SEPTIC TANK STUDIES 15 



line respectively. Table 10 shows the differential tests of table 8 expressed as 

 percentages of the total numbers of coliform cultures isolated from the soil 

 adjacent to particular disposal lines. 



In evaluating the data of the tables the writer takes the position, stated earlier 

 in this paper, that differences in the results obtained for the several disposal 

 lines from the three tanks should not be considered significant unless they were 

 pronounced. That was not true of the data obtained in this study. The results 

 may be summarized as follows. 



Gas-Positive Tests 



For the first sampling season, the greatest number of positive tests (50 to 65 

 percent) was obtained at the 48-hour reading. The number of positive tests at 

 24 hours was about balanced with the number at 72 hours. This meant, there- 

 fore, that roughly three-fourths of the gas-positive tests were obtained at 48 hours 

 or later. The control samples gave similar results. Escherichia coli from fresh 

 sewage pollution usually gives a positive gas test in 24 hours or less, while the 

 delated reactions usualh- are given by non-sewage soil t>pes or by false-positive 

 reactions. On the other hand, the soil types may give positive results in 24 hours. 

 All of this indicates that a majority of the coliform bacteria recovered from the 

 soil were either of non-sewage origin; or that, if they were of sewage origin, 

 their differential reactions had been altered by their sojourn in the soil. This 

 latter possibility has been suggested by several investigators but never proved. 



For the second sampling period the percentage of gas-positive results for 48 

 hours was again the largest; but it was lower and the 24-hour and 72-hour per- 

 centages were higher than for first sampling period. The majority of the gas- 

 positive tests were obtained in 48 or 72 hours of incubation as in the first period. 

 There was a considerable discrepancy between 48-hour results from the two 

 lines of the middle tank, but its significance, if any, is not apparent. 



The percentage of gas-negative tests was higher in the second period than in 

 the first. In the first period the percentages for one line from the middle tank 

 and one from the north tank were considerably higher than for the other lines. 

 However, the high percentage for the middle tank was for the line nearest the 

 tank, while the high percentage for the north tank was for the more distant line. 

 The percentages of gas-negative tests from the control soils were similar to those 

 from the several disposal lines except for the two high percentages just mentioned. 



The percentages of false-positive tests were higher, in general, for the first 

 sampling period than for the second. The distribution of values for the severa I 

 lines does not lend itself to any particular interpretation. In the first period the 

 highest value was for the control samples and the lowest, about half as great, was 

 for the near line of the middle tank. The value for the other line of this tank, 

 however, was nearly as large as that for the control. In the second period the 

 value for the more distant line of the north tank was the lowest of the series, and 

 that for the control samples was also low as compared to the high. 



In general, the results throughout the two seasons were heterogeneous and did 

 not indicate any preponderant values, either high or low, for any of the three 

 tanks. The preponderance of gas-positive values at 48 and 72 hours incubation 

 indicates that the cultures recovered were probably not from sewage but rather 

 were native to the soil. 



