120 TECHNICAL ANALYSIS OF 



Even when this method of experimenting was adopted, the 

 results obtained were always too high in comparison with 

 those obtained by means of the two nutrient liquids. Koctis 

 method cannot, therefore, be employed even in this form. 



If the hygienic method enabled us to determine with any 

 degree of certainty whether pathogenic bacteria were present 

 in a brewing water, we should, of course, always employ it in 

 addition to the other method ; as is known, however, this is, 

 unfortunately, not the case, and its value for our purpose is 

 at most that it affords a means for controlling the filters. 

 Even in this test it is highly important to carry out the 

 analysis in such a manner that the results obtained during a 

 lengthened period and at different places are strictly com- 

 parable. In the text-books the directions given with regard 

 to Koch's method are to leave the plate cultures for three to 

 four days at the ordinary room-temperature. To this it may 

 be remarked that, in the first place, during this period, as a 

 rule, only a small proportion of the bacteria present will show a 

 distinct development, and, when regarded as a percentage, this 

 ivill notj in all cases, be an expression of the total number of 

 bacteria. If accurate information is required with regard to 

 the bacterial contents of a water, the cultivation must be 

 continued for at least a fortnight. In one analysis, for 

 instance, the number of growths found after four days was 

 only one-tenth of those which developed after ten days, and 

 only one-fifteenth of those found after sixteen days. Like- 

 wise, the room temperature is very uncertain and variable, 

 and it will make no slight difference whether the cultures are 

 exposed to a temperature of 20 C. or 10-5 C. An analysis 

 may give a very different result during the summer from that 

 obtained during the winter. If the work is carried out during 

 the summer, and under fairly favourable conditions as regards 

 night temperature, the same water sample will give, under the 

 same conditions of culture, an appreciably different result, 

 according to whether the colonies which develop are examined 



