Article contents
Investigations on the use of the Milko-tester for routine estimation of fat content in milk
Published online by Cambridge University Press: 01 June 2009
Summary
Results of studies on the operation of the Milko-tester Mark II for the estimation of fat in both herd and individual milks are presented. Milko-tester readings were compared to the means of duplicate Gerber results.
It was found that, for 894 herd milks, a direct Milko-tester reading (G = M) had a standard error of estimate of ±0·071, whereas regression equation (G = 1·02 M −0·075) derived for the same data gave ±0·070. For 1133 individual milks, the equation (G = 1·03 M − 0·186) also had an accuracy of ±0·070, while the direct Milko-tester reading gave ±0·107. However, a simple subtraction of 0·077 from each Milko-tester value (i.e. G = M−0·077) for individual milks reduced the standard error to ±0·071.
It is concluded that the Milko-tester is satisfactory for herd milks, particularly for milks up to 3.8 % fat. For individual milks a special regression relationship is necessary to achieve comparable accuracy.
- Type
- Original Articles
- Information
- Copyright
- Copyright © Proprietors of Journal of Dairy Research 1967
References
REFERENCES
- 6
- Cited by