BY DR. BURDON-S ANDERSON. 191 



Then on the filter the clot is treated with distilled water at a 

 temperature of about 35 C C., the filtrate being alloAved to drop 

 into a measure-glass cooled in ice. It is of great importance 

 that this part of the process should be carried out with as little 

 loss of time as possible. I have found it a good plan to in- 

 close the clot in the filterer, and then to knead it repeatedly in 

 small quantities of warm water contained in the capsule ; the 

 products of all the extractions being collected on the same 

 filter, and received in the cooled beaker. A measured portion 

 (say ten cubic centimetres) is then transferred, with the aid of 

 a pipette, to a test-glass, to which alcohol is added drop \yy 

 drop from a burette. The precipitate formed by the first drops 

 of alcohol redissolves on shaking or stirring: as more alcohol 

 is added the precipitate at last remains undissolved. [By this 

 means the proportion of alcohol required, in order to diminish 

 the solvent power of the liquid sufficiently to render it prone 

 to crystallize, is determined.] Alcohol is then added to the 

 whole liquid, in proportion somewhat less than is required to 

 produce a permanent precipitate, The clear solution on being 

 left to itself, surrounded with iced water, soon begins to 

 crystallize. The crystals are separated by filtration, and washed 

 on the filter with ice-cold water containing a little spirit, and 

 subsequently with ice-cold water alone. To obtain the sub- 

 stance in a state of purity it must be subjected to recrystalli- 

 zation. For this purpose the crystals must be dissolved in 

 distilled water at 40C. and evaporated in vacuo, the process 

 being repeated until a product is obtained which on incinera- 

 tion leaves pure oxide of iron without trace of phosphoric 

 acid. 



Dr. Gamgee recommends the following process, which was 

 recently communicated to him by Professor Kiihne, and has 

 been successfully employed by him on three separate occasions. 

 Five hundred cubic centimetres of defibrinated blood of a dog 

 are mixed in a flask with 31 c. c. of pure ether, and thoroughly 

 shaken at intervals of a few minutes during an hour and a half 

 or two hours. The mixture is then placed in a cellar for about 

 twenty-four or thirty-six hours. The flask containing the lake- 

 red liquid is now surrounded with ice (not a freezing mixture) 

 for twelve hours, at the end of which time it is found to have 

 become converted into a magma of haemoglobin crystals. Dr. 

 Gamgee states that the only objection to this method consists 

 in the great difficulty of filtering the crystalline from the viscid 

 serous portion of the mixture. In laboratories where the cen- 

 trifugal apparatus is to be found, the magma may be placed in 

 tubes and submitted to excessively rapid rotation for three or 

 four hours, at the end of which time the haemoglobin will have 

 separated as a soft cake from the serum, which can be decanted. 

 Where no centrifugal apparatus can be obtained, the magma of 



