Beginning with half an hour and gradually increasing to about an hour (a little more for very large or very fat people,—­a little less for the small or thin) the daily massage is kept up through at least six weeks, and then if everything seems to be going along well, I direct the rubber or nurse to spend half of the hour in exercising the limbs as a preparation for walking.  This is done after the Swedish plan, by making very slowly passive and extreme extensions and flexions of the limbs for a few days, then assisted movements, next active unassisted movements, and last active movements gently resisted by nurse or masseuse.  When the patient is able to sit and stand, it is well to keep up and extend the number of these gentle gymnastic acts and to encourage the patient to make them habitual, or at least to keep them up for many months after the conclusion of treatment.[18]

At the seventh week massage is used on alternate days, and is commonly laid aside when the patient gets up and begins to move about.

In 1877, several of the members of the staff of the Infirmary for Nervous Disease, and especially my colleague, Dr. Wharton Sinkler, obliged me by studying with care the influence of massage on temperature, and some very interesting results were obtained.  In general, when a highly hysterical person is rubbed, the legs are apt to grow cold under the stimulation, and if this continues to be complained of it is no very good omen of the ultimate success of the treatment.  But usually in a few days a change takes place, and the limbs all grow warm when kneaded, as happens in most people from the beginning of the treatment.[19] The extremely low temperature of the limbs of children suffering with so-called essential paralysis is well known.  I have frequently seen these strangely cold parts rise, under an hour’s massage, six to ten degrees F. In such small limbs, the long contact of a warm hand may account for at least a part of this notable rise in temperature.  In adults this can hardly be looked upon as a cause of the rise of temperature produced by massage, first, because the long exposure of large surfaces incident to the process is calculated to lessen whatever increase of heat the contact of the hand may cause, and secondly, because this rise is a very variable quantity, and because occasionally some other and less comprehensible factors actually induce a fall rather than a rise in the thermometer as a result of massage.

In very nervous or hysterical women, ignorant of what the act of kneading may be expected to bring about, and especially in such as are thin and anaemic and have either a somewhat high or an unusually low normal temperature, we may find at first a slight fall of the thermometer, then a fairly constant rise, with some irregularities, and at last, as the health improves, a lessening effect or none at all.

The most notable rise is to be found in persons who, owing to some organic disease, have acquired liability to great changes of temperature.