•June 8, 1872.J 
THE PHARMACEUTICAL JOURNAL AND TRANSACTIONS. 
991 
substitution of Cl for OH had taken place, and ap¬ 
parently no action at all had ensued. 
(5). Tricodeia. —Tricodeia hydrochlorate was heated 
to 100 ° for 1 j hours with a large excess of strong HC1; 
on adding water to the product, a tarry substance was 
precipitated, whereas the original tricodeia hydrochlo¬ 
rate is readily soluble in dilute HC1; precipitated by 
Na 2 C0 3 , and the precipitate exhausted with ether, a 
a viscid non-crystalline hydrochlorate was obtained on 
agitation of the ethereal extract with HC1. The re¬ 
actions of this product appear to be identical with those 
of tricodeia, excepting that the reddish-purple tinge 
with Fe 2 Cl 6 appears instantaneously instead of only 
after standing a short time. Dried at 100 °, 
0*3070 grm. gave 0*756 C0 2 and 0*185 H 2 0. 
0*2480 „ 
0*1150 Ag-Cl. 
Calculated. 
Found. 
C 103 . 
1296 
68*03 
67*16 
H 120 . 
120 
6*30 
6*69 
Cl 6 . 
213 
11*18 
11*48 
N 6 •. 
84 
4*41 
Oj 2 •. •• .. 
192 
10*08 
103^-114^6^125 OH Cl 
1905 
100*00 
Hence this product has been formed by the reaction 
C 108 H m N e O 18 , 6 HCl= 6 H 2 O+O 108 H m N 6 O I 2 , 6 Ha, _ 
;and has the composition of a polymeride of “ apocodeia.’ ’ 
From the great similarity observed between this pro¬ 
duct and “ apocodeia” made by Matthiessen and Burn¬ 
side’s process,* it appears probable that the product of 
.the action of zinc chloride on codeia is a mixture of 
bodies of general formula (C 18 H 19 N0 2 )?i mHCI, in which 
the derivative where «=6 greatly predominates ; ex¬ 
periments on the action of zinc chloride on morphia 
®ow in progress in conjunction with Herr L. Mayer 
indicate^that mixtures are obtained in this case also. 
(c.) Bicodeia. —When pure dicodeia hydrochlorate is 
heated to 100 ° for one hour with a large excess of HC1, 
a change is produced expressible by the equation 
€ 72 H 84 N 4 0 12 ,4HC1+HC1= C 72 H 83 ClN 4 0 u ,4HCl-fH 2 0, 
which shows that the formula of this polymeride con¬ 
tains at least C 72 . Na 2 C0 3 throws down from the pro¬ 
duct a voluminous white precipitate, which differs in 
.appearance slightly from that of dicodeia, and turns 
green by exposure to air ; ether dissolves this precipi¬ 
tate, and on agitation with HC1 a viscid hydrochlorate 
is obtained which does not crystallize, but dries up to 
-a gum. Ee 2 Cl 6 gives a brown-purple tint, NO a H a 
blod-red, and 
K 2 Cr 2 0 7 +S0 4 H 2 
a lighter blood-red, none of which reactions occur with 
the original dicodeia. Dried at 100°, 
0*3200 grm. gave 
0*737 CO 
2 and 0*189 H a O. 
0*3260 „ 
0*172 Ag 
Cl. 
Calculated. 
Found. 
c 72 . 
y 
864 
-v 
63*50 
62*82 
h 87 . 
87 
6*39 
6*56 
Cl 5 . 
177*5 
13*04 
13*06 
n 4 . 
56 
4*12 
On. 
176 
12*95 
H 83 C1N 4 0 11 ,4HC1 
1360*5 
100*00 
III .—Action of Hydriodic Acid and Phosphorus on the 
Polymerides of Codeia. 
(a). Bicodeia. —When pure dicodeia is dissolved in a 
large excess of strong hydriodic acid (55 per cent. HI) 
and heated, together with a piece of phosphorus, to 
.ebullition until the boiling-point rises to 120 °, methyl 
* Proc. Roy. Soc., vol. xix. p. 71. 
iodide is given off and a considerable quantity of phos¬ 
phoric acid formed. The product, filtered through 
asbestos and precipitated with water, yields snow-white 
flakes that become yellow by exposure to air, and melt 
to a colourless oil at 100 ° when moist, although they do 
not fuse at that temperature when thoroughly dried. 
Dried at 100°, 
0*3155 grm. gave 0*5620 C0 2 and 0*1460 H 2 0. 
0*1895 „ 0*1190 Agl. 
Calculated. 
Found. 
Cj3G . « • . • . 
1632 
48*45^ 
48*58 
-0-161. 
161 
4*78 
5*14 
I 9 . 
1143 
33*94 
33*92 
N 8 . 
112 
3*33 
Oo# • « • « • • 
320 
9.50 
-i36H J53 IN 8 O 20 ,8HI 
3368 
100*00 
Hence this substance is formed by the reaction— 
2C 72 H 84 N 4 O 12 ,4HI+20HI= 
=I 8 +4H 2 0+8CH 3 I-f C 136 Hj S3 IN 8 O 20 ,8HI. 
The physical properties of this substance are almost 
identical with those of the bodies of analogous constitu¬ 
tion (containing C 13G ) formerly obtained from both 
codeia and morphia; carbonate of sodium throws down 
a precipitate almost insoluble in ether, showing that 
polymerization to the tetra-series has taken place; 
agitated with a large bulk of ether, this precipitate 
furnishes an extract, which, on agitation with dilute 
nitric acid and boiling with AgN0 3 and N0 3 H of the 
nitrate thus obtained, yields a precipitate of Agl, show¬ 
ing that iodine is contained in the precipitated base, 
The substance itself boiled with AgN0 3 and HN0 3 
produces a deep orange-coloured liquid, intermediate 
in tint between the blood-red produced by the derivatives 
of polymerized C 17 H 19 N0 3 , and the deep yellow by those 
of polymerized C ] 7 H. 21 N0 3 , a result confirmatory to some 
extent of the formula deduced from the analysis, this 
being capable of representation as— 
8 (C 17 H 20 NO 3 )+9HI - 4H 2 0. 
From this it appears pretty evident that the formulas 
hitherto attributed to the tetra bases (containing C 68 - C 72 ) 
are only half the true ones, which contain C J36 — C 144 . 
(5). Tetracodeia. —On treating tetracodeia in the same 
way and continuing the ebullition until the temperature 
reaches 130°, a brown syrupy liquid is finally obtained, 
which yields, on filtration through asbestos and pre¬ 
cipitation with water, a yellow brittle tar not fusible at 
100 ° when quite dry; methyl-iodide is produced in 
quantity during the action, but only traces of phosphoric 
acid, and this probably by atmospheric oxidation. Dried 
at 100 ° the tar gave these numbers : — 
0*3660 grm. gave 0*621 C0 2 and 0*149 H O. 
0*5520 „ 0*363 Agl. 2 
Calculated. 
Found. 
C 135 .. .. .. .. 
1632 
46*31 
46*27 
Hiss . 
158 
4*48 
4*53 
I ]0 . 
1270 
36*04 
35*54 
N s . 
112 
3*18 
o 22 . 
352 
9*99 
Oi36^i5oI 2 N 8 0 22 ,8HI 
3524 
100*00 
Hence this substance is formed by the reaction— 
C 144 H 168 N 8 0 24 ,8HI+ 10HI= 
= 8 CH 3 I+2H 2 0+C 136 H 150 I 2 N 8 O 22 , 8 HI. 
N0 3 H and AgN0 3 give a blood-red colouration with this 
product, showing, as the analytical numbers indicate, 
that it is derived from polymerized C 17 IIi 9 N0 3 , and not 
from polymerized Cj-H^NO^ or C 17 H 21 N0 3 . 
The foregoing results show that the methyl group m 
codeia is unaltered during the polymerization to dicodeia 
and to tetracodeia, and furnishes another proof of the 
conclusion come to in Part IV., § 2 , that the addition 
