12 On some Improvements in the 



ammonia, but not by citric acid, which ought therefore to be used, 

 whereas in other cases ammonia is the best solvent. 



Unless the stain is faint, a portion should be soaked in a few drops 

 of water in a watch-glass, the hquid squeezed out, allowed to stand a 

 short time in the glass, so as to dej^osit any small portions of the fabric, 

 and poured into one of the experiment cells. If the stain had been 

 recently made, and had not been changed by any special action, a 

 solution of haemoglobin would be obtained, and the various spectra 

 could be seen one after the other, as already described. If, however, 

 the stain were a few days or a few weeks old, we should obtain a 

 mixture of haemoglobin and methaemoglobin, or the latter alone. 

 The various spectra could then be developed, and compared side by 

 side with those from fresh blood, to be sure that there is complete 

 correspondence in the position and relative intensity of the bands. 

 The residue insoluble in water should then be dissolved in dilute 

 citric acid or ammonia, according to the nature of the fabric, and 

 the spectrum of deoxidized haematin developed. If insoluble in cold 

 citric acid or ammonia, hot ammonia should be tried, since the stain 

 might have been so heated as to coagulate the albumen. If it be 

 desirable to keep the specimen of deoxidized haematin for subsequent 

 reference, the cell may be covered with a piece of thin glass, and, 

 after removing the excess of liquid, the edge of the cover painted 

 round with gold-size. When properly managed, such an object 

 will show a perfectly good spectrum, even after many weeks. 



If therefore we have a sufficient amount of a moderately old 

 stain, we may easily see in succession the seven very different 

 spectra of the following solutions: — 1. Neutral methaemoglobin. 

 2. Alkaline methaemoglobin. 3. Deoxidized haemoglobin. 4. Oxi- 

 dized haemoglobin. 5. Acid haematin. 6. Alkahne haematin. 

 7. Deoxidized hamatin. If the amount was very small, only 

 Nos. 4 and 7 would show distinct bands, and the rest would be 

 characterized rather by their comparative absence ; and it must 

 always be borne in mind that Nos. 1 and 2 may be modified by 

 the presence of unaltered haemoglobin. No. 3 by that of dissolved 

 haematin, and Nos. 5, 6, and 7 by that of undecomposed haemo- 

 globin or methaemoglobin. 



It would be easy to obtain other preparations, and to see several 

 other spectra derived from blood, but it appears to me unneces- 

 sary, since the above are so remarkable and unique in the manner 

 in which they are produced, one after the other, especially by 

 deoxidization and reoxidization on stu'ring, which seldom occurs in 

 other colouring matters, that they afford as satisfactory a test for 

 blood as could be desired, and still more so when we consider not 

 only the general character of the spectra, but also the exact position 

 of the absorption-bands, and that some are so most unusually dis- 

 tinct. 



