the Colouring Matters of Madder. 193 



has done, is evidently impossible ; for no formnla of ordinary- 

 dimensions can be devised for rubian, which shall include within 

 itself the formula? of its five products of decomposition. If his 

 formula for rubian, C 32 H 18 18 , be correct, then his formula for 

 alizarine, C 10 H 3 O 3 , must be incorrect, as the number of atoms 

 of C in rubian must be a multiple of that in alizarine. On the 

 other hand, if his formulae for alizarine, rubiretine, and verantine 

 be admitted, then rubian must contain C 30 . The only decom- 

 position which M. Laurent's formulae explain is that of rubian 

 into rubianine and sugar, which would then be expressed by the 

 following equation : — 



Rubianine. Sugar. Rubian. 



C 20 H n O 9 + C 12 H 12 O 12 =C 32 H 19 O 17 + 4HO. 



Even in this case M. Laurent's formula for rubian requires 

 some alteration. 



M. Laurent objects to the equation C 56 H 34 O 30 =4C 14 H 5 O 4 

 + 14HO, whereby I explain the formation of alizarine from 

 rubian, because there is no authentic instance of a reaction with 

 fourteen equivalents of water. All that I can say in reply to 

 this objection is, that this must be the first authentic instance, 

 as I am constrained to admit it as an inference from positive 

 facts. M. Laurent has not given any confirmation of his for- 

 mula for alizarine. Had he, however, subjected a specimen of 

 the pure substance to analysis, he would have found that his 

 formula cannot for an instant be sustained, and that it does not 

 contain the same number of equivalents of hydrogen and oxygen. 

 M. Strecker, who adopts the same formula as M. Laurent, has 

 certainly given one analysis of alizarine ; but this analysis was 

 made with only 0*0650 grm. of substance, a quantity too small 

 to admit of any safe inference being drawn from its analysis. 

 The purest specimen of alizarine which I have ever prepared was 

 obtained by the decomposition of rubian with sulphuric acid. 

 The following table shows the composition of this specimen as 

 compared with that found by Strecker and Debus (as an average 

 of three determinations), and with the calculated composition 

 according to the two formulae : — 



It will at once be seen that the numbers found in my analysis 

 differ so widely from those calculated according to the formula 

 C 10 H 3 O 3 , that the latter cannot possibly be accepted as the true 

 one. An excess of 0-6 per cent, of hydrogen, as this formula 

 would presuppose, is unusually large ; an excess of 0*4 per cent. 



Phil. Mag. S. 4. Vol. 6. No. 38. Sept. 1853. 



