A small portion of the suspected sample is gently placed upon the surface of a beaker filled with cold water, and allowed to remain at rest for about fifteen minutes. If pure, the sample does not imbibe the water, but floats upon the surface without communicating much colour to it; if chicory or caramel be present, these substances rapidly absorb moisture and sink, producing brownish-red streaks in their descent, which, by diffusion, impart a very decided tint to the entire liquid. A similar coloration is caused by many other roasted roots and berries, but not so quickly or to so great an extent. The test may be somewhat modified by shaking the sample with cold water, and then allowing the vessel to stand aside for a short time. Pure coffee rises to the surface, little or no colour being imparted to the water; chicory, etc., fall to the bottom as a sediment, and give a brownish colour to the liquid.
If a small quantity of the sample is placed upon a clean plate of glass, and moistened with a few drops of water, the pure coffee berries remain hard, and offer resistance when tested with a needle; most grains employed for their adulteration become softened in their texture.
A considerable portion of the mixture is treated with boiling water and allowed to settle. Genuine coffee affords a clear and limpid infusion; many foreign grains yield a thick gummy liquor, resulting from the starchy and saccharine matters contained. An infusion of pure coffee, if treated with solution of cupric acetate and filtered, will show a greenish-yellow colour; if chicory be present, the filtrate will be reddish-brown. As a rule, samples of ground coffee which are much adulterated, pack together when subjected to a moderate pressure.
Owing to the low density of a coffee infusion (due to its almost entire freedom from sugar), as compared with that of the infusions of most roots and grains, it has been suggested by Messrs. Graham, Stenhouse and Campbell, to apply the specific gravity determination of the infusion obtained from the suspected sample as a means for detecting adulteration. The results afforded are fairly approximate. The solution is prepared by boiling one part of the sample with ten parts of water and filtering. The following table gives the densities, at 15°·5, of various infusions made in this manner:—
Assuming the gravity of the pure coffee infusion to be 1·0086, and that of chicory to be 1·0206, the approximate percentage of coffee, C, in a mixture, can be obtained by means of the following equation, in which D represents the density of the infusion:—
C = 1·00(1·020 - D) 12.
This was tested by mixing equal parts of coffee and chicory, and taking the specific gravity of the infusion; it was 1·01408, indicating the presence of 49 per cent. of coffee. Some idea of the amount of foreign admixture (especially chicory) in ground roasted coffee may be formed from the tinctorial power of the sample. It has already been mentioned that coffee imparts much less colour to water than do most roasted grains and roots. The table below shows the weights of various roasted substances which must be dissolved in 2·000 parts of water in order to produce an equal degree of colour:[11]—
The comparative colour test may also be applied as follows:[12]—One gramme each of the sample under examination, and of a sample prepared by mixing equal parts of pure coffee and chicory, are completely exhausted with water, and the infusions made up to 100 c.c. or more; 50 c.c. of the filtered extract from the suspected sample are then placed in a Nessler cylinder, and it is determined by trial how many c.c. of the extract from the standard mixture, together with enough distilled water to make up the 50 c.c., will produce the same colour. In calculating the chicory present, it is assumed that this substance possesses three times the tinctorial power of coffee.