It is with the deepest regret that the death
of the author of this thesis must be announced
with its issue.
Before the completion of this publication,
D R . LOOS accepted the appointment as Chemist
for the Copper Corporation of Chili, and sailed
for Chanaral in the latter part of June.

While

enroute he was stricken with yellow fever and
died on the seventeenth of July, nineteen hundred, in his twenty-fifth year.

A Study on Colophony Resin.

A DISSERTATION
Submitted in Partial Fulfilment of the Requirements for the
Degree of Doctor of Philosophy in the Faculty of Pure
Science of Columbia University

Hermann A. Loos, B.S., A.M.

EASTON, FA.:
THE CHEMICAL PUBLISHING COMPANY.
1900.

TABLE OF CONTENTS.
PAGE.

Acknowledgment
Introduction

4
5

*
P A R T I.—Colophony Resin.

Sources of Colophony
Anhydride Structure of Colophony Production of Anhydride by Melting Abietic Acid
Oxidation Products of Turpentine
Test for the Presence of Abietic Acid

6
6
10
11
11

P A R T II.—Preparation and Composition of Abietic Acid.
Historical Review of the Literature on Abietic Acid
Properties of Abietic Acid
Preparation of Pure Abietic Acid
Method of Trommsdorff, Maly, etc
" White Rosin "
" B l a c k Rosin "
" Gum Thus "
Fliickiger Method
" White Rosin "
" Black Rosiii "
'' Window Glass Rosin "
Direct Addition of Hydrochloric Acid
" White Rosin "
1
' Window Glass Rosin "
" Gum Thus "
Table of Analytical Results and Molecular Weight Determinations
Conclusion—Formula for Abietic Acid

12
18
20
21
21
23
24
25
25
26
27
28
28
29
30
31
31

P A R T III.—Salts of Abietic Acid.
Historical
Potassium Salt (Neutral)
Potassium Salt (Acid)
Sodium Salt (Neutral)
Sodium Salt (Acid;
Ammonium Salt
Other Salts
Effect of Hydrolization
Effect of Light
Resinate Colors
Conclusion
Biographical

33
34
36
40
40
41
42
42
42
43
45
46

ACKNOWLEDGMENT.
The author wishes to acknowledge his obligations to the officers of the chemical department who have aided him by
friendly advice and counsel, and more especially to Prof. C. E.
Pellew for the securing of samples and other necessary material, and to Mr. S. A. Tucker and Mr. M. T. Bogert for much
valuable assistance during the pursuit of this work.
H. A. I,.
INDUSTRIAL LABORATORY,
COLUMBIA

HAVKMKYER H A U , ,

UNIVERSITY.

INTRODUCTION.
In spite of the industrial importance of colophony resin, the
nature of its constituents is not thoroughly understood and
even the composition of its principal acid has long been a subject of controversy. Most of the work of a purely chemical
nature has been directed toward the determination of a formula
for abietic acid. The literature of the subject is full of contradictions and disparities of all sorts, and the following work
has been undertaken with the hope of throwing some light on
the disputed points and of determining the true composition of
abietic acid by means of products of the greatest attainable purity. Much work has already been done with rosin, of an analytical and a technical nature ; but this has been largely empirical, and it is probable that a better knowledge of the
chemistry of this substance, and of the character of the salts it
forms, will lead to many improvements in these directions.
The first part of this work is devoted to a discussion of the
sources and constituents of crude colophony, together with
some experimental work on the presence and formation of abietic anhydride, and the relation of the acid to the ordinary
oxidation products of turpentine. An historical review is then
given of the literature on abietic and other acids supposed pressent in colophony, and this is followed by a description of the
experimental work done in preparing pure abeitic acid and determining its formula. The last part of the thesis is devoted
to a description of the preparation and properties of some salts
of this acid and of the changes they were noticed to undergo as
the result of hydrolysis and the effect of sunlight.

PART L
COLOPHONY RESIN.
The resins form a physiological class of solid and semifluid
bodies, which result from constructive and destructive processes of plant life, and which show a great diversity in their
physical and chemical properties. The most important resin,
from an industrial point of view, is common rosin of colophony. This is obtained as a residue, from the natural resin
as exhuded from the tree, by removing the turpentine with a
steam distillation, and then keeping the product heated until
all water has been removed, which is indicated by a clarification of the melt. This process influences greatly the color
and general properties of the resulting rosin. Colophony is a
vitreous, brittle substance, breaking with a conchoidal fracture ;
it is insoluble in water, but soluble in most organic solvents.
It is almost odorless, and practically tasteless, and varies in
color from a pale yellow to a dark brown. The various grades
melt from 8o°-ioo° and the specific gravity varies from 1.045
to 1.085.

In the United States the principal rosin-producing trees1 are
Pinuspalustris (long-leaf pine), P. taeda (loblolly pine), and
P. cubensis (Cuban pine). The principal European sources2 are
P. laricio, P. pinaster, P. australis, P. taeda, sometimes also P.
maritima. In addition to these, other species sometimes furnish the commercial products sold as common rosin.
Colophony probably consists essentially of the anhydride of
abietic acid, although many authorities have denied this.
Fliickiger showed that masses of freshly exhuded rosin which
were clear and amorphous throughout gained in weight and
assumed a crystalline structure when exposed to water or moist
air. Other specimens, which were kept sealed and absolutely
dry for two years, underwent no change in weight or structure.
Maly obtained almost the calculated quantity of crystallizable
acid from amorphous rosin by the action of dilute alcohol. In
the following experiments by the author, the gain in weight
1
2

Report U. S. Dept. of Agric, Div. of Forestry, 1891 and 1892.
K. Dieterich : Anal. d. Harze, Balsame u. Gummiharze, p. n o .

7
approximates closely that required.by the hydrolysis of abietic
anhydride (C 88 H 65 0 ? ) to the free acid (C 19 H 28 O a ).
Some '' White Rosin'' was pulverized and dried over sulphuric acid in vacuum. Two portions were weighed in
small weighing bottles, covered with 60 per cent, alcohol, and
allowed to stand one week with frequent agitation. On carefully evaporating down, breaking up the mass, and drying alternately at ioo°, and over sulphuric acid, the following changes
in weight were noticed :
A.

Final weight
Original "
Gained "
Gained per cent.

1*4919
1.4512
0.0407
2.8

B.

1-9063
1-8475
0.0588
3.18

Abietic anhydride (molecular weight = 558) requires an increase of 3.226 per cent, in weight to convert it to the free
acid. Assuming 90 per cent, of abietic anhydride in this rosin,
would require an increase in weight of 2.90 per cent.
Lewkowitsch, 1 on the other hand, says: " T h e high acid
values, and especially the definite ether values, prove conclusively that colophony is not, as Maly maintains, an anhydride ;
viz., abietic anhydride, but consists chiefly of free acids and
small quantities of an anhydride.'' In the fourteen results he
refers to, the difference between acid value and saponification2
value,/. £., the ether value, varies from 16.4 to 36.11. Amsel3
finds the ether values in three determinations to be respectively 8, 7, and 7. Williams4 also finds results which are, in
general, lower than those mentioned by Iyewkowitsch. This
difference, it would seem, merely shows the presence of a small
quantity of free acid in colophony, and this is amply accounted
for by the partial hydrolysis the anhydride has undergone during the process of manufacture, or by the incomplete reversion
to anhydride during the melting down process. Moreover, in
1

" Com'l Anal, of Oils, Fats and Waxes,'* 2nd. ed., p. 235.
According to K. Dieterich, it is erroneous to speak of " saponification " values in connection with rosin, since no ethers are present.
3Ztschr. angew Chem., 1896, p. 430.
4
C. N., 58, p. 224.
8

8
the better grades of rosin in which the melting temperature
has been kept lower, the acid figures are found to be higher,
showing a greater proportion of free acid.
Perrenond, 1 V. Dietrich, 2 and Ducummun 3 hold that rosin
contains the free acid, with no anhydride whatever. They
base this view on the fact that crude rosin contains a smaller
percentage of carbon than abietic acid does, while, as anhydride, it should contain a larger percentage. Abietic acid
(C 19 H 28 0 2 ) contains 79.16 per cent, carbon, and the anhydride
81.7 per cent. Dietrich found 74 per cent, carbon in rosin,
and 77.5 per cent, in the product obtained on melting abietic
acid. An analysis of " White Rosin,'' made by the author,
resulted as follows:
0.2185 gram substance gave 0.6070 gram C0 2 and 0.1848
gram water, which corresponds to
Per cent.

c
H

75.75
9.40

It must not be forgotten, however, that other substances,
such as terpenes and their oxidation products, 4 are always
present in rosin, and these might tend to reduce the carbon
content of the whole. The first crystallizations of abietic acid
will, moreover, contain a smaller percentage of carbon than
than the pure product, and even a small quantity of impurity
will reduce the percentage of this constituent. A sample of
abietic acid, prepared by Maly's method, gave, after the second
crystallization, the following result on combustion:
0.2990 gram substance gave 0.5908 gram C0 2 and 0.1887
gram water, which equals
Per cent.

C

77-n

H

10.03

As to molten abietic acid, it is well known that volatile
*C. Z., 1885 ( I X ) , 1556, 1590.
Inaugural Dissertation, Bern, 1883.
3Inaugural Dissertation, Bern, 1885.
4
Terephthalic acid (C 8 H 6 0 4 ), an oxidation product of turpentine, contains 57.1 per cent carbon.
2

9
vapors are produced by its decomposition, consisting of terpenes, carbon dioxide, carbon monoxide, benzene hydrides,
etc. The mere fact that less carbon is present shows that a
decomposition has taken place, and this renders useless any
comparison of carbon content.
Perrenond and his co-workers assume that colophony consists
of crystals of abietic acid, imbedded in a resinous body. Crystals of this acid are very readily identified with the microscope ; polarized light will always detect them, even in the
presence of much amorphous matter. Wiesner 1 has shown
that most natural resins (7. e., such as are employed as removed from the tree, without melting or removing the turpentine by distillation) show crystals when examined in this way.
He covered the particles of rosin with turpentine, or some
solvent which dissolved the matrix to a greater extent than
the crystals. This same treatment was applied by the author
to all obtainable grades of commercial colophony, but in no
case could any crystals be found ; the smallest particles always
presented merely the chonchoidal fracture, which is characteristic of most resins. Only in the case of " Gum T h u s , " which
is a natural resin, could any crystals be detected. According
to Wiesner, the clarification of the crude rosin on melting during the preparation of colophony is due to a reversion of the
crystalline acid to anhydride. Lewkowitsch and Perrenond
state that the anhydride does not form on melting abietic acid,
and, in fact, the only definite anhydride that has been prepared artifically was made by Emmerling, 2 by heating the
crystalline acid with hydriodic acid to 145. °
Some experiments were made with the object of determining,
if possible, the changes produced in the crystalline acid by
heating at various temperatures above its melting-point. Some
pure acid was gently heated just above its melting-point. On
cooling slowly, large crystals were seen to form in the melt,
and when quite cold, a yellowish, opaque mass was left, which
was crystalline throughout, as shown by a microscopical ex1

Die technisch verwendeten Gummiarten Harze und Balsame,
p. 114.
2
Ber. d. chem. Ges., 12, 1441.

IO

animation. As the temperature of the melt increases, more
amorphous matter will be found on cooling. A product which
has been cooled from i75°-i8o° will leave a perfectly amorphous vitreous mass, which has all the appearance of the original rosin, excepting that it is more brittle. 1 But it was found
that that this mass could be crystallized directly from strong
alcohol, which can not be done with colophony. The latter
will leave an amorphous mass on evaporating a strong alcoholic
solution, and will only .crystallize after having been subjected to the hydrolyzing action of dilute alcohol, or strong alcohol with the addition of a mineral acid.
As already pointed out, analysis will give no indication of
the presence of the anhydride, but the product obtained, as
described, was decidedly different from the original colophony.
Its transparent appearance was probably due to a solution of
the free acid in some of the terpenes produced by the decomposition. Kraemer and Spilke 2 have shown that on heating
rosin under pressure, the terpene C18H38 is formed according
to the equation —
C 19 H 28 0 2 = C18H28 + C0 2 .
Some pure, dry abietic acid was heated for two hours at
180V 1 95° ; water and other vapors were given off. On
cooling, a brown vitreous mass was left. A saturated solution
of this product in alcohol did not crystallize, even on standing
several days. On the other hand, portions treated with dilute
alcohol or hydrochloric acid gave crystals of abietic acid in a
few hours. This product resembled, in all respects, the colophony from which the abietic acid had been originally made.
This would indicate that some anhydride is formed by a continued heating at this temperature.
Abietic acid has often been considered an oxidation-product
of turpentine. Terpenes contain no oxygen ; all resins do ;
hence it has been thought that the resins are oxidation-products of the terpenes. But in no case has it been possible to
1

When a shallow layer of melted abietic acid is allowed to cool,
it will contract and suddenly crack up into small pieces, with a distinctly audible decrepitation.
2
Ber. d. chem. Ges., 1899, pp. 2952 and 3614.

II

produce any of the known resins by oxidation of a balsam or
essential oil, nor has the reverse process been possible ; z. e.,
the reduction of a resin to a knowTn terpene. Resinous bodies
of some kind do, however, form by oxidation of essential oils
and there seems to be a loss of hydrogen in this process.
Heldt 1 claimed that resins are formed by a simple oxidation of
ethereal oils, there being a direct addition of oxygen and loss of
hydrogen in the form of water. Perrenond and Ducummun
deny the identity of abietic acid with the resin produced by
oxidizing terpentine. They examined turpentine that had
been treated with ozone and some that had resinified on standing several years, but in no case could any color reaction be obtained for abietic acid. The only product that could be identified was terephthalic acid.
Some experiments made by the author gave no better results. An old sample of turpentine, which was known to have
stood for five or six years, was distilled with steam ; the residue was washed with alkali and these washings were precipitated with acid. No indication of abietic acid could be obtained. Samples of terpentine were oxidized with hydrogen
peroxide and sodium peroxide. Especially with the latter reagent, the oxidation was very violent. The excess of turpentine was now distilled off with steam and the residue examined.
No crystalline product could be prepared from it, nor did any
of the tests for abietic acid indicate its presence. One of the
tes/cs employed was based on the fact that abietic acid will
give colored lakes with an aniline dye ; i. e., a double salt of
metal and a basic color seems to form (see p. 43 of this thesis).
The solubility of this substance in benzol serves as an indication of its presence. The above oxidation-products were
treated with dilute alkali; methyl violet was added and then
zinc sulphate. A small precipitate formed, but this, when
dried and treated with benzol, gave no colored solution.
In the preparation of abietic acid from rosin a yield of 80-95
percent, of crude acid is obtained. 1 More of the acid will
1

Ann. Chem. (Liebig), 63, 18.
In one case Mach obtained 98 per cent, of crude acid from a sample of French colophony.
2

12

slowly crystallize out from the mother-liquor, during several
months' standing. ! No other acids, crystalline or amorphous,
have been separated from the mother-liquors. The amorphous
acids which are often described in colophony consist mainly of
abietic acid which is held back in the mother-liquors. A small
quantity of turpentine and its oxidation-products, some of
which are acid in character, are also present. Colophony also
contains terpenes and other decomposition-products produced
by the breaking up of the abietic acid at the temperature required to drive off the last traces of water in the process of its
manufacture.
PART II.
ABIETIC ACID : HISTORY, PREPARATION, AND
COMPOSITION.
Previous to 1825, Ries, 2 in Vienna, obtained some crystalline bodies by acting on white pitch with acids. No analyses
nor attempts at identification were made. In 1826 Baup3 separated a crystalline body from the resin of P. adzes a id named
it '' acide abietique'' ; a similar body from Bordeaux resin
(probably P. maritima) he named " acide pinique" (apparently
identical with pimaric acid of Laurent). Unverdorben 4 showed
the acid character of colophony and described several of its
salts. On precipitating an alkali salt with acid he observed an
increase in weight which he ascribed to the formation of a
hydrate. 5 He described three acid bodies in ordinary resins
which he named respectively a-, /?-, and ^-resin. In colophony he describes seven constituents, the principal ones being
' * Sylvinsaure'' (ar-resin), a crystalline body which he declares
is identical with that obtained by Ries.
An amorphous
1

If the mother-liquor is stirred with a glass rod and then allowed
to stand several weeks the crystals will separate on the sides where
touched by the rod, very much like a precipitate of magnesium phosphate will.
2
Jahrbuch d. Polytech. Inst, zu Wien, I., p. 435 ; Repert. d. organ.
Chem. v. Fechner, p. 1291.
3
Ann. chim. phys., 31, 108.
4
P o g g . Ann., 7» 3*i '> 8, 405.
5
Ibid., 11, 27, 230,393.

13

body (/?-resin) which he named " Pininsaure" and a deeplycolored substance, *' colopholsaure'', which is the result of
superheating pininsaure.'' He describes other constituents of
rosin and salts of the acids but gives no analyses. Trommsdorff* prepared the crystalline acid by treating rosin with 60
per cent, alcohol, and on the strength of his analysis concludes
that it is an oxidation-product, not of temperature but of a
radical C10H1B, having one hydrogen atom less. He gives this
acid the formula C40H60O4,2 based on the composition of the
copper salt. Rose"' agrees with this formula but adds that
colophony is an oxidation-product of turpentine.
Caillot4 found an acid in Strasburg turpentine which he
called ' * abiesinsaure'' and which was subsequently believed to
be a mixture of sylvinic acid and pimaric acid. Laurent 5 believed this acid to be different from sylvic and to be identical
with the one he obtained from Colophonium du Bordeaux (P.
maritimd)and he accordingly gave it the name " Pimarsaure."
This body had, however, the same composition as sylvic and
pinic acid ; on distillation it formed pyromaric acid. Strecker 6
subsequently suggests the identity of pyromaric, pimaric, sylvic and, in fact, all the acids obtained from American colophony. Caillot7 again took up the work and agrees with Laurent on a melting-point of 1250 for pimaric acid. He found it
to be laevo-rotatory and when heated to ioo° it changed to a
dextro-rotatory acid, which melted at 208°.
Siewert8 declares that abietic acid is isomeric with pimaric
but differs from it in being slightly more soluble in alcohol.
He succeeded in obtaining a crystalline potassium salt of the
composition C80H29KO2, 3CaoH30Oa.
1

A n n . Chetn. (Liebig), 13, 169.
2 In t h i s case and wherever necessary, formulas and analyses based
on t h e old carbon equivalent have been changed and recalculated.
3 Ann. Chem. (Liebig), 13, 174.
4
J. de P h . et. de Ch., 16, 436 ; Bull. Soc. Chim. (Paris), 21 2 , 387 ;
and Ber. d. chem. Ges., 7,484.
5
Ann. Chem. (Ltebig), 34, 272; Ann. chim. p h y s . , 72, 383.
6
Ann. Chem. P h a r m . , 68, 338; Ann. Chem. (Liebig), 150, 131.
7
Ber. d. chem. Ges., 7, 484.
8
J s b . d. Chem., I, p. 572; and X I I , p. 508.

H
Maly1 prepared an acid from American colophony by the
method of Trommsdorff and Unverdorben. He could get no
crystalline potassium salt and hence claimed that Siewert's acid
was different from his, which he called abietic acid. He retains the name '' sylvinsaure'' for the acid produced from colophony by the action of hydrochloric or other mineral acids and
shows that no ether is formed when dry hydrochloric acid gas
acts on a solution of abietic acid in strong alcohol. Maly considers abietic acid to be dibasic and gives it a new formula ;
viz., C44H640B. He prepared an ethyl and a glyceryl ester and
made a series of interesting hydrocarbons by the action of phosphorus pentachloride. By the action of sodium amalgam he
obtained a body which he called hydro-abietic acid but which
shows no definite distinction from abietic acid itself.
Duvernoy 2 and Strecker 3 take exception to Maly's formula
for abietic acid and declare Siewert's formula, C20HS0Oa, to be
the correct one. Strecker declares abietic and silvic acids to be
identical. Duvernoy obtained a crystalline lead salt from this
acid which removed a supposed distinction between it and
pimaric acid. He also obtained a crystalline sodium salt, and
an acid, and a neutral salt of potassium, but could get no
crystalline salt of ammonium with abietic acid, and considered
this a distinction from pimaric acid. Maly4 maintains the
identity of pimaric and abietic acids, and finally admits that
while the latter has the formula C44H6406 there is also a small
quantity of the C20H30O2 acid present in colophony.
Fliickiger 5 led dry hydrochloric acid gas into an alcoholic
solution of rosin and obtained a body which was shown to be
identical with abietic acid, although Maly had claimed that
this would give an entirely different acid. Emmerling 0 pre1

M a l y : Jsb. d Chem., 1861, p. 389; J. p r a k t . Chem., 86, i n ;
Ibid., 92, 1 ; Ibid., 96, 145 ; Ann. Chem. (Liebig), 149, 244; Ibid., 161,
115 ; Chem. Centrbl., 3, 59 ; Sitzungsber. d. Wien. Acad., 44 ( I I ) , 128;
48 ( I I ) , 355; 5 0 ( H ) .
2
I n a u g u r a l Diss., Tubingen, 1865 ; Ann. Chem. ( L i e b i g ) , 148,143.
3 Ibid., 150, 131.
4
Ibid., 149, 244.
5
J. p r a k t . Chem., 131, 235.
6
Ber. d. chem. Ges., 12, 1441.

15

pared the acid by various methods and concluded that Fliickiger's acid is identical with abietic ; his analytical results on
the free acid and the salt agree with Maly's'formula, C 44 H 64 0 5 .
Emmerling prepared a bromine derivative and an impure acetyl
derivative, studied the action of hydriodic acid, of oxidizing
agents, of fused caustic potash and prepared a series of
hydrocarbons by a zinc chloride distillation. Kelbe, 1 while at
work on a process of purifying rosin oil, succeeded in preparing
abietic acid from this substance ; he also retains Maly's formula. Liebermann 5 admits the identity of abietic and sylvic
acids and declares them to have the same composition as
pimaric acid, C,0H,0Oa. He obtained a series of hydrocarbons by acting on these acids with hydriodic acid and red phosphorus and found that they rotated the plane of polarization in
opposite directions. Haller* confirmed this work, gave some
analyses of abietic acid, and determined its rotation to be
[«]D = - 5 3 ° .
Valente 4 separated a crystalline body from colophony,
which, while still in a crude condition, gave a melting-point of
165 0 and turned the plane of polarization to the right. He
made it from Bordeaux resin and although he calls it sylvic
acid he evidently had in hand the pimaric acid of Laurent. His
analyses agree with the formula C20H30O2 and he concludes
that Maly's acid does not exist and that Liebermann based his
examination on a mixture and not a pure substance. Perrenond5 and his co-workers, V. Dietrich6 and Ducummun, 7 next
took up this work. Dietrich examined American and European resins and found abietic acid in the former and pimaric
acid in the latter. These acids he found to agree in composition but to differ in melting-point and rotatory power (see table,
1

Ber. d. chem. Ges., 11, 2174; Ibid., 13, 888.
• Ibid., 171 1885.
3 Ibid., 18, 2165.
4
Atti della Reale Ace. dei Iyincei, I, 13, 1884 ; Ber. d. chem. Ges.,
18 (3)> 190.
5
Chem. Ztg., 9> 159°6
Etude compared sur l'acide abietique et l'acide pimarique(i883),
Bern.
7
fitude sur les acides crystallisables des Abitin£es (1885), Bern.

i6

p. 19) ; both are said to have the composition of anoxy-cymol,
C10H14O, although the analyses given differ considerably from
the theoretical requirements (see table, p. 18).
Based on the composition of the ammonium salt the formula
of pimaric acid is given as 4(C 10 H 14 O). Abietic acid, according to these authors, gives no crystalline ammonium salt but
merely a gelatinous soap. Ducummun found abietic acid in
the root resin of P. sylvestris and pimaric acid in the trunk resin
of the same tree.
Vesterberg 1 finally showed pimaric acid to be distinctly different from abietic acid. He obtained from French galipot resin,
three isomeric acids; the first is dextro-pimaric, which melts at
2 i o ° - 2 i i ° and hasarotation of [ « ] D = + 5 9 ° - 2 ; the second,
laevo-pimaric, melts at i4o°-i5o°, and has a rotation of [/*]D
— —272 0 . His analyses agree well with the formula C20H30O3
(see table, p. 18).
Mach, a pupil of Maly, gives a new formula for abietic acid ;
viz., C19H2eOa. He obtained a product of great purity from
several species of American colophony and his analyses average
78.97 per cent, carbon and 9.84 percent, hydrogen; the formula he proposes calls for 79.16 per cent, carbon and 9.73 per
cent, hydrogen. The distinctive feature of Mach's work, however, is the rational step he took in making a series of molecular weight determinations by Beckmann's method. His results vary from 269 to 344. This at once excludes Maly's formula, which requires a molecular weight of 672. Mach's formula calls for a molecular weight of 288, and seven determinations made with dilute solutions (which, as is well known,
give more accurate results in this method) vary from 269 to
298, and average at 285.5. I* is clearly seen that theseresults,
as well as the analytical data, agree very closely with the formula he proposes.
Mach's formula has, however, not been universally accepted.
In view of the great disparity in the work done on this substance, as pointed out in the preceding pages, the following
work was undertaken with the hope of throwing some light on
these much disputed points.
1

Ber. d. chem. Ges., 18, 3331; Ibid., 19, 2167; Ibid., 20, 3248.

17
TABLK

OF PUBLISHED ANALYSES OF ABIOTIC AND
PIMARIC A C I D S .

Abietic
c.

Acid,

Per cent.

Trommsdorff
Iyiebig
Rose

Siewert

Maly

Kraut 1
Emmerling

Kelbe
Haller
V. Dietrich
Ducummun

Mach

78.21
78.54
78.65
77.12
75.19
76.81
76.63
76.21
76.93
78.19
78.92
79.12
79.50
79.00
79-93
78.69
78.66
78.53
78.54
78.37
78.74
78.62
78.42
78.64
78.85
78.65
79.32
79.28
79-47
79.15
79.26
79.34
79.36
78.89
78.94
78.88
78.92

H.
Per cent.

9.86
9-79
9.82
9.42
9.46
9-36
9.27
943
9.64
9.95
9-97
9.68
9.85
9.63
9.63
10.00

9.97
9.75
9.72
9.68
9.68
9.59
9.64
10.00

9.78
9.62
9.63
9.72
9.56
9.58
9.62
9.64
9-65
9.81
9.86
9.81
9.84

i8
Abietic

Acid
c.

{continued).
H.
Per cent.

Per cent.

9.85
9.82
9.86
9.84
9.84
9.89
9.92
9.84
9-83
9-74

78-93
78.90
78.95
79.07
79.02
78.91
78.78
78.83
78.99
78.62
T h e o r y for C „ H , s O , 79.16
Pimaric
c.
L,aurent
Haller
Duvernoy
V. Dietrich
Ducummun

Per cent.

78.18

78.47
78.62
79.23
79.12
79.46

9-73
Acid.
H.
Per cent.

9.74
9.4I
9.62
9.95
IO.OO

9.63
9.68
9.60
952
9.96
IO.64

79.45
79.60
79.02
79.28
79.29
79.26
79.33

IO.I6
IO.OO

T h e o r y for C, 0 H S0 O, 79-47

994

Vesterberg
Mach

CHEMICAL AND PHYSICAI, PROPERTIES OF ABIETIC ACID.

Pure abietic acid is a white crystalline body insoluble in
water and soluble in most organic solvents. A table of solubilities is given on p. 22. The crystalline structure has been
studied by Siewert,1 Lang,2 Wulf,3 Fock,4 and Graber.5
x

3".

2

Zeit. fur die gesam. naturwissenschaften, Giebel u, Heintz, 14*

J. prakt. Chem., 96» 164.
3 Ber. d. chem. Ges., i3» 888.
4
Groth's Zeit. fur Krystallographie, 1883, VII, p. 58.
5
Monatsh. Chem., 15, 628.

19

Abietic acid is optically active both in the crystalline form
and in solution. Some of the observed rotations for abietic
acid are :
V a l e n t e : [ « ] D = ±37.87
Haller
= —53 0
Mach
= —56.97 (Abietic of a m. p. =
i53°-i54°.
= —67.34 1 (After an additional
= —66.94 j
crystallization.)
= —66.66 (After three additional
crystallizations.)
= —69.96 (French
colophony,
14th crystallization.)
The melting-point of abietic acid was found in the following
work to be i 5 9 ° - i 6 i ° , but was never sharp. Mach found that
all his products melted sharply at i53°-i54°. The meltingpoints recorded by different observers vary greatly, as is seen
in the table here given ; those of pimaric acid are also given
for comparison :
Abietic Acid.
[Sylvic acid.]

Siewert
Maly
Fliickig
Duvernoy
Kelbe
Bmmerling
Haller
Dietrich
Mach
L,aurent
Caillot
Siewert
Maly
Duvernoy
Haller
Dietrich
Vesterberg

1620
165°
f I20° (Soft)
j 135 0 (Melted)
1290
165°
139°
( 145°
(Soft)
( I 6 I ° - 2 ° (Melted)

165 0
i53°-i54°
Pimaric Acid.

"5°

{125 0
( 208 0 (Dextropimaric)
155°
165°
149°
( I20° (Soft)
0
\ 1490 (Melted)
148
f 2 i o ° - 2 i i ° (Dextropimaric)
\ i4o°-i50° (I<aevopimaric)

20

Preparation of Pure Abietic Acid and Determination
of its Formula.
There have been proposed six formulae for abietic acid.
The analyses on which they are based show, as a rule, great
variation. Rose's determinations of carbon content vary from
75.19 to 78.92 per cent.1 Machpointed outthat impure abietic
acid will give a lower carbon content and will approach in this
respect the results obtained by earlier observers.
In the following work abietic acid was prepared by different
methods from three typical grades of American colophony and
also from a natural resin from Pinuspalustris (the long-leaf
pine) one of the chief rosin-producing trees in this country.
The crystalline products were shown to be identical by microscopical examination, ultimate analysis, and determination of
molecular weight. The analytical results agree closely with
those obtained by Siewert1 and Mach and confirm in all respects the formula proposed by the latter ; /. e.y C19H3802.
Abietic acid will not crystallize directly from a solution of
rosin ; on evaporating a solution in alcohol, ether, chloroform,
benzol, or petroleum ether, an amorphous mass remains. Rosin
consists essentially of abietic anhydride, and this must first be
hydrolyzed to the free acid before crystallization can take
place. On saponifying rosin with alkali and then precipitating
with a mineral acid, free abietic acid is obtained but is contaminated with much amorphous matter and is in a condition
from which it can be crystallized only with the greatest difficulty. This is due to the fact that terpenes and other byproducts originally present are dissolved to some extent in the
rosin soap and are subsequently precipitated mechanically with
the abietic acid.
There are two general methods of preparation. The first is
the method of Trommsdorff in which the hydrolysis is pro1

Although Siewert proposes the formula C2oH3002 his analytical
results agree more closely with C19H2802.
Four analyses given by Siewert average 79.14 per cent, carbon and 9.70
per cent, hydrogen.
C19H2802 requires 79.16 per cent, carbon and 9.73 per cent, hydrogen.
C20H30O8 requires 79.47 per cent, carbon and 9.94 per cent, hydrogen.

21

duced by the aid of dilute alcohol, which brings the rosin particles into intimate contact with water. In the second, the
method of Fliickiger, the anhydride is changed to free acid by
passing a stream of dry hydrochloric acid gas into an alcoholic
solution of rosin. This method may be modified by adding
small quantities of mineral acid directly to such solution.
Mechanical devices such as pressure, suction filters, and centrifugals are indispensable for removing amorphous matter from
the crystalline products. As solvents for recrystallization,
methyl or ethyl alcohol, ether, petroleum ether, acetic acid,
or benzol may be used.
Method of Trommsdorff, Maly, Etc,
Portions of ioo grams " white rosin" were pulverized and
digested repeatedly with 70 per cent, alcohol. A brown, viscous layer settled on the bottom of the receptacle, and above
this a white flocculent layer slowly formed. After standing
two to three weeks the entire mass became granular. Frequent
trituration of this mass in a mortar was found to hasten the
separation. It was now filtered with suction or pressure and
dried below ioo°. This substance, when examined under the
microscope with polarized light, appears distinctly crystalline
in character but much amorphous matter is still present. The
melting-point of a portion dried in vacuum over sulphuric acid
was I O 2 ° - I O 5 ° . If the original product was not finely ground
the melting-point was always 3°-4° lower.
The product was
again digested with dilute alcohol and filtered by means of
suction. It was now dissolved in hot 95.5 per cent, alcohol,
diluted with water to beginning of turbidity, and allowed to
cool ; an oily layer separated which became granular in about
twenty-four hours. This was filtered and again crystallized ;
only after this process had been repeated four or five times
could good crystals be obtained, though they were still deeply
colored. The amorphous residue contaminated all crystallizations with the greatest pertinacity. To determine the best
medium for removing it, the combined mother-liquors were diluted, the separated substance dried over sulphuric acid, and
its solubility then compared with that of abietic acid. Satura-

22

ted solutions of abietic and the amorphous residue at 23 ° C.
were found to contain the following percentages of their weight
of substance :
Abietic acid.
Per cent.

Absolute alcohol
95.5 per cent, alcohol
70

"

"

Amorphous residue.
Per cent.

18.7
10.23

23.7
16.6
15.9

1.2

Equal parts 95.5 per cent
alchol and ether
37.9
38.2
Ether
37.8
39-7
Methyl alcohol
8.5
35.5
22.5
Benzol
61.1
In accordance with these results, dilute alcohol was used for
the first crystallizations (i. e., a hot solution of the substance
in strong alcohol was diluted), and methyl alcohol for the final
ones, since this medium was found to give much better crystals. If about one per cent, of water by volume is added to a
methyl alcohol solution of abietic acid, crystals from 6-8 mm.
across can sometimes be obtained.
Glacial acetic acid has
been suggested as a crystallizing medium. 1 This was found,
however, to give good crystals only when but little amorphous
matter was present; i. e., with a product that had already been
purified. In this case large crystals may be obtained ; if allowed to stand in the mother-liquor for a few days, however,
they will again go into solution probably with the formation
of some acetyl derivative.
The addition of water will again
precipitate abietic acid.
The above product, purified as described, gave a higher
melting-point with each crystallization, until after the fourteenth it melted at i54°-i6o°. It was tinted slightly yellow.
An additional crystallization gave no improvement in color nor
rise in melting-point.
Two analyses of this product gave the following results :
0.2007 gram substance gave 0.1776 gram H 2 0 . and 0.5802
gram C0 2 , corresponding to
Per cent.

Carbon
Hydrogen
1

Kelbe : Ber. d. chem. Ges., 13, 888.

78.83
9.83

23

o.2344 gram substance gave 0.2045 gram H 2 0, and 0.6782
gram C0 2 , corresponding to
Per cent.

Carbon
78.91
Hydrogen
9.70
A molecular weight determination made by the boiling-point
method, in a Beckmann apparatus, gave the following result
(X = 2530) :
0.2914 gram substance, dissolved in 31.252 grams acetic
acid, raised the boiling-point o°.o86, which corresponds to a
molecular weight of 274.
A determination by the freezing-point method, in a Raoult
apparatus, gave the following result (K = 3880) :
0.6916 gram substance, in 21.783 grams glacial acetic acid,
lowered the freezing-point o°.406, corresponding to ju = 303.
This method of preparation was laborious and time-consuming and required large quantities of solvents. The meltingpoint, as in the products from other methods of preparation,
was not sharp, probably due to decomposition beginning in the
vicinity of the melting-point. This product was also found to
have retained a slight quantity of impurity by the benzol and
hydrochloric acid test described on page 28.
The same process was now applied to some " black rosin."
This is a crude product of very dark color ; it is prepared from
the poorest quality of turpentine and consists mainly of the
residues from the turpentine stills. It is sometimes sold as
American galipot rosin but must not be confused with the
French galipot, from which it is entirely distinct. The difficulties of this method of preparation were simply manifolded
with this rosin. Fifty grams were pulverized and digested with
200 cubic centimeters of 70 per cent, alcohol. After standing
ten days the mass was pressed, washed, and again treated with
alcohol of the same strength.
The product was now crystallized fifteen times, 1 first from ethyl and finally from methyl
1

In making these crystallizations the product was dissolved in
strong alcohol with heat, diluted with cold water just to turbidity,
cooled rapidly with occasional shaking, allowed to stand one hour,
filtered with suction until almost dry, and then redissolved in strong
alcohol. Each crystallization was also washed with small quantities
of 70 per cent, alcohol.

24

alcohol. It now melted between i49°-i54°, was crystalline in
character but yellowish in color. The hydrochloric acid and
benzol test showed this product to be impure. No further
crystallizations were made.
0.2227 gram of the substance, dried over sulphuric acid in
vacuum, gave on analysis 0.1977 gram H 2 0 , and 0.6426 gram
CO,, corresponding to
Per cent.

Hydrogen
9.86
Carbon
78.66
While this method is very troublesome when applied to manufactured rosins (such as have been subjected to distillation and subsequent melting down), it was found very satisfactory for the treatment of a natural rosin. For this purpose,
what is commonly known as ' 'Gum Thus, "or crude turpentine, !
was used. This is an exhudation of the Pinus palustris: it is
a soft, yellowish, opaque resin of decided turpentine odor, containing 17 per cent, of volatile oil.2 The opacity is due to the
presence of some free abietic acid (produced from the anh}'dride by the action of moist air); a strong alcohol solution will
deposit this constituent in the form of a crystalline layer on
standing over night. The resin was pulverized, and 50 grams
were shaken up with 250 cubic centimeters of 70 per cent, alcohol. After standing four days, the mass was pressed, washed
twice with dilute alcohol, and recrystallized. Four additional
crystallizations were made. The product now melted at 159 0 161 °, and gave the following results on analysis : 0.2884 gram
substance gave 0.2562 gram water and 0.8975 gram COa, which
corresponds to
Per cent.

Hydrogen
Carbon

9.87
78.98

Determination of molecular weight by the boiling-point
method:
0.3495 gram substance dissolved in 18.070 grams absolute
1
Other synonyms are: Thus Atnericanum, Terebinthina, Common
Frankincense.
* U. S. Dispensatory, 1894, p. 1356.

25

ether caused a lowering of o°.i50 in the boiling-point, corresponding to a molecular weight of 272.
Use of Acids to Aid Hydrolisation.
Fliickiger's method consists in passing hydrochloric acid gas
into an alcoholic solution of rosin ; he dissolved the rosin in
seven times its weight of 70 per cent, alcohol, and then passed
the dry gas. Maly thought that the action of hydrochloric
acid formed a new acid distinct from abietic, which he called
'' Sylvinolsaure.'' Mach, however, showed that the product obtained in this way is identical with abietic acid. He passed dry
hydrochloric acid gas into a very concentrated solution of rosin
in 95.5 per cent, alcohol until small crystals began to form ;
the solution now solidified to a mass of crystals, and, after four
recrystallizations, gave a white product, which melted sharply
at i53°-i54°. This process was repeated with solutious of
rosin, both in dilute and in strong alcohol; they were found
to darken rapidly and to develop considerable heat; at the
same time there was formed a dark viscous substance, which
was very difficult to remove in subsequent crystallizations. To
avoid this, a very thorough cooling with a freezing-mixture
was necessary. The anise-like odor mentioned by Fltickiger
was plainly noticeable during this operation.
One hundred grams '' white rosin'' were dissolved in 700
cubic centimeters of 70 per cent, alcohol with heat; the solution was filtered, cooled, and placed in a freezing-mixture, and
dry 1 hydrochloric acid gas passed through. A thick oily layer
separated in about twenty minutes, and gradually became crystalline ; the crystals, however, were enclosed in a brown amorphous mass. The gas was allowed to pass for about one hour.
The precipitate was filtered with suction, washed with alcohol
of increasing dilution, and finally with sufficient water to remove all hydrochloric acid. The crude product was dissolved
in 100 cubic centimeters of 95 per cent, alcohol, water added
with stirring just to the point of cloudiness, and allowed to
cool. In the next crystallization, less water was used for di1

1

tion.

The gas is dried in order to avoid further dilution of the rosin solu-

26
lution, and the third was made from methyl alcohol. The
product was slightly yellowish in color, had a melting-point of
i58°-i59° } and gave the following results on analysis :
0.2261 gram substance gave 0.2022 gram water and 0.6546
gram C0 2 , corresponding to
Per cent.

Hydrogen
9.94
Carbon
78.96
Three further crystallizations from methyl alcohol gave good
white crystals having the same melting-point:
0.2590 gram of this substance gave 0.2309 gram water and
0.7592 gram C0 2 , accordingly,
Per cent.

Hydrogen
9.91
Carbon
79.03
A second analysis gave for 0.2078 gram substance, 0.1847
gram H 2 0 and 0.6011 gram C0 2 .
Per cent.

Hydrogen
Carbon

9.87
78.89

The following molecular weight determination was made by
the boiling-point method : 0.3152 gram substance dissolved in
absolute ether (17.221 grams) caused an elevation of o°. 132 in
the boiling-point; hence the molecular weight equals 293
(K = 2110). By the freezing-point method, 0.4011 gram substance dissolved in 28.645 grams glacial acetic acid caused a
lowering of o°.i33 in the freezing-point, which is equivalent to
ix = 266 (K = 3880).
Fifty grams "Black Rosin" were dissolved in 100 cubic
centimeters of 95.5 per cent, alcohol, diluted with 20 cubic
centimeters of water; the solution was filtered, and hydrochloric acid gas passed, until the beginning of crystallization.
Violent stirring now completed the crystallization. The precipitate, which was dark in color, was filtered by means of suction, washed, and eight recrystallizations made. The final product had a melting-point of i54°-i55°, was slightly yellowish
in color, and was shown to be impure by the benzol and hydrochloric acid test:

27

0.1982 gram substance gave, on analysis, o. 1761 gram H 2 0
and 0.5728 gram C0 2 .
Per cent.

Hydrogen
9.87
Carbon
78.81
Fifty grams '' Window Glass Rosin'' were dissolved in 150
cubic centimeters 95 per cent, alcohol and dry hydrochloric
acid gas passed in. Much heat was evolved, and large quantities of the usual brown by-products were formed. The mass
was filtered and washed as usual. Five crystallizations were
made from methyl alcohol and one from glacial acetic acid.
The latter was washed with alcohol and the crystals dried over
sulphuric acid in vacuum. The melting-point was
0.2239 gram of this substance gave, on analysis, o. 1944 gram
H 2 0 and 0.6483 gram C0 2 , corresponding to
Per cent.

Hydrogen
9.65
Carbon
78.98
0.3400 gram of the same substance, dissolved in 33.332 grams
glacial acetic acid, depressed the freezing-point o°.i38, corresponding to a molecular weight of 287 (K = 3880).
This method gives a pure product, with fewer crystallizations, than the method first employed, and in a much shorter
time.
However, not only is the objectionable by-product
mentioned above very difiicult to separate, but the product is
generally lumpy, and the original rosin is, therefore, not completely hydrolyzed, but is left partly in an amorphous condition. It seemed necessary then to limit the amount of hydrochloric acid introduced, and to use some means of mechanical agitation to prevent the formation of lumps. To this end,
some hydrochloric acid, 1.20 sp. gr., diluted with twice its
volume of alcohol, was added to a solution of rosin in 95 per
cent, alcohol contained in a bottle. The quantity of hydrochloric acid added was such that the water it contained would
dilute all the alcohol present to about 75 per cent. The bottle
was nowr wired to the chuck of a lathe and agitated for one hour.
The abietic acid separate^ in the form of a finely divided crystalline powder. It was subsequently found that a very much

28

smaller quantity of hydrochloric acid would suffice to give a
good precipitation, and, of course, in this case the obnoxious
by-products were smaller in quantity.
Other mineral acids may be used instead of hydrochloric; *
but this was found preferable to sulphuric or nitric. The chief
difficulty in the preparation of pure abietic acid is not in the
hydrolization of the anhydride, but in the separation of the
amorphous by-products (terpenes and their oxidation products).
The addition of a small quantity of hydrochloric acid greatly
diminishes the solubility of abieticacidin alcohol, while that of
the amorphous residue is, if anything, increased. It may be
that pure abietic acid is not affected by hydrochloric acid in
alcoholic solution. This is the basis of Twitchell's method
for the determination of rosin. To a solution of purified abietic
acid in alcohol was slowly added half its volume of hydrochloric acid, which caused an immediate separation. This was allowed to stand two months, when the crystals were filtered
and washed ; there was no change in the appearance of the
crystals nor in melting-point. The mother-liquor was yellow,
but this color was due to the action of hydrochloric on a small
quantity of impurity still remaining.
The color produced in this way was used as a test for the
presence of impurities in specimens of abietic acid. A small
quantity of the latter is dissolved in 5 cubic centimeters of
benzol in a test-tube.
Five cubic centimeters hydrochloric
acid, 1.20 sp. gr., are added and thoroughly shaken together.
Any impurity present will give a brown to a yellow color to
the lower layer ; this is a very delicate test.
One hundred and fifty grams ' * White Rosin '' were dissolved
in 300 cubic centimeters hot 95 per cent, alcohol. The solution
was filtered and allowed to cool. It was now violently agitated by means of a mechanical stirrer, while a mixture of 20
cubic centimeters of hydrochloric acid and 40 cubic centimeters of alcohol was slowly added. The solution grew cloudy
and small oily drops separated. In about forty-five minutes,
1

Mach used sulphuric acid in a similar manner for a preparation
whose analysis approximated closer than any other to the theoretical
requirement for C19H2802.

29
the solution had crystallized to a thick, creamy consistency,
the agitation preventing the formation of any lumps and consequent occlusion of mother-liquor. The mass was filtered
by means of suction, washed with very little dilute alcohol and
much water. The dried product was plainly crystalline in
character, of a light straw-color and weighed 112 grams. The
melting-point at this stage was i40°-i43°. After two crystallizations from ethyl alcohol and three more from methyl alcohol, the melting-point was i 5 9 ° - i 6 i ° . In the second and
third crystallization, 2 per cent, by volume of hydrochloric
acid was added to the hot solution.
0.2421 gram of the final product, dried over sulphuric acid
in vacuum gave 0.2097 gram water and 0.7024 gram C0 2 .
per cent.

Hydrogen
9.63
Carbon
79. n
0.2143 gram of the same substance gave 0.1908 gram water
and 0.6209 gram C0 2 , corresponding to
Per cent.

Hydrogen
Carbon

9.89
79.02

Two molecular weight determinations were made by the
freezing-point method. 0.3750 gram substance, dissolved in
17.976 grams glacial acetic acid, depressed the freezing-point
o°.288 ; JA = 281.
0.6303 gram substance in 22.243 grams glacial acetic acid
gave a depression of o°.364 ; /*= 302.
300 grams ' * Window Glass Rosin '' were dissolved in 300
cubic centimeters of hot 95 per cent, alcohol. The solution was
filtered, cooled, and then violently agitated, while a mixture
of 35 cubic centimeters of hydrochloric acid and 70 cubic centimeters of alcohol was slowly added. In about forty minutes,
the solution had solidified to a crystalline mass. This was filtered and washed, as in the above preparations. This product,
when dry, weighed 245 grams, and had a melting-point of
i44°-i48° ; it was, however, somewhat darker in color than
the preceding one. It was now recrystallized from alcohol:
a second and third crystallization were made with the aid of

3°
hydrochloric acid and three more with methyl alcohol. This
product was white and gave only a trace of color when tested
for impurities as above described.
0.2762 gram of the dry substance gave, on analysis, 0.2435
gram water and 0.8021 gram C0 2 .
Per cent.

Hydrogen
9.80
Carbon
79-21
0.2177 gram gave 0.1902 gram water and 0.6313 gram CO,.
Per cent.

Hydrogen
Carbon

9.71
79-°5

Four molecular weight determinations were made as follows,
using glacial acetic acid and the freezing point method:
Gram substance.

Grams solvent.

Depression.

Molecular weight.

O.4307
23.0675
0.262 0
277
2
0.3241
o-754
0.206 0
294
0.4698
22.687
0.275 0
292
0.2695
24.302
0.156 0
276
To 50 grams * 'Gum Thus'' dissolved in 100 cubic centimeters
of alcohol were added 10 cubic centimeters of hydrochloric acid
in 20 cubic centimeters of alcohol. The formation of crystals
from this resin was very much slower in this process. After
filtering and washing as usual, five crystallizations were made
from methyl alcohol.
0.1946 gram of the dry substance gave 0.1716 gram H a O
and 0.5643 gram CO a ; hence,
Per cent.

Hydrogen
Carbon

9.77
79.09

A molecular weight determination, made by the boiling-point
method, resulted as follows: 0.7772 gram of this substance
raised the boiling-point of 17.843 grams glacial acetic acid
0.357°; /* = 3°9In the accompanying table will be found a list of the analytical results and of the molecular weights obtained.
The
average for twelve determinations of carbon and hydrogen

3*

(omitting the analyses from the preparations of ' 'Black Rosin,''
which were shown to contain much impurity by the color test)
gives 9.80 per cent, hydrogen and 79.01 per cent, carbon. The
various products obtained show sufficient agreement to prove
their identity. The last method of preparation employed
seems to give products of the greatest purity, and their analysis gives a somewhat higher carbon content. The molecular
weight determinations show considerable variation.
TABLE OF ANALYTICAL RESULTS AND MOLECULAR WEIGHT
DETERMINATIONS.
Preparation.

Hydrogen.

1. Maly's method, '' White Rosin "
9.83
2. "
"
"
"
9.70
3. <«
"
. " Black R o s i n "
9.86
4. "
"
"GumThus"
9.87
5. Fliickiger's method, " White Rosin," 3d
crystallization
9.94
6. Fliickiger's method, " White Rosin,'' 6th
crystallization
9.91
7. Fliickiger'smethod,'' White Rosin,'' 6th
crystallization
9.87
8. Fliickiger's method, " Black Rosin " . . . 9.87
9.
"
"
" Window Glass Rosin "
9.65
10. Direct addition of HC1 to solution of
"White R o s i n "
9.63
11. Direct addition of HC1 to solution of
"White Rosin"
9.89
12. Direct addition of HC1 to solution of
" Window Glass Rosin "
9.80
13. Direct addition of HC1 to solution of
" Window Glass Rosin "
9.71
14. Direct addition of HC1 to solution of
"GumThus"
9.77
Average excluding Nos. 3 and 8, which
were shown to contain inpurities... 9.80
Theory for C 19 H 28 0 2
9.73

Carbon.

Molecular
weight.

78.83
78.91
78.66
78.98

274, 303
274,303

78.96

293,266

78.03

293, 266

78.89
78.81

293, 266

78.98

287

79.11

281,302

79.02

281,302
r 277, 294
I 292, 276
r 277, 294
1 292, 276

79.21
79.05

272

79.09

309

79.01
79.16

286
288

Hence the formulas proposed by Unverdorben (C40H60O4,
JX — 604), by Trommsdorff and Rose (C40H64O4, M = 608),
and by Maly C44H6406, J* = 672) are untenable. The other
proposed formulas require the following composition and
molecular weights:

32

The formula of Perrenond, Dietrich and Ducummun,
C10H14O, requires 80.0 per cent, carbon and 9.33 per cent, hydrogen and a molecular weight of 300 ; i. e., if abietic were
2 (C 1 0 H u O).
The formula of Siewert and Strecker, Ca0H30O8 (molecular
weight = 302), requires 79.47 per cent, carbon and 9.94 per
cent, hydrogen.
The only other possible formula (excepting that of Mach)
whose composition approaches closely to the result of the
above analyses, and which is within the limits of the molecular
weight determinations made, is C18H2ROa, requiring 78.83 per
cent, carbon and 9.49 per cent, hydrogen.
Finally, Mach's formula requires a molecular weight 0^288
and a composition of 79.16 per cent, carbon and 9.73 per cent,
hydrogen.
The results obtained in this work are accordingly expressed
most exactly by the formula C 19 H„O a .

33
PART I I I .
SALTS OF ABIETIC ACID.
Many salts of this acid have been described, and their metallic content determined ; the results obtained show a great disparity, but have, nevertheless, been frequently used as a proof
or even the basis of some of the proposed formulas. The original acid worked with was, in general, impure, and the products obtained were merely amorphous mixtures.
Metallic
abietates are made by double decomposition of metallic salts
and alkali abietates. The latter are not readily obtained in
a pure condition ; only two observers have described them as
crystalline. Maly described them as amorphous, and considered that this distinguished them from the salts of pimaric acid.
Now, as will be seen below, neutral salts, even when pure, will,
in the presence of water, quickly hydrolyze to an acid salt and
free alkali, the former precipitating from solution because of
its insolubility. To keep a neutral salt in solution, an excess
of alkali is necessary, which, in a double decomposition, will
precipitate metallic hydroxide ; on the other hand, too large
an excess will salt out alkali abietate. These are not conditions
that will give pure salts.
It might be thought that recrystallization would remove the
contaminants, but here again we meet with an hydrolysis of
even a metallic salt, probably to acid salts, or free acid and
metallic hydroxide. This process seems to be hastened by the
action of sunlight (see page 42). Even Mach, whose salts
were made with the greatest care, and whose analytical results
approach closely to the theoretical percentages, admits that
they cannot be accepted as a support for his formula.
The description of a few salts of abietic acid, given below,
will give some idea of the difficulties met with in their preparation and purification. It is probable, however, that conditions may be found under which pure crystalline salts may be
obtained; this work is not by any means concluded, and a
more complete study of these salts will be undertaken with the
view of obtaining further evidence for the formula proposed,

34

and of ascertaining the changes they undergo in the ordinary
analytical and technical processes.
POTASSIUM SAI/fS.

Neutral Salt, C^H^KO,
This salt is prepared with great difficulty, because of the
facility with which it will revert to the acid salt. Its preparation requires digestion of the acid with potassium hydrate, or
carbonate, for a long time, with complete absence of water.
Mach used the carbonate, and obtained a salt which contained
12.21 per cent, potassium. 1 A neutral salt was made by this
method, as follows:
Twenty grams pure abietic acid and 5 grams pure dry potassium carbonate were covered with 200 cubic centimeters of absolute alcohol and 100 cubic centimeters of ether, in a flask
connected with a return condenser, and boiled for 210 hours. 2
An amorphous precipitate formed at first, but in the course of
boiling, it went almost completely into solution. The product
was finally filtered hot to remove the slight excess of potassium carbonate, was concentrated slightly by boiling, and allowed to cool. An amorphous mass separated, which, after
filtration, was crystallized from alcohol and ether. The crystals were almost white, needle shaped, and melted at 143 0 .
The potassium content was determined by burning in a platinum crucible with concentrated sulphuric acid, to which was
occasionally added a few drops of nitric acid. Great care was
necessary in this operation to avoid loss by foaming over.
When the carbon had been completely oxidized, more sulphuric
and nitric acids were added to reconvert any reduced sulphide to sulphate ; this was evaporated, and the residue care1
This is the average of three determinations which gave respectively 12.09, 12.25, ai*d 12-30 per cent, potassium ; these results he
obtained by weighing the potassium carbonate remaining after a combustion for carbon and hydrogen.
2 This preparation and others like it were heated on an air-bath made
of an asbestos-lined box containing four 20 candle-power incandescent
lamps. The top of the condenser was closed with a calcium chloride
tube. In this way the preparation boiled fifteen hours per day for over
two weeks without requiring any attention.

35
fully ignited. 0.4227 gram substance treated in this way gave
o. 1102 gram potassium sulphate ; this equals 11.70 per cent,
potassium. The theoretical requirement for C 19 H 27 K0 9 is 11.96.
A better product was obtained by using potassium hydrate
instead of the carbonate.
Eight grams potassium hydrate were dissolved in 150 cubic
centimeters of absolute alcohol. Fifty grams pure abietic acid,
dissolved in 100 cubic centimeters of ether, were now added,
and the mixture boiled under a return condenser. The solution was clear at first, but in a few hours a thick white curd
was found, which again went almost completely into solution
in the course of four or five days ; on cooling at this time, a
copious mass of crystals formed. The boiling was continued
for 210 hours. A portion was removed, heated with a little
more alcohol and ether to complete the solution, filtered, and
allowed to crystallize. A second crystallization gave a product
which was almost white ; it melted at 149 0 , and was neutral in
action toward phenolphthalein.
0.3427 gram of this substance, treated with sulphuric acid,
as described, gave 0.0921 gram potassium sulphate, which is
equal to 12.07 P e r c e n t - potassium. (Theory: 11.96 per cent.)
Mach's salt contains a small quantity of impurity, as shown
by some analyses he made for carbon, hydrogen, and potassium. He ascribes this to an absorption of oxygen during the
long process of boiling ; however, three analyses give for oxygen (by difference) respectively, 10.41, 10.77, a n ^ IO -59 VeT
cent. ; while C19Ha,KO requires 10.82 per cent. This evidently
does not indicate an absorption of oxygen. His potassium
determinations gave results somewhat higher than the theoretical; but the method he employed, viz,, weighing the residual potassium carbonate after a combustion for carbon and
hydrogen, is not without objection.
The crystals are white, can be recrystallized from alcohol
and ether, and are more soluble in hot than in cold water ; the
solubility in water is greatly increased by the addition of a
small quantity of alkali, while a large quantity has a " saltingout 'y effect, and will cause a precipitation of the salt.
The solubility of the neutral salt was found to be as follows:

36
Solvent.

Ether
Absolute alcohol
Water

Percentage
content of saturated
solutions at 200.

No. of parts
solvent required.

1.8
5.8
about

55.6
17.2
30

When freshly prepared, neutral potassium abietate is boiled
with water and a clear solution is at first obtained, which will
rapidly give a copious flocculent precipitate of an acid salt, and
liberate free alkali. This can be shown by the addition of a
few drops of phenolphthalein. The reaction seems to take
place as follows:
n + 1 (C 19 H 27 K0 3 ) + nH 2 0 = C 19 H iT KO sl nC 19 H„O f + nKOH
The salt produced will be described more fully in the next section, on the acid salt. On keeping the neutral salt from three
to four months, it was found to have dimished in solubility and
changed in other properties, no doubt due to a spontaneous
decomposition.
Acid Potassium Salt, C^H^KO^, jCl9H^O^
Abietic acid has a tendency to form an acid salt under almost any conditions ; more strictly speaking, this is a neutral
salt with several molecules of acid of crystallization. The salt
that has been obtained consists of three parts of abietic to one
of neutral salt; it was first described by Siewert, and subsequently by Mach. It is stable and very insoluble, and is not
affected by the presence of some free alkali.
Following are the conditions under which this salt was observed to form :
1. A slight excess of potassium hydrate acting on a saturated
solution of abietic acid in alcohol (95 per cent, or less).
2. A slight excess of potassium carbonate boiled with an alcoholic solution of abietic acid for several hours.
3. From the freshly-prepared neutral salt, by boiling its
aqueous solution.
4. From the freshly-prepared neutral salt, by allowing its
aqueous solution to stand twenty-four hours or longer.
5. From a solution of abietic acid in aqueous potassium hydrate, by carefully regulating the quantities used.

37
i. To a hot concentrated solution of abietic acid in 95.5 per
cent, alcohol was added a slight excess of an alcoholic solution
of caustic potash. On cooling, acicular crystals separated,
which, when filtered and dried, were analyzed.
0.4807 gram substance gave 0.359 gram K 2 S0 4 = 3.35 per
cent, potassium.
Theory for C 19 H 27 K0 2 , 3C19H2802 = 3.28
per cent, potassium.
A portion of the preparation for neutral salt was removed
after boiling one day (see page 34). Needle-shaped crystals
were obtained on cooling, which, on recrystallizing, filtering,
and drying, were analyzed.
0.4190 gram substance gave 0.305 gram K a S0 4 = 3.27 per
cent, potassium. The melting-point was found to be 197° l
2. A saturated solution of abietic acid in 95.5 per cent, alcohol, was boiled one hour with an excess of dry, powdered
potassium carbonate. The solution was filtered hot and a few
cubic centimeters of ether added until cloudy. On cooling,
good crystals were obtained.
0.6022 gram substance gave 0.0481 gram K 2 S0 4 = 3.58 per
cent, potassium.
In one case, potassium carbonate was shaken in a large testtube with an excess of a dilute alcoholic solution of abietic
acid. On standing one day, large crystals of abietic acid separated on the sides ; after the third day, the tube was almost
filled with long needles of the acid salt.
3. Some freshly-prepared neutral potassium abietate was
dissolved in cold water, and a few drops of phenolphthalein
added — no color. On boiling violently, a copious, white,
curdy precipitate formed and the solution reddened. The
precpitiate was filtered and recrystallized twice from alcohol
and ether. The melting-point was 187 0 .
0.6282 gram substance gave 0.0511 gram K 2 S0 4 = 3.65 per
cent, potassium. Theory for C 19 H 27 K0 2 , 3C19H2802 = 3.28
per cent, potassium. This is evidently the acid salt, and its
formation may be expressed as follows :
4C 1 .H„KO, + 3 H , 0 = C 19 H a ,K0 2 , 3 l 9 CH ! e 0 5 + 3 K O H .
1

Mach's acid salt melted at 183°.

38

All attempts to determine the amount of alkali liberated
were futile, because of the uncertain action of phenolphthalein
as indicator. The salt thus obtained is but very slightly soluble in water, and the presence of a small quantity of free alkali seems to exert no influence. This reaction probably expresses what takes place on dissolving and diluting a rosin soap,
which is principally in the condition of a neutral salt, owing to
the excess of free alkali used, and the concentrated condition
of the solution from which it was made. It has been shown1
that the detergent value of an ordinary soap is due to the liberation of free alkali by the hydrolytic action of water, and that
this action is increased with the quantity of water used, and
decreased with the presence of free alkali. With the potassium abietate, this is evidenced by the separation of the acid
salt because of its great insolubility.
This reaction would, it seems, also explain the different results obtained in determining saponification figures for some
resins, according to whether hot or cold solutions are used. It
is known 2 that these figures are always highest when obtained
in a cold solution, and vice versa. In a hot solution, the tendency is to form the acid salt, and consequently less alkali is
necessary to reach the neutral point with the indicators ordinarily used.
Siewert's acid salt corresponds to this one in composition.
On boiling this salt, he says it decomposes to the free acid
(sylvic acid) and the neutral salt which goes into solution.
No such change was observed with the acid salts of abietic
acid, obtained by any of the methods described.
Duvernoy obtained crystalline neutral and acid salts of pimaric acid. The acid salt has the composition C20H39KO2,
2C20H30O2. On boiling a neutral salt of this acid, he says he
obtained the acid salt, but gives no data in support of this
contention.
1
Alder Wright, Jour. Soc. Arts, 23, 1885; Thorpe's " Dictionary of
Chemistry," III, p. 412. A tallow-resin soap which he examined liberated 1.7 per cent, of the alkali, originally present, on adding 10 parts
water, and 5.6 per cent, on adding 150 parts.
2
K. Dietrich: Ch. Rev. f. Fett u. Harz, Ind. 1897, p. 1, et seq.

39
4. A solution of freshly-prepared acid salt, to which a few
drops of phenolphthalein had been added, was allowed to stand
twenty-four hours. The solution had become red, and a precipitate had formed, which was filtered, dried, and analyzed.
0.3794 gram substance gave 0.0353 gram K 2 S0 4 = 4.17 per
cent, potassium. The conversion to acid salt was evidently
incomplete.
Ten grams pure abietic acid were covered with 1.5 grams
potassium hydrate dissolved in 400 cubic centimeters of water.
This was agitated and then allowed to stand without disturbance for three or four hours at 5o°-6o°. A scum of white
needle-shaped crystals, imbedded in much gelatinous matter,
had collected on the top of the solution and on the sides of the
beaker. The solution was now boiled, upon which a dense
precipitate formed, which had a silky texture, and could be
drawn out into long threads—the same appearance as the
crystallized acid salt when boiled with water. On filtering,
recrystallizing from alcohol and ether, and drying, the following analytical results were obtained :
0.5946 gram substance gave 0.0409 gram K 3 S0 4 , = 3.09 per
cent, potassium.
*
Properties of the Acid Salt.—This salt, when dry, is a white
felt-like mass, consisting of fine acicular crystals, which melt
at 197°.1 As already stated, it is very insoluble, and not affected by a slight excess of alkali. The following solubility
determinations have been made :
Solvent.

No. of parts solvent
required for
solution.

Percentage content
of s1 saturated solution at 20°.

Ether
2.7
Equal parts alcohol
and ether
9-5
Alcohol, absolute
31-3
3-8
"
95.5
70.0
3-7
Benzol
4.08
Petroleum
ether
(hot)

37-03

0.98

IO.5
3-2
26.3I
27.O3
24.5I
102.00

Mach found t h e melting-point of his salt to be 183°.

4o
SODIUM S A W S .

A crystalline neutral sodium salt can be obtained which is
similar to the potassium salt.
Three and a half grams sodium hydrate were heated with 75
cubic centimeters absolute alcohol, and to this was added a
solution of 25 grams abietic acid in 50 cubic centimeters ether
and 25 cubic centimeter absolute alcohol. The mixture was
heated in a flask connected with a return-condenser sealed
by a calcium chloride tube, for 210 hours. A copious precipitate formed, which only partially dissolved during boiling.
On cooling, acicular crystals formed which were yellowish
in color.
0.3972 gram of the dry substance gave 0.0981 gram Na 2 S0 4 ,
corresponding to 8.00 per cent, sodium. Theory for C 19 H 27 Na0 2
= 7.42 per cent, sodium.
After three crystallizations the salt was white in color. Two
analyses resulted as follows :
Substance.

Na 2 S0 4 .

Sodium.

Gram.

Gram.

Per cent.

O.5339
O.4962

O.0893
O.0826

5-42
5.39

This corresponds neither to C 19 H 27 Na0 2 nor to C 19 H 27 Na0 2 ,
3C 19 H 28 0 2 , which requires 1.96 per cent Na. It is evident that
here again an hydrolysis has taken place and the product is
probably a mixture of the neutral salt, and one or more acid
salts.
On boiling a solution of the neutral sodium salt containing a
slight excess of alkali a precipitate is obtained.
0.1750 gram of the dried salt gave 0.0191 gram Na 2 S0 4 =
3.54 per cent, sodium.
A sodium salt, C 19 H 27 Na0 2 ,C 19 H 28 0 2 , corresponding to the
ammonium salt obtained by Mach (C 19 H 27 NH 4 0 2 ,C 19 H 28 0 2 )
requires 3.85 per cent, sodium.
The neutral salt is crystalline ; that obtained after the third
crystallization melts at 216°. The following solubility determinations were made with this salt.

4i

Solvent.

Percentage content
of a saturated 0 solution at 20 .

Ether (hot)

0.22

"

20°

0.2I

Petroleum ether (hot) 0.21
Alcohol, absolute
2.00
95-5
16.86
Water (freshly prepared salt)
11.69
Water (salt that had
been standing three
months
10.16
Ammonium

Parts solvent
required for
solution.
456.OO
480.OO
480.OO
50.00

5.93
8.55
9.84

Salt.

Mach obtained a crystalline ammonium salt and removed
thereby the supposed distinction between abietic and pimaric
acids. He prepared the salt by adding ammonium carbonate
to an alcoholic solution of abietic acid, and also by passing a
mixture of ammonia gas and carbon dioxide into such solution.
His analytical results approach closely to the requirements for
C 19 H„NH 4 O s ,C 19 H 28 O s . A crystalline ammonium salt was obtained by the following method :
A solution of abietic acid in alcohol and ether was boiled
with a return-condenser while a slow current of ammonia gas
(dried by passing over soda-lime) was passed through. The
gas was passed into the boiling solution for one day ; the product was allowed to cool and the gas passed the same length of
time through the cold solution. A crystalline salt was obtained, which, after the second crystallization, had but a slight
tinge of color, and was neutral in reaction and in smell. The
addition of potassium hydrate to the salt liberated ammonia.
On adding water to an alcoholic solution of this salt and allowing to stand several days, crystals were found to separate.
On examining such crystals they were found to be triclinic
plates, to rotate the plane of polarization, and to melt at 1570 ;
accordingly they are free abietic acid. A complete hydrolysis
evidently takes place with the separation of ammonia and free
abietic acid. The dry salt, when kept any length of time, will
evolve ammonia to a noticeable extent; this decomposition
seems to be more rapid in direct sunlight.

42

Several attempts were made to determine the ammonia content by treating with a known excess of standard acid and
titrating back with standard alkali. The results were variable
and gave no decisive clue to the composition of the salt. One
of the determinations made is here given :
i. 0.6392 gram substance was treated with 10 cubic centimeters tenth-normal hydrochloric acid and 50 cubic centimeters
water. After digesting one hour, the product was filtered,
washed, and titrated with decinormal potassium hydrate.
5.36 cubic centimeters KOH were required, corresponding to
5.25 per cent. NH 4 .
C 19 H 27 NH 4 0, requires 8.78 per cent. NH 4 .
C 19 H U7 NH 4 0,,C 19 H 28 0 2 requires 3.09 per cent NH 4 .
OTHKR SAINTS OF ABIETIC ACID.

By the use of the potassium salt for double decomposition,
abietates were obtained of the following metals : Barium, calcium, magnesium, zinc, iron, manganese, chromium, and silver. Some of these were prepared in aqueous and some in
alcoholic solution. The products were all amorphous ; such
analyses as were made for metallic content gave results that
were quite close to the theoretical requirements for neuttal
salts.
It has been observed that crude metallic resinates vary in
solubility under the influence of sunlight. 1 The same phenomenon was observed with the pure salts obtained, notably
those of zinc and magnesium. Some zinc salt, which formed a
clear solution in benzol, was exposed to sunlight in a shallow
layer for two days (protected, of course, from laboratory fumes).
Now on treating with benzol a white residue remained which
dissolved in acid with slight effervescence, and was shown to
be zinc hydrate with a small quantity of carbonate. An excess of the unexposed salt, treated with benzol, gave a solution which contained 10.7 per cent, of its weight of substance,
while after exposure a similar solution (after decanting from
1

Miiller-Jacobs: Ztschr. angew. Chem., 1890.

43

the insoluble residue described) contained 55.1 percent of substance.
There has evidently been formed free abietic acid or an acid
salt more soluble in benzol. However, no crystals of abietic
acid could be obtained from this solution. Some of the silver
salt was exposed to light, treated with alcohol and ether, filtered, and diluted with water. In the amorphous mass obtained, crystals of abietic acid could be identified by the use of
the microscope and polarized light.
Maly observed that metallic abietates, which had been kept
some time, differed in properties from those freshly prepared ;
he found that some salts, especially that of barium, gained in
weight. This he ascribed to oxidation.
He prepared a zinc
salt by double decomposition,which contained 11.02 per cent. Zn
[(C 19 H ai 0 2 ),Zn requires 10.2 per cent. Z n ) ] . On dissolving in alcohol and precipitating with ether, and repeating this treatment
several times, a product was finally obtained which contained
17.11 per cent. zinc.
Colored Resinates.
Closely related to the salts just discussed are the colored
resinates described by Muller-Jacobs.' They are prepared by
adding a basic organic dyestuff to a dilute rosin soap, and then
precipitating by the addition of a solution of metallic salt. The
product obtained has retained the color of the dyestuff, is insoluble in water, slightly soluble in alcohol, but readily soluble
in benzol, turpentine, and most oils and fats. Perhaps the
most interesting feature of these products is their sensitiveness
to light. This property has made them the basis of a photoengraving process.2
The colored resinates seem to be double salts of the metallic
base and the color base with abietic acid. A large number
were prepared and studied with the hope of getting some clue
to their composition and the nature of the change they undergo
through the influence of light. The products are complex and
vary according to the conditions of their preparation. Deter1
Ztschr. angew. Chem., 1889, p. 432; Ding. poly. J., 273, 139;
Reimann's Farberzeit., 30, 1894.
* Muller-Jacobs: Ztschr. angiw. Chem., 1890.

44
minations of their metallic content were made and it was attempted to find the percentage of colorbase by a series of
colorimetric comparisons. The results obtained were unsatisfactory, but the work has not been completed and it is hoped
that more decisive results will be obtained.
The remainder of this discussion will be limited to a few observations made with these products. There are several facts that
point to a double salt structure for the colored resinates. It has
already been seen that abietic acid has a tendency to form acid
salts with the alkalies ; such salts have free carboxyl groups to
unite with the basic color, and the great insolubility of the
new product formed will cause its precipitation from aqueous
solution. Warm solutions in which the tendency to form acid
salts is greatest, give the best quality of resinate color. Cold
solutions give irregular products and require larger quantities
of metallic salt to precipitae them. To a cold solution of pure
neutral potassium abietate were added solutions of rosanilin
and zinc sulphate ; a small quantity of benzol soluble-colored
salt was produced. A much larger yield was obtained when
the solutions were heated or when the acid salt was originally*
used.
A solution of neutral potassium abietate was heated with
some methyl violet. On cooling with ice a benzol-soluble salt
was produced which contained potassium.
Some of the dry colored salt made from pure abietic acid was
exposed to sunlight in a shallow layer. It had previously been
washed until all water-soluble color was removed; after exposure and treatment with water, a deep color was imparted to
the latter, showing that free basic color had been produced.
Other products of the decomposition could not be identified ;
the action of light on the material in bulk, is slow. A thin
film of the substance placed under a negative is sufficiently exposed in from fifteen to thirty minutes, but in such case the
quantity of material is too small to work with.
A saturated benzol solution of a zinc-rosanilin abietate prepared in diffused daylight, contained 12. i per cent, of its weight
of substance ; a similar solution of the same material, after ex-

45
posure to sunlight for one day, contained 1.6 percent, of its
weight of substance.
It was attempted to prepare simple salts of abietic acid with
basic colors and with anilin, but none could be obtained.
Some experiments were made with the hope of utilizing the
colored salts as a test for rosin (e. g-., in soap) and perhaps of
devising a method of colorimetric determination.
Work in this direction will be continued.

CONCLUSION.
In the first part of this work further evidence was obtained
for the anhydride structure of rosin. It was also shown that
the anhydride will form on heating the acid, and that abietic
acid cannot be looked upon as an oxidation product of turpentine.
Pure abietic acid was produced by an improved method.
The examination of the products obtained confirmed in every
respect Mach's formula, C19H28Oa.
Some new salts were made and some new tests applied. It
was shown that the salts of abietic acid are decomposed by
the action of water, and more rapidly when exposed to sunlight. The products of the decomposition were in some cases
identified.

BIOGRAPHICAL.
Hermann A. L,oos was graduated from the College of the City
of New York in 1895, with t n e degree of Bachelor of Science.
1895-1897—Teacher in New York City public schools.
1895-1899—Teacher in New York City evening schools.
1896—Summer Course in Chemistry, New York University.
1897-1900—Attended under Columbia University Faculty of
Pure Science.
1898—Received the degree of Master of Arts ; Thesis :
" The Electrolytic Determination of Zinc as Amalgam."
1898-1899—Instructor in Chemistry at the East Side Evening High School, New York City.
1899-1900—University Fellow in Chemistry.
Publications—"A Study on the Metallic Carbonyls and their
Decompositions" (with V. Lenher), School of Mines Quarterly, 21, 182, " O n the Decomposition of Nickel Carbonyl
in Solution" (with V. L,enher). J. Am. Chem, S o c , 22, 114.