Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: 6/24/1996
Publication Date: N/A
Technical Abstract: Relationship of test-day somatic cell score (SCS) with milk yield on subsequent test days was examined to determine influence of test-day SCS on milk yield. Data were from 47,719 1st and 8,719 2nd lactations from Holstein herds in California, Pennsylvania, and Wisconsin. Eight successive test days were required for each lactation; a 1st lactation must have been present for 2nd lactation to have been included. For 1st lactation, mean SCS was lowest (2.36) for test day 2 and highest (2.94) for test day 10; corresponding means for 2nd lactation were 2.32 and 3.64. Multiple linear regressions for milk yield of test day i on SCS for all prior test days (1,2,...,i-1) were computed within herd-year. For 1st lactation, all coefficients for SCS of immediately prior test day (i-1) were negative and significant (P<.05). For each test-day milk yield, at least one regression coefficient was significant for SCS of test days 1,2,...,i-2, and all coefficients were negative. For 2nd lactation, multiple regression equations for predicting milk yield of test days 3-5 contained no significant coefficients other than for SCS of immediately prior test day (i-1) in contrast to equations for test days 2 and 6-10. For milk yields of test days 6-8 and 10, regression on SCS of test day i-2 was significant and negative. Several significant regressions on SCS of test days 1,2,...,i-3 were positive for equations for milk yields of test days 7-10. Although regressions of yield on SCS of adjacent test days were negative, regressions of yield on SCS of more distant test days (SCS during early lactation) tended to be positive. These results confirm validity of using SCS for estimation of milk loss, even for SCS measured prior to current test day, thus excluding dilution effects.