INTERPRETATION OF RESULTS 315 
The following is an example: Write down the percentage of plus 
(+) tests for the quantities examined expressed as decimals of 100. 
Take the differences between these percentages. Multiply each of 
these differences by the reciprocal of the quantity corresponding to the 
larger of the two percentages from which such difference is taken. The 
sum of these products will be the B. coli index. In the following exam- 
ple, the B. colt index may be taken as 0.019—equivalent to 19 B. colz 
per cubic centimeter 
a 
{ - 
Quantity of Per Cent Expressed as 
Reciprocals of 
=i ‘ater 4 Yasitive Decimals Differences. Quantita es, Product. 
100 13 8 0 1330 =| . 1090 OL 00109 
10 29 0 0290 .0260 .10 .00260 
1 03 0 0030 .0026 1.00 .00260 
0 1 0 04 0 0004 .0003 10.00 .00300 
0 Ol 0 Ol 0 0001 .0001 100.00 .01000 
In this B. colt index the committee has made no provision for per- 
centage reduction. It recommends that it be omitted for various 
reasons. The bacteria in an effluent from a filter are not derived from 
the raw water. It required quite a little time for a water to pass through 
a filter. The percentage removal is, to a certain extent, a function of 
the number in the raw water and does not vary with the number left 
in the filtered water. Other objections are also mentioned by the 
committee. 
Interpretation of Results. The accurate interpretation of results 
of analysis is a part of sanitary work which is often taken too lightly. 
This problem is rendered more difficult by the lack of an ideal indicator 
of pollution. No organism has been described which is always and only 
present in water and on foods which have received sewage pollution. 
Thresh (1913) gives an interesting discussion with regard to what should 
be regarded as polluted water. He sent questions to Houston and Wins- 
low seeking their opinions on the subject. He submitted the following 
questions to Houston of the Metropolitan Water Board: What bac- 
teriological proof would you consider concluswae as to the pollution of a 
water with sewage or manurial matter? Houston did not give a definite 
answer and stated that one was not possible. He advises that the final 
verdict should rest on data secured from all analyses. His experience 
seems to indicate that B. colt should be absent from 100 ¢.c. quantities 
of the water in a majority of representative samples, preferably not less 
