SPECIAL METHODS. 237 



reactions of the protoplasm, which never contains coloring 

 matters that show the reaction directly. 



In the large plasmodia of jEthalium septicum Reinke (I, 

 8) was able to determine the alkaline reaction macroscopical- 

 ly ; and he deduced the presence of a volatile alkali from the 

 -observation that a bluing of litmus-paper occurred when it 

 was not in direct contact with the plasmodium. But this 

 observation does not in any way exclude the presence of 

 other alkaline substances in the plasmodium ; and it has been 

 shown to be probable, by Schwarz (I, 33), that the alkaline 

 reaction of the protoplasm in the higher plants probably 

 does not depend on the presence of ammonia or ammonium 

 compounds. 



426. In these plants Schwarz (I, 20) attempted to deter- 

 mine the reaction of the protoplasm by killing cells that 

 naturally have a colored cell-sap by alcohol, heat, crushing, 

 or electricity, and then noting the color assumed by the dead 

 protoplasm. Again, he treated colorless cells in the same 

 way in a feebly acidified extract of borecole leaves which is 

 3^ellow-red, purple-red, or red-violet in an acid solution, 

 violet when neutral, and blue, blue-green, grass-green, yellow, 

 or yellowish orange when feebly alkaline. Schwarz found 

 that, after killing by electricity, the protoplasm of a few 

 cells was blue-green ; but in most, it was blue-violet, or red- 

 violet, and he therefore concluded that the reaction of the 

 protoplasm is alkaline. 



On the other hand, Arthur Meyer (III) observed that the 

 coloring matter of kale is not violet, but blue, when the re- 

 action is neutral, and that, when tin-foil electrodes are used 

 in conducting the current, a violet tin compound of the col- 

 oring matter is formed and taken up by the dead protoplasm, 

 and that various colorations of the extract may be caused 

 near the electrodes by the decomposition of the salts con- 

 tained in it. There can therefore be no doubt that Schwarz's 

 method cannot give trustworthy results. 



427. The most certain conclusions may be reached by 

 Pfeffer's methods of introducing artificially certain colored 



