1910.] PUBLIC DOCUMENT — No. 31. 153 



cent, of water and 10 per cent, of skim milk, without positively 

 detecting its presence by the aid of the refractometer. 



5. It is believed that the Zeiss refractometer will prove very 

 helpful in the detection of added water in milk. The evidence 

 furnished, however, must be considered in connection with that 

 secured by direct chemical analysis. 



It is believed that the percentage of ash in milk is likely to 

 prove fnlly as helpful in many cases as the index of refraction 

 in detecting the presence of added water. Mixed milk falling 

 substantially below .70 per cent, of ash must be regarded with 

 suspicion, and that testing below .65 per cent, of ash as watered. 



The impression held by some milk inspectors and producers, 

 that the immersion refractometer will detect very small amounts 

 of added water, is erroneous. Such an impression, firmly fixed 

 in the minds of unscrupulous producers and dealers might have 

 a salutary eft'ect, but it is not justified by results in actual 

 practice. 



