as the various instincts of fighting of different animals, which we connect with anger, and those of flight which we connect with fear-there is some evidence that these instincts may sometimes be excited and evoke their characteristic behaviour without arousing the particular emotions. Not only do experiments on animals appear to enforce this conclusion; but we have familiar experiences of our own. We are face to face with sudden danger; with calmness and rapidity we make the requisite movements, and find that we have escaped from the dangerous situation without feeling the emotion; but we often feel it retrospectively, if we reflect on the situation. We may also fight without feeling the emotion of anger. With calmness we may watch the movements that our opponent is about to make, and guard against them. At times there may be a rush of angry emotion; at others, the behaviour of the instinct and the acquired skill proceed without it. In many such cases we can see, just as in those fragmentary instincts of grasping or sitting or standing, that the movements are accomplished so rapidly that there is not time for the emotion to be felt. And it is precisely under the opposite conditions that we are most likely to feel intense emotion: where the action is not accomplished rapidly, but delayed. There are no fears so intense as those which arise in situations from which we cannot escape, where we are forced to remain in contemplation of the threatening events. There is no anger so intense as when the blood boils and all the sudden energy that comes to us cannot vent itself on our antagonist. The arrest of an instinct is that which most frequently excites the emotion connected with it; and therefore we feel the emotion so often before the instinctive behaviour takes place, rather than along with it. What appears to be invariably present when an instinct is aroused is not emotion but a feeling of impulse. Without impulse it would be no more than a compound, reflex action.

1 As those of Dr. Goltz on a dog from which the hemispheres had been removed; also others of Prof. Sherrington recorded in his work on 'The Integrative Action of the Nervous System.' See also Prof. Lloyd Morgan's Instinct and Experience,' ch. iii.

Preyer, in the observations of his child, remarks that the first clasping of a finger, when it is placed in the hand of the child, is not an instinctive movement involving a felt impulse, but a simple reflex. For he observes the reflex action on the fifth day, but the instinctive clasping not until the seventeenth week. The absence of the clasping in sleep, which he noticed, is "to be ascribed to the insufficient excitement of the nerves of the skin, and the diminution of reflex excitability in sleep." For when, he says, instead of putting a finger in the hollow of the child's hand, "I move my finger with a gentle rubbing movement back and forth upon the flat of his hand, he often clasps it quickly, almost convulsively, with his fingers, without waking." 2

If we cannot suppose that the excitement of an instinct is invariably accompanied by an emotion, still less by one of a particular quality, we may at least distinguish between those reflex actions that are preceded by a psychical impulse to perform them and those that are not, and confine the term "instinct" to the former. But still we cannot be sure that such simple actions as "clasping" what is placed in the hand may not sometimes be accompanied by an impulse that we feel and at other times respond so rapidly to the stimulus of a sensation as to exclude it.

(2) If an instinct may be excited without involving an emotion, an emotion may be incapable of being excited without involving the excitement of some instinct or other innate tendency. But the same emotion may include a variety of instincts in its system. A few examples will make this point clear. In the chapters on the primary emotions we shall consider in detail the variety of instincts or innate tendencies that are organised in these particular systems. We shall here confine ourselves to fear. In the system of that emotion as it exists in man and in many animals, there are included at least two instincts of flight and concealment. Sometimes these instincts are excited together or in close succession, sometimes one and not the other. An animal-and especially man with his greater plasticity and power of adaptation— may, under the influence of this emotion, either take to flight, 1 Op. cit. 'The Senses and the Will,' pt. ii. ch. xi.

2 Ibid.

and afterwards conceal himself, or, on the other hand, he may conceal himself without flight by creeping into cover, like a crab that buries itself in the sand, or withdraws into the crevice of a rock, or like a child that hides under the bed-clothes; or, on the other hand, he may take to flight and, like birds, when at a sufficient distance, feel secure without further protection. According to the theory we are considering, each one of these instincts ought to have a different primary emotion as its affective aspect; but we feel only the one emotion of fear in both cases. We do not deny that this fear has different sensations connected with it when we are running away or lying still; but it is in both cases the same primary emotion.

(3) In the third place, the same instinct may be connected with the systems of different emotions. That an instinct is organised in one does not prevent it being also organised in another. The same habits may be useful in different systems, so also the same instincts; just as the same emotions function over and over again in a variety of sentiments. The simple instincts of sitting, or standing, or grasping objects with the hand, the instincts of locomotion, evoking the peculiar instinctive movements by which an animal propels itself, whether by walking or flying, or swimming, or by sinuous movements are indispensable to the systems of many emotions, and may be excited through fear, anger and the appetites. Thus the instinct of flight in birds may be roused to activity through the emotions of fear, and of anger, and through the appetites of hunger and of sex. The combative instincts connected with the emotion of anger are found also in a modified form in connection with the enjoyment of play; as we see when dogs play at biting one another, and alternately give chase and take to flight. What instincts may not enter the system of this wonderful emotion of play?

If then an emotion is not the affective aspect of the excitement of an instinct, what is that which corresponds most closely to an instinct in consciousness? An impulse; and, consequent on it, the sensations that accompany the subsidiary motor response. But an emotion with its system is a more comprehensive fact than an instinct with its impulse, because in the former system there may be organised a

variety of instincts with their impulses, all of which subserve in different situations the same end of the emotion. There may be also included in an emotional system a number of acquired tendencies which increase the variety of methods at its disposal. Corresponding with these innate and acquired tendencies in the system of an emotion there will then be felt and included a variety of impulses. In the system of fear there are the impulses of flight and of concealment, and others, as we shall see, connected with different innate or acquired dispositions,-felt in different situations; but felt with one and the same primary emotion of fear.

Not only is the system of a primary emotion a more comprehensive fact than the system of an instinct, but in man it becomes also a much more important fact. An instinct has only one kind of behaviour connected with it, and, when the appropriate stimulus excites it, must tend to respond with this one kind of behaviour. The same emotion in man has a variety of different kinds of behaviour connected with it, both instinctive and acquired, and when an appropriate stimulus excites it, is not compelled to respond with the same kind of behaviour, but may select that one which in the situation is the most appropriate. We shall therefore conclude with the enunciation of the following tentative law: (26) Every primary emotion tends to organise in its system all instincts that are serviceable to its innately determined end, and to acquire many serviceable tendencies which modify such instincts.

3. Of the Significance of the Physiological Theory of Emotion for the Science of Character

As the science of character must conceive of the primary emotional systems as forces innately determined to pursue certain ends, and possessing certain hereditary and acquired methods for achieving them; and as moreover its principal problem in respect of these systems is to understand the laws of their action and interaction, so from the point of view of this science we have to judge of the bearing on it of the physiological theory of emotion. This physiological theory holds that the organic sensations, which are so conspicuous in intense emotions, are due to the alteration of the function of

different organs, and that these alterations of function have two contrasted effects, are either stimulating or depressing.

Now since the emotions are forces organised to achieve certain results or ends, then it is clear that these functional modifications connected with them will either further the attainment of these ends or impede them, and that in a given case they may have partly the one effect and partly the other; either by affording the emotional system the increase of strength which it needs, or by favouring it in some other way; or, on the other hand, by unduly weakening it, or by rendering it in some other way less fitted to attain its ends. Sometimes these opposite effects will be manifested by the same emotion under different conditions, or at different degrees of its intensity or prolongation. Fear, for instance, may be so depressing as to paralyse us, or so stimulating as to make us run as "if the devil was behind us." Prolonged fear is held to be always depressing, yet it may be so sustaining to man's inconstant will that he gives henceforth all his days to "working out his salvation with fear and trembling."

If an animal has to escape by instant flight, and is so paralysed by fear as to be unable to move, then these functional changes are unfavourable to the efficiency of the emotional system. In other cases where a creature has to lie still, and escape detection, the depressing effect of fear may be advantageous to it. Again the influence of these functional changes may be mixed. The great increase of muscular strength that accompanies anger is favourable to a system which is essentially aggressive, but the trembling which is often a feature of violent anger, unfits a man for delicate movements of attack and defence. Thus such functional changes belonging to the system of an emotion may be beneficial to it in one direction and detrimental in another; but we may expect that, as they are innate features of it, they are on the whole beneficial, and that natural selection has favoured their development.

The two significant features of the systems of the primary emotions on their physiological side are, then, the instincts which belong to them, and those functional changes which we have been considering. The one furnishes the proximate