We have hitherto considered two artificial groups and have found that they are dominated by two emotional ties. One of these, the tie with the leader, seems (at all events for these cases) to be more of a ruling factor than the other, which holds between the members of the group. Now much else remains to be examined and described in the morphology of groups. We should have to start from the ascertained fact that a mere collection of people is not a group, so long as these ties have not been established in it; but we should have to admit that in any collection of people the tendency to form a psychological group may very easily become prominent. We should have to give our attention to the different kinds of groups, more or less stable, that arise spontaneously, and to study the conditions of their origin and of their dissolution. We should above all be concerned with the distinction between groups which have a leader and leaderless groups. We should consider whether groups with leaders may not be the more primitive and complete, whether in the others an idea, an abstraction, may not be substituted for the leader (a state of things to which religious groups, with their invisible head, form a transition stage), and whether a common tendency, a wish in which a number of people can have a share, may not in the same way serve as a substitute. This abstraction, again, might be more or less completely embodied in the figure of what we might call a secondary leader, and interesting varieties would arise from the relation between the idea and the leader. The leader or the leading idea might also, so to speak, be negative; hatred against a particular person or institution might operate in just the same unifying way, and might call up the same kind of emotional ties as positive attachment. Then the question would also arise whether a leader is really indispensable to the essence of a group--and other questions besides. But all these questions, which may, moreover, have been dealt with in part in the literature of Group Psychology, will not succeed in diverting our interest from the fundamental psychological problems that confront us in the structure of a group. And our attention will first be attracted by a consideration which promises to bring us in the most direct way to a proof that libidinal ties are what characterize a group. Let us keep before our eyes the nature of the emotional relations which hold between men in general. According to Schopenhauer's famous simile of the freezing porcupines no one can tolerate a too intimate approach to his neighbour.[34]