METHODS OF TESTING BABCOCK MILK BOTTLES. 



135 



2. The thickness of the graduation lines is comparable with 

 the whole space. In fact, in some makes of bottles, the thickness 

 of the graduation lines is quite one-quarter of the whole space 

 between the lines. Hence the difficulty of reading the level, 

 mentioned in the preceding paragraph, is increased. 



3. It is difficult sometimes to avoid air bubbles in the mercury 

 when filling the bottle in Method III. Errors would arise from 

 these rising to the surface after a reading of the level is taken. 

 Also, a change of temperature taking place while testing would 

 cause a change in the total volume occupied by these air bubbles, 

 with a consequent change in the level of the mercury in the 

 neck. 



4. When testing by the third method, the large amount of 

 mercury filling the bottle causes appreciable errors if the 

 temperature varies while testing, as it is sometimes very liable 

 to do under ordinary conditions of working. For example, the 

 weight of the mercury filling a Babcock milk test bottle is 

 approximately 700 grams. The expansion of that amount of 

 mercury for a rise of temperature of 1°C. is nearly 0"01 c.c, 

 which is equivalent to 0'05 per cent, on the Babcock scale of 

 percentages. This is obviously a very serious error, and is 

 probably the chief difficulty encountered when using this method 

 of calibration. 



Under working conditions in the laboratory it was noticed 

 that the reading of the level of the mercury was usually lower 

 after allowing the bottle to stand for a few minutes. Now it is 

 conceivable that the large weight of mercury in the glass bottle 

 miglit cause an expansion in volume of the flask owing to the 

 elasticity of the glass. It is probable, however, that this is too 

 small to be apparent, and that the fall in the level of the mercury 

 in the neck is practically all due to the lowering of the tempera- 

 ture, for in pouring the mercury into the bottle a small, but 

 appreciable, rise in temperature takes place, as is shown by the 

 following experiment that was carried out: — 



Immediately after pouring mercury into a bottle up to 

 approximately the top of the scale, the temperature was carefully 

 observed and a reading of the level taken. These readings were 

 repeated at intervals of five minutes, and the results were as 

 follow : — 



M 



