The BabcocU Test. 55 



twenty-four or a thirty-two bottle tester is large enough, 

 and to be preferred to a larger tester, even if a large 

 number of samples are to be tested at a time. The 

 operator can use his time more economically in running 

 a machine of this size than one holding fifty or sixty 

 bottles; the work of filling or cleaning the bottles and 

 measuring the fat can be done while the tester is run- 

 ning if a double supply of bottles is at hand. Large 

 testers require more power than smaller ones, and when 

 sixty tests are made, at a time, the fat column in many 

 bottles will get cold, before the operator has time to 

 read them, unless special precautions are taken for 

 keeping the bottles warm. 



65. The tester should be securely fastened to a solid 

 foundation and set so that the revolving wheel is level. 

 The latter must be carefully balanced in order that the 

 tester may be run smoothly at full speed. A machine 

 that trembles when in motion is neither satisfactory nor 

 safe, and the results obtained are apt to be too low. 

 High-standing machines are more likely to cause trouble 

 in this respect than low machines, and should there- 

 fore be subjected to a severe test before they are ac- 

 cepted. 



If all sockets are not filled with bottles when a test 

 is to be made, the bottles must be placed diametrically 

 opposite one another so that the machine will be bal- 

 anced when run. The bearings should be kept clean and 

 be oiled with as much care as the bearings of a cream 

 separator. 



The cover of the machine should always be kept 

 closed while the bottles are whirled, and should not be 



