336 
Journal of Agricultural Research 
Vol. XXXI, No. 4 
of Manor loam soil was 333 c. c., the minimum 294 c. c., the average 
323 c. c. These figures show far less variation than the rates of per¬ 
colation. Most sets of six tubes varied little within the set—in one 
case only 6 c. c. Most sets were of air-dry soil. As the tubes are 
2 inches in diameter, applications of 258 c. c. per tube would be equal 
to approximately 3 gallons per square foot to secure percolation 
through 13 inches of potting soil and of 323 c. c. per tube would be 
approximately 4 gallons per square foot to secure percolation through 
14 inches of Manor loam soil. At the time these laboratory tests 
were started the field applications had not exceeded 2 quarts of for¬ 
maldehyde solution per square foot, and tests of field-treated soil 
had indicated that such applications could not be depended on to 
penetrate more than 4 or 5 inches. A few tubes were set up with 
soils to which definite amounts of water had been added, and formal¬ 
dehyde solution was then added to each tube at the rate of 2 quarts 
per square foot. The results were exceedingly erratic. To other 
sets of tubes applications were made at the rate of 1 gallon per square 
foot. Of 9 tubes set up with Manor loam put through an 8-mesh- 
to-the-inch sieve, 2 tubes showed penetration 9 to 10 inches, 5 tubes 
8 to 9 inches, and 2 tubes 7 to 8 inches at the end of 1 day. These 
results indicated that it would be necessary to use applications of 1 
gallon or more per square foot in order to secure the 8 inches of pene¬ 
tration required to reach all wart spores in the soil. 
Penetration of Formaldehyde Compared with that of Mercuric 
Chloride 
The experiments with formaldehyde indicated that the penetra¬ 
tion of formaldehyde in water solutions practically coincided with 
the penetration oi the water of the solution. The test for formal¬ 
dehyde while sensitive and accurate and relatively simple is far 
more tedious than the simple precipitation test with H 2 S for mercury. 
Preliminary determinations were made to see if the H 2 S precipitate 
test could be successfully used in soil solutions. No positive results 
were obtained with this test at any depth in either potting soil or 
Manor loam treated with 1 gallon per square foot of 1 to 1,000 
mercuric chloride solution. Three tubes of potting soil were then 
given applications of 1 per cent mercuric chloride at the rate of 3 
gallons per square foot. A faint mercury test was obtained in the 
percolate from each tube. These tests indicated rapid removal of 
the mercury from the solution as it passed through the soil. To 
retard this removal of the mercury, sodium chloride was added to 
the mercuric chloride solution at the rate of 5 parts of salt to 1 part 
of mercuric chloride. Salt was first added to mercury in our field 
treatments because Hartley (7) reported that the combination was 
P articularly severe on green plants when used in soil treatments, 
t was thought possible it would be equally severe in its action on 
the wart fungus. In studying the penetration of mercuric chloride 
it seemed wise to test the penetration of the solution actually used 
in the field experiments. The penetration was found to be very 
much better when salt was added to the solution. Manor loam 
with a moisture of 9.5 per cent and potting soil with a moisture of 
6.8 per cent were used for parallel sets of tubes treated with 1 per 
cent formaldehyde and witn 2 per cent mercuric chloride plus 10 
per cent sodium chloride, at the rate of 200 c. c. per tube (nearly 
