[Vor. 4 
184 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
A more pronounced reaction was obtained by igniting bits 
of tissue and triturating the residue on the glass plate with 
a 10 per cent solution of hydrochloric acid. The liquid was 
then drained off, a large drop of NaH(NH4) POs, applied, and 
the slide allowed to remain for several minutes in an atmos- 
phere of ammonia. 
Positive results were obtained when these tests were ap- 
plied to both lighter and darker areas. 
POTASSIUM 
The most reliable test for potassium is its reaction with 
platinie chloride, resulting in the formation of crystals of 
potassium chloroplatinate. A 10 per cent solution of platinic 
chloride is recommended for this test. The sections to be 
tested were mounted in a drop of alcohol, and a drop of pla- 
tinie chloride about one-tenth the size of the drop of alcohol 
was placed on the slide. The reagent and alcohol mount were 
then brought into communication by means of a glass needle. 
After several minutes crystals of potassium chloroplatinate, 
mainly in the form of octahedrons, but also in the form of 
hexahedrons and rhombohedrons, separated out. 
This result was checked by applying a test solution con- 
sisting of 2 grams of cobalt nitrite and 3.5 grams of sodium 
nitrite dissolved in 1 ce. of acetic acid diluted with water to 
7.5 ec. After the cessation of the liberation of nitric oxide 
fumes, the solution was diluted to a volume of 10 ce. Upon 
the application of this reagent to sections, minute granules 
of potassium cobalt nitrite separated out. The crystals were 
extremely small and were detected with difficulty. 
These tests were applied with equal success to both dis- 
eased and healthy tissue. 
PHOSPHORUS 
Phosphorus is generally detected by means of a solution 
of 1 gram of ammonium molybdate in 12 ce. nitric acid (sp. 
от. 1.18). In the presence of phosphorus, granules or small 
octahedrons of ammonium phosphomolybdate separate out. 
