AMD ON THE ABSOKPTION OF THE PRISMATIC COLOURS. 435 



advantage would have been of very considerable amount ; but I 

 have found this quite impracticable, and I venture to say, that 

 the separation of homogeneous green or homogeneous yellow 

 light, of any considerable intensity, cannot be eflFected by any 

 coloured media with which we are at present acquainted. 



Abandoning, therefore, all hopes of obtaining from colour- 

 ed media any farther improvement upon the microscope than 

 what I had formerly announced, it occurred to me, that the 

 object which I had in view might be obtained, if I could pro- 

 cure, from the combustion of inflammable substances, a homo- 

 geneous flame for illuminating microscopic objects. 



It had long been known, that a great quantity of homoge- 

 neous yellow light was created by placing salt or nitre in the 

 white flame of a candle, or in the blue and white flame of burn- 

 ing alcohol *. A light, however, generated in this manner, 

 was more fitted for a casual experiment, than for a permanent 

 source of illumination ; and as insalubrious vapours are disen- 

 gaged during the combustion of these salts, I did not avail my- 

 self of this method of obtaining yellow light. 



After numerous experiments, attended with much trouble 

 and disappointment, I found that almost all bodies in which 

 the combustion was imperfect, such as paper, linen, cotton, ^c. 

 gave a light in which the homogeneous yellow rays predomi- 

 nated ; — that the quantity of yellow light increased with the 

 humidity of these bodies ; — and that a great proportion of the 

 same light was generated, when various flames were urged me- 

 chanically by a blowpipe or a pair of bellows. 



As the yellow rays seemed to be the product of an imper- 

 fect combustion, I conceived that alcohol diluted with water 



would 



* Edinburgh Physical and Literary Essays, vol. ii. p. 34. ; and Dr Thomas 

 Young's Nat. Phil. vol. i. p. 438. Mr Herschel informs me, that sulphur in a 

 certain stage of its combustion produces a homogeneous yellow light. 



