OUTLINES OF A POLICY OF EEFOEM 275 



of all such examinations.' Since then the test has 

 been greatly perfected, so that the percentage of 

 error ought to be almost nil. Too much stress can- 

 not be laid upon the fact that frequently the best- 

 looking animals in a herd are the ones most seriously 

 affected. It is important, therefore, that every 

 animal should be regidarly tested, regardless of 

 appearances. 



Nothing less than a periodical test of every dairy 

 herd, and of every animal in every such herd, will 

 do. The infected animals should be destroyed, and 

 not merely segregated. This is the verdict of practi- 

 cally every serious student of the Danish experiment. 

 Segregation, breeding from diseased animals, or 

 fattening them for sale as second-class food for the 

 poor are compromises to be avoided. Not only do 

 such compromises prolong the process of eradicating 

 the disease, but lead to all kinds of corrupt and 

 fraudulent practices. So I am assured, and certainly 

 such results might be expected. However true this 

 latter charge may be of Denmark, it can scarcely be 

 doubted that it would be terribly true in this country. 

 Every infected animal should be destroyed, and its 

 carcass rendered unfit for use as food. 



And this at once raises the question of compensa- 

 tion, which the British government has found so 

 troublesome. It is plain that if we are ever to 

 eliminate bovine tuberculosis, the active support- 



