68 ANALYSIS OF MILK. 



as, unless there is a considerable difierence between the figures 

 actually determined, the figures from which a and 6 are calcu- 

 lated are so small as to be greatly afiected by experimental 

 error ; while, if the difference be large (as in the case of analyses 

 of cream and skim milk), it is found that the experimental error 

 is also increased. For this reason, and also for the reason that 

 the specific gravity of fat and solids not fat are themselves liable 

 to slight variations, it is necessary to deduce the formula from 

 a great many determinations, which means much labour in 

 calculating. 



In order to ascertain the specific gravities of fat and solids 

 not fat in milk, the author has calculated their value from over 

 200 analyses made by the most exact methods at his command, 

 and finds the following figures : — 



Specific gravity of fat, . . . 0-93 



,, ,, solids not fat, . . 1'616 



It is seen that the author's figure, calculated from actual 

 analyses, agrees with Fleischmann's determination of the specific 

 gravity of fat. 



As the specific gravity of milk does not vary much, it will not 



make an appreciable error if, instead of yp, the expression 



be used ; this form of calculation is much easier. 



The idea of deducting a relation between specific gravity, fat, 

 and total solids appears to have arisen with Behrend and Morgen, 

 who published a table. Shortly afterwards Clauznitzer and 

 Mayer, and Hehner published formulae, but as thejr were founded 

 on inaccurate data, they are now abandoned. 



Pleischmann and Morgen next pubUshed a formula in which 

 the specific gravity of fat was assumed to be 0-94 ; this was 

 corrected by Fleischmann after his determination of the specific 

 gravity of fat as 0-93. His formula is 



T = 0-2(365 jr + 1-2 F. 



Hehner and the author deduced the formula 

 T = 0-254 G + l-16d F.* 



This i^ in the less scientifically correct, but more convenient 

 form ; as it was found that milk differing appreciably in specific 

 gravity from 1-0322 did not give results which agreed well with 

 the formula, various approximations have been made to this. 



The author has calculated a formula which gives practically 



* In the original paper a slight correction for skim milks was included 

 in the formula ; this has now been abandoned. 



