The Babcock Test. 39 



fore represents two-tenths of one per cent, of fat when 

 17.6 cc. of inilk is measured out. The small divisions 

 are sufficiently far apart in most Babcock test bottles to 

 make possible the estimation of one-tenth of one per 

 cent, of fat in the samples tested. 



As the necks of Babcock test bottles vary in diameter, 

 each separate bottle must be calibrated by the manufac- 

 turers; the length of the scale is not, for the reason given, 

 apt to be the same in different bottles. 1 



If the figures and lines of the measuring scale become 

 indistinct by use, the black color may be restored by 

 rubbing a soft lead pencil over the scale, or by the use of 

 a piece of burnt cork after the scale has been rubbed 

 with a little tallow. On wiping the neck with a cloth or 

 a piece of paper the black color will show in the etchings 

 of the glass, making these plainly visible. 



45. Marking test bottles. Test bottles can now be 

 bought with a small band or portion of their neck or 

 body ground, or "frosted," for numbering the bottles 

 with a lead pencil. Bottles without this ground label 

 can be roughened at any convenient spot by using a wet 

 fine file to roughen the smooth surface of the glass. There 

 is this objection to the latter method that unless carefully 

 done, it is apt to weaken the bottles so that they will 

 easily break, and to both methods, that the lead pencil 

 marks made on such ground labels may be effaced dur- 

 ing the test if the bottles are not carefully handled. 



i A flat-bore test bottle and one with a brass collar and screw used for 

 opening and closing a small hole in the neck of the test bottle have lately 

 been placed on the market by the Wagner Glass Works of New York. 

 These have been tried by us, but no particular advantage over the round 

 necks was discovered. 



