102 
EIGHTH REPORT. 
CHARACTERISTICS OP UNDERGROUND WATERS. 
Hardness. It is rare to find an underground water supply in which 
the amount of carbonates of calcium and magnesium fall below 140 parts 
per million, and in a large part of the State sulphates are also present 
to increase the hardness. The softest waters appear to be obtained 
from sand dunes and very shallow wells in sand deposits on the old 
beds of lakes which extended somewhat beyond the limits of the present 
Great Lakes. There are also occasional waters from sandstone forma¬ 
tions, and from near the base of the drift (where a supply from under¬ 
lying sandstone is suspected) in which the waters are soft enough to 
be fair for boiler use. 
Salinity. The salt or brackish waters of Michigan are generally con¬ 
fined to rock formations lying at considerable depth, but there are parts 
of the Saginaw basin and also places near the borders of Lake St. 
Clair in which the waters in the glacial deposits are found to carry 
enough salt to render them objectionable for drinking. In such cases 
it is supposed that the salt Avater has risen into the glacial deposits 
from underlying rock formations, though it is barely possible that the 
drift deposits themselves carry enough saline matter in these localities 
to cause excessive salinity of the water. It is worthy of note, that 
throughout the southern portion if not throughout the entire State a 
small amount of salt is found in the waters from the drift where no 
contamination by human agencies has occurred. The amount of chlorine 
or sodium chloride found by an analyst is not therefore to be taken as 
a measure of contamination. References may be made to the report 
of i)r. Lane on Lower Michigan Mineral Waters (Water Supply Paper 
31) for numerous analyses which will throw light upon the saltness of 
water in various localities and at different geological horizons. 
Iron. In many of the underground waters iron is present in sufficient 
amount to color the boiler scale and also to induce the growth of iron- 
loving organisms such as crenothrix at points within the water pipes or 
system. The amount of iron necessary to induce the growth of these 
organisms is small compared with the total solids, less than five parts 
per million apparently being sufficient. Chemists have found that hav¬ 
ing the standpipe covered, and light excluded, will greatly retard organic 
growths in the water. It is a popular notion that the brown, slimy 
materia] found along the course of escaping water is mainly iron oxide; 
but in many cases it is an organic growth, the iron being present only 
as a stain. Many of the waters from deep Avells in the drift, however, 
contain considerable iron and it is also present in Avaters from rock 
formations. Thus at Battle Creek the waters from the Marshall sand¬ 
stone have been found by analyses made at the University to contain 
2.5 to 4 parts per million of iron and the city has Avisely delayed a 
a change to this supply until the effect of the iron has been determined. 
Other objectionable constituents. Mention need scarcely be made of 
bitter Avater (caused by the presence of magnesium sulphate and mag¬ 
nesium chloride) for it is found in only a feAV localities and is generally 
avoided for public water supplies, not only because it is not palatable, 
but because it scales and corrodes badly by dissociation of the sulphate 
and production of sulphuric acid. 
A much more common agent of corrosion is free carbonic acid gas 
