evolved, and on the completion of the reaction a dark-brown liquid remains. On evaporating to remove excess of acid, a brown syrup remains, and on the addition of water a brown powder is deposited. This resembles the brown compound produced by the action of potassa-lye upon chrysammic acid. If this brown powder is filtered off, the evaporation leaves a considerable quantity of oxalic acid. The filtrate from this, if mixed with potassa-lye, smells strongly of ammonia.

Chrysocyammic Acid.

The product of the reaction of cyanide of potassium and chrysammic acid. It is obtained by adding 1 part of chrysammic acid to a solution of 2 parts cyanide of potassium in 12 to 15 parts water at about 60°. The liquid becomes hot and evolves ammonia and hydrocyanic acid. After some hours' digestion there is formed an imperfectly crystalline precipitate, which is filtered, pressed, and purified by repeated recrystallisation. Pure chrysocyammic acid is then obtained. by solution in dilute nitric acid. When dry it has a metallic lustre, is insoluble in water, but soluble in alcohol. At 120° it loses 3 equivalents of water and detonates when heated. It is a monobasic acid, the general formula, according to Finkh, being MO.C1H2N3O11. The alkaline salts are crystalline, and soluble in water; most of the metallic salts are insoluble, and are obtained by double decomposition as darkred crystalline precipitates. The salts detonate when heated as violently as gunpowder.

2

The potash salt is a dark, crystalline powder, reddishbrown when dry and of a metallic lustre when pulverised. It dissolves in water with an intense deep-violet colour, and loses all its water at 120°.

The ammonium salt is obtained by decomposing the potash salt with sal-ammoniac, and forms dark-green needles.

It is more sparingly soluble and more readily crystallised than the potash salt.

The barium salt is obtained on precipitating the ammonium salt with chloride of barium; it is a red-brown powder, of metallic lustre when pulverised, sparingly soluble in pure water, but insoluble in saline liquids.

The lime salt resembles the baryta salt in its behaviour and mode of preparation.

The lead salt is obtained as a voluminous red precipitate, insoluble in water.

The silver salt is obtained from the potash salt by double decomposition. It is a reddish-brown precipitate, which, when dry, has a dark-red brassy lustre.

This compound was discovered by Schützenberger and Schiffert in commercial purpurin, and prepared by them from purpurin. It is obtained by allowing ordinary phosphorus to react upon an alkaline solution of purpurin. The latter dissolves without evolution of gas, and the reduction is complete in a few minutes. The colour of the solution passes from violet-red to red, and finally to brown. The product is poured into acidulated water, and the flocculent precipitate is collected and washed.

Another method has been described by Liebermann and Fischer. If nitrous acid is passed through a boiling alcoholic solution of purpurinamid, nitrogen escapes, and a compound is obtained, which, when purified by sublimation, is identical with Schützenberger's purpuroxanthin. It is formed according to the equation:

C1HNO1+NO2H+H2=С11Н ̧O1+N2+2H2O.

8

Schützenberger has also obtained it by the reduction of purpurin with stannous chloride in an alkaline solution.

As Rosenstiehl has succeeded in converting purpuroxanthin into purpurin by oxidation, and as Baeyer has ascertained the constitution of purpurin, it is evident that purpuroxanthin is the third bioxyanthraquinon derivative containing the two hydroxyls in the same benzol-nucleus, for which there remains only the position 1.3. Whilst Liebermann and Fischer state that purpuroxanthin does not dye with mordants, Schützenberger maintains that it dyes a yellow shade.

Purpuroxanthin forms a yellow mass, and sublimes readily in orange-yellow needles resembling pure alizarin.

In alcohol, acetic acid, and benzol, it dissolves readily, and is precipitated by water from its alcoholic solution as a transparent jelly, which gradually condenses into flocks. It dissolves in alkalis with a fine red colour. The lime and baryta salts dissolve in boiling water with an orange-red colour. It dissolves at a boiling heat in alum water, but is almost entirely redeposited on cooling.

If purpuroxanthin is heated with hydriodic acid (boilingpoint 127°) and common phosphorus, there is obtained in the first place a substance of a more greenish yellow colour which crystallises from benzol in splendid leaflets. The alkaline solution is brown, and is oxidised on exposure to the air, with reproduction of purpuroxanthin. With aluminous mordants it dyes shades similar to quercitron, and agrees with the formula C4H10O4, or C14H12O4. On prolonging the action of the hydriodic acid under the same conditions, both the hydrides of anthracen are obtained. If heated with zinc-powder, purpuroxanthin yields anthracen. In a boiling alkaline solution, purpuroxanthin takes up 1 atom of oxygen and purpurin is reproduced.

Frangulin, C20H20010

This glucoside is obtained from the bark of Rhamnus rangula. This is extracted at a boil with ammoniacal water,

the decoction is mixed with hydrochloric acid, the precipitate pressed, and extracted with boiling alcohol and sugar of lead. From the alcoholic solution thus purified the frangulin is precipitated by the addition of more sugar of lead.

The red precipitate is decomposed with sulphuretted hydrogen, and the frangulin is extracted from the sulphide of lead by means of boiling alcohol. The alcohol is mostly distilled off, and from the remainder the frangulin is precipitated by water and obtained in a state of purity by several repetitions of this last operation.

Frangulin forms a yellow mass which appears crystalline under the microscope. It is almost insoluble in cold water, sparingly soluble in cold alcohol and ether, more freely soluble in hot alcohol. Alkalis dissolve it with a fine red colour. The ammoniacal solution is colourless, but soon turns red. It melts at 226°, and has the properties of a feeble acid, but its salts have not been obtained. Acids decompose it into frangulic acid and sugar,

C20H20010=C14H2O4+C6H12O6.

Frangulic Acid, C11HO,.

This acid can be obtained also direct from the bark of the rhamnus if it is extracted with water containing caustic soda.

From alcohol it crystallises in orange-coloured quadratic leaflets or needles of the formula 2(C1,H) (OH)2O2+3H2O, which do not become perfectly anhydrous until heated to 180°, and melt between 252° and 254°. It is somewhat soluble in water, chloroform, and benzin; readily soluble in alcohol and ether. Alkalis dissolve it with a fine red colour, but acids separate frangulic acid unchanged from the solution. From a slightly ammoniacal solution it is precipitated by alkaline earths and many metallic salts. Fuming nitric acid dissolves frangulic acid, and a red mass remains on dilution

with water. This nitrated frangulic acid is probably the same substance which Casselmann describes as nitrofrangulic acid, and which he obtained along with oxalic acid on heating frangulin with fuming nitric acid. If heated with zincpowder, frangulic acid yields anthracen.

Dibromfrangulic Acid, C1H¿Br204.

Obtained by mixing an alcoholic solution of frangulic acid with excess of bromine. It is sparingly soluble in alcohol, and consequently separates from the mixture just mentioned. It forms a bright-red, light crystalline mass, which, under the microscope, appears to consist of fine short needles.

Difrangulic Acid, C2H14O7+2H20.

This acid may be regarded as the anhydride of frangulic acid, and it may be formed from 2 mol. of frangulic acid by the elimination of 1 mol. water.

Faust obtained this acid from the lead precipitates produced in the direct preparation of frangulic acid from rhamnus bark by extraction with dilute soda-lye. It has a strong resemblance to frangulic acid, but is of a darker colour. It melts at 248°-250°. This agreement in the melting-points is probably due to the circumstance that frangulic acid, when melting, or immediately previously, is converted into difrangulic acid by elimination of water. Kubly's aromin is also probably identical with frangulin, and aromic acid with frangulic or difrangulic acid.

Trioxyanthraquinons, С1H(OH)2O2.

In the trioxyanthraquinons, three hydrogen atoms of anthraquinon are replaced by hydroxyl, according to the posi