94 BOTANICAL GAZETTE [FEBRUARY 
results with former ones, that it is immaterial, as far as the per- 
centage of turning is concerned, whether the spore-containing layer 
is above or below the one containing the test substance. 
Other culture media were used, such as 10 per cent. and 25 per 
cent. glycerin, gelatin made up to contain 10 per cent. of glycerin 
and to contain 5 per cent. of cane sugar. With none of these were 
there more distinct indications of chemotropism than in former 
tests, in which non-nutrient gelatin was the culture medium. This 
would indicate that the available nutriment and the concentration 
of the medium have no effect. 
The final test along this line was made with silica jelly, a medium 
free from organic matter and of suitable consistency. ‘The method 
of preparation was that used by Moore (23), except that dialysis 
was accomplished satisfactorily with parchment paper. In order 
to secure the proper coagulation of the medium, it was necessary 
to add mineral salts to all media used. A nutrient salt solution, 
containing 18“of ammonium nitrate, 0.5%" of monobasic potas- 
sium phosphate, 0.25%" of magnesium sulfate, a trace of ferric 
chlorid, and 58" of glucose, in a volume of 1o0°*, was made the 
basis of the work. It would seem that if fungi show chemotropism 
under any conditions it would be by turning from a medium lacking 
some one or more of the elements necessary for full development, 
toward the diffusion centers of compounds supplying the missing 
element or elements, or toward a full nutrient solution. 
In the tests each compound in turn, except ferric chlorid, was 
omitted from the silica jelly containing the spores, and in each case 
jelly containing the omitted compound in proper proportion on the 
one hand, and full nutrient jelly on the other were used on the opposite 
side of the mica plate from the above-mentioned spore-containing 
layer. Control tests were carried on with jelly lacking one and the 
same compound on each side of the plate, and also with full nutrient 
jelly on each side of the plate. To reduce evaporation, the lower 
layer was covered with an unperforated mica plate. The results 
are given in Table V, in percentage of response by turning from _ 
the first-named medium in which the spores were sown to the con- 
trasted one. 
There is a very striking uniformity in the percentages, and this 
