When great thirst has been excited either by bodily labour or by external heat, it will generally be more effectually as well as safely quenched by drinking moderately of tepid than of cold fluids. A tepid draught--a cup of tea for example--produces no disturbing action in the stomach, and, being immediately absorbed, supplies the deficiency of liquid without changing the balance of the circulating fluids, thus relieving thirst very much in the same way as Dupuytren did by injecting tepid water into the veins; whereas cold drink, by the suddenness of its impression, disturbs the balance of the circulation, and excites a degree of reaction which increases the original discomfort. If the system is at the same time weakened by fatigue, cold drinks are always injurious. Of late years a change in the treatment of horses in this respect has taken place. Formerly it used to be considered dangerous to give water to a horse in a state of perspiration. Now, however, it is a common practice to allow post and stage-coach horses a little water before their run is completed, however warm they may be: but, in the first place, the quantity allowed is small, and is always given before the strength is exhausted; and, in the second place, the excitement of the subsequent exercise prevents any considerable disturbance in the balance of circulation, and thus ensures the safety of the animal. If even the same moderate quantity were given at the end of the stage, and the horse were then allowed to rest, inflammation would almost inevitably result. The same principle applies to the human frame, and affords an easy explanation of the occasional instances we see of persons _heated_ by exertion drinking cold water without injury. If the exertion have been merely sufficient to produce _excitement without fatigue_, and _heat without debilitating perspiration_, and especially if it be resumed after the draught, little or no harm will ensue. But if exhaustion or fatigue have been induced, and the individual be allowed to drink freely of the cold water and then rest, the probability of mischief resulting from it will be greatly increased. Many accidents result from losing sight of this distinction, and from acting alike in circumstances so essentially different. On the subject of _the kind of drink_ which ought to be taken a great deal might be said, were it necessary to discuss here the qualities of all the liquors that are in use. But as my purpose is very different, a few general remarks will be enough.