ARNATTO. 391 



remains for seven or eight days, or until the liquor 

 begins to ferment ; sometimes, indeed, weeks or 

 even months elapse before this effect is produced. It 

 is then strongly agitated with wooden paddles and 

 beaters to promote the separation of the pulp from 

 the seeds ; this operation is continued until these 

 have no longer any colouring matter adhering to 

 them. The turbid liquor is then passed through 

 close cane sieves, leaving the refuse seeds behind. 

 The mixture is now very thick, of a deep red colour, 

 and of an extremely unpleasant odour. On being 

 boiled the colouring matter is thrown up to the sur- 

 face in the form of scum, or otherwise the colour is 

 allowed to subside ; in either case the scum or the 

 precipitate must be boiled in coppers until reduced 

 to a consistent paste. It is then suffered to cool arid 

 made up into cakes, which are dried in the shade. 

 The liquor from which the colouring matter has been 

 removed is preserved under banana leaves until it 

 becomes heated by fermentation ; it is then re-boiled, 

 and the scum which rises is taken off. It then again 

 undergoes similar treatment, until no more colour re- 

 mains to be extracted. Instead of this tedious process, 

 which occasions diseases by the putrefaction induced, 

 and which at best affords only a spoiled product, M. 

 Leblond proposes simply to wash the seeds of ar- 

 uatto until entirely deprived of their colour, which 

 lies wholly in the pulpy part, and to precipitate the 

 colour by means of vinegar or lemon-juice, and then 

 to boil in the ordinary manner*. 



The experiments of M. Vauquelin made on some 

 arnatto berries imported by M. Leblond, confirmed 

 the efficacy of the process which he proposed, and 

 the dyers ascertained that arnatto obtained in this 

 manner had at least four times the value of that 

 which was procured in the ordinary manner. It was 

 * Ann. de Chimie, torn, xlvii. 



