ANALYSIS OF WATER AND INTERPRETATION OF RESULTS. 31 
but the waters that can be obtained in this way are few. Some of 
the oil wells produce water which is known to come from a certain 
depth, but generally its exact source is unknown. In the oil fields, 
where sharp variations in the character of the water may occur within 
short vertical distances, too much care can not be exercised in the 
collection of samples. When a sufficient number of authentic samples 
of typical waters in any area has been collected and analyzed it will 
then be possible, by comparison with these standards, to form an 
estimate of the probable source of any water. 
Authentic samples can usually be taken only while drilling the 
well or while repairing it later. Flowing waters may, of course, be 
easily sampled, and the only precaution necessary is to let the water 
flow long enough before sampling to insure complete washing out of 
the drilling water from the hole. When the well is being drilled by 
the standard method the collection of samples is not especially diffi¬ 
cult, and samples uncontaminated by water from upper sands may 
generally be obtained after the drilling water has been bailed out. 
Under some conditions it is possible to obtain a fairly satisfactory 
sample merely by running the bailer to a point near that at which 
the water is entering the drill hole. When the circulating or rotary 
systems are used the proper collection of samples is often difficult 
or even impossible. Waters that flow when the pressure is released 
may be sampled at the expense of slight trouble, and if a string of 
casing is to be set within a short distance of a strong water sand, a 
sample of this water may be obtained while drilling is suspended. 
However, the rotary drill tends to mud up or plaster the walls of the 
hole and thus shut off flows of water, and if one of these flows breaks 
forth later the driller is likely to suppose that it comes from a deeper 
sand. 
In view of these uncertainties it is evident that unless the location 
of the water can be positively determined its analysis should not be 
used as a standard for exact comparison. For this reason it is the 
more desirable to collect samples carefully, and label them accurately, 
whenever good ones can be obtained, as when a flowing water is 
encountered. 
CHEMICAL ANALYSIS. 
DETERMINATION OF CONSTITUENTS 
The common mineral constituents of natural waters may be divided 
into two groups, those which have definite chemical relations with 
some other constituent and those which are probably present as col¬ 
loids. The colloid group comprises silica, iron oxide, and alumina. 
These substances, like those in the former group, may perhaps be held 
in solution in some waters, but ordinarily they are considered to be not 
in actual solution, and are conventionally reported as the oxides 
silica (Si0 2 ), ferric oxide (Fe 2 0 3 ), and alumina (A1 2 0 3 ). The con- 
