a) The Cacao Bean Proper.

Just as the beans of the cacao fruit are included under the botanical concept “Seed”, so also their chemical constituents closely resemble those common to every other seed. There are the usual reserve stuffs inherited from the mother plant, which serve as sustenance for the yet undeveloped organs, and compare with albumen in the feathered world. Apart from the constituents incidental to all plant life at this stage, such as albumin, starch, water, fat, sugar, cellulose and mineral stuffs such as ash, the cacao seed has two other components peculiar to itself; Theobromine and Cacao-red. We adjoin a succession of chemical determinations respecting the quantitative proportions of these substances in the seed, and think further that we may be allowed to cite the results of fore-time investigators in this sphere, especially as their work has formed the basis for all future operations, and again, in view of the doubt which still prevails in scientific circles as to the “Normal” composition of the cacao bean.

Table 4.
Percentage Composition of the Hulled Bean.

Analyst	Payen[20]	Lampadius[20]	Mitscherlich[20]
Constituents percent	Undescribed	West Indies	Guayaquil	Caracas
1. Water	10·0	3·40	5·60	—
2. Nitrogenous matter	20·0	16·70	14·39	—
3. Theobromine	2·2	—	1·20	—
4. Fat	52·0	53·10	45-49	46-49
5. Cacao-red	—	2·07	3·50	—
6. Sugar	—	—	0·60	—
7. Gum and Starch	10·0	7·75	14·30	13·5
8. Woody fibre	2·0	0·90	5·80	—
9. Ash	4·0	3·43	3·50	—

Table 5.

Constituents percent	Caracas	Guayaquil	Trinidad	Puerto Cabello	Surinam
	Laube		Aldendorff
1. Water	4·04	3·63	2·81	2·96	3·76
2. Nitrogenous matter	14·68	14·68	15·06	15·03	11·00
3. Fat	46·18	49·04	48·32	50·57	54·40
4. Starch	12·74	11·56	14·91	12·94	—
5. Other non-nitrogenous matter	18·50	12·64	12·06	11·49	28·32
6. Woody fibre	4·20	4·13	3·62	3·07	—
7. Ash	3·86	3·72	3·22	3·94	2·35
	C. Heisch
Constituents percent	Granada	Bahia	Cuba	Para
1. Water	3·90	4·40	3·72	3·96
2. Nitrogenous matter	12·45	7·31	8·56	12·50
3. Fat	45·60	50·30	45·30	54·30
4. Starch	—	—	—	—
5. Other non-nitrogenous matter	35·70	35·30	39·41	26·33
6. Woody fibre	—	—	—	—
7. Ash	2·40	2·60	5·90	3·06

The analyses carried out by Zipperer in the year 1886 yielded the following results[21]:

Table 6.
A) Analysis of the Raw Shelled Bean (Kernel).

Constituents percent	Names of Sorts
Constituents percent	Ariba	Machala Guayaquil	Caracas	Puerto Cabello
1. Moisture	8·35	6·33	6·50	8·40
2. Fat	50·39	52·68	50·31	53·01
3. Cacaotannic acid, sugar, decomposition products, phlobaphene	8·91	13·72	10·76	7·85
4. Theobromine	0·35	0·33	0·77	0·54
5. Starch	5·78	8·29	7·65	10·05
6. Cellulose and proteins	22·10	14·45	19·84	15·83
	Proteins	Proteins	Proteins	Proteins
	to	to	to	to
	cellulose	cellulose	cellulose	cellulose
7. In the ratio	7·3:1	5:1	6·6:1	5·3:1
8. Ash	5·12	4·17	4·17	4·32
	Surinam	Trinidad	Port au Prince	Average
1. Moisture	7·07	6·20	6·94	7·11
2. Fat	50·86	51·57	53·66	51·78
3. Cacaotannic acid, sugar, decomposition products, phlobaphene	8·31	9·46	11·39	10·02
4. Theobromine	0·50	0·40	0·32	0·45
5. Starch	6·41	11·07	8·96	8·33
6. Cellulose and proteins	24·13	18·43	15·81	18·71
	Proteins	Proteins	Proteins	Proteins
	to	to	to	to
	cellulose	cellulose	cellulose	cellulose
7. In the ratio	8:1	6:1	5·25:1	6·2:1
8. Ash	2·72	2·87	2·92	3·60

In addition to these, there is an exhaustive succession of analyses conducted by Ridenour,[22] which we accordingly submit as Table 8. Following Filsinger,[23] we cannot regard these analyses as an absolutely trustworthy representation of the “Normal” composition of the cacao bean, the values in starch, albumin and ash considerably deviating from all that have been established up to the present time. Among more recent researches, we cite those carried out by Matthes and Fritz Müller.[24]

Table 7.
B) Analysis of the Raw Shelled Bean (Kernel).

Constituents percent	Names of Sorts
Constituents percent	Ariba	Machala Guayaquil	Caracas	Puerto Cabello
1. Moisture	8 ·52	6 ·25	7 ·48	6 ·58
2. Fat	50·07	52·09	49·24	48·40
3. Cacaotannic acid, sugar and phlobaphene	8 ·61	7 ·84	6 ·85	8 ·25
4. Theobromine	0 ·30	0 ·31	0 ·05	0 ·52
5. Starch	9 ·10	11·59	9 ·85	10·96
6. Cellulose and protein bodies	19·43	18·17	22·16	21·21
	Proteins	Proteins	Proteins	Proteins
	to	to	to	to
	cellulose	cellulose	cellulose	cellulose
7. In the ratio	6·5:1	6:1	7·7:1	7:1
8. Ash	3 ·89	3 ·75	3 ·92	4 ·08
	Surinam	Trinidad	Port au Prince	Average
1. Moisture	4 ·04	7 ·85	6 ·27	6 ·71
2. Fat	49·88	48·14	46·90	49·24
3. Cacaotannic acid, sugar and phlobaphene	8 ·08	7 ·69	7 ·19	7 ·78
4. Theobromine	0 ·54	0 ·42	0 ·36	0 ·43
5. Starch	10·19	8 ·72	12·64	10·43
6. Cellulose and protein bodies	24·39	23·06	21·82	21·43
	Proteins	Proteins	Proteins	Proteins
	to	to	to	to
	cellulose	cellulose	cellulose	cellulose
7. In the ratio	8:1	7·6:1	7·3:1	7·1:1
8. Ash	2 ·88	4 ·12	4 ·82	3 ·92

Table 8. Ridenour.

	Commercial Varieties
Constituents percent	Bahia	Surinam	Java	Trinidad	Roasted Trinidad	Ariba	Caracas
1. Fat	42·10	41·03	45·50	43·66	41·89	43·31	36·81
2. Theobromine	1·08	0·93	1·16	0·85	0·93	0·86	1·13
3. Albumin	7·50	10·54	9·25	11·90	12·02	10·14	10·59
4. Glucose	1·07	1·27	1·23	1·38	1·48	0·42	2·76
5. Saccharose	0·51	0·35	0·51	0·32	0·28	1·58	1·56
6. Starch	7·53	3·61	5·17	4·98	5·70	6·37	3·81
7. Lignin	7·86	3·90	6·10	5·65	5·87	4·62	3·28
8. Cellulose	13·80	16·24	13·85	13·01	19·64	14·07	16·35
9. Extractive by difference	8·99	13·53	8·90	8·31	5·84	9·00	12·72
10. Moisture	5·96	5·55	5·12	6·34	2·63	5·90	6·63
11. Ash	3·60	3·05	3·31	3·60	3·70	8·73	4·36
	Commercial Varieties
Constituents percent	Roasted Caracas	Granada	Tabasco	Machala	Maracaibo	Average
1. Fat	37·63	44·11	50·95	46·84	42·20	42·99
2. Theobromine	0·99	0·75	1·15	0·76	1·03	0·97
3. Albumin	12·36	9·76	7·85	12·69	11·56	10·51
4. Glucose	1·76	1·81	0·94	1·60	1·09	1·46
5. Saccharose	0·51	0·55	2·72	0·46	1·36	0·89
6. Starch	6·07	6·27	3·51	1·35	1·69	4·67
7. Lignin	9·05	5·55	6·44	5·95	7·16	5·95
8. Cellulose	11·69	13·49	12·57	11·32	17·32	14·44
9. Extractive by difference	9·22	9·72	9·26	9·02	6·79	9·30
10. Moisture	5·69	5·28	1·55	5·86	5·67	5·18
11. Ash	5·03	2·71	3·06	5·15	4·13	3·70

Table 9.

No.	Description	Moisture	Ether	Non-fatty dry substances	Mineral constituents	In water
		Moisture	Ether	Non-fatty dry substances	Mineral constituents	insoluble ash	soluble ash
		%	%	%	%	%	%
1	St. Thomas II	2·82	55·87	—	2·79	1·93	0·86
2	Java I	2·78	53·88	—	3·60	1·60	2·00
3	St. Thomas I	2·82	54·50	—	3·01	1·85	1·16
4	Caracas I	2·67	53·78	—	3·35	2·12	1·23
5	Puerto Cabello	3·34	53·29	—	3·58	1·73	1·85
6	Machala	2·93	53·98	—	3·34	2·10	1·24
7	Samana	2·94	55·28	—	3·10	1·85	1·25
8	Accra	2·94	53·94	—	3·19	1·84	1·35
		B. Percentages for the non-fatty dry substances.
1	St. Thomas II	—	—	41·36	6·536	4·672	1·864
2	Java I	—	—	43·34	8·306	3·692	4·614
3	St. Thomas I	—	—	42·68	7·053	4·311	2·742
4	Caracas I	—	—	43·55	7·692	4·868	2·824
5	Puerto Cabello	—	—	43·37	8·254	3·989	4·265
6	Machala	—	—	43·09	7·767	4·900	2·867
7	Samana	—	—	42·78	7·246	4·325	2·921
8	Accra	—	—	43·12	7·398	4·267	3·131
		C. Percentages for the total of ash.
1	St. Thomas II	—	—	—	—	71·49	28·51
2	Java I	—	—	—	—	44·45	55·55
3	St. Thomas I	—	—	—	—	61·12	38·88
4	Caracas I	—	—	—	—	63·38	36·62
5	Puerto Cabello	—	—	—	—	48·33	51·67
6	Machala	—	—	—	—	63·09	36·91
7	Samana	—	—	—	—	59·69	40·31
8	Accra	—	—	—	—	57·68	42·32

No.	Description	Alkali strength		Potassium Carbonate reckoned from Alkali strength of soluble ash	Pure ash (mineral stuffs minus Pot. Carb.)
		of the soluble ash	of the insoluble ash		Pure ash (mineral stuffs minus Pot. Carb.)
		cb. mm. Nitric acid.		%	%
1	St. Thomas II	3·6	4·8	0·25	2·54
2	Java I	10·4	6·8	0·72	2·88
3	St. Thomas I	2·6	5·0	0·18	1·83
4	Caracas I	4·6	4·8	0·32	3·03
5	Puerto Cabello	10·4	3·8	0·72	2·86
6	Machala	2·6	5·6	0·18	3·16
7	Samana	4·6	6·2	0·32	2·78
8	Accra	3·6	4·8	0·25	2·94
		B. Percentages for the non-fatty dry substances.
1	St. Thomas II	8·7	11·6	0·60	5·94
2	Java I	24·0	15·7	1·66	6·65
3	St. Thomas I	6·1	11·7	0·42	6·63
4	Caracas I	10·6	11·0	0·73	6·96
5	Puerto Cabello	24·0	8·8	1·66	6·59
6	Machala	6·1	13·0	0·42	7·35
7	Samana	10·8	14·5	0·74	6·50
8	Accra	8·3	11·1	0·58	6·82
		C. Percentages for the total of ash.
1	St. Thomas II	133·1	177·4	9·18	90·82
2	Java I	289·1	189·1	20·00	80·01
3	St. Thomas I	87·0	167·0	6·00	94·04
4	Caracas I	137·9	143·9	9·50	90·51
5	Puerto Cabello	290·7	106·6	20·10	79·89
6	Machala	78·5	—	5·40	94·59
7	Samana	149·0	200·0	10·20	89·79
8	Accra	112·2	150·0	7·8	92·16

No.	Description	Phosphoric acid			Silicic acid (SiO₂)	Ferric acid (Fe₂O₃)
		total	soluble in water	insoluble in water	Silicic acid (SiO₂)	Ferric acid (Fe₂O₃)
		%	%	%	%	%
1	St. Thomas II	1·0243	0·2474	0·7769	0·0154	0·0416
2	Java I	1·0753	0·4667	0·6086	0·0300	0·0224
3	St. Thomas I	1·1136	0·3621	0·7515	0·0122	0·0464
4	Caracas I	1·2708	0·3392	0·9316	0·0080	0·0184
5	Puerto Cabello	1·1433	0·4692	0·6741	0·0260	0·0207
6	Machala	1·2836	0·3647	0·9189	0·0116	0·0200
7	Samana	1·0881	0·3213	0·7668	0·0090	0·0560
8	Accra	1·1221	0·3672	0·3549	0·0082	0·0284
B. Percentages for the non-fatty dry substances.
1	St. Thomas II	2·4795	0·5989	1·8806	0·0373	0·1007
2	Java I	2·4790	1·0769	1·4021	0·0692	0·0517
3	St. Thomas I	2·6092	0·8484	1·7608	0·0286	0·1087
4	Caracas I	2·9180	0·7789	2·1356	0·0184	0·0422
5	Puerto Cabello	2·6361	1·0819	1·5542	0·0600	0·0477
6	Machala	2·9837	0·8481	2·1356	0·0269	0·0464
7	Samana	2·5435	0·7511	1·7934	0·0214	0·1309
8	Accra	2·6023	0·8516	1·7507	0·0191	0·0658
C. Percentages for the total of ash.
1	St. Thomas II	37·94	9·16	28·78	0·571	1·541
2	Java I	29·87	12·96	16·91	0·833	0·623
3	St. Thomas I	37·27	12·12	25·15	0·408	1·551
4	Caracas I	37·94	10·12	27·82	0·240	0·549
5	Puerto Cabello	31·94	13·11	18·83	0·727	0·578
6	Machala	38·42	10·92	27·50	0·346	0·597
7	Samana	35·12	10·37	24·75	0·295	1·806
8	Accra	35·18	11·51	23·67	0·258	0·889

Table 10. Commoner Varieties.

Key to Column Headings

C; Moisture
D; Ether extract
E; Mineral matter
F; Potassium Carbonate reckoned on alkali soluble in water
G; Pure ash (mineral matter minus K₂CO₃)
Ha; according to König, as modified by us
Hb; as yielded by the Wender process
I; Silicic acid (SiO₂)
J; Ferric oxide (Fe₂O₃)
K; Soluble in alcohol P₂O₅

No.	Description	C	D	E	F	G
No.	Description	%	%	%	%	%
1	Superior Ariba, Summer crop	6·95	26·17	7·45	2·07	5·38
2	Machala 81%, Thomé I 19%	5·94	28·79	7·06	1·99	5·07
3	Machala 53%, Thomé I 47%	6·47	25·73	7·15	2·14	5·01
4	Cameroon	6·36	26·41	7·05	2·33	4·72
5	Thomé I 73%, Samana 27%	7·97	24·90	6·89	2·29	4·60
6	Thomé II 60%, Samana 20%, Accra 20%	7·37	22·85	7·39	2·24	5·15
7	Accra 60%, Thomé II 40%	6·93	22·80	7·36	2·25	5·11
8	A}Same variety,	6·56	18·96	7·61	2·14	5·47
9	B}more defatted	6·06	24·75	7·16	2·01	5·15
10	C}less defatted	5·58	29·72	6·57	1·89	4·68
11	Monarch double Ariba(R. & Cie.)	7·59	14·80	8·32	2·32	6·00
12	Helios(R. & Cie.)	7·37	17·25	7·91	2·12	5·79
a	Ariba shells (R. & Cie.) very fine ground	7·17	14·00	7·40	2·20	5·20
b	germs, Ariba (R. & Cie.) very fine ground	6·64	18·02	6·93	2·43	4·50
No.	Description	Raw Fiber
		Ha	Hb	I	J	K
		%	%	%	%	%
1	Superior Ariba, Summer crop	4·20	4·60	0·0170	0·0522	0·0605
2	Machala 81%, Thomé I 19%	5·00	5·47	0·0172	0·0373	0·0625
3	Machala 53%, Thomé I 47%	5·20	5·42	0·0186	0·0513	0·0612
4	Cameroon	4·63	4·64	0·0160	—	0·0669
5	Thomé I 73%, Samana 27%	4·20	4·38	0·0167	0·0753	0·0690
6	Thomé II 60%, Samana 20%, Accra 20%	4·23	5·00	0·0208	0·0678	0·0726
7	Accra 60%, Thomé II 40%	4·06	4·40	0·0198	0·0545	0·0766
8	A}Same variety,	4·00	5·24	0·0390	—	—
9	B} more defatted	3·58	4·61	—	—	—
10	C} less defatted	3·20	4·42	—	—	—
11	Monarch double Ariba(R. & Cie.)	6·90	—	0·0420	—	0·0877
12	Helios(R. & Cie.)	6·40	—	0·0340	0·0400	0·0930
a	Ariba shells (R. & Cie.) very fine ground	7·49	—	0·2976	—	0·0383
b	germs, Ariba (R. & Cie.) very fine ground	7·42	—	—	—	0·0587

Table 11. Analysis of Cacao.
Dry product, defatted and free from alkali.

Key to Column Headings

C; Defatted and alkali-free dry products
D; Pure ash (mineral substances less K₂CO₃)
E; Ash insoluble in water
F; Alkalinity of the insoluble ash Nitric acid
Ga; total
Gb; soluble in water
Gc; insoluble in water
H; Silicic acid (SiO₂)
I; Ferric oxide (Fe₂O₃)
J; P₃O₅ soluble in alcohol
K; after König (modified)
L; as yielded by the Weender process

No.	Description	C	D	E	F
No.	Description	%	%	%	ccm
1	Thomé II	41·06	6·186	4·725	11·7
2	Java I	42·62	6·757	3·754	15·9
3	Thomé I	42·50	6·659	4·353	11·8
4	Caracas I	43·23	7·010	4·904	11·1
5	Puerto-Cabello	42·65	6·706	4·056	8·9
6	Machala	42·91	7·365	4·894	13·1
7	Samana	42·46	6·548	4·357	14·6
8	Accra	42·87	6·858	4·292	11·2
9	Ariba	64·81	8·301	—	—
10	Machala + Thomé I	63·28	8·013	—	—
11	Thomé + Machala	66·66	7·517	—	—
12	Cameroon	64·90	7·273	—	—
13	Thomé I + Samana	64·84	7·095	—	—
14	Thomé II, Samana + Accra.	67·54	7·625	—	—
15	Accra + Thomé II	68·02	7·513	—	—
16	A	72·34	7·561	—	—
17	B	67·18	7·666	—	—
18	C	62·80	7·452	—	—
19	Monarch Ariba (R. & Cie.)	75·29	7·969	—	—
20	Helios Ariba (R. & Cie.)	73·39	8·880	—	—
a	Shells	76·63	6·786	—	—
b	Germs	72·91	6·173	—	—
No.	Description	Phosphoric Acid (P₂0₅)
		Ga	Gb	Gc	H	I
		%	%	%	%	%
1	Thomé II	2·4947	0·6025	1·8922	0·0375	0·1013
2	Java I	2·5229	1·0950	1·4279	0·0704	0·0525
3	Thomé I	2·6202	0·8520	1·7682	0·0287	0·1091
4	Caracas I	2·9391	0·7846	1·1545	0·0185	0·0425
5	Puerto-Cabello	2·6807	1·1001	1·5806	0·0610	0·0480
6	Machala	2·9914	0·8499	2·1414	0·0270	0·0466
7	Samana	2·5626	0·7802	1·7824	0·0212	0·1319
8	Accra	2·6175	0·8565	1·7610	0·0191	0·0662
			Raw fibre
		J	K	L
9	Ariba	0·0933	6·48	7·10	0·0262	0·0806
10	Machala + Thomé I	0·0984	7·90	8·64	0·0272	0·0590
11	Thomé + Machala	0·0919	7·80	8·13	0·0280	0·0770
12	Cameroon	0·1030	7·13	7·15	0·0246	—
13	Thomé I + Samana	0·1064	6·48	6·75	0·0258	0·1162
14	Thomé II, Samana + Accra.	0·1075	6·27	7·40	0·0308	0·1004
15	Accra + Thomé II	0·1126	5·97	6·47	0·0290	0·0801
16	A	—	5·53	7·24	—	—
17	B	—	5·33	6·87	—	—
18	C	—	5·10	7·04	—	—
19	Monarch Ariba (R. & Cie.)	0·1165	9·16	—	0·0558	—
20	Helios Ariba (R. & Cie.)	0·1266	8·72	—	0·0446	—
a	Shells	0·0499	9·77	—	0·3884	0·0545
b	Germs	0·0805	10·18	—	—	—

1) See Table 9 A and Table 10.

The foregoing tables provide us with a general idea of the chemical constituents of the cacao bean, but their distinctive properties, both chemical and physical, still remain to be defined, with which we accordingly proceed, as such data will on the one hand enable us to grasp how loss may be avoided in the manufacture of cacao and chocolate wares, and at the same time render intelligible familiar processes connected therewith.

As we have seen, the following substances occur in cacao in varying amounts:

1. Water.
2. Fat.
3. Cacao-red.
4. Theobromine.
5. Albumen.
6. Starch.
7. Cellular tissue or cellulose.
8. Small percentages of grape and cane sugar.
9. Mineral or ash stuffs.

Like the majority of plants and plant products, the cacao bean consists of vesicles or cells, closed on all sides and arranged in a series of layers. They are constructed of cellular tissue or cellulose, and contain fat, albumen, water, starch, theobromine, cacao pigment, besides sugar and salts in inferior quantities.

1. Water or Moisture.

There is present in the bean from 6 to 8 percent of water, a factor which bodes well for the proper germination of the seed, as when this latter is deprived of moisture, e. g. in the course of a too thorough drying, it speedily decays. Water is still evident in small quantities even in the largest and almost withered beans, as will be seen on comparison of the foregoing analyses.

2. Fat.

As a constituent at the expense of which respiration is effected, fat remains one of the most important resources of plant. It has a twofold excellence in this connection, and firstly as a highly calorifacient and carboniferous substance, and again because such a reserve enables the living organism to oxidise with particular ease, wherefore it is found accumulated in somewhat significant measure in the majority of seeds. When seen under the microscope it appears either as round coherent masses, or as crystalline aggregates clearly distinguishable from the rest of the cell contents on treatment with a solution of osmic acid. The fat in the cacao bean usually amounts to from 50-56 percent, or one half of the total weight of the shelled beans; the shell also contains from 4 to 5 percent of fat.[25] The unfermented bean has frequently, in addition to its bitter taste, a most unpleasant flavour, attributable to the rancidity of its fatty contents.

The raw bean contains rather more fat than the roasted bean, for whilst the one averages from 50 to 55 percent, there is seldom more than 48-52 percent in the other. The cause of this phenomenon may be connected with the enrichment of the shells in fat, and in some instances, as when the beans are over-roasted, is to be ascribed to the chemical change which the play of burning heat on fatty bodies involves, when a destructive decomposition of the whole ensues, with formations of acroleine. Chemically considered, cacao butter consists of a mixture of so-called esters, or compounds connected with ether, such as the glycerides of fatty acids, and contains, in addition to stearine, palmatine, and laurine[26], the glyceride of arachidic acid. It was also formerly supposed that formic, acetic and butyric acids were among the constituents of this ingredient, but the view has been proved erroneous by Lewkowitsch[27]; similarly, the presence of theobromic acid alleged by Kingzett[28] has been called into question by Graf.[29]

Cacao butter is a fairly firm fat of pleasant taste and smell, which varies in colour between yellowish white and yellow. When freshly expressed, it has frequently a brownish shade, passing after a short time into a pale yellow, and turning almost white on long keeping. The brown colour is due to pigment in suspension, which becomes sediment in the course of melting, when the butter asumes a normal colour, referrible to pigment dissolved in the butter oils, and secondarily to a dissolution of the products of roasting in these liquids, rather than to any matter in suspension. The pleasant smell and taste of cacao butter is probably closely allied to the dissolved substances mentioned.

The fat extracted from cacao by solvents differs essentially from that obtained by hydraulic pressure, a fact overlooked in some of even the most recent experiments, and which therefore cannot be too strongly emphasised. Extracted fat is yellowish white, sometimes approximating to grey, and after having been kept a long time, the whole becomes tinged with an actual whiteness, which first attacks the outer surface, and then rapidly progresses towards the centre in concentric paths, and which is a sign of rancidity. Its fracture is partly granular, the smell is not so pronounced as that of expressed fat, being even unpleasant at times, as in the case of faulty wares (but compare page), and it has a keen taste. Cacao butter does not, as is generally supposed, keep better than other vegetable fats, but is equally liable to become rancid, as Lewkowitsch[30] demonstrates. By rancidity is denoted that state of offensive taste and smell acquired by fatty substances on longer or shorter keeping and especially when they are not properly stored. What chemical re-arrangements of the respective constituents this state presupposes is very questionable; though it appears from the experiments of Lewkowitsch[30] and others[31] that the formation of acids does not play as prominent a part as the experimenter is inclined to think, nothwithstanding the marked increase in quantity which may occur. The primary cause of rancidity will rather be found in the oxidation products of the glycerine contained in all fats.

The specific gravity of cacao butter varies considerably, according as it has been expressed or extracted by means of solvents. White[32] asserts that it can only be determined when the liquefied oil has been solidified several days. According to Rammsberger the specific gravity of expressed butter is 0·85; that of butter extracted by treatment with ether figures at 0·958. Hager gives the normal specific gravity of fresh cacao butter at 15° C. as from 0·95 to 0·952; stale butter 0·945 to 0·946, and the same figures have been confirmed by other investigations, though Dietricht gives 0·98 to 0·981 at 100° C. The melting point is generally regarded as 33° C.; there is in this respect, however, a great difference between the two descriptions of fat. Expressed fat which has been kept for some length of time melts between 34° C. and 35° C., and these figures remain constant, so that it is advisable to read the melting point of fat which has been in store some time rather than that of the fresh pressed product, and take this as a standard. All other fat shows a lower melting point.

As the melting point of freshly melted cacao butter shows considerable fluctuation, the liquid fat must be kept in darkness and cooled with ice for about a week, and the reading should not be taken before the expiration of this time, as only then is it possible to obtain any definite and final result.

Experiments on the melting point of cacao butter as carried out by Zipperer under special conditions yielded the following values; cf. also Table 12.

Kind of bean	Melting point raw	Centigrade roasted
Machala Guayaquil	34·5	34·0
Caracas	33·5	34·0
Ariba	33·75	31·5
Port au Prince	34·25	33·8
Puerto Cabello	33·50	33·0
Surinam	34·20	34·0
Trinidad	34·00	34·0

White and Oldham[33] give the following melting points:

Guayaquil	33·6-33·9
Granada	33·0-33·3
Trinidad	31·5-32·5
Caracas	33·0-33·6
Ceylon	33·9-34·2

Filsinger and Henking found[34]:

Cauca	32·1-32·4
Bahia	32·7-33·4
Porto Plata	33·1-33·6

These results vary somewhat, but the differences are to be ascribed to the methods employed and to the manner in which the observations of different experimenters are carried out. Generally it may be taken that the melting point should not be under 3° or over 35°C. The fat solidifies between 21·5° and 23° C. (solidifying point). The fatty acids from the fat melt at 48°-52° C.; they begin to solidify at 45° C., the solidifying ending generally at 51°-52° C. (see table 12).

Adulteration of cacao fat, as many experiments have shown, cannot be detected simply by deflections in the melting point. Björklund’s ether test,[35] which is very suitable for the detection of an admixture of extraneous substances like tallow, wax and paraffin, is carried out as described in paragraph....

Cacao fat, like all other fats, is saponified by alkalis, that is to say, forms a soap or a chemical compound of the fatty acids with alkalis such as potash, soda, ammonia etc. On the addition of a mineral acid to the soap a salt of the mineral acid and alkali is formed, with the separation of the fatty acid. The fatty acids are of two kinds:

1. The volatile acids or those which are volatile at 100°-110° C. or more easily with steam than other vapours. These usually exist only in very small quantity in cacao fat but may considerably increase in amount in the fat obtained from imperfectly fermented beans.[36]

2. The solid fatty acids are such as are fixed, and do not act in the manner above mentioned: cacao butter consists chiefly of the glycerides of these acids.

Björklund’s tests will only detect, as has been stated, admixtures of wax, paraffin, tallow and bodies of a relatively high melting point. Another method must therefore be adopted to detect fat of low melting points, as cocoa-nut fat, or liquid oils like cotton seed and sesame oils. The methods in use in connection with cacao butter are the determination of the iodine, saponification and acid values, finding the melting point of the fatty acids, the Reichert-Meissl number, and by means of Zeiss’ butyro-refractometer, its refractive index.

The iodine value indicates the amount of iodine percent absorbed by the fat, and is accordingly a measure of the unsaturated fatty acids. As these latter differ in amount in vegetable and animal fats, though constant for each separate kind, it is possible by means of this iodine value to recognise a genuine cacao fat and to detect adulteration. The determination of the iodine value is carried out by Hulbl’s[37] method, and according to Filsinger,[38] it is advisable to let the iodine solution act on the fat for from ten to twelve hours in diffused daylight. Before determining the iodine value in cacao fat, says Welmans[39] this substance should be dried at from 100-105°C. to expel the acroleine produced by too high roasting, at the same time avoiding too high a temperature, as acroleine can then be very easily reproduced. Filsinger has determined the iodine value of many varieties of cacao butter with the following results:

Kind	Iodine value:
Cauca	36·2-36·7
Bahia	36·8-37·1
Porto Plata	36·6-36·9
Ariba	35·1-36·8

Genuine cacao butter shows an average iodine value of from 33-37·5.[40]

The saponification value or Köttstorfer’s number[41] expresses the number of milligrammes of potassium hydrate required for the complete saponification of 1 gramme of fat, or in other words, the amount of potassium hydrate necessary to the saponification of the fat in thents percent. Filsinger[42] gives the amount as between 192 and 202 in genuine cacao butter, although it usually fluctuates between 194 and 195. Its determination is the means of detecting adulterations with cocoa-nut butter and its preparations.

The determination of the acid value has lately become of importance, especially since the introduction of the so-called Dutch Ha cacao or shell butter, which is obtained from cacao refuse and is often rancid. This value or number expresses the amount of potassium hydrate necessary to neutralise the free fatty acids in 1 gramme of fat, and it is therefore a measure of the amount of free fatty acid. As this constant has been variously stated, according to the methods adopted (Burstyn, Merz), the fact must be taken into account when comparing the literature on the subject. As the constants have been determined by two different methods (Merz, Burstyn), this must be taken into consideration when comparing the various data on the acid value of fats. Whilst the “Vereinbarungen” (No. 1, 1897) in a chapter on “Food Fats and Oils” still recognise two distinct methods in the determination of free fatty acids, as well as two different ways of recording the results (degree of acidity and free acid, calculated on the oily acids) there occurs in the supplement to the recent margarine code for Germany issued by the Chancellor on April 1st. 1898, entitled “Instructions for chemical research in fats and cheeses” under c) a dictum that there is only one absolute and precise procedure in the “Determination of free fatty acids (degree of acidity) These calculations are based on the Burstyn method, which we accordingly annex, more especially as it is now in universal use. It should be observed that the method of preparation and the age of the beans, as well as that of the fat all tend to increase the acid value.

The Reichert Meissl value expresses the percentage value of the volatile fatty acids present in the fat; as already mentioned, they amount to 1·6 ccm, in cacao fat extracted by solvents. Milk chocolate, says Welmans, yields a fat having a Reichert-Meissl value of 2·5, but compare page....

The determination of the refractive index in Zeiss butyrorofractometer is of value for ascertaining the purity of cacao butter, and it serves as a control on the iodine value, for according to Roques[43] the refractive index and the iodine value stand in equal relation, so that fat having a high refractive index gives a high iodine value and vice versa. The refractive index of cacao butter ranges between 1·4565-1·4578 at 40°C. corresponding to 46-47·8 on the scala of the Zeiss butyro-refractometer. The use of the latter is recommended by Filsinger as a preliminary test for cacao butter, since with a normal refraction it is not necessary to proceed further and determine the iodine, saponification and acid values, nor the melting point. In conclusion we annex table 12, where the respective constants for different varieties of cacao butter will be found tabulated.[44]

For further information on all these methods, the reader is referred to the excellent work of R. Benedict, entitled “Analysis of Fats and Waxes”: VII. Edition, Berlin.

Table 12.
Physical and Chemical Analyses of the Various Kinds of Pressed Stollwerck Cacao Butter.

	Accra	Ariba	Bahia	Guayaquil	Cameroon
	a) Fat
Point of refraction at 40° C	64·3	46·1	46·9	46·5	46·0
Melting Point (Polenske)⁽¹⁾	33·1	33·2	31·95	32·5	33·65
Freezing Point (Polenske)	20·0	21·55	19·35	19·8	20·95
Variations⁽²⁾ between Melting Point and Freezing Point (Polenske)	13·1	11·65	12·60	12·5	12·70
Reichert-Meissl number	0·49	0·33	0·38	0·55	0·33
Polenske⁽²⁾ number	0·50	0·50	0·60	0·42	0·40
Köttstorfer number	192·4	191·7	191·4	190·8	193·2
Hübl’s iodine value	35·24	34·89	37·87	36·54	34·0
Bellier’s reaction⁽⁴⁾	violet	as 1	as 1	as 1	as 1
R. Cohn’s reaction⁽⁵⁾
a) Fresh fat⁽⁶⁾	negative	"	"	"	"
b) Rancid fat	strong positive	weak positive	positive	weak positive	positive
	b) Fatty Acids⁽⁷⁾
Refractive index at 40° C	34·60	34·55	34·50	34·40	33·70
Melting Point⁽⁸⁾	52·90	52·95	51·80	52·90	52·00
v. Hübl’s iodine value	35·88	36·27	38·78	37·78	36·02
	Puerto Cabello	Thomé	Trinidad	Fluctuations of Analyses Values
	Puerto Cabello	Thomé	Trinidad	from	mean
	a) Fat
Point of refraction at 40° C	46·0	46·8	46·3	46·0-46·9	46·4
Melting Point (Polenske)⁽¹⁾	32·7	32·95	32·9	31·95-33·65	32·9
Freezing Point (Polenske)	20·8	18·60	20·66	18·6-21·55	20·2
Variations⁽²⁾ between Melting Point and Freezing Point (Polenske)	11·9	14·35	12·30	11·65-14·35	12·7
Reichert-Meissl number	0·41	0·55	0·55	0·33-0·55	0·45
Polenske⁽²⁾ number	0·40	0·55	0·55	0·4-0·6	0·49
Köttstorfer number	191·6	191·7	191·5	190·8-193·2	191·8
Hübl’s iodine value	32·72	37·24	33·72	32·72-37·87	35·28
Bellier’s reaction⁽⁴⁾	as 1	as 1	as 1	—	—
R. Cohn’s reaction⁽⁵⁾
a) Fresh fat⁽⁶⁾	"	"	"	—	—
b) Rancid fat	opal escence+	opal escence+	opal escence+	—	—
b) Fatty Acids⁽⁷⁾
Refractive index at 40° C	33·50	34·70	33·50	33·5-34·7	34·18
Melting Point⁽⁸⁾	51·45	52·05	52·50	51·45-52·95	52·32
v. Hübl’s iodine value	33·85	39·60	36·02	33·85-39·78	36·90

Remarks 1) Exact point of liquefaction difficult to observe; therefore the average of several readings must be taken.

2) Work from the Imperial Office of Health 1907, 26, 444-463.

3) Work out of the Imperial Office of Health 1904, 20, 545-558.

4) Central Journal for Germany 1908, 36, 100.

5) Journal for Popular Chemistry 1907, 16, 308.

6) Obtained at the expiration of a four weeks’ treatment as recommended by Erlenmeyer.

7) Non-volatile fatty acids, insoluble in water, from the determination of the Reichert-Meissl number.

8) Obtained as under a). Freezing Point in various cases, 1 to 8 equals 47·8—Melting Point minus Freezing Point: 52·3-47·8 4·5.

We have already stated that there is also cacao fat in the shells, and though it only amounts to some four or five percent, it has long been the care of experimenters to recover and realise that little as fully as possible. It is commercially known as Dutch IIa or artificial cacao butter, and cannot be obtained like the fat of the kernel by mechanical means, but is obtained by some cheap solvent like benzene. The traces of benzene are very difficult to hide, and consequently this shell butter has little commercial value and its manufacture is unremunerative.

Filsinger[45] gives the iodine value of shell butter as higher than that of kernel butter, and fixes it between 39 and 40: its acid value, especially if the fat is rancid, can reach 50-60° Burstyn, i. e. 50 to 60 ccm. normal alkali for 100 grammes of fat.[46] If the free acid of shell butter be counteracted with sodium or magnesium carbonate, the neutral fat then has the normal iodine value of pure cacao butter, namely 36·5. In a sample giving an abnormally high iodine value it is always necessary to determine the acid value, and if the latter be too high, the fatty acids must be removed, when if the sample be unadulterated, the normal iodine value will be obtained. It may be noted in passing that the high acid values occurring in shell butter may be due in part to the acidity of the benzene employed as a solvent.

Cacao butter has a considerable commercial value, and is consequently liable to adulteration with many inferior fats of vegetable origin. Among these are especially beef and mutton tallow, the purified fatty acids of palm-nut oil, wax, paraffin, stearic acid, dicka fat (nucoa butter, possibly) and cocoa-nut fat, as well as the numerous preparations of the last named, variously known in commerce as Mannheim cocoa-nut butter, vegetaline, lactine, finest plant butter, chocolate butter, laureol vegetable butter, palmin, kunerol etc. Other but less commoner are the sesame cotton-seed, arachidic, margarine and hazelnut oils.

For the detection of these and similar adulterates, the reactions and analytical methods described are all-sufficient. Benedict[47] discovers that the presence of wax and paraffin considerably diminishes the saponification value, cocoa, nut fat increases it and lowers the iodine value, whereas stearic acid raises the acid value.

	Melting point °C.	Melting Point of fatty acids °C.	Iodine value
Cacao butter	30-34·5	48-52	34-37·5
Oil of Almonds	—	14	93-101·9
Sesame oil	—	26-30	106·4-109
Earth-nut (Arachis) oil	—	27-31	92-101
Hazelnut oil	—	17-25	83·2-88
Cotton-seed oil	—	38-40	106-111
Oleo-margarine	32·4-32·5	42	43·8-48·5
Beef tallow	43-49	43-46	35·4-36·5
Wax	62-64	—	8·0-11
Paraffin	38-82	—	3·9-4
Stearic acid	71-71·5	—	—
Sebin	37·6-37·8	—	43·7-43·8
Cocoa-nut fat	20-28 chiefly 26·2-26·4	24-25	8-9
	Saponification value	Acid value	Refractive index in Zeiss’s butyrometer
Cacao butter	192-202	9·24-17·9	46-47·8 at 40° C.
Oil of Almonds	189·5-195·4	—	64-64·8 at 25° C.
Sesame oil	187-192	—	67-69 at 25° C.
Earth-nut (Arachis) oil	190-197	—	65·8-67·5 at 25° C.
Hazelnut oil	191·4-197·1	—	—
Cotton-seed oil	191-197	—	67·6-69·4 at 25° C.
Oleo-margarine	195-197·4	—	48·6 at 40° C.
Beef tallow	193·2-198	—	49 at 40° C.
Wax	97-107	19-21	—
Paraffin	—	—	—
Stearic acid	195-200	195-200	—
Sebin	192·4-192·6	—	—
Cocoa-nut fat	254·8-268·4	—	35·5 at 40° C.

The presence of cocoa-nut fat can also be shown by the etherification of the fatty acids with alcohol and sulphuric acid, when the characteristic odour of the ester of cocoa-nut acid occurs. Vegetable oils, such as almond, cotton-seed, arachidic, sesame and hazelnut oils, lower the melting point of the fatty acids and raise the iodine value. Sesame oil is easily detected by Baudouin’s reaction, yielding a raspberry coloration whilst pure cacao butter keeps a fine yellow or dark brown. It is possible to detect the presence of so minute a quantity as 1% of sesame oil, by means of Baudouin’s reaction.

The following table, containing the analytical determinations of all fatty substances which can possibly be employed in the adulteration of cacao butter, will serve to facilitate reference to this subject.

In addition to its use in the manufacture of certain cacao preparations and for lubricating parts of machinery which come into contact with the cacao etc. cacao fat is also used in perfumery and especially in pharmacy for making suppositaries, ointments, etc., but it is of no importance in soap making. As an edible fat, in the true sense of the word, like ordinary butter or lard, cacao butter is not used. It has been maintained by Benedikt[48] that when in the form of chocolate it is as easily digestible in the human organism as milk fat, which is generally regarded as offering most favourable conditions for absorbtion in the intestinal canal. The digestibility of both fats varies from 92·3 to 95·38 percent, and both, in this respect, stand very near to cocoa-nut fat from which the solid glycerides have been removed, and to ordinary butter, the former according to Bourot and Jean.[49] being digestible to the extent of 98 and the latter 95·8 percent.

Cacao butter is obtained as a by-product in the preparation of cocoa powder and in every country where cocoa powder is produced there is always a large trade in the former article. That is, apart from Germany, especially the case in Holland, where the monthly supply to the Amsterdam market is so large that during 1899 one firm alone—Van Houten—had 855 tons for sale. The average price of late years has considerably increased, and is now about 64-73 cents per kilogramme.

3. Cacao-red or Pigment.

The majority of investigators interested in the cacao bean have assigned its peculiar aroma and taste to the cacao-red which it develops. As previously pointed out, the young fresh bean is colourless, the pigment forming later, as can be observed in many vegetable colouring materials, such as oakand cinchona-red, madder, indigo and kola-nut red (from Sterculia acuminata). As the later investigations of Hilger[50] have shown, the fresh colourless cacao bean contains a diastasic ferment, as well as a glucoside body, which C. Schweitzer[51] has termed glocoside or cacaonin. The term glucoside may be noted in passing as including those bodies, the greater number of which occur in plants, and which by treatment with alkalis, acids or ferments are split up into an indifferent body and a sugar, generally glucose. These bodies may be chemically regarded as ethyl derivatives of the respective sugars. When the ripe, white seeds are dried, the cacao-glycoside is partly decomposed by the agency of the above-mentioned diastasic ferment and formations of grape sugar, pure non-nitrogenous cacao-red, together with theobromine and coffeine ensue. These substances, and likewise a certain amount of undecomposed cacao glycoside, can all be detected in the seed, which has by this time acquired a brownish to violet colour.

The unfermented bean, according to Schweitzer, has as much as 0·6% unaltered glucoside. Fermentation produces the same effect as drying, as here again the glycerine is not completely split up, for the cacao-red, isolated in the ordinary way, consists according to Hilger of a mixture of pure non-nitrogenous cacao-red and some glycoside.

The complete decomposition of the cacao glycoside can only be effected in a chemical manner, by boiling the finely divided and defatted seeds with dilute acids, a method which has made it possible to effect an exact determination of the diureides, as the treatment with acid sets free the totality of their theobromine and coffeine.

Schweitzer regards the molecule of cacao glycoside as an ester comprised of one molecule of non-nitrogenous cacao-red, six molecules of starch-sugar and one molecule of theobromine with double-sided attachment and having the hypothetrical formula C₆₀H₈₆O₁₅N₄.

Before the appearance of Hilger’s researches, all statements of a chemical nature respecting cacao-red related to a mixture of a pure non-nitrogenous pigment and the glycoside, which must in all cases be preliminarily obtained, before the pure pigment can be prepared. That can be done[52] by treating the roasted beans with petroleum ether, which removes the fat and part of the free theobromine then with water, to extract the remaining theobromine, coffeine, sugar and salts, and finally with alcohol, to extract the cacao-red. The alcoholic residue is then quickly dried on porous plates. The material thus obtained is a reddish brown amorphous bitter powder, which is scarcely soluble in water, easily so in alcohol or in dilute alkali, and is reprecipitated by acid from its alkaline solution. It gives a sublimate of theobromine when heated. When the substance is distilled with 5 percent of sulphuric acid, the added glycoside is completely decomposed into sugar, theobromine and the real cacao-red, which latter is represented by the formula C₁₇H₁₂(OH)₁₀. It appears to stand in near relation to tannin, which it resembles in yielding formic acid, acetic acid, and pyrocatechin by the action of caustic alkalis. The pure non-nitrogenous cacao-red, at present, is of exclusively scientific interest; for practical purposes only the crude cacao-red, cacao-red glycoside, as naturally existing in the bean, is of importance. The better and the more effectual the manner in which the beans have been prepared by fermentation, the more intense is the formation of the cacao red, especially its localisation in the cells and cell tissues. This is the reason that the variations in colour of different kinds of bean and the aqueous extracts which they yield are so distinct.

Especially is this noticeable in carelessly dried beans, in which the cotyledon tissue is of a dirty brown or yellow colour instead of being brown or violet; the pigment here is not restricted to separate cells but has the appearance of having penetrated into the contiguous albuminous cells. The bean contains 2·6-5 percent of the crude cacao-red; it is soluble in alcohol and in ether and partly so in hot water, and is completely extracted from the bean by weak acetic acid.

The crude cacao-red can be determined quantitatively by precipitating its solution with lead acetate, decomposing the lead precipitate with sulphuretted hydrogen and evaporating the filtrate containing the cacao-red to dryness.

The aqueous extract of the beans, which contains the cacao-red, is coloured greenish brown by alkalis, red by acids; acetates give a grey to yellowish colour; tincture of iodine, stannous chloride and mercurous nitrate give a rose to brown precipitate. Iron and copper salts produce grey precipitates which gradually become brown to black. Gelatine solution, containing alum, and albumin give copious yellow precipitates.

Stains produced on linen by the colouring matter of cacao-red can be removed by treatment with hot water and finally bleaching with a solution of sulphurous acid.

4. Theobromine.

All those materials which are regarded as stimulants, like coffee, tea, cacao, tobacco etc., owe their action to peculiar nerve stimulating bodies, which are present only in small quantity in the seeds or leaves of the respective plants and are termed by chemists alkaloids and diureides.

The physiologically active constituents of tea, coffee and cacao are considered, even up to to-day, by many authors as alkaloids or organic bases and especially ranked among the xanthine or purine bases. Recent investigations, however, separate these substances from the alkaloids in the strict sense and comprise them within a particular group of urea derivatives under the designation of ureides; the ureides of tea, coffee and cacao representing two molecules of urea, they are to be qualified as “diureides

A bitter substance in the cacao bean had already been observed by Schrader, but Woscressensky[53] in 1841 was the first to isolate the diureide, theobromine.

Theobromine is found in the unfermented and fermented beans in two forms; as free theobromine, which has been eliminated from the glucoside by the ferment in the drying and fermenting processes, and in combination with glucose and cacao-red as a glucoside, from which it can only be separated by chemical means.

Theobromine stands in near relation to caffeine, the diureide of tea and coffee, as will be seen from their chemical formulae—in which theobromine is shown to contain one methyl group CH₃, less, its place being taken by an hydrogen atom;

Caffeine		Theobromine
C₅HN₂O₃(CH₃)₃,		C₅H₂N₂O₃(CH₃)₂,

so that in all, theobromine falls short of caffeine by only one radical. Strecker[54] was the first to show the relation between the two substances, when he succeeded in converting caffeine into theobromine by the action of methyl oxide on silver theobromine for 24 hours at 100° C. Caffeine and silver iodide are then formed and can be separated by treatment with alcohol, which dissolves the caffeine, leaving the silver iodide undissolved.

E. Fischer[55] was shown the relation of theobromine and caffeine to uric acid by artificial synthesis of both substances from derivatives of both. Fischer, starting with monomethyl pseudo-uric acid, converted it into 7-methyl uric acid by distilling it with hydrochloric acid, and afterwards, by treating the lead salt of the latter with methyl iodide and ether, produced 3-7-methyl-uric acid. That acid was converted into dimethyldioxychlor-purine by treatment with a mixture of phosphorus oxychloride and phosphoric penta-chloride, with subsequent reduction into 3-7 dimethyl-6-amino-2-oxy-purine, from which, by the action of nitrous acid with loss of the amine group, theobromine was finally obtained. The synthesis of theobromine is a brilliant exploit of Fischer’s, and it is quite possible that at no distant period, when a simple and cheap method of production has been arrived at, synthetical theobromine will appear commercially as a rival of the natural product. At present there is no prospect of this being immediately realised, and cacao shells from which theobromine is now prepared are as yet in no danger of displacement by the new substitute, but still serve as a useful by-product in the manufacture of cacao.

Theobromine and caffeine, like the alkaloids or plant bases, have a distinct physiological and even toxic action if taken in too large quantities.

From the experiments of Mitscherlich it appears that theobromine has a similar action to caffeine, but is somewhat less active owing to its being less soluble in the gastric juice. Mitscherlich’s experiments with frogs, pigeons and rabbits show that 0·05 grammes killed a frog in 40 hours, 0·05 grammes a pigeon in 24 hours, and 1 gramme a rabbit in less than 20 hours. Death resulted in all cases from cramping of the spinal cord, producing either convulsions or subsequent paralysis.

The results of these experiments do not detract from the nutritive value of cacao, since the human organism requires ten times as much theobromine as rabbits to exhibit the slightest toxic symptom; in cacao mass containing 1 % not mentioned in discussion; just a head’s up to PP for S&R] theobromine, that would involve the consumption of 5 lbs. averdupois of chocolate at once, a practical impossibility. Similar conditions prevail in connection with the use of tea, coffee, and especially tobacco, where symptoms of poisoning have been occasionally noticed (the nicotine peril of excessive smokers) but it would seem that cacao and chocolate are the most favourably placed of these stimulants as regards such toxic action. It appears from the experiments of Albanese[56] Bondzynski, Gottlieb[57] and Rost[58] that 3 percent of the theobromine administered passed out in the urine unaltered, whilst on the other hand 20-30 percent of that decomposed in the organism is found again as monomethyl-xanthine.

The larger proportion of the monomethyl xanthine is heteroxanthine (= 7 Methyl-X) and the inferior 3 Methyl-X. The excretion of theobromine appears to be closely connected with the quantity of urine voided, which is especially increased by the administration of theobromine. Since 1890, as a result of W. v. Schröder’s[59] observations in 1888, that property of theobromine has had an extended application in practical therapeutics; theobromine has been used as a diuretic in kidney diseases, and, unlike all similar medicinal agents, it exercises no influence on the heart, a circumstance which essentially increases its therapeutic value. It can be employed for medicinal purposes, either uncombined or in the form of salicylate, acetate and certain double compounds, as sodium or lithium and theobromine salicylate or acetate.

The double compounds known as diuretin, agurin and uropherin are freely soluble in water and are therefore more readily absorbed into the system than pure theobromine, which is only with difficulty soluble in water. Through the establishment of theobromine as a medicinal agent, for which we are indebted to Chr. Gram[60] and G. See,[61] cacao husks, hitherto a waste product in the manufacture of cacao, have become of value for the preparation of theobromine, in which many of the largest German chemical factories are now engaged.

Fluctuations as regards the percentage of theobromine in the beans are so extraordinary that they can only be ascribed to the lack of prescribed and definite modes of procedure in fermenting, which obviously necessitates differences in the resulting products.

Eminger found from 0·88-2·34 percent of theobromine in the examination of a rather considerable number of commercial kinds of cacao beans and in the husks 0·76 percent of the diureide: C. C. Keller[62] has also found it in the leaves and in the pericarp. Cacao contains 0·05 to 0·36 percent of caffeine.

Theobromine is a permanent white powder, appears under the magnifying glass as small, white, prismatic or granular crystals. At first it has only a slightly bitter taste, which becomes more intense when it is kept in the mouth for some length of time; and indeed, the bitter taste of the cacao bean and its preparations is mostly due to theobromine. It sublimes at 220 ° C. without melting. This phenomenon explains why the over roasted bean, that is, the kernel of beans which by accident have been heated to more than 130-150 ° C. is poorer in theobromine than the husks. When heated to 310 ° C. theobromine melts to a clear liquid which re-crystallizes on cooling.

One part of absolutely pure theobromine dissolves according to Eminger in 736·5 parts of water at 18 ° C., in 136 parts at 100 ° C. in 5399 parts alcohol (90 %) at 18 ° C. in 440 parts at boiling (90 %) point and in 818 parts of boiling absolute alcohol. It dissolves in 21000 parts of ether at 17 ° C. in 4856 parts of methyl alcohol at 18 ° C. in 58·8 parts of chloroform at 18 ° C. and in 2710 parts of boiling chloroform[63]. Theobromine is partly decomposed by strong alkalis but by cautious addition of alkalis it forms compounds with them, which, are readily dissolved by solutions of sodium salicylate, acetate or benzoate. These double compounds under the name of diuretin, agurin and uropherin have lately become of therapeutic value.[64]

Sodium silicate and more particularly trisodiumphosphate according to Brissemoret[65] are great solvents of theobromine. One and a half molecules of the latter salt can dissolve one molecule of theobromine so that in this way it is possible to prepare a solution of nearly 2 percent. Phenol also dissolves a large quantity of theobromine, according to Maupy,[66] who has utilised this property for the determination of theobromine. The defatted cacao preparation is moistened with water and extracted with a mixture consisting of 15 percent of phenol and 85 percent of chloroform.

Theobromine, like caffeine, gives the so called murexide reaction when evaporated with chlorine water—forming amalic acid—and when a watch glass previously moistened with a little fluid ammonia is held over the last few drops at the end of the operation. The residue thus obtained has a violet colour, which serves to distinguish theobromine readily from other plant bases which do not belong to the xanthine group.

Although theobromine is the most valuable constituent of cacao beans, the importance attached to a greater or lesser amount in the beans as a commercial article was formerly much exaggerated.

The investigations of Dragendorff and others have shown that the value of various stimulants like tobacco, coffee and tea, does not entirely depend on the amount of alkaloid or diureide but partly also on the joint action of all the constituents of those articles, and it is particularly the aromatic bodies which determine their commercial value. Various kinds of coffee, for example, of inferior commercial value contain considerably more caffeine than the costly Mocca beans. The highly prized Havana tobacco ranges lower than the Sumatra kinds in nicotine content, and the same conclusion with regard to cacao would probably be correct. In support of this view, attention may be directed to the following analyses performed by Wolfram.[67]

Percentage of theobromine at 100° C.

Description	% Theobromine %
Caracas	In the bean	1·63	In the shells	1·11
Guayaquil (of considerably less value than the first)		1·63		0·97
Domingo		1·66		0·56
Bahia		1·64		0·71
Puerto Cabello (fine kind)		1·46		0·81
Tabasco		1·34		0·42
	Average	= 1·56%		= 0·76%

Excluding the theobromine in the shells which are not used in the preparation of cacao, it will be seen from the above table that the Caracas bean, which is the finest and dearest, has an amount of theobromine which is only equal to, or even a little less, than that in the inferior beans from Guayaquil and Domingo.

5. Albumin.

On the presence of albuminous bodies in the cacao bean, varying between 14-15 percent, depends to a great extent its nutritive value. The albumin in plants, unfortunately, is not to hand in a form suitable for direct absorption and assimilation in the animal organism, in fact, only a fraction of it is so available. Before considering the nutritive value of the albumin of the cacao bean it will be well to give attention to the general chemical and physical properties of albumin so far as a knowledge of them will assist in the elucidation of the subsequent matter.

Albuminous bodies or proteins occur either dissolved in the sap of plants or in a solid in the protoplasm of plant cells; also in the form of granular deposits (Aleuron granules[68]). In cacao they are apparently present in the three different conditions.

The term vegetable albumen, in its more restricted sense, is meant to designate a protein substance which is soluble in water and is coagulable by heat. The greater part of the proteid which exists in the seeds and sap of plants and is coagulable by heat, is not albumin but globulin, that is to say, it is insoluble in water, though dissolved by solutions of neutral salts. Whilst many protein substances in aqueous solution require a temperature of 100 ° C. before coagulating, or becoming insoluble under certain conditions, others coagulate at 65 ° C. Concentrated acetic acid dissolves all albuminous bodies with the aid of heat, concentrated nitric acid gives a yellow coloration (xantoprotein reaction). Albuminous substances are decomposed when heated to 150 ° C. developing a dark colour, swelling up and evolving an offensive smell, finally leaving behind a difficultly combustible coaly residue.

Globulins combine with aqueous solutions of alkalis such as potash, soda, ammonia etc. producing alkaline albuminates; with acids they form acid albuminates or syntonins. Both have the property in common, that whilst they are insoluble in pure water, they readily dissolve in slightly acidulated or alkaline water, as well as in weak saline solutions, and are then no longer coagulable by boiling.

Albuminous bodies are converted first into albumoses (proteoses), and then into peptons by gastric and intestinal digestion or by hydrolytic decomposition with acids or alkalis, also by the action of steam under pressure of many atmospheres, as well as by putrefaction. Albumoses, with the exception of hetero-albumose, are soluble in water. Peptons dissolve entirely and in that condition are absorbed by the animal organism.

Albumins are precipitated from their solutions by strong alcohol, and in that way Zipperer succeeded in precipitating 4·25 percent of albumin from the aqueous extract of Trinidad cacao, which corresponds to about 25 percent of the total amount of albumen in the bean.

The results of his investigation have shown that generally more soluble albumen is present in the unfermented than in the fermented bean. Consequently, it would appear that in the finer kinds of cacao beans, in which very careful fermentation has been carried out, the albumin, owing to fermentative alteration, is rendered less soluble.

The constitution of albumin is still not sufficiently known, despite the excellent experiments of E. Fischer on this subject; generally it is regarded as having the formula:

C	52·31-54·33%
H	7·13- 7·73%
N	15·49-17·60%
S	0·76- 1·55%
O	20·55-22·98%

Accepting a mean formula corresponding to the above figures as representation of the albumen (namely C₇₂H₁₁₂N₁₈SO₂₂), it becomes possible to obtain a quantitative determination of this constituent in the plants in which it is contained. There is, for instance, 16 % of nitrogen here. Starting from such a standpoint, and determining the percentage of Nitrogen contained in a plant, and multiplying by 6·25 (i. e. 16 %), the amount of albumen is obtained. For further particulars see paragraph 4. The albumen in cacao, as previously mentioned, is in the form of globulin, that is, in a less soluble form. In cacao preparations which are required for invalids, especially those with affections of the stomach, it is important to have the albumen in a more readily soluble condition. Various attempts have been made with cacao preparations to obtain that result, and later on, full illustrations and explanations will be given on this subject. First of all, however, it is desirable to consider the scientific methods employed to ascertain the relative digestibility or indigestibility of albumen.

Professor Stutzer[69] of Bonn has been engaged in determining the action of digestive ferments of the animal organism on alimentary substances, and has worked out a method by which it is possible to ascertain the proportion of albuminous substances which can be regarded as digestible.

The method depends upon the fact that salivary, gastric and intestinal digestion can be artificially imitated in the laboratory. But as the salivary secretion only digests starch and is difficult to obtain, malt diastase, which serves the same purpose, is used instead. On the other hand albuminous material is only digested by juices of the stomach and intestines as fresh obtained from the mucous membranes of the pig or ox. If we suppose an average of 16 percent of total albumen in cocoa powder, the following results would probably be given by Stutzer’s method:

Of 16 % of total albumen there are on an average:

Albumen:	corresponding to percentage of the total mass:
7·6% soluble in the stomach	47·5%	}65%
2·8% soluble in the intestines	17·5%	}65%
5·6% insoluble	35·0%
16·0%	100·0%

As shown by the experiments of Forster[70] however, artificial digestion does not correctly represent the actual consumption of nutriment in the human body. Forster’s experiments, in which cacao powder was administered to healthy men, gave a much higher value, in fact, 80 percent of the nitrogenous substance was digested, against 65 percent by Stutzer’s artificial method of digestion. The results obtained by artificial digestion must therefore be increased in that proportion.

6. Starch.

Starch is one of the most important constituents of cacao, as on the starch taken in conjunction with the fat and albumen depends the nutritive value of the cacao bean. As previously stated, cacao starch is one of the smallest kinds which occur in the vegetable kingdom; consequently it can easily be distinguished from the starch granules of other plants. Owing to their minuteness the concentric rings showing the stratified structure of the starch granules can only be distinguished with difficulty under the microscope. Cacao starch consists usually of globular granules, generally separate, but sometimes in aggregations of two or three. The appearance under the microscope of the starch granules is clearly shown in fig 7, which represents a section of Ariba cacao enlarged 750 times.[71]

Fig. 7.

a on the above represents the intercellular spaces, b the cell walls, c the starch granules, d the fat crystals, those being the contents and structural elements of the cacao cell that the microscope will at once distinguish.

Cacao starch has the usual properties of ordinary kinds of starch, namely:

1. It is gelatinised by hot water, that is to say, the water penetrates between the layers of starch granules, separating them and causing by its penetration a swelling up of the starch whereby a transparent mass know as “starch paste” is produced. It has been supposed that cacao starch is less easily gelatinised than the starch of other plants. According to investigations of Soltsien’s[72], which Zipperer unreservedly endorses, this is not the case, for under certain essential conditions, cacao starch gelatinises just as readily as other kinds of starch.

The blue coloration of starch with iodine.

This is said to take place more slowly with cacao than with other starches, though we have always found that once the cacao starch is gelatinised, a blue coloration appears immediately on adding a sufficiently strong solution of iodine.

There are certainly other materials in the cacao bean, such as fat, which by more or less enveloping the starch, prevent access of water to the starch granules and thus hinder gelatinisation; or again, the albumen and cacao-red may exert some retarding influence on the iodine reaction, especially if the iodine solution used is very dilute. Yet it is impossible to describe the reaction as slow.

According to Soltsien, if a mixture of two parts of cacao bean with one part of calcinated magnesia and water is heated, a clear-filtering decoction is obtained, which immediately assumes the blue colour on addition of iodine solution. On neutralising the filtrate with acetic acid, and adding 3-4 parts of strong alcohol, its starch is precipitated.

By boiling with dilute acids as well as by the action of ferments like the saliva, diastase etc., starch is converted into starch sugar (glucose, dextrose). The empirical formula for starch is C₆H₁₀O₅, that for starch sugar is C₆H₁₂O₆, so that in the conversion one molecule of water is introduced, wherefore its chemical nature is greatly changed, and especially in its becoming freely soluble in water. That alteration allows of starch being quantitatively determined, as the dextrose thus produced has the property of reducing an alkaline solution of copper sulphate (known as Fehling’s solution, after the discoverer); that is to say, the copper sulphate is converted into insoluble red cuprous oxide. As dextrose always precipitates a definite amount of cuprous oxide, the quantity of starch present can in that way be determined.

The chemical determination of starch is only in a limited degree effectual in the recognition of an admixture of foreign starch in cacao preparations. If more than 10-15 percent of starch (calculated on the crude bean) has been found, then it must be assumed that there has been an admixture of foreign starch, but chemistry affords no means by which foreign starch can be distinguished from the genuine starch of the cacao bean. For that purpose the foreign starch must be minutely observed under the microscope, which not only serves to detect its presence, but gives an approximate estimation of the amount present, and its origin. Great caution should be exercised, or the result may be easily exaggerated.

7. Cellulose or crude fibre.

We have already made the acquaintance of this material as the chief constituent of the cell walls and vascular tissues. Recent chemical investigations have shown that it consists of the anhydrides of hexose and pentose (sugar compounds) incrustated with many impurities, such as cacao-red, gum, mucilage etc. From a chemical point of view, cellulose has the same formula as starch, viz. C₆H₁₀O₅, or one of its multiples represented in formula. One of its chemical properties is solubility in ammonio-cupric sulphate, and affinity for alkalis such as potash, soda, ammonia, causes it to swell when they act on the cell fibres.

Weender’s process[73] as worked out by Henneberg is the one usually adopted for the determination of crude fibre in plants, although recently H. Suringar, B. Tollens[74] and more particular König[75] have pointed out that in Weender’s process the so-called pentosan, that is to say, the sugar-like constituent of the composition C₅H₁₀O₅, which comprises a not inconsiderable portion of the crude fibre, undergoes a disproportionate alteration, so that the analytical results thus obtained can by no means give an accurate representation of the amount of cellulose. The crude fibre must therefore be treated in such manner as to eliminate the pentosan. For this purpose the various methods of König, Matthes and Streitberger have been proposed, to which we shall return in Book 4. Filsinger, the meritorious experimenter on the subject of cacao, has by König’s method determined the amount of crude fibre in a series of different varieties of cacao bean, and obtained the following results as regards shelled and roasted beans.

	percent
1. Puerto Cabello	5·37
2. Java	3·97
3. Ariba Guayaquil I	4·10
4. Ariba Guayaquil II	4·07
5. Machala Guayaquil I	4·43
6. Para	4·01
7. Surinam Guiana	3·01
8. Bahia	2·81
9. Grenada	3·10
10. Guatemala	3·50
11. Machala Guayaquil II	3·58
12. Caracas	3·65
13. Samana	4·58
14. St. Thomé A I	4·13
15. St. Thomé A II	2·95
16. St. Thomé B	3·15
17. Haiti	3·12[76]

These new values may be provisionally regarded as normal. From these results not only can an idea of the functioning of the cacao shelling machine be obtained, but also the presence of any occasional admixture of husk in cacao preparations may be inferred, since the husk contains a great deal more crude fibre than the kernel. Therefore the determination of the crude fibre is an important item in the testing of cacao preparations, as there is no doubt that the presence of vegetable substances rich in crude fibre can be detected by the increase in the amount of cellulose.

8. Sugar and plant acids.

The presence of glucose in raw cacao beans was first pointed out by Schweitzer[77]. The sugar is formed by the action of the cacao ferment on the glucoside cacaonin during the processes of drying and fermentation. In addition to sugar, malic and tartaric acids have been observed. These substances, however, are only of interest to the plant physiologist and not to the manufacturer, so it is sufficient merely to notice them here in passing.

9. The mineral or ash constituents.

When cacao beans are ignited, the constituents of an organic nature are volatilised and only the non-volatile or inorganic constituents remain behind. These consist of potash, soda, lime, iron magnesia, combined with silicic acid, phosphoric acid, sulphuric acid and chlorine.

The amount of ash in raw and shelled cacao beans varies from 3-4 %. Tuchen[78] found 2·9-3 %, Trojanowski[79] 2·08-3·93 %, Zipperer[80] 2·7-4 %, L’Hote[81] 2·2-4 %, H. Beckurts[82] 2·20-3·75, J. Hockauf[83] 2·84-4·4 percent. Of those kinds which are now most in use, Ceylon gave 3·30 percent, Java 3·20 and Kameroon 2·95 percent. (Beckurts).

Quantitative analyses of the ash of the cacao beans have been made by several investigators, and the following table gives a series of the most complete analyses, made by R. Bensemann[84].

Table 14. Analysis of the ash of Cacao Beans by R. Bensemann.
The ash of the kernel free from husk dried at 100°C. contained:

Key to Column Headings

B = Maracaibo
C = Caracas
D = Trinidad
E = Machala
F = Porto Cabello
G = Mean

Insoluble respectively in dilute hydrochloric or nitric acid		B	C	D	E	F	G
a)	Volatile dessicated at 100° C.	0·142	0·076	0·144	0·074	0·198	0·127
b)	Fixed at red heat	0·312	1·663	0·553	0·630	1·075	0·846
	Soluble in dilute hydrochloric or nitric acid:
c)	Potassium oxide K₂O	35·889	33·844	30·845	30·686	29·989	32·251
d)	Sodium oxide Na₂O	0·515	0·766	1·964	4·173	3·427	2·169
e)	Calcium oxide CaO	4·118	5·030	4·638	3·112	2·923	3·964
f)	Magnesium oxide MgO	15·750	15·151	16·060	16·172	17·562	16·139
g)	Ferric oxide Fe₂O₃	0·182	0·217	0·491	0·629	0·303	0·364
h)	Aluminium oxide Al₂O₃	0·080	0·326	0·490	0·432	0·305	0·327
i)	Silicic acid SiO₂	0·214	0·211	0·169	0·134	0·240	0·194
k)	Phosphoric anhydride P₂O₅	27·741	29·302	28·624	37·000	35·274	31·588
l)	Sulphuric anhydride SO₃	2·632	2·740	3·957	2·042	3·952	3·065
m)	Chlorine Cl	0·295	0·341	0·427	0·279	0·085	0·285
n)	Carbonic anhydride CO₂	10·349	8·435	8·953	2·788	3·481	6·801
o)	Water H₂O	1·847	1·975	2·781	1·912	1·205	1·944
	Oxygen O equivalent to chlorine	0·066	0·077	0·090	0·063	0·019	0·064

In previously describing the aleuron granules of the cacao bean it was mentioned that they contain a comparatively large globoid. According to Molisch[85], when sections are cautiously heated on platinum foil, these globules are found in the ash. From their number they give a characteristic appearance to the ash of cacao beans, and thus may serve as a good means of identifying cacao, since they can be detected in the smallest quantity of a genuine cacao preparation.

A noteworthy fact may here be mentioned, namely the presence of a rather small amount of copper in the ash of cacao beans as well as the husks. Duclaux[86] was the first to point out this fact, which several other observers, such as Skalweit[87] and Galippe[88] have also confirmed. The amount of copper in the husk varies from 0·02 to 0·025 percent and in the beans from 0·0009-0·004 percent (Duclaux). Copper in similar amount is found in all kinds of beans and husks, and its presence is due to the absorption of copper by the plant from the soil, whence it gradually accumulates in the fruit.

b) The Cacao Shells.

Most of the constituents which exist in the cacao kernels are also to be found in the husks and the methods for isolating and determining them are the same in both cases. The composition of the husk, according to Laube and Aldendorff[89], is as follows:

Table 15.

Key to Columns

B. Amount of husk
C. Water
D. Nitrogenous substance
E. Fat
F. Non nitrogenous extractive
G. Woody fibre
H. Ash
I. Sand

	B	C	D	E	F	G	H	I
	Per cent
Caracas	20·09	7·74	11·68	5·99	35·29	12·79	8·32	18·62
Guayaquil	—	9·11	12·94	10·75	47·08	13·12	6·79	0·21
Trinidad	14·04	8·30	15·14	4·23	46·05	18·00	7·06	0·92
Puerto Cabello	14·92	6·40	13·75	4·38	47·12	14·83	6·06	7·46
Soconusco	18·58	6·48	19·12	6·48	39·39	15·67	8·15	4·71
Mean	16·33	7·83	14·29	6·38	45·79	14·69	7·12	5·90

Zipperer’s analysis[90] of the unroasted husks gave the following results:

Table 16.

Key to Columns

B. Surinam
C. Caracas
D. Trinidad
E. Puerto Cabello
F. Machala
G. Port au Prince

	B	C	D	E	F	G	H	I
	Per cent
Moisture	13·02	11·90	13·09	12·04	—	—	—	12·51
Fat	4·17	4·15	4·74	4·00	—	—	—	4·23
Cacao tannic acid soluble in 80% alcohol	5·10	3·80	4·87	9·15	—	—	—	4·58
Theobromine	0·33	0·30	0·40	0·32	—	—	—	0·33
Ash	7·31	16·73	7·78	8·99	—	—	—	10·20
Woody fibre	14·85	17·99	18·04	15·98	—	—	—	16·71
Nitrogen	—	2·25	2·13	—	—	—	—	2·19
Proportion of husk in the raw seeds	14·60	15·00	14·68	12·28	16·14	16·00	18·68	15·34

Roasted cacao husks contain according to G. Paris[91] the following constituents:

Moisture 12·57 percent, nitrogenous substance 14·69 percent, fat 3·3 percent, extractives 45·76 percent, crude fibre 16·33 percent and ash 7·35 percent.

50 grammes of the husks when boiled with 500 grammes of water give 25·08 percent extract, 20·68 % organic substance, 4·4 % ash, 0·21 % sugar (reducing substance), 0·79 % theobromine, 0·12 % percent acid, calculated as tartaric acid.

The following constituents have been found by R. Bensemann[92] in the ash of cacao husks:

Table 17[93].

	Maracaibo	Caracas	Trinidad	Machala Guayaquil	Porta Plata
	Per cent
Ash dried at 100° C.
I. insoluble in dilute hydrochloric or nitric acid:
a) Volatile dessicated at 100° C.	0·113	0·421	0·979	0·306	1·247
b) Fixed at red heat	1·917	47·711	29·315	37·662	51·513
II. Soluble in dilute hydrochloric or nitric acid:
c) Potassium oxide K₂O	31·517	11·812	25·866	23·117	12·174
d) Sodium oxide Na₂O	4·188	3·298	2·726	1·210	2·780
e) Calcium oxide CaO	10·134	4·458	5·097	3·503	4·401
f) Magnesium oxide MgO	9·546	4·703	5·206	4·837	4·090
g) Ferric oxide Fe₂O₃	0·647	0·931	0·339	0·958	0·462
h) Aluminium oxide Al₂O₃	0·281	1·554	0·710	1·854	1·046
i) Silicic acid SiO₂	1·180	7·975	2·416	4·321	6·780
k) Phosphoric anhydride P₂O₅	9·068	7·630	4·703	7·288	7·242
l) Sulphuric anhydride SO₃	3·041	1·478	3·398	1·741	2·012
m) Chlorine Cl	1·005	0·220	1·022	0·255	0·444
n) Carbonic anhydride CO₂	25·454	5·399	16·290	11·834	4·247
o) Water H₂O	2·135	2·499	2·263	1·171	1·662
p) Oxygen O equivalent to chlorine	0·226	0·049	0·290	0·057	0·100

As evidenced in the preceding examples, data as to the constituents of the cacao husk deviate considerably with different authors. Laube and Aldendorff, for instance, found 14-20 percent, while Zipperer obtained 12-18 percent of husks.

These discrepancies are mainly due to adhering sand and ferruginous earth collected during the drying and fermenting processes. If the beans are carefully collected and kept free from earthy substances, the percentage of husks as against that of the bean will appear much lower; it is, indeed, now possible to obtain properly treated beans which contain on an average only some 10 percent of husks, such as Ariba and Machala. The husks of these two varieties are exceedingly woody, and their amount sometimes reaches 15 per cent. The latest machinery for cleaning the beans effects so complete a separation of the husks from the kernel that very little of the former remains in the finished cacao preparation (less than 1 percent in thin-shelled beans and no more than 2 percent in thick-shelled beans such as Ariba). For some years it was not possible to effect so thorough a removal of the husk, so that there was always found an appreciably large amount of shells in the finished preparations, which rendered it difficult to detect adulteration. As, however, the quantity of ash present in the husk is double that in the kernel, it was possible to form an opinion as to the intentional admixture of shells from the increase of ash in cacao preparations. Hence the ash was always required to be determined when adulteration was suspected. Under existing conditions the addition of a quantity of shells sufficient to increase the percentage of ash present in the powder or chocolate is scarcely practicable, so that, for the purpose of detecting small additions, other methods must be resorted to, such as the estimation of the crude fibre or silica in the ash[94] with the aid of the microscope, in which it is possible to easily distinguish the forms of the cotyledon (kernel) mass and those of the husk. The diagram on page 14, Fig. 3, clearly shows the elementary forms of the cacao husk as represented by Mitscherlich. It illustrates a longitudinal section of the husk of Bahia beans, enlarged about 500 times, with six different cell elements in alphabetical order. First the compressed cells of the epidermis are to be seen on the exterior, in several parallel series and succeeded by moderately broad and thin-walled cellular tissue of the parenchyma, which sometimes presents large empty spaces (sch) the results of the loosening of the cell walls through the formation of mucilage. This cellular tissue (lp) is also permeated by bundles of spiral vessels (gfb), which, with the dry cells, are characteristic of the husk, as they exist only in very small quantity in the kernel. Then follow parallel rows of cells (lp) resembling epithelial cells; next comes a layer of cells with thick walls, the dry cells (st) and finally several rows of elongated ones (lp). The silver membrane (is) interposes between the husk and the kernel, fragments of which remain adhering to the shell after separation of the latter.

To conclude, we find that the husk of the cacao bean consists of the inner coat of fruit, called endocarp and other parts of the fruit covering, as well as the skin of the seed[95]. The following layers may be distinguished;

1. The pulp, (f in fig. 3) fragile large cells with frequent hiatus;

2. the endocarp (fe), a single layer of fragile, very narrow and irregularly arranged cells, but without hiatus;

3. the epicarp, or skin (se), polygonal and extended cells, with an outer wall of some thickness.

4. the parenchyma or cellular tissue (lp), consisting of large and multiform cells, with vascular bundles (gfb), the large mucilagenous or slime cells (sch) and

5. the sklerogenous or dry cells (st), a single layer of vessels shaped like a horseshoe, and thickening towards the interior, and in conclusion

6. the silver membrane (is), belonging to the earlier inner coat of fruit, and consisting of two single rows of fat-bearing cells.

In examination of the husks of the plane surface enlarged 160 times (fig. 8), it will be noticed that the characteristic epidermis (ep) consists of large and rather elongated but irregular polygonal cells. Frequently on the epidermis may be remarked a delicate network of the cells constituting the fruit pulp (p). Beneath the epidermis lies a very delicate transverse cellular layer (qu) followed by the parenchyma, as already stated. The remaining elementary forms are not readily observed on a plane surface but only in section, though we adjoin a few diagrams, showing the layers as isolated from the pericarp; namely, fig. 9 parenchyma, a layer of sklerogenous cells, fig. 10, and the silver membrane (is) with two superjacent Mitscherlich particles (tr) in fig. 11.

Fig. 8.

Fig. 9.

Fig. 10.

For microscopical examination, the husk must first be defatted with petroleum or ordinary ether and then treated with dilute chloral hydrate (8: 5) to assist the definition of the forms. An approximate estimation of the amount of husk in a cacao preparation can be made by means of the microscope, adopting Filsinger’s[96] levigation method, which consists of concentrating those elements of the cacao which are seldom seen even in suspension in water, and which sink to the bottom when repeatedly stirred in that liquid. To these belongs first of all the husk, and its presence and determination in the levigation method is accordingly greatly facilitated. The details of the method will be further described in treating of husk admixtures in cacao preparations.

Fig. 11.

Cacao shells are the only by-product in the cacao industry, and have been developed and exploited to such an extent, that a rational utilisation of the ever increasing quantities has become a matter of urgent necessity. They are not used in our industry, for an admixture of husk is not permissible, even in the inferior kinds of chocolate or cocoa powder, but must be regarded as an adulteration. It is true that they have been brought on the market as cocoa tea, and again, have been coated with sugar, to make them tasty; and to this day, candied husks constitute a favourite sweetmeat of the population of East Germany. But in this way only comparatively inferior quantities of the by-product were absorbed, and consequently projects of all kinds have been suggested to use up larger percentage. As we have seen, the fatty contents of the bean can be extracted with benzine, and there is a resultant 4 or 5 percentage of fat of inferior value, which is commercially known as “Dutch IIa Cacao Butter”; the defatted shells can be further used for the preparation of theobromine, as Zipperer has already noted in the first edition of this book.

Kathreiner’s successors in Munich[97] employ an extract of cacao shells prepared with hot water, in order to improve coffee berries during the roasting and to give a flavour to the coffee substitutes prepared from corn and malt. Cacao extract is also prepared from the shells[98] by first treating them with water or steam, and afterwards extracting with water, and finally evaporating as far as necessary. The thick extract thus prepared contains theobromine, and is intended for use either alone or as an addition to cacao powder and chocolate.

Strohschein in Berlin[99] prepares from the shells a thick liquid extract which he calls “Martol Its preparation was suggested by the fact that the cacao husk gives evidence of containing a considerable amount of iron. In “Martol”, the iron occurs as a tannate, and the preparation further contains theobromine, carbohydrates, and phosphoric acid. The preparation is said to be used as a medicinal remedy in chlorosis, yet has scarcely justified such a statement.

Alfred Michel of Eilenberg[100] utilises the shells in the preparation of a brown colouring material. The husks, free from impurities, are first soaked in soft water, with or without the addition of sulphuric acid, then washed and finally treated with a strong 35 % solution of caustic soda. From the alkaline solution, the colouring matter is precipitated with acid or acid metallic salt, collected on a filter, and again washed. Thus obtained, it is a dark reddish-brown paste, possessed of a vitreous fracture. The yield of colouring matter is from 20-25 % of the weight of the original shells. By re-treatment with alkali, the paste can be again obtained in solution and can be used as required, either in liquid or paste form. The colouring matter can be obtained in different tints, either by soaking the shells in more er less dilute sulphuric acid, or by precipitation from the alkaline solution at various temperatures, or yet again, by the addition of metallic oxides.

Boussignault[101] says that in Paris briquettes have been made from cacao shells, and twenty-two years ago, Zipperer[102] proposed to use them as fodder, especially for horses. Experimental work in that direction was instituted, but for various reasons, had to be abandoned. The question as to a rational working up of the husk of the cacao bean is once more receiving special consideration, more particularly since the publication by the “Association of German Chocolate Manufacturers” of a prize essay on the subject. The fodder value of the husks as determined by Märcker is apparent from the following figures:

Table 18.

Shells	free from dust, whole %	fine meal %	whole and dusty %
Moisture	9·08	6·50	9·95
Albumen	13·56	14·13	12·69
Albumen digestible	6·06	7·07	4·38
Fat	2·65	6·76	3·96
Raw fibre	29·14	25·80	21·55
Ash	6·32	6·44	7·26
Non-nitrogenous extractive	39·25	40·37	44·59

Feeding experiments which were carried out in certain agricultural institutes showed that the cacao husk stands in nutritive value between good meadow hay and wheaten bran, and is not only a fattening fodder for oxen, but also a valuable feeding material for cows and deer[103]. These results have been confirmed by Prof. Feruccio Faelli in Turin[104].

The advantages of cacao shells as fodder, when a comparison with bran is established, are at once apparent. Two hundredweight (that is to say, about 220 lbs. averdupois) cost only from six to seven shillings, whilst the price of bran varies between nine and ten shillings. The husks also keep better, for after having been stored eighteen months, Professor Faelli found that they had undergone no alteration, whilst on the other hand bran had become sour. A further advantage possessed by the husk is that it will absorb four times its weight of water against three times absorbed by bran. Cattle not only readily get accustomed to the fodder but subsequently take to it with eagerness. The best results were obtained with Dutch, Swiss and Parmesan milch cows. After 10 days feeding the butter and milk-sugar had increased, as well as the daily average yield of milk from 44 to 49·5 kilogrammes. As soon as the feeding with cacao husk was discontinued the yield of milk decreased. Faelli concludes that cacao husk, which can be used as a fodder up to 4 kilog. daily, exercises a very favourable influence on milch cows, and he purposes to continue the investigation with horses.

In a report on the Experimental Farms of Canada 1898, page 151, reference is made to the manurial value of the husks in enriching the soil with nitrogen and potash, a fact which had already been pointed out by Boussignault.

The future use of the husks appears therefore to be ensured, and it is to be hoped that it will allow of a permanent consumption of this by-product.