|
|
||||||||
Dairy Science Department, University of Florida, Gainesville 32611
ABSTRACT
Estimates of variation associated with sire of fetus effects were obtained from 14,147 Jersey records that originated in four southeastern states. After adjustment for age of cow and length of record, the complete model included sire of fetus, sire of cow, herd-year-season effects, and effects of all possible interactions. Deletion of some or all interactions was necessary for mixed model analyses. Completely random models yielded unrealistically high estimates for sire of fetus variances for milk and fat yields and for fat percent, as had previous research. Mixed model analyses suggested that sire of fetus effects were small, albeit real and measurable. As a portion of total variability they accounted for: in milk and fat yields, 1 to 2%; fat percentage, 1 to 4%; days open, 0 to 1%. Gestation lengths were affected by fetal sire, regardless of whether the model was random or mixed; variances were 9.3 to 13.6% of the total. Preliminary estimates of interactions involving sire of fetus, sire of cow, and herd-year-season suggested that they are not zero and warrant further study.
1 Florida Agricultural Experiment Station Journal Series Number 8490.
2 Gallina Blanca Purina, Barcelona, Spain.
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |