1865.] Sokby on Detection of Blood- Stains by Spectrum- Analysis. 211 



as described just below, from most readily passing into bsematin, wben 

 acted on by a weak acid. So far I bave been unable to decide 

 wbetber tbe difference consists in brown cruorine containing more 

 oxygen or water, or in its being only an isomeric modification of tbe 

 scarlet. However, since such a powerful oxidizing agent as perman- 

 ganate of potash does not produce it, and since it is not formed in 

 solution or when damp, the latter supposition appears to me to be the 

 most probable, unless indeed it contains less water. 



Professor Stokes has shown that weak acids convert cruorine into 

 hsematin,* and has described the spectrum produced by it in its 

 natural state, and when alkaline and deoxidized. On adding acetic 

 acid to a solution of either scarlet or brown cruorine, a spectrum is 

 obtained like that shown by No. 7. There is a well-marked absorp- 

 tion-band in the red, the centre being situated at l£ above D, and 

 therefore rather higher than that in the case of brown cruorine, and 

 very nearly in the position of Fraunhofer's line C. There is also 

 another at about 14 below D, and perhaps a third still lower, which 

 is only seen when the solution is more dilute, and even then so 

 obscurely that its existence appears doubtful. When dried on glass 

 the band in the red does not disappear ; and when the solution is 

 strong, ammonia causes a precipitate, as though hasmatin were much 

 less soluble in ammonia than either scarlet or brown cruorine. 

 However, the solution being somewhat dilute, using about Ird or ^th 

 of a grain of citric acid to prevent the subsequent precipitation of 

 oxide of iron, adding ammonia in decided excess, and introducing the 

 whole into a cell ; if about -yVtii or -joth of a grain of crystallized 

 protosidphate of iron be dissolved in the solution, we obtain the very 

 well-defined spectrum of deoxidized haematin shown by No. 8. The 

 centre of the upper band is | below D, and of the lower at 2. The 

 upper is about £ wide, very dark, and sharply denned ; the lower 

 nearly the same width, but much more faint. Neither is much shaded 

 at the edges, and both vary more in darkness than in width, on 

 increasing or decreasing the strength of the solution or the length of 

 the cell in which it is examined. When concentrated, and in a tube 

 £ an inch long, the width of the bands is much as shown, but the 

 shaded part is darker. In a tube |th of an inch in diameter, a 

 spectrum as distinct as No. 8 is given by Troth of a grain of liquid 

 blood, making on thin linen a spot about roth of an inch in diameter ; 

 and such a minute quantity will show the bands faintly, yet distinctly, 

 that ToWth of a grain of liquid blood would be quite enough to furnish 

 us with unmistakable evidence of its presence, if some little care and 

 trouble were taken over the experiment, and the solution examined in 

 a cell not more than roth of an inch in diameter. 



It thus appears that in various conditions blood yields no less than 

 eight very different spectra. At all events, we can most readily get 

 four or five, so characteristic, that with proper care there could be no 

 fear of confounding it with any other substance that I have hitherto 

 examined, and probably not with any in existence. However, in all 



* ' Proceed. K, S.' x'm. 357. 



