GIFT OF

CAROTIN

THE PRINCIPAL NATURAL YELLOW
PIGMENT OF MILK FAT

CHEMICAL AND PHYSIOLOGICAL RELATIONS OF

PIGMENTS OF MILK FAT TO THE CAROTIN

AND XANTHOPHYLLS OF GREEN

PLANTS

BY

LEROY SHELDON PALMER, B.S. in Ch.E., M.A.

SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

IN THE

GRADUATE SCHOOL

OF THE

UNIVERSITY OF MISSOURI

1913

CAROTIN

THE PRINCIPAL NATURAL YELLOW
PIGMENT OF MILK FAT

CHEMICAL AND PHYSIOLOGICAL RELATIONS OF

PIGMENTS OF MILK FAT TO THE CAROTIN

AND XANTHOPHYLLS OF GREEN

PLANTS

BY

LEROY SHELDON PALMER, B.S. in Ch.E, M.A.

SUBMITTED IN PARTIAL, FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

IN THE

GRADUATE SCHOOL

OF THE

UNIVERSITY OR MISSOURI

1913

TABLE OF CONTENTS

PART I— HISTORICAL

PAGE

I. INTRODUCTION 313

1. The Pigment of Butterfat as a Factor in the Colora-

tion of Milk 316

2. Object of Present Investigations 316

II. HISTORICAL.

1. The Pigments of Plant Origin.

(a) The Carotins.

(1) The Pigment of the Carrot , 317

(2) The Carotin of Green Plants 318

(3) Carotin in Flowers, Fruits and Seeds . . 321

(b) The Xanthophylls 322

2. The Pigments of Animal Origin.

(a) The Luteins 324

(b) The Chromophanes 325

(c) The Lipochromes 326

(1) The General Properties of the Lipo-

chromes 328

(2) The Lipochromes of Algae, Fungi, and

Bacteria 328

(3) The Lipochrome of Egg Yolk 329

3. The Physiological Relation Between Plant and Ani-

mal Lipochromes 330

III. SUMMARY 332

IV. BIBLIOGRAPHY 334

PART II— RELATIONS OF MILK FAT PIGMENTS TO CARO-
TIN AND XANTHOPHYLLS OF GREEN PLANTS

I. INTRODUCTION 339

II. METHODS OF ISOLATION 340

1. General Properties of Butter Fat Pigments 341

3

S99518

PAGE

III. METHODS OF IDENTIFICATION 342

1. Relative Solubility of Carotin and Xanthophylls 342

2. Adsorption Properties of Carotin and Xanthophylls . . 344
(a) Carotin and Xanthophylls of Green Plants .... 346

IV. IDENTIFICATION OF BUTTER FAT PIGMENTS 350

1. Standardization of Absorption Bands 355

V. CHARACTER OF PIGMENTS IN DIFFERENT BUTTER FATS . . . 356

1. Light Colored Fats. 2. After Carrot Feeding 357

3. Colostrum Milk Fat 358

4. Crystalline Form of Carotin from Butter Fat 359

VI. RELATION BETWEEN COLOR OF MILK FAT AND FOOD OF Cow 360

1. Yellow Pigments of Common Cattle Feeds 360

(a) Cottonseed Meal and Cottonseed Hulls 361

(b) Bleached Alfalfa Hay 362

(c) Yellow Corn 362

(d) The Carrot 362

2. The Feeding Experiments 364

VII. RELATION BETWEEN COLOR OF MILK FAT AND BREED OF

Cow 377

VIII. DISCUSSION OF RESULTS 383

IX. SUMMARY 386

X. BIBLIOGRAPHY 387

PART III— PIGMENTS OF BODY FAT, CORPUS LUTEUM
AND SKIN SECRETIONS OF THE COW

I. INTRODUCTION 391

II. METHODS OF IDENTIFICATION 392

III. THE PIGMENTS OF THE BODY FAT 394

1. Methods of Isolation 395

2. Identification of Pigments 395

(a) Experiment I 395

(b) Experiment II 396

4

PAGE

IV. RELATION BETWEEN COLOR OF BODY FAT AND FOOD OF

Cow 397

V. RELATION BETWEEN COLOR OF BODY FAT AND BREED OF

Cow 399

VI. THE PIGMENTS OF THE CORPUS LUTEUM 402

VII. THE PIGMENTS OF WAXY SECRETIONS OF JERSEY Cows . . 405

VIII. BODY FAT AND BLOOD SERUM PIGMENTS OF NEW-BORN

CALF 406

IX. DISCUSSION OF RESULTS 408

X. SUMMARY 410

XL BIBLIOGRAPHY 411

PART IV.
A. THE YELLOW PIGMENT OF BLOOD SERUM

I. INTRODUCTION 415

II. METHODS OF IDENTIFICATION 416

III. METHODS OF ISOLATION 417

IV. CHEMICAL IDENTIFICATION OF PIGMENT 418

V. DISCUSSION OF EXPERIMENTS 421

VI. PHYSIOLOGICAL RELATION BETWEEN CAROTIN OF BLOOD

SERUM AND FOOD OF Cow 422

VII. TRANSPORTATION OF CAROTIN AND XANTHOPHYLLS BY

BLOOD 427

VIII. THE HIGH COLOR OF COLOSTRUM MILK FAT 435

5

PAGE

IX. DISCUSSION OF RESULTS 437

X. SUMMARY 438

XI. BIBLIOGRAPHY 439

B. CAROTIN AND XANTHOPHYLLS DURING DIGESTION

I. INTRODUCTION 441

II. METHODS OF STUDY 441

III. ACTION OF DIGESTIVE JUICES 442

IV. CHARACTER OF PIGMENTS ALONG DIGESTIVE TRACT 444

V. THE EXCRETED PIGMENTS 445

VI. DISCUSSION OF RESULTS 445

VII. SUMMARY AND BIBLIOGRAPHY 446

C. THE PIGMENTS OF HUMAN MILK FAT

I. INTRODUCTION 447

II. EXPERIMENT I 447

III. EXPERIMENT II 448

IV. DISCUSSION OF RESULTS 449

V. SUMMARY . . 450

BIOGRAPHY 451

CAROTIN— THE PRINCIPAL NATURAL YELLOW PIGMENT OF

MILK FAT.

Its Relations to Plant Carotin and the Carotin of the Blood Serum,
Body Fat and Corpus Luteum.

LEROY S. PALMER AND C. H. ECKLES.

The investigations dealing with the natural yellow pigment of
milk fat will appear as a series of four bulletins as follows :

Part. I. A Review of the Literature Concerning the Yellow Plant
and Animal Pigments. Missouri Agricultural Experiment
Station Research Bulletin No. 9.

Part II. The Chemical and Physiological Relation of the Pigments
of Milk Fat to the Carotin and Xanthophylls of Green Plants.
Missouri Agricultural Experiment Station Research Bulletin
No. 10.

Part. III. The Pigments of the Body Fat, Corpus Luteum and
Skin Secretions of the Cow. Missouri Agricultural Experi-
ment Station Research Bulletin No. n.

Part IV. (A). The Yellow Lipochrome of Blood Serum. (B.)
The Fate of Plant Carotin and Xanthophylls During Diges-
tion. (C). The Pigments of Human Milk Fat. Missouri Ag-
ricultural Experiment Station Research Bulletin No. 12.
The present paper is the first of the series. As indicated it will
"be confined entirely to a review of the extensive literature in regard
to the yellow plant and animal pigments.

Part II will be a report of the chemical identification of the milk
fat pigment. It will also include a number of investigations showing
the relation between the amount of pigment in the milk fat and the
character of the ration and the breed of the cow.

Part III will consist of the data showing the chemical identification
of the pigments mentioned. Data will also be presented showing the
relation between the color of the body fat and the character of the

(313)

314 MISSOURI :AG$.^EXF. STA. RESEARCH BULLETIN No. 9.

ration and the breed of the cow. A brief experiment will also be
reported showing the absence of these pigments in the body of a new-
born Jersey calf.

Part IV (A) will report the chemical identification of the blood
serum pigment. It will show how blood carries the pigment and what
effect the character of the ration has upon the amount of pigment
carried by the blood and the amount secreted in the milk at the same
time. A brief study of the cause of the high color of colostrum
milk will also be reported. (B) This will consist of the report of a
few investigations relative to the fate of the carotin and xanthophyll?
of plants during their passage through the cow's body. (C) The
experiments reported here will show the character of the pigments
of human milk fat.

CAROTIN, THE PRINCIPAL YELLOW, P^tfMEN'rpf Miiac FAT. 315

A REVIEW OF THE LITERATURE CONCERNING THE YELLOW
PLANT AND ANIMAL PIGMENTS.

It has been the custom for generations to judge the quality of
dairy products to a large extent by their yellow color. This has been
carried to such an extent that the manufacturer of butter, whether
it be on a large or small scale, finds it impossible to market butter that
does not have a standard yellow color. The consumer of milk or cream
as a rule looks upon a yellow color as indicating the richness and quality
of the product. Although it is well known that the color has no
relation to the food value of milk or cream, the popular prejudice
is so strong that the producer of market milk has to take it into
account and try to supply a product with as much natural yellow
color as possible.

During part of the year, namely during the spring and early
summer and usually also in the early fall, the fresh green feeds which
the cows receive give the shade of yellow to the milk fat which the
consumer demands. During the winter months, or in summer if the
pastures become dry, this yellow color is wholly or in part absent
from the milk fat, and the butter manufacturer is then forced to
color the butter artificially, in order to maintain the required standard.

It is generally accepted as a fact that the breed of the cow has
a pronounced relation to the color of the milk fat and that the Guernsey
and Jersey breeds rank first in this respect. The breeders of this class
of cattle have emphasized this characteristic as one of the strong points
of their respective breeds. This characteristic of Guernsey and Jersey
breeds, as compared with the Holstein and Ayrshire breeds, has been
generally attributed to physiological differences. According to this
view, Guernsey and Jersey cattle are able to produce a higher colored
fat due to some inherent quality, just as they are able to produce a
higher percentage of fat in their milk. It is a well-known fact that
the skin and the secretions of the skin of Guernsey and Jersey cattle
have a higher yellow color than other breeds, and this characteristic
is looked upon by cattle breeders as an indication of the ability of
animals of these breeds to produce highly colored milk fat.

The body fat of Guernsey and Jersey cattle is also characterized
by a high yellow color and for this reason beef from these animals
is often looked upon with disfavor by the butcher and the consumer.

That the yellow color of butter has a relation to its market value
is shown by the fact that "color" has a place on the standard butter
score cards with a value of fifteen out of one hundred points. The

316 Mis£oOtfi;A^k.«^i>; STA: RESEARCH BULLETIN No. 9.

oleomargarin manufacturers have also recognized the value of color
and, so far as the law has permitted, have made a practice of coloring
oleomargarin in imitation of butter. When the law placed a tax on
artificially colored oleomargarin, or in some cases prohibited it entirely,
the manufacturers began using only the highest colored beef fats
that could be bought or mixed the oleomargarin with butter having
a high natural color, in order to produce the color they sought.

The Pigment of the Butter Fat as a Factor in the Coloration of Milk.

The more or less yellow color of cows' milk which is especially
evident in the cream and butter has not been attributed in all cases
to the same pigment. On the one hand a few authors have stated
that the pigment of butter is manifested in the familiar yellow color
of milk whey. This view originated with Blyth l who called the whey
pigment lactochrome and the view has found its way into a number
of texts. On the other hand a larger number of authors have ignored
the whey pigment and considered the lipochrome-like pigment of the milk
fat to be the only factor causing the yellow color of cream and butter.

The investigations which were carried on in this laboratory have
been the first to point out that the whey pigment and the butter fat pig-
ment are not identical but are distinct substances ; and that both are of
importance in causing the yellow color of milk. The pigment of the
butterfat is the more important of the two, however. The pigment
of the whey is of secondary importance, and is of an entirely different
nature. Its probable identity with urochrome, the specific urinary pig-
ment, has recently been shown by one of us.2

Object of the Present Investigations.

The present investigations were undertaken primarily to study the
chemical nature of the yellow butterfat pigment and to classify it from
a scientific standpoint. At the same time information was gathered
with the hope of ascertaining to what extent the generally accepted
views concerning the color of milk fat are correct in order to establish
a scientific basis for the subject which would be of value to those
interested in the handling of dairy products in a commercial way.

In the principal part of the investigation it was sought; (i) to
show the chemical and if possible the physiological relation of the
butter fat pigment to similar animal pigments such as the

1. A. W. Blyth, "Foods. Their Composition and Analysis" Text, 4th
Edition 1896, p. 239.

2. Lactochrome: The Yellow Pigment of Milk Whey, etc., by Leroy S.
Palmer and Leslie H. Cooledge. Missouri Agricultural Experiment Station
Research Bulletin No. 13;; Jour. Biol. Chem. XVII, p. 251 (1914).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 317

corpus luteum pigment, the body fat pigment, and the blood serum
pigment; and (2) to show the chemical and physiological relation of
the butterfat pigment to the carotin and xanthophylls of green plants.
In the secondary part of the investigation it was sought to study
the influence of certain factors which have both practical and scientific
bearing upon the color of the butterfat, among which are the breed
of the animal and the character of the ration, the latter in connection
with the chemical and physiological studies indicated above.

THE PIGMENTS OF PLANT ORIGIN.

The earliest researches on plant pigments dealt with the green
pigments. Caventon first called them chlorophyll in 1817. His work,
however, was preceded by the pioneers in this field, among which the
names of Grew, whose work is dated 1682, and Rouelle, Meyer,
Fourcroy, Berthellot, Senebier, Proust and Vanquelin are of historical
interest.

The Carotins.

The Pigment of the Carrot. The yellow pigment of the cultivated
carrot (Daucus Carota) has long been of interest to botanists and
chemists, the investigations of this body having extended over almost
one hundred years.

Wachenroder 1 was the first investigator of the carrot pigment.
He isolated it and called it Karotin. The work of Vanquelin and
Bouchardat 2 soon followed and a little later Zeise 3 took up the study.
He obtained the first crystals and assigned to them the chemical formula
C5 H10 or 10 (C5 H8).

Husemann 4 was the next investigator. He found six per cent
of oxygen in his pure preparation and gave the pigment the formula
C18 H24 O. A secondary pigment which he thought always accom-
panied the carotin in small amounts, he named hydrocarotin and gave
it the formula C18 H30 O.

It is to Arnaud 5 however that we are indebted for the first
thorough research in regard to the carrot pigment carotin. The crys-
tals which he obtained were flat, rhombic-shaped crystals, red orange
by transmitted light, and greenish blue by reflected light. They melted
at 168° C. He showed beyond a doubt that the pigment was simply

1. Dissertatio de Anthelminticis Gottingen 1826 ; also Geigers Magaz. Pharm.
33 p. 144 (1831); also Berzelius Jahresber. 12 p. 277 (1833).

2. Schweizg. Jour. Chem. 58, p. 95 (1830).

3. Lieb. Ann. 62 p. 380 (1847); Annal. Chem. Phys. (3) 20, p. 125 (1847).

4. Lieb. Annal. 117 p. 200 (1860).

5. Compt Rend. 102 p. 1119 (1886), p. 1319 (1887); Jour. Pharm. Chim.
14 p. 149 (1886).

318 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

an unsaturated hydrocarbon. He gave it the formula C26 H38 and the
iodine derivative the formula C26 H38 I2.

Eiiler and Nordenson 1 report the most recent investigations in
regard to the carrot pigment. They found their crystalline prepara-
tion to be mixed with crystals of xanthophyll ; they also showed that
the belief often advanced that carotin is chemically related to choles-
terol, is unfounded.

The Carotin of Green Plants. — Arnaud 2 was one of the first in-
vestigators to show that the carrot carotin is identical in properties
with a yellow constituent of chlorophyll, although the existence of
this yellow constituent of chlorophyll had long been the subject of
investigation.

Berzelius 3 first sought to isolate a yellow pigment from autumn
leaves by extracting with alcohol. He called it "Blattgelb" or xantho-
phyll, and expressed the belief that the pigment pre-existed along with
the green coloring matter of the leaf.

The subject subsequently received the attention of many investi-
gators. Fremy,4 Michels, Millardet, Miiller, Tinisnsseff, Gerland, Ran-
nenhoff, Askennasy, Stokes, Sorby,5 Tschirch,6 Kraus,7 Filhol,8
Hansen,9 Conrad,10 Wiesner,11 and many others took up the investiga-
tion.

Fremy designated the yellow pigment Phylloxanthin. Filhol
noticed that by treating crude alcoholic chlorophyll solutions with ani-
mal charcoal it was possible to remove the green constituent of the
mixture leaving a yellow colored solution, the color of which he be-
lieved was due to a pre-existing pigment or pigments associated with
the green one. Kraus confirmed the observations of Filhol, and was
the first to notice that when an alcoholic solution of chlorophyll is
shaken with benzoline (petroleum ether) the alcohol retains the yellow
coloring matter, the benzoline taking up the green constituent. Kraus'
investigation was also the first to show that the ordinary chlorophyll
spectrum was due partly to the green and partly to the yellow con-
stituent, which he called xanthophyll. Kraus' xanthophyll gave a

1. Zeit. f. Physiol. Chem. 56, p. 223 (1908).

2. Compt Rend. 100, p. 751 (1885); 104 p. 1293 (1887).

3. Ann. d. Chem. 21, p. 257 (1837).

4. Ann. Sc. Nat. 13, p. 45 (1860); Compt. Rend. 41, p. 189 (1865).
6. Proc. Roy. Soc. 21, p. 456 (1875).

6. Botan. Zeitung. 42, p. 817 (1884).

7. Flora, p. 155 (1875).

8. Compt. Rend. 39, pp. 9-184; 50, pp. 545 and 1182.

9. Sitz. ber. d. phys. Med. Ges. Wiirzberg (1883); and Arbeiten d. Botan.
Gessel. Wurzberg, 3, p. 127 (1884) and "Die Farbstoff des Chlorphylls" (1889).

10. Flora, Vol. 25 (1872).

11. Flora, Vol. (1874); Sitz. der. Wein. Akad. 89, 1. abts. p. 325.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 319

dark blue coloration with concentrated H2 SO4, and bleached very
quickly in the sunlight.

Sorby, using carbon bisulphide as the separator in place of benzo-
line, was the first to show that there is more than one yellow pigment
associated with chlorophyll.

Hansen's method of isolating the yellow pigments was still dif-
ferent. He treated the alcoholic extracts with caustic alkali, evaporated
the liquor to dryness and extracted the yellow pigment from the
residue with ether, the spectroscopic study of which led him to believe
that it exhibited three absorption bands. He believed also that it was
identical with the pigment of the carrot.

E. Schunck 1 obtained from all crude alcoholic chlorophyll ex-
tracts minute sparkling red crystals which deposited on standing, and
which he considered identical with the crystals which Bougarel 2 had
called erythrophyll, and which Hartsen3 has called crysophyll. This
pigment showed two absorption bands.

Tschirch 4 using Hansen's method, found two yellow coloring
matters, to which he gave the name xantho-carotin, showing three
bands, and xanthophyll proper which showed no bands.

Returning now to Arnaud's 5 work, we find that he identified
the red orange crystalline pigment which he obtained from spinach
leaves with the carotin of the carrot, both as regards to crystalline
form, melting point and chlorine derivatives.

We are indebted to Immendorff 6 for the confirmation of Arnaud's
results indicating that the carotin of green plants is identical with
the carotin of the carrot. Immendorff gave the pigment the formula
which Arnaud found for carotin, namely C26 H38. He states, however,
that the percentage composition of the pure pigment corresponded best
with Zeise's formula, C5 H8. Immendorff believed that carotin was
the only yellow pigment accompanying chlorophyll in the green leaf.

One of the most extensive publications in regard to carotin is
that by F. G. Kohl 7 This author also gives one of the best and most
voluminous compilations of the carotin literature that is to be found,
besides a large amount of experimental data. The literature is also
excellently reviewed by Tammes.8 Kohl gave carotin the formula

1. Proc. Roy. Soc. 44, p. 449.

2. Ber. Chem. Gessel, 10, p. 1173 (1877).

3. Arch. Pharm. 207, p. 166 (1875).

4. Botan. Zeitung. 42, p. 817 (1884); Ber. der Deutsch. Botan. Ges. 14, pt.
2, p. 76 (1896).

5. Compt Rend. 100, p. 751 (1885).

6. Landwirtschaftliche Jahrbiicher 18, p. 507 (1889).

7. Untersuch, Tiber d. Karotin, Leipzig. 1902.

8. Flora, p. 205 (1900).

320 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

C26 H38, and the iodine derivative C26 H38 I2. He also gave a detailed
description of the spectroscopic absorption of carotin. In ether and
carbon-bisulphide he measured three bands :

In ether In carbon bisulphide

I 49<>475 A I 510-485 *

II 455-445 * II 470-458 *

III 430-418 ^ III 437-425 A

Carotin is laevorotatory, according to Kohl, «D at 15° in chloroform
being— 30.17°.

Schunck * in his spectroscopic study of the yellow pigments of
leaves and flowers, described the properties of carotin. Schunck also
photographed the absorption bands of crysophyll (carotin) from the
daffodil leaf, from spinach, from the carrot and from grass, in alcoholic
solution. All of the carotin preparations showed the same three pro-
nounced bands situated between F and H the first band of which
lay almost directly upon the F line.

The most recent detailed investigation of the carotin of green
plants is that of Willstatter 2 and Meig, and a study of their data shows
that their results are to be accepted as the final proof of the chemical
constitution and properties of this pigment.

Willstatter and Meig describe the properties of carotin as follows :
Its crystals are copper colored plates of almost quadratic form, and
melt at 167.5° to 168° C. Its crystals are soluble with great difficulty
in hot ethyl alcohol and almost insoluble in cold ethyl alcohol, and
in methyl alcohol they are still less soluble; one gram of the crystals
requires 1.5 liters of petroleum ether (b. p. 30-50 °C.) for solution and
about 900 c.cm. of hot ethyl ether ; the crystals are difficultly soluble in
acetone, easily soluble in benzol, very easily soluble in chloroform
and instantly soluble in carbon bisulphide; the crystals are soluble in
concentrated sulphuric acid with an indigo blue color and are pre-
cipitated as green flakes on dilution with water.

The carotin obtained by Willstatter and Meig crystallized from
its deep red carbon bisulphide solution on addition of absolute alcohol,
but analysis showed that the crystals contained ^2 to ^ of a molecule
of alcohol of crystallization. The carotin showed the composition of
a pure hydrocarbon only after crystallization from low boiling point
petroleum ether. From this solvent the preparation of Willstatter
and Meig showed the composition C5 H7. A preparation of carotin
which the same authors obtained from the carrot showed the same

1. Proc. Roy. Soc. 72, p. 170 (1903).

2. Ann. der Chemie, 355, p. 1 (1907).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 321

composition. A molecular weight determination of both the carotin
from the carrot and from the Brennessel leaves, by the ebullioscopic
method in carbon bisulphide gave an average of 533 which corresponds
exactly with 8(C5 H7) or C40 H56. This shows that Arnaud's formula
of C26 H38 is not quite correct^ The same difference is brought out
by the analysis of the iodine derivative which Willstatter and Meig
also prepared.

The absorption bands of carotin were measured by Willstatter and
Meig and they coincided almost exactly with those given, for it by
Tschirch * and Monteverde2. They did not attempt to measure the
third band in the violet which they considered to be end absorption,
but measured only the two bands in the blue and indigo blue.

Willstatter and Meig Tschirch Monteverde

(alcohol sol.) (alcohol sol.) (petroleum ether sol.)

I 488-470 A I 487-470 A I 491-472 A

II 456-438 A II 457-439 A II 461-444 A

Carotin in Flowers, Fruits and Seeds. According to Czapek 3 caro-
tins have been identified in many flowers by Hansen,4 Immendorff,5
Kohl,6 Tammes,7 Hilger,8 and his pupils, Wirth,9 Pabst,10 Kirchner,11
Ehrung 12 and Schuler.13

Among the fruits, Arnaud,14 Passerini,15 Kohl,16 Schunck 17 and
Montanari 18 have investigated the tomato pigment and believed it to
be a carotin. Its identity as a truly isomeric carotin has recently been
proved by Willstatter and Escher.19 Schrotter20 has shown that the
pigment of the pumpkin is in all probability a carotin and Desmoliere 21
has identified carotin in the apricot.

1. Ber. d. deut. botan. Ges. 14, 76 (1896); 22, 414 (1904).

2. Acta Horti. Petropolitani XIII Nr. 9, 123 and 150 (1893).

3. Bichemie der Pflanzen, vol. I, p. 172, etc.

4. 5, 6, 7. Loc. cit.

8. Botan. Centr. 57, p. 335 (1894).

9. Dissert. Erlangen (1891).

10. Arch. Pharm. 230, p. 108 (1892).

11. Dissert. Erlangen 1892.

12. Botan. Cent. 69, p. 154 (1897).

13. Dissert. Erlangen. 1899.

14. Compt. Rend. 102, p. 1119 (1886).

15. Compt. Rend. 100, p. 875 (1885).

16. Loc. Cit.

17. Proc. Roy. Soc. 72, p. 172 (1903).

18. Le Staz. sp. agra. ital. 37, p. 909 (1904).

19. Zeit. Physiol. Chem. 64, p 74 (1910).

20. Vehr. Zool. bot. Gessel. 44, 298 (1895).

21. Chem. Centr. 2, p. 1001, 1902.

322 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

Among the seeds, Schunck 1 has found the annatto pigment to be a
carotin.

The Xanthophylls.

It was mentioned above that it has been found that a second class
of pigments usually accompanies carotin. Investigations of this class
of pigments, now called xanthophylls, has not been as extended as
that of carotin but the constitution and properties of the xanthophylls
are nevertheless at present established.

Sorby2 differentiated the pigments accompanying chlorophyll as
xanthophyll, orange xanthophyll, and yellow xanthophyll, all with
spectroscopic properties. J. Borodin 3 observed that besides carotin,
a second crystallizable yellow substance exists in leaves which is much
more soluble in alcohol than carotin and insoluble in benzine. Im-
mendorff4 denied the existence of more than one pigment as was
noted above. Monteverde5 confirmed Borodin's observations.
Tschirch 6 in 1896, showed that green leaves contain a second yellow
pigment which, however, showed no absorption bands. Tschirch called
the second pigment xanthophyll. The name, however, was a misnomer,
for Schunck 7 later showed that Tschirch was dealing with a group of
water and alcohol soluble pigments probably identical with the lich-
noxanthine described by Sorby.8 Tschirch9 later recognized the ex-
istence of a true second yellow crystallizable pigment.

Molisch 10 in his critical study of the yellow pigments left the
question of their plurality an open one, and Tammes n also left the
question undecided.

Schunck 12 in his widely known spectroscopic study of the yellow
pigments of plants and flowers, demonstrated beyond a doubt that a
second great group of pigments, which he designates the xanthophylls,
accompanies the crysophyll. He differentiated three different xantho-
phylls and designated them L. B. and Y. xanthophyll, respectively.

He found that the xanthophylls were all characterized by giving
the same color reactions in the dry state as crysophyll and three

1. Proc. Roy. Soc. 72, 1903.

2. Proc. Roy. Soc. 21, p. 457 (1875).

3. Melanges Biol. tir. d. bull d. L'Acad. Imp. d. St. Petersb. 11, p. 512 ( 1883 ) .

4. Loc. cit.

5. Loc. cit. p. 148 (1903).

6. Ber. d. d. Botan. Gessel, 14, p. 76 (1896).

7. Proc. Roy. Soc. 72 (1903).

8. Loc. cit.

». Ber. d. d. Botan. Gessel. 22, p. 414 (1904).

10. "Die Krystallization und der Nachweis des Xanthophylls (carotins)
in Blatte" (Ber. d. Deut. Botan. Ges. 14, p. 18 (1896).

11. Loc. cit. (1900).

12. Proc. Roy. Soc. 65 (1899); 68 (1901); 72 (1903.)

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 323

similar absorption bands in the violet region of the spectrum. The
bands of the xanthophylls, however, were all shifted somewhat
towards the blue with respect to the bands of crysophyll, the amount
of shifting depending on the xanthophyll, L xanthophyll being shifted
the least and Y xanthophyll the most. Schunck found the absorption
bands of the different xanthophylls especially characterized by the
action of their alcoholic solutions in the presence of HC1 and HNO3
the details of which are given in his latest paper.1

Schunck also made the very interesting discovery that the yellow
pigment of egg yolk and fowl serum shows the identical properties
of L xanthophyll both with respect to the position of the original
absorption spectra and also the action of acids upon the spectra. .

One of the most interesting and important studies of chlorophyll
and its accompanying yellow pigments was made by Tswett2 who
discovered and thoroughly investigated the adsorption properties of
these pigments. He was able to demonstrate the presence of at least
four different xanthophylls which he designates as xanthophylls
* «' a" and B. A more detailed review of this work will be given
in connection with a report of the present investigations. It is of
interest here especially on account of its historical position with respect
to the establishment of the chemical constitution of the xanthophylls.
It was Willstatter and Meig3 who isolated and identified the
crystalline xanthophyll pigment accompanying the carotin in green
plants and leaves, and, as noted above, Eiiler and Nordenson4 have
recently found xanthophyll crystals in their extracts from the carrot,
thus indicating a more general distribution of the xanthophylls in
connection with carotin than has been believed.

The results of the study of the crystalline xanthophyll show that
it is composed of carbon, hydrogen and oxygen in the proportion
C4t H56 O2 and is thus merely carotin dioxide.5 The pigment is further
distinguished from carotin by the color and shape of its crystals, which
are yellow or orange trapesium plates sometimes spear or wedge-shaped
which are characterized by a steel blue reflection. The pigment ex-
hibits an entirely different solubility toward petroleum ether and abso-
lute alcohol than carotin, being insoluble in the former and readily
soluble in the latter solvent. According to these authors, the pure

1. Proc. Roy. Soc. 72 (1903).

2. Ber. Botan. Gessel, 24, pp. 316 and 384 (1906).

3. Ann. der. Chemie, 355, p. 1 (1907).

4. Loc. cit

5. Willstatter and Meig point out the probable identity of xanthophyll with
the hitherto unexplained hydrocarotin found by Husemann.

324 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

crystals have a melting point of 172° C. (corrected) which is slightly
higher than the melting point of the carotin crystals ; and the absorption
bands of the pigment are slightly shifted toward the violet from the
corresponding bands of carotin, as was also shown by Schunck x for
the xanthophylls which he differentiated.

It might be readily assumed that xanthophyll is formed directly
from carotin in the plant. In fact Tschirch 2 has claimed that carrot
carotin goes over to xanthophyll in the air. Eiiler and Nordenson 3
do not credit this statement and state that, "One may well suppose
that in the plant, xanthophyll normally is formed from the carotin,
but outside of the plant it has not been possible to imitate this trans-
formation, the most skillful oxidation always leading to a much
higher oxidized product." Willstatter and Meig believed in this con-
nection that xanthophyll although carotin dioxide is not the end pro-
duct of the oxygen absorption of carotin in the plant. Monteverde 4
and Lyubimenko have recently claimed that chlorophyll and xantho-
phyll originate from the same colorless substance, carotin being a
complimentary product generated during the formation of cholorophyll,
but not necessarily from the xanthophyll.

The Pigments of Animal Origin.

The Luteins. — We will now direct our attention to a review of the
literature bearing upon the yellow pigments of so-called animal origin.
Thudichum 5 was one of the first to investigate the yellow animal pig-
ments and he classified a great many of them together with the yellow
pigments of plants under the name lutein, the name being taken from
the pigment of the corpus luteum. He states, "Various parts of
animals and plants contain a yellow crystallizable substance which has
hitherto not been defined, and which I call lutein. It occurs in the
corpora lutea of the ovaries of animals, the serum of the blood, the
cells of adipose tissue, in butter, in the yolks of eggs of oviparous
animals, in seeds such as maize, in husks and pulps of fruits such
as annatto, in roots such as carrots, in leaves such as those of coleus,
and in the stamens and petals of a great many flowers.''

It is unfortunate that none of the above statements are supported
by experimental evidence, for it can hardly be accepted that Thudichum
was able to obtain crystals of lutein from all the bodies in which he

1. Loc. cit.

2. Ber. Botan. Gessel. 22, p. 414 (1904).

3. Loc. cit.

4. Bull. Acad. Imper. Sc. St. Petersb. 30, p. 609 (1912).

5. Proc. Roy. Soc. 17, p. 253 (1869).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 325

claims to have found it, or was able to show all the properties which
he describes for the crystals which he evidently did obtain.

Crystalline animal pigments were apparently obtained before
Thudichum's claims in this regard. According to Krukenberg,1 Wittich 2
obtained crystals of a red pigment from Euglenia Sanguirubo, and
Piccolo and Lieben 3 found a crystalline animal pigment. Pouchet 4
a little later obtained a yellow crystalline pigment from lobsters.

The Chromophanes. — The early workers in the field of animal pig-
ments laid great emphasis upon the so-called color reactions, one of
which, the blue reaction which concentrated HNCX, was mentioned by
Thudichum. That a similar reaction is given by concentrated H2 SO4
was first noticed by Wittich in 1863, an<^ Buchholz also noticed it with
a fat pigment from a Ganglion cell of an invertebrate. Piccolo and
Lieben had also noticed the blue reaction with concentrated H2 SO4.
Besides Thudichum, Filhol 5 and Stadeler Q noticed the blue reaction
with concentrated HNO3. Stadeler attempted to isolate the egg yolk
pigment. He failed to do so, however, but attempted to establish the
difference between it and Bilirubin with which it bad been considered
identical. A little later a third reaction of the luteins was discovered
by Schwalbe,7 namely a blue-green color with a solution of iodine in
potassium iodide. Schwalbe first noticed the reaction with the cone-
globules of the retinas of birds and lizard's eyes. The red globules
gave a beautiful blue to blue-black color, and the yellow oil globules a
green to blue-green to blue. The pigments thus characterized were
called chromophanes by Schwalbe and the existence of these pigments
was a little later considerably extended by Capranica 8 who also made
use of the iodine reaction.

Kiihne9 took up the study of the chromophanes of the cone-
globules of bird retinas, and separated three pigments which he desig-
nated Rhodophan, Chlorophan and Xanthophan, respectively, according
to the color of their solutions.

Kiihne also studied the absorption spectra and color reactions of
the pigment of the egg yolk and the corpus luteum and compared them

1. Grundzuge einer vergleichenden Physiologie der Farbstoff und der
Farben; 1884.

2. Arch f. Path. Anat. 27, p. 573 (1863).

3. Giornals d. Scienze Natural! et. Economich. Palermo 2, p. 258 (1866).

4. Jour. d. L'Anat et. Physiol. 12, p. 12 (1876).

5. Compt. Rend. T. 39, p. 184, T. 50, pp. 545 and 1182.

6. Jour. f. Pract. Chem. 100, p. 149 (1867).

7. Hand D. Ges. Augenheilkunde von Graefe u. Saemisch I, p. 414 (1874).

8. Arch. f. Anat. Physiol. p. 283 (1877).

9. Untersuch, des Physiol. Universitat Heidelberg I, 4th Heft, p. 341
(1878); IV, p. 169 (1882); Jour. Physiol. 1, p. 109 (1878).

326 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

with these properties of the retinal pigments. A study of their spectro-
scopic absorption properties led him to believe that the pigments were
not identical.

Kiihne in his celebrated work on "Optochemie" occupied himself
somewhat again with the egg yolk pigment and called it Ontochrin
or Lecithochrin. He did not try to isolate it free from nitrogen, but
he did succeed in observing crystals. He again was careful to dis-
tinguish between the egg yolk pigment and the corpus luteum pigment,
which he at this time considered as extraordinarily closely related to
carotin.

The Lipochromes. Basing his work on the researches of Kiihne,
Krukenberg commenced a series of researches which extended from
1879 to 1886, the most important of which appeared in his "Verglei-
chende Physiologische Studien" * and especially in the paper, "Grund-
zuge einer vergleichenden Physiologic der Farbstoff und der Farben'r
which appeared in 1884. Krukenberg made an exhaustive study of
what had been done on animal pigmentation and included under one
head all those pigments which had previously been known as luteins,
carotin, zoonerythrin (tetronerythrin) and Kiihne's chromophanes, and
called them lipochromes.

Krukenberg believed that carotin, the pigment of the carrot, was
the best representative of the lipochrome coloring matters, and ac-
cepted Husemann's formula for carotin (C18H24O) as representing
the chemical composition of the lipochromes.

In regard to the origin of lipochromes Krukenberg believed, "It
is probable that in most cases they originate from fatty substances,
for frequently, if not without exception, they occur in company with
fat and allow themselves to easily go over into cholesterin-like bodies."

In 1885 Krukenberg 2 isolated a yellow lipochrome from the blood
serum of the ox by extracting the serum with amyl alcohol. The
solution showed two absorption bands, one enclosing the line F and
the other lying between F and G. A year later Halliburton 3 reported
that he extracted a yellow lipochrome from the blood serum of the
pigeon, hen, dove and tortoise by means of alcohol. Halliburton re-
ported an identical pigment in the body fat of these same animals.

MacMunn 4 was the next investigator of animal pigments, and
like Krukenberg, he extended the classification lipochrome to include

1. Zoonerythrin (Tetronerythin) : — Central, f. d. Medic. Wiss. 1879. Vergl.
Physiol. Studien I Reihe, II Abth. s. 67-71; III Abth. s. 114-115; IV Abth. s.
30-35; V Abth. s. 87-94; II Reihe, I Abth. s. 165-167; III Abth. s. 135).

2. Sitz, ber. d. Jen. Gessel. f. Med. 1885.

3. Jour. Physiol. 7, p. 324 (1886).

4. Philos Trans. Roy. Soc. 177, p. 247 (1886).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 327

the yellow constituent of chlorophyll or Hansen's "Chlorophyll Yellow."
He believed that the lipochromes were chemically closely related to
chlorophyll.1

MacMunn's greatest contribution to animal chromotology was in
iSSg.2 The pigments of a great many marine animals, Crustacea,
worms and sponges were examined and classified. Lipochromes were
found abundantly, MacMunn drawing a distinction as to whether the
lipochrome was a rhodophan or a chlorophan-like lipochrome.

In regard to some of the properties of the lipochromes MacMunn
states, as did Krukenberg, that they are sensitive to light, both in the
solid state and in solution, and yield in many cases cholesterin-like
substances. He believed that many of the plant lipochromes were
identical with the animal lipochromes.

It will be remembered that for a long time there were many
followers of the view that a close relationship existed between carotin
and cholesterol and that this view was only finally discredited by a
study of the pure crystalline pigment.

Cotte 3 recently carried out an investigation in which he sought
and claims to have shown that the lipochromes, both animal and
vegetable are intimately associated with cholesterol. Cotte's results
have been thoroughly disproved by Henze.4

Since the early work of Pouchet 5 and Maly 6 who distinguished
between yellow and red crustacean lipochromes many investigators
have classified the lipochromes according to their red or yellow color,
Newbigin 7 in a recent investigation of the pigments of the skin,
muscle and ovaries of the salmon, reports that he found two pigments
present, a red and a yellow, which he claims he was able to separate
from each other. Newbigin concluded from the color reactions of
the pigments that the red pigment was a true lipochrome while the
yellow pigment was not.

In regard to the yellow pigment, Newbigin says that, "It belongs
to a group of pigments that are apparently exceedingly widely dis-
tributed in the animal kingdom, but which have been little investigated.
They have been commonly confounded with the lipochrome pigments."

He extracted the pigment from the bright yellow body fat of a
cow and found it to have properties identical with the yellow pigment

1. Jour. Physiol. 9, p. 1 (1888).

2. Quart Jour. Micros. Sc. 30, p. 15 (1889).

3. Compt. Rend. Soc. Biol. 55, p. 812 (1903).

4. Zeit. Physiol. Chem. 41, p. 109 (1904).

5. Jour. d. 1'Anat. de la Physiol. 1, 12, 10 (1876).

6. Sitz. d. k. Akad. d. Wiss. zu. Wein. 83 (1831).

7. D. Noel Patton— Report of Inv. on Life Hist, of Salmon ( 1898 ) , Article XT.

328 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

of the salmon with the exception that it was very little soluble in
methyl alcohol, but dissolved readily in ether.

General Properties of the Lipochromes. It will not be out of place
to give a brief summary here of the general characteristics and prop-
erties of the lipochrome pigments as found up to this time.

Lipochromes 7 may be classed as salve-like, yellow or red or
orange colored residues, which have been obtained in needles or
rhombic plates, where they have been crystallized. They are soluble
in alcohol, ether, benzol, petroleum ether, amyl alcohol, chloroform,
carbon bisulphide, ethereal oils and fats with a yellow or yellow-
orange color. They are insoluble in cold and hot water and alkalies
and dilute acids, but are soluble in alcoholic alkaline solutions and
are unchanged when these solvents are heated. In alcohol or other
solvents they are unstable,, and readily bleach, as do the residues
from these solutions. The bleach product is unknown, but it is cer-
tainly not identical with cholesterol. On addition of concentrated
H2 SO4 or HNO3, the lipochromes give a color change of blue-green-
violet to brown. The color reactions are often interfered with by the
presence of a small amount of foreign substance. The lipochromes
generally give a blue-green coloration with a solution of iodine in
potassium iodide. Spectroscopically the lipochrome solutions show
two bands and sometimes three in the blue part of the spectrum,
and again they sometimes show no bands at all.

The lipochromes may be extracted from the fresh or dried tissues
in which they are found, by organic solvents, best by hot or cold
alcohol, ether, petroleum ether, carbon bisulphide or chloroform, the
choice of the solvent resting with whether some foreign pigment is
present. When fat is present, the pigment may be heated with alcoholic
alkali which will not saponify the lipochromes. The lipochromes can
be extracted from the soap with ether, petroleum ether, or chloroform,
either directly or after acidifying, or the lipochromes can be salted out
of their alkaline soap solutions with sodium chloride, and the lipo-
chromes obtained by extracting the precipitated soap with alcohol or
ether.

The Lipochromes of Algae, Fungi, and Bacteria. While a wide dis-
tribution of the lipochromes has already been mentioned, a review
of their literature would not be complete without mentioning their
distribution in algae, fungi and bacteria.

1. Summarized from "Lipochromes" by Franz Samuely. Alderhalden's
Biochemisches Handlexikon vol. 6, and Handbuch der Biochemischen Arbeits-
methoden, vol. 2.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 32^

Hansen l first showed the presence of lipochromes in algae and
Tammes 2 has lately shown their presence in a large number of these
plants. Zopf 3 has investigated the lipochromes of fungi and especi-
ally of bacteria, the first lipochrome-producing bacteria being pointed
out by him.

The Lipoehrome or Lutein of Egg Yolk. It will be readily agreed
that while some order has been attained in classifying the widely dis-
tributed animal pigments, by means of the convenient and flexible
classification "lipochromes," our knowledge of the animal pigments is
far from being satisfactory when compared with the status of the
orange and yellow plant pigments, the carotins and xanthophylls. The
science of animal chromotology should accordingly be exceedingly
grateful for the recent work of Willstatter 4 and Escher, on the lutein
of egg yolk, the result of which has been to throw new light upon
the constitution of the lipochromes of the higher animals and upon
their relations to the carotins and the xanthophylls.

The main pigment of the yolk of hen's eggs was isolated in
crystalline form by these investigators, and when in approximately
pure condition showed sufficiently close agreement with the constitution
of xanthophyll that the authors claim that the egg lutein on account
of its melting point (195-196° C. corrected) is a true isomer of the
crystalline xanthophyll of green plants. In all its other properties in-
cluding its spectroscopic absorption bands, the egg lutein was identical
with the crystalline plant xanthophyll.

It is worthy of note also that during the isolation of lutein a
minor constituent was noticed which gave every indication of being
closely related to carotin; but as it was present in very small amount
compared with the xanthophyll it was disregarded.

In concluding the review of this investigation it will be important
to mention that the authors state that one of them, i. e., Escher, is at
present investigating the pigment of the corpus luteum which they state
has been found to belong to the hydrocarbon or carotin group of
pigments.5

1. Arbeit Botan. Inst Wurzberg 3, 296 (1883).

2. Loc. cit.

3. Ber. Botan. Gessel. 9, 27 (1891).

4. Zeit Physiol. Chem. 76, pp. 214-225 (1912).

5. Note — Since writing the above, Dr. Escher has published his investiga-
tions which show that the corpus luteum pigment is in every respect identical
with the carotin of the carrot and of green plants. Zeit. f. Physiol. Chem. 83,
p. 198 (1913).

330 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

The Physiological Relation Between Plant and Animal Lipochromes.

With the review of the chemical side of this problem complete,
it yet remains to consider what has been shown in regard to relations
other than chemical, between the animal and plant pigments whose
properties are so nearly related and in many cases identical.

The literature has been found to be very brief on this point.
Newbigin 1 gives a rather extensive consideration of this subject and
attempts to explain the presence of the red and yellow pigments found
by him in the salmon organism. While he considered the most obvious
explanation to be that they were derived directly from the food, he
found a number of difficulties in the way of the acceptance of such
an explanation, the most important of which was that he was able
to show the presence of but a trace of only the yellow pigment in the
usual food of the salmon,

As to the possibility of transference of yellow pigments from
one organism to another, Newbigin points out what he believes to be
some evidence apart from the case of the salmon. He says, "Poulton 2
has shown by experiment that certain caterpillars derive their pigments
from their food. Again it is not uncommon to find fat of sheep and
cows dyed a deep yellow color. According to some authorities this
occurs quite sporadically without known cause, while according to
others, special foods, notably maize, are the important agents." New-
bigin says in this connection, "I have examined the yellow pigment of
maize, and compared it with the pigment from yellow fat. The maize
pigment gives the lipochrome reaction faintly with H2 SO4 distinctly
with HNO3 while the fat pigment gives no lipochrome reaction. In
other respects, in tint, solubility, etc., the pigments closely resemble
each other." Newbigin did not feel warranted to conclude from these
experiments that all yellow pigments of animals are derived from
their food, for with such a conclusion, he states, "It would be difficult
to understand why such colored fat should not be universal in herbivor-
ous animals, for all green parts of plants contain also a certain amount
of yellow pigment."

It seemed to Newbigin, however, that a reasonable explanation for
salmon, domesticated cattle and caterpillars would be to suppose that
when they ingest a moderate amount of colored fat in their food,
that they could utilize or eliminate the pigment, and so deposit colorless
fat in the tissues ; but when the ingestion of colored fat is in excess
of the actual requirements as it so often is, especially with domes-

1. Loc. cit.

2. Proc. Roy. Soc. 54, p. 417; Nat. Sci. 8, p. 98.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 331

ticated cattle, an elimination or utilization of the pigmented fat is
impossible and fat colored with the pigment in a more or less modified
condition is thus stored up.

There is abundant proof in this literature aside from the above
speculations that animals are able to lay up fat soluble dyes in the
organism and even eliminate them in the milk. Only recently Mendel
and Daniels 1 have shown that Sudan III and other fat soluble dyes
may be deposited in the organism in adipose tissue and bone marrow
when introduced into the organism either dissolved in fat or when
fed alone. When fed with fat or when fat was present in the alimen-
tary tract the dyes entered the organism through the lymphatics in
solution in fat, but when fat was absent, through the portal circulation
dissolved in bile in which they are nearly all soluble. In the latter
case the pigments did not pass beyond the liver unless fat was present
to transport them, in which case only they were subsequently found
in the blood. When fat stained food was fed to small animals (cats,
rats, guinea pigs, etc.) in lactation, and in one case with a goat, the
dye appeared in the milk shortly after the first feeding of the dye.
The same authors feeding fifteen grams of Sudan III to a Holstein
cow for three successive days were unable to detect the dye in the milk.
The authors also made the interesting observation that stained fat
does not traverse the placental barrier; the blood and foetus and
fat of the young born of Sudan-stained female cats and rats were free
from the dye.

Gogitidse2 fed hog fat (100 grams per day) colored with Sudan
III to a bitch and after two days found the dye in the milk. The
body fat did not show this coloration so soon and then not so clearly,
in fact only after long continued feeding of the stained fat.

Backhaus 3 studying the "Influence of Feed and Individuality on
the Taste and Healthfulness of Milk," says that a number of plants
influence the color of milk and butter. The same author conducted
several pigment feeding experiments with cows. Negative results were
obtained with respect to the milk when feeding Fuchsin, Bismark
brown, and curcuma powder, although the feces showed the pigments
abundantly. When feeding sodium fluorescin the urine was affected
but not the milk. When feeding methyl violet, however, the author
was able to show that this pigment was carried into the milk fat in
a reduced condition so that on contact with the air and with the aid

1. Jour. Biol. Chem. 13, No. 1, p. 72 (1912).

2. Zeit. f. Biol. 45, 353 (1904).

3. Berichte, Landwirt. Inst. U. Konigsberg 5 (1900).

332 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

of heat, the milk fat showed an intense blue coloration. The feces,
however, showed the pigment in an unchanged condition.

Summary.

The foregoing review of the literature has shown that the great
number of pigments that exist throughout the entire plant and animal
kingdoms have long been of interest from a scientific standpoint. The
pigments of botanical origin have been thoroughly and exhaustively
investigated. This is especially true of the yellow pigments carotin
and xanthophylls, and their chemical constitution and properties are
now established.

The yellow and orange pigments of plants were at first classified
in one group, and were called carotins, the name being derived from
the pigment of the carrot, which was the first one investigated. A
great many different names were given to this pigment as it was
independently discovered in various plants but the identity of these
pigments with the carrot pigment has now been established. It was
eventually discovered that the carotins are always accompanied, especi-
ally in green plants, by a second great class of pigments which have
been called xanthophylls, whose relation to carotin has but recently
been established.

As the work on plant pigmentation developed, it was recognized
that the general properties of a great many yellow pigments found in
animals were similar to the so-called carotins. The first investigators
classified these animal pigments under the name lutein, the name
being derived from the pigment of the corpus luteum, which was the
first one investigated. The name lutein was extended by the animal
chromotologists to include the carotins of plants and its related pig-
ments. Later, when the animal luteins had become generally recog-
nized by their association with fat, the name lutein was changed to
lipochrome and this designation was also extended to include all sim-
ilar pigments of both plants and animals.

The classification of the plant and animal pigments which is at
present generally accepted is to restrict the names carotin and xan-
thophylls to the two great classes of yellow plant pigments, and to
include under the name lutein or lipochrome only those yellow pig-
ments which are considered to be of animal origin.

The most recent work in the field of animal chromotology has
shown that the luteins can also be subdivided into carotin and xan-
thophyll groups depending on their chemical relation to the carotin or
xanthophylls of plant origin. Accordingly Schunck l has shown the

1. Loc. cit.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 333

spectroscopic identity of the egg yolk pigment with a xanthophyll
which he isolated from the yellow daffodil, the nasturtium, and green
leaves. Willstatter and Escher l have confirmed this with a chemical
analysis of the egg yolk pigment, showing it to be a true isomer
of the crystalline xanthophyll of green plants; they have called it
xanthophyll B. Escher 2 has recently published his investigation show-
ing that the pigment of the corpus luteum is identical in chemical
composition and properties with the carotin of green plants.

These recent discoveries have opened the way for an extension
of such investigations to other yellow animal pigments whose isolation
is rendered much more difficult by their association with very large
quantities of fat and other substances. These discoveries have also
raised the question whether any relation other than chemical exists
between the yellow animal and plant pigments. This question has
never been investigated. The investigations which will be reported in
the succeeding papers are the first to show that there is a definite
relation other than chemical between the yellow plant and animal pig-
ments.

1. Loc. cit.

2. Loc. cit.

334 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

BIBLIOGRAPHY.

1. Arnaud: Compt. rend. 100, p. 751 (1885); Jour. Pharm. Chim.

14; p. 149 (1886) ; Compt. rend. 102, p. 1119 (1886) ; 104, p.
1293 (1887).

2. Backhaus: Bericht. Landwirt. Inst. U. Konigsberg, 5 (1900).

3. Berzelius: Ann. d. Chem. 21, p. 257 (1837).

4. Borodin: Melanges Biol. tir. d. L'Acad. Imp. d. St. Petersb.

11, p. 512 (1883).

5. Bouchardat: Schweizg. Jour. Chem. 58, 95 (1830).

6. Bougerel: Ber. Chem. Gessel. 10, p. 1173 (1877).

7. Capranica: Arch. f. anat. u. Physiol. p. 283 (1877).

8. Conrad: Flora 25 (1872).

9. Cotte: Compt. rend. Soc. Biol. 55, p. 812 (1903).

10. Czapek: "Biochemie der Pflanzen" vol. i, p. 172 et seq.

11. Desmoliere: Chem. Centr. 2, p. 1001 (1892).

12. Ehrung: Botan. Cent. 69, p. 154 (1897).

13. Eiiler and Nordenson: Zeit. f. Physiol. Chem. 56, p. 223 (1908).

14. Filhol: Compt. rend. 39, pp. 9-184; 50, pp. 545 and 1182.

15. Fremy: Ann. Sc. Nat. 13, p. 45 (1860); Compt. rend. 41, p.

189 (1865).

16. Gogitidse: Zeit. f. Biol. 45, p. 353 (1904).

17. Halliburton: Jour. Physiol. 7, p. 324 (1886).

18. Hansen: Sitz. ber. d. phys. Med. Ges. Wiirzberg (1883);

Arbeiten d. Botan. Gessel. Wiirzberg. 3, p. 127 (1884) ; "Die
Farbstoff des Chlorophylls" (1889).

19. Hansen: Arbeit, botan. Inst. Wiirzberg 3, p. 296 (1883).

20. Hartsen: Arch. Pharm. 207, p. 166 (1875).

21. Henze: Zeit. Physiol. Chem. 41, p. 109 (1904).

22. Hilger: Botan. Centr. 57, p. 335 (1894).

23. Husemann: Lieb. Ann. 117, p. 200 (1860).

24. Immendorff: Landwirt. Jahrb. 18, p. 507 (1889).

25. Kirchner: Dissert. Erlangen (1892).

26. Kohl : "Untersuch. iiber d. Karotin," Leipsig 1902.

27. Kraus: Flora, p. 155 (1875).

28. Krukenberg: Central f. d. Med. Wiss. 1879. Vergl. Physiol.

Studien I Reihe, II Abth. pp. 67-71; III Abth. pp. 114-115;
IV Abth. pp. 30-35; V Abth. pp. 87-94; II Reihe i Abth. pp.
165-167; III Abth. p. 135.

29. Krukenberg: Sitz. ber. d. Jen. Gessel. f. Med. 1885.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT. 335

30. Kiihne: Utersuch. des. Physiol. Inst. U. Heidelberg I, 4th Heft

p. 341 (1878); Jour. Physiol. i, p. 109 (1878).

31. Kiihne: Untersuch. des. Physiol. Inst. U. Heidelberg 4, p. 169

(1882).

32. Maly: Sitz. d. k. Akad. d. Wiss. zu Wein 83 (1881).

33. MacMunn: Philos. Trans. Roy. Soc. 177, p. 243 (1886); Jour.

Physiol 9, p. I (1888); Quart. Jour. Micro. Sc. 30, p. is

(1889).

34. Mendel and Daniels: Jour. Biol. Chem. 13, No. i, p. 72 (1912),

35. Merejkowsky: Bull. d. la soc. Zool. d. France, 1883.

36. Molisch: Ber. d. d. Botan. Gessel. 14, p. 18 (1896).

37. Montanari: Le Stazioni Sperim. Agr. Ital. 37, p. 909 (1904).

38. Monteverde : Acta. Horti. Petropolatini XIII. Nr. 9, pp. 123 and

150 (1893).

39. Monteverde and Lyubimenko: Bull. Acad. Imper. Sc. St.

Petersb. 30, p. 609 (1912).

40. Newbigin: D. Noel Paton. "Report of Investigations on Life

History of the Salmon." 1898, Article XV, p. 159.

41. Pabst: Arch. Pharm. 230, p. 108 (1892).

42. Passerini: Compt. rend. 100, p. 875 (1885).

43. Piccolo and Lieben: Giornals d. Scienze Naturali et Economich.

Palermo 2, p. 258 (1866).

44. Pouchet: Jour. d. L'Anat. d. la Physiol. 12, p. 10 (1876).

45. Poulton: Proc. Roy. Soc. 54, p. 417; Nat. Sci. 8, p. 98.

46. Pringsheim: Untersuch. uber das chlorophylls I Abth., Berlin

1874.

47. Schrotter: Verb. Zool. Botan. Gessel. 44, p. 298 (1895).

48. Schuler: Dissert. Erlangen 1899.

49. Schunck: Proc. Roy. Soc. 65 (1899) ; 68 (1901), 72 (1903).

50. Schunck: Proc. Roy. Sec. 44, p. 449.

51. Schwalbe: Hand. d. Ges. Augenheilkunde von Graefe u.

Saemisch I, p. 414 (1874).

52. Sorby: Proc. Roy. Soc. 21, p. 456 (1875).

53. Stadeler: Jour. f. Pract. Chem. 100, p. 149 (1867).

54. Tammes: Flora, p. 205 (1900).

55. Thudichum: Proc. Roy. Soc. 17, p. 253 (1869).

56. Tschirch: Botan. Zeitung 42, p. 817 (1884).

57. Tschirch: Ber. der. d. Botan. Gessel 14, pt. 2, p. 76 (1896)

22, p. 414 (1904).

58. Tswett: Ber. der. d. Botan. Gessel. 24, pp. 316 and 384 (1906) ;

29, p. 630 (1911).

336 MISSOURI AGR. EXP. STA. RESEARCH BULLETIN No. 9.

59. Wachenroder: Dissertatio de Anthelminticis Gottigen 1826;

Geiger's Magaz. Pharm. 33, p. 144 (1831) ; Berzelius Jahres-
ber. 12, p. 277 (1833).

60. Wiesner: Flora (1874); Sitz. der Wein. Akad. 89, part i, p.

325.

61. Willstatter and Escher: Zeit. f. Physiol. Chem. 64, p. 74 (1910) ;

76, pp. 214-225 (1912).

62. Willstatter and Meig: Ann. d. Chemie 355, p. i (1907).

63. Wittich: Arch. f. Path. Anst. 27, p. 573 (1863).

64. Wirth: Dissert. Erlangen (1891).

65. Zeise: Lieb. Ann. 62, p. 380 (1847); Ann. Chem. Phys. (3)

20, p. 125 (1847).

66. Zopf : Ber. der d. Botan. Gessel. 9, p. 27 (1891).

CAROTIN— THE PRINCIPAL NATURAL YELLOW PIGMENT OF
MILK FAT— JPART II.*

Chemical and Physiological Relations of Pigments of Milk Fat to the
Carotin and Xanthophylls of Green Plants.

LEROY S. PALMER AND C. H. ECKLES

For a number of years the great variety of yellow animal pig-
ments have been classified under the general name lipochrome. Recent
investigations by Willstatter and Escher,1 and by Escher 2 have shown,
however, that some of these pigments are in reality very closely related
or identical with the carotin or xanthophyll pigments of plants. Will-
statter and Escher have analyzed the pure lutein of egg yolk and
found it to be isomeric with the crystalline xanthophyll of green
plants; and Escher has identified in the same manner the lipochrome
of the corpus luteum as a true carotin. The isolation of both pig-
ments was attended with great difficulty. In the case of the egg yolk
pigment the yolk of 6,000 hen eggs yielded only four grams of crude
crystalline pigment, while in the case of the corpus luteum pigment
less than 0.5 gram of crystals were obtained from 10,000 cows' ovaries.

The natural yellow pigment of butter is the most commonly ob-
served of all animaF lipochromes. It is also more important from a
commercial standpoint than any other animal pigment; the public
judges the richness of dairy products by their yellow color, and de-
mands that butter especially shall have a standard shade of yellow.
The pigment of butter fat, however, has been the least investigated
of all animal lipochromes. Thudichum's 3 classical investigation in-
cluded the pigment of butter fat under the general classification lutein,
which he proposed. No other study of the butter fat pigment has

1. Zeit. f. Physiol. Chem. 76, pp. 214-225 (1912).

2. Zeit. f. Physiol. Chem. 83, p. 198 (1913).

3. Proc. Roy. Soc. 17, p, 253 (1869).

* See Res. Bui. No. 9, p. 312, for statement of co-operation with U. S.
Dept. of Agriculture.

(339)

34° MISSOURI AGRICULTURAL EXPERIMENT STATION, BULLETIN NO. IO

been reported. It is usually classified in the current text books,1
however, according to Krukenberg's classification of lipochromes.

In view of the commercial importance attached to the butter fat
pigment and especially in view of the results of the most recent inves-
tigations in this field, it was recognized that a thorough investigation
of this pigment would be of great value both from a scientific as well
as a practical standpoint.

The present investigation was therefore undertaken for the pur-
pose of classifying the butter fat pigment as a true lipochrome and
also with respect to its relation to the carotin and xanthophylls of
green plants. It was also the purpose of the investigation here
recorded, to gather as much information as possible relative to the
influence of certain factors upon the color of butter, among which may
be mentioned the character of the ration and the breed of the cow.

METHODS OF ISOLATION,

The statement is frequently met in the literature2 and in the text
books and works 8 on oils and fats, that the pigment of butter or butter
fat appears in the unsaponifiable extracts along with cholesterol and
other substances.

A number of methods for obtaining the unsaponifiable matter of
butter fat are available and several were tried. The method finally
adopted for the isolation of the crude pigment was to saponify the
butter fat with a twenty per cent solution of alcoholic potash, using
2, c. c. for each gram of fat. Saponification was allowed to continue
for one half to one hour at the temperature of the boiling solution.
The soap was dissolved in three volumes of distilled water. After
cooling, the solution was shaken with an equal volume of pure ether
in a separatory funnel. The extraction was repeated with a fresh
volume of ether equal to one-half the volume of the soap solution.
It was found that this procedure would leave the soap colorless, if
no aldehyde resins had formed during saponification or none of these
colored bodies had been present in the alcoholic potash previous to its
addition to the fat. The ether extract containing the pigment and
other unsaponifiable matter was now freed from alkaline soap, by
shaking many times with excess water, carefully at first to avoid

1. Such as Hammarsten, "Text Book of Physiological Chemistry" and
Schaefer, "Text Book of Physiological Chemistry", etc.

2. Kirsten: Zeit. Nahr. Genuss. 5, p. 833 (1903).

3. Lewkowitsch: "Oils, Fats and Waxes." Vol. I, p. 371, (1909 Edition).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 34!

emulsions, and more vigorously with subsequent washings. When
the wash water no longer reacted alkaline toward phenolphthalein, the
ether was either dried over fused CaCl2 or, after standing several
hours, decanted from the precipitated moisture. The ether was then
evaporated, leaving a salve-like residue of various tints of yellow to
orange to red, depending upon the amount of fat used and the depth
of its original color.

Although the methods of study eventually adopted did not make
the procedure necessary, it was found possible to completely free
the pigment from its chief impurity, i. e., cholesterol, by means of
the- digitonin method of Windaus x for the quantitative estimation
of cholesterol. A hot one per cent solution of digitonin in ninety
per cent alcohol, when added to an alcoholic solution of the un-
saponifiable residues from butter fat, completely precipitated the
cholesterol as a colorless compound, leaving the pigment in solution.
This solution was still contaminated, however, with traces of fat and
lecithin decomposition products.

GENERAL PROPERTIES OF THE BUTTER FAT PIGMENT.

The unsaponifiable residues from butter fat, either crude or freed'
from cholesterol by digitonin, were readily soluble in hot alcohol and
in ether, chloroform, petroleum ether, etc. with a golden yellow color;
and in carbon bisulphide with a color which varied, according to the
concentration, from a red orange to a blood red color. The freshly
prepared crude, and also, the cholesterol-free residues usually gave
an instantaneous but transient purple color on adding a drop of con-
centrated sulphuric acid, a light green color quickly changing to a
greenish blue color on adding a drop of concentrated nitric acid, and
a dark blue color with the combined acids. These color reactions
were usually shown a little clearer by the cholesterol free residues.
Small amounts of impurities often interfered with the color reactions ;
in fact they were often rendered negative by some slight decompo-
sition of the pigment which was not apparent in the intensity of the
color. This was -due to the fact that the crude samples of butter fat
pigment were very unstable, quickly bleaching in the air, especially
with the aid of heat in the presence of a little water. It was necessary
therefore to take great care to have the ether solutions of the pigment
as free as possible from water before evaporation, or to transfer the
pigment to some solvent such as petroleum ether, which does not
absorb water so readily.

1. Zeit. f. Physiol. Chem. 65, p. 110 (1909).

342 MISSOURI AGRICULTURAL EXPERIMENT STATION, BULLETIN NO. IO

In addition to the above properties it was found that all fat-free,
but not necessarily cholesterol-free solutions of the pigment showed
spectroscopic absorption bands. In alcohol two sharp bands were
exhibited in the blue part of the spectrum; and in carbon bisulphide
these bands were nearly always accompanied by a third faint band
in the violet, which was now visible on account of the general shifting
of the bands toward the red end of the spectrum. The unstable char-
acter of the butterfat pigment required that the isolation be carried
out as rapidly as possible in order to preserve all the characteristic
properties of the pigment. This was especially true for the study of
the absorption spectra. Fifteen to thirty grams of fat were found
to yield sufficient pigment for a spectroscopic study. The use of
large quantities of fat (300 to 1,000 grams) always led to unsatis-
factory results.

The general properties of the butter fat pigment show that it is
to be classed as a true lipochrome. The chemical relation of the pig-
ment to the carotin and xanthophylls of green plants remains to
be shown.

METHODS OF IDENTIFICATION.

The nature of the substances with which the butter fat pigment
is associated at once precluded its isolation in sufficient quantity to
establish its chemical composition and molecular weight. It was
therefore necessary to adopt other methods of identification which
would be sufficiently accurate and characteristic that the final results
could not be mistaken.

The methods that were adopted were, (i) a study of the spec-
troscopic absorption properties, (2) a study of the relative solubility
properties, (3) a study of the adsorption properties with respect to
calcium carbonate; (4) an attempt was also made to study the crys-
talline form.

Before giving the results of our studies, some discussion will be
given of the relative solubility and adsorption properties of carotin
and the xanthophylls.

RELATIVE SOLUBILITY OF CAROTIN AND XANTHOPHYLLS.

M. Tswett * was the first one to publish a comprehensive state-
ment in regard to the relative solubility of the plant pigments in

1. Ber. der. Deut. Botan. Gessel. 24, pp. 316, 384 (1906).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 343

different solvents. He classified the solvents into three groups as
follows according to their relation to the leaf pigments.

"i. Alcohols (methyl, ethyl, amyl), acetone, acetaldehyde, ether,
chloroform: These solvents, acting on freshly cut up or dried leaves
dissolve out all pigments equally and completely."

"2. Petroleum ether or petroleum benzin: This solvent, acting
on fresh leaves finely ground with sand or emery, takes on a
more less yellow appearance, which is especially due to carotin,
but contains also other pigments. Dried leaves (at a low temperature)
likewise give up their carotin to this solvent, and in somewhat purer
condition."

"3. Benzol, xylol, toluol, and carbon bisulphide: These solv-
ents set intermediately between the first two groups."

Willstatter and Mieg J a little later approached the same problem
from another standpoint and showed that the methods used by Kraus 2
and Sorby 3 for demonstrating the presence of more than one pigment
in green plants, when properly applied could be made characteristic
properties of carotin and xanthophylls. Kraus shook his alcoholic
extracts of green leaves with petroleum ether and found that the
green pigment went into the petroleum ether leaving the alcoholic
solution yellow. Sorby shook his alcoholic extracts with carbon
bisulphide and found that the latter solvent contained the green pig-
ment while the alcohol was left yellow. Wilstatter and Mieg, apply-
ing these tests to the isolated carotin and xanthophyll pigments obtained
the following results:

"i. If methyl alcohol is added to a petroleum ether solution
of carotin so that the liquids do not mix, the carotin will remain for
the greatest part in the petroleum ether layer, the alcohol layer being
only slightly colored. If a trace of water is added, the methyl alcohol
layer will become colorless. The same phenomenon occurs with ethyl
alcohol, and one can start with an alcohol or benzol solution and
show the same thing."

"2. If carbon bisulphide is added to an alcoholic carotin solu-
tion and a little water added, the carbon bisulphide will separate and
will quantitatively contain the carotin."

"3. If an alcoholic solution of the xanthophylls is mixed with
petroleum ether and the liquids separated with a little water, by far the
greatest portion of the xanthophylls will be found in the alcohol layer."

1. Ann. der. Chemie 355 p. 8 (1907).

2. Flora, p. 155 (1875).

3. Proc. Roy. Soc. 21, p. 456 (1875).

344 MISSOURI AGRICULTURAL EXP. STA._, RESEARCH BULLETIN NO. IO

"4. If an alcoholic solution of xanthophylls is mixed with carbon
bisulphide and the solvents separated with water, the xanthophylls will
be distributed between both layers."

Schunck,1 using Sorby's method, showed that if alcoholic solu-
tions of xanthophylls are repeatedly shaken with carbon bisulphide
all the xanthophylls with spectroscopic absorption properties can be
extracted.

Tswett2 studying this question again in 1911 states that, "Accord-
ing to a well known rule, organic compounds are best soluble in solv-
ents of similar composition," and concludes that carotin, a hydrocarbon,
is therefore much more readily soluble in the hydrocarbons of the
aliphatic and cyclic group than in alcohols, a point well illustrated
by the above experiments of Willstatter and Mieg. Continuing, Tswett
states, "If one therefore shakes an eighty to ninety per cent alcoholic
solution of carotin with petroleum ether, the pigment goes almost
completely into the petroleum ether layer." "A pigment which in
the above mentioned two phase system occupies the lower alcoholic
layer, is therefore not a carotin." It may be added in the light of
Willstatter and Mieg's investigation that if the original solution before
differentiation was a mixture of carotin and xanthophylls, the lower
alcoholic layer will contain the xanthophylls.

ADSORPTION PROPERTIES OF CAROTIN AND XANTHOPHYLLS.

Considering now the so-called adsorption properties of the pig-
ments, we find that this striking characteristic was discovered and
elaborated by Tswett.3 This investigator found that by shaking a
perfectly anhydrous petroleum ether or carbon bisulphide solution of
the mixed pigments of green plants with an excess of dry calcium
carbonate, Inulin or Saccharose, all the pigments will be completely
adsorbed by the material with the exception of the carotin, which can
be readily washed out of the material with the free solvent. In the
case of petroleum ether solutions the green colored mass can now
be completely freed from all its pigments by means of petroleum ether
containing ten per cent absolute alcohol. If the resulting solution
is now shaken with eighty per cent alcohol the petroleum ether layer
will contain the chlorophyll pigments and the alcohol layer the xan-
thophylls. "If to the petroleum ether solution of the mixed pigments

1. Proc. Roy. 72 (1903).

2. Ber. der. Deut. Botan. Gessel. 29, p. 630 (1911).

3. Ber. der. Deut. Botan. Gessel. 24, p. 316 and 384 (1906).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 345

is added adsorption material only sufficient to destroy the fluorescence,
both the carotin and xanthophylls remain in solution," and can be
separated by means of a differentiation between the petroleum ether
and eighty per cent alcohol, or, "By treating the solution with more
adsorption material, after pouring it away from the first, and the
xanthophylls then freed from combination with the adsorption material
by means of alcoholic petroleum ether/' In addition to the above,
Tswett made the interesting discovery that the pigments which are
adsorbed by the various materials suggested, can to a certain extent
displace one another in the adsorbing material. As an example one
finds that, "If a petroleum ether solution of the mixed pigments is
filtered through a column of adsorption material (such as CaCO3
packed tight in a glass tube) the pigments will be separated from one
another from top to bottom in differently colored zones, proportion-
ately to their degree of adsorption." This
separation will be complete if a stream of
pure solvent is put through the column
after the pigment has been adsorbed in
the upper part of the column. As stated
by Tswett, "Like the rays of light in the
spectrum, so the different components of
a pigment mixture are actually separated
in the CaCO3 column, and may thus be
qualitatively estimated." Tswett calls
such an experiment a "chromotogramm."
He found carbon bisulphide to be one of
the most useful solvents for a chromoto-
graphic analysis, on account of the bril-
liant color which all pigments show in
this solvent.

In describing the technique for the
chromotographic analysis, the author
mentions the following essential points.
A very finely divided material with not
too strong adsorption properties should
be used for the adsorbator. (CaCO3 was
found to answer these qualifications
best.) A glass tube is now prepared 10 to
20 m. m. in diameter and 15 to 20 c. m.
long, one end of which is drawn out to

a narrow diameter, at which end the opening is fused in a little to
form a base for the deposition of the adsorbing material. A small

FIGURE I.

346 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

piece of cotton is placed in the small end of the tube and the per-
fectly dry CaCO3 poured in and firmly tamped down to a homogenous
texture. The chromotographic apparatus may now be arranged accord-
ing to Figure I, and the filter flask attached to the suction pump. The
CaCO3 is now moistened with a little of the solvent to be used, (this
is very necessary) and a certain amount of the liquid which is to be
studied poured on the CaCO3. A stream of pure solvent is subse-
quently established and the different adsorption zones will then spread
out and reach their definite maximum differentiation. All unadsorbed
substances will be completely washed away and, "Substances which
form truly dissociable adsorption compounds with the CaCO3 pass
slowly, 'ringwise' through and can be taken up each by itself at the
mouth of the tube."

Carotin and Xanthophyll of Green Plants.

After selecting the methods of differentiation and characteriza-
tion to be applied to the milk fat pigment, it was considered desirable
to ascertain whether they were sufficiently characteristic for a com-
plete identification, should the milk fat pigment be found to be either
a carotin or a xanthophyll. The following experiment was accord-
ingly carried out.

About fifteen grams of air dry and finely divided alfalfa hay
which had a deep green color, was let stand for several days, with
shaking, under pure carbon bisulphide. The resulting deep olive brown
fluid was concentrated to about 25 c. c. at a low temperature. A
glass tube about eight inches long and one-half inch in diameter,
the last two inches of which were drawn out to a small opening,
was now filled and packed with pure CaCO3, which had been pre-
viously dried for two hours at 150° C. The CaCO3 was tamped
in a small portion at a time by means of a small cotton wad and a
heavy glass rod. The small end of the tube was now inserted through
a one hole rubber stopper and fitted tightly into a side neck test tube.
The apparatus was then attached to a suction pump. A stream of pure
carbon bisulphide was passed through the column. When the CaCO3
had become thoroughly moistened, 2 to 3 c. c. of the alfalfa extract was
poured into the top of the column, vigorous suction being maintained
all the time. When the extract had passed entirely into the CaCO3
and occupied about one inch of the column, a stream of pure carbon
bisulphide was run through. As the pigment passed through it dif-
ferentiated itself into a number of green and yellow zones, the least

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 347

adsorbed pigment at the bottom of the differentiation having a rose
color. The stream of carbon bisulphide was continued until the rose
colored solution began to drop from the lower end of the tube. The
appearance of the column at this time is shown in Figure II. The
stream of carbon bisulphide was now continued until the rose colored
zone had entirely passed through.

According to Tswett this zone contained only carotin. The beau-
tiful rose or red orange solution that was obtained was studied as
follows : It was first examined in the spectroscope where it showed two
distinct bands, and a faint third one. (See Table i.) The carbon bisul-
phide solution was then evaporated. The
residue was on orange yellow solid. A
portion of it gave a beautiful blue colora-
tion with concentrated H2SO4.

The remainder of the residue was
dissolved in hot ninety-five per cent alco-
hol, and the phystosterol which precipi-
tated out on cooling filtered off. The fil-
trate was divided into two portions, pe-
troleum ether (b. p. 30-50° C) being
added to one and pure carbon bisulphide
to the other. On separation of the re-
spective solvents with a little water, the
pigment was found quantitatively in the
petroleum ether and carbon bisulphide
layers respectively. The pigment was
again put into alcohol, and in this solv-
ent showed two strong absorption bands
and end absorption. (See Table i.)

Potassium hydroxide was now added
to the alcoholic solution and the solution
boiled for several hours. The alkaline
solution was diluted with three volumes
of distilled water and extracted with an
equal volume of ether. The ether com-
pletely and readily extracted the pigment, FIGURE II.
showing it to be unsaponifiable. The golden yellow ether solution was
washed free from alkali with water and evaporated to dryness. The
residue was taken up with alcohol. This solution showed only two
bands. (See Table i.)

The alcoholic solution was now tested again for its relative
solubility toward petroleum ether (b. p. 30-50° C) and carbon bisul-

3

Colorless.

Zone I: Yellow.

Colorless.

/ Zone II:
I Yellow.

Colorless.

Green Zone.

/Zone III:

\ Yellow.
Colorless.

/ Zone IV:

\ Orange- Yellow.

f Greenish-

1 Brown Zone.

Zone V: Rose.

Cotton Ping.

348 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

phide. The pigment was again found quantitatively in the petroleum
ether and carbon bisulphide. A portion of pigment was again put into
alcohol and the solution made strongly alkaline with solid sodium
hydroxide. On addition of much sodium chloride to this solution
the pigment was not precipitated. The remainder of the pigment
was now dissolved in carbon bisulphide, after evaporation of the solu-
tion (petroleum ether), and the carbon bisulphide solution filtered
through the CaCO3 column again. It passed through unadsorbed
as a rose colored zone.

TABLE 1. ABSORPTION (a) BANDS OF ALFALFA CAROTIN

In CS2

In C2H6OH

In C2H6OH
(after ^
saponification)

Band I
Band II
Band III

225-245
259-279
300-320

256-275
300-320
345-. . .

256-277
303-320

(a) Note: It should be noted that all spectroscopic measurements, both
this and subsequent ones, were made according to an arbitrary scale which was
attached to the spectrometer. This scale was always set to a fixed standard be-
fore studying each pigment, the standard being produced by a sodium flame
with the spectrometer slit closed to furnish the narrowest possible line. The
spectrometer was equipped with a crown glass prism and lenses and had a nar-
row dispersion.

The study of the alfalfa carotin showed conclusively that its
adsorption, spectroscopic and solubility properties were clear and
characteristic and were unchanged by boiling in alcoholic potash.
It was also shown that the pigment could not be salted out of its
sodium alcoholate solution with common salt when the pigment was
free from fat. If the solution had contained much soap the pigment
would in all probability have been precipitated with the soap in
the salting out process. The object of the test was to see whether
this was or was not a characteristic test for a comparatively pure car-
otin. Newbigin 1 claimed to have found a true lipochrome which
could be salted out of its alkaline solution.

1. D. Ndel Paton. "Investigations on Life History of the Salmon." 1898,.
Art. XV. p. 159.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 349

After studying the carotin pigment, more carbon bisulphide was
run through the column shown in Figure II. until the lower green zone
was washed out. This was removed from the test tube and the stream
of carbon bisulphate continued until the two least adsorbed orange-yel-
low zones had been washed out. The carbon bisulphide solution of these
pigments had a golden yellow color with a faint green tinge. Before
the spectroscope the solution showed two chlorophyll bands in the
red, and three very clear bands in the blue. (See Table 2.)

The carbon bisulphide was evaporated off and the residue dis-
solved in hot eighty per cent alcohol, the phytosterol which precipi-
tated on cooling being filtered off. The alcoholic filtrate was now
extracted with petroleum ether (b. p. 30-50° C), which took up some
green color. This was continued until no more green was extracted.
The golden yellow alcoholic xanthophyll solution was examined in
the spectroscope. No chlorophyll bands were visible but only two
strong bands in the blue and sharp end adsorption.

The alcoholic solution was now saponified with potassium hydrox-
ide and the diluted soap extracted with ether in the usual way. The
color immediately went into the ether, the separation being complete
with one extraction. After washing free from alkali the ether was
evaporated. A portion of this residue as well as a portion of the res-
idue of the solution before saponification, showed a beautiful green-
ish blue coloration with concentrated H2SO4. The remainder when
dissolved in alcohol, gave a golden yellow solution showing two bands
and end absorption. (See Table 2.)

TABLE 2. ABSORPTION BANDS OF ALFALFA XANTHOPHYLLS

InCS2

In C2H6OH

In C2H6OH
(after
saponification)

Band I
Band II
Band III

232-250
273-293
312-330

265-282
306-326
357-. ..

265-286
306-326
355-. . .

A little HC1 added to a portion of the alcoholic solution gave no
blue coloration. Solubility tests on the saponified pigment showed
that petroleum ether extracted no color from an alcoholic solution on
dilution with a little water, while carbon bisulphide extracted about an
equal portion.

35O MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. IO

These tests showed conclusively that this pigment belongs to the
oc group of xanthophylls ; and that its adsorption, solubility and
spectroscopic properties which are characteristically different from
those of carotin, are unaltered by treatment with alcoholic potash.

The remaining xanthophyll pigment was so firmly held in com-
bination with the CaCO3 that carbon bisulphide would not wash it
out. A stream of ten per cent alcoholic petroleum ether was there-
fore run through the column, washing out all the remaining pigments.
The resulting solution showed only faint absorption bands, indicat-
ing that the pigments noted above, i. e. carotin and xanthophyll oc,
are the principal yellow pigments of the alfalfa hay.

IDENTIFICATION OF THE PIGMENT OF BUTTER FAT.

With the properties of carotin and xanthophylls well established,
attention was next directed to the butter fat pigment. The follow-
ing experiments were carried out, the results of which are very
striking.

Experiment i.

Fifteen grams of very yellow butter fat from a Jersey cow who
was on fresh, green, fall grass was saponified in the usual way with
alcoholic potash, taking great care to avoid the presence or formation
of the colored alcohol decomposition products, the aldehyde resins.
After dilution the soap was extracted with ether. The ethereal extract
was washed free from alkali with distilled water and evaporated into
ninety-five per cent alcohol. An equal volume (100 c. c.) of petroleum
ether (b. p. 30-50° C.) was now added to the alcoholic solution and
just enough water to cause a separation of the alcohol and petroleum
ether. The golden yellow petroleum ether layer which resulted, con-
tained practically all the color. The alcohol layer was drawn off
and extracted with fresh volumes of petroleum ether until only a
trace of color went into the petroleum ether layer. All the petroleum
ether extracts were then combined and extracted with eighty per
cent alcohol. A mere trace of color went into the alcohol. The
alcohol solutions were combined.

The Petroleum ether solution — This contained by far the greatest
part of the total pigment. The solution was evaporated quickly at
a temperature below 50° C., leaving a red oily residue which instantly
dissolved in carbon bisulphide with a deep red orange color. After
adjusting the concentration for the 10 m. m. cell, so that the bands

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 351

were all plainly visible, this solution showed three distinct absorption
bands. (See Table 3.)

The carbon bisulphide solution was evaporated to dryness at the
lowest possible temperature, the residue taken up in 50 c. c. of hot
ninety-five per cent alcohol and hot one per cent digitonin in ninety
per cent alcohol added until no more precipitate came down. The
digitonin-cholesteride was filtered off, the filtrate evaporated to dry-
ness, and the residue dissolved once more in carbon bisulphide. The
solution still showed the three bands. On evaporation it left a golden
yellow oil which solidified on cooling to a reddish yellow salve. Con-
centrated H2 SO4 added to the residue gave a blue green color which
slowly changed to a purple color.

The alcoholic solution. — This was evaporated to dryness. When
very concentrated it showed a little color, and the carbon bisulphide
solution of the residue had a light orange color when it had a volume
of i % c. c. When viewed in a 25 m. m. cell this solution showed
three absorption bands. (See Table 3.)

TABLE 3. ABSORPTION BANDS OF BUTTERFAT PIGMENTS.

Petroleum ether soluble pigment
In CS2 solution

Alcohol soluble
pigment in
CS2 solution

Crude

Free from
Cholesterol

Band I
Band II
Band III

222-240*
257-276
299-319

224-242
260-277
301-316

229-247
269-288
309-328

The striking results of this experiment were the remarkable
similarity of the solubility and spectroscopic properties of the main
butter fat pigment to carotin, and the indications of a secondary minor
constituent of the butter fat pigment, whose solubility and spectro-
scopic properties were strikingly similar to xanthophyll.

It at once became evident that should these observations be con-
firmed, it would be not only profitable, but essential to ascertain
whether the presence of secondary xanthophyll-like pigments is nor-
mal to butter fat under other conditions of coloration, such as in
light colored butter fat, colostrum butter fat and other conditions.

That the presence of a secondary pigment in the fat under inves-
tigation was confirmed and its character more clearly identified is
shown by the following experiment.

352 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

Experiment 2.

Fifteen grams of the fat was treated as in Experiment i. In ad-
dition the carbon bisulphide solution of the unsaponifiable ether extract
was filtered through a column of CaCO3 in a manner identical with the
chromotographic experiment with alfalfa hay. As far as could be de-
tected with the eye all the pigment passed quite rapidly through as an
unadsorbed rose colored zone, which spread out considerably in its
passage through the column but showed no differentiation into zones.
The absorption bands of the filtered pigment were identical with the
bands of carotin.

After the carbon bisulphide had washed out all the pigment and
was passing through colorless, a stream of petroleum ether contain-
ing ten per cent alcohol was run through the column. As it passed
through it gathered a zone of yellow color, leaving the column pure
white. This pigment was collected at the mouth of the tube, its solu-
tion evaporated, and the residue dissolved in carbon bisulphide. The
light-orange colored solution showed two strong absorption bands and
a third fainter one.

The carbon bisulphide solution of the main pigment was now evap-
orated to dryness and the residue dissolved in ninety-five per cent
alcohol. The alcohol was diluted with water to an eighty to ninety per
cent solution and extracted with petroleum ether (b. p. 30-50° C). The
bulk of the pigment went into the petroleum ether and a second
extraction with fresh petroleum ether took out still more pigment. A
third extraction with fresh petroleum ether, however, left the alcohol
layer considerably more colored than the petroleum ether layer. The
alcohol layer was now evaporated and the residue dissolved in
carbon bisulphide, giving an orange yellow solution which showed
three strong absorption bands. There seemed to be three or four
times as much of this pigment as of the xanthophyll which had been
adsorbed by the CaCO3 in the chromotogramm, and together they
probably amounted to ten per cent of the total pigment.

All the xanthophyll pigments were now combined (they were
all in carbon bisulphide solution) and the resulting solution analyzed by
means of a chromotogramm. As the orange-yellow solution was washed
through the column by a stream of carbon bisulphide it took on
the appearance as shown in Figure III. Zones two and three were
collected together, and showed three absorption bands. (See Table 4.)
Their solution was evaporated and the residue dissolved in petroleum

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 353

ether. Only a part of the residue would dissolve readily, and the
remainder only on addition of a little absolute alcohol. Eighty per cent
alcohol extracted a large part of the color from the alcoholic petroleum
ether solution. The portion which remained in the petroleum ether was
transferred to carbon bisulphide in which it showed three absorption
bands. (See Table 4.) The portion extracted by the eighty per

cent alcohol was also transferred to car-
bon bisulphide. The latter solution showed
three absorption bands but the third was
faint and was not measured. (See Ta-
ble 4.)

Alcoholic solutions of both portions
of pigment showed no color change on
the addition of a little concentrated HC1.
There was also no effect on the absorp-
tion bands.

Zone i of the chromotogramm was of
a pure yellow color. It was completely
adsorbed by the CaCO3 with respect to
CS2 and was evidently the same pigment
which had been adsorbed in the first chro-
motogramm of the combined carotin and
xanthophyll-like pigments. A stream of
alcoholic petroleum ether readily washed
it out of the column as it did in the first
chromotogramm. In carbon bisulphide so-
lution the pigment had a light orange
color, and showed two brilliant absorp-
tion bands and a third fainter one. (See
Table 4.)

The pigment showed the three bands
in alcoholic solution as well as in car-
bon bisulphide. When in alcohol, it gave
no color reaction with a little concentrated

HC1, and there was also no immediate effect upon either the intensity
or position of the absorption bands. In the solid state this pigment
gave a transient greenish-blue color with concentrated H2SO4.

Colorless.

Zone I:
Yellow.

Colorless.

Zone II:
Orange,
shading
from light
to dark.
Zone III:
Deep
Orange.

Colorless.

Cotton Plug.

FIGURE III.

354 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. IO

For the sake of comparison with the carotin bands of the alfalfa
hay, the alcoholic solution of the main butter fat pigment was exam-
ined in the spectroscope. The results are given in Table 5.

TABLE 4.

ABSORPTION BANDS OF BUTTER FAT XANTHOPHYLLS SHOWN IN
FIGURE III.

Zone I.
CSa solution

Zones II & III
C2H6OH solution

Zones II & III

(P. ether sol.
Part)
InCS2

(Alcohol
sol. Part)
InCS2

Band I
Band II
Band III

232-249
271-288
313-330

263-280
306-325
355-. . .

230-249
268-289
312-330

231-250
272-240
Not measured

TABLE 5. COMPARISON OF BANDS OF CAROTIN OF ALFALFA AND BUTTERFAT.

(C2H6OH SOLUTION).

Alfalfa Carotin

Butterfat Carotin

Band I

256-275

256-274

Band II

300-320

298-312

Band III

345-. . .

345-. . .

The result of this experiment was not only to confirm the re-
markable similarity of the main butter fat pigment of this partic-
ular fat to the carotin of alfalfa hay, but also to confirm the presence
of a secondary constituent practically identical with the xanthophylls.
In addition it was found that these xanthophylls, like the xanthophylls
of alfalfa hay, could be chromotographically separated into three
constituents. The two main constituents seemed to be closely related
to carotin in adsorption properties so that their presence could not
be detected in the presence of a large amount of carotin until first
separated from the main pigment with the aid of their relatively
greater solubility in alcohol with respect to petroleum ether. The
third constituent of the secondary group was more nearly related
to a true xanthophyll in all its properties, including its adsorption by
CaCO3. When classified with respect to the action of their alcoholic so-
lutions toward a little concentrated HCl all the xanthophyll pigments
apparently belong to the group which Tswett calls oc xanthophylls.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 355

STANDARDIZATION OF ABSORPTION BANDS OF CAROTIN AND

XANTHOPHYLLS.

Before proceeding further it may be well to make a few state-
ments in regard to the spectroscopic properties of the various pig-
ments which have been under consideration and which are still to
be considered. It will be obvious that measurements of absorption
bands could not be made in this work with the accuracy attained
by Willstatter and his collaborators who used solutions of standard
strength and a spectroscope of great exactness. Small differences
in measurement among xanthophylls or between carotins should ac-
cordingly be disregarded. It was merely attempted in every case
to secure solutions of such concentration that the bands were of as
nearly the same intensity as could be detected with the eye before
making measurements. It was not always possible to use the same
thickness of cell to secure the required intensity of the bands, small
amounts of pigment naturally requiring a greater depth of solution
than large amounts.

In order to have standard spectroscopic properties for future
comparison, carotin and xanthophylls were extracted from the car-
rot, and a careful study made of the spectroscopic properties of each
pigment.

A grated carrot was boiled in water for about two hours, the
water squeezed out of the pulp and the pulp dried on the steam
bath. It was pulverized and extracted with ether in a Soxhlet extrac-
tor until colorless. The ether was evaporated into ninety-five per
cent alcohol at a low temperature. The resulting solution was diluted
with a little water to eighty per cent and the pigment carefully
separated between petroleum ether (b. p. 30-50° C) and the eighty
per cent alcohol. Both portions of the pigment were carefully trans-
ferred to pure carbon bisulphide and the solutions adjusted for the
10 m. m. cell until all bands were distinct and as nearly as possible
of equal intensity. With the spectrometer set at the sodium line stand-
ard the positions of the absorption bands were standardized. The
color of the solutions was measured also, by means of the Lovibond
tintometer. (See Figure V.)

356 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

This data is given in Table 6.

TABLE 6. SPECTROSCOPIC STANDARD OF CAROTIN AND XANTHOPHYLLIS. (FROM

THE CARROT.)

Pigment

Layer

Color Absorption Bands

Yellow

Red

Light

Band I

Band II

Band III

InCS2

10 m.m
10 m.m

26.0
17.0

5.5
4.4

1.0
0.5

225-242
233-253

261-278
272-291

301-319
312-330

Carotin
Xanthophyll

In C2H6OH

10 m.m
10 m.m

257-275
263-280

303-318
305-325

345-364
355-. . .

Carotin
Xanthophyll

It will be noticed that the relative position of the bands of car-
otin and xanthophylls is more characteristic in carbon bisulphide
than in alcohol. On this account, nearly all subsequent spectro-
scopic studies were made in carbon bisulphide solution.

It will be noticed that both pigments showed three bands. The
difference in the color of the carbon bisulphide solutions of the
carotin and xanthophyll pigments when showing bands of equal inten-
sity is also especially noteworthy. To the eye, the carotin solutions
are a deep red orange while the xanthophyll solutions are a much
purer orange color. It is worthy of mention that this characteristic
difference was so noticeable throughout all this study, that in all
cases the color of the carbon bisulphide solution of unknown pig-
ments was recorded in connection with the measuring of the absorp-
tion bands.

CHARACTER OF THE PIGMENTS IN DIFFERENT BUTTER FATS.

It was stated previous to the confirmation of the presence of
secondary xanthophyll-like pigments in the particular butter fat
studied, that if the presence of these pigments could be confirmed
it would necessary to establish such a fact as either characteristic
of all butter fat or as incidental only to the particular fat studied.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 357

A wide variety of fats of different color, from different breeds of
cows and produced under widely different conditions of feeding, etc.,
were available for this study. The technique of these experiments
was identical with that used for the high colored fat from the
Jersey cow on grass recorded in Experiments I and 2 above.

The Pigments of Light Colored Fats. — The character of the pigment
in three light colored fats was tested. The fats represent three
subsequent periods in a feeding experiment of a pure bred Ayrshire
cow in which the color was practically eliminated from the butter
fat. (See Table 13 of feeding experiments for record of this experi-
ment.) The color of the fat and the character of the pigments
are shown in the following Table 7.

TABLE No. 7.

Color of Fat

Yellow Red

6.0

2.5

1.4

1.0

0.7

0.5

Character of Pigments.

Petroleum ether quant, extracted cholesterol-free pig-
ment from 80 per cent alcohol.

Solubility test showed carotin and xanthophylls.

Chromotogramm of xanthophylls showed adsorbed
yellow constituent and unadsorbed orange zone.

Pigment from 240 gm. fat had color of 35 yellow, 1. 0
red. Solubility test showed both carotin and xan-
thophylls, as did also absorption bands.

The Pigments of Butter Fat After Carrot Feeding. — The fat tested
was taken during a carrot feeding experiment with the same cow
used in the above experiments. This feeding experiment directly
followed the third period of very light colored fat. The color of the
fat was 28 yellow and 1.4 red. Solubility tests on the pigment
showed carotin and a very small amount of xanthophyll. The petro-
leum ether soluble part of the pigment gave a CS2 chromotogramm
of a single unadsorbed rose colored zone. The filtered pigment
showed three well defined absorption bands. I, 225-245; II, 263-
283; III, 301-320.

358 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

The Pigments of the Fat from Colostrum Milk. — It is a well known
fact that the first milk drawn after parturition always has a high
yellow color. It is not generally known, however, that this high
color is usually due 1 entirely to the suspended fat globules. We have
many times observed, not only in connection with this study but
also in connection with numerous studies dealing with the chemical
composition of milk, that when the fat is entirely removed from
colostrum milk the skim milk has the appearance of ordinary skim
milk, and the butter and the rendered fat have a depth of color
which is never equaled at any subsequent stage of the lactation period.
This characteristic of colostrum milk is common to all breeds of cows,
and the high color of the fat continues in cows of all breeds for a
short time after parturition and then gradually falls off. Table 8 gives
the color of the milk fat of several cows shortly after parturition and
again a week or two later. The color readings are the Lovibond tin-
tometer readings of a one-inch layer of melted, rendered fat.
TABLE No. 8. COLOR OF THE FAT OF COLOSTRUM MILK.

Cow No.

Breed

Roughage

lCQ»

Days after
Parturi-
tion.

Color

Yellow

Red

Light

301

Ayrshire

Alfalfa(a)

4

78.0

3.5

1.0

301

Ayrshire

Alfalfa

26

71.0

1.5

0.5

300

Ayrshire

Alfalfa

4

71.0

3.5

1.0

300

Ayrshire

Alfalfa

20

68.0

2.8

1.0

2

Jersey

Alfalfa

13

68.0

2.6

0.5

2

Jersey

Alfalfa

22

57.0

2.5

0.5

2

Jersey

Alfalfa

2

54.0

4.3

1.0

2b

Jersey

Alfalfa

20

50.0

2.5

1.0

20

Jersey

Alfalfa

2

47.0

4.8

1.0

20

Jersey

Alfalfa

25

47.0

2.0

0.5

206

Holstein

Alfalfa

1

50.0

4.7

0.3

206

Holstein

Alfalfa

5

54.0

2.0

0.2

(a) The Alfalfa hay was rich in carotin and xanthophylls.

(b) Second sample taken after next parturition.

This phenomenon at once offered the interesting problem of

the relation of the colostrum pigment to the pigment of normal

butter. A close study was accordingly made of the pigment of the

fat from the first milk drawn after parturition. The cow selected

1. Colostrum milk is occasionally contaminated with blood.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 359

for study was a pure bred Holstein. The fat tested had a very high
color; a one inch layer gave a reading of 64 yellow, 5.0 red and
i.o light in the Lovibond tintometer.

The unsaponifiable pigment and impurities from fifteen grams
of the fat had a golden yellow color in ether and in carbon bisulphide
a blood red color. This solution was analysed chromotographically.
The entire pigment passed through unadsorbed as a red orange or
rose colored zone, leaving no adsorbed zones and no pigment behind
in the CaCO3 which could be washed out with ten per cent alcoholic
petroleum ether. The filtered solution showed two strong absorption
bands and a third faint one. (See Table 9.)

The pigment was now analyzed according to its proportionate
solubility in petroleum ether (b. p. 30-50° C) and eighty per cent
alcohol, and was thus divided into two portions, a major portion ex-
tracted by the petroleum ether and a very minor portion which the pe-
troleum ether would not extract from the eighty per cent alcohol.

The carotin-like pigment thus obtained was freed from choles-
terol by the digitonin method and its bands again measured in carbon
bisulphide solution. (See Table 9.)

The residue from this solution gave a beautiful transient blue
color with concentrated H2SO4 and a very transient blue-green color
with concentrated HNO3.

The eighty per cent alcohol soluble pigment showed three absorp-
tion bands in carbon bisulphide solution, the first two being a little
more intense than the third. (See Table 9.) While in this solution
the pigment was analyzed by means of a chromotogramm and showed
two zones, a primary little adsorbed zone of orange color, and a sec-
ondary more adsorbed zone of yellow color. Hydrochloric acid gave
no color reaction with the alcoholic solution of the pigment of either
zone.

TABLE 9. ABSORPTION BANDS OF PIGMENTS OF COLOSTRUM MILK FAT.

Combined Pigment
(CS2 Solution)

Carotin
(CS2 Solution)

Xanthophylls

(CS2 Sol.)

(C2HBOH Sol.)

Band I
Band II
Band III

223-240
260-278
300-320

224-242
259-278
302-319

232-249
272-291
312-330

264-281
306-326
356-. . .

Crystalline Form of Carotin From Butter Fat. — The great concen-
tration of pigment in the fat from colostrum milk seemed to offer
an excellent opportunity to at least attempt the isolation of the pig-

360 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

ment in crystalline form. The pigment was isolated in the usual
way from forty grams of very high colored colostrum fat from a
Jersey cow. Great care was taken during the isolation to avoid
aldehyde resin pigments. This was successful. The ether solution
of the unsaponifiable substances was evaporated at 35° C. and the
residue dissolved at once in carbon bisulphide. This solution, which
had a blood red color, was concentrated to 2 c. cm. volume at a low
temperature and an excess of cold absolute alcohol added. There
was no immediate crystallization, but after standing several days
there were deposited a number of yellow crystals. These were at
first thought to be crystals of the pigment, for similar crystals were
obtained in the same manner from the corpus luteum pigment. The
crystals were perfectly formed double pyramidal forms, but proved
to be crystals of sulphur which evidently arose from the carbon bisul-
phide used. A further attempt at crystallization of the pigment was
abandoned. It undoubtedly would prove successful if sufficient pure
material could be obtained.

THE RELATION BETWEEN THE COLOR OF THE MILK FAT
AND THE FOOD OF THE COW.

General observation for no doubt hundreds of years, at least,
ever since butter has become of importance in the diet of man, has
shown that green feeds of all kinds, especially fresh green grass
greatly increase the color of butter fat. Other feeds, such as car-
rots, beets and yellow corn have been said to have the same effect.
It has never been the subject of a scientific investigation however,
to show just what relation exists between the food of the cow and
the color of the milk fat. With a chemical relation established be-
tween the milk fat pigments and carotin and xanthophylls, the rela-
tion of the color of the milk fat to the food seems to be readily
explained on the ground that the foods that have been observed to
cause the highest colored butter, namely, green foods, carrots, etc.
are those which are especially rich in carotin and xanthophylls, par-
ticularly carotin, as in the case of the carrots. Indeed we can go
still further and definitely state that these pigments must be abund-
antly present in the food before the milk fat will show a high color,
as will be demonstrated by the experiments which are about to be
reported.

Character of the Yellow Pigments of the Common Cattle Feeds. —
The chemical study of the yellow pigment of milk fat, shows that
its principal constituent belongs to the hydrocarbon or carotin group
of pigments, although it also contains xanthophylls as a very minor
secondary constituent. It accordingly became necessary to study the

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 361

nature of the pigments of various common cattle foods before con-
ducting any feeding experiments to prove that there is a direct rela-
tion between the presence of carotin in the butter fat and the presence
of carotin in the food. It was at first merely sought to show the
presence or absence of unsaponifiable yellow pigments in various
foods, by extracting them with hot alcohol, saponifying the extract
with potassium hydroxide and extracting the pigment from the soap
with ether. From this standpoint, it was found that among the
grains and concentrates, wheat bran, linseed meal, dried brewer's grains
and cottonseed meal all showed the presence of small amounts of un-
saponifiable yellow pigments, while white corn was found to be the only
common grain absolutely free from such pigments. Yellow corn was
of course found to be rich in unsaponifiable yellow pigments. Some
roughages such as corn silage and cottonseed hulls were found to
contain small amounts of unsaponifiable pigments, while wheat straw
and oat straw were practically free from them. The hays were
found to be the most variable of all feeds. All green hays,1 such
as alfalfa, first-class clover and the very best timothy were found
to contain considerable amounts of unsaponifiable yellow pigments,
the amounts varying with the greenness of the hay. Bleached hays,
whether timothy, clover, or alfalfa were more or less free from un-
saponifiable yellow pigments.

Some feeds were investigated more particularly with a purpose
of showing the character of the unsaponifiable yellow pigments. With-
out reporting the experimental details in all cases, but merely stat-
ing that the methods of analysis were spectroscopic, chromotographic,
and solubility methods, the following results were obtained.

Cotton Seed Meal and Cotton Seed Hulls. — It was found that the
unsaponifiable yellow pigments of cottonseed meal and cottonseed
hulls were due entirely to the oil they contain; and further that this
oil which is known to be characterized by its yellow color contains
equal proportions of carotin and xanthophyll. The carotin is the
usual one met in other places, while the xanthophyll is made up of
at least five different constituents according to their adsorption prop-
erties. The chief xanthophyll is not adsorbed to any extent by CaCO3
from its carbon bisulphide solution, and in this solvent shows absorption
bands of characteristic position, shifted considerably toward the blue
from the normal xanthophyll bands. In CS2 the bands measured:
I, 238-260; II, 285-303; III, 355—. The remainder of the xanthophylls
were so firmly held by the CaCO3 that they could not be readily

1. By green^ hay is meant hay that has been cured under such conditions
that it still retains a large part of the green color which characterizes its uncut
condition. The green alfalfa hay referred to throughout this paper was west-
ern cured alfalfa which had a remarkably bright green color.

362 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

washed out with a stream of carbon bisulphide. They were all
washed out however, by a one per cent alcoholic petroleum ether
solution.

Bleached Alfalfa Hay. — This hay was quite free from green
stalks and as it had been found palatable to the cows it was of special
interest as an experimental roughage for non-pigmented feeding
studies. A selected sample of the hay was ground up fine, and
extracted with ten per cent alcoholic petroleum ether. The light
grqen colored extract showed no indication of either carotin or
xanthophyll when analyzed by means of the chromotogramm, and no
yellow pigment was extracted from its alcoholic solution by petro-
leum ether or by carbon bisulphide.

Yellow Corn. — The unsaponifiable yellow pigment of yellow corn
is in all probability in the oil. It was found to be composed of two
constituents, the largest part of which is a xanthophyll-like pigment,
showing absorption bands in alcoholic and carbon bisulphide solu-
tions lying close to the normal xanthophyll bands. In CS2 the bands
measured: I, 229-251; II, 274-291; III, not measured. It was not
adsorbed from either petroleum ether or carbon bisulphide by CaCO3
but passed through as a yellow or orange zone. Its carbon bisulphide
solutions were orange colored. Petroleum ether readily extracted
the pigment from its concentrated eighty per cent alcoholic solu-
tion, but it could be completely re-extracted from its petroleum ether
solution by fresh eighty per cent alcohol. In this respect it differed
from any xanthophyll-like pigment yet investigated. The pigment
was more soluble in carbon bisulphide than in eighty per cent alcohol
and in this respect favored carotin. On warming its alcoholic solu-
tion containing a little concentrated hydrochloric acid, the color of
the solution changed to a light greenish blue, with the fading of
the absorption bands. The minor constituent of the corn pigment
had the spectroscopic, solubility and adsorption properties of carotin.

The Carrot. — It was planned to conduct some feeding experiments
with carrots, and a special study of its relative proportion of carotin and
xanthophyll was accordingly considered advisable. A large well-
colored carrot was scraped, chopped fine, and boiled in water for
one hour. The softened tissue was poured onto cheesecloth and
the water squeezed out of the pulp. The pulp was pressed through
a wire gauze, dried and powdered. The meal was extracted with
carbon bisulphide and a portion of the blood red solution analyzed
by means of a chromotogramm. By far the greatest part of the pig-
ment passed through unadsorbed as a rose colored zone, leaving a
small amount of adsorbed pigment in the column which was not
differentiated into zones but which was readily washed out with alco-

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 363

holic petroleum ether. The proportion of xanthophyll to the entire pig-
ment in this test was very small, not over three or four per cent.

The rose colored carotin solution showed three beautiful ab-
sorption bands, all of equal intensity. The measurements of the
bands were completely in accord with the standards given in Table 6.
The remainder of the carbon bisulphide extract of the dried car-
rot meal was evaporated into alcohol. The alcohol was diluted with
water to an eighty per cent solution and extracted with petroleum
ether (b. p. 3O°-5o° C) until no more color was extracted. The alco-
holic solution of the xanthophylls which remained was combined with
the solution of the xanthophylls washed out of the chromotogramm
with alcoholic petroleum ether. The combined solutions were evapor-
ated to dryness and the residue dissolved in a small amount of carbon

bisulphide. This solution showed three
absorption bands and end absorption. The
measurements were the same as given in
Table No. 6.

This carbon bisulphide solution was
orange colored when concentrated and
gave the chromotogramm shown in fig-
ure IV. There was a small amount of car-
otin pigment present which being unad-
sorbed was readily washed out and caught
at the lower end of the tube. Zone VIII
zone ii: Orange. was aiso on\y siightly adsorbed and was
colorless. washed out by a stream of carbon bisul-

- zone ni: orange, phide. When this zone was washed out
the suction was stopped. The CaCO3
tinted zones were then removed separately
from the column with the aid of a knife;
52 was drawn off in each case at a

Colorless.

Zone I: Yellow.

Colorless.

colorless.

Colorless.

Zone V: ,* ^o

Yellow-Orange, the Cb,

zonevi* Green ^ow temPerature ? an^ the pigment washed

Colorless.

/ Zone VII:
\ Red-Orange.

Colorless.

7

f Zone VIII:
\ Faint Red
( Orange.

Rose.

Cotton Plug.

FIGURE IV.

out with ninety-five per cent alcohol.

Zone II was found to contain the
most pigment, judged from the color of
its alcoholic solution, with the lower zones
about the same with the exception of Zone
V which had very little pigment. Zone
I contained considerably more pigment
than was indicated by the color of its
zone in the chromotogramm. The spec-
troscopic properties of all the xanthophyll
constituents were studied with respect to
the normal xanthophyll bands. The results

364 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

are shown in Table 10. The observations were not made by observing
all the pigments in the same volume but the volumes were adjusted
in each case to give the best possible bands.

It will be noticed that with the exception of Zone I, all the xan-
thophylls showed the three normal xanthophyll bands to some degree of
intensity. Similarly hydrochloric acid had no effect on any of these so-
lutions or on the bands. Nitric acid had the same effect on the pigments
of Zones II, VII, and VIII causing the solutions to fade with the disap-
pearance of the bands. The effect of nitric acid on the pigments of
Zones III and IV was somewhat different. In these two cases a fourth
well-developed band appeared before the solution lost its color and
the three normal bands faded. The pigment of Zone I was entirely
different from any of the others. The normal first band of the
xanthophylls was entirely missing and the alcoholic solution turned
a distinct bluish green on the addition of a little concentrated hy-
drochloric acid. This color persisted for 24 hours, long after the
absorption bands had disappeared. This was evidently the xanthophyll
B of Tswett and the xanthophyll B of C. A. Schunck.

TABLE No. 10.

SPECTROSCOPIC PROPERTIES OF THE XANTHOPHYLLS OF THE
CARROT.

Zone

Band I

Band II

Band III

End Absorption

I
II

Missing
Fair

Good
Fair

Good
Fair

Very faint
None

III

Very strong

Very strong

Weak

None

IV
V

Strong
Very faint

Very strong
Good

Faint
Good

None
None

VII
VIII

Very strong

Very strong
Good

Very faint
Very faint

None
Very faint

THE FEEDING EXPERIMENTS.

The foregoing studies indicated that the foods best adapted
for non-pigmented rations were bleached hays and cottonseed hulls
for roughages and white corn and cottonseed meal for grains. To
study the variation in the color of butter fat, the ration of various
cows was changed to one containing the smallest possible amounts
of carotin and xanthophylls, and the butter fat studied colorimetrically
during this time. The procedure for the butter fat was in each case
as follows : The milk of one or two milkings was separated by means
of a centrifugal hand separator and the cream churned by hand in

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 365

quart bottles. The butter thus obtained was rendered at a tempera-
ture of 50° to 60° C. and the rendered fat filtered. The pure filtered
fat was analyzed colorimetrically by means of the Lovibond tintometer
and its standard color glasses. The color of the fat was always com-
pared in one-inch layer. The Lovibond tintometer is shown in Fig-
ure V.

FIGURE V.

The solution (in the present case melted butter fat) whose color
is to be measured is placed in a cell with glass ends (one inch apart
in all this work), and the color matched by standard color glasses
of various units of yellow, red or blue, and the color of the solution
read by adding together the various glasses of color used to match
the unknown color. Melted butter fat having an orange tint requires
only yellow and red to match its color. All readings are made with
the instrument pointing towards the daylight (not sunlight). The
instrument is quite sensitive towards the yellow glasses below 25 units
of yellow but the sensitiveness decreases considerably above 40 units
of yellow. In other words, it is possible to match the exact color of
an "unknown" much more closely when its color is below 30 to 35
units of yellow than when its color is above this value. In a great
many cases, and this nearly always applies to butter fat, it is possible
to match the tint of the fat but the color of the fat is more brilliant
than that of the combined standard glasses. In this case an exact
match can be obtained by "damping down" the butter fat color by
inserting in front of it equal units of the three colors, yellow, red
and blue, and recording this as "light."

Before reporting the data dealing with the variation in the color
of the butter fat it may be possible to convey some idea of what the
various colors mean when applied to butter fat by stating that ren-
dered "June" butter in the one-inch cell will give a color of from 80
to 60 units of yellow. Color readings between 45 and 25 units of

366 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. IO

yellow would accordingly indicate a fairly well colored to light col-
ored butter, between 20 and 8 units of yellow would be called light
to very light colored butter, while below these limits ranging down
to i or 2 units of yellow would be called white to "dead" white,
especially if the fat was still in the form of butter.

Experiment i.

The ration of Cow No. 57, a pure bred Jersey, was changed
from a ration rich in carotin and xanthophylls to a ration containing
a very small amount of these pigments. The ration rich in carotin
and xanthophylls consisted of alfalfa hay and yellow corn. The ration
poor in these pigments was composed of bleached clover hay and
white corn. The results of the experiment are shown in Table n.

The change from a ration rich in carotin and xanthophylls to
one poor in these pigments caused the color of the butter fat to drop
from 43 units of yellow to 8.5 units of yellow, from a well colored
to a very light colored fat. This change of color was very gradual
and required 29 days. It should be stated, however, that the cow
did not relish her non-pigmented ration. She lost weight regularly,
and her milk production fell off a great deal. It was apparent that
the animal was drawing heavily during this entire period from a
storage of pigment in her body. It will be shown in the subsequent
papers of this series that in this experiment the blood and also the
body fat were supplying the pigments for the milk fat.

It may be stated that a slow lowering of the color of the milk
fat, such as took place in this experiment, would be normal for all
Jersey cows whose ration is changed to an unpalatable, non-pigmented
one like that used in this experiment. The explanation for this is
found in the high color of the body fat of this breed of cows. We
therefore have here a clear explanation of why Jersey cows will some-
times apparently give yellow milk fat during the winter months when
their food is almost or entirely lacking in carotin and xanthophylls.
Under these conditions if the body fat is called upon to supplement
the digestion products of the food in the production of milk fat at
the same time the blood serum storage of pigments is being drawn
upon, it is clear that the reduction in color of the milk fat will be very
gradual, as in the case of Cow No. 57, and a complete elimination of
color may require a long period of time.

Continuing the discussion of the experiment it is seen that when
the color of the milk fat had dropped to 8.5 units of yellow the white
corn in the ration was replaced by mixed corn, white and yellow

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 367

(mostly yellow). Later this was replaced by yellow corn entirely.
The roughage remained the same, i. e. bleached clover hay. The
table shows that yellow corn had no effect whatever upon the color
of the milk fat. There seemed to be a very slight effect at first but
the color of the fat soon dropped back to 8.0 units of yellow.

TABLE No. 10. THE EFFECT OF A CAROTIN AND XANTHOPHYLL-FREE RATION
ON THE COLOR OF MILK FAT. JERSEY Cow No. 57.

Pounds

Color of butter fat.

Date

hay

Pounds grain fed

per
day

per day

Yellow

Red

Light

1912

Corn

Mixture

Mar. 1

15 (a)

8

43.0

2.0

0.2

2

15(b)

8

3

15

4 white

4

4

15

4 white

4

5

15

4 white

4

6

15

8 white

7

15

8 white

33.0

2.0

0.2

8

15

8 white

9

15

8 white

29.0

1.7

0.2

10

15

8 white

11

15

8 white

33.0

1.8

0.2

12

15

8 white

26.0

1.6

0.2

13

15

8 white

26.0

1.7

0.2

14

15

8 white

22.0

2.0

0.2

15

15

8 white

22.0

2.0

0.2

16

15

8 white

21.0

1.8

0.2

17

15

8 white

18

15

8 white

18.0

1.7

0.2

19

15

8 white

19.0

1.6

0.2

20

15

8 white

18.0

1.6

0.2

21

15

8 white

22

15

8 white

12.0

1.6

0.2

23

15

8 white

24

7

10 white

25

7

10 white

11.0

1.6

0.2

26

7

10 white

11.0

1.6

0.2

27

7

10 white

10.0

1.5

0.1

28

7

10 white

10.0

1.5

0.1

29

7

10 white

8.5

1.8

0.2

30

7

10 white

9.0

1.7

0.2

31

7

10 mixed

10.0

1.8

0.2

(a) Alfalfa.

(b) Clover— Mar. 2 to Mar. 31.

The 7 Ibs. of clover hay was now replaced by 10 Ibs. of alfalfa
hay rich in carotin and xanthophylls. The effect on the color of the

368 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

TABLE 11 (CONTINUED). EFFECT OF FEEDING YELLOW CORN TO A Cow GIV-
ING Low COLORED MILK FAT. JERSEY Cow No. 57.

Date

Pounds
hay
per
day

Pounds
grain fed

day

Color of butter fat

Yellow

Red

Light

1912

Corn

Apr. 1

7 (a)

10 mixed

8.5

1.7

0.2

2

7

10 mixed

8.0

2.1

0.2

3

7

10 mixed

8.0

.9

0.2

4

7

10 mixed

12.0

.5

0.2

5

7

10 mixed

12.0

.7

0.2

6

7

10 mixed

11.0

.8

0.2

7

7

10 yellow

8

7

10 yellow

11.0

.8

0.2

9

7

10 yellow

8.0

.7

0.2

10

7

10 yellow

9.0

.7

0.2

11

10(b)

10 yellow

10.0

.5

0.2

12

10

10 yellow

12.0

.6

0.2

13

10

10 yellow

15.0

.6

0.2

14

10

10 yellow

20.0

.6

0.2

15

10

10 yellow

33.0

.7

0.2

16

10

10 yellow

36.0

.6

0.2

17

10

10 yellow

38.0

.6

0.2

18

10

10 yellow

43.0

.6

0.2

19

10

10 yellow

45.0

2.0

0.5

21

10

10 yellow

43.0

1.8

0.5

22

10

10 yellow

43.0

1.7

0.5

23

10(c)

10 yellow

46.0

1.7

0.5

24

10

10 yellow

40.0

1.8

0.5

25

10

10 yellow

43.0

1.8

0.5

27

10

10 yellow

52.0

2.0

0.5

30

10

10 yellow

57.0

2.1

0.5

May 2

10

10 yellow

64.0

2.1

0.5

4

10

10 yellow

64.0

2.2

0.5

5-10(d)

Pasture only

80.0

2.5

0.5

a) Clover— Apr. 1-10.

b) Alfalfa— Apr. 11-22.

(c) Alfalfa and pasture — Apr. 23-May 10.

(d) Sample taken May 10.

milk fat was immediate. At the end of seven days the color had
increased to 43 units of yellow, the maximum supplied by this rough-
age. Later the ration was supplemented by some fresh pasture grass.
The color of the milk fat increased to 64 units of yellow. Still later
the cow was turned out to pasture alone. The color then reached
the maximum we have observed, i. e. 80 units of yellow.

Experiment No. 2.

This experiment was conducted with Cow No. 301, a pure bred
Ayrshire. A short time previous to the experiment here reported this

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 369

cow was in very poor flesh resulting from a feeding experiment in
which she was heavily underfed. Her ration during her underfeeding
was composed of alfalfa hay rich in carotin and xanthophylls and a
grain mixture of corn, bran and linseed meal. At the end of the
underfeeding experiment the ration was changed to bleached alfalfa
hay, the pigmentation of which was reported above, and white corn
and cottonseed meal. Enough of this ration was given to bring the
cow back to normal feeding conditions. At the end of this time the
experiment here reported was begun. The results are given in
Table 12.

At the end of ten days the color of the milk fat had dropped to
9 units of yellow. When it was apparent after a week's further trial
that the color had reached at least a temporary minimum value for
this ration, the grain was changed to yellow corn entirely. This was
finally increased to 12 Ibs. per day. The result was very clearly in
accord with that of Experiment No. I, showing that yellow corn is
not a source of pigment for the milk fat of dairy cows.

TABLE No. 12. EFFECT OF NON-PIGMENTED RATION AND A RATION CONTAIN-
ING YELLOW CORN UPON THE COLOR OF MILK FAT. AYR-
SHIRE Cow No. 301.

Date of

Pounds
alfalfa

Pounds

Pounds
cotton-

Color of 1

cutter fat

sample

hay

corn

seed
meal

Yellow

Red

1912

October 3

16

6(a)

2

27.0

.7

13

16

6

2

9.0

.7

17

16

6

2

6.0

.5

21

16

6

2

9.0

.2

24

16

8(b)

7.5

.2

25

16

8

9.0

.5

26

16

8

10.0

1.5

27

16

8

9.0

1.5

28

16

8

9.0

1.5

29

16

8

10.0

1.5

30

16

8

8.5

1.2

31

16

8

10.0

1.2

November 1

16

12

9.5

1.2

2

16

12

10.0

1.2

3

16

12

8.5

1.2

4

16

12

8.0

1.2

(a) Oct. 3-21, white corn.

(b) Oct. 24- Nov. 4, yellow corn.

370 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. IO

Experiment No. 3.

This feeding experiment was conducted with the same cow as
the preceding experiment and immediately followed that experiment.

It seemed very probable that the reason the color of the milk
fat in Experiments i and 2 could not be lowered more than the uni-
form figure found in both experiments, i. e., about 8 units of yellow,
was due to the fact that the ration was supplying a small amount of
pigment and also to the fact that the normal storage, that in the
blood serum, had not been exhausted. It seemed reasonable to sup-
pose, therefore, that if the first factor was eliminated at the outset,
the second would necessarily also be eliminated if the experiment was
continued for a sufficient length of time. The experiment here re-
ported was for the purpose of testing the validity of this supposition,
and also for the purpose of ascertaining to how low a point the color
of the milk fat could be reduced. The ration chosen was one which
would supply the least amount of carotin and xanthophylls. It was
composed of cottonseed meal and cottonseed hulls. The results of
the experiment are given in Table 13.

The supposition stated above was fully borne out by the long
continued feeding of a practically non-pigmented ration. At the end
of 52 days' feeding, the cow was producing absolutely colorless butter.
It was only when the rendered butter was viewed in the tintometer
that a very slight amount of color could be detected. This very
slight amount of color was due to the fact that the normal storage
of pigment in the body, that in the blood serum, had not been com-
pletely exhausted, as will be shown in a subsequent paper of this
series. It is probable also that the body was being drawn upon for
some of its pigments, for the animal suffered somewhat from under-
feeding on this ration. It is also possible that a very small amount
of carotin was being supplied by the oil in the cottonseed meal and
hulls. For practical purposes such a supply of pigment would of
course be considered absolutely negative.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 37!

TABLE No. 13. EFFECT OF A LONG-CONTINUED FEEDING OF A NON-PIGMENTED

RATION UPON THE COLOR OF MILK FAT. AYRSHIRE

Cow No. 301.

Date of
sample

Pounds
alfalfa
hay

Pounds
cotton-
seed
hulls

Pounds
cotton-
seed
meal

Pounds
Corn

Color of butter fat

Yellow

Red

1912-13

Nov. 5

6.5

12

9.0

1.2

6

4.5

0.5

0.5

10

8.0

1.2

7

8.0

1.0

1.0

8

9.0

1.1

8

8.0

2.0

2.0

7

8.0

1.0

9

8.0

2.0

2.0

6

7.0

1.0

10

7.0

3.0

3.0

6

6.5

1.0

11

6.0

3.0

3.0

5

6.0

1.0

12

6.0

4.0

4.0

4

5.5

0.9

13

6.0

4.0

4.0

4

5.0

0.8

14

3.0

5.0

5.0

2

5.0

0.8

15

2.0

6.0

6.0

5.0

0.9

16

10.0

6.0

4.0

0.7

17

12.0

6.0

4.5

0.8

18

12.0

7.0

4.0

0.7

19

12.0

7.0

3.5

0.7

20

12.0

8.0

3.5

0.7

21

13.0

8.0

3.5

0.7

22

14.0

8.0

3.5

0.7

23

14.0

8.0

3.2

0.7

24

16.0

8.0

3.1

0.7

25

16.0

8.0

3.0

0.7

26

16.0

8.0

2.6

0.5

27

16.0

8.0

3.2

0.7

28

16.0

8.0

2.5

0.7

29

* ,

16.0

8.0

3.0

0.7

Jan. 2

16.0

8.0

1.4

0.5

7

16.0

8.0

1.3

0.4

It will be of interest to state in connection with this experiment
that it furnished the three "light colored" fats for the studies of the
proportion of carotin and xanthophyll which were reported above.

Experiment 4.

The color of the butter 'fat of Cow No. 301 was now so low that
the conditions were considered ideal for one or two additional im-
portant investigations, first a confirmation of the apparently negative
effect of feeding yellow corn which was obtained in Experiments I
and 2, and second a study of carrot feeding. In the latter study the
results would be especially interesting in view of the fact that the
pigment fed would be almost pure carotin. The hay used in this
experiment was a very light colored timothy hay which was not quite
so free from carotin and xanthophylls as the bleached alfalfa of Expe-
riment 3, but which apparently had no effect on the color of the fat

372 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

as the data will show. The results of the two studies are given in
Table 14.

In regard to yellow corn feeding, it is seen that replacing 2 Ibs.
cottonseed meal and 4 Ibs. white corn by 6 Ibs. yellow corn had no
influence upon the color of the milk fat. If yellow corn has any value
as an aid in the production of yellow milk fat, it would certainly
have been evident in this experiment where the milk fat was prac-
tically colorless before feeding the yellow corn. This experiment
therefore gives conclusive proof of the negative value of yellow corn
in the pigmentation of milk fat. The reason for this will be discussed
more fully in the general discussion of the experiments.

TABLE No. 14. EFFECT OF A RATION CONTAINING YELLOW CORN AND OF A
RATION CONTAINING CARROTS UPON THE COLOR OF MILK FAT.
AYRSHIRE Cow No. 301.

Date of

Pounds
rough-

Pounds
carrots

Pounds
cotton-

Pounds
corn

Color of 1

Dutter fat

sample

age(a)

seed meal

Yellow

Red

1912

Jan. 3

14

4

4 white

21

12

4

4 white

24

12

4

4 white

1.2

0.4

25

12

4

4 white

28

12

2

6 yellow

29

12

2

6 yellow

1.4

0.5

30

12

2

6 yellow

1.5

0.4

31

12

2

6 yellow

1.7

0.4

Feb. 1

12

2

6 yellow

1.7

0.4

2

12

2

6 yellow

2.0

0.5

3

12

2

6 yellow

1.8

0.5

4

12

2

6 yellow

2.0

0.5

5

12

2

6 yellow

1.8

0.3

6

12

2

6 yellow

2.0

0.5

7

12

6

4

4 white

2.2

0.5

8

12

20

4

4 white

2.2

0.5

9

12

30

4

4 white

2.3

0.5

10

12

40

4

4 white

3.2

0.6

11

12

40

4

4 white

4.5

0.7

12

12

50

4

4 white

5.5

1.0

13

12

50

4

4 white

8.0

1.0

14

12

50

4

4 white

11.5

.0

15

12

30

4

4 white

16.0

.1

16

12

30

4

4 white

15.0

.2

17

12

20

4

4 white

18.0

.3

18

20

19.0

. 1

19

3

15.0

.1

20

5

10.0

.3

21

3

10

1

1 white

8.0

.2

22

3

10

2

2 white

9.0

.3

(a) Roughage consisted of 2 parts bleached timothy hay and 1 part cotton-
seed hulls.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 3/3
TABLE No. 14. (CONTINUED)

Date of

Pounds
rough-

Pounds

Pounds
cotton-

Pounds

Color of 1

cutter fat

sample

age(a)

carrots

seed
meal

corn

Yellow

Red

1913

Feb. 23

12

20

4

4 white

11.0

.3

24

12

20

4

4 white

21.0

.3

25

12

20

4

4 white

28.8

.2

26

12

20

4

4 white

36.0

.8

27

12

20

4

4 white

27.0

.3

28

12

20

4

4 white

24.0

.3

Mar. 1

12

20

4

4 white

24.0

.3

2

12

20

4

4 white

28.0

.4

3

12

20

4

4 white

26.0

.4

4

12

20

4

4 white

26.0

.3

5

12

20

4

4 white

24.0

.4

6

12

50

4

4 white

24.0

.3

7

12

4

4 white

23.0

.3

8

12

4

4 white

19.0

.2

9

12

4

4 white

19.0

.2

10

12

4

4 white

18.0

.6

11

12

4

4 white

17.0

.1

12

12

4

4 white

14.0

.1

13

14

4

4 white

11.0

.1

14

14

4

4 white

12.0

.2

15

14

4

4 white

12.0

1.2

17

14

4

4 white

10.5

1.2

21

14

4

4 white

7.5

1.2

24

12

4

4 white

7.0

1.0

30

12

4

4 white

7.5

1.3

(a) Roughage consisted of 2 parts timothy hay and 1 part cottonseed hulls.

Note: During the period from 2:18 p. m. to 2:23 a. m. the cow was badly
off feed and her entire ration was withdrawn for a few days. She soon re-
covered, however, and was put back on the experimental ration, not feeding so
many carrots, however.

In regard to the effect of feeding carrots, the data brings out
several interesting points. The carrots were added to the ration on
February 7, and the amount was rapidly increased to 50 Ibs. per day.
The cow ate them with a good deal of relish for about a week when
she began to refuse part of them and finally went off feed entirely,
making it necessary to withdraw her entire ration for a day or two.
The effect of this carrot feeding period upon the color of the milk
fat was to increase the color from 1.8 to 19 units of yellow. This in-
crease was not nearly as great, however, as was expected considering
the large amount of carrots that was fed and the length of time they
had been in the ration, i. e. eleven days.

374 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. IO

It is clear from a study of the subsequent data that this abnormal
result was due to some physiological disturbance that finally resulted
in the cow going entirely off feed. As soon as the carrots were re-
moved from the ration, the color of the fat dropped at once to 8 units
of yellow. However, when the animal had recovered from her attack
of indigestion, and the carrots were again added to the ration, the
effect upon the color of the milk fat was perfectly normal. On the
fourth day of feeding 20 Ibs.. of carrots the color had reached a max-
imum of 36 units of yellow. The color later dropped a little with
an increase in fat production which is not shown in the table, but
remained in the neighborhood of 28 units of yellow until the carrots
were removed. The color then began to drop slowly in the normal
way. After dropping to 7 units of yellow on the twenty-third day
after the carrots had been removed, the experiment was stopped.

This experiment furnished the samples of butter fat whose pro-
portion of carotin and xanthophyll were studied and reported in an
earlier part of this paper, namely the fat after carrot feeding.

Experiment 5.

This was a second carrot feeding experiment using another cow,
i. e. Cow No. 221, a pure bred Holstein cow. The experiment was
not as successful as was hoped because of the peculiar appetite of the
cow. She refused to eat more than ten pounds of the carrots per day
except on two days so the experiment was discontinued. The data
are given in Table 15. Notwithstanding the peculiar appetite of the
cow it is interesting to note that the feeding of only 10 Ibs. of carrots
per day for 8 days was sufficient to bring the color of the milk fat
almost back to the starting point, i. e. 26 units of yellow.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 375

TABLE No. 15. EFFECT OF A NON-PIGMENTED RATION AND OF A RATION CON-

TAINING CARROTS UPON THE COLOR OF MILK FAT.

HOLSTEIN Cow No. 221.

Pounds

Color of butter fat

Date of

Pounds

Pounds

cotton-

Pounds

sample

hay(a)

carrots

seed
meal

corn

Yellow

Red

1912
Nov. 29

Normal herd ration containing green
alfalfa hay.

26.0

1.5

Dec. 9

16

4

4

8.0

1.4

12

16

4

4

7.0

1.3

14

16

4

4

6.0

1.1

17

16

10

4

4

6.0

1.3

18

16

10

4

4

7.0

.5

19

16

10

4

4

9.0

.5

20

16

10

4

4

8.0

.5

21

16

10

4

4

8.0

.5

22

16

10

4

4

10.0

.5

23

16

10

4

4

17.0

.3

24

16

30

4

4

17.0

.2

(a. m.) 25

16

20

4

4

14.0

.5

(p. m.) 25

16

0

4

4

20.0

.5

(a. m.) 26

16

0

4

4

18.0

.5

(p. m.) 26

16

0

4

4

17.0

.5

27

16

0

4

4

18.0

.5

28

16

0

4

4

12.0

.5

1913

Jan. 3

16

0

4

4

7.5

.5

22

Herd ration since 1-4-13

14.0

.2

(a) The hay was a mixture of equal parts of light colored timothy and
bleached alfalfa.

Experiment No. 6

This was a feeding experiment the results of which were ex-
pected to corroborate those of Experiment i, and show that the colon
of the milk fat of Jersey cows is as much dependent on the food as
those of other breeds. The variation in the feed and the resulting
color of the milk fat is shown in Table 16. A pure bred Jersey cow,
No. 59, was used for this experiment.

376 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

TABLE No. 16. EFFECT OF A NON-PIGMENTED RATION UPON THE COLOR OF
MILK FAT. JERSEY Cow No. 59.

Date of
feeding

Pounds
grain
mix-
ture(a)

Pounds
corn
silage

Pounds
hay

Date of
sample

Color of butter fat

Yellow

Red

Light

1911

11.6
11.
11.
11.
11
11.5
12.

0
10
10
10

10

Corn
Stover
5

5

14(b)
9(c)
8
4
4
8(d)
8

April 2
April 14
April 23
May 9
May 20
June 14
June 29

46.0
6.0
4.0
3.0
3.0
47.0
26.0

1.8
1.5
1.4
1.5
1.5
1.8
2.0

0.5
0.2
0.2
0.2
0.2
0.5
0.2

March 11
to
April 2

April 3
to
April 4

April 15
to
April 23

April 24
to
May 8

May 10
to
May 20

May 21
to
June 14

June 15
to
June 29

(a) The grain mixture was 5 Ibs. corn and 6 Ibs. cottonseed meal.

(b) Green alfalfa hay from March 11 to April 2.

(c) Bleached timothy hay from April 3 to May 20.

(d) Green alfalfa hay beginning May 21.

The results of this experiment show conclusively that Jersey
cows are as much dependent upon their food for the pigments of the
milk fat as other breeds of cows. In addition the experiment offers
excellent proof of some statements previously made in explanation
of the gradual lowering of the color of the milk fat of Cow No. 57
in Experiment No. i. It was stated there that the yellow body fat
was supplementing the normal storage of pigment on account of the
unpalatableness of the ration. In the present experiment we have an
example of the effect of changing the ration to a non-pigmented one

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 377

without causing the animal to draw upon any storage of pigment
other than the normal one of the blood serum. The result was that
the color of the milk fat dropped from 46 units of yellow to 6 units
of yellow in twelve days, whereas in Experiment I it required thirty
days to bring about a similar change of color. The very low color
of the fat which was reached in Experiment 6 also indicates that
only the normal storage was being drawn upon. Experiment 6 also
shows that corn silage is not a source of pigment for the milk fat.
The chemical changes which take place in this roughage evidently also
largely destroy the carotin and xanthophylls. The chemical studies of
the pigments of corn silage, which were reported above, showed this to
be the case.

RELATION BETWEEN COLOR OF MILK FAT AND BREED OF COW.

The foregoing experiments have shown conclusively that dairy
cows, exclusive of breed, are dependent on the carotin and xanthophylls
in their feed for the pigment of their milk fat, in other words, that
they cannot produce the pigment which is thus secreted. The ques-
tion is at once raised as to wherein lies the so-called breed character-
istic which is so much emphasized by the breeders of Guernsey and
Jersey cattle? It will not be denied that a breed characteristic does
exist in connection with the color of butter fat. We believe, how-
ever, that the data now to be presented will show that this breed
characteristic has been overemphasized.

Since the butter fat is dependent upon the food of the cow for
its color, it was necessary to compare the color of the butter fat
of the different breeds under comparative feeding conditions, in order
to obtain a correct estimate of the breed relation.

It would naturally be expected that the most favorable condition
for studying the accuracy of the views held by the cattle breeders and
others that some breeds of cows, such as the Jersey and Guernsey,
are color producers while other breeds, such as the Holstein, are not
color producers would be a comparison of the color of the combined
fat of several cows of each breed. Table 17, which follows, gives
such a comparison taken from animals in one herd. The milk and
fat production of the various cows varied widely. The comparison
was made during the winter months, the only source of pigment being
a more or less variable quantity of green alfalfa hay in the ration,
which was, however, the same for all the animals.

378 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

TABLE No. 17. RELATION OF BREED TO COLOR OF MILK FAT.

Color

of butter fc

it

Breed

Yellow

Red

Light

Jersey

50.0

2.1

0.2

Ayrshire

38.0

1.7

0.2

Shorthorn

34 0

1 6

0.2

Holstein

31 0

1.7

0.2

The most striking fact brought out by this table is that the ques-
tion of the color of the fat produced by the four breeds represented
is not one of presence or absence of color, but rather a question of
relative color. The fat from the Jersey cows was unquestionably the
highest colored of the four samples but the fat from the Holsteins
also had a very good color, although the butter would probably have
been scored as "slightly low in color."

This point of relative color production is also clearly shown
when comparing the fat produced by individual members of the breeds.
Table 18 shows the color of the fat from two Jerseys and one Hol-
stein cow under feeding conditions most favorable for the maximum
color. These animals were producing about the same amount of but-
ter fat, and the roughage of their ration consisted for the most part
of freshly-cut soybeans, very rich in carotin and xanthophylls.

TABLE No. 18.

SHOWING RELATIVE COLOR PRODUCTION BY DIFFERENT IN-
DIVIDUALS.

Color

Cow No.

Breed

Feed

Yellow

Red

Light

59

Jersey

Fresh green soybeans
and grain

54.0

2.5

1.0

34
208

Jersey
Holstein

Ditto
Ditto

60.0
29.0

2.5
1.8

1.0
0.5

When comparing the color of the fat produced by individual
members of the Jersey and Holstein breeds under feeding conditions
favorable for only a moderate amount of color in the fat, the relative
color production of the breeds very nearly approaches unity. This is

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 379

especially true when the fat production and the actual proportion of
the ration furnishing the pigments are taken into account. Such a
comparison is shown in Table 19.

TABLE No. 19. RELATIVE COLOR PRODUCTION UNDER SPECIAL FEEDING

CONDITIONS.

Cow

No.

Breed

Pounds
green
alfalfa
hay

Green feed
(Dry
matter
basis) (a)

Pounds
milk
fat
per day

Color

Yellow

Red

34
41
11
13
208
211
219

^ ersey
'm ersey
^ ersey
. ersey
iolstein
Holstein
Holstein

10.0
12.0
8.0
8.0
15.0
12.0
12.0

42%
41%
40%
45.5%
35%
28%
28%

1.34
1.78
, 1.51
0.58
1.23
2.13
1.50

29.0
29.0
24.0
33.0
19.0
17.0
19.0

1.6
1.6
1.3
1.6
1.6
1.5
1.7

(a) Per cent of moisture free green feed in total ration.

The most interesting feature of the above table is to note that
on the basis of the per cent of green dry matter in the ration, the
Jersey cows produced the highest colored fat because they received
the highest per cent of green dry matter. By taking into considera-
tion also the difference in fat production, an interesting calculation
can be made with the figures of average color, fat production and per
cent of green dry matter in the ration, which will cause the relative
color production of the two breeds to approach almost unity.

Jerseys

Holsteins

Average per cent of green feed in ration

42 1%

30%

Average fat production

1 301bs.

1. 6 Ibs

Average color of fat (Units of yellow)

29 0

18 0

If it be assumed for the moment that there is no breed charac-
teristic we can say that 30% green feed in the ration of the Holsteins
produces 18 units of yellow in the fat for the same reason that 42%
green feed in the ration of the Jerseys produces 29 units of yellow
in their fat. If this is true then the following proportion would be a
true one, i. e. :

42:29 : : 30:18

The product of the means is not quite equal to the product of the
extremes but gives the result,

870 = 756

380 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

If the amount of fat produced is taken into consideration and each side
of this equation is multiplied by the corresponding amount of fat we
have the result,

870 X 1.3 = 756 X 1.6
or 1,131 = 1,210

or i = 1.07, which is very near unity.

The relation between the breed of the cow and the color of the
fat under two different conditions of feeding is well illustrated by
Tables 18 and 19. The color of the fat produced by cows No.. 34
and 208 is given under both heavy and moderate pigment feeding.
The data in Table 19 were obtained a number of weeks after that in
Table 18. The figures show that the change from heavy to moderate
pigment feeding caused the color of the milk fat of the Jersey cow
to drop 50% while a similar change in the feed of the Hblstein cow
caused a color drop of only 35%.

The relation of the breed to the change in color produced by a
change in the ration is also well illustrated in the following table
No. 20.

TABLE No. 20. COLOR PRODUCTION IN DIFFERENT BREEDS AS AFFECTED BY

CHANGES IN RATION.

Cow No.

Breed

Date

Grams
fat
produced

Color of fat

Yellow

Red

213
213

Holstein
Holstein

3-11-13
4-10-13

122
135

8.5
54.0

1.4
1.8

220
220

Holstein
Holstein

3-11-13
4-10-13

167
208

3.0
22.0

0.7
1.2

303
303

Ayrshire
Ayrshire

3-11-13
4-10-13

213
263

2.5
16.0

0.6
1.1

16
16

Jersey
Jersey

3-11-13
4-10-13

304
363

11.0
64.0

1.7
2.0

57
57

Jersey
Jersey

3-11-13
4-10-13

240
263

5.2
54.0

1.2
1.7

64
64

Jersey
Jersey

3-11-13
4-10-13

281
358

4.7
47.0

1.5
1.6

The first sample for each cow in the above table represents the
result of a long continued feeding of a ration almost entirely lacking
in carotin and xanthophylls. The second sample represents one month's
feeding of a ration rich in these pigments, the ration including a

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 381

plentiful supply of green alfalfa hay and some fresh green grass.
The data bring out two points worthy of emphasis. One of these
is that it is possible to find pure bred Holstein cows entirely lacking
in the so-called breed characteristic of color production. Holstein
Cow No. 213 for instance produced as much color in her fat in both
periods as any of the Jerseys. The low color of the milk fat of cows
No. 220 and No. 303 in the second period can only be explained at
present on the ground that it was due to some inherent characteristic
of the animals, which for lack of a better term may be called breed
characteristic. The other point brought out by the data is merely in
emphasis of the results obtained in the feeding experiments showing
that all breeds of cows suffer alike in regard to the color of their
milk fat when the pigments carotin and xanthophylls are withdrawn
from their food. At this time there is no breed characteristic.

Not only does the breed characteristic disappear when the source
of the pigment is withdrawn, but it also disappears for all cows at the
time of maximum color in the fat, i. e. immediately after parturition.
Data was given in Table 8 showing the high color of the colostrum
milk fat for cows of three breeds. There was certainly no breed
characteristic evident there.

There is one other breed difference yet to be considered, which
has led, probably more than anything else, to the belief that Jersey
and Guernsey cows can produce yellow butter fat at any time, regard-
less of feed. This difference has to do primarily with the storage of
pigment in the body, and its discussion belongs properly to the two
subsequent papers of this series. A brief statement here in regard
to it however will prevent a doubt arising in the minds of some read-
ers, whose practical experience is apparently contrary to the experi-
mental evidence here offered.

Stating the question in hypothetical form, it may be said that
if a Jersey (or Guernsey) and a Holstein cow, both giving well-colored
milk fat, the possibility of which cannot be denied in the light of the
evidence which has been offered on this point, are put upon dry feed
containing little or no carotin and xanthophylls, the color of the milk
fat will drop much faster with the Holstein cow than with the Jer-
sey (or Guernsey) cow, unless great care is taken to provide a ration
as nourishing and palatable as the previous pigmented one. The re-
sult will be that the Jersey (or Guernsey) cow will appear to be pro-
ducing colored milk fat on a non-pigmented ration. The explanation
for this has already been given in connection with feeding Experi-
ments Nos. i and 6, and lies in the fact that the body fat of Jersey
and Guernsey cows furnishes a supplementary storage of pigments

382 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

not usually found in other breeds. It will be shown in a subsequent
paper that if the body fat which furnishes the supplementary pig-
ments in the case of the Jersey (or Guernsey) cow is laid on with
a non-pigmented ration, it will be as colorless as is often seen in the
case of the body fat of Holstein cows. If this were true in the
hypothetical case described above, there would have been no breed
characteristic evident, for it will also be shown in a subsequent paper
that the normal storage of pigment, i. e. that of the blood serum, is
practically the same for all breeds of cows.

Sufficient evidence has been presented to permit a repetition of
a previous statement, namely that the relation of the breed to the
color of the milk fat has received more emphasis than a study of the
question will warrant. The color of the milk fat is primarily depend-
ent upon the character of the food and the fact that some breeds of
cows give less color in their milk fat than other breeds will probably
be found to be only an apparent one when all the factors which come
into play are known.

CAROTIN, THE PRINCIPAL \ELLOW PIGMENT OF MILK FAT 383

DISCUSSION OF RESULTS.

It was the primary object of this investigation to classify the
natural yellow pigment of milk fat both as an individual and also
in relation to the two well-known yellow classes of plant pigments, car-
tin and xanthophylls, whose general properties have often been observed
to be closely related to various yellow pigments of so-called animal
origin.

Basing the study upon a number of well-defined, characteristic
physical and chemical properties of carotin and xanthophylls, it has
been shown that the principal pigment of milk fat is a member
of the fast widening group of hydrocarbon pigments, the carotin
of green plants. In addition it has been shown that the milk fat
carotin nearly always has associated with it one or more minor
constituents whose general properties and characteristics are identical
with the xanthophyll group of pigments. Two and possibly three
xanthophyll constituents were found in one sample of high colored
butter fat.

In addition to the establishment of a chemical relation between
carotin and xanthophylls and the yellow lipochrome of milk fat, it
has been possible to demonstrate a much more significant fact, namely
that this lipochrome whose origin has hitherto been considered to be
in the animal body is in reality merely the carotin and xanthophylls
of the food, which are absorbed by the body and subsequently secreted
in the milk fat. Numerous feeding experiments show that when the
food is deficient in carotin and xanthophylls for a period of time, the
milk fat slowly decreases in color and eventually approaches a color-
less condition. The experiments also show that when foods rich in
carotin and xanthophylls are given to a cow whose milk fat is deficient
in lipochrome, the color of the milk fat at once increases in propor-
tion to the amount of pigments fed. This is true regardless of whether
the carotins and xanthophylls are associated with chlorophyll as in
green feeds, or whether chlorophyll is completely absent and xan-
thophylls almost so, as in carrots.

The experiments show in addition that small amounts of carotin,
such as are present in the oil of cottonseed meal have apparently no
effect on the color of the butter fat. It is not clear, however, whether
this is due to the smallness of the amount of carotin or to the state
in which it exists in the food, i. e., dissolved in oil. There is some
evidence on both sides. Mendel and Daniels * have recently found

1. Jour. Biol. Chem. 13, No. 1, p. 72 (1912).

384 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. IO

that when fifteen grams per day of Sudan III dissolved in oil was
fed to a cow for three successive days, there was no indication of
the dye in the milk. On the other hand, in the feeding experiments
with Cow No. 301 where the bleached alfalfa hay was changed to
timothy hay containing a small amount of carotin, there was also
apparently no effect on the color of the milk fat.

It is especially noteworthy that all of the above feeding experi-
ments which involved yellow corn are united in pointing to its inabil-
ity to impart any color to the butter fat. This result is not so sur-
prising, however, when viewed in the light of the character of the
pigment of yellow corn as shown in the chemical studies. It was
found there that the pigment is largely a xanthophyll. It may be
stated that the butter fat of Cow No. 301 during the last yellow corn
experiment failed to show the presence of the corn xanthophyll, when
subjected to careful examination.

The feeding of carotin in the form of carrots to a cow giving as
low colored milk fat as Cow No. 301 gave an excellent opportunity to
study the proportion of carotin and xanthophyll in the resulting well-
colored butter fat. This investigation was reported in connection
with the study of the proportion of carotin and xanthophyll in butter
fat under varying conditions of production. It was found that the
xanthophylls were practically absent from the fat. The conclusion
is that the xanthophylls must be present in the food in large excess,
as in grass, before they will appear in the butter fat.

It was mentioned above, and the fact is worthy of special notice,
that when the carotin and xanthophylls are withdrawn from the food
the falling off in color of the butter fat is sometimes very slow. In the
case of Jersey cow No. 57, it required twenty-seven days for the butter
fat to drop in color from 43 to 8.5 units of yellow. Normally, it should
require much less time for the color to drop this amount. For instance
it required only 12 days for a similar drop to be brought about in the
color of the milk fat of Jersey cow No. 59. In explanation of this
difference it may be stated that upon a normal plane of nutrition, the
blood serum furnishes the pigment for the milk fat. When the plane
of nutrition is below normal in a lactating cow the body fat is drawn
upon to aid in the production of milk fat and also for other purposes.
If the body fat thus utilized has a high yellow color, as is usually
the case in Jersey and Guernsey cows, the normal storage of pigment
for the milk fat will be continually, at least partially, replenished.
The reduction in color of the milk fat will then be much slower than
normal. It was stated above and will bear repetition, that we have
here an explanation of why Jersey cows apparently often produce high
colored milk fat on a low pigmented ration, as during the winter months.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 385

The results of these experiments are of considerable practical
importance. It is readily seen, for instance, that the peculiar popular
conception of the enhanced value of butter on account of a high
yellow color is absolutely without foundation. It is furthermore seen
that the prevailing opinion among some cattle breeders that Guernsey
and Jersey cows are able to synthetically produce a high colored but-
ter fat under all conditions is also unfounded. It has been shown
that all breeds of cows will produce well-colored butter fat under
proper feeding conditions. The reverse has been shown to be espe-
cially true, namely that a cow, regardless of breed, cannot produce
high colored butter fat under normal conditions, unless the food con-
tains the pigments which are utilized for that purpose.

With our present knowledge, however, we would not be justified
in saying that there is no breed characteristic in connection with the
color of butter fat. Under apparently equal conditions Jersey and
Guernsey cows usually give higher colored milk fat than Holstein or
Ayrshire cows. We have been able to offer some evidence, however,
showing that during a moderate pigmentation of the milk fat, this
difference largely disappears when the amount of fat produced and
the proportion of the ration which is the source of the pigment are
taken into consideration. Further experiments would be required to
ascertain whether this is true under all conditions of milk fat pig-
mentation or only true for moderate pigmentation. The data already
at hand at least entirely justify the statement that the so-called breed
characteristic has been given more emphasis than is warranted by an
actual study of the facts.

The results of our experiments are furthermore of considerable
physiological significance. A direct source for the lipochromes of the
cow has been established, which opens the question of a similar
source for all animal lipochromes. The lipochrome of milk fat has
been increased or decreased with great ease by merely varying the
food of the cow. Such a result throws great doubt upon any physio-
logical significance which the lipochromes have been supposed to exert
in the animals in which they have been found. Apparently it is merely
a question of the inability of the animal body to throw off the excess
of carotin and xanthophylls contained in its food. Experiments will
reported in a later paper of this series showing that the blood serum,
of the cow at least, very rapidly takes up the carotin (especially) of
the food and carries it through the body in combination with an albu-
min of the serum.

386 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. IO

SUMMARY.

1. The fat of cows' milk owes its natural yellow color to the
pigments carotin and xanthophylls, principally carotin, the well-known,
wide-spread, yellow vegetable pigments found accompanying chloro-
phyll in all green plants.

2. The carotin and xanthophylls of milk fat are not synthesized
in the cow's body, but are merely taken up from the food and subse-
quently secreted in the milk fat.

3. When food practically free from carotin and xanthophylls,
such as the cow usually receives during the winter months, is given
to a milk-giving cow, the immediate supply of these pigments in the
organism is greatly depleted and may be entirely used up, on account
of the constant drain upon the supply by the milk glands. The but-
ter fat accordingly approaches a colorless condition in proportion to
the supply of carotin and xanthophylls in the system, the length of
time these pigments are kept out of the food, and also, very probably,
in proportion to the amount of milk fat being produced.

4. If food rich in carotin and xanthophylls is given to a milk-
giving cow whose milk fat has become practically colorless by reason
of the above conditions, the organism will at once recover its lost
pigments and the milk fat will increase in color in proportion to the
amount of carotin and xanthophylls, especially carotin, in the food.
Fresh green grass probably being the richest in carotin of all natural
dairy cattle feeds, accordingly produces the highest colored butter.

5. There is some difference among different breeds of dairy
cows in respect to the maximum color of the milk fat under equally
favorable conditions for the production of a high color. Each breed
of cows, however, will undergo the same variation in color of the
milk fat which follows a withdrawal or addition of carotin and xan-
thophylls, especially carotin, to the food. Under some conditions,
also, the apparent breed characteristic largely disappears. The popu-
lar opinion in regard to the breed characteristic has been overempha-
sized, and statements in regard to it should in the future be qualified
with a statement of the conditions of feed, etc.

6. Under normal conditions cows of all breeds produce very high
colored milk fat for a short time after parturition. The pigments
of the fat at this time are identical with the normal pigments of the
fat. Their increase at this time is probably due to the physiological
conditions surrounding the secretion of the milk of the freshening
animal.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 387

BIBLIOGRAPHY.

1. Escher: Zeit. f. Physiol. Chem. 83, p. 198 (1913).

2. Kirsten: Zeit. Nahr. Genussm. 5, p. 833 (1902).

3. Kraus: Flora, p. 155 (1875).

4. Lewkowitsch: "Oils. Fats and Waxes," vol. i, p. 371 (1909 Edi-
tion) .

5. Mendel and Daniels: Jour. Biol. Chem. 13, No. i, p. 72 (1912).

6. Newbigin: D. Noel Paton — "Reports of Investigations on Life
History of Salmon" (1898), Article XV, p. 159.

7. Schunck: Proc. Roy. Soc. 72 (1903).

8. Sorby: Proc. Roy. Soc. 21, p. 456 (1875).

9. Thudichum: Proc. Roy. Soc. 17, p. 253 (1869).

10. Tswett: Ber. d. Deut. Botan. Gessel. 24, pp. 316 and 384 (1906) ;

29, p. 630 (1911).
n. Willstatter and Escher: Zeit. f. Physiol. Chem. 76, pp. 214-225

(1912).

12. Willstatter and Mieg: Ann. d. Chemie 355, p. i (1907).

13. Windaus: Zeit. f. Physiol. Chem. 65, p. no (1909).

CAROTIN— THE PRINCIPAL NATURAL YELLOW
PIGMENT OF MILK FAT.*— PART III. ;,

The Figments of the Body Fat, Corpus Luteum and Skin Secretions

of the Cow.

LEROY S. PALMER AND C. H. ECKLES

Recent investigations in regard to yellow animal pigments have
shown that some of them are closely related chemically or identical
with yellow pigments of plant origin. Willstatter and Escher * have
found that the pigment of egg yolk is isomeric with a crystalline xan-
thophyll of green plants, and Escher2 has found the pigment of the
corpus luteum to be identical with the widely distributed hydro-
carbon pigment, the carotin of fruits, flowers and green plants.

In a study of the commonly observed yellow lipochrome of but-
ter fat we have found3 that it is composed principally of a pigment
identical with carotin, with one or more minor constituents which
are evidently identical with the xanthophyll pigments. We have fur-
thermore shown by an unbroken chain of evidence that these pigments
are present in the milk fat as a result of feeding a ration containing
an abundant amount of these pigments. The presence of these pig-
ments in milk fat is therefore not due to any synthetic powers which
the animal possesses, but merely to the fact that the organism ab-
sorbs the pigments along with the products of food digestion and
subsequently secretes them dissolved in the milk fat. We were ac-
cordingly able to vary the amount of pigment in the milk fat by sim-
ply choosing the proper feeds, i. e., either deficient in carotin and
xanthophylls or very rich in these pigments. The above relations
between the carotin and xanthophylls of milk fat and the carotin
and xanthophylls of feeds were found to hold good for all breeds
of dairy cows.

1. Zeit. f. Physiol. Chem. 76, pp. 214-225 (1912).

2. Zeit. f.Physiol. Chem. 83, p. 198 (1913.

3. Research Bulletin No. 10 Missouri Agr. Exp. Sta; Jour. Biol. Chem.
17, p. 191 (1914).

*See Foreword, Part I, for statement of co-operation with Dairy Division,
U. S. Department of Agriculture.

(391)

392 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. II

The establishment of the chemical identity of the pigment of
milk fat and of its simple physiological relation to the carotin and
xanthophylls of green plants at once opened the question of a sim-
ilar relation of the pigment which is so often observed in the body
fat of cows, especially those of certain breeds, such as the Jersey and
Guernsey. The question is also raised as to the presence of xan-
thophylls in the corpus luteum pigment. In addition an interesting
question is opened as to whether the yellow skin secretions of cer-
tain breeds of dairy cows, which is often interpreted as indicating
the ability of these animals to secrete yellow milk fat, is also due
to the same pigments that characterize the butter fat.

The present investigation was undertaken for the purpose of
studying these questions. In addition some information was gathered
relative to the relation of the breed of the cow to the amount of color
found in the body fat.

METHODS OF IDENTIFICATION.

The general methods of studying and identifying the pigments
of the body fat, corpus luteum and skin secretions of the cow were
the same as were used in the study of the pigment of milk fat.. A
detailed account of these methods may be found in the preceding
paper of this series, which deals with the milk fat pigment.

These methods were a study of what we have called the spec-
troscopic, solubility, and adsorption properties of the pigments. The
methods were confined to characteristic, physical and chemical prop-
erties of the pigments for the same reasons that they were used for
the study of the milk fat pigment, namely because, in the case of the
body fat at least, the very large amount of fat with which the pig-
ments are associated precludes their isolation in sufficient quantity
for chemical analysis. In the case of the pigments of the corpus
luteum and skin secretions not enough material was available for
isolating any great quantity of pigment.1

The Spectroscopic Properties. — It was found that carotin and
xanthophylls isolated from green alfalfa hay, carrots or other plants
rich in these pigments showed characteristic absorption bands when
viewed in a spectroscope of narrow dispersion. When the spectro-
scope was set at an arbitrarily chosen standard, each class of pig-
ments exhibited bands in characteristic position, especially in carbon
bisulphide solution, and could then be readily identified. The arbitrary

1. Escher succeeded in isolating less than 0.5 gram of impure crystals of
corpus luteum pigment from 10,000 cows' ovaries.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 393

standard was obtained by fixing the arbitrary scale attached to the
spectroscope at a constant figure which was furnished by a sodium
flame, the spectrometer slit being closed to furnish the narrowest
possible line. This standard did not of course give absolute meas-
urements of absorption bands but merely a means of comparing the
position of bands of various solutions, which was the desired end
in view. Before measuring the bands of an unknown pigment, the
strength of its solution was adjusted to give bands of as nearly the
same intensity and clearness as the bands whose arbitrary measure-
ments furnished the standard. The arbitrary standards for the ab-
sorption bands of carotin and xanthophylls, which were adopted, are
given in Table I.

TABLE 1. — SPIJCTROSCOPIC STANDARDS or CAROTIN AND XANTHOPHYIXS.

Pigment

Solvent

Measurement of absorption bands.

Band I

Band II

Band III

Carotin
Carotin

CS,

C2H5(OH)

225—242
257—275

261—278
303—318

301—319
345—364

Xanthophyll
Xanthophyll

CS2

C2H5(OH)

233—253
263—280

272—291
305—325

312—330

355—.. .

The Solubility Properties. — The relative solubility properties of
carotin and xanthophylls are based on the fact that organic compounds
are best soluble in solvents of similar composition. Accordingly car-
otin, which is a hydrocarbon, is much more soluble in hydrocarbons
like petroleum ether than in the alcohols. Similarly the xanthophylls
are much more soluble in the alcohols than in a hydrocarbon like
petroleum ether. These phenomena, as stated in the preceding paper
of this series, were discovered and elaborated by Tswett l and by
Willstatter and Mieg.2 At this station they were found to be very
characteristic of carotin and xanthophylls and in addition very use-
ful for separating and differentiating the carotin and xanthophyll
constituents, not only of plants but also of the milk fat pigment.
Thus the xanthophyll constituents of a mixed pigment could be
readily separated by shaking the petroleum ether solution of the

1. Ber. d. Deut. Botan. Gessel. 24, pp. 316, 384 (1906); 29, p. 630 (1911).

2. Ann. d. Chemie. 355, p. 1 (1907).

394 MISSOURI AGRICULTURAL EXP. STA._, RESEARCH BULLETIN NO. II

mixed pigment with eighty to ninety per cent alcohol. Or if an
eighty to ninety per cent alcoholic solution of the mixed pigments
was shaken with petroleum ether, the latter solvent would completely
extract the carotin, leaving the xanthophylls in the alcohol. By the
second method especially, it was possible to show the presence of
xanthophyll pigments in butter fat which could not be extracted
from their alcoholic solution by petroleum ether.

The Adsorption Properties. — These properties were discovered by
Tswett.1 They are based on the fact that carotin and the xan-
thophylls show a great difference in regard to the ease with which
they enter into combination with certain finely divided organic and
inorganic compounds, such as Inulin, Saccharose or CaCO3 For
instance, carotin is not adsorbed at all by CaCO3 from its perfectly
anhydrous carbon bisulphide or petroleum ether solutions, while the
xanthophylls are adsorbed to a greater or kss extent. Briefly then,
it has been found that if a carbon bisulphide solution (in which
the pigments have an unusually brilliant color) of the mixed pig-
ments is filtered slowly through a column of CaCO3, previously mois-
tened with the solvent, the pigments will be differentiated into zone-
like rings as they pass through the column, depending on their ad-
sorption affinity towards the CaCO3. Carotin being unadsorbed will
pass through first as a rose or red orange colored zone, with the
various xanthophylls distributed above as zones of different shades
of yellow or orange. The xanthophylls which are completely ad-
sorbed by the CaCOs can be washed out afterwards by a stream of
petroleum ether containing ten per cent absolute alcohol.

THE PIGMENTS OF THE BODY FAT.

The pigment of the body fat of the cow has never been subjected
to a critical examination. Newbigin 2 reports the only attempt to
identify it. He extracted the pigment from a sample of bright
yellow body fat and compared its properties with those of a yellow
pigment which he isolated from the salmon, which pigment, he says,
belongs to a widely distributed group of animal pigments commonly
confounded with the lipochrome pigments. He found the body fat
pigment very similar in properties to the yellow non-lipochrome
pigment. It did not gfive the lipochrome properties and was very
little soluble in methyl alcohol. Newbigin also compared the body

1. Ber. d. Deut. Botan. Gessel. 24, pp. 316, 384 (1906).

2. D. Noel Patton, Report of Inv. on Life History of Salmon (1898),
Article XN, p. 159.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 395

fat pigment with the pigment from maize, with the result that, "The
maize pigment gave the lipochrome reaction faintly with H2SO4,
distinctly with HNO3 while the fat pigment gave no lipochrome re-
action. In other respects, in tint, solubility, etc., the pigments closely
resembled each other." The experience of this station in studying
the lipochrome properties of the milk fat pigment seems to indi-
cate that Newbigin's results were due to the fact that he was work-
ing with decomposed pigment. Nevertheless particular attention was
paid to the lipochrome properties of the body fat pigment.

Method of Isolation.

The method was the same as that used for isolating the milk
fat pigment. It consisted in careful saponification of the fat with
alcoholic potash (2 c.c. of 20% solution per gram of fat) and sub-
sequent extraction of the unsaponifiable matter from the diluted
soap (3 volumes of water to one of soap) with ether. The ether ex-
tract was sometimes purified by re-saponification and re-extraction
with ether, and sometimes was freed from cholesterol with digitonin.
Only small amounts of fat were used for each test, i. e., 25 to 30
grams, as the studies of the milk fat pigment showed that the best
results could thus be obtained.

Identification of Pigments.

Only two typical experiments showing the character of the body
fat pigments will be reported.

Experiment I.

Twenty-five to thirty grams of pure rendered kidney fat from
a Jersey cow was used for this experiment. The fat had a high
yellow color testing in the one-inch tintometer cell 54 yellow, 2.3 red.
The unsaponifiable matter of the fat had a golden yellow color in
ether solution and a blood red color in carbon bisulphide solution.
The carbon bisulphide solution left no adsorbed zone in the CaCO3
when analyzed chromotographically, but passed through unadsorbed
as an orange-red or rose colored zone. A very small amount of
pigment was left in the column, however, which was readily washed
out by a stream of alcoholic petroleum ether.

The main pigment was now examined with respect to its solu-
bility relations toward petroleum ether and 80 per cent alcohol. A
very minor constituent was thus obtained more soluble in 80 per

396 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. II

cent alcohol than in petroleum ether (b. p. 30-50). This constituent,
when combined with the pigment adsorbed by the CaCO3 in the
chromotogramm, did not show sufficiently sharp absorption bands
for accurate measurements.

The main petroleum ether soluble pigment was transferred to
carbon bisulphide in which it showed two strong bands and a third
faint one, the measurements of which are given in Table 2. The
residue from this solution gave a greenish-blue color with concen-
trated sulphuric acid.

, Experiment II.

An unusually high colored fat taken from the back of a Jersey
cow was used for this experiment. The pure rendered fat tested
in the one-inch tintometer cell 80 yellow, 2.7 red. Thirty grams
of this fat was saponified with 300 c. c. of 5 per cent alcoholic pot-
ash by dissolving the fat in the hot alkaline solution, letting stand
for 24 hours in the cold with frequent shaking and finally boiling on
the steam bath for one half hour. By using this procedure not a
trace of foreign pigment was developed. One extraction with ether
rendered the diluted soap colorless. The golden yellow ether ex-
tract, after purification, was concentrated into 50 c. c. of absolute
alcohol. On careful analysis of this solution it was found possible
to separate its pigment into a major pigment which was readily ex-
tracted by petroleum ether and a minor pigment which could not be
extracted by petroleum ether. The main petroleum ether soluble
pigment was readily soluble in carbon bisulphide with a blood red
color, and in this solvent showed two strong absorption bands and
a third faint one, the measurements of which are given in Table 2.
Analyzed chromotographically, the carbon bisulphide solution passed
through as an unadsorbed beautiful rose colored zone. There was
no differentiation. The residue from the solution gave a deep blue
color with concentrated sulphuric acid.

The alcohol soluble pigment, which probably comprised several
per cent of the total, was transferred to ether by diluting the alco-
holic solution with much water in a separatory funnel. Petroleum
ether was added, precipitating some water, and the ethereal solu-
tion washed with water until clear. The solution was now evap-
orated and the yellow residue dissolved in carbon bisulphide, giving
a yellow-orange solution which showed two fine absorption bands, and
a third fainter one, the measurements of which are given in Table
2. The bands seem to be shifted more toward the blue than the
usual xanthophyll bands.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 397

The orange-yellow carbon bisulphide solution was now analyzed
chromotographically. Only one pigment was present, which passed
through the column very slowly as a narrow orange zone, leaving
no pigment in the CaCO3 which could be washed out with alcoholic
petroleum ether.

TABLE 2. — ABSORPTION BANDS OF CAROTIN AND XANTHOPHTTLLS OF BODY FAT.

Experiment
No.

Measurements of absorption bands

Carotin

Xanthophylls

1.

Band I
Band II
Band III

223—242
259—279
300—319

2.

Band I
Band II
Band III

224—243
262—288
302—322

Band I 235—252
Band II 278—302
Band III 315—335

It must necessarily be concluded from these experiments that
Newbigin's "inert" class of lipochromes does not exist in the body
fat of the cow, but, on the other hand, the pigment of this fat is, like
the butter fat pigment, composed of a major carotin and one or more
minor xanthophyll constituents, all of which also show the proper-
ties of lipochromes. It is to be noted also that the number of xan-
thophylls in the body fat varies, as was found to be the case in the
butter fat studies.

The Relation Between the Cofor of the Body Fat and the Food of the Cow.

Numerous feeding experiments in connection with the study
of the pigment of milk, reported in the preceding paper of this series,
showed that the carotin and xanthophylls which were found to char-
acterize the milk fat were present there on account of the fact
that the food contained these pigments. Since the pigment of the
body fat is also composed of carotin and xanthophylls it is natural
to suppose that it is, like the pigment of milk fat, derived from the
food, the carotin and xanthophylls being carried to the fat depots and
fat synthesizing body cells in the same manner that they are carried to
the milk glands.

In order to obtain evidence of this fact, however, the following
experiment was undertaken. Two barren and dry Jersey cows in

398 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. II

moderate flesh were fed wheat straw alone for sixty days or until the
animals had lost as much fat as was considered necessary for the
second part of the experiment. The daily ration of cow No. 25
was then changed to 9 pounds of yellow corn and 20 pounds of green
alfalfa hay, which was rich in carotin and xanthophylls. Cow No. 21
was given a daily ration averaging 11.4 pounds of white corn and
14 pounds of bleached clover hay, very deficient in carotin and xan-
thophylls. Cow No. 25 was slaughtered at the end of 81 days. Her
gain in weight during this period was 160 pounds. Cow No. 21
was slaughtered at the end of 95 days. She had gained materially
in condition during her "fattening" period although the scales showed
little gain in weight. This was probably due to a much greater "fill"
when receiving wheat straw. Samples of fat from various parts of
the body were taken from each cow at slaughtering and used for
color readings. The results aire given in Table 3. The colorimetric
readings in this and subsequent tables were made on the rendered,
melted fat, measured by the Lovibond Tintometer. A complete de-
scription of this instrument may be found in the preceding paper of
this series.

TABLE 3. — THE RELATION OF FEED TO COLOR OF BODY FAT.

Color

of fat.

Part of body.

<

:ow No

. 25.

Cow I

^o. 21.

Yellow

Red

Light

Yellow

Red

Light

Rib plate .

50

2.3

1.0

1.4

0.1

0

Caul.

47

2.1

1.0

3.6

0.5

0

29

1.3

1.0

8.0

1.0

0

Around ovaries and uterus

49

2.3

1.0

2.5

0.3

0

Attached to fourth stomach ....

33

1.6

1.0

24.0

1.7

.0

In pelvic cavity . .

50

2.3

1.0

47,0

2.1

.0

Kidney

54

1.6

1.0

50.0

2.1

.0

Crops .

50

2.3

1.0

47.0

1.9

.0

Over last rib

47

2.3

1.0

50.0

2.1

.0

Over outside chuck. .

47

2.0

1.0

47.0

1.8

.0

The results of this experiment are even more striking when
the amount of fat on the various parts of the bodies of the two
cows is taken into consideration. Aside from the kidney fat and
pelvic cavity fat, which were probably not disturbed to any extent
during the starvation period, and which furthermore were of equal
color in the two animals, the largest proportion of the entire fat

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 399

of the two cows was represented by the caul fat. The caul apron
of Cow 25 had a net weight of 12,132 grams; that of Cow 21 a net
weight of 7,364 grams. It is here that the great difference in color
is noticeable. In fact, if all the fat on the body of Cow 21 had
been rendered and its color compared with the same from Cow 25,
the difference in color would have been very marked.

There can be no doubt that the above data is conclusive as to
the effect of feeding a non-pigmented ration to a fattening cow whose
fat had been largely eliminated by starvation previous to the feed-
ing of the colorless ration. In other words, it is apparent that the
body fat of the Jersey cow is colored primarily because the food is
rich in carotin and xanthophylls during the time the fat is on. The
blood serum of both cows contained carotin at the time of slaugh-
tering. Unfortunately no comparison was made of the amount in
the serum of the two cows. No doubt there was a much smaller
amount in the blood serum of Cow 21, than in the serum of Cow 25. In
both cows there was no known path of elimination of the blood pig-
ment during the starvation period, the cows being both dry and
barren. During the fattening period of Cow 25, the demands made
on the blood store by the body fat producing cells, was replenished
by the food. In the case of Cow 21, however, there was no replen-
ishing source, and the amount in the serum must have greatly
diminished.

The data resulting from this experiment have some significance
aside from the relation of the body fat to the food of the cow. The
fact that the outside fats of the two cows were of equal color, and
the inside fats, especially the caul and rib plate fat, were of such
widely different colors, would seem to indicate what fats are drawn
upon first in starvation in this class of animals, and what fats are
laid on first during fattening.

Relation Between Color of Body Fat and Breed of Cow.

Considerable attention was given in connection with the study
of the milk fat pigment, to a study of the relation of the breed to
the amount of pigment in the fat. This study showed that the re-
lation of breed to milk fat coloration is a relative one, Holstein and
Ayrshire cows producing well-colored butter fat as well as Jerseys
and Guernseys under proper feeding conditions. It was also shown
that under certain conditions there was no difference in milk fat
coloration among the different breeds. These results naturally raised
the question whether a similar study of the body fat pigmentation

4OO MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. II

would lead to the same conclusions. It is not uncommon to find
the body fat of Jersey and Guernsey cows with a high yellow color.
This has led to a general belief that this phenomenon is a character-
istic of only these breeds of cows, As a matter of fact butchers
and also the consumer look with disfavor upon beef from these
animals on account of this high color of the fat. Although Table
3, above, shows very clearly that the color of the body fat of Jersey
cows is as much dependent upon the feed as the color of the milk
fat, it was nevertheless important to study the coloration of the
body fat, of the different breeds, which had accumulated under ordi-
nary conditions. In this way the normal breed relation could be
determined.

Only a few animals were available for this study. Besides the
data for the two Jersey cows given in Table 2, we have the colori-
metric study of the body fat of one Jersey and 3 Holstein cows.
The data from these animals is given in Table 4.

TABLE 4. — RELATION OF BREED TO COLOR OF BODY PAT.

Cow No. 2,

Cow No. 207,

Cow No. 226,

Cow No. 221

Jersey

Holstein

Holstein

Holstein

Part of body

Yellow

Red

Yellow

Red

Yellow

Red

Yellow

Red

Rib plate

54

1.5

47

1.7

15.0

1.2

Crops

63

1.8

14

1.2

.

21.0

1.2

Thoracic cavity

17

1.0

36

1.5

'9

1.0

6.0

1.2

Caul

50

1.7

54

1.7

18

1.1

12.0

1.2

Pelvic cavity

47

1.5

61

1.8

. .

10.0

1.2

Over last rib

63

1.8

17

1.3

.

23.0

1.2

Ovaries, uterus

62

1.8

v

Chuck

54

i'.7

14

1.0

22!

1.2

Kidney

47

1.5

64

1.8

20.

1.2

Stomach

24

i!o

11.

1.2

The most important points presented in this table are the wide
difference between the color of the fats of Holstein Cow No. 207
and the other two Holstein cows ; and the wide difference between
the color of the inside and outside fats of Holstein Cow No. 207.
The first point is possibly due to an individual characteristic of Cow
No. 207, although it is not known under what conditions the fat
was formed. It should be stated in connection with the data of
Cow No. 226 that the animal died in parturition, and the reason
so few samples of fat are recorded is due to the fact that the animal
had no fat on the body at those particular places. In regard to the

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 4OI

data on Cow No. 207, it is to be noticed that the inside fat all had
a color equal or greater than the corresponding fat of Jersey Cow
No. 2, while the outside fats were uniformly much lighter in color.
In explanation of this result it may be said that milk cows are known
to lay on fat first on the inside of their body, and we have data to
show, not only that this particular Holstein cow normally produced
high-colored milk fat under favorable feeding conditions, but also
that the laying on of most of the fat, whose color is shown in the
table, was during the summer when her ration was largely fresh
green grass.

This does not hold true for Holstein Cow No. 221, for much of
her fat was also laid on while on grass. It should be added too
that the milk fat of this cow was never known to have a very high
color. This was brought out especially in a carrot-feeding experi-
ment with this animal which was reported in the preceding paper
of this series. The maximum color obtained in that experiment was
practically the same as the maximum color found in her body fat.
There seems to be a breed characteristic evident here, but owing
to the high color readings obtained from Holstein Cow No. 207, it
may be due to the individual rather than to the breed.

Perhaps the most important point brought out by this data is
that the color of the body fat of any individual, regardless of breed,
laid on under given feeding conditions is practically the same as the
color of the milk fat under the same conditions.

Another point which should be mentioned here, but which will
be more readily understood in the light of the results which will
be given in a subsequent paper, is that this difference between indi-
viduals is not due to lack of carotin in the blood. The amount of
carotin in the blood of Cow No. 221 at the time of slaughtering was
as great as is found in the blood of a Jersey cow receiving the same
feed.

Our data is not sufficiently extensive to warrant any conclusions
as to the normal difference in body fat pigmentation between the
different breeds. Very probably it is considerably greater than the
normal difference between the milk fat pigmentation of the different
breeds. The reason for this is not evident from our present knowl-
edge of the physiology of pigmentation. The fact that such a wide
difference often does exist is of considerable importance, however,
in explaining why the milk fat of Jersey and Guernsey cows often
has a higher color than can be explained by the character of the
ration. Reference was made to this in connection with the feeding
experiments reported in the preceding paper on the milk fat pigment.

4O2 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. II

It was there shown that changing the ration of a Jersey cow from
one rich in carotin and xanthophylls to an unpalatable one very poor
in these pigments did not result in an immediate lowering of the
color of the butter fat, but resulted rather in a gradual reduction in
color, extending over a considerable period of time. The animal
at the same time usually lost weight. This fact taken in connection
with the normal high color of the body fat of Jersey cows which
was brought out in the present experiments, gives a clear explana-
tion of the entire phenomenon. The pigments of the body fat were
being drawn upon, or rather the utilization of the body fat for energy
liberated pigments which furnished a partial temporary supply for
the milk fat. The milk fat of cows, whose body fat lacked this high
color would, therefore, under similar conditions lose color very much
faster. The high color of the milk fat of Jersey cows on a nonpig-
mented ration is, therefore, due to the fact that their body fat has
a normal high yellow color.

THE PIGMENTS OF THE CORPUS LUTEUM.

Viewed in the light of the foregoing investigations it is not
surprising that Escher l has found that the corpus luteum pigment
belongs to the carotin group, thus establishing its identity with the
principal milk fat and body fat pigments. In view of the plurality that
has been established for both the milk and body fat pigments it
became at once important to study the corpus luteum pigment in
this connection also. Only a few corpora lutea were available for
the study, in fact the ovaries of only six cows at different times were
available for examination and in three cases only were well-devel-
oped corpora lutea found. Of Jersey Cows No. 21 and No. 25 slaugh-
tered at the same time, only Cow No. 25 had a corpus luteum of any
development. Jersey Cow No. 8 and a Hereford cow were slaugh-
tered at the same time but only the beef bred cow had a well-de-
veloped corpus luteum. Cow No. 207 slaughtered at another time
had no well-developed corpus luteum but there were the remains
of a number of former corpora lutea and one just developing. Hoi-
stein Cow No. 221 slaughtered some time later, had a well-developed
corpus luteum.

The investigations of the corpora lutea of the Jersey cows, Nos.
21 and 25, were carried out on the combined pigments previous to
the discovery of the xanthophyll constituent of butter fat pigment,
but the data obtained is nevertheless very instructive.

1. Loc. cit.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 403

The corpora lutea were carefully cut away from the surround-
ing tissue, ground up with sand, and extracted with ether. In this
solvent and in alcohol, the pigment showed two absorption bands
in the blue part of the spectrum. Solubility tests on the alcohol
solution showed that petroleum ether and carbon bisulphide ex-
tracted almost all the pigment. That which was not extracted was
treated with hot alcoholic potash and the soap extracted with ether in
which the pigment all readily went. The saponified pigment was trans-
ferred to alcohol and freed from cholesterol with digitonin. After
concentrating, the cholesterol-free filtrate was extracted with carbon
bisulphide. Not all the pigment was extracted even with two ex-
tractions, and petroleum ether extracted no color from the remain-
ing light-yellow alcoholic solution. The experiments with the sec-
ondary pigment were not carried farther at this time as the signifi-
cance of its presence was not appreciated, but viewing the data in
the light of the results of the milk fat and body fat investigations
it is evident that a secondary xanthophyll pigment was present here.
This has been emphasized because the results of the investigations
subsequently conducted were unfortunately vitiated because of un-
expected aldehyde resin colorations which developed during saponi-
fication. It was shown by a special study that these reddish yellow
bodies when present in considerable quantity are extracted from
the diluted alkaline solutions by ether, but are not readily extracted
from alcohol by petroleum ether. Consequently they interfere with
a proper study of the pure pigments. Such a result was obtained
in the study of the corpora lutea pigments of Cow No. 8 and the
Hereford cow. The only noteworthy result of that investigation was
to obtain a beautiful rose colored unadsorbed zone in a chromoto-
gramm of a carbon bisulphide solution of the combined pigment.
This solution showed three absorption bands, the measurements of
which are given in Table 5.

The next investigation was with the corpora lutea of Hoi stein
Cow No. 207. As stated above there was no well-developed corpus
luteum, the largest part of the pigment obtained being from the re-
mains of several former corpora lutea which were present as small
red colored patches about the size of a pin head. These were care-
fully cut out and macerated with a little sand and CaSO4 and extracted
with carbon bisulphide for several hours. The solution, of about
25-50 c. c. volume, had a deep orange-red color, which showed" three
beautiful bands, the third band being considerably fainter than the
first two. The measurements are given in Table 5.

404 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. II

A chromotogramm of this solution showed only one rose col-
ored zone which passed rapidly through the CaCO3 column. A little
of the solution was evaporated into absolute alcohol and after mak-
ing the alcohol eighty per cent, the pigment was studied in regard
to its solubility properties toward petroleum ether (b. p. 30-50° C)
and carbon bisulphide respectively. In both cases the alcohol was left
absolutely colorless. In this case then, where the pigment was chiefly
from the remains of former corpora lutea, carotin was the only pig-
ment present.

TABLE 5. — ABSORPTION BANDS OF CORPUS LUTEUM CAROTIN IN CARBON

BISULPHIDE.

Bands

From Hereford Cow No. 8.
Good corpus luteum.

From Holstein Cow No. 207.
Remains of corpus luteum..

Band I
Band II
Band III

225—242
262—282
305—320

225—242
262—285
305—320

The final investigation was with the well-developed corpus lu-
teum from Holstein Cow No. 221. Both ovaries of the cow were
ground up with sand and plaster of paris and the mass extracted
with petroleum ether (b. p. 30-50° C.) until the extract was color-
less. The pigment thus extracted was carefully differentiated be-
tween the petroleum ether and 85 per cent alcohol. No pigment
whatever was extracted by the alcohol. The pigment was then sub-
mitted to saponification with KOH after transferring to alcohol. No
aldehyde resin pigments formed during saponification. The pig-
ment was extracted from the soap with ether. The ether extract
was thoroughly washed with water as usual and then concentrated at
a low temperature with the constant addition of petroleum ether so
that the pigment finally remained in petroleum ether solution. This
solution was then shaken with 85 per cent alcohol. The alcohol
extracted no pigment whatever. In this case then, although a nor-
mal corpus luteum was used, and the entire pigmented extract sub-
mitted to saponification, carotin was the only pigment present.

The result of this study was to confirm the results of Escher *
that the corpus luteum pigment is identical in properties with caro-
tin. In addition we have shown that this pigment, like the prin-
cipal pigments of milk fat and body fat, may have associated with

1. Loc. cit.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 405

it small quantities of xanthophyll pigment.. It is possible, however,
that these xanthophylls are present in the fat which may be extracted
along with the carotin of the corpus luteum.

THE PIGMENTS OF THE WAXY SECRETIONS IN THE EARS AND
ON THE SKIN OF JERSEY COWS.

It was stated in the introduction that the secretions of the skin
of Jersey and Guernsey cows is often considered as indicating the
ability of these breeds to secrete yellow milk fat. It was accordingly
thought that a brief investigation of this pigment would be of inter-
est and possibly of some scientific value.

The yellow skin secretion of Jersey cows is especially abundant
in the ears. A few grams of the yellow waxy matter was accord-
ingly scraped from the ears of several pure bred Jersey cows and
the wax macerated with ether, which readily dissolved away the
pigment and some fatty matter, giving a bright yellow solution. The
ether solution was concentrated to low volume and diluted with
about loo c. c. of 2 per cent alcoholic potash and the solution boiled
on the steam bath for 30 minutes. The pigment was extracted from
the soap solution with ether in the usual way. The extraction of
the pigment was easy and complete. The ether solution was freed
from alkali as usual and then diluted with some petroleum ether. The
slightly cloudy solution which resulted was washed with water until
clear and evaporated into absolute alcohol. The alcohol was now
diluted with petroleum ether (b. p. 30 to 50° C) and water added
sufficient to cause separation. Several extractions with petroleum
ether resulted in the division of the original pigment into a major
petroleum ether soluble pigment and a minor pigment which could
not be extracted from 80-90 per cent alcohol with petroleum ether.

The petroleum ether pigment gave a red orange carbon bisulphide
solution showing the carotin absorption bands:

I. 224—243

II. 263—287

III. 303—320

and a beautiful rose colored unadsorbed zone in the CaCO3 chromoto-
gramm.

The 80 per cent alcohol soluble pigment which amounted tc two
or three per cent of the entire pigment, gave a yellow-orange car-
bon bisulphide solution. The solution showed only one absorption
band however, the other bands being obscure.

I. 232—254

406 MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. II

It is thus seen that the yellow pigment of the skin secretions
of the Jersey cow is identical with the other yellow lipochromes of
the body and like them belongs chiefly to the carotin group of pig-
ments.

THE BODY FAT AND BLOOD SERUM PIGMENTS OF THE NEW-BORN

CALF.

Carotin and xanthophylls having been found to be normal con-
stituents of the body fat of cows which had been fed green feeds or
other feeds containing an abundant amount of these pigments, an in-
teresting question was raised as to whether these pigments are present
in the body of the new-born calf. If these pigments should be found
to be entirely absent from the new-born calf, additional proof would
therefore be offered that these pigments were the result of subse-
quent feeding. The presence of carotin and xanthophylls in the new-
born calf, however, would not be proof that these pigments cannot
arise from the food, but would merely indicate that they were able
to traverse the placental barrier from the mother whose body is
normally rich in these pigments. In this connection the question
would be especially interesting in view of the fact that Mendel and
Daniels l have recently found that fat soluble dyes, such as Sudan
III, do not traverse the placental barrier of small animals such as
cats and rats, whose milk fat and body fat, however, is readily tinted
as the result of feeding the dyes.

In order to study this question, the following experiment was
carried out:

A new-born pure bred Jersey calf weighing 50 pounds was not
allowed to suckle its mother but was slaughtered a few hours after
its birth.

Five hundred c. c. of the blood was caught in a cylinder and
allowed to clot. After standing 48 hours, 250 c. c. of serum was
obtained. The proteins were precipitated from the serum with al-
cohol and were filtered off on a Biicher funnel with suction. They
had a reddish gray color. They were rubbed up to a paste with
absolute alcohol in a mortar and then extracted with boiling abso-
lute alcohol. The extract was absolutely colorless. The alcoholic
filtrate from the precipitated proteins had a greenish-yellow color.
It was concentrated to 50 c. c. and absolute alcohol added, precipitat-
ing a little protein. The filtrate had a beautiful greenish-yellow color
but the pigment was not extracted by carbon bisulphide, petroleum

1. Jour. Biol. Chem. 13 No. 1, p. 72 (1912). ~***

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 407

ether, or ether, but seemed to be partly thrown down by lead acetate
and by saturation of a dilute alcoholic solution with (NH4)2SO4.
Acid mercuric nitrate solution decolorized the alcoholic solution at
the same time throwing down a white precipitate.

The only conclusion that can be drawn from this experiment
is that the blood serum of the new-born calf is free from carotin.
A small amount of an unknown pigment was present which was
readily soluble only in alcohol, and insoluble in water, ether, carbon
bisulphide, and petroleum ether.

There was practically no fat on the body except a little around
the kidneys and in the tissue of the caul apron. In the body the
latter tissue had a slight brownish color. All the fat tissue that
could be obtained was ground up, rendered and filtered. About 40
grams in all was obtained. The rendered fat had a slight yellow
color giving a tintometer reading in i-inch layer of 4 yellow and
.3 red. When solid the fat had a greenish -yellow tint. Thirty grams
of the fat was saponified with alcoholic potash and the soap extracted
with ether. It was possible to differentiate the small amount of
pigment thus obtained so that it was about equally divided between
petroleum ether and 80 per cent alcohol. In carbon bisulphide solu-
tion these portions showed their relation to carotin and xanthophyll
both in the spectroscope and chromotogramm. Both portions showed
two beautiful bands which measured as follows:

Body Fat Carotin Body Fat Xanthophyll

(In CS2) (In CS2)

Band I 225 — 244 Band I 235 — 250

Band II 263—280 Band II 270—285

The results of this experiment show that a small amount of
carotin and xanthophyll are present in the body fat of the new-born
Jersey calf. The results present the apparent anomaly, however, of
the presence of the pigments in the body fat and their absence in
the blood. In explanation of this it may be said that the body fat
of the new-born calf, the amount of which is very small indeed,
probably arises from the fat of the mother, being transferred to the
foetus a very small quantity at a time. The small quantity which
would be present in the blood stream under these conditions, i. e.,
dissolved in fat, would not have been detectable by the method used
for the investigation of the pigments of the blood serum. It is abso-
lutely certain that there were none of these pigments present in
the blood serum in the way in which they are normally found in the
mature animal.

408 MISSOURI AGRICULTURAL EXP. STAV RESEARCH BULLETIN NO. II

The results of the investigation are of further value in indicat-
ing that under proper feeding conditions, it might be possible to
raise even a Jersey cow with practically none of the characteristic
carotin and xanthophyll pigments in her body.

DISCUSSION OF RESULTS.

The results of the experiments reported in this paper are in
perfect accord with those of the preceding paper. The discovery of
the carotin and xanthophyll nature of the milk fat pigment would
lead quite naturally to the supposition that the other lipochrome pig-
ments of the body of the cow are of the same character. This sup-
position was fully borne out by the result of experimental study.
The yellow lipochromes of the body fat, of the corpus luteum and
of the skin secretions were found to be composed principally of car-
otin with one or more minor xanthophyll constituents.

In addition to the establishment of the chemical relation of these
pigments to the carotin and xanthophyl of green plants in the case
of the body fat at least it has been possible to show that the pig-
ments are derived from the food in a manner identical with pig-
ments of the milk fat. The carotin and xanthophylls of the corpus
luteum and skin secretions must therefore be derived from the same
source.

Viewing the results from a physiological standpoint, it is seen
that the establishment of such a source for these pigments and the
ease with which they are therefore increased and decreased,1 throws
great doubt upon any physiological significance which these pigments
have been supposed to exert in the animal body. In the case of the
corpus luteum for instance, the accumulation of the carotin during
the formation of this body is merely a phenomenon incidental to the
rupture of the Graafian follicle and the subsequent formation of the
cellular tissue around the central blood clot, and to the fact that the
blood serum is normally very rich in carotin, as will be shown in
the following paper. This, of course, does not explain the mechanism
of the accumulation of the carotin-containing cells around the ruptured
Graafian follicle. The chemical combination of the carotin in the
blood serum is no doubt of importance in this connection.

The popular opinion that the body fat of Jersey cows is nor-
mally characterized by a higher yellow color than Holstein cows has
been at le'ast partially confirmed by experimental study, although it

1. This is especially true of the milk fat and, as will be shown in the
succeeding paper, the blood serum.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 409

was found that Holstein cows may also possess high-colored body
fat. At least there seems to be more breed characteristic in this
respect, than in the case of the pigmentation of the milk fat. There
is no foundation, however, for the belief that beef has a lower value
because its fat has a high color. If this pigment is the same as is
demanded by the consumer for butter, why should not beef with high-
colored fat also be more desirable ? It is recognized of course that
some of the unfavorable attitude toward beef with highly colored
fat arises partially from the fact that it indicates that the beef prob-
ably came from a dairy cow. The two ideas are nevertheless very
closely associated.

The normally high color of the body fat of Jersey cows and
also of those of the Guernsey breed, explains why cows of these breeds
often appear to be producing well-colored butter on a ration deficient
in carotin and xanthophylls. Several statements in regard to this
have already been made. This will bear repetition, however, because
the subject is an important one. Briefly, it may be said that when
cows whose body fat has a high yellow color are put upon a ration
deficient in carotin and xanthophylls and also, as is usually the case
with such rations, deficient in food value, the body fat is called upon
to furnish energy value for the animal and also in many cases to
supplement the food digestion products in the production of milk
fat. It is readily seen that in such cases an important source is
opened up for pigments for the milk fat. Just how important this
source could be would depend upon the amount of highly colored
body fat available for the needs of the body, and upon the rapidity
with which it would be used up. If our experimental data are cor-
rect showing that the inside fats, such as the caul fat and rib plate
fat, are the first drawn upon in starvation of this class of animals,
then the amount of available highly colored fat would be rather
large. Dairy cows usually have a fairly abundant amount of these
fats, especially the caul fat. It is thus readily seen that a continuous
drawing upon these inside fats for a long period of time would result
in a very slow and gradual reduction of the color of the milk fat.
The deduction that the animal was actually producing colored milk
fat on a carotin-xanthophyll-free ration would, therefore, be quite nat-
ural but nevertheless entirely false.

In a similar manner it is readily seen why the breeders of Jer-
sey and Guernsey cattle have been led to believe that the yellow skin
secretions of these breeds are indicative of their ability to produce
yellow milk fat. It is interesting to find that the yellow pigments
of these secretions are carotin and xanthophylls. It should be clearly

4IO MISSOURI AGRICULTURAL EXP. STA., RESEARCH BULLETIN NO. II

borne in mind, however, that the only indication that a cow will secrete
yellow milk fat is that the food contains an abundance of carotin and
xanthophylls.

SUMMARY.

1. The yellow lipochrome of the body fat, corpus luteum and
skin secretions of the cow, like the lipochrome of butter fat, is com-
posed principally of a pigment whose physical and chemical properties
are identical with the carotin of green plants. The same pigment may
have associated with it one or more minor constituents whose physi-
cal and chemical properties are identical with the xanthophylls of
green plants.

2. The carotin and xanthophyll pigments of the body fat are
derived from the food of the cow. The body fat of Jersey cows
formed on a ration deficient in carotin and xanthophylls, is devoid
of color.

3. The body fat of Jersey and Guernsey cows is usually char-
acterized by a higher yellow color than cows of other breeds. This
is of great importance in explaining why cows of these breeds may
sometimes show a much slower elimination of the pigment from
milk fat on a non-pigmented ration, as during the winter months.
In these cases the body fat furnishes a supplementary source of pig-
ments for the milk fat.

4. The yellow body fat of Jersey and Guernsey cows should
not be a point against the use of these animals for beef. The pig-
ments here are the same as those for which the consumer will pay a
higher price when present in butter.

5. The breeders of Jersey and Guernsey cattle are probably
correct in their belief that the yellow skin and skin secretions of
these animals are characteristic of the breeds. It is not correct, how-
ever, that this characteristic is indicative of the ability of these ani-
mals to secrete yellow milk fat under all conditions. The only indi-
cation of this is whether the food contains an abundance of carotin
and xanthophylls.

6. The blood serum of the new-born Jersey calf is free from
carotin and xanthophylls. The small amount of fat on the body is
tinted very faintly with these pigments.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 411

BIBLIOGRAPHY.

1. Escher: Zeit. f. Physiol. Chem. 83, p. 198 (1913).

2. Mendel and Daniels: Jour. Biol. Chem. 13, No. i, p. 72 (1912).
Salmon" 1898, Article XV, p. 159.

3. Newbigin: D. Noel Paton, "Report of Inves. on Life History of

4. Tswett: Ber. d. Deut. Botan. Gessel. 24, pp. 316, 384 (1906).

5. Tswett: Ber. d. Deut. Botan. Gessel. 29, p. 630 (1911).

6. Willstatter & Escher: Zeit. f. PhysioL Chem, 76, pp. 214-225
(1912).

7. Willstatter and Mieg: Ann. d. Chem. 355, p. i (1907).

CAROTIN— THE PRINCIPAL NATURAL YELLOW PIGMENT
OF MILK FAT— Part IV, *

A. The Yellow Pigment of Blood Serum.

B. Carotin and Xanthophylls During Digestion.

C. The Pigments of Human Milk Fat.

LEROY S. PALMER and C. H. ECKLES.

A. THE YELLOW PIGMENT OF BLOOD SERUM

Very few investigations have dealt with the so-called lutein of the
blood serum. Thudichum 1 was the first to mention it and classify
it as a lutein. Schunck,2 a number of years later, showed that the
lutein of fowl serum was spectroscopically identical with the L.
xanthophyll which he isolated from yellow flowers and green plants.
Halliburton 3 also studied the lutein of the serum of the hen, but the
pigment isolated by him had evidently lost its spectroscopic properties
in view of Schunck's investigation. Finally, Krukenberg4 extracted
the lutein from ox serum by shaking with amyl alcohol. The extract
showed two absorption bands. He used the designation lipochrome
for the pigment. His work is usually mentioned in the present text
books of physiological chemistry.

Recent investigations in connection with various animal luteins
or lipochromes have shown that they may be classified as belonging
to the widely distributed carotin or xanthophyll groups of pigments
of the vegetable world. Willstatter and Escher5 have identified the
pigment of egg yolk as an isomer of the crystalline xanthophyll of
green plants ; and Escher 6 has shown that the principal corpus luteuni
pigment is identical with the carotin of plants. Extending this work

*See Research Bulletin No. 9, p. 312, for statement of Co-operation with
TJ. S. Dept. of Agriculture.

1. Proc. Roy. Soc. 17, p. 253 (1869).

2. Proc. Roy. Soc. 72, p. 165 (1903).

3. Jour. Physiol. 7, p. 324 (1886).

4. Sitz, Ber. d. Jen. Gessel. (1885).

5. Zeit. f. Physiol. Chem. 76, pp. 214-225 (1912).

6. Zeit f. Physiol. Chem. 83, p. 198 (1913).

(415)

4l6 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

we have shown that the yellow lipochromes of the milk fat and body
fat of cows are also composed principally of carotin, altho both have
associated with them one or more minor xanthophyll constituents. In
addition we have shown conclusively that these pigments originate
from the food of the cow. They are therefore not products of animal
synthesis but merely substances assimilated with the digestion products
of the food and subsequently secreted in the milk fat or laid up in the
body fat. The studies leading to these results are given in the two
preceding papers of this series.1

Up to this time the experimental evidence pointing to the above
stated physiological relation between the carotin and xanthophylls of
plants and the pigments of butter fat and body fat has been based upon
feeding experiments in which the relation between the amount of these
pigments in the food and the color of the milk fat and body fat was
carefully studied. It was recognized however that the evidence would
not be absolutely complete until the means of transporting the food
pigments to the milk fat and body fat could be established. A
close study of the yellow lipochrome of the blood serum, in a man-
ner similar to the preceding studies of the milk fat and body fat
pigments, naturally seemed to offer the most ready means of establish-
ing the physiological relation between the plant pigments and the lipo-
chromes of milk fat and body fat.

The present investigation was therefore undertaken for the pur-
pose of studying the yellow lipochrome of the blood serum in
regard to its chemical and physiological relations to the carotin and
xanthophylls of green plants and to these pigments when found in the
milk fat and body fat of the cow. It was believed that this investigation
would serve the twofold purpose of establishing the connecting physi-
ological link between these plant and animal pigments and also scien-
tifically classifying the blood serum lutein of the cow which pigment
has never been the subject of close investigation.

METHODS OF IDENTIFICATION.

The methods used for identifying the pigment of the blood
serurn were the same as were used in the study of the milk fat and
body fat pigments. They consisted in the application to the isolated
pigment of the characteristic physical and chemical properties of
carotin and xanthophylls. These properties were the position of the
spectroscopic absorption bands, the relative solubility toward petroleum
ether and 80 to 90 per cent alcohol, and the adsorption affinity toward

1. Also Jour. Biol. Chem. 17, No. 2, pp. 191, 211 (1914).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT

417

calcium carbonate. A detailed description of these properties when
applied to both the plant carotin and xanthophylls and the pigments
of milk fat, body fat, corpus luteum and skin secretions of the cow
was given in the two preceding bulletins 1 of this series, and need not
be repeated here. The measurements of the spectroscopic absorption
bands of the carotin and xanthophylls which are used for comparison
were made according to an arbitrarily fixed and standard scale. It
may not be out of place therefore, to repeat here a table which was
given in the paper immediately preceding this one, showing these
standard measurements. This table is given below as Table I. The
measurements in carbon bisulphide solution only are given.

TABLE No. 1. — SPECTROSCOPIC STANDARDS OF CAROTIN AND XANTHOPHYLLS.

Pigment

Solvent

Measurements of absorption bands

Band I

Band II

Band III

Carotin
Xanthophyll

CS2
CS2

225—242!
233—253:

261—278
272—291

301—319
312—330

METHODS OF ISOLATION.

The study of the blood serum lutein required considerable pre-
liminary study of methods of isolation. The amyl alcohol method of
Krukenberg 2 was not considered suitable on account of the high boil-
ing point of the solvent. The method used by Schunck 3 seemed to be
much better suited for the work. He precipitated the proteins from
the serum with alcohol, and as the proteins carried down the lutein he
was able to isolate it by extracting the precipitated proteins with boil-
ing absolute alcohol. Preliminary investigations of the blood serum
lutein using Schunck's method showed, however, that it was applicable
only to serum free from dissolved red blood corpuscles. When
hemoglobin was present it was always carried down with the protein
and some of the red color dissolved in the subsequent alcohol extract.
In addition, the method did not seem to be a quantitative one, some of
the lutein invariably being found in the dilute alcoholic filtrate from
the precipitated proteins. These investigations showed however, that
in every case both petroleum ether and carbon bisulphide almost quanti-

1. Research Bulletins Nos. 10 and 11, Missouri Agr. Exp. Sta.; also Jour.
Biol. Chem. 17, pp. 191, 211 (1914).

2. Loc. cit.

3. Loc. cit

4l8 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

tatively extracted the yellow pigment from its alcoholic solution on
dilution with a little water. This result at once indicated the carotin
nature of the blood serum lutein, and this was confirmed by the experi-
ments reported below.

The methods used for isolating the pigment in these studies varied
somewhat in detail but were all based upon a preliminary, more or
less complete dessication of the blood serum by calcium sulphate
(plaster of Paris). The details are given in connection with the report
of the experiments.

CHEMICAL IDENTIFICATION OF THE PIGMENT.

The following experiments were conducted to show the chemical
relation of the blood serum lutein to the carotin and xanthophylls.

Experiment I.

About 20 cubic centimeters of golden yellow serum from Jersey
Cow No. 8* was mixed with plaster of Paris until almost dry, dried
for a few minutes on the steam bath, the mass pulverized, and shaken
with successive volumes of petroleum either in an Erlenmeyer flask
until no more color appeared in the petroleum ether. The extract was
light yellow in color and no color was extracted from the concentrated
solution by 80 per cent alcohol. The plaster of Paris mass was now
shaken with successive proportions of petroleum ether containing 10 per
cent absolute alcohol, until the extraction was colorless. The resulting
extract had a deep yellow color containing many times as much pigment
as the extract with petroleum ether alone. This solution, after con-
centration, was extracted with 80 per cent alcohol. Apparently no
color was extracted. The petroleum ether solutions were combined,
evaporated to dryness and the residue dissolved at once in carbon
bisulphide giving a deep red-orange solution which showed 3 absorp-
tion bands, Band III being much fainter than the other two. The
measurements of the bands are given in Table No. 2.

Experiment II.

Fifty cubic centimeters of the same serum was completely dessi-
cated with plaster of Paris and the pulverized mass shaken with absolute
alcohol and ether until no more color was extracted. The ether was
distilled off and the golden-yellow alcoholic solution saponified with

*Note: — The serum in this case and in all subsequent cases was obtained
by allowing the freshly drawn blood to clot in a tall cylinder or jar and
the serum which pressed out on standing syphoned off into glass stoppered
bottles.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 419

KOH. After dilution with water, the soap was extracted with ether.
The ether was washed, filtered and evaporated into absolute alcohol.
The alcohol was diluted to an 80 per cent solution and extracted with
petroleum ether (b.p.3O-5o°C.) until no more color was extracted. The
alcohol layer was left with quite a little color, but by far the greatest
part of the color was in the petroleum ether extracts. The petroleum
ether soluble pigment gave a deep red carbon bisulphide solution which
showed 3 absorption bands, the third being faint. The measurements
of the band are given in Table No. 2.

The 80 per cent alcohol soluble pigment showed no clear absorp-
tion bands.

Experiment III.

Fifty cubic centimeters of the same serum was mixed to a thick
paste with plaster of Paris, and the pasty mass shaken thoroughly
with 700 c.c. of hot 95 per cent alcohol. All the color was extracted,
a second extraction with fresh alcohol being colorless. The yellow
extract was concentrated to 100 c.c. An equal volume of 10 per cent
alcoholic potash was added and the solution boiled on the steam bath
for one hour. No aldehyde resin pigments formed. The alkaline
solution was diluted with 3 volumes of distilled water and extracted
with two-thirds of its volume of ether. All the color was extracted
by the one extraction. After washing and filtering, the golden-yellow
extract was evaporated to dryness and the residue taken up at once
with petroleum ether (b.p.3O-5o°C.) This solution was now thor-
oughly shaken with 80 per cent alcohol until no more color was ex-
tracted. Fresh petroleum ether extracted a little color from the alco-
holic extract, which was not re-extracted by fresh 80 per cent alcohol.
The blood serum lutein was now divided into a major and minor
pigment, the major being insoluble in 80 per cent alcohol in the pres-
ence of petroleum ether and the minor being insoluble in petroleum
ether in the presence of 80 per cent alcohol.

The petroleum ether soluble pigment had a blood-red color in CS2
solution and showed 3 absorption bands, the measurements of which
are given in Table No. 2.

Analyzed chromotographically it passed through CaCO3 un-
adsorbed as a beautiful rose-colored area, leaving no adsorbed zones.

The 80 per cent alcohol soluble pigment was transferred to ether
by diluting its alcoholic-ether solution with much water, and from
the ether to carbon bisulphide after evaporation of the former. In
carbon bisulphide it gave an orange-yellow solution showing two
absorption bands in the 25 m.m. cell. The measurements of the bands
are given in Table No. 2.

42O MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

Chromotographic analysis showed the presence of only one pig-
ment which very slowly passed through the CaCO3 as a yellow zone.

Experiment IV.

Serum from Jersey Cow No. 2, to the amount ot 275 cubic
centimeters, was dessicated with a little more than the calculated
amount of plaster of Paris necessary to take up the water, and after
setting over night the hard mass was pulverized in a mortar. The
powder was moistened with 95 per cent alcohol and shaken with
petroleum ether (b.p.3O-5o°C.) until the petroleum ether extracted
no more color, and then with ether until that extract was colorless.1
The petroleum ether extract was concentrated to 50 c.c. and the
solution added to the alcohol-free ether-alcohol extract, which had
been concentrated to about 150 c.c. An equal volume of 4 per cent
alcoholic potash solution was now added to the combined ethereal
solutions, the ethers evaporated off and the alcoholic solution boiled
on the steam bath for a few minutes. The pigment was then ex-
tracted from the alkaline alcoholic solution in the usual way and when
in alcoholic solution was analyzed with respect to petroleum ether
and eighty to ninety per cent alcohol. Two pigments were thus
obtained with proportions of perhaps 95 and 5 per cent of the total.

The petroleum ether soluble pigment gave a red colored residue
which dissolved instantly in carbon bisulphide with a blood-red color,
and showed the most beautiful absorption bands yet seen for this
pigment. Three bands were visible, the third band being considerably
fainter than the other two bands but clear and distinct. The measure-
ments are given in Table No. 2.

1. The mechanism of this method of obtaining the blood serum pigment
is so interesting, its advantages so striking and its results so satisfactory,
that it is worthy of some discussion.

It appears that the addition of just sufficient alcohol (either absolute or
95 per cent) to thoroughly moisten the dessicated serum liberates the main
lutein pigment in such a way that when the moistened mass is shaken with
petroleum ether the result is the same as if an 80 per cent to 90 per cent
alcoholic solution of the isolated pigment is shaken with petroleum ether.
There is the additional advantage however, that the CaSO4 prevents the
formation of emulsions and holds the alcoholic solution so firmly fixed in the
paste that the petroleum ether can be poured away and the use of a
separatory funnel be entirely dispensed with. When all the pigment more
soluble in petroleum ether than in 80 per cent alcohol is thus extracted,
any pigment which remains can be readily extracted with ether which mixes
readily with the dilute alcohol. The pigment thus extracted can be readily
freed from alcohol by shaking with water leaving the last as well as the
first pigment extracted, in low boiling point solvents, an additional decided
advantage in view of the ease with which they are oxidized. It should be
added however that the method for extracting the pigment more soluble in
alcohol than in petroleum ether does not apply well for serum containing
much haemoglobin for in this as well as all other alcohol methods the red
pigment is somewhat soluble in the dilute alcohol.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT

421

Analyzed by means of a chromotogramm the solution showed
only a wide, quickly filtering, unadsorbed, rose-colored zone.

The 80 per cent alcohol soluble pigment was transferred to ether
and then to carbon bisulphide, giving in the latter an orange-yellow
solution showing one band and much end-absorption. In the chromo-
togramm it showed 2 zones close together, an upper orange zone and
a lower canary yellow zone. The carbon bisulphide solution of the
orange zone showed 2 absorption bands and end-absorption. The
measurements are given in Table No. 2.

TABLE No. 2. — SPECTEOSCOPIC ABSORPTION BANDS OF BLOOD SERUM CAROTIN AND

XANTHOPHYLLS.

Absorption bands

Experiment

Solvent

No.

Carotin

Xanthophylls

1

I.

225—242

CS2

II.

263—286

III.

305—322

I.

223—243

2

CS2

II.

262—285

III.

302—325

I.

225—242

I. 232—254

3

CS2

II.

263—286

II. 273—295

III.

305—325

III.

4

CS2

I.

223—242

I. 232—252

II.

262—286

II. 270—292

III.

300—320

III. 310—

DISCUSSION OF EXPERIMENTS.

It must be concluded from the above experiments that the princi-
pal lipochrome of the blood serum of the cow is identical with that
of the milk fat, body fat, and corpus luteum, and as in the case of
these pigments, with the carotin of green plants.

It appears also from the above investigations that a small portion
of the blood lutein pigment is composed of xanthophylls. It was found
to be much more difficult to show their presence in the blood serum
than in the body fat or butter fat. The reason for this is not perfectly
clear but a close study of the investigations throws some light on the
question. It will be noticed that it required complete extraction of

422 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

a comparatively large amount of serum with ether and subsequent
saponification of the fat thus extracted to really demonstrate the pres-
ence of xanthophylls. It is a well-known physiological fact that the
proportion of fat in blood serum is comparatively small.* When
coupled with the above observations, this seems to indicate a relation
between the xanthophylls and the fat carried by the blood. Some
observations which will be reported later, in connection with the
fate of the carotin and xanthophylls during digestion, will furnish
more evidence in this same direction.

PHYSIOLOGICAL RELATION BETWEEN CAROTIN OF BLOOD SERUM

AND FOOD OF COW.

After establishing the chemical relation between the principal
blood serum lipochrome and the carotin of the food, it became im-
portant to establish a similar relation from a physiological standpoint.
Very fortunately this was recognized previous to conducting some of
the important feeding experiments which showed the relation between
the color of milk fat and the food of the cow, and which were re-
ported in Research Bulletin No. io/ Missouri Agr. Exp. Sta., the
second bulletin in this series. It was accordingly arranged to study
the variation in the amount of carotin and xanthophylls in the blood
serum during portions of these feeding experiments. In this way it
could be determined what relation exists between the amount of carotin
and xanthophylls in the blood serum and the amount of these pigments
in the milk fat, as well as the relation between the amount of carotin
and xanthophylls in the serum and amount of these pigments in
the food. Such a study required the devising of some method of
analysis whereby the color of the various blood serums could be com-
pared with each other and also with the color of the butter fat.
The following method was adopted.

In the case of live cows whose blood was to be tested, a trocar
was inserted in the jugular vein and 200 to 250 c. cm. of blood drawn
off into a glass cylinder. As soon as the blood had clotted and suf-
ficient serum had pressed out, two 10 cc. portions were pipetted off
and carefully dessicated with an excess of plaster of Paris. The
powdered mass was moistened with absolute alcohol, and the color
extracted immediately by shaking with the selected solvent until color-
less. For one sample the solvent was ether and for the other sample
the solvent was petroleum ether. In all the studies the petroleum

*Nate: — Hammerstein gives .1 to .7%.

1. Also Jour. Biol. Chem. 17, p. 19! (1914).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT

423

ether extract proved to be the easiest to handle as it was practically
free from water. The extract in each case was carefully evaporated
to a volume of i to 2 cubic centimeters and then made up to 12^/2
cubic centimeters with absolute alcohol, this volume being just suf-
ficient to fill the one-inch cell of the Lovibond Tintometer. The
solutions were analyzed at once in the Tintometer and their color
readings recorded. Duplicate determinations were thus obtained.
This was considered necessary since the method was not free from
error due to possible bleaching of the extracted pigments. The entire
procedure was carried out as quickly as possible. The results of the
duplicate determinations were averaged.

The first series of observations of the color of the milk fat and
blood serum corresponding to various pigmented rations was made
with Ayrshire cow No. 301. These feeding experiments and the result-
ing variation in the color of the milk fat were shown in detail in
Tables 12, 13 and 14 of the preceding bulletin dealing with the milk
fat pigment. The relation between the character of the ration, and
the color of the milk fat and blood serum at stated intervals during
the feeding experiments is shown in Table No. 3, below.

TABLE No. 3. — THE RELATION OF THE CHARACTER OF THE RATION TO THE COLOR

OF THE MILK FAT AND BLOOD SERUM.

AYRSHIRE Cow No. 301.

Date of
sample.

Feed of cow

Butterfat

Serum

Yellow

Red

Yellow

Red

1913

Jan. 7
Jan. 24
Feb. 7

Mar. 1
Mar. 6
Mar. 27

Cottonseed meal and cotton-
seed hulls only

1.3
1.2

2.0
24.0

24.0
7.0

0.4
0.4

0.5
1.3

1.4
1.0

3.3
2.6

4.9
54.0

47.0
26.0

0.5
1.1

1.2
1.8

1.5
0.7

Cottonseed hulls, timothy hay
and white corn

Timothy hay, cottonseed
hulls, cottonseed meal and
yellow corn

Timothy hay, cottonseed
hulls, cottonseed meal,
yellow corn, and 20 Ibs.
carrots per day
Timothy hay, cottonseed
hulls, cottonseed meal, yel-
low corn, and 20 Ibs. car-
rots per day

Timothy hay, cottonseed
hulls, cottonseed meal, and
yellow <~"orn

424 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

The above table shows in a very striking manner that the amount
of carotin in the blood serum of the lactating cow, as well as the
amount of carotin and xanthophylls in the milk fat, is dependent upon
the ration. The figures in the table also show very clearly why the
butter fat in the first two cases was not absolutely colorless. These
two samples were taken at the end of a long-continued feeding of
a ration almost entirely lacking in carotin and xanthophylls, which
was planned for the purpose of eliminating as far as possible the
color from the milk fat. It is evident that this was not accomplished
in the strictest sense of the word because the blood serum in both
cases still contained a small amount of carotin.

The results of these studies were so striking that it was con-
sidered advisable to confirm them if possible with other animals.
We fortunately had at hand, 6 cows of 3 different breeds, all pure
bred animals, which were in ideal condition for such an experiment.
They had all just completed an experiment in which their feed for
12 to 14 weeks had been essentially a non-pigmented one, being made
up of cottonseed meal, corn stover and very light-colored timothy
hay. A night and morning milking of each cow was combined
and after determining the percentage of fat in the combined sample l
the milk was separated, the cream churned, and the color of the
rendered butter fat observed in the Tintometer. The same day, sam-
ples of blood were drawn from each animal and the color of the
serum determined by the method described above. The feed of the
cows was now changed so that it was largely made up of alfalfa
hay, rich in carotin and xanthophylls, and later a little fresh green
pasture grass. Thirty days after the cows had been on this feed
the first experiment was repeated and the color of the butter fat
and blood serum again observed. The results of this experiment
are given in Table No. 4.

1. The weight of the combined sample was also recorded.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 425

TABLE No. 4. — RELATION BETWEEN CHARACTER OF RATION AND AMOUNT OF
PIGMENT IN MILK FAT AND BLOOD SERUM.

Cow

No.

Breed

Date

Grams
fat.

Butterfat

Serum

Yellow

Red

Yellow

Red

213
213

Holstein
Holstein

3-11-13
4-10-13

122
135

8.5
54.0

1.4
1.8

6.0
48.0

0.7
1.1

220 JHolstein
220 iHolstein

I

3-11-13
4-10-13

167
208

3.0
22.0

0.7
1.2

7.0
41.0

0.8
1.0

303 Ayrshire
303 (Ayrshire

3-11-13
4-10-13

213
263

2.5
16.0

0.6
1.1

11.0
40.0

0.9
1.0

16
16

Jersey
Jersey

3-11-13
4-10-13

304

363

11.0
64.0

1.7
2.0

10.0
45.0

0.9
1.1

57
57

Jersey
Jersey

3-11-13
4-10-43

240
263

5.2
54.0

1.2
1.7

13.0
57.0

1.1
1.8

64
64

Jersey
Jersey

3-11-13
4-10-13

281
350

4.7
47.0

1.5
1.6

7.5
45.0

0.7
1.0

The results of this experiment are as striking and conclusive
as in the experiment with Cow No. 301, and show that there is a
direct relation between the amount of carotin in the food and the
amount of lutein in the blood serum, just as there is a direct relation
between the presence of an excess of carotin in the food and the
production of a high-colored butter fat. It is necessarily true also
that there is a direct relation between the color of the butter fat
and the amount of lutein in the blood serum. A small amount of
lutein in the blood serum will always mean a light-colored butter fat.
It does not appear to be necessarily true, however, that a high-colored
serum will be accompanied by a high-colored butter fat. The only
conclusion in this connection that can be drawn from the data of Tables
3 and 4 is that an increase in the color of the blood is accompanied by
an increase in the color of the fat. The actual color of the fat under
these conditions is apparently dependent upon a number of conditions
which are not explained by this data. The amount of fat being
produced and the breed of the animal are both factors which probably
influence the color of the fat. There is certainly a wider difference

426 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

between the color readings of the butter fats than between the color
readings of the blood serum extracts during the second part of the
experiment whose data are given in Table No. 4.

It is possible that this difference may be explained on the ground
that the albumin content of the milk is in some way closely related
to the color of the milk fat. At any rate the data given in the
following table admit of such interpretation. The conditions of
the animal preceding the data were as follows : Cow No. 301 had
been subjected to severe underfeeding, i. e., she received only about
70 per cent of the food required to produce her milk and maintain
her body weight. The food she received during this time was com-
posed of about 5 pounds of a mixture of corn, bran and linseed meal,
and about 7 pounds of alfalfa hay, a ration moderately rich in carotin.
Her ration was then changed to one practically free from carotin,
consisting of white corn, cottonseed meal and bleached alfalfa hay.
The cow was brought back to a normal plane of nutrition on this
ration. The immediate effect on the composition of the milk and
the color of the milk fat is shown in Table 5. The subsequent
effect upon the color of the milk fat is given in Table No. 15 of
the preceding Bulletin of this series which dealt with the milk fat
pigment.

TABLE No. 5. — A POSSIBLE RELATION BETWEEN THE ALBUMIN OF MILK AND THE
COLOR OF THE MILK FAT. *
AYRSHIRE Cow No. 301.

Milk fat

Total

Casein

Albu- Color of fat

Date

Hay

Grain

per day

protein

min '

1912

Lbs.

Lbs.

Grams

Grams

Grams

Grams Yellow

Red

(c) i

9-23-24

7.2fa)

4.9

254

209

138

37 15

1.8

9-25-26

8. Kb)

4.9

249

203

134

36

15

1.8

9-27-28

11.0

6.0

262

222

138

47

19

1.8

9-29-30

14.0

7.0

272

230

135

57

21

1.7

10-1-2

14.5

7.5

267

237

132

67

28

1.8

10-3-4

9.8

8.0

276

220

139

46

20

1.7

(a) Alfalfa hay, rich in carotin.

(b) Bleached alfalfa hay, free from carotin.

(c) Calculated.

The data in Table No. 5 shows that although the conditions of
feed were such that a decline in color would be expected, the reverse
was found. The point to be emphasized is that the sharp increase
in the color of the milk fat was coincident with an increase in the

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT

427

albumin content of the milk. The result points to the probabilit)
of a relation between the higher color and the increase in albumin.
Additional evidence pointing to the same relation is the presence of
extremely high color along with a very abnormal amount of albumin
in colostrum milk, as pointed out later in this paper.

An attempt was made to determine if any definite relation
exists between the albumin and color as found in the milk of
various individuals. The milk of 12 cows representing the Jersey,
Holstein and Ayrshire breeds was used. The feed received was
pasture grass and some grain. The color of the milk fat and the
percentage of albumin were determined for each animal. The results
of the study are given in Table No. 6.

TABLE No. 6. — RELATION BETWEEN THE AI/BUMIN CONTENT OF MILK AND THE
COLOR OF MILK FAT, UNDER NORMAL CONDITIONS.

Pounds

Grams

Grams

Grams

Color of fat.

Breed

Cow No.

milk

protein

albumin

fat

per

per day

per day

per day

day

Yellow

Red

j

14

18.4

311.0

24.45

459.0

64.0

2.7

j

64

12.9

254.0

22.65

380.2

64.0

2.8

j

57

10.8

202.0

22.02

267.0

64.0

3.0

A.

303

20.2

310.0

33.20

363.5

50.0

1.7

H.

221

15.1

270.0

27.05

257.0

43.0

1.8

H.

222

14.5

233.8

20.12

202.5

33.0

1.3

j\

16

20.8

339.0

35.75

414.0

64.0

3.1

j.

317

13.0

234.5

41.00

259.0

80.0

3.5

H.

225

13.3

200.0

22.35

150.9

47.0

1.8

H!

220

16.3

525.0

35.17

192.5

24.0

1.7

A.

301

18.7

282.5

39.00

273.0

54.0

2.0

H.

213

11.4

161.5

19.05

144.2

64.0

2.7

"J." stands for Jersey, "H." for Holstein, "A." for Ayrshire.

There appears to have been no relation between the albumin of
the milk and the color of the milk fat of these animals. It is not
considered however, that these results are conclusive either as prov-
ing or disproving the supposition that such a relation exists. Our
knowledge of the subject is too limited at present to enable us to
control all the factors that enter into the question.

THE TRANSPORTATION OF CAROTIN AND XANTHOPHYLLS BY THE

BLOOD SERUM.

When it had been shown conclusively that the lutein of the
blood serum of the cow is composed of the carotin and xanthophyll
pigments of the food, taken up along the digestive tract and transmitted
by means of the blood to the fat synthesizing cells of the milk glands

428 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

and body tissue, it became a matter of considerable importance to
ascertain how the blood carries these pigments through the body.
It has already been shown that there is a strong possibility that the
minor constituent of the pigment, i.e., the xanthophylls, are carried
in the serum dissolved in the fat. When considering such physiological
phenomena as the great volume of blood in circulation in an animal
as large as the cow and the rapidity with which it circulates, it
would seem very probable that even the very small percentage of
fat in the blood is sufficient to account for all the pigment, both
carotin and xanthophylls, which is presented to the milk glands and
body cells. On the other hand, when one considers the very large
proportion of carotin which is present in any given quantity of blood
serum of a cow receiving a ration rich in carotin, it must be concluded
that the fat plays little if any part in the transportation of this
pigment. The studies that are to be presented will, therefore, consider
only the carotin of the blood serum, since it is this pigment that com-
prises by far the greatest proportion of the lutein of the serum.

It might be considered that the carotin is carried by the serum
merely in simple solution. In fact, Thudichum l stated that the lutein
of the blood is in solution in the serum. This seems to be very
probable especially in view of the fact that Krukenberg2 found that
it could be extracted from the serum by means of amyl alcohol.

Many facts can be presented, however, that go to show that
the carotin does not exist in the serum in simple solution. In the
first place neither carotin from plants nor the carotin of the serum
itself are, when isolated, taken up to any extent when treated with
the pure blood serum. Blood serum almost free from carotin from
natural causes, showed no indication of having taken up the carotin
in either case when poured over the pure amorphus pigments ; and the
serum itself showed no increase in the amount of color that could be
extracted by petroleum ether after dessication with plaster of Paris
and moistening with alcohol.

In addition to the above the following observations were made : 3

I. Five c.c. portions were shaken vigorously with equal volumes
of petroleum ether, ether, CS2 and amyl alcohol respectively. All
extracts were colorless except in the case of the amyl alcohol which
was golden-yellow, and showed the carotin absorption bands both

1. Loc. cit

2. Loc. cit.

3. Except where stated most of work about to be reported was done
with a golden-yellow, high-colored serum from Ayrshire cow No. 301. The
serum was obtained from this cow by drawing 250 c.c. of blood from the
jugular vein and allowing it to clot and the serum to press out. The serum
was free from red corpuscles.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 429

in carbon bisulphide and alcohol, the solution in the first solvent
showing three distinct bands. This shows that the lipochrome which
Krukenberg extracted from ox serum was the pigment which we
have identified as carotin. On addition of alcohol to the other mixtures
the solvents in each case completely extracted the pigment on shaking.

2. Five c.c. portions were dessicated with plaster of Paris and
shaken with ether, petroleum ether, and carbon bisulphide, respectively.
A mere trace of color was extracted in each case. When a little
absolute alcohol was added however, in all cases the solvents became
well colored.

3. Five c.c. of serum was diluted with twenty volumes of water,
without causing any precipitation of the pigment.

4. Twenty-five c.c. of serum was treated with successive por-
tions of saturated (NH4)2 SO4 solution to the following per cent
saturations : 28-35, 36-40, 45-46, and finally to one-half saturation. The
fractionally precipitated globulins were in every case practically free
from carotin, and the half saturated globulin free serum was golden-
yellow. The color was entirely precipitated from a portion of this
solution on complete saturation with (NH4)2 SO4. The remainder
was acidified with a few drops of i^ per cent acetic acid and heated
to about 80 °C. The coagulated albumins carried down only a small
part of the color. The entire pigment was precipitated from the
filtrate, however, on complete saturation with (NH4)2 SO4 in sub-
stance, the light precipitate which came down being deep yellow in
color. This deep yellow precipitate was readily soluble in water
giving a clear yellow aqueous solution from which neither ether nor
petroleum ether would extract any color until the protein in the solu-
tion had first been coagulated with alcohol.

5. Five c.c. of serum was diluted to 25 c.c. with distilled water
and the solution saturated with Mg SO4 in substance. The globulins
were filtered off. The filtrate was golden-yellow. Acetic acid was added
to a concentration of I per cent. The precipitated albumins were
bright yellow, leaving the solution colorless. Petroleum ether and
carbon bisulphide extracted a slight amount of color from this pre-
cipitate on long contact. After the addition of a little alcohol, how-
ever, both solvents readily extracted the color.

6. One hundred c.c. of serum from Jersey Cow No. 25 was
diluted with several volumes of water, a pinch of NaCl and a few
drops of glacial acetic acid added and the solution heated quickly to a
temperature just below the boiling point. The coagulum which formed
contained a very little pigment but the filtrate was golden-yellow. No
color could be extracted from the filtrate by carbon bisulphide, or by

43° MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

ether, even after making strongly alkaline with potassium hydroxide.
Amyl alcohol extracted the pigment.

7. Experiment 6 repeated on 200 c.c. of the same serum gave the
same result. The entire pigment in the golden-yellow nitrate was
coagulated by boiling. The coagulum was not soluble in water.

8. Fifty c.c. of serum (Cow No. 2) was diluted to 350 c.c. with
water, a pinch of salt added and the solution heated on the steam
bath, with stirring until cloudiness appeared. On adding a few drops
of glacial acetic acid, a sharp coagulation took place. On filtering,
the filtrate was bright yellow in color. On saturation of the filtrate
with (NH4)2 SO4 in substance, a comparatively small amount of deep
yellow precipitate was thrown down leaving a colorless supernatant
solution. The yellow precipitate, which was contaminated with a little
(NH4)2 SO4, was readily soluble in water, giving a perfectly clear
yellow solution from which the yellow color was again entirely thrown
down on saturation with (NH4)2 SO4 in substance, or on the addition
of mercuric nitrate. The latter precipitate when still moist would not
give up its color to petroleum ether until first moistened with absolute
alcohol. The bright yellow pigment now found in the petroleum ether
gave a red-orange CS2 solution which showed the three carotin
absorption bands.

9. Two 350 c.c. portions of serum (Jersey Cow No. 2) were
treated respectively as follows :

Portion A was treated with (NH4)2 SO4 in substance to one half
saturation, according to the formula

VC2
X=

18.158 — .54 C2

Where V — original volume of protein solution.
C = desired saturation as grams in 10 c.c.
X = grams to be added to give the required
saturation.

The globulins which precipitated carried down some of the pig-
ment, but on dissolving them in 150 c.c. of warm water containing some
(NH4)2 SO4, and adding (NH4)2 SO4 to half saturation, they were
thrown down practically colorless. The yellow filtrate from this pre-
cipitation was added to the other globulin-free filtrate and the com-
bined solutions diluted to 1500 c.c. with distilled water. This solution
was now raised to a temperature of 75 °C in a water bath. 15 c.c. of
i% per cent acetic acid added and the temperature raised to 80 °C,
when a sharp coagulation occurred. The solution was filtered, giving a

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 43!

perfectly clear bright yellow filtrate. It was saturated with (NH4),>.
SO4 in substance, throwing down a small amount of deep yellow
precipitate. The precipitate was filtered off on a large (n-inch)
Biichner funnel, using suction, so that the layer of yellow protein
would be as free as possible from occluded (NH.4)2 SO4. The golden-
yellow precipitate was sucked as dry as possible on the funnel and the
sticky mass covering the paper in very thin layer dissolved in warm
water, in which it was readily soluble, and the clear yellow solution
set aside.

Portion B was diluted with an equal volume of distilled water,
a little sodium chloride added, and the solution raised to a temperature
of 75 °C. in a water bath. Acetic acid was now added carefully until
a heavy definite coagulation took place. The coagulated proteins car-
ried down some of the pigment but by far the greatest part was in the
clear yellow filtrate. This filtrate was saturated with (NH4)2 SO4
in substance and the precipitated pigmented protein filtered off in the
same way as in the case of portion A. After being made comparatively
dry by suction, the deep yellow residue was readily soluble in a small
amount of cold distilled water.

The two similar solutions from A and B were now combined and
filtered on a small Biichner funnel through several layers of fine paper
to free it from dirt and other foreign matter introduced by the (NHt)2
SO4. The golden-brownish-yellow filtrate of about 250 c.c. volume
had a faint cloudiness when viewed by transmitted light and contained
some (NH4)2 SO4. That the pigment of this solution was carotin was
shown by the fact that when an equal volume of alcohol was added to
15 c.c. of the solution and the mixture was shaken with petroleum
ether, the petroleum ether rose to the top as a golden-yellow solution,
leaving the lower cloudy alcoholic layer colorless. The pigment in
the petroleum ether layer gave a red-orange carbon bisulphide solution,
and in this solvent showed the usual carotin absorption bands.

The aqueous solution was now dialysed in a parchment bag
against running water for eight days. At the end of this time the
solution was still giving a precipitate with barium chloride indicating
that the solution was not free from (NH4)2 SO4. No protein crystal-
lization had occurred, but decomposition had begun, for the solution
was cloudy, and showed a very fine coagulation. This coagulum was
filtered off. It had a dirty brown color and when almost dry was quite
sticky. It was not soluble in water, but both in the dry state and in
suspension in water it gave up a golden-yellow color to ether, leaving-
the precipitate dirty white in color. The extracted pigment showed
the three absorption bands of carotin in carbon bisulphide solution..

432 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

In the solid state the pigment was insoluble in absolute alcohol but
readily soluble in petroleum ether — absolute alcohol, from which the
petroleum ether readily extracted it on dilution with a little water.
The solid pigment was also very difficultly soluble in alcoholic potash.
After saponification for one half hour and extraction by ether, the
pigment, in the solid state, was fused with a little solid sodium
hydroxide and potassium nitrate in a porcelain crucible. The flux was
dissolved in hot water and the solution evaporated to dryness on the
steam bath in the presence of an excess of c. p. HNO3. As much as
possible of the residue was dissolved in hot water containing some c,
p. HNOS, the solution filtered and an equal volume of pure molybdate
solution added to the 100 c.c. of filtrate. On digestion at 60° C, for
several hours there was a distinct yellow precipitate of ammonium
phosphomolybdate.

Returning to the slightly cloudy but yellow aqueous filtrate from
the dialysed solution, we found that the color could be entirely thrown
down; (i) by acid lead acetate as a light yellow precipitate which
bleached almost entirely in 12 hours, but from which petroleum ether
extracted a faint yellow color after soaking in alcohol for about one
hour; (2) by nitric acid mercuric nitrate solution as a bright yellow
precipitate which was very stable and gave up its color to petroleum
ether only after soaking in alcohol; 1 (3) by neutral ten per cent solu-
tion of AgNO3 as a deep yellow precipitate which was stable although
darkening badly as the AgNO3 oxidized in the light, but readily giving
up its color to petroleum ether on addition of alcohol to the precipitate,
the pigment thus extracted showing the carotin bands in CS2 solution ;
(4) on saturation with (NH4)2 SO4 in substance as a deep yellow
precipitate which was not soluble in water but when suspended in water
gave up its color to petroleum ether only after the addition of abso-
lute alcohol; (5) on addition of an excess of alcohol as a yellow pre-
cipitate which when dry gave up no color to petroleum ether alone,
but to alcoholic petroleum ether gave up a yellow pigment which was
quantitatively found in the petroleum ether on separation of the alco-
hol with a little water; (6) on heating the neutral solution to boiling
as a yellow coagulum insoluble in water and giving up no color to hot
alcohol or petroleum ether.

In addition to the above observations the following may be men-
tioned. In working with a large number of samples of blood serum

1. The pigment thus extracted showed the three carotin absorption bands
in carbon bisulphide solution; in alcoholic solution it gave a. pronounced
precipitate of digitonin-cholesteride on addition of hot one per cent digitonin
solution in ninety per cent alcohol.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 433

it was often noticed that when the serum had stood for some time
in a closed bottle in contact with a little supernatant air, an orange-
yellow scum always came to the top. This was found to be a water
insoluble protein which would not give up any color to petroleum
ether when its aqueous suspension was shaken with that solvent, but
when an equal volume of absolute alcohol was added and the shaking
with petroleum ether repeated, the latter solvent rose to the top as a
beautiful yellow layer. The pigment thus extracted gave a red-orange
carbon bisulphide solution showing the three carotin absorption bands.

In order to show more conclusively the character of the protein
with which the serum carotin is evidently combined, the coagulation
temperature of the protein was determined. For this purpose 150 c.c.
of serum was diluted with an equal volume of a saturated solution of
ammonium sulphate. After filtering off the precipitated globulins, por-
tions of the globulin-free filtrate were submitted to fractional coagu-
lation. It was found that on carefully elevating the temperature to
80° C. and holding it at that temperature for a short time, the filtrate
from the coagulated albumins still yielded a large amount of carotin on
addition of alcohol and shaking with petroleum ether. On the other
hand the coagulated albumins yielded a comparatively small amount
of carotin. A similar result was obtained at temperatures of 81°, 82°,
83°, 84°, 85°, and 85.5° C., although the amount of carotin in the
filtrate rapidly decreased with the increase in coagulation temperature.
At 86° C., however, the pigmented protein had completely coagulated,
and the filtrate yielded no carotin on treatment with alcohol and
petroleum ether. The coagulation temperature limits of the pigment
carrying protein therefore lie between 80° and 86° C., when the pro-
tein is in half saturated ammonium sulphate solution. There is no
marked coagulation at the lower temperature, but it is completely
coagulated at the upper temperature.

The coagulation temperature of the protein which carries the
carotin in the blood was studied further with an aqueous solution of
the protein obtained in a manner similar to the one used in obtaining
the protein for the study previously reported. Briefly, an equal volume
of saturated (NH4)2 SO4 solution was added to 200 cubic centimeters
of blood, rich in carotin, from Holstein Cow No. 221. The globulins
were filtered off and the golden-yellow filtrate heated carefully in a
water bath to a temperature of 79° C. The coagulated proteins were
filtered off. The yellow filtrate was saturated with (NH4)2 SO4
in substance and let stand several hours. The golden-yellow pre-
cipitate was filtered off on a Buchner funnel. After allowing to suck

434 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

quite dry, all crystals of (NH4)2 SO4 were removed with a spatula
and the protein dissolved in about 75 cubic centimeters of water, in
which it was readily soluble. This deep yellow solution was neutral
in reaction. It contained a small amount of (NH4)2 SO4, the amount
of which was determined quantitatively. (NH4)2 SO4 was added
to a portion of this solution to bring the concentration up to a normal
solution. An equal volume of saturated (NHt)2 SO., solution was
added to the remainder, giving a solution between 3 and 4 normal and
one similar to the one whose coagulation temperature was observed
above. The coagulation temperature of the pigmented protein of these
two solutions was carefully studied. Both solutions were found to
contain a small amount of colorless protein which coagulated between
65° and 75° C. This was filtered off, the nitrate retaining its original
yellow color. This nitrate was then studied further.

The 3-4 normal or one half saturated (NH4)2 SO4 solution acted
in a manner identical with the solution whose study is recorded above.
The first opalescence appeared between 79° and 80° C., and complete
coagulation did not take place until the temperature was raised to
86° C.

It was not found possible to cause a clear coagulation of the
pigmented protein in the neutral normal solution of (NH4).. SO4 even
when the temperature was raised to 90° C. Opalescence began, how-
ever, between 82° and 82.5° C. Coagulation was readily obtained
when the solution was heated to 89° C. in the presence of a very little
HC1. (3 drops of a */2 normal HC1 solution were added to 10 c. c. of
solution.)

The large amount of evidence which has now been submitted in
regard to the transportation of the carotin in the blood serum will
justify but one conclusion, namely that the carotin exists in the blood
in conjugation with one of the proteins. The evidence will also justify
the conclusion that the protein with which the carotin is combined is
an albumin.

Summarizing the evidence, we have shown that the carotin car-
rying protein is precipitated from its solution in the serum or from its
aqueous solution, on complete saturation only with ammonium sulphate,
or by saturation with magnesium sulphate only in one per cent acetic
acid solution, or by heating its half saturated ammonium sulphate
solution to 86° C. ; the protein may also be coagulated by alcohol, or
by boilding its solution in the presence of acetic acid. As in all salting
out methods for the precipitation of proteins, the pigmented protein
is readily soluble in water after being thrown down by ammonium

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 435

sulphate, but is no longer soluble in water after being coagulated by
heat or alcohol.1 The protein is not coagulated by saturation of its
solution with sodium chloride.

Not very much can be said in regard to the character of the
combination of the carotin with the albumin The combination is evi-
dently a firm one, and is broken down only in the presence of alcohol
so that the pigment can be extracted by ether or petroleum ether. The
union is also broken down by dialysis, or at least rendered less firm,
but is not broken down when the protein is precipitated as a lead, silver
or mercury salt. It is interesting to notice that cholesterol and a phos-
phorus-containing body (probably lecithin) are mixed up in some way
in the combination of albumin and carotin, the liberated carotin from
the dialysed pigmented albumin yielding both cholesterol and phospho-
rus. We propose the name caroto-albumm or luteo-albumin for the
chromo-protein which transmits the carotin from the food to the milk
glands and fat synthesizing body cells of the cow.

The finding of this highly unsaturated hydrocarbon carotin pig-
ment in combination with one of the albumins of the blood, probably
similar to the combination of the haematin in the haemoglobin of the
red blood corpuscles, at once raises some impiortant questions as to
a possible physiological significance which might be attached to the
presence of the pigment. One can only suggest that like the haemo-
globin the luteo-albumin may be of importance in connection with the
oxygen supply of the body. This is not probable. The ease with which
the carotin is increased and decreased in the blood serum as shown by
the feeding experiments, seems to preclude the possibility of the
carotin being absolutely essential to the life of the cow.

A STUDY OF THE HIGH COLOR OF COLOSTRUM MILK FAT.

Considerable data was given in the preceding Bulletin of this series,
which showed that colostrum milk fat from all breeds of cows is
characterized by a very high content of carotin. In view of the results
obtained in the study of the pigment of the blood serum, it seemed
very probable that this interesting phenomenon was due to a great
accumulation of the carotin in the blood serum just previous to par-
turition. In order to obtain some definite experimental evidence in
support of this supposition, blood was drawn from the jugular vein of
a pure bred Jersey cow (No. 23), when she was dry and three days
previous to parturition. The amount of color in 10 c.c. of the blood

1. After standing a short time under the alcohol.

436 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

was determined in the manner previously described. Thirty days
after calving the blood was tested again. The color of the milk fat
was determined at this time also. The color of the blood serum of
another Jersey cow was determined at another time twelve hours
previous to the time it was judged she would give birth to a calf. The
data obtained in the two experiments are given in Table 7.

TABLE No. 7. — RELATION OF BLOOD SEBUM COLOB TO COLOB OF COLOSTRUM

MILK FAT.

Remarks.

Color of
serum

Color of
fat.

Yellow

Red

Yellow

Red

Three days before parturition (Jersey
Cow No. 23.) .....

42.0
50.0
29.0

0.8
2.0
0.2

64.0

1.8

Thirty days after parturition (Jersey
Cow No. 23.)..

Twelve hours before parturition (Jersey
Cow No. 2.)

It is readily seen that another explanation must be sought for
the high color of colostrum milk fat, other than an accumulation of
carotin in the blood. No doubt a certain amount of storing up of
carotin does occur if a cow is dry previous to parturition and the
serum is low in color at the time of drying up, it being supposed
of course that the food contains a plentiful supply of carotin. The
data presented in Table 7, when coupled with the data in Tables 3
and 4, show very clearly that under normal conditions the amount
of pigment carried by the serum does not exceed a certain maximum
point that appears to be practically the same for all cows, regardless
of breed. This is not abnormal when it is considered that the carotin
of the blood serum is in combination with a protein which no doubt
comprises a more or less constant proportion of the blood.

This result forces us to the same conclusion reached in connection
with the study of the physiological relation between food, blood serum
and milk fat carotin, namely that other factors, among which may
be the composition of the milk, must be taken into consideration in
explaining the pigmentation of milk fat. In the case of the high
color of colostrum milk, some facts stand out that seem to have a
special bearing upon the phenomenon. For instance it is a well-known
fact that the milk drawn for the first few days after parturition has

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 437

a very abnormal composition which is characterized by a very low
fat percentage and a very high protein content, the largest proportion
of which is albumin. Unpublished data are at hand which show a
composition of some colostrum milks of 1.3 per cent fat and over 4.5
per cent albumin. When it is considered that the carotin is carried
by the blood in combination with an albumin, and when it is also
taken into account that the source of the lactalbumin is undoubtedly,
at least partially, the serum albumin, a most plausible explanation
of the high color of colstrum milk fat is at once apparent. It is also
apparent that this high color will continue until the milk has reached
a normal composition or until the blood supply is depleted. That
this will occur regardless of breed is also readily explained since data
show that the maximum color of the blood serum does not materially
differ with different breeds.

DISCUSSION OF RESULTS.

The results of the foregoing studies in regard to the yellow
lipochrome of the blood serum of the cow do not require any extended
discussion. Following the interesting discoveries set forth in the
preceding papers in regard to the nature of the pigments of milk
fat and body fat and their simple physiological relation to the carotin
and xanthophylls of the food which the cow receives, it was not sur-
prising to find that the hitherto practically unknown lipochrome of
the blood serum of the same animal is also chiefly carotin and bears
the same relation to the food as the milk fat carotin. We are thus
able to establish the connecting link between the food carotin and
the carotin of the milk fat, body fat carotin, corpus luteum, etc., of
the cow.

One of the most important results of this study was the dis-
covery that the carotin is not transmitted to the milk glands and body
cell from the food by means of simple solution in the blood serum,
but is on the other hand carried through the body in combination
with an albumin of the serum.1 This fact is undoubtedly of con-
siderable importance in connection with the entire phenomenon of
the pigmentation of the milk fat. It may be safely predicted that
all the factors which surround this phenomenon are in some way
dependent upon this fact, and all these factors will not be known until
it is clearly understood what part this caroto- (or luteo-) albumin

1. Incidentally this discovery has resulted in the addition of a new
chromoprotein to the list of conjugated proteins. This is itself of consider-
able physiological interest.

MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

plays in the formation of the milk fat. The same holds true for the
body fat.

The readily demonstrated fact that the withdrawal of carotin
from the food results in a marked decrease in the color of the milk
fat being secreted or in the body fat being formed, shows that the
albumin which carries the carotin in the blood serum does play a
definite part in the formation of both milk fat l and body fat and no
doubt also in the formation of the corpus luteum.

The whole phenomenon offers many difficult and interesting prob-
lems for future study. Many of these when solved will undoubtedly
throw light upon the chemistry of the mechanism of milk secretion.

SUMMARY.

1. The well-known lipochrome of the blood serum of the cow is,
like the lipochrome of the milk fat, body fat, etc., of the same animal,
composed principally of carotin, the widespread hydrocarbon pigment
of plants. Associated in small quantity with the carotin of the serum,
probably dissolved in the fat of the blood, are one or more xanthophyll
pigments, which are always found in more or less variable quantities
associated with the carotin of plants.

2. The carotin and xanthophylls of the blood serum are derived
from the food and furnish the normal source for these pigments in
the milk fat and body fat, etc. A variation in the quantity of these
pigments in the food results in a corresponding variation in the amount
found in the blood serum and milk fat. Body fat formed during this
time will be also affected.

3. The carotin is carried by the blood serum in combination with
an albumin. The combination is a very firm one. Lecithin and cho-
lesterol are probably a part of the combination. We propose the name
caroto-albumin for this new chromo-protein of the blood.

4. The caroto-albumin of the blood serum of the cow is probably
of importance in the formation of the milk fat, body fat and corpus
luteum of the cow. It is doubtful if this new pigmented protein is
of importance in the oxygen respiration of the body.

5. The lactalbumin of cows' milk may, among other factors,
be related to the color of the milk fat. There appears to be a special
relation here in connection with the high color and the high albumin
content of colostrum milk.

1. The presence of both cholesterol and lecithin in the caroto-albumin may
explain the origin of these lipoids, as well as carotin in butter fat.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 439

BIBLIOGRAPHY.

1. Escher: Zeit. f. Physiol. Chem. 83, p. 198 (1913).

2. Halliburton: Jour. Physiol. 7, p. 324 (1886).

3. Krukenberg: Sitz. Ber. d. Jen. Gessel. f. Med. (1885).

4. Palmer and Eckles : Missouri Agri. Exp. Sta. Research Bulletins
Nos. 10, and n (1914), also Jour. Biol. Chem. 17, pp. 191, 211

5. Schunck: Proc. Roy. Soc. 72, p. 165 (1903).

6. Thudichum: Proc. Roy. Soc. 17, p. 253 (1869).

7. Willstatter and Escher: Zeit. f. Physiol. Chem. 76, pp. 214-225
(1912).

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 44!

B, CAROTIN AND XANTHOPHYLLS DURING DIC-ESTION

The establishment by us l of both a chemical and physiological
relation between the carotin and xanthophylls of plants and the yellow
lipochromes of the milk fat, body fat, blood serum and corpus luteum
of the cow has shown that it is the carotin that is by far the more
important in pigmentation of the animal body. It is a well-known
fact that xanthophylls are as abundantly and sometimes more abun-
dantly distributed in vegetable matter than carotin. The question
naturally arises then, why carotin is the pigment which is principally
taken up by the cow's body, and why the xanthophylls appear there
only in very small quantity. This seemed to us to be an important
physiological question.

It will readily be recognized that a question of this nature is not
easily answered. It may therefore be stated in advance that the results
of our studies were not as satisfactory as was anticipated. The
data are presented, however, for what value they may possess, since
opportunity was not presented for a further study of the question.
The data are of some value, at least, in that a number of facts are
presented which are sufficiently related to advance a fairly acceptable
theory in regard to the question.

METHODS OF STUDY

Several methods of study which did not appear to offer many
difficulties, seemed available, by which it was thought light could be
thrown on the question. One method was to study the action of the
various digestive fluids, both natural and artificial, on fresh crude
residues of the amorphous carotin and xanthophylls of plants. An-
other method was to study the nature of the unsaponifiable pigment
extracts at various places along the digestive tract of the cow. A
third method was closely related to the second and consisted in a study
of the unsaponifiable yellow pigments excreted under conditions where
unassimilated or undestroyed carotin and xanthophylls of the food
would be likely to appear unchanged in the feces. Lack of time made
a thorough study of all the methods impossible so that only the
significant features of the results of each study will be given.

1. Missouri Agricultural Experiment Station Research Bulletins Nos. 10
and 11, this Bulletin, p. 415 (1914); Jour. Biol. Chem. pp. 191, 211, 223 (1914).

442 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

THE ACTION OF DIGESTIVE JUICES

The following solutions were added to equal portions of carotin
and xanthophylls * in test tubes, and the tubes plugged with cotton
and set aside at 40° C. Observations for signs of decomposition were
made every day for five days.

Tube i. Five cc. of 0.25 per cent HC1 solution of pepsin.

Tube 2. Five cc. of 0.25 per cent HC1 solution of filtered gastric
juice from the fourth stomach of a Jersey cow.

Tube 3. Five cc. of 0.25 per cent Na2CO3 solution of trypsin.

Tube 4. Five cc. of 0.25 per cent Na2CO3 solution of extract
from pancreas of a Jersey cow.

Tube 5. Five cc. of 0.25 per cent Na2CO3 solution of trypsin
plus 5 cc. of fresh bile from a Jersey cow.

Tube 6. Five cc. of 0.25 per cent NaCO solution of pan-
creatic extract2 plus 5 cc. of fresh bile.

1. The carotin and xanthophylls were isolated as follows: 200 grams of
air-dried, powdered, green alfalfa leaves were shaken with three litres of 10
per cent alcoholic petroleum ether for two days, and then with 1 litre of
CS2, until the solvent had taken up as much pigment as possible. The carotin
and xanthophylls were isolated from each extract and combined. Each
solution was now concentrated to 50 cc. and divided into ten parts. These
were put into test tubes and the solvent driven off at a low temperature.
The residues were used for the studies reported above.

The carotin and xanthophylls were isolated from the alcoholic petroleum
ether extract as follows: The xanthophylls were removed from the extract
by shaking with an equal volume of 80 per cent alcohol. The carotin in
the petroleum ether was now freed from chlorophyll by shaking with an
excess of CaCO3, the solution was now evaporated into alcohol and trans-
ferred to ether by diluting with much water after the addition of ether.
The solution was freed from traces of chlorophyll that had escaped absorp-
tion by the CaCO3 by shaking with 30 per cent alcoholic potash. The ether
was then freed from alkali with distilled water. This ether solution of
carotin was combined with the similar solution obtained from the CS2, ex-
tract as described below. The 80 per cent alcohol, containing the xantho-
phylls, was partially freed from chlorophyll by shaking with moist animal
charcoal for one hour. The pigments were then transferred to ether, the
remainder of the chlorophyll being removed by 30 per cent alcoholic potash
as in the case of the carotin. The ether solution was then washed free
from alkali and added to the xanthophylls obtained from the CS2 extract
as described below.

The carotin and xanthophylls were isolated from the CS2 extract as follows:
The extract was concentrated into 95 per cent alcohol and after filtering was
saponified with KOH. The pigments were extracted from the soap with
ether. The ether was washed free from alkali and evaporated into alcohol.
The carotin and xanthophylls were separated by differentiation between
petroleum ether (b. p. 30-50 °C.) and the alcohol.

2. The pancreatic extract was prepared by extracting a freshly ground
cow's pancreas with 150 cc. of 30 per cent alcohol for 24 hours, straining
off the extract, filtering and neutralizing with KOH and 0.5 per cent
Na£CO3. To prepare the above solution an equal volume of 0.5 per cent
Na2CO3 was added.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 443

Tube 7. Five cc. of neutral solution of pancreatin.

Tube 8. Five cc. of neutral pancreatic extract.

Tube 9. Five cc. of neutral pancreatic solution plus 5 cc. of fresh
bile.

Tube 10. Five cc. of neutral pancreatic extract plus 5 cc. of bile.

The pepsin, trypsin and pancreatin were Merck's U. S. P. prepara-
tions.

A set of ten' tubes was also prepared containing equal portions of
the xanthophylls of yellow corn.1

The following results were obtained. Carotin: Bleaching oc-
curred only in the tubes containing neutral and alkaline pancreatic
extracts. In the same tubes plus bile there was no decoloration.
The bile had no solvent action on the carotin, which was in marked
contrast to the xanthophylls, as noted below. Xanthophylls : The pig-
ments in tubes I, 3 and 4 were largely decolorized at the end of the
second day, while those in tubes 2, 7 and 8 retained their color after
the fifth day. No observations could be made on the tubes containing
bile until the fifth day on account of the fact that the bile had com-
pletely dissolved the pigments as soon as it was added. The pigments
were examined by desiccating the contents of the tubes with plaster
of Paris and extracting with ether. Marked bleaching had occurred
in all the bile tubes. Corn Xanthophylls : There was marked destruc-
tive action of these pigments in all the tubes except those containing
bile. The corn xanthophylls, like the xanthophylls from the alfalfa,
were readily soluble in bile.

The most significant feature of the above results is the marked
difference in the solubility of carotin and xanthophylls in bile, the
surprising result being the very slight solubility of the carotin. This
was confirmed quantitatively using carotin from another source and
the bile from several different cows. The results are given in Table i.
The carotin used was a freshly prepared ether solution of carotin from
the carrot. Equal volumes of this solution were evaporated at a low
temperature and the residues treated with 10 cc. of bile from each of
four cows. After standing for several days with frequent shaking
the bile was filtered and 5 cc. of the filtrate desiccated with plaster
of Paris. This was extracted with ether until colorless. The extract
in each case was concentrated to a low volume, made up to 12.5 cc.
with absolute alcohol, and the color of the solution measured in the
Lovibond Tintometer.

1. This was the unsaponifiable pigment of the corn which was more
soluble in 80 per cent alcohol than in petroleum ether.

444 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12
TABLE No. 1. — THE SOLUBILITY OF CAROTIN IN BILE.

Experi-

Source

Carotin used

Carotin in bile

Blank*

ment

of

No.

bile

Yellow

Red

Yellow

Red

Yellow

Red

1.

Jersey

57.0

2.0

3.0

0.6

1.0

0.2

2.

Angus

57.0

2.0

9.0

0.8

1.0

0.2

3.

Holstein

57.0

2.0

10.0

0.9

1.0

0.2

4.

Holstein

57.0

2.0

10.5

1.0

1.0

0.2

*The blank is the amount of color extracted from 5 c.c, of bile alone,
after desiccation with plaster of Paris.

An interesting feature in the above table is the apparent greater
solubility of carotin in the bile of HJolstein cows, than in the bile of
Jersey cows. If this is confirmed by future study, considerable sig-
nificance could be attached to it in explaining, at least partly, the
differences between the two breeds in the amount of carotin that is
secreted in the milk fat.

CHARACTER OF THE PIGMENTS ALONG THE DIGESTIVE TRACT

The plan in this part of the study was to examine the pigments
which could be extracted from the material at various places along the
digestive tract of several cows. Material was obtained from one Hol-
stein cow and two Jersey cows at slaughtering, from each of the three
stomachs just before the food entered the next part of the digestive
tract, from three places in the small intestines, from the caecum, and
from the large intestine. One or two hundred grams of material were
either dried on the steam bath or desiccated with plaster of Paris, and
the resulting mass in either case extracted with CS2. The solubility,
spectroscopic, and adsorption properties of the extracted pigments were
carefully noted. The pigments were thus differentiated into carotin
and xanthophyll constituents as well as classified as belonging to
either of the two groups.

The results of the study were not satisfactory, in that there was
no uniformity among the several cows in regard to the character
of the pigments found at any particular place, although all the animals
were receiving a ration which should have furnished an excess of both
carotin and xanthophylls. The reason for this is not obvious. It
might be thought that the partial drying in some cases destroyed the
pigments. Possibly this occurred to some extent, but it would not

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 445

account for the lack of uniformity where this method of desiccation
was not employed.

No further discussion will be given this study. Mention has been
made of it merely because the method seems to be a valuable one,
and will warrant further application.

THE EXCRETED PIGMENTS

For this study the feces of a cow were examined, in a feeding-
experiment where the carotin and xanthophylls were furnished by
the feeding of carrots only. The balance of the ration was composed
of grain and timothy hay almost free from carotin and xanthophylls.

The method of demonstrating the character of the pigments
in the feces was to desiccate a quantity of fresh feces with plaster
of Paris and extract the mass with pure carbon bisulphide. The ex-
tract was concentrated and studied spectroscopically, and also by means
of a Tswett chromotogramm. The relative solubility properties of the
pigments thus found were studied, and also the spectroscopic properties
of the pigments thus separated.

In this way it was found that when the cow was receiving 50
pounds of carrots per day, both carotin and xanthophylls were abun-
dantly present in the feces. This continued for six days after the
carrots were withdrawn from the ration, although it was possible
to detect but little xanthophyll during this time.

DISCUSSION OF RESULTS

Combining the results of the above experiments, the appearance
of carotin in the cow's system when fed in excess may be explained
on the ground of its greater stability toward the digestive processes,
as shown by the digestion experiments, and the abundant appearance
of the pigment in the feces. The failure of the xanthophylls to
appear to any extent in the cow's system may be due similarly to the
fact that they are apparently more easily destroyed 1 during digestion.
Some of them that escape destruction are undoubtedly taken up by
the bile and thus enter the system through the portal circulation.
Some oxidation probably takes place in the liver. If fat is present
to any extent some of the xanthophylls will evidently be taken up and

1. Willstatter and Mieg. (Ann. d. Chem. 355, p. 1, 1907), state that
xanthophylls are very sensitive toward acids. This would lead one to
expect that they would be largely destroyed by the gastric juice. Our
results were contradictory in this respect. We found an artificial gastric
juice to destroy the xanthophylls but the natural gastric juice from the
fourth stomach of a cow apparently had no effect on them.

446 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

enter the blood dissolved in fat. In this connection it is of interest
to recall that we have shown 1 that there is evidence to indicate that
what xanthophylls can be found in the blood are present dissolved
in fat.

An additional possible explanation of this whole question should
not be overlooked, however, namely, that the difference in the pro-
portion of carotin and xanthophylls taken up by the cow's body
may be due entirely to the difference in chemical composition
between carotin and xanthophylls. Carotin is an unsaturated hydro-
carbon and is furthermore capable of combining with a protein of the
blood, as we have shown.2 The xanthophylls, on the other hand, are
carbon, hydrogen and oxygen compounds, in fact are chemically caro-
tin-dioxides. Although still unsaturated bodies, their slight difference
in composition from carotin, may prevent their combination with the
serum albumin, thus making it impossible for them to appear to any
extent in the blood and fatty formations of the cow's body. If fat
played a greater part in the food of the cow, the xanthophylls would
undoubtedly appear to a greater extent in the body of this animal.

SUMMARY

1. Carotin is assimilated from the food of the cow in preference
to xanthophylls partly because of its greater stability toward the
juices of the digestive tract. Xanthophylls are much more soluble
in bile than carotin,3 which probably accounts for their appearance
in the fat of the blood.

2. It is probable that carotin forms by far the greater part of
the lipochromes of the cow's body chiefly on account of its ability
to form a compound with one of the proteins of the blood. The
xanthophylls, being of different composition, probably are not capable
of forming such a compound.

BIBLIOGRAPHY

1. Fischer and Rose: Zeit. f. Physiol. Chem. 88, p. 331 (1913).

2. Palmer and Eckles: Missouri Agricultural Experiment Station
Research Bulletins Nos. 10 and n (1914); Jour. Biol. Chem.
17, pp. 191, 211, 223 (1914).

3. Willstatter and Mieg: Ann. d. Chem. 355, p. I (1907).

1. This Bulletin, page 422; Jour. Biol. Chem. 17, p. 211, 1914.

2. lUd.

3. A confirmation of the very slight solubility of carotin in bile is seen
in the recent finding of Fischer and Rose (Zeit f. Physiol. Chem. 88, p. 331,
1!913), that the gall stones of cows contain crystallizable carotin. No
xanthophylls were found.

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 447

C. THE PIGMENTS OF HUMAN MILK FAT

The discovery of the chemical and physiological relations of
the pigments of the fat of cow's milk to the carotin and xanthophylls
of plants naturally opens the question whether the pigments which
characterize the fat of other animals are of a similar character.
Opportunity was not afforded to study this question with any domestic
animals other than the cow. An opportunity, however,, was offered
to investigate the character of the pigments which sometimes give a
high color to the fat of human milk.

The methods used for studying the character of the pigments
were the microscopic ones used in the preceding studies. The adsorp-
tion properties were not studied, however, the demonstration being
confined to the observation of the absorption bands and the relative
solubility properties.

The fat from two samples of human milk from different sources
was used. Very little was known in regard to one of the samples,
it having been sent to the laboratory for analysis by a well-known
physician of the community. The other sample was taken by one of
us from a woman who had just given birth to a child, and represented
a portion of the milk of each day of the first few days of lactation.
Some further observations in regard to this sample will be reported
below.

Experiment No. i

This was the sample in regard to which very little was known,
with the exception that it was a bona fide sample of human milk.
The milk had a faint yellow tint. The volume of milk used was
approximately 125 c.cm. The milk contained about 3.5 per cent fat
and therefore yielded a little over 4 grams of fat. The fat was
obtained from the milk by precipitating it along with the proteins.
To do this the milk was acidified with acetic acid, a pinch of salt
added, and the milk brought to a boil. The precipitated proteins,
when filtered off, had a bright yellow color, due to occluded fat.
The fat was dissolved out with hot 95 per cent alcohol.

After concentrating the alcoholic extract, the fat was saponified
by adding a small piece of KOH and boiling for about one hour.
The pigment was readily extracted from the soap by ether, after
dilution with water. ' The golden-yellow ether solution was washed
with water and evaporated to dryness. The residue dissolved at once
in carbon bisulphide with a red-orange color and in this solution

448 MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

showed two beautiful absorption bands, and possibly a third. The
CS2 was carefully evaporated. A part of the residue which remained
was difficultly soluble in absolute alcohol, but readily dissolved when
a little petroleum ether was added. When differentiated between
petroleum ether and 80 per cent alcohol the combined pigment was
readily divided into two apparently equal proportions with perhaps
slightly more color in the petroleum ether layer.

The pigment of the petroleum ether layer gave a red-orange
carbon bisulphide solution showing two strong absorption bands and
a third faint one, the measurements of which are given in Table No. i
below. j

TABLE No. 1. — ABSORPTION BANDS OF CAROTIN AND XANTHOPHYLLS FROM
HUMAN MILK FAT.

Experi-
ment.

Measurements of absorption bands.

Carotin

Xanthophylls.

No. 1

I.
II.
III.

225—244
262—280
300—319

I. 234—253
II. 275—293
III. 320—

No. 2

I.
II.
III.

225—242
265—282

I. 232—252
II. 273—293
III. 312—330

The pigment of the alcoholic layer gave a yellow-orange, carbon
bisulphide solution showing two good absorption bands and end absorp-
tion, the measurements of which are given in Table i.

Experiment No. 2

As stated above, this sample of human milk was taken by one
of us and represented the milk of the first few days of lactation
including the colostrum milk. The milk itself was characterized
by a high yellow color and the fat which rose to the top of the
sample had a very deep yellow color. About 350 c.cm. of milk were
obtained. The fat percentage being between 5 and 6, nearly 20
grams of fat were yielded for the study of the pigments.

The fat was obtained from this sample of milk in a manner
very similar to that used in the preceding experiment. The proteins
and fat were precipitated together by adding a little salt and also con-

CAROTIN, THE PRINCIPAL YELLOW PIGMENT OF MILK FAT 449

siderable ammonium sulphate, acidifying with acetic acid and bringing
to a boil. The precipitate was filtered off on a Bikhner funnel. The
layer of protein and fat had a golden-yellow color. The fat was
extracted with hot alcohol and ether. The golden-colored extract
was evaporated to dryness and the fat dissolved away with ether.
Alcohol was added and also 5 grams of KOH and saponification of
the fat allowed to, proceed on the steam bath for one-half hour. The
pigment was extracted from the diluted soap with ether. After
thorough washing with distilled water, the ether was evaporated care-
fully to dryness. The residue had a deep red color. It dissolved
at once in petroleum ether (b.p. 3O°-5o°C.).

The pigment in this solution was now differentiated between the
petroleum ether and 80 per cent alcohol. In this way it was divided
into two portions which were about equal as far as could be detected
by the color of the two solutions, with perhaps slightly more color
in the 80 per cent alcohol.

The pigment in the petroleum ether layer gave a blood-red colored
carbon-bisulphide solution which showed two absorption bands and
considerable end absorption. The measurements of these bands are
given in Table i.

The pigment in the 80 per cent alcohol layer gave an orange-
colored carbon-bisulphide solution which showed three distinct absorp-
tion bands. The measurements of these bands are given in Table No. I.

DISCUSSION OF RESULTS

The results of the above experiments show very clearly that the
fat of human milk may be tinted with the same pigments found in
the fat of cow's milk. The relative proportion of carotin and
xanthophylls in human milk fat is much more nearly equal than in
the fat of cow's milk. This is not surprising when it is considered
that there is strong evidence that the xanthophylls are 'conveyed
through the body dissolved in fat, and when it is also considered
that fat plays a much greater part in human food than in the food
of the cow. i

An especially interesting fact brought out by these brief studies
is that colostrum milk fat of the human is characterized by a very
high color just as is the case with the fat of the colostrum milk
from cows. In the experiment here reported, one of us had occasion
to observe that after about ten days the milk fat from the same woman
was very much lighter in color than during the first few days of
lactation. The milk was also observed at intervals for a period of

45O MISSOURI AGRICULTURAL EXP. STA. RESEARCH BULLETIN NO. 12

several months. Considerable variation in the color of the fat was
noticed. Although it was not possible to accurately trace the cause
of this variation, as we did in the case of cows in an earlier paper
of this series, it was undoubtedly due to changes in diet.

In conclusion it may be stated that all students of human anatomy
are familiar with the fact that the fat on the human body is often
characterized by a marked yellow color. In view of the fact that the
pigments of the milk fat and body fat of the cow are identical, it
must therefore be concluded that the pigments of the milk fat and
body fat of humans are identical.

SUMMARY*

1. The fat of human milk may be tinted by carotin and xantho-
phylls, the pigments which characterize the fat of cows' milk. The
relative proportion of carotin to xanthophyll in human milk fat is
much more nearly equal than in the fat of cows' milk.

2. The colostrum fat of human milk is characterized by a very
high color as is the case with the fat of the colostrum milk of cows.

3. The pigment of human body fat is no doubt identical with the
pigment of human milk fat.

*See page 438 for summary of "The Yellow Pigment of Blood Serum."
See page 446 for summary of "Carotin and Xanthophylls During Digestion."

BIOGRAPHY

Leroy Sheldon Palmer was born in Rushville, Illinois, on March
23, 1887. He received his common school education in the public
schools of the city of St. Louis, Missouri, graduating from the Central
High School of that city in June, 1905. He entered the School of
Engineering of the University of Missouri in September, 1905, and
received the degree of B.S. in Chemical Engineering in June, 1909.
During the summer of 1909 he was Chemical Assistant for the United
States Bureau of Fisheries, the work being conducted under the direc-
tion of Dr. Chas. W. Greene at Columbia, Missouri. He was ap-
pointed Fellow in Chemistry at the University of Missouri for the
year 1909-1910, but resigned in October, 1909, to become Assistant
Chemist in the Co-operative Government Dairy Research Laboratory
of the University of Missouri, being appointed by the Dairy Division
of the Bureau of Animal Industry of the United States Department
of Agriculture. He pursued graduate work in the University of Mis-
souri during the years 1909-1910 and 1910-11, and received the degree
of M.A. in June, 1911. He was appointed by the Dairy Division of
the United States Department of Agriculture in October, 1911, as the
Government Representative in the Co-operative Dairy Research Lab-
oratory of the University of Missouri. He pursued work in the
Graduate School of the University of Missouri during 1911-1912 and
1912-13. He was appointed Assistant Professor of Dairy Chemistry
and Assistant Chemist to the Experiment Station in the Department of
Agricultural Chemistry by the University of Missouri in April, 1913.

451

LD 21-95m-7,'37

/

7*3

UNIVERSITY OF CALIFORNIA LIBRARY