4COCI2 + 2H2O + 2CI2.

2C0C1., + 2H,0 = 2C0., + 4HC1.

2CHCi3 + 201., = 2CCI4 + 2HC).

6CHCI3 + 30, = 2C0C1., + 2CCI4 + 200, + 6HC1.

In harmony with this view, chlorine, water, and carbonyl chloride
are found in the early stages, the chlorine being first recognized, and
disappearing with the water at a more advanced stage, and the carbonyl

CHLOKOFORM. 687

chloride reaction being invariably obtained, not only in the early but
also in the most advanced stage met with.

The German Pharmacopoeia requires chloroform used for anaesthesia
to remain colourless for forty-eight hours when shaken with H.,SO^ ;
and also that if 20 c.c. be shaken with 15 c.c. of H2SO4 and 4 drops of
formaldehyde solution in a glass stoppered flask previously rinsed out
with sulphuric acid, the acid should remain colourless for half an hour.

The presence of aldehyde or acetone is indicated by boiling the
sample with aqueous potash, which is darkened if aldehyde or acetone
be present.

Chloroform made from pure ethyl alcohol is naturally easier to
purify than when made from methylated spirit or acetone. But by
very careful purification it can be made from the two last-named
sources to satisfy the requirements of the British Pharmacopoeia,
which now includes chloroform from any source so long as it is suffici-
ently pure. For purposes of comparison pure chloroform may be
made from chloral hydrate, or by crystallization with salicylide
(Anschutz's process).

Chloral Hydrate. — This body is a combination of trichloraldehyde
with water to form the crystalline trichlorethylidene glycol
CCI3 . CH(OH),.

The Pharmacopoeia requires it to be soluble in less than its weight
of water, alcohol, or ether, and in four times its weight of chloroform.
The aqueous solution is neutral or faintly acid to litmus. When
gantly melted it commences to solidify at 48-9°. It boils at 94'4'' to
96-7°. It leaves no residue on heating. If 4 grms. be heated with
30 c.c. of normal soda solution, not more than 6 c.c. of normal H2SO4
should be required to neutralize the alkali remaining after the reaction.
A solution in chloroform when shaken with HgSO^ imparts no colour
to the acid. If 1 grm. be warmed with 6 c.c. of water and 0*5 c.c. of
10 per cent KOH, the mixture filtered and iodine solution added
until the liquid is of a deep brown colour and the whole set aside for
an hour, no deposit of iodoform should take place (absence of chloral
alcoholate). Its aqueous solution should yield no precipitate with
silver nitrate solution (absence of free chlorides).

The official tests for chloral hydrate are not satisfactory. Many
good samples do not commence to solidify after melting until a much
lowor temperature than that given, the United States Pharmacopoeia
allowing a wide range of temperature. At the same time a low solidify-
ing point indicates excess of water, and a temperature of 47° should be
insisted on. The quantitative test should be carried out in the cold,
when accurate results are obtained, which is not the case when the
mixture is heated. Each c.c. of normal alkali used is equivalent to
0-1475 chloral, or 0*1665 gr. of chloral hydrate.

A useful method of examining chloral hydrate is that proposed by
K. Miiller. Twenty-five grms. are placed in a finely graduated tube
and a slight excess of strong caustic potash solution added. The tube
must be kept cool. After remaining for two hours, the liquid becomes
clear and separates into two layers, the lower layer being chloroform.
The temperature is adjusted to 17'. The number of c.c. multiplied by

688 FOOD AND DRUGS.

1-84 gives the weight of anhydrous chloral in the sample used, or by
2-064 the weight of chloral hydrate. Carefully carried out, this pro-
cess yields results which are accurate to within 0'5 per cent.

Frequently a slightly higher yield of chloral than the theoretical
(89-1 per cent chloral) is obtained, owing to the fact that this drug is
often made not quite fully hydrated, in order to avoid deliquescence, or
even, in hot weather, liquefaction.

r/;

Sii

t— I

•3

£3

>_3

-S

£5

« ^^

o 8 M

«^'-^ to

C ^ £ QJ

'111

■-Co
pus

'•> S S

« O 6C

« •« q

-^ « a

&» "3

o^- § o

O Os ^ X

^ Oi

-^H e S g
>-~! .2 b

c3 u

a.s

o ** b ^

'is

a 0}

* .
a

•a^qissituaad

•8{qiSSTUU8d

■w 5 o

.5 8.

C
_ O

.2 a

2-1 §

08 t:

.2 a

VOL.

689

0-^5 grm., dissolved quickly in water with 1-25 grms. of
NaHCO^ should tohen well shaken with 3 successively added
drops of HCl, decolorize 50-8 to 50-9 c.c. of decinormal iodine
solution (99-8 per cent As^Og).

Melting-point is for pure acid. Acid from gum benzoin may
melt at 120". Heated in closed vessel with CaCO.„ mass
dissolved in dilute HNO3 afid AgNO;j added, only slightest
cloudiness should result (absence of chlorobenzoic acid).
When warmed with its own weight of potassium perman-
ganate and 10 times its loeight of dilute sulphuric acid, it
should not develop odoiir of benzaldehyde {absence ofcimiamic
acid). 0-2 grm. in 10 c.c. of water does not decolorize 2 drops
of 1 per cent potassium permanganate at once.

When carefully heated to expel combined water it yields 66-4
per cetit of anhydride.

Specific gravity at melting-point, 1-060 to 1-066. Optically
inactive. One part of phenol, liquefied by the addition of
10 per cent of water, added to 1 volume of glycerin, forms
a clear liquid which is not rendered turbid with 3 volumes
more of water {absence of cresol).

improperly
purified

V 9

•axqissiuuad
•uoinira j'ad pcarj

«> § ^1

1 in 3

1 in 30
1 in 27

Solubility in

water at
ordinary t°.

1 in 100
1 in 400

1 in 30
lin25

1 1 ^ 1 Ijl

121-4°

not
beloiv

38-8°

Acid, arsenious
Acid, benzoic

Acid, boric
Acid, carbolic

690

§ o

« <3

Si' *9 •

»o -E! ■'• *i

-« e?

o S

rS; CO

" ^ a

^^•<^• O § s? O C

« I ^ ^ " S.i

a'^

. e ^ -^

r^ I * S

■^ i-i •'3 n3
*>• - £3 -►?

»^ -. S3 _g
t« S S o

|§ § ft

!Z5l2

§1

k. ^

«e, e»5

Z. t

S -i

CNJ «

«0 h4

&»»p

•5S

O 01

»0 00

Si c

■r(<

1 §

c
o

«o

0
_ o

■s^

8.2

^ o

-I

.sa
s

B:a

-a .2

•c

691

8 Ck

1 1

«1'

2 g^ ^

II «'S5

e-^8

2 ^ -

S 2S c
« s e .^

^ «' ^'^

■lg^^

•8 >

^ ^ ^ "TS o —

03 g § CO 0«S te

^ 2 ^
t|8

I'll

r«H i» ^ •«» DO

S " -S '^5 ^

&5 ^00

•5^:2 J 1 1

,?0

OkTJ

•8iqi8Stuuad

•aiqissiuuad
•noinini Jed pcai

_ o

.a s.

^ o

.2 a

in^-

^ o

a
^ o

^ g

Solubilit

water

ordinary

"oS "S

•3'^

is-s

*■■£

9 S*

.2 a

.2 84

692

I! IJ if 1 iiliiiil Is^ llill II

ii i^ll ;;i siillfl!^ ilS^- ill I si

c 3
.2 ^

"g,

»o

« §

■*

* -i eg

2 g - §-

^1! I os-S

eg,' a §,

■'2 '2

&4 a, Oi

— .2 3 — ' .2 — .2 — .2 —'2

.s s^ s ■ .s a c a .s a .s a

o c o o o o

! ! 1 1 I

1 1^1 1 ■ II 1

a -2 .2 .2

6 I o 5 ^ § ■§

J3 O

-<i

693

c —

-1

5
1

-2

O c3

§•5

•3

8 «0 •

CO-

5 ^
•2 .2

o o „• S

S a

«1
li

B 00

>-(

w :::s

«»

.8S

1-2

-1

§^

rs: ft

?i ^

-:s

« s

Oi'l

o8§

III

1
1

"ft

52;

3 =

n3
1

tsO

•8iqiS8IUIJ9d

•aotiiiai aad SQS^y

i§

•atqiseinijad

•aoTHira jad psyj

.S*i

—

o .

^io

— .2

_ o

11.1

<=" -^

•^1

.aa

.2 a

1^*

iH ^

'N ""

2
ft

5s t

^ -H

5 ® 55

.2.S

-» 8

(M '-(

DC O

l"o^

J«

05 H* ex.

CLi

—

■ia

„

TS"

-<ji

•<!

694

^H,^

■5;

•SI

::^g

rS -«;

5 CO

•"^

g ?J

IS "i 1 I £ s ir .si's- |&--«||
|>^ |l :l III |4 i^S Ills I.

2 W

S| I I

$ 2 M

1 *

"Ttl

«D

s
o

«C

I 51 I ,1

r^ iH

.2 a

g a 8 g Si a

s §

2 a«. a SoSa©

« §^§o 3-i2r2 3"S

a§ a c § -3:2 -a g g? -a^

11 I ii ii il 11 1?

695

fr.

•^

t>^

CO •»-■ . m -ka

■2.5

fume
ms. 0
Igrm
exces

abou

^
e

t &a ^c-S

^-i.^

h HNO.j until no r
eld 1-7' to to 1-8
I rarely obtained),
poured into a lari
d dried should wei

-S3

;25

1
I

nd warmed witl
hen ignited, yi
is 2 grm., but is
1 and the liquid
litate washed an

■I

■1

jj ^ « o »

II «

5 grm. dissolved
and then mixed \

jQ iodine before
100 per cent Sb^'

grms., moistened
are evolved and
oxide (official tigui
be dissolved in H
of water, the preci
0-85 grm.

arm. dissolved in

60-2 to 60-7 c.c. Tq

manent colour {■-
that the Pharma
water !

'-I

£1

•aiqissituaad

0) C

»H

^ s- -S

•uoniiui lad «o''SV

6 E ^,

•aiqissiuijad

•uoiI|iui 196. pBa^

.2 -^

SI'S

IS'

9
1

ilubility in
water at
dinary t«.

6..

S s

^•

O 3

S -C3

"£

•^ a

•5

•r*

fl 'B

-13

<j 55

<tl

696

5* ir

fl
o

SO
OS

fl
1

oxycarbonate)

-2

■0

|i
If

is
il

:<5

III

e>3

i
<

;si
1

'4<
1

S
0

"fl

is
11

^11

-2

03 *!

2
2

§> CO

i fl

Ol §

1
1

s i «

--^

« s

s a

f-i o

'o '3

^ 06

O 60

OQ U

<x>

_fl

r-i

—

• 1

—

1
CD

—

•tS fl

Ic a

'"^1

= -§

2 0

»§

'4S

•43 .2p

§1

fl
1

'fl
0)
OQ

s s

<tj

»4

697

+a . -w

t*-

OJ X ^" »-i

a
S

isl

S-l

^.^e

sr-i^

^ ^S

.-;•-" {"-its

=11 s

^■^•1,

•is

i.a

,5»

i!3

o
a

n3
1

S OQ

1"°!

^ So

.2
'■B

1
£

3
&<
O

ij .Ti 04 t-

Is J

2 t^'^cb

§ 0

Ki-
ll

""I

•^ a

o §

3 .„ O 00
O o ^

as.

*" CO

8 ''*

'»K

■^

•-I

fl I. a a

"-^

«i

igij

* 8z.2

•aiqissiuijad
•uoiiiiui jad ^o^^SV

■*

«D

•8|qi8SXIUI8d

•iioiniui aad pBa^

5 pointed out that th(
BiNO, . H2O, which
d to the formula (Bi
anufactured by Eng]

Ttt

' 'i'l-i

III

-2

g
a

|5S|

i^=3 P fl

^ s s 1

■t-9

5j2Sa

?W to eg

(-!

^
••-•

•■2 2 «

•2 IS

.2 6-^.

« §

ta a 00 fl

1 8;:^ §

698

-2:2

as .12

v^ O

^ eg

"^ o

*» ii « -*f .— J3 'JS C

ti'TS ^ '-' se »i o — S
-■S 212 bc§ -r-^ S

=^ ^ ^,.S -S -2 >;. O-S
— »-i c _ « >-

- -f!; § S ;: " -eg

2 '« J iJ 'c s

"^■^ ajgCg^S

Its O'

S 3
O O

;2 0so
goo

I 2 . -

111 ^S-^^-

I »= « oTJ °g'^

o S S e^ ce ® § i^

^ ►S &<

o :3

:t5 oe

2 8

§ i

^ w c o
-« go/;'

a.
S? .§ .-^ -

« c 2

S "S 2 02

-^ S:S §

5'2 ^ g

s» § 2 a

1 1

«3

a
0

I I

I 1 I

I I

I I 1

^ eg *S

pq eg

eS ^
O

699

ID '^

•-3 ^

X o ■^

(D CO g

-^ Oi o

-2 ^

« s

^ a

O ~

^ n-i O
1^ o «

si 3 ^

.«n3 «

oo OB

•^ ^* -2 J*

s o So

2 « S o

O Si « fl

^ s 3 "^

?- o 0-3

^t2 CO '^

«N ^ ^ W

fO o a> -w
CO

•J. g e «
?: ^ 5 =^

«0 ,,_, 05 s

go ^^
00 s s fe

O) 'T3 rg 00

-«'H -3 5

^ ° -'^

a c

1 :5 o o)

2 «

on e

'aiqissiuijad

•a^qissinuad

»^

•a

(£

Si
0

3
0

700

:2 S ^

.2 •? c

o ?: I o

S - i S

5 »'Sa?

w K O

t3 ,

o to^ ■

b''^ '^

'b o

i sit

fc- SI'S £1,

J-1 C oo »

£ B "^ e

-111"

III I

=e .-2 « §
3 *2 CO -"^ g

o o « S e

® § is

c8 00 II

& CO

=2 <4j S

O i, o

So ^

c8 S ^

»-i " K '—

^ ^■'< o

X to OS i-*

'So®

oO-l
J-1 ^ «

ce g s

S ** §

^ ^ CO

2 § «

o j;

So" 02
'^TS II

0*0 0
CO J. -«

S CMC

*• « e

o S
il

-3 §

9 3

XI S3

■El

•3

3 OS

701

'§

with
ury.
rous

>-<

'"1

1
1

fi
is

SIS

2-1

.2
.1

s ^

si
II

"^ « -J

S g s.

III

1 s

ill

3
t

Ills

2
"2

1
1

8
1

1-^
Hi

2 S

£1

•9tqiRciraj[9d

•uoinra aad *'0^gv

•eiqissiraiad

•nonn™ •'^^ r^^T

S-io'

«l

IS"

bO o

•2 1

lability
vater at
rlinary t

1-1
C

r-(

3:2

.2 S<

-S3
o s

a-o

ri c

p»

•

el-

.-H S

I
^

1
1

^ 1

a
■c

702

^3 5

IS.

Ij — « s

^- « ^ G

^ S» i-H 33
03

CO 5.

f-'^

-^"^

S'c.

^'t>

'^ ?>i

«»

s s

n-(

■^ g

-«s

^ '<-> Qi 9i

S2^i

!$ ^ o a
s «* o

S tn oe .t^

« -, «S >H

s « ^

c^ i — '^

'^.« ° a S

•s ° 0 a §

K CO

C .^ -5

2 S t^

'^ ft

*» ""Sir

« -2

^ S So S.

« o

5S 6C oj

S <s i

OB g »

as *; 05

i s ^ ^

6C ^

I I

0 q s
.2'S ^

1 1 1 1 III ^

1 1 1 1 III 0

1 1 1 1 '11- c

till 0

g a

•S.S o

o "o

•Of

8 a

I I

g|?

I I

■I I

6C CO &C S

^:5 ^2

go 08 o

a g a

td as

703

g a

^
«

„ o

!i.

« 2

o ^

K-^

-S'^

'^1

CO a

■p-^"

i^w"

"^^

«l

Principal Tests.

1 *z

II!

.11

lla

^'1

•is

® S o

a
1

fell

»««

i§

ig'

|gg

r«

-1, "^ ^
5^

0) 0

ora

£|

•aiqissTuijad

•uoiiitui jad 80'^v

•eiqissiuiiad

•uoniira Jad pwi

= -S

— a .

-H .2
08 "E

— .2

_ o

.3 $8

'" o

— o

9 ^

.S ^

c a

«s _o:

Solublo i
water at
rdinary 1

— .2
"3 -£

„ o

=i|

— .2

— .2
15 "C

■-I

s a

aa
s

•i

§§

§1

lis

3 C CO

^^1

S'l

eS a

cr c

.2^-S

J-

704

•J

'X3

precipitate
contain at
s mercuric

«5

d
■^

-2

1S?^

-3

^•S

-1-^

'P.

•9 s

-4

2
0

1-^

ll
11.

■3

•2

'S.

3 ^

c
0

I0"

2 ^:^

S
1

0
d

^k^

r^w

^ ^

io^

111

„ caco

s-1

>-4

"S CO

ii
Ig

'^0

la
"3

g
2 W

-w

1.^

^^^

1.0

CV}

1 .

u '"'

—

— < .2

"oi

't.

— « .2
Is 15

= 8,

== n,

«8,

•-" o

•~

'"'

^ 0

_ 0

^'•c

"3 ■£

^•^

is'-c

•^

■3 "S

=i-^

0 8L

C P4

a B4

-^ 2

■"

....

"^2

ft
1

.2 «3

•s

»3

s
cr

§1

2
1^

1-1

d<a

»-5

J "=

iJ c

VOL.

705

e ^ ^ S

^ g i. ec -!i

2 -^ ^ St ^

■» «" e ^

-^ ^ Sft"^ "-< "^-^

C> "^ rs. **

^ g ^ §

'e « <= 5;

^ g « ^

So S

-«• °

131

o « s

S-a

o ^

< .
S

r-J t^'S

O CO -S
96 o.

sr?

^O o

II
II

•a^qissuujad

•aiqissioijad
•aoniiin jad p^aq;

-^^
s

I I

'oo

— o

ill

-•■e

d
— .2

'a'Z

•" o

ll
^1

1
o

'SI

q. plumbi
bacetatis
fortis

■a'i

706

CO
•^ 00

s ^

K C5

« II '«

:%i

g ^ =c JD
►N V> - ft

•"S oi

O W5

a "^ ^ -*=
•?«~ H

a icS
'i So

a|^
oil

C<1 t3 ."tS
O « ft

'S
<v

o »*^
« fl Ss

8|^

.-( ? as
o ci a

O 0) 03

• CO

-- a-
febg a

O a a8 ^*

o a 0) o

2 .s 2 ^ a

5 3 O

^ X * c S

-* .i 3 2-2

|a£-r

<!- ,

o3 a>

5^-1 o« o

O § 1» o .2

^ * o 2
§00 « or

-s'a •'^^

-S: •§ 6 •- -2 "3
S^ . > £ "^

»* U o 03 «i

= "211

•o ^.-^ — g

^ I « ^ o'

.J*

0 "^"^ q
._ ^ CO . n

. " ^ o bo
^-^ ton

^ • STS 3 t^

?. ^ ^"3 ° i
a

Kg II '^ .t3

■^ J» +a S

O «

"Id 5 a ID 0

^ o

25
«a a

-H .2

aa
2
&4

aS hi

_ _o

O M

2*—
13 2

■■" o

707

>> s

'vns

rQ W J>. QJ w

IH .1-1

» JO eg

:i^

0 a

C5 -3

1 s

1°

3 to

g'S

9 o

• CD
«5^

■75

■J 5

1.
0^

§1

;S5 §

■V

M
§

^^-d

-« s

® § c

^
«

-s-3

8,-?

*5 i5

00.2

->a

0--

«5-2

a*-^

«o

"oii

feb-S

•aiqissiuuad

•uoiiiixn aad ^'O^SV

•9iqissiuiJ8d

•tioii[iui iad p«8'j

ii-B

r-l

3 « 8

s
'6

— _, e«

^§

■"

•s^^

,~ <=«

-;!^

'*< i-i

"S^

1^1

..-1

-^o

c3 2^

I— I cc

Magnesia lev

Magnesia pond

rosa

gnesii ca
onas levis
gnesii ca
as pondero

oS ^ c8 fl

S ^o

a rv

■>? 'W «?

708

^ CO O
^ *~ -=3

o.-| o
^e g

>j^ o

^ s «

w -^ j.
••M Si ^~»

0> OD 00 00

s^^o o .2
« « c; >i

-2 TS o ^

- ° S "
•-2 « -3 -

^ ^ C/2 -3

• s^-S £

1 - o £ -S
00

.§---

^ ^ 6 c .£ ^

• — Si -ri o

^s-^^i^li

«j -S: ^ '^ -22

•g . S o O « s-

• o J u 173 Tj "^
S ^ . <a 2 fl
^ CO 00 >-t o 3

« 8

s 9 ^

"^ § s «5 .2 c -c

2 o g ^ o a =c

'^*'_§ «

O ii ^ 00

cc ^ aS ^

t -3 J

*«- « o -iS

^^« a

'-s'« S §

"^ « sf g

^ • - - -s s

. g c »^ 2

* -c: ^

t^ -5 -is 3

*= <» b
^ -Si v5
Si -iis "^

*>> 08

2 t

I ^

05 i-i '^ s5

J », S 8

** ^ «- o

^ o

1«

8 So .S

■^ Xi ^

'-H * 08

--^ SO

.§ fl

►5 fl

S P^

709

S ..13

'^ s s 2
--lis.

3^-2 §

-^ r J IS

® S fl §
f *S .2 ^

•-^ CD

^ a

OS «S

O 53 n3

O) (U M

5 oS "^
» hrt »

^g
O o

«^

11-2

a I

Q ^ 01

■<> ^ O)

1 .2 -5 c5
8 2-0

« 08 -in

►S CO t3 S

^ 'S DO 3

K) IN

-is «§
.. « »-

f^ <» OS

-» ^ I

lis

•8]qi8SlUIJ9d

•uoiiiira J9d ^O^V

•9{qissiraj9d

III

s a

II r

2 «Oeq

8 1-2

710

Oh O

3 a

(^ o

o eu

1 a

^ II

•M o

o -^

S, OS

•5 o
8 «

■gSa

e^^-^
^>g

§'5c

';s -2 s ^

'^^b S-&

^ §

It ^

— > 'C ^

c-n o

go «

^S s
I « S

• S O 00

111

^ o2
o «

^ « ^

S « o
w «'©

o< « to

TS O

.2 ^ >»

2 <^ »

C '^ -w

O.Si.««

05 S K

"^^^

O w oj

W o >

? H £:

o"5 S

"^8 £
!^ s a

04 ca

5 5

S 8 *

* a

o
o as

01 -5-.

£j

o «

^ s

3.2

08 O

all*

I I

05 .0

I I

.5 a 8

8 a

^ 8

I I

1 1

I . I

._ oS
Du,.2

5-s

I a
5|

;3 X!

;2S

O J

3 -3

711

■1 "^

CO ~

o ^

e.§H

ti.l
lit

:S *• a

'o — 'o
. C « M

o «^

OD I o

fee*

•9tqi8STUU9d

•ajqissxraigd

.-822

712

§1

45 o

C a

g^^

-1

CO V

* Til *

. fi OB

§ t J

s^ o o

S= 5 ^ a>^
i» V. ^ -o £ 2

Cft Oi O cS S J3
Cr> S» ^ to 05 C

"1 o 'S S

S § ?: .

J 3 i £

-I< o ;? ^

CO c3 QQ

11^^^

^^is

o ^

2 -e' S =

CO r-- C?^^

2 S

go „ ,_

S K c o

0 0-2-2^
^ «■ 5 a "^

K •= o O ^
. . S o 5

OiQ O « 8

^«^ §•§ I
« II ^ .2 ^

t> 2te*o

§ S «
^§1

^ 0)

-3 a

8 a

cn a
2

^
g

713

e o s

^11 =

.2 f« -2 -^ « oj

fc "» 2, « K . — ® •

^s 2 « S g^g g

^ o

•^ o

8 g

O "

S S o g >?^
Is to •«

so
^^

a to

o .12
o ^

2 c
o o

O • !3

to ^

a! <5

S#'S| .^ * §

«^ w _, s

s e

5 ^i g

S 2 c
s *2 .2

^ CO ^

^8 g 8 o
02:2 -2T3

«, V5 -« CC O
■" ,^ -W —

.^ o e o .2

^ I ^ -^ g B
eg §1 o|

S'? 1^1 1^

o h 00 -« .t»

i e § § § S

i^.g V o ® ^

•9|qiP8nui9d I 5o o

•-5 ^

I I

oS O
§ 2

1
02

•a

714

ill

CO C5 g

^ o a
.'s.s

?? « cc
« . Pn

^ O o

1"l

<J0

-«:

>*>

£•

8 S OS
•» f« CQ

^'2

S'®

■5: os

. 02

g o -x » ■*»

*B ii Cj CO

.Ills

i-.S^

^ s " *

«■ 00 » a

g <s> o H

8.5

to O >= i.

»5 l^ ^^n3 ""
"^ ^ 8 O O
§ -S? ^ 02 ai

e 8 S-

11 « .S ^ »
& ^ w s a

5^ ^

"Ho

•§ *l

1 = i "^ ^

a g e ce -;:

^ S « te '«

§ 8 E S'g'a

^ I « c .2 a

.g -5 g O^ sS

il = s ^^

;^ rs *s to fc» 2

^ <^. V. 4a aiH m

-5 3

1 1 1 1 1

440°

1 1 1 1 ■ 1

about
11.5°

eS
3

a
o

716

tiD?C .

6 S^

dm',
^ O II

1^2

13 i*

-S .^ fl

a 00 >■

o *'
o

-« rr.

«^ o hrt

«^ r"

§03 O

g *" <u

g o >

r2 g .2

£ o

O CT'

g N o

fee :.

DO O

o O

o «>• g •»»

■= s i a.

-sS o"

3P §

g c

■^ -s '^ ^

•J . - §«
O

1 I

•aiqtssiuijgd ,

•aiqissiraaad
•noi]jira jad p^aq;

O i=^0

^-o

1 1 1 1 1 1 II

M. Pt.

51°

716

INDEX.

Abel's flash-point (of turpentine), 623.

Abies balsamea, 476.

Abraham on concentrated liquors, 497.

Absolute alcohol, 273.

Acacia, 428, 430.

— as an adulterant, 433.

— catechu, 444.

— cutch, 444.

— Senegal, 430.

— sieberiana, 432.
Acarus domesticus, 83.
Acetal in brandy, 288.
Acetanilide in vanillin, 266, 269, 270.

— properties of, 689.

— test for. 709. 712.
Acet-eugenol, 266.
Acetic acid, as constituent of oils, 235,

446.

in cider, 351.

in vinegar, 248, 250, 255.

microscopic reagent, 424.

pharmacopoeial, 689.

— aether, 694.

— aldehyde in brandy, 288.

— esters in rum, 304.
Acetin in spermaceti, 645.
Acet-iso-eugenol, 267.
Acetone in oil of tea, 2.

— iodoform, reaction of, 274.
Acetum ipecacuanhse, 566.
Acetylated resin, value of guiacum, 458.
Acetyl-chlorohydrose, 117.

— palmitate, 645.
-vanillin. 266.

Achorium intybus, 31.
Achroo-dextrin, 149.
Acidum aceticum, 424, 689.

— boricum, 654, 690.

— carbolic, 690.

— citricum, 649, 654, 691.

— gallicum, 691.

— hydrobromicum, 691.

— hydrochloricum, 655, 691.

— hydrocyanic, 692.

— lacticum, 692.

— nitricum. 655, 692.

— nitro-hydrochloric, 692.

— oleic, 693.

— phosphoric, 693.

— salicylicum, 658.

Acidum sulphuricum, 655, 6J3.

— sulphurosum, 694.

— tartaricum, 654, 694.

— tannicum, 694.
Aconine, 506.
Aconite, 506.
Aconitine, 507.
Aconitum napellus, 506.
Acorns in coffee, 32.
Acrinyl-isothiocyanate, 24.

— thiocyanate, 214.
Adams' method in milk analysis, 47.

— process for condensed milk, 74.
Adenine, 401,410.
Adeps lanae, 632, 658.
Adulteration of almond oil (essential).

258.
(fixed), 633.

— of annatto, 246.

— of balsam of Peru, 439.

— of beeswax, 642.

— of bread, 180.

— of Canada balsam, 477.

— of catechu, 444.

— of cheese, 85.

— of chemicals, 689.

— of chloroform, 685.

— of cider, 857.

— of cinnamon, 229.

— of cloves, 222.

— of cochineal, 239.

— of cocoa, 19.

— of cocoa butter, 27.

— of cod liver oil, 638.

— of coffee, 31.

— of copaiba, 448, 451.

— of cornflour, 182.

— of fatty oils, 632.
~ of flour, 177.

— of honey, 150.

— of ipecacuanha, 560.

— of lard, 108.

— of linseed oil, 636.

— of mace, 236.

— of milk, 58.

— of mustard, 216.

— of nutmegs, 232.

— of oatmeal, 182.

— of oil of lemons, 261.

— of olive oil, 117.
I — of opium, 581.
j — of paprika, 213.

717

718

INDEX.

Adulteration of pastes, 184.

— of pepper, 203, 204.

— of saccharin, 673.

— of saffron, 240.

— of scammony, 481.

— of spermaceti, 645.

— of storax, 493.

— of sugar, 142.

— of tea, 1.
Allen, 7.

— leaves used for, 7.

Wanklyn, 7.

— of turmeric, 245.

— of turpentine, 623.

— of vanilla, 265.

— of vinegar, 250.

— of wine, 311.
Aether, 694.
African copaiba, 448.
Agrostemma githago, 178.
Albaspidin, 469.
Albumen (in milk), 58.

— pepsine testing by, 472.

— wine clarification by, 311, 312.
Alburaenoid nitrogen in cocoa, 20.
Albuinenoids, 339.

— in bread, 174.

Albumoses in meat extract, 397, 406.
Alcock on jalap, 572.
Alcohol, 273.

— determination of, 283, 291.

— glycerine ratio in wine, 336.
Alcoholic beverages, 273.
Alcohol in beer, 345.

— in cider, 424.

— in wine, 813, 314, 315, 336.

— table, 275, 276, 277, 278, 279.
Aldehydes in alcohol, 295.
Aldehyde, determination of, in alcohol,

291.

— in brandy, 2S9, 290, 291.

— in nitrous ether, 487.

— in rum, 305, 306.

— in whisky. 301.

— in wine, 329.

— properties of, 709.

— sugars, 121, 122.
Ale, 337.

Aleppo scammony, 481.
Aleurone grains, 418, 419.
Alexandrian senna, 483, 484.
Alicante saffron, 240. ,

Alkali in cocoa, 24.
Alkalinity of ash in tea, 6.
Alkaloid in pepper, 203.
Alkaloids, estimation of, 504.

— in beer, 347.
Alkanet, tincture of, 424.

Allen & Scott Smith on opium prepara-
tions, 589.

Allen & Stokes on analysis of milk, 46.

Allen-Marquardt's process for alcohol,
297, 299, 309.

— on altered milk, 77.

Allen- Marquardt on analysis of coffee,
36.

— on beer, 348.

— on bread, . 80.

— on caffeine, 38, 39.

— on cheese analysis. 85.

— on cider, 349, 350, 351, 353, 354, 355.

— on factitious coffees, 33.

— on ginger, 193.

— on ipecacuanha, 567.

— on meat analysis, 374.

— on meat extract, 404.

— on milk analyses, 53.

— on pepsine, 473, 476.

— on saccharin, 675.

— on specific gravity of coffee, 34.

— on strychnine, 579.

— on sugar, 124.

— on turpentine, 623.

— on whey, 77.

Allen's method of preparing sugar for

polarimeter, 131.
Allspice, 224, 227, 619.

— microscopic examination of, 227,

228.
Allyl disulphide, 438.

— isothiocyanate, 217, 218, 219, 220.

— sulphide, 438.

— thiocarbamide, 217.

— thio-urea, 218, 219.
Almond essence, 257.

— oil, 633.

adulteration of essential, 258.

analysis of, 635.

— shells in pepper, 205, 206.
Aloe barbadensis, 507.

— chinensis, 507.

— vera, 507.
Aloe-emodin, 511.
Aloeresinotannol, 508.
Aloes, 347, 348, 507.

— detection of, 510.

— in beer, 512.
Aloin, 508, 512.
Altered milk, 77.
Alum, 695.
Alum-carmine, 425.
Alumina, 189.

Alum in flour. 179, 180.

Amagat & Jean's oleo-refractometer, 97.

Amber syrup, 147.

American annatto, 246.

— frankincense, 480.

— storax, 492.

— turpentine, 478, 622.

— whisky, 303.

Amides, nitrogen as, in cheese, 87.
Amine bases in brandy, 288.
Amines of fatty series, 385, et seq.
Ammonia, 697.

— in meat extract, 406.
Ammoniacal nitrogen in meat extract,

404.
Ammoniacum, 428, 434.

INDEX.

719

Ammoniated tincture of ergot, 550.
of opium, 588.

— — of quinine, 534.
Ammonii citratis, 660.
Ammonium carbonate, 484, 695.

— compounds, 695.
Ammoresinotannol, 435.
Amthor on vinegar, 252.
Amygdalin, 257, 258.
Amygdalus communis, 256.

Amyl alcohol, 286, 288, 297, 299, 3';7

330, 331, 501.
Amylamine, 359.
Amyl-methyl-ketone, 225.
Amyl nitrite, 695.
Amylodextrin, 150.
Analysis of alcohol, 274 et seq.

— of annatto, 246, 247.

— of barley malt, 338.

— of beer, 342.

— of butter, 90.

— of butter fat, 95.

— of cheese, 84, 8'.

— of cider, 349, 35 ).

— of cider vinegar, 255.

— of cinnamon, 229.

— of cocoa, 16, 20.

— of coffee, 34.

— of condensed milk, 73.

— of Demerara rum, 306

— of ginger, 194

— of Jamaica rum, 805.

— of mace, 237.

— of meat extract, 401, 407.

— of meats, 370, 371.

— of milk, 46.

— of mustard, 216.

— of nutmegs, 233.

— of olive oil, 112.

— of opium, 582.

— of paprika, 213.

— of pastes, 184.

— of pepper, 199, 200.

— of suet, 111.

— of turmeric, 244.

— of vanilla, 266, 270.

— of vinegar, 249.

— of whisky, 303.

— of wine, 313, 314, 317, 318, 319, 320,

321, 324, 325, 383.
Angelic acid, 608.
Aniline blue in rice, 182.

— colours in wine, 381. '

— dyes, 246.

— yellow in butter, 93.
in rice. 183.

— orange in milk, 71.
Animal alkaloids. 358.
Aniseed, 428.

— oil, 606.
Annatto, 184, 246.

— adulteration of, 246.

— analysis of, 246, 247.

— in butter, 91, 93.

Annatto in cream, 70, 71.

— microscopic examination of, 247.

— tests for, 71.
AnatoUau liquorice. 466, 467.
Anethum sowa, 428, 606.

— oil, 606. »
AnUiemis nobilis, 609.
.Anthra-glucosides, 599.
Anthraquinone, 598.

— reaction of aloes, 511.
Antimonii oxidum, 659, 696.
Antimonium nigrum purificatum, 659,

661.

— sulphuratum, 659, 661, 696.

— tartratum, 659, 696.
Antimony, compounds of, 696.
Antrich on cocaine, 539.
Apomorphine hydrochloride, 594.
Apparatus for microscopical analysis,

416.

— in wine analysis. 326.
Apple juice, 217, 349, 350.
Apricot kernel oil, 634.

(essential), 257.

Arabic acid, 430.
Arabin, 431.
Arabinose, 23, 433, 470.
Arachidic acid in butter, 89.
Arachis, 112, 113.

— oil, 108, 109, 116, 682.
Araroba, •i36, 511.
Armagnac brandy, 287.
Armstrong on terebene, 623.
Arndt on ipecacuanha, 561.

Arnold & Mentzel on wine analysis, 332.
Aromatic oxygenated bases, 360.

— ptomaine, 35'J, 366.

— spirit of ammonia. 235, 484.
Arrowroot as adulterant in cocoa. 69.

— starch, 170.
Arsenates, tests for, 649.
Arsenic, 636.

— in malt, 840.

— tests for, 648, 650.
Arsenii iodidum. 697.
Arsenious acid, 690.
Arsenites, tests for, 649.
Artificial benzaldehyde, 258.

— cognac oil, 2S8. 2S9.

— mustard oil, 218.
Asafcetida, 435, 437.
Asaresinotannol. 437.
Asbestos in drying of milk, 46.
Ash in coffee, 34.

— composition, cane and beet sugar,

Monier's analysis. 137.

— determination, commercial sugar,

136.

— standards of drugs, 428.
Ashby on vinegar, 251.
Asjjergillus glaucus, 83.
Aspidinol, 469.

Aspidium filix-mas, 429, 468.
Aspinall on rum, 304.

720

INDEX.

Astragalus gummifer, 432.
Atropa belladonna, 513.
Atropine, 515, 520, 602.
Attenuation of the wort, 344.
Austrian oil of turpentine, 622.
Avi-Lallemant on butter analysis, 95.
Avena saliva, 169.
Azo dyes to colour milk, 70, 71.
— yellow, 184.

Babcock's process in fat separation of

milk, 51.
Bacillus acidi lactici, 80.
paralactici, 80.

— coli communis, 81.
Bacteria in milk, 79.
Bacterium caucasinm, 80.

— lactis, 79.

Badouin's test for olive oil, 116.

Bahia copaiba, 448.

Baier & Neumann's analysis of choco-
late, 29.

Baker & Hutton's process for condensed
milk, 75.

Baking powders, 187.

Balsam of copaiba, 447.

— of Peru, 438.

adulterants of, 439.

— of Tolu, 439, 442.
Balsamodendron myrrha, 470.
Banda mace, 234, 236.

Banana as adulterant in cocoa, 19.

— starch, 171.
Barbadoes aloes, 510.
Barbaloin, 508, 512.
Barclay on belladonna, 518.

— on cod liver oil, 639.
Barger on ergot, 550.
Barium glycyrrhizate, 463.
Barium salts in cayenne. 209.

— sulphate in bread, 186.

in pepper, 204.

in saffron, 241.

Barker & Russell on cider, 352, .353.
Barley, 187.

— as adulterant in cocoa, 19.

— in coffee, 33.

— flour, 172.

— malt, 337.

— meal in oatmeal, 182.

— starch, 168.

— sugar, 117.
Barrows on flesh, 358.
Bartley on pepsine, 474.
Bassoric acid, 433.
Bassorin, 433, 435.
Bates' saccharometer, 339.

Bauman on flour, 179. |

Bavaria, restrictions on malt liquors,

337.
Bdellium, 470, 471. j

Beans, 187.

Bean starch, 171.
Becchi, test for olive oil, 115, 116.
Beckmann on alcohol analysis, 299.
Beckurts on belladonna. 515.

— on nux vomica, 575.
Bedford & Jenks on alcohol, 299.
Beechwood tar, 478.

Beef suet. 111.
Beer, 337, 341.

— preservatives in, 348.
Beeswax, 641.

Beetles in drugs, 423.

Beetroot sugar, 135.

Beet sugar molasses, 147, 304.

Beimling's centrifugal apparatus, 51.

Bell on alcohol analysis, 297.

— on turmeric, 243.

Bell's process, milk analysis, 51.
Belladonna, 565

— microscopic examination of, 513,.

514.

— ointment, 519.

— plaster, 520.

— root, 427, 513.

Benitvoglio method for paste analysis,

185.
Bennet on lead tests, 670.
Benzaldehyde, 257.
Benzene, 512.

Benzoic acid, 69, 253, 266, 266, 267, 383,
440, 441, 690.

in saccharin, 676.

use of, 385.

Benzoic esters, 440.

Benzoin, 440.

Benzol, 697.

Benzoyl-chloride test for alcohol, 279.

Benzoyl-eugenol, 226.

Benzyl alcohol, 439.

— benzoate, 440.

— cinnamate, 439, 440.

Bernard Dyer, glucose syrup formulae,
146.

Berthelot's test for alcohol, 279.

Berberine, 551, 554.

Betaine, 359, 361.

Bhang, 442.

Bianchi & Di Nola on saccharin, 675.

Bieber's test for almond oil, 634.

Bigelow & Cook on meat extract, 407,
403.

Bigelow & McElroy's analysis of con-
densed m.lk, 75.

Bird on aromatic spirit of ammonia,
484.

— on belladonna, 515. 517.

— on chemical tests, 662.

— on concentrated liquors, 497.

— on ipecacuanha, 564.

— on nux vumica, 575.

— on opium preparations, 588.
Bird's apparatus for ipecacuanha, 565.

— arsenic apparatus, 664.
Bisabol myrrh, 47 , 471, 472.

INDEX.

721

Bismarck brown as microscopical re-
agent, 424.
Bismuth compounds, 697, 698.

— carbouas, 659.

— oxidum, G59.

— salicylas, 660.

— subnitras, 660.
Bitter almond oil, 257.
Bixa orellana, 71, 246.
Bixin, 246.

Blackberry leaf in tea, 15.
Black catechu, 444.

— grain cochineal, 238.

— mustard. 214, 215.

— pepper, 198, 199, 200, 201.

— tea, 1.

Blarez on wine analysis, 336.
Bleached flours, 181.
Bleaching in microscopic analysis, 420.
Blichfeldt's process for butter, 102.
Blown tins (preserved meats), 372.
Blyth on cayenne pepper, 209.

— on milk preservatives, 60.
Board of Agriculture on butter, 88.

— of Agriculture's regulations re milk,

41.
Bock, 341, 342.
Bodmer, Leonard & Smith's composition

of golden syrup, 144.
' examination of golden syrup,

146, 147.
Boheic acid, •!.
Bombay mace, 236, 237.
Bomer on lard, 109.

— on meat extract, 406, 407.
Booth on analysis of chocolate, 21.
Booth's analysis of cocoa, 17.
Borates in cider, 349.

— in meat extract, 413.

— in milk, 61.
Borax, 5, 61, 378, 654.

— in cider, 355.

— packed hams, 379.

— test for tragacanth, 434.
Bordas' process in wine analysis, 328.
Bordeaux wine, 32J.

Borgherio on rice, 183.
Boric acid, 61, 384, 654, 690.

— — as preservative, 380.

detection of, in cider, 353.

in meat extracts, 413.

in beer, 349.

in cider, .352, 353.

in cream, 67, 68.

in milk, 60, 61.

Borntrjiger on aloes, 510.

Boseley, analysis of marmalade, 1B4.

Bottinger's process in wine analysis, 328.

Botulism, 361.

Boucard's bacillus, 80.

Bourbon vanilla, 267.

— whisky, 303.
BDvril, 404.

Braeutigan on colocynth, 543.

Brandy, 286, 808.

— definition of, 286.
Bran in coffee, 32, 83.
Brannt on asafoetida, 438.
Brassica, 221,

Braiitigam & Edelmann, 393.
Brazil wood, 240, 241.
Bread, adulterants in, 180.
Brieger o.i botulism, 361. •

— on flesh decomposition, 359, 860.

— on ptomaine-separation, 362, 363.
British brandy, 286.

— I harmaceutical codex, 521, 542, 606,

698.

— pharmacopoeia, 116, 147, 150, 154,

193, 208, 214, 217, 222, 225, 227, 228,
231, 235, 236, 238, 239, 240, 256,
261, 273, 288, 414, 424, 436, 437,
438, 440, 442, 444, 447, 455, 456,
457, 458, 459, 460, 462,. 469, 470,
472, 476, 477, 481, 483, 484, 486,
490, 492, 494, 495, 497, 506, 507,
508, 512, 513, 516, 518, 519, 520,
522, 525, 533, 534, 535, 537, 538,
541, 547, 548, 5 19, 550, 551, 555,
556, 560, 561, 566, 569, 570, 571,
572, 574, 578, 579, 580, 582, 584,
590, 593, 595, 697, 602, 606, 608,
609, 610, 611, 612, ()13, 614, 618,
619. 620, 621, 622, 626, 632, 634,
636, 637, 639, 646, 648, 649, 668.
673, 677, 682, 683, 684, 686, 686.

— wines, 310.
Brsemer's reagent, 424.
Bromine thermal value, 629.
Brooks' analysis of paprika, 213.
Brooks on pepper, 201.

Brown & Heron on malt, 162.

Brown & Millar's table for converting

sugar, 129.
Brown mustard, 221.

— on chloroform, 685, 686.

— on malt, 166.

Brown's analysis of cider, 360, 351.
Brucine, 574.

— separation from strychnine, 577.
Bryan, precipitation of dextrose, 131.
Buchanan on canned foods, 373, 374.
Buckwheat, 187.

— flour, 172.

— in coffee, 33.
Buisine on beeswax, 641.
Bulgarian bacillus, 80, 82.
Burgundy, 309, 310, 312, 313.

— pitch, 477.
Burkea Africana, 432.
Burton ales, 342.
Butter, 88, 628, 630.

— analysis of, 90.

— determination of mineral matter,.

90.

— fat, analysis of, 95.

composition of, 89.

specific gravity of, 95.

VOL. I.

722

INDEX.

Butter, iodine value of, 95.

— microscopical examination of, 105.
Buttermilk, 80.

Butter, " rancidity " of, 89.

— refractive index of, 96, 97, 98, 99.

— saponification, 95.

— substitutes, 94.

— under polarized light, 106.

— volatile fatty acids, 97.
Butyl alcohol, 288, 297, 299.
Butylamine, 359, 365.
Butyl chloral hydrate, 699.
Butyric acid, 491.

in butter, 89.

— esters, 288, 304.
Butyrin, 42, 89.
Butyro-refractometer, 96, 632.

Cacao beans, 16.
Cadaverine, 359, 360, 365.
Cadinene, 471, 608.
Cadmium butyrate, 102.

— value of cocoanut fat, 102.
Caesar & Loretz on strophanthus, 604.
Caffeine, I, 2, 3, 8, 9,. 31.

— determination of, 38.

— (pure), 10.
Caffetannic acid, 31.

determination of, 37.

Calabrian liquorice, 46B.
Calcii hypophosphis, 657, 699.
Calcium compounds, 699.

— dinitro-a-naphtholate, 217.

— salt of Arabic acid, 430.

— sucrate, 70.

— sulphate in catechu, 44 J.
in wine, 328.

Calendula in saffron, 241.
Calx, 699.

— chlorinata, 699

— sulphurata, 699.
Cambogia, 458.
Camellia japonica, 15.

— sasanqua, 15.
'Campbell-Brown on pepper, 207.
•Camphene, 261.

•Camphor, 588, 6S2.
•Camphorated oil, 683.
■Camphor liniment, 112.
-Canada balsam, 476, 622.

— turpentine, 476.
'Canadinic acid, 477.
Canadolic acid, 477.
Canadoresene, 477.
Canadinolic acid, 477.
Cane-molasses, sugar values of, 144.
Cane sugar, 117, 316, 492.

in condensed milk, 74.

Cannabihene 442.
Cannabinol, 442.
Cannabis indica, 442.

— sativa, 442.

Canned meats, 369.
Cantharides, 522,
Cantharadin, 522.

— determination of, 523, 524.
Cantharis vesicatoria, 522.
Capaloin, 5(,'8.

Cape aloes, 510.

— saffron, 242.
Capivi, 447.

Capric acid in butter, 89.

Caprinin, 42.

Caproic acid in butter, 89.

— esters, 288.
Caproin, 89.

Caprylic acid in butter, 89.
Caprylin, 42.
Capsaicin, 212.
Capsicum, 193, 217.

— annum, 208, 213.

— detection of, in ginger, 198.

— fastigatum, 208.

— minimum. 208.

— putescens, 208.
Caramel, 117.

— as colouring lor milk, 70.

— in vanilla, 268.

— in vinegar, 252.
Caramelized sugar in coffee, 32.
Caramel, tests for, 72.

— tests for, in vanilla, 269.
Caraway fruit, 609.

— oil, 609.
Carbohydrate foods, 117.
Carbohydrates, 338, 340.

— in ginger, 196.

— of honey, 150.
Carbolic acid, 690.
Carbo ligni, 700.
Carbon bisulphide, 700.
Carbonic acid gas in wine, 309.

in beer, 346.

Cardamoms, 444, 484.

— microscopical illu^^tration of, 447.
Carles on lead tests, 671.
Carmine, 425.

— lake, 239.
Carminic acid, 239.
Carnauba wax, 642.
Carnine, 401.
Carnitine, 401.
Carophyllum, 222.
Carpathian oil of turpentine, 622.
Carrotin in butter, 91, 93.
Carter Bell on milk, 44.

Carter Bell's analysis of bread, 181.
Carthagena ipecacuanha, 5t0.
Carthamus in saffron, 241.
Carvone, 606, 609.

— phenylhydrazone, 610.
Cascara, 611.

Casein, 42, 58, 78, 311.

— in chocolate. 30.
Caseinogen. 78, 79.
Caseoses. nitrogen as, 87.

INDEX.

723

Cassal & Gerrans on boron compound
in milk, 62.

on cider analysis, 356.

Cassia, 422.

— acutifolia, 483.

— angustifolia, 483.

— bark, microscopic illustration of,

231.

— oil, 232.

Castor oil, 112, 636, 639.
Catechin. 444.

— in tea, Allen 14.
Catechu, 5, 444.
Catechu-tannic acid, 444.
Cathartic acid, 598.
Catwell on opium 582.
Caustic potash, 710.
Cayenne annatto, 246.

— in ginger, 212.

— pepper, 191, 208.

adulteration of, 208.

in mustard, 217.

Cazeneuve's test in wine analysis, 381.
Cellulose in cocoa, 21.
Centrifugal apparatus for milk, 51.
Cera alba, 641.

— riava, 641.
Ceresin wax, 642, 644.
Cerii oxalas, 656, 700.
Cerotic acid, 235, 642.
Ceylon nutmegs, 235.

— vanilla, 267.

" Ceylon wilds " cardamoms, 445.
Chace on lemon oil, 264.
Chalk in catechu, 444.

— in pepper, 203, 204.
Chambers on gin, 307.
Chamomile, 608.
Champagne, 312, 313.

— brandies, 289.
Chapman on beer, 347.

— on meat extract, 411.
Characters of glucose, 148.

— of pure golden syrup contrasted

with glucose syrup, 145:

Charas, 142.

Chattaway, Pearmain & Moore's analy-
sis of cheese, 84.

Cheese, 83.

— adulteration of, 85.

— analysis of, 85.
Chemicals, table of, 689.

" Chemistry of essential oils," 364.
Chevreau on pepper, 206.
Chicory, 31.

— root, 32.
ChilUes, 208. 211.
China clay in coffee, 32.
Chinese cassia tree, 229.

— flies, 522.
Chiretta, 347.

Chloral hydrate, 421, 424, 687.

— — in microscopical analysis, 419.

— iodine, 424.

Chloride of tin in Demerara sugar, 143.
Chlorinated lime in microscopical
analysis, 425.

— soda in microscopical analysis, 425.
Chloroform, 425, 684.

Chlorophyll, 421, 422.
Chlorzinciodine. 425.
Chocolate, 27.

— composition of, 28.
Cholesterol, 89, 630, 638.

— in cocoa butter, 26.
Choline, 359, 361.
Chromic acid, 691.
Chrysammic acid, 511.
Chrysarobin, 436.
Chrysophaneiu, 598.
Chrysophanic a?id, 436, 484, 598, 599,

600.
Ohrysophanohydroanthone, 436.
Church on flour, 187.
Churrus, 442.

Ciamician & Silber on turmeric, 243.
Cider, 349

— adulteration of, 357.

— analysis of, 349, 350.

— vinegar, 248, 254, 350.
Cinchona, 525.

— alkaloids, separation of, 530.
Cinchotiaceae, 30.

Cinchona, microscopical analysis of,
526.

— succirubra, 525.
Cinchonidine, 526, 531, 532.
Cinchonine, 526, 531, 532.
Cineol, 446

Cinnabar in cayenne pepper, 208.

Cinnamate, 493.

Cinnamates in storax, 493.

Cinnamein, 439.

Cinnamic acid, 439, 440, 493.

— aldehyde, 231, 232, 471.

— ester, 440.
Cinnamon, 227, 228, 422.

— adulteration of, 229, 231.

— analysis of, 229.

— essential oil of, 230.

— microscopical illustration of, 229.

— oil. 232.
Cinnaniomum camphora, 682.

— cassia, 229.

— zeylanicum, 228.
Cinnamyl esters. 493.

— cinnamate, 439, 493.
Citral, 261, 263.

— in oil of lemons, 261.
Citraptene. 261.

" Citrated " milk. 79.
Citric aci I, 691.
Citrullol, 541.
CitruUus colocynthis, 541.
Claassen on opium. 593.
Claret, 310, 312, 313.
Clarifying sugar. Pellet, 132.
Clardon on brandy, 288.

724

INDEX.

Clay in pepper, 204.
Clerget's process for sugars, 146.
Clove oil in cinnamon oil, 231.
Cloven ^ 222.

— mi'croscopic examination of, 224,
225.

Coagulable proteins, 390.

— albumen, 405.
Coal-tar colours in wine, 331.

— dye in cream, 72.

— dyes in butter, 91.

— for colouring saffron, 240.

— lake in cayenne pepper, 209.

— solvent naphtha in turpentine, 624.

— yellow, 241.
Coca, 535.
Cocaine, 535, 538.

— determination of, 537.

— hydrochloride, 5B-^.
Cocamiue, 535, 538.
Coccus cacti, 238.
Cochineal, 191, 238, 250.

— adulteration of, 239.
Cochlospermum gossypium, 434.
Cocoa, adulterants of, 19.

— analysis of, 20.

— and chocolate, 16.

— butter, 26.

adulteration of,. 27.

composition of, 26.

substitutes for, 27.

— husks, analysis, 16, 18.

— microscopic examination of, 25.

— nibs, analysis of, 17.
Cocoanut fat, 97, 101, 102.

determination of, 103.

— oil, 101, 103, 105, 108, 109, 628, 632.
Cocoa shells, 26.

Cocoanut shells in pepper, 205.

Codeine, 582, 593.

Codfish bases, 359.

Cod liver oil, 359, 637, 640.

Coffee, 30.

— adulterants in, 31.

— analysis of, 34.

— Arahica, 15, 30.

— ash in, 34.

— constituents of, 31.

— specific gravity of, 34.
Cognac, 286.
Colchicine, 539, 541.
Colchicum, 539.

Collagen, determination of. 390.
Colledge on cantharides, 522.
Collidine, 359, 360, 366.
Collingwood Williams on rum, 304.
Colocynth. 541.
Colocynthin, 542.

Colophony, 437, 439, 440, 477, 478, 480,
482, 493.

— (Storch-Morawski reaction), 458.
Colorimeters, 291, 292.
Colorimetric method in cider analysis,

356.

Colostrum, 42, 44.

Colouring matter in alcohol, 298.

in butter, 91.

^ in cieam, 70.

in vinegar, 252.

in wine, 330.

Colouring matters, reactions of, 93.
Colour reactions of alkaloids, 503.
Commercial cane sugar and products,
135.

— glucose, 147, 353.

— saccharin, 673.

— sugar, ash determination, 136.
Commiphorinic acid, 470.
Composition of commercial glucoses,

148.
Compound liquorice powder, 467.

— mixture of senna, 484.

— rhubarb pills, 510.

— tincture of camphor, 588.
Concentrated liquors of the British

Pharmacopoeia, 496.
Condensed milk, 72.

analysis of, 73.

Coniferous honey, 150.

Coniine, 543.

Conium, 543.

Conroy on ipecacuanha, 558.

Constituents of glucose, 147, 148.

Converting optical rotations of sugar.

Brown & Millar's table, 129.
Convolvolin, 571.
Convolvulus scammonica, 481.
Cook on meat extract, 412.
Cook's method for meat extract analysis,

410.
Copaiba, 439, 447.

— acid and ester values, 450.

— adulterants in table, 452.

— African, author's rotation table, 452.

— balsams of, 449.

— colour reaction, 450.

— determination of essential oil in,

450.

— essential oil of optical rotation

tables, 450, 451.

— essential oils in, 449.

— fatty oils in, 453.

— oils. Cocking on, 462.

— resins in, 449.

— South American, 452.

Copaibas, true and adulterated, difference

in rotation, 453.
Copaiba, turpentine in, 453.
Copper in preserved meats, 376.
Corallin-soda, 421, 425.
Cordonnier's double stain, 420, 425.
Coriander seed in pepper, 208.
Corindine, 359, 360, 367.
Corn-cockle seeds in flour, 178.
Cornelison's method, butter-colouring,

detection of, 92.
Corn flour, 181, 184.
Corn oil, 630.

INDEX.

725

Cotton seed oil, 105, 108, 109, 110, 112,

115, 630, 632.
Cough mixtures, detection of opium in,

689, 590.
Coumarin, 266, 268, 269, 270.
C jwley & Catford on cardamoms, 445.

on chemical tests, 666.

Crampton & Simons on vanilla, 269.
Crampton & Tolman on whisky, 303.
Cream cheese, 83.

— colouring matter in, 70.

— fat in, 52.

— of tartar, 188, 254.

— preservatives in, 67.
Creatin, 401, 409.
Creatinin, 401, 409, 411.
Creosol, 455, 679.
Creosote, 453.

— analysis of (Behal & Choay), 455.

— beechwood, 453.

— description of, 453.

— properties of, 453.
Cresol, 678.
Cresotic acid, 679.
Creta preparata, 700.

Cribb & Richard's analysis of cocoa, 17.

analysis of chocolate, 21.

Cripps & Brown on cloves, 224.
Cripps & Dymond on aloes, 511.
Cripps & Whitby on ipecacuanha, 561.
Cripps on conium, 544.

— on ipecacuanha, 563.
Crismer's test for butter fat, 104.
Crocin, 240, 243.

Crocus sativus, 240.

Cross & Bevan on analysis of cocoa, 23.

Croton oil, 634.

— tiglium, 634.
Crude drugs, 427.

— fibre in flour, 177.
determination of, 191.

Cubebic acid, 456.
Cubebs, 456.

— adulterants, 456.

— genuine determined, 456.
Cuminic aldehyde, 471.

Ouniasse's table for alcohol determina-
tion, 292.
Cupreine, 526. 533.
Curcuma longa, 243.

— rotunda, 243.
Curcumin, 243, 245.
Curd soap, 647.

Curtman on salicylic acid, 679.
Cutch, 444.
Cuttle fish, 359.

Dandelion root, silica in, 35.
Darnel in flour, 178.
Date stones in coffee, 40.
powdered, diagram, 461.

— vinegar, 248.

Datura, stramonium, 602.
Davies on strychnine, 580.
Decoction of aloes, 512.
Decomposition of flesh, 358.
Decyl aldehyde, 261.
Demerara crystals, dyeing, 142.

— rum, 306.

Denis on almond oil, 259.

Denner on almond oil, 259.

Dennstedt on lard, 109.

Departmental committee on food pre-
servatives, 92.

on preservatives and colouring

matters. 379.

de Raczkowski's process in wine analysis,
328.

De Vrij on cinchona bark, 527, 528.

Dextrin, 119, 145, 146, 147, 148, 149,
174, 255, 256, 257, 339, 342, 431,
460.

— estimation of, 166.

— in beer, 346.
Dextro-and Isevo-rotation, 145.
Dextrocamphene, 235.
Dextropinene, 235.

Dextrose, 117, 118, 119, 121, 123, 132,
146, 147, 149, 150, 256, 257, 258,
460, 490.

Diacetylchrysarobin, 436.

Diastase, 160, 273.

— action of, 155.

— Ling on, 169.

— solids in, 159, 160.

Diastatic value, determination of, 159.

of malt, 339.

Diazobenzene butyrate, 69.

— potassoxide, 59.
Dichrysarobin, 436.
Diethylamine, 359, 364.
Digitaligenin, 545.
Digitalin, 545, 646.
Digitalis, 545.
Digitalose, 545.
Digitoxin, 546, 547.

Diagnostic characters of leaf of tea, 15.
Dihydrocollidine, 359, 360, 367.
Dihydroxystearic acid, 639.
Dill fruit, 606.

— oil, 606.
Dimethyl-guaiacol, 465.
Dinitro cresylate of sodium, 241.
Dipentene, 235, 622.
Disinfection of wine barrels, 332.
Distilled milk, tests with, 65.

— water, 697.
DitLe on honey, 152.

" Doctoring " champagne, 289.
Dorema ammoniacum, 434.
Doreraus on canned meats, 372.
Dott on nitrous ether, 489.

— on opium, 584.

Double staining in microscopic analysis,

420.
Dowzard on benzoin, 442.

726

INDEX.

Dowzard on opium, 587.

— on saffron, 242.

Dragendorff's method of ptomaine
separation, 364, 369.

— reagent, 503.
Drugs, 427.

Dubois' analysis of chocolate, 28.

Dubosq's colorimeter, 292.

Dunbar on almond oil, 259.

Dunstan & Henry on podophyllum,

695.
Dunstan & Ransom on belladonna, 515,

516.
Dunstan & Robinson on arsenic tests,

650.
Dunstan & Short on nux vomica, 574.
Dunstan & Robinson on chemical tests,

662.
Dupont on lard, 109.
Dupre's process for bread analysis, 180.

— test in wine analysis, 330.
" Durum " wheat, 183.

Dyer & Gilbard on ginger, 197.

— on vinegar, 248.

Dyer's analysis of dried chicory, 33.
Dyes in foodstuffs, 385.

Earthnuts, 187.

East African vanilla, 267.

East Indian senna, 483.

Easton's syrup, 580.

Eau-de-vie, 286.

Eber on botulism, 361.

Elaterium, 548.

Elettaria cardamomum, 444.

Embrey on flour. 179.

Emmett & Grindley on meat extract,

409.
Emodin, 483, 508, 512, 599, 600.
Emulsin, 257.
English and German beers, difference

between, 342.
English lager, 341.
Enzymes, 389.
Ergot, 549.

— in flour, 178.

Eriksson on liquorice root, 465.
Erythrcdextrin, 150.
Erythroxylon coca, 535.
Essence of lemon, 260.

— of vanilla, 263.

Essential oils, 191, 193, 198, 214, 236,
243, 257, 260, 261, 308, 435, 437,
446, 447, 448, 470, 606.

— oil in cloves, 224.

in mustard, 217, 221.

in nutmegs, 233, 234, 235.

of almonds, 257.

of asafoetida, 438.

of aniseed, 606.

. of cade, 608.

of cajaput, 609.

Essential oil of caraway, 609.

of cinnamon, 230, 610

of chamomile, G08.

■ of cloves, 223, 610.

of copaiba, 610.

of coriander, 610.

of cubebs, 610.

of dill, 606.

of eucalyptus, 611.

of juniper, 612.

of lavender, 613.

of lemon, 260.

of mace, 238.

of peppermint, 614.

of pimento, 227.

of pine, 619.

of rose, 619.

of rosemary, 620.

of spearmint, 618.

of santal, 621.

of turpentine, 622.

Ester in spirits, 289, 290, 291, 300, 301
303, 305, 306, 307.

— determination of, 295.

— of eugenol, 225.

Ether standard for spirits, 289.
Ethereal tincture of lobelia, 671.
Ether extract of cloves, 223.

of pepper, 209.

of mace, 237.

Ethyl acetate, 250, 288.
Ethylidene diamine, 365.
Ethyl alcohol, 273, 281, 282.

refractive index of, 284.

Ethylamine. 364.
Ethyl cinnamate, 493

— nitrite in nitrous ether, 487.
Eucalyptol, 611.

Eucalyptus oil, characters and tests of,

611.
Eugenia caryophyllata, 222.
Eugenol, 225, 231, 235, 266, 471.
Evans on liquorice, 467.
Ewell & Prescott on salicylic acid,

678.
Export lager, 341.
Extract gravity of beer, 344.

— of belladonna, 616, 618.

— of cinchona, 534.

— of coca, 537.

— of ergot, 549.

— of hydrastis, 554.

— of hyoscyamus, 556.

— of ipecacuanha, 565, 566.

— of jaborandi, 670.

— of liquorice, 467.

— of malt, 155, 340.

— of meat, 396.

— of nux vomica, 576, 578.

— of opium 686.

— of strophanthus, 605.

— of yeast, 336.

Eykman's apparatus for nitrous ether,
4S8.

INDEX.

727

*' Facing " of rice, 182.
Fairley on vinegar, 248.
False mace, 236.

— myrrh, 470.

Farr & Wright on belladonna, 518.

on colchicum, 53J, 541.

on conium, 544, 545.

— — on ipecacuanha, 566.
on jaborandi, 569.

— — on nux vomica, 575.
on stramonium, 603.

Fat, determination of, in butter, 90.

— in coffee, 35.

in flour, 175.

in milk, 47.

— differences, Bremer on, 894.

— ii chocolate, 20.

— in cocoa, 20,

Fats of the Pharmacopoeia, 632.

Fatty oils, 626.

Fawsett on aloes, 510.

Fehling's solution, 119, 122, i:i3, 124,
126, 127, 128, 160, 161, 162, 177,
189, 191, 204, 196, 216, 242, 245,
256, 257, 322, 323, 340, 345, 346,
410, 430, 465, 468, 673.
milk sugar determined by, 66.

Fennel, 224, 46S.

Ferri arsenas, 700.

Ferric chloride, solution of, 425.

Ferrum, 660, 662, 701.

— compounds of, 701.

— redactum, 661, 662.
Ferula fostida, 437.

— galbaniflua, 456.

Fibre, determination of, in starch, 191.

— in cocoa, 21.
Fibres in saffron, 241.
Figs in coffee, 37.
FiHcic acid, 469,
Filicmylbutanone, 469.
" Filled " cheese, 85.
Filmarone, 46J,

Fischer on salicylic acid, 678.
Fixed ether extract, 191.

— oil in mustard, 216.
-of mustard, 222.

— oils, 632.
Fiavaspidic acid, 469.
Flavoured Jamaica rum, 305.
Flavouring essences, 191.
Flesh, decomposition of, 358.

— foods, 358.
Fleury on vinegar, 254.
Flour, a3idity of, 177,

— adulteration of, 177.

— in opium, 5^2.

Fluid extracts, standards for, 496.
Fluorides, detection of, in butter, 94.

— in beer, 349,

♦' Food adulteration " (J, P, Battershall),

Formaldehyde, 281, 282, 332, 383.

— determination of, in milk, 66.

— in milk, 60, 63, 64, 66.
Formic acid, 235, 446.
Forster on mustard, 220,
Fortified wine, 336,
Fractionating still. 287.
French cider, 349, 351, 352.

— codex, test for olive oil, 115.

— digitalin, 546.

— Government on brandies, 286.

— laws for wine, 310,

— official method for determination

of alcohol, 291, 295, 296, 298, 300,
302.

— — methods in wine analysis, 830.

— oil of rose, 619,

of turpentine, 478, 494, 622.

Frerichs on opium preparations, 594.
Fresenius & Urunhiit on salicylic acid,

6S1,
Fresenius & Popp on meat extracts, 413,
Fresenius' method for pastes, 185.
Freyer's process for salicylic acid, 681,

682,
Friedrichs on myrrh, 470.
Frohde's reagent, 543.
Fructose, 120.
Fruit sugar, 120.

Fuller's earth in vanilla testing, 269.
Furfurol, 225, 289, 3U5. 306. 470.

— estimation of, in alcohol, 293.

— test for sesame oil, 116.
Fusel oil, 288, 296.

Gadamer's method for mustard oil, 219,
Gadinene, 359, 361, 367.
Galactose, 119, 121. 154, 433, 470,
Galbanum, 435, 456.

— qualities of, 457.
Gallic acid, 691, 694.
Gallisin, 148,
Gambler, 5, 444,
Gamble on sour milk, 78.
Gamboge, 458.

— adulterants, 459.

— examination of, 459.

— kinds of, 459.

— nature, 458, 459,

— reaction for starch, 459.

— acid, 458.
Game, 858.
Ganjah, 442,
Garcinia Hanburii, 458.
Garnett & Grier on ginger, 196.
Garnett on ginger, 198.

Garsed & Collie's cocaine process, 537,
Gautier on flesh decomposition, 359,
360,

— on wine analysis, 331.

— table of tea infusions, 3.
Gelatin, 359.

728

INDEX.

Gelatin, determination of, 391.
Gelatine, 311, 312, 372, 413.

— composition of, 414.

— solution, 11.

— test in wine analysis, 330.

— in cream, 69.

Gelatin in meat extract, 41C,
Gelsemium, 550.
Geneva gin, 307.
Genevre, 307
Gentian, 347, 348, 459.

— adulterants, 460.
Gentiana lutea, 459.
Gentiamarin, 460.
Gentianose, 460.

Gentian, powdered, diagram, 461.

— tincture of, 460.
Gentiin, 460.
Gentiobiose, 460.
Gentiogenin, 460.
Geotiopicrin, 459.
Geraniol, 261.

Gerber's centrifugal apparatus, 51, 52.
German and English beers, difference
between, 342.

— ciders, 351.

— digitalin, 546.

— lager, 341.

— official method in wine analysis,

316, 322, 323, 326.

process for determming alcohol,

296.

— oil of turpentine, 622.

— pharmacopceia, 524, 594, 687.
Gerrard's experiments on cayenne, 209.

— process for sugar, 127.
German sausages or wursts, 386.
Gilbert on boron in milk, 61.

Gilg, Thoms, & Schedel on strophanthus,

605.
Gin, 307.
Ginger, 193, 224.

— ale, 196.

— analysis of, 194.
Gingerine, 198.
Ginger in pepper, 205.

— microscopic examination of, 197.
Gingerol, 196. ^

Gladhill on pepper, 199.

Glass-packed meats, 382,

Gliadin, 174, 175.

Globulin, 174.

Gluco-amylins, 148.

Glucosazone, 121.

Glucose, 119, 124, 146, 147, 189, 214,

221, 256, 316, 349, 350, 366, 465,

466, 491, 492, 545.

— and maltose, separation of, 148,

149.

— experiments by Wylie, 148.

— in rice, 182.

— in saffron, 211.

— syrup, 147.

constituents of, 146.

Glucose syrup, formulae for calculating

percentage of, 146.
Glucosides, 2, 214, 217, 257, 598, 599.
Glucuronic acid, 465,
Glue, 413.

Glusidum, 558, 673.
Gluten, 174, 177, 183.
Glutenin, 174.
Glycerin, 273, 421, 425, 643, 692.

— in rice, 182.

— in saffron, 241.

— in wine, 313, 314, 326, 336.

— in beer, 345.

— preparation in microscopical analy-

sis, 419.
Glycerinum, 662, 702.
Glycogen, 392, 393.

— determination of, 393, 394.

— table . f, 393,
Glycyrrhetinic acid, 465.
Glycyrrhiza glabra, 462.
Glycyrrhizin, 463. 465,466.

— Cederberg's process, 464.

— in liquorice, 463.
Goa powder, 436.
Golden syrup, 144.

— syrups or treacles, pure, composition

of, table, 144.
Gordin & Prescott on hydrastis, 552.
Gorgonzola, adulteration of, 85.
Goske on analysis of cocoa, 22.
Gottlieb's method of milk analysis, 51.
Graham, Redwood & Hobhouse's table

for beer, 344,
Graham, Stenhouse & Campbell on

coffee, 35, 37.
Granada cocoa bean, 17.
Granilla, 238.
Granulose, 166.
Grapes, boric acid in, 356.
Gravimetric process for wine analysis,

323.
Green extract of belladonna, 518.

of hyoscyamus, 556.

Greenish & Braithwaite on insect pests,

423.
Greenish & Collin on coca, 536.

on digitalis, 546.

on microscopic analysis, 422.

on starch, 168, 169, 170, 171.

on stramonium, 602.

Greenish & Wilson on cantharides, 522.
Greenish on cardamoms, 445,

— on cinnamon, 229.

— on cocoa shells. 26.

— on microscopical analysis, 416, 422.

— on myrrh, 471.

— on senna, 484,
Green tea, 1,
Griebel on mace, 237.
Grier on ginger, 198.
Griess-Ilosvay's reaction for bread, 181,
Ground rice in pepper, 205,
Griitzner on mustard oil, 319,

INDEX.

729

Guaiacic acid, 457.
Guaiacol, 455.
Guaiaconic aci i, 467.
Guaiacum, 457, 482.

— adulterants, 45S.

— ammoniated tincLure of, 458.

— genuine, characters, 458.

— officinale, 457.
Guaiareiic acid, 457.
Guanidine, 359, 366.
Guanin, 410.
Guaza, 442.

Guichard on gum arable, 431.
Guibourt on ipecacuanha, 558.
Guignes on scammony, 482.
Gum, 435, 480.

— acacia, 435.

— arable, 430.

— Benjamin, 440.

— resin, 437.

— tragacanth, 203.

Gunning- Arnold method in meat extract,

402.
Gunning's method in meat extract, 402,

408.

modification of, 403.

Gunn on coca, 536.
Giinther's bacillus, 80, 82.
Gurjun balsam, 448,
Gutzeit's test for arsenic, 663.

Halphen test for cottonseed oil, 115, 116.

Ham and bacon (borax-treated) as pre-
servatives, 379.

Hamill on preservatives in cream, 67

Hamill's recommendations on cream
preservatives, 68.

Hammarsten's method in analysis of
chocolate, 29.

Hanausek on pepper, 208.

Hanus on cinnamon, 232

— on vanilla, 267.
Hammill on rice, 182.
Hard soap, 646.
Hard paraffin, 646.
Hardy's test for alcohol, 280.
Haricot bean flour, 172.

Harnack & Hildebrand on opium pre-
parations, 5P5.
Harrison & Gair on malt, 161, 162, 103,
164, 165.
on strychnine, 580.

— on maltose, 160.
Hashish, 442.
Hassel on pepper, 211.

Hasse on opium preparations, 594.

Haupt on botulism, 361.

Harvey & Wilkie on lead test, 669, 671.

Harvey on salicylic acid, 680.

Hawkin on cocaine, 539.

Hazel nut oil, 633.

Heerabolene, 471.

Heeraboresene, 470.
Hefelmann on mace, 238.

— on saccharin, 674.
Hehner & Mitchell test, 629, 6.30.
Hehner & Richmond's milk formula, 53,

54.
Hehner & Skertchly on analysis of

cocoa, 23.

on coffee, 38.

Hehner on boron in milk, 61, 62.

— on chemical tests, 666.

— on detection of fluorides, 94.

— on meat extract, 404, 409.

— on olive oil, 115.

— on pepper, 206.

— on sugar, 127.

— on vinegar, 250.

Hehner's test for detecting formalin in

milk, 64.
Hehner value of fatty acids, 627.
Heisch on pepper, 191.
Heliotropin, 268.
Henderson on belladonna, 520.
Herabol myrrh, 470.
Herapath's reaction for quinine, 533.
Hercod & Maben on pepsine, 473.
Herring pickle, 359.
Heisch's analysis of cocoa, 16.
Hess & Prescott on vanilla, 269.
Hesse's hexacetyldichrysarobin, 436.
Hewitt on alcohol analysis, 295.
Hexabiose, 117, 121.
Hexamethylene tetramine in wines, 336.
Hexoses, 119, 121.
Hexylamine, 365.
Higher alcohols, 308.

determination of, 296.

Hilger on botulism, 361.

• — on vinegar, 252.

Hill on lead tests, 671.

Hiltner on oil of lemons, 264.

Hinkel's method in alcohol analysis, 281.

Hirschsohn on aloss, 510.

Histidine, 401.

Hock, 312, 313.

Hoffmann & Hilger on flour, 178.

Hog tragacanth, 432.

Hollands gin, 307.

Holmes on belladonna, 514.

— on ipecacuanha, 557.

— on strophanthus, 604.
HommoUe's digitalin, 546.
Homonataloin, 508.
Homopiperidinic acid, 359.
Homopyrocatechin, 456.
Honey, 119, 120, 149, 491.

— adulteiation of, 150.

— adulterants, reducing values of, 152.

— analysis, 153.

— genuineness of, 152, 153.

— microscopic examination, 150.

— optical rotation of, 151, 152.

— rotatory power of adulterants, 151,

152.

730

INDEX.

Honey, specific gravity of, 151.

— tannin precipitate of, 154.
Hooper on cannabis indica, 442.
Hops, 337, 347.

Hordenm distichon. 168.
Horse flesh, 359, 391.

determination of, 394, 395.

Howard on cinchona bark, 528.
Hubl's iodine process for fat, 628, 629.
Hudson V. Bridge, 491, 566.
Hungarian oil of juniper, 612.

of turpentine, 622.

Hiippe on bacteria in milk, 79, 80,

82.
Husk in cocoa powders, 22.
Hydrargyrum, 703.

— compounds, 703.
Hydrastine, 554.
Hydrastis, 551.
Hydrobromic acid, 455, 691.
Hydrocarbon wax, 643.
Hydrochloric acid, 426, 691.
Hydrocinchonidine, 526.
Hydrocinchonine, 526.
Hydrocyanic acid, 253, 692.
Hydrogen peroxide in cream, 67.
Hydrolysed sucrose, 144.
Hydroxy-phthalic acid, 678.
Hymenopterse, 149.
Hyoscine, 515, 521.
Hyoscyamine, 515, 521, 602.
Hyoscyamus, 555.
Hypoxanthine, 401, 410.

Idioblasts, 15, 11.
Ulicium verum, 606.
Imitation coffee, 32.

— rum, 304.
Indian beeswax, 642.

— corn, 181.

— gamboge, 459.

— gum, 434.

— saffron, 243.

— senna, 484,
Indigo as a tea-facing, 5.
Inosite in wine, 335,
Insect pests, 423.

— wax, 642.

Inversion values, sucrose, Herzfeld,
133.

table, 133.

Invert sugar, 118, 121, 123, 357, 492.

estimation of, 137.

lodeosine, 505.

Iodide method for glycerine determina-
tion, 328.
Iodine green, 425.

— value of butter, 95.

of fatty acids, 627.

of lard, 109.

of mace, 236.

of olive oil, 112.

Iodine, water as a microscopic reagent,

426.
Iodoform, 703.

— test for alcohol, 274.
lodopotassium iodide, 426,
lodum, 656, 703.
Ipecacuanha, 556.

— colour reactions of, 567, 568.

— preparations, 566.

— wine, 567,
Ipomcea orizabensis, 481.

— purga, 571.

— sitnulans, 571.
Ipuranol, 541.
Irish whisky, 302.
Iron, 691, 701,

— and quinine citrate, 534.

— compounds of, 701,

— oxide, 246.

in adulteration of cheese, 85,

Isinglass, 414,
Iso-atropyl-cocaine, 535,
Isobarbaloin, 512.
Isobutyl alcohol, 288.
Isocholesterol, 633.
Isoemodin, 483, 598.
Isoeugenol, 235, 266.
Isolinolenic acid, 636.
Isomylamine, 365.
Isopropyl alcohol, 289, 491.
Isopropylamine, 364,
Isovanillin, 266.
Isoricinoleic acid, 639.
Italian pastes, 183, 184.

Jaborandi, 569,

Jackson & McGeorge on vanilla, 270,

Jalap, 571,

Jalapin, 571.

Jalap resin, 482,

Jamaica ginger, 195, 196.

— rum, 303, 304.
analysis of, 305,

Jam analysis, formulae, 135.

Roseley's method, 134, 135.

Jauke's process for tannin, 10.

Japan wax, 628, 642, 644.

Java vanilla, 267,

Jean-Amagat refractometer, 632,

Johnson on analysis of coffee, 36,

Jorgensen on mustard, 221,

Jorgensen's process for saccharin,.

677,
Jorissen on salicylic acid, 679.
Jowett & Potter on chrysarobin, 436.
Jowett on maltose, 160.
Juckenack & Hilger on caffeine, 39.
Juckenack's method, 184, 185.
Juniper, 307.

~ berries, 308.

— oil, 612.
Juniperus oxycedrus, 608.

INDEX.

731

Kaolin, 203.

Kastle on saccharin, 076.

Keller on digitoxin, 547.

— on ergotin, 549.

— on ipecacuanha, 563.
Keller's process for belladonna, 515.
Kempner on botulism, 361.
Kephir, 80.

Kerner on flesh, 358.
Ketone sugars, 122.
Kino, 460.

Kino-tannic acid, 460.
Kino, tincture of, 462.
Kirkby on ipecacuanha, 559.
Kirschner on butter, 101.
Kjeldahl-Gunning method, 169.

in meat extract, 402.

Kjeldahl on malt, 161.
Kjeldahl's law, 165.

— process, 86, 87, 90, 340, 369. 373,

402, 403, 405, 406.

for milk, 78.

in milk analysis, 57.

Knapp on sugars, reducing, 127, 128.
Koettstorfer value of fa.t, 626.
Kolbe, on salicylic acid, 678.
Konig on adulteration of coffee, 31.

— on beer analysis, 342.

— on canned meats, 369.

— on meat extract, 406, 407.
Konig' s analysis of vanilla, 160.

of cocoa, 16.

Koningh on boron in milk, 61.
Kottmayer on ipecacuanha, 563.
Koumiss, 80, 81.
Kraemer on saffron, 241.
Kraft on liquorice, 469.
Kremel on aloes, 511.

— on pepsine, 476.
Kulisch on cider, 351.
Kunze on analysis of cocoa, 23.
Kunz's method in wine analysis, 334.
Kurkum, 243.

Labelling of canned goods, 3 :3,
Lactalbumin, 42.
Lactic acid, 274, 692.

bacilli, 79, 81.

in cheese, 85.

Lactoglobulm, 42.

Lactose, 42, 55, 56, 57, 78, 79, 117, 119,
123, 154.

— birotation of, 154.

— characters of, 154.

— determination of, in condensed milk,

74.

— in chocolate, 29.
Lager beers, 341, 342.

Landolt's formula for camphor, 682.
Lard, 106, 630, 646.

Lard adulteration of, 108.

— melting-point of, 108.

— microscopic examination of, 110.
Laurie acid in butter, 89.
Lavender oil, 613.

adulterants, 614.

ester contents, 613.

— — varieties, 613.

La Wall and Bradshaw, on ginger, 198.

Lawson on annatto, 246.

Leach & Lythgoe on alcohol analysis^

283.
Leach on cider vinegar, 255.

— on mace, 236.

— on pepper, 203.

— on spices, 191.

— on turmeric, 243.

Leach's formula for calculating glucose

syrup, 146.
Lead in chemicals, 649, 668.
Lead, 710.

— chromate, 204.

— compounds, 710.

— number, 270, 272.
of vanilla, 270.

— tests for, 648.
Leaf lard, 106, 107.

Leeds' method, butter colouring, detec-
tion of, 92.

Leffmann & Beam's analysis of con-
densed milk, 75.

Leffmann-Beam process for butter fat,
102.

in milk analysis, 52, 53.

centrifugal apparatus, 51.

Leger on cantharides, 525.

Leger's reaction for aloes, 609.

Legislation on preserved meats, 377,
378, 383.

Legler's method in wine analysis, 334.

Leguminous starch, 177.

Lehmann on canned foods, 374.

Lemon essence, 260.

— grass oil, 261.

— oil, microscopical illustration of,

160.
Lentil flour, 173.
Lentils, 187.
Lentil starch, 172.
Lenz on pepper, 204.
Leonard & Smith on camphorated oil,

683.
Leonard on Hehner's formaldehyde test,

64.
Leucine, 401.
Leucomaines, 358, 359.
Levant galbanum, 457.
Levulose, 117, 118, 120, 121, 123, 149,

150.
Lewinsohn on myrrh, 471.
Lewkowitsch on almond oil, 634.

— on butter, 104.

— on cod liver oil, 639.

— on lard, 106, 107, 109.

732

INDEX.

Lewkowitsch on olive oil, 116, 116.
Leys' test for milk, 247.
Xiiebermann on resin, 478.
Liebig's extract of meat, 397.
Lignified tissue in microscopical analy-
sis, 419.
Lime, 705.
Limed ginger, 195.
Lime method in wine analysis, 328.
Liming nutmegs, 233.
Limonene, 261, 446, 606, 609.

— nitroso-chloride crystals, 265.
Linalool, 261.

Ling & Maclaren, sugar values of cane

molasses, 144.
Ling & Rendle on malt, 163.
Ling on malt, 165.

— on maltose, 160.
Ling's formula for malt, 164.

— method in malt analysis, 339.
Liniment of belladonna, 517.
Linimentum camphorse, 683.

— opii, 587.
Linoleic acid, 235.
Linolenic acid, 636.
Linolic acid, 636.
Linseed oil, 636.
Lintner on malt, 161, 165.

— value for malt, 164.
in malt, 339, 340.

Linum usitatissimum, 636.
Liquidambar 07-ie7italis, 492.

— styraciflua, 492.
Liquid cochineal, 239.

— extract of belladonna, 516.

of cinchona, 534.

of coca, 537.

of ergot, 549.

of hydrastis, 554.

of ipecacuanha, 565, 566.

of jaborandi, 570.

of liquorice, 467, 512.

of nux vomica, 577.

of opium, 586.

— paraffin, 646.

— tar, 477.
Liquor ammonise, 704.

acetatis, 704.

citratis, 704.

fortis, 655.

— arsenicalis, 704.

— arsensii hydrochlor., 704.

— bismuthi, 660.

et ammonii citratis, 662, 704.

— calcis, 705.

chlorinatae, 705.

— ethyl nitritis, 490.

— ferri acetatis, 661, 705.

perchloridi, 705.

fortis, 661, 662.

pernitratis, 661.

persulphatis, 705.

— hyd. peroxidi, 706.

— hydrarg. nitratis acidus, 705.

Liquor hydrarg. perchloridi, 705.

— hydrogenii peroxidi, 667, 706.

— iodi fortis, 706.

— magnesii carbonatis, 706.

— plumbi subacetatis, 706.

— potassse, 707.

— potassii permanganatis, 707.

— sodii arsenatis, 707.
chlorinatae, 707.

— strychninae hydrochloridi, 580.

— zinci chloridi, 707.
Liquorice, 484.

— root, 462.

analysis of, 463.

Hafner's method of analysis,

463.
microscopic examination, 462.

— sugars, 464.

Lister on bacteria in milk, 79.
Lithii carbonas, 707.

— citras, 707.
Lithium, 708.
Liversege on camphor, 684.
Lobelia, 571.

Lobeline, 571.
Laevo-pinene, 261.
Logwood, 240, 241, 242.
Loluim temulentum, 178.
Long nutmeg, 233.

— pepper, 201, 202, 207.
Lowenthal's calculations for tannic acid,

etc., 11, 12.

— permanganate process, 444.

— process in wine analysis, 332.
Lucas on squills, 491.

Lupin seeds in coffee, 32.
Lutein, 184.
Lyons on coca, 536.

— on colchicine, 540.
Lythgoe-Babcock's method in cheese

analysis, 85.
Lythgoe on analysis of coffee, 37.

— on cider vinegar, 255.

Maben on hydrastis, 551.
Macaroni, 183.
Macassar mace, 236, 237.

— nutmeg, 233.
Mace, 224, 2-35, 236.

— adulteration of, 236.

— microscopical examination of, 238.
MacEwan on nitrous ether, 486.
MacFadden on preserved meats, 376.
MacFarlane on analysis of coffee,

36.
Maclagan on cocaine, 539,
Madeira, 313.
Magnesia, 699. 708.
Magnesii carbonas, 708.

— sulphas, 708.
Magnetic iron filings in tea, 7.

ore in tea, 5.

INDEX.

733

Mahrhofer's process for starch deter-
mination, 3S7.
Maisch on saffron, 240.
Maize, 177, 179, 187.

— as adulterant in cocoa, 19.

— flour, 172, 181.

— oil, 110, 628, 632.

— starch, 147, 168.

— whisky, 302.
Malabar cardamoms, 445.

oil, 446.

Male fern, 429, 468.

Malic acid, 254, 255, 256, 351, 356,

504.
Malt, 249, 250.

— Analysis Committee of Institute of

Brewing, 161.

— composition of, 156.

— analysis of, 156.

— diastatic value of, 156, 157, 158.

— extract, 155,

— extractive matter of, 339.

— in vinegar, 253.

— liquors, 337.

— spirit, 301.

— valuation of, 338.

— vinegar, 248, 256.

— whisky, 302.

Maltose, 117, 118, 123, 124, 145,146,147,
149, 155, 160, 162, 177, 339, 345.

— specific rotatory power, 155.
Mandelic acid, 258.
Mangalore cardamoms, 445.
Manihot ^dilissima, 171.
Manipulations of wine, 310.
Mannite, 149.

Mann on strophanthus, 604.
Mansfield on lard, 109.
Maple, 117.

Maple products, Bryan, 136.
Maracaibo copaiba, 448.
Maranham copaiba, 448.
Maranta, 170.
Margarine, 94, 101.

— Act, 94.

— cheese, 83, 85, 93.
Marigold in butter, 93.

— in saffron, 242.
Marmalade analysis, Boseley, 134.

formulae, 135.

Marquardti's process for alcohol, 296.
Marsala, 313.

Marsh-Berzelius test for arsenic, 666.
Marsh's test for arsenic, 667.
Martelli on pepper, 206.
Martinez on butter, 90.
Martinique rum, 307.
Martius yellow, 184.

in butter, 93.

in mustard, 216.

Massol's bacillus, 80.
Mastic, 482.

Matthes & Rammstedt on hydrastis,
554.

Matthews & Parker, golden syrup analy-
sis, 147.
Matzoon, 80.

Mayer & Merling on lithium, 708.
Mayer's reagent, 502, 520.
McGill on analysis of coffee, 36.

— on cloves, 223.
Meat bases, 391, 401.

Meat extract, analysis of, 407.

composition, 398, 399.

fluid, 398.

how produced, 398.

nitrogen in, 400.

Meat fibre in meat extract, 404.

— fibrin, 359.

— juice, 398.

— juices analysis, 400.
Mechanical separation in microscopic

analysis, 421.
Meconic acid in cough mixtures, 589»

590.
Medicus & Schwab, determination of

stare b, 388.
Melaleuca, 609.
Mentha arvensis, 615.
Mentha viridis, 618.
Menthol, 708.

— determination of, 615, 617.
Menthone, 617.

Merchandise Marks Act, 309, 310.
Merck on digitalin, 646.

— on squills, 490.

Mercury salts, reduction of, 127.

Merson on cochineal, 239.

Messina oils, 261.

Messinger & Vortmann on salicylic acid,.

680, 682.
Metacresol, 471.

Meta-diamido-benzoic acid, 385.
Meta-dinitro-benzoic acid, 385.
Metallic contamination in vinegar, 252.
Metanil yellow, 184, 186.
Metchnikoff on bacteria in milk, 82.
Metroxylo'ii sagu. 170.
Methyl-acetol, 253.

Methyl alcohol, 225, 274, 281, 282, 300»
'309, 695.

detection of, 280.

determination of, 282.

in nitrous ether, 490

refractive index of, 284.

Methylated spirit, 274.
Methyl-benzoyl ecgonine, 535, 537.
Methyl-cinnamyl ecgonine, 535, 538.
Methyl ether of dichrysarobin, 436.

— gadinene, 359, 361.

— guanidine, 359, 360, 366, 401.

— morphine, 593.

— protocatechuic aldehyde, 266.

— salicylate, 2.
Mexican scammony, 481.

— vanilla, 267.

Micko on meat extract, 411.

— on rum, 307.

734

INDEX.

Microscope for analysis, 416.
Microscopical analysis, 416.

— examination of allspice, 227, 228.
of annatto, ^47.

of belladonna root, 513, 514.

of butter, 105.

— ot cinchona, 526.

of cloves, 224, 225.

of coca, 535, 536.

of cocoa, 25.

of colocynth, 542.

of coffee, 40.

of digitalis, 546.

of drugs, 427.

of flour, 187.

of ginger, 197.

of honey, 150.

of hyoscyamus, 556.

of ipecacuanha, 558, 559, 561.

of jalap, 573.

of iard, 110.

of mace, 238.

of mustard, 216, 217.

of nutmegs, 233.

of opium, 582.

of pepper, 206, 207.

of pepper, cayenne, 211, 212.

of rhubarb, 601.

of saffron, 242.

of senna, 484.

of starch, 167.

of stramonium, 602.

of turmeric, 245.

TMilk, 41.

— adu teration of, 58.

— amount of reagents in analysis,

53.

— analysis of, 46.
of altered, 77.

— annatto colouring in, 247.

— bacteria in, 79.

— composition of, 42.

— condensed, 72.

— determination of fat, 47.
of mineral matter, 47.

— evening, analysis of, 44.

— frozen, 45.

— infected, 45.

— • morning, analysis of, 44.

— poisonous, 59.

— polarimetric determination of, 53.

— powdered, 72.

— preservatives, 60.

— proteids of, 57.

— ropy, 45.

— specific gravity of, 46.

— sugar, 42, 124.

determination of, in milk, 53.

in cheese, 86.

volumetric determination of, 56.

— tuberculous, 45.
Millard on podophyllum, 597.
Milliau on oUve oil, 115, 116.
3Iills' colorimeter, 292.

Mineral adulterants in mustard, 216.
of flour, 179.

— matter in mace, 236.

determination of, 195.

in molasses, 145.

in pepper, 202, 209.

in saffron, 241.

in vinegar, 252.

in wine, 316.

of extract of malt, 401.

— oil, 637.

Mitchell & Smith on alcohol analysis,

297.
Mitchell on cloves, 223.

— on flesh foods, 359.

— on ptomaines, 364.
Mitchell's analysis of cinnamon, 228.
Mixtures of sucrose, invert sugar and

glucose, 134.
Moller on cinnamon, 231.

— on nutmeg, 234.

— on turmeric, 245.
Mohler's test in microscopy, 385.
Molasses, 307.

— in coffee, 33.

— spirit, 304.

— treacle, golden syrup, analysis, 144.
Monamines, 359, 364.

Monhaupt on butter, 101.
Monophenols, 455.
Moor's analysis of drugs, 427.
Morin on brandy, 288.
Morphine, 582, 590.

— acetate, 590.

— detection of, 592.

— hydrochloride, 591.

— tartrate, 591.
Moschus moscJiiferus, 469.
Moselle, 312, 313.

" Mother cloves," 222, 224.
Mucic acid, 470.
Mucilage, 214, 424, 426.

— in microscopical analysis, 419.
Miiller on chloroform, 687.
Mulliken & Scudder's test for alcohol,

280.
Musa sapientium, 171.
Muscarine, 359, 361, 367.
Musk, 469.
Muskone, 469.
Mussels, 359.
Mustard, 213, 622.

— adulteration of, 216.

— analysis of, 216.

— microscopic examination of, 216,

217.

— oil, 214, 632.

— plasters, 214.

— turmeric in, 244.
Muter on nitrous ether, 489.

— on salicylic acid, 678.
Muter's analysis of cheese, 84.

— method in milk analysis, 58.
Mutton suet. 111.

INDEX.

735

Mycoderma aceti, 248.
Mydaleiue, 3G2, 867.
Mydatoxine, 859, 361.
Mydine, 360, 867.
Mylabris cichorii, 522.
Myricin, 642.
Myricyl palmitate, 642.
Myristica argentea, 233, 236.
Myristica fatua, 236.

— fragrans, 232, 236.

— Malabarica, 236.
MyristiC acid. 235, 636. .

in butter, 89.

Myristicin, 42, 235, 238.
Myronic acid, 214.
Myrosin, 214.
Myroxylon pereira;, 438.

— toluifera, 439.
Myrrh, 470.
Myrrholic acid, 470.
Myrtle wax. 642.
Mysore cardamoms, 445.

— oil, 446.
Mytilotoxine, 359, 367.

Nagelvort on opium, 586.
Naphthol, 709.

— yellow, 184, 186.
Narcotine, 582.
Natal aloes, 508.
Nataloin, 503.

Nativelle's crystallized digitalin, 546.

Natural wines, 335.

Naylor & Bryant on belladonna, 518.

on ipecacuanha, 566.

Nelson on gineer, 198.

Nencki on flesh decomposition, 359.

Neosine, 401.

Neuridine, 859, 360, 366.

Neurine, 859, 360, 367.

Neutral spirit, 300, 304, 308.

Niehl on glycogen, 394.

Nitrate of potash in saffron, 24J..

— of silver, 697.
Nitric acid, 692.
Nitro-benzene, 258.

— cresylate of sodium, 242.
Nitrogen, determination of, in cheese,

86.

in pepper, 204.

Nitrohydrociiloric acid, 693.
Nitrogen in cocoa, 22.

— in vinegar, 252.

— separation of, in meat extract, 402.
Nitrogenous matter in malt, 340.
Nitrotoluene, 258.

Noodles, 184.

Nopalea coccinellifera, 238.

Nopel, 238.

Nutmeg, 232, 619.

— butter, 234, 235.
Nutmegs, adulteration of, 232.

Nutmegs, analysis of, 233.
— microscopic examination of, 233.
Nut oil in paprika, '-'13.
Nux vomica, 565, 574.

Oat flour, 172.
Oatmeal, 182, 187.
Oat starch, 169.
Oats in coffee, 83.
Oenanthic ether, 288.
Official French methods in wine analy-
sis, 380, 337.

— German methods in wine analysis,

316, 822, 323, 326.
Ogden on cloves, 238.
Ogden's analysis of cinnamon, 228.
Oil in microscopical analysis, 418.

— in paprika, 213.

— of almonds (essential), 257.
(fixed), 638.

— of aniseed, 606.

— of cajaput, 609.

— of caraway, 609.

— of castor, 639.

— of chamomile, 608.

— of cinnamon, 230.

— of cloves, 225, 266.

— of cod liver, 687.

— of copaiba, 448.

— of coriander, 610.

— of cubebs, 610.

— of croton, 634.

— of dill, 606.

— of eucalyptus, 611.

— of juniper, 612.

— 01 lavender, 618.

— of lemons. 261.

— of linseed, 636.

— of mace, 238.

— of mirbane, 258.

— of mustard, 217.

— of nutmeg, specific gravity of, 235.

— of olive. 111.

— of pimento, 227.

— of peppermint, 614.
table. 616.

— of pine, 619.

— of rose, 619.

— of rosemary, 620.

— of santal, 621.

— of spearmint, 618.

— of theobroma, 641.

— of turmeric, 246.

— of turpentine, 478, 622.

— preparation for microscopic analy-

sis, 420.
Olea Europcea, 111.
Oleic acid, 112, 235, 693.

— - in butter, 89.
Olein, 42, 112, 174.
Oleomargarine, 106.
Oleo-resiu in cayenne, 209.

736

INDEX.

Oleo-resin refractometer, 97.
Oleum anethi, 606.

— anisi, 606.

— anthem idis, 608.

— cadinum, 608.

— cajapnti, 609.

— carui, 609.

characters, 609.

— caryophylli, 610.

— cinnamoni, 610.

— copaibae, 610.

— coriandri, 610.

characters and tests, 610.

— cubebse, 610.

characters and tests, 610.

— eucalypti, 611.

determinations, 611.

— juniperi, 612.

characters and tests, 612.

— lavandulae, 613.

— limonis, 614.

— menthae, characters and tests, 614.
piperitse, 614.

viridis, 618. ,

— myristicae, 619.

— pimentae, 619.

— pini, 619.

— rosse, 619.

— rosmarini, 620.

— sinapis volatile, 622.

— terebinthinse, 622.
Olive oil, 111, 630, 639.

adulteration of, 112.

in paprika, 213.

soap, 646.

• — stones in pepper, 205, 206.
Opium, 501, 581.

— preparations, 586.
Optical values in molasses, 146.
Opuntia, 238.

Orange oil in oil of lemons, 261.

Ordonneau on brandy, 288.

Orellin, 246.

Orfila's method in meat analysis,

374.
Orizaba jalap resin, 482.
Orthc-hydroxy-benzoic acid, 677.
Orijza sativa. 169.
Osazones, 117, 118, 120, 121.
Osborne & Vorhees on flour, 174.
Ostertag on botulism. 361.
O'Sullivan on malt, 338.

— on tragacanth. 432.

— process for flour, 176, 177.
Otto of roses, 620.
Oudemann on chloroform, 685.
Over-proof spirit values, 274, 276, 277,

278.
Oxide of mercury test in wine analysis,
331.

— of silver, 697.
Oxyanthraquinone. 511.
Oxygenated ptonmines, 359, 367.
Oxymel of squills, 491.

Paal & Amberger on butter, 102.
Pabst on pepper, 206.
Palermo oils, 261.
Palladino on gum arable, 431.
Palmitic acid, 636, 693.

in butter, 89.

Palmitin, 42, 174.
Palm keinel oil, 101.

— nut oil, 27.
Panum on flesh, 358,
Papaver soinniferiim, 581,
Paprika, 212, 213,
Paracasein lactate, 87.
Para copaiba, 448,
Paracresol, 455.
Para-cresotic acid, 678,
Paracuraic acid ester, 508,
Paraformaldehyde, 332.
Para-hydroxy-benzoic acid, 678,
Paraldehyde, 709.
Paranuclein, nitrogen as, 86,
Paraphenetidin, 709.
Parasites in meat, 391, 392.
Paregoric, 588, 589,
Parenchymatous tissue, 460.

Paris Municipal Laboratory, analyses

of, 352.
Parry on cloves, 222, 223.

— on pepsine, 473.

Partheil & Van Haaren on camphor,

682.
Partheil's process in wine analysis, 327.
Parvoline, 359, 360, 366,
Paste annatto, 246.
Pasteur on bacteria in milk, 79.

— on wine analysis, 336.
Pastureau on vinegar, 253.
Patent still spirits, 287, 301.
Pathogenic bacteria in milk, 79.
Paul & Cownley on cayenne, 38.

on ipecacuanha, 569.

Paul on cocaine, 539.

Pavy's method for determining sugar,

126.
Peach kernel oil, 633.

— kernels, 257.
Pea flour, 172, 173.
Pear juice, 350.
Pearmain on araroba, 436.
Pea starch, 172,

Peas, 187,

— in coffee, 33.
Pectin, 357.

Penang benzoin, 440, 441,

Penfield & Sperry on boron in milk, 61,

Pennington's process for meat analysis,

389,
Pcntamethylcnc-diamine, 359, 365.
Pentosans, 23.

— determination of, 38, 206.
Peony in saffron, 242.
Pepper, 198.

INDEX.

737

*epper, adulteration of, 203, 204.

— analysis of, 199, 200.
Peppercorns, 205.

Pepper, microscopical examination of,
211, 212.

— microscopic examination of, 207.
Peppermint oils, table of characters, 616.

— Power & Kleber, 615.
Pepsine, 472.
Peptone, 472, 473, 474.
Peptones, 391.

— in meat extract, 405, 406.

— nitrogen as, 87.
Perrin on wine analysis, 335.
Perry, 352, 356.

— adulteration of, 357.
Peruresinotannol, 439.
Petit cidre, 352.

— Poire, 352.
Petri on flour, 178.
Petroleum, 646.
Pfeiffer on coca, 535.

Pfyl & Scheitz on saffron, 242.
Pharmaceutical codex. 162, 448.
Phaseolus vulgaris, 171.
Phellandrene, 611.
Phenacetin, 709.
Phenazone, 709.
Phenol, 238, 678.
Phenoloid compounds, 455.
Phenol-phthalein, 390.

— action on milk, 42.
Phenols, 453.
Phenylhydrazine, 609,

— compounds of sugar, 121.
Phenyl-propyl alcohol, 493.
Phloroglu3in, 426.
Phloroglucinol, 16.
Phosphates in cider, 353.
Photomicrograph of lemon oil, 265.
Phosphoric acid, 255, 693.

in beer, 346.

in vinegar, 252.

in baking powders, 187.

Phosphorus, 657, 709.
Phosphotungstic acid, 503.
Physeter macrocephalus, 645.
Physiological alkaloids, 358.
Phytolacca decnndra, 330.
Phytosterol, 630.
Phytosteryl acetate, 638.

test, 109.

for butter, 105.

Picallili, 243.

Picea excelsa. 477.

Picraconitine, 506.

Picric acid, 184, 186, 426, 511, 703.

Picrocrocin, 240.

Picrotoxine, 709.

Pierrserts' sugar formula, 134.

Piesse & Stansell on mustard, 214, 221.

Piettre on glycogen, 393.

Pilocarpine, 56 J, 570.

— hydrochloride, 570.

— nitrate, 570.

VOL. I.

Pilocarpus jaborandi, 569.
Pimaric anhydride, 477.
Pimenta officinalis, 227.
Pimento, 20-!, 212-3, 227.
Pimpiiiella anisnm, 606.
Pinene, 264, 438, 471, 622.

— nitroso-chloride crystals, 265.
Pinewood fibre, 423.

Piuth method for paste analysis, 185.
Pinus palustris, 480.

— ptimilio, 619.

— sylvestris, 477, 622.
Piperidine, 198.
Piperine, 198.

Piper longutn, 207.

— nigrum, 198.

— ofhcinarum, 207.
Pisum sativum, 172.
Pitch, 477.

Pix liquida, 477.
Plastering of wine, 328, 338.
Plaster of paris in bread, 180.
Platinum chloride, alkaloidal reagent,

503.
Poaya, 558.
Podophyllotoxin, 597.
Podophyllum, 595.
Podwyssozki & Kiirsten on podophyllum,

595.
Poisonous milk, 59.
Poivrette, 205.

Polarimetric determination of condensed
milk, 74.

of milk, 53.

of sugar, 128.

Polarization in wine, 313, 316.

— Clerget's process, 131.

— of cider vinegar, 256.
Polarized light, butter under, 106.

in butter examination, 105.

Polenske on butter, 102, 103.
Poly-arabinan-trigalactane-geddic acids,

433.
Polyphenols, 456.
Pomegranate, boric acid in, 356.
Poppy in saffron, 241, 242.

— seed oil. 112, 116, 632, 633, 634.
Port, 309, 310, 312, 313.

Porter, 338, 341, 342.

— on preserved meats, 377.
Potash, 426.

— preparation for microscopic analy-

sis, 420.

— soap, 647.
Potassii chloras, 657.

— iodidum, 654.

— nitras, 657.

— permanganas, 657.

— sulphas, 654.
Potassium bitartrate, 254.

— carbouate, 246.

— compounds, <10, 711, 712.

— glycyrrhizinate, 464.

— myronate, 214.

— sulphate, 329, 337.

738

INDEX.

Potato alcohols, 300.

— flour, 181, 184.

— spirit, 289.

— starch, 177.

Pot still process, 302.
Potted .meats, 381.
Powdered milk, 72.
Powders, insect pests in, 423.
Power & Moore on colocynth, 541.

on elaterium, 648.

Power & Galway on nutmeg, 235.

Power on jalap, 572.

Precipitation of levulose, Bryan, 131.

Prepared storax, 492.

Preparing sugar for polarimeter, 131.

Preservatives in beer, 348.

— in British preserved meats, 378.

— in foods, 385.

— in glass-packed meats, 382.

— in imported canned goods, 376.

— in milk, 60.

— in sausages, 384.

— in tinned meats, 376.
Preserved peas, 372.

" Process " butter, 105.
Proctor on saccharin, 674.

— on saffron, 242.

Prollius & Fliickiger on opium, 583.
Prollius' process for cinchona bark, 527.
Proof spirit, 274.
Propionic esters, 288.
Propyl alcohol, 286, 288.

— amine, 359, 364.

— guaiacol, 453.
Proteids, 160.

— determination of, in condensed milk,

74.

— determination of, in flour, 176.

— estimation of, 166.

— in beer, 346.

— of milk, determination of, 57.
Proteins, 389, 407.
Proteolytic bacteria, 81.
Proteose, 174.

Proteoses, 391, 408.
Protoplasm, 419.
JPrtmus armeniaca, 257.
Prussian blue, test for, 15.
Pseudocedrela kotchyi, 432.
Psychotria ipecacuanha, 556.
Psychotrine, 569.
Pterocarpus marsupium, 460.
Ptomaines, 358, 359.
• — in milk, 59.

— poisoning, 360.

— 8 paration of, 362.
Public Health Act, 383.
Puckner on asafcetida, 437.

Pulvis ipecacuanhse compositus, 567.
Putrescine, 359. 360, 365.
Pure alcohol, 273.

— golden syrup and glucose syrup,

contrasted table, 145.

— maple products, Bryan's analysis,

136.

Purified storax, 493.
Pyridine, 501.
Pyrocatechin, 456.

Quantitative colorimetric lead test, 668.

Quassia, 347.

Quercetin, 2.

Quinamine, 526.

Quince, boric acid in, 366.

Quinicine, 526.

Quinidine, 626, 631, 633. .

Quinine, 626, 533.

— detection of, 533.

— sulphate, 533.

— wine, 536.
Quinoline, 601.

— yellow, 184.

Raffinose, 144, 151.

Ramsay on chloroform, 686.

Rancidity in butter, 89.

Rape oil, 112.

Rapeseed oil, 632.

Raw's method in wine analysis, 334.

Raynaud's method in wine analysis,

327.
Reagents for microscopical examination

of foods and drugs, 424.
Rectified spirit, 273.
" Reduced extract " of wine, 337.
Reducing powers of sugar, 127,

— sugar, 122, 431.

in cider vinegar, 266.

in beer, 346.

Red wine, 309, 322.

Refractive index for fats and oils, 632.

of coffee, 37.

of beer, 345.

of gin, 309.

of oil of lemon, 263.

of sugar solution, 137.

— indices of butter, 98, 99.

of ethyl alcohol and methyl

alcohol, 283.

— values of butter, 96.
Refractometer number, 632.
Reichert on butter, 97, 100.
Reichert-Meissl values of butter fat, 97,

102.

for croton oil, 634, 636.

for fatty acids, 631.

Reichert Wollny value for fatty acids,

631.
Reichert s process for cocoanut fat, 103.
Reid on saccharin, 674.
Reinsch test for arsenic, 667.
Remijia, 526.

Remsen & Burton on saccharin, 674.
Renard on olive oil, 114.
Rennet, 78, 79.
Resin, 191, 196, 198, 243, 247. 2Gh, 435,

447, 478, 480, 481, 595, 596, 642,

644.

INDEX.

739

Resin in microscopical analysis, 418.

— in pepper, 203.

— value of jalap, 572.
Rhaberone, 538.
Rhamnose, 2.
Rhein, 598, 599.
Rheoclirysidin, 598, 539.
Rheochrysin, 598, 533.
Rheopurgarin, 538.
Rheopurgin, 598.
Rheum officinale, 597.

— palmabum, 597.
Rhubarb, 511, 512, 597.

— turmeric in, 244.
Rice, 182, 187.

— as adulterant in cocoa, 19.
Rice flour, 172, 181, 184.

in potted meats, 384.

— meal, 177.

— starch, 163.

Richardson on cayenne pepper, 208.

— on cloves, 223.

— on papper, 193.

— on rice, 183.

— on tannin value, 192.
Richardson's analysis of ginger, 195.
Riche & Bardy's method in alcohol

analysis, 280.
Richmond & Boseley on milk analysis, 57.

on milk, 63, 64.

on Trillat's test, 65.

on analysis of butter, 93.

Richmond & Harrisan's method in cider

analysis, 356.
Richmond & Hehner's analysis of milk,
52.

— on milk, 42, 43, 44, 45.

— on milk analysis, 55.
Richmond's analysis of cream, 69.
Ricinoleic acid, 639.

Rideal on formaldehyde in milk, 63.
Ridenour's analysis of cocoa, 18.
Riley on imitation coffee, 32.
Rceser on mustard oil, 219.
Rogerson on jalap, 572.
Roll annatto, 246.
Ropy milk, 45.
Roques on cider, 351.
Rosa ce7itifolia, 619.

— damascena, 619.
Rosemary oil, 620.
Rosenbladt & Gooch on milk, 61.
Rose's method in milk analysis, 51.

— process for determination of alcohol,

293.

Rose-Stutzer-Windisch method for de-
termining alcohol, 296.

Rosin, 477.

— oil, 637.

— spirit, 624.
Rosmarinus officinalis, 620.
Rosolic acid, 505.

Rotation in sucrose solutions, 132.
Rothenfusser's process in cream analy-
sis. 70.

Rottger on pepper, 204.

Royal Commission on potable spirits,

286, 287, 2S9, 301, 304, 307, 308.
Ru^geri on oil, 114.
Rum, 287, 304.
Rupp on milk analysis, 56.
Russian oil of turpentine, 622.
Rye, 177, 187.

— flour, 172.

— in coffee, 33, 37.

— starch, 168.

— whisky, 302, 303.

Saccharin, 121, 147, 273, 309, 310, 313,
387, 357, 465, 673.

— detection of, 674.

— determination of, 674.

— in cream. 08, 69.

— in beer, 348, 677.

— in cider, 349.
Saccharum, 121, 147.
Saccharomyces cerevisicB, 341.

— ellipsoideus, 309.
Saccharose, 117, 46 ), 466.
Sacchse on sugar reducing, 127, 128.
Safflower in butter, 93,

— in saffron, 242.
Saffron, 184, 240.

— adulteration of, 240.

— in butter, 91, 93.

— in Italian pastes, 183.

— microscopic examination of, 242.

— valuation of, 242.
Safrol, 235.

Sago as adulterant in cocoa, 19.

— starch, 170.
Salad oil, 111.

Sale of Pood and Drugs Acts, 1, 41, 67,
83, 85, 88, 9., 97, 187, 216, 802,
308, 309, 378, 512.
Salicin, 712.

Salicyhc acid, 69, 252, 313, 435, 498,
535, 677.

detection of, 679.

in beer, 348.

in cream, 70.

in saccharin, 676.

Salol, 712.
Sal volatile, 235.
Sandarac, 432.
Sand in chicory, 35.

— in pepper, 204.
Sangle-Ferriere & Cuniasse on alcohol

analysis, 282.
Santal oil. 621.

— wood (red), 241.
Santalum album, 621.
Santonin, 713.

Sapo animalis, 647, 658.

— durus, 647, 658,

— mollis, 647.
Saponification of butter, 95.

— value of fat, 626.

740

INDEX.

Saprine, 359, 360, 366.
Saprophytic bacteria in milk, 79.
Sauce, 488.

Sausage poisoning, 361.
Sausages, 386.

— analysis, 386, 387.

— colouring matter, 295, 396.

— examination of, 413.

— fats in, 387.

— nitrogen in, 388.

— starch in, 388.
Sawdust in coffee, 32, 33.
Sayre on senna, 484.
Scammony, 481, 571.
Scammonin, 572.
Schacht on chloroform, 685.
Schenck, 341, 372.

Schidrowitz & Kaye, whisky analysis,
302.

process in alcohol determination,

296, 297, 299.

— on opium, 585.

Schiff's reagent in milk analysis, 63.
Schindler's reaction for mace, 237.
Schittenhelm's method for meat extract

analysis, 410.
Schlegel's method for pastes, 185.
Schleibler's reagent, 503.
Schlicht's method for mustard analysis,

220.
Schmitz-Dumont test for tropeolins, 186.
Schmceger on milk analysis, 55.
Schneider on microscopic analysis, 422.
Schouten's reaction for aloes, 509.
Schreiber's process for hydrastis, 553.
Schryver on canned foods, 373.

— on tinned meats, 373.

Schulze's maceration mixture, 420, 426.

Scillain, 490.

Scillin, 490.

Scillipicrin, 490.

Scillitoxin, 490.

Scopola carniolica, 521.

Scopolamine, 602.

Scotch ales, 342.

Scotch whisky, 302.

Scott Smith on ipecacuanha, 667.

Scoville on tragacanth, 434.

Searl on extract of meat, 410.

Sebelien's process in milk analysis, 58.

Secale'cereale, 168.

Secondary constituents of spirits, 308.

Seidell on salicylic acid, 682.

Self & Greenish on cantharides, 524.

Selmi on flesh decomposition, 359.

Semmler on asafcetida, 438.

Semolina. 183.

Senkpiehl on botulism, 361.

Senna, 468, 483, 512.

— emodin, 484.
Sesame, 112.

Sesame oil, 109, 116, 632.
Sesquiterpene caryophyllene, 225

— limene, 261.
Shale naphtha, 624.

Shark liver oil, 628, 638.

Sharp on strophanthus, 604.

Shaw's "Essay on Distilling " 307.

Shea butter, 27.

Sherry, 309, 310. 312, 313, 567.

— plastering of, 329.

Shrewsbury & Knapp's process for cocoa-
nut fat, 103.
Siam benzoin, 440, 441.
Silent spirit, 288, 300, 301, 303.
Silica in chicory, 35.
Simon on rum, 307.
Sinalbin, 214, 221.
Sinapine sulphate, 221.
Sinapis alba, 213.

— nigra, 213.

— junca, 213.
Sinigrin, 214.
Sitodrepa panicea, 423.
Silver, compounds of, 697.

— grain cochineal, 238, 239.
Sjerning's method in meat extract analy-
sis, 408.

Skimmed milk, 41. •*

cheese, 86.

Smith on bismuth salts, 698.

Soap, 646.

Soaps, 626.

Socotrine aloes, .''07, 510.

Sodii hypophosphis, 657.

— iodidum, 654.

— phosphas, 654.
Sodium, 715.

— compounds, 713, 714, 715.

— sulphindigotate, 11.
Soft paraffin, 646.

— soap, 646.

Soleil-Dubosq polarimeter, 130.
Soleil- Ventzke-Sch eibler polarimeter,

133.
Solereder on microscopic analysis, 422.
Solid and liquid glucoses, composition

of, 149.

— glucose, 147, 148.
Sonnenschein on flesh decomposition,

358, 360.
Sonnenschein's reagent, 502.
Sorghum plant, 117.
Soudan glycerine. 424.
Souring of milk, 78.
Sour milk, 80.

estimation of fat in, 77.

Soxhlet apparatus for milk fat, 47.
Soxhlet's aerometric apparatus, 48.

— fat table, 49.
Spaeth on nutmeg. 234.
Spaghetti, 183, 184.
Spanish flies, 522.

— liquorice, 466, 467.
Sparkling wines, 311.
Spermaceti, 642, 645.
Spices, 191.

Spirit indication, 344.

— of camphor, 6S3.

— of nitrous ether, 485.

INDEX.

741

Spirits of the pharmacopoeia, 498.

— of wine, 284.
S2)orodo7ie7fia casei, 83.
Spurious ipecacuanha, 558.

Squibb's method for cinchona bark,

427.
Squills. 490.
" Standard ' lard, 107.

— polarimeters, 130.
Stannous chloride, 649.

Starch, 117, 161, 162, 166, 189, 204, 224,
339, 424.

— as adulterant in cocoa, 19.

— determination of, in flour, 176.
of spices, 191.

— glucose syrup, 145.

— in annatto, 246.

— in catechu, 444.

— in cocoa, 20.

— in cream, 69.

— in microscopical analysis, 418,

— microscopic examination of, 167.

— in mustard, 216.

— in pepper, 211.

— in potted meats, 384.

— in rice, 182.

— in turmeric, 245

— removal of, in microscopic analysis,

420.

— sugar, 119.

Stas-Gautier method of ptomaine separa-
tion, 363.
Stearic acid, 639, 642, 644, 693.

in butter, 89.

Stearin, 42, 108. 109, 110, 112.

— of cocoanut oil, 27.
Stebbings on pepsine, 476.
Sterculia urens, 434.
Stevenson on pepper, 208.

Stierlin's method in wine analysis, 327.

Stockholm tar, 477.

Stoddart on pepper, 203.

Stokes & Bodmer's analysis of condensed

milk, 74.
Storax, 439, 492.
Stout, 337, 338, 341.
Stramonium, 602.
Strength of sugar solutions, sp. gr.

table, 143.
Strophanthin, 603, 605.
Strophanthus, 603.
Strychnine, 574, 578.

— separation from brucine, 577.
Strychnos nux vomica, 674.
Stutzer on meat extract, 401.
Stiitzer's analysis of cheese, 88.
Styracin, 439.

Styrax, 440, 442.

— calamitus, 492.
Styrol, 493.

Substitutes for cocoa butter, 27.
Succinic acid, 313.

in wine, 334.

Sucrate of lime in cream, 69.
.Sucrose, 76, 117, 123, 349, 350, 356.

Sucrose, analysis, 136.

— impurities, percentages of, by speci-

fic gravity table, 142.

— in chocolate, 29.

— in cream, 70.

— in genuine golden syrup, 147.

— rotation, Clerget's formulae, 133.
inversion values, Herzfeld, 138.

Suet, 111, 646.

Sugar, 117, 250, 266. 270, 307, 342, 692.

— beet, 117.

— candy, 118.

— cane, 117.

— adulteration 142.

— as adulterant in cocoa, 19.

— as reducing agent, 122.

— in cocoa, 29.

— in mustard, 217.

— in vinegar, 253.

— in wine, 313, 314, 316, 324, 325, 837.

— of milk, 154.

specific rotatory power, 154.

~ polarimetric dttermination of, 128.

— relative reducing power, 127.

— solutions, strength of by sp. gr., 148.

— vinegar, 248, 256.
Sulphamido-benzoic acid, 674.
Sulphates in wine, 328.
Sulphate of atropine, 521.

— of lime in saffron, 241.

— of soda in saffron, 241.
Sulphates in beer, 348.
Sulphomolybdic acid, 504.
Sulphonal, 715.
Sulphur, 311, 715.

— iodide, 715.

— prsecipitatum, 658, 662.

— sublimatum, 662.
Sulphuric acid, 693, 694.

baking powders, 187, 189.

Sulphurous acid, 383, 384, 694.

in cider, 349.

in wine, 329.

Sumatra benzoin, 440, 441.
Sunflower oil, 632.
Sutton on mustard, 221.
Swedish oil of turpentine, 622.
Sycamore leaf in tea, 15.
Sylvestrene, 622.
Sylvic acid, 478.
Syntonin, 405, 474, 476.
Syrian tragacanth, 432.
Syrup, 118.

— of hypophosphites, 580.
Syrupus glucosi, 654.

Table comparing polarimeters, 131.

— of chemicals, 689.
Tables of water in syrups by refrjictive

indices, 138-141.
Tahiti vanilla, 267.

Taka-diastase in malt analysis, 158, 159.
Takamine in malt, 158.

742

INDEX.

Talc in rice, 182.
Tallow, 642.

— as an adulterant of cheese, 85.
Tamba on botulism, 361.
Tampico jalap, 571, 573.

resin, 481, 482.

Tankard on boron in milk, 62.
Tankard's analysis of gelatine, 414.
Tannic acid, 1, 211, 313, 444, 504, 691, 694

in cider, 352.

in cloves, 224.

— in wine, 332.

Tannin, 2, 3, 8, 10, 314, 350, 357, 424.

— test, copper acetate process, 10.
Eder's process, 10.

Fletcher & Allen's, 10, 11.

gelatine process, 13.

hide powder, 13.

Janke's, 10.

lead acetate process, 10.

Lowenthal's, 11.

Procter's, 14.

Tatlock & Thompson, 14.

Vignon, 13.

— value of spices, 192.
Tanret on ergot, 550.
Tapioca starch, 171.

Tartaric acid, 217, 250, 254, 311, 313,
314, 694.

in wine, 331, 332.

baking powders, 187, 189.

Tartrates in wine, 311.
Tartrazine, 184.

Tatlock & Thompson, determination of
caffeine, 9.

on coffee, 34.

Taylor on scammony, 481.

— podophyllum, 597.
Tea, 1.

— adulterants, 5.

— adulteration of, 1.

— albuminou§ matter, 2

— aqueous extract of, 7.

— author's examinations for ash, 7.
table of alkalinity of ash, 6.

— carbohydrates, 2.

— cellulose, etc., in, 2.

— Ceylon, 2.

— China, 3.

— chlorophyll, 2.

— customs laboratory's table of ash

and siliceous matter in, 6.

— diagnostic characters of structure,

15.

— Eder's composition of, 2.

— essential oil, 2.

— exhaustion of, 8.

— infusions, Geisler, 34.

— Natal, table of analysis. Imperial

Institute, 5.

— Nyasaland, table of analysis, 4.

— structural and microscopic examina-

tion of, 15.

— table of difference of ash between

genuine and exhausted tea leaves, 7.

Tea-table extractive matter in, 9,

— Tatlock & Thompson's process of

aqueous extract, 8.

— test for indigo, 15.

— venation and serration of leaf, 15.

— Wanklyn's table of ash in tea

adulterants, 7.

— water, soluble ingredients in, 8.
Teed on lead determination, 670, 671.
Tenareze brandy, 287.

Terebene, 493.

Terpineol, 446.

Tertoni on saccharin, 676.

Tests for chemicals, 689.

Tetrahydro-proto-catechuic acid, 444.

Tetramethylene-diamine, 359, 365.

Thalleoqum reaction, 504.

Thea, 1.

— assamica, 1.

— bohea, 1.

— sinensis, 1.

— viridis, 1.
Theine, 1.
Theobroma, 16, 17.
Theobromine in cocoa, 22.
Thiosinamine, 221.
Thompson on boron in milk, 61.
Thomson's process in cider analysis, 353-
Thorpe & Holmes on alcohol, 284.

on alcohol analysis, 282, 283.

on spirits, 498.

— on chemical tests, 668.
Thresh on cayenne pepper, 208.

— on pepper, 212.
Thus, 480.
Thymol, 389, 716.

Tiemann & Haarman on vanilla, 267.
Tiglic aldehyde, 457.

— acid, 608.
Tincture of aconite, 500.

— of aloes, 512.

— of asafcetida, 437.

— of belladonna, 517.

— of benzoin, 441.

— of camplior, 588.

— of cannabis indica, 443.

— of cantharides, 525.

— of cardamoms, 447.

— of catechu, 444.

— of cinchona, 534.

— of colchicum, 541.

— of conium, 545.

— of digitalis, 548.

— of gelsemium, 550.

— of hydrastis, 554.

— of hyoscyamus, 555.

— of jaborandi, 569.

— of jalap, 574.

— of myrrh, 472.

— of nux vomica, 578.

— of opium 586.

— of paregoric, 588.

— of podophyllum, 597.

— of quinine, 534.

— of senna, 484.

INDEX.

743

Tincture of squills, 491.

— of stramonium, 603.

— of strophanthus, 605.

— of tolu, 440.

Tinctures, standards for, 495, 496.

Tin in preserved meats, 373.

Tinned meats, 378.

Tinnivelly seuna, 483.

ToUens on analysis of cocoa, 23.

Tolman & Hillyer on alcohol analysis,

298.
Toluresinotannol, 440.
Tonka beans, 268.
Tortelli on olive oil, 114.
Tous les mois arrowroot, 170.
Toxigenes, 361.
Tragacanth, 432.

— adulteration of, 433.
Tragacanthose, 433.

Treacle, ash of (Winter Blyth), 144.

— or molasses, 144.
Triacetyl chrysarobin. 436.
Trichina S2nralis, 377.
Trichinosis, 361.
Triethylamine, 359, 364.

Trillat's process in wine analysis, 326.

— test for formaldehyde, 65.
in alcohol analysis, 281.

Trimethylamine, 359, 364.
Trimethyl-xanthine, 2.
Trimorphine, 595.
Trimyristin, 235.
Tiinidad cocoa bean, 17.
Triticum, 167.

— sativum, 173.
Tropeolins, test for, 186.
Tropococaine, 535.

Trowbridge & Grindley on meat analy

sis, 389.
Truchon on saccharin, 676.
Truelle on cider analysis, 350.

— on pear juice, 357.
Truxilline, 535.

Tschirch & Edner on rhubarb, 699.

— on aloes, 509, 510

— on myrrh, 470.
Tubercle bacilli in milk, 45.
Tucholka on myrrh, 471.
Turmeric, 184, 191, 217, 238, 243.

— adulteration of, 245.

— analysis of, 244.

— detection of in rhubarb, 600.

— in butter, 91, 93.

— in mustard 216.

— microscopic examination of, 245.

— oil of, 246.

— test for, 186, 244.
Turpentine, 264, 265, 477.

— in oil of lemons, 261.

— oil, 624.

Twitchell & Gladding on resin, 479.

Types of sugar, 135.

Tyrer & Wertheimer on terebene, 494.

Tyrosamine, 360.

Tyrosine, 401.

Tyrotoxicon, 59, 60.
Tyrotoxine, 367.

Ultramarine in rice, 182.
Umbelliferone, 435.
Umney's analysis of mace, 237.
Umney on aniseed, 607.

— on belladonna, 518.

— on cloves, 226.

— on cochineal, 239.

— on oleum pini, 619.

— on podophyllum, 595.

— on drugs, 427.

Unaltered proteids in meat extract, 404.
Uiwaria gambier, 444.
Under proof, 274, 275, 276.
Unfermentable carbohydrates, 148.
United States Meat Inspection Act, 377,
379.

pharmacopoeia, 280, 612, 687.

Unsaponifiable matter, 627.
Urginea scilla, 490.
Uric acid, 2.

Valenta's test for butter fat, 104.
Vandam's process for cocoanut fat, 103.
Vanderplaten on nutmegs, 233.
Van Ermengem on botulism, 361.
Van Ketel & Antusch on cheese analy-
sis, 87.
Vanilla, 16, 265.

— adulteration of. 160.

— analysis of, 266, 270.

— essence of, 268.

— planifolia, 265.

Vanillin, 265, 266, 267, 268, 269, 437,
439, 440, 493.

— in chocolate, 28.

— wtffa-nitro-benzhydrazone, 267.
Van Rombrugh on tea, 2.
Vaporimeter, 315.

Vasey on alcohol analysis, 295.

— on brandy, 287.

Vasey's analysis of potable spirits, 303,

308.
Vaughan on poisonous milk, 58.
Venation and serration of leaf in tea, 15.
Venetian turpentine, 622.
Ventzke scale polarimeter, 491.
Ventzke's polarimeter, 130.
Verley & Boising's method for cloves,

226, 227.
Vermicelli, 183, 184.
Victoria yellow, 184, 186.

in butter, 93.

Vieth & Richmond on milk, 43.
Vieth's analy.sis of cream, 69.
Vigneron's process for cinchona assay

529, 5i4:.
Villavecchia on olive oil, IIG
Vinegar, 191 217, 247.

744

INDEX.

Vinegar Act of 1818, 250.

— adulteration of, 250.

— analysis of, 249.

— of ipecacuanha, 566.
— ' of squills, 491.

— preservatives, 252.
Vinum colchici, 641.

— ipecacuanhas, 567.
Violette on butter fat, 89.
Vitali on salicylic acid, 680.
Vitiatine, 401.

Vogel op flour, 178.
Vogl on microscopic analysis, 422.
Voigtlander on lard, 109.
Volatile ether extract, 191.

— fatty acids of butter, 97.

— oil, determination of, 192.
Volhard's method for mustard oil, 219.

Wagner's reagent, 503.

Wallach on lemon oil, 264.

Waller & Martin on mustard analysis,

216.
Wallis on capsicum, 21X.
Walnut shells in pepper, 206.
Wanters on saffron, 242.
Warden & Bose on ptomaines, 368, 369.
Warington's colorimetric test for lead,

668, 670.
Water in alcohol, 280.

— in molasses, 145.

— in syrups, tables of, 137-41.

— preparation for microscopical analy-

sis, 418.
Watts & Tempany's analysis of con-
densed milk, 75.
Waxes, 632.

Weber on cardamoms, 446.
Weil on bread, 181.
Weiss' tables for wine analysis, 323.
Werner-Schmidt process for sourmilk,

77.

in analysis of milk. 50.

Werner-Schmidt's method in cheese

analysis, 85.
Wneat as adulturant in cocoa, 19.

— bran, 187.

— flour, 172.

in mustard, 216.

— in coffee, 32, 33, 37.

— starch, 167.

Whey, examination of, 77.
Whisky, 287, 299, 301.
White mustard, 214, 215, 221.
oil, 221.

— on aromatic spirit of ammonia,

484.

— pepper, 198.

— wine, 309.
vinegir, 253.

White wines, 322.

Wijs' value for fat, 628, 629.

Wild mace, 236.

Wiley & Ewell on milk analysis, 56.

Wiley on alcohol. 284.

— on milk analysis, 55.
Wilson on flour, 179.
Wimmel on ipecacuanha, 558.
Windisch on wine.

Wine, 217, 309.

— adulteration of, 311.

— analysis of, 313.

significance of results, 336.

— classification of, 309.

— vinegar, 248, 253.

Winton & Bailey on vanilla, 269.

— & Lott on vanilla, 271.

— & Silverman on vanilla essence,

268.

— on cloves, 223.

— on microscopic analysis, 422.
Winton's analysis of cinnamon, 228,

— lead number, 192, 256.
Wirthle on canned foods, 374.
Wollny on butter, 97.

— on milk analysis, 53.
Woll's analysis of cheese, 83, 84.
Wood on ergot, 550.

Wool fat, 632.

— wax, 633.
Wort. 339, 340.

— colour of, 339.

Wursts or German sausages, 386.

Xantho-creatinine, 401.

Xanthin bases, determination of, 409.

Xanthine, 401.

Xylanbassoric acid, 433.

Xylene, 396.

Xylol balsam, 424.

Xylose, 23, 433.

Yeast, 309, 337, 341, 342

— extract, 410.

— illustrations of, 341.
Yoghourt, 80.

Young on bread, 180.

Zea mafs, 168.

Zeisel & Fanto's process in wine analy-
sis, 328.
Zeiss-Abb6 instrument, 632.
Zeiss' butyro-refractometer, 96.
Zinc compounds, 716.
Zingiber officinale, 193.
Zueler on flesh decomposition, 358, 360.

ABtKDEEN : THE UNIVERSITY PRESS

.-^

University of Toronto

library

—5

CD
O

DO NOT /^

iH H

REMOVE #

THE //

rt CO

■g^

CARD

1-3 U

CO ^

FROM ^

•» o •

THIS \

POCKET \^

Acme Library Card Pocket

Under Pat. "Ref. Index FUe"

cfl

Made by LIBRARY BUREAU

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Food and drugs"

See other formats