Multivariate analysis for characteristics of heat tolerance in horses in Brazil

The environment in which the horse is reared affects its ability to maintain thermal balance which is in turn related to thermal characteristics and regulatory physiological mechanisms. In this study a multivariate analysis of physiological traits in relation to heat tolerance in horses was carried out in the Federal District, Brazil. The aim was to test the ability of these analyses to separate groups of animals and determine which physiological traits are most important in the adaptation to heat stress. Forty adult horses (4 to 13 years) were used, ten from each of four different genetic groups (English thoroughbred, Brazilian showjumper, crossbred and Breton). The traits examined included heart and breathing rate, rectal temperature as well as blood parameters. The data underwent multivariate statistical analysis including cluster, discriminate and canonical using Statistical Analysis System - SAS ® procedures CLUSTER, STEPDISC, CANCORR and DISCRIM. The tree diagram showed clear distances between groups studied and canonical analysis was able to separate individuals in groups. The discriminate analysis identified the variables which were most important in separating these groups. The multivariate analysis was able to separate the animals into groups with RR, HR and RT being important in this separation.     

Use of heat tolerance traits in discriminating between groups of sheep in central Brazil

The animal and its environment make up an integrated system, where each acts on the other. Tropical regions are characterized by high levels of solar radiation and environmental temperature which may adversely affect animal production. This study carries out a multivariate analysis of physical and physiological traits in sheep in the Federal District of Brazil to test the ability to separate groups of animals and determine which traits are most important in the adaptation of animal to heat stress. The variables studied included coat thickness, number and length of hairs, pigmentation of the skin and coat, number of sweat glands as well as heart and respiratory rates, rectal and skin temperatures, sweating rate, and blood parameters. Five groups of ten animals were used depending on breed (Bergamasca, crossbred, or Santa Inês) or coat color (Santa Inês—brown, white, and black). The data underwent multivariate statistical analyses including cluster, discriminate, and canonical, using Statistical Analysis System—SAS®. The tree diagram showed clear distances between groups studied and canonical analysis was able to separate individuals in groups, especially Bergamasca and white Santa Inês. The canonical correlation redundancy analysis showed that coat reflectance as well as hair length and number of hairs per unit area were the most useful in explaining changes in physiological traits. Skin and coat traits such as hair length, coat reflectance, percentage of epithelial area occupied by sweat glands, skin reflectance and thickness, as well as heart and breathing rates were the most important in separating these groups.