64 



If our experiment stations had taken this matter up in a practical 

 manner, I am sure they would long ago have demonstrated that much 

 of the faulty butter on the market is due to the water. 



From Mr. Boggild's excellent book " Danish Dairying," I take 

 the illustrations Fig. 67=68, which represents a galvanized iron filter. 

 The cross section, Fig. 68, shows first a loose perforated wooden bot- 

 tom, then a layer of pebbles, then gravel, then sand, then another 

 perforated bottom. On this there is a layer of charcoal and then a 

 layer of scrap iron. The upper bottom has only one hole in the cen- 

 tre, and is covered with pebbles. The height is 3 ft., 6 in., and the 

 filter is filled with the above mentioned materials at least two= thirds. 

 Fig. 67 represents the manner in which the filter (a) is fixed on the 

 wall, with the supply pipe (e) and its cock (f) provided with a 

 rod (g). 



In order always to have filtered water in stock a storage tank (b) 

 is provided. The latter ought, however, to have a cover not shown in 

 the illustration. 



I am of the opinion that the water before filtering or after, ought 

 to be boiled and then cooled, unless indeed the Utopian age were here 

 when every creamery has a 



PASTEUR FILTER. 



This filter which I regret to 

 say requires a pressure of at least 

 20 Ibs. to the square inch to do 

 practical work, and which is rather 

 expensive, is not only a filter, but 

 a complete sterilizing apparatus, as 

 no microbe, no germs of microbes 

 even, can pass through those won- 

 derful hollow " candles " made of 

 a composition of unglazed por- 

 selain, prepared by Pasteur's as- 

 sociate, Prof. Chamberland. 



The idea of sterilizing milk 

 this way lay near and would ob- 

 viate the dreaded boiled flavor, but 

 alas and alack, this filter is so 

 powerful that only a very clear 

 would be the result. 



