May 12, 1923 
Effect of Decomposition on Concrete THe 
489 
end of the line, where the lower part of the tile was entirely protected by 
soft mineral soil, little disintegration had taken place, but the upper 
part was badly pitted where peat had lain against it (PI. 2, A). 
RECJSNT IrlMH DRIFT 
Walworth and Ozaukee Counties were taken as representative of the 
heavy limestone drift area of recent formation and rich in lime. The 
peat swamps in this area are very narrow, have been subject to heavy 
wash, and are high in mineral matter. The soils are comparatively 
tight clays and loams and seepage is negligible. Consequently, the 
ground water in summer descends below the tile lines and, except for 
short intervals, usually remains there throughout the growing season 
(PI. 2, B). Tile No. 2 came from near the mouth of a short drain where 
it was not subjected to standing water. No. 3, however, came from 
opposite the mouth of a low draw where there would be considerable 
seepage from the surrounding highland. The peat areas in this forma¬ 
tion are comparatively small, and as the rate of disintegration is slow 
it is not expected that it will cause serious difficulty. 
Ol,D RESIDUAI, SOII,S 
Richland County was taken as representative of the unglaciated area, 
and here it was found that in the acid residual soils the tile broke down 
wdth comparative rapidity. 
ACID sandy SOII^ 
Wood and Juneau Counties were taken as representative of the sand¬ 
stone area. These soils are for the most part very strongly acid. In 
this district concrete tile set in peat broke down within a year after 
they were laid. The sewer tile shown in Plate 4, B, was laid in clay soil 
high in lime. The outside gave no sign of disintegration but the inside 
was somewhat pitted. This area appeared to be the most destructive 
to concrete of any that were investigated. Throughout this whole sandy 
district the use of concrete in marshes appears to be only a temporary 
expedient. Opportunity was offered to examine the concrete culverts 
on the roads trough the cranberry marshes of Wood County. Condi¬ 
tions here offer the most severe test that could be devised, for the acid 
waters of the cranberry marshes and reservoirs remain against the con¬ 
crete throughout the season. All of the culverts and bridges examined 
showed signs of disintegration, the destruction having penetrated into 
the solid concrete. Plate 3, B, shows this very strongly. 
GRANITE SOII,S 
Waupaca County was considered typical of the granite soils. An 
observation made at Wyanwega was extremely interesting, for the tile 
were laid not in peat, but in sand underlying muck. An area of some 
53 acres was tiled in 1915. In 1917 the tile system began to give trouble. 
In 1918 several breakdowns occurred, owing to the complete collapse of 
the tile at those points. In 1919 there were over 20 breakdowns, one 
of which was about 100 feet long. Strange to say, in spite of its thicker 
wall, the 8-inch outlet main seemed to suffer the most. Plate 4, A, shows 
samples from this system. In no case except No. 4 did the peat come 
in contact with the tile. No. 3 was taken from a tile line on a flat piece 
of high ground. There was no peat on this line. 
