88 PRINCIPLES AND PRACTICE OF MILK HYGIENE 
discover the condition. After these cows were removed, 
the milk from these farms ceased to infect guinea pigs. 
No clinical cases of udder tuberculosis were found in the 
other 86 herds examined, but on these farms cows had 
been sold between the time the milk samples were col- 
lected and the herd was examined, or the farmer had been 
buying milk from other sources when the samples were 
collected. 
Friis inoculated guinea pigs with samples of mixed 
milk from 28 dairy farms in and about Copenhagen and 
demonstrated tubercle bacilli in the samples from four 
farms. On one of these farms two cows with tuberculosis 
of the udder were found and one cow with udder tuber- 
culosis was found on another, while on the other two 
farms cows were found showing physical symptoms of 
tuberculosis in other organs. There is no doubt that the 
other twenty-four farms contained cows which would have 
reacted to the tuberculin test. 
The milk from 12 non-clinical reactors was tested for 
tubercle bacilli by Klein and Campbell by injection into 
guinea pigs. ‘These cows were in a stable with 12 other 
non-clinical reactors which were not included in the ex- 
periment because they were approaching the end of the 
lactation period. The stable was light, well-ventilated, of 
suitable size and clean. The cows were cleaned with a 
curry-comb and brush and the udders wiped with a damp 
cloth before each milking. All the cows in the stable were 
examined by inspection, palpation and auscultation when 
the experiment began and no symptoms indicating tuber- 
culosis were found. They were all in a good, thrifty 
condition and none had a chronic cough. (AI suspici- 
ous animals had been previously removed.) The 12 cows 
used in the experiment were arranged in groups of three 
