**Acknowledgements**

available) that can complement the ways of reducing temperature in farms through costly heat abatement systems. Formerly described results show that applying random regression methods to the data from milk recording, together with the climatic information from the meteorological stations close to the farms, goats can be genetically evaluated for their response to HS and robust or tolerant animals can be selected. Selection criteria can focus on increasing the tolerance threshold or the slope of decay of the considered trait, but the estimation of the genetic value for the threshold of tolerance of each animal in goats is not easy due to the scarce and noisy information available (only six‐ to eight‐test day per lactation) [61]. However, in all studies, in both dairy cattle and goats, it has been found that most of the observed variability of the response is associated to the production level (the intercept coefficient of the response) and only a small fraction of the variation is associated to the slope coefficient. Furthermore, there is an antagonistic relation between the intercept and the slope. Therefore, selecting for a lower decay of yield (lower negative or positive slope) would lead to a negative effect on the level of production. Carabaño et al. [61, 62] have proposed using canonical variables resulting from the eigendecomposition of the additive genetic random regression coefficient (co)variance matrices derived from the norm of reaction models formerly described. The canonical variable explaining the largest proportion of the variation of genetic values of animals across the range of values of THI for milk yield is linked to the production level and only a small proportion of about 10% can be used to select animals tolerant to HS without compromising the level of production. On the contrary, in the case of milk components there is a canonical variable explaining a larger part (up to 25%) of the genetic variation for heat tolerance independent of the production level. This could be a good selection criterion to get heat‐tolerant animals for the trait fat plus protein yield, which is the most important trait in goats' selection programmes. According to Carabaño et al. [61], genomic information may play an important role in identifying genomic variants present in animals showing high production levels and a low rate of decay due to HS. Other possible tools to be used for selecting heat‐tolerant animals could be the use of single nucleotide polymorphism (SNP) markers, found through GWAS analyses, associated with physiological indicators of the response to HS like rectal temperature and respiration and sweating rate. Biomarkers determined in milk by means of mid‐infrared technology, routinely used for milk composition analyses in milk‐recording programmes, and other recent discoveries of genetic mechanisms involved in heat tolerance through transcriptomics and proteomics may also contribute to find selection tools to improve the response to HS without impairing

Traditionally, it has been considered that goats are better adapted to semiarid and hot climates than cattle and sheep; however, the results of the studies carried out in native Spanish breed formerly describe showed that goats also suffer some physiological effects derived

A negative effect on milk traits of both high and low temperatures was observed in three

from the exposure to high temperatures with a negative impact on yields.

Spanish native breeds of goats. The effect is higher in high‐yielding animals.

production levels [61].

**7. Conclusions**

340 Goat Science

A part of the results of experiments referred to in this chapter have been obtained in the framework of the projects AGL2007‐66161‐C02 and INIA‐FEDER: RTA2011‐00108, financed by Spanish (Ministry of Economy) and European (ERDF) funds, with the collaboration of the Spanish Meteorological State Agency (Agencia Estatal de Meteorología—AEMET), which provided the meteorological records, and the breeders associations of Murciano‐Granadina (CAPRIGRAN), Malagueña (CABRAMA), Florida (ACRIFLOR) and Payoya goats which allowed for the use of their data bases.
