When natural tannins are boiled with dilute sulphuric acid, they are decomposed, generally yielding glucose, often also insoluble red bodies, called phlobaphenes. These bodies, when fused with caustic potash, give protocatechuic acid, and either acetic acid or a peculiar sugar called phloroglueine. Hlasiewetz gives the following table of the derivatives of tannin so treated :— From Hlasiewetz's table, it would appear that a very large proportion of tannins yield glucose on digestion with a dilute mineral acid. They appear, however, to differ materially from the class of bodies known as glucosides, which are generally crystalline., and yield their glucose very easily under the influence of acids, while tannins (except perhaps morintannic acid) are amorphous, and require somewhat prolonged digestion. Hlasicwetz suggests that they may be " gummides," or compounds containing gum or dextrine, which is converted by digestion with acid into glucose (see pp. 1914-21). But it is by no means certain that glucose should be regarded as an essential constituent of tannins, since, by repeated precipitations with lead acetate, and decomposition of the taunate with sulphuretted hydrogen, gallotannic acid nnay be prepared almost free from any glucose, yielding constituent, and retaining all its characteristic properties. Schiff, by acting on gallic acid with phosphorus oxychloride, has produced gallotannic acid synthetically, and this artificial tannin yields no glucose, but only gallic acid, by digestion with acids. This synthesized tannin appears to be digallic acid, formed from two molecules of gallic acid by the simple abstraction of one molecule of water. Thus, its constitutional formula is Natural gallotanuic acid is probably a molecular compound of this digallic acid with glucose or dextrine.
In digesting ordinary gallotannic acid, from gall-nuts, sumach, myrabolans, and other sources, with sulphuric add, or in its decomposition by natural fermentation, besides glucose and gallic acid, a varying quantity of a grey or fawn-coloured deposit of ellagic acid is always produced. This is closely allied to tannic (iligallic) acid, differing from it only by the abstraction of 2 atoms of hydrogen, and may be produced from it by heating it with dry arsenic acid (which oxidizes the hydrogen). It is, however, probable that it is produced in the cases named from a peculiar tannic acid, called by Lowe ellagitannic acid, which exists in divi-divi, myrabolans, pomegranate-rind, and other materials, in mixture with ordinary gallotannic aoid. This body is what Hlasiewetz calls " pomegranate tannin." Oak-bark and valonia yield abundant light-coloured deposits, which are probably ellagic acid.
On heating dry gallotannlo or gallic acids to I80°-210° (356°-410° F.), they are decomposed, and partially sublime in white prismatic plates of pyrogallol, or pyrogenic acid, while a black residue of metagallic acid remains. By heating a solution of gallio acid in glycerine to 200°
(392° F.), it is completely converted into pyrogallol, without production of metagallic acid.
Pyrogallol (C„1160,) has a feebly acid and very bitter flavour. It fuses at 131° (268° F.). is soluble in less than 3 parts of cold water, and still more soluble in hot. It is al,o soluble in ale,ohol and ether, but not in ab.olute ohloroform. In presence of alkalies, it absorbs oxygen with great avidity, turning brown or black. It reduces Fehling's solution, and those of gold and silver, and gives purplish-hlacks with ferric and ferrous salts. Probably its most characteristic reaction is the fine but very fleeting purple coloration which it gives with lime-water. Since all vegetable extracts containing gallic or gallotannio acids yield pyrogallol, its production has been used as a test for the presence of these bodies by Stenhouse, thue distinguishing again between the two classes of tannins derived from gallic and protocatechuic acids respectively. It is a singular fact that although oak-bark and valonia tannins give blue-blacks with iron salts, and yield abundant white deposits supposed to be ellegic acid, they give no pyrogallic acid on heating; but on the contrary, are stated by Johansen to yield protocatechuic acid on fusion with potaell. They thus present important differences from gallotannic acid on tbe one hand and from meet red-yielding tannins on the other, and urgently demand further investigation. Experiments made hy the writer suggeett he probability that oak-bark tannin contains phloroglucin as well as protocatechuic acid.
Associated with different species of catechutannic acids in catch and gambier, considerable portione of a white crystalline body (or possibly clfies of bodies) arc found. This is c.itechin. It is contained in large quantity in cube gambicr, forming a great part of the pale-coloured crystalline interior. It is readily soluble in boiling water, but very slightly so in cold ; hence, on allowing a boiling solution of gambler to cool, it is deposited in large quantities as a whitish sediment. Its solution does not precipitate gelatine. Its relation to the tannins of cutch and gambier is not well made out, but they seem to be anhydrides. Its lower anhydrides are soluble in water, and precipi tate gelatine ; while with every suceessive molecule of water which they give up, they become more insoluble, and the higher anhydrides are reds similar to phlobaphenc, insoluble in water and ether, but soluble in alcohol. Thus to an extent the relations between catechin and the catechu tannins are similar to those between gallic acid and gallotannic acid, but catechin is a body of much more complicated structure than gallic acid. Its formula is probably C,„H,,0„ and both it and its anhydrides, when fused with potash, yield protocatechuic acid and pliloroglucin.