236 
THE PHARMACEUTICAL JOURNAL AND TRANSACTIONS. [September 21 , 1872. 
line principle, generated by the abstraction of the ele¬ 
ments of water from that substance. 
This view was supported by the undoubted fact al¬ 
ready alluded to, that by repeated washings a portion is 
rendered permanently soluble, and, on the other hand, 
that pure aloin heated for a long time on the water- 
bath is partly resinified. 
Experiments with the soluble resin A. 
Three combustions were made in oxygen gas, the 
substance being weighed in a platinum boat which was 
introduced into the combustion tube behind a long 
column of granular oxide of copper. 
1. From Barbadoes aloes— 
•2968 of substance dried as in all the following ex¬ 
periments at 220° Fahr. left -0003 ash, and gave 
•670 CO, and ’1510 H,0. 
2. The same. 
•3162 gave *0005 of ash, and -7115 CO., and ’154 
H 2 0. 
3. Socotrine. 
•3275 gave ‘7260 C0 2 and -1680 H,0. 
These data correspond respectively to the following 
percentages: — 
C (1) 61-60 (2) 61-46 (3) 60-46 
II 5-65 5-42 5-69 
Considering the nature of the substance analysed, 
these numbers agree with each other satisfactorily. 
A portion of the resin to which the analyses (1) and 
(2) refer was dissolved in rectified spirit, and the solu¬ 
tion deluged at once with bromine water in excess, A 
precipitate formed which was collected, washed, and 
dried in vacuo over sulphuric acid. The bromine in it 
was determined by heating in a sealed tube with nitric 
acid and silver nitrate. 
•3145 gave *302 Ag. Br. 
corresponding to 
40-85 per cent, of Bromine. 
Now, assuming this resin A to have the constitution 
which we believe we have a right to assign to it, its 
formula would be 
C 6S h 70 0 27 . 
That is to say two molecules of aloin minus a molecule 
of water. 
2 (C 34 H 36 0 14 ) _ 11,0 = C 63 H ro 0 2r 
The percentages of carbon and hydrogen required by 
this formula are given below, together with the experi¬ 
mental numbers— 
Theory. Experiment. 
12 3 
C rs 816 61-91 61-60 61-46 60-46 
H ro 70 5-31 5-65 5-42 5*69 
(l ? 432 — — _ _ 
kfis H 58 Br 12 0 27 . 
C'6S 
Theory. 
816* 
Has 
58 
Br,, 
960 
O 
to 1 
M 1 
432 
Experiment. 
42-36 
40-85 
From this comparison it will _ be seen that, so far as 
composition is concerned, our view is completely con¬ 
firmed. 
Further evidence which seems to be conclusive was 
obtained by heating a portion of the resin with water in 
a sealed tiibe to 300° F. for several hours; when cold the 
tube was opened, the solution filtered from the undissolved 
residue and evaporated to a syrup. It could not, how¬ 
ever, be made to crystallize. 
I his was perhaps to have been expected, contaminated 
as it was with a considerable quantity of resinous im¬ 
purity. But by adding bromine water to a part of the 
solution a yellow precipitate was obtained, which, when 
dissolved in alcohol and left to spontaneous evaporation, 
furnished crystalline tufts which were visible under the 
microscope. 
By boiling a portion of the same substance for several 
hours with dilute hydrochloric acid in a flask from which 
the air was excluded, a larger portion was rendered 
permanently soluble. The solution was strongly bitter 
and gave a precipitate with bromine water. 
We also found that when resin A was boiled with, 
nitric acid, a large quantity of chrysammic acid was 
formed, together with picric and oxalic acids and car¬ 
bonic anhydride. It yielded the same products, in fact, as 
are obtained from barbaloin under the same circum¬ 
stances. 
We thought it likely that, conversely, crystallized 
barbaloin would be readily converted into resin by the- 
action of dehydrating agents, but we found that on 
heating it with chloride of zinc, or with concentrated 
hydrochloric acid in sealed tubes, till on gradually raising 
the temperature a change was visible in the contents of 
the tube, the condensation seems to proceed at one step 
too far, and the result is a black insoluble pulverulent 
matter which we have not yet examined further. 
The previously recounted experiments appear, how¬ 
ever, to establish the view that that portion of the resi- 
noid matter of aloes which is soluble in hot water 
is a kind of anhydride of aloin standing in the 
same relation to the crystalline substance that ether 
does to alcohol, or, as recently shown, that tannic acid 
bears to gallic acid. In both cases a crystalline body is 
converted by condensation and the loss of the elements, 
of water into an uncrystallizable body. 
2(C ? H c O ) - H 2 0 = O u H,„0 3 
Gallic Acid. Tannic Acid. 
^ (^ 34 ^ 136 ^ 14 ) H 2 0 = c C8 h 70 o 27 
Barbaloin. Aloe Resin A. 
One practical conclusion that may be drawn from these 
experiments is, that in preparing extract of aloes, the- 
quantity of product may be much augmented, probably 
without deterioration of its quality, by treating the de¬ 
posited resin several times with water. 
Experiments with the insoluble resin Ii. 
This substance purified in the manner already described 
was analysed by combustion in oxygen gas as in the last 
case. 
1. Barbadoes 
•2980 gave -6795 C0 2 and -1520 H 2 0. 
2. Barbadoes 
*3225 gave -7410 C0 2 and *1635 H 2 0. 
3. Socotrine 
•3410 gave -0015 ash, and *7710 CO, and -1670 
H 2 0. 
Calculating from these data, the percentages of carbon 
and hydrogen are as follows:— 
1. 2. 
3. 
C. 62-18 60-27 
61-94 
H. 5-66 5-63 
5-46 
It will be seen in these analyses that the numbers ob¬ 
tained are not perfectly concordant, and this might in 
fact be almost expected from the indefinite and change¬ 
able character of the substance operated upon. More - 
over, it is possible that we may not have been wholly 
successful in removing from it traces of foreign matters 
soluble in rectified spirit. Nevertheless, we incline to 
the belief that the composition of this substance is more 
nearly represented by the numbers obtained in the first 
and third analysis than by those of the second. And 
we regard it as not improbable that this insoluble part 
of the resin may result from the simultaneous oxidation 
and condensation of crystallizable barbaloin. Suppose, 
for instance, two molecules of barbaloin to lose a molecule 
of water, and at the same time four atoms of hydrogen : 
2 (CwH-aOu) - H.O-H, = C B H w O^. 
The formula to which such an hypothesis would lead, 
would be 
^6sT7g6^27> 
