44 DRS. E. SCHUNCK AND H. ROEMER ON SOME 



colour. Several crystallizations from alcohol were requisite 

 to render it quite pure. Sometimes it yielded golden-yel- 

 low micaceous scales ; but these on recrystallization always 

 gave needles. The needles contain water of crystallization, 

 which is driven off by heating to ioo°, the crystals losing 

 their lustre by drying. The scales seem to be anhydrous. 



The properties of isoanthraflavine are very similar to those 

 of anthraflavine. Its melting-point is above 330°. When 

 slowly heated between watch-glasses, it yields a sublimate 

 consisting of star-shaped groups of yellow needles, a car- 

 bonaceous residue being left, which is more considerable 

 than in the case of anthraflavine. 



It is a little more soluble in boiling water than anthra- 

 flavine, the solution being at first yellow, but acquiring a 

 reddish tint on continued boiling. It dissolves in concen- 

 trated sulphuric acid in the cold, and in alcohol and ether 

 on boiling ; but it is almost insoluble in benzol and chlo- 

 roform. It dissolves in caustic alkalies with a cherry-red 

 colour, and may thus be easily distinguished from anthra- 

 flavine, the alkaline solutions of which are yellow or, when 

 concentrated, reddish yellow. In concentrated sulphuric 

 acid isoanthraflavine dissolves with a cherry -red, anthrafla- 

 vine with a yellow colour. The two substances may also be 

 readily distinguished by their behaviour towards lime- and 

 baryta-water, in which isoanthraflavine dissolves even in the 

 cold, yielding red solutions, whilst anthraflavine only dis- 

 solves in boiling baryta- water, and is almost insoluble in 

 lime-water at all temperatures. An alcoholic solution of 

 isoanthraflavine gives no precipitate with an alcoholic solu- 

 tion of acetate of lead. Isoanthraflavine imparts no colour 

 whatever to mordants ; and its behaviour in this respect is 

 the same as that of most of the other isomerides of alizarine. 

 In some of its properties it strongly resembles purpuroxan- 

 thine ; but having prepared a specimen of the latter 



