GENERAL E.XCiriNG CAUSES 37 



good, and each of them may be employed according to 

 circumstances. It is obvious that certain circumstances 

 may make it necessary to adopt one or other method. 

 I need only mention as an example that, after severe 

 loss of water, such as occurs in consequence of long- 

 continued exertion, the animal should always be allowed 

 to drink before it is fed, as otherwise it will not feed 

 well. Although all methods of watering are equally 

 good for the horse, it is not desirable to change un- 

 necessarily from one method to another. Animals, or 

 at least some of them, appear to be not altogether in- 

 different to such a change. We observed in our experi- 

 ments that whenever a change was made from the plan 

 of watering after food to that of watering before, the 

 appetite fell off for some days ; not that they did not 

 consume the w^hole of the food given to them, but for 

 some days together they did not eat with the same 

 avidity, and took a longer time to consume their rations 

 completely. A similar effect was not observed when the 

 change was from watering before to watering after food, 

 or from watering after to watering during meals, or when 

 the change was in the opposite direction to the last. It 

 is possible that this method of watering before food, 

 until the animal has become accustomed to it, produces 

 a certain feeling of satiation. Further experiments would 

 be necessary to show to what extent, apart from the bad 

 effects of an excessive consumption of water, the plan of 

 watering after meals acts injuriously on digestion in the 

 case of foods that are difficult of digestion or tend to 

 induce tympany, such as grains provided w^ith an 

 envelope.' 



(c) Errors in the Proper Distribution of Work, 

 Rest, Watering, and Feeding.— Of all the baneful 

 effects of civiHzing the horse that tend to the production 



