283 Mr. G. S. Johnson. 



nally present in the flesh extract would undergo conversion into 

 kreatine, and consequently that the kreatine which is ultimately 

 obtained may have resulted either partly or entirely from a conver- 

 sion of kreatinin into kreatine, in short that the kreatine is a mere 

 product from the flesh, not a true educt. My first experiment was 

 designed to test this theory. 



Experiment I. 70 Ib. of fresh butcher's beef was finely cut up by 

 a sausage machine, after being freed as far as possible from fat and 

 bone, and thoroughly incorporated with water by hand-kneading. 

 By means of a specially designed screw-press, which was made for 

 me by Messrs. Farrow and Jackson, the aqueous extract was removed 

 from the fibre. This process was repeated with fresh additions of 

 water, until four extracts were obtained, each extract being sepa- 

 rately examined. 



The albuminoid substances were separated from all the extracts by 

 heat (about 80 C.), most of the colouring matter being carried down 

 by the coagulated albumen. The filtrates were then concentrated by 

 evaporation over steam, until a scum began to form upon the surface, 

 after which the further concentration was effected in vacuo over 

 sulphuric acid, by means of a Carre's freezing machine. No chemical 

 agent was added to any of the extracts at any time, until the con- 

 centration was complete. 



In each case the residue obtained by concentrating these extracts 

 was partly crystalline and partly gelatinous. 



Complete separation of the crystalline from the amorphous matter 

 was readily effected by means of dilute alcohol, which left the crystal- 

 line matter undissolved. 



Finally, by fractional crystallisation from water, the crystalline 

 matter was resolved into an organic and an inorganic substance. 

 The organic crystals were kreatine. 



The inorganic salt was acid potassium phosphate, KHoPO^ 

 The results of this experiment prove 



1st, that kreatine may be obtained from the watery extract of flesh 

 whose natural acidity has not been, chemically neutralised, so that 

 hydrolytic conversion of kreatinin into kreatine is most improbable ; 

 and, 



2ndly, that the presence of phosphates in solution does not inter- 

 fere as has been stated to be the case by some observers with the 

 crystallisation of kreatine from extract of flesh. 



Inasmuch as I am not aware that acid potassium phosphate has 

 been actually obtained in crystals from the watery extract of muscle 

 hitherto, I have thought it advisable to give analytical results which 

 prove the identity of the substance. 



The salt is much more soluble in water than the kreatine with 

 which it is associated, so that the separation of the two substances is 



