Sources of individual variation to heat tolerance in a rabbit line

The objective of the present study was to quantify the extent of the interaction between temperature and animal-specific effects on prolificacy and average weaning weight in rabbit does and to assess to what extent this interaction has a genetic determination. For this purpose, 18,491 total born (TB) records from 5,929 Caldes does, as well as 16,868 weaned kits (NW) and average weaning weight (AvgWW) records from 5,353 does of the same line were used. Four models were considered, ranging from the simplest, which assumed no effect of temperature on the traits studied, to the most complex, which allowed for different slopes of response and thresholds for heat tolerance for each animal. Deviance information criteria (DIC) were adopted as model choice criteria. The model postulating that each animal is subjected to a linear change in its performance when the temperature reaches a certain animal-specific threshold was preferred for all the considered traits. This interaction has the greatest effect for AvgWW; the ratio between the variance associated to the intercept and the total variance declined from 29.1% at 19°C to 19.4% at 25°C. This decline results from the increase in the variation associated with the interaction of the individual effects with temperature and also from the covariance between this interaction term and the intercepts. An important part of this interaction has a genetic origin, particularly for AvgWW and NW, for which the estimated heritability increased from 0.11 to 0.24 and 0.06 to 0.10, respectively, from 19 to 25°C. Similarly, it was found that the proportion of the total variance for the individual effects involved in the interaction with temperature which have a genetic origin is high for all the traits, around one-half for the variation on the linear changes in the performance and more than one-third for the variation on the onsets for heat stress. However, the error associated with these estimates is large and results must be considered with caution. Thus, in conclusion the extent of the interaction between individual variation and in farm temperature seems to be important and has a genetic origin. Nevertheless, the accuracy of the estimated genetic parameters is low.