Genetic and environmental parameters for racing time at different distances in Brazilian Thoroughbreds.

The aim of the present study was to investigate genetic parameters for racing time in Thoroughbred horses racing at distances between 1000 and 1600 m subdivided into 100-m intervals. The data provided by TURFETOTAL Ltda comprised races that occurred in the Gávea and Cidade Jardim race tracks over a period of 11 years (1992-2002) and consisted of 32,145 races and 238,890 time records. The variance components necessary to obtain the heritability and repeatability estimates of the traits studied were estimated with the MTDFREML program, and animal age at race (3 years old or younger, 4, 5 and older than 5 years), sex (male and female), number of races (1-32,145), and postposition at start (1-11) as fixed effects, and animal and permanent environmental random effects were included in a one-trait animal model. Males were significantly superior to females at all distances. Excluding the 1100 m distance, animals 4 years of age were significantly faster than the mean of the other ages for all distances analysed. Horses older than 5 years showed a significantly lower performance than the mean of the other ages for all distances analysed, except for the 1100 m. Postpositions one and two did not differ significantly from one another for any of the distances analysed. These two inner positions both together varied from the other positions depending on race length. The components of additive genetic and permanent environmental variance varied in a similar way, tending to decrease with increasing racing distance, and the other temporary environmental variance almost doubled from 1000 to 1600 m. As was the case for the additive genetic and environmental variances, heritability and repeatability estimates tended to decrease with increasing distance, indicating that selection based on racing time will be less successful when the racing distance increases.